pdf LRI Scientific report 2012

LONDON RESEARCH INSTITUTE SCIENTIFIC REPORT 2012

Cover image Montage of human cells at different stages of the division process (chromosomes in blue, mitotic spindle microtubules in green, and the cortical actin cytoskeleton in red). Image supplied by the Cell Division and Aneuploidy Group, Clare Hall.

SCIENTIFIC REPORT 2012

CONTENTS

DIRECTOR’S INSTITUTE INTRODUCTION FRANCIS CRICK PROGRESS REPORT 2012 SPECIAL REVIEW: DAVID ISH-HOROWICZ SPECIAL REVIEW: GORDON PETERS RESEARCH HIGHLIGHTS

5 8 10 14 18

Caetano Reis e Sousa Immunobiology Erik Sahai

RESEARCH GROUPS – LINCOLN’S INN FIELDS 27 Facundo D Batista 28

Molecular Neuropathobiology

Lymphocyte Interaction

Gene Expression Analysis 30

Mammalian Genetics Dominique Bonnet

34

38

42

46

50

54 56

Molecular Oncology

Sharon Tooze Richard Treisman Frank Uhlmann Helen Walden Michael Way

RESEARCH GROUPS – CLARE HALL Simon Boulton Peter Cherepanov Alessandro Costa

60

Architecture and Dynamics of Macromolecular Machines Vincenzo Costanzo DNA Damage and Genomic Stability

2

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

82 84 86 88 90

93 94 96

Chromatin Structure and Mobile DNA 58

Protein Phosphorylation Gordon Peters

80

DNA Damage Response

Cell Cycle Peter J Parker

Takashi Toda

Cell Motility

Structural Biology Paul Nurse / Jacqueline Hayles

78

52

Tumour Host Interaction Neil Q McDonald

Barry Thompson

Protein Structure Function

Lymphatic Development Ilaria Malanchi

76

Chromosome Segregation 48

Computational Biology Taija Makinen

Nicolas Tapon

Signalling and Transcription

Cell Biophysics Nicholas Luscombe

74

Secretory Pathways 44

Developmental Genetics Banafshe Larijani

Charles Swanton

Cell Regulation

Developmental Signalling David Ish-Horowicz

72

Epithelial Biology 40

Immuno Surveillance Caroline Hill

Thomas Surrey

Apoptosis and Proliferation Control

Vascular Biology Adrian Hayday

70

Translational Cancer Therapeutics 36

Signal Transduction Holger Gerhardt

Martin R Singleton

68

Microtubule Cytoskeleton

Telomere Biology Julian Downward

Almut Schulze

66

Macromolecular Structure and Function 32

Haematopoietic Stem Cell Julia Promisel Cooper

64

Tumour Cell Biology Giampietro Schiavo

Axel Behrens

62

98

100

John FX Diffley

102

Chromosome Replication Peter Karran

104 106 108 110 112

115 117 118

Bioinformatics and Biostatistics Paul Bates

119

Biomolecular Modelling Research Ruth Peat

120

Cell Services Lucy Collinson

Alexander Tournier

132

Stephane Mouilleron

132

Anne Vaahtokari

133

Super-Resolution Microscopy

Advanced Sequencing Aengus Stewart

131

Protein Structure

Genetic Recombination TECHNOLOGY CORE FACILITIES Nik Matthews

Francois Lassailly

Mathematical Modelling

DNA Damage Tolerance Stephen C West

131

In Vivo Imaging

Mechanisms of Gene Transcription Helle Ulrich

Ali Alidoust Fermentation

Cell Division and Aneuploidy Jesper Q Svejstrup

130

Transgenics

Mammalian DNA Repair Mark Petronczki

Ian Rosewell

RESEARCH PUBLICATIONS AND THESES 135 RESEARCH PUBLICATIONS 136 THESES 154 INSTITUTE INFORMATION 155 ADMINISTRATION 156 ACADEMIC PROGRAMME 158 SEMINARS AND CONFERENCES 160 EXTERNAL FUNDING 162 INSTITUTE MANAGEMENT 164 CONTACT DETAILS IBC

121

Electron Microscopy Graham MG Clark

122

Equipment Park Gordon Stamp

123

Experimental Histopathology Derek Davies

124

Fluorescence Activated Cell Sorter Mike Howell

125

High-Throughput Screening Daniel Zicha

126

Light Microscopy Nicola O’Reilly

127

Peptide Synthesis Bram Snijders

128

Protein Analysis and Proteomics Svend Kjaer Protein Purification

129

Image of the spleen showing B cells (red), T cells (blue) and CD169+ macrophages (green).

CONTENTS 3

Top: Clare Hall Laboratories. Above: Lincoln's Inn Fields Laboratories.

4

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

DIRECTOR’S INSTITUTE INTRODUCTION

Richard Treisman

Alessandro Costa

2012 saw the fiftieth anniversary of completion of the Lincoln’s Inn Fields laboratories, and ten years since the establishment of Cancer Research UK and the LRI. To mark the occasion a special LRI 10th Anniversary Symposium, organised by Steve West and Julian Downward, was held in June at our neighbours, the Royal College of Surgeons. Presentations by former graduate students, postdocs and group leaders from the LRI, its predecessor the ICRF, and the current LRI gave a vivid picture of how the Institute continues a long tradition of scientific innovation and excellence. Group leader recruitments and departures are essential for a research institute such as LRI to refresh its outlook and adapt to new research directions. While there have been no departures in 2012, Alessandro Costa opened his Macromolecular Machines laboratory at Clare Hall in March, while Nick Luscombe, a joint LRI-UCL appointee, moved his Computational Biology group from EBI to Lincoln’s Inn Fields in May. Bram Snijders joined LRI as head of the Proteomics and Protein Analysis technology core laboratory, following the departure of Mark Skehel to Cambridge. Bram is already building links with

NIMR scientists pending our move to the new Crick laboratories in 2015. Two junior group leaders, Nate Goehring and Dinis Calado, from MPI-CPG (The Max Planck Institute of Molecular Cell Biology and Genetics), Dresden and MDC (The Max Delbrück Center for Molecular Medicine), Berlin respectively, will join LRI next year to develop programmes in cell polarity and mouse models of B cell leukemia. Scientific review is essential for ensuring appropriate allocation of our core resources, and for Cancer Research UK to evaluate the

Nick Luscombe

Bram Snijders

Jessica Strid winner of the 2011 Hardiman-Redon Prize, presenting her work at the LRI 10th Anniversary Symposium.



DIRECTOR’S INSTITUTE INTRODUCTION

5

False-coloured scanning electron micrograph of connective tissue. Image: Anne Weston, Electron Microscopy Facility.

performance of LRI research groups alongside that of its grantees. Reviews involve a site visit by up to six external scientific reviewers usually from overseas, and consultation of several peer referees per programme, and take place over 1-2 days. Promotions from Junior to Senior Group Leader are also subject to stringent review, and Helen Walden was successfully promoted to Senior Group Leader in July. It was also an extremely busy year for senior group leader review at LRI, with no fewer than 21 programmes being evaluated, at four separate review visits. Gratifyingly, no fewer than six scored at the highest level ('O-O' for the cognoscenti). Congratulations to all those reviewed and thanks to the CRUK Science Research and Funding department and LRI admin for ensuring that the process ran smoothly. The standing of LRI scientists within their fields is recognised in numerous ways including membership of funding review panels, both within Cancer Research UK and with other UK and overseas funding organisations. Holger Gerhardt and Adrian Hayday are members of the Cancer Research UK Science Committee, Charlie Swanton and Peter Parker are members of the Biomarker review panel, Julian Downward is vice-chair of the Drug Discovery review panel, and Sally Leevers and Adrian are members of the Training Board. Other marks of recognition are scientific honours and awards, and it is always a pleasure to record these. In 2012 Simon Boulton was elected Fellow of the Academy of Medical Sciences, and Thomas Surrey and Axel Behrens were elected members of EMBO. Julian Downward was elected an honorary

6

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Fellow of the Royal College of Physicians, no mean feat for a non-clinical scientist. We were also delighted to learn of Julian Lewis’ election to the Royal Society, following his retirement from LRI last year. Congratulations also to Julian Downward, Thomas Surrey and Helle Ulrich, who were recipients of ERC Advanced research grants during the year, and to Lucy Collinson, who won a consortium grant from the MRC for development of novel microscopy techniques. It is also very pleasing to review 2012 awards to LRI junior researchers. Anne Weston (EM core) and Kuan-Chung Su (Petronczki group) both received Wellcome Trust Image Awards. KuanChung for the spiral view of mitosis and Anne received two awards for her false-coloured scanning electron micrograph of a Diatom frustule and Connective Tissue. Anne also won first prize in the Royal Microscopy Society micrograph competition. Graham Bell (Thompson Group) won the Richard Hambro Poster Prize at the 2012 NCRI meeting in Liverpool, and Ashley Humphries (Way Group) won a poster prize at the EMBO meeting in Nice. LRI students performed exceptionally well at this year’s International PhD Student Cancer Conference in Amsterdam, with Rafal Lolo (Boulton Group) and Risa Mori (Toda Group) winning first and second place talk awards, and Jeroen Claus (Parker Group) winning the poster prize. As a fundraising charity, Cancer Research UK is raising the capital funds for its contribution to the Crick Institute project through a special appeal, 'Create The Change', which is distinct from the

Winner of Wellcome Image Award Kuan-Chung Su and Mark Petronczki. Composite confocal micrograph uses time-lapse microscopy to show a cancer cell (HeLa) undergoing cell division (mitosis). The DNA is shown in red, and the cell membrane is shown in cyan.

Charity’s day-to-day fundraising activities. The Create The Change board is chaired by Charles Manby, who brings considerable experience in the investment-banking arena and as leader of numerous fundraising campaigns in the charitable and education sectors. The board, whose membership was announced in November, consists of influential and well-connected senior volunteers who will advise on and engage

August 1962 the first phase of a new purpose built research institute in Lincoln’s Inn Fields was completed.



prospective donors. LRI staff have continued to play their part in Create the Change events over the year, which saw the total contribution pass £25 million, good progress toward the overall campaign goal of £100 million. As LRI becomes fully integrated into the Crick project, an increasing number of activities are being planned jointly with NIMR and the Crick University partners. Our 2011 group leader recruitment round already saw our Faculty Committee and interview panels augmented by Chris Boshoff or Tariq Enver from UCL, and Jim Smith from NIMR. In 2012, the process has been further integrated, with recruitment undertaken under a formal LRI-Crick banner with joint interviews at the Crick offices. The LRI postdoc retreat now also takes place under Crick auspices, and a successful first joint meeting was held at KCL this year. A third Crick group leaders’ retreat will be held in early 2013 with colleagues from NIMR and the three Crick university partners. The ten years since the foundation of Cancer Research UK and the birth of LRI have seen substantial change: the new CRUK Cambridge Research Institute has opened; the Gray Institute has been reborn in Oxford in partnership with the MRC as the Gray Institute for Radiation Oncology and Biology; the Beatson Institute has moved into a splendid new building; and the Paterson has become part of Manchester University. As our second decade begins much hard work is going into ensuring the success of the next big change, our transition into the new Francis Crick Institute.

DIRECTOR’S INSTITUTE INTRODUCTION

7

FRANCIS CRICK INSTITUTE PROGRESS REPORT

2012 has been the year when the Francis Crick Institute has begun to take tangible and substantial physical shape. Planning for the organisation and operation of the new Institute, and the migration of the founding Institutes to the new building started in earnest this year. Scientific integration between the founding Institutes has continued, while the Crick has begun to increase its profile in the scientific community in London with a series of symposia, seminars and workshops. There have been a number of changes to the Crick Executive during the year. Jim Smith and Richard Treisman joined formally as Research Directors, Keith Peters as Crick Clinical Consultant Director, and Amanda Towse as HR Director, while Geoff Dobson and David Livesley left the project. Phil Butcher has joined the Crick, seconded from the Sanger Institute, to lead on IT, while LRI’s Ava Yeo and NIMR’s John Wills have joined on a part-time basis as transition leads for Science Operations and Science Facilities respectively. Work to develop the Crick transition gathered pace, initially under the direction of David Livesley and latterly Andy Smith, producing a set of initial transition proposals. Amongst these were included science and operational strategy, science technology platforms, scientific equipment requirements, IT strategy, biological resources, stores and logistics, organisational model and HR issues, and an outline

plan for migration to the new laboratories. The proposals were reviewed favourably by the Crick Board in December. In parallel with this work, the Crick has been further developing its scientific and operational strategy through discussions with its University partners concerning interdisciplinary interactions, and with both the Universities and Institute group leaders on the research programme. An important aspect of this has been the matching of the proposed Crick PhD training programme to the different regulations in force at the partner universities. The third Crick Group leaders retreat, which will be held at Ashridge in Spring 2013, will consider aspects of the developing Crick scientific strategy, with the scientific programme reflecting the potential for interdisciplinary interactions afforded by our the Crick University partners.

Progress at the site of the new Francis Crick Institute. Left panel January 2012. Right panel December 2012.

8

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Audience participating at the Second Crick Symposium in November on Infection and Immunity.

Perhaps the most important Crick event during the year from the perspective of its future scientific programme was the Science Assessment Process. While both MRC and CRUK subject the science at NIMR and LRI to quinquennial review to determine quality and resourcing, the criteria used by each differ somewhat. The Crick Science Assessment was undertaken to provide a unified view of the scientific programmes at the two institutes, in order to enable Crick to allocate its resources appropriately once the new Institute is up and running. An international panel of over twenty senior scientists was asked to review NIMR and LRI programmes against a Crick quality standard set to be comparable with the most prestigous funding schemes worldwide. The SAP was undertaken in two stages, in July and October. 2012 saw huge progress on the Crick building project. From a 3.5 acre hole in the ground in January, the year has seen excavation of a second basement level at the same time as construction back up to ground level and beyond. By year’s end, the structure had reached level 5 at the west end of the building: a trip to the nascent laboratory floors allowed appreciation of their scale, and gave a good feel for how the laboratory write up spaces look onto the atrium space on one end and external world at the other. The building programme is on schedule, with topping-out expected in early 2013, followed by the fit-out which is scheduled to be complete in June 2015. With construction underway, planning work has moved to the interior design and furnishings of the new laboratories. At the turn of the year a fullsize mockup of a Crick laboratory interior was built in a warehouse close to Heathrow airport. Laboratory furnishing bids were formally reviewed alongside it by the Crick, with substantial participation from



Institute and university staff. Following selection of the successful bidder, the mockup was fitted out for open viewing, with much positive feedback from institute staff concerning the laboratory layout and furnishings. During the period leading up to the formal opening of the Crick Institute, the Crick will be running a programme of symposia, workshops and seminars to raise its profile in the research community. Nic Tapon (LRI) and Alex Gould (NIMR) are leading on the development of the programme. The first two Crick Symposia, held in July and November, focussed on ‘Metabolism in health and disease’ and ‘Immunity and Infection’ and have attracted audiences of several hundred from London and further afield. The third symposium, on ‘Genes to Therapies’ will be held in March 2013. The Crick Symposia programme is complemented by Crick-sponsored workshops and seminars held on a variety of topics. In addition to raising the profile of the Crick in the scientific community, these events are generating closer links between researchers at LRI, NIMR and the Crick University Partners. As 2013 opens, another major milestone in the development of the new Institute will be passed, with the start of formal consultation on the transfer of affected staff from Cancer Research UK and the MRC to the Francis Crick Institute. This consultation will be done via elected staff representatives sitting on a new LRI Staff Consultative Forum. This will replace the LRI Crick Institute Staff Discussion forum, which has been the main venue for informal discussion of Crick matters with staff until now. At the time of writing, ballots are being cast for elections to the new body, which will meet for the first time early in the New Year. Richard Treisman

DIRECTOR’S INSTITUTE INTRODUCTION

9

SPECIAL REVIEW: DAVID ISH-HOROWICZ How the fruitfly got its stripes and other Just-so Stories

David Ish-Horowicz

I never actually considered being anything other than a research scientist (although I expected to become a chemist, had no understanding or interest in biology, and didn’t discover genetics till halfway through my PhD). It would be fun; I would meet interesting people, and be able to choose my own working hours and working clothes – no suits or ties! And there was the compelling idea that I might discover something novel and important. It took me a little while to realise that, to misquote Peter Parker (Spider-Man, not the research scientist), great advantages imply great responsibility, in particular, the many hours of hard work needed to cross many troughs and climb a few exhilarating peaks. Of course, it also required good luck, sometimes making it myself and, more often, stumbling into it. Taking flight My good fortune began with accidentally choosing to pursue a PhD at the MRC Laboratory of Molecular Biology, working on tRNA with Brian Clark as a forgiving supervisor who ignored my initial indolence. It continued when I had to make a last-minute switch for my postdoc to Walter Gehring’s brilliant lab in Basel. Towards the end of the latter period, and despite my lack of coherent results, I was hired as a junior group leader at the Imperial Cancer Research Fund Mill Hill Labs by John Cairns, after he’d decided that developmental biology and genetics would be key to understanding cancer. Typically, John was far ahead of his time in appreciating the need to study ‘normal’ in vivo gene function in order to also understand pathological contexts. Unlike hirings today, he didn’t need to pay heed to my (lack of) postdoc publications. In early 1977, I arrived at the Labs, which were housed in a ramshackle set of buildings next to NIMR. I joined a small community of exciting young scientists, including Jonathan Slack, Jeff Williams and Brigid Hogan, with similar ideas of what constituted the interesting questions, albeit with differing views of how to tackle them. DNA cloning was still in its infancy, and I and my dedicated, long-suffering and long-serving

10

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

colleague, Sheena Pinchin, spent the next few years honing molecular biology skills while working on heat shock genes and extrachromosomal transposons. Stripes, segmentation, sex and structure By 1980, I knew that I wanted to work on Drosophila segmentation, based on the classic genetic screen of Janni Nüsslein-Volhard and Eric Wieschaus that deservedly won them the Nobel Prize in 1995. They had been my colleagues while I was a postdoc in Walter Gehring’s lab but, although I was fully informed of their work, I delayed an immediate switch of research direction because cloning of specific developmental genes seemed ‘too difficult’ at that time. It was a further three years before my group embarked on the experiments that led to Phil Ingham, Ken Howard and me cloning the hairy ‘pair-rule’ gene that is required for embryonic segment formation1. Janni was sceptical about our choice – it didn’t have a sufficiently ‘pretty’ mutant phenotype – but the gene proved a winner. Not only was it transcribed in stripes that corresponded to its effects on patterning (Figure 1), but later experiments showed that the striped pattern, i.e. the segmented bodyplan, is hardwired into the Drosophila genome, with individual or small groups of stripes being encoded by

distinct ‘stripe enhancers’. Even today, these stripes are the most astounding and best-studied example of how enhancers, activators and repressors combine to give spatially-restricted transcription. During this period, the lease on the Mill Hill Labs expired, and the ICRF’s Director, Walter Bodmer, did us the (initially under-appreciated) favour of founding a new Developmental Biology Unit (DBU) in Oxford, under the direction of Richard Gardner. I moved cities somewhat reluctantly in 1985, not realising how much fun Oxford would be both scientifically and socially. Within a couple of years, Jonathan Slack and I had been joined as group leaders by Rosa Beddington, Julian Lewis and Phil Ingham. The quality of these colleagues, our common research agenda, and the presence of all the main model organism systems within a small unit, made for a heady intellectual brew and some wonderful science. My group continued its studies of hairy, showing that it encodes a basic-helix-loop-helix repressor protein2,3, one of several related proteins with immediate relevance to a variety of other developmental contexts. Perhaps our most unexpected (and ‘sexiest’) result was Susan Parkhurst’s finding that early, ectopic hairy

expression selectively killed female embryos 4 . Drosophila has a radically different mode of sex determination from humans: flies with 2 X-chromosomes develop as females, and 1X animals as males. Our work showed this difference is triggered by the balance between X-linked transcriptional activators and autosomal transcriptional repressors. A fortunate meeting at a conference led to a collaboration with Roger Brent (then at Massachusetts General Hospital) in which Ze’ev Paroush used the fledgling yeast two-hybrid system to show that Hairy represses transcription by recruiting a non-DNA-binding protein, Groucho, to target genes 5. This was perhaps the first demonstration of a metazoan co-repressor. Subsequent work by us and others showed that Groucho and its vertebrate homologues are co-repressors for a very wide range of other transcription factors. Many years later, a collaboration with Laurence Pearl (then at the Institute of Cancer Research) used a combination of genetics and structural biology to explain Groucho’s promiscuity in choosing partners 6. How chickens know the time Drosophila and vertebrates appear to differ radically in their morphogenesis and modes of

Figure 1 Striped expression of two pair-rule gene transcripts. Double-label in situ hybridisation for hairy (green) and fushi tarazu (red) transcripts shows striped expression in syncytial blastoderm embryos. Transcripts accumulate exclusively apically of the layer of peripheral nuclei (blue). From [13].



SPECIAL REVIEW

11

early patterning. In particular, it was believed that their nervous systems had evolved separately. Nevertheless, Julian Lewis and I decided to combine forces to see if recent advances in the understanding of Drosophila neurogenesis might offer new insights into vertebrate neural development. This project paved the way to a long and fruitful association between our labs. The collaboration began with Domingos Henrique cloning a wide variety of vertebrate homologues of important Drosophila genes, including the first cloning of Notch ligands in vertebrates. Most important of the latter was chick Delta1 (Dll1 in mouse). Domingos’s classic experiments in chick and frog showed that Delta-Notch signalling regulates progenitor cells’ decisions whether to self-renew or differentiate, similar to one of Notch’s major roles in Drosophila 7,8 (Figure 2). Not only did this work indicate the evolutionary conservation of neural patterning mechanisms, but it provided the tools for others to show that Notch signalling regulates many different cell-fate decisions during vertebrate development. Much later, it became clear that Notch is actually a major human oncogene for T-cell acute lymphocytic leukaemia. Domingos’s work also provided a most unexpected entrée to another fundamental question. Segmentation in flies and vertebrates are driven by very different mechanisms, occurring simultaneously in the former and sequentially in the latter. Domingos cloned chick hairy genes and, with Olivier Pourquié (then in Marseille), found that their expression oscillates in Figure 2 Notch signalling is required to maintain neural progenitors in the chick retina. (A) Overactivating Notch signalling (green cells) suppresses differentiation into Islet-1-expressing neurons (red). (B) Blocking Notch signalling leads to massive premature and excessive production of neurons that extend throughout the neuroepithelium. Affected domains are marked with white bars. From [8].

A

B

12

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

precursor cells with the same periodicity as segment (somite) formation (Figure 3) 9. This cyclic transcription provided the first evidence supporting a longstanding ‘clock-and-wavefront’ theoretical model for reiterated somitogenesis 11. Thus, not only did hairy provide a common theme for our studies of segmentation in Drosophila and vertebrates, but the realisation that dynamic Notch signalling is also linked to the segmentation clock marked an unforeseen convergence with our studies of vertebrate neurogenesis. After ten years in Oxford, another move beckoned. Developmental biology had undergone a dramatic rise in status, and the relevance to cancer (and everything else) of studying ‘normal’ gene function in a complex, genetically-tractable animal was being increasingly appreciated. Thanks to the Unit’s focus and hands-off management, together with a supportive and interactive environment, it was too successful to be left in Oxford. Such thinking triggered Paul Nurse, the new Deputy Director of ICRF, to transfer the DBU scientists to Lincoln’s Inn Fields. Of course, we were disappointed at the dismemberment of such an incredibly productive unit, but at least it was for the right reasons, to bring new ways of thinking to an already exciting hub of science. The new environment also served to revitalise our work. Motoring RNA around the cell In parallel with studying Drosophila segmentation, our work began to broaden to include a very different aspect of embryonic patterning – intracellular localisation of mRNAs by molecular motors. This process underlies establishment of the major body axes during Drosophila oogenesis, and contributes to patterning of our old friend, the early embryo. The topic was first pursued at the DBU by Ilan Davis when he was a graduate student 12 . Much later, breakthrough experiments at LIF from Sabbi Lall and Simon Bullock developed injection assays showing that fluorescentlylabelled pair-rule and other RNAs localise in a microtubule-dependent manner, and recruit and require the dynein accessory proteins, Bicaudal-D and Egalitarian 13. These rapid in vivo assays for analysing signal structure and activity ultimately led to similar studies of RNA localisation during oogenesis (by Ilan Davis, initially in Edinburgh and then Oxford), and the

Lfng

Figure 3 Cyclic expression of the mouse lunatic fringe segmentation gene in the PSM. In situ hybridisation of three embryos showing three sequential phases of cyclic transcription. Arrowheads and lines shows positions of newest and incipient somite boundaries; posterior is downwards. From [10].

determination of the first (and so far only) solution structure of one such signal (with Simon Bullock and Peter Lukavsky, MRC Laboratory of Molecular Biology) 14 . The structure identified novel forms of double-stranded RNA, and suggested that these contribute to signal specificity, work now being continued by Simon in Cambridge.

‘unknown unknowns’ surely outnumber the ‘known unknowns’, and we will need all possible research weapons if we’re to really understand how genes work in vivo and how they can go wrong. Let’s hope that the critical thinking and scientific precision of Drosophila research continues to play a prominent role in this quest.

A fresh start After more than 15 years in LIF, it’s time for me to move again, and to hang out with a new array of colleagues. I’m still amazed at the tremendous pace of current biological advance, addressing fundamental questions with a speed that was inconceivable thirty years ago when I first chose to work with Drosophila. Yet we still consistently overestimate how much we really know; the

I should acknowledge my incalculable debt to ICRF/CR-UK for their support, and to the many wonderful friends, colleagues and collaborators (only a few of which I’ve been able to name) who have helped make science so much fun. I’ve had the privilege of working during a “golden age” of molecular and cellular developmental biology, and of consorting with inspiring colleagues. Hopefully I’ve inspired a few in turn.

References

1. Ingham PW, Howard KR, Ish-Horowicz D. Transcription pattern of the Drosophila segmentation gene hairy. Nature. 1985; 318:439-45

2. Ish-Horowicz D, Pinchin SM. Pattern abnormalities induced by ectopic expression of the Drosophila gene hairy are associated with repression of ftz transcription. Cell. 1987; 51:405-15 3. Rushlow CA, Hogan A, Pinchin SM, Howe KR, Lardelli MT, et al., The Drosophila hairy protein acts in both segmentation and bristle patterning and shows homology to N-myc. EMBO J. 1989; 8:3095-103 4. Parkhurst SM, Bopp D, Ish-Horowicz D. X:A ratio in Drosophila is transduced by helix-loop-helix proteins. Cell. 1990; 63:1179-91 5. Paroush Z, Finley RLJ, Kidd T, Wainwright SM, Ingham PW, et al., Groucho is required for Drosophila neurogenesis, segmentation and sex-determination, and interacts directly with Hairy-related bHLH proteins. Cell. 1994; 79:805-15.

8. Henrique D, Hirsinger E, Adam J, le Roux I, Pourquié O, Ish-Horowicz D, Lewis J. Maintenance of neuroepithelial progenitor cells by Delta-Notch signalling in the embryonic chick retina. Curr Biol. 1997; 7:661-670 9. Palmeirim I, Henrique D, Ish-Horowicz D, Pourquié O. Avian hairy gene expression identifies a molecular clock linked to vertebrate segmentation and somitogenesis. Cell. 1997; 91:639-48 10. Morales AV, Yasuda Y, Ish-Horowicz D. Periodic lunatic fringe expression is controlled during segmentation by a cyclic transcriptional enhancer responsive to Notch signaling. Dev Cell. 2002; 3:63-74 11. Cooke J, Zeeman EC. A clock and wavefront model for control of the number ofrepeated structures during animal morphogenesis. J Theor Biol. 1976; 58:455-76 12. Davis I, Ish-Horowicz D. Apical localisation of pair-rule transcripts requires 3' sequences and limits protein diffusion in the Drosophila blastoderm embryo. Cell. 1991; 67:927-40

6. Jennings BH, Pickles LM, Wainwright SM, Roe SM, Pearl LH, et al., Molecular recognition of transcriptional repressor motifs by the WD domain of the Groucho/TLE corepressor. Mol Cell. 2006; 22:645-55

13. Bullock SL, Ish-Horowicz D. Conserved signals and machinery for RNA transport in Drosophila oogenesis and embryogenesis. Nature. 2001; 414:611-6

7. Henrique D, Adam J, Myat A, Chitnis A, Lewis J, Ish-Horowicz D. Expression of a Delta homologue in prospective neurons of the chick. Nature. 1995; 375:787-90

14. Bullock SL, Ringel I, Ish-Horowicz D, Lukavsky PJ. A'-form RNA helices are required for cytoplasmic mRNA transport in Drosophila. Nat Struct Mol Biol. 2010; 17:703-9.



SPECIAL REVIEW

13

SPECIAL REVIEW: GORDON PETERS How did it come to this? From MMTV to Polycomb via FGF3, Cyclin D, INK4a and ARF.

Gordon Peters

As I negotiated the double doors at 44 Lincoln's Inn Fields in July 1977, I never imagined that I’d be doing the same 35 years later. I had just returned from a post-doc position in the US and had the vague idea that spending two or three years at the ICRF could be a stepping-stone to a junior lectureship somewhere in the UK. In hindsight, I was incredibly fortunate that Robin Weiss offered me bench-space, my own scientific officer and freedom to do what I wanted. It was pretty scary as in the privileged and stimulating environment that prevails at LIF, you are expected to aim high. In those formative years, I learnt the importance of picking the right questions and what follows is an account of where this led me and why, intended as entertainment rather than celebration. I would have liked to mention all the people who were part of the adventure but the list is frighteningly long.

Getting going with tRNAs As a post-doc with James Dahlberg at the University of Wisconsin, I had become interested in the host tRNAs that are incorporated into retrovirus particles to prime reverse transcription. At LIF, among the first people I encountered was Clive Dickson who, as well as being one of the nicest guys you could hope to meet, was an expert on mouse mammary tumour virus (MMTV). It wasn’t hard to spot an opportunity. It turned out that MMTV uses tRNALys3 as a primer, rather than the tRNAPro used by most murine retrovirus and by HTLV. Not a huge deal, except that HIV also uses tRNALys3. Looking for oncogenes Clive and I quickly realised the benefits of symbiosis and began a collaboration that would last for many years, including my 8-year posting to the ICRF labs at Bart’s. I was also fortunate to recruit Sharon Brookes as my scientific officer and it speaks volumes that she is running the lab to this day. Back then, the real action in the retrovirus field was not the proteins that would eventually be targeted by anti-retroviral drugs but the identification of oncogenes that had been incorporated into the viral genome at the expense

14

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

of the structural proteins: the v-onc genes like Ras, Src, Fos, Myc etc. The problem with MMTV was that it belonged in the ‘boring’ category of virus that did not contain a viral oncogene and did not transform cells in culture. The boring label was unjustified as these nontransforming viruses cause tumours in animals at high frequency, albeit after a long latency. All became clear when Hayward and his colleagues showed that in lymphomas induced by avian leukosis virus, there is invariably a copy of the viral genome (provirus) integrated next to c-Myc. Cue the scramble to find novel oncogenes by identifying common proviral integration sites in panels of tumours, against the background of endogenous proviruses. The MMTV field was quick off the mark, and in 1982, Nusse and Varmus identified a new gene that they called Int-1 (integration site 1). We were taking a similar approach but in a different mouse background and, luckily for us, came upon a different target gene 1. Almost by accident it became known as Int-2, not a great PR move. I recall hearing Int-1 being described as a “minor oncogene”, but the speaker was clearly not an epidemiologist, nor a clairvoyant. Every MMTV-induced tumour has a

Figure 1 Location and orientation of MMTV proviruses in Int-1 and Int-2. From [2].

Int-1/ch15

Int-2/ch7

provirus at one or other Int locus and one of my personal highpoints was our Nature paper 2 showing that many tumours had insertions at both Int-1 and Int-2, one the first examples of oncogene co-operation in vivo (Figure 1). Identity crisis It did not take long to spot that Int-1 was related to the Drosophila wingless gene and it was subsequently rebranded as Wnt-1, the founding member of that illustrious family. For us, the ‘brand refresh’ was less favourable; the top spots went to acidic and basic fibroblast growth factors and Int-2 was designated as Fgf-3. At first it felt like a misnomer; until we cloned the Xenopus homologue, some years later, we struggled to find evidence that the protein behaved as a secreted growth factor. That said, times were good and we were the acknowledged world leaders in the Int-2 field. The reality, of course, was that until the embryologists became involved, we were the only people working on Int-2/Fgf-3 and its frustrating idiosyncrasies. Figure 2 Amplification of 11q13 markers in human breast cancers. From [3].



The human touch The standard justification for working on MMTV was that it could provide insights relevant to human breast cancer and with this in mind, we cloned and sequenced the human homologue, INT2/FGF3, and mapped it to chromosome 11q13. With help from the ICRF Clinical Oncology Unit at Guys Hospital and generous collaborators, we found that INT2/FGF3 and the adjacent HST1/FGF4 gene are amplified in about 15% of primary breast cancers, and similar observations were made for other tumour types. It even had prognostic significance but our bubble of excitement was soon burst when we found that neither of the FGFs is expressed in breast cancer cells, irrespective of amplification. Here was a proverbial fork in the road: should we continue working on FGFs and their receptors or should we follow the DNA amplification lead? For a while we did both but Clive gradually converted to mouse embryology and transgenics, whereas I became obsessed by the DNA. There had to be another gene on the amplicon that was important for cancer and all we had to do was find it. Simple! The plan was to identify new candidates on 11q13 by long-range mapping and to use additional probes to establish the minimum amplified region in our panel of breast cancers. This was 1991, and in one of the coincidences that make science so fascinating, six different groups identified the same gene in the same year but for different reasons. With the best of intentions, we found ourselves collaborating with two of them, and that the anonymous locus (D11S287) that we mapped on the 11q13 amplicon 3 was in fact cyclin D1 (Figure 2).

SPECIAL REVIEW

15

A bit of a l’ARF Although the evidence implicating p16INK4a in tumour suppression is pretty conclusive, the locus specifies a second protein, named ARF, whose amino terminal half is encoded by a unique exon (1β) while the carboxy-terminal half uses an alternative reading frame in the second exon of p16INK4a. There is no sequence similarity between the ARF and INK4 proteins but with curious echoes of the p16INK4a/pRb loop, ARF is able to cause a senescence-like arrest by activating the p53 pathway, and is negatively regulated by p53 7. It therefore has credentials as a tumour suppressor and in mouse models this is clearly the case. Indeed, in mouse cells, Arf seems to play a more prominent role in senescence than Ink4a.

ts8 SV

Se

nT

IG

3

T IG ng

IG 1

Yo u

nT Se

Yo u

3

p16 comes of age It quickly became clear that the p16INK4a- cyclinDCDK4-pRb pathway is among the most frequent targets of genomic alterations in human cancer, and that there is a striking inverse correlation between p16INK4a and pRb status. Although this is generally viewed as a defect in cell cycle control, p16INK4a is an extremely stable protein whose levels do not fluctuate appreciably in cycling cells. It was when we were trying to work out how pRb regulates the INK4a promoter, using primary fibroblasts and SV40 T-Ag, that Eiji Hara noticed that the most impressive change in p16INK4a levels occurred when the cells became senescent (Figure 3). We were not alone in making the connection, but I still rank Eiji’s modest paper as one of our most significant 4 .

T IG 1

Senescence is now recognised as a general response to aberrant proliferative stimuli, for example from an activated oncogene, and markers of senescence, including up-regulation of p16INK4a, are evident in premalignant lesions in vivo. Importantly, virtually all established human tumour cell lines have defects in either p16INK4a or pRb, which is why we do most of our work in primary cells. Many of the characteristics of senescence can be induced in early passage fibroblasts by ectopic expression of p16INK4a 5 and we have used this as a basis for functional tests of germline p16INK4a variants associated with melanoma. Through our contacts with clinical geneticists, we also gained access to fibroblasts from very rare patients that have germline mutations in both alleles of INK4a 6. We thought we’d been clever to corner this niche in the market but shRNA put paid to that.

ng

Figure 3 Accumulation of INK4a RNA in senescent and SV40 transformed fibroblasts. From [4].

The G1/S bandwagon From being the only people working on FGF3, we now found ourselves in the middle of a gold rush as families of cyclins, CDKs and CDK inhibitors were uncovered at breathtaking speed and the basics of the G1/S phase transition in mammalian cells took shape. The concept of cyclin D1 as a proto-oncogene gained wide acceptance but proved hard to demonstrate experimentally. In the event, its credentials were soon reinforced by signs that some CDK inhibitors act as tumour suppressors. The stand-out candidate was p16INKa, which binds directly to CDK4 and CDK6 and blocks their association with D-cyclins. Discovered by Beach and his colleagues in 1993, p16INK4a hit the headlines again the following year when it was identified by groups searching for a melanoma susceptibility gene on human chromosome 9p21.

– INKa

– γ-actin

16

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

It is widely assumed that ARF is also a tumour suppressor in human cells but the evidence remains circumstantial. All the testable functions of ARF, based on ectopic expression, can be attributed to the residues encoded by exon 1β, but the protein behaves like molecular Velcro. Despite being poorly conserved, it is reputed to interact with literally dozens of important cellular proteins and to modulate their activity in various ways, in some cases by escorting its clients to the nucleolus. Our data suggest a different scenario (Figure 4) and we suspect that ARF might be sequestered in the nucleolus to keep it out of harms way 8 . Unlike p16INK4a, we have never found point mutations that convincingly inactivate ARF and the patient-derived fibroblasts are specifically defective for INK4a. Interestingly, these cells become tumorigenic following expression of telomerase, MYC and RAS, reaffirming the importance of p16INK4a in tumour suppression 9. Gratifying though this was, we seemed to have come full circle to ponder the mysteries of oncogene cooperation again. Time for another re-think.

Figure 4 Induction of human ARF causes stabilisation of MDM2 and p53 without relocating them to the nucleolus. From [8].

IPTG:

–

+

p14ARF

p53

MDM2

Ending on a Hi-Seq We have a pretty good idea what p16INK4a can do but know a lot less about how its expression is regulated. From an evolutionary perspective, it seems bizarre that two (three including INK4b) tumour suppressors should end up side by side in the genome unless they are co-ordinately regulated in some contexts. The Polycomb group (PcG) proteins are obvious candidates for this level of control and mice lacking specific PcG proteins have defects that are attributable to derepression of INK4a. What we hadn’t bargained for is the huge number of PcG proteins that participate in the regulation of INK4a and other target genes but, as at every stage of my career, the inspirational environment and amazing resources at LIF have encouraged us to take on the challenge. It’s too soon to take stock of our recent efforts (see Molecular Oncology report pages 60-61) but these are exciting times and it will be frustrating to leave many tantalising questions unanswered.

p53

References

1. Peters G, Brookes S, Smith R, Dickson C. Tumorigenesis by mouse mammary tumor virus: evidence for a common region for provirus integration in mammary tumors. Cell. 1993; 33:369-377 2. Peters G, Lee AE, Dickson C. Concerted activation of two potential proto-oncogenes in carcinomas induced by mouse mammary tumour virus. Nature. 1986; 320:628-631 3. Lammie GA, Fantl V, Smith R, Schuuring E, Brookes S, Michalides R, Dickson C, Arnold A, Peters G. D11S287, a putative oncogene on chromosome 11q13, is amplified and expressed in squamous cell and mammary carcinomas and linked to BCL-1. Oncogene. 1991; 6:439-444 4. Hara E, Smith R, Parry D, Tahara H, Stone S, Peters G. Regulation of p16CDKN2 expression and its implications for cell immortalization and senescence. Mol Cell Biol. 1996; 16:859-867 5. McConnell BB, Starborg M, Brookes S, Peters G. Inhibitors of cyclin-dependent kinases induce features of replicative senescence in early passage human diploid fibroblasts. Curr Biol. 1998; 8:351-354



6. Brookes S, Rowe J, Ruas M, Llanos S, Clark PA, Lomax M, James MC, Vatcheva R, Bates S, Vousden KH, et al., INK4adeficient human diploid fibroblasts are resistant to RAS-induced senescence. EMBO J. 2002; 2:2936-2945 7. Stott FJ, Bates S, James MC, McConnell BB, Starborg M, Brookes S, Palmero I, Ryan K, Hara E, Vousden KH, et al., The alternative product from the human CDKN2A locus, p14(ARF), participates in a regulatory feedback loop with p53 and MDM2. EMBO J. 1998; 17:5001-5014 8. Llanos S, Clark PA, Rowe J, Peters G. Stabilization of p53 by p14ARF without relocation of MDM2 to the nucleolus. Nat Cell Biol. 2001; 3:445-452 9. Drayton S, Rowe J, Jones R, Vatcheva R, Cuthbert-Heavens D, Marshall J, Fried M, Peters G. Tumor suppressor p16INK4a determines sensitivity of human cells to transformation by cooperating cellular oncogenes. Cancer Cell. 2003; 4:301-310

SPECIAL REVIEW

17

RESEARCH HIGHLIGHTS

Every year the Report features significant research advances made by Institute scientists, summarising the findings in terms accessible to the nonspecialist scientific reader. The winner of the HardimanRedon Prize, awarded to a junior researcher who has made outstanding contributions to a highlighted publication during the year, is chosen from amongst these papers. The 2012 prize is awarded to Madhu Kumar, a postdoc with Julian Downward, for his work on the role of GATA2 in Ras-dependent tumourigenesis. The GATA2-driven oncogene network is requisite for RAS oncogene-driven no-small cell lung cancer. Kumar MS, Hancock D, Molina-Arcas M, Steckel M, East P, Diefenbacher M, Armenteros-Monterroso E, Lassailly F, Matthews N, Nye E, Stamp G, Behrens A, Downward J. Cell 2012; 149(3):642-55

Figure 1 Loss of GATA2 induces KRAS mutant lung tumour regression and can be recreated with clinically available drugs. A. When GATA2 is lost, primary KRAS mutant lung tumours undergo a marked tumour regression. B. Relative to placebo treated animals (PBS) and treatment with single agents bortezomib (BTZ) and fasudil (Faz), combined treatment with bortezomib and fasudil (BTZ+Faz) causes a marked clearance of KRAS mutant lung tumours.

A

B

Lung cancer is the most common cancer type worldwide, causing over one million deaths per year. Nearly a quarter of lung cancer patients have mutations in the KRAS oncogene. However, in spite of over thirty years of knowledge about KRAS, there is a lack of therapeutic options for this subset of cancers, representing a significant unmet clinical need. In this study, we followed up a screen for genes that are required for the progression of KRAS mutant lung cancers. We found that loss of the gene GATA2, a regulatory gene controlling the expression of other genes, caused the regression of KRAS mutant lung cancers. We then determined the pathways by which loss of GATA2 killed KRAS mutant tumours. Intriguingly, we found suppression of two GATA2-controlled pathways with clinically available drugs induced regression of KRAS mutant lung tumours. Thus, this 'GATA2 network' constitutes a new basis for therapeutic targeting of KRAS lung cancer with drugs already in clinical use (Figure 1). Asymmetric segregation of polarized antigen on B cell division shapes presentation capacity. Thaunat O, Granja AG, Barral P, Filby A, Montaner B, Collinson L, Martinez-Martin N, Harwood NE, Bruckbauer A, Batista FD. Science 2012; 335(6067):475-9 B cells are activated following the delivery of two temporally separated signals. The first signal is initiated by specific antigen binding to the B cell receptor (BCR) that leads to internalization, an accumulation in particular in the endosomal compartment. Subsequently, antigen within the endosomes is processed, loaded on major

18

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

c-Jun

p75NTR/DAPI

Figure 2 Confocal imaging of 7 days post-injury longitudinal sciatic nerve sections from wild-type mice(Junf/f ) and animals lacking c-Jun specifically in Schwann cells (JunΔSC). Double immunofluorescence for the SC progenitor marker p75 neurotrophin receptor (p75NTR; green, left panel) and c-Jun (red, middle pannel). Merge images are shown in the right panel. DNA (blue) was counterstained with DAPI (blue). Arrows indicate p75NTR+ SC progenitors that co-express c-Jun.

histocompatibility complex (MHC), and presented on the B cell surface. This antigen presentation allows for the recruitment of cognate CD4+ T helper cells and delivery of the second signal to allow maximal B cell activation. In this study we have investigated what happens to these antigencontaining compartments as activated B cells divide. We observed that the polarization of these compartments is maintained throughout cell division, such that antigen is asymmetrically segregated between daughter B cells. Intriguingly, the antigen inheritance correlates with the capacity for antigen presentation to T cells. On this basis, we have suggested that either the retention, or rapid elimination, of antigen might play an important role in the effective induction of humoral immune responses by influencing the fate of B cells. c-Jun in Schwann cells promotes axonal regeneration and motoneuron survival via paracrine signaling. Fontana X, Hristova M, Da Costa C, Patodia S, Thei L, Makwana M, Spencer-Dene B, Latouche M, Mirsky R, Jessen KR, Klein R, Raivich G, Behrens A. Journal of Cell Biology 2012; 198(1):127-41 Nerve injury can lead to debilitating movement impairment. In order to regain functionality, neurons not only have to survive the insult but also re-elongate their axons, often over very long distances, to achieve reconnection with their peripheral target muscles. That poses the question, what are the factors that make a neuron regrow? Schwann cells wrap an insulating myelin sheath around healthy peripheral axons and loss of Schwann cells, e.g. in multiple sclerosis, causes severe defects in neuronal function. We identified an additional role of Schwann cells in neuronal regeneration. The transcription factor c-Jun is strongly expressed in Schwann cells after injury (Figure 2), and absence of c-Jun in Schwann cells impaired neuronal survival and axonal regeneration. c-Jun functions by directing the synthesis of neurotrophic factors, which nurture



Merge

the injured neurons, and promote their regeneration. Our study illustrates that many cell types need to work together to repair nerve damage, highlighting the complexity of tissue regeneration (Figure 2). MMS19 links cytoplasmic Fe-S cluster assembly to DNA metabolism. Gari K, Leon Ortiz AM, Borel-Vannier V, Flynn H, Skehel JM, Boulton SJ. Science 2012; 337:243-5 The function of many DNA metabolism proteins depends on their ability to coordinate an ironsulfur (Fe-S) cluster. Biogenesis of Fe-S proteins is a multi-step process that takes place in mitochondria and the cytoplasm, but how it is linked to nuclear Fe-S proteins is not known. MMS19 is a highly conserved HEAT-repeat protein that impacts on many different cellular processes but how it functions has remained elusive. In this study, we establish that MMS19 functions as part of cytoplasmic Fe-S assembly machinery to facilitate Fe-S cluster transfer to target Fe-S proteins. In the absence of MMS19, failure to transfer Fe-S clusters to target proteins confers Fe-S protein instability, affecting many key Fe-S DNA metabolism enzymes. This study provided molecular insight to explain the previously reported phenotypes associated with MMS19 deficiency (such as DNA repair and proliferation defects) and why defects in mitochondrial Fe-S cluster biogenesis confer genome instability. RTEL1 dismantles T-loops and counteracts telomeric G4-DNA to maintain telomere integrity. Vannier JB, Petalcorin MIR, Pavicic-Kaltenbrunner V, Ding H, Boulton SJ. Cell 2012; 149:795-806 Telomeres are complex nucleo-protein structures, which function to maintain and protect the end of linear chromosomes. RTEL1 (Regulator of TElomere Length) is a helicase that is important for normal telomere length homeostasis but how it functions at telomeres was not known. In this study we establish that RTEL1 performs two distinct roles at

RESEARCH HIGHLIGHTS

19

Figure 3 RTEL1 unwinds T-loop structures to prevent catastrophic telomere processing by the SLX4 resolvasome. In RTEL1-/- cells, persistent T-loops are excised from chromosome ends by the SLX4 complex resulting in the formation of T-circles.

telomeres in T-loop and G4-DNA disassembly. Failure to dismantle T-loops in RTEL1-deficient cells resulted in rapid changes in telomere length and telomere loss as a result of catastrophic telomere processing by the SLX4 nuclease complex, which resolves the T-loop as a circle. RTEL1 also counteracts the formation of telomeric G4-DNA structures, which hinder DNA replication and are a major source of telomere fragility. Taken together, our findings define the mechanistic basis of T-loop disassembly and provide insight into the source and prevention of telomere fragility (Figure 3).

regulate telomerase, this study prompts a model for telomere length homeostasis. The study shows that Taz1, but not Rap1, is required for restriction of telomerase activity to specific cell cycle stages, and that rap1Δ telomere elongation is a consequence of insufficient telomeric coverage by Taz1. Moreover, the stalled telomeric replication forks conferred by loss of Taz1 attract telomerase regardless of telomere length. Hence, Taz1 appears to enforce the spatial and temporal regulation of telomerase by controlling the passage of replication forks through telomeres.

Taz1 enforces cell-cycle regulation of telomere synthesis. Dehé PM, Rog O, Ferreira MG, Greenwood J, Cooper JP. Molecular Cell 2012; 46(6):797-808

F-actin is an evolutionarily-conserved damageassociated molecular pattern recognized by DNGR-1, a receptor for dead cells. Ahrens S, Zelenay S, Sancho D, Hanč P, Kjær S, Feest C, Fletcher G, Durkin C, Postigo A, Skehel B, Batista FD, Thompson B, Way M, Reis e Sousa C, Schulz O. Immunity 2012; 36(4):635-45

The loss of a single telomere triggers cell cycle arrest in checkpoint-competent cells, or catastrophic genome destabilization in checkpoint-mutated cancer cells. Hence, cells ensure that telomerase acts preferentially on the shortest telomeres, thus preventing sporadic telomere loss. However, the mechanisms governing this telomere length homeostasis remain unclear. By examining how the highly conserved telomere proteins Taz1 and Rap1

20

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

The immune system is capable of responding to tissue injury by recognising exposure of intracellular molecules, referred to as damageassociated molecular patterns (DAMPs). DAMPs are exposed when cells lose their membrane integrity during cell death and are thought to be detected by receptors on innate immune cells such as dendritic

Cell death leads to the exposure of endogenous molecules, which are thought to activate dendritic cells and thereby promote adaptive immunity against antigens present in dead cells. DNGR-1 is a dendritic cell-restricted receptor that recognises filamentous actin exposed in dead cells and contributes to the cytotoxic T cell response against dead cell-associated antigens. Contrary to our expectations, DNGR-1 does not mediate activation of dendritic cells by dead cells. Rather, it controls endocytic handling of necrotic cell material to favour antigen extraction from the corpses. Consequently, DNGR-1 contributes to priming of cytotoxic T cell responses in both non-infectious settings and during cytopathic virus infection. The existence of a dedicated receptor for crosspresentation of dead cell-associated antigens and its demonstrable impact on antiviral responses underscores the importance of dead cell sensing in immunity in both sterile and non-sterile conditions.

Figure 4 In vitro polymerised actin filaments stained with an extracellular domain of DNGR-1 (green).

cells (DCs). We have previously identified DNGR-1 as a DC-specific receptor for dead cells that recognises an unknown DAMP. In this study, we showed that the DNGR-1-ligand is filamentous actin (F-actin), an evolutionarily conserved cytoskeletal component of all eucaryotic cells. Given its evolutionary conservation, abundance and stability, F-actin may play a more general role, beyond DNGR-1 recognition, in alerting the immune system to tissue damage (Figure 4). The dendritic cell receptor DNGR-1 controls endocytic handling of necrotic cell antigens to favor cross-priming of CTLs in virus-infected mice. Zelenay S, Keller AM, Whitney PG, Schraml BU, Deddouche S, Rogers NC, Schulz O, Sancho D, Reis e Sousa C. Journal of Clinical Investigations 2012; 122(5):1615-27

Centralspindlin links the mitotic spindle to the plasma membrane during cytokinesis. Lekomtsev S, Su KC, Pye VE, Blight K, Sundaramoorthy S, Takaki T, Collinson LM, Cherepanov P, Divecha N, Petronczki M. Nature 2012; 492(7428):276-9 Cell division underpins human development, regeneration and reproduction. Errors during cell division cause genetic imbalances that can fuel tumour development and progression. For successful cell division two key events have to be tightly coordinated: (1) the partitioning of the genetic material (chromosome segregation) and (2) the splitting of the cell envelope and consequently the entire cell volume (cytokinesis). To achieve this, animal cells use the same

Figure 5 A pair of human daughter cells just prior to the final cut. Centralspindlin is labeled in red, the mitotic spindle in black and daughter nuclei (DNA) in cyan.



RESEARCH HIGHLIGHTS

21

structure. The structure revealed that Dbf4 wraps around Cdc7, stabilizing the enzyme in the active conformation. To help cancer drug developers, we co-crystallized the kinase with a range of known Cdc7 small molecule inhibitors. Our structures elucidated the interactions these small molecules make in the Cdc7 active site and will be invaluable in the development more specific and potent inhibitors of the S-phase kinase (Figure 6).

Figure 6 The crystal structure of human Cdc7 kinase.

EBs recognize a nucleotide-dependent structural cap at growing microtubule ends. Maurer SP, Fourniol FJ, Bohner G, Moores CA, Surrey T. Cell 2012; 149(2):371-82

structure, the mitotic spindle, to segregate sister genomes to opposite sides and to define the centre of the cell for cytokinesis. How the mitotic spindle is connected to the cell envelope to control the final cut during cytokinesis was unknown. We found that a protein complex called centralspindlin links the centre of the mitotic spindle to the cell envelope (Figure 5). This link is essential for the successful division of human cells by allowing the mitotic spindle to hold on to the cell envelope until the final cut. Crystal structure of human CDC7 kinase in complex with its activator DBF4. Hughes S, Elustondo F, Di Fonzo A, Leroux FG, Wong AC, Snijders AP, Matthews SJ, Cherepanov P. Nature Structural and Molecular Biology 2012; 19(11):1101-7 Proliferation of human cells requires duplication of the 3,200-Mbp (mega base pairs) genome during each round of cell division. This colossal task depends on an army of molecular machines to start and complete DNA synthesis in a precise and timely fashion. Cdc7 kinase is one of the key licensing factors acting during initiation of DNA synthesis. Together with S-phase cyclin-dependent kinases (CDKs), Cdc7 marshals a cascade of protein-protein interactions leading to the assembly of the DNA replication machinery. Due to its pivotal roles in cell proliferation, Cdc7 is an emerging target for the development of cancer therapeutics, and structural information is urgently needed to aid in optimization of small molecule inhibitors of this kinase. We were able to grow crystals of human Cdc7 bound to its activator protein Dbf4 and determined its three-dimensional

22

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

In living cells of many organisms such as animals, plants and fungi, filaments called microtubules form a scaffold for the interior organisation of the cell. Through dynamic reorganization microtubules also build the machinery required to segregate the genetic information during cell division. The cell controls microtubule organization with the help of a special class of proteins which bind selectively to the ends of growing microtubules and many of these microtubule end-tracking proteins are mutated in cancer cells. EB1 is an especially important protein, because it recruits almost all other end tracking proteins. In collaboration with Carolyn Moores, Birkbeck College in London, we have discovered where exactly on the microtubule surface EB1 binds. We discovered that EB1 binds to a particular microtubule structure that is vital for their dynamic properties. This finding contributes to a better understanding of how end tracking proteins work in the cell and how regulators of microtubule dynamics can act at the molecular level (Figure 7). DBIRD complex integrates alternative mRNA splicing with RNA polymerase II transcript elongation. Close P, East P, Dirac-Svejstrup AB, Hartmann H, Heron M, Maslen S, Chariot A, Söding J, Skehel M, Svejstrup JQ. Nature 2012; 484:386-9 The interactome of chromatin-associated mRNP particles was characterized by protein purification and mass spectrometric analysis. This led to the identification of a novel protein complex, named DBIRD, which binds directly to RNA polymerase II (RNAPII). DBIRD regulates alternative splicing of a large set of exons embedded in A/T-rich DNA, and is present at the affected exons. RNAi-mediated DBIRD depletion results in region-specific decreases in transcript elongation speeds, particularly across areas encompassing affected exons. Together, these data indicate that DBIRD complex acts at the interface between mRNP particles and RNAPII, integrating transcript elongation with the regulation of alternative splicing.

Figure 7 Reconstruction of a microtubule from cryo-electron microscopy images. Left: the microtubule binding domain of fission yeast EB1 is shown in green. Right: Pseudoatomic model showing 4 tubulin dimers (blue and purple) and fission yeast EB1 (green).

RNA Polymerase II collision interrupts convergent transcription. Hobson D, Wei W, Steinmetz LM, Svejstrup JQ. Molecular Cell 2012; 48(3):365-74 Antisense transcription is prevalent in eukaryotic genes, so what happens when RNA polymerase II (RNAPII) molecules collide head-to-head? This paper shows that polymerases transcribing opposite DNA strands cannot bypass each other. RNAPII stops, but does not dissociate upon collision both in vitro and in vivo, but removal of collided RNAPII from the DNA template can be achieved via ubiquitylation-directed proteolysis. Indeed, in cells lacking efficient RNAPII polyubiquitylation, the half-life of collided polymerases increases, so that they can be detected between convergent genes. These results provide new insight into fundamental mechanisms of gene traffic control, and point to an unexplored effect of antisense transcription on gene regulation via polymerase collision (Figure 8).

Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C. New England Journal of Medicine 2012; 366(10):883-92 This work demonstrates the evolution of two renal cancers between the primary and metastatic sites, demonstrating diversity in the somatic mutational spectrum, ploidy profiles and DNA copy number between biopsies from the same tumour. The extent of intratumour heterogeneity is demonstrated for several “convergent” yet distinct somatic loss of function events in SETD2, PTEN and KDM5C in different regions of the same tumour.

Figure 8 Left: Nucleic acid-centric view of convergent transcription. Right: Protein-centric view of elongation, showing a crystallographic model of convergently transcribing RNAPII elongation complexes (DNA in green/blue; RNA in red).

Polymerase bypass



Polymerase collision and arrest (?)

RESEARCH HIGHLIGHTS

23

A

C

R3

R2 R4

R5

R9

R1 R2 R1 R2 R9 R9 R8 R8R3R3 R5R5 KDM5C (missense & frameshi) KDM5C (missense & frameshi) mTOR (missense ) mTOR (missense )

R2 R4

Hilum

R1

Hilum

Figure 9 A: Multiple tumour regions were biopsied from a primary renal cell carcinoma (R1-R9) and R1from two associated metastatic sites (M1, M2a+b; not shown) from patient 1. R3 of B: Regional distribution non-synonymous mutations (red indicates the presence, R5 blue the absence of a mutation in a region; gene names in green: mutation was validated and in red: validation of R9 the mutation failed). C: Phylogenetic reconstruction of the history of this tumour shows branched evolution and convergence (multiple independent R7 SETD2/KDM5C mutations).

SETD2 (splice site)

NL

NL

SETD2 (splice site) R4a

R6 Ubiquitous

R6 R8

R7 10 R8 cm

R4b

R4b

SETD2 (frameshi)

SETD2 (frameshi)

Ubiquitous

Shared Primary

VHL

Shared Metastasis

Shared Primary

Private

Shared Metastasis Private

R4a

VHL SETD2 (missense) KDM5C (splice site)

SETD2 (missense) KDM5C (splice site)

M1 M2b

M2a

M1 M2b

10 cm

M2a

B Shared Primary

Shared Metastasis

Private

IL12RB2 BCAS2 IFI16 FCAMR PLB1 ALS2CR12 C2orf21 VHL SGOL1 KLHL18 SSR3 CLCN2 WHSC1 ATXN1 DOPEY1 CCR6 INTS1 PTPRZ1 ZC3HC1 EXT1 RALGDS MSRB2 EIF4G2 ANO5 C11orf68 MRPL51 KDM2B TOX4 NUSAP1 TCF12 ZC3H18 DDX52 ZNF519 AKAP8 CYP4F3 KIAA0355 WDR62 KLK4 IGLON5 NLRP7 MAGEB16 SESN2 CCBL2 SETD2 PLRG1 CASP2 SSNA1 TH PPFIA1 CDKN1B WSCD2 ZNF780A PPP6R2 MTOR UGT2A1 ABHD11 GALNT11 RIMBP2 PSMD7 CENPN SOX9 NPHS1 RBFOX2 KDM5C KDM5C SATL1 FLNA ITGB3 LATS2 DIRAS3 NGEF ZNF493 SPATA21 DDX58 DAPK1 ALKBH8 KL ERCC5 DIO1 PIAS3 MR1 C3orf20 SETD2 TNIK LIAS FBXO1 AKAP9 ITIH5 WDR24 MYH8 TOM1 SBF1 KDM5C USP51 NAP1L3 ADAMTSL4 DUSP12 SLC2A12 RAB27A CIB2 RPS8 FAM129B PHF21B HDAC6 MAP3K6 MAMLD1 RLF DNMT3A HMG20A ZNF521 MMAB DACH2 SLC2A1 TM7SF4 ANKRD26 CD44 KRT4 KIAA1267 C3 ADAMTS10 IFNAR1 BCL11A PLCL1 SETD2 KIAA1524 NRAP HPS5 DIXDC1 LAMA3 CDH19 SUPT6H WDR7 C2orf85

Ubiquitous

M2b M2a M1 R4 R9 R8 R5 R3 R2 R1

The role of intratumour heterogeneity in the acquisition of drug resistance is demonstrated for a heterogeneous mutation near the mTOR kinase active site that results in constitutive activation of the mTOR kinase, only in regions of the primary tumour that harbour this mutation. These data suggest that tumour genomics analyses to stratify patients for distinct targeted approaches will need to consider the clonal dominance of the target as well as the potential regional distribution of tumour sub clones between sites of disease. Intratumour heterogeneity may also contribute to the emergence of resistant sub clones over time and therapeutic failure (Figure 9). Salt-inducible kinases regulate growth through the Hippo signalling pathway in Drosophila. Wehr MC, Holder MV, Gailite I, Saunders RE, Maile TM, Ciirdaeva E, Instrell R, Jiang M, Howell M, Rossner MJ, Tapon N. Nature Cell Biology 2012; doi:10.1038/ncb2658 In order to prevent overgrowth and cancer formation, cell growth and proliferation must be kept in check throughout our lives. Using the development of fruit flies as a model system has led to the discovery of the Hippo pathway, a key repressor of tissue growth which is affected in a variety of human tumours. Much research has gone into figuring out the signals that ensure the adequate levels of Hippo signalling, leading to the idea that appropriate tissue architecture prevents runaway proliferation via the Hippo pathway,

24

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

though this is by no means the complete story. In this paper, we use a clever new technique called the Split-TEV system to screen thousands of genes for new Hippo pathway regulators. This leads us to the Sik (Salt-inducible kinases), which are part of a family of nutrient/energy sensors. Our work therefore suggests that the Hippo pathway can respond to metabolic cues as well as to tissue architecture (Figure 10). Positive feedback and mutual antagonism combine to polarize Crumbs in the Drosophila follicle cell epithelium. Fletcher GC, Lucas EP, Brain R, Tournier A, Thompson BJ. Current Biology 2012; 22(12):1116-22 Most human cancers arise from the ‘epithelial tissues’ of the body, those that form layers or tubes such as the skin, intestine and lung bronchi. Cancer cells multiply more than normal cells, but only become highly dangerous when they lose their normal epithelial form. For example, a normal epithelial cell has a columnar shape, while malignant cancer cells have a more disorganised shape and can escape from the epithelium to invade other tissues. The normal shape of epithelial cells depends on a molecular orientation system that tells each cell which way is ‘up’ and ‘down’ relative to the other cells in the tissue. When this system is broken, an epithelial no longer knows up from down and so cannot organise itself into its normal epithelial shape. We have studied the molecules that define up and

Figure 10 Confocal micrograph of a Drosophila wing imaginal disc stained for the apical protein Expanded in red and DNA in blue. The GFP positive areas in green are depleted for the Hippo kinase, which induces Expanded expression.

down in epithelial cells and found that ‘up’ molecules and ‘down’ molecules separate according to a simple set of principles (Figure 11). TBC1D14 regulates autophagosome formation via Rab11- and ULK1-positive recycling endosomes. Longatti A, Lamb CA, Razi M, Yoshimura S, Barr FA, Tooze SA. Journal of Cell Biology 2012; 197(5):659-75 Autophagy, or self-eating, is essential for cell homeostasis, development, and protects against infection and many human diseases, in particular neurodegeneration and cancer. Autophagosomes, which are specialized membrane vesicles, capture and target cellular material for degradation, thus enabling autophagy. To exploit autophagy as a way to prevent disease we must understand how these autophagosomes form. Small GTPases of the Rab Figure 11 Columnar epithelial cells from Drosophila ovarian follicles.

family are required for all vesicle formation, so in this study we systematically inactivated all Rab proteins by overexpression of all predicted GTPase activating proteins, or RabGAPs. Our work revealed multiple RabGAPs are required, but one, TBC1D14, co-localized with the essential autophagy kinase ULK1. TBC1D14 together with Rab11 to controls vesicles from the recycling endosome, regulating growth factor trafficking and signalling. Importantly, ULK1-positive TBC1D14-containing vesicles are delivered to the forming autophagosome from the recycling endosome, potentially linking the regulation of nutrient and growth factor uptake with autophagy. Clathrin potentiates vaccinia-induced actin polymerisation to facilitate viral spread. Humphries AC, Dodding MP, Barry DJ, Collinson LM, Durkin CH, Way M. Cell Host and Microbe 2012; 12:346-359 Like many pathogens, vaccinia virus hijacks the actin cytoskeleton of its host to assist the spread of infection. Our recent work has provided further insight into how the virus uses actin at the plasma membrane to enhance its spread. We found that clathrin, a molecule that is well known to play a role in recycling and trafficking of proteins, is recruited to the virus as it leaves the cell. Clathrin is known to play a role in viral entry, however, this is the first time clathrin has been shown to promote the spread of a virus. We found that clathrin clusters viral proteins within the plasma membrane underneath the virus. This creates an ordered signalling platform that makes viral induced actin polymerization more efficient. Our work has highlighted an unappreciated role for clathrin, which is emerging as a more adaptable and versatile molecule than traditionally thought.



RESEARCH HIGHLIGHTS

25

Wing imaginal discs carrying double FLP/FRT clones (GFP in green, mCherry in red). Image: Nic Tapon, Apoptosis and Proliferation Control Group

26

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

LINCOLN’S INN FIELDS

The London Research Institute, Lincoln’s Inn Fields research laboratories are located in the centre of London. There are 32 research groups working within the broad research themes of cellular regulatory mechanisms, biology of tumours and tissues, and genomic integrity and cell cycle. By carrying out basic research, the Institute will continue to increase the understanding of cancer biology. The researchers are supported by an excellent range of Technology Core Facilities.



LINCOLN’S INN FIELDS

27

LYMPHOCYTE INTERACTION www.london-research-institute.org.uk/research/facundo-batista

Group Leader

Facundo D Batista Postdoctoral Scientists Francesca Gasparrini Selina Keppler Nuria Martínez-Martín Pieta Mattila Graduate Students Marianne Burbage Angelo Costello Christoph Feest Mauro Gaya Scientific Officers Carol Fong (BliNK Therapeutics Ltd.) Naomi Hardwood Beatriz Montaner Irene Sanjuan-Nandin (BliNK Therapeutics Ltd.) Microscope Developer Andreas Bruckbauer Visiting Workers Patricia Barral Maria Dolores Sánchez-Niño

By making antibodies, B cells occupy a vital part of our immune system and protect us from a variety of different pathogens, such as bacterial viruses and cancer. Efficient B cell responses are characterised by the fast production of an initial protective antibody wave, but subsequently by the generation of B cell memory. This refers to good B cells being selected and kept alive for a very long period of time, allowing protective antibodies to be produced throughout and individual’s life. However, this is not without risk and consequently, the steps leading to B cell activation must be stringently controlled to prevent the stimulation of bad B cells, which can in turn lead to autoimmune diseases. Research in the Lymphocyte Interaction Group endeavours to drive a broad and comprehensive understanding of the cellular and molecular events that lead to B cells becoming activated. We address this by combining the power of genetics with biochemistry and advanced imaging technology, both at high-resolution and in vivo. B cells segregate antigen asymmetrically on division B cell activation is initiated when the B cell receptor (BCR) specifically binds antigen. This is know as ‘signal 1’ and generally occurs in vivo on the surface of antigen presenting cells. Binding of the antigen by the BCR leads to the generation of intracellular signalling, which is followed by antigen acquisition, internalisation, degradation and presentation to T cell. During this cognate interaction, T cells give B cells what is referred as ‘signal 2’, providing them with the remaining key signals which enable them to become fully activated. Recently, we reported a novel role for macrophages (which are located in the subcapsular sinus area of the lymph nodes) in presenting antigen to B cells in vivo, in other words providing them with signal 1 (Carrasco and Batista, 2007; Immunity. 27: 160-71). One striking feature we noticed is that during the interaction the antigen appears to be polarised to the site of contact. Following up on this observation, we recently reported (Thaunat et al., 2012; Science. 335: 475-9) that after antigen extraction from these

28

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

macrophages, B cells are able to retain antigen in a polarised way and for prolonged periods of time. Interestingly, we also showed that antigen polarisation is maintained during and after B cell division, suggesting that during division B cells can generate different daughter cells as a result of asymmetric cell division, in which each daughter receives a different load of antigen. To investigate this, we developed an assay to simultaneously track antigen and B cell division and observed the rapid emergence of a population containing little or no antigen, as well as the maintenance of B cells with high levels of antigen. By comparing this experimental data with mathematical models, we established that the asymmetric segregation of antigen does indeed occur on B cell division. Indeed, the polarization of acquired antigen into a small number of intracellular compartments would be expected to promote the asymmetric segregation of antigen. But what is the functional significance of asymmetric antigen segregation to dividing B cells? We went on to show that the amount of antigen present in the B cell correlates with the

Figure 1 This Super-resolution image shows the B cell receptor exists in nano-sized clusters within the plasma membrane. It is reconstructed from many localisations of single dye molecules which are stochastically activated and then imaged one by one. This new dStorm microscopy technique was successfully implemented in our lab to study changes in receptor organisation during B cell activation.

capacity to present antigen in complex with MHC and recruit the T cell help required for activation. On the basis of these findings, we postulate that B cells receiving a greater antigen inheritance following division are more able to compete for limited T cell help, facilitating somatic hypermutation and class switch. In addition, those daughters receiving little or no antigen may promote affinity maturation, as these cells have diminished capacity for presentation and their survival depends entirely on the acquisition of new antigen. In this way, those B cells expressing newly mutated BCR with a greater affinity for antigen will be selected, and retained, for further rounds of selection. As such, this unequal partitioning of antigen may be critical for the induction of effective humoral immune responses. The actin cytoskeleton and microtubule network participate in B cell activation Another active part of research in the lab has been the focus on trying to understand the process of B cell activation at high molecular resolution. One principle emerging from our work is that the global distribution and organisation of the BCR in the B cell membrane, determines its capacity to mediate signalling and consequent B cell activation (Harwood and Batista, 2010: Annu Rev Immunol. 28: 185-210). In contrast to the classical fluid mosaic model, the current view of the plasma membrane is that membrane lipids and proteins are not generally free to diffuse laterally in the plane of the membrane. Previously, we have tracked single particles of BCR and have shown that BCR diffusion is indeed restricted by an ezrin-defined actin network (Treanor et al., 2010: Immunity. 32: 187-99; Treanor et al., 2011: J Exp Med. 208: 1055-68). Recently, in consultation with



our resident microscopy developer, Andreas Bruckbauer, we have been successful in implementing super-resolution microscopy methods, including Photo-Activated Localisation Microscopy (PALM) and direct Stochastic Optical Reconstruction Microscopy (dSTORM). This new type of methodology, combined with novel imaging analysis, has allowed us to visualise and establish that on the surface of naïve B cells the BCR already exists in a pre-clustered state, previous to any activation. Surprisingly, we observed that disruption of this cytoskeleton network using pharmacological reagents triggers a robust activatory signal without dramatically affecting the overall receptor distribution on the cell surface. In line with this, we discovered that this signal, not only requires BCR but also the positive regulator CD19. This critical observation clearly demonstrates that the increased likelihood of BCR collisions following release from cytoskeleton constraints is insufficient to mediate B cell signalling. Instead, these results reveal a novel, previously unappreciated, layer of complexity in the organisation of the B cell membrane, involving the regulated segregation of BCR and of co-receptors, such as CD19. These findings also conceptually challenge the general believe that B cell activation is simply the result of BCR ligation or clustering. This leads us to suggest an alternative current working hypothesis, whereby the removal of cytoskeleton constraints allows the reorganisation of compartments in the plasma membrane, such that BCR interactions with positive co-receptors are facilitated while probably interactions with negative co-receptors could be prevented. Publications listed on page 136

LINCOLN’S INN FIELDS

29

MAMMALIAN GENETICS www.london-research-institute.org.uk/research/axel-behrens

Group Leader

Axel Behrens Postdoctoral Scientists Sophia Blake Atanu Chakraborty Catherine Cremona Markus Diefenbacher Xavier Fontana Ralph Gruber Omar Khan Rocio Sancho Graduate Students Rute Ferreira Hendrik Messal Kay Penicud Tianyi Zhang Scientific Officers Clive Da Costa Nnennaya Kanu

30

Every organ harbours adult stem cells which have the potential for long-term replication, together with the capacities of self-renewal and multi-lineage differentiation. These stem cells function in tissue homeostasis and contribute to regeneration in response to injury. In addition, many cancers are caused by transforming mutations occurring in tissue-specific progenitor cells. Our major focus is to elucidate the molecular mechanisms governing stem cells function and cancer. Fbw7 repression by Hes proteins creates a feedback loop that modulates Notch-mediated stem cell fate decisions Notch signalling controls cell differentiation decisions in a wide range of species, primarily via a process of lateral inhibition. Notch activation at the cell surface stimulates an exchange of signals that repress Notch ligand production, amplifying initial differences in Notch activation levels in neighbouring cells and resulting in unequal cell differentiation decisions. Imbalances in this feedback system result in alterations in cell fate, affecting tissue function. We found that the downstream Notch signalling effector HES5 directly represses transcription of a crucial component of an SCF-type E3 ubiquitin ligase, Fbw7β. Fbw7 mediates degradation of an array of target proteins, including Notch, and therefore reduced Fbw7 expression results in increased Notch signalling. This creates a positive feedback loop that controls Notch levels in cells. Fbw7Δ/+ heterozygous mice showed haploinsufficiency for Notch degradation, causing impaired intestinal progenitor cell and neural stem cell differentiation. Notably, concomitant inactivation of HES5 rescued both phenotypes and restored normal stem cell differentiation potential. In silico modelling suggests that the Notch/ HES5/ Fbw7β positive feedback loop is responsible for the Fbw7 haploinsufficiency phenotype and its rescue by HES5 inactivation. Thus repression of Fbw7β transcription by Notch signalling is an essential mechanism that is coupled to and required for the correct specification of cell fates induced by lateral inhibition.

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

RACO1 defines the molecular pathway that links growth factor signalling to c-Jun/AP-1 activation The AP-1 transcription factor is a heterodimeric complex of various Jun, Fos and ATF-2 family members and mediates diverse cellular responses ranging from cell proliferation and differentiation to tumourigenesis and apoptosis. c-Jun is essential for cell proliferation and transformation, by controlling cell cycle regulator genes including cyclin D1 and cdc2. c-Jun null fibroblasts display severe proliferation defects, impaired cell cycle re-entry after serum withdrawal and inability to undergo transformation by oncogenic Ras. However, the molecular connection between c-Jun and growth factor signalling was enigmatic. We have recently described a novel c-Jun coactivator, RING domain AP-1 co-activator 1 (RACO1), that links growth factor/ oncogenic Ras signalling to AP-1 activation. Ras activation stimulates RACO1 function by increasing protein stability. Mechanistically, RACO1 stability is controlled by the competition of degradative K48- and non-degradative K63-linked ubiquitination. Upon activation of the Ras-MEK-ERK pathway, K63-linked ubiquitin chains are attached to the same residues targeted for degradative K48-linked ubiquitination, thereby resulting in enhanced protein levels and consequent upregulation of c-Jun target genes. This work identified RACO1 as the much soughtafter missing link between Ras signalling and AP-1 activation.

Canonical signal (e.g. IR)

Figure 1 Competition for ATM binding controls ATM signalling. In wild-type cells, NBS1 and ATMIN compete for ATM binding and both proteins contribute to ATM substrate phosphorylation. In NBS1-deficient cells, ATM signalling in response to irradiation is greatly reduced, but some ATMINdependent non-canonical ATM signalling is increased. In ATMINdeficient cells, NBS1 binding to ATM and ATM substrate phosphorylation after IR is increased. In double mutant cells, ATM signalling is abrogated.

Competition between NBS1 and ATMIN controls ATM signalling pathway choice The protein kinase mutated in ataxia telangiectasia, ATM, is activated by several distinct stimuli, and these require different cofactors for signalling. Canonical ATM activation by DNA double-strand breaks requires the MRN complex component NBS1, whereas signalling induced by changes in chromatin structure requires the ATM INteractor protein ATMIN. However, the mechanism by which different cofactors decide ATM’s signalling response is not known. Our research has now shown that NBS1 and ATMIN proteins compete for ATM binding, and that this mechanism controls ATM function. Double-strand break-induced ATM substrate phosphorylation via NBS1 was increased in atmin-mutant cells. Conversely, NBS1 deficiency resulted in increased ATMIN-dependent ATM signalling. Thus absence of one cofactor increased flux through the alternative pathway. NBS1 and ATMIN bind ATM via similar motifs, and excess ATMIN impaired double-strand break-induced ATM activation and MBS1/ATM interaction, suggesting that the proteins compete for ATM binding directly. NBS1/ATMIN double deficiency



Non-canonical signal (e.g. hypotonic stress)

resulted in complete abrogation of ATM signalling and increased radiosensitivity compared to nbs1 mutants, suggesting that ATMIN mediates some ATM signalling from double-strand breaks in the absence of NBS1. Despite the profoundly radiosensitive phenotype of the double mutant, ATMIN deficiency rescued the cellular lethality of NBS1-deficient cells, and partly restored intestinal crypt cell proliferation and villus structure. Hence ATMIN and NBS1 mediate all ATM signalling by double strand breaks, and increased ATMINdependent ATM signalling explains the different phenotypes of nbs1- and atm-mutant cells. Thus antagonism and redundancy of ATMIN and NBS1 is a crucial regulatory mechanism for ATM signalling and function.

Publications listed on page 136

LINCOLN’S INN FIELDS

31

HAEMATOPOIETIC STEM CELL www.london-research-institute.org.uk/research/dominique-bonnet

Group Leader

Dominique Bonnet Clinical Scientist Yasmin Reyal Postdoctoral Scientists Alessandro Di Tullio Katie Foster Arnaud Gandillet Ashley Hamilton Lourdes Lopez-Onieva Kevin Rouault-Pierre Graduate Students Alessandra Audia Amy Bradburn Scientific Officers Linda Ariza-McNaughton Erin Currie Fernando dos Anjos Afonso Jashu Patel Visiting Workers Irene Pizzitola David Taussig

The group is interested in studying human normal haematopoietic stem cells (HSCs) and leukaemic stem cells (LSCs). Over the past few years, it has become clear that both normal HSCs and LSCs are more heterogeneous than previously thought. Our aim now is to better characterise the differences between the different HSCs or LSCs. The lab is also involved in studying the role of the microenvironment. For that we have developed in vivo imaging techniques allowing us to visualise and define the normal and leukaemic stem cells niche in vivo. We hope to dissect the role of different components of the stem cell niche and whether we can intervene in this niche to expand normal stem cells and eradicate leukaemic stem cells (LSCs). All these projects should shed light on pathways or interactions that are more specifically used by LSC and where therapeutic intervention might be developed.

Leukaemic stem cell Recently we and others have defined a new heterogeneity at the LSC level (Taussig et al., 2008; Blood. 112: 568-75; Taussig et al., 2010; Blood. 115: 1976-84; Sarry et al., 2011; J Clin Invest. 121: 384-95; Eppert et al., 2011; Nat Med. 17: 1086-93; Goardon et al., 2011; Cancer Cell. 19: 138-52). Indeed it is becoming clear now that LSCs in acute myeloid leukaemia (AML) are not restricted to CD34+CD38- cells but can harbour different phenotype CD34+CD38+, CD34-CD38+ or CD34-CD38-. The phenotype of the LSCs varied from patient to patient and even in the same patient more than one fraction of LSCs can be identified at one time-point. Thus, for each AML patient studied we could not rely on the phenotype to define LSC but have to perform functional analysis (i.e. xenotransplantation assay). What is unclear is the relationship between the different LSC fractions and how these different clones of LSCs variegate over time. To answer these different questions, we are now evaluating the genetic variegation of the LSC compartment over time as well as looking at the difference in chemoresistance between the different clones of LSCs.

32

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

CEBPA mutant alters human HSC C/EBPa (CEBPA) is mutated in approximately 8% of AML in both familial and sporadic AML and, with FLT3 and NPM1, and has received the most attention as a predictive marker of outcome in patients with normal karyotype disease. Mutations clustering to either the N- or C-terminal (N-and C-ter) portions of the protein have different consequences on the protein function. In familial cases the N-ter form is inherited with patients exhibiting long latency period before the onset of overt disease, typically with the acquisition of a C-ter mutation. Despite the essential insights murine models provide, the functional consequences of wild-type C/EBPa in human haematopoiesis and how different mutations are involved in AML development have received less attention. Our recent data underlines the critical role of C/EBPa in human haematopoiesis and demonstrates that C/EBPa mutations (alone or in combination) are insufficient to convert normal human hematopoietic stem/progenitors cells (HSC/HPCs) into leukemic initiating cells, although individually each altered normal haematopoiesis. It provides the first insight into the effects of N- and C-ter mutations acting alone and to the combined

Figure 1 A. Transversal section of a tibia showing in purple the bone and in red vessels. This shows the complex and dense vascularisation in long bone even in the zone close to the bone (endosteal region) thought to be devoid of vessels. B–C. 3D tracking of human leukaemia (HL60 cell line) in live bone marrow.

A

B

C

effects of N/C double mutants. Our results mimicked closely what happens in CEBPA mutated patients. One question, which is still open, is how many mutations do exist in these C/EBPa mutated AML patients and how many are needed to trigger leukaemia (Quintana-Bustamante et al., 2012; Leukemia. 26: 1537-46). Cross-talk between LSCs and their microenvironment It has become clear both in solid tumours and in leukaemia that cancer stem cells depend on their



microenvironment to grow and expand. In AML, we know that LSCs cannot be maintained ex vivo without the addition of a stroma support indicating that LSCs are dependent on their microenvironment for their survival/maintenance. Understanding the crosstalk between LSC and their microenvironment is thus crucial to better understand this dependency and potentially use this to target LSC in vivo. We thus started a project trying to better define the factors involved in this crosstalk. AML samples were co-cultured ex vivo with mesenchymal stroma cells (MSC) and after one week, micro-array analysis was performed on sorted stroma cells. Using pathway analysis and the Gene Go program, we built a network of the combined datasets. Further studies will evaluate the effect of the different key factors. If any of these factors affect the growth/differentiation/ apoptosis of AML, we will further investigate one/ more pathway(s) in more detail in vitro but also in vivo using knock-down approaches. Stem cell niche HSCs reside in specialized bone marrow (BM) environments called stem cell niches (HSCNs). Three types of HSCNs have been reported so far, involving osteoblasts (osteoblastic niche), blood sinusoids (vascular niche) and mesenchymal stem cells. However the precise localization, composition and regulation of the niche(s) remains highly controversial. In order to better characterize bone marrow microenvironment, stem cells and their niches, we have developed different technologies for in vivo contrasting procedures as well as for tracking normal and leukemic cells in vivo, combining whole body near infrared fluorescence, bioluminescence imaging, intravital microscopy of intact live bone marrow as well as histology and flow cytometry. Using improved contrasting procedures for the BM endothelium we demonstrated that osteoblastic niches are in close proximity to vascular niches in flat but also in long bones. We believe that the combined used of advanced multimodal and multiscale analysis of the bone marrow will very likely contribute to shed new lights on our understanding of hematopoietic stem cells and their niches in health and disease (Lassailly et al., 2010; Blood. 115: 5347-54; Lassailly et al., In revision). Using this technique and adding time-lapse imaging, we are visualizing the arrival of human HSCs in the bone marrow and following their behaviour over time. This should allow us to better define the niche and see whether LSCs used the same niche as normal HSCs (Foster, In Preparation).

Publications listed on page 137

LINCOLN’S INN FIELDS

33

TELOMERE BIOLOGY www.london-research-institute.org.uk/research/julie-cooper

Group Leader

Julia Promisel Cooper Postdoctoral Scientists Hani Ebrahimi Alfonso Fernandez-Alvarez Michael Klutstein Graduate Students Martina Begnis Alex Fennell Misa Ogura Sophie Zaaijer Yuan Zhao Scientific Officers Cecile Bez Jessica Greenwood Christopher Pitt Nadeem Shaikh Masters Student Clement Bertholet

Chromosome ‘linearity’ is so universal among eukaryotes that we assume it to be evolutionarily advantageous. However, formidable challenges are presented by linear chromosome termini, as their resemblance to damage-induced DNA breaks can expose them to degradation and fusion. If left unchecked, these pathways cause genomic instability and cancer. Telomeres protect chromosome ends from these events, and solve the ‘DNA end replication problem’ – the inability of DNA polymerases to fully replicate the ends of linear molecules – by engaging telomerase, a reverse transcriptase whose integral RNA subunit templates synthesis of telomere repeats. We study the spectrum and mechanisms of telomere function. Human stem cells express telomerase, insufficient levels of which cause diseases reflecting stem cell failure. However, telomerase expression is downregulated in most somatic cells, thus limiting cellular lifespan. To overcome this limit to proliferation, cancer cells activate telomerase or an alternative mode of telomere maintenance. Hence, while loss of telomere function promotes early tumorigenesis, the genomic instability precipitated by telomere loss eventually halts cancer cell proliferation, making telomeres intriguing universal anti-cancer targets. Fission yeast telomeres are remarkably similar to those of human but provide precise genetic manipulability. Taz1 (ortholog of human TRF1/ TRF2) binds telomeric DNA, regulates numerous telomere functions; other components of human ‘shelterin’ are also found in fission yeast and we are building an integrated picture of how these proteins protect chromosome ends. Control of telomere replication and segregation We had previously shown that ‘naked’ telomere repeats obstruct replication fork progression, and that stalled telomeric replication forks can lead to telomeric entanglements that impede chromosome segregation (Figure 1). We are focusing on the role of Rif1, a conserved replication/repair factor whose activities promote the conversion of stalled telomeric replication forks to lethal entanglements. We find that Rif1

34

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

localizes to the separating telomeres at anaphase, suggesting a Rif1-controlled final step in chromosome resolution at mitosis. In cells expressing telomerase, homeostatic mechanisms ensure that telomere lengths are maintained within a specific range, preventing individual telomeres within the cell from sporadically triggering cell death. This homeostasis requires tight control of telomerase both in time (it acts only after completion of replication) and space (it acts preferentially on the shortest telomeres). However, the key principles that underlie regulated access to telomeres remain mysterious. Taz1 and Rap1 are required to prevent excessive telomeric elongation, but we find that their mechanisms of action differ. Taz1 restricts telomeric overhang generation and telomerase activity to S-phase while Rap1 is dispensable for this timing. Furthermore, although taz1Δ telomeres experience transient attrition due to replication fork collapse, this is balanced not only by temporal expansion of the telomerase activity period but also by unusually high telomerase levels, suggesting that collapsed forks generate powerful telomerase substrates. Sheltering without canonical shelterin Telomerase-negative cells can occasionally acquire the ability to maintain telomeres via recombination. Fission yeast can also survive telomere loss by chromosome circularization;

circular chromosomes allow viability but confer several conspicuous defects, like extreme hypersensitivity to agents that induce DNA damage. We identified a third, damage-resistant subclass of telomerase-minus cells that survive using an intriguing strategy that we dubbed ‘HAATI’ (heterochromatin amplification-mediated and telomerase independent). In HAATI cells, telomere repeats are absent but large tracts of ‘generic’ heterochromatin jump to each chromosome end. This heterochromatin, along with a non-telomeric single strand overhang at the chromosome terminus, is thought to recruit the end-protection protein, Pot1, to chromosome ends; Pot1 is in turn crucial for maintenance of HAATI chromosome linearity. Our observations about HAATI have a number of interesting implications. They imply that Pot1 can bind and protect chromosome ends in the absence of its cognate DNA binding sequence. The mechanism of heterochromatin jumping in HAATI cells may shed light on general principles of heterochromatin maintenance and repair. Moreover, HAATI resembles the chromosome end-maintenance strategy found in Drosophila melanogaster, which lacks specific telomere sequences but nonetheless assembles terminal heterochromatin structures that associate with specific end-protection factors. The existence of such a strategy in organisms as diverse as fission Figure 1 In wild type cells, telomeres (in green) segregate after bulk chromatin (in red) but nonetheless segregate efficiently. In the absence of Taz1, telomere entanglement leads to a readily visible failure of chromosome segregation (below).

yeast and flies suggests that HAATI also occurs in human cells and may be a way for telomeraseminus cancer cells to achieve unlimited proliferation. We are investigating HAATI from each of these perspectives. We have tantalizing data suggesting a role for the RNAi pathway in controlling HAATI formation and a role for nuclear compartmentalization in maintaining the HAATI survival strategy; understanding these roles should illuminate principles of genome-wide heterochromatin regulation. Meiotic telomeres – regulators of centrosomes and centromeres During meiosis, the specialized cell cycle in which cells halve their ploidy to allow sexual reproduction, telomeres take on fascinating and crucial roles. Telomere clustering during early stages of meiosis, or ‘bouquet formation’, is observed throughout the Eukaryota. We found previously that in the absence of the bouquet (e.g. in taz1Δ cells), the spindle pole body (SPB; fission yeast centrosome equivalent) fails to divide properly and form a spindle at meiosis I, and often dissociates from the nucleus. Thus, the highly conserved bouquet plays an unanticipated role in controlling spindle formation. Our recent studies have defined at the molecular level how SPBs differ in the presence and absence of the bouquet. We have found that in the absence of the bouquet, contact between centromeres and the SPB can confer proper spindle formation, suggesting interchangeable functions shared by telomeres and centromeres. Surprisingly, we have also found that the bouquet affects centromere assembly and the attachment of chromosomes to the spindle during meiosis (Figure 2). This work raises new questions about interactions between different chromosome regions. Publications listed on page 138

Figure 2 In meiotic cells lacking the telomere bouquet, some centromeres fail to attach to the spindle.



LINCOLN’S INN FIELDS

35

SIGNAL TRANSDUCTION www.london-research-institute.org.uk/research/julian-downward

The Signal Transduction Group is interested in the mechanisms by which cancer cells become addicted to growth and survival signals generated by activated oncogenes and loss of tumour suppressor genes. We particularly focus on identifying unique dependencies of oncogene addicted cancer cells that might be targetable in the therapy of human cancer. Group Leader

Julian Downward Postdoctoral Scientists Esther Castellano-Sanchez Elza de Bruin Ralph Fritsch Nadia Godin-Heymann Madhu Kumar Miriam Molina Arcas Miguel Murillo Martin Clare Sheridan Britta Weigelt Graduate Students Catherine Cowell Georgios Vlachogiannis Scientific Officers Elena Armenteros-Monterroso David Hancock Patricia Warne Visiting Worker Inge de Krijger

Investigation of mechanisms of oncogene driven transformation and drug resistance Much of the work in the laboratory has focused on the RAS family of oncogenes and the signalling pathways that they control. RAS genes are activated by point mutation in some 20% of all human tumours and are known to play a key role in the establishment of the transformed phenotype. While the signalling pathways activated by RAS are well characterised (Downward, 2003; Nat Rev Cancer. 3: 11-22), it remains a major challenge to identify what proteins are selectively important in the establishment and maintenance of the transformed phenotype and may therefore act as potential therapeutic targets for cancer treatment. In order to investigate novel aspects of these pathways in cancer cells, we have employed functional genomics approaches using posttranscriptional gene silencing by genome-scale libraries of RNA interference agents. When cells become progressively transformed during the evolution of cancer, they suffer stresses that are not seen by normal cells and become increasingly dependent on stress management pathways. This means that the tumour cells show a unique set of dependencies, both on the oncogenic drivers and also on stress handling pathways, sometimes termed oncogene addiction and non-oncogene addiction, respectively (Luo et al., 2009; Cell. 136: 823-37). We have investigated these dependencies by RNAi screening, comparing a cancer cell line containing an activated KRAS allele with a normal (‘isogenic’) derivative in which this has been removed (Wang et al., 2010; Oncogene. 29: 4658-70), and also using a panel of thirty or so cancer cell lines, half of which were mutant and half wild type for KRAS (Steckel et al., 2012; Cell Research. 22: 1227-45). This approach

36

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

uncovered proteins whose therapeutic targeting might be expected to provide differential toxicity towards KRAS mutant tumour cells. One example of a determinant of non-oncogene addiction found in this way, the transcription factor GATA2, has been investigated in detail. Using genetic mouse models, we could show that the development and continued maintenance of KRAS induced lung cancer is uniquely dependent on the expression of GATA2: when this is blocked, the lung adenocarcinomas regress completely. Although, as a transcription factor, GATA2 itself is not likely to be a good drug target, using a range of methods to investigate the transcriptional programme controlled by GATA2 we identified a number of downstream pathways that can be inhibited by existing drugs. Using a combination of two such drugs we have been able to show an impressive therapeutic response in KRAS induced mouse lung cancer models (Kumar et al., 2012; Cell. 149: 642-55). We are now investigating the possibility of studying this drug combination in a clinical setting. The role of phosphatidylinositol 3-kinase in RAS-driven oncogenesis RAS proteins signal through direct interaction with a number of effector enzymes, including type I phosphatidylinositol (PI) 3-kinases. We have generated mice with mutations in the RAS binding domain (RBD) of the Pik3ca gene encoding the PI 3-kinase catalytic p110α isoform that block its ability to interact with RAS (Gupta et al., 2007; Cell. 129: 957-68). These mice are highly resistant to endogenous KRAS oncogene induced lung tumourigenesis and HRAS oncogene induced skin carcinogenesis. The interaction of RAS with p110α is thus required in vivo for RAS-driven tumour formation. The demonstration of the importance of the RAS/PI 3-kinase interaction in

Figure 1 The effect of loss of p110α interaction with RAS on preexisting lung tumours in mice. A KRAS induced mouse lung cancer model (KRAS LA2, Johnson et al., 2001; Nature. 410: 1111-6) was used to study the effect of removing RAS interaction with p110α on pre-existing lung adenocarcinomas, using an inducible tamoxifen inducible Cre recombinase system. Reconstructions of representative lung tumours measured by micro CT are shown and compared between animals expressing either a single allele of RBD mutant or wild type p110α both before (week 0) and 19 weeks after deletion of the other allele of p110α, which was floxed wild type. Loss of p110α interaction with RAS leads to initial partial regression and then to long-term stabilisation of the lung tumours.

tumourigenesis raises the prospect that agents that disrupt this interaction might have particular value in cancer therapy. This work is being further pursued by the generation of mice with inducible expression of the inactivating mutation in the RAS binding domain of p110α so that the requirement of this interaction for tumour maintenance, rather than simply tumour initiation and development, can be assessed. It appears that the RAS/p110α interaction plays a role in tumour maintenance, although it is not as critical as it is for tumour formation (Figure 1). Furthermore, we have also created a mouse with inactivating mutations in the RAS binding domain of p110β, the other ubiquitously expressed PI 3-kinase catalytic subunit isoform, and are testing the effects of this



mutation on tumour initiation and maintenance, especially in the context of PTEN deletion, where p110β is thought to be particularly important. Our investigations with p110β have led us to the surprising observation that this isoform is not controlled by direct interaction with RAS, unlike p110α, γ and δ, but rather that the RBD of p110β interacts directly with a number of other small GTPases with distinct biological function. This has led us to a significantly revised model of how extracellular stimuli, especially those signaling through G protein coupled receptors, activate the PI 3-kinase activity of p110β, and the importance of this mechanism in cancer metastasis and also fibrosis.

Publications listed on page 138

LINCOLN’S INN FIELDS

37

VASCULAR BIOLOGY www.london-research-institute.org.uk/research/holger-gerhardt

Group Leader

Holger Gerhardt Postdoctoral Scientists Katie Bentley Raquel Blanco Claudio Franco Martin Jones Eleonora Lapi Graduate Students Irene Aspalter Véronique Gebala Filipa Neto Benedetta Ubezio Scientific Officers Jane Babbage Russell Collins Anan Ragab Dominique Sauvaget

The Vascular Biology Group aims to unravel the basic cellular principles and the molecular control of blood vessel patterning in development and disease. Blood vessels are critical for tissue growth and healthy organ function. Effective blood vessel function requires that the endothelial cells lining the vessel assemble a regular network of interconnected tubes with adequate diameter and branching frequency support blood flow. As different organs serve distinct functions, and possess different metabolic requirements, blood vessel patterning bears organ specific characteristics. We use a cell biology approach in various model systems in vivo and in vitro, in combination with computational modelling to investigate how individual endothelial cells respond to signals from the tissue and communicate with each other in order to orchestrate behaviour leading to functional network formation. Recently we discovered a surprising degree of dynamic endothelial cell rearrangements during sprouting angiogenesis (Jakobsson et al., 2010; Nat Cell Biol. 12: 943-953). When generating chimeric mouse embryos through blastocyst injections of genetically tagged embryonic stem cells, cell proliferation leads to clonal expansion in many epithelial structures and organs. Also in tissues that are constantly renewed during adult life, such as the gut epithelium, clonal expansion leads to clear lineage boundaries within the epithelium. Surprisingly, such regional clonality is not present in the vascular endothelium. Instead, daughter cells arising from cell division disperse in the vascular network, indicating that endothelial cells migrate even after the vessel has formed. Indeed, in the zebrafish brain, we observed this dynamic rearrangement both in new sprouts and in already perfused vessels. Several of our current projects investigate mechanisms and function of this dynamic endothelial cell rearrangement during sprouting and vascular pattern formation. Integrating computational modelling and experimentation, we identified that a key driving force for endothelial rearrangements is differential adhesion. Katie Bentley previously,

38

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

together with Paul Bates in the Biomolecular Modelling Group, developed an agent based computational model of endothelial tip and stalk selection incorporating a feedback loop between VEGFR2 activation, Notch ligand Dll4 production, Notch signalling, and Notch dependent downregulation of VEGFR2 (Bentley et al., 2008; J Theor Biol. 250: 25-36). This feedback loop operates between neighbouring endothelial cells establishing dynamic and competitive tip/stalk selection through lateral inhibition. In order to model physical properties, dynamic cell shape changes and cell intercalation, Katie developed a sophisticated modelling platform that incorporates the cellular potts model with differential adhesion into the agent based model of tip/stalk selection. The basic assumption of differential adhesion (Graner and Glazier 1992; Phys Rev Lett. 69: 2013-16) is that a cellular system will strive to achieve an energy minimum in which more strongly adhesive cells prefer to neighbour each other. As a consequence, cells with different cadherin levels or different cadherin molecules will sort in a homophilic way, driving a highly mosaic situation towards clusters of highly adhesive cells with weakly adhesive cells apart. In order to investigate the hypothesis of differential adhesion in the angiogenic sprout, Katie modelled two distinct mechanisms that affect adhesion and polarity. The first one is based on previous studies showing that VEGFR2 signaling promotes endocytosis and turnover of vascular

A Figure 1 Colour heat-map representation of VE-cadherin patch classifications in mouse retina under normal conditions A. After Notch inhibition (DAPT) B. After intraocular VEGF injection. C. Note the highly differential colour pattern including strongly inhibited patches neighbouring highly active patches in normal branching conditions. Notch inhibition shifts the pattern to highly active, where as high VEGF leads to clustering of larger regions with similar patch classifications, correlating with loss of branching.

B

C

endothelial cadherin (VE-cadherin) at the cell junction. A cell with higher VEGFR2 levels and activity would accordingly possess more dynamic VE-cadherin turnover, and thus reduced adhesion. Indeed, VEGF-A leads to increased vascular permeability partly through this effect on endothelial cell junctions. The second mechanism is based on a simpler computational model she developed previously, which predicted that cells with lower Notch activity are more motile, in a polarised fashion towards the tip. Recent in vitro studies support this scenario (Arima et al., 2011; Development. 138: 4763-76). We incorporated a Notch regulated polarity component that affects directional cell protrusions mediating intercalation and then tested through many combinatorial simulations whether indeed differential adhesion, polarity and/or Notch regulation were all or in part required to generate realistic rearrangement behaviour. We validated the simulations against a large data set from time-lapse movies containing mosaic cell competition situations that affect the propensity of individual cells to move and reach the tip position. Surprisingly, only simulations that combined differential adhesion between the endothelial cells, as opposed to equally strong or equally weak adhesion regimes amongst all cells, and the existence of polarized Notch regulated protrusions completely matched all experimental data. Closer analysis of the underlying principles and interactions illustrated that differential adhesion cooperates with polarized protrusions to drive long-range movement and intercalation. In fact, these mechanisms appear to reinforce the robustness and efficacy of lateral inhibition in establishing the observed salt-and-pepper selection of tip and stalk cells. As a consequence, differential adhesion and lateral inhibition unite to shape the frequency of branching and vessel extension through intercalation. In order to test whether VE-cadherin is indeed regulated and patterned in a differential manner within the sprout, we used time-lapse imaging of vascular sprouts expressing a GFP-tagged allele of VE-cadherin in the endogenous gene locus. Further, we developed image segmentation and classification of VE-cadherin profiles in the mouse retina and embryonic stem cell derived sprouting assays, identifying for the first time that



VE-cadherin shows a highly differential pattern that is Notch dependent. Segmentation and classification across retinas of different developmental stages and under Notch inhibition illustrated that differential VE-cadherin patterning correlates with phases of active cell rearrangements, and that cells that are more active, consistently show more endosomal VE-cadherin and contractile, serrated junctions indicative of movement through a cell sheet. Mature vessels, comprised of quiescent endothelium in contrast showed no endosomal VE-cadherin and no differential patterns. To gain insight into the role of these new rearrangement mechanisms, Katie modelled the effects of pathological VEGF levels, similar to situations in ischemic neoangiogenesis or in the tumour microenvironment. In previous work, we found that high VEGF levels lead to synchronization of Dll4/Notch signalling within the endothelial populations such that cells synchronously alternate between all attempting active sprouting or all adopting the inhibited state. In the new model, high VEGF levels also caused this synchronization, but additionally showed that synchronization disrupts the differential patterning, halting all effective cell rearrangements. Instead of intercalating, the model predicts that cells will cluster and vessels become stunted and dilated. Testing these new predictions, we studied VE-cadherin patterning using the image segmentation and patch classification in retinas following VEGF-A injection (Figure 1), and in vessels in mouse glioblastoma. In both cases vessel diameter control is lost, and vessel become tortuous and show glomerular clusters. In addition, the differential pattern of VE-cadherin between individual cells is lost, instead large clusters of adjacent cells show similarly active or inhibited junctions, indicating that the endothelial population switches from intercalation that drives effective branching, to clonal expansion and clustering, instead driving vessel diameter increase. Ongoing studies into the molecular components that regulated differential VE-cadherin patterning between the endothelial cells under VEGF and Notch stimulation will hopefully soon shed more light on this fascinating new behaviour underlying branching and patterning in angiogenesis.

Publications listed on page 138

LINCOLN’S INN FIELDS

39

IMMUNO SURVEILLANCE LABORATORY www.london-research-institute.org.uk/research/adrian-hayday

Immunogenicity describes the capacity to activate the immune system to make protective responses. The past fifteen years has largely elucidated the immunogenic pathways by which we detect infectious agents. We now seek to define the molecular interactions responsible for the immunogenicity of more generalized tissue disruption, such as that caused by toxins, physical damage, or cell transformation. We term this ‘Lymphoid Stress Surveillance’ (LSS). Group Leader

Adrian Hayday Postdoctoral Scientists Lucie Abeler-Dorner Livija Deban Rosie Hart Fernanda Kyle Marie-Laure Michel Olga Sobolev Jessica Strid Mahima Swamy Gleb Turchinovich Pierre Vantourout Melanie Wencker Clinical Scientists Deborah Enting Rick Woolf Yin Wu Graduate Students Bodhi Hunt Rafael di Marco Barros Scientific Officer Anett Jandke

40

Despite identifying some key molecular and cellular players, critical gaps need to be filled before this pathway can be generally accepted and contextualized. In particular, LSS should have profound relevance for tumour immunology since it considers the immune potential to recognize and respond to cells that are en route to malignant transformation. In particular, we seek to understand how LSS ‘learns’ how to distinguish dysregulated cells from healthy cells, thereby avoiding chronic inflammation. In parallel with our basic studies, we maintain clinic-based investigations at Guy’s Hospital. Tissue stress and allergy To better understand LSS, we have been developing new experimental approaches, and were delighted at their enthusiastic endorsement by the Quinquennial Review Committee in May 2012. Thus, a major unexpected finding in 2011 was that when skin-resident T cells engage Rae-1, an MHC class I-like molecule upregulated on physically-stressed tissue, the response – mediated by the activating receptor, NKG2D – leads rapidly to downstream IgE production, most commonly associated with allergies. This evoked a 20-year old hypothesis that IgE may expel toxins and prevent their systemic dissemination, thus protecting tissues. It also suggested a novel route to IgE, and hence to allergies. To pursue this, we adopted a nucleic acid sequencing approach to assess the quality of IgE induced by tissue-stress compared to that induced by exposure to defined allergens. Second, we began collaborating with Professor Ian Kimber (Manchester) to test the contribution of LSS to symptomatic responses (e.g. atopic dermatitis) to environmental and/or industrial sensitizing agents. And third, in

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

collaboration with LRI specialist services, we combined in situ hybridization with immunohistochemistry to identify the B cells responsible for IgE production following tissue disruption. What is stress? The stress-response of tissue-associated lymphocytes simultaneously invokes three questions: what conditions upregulate NKG2D ligands; what other potential ‘stress antigens’ are co-induced; and is there simultaneous downregulation of molecules that might ordinarily brake the response? We have identified a mechanism by which epidermal growth factor dramatically increases the RNA stability for MICA – a major human NKG2D ligand, widely associated with cancer. Thus, LSS functions in situations beyond the DNA-damage response to mutagenic stress. In collaboration with Professor B Wollscheid (Zurich), we also have employed ‘Cell Surface Capture’ proteomics to define additional molecular changes to the surface of epithelial cells subject to different ‘stresses’, and are examining the impact of those changes on local T cells. Thirdly, we are employing improved imaging techniques to visualize mouse intraepidermal T cell interactions with epithelial before and after ‘skin-stress’. Under such circumstances, we note the downregulation of Skint-1, an Immunoglobulin superfamily molecule that we and our collaborators first isolated, that is necessary for normal intraepidermal T cell development and that plays a fundamental role in tolerising T cells more generally (Roberts et al., 2012; Immunity. 36: 427-437). In sum, we seek to describe how discrete responses of epithelial cells to body surface stresses are integrated so as to orchestrate

1

Figure 1 IL-17-producing gamma delta T cells induce disease. There is a rapid increase in IL-17-producing γδT cells when mice are painted with aldara, an agent containing imiquimod (IMQ), a clinicallyemployed anti-cancer agent that can induce psoriasis as a side effect. The IL-17-producing γδ T cells are exquisitely sensitive to the cytokine, IL-7, and the simultaneous application of anti-IL-7 antibody inhibits the IMQ-induced increase in cell numbers. This likewise inhibits erythema and other aspects of the psoriasis-like pathology, demonstrating the critical contributions of IL-17-producing γδ T cells to disease (From Michel et al., 2012).

2

Figure 2 IL-7 selectively regulates STAT3 in IL-17-producing gamma delta T cells. The depicted images are from Image Stream. Image 1 shows a cell: in the top row, the cell stains purple for CD44 (2) and green for the γδ T cell receptor (3), and is an IL-17-producing cell. Image 5 shows that phosphorylated STAT3 is detectable (red) in response to IL-7. Conversely, in the lower row, the cell stains green for the gamma delta T cell receptor, and yellow for CD27 and is an interferonproducing cell. It does not activate STAT3 in response to IL-7 (from Michel et al., 2012). Figure 3 Integrated responses of different skin cell types to stress. The y-axis depicts membrane staining for the expression of a protein, CD45 by cells isolated from the outer layer of the skin (epidermis). The cells expressing high levels (light grey) are skin-resident T cells; those expressing medium levels (grey) are skin-resident dendritic cells (Langerhans cells); the cells expressing no CD45 (black) are epithelial cells. The x-axis depicts cytokine staining. When the tissue is dysregulated (right-hand side) the cytokine IL-6 (box IL-6+) is detected among epithelial cells relative to control treatment (left-hand side) but is made by no other cells. The simultaneous response of the T cells is to make IL-13 (bottom panel) that is again not made by any others cell-types. Hence, environmental agents that dysregulate a body surface can provoke complex responses that are the integration of individual cell-type responses.

3

potentials. Moreover, we have found that intraepithelial lymphocytes (IEL) display equivalent functional responses in the gut or skin, while sub-epithelial compartments in either location likewise express similar but distinct functional potentials. We found that these potentials are pre-programmed in the thymus during T cell development. Specifically, intraepidermal T cells ‘learn’ function by engaging thymic epithelial cells expressing Skint-1 (Turchinovich and Hayday, 2011; Immunity. 22: 59-68). More recently, we found that Skint-1 also sets the response threshold for skin IEL. Conspicuous parallels with intestinal IEL have again surfaced, but since these cells are Skint-1independent, we can presumably identify Skint-1equivalent molecules that regulate gut immune responses. Moreover, when systemic T cells enter the gut epithelium, for example following systemic infection, equivalent response thresholds are imposed, without which the epithelia would descend into chronic inflammation. By use of mutants, we have identified a regulatory molecule that seemingly controls this, and that thereby regulates gut immune responses. T cells that orchestrate A key signature of LSS is that tissue T cells sense the environment and orchestrate down-stream events, rather than simply targeting the tissue as part of a systemic immunoprotective response. Thus, we have found that dermal γδ T cells producing IL-17 are rapidly responsive to skin exposure to a drug that can induce psoriasis-like symptoms, and are themselves regulated by the cytokine IL-7 (Michel et al., 2012; PNAS. In Press). By blocking IL-7 we could reduce pathology, with implications for other inflammatory diseases, notably multiple sclerosis, with which IL-7 is associated. Together with I. Kimber we also found such cells to be critical to atopic dermatitis and are now actively pursuing parallel studies in humans.

immune responses (Willcox et al., 2012; Nature Immunology. 13: 872-79). In parallel we are investigating how each component of lymphoid stress surveillance may facilitate immune recognition of carcinoma, specifically prostate cancer and triple negative breast cancer that are a focus of our clinical studies at Guy’s hospital. Learning surveillance T cells associated with skin or gut fall into specific micro-anatomical compartments, e.g. epidermis and dermis, that have distinct functional



We’re not all adults The majority of immunology is learned from adult mice or humans, whereas neonates experience the most catastrophic exposure to the environment. We hypothesized that LSS may be particularly important to newborns, and found that human IL-17-producing γδ cells invoked in LSS are massively enriched in neonates. Hence, we have developed a major longitudinal study of newborns to better characterize the development of immune function. Publications listed on page 139

LINCOLN’S INN FIELDS

41

DEVELOPMENTAL SIGNALLING www.london-research-institute.org.uk/research/caroline-hill

Group Leader

Caroline Hill Postdoctoral Scientists Marco Briones-Orta Joanne Harding Claire Heliot Anassuya Ramachandran Marie-Christine Ramel Antonius van Boxtel Pedro Vizán Graduate Students Tessa Gaarenstroom Daniel Miller Sabine Reichert Scientific Officers Debipriya Das Ilaria Gori Rebecca Randall

42

Our work focuses on the signalling and function of Transforming Growth Factor β (TGF-β) superfamily ligands, which include TGF-βs, Activins, Nodals, BMPs and GDFs. Many of these ligands play a central role in early vertebrate development, acting as morphogens in the specification and patterning of the germ layers. Moreover, aberrant TGF-β, BMP and Nodal signalling is implicated in many different types of cancer. Our goal is to understand the mechanism whereby these ligands signal from the plasma membrane to the nucleus, determine how they function and are regulated in embryonic development, and how these signalling pathways are hijacked by tumour cells to promote growth of primary tumours and metastasis to distant sites. Mechanism of TGF-β/Smad signalling The best characterised pathway downstream of TGF-β superfamily ligands is the Smad pathway. Several years ago we built a predictive computational model of the core intracellular Smad signal transduction module of the TGF-β pathway, using a wealth of quantitative and qualitative data generated in my lab. The scope of that model was an acute TGF-β stimulation, followed by pathway termination using the receptor inhibitor SB-431542. However, understanding how signal transduction pathways function in vivo in complex physiological and pathological contexts requires a detailed knowledge of signalling dynamics over extended times. We have shown for the first time that TGF-β exposure desensitises cells, rendering them refractory to further stimulation for approximately 24h. This loss of signalling competence also explains the attenuation of signalling observed after prolonged ligand stimulation. To shed light on the underlying mechanism we developed a new computational model by designing a receptor module that incorporated these novel experimental findings and merging this module with our original Smad signal transduction module (Figure I). Fitting the comprehensive model to our kinetic data revealed that both signal attenuation and refractory behaviour are caused by the fast depletion of signalling-competent receptors upon TGF-β

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

binding and their slow replenishment, which is the rate-limiting step for regaining signalling competence. This kinetic model allows the prediction of TGF-β superfamily responses in in vivo contexts such as tumourigenesis. We predict that tumours that receive a constant supply of TGF-β would maintain only a very low level of signalling, allowing expression of a subset of TGF-β genes, but keeping silent those which require a strong acute signal. This may explain some aberrant TGF-β responses in tumours in vivo (Vizán et al., Submitted). We recently discovered a novel arm of TGF-β/ Smad signalling through Smad1 and Smad5, which occurs in many epithelial and cancer cells in addition to the well-characterised signalling though Smad2 and Smad3. This leads to formation of so-called mixed R-Smad complexes between phosphorylated Smad2/3 and phosphorylated Smad1/5. We are investigating the function of this novel arm of TGF-β signalling in both untransformed cells and in tumourigenesis. We have shown that the ability of TGF-β signalling to inhibit BMP-induced transcriptional responses is mediated by phosphorylated Smad3–Smad1/5 complexes, acting at the level of BMP-responsive promoter elements. Moreover, these particular mixed R-Smad complexes are also important for preventing TGF-β inducing transcription of BMP-responsive genes, as shown in Smad3

Figure 1 The topologies of the receptor and Smad networks used for computational modelling of the TGF-β pathway. TGF-β binds competent receptors at the surface (R Scom), which become pre-active (RT) until they are internalised (Ract) and can phosphorylate cytoplasmic Smad2. Receptors need to mature (kmat) intracellularly (RIcom) before they can signal. Receptors are subject to synthesis (k syn) and degradation (kd) rates, with a factor (D) to increase degradation of ligand-bound receptors. In both the nucleus and the cytoplasm, phosphorylated Smad2 (PSmad2) forms heteromeric complexes with Smad4 and homomeric complexes. Smad2, monomeric PSmad2, and Smad4 shuttle reversibly between nucleus and cytoplasm. Heteromeric complexes and homomeric complexes are imported into the nucleus faster than monomeric Smad2 by a factor CIF (complex import factor), but cannot be exported. The Smad2 phosphatase (PPase) is nuclear and irreversibly dephosphorylates monomeric PSmad2. Finally, the receptor kinase is reversibly blocked by the inhibitor SB431542.

knockdown experiments. Smad3 is therefore not only required for mediating TGF-β’s inhibitory effects on BMP signalling, but also plays a critical role in restricting the transcriptional output in response to TGF-β (Grönroos et al, 2012; Mol Cell Bio. 32; 2904–16). The role of the E3 ubiquitin ligase Arkadia in tumourigenesis We identified the E3 ubiquitin ligase Arkadia several years ago as absolutely required for a subset of TGF-β responses via Smad3. It functions by inducing degradation of the transcriptional co-repressors Ski and SnoN in response to ligand. We have been focusing on its role in tumourigenesis, and in collaboration with the Tumour Cell Biology Group have shown that it has potent tumour-promoting activity. In three different cell lines that require TGF-β for metastasis: MDA-MB-231 (human breast carcinoma), MTLN3E (rat breast carcinoma), and B16 (mouse melanoma) we have shown that knockdown of Arkadia results in a substantial inhibition in the ability of these cells to colonise lungs of immunodeficient mice. In vitro co-culture experiments with HUVEC cells, as well as RNA seq analysis of TGF-β-responsive genes that require Arkadia, suggest that Arkadia is required for extravasation (Briones-Orta et al., Cancer Res. In Press).



Spatial regulation and function of Nodal and BMP signalling in early vertebrate development It is becoming clear that in addition to TGF-β, BMPs and Nodal also play an important role in cancer. Many of their roles in cancer represent a redeployment of their roles in early development and in tissue homeostasis. One of our major goals therefore is to understand how Nodal and BMP activity is spatially and temporally regulated during early vertebrate development and to determine their function in this context. To this end we have generated transgenic zebrafish lines expressing fluorescent reporters specific for these particular pathways. This year we have been focusing on determining how the ventral to dorsal BMP gradient is formed during the blastula stages of zebrafish development, using our BRE-mRFP line that contains an mRFP reporter driven by Smad1/5–Smad4-binding elements that respond to BMP/GDF signalling. In contrast to the diffusion/transport-based models of BMP gradient formation in Drosophila, which have informed most of our thinking about signalling gradient formation to date, we have shown that establishment of the BMP activity gradient in early zebrafish embryos is determined by graded expression of the BMP ligands (Ramel and Hill, Dev Biol. In Press). A little later in development, at the beginning of somitogenesis, we observed a horseshoe-shaped domain of BMP activity around the anterior neural plate, which is remodelled over the subsequent few hours to form an additional inner domain which marks the neural crest. We have shown that this BMP activity is required for neural crest formation and are currently investigating how these domains are generated. We have generated another zebrafish line, the ARE-EGFP line, that contains an EGFP reporter that binds Smad2–Smad4–FoxH1 complexes in response to Nodal signalling to understand how spatial Nodal activity is controlled, and we are also investigating how the dynamics of Nodal signalling are regulated. This work is being complemented by another project in the lab focused on how Nodal and the related ligand, Activin regulate transcription in the context of chromatin. Publications listed on page 139

LINCOLN’S INN FIELDS

43

DEVELOPMENTAL GENETICS www.london-research-institute.org.uk/research/david-ish-horowicz

Group Leader

David Ish-Horowicz Postdoctoral Scientists Rippei Hayashi Nathaniel Hoyle Cristian Soza-Ried Graduate Student Sophie Liddell Scientific Officers Sheena Pinchin Stephen Wainwright

The organised complexity of all multicellular animals arises from a single-celled fertilised egg, and relies on sets of evolutionarilyconserved genetic pathways that are reused many times during development. The Developmental Genetics Group examines the establishment of cell asymmetries, and how they pattern embryos to generate the complex spatial architecture of the adult. We are particularly interested in how molecular motors drive intracellular asymmetries, and in the oscillator circuitry that controls the process of segmentation along the vertebrate body-axis. The former questions are being addressed in the fruitfly Drosophila, whose sophisticated genetics allows rapid gene identification and functional analysis. We study the latter in model vertebrates – chick, mouse and zebrafish – focussing on the kinetics of gene expression and how they can contribute to the periodicity of segment formation. Genetic screen for novel factors regulating mRNA transport in Drosophila Patterning of the two major body axes in Drosophila takes place within the developing oocyte, via localised mRNA transcripts that establish asymmetries that are propagated and elaborated following egg fertilisation. gurken (grk) transcripts, which encode a TGFα-like protein, localise posteriorly in early oocytes to define the anteroposterior axis. Later, they accumulate anterodorsally to establish dorsoventral polarity. At each stage, Gurken protein establishes polarity by signalling to overlying accessory ‘follicle cells’ that, in turn, further pattern the oocyte. We have established a genetic screen for factors required for selective localisation of gurken transcripts that are labelled in vivo via a fluorescent binding-protein. Examination of dissected ovaries under the microscope readily reveals newly-induced mutations that disrupt gurken accumulation, and also mutations that affect oocyte specification and morphology. Using standard genetic techniques combined with whole genome sequencing, we have identified and

44

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

mapped the molecular lesions responsible for mutations in more than 20 genes. These include several new alleles of armitage, a gene that encodes an RNA helicase required for the piwi pathway of RNA-mediated transposon-silencing in the germline. These alleles include the first completely null mutants, with defects in follicle cell development that was unaffected by previous, slightly active alleles. Our results reconcile the various genetic data indicating that Piwiassociated small RNAs (piRNAs) also silence transposons in follicle cells. Amongst the other mutations we recovered is an allele of the molecular motor component, kinesin light chain (klc), which was identified in combination with a co-induced mutation affecting piRNA activity, due to localisation of grk transcripts anteriorly but not dorsally. Loss of klc activity alone leads to misorganisation of the oocyte microtubule cytoskeleton and weak grk mislocalisation. Oocyte centrosome behaviour is also defective. Normally, all 16 centrosomes in an egg-chamber are transported into the oocyte where they aggregate into a single, large centrosome. In klc oocytes, the centrosomes are

transported into the oocyte but cluster only loosely. The cluster fails to localise correctly within later oocytes, thereby contributing to the disrupted cytoarchitecture and grk transcript mislocalisation. Transcriptional kinetics in eukaryotic cells and embryos The sequential production of body segments in vertebrate embryos is regulated by a molecular oscillator (the segmentation clock) that drives cyclic transcription of genes involved in positioning intersegmental boundaries. Many pathways cycle during segmentation, and it is not clear which one (or ones) determine the clock’s period and, thereby, somite size and number. Oscillators depend on delayed negative-feedback circuitry, and mathematical modelling indicates that clock period is a direct function of the total delay-kinetics of the circuit (Lewis, 2003: Curr Biol. 13; 1398-408). To define the pacemaker oscillator, one should modify its period, which implies altering the delays such as those associated with synthesis of clock component mRNAs. To identify rate-limiting transcriptional steps, we have used zebrafish, chick and mouse embryos (whose clock periods are 30, 90 and 120 min, respectively) to measure the delays for three Notch-modulating genes that cycle during segmentation, Lunatic fringe (Lfng), Hes7/her1 and Nrarp. Initial experiments using cultured cells with inducible Lfng or Hes7 expression indicated that transcriptional elongation was very rapid, and unlikely to cause significant delays. However, mRNA splicing and export were much slower and could substantially delay transcript production. To

test if these results apply in vivo, we measured transcriptional delays in developing embryos, using in situ hybridisation with fluorescent probes. The clocks of neighbouring cells are synchronised locally, such that the spatial displacements between expression domains can be used to measure temporal delays. By using probes specific for different transcript forms we could estimate delays due to elongation, processing and export from the nucleus (Figure 1). We were unable to visualise elongation delays in chick and mouse embryos, indicating that primary transcription is complete within 2-3 min. However, intron splicing takes ~7-10 min in chick and ~10-12 min in mouse, implying that removal of the last intron contributes significantly to clock period. Even longer delays, 10 and 15 min in chick and mouse respectively, are associated with export of processed mRNAs from the nucleus into the cytoplasm, indicating that this process provides the predominant transcriptional delay contributing to the clock, and may contribute to species-specific differences in clock period. Our results indicate that the kinetics of mRNA production account for much of the clock period in the three species examined, and provides strong support for delayed autorepression as the underlying mechanism of the segmentation clock.

Publications listed on page 139

Figure 1 Fluorescent in situ hybridisation of mouse embryo visualising delays due to processing and export of Hes7 transcripts in the presomitic region. Anterior margins of the unspliced (green; arrow) and spliced (red; arrowhead) transcript domains are offset, reflecting the time-delay associated with splicing. Nuclear and cytoplasmic spliced transcripts can be distinguished via staining of nuclei (blue), allowing measurement of the export delay. Horizontal bars mark boundaries between formed somites.



LINCOLN’S INN FIELDS

45

CELL BIOPHYSICS www.london-research-institute.org.uk/research/banafshe-larijani

Cell Biophysics draws upon the physical sciences to develop novel avenues for investigation of mechanisms involved in protein and lipid interactions and dynamics at various subcellular membrane compartments (plasma and nuclear membrane). Our focus is to develop a global view on the effect of membrane dynamics on signalling and vice versa. Group Leader

Banafshe Larijani Postdoctoral Scientists Marie-Charlotte Domart Selvaraju Veeriah Clinical Scientist Julien de Naurois Graduate Students Gary Hong Chun Chung Nirmal Jethwa Scientific Officers Richard Byrne Veronique Calleja Consultant Engineers Christopher Applebee Pierre Leboucher

Modulation of lipid species composition defines both the morphological changes of the membrane and the ‘local signal’. Perturbations caused by changes in membrane curvature and morphology due to phospholipid composition variations affect the affinity of proteins to be targeted to appropriate membrane compartments. The regulation of membrane protein-lipid and proteinprotein interactions will be affected by the compositional and physical properties of the membrane bilayers. We focus studies, on the polyphosphoinositides and their derivatives. These phospholipids have been mainly recognised as second messengers and their effect on membrane dynamics and morphology has not been correlated with their role as signalling molecules. Using chemical biology tools our goal is to study how phosphoinositide-dependant protein interactions and dynamics are affected at both compartments, upon the acute and localised intervention of the phosphoinositides. Membranes, Morphology and Function Membrane morphology and function depends on associated proteins, but important events such as fusion and fission or organelles shaping, involve alterations of lipid composition. We investigate the in vivo role of lipids in ensuring the correct morphology of subcellular compartments such as the nuclear envelope. Regulation of nuclear envelope dynamics is an important example of the universal phenomena of membrane fusion and fission. The nuclear envelope (NE) is disassembled and reassembled at each mitosis in a typical animal cell. The NE is not a ‘passive’ membrane in that its reformation is central to proper cell functioning. To understand the complex architecture of the NE is important, as its correct formation is a requirement for all

46

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

cells that have an open mitosis. Dysfunction of NE assembly leads to different forms of human disease, cancer being amongst one of them. In the past years there has been much debate on the mechanism of NE assembly, and to date this complex issue has not been fully resolved. We have shown the involvement of lipids in the regulation of NE assembly, by using the formation of the male pronucleus in echinoderms as a simple model to investigate the molecular mechanisms involved in NE assembly. The process of male pronuclear formation is comparable to NE assembly in somatic cells. The discovery of a non-(ER) membrane fraction with atypical levels of phosphoinositides (60 mol%) has led to the hypothesis that this family of lipids are not only transient signals but that they affect membrane dynamics rendering them both signalling and membrane modifying molecules. To provide evidence for this hypothesis we have developed solid-state nuclear magnetic resonance spectroscopy methods to define the structure and dynamics of the natural membrane domains in our model. Our novel research using localised cellular intervention has demonstrated that one type of lipid, diacylglycerol (DAG), is essential for the formation of the NE (Figure 1). Without affecting other subcellular compartments, we locally depleted this lipid, resulting in a fragmented envelope and cell death. However, the cells, which were not affected by the localized intervention, lived and continued dividing. The fragmented NE phenotype was rescued to reform completely, by the addition of DAG enriched vesicles to the DAG-depleted cells. Once the nuclear envelope formed in its entirety the cell survived through further divisions. In this pioneering work we demonstrate that proteins are not the sole

Figure 1 Comparison of 3D models reconstructed from CLEM serial images of DAG-depleted (left panel) and DAG-rescued (right panel) HeLa cells shows the NE reformed in the presence of 1,2 DAG.

regulators in nuclear envelope formation and importantly these molecular processes are conserved from echinoderm embryos to mammalian cells. Our future research in the regulation of nuclear envelope assembly will investigate the role of fusogenic vesicles, transfer proteins, phosphoinositides and their modifying enzymes both in mammalian and echinoderm oocytes, providing mechanistic insights in membrane fusion across the deuterostome superphylum. Phosphoinositide dynamics at the plasma membrane and their effect on the regulation of AGC kinases in vivo The serine/threonine kinases PKB/Akt and PDK1 have received considerable attention over recent years and have become the focus of drug targeting for cancer therapy since a major role for these proteins in tumour progression has emerged. Our quantitative imaging approach has provided novel insights into the mechanism of in vivo conformational dynamics of PKB/Akt and PDK1. We will further our investigation by assessing the mechanism of interaction between PKB/Akt and PDK1 and will use 3D cell models to have a deeper understanding about complex formation and dynamics of these onco-proteins in an environment that is closer to a solid tumours. Our aim is to also to modify acutely PtdIns (3,4,5) P3 at the plasma membrane using the heterodimerisation tool, and determine whether these kinases will localise to another compartment and still maintain their activity within those compartments. After locally perturbing phosphoinositide metabolism at the plasma membrane, we will monitor the phosphorylation state, localisation and conformation of our PKB-FRET biosensor by two-photon fluorescence lifetime imaging. The local depletion of the effective phosphoinositides at the plasma membrane will be assessed by lipid



mass spectrometry that we have developed in our laboratory. PKB/Akt activation state determinant and other downstream effectors of growth factor receptors in breast and colon carcinoma Molecular targets in cancer therapy are identified by the relationship between their expression, pathogenesis and clinical outcome. Therefore in a clinical environment, the quantitative assessment of the degree of target molecule expression may provide an indication of both prognosis and whether or not a patient is to benefit from the targeted treatment. In tumours, the activation status of proteins, rather than their over expression, would enable a more precise predictive parameter for determining the status of the target. Actual methods for assessing oncoprotein activation are based on immunohistochemistry, which is limited by the subjectivity of manual scoring and poor specificity reflected in the use of one-site assays. To address these points we have developed a generic high throughput methodology that combines the spatio-temporal and quantitative attributes of time resolved FRET detected by multiple frequency domain lifetime imaging with the sensitivity of the a signal amplification system. We were able to identify molecular heterogeneity of an PKB/Akt, between different regions of interest within the same tumour core and between various cores within the same patient. Our findings suggest that the coincidence amplified-FRET assay is a highly sensitive and specific assay for the detection of molecular heterogeneity that benefits from a high and quantifiable dynamic range. This new methodology will have direct clinical applications in guiding targeted therapeutics and in developing companion diagnostic tools.

Publications listed on page 139

LINCOLN’S INN FIELDS

47

COMPUTATIONAL BIOLOGY www.london-research-institute.org.uk/research/nicholas-luscombe

Cellular life must recognise and respond appropriately to diverse internal and external stimuli. By ensuring the correct expression of specific genes at the appropriate times, the transcriptional regulatory system plays a central role in controlling many biological processes: these range from cell cycle progression and maintenance of intracellular metabolic and physiological balance, to cellular differentiation and developmental time-courses. Group Leader

Nicholas Luscombe Postdoctoral Scientists Borbala Gerle Anabel Prieto Alessandra Vigilante Katharina Zarnack Judith Zaugg Graduate Students Filipe Cadete Maria Dermit Robert Sugar Scientific Officer Inigo Martincorena

Numerous diseases result from a breakdown in the regulatory system and a third of human developmental disorders have been attributed to dysfunctional transcription factors (TFs). Furthermore, alterations in the activity and regulatory specificity of TFs are now established as major sources for species diversity and evolutionary adaptation. Indeed, increased sophistication of the regulatory system appears to have been a principal requirement for the emergence of metazoan life. Much of our basic knowledge of transcriptional regulation has derived from molecular biological and genetic investigation. In the past decade, the availability of genome sequences and development of new laboratory techniques have generated (and continue to generate) information describing the function and organisation of regulatory systems on an unprecedented scale. Genomic studies now allow us to examine regulatory systems from a whole-organism perspective; on the other hand however, many observations made with these data are unexpected and appear to complicate our view of gene expression control. The continued flood of biological data means that many interesting questions require the application of computational methods to answer them. The combination of computational biology and genomics enables us to uncover general principles that apply to many different biological systems; any unique features of individual systems can then be understood within this broader context. The Computational Biology Group applies computational and genomic methods to answer

48

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

three main questions: (i) How is gene expression regulated? (ii) How do these mechanisms control interesting biological behaviours? (iii) How does gene regulation interact with evolutionary processes? Much of our work until now has been purely computational, either analysing publicly available data or in collaboration with experimental laboratories performing functional genomic investigations. Research highlights Evolution of local mutation rates. By comparing the sequences of 34 E. coli genomes, we demonstrated that local mutation rates have been evolutionary optimised (Martincorena et al., 2012; Nature. 485(7396): 95-8; Martincorena and Luscombe, BioEssays. In Press). Mutation rates follow a ‘risk management’ strategy, prioritising the cells’ resources to protect ‘important’ genes. The study settled a 50-year old question and revised a central tenet of evolutionary theory. It also raised intriguing new questions about how DNA-repair mechanisms are deployed across the genome. Epigenetic mechanisms for dosage compensation. In collaboration with the Akhtar laboratory (Max Planck Institute for Immunobiology and Epigenetics), we dissected chromosome-wide mechanisms involved in the precise two-fold up-regulation of male X-linked genes during fly dosage compensation. We discovered the first context-dependent activity of a histone acetyltransferase (Kind et al., 2008; Cell. 133(5): 813-28), and we reported the one of the first

laboratory, we developed nucleotide-resolution, genome-wide techniques to identify protein-RNA interactions (Konig et al., 2010; Nat Struct Mol Biol. 17(7): 909-15). We demonstrated how hnRNP C binds to enhanced and repressed splice sites. Recently, we discovered how competitive binding between U2AF65 and hnRNP C at protects the transcriptome from the detrimental exonisation of thousands of Alu elements (Zarnak et al., Cell. In Press).

Figure 1 A schematic diagram showing that neutral mutation rates in the E. coli genome are heterogeneous and non-randomly distributed.

genome-wide roles of nucleoporin subunits in gene regulation (Vaquerizas et al., 2010; PLoS Genetics. 6(2): e1000846). Recently, we accurately measured the two-fold increase in Pol II-binding at promoters, demonstrating that dosage compensation is controlled by enhanced initiation (Conrad et al., 2012; Science. 337(6095): 742-6). Statistical models of gene expression in fly development. Using compiled in situ hybridisation images from the Virtual Embryo dataset, we developed statistical models that for the first time accurately reproduce even skipped expression. Importantly, the models precisely forecast behaviours beyond the training data, making them truly predictive (eg, effects of regulatory perturbations). The study generated experimentally testable hypotheses and provided new insights into the underlying mechanisms of transcriptional regulation (Ilsley et al., Submitted). Prevention of aberrant exonisation of Alu elements. In collaboration with Jernej Ule’s Figure 2 Predicted expression of eve 2 in A. wild-type and B. gt mutant. C. In situ hybridisation image of gt mutant (From Small et al., 1992; EMBO J. 11(11): 4047-57).

A

B

C

Future work Nuclear organisation of chromosomes. It is increasingly appreciated that the spatial organisation of chromosomes profoundly influences gene expression; however the details of how this is achieved are poorly understood. We will build on our successful collaborations with the Akhtar laboratory to study the effects of X-chromosomal positioning on dosage compensation. Excitingly, we recently initiated collaborations with Peter Fraser (Babraham Institute), a world-expert on ChIA-PET and Hi-C, to investigate nuclear organisation in mammalian cells. Gene regulation in disease states. We will apply our basic knowledge of gene regulation to disease systems. There are indications that bacterial infections cause changes to the host’s regulatory system, so affecting expression patterns. We have initiated collaborations with Richard Hayward (Department of Structural and Molecular Biology, UCL) to apply genomic techniques to investigate the prevalence of these effects, and the influence they have on the progression of bacterial infections. Gene regulation and DNA-damage repair. A major implication of our mutation rate study is that highly expressed genes are preferentially protected from DNA damage; however mechanisms such as transcription-coupled repair do not explain our observations. There are early indications that similar mechanisms operate in cancer. DNA damage repair is traditionally studied from a molecular perspective: we will collaborate with LRI laboratories examine this phenomenon from a genomic viewpoint also. This will dramatically improve understanding of how DNA damage repair operates on a genome-wide scale.

Publications listed on page 140



LINCOLN’S INN FIELDS

49

LYMPHATIC DEVELOPMENT www.london-research-institute.org.uk/research/taija-makinen

Group Leader

Taija Makinen Postdoctoral Scientists Ines Martinez-Corral Andrea Taddei Florence Tatin Graduate Students Lukas Stanczuk Andres Vicente Scientific Officer Sherry Xie

The lymphatic vasculature constitutes an intricate network of vessels critical for fluid homeostasis, immune surveillance and fat absorption. In cancer, metastatic tumour cells can exploit the lymphatic vasculature and spread via the lymphatic vessels to lymph nodes. Our goals are to understand the mechanisms regulating developmental and pathological lymphangiogenesis. This knowledge will be instrumental in designing pro- or anti-lymphangiogenic therapies for lymphatic disorders. The mature lymphatic vasculature consists of two differentiated vessel types, lymphatic capillaries and collecting vessels, which have distinct functions in the uptake and transport of lymph, respectively. Failure of the lymphatic vessels, caused by a genetic defect, malformation or damage following surgery or radiation therapy, can lead to lymphoedema, which is a progressive and lifelong condition characterised by gross swelling of the affected limb. For example, women who have been treated for breast cancer, and in particular those who receive axillary lymph node dissection and/or axillary radiation therapy, are at risk for developing lymphedema of the arm. Although some progress has been made in pre-clinical studies of curative therapies for lympohedema, the currently available treatments can only manage the symptoms, for example, by manual drainage, massage or compression garments. The ability of the lymphatic vessels to generate and maintain lymph flow relies on contractions of smooth muscle cells (SMCs) in the walls of the collecting vessels and the action of luminal valves that prevent backflow of lymph. We have previously characterised lymphatic valve morphogenesis and identified new molecular regulators of this process. This year, we completed a study aimed at gaining a better understanding of the regulation of lymphatic vessel wall assembly and lymphatic pumping (Lutter et al., 2012; J Cell Biol. 197: 837-49).

50

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

The role of smooth muscle cells in the formation of functional lymphatic vessels In order to better understand how functional collecting lymphatic vessels form, we first established a timeline and a defined set of markers of lymphatic collecting vessel differentiation. We found that SMC recruitment onto the developing collecting lymphatic vessels coincided with downregulation of LYVE-1, a marker of lymphatic capillaries, and assembly of vascular basement membrane. Using Affymetrix expression analysis, we then identified specific extracellular matrix components of the lymphatic in comparison to blood vascular basement membrane. Among 14 identified genes we found Reelin, a large glycoprotein previously implicated in the development of the nervous system. Congenital lymphedema observed in some human patients carrying mutations in RELN gene suggested a function also in lymphatic development, making Reelin an interesting gene for further studies. To investigate the physiological function of Reelin in lymphatic development we analysed the vasculature of Reelin deficient mice. We found that the initial lymphatic vessel networks formed normally in the mutants, however, their maturation into functional collecting vessels was defective. Reelin mutant vessels maintained an abnormally high level of LYVE-1 expression and showed reduced SMC coverage when compared to wild type collecting vessels (Figure 1). In addition, the mutant vessels were enlarged,

Figure 1 Confocal tile scan images of ear whole mounts showing dermal collecting lymphatic vessels in wild type and Reelin mutant mice. The tissue was stained for alphaSMA (red, smooth muscle cell marker), LYVE-1 (green, lymphatic capillary marker) and podoplanin (blue, lymphatic vessel marker). Overlapping blood vessels were masked to highlight collecting vessels only. (From Lutter et al., 2012).

displayed irregular vessel diameters and had impaired function. Our further studies showed that SMC contact stimulated release and proteolytic processing of endothelium-derived Reelin. Lymphatic endothelial cells in turn responded to Reelin by upregulating the expression of MCP1/CCR2, a known regulator of SMC recruitment and migration. This suggests an autocrine mechanism for Reelin in controlling endothelial expression of factors involved in SMC recruitment. Together, our results show a critical function for Reelin in the formation of functional lymphatic vessels and uncover a mechanism by which Reelin signalling is activated by communication between the two cell types of the collecting lymphatic vessels, smooth muscle and endothelial cells. They also reveal a previously unrecognised important function for SMCs in regulating lymphatic vessel morphogenesis and function.



Identification and functional characterisation of novel causative genes for primary lymphoedema Identification of genetic mutations in human hereditary lymphoedema and functional characterisation of these genes using various animal models have greatly increased our understanding of normal lymphatic development. Our collaborators at St George’s Hospital in London recently identified GATA2 and KIF11 as novel causative genes for hereditary lymphoedema. We have generated mouse models for these genes in order to investigate their function during different stages of development and in adult mice. We expect that characterisation of these mice will provide novel insights into normal lymphatic development but also into pathophysiological mechanisms involved in lymphoedema.

Publications listed on page 140

LINCOLN’S INN FIELDS

51

TUMOUR HOST INTERACTION www.london-research-institute.org.uk/research/ilaria-malanchi

A cancer cell surrounded directly by the normal host environment of a tissue would be unable to develop into a tumour. Only by modifying their environment can cancer initiating cells survive, proliferate and build a supportive stromal structure consisting of tumour associated host cells locally or systemically recruited. The aim of our lab is to understand the crosstalk tumour cells establish with the host organism to allow both tumour onset and metastatic progression. Group Leader

Ilaria Malanchi Postdoctoral Scientist Luigi Ombrato Graduate Students Yaiza Martin del Pozo Stefanie Wculek Scientific Officer Robert Moore

The functional association of cancer cells with their surrounding tissues forms a complex structure reminiscent of an ‘organ’ that changes as malignancy progresses. Tumours, similarly to a normal organ of the body, retain only a small fraction of cells, termed cancer stem cells (CSCs), with the potential of sustaining long-term growth. They account for a small percentage of the total cancer cell population and are the precursor of all cancer cells with the ability to replace part of or the entire tumour if required. The central role of CSCs appears to be maintained during tumour progression. We provided direct evidence that CSCs drive metastatic colonization and, similarly to normal tissue stem cells, rely on signals derived from their microenvironment for controlling their fate decision, behaviour and maintenance. This complex microenvironment is named ‘niche’ (Figure 1). We aim to clarify how the tumour creates its niche and how these required extrinsic signals impact on cancer cell intrinsic pathways and behaviour. Tumor-host interactions extend well beyond the local tissue microenvironment and tumours not only respond to, but also actively perturb host organs at distant anatomic sites. For instance, inflammatory cells, once assumed to act only to attenuate tumour development, clearly play a role in tumour promotion and malignancy and indeed inflammatory conditions are able to increase the speed of tumour progression. Furthermore, tissue injury has an important role in the onset of malignant diseases, with chronic inflammation being the most well known critical carcinogenetic risk factor. Part of our lab focuses in particular on components of the immune system as part of the tumour niches that favour onset of tumour and metastatic establishment.

52

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Cancer stem cells and metastatic niche The ability of cancer cells to induce a permissive environment discriminates successfully metastasising cells from the unsuccessful ones (Figure 1). Which signals from the niche are essential for metastasising cells? To answer this question it is important to consider that cells successfully metastasising belong to the cancer stem cell population of the tumour. In a spontaneous metastasising breast tumour model, we observed about 20-fold increase in the percentage of CSCs in the initial phase of metastatic colonization. Importantly, the CSC enrichment in this initial phase is transitory. During metastatic progression the CSC population will reduce its relative amount and fully blown metastasis will maintain the same CSC frequency as the primary tumour. This indicates a cross talk with the rest of the tumour mass to regulate CSC number. We are exploring the cross talk between CSCs and host cells by generating gene expression profiles. We have produced a CSC gene expression signature and compared it to the early and late phase of metastatic establishment to the lung. In line with the previously observed CSC enrichment, cluster analysis shows an overall similarity between the early metastasis and the CSC up-regulated genes (Figure 2). We have selected some of those matching genes to functionally test their impact on metastatic potential in vivo and CSCs potential in vitro. Cancer stem cells intrinsic signals We also gained evidence that Wnt signalling is a specific essential intrinsic regulator of CSC function. In tumours as well as in metastasis Wnt activity is primarily concentrated in the CSC population. We are now investigating the

molecular regulation of this signalling pathway in CSC maintenance and self-renewal at the metastatic site by generating the gene profile of cancer cells with or without active Wnt signalling. Cancer cells at the metastatic site will be discriminated for their Wnt pathway activation using an in vivo reporter construct. Pro-tumorigenic activity of neutrophils The pro-tumorigenic and pro-inflammatory activity of some innate immune cells has been characterized, however, few studies concentrated on neutrophils. Neutrophils are the first cells that migrate towards the inflammatory site and are crucial to amplify the inflammation response. We found these cells present in at the primary tumour site and at the metastatic site even before metastatic establishment (Figure 1). We propose a direct role of neutrophil recruitment in tumour initiation and metastasis. Transgenic mice completely resistant to inflammation-dependent

skin carcinogenesis, showed a strong lack of neutrophil recruitment. Tumorigenesis could be restored when skin recruitment of neutrophils was enforced in the skin. Furthermore, in the absence of neutrophil recruitment, chemically induced lung carcinomas are strongly reduced in number and size. Finally, in a model of mammary gland tumours, the metastatic lungs show a 10 to 20-fold increase in neutrophils compared to control lungs. Remarkably, failure of neutrophil recruitment resulted in a strong reduction in lung metastasis. Using genetic and chemical strategies to block neutrophils in tumour and metastatic models we aim to clarify the tumour promoting functions of neutrophils. We are now investigating a direct effect of neutrophils on the different tumour cell populations as well as indirect neutrophildependent modulations over other immune-cell components of the tumour microenvironment.

Figure 1 Schematic representation of the stepwise tumour evolution with the surrounding microenvironment (niche) at the primary site (upper panel) and at the metastatic site (lower panel). Note the presence of systemic changes at distant site (lower panel) after primary tumour establishment defining a pre-metastatic micro-environment. These changes impact metastatic outcome.

Figure 2 A. Design of the microarray analysis: cancer cells were isolated from the metastatic site after one week (colonization phase, enriched in CSCs) or after four weeks (metastatic phase). Gene expression profiles were compared with the primary tumour Cancer Stem Cells (CSCs) gene profile. B. Cluster analysis of the indicated expression profiles.



LINCOLN’S INN FIELDS

53

STRUCTURAL BIOLOGY www.london-research-institute.org.uk/research/neil-mcdonald

Group Leader

Neil Q McDonald Postdoctoral Scientists Charlie Nichols Ivan Plaza-Menacho Peter Saiu Erika Soriano Graduate Students Emily Burns Kerry Goodman Marina Ivanova Scientific Officers Maureen Bowles Phillip Knowles John Lally Judith Murray-Rust Andrew Purkiss

Growth factor pathway deregulation is a hallmark of cancer and is frequently associated with altered catalytic functions of pathway components. Our research aim is to understand the structural biology underlying the assembly, activation and deregulation of growth factor signalling complexes. Our current research investigates how signals arriving at the cell membrane stimulate the RET receptor tyrosine kinase, how protein partners and substrates are selected by the atypical PKC kinases close to the plasma membrane, as well as the basis for growth factor-stimulated transcriptional responses. The lab uses both structural (crystallography, EM and SAXS) and biophysical methods (calorimetry and fluorescence polarisation), complemented by both enzymatic and cell-based assays, in pursuit of our research goals. One significant research highlight this year was our publication of a helical assembly comprised of three G-actin molecules and the RPEL domain from the PP1-binding protein, Phactr1 (Mouilleron et al., 2012; Structure. 20: 1960-70). Cell surface receptor activation and oncogenic deregulation We are studying the RET proto-oncogene, a transmembrane-spanning receptor tyrosine kinase that is activated upon binding a GDNF family ligand and a co-receptor (GFRα). RET is crucial for embryonic development and its mutation underlies three human diseases (Hirschsprung’s disease, kidney agenesis and cancer). Our current activities are focused on a detailed structural characterization of the architecture of the RET extracellular domain by using a combination of crystallography and low angle X-ray scattering. We wish to understand how both loss-of-function (l-o-f) and gain-offunction disease mutations perturb RET structure and activation. More recently we have shown that the first two (of four) cadherin domains adopt a clamshell arrangement containing residues that constitute a folding bottleneck for the entire receptor. Elimination of just two amino acids from this region of RET alleviates its intrinsic ‘poor’ folding properties and rescues the majority of l-o-f

54

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Hirschsprung’s disease mutations such that they can be relocated to the plasma membrane where they can bind ligand/co-receptor. To explore the folding bottleneck within RET further, we have identified and are characterising proteins that are able to either enhance or restrict the maturation and export of RET to the plasma membrane. Efforts to reconstitute vertebrate RET ternary complexes containing both ligand and co-receptor have been successful, permitting structural analyses of the complex to identify ligand-dependent conformational changes arising upon ternary complex formation. A consequence of RET ligand occupancy and ‘activation’ is the trans-phosphorylation of internal tyrosine sites within the tyrosine kinase catalytic domain. Our careful analysis of the kinetics and order of tyrosine autophosphorylation within the intracellular domain of wild type and oncogenic RET mutants reveals surprising differences suggesting how autoinhibitory mechanisms are subverted by RET oncogenic mutation.

Polarity signalling assemblies at the plasma membrane Atypical PKC isoforms associate with the polarity proteins Par-3 and Par-6 to form a conserved apical polarity complex (PAR complex), found in both vertebrates and invertebrates. In epithelial cells, the PAR complex (Par-3/Par-6/aPKC) is localised asymmetrically within the apical, rather than the basolateral membrane and is critical for the formation of adherens junctions. Defects in the regulation of apical-basal polarity are associated with both tissue overgrowth and tumourigenesis and are consistent with an oncogenic role for aPKC as well as tumour suppressor roles for other polarity determining factors. We have been interested in how the aPKC isoforms (iota and zeta) are regulated and how they selectively target their substrates. The PAR complex phosphorylates substrates such as Par-3,

Figure 1 Phactr-1 (top– solid surface) and MRTF-A (bottom – solid surface) form closed helical assemblies with multiple G-actin molecules (ribbons) via their RPEL motifs (red surface), thereby restricting access of their intrinsic nuclear localisation sequences (dark grey surfaces) to importins, resulting in a cytoplasmic location.

A

B

Kibra, LLGL2 and Numb leading to their dissociation from the membrane. However some of these same aPKC substrates are able to bind and directly inhibit its serine/threonine kinase catalytic activity suggesting a dual action. However, many of the molecular details underlying the inhibitory or substrate-driven interactions remain obscure at present. To understand the dual action function of aPKC partners, we have reconstituted the PAR complex and sub-complexes to determine the regulatory effect of partner proteins on aPKC kinase activity. Our structure of a PAR sub-complex indicates how Par-3 can function as both a high affinity inhibitor and as a low affinity substrate. We are currently probing the in vivo role of these functions in cells and in flies through collaborations with both the Protein Phosphorylation and Epithelial Biology Groups to validate our in vitro findings. Propagation of growth factor-initiated signals into the nucleus Growth factor signals alter actin dynamics and thereby influence many aspects of cell shape and cell motility. Signalling to actin-binding protein regulators leads to a dramatic fluctuation in the levels of monomeric actin (G-actin) following polymerisation to F-actin. Recent studies have identified a molecular G-actin sensor, called the RPEL domain, which contains a tandem array of three RPEL motifs, each of which is competent to bind a G-actin molecule. The domain is present in two otherwise unrelated protein families; the MRTF family of serum response factor (SRF) transcriptional co-activator proteins and the Phactr family of actin and PP1 phosphatase-binding proteins. We are interested in how growth factors alter the nuclear import and functional properties of both the MRTF and Phactr families. Fluctuating G-actin levels influence both the stoichiometry of G-actin molecules loaded onto the RPEL proteins and their subcellular localisation. We have collaborated closely with the Transcription Group to determine crystal structures of multivalent G-actin complexes of RPEL repeat proteins (Figure 1). These structures form closed helical assemblies with G-actin subunits regularly spaced around the RPEL domain crank to make cooperative interactions. These G-actin assemblies form an unprecedented arrangement compared to all previous actin-binding protein complexes, including F-actin. Moreover, our structures explain how G-actin interaction alters the subcellular localisation of MRTF-A, by inhibiting nuclear import in ‘unstimulated’ cells through competing with importin α-β binding. Growth factor stimulation leads to a drop in G-actin levels with the consequence that RPEL protein nuclear localisation sequences become accessible to importins resulting in net migration into the nucleus.

Publications listed on page 141



LINCOLN’S INN FIELDS

55

CELL CYCLE www.london-research-institute.org.uk/research/paul-nurse

Group Leaders

Paul Nurse Jacqueline Hayles Research Co-ordinator Ryoko Mandeville Postdoctoral Scientists Pilar Gutierrez-Escribano Kazunori Kume Francisco Rivero Navarro Louise Weston Graduate Students Matthew Swaffer Elizabeth Wood Scientific Officers Linda Jeffery Juan-Juan Li Naomi Moris Visiting Worker Maja Sinn

56

This laboratory investigates the cell cycle, cell form and cellular growth, using the fission yeast Schizosaccharomyces pombe as a model organism. Fission yeast is a single celled organism with a typical eukaryotic cell cycle, a regular cylindrical shape and powerful genetic, genomic, biochemical and cell biological methodologies available for its study. These features make it an ideal model organism for studying these processes. During the cell cycle the genome is accurately replicated and segregated to daughter cells, and errors in transmission of the genetic material can lead to mutations and chromosomal rearrangements. In addition failure to accurately reproduce and maintain cell form and regulate cell growth can disrupt tissue architecture and cell function. Study of these processes, which are fundamental to all organisms, is therefore important for understanding the underlying mechanisms that may lead to diseases such as cancer. Cell cycle The major control regulating entry into mitosis in all eukaryotes is triggered by activation of the cyclin-dependent kinase CDK1. CDK1 activity is regulated by phosphorylation/dephosphorylation of a conserved tyrosine 15 residue (Y15) in response to nutritional and cell geometry sensing pathways. Francisco Navarro has carried out a near genome wide screen to identify new elements involved in the regulation of mitotic entry by screening for gene deletion mutants that prematurely enter mitosis at a small cell size (Navarro and Nurse, 2012; Genome Biol. 12: R36). He screened 3000 viable mutants and identified 18 genes that when deleted advanced cells into mitosis at a small cell size, seven of which have not been previously implicated in regulation of the cell cycle (ski3, snf5, sol1, sgf73, pab1, SPBC19F8.02 and SPAC27E2.03c). Several mutants showed only a modest reduction in cell size at mitosis, but by combining different mutations Francisco found that cell length at division was reduced further in an additive manner (Figure 1), suggesting that these mutations may be affecting different pathways regulating the G2/M transition. To determine whether any of these 18 genes acted

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

independently of the known pathways affecting entry into mitosis he carried out pairwise crosses with sty1Δ (nutritional sensing pathway) and cdr1Δ (cell geometry pathway). Two genes nif1 and ski3 acted upstream of both pathways and five genes (ppa2, sol1, snf5, zfs1, and clp1) showed a reduced cell length at mitosis independently of sty1Δ and cdr1Δ. Further analysis of these 5 genes showed that snf5, sol1 and zfs1 all acted independently of CDK1 Y15 phosphorylation; the small cell size was not the result of increased CDK1 complex levels or of Rum1 (CDK1 inhibitor) deregulation. Although the regulation of entry into mitosis has been studied extensively for many years this genome wide screen has identified new pathways that affect the timing of mitosis independently of known pathways and Y15 phosphorylation of CDK1. Cell form To investigate how cells form a polarised growth zone of the correct size Felice Kelly (Rockefeller) examined de novo growth zone formation using fission yeast spheroplasts. Spheroplasts have no cell wall, have a round shape with depolarised growth proteins, and have a disorganised

14.3+/-0.9

11.4+/-0.5

9.6+/-0.6

Wild type

ski3Δzfs1Δ

ski3Δzfs1Δppa2Δ

Figure 1 Divding cells carrying additive combinations of mutations that confer a small cell phenotype. Cell wall is stained with Blankophor and cell length at division in microns is indicated.

cytoskeleton. On recovery they become polarised and form a normal rod shaped cell of the correct width (Kelly and Nurse, 2011; Mol Cell Biol. 22: 3801-11). These features make them a good model for examining de novo formation of a growth zone. As spheroplasts recovered a polarised patch formed in the round spheroplast, demonstrating that a growth zone can organise dependently of cell shape. Time lapse microscopy following Rga4 (a Cdc42 GAP) and Scd2 (a Cdc42 scaffold protein) looked at the sequential organisation of the newly formed growth zone. Rga4 was initially localised throughout the spheroplast membrane, Scd2 first formed a stable patch in the same location as the future growth zone and this was followed by exclusion of Rga4 from this region. In the absence of Rga4 or Scd2 a growth zone could still form indicating that these genes are not required for polarization, however the size of the growth zone was wider than normal. Rga4 and Scd2 are localised to the growth zone in exponentially growing cells and potentially could carry the memory of the position of the growth zone in regenerating spheroplasts. However Felice found that this is unlikely as neither of these proteins remained localised to the cell membrane during spheroplast formation. These results indicate a potential role for Cdc42 activation in the formation of a growth zone of the correct size independently of pre existing cell shape.



8.4+/-0.5

ski3Δzfs1Δ ppa2Δsnf5Δ

7.2+/-1.3

ski3Δzfs1Δ ppa2Δsnf5Δ clp1Δ

Genomics and chemical biology Osami Niwa, a visiting worker in the lab, used the genome wide gene deletion collection to further investigate the regulation of cell ploidy (Tange et al., 2012; PLoS Genet. 8: e10002776). He screened the deletion library to identify genes important for the viability of aneuploid spores generated by triploid meiosis or DNA damage. Aneuploid spores can remain viable for a period after germination during which the effects of aneuploidy may be reduced thus allowing colonies to form. He found that a conserved regulator of mRNA turnover, CCR4-NOT complex was important for reducing the effects of aneuploidy. Chemical inhibitor molecules are useful for analysing various cellular processes. In fission yeast the use of chemical inhibitors has been limited by a very effective multidrug resistance response. To overcome this Ai Takemoto in collaboration with Tarun Kapoor’s laboratory has constructed and tested a multidrug sensitive fission yeast strain (Kawashima et al., 2012; Chem Biol. 19: 893-901). Five genes have been deleted, two ABC transporters encoded by brf1 and pmd1, two MFS transporters encoded by msf1 and caf5 and the transcription factor encoded by pap1. The strain was found to be sensitive to a range of diverse chemical inhibitors and will be useful for analysing the effects of chemicals on fission yeast cells. Publications listed on page 141

LINCOLN’S INN FIELDS

57

PROTEIN PHOSPHORYLATION www.london-research-institute.org.uk/research/peter-parker

Group Leader

Peter Parker Clinical Scientist Julien de Naurois Postdoctoral Scientists Nicola Brownlow Tanya Pike Philippe Riou Marta Sanz-Garcia Frances Willenbrock Graduate Students Jeroen Claus Nick Peel Yixiao Zhang Scientific Officers Angus Cameron Richard Whelan Masters Student Naveen Joshi

The existential behaviour of transformed cells is a function of somatic change. These changes determine dysfunctional cell autonomous properties, as well as aberrant responses to, or ignorance of, the host environment. The manner in which these behaviours are executed reflect altered functions (gains and losses) typically associated with cellular regulatory networks. The clonal retention, gain and loss of function of individual regulatory elements reflecting the competitive growth and survival pressures associated with the tumour niche. Seeking out the drivers for the retained/ emergent properties, and where and when they become hard-wired as others are lost, provides a basis on which to seek out targets for intervention. The protein kinases remain one of the most important regulatory elements in physiology and pathophysiology, with cancer-associated somatic mutations being disproportionately associated with these proteins. Amongst these transducers, the Protein Phosphorylation Laboratory has focused on the PKC superfamily, which represents >2% of the human kinome. Our work has been directed towards particular cancer-associated properties including: aberrations in cell division, survival controls and migration/invasion. In these areas we have investigated the roles of individual PKC family members and in particular are concerned with molecular mechanisms. The impact that our understanding has on opportunities for intervention and on the wider cancer signalling field are predominant in our thinking. Cell division We have pursued the observation that PKCε loss or inhibition triggers cytokinesis failure in certain model systems. Screening to assess potential mechanisms has identified PKCε substrates that may contribute to this behaviour. These have been identified by candidate approaches and PKCε phosphorylation sites mapped using arrays. Antisera directed at these sites have been raised to assess the circumstantial evidence for involvement in the control of cytokinesis. In parallel we have generated mutant, phosphorylation incompetent and

58

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

phosphomimetic, forms of the candidates to assess more directly whether their phosphorylation may be involved in this control. The frequency of failure at cytokinesis is exacerbated by PKCε loss or inhibition and we have identified an earlier step in cell division that appears to be responsible. Notably, we can inhibit acutely PKCε at this earlier stage using a relatively selective inhibitor and demonstrate a requirement for the endogenous protein, corroborating what has been observed for the ectopically expressed gatekeeper mutant that is sensitive to the drug NaPP1. We are currently working to determine what role PKCε plays in mitosis and how this is coupled to its requirements in cytokinesis. Finally we continue to develop insight into the specificity of this behaviour in light of the relative normality of the knock-out mouse model. The potential for selective intervention with a very favourable therapeutic index is an extremely attractive prospect. Apoptosis controls There is a wealth of evidence for roles of PKC isoforms in the control of survival pathways, typically involving anti-apoptotic inputs and engaging through diverse mechanisms. For example, PKCα in glioblastoma cells exerts an anti-apoptotic effect that appears to be

independent of its catalytic activity. For PKCδ there is evidence that it can both inhibit apoptosis as well as promote apoptosis, depending on circumstances. We are particularly interested in understanding how this protein kinase integrates contextual signalling information to switch between these extreme cellular decisions. Within the same model system we are mapping events associated with these polarised aspects of PKCδ behaviour, to define at a molecular level the alterations in PKCδ that confer this switching behaviour. This will enable us to trace back up the input pathways to define how context modifies PKCδ behaviour to exert these distinct outputs. Migratory pathways In our collaborative work on HGF-dependent migration, (Dr Stephanie Kermorgant; Queen Mary University London), mechanistic insights have developed from an siRNA library screen for which a series of hits were validated. In parallel we have identified distinct mechanisms involved in spatially separated signals directed at the actin cytoskeleton. The evidence indicates that there is spatial and temporal regulation of receptor coupling events, potentially enabling distinct controls to operate from different compartments to which the ligand bound receptor, c-Met, is trafficked. Notably, during the course of analysing the regulators of HGF-induced c-Met traffic through these compartments, we found that the kinesin inhibitor ATA blocks HGF induced signals and that unexpectedly, this is directed at c-Met itself through what appears to be an unusual allosteric mechanism. The behaviour of this inhibitor suggests that there may be opportunities to develop ATP non-competitive inhibitors of c-Met.

Work on the aPKC-Exocyst>ERK promigratory pathway has yielded a series of candidate targets downstream of plasma membrane ERK. We are following up on a number of these with a particular view towards how they impact on focal adhesion turnover, a property well-established on this aPKC pathway. Related to this, we have identified Dynein Intermediate Chain 2 (DIC2) as a downstream target of aPKC in these migratory models and shown that it interacts with paxillin in a manner controlled by an aPKC dependent phosphorylation. DIC2 appears to play a crucial role as part of the dynein-dynactin complex in the regulation of focal adhesion turnover downstream of aPKC (Figure 1). Building on our work with the PKN family in migratory behaviour, we have completed the generation of the family of knock-outs. We are currently assessing phenotypes for these. PKC structure, function and intervention Work continues in collaboration with CRT on the development of aPKC inhibitors. This includes structural work with Neil McDonald (Structural Biology Group), functional studies in vivo with Banafshe Larijani (Cell Biophysics Group), alongside mechanistic studies and substrate/ biomarker screens. In the last context, we have identified multiple partners and substrates for aPKCs and are working up a range of reagents and assays to determine utility as biomarkers and to understand involvement in the many outputs of the aPKCs.

Publications listed on page 141

Figure 1 Dynein Intermediate Chain 2 (DIC2) is involved in aPKC-dependent focal adhesion turnover. The working model depicts the delivery of the dynein-dynactin complex to a focal adhesion complex where DIC2 interacts with the focal adhesion protein paxillin. Consequent to this, the focal adhesion is invaginated/endocytosed and/or disassembled. The subsequent phosphorylation (‘P’) of DIC2 through an aPKC dependent mechanism permits DIC2 to recycle and effect further focal adhesion turnover.



LINCOLN’S INN FIELDS

59

MOLECULAR ONCOLOGY www.london-research-institute.org.uk/research/gordon-peters

Group Leader

Gordon Peters Postdoctoral Scientists Ana O’Loghlen Richard Palermo Helen Pemberton Graduate Student Hollie Chandler Scientific Officer Sharon Brookes

We have a long-standing interest in the INK4a-ARF-INK4b tumour suppressor locus and its role in the implementation of senescence, the state of profound growth arrest that is elicited by various forms of cellular stress. In recent years, the emphasis of our research has shifted from the functions of the INK4a-ARF-INK4b gene products to the mechanisms that regulate their expression, and in particular the role of Polycomb group (PcG) proteins. PcG complexes are important for establishing the patterns of gene expression in each cell and are critical for the maintenance of pluripotency. They are also implicated in cancer, at least in part via suppression of INK4a-dependent senescence. Curiously, the INK4a locus is regulated by an extensive variety of PcG complexes and efforts to elucidate the underlying mechanisms have drawn us to genome-wide approaches in which we are considering fundamental aspects of PcG function.

Multiple PRC1 complexes bind simultaneously at the INK4a locus PcG proteins participate in multi-component complexes that modify and bind to histone tails. We are primarily interested in Polycomb repressive complex 1 (PRC1) because of genetic and biochemical evidence implicating several of its components in the regulation of INK4a. In Drosophila, the canonical PRC1 complex contains equimolar amounts of the Polycomb (Pc), Posterior sex combs (Psc), Polyhomeotic (Ph) and Sex combs extra (Sce) proteins but as each has several orthologues in mammalian cells, there can be many different permutations of PRC1. The reasons for the expansion remain unclear as there is currently little evidence for functional or target specificity. By developing antibodies that support chromatin immunoprecipitation (ChIP) of the endogenous proteins, we are now able to locate four members of the Pc family, two Psc proteins, all three Ph orthologues and both Sce proteins at specific loci. Importantly, we find that all 11 proteins associate with the INK4a locus in human diploid fibroblasts (HDF) and that the binding profiles are indistinguishable. Moreover, sequential ChIP

60

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

implies that different members of the Pc, Ph and Sce families, and by implication multiple PRC1 complexes, associate with the same DNA. Genome wide binding profiles of PRC1 complexes To determine whether the situation at INK4a applies to other target genes, we performed ChIP-seq (ChIP coupled followed by DNA sequencing), with the help of the ASF and BABS, in two different strains of HDF. These analyses showed that three Pc orthologues (CBX6, CBX7 and CBX8) and both Sce orthologues (RING1 and RING2) co-localize at several hundred sites in the genome. The binding profiles and positions relative to potential target genes are remarkably varied and although they generally coincide with regions marked by H3K27me3, the PRC1 peaks are more sharply defined and have inherent architecture. In many cases, the profiles are precisely conserved between the two strains of HDF but there are an equivalent number of loci where the binding patterns are distinct. The HOX gene clusters are notable examples (Figure 1) and the data would be consistent with the idea that fibroblasts from different anatomical sites (breast versus foreskin) have characteristic gene expression signatures. Although PRC1 occupancy

Figure 1 A. ChIP-seq profiles of CBX6, CBX7, CBX8, H3K27me3 and H3K4me3 across the HOXD cluster in human fibroblasts from different anatomical sites. B. RNA expression profiles across the HOXD cluster for the different fibroblast strains, as determined by RNA-seq.

A

B

generally equates with reduced expression (as judged by RNA-seq), this not always the case and over 50% of the PRC1-associated genes are detectably expressed. Interestingly, shRNAmediated knockdown of a single PRC1 component causes both positive and negative changes in a subset of the putative target genes. Taken together, our data favour the idea that multiple variants of PRC1 congregate in structures analogous to the Polycomb bodies described in Drosophila. In Drosophila, multiple copies of the canonical PRC1 complex form visible nuclear foci that are responsible for long-range interactions between distant loci. We are currently aiming to consolidate these ideas and to extend the analyses to include different cell types, notably B-lymphocytes and their EBV-infected derivatives. The role of CK2 in Polycomb complexes The catalytic and regulatory subunits of casein kinase II (CK2α, Ck2α’ and CK2β) often co-purify with PRC1 complexes and it is conceivable that CK2 either phosphorylates a component of the PRC1 complex or is used by PRC1 to phosphorylate other chromatin-bound proteins. Co-precipitation and peptide binding assays have confirmed that CK2 associates with all five CBX homologues and in the case of CBX7, this was shown to depend on residues within the conserved Pc box near the C-terminus. Peptide arrays identified three hydrophobic residues that are important for the interaction with CK2. Based on published



crystallographic data, these critical residues are predicted to lie on a surface exposed interface formed by a β-sheet interaction between the C-terminal regions of CBX7 and RING2. Modelling suggested that it should be possible to generate minimally mutated forms of CBX7 that are incapable of binding to CK2. Such mutants would clarify the role (if any) for CK2 in transcriptional regulation by CBX7. However, the mutations tested thus far also abrogate binding between CBX7 and RING2. An alternative hypothesis would be that the association of CK2 with PRC1 relates to DNA repair, as there is accumulating evidence that PcG complexes are recruited to sites of DNA damage. To investigate this possibility, in the context of other work in the lab, we established HDFs that express a tamoxifen-regulated version of the AsiS1 restriction enzyme. Staining for γ-H2AX and 53BP1 confirms that the system operates as expected, both in the fibroblasts and in the U20S cell line commonly used to visualise PcG bodies. However, detection of PRC1 proteins by either immunofluorescence or genome-wide ChIP-seq suggests that they are not recruited to AsiS1 sites in this system and extensive DNA damage does not mobilise PRC1 complexes from known binding sites.

Publications listed on page 141

LINCOLN’S INN FIELDS

61

IMMUNOBIOLOGY www.london-research-institute.org.uk/research/caetano-reis-e-sousa

Group Leader

Caetano Reis e Sousa Postdoctoral Scientists Sophie Acton Safia Deddouche Julie Helft Jatta Houtari Barbara Schraml Annemarthe van der Veen Paul Whitney Santiago Zelenay Graduate Students Susan Ahrens Delphine Goubau Pavel Hanč Janneke van Blijswijk Scientific Officers Sonia Lee Neil Rogers Oliver Schulz

Dectin-1 is a myeloid C-type lectin receptor (CLR) that binds β-glucans on fungal cell walls. We have previously found that Dectin-1 possesses a tyrosine-based ‘hemITAM’ motif in its cytoplasmic tail that allows for direct recruitment and activation of Syk kinase (Rogers et al., 2005; Immunity. 22: 507-17). A few other Dectin-1-like myeloid CLRs possess the same motif and might therefore also signal via Syk (Sancho and Reis e Sousa, 2012; Annu Rev Immunol. 30: 491-529). Over the last year, we have made significant progress on dissecting the role of one such receptor, DNGR-1, in terms of its expression pattern, function and ligand specificity. We initially reported that DNGR-1 (also called CLEC9A) is a DC-restricted CLR recognising an unidentified intracellular ligand that is ubiquitously expressed in all cell types and is exposed upon loss of membrane integrity at the point of cell necrosis (Sancho et al., 2009; Nature. 458: 899-903). We further showed that DNGR-1deficient mice have a defect in crosspriming of cytotoxic T cells against antigens associated with dead cells (Sancho et al., 2009). Our most recent work was aimed at deciphering whether this is because of a unique role of DNGR-1 in decoding the adjuvanticity of dead cells or relates instead to a function of the receptor in regulating antigen extraction from corpses. In addition, we use a multi-pronged experimental approach to identify the DNGR-1 ligand. Finally, we expanded on our observations that DNGR-1 can be used as a marker for mouse ‘CD8α+-like’ DC and can further be utilised to identify their human equivalents. These three lines of research are expounded below. DNGR-1 as a marker of BATF3-dependent DC across species The CD8α+ DC family in mice comprises the CD8α+ DC in lymphoid organs as well as CD8α– cells in non-lymphoid tissues that express CD103 but not CD11b. These two groups of cells share phenotypic and functional similarities, as well as a unique developmental dependence on the transcription factor BATF3. Together with the Haematopoietic Stem Cell Group, we previously characterised DNGR-1+ human DC in human spleen and in the

62

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

spleens of humanised mice which resemble mouse CD8α+ DC in phenotype and function (Poulin et al., 2010; J Exp Med. 207: 1261-71). Other groups found similar cells in human blood and tonsil (Bachem et al., 2010; J Exp Med. 207: 1273-81; Crozat et al., 2010; J Exp Med. 207: 1283-92; Jongbloed et al., 2010; J Exp Med. 207: 1247-60). However, it was not clear whether such cells are also present in human non-lymphoid organs. Furthermore, the identification of ‘CD8α+ DC-like’ cells across different tissues and species remained problematic because of the lack of a unique marker that could be used to unambiguously define lineage members. We have continued to collaborate with the Haemopoietic Stem Cell Group and were recently able to show that mouse CD8α+ DC and mouse CD103+ CD11b – DC share high expression of DNGR-1 (Poulin et al., 2012; Blood. 119: 6052-62). We further showed that DNGR-1 uniquely marks a CD11b – human DC population present in both lymphoid and non-lymphoid tissues of humans and humanised mice (Poulin et al., 2012). Finally, we demonstrated that knockdown of BATF3 selectively prevents the development of DNGR-1+ human DC in vitro (Poulin et al., 2012). Thus, our data indicate that high expression of DNGR-1 specifically and universally marks a unique DC subset in mouse and human (Poulin et al., 2012). Evolutionarily-conserved BATF3 dependence justifies classification of these DNGR-1hi DC as a distinct DC lineage and further suggests that mouse experiments using CD8α+ DC might be directly translateable to man.

Figure 1 DNGR-1 ligand overlaps with the actin cytoskeleton. The image depicts a Drosophila melanogaster ovary surrounded by a smooth muscle sheet. The organ was stained with an extracellular domain of mouse DNGR-1, which highlights the F-actin in muscle striations (red).

DNGR-1 regulates the antigenicity of dead cells Dectin-1 signalling via Syk in response to β-glucan particles potently activates DC and renders them competent to prime T cell responses (LeibundGutLandmann et al., 2007; Nat Immunol. 8: 630-8; LeibundGut-Landmann et al., 2008; Blood. 112: 4971-80). Because DNGR-1 also signals via Syk, we hypothesised that it might function to activate DC in response to encounter with cell corpses. Surprisingly, we found that DNGR-1 signalling does not activate mouse DC or other myeloid cells (Zelenay et al., 2012; J Clin Invest. 122: 1615-27). Rather, its role in crosspriming correlates with its ability to divert necrotic cell cargo into a recycling endosomal compartment, which favours crosspresentation (Zelenay et al., 2012). As a consequence, DNGR-1 regulates crosspriming not only in non-infectious settings such as upon immunization with antigen-bearing dead cells but also in highly immunogenic situations such as after mouse infection with herpes simplex virus type 1 (Zelenay et al., 2012). Similarly, our colleague, David Sancho, showed that loss of DNGR-1 impairs the CD8+ cytotoxic response to vaccinia virus, especially against those antigens and virus strains that are most dependent on cross-presentation (Iborra et al., 2012; J Clin Invest. 122: 1628-43). The existence of a dedicated receptor for crosspresentation of cell-associated antigens and its demonstrable impact on antiviral responses in mice underscores the importance of crosspriming in immunity and suggests that antigenicity and adjuvanticity can in some instances be decoded by distinct innate immune receptors.



DNGR-1 binds to F-actin Our efforts to understand DNGR-1 function were marred by our lack of knowledge of the identity of the DNGR-1 ligand. Initial biochemical characterisation had suggested that the ligand was proteinaceous in nature and that it corresponded to a pre-formed component of healthy cells that was exposed in cell corpses that had lost membrane integrity (Sancho et al., 2009). We therefore went on to set up a biochemical assay and tested for the presence of ligand in cells from different species. This revaled that the DNGR-1 ligand is evolutionarily conserved, being preserved from yeast to man (Ahrens et al., 2012; Immunity. 36: 635-45). We then designed a ligand purification scheme and analysed the results by mass spectrometry. In parallel, we used microscopy to assess the distribution of DNGR-1 ligand in different cells and tissues. The two types of approach gave remarkably concordant results and led to the conclusion that the DNGR-1 ligand corresponds to the F-actin component of the cellular cytoskeleton (Figure 1) (Ahrens et al., 2012). Consistent with that notion, we showed that dead cells retain enough F-actin content to be able to trigger DNGR-1 signalling (Ahrens et al., 2012). The identification of F-actin as a DNGR-1 ligand suggests that cytoskeletal exposure is a universal sign of cell damage that can be targeted by the innate immune system to initiate immunity (Ahrens et al., 2012).

Publications listed on page 141

LINCOLN’S INN FIELDS

63

TUMOUR CELL BIOLOGY www.london-research-institute.org.uk/research/erik-sahai

Group Leader

Erik Sahai Postdoctoral Scientists Fernando Calvo Eishu Hirata Chris Madsen Danielle Park Graduate Students Stefanie Derzsi Nil Ege Cerys Manning Scientific Officers Steven Hooper Robert Jenkins

Research in the Tumour Cell Biology Group is focused on understanding cancer invasion and the tumour micro-environment. We are particularly interested in the molecular mechanisms that cancer cells use to move in complex three dimensional environments and the role of stromal fibroblasts in generating an environment that enables cancer cell invasion. Further, we hope to understand how therapeutic interventions affect these processes, and how they may be more effectively manipulated to improve cancer outcomes.

Cells can physically couple their contractile actomyosin cytoskeleton to matrix adhesions and/ or the plasma membrane. The relative level of coupling to matrix adhesions and the plasma membrane has profound consequences for cell migration. Linkage to the ECM is advantageous for mesenchymal migration and movement on 2D substrates, whereas linkage to the plasma membrane can generate hydrostatic pressure that aids cell migration in other contexts. Chris Madsen has been working to understand the mechanism that determines what the contractile actomyosin cytoskeleton is linked to. Working together with the Epithelial Biology Group he has uncovered many novel regulators of cell migration. These molecules are important during embryonic development and also for cancer cell invasion. Chris has identified an important role for phosphatases in regulating Myosin Light Chain and Ezrin, Radixin, Moesin family proteins. These molecules are critical for actomyosin contraction and actin-plasma membrane linkage, respectively. The mechanism co-ordinately inactivating them through dephosphorylation is critical for migratory and metastatic behaviour of breast cancer cells. Eishu Hirata and Cerys Manning have been working on identifying the signalling pathways that are upregulated in invasive cancer cells. They have been utilising a combination of reporter construct and FRET-based approaches to monitor the activity of signalling pathways. This has enabled changes in cell signalling to be observed in vivo with cellular resolution. Cerys has then combined insights from intravital imaging with

64

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

fluorescence activated cell sorting to purify invasive sub-populations of cells and performed genomic analyses. This work has revealed that an interesting role for chromatin modifying enzymes in regulating the invasive phenotype in melanoma models. Eishu has used similar approaches in combination with administration of chemotherapeutic agents. He has also exploited the use of titanium imaging windows to obtain longitudinal data about signalling in tumours before and after the administration of therapeutic drugs. This has revealed that invasive cells are resistant to therapeutic agents. Interestingly, this resistance does not seem to be an intrinsic property of the cancer cells, but is conferred by micro-environmental factors that are found in vivo. Cerys has used imaging windows in a similar manner to study the heterogeneous response of tumour blood vessels to anti-angiogenic drugs. This has revealed that resistance to antiangiogenic therapy may result from pre-existing microenvironments, and not as an adaptive response to the therapy. In many cases, cancer cells move along or through physical tracks in the extra-cellular matrix that are generated by stromal cells. Cancer-Associated Fibroblasts (CAFs) are able to generate these tracks through the extra-cellular matrix. Fernando Calvo has isolated CAFs from different stages of murine breast cancer and normal non-malignant mammary glands. He has used these to learn about the factors that make CAFs distinctive. He has identified several signalling pathways that are constitutively activated in CAFs. In particular, numerous inflammatory cytokine signalling

Figure 1 Word cloud showing the key words from the Tumour Cell Biology future research proposal in 2012.

pathways are prominently activated in CAFs. These findings are in agreement with our previous findings that Traf6-dependent expression of TNFa by squamous cell carcinoma cells leads to activation of CAFs (Chaudhry et al., 2012; Oncogene). Nil Ege is now establishing methods to image the activity of these signalling pathways in fibroblasts in organotypic models and in vivo. This will enable us to study the activation of CAFs in situ. Stefanie Derzsi is also investigating the interplay between CAFs and cancer cells, but with a focus on signalling associated with direct contacts between cancer cells and CAFs. She has been greatly assisted by analytic imaging methods developed by Rob Jenkins. Stefanie has found a variety of molecular contacts between cancer cells and CAFs. Further, direct contact between cancer cells and CAFs triggers similar changes in cell signalling and gene expression to those that we observe in the invasive cells purified by Cerys. This suggests that contact with CAFs may be a key factor in promoting the invasive behaviour that we observe in vivo. We are always looking to relate our experimental models with the clinical setting. Steve Hooper is

liaising with clinicians at various locations in London to study the invasion of squamous cell carcinoma. He is imaging patterns of invasion in primary explants. This is helping to confirm ideas emerging from our organotypic models and also suggesting new hypotheses. Further, he is establishing human CAFs in culture that we are using to corroborate findings from the analysis of murine CAFs performed by Fernando. A notable feature of Steve’s analysis is the prominence of collective invasion. Danielle Park joined the group in October and she will extend our group’s analysis of collective invasion. Clearly, there is still much that we do not know. Over the coming years we will focus on enhancing our understanding of the molecular interactions of CAFs with cancer cells and the physical properties of the extra-cellular environment. We are particularly interested in learning more about the role of these interactions in modulating the response to therapeutic agents and how they promote metastatic spread.

Publications listed on page 142

Figure 2 Intravital of melanoma (purple) and associated blood vessels (yellow).



LINCOLN’S INN FIELDS

65

MOLECULAR NEUROPATHOBIOLOGY www.london-research-institute.org.uk/research/gipi-schiavo

Group Leader

Giampietro Schiavo Postdoctoral Scientists Solene Debaisieux Olga Martins de Brito Michael Parkinson Graduate Students Kinga Bercsenyi Katherine Gibbs Nathalie Schmieg Martin Wallace Scientific Officers Matthew Golding Marco Terenzio Claire Thomas Visiting Worker Pilar Cabo

Neuronal function relies on the formation and maintenance of a complex axonal and dendritic network, which is actively remodelled in response to synaptic activity and specific signals. The homeostatic control of these processes requires the tight regulation of trafficking mechanisms to facilitate the intracellular transport of organelles and cargoes, such as mRNA, mitochondria and signalling endosomes, which in turn are necessary to ensure neuronal growth, differentiation and survival. Over the past five years, our laboratory has developed a major interest in the analysis of the spatio-temporal control of signalling endosomes and in particular of neurotrophin signalling. The signalling cascades activated by neurotrophins in the neuronal periphery are regulated by the trafficking of internalised receptors towards the cell body, where they elicit transcriptional responses. In spite of the importance of this process in health and disease, the nature of the cellular machinery controlling neurotrophin receptor trafficking in neurons has not yet been fully elucidated.

receptors. This ligand-dependent re-routing alters the repertoire of signalling-competent neurotrophin receptors at the plasma membrane, which results in attenuated phosphorylation of TrkB, and reduced activation of its downstream effectors, such as AKT and MAPK. These findings, together with our observation that BICD1 expression is restricted to the developing nervous system when neurotrophin receptor expression peaks, indicate that BICD1 plays a key role in regulating neurotrophin signalling by modulating the endosomal sorting of internalised ligandactivated neurotrophin receptor complexes.

BICD1 regulates the sorting and intracellular signalling of neurotrophin receptors We have recently identified new players of this process by screening a targeted siRNA library for effects on the dynamics of endosomes containing neurotrophin receptor complexes in motor neurons differentiated from mouse embryonic stem cells. Depleting Bicaudal D homolog 1 (BICD1) increased the intracellular accumulation of ligand-bound TrkB, a tyrosine receptor kinase activated by brain derived neurotrophic factor (BDNF), and the non-catalytic neurotrophin receptor p75NTR. BICD1 has been shown to play a role in cytoplasmic dynein-mediated transport of mRNA in Drosophila, synaptic clathrin reclying, and has more recently been described as having an important role in the development and function of the Drosophila and C. elegans nervous systems.

A motor driven mechanism for cell length sensing The long-range trafficking of neurotrophin receptors from the soma to the neuronal periphery and the retrieval of the activated neurotrophin-bound complexes to the cell body is mediated by fast axonal transport. Axonal transport is paramount for neuronal survival and deficits in this process have been linked to widespread neurodegenerative diseases. However, we have recently found in a collaborative study with Mike Fainzilber at the Weizmann Institute, that in addition to its homeostatic role, axonal transport also functions as cell size sensor mechanism in different cell types.

BICD1 depletion affects neurotrophin receptor trafficking by disrupting endosomal sorting, reducing lysosomal degradation and increasing the plasma membrane recycling of neurotrophin

66

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Size homeostasis is a fundamental problem in cell biology, yet it remains unclear how large cells such as neurons assess their own size or length. We examined a role for molecular motors, such as cytoplasmic dynein and kinesins, in intracellular length sensing. Computational simulations suggest that spatial information can be encoded

Figure 1 Motor neuron derived from differentiated mouse embryonic stem cells in a microfluidic chamber. The panel displays the axons of two neurons differentiated from HB9::GFP mouse embryonic stem cells stained for the pan neuronal marker βIII tubulin (blue) and the axonal specific marker SMI32 (red). Only one of the two neurons expresses GFP (green, on the right) and can be unambiguously identified as a motor neuron).

by the frequency of an oscillating retrograde signal arising from a negative feedback loop between bidirectional motor-dependent signals. The model predicts that decreasing either or both anterograde or retrograde signals should increase cell length. This prediction was confirmed upon application of siRNAs for specific kinesin and/or dynein heavy chains in adult sensory neurons. Dynein heavy chain 1 mutant sensory neurons also exhibited increased lengths both in vitro and during embryonic development. Moreover, similar length increases were observed in mouse embryonic fibroblasts expressing a single point mutant of dynein heavy chain 1. Our results indicate that molecular motors critically influence intracellular length sensing and growth control. Botulinum neurotoxin transport in central nervous system neurons Several pathogens and virulence factors exploit the axonal retrograde transport pathway used by neurotrophin receptor complexes to reach the neuronal soma. In the last five years, our laboratory has played a major role in defining the mechanism of axonal transport of tetanus neurotoxin and several viruses, such as poliovirus and canine adenovirus type 2. Recent evidence suggests that botulinum neurotoxins (BoNT), which are now widely used in human therapy, also enter this pathway and cause central effects upon injection at distal sites, both in humans and animal models. This is surprising, given that the striking differences between the clinical symptoms of



tetanus (spastic paralysis) and botulism (flaccid paralysis) have been ascribed to the different fate of these neurotoxins once internalised in motor neurons. Whilst tetanus toxin is known to undergo transcytosis into inhibitory interneurons and block the release of inhibitory neurotransmitters in the spinal cord, BoNTs were considered to act only locally by blocking acetylcholine release at the neuromuscular junction. We recently solved this conundrum by demonstrating that BoNT/A and BoNT/E are internalised by spinal cord motor neurons and undergo fast axonal retrograde transport in non-acidic axonal carriers that partially overlap with those containing neurotrophin receptors and tetanus toxin, following a process that is largely independent of stimulated synaptic vesicle endo-exocytosis. Whereas for tetanus neurotoxin these peripheral and central effects coincide, for BoNT/A, the central effect lags behind the onset of peripheral symptoms. Consequently, the blockade of acetylcholine release at the neuromuscular junction ‘masks’ any alteration in motor neuron firing caused by the central action of BoNT/A, thus leaving the injected muscle persistently flaccid. Our findings offer a sound explanation as to the mechanism of action of these important virulence factors in vivo and provide the rationale for novel uses of BoNTs in the clinic.

Publications listed on page 142

LINCOLN’S INN FIELDS

67

GENE EXPRESSION ANALYSIS www.london-research-institute.org.uk/research/almut-schulze

Group Leader

Almut Schulze Postdoctoral Scientists Emma Ferber Barrie Peck Susana Ros Graduate Students Franziska Baenke Heike Miess Scientific Officer Beatrice Griffiths

The hallmarks of cancer include uncontrolled proliferation, reduced cell death and the loss of tissue homeostasis. The loss of normal control of cell growth and proliferation is the consequence of aberrant regulation of cellular signalling pathways through the activation of oncogenes or loss of tumour suppressor function. Altered cellular metabolism has emerged as an important feature of tumour cells. The focus of our work is to understand how oncogenic signalling pathways regulate metabolic processes and how this contributes to the development or maintenance of the cancerous state. Metabolic reprogramming is now recognised as an important hallmark of cancer. Our work has highlighted several important areas of crosstalk between oncogenic signaling and metabolic regulation in cancer cells. Altered metabolic activity is important to support the rapid proliferation of cancer cells but also makes them highly vulnerable to perturbation (Schulze and Harris 2013; Nature. 7;494(735):130). Therefore, metabolic enzymes are likely to provide promising targets for cancer therapy (Jones and Schulze, 2012; Drug Discov Today. 17: 232-41). Regulation of mitochondrial activity and reactive oxygen metabolism by FOXO3a The PI3-kinase/Akt signalling pathway is frequently activated in human cancer, and it is believed that it is vital for the survival of cancer cells. Among the downstream effectors of this signalling pathway are the FOXO proteins, a class of transcription factors that are phosphorylated and inhibited in response to Akt activation. FOXO proteins have been implicated in cell cycle control, cellular redox balance, tumour suppression and aging. Using gene expression analysis, we have demonstrated that FOXO3a activation results in the global downregulation of many nuclearly encoded genes with mitochondrial function. This regulation was independent of known regulators of mitochondrial gene expression, but required the repression of c-Myc function. Mitochondria are central hubs for cellular bioenergetics and an

68

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

important source of reactive oxygen species (ROS). While excess ROS can cause oxidative damage to cellular structures, free oxygen radicals have also been recognised as important mediators of cellular signalling. We found that FOXO3a activation altered mitochondrial morphology and function resulting in decreased oxygen consumption and reduced ROS accumulation. This reduction in cellular ROS levels was independent of SOD2, the enzyme previously identified as the major ROS-detoxification factor downstream of FOXO3a, but was restored by constitutive activation of c-Myc. The opposing effects of FOXO3a and c-Myc on mitochondrial function could be linked to reactive oxygen metabolism in hypoxic cells. We could also show that inhibition of c-Myc function by FOXO3a blocked HIF-1α stabilisation in hypoxic cells suggesting that regulation of mitochondrial ROS production contributes to the tumour suppressor function of FOXO3a (Ferber et al., 2012; Cell Death Differ. 19: 968-79). Inhibition of SREBP induces ER-stress by altering cellular lipid composition and blocks cancer cell growth Regulation of lipid metabolism via activation of sterol regulatory element binding proteins (SREBPs) has emerged as an important function of the PI3-kinase/Akt signalling axis. Although the contribution of dysregulated Akt/ mTORC1 signalling to cancer has been investigated extensively and altered lipid metabolism is observed in many tumours (Santos and Schulze

2012; FEBS J. 279: 2610-23), the exact role of SREBPs in the control of biosynthetic processes required for Akt-dependent cell growth and their contribution to tumourigenesis was only poorly understood. We demonstrated that inhibition of SREBP causes distinct changes to cellular lipid composition and cellular bioenergetics. Inhibition of SREBP also caused the engagement of the ER-stress pathway leading to inhibition of global protein synthesis. We found that SREBP function is crucial for the survival of cancer cells, particularly under conditions of limited extracellular lipid availability. Moreover, silencing of SREBP1 induced ER-stress in human glioblastoma cells and prevented tumour growth in a xenograft model (Griffiths et al., Cancer and Metabolism. In Press). Our results demonstrate the importance of SREBP in the coordination of lipid and protein biosynthesis, two processes that are essential for cell growth and proliferation. They also show that regulation of lipid composition by SREBP is essential to maintain cellular integrity and prevent the activation of stress response pathways and cell death.

Figure 1 Role of PFKFB4 in supporting prostate cancer cell survival A. Effect of PFKFB4 depletion on tumour formation in prostate cancer cells. B. Histology of control tumours and tumours after depletion of PFKFB4. Arrows indicate apoptosis. C. Metabolic effects of PFKFB4 depletion in prostate cancer cells. Loss of PFKFB4 results in increased synthesis of Fru-2,6-BP and allosteric activation of PFK1. This enhances glycolytic flux and depletes metabolites from the pentose phosphate pathway. Reduced NADPH production causes ROS accumulation and cell death.

A

Functional metabolic screen identifies PFKFB4 as an important regulator of prostate cancer cell survival Alterations in metabolic activity contribute to the proliferation and survival of cancer cells. We investigated the effect of siRNA-mediated gene silencing of 222 metabolic enzymes, transporters and regulators on the survival of three metastatic prostate cancer cell lines and a non-malignant prostate epithelial cell line. This approach revealed significant complexity in the metabolic requirements of prostate cancer cells and identified several genes selectively required for prostate cancer cell survival. Among these genes was PFKFB4, an isoform of phosphofructokinase 2 (PFK2). We demonstrated that PFKFB4 is required to balance glycolytic activity and anti-oxidant production to maintain cellular redox balance in prostate cancer cells. Depletion of PFKFB4 also inhibited tumor growth in a xenograft model, indicating that its function is essential under physiological nutrient levels (Ros et al., 2012; Cancer Discovery. 2: 328-43). These results suggest that PFKFB4 is a potential target for the development of anti-neoplastic agents. Publications listed on page 143

B

C



LINCOLN’S INN FIELDS

69

MACROMOLECULAR STRUCTURE AND FUNCTION www.london-research-institute.org.uk/research/martin-singleton

Group Leader

Martin R Singleton Postdoctoral Scientists William Chao Matthew Webster Silva Zakian Graduate Students Avradip Chatterjee Thibaud Perriches Ben Wade

Our laboratory studies the molecular mechanisms of eukaryotic chromosome segregation. This process requires that microtubules of the mitotic spindle are anchored to bioriented sister chromatids at anaphase onset. This connection is mediated by a large multi-protein complex, the kinetochore, and the sister chromatids are in turn held together by another large complex known as cohesin. We are interested in determining the atomic structures of some of the individual proteins in these complexes, and understanding how their overall architecture and function aids the mitotic spindle in bringing about the rapid and accurate dissemination of the genome. We employ a combination of biophysical and biochemical techniques to address these questions, primarily those of X-ray crystallography for high-resolution studies and electron microscopy to analyse larger complexes.

Scientific Officer Xiao Hu

Centromere establishment The centromere is the unique chromosomal locus upon which the kinetochore assembles. Centromere identity in eukaryotes is determined by the presence of nucleosomes containing a specialised version of histone H3, generically known as CenH3. It is essential that the incorporation of CenH3 nucleosomes into chromatin is restricted to a single site, which is maintained throughout the cell cycle. How does this occur? In budding yeast, it appears that the initial signal is the binding of the 430 kDa CBF3 complex to the conserved CDEIII element of the 125 b.p. centromere sequence. CBF3 contains four subunits; Ndc10, Cep3, Ctf13 and Skp1 with a presumed 2:2:1:1 stoichiometry. We have previously determined the crystal structure of Cep3, and since been working on Ndc10. This is a large, multi-domain protein with no obvious sequence motifs that provide clues as to its function. We have recently solved the crystal structure of the amino-terminal half of this protein (Figure 1A). It forms a predominantly α-helical structure, wrapped around a narrow β-sheet, giving the domain a 'W' like appearance. Searches for structurally homologous proteins in the protein databank surprisingly demonstrated Ndc10 to have the same fold as the tyrosine recombinase

70

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

family, which includes a wide variety of proteins such as Cre and type IB topoisomerases (Figure 1B). We biochemically determined that this domain had a non-specific DNA-binding activity, and by aligning the structure to that of the previously determined Cre-DNA complex, we were able to assign a DNA-binding surface on Ndc10 (Figure 1C), which was subsequently verified by mutagenesis experiments. The extreme N-terminal of the protein forms a helical bundle that is also present in Cre, however it adopts a substantially different conformation in Ndc10 (Figure 1B). In the Cre protein, this bundle is responsible for binding in the major groove of the loxP site, as well as protein tetramerisation. Neither of these activities are relevant to Ndc10 activity, so the conservation and function of the domain remain mysterious. All other members of this recombinase family make incisions in the phosphodiester DNA backbone via a conserved catalytic tyrosine residue. This residue, and the secondary structure element upon which it is located are absent in the Ndc10 structure, suggesting that the primary function of the complex is to provide a nonspecific DNA binding activity. Why this particular DNA-binding motif has been employed at the

Figure 1 A. Ribbon diagram of the budding yeast Ndc10 N-terminal domain. B. Three-dimensional superimposition of Ndc10 (blue) against the Cre recombinase (yellow, pdb i.d. 2crx). The N-terminal helical bundle of Cre is in a substantially different conformation to the structurally homologous Ndc10 N-terminal (indicated by arrow). C. Electrostatic surface of Ndc10 with a segment of duplex DNA modelled by inference from the Cre-DNA structure. D. Schematic diagram of the yeast centromere showing key DNA and protein components.

A

B

C

D

centromere is unclear, but it may be that subtle binding-induced changes in the DNA conformation are required to facilitate centromeric nucleosome loading. In this model, the Cep3 subunit provides site-specificity to CBF3, while Ndc10 participates more directly in the nucleosome-loading process (Figure 1D). Exactly how these proteins interact with each other, the CDEIII element, and the centromeric nucleosome present interesting questions for further study. Sister chromatid cohesion Metaphase sister chromatids are held together by the ring-shaped cohesin complex, which opposes spindle forces until proteolytic cleavage of the Scc1 subunit at anaphase onset. Cohesin is initially loaded onto chromosomes in the late G1 phase of the cell cycle. However, at this point the association between cohesin and chromatin remains reversible and is not capable of generating cohesion. Acetylation of the Smc3 subunit of cohesin by a replication-fork associated acetyltransferase, Eco1 switches the complex to a cohesion-competent state. How this relatively



small covalent modification of Smc3 triggers such a profound change in the activity of cohesin remains unclear. Studies by several laboratories have implicated the action of two members of the cohesin complex, Wpl1 and Pds5 as being critical to this process. Genetic studies have shown that Eco1 loss-of-function lethality can be suppressed by mutations in either of these proteins, leading to the idea that they have a destabilising effect on cohesin binding, which can be relieved by the counteracting acetylation of Smc3. We have recently solved the structure of the Wpl1 protein, and demonstrated that it can interact directly with the ATPase head of the Smc3 protein as well as Pds5. We are trying to determine how this binding is modulated at different stages of the cohesin cycle, and what the functional implications of the interaction on cohesin activity might be. In parallel, we are studying the substrate recognition properties of the Eco1 acetyltransferase, as it has been reported to have multiple activities including a possible role during double-strand break repair. Publications listed on page 143

LINCOLN’S INN FIELDS

71

MICROTUBULE CYTOSKELETON www.london-research-institute.org.uk/research/thomas-surrey

Group Leader

Thomas Surrey Postdoctoral Scientists Todd Fallesen Franck Fourniol Sebastian Maurer Johanna Roostalu Martina Trokter Graduate Students Hella Baumann Christian Duellberg Rupam Jha Scientific Officers Nicholas Cade Iris Lueke Masters Student Pauline Vercruysse Diploma Student Gergo Bohner

72

Various filamentous structures exist inside living cells, forming the cytoskeleton. Microtubules are protein polymers with a tubular shape that are made from tubulin subunits. During the normal cell cycle, microtubules switch between periods of growth and shrinkage, a property called dynamic instability. It is crucial for the living cell to regulate this dynamic instability, which must be properly adjusted depending on the state of the cell and its local environment. Microtubules are especially important during cell division when, in conjunction with other proteins, they build the spindle that segregates the genetic material to the two daughter cells. We study these components of the microtubule cytoskeleton using cell biology, biochemistry, and advanced fluorescence and electron microscopy. This enables us to understand the molecular mechanisms underlying the dynamic cytoskeleton behavior, which is of crucial importance to maintain cells in a healthy state. Insight into the mechanics and the length scale of nuclear divisions from a new ex vivo system Amongst others, there are two fundamental questions in cell biology: what drives the movements of macromolecular assemblies inside living cells, and what determines where large intracellular assemblies are positioned? These two questions are related when movements determine how and where objects are positioned. Correct positioning can be crucial for proper development of an organism: a dramatic example is the early development of the model organism Drosophila (the fruit fly). The first nuclear divisions take place in a single-cell embryo (a syncytium), before individual cells finally form, that then continue dividing. Little is known about how nuclei are correctly distributed and positioned in the early syncytial embryo. In collaboration with the laboratory of Anne Ephrussi, European Molecular Biology Laboratory in Heidelberg, Germany, we have developed a new assay that allows the observation of nuclear divisions in an extract made from individual Drosophila embryos (Telley et al., 2012; J Cell Biol., 197: 887-95; Telley et al., 2012; Nat. Protocols. In Press). Remarkably, the extract retained many properties of entire

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

embryos despite the absence of the embryo cortex (its outer layer). Using laser microsurgery, the new assay allowed us to clearly show which parts of the microtubule cytoskeleton cause the movement of the genetic material (the DNA) and its distribution through embryonic space. We could also demonstrate that the processes inside the embryo at this developmental stage lead to a separation of nuclei by a specific distance that also appears to exist in other insect species. Beyond allowing a better understanding of Drosophila development, this novel ex vivo assay promises to become an extremely useful tool to investigate general aspects of spindle mechanics. Structural insight into microtubule end binding and dynamic instability In all living cells which have a nucleus, i.e. cells from species ranging from yeast to human, proteins exist which bind transiently to the end region of growing microtubules. These ‘plus-end tracking’ proteins control and mediate many essential microtubule functions, and several of them are mutated in cancer cells. One of these proteins, EB1, has a key function because it recruits almost all other plus-end tracking proteins

Figure 1 Reconstruction of a microtubule from cryo-electron microscopy images. Left: the microtubule binding domain of fission yeast EB1 is shown in green. Right: Pseudo-atomic model showing 4 tubulin dimers (blue and purple) and fission yeast EB1 (green).

(Bieling et al., 2007; Nature. 450: 1100-5; Bieling et al., 2008; J Cell Biol. 183: 1223-33); hence, it was vital to understand how EB1 could bind selectively to growing microtubule ends. To answer this question, we prepared microtubules in vitro composed wholly of tubulin subunits in a form that they normally only have at growing microtubule ends (Maurer et al., 2011; PNAS. 108: 3988-93); we were then able to visualise how EB1 proteins bind to these microtubules using cryoelectron microscopy and single particle reconstruction methods. In collaboration with Carolyn Moores, Birkbeck College, London, we determined exactly where EB1 binds on the microtubule surface; furthermore, we were able to generate a pseudo-atomic model which revealed for the first time the specific nature of the molecular interactions at high resolution (Maurer et al., 2012; Cell. 149: 371-82). We discovered that EB1 binds to a microtubule structure that is vital for the dynamic properties of the microtubule. This finding provides new insight into the molecular mechanism of dynamic instability and contributes to a better understanding of how other regulators of microtubule dynamics act at the molecular level. Artificial regulators of microtubule dynamics Several anti-cancer drugs affect microtubule dynamic instability and therefore can arrest cell division. Nowadays, novel artificial proteins with



designed properties can be generated that act as inhibitors of microtubule dynamics with different properties. In one of our projects (in collaboration with Marcel Knossow, CNRS, Gif-sur-Yvette, France) we have characterised such a new protein (Pecqueur et al., 2012; PNAS. 109: 12011-6). However, it is a challenge to deliver such proteins or conventional drugs to a defined area of tissue or to a distinct region in a cell. In collaboration with the group of Aranzazu del Campo at the Max Planck Institute for Polymer Research in Mainz, Germany, we have made progress towards such targeted effects: Paclitaxel, a commonly used small-molecule inhibitor of microtubule dynamics, was chemically modified by our collaborators to make it ‘light activatable’. We were able to show that this ‘caged’ Paclitaxel could indeed be transformed from an inactive state to an active, inhibitory state by light illumination (Gropeanu et al., 2012; PLoS One. 7: e43657). Using local illumination, this allowed us to affect microtubule dynamic instability in only one pre-defined part of an in vitro culture of living cells. This new compound has the potential to become a useful tool for cell biologists studying microtubule cytoskeleton dynamics, and possibly to be further developed into a drug that can be applied to tissues allowing targeted treatment.

Publications listed on page 143

LINCOLN’S INN FIELDS

73

TRANSLATIONAL CANCER THERAPEUTICS www.london-research-institute.org.uk/research/charles-swanton

Group Leader

Charles Swanton Postdoctoral Scientists Xin Yi Chloe Goh Carlos Lopez-Garcia Pierre Martinez Sarah McClelland Claudio Santos Clinical Fellow Marco Gerlinger Graduate Students Rebecca Burrell Andrew Crockford Sally Dewhurst Nicholas McGranahan Scientific Officers Eva Gronroos Andrew Rowan

74

Cancer drug resistance is almost inevitable in the majority of advanced metastatic solid tumours. Our group and others have demonstrated intratumour genetic heterogeneity (ITH) between primary and metastatic sites and within individual primary tumours and shown that chromosomal instability, a pattern of genomic instability that contributes to ITH, is associated with poor outcome and multidrug resistance. Understanding how tumours generate ITH, adapt to new environments and resist therapeutic intervention is the major research focus of the laboratory.

Understanding the causes and consequences of intratumour heterogeneity Recent technological advances have permitted higher resolution and more rapid analysis of individual cancer genomes at the single nucleotide level revealing extensive inter-tumour heterogeneity with limited somatic alterations shared between tumours of the same histopathological subtype. ITH has been documented by our group and others both within individual tumour biopsies and spatially separated between biopsies of the same tumour. Sequential analysis of tumours has also revealed evidence that ITH temporally evolves during the disease course. ITH has implications for predictive or prognostic biomarker strategies, where the tumour subclone that may ultimately influence therapeutic outcome may evade detection due to its absence or presence at low frequency at diagnosis or due to its regional separation from the tumour biopsy site. Emerging evidence suggests that tumours with greater diversity have a worse clinical outcome and an increased propensity for resistance to common anti-cancer treatments. However, despite this relationship, extremes of tumour diverstity may be associated with relatively improved survival outcomes, suggesting that tumour diversity may be therapeutically exploitable. Our laboratory is focused on understanding a mechanistic basis for tumour heterogeneity and its impact upon tumour evolution. We are adapting tumour sampling methodologies to the challenges of intratumour

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

heterogeneity within clinical trials focused on identifying new drug resistance mechanism (Figure 1) and studying tumour adaptation during treatment and disease progression in an attempt to identify drivers of tumour diversity that may be therapeutically tractable. 1. Temporo-spatial heterogeneity of solid tumours and defining convergent evolutionary events We are using multiregional genomics analysis of renal, breast and non-small cell lung cancers to define phylogenetic clonal relationships and patterns of disease progression and tumour evolution (Swanton 2012; Cancer Res. 72: 4875-82). We have proposed a tumour phylogenetic tree model that defines trunk drivers as optimal therapeutic targets present within all tumour subclones, whilst attempting to map tumour diversity in terms of heterogeneous somatic events that may result in the acquisition of drug resistance present in the tumour branches (Yap et al., 2012; Sci Trans Med. In Press). We have defined parallel evolutionary events, where the same gene is inactivated three times in three distinct ways in three different regions of the same tumour (Figure 2). Our work is revealing several examples of this phenomenon which suggests that tumour progression may be predictable and hence therapeutically tractable, due to tumour phenotypic dependence on inactivation or activation of similar pathways in genetically distinct regions of the same tumour.

Figure 1 The Challenges of Intratumour Heterogeneity subject to investigation by the Translational Cancer Therapeutics Group (Taken from Swanton 2012; Cancer Res. 72: 4875-82).

Figure 2 Parallel evolution of three distinct SETD2 loss of function events in three different regions of a primary and paired metastatic Renal cell carcinoma from the same patient (adapted from Gerlinger et al., 2012; N Engl J Med. 366: 883-92). SETD2 is a H3K36 trimethylase. Immunohistochemistry of H3K36me3 revealed absent me3 staining in all tumour regions in contrast to a positive SETD2 wild type control. Black arrows tumour nuclei, white arrows stroma.

2. Understanding the relationships between tumour diversity and clinical outcome Our laboratory is studying the relationships of chromosomal structural and numerical instability with tumour metastases initiation and therapeutic outcome in solid tumours using sector ploidy profiling and FISH based approaches to quantify CIN in whole tissue sections. Efforts to model genomic instability in vitro and in vivo are ongoing using functional and integrative genomics approaches. By the end of 2013 we will have completed the analysis of 3600 breast tumours treated within the TACT adjuvant primary breast cancer clinical trial to address the relationships of breast cancer numerical chromosomal instability with outcome.



3. Towards a mechanistic basis for tumour heterogeneity. We hypothesised that genomic instability may be driven by underlying ordered genomic changes within unstable cancer genomes. Using tumour genomics datasets we have defined such ordered chromosomal events in highly genomically unstable lung, breast and colorectal cancers. We are currently running a panel of cell cycle genomic checkpoint assays to address the functional relevance of genes encoded within these genomic regions to attempt to identify drivers of tumour heterogeneity in solid tumours.

Publications listed on page 144

LINCOLN’S INN FIELDS

75

APOPTOSIS AND PROLIFERATION CONTROL www.london-research-institute.org.uk/research/nic-tapon

Group Leader

Nicolas Tapon Postdoctoral Scientists Eunice Chan Ieva Gailite Yanlan Mao Paulo Ribeiro Ohad Shaham Graduate Students Pedro Gaspar Rachael Shaw Yanxiang Zhou Scientific Officers Birgit Aerne Maxine Holder Master Student Lennart Kester

Organ development in multicellular organisms requires a strict control of cell proliferation, growth and death. Genetic screens in Drosophila melanogaster have identified the Hippo (Hpo) pathway as one of the major signalling pathways required for tissue size control. Our lab aims to identify new members of the Hpo pathway and to understand how the pathway functions to restrict tissue size during development and in adult stem cells. In addition, we are more broadly interested in how tissue architecture is coupled to growth control. We use a combination of genetics, cell biology, biochemistry and mathematical modelling to tackle these issues.

The Hpo pathway controls the final tissue and organ size by both inhibiting cell proliferation and promoting apoptosis. At the core of the pathway lies a kinase cascade comprised of the Ste20related kinase Hpo and the Dbf2-related kinase Warts (Wts). Hpo and Wts, together with their respective scaffold proteins Salvador (Sav) and Mob as Tumour Suppressor (Mats), phosphorylate and inhibit the pro-growth transcriptional co-activator Yorkie (Yki). The upstream signals that modulate Hpo pathway activity remain poorly understood. Through the KEM and AMOT complexes, which associate with the apical polarity protein Crumbs (Crb), and junctional components such as α-catenin, Hpo signalling is believed to respond to cell density. Fat signalling has been proposed to mediate the effects of morphogens on Hpo signalling. Finally, mechanical forces, sensed through the actin cytoskeleton, have also been proposed to modulate Yki/YAP activity, though the molecular mechanism remains unclear. Thus, the Hpo pathway is believed to respond to a variety of signals such as local tissue architecture (cell crowding, tissue mechanics) as well as long-range patterning cues (morphogen gradients), but the details of upstream signalling still remain to be elucidated. RNAi screening approaches With the help of the High-Throughput Screening Facility at the LRI, we have devised strategies to

76

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

identify Hpo pathway members by performing high-thoughput RNAi screens in tissue culture. Our first attempt was using a luciferase-based reporter for the activity of Yki, the transcriptional coactivator that is negatively regulated by Hpo signalling. By combining the RNAi screening data with a proteomic map of the Hpo pathway generated by Affinity-Purification/Mass Spectrometry by our collaborator Matthias Gstaiger at ETH in Zürich, we identified the STRIPAK PP2A complex as a major phosphatase that antagonises Hippo activation. Recently, we have sought to create more sophisticated readouts of Hippo pathway activity. We have used the Split-TEV technique, which is based on the functional complementation of previously inactive TEV protease fragments fused to potential interaction partners of choice. Using this method, we can detect the protein-protein interaction between Yki and 14-3-3, which is dependent on Yki phosphorylation by Wts. This readout has enabled us to screen the Drosophila genome for proteins that modulate Wts activity in Drosophila S2R+ tissue culture cells. The SIK kinases repress Hpo signalling Using the Split-TEV-based assay for Wts activity described above, we identified the Salt-inducible kinases (Sik2 and Sik3) as negative modulators of Hpo signalling. In vivo overexpression of active Sik2 induces tissue overgrowth and upregulation of the Yki transcriptional target expanded (Figure

1). The effect of Sik2/3 on Hpo signalling is mediated, at least in part, by Sav phosphorylation on S413. Our data suggest that Sav phosphorylation by Sik reduces its ability to efficiently scaffold the Hpo/Wts core kinase complex, thereby reducing Yki inhibitory phosphorylation. In agreement with this notion, Sav-S413A displayed an enhanced ability to reduce growth in vivo. Siks belong to the AMPK (AMP-activated protein kinase) family, which are regulated by the LKB1 (Liver kinase b1) tumour suppressor. Members of this family have been implicated in regulating metabolism in response to nutrient/energy levels, like AMPK itself, which represses mTOR (mammalian Target of Rapamycin) signalling in response to high AMP/ATP ratios. Siks play a major role in inhibiting gluconeogenesis in the liver in response to high glucose levels through inhibitory phosphorylation of the transcriptional coactivator CRTC2 (CREB-regulated transcription coactivator 2)/TORC2, and activatory phosphorylation of the Histone Deacetylase HDAC4, a function which appears to be conserved in Drosophila. The Siks are under hormonal control by Insulin receptor (InR) signalling via the downstream kinase Akt, which phosphorylates and activates Sik2/3, and the glucagon homologue Adipokinetic Hormone (AKH), which signals through a G-protein coupled receptor (GPCR) and PKA, which phosphorylates and inhibits Siks. Under fasting

conditions low Insulin and high AKH activity combine to shut down Sik3, thereby promoting gluconeogenesis and inducing mobilisation of Fat Body (FB - the fly liver equivalent) lipid stores to restore circulating glucose levels and energy homeostasis. Our work suggests a role for Sik2/3 in growth control during development. Analogously to InR signalling in Drosophila, which promotes both nutrient storage in the FB and developmental growth of peripheral tissues, the Sik kinases might couple Hpo pathway activity to nutrient or energy availability, ensuring that Yki is only able to drive tissue growth under favourable conditions. Recent work has established a connection between SIK kinases and cancer. In human patients with high-grade serous ovarian cancers, high SIK2 expression correlated with poor prognosis, and Sik2-depleted ovarian cancer cell lines showed decreased growth, while SIK3 promoted proliferation when overexpressed in ovarian cancer cells. Furthermore, SIK2 has been identified as a candidate oncogene from lung adenocarcinoma patient samples. We showed that SIK2 promotes YAP-dependent transcription in human cells. SIK2 inhibitors may therefore prove attractive candidates to boost Hpo pathway activity in ovarian tumour cells, though such a strategy may be less effective in tumours harbouring YAP amplifications.

Publications listed on page 144

Figure 1 Activated Sik2 promotes the expression of the Yki target Expanded. Expression of Hpo RNAi (A, A’) and Sik2-S1032A (C, C’), but not wild type Sik2 (B, B’) in actin flip-out clones (marked positively with GFP in green), leads to the upregulation of Ex (red or white) protein levels at the apical membrane of wing imaginal discs.



LINCOLN’S INN FIELDS

77

EPITHELIAL BIOLOGY www.london-research-institute.org.uk/research/barry-thompson

Group Leader

Barry Thompson Postdoctoral Scientists Eliana Lucas Neil Pearson Graduate Students Graham Bell Mariana Campos Ichha Khanal Clara Sidor Scientific Officer Georgina Fletcher

How animal cells co-operate to build tissues remains a fundamental unsolved problem in biology. In order to construct tissues of particular forms, cell must be able to orient their behaviour relative to one another. We examine the molecular determinants of cell polarity and how these determinants orient cell behaviour in vivo. The relevance of this work for human disease is illustrated by the fact that defects in systems of cell polarity can lead to developmental abnormalities and tumour formation. We focus particularly on the epithelial tissues of the fruit fly Drosophila, which offers the possibility of live-imaging of epithelial tissue development as well as powerful genetic analysis. We also apply computational modelling of cell polarity and tissue morphogenesis to understand the principles that govern polarization of determinants and how these determinants can influence cell shape, cell division, cell migration and tissue morphogenesis. Apico-basal polarity in epithelia: the Crumbs system Epithelial polarity is crucial to orient the behaviour of epithelial cells. The plasma membrane of epithelial cells is segregated into distinct apical and basolateral domains. The set of evolutionarily conserved determinants of apical and basolateral polarity are known, but how these determinants actually polarise remains unsolved. In our genetic screen, we identified multiple components of the endocytic trafficking machinery as being essential for epithelial polarity in Drosophila. This finding led us to re-examine the mechanism by which epithelial polarity is maintained in the follicle cell epithelium (Figure 1). Previous work suggested that polarity involves mutual antagonism between apical and basolateral determinants. However, we find that mutual antagonism alone is not sufficient to generate stable polarity in computer models. Our results indicate that the apical polarity determinants engage in a positive feedback loop based on regulated endocytosis of the key transmembrane apical determinant Crumbs. This positive feedback loop is essential for polarity and,

78

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

when combined with mutual antagonism, is sufficient to generate spontaneous and robust polarity in computer models. This work provides a new principle for understanding the maintenance of epithelial polarity. We now aim to unravel the mechanism by which Crumbs can self-recruit to the plasma membrane to drive positive feedback. Clearly, one key regulatory step is the regulation of Crumbs endocytosis – concentration of Crumbs at the membrane must be capable of inhibiting its own endocytosis. Crumbs requires other apical determinants, such as cdc42, aPKC, Par6 and Sdt to stabilise itself at the membrane, but how these determinants prevent Crumbs endocytosis remains unclear. We have recently identified a role for the Spectrin cytoskeleton in anchoring Crumbs at the plasma membrane. Loss of the Spectrin cytoskeleton results in a strong reduction of Crumbs localisation at the plasma membrane. We now wish to understand how Crumbs is linked to Spectrins. Good candidate molecules are the FERM domain proteins Expanded, Merlin and Moesin, which bind to the Crumbs intracellular domain. At least one of these proteins, Moesin,

Figure 1 Crumbs polarizes through a combination of positive feedback among the apical determinants and mutual antagonism between apical and basolateral determinants. Positive feedback is achieved through self-recruitment of Crumbs via Crumbs-Crumbs interactions and recruitment of cdc42, aPKC, Par6, Sdt, Expanded and Kibra. Mutual antagonism involves Lgl being removed from the apical domain upon aPKC phosphorylation and Crumbs being removed from the basolateral domain upon Lgl inhibition of aPKC kinase activity.

has been previously reported to interact with Spectrins. Our preliminary results indicate that Expanded and Merlin also bind directly to Spectrins. We now aim to remove all three of these FERM domain proteins by construction of a triple mutant and to examine the consequences for Crumbs localisation. We are also interested in testing whether anchorage of Crumbs to Spectrins via FERM domain proteins might be regulated by other apical determinants and in particular by aPKC phosphorylation. Another key step in Crumbs self-recruitment is that ability of Crumbs to polarise its own delivery to the plasma membrane. Crumbs arrives at the plasma membrane via Rab11-positive endosomes, which are a conduit for both newly synthesised Crumbs and recycled Crumbs. The Exocyst complex has also been implicated in helping to deliver Crumbs from Rab11-positive endosomes to the plasma membrane. Both the Exocyst complex and Rab11-positive endosomes are localised at or



near the apical membrane domain in the follicle cell epithelium. We are interested in how these key regulators of Crumbs trafficking are themselves polarised. A key candidate identified in our genetic screens is Kibra, a WW and C2 domain protein that functions in parallel with Expanded and Merlin. We are now interested in testing whether Kibra interacts with the Exocyst complex to help deliver Crumbs to the apical membrane. If so, this would explain why simultaneous loss of both Kibra and Expanded causes strong crumbs-like phenotypes in Drosophila – because loss of both proteins interferes with both delivery and subsequent anchorage of Crumbs to the plasma membrane. This work promises to generate new mechanistic insights into how a key polarity determinant becomes polarised in epithelia.

Publications listed on page 145

LINCOLN’S INN FIELDS

79

CELL REGULATION www.london-research-institute.org.uk/research/takashi-toda

Group Leader

Takashi Toda Postdoctoral Scientists Isabelle Jourdain Yuzy Matsuo Akiko Nishi Hirofumi Takada Graduate Students Aldona Kutkowska Risa Mori Ngang Heok Tang Scientific Officer Hirohisa Masuda

High-fidelity chromosome transmission lies at the heart of successive and successful cell division. Any errors in this process would result in birth defects and/or aneuploidy, the hallmark of many human cancers. Two different cytoskeletal systems, microtubules and actin filaments, coordinately play seminal roles in equal partition of duplicated chromosomes. Microtubules form bipolar spindles, which pull each pair of sister chromatids towards the opposite poles. The actin cytoskeleton, on the other hand, is required for cell morphogenesis and physical cell separation (cytokinesis), thereby producing two offspring cells. The long-term goal of our laboratory is to understand the molecular mechanisms underlying how these cytoskeletons ensure faithful chromosome segregation. During 2012, we have made the following two progresses. One is the identification of a conserved key molecule that regulates the global organisation of the actin cytoskeleton. The other is the finding of a novel ubiquitin-proteolysis pathway that is critical to meiotic spindle assembly during gametogenesis. Identification of fission yeast Sec3 as a central player essential for the global organisation of the actin cytoskeleton Actin, one of the most ubiquitously conserved proteins across the eukaryotes, plays diverse roles in cell motility, cell-cell contact, transcription and cell division. To execute a myriad of these cellular functions, actin interacts with multiple actinbinding proteins, thereby displaying diverse morphologies inside the cell; in human cells, actin assembles into at least 15 distinct structures. In many tumour cells, theses actin structures are altered due to its misregulation. Therefore understating the mechanism by which actin structures are properly organised is critical for researchers to tackle cancers and develop cancer therapeutics. It is assumed that all these actin structures, despite their diverse forms, are under the control of an upstream signalling molecule. In fact the small GTPases such as Cdc42 are proposed to be such a master regulator, yet our knowledge of how Cdc42 executes its commanding role towards downstream actin-mediated events

80

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

remains largely limited. In the current work, using fission yeast Schizosaccharomyces pombe, we have uncovered a conserved downstream molecule of Cdc42. In the fission yeast, actin comprises only three structures, membranebound endocytic patches, intracellular transportmediating cytoplasmic cables and the medial cytokinetic actomyosin ring. Sec3 is a component of the exocyst complex, which plays essential roles in many exocytic processes including exocytosis, polarised growth and cell division. A canonical model for exocyst function is that this complex provides spatiotemporal information for the recruitment and tethering of Golgi derived secretory vesicles to the plasma membrane prior to vesicle fusion. From genome-wide screening for novel cell-cycle regulators, we have identified the fission yeast homologue of Sec3 as a critical regulator for the actin cytoskeleton (Ikebe et al., 2011; Biosci Biotechnol Biochem. 75: 2365-70; Jourdain et al., 2012; Traffic. 13: 1481-95). We have shown that in

Figure 1 Sec3 controls actin organisation in fission yeast. Actin structures in live fission cells stained by LifeAct-GFP. Wild type and temperature sensitive sec3 mutant cells (sec3-916) were grown at 36°C. Three types of the actin cytoskeleton are visible in wild type cells (the left panel). Filamentous cables serve as tracks for vesicular transport (top two cells); cortical patches at the cell tips ensure the endocytic uptake of material (top two cells) and the medial cytokinetic ring constricts to separate two daughter cells (left-most cell). All these structures are absent (cables) or altered (patches and ring) in sec3 mutants (the right panel). Bar, 10 μm.

Figure 2 A model of Sec3 function. As part of the exocyst complex, Sec3 (green ovals) has a canonical role in tethering secretory vesicles to the plasma membrane prior to fusion (the top right hand corner of the figure). Sec3 also physically binds actin-binding proteins including Formin-For3 and a membrane-bound protein Sla2, by which these two proteins are recruited and anchored to the proper cortical sites (cell tips) on the plasma membrane. For3 and Sla2 are required for actin cable assembly (red background) and clathrin-mediated endocytosis (orange background) respectively. In addition Sec3 regulates cytokinesis by interacting with currently hypothetical factors (?) that direct the constriction of the cytokinetic actomyosin ring (the bottom left corner in the blue background). Sec3 therefore bridges the secretion machinery to all three actin structures in fission yeast. Sec3 is under the control of the upstream polarity cues (GTPase-Cdc42), the molecular mechanisms of which currently remain to be determined.

sec3-916

LifeAct-GFP

wild-type

addition to its canonical roles in vesicle trafficking, Sec3 also plays a key role in the global organisation of actin structures. As shown in Figures 1 and 2, in the absence of Sec3 function, all the actin structures are altered; actin patches that are normally polarised to the cell tips delocalise, actin cables disappear and cytokinetic rings are deformed. Importantly these cells lose cell polarity, which is one of the most common characteristics of tumour cells. As Sec3 is conserved through evolution, the characterisation of fission yeast and human Sec3 will help our understanding of the molecular mechanism by which normal cells become cancerous. A novel ubiquitin ligase important for spindle microtubule formation during meiosis The ubiquitin-proteasome system plays a pivotal role in various cellular processes, including cell cycle progression, cell proliferation and differentiation. Substrate proteins are ubiquitylated by the enzymatic cascade consisting of ubiquitin-activating enzyme (E1), ubiquitinconjugating enzyme (E2) and ubiquitin ligase (E3). These ubiquitin transferase reactions result in the formation of polyubiquitin chains on substrates, which are recognised by the 26S proteasome,



followed by rapid irreversible degradation. Among the three enzymes, the E3 ubiquitin ligases are central to determining the timing and specificity of substrate proteolysis. There are two conserved ubiquitin ligases that regulate cell cycle progression, Anaphase Promoting Complex/ Cyclosome and Skp1-Cdc53/ Cullin-1-F-box (SCF). The SCF complex contributes to a variety of mitotic events including mitotic spindle formation, but its function during meiosis, a developmentally programmed specialised cell division to produce haploid gametes from a diploid precursor, is not well understood.

We have unveiled a novel function of the SCF in meiotic spindle formation. The skp1 temperature-sensitive mutant exhibited abnormally bent spindles during meiosis I, leading to erroneous distribution of spindle microtubules in cells undergoing meiosis II and resultant chromosome mis-segregation. We have found that the meiotic bent spindle in skp1 mutant cells was due to a hypertension generated by chromosome entanglement. Interestingly, the spindle bending was efficiently suppressed by inhibiting the formation of double strand break, indicating that the entanglement was generated via the meiotic recombination machinery. Consistently, in skp1 cells Rad51-Rad52 foci, which normally emerge only transiently during an early stage of meiosis, persisted for a prolonged period until meiosis I, leading to the accumulation of recombination intermediates. Intriguingly abnormal bent spindles were also observed in the mutant of Fbh1, a conserved F-box protein containing the DNA helicase domain. We have further shown that SCFFbh1 is likely to function in the resolution of meiotic recombination intermediates, thereby ensuring the formation of proper bipolar spindle microtubules and chromosome segregation (Okamoito et al., 2012; PLoS One. 7: e30622). As far as we are aware, this is the first report on a critical role for the SCF ubiquitin ligase in meiotic spindle assembly.

Publications listed on page 145

LINCOLN’S INN FIELDS

81

SECRETORY PATHWAYS www.london-research-institute.org.uk/research/sharon-tooze

Group Leader

Sharon A Tooze Postdoctoral Scientists Eyal Kalie Christopher Lamb Martina Wirth Graduate Students Hannah Dooley Justin Joachim Scientific Officers Harold Jefferies Minoo Razi

Autophagy is a highly conserved, membrane-mediated pathway that delivers cytoplasmic components to the lysosome for degradation. Recent developments have highlighted the essential role of autophagy in promoting cell survival in response to extracellular stress, infection, and pathological conditions, such as neurodegeneration and tumorigenesis. Importantly, autophagy is required for development and tissue homeostasis, and a further understanding of the molecular mechanisms underlying the process is essential to exploit the potential for manipulation of autophagy to treat disease. Autophagy requires intracellular membrane compartments, such as the endoplasmic reticulum, Golgi complex, endosomes and lysosomes, alongside the dedicated protein machinery, the Atg (autophagy related) proteins. My laboratory studies how the Atg proteins and trafficking proteins function during acute amino-acid withdrawal in the context of normal intracellular trafficking pathways to elucidate the molecular basis of autophagy.

Masters Student Javier Hervas

Introduction Macroautophagy (here referred to as autophagy) is non-selective autophagy (self-eating) of cytoplasmic proteins and organelles. Autophagy occurs at basal levels in all eukaryotic cells and is upregulated during amino acid deprivation. Upregulation of autophagy increases the flux through the autophagosome which fuses with the lysosome to generate degradative autolysosomes, thus replenishing the cytosolic pool of amino acids, lipids and macromolecules. Formation of autophagosomes occurs within minutes of amino acid deprivation and utilizes 18 out of the 36 Atg proteins identified in yeast. The ULK complex, which contains ULK1 and 2 (Atg1 homologues), Atg13, FIP200 (Atg17 homologue), and Atg101, the most upstream complex in the pathway, is negatively regulated by the master growth and energy sensors mTORC1 and AMPK. Inhibition of mTORC1, and activation of AMPK, triggers activation of the ULK complex and the class III phosphatidylinositol 3-kinase (PI3K) complex. Co-incident with activation, the ULK and PI3K complexes translocate to the phagophore.

82

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Downstream effectors of the ULK and PI3K complex include WIPI proteins which bind PI3-phosphate (PI3P), and the ubiquitin-like conjugation systems, which produce LC3-PE (LC3-II), Atg12-Atg5-Atg16. The concerted action of the Atg protein machinery results in the formation of a double-membrane autophagosome from the phagophore at specialized sites on the ER known as omegasomes. This membrane grows and within minutes after starvation expands to sequester cytosol and organelles, and then closes forming an autophagosome that can be up to 2 microns in diameter. While the identity of the mammalian Atg protein machinery is now known, many questions remain unanswered about the regulation of membrane dynamics, the source of the rapidly expanding autophagosomal membrane, and the interaction between the autophagosome and the other subcellular compartments. We address these questions by studying a subset of Atg proteins, and identifying novel regulators using small scale and genome wide screens under conditions of amino

Figure 1 An overview of the mammalian autophagy pathway, illustrating the sequential membrane compartments involved in the process of making an autophagosome, leading to the final stage, the autolysosome. Many of the steps illustrated by arrows are controlled by vesicular trafficking components. Contribution of membrane from the recycling endosome to the autophagosome is controlled by Rab11 and its effector TBC1D14 (Longatti et al., 2012). Atg9, a resident of the Golgi, the Atg9 compartment and endosomes, may mediate the vesicular transfer of membrane from anyone of these sources for the initiation and rapid expansion of the autophagosome (Orsi et al., 2012).

acid starvation. In addition to our cell biological approach we are developing a genetic models including mice lacking both ULK1 and ULK2, and zebrafish to study autophagy in vivo models. Membrane dynamics: role of recycling endosomes and Atg9 While recent data supports the role of the ER in the formation of the phagophore, it is likely that Golgi and other compartments provide additional membranes. To understand the contributions of other membranes we focused on Rabs proteins, small GTPases that mediate vesicular fusion. To identify the Rab proteins required for autophagy we screened by overexpression a library of RabGAPs, GTPase activating proteins that inactivate Rabs (Longatti et al., 2012: J Cell Biol. 197: 659-75). Our screen revealed 11 RabGAPs which when overexpressed reduced autophagy, and one, TBC1D14, which robustly co-localized with ULK1. TBC1D14 is present on Golgi and recycling endosomes (RE), and its trafficking is sensitive to amino-acid withdrawal. Chris Lamb and Minoo Razi working together on the function of TBC1D14 revealed that TBC1D14 is an effector of Rab11, a Rab which controls the RE compartment. Activated Rab11 is required for autophagy and during starvation REs containing the ULK complex are translocated to forming autophagosomes. Chris Lamb is now exploring how TBC1D14 coordinates RE traffic to autophagosomes. Our hypothesis is that this occurs through interaction with Rab11 and effectors of Rab proteins in a regulated manner. Atg9, the only membrane spanning Atg protein so far identified is, like the ULK complex, found on RE and as shown by Hannah Dooley interacts with transferrin receptor. However, Atg9 is also present on Golgi and late endosomes (Young et al., 2006; J Cell Sci. 119: 3888-900) and in a unique



compartment, the ‘Atg9 compartment’. Atg9positive vesicles transiently interact with both phagophore and autophagosome membranes, and we have hypothesized that one function of Atg9 is to allow a rapid expansion of the autophagosome during amino acid starvation (Orsi et al., 2012; Mol Biol Cell. 23: 1860-73). Studies initiated by Javier Herras and Harold Jefferies, in collaboration with the Cell Biophysics Group, have begun to address the possibility that the expansion mediated by Atg9 may be by delivery of lipids via vesicular transfer or proteinmediated transfer from Atg9-positive vesicles. Novel regulators of autophagy To identify novel regulators of autophagy we performed a siGenome wide screen under amino acid starvation and identified several putative candidates (McKnight et al., 2012; EMBO J. 31: 1931-46). SCOC, a short coil-coil protein, is a Golgi-localized protein that we showed interacts with ULK1 and UVRAG, dependent on FEZ1. Given that ULK1 and UVRAG (a subunit of the PI3K complex) act sequentially in autophagy, we are testing the hypothesis that SCOC may regulate the progression of the autophagosome membrane and maturation. Martina Wirth in collaboration with Stephane Mouilleron (Protein Structure unit) is currently exploring the SCOC complexes using a structure-function approach. A second novel autophagy regulator is WAC, a WW domain containing adaptor with coiled-coil protein, which we have shown is required for autophagy and may regulate ubiquitinated protein turnover. As WAC is found in both a nuclear and cytoplasmic pool, Justin Joachim is dissecting the function of WAC in these two compartments.

Publications listed on page 145

LINCOLN’S INN FIELDS

83

SIGNALLING AND TRANSCRIPTION www.london-research-institute.org.uk/research/richard-treisman

SRF (Serum Response Factor) is central to the cellular response to extracellular signals controlling cell proliferation and differentiation, and the regulation of muscle-specific gene expression. SRF associates with two types of regulatory cofactors, controlled by extracellular signals. Group Leader

Richard Treisman Associate Scientist Patrick Costello Postdoctoral Scientists Jessica Diring Cyril Esnault Charles Foster Anastasia Mylona Graduate Students Francesco Gualdrini Richard Panayiotou Maria Wiezlak Scientific Officers Diane Maurice De Coulon Mathew Sargent

The TCF cofactor family SAP-1, Elk-1 and Net is controlled by the classical Ras-ERK pathway; in contrast the MRTF proteins, MAL/MRTF-A and Mkl2/MRTF-B, are controlled by the Rho-actin pathway, in which actin itself functions as a signalling molecule. The Rho-actin pathway is thought to underpin the emerging role of SRF as a master regulator of cytoskeletal gene expression essential for processes such as cancer cell motility and invasiveness. The MRTFs are controlled by signal-dependent alterations in the interactions between G-actin and their regulatory RPEL domains, which arise through growth factor- or mechanical stress-induced fluctuations in cellular G-actin concentration. During the year our studies have established the generality of RPEL proteins as targets for regulation through Rho-actin signalling, and shed new light on the molecular mechanisms involved. Actin in control of MRTF activity Actin binding controls MRTF nucleocytoplasmic shuttling by regulating access of the importin α–β complex to nuclear import signals within the RPEL domain, and by facilitating Crm1-dependent MRTF nuclear export. Growth factor stimulation also induces phosphorylation of the RPEL domain. Richard Panayiotou has shown that this impairs actin binding, and is now testing the significance of this for the dynamics of MRTF nucleocytoplasmic shuttling. Actin-RPEL domain interaction also appears to regulate transcriptional activation by the MRTFs. We previously showed that blockade of the nuclear export factor Crm1 confines MRTF-A to the nucleus in resting cells without activating transcription of model target genes, even through chromatin immunoprecipitation analysis shows MRTF-A-SRF complexes bound to their

84

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

promoters. This transcription inhibition can be relieved by disruption of MRTF-A/actin interaction, for example by the actin-binding drug cytochalasin D. Interestingly, several established cancer cell lines also show constitutive MRTF nuclear localisation without transcriptional activation of SRF targets, and Charlie Foster is investigating the basis for this. Francesco Gualdrini is using two strategies to investigate the basis for this inhibition. Using RNAseq, he has characterised which genes are activitated by CD, irrespectively of blockade of Crm1, and showh that these genes are predominantly MRTF-SRF targets. He is now investigating how transcriptional activation is blocked, using ChIPseq approaches to assess recruitment of RNA polymerase II, components of the transcriptional machinery, and chromatin modifications. Francesco is using biochemical approaches to study how actin affects assembly of MRTF-SRF complexes in vitro complementing structural studies of MRTF-SRF-DNA complexes by Anastasia Mylona. The Phactr family and other RPEL proteins The Phactr family of PP1-binding proteins is implicated in human diseases including Parkinson’s, cancer and myocardial infarction. Each of the four Phactr proteins contains a conserved C-terminus required for interaction with PP1, and four G-actin binding RPEL motifs: an N-terminal motif, abutting a basic element, and a C-terminal triple RPEL repeat, which overlaps the PP1 binding sequences. Maria Wiezlak demonstrated that in fibroblasts, the subcellular localisation of each Phactr protein is predominantly cytoplasmic, but subtly different. Only Phactr1 exhibits Rho-actin regulated nuclear accumulation, and this was also observed with endogenous Phactr1 in the melanoma cell line CHL1. Phactr1 cytoplasmic localisation in resting

coordination of cellular behaviour with G-actin concentration. Jessica Diring is investigating the properties of other RPEL proteins identified through motif searches.

Figure 1 Down regulation of Phactr1 disperses actin stress fibres and focal adhesions in CHL-1 melanoma cells. Cells were treated with control or Phactr1 siRNAs and stained for F-actin (green) and paxillin (red).

cells requires the C-terminal RPEL motifs: RPEL mutants unable to bind actin are constitutively nuclear. Unlike the MRTFs, however, nuclear export of Phactr1 is independent of Crm1. Phactr1 nuclear accumulation is importin α–β-dependent, and with Jessica Diring, Maria showed that G-actin binds competitively with importin α–β to nuclear import signals associated with the N- and C-terminal RPEL motifs. All four motifs are required for G-actin to inhibit serum-induced Phactr1 nuclear accumulation. To gain insight into Phactr RPEL function Stephane and Maria conducted biochemical and structural studies. Gel filtration showed that Phactr1 Cterminal RPEL domain formed a trivalent complex with actin, which was crystallised, revealing a complex with one actin bound to each RPEL motif. Surprisingly, the relative orientations of the RPEL-actins in this complex are essentially identical to those of the RPEL-actins in the pentameric actin•MRTF complex previously characterised in the laboratory. This similarity reflects the formation of secondary actin contacts between the actin-bound RPEL motifs and their neighbouring actins, and Stephane showed that these contacts mediate cooperative actin binding to the Phactr1 C-terminal RPEL domain. Mutational analysis showed that secondary contacts between the N-terminal RPEL motif and the C-terminal domain mediate formation of the actin-inhibited state of Phactr1. Phactr1 C-terminal RPEL mutants that cannot bind G-actin induce aberrant actomyosin structures, and this requires their nuclear accumulation PP1 binding. With Jessica, Maria showed that G-actin binds competitively with PP1 to the Phactr1 C-terminal region. Maria showed that in CHL-1 melanoma cells, Phactr1 is required for stress fibre assembly, motility, and invasiveness, working with Jasmine Abella from the Cell Motility group. These findings suggest that Phactr1 G-actin sensing allows its coordination with F-actin availability, and more generally, that RPEL proteins allow



Global analysis of the SRF network There has been as yet no comprehensive assessment of the targets of the TCF and MRTF cofactor families in the growth factor response, and the significance of the SRF-linked signals for the transcription process – whether they control RNA polymerase recruitment or elongation, and at which step – has remained unclear. Cyril Esnault and Francesco Gualdrini are addressing these issues in collaboration with Aengus Stewart’s Bioinformatics and Biostatistics group. Cyril has used ChIPseq and RNAseq, using specific inhibitors to block signalling through the Rho-actin and Ras-ERK signalling pathways, and cytochalasin D to specifically activate MRTF signalling. Two-thirds of SRF binding events are increased upon serum stimulation, reflecting MRTF recruitment and increased nucleosome displacement. Cofactor recruitment by SRF is indeed gene-specific, with 70% of SRF sites recruiting MRTFs, 4% TCFs, 3% both, and 24% neither. Over 95% of MRTF binding events are SRF-associated, and SRF is thus their only effective target in this system. Cyril has identified over 500 direct target protein-coding genes for SRF linked signalling, associated with cytoskeletal processes (structure, adhesion, motility), cell growth and cell cycle, blood vessel development, hexose transport, transcriptional regulation, and core circadian regulators, consistent with prior findings from Ueli Schibler’s group that serum stimulation resets the fibroblast circadian clock. SRF also activates a number of long non-coding and miRNA precursor genes, which Cyril is investigating further. Patrick Costello and Diane Maurice have continued our analysis of the role of the SRF network in haematopoietic and T cell development in the mouse. Patrick has shown that SRF is required for effective colonisation of the bone marrow by haematopoietic stem cells during development, and is testing whether this reflects a defect in their response to chemokines. Diane has demonstrated enhanced memory cell formation in SAP-1 null animals, and is investigating the basis for this, focussing on the role of the SAP-1 target gene Egr-2.

Publications listed on page 146

LINCOLN’S INN FIELDS

85

CHROMOSOME SEGREGATION www.london-research-institute.org.uk/research/frank-uhlmann

Group Leader

Frank Uhlmann Postdoctoral Scientists Celine Bouchoux Sebastian Heeger Yasutaka Kakui Thomas Kuilman Lidia Lopez Serra Yasuto Murayama Graduate Students Vanessa Borges Adrian Charbin Molly Godfrey Meghna Kataria Ainhoa Mariezcurrena Rahul Thadani Scientific Officer Maria Ocampo-Hafalla Masters Student Noemie Scheidel

Aneuploidy, i.e. missing or supernumerary chromosomes in the cell nucleus, is a hallmark of malignant tumour progression. A large number of genes that orchestrate faithful chromosome segregation during mitotic cell divisions are tumour suppressors or turn into potent oncogenes if misregulated. The aim of the Chromosome Segregation Group is to investigate the function of these genes and the cellular mechanisms that safeguard accurate chromosome segregation. In particular, we are investigating the contribution of structural chromosomal proteins to sister chromatid cohesion and chromosome condensation, processes that ensure faithful segregation of centimetre-long chromosomal DNA molecules within micrometre-sized cells. We also investigate how the kinases and phosphatases of the cell division cycle machinery bring about ordered progression though mitosis. Chromosome resolution and tumourigenesis The chromosome replication and segregation cycle forms the basis for the inheritance of a cell’s genetic information from one generation to the next. Mistakes in the process not only lead to missegration of chromosomes and thereby gain or loss of genomic information, they can also have serious knock-on effects on other aspects of cell division. This is in particular true for chromosome resolution in anaphase. The genomic DNA that makes up the chromosomes is replicated during S-phase of the eukaryotic cell cycle. After its replication, as a consequence of the reactions involved in replication termination, the two sister chromatids remain entangled with each other by what is called topological catenation. This catenation has to be removed by an enzyme, called topoisomerase II, which transiently cleaves DNA strands to allow the intertwinings to resolve. Topoisomerase II is therefore an essential protein that facilitates sister chromatid resolution in anaphase. If resolution fails, chromosome bridges persist during anaphase. These will get into the way of the cytokinetic furrow that constricts to separate the two new daughter cells. In response, the furrow often retracts and cytokinesis fails,

86

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

leaving behind a tetraploid progeny cell that carries both sets of sister chromatids. It is thought that these tetraploid cells are a major source for the development of tumours that are characterised by further chromosomal aberrations. The condensin complex and chromosome resolution The condensin complex is a major structural component of mitotic chromosomes. Ongoing research in our laboratory investigates how condensin makes possible the compaction of centimeter-long DNA strands into micrometersized chromosomes. In addition to its hallmark role in chromosome compaction, condensin is also required for chromosome resolution. Cells lacking condensin display anaphase chromosome bridges reminiscent of cells lacking topoisomerase II. The reason for these anaphase bridges has remained controversial, various hypotheses that could explain them have been put forward. We tested whether anaphase bridges in the absence of condensin might be the consequence of persistent sister chromatid catenation in mitosis. To do so, we developed an assay to follow the catenation status of sister chromatids as

Figure 1 Condensin promotes resolution of sister chromatid catenanes following their production in S-phase. A. Cells containing, or depleted of, condensin were synchronised in the G1 phase of the cell cycle and released for synchronous progression through S-phase and mitosis. FACS analysis of DNA content shows cell cycle progression through S-phase and return to G1 following mitosis and cytokinesis. In the absence of condensin, chromosome segregation fails, leading to daughter cells with abnormal DNA content. A 21.2 kb circular minichromosome (pS14-8), that the cells contained, was detected by Southern blotting using a minichromosome-specific probe. The topological isoforms of the minichromosome are indicated to the right (RC, relaxed catenanes; MC, mixed catenanes; RM, relaxed monomer; SC, supercoiled catenanes; L, linear form; SM, supercoiled monomer). Catenated forms appear during S-phase, their complete resolution by the time of mitosis depends on the presence of condensin. C. Immunofluorescent staining of the mitotic spindle using an α-tubulin antibody and visualisation of the chromosomes using the DNA stain 4’,6-diamidino-2-phenylindole (DAPI) shows anaphase bridges in the absence, but not in the presence, of condensin.

A

B

C

budding yeast cells pass through the cell division cycle. Because the catenation status of linear chromosomes cannot be studied using conventional gel electrophoretic techniques, we utilised a large circular minichromosomes in budding yeast. This minichromosome has many of the same features as authentic yeast chromosomes but, in addition, its catenation status during the cell cycle can be followed by gel electrophoresis (Figure 1). Using a series of enzyme treatments, we assigned the numerous topological forms of the minichromosome that appear and disappear during cell cycle progression. We confirmed that catenated forms of the minichromosome are the product of DNA replication in S-phase, and that these catenanes are resolved again with the help of topoisomerase II. A large fraction of catenanes resolve very soon after DNA replication, followed in wild type cells by the last catenanes that disappear around the time of mitosis. Strikingly, when we inactivated condensin catenation of a subset of the replication products again proceeded swiftly, however a substantial fraction of catenanes failed to be resolved and persisted throughout mitosis and into the beginning of the next cell cycle. At the same time, we saw that chromosome segregation remained incomplete and anaphase bridges stretched across the cell cleavage site. This demonstrates that condensin is required to promote complete decatenation of sister chromatids during mitosis, offering an explanation for the chromosome segregation failure and anaphase bridges in its absence.



Outlook What is the mechanism by which condensin promotes sister chromatid decatenation? While we found that the chromosomal localisation pattern of the two protein overlaps, we were unable to detect a direct protein interactions between the two players. It therefore seems likely that DNA reconfiguration by condensin helps to provide a better substrate on which topoisomerase II can act. In addition, our observation that some, but not all, catenanes require condensin for their resolution raises the question as to what types of topological forms are aided in their resolution by condensin. Are they complex intertwinings, like knots, that only occasionally form? A better characterisation of the nature of topological links between sister chromatids will be required to answer these questions. Another important question to address is the origin and fate of catenanes along authentic, long linear chromosomes. We will need to develop techniques to isolate subdomains of linear chromosomes and covert them into circular form, to be able to assess their catenation status over time. Together this will aid our understanding of how very long DNA molecules that make up eukaryotic genomes are duplicated and neatly segregated while avoiding persisting entanglements that could set off a series of unwanted events leading to tumourigenesis.

Publications listed on page 146

LINCOLN’S INN FIELDS

87

PROTEIN STRUCTURE AND FUNCTION www.london-research-institute.org.uk/research/helen-walden

Group Leader

Helen Walden Postdoctoral Scientists Julio Martinez-Torres Ateesh Sidhu Graduate Students Charlotte Hodson Jennifer Miles Scientific Officer Lynn Burchell

My lab aims to unravel the molecular mechanisms of ubiquitination. Specifically, we aim to understand substrate recognition, E3 ligase regulation and cooperation between E2-conjugating and E3-ligase enzymes to assemble the correct ubiquitin signal. We study two medically relevant ubiquitin ligases that sit at either end of the scale in terms of specificity and selectivity. First, the multi-subunit Fanconi Anemia (FA) core complex, is a model for exquisite specificity and selectivity as it is responsible for the monoubiquitination of two structurally-related substrates. Second, Parkin, a single polypeptide RING-InBetweenRING-RING (RBR) ligase, has a diverse array of putative substrates, collaborates with multiple E2s and mediates several different types of modification.

We aim to define the rules governing tolerance and promiscuity in substrate selection and modification, and to establish paradigms for regulation of E3s. Understanding how these very different E3 ubiquitin ligases function, on a molecular level, will provide insights into their associated disease states as well as general mechanisms of ubiquitination. Ubiquitination is a post-translational modification through the covalent attachment of the 76 amino acid ubiquitin to a substrate. This signal determines the cellular fate of the target and drives many biological processes including cell cycle progression, DNA damage repair, and endocytosis. Dysfunction in the ubiquitin system is frequently associated with disease states, including cancer and neurodegeneration. Outstanding questions in the field relate to how specificity and promiscuity towards substrates to be ubiquitinated are encoded into the conjugation machinery, including; 1) how is a specific ubiquitination substrate recognised, 2) how is a given modification selected, 3) how is the ubiquitin conjugation machinery regulated.

88

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

The Fanconi Anemia core complex The first model we use is the Fanconi Anemia (FA) core complex. The FA DNA repair pathway is a multimembered, complex pathway required for the repair of interstrand crosslinks (Figure 1). A critical step in the pathway is the specific monoubiquitination of only two substrates, FANCD2 and FANCI. This is achieved by the FA core complex, which has at least 10 subunits, one of which has ubiquitin ligase activity, FANCL. We have already solved the structure of FANCL and determined functions for the distinct FANCL domains. We now wish to understand how mutations in any member of the core complex lead to the loss of FANCD2-ubiquitination, and consequently, Fanconi Anemia. We are taking several approaches to address this question. We have developed and optimised robust in vitro assays to identify what components of the core complex may stimulate or inhibit FANCLdependent ubiquitination of FANCD2 (Figure 2A). We are also characterising the complex structurally and biophysically as nothing is yet known about the stoichiometry or assembly of the core complex. These are ambitious aims, but we have many of the materials now in hand, including milligramme quantities of each of the components, and subcomplexes of the FA pathway (Figure 2 B, C).

Figure 1 The Fanconi Anemia DNA repair pathway. A model of the FA pathway at a stalled DNA replication fork (grey), caused by an interstrand crosslink (ICL). FANCM and its associated proteins are green, which assemble on DNA at the fork. The core complex is in blue, with the ligase subunit in mauve. Substrate FANCI and FANCD2 are in orange/yellow, with ubiquitin in purple.

The RBR ligase, Parkin Our second model is the RING-InBetweenRINGRING ligase Parkin. In contrast to the FA core complex, Parkin is a single polypeptide with a vast array of unrelated putative substrates (more than 30 currently reported), interactors, E2s, and ability to catalyse multiple modifications. Recent data from ourselves and others shows the RBR ligases are a distinct class of E3s, that are not yet mechanistically understood in terms of ubiquitin transfer, substrate selection and regulation. Therefore Parkin is an excellent model for obtaining fundamental insights into RBR ligase function, investigating promiscuity in substrate recognition and understanding E3 regulation. In addition it is an important protein to understand in the context of its related disease states, neurodegeneration and cancer. Originally, mutations in parkin were identified as a genetic component of Parkinson’s disease. Parkinson’s disease is a neurodegenerative disorder characterised by the loss of dopaminergic neurons from the substantia nigra, and the presence of Lewy Bodies, which are aggregated inclusions rich in ubiquitin and α-synuclein. Autosomal Recessive Juvenile Parkinsonism (ARJP) is one of the most common familial forms of

A

Figure 2 A. Ubiquitination of FANCD2 by the core complex. B. Purified components of the FA core complex. C. Purified stable subcomplex of 4 FA proteins.



B

Parkinson’s disease, and 50% of cases are linked to mutations in the parkin gene. Mutations and alterations in parkin are not only found in ARJP, but also occur frequently in a wide range of malignancies including glioblastoma, breast, and lung cancer. These observations have led to the suggestion that Parkin is a putative tumour suppressor. Consistent with this, parkin-null mice display increased carcinomas. Current evidence also suggests a pivotal function for Parkin in the clearance of damaged mitochondria during mitophagy. We have established cell-based and in vitro assays for assaying Parkin activity. We have also expressed and purified many of the pathogenic mutations (there are 77 known point mutations in Parkin that lead to ARJP), in order to characterise the behaviour of Parkin mutations to better understand the disease. We are also testing the hypothesis that Parkin functions through adaptors to recruit potential substrates. We are also investigating the molecular basis of the Parkin/ PINK1 (PINK1 is Pten-induced kinase 1, and is also mutated in some cases of ARJP) interaction that is required for clearance of mitochondria. Publications listed on page 146

C

LINCOLN’S INN FIELDS

89

CELL MOTILITY www.london-research-institute.org.uk/research/michael-way

Group Leader

Michael Way Postdoctoral Scientists Jasmine Abella David Barry Joseph Cockburn Yutaka Handa Graduate Students Sara Donnelly Charlotte Durkin Chiara Galloni Ashley Humphries Xenia Snetkov Scientific Officers Theresa Higgins Antonio Postigo

90

Viruses hijack and subvert a diverse range of cellular processes from the moment they enter a cell until their newly replicated progeny leave. Understanding exactly how they take advantage of their host offers a wonderful opportunity to obtain fundamental insights into a wide variety of cellular processes. We are using a variety of quantitative imaging and biochemical approaches to study how vaccinia virus uses its unwilling host as a model system to understand the regulation and function of Src and Rho GTPase signalling networks, microtubule-based transport, cytoskeletal dynamics and cell migration. In addition, we are also investigating a number of other topics. These include the cellular function of Tes, a tumour suppressor that negatively regulates Mena-dependent cell migration as well as the mechanisms which control invadopodia formation and function. Signalling networks and regulation of actin polymerization The correct spatial and temporal regulation of actin polymerization by phospho-tyrosine based signalling networks is essential for many fundamental cellular processes, including cell migration during development and throughout the lifetime of multicellular organisms. Unfortunately, misregulation of phospho-tyrosine based signalling networks can be devastating for example if it stimulates tumour cells to undergo metastasis and establish new tumours. Understanding how phospho-tyrosine based signalling networks promote actin polymerization is a formidable task. Central to this challenge is identifying how specificity is generated and regulated within a network by a common set of modular domains with modest binding affinities and multiple interaction partners. Any analysis is further complicated by higher order oligomerization states and the co-operative nature of interactions within the network. Lastly, signalling networks are not static, but highly dynamic systems that are constantly adapting to the ever changing needs of the cell. Unravelling exactly how signalling networks promote actin-based motility will therefore require a molecular understanding and detailed knowledge of the

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

interactions, dynamics and stoichiometry of proteins in the network. Unfortunately, many signalling networks controlling actin polymerization are not readily amenable to such quantitative analyses as their components and/or activation is often transient and dispersed. In contrast, the signalling pathway that is hijacked by vaccinia virus to induce actin polymerization is highly localized and sustained (Weisswange et al., 2009; Nature. 458: 87-91). Clathrin enhances vaccinia induced actin polymerization During vaccinia virus infection, cell-associated enveloped viruses (CEV) attached to the plasma membrane are able to induce Arp2/3 complexdependent actin polymerization to enhance their spread into neighbouring cells. CEV stimulate actin polymerization by inducing an outside-in signalling cascade that locally activates Src and Abl family kinases. Activation of these kinases results in the phosphorylation of tyrosine 112 and 132 of A36, an integral viral membrane protein that becomes localized beneath CEV when the virus fuses with the plasma membrane. Phosphorylation of tyrosine 112 and 132 of A36 leads to the recruitment of a signalling network, consisting of Grb2, Nck, WIP and N-WASP, the latter of which

actin polymerization and the actin tails that do form are less stable, which ultimately leads to reduced viral spread. Counter intuitively, the loss of AP-2 or clathrin recruitment results in longer and faster actin tails that disassemble slower. These morphological changes suggest that the individual actin filaments within the tail are longer and require more time to disassemble given the increased speed of the virus. The faster rate of virus movement also suggests that the net actin polymerization beneath the extracellular virus must be increased. The question is how does AP-2 and clathrin recruitment brings about these changes in behaviour when they are not associated with CEV undergoing actin-based motility?

Figure 1 A structured illumination image acquired using an OMX microscope, which shows three vaccinia particles (red) each nucleating an actin tail (red), the A36 viral protein is tightly clustered beneath the virus at the site of actin nucleation (green).

activates the Arp2/3 complex to promote robust actin polymerization beneath the virus. The signalling network that vaccinia uses to stimulate actin polymerization is at the heart of a number of cellular processes, including the formation of invadopodia during tumour cell invasion. Vaccinia therefore provides an excellent model to understand exactly how a co-operative phosphotyrosine-based Nck and N-WASP signalling network functions to stimulate actinpolymerization. Indeed, our recent analysis of the dynamics of the proteins in the vaccinia-signalling network using Fluorescence Recovery After Photobleaching (FRAP) has shown that the stability of N-WASP in the system regulates the rate of Arp2/3 complex-dependent actin-based motility (Weisswange et al., 2009). We have made great progress in understanding how vaccinia stimulates actin polymerization. However, the events that occur during and immediately after the fusion of the virus with the plasma membrane prior to the activation of Src and actin polymerization are still largely unknown. We have now found that vaccinia recruits clathrin in an AP-2 dependent fashion immediately after it fuses with the plasma membrane (Humphries et al., 2012; Cell Host and Microbe. 12: 346-59). Live cell imaging, however, reveals that when the virus initiates actin-based motility the clathrin does not remain associated with CEV. Nevertheless, the transient recruitment of clathrin impacts on the ability of the virus to induce and sustain actin-based motility. In the absence of clathrin recruitment it is harder for the virus to stimulate



Using structured illumination microscopy we have found that AP-2 recruitment induces increased clustering of A36 and N-WASP at the site of actin tail formation. The increased clustering of N-WASP in turn enhances its stability, leading to more efficient activation of the Arp2/3 complex and robust actin polymerization. We envisage that there is a threshold and optimal density of phosphorylated A36 required to initiate and sustain actin tails. In the wild type situation, clathrin helps facilitate a level of A36 clustering that assembles an organized signalling network sufficient to induce sustainable actin polymerization. In the absence of clathrin recruitment, fewer viruses reach the threshold of A36 clustering required to efficiently nucleate actin polymerization. Moreover, even when the threshold is achieved, the organization of the signalling network is sub-optimal, resulting in an altered output and changes in actin tail morphology and stability. Consistent with this suggestion, we found that artificially reducing the number of phosphorylation competent A36 molecules on the virus phenocopies the loss of clathrin recruitment, as actin tails are longer and faster. Further studies aimed at analysing how the density of A36 directly correlates with signalling network dynamics and output offer an excellent opportunity to further our understanding of the initiation and regulation of phospho-tyrosine based signalling networks, as well as the parameters that define continuous actin propulsion in a freely exchanging system.

Publications listed on page 146

LINCOLN’S INN FIELDS

91

The cellular actin cytoskeleton (yellow) and DNA (blue) in a human fibroblast. Image: Mark Petronczki, Cell Division and Aneuploidy Group.

92

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

CLARE HALL

The London Research Institute Clare Hall Laboratories are located North of London. The main focus of the research for the 10 Laboratories housed on the Clare Hall campus is genome integrity; including DNA repair, recombination and replication, cell cycle control and transcription. The researchers are supported by an excellent range of Technology Core Facilities.



CLARE HALL

93

DNA DAMAGE RESPONSE www.london-research-institute.org.uk/research/simon-boulton

Group Leader

Simon Boulton Postdoctoral Scientists Carrie Adelman Ross Chapman Kerstin Gari Ana Maria Leon Ortiz Zuzana Licenikova-Horejsi Grzegorz Sarek Jennifer Svendsen Antonia Tomas-Loba Jean-Baptiste Vannier Graduate Students Rafal Lolo Martin Taylor Scientific Officers Valerie Borel-Vannier Julie Martin Mark Petalcorin

DNA is highly susceptible to damage and must be repaired correctly to prevent genome instability. Failure to correctly repair DNA damage is the underlying cause of a number of hereditary cancer predisposition syndromes such as Fanconi anemia and Blooms. The long-term aim of my lab is to understand how DNA double-strand break (DSB) repair pathways, such as non-homologous end joining (NHEJ) and homologous recombination (HR), are regulated in mitotic cells and during meiosis. We also have an active interest in understanding how these pathways impact on human diseases such as cancer. Links between Fe-S biogenesis and genome stability The function of many DNA metabolism proteins depends on their ability to coordinate an iron-sulfur (Fe-S) cluster. Biogenesis of Fe-S proteins is a multi-step process that takes place in mitochondria and the cytoplasm, but how it is linked to nuclear Fe-S proteins is not known. MMS19 is a highly conserved HEAT-repeat protein that was first identified in budding yeast as being required for the removal of (UV)-induced pyrimidine dimers. mms19Δ cells are impaired for both nucleotide excision repair (NER) and RNA polymerase II transcription and display reduced protein levels of the TFIIH component Rad3. Human MMS19 also interacts with two components of TFIIH and is also believed to play a role in mitotic spindle formation and chromosome segregation. Surprisingly, Fission yeast Mms19 is part of a different complex, which links DNA replication to heterochromatin silencing. A molecular explanation of how MMS19 impacts on so many different processes has remained elusive. We found MMS19 as a novel interaction partner of the Regulator of Telomere Length protein RTEL1. As RTEL1 belongs to the same family of Rad3-like SF2 helicases as XPD, the Fanconi Anemia protein J (FANCJ), and the Warsaw Breakage Syndrome protein ChlR1, which all bind an Fe-S cluster, we reasoned that MMS19 may be involved in Fe-S biogenesis. Indeed, we found that MMS19 forms a complex with the cytoplasmic Fe-S assembly (CIA)

94

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

proteins CIAO1, IOP1, and MIP18. Cytoplasmic MMS19 also binds to multiple nuclear Fe-S proteins involved in DNA metabolism, including RTEL1, XPD, FANCJ, DNA polymerase ε, DNA primase, DNA2 and others. In the absence of MMS19, a failure to transfer Fe-S clusters to target proteins is associated with Fe-S protein instability and pre-implantation lethality of Mms19 knockout mice. Based on these observations, we propose that MMS19 functions as part of the CIA machinery to facilitate Fe-S cluster transfer to target Fe-S proteins (Figure 1). These findings provide molecular insight to explain the previously reported phenotypes associated with MMS19 deficiency (such as DNA repair and proliferation defects) and why defects in mitochondrial Fe-S cluster biogenesis confer genome instability (Gari et al., 2012; Science. 337: 243-5). Distinct roles for RTEL1 in maintaining vertebrate telomeres RTEL1 is a helicase that was originally identified by mapping of loci that control telomere length differences between M. musculus and M. spretus. RTEL1 plays a critical role in genome stability as knockout mice are embryonic lethal and cells derived from these mice exhibit telomere fragility and loss (Figure 1). Despite considerable interest in the telomere field the mechanistic basis of this defect was not known. We previously identified RTEL1 as a key regulator of HR in a genetic screen for anti-recombinases. rtel-1 mutant worms and RTEL1 depleted human cells are hyperrecombinogenic and DNA damage sensitive and

Figure 1 MMS19 functions with the cytoplasmic Fe-S assembly pathway to facilitate the incorporation of Fe-S clusters into target proteins, many of which play essential roles in DNA metabolism.

Figure 2 RTEL1 unwinds T-loop structures to prevent catastrophic telomere processing by the SLX4 resolvasome. In RTEL1-/- cells, persistent T-loops are excised from chromosome ends by the SLX4 complex resulting in the formation of T-circles.

biochemical studies revealed that human RTEL1 promotes the disassembly of D loop recombination intermediates in vitro. More recently, we established that RTEL-1 is required to limit excessive crossing over during C. elegans meiosis. Our findings raised the possibility that the phenotype of the RTEL1 deficient mice reflects a role for RTEL1 in controlling HR at telomeres. Telomeres are specialized protein-DNA complexes that function to protect the end of linear chromosomes. This is achieved in several ways; telomeres recruit and regulate telomerase to solve the end replication problem, the telomere masks the DNA end from being detected by the DNA damage checkpoint, and telomere-binding proteins facilitate replication through telomeric repeats. It has been proposed that telomeres protect chromosome ends by sequestration of the 3' telomeric end into sub-telomeric duplex TTAGGG repeats resulting in the formation of a T loop structure. Formation of the T loop structure requires both HR and telomere specific proteins



for its assembly and is therefore believed to resemble a D loop HR intermediate. Because human RTEL1 can disassemble D loops in vitro, it is possible that the stochastic loss of telomeres in Rtel1 deficient cells may reflect a failure to correctly disassemble the T loop structure. Using conditional RTEL1 knockout mouse cells we have shown that T-loop sized telomeric circles (TCs) accumulate in Rtel1-/cells coincident with telomere loss. Inhibiting DNA replication or depleting the SLX4 nuclease complex blocks TC formation and rescues telomere loss in Rtel1-/- cells. In contrast, telomere fragility in Rtel1-/- cells is SLX4 independent and is exacerbated by inhibiting replication or stabilizing G-quadruplex structures that are known to form at telomeres. Consistent with these findings we have found that RTEL1 efficiently binds to and unwinds G4-DNA structures in vitro. Collectively, our results define a general mechanism of TC formation in cells and reveal two distinct functions for RTEL1 in telomere maintenance: it disassembles T-loops to prevent catastrophic telomere excision by the SLX4 complex and also interacts with the replisome to facilitate replication of telomeric repeats (Figure 2. Vannier et al., 2012; Cell. 149: 795-806). We are currently employing genetic screens in C. elegans, proteomic approaches in mammalian cells and mouse models to investigate the regulation of RTEL1 function at telomeres and during semi-conservative DNA replication.

Publications listed on page 147

CLARE HALL

95

CHROMATIN STRUCTURE AND MOBILE DNA www.london-research-institute.org.uk/research/peter-cherepanov

Group Leader

Peter Cherepanov Postdoctoral Scientists Siobhan Hughes Paul Lesbats Daniel Maskell Graduate Student Samual Dick Masters Student Ofori Jones Scientific Officers Nicola Cook Valerie Pye

Our laboratory focuses on the structural biology of chromatin function and its interactions with retroviral DNA integration machinery. Using X-ray crystallography and complementary biochemical approaches we aim to elucidate three-dimensional structures and mechanisms of biological machineries involved in regulation of eukaryotic DNA replication, gene expression and retroviral integration. This year we reported a series of crystal structures of human Cdc7 kinase, which is one of the key players in regulation of cell cycle progression. These structures will aid in the development of cancer therapeutics that act by blocking activity of Cdc7. Retroviral integration machinery and its interactions with chromatin and cellular factors Retroviruses are unique in their ability to efficiently insert their genomes into host cell DNA. This property makes them ideal vehicles for gene therapy. Yet, uncontrolled integration of retroviral vectors poses an inherent risk of insertional mutagenesis and consequently, serious side effects. The catalytic events associated with integration are carried out by the virus-derived enzyme integrase (IN). Acting during early stages of the viral replication cycle, IN is responsible first for 3'-processing, the reaction in which two or three nucleotides are removed from 3'-viral DNA ends, leaving 3'-hydroxyl groups. IN subsequently catalyzes strand transfer, wherein it uses the 3'-hydroxyls to attack a pair of phosphodiester bonds in host cell DNA. Retroviral IN is structurally and mechanistically related to a diverse group of nucleotydiltransferases that includes bacterial and eukaryotic transposases, V(D)J recombinase RAG1, ribonuclease H, and the catalytic subunit of the RNA-induced silencing complex (RISC) Ago. The superfamily is distinguished by the characteristic structural fold, organization of the active site and the mechanism of metal-dependent catalysis. As one of the three essential HIV-1 enzymes, IN is an important target for anti-HIV/AIDS drug development. This year we published a series of snapshots of a retroviral IN catalyzing the 3'-processing and strand transfer reactions within crystals (Hare et

96

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

al., 2012; EMBO J. 31: 3020-8). The new structural data provided unprecedented insights into the positions of the metal ions and chemical reacting groups in the IN active site (Figure 1). This highlighted a substrate mimicry utilized by clinical HIV-1 strand transfer inhibitors in their mode of binding to IN and explained why these small molecules have been ineffective against the 3'-processing reaction. Based on the structures we suggested potential ways of improving strand transfer inhibitors. We are now investigating the relationship between retroviral integration and the chromatin structure. Currently, our main questions are how retroviral IN-DNA complexes capture native nucleosomes and how they interface with the cellular chromatin remodeling machinery. The answers will eventually help to identify novel drug targets in the battle with HIV and to improve vectors for gene therapy applications. Initiation and control of eukaryotic DNA replication Eukaryotic chromosomal DNA replication is initiated at multiple origins at the onset of and throughout S phase. Replisome assembly and initiation of DNA synthesis at individual origins critically depend on activities of S-phase cyclindependent kinases and Cdc7. Both types of kinases are regulated by their respective activating subunits (cyclins and Dbf4, respectively). Phosphorylation of MCM2-7 by Cdc7 allows

Cdc7 and Dbf4 are overexpressed in many cancers and tumour cell lines. Due to its pivotal role in cell proliferation, Cdc7 is emerging as a target for the development of cancer therapeutics. To-date, several classes of ATP competitive inhibitors of Cdc7 have been reported, and at least two compounds have advanced to clinical trials. The development of selective kinase inhibitors is complicated by the sheer size of the human kinome and conservation within kinase active sites. Fortunately, structural information can greatly aid in inhibitor design and optimisation.

Figure 1 Active site of IN freeze-trapped prior to 3'-processing. IN is shown as cartoons. Selected amino acid side chains and viral DNA are shown as sticks. The scissile phosphodiester bond is indicated with an arrowhead. Metal ions (Mn2+) and associated water molecules are shown as purple and red spheres, respectively. The water molecule participating in the reaction as a nucleophile is indicated (Wnuc). Adapted from Hare et al., 2012.

recruitment of essential replication initiation factors en route to replisome assembly. In addition, Cdc7 plays important roles in the intra-S-phase checkpoint, chromosome cohesion, mitotic exit and meiosis. This year we reported a crystal structure of the human Cdc7-Dbf4 heterodimer, which elucidated the active form of this essential S-phase kinase (Hughes et al., 2012; Nat Struct Mol Biol. 19: 1101-7). The structure revealed that Dfb4 wraps around Cdc7, burying some 6,000 Å2 of hydrophobic molecular surface in a bipartite interface with Cdc7 (Figure 2). Phylogenetically conserved Dbf4 motifs -M and -C interact with the C and N lobes of Cdc7, respectively. Using a combination of biochemical approaches, we further demonstrated that motif-C comprises the effector domain of Dbf4 stabilizing the canonical αC helix of Cdc7 in the active conformation.

To help drug developers, we co-crystallized the Cdc7-Dbf4 complex with a series of ATPcompeting small molecules, including PHA767491 and XL413. The resulting structures elucidated interactions the compounds make in the active site of Cdc7 (Figure 2) and will be invaluable for the development of more specific and potent ATPcompeting Cdc7 inhibitors. Furthermore, the details of the Cdc7-Dbf4 interface may allow development of small molecules targeting pockets on the surface of the catalytic subunit that are involved in interaction with Dbf4. Engaging less conserved regions of the kinase, such inhibitors will far less likely suffer from unwanted off-target activities. Our structures are already helping Cancer Research Technology in their Cdc7 inhibitor development program. We are now focusing on finer details of regulation of Cdc7 activity during normal S phase and cell cycle arrest. We are also interested in other essential factors involved in the initiation and control of eukaryotic DNA replication.

Publications listed on page 147

Figure 2 Overall structure of the Cdc7-Dbf4 heterodimer (left) and views of the Cdc7 active site engaged with inhibitors (right). On the left panel, the protein chains are shown as cartoons with Cdc7 coloured in green (canonical part of the N lobe structure), purple (canonical C lobe structures) and yellow (structures unique to Cdc7) and Dbf4 in orange; Cdc7 αC helix is highlighted in dark green. On the right panel, Cdc7 is shown is space-fill mode and PH767491 or XL413 as sticks; Cdc7 surface is coloured according to conservation, with least conserved atoms are shown in red and those most conserved in grey. Adapted from Hughes et al., 2012.



CLARE HALL

97

ARCHITECTURE AND DYNAMICS OF MACROMOLECULAR MACHINES http://www.london-research-institute.org.uk/research/alessandro-costa

Group Leader

Alessandro Costa Postdoctoral Scientist Jin Chuan Zhou Graduate Student Paolo Swuec Scientific Officer Adelina Davies

A hallmark of cancer cells is genomic instability, arising from errors in the mechanisms that maintain gene copy number and chromosome ploidy. Our research aims to understand how macromolecular machines involved in DNA replication and repair function to preserve chromosome integrity. To address these issues, we employ a combination of single-particle Electron Microscopy, Molecular Modelling and Biochemistry, to generate mechanistic models that explain the basis of key nucleic acid transactions. For example, we are interested in understanding how DNA unwinding and synthesis are coupled during chromosome duplication or how homologous recombination intermediates are processed to ensure that genome stability is maintained. By describing the architecture and dynamics of important DNA manipulation machineries, we seek to establish a molecular framework that explains how higher eukaryotes respond to DNA damage and how cell proliferation is regulated to avoid tumorigenesis. DNA replication – Coordinating DNA unwinding and synthesis In all domains of life the replicative helicase is a ring-shaped, hexameric ATPase motor that is generally loaded onto an origin of replication to unwind duplex DNA and provide the singlestranded DNA template for replicative polymerases. The eukaryotic helicase has three interesting features. First, it is the only replicative helicase composed of six distinct polypeptides (called Minichromosome maintenance complex, Mcm2-7), which have evolved to interact with different components within the replication machinery (the Replisome). Second, unlike in other systems, helicase loading at eukaryotic origins is not sufficient to initiate DNA replication in vivo: rather, additional events (covalent modifications/recruitment of activating proteins) are needed to switch on the DNA unwinding function. Third, the isolated motor assembly has poor helicase activity in vitro, while DNA unwinding is greatly stimulated by five ancillary factors called Cdc45 and the tetrameric GINS complex. Single-particle Electron Microscopy has been used to describe the mechanism of

98

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

activation of the eukaryotic replicative helicase motor. The isolated Mcm2-7 hexamer adopts an open configuration with a discontinuity at the Mcm2/5 interface. When associated with its activators however, the Mcm2-7 complex is locked into a planar configuration and Cdc45/GINS bind to the side of this motor assembly effectively sealing off the Mcm2-5 gate. This event is required to switch on the motor’s ATPase and translocase functions. The structure of the CMG helicase represents an initial step towards obtaining an architectural description of the eukaryotic replication machinery at the fork and raises a new set of questions. For example: how does the replicative helicase pause when it senses DNA damage, to allow intervention by the DNA repair machinery? Also, how is fork restart regulated and how are DNA unwinding and DNA synthesis coupled at the replication fork? To address these questions, we are studying the architectural relationships among the helicase and the three replicative polymerases Pol alpha, delta and epsilon.

The key player in this process is the four-member Dissolvasome complex, including Topoisomerase IIIα, the RMI1/2 factors and the BLM helicase (Figure 1), whose mutation is linked to genetic disease and cancer development.

A key role in this context is played by the Replication Pausing Complex, including Tipin, Tim1, Claspin and And1, structural proteins that tether the CMG helicase to the replicative polymerases and couple their activities. These factors are of primary importance for the maintenance of chromosomal integrity under replication stress conditions, as they keep the CMG from translocating, when replicative polymerases stall. Describing the structure of the Replication Pausing Complex, alone or associated with isolated helicase/polymerase members, will help understand the mechanisms of fork progression, stalling and restart.

Figure 1 The Dissolvasome complex. The atomic structures of various subunit domains are available, but little is known about their supramolecular architecture and functional interplay.

While the mechanism of Holliday Junction dissolution is still unclear, a wealth of information is available, on various orthologs of the Dissolvasome assembly. For example, Topoisomerase IIIα belongs to the type-IA class of topoisomerases, pad-lock shaped enzymes that effect changes in DNA topology through a ‘strandpassage’ mechanism. Here, a DNA single strand is cleaved and physically opened, a second DNA segment is passaged through the gate, after which the broken strand is resealed. RMI1 and RMI2, OB-fold containing factors, have been implicated in nucleic acid engagement or in protein-protein interactions. A recent study in yeast indicates that RMI1 is instrumental in stabilising a covalentlybound, open form of the Topoisomerase IIIα DNA gate, a configuration that would favour DNA-strand passage. BLM contains a RecQ-type DNA helicase domain, whose ATPase function is required for DNA opening and translocation. Remarkably, the presence of BLM but not its hydrolase activity is needed to lock the topoisomerase gate. This observation indicates that BLM has a structural, rather than catalytic role that effects DNA strand passage and suggests that BLM’s roles in branchmigration and decatenation might be uncoupled.

DNA repair – Mechanism of DNA-crossover suppression Double-strand breaks can be repaired by homologous recombination in all cells. At the end of this process Holliday Junctions, covalent linkages between donor sequences, can be resolved by the action of a nuclease. This event results in the exchange of genetic information between two DNA segments (DNA crossover), which can lead to the rise of deleterious mutations. To prevent this, eukaryotes have developed a strategy to dissolve homologous recombination intermediates back to their pre-recombination state (DNA-crossover suppression).

RMI2 RMI1 (CTD) Topoisomerase IIIα RMI1 (NTD)

BLM (CTD)

BLM Helicase

Despite these recent advances, many issues remain unresolved and a mechanistic understanding of Holliday Junction dissolution is still lacking. For example, i) how Topoisomerase IIIα and BLM are related in space remains obscure, ii) the role of RMI1/2 in the decatenation reaction is unclear iii) whether ATP binding and hydrolysis by BLM affect Topoisomerase IIIα movements is unknown. To address these issues, we have initiated a study of the human Dissolvasome architecture in various functional states. We aim to provide a complete mechanistic description of Holliday Junction dissolution, by describing how BLM and Topoisomerase IIIα coordinate their activity to suppress DNA crossovers.

Publications listed on page 147



CLARE HALL

99

DNA DAMAGE AND GENOMIC STABILITY www.london-research-institute.org.uk/research/vincenzo-costanzo

Group Leader

Vincenzo Costanzo Postdoctoral Scientists Antoine Aze Alessia Errico Simona Fiorani Fabio Puddu Tamon Tanaka Maria Vinciguerra Graduate Students Andrew Crockford Gabriele Piergiovanni Scientific Officers Nicola Brown Julian Gannon Sarah Smith

In the last few years my lab has explored mechanisms underlying the maintenance of genome stability in higher eukaryotes. Although many of the players involved in this process are known the biochemical mechanisms underlying their function are poorly understood. We took advantage of the vertebrate Xenopus laevis egg extract to study the complex biochemistry behind these processes. This approach was complemented by the use of genetic methods based on selective gene inactivation in mammalian cells, and more recently, in transgenic mice. Using these strategies we have studied the function of proteins such as ATM, ATR, the Mre11/ Rad50/Nbs1 (MRN) complex, the Tipin-Tim complex and Rad51 in cell cycle, DNA replication and DNA repair. These studies will expand our knowledge on the molecular aspects of DDR mediated control of DNA replication in unchallenged and perturbed conditions by highlighting the role of DDR proteins in DNA replication initiation, DNA replication fork structure maintenance and nucleotide biosynthetic pathways regulation. We expect that findings arising from this project will shed new light on those mechanisms, which if deregulated, contribute to occurrence of genome instability in human cancer. The role of DNA repair and DNA damage response proteins in unchallenged DNA replication Despite the high degree of conservation of the single proteins the requirement for cell survival is profoundly different for many of DNA repair proteins across different species. For example, inactivation of important Homologous Recombination (HR) genes such as Mre11, Rad50 or Nbs1 is lethal in mice, indicating that they complex fulfils essential tasks for cell survival in complex organisms. In contrast, mutations of these genes in S. cerevisiae cause have more limited effects. The reasons behind this discrepancy are unclear. Using the Xenopus egg extract we have analyzed the role in DNA replication of HR genes such as Rad51 and Mre11.

100

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Rad51 is the eukaryotic ortholog of RecA in E. coli. This protein plays a central role in HR during meiosis as well as during DSB repair and is regulated by the tumor suppressor gene BRCA2. We found that Rad51 is required to prevent the accumulation of ssDNA lesions during unchallenged DNA replication suggesting that Rad51 is directly required at DNA replication forks for uninterrupted and accurate replication of both damaged and undamaged templates. Rad51 could have a protective role towards nascent DNA chains and the observed extended ssDNA stretches at the fork could result from increased susceptibility to exonucleolytic degradation. We confirmed this hypothesis showing that nascent DNA strands are actually degraded by the Mre11 nuclease in the absence of Rad51. Indeed, suppression of Mre11 activity also suppresses ssDNA gap formation.

following replication fork stalling. Timeless (Tim1), Tipin and Claspin have been identified, both in yeast and higher eukaryotes, as members of a ‘Replication Pausing Complex’ that contributes to fork stabilization, fork recovery and checkpoint activation. We have previously investigated the role of Tipin and Tim1 during unchallenged DNA replication in Xenopus laevis egg extract. Tipin and Tim1 form a complex independent of Claspin in egg extract that promotes DNA synthesis when dormant origins are suppressed by stabilizing the binding of polymerase α to DNA.

Figure 1 Replication fork progression is discontinuous and might require polymerase α mediated re-priming of DNA synthesis downstream replication stalling sites. The Tipin/ Tim1 complex might promote polymerase α dependent replication re-priming in unchallenged and perturbed S-phase.

We will now deplete the Tipin/Tim1 complex using antibodies already available, monitor DNA replication and analyse replication intermediates by EM in the presence or absence of Rad51 and/or active Mre11. In case Tipin/Tim1 is required for efficient re-priming we expect a reduction of ssDNA gap accumulation. Our findings have uncovered an unanticipated role in DNA replication for proteins previously thought to be involved only in DNA repair. We believe that this approach will lead to important breakthroughs in our understanding of DNA replication processes involved in maintenance of genome stability. Therefore, Rad51 appears to limit the extent of the resection, which progresses to pathological levels in its absence. The accumulation of ssDNA gaps is likely due to the discontinuous progression of replication forks, which stall in the presence of DNA lesion or structures that block replicative polymerase progression, leaving gaps behind the moving replisome, which are then repaired by Mre11 and Rad51 dependent post-replicative DNA repair. The experimental approach we adopted has led to striking and unexpected results, demonstrating that Rad51 is a replication factor, assisting continuous DNA synthesis by preventing degradation of nascent strands at stalled forks. The discontinuous nature of replication forks might require continuous polymerase α mediated re-priming of DNA synthesis downstream replication stalling sites. How this re-priming occurs is unclear. We are now exploring the possibility that the Tipin/Tim1 complex promotes polymerase α dependent replication re-priming in unchallenged and perturbed S-phase (Figure 1). Replisome components are stably associated with replication forks to ensure replication resumption



To gain more insights on molecular mechanisms underlying Tipin/Tim1 function we have recently performed a mass spectrometry analysis of the Tipin/Tim1 complex immuno-purified from egg extract with specific antibodies. Using this approach we have isolated several potential partners of the Tipin/Tim1 complex. Among these our preliminary data indicated that Tipin/Tim1 interacts with proteins involved in chromatin metabolism. In this part of the project, we are exploring the implications of Tipin/Tim1 interaction network with the for replication fork metabolism. The use of the Xenopus system, in which transcription is inactive will facilitate the dissection of the replication-associated function of chromatin metabolism proteins from the more characterised transcriptional role of these proteins. We have already cloned Xenopus orthologs of many of these proteins. We are in the process of producing recombinant proteins that will be used for the production of specific antibodies to characterise these interactions and perform functional studies.

Publications listed on page 147

CLARE HALL

101

CHROMOSOME REPLICATION www.london-research-institute.org.uk/research/john-diffley

Group Leader

John FX Diffley Postdoctoral Scientists Dominik Boos Stephanie Carter Gideon Coster Max Douglas Jordi Frigola Belén Gómez-González Khalid Siddiqui Mona Yekezare Graduate Students Tom Deegan Agnieszka Janska Amina Mehanna Kenneth On Scientific Officers Lucy Drury Anne Early

102

We synthesize more than a light year of DNA during the course of our lives. As a consequence, DNA replication must be extremely accurate and efficient to ensure that all of our cells inherit a complete and intact complement of genetic material during cell proliferation. Understanding the mechanism and regulation of DNA replication is therefore crucial for our basic understanding of all cell proliferation. DNA replication and DNA damage checkpoints also have an important place in cancer biology. For example, it has become clear in recent years that replicative stress caused by oncogenes activates DNA damage checkpoints in precancerous cells, which induces senescence or apoptosis, and thus acts as a barrier for oncogenesis. Yet we know little about the nature of this replicative stress – is it caused by damage to the DNA template, defects in initiating replication or defects in the elongation machinery? Towards reconstitution of eukaryotic DNA replication DNA replication in eukaryotic cells initiates from many replication origins distributed on multiple chromosomes. Stable genome inheritance requires that these origins initiate efficiently during S phase, and that re-initiation from these origins is subsequently prevented. The logic of how the activation of multiple replication origins is limited to once per cell cycle is now well understood (Figure 1): origins are first ‘licensed’ during G1 phase with a pre-replicative complex (pre-RC) containing the Mcm2-7 helicase; the helicase is then activated during S phase, when pre-RCs can no longer be assembled. Mcm2-7 is loaded onto origin DNA by the combined action of three ‘licensing factors’: the six-subunit Origin Recognition Complex (ORC), Cdc6 and Cdt1. The activation of the helicase involves the recruitment of Cdc45 and the heterotetrameric GINS complex into a CMG (Cdc45, Mcm2-7, GINS) complex. This step requires a set of ‘firing factors’ including the two protein kinases, Cyclin Dependent Kinase (CDK) and Dbf4 Dependent Kinase (DDK), the leading strand DNA polymerase ε and a host of other essential proteins of unknown function including Sld2, 3, 7, Dpb11 and Mcm10. CDK plays

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

a second crucial role in regulating replication by inhibiting the licensing step. Thus, the oscillations of CDK activities in the cell cycle – low in G1 phase, and high in the remainder of the cell cycle – are critical to ensure origins are activated just once in each S phase. The DNA damage checkpoint kinase Rad53 also plays a key role in stabilising stalled replication forks and inhibiting late replication origin firing after DNA damage. We have previously developed a system for pre-RC assembly with purified proteins (Remus et al., 2009; Cell. 139: 719-30). We assemble pre-RCs on DNA that has been coupled to paramagnetic beads (‘B’ in Figure 2) via a streptavidin-biotin linkage. Figure 2 summarises some of the complexes that can be distinguished in this system. When ATP is present, only ORC and Mcm2-7 are stably bound to DNA. High salt wash (HSW; 0.5M NaCl) of the isolated beads removes ORC, but not the Mcm2-7 double hexamer, which remains bound topologically to duplex DNA. We refer to the retention of Mcm2-7 on DNA after high salt extraction as ‘loading’. ORC, Cdc6 and Cdt1•Mcm2-7 are retained on the DNA when the reaction is supplemented with the nucleotide analogue ATPγS but challenge with high salt under

Figure 1 The Chromosome Replication Cycle. A simplified version of the chromosome replication cycle. Details are discussed in the text.

these conditions removes all of the constituents. We refer to this salt-sensitive retention of Mcm2-7 subunits as ‘recruitment’. We have used this system to examine the mechanism by which the Mcm2-7 complex is loaded into a double hexamer bound around double stranded DNA. We have found that the initial recruitment of Mcm2-7 occurs via a conserved domain at the C-terminus of Mcm3. Binding of Mcm3 to ORC•Cdc6 activates the ATPase of ORC•Cdc6 and ATP hydrolysis can, if all pre-RC components are not present, lead to release of Mcm3 in a proofreading-like reaction. Alternatively if all of the pre-RC components are present, ATP hydrolysis promotes the loading reaction. We have found that the ATPase activity of the Mcm2-7 subunits is also crucial for licensing, and

we are currently trying to understand this requirement. In addition to using this system to examine the mechanism of helicase loading, we are also using it to examine the mechanism of helicase activation. We have recently shown that we can activate the pre-RCs we assemble with purified proteins in an S phase extract prepared from cells that overexpress a set of firing factors. We anticipate this system will be extremely useful to understand how the helicase is activated, how the replisome assembles and, in the longer term, how other activities such as inheritance of chromatin structures, sister chromatid cohesion and post-replication repair are coupled to DNA replication. Publications listed on page 148

Figure 2 Pre-RC Assembly in vitro. An outline of pre-RC assembly with purified proteins. Details are described in the text. HSW is high salt wash (0.5M NaCl).



CLARE HALL

103

MAMMALIAN DNA REPAIR www.london-research-institute.org.uk/research/peter-karran

Group Leader

Peter Karran Postdoctoral Scientists Quentin Gueranger Elizabeth McAdam Graduate Students Melisa Guven Matt Peacock Scientific Officers Reto Brem Peter MacPherson

The work of the Mammalian DNA Repair Group is currently concerned with the ultraviolet radiation (UVA)-mediated induction and repair of lesions caused by sulphur-containing DNA base analogs. The thiopurines, 6-thioguanine (6-TG), 6-mercaptopurine, and its prodrug azathioprine are used to treat leukemia and autoimmune disorders. Their use is associated with side effects, the most significant of which are sun sensitivity and a very high skin cancer incidence. Part of our work aims to understand the reasons for the increased cancer risk. Sulfur-containing base analogs are photosensitizers that combine with UVA to cause cell death. This interaction has clinical potential. Photochemotherapy, which combines synergistic ultraviolet radiation and photosensitizing drugs, is an established therapeutic strategy and UVA is relatively harmless. Existing photochemotherapies are associated with an increased skin cancer risk and there is a need to develop alternatives employing less radiation and with improved targeting. Our work has indicated that the thiopyrimidines 4-thiothymidine (4-thioTdR) and 4-thio-5-bromodeoxyuridine (4-thio-BrdU) might offer safe and effective alternatives to existing photosensitizers. Thiopurines, skin cancer and DNA repair In common with the other thiopurines, the immunosuppressant azathioprine is metabolised to nucleotides and incorporated into DNA as 6-TG. We previously described UVA photosensitivity in azathioprine patients that reflects the ability of DNA 6-TG to absorb energy from UVA. Since canonical DNA bases do not absorb UVA to a significant extent, photoactivation of DNA 6-TG generates novel forms of DNA damage that the cell may find difficult to repair. This is particularly true for the DNA interstrand crosslinks (ICLs) that are a significant DNA 6-TG photoproduct. Cells from patients with the hereditary disorder Fanconi anemia (FA) are extremely sensitive to agents the produce ICLs. They are very sensitive to the combination of DNA 6-TG and UVA and are also unexpectedly sensitive to 6-TG without radiation. Reto Brem has shown that this is because exposure to 6-TG induces a condition of

104

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

oxidative stress - an excess of intracellular reactive oxygen species (ROS). Excess ROS are generated because the thiopurine depletes cellular antioxidant defences and thereby increases steady state ROS levels. The excess ROS react with DNA 6-TG to generate ICLs and other lesions. FA cells provide a sensitive detector for this DNA damage. Their 6-TG sensitivity can be reversed by treatment with ROS scavengers. These findings have led to the suggestion that 6-TG combined with antioxidants might represent an alternative, and safer, thiopurine treatment option. UVA comprises > 90% of the ultraviolet in sunlight. It is normally not a serious threat to DNA and the minor UVB component is responsible for the direct introduction of potentially mutagenic and carcinogenic DNA lesions. Incomplete repair of this DNA damage results in the generation of characteristic signature mutations in skin tumors

Figure 1 DNA repair in extracts of 6-TG/UVA treated cells. A. Nucleotide excision repair (NER). Extracts prepared from HeLa cells grown for 24h in 6-TG and UVA as indicated were incubated with a covalently closed circular duplex DNA containing a single cisplatin intrastrand crosslink that is a recognized substrate for repair by NER. Repair-mediated incision on either side of the lesion generates short oligonucleotides (approx. 26nt). These are then selectively radiolabelled with Sequenase to generate 29-32nt products that are separated on denaturing gels. 6-TG treatment reduces NER efficiency and repair is undetectable following combined 6-TG/UVA. M = size markers. B. Non-homologous end joining. Extracts prepared from untreated cells or cells treated with 6-TG+UVA as indicated were incubated with radiolabelled linearized plasmid DNA and the products analysed by gel electrophoresis. End joining generates dimer, trimer and tetramer forms as indicated. 6-TG/UVA treatment abolishes this end joining.

A

in the general population. The same spectrum of mutations characterises the skin tumors of immunosuppressed individuals. This suggests that DNA repair may be compromised in sun exposed skin of azathioprine patients and our previous work suggested that nucleotide excision repair (NER) – the pathway responsible for removing potentially mutagenic UVB photoproducts – might be inhibited by 6-TG/UVA treatment. Consistent with this, Quentin Gueranger has shown that DNA 6-TG/UVA treatment increases the susceptibility of cultured human cells to mutation by low doses of UVB. At least in part, this reflects altered efficiency of the DNA repair proteins and NER by extracts of treated cells is significantly reduced. The effect on DNA repair appears to be a general one and Matt Peacock has shown that extracts prepared from treated cells are also defective in rejoining of DNA double strand breaks – an important protection against genomic instability (Figure 1). We believe that by causing oxidation of DNA repair proteins, photochemically generated singlet oxygen (1O2) is partly responsible for a general DNA repair inhibition. Thiopyrimidine photoactivation as a photochemotherapeutic option Thiopyrimidines are extensively incorporated into DNA of cultured cells via the deoxynucleoside salvage pathway. They are non-toxic but also cause extensive cell death in combination with low UVA doses. Lizzie McAdam and Reto Brem are examining the mechanism(s) underlying the synergistic cytotoxicity of the thymidine analogs, 4-thioTdR and 4-thioBrdU with UVA. They have shown that, unlike 6-TG, neither DNA thiopyrimidine generates a burst of ROS following UVA activation. This suggests that the photochemical reactions of the DNA thiopyrimidines predominantly involve nucleobase free radicals rather than 1O2. As with 6-TG, ICLs are a major detectable photoproduct of both analogs. In contrast to 6-TG/UVA, however,



B

FA cells are only mildly sensitive to the UVA activation of DNA thiopyrimidines. Working with synthetic oligonucleotides that contain a single 4-thioT base, Peter Macpherson has shown that crosslinking involves 4-thioT and its complementary adenine and that the crosslink can be unhooked by depurination of the affected adenine. Depurination may offer the cell an alternative to ICL processing by the canonical FA pathway in protection against the cytotoxicity of these ICLs. The base excision repair (BER) pathway of DNA damage has a significant impact on thiopyrimidine/UVA sensitivity. Cells defective in the OGG-1 DNA glycosylase, which protects against the effects of oxidized DNA guanine are sensitive to 4-thioTdR/UVA. A defect in the UNG DNA glycosylase that removes uracil from DNA confers sensitivity to 4-thioBrdU/UVA. Each of these BER enzymes has a limited substrate range and may offer clues to the identity of the lethal DNA photoproducts. There are potential clinical implications of our findings. Firstly, the demonstration that UVA activation of DNA 6-TG which underlies the skin photosensitivity of azathioprine patients reflects the generation of ROS and depletion of cellular antioxidant defences indicates that application of suitable antioxidants might provide protection against both photosensitivity and skin cancer development. Secondly, the observation that the thiopyrimidine nucleosides act by a different mechanism that does not involve the generation of a highly damaging burst of 1O2 indicates that they might provide an alternative approach to cancer treatment. Their extreme synergy with very low, harmless UVA doses suggests a possible application in treatments analogous to the currently used photodynamic therapy or extracorporeal photopheresis.

Publications listed on page 148

CLARE HALL

105

CELL DIVISION AND ANEUPLOIDY www.london-research-institute.org.uk/research/mark-petronczki

Group Leader

Mark Petronczki Postdoctoral Scientists Sergey Lekomtsev Laurent L’Epicier-Sansregret Murielle Serres Maria Dolores Vázquez Novelle Graduate Students Kristyna Kotynková Kuan-Chung Su Sriramkumar Sundaramoorthy Scientific Officer Tohru Takaki

Every second several hundred thousand cells in our body duplicate themselves through a process known as cell division. To generate healthy and viable cells the division process has to accurately partition all 46 chromosomes to daughter cells. Our group uses animal cell systems to investigate the molecular mechanisms underlying cell division and the consequences of genomic imbalances caused by cell division errors. We are particularly interested in the process of cytokinesis, the final step of cell division that partitions segregated sister genomes and leads to the birth of new daughter cells. Recently, we were able to define an essential molecular link between the cell envelope and the mitotic spindle, a key scaffold within dividing cells. Our results suggest that this link plays a key role in facilitating the final cut during cytokinesis in animal cells. Cytokinesis – a tale of microtubules, actin and the plasma membrane During cytokinesis the cytoplasm of a dividing mother cell is split by the ingression of a cleavage furrow. This process is powered by the contraction of a plasma membrane-associated actomyosin network, called the contractile ring. Subsequently, a final membrane scission event, called abscission, severs the intercellular bridge that connects nascent daughter cells (Figure 1C). A key requirement for successful cell division is that the splitting of daughter cells occurs between segregated sister genomes. To achieve this, animal cells ingeniously link both chromosome segregation and cytokinesis to the same cellular macrostructure, the microtubule-based mitotic spindle. First sister genomes are pulled to opposite sides by spindle microtubules. Then, motor proteins and microtubule-associated factors assemble an array of overlapping microtubule bundles between sister genomes. This structure is known as the spindle midzone and acts as a signalling hub that promotes the ingression of the cleavage furrow at the equatorial membrane. Furrow ingression compresses midzone microtubules and leads to the formation of an intercellular bridge with the midbody at its

106

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

centre (Figure 1B and 1C). The midbody is a dense proteinaceous structure that recruits abscission proteins for the final membrane severing reaction and has been proposed to act as an anchor for the cleavage furrow membrane after contractile ring disassembly. Thus, substructures of the late mitotic spindle, such as the midzone and midbody, are key regulators of cytokinetic events at the plasma membrane. They help define the centre of the cell for the cleavage apparatus and provide structural support for the final abscission event. Last year, we reported how the midzoneassociated Rho guanine exchange factor Ect2 helps to deliver the signal for cleavage furrow formation to the cell envelope in a spatially and temporally controlled manner (Su et al., 2011; Developmental Cell. 21: 1104-15). Holding on to the cell envelope for the final cut Despite the importance of the mitotic spindle apparatus for cytokinesis in animal cells, direct interactions between core components of the spindle midzone or midbody and the plasma membrane have not been identified. Recently, we were able to define a plasma membrane tethering activity in centralspindlin (Figure 1C), a conserved core component of the spindle midzone and

midbody (Lekomtsev et al.,2012; Nature. 492: 276-9). Centralspindlin is a heterotetramer that contains two molecules of the kinesin motor protein Mklp1 and two molecules of the Rho GTPase activating protein MgcRacGAP (Figure 1C). Centralspindlin can be considered a master regulator of cytokinesis in animal cells. It organises overlapping bundles of microtubules at the midzone and midbody, promotes cleavage furrow formation by controlling Rho GTPases, and recruits abscission factors. The centralspindlin subunit MgcRacGAP contains a highly conserved but uncharacterised C1 domain (Figure 1A and 1C). We discovered that MgcRacGAP’s C1 domain acts as a plasma membrane interaction domain by binding to the polyanionic phosphoinositide lipids PI4P and PI(4,5)P2 in the inner leaflet of the cell envelope. In collaboration with the Chromatin Structure and Mobile DNA group at Clare Hall, we determined the structure of MgcRacGAP’s C1 domain using X-ray crystallography. The domain adopts a characteristic C1 domain fold with two zinccoordinating centres as a structural base and a superimposed loop β-sheet arrangement (Figure 1A). Computational, cellular and biochemical analyses suggest that the C1 domain interacts with the plasma membrane through a hydrophobic cap (green) that penetrates into the hydrocarbon core of the lipid bilayer (dashed black line) and a basic jacket (purple) that interacts with the anionic

A

Figure 1 Centralspindlin links the mitotic spindle to the plasma membrane during cytokinesis. A. Model of MgcRacGAP C1 domain docking to a lipid bilayer. The structure of MgcRacGAP’s C1 domain was determined by X-ray crystallography and refined to a resolution of 2.2 Å. B. Confocal imaging of dividing human cells stably expressing a membrane marker protein (MyrPalm-mCherry, red) together with wild-type or K292L mutant MgcRacGAP (white). Both wild-type and mutant MgcRacGAP accumulate at the spindle midzone and midbody during cleavage furrow ingression. The K292L mutation in MgcRacGAP’s C1 domain causes the detachment of the plasma membrane from the midbody and subsequent cytokinesis failure (arrowheads). C. Model. (Reprinted from Lekomtsev et al., 2012).

heads of phosphoinositide lipids (Figure 1A). Importantly, mutations in the hydrophobic cap and in basic jacket residues of MgcRacGAP’s C1 domain (such as K292L) not only abrogate membrane interaction of the protein and but also abolish cytokinesis in human and chicken cells (Figure 1B). Using hybrid proteins and chemical genetic approaches we could demonstrate that artificial tethering of centralspindlin to the plasma membrane restores successful cell division in the absence of MgcRacGAP’s C1 domain. Thus, centralspindlin’s interaction with the plasma membrane is a conserved and essential requirement for cytokinesis in animal cells. While C1 domain function is dispensable for the formation of the midzone and midbody, centralspindlin’s membrane interaction promotes contractility and is essential for the attachment of the plasma membrane to the midbody (Figure 1B). Mutations in MgcRacGAP’s C1 domain cause the detachment of the plasma membrane from the midbody and subsequent cell division failure (Figure 1B). Thus, our analysis suggests that centralspindlin links the mitotic spindle to the plasma membrane to secure the final cut during cytokinesis in animal cells (Figure 1C) (Lekomtsev et al., 2012). Our work could also provide a mechanism and the structural basis for the anchoring of the cleavage furrow to the midbody, a long-postulated function of this organelle. Publications listed on page 148

C

B



CLARE HALL

107

MECHANISMS OF GENE TRANSCRIPTION www.london-research-institute.org.uk/research/jesper-svejstrup

Group Leader

Jesper Q Svejstrup Postdoctoral Scientists Stefan Boeing Andreas Ehrensberger Theo Kantidakis Hannah Mischo Marco Saponaro Yuming Wang Laura Williamson Graduate Students David Hobson Michael Lim Michael Ranes Kotryna Temcinaite Marcus Wilson Scientific Officers Barbara Dirac-Svejstrup Michelle Harreman Jane Walker

While cellular processes maintaining genome integrity and allowing faithful replication after DNA damage are of utmost importance for the long-term survival of cells and organisms, the key immediate response of cells suffering genotoxic insult is arguably to maintain gene expression. Indeed, without continued transcription, cells cannot proceed through the cell cycle, and even non-dividing cells will perish. The overall aim of our research is to understand the basic mechanisms underlying RNA polymerase II (RNAPII) transcription, but in particular how transcript elongation interfaces with other processes on DNA, such as mRNA splicing, chromatin remodeling, DNA repair, replication and recombination. We also investigate what happens during transcription stress, such as that caused by DNA damage. We believe that a detailed insight into the basic mechanisms of transcript elongation will make it possible for us to understand certain human diseases, and thereby in time hopefully how to treat them. The interface between transcription and other DNA-related processes We use biochemical, genetic and cell biological approaches in eukaryotes, primarily mammalian cells and yeast, to shed light on the process of transcript elongation and its interface with other DNA-related processes. Our published work in 2012 includes studies of transcriptional collision in yeast, the coupling between transcript elongation and alternative mRNA splicing in human cells, and the identification of new subunits of the human Elongator complex. Antisense non-coding transcripts, genes-withingenes, and convergent gene pairs are prevalent among eukaryotes. The existence of such transcription units raises the question of what happens when RNA polymerase II (RNAPII) molecules collide head-to-head. We previously studied collision between RNAPII elongation complexes transcribing the same DNA strand (head-to-tail collision). This study indicated that dynamic interactions between conformationally elastic elongation complexes make significant and fundamental contributions to transcript

108

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

elongation (Saeki and Svejstrup, 2009; Mol Cell. 35: 191-205). The situation is different when RNAPII molecules transcribe opposite DNA strands; here, the approaching transcription ‘bubbles’ should in theory be able to pass each other. On the other hand, the large size and extraordinary stability of the eukaryotic elongation complex might make bypass difficult or impossible (Figure 1). To study the fundamental consequences of transcriptional collision we used a combination of biochemical and genetic approaches in yeast to show that polymerases transcribing opposite DNA strands cannot bypass each other (Hobson et al., 2012; Mol Cell. 48: 365-74). RNAPII stops, but does not dissociate upon head-to-head collision in vitro, suggesting that opposing polymerases represent insurmountable obstacles for each other. Headto-head collision in vivo also results in RNAPII stopping, and removal of collided RNAPII from the DNA template can be achieved via ubiquitylationdirected proteolysis. Indeed, in cells lacking efficient RNAPII poly-ubiquitylation, the half-life of collided polymerases increases, so that they can be detected between convergent genes. These

Figure 1 RNAPII Collision is a block to transcript elongation in vivo. A. Nucleic acid-centric view of convergent transcription. B. Protein-centric view of elongation, showing a crystallographic model of convergently transcribing RNAPII elongation complexes (Kornberg, 2007). (DNA in green/blue; RNA in red). Our results show that colliding RNAPII molecules cannot bypass each other (Hobson et al., 2012).

A

B

results provide new insight into fundamental mechanisms of gene traffic control, and point to an unexplored effect of antisense transcription on gene regulation via polymerase collision (Hobson et al., 2012). Alternative mRNA splicing is the main reason that vast mammalian proteomic complexity can be achieved with a limited number of genes. Splicing is physically and functionally coupled to transcription, and is greatly affected by the rate of transcript elongation. As the nascent pre-mRNA emerges from transcribing RNA polymerase II (RNAPII), it is assembled into a messenger ribonucleoprotein (mRNP) particle, which is its functional form and determines the fate of the mature transcript. However, factors that connect the transcribing polymerase with the mRNP particle and help integrate transcript elongation with mRNA splicing remain obscure. We characterised the interactome of chromatinassociated mRNP particles (Close et al., 2012a; Nature. 484: 386-9). This led to the identification of Deleted in Breast Cancer 1 (DBC1) and a protein we named ZIRD as subunits of a novel protein complex, named DBIRD, which binds directly to RNAPII. DBIRD regulates alternative splicing of a large set of exons embedded in A/T-rich DNA, and is present at the affected exons. RNAi-mediated DBIRD depletion results in region-specific decreases in transcript elongation, particularly across areas encompassing affected exons. Together, these data indicate that DBIRD complex acts at the interface between mRNP particles and RNAPII, helping integrate transcript elongation with the regulation of alternative splicing (Close et al., 2012a). Ubiquitylation is a highly diverse and complex post-translational modification for the regulation of protein function and stability. Studies of



ubiquitylation have, however, been hampered by its rapid reversal in cell extracts, for example through the action of de-ubiquitylating enzymes (DUBs). We developed a novel ubiquitin-binding protein reagent, MultiDsk, composed of an array of five UBA domains from the yeast ubiquitinbinding protein Dsk2, fused to GST (Wilson et al., 2012; PLoS One 7: e46398). MultiDsk binds ubiquitylated substrates with unprecedented avidity, and can be used as both an affinity resin to study protein ubiquitylation, and to effectively protect ubiquitylated proteins from the action of DUBs and the proteasome in crude cell extracts. The Elongator complex is composed of 6 subunits (Elp1-Elp6) and promotes RNAPII transcript elongation through histone acetylation in the nucleus as well as tRNA modification in the cytoplasm. The identity of human ELP1 through ELP4 has previously been reported, but human ELP5 and ELP6 have remained uncharacterised. We found that DERP6 (ELP5) and C3ORF75 (ELP6) encode these subunits of human Elongator. The importance of these subunits was investigated by a combination of biochemical and cellular analysis. We found that DERP6/ELP5 is required for the integrity of Elongator and directly connects ELP3 to ELP4. The migration and tumorigenicity of melanoma-derived cells are significantly decreased upon Elongator depletion through the ELP1 or ELP3 subunits. Depletion of DERP6/ELP5 and C3ORF75/ELP6 has similar defects, indicating that DERP6/ELP5 and C3ORF75/ELP6 are essential for Elongator function. Together, our data identify DERP6/ELP5 and C3ORF75/ELP6 as key players for migration, invasion and tumorigenicity of melanoma cells, as integral subunits of Elongator (Close et al., 2012b; J Biol Chem. 287: 32535-45).

Publications listed on page 148

CLARE HALL

109

DNA DAMAGE TOLERANCE www.london-research-institute.org.uk/research/helle-ulrich

Group Leader

Helle Ulrich Postdoctoral Scientists Laure Gonzalez Magdalena Morawska-Onyszczuk Oliver Santt Tomio Takahashi Christopher Williamson Ronald Wong Graduate Students Jennifer Banerjee Elizabeth Colby Scientific Officers Adelina Davies Jonathan Lowther Jacqueline Marshall

DNA is susceptible to a variety of insults from exogenous and endogenous sources. In contrast to DNA repair systems, which usually rely on the excision and subsequent re-synthesis of the damaged region to restore the original sequence information, DNA damage tolerance mechanisms allow the bypass of lesions without their actual removal. They ensure the completion of DNA replication on damaged templates and are therefore essential for survival of a cell in the presence of genotoxic agents. As lesion bypass is often associated with damage-induced mutations, however, its activity needs to be tightly controlled. Our research aims at understanding the mechanisms and signals by which two small posttranslational protein modifiers, ubiquitin and SUMO, promote damage tolerance and limit the accumulation of unwanted mutations. Control of DNA damage bypass by ubiquitin and SUMO Posttranslational protein modifications generally modulate the functions of their targets by creating or modifying interaction surfaces. In order to understand the biological signals elicited by such events, it is essential to identify the downstream effector proteins that recognise the modifier in conjunction with the target. For several years our lab has been studying how posttranslational modifications of an essential replication factor, PCNA, govern the efficiency and accuracy of DNA lesion bypass. PCNA is a ring-shaped molecule that acts as a processivity factor for replicative DNA polymerases as well as an interaction platform for dozens of proteins involved in various DNA transactions. In response to DNA damage or replication fork problems, PCNA is ubiquitylated on a highly conserved lysine, K164. Whereas monoubiquitylation activates translesion synthesis by damage-tolerant DNA polymerases, polyubiquitylation is required for an error-free pathway of damage avoidance possibly involving template switching. During undisturbed replication, PCNA in budding yeast is modified by SUMO, which limits unscheduled recombination. We now understand the molecular consequences of PCNA monoubiquitylation and sumoylation in

110

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

fairly good detail. The monoubiquitin moiety on PCNA is directly recognised by a series of damagetolerant polymerases, which leads to their recruitment upon replication fork stalling. Sumoylated PCNA is bound by an antirecombinogenic helicase, Srs2, which inhibits the association of recombination factors at sites of DNA replication. In contrast, we still know relatively little about how polyubiquitylation of PCNA drives template switching. Our research over the past year has given new insight into the recognition of modified PCNA and has revealed an interesting cross-talk between ubiquitin and SUMO that contributes to the activation of DNA damage bypass. Identification of Mgs1 as an effector of ubiquitylated PCNA The AAA+ ATPase Mgs1, a PCNA interactor that contributes to genome maintenance in a poorly defined manner, harbours a ubiquitin-binding zinc finger (UBZ) motif. We were able to show that this domain facilitates the recruitment of Mgs1 to chromatin at sites where PCNA is ubiquitylated in response to DNA damage (Saugar et al., 2012; Nucleic Acids Res. 40: 245-57). By means of genetic analysis, we found that those activities of Mgs1 that become apparent in response to DNA damage indeed depend on the ubiquitylation of PCNA,

thus confirming that Mgs1 acts as a downstream effector of the modified clamp. Using linear fusions of ubiquitin and PCNA as mimics of ubiquitylated PCNA, we demonstrated that the interaction between Mgs1 and PCNA is enhanced by a monoubiquitin unit and even more so by polyubiquitin chains on PCNA (Figure 1). Thus, our findings indicate that Mgs1 might contribute to DNA damage bypass in a regulatory manner. However, our previous results (Zhao and Ulrich, 2010; Proc Natl Acad Sci. USA. 107: 7704-9) as well as unpublished data (Takahashi and Ulrich, 2012) indicate that linear polyubiquitin chains cannot substitute for the genuine, K63-linked chains that are attached to PCNA in vivo. Hence, the rather non-specific manner in which Mgs1 recognises ubiquitin makes it unlikely that the action of Mgs1 alone initiates the polyubiquitin-dependent template switching pathway. Cross-talk between ubiquitin and SUMO on PCNA The notion that PCNA sumoylation is targeted predominantly towards the same lysine as ubiquitylation raises the question of how the Figure 1 Mgs1 preferentially binds to polyubiquitylated PCNA. A. Domain structure of Mgs1. The asterisk indicates the conserved residue mutated to alanine in Mgs1*. B. Interactions of Mgs1 with ubiquitylated PCNA, analysed by pull-down assays in vitro. GST-tagged Mgs1 and Mgs1* were immobilized on glutathione Sepharose and assayed for binding to linear fusion constructs of ubiquitin and PCNA as indicated. Input and bound material were compared by anti-PCNA western blotting (Figure modified from Saugar et al., 2012).

A

Figure 2 Rad18 preferentially ubiquitylates sumoylated PCNA. A. Domain structure of Rad18. Asterisks indicate conserved residues mutated to alanine in the SIM* mutant. B. Activity of Rad18 towards PCNA versus a linear SUMO-PCNA fusion construct, assayed in vitro with purified proteins. Reactions included ubiquitin-E1, Rad6-Rad18, ubiquitin, ATP and PCNA or SUMO-PCNA, and products were identified by anti-PCNA western blotting (Figure modified from Parker and Ulrich 2012).

A

B

B



transition between the replication-associated sumoylated form of PCNA to the damage-induced ubiquitylated form is managed. We had previously reconstituted PCNA sumoylation and ubiquitylation with purified proteins in vitro and confirmed that the requirements for each of the modifications in vitro closely match the physiological situation in the yeast cell (Windecker and Ulrich, 2008; J Mol Biol. 376: 221-31; Parker et al., 2008; EMBO J. 27: 2422-31; Parker and Ulrich, 2009; EMBO J. 28:3657-66). To our surprise, we found that in this purified system ubiquitylation and sumoylation of PCNA did not compete with each other, but that the presence of the sumoylation enzymes instead enhanced the efficiency of PCNA monoubiquitylation. Experiments using a linear fusion of SUMO to the N-terminus of PCNA revealed that this stimulatory effect was attributable to sumoylation of PCNA itself (Figure 2). Consistent with this notion, we identified a SUMO-interacting motif (SIM) in the ubiquitin ligase (E3) Rad18, responsible for PCNA monoubiquitylation. We found that this motif indeed mediated the enhanced activity of Rad18 towards sumoylated PCNA (Parker and Ulrich, 2012; Nucleic Acids Res. 40: 11380-8). The positive effect of the SIM-SUMO interaction was reproduced in vivo, where mutation of the SIM or abolishment of PCNA sumoylation led to significant defects in PCNA ubiquitylation and damage bypass. Hence, our findings have identified Rad18 as an additional effector of sumoylated PCNA and at the same time a member of a recently identified class of enzymes, the so-called SUMOtargeted ubiquitin ligases. Our results indicate that the PCNA-SUMO conjugate is likely the physiological substrate of Rad18 and suggest a new mechanism by which damageinduced ubiquitylation is specifically targeted to those molecules of PCNA that are actively engaged in replication.

Publications listed on page 149

CLARE HALL

111

GENETIC RECOMBINATION www.london-research-institute.org.uk/research/stephen-west

Group Leader

Stephen C West Postdoctoral Scientists Gary Chan Miguel Gonzalez Blanco Laurent Malivert Maria Jose Martin Pereira Joao Matos Kanagaraj Radhakrishnan Shriparna Sarbajna Joanna Soroka Haley Wyatt Graduate Student Alessandra Pepe Scientific Officers Michael McIlwraith Rajvee Shah

Mammalian cells possess a large repertoire of DNA repair processes that maintain the integrity of our genetic material. But some individuals carry mutations in genes required for DNA repair, and this often leads to inheritable disease. An important repair process involves recombination, and defects in this process have been linked with cancer predisposition, in particular breast cancers caused by mutation of the BRCA2 gene, acute leukemias associated with Fanconi anemia, and a wide range of cancers found in individuals with the chromosome instability disorder known as Bloom’s Syndrome. The focus of our research is to determine the molecular mechanisms of recombinational repair, and to define why defects in these processes cause cancers. Our genetic material (DNA) is continually subjected to damage, either from endogenous sources such as reactive oxygen species produced as by-products of oxidative metabolism, from the breakdown of replication forks during cell growth, or by agents in the environment such as ionising radiation or carcinogenic chemicals. To cope with such damage, cells employ elaborate and effective repair processes that are each specialised to recognise different types of lesions in DNA. These repair systems are essential for the maintenance of genome integrity. Some individuals, however, are genetically predisposed to crippling diseases or cancers that are the direct result of mutations in genes involved in the DNA damage response. The BRCA2 tumour suppressor For several years we have been interested in the mechanisms of homologous recombination, how they contribute to the repair of DNA doublestrand breaks, and how they promote genome stability. The process of homologous recombination (HR) requires a number of proteins including RAD51, RAD52, RAD54, the RAD51 ‘paralogs’ (RAD51B, RAD51C, RAD51D, XRCC2, XRCC3), BRCA2, PALB2 and RP-A. Many of these proteins have been purified in this laboratory, and we use biochemical, cytological and molecular biological approaches to understand how they function within the cell to repair DNA breaks.

112

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Since women carrying BRCA2 mutations have a 70% chance of developing breast cancer, we are determined to understand the precise role of the BRCA2 tumour suppressor in DNA repair mediated by recombination. Towards this goal, we recently succeeded in purifying the BRCA2 protein from human cells. Using biochemical analyses and electron microscopic visualisation techniques, we found that the protein binds specifically to singlestranded DNA (ssDNA), or to regions of ssDNA present at replication-fork structures. Importantly, BRCA2 interacts directly with the RAD51 recombinase and directs RAD51 to bind to ssDNA at sites where DNA repair reactions are initiated. Without BRCA2, RAD51 fails to localise to sites of DNA damage, providing a molecular basis for the role of BRCA2 in the maintenance of genome stability and suggests why mutations in this protein lead to tumourigenesis. Current work is focussed on determining the three-dimensional structure of BRCA2 protein and will define how BRCA2 facilitates the formation of the RAD51 nucleoprotein filaments that initiate recombinational repair. Genome instability disorders linked to defects in the resolution of recombination intermediates Individuals with Bloom’s Syndrome (BS) suffer from a genetic disorder that leads to dwarfism,

Figure 1 Model for the timing and control of joint molecule resolution in mitotic and meiotic cells. In mitotic cells, the endonuclease activities of the two crossoverpromoting endonucleases MUS81-EME1 (Mus81-Mms4 in yeast) and Gen1 (Yen1 in yeast) are restrained until late in the cell cycle. This control mechanism ensures that most joint molecules are resolved into non-crossover products (thus avoiding loss of heterozygosity and sister chromatid exchanges). In contrast, in meiosis, Mus81 endonuclease is activated in a timely fashion in order to form the crossovers necessary for proper chromosome segregation, with Yen1 providing a ‘safeguard’ function.

immunodeficiency and reduced fertility. BS patients also develop various types of cancers, often at a young age. Cells derived from individuals with BS exhibit an extreme form of genome instability, the hallmark feature of which is an elevated frequency of sister chromatid exchanges. Bloom’s syndrome is caused by mutations in the BLM gene, which encodes BLM helicase, a protein that forms a complex with topoisomerase IIIα and the RMI1 and RMI2 proteins. We have purified this complex, known as the BTR complex, and are currently determining its structure and mechanism of action. The BTR complex plays an important role in the resolution of joint molecules that arise through recombination. However, in addition to the BTR complex, there are two other mechanisms for the processing of joint molecules that involve the MUS81-EME1 and GEN1 endonucleases. We found that inactivation of MUS81 and GEN1 from cells derived from Bloom’s syndrome patients led to an unusual aberrant chromosome morphology and cell death (Wechsler et al., 2011; Nature. 471: 642-6). Our analysis showed that the BTR complex normally resolves joint molecules in a manner that specifically avoids sister chromatid exchanges (and loss of heterozygosity when recombination occurs between homologous chromosomes rather than sister chromatids), and that, in the absence of BTR, joint molecule resolution is mediated by the two nucleolytic pathways for resolution requiring MUS81-EME1 or GEN1. Use of these alternatives allows the cell to separate recombining chromosomes, but also comes at a heavy price since BS cells exhibit genome instability and patients suffer a broad range of early onset cancers.



Regulation of nucleases that determine our genetic make-up Knowing that cells possess three distinct mechanisms for the resolution of joint molecules left us with a puzzle – how is it that mitotic cells use the BTR complex for joint molecule resolution rather than MUS81-EME1 or GEN1? Conversely, how is it that meiotic cells preferentially use the MUS81-EME1 and GEN1 pathways to promote chromosome segregation and form the crossovers necessary for the bipolar orientation and segregation of our maternally and paternally inherited homologous chromosomes? Mitotic and meiotic cells appear to possess similar pathways of resolution, but the way that they are used or regulated is clearly different. Our work led us to show that the specialized chromosome segregation patterns of meiosis and mitosis, which require the coordination of recombination with cell cycle progression, are achieved by regulating the timing of activation of the two crossover-promoting endonucleases (Matos, Blanco et al., 2011; Cell. 147: 158-72). In yeast meiosis, we discovered that Mus81-Mms4 (the ortholog of MUS81-EME1) and Yen1 (the equivalent of GEN1) are controlled by phosphorylation events that modulate their activities throughout the cell cycle. Mus81-Mms4 was hyper-activated by Cdc5-mediated phosphorylation in meiosis I, in order to generate the crossovers necessary for chromosome segregation. In contrast, Yen1 was activated in meiosis II, where it catalyses the resolution of persistent Holliday junctions that would otherwise block chromosome segregation. In both yeast and human mitotic cells, similar regulatory networks are thought to restrain both nuclease activities until mitosis, biasing the outcome of recombination towards non-crossover products while also ensuring the elimination of any persistent joint molecules. Mitotic regulation of these nucleases thereby facilitates chromosome segregation while limiting the potential for loss of heterozygosity and sister-chromatid exchanges.

Publications listed on page 149

CLARE HALL

113

False-coloured scanning electron micrograph of connective tissue. Image: Anne Weston Electron Microscopy Facility

114

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

TECHNOLOGY CORE FACILITIES

The London Research Institute benefits from access to a wide range of high quality research services. Scientific support for researchers at the LRI is provided by some nineteen core facility groups of various sizes run by the Institute. These aim to provide state of the art facilities for LRI researchers that are proactive in enabling the research groups to carry out world leading science.



TECHNOLOGY CORE FACILITIES

115

Technology core facilities provided centrally at the LRI allow access to cutting edge equipment and instruction in its correct and effective usage. Within these facilities, services may either be run by dedicated staff or by researchers themselves with appropriate service staff input.

Julian Downward Associate Director, LRI

116

LRI Core Facilities include the following: Light Microscopy provides conventional, confocal, multiphoton and automated microscopy, time-lapse video and microinjection services. Electron Microscopy has a field emission scanning electron microscope and transmission electron microscopes. The FACS Facility provides a comprehensive and continuously evolving flow cytometry service. The Experimental Histopathology Facility analyses the phenotypes of transgenic mice as well as providing expertise in human histopathology, in situ hybridisation techniques, laser capture microdissection and automated image acquisition using the Ariol system. The Equipment Park provides DNA Sanger sequencing, robotic nucleic acid preparation, quantitative PCR, gel imaging systems and HPLC micro-purification. The Advanced Sequencing Facility provides next generation DNA sequencing with an Illumina HiSEQ 2000, Illumina Genome Analyzer IIx and an Ion Torrent PGM. The High Throughput Screening Facility brings together the equipment, personnel and expertise needed to carry out and interpret large scale screening assays, such as automated high content cell-based screening of genome-wide RNA interference libraries. The Protein Purification Facility specialises in the production of pure recombinant proteins for structural studies, using baculoviral, bacterial and mammalian tissue culture systems. The Bioinformatics and Biostatistics Facility provides support for all the Institute’s bioinformatics needs, ranging from high throughput sequencing data analysis and high throughput screen interpretation to global gene expression analysis. Cell Services provides a wide range of quality controlled cells and media, as well as the production of monoclonal antibodies from

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

hybridoma lines. The Peptide Synthesis Facility provides custom made peptides, peptide arrays and cross-linking reagents. The Transgenics Facility provides the latest methodology for the generation of genetically modified mice. Provision of proteomics technology within the Institute has recently seen the concentration of all mass spectrometry equipment and expertise into a single facility at Clare Hall, which has been led by Bram Snijders since May 2012. The facility has six MS systems, including two Orbitraps. As part of the move towards the Francis Crick Institute, we have begun to coordinate the provision of mass spectrometry at the LRI with the National Institute of Medical Research at Mill Hill. Bram Snijders now oversees proteomics at both Institutes, with further systems located at Mill Hill, along with a newly formed Metabolomics Facility. It is envisaged that in the coming year further technology core facilities will be integrated in this way in the run up to the move to the Francis Crick Institute in late 2015. Other recent and ongoing developments have seen the appointment Stephane Mouilleron as an Associate Scientist to head our Protein Structure unit, which provides X-ray crystallography support for the structural biology groups working within the institute. The quality and development of LRI core facilities is driven by user committees, made up of representatives of the service provider, users and management which provide advice on technical advances, prioritise projects when facilities are limited, and act as a focal point for interactions with researchers. In addition, a programme of review by external experts every three years ensures that LRI services remain cutting edge.

ADVANCED SEQUENCING www.london-research-institute.org.uk/technologies/advanced-sequencing

The Advanced Sequencing Facility offers four main methods: ChIP-seq, Paired End sequencing (e.g. whole genome), RNA-seq and targeted enrichment sequencing (e.g. whole exome). The ASF is at the forefront in adopting, developing, and adapting these technologies and protocols. Head

Nik Matthews Staff Sharmin Begum Ben Phillimore Adam Rabinowitz

The interest in, and huge uptake of sequencing, since the advent of second generation technologies has been quite astonishing. Radical changes in sequencing technology have evolved from initial chain termination methodology to sequencing by synthesis or ligation. These two different sequencing technologies have enabled unheralded quantities of sequence data to be produced. Technology At the ASF, we have three second generation sequencing machines: the Illumina GAIIx, the Illumina HiSeq 2500 and the Ion Torrent PGM. All these rely on sequencing by synthesis but in two very different ways. The illumina machines detects fluorophores bound to the bases, whilst the Ion Torrent exploits the release of hydrogen ions when the nucleotides are incorporated, measuring this change in pH. (Figure 1) Expertise ChIP-seq is the evolution of ChIP-ChIP to study DNA-associated proteins. This technique is used to study heritable information and epigenetic patterns in cancer cells and much more. Paired End sequencing, (whole genome sequencing), makes it possible to sequence the human genome to between 15-20X (mean) coverage on the GA and much more (80X +coverage can be gained from the high throughput HiSeq2500). This enables us to sequence all the model organisms,

Figure 1 Each nucleotide is separately passed over approximately ten million micro-wells. In each well there is one bead that has around one million bound PCR fragments of the same template. If the base is incorporated there is a release of Hydrogen ions and a pH change. This gives a signal and knowing what base being used you can build up digital graphs of the bases in the fragments within the wells. Picture courtesy of Life Technologies.

including human, to a level of accuracy that is very important in examining rare variants. Further to this, target enrichment is used to sequence specific contiguous or non-contiguous areas within any genome. Custom designed regions of interest, or whole exome human and mouse, can be selected for sequencing. Sequencing the transcriptome (RNA-seq) has recently gained increasing prominence. Unlike the genome, which apart from mutations is relatively static, the transcriptome can change due to internal and external conditions. With RNA–seq, it is not only possible to study the actively expressed genes (mRNA) but also the non-coding genes (ncRNA) in the sample. Projects A highlight for ASF in 2012 was a successful collaboration with the Translational Cancer Therapeutics group (Gerlinger et al., 2012; N Engl J Med. 366(10):883-92). Multiregional sampling was undertaken from two patients with metastatic renal clear cell carcinoma. We used targeted enrichment, in this case whole exome, from fresh frozen samples from patients 001 and 002. These whole exome libraries were sequenced with 72bp/75bp paired-end reads on Illumina GAIIx/ HiSeq platforms. Phylogenetic analyses of primary tumour for both patients show branching phylogenetic trees rather than linear evolution. Furthermore, in different regions good and poor prognosis markers were detected from geneexpression signatures. Underestimation of the mutational burden of heterogeneous tumours when analysis is performed from a single tumour biopsy can lead to difficulties in validation of oncology biomarkers. It may also contribute to therapeutic failures by Darwinian selection of pre-existing drug resistant clones. Publications listed on page 149



TECHNOLOGY CORE FACILITIES

117

BIOINFOMATICS AND BIOSTATISTICS www.london-research-institute.org.uk/technologies/bioinfomatics-and-biostatistics

Datasets in biology are still increasing in size and complexity. To illustrate this two projects we are collaborating on have surpassed 10Tb of analytical data. Most of this data originates from experiments involving multiple comparisons between ChIP, RNA and DNA derived sequence frequently with public domain data from international efforts such as ENCODE and the Cancer Genome Atlas. Head

Aengus Stewart Staff Probir Chakravarty Philip East Mickael Escudero Stuart Horswell Gavin Kelly Anna Lobley Richard Mitter Harshil Patel Max Salm

Figure 1 Multi-region sequencing of a Glioblastoma reveals a complex chromosomal re-arrangement. Data courtesy of Translational Cancer Therapeutics Group.

It often requires the specialist skills of more than one bioinformatician, bringing mathematical and statistical methods, in addition to genomic and specialist domain knowledge to bear on the biological question. They are computationally challenging as well and routinely exceed the capability of server computing as a result the Bioinformatics and Biostatistics Facility rely heavily on the High Performance Computing Facility at LRI. Detecting clustering within genomic or proteomic data Frequently we are asked seemingly straightforward biological questions, but the mathematical machinery required to answer these questions can be deceptively complex. One such case is detecting whether a type of genomic feature, such as a set of binding sites, are randomly distributed or are alternatively clustered in some way. This question can be interpreted in many ways, and only recently have we found a sound and meaningful formulation. Previously we have set biologically motivated, but statistically arbitrary, thresholds – e.g. what proportion of observed binding sites have another binding site within 2kb? We can evaluate if the data are significantly clustered, but what if

118

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

we have chosen the wrong distance threshold? We have now adapted a technique from spatial statistics, called Ripley’s K, to give us an overview of clustering at all possible scales and we have now used this successfully in a number of projects. For example looking at CDK motif distribution within a set of proteins. Intra-tumour heterogeneity Working with the Translational Cancer Therapeutics Group, we have been investigating Intra-Tumour Heterogeneity using Next Generation Sequencing. Since each individual sample was, by definition, potentially heterogeneous, we performed power calculations to determine the appropriate minimal coverage required to detect rare variants. The detection of somatic single nucleotide variants required the construction of a bespoke pipeline to perform both absolute count based calling and statistically based analyses, taking into consideration variables such as; stromal contamination; differential coverage in germline versus tumour samples; ploidy differences; localized copy number variability; structural variants; the presence of sub-clonal populations in the sample; read directionality biases; and other ‘systematic’ artefacts relating to the sequencing platforms. The complexity and sheer volume of the data in this project is extremely challenging. The analyses showed that of the somatic, non-synonymous variants detected, only around a third were present in all nine regions from the primary tumour and the metastases, with a further third found to be in most primary regions but not the metastases, and the remaining third were largely unique to individual regions. Publications listed on page 149

BIOMOLECULAR MODELLING RESEARCH www.london-research-institute.org.uk/research/paul-bates

Head

Paul A Bates Staff Raphael Chaleil Postdoctoral Scientists Tammy Cheng Mieczyslaw Torchala Graduate Students Rudi Agius Sakshi Gulati Melda Tozluoglu

In this group we study fundamental and challenging problems in both structural and systems biology; in particular, how macromolecules interact at the atomic level to facilitate cellular events. Much of the work involves the design of novel computer algorithms that are based upon the principles of physics and evolutionary biology. Output from these algorithms are often displayed along a time course, indicating, for example, how macromolecules diffuse and interact in the cytosol and how cancer cells migrate through the extracellular matrix. These simulations are proving to be important in helping to interpret experimental data and suggest further experiments to probe complex molecular systems. Outlined below is one of the algorithms we have developed to study the consequences of missense protein mutations in complex biological systems. A structural systems biology approach for quantifying the systemic consequences of missense mutations in proteins Gauging the systemic effects of non-synonymous single nucleotide polymorphisms (nsSNPs) is an important topic in the pursuit of personalized medicine. However, it is a non-trivial task to understand how a change at the protein structure level eventually affects a cell’s behaviour. This is

Figure 1 Schematic diagram of the approach taken to model the systemic impact of point mutations within proteins such as those obtained from the genome wide association studies reporting single nucleotide polymorphisms, see (Cheng et al., 2012; PLoS Comput Biol. 8(10):e1002738).

because complex information, at both the protein and pathway level, has to be integrated. With respect to the fact that the idea of integrating both protein and pathway dynamics to estimate the systemic impact of missense mutations in proteins remains predominantly unexplored, we investigate the practicality of this approach by formulating mathematical models, combined with experimental data, to study missense mutations. We explore the practicality of this structural systems approach by formulating ordinary differential equations (ODEs) to describe two biological systems: (1) the G2 to M transition mechanism that controls the cell length in fission yeasts; (2) the human Erk signalling pathway. The study shows that the systemic impact of point mutations can be reasonably gauged through a systemic impact factor (SIF) that is a function of free energy change of host proteins (proteins that contain point mutations) and the sensitivity of host proteins in terms of regulating the expression profile of a downstream reporter protein (see Figure for an overview of the methodology).

Publications listed on page 150



TECHNOLOGY CORE FACILITIES

119

CELL SERVICES www.london-research-institute.org.uk/technologies/cell-services

Our dedicated team provides LRI research groups with all their cell culture needs. We work together to fulfill cell culture requirements, purified antibody, cell authentication, customised media/plates and mycoplasma screening requests as well as providing washroom services at Clare Hall. There have been several achievements this year, including:

Staff Amy Bailey Arron Charman Trevor Cooper Warren Cooper Darren Haines Darren Harvey Marley Holding-Pillai Rachel Horton-Harpin Spencer Horton-Harpin Samantha Kenton Julie Morrin Christine Saunders Martin Saunders Debbie Schofield Sonal Sheth Karen Stoughton Scott R. Taylor Mark Thorlby Paul Willis

CHD1 collaboration using murine embryo stem cells Cell Services has been involved in a long-term project requiring large numbers of mouse embryonic stem (ES) cells to be cultured regularly for protein purification. Despite the challenges associated with culturing ES cells, more than 1010 cells are consistently cultured successfully. These large quantities of cells are required to purify proteins for biochemical characterization, and thereby offer a unique approach to studying stem cell biology. The cells from these preparations have been used to isolate a key enzyme for the maintenance of pluripotency, CHD1. The Mechanisms for Gene Transcription group are currently attempting to purify and characterize a

M

oc

k

Figure 1 CHD1 purified from ES cells.

D1 -F LA G

Ruth Peat

CH

Head

CHD1 Cell_Services_Figure_1.tif

120

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

number of other enzymes involved in gene expression and pluripotency (Figure 1). Preliminary experiments indicate that such purifications will indeed be possible with the continued support of Cell Services. Development of a capillary electrophoresis method for non-human profiling Our quality control and development team have been working in collaboration with specialists at Sigma Aldrich to try and develop a capillary electrophoresis method for validation of non-human cell lines. It is hoped that this will complement our existing STR profiling technique, replacing our DNA Fingerprinting method and offer us a faster, more efficient process. In addition the team have been investigating an alternative method for confirming the species of our cell lines. Currently we use Isoenzyme Analysis, however a new PCR-based method is being developed which involves amplification of a specific ‘barcode’ region of DNA that can be used to identify the species. This should also be a more cost effective approach. Cell culture training This year we were pleased to be included in the Core Facilities training programme, which offers new PhD students the opportunity to attend various training courses related to the field of research they will be involved in. The training was well attended and we received very positive feedback. In addition our department continues to provide student tours and school placements as part of the LRI work experience initiative.

ELECTRON MICROSCOPY www.london-research-institute.org.uk/technologies/electron-microscopy

Head

Lucy Collinson Staff Ken Blight Raffaella Carzaniga Christopher Peddie Anne Weston Masters Student Erica Lam Chao Liu Charlotte Melia Stephen Woods

Figure 1 3D rendering of a B cell from 3View data (nucleus, green; plasma membrane, blue; bHEL, red). With Lymphocyte Interaction Group (Thaunat et al, 2012). Figure 2 Z-slice from a 3D reconstruction of a whole G361 cell, imaged by soft X-ray tomography (intracellular bridge highlighted). With L.Duke and M.Razi. Figure 3 Fluorescence and electron contrast in a 70nm resin section (GFP-C1, green; H2B-mCherry, red). With Cell Biophysics Group. Figure 4 Comparison of manual (2-3 weeks) and semi-automated (1 day) segmentation of a HeLa cell nucleus. With L.Pizarro and Cell Biophysics Group.

Electron microscopy is used to image the structure of molecules, cells and tissues at sub-nanometer resolution. In transmission electron microscopy (TEM), samples are cut into ultrathin sections of approximately 80-100nm so that the electron beam can pass through the sample and form an image on the detector below. In scanning electron microscopy (SEM), the electron beam is scanned over the surface of the sample to produce topographical or compositional information from the surface layer only. The Electron Microscopy Unit (EMU) The EMU is a Technology Core Facility providing the equipment and expertise necessary to image the structure of molecules, cells and tissues at high resolution. The EMU team consists of five experienced post-doctoral electron microscopists working closely with research groups at the LRI to plan, optimise and implement high-resolution imaging experiments. We are experts in preparing, imaging and interpreting a wide range of samples including DNA, proteins, viruses, yeast, cells, tissues, fruit flies, worms and zebrafish. We use a range of EM techniques including negative staining, resin embedding, ultrathin sectioning, serial sectioning, electron tomography, cryosectioning and immunolabelling, plunge freezing, high pressure freezing and correlative light electron microscopy (CLEM). The facility has two FEI Tecnai 120kV TEMs and a JEOL 6700F field emission SEM. The EMU is well connected to national and international EM facilities and imaging groups, with a strong track record of publication and invited conference presentations on new tools and techniques. New in 2012 We are very excited to announce that we have installed a Zeiss Sigma 3View SEM for automated



serial imaging of cells and tissues (Armer et al., 2009; PLoS One. 4: e7716; Stenzel et al., 2011; EMBO Rep. 12: 1135-43; Thaunat et al., 2012; Science. 335: 475-9; Bushby et al., 2012; Methods Cell Biol. 111: 357-82). This will allow us to automatically collect 3DEM images over hundreds of microns at nanometer resolution (Figure 1). This year, we have focused on three main areas of development: 1. Correlative cryofluorescence and soft X-ray tomography for imaging whole unstained mammalian cells in near-native state (collaboration with Liz Duke at Diamond Light Source, Secretory Pathways and Cell Biophysics groups, Figure 2). 2. Maintenance of GFP fluorescence through sample processing for integrated light and electron microscopy (Figure 3). 3. Development of algorithms for automated segmentation of biological structures in electron micrographs (collaboration with Stephen Woods, Chao Liu and Luis Pizarro at Imperial College, Figure 4).

Publications listed on page 150

TECHNOLOGY CORE FACILITIES

121

EQUIPMENT PARK www.london-research-institute.org.uk/technologies/equipment-park

The Equipment Park provides access to state of the art molecular biology instrumentation and offers instruction in the correct and efficient use of the technologies involved. The range of equipment is constantly reviewed and specific requests from research laboratory heads are encouraged. Head

Graham MG Clark Staff Vicky Dearing Muthulakshmi Muthusamy Olga O’Neil David Philips Ramin Sadri

Figure 1 Demonstration of Western blot detection using the LI-COR and Image Quant systems. The LI-COR system allows dual detection of distinct antigens simultaneously by using different infrared secondary antibodies. A. Here the antibody against a specific phosphorylation site in PKC is detected through one channel (680) while the total protein is detected through the other (800). The large dynamic range of the LI-COR system makes it ideal for accurate quantisation of Western blots. Alternatively the more classic HRP conjugated secondary antibodies can be visualised using ECL 9enhanced cheluminescence) and the Image Quant LAS400 system. B. This system employs a high resolution LCD camera to detect chemiluminescent signal.

122

The technologies include: The Beckman FX liquid handling robot is a dual ’pod’ automation system for parallel pipetting, and other liquid transfer techniques. Its low volume capability helps miniaturize reaction setups and save on costly reagents. The system has a stacker carousel, which feeds micro-titre plates and tip-boxes into the system, allowing multi-plate transfers. A Span-8 configuration featuring independent well access is also available for tube-based operations and where flexible access to wells in a plate is needed. Applications include: • High-density replication • Assay plate set-up (PCR, quantitative PCR (384 well), ELISA, kinase etc. • PCR and sequencing reaction clean-up using paramagnetic technology. The QIAgility is a bench-top instrument for automated setup of PCR reactions that is able to handle a wide variety of tube and plate formats. The system performs the preparation of master mix from individual reaction components and dilutions of standard series. Optionally available UV light and HEPA filter help to reduce the risk of sample carry-over. A

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

B

In addition to PCR setup, the flexibility of the system allows high-precision pipetting applications including: • Normalization of DNA and RNA concentration • Transfer of liquid samples from one tube format to another • Serial dilutions with variable dilution ratios • Sample pooling PherastarPlus micro-plate reader – premier multi-detection HTS micro-plate reader with Simultaneous Dual Emission in all modes. The reader is able to perform all leading non-isotopic detection technologies including: • Fluorescence Intensity • Fluorescence Polarization • Time-Resolved Fluorescence • TR-FRET • Luminescence • Absorbance UV/Vis Li-Cor Odyssey Infrared Imager – the imaging system offers a different way to analyse blots and gels. Odyssey is uniquely equipped with two infrared channels for direct fluorescence detection enabling simultaneous probing of two separate targets on the same gel, e.g. Western blots. ImageQuant LAS 4000 – a digital (CCD) imaging system for sensitive, quantitative imaging of gels and blots without film, by chemiluminescence and fluorescence. The high sensitivity and wide dynamic range is designed to capture the signals from ECL western blotting reagents. A wide range of visible fluorescent dyes can be imaged via Red, Green and Blue epi-illumination.

EXPERIMENTAL HISTOPATHOLOGY www.london-research-institute.org.uk/technologies/experimental-histopathology

The EHP Facility provides advice, training and expertise in a range of techniques to analyse cells and tissues from experimental models and human tissue banks. The lab will undertake the following procedures: • Rodent histopathology and comparative human Head

Gordon Stamp Operational Head Emma Nye Staff Tamara Bunting Kornelia Fritsch Bradley Spencer-Dene Richard Stone

cancer pathology reporting • Mouse dissection service and dissecting microscopy • Optimal handling/fixation options of fresh tissue • Mouse developmental analysis including embryonic lethal phenotypes • Histological sectioning/staining of frozen/fixed tissue • High resolution photomicroscopy • Immunohistochemistry (IHC) - >250 mouse orientated antibodies • Novel antibody optimisation for IHC • DNA/RNA extraction from FFPE • In Situ Hybridisation (ISH) using non-isotopic non-fluorescent methods on slides and whole mount embryos and organs using the InsituPro robot • 3D volume rendered reconstruction • Morphometric analysis - NIS Elements software platform for object classification/measurement • Laser Capture Microdissection (Leica LMD 7000) • Laser Scanning Microscopy using the Ariol SL50 system with Genetix software as follows; – IHC module for quantitating membrane, nuclear and cytoplasmic expression – Fluorescent module for FISH and immunofluorescent capture – Tissue Microarray module – General Morphometric Image Analysis module (ploidy, angiogenesis, area/volume) EHP encourages participation by graduate students and postgraduate scientists, and we

Figure 1 A. Brain section 12μm, before dissection, objective 63x, stained with Toluidine blue. Detection of cells by AVC+software. The selected shapes are within the chosen detection criteria. B. After dissection.

A

B

provide individual training and expert advice in photomicroscopy, tissue handling, IHC, Morphometry, LMD and other laboratory techniques. Technology highlight – Laser Capture Microdissection (LCM) Successful molecular biological analysis can depend on isolating specific cell populations from surrounding tissue components. This may be important in defining gene expression profiles of tumour cells versus desmoplastic stroma, or cellular subpopulations in a complex tissue, which are impossible to analyse with tissue homogenates. The Leica LMD 7000, a modified upright microscope that is user-friendly, is a highly effective method of collecting directly visualised cells for: • Real -time PCR (quantitative PCR) • Genomic DNA analysis and NGS • Reverse transcription PCR • RNA analysis (total and specific mRNA isolation) • Proteomics (2-D SDS PAGE, LC MS, Western blotting) Using specially coated slides, upon which a section of tissue 5-50 microns thick is mounted with a light counterstain, a fine UV laser cuts the slide coating film around the region or cells of interest, which then drops under gravity into a collection vial, ensuring no contamination. The image is retained to ensure QC of the sample. For advanced proteomic work frozen tissue may be used. Living cells from culture for subcloning or DNA/RNA analysis.

Publications listed on page 151



TECHNOLOGY CORE FACILITIES

123

FLUORESCENCE ACTIVATED CELL SORTING www.london-research-institute.org.uk/technologies/facs

Head

Derek Davies Staff Laura Bazley Julfa Begum Joana Cerveira Andy Filby Carl Henderson Sukhveer Purewal Kirsty Sharrock Masters Student Saleha Patel

The FACS Facility at the LRI is a dedicated scientific service offering an extensive flow cytometry analysis and sorting facility. Flow cytometry is a sophisticated form of fluorescence microscopy where cells in suspension pass one by one through a laser beam and emitted fluorescence can be captured and measured. Any part of a cell or any function of a cell that can be tagged by a fluorochrome may be measured by flow cytometry which makes it an essential technique in many biological applications.

Equipment Analytical cytometers. The FACS Facility has 6 analytical cytometers including one platereading cytometer. These may all be useroperated and we offer a one to one training for all new users of the Facility. FACS Calibur – 4 fluorescence detectors, 2 lasers (488nm and 633nm) LSRII: 13 fluorescence detectors, 4 lasers (355nm, 405nm, 488nm, 633nm) LSRII-SORP: 16 fluorescence detectors, 4 lasers (405nm, 488nm, 561nm, 648nm) LSR Fortessa: 15 fluorescence detectors, 5 lasers (355nm, 405nm, 488nm, 561nm, 638nm) LSR Fortessa: 18 fluorescence detectors, 5 lasers (355nm, 405nm, 488nm, 561nm, 638nm) MACSQuant VYB: 8 fluorescence detectors, 3 lasers (405nm, 488nm, 561nm)

Figure 1 MACSQuant VYB flow cytometer (Miltenyi Biotec) capable of absolute cell counting in a 96-well plate format.

Cell sorters. These are able to retrieve up to six specifically defined populations so that cells may be recovered for re-culture, functional assays or RNA or DNA recovery. Only members of the FACS Facility operate the sorters but experiments are scheduled and planned in close collaboration with our users. All sorters are housed in Class 2 Microbiological Safety Cabinets. MoFlo: 9 fluorescence detectors, 3 lasers (355nm, 488nm, 648nm) FACS Aria III: 13 fluorescence detectors, 4 lasers (405nm, 488nm, 561nm, 640nm) Influx: 14 fluorescence detectors, 4 lasers (405nm, 488nm, 561nm, 640nm) Other services The members of the FACS Facility are available to provide advice on the design of experiments, sourcing and supply of reagents, data analysis, presentation and interpretation as well as troubleshooting machines and experiments. We also develop and introduce new techniques and technologies that would be useful to our users. We collaborate closely with the group’s users and this has led to several recent publications particularly with groups involved in stem cell investigation, imaging flow cytometry or where DNA analysis and cell kinetic information is required. A major recent achievement was the award of ‘Paper of the year’ in the journal Cytometry for the Facility’s work in imaging and cell division.

Publications listed on page 151

124

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

HIGH THROUGHPUT SCREENING www.london-research-institute.org.uk/technologies/high-throughput-screening

The High Throughput Screening Facility enables research groups to access large-scale screening technologies. Primarily this takes the form of genome-wide siRNA screens although other types of screening are also popular.

Head

Michael Howell Staff Rachael Instrell Ming Jiang Rossella Rispoli Becky Saunders

As in previous years our screening work with LRI researchers this year has covered a broad range of biology from Vaccinia virus replication, to identifying small molecule inhibitors of cancer relevant proteins to characterising the cellular response to UV damage (in collaboration with the Cell Motility, DNA Damage and Genome Stability and Mechanisms of Gene Transcription Groups). After five years of operation, the facility has amassed a huge quantity of data that we maintain in a bespoke database. This database is made up of approximately 100 intricately linked tables with more than 25 million data entries in total. We have

created an interface to enable LRI researchers to explore and query this data resource (Figure 1) where they can browse through the details of individual screens right down to the raw data if necessary or can view what role favourite genes might play in a range of alternative biological settings. Such a resource can, for example, help provide clues as to the functions of genes identified as part of mutational analyses of cancers.

Publications listed on page 152

Figure 1 The HTS database is accessible to all LRI researchers. They can browse through all of the projects conducted in the facility (1), drilling down through the details of the methodology (2) to the final datasets (3) in an interactive manner (4). This central data resource also allows users to see if their favourite gene has biological functions in other contexts or if a biological function can be ascribed to an uncharacterised gene by analysing its effects across all of our unbiased screening data (5).



TECHNOLOGY CORE FACILITIES

125

LIGHT MICROSCOPY www.london-research-institute.org.uk/technologies/light-microscopy

We provide services in multi-dimensional imaging with fixed and live specimens using confocal microscopy and low-light-level imaging. Support is also available for image processing and motion analysis. In addition, we pursue collaborative research in application of these techniques.

Head

Daniel Zicha Staff Deborah Aubyn Trevor Duhig Peter Jordan Alastair Nicol

Imaging technology overview Laser scanning confocal microscopes LSM 780s (Zeiss) including a multiphoton system with a tuneable Chameleon Ultra II laser (Coherent), LSM 710s (Zeiss), SP5 (Leica) and LSM 510s (Zeiss), UltraVIEW spinning-disk confocal imaging system (PerkinElmer), low-light-level imaging systems based on Metamorph software (Molecular Devices) including a confocal high content screening system Discovery 1 (Molecular Devices), and a microinjection system (Eppendorf) have been configured for contrast enhancement, high resolution 3D and dynamic imaging of biological specimens in multiple fluorescence channels with optical sectioning using motorised focus at multiple fields. Detailed information on functionality of individual imaging systems is presented on the Intranet. Image processing and statistical analysis can be employed for deconvolution, colocalisation, automatic or

Figure 1 Example micrographs of LoVo cells acquired by laser scanning confocal microscopy. 4nM Hu-r-βGBP was added at hour 6 after seeding and cells were fixed at hour 48. Inhibition of PI3K activity is followed by changes of cell morphology brought about by cytoskeletal rearrangement.

interactive segmentation of cells and intracellular structures, morphometry, and tracking using Huygens (SVI), Volocity (Improvision), Imaris (Bitplane), AQM/ iQ (Kinetic Imaging/ Andor), Metamorph (Universal Imaging), MATLAB (MathWorks) and Mathematica (Wolfram Research). New functionality The ‘High Dynamic Range’ (HDR) imaging mode is now available on the LSM 780 confocal systems and has the ability to automatically change laser power during imaging depending on the distribution of intensities in the image. This is based on a similar idea introduced in Nikon’s Controlled Light Exposure Microscopy (CLEM) but the HDR is simpler in that it only utilises three distinct levels of laser power. Collaborative research highlights Quantitative imaging using laser scanning confocal microscopy and high content screening system Discovery 1 has been utilised in a collaborative project: Mallucci L, Shi DY, Davies D, Jordan P, Nicol A, Lotti L, Mariani-Costantini R, Verginelli F, Wells V, Zicha D. Killing of Kras-Mutant Colon Cancer Cells via Rac-Independent Actin Remodeling by the βGBP Cytokine, a Physiological PI3K Inhibitor Therapeutically Effective In Vivo. 2012; Mol Cancer Ther. 11: 1884-93. The morphological response illustrated in Figure 1 was quantified as a significant increase (ANOVA P < 0.001) of spread area, defined by F-actin, with control cells at 193.1±2.7 μm2 (N=707) and βGBP treated cells at 250.4±6.7 μm2 (N=179).

Publications listed on page 152

126

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

PEPTIDE SYNTHESIS www.london-research-institute.org.uk/technologies/peptide-synthesis

The Peptide Synthesis Facility provides peptides and peptide arrays to LRI scientists. We have worked with 31 LRI groups this year providing peptides and peptide arrays. Peptides have ranged from alpha factor to highly modified and long peptides (> 60 amino acids) for protein interaction and structural studies.

Head

Nicola O’Reilly Staff Ganka Bineva-Todd Stefania Federico Dhira Joshi Darryn Mark

We make peptides of many lengths and modifications including biotin addition, dye labels, phosphorylation, methylation, acetylation, farnesylation, branched peptides and peptides linked by disulphide bridges. We are keen to make unusual peptides or peptide-based reagents, which can enable scientists to further their research. We can aid in design of peptides and conjugation and immunisation strategies of peptides for antibody generation. We have around 20 peptides in stock that are used to elute proteins from columns, prime immune cells and synchronise yeast.

Figure 1 Cy5 NHS ester.

This year we have made cy5 labelled peptides for the first time. For flow cytometry Cy5 is a ‘brighter’ fluorophor which does not bleach as fast as fluorescein. It is also useful in FRET experiments in which GFP- and Cherry- labelled proteins are present because the absorption spectrum of Cy5 is not near the emission spectrum of GFP. In order to label peptides with Cy5, peptides are assembled on the solid support as per standard solid phase peptide synthesis. Cy5 is incorporated on to an available amine group (which is generally the N-terminus of the peptide) by using the N-hydroxysuccinimide modified dye. Two-to-five fold excess of dye over peptidyl resin is dissolved in dimethylformamide. Diisopropylethylamine is added to keep the pH basic because the reaction of NHS esters with amines is strongly pH dependant. The reaction is allowed to proceed overnight. Excess reactants are washed away and the peptide is cleaved and deprotected. A dark blue precipitate results. The peptide is then purified and can undergo further modification (such as being cross-linked to transport peptides or incorporated into proteins) as required.

Figure 2 Cy5 and Fluorescein labelled peptide.

Publications listed on page 152



TECHNOLOGY CORE FACILITIES

127

PROTEIN ANALYSIS AND PROTEOMICS www.london-research-institute.org.uk/technologies/protein-analysis-and-proteomics

Mass spectrometry is an emerging technology that can impact on all disciplines of the biomedical sciences. The LRI Protein Analysis and Proteomics facility at Clare Hall houses a state-of-the art instrument suite and a team of experts that actively seeks to develop and apply novel methods aligned with the needs of researchers at the LRI. Arguably, mass spectrometry has made its biggest impact in the field of proteomics. It is now feasible to identify and quantify thousands of proteins from a complex sample in a relatively short time. The Protein Analysis and Proteomics Facility aims to cover all aspects of protein mass spectrometry. Predominantly, the infrastructure is designed to enable the peptide centred approach for protein identification, quantification (SILAC, iBAQ, isobaric labelling), and the discovery, characterisation and quantification of post-translational modifications (PTMs).

Head

Bram Snijders Staff Karin Barnouin Andrea Di Fonzo Vesela Encheva David Frith Helen Flynn Sarah Maslen

Figure 1 Spectrum of peptide IAAELQCLTVAGGQDNVMGVK. Cysteine-107 of the DDK subunit CDC7 is protected from alkylation with Iodoacetamide when in complex with its heterodimeric partner DFB4. BLUE: Isotope envelope of the light reaction of peptide. RED: Isotope envelope resulting from the ‘heavy’ reaction. The light reaction was performed under SDS denaturing conditions. The heavy reaction was performed under native conditions. A. CDC7 +DFB4, cys-107 was protected from heavy alkylation. B. CDC7 alone, cys-107 was exposed and available for the alkylation reaction. C. BLUE reaction under denaturing conditions. D. RED reaction under denaturing conditions.

Global analysis of post-translational modifications Protein phosphorylation plays a major role in defects and effects of cellular signalling pathways. Currently, mass spectrometry is the only technology that can give insight in the site-specific dynamics of this and other protein PTMs for thousands of modified residues at the same time. We have established a workflow that can characterise >10000 phosphosites per cell lysate. The workflow requires a number of PTM enrichment and affinity purification steps and rigorous statistical data analysis. The field of PTM analysis by mass spectrometry is fast-moving and a continuing focus for the facility. Interaction proteomics The co-immunoprecipitation of proteins followed by their identification by mass spectrometry is a

B

C

100

100

100

90

90

90

90

80

80

80

70

70

70

70

60

60

60

60

50

50

50

50

40

40

40

40

30

30

30

30

20

20

20

20

10

10

0

0 1088

128

1090

1092

1094

1086

1088

1090

1092

1094

Mass spectrometer ThermoScientific LTQ Orbitrap XL/ETD ThermoScientific Q-Exactive ThermoScientific Velos Orbitrap Waters Synapt HDMS Agilent 6510 AB Sciex 5800 TOF/TOF

80

10 0 1086

Structural proteomics Native mass spectrometry can be used to gain insight into the composition and shape of proteins and complexes. In addition, a number of peptide centred approaches such as chemical crosslinking and surface labeling can yield further low resolution protein structure. With the Chromatin Structure and Mobile DNA Group we used an alkylation protection assay to study the solvent accessibility of the CDC7 Cys107 in the CDC7–DBF4 complex (see publications). These mass spectrometry based methods are attractive since compared to other structural approaches they can be performed on heterogeneous samples and require low amounts of material.

D

100

RELATIVE ABUNDANCE

RELATIVE ABUNDANCE

A

powerful strategy for the discovery of novel protein complexes. Despite this, it can be challenging to distinguish bona fide interactors from contaminating proteins on the basis of a binary classifier alone (absence or presence of identification). The facility therefore maintains a database of protein interactions and common contaminants that are easily accessible by users and can be used for comparison and filtering of the data. Furthermore, quantitative information from spectral counts or peptide signal intensities can be used as the input for more powerful classification algorithms embedded in the software.

10

1088

1090 m/z

1092

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

1094

0 1086

1088

1090 m/z

1092

1094

Publications listed on page 153

PROTEIN PURIFICATION www.london-research-institute.org.uk/technologies/protein-purification

The 5th year of the Protein Purification Facility (PPF) has been brimming with activity and seen new developments broadening the functional profile of the facility.

Head

Svend Kjær Staff Annabel Borg Roger Rajesh George Sara Elizabeth Kisakye-Nambozo

PPF – core operations PPF is an approachable local solution to challenges LRI researchers may face in the process of producing and characterizing proteins. PPF supports its users at all steps from cloning strategies to pure material with advice, reagents, instruments and hands-on. The base of the operation is a comprehensive vector bank, which offers ready-to-go expression vectors with a variety of affinity-tags and protease cleavage sites. The core technology for protein expression remains the use of insect cells infected with engineered baculovirus. The advantage of insect cells is their ability to express large, multi-domain proteins, and even multi-protein complexes by use of the Multibac system, with native or near-native

essential information of the antibody and allowing modification of binding and other properties of the Fab. The Fab fragments are valuable reagents with a multitude of applications as cocrystallization reagents, for cell-surface receptor blocking experiments (and hence therapeutic agents) as well as general protein detection applications.

post-translational modifications (PTM’s). In addition to insect cells, PPF also provides advice on E.coli expression and established during 2012, a platform for transient large-scale protein expression in mammalian cell suspension cultures as a complement to stably transfected cell-lines for secreted proteins.

The year ahead PPF intends to maintain its high level of activity, optimise work-flows and wishes to initiate new collaborations and technological innovations. One particular area of focus will be in vitro reconstitution of protein complexes, where the interactions are PTM-dependent. Moreover, exploration of new expression systems such as suspension culture S2 cells is underway and will offer a new strategy for expression of secreted proteins.

Figure 1 The Octet measuring proteinprotein interactions in real-time by biolayer interferometry.

Cloning and expression of Fab antibody fragments The transient mammalian expression has found applications for production of Fab antibody fragments. PPF has implemented techniques for rapid amplification and cloning of VH and VL genes from hybridoma cell-lines, thus capturing the



New instruments and applications During 2012 PPF expanded the instrument portfolio with an OctetRed96 (Figure 1) and a Wyatt MALS (Multi Angle Laser Scattering) system. The Octet is based on biolayer interferometry to detect macromolecular interactions in real time in a ’walk-up-and-use’ 96-well format, revealing the kinetic profile as well as the strength of the interaction. The Octet is a versatile instrument also enabling quantification of proteins in complex mixtures such as lysates or serum. Amongst other uses, Pavel Hanc from the Immunobiology Group has applied the Octet for studies on the interaction of a F-actin ligand with F-actin polymerised on the Octet biosensor. The Wyatt MALS system consists of a Dawn8 unit connected to an Optilab rEX refractive index detector, which in concert allows determination of the absolute molecular weight of proteins and protein complexes regardless of shape.

Publications listed on page 153

TECHNOLOGY CORE FACILITIES

129

TRANSGENICS

The Transgenic Facility is based at Clare Hall and comprises a team of seven. The facility aims to assist those groups that work with mouse models providing the means to archive and rederive strains and the techniques to produce novel mouse strains to advance research projects.

Head

Ian Rosewell

Techniques The specialist techniques that we provide centre on embryo transfer. The embryos that are transferred can come from a multitude of sources, from simple collection for the purposes of a rederivation or from in vitro fertilisation (IVF) or from a cryopreserved sample. They might also stem from a microinjection procedure, either pronuclear injection, to deliver cDNA, BAC DNA, RNA or nucleases into the mouse germline, or embryonic stem (ES) cell providing for chimera generation for the production of, for example, gene-targeted strains. Developments Though ES cells are transfected in house a particular increase in the last two years has stemmed from the injection of imported ES cells. ES cells harbouring combined conditional and reporter alleles and based on inbred background are readily sourced from the International

Figure 1 First ES cell colonies arising after dissaggregation of an outgrowth formed by a blastocyst in culture.

Knockout Mouse Consortium. These have proved to be reliable and coverage is now such that cells are available with a mutation in many of the known genes. ES cell derivation is now a routine and rapid procedure. The resulting ES cell lines are used in a multitude of experiments, for in vitro differentiation protocols or to repeatedly generate chimera cohorts from complex genotypes, overcoming breeding bottlenecks. A number of ES cell lines of differing genetic background, have been produced, characterised and used in routine targeting projects. These cell lines, and support for their growth, are freely available. The means for the cryopreservation of mouse strains is most often achieved with the collection and freezing of sperm from the epidiymus avoiding several breeding steps. IVF is then used to gauge the viability of stored sperm. Protocols have become available in the last year, which have significantly improved results that can be achieved with thawed sperm from an IVF procedure. The three developments outlined above, as well as improving the services that we provide, all provide a quantifiable ‘3R’s’ benefit. The 3R’s (reduction, replacement and refinement) is a principle of animal based research, which is a key guiding principle in our work.

130

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

FERMENTATION

IN VIVO IMAGING www.london-research-institute.org.uk/ technologies/in-vivo-imaging

Fermentation

Alireza Alidoust Namita Patel

The Fermentation Unit is based at the Clare Hall Laboratories and provides pilot-plant scale production of microbiological organisms for LRI scientists. Batches of ten to one hundred litres of yeast, bacteria or nematodes can be produced under stringently controlled conditions using four state-of-the art fermenters. It is possible to run batch or fed-batch cultures where additional nutrient or induction chemicals can be added at any point during the run. Computer control of the main growth parameters (pH, temperature, dissolved oxygen, aeration and agitation) allows a wide spectrum of growth conditions and reproducibility. The types of organisms we grow routinely are yeasts (Saccharomyces cerevisiae, Schizosaccharomyces pombe and Pichia pastoris), the bacteria Escherichia coli and Caenorhabditis elegans (nematode) for whole cell production or protein purification. The service also offers breakage of cells using one of three available methods (cell disrupter, ball mill and freezer mill).

The In Vivo Imaging Facility: non-invasive imaging from whole body to subcellular scale. An important part of this year has been dedicated to improving and consolidating the use of micro-CT imaging for longitudinal quantification of autochthonous lung tumours (Kumar et al., 2012; Cell. 149: 642-55). Acquisition of the RespGate software improved retrospective gating. Automating image segmentation is one of our priorities. [18]F-radiolabelled glucose is routinely used to analyse tumour metabolism in cancer patients. In collaboration with the Haematopoietic Stem Cell Group we have obtained and characterized nearinfrared labelled 2-Deoxyglucose (NIR-2DG), which might offer the opportunity to perform similar analyses without radioactivity (Figure 1). Needs to diagnose pancreatic and prostate cancers lead us to evaluate the usefulness of highresolution ultrasound imaging. This technology will become available in the near future. The National Centre for Replacement, Reduction and Refinement of Animals in Research (NC3Rs) has recognised the quality of the work undertaken in the facility and decided to fund some of our projects.

In Vivo Imaging

Francois Lassailly

We have launched the LIVIm (LRI In Vivo Imaging) Club, which now provides the first UK platform dedicated to these interdisciplinary exchanges. Publications listed on page 153

Fermentation Figure 1 New Brunswick fermenter.

In Vivo Imaging Figure 1 Orhtotopic human acute myeloid leukaemia (AML) imaged thanks to the co-registration of bioluminescence (BLI, top) and near-infrared imaging (NIRI, bottom) after administration of a NIR-2DG probe.



TECHNOLOGY CORE FACILITIES

131

Mathematical Modelling

Alexander Tournier

MATHEMATICAL MODELLING

PROTEIN STRUCTURE

This unit uses mathematical and computational approaches to study complex biological systems. To maximise biological relevance, this work is done in tight collaboration with the research groups in the building.

X-ray crystallography can provide highly detailed molecular structure of large molecular assemblies. Elucidating the atomic details of molecular interactions is particularly important to understand the biological function of protein/ enzyme at an atomic level.

A recent collaboration with the Apoptosis and Proliferation Control Group has investigated how mechanical forces can shape epithelial tissue during growth and morphogenesis. To this end we developed a computational model simulating the growth of the wing epithelium. We have also developed advanced image analysis techniques to extract and quantity complex spatial and dynamical data from the high-resolution time-lapsed microscopy data. In collaboration with the Epithelial Biology Group, we have developed a stochastic model to explore the mechanisms that underpin the establishment of polarity in epithelial cells precursors. Publications listed on page 153 A

B

C

D

Protein Structure

Stéphane Mouilleron

The Protein Structure unit was recently formed in order to provide to any biology group from the LRI the opportunity to undertake structural studies of their favourite proteins in-house at the LRI. This allows groups, collaborative access to the state-ofthe-art high throughput crystallisation robots and regular access to the ESRF (Grenoble, France) and Diamond synchrotron (Oxford, UK) to collect high quality data from protein crystals. Scientific projects The unit is currently running two successful collaborations, one with the Signalling and Transcription, and Structure Biology Groups which lead to the resolution of six crystal structures of the Phactr1 protein bound to G-actin (Figure 1A) (Mouilleron et al., 2012; Structure. 20: 1960-70; Wiezlak et al., 2012; J Cell Sci.) and a second one with the Secretory Pathways Group for which we recently solved the structure of SCOC coiled-coil domain which reveals an imperfect dimer interface (Figure 1B). These structures have provided a strong insight into the molecular basis for Phactr1 and SCOC function. Publications listed on page 153 A

Mathematical Modelling Figure 1 Stochastic simulation of the establishment and maintenance of polarity in cells as precursors to the formation of epithelia. A. The simple case of a self-recruiting membrane protein. B. Two membrane proteins in the absence of any interaction. C. Green membrane protein takes off red membrane protein. D. Red membrane protein establishes stable domain by self-recruitment in the presence of antagonising green membrane protein.

B

Protein Structure Figure 1 A. Structure of the Phactr1 RPEL domain crank (red and white solid rendering) bound to three G-actin molecules (pale blue, green, and pink ribbon). B. Structure of the parallel dimeric coiled-coil of SCOC.

132

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

SUPER-RESOLUTION MICROSCOPY www.london-research-institute.org.uk/ technologies/super-resolution-microscopy

Why use super-resolution microscopy? Many important biological and cellular structures such as viruses, bacteria, and synapses are so small that they are difficult to study using conventional light microscopy. Super-resolution microscopy refers to new light microscopy techniques that have improved resolution and that can be used without the laborious sample preparation required for electron microscopy. Super-Resolution Microscopy

Anne Vaahtokari

Our super-resolution microscope The LRI super-resolution microscope is DeltaVision OMX V3 (API/GE Healthcare) instrument that uses structured illumination to provide threedimensional super-resolution images for up to three fluorescent channels. The OMX is equipped with five lasers and four EMCCD cameras. It takes about a minute to acquire raw images throughout a mammalian cell and five minutes to reconstruct the final super-resolution image. Super-resolution microscopy projects The research projects in which the OMX microscope has been used vary from studies of viral infection of cells to establishment of cell movement during angiogenesis. In addition to mammalian cells and tissue sections, we have imaged bacteria, yeast, and developing Drosophila tissues. Recently, super-resolution imaging of live yeast cells has successfully been accomplished.

Figure 1 COS7 cell showing DNA labelled with Hoechst (blue) and GFP-Lamin B receptor (grey) in endoplasmic reticulum and nuclear envelope. In collaboration with Charlotte Melia (MSc student, Cell Biophysics Group).



TECHNOLOGY CORE FACILITIES

133

B cells recognise stimulating antigens on a surface and spread over that surface. During spreading they alter their cytoskeleton (shown are microtubules in red and actin in green) and later divide due to the stimulation. This image was acquired with structured illumination microscopy, a new high-resolution imaging method available at the LRI. Image: Lymphocyte Interaction Group.

134

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

RESEARCH PUBLICATIONS

RESEARCH PUBLICATIONS THESES SUBMITTED 2012



RESEARCH PUBLICATIONS

135

RESEARCH PUBLICATIONS

Facundo Batista (page 28) Lymphocyte Interaction Group Primary Research Papers Ahrens S, Zelenay S, Sancho D, Hanč P, Kjaer S, Feest C, Fletcher G, Durkin C, Postigo A, Skehel B, Batista FD, Thompson B, Way M, Reis e Sousa C, Schulz O. F-actin is an evolutionarily-conserved damage-associated molecular pattern recognized by DNGR-1, a receptor for dead cells. Immunity. 2012;36(4):635-45

Barral P, Sánchez-Niño MD, van Rooijen N, Cerundolo V, Batista FD. The location of splenic NKT cells favours their rapid activation by blood-borne antigen. EMBO J. 2012;31(10):2378-90 Lösing M, Goldbeck I, Manno B, Oellerich T, Schnyder T, Bohnenberger H, Stork B, Urlaub H, Batista FD, Wienands J, Engelke M. The Dok-3/Grb2 signal module attenuates Lyn-dependent activation of Syk in B cell antigen receptor Microclusters. J Biol Chem. 2012;doi:10.1074/jbc.M112.406546 Thaunat O, Batista FD. Daughter B cells are not created equal: Asymmetric segregation of antigen during B cell division. Cell Cycle. 2012;11(12):2219-20

Thaunat O, Granja AG, Barral P, Filby A, Montaner B, Collinson L, Martinez-Martin N, Harwood NE, Bruckbauer A, Batista FD. Asymmetric segregation of polarized antigen on B cell division shapes presentation capacity. Science. 2012;335(6067):475-9 Other Publications Harwood NE, Barral P, Batista FD. Neutrophils-the unexpected helpers of B-cell activation. EMBO Rep. 2012;13(2):93-4

Axel Behrens (page 30) Mammalian Genetics Group Primary Research Papers Arthur-Farraj PJ, Latouche M, Wilton DK, Quintes S, Chabrol E, Banerjee A, Woodhoo A, Jenkins B, Rahman M, Turmaine M, Wicher GK, Mitter R, Greensmith L, Behrens A, Raivich G, Mirsky R, Jessen KR. c-Jun reprograms Schwann cells of injured nerves to generate a repair cell essential for regeneration. Neuron. 2012;75(4):633-47

Fontana X, Hristova M, Da Costa C, Patodia S, Thei L, Makwana M, Spencer-Dene B, Latouche M, Mirsky R, Jessen KR, Klein R, Raivich G, Behrens A. c-Jun in Schwann cells promotes axonal regeneration and motoneuron survival via paracrine signaling. J Cell Biol. 2012;198(1):127-41 Izumi N, Helker C, Ehling M, Behrens A, Herzog W, Adams RH. Fbxw7 controls angiogenesis by regulating endothelial notch activity. PLoS One. 2012;7(7):e41116 Kumar MS, Hancock DC, Molina-Arcas M, Steckel M, East P, Diefenbacher M, Armenteros-Monterroso E, Lassailly F, Matthews N, Nye E, Stamp G, Behrens A, Downward J. The GATA2 transcriptional network is requisite for RAS oncogene-driven non-small cell lung cancer. Cell. 2012;149(3):642-55

Electron microscopy of B cells as they spread out through membrane protrusions to gather antigen coated on a surface

136

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Reynolds N, Latos P, Hynes-Allen A, Loos R, Leaford D, O’Shaughnessy A, Mosaku O, Signolet J, Brennecke P, Kalkan T, Costello I, Humphreys P, Mansfield W, Nakagawa K, Strouboulis J, Behrens A, Bertone P, Hendrich B. NuRD suppresses pluripotency gene expression to promote tanscriptional heterogeneity and lineage commitment. Cell Stem Cell. 2012;10(5):583-94

Ruff C, Staak N, Patodia S, Kaswich M, Rocha-Ferreira E, Costa CD, Brecht S, Makwana M, Fontana X, Hristova M, Rumajogee P, Galiano M, Bohatschek M, Herdegen T, Behrens A, Raivich G. Neuronal c-Jun is required for successful axonal regeneration, but the effects of phosphorylation of its N-terminus are moderate. J Neurochem. 2012;121(4):607-18 Vinuesa AG, Sancho R, García-Limones C, Behrens A, Ten Dijke P, Calzado MA, Muñoz E. Vanilloid Receptor-1 regulates neurogenic inflammation in colon and protects mice from colon cancer. Cancer Res. 2012;72(7):1705-16 Zhang T, Penicud K, Bruhn C, Loizou JI, Kanu N, Wang Z, Behrens A. Competition between NBS1 and ATMIN controls ATM signalling pathway choice. Cell Rep. 2012;2(6):1498-504

Dominique Bonnet (page 32) Haematopoietic Stem Cell Group Primary Research Papers Griessinger E, Jayasinghe SN, Bonnet D. Aerodynamically assisted bio-jetting of hematopoietic stem cells. Analyst. 2012;137(6):1329-33

Ibrahim EE, Babaei-Jadidi R, Saadeddin A, Spencer-Dene B, Hossaini S, Abuzinadah M, Li N, Fadhil W, Ilyas M, Bonnet D, Nateri AS. Embryonnic NANOG activity defines colorectal cancer stem cells and modulates through AP1 and TCF-dependent mechanisms. Stem Cells. 2012;30(10):2076-2087 Jaganathan BG, Bonnet D. Human mesenchymal stromal cells senesce with exogenous OCT4. Cytotherapy. 2012;14(9):1054-63 Kallinikou K, Anjos-Afonso F, Blundell MP, Ings SJ, Watts MJ, Thrasher AJ, Linch DC, Bonnet D, Yong KL. Engraftment defect of cytokine-cultured adult human mobilized CD34(+) cells is related to reduced adhesion to bone marrow niche elements. Br J Haematol. 2012;158:778-87 Patel S, Zhang Y, Cassinat B, Zassadowski F, Ferré N, Cuccuini W, Cayuela JM, Fenaux P, Bonnet D, Chomienne C, Louache F. Successful xenografts of AML3 samples in immunodeficient NOD/shi-SCID IL2Rγ(-/-) mice. Leukemia. 2012;26(11):2432-5

Poulin LF, Reyal Y, Uronen-Hansson H, Schraml B, Sancho D, Murphy KM, Håkansson UK, Ferreira Moita L, Agace WW, Bonnet D, Reis E Sousa C. DNGR-1 is a specific and universal marker of mouse and human Batf3-dependent dendritic cells in lymphoid and non-lymphoid tissues. Blood. 2012;119(24):5722-30



Protein crystals from purified proteins.

Quintana-Bustamante O, Smith LL, Griessinger E, Reyal Y, Vargaftig J, Lister TA, Fitzgibbon J, Bonnet D. Overexpression of wild-type or mutants forms of CEBPA alter normal human hematopoiesis. Leukemia. 2012;26(7):1537-46 Tamoutounour S, Henri S, Lelouard H, de Bovis B, de Haar C, van der Woude CJ, Woltman AM, Reyal Y, Bonnet D, Sichien D, Bain CC, Mowat AM, Reis e Sousa C, Poulin LF, Malissen B, Guilliams M. CD64 distinguishes macrophages from dendritic cells in the gut and reveals the Th1-inducing role of mesenteric lymph node macrophages during colitis. Eur J Immunol. 2012;42(12):3150-66 Vargaftig J, Taussig D, Griessinger E, Ansjos-Afonso F, Lister TA, Cavenagh J, Oakervee H, Gribben J, Bonnet D. Frequency of Leukemia initiating cells does not depend on the xenotransplantation model used. Leukemia. 2012;26(4):858-60 Willems L, Chapuis N, Puissant A, Maciel TT, Green AS, Jacque N, Vignon C, Park S, Guichard S, Herault O, Fricot A, Hermine O, Moura IC, Auberger P, Ifrah N, Dreyfus F, Bonnet D, Lacombe C, Mayeux P, Bouscary D, Tamburini J. The dual mTORC1 and mTORC2 inhibitor AZD8055 has anti-tumor activity in acute myeloid leukemia. Leukemia. 2012;26(6):1195-202 Other Publications Valent P, Bonnet D, De Maria R, Lapidot T, Copland M, Melo JV, Chomienne C, Ishikawa F, Schuringa JJ, Stassi G, Huntly B, Herrmann H, Soulier J, Roesch A, Schuurhuis GJ, Wöhrer S, Arock M, Zuber J, Cerny-Reiterer S, Johnsen HE, Andreeff M, Eaves C. Cancer stem cells definitions and terminology: the devil is in the details. Nat Rev Cancer. 2012;12(11):767-75

RESEARCH PUBLICATIONS

137

Julia Promisel Cooper (page 34) Telomere Biology Group Primary Research Papers Dehé PM, Rog O, Ferreira MG, Greenwood J, Cooper JP. Taz1 enforces cell-cycle regulation of telomere synthesis. Mol Cell. 2012;46(6):797-808

Greenwood J, Cooper JP. Non-coding telomeric and subtelomeric transcripts are differentially regulated by telomeric and heterochromatin assembly factors in fission yeast. Nucleic Acids Res. 2012;40(7):2956-63 Other Publications Cooper JP, Youle RY. Balancing cell growth and death. Curr Opin Cell Biol. 2012;24(6):802-3

Ebrahimi H, Cooper JP. Closed mitosis: A timely move before separation. Curr Biol. 2012;22(20):R880-2

of synthetic lethal interactions in KRAS oncogenedependent cancer cells reveals novel therapeutic targeting strategies. Cell Res. 2012;22(8):1227-45

Julian Downward (page 36)

Other Publications Cully M, Downward J. Assessing cell size and cell cycle regulation in cells with altered TOR activity. Methods Mol Biol. 2012;821:227-37

Signal Transduction Group Primary Research Papers Bogel G, Gujdar A, Geiszt M, Lanyi A, Fekete A, Sipeki S, Downward J, Buday L. Frank-ter Haar syndrome protein Tks4 regulates EGF-dependent cell migration. J Biol Chem. 2012;287(37):31321-9

Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366(10):883-92 Kumar MS, Hancock D, Molina-Arcas M, Steckel M, East P, Diefenbacher M, Armenteros-Monterroso E, Lassailly F, Matthews N, Nye E, Stamp G, Behrens A, Downward J. The GATA2-driven oncogene network is requisite for RAS oncogene-driven no-small cell lung cancer. Cell. 2012;149(3):642-55 Lee AJ, Roylance R, Sander J, Gorman P, Endesfelder D, Kschischo M, Jones NP, East P, Nicke B, Spassieva S, Obeid LM, Juul Birkbak N, Szallasi Z, McKnight NC, Rowan AJ, Speirs V, Hanby AM, Downward J, Tooze SA, Swanton C. CERT depletion predicts chemotherapy benefit and mediates cytotoxic and polyploid-specific cancer cell death through autophagy induction. J Pathol. 2012;226(3):482-94 Steckel M, Molina-Arcas M, Weigelt B, Marani M, Warne PH, Kuznetsov H, Kelly G, Saunders B, Howell M, Downward J, Hancock DC. Determination

138

Confocal laser scanning micrograph of murine venous valve stained with antibodies against lymphatic regulators Integrinalpha9 (red) and Prox1 (green).

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Molina-Arcas M, Downward J. How to fool a wonder drug: truncate and dimerize. Cancer Cell. 2012;21(1):7-9 Weigelt B, Downward J. Genomic determinants of PI3K pathway inhibitor response in cancer. Frontiers in Oncology. 2012;2:109

Holger Gerhardt (page 38) Vascular Biology Group Primary Research Papers Cesca F, Yabe A, Spencer-Dene B, Scholz-Starke J, Medrihan L, Maden CH, Gerhardt H, Orriss IR, Baldelli P, Al-Qatari M, Koltzenburg M, Adams RH, Benfenati F, Schiavo G. Kidins220/ARMS mediates the integration of the neurotrophin and VEGF pathways in the vascular and nervous systems. Cell Death Differ. 2012;19(2):194-208

Fraccaroli A, Franco CA, Rognoni E, Neto F, Rehberg M, Aszodi A, Wedlich-Söldner R, Pohl U, Gerhardt H, Montanez E. Visualization of endothelial actin cytoskeleton in the mouse retina. PLoS One. 2012;7(10):e47488 Other Publications Blanco R, Gerhardt H. VEGF and Notch in tip and stalk cell selection. Cold Spring Harb Perspect Med. 2012;pii:a006569.doi:10.1101/cshperspect.a006569

Franco C, Gerhardt H. Tissue engineering: Blood vessels on a chip. Nature. 2012;488(7412):465-6

Adrian Hayday (page 40)

Caroline Hill (page 42)

Immuno Surveillance Group

Developmental Signalling Group

Primary Research Papers Knight J, Spain SL, Capon F, Hayday A, Nestle FO, Clop A. Wellcome Trust case control consortium; genetic analysis of psoriasis consortium; I-chip for psoriasis consortium, Barker JN, Weale ME, Trembath RC. Conditional analysis identifies three novel major histocompatibility complex loci associated with psoriasis. Hum Mol Genet. 2012;21(23):5185-92

Primary Research Papers Grönroos E, Kingston IJ, Ramachandran A, Randall RA, Vizán P, Hill CS. Transforming growth factor β inhibits bone morphogenetic protein-induced transcription through novel phosphorylated Smad1/5-Smad3 complexes. Mol Cell Biol. 2012;32(14):2904-16

Michel ML, Pang DJ, Haque SF, Potocnik AJ, Pennington DJ, Hayday AC. Interleukin 7 (IL-7) selectively promotes mouse and human IL-17producing γδ cells. Proc Natl Acad Sci USA. 2012;109(43):17549-54 Modi BG, Neustadter J, Binda E, Lewis J, Filler RB, Roberts SJ, Kwong BY, Reddy S, Overton JD, Galan A, Tigelaar R, Cai L, Fu P, Shlomchik M, Kaplan DH, Hayday A, Girardi M. Langerhans cells facilitate epithelial DNA damage and squamous cell carcinoma. Science. 2012;335(6064):104-8 Roberts NA, White AJ, Jenkinson WE, Turchinovich G, Nakamura K, Withers DR, McConnell FM, Desanti GE, Benezech C, Parnell SM, Cunningham AF, Paolino M, Penninger JM, Simon AK, Nitta T, Ohigashi I, Takahama Y, Caamano JH, Hayday AC, Lane PJ, Jenkinson EJ, Anderson G. Rank signaling links the development of invariant γδ T cell progenitors and Aire(+) medullary epithelium. Immunity. 2012;36(3):427-37

Other Publications Hill CS, van Aelst L. Signaling and gradients: what’s going down? Curr Opin Cell Biol. 2012;24(2):155-7

Hill CS. Inhibiting the Inhibitor: The role of RNF12 in TGF-β superfamily signaling. Mol Cell. 2012;46(5):558-9 Ramel MC, Hill CS. Spatial regulation of BMP activity. FEBS Lett. 2012;586(14):1929-41

David Ish-Horowicz (page 44) Developmental Genetics Group Primary Research Papers Stauber M, Laclef C, Vezzaro A, Page ME, Ish-Horowicz D. Modifying transcript lengths of cycling mouse segmentation genes. Mech Dev. 2012;129(1-4):61-72

Banafshe Larijani (page 46) Cell Biophysics Group

Tsoi LC, ....Hayday A, et al. Identification of 15 new psoriasis susceptibility loci highlights the role of innate immunity. Nature Genetics. 2012;44:1341-8

Primary Research Papers Byrne RD, Applebee C, Poccia DL, Larijani B. Dynamics of PLCγ and Src Family Kinase 1 interactions during nuclear envelope formation revealed by FRET-FLIM. PLoS One. 2012;7(7):e40669

Willcox CR, Pitard V, Netzer S, Couzi L, Salim M, Silberzahn T, Moreau JF, Hayday AC, Willcox BE, Déchanet-Merville J. Cytomegalovirus and tumor stress surveillance by binding of a human γδ T cell antigen receptor to endothelial protein C receptor. Nat Immunol. 2012;13(9):872-9

Calleja V, Leboucher P, Larijani B. Protein activation dynamics in cells and tumor micro arrays assessed by time resolved förster resonance energy transfer. Methods Enzymol. 2012;506:225-46

Zhang X, Volpe EA, Gandhi NB, Schaumburg CS, Siemasko KF, Pangelinan SB, Kelly SD, Hayday AC, Li DQ, Stern ME, Niederkorn JY, Pflugfelder SC, De Paiva CS. NK cells promote Th-17 mediated corneal barrier disruption in dry eye. PLoS One. 2012;7:e36822

Domart MC, Hobday TMC, Peddie CJ, Chung GHC, Wang A, Yeh K, Jethwa N, Zhang Q, Wakelam MJO, Woscholski R, Byrne RD, Collinson LM, Poccia D, Larijani B. Acute manipulation of diacylglycerol reveals roles in nuclear envelope assembly and endoplasmic reticulum morphology. PLoS One. 2012;7(12):e51150

Other Publications Abeler-Dorner L, Swamy M, Williams G, Hayday AC, Bas A. Butyrophilins: an emerging family of immune regulators. Trends in Immunology. 2012;33:34-41

Jethwa N, Fili N, Larijani B. Acute depletion of plasma membrane phospholipids – dissecting the roles of PtdIns(4)P and PtdIns(4,5)P2. Journal of Biol Chem. 2012;5(4):137-9

Hayday A, Tigelaar R. Casting new light on the TCR. Nat Immunol. 2012;13(3):209-11



RESEARCH PUBLICATIONS

139

Other Publications Kahramanoglou C, Prieto AI, Khedkar S, Haase B, Gupta A, Benes V, Fraser GM, Luscombe NM, Seshasayee AS. Genomics of DNA cytosine methylation in Escherichia coli reveals its role in stationary phase transcription. Nat Commun. 2012;3:886

König J, Zarnack K, Luscombe NM, Ule J. Protein-RNA interactions: new genomic technologies and perspectives. Nat Rev Genet. 2012;13(2):77-83 Human chromosomes in three different flavours of structural compaction.

López DJ, Egido-Gabas M, López-Montero I, Busto JV, Casas J, Garnier M, Monroy F, Larijani B, Goñi FM, Alonso A. Accumulated bending energy elicits neutral sphingomyelinase activity in human red blood cells. Biophys J. 2012;102(9):2077-85 Other Publications Byrne RD. The nuclear membrane as a lipid ‘sink’ – linking cell cycle progression to lipid synthesis. J Chem Biol. 2012;5(4):141-2

Larijani B, Poccia DL. Effects of phosphoinositides and their derivatives on membrane morphology and function. Curr Top Microbiol Immunol. 2012;362:99-110

Nicholas Luscombe (page 48) Computational Biology Group Primary Research Papers Conrad T, Cavalli FM, Holz H, Hallacli E, Kind J, Ilik I, Vaquerizas JM, Luscombe NM, Akhtar A. The MOF chromobarrel domain controls genome-wide H4K16 acetylation and spreading of the MSL complex. Dev Cell. 2012;22(3):610-24

Conrad T, Cavalli FM, Vaquerizas JM, Luscombe NM, Akhtar A. Drosophila dosage compensation involves enhanced Pol II recruitment to male X-linked promoters. Science. 2012;337(6095):742-6 Lam KC, Mühlpfordt F, Vaquerizas JM, Raja SJ, Holz H, Luscombe NM, Manke T, Akhtar A. The NSL complex regulates housekeeping genes in Drosophila. PLoS Genet. 2012;8(6):e1002736 Reimand J, Aun A, Vilo J, Vaquerizas JM, Sedman J, Luscombe NM. m:Explorer – multinomial regression models reveal positive and negative regulators of longevity in yeast quiescence. Genome Biol. 2012;13(6):R55 Martincorena I, Seshasayee AS, Luscombe NM. Evidence of non-random mutation rates suggests an evolutionary risk management strategy. Nature. 2012;485(7396):95-8

140

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Taija Makinen (page 50) Lymphatic Development Group Primary Research Papers Chen CY, Bertozzi C, Zou Z, Yuan L, Lee JS, Lu M, Stachelek SJ, Srinivasan S, Guo L, Vicente A, Mericko P, Levy RJ, Makinen T, Oliver G, Kahn ML. Blood flow reprograms lymphatic vessels to blood vessels. J Clin Invest. 2012;122(6):2006-17

Lutter S, Xie S, Tatin F, Makinen T. Smooth muscleendothelial cell communication activates Reelin signaling and regulates lymphatic vessel formation. J Cell Biol. 2012;197(6):837-49 Ostergaard P, Simpson MA, Mendola A, Vasudevan P, Connell FC, van Impel A, Moore AT, Loeys BL, Ghalamkarpour A, Onoufriadis A, Martinez-Corral I, Devery S, Leroy JG, van Laer L, Singer A, Bialer MG, McEntagart M, Quarrell O, Brice G, Trembath RC, Schulte-Merker S, Makinen T, Vikkula M, Mortimer PS, Mansour S, Jeffery S. Mutations in KIF11 cause autosomal-dominant microcephaly variably associated with congenital lymphedema and chorioretinopathy. Am J Hum Genet. 2012;90(2):356-62 Sabine A, Agalarov Y, Maby-El Hajjami H, Jaquet M, Hagerling R,Pollmann C, Bebber D, Pfenniger A, Miura N, Dormond O, Calmes JM, Adams RH, Makinen T, Kiefer F, Kwak BR, Petrova TV. Mechanotransduction, PROX1, and FOXC2 cooperate to control connexin37 and calcineurin during lymphatic-valve formation. Dev Cell. 2012;22(2):430-45 Wallgard E, Nitzsche A, Larsson J, Guo X, Dieterich LC, Dimberg A, Olofsson T, Pontén FC, Mäkinen T, Kalén M, Hellström M. Paladin (X99384) is expressed in the vasculature and shifts from endothelial to vascular smooth muscle cells during mouse development. Dev Dyn. 2012;241(4):770-86 Other Publications Bazigou E, Makinen T. Flow control in our vessels: vascular valves make sure there is no way back. Cell Mol Life Sci. 2012;doi:10.1007/s00018-012-1110-6

Neil McDonald (page 54) Structural Biology Group Primary Research Papers Bowles M, Lally J, Fadden AJ, Mouilleron S, Hammonds T, McDonald NQ. Fluorescence-based incision assay for human XPF-ERCC1 activity identifies important elements of DNA junction recognition. Nucleic Acids Res. 2012;40(13):e101

Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366(10):883-92 Mouilleron S, Wiezlak M, O’Reilly N, Treisman R, McDonald NQ. Structures of the Phactr1 RPEL domain and RPEL motif complexes with G-actin reveal the molecular basis for actin binding cooperativity. Structure. 2012;20(11):1960-70 Wiezlak M, Diring J, Abella JV, Mouilleron S, Way M, McDonald NQ, Treisman R. G-actin regulates shuttling and PP1 binding by the RPEL protein Phactr1 to control actomyosin assembly. J Cell Sci. 2012;doi:10.1242/jcs.112078

Paul Nurse/Jacqueline Hayles (page 56) Cell Cycle Group Primary Research Papers Kawashima S, Takemoto A, Nurse P, Kapoor TM. Analyzing fission yeast multi drug resistance mechanisms to develop a genetically tractable model system for chemical biology. Chem Biol. 2012;19:893-901

Kelly F, Nurse P. De novo growth zone formation from fission yeast spheroplasts. PLoS One. 2011;6:e27977 Li JJ, Schnick J, Hayles J, Macneill SA. Purification and functional inactivation of the fission yeast MCM(MCM-BP) complex. FEBS Lett. 2011;585:3850-5 Navarro FJ, Nurse P. A systematic screen reveals new elements acting at the G2/M cell cycle control. Genome Biol. 2012;13(5):R36 Tange Y, Kurabayashi A, Goto B, Hoe KL, Kim DU, Park HO, Hayles J, Chikashige Y, Tsutumi C, Hiraoka Y, Yamao F, Nurse P, Niwa O. The CCR4-NOT complex is implicated in the viability of aneuploid yeasts. PLoS Genet. 2012;8(6):e1002776



Other Publications Navarro FJ, Weston L, Nurse P. Global control of cell growth in fission yeast and its coordination with the cell cycle. Curr Opin Cell Biol. 2012; pii: S09550674(12)00178-0

Nurse P. Finding CDK: Linking yeast with humans. Nat Cell Biol. 2012;14(8):776 Nurse P. The cell cycle. Preface. Philos Trans R Soc Lond B Biol Sci. 2011;366:3493 Nurse P. In answer to questions about the Francis Crick Institute. Lancet. 2012;379,2427-8 Nurse P. Culture of creativity. Interview by Fred Guterl. Sci Am. 2012;307(4):52-3

Peter Parker (page 58) Protein Phosphorylation Group Primary Research Papers Milanovic M, Radtke S, Peel N, Howell M, Carrière V, Joffre C, Kermorgant S, Parker PJ. Anomalous inhibition of c-Met by the kinesin inhibitor aurintricarboxylic acid. Int J Cancer. 2012;130(5):1060-70

Rosse C, Boeckeler K, Linch M, Radtke S, Frith D, Barnouin K, Morsi AS, Hafezparast M, Howell M, Parker PJ. Binding of dynein intermediate chain 2 to Paxillin controls focal adhesion dynamics and migration. J Cell Sci. 2012;125:3733-8

Gordon Peters (page 60) Molecular Oncology Group Primary Research Papers O’Loghlen A, Muñoz-Cabello AM, Gaspar-Maia A, Wu HA, Banito A, Kunowska N, Racek T, Pemberton HN, Beolchi P, Lavial F, Masui O, Vermeulen M, Carroll T, Graumann J, Heard E, Dillon N, Azuara V, Snijders AP, Peters G, Bernstein E, Gil J. MicroRNA regulation of Cbx7 mediates a switch of Polycomb orthologs during ESC differentiation. Cell Stem Cell. 2012;10(1):33-46

Caetano Reis e Sousa (page 62) Immuobiology Group Primary Research Papers Ahrens S, Zelenay S, Sancho D, Hanč P, Kjær S, Feest C, Fletcher G, Durkin C, Postigo A, Skehel B, Batista FD, Thompson B, Way M, Reis e Sousa C, Schulz O. F-actin is an evolutionarily-conserved damage-associated molecular pattern recognized by DNGR-1, a receptor for dead cells. Immunity. 2012;36(4):635-45

RESEARCH PUBLICATIONS

141

Finney BA, Schweighoffer E, Navarro-Núñez L, Bénézech C, Barone F, Hughes CE, Langan SA, Lowe KL, Pollitt AY, Mourao-Sa D, Sheardown S, Nash GB, Smithers N, Reis e Sousa C, Tybulewicz VL, Watson SP. CLEC-2 and Syk in the megakaryocytic/ platelet lineage are essential for development. Blood. 2012;119(7):1747-56

Friedl P, Sahai E, Weiss S, Yamada KM. New dimensions in cell migration. Nat Rev Mol Cell Biol. 2012;13(11):743-7

Iborra S, Izquierdo HM, Martínez-López M, Blanco-Menéndez N, Reis E Sousa C, Sancho D. The DC receptor DNGR-1 mediates cross-priming of CTLs during vaccinia virus infection in mice. J Clin Invest. 2012;122(5):1628-43

Giampietro Schiavo (page 66)

Poulin LF, Reyal Y, Uronen-Hansson H, Schraml B, Sancho D, Murphy KM, Håkansson UK, Ferreira Moita L, Agace WW, Bonnet D, Reis e Sousa C. DNGR-1 is a specific and universal marker of mouse and human Batf3-dependent dendritic cells in lymphoid and non-lymphoid tissues. Blood. 2012;119(25):6052-62

Primary Research Papers Cesca F, Yabe A, Spencer-Dene B, Scholz-Starke J, Medrihan L, Maden CH, Gerhardt H, Orriss IR, Baldelli P, Al-Qatari M, Koltzenburg M, Adams RH, Benfenati F, Schiavo G. Kidins220/ARMS mediates the integration of the neurotrophin and VEGF pathways in cardiovascular and nervous system development. Cell Death Differ. 2012;19(2):194-208

Tamoutounour S, Henri S, Lelouard H, de Bovis B, de Haar C, van der Woude CJ, Woltman AM, Reyal Y, Bonnet D, Sichien D, Bain CC, McI Mowat A, Reis e Sousa C, Poulin LF, Malissen B, Guilliams M. CD64 distinguishes macrophages from dendritic cells in the gut and reveals the Th1-inducing role of mesenteric lymph node macrophages during colitis. Eur J Immunol. 2012;doi:10.1002/eji.201242847 Zelenay S, Keller AM, Whitney PG, Schraml BU, Deddouche S, Rogers NC, Schulz O, Sancho D, Reis e Sousa C. The dendritic cell receptor DNGR-1 controls endocytic handling of necrotic cell antigens to favor cross-priming of CTLs in virus-infected mice. J Clin Invest. 2012;122(5):1615-27 Other Publications Sancho D, Reis e Sousa C. Signaling by myeloid C-type lectin receptors in immunity and homeostasis. Annu Rev Immunol. 2012;30:491-529

Erik Sahai (page 64) Tumour Cell Biology Group Primary Research Papers Chaudhry SI, Hooper S, Nye E, Williamson P, Harrington K, Sahai E. Autocrine IL-1b-TRAF6 signalling promotes squamous cell carcinoma invasion through paracrine TNFa signalling to carcinoma-associated fibroblasts. Oncogene. 2012;doi:10.1038/onc.2012.91

Hirata E, Yukinaga H, Kamioka Y, Arakawa Y, Miyamoto S, Okada T, Sahai E, Matsuda M. In vivo fluorescence resonance energy transfer imaging reveals differential activation of Rho-family GTPases in glioblastoma cell invasion. J Cell Sci. 2012;125(Pt 4):858-68

142

Other Publications Friedl P, Locker J, Sahai E, Segall JE. Classifying collective cancer cell invasion. Nat Cell Biol. 2012;14(8):777-83

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Molecular Neuropathobiology Group

Cogli C, Progida C, Thomas CL, Spencer-Dene B, Donno C, Schiavo G, Bucci C. Charcot-Marie-Tooth type 2B disease-causing RAB7A mutant proteins show altered interaction with the neuronal intermediate filament peripherin. Acta Neuropathol. 2012;doi:10.1007/s00401-012-1063-8 Restani L, Giribaldi F, Manich M, Bercsenyi K, Menendez G, Rossetto O, Caleo M, Schiavo G. Botulinum neurotoxins A and E undergo retrograde axonal transport in primary motor neurons. PLoS Pathog. 2012;doi:10.1371/journal.ppat.1003087 Rishal I, Kam N, Perry RB, Shinder V, Fisher EM, Schiavo G, Fainzilber M. A motor driven mechanism for cell length sensing. Cell Report. 2012;1(6):608-16 Scholz-Starke J, Cesca F, Schiavo G, Benfenati F, Baldelli P. Kidins220/ARMS is a novel modulator of short-term synaptic plasticity in hippocampal GABAergic neurons. PLoS One. 2012;7(4):e35785 Wade A, Thomas C, Kalmar B, Terenzio M, Garin J, Greensmith L, Schiavo G. Activated leukocyte cell adhesion molecule (Alcam) modulates meurotrophin mignaling. J Neurochem. 2012;121(4):575-86 Other Publications Kuta A, Hafezparast M, Schiavo G, Fisher EMC. Genetic insights into mammalian cytoplasmic dynein function provided by novel mutations in mouse. Dyneins. S King Ed, Academic Press, London (UK) 2012;483-504

Naumann M, Dressler D, Hallett M, Jankovic J, Schiavo G, Segal KR, Truong D. Evidence-based review and assessment of botulinum neurotoxin for the treatment of secretory disorders. Toxicon. 2012;doi:10.1016/j.toxicon.2012.10.020

Martin Singleton (page 70) Macromolecular Structure and Function Group Primary Research Papers Perriches T, Singleton MR. The structure of the yeast kinetochore Ndc10 DNA-binding domain reveals an unexpected evolutionary relationship to tyrosine recombinases. J Biol Chem. 2012;287(7):5173-9

Zhang G, Kelstrup CD, Hu XW, Hansen MJ, Singleton MR, Olsen JV, Nilsson J. The Ndc80 internal loop is required for recruitment of the Ska complex to establish end-on microtubule attachment to kinetochores. J Cell Sci. 2012;125:3243-53

Thomas Surrey (page 72) Image of a frozen section of the spleen showing germinal centers (red), B cell area (blue) and T cell area (green).

Neubrand VE, Cesca F, Benfenati F, Schiavo G. Kidins220/ARMS as functional mediator of multiple receptor signalling pathways. J Cell Sci. 2012;125(8):1845-54 Thige AP and Schiavo G. Botulinum neurotoxins: mechanism of action. Toxicon. 2012;doi:10.1016/j. toxicon.2012.11.011

Almut Schulze (page 68) Gene Expression Analysis Group Primary Research Papers Ferber EC, Peck B, Delpuech O, Bell GP, East P, Schulze A. FOXO3a regulates reactive oxygen metabolism by inhibiting mitochondrial gene expression. Cell Death Differ. 2012;19(6):968-79

Ros S, Santos CR, Moço S, Baenke F, Kelly G, Howell M, Zamboni N, Schulze A. Functional screen identifies 6-phosphofructo-2-kinase/fructose-2,6biphosphatase (PFKFB4) as an important regulator of prostate cancer cell survival. Cancer Discovery. 2012;2(4):328-43 Other Publications Jones NP, Schulze A. Targeting cancer metabolism – aiming at a tumour’s sweet-spot. Drug Discov Today. 2012;17(5-6):232-41

Ros S, Schulze A. Linking glycogen and senescence in cancer cells. Cell Metab. 2012;16(6):687-8 Santos CR, Schulze A. Lipid metabolism in cancer. FEBS J. 2012;279(15):2610-23

Microtubule Cytoskeleton Group Primary Research Papers Gropeanu RA, Baumann H, Ritz S, Mailänder V, Surrey T, del Campo A. Phototriggerable 2',7-Caged Paclitaxel. PLoS One. 2012;7(9):e43657

Maurer SP, Fourniol FJ, Bohner G, Moores CA, Surrey T. EBs recognize a nucleotide-dependent structural cap at growing microtubule ends. Cell. 2012;149(2):371-82 Pecqueur L, Duellberg C, Dreier B, Jiang Q, Wang C, Plückthun A, Surrey T, Gigant B, Knossow M. A designed ankyrin repeat protein selected to bind to tubulin caps the microtubule plus end. Proc Natl Acad Sci USA. 2012;109(30):12011-6 Telley IA, Gáspár I, Ephrussi A, Surrey T. Aster migration determines the length scale of nuclear separation in the Drosophila syncytial embryo. J Cell Biol. 2012;197(7):887-95 Trokter M, Mücke N, Surrey T. Reconstruction of the human cytoplasmic dynein complex. Proc Natl Acad Sci USA. 2012;109(51):20895-900 Other Publications Dogterom M, Surrey T. Microtubule organization in vitro. Curr Opin Cell Biol. 2012;doi:pii:S09550674(12)00192-5. 10.1016/j.ceb.2012.12.002.

Duellberg C, Fourniol FJ, Maurer SP, Roostalu J, Surrey T. End-binding proteins and Ase1/PRC1 define local functionality of structurally distinct parts of the microtubule cytoskeleton. Trends Cell Biol. 2012;pii:S0962-8924(12)00181-X Trokter M, Surrey T. LIS1 clamps dynein to the microtubule. Cell. 2012;150(5):877-9

Schulze A, Harris AL. How cancer metabolism is tuned for proliferation and vulnerable to disruption. Nature. 2012;491(7424):364-73



RESEARCH PUBLICATIONS

143

Charles Swanton (page 74) Translational Cancer Therapeutics Group Primary Research Papers Chapman MH, Tidswell R, Dooley JS, Sandanayake NS, Cerec V, Deheragoda M, Lee AJ, Swanton C, Andreola F, Pereira SP. Whole genome RNA expression profiling of endoscopic biliary brushings provides data suitable for biomarker discovery in cholangiocarcinoma. J Hepatol. 2012;56(4):877-85

Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366(10):883-92 Gerlinger M, Santos CR, Spencer-Dene B, Martinez P, Endesfelder D, Burrell RA, Vetter M, Jiang M, Saunders RE, Kelly G, Rioux-Leclercq N, Stamp G, Patard JJ, Larkin J, Howell M, Swanton C. Genomewide RNA interference analysis of renal carcinoma survival regulators identifies MCT4 as a Warburg effect metabolic target. J Pathol. 2012;227(2):146-56 Lee AJ, Roylance R, Sander J, Gorman P, Endesfelder D, Kschischo M, Jones NP, East P, Nicke B, Spassieva S, Obeid LM, Juul Birkbak N, Szallasi Z, McKnight NC, Rowan AJ, Speirs V, Hanby AM, Downward J, Tooze SA, Swanton C. CERT depletion predicts chemotherapy benefit and mediates cytotoxic and polyploid-specific cancer cell death through autophagy induction. J Pathol. 2012;226(3):482-94 Lu L, Teixeira VH, Yuan Z, Graham TA, Endesfelder D, Kolluri K, Al-Juffali N, Hamilton NJ, Nicholson A, Falzon M, Kschischo M, Swanton C, Wright NA, Carroll B, Watt FM, George J, Jensen KB, Giangreco A, Janes SM. LRIG1 regulates cadherin-dependent contact inhibition directing epithelial homeostasis and preinvasive squamous cell carcinoma development. J Pathol. 2012;doi:10.1002/path.4148 Natrajan R, Mackay A, Lambros MB, Weigelt B, Wilkerson PM, Manie E, Grigoriadis A, A’hern R, van der Groep P, Kozarewa I, Popova T, Mariani O, Turaljic S, Furney SJ, Marais R, Rodruigues DN, Flora AC, Wai P, Pawar V, McDade S, Carroll J, Stoppa-Lyonnet D, Green AR, Ellis IO, Swanton C, van Diest P, Delattre O, Lord CJ, Foulkes WD, Vincent-Salomon A, Ashworth A, Henri Stern M, Reis-Filho JS. A whole-genome massively parallel sequencing analysis of BRCA1 mutant oestrogen receptor negative and positive breast cancers. J Pathol. 2012;227(1):29-41

144

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Staples CJ, Myers KN, Beveridge RD, Patil AA, Lee AJ, Swanton C, Howell M, Boulton SJ, Collis SJ. The centriolar satellite protein Cep131 is important for genome stability. J Cell Sci. 2012;125(20):4770-9 Other Publications Fennell DA, Swanton C. Unlocking Pandora’s box: personalising cancer cell death in non-small cell lung cancer. EPMA J. 2012;3(1):6

Fisher R, Larkin J, Swanton C. Inter and intratumour heterogeneity: a barrier to individualized medical therapy in renal cell carcinoma? Front Oncol. 2012;2:49 Larkin J, Goh XY, Vetter M, Pickering L, Swanton C. Histone and chromatin regulation in renal cell carcinoma: opportunities for therapeutic intervention? Nature Reviews Urology. 2012;9(3):147-55 Lee AJ, Swanton C. Tumour heterogeneity and drug resistance: personalising cancer medicine through functional genomics. Biochem Pharmacol. 2012;83(8):1013-20 McGranahan N, Burrell RA, Endesfelder D, Novelli MR, Swanton C. Cancer chromosomal instability: therapeutic and diagnostic challenges. EMBO Rep. 2012;13(6):528-38 Swanton, C. Intratumor heterogeneity: evolution through space and time. Cancer Res. 2012;72(19):4875-82 Swanton C. Plasma-derived tumor DNA analysis at whole-genome resolution. Clin Chem. 2012;doi:10.1373/clinchem.2012.197053 Weigelt B, Reis-Filho JS, Swanton C. Genomic analyses to select patients for adjuvant chemotherapy: trials and tribulations. Ann Oncol. 2012;23(10):211-8 Yap TA, Swanton C, de Bono JS. Personalization of prostate cancer prevention and therapy: are clinically qualified biomarkers in the horizon? EPMA J. 2012;3(1):3 Yap TA, Gerlinger M, Futreal PA, Pusztai L, Swanton C. Intratumor heterogeneity: seeing the wood for the trees. Sci Transl Med. 2012;4(127):127ps10

Nicolas Tapon (page 76) Apoptosis and Proliferation Control Group Wehr MC, Holder MV, Gailite I, Saunders RE, Maile TM, Ciirdaeva E, Instrell R, Jiang M, Howell M, Rossner MJ, Tapon N. Salt-inducible kinases regulate growth through the Hippo signalling pathway in Drosophila. Nat Cell Biol. 2012;doi:10.1038/ncb2658

Other Publications Tapon N, Harvey K. The Hippo pathway-From top to bottom and everything in between. Semin Cell Dev Biol. 2012;23(7):768-9

Barry Thompson (page 78) Epithelial Biology Group Primary Research Papers Fletcher GC, Lucas EP, Brain R, Tournier A, Thompson BJ. Positive feedback and mutual antagonism combine to polarize crumbs in the Drosophila follicle cell epithelium. Curr Biol. 2012;22(12):1116-22

Ahrens S, Zelenay S, Sancho D, Hanč P, Kjær S, Feest C, Fletcher G, Durkin C, Postigo A, Skehel B, Batista FD, Thompson B, Way M, Reis e Sousa C, Schulz O. F-actin is an evolutionarily-conserved damage-associated molecular pattern recognized by DNGR-1, a receptor for dead cells. Immunity. 2012;36(4):635-45 Other Publications Thompson B. The delicate choreography in growing epithelia. J Cell Biol. 2012;198(3):268-9

Thompson BJ. Cell polarity: models and mechanisms from yeast, worms and flies. Development. 2012;doi:10.1242/dev.083634 Thompson BJ, Perez F, Vaccari T. The young and happy marriage of membrane traffic and cell polarity. EMBO Rep. 2012;13(8):670-2

Takashi Toda (page 80) Cell Regulation Group Primary Research Papers Penney M, Samejima I, Wilkinson CR, McInerny CJ, Mathiassen SG, Wallace M, Toda T, Hartmann-Petersen R, Gordon C. Fission yeast 26S proteasome mutants are multi-drug resistant due to stabilization of the Pap1 transcription factor. PLoS One. 2012;7(11):e50796

Jourdain I, Dooley HC, Toda T. Fission yeast Sec3 bridges the exocyst complex to the actin cytoskeleton. Traffic. 2012;13(11):1481-95 Okamoto SY, Sato M, Toda T, Yamamoto M. SCF ensures meiotic chromosome segregation through a resolution of meiotic recombination intermediates. PLoS One. 2012;7(1):e30622

Drosophila melanogaster ovary surrounded by a smooth muscle sheet. The organ was stained with an extracellular domain of DNGR-1, which highlights the F-actin in muscle striations (red). The nuclei of nurse, follicle epithelial and muscle cells are stained green.

Sharon Tooze (page 82) Secretory Pathways Group Primary Research Papers Kraft C, Kijanska M, Kalie E, Siergiejuk E, Lee SS, Semplicio G, Stoffel I, Brezovich A, Verma M, Hansmann I, Ammerer G, Hofmann K, Tooze S, Peter M. Binding of the Atg1/ULK1 kinase to the ubiquitin-like protein Atg8 regulates autophagy. EMBO J. 2012;31(18):3691-703

Lee AJ, Roylance R, Sander J, Gorman P, Endesfelder D, Kschischo M, Jones NP, East P, Nicke B, Spassieva S, Obeid LM, Juul Birkbak N, Szallasi Z, McKnight NC, Rowan AJ, Speirs V, Hanby AM, Downward J, Tooze SA, Swanton C. CERT depletion predicts chemotherapy benefit and mediates cytotoxic and polyploid-specific cancer cell death through autophagy induction. J Pathol. 2012;226(3):482-94 Longatti A, Lamb CA, Razi M, Yoshimura S, Barr FA, Tooze SA. TBC1D14 regulates autophagosome formation via Rab11- and ULK1-positive recycling endosomes. J Cell Biol. 2012;197(5):659-75 McKnight NC, Jefferies HB, Alemu EA, Saunders RE, Howell M, Johansen T, Tooze SA. Genome-wide siRNA screen reveals amino acid starvation-induced autophagy requires SCOC and WAC . EMBO J. 2012;31(8):1931-46 Orsi A, Razi M, Robinson D, Weston AE, Collinson LM, Dooley H, Tooze, SA. Dynamic and transient interactions of Atg9 with autophagosomes, but not membrane integration, are required for autophagy. Mol Biol Cell. 2012;23(10):1860-73 Stanley P, Tooze S, Hogg N. A role for Rap2 in recycling of the extended conformation LFA-1 and T cell function and T cell function. Biol Open. 2012;1(11):1161-8



RESEARCH PUBLICATIONS

145

Other Publications Petronczki M, Uhlmann F. Cell Biology. ESCRTing DNA at the cleavage site during cytokinesis. Science. 2012;336(6078):166-7

Thadani R, Uhlmann F, Heeger S. Condensin, Chromatin Crossbarring and Chromosome Condensation. Curr Biol. 2012;22:R1012-R1021

Helen Walden (page 88) Protein Structure Function Group Wing imaginal discs carrying double FLP/FRT clones (GFP in green, mCherry in red, anti-Expanded in blue).

Other Publications Joachim J, Wirth M, McKnight NC, Tooze SA. Coiling up with SCOC and WAC: Two new regulators of starvation-induced autophagy. Autophagy. 2012;8(9): 1397-40

Kalie E, Tooze SA. Tales of the autophagy crusaders. EMBO Rep. 2012;13(3):175-7 Klionsky DJ, ….. Tooze, SA. et al., Guidelines for the use and interpretation of assays for monitoring autophagy. Autophagy. 2012;8(4):445-544 Longatti A, Tooze SA. Recycling endosomes contribute to autophagosome formation. Autophagy. 2012;8(11) Reggiori F, Tooze SA. Autophagy regulation through Atg9 traffic. J Cell Biol. 2012;198(2):151-3

Richard Treisman (page 84) Signalling and Transcription Group Primary Research Papers Mouilleron S, Wiezlak M, O’Reilly N, Treisman R, McDonald NQ. Structures of the Phactr1 RPEL domain and RPEL motif complexes with G-actin reveal the molecular basis for actin binding cooperativity. Structure. 2012;20(11):1960-70

Wiezlak M, Diring J, Abella JV, Mouilleron S, Way M, McDonald NQ, Treisman R. G-actin regulates shuttling and PP1 binding by the RPEL protein Phactr1 to control actomyosin assembly. J Cell Sci. 2012;doi:10.1242/jcs.112078

Frank Uhlmann (page 86) Chromosome Segregation Group Primary Research Papers O’Reilly N, Charbin A, Lopez-Serra L, Uhlmann F. Facile synthesis of budding yeast a-factor and its use to synchronize cells of a mating type. Yeast. 2012;29(6):233-40

146

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Primary Research Papers Burchell L, Chaugule VK, Walden H. Small, N-terminal tags activate Parkin E3 ligase activity by disrupting its autoinhibited conformation. PLoS One. 2012;7(4):e34748

Kondapalli C, Kazlauskaite A, Zhang N, Woodroof HI, Campbell DG, Gourlay R, Burchell L, Walden H, Macartney TJ, Deak M, Knebel A, Alessi DR, Muqit MMK. PINK1 is activated by mitochondrial membrane potential depolarization and stimulates Parkin E3 ligase activity by phosphorylating Serine 65. Open Biology. 2012;2(5):120080 Other Publications Hodson C, Walden H. Towards a molecular understanding of the Fanconi Anemia core complex. Anemia. 2012;2012:926787

Walden H, Martinez-Torres RJ. Regulation of Parkin E3 ubiquitin ligase activity. Cell Mol Life Sci. 2012;69(18):3053-67

Michael Way (page 90) Cell Motility Group Primary Research Papers Ahrens S, Zelenay S, Sancho D, Hanč P, Kjær S, Feest C, Fletcher G, Durkin C, Postigo A, Skehel B, Batista FD, Thompson B, Way M, Reis e Sousa C, Schulz O. F-actin is an evolutionarily-conserved damage-associated molecular pattern recognized by DNGR-1, a receptor for dead cells. Immunity. 2012;36(4):635-45

Dart AE, Donnelly SK, Holden DW, Way M, Caron E. Nck and Cdc42 co-operate to recruit N-WASP to promote FcγR-mediated phagocytosis. J Cell Science. 2012;125:2825-30 Humphries AC, Dodding MP, Barry DJ, Collinson LM, Durkin CH, Way M. Clathrin potentiates vacciniainduced actin polymerization to facilitate viral spread. Cell Host and Microbe. 2012;12:346-59 Lynn H, Horsington J, Ter LK, Han S, Chew YL, Diefenbach RJ, Way M, Chaudhri G, Karupiah G, Newsome TP. Loss of cytoskeletal transport

during egress critically attenuates Ectromelia virus infection in vivo. J Virol. 2012;86:7427-7443 Postigo A, Way M. The vaccinia virus-encoded Bcl-2 homologues do not act as direct Bax inhibitors. J Virol. 2012;86(1):203-13 Wiezlak M, Diring J, Abella JV, Mouilleron S, Way M, McDonald NQ, Treisman R. G-actin regulates shuttling and PP1 binding by the RPEL protein Phactr1 to control actomyosin assembly. J Cell Sci. 2012;125:5860-5872 Other Publications Way M. A fresh start-but business as usual. J Cell Sci. 2012;125:1-2

Simon Boulton (page 94) DNA Damage Response Group Primary Research Papers Chapman JR, Sossick AJ, Boulton SJ, Jackson SP. BRCA1-associated exclusion of 53BP1 from DNA damage sites underlies temporal control of DNA repair. J Cell Sci. 2012;125(Pt 15):3529-34

Gari K, Leon Ortiz AM, Borel-Vannier V, Flynn H, Skehel JM, Boulton SJ. MMS19 links cytoplasmic FeS cluster assembly to DNA metabolism. Science. 2012;337:243-5 Moldovan G, Dejsuphong D, Petalcorin MIR, Hofmann K, Takeda S, Boulton SJ, D’Andrea A. Inhibition of homologous recombination by the PCNA-interacting protein PARI. Mol Cell. 2012;45: 75-86 Staples CJ, Myers KN, Beveridge RD, Patil AA, Lee AJ, Swanton C, Howell M, Boulton SJ, Collis SJ. The centriolar satellite protein Cep131 is important for genome stability. J Cell Sci. 2012;125(20):4770-9 Vannier J-B, Petalcorin MIR, Pavicic-Kaltenbrunner V, Ding H, Boulton SJ. RTEL1 dismantles T-loops and counteracts telomeric G4-DNA to maintain telomere integrity. Cell. 2012;149:795-806 Other Publications Chapman JR, Taylor MRG, Boulton SJ. Playing the end game: regulating the fate of DNA double strand breaks. Mol Cell. 2012;47(4):497-510

Hare S, Maertens GN, Cherepanov, P. 3ʹ-Processing and strand transfer catalysed by retroviral integrase in crystallo. EMBO J. 2012;31(13):3020-8 Hughes S, Elustondo F, Di Fonzo A, Leroux FG, Wong AC, Snijders AP, Matthews SJ, Cherepanov P. Crystal structure of human CDC7 kinase in complex with its activator DBF4. Nat Struct Mol Biol. 2012;19(11):1101-7 Lekomtsev S, Su KC, Pye VE, Blight K, Sundaramoorthy S, Takaki T, Collinson LM, Cherepanov P, Divecha N, Petronczki M. Centralspindlin links the mitotic spindle to the plasma membrane during cytokinesis. Nature. 2012;492(7428):276-9 Métifiot M, Maddali K, Johnson BC, Hare S, Smith SJ, Zhao XZ, Marchand C, Burke TR, Hughes SH, Cherepanov P, Pommier Y. Activities, crystal structures and molecular dynamics of dihydro-1Hisoindole derivatives, inhibitors of HIV-1 integrase. ACS Chem Biol. 2012;doi:10.1021/cb300471n Wang H, Jurado KA, Wu X, Shun MC, Li X, Ferris AL, Smith SJ, Patel PA, Fuchs JR, Cherepanov P, Kvaratskhelia M, Hughes SH, Engelman A. HRP2 determines the efficiency and specificity of HIV-1 integration in LEDGF/p75 knockout cells but does not contribute to the antiviral activity of a potent LEDGF/ p75-binding site integrase inhibitor. Nucleic Acids Res. 2012;40(22):11518-30 Other Publications Engelman A, Cherepanov P. The structural biology of HIV-1: mechanistic and therapeutic insights. Nat Rev Microbiol. 2012;10:279-290

Alessandro Costa (page 98) Architecture and Dynamics of Macromolecular Machines Group Primary Research Papers Lyubimov AY, Costa A, Bleichert F, Botchan MR, Berger JM. ATP-dependent conformational dynamics underlie the functional asymmetry of the replicative helicase from a minimalist eukaryote. Proc Natl Acad Sci USA. 2012;109(30):11999-2004

Vincenzo Costanzo (page 100) DNA Damage and Genomic Stability Group

Peter Cherepanov (page 96) Chromatin Structure and DNA Group Primary Research Papers Gupta K, Curtis JE, Krueger S, Hwang Y, Cherepanov P, Bushman FD, Van Duyne GD. Solution conformations of prototype foamy iirus Integrase and its stable synaptic complex with U5 viral DNA. Structure. 2012;20(11):1918-28



Primary Research Papers Lazzaro F, Novarina D, Amara F, Watt DL, Stone JE, Costanzo V, Burgers PM, Kunkel TA, Plevani P, Muzi-Falconi M. RNase H and postreplication repair protect cells from ribonucleotides incorporated in DNA. Mol Cell. 2012;45(1):99-110

RESEARCH PUBLICATIONS

147

Ray Chaudhuri A, Hashimoto Y, Herrador R, Neelsen KJ, Fachinetti D, Bermejo R, Cocito A, Costanzo V, Lopes M. Topoisomerase I poisoning results in PARP-mediated replication fork reversal. Nat Struct Mol Biol. 2012;19(4):417-23 Zhao Z, Oh S, Li D, Ni D, Pirooz SD, Lee JH, Yang S, Lee JY, Ghozalli I, Costanzo V, Stark JM, Liang CA. A dual role for UVRAG in maintaining chromosomal stability independent of autophagy. Dev Cell. 2012;22(5):1001-16 Other Publications Errico A, Costanzo V. Mechanisms of replication fork protection: a safeguard for genome stability. Crit Rev Biochem Mol Biol. 2012;47(3):222-35

John Diffley (page 102) Chromosome Replication Group Primary Research Papers Mehanna A, Diffley JF. Pre-replicative complex assembly with purified proteins. Methods. 2012;57(2):222-6 Other Publications Boos D, Frigola J, Diffley JFX. Activation of the replicative DNA helicase: breaking up is hard to do. Curr Opin Cell Biol. 2012;24(3):423-30

Douglas ME, Diffley JF. Replication timing: the early bird catches the worm. Curr Biol. 2012;22(3):R81-2

Peter Karran (page 104) Mammalian DNA Repair Group Primary Research Papers Brem R, Karran P. Multiple forms of DNA damage caused by UVA photoactivation of DNA 6-thioguanine. Photochem Photobiol. 2012;88(1):5-13

Brem R, Karran P. Oxidation-mediated DNA crosslinking contributes to the toxicity of 6-thioguanine in human cells. Cancer Res. 2012;72(18):4787-95

Hofbauer GF, Attard NR, Harwood CA, McGregor JM, Dziunycz P, Iotzova-Weiss G, Straub G, Meyer R, Kamenisch Y, Berneburg M, French LE, Wüthrich RP, Karran P, Serra AL. Reversal of UVA skin photosensitivity and DNA damage in kidney transplant recipients by replacing azathioprine. Am J Transplant. 2012;12(1):218-25 Nakao S, Zhang S, Vaara M, Syväoja JE, Lee MY, Tsurimoto T, Karran P, Oda S. Efficient long DNA gap-filling in a mammalian cell-free system: A potential new in vitro DNA replication assay. Biochimie. 2012;doi:10.1016/j.biochi.2012.09.031 Reelfs O, Karran P, Young AR. 4-thiothymidine sensitization of DNA to UVA offers potential for a novel photochemotherapy. Photochem Photobiol Sci. 2012;11(1):148-54 Other Publications Attard NR, Karran P. UVA photosensitization of thiopurines and skin cancer in organ transplant recipients. Photochem Photobiol Sci. 2012;11(1):62-8

Mark Petronczki (page 106) Cell Division and Aneuploidy Group Primary Research Papers Lekomtsev S, Su KC, Pye VE, Blight K, Sundaramoorthy S, Takaki T, Collinson LM, Cherepanov P, Divecha N, Petronczki M. Centralspindlin links the mitotic spindle to the plasma membrane during cytokinesis. Nature. 2012;492(7428):276-9 Other Publications Petronczki M, Uhlmann F. Cell Biology. ESCRTing DNA at the cleavage site during cytokinesis. Science. 2012;336(6078):166-7

Jesper Svejstrup (page 108) Mechanisms of Gene Transcription Group Primary Research Papers Close P, East P, Dirac-Svejstrup AB, Hartmann H, Heron M, Maslen S, Chariot A, Söding J, Skehel M, Svejstrup JQ. DBIRD complex integrates alternative mRNA splicing with RNA polymerase II transcript elongation. Nature. 2012;484:386-9

Close P, Gillard M, Ladang A, Jiang Z, Papuga J, Hawkes N, Nguyen L, Chapelle JP, Bouillenne F, Svejstrup JQ, Fillet M, Chariot A. DERP6 (ELP5) and C3ORF75 (ELP6) regulate tumorigenicity and migration of melanoma cells as subunits of Elongator. J Biol Chem. 2012;287(39):32535-45

Dividing human cells, microtubules in red and splitting sister genomes in green.

148

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Hobson, DJ, Wei W, Steinmetz LM, Svejstrup JQ. RNA Polymerase II collision interrupts convergent transcription. Mol Cell. 2012;48(3):365-74

Wilson MD, Saponaro M, Leidl MA, Svejstrup JQ. MultiDsk: A ubiquitin-specific affinity resin. PLoS One. 2012;7(10):e46398 Other Publications Ehrensberger AH, Svejstrup JQ. Reprogramming chromatin. Crit Rev Biochem Mol Biol. 2012;47(5):464-82

Svejstrup JQ. Transcription: another mark in the tail. EMBO J. 2012;31(12):2753-4 Wilson MD, Harreman H, Svejstrup JQ. Ubiquitylation and degradation of elongating RNA polymerase II: the last resort. Biochim Biophys Acta. 2012;pii: S18749399(12)00145-9

Helle Ulrich (page 110) DNA Damage Tolerance Group Primary Research Papers Parker JL, Ulrich HD. A SUMO-interacting motif activates budding yeast ubiquitin ligase Rad18 towards SUMO-modified PCNA . Nucleic Acids Res. 2012;40(22):11380-8

Saugar I, Parker JL, Zhao S, Ulrich HD. The genome maintenance factor Mgs1 is targeted to sites of replication stress by ubiquitylated PCNA. Nucleic Acids Res. 2012;40(1):245-57 Other Publications Davies AA, Ulrich HD. Detection of PCNA modifications in Saccharomyces cerevisiae. Methods Mol Biol. 2012;920:543-67

Finley D, Ulrich HD, Sommer T, Kaiser P. The ubiquitinproteasome system of Saccharomyces cerevisiae. Genetics. 2012;192(2):319-60 Parker JL, Ulrich HD. In vitro PCNA modification assays. Methods Mol Biol. 2012;920:569-89 Ulrich HD. Ubiquitin, SUMO, and phosphate: how a trio of posttranslational modifiers governs protein fate. Mol Cell. 2012;47(3):335-7 Ulrich HD. Ubiquitin and SUMO in DNA repair at a glance. J Cell Sci. 2012;125(2):249-54

Stephen West (page 112) Genetic Recombination Group Primary Research Papers Blackford AN, Schwab RA, Nieminuszczy J, Deans AJ, West SC, Niedzwiedz W. The DNA translocase activity of FANCM protects stalled replication forks. Hum Mol Genet. 2012;21(9):2005-16



Muñoz-Galván S, Tous C, Blanco MG, Schwartz EK, Ehmsen KT, West SC, Heyer WD, Aguilera A. Distinct roles of Mus81, Yen1, Slx1-Slx4, and Rad1 nucleases in the repair of replication-born double-strand breaks by sister chromatid exchange. Mol Cell Biol. 2012;32(9):1592-603

Advanced Sequencing (page 117) Nik Matthews Primary Research Papers Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366(10):883-92

Kumar MS, Hancock D, Molina-Arcas M, Steckel M, East P, Diefenbacher M, Armenteros-Monterroso E, Lassailly F, Matthews N, Nye E, Stamp G, Behrens A, Downward J. The GATA2-driven oncogene network is requisite for RAS oncogene-driven no-small cell lung cancer. Cell. 2012;149(3):642-55

Bioinformatics and Biostatistics (page 118) Aengus Stewart Primary Research Papers Archibald KM, Kulbe H, Kwong J, Chakravarty P, Temple J, Chaplin T, Flak MB, McNeish IA, Deen S, Brenton JD, Young BD, Balkwill F. Sequential genetic change at the TP53 and chemokine receptor CXCR4 locus during transformation of human ovarian surface epithelium. Oncogene. 2012;31(48):4987-95

Arthur-Farraj PJ, Latouche M, Wilton DK, Quintes S, Chabrol E, Banerjee A, Woodhoo A, Jenkins B, Rahman M, Turmaine M, Wicher GK, Mitter R, Greensmith L, Behrens A, Raivich G, Mirsky R, Jessen KR. c-Jun reprograms Schwann cells of injured nerves to generate a repair cell essential for regeneration. Neuron. 2012;75(4):633-47 Close P, East P, Dirac-Svejstrup AB, Hartmann H, Heron M, Maslen S, Chariot A, Söding J, Skehel M, Svejstrup JQ. DBIRD complex integrates alternative mRNA splicing with RNA polymerase II transcript elongation. Nature. 2012;484(7394):386-9 Ferber EC, Peck B, Delpuech O, Bell GP, East P, Schulze A. FOXO3a regulates reactive oxygen metabolism by inhibiting mitochondrial gene expression. Cell Death Differ. 2012;19(6):968-79

RESEARCH PUBLICATIONS

149

Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366(10):883-92

of synthetic lethal interactions in KRAS oncogenedependent cancer cells reveals novel therapeutic targeting strategies. Cell Res. 2012;22(8):1227-45

Gerlinger M, Santos CR, Spencer-Dene B, Martinez P, Endesfelder D, Burrell RA, Vetter M, Jiang M, Saunders RE, Kelly G, Dykema K, Rioux-Leclercq N, Stamp G, Patard JJ, Larkin J, Howell M, Swanton C. Genome-wide RNA interference analysis of renal carcinoma survival regulators identifies MCT4 as a Warburg effect metabolic target. J Pathol. 2012;227(2):146-56

Jones DT, Lechertier T, Mitter R, Herbert JM, Bicknell R, Jones JL, Li JL, Buffa F, Harris AL, Hodivala-Dilke K. Gene expression analysis in human breast cancer associated blood vessels. PLoS One. 2012;7(10):e44294

Kulbe H, Chakravarty P, Leinster DA, Charles KA, Kwong J, Thompson RG, Coward JI, Schioppa T, Robinson SC, Gallagher WM, Galletta L; Australian Ovarian Cancer Study Group, Salako MA, Smyth JF, Hagemann T, Brennan DJ, Bowtell DD, Balkwill FR. A dynamic inflammatory cytokine network in the human ovarian cancer microenvironment. Cancer Res. 2012;72(1):66-75 Kumar MS, Hancock DC, Molina-Arcas M, Steckel M, East P, Diefenbacher M, Armenteros-Monterroso E, Lassailly F, Matthews N, Nye E, Stamp G, Behrens A, Downward J. The GATA2 transcriptional network is requisite for RAS oncogene-driven non-small cell lung cancer. Cell. 2012;149(3):642-55 Lee AJ, Roylance R, Sander J, Gorman P, Endesfelder D, Kschischo M, Jones NP, East P, Nicke B, Spassieva S, Obeid LM, Juul Birkbak N, Szallasi Z, McKnight NC, Rowan AJ, Speirs V, Hanby AM, Downward J, Tooze SA, Swanton C. CERT depletion predicts chemotherapy benefit and mediates cytotoxic and polyploid-specific cancer cell death through autophagy induction. J Pathol. 2012;226(3):482-94 Lewis A, Davis H, Deheragoda M, Pollard P, Nye E, Jeffery R, Segditsas S, East P, Poulsom R, Stamp G, Wright N, Tomlinson I. The C-terminus of Apc does not influence intestinal adenoma development or progression. J Pathol. 2012;226(1):73-83 Ros S, Santos CR, Moco S, Baenke F, Kelly G, Howell M, Zamboni N, Schulze A. Functional metabolic screen identifies 6-phosphofructo-2kinase/fructose-2,6-biphosphatase 4 as an important regulator of prostate cancer cell survival. Cancer Discov. 2012;2(4):328-43 Steckel M, Molina-Arcas M, Weigelt B, Marani M, Warne PH, Kuznetsov H, Kelly G, Saunders B, Howell M, Downward J, Hancock DC. Determination

150

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Salm MP, Horswell SD, Hutchison CE, Speedy HE, Yang X, Liang L, Schadt EE, Cookson WO, Wierzbicki AS, Naoumova RP, Shoulders CC. The origin, global distribution, and functional impact of the human 8p23 inversion polymorphism. Genome Res. 2012;22(6):1144-53

Biomolecular Modelling Research (page 119) Paul Bates Primary Research Papers Cheng TM, Goehring L, Jeffery L, Lu YE, Hayles J, Novak B, Bates PA. A structural systems biology approach for quantifying the systemic consequences of missense mutations in proteins. PLoS Comp Biol. 2012;doi:10.1371/journal.pcbi.1002738

Moal IH, Bates PA. Kinetic rate constant prediction supports the conformational selection mechanism of protein binding. PLoS Compl Biol. 2012;8(1):e1002351 Torchala M, Chelminiak P, Bates PA. Mean firstpassage time calculations comparison of the deterministic Hill’s algorithm with Monte Carlo simulations. Eur Phys J B. 2012;85(4):116 Other Publications Cheng TM, Gulati S, Agius R, Bates PA. Understanding cancer mechanisms through network dynamics. Brief Funct Genomics. 2012;11(6):543-60

Electron Microscopy (page 121) Lucy Collinson Primary Research Papers Domart MC, Hobday TMC, Peddie CJ, Chung GHC, Wang A, Yeh K, Jethwa N, Zhang Q, Wakelam MJO, Woscholski R, Byrne RD, Collinson LM, Poccia DL, Larijani B (2012). Diacylglycerol controls in vivo ER morphology and nuclear envelope fusion. PLoS ONE. 2012;7:e51150

Humphries AC, Dodding MP, Barry DJ, Collinson LM, Durkin CH, Way M. Clathrin potentiates vacciniainduced actin polymerization to facilitate viral spread. Cell Host and Microbe. 2012;12:346-359

Kozik P, Hodson NA, Sahlender DA, Simecek N, Soromani C, Wu J, Collinson L, Robinson MS. A human genome-wide screen for regulators of clathrin-coated vesicle formation reveals an unexpected role for the V-ATPase. Nat Cell Biol. 2012;doi:10.1038/ncb2652 Lekomtsev S, Su K-C, Pye VE, Blight K, Sundaramoorthy S, Takaki T, Collinson LM, Cheperanov P, Divecha N, Petronczki M (2012). Centralspindlin links the mitotic spindle to the plasma membrane during cytokinesis. Nature. 2012;492(7428):276-9 Orsi A, Razi M, Robinson D, Weston AE, Collinson LM, Dooley H, Tooze, SA. Dynamic and transient interactions of Atg9 with autophagosomes, but not membrane integration, are required for autophagy. Mol Biol Cell. 2012;23(10):1860-73 Thaunat O, Granja AG, Barral P, Filby A, Montaner B, Collinson L, Martinez-Martin N, Harwood NE, Bruckbauer A, Batista FD. Asymmetric segregation of polarized antigen on B cell division shapes presentation capacity. Science. 2012;335(6067):475-9 Other Publications Bushby AJ, Mariggi G, Armer HE, Collinson LM. Correlative light and volume electron microscopy: using focused ion beam scanning electron microscopy to image transient events in model organisms. Methods Cell Biol. 2012;111:357-82

Experimental Histopathology (page 123) Gordon Stamp Primary Research Papers Cesca F, Yabe A, Spencer-Dene B, Scholz-Starke J, Medrihan L, Maden CH, Gerhardt H, Orriss IR, Baldelli P, Al-Qatari M, Koltzenburg M, Adams RH, Benfenati F, Schiavo G. Kidins220/ARMS mediates the integration of the neurotrophin and VEGF pathways in cardiovascular and nervous system development. Cell Death Differ. 2012;19(2):194-208

Chaudhry SI, Hooper S, Nye E, Williamson P, Harrington K, Sahai E. Autocrine IL-1b-TRAF6 signalling promotes squamous cell carcinoma invasion through paracrine TNFa signalling to carcinomaassociated fibroblasts. Oncogene. 2012;doi:10.1038/ onc.2012.91 Fontana X, Hristova M, Da Costa C, Patodia S, Thei L, Makwana M, Spencer-Dene B, Latouche M, Mirsky R, Jessen KR, Klein R, Raivich G, Behrens A. c-Jun in Schwann cells promotes axonal regeneration and motoneuron survival via paracrine signaling. J Cell Biol. 2012;198(1):127-41



Gerlinger M, Santos CR, Spencer-Dene B, Martinez P, Endesfelder D, Burrell RA, Vetter M, Jiang M, Saunders RE, Kelly G, Rioux-Leclercq N, Stamp G, Patard JJ, Larkin J, Howell M, Swanton C. Genomewide RNA interference analysis of renal carcinoma survival regulators identifies MCT4 as a Warburg effect metabolic target. J Pathol. 2012;227(2):146-56 Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366(10):883-92 Kumar MS, Hancock D, Molina-Arcas M, Steckel M, East P, Diefenbacher M, Armenteros-Monterroso E, Lassailly F, Matthews N, Nye E, Stamp G, Behrens A, Downward J. The GATA2-driven oncogene network is requisite for RAS oncogene-driven no-small cell lung cancer. Cell. 2012;149(3):642-55 Lewis A, Davis H, Deheragoda M, Pollard P, Nye E, Jeffery R, Segditsas S, East P, Poulsom R, Stamp G, Wright N, Tomlinson I. The C-terminus of Apc does not influence intestinal adenoma development or progression. J Pathol. 2012;226(1):73-83 Patalay R, Talbot C, Alexandrov Y, Lenz MO, Kumar S, Warren S, Munro I, Neil MA, König K, French PM, Chu A, Stamp GW, Dunsby C. Multiphoton multispectral fluorescence lifetime tomography for the evaluation of Basal cell carcinomas. PLoS One. 2012;7(9):e43460 Shearman A, Stamp G, Tekkis P, Tan E. Pan-enteric diaphragm disease. Colorectal Dis. 2012;doi: 10.1111/j.1463-1318.2012.03102.x Valorani MG, Montelatici E, Germani A, Biddle A, D’Alessandro D, Strollo R, Patrizi MP, Lazzari L, Nye E, Otto WR, Pozzilli P, Alison MR. Pre-culturing human adipose tissue mesenchymal stem cells under hypoxia increases their adipogenic and osteogenic differentiation potentials. Cell Prolif. 2012;45(3):225-38

Fluorescence-Activated Cell Sorting (page 124) Derek Davies Primary Research Papers Abdullah Thani NA, Sallis B, Nuttall R, Schubert FR, Ahsan M, Davies D, Purewal S, Cooper A, Rooprai HK. Induction of apoptosis and reduction of MMP gene expression in the U373 cell line by polyphenolics in Aronia melanocarpa and by curcumin. Oncol Rep. 2012;28(4):1435-42

RESEARCH PUBLICATIONS

151

Ros S, Santos CR, Moco S, Baenke F, Kelly G, Howell M, Zamboni N, Schulze A. Functional metabolic screen identifies 6-phosphofructo-2kinase/fructose-2,6-biphosphatase 4 as an important regulator of prostate cancer cell survival. Cancer Discov. 2012;2(4):328-43 Rosse C, Boeckeler K, Linch M, Radtke S, Frith D, Barnouin K, Morsi AS, Hafezparast M, Howell M, Parker PJ. Binding of dynein intermediate chain 2 to Paxillin controls focal adhesion dynamics and migration. J Cell Sci. 2012;125:3733-8 Labelling of endothelial single cells with membrane-targeted GFP. Tip cell in a whole-mount P5 mTmGPdgfbiCre retina stained for isolectin B4.

Filby A, Davies D. Reporting imaging flow cytometry data for publication: Why mask the detail? Cytometry A. 2012;81(8):637-42 Mallucci L, Shi DY, Davies D, Jordan P, Nicol A, Lotti L, Mariani-Costantini R, Verginelli F, Wells V, Zicha D. Killing of Kras mutant colon cancer cells via Racindependent actin remodeling by the bGBP cytokine a physiological PI3K inhibitor therapeutically effective in vivo. Mol Cancer Ther. 2012;11(9):1884-93 Thaunat O, Granja AG, Barral P, Filby A, Montaner B, Collinson L, Martinez-Martin N, Harwood NE, Bruckbauer A, Batista FD. Asymmetric segregation of polarized antigen on B cell division shapes presentation capacity. Science. 2012;335(6067):475-9 Other Publications Davies D. Cell separations by flow cytometry. Methods Mol Biol. 2012;878:185-99

Siebring-van Olst E, Vermeulen C, de Menezes RX, Howell M, Smit EF, van Beusechem VW. Affordable Luciferase Reporter Assay for Cell-Based HighThroughput Screening. J Biomol Screen. 2012;doi:10.1177/1087057112465184 Staples CJ, Myers KN, Beveridge RD, Patil AA, Lee AJ, Swanton C, Howell M, Boulton SJ, Collis SJ. The centriolar satellite protein Cep131 is important for genome stability. J Cell Sci. 2012; 125(20):4770-9 Steckel M, Molina-Arcas M, Weigelt B, Marani M, Warne PH, Kuznetsov H, Kelly G, Saunders B, Howell M, Downward J, Hancock DC. Determination of synthetic lethal interactions in KRAS oncogenedependent cancer cells reveals novel therapeutic targeting strategies. Cell Res. 2012;22(8):1227-45 Wehr MC, Holder MV, Gailite I, Saunders RE, Maile TM, Ciirdaeva E, Instrell R, Jiang M, Howell M, Rossner MJ, Tapon N. Salt-inducible kinases regulate growth through the Hippo signalling pathway in Drosophila. Nat Cell Biol. 2012;doi:10.1038/ncb2658

Light Microscopy (page 126) High Throughput Screening (page 125)

Daniel Zicha

Michael Howell

Primary Research Papers Mallucci L, Shi DY, Davies D, Jordan P, Nicol A, Lotti L, Mariani-Costantini R, Verginelli F, Wells V, Zicha D. Killing of Kras mutant colon cancer cells via Racindependent actin remodeling by the bGBP cytokine a physiological PI3K inhibitor therapeutically effective in vivo. Mol Cancer Ther. 2012;11(9):1884-93

Primary Research Papers Gerlinger M, Santos CR, Spencer-Dene B, Martinez P, Endesfelder D, Burrell RA, Vetter M, Jiang M, Saunders RE, Kelly G, Rioux-Leclercq N, Stamp G, Patard JJ, Larkin J, Howell M, Swanton C. Genomewide RNA interference analysis of renal carcinoma survival regulators identifies MCT4 as a Warburg effect metabolic target. J Pathol. 2012;227(2):146-56

McKnight NC, Jefferies HB, Alemu EA, Saunders RE, Howell M, Johansen T, Tooze SA. Genome-wide siRNA screen reveals amino acid starvation-induced autophagy requires SCOC and WAC. EMBO J. 2012;31(8):1931-46 Milanovic M, Radtke S, Peel N, Howell M, Carrière V, Joffre C, Kermorgant S, Parker PJ. Anomalous inhibition of c-Met by the kinesin inhibitor aurintricarboxylic acid. Int J Cancer. 2012;130(5):1060-70

152

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Peptide Synthesis (page 127) Nicola O’Reilly Primary Research Papers Mouilleron S, Wiezlak M, O’Reilly N, Treisman R, McDonald NQ. Structures of the Phactr1 RPEL domain and RPEL motif complexes with G-actin reveal the molecular basis for actin binding cooperativity. Structure. 2012;20(11):1960-70

Mouilleron S, Wiezlak M, O’Reilly N, Treisman R, McDonald NQ. Structures of the Phactr1 RPEL domain and RPEL motif complexes with G-actin reveal the molecular basis for actin binding cooperativity. Structure. 2012;20(11):1960-70 O’Reilly N, Charbin A, Lopez-Serra L, Uhlmann F. Facile synthesis of budding yeast a-factor and its use to synchronize cells of a mating type. Yeast. 2012;29(6):233-40

Protein Structure (page 132) Stephane Mouilleron Primary Research Papers Bowles M, Lally J, Fadden AJ, Mouilleron S, Hammonds T, McDonald NQ. Fluorescence-based incision assay for human XPF-ERCC1 activity identifies important elements of DNA junction recognition. Nucleic Acids Res. 2012;40(13):e101

Bram Snijders

Mouilleron S, Wiezlak M, O’Reilly N, Treisman R, McDonald NQ. Structures of the Phactr1 RPEL domain and RPEL motif complexes with G-actin reveal the molecular basis for actin binding cooperativity. Structure. 2012;20(11):1960-70

Primary Research Papers Gari K, Leon Ortiz AM, Borel-Vannier V, Flynn H, Skehel JM, Boulton SJ. MMS19 links cytoplasmic FeS cluster assembly to DNA metabolism. Science. 2012;337:243-5

Wiezlak M, Diring J, Abella JV, Mouilleron S, Way M, McDonald NQ, Treisman R. G-actin regulates shuttling and PP1 binding by the RPEL protein Phactr1 to control actomyosin assembly. J Cell Sci. 2012;doi:10.1242/jcs.112078

Protein Analysis and Proteomics (page 128)

Hughes S, Elustondo F, Di Fonzo A, Leroux FG, Wong AC, Snijders AP, Matthews SJ, Cherepanov P. Crystal structure of human CDC7 kinase in complex with its activator DBF4. Nat Struct Mol Biol. 2012;19(11):1101-7

Protein Purification (page 129) Svend Kjaer Primary Research Papers Ahrens S, Zelenay S, Sancho D, Hanč P, Kjær S, Feest C, Fletcher G, Durkin C, Postigo A, Skehel B, Batista FD, Thompson B, Way M, Reis e Sousa C, Schulz O. F-actin is an evolutionarily-conserved damage-associated molecular pattern recognized by DNGR-1, a receptor for dead cells. Immunity. 2012;36(4):635-45

In vivo Imaging (page 131)

Other Publications

Nancy Hogg

Acharyya S, Oskarsson T, Vanharanta S, Kim J, Morris PG, Manova-Todorova K, Leversha M, Hogg N, Norton L, Brogi E, Massagué J. A CXCL1 paracrine network links cancer chemoresistance and metastasis. Cell. 2012;150:165-178 Stanley P, Tooze S, Hogg N. A role for Rap2 in recycling of the extended conformation LFA-1 and T cell function and T cell function. Biol Open. 2012;1(11):1161-8 Svensson L, Stanley P, Willenbrock F, Hogg N. The Gaq/11 proteins contribute to T lymphocyte migration by promoting turnover of integrin LFA-1 through recycling. PLoS One. 2012;7(6):e38517

Francois Lassailly

Svend Petersen-Mahrt

Primary Research Papers Kumar MS, Hancock DC, Molina-Arcas M, Steckel M, East P, Diefenbacher M, Armenteros-Monterroso E, Lassailly F, Matthews N, Nye E, Stamp G, Behrens A, Downward J. The GATA2 transcriptional network is requisite for RAS oncogene-driven non-small cell lung cancer. Cell. 2012;149(3):642-55

Rangam G, Schmitz KM, Cobb AJA, Petersen-Mahrt SK. AID enzymatic activity is inversely proportional to the size of cytosine C5 orbital cloud. PLoS One. 2012;7:e43279

Mathematical Modelling (page 132) Alexander Tournier

Willmann KL, Milosevic S, Pauklin S, Schmitz KM, Rangam G, Simon MT, Maslen S, Skehel M, Robert I, Heyer V, Schiavo E, Reina-San-Martin B, Petersen-Mahrt SK. A role for the RNA pol IIassociated PAF complex in AID-induced immune diversification. J Exp Med. 2012;209(11):2099-111

Fletcher GC, Lucas EP, Brain R, Tournier A, Thompson BJ. Positive feedback and mutual antagonism combine to polarize crumbs in the Drosophila follicle cell epithelium. Current Biology. 2012;22(12):1116-22



RESEARCH PUBLICATIONS

153

THESES

Franziska Baenke Gene Expression Analysis Metabolic dependencies of breast cancer cells

Hollie Chandler

David Hobson

Delphine Goubau Immunobiology Group The innate immune response to viruses: a look into cytosolic nucleic acid sensing

Vanessa Borges Chromosome Segregation Group Establishment of sister chromatid cohesion during DNA replication in Saccharomyces cerevisiae

David Hobson Mechanisms of Gene Transcription Group RNA Polymerase II collision and its role in transcript elongation

Rebecca Burrell Translational Cancer Therapeutics Group Replication stress links structural and numerical chromosomal instability in colorectal cancer

Cerys Manning Tumour Cell Biology Group Heterogeneity in melanoma and the microenvironment

Hollie Chandler Molecular Oncology Group Association of CK2 with Polycomb complexes and its functional implications

Risa Mori Cell Regulation Group The Tubulin folding pathway: Rolse of cofactor C/ Tbc 1 and small GTPase Arl2/Alp41

Adrian Charbin Chromosome Segregation Group The role of condensin in chromosome resolution

Kay Penicud Mammalian Genetics Group The role of ATM regulatory proteins in DNA damage signalling

Avradip Chatterjee Macromolecular Structure Function Group Structural and biochemical studies of cohesin regulation by Wpl1 Kuan-Chung Su

Catherine Cowell Signal Transduction Group Novel mechanisms of resistance to EGFR tyrosine kinase inhibitors in non-small cell lung cancer Sara Donnelly Cell Motility Group Molecular Dissection of Nick:WIP:N-WASP signalling networks

Rachael Shaw Apoptosis and Proliferation Control Group Regulation of Drosophila intestinal regeneration by the Hippo pathway Kuan-Chung Su Cell Division and Aneuploidy Group Control of cleavage furrow formation during cytokinesis in human cells Melda Tozluoglu Biomolecular Modelling Multiscale Modelling of Cancer Cell Motility

Risa Mori

Charlotte Durkin Cell Motility Group Vaccinia virus as a tool to study novel mechanisms of host cell contraction and blebbing Kerry Goodman Structural Biology Group RET tyrosine kinase architecture, protein interactions and chemical inhibition Rachael Shaw

154

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Yuan Zhao Telomere Biology Group Pot1 phosphorylation regulates telomere function

INSTITUTE INFORMATION

ADMINISTRATION ACADEMIC PROGRAMME SEMINARS AND CONFERENCES EXTERNAL FUNDING INSTITUTE MANAGEMENT



INSTITUTE INFORMATION

155

ADMINISTRATION

Director of Operations Ava Yeo PhD Administration Team Charis Ashton David Bacon Lucy Davinson Caroline Doran Sabina Ebbols* Nicola Hawkes PhD David Hudson PhD Jane Kirk PhD Sally Leevers PhD Sophie Lutter PhD Mubanga Mwelwa* Athena Nicolaou* Emma Rainbow Kim Rowan Michelle Trowsdale* Tom Wallace Finance Team Catherine Cunningham PhD Betty Chen Azizur Rahman Grants Team Holly Elphinstone Kaya Chatterji HR Team Ruth Attenborough* Heather Campbell Emma Collins* Laboratory Management Team Nigel Peat Mark Johnson Fiona Johnson Elizabeth Li Hans Nicolai

LRI administration The LRI Administration team provides the Director with the administrative infrastructure and support to ensure the smooth running of the Institute. The team led by the LRI Director of Operations is responsible for the academic infrastructure through the administration of the academic committees for students and postdocs, management of LRI Technology Core Facilities, IT, finance, co-ordinating Institute wide initiatives and providing general administrative support to the Research Laboratories. During the year, the LRI has become responsible for running the Cancer Research UK library. Graduate Student Administration This year 18 graduate students joined the 4-year LRI PhD Programme. The training programme provides continual support for the students throughout their studentship at the LRI. Specialised Training Programme The LRI Software Tutor, David Bacon, runs specially designed courses to enable students to use commercially available software efficiently and effectively in the course of their research. Classes are held regularly at Clare Hall and Lincoln’s Inn Fields and include Adobe Illustrator, PowerPoint, Endnote, Word etc. Targeted support is given to students as they write up their thesis in their final year. Students also receive guidance and training on maintaining image integrity when using Photoshop editing software. All students attend a special internal seminar where the Institute’s Image Processing Guidelines are outlined to which all students and staff have to adhere to. The seminar introduces the concept of image integrity, and highlights the importance of correct processing techniques for digital editing. This year, our Software Tutor was invited by the European Institute of Oncology (IFOM), in Milan, to give a seminar on image integrity, where he outlined the LRI guidelines. Postdoctoral Fellows Administration Postdocs at the LRI continue to be supported by the LRI Postdoctoral Training Programme, developed collaboratively by postdocs, group leaders and the Academic Director. The programme supports the postdocs throughout

156

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

their time at the LRI, from induction to all the facilities and activities available, to development reviews at key stages of their training and careers advice and workshops. Postdocs attend Postdoc Consultative Meetings (PDCMs) throughout the year enabling communication between the Institute and postdocs. Postdocs are also represented on committees such as the Technology Core Facility User Groups ensuring that they contribute to Institute activities. This year, the first Francis Crick Institute Postdoc Retreat was organised by postdocs from the LRI and from the National Institute for Medical Research (NIMR). 180 postdocs attended from the LRI, NIMR, Kings College London, University College London and Imperial College London, and held at Kings College Waterloo Campus. Database Development This year we implemented a new system for recording and managing the import and export of biological materials in conjunction with the Biological Resources Unit. We continued to upgrade and implement systems to support the smooth running in various areas of the Institute’s administration. The systems supporting postdoctoral recruitment were revised and upgraded for this year’s recruitment drive. Administrative Support for Group Leaders Group leaders receive comprehensive secretarial and administrative support from research administrators. The procedures and processes are continually assessed and improvements implemented to ensure a smooth and efficient service is provided. Laboratory Management Services The Laboratory Services and Support team works closely with the research laboratories and technology core facilities at the LRI, playing an important role in support of their scientific activities. As well as looking after refurbishment projects, communal equipment and the containment facility, the team also includes the Electronics Department, Stores Team and the Fly Facility Service. The upkeep and replacement of communal equipment is an important part of our work, and ensures that the scientists are provided with the appropriate facilities.

Health and Safety Tim Budd PhD Mandy Marshall John Richmond Nicholas Tidman IT Jeremy Olsen Claire Brewer Marion Edwards Andy Foster* Jacki Goldman Simon Grierson Ellen Gyapong* Mark Henshall Mat Hillyard Andrew Jordan Chris Manser Santosh Nittala Wing Poon* Phil Spratt Mark Tomlinson* Research Administrators Jessica Adams Katherine Ames Sarah Baker* Aleksandra Banasiak* Helen Batley* Alice Birch Victoria Hill* Nicola Howes Sophia Kontakkis Jackie Martinez Aileen Nelson Mary Nicolaou Anastasia Photiou

*= Part year.

Our team liaises closely with both the Health and Safety and Property Services Departments, ensuring co-operation across all areas at the LRI. This is particularly important when setting up new research groups. Currently plans are underway for the design and refurbishment of two areas for new groups due to open in March 2013. At the Clare Hall Laboratories there have also been several refurbishment projects with a Category 2 facility opening in January 2012 and work has been carried out to expand the IT facility. In support of the Scientific Officer (SO) community we help, when required, with the running of laboratories, giving advice with equipment repair and maintenance as well as other technical support. In addition we also organise meetings for the SOs to get together and discuss both administrative issues as well as sharing scientific knowledge. Workshops are held to provide training for specialised equipment such as centrifuges. We are currently supporting the planned role out of the new ‘Purchase to Pay’ (P2P) purchasing system, which Cancer Research UK is implementing. Health and Safety The Health and Safety team at the LRI provides advice, training and support in all aspects of welfare and safety throughout the Institute, whether for scientific, administrative or maintenance work. This year the team have been involved in launching and running a new accident and incident reporting system which allows all staff to report undesirable events online. Information Technology (IT) The LRI IT team have been involved with a number of core activities. • Implementing major expansion to the new Isilon storage system, increasing capacity for science data storage • Transitioning to managing IT networks for LRI including implementing a new authentication domain for LRI, provision of an LRI specific wireless service and rollout of wireless facilities to specific labs • A major upgrade (the first for 10 years) of the job scheduling system for the HPC facilities • A major increase in the number of compute cores for the HPC compute farm, increasing capacity by over 50% • Implementing a resilient virtualised infrastructure across redundant sites New recruitment, restructuring of staff roles and reallocating resources across the range of LRI IT activities and sites have also been carried out. Jeremy Olsen, Head of IT, was appointed as Head of Core ICT at the Crick with effect from June 2012 on a secondment. Development of Crick strategy and design of systems for the commissioning of the new building at Brill Place continues and work



is being done on transition activities and the convergence of IT systems from both the founder institutes. LRI Schools Outreach 2012 has been a busy year for the London Research Institute’s outreach programme. This programme aims to engage the public, in particular school children, in our research and we encourage all staff to contribute to these activities. This year the LRI schools work experience scheme hosted students on week-long placements at Clare Hall, and for the first time at Lincoln’s Inn Fields. All students spent time in the research laboratories as well as core technology facilities to appreciate the range of research being undertaken and shown start of the art equipment. Additionally this year, four Nuffield Science Bursary places were offered to 6th Form Students across London. The successful projects were carried out in the Electron Microscopy Facility, who had two students, the High Throughput Screening Facility and the Telomere Biology group at LIF. All four students thoroughly enjoyed working alongside the LRI scientists and really valued the opportunity. The LRI organised the first 'Ask a (Nobel) Scientist' event which was held on 7th March and hosted at the Wellcome Trust. Over one hundred school pupils came to the event to pose diverse science related questions to Sir Tim Hunt, and Julie Cooper. Alok Jha, science correspondent at the Guardian invigilated the event. This year, the event will take place on 27th March. Another first was the LRI Schools Day, which was held on 14th November. The day was a great success with 50 students attending from five local schools to enjoy lectures from staff, tours of the research labs and some hands on experiments. We also attended various events to allow students aged 13-17 to discuss careers in science with our young scientists, such as the Futures Fair at the Royal College of Surgeons, the Somers Town Festival – START, engaging with the local community close to the Francis Crick Institute, and the London Regional “Big Bang”.

LRI Graduate Student demonstrating bioinformatics to pupils attending the first LRI Schools Day.

ADMINISTRATION AND MANAGEMENT

157

ACADEMIC PROGRAMME

The LRI has about 100 graduate students and nearly 200 postdoctoral fellows from all over the world carrying out research and following training programmes designed to support them during their time at the LRI and in their future careers. They excel in generating much of the LRI’s research output - their success is clearly illustrated in the research highlights and publications sections of this report.

LRI Graduate Students Every year, the LRI organises a highly competitive selection process designed to recruit outstanding graduate students passionate about carrying out research leading to a PhD. There were 1244 applications to the 2012 PhD programme, and 18 new students, including one MB-PhD student, were recruited to start their PhDs in our Lincoln’s Inn Fields and Clare Hall Laboratories in September. Approximately 80% of LRI students are funded by Cancer Research UK (CRUK) studentships, with the remaining 20% funded via scholarships from external agencies, such as Boehringer Ingelheim Fonds and A*STAR, or by grants from organisations including Breast Cancer Campaign and the European Research Council. By starting their PhDs at the same time, progressing together through the PhD programme, LRI students maintain close links with other students in the Institute, particularly within their own year group, providing a strong peer-support network. This network is fostered from the beginning of the PhD programme, when students attend a three day induction during their first week. This induction serves two important purposes: to prepare the students for undertaking their PhD at the LRI, and to enable them to get to know each other and other key members of staff. To this end, interactive sessions are scheduled to introduce new students to the extensive core facilities available, as well as sessions on designing experiments, working in the laboratory and participating in journal clubs. Students are registered at University College London, and have access to training and facilities there, in addition to in-house training provided by the LRI. Alongside supervision within the laboratory, students also get scientific advice from their thesis committees,

158

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

with whom they meet at key points throughout their PhD. Further support is available from Sally Leevers, the Academic Director, who oversees the academic and administrative aspects of the graduate programme, Sabina Ebbols and Sophie Lutter, the Research Managers for Graduate Studies, and Graduate Student Advisors – group leaders who provide oversight and advice on the student Programme. As part of the PhD programme, students receive extensive presentation skills training, and are asked to present their work at key milestones throughout their PhD. This starts with the 10 minute talks that students give three months into their PhD, and continues with a second year seminar following their upgrade from MPhil to PhD, and an ‘Exit’ seminar, presented immediately before their viva examination. Students also often interact directly with CRUK, by giving talks to supporters, or getting involved in outreach and/or fundraising activities. Therefore, by the end of their PhDs, LRI students are well-practised in communicating to both scientific and non-scientific audiences, which is reflected in the number of awards and prizes given to our students for their oral and poster presentations. At the EMBO meeting in Nice in September, Ashley Humphries (Cell Motility Group) won a poster prize, and at the National Cancer Research Institute Conference in November, Graham Bell (Epithelial Biology Group) won the Richard Hambro Student Poster Prize and Heike Meiss (Gene Expression Analysis Group) was awarded the AstraZeneca Student Prize. Rafal Lolo (DNA Damage Response Group), and Risa Mori (Cell Regulation Group) won first and runner-up talk prizes respectively at the International PhD Student Cancer Conference, held at the Netherlands Cancer Institute in June. Jeroen Claus (Protein Phosphorylation Group) won the best poster prize at the same Conference.

postdocs and make them aware of the facilities and activities available, and to introduce them to the Core Technology Facilities that will facilitate their research. Soon after starting, postdocs submit a project proposal and throughout their programme, they have annual career development reviews with their group leaders. These reviews focus on the postdocs’ science, taking a broader perspective than normal day-to-day conversations, and provide structure and focus for postdocs’ scientific and career development to aid them in their future career. Postdocs also present seminars mid-way and at the end of their fellowship. Intake of 2012 PhD Students

The next annual International PhD Student Cancer Conference will be hosted by the LRI at UCL and 44 Lincoln’s Inn Fields in June 2013. An organising committee made up of students from the LRI has been busy finding a venue, accommodation and keynote speakers for the Conference, as well as organising the scientific and social programme. This enjoyable and stimulating meeting has an extremely high scientific standard, and is attended by students from CRUK core-funded institutes as well as PhD students from top cancer research institutes across Europe. In addition to their busy academic and laboratory schedule, the students at the LRI organise a number of other scientific and social events via the Graduate Student Committee, which also provides a channel of communication between the students and LRI management. This years’ events included the Clare Hall Easter Event, where the keynote speakers were Dr David Neuhaus from the MRC Laboratory of Molecular Biology, Cambridge, and Dr Henry Chapman from the Centre for Free-Electron Laser Science, Hamburg, and the Christmas Lecture, given by Professor Colin Blakemore from the Department of Physiology, Anatomy and Genetics at the University of Oxford. LRI Postdoctoral Fellows In 2012, the total number of postdocs at the LRI remained stable at just under 200. Postdocs are initially awarded a 4-year fellowship, which is funded by CRUK. However, the LRI encourages postdocs to apply to other sources of funding where possible, for example from the European Molecular Biology Organisation, the Human Frontiers Science Programme and Marie Curie. Around half of our postdocs were funded this year by a competitively awarded fellowship or grant. The LRI Postdoctoral Training Programme, developed collaboratively by postdocs, group leaders and the Academic Director, has been running since April 2010. The programme starts with an induction to orientate



Postdocs are invited to attend Postdoc Consultative Meetings (PDCMs). PDCMs provide an opportunity for postdocs to discuss issues that affect them within the LRI and to communicate with Administration and Management. In addition, postdoc representatives attend relevant meetings including the Fellowships Committee to feedback to the PDCM what was discussed at these meetings. The annual postdoc retreat provides a forum for postdocs to meet and participate in different career development activities. This year LRI postdocs collaborated with postdocs from the National Institute for Medical Research (NIMR) to organise the first Francis Crick Institute Postdoc Retreat. The aim of this retreat was to foster collaboration and interaction between the Institutes as well as with the academic partners who will come together to form the Francis Crick Institute in 2015. The event was attended by 180 postdocs from the LRI, NIMR, King’s College London, University College London and Imperial College London, and held at the King’s College Waterloo Campus. The keynote speaker was Professor Peter Lawrence FRS, who talked about his career, his mentors and the future of science. This was followed by a ‘Speed Networking’ activity, which was designed to facilitate future scientific relationships with postdocs from different disciplines and institutes. The Retreat also included careers workshops where participants could get advice on academic career planning, job interviews and CVs. The day ended with an interactive debate on the topic of the future of scientific publishing before the Retreat concluded with a drinks and canapés evening which members of the Crick Executive Committee also attended. Career Development Activities This year, the LRI organised various career development activities, including talks on working in technology transfer, scientific publishing, and science policy, and workshops on obtaining postdoc positions and career development awards. In addition, a series of interactive Postdoc/PI discussion group sessions were held, to advise and prepare postdocs applying for independent researcher positions.

ACADEMIC PROGRAMME

159

SEMINARS AND CONFERENCES

The London Research Institute hosts a Special Seminar Series to invite external speakers from around the world to present their work, covering a broad spectrum of cutting-edge topics within the areas of genome integrity, signal transduction, structural biology, immunology and developmental biology. There are also a number of Special Interest Groups within the different areas of interest within the institute, which are open to external visitors to attend, providing a unique networking opportunity to encourage collaboration within London and the surrounding area. A selection from this year’s programme are listed below:

Special Seminars 2012 Elliot Meyerowitz, University of Cambridge, UK Physical and chemical signals control plant stem cells.

Gary Karpen, UC Berkeley, USA Chromatin regulation of genome stability.

Anjana Rao, La Jolla Institute for Allergy and Immunology, USA Signalling to gene expression.

Immunology

Geraldine Seydoux, The John Hopkins University School of Medicine, USA Breaking symmetry: polarization of the C. elegans embryo.

Dana Philpott, University of Toronto, Canada Nod proteins in infection and inflammation. Jim Kaufman, University of Cambridge, UK MDV and DFTD: the devil is in the details. Signalling

Donald Ingber, Harvard University, USA Can cancer be reversed by engineering the tumour microenvironment? Developmental Biology Allison Bardin, Institut Curie, France Balancing stem cell self-renewal and differentiation in the fly intestine. Amanda Fisher, Imperial College London, UK How does reprogramming work? Genes to Cells Jon Pines, University of Cambridge, UK Altering the balance of power in mitosis: mutual antagonism between the spindle assembly checkpoint and the APC/C.

160

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Dario Alessi, University of Dundee, UK Disruptions on the highways of cell signalling. Reuben Shaw, The Salk Institute, San Diego, USA The LKB1/AMPK tumor suppressor pathway coordinates growth, metabolism, and autophagy. Advanced Bioimaging Erik Schäffer, Biotechnology Centre, Dresden, Germany Illumin nanomechanics with optical tweezers: how cellular machines recombine DNA & work against protein friction. Christian Eggeling, University of Oxford Fluorescence (STED) super-resolution microscopy for biomedical research.

Conferences 20th April Postdoc Retreat The first Francis Crick Postdoc Retreat took place on 20th April, and was organised by postdocs from the LRI and the National Institute for Medical Research (NIMR). 180 postdocs from the LRI, NIMR, King’s College London, University College London and Imperial College London attended the event held at the King’s College Waterloo Campus. The keynote speaker was Professor Peter Lawrence FRS, who talked about his career, his mentors and the future of science. This was followed by a ‘Speed Networking’ activity, which was designed to facilitate future scientific relationships with postdocs from different disciplines and institutes. The Retreat also included careers workshops where participants could get advice on academic career planning, job interviews and CVs. The day ended with an interactive debate on the topic of the future of scientific publishing followed with drinks and canapés, which members of the Crick Executive Committee attended. 25th-26th June LRI 10th Anniversary Symposium At the end of June, the LRI held it’s 10th Anniversary Symposium at the Royal College of Surgeons, to celebrate the success of the 10 years of the LRI following the merger in 2002 of Imperial Cancer Research Fund and Cancer Research Campaign to form Cancer Research UK. The Anniversary Symposium brought together 350 past and present researchers who have worked at the Clare Hall or Lincoln’s Inn Fields Laboratories. Presentations were given by past group leaders or postdocs and students that have now established their own research groups around the world. As well as several current group leaders, the 2011 HardimanRedon winner Jessica Strid also presented her most recent research. The second day of talks concluded with celebratory drinks and birthday cake, which was cut by Richard Treisman, Director of the LRI, and everyone across both the sites joined in to help mark the occasion.

6th-8th June International Graduate Student Conference This year’s International PhD Student Cancer Conference, was hosted and organised by students at the Netherlands Cancer Institute (NKI) in Amsterdam. The annual two and a half day conference is held with other Cancer Research UK institutes (Beatson, Cambridge Paterson and the LRI) as well the NKI, and the Spanish National Cancer Centre (CNIO) in Madrid. All attendees presented their research – which covered many topics related to cancer, from basic biology to clinical aspects of the disease. The scientific talks and poster sessions were mixed with social activities. The keynote talks from Ab Osterhaus (Erasmus University Medical Centre, Rotterdam), and Jonathan Yewdell (National Institute of Allergy and Infectious Diseases,USA) provided a conference highlight and provoked a lot of questions and discussion. The 2013 conference will be held at the the LRI on 19th-21st June, and preparations are fully on the way to welcome the students to our Institute. Dates for 2013 20th May Francis Crick Postdoc Retreat 19-21st June International Graduate Student Conference 6th September David Ish-Horowicz’s Symposium

Delegates from the LRI and National Institute for Medical Research at the 2012 Francis Crick Postdoc Retreat.



SEMINARS AND CONFERENCES

161

EXTERNAL FUNDING Grants Association for International Cancer Research Axel Behrens – Mammalian Genetics John Diffley – Chromosome Replication Jesper Svejstrup – Mechanisms of Gene Transcription Bayer Charles Swanton – Translational Cancer Therapeutics Breast Cancer Campaign Mark Petronczki – Cell Division and Aneuploidy Stephen West – Genetic Recombination Cancer Research UK Discovery Committee Grant Almut Schulze – Gene Expression Analysis Cephalon Peter Parker – Protein Phosphorylation European Commission Axel Behrens – Mammalian Genetics Simon Boulton – DNA Damage Response Peter Cherepanov – Chromatin Structure and Mobile DNA Julie Cooper – Telomere Biology Vincenzo Costanzo – DNA Damage and Genomic Stability John Diffley – Chromosome Replication Julian Downward – Signal Transduction Caroline Hill – Developmental Signalling Caetano Reis e Sousa – Immunobiology Giampietro Schiavo – Molecular Neuropathobiology Jesper Svejstrup – Mechanisms of Gene Transcription Charles Swanton – Translational Cancer Therapeutics Richard Treisman – Signalling and Transcription John Diffley – Chromosome Replication Stephen West – Genetic Recombination

Liliane Bettencourt Prize for Life Sciences Award Caetano Reis e Sousa – Immunobiology Lister Institute Vincenzo Costanzo – DNA Damage and Genomic Stability Holger Gerhardt – Vascular Biology Louis Jeantet Richard Treisman – Signalling and Transcription Stephen West – Genetic Recombination Medical Research Council Axel Behrens – Mammalian Genetics Peter Cherepanov – Chromatin Structure and Mobile DNA Charles Swanton – Translational Cancer Therapeutics Motor Neurone Disease Association Giampietro Schiavo – Molecular Neuropathobiology National Institute of Health Peter Cherepanov – Chromatin Structure and Mobile DNA National Centre for the Replacement, Refinement and Reduction of Animals in Research Francois Lassailly – In Vivo Imaging Novartis Charles Swanton – Translational Cancer Therapeutics Rosetrees Trust Charles Swanton – Translational Cancer Therapeutics The Royal Society Simon Boulton – DNA Damage Response Banafshe Larijani – Cell Biophysics

European Molecular Biology Organisation (Young Investigator Programme) Mark Petronczki – Cell Division and Aneuploidy Barry Thompson – Epithelial Biology Helen Walden – Protein Structure Function

Swiss Bridge Stephen West – Genetic Recombination

Genetech Dominique Bonnet – Haematopoietic Stem Cell

Wellcome Trust Paul Nurse/Jacqueline Hayles – Cell Cycle Julian Downward – Signal Transduction

Weizmann Institute Giampietro Schiavo – Molecular Neuropathobiology

Human Frontier Science Project Thomas Surrey – Microtubule Cytoskeleton Inflammatory Breast Cancer UK Charles Swanton – Translational Cancer Therapeutics Leducq Holger Gerhardt – Vascular Biology Leukaemia and Lymphoma Research Dominique Bonnet – Haematopoietic Stem Cell Gordon Peters – Molecular Oncology

162

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

Fellowships Australian National Health and Medical Research Council Paul Whitney – Immunobiology Dutch Cancer Society Thomas Kuilman – Chromosome Segregation

European Molecular Biology Organisation Jasmine Abella – Cell Motility Sophia Blake – Mammalian Genetics Andreas Ehrensberger – Mechanisms of Gene Transcription Alfonso Fernandez Alvarez – Telomere Biology Franck Fourniol – Microtubule Cytoskeleton Belen Gomez-Gonzalez – Chromosome Replication Ralph Gruber – Mammalian Genetics Madhu Kumar – Signal Transduction Laurent L’Epicier Sansregret – Cell Division and Aneuploidy Laurent Malivert – Genetic Recombination Nuria Martinez-Martin – Lymphocyte Interaction Fabio Puddu – DNA Damage and Genomic Stability Johanna Roostalu – Microtubule Cytoskeleton Grzegorz Sarek – DNA Damage Response Annemarthe Van der Veen – Immunobiology

Medical Research Council Marco Gerlinger – Translational Cancer Therapeutics Yanlan Mao – Apoptosis and Proliferation Control Naito Foundation Hirofumi Takada – Cell Regulation Swiss National Science Foundation Kanagaraj Radhakrishnan – Genetic Recombination Wellcome Trust Sophie Acton – Immunobiology Esther Arwert – Tumour Cell Biology Ross Chapman – DNA Damage Response Hannah Mischo – Mechanisms of Gene Transcription Alison Twelvetrees – Molecular Neuropathobiology Frances Willenbrock – Protein Phosphorylation

Fundacion Alfonso Martin Escudero Ines Martinez-Corral – Lymphatic Development Student Fellowships Fundación Espanola Para La Ciencia y la Tecnología Marta Sanz-Garcia – Protein Phosphorylation Fundación Ramon Areces Pilar Gutierrez-Escribano – Cell Cycle German Research Foundation (DFG) Markus Diefenbacher – Mammalian Genetics Selina Keppler – Lymphocyte Interaction

Agency for Science, Technology and Research Tianyi Zhang – Mammalian Genetics Austrian Academy of Science Scholarship Irene Aspalter – Vascular Biology Breast Cancer Campaign PhD Grant Sriramkumar Sundaramoorthy – Cell Division and Aneuploidy

Human Frontier Science Project Raquel Blanco – Vascular Biology Kerstin Gari – DNA Damage Response Joao Matos – Genetic Recombination Pieta Mattila – Lymphocyte Interaction

Boehringer Ingelheim Fund Rahul Thadani – Chromosome Segregation Janneke Van Blijswijk – Immunobiology Yanxiang Zhou – Apoptosis and Proliferation Control

Instituto Pasteur Fondazione Cenci Bolognetti Francesca Gasparrini – Lymphocyte Interaction

Bogue Travel Fellowship – UCL Rafal Lolo – DNA Damage Response

Italian Foundation for Cancer Research (FIRC) Fabio Puddu – DNA Damage and Genomic Stability

Global Excellence Award – UCL Ainhoa Mariezcurrena – Chromosome Segregation

Japan Society for the Promotion of Science Yukata Handa – Cell Motility Yasuto Murayama – Chromosome Segregation

Overseas Research Scholarship Sakshi Gulati – Biomolecular Modelling Michael Lim – Mechanisms of Gene Transcription Rahul Thadani – Chromosome Segregation

Marie Curie Marco Briones-Orta – Developmental Signalling Livija Deban – Immuno Surveillance Safia Deddouche – Immunobiology Xavier Fontana – Mammalian Genetics Claudio Franco – Vascular Biology Michael Klutstein – Telomere Biology Lourdes Lopez Onieva – Haematopoietic Stem Cell Olga Martins de Brito – Molecular Neuropathobiology Sebastian Maurer – Microtubule Cytoskeleton Miriam Molina Arcas – Signal Transduction Clare Sheridan – Signal Transduction Antonia Tomas Loba – DNA Damage Response Pierre Vantourout – Immuno Surveillance Santiago Zelenay – Immunobiology



Medical Research Council Martin Wallace – Molecular Neuropathobiology Portuguese Government Pedro Gaspar – Apoptosis and Proliferation Control Fundação para a Ciência e a Tecnologia Mariana Campos – Epithelial Biology Filipa Neto – Vascular Biology

EXTERNAL FUNDING

163

INSTITUTE MANAGEMENT

London Research Institute Management Committee Richard Treisman PhD FRS (Chair) Giampietro Schiavo PhD Julian Downward PhD FRS John Diffley PhD FRS Michael Way PhD Stephen West PhD FRS Ava Yeo PhD London Research Institute Faculty Committee Richard Treisman PhD FRS (Chair) Michael Way PhD (Co-Chair) Simon Boulton PhD Holger Gerhardt PhD (from October 2012) Caroline Hill PhD (from October 2012) Caetano Reis e Sousa PhD (until October 2012) Erik Sahai PhD Jesper Svejstrup PhD FRS (from October 2012) Nic Tapon PhD (until October 2012) Frank Uhlmann PhD Helle Ulrich PhD (until October 2012) Ava Yeo PhD (in attendance) Julian Downward PhD FRS (ex officio) John Diffley PhD FRS (ex officio) London Research Institute Fellowships Committee John Diffley PhD FRS (Chair) Nic Tapon PhD (Deputy Chair) Simon Boulton PhD Adrian Hayday PhD Ilaria Malanchi PhD (joined April 2012) Neil McDonald PhD Peter Parker PhD FRS Sharon Tooze PhD Takashi Toda PhD (joined April 2012) Helle Ulrich PhD Helen Walden PhD (joined April 2012) Sally Leevers PhD (ex officio) Ava Yeo PhD (in attendance)

164

SCIENTIFIC REPORT 2012 LONDON RESEARCH INSTITUTE

London Research Institute Graduate Students Advisors Committee Sally Leevers PhD (Chair) Peter Cherepanov PhD (from Feb 2012) Caroline Hill PhD (from Nov 2012) Gordon Peters PhD (until Nov 2012) Mark Petronczki PhD (until Feb 2012) Erik Sahai PhD Thomas Surrey PhD (from Feb 2012) Charles Swanton PhD (from June 2012) Nic Tapon PhD (from Nov 2012) Helle Ulrich PhD Helen Walden PhD (until June 2012) Michael Way PhD (until Feb 2012) Dr Kaila Srai PhD (UCL) Ava Yeo PhD (in attendance) Sabina Ebbols (in attendance) Sophie Lutter PhD (in attendance)

CONTACT DETAILS

London Research Institute Scientific Report 2012 Kings Cross St Pancras

ON ST EU

AD RO

Euston

GR

Lincoln’s Inn Fields Laboratories 44 Lincoln’s Inn Fields London WC2A 3LY United Kingdom

NR S IN AY

U SO ON PT

AM

TH RO W Holborn

LBORN

O HIGH H

500 metres

No. 44

G

RE AC

AL

N LO

Lincoln’s Inn Fields

YCH DW

Temple

Covent Garden

Leicester Square

D

AN

OO RL

TR

BR

ES TH

N

TE WA E

IDG

Cancer Research UK Registered Charity in England & Wales (1089464), Scotland (SC041666) and the Isle of Man (1103). Registered as a company limited by guarantee in England and Wales (4325234) and the Isle of Man (5713F) Registered address: Angel Building, 407 St. John Street, London EC1V 4AD

D OA

Russell Square

Clare Hall Laboratories Blanche Lane South Mimms Potters Bar Hertfordshire EN6 3LD United Kingdom

Charing Cross Waterloo

Tel 44(0) 20 7242 0200 www.cancerresearchuk.org GR EYH O

ISSN 1479-0378 Copyright © 2012 Cancer Research UK

UN D

E

AN

EA

L HE

RL

LA NE

NC

A

SL

BL

AN

E BIGNELLS CORNER

M2

Printed by Williams Lea

5 NS R LBA

ST A OAD

Clare Hall Laboratories

JUNCTION 29

NE

E LA

NCH

BLA

BIGNELLS CORNER

A1081

A1

FR

OM

LO ND

ON

N

London Research Institute Cancer Research UK Lincoln’s Inn Fields Laboratories 44 Lincoln’s Inn Fields London WC2A 3LY United Kingdom Tel: +44 (0) 20 7242 0200 http://www.london-research-institute.org.uk/

cruk.org london-research-institute.org.uk