UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Roles of the DOG-1 and JRH-1 helicase-like proteins in DNA repair in Caenorhabditis elegans Youds, Jillian L. 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2007_spring_youds_jillian.pdf [ 10.76MB ]
Metadata
JSON: 24-1.0066819.json
JSON-LD: 24-1.0066819-ld.json
RDF/XML (Pretty): 24-1.0066819-rdf.xml
RDF/JSON: 24-1.0066819-rdf.json
Turtle: 24-1.0066819-turtle.txt
N-Triples: 24-1.0066819-rdf-ntriples.txt
Original Record: 24-1.0066819-source.json
Full Text
24-1.0066819-fulltext.txt
Citation
24-1.0066819.ris

Full Text

Roles of the DOG-1 and JRH-1 helicase-like proteins in D N A repair in Caenorhabditis elegans by Jillian L . Youds  B.Sc.,  Simon Fraser University, 2003  A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E REQUIREMENTS FORTHE DEGREE OF D O C T O R OF PHILOSOPHY in  The Faculty o f Graduate Studies (Medical Genetics)  T H E UNIVERSITY OFBRITISH C O L U M B I A December, 2007  © Jillian L . Youds, 2007  ABSTRACT Helicases perform vital roles in the cell by unwinding D N A to make it accessible for the essential processes o f replication, transcription and repair. In Caenorhabditis  elegans, the D O G -  1 helicase-like protein is required for polyG/polyC-tract (G/C-tract) maintenance, as dog-l animals have a mutator phenotype characterized by deletions that initiate in G/C-tracts. D O G - 1 may unwind secondary structures that form in polyguanine D N A during lagging strand replication. In order to more completely understand the role o f dog-1, genetic interactors were identified, dog-1 functionally interacts with the him-6/BLM  helicase. Absence o f  recombinational repair-implicated proteins in the dog-1 background, including H I M - 6 / B L M , R A D - 5 1 , B R D - 1 / B A R D 1 and H I M - 9 / X P F , as well as the trans-lesion synthesis polymerases polKMD po/7 increased the frequency o f animals with G/C-tract deletions, indicating that these pathways are important mechanisms for repair at G/C-tracts in the absence o f D O G - 1 . These data support the hypothesis that persisting D N A secondary structures can cause replication fork stalling, which can be resolved by deletion-free or deletion-prone mechanisms.  D O G - 1 has highest sequence identity to human B R 1 P 1 / F A N C J , which is mutated in patients from the Fanconi Anemia ( F A ) subgroup J. D N A damage sensitivity experiments indicated that, like chicken F A N C J cells, dog-1 mutants were not significantly sensitive to D N A damage from X-ray or UV-irradiation, but were extremely hypersensitive to the D N A interstrand cross-linking agent UVA-activated trimethylpsoralen. Thus, D O G - 1 appears to have a conserved role in cross-link repair and is the C. elegans F A N C J homolog. Characterization o f the dog-1/FANCJ-relatsd helicase, Jrh-1, revealed that mutants for this putative helicase are moderately sensitive to cross-linking agents, dog-1 jrh-1 double mutants displayed a synthetic lethal phenotype characterized by excessive recombination intermediates and mitotic catastrophe in the germline. However, absence o f JRH-1 did not have any effect on G/C-tract deletions, indicating that J R H - 1 does not have a redundant function with D O G - 1 at G/C-tracts. Absence o f JRH-1 reduced the fitness o f eTl and nTl translocation hétérozygotes, but not translocation homozygotes, jrh-1 was synthetically lethal with him-6/BLM and with the endonuclease mus-81, suggesting a possible role for J R H - 1 in regulating the balance between different types o f repair.  TABLE OF CONTENTS Abstract  ii  Table o f contents  iii  List o f tables  vii  List o f figures  viii  List o f abbreviations  x  Acknowledgements  xiii  Co-authorship statement  xiv  C H A P T E R 1 : Introduction 1.1 Pathways for D N A repair  1  1.1.1 Double strand break repair  2  1.1.2 Mismatch repair  4  1.1.3 Nucleotide excision repair  5  1.1.4 Lesion bypass  6  1.1.5 Interstrand cross-link repair 1.2 Helicases involved in D N A repair  8 11  1.2.1 The R e c Q helicases  11  1.2.2 The Srs2 helicase  15  1.3 C. elegans as a model for the study o f D N A repair  16  1.4 PolyG/polyC-tracts and the D O G - 1 helicase-like protein in C. elegans  18  1.5 Thesis objectives  19  CHAPTER 1 REFERENCES  20  C H A P T E R 2: dog-l and repair o f G/C-tracts in C. elegans 2.1 Introduction  28  2.2 Materials and Methods  29  2.2.1 W o r m lysis and P G R detection oidog-l(gklO)  29  2.2.2 Construction of double mutant strains  30  2.2.3 D A P I staining and measurement of mitotic zone nuclei  31  2.2.4 The vab-1 G/C-tract deletion assay  31  2.2.5 R N A i feeding method for the translesion synthesis polymerases  33 iii  2.3 Results  34  2.3.1 dog-1 functionally interacts with him-6/BLM.  34  2.3.2 Increased G/C-tract deletions in dog-1; him-6  38  2.3.3 Homologous recombination is required for G/C-tract stability in dog-1 mutants...42 2.3.4 Translesion synthesis is required for G/C-tract stability in dog-1 mutants  44  2.3.5 Nucleotide excision repair and non-homologous end-joining are not essential for repair at G/C-tracts  45  2.3.6 Base pair mutations frequently flank the G/C-tract deletions 2.4 Discussion  46 48  2.4.1 dog-1 is likely to function during repHcation  48  2.4.2 Homologous recombination and translesion synthesis are required for genome stability in dog-1 animals  49  2.4.3 H I M - 6 could function in RAD-51-dependent or -independent pathways  52  2.4.4 Loss o f N E R or N H E J does not affect deletion formation  53  2.4.5 A model for repair at G/C-tracts  54  CHAPTER 2 REFERENCES  C H A P T E R 3: dog-1 is the C. elegans BRIPl/FANCJ  57  homolog  3.1 Introduction  62  3.2 Materials and Methods  63  3.2.1 Measuring apoptosis in the C. elegans germline by S Y T 0 1 2 staining  63  3.2.2 D N A damage sensitivity assay - X-ray treatment  63  3.2.3 D N A damage sensitivity assay - UVC-irradiation treatment  64  3.2.4 D N A damage sensitivity assay - trimethylpsoralen+UVA treatment  64  3.2.5 D A P I staining following T M P + U V A treatment  66  3.3 Results 3.3.1 dog-1 is the reciprocal b e s t - B L A S T o f human BRIPl/FANCJ  67 67  3.3.2 dog-1 animals have increased D N A damage-induced apoptosis in the germline....69 3.3.3 dog-1 animals are sensitive to D N A interstrand cross-hnking agents  71  3.3.4 dog-1 mutants display evidence o f repair defects after cross-linking treatment  74  3.4 Discussion  77  3.4.1 Phenotypes o f the dog-1 mutant suggest D N A damage under normal conditions...77 3.4.2 D O G - 1 is the C. elegans B R I P l / F A N C J homolog  78  3.4.3 Implications o f sequence conservation between D O G - 1 and B R I P l / F A N C J  78  3.4.4 Relationship o f D O G - l / F A N C J to the F A pathway  81  3.4.5 H o w does the role for dog-1 in I C L repair relate to G/C-tracts?  82  3.4.6 Future experiments with dog-1  83  CHAPTER 3 REFERENCES  85  C H A P T E R 4: Characterization o f the D O G - l / F A N C J - r e l a t e d helicase JRH-1 4.1 Introduction  87  4.2 Materials and Methods  88  4.2.1 S Y T O 1 2 staining  88  4.2.2 D N A damage sensitivity experiments  88  4.2.3 D A P I staining  88  4.2.4 A n t i - R A D - 5 1 antibody staining  89  4.2.5 Strains and strain constructions  90  4.2.6 Measuring meiotic recombination  92  4.3 Results  93  4.3.1 jrh-1 is a member o f the dog-1 family and encodes a helicase-like protein  93  4.3.2 Phenotypes o f jrh-1 mutant animals  95  4.3.3 jrh-1 mutants are not sensitive to D N A damage induced by U V C or X-ray irradiation  96  4.3.4 jrh-1 mutants are moderately sensitive to D N A interstrand cross-linking agents...97 4.3.5 Loss of JRH-1 function is synthetically lethal in the absence o f D O G - 1  99  4.3.6 JRH-1 is not required for G/C-tract stability  102  4.3.7 jrh-1 dog-1 animals show extensive R A D - 5 1 foci in the mitotic germline  102  4.3.8 Loss o f function o f JRH-1 in eTl translocation hétérozygotes  104  4.3.9 Meiotic recombination is increased in jrh-1 mutants  108  4.3.10 him-6/BLM and mus-81 functionally interact with jrh-1  110  4.4 Discussion  113  4.4.1 Phenotypes of jrh-1 mutant animals support a role in replication or repair  113  4.4.2 JRH-1 is required for repair in the absence o f D O G - 1  115  4.4.3 Absence o f JRH-1 specifically affects eTl and nTl translocation hétérozygotes..! 16 4.4.4 J R H - 1 might have a function similar to the yeast helicase Srs2 4.4.5 A role for JRH-1 at telomeres?  120 122  CHAPTER 4 REFERENCES  124  C H A P T E R 5: General Discussion 5.1 C elegans as a model to study interstrand cross-link repair  128  5.2 A potential role for human B R I P l / F A N C J at G-rich D N A  129  5.3 The D O G - l / F A N C J - r e l a t e d helicase, J R H - 1 , functions in D N A repair  130  5.4 A n extended model for repair at G/C-tracts in the dog-1 mutant  131  5.5 Future studies on D O G - 1 and the D O G - 1 helicase-like proteins  134  CHAPTER 5 REFERENCES  136  A P P E N D I X 1: Primer sequences  138  LIST OF TABLES Table 1.1: M a i n pathways for D N A repair and the proteins involved  2  Table 2.1: Phenotypes o f N 2 , dog-1, him-6 and dog-1; him-6 animals  36  Table 2.2: Average nuclei and largest nuclei sizes in N 2 , dog-1, him-6, and dog-1; him-6 animals  38  Table 2.3: Deletions observed in the vab-1 G/C-tract in various repair mutants in the dog-1 background  41  Table 2.4: Sequences around vab-1 G/C-tract deletion breakpoints  47  Table 4.1 : Frequency o f dog-1 and jrh-1 single mutant and jrh-1 dog-1 double mutant animals with vab-1 G/C-tract deletions Table 4.2: Viability of eTl/dpy-18;  eTl/unc-46 and eTl/eTl  102 with and without jrh-1  (tml866).106  Table 4.3: Meiotic recombination frequency in the absence o f JRH-1  110  Table 4.4: Phenotypes ofjrh-1, him-6 and jrh-1; him-6 animals  Ill  Table 4.5: Phenotypes ofjrh-1 and mus-81 single mutants and jrh-1 mus-81 double mutants...! 12  LIST OF FIGURES Figure 1.1 : A speculative model for interstrand cross-link repair by the Fanconi anemia pathway  10  Figure 1.2: Proposed roles for Mus81, S g s l and Srs2 in repair  13  Figure 1.3: Schematic o f the C. elegans germline  17  Figure 2.1: Viability o f independent lines oï dog-1, him-6 and dog-1; him-6 animals  35  Figure 2.2: Germline mitotic zones o f N 2 , dog-1, him-6, and dog-1; him-6 animals  37  Figure 2.3: The vab-1 deletion assay  39  Figure 2.4: A model for repair o f secondary structures forming at G/C-tracts  55  Figure 3.1 : Protein sequence alignment of C. elegans D O G - 1 and human B R I P l / F A N C J  67  Figure 3.2: Conserved domains o f C. elegans D O G - 1 , G. gallus B R I P l and H. sapiens BRIP1.68 Figure 3.3: Phylogram showing the estimated evolutionary relationship between human B R I P l / F A N C J , chicken B R I P l / F A N C J , C. elegans D O G - 1 , human N H L l , chicken R T E L , and C. elegans F25H2.13  69  Figure 3.4: Number of SYT012-stained corpses in N 2 , dog-1 and dog-1 cep-1 animals  70  Figure 3.5: Sensitivity of N 2 and dog-1 animals to UVC-irradiation and X-ray treatment  71  Figure 3.6: Embryonic survival in the progeny of N 2 and dog-1 animals following no treatment, U V A only, T M P only or T M P + U V A treatment  72  Figure 3.7: Variability between trials o f T M P + U V A treatment  73  Figure 3.8: Sensitivity of N 2 and dog-1 animals to T M P + U V A treatment  74  Figure 3.9: Number of SYT012-stained corpses in N 2 and dog-1 animals under normal conditions and N 2 and dog-1 animals 24 hours after D N A cross-linking with T M P + U V A treatment  75  Figure 3.10: D A P I stained N 2 and dog-1 mutant germlines after T M P + U V A treatment  76  Figure 4.1: Phylogram of genes in the dog-1 and RecQ helicase families  93  Figure 4.2: Comparison o f the domain structure o f D O G - 1 and JRH-1  94  Figure 4.3: jrh-1 gene structure  95  Figure 4.4: Number of SYT012-stained apoptotic corpses in N 2 and jrh-1 mutant animals  96  Figure 4.5: Sensitivity o f N 2 and jrh-1 animals to UVC-irradiation and X-ray treatment  97  Figure 4.6: Embryonic survival in the progeny of N 2 and jrh-1 animals following no treatment, U V A only, T M P only or cross-linking by T M P + U V A treatment  98  Figure 4.7: Sensitivity o f N 2 , dog-1 and jrh-1 animals to T M P + U V A treatment  99  Figure 4.8: Comparison o f N 2 and jrh-1 dog-1 double mutant animals  100  Figure 4.9: Comparison o f whole D A P I stained N 2 , dog-1, jrh-1 and jrh-1  dog-1  double mutant germlines  101  Figure 4.10: Mitotic zones o f D A P I stained N 2 , dog-1, jrh-1 and jrh-1 dog-1 double mutant germlines  101  Figure 4.11: R A D - 5 1 staining of jrh-1 dog-1 and hT2 dog-1/jrh-1 dog-1 controls Figure 4.12: Competition assay oï eTl/dpy-18; eTl/unc-46  and jrh-1; eTl/dpy-18;  104 eTl/unc-46  animals Figure 4.13: Competition assay of nTl/+; nTl/dpy-U  107 and jrh-1; nTl/+;  nTl/dpy-11  animals.108  Figure 4.14: Fitness o f 20 lines of N 2 , him-6 and jrh-1 single mutants and jrh-1; him-6 double mutants Figure 4.15: M o d e l for the affect of JRH-1 absence on eTl hétérozygotes  Ill 118  Figure 5.1: A n extended model for repair of replication forks stalled at G/C-tract secondary structures  134  LIST OF ABBREVIATIONS  AT  Ataxia Telangiectasia  ATL-1  C. elegans Ataxia Telangiectasia-Like  ATLD  Ataxia Telangiectasia-Like Disorder  ATM  Ataxia Telangiectasia Mutated  ATR  Ataxia Telangiectasia Related  BACH  B R C A 1-interacting C-terminal Helicase  BARD  B R C A 1-Associated R I N G Domain  BCH-1  C. elegans B A C H 1-related (also known as JRH-1 and S P A R - 1 )  BLAST  Basic L o c a l Alignment Search Tool  BLM  B L o o M syndrome protein  BRCA  Breast Cancer Associated  BRC-1  C. elegans B R C A l homolog  BRD-1  C. elegans B A R D l homolog  BRIPl  B R C A l - I n t e r a c t i n g Protein  BSA  Bovine Serum A l b u m i n  CEP-1  C. Elegans P53 homolog  CS  Cockayne Syndrome  DAPI  4'-6-DiAmidino-2-PhenylIndole  DNA-PKcs  D N A - P r o t e i n Kinase catalytic subunit  DOG-1  C. elegans Deletions O f Guanine  DPY  C elegans Dumpy  ERCCl  Excision Repair Cross-Complementing  FA  Fanconi Anemia  FANC  Fanconi Anemia protein  FAAP  Fanconi Anemia-Associated Protein  FCD-2  C elegans F A N C D 2 homolog  FEN  Flap EndoNuclease  GFP  Green Fluorescent Protein  GG-NER  Global Genome Nucleotide Excision Repair  HJ  Holliday Junction  HIM  H i g h Incidence o f Males  HRR  Homologous Recombination Repair  ICL  Interstrand Cross-Link  IPTG  IsoPropyl-beta-D-ThioGalactopyranoside  JRH-1  C. elegans D O G - l / F A N C J - R e l a t e d Helicase (also known as B C H - 1 and S P A R - 1 )  LB  Luria-Bertani M e d i a  LIG4  Ligase 4  MSH  MutS Homolog  MLH  M u t L Homolog  NGM  Nematode Growth M e d i a  PMS  PostMeiotic Segregation increased  MMR  Mismatch Repair  MRE  Meiotic REcombination homolog  MRN  M R E 1 1 / R A D 5 0 / N B S 1 complex  MRT  C. elegans M o R T a l geririline  NBS  Nijmedgen Breakage Syndrome  NER  Nucleotide Excision Repair  NHL  N o v e l Helicase-Like  NHEJ  Non-Homologous End-Joining  PBS  Phosphate Buffered Saline  PCNA  Proliferating C e l l Nuclear Antigen  PCR  Polymerase Chain Reaction  PVL  C. elegans Protruding V u L v a e  RAD-51  R A D i a t i o n sensitivity abnormal  RFC  Replication Factor C  RNAi  R N A interference  RPA  Replication Protein A  RPM  Revolutions Per Minute  RTEL  Regulator TElomere Length  SCE  Sister Chromatid Exchange  SDSA  Synthesis Dependent Strand Annealing  SGS  Suppressor Growth Slow  SPAR-1  C. elegans SuPressor Aberrant Recombination (also known as B C H - 1 and JRH-1)  SRS  Suppressor Rad Six  ssDNA  single stranded D N A  TC-NER  Transcription Coupled Nucleotide Excision Repair  TFIIH  Transcription Factor II H  TILLING  Targeting Induced Local Lesions in Genomes  TLS  Trans-Lesion Synthesis  TMP  TriMethylPsoralen  TRF  Telomere Repeat Factor  TTD  TrichoThiDystrophy  UNC  C. elegans UNCoordinated  VAB  C. elegans Variable A B n o r m a l  WRN  W e R N e r syndrome protein  XP  Xeroderma Pigmentosum  XP-V  Xeroderma Pigmentosum-Variant  XRCC  X-ray Repair Cross-Complementing  ACKNOWLEDGEMENTS  The work completed in this thesis could not have been accomplished without the support of multiple individuals, who I wish to acknowledge here. I would like to thank A n n Rose for her excellent supervision and guidance throughout the past four years. I am in debt to Nigel O ' N e i l for sharing his knowledge and ideas and for his mentorship in all aspects o f this project. Thanks to all members o f the Rose lab for their support, discussions and general positive presence in the lab, and especially to Shir Hazir for pouring all those plates. Thanks also to collaborators Simon Boulton, Louise Barber and Jordan Ward for sharing data and discussing ideas. I would like to recognize my supervisory committee members, Peter Lansdorp, D o n Riddle and Hugh Brock for contributing their time and ideas to this project. Also, I would like to acknowledge D o n Riddle, David Baillie, Phil Hieter and K i r k McManus, D o n Moerman and M a r k Edgley and their laboratories for sharing lab equipment. I am grateful for financial support from the Natural Sciences and Engineering Research Council, the Michael Smith Foundation for Health Research and the National Cancer Institute o f Canada. Finally, I wish to thank Ngan Huynh and my parents and family for their continued love, encouragement and support.  CO-AUTHORSHIP STATEMENT A version o f Chapter 2 o f this thesis has already been published: Youds J L , O ' N e i l N J , Rose A M (2006) Homologous recombination is required for genome stability in the absence o f DOG-1 in Caenorhabditis  elegans. Genetics 173(2): 697-708. The experiments reported in the  published manuscript were designed by J L Y , N J O and A M R . Experiments were carried out by J L Y . J L Y wrote the paper, which was edited by N J O and A M R .  CHAPTER 1: Introduction The accurate transmission o f genetic information is in a delicate balance between maintaining genome stability and allowing for mutational events that drive evolution. D N A repair pathways function i n the preservation o f genome integrity by detecting damage, removing lesions and restoring the D N A sequence. A s lesions in the D N A can arise frequently from both endogenous and exogenous sources, and mutations can be heritable, D N A repair genes are particularly important in ensuring faithful copying of D N A during replication. If the D N A is not accurately replicated and transmitted, birth defects, neurological disorders, premature aging, and development o f cancer can resuh (Vinson & Hales, 2002; M i r k i n , 2006; Woodruff & Thompson, 2003; Andressoo etal, 2006; Hoeijmakers, 2001).  1.1 Pathways for DNA repair When damage to the D N A occurs, a specific series o f events takes place. First, the damage must be detected and the D N A damage checkpoint activated. The checkpoint functions to halt the cell cycle until the damage can be repaired and the cell cycle can be resumed. T w o central components o f the D N A damage checkpoint are the kinases A T M (Ataxia Telangiectasia Mutated) and A T R (Ataxia Telangiectasia mutated and Rad3-related; Harrison & Haber, 2006). A T M responds to double strand breaks, whereas A T R responds to lesions that generate singlestranded D N A , such as stalled replication forks (Harrison & Haber, 2006). Downstream of the D N A damage checkpoint, a series o f transducers and effectors function to either repair the damage or execute apoptosis. There are multiple pathways that can carry out repair, and the effectors that complete the repair depend upon the type o f D N A lesion that is present. The following sections describe the general mechanisms and components o f the major repair pathways relevant to the work described in this thesis (Table 1.1).  Lesion  Repair Pathway  Proteins Involved  Double strand breaks  Homologous recombination repair  ATM, MREl 1/NBS1/RAD50, RAD51, RAD52, RAD54, RFA, BRCAl, BARDl, BLM, BRCA2  Double strand breaks  Non-homologous end-joining  KU70/80, Artemis, DNA-PKcs, LIG4/XRCC4, BRCAl, SlRTl  Nucleotide mismatches, insertion/deletions  Mismatch repair  MSH2, MSH3, MSH6, MLHl, PMSl, PMS2, FENl  Bulky adducts, photoproducts  Nucleotide excision repair  XPA, XPB, XPC/HHR23B, XPD, XPF/ERCCl, XPG, L1G1,CSA CSB, DDBl  Replication fork stalls  Translesion synthesis  Interstrand cross-links, replication fork blocking lesions  Interstrand cross-link repair  POLii, POLi,  P O L k , POLÇ  REVl  ATR, FANCA, FANCB, FANCC, FANCD1/BRCA2, FANCD2, FANCE, FANCF. FANCG, FANCI, FANCJ/BRIPl, FANCL, FANCM, FANCN/PALB2, BRCAl, BLM, XPF/ERCCl, TOPOnitt, FAAPlOO, MREl 1/NBS1/RAD50, TLS pols  Tablel.l: A list o f several o f the main pathways for D N A repair and many o f the proteins involved in these pathways. Mechanisms for these pathways are described in the text that follows.  1.1.1 Double strand break repair Double strand breaks can occur as a result o f exogenous agents, such as ionizing radiation, or by endogenous causes, such as reactive oxygen species or when D N A containing a single strand break is replicated (Scott & Pandita, 2006). Furthermore, double strand breaks are intentionally created during meiosis. Therefore, double strand breaks are a common lesion and must be managed effectively in order to prevent chromosomal instability. T w o main pathways exist to repair double strand breaks: homologous recombination repair and non-homologous endjommg.  There are several double strand break repair deficiencies in humans that are characterized by various genome instability phenotypes such as cancer predisposition, immunodeficiency  and/or developmental defects. These include Ataxia Telangiectasia (AT) and Seckel syndrome, which are caused by defects in the checkpoint proteins A T M and A T R , respectively (Canman & L i m , 1998; O ' D r i s c o l l et al, 2003; reviewed in Scott & Pandita, 2006). Defects in these proteins result in difficulty recognizing the D N A damage and transducing signals in response to the damage (Scott & Pandita, 2006). Ataxia Telangiectasia-like disorder ( A T L D ) and Nijmegen Breakage syndrome ( N B S ) are caused by mutations in the M R E l 1 and N B S l components o f the M R E l 1 / R A D 5 0 / N B S 1 complex ( M R N complex), which functions in the early detection o f the break, as well as in double strand break end processing (Taylor et al, 2004; Matsuura et al, 1998). Mutations in components o f the two pathways that carry out double strand break repair are also causes for human diseases. One subtype o f the cancer susceptibility syndrome Fanconi anemia is caused by defects i n homologous recombination repair i n the absence F A N C D 1 / B R C A 2 . L I G 4 syndrome, which is characterized by developmental delay, skin abnormalities and cellular radiosensitivity, is caused by mutation i n ligase4 that results in faulty non-homologous end-joining (Scott & Pandita, 2006).  The homologous recombination repair (HRR) pathway functions to repair breaks using a homologous sequence as a template. This type o f repair occurs primarily during replication, in S- and G2-phase, and is promoted by B R C A l (reviewed in W y m a n & Kanaar, 2006; Powell & Kachnic, 2003). H R R takes place when single-stranded 3 ' D N A overhangs are generated; this process involves the M R N complex. The R A D 5 1 recombinase loads onto the single-stranded D N A end and is stabilized by B R C A 2 (Esashi et al, 2007; Davies & Pellegrini, 2007); this facilitates the invasion o f the template strand, creating a branched structure resembling a Holliday junction (HJ). The H J can be resolved to give crossover or non-crossover outcomes, and different mechanisms and enzymes might be used for H R R depending on the structure of the  broken D N A ends (Wyman & Kanaar, 2006). Because it uses a template for repair, H R R is generally error-free.  In non-homologous end-joining (NHEJ), the other main pathway for double strand break repair, blunt D N A ends or ends with small complementary overhangs are pieced together. N H E J might be the main pathway for double strand break repair during the G1-phase o f the cell cycle, when no sister chromatid is available for H R R (reviewed in Burma et al, 2006). Prior to N H E J , D N A end processing can occur depending on the presence o f gaps or end-blocking groups; end processing can involve various enzymes, such as Artemis (Burma et al., 2006). During N H E J , the K U 7 0 / K U 8 0 heterodimer recognizes and binds the D N A end, then recruits D N A - P K c s (Feldmann et al, 2000; Walker et al, 2001). A s a kinase, the active D N A - P K c s activates and recruits Ligase I V / X R C C 4 , which is responsible for ligation o f the D N A ends (Martin & M a c N e i l l , 2002; B u r m a et al, 2006). Although N H E J repairs breaks with high efficiency, it tends to be error-prone because it does not use a template for repair. N H E J is associated with gross chromosomal rearrangements that can result from the joining o f improper D N A ends (Varga & Apian, 2005). However, some studies have suggested that B R C A l might function in N H E J to reduce its mutagenic potential (Bau et al., 2004; Zhuang et al., 2006).  1.1.2 Mismatch repair Base substitutions and small insertion-deletion mismatches can occur in the D N A as a result o f errors introduced by the D N A polymerase. The D N A mismatch repair ( M M R ) pathway is the main mechanism that acts in repair of substitutions and insertion-deletions. Mutations in components o f the M M R pathway can lead to microsatellite instability that is characteristic o f hereditary non-polyposis colorectal cancer and various sporadic cancers (Hoeijmakers, 2001).  M M R is initiated when P C N A recruits the M M R proteins to the site of the damage. The M S H 2 / M S H 6 heterodimer binds to single base mismatches or insertion deletion mismatches, while the M S H 2 / M S H 3 heterodimer binds insertion deletion mismatches up to 16 nucleotides long (reviewed in Kunkel & Erie, 2005). M L H 1 / P M S 2 and M L H l / P M S l interact with the M S H complexes, also participating in the repair (Kunkel & Erie, 2005). The new strand o f D N A containing the error is excised and resynthesized; several proteins are implicated in these steps, including polymerase ô, R P A , R F C , and exonucleases such as F E N l (Hoeijmakers, 2001 ; Kunkel & Erie, 2005). Several o f the mismatch repair proteins, including M L H l and M L H 3 are reported to be involved in both M M R and meiotic recombination, as least in yeast (Wang et al., 1999).  1.1.3 Nucleotide excision repair Bulky adducts induced by chemicals and photoproducts caused by UV-irradiation are among the lesions repaired by the nucleotide excision repair pathway ( N E R ) . Defects in N E R proteins cause Xeroderma Pigmentosum (XP), Cockayne syndrome (CS), and tricothiodystrophy (TTD; reviewed in de Boer & Hoeijmakers, 2000). X P is characterized by skin cancer predisposition due to extreme photosensitivity, while C S and T T D patients display a spectrum o f neurological abnormalities as a consequence o f N E R deficiency (de Boer & Hoeijmakers, 2000).  Two sub-pathways o f N E R exist: these are global genome repair ( G G - N E R ) and transcription coupled repair ( T C - N E R ) . G G - N E R acts on lesions throughout the genome, whereas T C - N E R is triggered by lesions that block the transcription o f genes that are actively being transcribed (de Boer & Hoeijmakers, 2000). Lesions that distort the D N A helix are recognized and bound by the X P C / H H R 2 3 B complex in the early steps o f G G - N E R (Sugasawa et al, 1998). The stalled R N A polymerase could be a signal for T C - N E R that is recognized by  C S B and X P G , which might function in recruiting T F I I H to remodel the transcription bubble before T C - N E R can be carried out (Sarker et al, 2005). In both G G - N E R and T C - N E R , X P A functions in an early step, where it might verify the damage or orchestrate other proteins, such as R P A , and the T F I I H and E R C C l / X P F complexes in the repair process (de Boer & Hoeijmakers, 2000). The D N A helix around the site o f the lesion is locally unwound to allow access by the repair machinery; this might involve the X P B and X P D helicases that are associated with the T F I I H complex (Evans et al, 1997). X P G and E R C C l / X P F are involved i n excision either side of a 24-32 nucleotide sequence containing the lesion (O'Donovan et al, 1994; Sijbers et al, 1996; de Boer & Hoeijmakers, 2000). D N A synthesis occurs across the gap, and this involves polymerases ô and s, along with other replication factors (Shivji et al, 1995). N E R is completed when the newly synthesized D N A is ligated, likely by D N A ligase I (Barnes et al, 1992).  1.1.4 Lesion bypass A n y lesion that interferes with the progress o f the replication fork can either be repaired before replication resumes, or the damaged site can be bypassed and repaired after replication is complete. A s D N A damage is constantly occurring and repair can be a slow process that does not always fully complete, lesion bypass is an important mechanism for overcoming replication fork blockages (Lehmarm, 2006). When lesions occur on the lagging strand, one possible outcome is that a gap might be left in the Okazaki fragment that corresponds to the site o f the lesion, and the gap can be repaired after replication is complete by post-replication repair pathways (West et al, 1981). There are also several means to bypass replication fork blocking lesions that do not leave gaps in the nascent D N A strand; these include replication fork regression followed by template switching and translesion synthesis.  A stalled replication fork may be regressed in order to facilitate the bypass o f a lesion that blocks the fork. Fork regression involves the unwinding o f the newly synthesized D N A from the parental strands and the re-pairing of the two nascent strands together to form a 4-way D N A junction (Higgins et al, 1976; Grompone et al, 2004; Flores et al, 2001). The pairing o f the two nascent strands might allow lesion bypass through template switching, as the longer nascent strand could act as a template for the other nascent strand to replicate past damage that is present on the parental strand (Higgins et al, 1976). Following bypass, the D N A junction must be reset in order for replication to continue. Fork resetting is suggested to occur by way o f helicases, such as those in the R e c Q family, that could branch migrate Holliday junction-type structures to reset the fork without homologous recombination (Constantinou et al, 2000; Karow et al, 2000). The RecQ helicases have multiple roles in repair that w i l l be discussed in greater detail i n section 1.2.1. Alternatively, fork resetting might occur through cleavage o f the D N A junction by structure-specific endonucleases such as Mus81 (Whitby et al, 2003). Mus81 can cleave a range of abnormal structures that might form when replication is blocked, including branched substrates, pseudoreplication forks and D-Ioops (Whitby et al, 2003; Hollingsworth & B r i l l , 2004). In addition, Mus81 might be involved in synthesis dependent strand annealing ( S D S A ) , where it could cleave the protruding 3' D N A end, thus preventing the D N A end from reinvading the donor duplex and by doing so, would avoid the formation o f a homologous recombination repair intermediate that could be resolved as a crossover (Robert et al, 2006). Furthermore, Mus81 is synthetically lethal with the RecQ helicase Sgsl in yeast, and this lethality is dependent on the recombination protein R a d S l , indicating that Mus81 acts on some type o f homologous recombination intermediate on which Sgsl also acts (Fabre et al, 2002; Whitby et al, 2003). Although the exact roles o f Mus81 are still under debate, it appears to rescue stalled replication forks, and may also generate meiotic crossovers in yeast (Hollingsworth & B r i l l , 2004).  Damage bypass can also be carried out by the translesion synthesis (TLS) polymerases, which include the Y-family polymerases Polt], P o h ,  PoIk  and R e v l , as well as the B family  polymerase PolÇ (Sweasy et al., 2006). The T L S polymerases have active sites with open structures that allow the accommodation o f altered bases, but cause these polymerases to be error-prone, unlike the normal replicative polymerases (Lehmann, 2006). Different polymerases can act on different lesions. For example, Polr] catalyzes synthesis past cyclobutane dimers that are induced by UV-irradiation (Gibbs et al, 2005); mutations in Polt) cause Xeroderma Pigmentosum-variant ( X P - V ) due to defects in the ability to bypass UV-induced lesions (Sweasy et a/., 2006). O n the other hand, PolÇ and R e v l are required for bypass o f 6-4 photoproducts (Nakajima et al, 2004; Gibbs et al, 2005). During replication, the T L S polymerases associate with the polymerase sliding clamp P C N A (Kannouche et al., 2004; Lehmaim, 2006). When the replication fork encounters damage, P C N A becomes modified by ubiquitylation (Hoege et al, 2002), and this event might stabilize the association of the T L S polymerases with P C N A , thereby facilitating polymerase switching that allows bypass o f the lesion (Bienko et al, 2005; Lehmann, 2006).  1.1.5 Interstrand cross-link repair D N A interstrand cross-links are one o f the most toxic types o f damage because crosslinks prevent the separation o f the two D N A strands o f the double helix, thereby blocking replication and transcription. The pathway that carries out the removal o f interstrand cross-links is known as the Fanconi pathway, based on the identification o f a syndrome characterized by defects in this pathway. Fanconi anemia ( F A ) is a rare cancer susceptibility syndrome associated with various congenital abnormalities, including skeletal malformations, renal abnormalities and hematological problems (Levitus et al, 2006). Individuals with F A commonly exhibit  chromosomal instability and are predisposed to developing various types o f cancers, including acute myeloid leukemia and squamous cell carcinomas (Venkitaraman, 2004). Because o f a high incidence o f bone marrow failure and cancer development, the average life expectancy for F A patients is less than 20 years (Levitus et al, 2006). Currently, 13 F A complementation groups are known, including A , B , C , D 1 / B R C A 2 , D 2 , E , F, G , I, J / B R I P l , L , M and N / P A L B 2 (Taniguchi & D ' A n d r e a , 2006; Reid et al, 2007). A different gene is defective in each complementation group, which is likely one reason why the clinical phenotypes o f the disease are somewhat heterogeneous. The largest subgroup o f F A patients has defects in the F A N C A gene, followed by F A N C C and F A N C G (Levitus et al, 2006). O n the cellular level, F A cells are highly sensitive to agents that cause D N A interstrand cross-links (ICLs), but are not hypersensitive to other types o f D N A damaging agents, although a few groups have reported some controversial sensitivity to ionizing radiation (Levitus et al, 2006; Bigelow etal,  1979;  Carreau et al, 1999; Garcia-Higuera et al, 2001).  A key protein in the F A pathway is F A N C D 2 . In response to detection o f D N A crosslinks by A T R , which recognizes replication fork stalling caused by the cross-link, F A N C D 2 and F A N C I , which make up the ID complex, are monoubiquitylated by the F A core complex (reviewed in Thompson, 2005; Niedemhofer et al, 2005; Smogorzewska et al 2007). The F A core complex is made up o f proteins from multiple complementation groups, including F A N C A , F A N C B , F A N C C , F A N C E , F A N C F , F A N C G , F A N C M , F A N C L , as well as the FA-associated protein F A A P l O O ; the complex also associates with the B l o o m ' s syndrome protein B L M and T O P I I I a (reviewed in Thompson, 2005 and Niedemhofer et al, 2005; Gurtan & D ' A n d r é a , 2006). Monoubiquitylation o f the ID complex appears to be central to this pathway because it causes F A N C D 2 and F A N C I to be recruited to D N A repair foci, where F A N C D 2 interacts with other repair proteins including B R C A l , F A N C D 1 / B R C A 2 and the M R E l 1/RAD50/NBS1  (MRN) complex (Gurtan & D'Andrea, 2006). The XPF/ERCCl endonuclease might be involved in cross-link unhooking, or excision of the cross-link (Levitus et ai, 2006). The FA pathway is further complicated by interactions with other repair pathways. Potentially involved with the FA proteins in cross-link repair are the translesion synthesis polymerases, which might act to repair the ssDNA gap opposite the partially excised cross-link, and the homologous recombination proteins, which might be required to restart the replication fork following removal of the cross-link (Levitus et al, 2006). While the function of the FA core complex is becoming clearer, the components of the pathway and mechanism acting downstream of FANCD2 are not well known (Figure 1.1).  Figure 1.1: A speculative model for interstrand cross-link repair by the Fanconi anemia pathway (adapted from Thompson [2005], Niedernhofer et al. [2005], and Levitus et al. [2006]).  One effector protein thought to act downstream of FANCD2 is the helicase BRIPl/FANCJ, also known as BACHl (Levitus et al, 2006). Patients from Fanconi anemia complementation group J have biallelic mutations in BRIPl/FANCJ individuals with monoallelic BRIPI/FANCJ  (Le\\tus et al, 2005), while  mutations have increased susceptibility to breast  cancer (Seal et al, 2006). Based on the Fanconi anemia phenotype and studies o f  BRIPl/FANCJ  chicken cell lines (Bridge et al, 2005), B R I P l / F A N C J is thought to have a role in the repair o f interstrand cross-links. Data from chicken cell lines indicating that F A N C D 2 foci formation is not affected by the absence o f B R I P l / F A N C J has led to the proposal that B R I P l / F A N C J functions downstream o f F A N C D 2 (Bridge etal, 2005). However, B R I P l F A N C C double mutant cells did not show epistasis in cross-link sensitivity; rather, the double mutants had greater sensitivity to cisplatin treatment than B R I P l / F A N C J or F A N C C single mutants (Bridge et al, 2005). This raises the possibility that B R I P l / F A N C J might function i n a pathway parallel to the core complex and F A N C D 2 . Furthermore, it is not known on what substrates B R I P l / F A N C J acts, or with which proteins it interacts. While much is known about the F A core complex and F A N C D 2 , there are many questions remaining regarding effector proteins and how the F A pathway removes interstrand cross-links.  1.2 Helicases involved in DNA repair Helicases play a major role in D N A metabolism, with functions that range from replication to recombination to transcription. Helicases are implicated in multiple processes that preserve genome integrity, including nucleotide excision repair, homologous recombination repair, cross-link repair and telomere maintenance. Several o f the main repair helicases are described i n the following section.  1.2.1 The RecQ helicases The RecQ helicases are a related group o f evolutionarily conserved helicases that function in multiple roles in D N A repair and genome stability. In yeast there is a single RecQ helicase, while in humans there are five known RecQ helicases. In Saccharomyces  cerevisiae,  absence o f the R e c Q helicase Sgsl causes sensitivity to hydroxyurea treatment, indicating a role 11  during replication (Frei & Gasser, 2000). Sgsl is required for the stable association of polymerases a and s at stalled replication forks, suggesting that Sgsl might function in resolving aberrantly paired structures at stalled replication forks to allow polymerases to access singlestranded D N A (Cobb et al, 2003). S g s l mutants show a 2.5-fold increase in spontaneous crossover, suggesting that this protein contributes to preventing crossover outcomes of repair (Robert et al, 2006). Similar to human B L M , Sgsl might process homologous recombination intermediates through a mechanism termed double Holliday junction dissolution that leads to non-recombinational outcomes (Bennett et al, 1999; Karow et al, 2000; W u & Hickson, 2003; Liberi et al, 2005). Sgsl has a synthetic slow growth or lethal phenotype together with several genes impUcated i n repair, including Mus81 and Srs2, suggesting that S g s l might resolve repair intermediates that other endonucleases or helicases can also resolve (Figure 1.2; Khakhar et al, 2003).  D-loop repair intermediate  Mus81-dependent repair  Sgsl/BLM-dependent repair (double Holliday junction dissolution)  Synthesis dependent strand annealing (Srs2)  Srs2  MusSI-X  A*— Mus81  l^Dk^ K.  Figure 1.2: Proposed roles for Mus81, Sgsl and Srs2 in repair leading to non-recombinational outcomes. MusSl cleaves double Holliday junctions, while Sgsl can unwind the same substrate, together with Top3, by double Holliday junction dissolution. Srs2 removes RadSl from the invading DNA filament to promote synthesis dependent strand annealing (SDSA). MusSI might also function in SDSA to cleave the overhanging 3' DNA end. Figure adaptedfromOsman & Whitby (2007) and Lorenz & Whitby (2006).  Mutations in three of the five human RecQ helicases cause Bloom's syndrome (BLM), Werner's syndrome (WRN), and Rothmund-Thomson syndrome {RECQ4; Mankouri & Hickson, 2004). These syndromes have some overlapping characteristics and are associated with predisposition to cancer and/or premature aging (Khakhar et ai, 2003). Bloom's syndrome is characterized by growth retardation, sunlight sensitivity and predisposition to cancer (Cheok et al, 2005), whereas Werner's syndrome phenotypes include slow growth, premature aging and predisposition to sarcomas (Bohr, 2005).  BLM  cells display an approximately 10-fold elevation in frequency o f sister chromatid  exchanges (SCEs), suggesting that B L M functions to suppress these types o f events (Ellis et ai, 1999; Mankouri & Hickson, 2004). B L M can unwind a variety o f different D N A structures, including forked D N A duplexes, Holliday junction recombination intermediates and G quadruplex D N A (Mohaghegh et al, 2001; Sun et al, 1998; Cheok et al, 2005). B L M also catalyzes regression o f a model replication fork, which might be an early step in the response to replication fork stalling (Machwe et al, 2006). B L M interacts with the recombination protein R A D 5 1 as well as with the single-stranded binding protein R P A and the topoisomerase T O P I I I a , which catalyses the strand passage o f D N A to alter its topology (Cheok et al, 2005; W u et al, 2000; W u & Hickson, 2003). Together, B L M and T O P I I I a function in a process known as double Holliday junction dissolution, whereby homologous recombination repair intermediates are resolved into non-crossover outcomes, explaining the elevated S C E s observed in  BLMCQWS  (Wu & Hickson, 2003).  WRN cells show an increased frequency o f chromosomal rearrangements, such as translocations and deletions (Bachrati & Hickson, 2003; Baynton et al, 2003). L i k e B L M , W R N can unwind blunt duplexes containing single-stranded bubbles, synthetic Holliday junctions and G-quadruplex D N A structures (Mohaghegh et al, 2001 ; Sun et al, 1998); W R N can also catalyze replication fork regression in vitro (Machwe et al., 2006). W R N interacts with R A D 5 2 and N B S l , both o f which function in homologous recombination (Baynton et al, 2003; Cheng et al, 2004). W R N also associates with the K U 7 0 / 8 0 heterodimer, which functions in non-homologous end-joining (Cooper et al, 2000). Therefore, W R N might have roles in two pathways for double strand break repair, and defects in these pathways in the absence o f W R N might explain the chromosomal rearrangements observed in WRN CQWS (Bohr, 2005). The roles  for W R N in double strand break repair appear to be related to a function in telomere maintenance. Both B L M and W R N physically and functionally interact with the telomere repeat binding factor T R F 2 (Opresko et al, 2002), and these RecQ helicases might function in homologous recombination repair pathways that elongate telomere ends (Bohr, 2005). W R N is required for the efficient lagging strand synthesis of G-rich D N A at telomeres, further supporting an important role for W R N at telomeres (Crabbe et al, 2004).  1.2.2 The Srs2 helicase A l o n g with the R e c Q helicases, several other helicases also play significant roles in repair during replication. One o f these is Srs2, a helicase thus far only identified in yeast. Similar to S g s l , S. cerevisiae Srs2 has an important role in preventing hyper-recombination, as srs2 mutants show a 4-fold increase in spontaneous crossovers (Macris & Sung, 2005; Robert et al, 2006). Srs2 may prevent homologous recombination repair by channeling recombination repair intermediates into the Rad6-mediated post-replication repair pathway (Macris & Sung, 2005). In vitro, Srs2 interacts with the Rad51 recombinase and dislodges Rad51 from single-stranded D N A , thereby acting as an anti-recombinase (Krejci et al, 2003; Veaute et al, 2003). Specifically, Srs2 might channel repair intermediates into the synthesis-dependent strandannealing pathway ( S D S A ) by dissociating the invading D N A strand from the D-loop recombination intermediate (Figure 1.2; Ira et al, 2003). Srs2 also physically interacts with sumoylated form o f the polymerase processivity factor P C N A , which is associated with the replication fork (Hoege et al, 2002; Pfander et al, 2005). This interaction links Srs2 to the replication fork, suggesting a role in preventing unwanted homologous recombination as a means for repair during replication.  1.3 C. elegans as a model for the study of DNA repair A s D N A repair pathways are very complex, one means to begin to understand these pathways is to study a model organism, such as Caenorhabditis  elegans. C elegans is one o f the  most well-characterized model organisms. The nematode eats bacteria and can be cultured without difficulty on small petri plates; thus, it is inexpensive and easy to maintain, and requires little space. The nematode takes only three days to develop from egg to fertile adult, and both hermaphrodites and males exist, making genetic manipulation relatively quick and easy. Furthermore, each adult produces 300 progeny, allowing large populations for experiments. Because the nematode has been used as a model organism since the 1970s, there are numerous resources available to aid in research, and many methods for working with the organism have already been established. The genetics o f C. elegans is an especially powerful tool. There are many knockout mutants available from the Caenorhabditis  Genetics Centre, and genetic  interactions in the nematode can be easily assayed by making double mutants and observing the phenotypes o f the animals. The ease o f generating double mutant strains is a great advantage o f C. elegans as a model system, as it takes only a few weeks to make double mutants i f the single mutant strains are available.  Many methods to study D N A replication and repair in C. elegans focus on the germline (Gartner et al., 2004). The germline is an excellent tool for the study o f repair because of the manner in which it is structured. The germline is organized in a temporal-spatial manner such that nuclei can be observed as they progress from mitosis into and through the various stages o f meiosis from distal to proximal regions. The germline begins at the distal tip, where nuclei undergo mitosis in the mitotic zone, which acts as the stem cell compartment for the germline. Beyond the mitotic zone, nuclei progress into the transition zone, where meiosis begins, and nuclei can be observed with a characteristic crescent shape. Nuclei then progress into the  pachytene stage o f meiosis, followed by diplotene and diakinesis, where the highly condensed D N A can be observed as six bivalents that correspond to the homologous chromosome pairs. Following diakinesis, the nuclei cellularize to become oocytes and fertilization occurs (Figure 1.3).  Distal Tip  litotic Zone  Transition Zone 1 r  1 I  Meiotic Pachytene  Oogenesis Early Embryos  Spermatheca  Figure 1.3: Schematic o f the C. elegans germline.  Damaged D N A results in visible changes in the C. elegans germline. For example, in response to D N A damage in the mitotic nuclei, the S-phase checkpoint w i l l arrest these nuclei in order to allow time for repair to occur before replication proceeds (Ahmed et al., 2001). S-phase arrest results in visibly enlarged nuclei and fewer total nuclei in the mitotic zone (Ahmed et al., 2001). Damaged nuclei can also undergo apoptosis in the germline, and apoptotic nuclei can be observed in pachytene at the bend region o f the gonad (Gartner et al., 2004). In addition, fragmented chromosomes and chromatin bridges, as well as other damage can also be observed throughout the germline, allowing for the evaluation o f defects in any D N A repair mutant.  c elegans is particularly useful for the study o f D N A repair because many o f the repair proteins and pathways present in human cells are conserved in the nematode. For example, the tumor suppressor p53, which is the most frequently mutated gene in human cancers (Attardi, 2005), has been identified i n C. elegans as cep-I (Schumacher et al., 2001; D e n y et al, 2001). The BRCAl  and BRCA2 genes implicated in familial breast cancer also have orthologs in C.  elegans, known as brc-1 and brc-2 (Boulton et al., 2004; Martin et al., 2005). Multiple repair helicases are conserved i n C. elegans, including him-6/BLM, wrn-l/WRN, rcq-5/RECQ5,  Y50D7A.2/XPD and  among others (Boulton etal, 2002; O ' N e i l & Rose, 2006; vmw.wormbase.org).  Furthermore, pathways including homologous recombination repair, non-homologous endjoining, mismatch repair, nucleotide excision repair and interstrand cross-link repair are conserved mechanisms for repair in C. elegans ( O ' N e i l & Rose, 2006; CoUis et al, 2006). For example, Meyer et al (2007) have shown that the repair o f U V C - i n d u c e d lesions is kinetically and genetically similar to the repair process in humans.  1.4 polyG/polyC-tracts and the DOG-1 helicase-like protein in C. elegans Aside from being a model for the study o f repair, C. elegans has been useful in defining novel functions for genes involved in replication and repair. In C. elegans, the D O G - 1 helicaselike protein has been shown to play a role in genome maintenance at polyG/polyC-tracts (G/Ctracts; Cheung et al, 2002). Cheung et al (2002) reported that dog-l(gklO)  mutants displayed a  mutator phenotype whereby certain mutant phenotypes frequently appeared. Mapping o f several of the visible notched head mutations showed that deletions beginning in the G/C-tract in the vab-1 gene were responsible for the phenotype. Examination o f a number o f other G/C-tracts in the genome revealed that approximately half o f the G/C-tracts greater than 18 nucleotides in length had deletions in the absence o f D O G - 1 (Cheung et al, 2002). Single stranded G-rich D N A might form secondary structures, such as G-quadruplex, that might create blockages and  cause replication fork stalling or collapse (Arthanari and Bolton, 2001; Woodford et al, 1994). G-rich D N A sequences with the potential to form D N A secondary structures are prevalent in the C. elegans and human genomes (Todd et al. 2005), and occur at telomeres, meiotic recombination hotspots, and gene promoters, among other sites (Simonsson 2001; Maizels, 2006). The C. elegans genome contains 396 G/C-tracts o f 18 or more consecutive guanines (www.wormbase.org), suggesting that there are many sites that are potentially vulnerable to deletions in the absence o f D O G - 1 . The G/C-tract deletions observed by Cheung et al (2002) in dog-1 mutants were typically a few hundred base pairs long, and initiated in the 3' end o f the G tract, extending upstream for various distances. Based on this data, Cheung et al (2002) proposed that D O G - 1 might be involved in unwinding D N A secondary structures that occur in tracts o f 18 or more guanines during lagging strand replication, dog-1 was also identified as the C. elegans gene most similar to human BRIPl  (Cheung et al., 2002), which has since been  identified as the F A gene F^A^CJ(Levitus et al, 2005; Bridge et al, 2005).  1.5 Thesis objectives The goals o f the work documented in this thesis were as follows: 1. ) Use dog-1 as a tool to determine the repair pathways that function at replication forks stalled by G/C-tract secondary structures in C. elegans. 2. ) Investigate the role o f D O G - 1 in D N A repair in C. elegans. 3. ) Characterize the function o f the dog-1/FANCJ-vdatcd C. elegans.  helicase JRH-1 i n D N A repair in  Chapter 1 References Andressoo JO, Hoeijmakers J H , & Mitchell J R (2006) Nucleotide excision repair disorders and the balance between cancer and aging. Cell Cycle 5: 2886-2888 Arthanari H & Bolton P H (2001) Functional and dysfunctional roles o f quadruplex D N A in cells. Chem Biol 8: 221-230 Attardi L D (2005) The role of p53-mediated apoptosis as a crucial anti-tumor response to genomic instability: lessons from mouse models. Mutat Res 569: 145-157 Bachrati C Z & Hickson ID (2003) RecQ helicases: suppressors o f tumorigenesis and premature aging. Biochem J 374: 577-606 Barnes D E , Tomkinson A E , Lehmann A R , Webster A D , & Lindahl T (1992) Mutations in the D N A ligase I gene o f an individual with immunodeficiencies and cellular hypersensitivity to DNA-damaging agents. Cell 69: 495-503 Bau D T , F u Y P , Chen S T , Cheng T C , Y u J C , W u P E , & Shen C Y (2004) Breast cancer risk and the D N A double-strand break end-joining capacity o f nonhomologous end-joining genes are affected by B R C A l . Cancer Res 64: 5013-5019 Baynton K , Otterlei M , Bjoras M , von Kobbe C , Bohr V A , & Seeberg E (2003) W R N interacts physically and functionally with the recombination mediator protein R A D 5 2 . J Biol Chem 278: 36476-36486 Bennett R J , K e c k J L , & Wang J C (1999) Binding specificity determines polarity o f D N A unwinding by the S g s l protein o f S. cerevisiae. J Mol Biol 289: 235-248 Bienko M , Green C M , Crosetto N , Rudolf F, Zapart G , Coull B , Kannouche P, Wider G , Peter M , Lehmann A R , Hofmarm K , & D i k i c I (2005) Ubiquitin-binding domains in Y-family polymerases regulate translesion synthesis. Science 310: 1821-1824 Bigelow S B , Rary J M , & Bender M A (1979) G 2 chromosomal radiosensitivity in Fanconi's anemia. Mutat Res 63:189-199 Bohr V A (2005) Deficient D N A repair in the human progeroid disorder, Werner syndrome. Mutat Res 577: 252-259 Boulton SJ, Martin JS, Polanowska J, H i l l D E , Gartner A , & V i d a l M (2004) B R C A l / B A R D l orthologs required for D N A repair in Caenorhabditis elegans. Curr Biol 14: 33-39 Bridge W L , Vandenberg C J , Franklin RJ, & H i o m K (2005) The B R I P l helicase functions independently o f B R C A l in the Fanconi anemia pathway for D N A crosslink repair. Nat Genet 37: 953-957 Burma S, Chen B P , & Chen D J (2006) Role o f non-homologous end joining (NHEJ) in maintaining genomic integrity. DNA Repair (Amst) 5: 1042-1048  Canman C E & L i m D S (1998) The role o f A T M in D N A damage responses and cancer. Oncogene 17: 3301-3308 Carreau M , A l o n N , Bosnoyan-Collins L , Joenje H , & Buchwald M (1999) Drug sensitivity spectra in Fanconi anemia lymphoblastoid cell lines o f defined complementation groups. Mutat Res 435: 103-109 Cheng W H , von Kobbe C , Opresko P L , Arthur L M , Komatsu K , Seidman M M , Carney JP, & Bohr V A (2004) Linkage between Werner syndrome protein and the M r e l 1 complex via N b s l . J BiolChem 279:21169-21176 Cheok C F , Bachrati C Z , Chan K L , R a l f C , W u L , & Hickson ID (2005) Roles of the Bloom's syndrome helicase in the maintenance of genome stability. Biochem Sac Trans 33: 1456-1459 Cheung I, Schertzer M , Rose A , & Lansdorp P M (2002) Disruption o f dog-1 in Caenorhabditis elegans triggers deletions upstream o f guanine-rich D N A . Nat Genet 31: 405-409 Cobb J A , Bjergbaek L , Shimada K , Frei C , & Gasser S M (2003) D N A polymerase stabilization at stalled replication forks requires M e c l and the RecQ helicase Sgsl. EMBO J 22: 4325-4336 Collis SJ, Barber L J , Ward J D , Martin JS, & Boulton SJ (2006) C. elegans F A N C D 2 responds to replication stress and functions in interstrand cross-link repair. DNA Repair (Amst) 5: 1398-1406 Constantinou A , Tarsounas M , K a r o w J K , Brosh R M , Bohr V A , Hickson I D , & West S C (2000) Werner's syndrome protein ( W R N ) migrates Holliday junctions and co-localizes with R P A upon replication arrest. EMBO Rep 1: 80-84 Cooper M P , Machwe A , Orren D K , Brosh R M , Ramsden D , & Bohr V A (2000) K u complex interacts with and stimulates the Werner protein. Genes Dev 14: 907-912 Crabbe L , Verdun R E , Haggblom C I , & Karlseder J (2004) Defective telomere lagging strand synthesis i n cells lacking W R N hehcase activity. Science 306: 1951-1953 Davies O R & Pellegrini L (2007) Interaction with the B R C A 2 C terminus protects R A D 5 1 - D N A filaments from disassembly by B R C repeats. Nat Struct Mol Biol de Boer J & Hoeijmakers J H (2000) Nucleotide excision repair and human syndromes. Carcinogenesis 21: 453-460 D e n y W B , Putzke A P , & Rothman J H (2001) Caenorhabditis meiosis, and stress resistance. Science 294: 591-595  elegans p53: role in apoptosis,  Ellis N A , Proytcheva M , Sanz M M , Y e T Z , & Gemian J (1999) Transfection o f B L M into cultured bloom syndrome cells reduces the sister-chromatid exchange rate toward normal. Am J Hum Genet (iS: 1368-1374 Esashi F, Galkin V E , Y u X , Egelman E H , & West S C (2007) Stabilization o f R A D 5 1 nucleoprotein filaments by the C-terminal region of B R C A 2 . Nat Struct Mol Biol 14: 468-474  Evans E , Moggs J G , Hwang JR, Egly J M , & Wood R D (1997) Mechanism o f open complex and dual incision formation by human nucleotide excision repair factors. EMBOJW. 6559-6573 Fabre F, Chan A , Heyer W D , & Gangloff S (2002) Alternate pathways involving Sgsl/Top3, M u s S l / Mms4, and Srs2 prevent formation o f toxic recombination intermediates from singlestranded gaps created by D N A replication. Proc Natl Acad Sci USA 99: 16887-16892 Feldmann E , Schmiemann V , Goedecke W , Reichenberger S, & Pfeiffer P (2000) D N A doublestrand break repair in cell-free extracts from Ku80-deficient cells: implications for K u serving as an alignment factor in non-homologous D N A end joining. Nucleic Acids Res 28: 2585-2596 Flores M J , Bieme H , Ehrlich S D , & M i c h e l B (2001) Impairment o f lagging strand synthesis triggers the formation o f a R u v A B C substrate at replication forks. EMBO J 20: 619-629 Frei C & Gasser S M (2000) RecQ-like helicases: the D N A replication checkpoint connection. J Cell Sci 113 (Pt 15): 2641-2646 Garcia-Higuera I, Taniguchi T, Ganesan S, M e y n M S , Timmers C , Hejna J, Grompe M , & D'Andrea A D (2001) Interaction o f the Fanconi anemia proteins and B R C A l in a common pathway. Mol Cell 7: 249-262 Gartner A , MacQueen A J , & Villeneuve A M (2004) Methods for analyzing checkpoint responses in Caenorhabditis elegans. Methods Mol Biol 280: 257-274 Gibbs P E , M c D o n a l d J, Woodgate R , & Lawrence C W (2005) The relative roles in vivo of Saccharomyces cerevisiae P o l eta, Pol zeta, R e v l protein and Pol32 in the bypass and mutation induction o f an abasic site, T - T (6-4) photoadduct and T - T cis-syn cyclobutane dimer. Genetics 169: 575-582 Grompone G , Ehrlich D , & M i c h e l B (2004) Cells defective for replication restart undergo replication fork reversal. EMBO Rep 5: 607-612 Gurtan A M & D'Andrea A D (2006) Dedicated to the core: understanding the Fanconi anemia complex. DNA Repair (Amst) 5: 1119-1125 Harrison J C & Haber J E (2006) Surviving the breakup: the D N A damage checkpoint. Annu Rev Genet 40:209-235 Higgins N P , Kato K , & Strauss B (1976) A model for replication repair in mammalian cells. J Mol Biol 101:417-425 Hoege C , Pfander B , M o l d o v a n G L , Pyrowolakis G , & Jentsch S (2002) RAD6-dependent D N A repair is linked to modification o f P C N A by ubiquitin and S U M O . Nature 419: 135-141 Hoeijmakers J H (2001) Genome maintenance mechanisms for preventing cancer. Nature 411: 366-374 Hollingsworth N M & B r i l l S J (2004) The M u s S l solution to resolution: generating meiotic crossovers without HoUiday junctions. Genes Dev 18: 117-125  Ira G , M a i k o v a A , Liberi G , Foiani M , & Haber J E (2003) Srs2 and Sgsl-Top3 suppress crossovers during double-strand break repair in yeast. Cell 115: 401-411 Kannouche P L , W i n g J, & Lehmann A R (2004) Interaction o f human D N A polymerase eta with monoubiquitinated P C N A : a possible mechanism for the polymerase switch in response to D N A damage. Mol Cell 14: 491-500 Karow J K , Constantinou A , L i J L , West S C , & Hickson ID (2000) The Bloom's syndrome gene product promotes branch migration o f Holliday junctions. Proc Natl Acad Sci USA 91: 65046508 Khakhar R R , Cobb J A , Bjergbaek L , Hickson ID, & Gasser S M (2003) R e c Q hehcases: multiple roles in genome maintenance. Trends Cell Biol 13: 493-501 Krejci L , V a n K o m e n S, L i Y , Villemain J, Reddy M S , K l e i n H , Ellenberger T, & Sung P (2003) D N A helicase Srs2 disrupts the Rad51 presynaptic filament. Nature 423: 305-309 Kunkel T A & Erie D A (2005) D N A mismatch repair. Annu Rev Biochem 74: 681-710 Lehmann A R (2006) Translesion synthesis in mammalian cells. Exp Cell Res 312: 2673-2676 Levitus M , Joenje H , & de Winter JP (2006) The Fanconi anemia pathway o f genomic maintenance. Cell Oncol 28: 3-29 Liberi G , Maffioletti G , Lucca C , Chiolo I, Baryshnikova A , Cotta-Ramusino C , Lopes M , Pellicioli A , Haber J E , & Foiani M (2005) Rad51-dependent D N A structures accumulate at damaged replication forks in sgsl mutants defective in the yeast ortholog o f B L M RecQ helicase. Genes Dev 19: 339-350 Lorenz A & Whitby M C (2006) Crossover promotion and prevention. Biochem Soc Trans 34: 537-541 Machwe A , X i a o L , Groden J, & Orren D K (2006) The Werner and B l o o m syndrome proteins catalyze regression o f a model replication fork. Biochemistry 45: 13939-13946 Macris M A & Sung P (2005) Multifaceted role of the Saccharomyces cerevisiae Srs2 helicase in homologous recombination regulation. Biochem Soc Trans 33: 1447-1450 Maizels N (2006) Dynamic roles for G4 D N A in the biology o f eukaryotic cells. Nat Struct Mol Biol 13: 1055-1059 Mankouri H W & Hickson ID (2004) Understanding the roles o f RecQ helicases in the maintenance o f genome integrity and suppression o f tumorigenesis. Biochem Soc Trans 32: 957958 Martin JS, Winkelmann N , Petalcorin M I , Mcllwraith M J , & Boulton SJ (2005) R A D - 5 1 dependent and -independent roles o f a Caenorhabditis elegans BRCA2-related protein during D N A double-strand break repair. Mol Cell Biol 25: 3127-3139  Matsuura S, Tauchi H , Nakamura A , Kondo N , Sakamoto S, Endo S, Smeets D , Solder B , Belohradsky B H , Der Kaloustian V M , Oshimura M , Isomura M , Nakamura Y , & Komatsu K (1998) Positional cloning o f the gene for Nijmegen breakage syndrome. Nat Genet 19: 179-181 Meyer J N , B o y d W A , A z z a m G A , Haugen A C , Freedman J H , V a n Houten B (2007) Decline o f nucleotide excision repair capacity in aging Caenorhabditis elegans. Genome Biol 8(5): R70. M i r k i n S M (2006) D N A structures, repeat expansions and human hereditary disorders. Curr Opin Struct Biol 16:351-358 Mohaghegh P, K a r o w J K , Brosh Jr R M , J r , Bohr V A , & Hickson I D (2001) The Bloom's and Werner's syndrome proteins are D N A structure-specific helicases. Nucleic Acids Res 29: 28432849 Nakajima S, L a n L , Kanno S, Takao M , Yamamoto K , Eker A P , & Yasui A (2004) U V lightinduced D N A damage and tolerance for the survival o f nucleotide excision repair-deficient human cells. J Biol Chem 279: 46674-46677 Niedemhofer L J , Lalai A S , & Hoeijmakers J H (2005) Fanconi anemia (cross)linked to D N A repair. Cell 123: 1191-1198 O'Donovan A , Davies A A , Moggs J G , West S C , & Wood R D (1994) X P G endonuclease makes the 3' incision in human D N A nucleotide excision repair. Nature 371: 432-435 O'Driscoll M , Ruiz-Perez V L , Woods C G , Jeggo P A , & Goodship J A (2003) A splicing mutation affecting expression o f ataxia-telangiectasia and Rad3-related protein ( A T R ) results in Seckel syndrome. Nat Genet 33: 497-501 O ' N e i l N J & Rose A M (2006) D N A repair. WormBook doi/10.1895/wormbook. 1.54.1, http://wvvw.wormbook.org Opresko P L , von Kobbe C , Laine JP, Harrigan J, Hickson I D , & Bohr V A (2002) Telomerebinding protein T R F 2 binds to and stimulates the Werner and B l o o m syndrome helicases. J Biol Chem 277:41110-41119 Osman F & Whitby M C (2007) Exploring the roles o f M u s S l - E m e l / M m s 4 at perturbed replication forks. DNA Repair (Amst) 6: 1004-1017 Pfander B , Moldovan G L , Sacher M , Hoege C , & Jentsch S (2005) S U M O - m o d i f i e d P C N A recruits Srs2 to prevent recombination during S phase. Nature 436: 428-433 Powell S N & Kachnic L A (2003) Roles of B R C A l and B R C A 2 in homologous recombination, D N A replication fidelity and the cellular response to ionizing radiation. Oncogene 22: 57845791 Reid S, Schindler D , Hanenberg H , Barker K , Hanks S, K a l b R, Neveling K , K e l l y P, Seal S, Freund M , W u r m M , Batish S D , Each F P , Yetgin S, Neitzel H , Ariffin H , Tischkowitz M , Mathew C G , Auerbach A D , Rahman N (2007) Biallelic mutations in P A L B 2 cause Fanconi anemia subtype F A - N and predispose to childhood cancer. Nat Genet 39(2): 162-4  Robert T, Dervins D , Fabre F , & Gangloff S (2006) M r c l and Srs2 are major actors in the regulation o f spontaneous crossover. EMBO J 25: 2837-2846 Sarker A H , Tsutakawa S E , Kostek S, N g C , Shin D S , Peris M , Campeau E , Tainer J A , Nogales E , & Cooper P K (2005) Recognition o f R N A polymerase II and transcription bubbles by X P G , C S B , and T F I I H : insights for transcription-coupled repair and Cockayne Syndrome. Mol Cell 20: 187-198 Schumacher B , Hofmann K , Boulton S, & Gartner A (2001) The C. elegans homolog o f the p53 tumor suppressor is required for D N A damage-induced apoptosis. Curr Biol 11: 1722-1727 Scott SP & Pandita T K (2006) The cellular control o f D N A double-strand breaks. J Ce// Biochem 99: 1463-1475 Seal S, Thompson D , Renwick A , Elliott A , K e l l y P, Barfoot R, Chagtai T, Jayatilake H , Ahmed M , Spanova K , North B , M c G u f f o g L , Evans D G , Eccles D , Breast Cancer Susceptibility Collaboration ( U K ) , Easton D F , Stratton M R , & Rahman N (2006) Truncating mutations in the Fanconi anemia J gene BRIP] are low-penetrance breast cancer susceptibility alleles. Nat Genet 38: 1239-1241 Shivji M K , Podust V N , Hubscher U , & Wood R D (1995) Nucleotide excision repair D N A synthesis by D N A polymerase epsilon in the presence o f P C N A , R F C , and R P A . Biochemistry 34: 5011-5017 Sijbers A M , de Laat W L , A r i z a R R , Biggerstaff M , W e i Y F , Moggs J G , Carter K C , Shell B K , Evans E , de Jong M C , Rademakers S, de Rooij J, Jaspers N G , Hoeijmakers J H , & Wood R D (1996) Xeroderma pigmentosum group F caused by a defect in a structure-specific D N A repair endonuclease. Cell 86: 811-822 Simonsson T (2001) G-quadruplex D N A structures-variations on a theme. Biol Chem 382: 621628 Sugasawa K , N g J M , Masutani C , Iwai S, van der Spek P J , Eker A P , Hanaoka F, Bootsma D , & Hoeijmakers J H (1998) Xeroderma pigmentosum group C protein complex is the initiator o f global genome nucleotide excision repair. Mol Cell 2: 223-232 Smogorzewska A , Matsuoka S, Vinciguerra P, M c D o n a l d E R 3rd, Hurov K E , L u o J, B a l l i f B A , G y g i SP, Hofmann K , D'Andrea A D , Elledge S J (2007) Identification o f the F A N C I protein, a monoubiquitylated F A N C D 2 paralog required for D N A repair. Cell 129(2):289-301. Sun H , Karow J K , Hickson ID, & Maizels N (1998) The Bloom's syndrome helicase unwinds G 4 D N A . J Biol Chem 273: 27587-27592 Sweasy J B , Lauper J M , & Eckert K A (2006) D N A polymerases and human diseases. Radiât Res 166:693-714 Taniguchi T & D'Andrea A D (2006) Molecular pathogenesis o f Fanconi anemia: recent progress. 5/00^/107:4223-4233  Taylor A M , Groom A , & B y r d P J (2004) Ataxia-telangiectasia-like disorder ( A T L D ) - i t s clinical presentation and molecular basis. DNA Repair (Amst) 3: 1219-1225 Thompson L H , H i n z J M , Yamada N A , & Jones N J (2005) H o w Fanconi anemia proteins promote the four Rs: replication, recombination, repair, and recovery. Environ Mol Mutagen 45: 128-142 Todd A K , Johnston M , & Neidle S (2005) Highly prevalent putative quadruplex sequence motifs in human D N A . Nucleic Acids Res 33: 2901-2907 Varga T & A p i a n P D (2005) Chromosomal aberrations induced by double strand D N A breaks. DNA Repair (Amst) 4: 1038-1046 Veaute X , Jeusset J, Soustelle C , Kowalczykowski S C , L e C a m E , & Fabre F (2003) The Srs2 helicase prevents recombination by disrupting Rad51 nucleoprotein filaments. Nature 423: 309312 Venkitaraman A R (2004) Tracing the network connecting B R C A and Fanconi anaemia proteins. Nat Rev Cancer 4: 266-276 Vinson R K & Hales B F (2002) D N A repair during organogenesis. Mutat Res 509: 79-91 Walker J R , Corpina R A , & Goldberg J (2001) Structure o f the K u heterodimer bound to D N A and its implications for double-strand break repair. Nature 412: 607-614 Wang T F , Kleckner N , & Hunter N (1999) Functional specificity o f M u t L homologs in yeast: evidence for three M l h l - b a s e d heterocomplexes with distinct roles during meiosis in recombination and mismatch correction. Proc Natl Acad Sci U SA 96: 13914-13919 West S C , Cassuto E , & Howard-Flanders P (1981) Mechanism o f E. coli R e c A protein directed strand exchanges in post-replication repair o f D N A . Nature 294: 659-662 Whitby M C , Osman F , & D i x o n J (2003) Cleavage o f model replication forks by fission yeast M u s S l - E m e l and budding yeast M u s S l - M m s 4 . J Biol Chem 278: 6928-6935 Woodford K J , H o w e l l R M , & Usdin K (1994) A novel K(+)-dependent D N A synthesis arrest site in a commonly occurring sequence motif in eukaryotes. J Biol Chem 269: 27029-27035 Woodruff R C & Thompson JN,Jr (2003) The role o f somatic and germline mutations in aging and a mutation interaction model o f aging. J Anti Aging Med 6: 29-39 Wormbase website: www.wormbase.org. Release WS178, August 2007 W u L , Davies S L , North P S , Goulaouic H , Riou JF, Turley H , Gatter K C , & Hickson ID (2000) The Bloom's syndrome gene product interacts with topoisomerase III. J Biol Chem 275: 96369644 W u L & Hickson ID (2003) The Bloom's syndrome helicase suppresses crossing over during homologous recombination. Nature 426: 870-874  Wyman C & Kanaar R (2006) D N A double-strand break repair: all's well that ends w e l l . Annu Rev Genet 40: 363-383 Zhuang J, Zhang J, Willers H , Wang H , Chung J H , van Gent D C , Hallahan D E , Powell S N , & X i a F (2006) Checkpoint kinase 2-mediated phosphorylation o f B R C A l regulates the fidelity o f nonhomologous end-joining. Cancer Res 66: 1401-1408  CHAPTER 2^ dog-1 and repair of G/C-tracts in C. elegans  2.1 Introduction Previous work by Cheung et al. (2002) described a role for the dog-1 gene i n maintenance o f polyG/polyC-tracts (G/C-tracts) throughout the C. elegans genome. Cheung et al (2002) proposed that D O G - 1 unwinds secondary structures in poly guanine D N A during lagging strand synthesis, as in the absence o f D O G - 1 , a mutator phenotype characterized by deletions that initiate i n polyguanine tracts is observed. Persisting secondary structures might cause replication fork stalling in dog-1 mutants, which could lead to sequence loss i f the secondary structures are not resolved by some alternative means in the absence o f D O G - 1 . Many possibilities exist for G/C-tract resolution in the absence o f D O G - 1 . Other helicases such as those in the R e c Q family might directly unwind the secondary structure, or the polyguanine structure might cause replication fork stalling, which could require homologous recombination to repair and restart the fork. If the stalled fork leads to a double strand break, either the homologous recombination repair or non-homologous end-joining pathways might be employed for repair. Alternatively, the secondary structure might be bypassed using a translesion synthesis polymerase, or the structure could be treated as a bulky adduct that is repaired through the nucleotide excision repair pathway. The aim o f the experiments described in this chapter was to examine the possible interaction o f dog-1 with genes involved in repair, and to understand which repair pathways are required for G/C-tract maintenance in the absence o f D O G - 1 .  ' A version o f this chapter has been published, see: Youds J L , O ' N e i l N J & Rose A M (2006) Homologous recombination is required for genome stability in the absence o f D O G - 1 in Caenorhabditis elegans. Genetics 173(2): 697-708.  2.2 Materials and Methods 2.2.1 Worm lysis and P C R detection of dog-1 (gklO) A l l worm strains were handled as described previously (Brenner, 1974) and were grovra at 20°C unless otherwise indicated. The strain V C 1 3 dog-1 (gklO) is a knockout allele of dog-1, containing a 2030bp deletion beginning in exon 2 and extending into exon 9, where a premature stop codon is introduced (Cheung et al, 2002). In constructing all double and triple mutant strains, dog-1 (gklO) was detected by single worm lysis and P C R o f the dog-1 gene by the following method. Individual worms were picked into tubes, prepared with 5\i\ lysis buffer ( l O m M T r i s - H C l , 5 0 m M K C l , 2 . 5 m M M g C b , 0.45% N P 4 0 , 0.45% Tween20, 0.01% gelatin, lOOiLig/ml Proteinase K ) , and incubated at -70°C for 10 minutes, 56°C for 1 hour, then 95°C for 15 minutes to inactivate the ProteinaseK enzyme. P C R reagents were added directly to the prepared D N A to make a 25^1 reaction with the following components: I X P C R buffer (50mM Tris p H 8.3, 5 0 m M K C l , 3 5 ^ M Na2HP04, 0.05% Tween-20), 2 . 0 m M M g C b , 0.25mM each d N T P , 0 . 4 m M each primer dogL2, dogR2 and dogM2 (see Appendix 1 for primer sequences) and 1 unit o f Taq polymerase (Invitrogen, N e w England Biolabs). P C R was carried out under the following conditions: 94 °C for 4 minutes followed by 34 cycles o f 94°C for 30 seconds, 55°C for 30 seconds and 72°C for 1 minute 30 seconds, and a final elongation step o f 72''C for 10 minutes. P C R products were visualized by electrophoresis on 1% agarose gel stained with ethidium bromide. Gels were imaged using a GelDoc 2000 imager (Biorad).  Primers dogL2 and R 2 are located external to the gklO deletion, while dogM2 is located within the region deleted in gklO. In wild type animals, dogM2 and L 2 produce a 604bp product, while dogR2 and L 2 produce a 3003bp product, which was rarely observed because the smaller band was preferentially amplified. In gklO animals, the d o g M 2 primer has no  recognition site because it is located in the deleted region, and the dogR2 and L 2 primers produce a product that is 203Obp shorter than the w i l d type product (due to the gklO deletion), resulting in a 972bp band. Animals heterozygous for the gklO deletion were detected by the presence of both the 604bp w i l d type and 972bp gkJO P C R products.  2.2.2 Construction of double mutant strains The strains used in this chapter include: V C 1 3 dog-1 (gklO), C B 1 4 7 9 him-6 V C 1 9 3 him-6 (ok412), V C 1 7 4 wrn-l(gk99), him-9 (el487), T G 9 dpy-13(el84)  (el423),  V C 6 5 5 brd-1 (gk297), T J l cep-1 (gkl38), CB1487  rad-51 (lg8701)/nTl[let-?(m435)],  RB864x;7a-7 (ok698),  RB873 lig-4 (ok716), R B 9 6 4 cku-80 (ok861) and TM\29%fcd-2(tml298).  To prevent the  accumulation o f mutations in the mutator background, dog-1 (gklO) was routinely crossed to N 2 males and the dog-1 deletion was resegregated multiple times. In each case, the homozygous presence o f the gklO deletion was verified by P C R , as described in section 2.2.1.  In constructing all double and triple mutant strains, dog-1 (gklO) males were crossed to the strain containing the mutant gene o f interest. Double heterozygous progeny were allowed to self-fertilize, and their progeny were plated individually. After laying for 48 hours, these animals were lysed, the lysates split into 2 tubes, and P C R s were carried out to test for the homozygous presence o f dog-1 (gklO) and homozygous mutation o f the gene o f interest. In all cases except for him-6(el423)  and him-9(el487),  two external and one internal primers were  designed in order to detect deletion alleles o f the strains listed above (see Appendix 1 for primer sequences). In the case o f him-9(el487),  homozygous him-9 animals were detected by the high  incidence o f males (Him) phenotype among progeny, as well as a set o f P C R s . him-9(el487)  is a  large rearrangement involving the mab-3 gene on chromosome II ( N . O ' N e i l , p. communication), and can be detected by the absence o f P C R product with the use o f primers N O N 5 0 and 53  (where a product is detected i n w i l d type animals), and the presence o f P C R product with the use of primers N O N 5 0 and 87 (where no product is detected in w i l d type animals). In the case o f the point mutation him-6(el423),  homozygous him-6 animals were detected by the H i m  phenotype among progeny.  2.2.3 DAPI staining and measurement of mitotic zone nuclei One day prior to staining, L 4 stage animals were picked onto fresh plates and allowed to age 24 hours. One-day old adults were picked into 10^1 o f M 9 buffer in a watch-glass and 200)j.l of 150nM 4',6-diamidino-2-phenylindole dihydrochloride ( D A P I ; Sigma) in ethanol was added. The animals were placed in the dark and allowed to incubate for 1 hour. Animals were destained by soaking in M 9 buffer overnight at 4°C in a humid chamber. DAPI-stained animals were mounted on 3% agarose slides and viewed at 4 0 X magnification on a Zeiss Axioscope fluorescent microscope. A Retiga 2000R camera (Qimaging) and Openlab 4.0.2 software (Improvision) was used to image the germlines. Using the measure tool i n Adobe Photoshop 7.0, the width o f individual nuclei in the mitotic region (extending from the distal tip to the first transition zone nuclei) o f each gonad arm was measured. For each mitotic region, ten randomly chosen nuclei were measured and in addition, the single largest nucleus was measured.  2.2.4 The vab-1 G/C-tract deletion assay The occurrence o f G/C-tract deletions in all single, double and triple mutants was measured by P C R amplification o f the G/C-tract located within the vab-1 gene on chromosome II. L4-stage animals o f the genotype o f interest were picked to fresh plates 24 hours before D N A preparation. D N A o f individual worms was prepared by the method described in section 2.2.1. G/C-tracts were amplified in each animal by a nested P C R assay modified from Cheung et al. (2002). External P C R reactions were carried out in 25^1 volumes, containing I X G - P C R buffer  (50mM Tris p H 8.3, 35^M Na2HP04, 0.05% Tween-20), 2 . 0 m M M g C b , 0.25mM eacii d N T P and 0.4mM o f eacli primer external primer (vabNl and vabN2, see appendix 1 for sequences). The external P C R program was 94 °C for 4 minutes followed by 34 cycles o f 94°C for 30 seconds, 58°C for 30 seconds and 72°C for 1 minute 30 seconds, and a final elongation step o f 72°C for 10 minutes. One microlitre o f D N A from the external reaction was used as the template for a second internal P C R . The internal P C R reagents were the same as those used for the external P C R , except internal primers (152 and 153, see appendix 1) were used i n place o f the external primers. The internal P C R program was the same as the external program, except primers were annealed at 62°C and the extension time was 1 minute. P C R products were electrophoresed on 1% agarose gels and stained with S Y B R Green (Molecular Probes) for nucleic acid visualization. Gels were imaged using a G e l Doc 2000 (Bio-Rad). Selected deletions were sequenced (Nucleic A c i d and Protein Services, U B C ) in order to confirm specific amplification o f the sequence o f interest and the presence o f deletions initiating at the G/C-tract.  A second G/C-tract was tested to determine whether the increased deletion formation observed in the double and triple mutants was specific to the vab-1 G/C-tract; this second G / C tract is located within the cosmid R144 on chromosome III. The R144 G/C-tract was P C R amplified using the primers 398 and 399 (see Appendix 1). The R144 G/C-tract P C R program was 94 °C for 4 minutes followed by 34 cycles o f 94°C for 30 seconds, 62°C for 30 seconds and 72°C for 1 minute 30 seconds, and a final elongation step o f 72°C for 10 minutes. R144 G / C tract P C R products were visualized in the same manner as those for the vab-1 G/C-tract. A 4.4fold increase in deletions was observed in dog-1; him-9 animals (18 deletions in 125 individuals) compared to dog-1 animals (4 deletions in 123 individuals), indicating that the increased instability is not specific to the vab-1 G/C-tract, but also occurs at other G/C-tracts.  2.2.5 R N A i feeding method for the translesion synthesis polymerases Single colonies ofE. coli containing the F53A3.2 ipolr]; www.wormbase.org) and F22B7.6 (pol/c, www.wormbase.org) R N A i constructs were grown overnight i n L B medium with 50|j.g/mL ampicillin and 12.5jj.g/mL tetracycline. 60)^1 o f the concentrated culture was added drop-wise to N G M plates containing 50|Lig/mL ampicillin, 12.5)^g/mL tetracycline and 0 . 5 m M IPTG. The next day, L4-stage animals were placed on the plates to feed and were transferred to fresh R N A i plates every 2 days. F l progeny were picked to fresh R N A i plates at the L4-stage and were lysed as 1-day old adults 24 hours later for G/C-tract P C R .  2.3 Results 2.3.1 dog-1 functionally interacts with him-6/BLM Because the R e c Q hehcases B L M and W R N can unwind G-quadruplex (G4) D N A in vitro (Huber et al, 2002; Mohaghegh et al, 2001 ; Sun et al, 1998), it was asked whether loss o f function o f the C. elegans 5 Z M ortholog, him-6 (Wicky et al., 2004) or PFZ/V homolog, wrn-1 (Lee et al, 2004), would exacerbate the phenotype o f dog-1 mutant animals, dog-1; wrn-1 animals had normal viability; however, dog-1; him-6 animals displayed reduced viability. To explore the interaction between these two genes, the fitness o f dog-1; him-6 double mutants was assayed by maintaining independent lines o f the animals at 20°C and 25 °C, as higher temperatures intensify the effects o f genome stability mutants in C. elegans (Ahmed & Hodgkin, 2000). Each generation, a single L4-stage animal was transferred to a fresh plate, and a line was scored as inviable when the parent worm was sterile or only laid inviable embryos. A t both 20°C and 25°C, dog-1 mutants survived fairly well, with only 3 o f 20 lines becoming sterile by the lO"' generation at 20°C and 5 o f 20 lines becoming sterile in 10 generations at 25°C. A decline in viability was observed for him-6 animals in which 8 o f 20 lines became sterile at 20°C, and 19 o f 20 lines became sterile by the lO"^ generation at 25 °C, in accordance with data from Grabowski et al (2005). A t 20°C, 18 out o f 20 lines o f dog-1; him-6 animals had become sterile by the lO"" generation (Figure 2.1 A ) . The decline in viability was exacerbated when dogI; him-6 animals were grovm at 25 °C, as all 20 lines had become sterile by the 4"" generation (Figure 2.IB). Thus, dog-1; him-6 double mutants exhibit reduced fitness that is much more severe than either o f the single mutants.  0  5  10  Generation  Generation  Figure 2.1: Viability o f independent lines o f dog-1, him-6 and dog-1; him-6 animals at A ) 20°C and B ) 25°C. Figure adapted from Youds et al (2006).  To examine these double mutants in more detail, the progeny o f dog-1; him-6 and dog-1; wrn-1 animals were scored for viability, embryonic lethality, brood size and the frequency o f spontaneous males (which are X O compared to X X hermaphrodites). The high incidence o f males (Him) phenotype is a result o f frequent losses o f a whole or part o f one X chromosome, and is often observed in D N A damage checkpoint or repair mutants (van Haaften et al, 2004). dog-1; wrn-1 animals had normal viability and did not show elevated levels o f embryonic lethality or males. However, dog-1; him-6 double mutants had significantly fewer viable progeny than him-6 animals (t-test p=2.45''°) and fewer viable progeny than would be expected for an additive effect o f the two genes (average number o f viable progeny observed was 28, compared to 47 expected; Table 2.1 ). The average brood size o f the double mutants was no different from the expected (average brood size observed was 185 compared to 194 expected; Table 2.1), and the reduction in viable progeny could be accounted for by increased embryonic lethality (average percent embryonic lethality observed was 73% compared to 53% expected;  Table 2.1). In dog-1; him-6 broods there was a significantly higher percentage o f male progeny than in him-6 broods (t-test p=2.89"''); the incidence o f males was also higher than expected for an additive effect o f the two genes (average percent males observed was 20% compared to 13%) expected; Table 2.1).  Genotype  N2 (n-15) dog-1 (n=37) him-6 (n=--43) dog-1; him-6 (n=41)  Percent Viable Progeny 99.9 ± 0.05 97.1±0.9 46.9± 2.0 27.6± 1.8  Total Brood Size 271 ± 13 239 ± 7 220 ± 11 1 8 5 ± 13  Percent Embryonic Lethality 0.1 ± 0 . 0 5 2.9 ± 0.9 53.1 ± 2 . 0 72.5 ± 1.8  Percent Males  0.2 ± 0.07 0.2 ± 0.07 12.6 ± 0 . 6 1 9 . 7 ± 1.1  Table 2.1: Phenotypes o f N 2 , dog-1, him-6 and dog-1; him-6 animals, n represents the number of parent worms whose progeny were scored. A l l values are ± standard error o f the mean. Table adapted from Youds et al. (2006).  To further examine the dog-1; him-6 animals, the vital dye S Y T 0 1 2 was used to observe levels o f apoptosis in the single and double dog-1; him-6 mutants. A s reported in section 3.3.2, 0.8 ± 0.2 (n=44 gonad arms) SYT012-stained corpses per gonad arm were observed in N 2 animals, and an average o f 3.3 ± 0.2 (n=77) corpses were observed in dog-1 mutants. In him-6 single mutants, an average o f 4.7 ± 0.3 (n=47) SYT012-stained corpses were present per gonad arm, similar to findings reported by Grabowski et al. (2005). K i m et al. (2005) have shown that the elevated levels o f apoptosis in him-6(ell04)  animals are dependent on the cep-l/p53-  controlled D N A damage checkpoint. In dog-1; him-6 animals, an average o f 4.0 ± 0.3 (n=46) corpses were present per gonad arm, indicating that no more cells are undergoing apoptosis in the double mutant than in either o f the single mutants.  To investigate the basis o f the dog-1; him-6 phenotype with respect to replication, D N A in the germline mitotic zone o f dog-1; him-6 animals was visualized by D A P I staining. In C.  elegans, the germline nuclei undergo mitosis in a syncytium in the distal region o f the gonad. These nuclei are spatially separated from the more proximal germ nuclei undergoing the early stages o f meiosis by a region known as the transition zone, in which nuclei transitioning from mitosis to meiosis have a characteristic crescent shape. When genotoxic stress is present in the mitotic zone, the D N A damage checkpoint w i l l arrest nuclei, presumably in order to allow repair to occur and replication to complete (Gartner et al, 2000). This S-phase arrest is characterized by enlarged nuclei and an overall reduction in the number o f nuclei within the germline mitotic region (Ahmed et al., 2001; Garcia-Muse & Boulton, 2005). A severe mitotic arrest, in which there are only a few very large nuclei that occupy the space o f the entire mitotic zone, is observed in mutants such as atl-l/ATR  or after treatment with genotoxic agents (Garcia-Muse &  Boulton, 2005; A h m e d et al, 2001); this was not the case in dog-1; him-6 animals. However, mitotic nuclei were clearly enlarged in dog-1; him-6 animals when compared to N 2 , dog-1 or him-6 (Figure 2.2).  N2  dog-1  Figure 2.2: Germline mitotic zones o f A ) N 2 , B ) dog-1, C ) him-6, and D ) dog-1; him-6 animals. Figure adapted from Youds et al. (2006).  In order to quantify the size o f germline mitotic nuclei in dog-1; him-6 animals, measurements o f the average diameter o f nuclei within the mitotic regions o f N 2 animals, dog-1  and him-6 single mutants and dog-J; him-6 double mutants were made. N 2 and dog-J animals had similar average mitotic nuclei sizes (Table 2.2; t-test p=0.30). Consistent with him-6 having a role in replication, him-6 animals had a significantly larger average mitotic nuclei size than N 2 animals (Table 2.2; p=0.00086). However, the average size o f nuclei in the mitotic region o f dog-1; him-6 animals was significantly larger than N 2 animals or either o f the single mutants (Table 2.2; p = l . 6 6 e " ' l p=1.37e"'^, p=5.76e"'^ in t-tests w i t h N 2 , dog-I, and him-6, respectively). When the diameter o f the single largest nucleus in each mitotic zone was measured to determine the average largest nucleus size, N 2 and dog-] animals had similar average largest nucleus sizes (Table 2.2; p=0.14). him-6 animals had a slightly greater average largest nucleus size (Table 2.2; p=0.0013 in t-test with N2). Again, the average size o f the largest nucleus in dog-]; him-6 mitotic regions were much larger than those in N2 animals or dog-] or him-6 single mutants (Table 2.2; p=1.60e"'\ p=1.96e''^ p=2.60e"" in t-tests w i t h N 2 , dog-l, and him-6, respectively). The larger size o f mitotic nuclei in dog-]; him-6 animals was interpreted to be a response to replicative stress in dog-]; him-6 double mutants.  Genotype N 2 (n=26) dog-1 (n-26) him-6 (n=24) dog-]; him-6 (n=23)  Average Nuclei Size (itm)  Average Largest Nucleus Size ()xm)  3.14±0.10 3.21 ± 0 . 0 9 3.38±0.10 4.29 ± 0 . 1 4  4.19 ± 0 . 0 6 4.40 ± 0.07 6.12±0.15  4.04 ± 0.08  Table 2.2: Average nuclei and largest nucleus sizes in N 2 , dog-], him-6, and dog-]; him-6 animals, n refers to the number o f mitotic zones in which nuclei were measured. Ten nuclei were measured per mitotic zone. A l l values are ± standard error o f the mean. Table adapted from Youds et al. (2006).  2.3.2 Increased G/C-tract deletions in dog-1; liim-6 animals Previously, D O G - 1 was shown to function in the maintenance o f G/C-tracts. as deletions initiating at these tracts were observed in dog-1 mutants (Cheung et al., 2002). Therefore, it was 38  tested whether or not the decreased viability and replicative stress in dog-1: him-6 animals might be explained by an overlapping function of DOG-1 and H l M - 6 in processing G-rich D N A during replication. If this were the case, G/C-tract deletions would be expected in him-6 animals, and an increased number o f deletions would be observed in dog-1; him-6 animals compared to dog-1 single mutants. The assay for G/C-tract deletions is PCR-based, using sets o f nested primers to ensure specific amplification o f the G/C-tract o f interest. For the assay, the G/C-tract adjacent to exon 5 o f the vab-1 gene located on chromosome II was chosen (Figure 2.3 A , B ) . This tract was previously shown to delete in some dog-1 animals (Cheung et al., 2002). ^Vhole single animals were used for the assay, and as animals have differences in germline proliferation that are dependent on age, G/C-tract amplification was performed only on 1-day old aduhs. B y this means, approximately the same number o f cells was sampled in each animal. Deletions were consistently detected in the vab-1 G/C-tract in 11% o f dog-1 animals assayed, thus allowing the straightforward identification o f increases or decreases in the number o f animals with deletions of this particular G/C-tract in a repair mutant background (Figure 2.3 D).  1  vab-1 t H — •  2  3  0  exon 5 ^  I  _>  4  ~—Qr-~-nt-  I  5  —Q  6  —  7  O  8  1  9  !0  f1--S  1 I  ^ ^og-i 1  1  2  3  4  5  8  7  6  9  10  1kb doj?-1, œpaïf  mutant  5  Figure 2.3: The vab-1 deletion assay. A ) Structure o f the vab-1 gene. B ) Position o f the G / C tract (black box) is just left o f exon 5. External and internal primers are indicated by arrows. C ) A typical G/C-tract deletion brings the primers closer together to produce a smaller P C R product. D) A sample o f the deletions observed by electrophoresis on agarose gel. Deletions were present in 11% o f dog-1 mutants. The frequency o f animals with deletions increased when mutations i n repair genes, such as him-6, were introduced into the dog-1 background. Figure adapted from Youds et al. (2006).  Based on the lack o f genetic interaction between dog-] and wrn-l, it was not expected that an elevated number o f animals with deletions would be observed for dog-]; wrn-] double mutants compared to dog-] single mutants. In fact, dog-]; wrn-] animals showed only a small, non-significant increase in the number o f animals with deletions relative to dog-] mutants (Table 2.3). When him-6 mutants were tested for G/C-tract delefions, none were detected, suggesting that H I M - 6 does not function specifically to unwind G-rich D N A secondary structures i n the presence o f D O G - 1 . However, the number o f dog-]; him-6 animals with deletions was increased 3.2-fold over the number o f dog-] animals with deletions (Table 2.3). This suggests that in the absence o f dog-] function, H I M - 6 prevents the formation o f deletions initiating in G/C-tracts.  Genotype  dog-1 wrn-1 dog-1; wrn-1 him-6 dog-1; him-6 dpy-13 rad-51 dog-1; dpy-13 rad-51 him-9 dog-1; him-9 dog-1; him-9; dpy-13 rad-51 dog-1; him-9; him-6 brd-1 dog-1; brd-1 dog-1; him-9; brd-1 cep-1 dog-1; cep-1 fcd-2 dog-1; fcd-2 poh] ( R N A i ) dog-1; polr/(RNAi) polK(KN Ai) dog-IpoliciRN Ai) xpa-1 dog-1; xpa-1 cku-80 dog-1; cku-80 lig-4 dog-1; lig-4  Number of Animals Assayed 228 108 103 100 157 102 179 108 178 108 212 101 121 100 102 107 120 150 100 108 121 101 119 95 66 54 90 53  Number with Deletions 26 2 20 0 58 1 84 0 65 57 116 2 44 42 0 33 0 50 0 28 1 27 0 15 0 6 0 8  Percentage of Animals with Deletions 11.4 1.9 19.4 0 36.9 0.98 46.9 0 36.5 52.8 42.6 2.0 36.4 42.0 0 30.8 0 33.3 0 25.9 0.82  p-value i n t-test with dog-1  Fold Increase Relative to dog-1 1.0  0.32'  1.7  0.0094'  3.2  0.00012'  4.1  0.0027' 0.17" 0.66"  3.2 4.6 3.7  0.0018' 0.98"  3.2 3.7  0.0040'  2.7  0.0032'  2.9  0.0020'  2.3  26.7 0 15.7  0.025'  2.3  0.70'  1.4  11.1  0.95'  1.0  15.1  0.72'  1.3  Table 2.3: Deletions observed in the vab-1 G/C-tract in various repair mutants in the dog-1 background. Number with deletions is the number o f individual animals that showed one or more deletions in the vab-1 G/C-tract as determined by P C R . ' represents p-value in t-test with dog-1. " represents p-value in t-test with dog-1; him-9. t-tests against other applicable double mutants did not give significant p-values. Table adapted from Youds et al. (2006).  2.3.3 Homologous recombination is required for G/C-tract stability in dog-1 mutants The finding that absence o f H I M - 6 in the dog-1 mutant background leads to an increase in the number o f animals with G/C-tract deletions led to the question o f whether other pathways function to prevent G/C-tract deletions. To test i f homologous recombination repair might be involved in the maintenance o f G/C-tracts in dog-1 mutants, dog-1; dpy-13 rad-51 mutants were constructed. The dpy-13 mutation was used as a phenotypic marker to identify rad-51 homozygotes. R A D - 5 1 is the C. elegans homolog o f the bacterial R e c A protein, which has a crucial role in homologous recombination (Rinaldo et al., 1998). In C. elegans, rad-51 has a maternal effect lethal phenotype. Therefore, dpy-13 rad-51 homozygotes that had segregated from dpy-13 rad-51 hétérozygotes were selected and assayed for deletions. Deletions were very rare in dpy-13 rad-51 animals. The number o f animals with deletions in dog-1; dpy-13 rad-51 mutants was increased more than 4-fold over the number i n dog-1 single mutants (Table 2.3). The number o f animals with deletions was also assayed in him-9 and dog-1; him-9 double mutants, him-9 is the ortholog o f the human XPF gene ( N . O ' N e i l , unpublished data), which is implicated in homologous recombination in yeast and in gene targeting i n mammalian cells, as well as in cross-link repair (Klein, 1988; Niedernhofer et al, 2001 ; Niedernhofer et al., 2004; Sargent etal, 2000; Schiestl & Prakash, 1988; Schiestl & Prakash, 1990). N o deletions were observed in him-9 single mutants. The number o f animals with deletions in dog-1; him-9 double mutants was increased 3.2-fold over dog-1 single mutants (Table 2.3). These results suggest that both H I M - 9 and R A D - 5 1 function in the maintenance o f G/C-tracts in dog-1 animals. While these data support a role for homologous recombination in repair at G/C-tracts in dog-1 animals, this pathway apparently is not sufficient to prevent deletions altogether.  To test whether R A D - 5 1 and H I M - 9 function in the same pathway, a dog-1; him-9; dpy13 rad-51 mutant was constructed and these animals were assayed for the occurrence o f vab-1  G/C-tract deletions. These mutants exhibited a number o f deletions similar to those in dog-1; dpy-13 rad-51 animals (Table 2.3), indicating that R A D - 5 1 and H I M - 9 function in the same pathway. To determine i f H I M - 6 might also function in homologous recombination or act in a separate pathway, dog-1; him-9; him-6 triple mutants were constructed. The triple mutants had increased numbers o f animals with deletions relative to both dog-1; him-9 and dog-1; him-6 (Table 2.3). However, the variability o f the number o f animals with deletions in individual experiments with dog-1; him-9; him-6 ranged from 25.7% to 68.4%), resulting in a high t-test value. The variation i n the number o f animals with deletions measured in the dog-1; him-9; him6 triple mutant was much greater than in any other single, double or triple mutant. Because o f this, it was difficult to assess whether H I M - 9 and H I M - 6 function in the same or different pathways to suppress G/C-tract deletions in dog-1 mutants.  B R C A l and B A R D l have been implicated in the homologous recombinational repair response to stalled replication forks in mammalian cell lines (Scully et al, 1997; Scully et al., 2000). Thus, it was investigated whether brd-1, the ortholog o f human BARDl  (Boulton et al,  2004), might be involved in repair at G/C-tracts in dog-1 mutants. The number o f animals with deletions in dog-1; brd-1 was elevated 3.2-fold over the number o f dog-1 animals with deletions (Table 2.3). Construction o f a dog-1; brd-1; dpy-13 rad-51 mutant was attempted in order to determine whether or not brd-1 was epistatic to rad-51, but these animals were not viable. However, in dog-1; brd-1; him-9 triple mutants, the number o f animals with G/C-tract deletions was not significantly different from the number with deletions in either the dog-1; him-9 or dog1; brd-1 strains (Table 2.3). Based on this data, the inviability o f the dog-1; brd-1; dpy-13 rad51 mutant was not interpreted to indicate that BRD-1 and R A D - 5 1 function in separate repair pathways; rather, checkpoint or other functions o f B R D - 1 might be required in the absence o f  R A D - 5 1 . These results, taken with others, suggest that B R D - 1 , R A D - 5 1 and H I M - 9 function i n homologous recombination repair at G/C-tracts in dog-1 animals.  Evidence from mammalian cell lines has shown that p53 localizes to sites o f stalled replication forks, where it is proposed to modulate homologous recombination repair (Sengupta et al, 2003; Yang et al, 2002). Thus, the effect of absence o f the C. elegans p53 ortholog C E P 1 on deletions in the dog-1 background was tested. The number o f animals with deletions observed i n dog-1 cep-1 double mutants was elevated 2.7-fold over the number in dog-1 animals (Table 2.3). This resuh could suggest that CEP-1 is also involved i n G/C-tract maintenance i n absence o f D O G - 1 . However, it is possible that the increase in the number o f animals with deletions might be due to reduced apoptosis o f cells with damage in the absence o f C E P - 1 .  A s part o f its role in interstrand cross-link repair, a central component o f the Fanconi repair pathway, F A N C D 2 , has been suggested to stabilize stalled replication forks (reviewed in Thompson et al, 2005). I f replication fork stalling occurs in the absence o f D O G - 1 , there might be a need for fork stabilization. Therefore, the effect o f absence o f the C. elegans homolog o f FANCD2,fcd-2  (Dequen et al, 2005; Collis et al, 2006), in the dog-1 background was tested.  In fcd-2 mutants, no G/C-tract deletions were observed. However, a 2.9-fold increase in the number o f animals with deletions was observed in dog-1; fcd-2 double mutants compared to dog1 single mutants (Table 2.3), suggesting that F C D - 2 also has a role in deletion-free repair in the absence o f D O G - 1 .  2.3.4 Translesion synthesis is required for G/C-tract stability in dog-1 mutants Polymerases alternative to the typical replicative polymerases are frequently involved in lesion bypass to allow continued replication past various types o f D N A damage (Lehmann, 2002;  Prakash et al, 2005). In particular, polr| and p o k accumulate at replication forks stalled by D N A damage (Bergoglio et al, 2002; Kannouche et al, 2001 ; Kannouche et al, 2003). A n apparent question was whether these polymerases might also function at G/C-tracts in dog-1 mutants. Because there are currently no translesion synthesis mutants available in C. elegans, R N A i was used to knock-down polr] and p o k i n the dog-1 background and the presence o f G / C tract deletions was assayed. Deletions were rarely observed inpolr^ orpolK(KHA\)  control  animals, but were observed in 25.9% o f dog-1; polrj ( R N A i ) and 26.7%> o f dog-1; polK ( R N A i ) animals; both o f these were significant increases over the number o f dog-1 animals with deletions (Table 2.3). A third translesion synthesis polymerase, rev-1, which preferentially incorporates a C opposite template G (Prakash et al, 2005), also exists in C. elegans and is implicated at stalled forks (Mukhopadhyay et al, 2004). However, the embryonic lethal revi ( R N A i ) phenotype prevented the application o f the assay.  2.3.5 Nucleotide excision repair and non-homologous end-joining are not essential for repair at G/C-tracts If G/C-tract secondary structures that persist in the absence o f D O G - 1 are detected as bulky lesions on the D N A , the structures might be repaired through the nucleotide excision repair ( N E R ) pathway, which is the primary pathway for removal o f bulky adducts. Therefore, it was assayed whether N E R might play roles in either preventing or causing deletions at G / C tracts in absence o f D O G - 1 . The role o f N E R in repair o f G/C-tracts in the dog-1 background was assayed using an xpa-1 mutant, which is the ortholog o f human XPA (Park et al, 2002). X P A has an essential role in N E R . The dog-1 xpa-1 double mutant was constructed; these animals did not exhibit a number o f deletions significantly different from dog-1 single mutants (15.8%); p=0.70; Table 2.3), indicating that N E R is not a principal means for G/C-tract maintenance.  A s an alternative to G/C-tract secondary structures being treated as bullcy lesions, it is possible that the structure causes replication fork stalling that leads to double strand breakage after a subsequent round o f replication. In this instance, the double strand break could be repaired through either the homologous recombination repair or non-homologous end-joining (NHEJ) pathways. To test whether or not N H E J might be active at G/C-tracts in the dog-1 background, mutations in lig-4 and cku-80 were utilized; these genes are homologs o f the mammalian N H E J genes DNA ligase IV and Ku80 (Daley et al., 2005). In dog-1; lig-4 and dog1; cku-80 double mutants, the number o f animals with deletions was not significantly different than the number observed i n dog-1 single mutants (15.1%, p=0.72; 11.1%, p=0.95, respectively; Table 2.3). Furthermore, each o f the dog-1; lig-4 and dog-1; cku-80 double mutant strains appeared to be healthy. These data indicate that N H E J is not a key mechanism for deletion-free repair or deletion formation at G/C-tracts. However, this does not preclude the involvement o f other components o f the N E R or N H E J pathways in the homologous recombination repair that is clearly required for genome stability in absence o f D O G - 1 .  2.3.6 Base pair mutations frequently flank the G/C-tract deletions The above data indicate that each o f the genes him-6, rad-51, him-9, brd-1, cep-1, fcd-2, polfj. and jDo/Arhave roles in preventing deletions at G/C-tracts in the absence o f dog-1. However, the mechanism by which the deletions are formed is unknown. During the course o f the investigation, 30 o f the deletions that were detected in the G/C-tract assay were sequenced (selected sequences are shown in Table 2.4). Nine o f the sequenced deletions contained one or more base pair mutations within a few bases of the deletion site. In most cases, the mutations were single base pair changes, but multiple base pair changes and base pair deletions were also observed. These base pair mutations occurred with the same frequency in dog-1 single mutants  (4 out o f 15 deletions) as in homologous recombination repair mutants in the dog-1 background (5 out o f 15 deletions). This observation suggests that a mutagenic mechanism is involved in the formation o f small deletions in dog-1 single mutants and that the same mechanism acts more frequently when other repair pathways, such as homologous recombination repair, are impaired.  Genotype of animal wild type  Deletion sequence left o f break ^^^gaaaccccccccccccccccccccat  dog-1  w i l d type  gaaac  gaaaccccc  tac--c-gaccacaagttaccgggcc  (lOlbp deletion)  gttgtctacttccgaccacaagttac  "*™^gaaaccccccccccccccccccccat ^ ' " ' ' ^ a g a a a a t t a t a c a a c t a t g c c t g a a t gaaacccccc  (334bp d e l e t i o n )  agaaaattatacaactatgcctgaat  ^*™'gaaaccccccccccccccccccccat ^ ^ ^ ' ^ a a c t t g t c t a c t t c c g a c c a c a a g t t gaaacccccccc  (96bp d e l e t i o n )  ^^™^gaaaccccccccccccccccccccat  dog-1 ; dpy-13 rad-51  wild type  (108bp d e l e t i o n )  gaaacccccccc  dog-1; him-9  wild type  gaatggttcgtgtattccgaatgcgt  ^*™'gaaaccccccccccccccccccccat ^ * ^ ' * c t t g t c t a c t t c c g a c c a c a a g t t a c  dog-1; him-9  wild type  (163bp d e l e t i o n )  ^^'"'^gaaaccccccccccccccccccccat ^^^^'tacttccgaccacaagttaccgggcc  dog-1; brd-1  wild type  tagtatcaaaacgtctggcgagtgcg  "'™^gaaaccccccccccccccccccccat ^ ^ ^ ^ ^ g a a t g g t t c g t g t a t t c c g a a t g c g t  dog-1; brd-1  wild type  ^^''''tggtatcaaaacgtctggcgagtgcg  gaaaccccccccc(233bp deletion)  dog-1  wild type  Deletion sequence right o f break  gaca  {126bp d e l e t i o n )  ^*™'gaaaccccccccccccccccccccat  dog-1; dpy-13 rad-51  gaaacccccc  (271bp d e l e t i o n )  aacttgtctacttccgaccacaagtt  "^'^''caagttaccgggccaaaagagactga caagttaccgggccaaaagagactga  gtggatattctcaaattgccgattcg gtggatattctcaaattgccgattcg  Table 2.4: Sequences around vab-1 G/C-tract deletion breakpoints. Genotype o f animal indicates that from which vab-1 G/C-tract deletion was isolated. W i l d type sequence is shovra above for each deletion. Deletion sequence is shown below with deletion size in brackets. Superscripts indicate the sequence location within the cosmid. Mismatches are indicated by holding. Table adapted from Youds et al. (2006).  2.4 Discussion 2.4.1 DOG-1 is likely to function during replication Genomic instability such as that observed in the dog-1 mutant can have several causes, including defects during replication or transcription. For example, Chavez and Aguilera (1997) showed that blocks in transcriptional elongation in yeast HPRl  mutants are responsible for  genomic instability associated with deletions between direct repeats. More recently, L i and Manley (2005) reported that genome stability can be affected by the absence o f the protein splicing factor A S F / S F 2 because R N A : D N A hybrids form on the nontemplate strand o f transcribed genes, leading to rearrangements. Several lines o f evidence suggest that D O G - 1 functions during replication and that the deletions observed in dog-1 mutants occur during replication and are not transcription-based. First, microarray analysis showed that dog-1 was expressed at 7.67-fold higher levels in wild type animals compared to glp-4 animals (which lack virtually all germline cells), indicating that dog-1 expression is enriched in the germline, where nuclei in the mitotic region are constantly undergoing replication (Reinke et al, 2000).  Second,  the observed deletions always initiate in the G/C-tract and extend i n the same direction, suggesting a mechanism for deletion formation that could be related to lagging strand replication (Cheung et al, 2002). Deletions occur at G/C-tracts that are located at intergenic sites as well as within introns, suggesting that the deletions are not dependent on transcription (Cheung et al, 2002). The data presented here shows that loss of both D O G - 1 and H I M - 6 results in elevated levels o f embryonic lethality in the progeny o f dog-1; him-6 animals. Such an interaction between dog-1 and him-6 would be expected i f these two genes have some parallel or overlapping function. Similar to dog-1, him-6 has elevated expression in the germline (Reinke et al, 2000), and K i m et al. (2005) have shown that it functions in cells undergoing active replication and division, including the mitotic germline stem nuclei and early embryonic cells. Further supporting a role for D O G - 1 during replication, enlarged germline mitotic nuclei were  observed in dog-1; him-6 animals, suggesting replicative stress. Finally, results from the G / C tract deletion assay demonstrate that homologous recombination repair and translesion synthesis polymerase genes, which function during replication (Barbour & X i a o , 2003; Lehmann, 2002; Lundin et al, 2003; Prakash et al, 2005; Saleh-Gohari et al, 2005), play an important role in preventing G/C-tract deletions in the dog-1 mutant. Thus, it is probable that the c?og-7-dependent deletions are formed during or immediately following replication.  2.4.2 Homologous recombination and translesion syntliesis are required for genome stability in dog-1 animals Several reports have indicated that homologous recombination functions at stalled or collapsed replication forks to carry out error-free repair (Barbour & X i a o , 2003; Lundin et al, 2003; Saleh-Gohari et al, 2005). The data presented here show that homologous recombinationassociated proteins function at G/C-tracts in C. elegans dog-1 mutants. Absence of the R A D 5 1 , X P F and B A R D l homologs, R A D - 5 1 , H I M - 9 and B R D - 1 in the dog-1 background led to an increase in the number o f animals with G/C-tract deletions, suggesting that these three proteins promote deletion-free repair at G/C-tracts. Epistasis experiments place B R D - 1 , R A D - 5 1 and H I M - 9 in the same repair pathway.  In C. elegans, as in mammals, C E P - l / p 5 3 is required for D N A damage-induced apoptosis (Derry et al, 2001 ; Schumacher et al, 2001). Recent mammalian studies have suggested that p53 can also modulate homologous recombination repair during replication (Sengupta et al, 2003; Sengupta & Harris, 2005). This modulation o f repair might be related to the ability of p53 to recognize mismatches within heteroduplex, and thereby suppress erroneous repair (Dudenhoffer et al, 1998; Y u n et al, 2004), or to its ability to regulate the processing of Holliday junctions by B L M (Yang et al, 2002). A 2.7-fold increase was observed in the number  of dog-l cep-1 animals with deletions compared to dog-1 animals. This could be interpreted to suggest that CEP-1 promotes maintenance o f G/C-tracts in dog-1 mutants. However, data documented in section 3.3.2 shows that CEP-1-mediated apoptosis is increased i n dog-1 mutants. Based on this data, another possible interpretation is that more deletions were observed i n dog-I cep-1 animals because C E P - 1 was not present to activate apoptosis in cells with D N A damage resulting from absence o f D O G - 1 . Therefore, the individual animals sampled might have contained cells with deletions that would normally have undergone apoptosis, thus leading to higher numbers o f animals with deletions. This is unlikely to be the case because i f it were, the assay should have produced a noticeable increase in germline deletions in which no w i l d type P C R band was visible (i.e., all cells in the entire animal having the same deletion), compared to animals that are mixed populations o f deletions and w i l d type sequence.  However, the  possibility that reduced apoptosis is the reason for the elevated frequency o f dog-1 cep-1 animals with deletions cannot be ruled out.  In addition to the homologous recombination repair-implicated genes, mutation  offcd-2  in the dog-I background led to an increase in deletion frequency, fcd-2 is the C. elegans homolog o f human FANCD2,  and the monoubiquitylated F C D - 2 protein is reportedly recruited  to sites o f stalled replication forks and interstrand cross-links (Dequen et al., 2005; Collis et al., 2006). Using the G/C-tract deletion assay, the frequency o f deletions in dog-1; fcd-2 mutants was similar to the frequencies detected in homologous recombination repair mutants in the dog-1 background. In humans, F A N C D 2 functions in interstrand cross-link repair (reviewed in Levitus et al., 2006), and also responds to D N A replication stress following treatment with hydroxyurea or aphidicolin (Howlett et al., 2005). Following D N A damage, F A N C D 2 becomes monoubiquitylated and colocalizes with homologous recombination repair proteins R A D 5 1 and B R C A l at sites o f damage (Taniguchi et al., 2002); this F A N C D 2 activity might help to stabilize  the replication fork to allow repair through error-free processes such as homologous recombination repair (reviewed in Thompson et ai, 2005). There are indications that F A N C D 2 might promote homologous recombination repair, as reduced levels o f homology-directed repair are observed in the absence o f F A N C D 2 , though the reduction i n recombination repair is small compared to that observed i n the absence o f R A D 5 1 (Nakanishi et al, 2005). In C. elegans, F C D - 2 might play a similar role in promoting homologous recombination repair in response to replication stress. The finding o f increased number o f animals with deletions in dog-1; fcd-2 could be explained by reduced, though not eliminated, homologous recombination repair i n the absence o f F C D - 2 . However, no functional linkage between F C D - 2 and the homologous recombination repair pathway has been demonstrated in C. elegans, as the gene has only recently been characterized i n this organism. Therefore, it is equally possible that replication forks stalled by G/C-tract secondary structures could be repaired through F C D - 2 as part o f a pathway that is separate from the homologous recombination repair machinery. In either case, the increased deletions observed in dog-1 ; fcd-2 animals suggest that F C D - 2 is involved in deletion-free repair at G/C-tracts.  The G/C-tract deletion assay was also used to test whether translesion synthesis is required for genome stability i n dog-1 mutants. Knock-down o f either polrjor pol/cin the dog-1 background increased the number o f animals with deletions. Recent reports have shown that polrj, in addition to a function in translesion synthesis, interacts with R A D 5 1 and has the ability to extend D-loops as part o f the reinitiation o f D N A synthesis by homologous recombination repair (Kawamoto et al., 2005; Mcllwraith et al., 2005). Thus, the role o f polrj'm dog-1 mutants might be related to its function in homologous recombination. It is possible that the increase in the number o f animals with deletions observed in dog-1; polrj ( R N A i ) and dog-1 ; polK {KHM) animals was not as high as the number seen in homologous recombination mutants in the dog-1 51  background because R N A i was used to knock-down these polymerases. Therefore, low levels o f the polîj and poltc gene products might have been present.  2.4.3 HIM-6 could function in RAD-51-dependent or -independent pathways The role for H I M - 6 at the replication fork is not clearly defined, as H I M - 6 might function either to unwind G/C-tract secondary structures or to resolve stalled forks through a R A D - 5 1 dependent or independent mechanism in dog-] mutants. In vitro, B L M preferentially unwinds G4 D N A , suggesting that it might prevent replication fork stalling by unwinding these structures in vivo (Huber et al., 2002; Mohaghegh et al., 2001 ; Sun et al., 1998). Because no G/C-tract deletions were observed in him-6 animals, it can be concluded that in the presence o f D O G - 1 , H I M - 6 does not play a key role in unwinding G/C-tract secondary structures. However, it is possible that H I M - 6 might unwind these structures when D O G - 1 is absent. It has also been proposed that B L M plays a role in replication fork restoration by catalyzing branch migration to prevent the need for homologous recombination repair (Karow et al., 2000). In addition, B L M interacts with R A D 5 1 ( W u et al., 2001), and can act on recombination intermediates such as Holliday junctions in vitro (Karow et al., 2000; W u & Hickson, 2003). B L M may resolve these intermediates through a non-recombinational mechanism termed double Holliday junction dissolution, whereby exchange o f flanking sequences is prevented ( W u & Hickson, 2003). Wicky et al. (2004) proposed that H I M - 6 functions downstream o f R A D - 5 1 in C. elegans. The findings presented here show that the number o f G/C-tract deletions in dog-I; him-6 animals was increased, indicating that deletion-free repair is compromised in the double mutants. dog-I; him9; him-6 triple mutants had highly variable viability and numbers o f animals with deletions, in contrast to other single, double and triple mutants. This variability made it difficult to conclusively assign a role to H I M - 6 , as epistasis in the frequency o f animals with deletions was unclear. H I M - 6 might function downstream o f R A D - 5 1 to resolve G/C-tract repair intermediates  through a non-recombinational mechanism, similar to B L M in human cells. Another possibility is that H I M - 6 functions in a separate pathway that promotes fork restoration independent o f R A D - 5 1 in dog-I mutants.  In S. cerevisiae, the H I M - 6 / B L M homolog Sgsl may have both R A D 5 1 dependent and independent roles ( l i & B r i l l , 2005). One further clue to the role o f him-6 in C. elegans is the finding that him-6 has a genetic interaction with dog-I, as evidenced by the synthetic sick phenotype and replication stress observed in dog-1; him-6 double mutants. Other homologous recombination repair mutants, such as brd-I and him-9, together with the dog-1 mutant were healthy, indicating that no functional interaction was present. However, absence of any o f the homologous recombination repair genes or him-6 in the dog-1 background led to elevated numbers o f animals with deletions. Thus, him-6 has an interaction with dog-I that is different from the genes in the homologous recombination repair pathway. This suggests that him-6 has roles in common with homologous recombination repair, such as preventing deletion formation through a R A D - 5 1 dependent mechanism, and also independent from R A D - 5 1 , such as preventing replication stress, possibly by stabilizing the replication fork across repetitive sequences or resolving stalled forks through a R A D - 5 1 independent mechanism. Therefore, it is likely that H I M - 6 functions in multiple roles in C. elegans, as S g s l does in S. cerevisiae.  This  interpretation could explain the unclear epistasis in the number o f animals with deletions observed for dog-1; him-9; him-6 triple mutants.  2.4.4 Loss of N E R or N H E J does not affect deletion formation Neither N E R nor N H E J are required for preventing deletions when D O G - 1 is absent because the number o f animals with deletions did not increase when essential elements o f these mechanisms were knocked out. Although N E R and N H E J do not appear to play important roles  at G/C-tracts in dog-J mutants, components o f the N E R and N H E J pathways might function to maintain genome stability in dog-1 animals. It is well-established that certain components o f repair pathways can function in different repair mechanisms. For instance, X P F / RadlO is a component o f the N E R pathway, but is also involved in the trimming o f non-homologous ends from double strand breaks during synthesis-dependent strand annealing in yeast (Ivanov & Haber, 1997) and functions in cross-link repair in mammalian cell lines (Hoy et al., 1985; Kuraoka et al., 2000). The finding that absence o f the C. elegans X P F homolog H I M - 9 i n the dog-1 background elevates the frequency of animals with G/C-tract deletions, but that absence o f X P A - 1 does not change the frequency o f animals with deletions illustrates the point that while N E R is not specifically involved in G/C-tract maintenance, other genes that act in multiple pathways including N E R are involved. Though neither N E R nor N H E J appear to be the main mechanisms for preventing deletions at G/C-tracts, it is entirely possible that other components of these pathways do function in repair at G/C-tracts through other pathways such as homologous recombination repair.  2.4.5 A model for repair at G/C-tracts Because D O G - 1 appears to function during replication and both homologous recombination and translesion synthesis are required for genome stability in the absence o f D O G - 1 , a likely scenario is that persisting G/C-tract secondary structures can lead to replication fork stalling i n dog-1 mutants. D O G - 1 could be the primary means for preventing replication fork stalling by unwinding secondary structures, and in the absence o f D O G - 1 , homologous recombination repair or translesion synthesis can carry out repair and allow replication to proceed (Figure 2.4). H I M - 6 may act downstream o f R A D - 5 1 , resolving repair intermediates to give non-recombinational outcomes, but it is also possible that H I M - 6 functions to repair the stalled fork through a R A D - 5 1 independent mechanism, or H I M - 6 might function in both of  these roles. Mutagenic repair, which results in the observed small deletions, occurs at a low frequency in the absence o f D O G - 1 , but when both D O G - 1 and the deletion-free repair pathways are impaired, the deletion-generating mechanism functions more frequently.  G-rich D N A s e c o n d a r y structure forms during replication  /  \  no D O G - 1 present: alternative r e p a i r of s t a l l e d fork r e q u i r e d  /  \  deletion-prone repair (mechanism unknown)  structure r e s o l v e d by D O G - 1  deletion-free repair  ^  /  X  trans-lesion synthesis  \  \  replication fork resolution by HIM-6  RAD-51  HIM-9 recombinational outcome  HIM-6 non-recombinational outcome  Figure 2.4 A model for repair o f secondary structures forming at G/C-tracts. Normally, D O G - 1 would be present to unwind the secondary structure, but in its absence, the secondary structure persists, causing replication fork stalling. The stalled fork can be repaired in a deletion-free manner through translesion synthesis by either polti or p o k , or through homologous recombination repair, involving B R D - 1 , R A D - 5 1 , H I M - 9 and possibly H I M - 6 . H I M - 6 might also carry out deletion-free repair through a R A D - 5 1 independent mechanism. Alternatively, the deletion-prone repair pathway might act through an unknown mechanism to restore the replication fork, leaving a deletion at the site o f the secondary structure. Figure adapted from Youds etal. (2006).  O f the deletions obtained in the G/C-tract deletion assay, 30 were sequenced and about one-third o f these had base pair mutations within a few bases o f the deletion site. The base pair mutations occurred with equal frequency in dog-] single mutants as in homologous  recombination repair mutants in the dog-1 baclcground, suggesting that the deletions are formed by the same mechanism in dog-1 mutants as in the absence o f both D O G - 1 and homologous recombination. The deletions might occur through this mutagenic mechanism when secondary structures persist due to the absence o f D O G - 1 and homologous recombination repair and/or translesion synthesis are unable to repair the stalled fork. None o f the repair genes tested i n the G/C-tract deletion assay led to a decrease in the frequency o f animals with deletions relative to the dog-1 frequency; therefore, the pathway for deletion formation remains unknown. Because none o f the pathways tested were responsible for deletion formation, it is likely that the deletions are formed through a mechanism that is directly linked to replication. For instance, the deletions could be formed i f a gap in the lagging strand occurs opposite the secondary structure, followed by joining o f the ends opposite the gap. This type o f mechanism would account for the variable size o f the deletions, as the size o f the gap might vary depending on the location of the secondary structure within the lagging strand Okazaki fragment, and whether there are other replication fork stalls nearby.  The data presented in this chapter shows that him-6 and other homologous recombination repair-implicated genes, including brd-1/BARDl,  rad-51/RAD51,  synthesis polymerases polrj and polK, as well as cep-l/p53  him-9/XPF, and the translesion  and fcd-2/FANCD2  promote deletion-  free repair i n the absence o f D O G - 1 . When any o f H I M - 6 , R A D - 5 1 , H I M - 9 , B R D - 1 , polri, p o k , CEP-1 or F C D - 2 are impaired in the dog-1 background, animals exhibit an increase in small deletions initiating in G/C-tracts. These findings underscore the utility o f C. elegans for the study o f D N A repair, as individual animals can be assayed to measure the contribution o f various repair pathways to genome stability, and highlight the multiple layers o f repair that are present in cells to maintain genome integrity.  Chapter 2 References Ahmed S, A l p i A , Hengartner M O , & Gartner A (2001) C. elegans R A D - 5 / C L K - 2 defines a new D N A damage checkpoint protein. Curr Biol 11: 1934-1944 Ahmed S & Hodgkin J (2000) M R T - 2 checkpoint protein is required for germline immortality and telomere replication in C elegans. Nature 403: 159-164 Barbour L & X i a o W (2003) Regulation o f alternative replication bypass pathways at stalled replication forks and its effects on genome stability: a yeast model. Mutat Res 532: 137-155 Bergoglio V , Bavoux C , Verbiest V , Hoffmann JS, & Cazaux C (2002) Localisation o f human D N A polymerase kappa to replication foci. J Cell Sci 115: 4413-4418 Boulton S J, Martin JS, Polanowska J, H i l l D E , Gartner A , & V i d a l M (2004) B R C A l / B A R D l orthologs required for D N A repair in Caenorhabditis elegans. Curr Biol 14: 33-39 Brenner S (1974) The genetics o f Caenorhabditis elegans. Genetics 77: 71-94 Chavez S & Aguilera A (1997) The yeast H P R l gene has a functional role in transcriptional elongation that uncovers a novel source o f genome instability. Genes Dev 11: 3459-3470 Cheung I, Schertzer M , Rose A , & Lansdorp P M (2002) Disruption o f dog-1 in Caenorhabditis elegans triggers deletions upstream o f guanine-rich D N A . Nat Genet 31: 405-409 Collis SJ, Barber L J , Ward J D , Martin JS, & Bouhon SJ (2006) C. elegans F A N C D 2 responds to replication stress and functions in interstrand cross-link repair. DNA Repair (Amst) 5: 1398-1406 Daley J M , Palmbos P L , W u D , & Wilson T E (2005) Nonhomologous end joining i n yeast. Annu Rev Genet 39: 431-451 Dequen F , St-Laurent JF, Gagnon S N , Carreau M , Desnoyers S (2005) The Caenorhabditis elegans FancD2 ortholog is required for survival following D N A damage. Comp Biochem Physiol B Biochem Mol Biol 141(4): 453-60 D e n y W B , Putzke A P , & Rothman J H (2001) Caenorhabditis meiosis, and stress resistance. Science 294: 591-595  elegans p53: role in apoptosis,  Dudenhoffer C , Rohaly G , W i l l K , Deppert W , & Wiesmuller L (1998) Specific mismatch recognition in heteroduplex intermediates by p53 suggests a role in fidelity control o f homologous recombination. Mol Cell Biol 18: 5332-5342 Garcia-Muse T & Boulton SJ (2005) Distinct modes o f A T R activation after replication stress and D N A double-strand breaks in Caenorhabditis elegans. EMBOJ24: 4345-4355 Gartner A , Milstein S, A h m e d S, Hodgkin J, & Hengartner M O (2000) A conserved checkpoint pathway mediates D N A damage-induced apoptosis and cell cycle anest in C. elegans. Mol Cell 5: 435-443  Gerring S L , Spencer F , & Hieter P (1990) The C H L 1 ( C T F 1) gene product of Saccharomyces cerevisiae is important for chromosome transmission and normal cell cycle progression in G 2 / M . £ M 5 0 J 9 : 4347-4358 Grabowski M M , Svrzikapa N , & Tissenbaum H A (2005) B l o o m syndrome ortholog H I M - 6 maintains genomic stability in C. elegans. Mech Ageing Dev 126: 1314-1321 Howlett N G , Taniguchi T, Durkin S G , D'Andrea A D , & Glover T W (2005) The Fanconi anemia pathway is required for the D N A replication stress response and for the regulation o f common fragile site stability. Hum Mol Genet 14: 693-701 Hoy C A , Thompson L H , Mooney C L , & Salazar E P (1985) Defective D N A cross-link removal in Chinese hamster cell mutants hypersensitive to bifunctional alkylating agents. Cancer Res 45: 1737-1743 Huber M D , Lee D C , & Maizels N (2002) G4 D N A unwinding by B L M and S g s l p : substrate specificity and substrate-specific inhibition. Nucleic Acids Res 30: 3954-3961 Ii M & B r i l l S J (2005) Roles o f SGS 1, M U S 8 1 , and R A D 5 1 i n the repair o f lagging-strand replication defects in Saccharomyces cerevisiae. Curr Genet 48: 213-225 Ivanov E L & Haber J E (1995) R A D l and R A D I O , but not other excision repair genes, are required for double-strand break-induced recombination in Saccharomyces cerevisiae. Mol Cell 5 / 0 / 15:2245-2251 Kannouche P, Broughton B C , V o l k e r M , Hanaoka F , Mullenders L H , & Lehmann A R (2001) Domain structure, localization, and function o f D N A polymerase eta, defective in xeroderma pigmentosum variant cells. Genes Dev 15: 158-172 Kannouche P, Fernandez de Henestrosa A R , Coull B , V i d a l A E , Gray C , Zicha D , Woodgate R , & Lehmann A R (2003) Localization o f D N A polymerases eta and iota to the replication machinery is tightly co-ordinated in human cells. EMBO J 22: 1223-1233 Karow J K , Constantinou A , L i J L , West SC, & Hickson ID (2000) The Bloom's syndrome gene product promotes branch migration o f Holliday junctions. Proc Natl Acad Sci U SA 91: 65046508 Kawamoto T, A r a k i K , Sonoda E , Yamashita Y M , Harada K , K i k u c h i K , Masutani C , Hanaoka F, Nozaki K , Hashimoto N , & Takeda S (2005) Dual roles for D N A polymerase eta in homologous D N A recombination and translesion D N A synthesis. Mol Cell 20: 793-799 K i m Y M , Yang I, Lee J, & K o o H S (2005) Deficiency o f Bloom's syndrome protein causes hypersensitivity o f C elegans to ionizing radiation but not to U V radiation, and induces p53dependent physiological apoptosis. Mol Cells 20: 228-234 K l e i n H L (1988) Different types o f recombination events are controlled by the R A D l and R A D 5 2 genes o f Saccharomyces cerevisiae. Genetics 120: 361-1)11  Kuraoka I, Kobertz W R , A r i z a R R , Biggerstaff M , Essigmann J M , & Wood R D (2000) Repair o f an interstrand D N A cross-link initiated by E R C C l - X P F repair/recombination nuclease. J Biol Chem 275:26632-26636 Lee SJ, Y o o k JS, H a n S M , & K o o H S (2004) A Werner syndrome protein homolog affects C. elegans development, growth rate, life span and sensitivity to D N A damage by acting at a D N A damage checkpoint. Development 131: 2565-2575 Lehmann A R (2002) Replication o f damaged D N A in mammalian cells: new solutions to an old problem. Mutat Res 509: 23-34 Levitus M , Joenje H , & de Winter JP (2006) The Fanconi anemia pathway o f genomic maintenance. Cell Oncol 28: 3-29 L i X & Manley J L (2005) Inactivation o f the S R protein splicing factor A S F / S F 2 results in genomic instability. Cell 122: 365-378 Lorenz A & Whitby M C (2006) Crossover promotion and prevention. Biochem Soc Trans 34: 537-541 Lundin C , Schultz N , Arnaudeau C , Mohindra A , Hansen L T , & Helleday T (2003) R A D 5 1 is involved i n repair o f damage associated with D N A replication in mammalian cells. J Mol Biol 328:521-535 Mcllwraith M J , Vaisman A , L i u Y , Fanning E , Woodgate R , & West S C (2005) Human D N A polymerase eta promotes D N A synthesis from strand invasion intermediates of homologous recombination. Mol Cell 20: 783-792 Mohaghegh P, K a r o w J K , Brosh Jr R M , J r , Bohr V A , & Hickson ID (2001) The Bloom's and Werner's syndrome proteins are D N A structure-specific helicases. Nucleic Acids Res 29: 28432849 Mukhopadhyay S, Clark D R , Watson N B , Zacharias W , & McGregor W G (2004) R E V l accumulates i n D N A damage-induced nuclear foci in human cells and is implicated i n mutagenesis by benzo[a]pyrenediolepoxide. Nucleic Acids Res 32: 5820-5826 Nakanishi K , Yang Y G , Pierce A J , Taniguchi T, Digweed M , D'Andrea A D , Wang Z Q , & Jasin M (2005) Human Fanconi anemia monoubiquitination pathway promotes homologous D N A repair. Proc Nad Acad Sci U S A 102: 1110-1115 Niedernhofer L J , Essers J, Weeda G , Beverloo B , de Wit J, Muijtjens M , Odijk H , Hoeijmakers J H , & Kanaar R (2001) The structure-specific endonuclease E r c c l - X p f is required for targeted gene replacement in embryonic stem cells. EMBO J 20: 6540-6549 Niedernhofer L J , Odijk H , Budzowska M , van Drunen E , Maas A , Theil A F , de Wit J, Jaspers N G , Beverloo H B , Hoeijmakers J H , & Kanaar R (2004) The structure-specific endonuclease E r c c l - X p f is required to resolve D N A interstrand cross-link-induced double-strand breaks. Mol Cell Biol 24: 5776-5787  Park H K , Y o o k JS, K o o H S , Choi IS, & A h n B (2002) The Caenorhabditis elegans X P A homolog o f human X P A . Mol Cells 14: 50-55 Prakash S, Johnson R E , & Prakash L (2005) Eukaryotic translesion synthesis D N A polymerases: specificity of structure and function. Annu Rev Biochem 74: 317-353 Reinke V , Smith H E , Nance J, Wang J, V a n Doren C , Begley R , Jones SJ, Davis E B , Scherer S, Ward S, & K i m S K (2000) A global profile of germline gene expression in C elegans. Mol Cell 6:605-616 Rinaldo C , Ederle S, Rocco V , & L a Volpe A (1998) The Caenorhabditis elegans R A D 5 1 homolog is transcribed into two alternative m R N A s potentially encoding proteins o f different sizes. Mol Gen Genet 260: 289-294 Saleh-Gohari N , Bryant H E , Schultz N , Parker K M , Cassel T N , & Helleday T (2005) Spontaneous homologous recombination is induced by collapsed replication forks that are caused by endogenous D N A single-strand breaks. Mol Cell Biol 25: 7158-7169 Sargent R G , Meservy J L , Perkins B D , K i l b u m A E , Intody Z , Adair G M , Nairn R S , & W i l s o n J H (2000) Role of the nucleotide excision repair gene E R C C l i n formation o f recombinationdependent rearrangements in mammalian cells. Nucleic Acids Res 28: 3771-3778 Schiestl R H & Prakash S (1990) R A D I O , an excision repair gene o f Saccharomyces cerevisiae, is involved in the R A D l pathway o f mitotic recombination. Mol Cell Biol 10: 2485-2491 Schiestl R H & Prakash S (1988) R A D l , an excision repair gene o f Saccharomyces also involved in recombination. Mol Cell Biol 8: 3619-3626  cerevisiae, is  Schumacher B , Hofmann K , Boulton S, & Gartner A (2001) The C. elegans homolog of the p53 tumor suppressor is required for D N A damage-induced apoptosis. Curr Biol 11: 1722-1727 Scully R, Chen J, Ochs R L , Keegan K , Hoekstra M , Feunteun J, & Livingston D M (1997) Dynamic changes o f B R C A l subnuclear location and phosphorylation state are initiated by D N A damage. Cell 90: 425-435 Scully R, Puget N , & Vlasakova K (2000) D N A polymerase stalling, sister chromatid recombination and the B R C A genes. Oncogene 19: 6176-6183 Sengupta S & Harris C C (2005) p53: traffic cop at the crossroads o f D N A repair and recombination. Nat Rev Mol Cell Biol 6: 44-55 Sengupta S, L i n k e SP, Pedeux R, Yang Q, Famsworth J, Garfield S H , Valerie K , Shay J W , Ellis N A , Wasylyk B , & Harris C C (2003) B L M helicase-dependent transport o f p53 to sites o f stalled D N A replication forks modulates homologous recombination. EMBOJ22: 1210-1222 Sun H , K a r o w J K , Hickson I D , & Maizels N (1998) The Bloom's syndrome helicase unwinds G 4 D N A . J Biol Chem 273: 27587-27592  Taniguchi T, Garcia-Higuera I, Andreassen P R , Gregory R C , Grompe M , & D A n d r e a A D (2002) S-phase-specific interaction o f the Fanconi anemia protein, F A N C D 2 , with B R C A l and R A D 5 1 . 5 W 100: 2414-2420 Thompson L H , H i n z J M , Yamada N A , & Jones N J (2005) H o w Fanconi anemia proteins promote the four Rs: replication, recombination, repair, and recovery. Environ Mol Mutagen 45: 128-142 van Haaften G , Plasterk R H , & Tijsterman M (2004) Genomic instability and cancer: scanning the Caenorhabditis elegans genome for tumor suppressors. Oncogene 23: 8366-8375 Wicky C , A l p i A , Passannante M , Rose A , Gartner A , & Muller F (2004) Multiple genetic pathways involving the Caenorhabditis elegans Bloom's syndrome genes him-6, rad-51, and top3 are needed to maintain genome stability in the germ line. Mol Cell Biol 24: 5016-5027 Wormbase website: www.wormbase.org. Release WS178, August 2007 W u L , Davies S L , Levitt N C , & Hickson ID (2001) Potential role for the B L M helicase i n recombinational repair v i a a conserved interaction with RAD51. J Biol Chem 276: 19375-19381 W u L & Hickson ID (2003) The Bloom's syndrome helicase suppresses crossing over during homologous recombination. Nature 426: 870-874 Yang Q, Zhang R, Wang X W , Spillare E A , Linke SP, Subramanian D , Griffith J D , L i J L , Hickson I D , Shen J C , Loeb L A , Mazur SJ, Appella E , Brosh R M , J r , Karmakar P, Bohr V A , & Hanis C C (2002) The processing o f Holliday junctions by B L M and W R N helicases is regulated hyp53. J Biol Chem 277: 31980-31987 Youds J L , O'Neil N J , & Rose A M (2006) Homologous recombination is required for genome stability in the absence o f D O G - 1 in Caenorhabditis elegans. Genetics 173: 697-708 Y u n S, Lie-A-Cheong C , & Porter A C (2004) Discriminatory suppression o f homologous recombination by p53. Nucleic Acids Res 32: 6479-6489  CHAPTER 3^ dog-1 is the C. elegans BRIPl/FANCJ homolog  3,1 Introduction The human gene most similar to dog-1 is BRlPl.  Patients with the cancer susceptibility  syndrome Fanconi A n e m i a ( F A ) in subgroup J have mutations in BRIPl/FANCJ  (Bridge et al,  2005; Levitus et al, 2005; Levran et al, 2005). The main diagnostic characteristic o f F A is an extreme hypersensitivity to agents that cause D N A interstrand cross-links, indicating a role for the F A proteins in repair o f cross-links. In accordance with a function in cross-link repair, BRIPl/FANCJ  chicken cell lines are specifically sensitive to D N A cross-linking agents, but are  not sensitive to other types o f D N A damage (Bridge et al, 2005). If dog-1 were the C. elegans BRIPl/FANCJ to BRIPl/FANCJ  homolog, dog-1 mutants should show a similar D N A damage sensitivity profile cells. A functional similarity between D O G - 1 and B R I P l / F A N C J might make  C. elegans a suitable model for study o f the repair pathway involved in F A . Therefore, the goal of the experiments described in this chapter was to determine whether or not D O G - 1 , the B R I P l / F A N C J homolog in C. elegans, has a conserved function in interstrand cross-link repair.  A version o f this chapter has been submitted as part o f a collaborative work: Jillian L . Youds, Louise J. Barber, Jordan D . Ward, Spencer J. Collis, Nigel J. O ' N e i l , Simon J. Boulton and A n n M . Rose (2007). D O G - 1 is the Caenorhabditis elegans B R I P l / F A N C J homolog and functions in interstrand cross-link repair. Manuscript submitted to Molecular and Cellular Biology September 05, 2007.  3.2 Materials and Methods 3.2.1 Measuring apoptosis in the C. elegans germline by SYT012 staining L4-stage animals were picked to fresh plates 24 hours before staining. One-day o l d adults were picked into 50ul o f 3 3 ^ M S Y T 0 1 2 in M 9 buffer (Molecular Probes) in eppendorf tubes and incubated for 3 hours in the dark. After incubation, animals were destained by placing them on N G M OP50 seeded plates to feed for 1 hour. After destaining, animals were placed in 2 m M levamisole on 3% agarose pads for viewing on a Zeiss Axioscope fluorescent microscope with 4 0 X objective. The number of SYT012-stained bodies per gonad arm was counted for each animal. A l l values reported are ± standard error o f the mean.  3.2.2 D N A damage sensitivity assay - X-ray treatment The sensitivity o f dog-I mutants to D N A double strand breaks was measured by exposing the strain to X-rays and scoring viability. L4-stage animals were picked to fresh plates to obtain a synchronized group o f animals. These animals were aged 24-hours so that animals were 1-day old adults on the day o f the experiment. In each case, dog-I was tested alongside an N 2 (wild type) control and a positive control, such as brd-I(gk297)  or brc-I(okI26I),  which are  hypersensitive to double strand breaks. Plates o f the strains to be tested, each containing approximately 20 animals, were placed face-up without lids in the X-ray machine and exposed to either 3000 or 4500 Rads o f X-ray (use o f X-ray machine was kindly allowed by D . Baillie at Simon Fraser University). Control plates were not exposed to X-rays. Following treatment, the animals were allowed to recover overnight at 20°C. Animals were plated 5 per plate and allowed to lay for a 4 hour interval that was the period 22-26 hours post-treatment, then the adults were removed. The number o f dead eggs versus hatching larvae was scored 24 to 48 hours after laying in order to calculate the percentage o f progeny surviving the X-ray treatment.  3.2.3 D N A damage sensitivity assay - UVC-irradiation treatment The sensitivity of dog-1 animals to D N A adducts such as thymidine dimers was tested by exposing animals to UVC-irradiation and scoring viability. L4-stage animals were picked to fresh plates to obtain a synchronized population. These animals were aged 24-hours so that animals were 1-day old adults on the day of UVC-irradiation treatment. In each case, dog-1 was tested alongside an N2 (wild type) control. Plates of each genotype to be tested, containing approximately 20 animals, were placed face-up without lids equidistant from the bulbs in a U V cross-linker (spectrolinker XL-1000, Spectronics Corporation) with 254imi bulbs (spectrolinker use was kindly allowed by D. Riddle, UBC). Animals were exposed to one of the control or test conditions: no U V , 50J U V , or lOOJ U V , and allowed to recover overnight at 20°C. Animals were plated 5 per plate and allowed to lay eggs for the 4 hour period that was 22 to 26 hours post U V C treatment. Following this, adults were removed from the plates. The number of dead eggs versus hatching larvae was scored 24-48 hours after laying and the percentage of progeny surviving U V C treatment was calculated.  3.2.4 D N A damage sensitivity assay - trimethylpsoralen+UVA treatment The sensitivity aï dog-1 mutants to D N A interstrand cross-links was measured by way of trimethylpsoralen (TMP) with U V A treatment. Initial TMP+UVA sensitivity experiments were carried out by the follov^ng protocol: prior to treatment, 3-4 healthy plates of N2 (wild type) animals and the strains to be tested were washed into 15mL falcon tubes using M9 buffer (22mM KH2PO4,22mM Na2HP04, 85mM NaCl, ImM MgS04) containing TritonXlOO (lOO^il/L). The worms were centrifiiged at ISOOrpm for 2min, and the buffer removed. M9/TritonX100 buffer was added to make the final volume 5mL per tube. A T M P stock (trioxsalen, Sigma) of 2.5mg/mL was made by adding 40mL acetone to the lOOmg bottle of TMP. Note that TMP does not dissolve well in M9 buffer, and even in acetone takes time and agitation to dissolve. Once  fully dissolved, 20fil o f the 2.5mg/mL T M P stock solution was added to each tube o f worms i n 5mL of M9/TritonX100 to give a final concentration o f 10|ag/mL. The tubes o f worms were placed on a Nutator mixer and allowed to incubate in T M P for 1 hour at room temperature in the dark. Following this, the tubes were centrifuged at ISOOrpm for 2 minutes. The T M P solution was removed using a vacuum aspirator, leaving 0.5mL o f worms i n liquid in each tube. T o each tube, 1.5mL o f M9/TritonX100 was added (for a total volume o f 2mL) and the animals were transferred into individual wells o f 6 well-plates (2mL worms per well). Animals were exposed to U V A at 340j4.W/cm^ for 90 seconds (use o f U V apparatus was courtesy o f M . Edgley and D . Moerman, U B C ) . Control samples were treated only with T M P , or only with U V A , or with neither.  Following U V A exposure, the animals were plated on fresh plates and allowed to recover for several hours. Young adult animals were plated 5 per plate and left overnight at 20°C. O n the day following treatment, animals were plated to fresh plates and allowed to lay for a 4 hour period that was 22 to 26 hours post-treatment, after which the adult animals were removed from the plates. The number o f dead eggs versus hatching larvae was scored 24-48 hours after laying in order to determine the percentage o f progeny surviving the T M P + U V A treatment. Results obtained from experiments conducted by this protocol were only for a single level o f crosslinking treatment.  In later experiments, the T M P + U V A protocol was modified so that a range o f crosslinking intensities could be tested to construct a kill-curve for cross-link sensitivity. The modified protocol was as follows: on the day prior to treatment, L4-stage animals o f the genotypes o f interest were picked to fresh plates in order to obtain a synchronized population. On the day o f the experiment, the 1-day old animals were washed into tubes with  M9/TritonX100 buffer, as indicated above, and treated witli 10|j,g/mL T M P . A l l parts o f the protocol were the same, except that the worms were exposed to U V A at 550)iW/cm^ for 10, 20 or 30 seconds (these conditions being equal to approximately 55, 110 and 165J o f U V A ) in a 50\xl volume o f buffer.  3.2.5 DAPI staining following T M P + U V A treatment N 2 and dog-1 animals were exposed to T M P + U V A treatment as described at the begiiming o f section 3.2.4. A few hours after treatment, young adults were picked to a separate plate and allowed to recover until 24 hours post-treatment. Animals were washed off the plates with M 9 containing TritonXlOO into eppendorf tubes and washed three times with M9/TritonX100. A s much buffer was removed as possible and 200|il o f 150nM 4',6-diamidino2-phenylindole dihydrochloride ( D A P I ; Sigma) in ethanol was added. The animals were placed in the dark and allowed to incubate for 1 hour. D A P I was removed and animals were washed twice in M9/TritonX100, then destained by soaking i n I m L o f M 9 buffer overnight at 4 ° C . DAPI-stained animals were mounted on 3% agarose slides and viewed at 4 0 X magnification on a Zeiss Axioscope fluorescent microscope. A Retiga 2000R camera (Qimaging) and Openlab 4.0.2 software (Improvision) was used to image the germlines.  3.3 Results 3.3.1 dog-1 is the reciprocal best-BLAST hit of human BRIP1/FANCJ  BRIPl/FANCJ  has been recognized as the human gene most similar to dog-1 (Cheung et  al, 2002). In fact, D O G - 1 and B R I P l / F A N C J are reciprocal b e s t - B L A S T hits between C. elegans and humans, with 3 1 % identity and 50% similarity (http://www.ncbi.nlm.nih.gov/BLAST/; Figure 3.1).  CeDOG-1 HsFANCJ CeDOC-1 HsFANCJ  1 - - - - - - - - -MSSSDAFWRMF"EjNKNKGKSNlRSAFQVVKFEOPSl~STEPDDKraP[IHHElAGLMlKNPAK| 1 MSSMWSEYTIGCVKIYFPYKE|YPSQLAMMNSILRGLNSKOHCLLESPTGSG|3S[1ALLCSALAWOQSLS| 72 Y E j n i M M L C V P V R V P R G L S L Y S T Q K L M l V K l l . T M . K N S 81 E V W I S C C C A C H S K D F Ï N N D M N Q G T S R H F N Y P S Ï P P S E  ,\ LGESPl [GTSSTCQ  :,KTM  IPEKTI  IRKRVQIKNDE ADEGVSEKA  CHWLKQYn: SBKKQASQ" iKKOASnYRDENDDFOVreraKR  CeDOG ^1 HsFANCJ  1 49 161  CeDOG-1 HsFANCJ  229 240  CeDOG - 1 HsFANCJ  389 389  CeDOG- 1 HSFANCJ  469 MLRWIRQVSTEAKÏFARGGQI 4 60 DYESACKI WS(.NEMLL I L H K M T  CCUOC- 1 HsFANCJ  5 49 P S S T A l VC 1 EKWLYFQSYraGNQQjJQSTYRLN 1S I EP 1 Nmgc.rg|NH IKDADVSMS ISFGIflPRPri IRSS AGPRNMQYKEE 5 13 A R E V P V I S A S 1 Q I M L K G L [ | M V L D Q L F R raJls|3SAl)l)YKl AlQUrYgWlffl - • QIDISD  CeDOG-1 HsFANCJ  629 AWtBADAAADCDDWKDPSMSETCHKPISECCKTTISLiWlMStaWC.srai-mFNElRSreWAISr^^ KTQJgMEraKoav 569 CLjlVLPKNK KRSRQKTAVHVLN ' " " [HTraLNnwvAlssI^iNGKvortmiiMwrMisiiffKsrssiïïm 709 â!GDOVtlNKDNlFAAVl..PI[WBFfflNRIOCnYRfMlSDPESS[fftC.rai MTJI 1 KYWi.sNmpAr>wrtrairsE«RvniDonna'QCMIRNSl 636 lANHiOKNSQvwvGriGsaSKjgRNLCAHFQiîijEï- -FEraoDfav!iBli.r.i.sCW(riMsofBligaliSHKiJlËKii5 1. RWLS'IGL  IHGFSGKTmvVFTNDrBlPLRSKVYFEPPFEEEFIEPViaVKNDKIjaEABlWEHDFTPHTPGFUPPVTLKraELEPVtagPEPV jRPLETTQgl RKRHCFgTEVHNLDAKVDSGKTVKLN - SgLEK I NSiaSPIBKPPGHCSRCCCSnKOGNSotaSSNTlfaaDHTG  TtacrCLPRXRIM 'VKjBFSraLPSJAK I L K H I gEQS NSAIJS 1 G C S F K S ARKHADlSOYfflK SDHJ ILLDGKNGKSCYFY VVCNFNRNEKI KiB: GTRTHKQIAQI TE{gLK|aTAQSGVP - M| CeDOG ~ 1 309 AMKPRFEKAL IRDHLERNGTVVFDMEKIJvra'rLAISyP OLCPYFSsHNia 1 w I rmTiTiiwigwsFsraivffii'i n a N i a s m v o i R N S HsFANCJ 314 H G V H K I S D Q H T I Q T F Q G MCKAWDHEIILVSLGKK LGKKLKACPY'iY I I A B F J I I s n h i m i a w a Y N n i i . ! S < \ U I I S F S M [ S E N L K F O  CeDOG - 1 HsFANCJ  K .sjgip S[  iKEraDDSILSL RIKraKA\'mrELDKTGIEErai,li|lIF|Drai:fflULKEFRCHLLYLraH[itL  VILDEAHNIED VILDEAHNIED  IVQBR - -  - FA[aDl.lJSsMVNNN - - IR|3KDJjE[|I.J3/\ncCSLI N V V L E A N A Ë Y J J V F R  ISSANLFTSLTDPENGVDLFTPPLIPKHAAVAAS iFP- ILQGH  JI IKHN JviOKF  PEMQYFDAFK IKISPI YGKEE  CeDOG-1 HsFANCJ  789 M R Q Ji gMKravHLV-taiRRSSELTSVMjBQFDAA 1 FTEl'SRFG AN 1 Nfflsrivn-IMFiaWaMïraa 1 flTâ XrtrKRCTgvMISV a»TSvr3flAM[E 714 WHNl Liai, viy 11^ I vfa|o G G E K T N - - FTBELLOvY'.ffiA 1 KYKGEKpfflAili A'ff'^cITBTOllSTagi HJssiiTiJNCTa -\|fi- r 1 atlsi'-fSni V KlT CeDOG - 1 869 EQNE QAYRALNQALGRC SARVS SA 2MLM im ÎSLliRO IGNl.VGl HSFANCJ 792 IPBR QAYRALNQALGRC IROÈ K H l l 'LILV{J;D|FRNNPSRYI CeDOG 1 HSFANCJ  949 867  CeDOG- 1 HsFANCJ  947  CeDOG-1 HSFANCJ  1027  ILKSYPSIjKEFNANF 1OHHSTQESALESLA  I QRRH,\nEKAKKENFCE  LSKKHOKHLNVS1KDRTN1ODNESTLEVTSLKYSTPPYLLEAASHLSPENFVEDEAKICVOEL  QCPK11 l'KNSPLPSS11SRKEKNDPVFLEEACKAEK1V1SRSTSPTFNKQlKRVSVVSSFNSLGOYFTGK1PKAÏPELGSS  CeDOG 1 HsFANCJ 1 107 CeDOG-1 HsFANCJ  So  1 187  Figure 3.1: Protein sequence alignment o f C. elegans D O G - 1 and human B R I P l / F A N C J . Identical amino acids are shown on black background, similar amino acids are shown on grey background. Sequences aligned using ClustalW 1.83 and shaded with BoxShade 3.21.  The similarity between the proteins is further exemplified by the fact that C. elegans D O G - 1 , Gallus gallus B R I P l / F A N C J and Homo sapiens B R I P l / F A N C J contain the same conserved domains, including D E x D c 2 , a domain characteristic o f the D E A D - I i k e helicases superfamily, D E A D _ 2 , a conserved region o f a number of R A D - 3 - l i k e D N A - b i n d i n g helicases, H E L I C c 2 , which is a helicase superfamily C-terminal domain, and D i n G , a RAD-3-related helicase domain found in proteins associated with D N A replication, transcription, recombination and repair (http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml; Figure 3.2). One important difference between D O G - 1 and human B R I P l / F A N C J is the protein size: B R I P l / F A N C J is 1249 amino acids in length and contains 266 amino acids at its C-terminal end that do not exist in the 983 amino acid D O G - 1 protein (Figure 3.2). It is the C-terminal end o f human B R I P l / F A N C J that interacts with the breast cancer-associated protein B R C A l ; specifically, amino acids 988 to 993 of B R I P l / F A N C J have been shown to interact with B R C A l (Cantor et al, 2001; Shiozaki et al, 2004), and these residues are not conserved in D O G - 1 .  250  1  Caenorhaditis elegans |  500  DExOc2  OinG  750  ,  ^DExPc2  j  |  DinG  HEUCc2  |  DinG  DEAD_2 ^  „  „  Ga/^sga//os  250  ^  | D6xDc2 DinG  |  500  D£xDc2  |  750  ,non  HEUCc2  ^  750  ioqO  1219  _|  I  175?  j  DinG DEAD_2  Homo sapiens BRIPl  1  I  , OExOc2  250  500  I  I  HEUCC2  DEAD_2  Figure 3.2: Conserved domains o f C. elegans D O G - 1 , G. gallus B R I P l / F A N C J and H. sapiens B R I P l / F A N C J as determined by the Conserved Domain Database (http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml). Black box indicates the B R C A l interacting region in human B R I P l / F A N C J .  The next most similar C. elegans gene to human BRIPl/FANCJ  is F25H2.13.  These  proteins share 27% identity and 45% similarity, indicating that F25H2.13 and B R I P l / F A N C J are less well-matched than are D O G - 1 and B R I P l / F A N C J . Furthermore, phylogenetic analysis indicates that F25H2.13 is evolutionarily more similar to the human N H L and chicken R T E L proteins than it is to B R I P l / F A N C J (http://www.ebi.ac.uk/clustalw/), supporting the hypothesis that D O G - 1 , not F25H2.13, is the most likely C. elegans B R I P l / F A N C J homolog (Figure 3.3).  Homo sapiens BRIPl/FANCJ — Gallus gallus BRIPt/FANCJ — Caenohabditis elegans DOG-1 — Homo sapiens NHLl — Gallus gallus RTEL — Caenohabditis elegans F25H2.13 Homo sapiens BLM Gallus gallus BLM Caenohabditis elegans HIM-6 Caenohabditis elegans HEL-1  Figure 3,3: Phylogram showing the estimated evolutionary relationship between human B R I P l / F A N C J , chicken B R I P l / F A N C J , C. elegans D O G - 1 , human N H L l , chicken R T E L , and C. elegans F25H2.13. The B l o o m Syndrome proteins, human B L M , chicken B L M and C. elegans H I M - 6 , as well as the C. elegans R N A helicase H E L - 1 are included as an outgroup. Phylogram created using ClustalW (http://www.ebi.ac.uk/clustalw/).  3.3.2 dog-1 animals have increased DNA damage-induced apoptosis in the germline If dog-1 is the homolog o f BRIPl/FANCJ  and has a role in D N A repair, it might be  expected that the dog-1 (gk 10) strain would show some indication o f D N A damage or stress under normal conditions. Upon initial examination, the dog-1 (gklO) strain is relatively healthy, and the only obvious difference between dog-1 and N 2 (wild type) animals is a slightly reduced brood size o f 239 ± 7 for dog-1 compared to 271 ± 13 for N 2 . After observing the strain over multiple generations, the mutator phenotype reported by Cheung et al. (2002) is also apparent. The mutator phenotype is evidence for defects in the ability o f dog-l animals to maintain the D N A . Persisting D O G - 1 substrates, such as G/C-tract secondary structures or other unrepaired  intermediates, could be problematic during replication and might be detected as a type o f D N A damage or lesion. To test whether or not dog-1 mutants had indications o f D N A  damage under  normal conditions, the vital dye S Y T 0 1 2 was used to observed apoptotic corpses in the germline. In N 2 (wild type) animals, an average of 0.8 ± 0.2 (n=44 gonad arms) S Y T 0 1 2 stained corpses per gonad arm was observed, while in dog-1 animals, an average o f 3.3 ± 0.2 (n=77) corpses were present per gonad arm (Figure 3.4). Thus, apoptosis is elevated in the absence o f D O G - 1 . To determine whether the increased apoptosis in dog-1 mutants was the resuh of unrepaired D N A  damage, a dog-1 cep-1 double mutant was constructed and S Y T 0 1 2 -  stained. C E P - 1 is the C. elegans p53 homolog that is required for D N A apoptosis (Deny et ai, 2001 ; Schumacher et ai, 2001). Loss o f CEP-1  damage-induced abrogated the increased  apoptosis observed in dog-1 mutants (0.8 ± 0.1 corpses per gonad arm [n=49]; Figure 3.4), indicating that the increased apoptosis in dog-1 animals is dependent on the D N A checkpoint protein C E P - 1 . Therefore, dog-1 mutants contain D N A 1 checkpoint. Elevated levels o f apoptosis triggered by the D N A  damage  damage that triggers the cepdamage checkpoint might  account for the reduced brood size i n dog-1 mutants.  E »10  <  c 8-9 3 1Of 6-7 s. 4-5 uo o 2-3 JCXXXXXXXXXX È E xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx 2  N2  Ê >10 < I 6-7 I I » Ê i 0-1  X  >10  X  B-9  xxxxx  £ 6-7  xxxxxxxxxx  & 4-S  xxxxxxxxxxxxxxxxxxxx xxxxxxx dog-1  xxxxxxxx e  i 01 xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx dog-1  cep-1  Figure 3.4: Number o f SYT012-stained corpses in N 2 , dog-1 and dog-l cep-1 animals. The elevated apoptosis observed in dog-1 mutants is abrogated in the absence o f C E P - l / p 5 3 .  3.3.3 dog-1 animals are hypersensitive to DNA interstrand cross-linking agents A diagnostic cliaracteristic o f F A cells is specific hypersensitivity to D N A interstrand cross-linking agents (reviewed in Levitus et al, 2006). Chicken B R I P l / F A N C J cells are highly sensitive to D N A damage in the form o f interstrand cross-links, but are not sensitive to D N A lesions generated by X-ray or UV-irradiation treatment (Bridge et al, 2005). If D O G - 1 is the homolog o f B R I P l / F A N C J , the dog-l mutant should show a similar sensitivity to cross-linking agents. In order to test this hypothesis, a series o f D N A damage sensitivity tests were carried out on the dog-1 strain. N 2 animals and dog-l mutants were exposed to 0, 50 or lOOJ of U V C irradiation (254rmi), and embryonic survival was measured in the broods 22-26 hours following treatment, dog-l animals were no more sensitive to the U V C treatment than were N 2 animals (ttest p=0.60 at 50J; p=0.28 at lOOJ; Figure 3.5). Similarly, after exposure to 0, 3000 or 4500 Rads o f X-ray, dog-1 animals did not show significant hypersensitivity compared to N 2 animals (p=0.28 at 3000 Rads; p=0.58 at 4500 Rads; Figure 3.5), though sensitivity was observed in the positive control strain brd-l/BARDl,  as expected.  B  100 90 80 70  N2(wild type) dog-1  60 50 S  40  4) Û.  30 20 10 0  -r  1500 D o s e of U V î M m (J)  T  3000  4500  D o s e of X - r a y ( R a d s )  Figure 3.5: Sensitivity o f N 2 and dog-l animals to A ) UVC-irradiation and B ) X-ray treatment. Values plotted represent single experiments. N o significant difference in embryonic survival between N 2 and dog-l after U V C or X-ray treatment was replicated in 3 independent experiments.  Like chicken B R I P l / F A N C J cells, dog-1 mutants were not sensitive to X-ray or U V irradiation-induced D N A damage, but these animals were sensitive to D N A cross-linking agents. Cross-linking with trimethylpsoralen and U V A ( T M P + U V A ) treatment caused significant embryonic death in dog-1 mutants compared to N 2 animals. In these experiments, N 2 and dog1 animals were incubated in M 9 buffer containing 10)ig/ml T M P for 1 hour, then exposed to U V A at 340)aW/cm' for 90 seconds in a 2ml volume o f liquid. During this treatment, the T M P intercalates with the D N A during the time that the nematodes are incubated in the T M P . Upon U V A exposure, D N A - b o u n d T M P monoadducts are formed when a photon is absorbed, and cross-links between two nearby monoadducts are formed when a second photon is absorbed (Johnston et al, 1981). Control experiments including no treatment, U V A only and T M P only conditions indicated that dog-1 mutants were specifically sensitive to the T M P + U V A crosslinking treatment (no treatment p=0.43; U V A only p=0.18; T M P only p=0.50; T M P + U V A p=6.1xl0''^; Figure 3.6).  0N2  m dog-1  • no treatment  UVA only  TMP only  TmUVA  Treatment  Figure 3.6: Embryonic survival in the progeny of N 2 and dog-1 animals following no treatment, U V A only, T M P only or T M P + U V A treatment.  T M P + U V A sensitivity experiments were carried out on 5 independent occasions, and it was determined tiiat tiie treatment leads to some variability in severity o f cross-linking due to experimental conditions; for example, the activity of the T M P or the number and density o f worms exposed to U V A would vary slightly between experiments. Thus, embryonic survival o f N2 animals varied somewhat between experiment trials, but in every trial, dog-1 animals were significantly hypersensitive to the treatment, compared to N 2 animals (trial 1 p=0.0056; trial 2 p=0.00015; trial 3 p=5.8xlO-^; trial 4 p=3.4xl0"^; trial 5 p=6.1xl0"^ Figure 3.7).  03 > D CO O  'c o  £^  n E HI c  Q) O I—  0) ÛL  r-i-]  90 80 70 60  mm  50  * dog-l  40 30 20 10 0  • 3  4  T M P + U V A Trial  Figure 3.7: Variability between trials o f T M P + U V A treatment. Although severity o f sensitivity varied between trials, dog-1 was shown to be significantly hypersensitive in all 5 trials o f T M P + U V A treatment.  Because o f the variability in this protocol, and in order to more closely match our test conditions with those o f our collaborators in the Boulton lab, the protocol was adapted slightly. A smaller, more uniform number o f animals were treated in each experiment, and survival was observed at 3 different doses o f U V A in order to construct a k i l l curve for the conditions. Approximately 150 animals were incubated in 10|.ig/ml T M P for 1 hour, then exposed as groups o f approximately 50 animals to U V A at 550).iW/cm^ for 10, 20 or 30 seconds (being equal to 55,  110 and 165 J o f U V A ) . Similar to previous T M P + U V A treatments, significant hypersensitivity was observed in dog-1 mutants compared to N 2 animals at every dose tested (p=0.0041 at 10^g/ml T M P + 55J U V A ; p=9.0xl0"'' at 10|ag/ml T M P + 1 lOJ U V A ; p=8.2xl0"' at lO^g/ml T M P + 165J U V A ; Figure 3.8). Some variability between experiments was still observed under these treatment conditions, and this may be due to the nature o f T M P itself Importantly, the experiments showed that dog-1 is consistently hypersensitive to cross-linking treatment by TMP+UVA.  1 0 0 !!» 90  to >  80  3  60 •  c  40  70 •  CO  CL  50 30  dog-1  T ±  20 • 10 • 0 0  N2(wild type)  T i  —1— 50  100  T 1 —I— 150  —I 200  1 0 u g / m l T M P + D o s e of U V A (J)  Figure 3.8: Sensitivity of N 2 and dog-1 animals to trimethylpsoralen/UVA ( T M P + U V A ) treatment with lO^ig/ml T M P and 55, 110 or 165J U V A .  Experiments carried out by the Boulton lab also indicate that dog-1 is specifically hypersensitive to other cross-linking agents, including nitrogen mustard and cisplatin (L. Barber & S. Boulton, p. communication).  3.3.4 dog-1 mutants display evidence of repair defects after cross-linking treatment The dog-1 mutant hypersensitivity to interstrand cross-links detailed above could be explained by two different hypotheses. One possibility is that the checkpoint that detects D N A  interstrand cross-links might not be ftinctional in dog-1 mutants. The other possibility is that the cross-link repair process itself is impaired in the absence o f D O G - 1 . In order to test the activity of the D N A damage checkpoint after cross-linking treatment, S Y T 0 1 2 staining was carried out on N 2 and dog-1 mutants 24 hours after T M P + U V A treatment. A s described in section 3.3.2, N2 animals under normal conditions had an average o f 0.8 ± 0.2 (n=44 gonad arms) S Y T 0 1 2 stained corpses, while dog-1 animals under normal conditions had an average o f 3.3 ± 0.2 (n=77) corpses present per gonad arm. Twenty-four hours after cross-linking treatment with T M P + U V A , the average number o f corpses in N 2 gonad arms had increased to 2.6 ± 0.2 (n=43), while in dog-1 animals, the average number o f corpses per gonad arm had increased to 5.3 ± 0.4 (n=27; Figure 3.9).  É *10  I  xxxxxxxxxxxxxxxxxx  X  I 6-7  xxx  5 4-5  xxxxxx  X  xxxxxxx  8-9  I  2-3  E  ê 0-1  xxxxxxxxxxxx xxxx  dog-1  N2  D  § »10  I xxxxxxxxx xxxxxxxxxxx xxxxxxx  8-9  X XX  I 6-7 s.  XXXXXXXXX  o 4-5  XXX  È  2-3  XXXXXXXX  X  E  N2 24 hrs post T M P + U V A  ^ 0-1  dog-1 24 hrs post T M P + U V A  Figure 3.9: Number o f SYT012-stained corpses in A ) N 2 and B ) dog-1 animals under normal conditions and C ) N 2 and D) dog-1 animals 24 hours after D N A cross-linking with T M P + U V A treatment. D N A damage-induced apoptosis is increased equally in N 2 and dog-1 animals following cross-linking.  Apoptosis was increased by approximately 2 corpses per gonad arm in both N 2 and dog-1 animals 24 hours after cross-linking treatment, indicating that the D N A damage checkpoint is equally active in N 2 and dog-1 animals. Based on this data, the embryonic lethality in the progeny o f dog-1 mutants after cross-linking treatment cannot be explained by defects in the D N A damage checkpoint; rather, defects in the cross-link repair process in dog-1 mutants must be the cause o f lethality following T M P + U V A treatment.  The germlines o f N 2 and dog-1 animals were visualized with D A P I staining 24 hours after cross-linking treatment so that defects in repair could be observed. Chromatin bridges were present in the early embryos o f dog-1 mutants, indicating that in these animals there is D N A bound together that becomes stretched apart when the early embryonic cells try to complete mitosis (Figure 3.10). These chromatin bridges might represent unrepaired interstrand crosslinks in dog-1 mutants following D N A cross-linking, as they are not present in N 2 animals after T M P + U V A treatment, and are also not observed in N 2 or dog-1 mutants under normal conditions. These data, taken with others, support the hypothesis that dog-1 is the C. elegans homolog o f human BRIPl/FANCJ,  and has a conserved role in interstrand cross-link repair.  A  Figure 3.10: D A P I stained A ) N 2 and B) dog-1 mutant germlines 24 hours after T M P + U V A treatment. Chromatin bridges (arrows) are present in the early embryos o f dog-1 but not N 2 animals.  3.4 Discussion 3.4.1 Phenotypes of the dog-1 mutant suggest D N A damage under normal conditions The G/C-tract deletion phenotype o f dog-1 mutants reported by Cheung et al. (2002) is unambiguous evidence for the role o f dog-1 in maintaining genome stability. The dog-1 strain also displays several additional phenotypes that are suggestive o f a D N A repair mutant. Apoptosis is elevated in the germlines o f dog-1 animals compared to N 2 , and is dependent on cep-l/p53, indicating that the D N A damage checkpoint is active i n dog-1 mutants under normal conditions. It is likely that the D N A damage checkpoint is detecting endogenous problems associated with G/C-tracts in the absence o f D O G - 1 . These might be persisting secondary structures that are difficult to resolve or repair due to local chromatin geography or the checkpoint might be activated in response to a G/C-tract deletion that has occuned in an essential gene, leading to apoptosis. The elevated apoptosis in dog-1 mutants might account for the reduced brood size observed in the strain. The dog-1 strain also shows a slightly elevated level o f embryonic lethality among progeny, approximately 2.9% embryonic lethality versus 0.1%) in N 2 (reported i n section 2.3.1). The potential problems at G/C-tracts described above might also be the reason for the slightly elevated level o f embryonic lethality observed in the dog-1 strain. Whereas major problems at G/C-tracts might result i n apoptosis i n the germline, i n less severe cases, the nucleus might survive to become an embryo, only to die when the stalled replication fork becomes a greater problem during the rapid cycles o f replication that occur during embryonic development or when the deleted essential gene function is required in embryogenesis. Together, the G/C-tract deletion phenotype, elevated germline apoptosis and elevated embryonic lethality support a role for dog-1 in maintaining the D N A .  3.4.2 DOG-1 is the C. elegans B R I P l / F A N C J homolog Reciprocal b e s t - B L A S T analysis showed that D O G - 1 is the C. elegans protein most similar to human B R I P l / F A N C J . The two proteins contain the same conserved domains and similar domain architecture. Furthermore, an evolutionary analysis including the next most similar C. elegans protein, F25H2.13, indicates that D O G - 1 is evolutionarily most similar to B R I P l / F A N C J . In humans, a central function o f B R I P l / F A N C J is to carry out I C L repair as F/iA^CJ cells lines are extremely sensitive to ICL-inducing agents and a subset o f F A patients have mutations in BRIP J/FANCJ  (Levitus et al., 2005; Levran et al., 2005). It is shown here that  dog-] has a conserved role in I C L repair as mutants are highly sensitive to cross-linking treatment at all doses o f T M P + U V A tested. The D N A damage checkpoint is activated in dog-1 mutants after cross-linking treatment, indicating that the defect i n these animals is not an inability to detect the D N A damage. Furthermore, chromatin bridges are visible in early embryos following cross-linking treatment, suggesting the presence o f unrepaired cross-links in dog-1 mutants. These findings, together with the sequence similarity between D O G - 1 and B R I P l / F A N C J , support the hypothesis that DOG-1 is the B R I P l / F A N C J homolog in C. elegans and has a conserved role in the repair o f interstrand cross-links. Based on this data, the C. elegans dog-1 mutant might make a useful model for study o f the interstrand cross-link repair pathway, which, in human cells, is extremely complex.  3.4.3 Implications of sequence conservation between DOG-1 and B R I P l / F A N C J While B R I P l / F A N C J has been identified as a Fanconi anemia protein, B R I P l / F A N C J (also referred to as B A C H l ) was first recognized as having a role in double strand break repair. The C-terminal end o f human B R I P l / F A N C J interacts with B R C A l ; specifically, amino acids 988-993, and the motif Ser990-X-X-Phe993, are required for this interaction (Cantor et al., 2001 ; Shiozaki et al., 2004). Cantor et al. (2001) reported that double strand break repair was impaired  by mutations in B R I P l / F A N C J that affect its ability to hydrolyze A T P , but repair was normal when a second B R I P l / F A N C J mutation that prevented binding to B R C A l was introduced. Thus, B R I P l / F A N C J appears to be involved in double strand break repair through its interaction with B R C A l . Germline mutations in B R I P l / F A N C J that affected helicase activity were detected in two patients with either early onset or familial breast cancer who did not have B R C A l or B R C A 2 mutations (Cantor et al, 2001; Cantor et al, 2004). However, multiple other studies have failed to demonstrate the involvement o f B R I P l / F A N C J in n o n - B R C A l / B R C A 2 associated familial breast cancer (Rutter et al, 2003; Karppinen et al, 2003; L u o et al, 2002; Lewis et al, 2005; Vahteristo et al, 2006), indicating that B R I P l / F A N C J germline mutations are a rare class and do not explain all those cases o f familial breast cancer which are not caused by B R C A l or B R C A 2 . Several studies have shown that B R I P l / F A N C J is mutated in patients from Fanconi anemia subgroup J (Levran et al, 2005; Levitus et al, 2005; Litman et al, 2005). Clearly, germline mutations o f B R I P l / F A N C J lead to a different outcome than do inherited B R C A l mutations. In fact, Peng et al (2007) have shown that the role o f B R I P l / F A N C J i n cross-link repair is independent o f B R C A l , and requires an interaction with the mismatch repair protein M L H l that involves lysines 141 and 142 o f B R I P l / F A N C J . It appears that B R I P l / F A N C J has multiple roles, both in interstrand cross-link repair and in double strand break repair in humans. Some authors have speculated that B R C A l might be functionally linked to the cross-link repair pathway through its interaction with B R I P l / F A N C J , but evidence to support this proposal has yet to be shown (Cantor & Andreassen, 2006).  Recent work on chicken B R I P l / F A N C J cell lines has also been informative with respect to the function o f B R I P l / F A N C J . Chicken B R I P l / F A N C J has 54% identity and 68% similarity to human B R I P l / F A N C J , and as is the case for C. elegans, the chicken and human proteins differ most at the C-terminal end (Bridge et al, 2005). Although the chicken and human  proteins are similar i n length, chicken B R I P l / F A N C J does not contain the Ser990-X-X-Phe993 motif that is required for B R C A l interaction (Bridge et al, 2005; Shoizaki et al, 2004). Bridge etal (2005) showed that chicken B R I P l / F A N C J cells were not significantly sensitive to double strand breaks induced by X-rays, and the cells could carry out normal homology-directed repair of an I-Scel-induced D N A double strand break. However, chicken B R I P l / F A N C J cells were highly sensitive specifically to interstrand cross-linking agents, supporting a conserved role for chicken B R I P l / F A N C J i n cross-link repair but not double strand break repair (Bridge et al, 2005).  The findings reported here for C. elegans mirror those observed in chicken cells. D O G - 1 and B R I P l / F A N C J have 50% similarity and 3 1 % identity, and the domain structure o f the two proteins is similar, making the proteins reciprocal b e s t - B L A S T hits between C. elegans and humans. However, 348 amino acids at the C-terminal end o f human B R I P l / F A N C J are not present in D O G - 1 . Because D O G - 1 is missing the B R C A l - i n t e r a c t i n g domain, it is unlikely to interact with the C. elegans B R C A l ortholog B R C - 1 . dog-1 mutants are not significantly sensitive to X-ray treatment, indicating that there are no defects in the double strand break repair pathway in the absence o f D O G - 1 , which has been shown to involve B R C - 1 / B R C A l in C. elegans (Boulton et al, 2004). Therefore, the proposed role for human B R I P l / F A N C J i n double strand break repair is unlikely to be conserved in C. elegans. Conversely, the role for B R I P l / F A N C J in cross-link repair is clearly conserved in C. elegans, as dog-1 mutants are highly sensitive to agents that induce D N A cross-linking. Thus, it appears that B R I P l / F A N C J has an evolutionarily conserved role in interstrand cross-link repair (C. elegans, chicken and human), but only the human protein, containing the B R C A l interacting motif, also appears to function in double strand break repair. A s several independent studies have shown that B R I P l / F A N C J is mutated in multiple patients with Fanconi anemia (Levran et al, 2005; Levitus  et al, 2005; Litman et al, 2005), it might be argued that the role for B R I P l / F A N C J in interstrand cross-link repair is more central to maintaining genome stability than is its role in double strand break repair.  3.4.4 Relationship of D O G - l / F A N C J to the F A pathway In human cells, B L M and components o f the Fanconi repair pathway are functionally linked (Pichieni et al, 2004; Meetei et al, 2003; Hirano et al, 2005). B L M cells are sensitive to cross-linking agents and B L M localizes with F A N C D 2 at stalled replication forks following exposure to hydroxyurea or cross-linking treatment (Pichieni et al, 2004). Furthermore, the Fanconi core complex is required for cross-link dependent B L M phosphorylation after crosslinking treatment (Pichierri et al, 2004). Hirano et al (2005) showed that the rate of sister chromatid exchange for F A N C C B L M double mutant chicken cell lines was similar to that o f B L M cells, suggesting that these two proteins are part o f the same pathway. In chapter 2, a genetic interaction between dog-1 and him-6/BLM  Mias described. That dog-1 and  him-6/BLM  show a synthetic sick phenotype in C. elegans suggests that dog-1 is somehow associated with him-6/BLM, but that the two proteins function in separate pathways. This idea is consistent with the finding o f increased frequency o f deletions in dog-1; him-6 animals, as both proteins appear to be able to resolve or repair G/C-tract structures, but likely do so through different means. I f H I M - 6 / B L M functions as part o f the Fanconi interstrand cross-link repair pathway in C. elegans, as it does in humans, this would imply that DOG-1 functions in a separate pathway from H I M 6 / B L M and the F A pathway.  Whether human B R I P l / F A N C J functions directly as part o f or in parallel to the F A pathway remains an open question. N o data is available from human cells, and data from chicken B R I P l / F A N C J cell lines is unclear. Although F A N C D 2 focus formation is normal in  B R I P l / F A N C J chicken cells, which might suggest that B R I P l / F A N C J acts downstream o f F A N C D 2 , interstrand cross-link sensitivity in the F A N C C F A N C J double mutant is additive, suggesting that these proteins function in parallel pathways (Bridge et al, 2005). The F A pathway is likely a complex pathway with multiple branching points, and so it may be difficult to separate and clearly identify the interactions i n this pathway, particularly i n higher organisms. Data from the Boulton lab indicates that in C. elegans, F C D - 2 / F A N C D 2 focus formation is not affected by the absence o f D O G - 1 , and this could be interpreted to mean that D O G - 1 functions either downstream or in parallel to F C D - 2 / F A N C D 2 ( L . Barber & S. Boulton, p. communication). In addition, dog-1 fcd-2 double mutants show epistasis in their sensitivity to the cross-linking agent nitrogen mustard, suggesting that D O G - 1 and F C D - 2 / F A N C D 2 function in the same pathway ( L . Barber & S. Boulton, p. comm). This result initially appears difficult to reconcile with the finding that G/C-tract deletions are increased i n dog-1; fcd-2 double mutants. One possibility is that D O G - 1 has separate functions in I C L repair and at G/C-tract secondary structures. F C D - 2 / F A N C D 2 functions at stalled replication forks that may be the result o f interstrand cross-links or other fork blocking lesions (Collis et al, 2006). D O G - 1 might function in the F C D - 2 / F A N C D 2 pathway for I C L repair, while it functions independently from F C D 2 / F A N C D 2 at G/C-tracts. In the absence o f D O G - 1 , F C D - 2 / F A N C D 2 could be required at replication forks stalled by G/C-tract secondary structures. Further research w i l l be required i n order to determine the role o f D O G - l / F A N C J relative to the F A pathway, but C. elegans holds much promise as a simplified model in understanding this pathway.  3.4.5 How does the role for dog-1 in I C L repair relate to G/C-tracts? D O G - 1 has two seemingly unrelated functions i n resolving secondary structures at G / C tracts and in repairing interstrand cross-links; however, there may be some relationship between these two roles. It is possible that DOG-1 is a helicase that recognizes multiple different  substrates, and acts at replication forks that are stalled by either a secondary structure or crosslinked D N A . However, it is more likely that DOG-1 has a certain type o f structure that it recognizes, and there is some structural similarity between a secondary structure and cross-link or cross-link repair intermediate. The roles for DOG-1 at G/C-tracts and in I C L repair might be separate, but have a common substrate structure that is recognized by D O G - 1 . A biochemical analysis of the substrates that D O G - 1 can unwind would determine whether or not this hypothesis is correct. Such experiments might measure the ability o f DOG-1 to unwind G quadruplex D N A , double stranded D N A , forked duplexes, and other structures. Regardless o f whether or not a G/C-tract secondary structure and a D N A cross-link or cross-link repair intermediate are structurally similar, the finding that D O G - 1 has roles both at G/C-tracts and i n cross-link repair hints at a possible function for human B R I P l / F A N C J at both o f these types o f lesions.  3.4.6 Future experiments with dog-1 While the data presented here strongly suggests that D O G - 1 is the B R I P l / F A N C J ortholog, the ultimate proof would be to rescue the interstrand cross-link sensitivity o f chicken and human BRIPl/FANCJ  cell lines with dog-1 c D N A . Several additional experiments could  also be carried out using dog-I to find new genes involved in replication and repair. For example, a large-scale R N A i screen for genes that are synthetic lethal with dog-1 would likely identify multiple new genes that are required for genome stability. dog-I would be a good choice for such a screen because the dog-1 (gklO) strain is quite healthy and sterile animals are rare, which would reduce the chance of finding false positives in the screen. In addition, new proteins acting in the cross-link repair pathway might be identified through a yeast two-hybrid screen with D O G - 1 , or with other components o f the repair pathway, such as F A N C D 2 or BRCA2/FANCD1.  The findings reported in this chapter support a role for D O G - 1 in interstrand cross-link repair that is functionally similar to the human homolog B R I P l / F A N C J . Considering that all o f the basic components o f the F A pathway are conserved in C. elegans, including F A N C J , F A N C D 2 , F A N C M , and F A N C L , among others (Collis et al., 2006), the nematode could be a useful model for study o f the interstrand cross-link repair pathway with respect to F A . Furthermore, the role o f D O G - 1 at G/C-tracts might shed light on endogenous D N A structures as a potential contributor to genome instability in the cancer susceptibility syndrome.  Chapter 3 References Boulton SJ, Martin JS, Polanowska J, H i l l D E , Gartner A , & V i d a l M (2004) B R C A l / B A R D l orthologs required for D N A repair in Caenorhabditis elegans. Curr Biol 14: 33-39 Bridge W L , Vandenberg C J , Franklin R J , & H i o m K (2005) The B R I P l helicase functions independently o f B R C A l in the Fanconi anemia pathway for D N A crosslink repair. Nat Genet 37: 953-957 Cantor S, Drapkin R, Zhang F , L i n Y , Han J, Pamidi S, & Livingston D M (2004) The B R C A l associated protein B A C H l is a D N A helicase targeted by clinically relevant inactivating mutations. Proc Natl Acad Sci USA 101: 2357-2362 Cantor S B & Andreassen P R (2006) Assessing the link between B A C H l and B R C A l i n the F A pathway. Cell Cycle 5: 164-167 Cantor S B , B e l l D W , Ganesan S, Kass E M , Drapkin R, Grossman S, Wahrer D C , Sgroi D C , Lane W S , Haber D A , & Livingston D M (2001) B A C H l , a novel helicase-like protein, interacts directly with B R C A l and contributes to its D N A repair function. Cell 105: 149-160 Cheung I, Schertzer M , Rose A , & Lansdorp P M (2002) Disruption o f dog-1 in Caenorhabditis elegans triggers deletions upstream o f guanine-rich D N A . Nat Genet 31: 405-409 Collis SJ, Barber L J , Ward J D , Martin JS, & Boulton SJ (2006) C. elegans F A N C D 2 responds to replication stress and functions in interstrand cross-link repair. DNA Repair (Amst) 5: 1398-1406 Derry W B , Putzke A P , & Rothman J H (2001) Caenorhabditis meiosis, and stress resistance. Science 294: 591-595  elegans p53: role in apoptosis,  Hirano S, Yamamoto K , Ishiai M , Yamazoe M , Seki M , Matsushita N , Ohzeki M , Yamashita Y M , Arakawa H , Buerstedde J M , Enomoto T, Takeda S, Thompson L H , & Takata M (2005) Functional relationships o f F A N C C to homologous recombination, translesion synthesis, and B L M . £/kffiO J 24: 418-427 Johnston B H , K u n g A H , Moore C B , & Hearst J E (1981) Kinetics o f formation o f deoxyribonucleic acid cross-links by 4'-(aminomethyl)-4,5',8-trimethylpsoralen. Biochemistry 20: 735-738 Karppinen S M , Vuosku J, Heikkinen K , AUinen M , & Winqvist R (2003) N o evidence of involvement o f germline B A C H l mutations in Finnish breast and ovarian cancer families. Eur J Cancer 39:366-371 Levitus M , Joenje H , & de Winter JP (2006) The Fanconi anemia pathway o f genomic maintenance. Cell Oncol 28: 3-29 Levitus M , Waisfisz Q , Godthelp B C , de Vries Y , Hussain S, Wiegant W W , ElghalbzouriMaghrani E , Steltenpool J, Rooimans M A , Pals G , Arwert F , Mathew C G , Zdzienicka M Z ,  H i o m K , De Winter JP, & Joenje H (2005) The D N A helicase B R I P l is defective in Fanconi anemia complementation group J. Nat Genet 37: 934-935 Levran O, Attwooll C , Henry R T , M i l t o n K L , Neveling K , R i o P, Batish S D , K a l b R, Velleuer E , B a n a l S, Ott J, Petrini J, Schindler D , Hanenberg H , & Auerbach A D (2005) The B R C A l interacting helicase B R I P l is deficient in Fanconi anemia. Nat Genet 37: 931-933 Lewis A G , Flanagan J, Marsh A , Pupo G M , Mann G , Spurdle A B , Lindeman G J , Visvader J E , Brown M A , Chenevix-Trench G , & Kathleen Cuningham Foundation Consortium for Research into Familial Breast Cancer (2005) Mutation analysis o f F A N C D 2 , B R I P l / B A C H l , L M 0 4 and S F N in familial breast cancer. Breast Cancer Res 7: R1005-16 Litman R, Peng M , Jin Z , Zhang F, Zhang J, Powell S, Andreassen P R , & Cantor S B (2005) B A C H l is critical for homologous recombination and appears to be the Fanconi anemia gene product F A N C J . Cancer Cell 8: 255-265 Luo L , L e i H , D u Q, von Wachenfeldt A , K o c k u m I, Luthman H , Vorechovsky I, & Lindblom A (2002) N o mutations i n the B A C H l gene in B R C A l and B R C A 2 negative breast-cancer families linked to 17q22. IntJ Cancer 98: 638-639 Meetei A R , Sechi S, Wallisch M , Yang D , Young M K , Joenje H , HoatHn M E , & Wang W (2003) A multiprotein nuclear complex connects Fanconi anemia and B l o o m syndrome. Mol Cell Biol 23: 3417-3426 Peng, M . , R. Litman, J. X i e , S. Sharma, R. M . Brosh Jr, and S. B . Cantor (2007) The F A N C J / MutLalpha interaction is required for correction o f the cross-link response in F A - J cells. EMBO J. 26: 3238-3249 Pichierri P, Franchitto A , & Rosselli F (2004) B L M and the F A N C proteins collaborate in a common pathway i n response to stalled replication forks. EMBO J 23: 3154-3163 Rutter J L , Smith A M , Davila M R , Sigurdson A J , Giusti R M , Pineda M A , Doody M M , Tucker M A , Greene M H , Zhang J, & Struewing JP (2003) Mutational analysis o f the B R C A l interacting genes Z N F 3 5 0 / Z B R K 1 and B R I P l / B A C H l among B R C A l and BRCA2-negative probands from breast-ovarian cancer families and among early-onset breast cancer cases and reference individuals. Hum Mutat 22: 121-128 Schumacher B , Hofmann K , Boulton S, & Gartner A (2001) The C. elegans homolog of the p53 tumor suppressor is required for D N A damage-induced apoptosis. Curr Biol 11: 1722-1727 Shiozaki E N , G u L , Y a n N , & Shi Y (2004) Structure o f the B R C T repeats o f B R C A l bound to a B A C H l phosphopepfide: implications for signaling. Mol Cell 14: 405-412 Vahteristo P, Yliannala K , Tamminen A , Eerola H , Blomqvist C , & Nevanliima H (2006) B A C H l Ser919Pro variant and breast cancer risk. BMC Cancer 6: 19  CHAPTER 4^ Characterization of the DOG-l/FANCJ-related helicase BCH-l/JRH-1  4.1 Introduction In C. elegans, there are 4 D O G - 1 helicase-like proteins encoded by four genes: F33H2.1 (dog-1), F25H2.13 (jrh-l), M03C11.3 and Y50D7A.2.  These genes have been implicated i n  genome stability and replication; Y50D7A.2 is a homolog o f the X P D hehcase involved i n nucleotide excision repair, M03C11.2 is a homolog o f the yeast helicase C h l l , which functions i n sister chromatid cohesion, and dog-1 has an important role i n G/C-tract maintenance and D N A repair as the homolog o f human BRIPl/FANCJ.  B C H - l / J R H - 1 is most similar to the human  N H L (novel helicase-like) protein, about which little is known. The mouse homolog o f JRH-1 is R T E L (for regulator o f telomere length), which is proposed to have a role in maintaining genome stability and telomere length (Ding et al., 2004). Loss o f R T E L function results i n an embryonic lethal phenotype, limiting its study in the mouse (Ding et al, 2004). Therefore, investigation was made into the function o f C elegans bch-IIjrh-1, l/FANCJ-relaXod  hereafter referred to as jrh-1 (for dog-  helicase). Because C. elegans jrh-1 mutants are viable and the construction o f  double mutant strains is relatively fast and easy in the nematode, C. elegans is a useful system in which to examine genetic interactions with jrh-1.  Understanding the function o f this gene in C  elegans w i l l add to our understanding o f the functions o f the D O G - 1 helicase-like proteins and might shed light on the role of N H L in humans.  ' A version o f this chapter w i l l be submitted as part o f a collaborative work with the Boulton lab in which the name spar-I w i l l be proposed for F25H2.13 (S. Boulton, p. communication).  4.2 Materials and Methods Unless otherwise indicated, the tml866 allele o f jrh-1 was used for all experiments described in this chapter.  4.2.1 SYT012 staining Methods were performed as described in section 3.2.1  4.2.2 D N A damage sensitivity experiments Methods were performed as described in sections 3.2.2-3.2.4  4.2.3 DAPI staining N 2 , dog-1, jrh-1 and dog-1 jrh-1 animals were staged by picking L4-stage animals and staining 24 hours later. Animals were washed off the plates with M 9 buffer containing TritonXlOO (lOO^il/L) into microfuge tubes and washed three times with M9/TritonX100. The buffer was removed and 200\i\ o f 150nM 4',6-diamidino-2-phenylindole dihydrochloride ( D A P I ; Sigma) in 95% ethanol was added. The animals were placed i n the dark and allowed to incubate for 1 hour. D A P I was removed and animals were washed twice in M9/TritonX100, and destained by soaking in 1ml o f M 9 buffer overnight at 4 ° C . DAPI-stained whole animals were mounted on 3% agarose slides and viewed at 4 0 X magnification on a Zeiss Axioscope fluorescent microscope. A Retiga 2000R camera (Qimaging) and Openlab 4.0.2 software (Improvision) was used to image the germlines. Extracted germlines were prepared by the antibody staining protocol described in section 4.2.4 and imaged as above.  4.2.4 Anti-RAD-51 antibody staining On the day o f staining, 1-day old animals were picked to fresh plates. Slides were prepared by coating with poly-L-lysine solution. Animals were washed off o f plates with M9/TritonX100 buffer, centrifuged at 1200rpm for 2min, then washed I X with M 9 buffer. M 9 was removed from the tubes, leaving the animals in a small volume o f buffer. Worms were pipetted onto slides in 2 m M levamisole and gonads were extracted using 23-gauge needles to cut just below the pharynx. Coverslips were placed over the worms and pressed down lightly to further extrude gonads. The slides were placed onto a cold metal block in dry ice and allowed to freeze for 10 minutes. A razor blade was used to crack off coverslips, and slides were plunged into cold methanol in dry ice for 5 minutes, then cold acetone i n dry ice for 5 minutes. Slides were allowed to dry following removal from the acetone. A P A B pen (Sigma) was used to outline the area o f the specimens on each slide. Slides were blocked to avoid non-specific binding by incubating i n P B S with 0.2% Tween20 and 3Vo Bovine Serum A l b u m i n ( B S A ) for 1 hour in a humid chamber at room temperature. Slides were rinsed 3 X in P B S with 0.2% Tween20 and incubated with 50\xl o f a 1:200 solution o f anti-RAD-51 antibody (a gift from Adriana LaVolpe) in P B S with 3% B S A in a humid chamber overnight at 4 ° C . Slides were washed 3 X i n P B S with 0.2% Tween20 and incubated with 50^1 o f F I T C conjugated goat antirabbit secondary antibody (Molecular Probes) at 1:2500 dilution in P B S with 3% B S A and 0.5|Lig/mL D A P I in a humid chamber. After 1 hour of incubation in secondary antibody, slides were washed 3 X in P B S with 0.2% Tween20, excess liquid was removed and 10^1 o f D A B C O anti-fade reagent was added. Slides were covered with coverslips and viewed on a Zeiss Axioplan 2 fluorescent microscope with lOOX objective. Imaging was done with a Photometries Coolsnap H Q camera and Metamorph 6.1r6 software (Universal Imaging Corporation).  4.2.5 Strains and strain constructions The strains used in this chapter include K R 3 4 9 9 dpy-11 unc-42, K R 3 4 2 dpy-18 unc-25, K R 1 8 0 dpy-17 unc-36, FX1866jrh-1 (tml866), V C 1 3 dog-l(gklO), VhT2[bli-4(e937) Is48] (I;III), B C 2 2 0 0 eTl[unc-36j/dpy-18; nTl/+{\W); nTl/dpy-ll{V\  \C50^acl-3(ok726)  and FX1937 mus-81(tml937).  LM99  smn-l(ok355)  eTl[unc-36]/unc-46,  IV/nTl fqls51j  {IY;Y), V C 1 9 3  KR4145 him-6(ok412)  For the jrh-1 dog-1, jrh-1 mus-81, and jrh-1; him-6 double  mutant strains, the hT2 balancer was used to maintain the strains as heterozygous for jrh-1 (in the cases ofjrh-1 dog-1 and jrh-1; him-6) and heterozygous for both jrh-1 and mus-81 (in the case o f jrh-1 mus-81) because the strains could not be maintained as homozygotes due to severe sickness/lethality. hT2 is a balanced translocation of chromosomes I and III and was used to balance genes on chromosome I.  hT2[gfp] dog-1 (gkl0)/jrh-l(tml866) V C 1 3 dog-l(gklO)  dog-1 (gklO)  males were created by heat-shocking the strain at 30°C for 6 hours.  dog-1 (gklO) males were crossed to F X 1866jrh-1 (tml866) hermaphrodites. Putative cross progeny were allowed to self fertilize and their progeny were plated individually. P C R was used to verify animals o f the genotype jrh-l(tml866) crossed to L M 9 9 smn-l(ok355)  dog-l(gklO)/+  dog-l(gklO).  N 2 males were  VhT2[bli-4(e937) Is48] (I;III) hermaphrodites in order to  generate hT2[bli-4(e937) Is48]/+ males. The hT2 translocation in this strain is marked by the presence o f a G F P insertion ils48), which can be observed on a G F P microscope as G F P expression i n the pharynx o f animals with the translocation. GFP-expressing hT2[bli-4(e937) Is48]/+ males were crossed to jrh-1 (tml866) dog-1 (gklO)/+ dog-1 (gklO) hermaphrodites. G F P expressing progeny o f this cross were plated individually and allowed to lay for 48 hours. Adults were picked off these plates after laying, and were lysed and tested for the heterozygous presence o f jrh-1 (tml866) and dog-1 (gklO). From any plates testing positive, GFP-expressing  progeny were plated individually, allowed to lay for 48 hours, then adults were tested for the homozygous presence o f dog-1 (gklO) in order to obtain the strain hT2[gfp] dog-1 (gklO)/jrhl(tml866)  dog-l(gklO).  The same method was used to construct hT2[gfp]/jrh-l(tml866);  him-  6(ok412).  jrh-1; eTl[unc-36]/dpy-18;  eTl[mc-36]/unc-46  BC2200 males o f genotype eTl[unc-36]/dpy-18;  eTl[unc-36]/unc-46  FX1866jrh-l(tml866) hermaphrodites. F l males o f genotype jrh-l/+; were crossed to jrh-l/+;  eTl[unc-36]/+;  eTl[unc-36]/+  were mated to  dpy-18/+;  unc-46/+  hermaphrodites. L4-stage progeny o f  this cross were plated individually and allowed to lay for 48 hours and then were lysed. P C R was used to determine the presence of the jrh-1 (tml866) homozygous deletion among the animals. The presence of eTl[unc-36]/dpy-18;  eTl[unc-36]/unc-46  was confirmed by  observation o f wild type, Dpy-18 Unc-46 and Unc-36 phenotypes i n the progeny. A similar method was used to construct the jrh-1; nTl[gfp]/+;  nTl/dpy-U  strain, using G F P expression i n  the pharynx [qls517 as a marker for nTl.  hT2[gfp]/jrh-l(tml866)  mus-81 (tml937)  F X 1 9 3 7 mus-81 (tml937) males were crossed to FX1866jrh-1 (tml866) hermaphrodites. Putative cross progeny were allowed to self and 100 o f their progeny were plated individually. P C R was used to find animals o f the genotype jrh-1 (tml866) mus-81 (tml 937)/jrh-l (tml866) +. N 2 males were crossed to L M 9 9 smn-l(ok355) yhT2[bli-4(e937)  Is48] (I;III) hermaphrodites in  order to obtain hT2[bli-4(e937) Is48]/+ males. GFP-expressing hT2[bli-4(e937) Is48]/+ males were crossed to jrh-l(tml866)  mus-81(tml937)/jrh-l(tml866)  + hermaphrodites. G F P -  expressing progeny o f this cross were plated individually and allowed to lay for 48 hours, as  which time adults were removed, lysed and tested for the heterozygous presence o f jrhl(tml866)  and mus-81 (tml937).  4.2.6 Measuring meiotic recombination The frequency o f meiotic recombination was measured in the intervals between dpy-11 and unc42, dpy-18 and unc-25, and dpy-17 and unc-36. Individual animals o f genotype dpy unc/+ + and jrh-1; dpy unc/+ + were plated and transferred daily for four days. In each o f the broods, the number o f wild-types, D p y U n c , Dpys and Unes were scored. Recombination frequency (p) was calculated using the equation: p=l-(l-2(RyT))^ Where:  R=numbers o f recombinants=2(largest recombinant class) T=total progeny=4/3 (number o f wild-types + 1 recombinant class)  The map distance was calculated as pxlOO. Confidence intervals for each recombination frequency were calculated using the statistics o f Crow and Gardner (1959).  4.3 Results 4.3.1 jrh-1 is a member of the dog-1 family and encodes a helicase-like protein The C. elegans dog-1 gene family contains four genes: dog-1, F25H2.13, Y50D7A.2 and M 0 5 C / / . 2 (Figure 4.1).  F33H2.1 {dog-1) - F25H2.13 {jrh-1) Y50D7A.2 (unassigned) 103C11.2 (unassigned) M03C11.2 (unassiqne T04A11.6 {him-6)  dog-1 helicase-like family RecQ helrcase family  Figure 4.1: Phylogram o f genes in the dog-1 and RecQ helicase families.  dog-1 is essential for G/C-tract maintenance in C. elegans (Cheung et al, 2002). Y50D7A.2 is an X P D - l i k e gene, suggesting a possible role in nucleotide excision repair (www.wormbase.org). M03C11.2 is similar to the yeast C h l l helicase, which is thought to function in sister-chromatid cohesion and has a role in maintaining genome integrity following D N A damage (Gerring et al, 1990; Spencer et al, 1990; Skibbens, 2004; Laha et al, 2006). N o viable mutants were available for either Y50D7A.2 or M03C11.2.  F25H2.13 was chosen for further study as mutants in the  gene were viable, and the gene had not been studied in C. elegans. F25H2.13 and D O G - 1 are closely related, having 29% amino acid identity and 48% similarity (http://www.ncbi.nlm.nih.gov/BLAST/). F25H2.13 has been named bch-1 for and is also known as jrh-1 for dog-l/FANCJ-rdated  BACHl-xdaXea,  helicase.  There are three domains in common between JRH-1 and D O G - 1 : D i n G , a RAD-3-related helicase domain, H E L I C c 2 , which is a helicase superfamily C-terminal domain, and D E x D c 2 , the D E A D - l i k e helicases superfamily domain that contains a D E A H - b o x . Despite their similar  domains, the two proteins have different structure (Figure 4.2). D O G - 1 contains an additional domain, D E A D _ 2 , which is a conserved region found in R A D - 3 - l i k e D N A - b i n d i n g helicases and is not present in J R H - 1 .  250  1  F33H2.1 ( D O G - 1 ) |  DExDc2  DExDc2  DinG  600  750  |  |  DinG  HËLICc2  983  DinG  DEAD 2  250  1  F25H2.13{JRH-1)  l  500  i  DExDc2  750  HEUCc2  |  DinG  Figure 4.2: Comparison o f the domain structure o f D O G - 1 and J R H - 1 . Both proteins contain domains that are characteristic of helicases, as determined by the Conserved Domain Database (http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml).  The human gene most similar to jrh-1 is NHL. JRH-1 and N H L share 34% identity and 53%) similarity (http://www.ncbi.nlm.nih.gov/blast/). Currently, there are no published reports on NHL function in human cells. The mouse gene most similar is RTEL, an essential gene that is proposed to fimction in the regulation o f telomere length (Ding et al, 2004).  In C. elegans, the jrh-l(tml866)  mutant is a deletion o f 1346bp begiiming in the first  intron and extending into intron 5, including most o f the D E x D c 2 domain and the D E A H - b o x . The tml866 deletion causes the sequence beyond the deletion to be out o f the normal reading frame and introduces a stop codon in exon 6 (Figure 4.3). null allele of jrh-1.  Therefore, tmlS66 is likely to be a  irh-1(tm1866)  Figure 4.3: jrh-] gene structure, showing the location and result o f the tml866 deletion. The location of the vc8 point mutation is indicated by an arrow. Gene structure adapted from www.wormbase.org.  A second allele of Jrh-1 was generated by T I L L I N G (Targeting Induced Local Lesions in Genomes; Gilchrist et al., 2006). This allele is a point mutation, known as vc8, which is a cytosine to thymine change at nucleotide 1165; this mutation causes a serine to phenylalanine change at amino acid 265 (a conserved residue). The jrh-1 (vc8) and jrh-1 (tml866) mutants are viable, though both have reduced viability.  4.3.2 Phenotypes of jrh-1 mutant animals In examining the jrh-1 (tml866) strain, several phenotypes were observed, jrh-1 mutant animals have a reduced brood size, with an average progeny number o f 68 ± 12 (n=20), which is approximately 4-fold lower than wild type. The animals also display a slow growth phenotype, and take approximately 4.5 days to develop from egg to gravid adult at 20°C, compared to 3.5 days f o r N 2 animals. In addition, 21 percent (12 out o f 56) o f jrh-1 mutants display a protruding vulva phenotype (Pvl), a phenotype that can be caused by unrepaired D N A damage (Weidhaas et al., 2006). These phenotypes might suggest a possible role for jrh-1 during replication or repair.  If jrh-1 mutants contain unrepaired D N A damage, an increase in apoptosis would be observed in the germline as described in Chapter 3. In order to test whether or not jrh-1 mutant  animals display elevated apoptosis that could be due to the presence o f unrepaired D N A damage, the vital dye S Y T O 12 was used. Apoptosis was measured as described previously (Youds et al, 2006 and Chapter 3 o f this thesis). N 2 animals had an average of 0.8 ± 0.2 SYT012-stained corpses per gonad arm (n=44). jrh-1 animals had a significantly higher average o f 2.5 ± 0.3 SYT012-stained corpses (n=57; Figure 4.4). Thus, jrh-1 mutant animals have elevated apoptosis in the germline. The dependence o f the elevated apoptosis on the p53 ortholog cep-1 was not tested due to the tight genetic linkage o f jrh-1 to cep-1 on chromosome I, which made it difficult to construct the double mutant.  Ê  < -a  c o  >10  ro  3 xxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx  N2  J3 3  8-9  X  6-7  xxxx  4-5  xxxxxxxxxxx  2-3  xxxxxxxxxxxxxxxxxxxxxx  0-1  xxxxxxxxxxxxxxxxxxx  jrh-1  Figure 4.4: Number o f SYT012-stained apoptotic corpses in A ) N 2 and B ) jrh-l mutant animals under normal conditions.  4.3.3 jrh-1 mutants are not sensitive to D N A damage induced by U V C or X-ray irradiation Because D O G - 1 has a role in repair o f D N A interstrand cross-links, one possibility was that JRH-1 might also function in D N A repair. In order to test for a role in repair, jrh-1 mutants were tested for sensitivity to D N A damage in the form o f bulky photoproducts (induced by UVC-irradiation) and double strand breaks (induced by X-rays). N 2 and jrh-1 mutant animals were exposed to the D N A damaging agents and sensitivity was measured by scoring embryonic survival 22 to 26 hours following exposure to each agent, jrh-1 mutants were not significantly  sensitive to treatments o f 50 or lOOJ of UVC-irradiation, compared to N 2 animals (t-test p=0.76 at 50J U V C ; p=0.16 at lOOJ U V C ; Figure 4.5A). Similarly, no significant difference in embryonic survival was observed between jrh-1 and N 2 animals treated with 3000 or 4500Rads of X-ray irradiation (p=0.13 at 3000Rads X-ray; p=0.97 at 4500Rads X-ray; Figure 4.5B). These results indicate that JRH-1 does not function in the repair o f bulky adducts or double strand breaks.  B  100 90 80 • 70 •  N2  60 •  —é^ jrh-1  50 •  S 2 a.  40 • 30 • 20 • 10 • 0 0  D o s e of UV254nm (J)  -r  1500  —1 3000  1 — 4500  D o s e of X - r a y ( R a d s )  Figure 4.5: Sensitivity o f N 2 and jrh-1 animals to A ) UVC-irradiation and B ) X-ray treatment. Values plotted represent single experiments. N o significant difference in embryonic survival between N 2 and jrh-1 after U V C or X-ray treatment was replicated in 3 independent experiments.  4.3.4 jrh-1 mutants are moderately sensitive to D N A interstrand cross-linking agents D O G - 1 was shown to function specifically in interstrand cross-link repair (Chapter 3 o f this thesis); therefore, JRH-1 was also tested for a role in cross-link repair. Two slightly different protocols were used to assess the sensitivity of jrh-1 mutant animals to interstrand cross-linking agents (described in section 3.2.4). Trimethylpsoralen ( T M P ) was used together with both low and high powers o f U V A . When jrh-1 animals were exposed to 10|ig/ml T M P followed by 90 seconds o f U V A at 340)jW/cm^ in a 2ml volume o f buffer, no significant  sensitivity to control or cross-linking treatment was observed, compared to N 2 (no treatment p=0.08; U V A only p=0.15; T M P only p=0.83; T M P + U V A p=0.08; Figure 4.6).  CD 100  >  CO o 'c o  90 80 70 60-  0N2  50-  E  lU c O Q) Q. i_  Ujrf)-1  4030-  20 1 10 0no treatment  U V A only  TMP only  TMPi-UVA  Treatment  Figure 4.6: Embryonic survival in the progeny o f N 2 and jrh-1 animals following no treatment, U V A only, T M P only or cross-linking by T M P + U V A treatment.  jrh-1 mutants were also treated for interstrand cross-link sensitivity using a range o f higher U V A power in order to construct a kill-curve. In these experiments,yr/j-7 animals were exposed to lO^g/ml T M P followed by 10, 20 or 30 seconds o f U V A at a power of 550(aW/cm^ in a 50jil volume o f buffer (being approximately equal to 55J, 1 lOJ and 165J U V A ) . A t this higher power, jrh-1 did not show significant sensitivity to U V A only, compared to N 2 animals (p=0.63 at 1 lOJ U V A only). However, jrh-1 was significantly sensitive to the higher doses o f crosslinking treatment (p=0.03 at lO^g/ml T M P + 55J U V A ; p=0.006 at 10)ig/ml T M P + 1 lOJ U V A ; p=0.009 at 10|ag/ml T M P + 165J U V A ; Figure 4.7).  10 u 1 0  1 50  1 100  1 150  1 200  1 0 u g / m l T M P + D o s e of U V A (J)  Figure 4.7: Sensitivity o f N 2 , dog-1 and jrh-1 animais to trimethylpsoralen with U V A ( T M P + U V A ) treatment with 10|xg/ml T M P and 55, 110 or 165J U V A . jrh-1 is significantly sensitive at higher doses o f cross-linking treatment.  Unlike dog-1 mutant animals, which are highly sensitive to cross-linking,7>A-7 animals show moderate sensitivity to T M P + U V A treatment, and only when the treatment uses a higher power o f U V A . jrh-1 appears to be sensitive only when high levels o f cross-linking are present, suggesting that it might play a minor role in cross-link repair.  4.3.5 Loss of JRH-1 function is synthetically lethal in the absence of DOG-1 The phenotypes o f jrh-1 mutant animals and the results o f the D N A damage sensitivity experiments described above suggested a possible role for jrh-1 in D N A repair. In order to test whether or not jrh-1 might function in D N A repair at stalled replication forks, the dog-1 mutant was utilized. Replication fork stalling is proposed to occur in dog-1 mutants as a result o f persisting G/C-tract secondary structures (Youds et al., 2006). If JRH-1 were required for repair at stalled replication forks, a genetic interaction might be expected between jrh-1 and dog-1.  The hT2 balancer, which heterozygously maintains jrh-1, was used to construct the jrh-1  dog-1  double mutant. The dog-1 single mutant animals were healthy, while jrh-1 animals were somewhat sick and slow growing, but could be maintained without difficulty, jrh-1  dog-1  double mutants grew to adulthood, but did not lay any eggs, resulting in complete sterility in the absence o f both o f these genes. Double mutants segregated from the hT2[gfp] dog-1/ jrh-1 dog1 balanced strain also displayed protruding vulvae (Pvl) with 100% penetrance, which is a synergistic enhancement o f the P v l phenotype rarely observed i n dog-1 mutants and present in only 20%) o f jrh-1 single mutants (Figure 4.8). Thus, the double mutant is a synthetic lethal with an enhanced P v l phenotype.  Figure 4.8: Comparison o f A ) N 2 and B ) jrh-1 dog-1 double mutant. The uterus is indicated by small arrows, and the vulva by the arrowhead. Absence o f embryos in the uterus and protruding vulva can be observed in jrh-1 dog-1.  One explanation for the synthetic lethality o f jrh-1 dog-1 animals is that JRH-1 is essential for repair at replication forks that stall in the absence o f D O G - 1 . To look for evidence of defects during replication, double mutant germlines were visualized with D A P I staining, dog1 and jrh-1 single mutant germlines were similar to those o f w i l d type animals, but jrh-1  dog-1  germlines displayed severe defects. The double mutant germlines were underproliferated, and had fragmented chromosomes and enlarged mitotic nuclei, the latter o f which are indicative o f an activated replication checkpoint (Ahmed et al, 2001; Figures 4.9 and 4.10).  Figure 4.9: Comparison o f whole D A P I stained A ) N 2 , B ) dog-1, C) jrh-1 and D) jrh-1 dog-1 double mutant germlines. Massive proliferation defects are evident in jrh-1 dog-1.  Figure 4.10: Mitotic zones o f D A P I stained A ) N 2 , B ) dog-1, C) jrh-1 and D) jrh-1 dog-1 double mutant germlines. The reduced number o f enlarged mitotic nuclei in jrh-1 dog-1 indicate activation o f the mitotic replication checkpoint.  Germline replication defects were so severe in jrh-1 dog-1 that the transition zone was not observed and no nuclei progressed into meiosis, leading to the sterile phenotype observed i n the double mutants. These data indicate that i n the absence o f both o f these helicase-like proteins germline replication has fatal consequences. When neither protein is present, mitotic catastrophe results.  4.3.6 JRH-1 is not required for G/C-tract stability D O G - 1 has been shown to be required to prevent G/C-tract deletions (Cheung et al, 2002), and the frequency o f deletions in dog-l mutants is increased in the absence o f homologous recombination repair (Youds et al, 2006). synthetic lethality observed in jrh-l  Could defects at G/C-tracts cause the  dog-l mutants? In the absence o f D O G - 1 , secondary  structures are proposed to form, some of which might be resolved by J R H - 1 . In the absence o f both D O G - 1 and J R H - 1 , an increased number o f replication blockages at G/C-tracts might result in the mitotic arrest observed. To test this, the frequency o f G/C-tract deletions in jrh-l l dog-1 animals was determined using the assay for deletions at the \ah-l  and jrh-  G/C-tract (Youds et  al, 2006; section 2.2.4). N o deletions were observed in jrh-1 single mutants. N o r was there a significant difference in the frequency of jrh-1 dog-l animals with deletions compared to dog-1 single mutants, using either jrh-1 allele (point mutation vcS or tml866 deletion; Table 4.1). Thus, dog-l and jrh-l  do not have redundant roles at G/C-tracts.  Genotype dog-1 jrh-1 (tml866) jrh-1 (tml866) dog-1 jrh-1 (vc8) dog-1  Deletion Frequency 11.4 0 6.9 12.5  Table 4.1: Frequency o f dog-l and jrh-l with vab-l G/C-tract deletions.  Number of animals tested 228 100 116 112  p-value in t-test with dog-1  0.30 0.81  single mutant and jrh-1 dog-1 double mutant animals  4.3.7 Jrh-1 dog-1 animals show extensive RAD-51 foci in the mitotic germline A more likely explanation for the synthetic lethality between jrh-1 and dog-1 mutants is that JRH-1 is required for D N A repair at replication forks stalled due to G/C-tract secondary structures occurring in the absence o f D O G - 1 . In this case, jrh-1 dog-1 animals would show signs of D N A damage caused by incomplete repair o f stalled forks, such as breaks or persisting  repair intermediates. In order to investigate whether or not jrh-1 dog-1 double mutants had increased D N A damage, anti-RAD-51 antibody was used to stain jrh-1 dog-1 animals. R A D - 5 1 is the C. elegans R e c A homolog required for strand invasion during double strand break (homologous recombination) repair and during meiotic recombination (Rinaldo et al, 1998 & 2002). R A D - 5 1 foci are observed rarely i n the mitotic zone o f the germline o f wild type animals. Foci appear beginning in the transition zone, where R A D - 5 1 localizes to sites o f meiotic double strand breaks during late zygotene to early pachytene; the foci gradually disappear i n late pachytene as the double strand breaks are repaired ( A l p i et al, 2003). When anti-RAD-51 was used to stain dog-1 and jrh-1 single mutants, a few rare foci in the mitotic zone, and many foci beginning in the transition zone and persisting into late pachytene were observed, as expected (Figure 4.11 A , C ) . Similar staining was observed in the control strain, hT2 dog-l/jrh-1  dog-1. However,y>/z-7 dog-1 double mutant animals displayed extensive R A D -  51 foci throughout the mitotic zone arrested nuclei (Figure 4.1 I B , D ) . The abnormal R A D - 5 1 staining pattern indicates that an excess o f homologous recombination repair intermediates are present in the mitotic zones of jrh-1 dog-1 animals.  AhT2 dog-1/jrh-1 dog-1  control  B jrh-1 dog-1  ChT2 dog-1/jrh-1 dog-1 mitotic z o n e  D jrh-1 dog-1 mitotic z o n e  Figure 4.11: Anti-RAD-51 staining of A) hT2 dog-1/jrh-1 dog-1 controls and B) jrh-1 dog-1 germlines. Mitotic zones of C) hT2 dog-l/jrh-1 dog-1 controls and D) jrh-1 dog-1 show excessive RAD-51 foci in the mitotic zones of jrh-1 dog-1 animals, but not in hT2 dog-l/jrh-1 dog-1 control animals. DAPI is shown in green, anti-RAD-51 is shown in red.  4.3.8 Loss of function of JRH-1 in eTl translocation hétérozygotes Homologous recombination repair of double strand breaks during meiosis requires the presence of a homolog. In the case of the translocation eTl (II1;V), the right arm of chromosome 111 is exchanged with the left arm of chromosome V and homolog pairing for those regions is prevented during meiosis (Rosenbluth & Baillie, 1981 ; Edgley et al, 1995). Thus, it is expected that homologous recombination repair of meiotic double strand breaks will not occur in the translocated regions. Therefore, other repair mechanisms must act at meiotic double strand breaks that occur within the translocated portions of the chromosomes. To explore the effect of absence of JRH-1 on repair of the translocated chromosomes, a strain containingjV/?-/ and eTl along with visible markers for chromosomes III and V was constructed. The eTl strain used was  BC2200, which has the genotype eTl[unc-36]/dpy-18(III);  eTl[unc-36J/unc-46(V),  where eTl is  marked by a mutation in unc-36 caused by the translocation breakpoint. The strain segregates with a ratio o f 4 w i l d types (translocation hétérozygotes) to 1 dpy-18; unc-46 (DpyUnc homozygotes) to 1 eTl [unc-36] (Unes that are eTl homozygotes). It quickly became evident that when the jrh-1 mutation was present, the strain had an unusual phenotype. Within a few generations in the jrh-1 background, the homozygotes, dpy-18; unc-46 and unc-36 animals, outcompeted the w i l d type translocation hétérozygotes. Repeat experiments with several different constructions o f the jrh-l;  eTl/dpy-18;  eTl/unc-46  strain always produced the same result. In  the control B C 2 2 0 0 strain, the homozygotes never out-competed the translocation hétérozygotes. Under normal conditions, homozygote animals should never out-compete the translocation hétérozygote animals because the translocation hétérozygotes are more prevalent in the progeny of the heterozygous eTl strain. Therefore, segregants from the jrh-1; eTl/dpy-18;  eTl/unc-46  strain were scored in order to examine brood size and viability. The brood size of jrh-1; eTl/dpy-18;  eTl/unc-46  animals was smaller than expected (observed average o f 36 eggs laid  per animal, compared to 45 expected). However, the brood size o f jrh-1; eTl/eTl  animals was  slightly greater than expected (observed average o f 56 eggs laid per animal, compared to 49 expected). Thus, the absence o f JRH-1 synergistically reduced the brood size o f the eTl translocation hétérozygotes, but not o f eTl translocation homozygotes (Table 4.2).  Genotype  Total brood per animal  N2  271  Percent o F eggs reaching adulthood 99.9  Number of adult progeny per animal 271  jrh-1  68  86.0  58  eTl/dpy-18; eTl/unc-46  180  47.1  85  eTl/eTl  194  96.9  188  jrh-1; eTl/dpy-18; eTl/imc-46  36  40.3  15  56  77.2  43  jrh-1:  eTl/eTl  Table 4.2: Viability oï eTl/dpy-18;  eTl/unc-46  and eTl/eTl  with and without jrh-1  (tml866).  The affect o f the jrh-1 mutation on eTl translocation hétérozygotes was further examined using a competition assay, in which multiple lines o f eTl/dpy-18; eTl/dpy-18; eTl/unc-46  eTl/unc-46  and jrh-1;  animals were maintained by serially transferring a piece o f agar  containing multiple animals ("chunking out" the crowded plate) and observing the transfer at which no heterozygous animals could be observed on the plate. W i l d type heterozygous animals survived in all 20 lines o f eTl/dpy-18; of jrh-1; eTl/dpy-18;  eTl/unc-46  eTl/unc-46 throughout the experiment. However, in lines  animals, wild type hétérozygotes were quickly out-competed  by eTl [unc-36] and dpy-18; unc-46 homozygotes. W i l d type hétérozygotes were out-competed by homozygous animals in all 20 lines after 9 serial transfers oï jrh-1; eTl/dpy-18; (Figure 4.12).  eTl/unc-46  18 -j o  16 -I  o  in c  14 -  0)  12 ^  «  10  3 0  -*  jrh-1: eT1/dpy-18; eT1/unc-46  —m— eT1/dpy-18, eT1/unc-46  8 i [  N  S s. 1  6 4 2  n  i—-0  1  2  3  4  5  6  7  8  1  9  N u m b e r of Serial T r a n s f e r s  Figure 4.12: Competition assay o f eTl/dpy-18; eTl/unc-46 animals.  and jrh-1; eTl/dpy-18;  eTl/unc-46  In order to test whether or not the affect o f absence o f JRH-1 on translocation hétérozygotes was general, and not specific to the eTl translocation, a second translocation was used.  The strain jrh-1; nTI/+ (IV); nTl/dpy-11  control strain nTl/+; nTl/dpy-11 eTl.  ( V ) was constructed and compared to the  in a competition experiment similar to that carried out with  Unlike eTl, homozygous nTl is lethal while dpy-11 is healthy, so it was not surprising that  heterozygous animals were no longer present on 12 o f 20 nTl/+; nTl/dpy-11  control plates after  8 serial transfers. Nevertheless, heterozygous animals were more quickly out-competed in lines of jrh-1; nTl/+; nTl/dpy-11  animals, similar to the observations for eTl translocation  hétérozygotes in the absence o f JRH-1 (Figure 4.13).  I-  16 -  « 3  10-  c  •  jrh-1: nT1/+; nTVdpy-11  -*—nTi/+; nT1/dpy-11  0 2  3  4  5  6  7  8  Number of Serial Transfers  Figure 4.13: Competition assay o f nTI/+; nTl/dpy-U  and jrh-l;  nTl/+; nTl/dpy-11  animals.  In translocation lieterozygotes, lack of meiotic crossing-over in the region o f the translocation has been attributed to the lack o f paired homologs. In the experiments described here, those individuals (translocation hétérozygotes) lacking a paired homolog to repair from had reduced viability relative to individuals with a homolog. Thus, it would appear that defects in jrh-1 mutant strains are made worse by the absence o f a homolog to repair from, implicating JRH-1 in repair processes that are alternatives to homologous recombination repair. The reduced brood size and fitness ofjrh-1; eTlfunc-36J/dpy-18;  eTl [unc-36]/unc-46  and jrh-1;  nTl/+;  nTl/dpy-11 resulted from the absence o f JRH-1 in regions lacking a homologous pairing partner. This result suggests that chromosomes that cannot repair by recombining with their homologous partner encounter problems in the absence o f J R H - 1 .  4.3.9 Meiotic recombination is increased in jrh-1 mutants Based on the synthetic lethality of jrh-1 with dog-1, JRH-1 might be required for repair at stalled replication forks. Evidence for jrh-I function during replication is based on its genetic  interaction with dog-1 as well as its role in reducing the number o f R A D - 5 1 foci in the mitotic germline. The affect o f JRH-1 on translocation hétérozygotes also argues for a role during meiotic double strand break repair that does not involve the homologous chromosome. One possibility is that JRH-1 controls the balance between repair processes by channeling repair intermediates away from the homologous recombination repair pathway into other repair pathways. I f this is the case, a prediction is that recombination would be increased in the absence o f J R H - 1 . To test this hypothesis, meiotic recombination was measured in jrh-1 mutants between three sets o f visible markers: dpy-11 and unc-42, dpy-18 and unc-25, and dpy-17 and unc-36. For the markers dpy-11 and unc-42, in the control strain, 114 recombinants were observed i n 3135 total progeny. Thus, in a wild type background, the map distance was 3.71cM with a 9 5 % confidence interval ranging from 3.05 to 4.44. In the jrh-1; dpy-11 unc-42/++ strain, 66 recombinants were observed in 723 total progeny. In the absence o f J R H - 1 , the map distance between dpy-11 and unc-42 was 9.59 with a 95% confidence interval o f 7.37 to 12.25 (Table 4.3). Between dpy-18 and unc-25 in a wild-type background, 238 recombinants were observed i n 3483 total progeny, giving a map distance o f 7.08cM with a 95% confidence interval of 6.16 to 8.02. In the jrh-1 background, 100 recombinants were observed in 776 total progeny for the dpy-18 unc-25 interval, giving a map distance o f 13.83cM between these two markers, with a 95% confidence interval o f 10.93 to 16.95 (Table 4.3). In a third interval, between dpy-17 unc-36, 32 recombinants were observed in 3299 total progeny, resulting in a map distance o f 0.97cM in a wild-type background, with 95% confidence interval o f 0.65 to 1.35. In the jrh-1 mutant, 18 recombinants were observed in 459 total progeny, resulting in a map distance of 4.00cM with 95% confidence interval o f 2.47 to 6.24 (Table 4.3). These data represent a significant increase in meiotic recombination frequency in all intervals tested in the absence o f J R H - 1 , suggesting that JRH-1 suppresses recombination.  Genotype  Total Progeny  Recombinants  Map Distance in c M (95% CI)  dpy-11 unc-42/++ ( V )  3135  114  3.71 (3.05-4.44)  jrh-1; dpy-11 ut7c-42/++ ( V )  723  66  9.59 (7.37-12.25)  3483  238  7.08 (6.16-8.02)  jrh-1; dpy-18 unc-25/++ (HI)  776  100  13.83 (10.93-16.95)  dpy-J 7 unc-3 6/++{111)  3299  32  0.97 (0.65-1.35)  jrh-1; dpy-17 unc-36/++ (III)  459  18  4.00 (2.47-6.24)  dpy-18  unc-25/++(UY)  Table 4.3: Meiotic recombination frequency in the absence o f J R H - 1 . The number of w i l d types and DpyUncs (non-recombinants), Dpys and Unes (recombinants) were scored for three intervals, which included both gene cluster {dpy-11 unc-42 and dpy-17 unc-36) and chromosome arm regions {dpy-18 unc-25).  4.3.10 him-6/BLM and mus-81 functionally interact with jrh-1 A function for JRH-1 in preventing excessive recombination repair and thereby promoting alternative means for repair is reminiscent o f the function the Srs2 helicase in yeast. Srs2 removes Rad51 from D N A filaments, thereby acting as an anti-recombinase (Krejci et al., 2003; Veaute et al, 2003). This would be important in situations where inhiating homologous recombination repair would be detrimental. In S. cerevisiae, toxic recombination intermediates are thought to be prevented by separate pathways involving Srs2, the R e c Q helicase S g s l , and Mus81 (Fabre et al, 2002). If JRH-1 functions similar to Svs2, jrh-1 should show a genetic interaction with the C. elegans homologs of Sgsl and Mus81, analogous to the yeast data.  In C. elegans, the sgsl homolog is him-6/BLM  {Wicky et al, 2004). The double mutant  jrh-1; him-6 was constructed. While 48.5% o f him-6 progeny and 86%) o f jrh-1 progeny survived to adulthood, only 7.1%o o f jrh-1; him-6 progeny survived to adulthood in the first  generation (Table 4.4). Thus, the viability o f these mutants was strongly reduced in the double mutant.  Genotype  Total Brood Size  jrh-1 (n=20) J him-6 (n=20) jrh-1; him-6 (n=20)  68± 12 214± 14 54± 11  Percent Embryonic Lethality 3.0 ±0.9 49.7 ± 1.4 72.1 ±2.5  Percent Larval Lethality 9.9 ±2.6 1.8 ±0.4 20.8 ±2.5  Percent Viable Progeny 86.0 ±9.0 48.5 ± 1.6 7.1 ± 1.6  Percent Males (of viable progeny) 0.2 ± 0.2 12.1 ± 1.0 14.5 ±3.8  Table 4.4: Phenotypes of jrh-1, him-6 and jrh-1; him-6 animals, n represents the number of parent worms whose progeny were scored. A l l values are ± standard error o f the mean.  In order to further examine the jrh-1; him-6 phenotype, multiple lines of jrh-1;  him-6  were maintained at 2 0 ° C and examined over time to determine the fitness o f the strain compared to jrh-1 and him-6 single mutants. For him-6 animals, 12 out o f 20 lines survived beyond 10 generations, whereas 6 o f 20 lines o f jrh-1 animals survived beyond 10 generations, jrh-1; him-6 animals displayed much reduced fitness compared to either single mutant, as none o f the 20 lines of the double mutants survived beyond 6 generations, indicating that jrh-1; him-6 is synthetic lethal (Figure 4.14).  in  0) (1)  him-6  0  2  4  6  8  10  Generation  Figure 4.14: Fitness o f 20 lines o f N 2 , him-6 and jrh-1 single mutants, and jrh-1; him-6 double mutants maintained at 20°C for 10 generations. Ill  To test whether or not jrh-1 had a genetic interaction with mus-81, the double mutant was constructed. The jrh-1 mus-81 double mutant was synthetic lethal, as jrh-1 mus-81 adult animals laid a small number o f eggs, but none hatched (Table 4.5).  Genotype  Total Brood Size  jrh-I (n=20) mus-81 (n=20) jrh-1 mus-81 (n=30)  68± 12 153± 13 22 ± 7  Percent Embryonic Lethality 3.0 ±0.9 15.2±6.0 100  Percent Viable Progeny 86.0 ±9.0 84.8 ±6.0 0  Table 4.5: Phenotypes ofjrh-1 and mus-81 single mutants and jrh-1 mus-81 double mutants, n represents the number o f parent worms whose progeny were scored. A l l values are ± standard error of the mean.  In C. elegans, him-6 and mus-81 have a synthetic lethal phenotype (N. O ' N e i l , p. communication). Thus, as in yeast, these genes interact in C. elegans, supporting the proposal that these genes represent separate pathways. These data are consistent with a role for JRH-1 in C. elegans that is similar to that o f Srs2 in yeast.  4.4 Discussion 4.4.1 Phenotypes oîjrh-1  mutant animais support a role in replication or repair  The jrh-1 (tml866) strain observed under normal growth conditions has several phenotypes that are suggestive o f defects in repair, jrh-1 mutants display a slow growth phenotype. However, unlike clk-2 mutants, which have slow development and elongated lifespan (Lakowski & H e k i m i , 1996), they do not live longer than N 2 animals. The slow growth of jrh-1 mutants could be related to repair defects requiring additional developmental time. A protruding vulva phenotype, like that present in approximately 20% of jrh-1 mutants, has been attributed to vulval cell death in N 2 animals following D N A damage (Weidhaas et al., 2006). Elevated apoptosis in the jrh-1 germline, similar to dog-1 mutants, is consistent with defects in repair that activate the cep-l/p53 D N A damage checkpoint, triggering apoptosis. The dependence o f the elevated apoptosis on the cep-l/p53 pathway was not tested due to the tight genetic linkage of jrh-1 and cep-1 on chromosome I; therefore, it is also possible that the elevated apoptosis is not D N A damage-induced, but might be physiological apoptosis.  Further evidence for a role in repair comes from the results o f the D N A damage sensitivity assays on jrh-1 mutants, jrh-1 mutants are not significantly sensitive to damage induced by X-rays, indicating that JRH-1 does not function in the homologous recombination repair or non-homologous end joining pathways, which are the major pathways for double strand break repair (reviewed in W y m a n & Kanaar, 2006). S i m i l a r l y , j r ^ - / mutants do not display significant hypersensitivity to the bulky adducts induced by UVC-irradiation (reviewed in D i p et al, 2004), indicating that jrh-1 does not play a role in nucleotide excision repair. However, yr/z-7 animals were moderately sensitive to cross-linking treatment by T M P + U V A , suggesting a possible role for JRH-1 in the cross-link repair pathway.  Compared to dog-I mutants,yr/2-i animals were less sensitive to T M P + U V A . dog-I mutants were highly sensitive to cross-linking treatment at all o f the powers tested, whereas jrh-I mutants only showed significant sensitivity at higher doses, when the animals were treated with higher power U V A (i.e., more photons per second). These differences in sensitivity might reflect the different effects o f l o w and high power U V A on the rate o f monoadduct (TMP-bound D N A ) versus cross-link ( D N A - D N A ) formation, and the activity o f JRH-1 and D O G - 1 on these repair substrates. During cross-linking treatment, T M P intercalates with the D N A . Upon U V A exposure, D N A - b o u n d T M P monoadducts are formed when a photon is absorbed, whereas crosslinks between two nearby monoadducts are formed when a second photon is absorbed (Johnston et al, 1981). Damage from T M P + U V A exposure is in the form o f both T M P - D N A monoadducts and intra- and interstrand cross-links. The proportion o f cross-links to monoadducts should increase with higher U V A power, dog-1 mutant animals are extremely sensitive at all doses, possibly because D O G - 1 is involved in the repair o f both types o f lesions. This would be consistent with the role o f D O G - 1 in the maintenance o f G/C-tracts. J R H - 1 , on the other hand, appears to be involved preferentially in repair o f cross-links because it shows moderate sensitivity only at higher power o f cross-linking. A n alternative explanation for the different sensitivities o f dog-1 and Jrh-1 is that the two gene products function in different parts of the cross-link repair pathway. D O G - 1 might function early in cross-link repair, where it could be required at all cross-linked substrates. Conversely, JRH-1 might act late in the pathway, for example, in the restart the replication fork after the cross-link has been removed. There are several different ways that a replication fork can be restarted (reviewed i n Branzei & Foiani, 2005); therefore, the pathway might branch at this point, causing the absence o f JRH-1 to have only a minor effect on viability following cross-linking treatment.  4.4.2 JRH-1 is required for repair in the absence of DOG-1 The synthetic lethaUty between jrh-1 and dog-1 also supports a role for JRH-1 i n repair. jrh-1 dog-1 double mutants are completely sterile and have enlarged mitotic nuclei and an overall reduction in the number o f nuclei in the mitotic zone. This germline mitotic catastrophe is evidence o f the activated S-phase replication checkpoint, similar to what is observed i n hydroxyurea treated N 2 animals (Ahmed et al, 2001). In the absence o f both helicase-like proteins, the progression o f the germline nuclei is arrested. Therefore, JRH-1 is clearly required in the absence o f D O G - 1 . D O G - 1 maintains G/C-tracts, which delete in its absence (Cheung et al, 2002). Examination o f the vab-1 G/C-tract in jrh-1 animals revealed no deletions. In addition, the frequency o f jrh-1 dog-1 mutants with deletions was not significantly different from the dog-1 frequency, indicating that jrh-1 does not have a redundant role with dog-1 at G / C tracts. The finding that JRH-1 is required in the absence o f D O G - 1 , but that the two proteins do not have redundant functions suggests that JRH-1 functions i n repair at stalled replication forks.  When jrh-1 dog-1 animals were examined for signs o f D N A damage, excessive R A D - 5 1 foci were observed. One possibility is that the R A D - 5 1 foci are the result o f inefficient completion o f homologous recombination repair in the absence o f J R H - 1 . However, the data do not support a role for JRH-1 in homologous recombination repair because in jrh-1 dog-1 double mutants there is no change in the frequency of animals with G/C-tract deletions compared to dog-I single mutants. This result differs from the increase in deletions observed when components o f homologous recombination repair are absent in the dog-1 background (Youds et al, 2006; Chapter 2 o f this thesis). In addition,yr/z-7 mutants are not sensitive to D N A double strand breaks from X-ray treatment, indicating that the homologous recombination repair and non-homologous end-joining pathways are functional in jrh-1 mutants. Nevertheless, that the R A D - 5 1 foci appear on arrested mitotic nuclei in jrh-1 dog-1 animals would be consistent with a  role for JRH-1 in resolving or removing R A D - 5 1 from repair intermediates. The unresolved foci lead to activation o f the replication checkpoint; this is an indication that severe D N A damage must be present in jrh-1 dog-1 animals.  4.4.3 Absence of JRH-1 specifically affects eTl and nil  translocation hétérozygotes  The translocations eTl and nTl were used to block homologous recombination repair o f meiotic double strand breaks in the translocated regions in a jrh-1 mutant background. During meiosis C. elegans, chromosomes pair and synapse with their homologs, and double strand breaks are created along the chromosomes by the SPO-11 protein in late zygotene to early pachytene (Dernburg et al, 1998). One o f these double strand breaks becomes a substrate for meiotic crossover (exchange o f flanking markers between homologs) during pachytene. Other breaks are repaired through gene conversion (no exchange o f flanking markers).  In the case o f the heterozygous reciprocal translocation eTl, the translocated portions o f chromosomes III and V are not paired with their homologs. In C. elegans, homolog recognition regions determine the chromosome pairing partners ( M c K i m et al, 1988). Therefore, in eTl translocation hétérozygotes, chromosome V with the translocated right arm o f chromosome III is paired with chromosome V , while chromosome III with the translocated left arm o f chromosome V is paired with chromosome III (Rosenbluth & Baillie, 1981). In the translocated regions o f chromosomes III and V , meiotic recombination does not occur because o f the lack o f a homolog. A recent report by Smolikov et al. (2007) indicates that when homologous recombination (interhomolog) repair is impaired, sister chromatid-directed repair contributes to the repair o f meiotic double strand breaks. Normally, the synaptonemal complex between sister chromatids acts as a barrier to prevent meiotic double strand break repair between sister chromatids, promoting repair between homologs (Schwacha & Kleckner, 1997; Wan et al, 2004; Webber et al, 2004; N i u et  al, 2005). However, sister chromatid-directed repair can occur once the synaptonemal complex breaks down between sister chromatids in late pachytene. A l p i et al (2003) have shown that in animals heterozygous for the eTl translocation, R A D - 5 1 foci persist into late pachytene in the non-homologously paired regions chromosomes III and V . Thus, repair o f meiotic double strand breaks on the translocated chromosomes in eTl hétérozygotes likely occurs by sister chromatiddirected repair after the dissolution o f the synaptonemal complex between sister chromatids.  Translocation hétérozygote animals lacking JRH-1 had reduced broods and fitness compared to translocation homozygotes in the jrh-1 background. The phenotype of jrh-1; eTl/dpy-18;  eTl/unc-46  animals could be explained by defects in sister chromatid-directed  repair in the absence o f J R H - 1 . Under this hypothesis, in jrh-1; eTl/dpy-18;  eTl/unc-46  animals, any meiotic double strand breaks in the translocated region would not be able to complete sister chromatid-directed repair because JRH-1 is absent. The lack o f repair would result in meiotic double strand breaks that persist and lead to fragmented chromosomes later in diakinesis, and ultimately lethality. A n t i - R A D - 5 1 staining done on jrh-1; eTl/dpy-18;  eTl/unc-  46 animals by collaborators in the Boulton lab showed that R A D - 5 1 staining persists beyond late pachytene, even longer than the R A D - 5 1 foci observed in eTl hétérozygotes. Fragmented chromosomes are also present at diakinesis in the translocation hétérozygotes in the absence o f JRH-1 (J. Ward & S. Boulton, p. communication). These observations suggest that sister chromatid-directed repair is compromised in the absence o f J R H - 1 . Figure 4.15 summarizes a model for how the absence o f JRH-1 affects viability oïeTl  translocation hétérozygotes.  A Wild type C. elegans germline Mitosis  -  o  O  Leptotene  O  O  O  Pachytene  Zygotene  O  O  O  O  o o o ° oo// o p> ^ ©  Q  Q  Q  O  © ® o  O  O  Oogenesis  Diakinesis  Diplotene  O  U  0  O  n o ^ o  0  o  o  meiolic nSR meiotic crossing formation or repair by gene  eT1/+  wild type  jrh-1  jrh-1: eT1/+  Zygotene  Legend homologous  Pachytene (early)  — —  chromosomes  MM  Sister chromatids synaptonemal complex  Pachytene (late)  Diplotene  ^  "~y" —r\-  meiotic double strand break  -^m^ eT1 translocated chromosome  DC  X ^  meiotic crossover event sister chromatid exchange gene conversion event  Diakinesis  Figure 4.15: Model for the affect of JRH-I absence on eTl hétérozygotes. A ) Schematic of the wild type C. elegans germline. Nuclei in the mitotic zone undergo mitosis, serving as the stem cell compartment for the germline. Nuclei progress through the transition zone, where the characteristic crescent shaped D N A is visible within nuclei, and begin the early stages of meiosis. SPO-11-induced double strand breaks occur in late zygotene/early pachytene (indicated by red dots) and are resolved by meiotic crossing-over or through gene conversion. In diakinesis, six bivalents can be observed in each nucleus. B) Schematic of activity during meiosis. In wild type animals, homologous chromosomes pair and synapse. SPO-11-induced double strand breaks are formed along the chromosomes. One break per chromosome is resolved by crossing-over between homologs. Other breaks are repaired by gene conversion between homologs. In the case of a heterozygous eTl translocated chromosome, breaks that occur within the translocated region cannot be repaired through gene conversion or homologous crossover due to lack of homology. Instead, these breaks must be repaired through a repair mechanism using the sister chromatid. This must occur after dissolution of the synaptonemal complex (SC), as the SC prevents sister chromatid exchange. In jrh-l mutants with the heterozygous eTl translocation, homologous recombination repair is not possible in the translocated region, and the sister chromatid must be used as a template for repair. Defects in sister chromatid repair due to the absence of JRH-1 lead to fragmented chromosomes after the breakdown of the SC in jrh-1; eTl/+ animals.  If JRH-1 functions in sister chromatid-directed repair, it could do so by two possible roles. JRH-1 could act directly in the repair process, or it could function to promote sister chromatid-directed repair. Repair from the sister chromatid occurs frequently during replication, thus it is unlikely that any mutant lacking the ability to repair damage using the sister chromatid would be viable. A more plausible flinction for JRH-1 is that it regulates the balance between different types o f repair. JRH-1 might channel repair intermediates away from homologous recombination repair, thereby promoting sister chromatid-directed repair. If this were the case, in the absence o f J R H - 1 , homologous recombination repair might occur inappropriately and/or excessively. Indeed, meiotic recombination was significantly increased in all three visible marker intervals tested in jrh-1 mutants compared to wild-type controls, supporting the idea that JRH-1 functions in the prevention o f excessive recombination repair. Importantly, increases in meiotic recombination were observed for both gene cluster and chromosome arm regions, suggesting a global increase in recombination in the absence o f JRH-1 rather than an alteration in the distribution o f meiotic crossovers.  Inappropriate homologous recombination can cause chromosomal aberrations (Chaganti et al, 1974; Whoriskey et al, 1991; W u & Hickson, 2001; W u , 2007), and is thus problematic. In dog-1 mutants, for example, unresolved secondary structures caused by G/C-tracts could put a heavy demand on repair mechanisms. In the presence o f J R H - 1 , many repair intermediates could be directed away from homologous recombination repair into other repair pathways. In the absence o f J R H - 1 , a greater number of recombination repair events may be initiated, resulting in excessive R A D - 5 1 foci and ultimately, mitotic catastrophe in jrh-1 dog-1 mutants. The channeling o f repair intermediates into certain pathways would be important to ensure efficient  replication, and loss o f this function may explain the slow growth and protruding vulvae defects observed in jrh-1 mutants.  4.4.4 JRH-1 might have a function similar to the yeast helicase Srs2 Fabre et al. (2002) have shown that in S. cerevisiae, toxic homologous recombination repair intermediates are prevented by separate pathways involving Srs2, Mus81 and the RecQ helicase S g s l . Srs2 is thought to do this by removing R A D 5 1 from recombination repair intermediates, thereby dissociating the D-loop, and promoting non-recombinogenic types o f repair, such as synthesis-dependent strand annealing ( S D S A ; Robert et al, 2006; reviewed in Macris & Sung, 2005). M u s B l has multiple possible functions; M u s S l might be able to cleave double Holliday junctions, resulting in a non-recombinational outcome o f repair following replication (Osman & Whitby, 2007). In addition, M u s S l might be involved in synthesisdependent strand annealing ( S D S A ) , where it could prevent the D N A end from reinvading the donor duplex thereby avoiding the formation of a recombination repair intermediate (Robert et al, 2006). S g s l / B L M processes recombination repair intermediates through double Holliday junction dissolution, which leads to non-crossover outcomes, and is also able to dissociate D loop recombination repair intermediates, thereby providing additional means for nonrecombinational outcome o f repair o f a stalled fork (Bennett et al, 1999; Karow et al, 2000; W u & Hickson, 2003; Liberi et al, 2005; Bartos et al, 2006). Thus, it appears that all three of these genes have a role in preventing recombinational outcomes o f repair, each by a different mechanism (Refer to Figure 1.2).  Analysis o f double mutants in S. cerevisiae has been used to separate the functions o f Srs2, M u s S l and S g s l . Knock-out o f sgsl and srs2 causes a slow growth phenotype that is suppressed by disrupting homologous recombination through mutation o f rad51 (Fabre et al,  2002; Schmidt & Kolodner, 2006). Similarily, mutation o f mus81 and sgsl causes a synthetic lethal phenotype, while mus81 srs2 double mutants display mild growth defects (Mullen et al, 2001; Fabre et al, 2002). Based on this data, sgsl and srs2 represent separate pathways, as do mus81 and sgsl. The relationship between srs2 and musSl is less clear, as these genes might be part o f the same pathway but also might function independently in S. cerevisiae (Fabre et al, 2002).  In C. elegans, the genetic interactions shown here between mus81, him-6 and jrh-1 support the idea o f an srs2-like role for jrh-1.  Similar to sgsl and mus81 in yeast, him-6 and  mus-81 have a synthetic lethal phenotype in C. elegans (N. O ' N e i l , p. communication), indicating that him-6 and mus-81 function in parallel pathways. Mutation o f jrh-I causes a synthetic lethal phenotype with him-6, reminiscent o f yeast srs2 sgsl double mutants. In addition, jr/z-7 was synthetic lethal with mus-81. The interaction between srs2 and mus81 in S. cerevisiae is lesser than that observed between jrh-1 and mus-81 in C. elegans, but this may be due to the differences between single-celled and multicellular organisms. Based on the genetic data alone, it is not clear whether JRH-1 functions to prevent homologous recombination repair by the same means as Srs2. However, the genetic interactions o f jrh-1 i n C. elegans are similar to those o f srs2 in yeast.  In future, molecular experiments should be carried out to confirm the genetic findings for jrh-1 reported here. Biochemical studies should test whether or not JRH-1 is able to dissociate R A D - 5 1 protein from D N A filaments, as Srs2 does (Krejci et al, 2003; Veaute et al, 2003). Helicase assays that test the activity o f JRH-1 on different substrates, such as recombination intermediates, forked duplexes and D N A secondary structures would be informative. In addition, interspecies rescue experiments would provide definitive evidence that JRH-1 functions  in tlie same manner as Srs2. In tliese experiments,7>/z-i could be introduced into yeast srs2 mutants to try to rescue the sickness o f the strain, as well as the synthetic lethality o f srs2 sgsl double mutants. Similar experiments could be done with human NHL cells or with mouse  RTEL,  in which srs2 replaces these genes in order to test whether the human and mouse jrh-1 homologs have functional similarity to srs2.  4.4.5 A role for JRH-1 at telomeres? Findings from the mouse suggest that R T E L functions at telomeres, where it is proposed to unwind G-rich telomere repeat structures, thereby helping to maintain telomere length (Ding et al, 2004). R T E L is thought to function in a similar manner i n maintaining genome stability at G-rich D N A sites throughout the genome (Ding et al, 2004). The telomeres of jrh-1 animals were not specifically examined as part o f this study. C. elegans telomere maintenance mutants such as mrt-2 have a mortal germline (Mrt) phenotype that is characterized by the inability to maintain the strain beyond several generations, presumably due to progressive telomere shortening (Ahmed & Hodgkin, 2000). jrh-1 mutants do show a M r t phenotype when individual lines are propagated by transferring single individuals (Figure 4.12), though the strain can be maintained indefinitely by transferring multiple animals. The chromosome fusion events that occur in mrt-2 mutant animals (Ahmed & Hodgkin, 2000) were not observed in jrh-1 mutants. However, it is easy to see how regulating or preventing homologous recombination between telomeres could be an important function for J R H - 1 . If too much homologous recombination occurred between telomeres, this would lead to gains at some telomeres, and losses at others, which could be problemafic. Thus, the role for JRH-1 in genome stability could be relevant to telomeres, as well as replication forks and other repair intermediates. The role for JRH-1 at other sites throughout the genome, such as stalled replication forks, might overshadow a more minor role at telomeres in C. elegans.  Future experiments could examine jrh-1 for telomere defects,  and any aberrations might be more obviously detectable in a cep-l/p53 background because the absence o f the D N A damage checkpoint would exaggerate any telomere maintenance defects.  The experiments described in this chapter characterize the function o f the related gene jrh-1.  dog-l/FANCJ-  The excessive R A D - 5 1 foci in jrh-1 dog-1 animals along with the affect o f  the jrh-1 mutation on translocation hétérozygotes, increased meiotic recombination, and the genetic interactions between jrh-1 and dog-1, him-6 and mus-81 support a possible role for J R H 1 in channeling repair intermediates away from homologous recombinational repair. This activity would be particularly important following replication fork stalling, such as that at G / C tract secondary structures i n the dog-1 mutant, or following removal o f a lesion that has stalled a replication fork, such as an interstrand cross-link.  Chapter 4 References Ahmed S, A l p i A , Hengartner M O , & Gartner A (2001) C. elegans R A D - 5 / C L K - 2 defines a new D N A damage checkpoint protein. Curr Biol 11: 1934-1944 A l p i A , Pasierbek P, Gartner A , & L o i d l J (2003) Genetic and cytological characterization o f the recombination protein R A D - 5 1 in Caenorhabditis elegans. Chromosoma 112: 6-16 Bartos J D , Wang W , Pike J E , & Bambara R A (2006) Mechanisms by which B l o o m protein can disrupt recombination intermediates o f Okazaki fragment maturation. J Biol Chem 281: 3222732239 Bennett R J , K e c k J L , & Wang J C (1999) Binding specificity determines polarity of D N A unwinding by the S g s l protein o f S. cerevisiae. J Mol Biol 289: 235-248 Branzei D & Foiani M (2005) The D N A damage response during D N A replication. Curr Opin Cell Biol 17: 568-575 Chaganti R S , Schonberg S, & German J (1974) A manyfold increase in sister chromatid exchanges in Bloom's syndrome lymphocytes. Proc Natl Acad Sci US All: 4508-4512 Cheung I, Schertzer M , Rose A , & Lansdorp P M (2002) Disruption o f dog-1 in Caenorhabditis elegans triggers deletions upstream o f guanine-rich D N A . Nat Genet 31: 405-409 Crow & Gardner (1959) Confidence intervals for the expectation o f a Poisson variable. Biometrika 46: 441-7. Dernburg A , M c D o n a l d K , Moulder G , Barstead R, Dresser M , Villeneuve A (1998) Meiotic recombination in C. elegans initiates by a conserved mechanism and is dispensible for homologous chromosome synapsis. Cell 94(3): 387-398. Ding H , Schertzer M , W u X , Gertsenstein M , Selig S, K a m m o r i M , Pourvali R, Poon S, Vulto I, Chavez E , T a m P P , Nagy A , & Lansdorp P M (2004) Regulation o f murine telomere length by Rtel: an essential gene encoding a helicase-like protein. Cell 117: 873-886 Dip R, Camenisch U , & Naegeli H (2004) Mechanisms o f D N A damage recognition and strand discrimination in human nucleotide excision repair. DNA Repair (Amst) 3: 1409-1423 Doe C L , A h n JS, D i x o n J, & Whitby M C (2002) M u s S l - E m e l and R q h l involvement in processing stalled and collapsed replication forks. J Biol Chem 211: 32753-32759 Edgley M L , Baillie D L , Riddle D L & Rose A M (1995) Genetic balancers. Caenorhabditis elegans, modern biological analysis o f an organism. Academic press, San Diego, 147-184. Fabre F, Chan A , Heyer W D , & Gangloff S (2002) Ahemate pathways involving Sgsl/Top3, M u s S l / M m s 4 , and Srs2 prevent formation o f toxic recombination intermediates from singlestranded gaps created by D N A replication. Proc Natl Acad Sci U S A 99: 16887-16892  Gerring S L , Spencer F , & Hieter P (1990) The C H L 1 ( C T F 1) gene product of Saccharomyces cerevisiae is important for chromosome transmission and normal cell cycle progression in G 2 / M . EMBO J 9: 4347-4358 Gilchrist E J , O'Neil N J , Rose A M , Zetka M C , & Haughn G W (2006) T I L L I N G is an effective reverse genetics technique for Caenorhabditis elegans. BMC Genomics 7: 262 Johnston B H , K u n g A H , Moore C B , & Hearst J E (1981) Kinetics o f formation o f deoxyribonucleic acid cross-links by 4'-(aminomethyl)-4,5',8-trimethylpsoralen. Biochemistry 20:735-738 Karow J K , Constantinou A , L i J L , West S C , & Hickson ID (2000) The Bloom's syndrome gene product promotes branch migration o f Holliday junctions. Proc Nad Acad Sci USA 91: 65046508 Krejci L , V a n K o m e n S, L i Y , V i l l e m a i n J, Reddy M S , K l e i n H , Ellenberger T, & Sung P (2003) D N A helicase Srs2 disrupts the Rad51 presynaptic filament. Nature 423: 305-309 Laha S, Das SP, Hajra S, Sau S, Sinha P (2006) The budding yeast protein C h l l p is required to preserve genome integrity upon D N A damage in S-phase. Nucleic Acids Res 34(20): 5880-5891 Lakowski B & H e k i m i S (1996) Determination o f life-span in Caenorhabditis clock genes. Science 272: 1010-1013  elegans by four  Liberi G , Maffioletti G , L u c c a C , Chiolo I, Baryshnikova A , Cotta-Ramusino C , Lopes M , Pellicioli A , Haber J E , & Foiani M (2005) Rad51-dependent D N A structures accumulate at damaged replication forks in sgsl mutants defective in the yeast ortholog o f B L M RecQ helicase. Genes Dev 19: 339-350 Macris M A & Sung P (2005) Multifaceted role o f the Saccharomyces cerevisiae Srs2 helicase in homologous recombination regulation. Biochem Soc Trans 33: 1447-1450 M c K i m K S , H o w e l l A M , Rose A M (1988) The effects o f translocations on recombination frequency i n Caenorhabditis elegans. Genetics 120(4): 987-1001. Mullen JR, Kaliraman V , Ibrahim SS, & B r i l l SJ (2001) Requirement for three novel protein complexes in the absence o f the Sgsl D N A helicase in Ssccharomyces cerevisiae. Genetics 157: 103-118 N i u H , Wan L , Baumgartner B , Schaefer D , L o i d l J, HoUingsworth N M (2005) Partner choice during meiosis is regulated by Hopl-promoted dimerization o f M e k l . Mol Biol Cell 16(12): 5804-18. Osman F & Whitby M C (2007) Exploring the roles o f M u s 8 1 - E m e l / M m s 4 at perturbed replication forks. DNA Repair (Amst) Rinaldo C , Bazzicalupo P, Ederle S, Hilliard M , & L a Volpe A (2002) Roles for Caenorhabditis elegans rad-51 in meiosis and in resistance to ionizing radiation during development. Genetics 160:471-479  Rinaldo C , Ederle S, Rocco V , & L a Volpe A (1998) The Caenorhabditis elegans R A D 5 1 homolog is transcribed into two alternative m R N A s potentially encoding proteins of different sizes. Mol Gen Genet 260: 289-294 Robert T, Dervins D , Fabre F, & Gangloff S (2006) M r c l and Srs2 are major actors in the regulation o f spontaneous crossover. EMBO J 25: 2837-2846 Rosenbluth R E & Baillie D L (1981) The genetic analysis o f a reciprocal translocation, V ) , in Caenorhabditis elegans. Genetics 99: 415-428  eTlÇlll;  Schmidt K H & Kolodner R D (2006) Suppression o f spontaneous genome rearrangements in yeast D N A helicase mutants. Proc Nad Acad Sci iyS'y^ 103: 18196-18201 Schwacha A & Kleckner N (1997) Interhomolog bias during meiotic recombination: meiotic function promote a highly differentiated interhomolog-only pathway. Cell 90: 1123-1135. Skibbens R V (2004) C h l l p , a D N A helicase-like protein in budding yeast, functions in sisterchromatid cohesion. Genetics 166: 33-42 Smolikov S, Eizinger A , Hurlburt A , Rogers E , Villeneuve A M , & Colaiacovo M P (2007) Synapsis-defective mutants reveal a correlation between chromosome conformation and the mode o f double-strand break repair during C. elegans meiosis. Genetics Spencer F , Gerring S L , Connelly C , & Hieter P (1990) Mitotic chromosome transmission fidelity mutants in Saccharomyces cerevisiae. Genetics 124: 237-249 Veaute X , Jeusset J, Soustelle C , Kowalczykowski S C , L e C a m E , & Fabre F (2003) The Srs2 helicase prevents recombination by disrupting Rad51 nucleoprotein filaments. Nature 423: 309312 Wan L , de los Santos T, Zhang C , Shokat K , HoUingsworth N M (2004) M e k l kinase activity functions downstream o f R E D 1 in the regulation o f meiotic double strand break repair in budding yeast. Mol Biol Cell 15: 11-23 Webber H A , Howard L , B i c k e l S E (2004) The cohesion protein O R D is required for homolog bias during meiotic recombination. JCell Biol 164: 819-829 Weidhaas J B , Eisenmann D M , Holub J M , & Nallur S V (2006) A Caenorhabditis elegans tissue model of radiation-induced reproductive cell death. Proc Nad Acad Sci USA 103: 9946-9951 Whoriskey S K , Schofield M A , & M i l l e r J H (1991) Isolation and characterization o f Escherichia coli mutants with altered rates o f deletion formation. Genetics 127: 21-30 Wormbase website: www.wormbase.org. Release WS178, August 2007 W u L (2007) Role o f the B L M helicase in replication fork management. DNA Repair (Amst) W u L & Hickson ID (2003) The Bloom's syndrome helicase suppresses crossing over during homologous recombination. Nature 426: 870-874  W u L & Hickson I D (2001) R e c Q helicases and topoisomerases: components o f a conserved complex for the regulation o f genetic recombination. Cell Mol Life Sci 58: 894-901 Wyman C & Kanaar R (2006) D N A double-strand break repair: all's well that ends w e l l . Annu Rev Genet 40: 363-383 Youds J L , O'Neil N J , & Rose A M (2006) Homologous recombination is required for genome stability in the absence o f D O G - 1 in Caenorhabditis elegans. Genetics 173: 697-708  CHAPTER 5: General Discussion 5.1 C. elegans as a model to study interstrand cross-link repair dog-1 has a conserved function with human BRIPl/FANCJ.  In humans, the F A pathway  includes at least 13 proteins: F A N C A , B , C , D 1 / B R C A 2 , D 2 , E , F , G , I, J / B R I P l , L , M and N / P A L B 2 (Taniguchi & D ' A n d r e a , 2006; Reid et al, 2007). Only 5 o f these proteins, F A N C D 1 / B R C A 2 , F A N C D 2 , F A N C J / B R I P l , F A N C L and F A N C M , have known homologs i n C. elegans (Dequen et al, 2005; Collis et al, 2006; Patel & Joenje, 2007). The other F A proteins might exist i n C. elegans, but might not have been identified due to lack o f sequence conservation. Alternatively, the F A proteins currently identified in C. elegans might represent the minimum set o f proteins that make up the F A pathway i n lower organisms. Therefore, C. elegans might represent a simplified model for the F A interstrand cross-link repair pathway.  In humans, F A N C M becomes hyperphosphorylated after D N A damage, and is involved in the early steps o f F A core complex assembly, while components o f the core complex including F A N C L function i n the ubiquitylation o f F A N C D 2 (Meetei et al, 2005; Levitus etal, 2006). In C. elegans, the F A N C M homolog might function in the initial response to the stalled replication fork, while the F A N C L homolog could be responsible for the monoubiquitylation o f F C D - 2 / F A N C D 2 . Because the downstream components o f the F A pathway are not well understood, study o f D O G - 1 and its function relative to the other C. elegans F A proteins could be an important means to dissect this pathway. In particular, identification o f the substrates on which D O G - 1 functions and examination o f possible interactions between D O G - 1 and B R C 2 / B R C A 2 w i l l be informative in understanding the response to interstrand cross-links downstream o f F C D - 2 / F A N C D 2 .  Current knowledge o f D O G - 1 in C. elegans might apply to human B R I P l / F A N C J , which has not yet been extensively studied. Data described in this thesis indicate that the homologous recombination repair and translesion synthesis proteins are crucial in preventing deletions at G/C-tracts in the absence o f D O G - 1 . Though it is not clear whether G/C-tract secondary structures and interstrand cross-links represent a common substrate, it is possible that homologous recombination repair and translesion synthesis w i l l be important for maintaining genome stability i n the absence o f B R I P l / F A N C J . Furthermore, BRIPl/FANCJ  might show  conserved genetic interactions with BLM, MUS81 and NHL in humans, just as dog-1 has been shown to functionally interact with him-6/BLM, mus-81 and jrh-I/NHL  i n C elegans.  5.2 A potential role for human B R I P l / F A N C J at G-rich D N A There a multiple sites o f G-rich D N A in the human genome, including at the telomeres, ribosomal D N A , certain microsatellites, and the immunoglobulin heavy chain switch regions (Maizels, 2006). Algorithms that search for G-rich sequences that could potentially form secondary structures, such as G-quadruplex, have identified more than 300,000 sites in the human genome (Huppert & Balasubramanian, 2005). These sites have the potential to form secondary structures whenever the D N A is single stranded, as it is during replication and transcription. Therefore, it seems reasonable that there might be a protein that unwinds these types o f structures i n humans, as D O G - 1 has been proposed to do in C. elegans. Based on the sequence and other functional similarity between DOG-1 and B R I P l / F A N C J , it is possible that B R I P l / F A N C J might carry out this function in humans. If B R I P l / F A N C J acts both in cross-link repair and in maintaining G-rich D N A by unwinding secondary structures in these sequences, what are the implications for Fanconi anemia? A real possibility is that the formation o f endogenous secondary structures is a source o f genome instability i n F A patients.  Generally, exogenous agents are required for D N A cross-link formation, and although cross-links are a very toxic type o f damage, D N A cross-linking is likely a rare event. However, endogenously forming D N A secondary structures could form very frequently, during any time when the D N A is single stranded. If the protein that normally manages these structures is absent, there would be great potential for genome instability at G-rich D N A sites throughout the genome. F A patients show evidence o f genome instability including chromatid breaks, large deletions and chromosomal rearrangements (Hintz et al, 2006; Levitus et al, 2006). The genome instability in F A patients might be a result of problematic D N A secondary structures that lead to deletions, rearrangements and other chromosomal aberrations i f the lesions are repaired through erroneous means. This hypothesis is currently being tested by several groups i n the D N A repair community who are sequencing G-rich D N A sites in F A group J patients. It w i l l be interesting to see whether or not the roles for D O G - 1 in C. elegans can lead to greater understanding o f F A in humans.  5.3 The DOG-l/FANCJ-related heUcase, JRH-1, functions in D N A repair Examination o f a helicase-like protein closely related to dog-1 showed that jrh-1 also has a function in D N A repair, but that it differs from the function oï dog-1. jrh-1 mutants and dog-1 mutants have different sensitivities to the interstrand cross-linking treatment trimethylpsoralen with U V A ; dog-1 animals are extremely hypersensitive, while jrh-1 animals are moderately sensitive. J R H - 1 and D O G - 1 do not have redundant roles at G/C-tracts. However, j r / z - i dog-1 double mutant animals displayed a synthetic lethal phenotype that appeared to be the result o f excessive homologous recombination repair intermediates. Based on the affect o f the jrh-1 mutation on translocation hétérozygote animals, along with other data, the favored hypothesis is that JRH-1 might be required to channel repair intermediates away from the homologous recombination repair pathway and into other repair pathways, such as sister chromatid-directed  repair. Preventing excessive homologous recombination repair is a function analogous to the role o f Srs2 i n yeast. Similar to yeast srs2 mutants, jrh-1 was shown to have a synthetic lethal phenotype with him-6/BLM  and mus-81, supporting the hypothesis that JRH-1 might have a  function similar to Srs2. In vitro protein work w i l l be required to determine whether or not J R H 1 has the same properties as Srs2 and is able to displace R A D - 5 1 from single stranded D N A . I f JRH-1 were the functional homolog o f Srs2, this would be exciting because no Srs2-like gene has been identified in higher organisms to date. Nonetheless, Srs2 plays such an important role in removing R A D 5 1 from invading D N A filaments that it could be expected that a protein or proteins with a similar function would exist in organisms other than yeast.  In mice, the protein most similar to J R H - 1 , Rtel, is proposed to function at telomeres (Ding et ai, 2004). Based on the data presented here for C. elegans, as well as the study o f Rtel in mouse, it is likely that human homolog, N H L , could have a function in maintaining genome stability at sites such as stalled replication forks and telomeres, where it might function in preventing excessive or inappropriate homologous recombination repair thereby promoting other repair mechanisms. A s D O G - 1 appears to be the C. elegans B R I P l / F A N C J homolog, the genetic interaction between dog-1 and jrh-1 in C. elegans might also be conserved in humans, between BRIPl/FANCJ  and NHL. I f conserved, this interaction is potentially relevant to  Fanconi anemia patients. N H L might perform a function that is essential in the absence of B R I P l / F A N C J . If this is the case, targeted knock-out o f NHL function in cancer cells in Fanconi group J patients might prove to be a useful therapeutic means to destroy tumor cells.  5.4 An extended model for repair at G/C-tracts in the dog-J mutant In Youds et al. (2006) and in Chapter 2 o f this thesis, a model was proposed in which replication forks stalled at G/C-tract secondary structures could be repaired through deletion-free  mechanisms such as translesion synthesis, homologous recombination repair or through H I M - 6 functioning independently o f R A D - 5 1 , or by an unknown deletion-prone mechanism. The new data described in the remainder o f this thesis present an opportunity to add to this model, now including J R H - 1 , M U S - 8 1 and H I M - 6 in a separate branch o f the model.  Mutations i n genes involved in homologous recombination repair in the dog-1 background lead to an increase in the frequency o f animals with deletions, but do not affect viability o f the animals (Youds et al, 2006). O n the other hand, mutations in any ofjrh-1, mus81 or him-6 in the dog-1 background lead to synthetic lethal (Jrh-l dog-1) or synthetic sick phenotypes {mus-81 dog-1 and dog-1; him-6 [mus-81 dog-1 shovm by N . O ' N e i l , p. communication; Youds et al, 2006]). dog-1; him-6 animals had increased G/C-tract deletions, while jrh-1 dog-1 and mus-81 dog-1 double mutants had a similar number o f deletions to the 11% observed in dog-1 single mutant animals (for mus-81 dog-1 13.7% o f animals had deletions, n=255). Thus, it appears that the frequency o f small G/C-tract deletion formation does not correlate with viability o f any o f the double mutants tested i n this thesis.  Based on the decreased viability in the absence of jrh-1, mus-81 or him-6, it is possible that these genes represent the preferred means for stalled replication fork repair in dog-1 mutants. If JRH-1 functions similar to Srs2 in yeast, JRH-1 might be involved in regulating the balance between homologous recombination and other types o f repair, jrh-1 dog-1 double mutants display mitotic catastrophe in the germline and this could be due to the inability to shunt some repair intermediates away from homologous recombination repair into other pathways in the absence o f J R H - 1 . Both M U S - 8 1 and H I M - 6 / B L M are implicated in resolving repair intermediates; therefore, the synthetic sick phenotypes o f mus-81 dog-1 and dog-1; him-6 might be because repair o f stalled forks can still occur but is less efficient in the absence o f either  M U S - 8 1 or H I M - 6 / B L M . Both mus-81 and him-6 are proposed to have multiple roles; therefore, while M U S - 8 1 and H I M - 6 might function in separate pathways for R A D - 5 1 independent repair, these proteins likely also have additional functions in replication and repair. In accordance with this idea, dog-1; him-6 double mutants show increased numbers o f animals with deletions as well as reduced viability possibly because him-6 can function both downstream o f R A D - 5 1 as well as independently. Additional biochemical and genetic experiments w i l l be required to dissect the complex interaction between these repair proteins.  In the absence o f other error-free repair mechanisms involving M U S - 8 1 or H I M - 6 , most G/C-tract stalled forks would be repaired through homologous recombination repair, which does not generate small deletions, but likely leads to reduced viability because o f excessive crossingover that could cause larger genomic aberrations. The deletion-prone pathway is always active at low levels in dog-1 mutants. The activity o f the deletion-prone pathway increases when homologous recombination repair substrates are not formed, leading to an increased frequency o f animals with deletions. A s none o f the genes tested in combination with dog-1 led to a decrease in animals with deletions, it is possible that the deletions are generated when no repair is initiated, and a gap occurs in the lagging strand at the site corresponding to the G/C-tract secondary structure. Figure 5.1 describes this model, highlighting the multiple repair mechanisms that can fonction at a stalled replication fork to maintain genome stability.  G-rich D N A s e c o n d a r y structure forms during replication  /  \  n o D O G - 1 p r e s e n t : a l t e r n a t i v e r e p a i r of s t a l l e d fork r e q u i r e d ^  deletion-prone repair (mechanism unknown)  trans-lesion synthesis  recombinational repair (RAD-51)  / \  HIM-9  HIM-6  recombinationai outcome  non-rocombinationai outcome  f  structure r e s o l v e d by D O G - 1  sister chromatid-directed repair  J \  MUS-81  HIM-6  Figure 5.1 A n extended model for repair o f replication forks stalled at G/C-tract secondary structures. Normally, D O G - 1 would be present to unwind the G/C-tract secondary structure, but in its absence, the secondary structure persists, causing replication fork stalling. The stalled fork can be repaired through various deletion-free means. Sister chromatid-directed repair would be the preferred mechanism, and JRH-1 might function to promote this pathway, while M U S - 8 1 and H I M - 6 resolve the repair intermediates in this pathway. The stalled fork could also be repaired during replication through translesion synthesis by either po\r\ or p o l K , or through homologous recombination repair, involving B R D - 1 , R A D - 5 1 , H I M - 9 or H I M - 6 . Altematively, the deletion-prone repair pathway might act through an unknovm mechanism that results in a deletion at the site o f the secondary structure.  5.5 Future studies on DOG-1 and the DOG-1 heUcase-iike proteins There are several potential avenues for investigation that branch from the work completed i n this thesis. A s dog-] can now be used as a model to study the interstrand cross-link repair pathway, one way to identify new players in the cross-link repair pathway might be to carry out a large scale yeast two-hybrid screen to look for proteins that interact with D O G - 1 or other components o f cross-link repair, such as F C D - 2 / F A N C D 2 . A s previously mentioned, study o f the substrates that JRH-1 and D O G - 1 can unwind would be informative in determining whether or not JRH-1 functions i n a manner similar to Srs2, as well as in understanding the range o f substrates that these helicase-like proteins might unwind. Finally, there are two members o f the dog-] helicase-like family, M03C]].2  and Y50D7A.2, which have not been studied in C.  elegans. M 0 3 C 1 1 . 2 is similar to yeast helicase C h l l , which functions in sister chromatid cohesion (Gerring et al., 1990; Spencer et al., 1990; Skibbens, 2004), while Y 5 0 D 7 A . 2 has similarity to the nucleotide excision repair helicase X P D (www.wormbase.org). When viable mutations or knock-outs o f these genes are available, characterization o f M03C11.2 and Y50D7A.2 is likely to reveal roles for these genes in replication and/or repair, similar to the roles of dog-1 and jrh-1 described here. Potential interactions between M03C11.2 and Y50D7A.2 and both genes with dog-1 and jrh-1 should also be tested. It w i l l be interesting to see how the genes in the dog-1 helicase-like family function together to maintain genome stability during replication and repair.  Chapter 5 References Cheung I, Schertzer M , Rose A , & Lansdorp P M (2002) Disruption of dog-1 in Caenorhabditis elegans triggers deletions upstream o f guanine-rich D N A . Nat Genet 3\: 405-409 Collis SJ, Barber L J , Ward J D , Martin JS, & Boulton S J (2006) C. elegans F A N C D 2 responds to replication stress and functions in interstrand cross-link repair. DNA Repair (Amst) 5: 1398-1406 Dequen F, St-Laurent J F , Gagnon S N , Carreau M , Desnoyers S (2005) The Caenorhabditis elegans FancD2 ortholog is required for survival following D N A damage. Comp Biochem Physiol B Biochem Mol Biol 141(4): 453-60 Ding H , Schertzer M , W u X , Gertsenstein M , Selig S, K a m m o r i M , Pourvali R , Poon S, Vulto I, Chavez E , Tam PP, Nagy A , & Lansdorp P M (2004) Regulation o f murine telomere length by Rtel: an essential gene encoding a helicase-like protein. Cell 117: 873-886 Gerring S L , Spencer F , & Hieter P (1990) The C H L 1 ( C T F 1) gene product of Saccharomyces cerevisiae is important for chromosome transmission and normal cell cycle progression i n G 2 / M . EMBO J 9: 4347-4358 Hinz J M , N h a m P B , Salazar E P , & Thompson L H (2006) The Fanconi anemia pathway limits the severity o f mutagenesis. DNA Repair (Amst) 5: 875-884 Huppert J L & Balasubramanian S (2005) Prevalence o f quadruplexes in the human genome. Nucleic Acids Res 33: 2908-2916 Joenje H & Patel K J (2001) The emerging genetic and molecular basis o f Fanconi anaemia. Nat Rev Genet 2: 446-457 Levitus M , Joenje H , & de Winter JP (2006) The Fanconi anemia pathway o f genomic maintenance. Cell Oncol 28: 3-29 Maizels N (2006) Dynamic roles for G4 D N A in the biology o f eukaryotic cells. Nat Struct Mol Biol 13: 1055-1059 Meetei A R , Medhurst A L , L i n g C , X u e Y , Singh T R , Bier P, Steltenpool J, Stone S, Dokal I, Mathew C G , Hoatlin M , Joenje H , de Winter JP, & Wang W (2005) A human ortholog of archaeal D N A repair protein H e f is defective in Fanconi anemia complementation group M . Nat Genet 37: 958-963 Patel K & Joenje H (2007) Fanconi anaemia and D N A replication repair. DNA Repair 6(7): 88590. Reid S, Schindler D , Hanenberg H , Barker K , Hanks S, K a l b R, Neveling K , K e l l y P, Seal S, Freund M , W u r m M , Batish S D , Each F P , Yetgin S, Neitzel H , Ariffin H , Tischkowitz M , Mathew C G , Auerbach A D , Rahman N (2007) Biallelic mutations in P A L B 2 cause Fanconi anemia subtype F A - N and predispose to childhood cancer. Nat Genet 39(2): 162-4  Skibbens R V (2004) C h l l p , a D N A helicase-like protein in budding yeast, functions i n sisterchromatid cohesion. Genetics 166: 33-42 Spencer F , Gerring S L , Connelly C , & Hieter P (1990) Mitotic chromosome transmission fidelity mutants in Saccharomyces cerevisiae. Genetics 124: 237-249 Taniguchi T & D'Andrea A D (2006) Molecular pathogenesis o f Fanconi anemia: recent progress. Blood 107: 4223-4233 Youds J L , O'Neil N J , & Rose A M (2006) Homologous recombination is required for genome stability i n the absence o f D O G - 1 in Caenorhabditis elegans. Genetics 173: 697-708  Appendix 1; Primer sequences  Primer sequences for detection of G/C-tracts and deletion alleles of genes Gene Amplified dog-l(gklO)  Name o f Primer dogL2 dogM2 dogR2  GGA GTA TAG AAC GTG TTT CG GCT CTT CTT TCA ATG TGA CGG CGT CCA CAT CAA CAG AAC C  55  wild type product 604 & 3003bp, deletion product 972bp  him-6(ok412)  him-6EL him-6 ER him-6 IR2  TTT TCG TGT TGC GGT TCG ACG CCA GTC GTA GTG TTT CC GTGTTCTTGGATGGTGGC  57  wild type product 700 & 2398bp, deletion product 500bp  wm-l(gk99)  WRN EL WRNIR WRNER  GCT GCA GAA TTG AAG AGA AAG CAT CAC TTA TCA TCT GTG CAT G CGA GTT CTC AGA GTG TAT CC  54  wild type products 455& 928bp, deletion product 732bp  brd-1 (gk297)  brd-1 EL brd-lIR brd-1 ER  GCA AAG ACT TGT TAA GTA AGG CCA GAA TTG GAT ATA TTC CAC GCT ATC GAG TGT GTT AAA TG  55  wild type products 743 & 1636 bp, deletion product 1254bp  cep-1 (gkI38)  cep-1 ER cep-1 IR cep-1 IR2  CGA CGG AGA TTG ACA GTT TTC G GGA ATT ATT TGC CGA TTT TCT C GAA ATT ATG TCT GAA TTT ACC C  55  wild type product 1532 & 2784bp, deletion product 1123bp  him-9 (el487)  NON50 NON53 NON87  GGACAGTACTCTCGGAGATT CGAACTGTATCAAATTGGTCTG CACATTGTCCGCTTGTGTC  56  NON50/53 ~lkb product only in wild type; NON50/87 ~800bp product only in him-9  rad-51 (lg8701)  rad-51 EL rad-51 IL rad-51 ER  CTC GAT CTA CCA TAC TAA AGC GGT AAT AAT TAC AGC GAC ACC GTC ATA ATT TGA TCT CCC GAC  58  wild type products 764 & 1995bp, deletion product 1055bp  xpa-1 (ok698)  xpa-1 EL xpa-1 IR xpa-1 ER  CAA TGG CAA TTT GCT AGT ATT TC CTG GCA CTT CAG ATA CAA CTT C CAT GCG CGT AAT ATA TGT GTA G  58  wild type products 476 & 1494bp, deletion product 581 bp  lig-4 (ok716)  lig-4 EL lig-4 IR lig-4 ER  CGT ATT TGT GGT ATT ACC CGG CTG TTC TCT TCA CAA CGA TTC C GAT CAT CTT TAT TGC AAC GTT CG  58  wild type products 766 &2120bp, deletion product 578bp  cku-80 (ok861)  cku-80 EL cku-80 IR cku-80 ER  GCC GTT AGT GAA AGT AAT GCA G CAC CGG GAG ATG GAT TCC AG CCT CAT CTG GTT GTG TCA TAT TC  58  wild type products 636 &2312bp, deletion product 744bp  tml298-EL tmI298-IR tml298-ER  CTA GCC AAT CAG ATG GAG TG GGA GCC TCT GGA ATG ATG CAG AAG CGA GCA AGC GCG  58  wild type product 502 & 982bp, deletion product 744bp  fcd-2(tml298)  Sequence (5'-3')  Anneal  atrC)  Products  Gene Amplified mus-81 (tml937)  Name o f Primer C43E11.2F C43E11.2R C43E11.21  AGG TAT TTG GCA GAC TTA CC GGC TGA ATG GAA CAC CCG AA GAG CTT CCG ATC TTC TTG C  Jrh-1 (tml866)  tml866-EL tml866-IR tml866-ER  CCT GTG TGG TGT GTG ATT AA CCT AGA ATT GTG GTT TTG ACC CAA TTC TGT AGA CGT ACA ATC  58  wild type product 784 & 2527bp, deletion product 1181bp  vab-1 G/C-tract external  vabNl vabN2  CGATTCCAACAATTGGTAAATACC AATATTTGCTAAACCTATTGTTGCC  58  wild type product 890bp  vab-1 G/C-tract internal  152 153  CGA CGA AAA ATG CAG AAT TTG GC AGG TGT GTG TGC ATA CCT CCG  62  wild type product 499bp  R144 cosmid G/C-tract  398 R144-IN2  CAT ATG GAT TGG CAT GTG AAG CA CTG CCT ACA GTA GTC TTT GCG  62  wild type product 1028bp  ZK377 cosmid A/T(27) tract  ZK3-L ZK3-R  CGT CTG GAG TTT TTT GTA TTC GCA TCG TGA TGA GTG GAT AC  56  wild type product 856bp  Y77E11A cosmid A/T(22i tract  Y77-L Y77-R  CGA ATT TTT CGC CAT TTT CAG GCT CCG TGT GCA TTG TAC C  56  wild type product 918bp  F12F6 cosmid  F12-L F12-R  CAC ATA CAA CCA TCG TCT CC CTT GAA ACG AAT TAC ATA TTG AG  56  wild type product 860bp  F46-EX1 F46-EX2  CCT CGA CAG TGA AAA TAA TAA AC GTA GCT GAT TGG TTC AGG TTC  62  wild type product 915bp  (CAG)(8) tract  F46H6 cosmid G/Ctract  Sequence (5'-3')  Anneal at C O 58  Products wild type product 406 & 2070bp, deletion product 879bp  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0066819/manifest

Comment

Related Items