Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of DNA mismatch repair genes in genome stability and carcinogenesis Baross-Francis, Agnes 2000

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

Item Metadata

Download

Media
831-ubc_2000-565017.pdf [ 5.58MB ]
Metadata
JSON: 831-1.0089720.json
JSON-LD: 831-1.0089720-ld.json
RDF/XML (Pretty): 831-1.0089720-rdf.xml
RDF/JSON: 831-1.0089720-rdf.json
Turtle: 831-1.0089720-turtle.txt
N-Triples: 831-1.0089720-rdf-ntriples.txt
Original Record: 831-1.0089720-source.json
Full Text
831-1.0089720-fulltext.txt
Citation
831-1.0089720.ris

Full Text

T H E ROLE OF D N A M I S M A T C H REPAIR GENES IN G E N O M E STABILITY A N D CARCINOGENESIS  by AGNES BAROSS-FRANCIS M.Sc. Bioengineering, Technical University of Budapest, Hungary, 1993 B.Sc. Chemical Engineering, Technical University of Budapest, Hungary, 1991 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in  THE FACULTY OF GRADUATE STUDIES  (Department of Medicine and the Genetics Graduate Program)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA June 2000 © Agnes Baross-Francis, 2000  In  presenting  degree freely  at  this  the  thesis  in  partial  fulfilment  University  of  British  Columbia, I agree  available for  copying  of  department publication  this or of  reference and study.  thesis by  this  for  his thesis  scholarly  or for  her  Department  of  The University of British Columbia Vancouver, Canada  DE-6  (2788)  I further  purposes  gain  the  requirements that the  agree  may  representatives.  financial  permission.  of  It  shall not  be is  an  advanced  Library shall make  that permission  for  granted  head  by  understood be  for  the that  allowed without  it  extensive of  my  copying  or  my  written  Abstract  DNA mismatch repair (MMR) deficiency is associated with an increased mutational burden and predisposition to certain malignancies. Relatively little is known, however, about mutant frequencies within MMR-deficient primary tumors. Thymic lymphomas from Mshl ' mice were thus analyzed using a lad-based transgenic 1  shuttle-phage mutation detection system. All tumors exhibited greatly elevated lad gene mutant frequencies, ranging from 3.2- to 17.4-fold above the ~15-fold elevations present within normal Mshl''' thymi. In addition, individual lad genes harboring multiple changes were found in the tumors. To investigate whether hypermutation was a feature of all tumors arising in mismatch repair-deficient mice, lad transgene mutant frequencies were obtained from tumors of mice deficient for Pmsl and/or Mshl. While lad gene hypermutation was again clearly evident in Msh2 ~ Pms2~'~ and Mshl' ' Pmsl' ' thymic lymphomas, three non+/  1  1  thymic Mshl deficient tumors failed to show elevated frequencies of mutation in lad when compared to a normal tissue within the respective mice. The elevated mutant frequencies in the lymphoid tumors, and the finding of multiple clustered mutations in lad genes from these tumors, suggested that they were possibly generated by a lymphoma-specific hypermutational mechanism. Similar to Mshl deficient mice, mice rendered deficient in Mlhl or Pmsl via gene targeting are also prone to tumorigenesis, particularly lymphomas. According to the recent model of mammalian MMR, these two proteins function as a heterodimer. However, while Mlhl' ' mice develop small intestinal adenomas and adenocarcinomas, 1  Pmsl''' animals remain free of such tumors. To establish whether this discrepancy might be associated with a quantitative and/or qualitative difference in genomic instability between Mlhl' ' and Pmsl' ' mice, we determined small intestinal epithelial cell DNA 1  1  mutant frequencies and spectra using the lad reporter system. We found that C:G->T:A ii  transitions were significantly elevated in Mlhr  versus Pms2~'~ mice, leading to a 1.5-fold  lacl mutant frequency increase in these animals. We hypothesize that this finding may explain, in part, why MlhT ' mice, but not Pmsl' ' mice, develop tumors at this site. 1  1  Furthermore, the difference in the lacl mutational spectrum of MlhT'' and Pmsl''' mice suggests that MLH1 may be involved in a PMS2-independent repair pathway particularly towards D N A lesions that result in C:G->T:A transition mutations.  iii  Table of Contents  Abstract  ii  Table of Contents  iv  List of Tables  vii  List of Figures  viii  List of Abbreviations  ix  Acknowledgements  x  Dedication  xi  Chapter 1 Introduction  1  1.1 D N A damage and repair pathways  1  1.2 The D N A mismatch repair system  3  1.2.1 Mismatch repair in Escherichia coli 1.2.2 Eukaryotic mismatch repair  3 6  1.3 Mismatch repair deficiency and carcinogenesis  11  1.4 MMR-deficient mouse models  12  1.5 The transgenic lacl shuttle phage in vivo mutation detection system  13  1.6 Thesis goals  20  1.7 Thesis summary  20  Chapter 2 Materials and Methods  23  2.1 Transgenic mice  23  2.1.1 Breeding strategies  23  2.1.2 Preparation of tail D N A for genotype testing of mice  23  2.1.3 Determination of the Msh2 genotype  24  2.1.4 Determination of the Pmsl genotype  24  2.1.5 Determination of the Mlhl genotype  27  2.1.6 Determination of the lacl genotype  27  iv  2.1.7 Agarose gel electrophoresis  28  2.1.8 Obtaining mouse tissues  28  2.1.9 Characterization of tumors  28  2.1.10 Isolation of small intestinal epithelial cells  28  2.2 Measurement of lacl mutational frequency  29  2.2.1 Isolation of genomic D N A  29  2.2.2 Preparing the SCS-8 plating culture  30  2.2.3 Performing the packaging reaction.....  30  2.2.4 Plating the packaged D N A samples  31  2.2.5 Screening and analyzing the assay trays  31  2.2.6 Verifying putative mutant plaques  31  2.2.7 Statistical analysis of mutant frequencies  32  2.3 Determination of lacl mutational spectrum  32  2.3.1 Amplification of mutant lacl genes  32  2.3.2 Purification of lacl template for sequencing  33  2.3.3 Sequencing of lacl mutants  33  2.3.4 Sequence analysis for mutation detection  33  2.3.5 Statistical analysis for the comparison of mutational spectra  34  Chapter 3 Thymic tumors of MSH2-deficient hosts exhibit dramatic increases in genomic instability  39  3.1 Introduction  39  3.2 Results  40  3.2.1 Mutant frequency mMshl' ' thymic lymphomas  40  3.2.2 Mutant frequency of lAshl' ' normal tissues  45  3.2.3 Mutation spectrum of M&hl' ' thymic tumors  50  1  1  1  3.2.4 Multiple mutations in single lacl genes 3.3 Discussion  51 54  v  Chapter 4 Tumors arising in D N A mismatch repair-deficient mice show a wide variation in mutant frequency  57  4.1 Introduction  57  4.2 Results  59  4.2.1 Mutant frequency of thymic lymphomas from PMS2 deficient and Mshl''' Pmsl''' double knock-out mice  59  4.2.2 Mutant frequency of MMR deficient non-thymic tumors  64  4.2.3 Mutational clusters in single lad genes are specific to thymic lymphomas  65  4.3 Discussion  65  Chapter 5 Elevated mutant frequencies and increased C:G—>T:A transitions in Mlhl''' versus Pmsl''' murine small intestinal epithelial cells  70  5.1 Introduction  70  5.2 Results  72  5.2.1 Mutant frequencies of Mlhl''' and Pmsl''' small intestinal epithelial cells  72  5.2.2 Mutant frequencies of small intestinal epithelial cells from wild type mice....77 5.2.3 Mutational spectra from Mlhl''' and Pmsl''' small intestinal epithelial cells 5.3 Discussion  77 81  Chapter 6 Discussion  88  6.1 Summary of thesis  88  6.2 D N A mismatch repair-deficiency leads to tumorigenesis via subsequent mutations in key oncogenes and tumor suppressors  89  6.3 Loss of other roles of MMR proteins may also contribute to carcinogenesis  90  6.4 Potential future directions  92  References  95  vi  List of Tables  Table 3.1 Spontaneous mutant frequencies for Mshl''' thymic lymphomas and normal tissues... 41  Table 3.2 lad mutation spectrum ofMshT'' thymic lymphomas, and normal thymi.. .46  Table 3.3 Multiple mutations occurring in single lad genes from Mshl'''  thymic  lymphomas and an Mshl''' normal thymus.. .52  Table 4.1 Spontaneous lad mutant frequencies for MSH2 and PMS2 deficient tumors and normal tissues... 60  Table 4.2 Multiple mutations occurring within single lad genes rescued from Mshl''' Pmsl''' and Mshl '~PmsT'' thymic lymphomas...66 +  Table 5.1 Spontaneous lad mutant frequencies for small intestinal epithelial cells of Mlhl''', Pmsl' ' and wild type mice...73 1  Table 5.2 Spontaneous lad mutation spectra from small intestinal epithelial cells of MM'''  and Pmsl''' mice...78  vii  List of Figures  Figure 1.1 Mismatch repair in E. coli.. A  Figure 1.2 Mismatch repair in eukaryotes...7  Figure 1.3 Function of the lacl gene... 15  Figure 1.4 The transgenic lacl mutation detection system... 17  Figure 2.1 Agarose gels resolving PCR fragments from genotyping reactions...25  Figure 2.2 Sequence of the lacl gene.. .35  Figure 2.3 Sequence alignment of a mutant with the wild type lacl gene.. .37  Figure 3.1 Spontaneous mutant frequencies for MshT'' thymic lymphomas and normal tissues... 43  Figure 3.2 lacl mutation spectrum oiMshl''' thymic lymphomas, and normal thymi...48  Figure 4.1 lacl mutant frequencies of D N A mismatch repair-deficient tumors and normal brains.. .62  Figure 5.1 lacl mutant frequencies for small intestinal epithelial cells of Mlhl'', Pmsl''' and wild type mice...75  Figure 5.2 Predicted distributions of the mutational spectra of 150 MlhT'' and 100 Pmsl''' lacl mutants obtained from small intestinal epithelial cell DNA...82  viii  List of Abbreviations  BER  base excision repair  HNPCC  hereditary non-polyposis colorectal cancer  IDL  insertion/deletion loop  MF  mutant frequency  MLHQ-3)  MutL homolog (1-3)  MMR  mismatch repair  MS  mutation spectrum  MSI  microsatellite instability  MSH(l-6)  MutS homolog (1-6)  NER  nucleotide excision repair  pfu  plaque forming unit  PMS(l-2)  post-meiotic segregation mutant (1-2) (also MutL homolog)  SD  standard deviation  X-gal  5-bromo-4-chloro-3-indolyl p-galactopyranoside  ix  Acknowledgements  I would like to thank everyone who helped me with this thesis directly or indirectly. I am very grateful to my supervisor, Dr. Frank R. Jirik. I am indebted to my supervisory committee, Dr. Ann M. Rose, Dr. Muriel J. Harris and Dr. Robert Kay for their helpful comments. I would like to thank all the past and present members of the Jirik lab, especially Janice E. Penney, Leigh Anne Waddleton and Cristin Fletch for maintaining the various mouse lines; Scott Pownall and Teresa McKernan for generating the BC-1 transgenic line; Naila Makhani, M . Kate Milhausen, and Lorraine Spence for technical assistance; and Dr. Susan E. Andrew for helpful discussions. I am grateful to Dr. Tak W. Mak for providing the MshT'' mouse line; and Dr. R. Michael Liskay for providing the Pms2''' and Mlhl''' mice. I also thank Keith Fichter and David Spear at the C M M T D N A Sequencing Core Facility, and Dr. Gareth Jevon for his help in characterizing the mismatch repair-deficient tumors. Finally, I am grateful to the Natural Sciences and Engineering Research Council of Canada for my graduate scholarship.  Dedication  To my Mother, Grandparents, and Great Uncle, without whom I could not have achieved this. To my Sister, Erzsi, to whom the world belongs. Being so far away, I missed them greatly. I hope I can make them proud.  Mamanak, Nagyinak, Nagypapanak es Keresztpapanak, akik nelkiil mindezt nem erhettem volna el. Hugomnak, Erzsinek, aki elott all az egesz jovo. Mindannyian nagyon hianyoztak. Remelem, biiszkek lesznek ram.  xi  Chapter 1 Introduction  1.1 D N A damage and repair pathways It appears that cellular DNA, rather than being inert, continuously undergoes damage, repair and resynthesis (Loeb and Loeb, 2000). Mutations stem from D N A damage caused by both environmental and cellular sources, such as spontaneous hydrolytic degradation, exogenous and endogenous chemicals, physical stress, ionizing radiation and D N A polymerase misincorporation. Mutations are necessary for adaptation and evolution; however, high mutation rates increase cancer susceptibility and may endanger survival of the species. For their correction, different D N A repair pathways exist with varying, and to some extent, overlapping specificities. Three main strategies of excision repair, described below, are employed by cells to prevent point mutations; besides, several mechanisms have evolved to repair double strand D N A breaks. The base excision repair (BER) system involves the replacement of a single damaged nucleotide in the D N A with a normal residue. Most D N A lesions repaired by BER are caused by spontaneous hydrolytic deamination or depurination of D N A , reactive oxygen species, U V irradiation, alkylating agents and D N A polymerase misincorporations. The pathway involves excision of an altered D N A base in free form by a D N A glycosylase, followed by introduction of a strand break 5' of the base-free sugar by an AP endonuclease. D N A polymerase B removes the sugar and fills in the single-nucleotide gap, and a ligase seals the resulting nick. In addition, there is a secondary BER pathway, which differs in the later stages of the repair process, where the repair patch is several nucleotides long (Lindahl et al, 1997; Schmutte and Fishel, 1999). The nucleotide excision repair (NER) process acts on a wide variety of adducts in D N A and is most effective on bulky or helix-distorting lesions, such as UV-induced 1  pyrimidine dimers. The helix-altering lesion is recognized by a complex of NER enzymes, then a 20-35 nucleotide long region around the damage is opened up. This is followed by dual incision on the 3' and 5' sides of the damage by the XPG and ERCC1XPF nucleases, respectively. Repair synthesis is carried out by a PCNA-dependent polymerase 8 or e holoenzyme and the patch is ligated to finish the process. A specific type of NER is the transcription-coupled repair (TCR) pathway, which involves various NER enzymes and acts on genes transcribed by R N A polymerase II (Lindahl et al., 1997; Schmutte and Fishel, 1999). The mismatch repair (MMR) system, which is described below in greater detail, primarily enhances D N A replication fidelity. MMR repairs D N A mismatches arising by incorporation of inappropriate nucleotides by D N A polymerases during D N A replication, or even during gap resynthesis in the course of D N A repair. Misincorporations can also be triggered by damaged bases in the template strand of D N A , or by changes in cellular dNTP concentrations. The MMR system also appears able to recognize a variety of damaged nucleotides in D N A (Lindahl et al., 1997; Schmutte and Fishel, 1999). Mutations in D N A repair proteins that render them defective, or mutations in D N A polymerases that make them error-prone lead to a mutator phenotype, which may be the first step in carcinogenesis. There is increasing evidence for the presence of numerous mutations in cancer cells, including point mutations, microsatellite instability, gene amplifications, chromosomal aberrations and aneuploidy. It has been estimated that the spontaneous mutation rate observed in normal cells is inadequate to account for the large number of mutations apparently required for tumorigenesis; it has thus been postulated that cancer cells manifest a 'mutator' phenotype (Loeb, 1991; Loeb and Christians, 1996). Further discoveries of the biochemical mechanisms that lead to increased mutagenesis may open up new possibilities to cancer diagnostics and therapy. 2  1.2 The DNA mismatch repair system The long patch D N A mismatch repair system, which is primarily responsible for the correction of mispaired bases incurred during D N A replication, was first described in Escherichia  coli. The key components of this system (described below) are well  conserved in prokaryotes and eukaryotes, including yeast and mammals, as well as in plants (Culligan et al, 2000; Kolodner, 1996).  1.2.1 Mismatch repair in Escherichia coli The E. coli MutHLS mismatch repair pathway recognizes and repairs all single base mispairs with variable efficiencies,  with the possible exception of C:C.  Furthermore, this system also repairs small insertion/deletion loops (IDLs) (Kolodner, 1995; Marra and Schar, 1999; Modrich and Lahue, 1996). During D N A replication, the bacterial MMR system is targeted to the nascent unmethylated daughter strands. E. coli D N A is normally methylated at GATC sites by the Dam methylase; however, after replication the daughter strand remains transiently unmethylated. The MutS protein homodimer binds to D N A at the site of a mismatched base. The MutL homodimer then interacts with MutS, and couples the mismatch recognition to MutH, which cuts the D N A strand at the nearest hemimethylated GATC site that is either downstream (3') or upstream (5') of the mispair (Kolodner, 1996; Kolodner, 1995; Prolla, 1998). The repair reaction then proceeds by exonucleolytic degradation of the nicked daughter D N A strand beginning at the incised GATC site, then continuing on past the position of the mismatch (Figure 1.1). This step requires UvrD (MutU/helicase II), and any one of the single stranded D N A exonucleases: Exo I, Exo VII, Exo X, or RecJ protein (Buermeyer et al., 1999). Following the excision step, resynthesis of the new D N A strand is mediated by D N A polymerase III holoenzyme, single-strand DNA-binding protein (SSB) and D N A ligase (Kolodner, 1996; Kolodner, 1995; Marra and Schar, 1999). It is important to 3  Figure 1.1 Mismatch repair in E. coli T h e a c t i o n o f M M R is s h o w n at the r e p l i c a t i o n f o r k of E. coli. R e p a i r is i n i t i a t e d b y b i n d i n g of M u t S to the m i s m a t c h , f o l l o w e d b y M u t L b i n d i n g a n d a c t i v a t i o n o f M u t H . M u t H n i c k s the u n m e t h y l a t e d d a u g h t e r D N A s t r a n d at the nearest G A T C site, t h e n e x o n u c l e a s e - m e d i a t e d e x c i s i o n e x t e n d i n g f r o m the n i c k t h r o u g h the m i s m a t c h takes place. T h i s is f o l l o w e d b y resynthesis to fill the single s t r a n d e d g a p .  4  5  note that the requirement for MutH and the use of Dam methylation for strand discrimination is limited to certain bacteria. Other bacteria, such as Streptococcus pneumoniae have a MMR system closely related to that of E. coli, but lacking a MutH  homolog or DNA methylation as a mechanism to distinguish between newly replicated and parental DNA (Kolodner, 1996).  1.2.2 Eukaryotic mismatch repair  The current model of eukaryotic MMR closely resembles that of E. coli, with two major differences. In yeast and mammals no MutH homologue has yet been identified, and GATC methylation is not used for distinguishing the parental and daughter DNA strands. It has been proposed that strand discrimination might be mediated by the presence of strand discontinuities in the newly synthesized DNA, or perhaps by CpG methylation in mammals; however, these hypotheses still require experimental validation (Bellacosa et al, 1999; Jiricny, 1998). Another important difference is that the MutS and MutL homologs function as various heterodimers rather than homodimers. Eukaryotic mismatch recognition is carried out by orthologs of MutS. In Saccharomyces cerevisiae and mammals six MutS paralogs have been identified, MSH1-  MSH6, three of which, MSH2, MSH3, and MSH6 function in a MutHLS-like mismatch repair pathway. MSH2 and MSH6 (GTBP/pl60) form a heterodimer called MutSa, which recognizes and binds with highest affinity to single base mismatches and IDLs of one nucleotide. Another factor, MutSP, composed of MSH2 and MSH3, mainly binds substrates with extrahelical nucleotides, where its affinity seems to increase with the size of the IDL (Figure 1.2). However, both complexes seem to be able to correct basebase mismatches and small IDLs to a certain extent (Johnson et al., 1996; Marsischky et al., 1996; Umar et al., 1998). In most cell lines MutSa is present at substantially higher concentrations, and thus appears to play the predominant role in repair  6  Figure 1.2 Mismatch repair in eukaryotes Recognition of mismatches by MutSoc or MutS(3 and the following binding of MutLoc or MutLB is shown for Saccharomyces cerevisiae. On the figure the names of the yeast proteins are shown. The closest mammalian homolog of yeast PMSI is mammalian PMS2, and the more closely related mammalian homologs of yeast MLH3 are mammalian PMSI and MLH3.  7  8  (Buermeyer et al, 1999; Jiricny, 1998; Kolodner, 1996; Kolodner and Marsischky, 1999; Marra and Schar, 1999). MutS and its various eukaryotic homologs possess an ATPbinding and an ATP-hydrolysis activity, and it has been demonstrated that the dimers are able to bind mismatched D N A in the presence of ADP, but unable to do so when complexed with ATP (Fishel, 1998; Jiricny, 1998). The MutS orthologs MSH1, MSH4 and MSH5 do not contribute to mismatch or IDL correction in nuclear D N A (Marra and Schar, 1999). MutLa, which is a dimer composed of eukaryotic homologs of MutL, namely MLH1 and PMSI in yeast, or MLH1 and PMS2 in mammals (PSM2 is the closest mammalian ortholog to yeast PMSI) functions in MMR by binding to the MutS related complexes (Jiricny, 1998; Kolodner and Marsischky, 1999). In yeast, a second heterodimer composed of MLH1 and MLH3, MutLB, also participates in MMR by repairing a subset of frameshift errors (Figure 1.2) (Flores-Rozas and Kolodner, 1998). The presence of a human MutLB complex containing h M L H l and hPMSl has recently been demonstrated (Raschle et al, 1999), but thus far no biochemical function has been ascribed to this heterodimer. A n additional mammalian MutL ortholog, MLH3,  has  recently been cloned and the protein has also been shown to interact with h M L H l (Lipkin et al, 2000). It is possible that the mammalian MLH1/PMS1 or M L H 1 / M L H 3 complexes also function in MMR. MutL and its eukaryotic homologs bind to MutS homolog complexes in the presence of ATP, and dissociate following hydrolysis of ATP to ADP (Ban et al, 1999; Kolodner and Marsischky, 1999). As mentioned above, there is no known eukaryotic ortholog to MutH of E. coli. However, a novel human methyl-CpG-binding endonuclease, MED1 (MBD4) has been identified, which might be a functional analog (but not sequence homolog) of MutH. MED1 interacts with MLH1, binds to methyl-CpG-containing D N A , and possesses endonuclease activity. Cell lines transfected with a dominant negative mutant of MED1 demonstrate microsatellite instability. (Microsatellites are very common and highly 9  polymorphic arrays of tandem repeats of 1-4 nucleotides within eukaryotic genomes (Weber and May, 1989). They are prone to insertion/deletion mutations, especially in tumors, which can be detected by PCR amplification of the repeats and high resolution gel electrophoresis (Shibata et al., 1996).) Thus MED1 may be involved in mammalian MMR (Bellacosa et al., 1999), but its role in directing repair to the nascent D N A strands remains to be shown. On the basis of the current knowledge about the E. coli MutHLS mismatch repair system, there are a host of other proteins required for eukaryotic MMR, such as exonucleases, helicases, and other enzymes and cofactors required for D N A synthesis (Kolodner, 1996). Although the whole eukaryotic pathway has not yet been fully resolved, some progress has been made in the identification of some enzymes that might play roles in MMR. Exonuclease I (encoded by the EXOZ gene) was first described in Saccharomyces pombe, then in S. cerevisiae, and recently in mammals (Kolodner and Marsischky, 1999). Exonuclease I is a 5'—>3' exonuclease, and it interacts with MSH2. A second exonuclease implicated in MMR, the endo/exonuclease FEN1 (RAD27) plays an important role in processing the 5' ends of Okazaki fragments (Kolodner and Marsischky, 1999). Mutations in RAD27 cause a mutator phenotype resembling MMR deficiency. Furthermore, FEN1 (RAD27) interacts with PCNA, a protein that also interacts with other MMR components (Kolodner and Marsischky, 1999). A number of replication factors have also been implicated in eukaryotic MMR, such as D N A polymerase 5, the single-stranded DNA-binding protein RPA, the proliferating cell nuclear antigen PCNA, as well as RFC, a factor required to load P C N A onto D N A . Further genetic, biochemical and protein interaction studies are needed to confirm the roles of these enzymes in the M M R pathway, as well as to identify additional components (Kolodner and Marsischky, 1999).  10  1.3 Mismatch repair deficiency and carcinogenesis Consistent with the hypothesis that an early event in tumorigenesis may be a gene mutation that confers a mutator phenotype (Loeb, 1991; Loeb and Christians, 1996), mutations in D N A mismatch repair genes have been linked to microsatellite instability and predisposition to cancer. Hereditary non-polyposis colorectal cancer (HNPCC) is an autosomal dominant disorder with high penetrance, that accounts for at least 5% of all colorectal cancers (Buermeyer et al., 1999; Toft and Arends, 1998). Affected individuals tend to develop tumors at 42 years as a mean age at diagnosis, some develop colorectal cancer alone (Lynch I type syndrome), o r may develop additional tumors in several other tissues leading to endometrial or ovarian cancers and malignancies of the stomach, pancreas, small intestine, skin, breast, and urinary tract (Lynch II type syndrome) (Prolla, 1998; Toft and Arends, 1998; Umar and Kunkel, 1996). To date, heterozygous germline mutations in four mammalian D N A mismatch repair genes have been associated with HNPCC: these are hMSH2 and hMLHl, extent hPMSl  and hPMSl.  and to a lesser  Mutations in hMSH6 have also been demonstrated in cancer  prone families that did not strictly meet all the classification criteria for H N P C C (Kolodner, 1995; Prolla, 1998; Toft and Arends, 1998; Wang et al., 1999b). Tumor cells from H N P C C individuals show losses of both alleles of a M M R gene, initially with an inherited mutation in one allele and later, a loss of function mutation of the second allele resulting in loss of heterozygosity (LOH). Thus, M M R genes behave as tumor suppressor genes. Homozygous germline inactivation of a mismatch repair gene has thus far been reported only in two families, where hMLHl -deficient children developed leukemias and/or lymphomas and signs of neurofibromatosis type 1 at an early age (Ricciardone et al, 1999; Wang et al., 1999a). Most, but not all MMR-deficient tumors demonstrate microsatellite instability (MSI), as well as an increase in point mutations due to base S u b s t i t u t i o n s . Beside the familial cancer predisposition syndromes, inactivation of M M R genes and microsatellite instability have been found in a wide 11  variety of sporadic cancers as well (Prolla, 1998; Toft and Arends, 1998). In addition, M M R genes can also be inactivated by transcriptional silencing, as it has been demonstrated for hMLHl  and hMSH2 in various human tumors (Curia et al, 1999;  Leung et al, 1999; Simpkins et al, 1999; Wheeler et al, 1999). Mismatch repair deficiency presumably leads to tumorigenesis due to elevated mutation rates, which in turn promote the accumulation of mutations in critical growth control and survival genes; however, the reasons why HNPCC individuals fail to develop tumors in all tissues with equal probability remains unclear.  1.4 MMR-deficient mouse models Several genes implicated in MMR have been mutated in mice via gene targeting providing an opportunity to study the biological roles of various M M R genes in mutagenesis and carcinogenesis. Interestingly, all the mice rendered deficient in MMR genes thus far are viable and show no obvious developmental defects (Heyer et al, 1999). Mshl  deficient mice develop normally and are fertile. However, homozygous  mutants have a reduced lifespan, compared to heterozygous and wild-type mice, due to the frequent development of T-cell lymphomas between 4 and 6 months of age. These mice also develop gastrointestinal adenomas and carcinomas, skin neoplasms, and other tumors if they survive more than 6 months. Tumors and normal cells of  Mshl ' 1  mice demonstrate microsatellite instability (de Wind et al, 1995; Reitmair et al, 1995). Msh6 knockout mice are fertile and susceptible to similar types of tumors as  Mshl'''  hosts (Edelmann et al, 1997). However, these mice survive longer, as they tend to develop these neoplasms by about one year of age. For example, 50% of Mshl''' animals die by the age of 6 months, whereas 50% of Msh6~'~ mice die between the ages of 10-11 months. Consistent with the role of Msh6 in MMR, Msh6''' tumors show a low rate of MSI (Edelmann et al, 1997; Heyer et al, 1999). Interestingly, while targeted inactivation 12  of Msh3 i n m i c e m a y l e a d to t u m o r s i n o l d m i c e , the p h e n o t y p e of m i c e deficient for b o t h Msh6 a n d Msh.3 resembles the p h e n o t y p e of Mshl' ' 1  hosts ( E d e l m a n n et al., 2000;  H e y e r et al., 1999). T h i s is i n k e e p i n g w i t h the roles o f M u t S h o m o l o g s i n the c u r r e n t e u k a r y o t i c m o d e l of M M R , w h e r e there is p a r t i a l r e d u n d a n c y b e t w e e n the M S H 6 a n d M S H 3 p r o t e i n s i n the r e p a i r of frameshift e r r o r s . D e f i c i e n c y for M S H 2 results i n a stronger p h e n o t y p e , as it leads to the lack of b o t h M u t S a a n d M u t S B . Mlhl  deficient m i c e h a v e a r e d u c e d s u r v i v a l (50% d e a d b y the age of 6 m o n t h s )  d u e to t h e i r s u s c e p t i b i l i t y to T c e l l l y m p h o m a s a n d g a s t r o i n t e s t i n a l a d e n o m a s c a r c i n o m a s , s i m i l a r to Msh2 deficient hosts. Mlhl'''  and  t u m o r s d i s p l a y h i g h l e v e l s of M S I .  m t e r e s t i n g l y , s i m i l a r to Mlhl deficient m i c e , h o m o z y g o u s g e r m l i n e loss of Mlhl  leads to  t u m o r s o f h e m a t o p o i e t i c o r i g i n i n h u m a n s ( R i c c i a r d o n e et al., 1999; W a n g et al., 1999a). Mlhl  d e f i c i e n c y i n m i c e leads to i n f e r t i l i t y i n b o t h sexes, d u e to a l a c k of s p e r m  p r o d u c t i o n i n m a l e s , a n d a l a c k of m a t u r e o o c y t e s i n females. M a l e s d e m o n s t r a t e p a c h y t e n e arrest i n m e i o s i s (Baker et al., 1996; E d e l m a n n et al., 1996; P r o l l a et al., 1998). Pms2 d e f i c i e n c y i n m i c e leads to s u s c e p t i b i l i t y to l y m p h o m a s a n d to a lesser extent, s a r c o m a s , w i t h the t u m o r s d i s p l a y i n g M S I . H o w e v e r , i n contrast to Mlhl'''  animals,  Pms2''' m i c e d o not d e v e l o p intestinal t u m o r s . Ptns2''' female m i c e are fertile, b u t Pms2''' m a l e s are infertile d u e to a b n o r m a l c h r o m o s o m e p a i r i n g i n m e i o s i s (Baker et al, 1995; P r o l l a et al, 1998). T h e discordance i n the p h e n o t y p e s of Mlhl'''  a n d Pms2''' m i c e m a y be  d u e to different roles of the t w o proteins i n M M R a n d / o r i n other c e l l u l a r processes.  1.5 The transgenic lacl shuttle phage in vivo mutation detection system O v e r 200 different assay systems h a v e b e e n d e v e l o p e d to assess the g e n o t o x i c potentials o f m u t a g e n i c c o m p o u n d s as w e l l as genetic defects affecting D N A m u t a t i o n s ( K o h l e r et al, 1990). Reverse m u t a t i o n assays are w i d e l y u s e d i n m i c r o o r g a n i s m s . T o s t u d y m u t a g e n i c effects i n m a m m a l s , several in vitro assays h a v e b e e n d e v e l o p e d u s i n g m a m m a l i a n cells ( M i r s a l i s et al., 1995). These h a v e the a d v a n t a g e of b e i n g cost-effective;  13  however, they are often inefficient predictors of outcomes in whole animals, as there may be differences in the rate of cell division, metabolic activation or D N A repair processes (Kohler et al, 1990; Mirsalis et al, 1995). Transgenic animals carrying mutational target genes offer an invaluable tool for the investigation of in vivo outcomes of mutagenic processes (Mirsalis et al., 1995). The currently available systems differ with respect to the type of target gene used, number of copies, integration site, as well as the strain of mouse. The bacterial lad gene of the lac operon has been widely used for the analysis of spontaneous and induced mutations in several assay systems, in part due to the ease of using a colorimetric assay to rapidly screen for mutations. The LacI protein forms a homotetramer and represses the transcription of the lacZ gene by binding to the lac operator. The coding sequence of the lad gene contains 1083 base pairs including the termination codon, the first transcribed nucleotide is at position 1, and the translated codons from position 29 to 1111. The lad gene encodes a polypeptide of 359 amino acids. The first 59 amino acids form the D N A binding domain, followed by a core domain (60-359) which is involved in oligomerization and inducer binding (de Boer and Glickman, 1998). When there is a mutation that prevents the formation of a functional tetramer, transcription of the lacZ gene can occur (Figure 1.3). In most transgenic constructs, such as the Big Blue® strain (Stratagene) (Kohler et al., 1991), retrievable bacteriophage vectors contain the lad gene as a target for mutagenesis and the amino-terminal or a portion of the lacZ reporter gene. Highly efficient phage packaging extracts enable the recovery of the bacteriophage vector from the genomic D N A of any tissue. If there is an inactivating mutation in the lad gene, the amino-terminus of the lacZ gene is expressed from the incoming phage. This fragment complements the carboxy-terminal fragment provided by an appropriate host cell, resulting in P-galactosidase activity, that can be detected by a colorimetric screening assay (Figure 1.4) (de Boer and Glickman, 1998; Kohler et al, 1991). The hydrolysis of X14  Figure 1.3 Function of the lacl gene The lac operon includes the lacl, lacO and lacZ genes (A). The lacl protein forms a tetramer (B), which binds to the lacO region and prevents the transcription of lacZ (C). If a mutation in lacl renders the repressor complex defective, this allows for lacZ expression (D) and formation of the a- and co-portion of B-galactosidase (E).  15  16  Figure 1.4 The transgenic lad mutation detection system After harvesting the mouse tissues genomic D N A is prepared. The shuttle vector is excised from the high molecular weight genomic D N A using a A,-phage packaging extract. Rescued phages are plated in the presence of X-gal on lawns of the bacterial cell host, and the lad mutant frequencies are established by the ratio of mutant (blue) to nonmutant (colorless) plaques.  17  18  gal substrate by this enzyme results in formation of a blue plaque. The ratio of blue plaques to the total number of plaques is a measure of mutant frequency. Mutations detected by the blue-white color screening can be further analyzed through sequence analysis of the lacl target gene providing valuable information on the mutational spectrum (Kohler et al, 1991). The BC-1 lacl transgenic mouse strain was developed in our laboratory to facilitate mutational studies in murine hosts. Similarly to other existing lacl strains, a concatamerized lambda-phage shuttle vector contains the lacl mutational target gene and lacZa as a reporter. In contrast to other strains, the BC transgene was situated within rearranged murine immunoglobulin heavy chain locus inserted within the lambda-phage arms. The transgene has integrated on chromosome 19 in about 30 copies (Andrew et al, 1996). The transgene is transcriptionally silent. The interpretation of lacl mutational spectra is greatly facilitated by knowing the set of sites and mutations that can be recovered by the assay system. To date, over 10,000 mutants have been sequenced from lacl transgenic animals recovered from various tissue types from control animals or after treatment with a wide variety of mutagens (The Big Blue Website; de Boer and Glickman, 1998). Thus far, inactivating mutations at over 600 sites have been reported from transgenic animals, and new sites are still being recovered (The Big Blue Website; de Boer and Glickman, 1998). These mutations include all types of base substitutions, as well as insertion and deletions. Furthermore, a number of sites of inactivating mutations have been recovered from bacteria but not yet from animals (de Boer and Glickman, 1998). The lacl gene has been the most frequently used target in mutational studies. Due to the wide variety of inactivating mutations and high mutational saturation, the lacl forward mutational assay system is an excellent tool to study mutational specificity and mechanism. However, as mutational frequencies and spectra are influenced by the sequence, chromosomal location, methylation status and transcriptional activity of the target 19  genes (Dogliotti, 1996), the use of additional types of mutational assay systems will undoubtedly lead to a more complete understanding of each mutagenic mechanism.  1.6 Thesis goals  1. With the hypothesis that the mutational load may change during the evolution of MMR deficient tumors, mutant frequencies and spectra were investigated within Mshl''' murine thymic lymphomas, the predominant tumor type arising in these mice, using the BC-1 lacl-based transgenic shuttle-phage mutation detection system.  2. Next it was determined whether the dramatic increases in genomic instability found within Mshl''' thymic lymphomas compared to normal Mshl''' tissues are also a feature of PMS2-deficient thymic tumors and other D N A mismatch repair-deficient non-thymic cancers.  3. The hypothesis, that the different susceptibility of Mlhl''' and Pmsl''' mice to intestinal tumor formation might be due to a subtle difference in genomic instability, likely due to the differential roles of MLH1 and PMS2 in the D N A mismatch repair pathway, was tested.  1.7 Thesis summary This thesis is composed of-three parts. In the first section (Chapter 3), a quantitative and qualitative assessment of mutations within Mshl'''  murine thymic  lymphomas using a lad transgenic shuttle-phage mutation detection system was undertaken. Normal tissues of Mshl''' mice have been shown to demonstrate -10-15 fold elevations in lacl mutant frequency compared to wild type animals (Andrew et al., 1997). We addressed the question if this increase seen in normal tissues was sufficient to 20  induce tumor formation, and if the mutational load was changed in the tumors. Interestingly, we found further increases in lad mutant frequency in the Mshl''' thymic tumors, and in addition, a number of lad genes exhibited multiple mutations (BarossFrancis et al, 1998). In the second part (Chapter 4), we determined whether the above findings could be extended to tumors of mice lacking another MMR protein, PMS2. We also addressed the question of whether increased mutant frequencies would also be a feature of relatively rare, non-thymic tumors in MMR-deficient animals. While increased lad mutant frequencies and multiple mutations were again found in thymic tumors deficient for Pmsl or both Mshl and Pmsl, three non-thymic Mshl''' tumors did not demonstrate mutant frequency elevations when compared to normal Mshl''' tissues in the same host. These findings suggested that the mutation rate present in MMRdeficient normal tissues was sufficient for tumorigenesis, and that the elevated mutant frequencies within the thymic tumors might be a result of a lymphoma-specific secondary mutator mechanism. Furthermore, the increased genomic instability within the thymic lymphomas may explain, in part, the increased rate of tumorigenesis at this site (Baross-Francis et al., 2000). In the third section (Chapter 5) we investigated the roles of MLH1 and PMS2 mammalian MutL homologs in MMR and neoplasia. According to the recent model of mammalian MMR, these two proteins function as a heterodimer. However, while Mlhl' ' 1  mice frequently develop small intestinal tumors, Pmsl''' animals remain free of such cancers. Using the lad mutational reporter system, we found significantly higher mutant frequencies and increased C:G—>T:A transitions within the small intestinal epithelial cells oiMlhl' ' 1  versus Pmsl' '. This finding may explain, in part, the differential 1  susceptibility to tumors at this site.  21  Publications arising from work in this thesis  Baross-Francis, A., Andrew, S.E., Penney, J.E., and Jirik, F.R. 1998. Tumors of D N A mismatch repair-deficient hosts exhibit dramatic increases in genomic instability. Proc. Natl. Acad. Sci. USA, 95: 8739-8743.  Baross-Francis, A., Milhausen, M.K., Andrew, S.E., Jevon, G., and Jirik, F.R. 2000. Tumors arising in D N A mismatch repair-deficient mice show a wide variation in mutant frequency as assessed by a transgenic reporter gene. Carcinogenesis, 21: 12591262.  Baross-Francis, A., Makhani, N . , Liskay, R.M, and Jirik, F.R. 2000 Elevated mutant frequencies and increased C:G—>T:A transitions in Mlhl''' versus Pmsl''' murine small intestinal epithelial cells. Submitted for publication.  22  Chapter 2 Materials and Methods  2.1 Transgenic mice The BC-1 lacl shuttle phage transgenic mouse line was previously developed in Dr. Frank R. Jirik's laboratory (Andrew et ah, 1996). The Mshl knockout mice were given to us by Dr. Tak W. Mak, and the Pmsl and Mlhl  deficient lines were provided by Dr.  R. Michael Liskay. Mice were viral antibody-free and housed in a barrier facility according to institutional guidelines.  2.1.1 Breeding strategies MshT'- (Reitmair et al, 1995), PmsT'' (Baker et al, 1995) and MlhV ' 1  (Baker et al,  1996), mice were crossed with the BC-1 lacl transgenic line. Hemizygous BC-1 mice carry -30 copies of a A,-phage shuttle vector transgene containing the lacl reporter gene (Andrew et al, 1996). Mshl,  Pmsl, or Mlhl heterozygous lacP mice were bred to obtain  lacP knockouts. In order to produce Mshl  and Pmsl  double knockout lacP mice, Mshl  and Pmsl knockout or heterozygous lacP parents were used, except for Pmsl' ' males, 1  which are sterile (Baker et al, 1995).For the Mlhl  line both male and female knockout  mice are infertile (Baker et al, 1996), thus only Mlhl  heterozygous lacP parents were  used for breeding. After crossing the MshT'', PmsT'' and MlhV''  mice (C57BL/6) with  the BC-1 lacl transgenic line (BALB/c), the resulting mice are of a mixed C57BL/6 and BALB/c background.  2.1.2 Preparation of tail D N A for genotype testing of mice For genotyping, 0.5 cm tail clips, obtained from anesthetized mice, were digested in 300 ul lysis buffer containing 1.2 mg/ml proteinase K, 50 m M Tris-HCl p H 8.0, 10 mM EDTA, 0.1% SDS, and 100 mM NaCl. Following heat-inactivation of the proteinase at 100 °C for 10 minutes, 1 u.1 of a 20-fold dilution was used in PCR reactions to 23  determine the Mshl  (Reitmair et al, 1995), Pmsl  (Baker et al, 1995), Mlhl  (Baker et al,  1996), and lad genotypes (Andrew et al, 1996).  2.1.3 Determination of the Mshl genotype A three primer PCR assay was used to distinguish between the 3' boundary of the wild type and the targeted alleles using the following primers: GCTCACTTAGACGCCATTGT;  L916, A A A G T G C A C G T C A T T T G G A ;  U771, Lllll,  GCCTTCTTGACGAGTTCTTC (Reitmair et al, 1995). Primers U771 and L916 amplify an -170 bp fragment from the wild type Mshl allele, whereas primers 17772 and  Lllll  yield a 460 bp fragment identifying the targeted allele. PCR was performed in a 50 ul reaction containing -100 ng of genomic DNA, 40 pmol of L916, and 20 pmol of 17772 and L2222 primers, 0.2 pmol of each dNTP, 1.5 m M M g C l  2  and 0.25 U of Taq  polymerase (Pharmacia). Following an initial denaturation step at 94 °C for 10 minutes, 40 cycles of PCR at 94 °C for 1 min, 62 °C for 30 sec, 72 °C for 2 min, were performed. The resulting D N A fragments were separated in a 2% agarose gel. A picture a gel is shown on Figure 2.1 (A).  2.1.4 Determination of the Pmsl genotype Three oligonucleotide primers were used for a PCR reaction to distinguish the 5' boundary  of  the  endogenous  TTCGGTGACAGATTTGTAAATG;  from Pl-1,  targeted  allele:  TTTACGGAGCCCTGGC;  Pl-1, P2-3,  T C A C C A T A A A A A T A G T T T C C C G (Baker et al, 1995). Primers P2-2 and P2-2 give an -180 bp product that is a diagnostic of the targeted allele, whereas primers P2-2 and P23 yield the wild type allele product of -300 bp. PCR was performed in a 50 |il reaction containing -100 ng of genomic DNA, 40 pmol of P2-2, and 20 pmol of P2-2 and Pl-3 primers, 0.2 pmol of each dNTP, 1.5 m M MgCl , and 0.25 U of Taq polymerase 2  (Pharmacia). The cycling conditions were 94 °C for 5 min, 61 °C for 1 min, 72 °C for 1 24  Figure 2.1 Agarose gels resolving PCR fragments from genotyping reactions Typical results are shown for testing A) Mshl,  B) Pmsl, C) Mlhl  and D)lacl genotypes.  D N A molecular weight markers are in the first lane (left side) of the gels. 1 kb D N A ladder (Gibco BRL) was used for A , B and D; 100 bp D N A ladder (New England Biolabs) was used for C.  25  A) Msh2 PCR  B) Pmsl PCR  u  —>  * < << <  •£  a + + + + +  + + + +  |- 300bp 180bp  460bp •170bp  C) Mlhl PCR  D) lacl  PCR  — —  Q  -  -+-+-+-++  — 260bp 200bp  26  —  - , ^ , U 500bp  min; 94 °C for 1 min, 59 °C for 1 min, 72 °C for 1 min; 94 °C for 1 min, 57 °C for 1 min, 72 °C for 1 min; 94 °C for 1 min, 55 °C for 1 min, 72 °C for 1 min; 94 °C for 1 min, 55 °C for 30 sec, 72 °C for 1 min (35 cycles); followed by an extention at 72 °C for 5 min. The PCR products were separated in a 2% agarose gel. A picture a gel is shown on Figure 2.1 (B).  2.1.5 Determination of the Mlhl genotype A PCR reaction was used to distinguish the 3' boundary of endogenous and targeted Mlhl  alleles with the following oligonucleotide primers:  A G G A G C T G A T G C T G A G G C ; MLH1-U,  TTTCATCTTGTCACCCGATG;  G A T C T C G A C G G T A T C G A T A A G C (Baker et al, 1996). Primers MLHl-a  and  MLHl-a, MLH1-T5, MLH1-U  yield a -260 bp product diagnostic of the untargeted allele, compared to MLHl-a  and  MLH1-T5, which give the targeted allele product of -200 bp. PCR was performed in a 25 ul reaction containing -100 ng of genomic DNA, 40 pmol of MLHl-a, 7/5 and MLH1-U,  20 pmol of MLH1-  0.2 mM of each dNTP, 1.5 mM MgCl , and 0.25 U of Taq polymerase 2  (Pharmacia). The cycling conditions were 94 °C for 4 min; followed by 30 cycles of 94 °C for 1 min, 54 °C for 1 min, 72 °C for 1 min; and a final extention of 72 °C for 3 min. The resulting D N A fragments were separated in a 3% agarose gel. A picture a gel is shown on Figure 2.1 (C).  2.1.6 Determination of the lad genotype lad transgenic mice were identified by PCR using primers spanning the lad gene: -11, G A C A C C A T C G A A T G G T G C ; and 487, C T G G T C A G A G A C A T C A A G . The -11 and 487 primers yield an -500 bp product diagnostic of the transgene. PCR was performed in a 25 u,l reaction containing -100 ng genomic DNA, 10 pmol of -11 and 487 primers, 0.1 m M of each dNTP, 1.5 m M of MgCl , and 0.25 U of Taq polymerase (Pharmacia). 2  The PCR cycling conditions were 95 °C for 3 min, followed by 35 cycles of 95 °C for 45  27  sec, 61 °C for 1 min, and 72 °C for 1 min (Andrew et al, 1996). The PCR products were separated in a 2% agarose gel. A picture a gel is shown on Figure 2.1 (D).  2.1.7 Agarose gel electrophoresis D N A fragments were resolved by electrophoresis at 100-150V in agarose gels (23% agarose in T A E buffer (40 m M Tris-acetate, 1 m M EDTA)). The agarose gels containing the D N A were stained in 1 u.g/ml ethidium bromide solution for 45-60 minutes. D N A was visualized by UV illumination and the image was captured using a Gel Doc 1000 gel documentation system (Bio Rad).  2.1.8 Obtaining mouse tissues Mice at the ages specified in the relevant chapters were sacrificed by carbondioxide inhalation. The tissues or tumors were rapidly removed and flash frozen in liquid nitrogen, before storage at -80 °C until used for D N A isolation.  2.1.9 Characterization of tumors MMR-deficient mice that appeared moribund were sacrificed by carbon-dioxide inhalation. Tumor types were established by necropsy, and histological examination of hematoxylin and eosin-stained tissue sections.  2.1.10 Isolation of small intestinal epithelial cells The middle one-third (-15 cm) of the small intestine of each mouse was flushed out with sterile phosphate-buffered saline, then inverted using a probe. The inverted small intestine was placed in 3 ml of sterile buffer containing 75 m M KCI and 20 m M EDTA, and then cracked by pulling it up and down several times using a 5 ml syringe to break the epithelial cells from the intestinal wall (The Big Blue Website). The solution  28  containing the epithelial cells was flash-frozen in liquid nitrogen prior to D N A isolation.  2.2 Measurement of lacl mutational frequency  2.2.1 Isolation of genomic D N A Frozen tissue or tumor (-100 mg) was transferred to a 7 ml Wheaton Dounce tissue grinder containing 3 ml douncing buffer (6 mM Na HP0 130 mM NaCl, 13 m M 2  4/  KCI, 1.5 m M K H P 0 , and 10 mM EDTA, p H 8.0). The tissue was homogenized using a 2  4  Wheaton pestle B and transferred to a 50 ml conical tube. (For the small intestinal epithelial cells the procedure described in 2.1.10 was followed instead of the above.) 3 ml of Proteinase K solution (2 mg/ml Proteinase K, 2% SDS, and 100 m M EDTA, p H 7.5) was quickly added and mixed by inverting the tube a few times. The mixture was incubated in a 55 °C water bath for 3 hours. A n equal volume of phenol/chloroform saturated with TE (10 m M Tris-HCl, p H 7.5,1 m M EDTA, p H 7.5) was added and the mixture was inverted 20 times until an emulsion formed. The emulsions were then centrifuged at 1000 g for 10 minutes and the aqueous phase was transferred to a new tube with a disposable large-bore transfer pipet. This phenol/chloroform extraction was repeated followed by an extraction with 5 ml of chloroform. Two volumes of 100% ethanol were added to the final aqueous phase and mixed by swirling and inversion until a visible D N A precipitate formed. The D N A precipitate was then transferred to a sterile 1.5 ml microcentrifuge tube and let dry for 5 minutes. The genomic D N A was dissolved in 80-300 |il TE buffer for a minimum of 12 hours at room temperature, then stored at 4 °C (Kohler et al, 1990).  29  2.2.2 Preparing the SCS-8 plating culture The E. coli SCS-8 (recAl, endAl, mcrA, A(mcrBC-hsdRMS-mrr),  A(argF-lac)U169,  <j)80dlacZAM15, Tn20(tet )) host strain (Stratagene) was maintained by streaking a r  colony onto an NZY-tetracycline agar plate. The bacterial streak plate was incubated overnight in a stationary 37 °C air incubator and then stored at 4 °C for up to four weeks. To prepare the SCS-8 plating culture, 20 ml of NZY medium, supplemented with 0.25% (w/v) maltose and 12.5 mM MgS0 was inoculated with a single colony of E. coli 4  SCS-8 cells and grown at 37 °C overnight with shaking at 200 rpm. The culture was then centrifuged at -1000 g for 10 minutes to pellet the bacterial cells. The supernatant was discarded and the pellet was gently resuspended in sterile 10 m M MgS0 at OD =0.5. 4  600  The SCS-8 plating culture was stored at 4 °C for up to two weeks (Kohler et al., 1991; Kohler et al, 1990).  2.2.3 Performing the packaging reaction The lambda shuttle vector was recovered from the genomic D N A using Transpack in vitro lambda packaging extract (Stratagene). 8 u l of genomic D N A (adjusted to 0.5 mg/ml) was incubated with the first packaging extract (orange tube, Stratagene) for 90 minutes at 30 °C, and then 10 u l of the second extract (blue tube, Stratagene) was added and incubated for another 90 minutes. The reaction was terminated by addition of 972 u l of SM buffer (10 mM NaCl, 8 m M MgS0 , 50 mM Tris4  HCl, p H 7.5, 0.01% gelatin). The terminated reaction was kept at 4 °C until used for plating. If the entire packaged D N A was not plated on the same day it was packaged, 50 u l of chloroform was added per ml of packaged D N A sample, gently vortexed and stored at 4 °C for up to one week (Kohler et al, 1991).  30  2.2.4 Plating the packaged D N A samples Rescued phage were plated on SCS-8 (Stratagene) bacterial lawns containing Xgal. 1.5 ml of OD =0.5 SCS-8 plating culture was aliquoted into a 50 ml conical tube for 600  each packaging reaction to be plated. 200 u.1 of packaging reaction containing rescued phage from mouse genomic D N A were added to each tube of host cells and incubated at 37 °C for 15 minutes. 35 ml of molten NZY top agarose (0.35% agarose containing 1.5 mg/ml X-gal) was added to each tube, mixed by swirling, and immediately plated onto 25 cm x 25 cm assay trays containing NZY agar. These dishes were then incubated overnight at 37 °C (Kohler et al, 1991).  2.2.5 Screening and analyzing the assay trays Plates were examined for the presence of blue mutant plaques on a background of nonmutant colorless plaques. The ratio of blue plaques to colorless plaques was taken as a measure of lacl mutant frequencies. Density of the plaques was limited to 15,000 plaque-forming units (pfu) per plate (25 cm x 25 cm) to ensure accuracy in detection of plaques with the mutant phenotype (Kohler et al, 1991). Mutant frequencies were determined from 4-10 packaging reactions per sample. The mutant frequency variation between different packaging reactions was not bigger than between plates from the same packaging reaction (0.5-2 fold differences from final result).  2.2.6 Verifying putative mutant plaques Blue mutant plaques were picked from the assay trays with wide-bore pipet tips and transferred to sterile microcentrifuge tubes containing 500 ul SM buffer and 50 u.1 chloroform. The tubes were incubated for overnight at 4 °C to allow the phage particles to elute from the agar plug. 2 ul of the 1:20 dilution of the eluted phage was mixed with 200 u.1 SCS-8 plating culture in a 15 ml tube and incubated at 37 °C for 15 minutes. 3 ml of molten NZY top agarose containing 1.5 mg/ml X-gal was added to each tube and the 31  content was poured on a 100-mm NZY agar plate. The plates were incubated overnight at 37 °C. The mutant plaques that replated with a blue phenotype were isolated with wide-bore pipet tips and transferred to 200 u.1 SM buffer mixed with 25 u.1 chloroform. The tubes were stored at 4 °C for at least overnight to allow the phage particles to elute from the agar plug. The eluted phage templates were later used for amplification of mutant lacl genes (Andrew et al., 1996).  2.2.7 Statistical analysis of mutant frequencies Mean mutant frequencies resulting from 3-5 animals per group were compared using Student's f-test. Significant differences were determined at a 95% level of confidence (p<0.05).  2.3 Determination of lad mutational spectrum  2.3.1 Amplification of mutant lacl genes Following isolation and verification of single mutant clones, the 1.2 kb lacl genes were amplified by PCR (Andrew et al., 1996) of phage templates using the following lacl primers: -22, G A C A C C A T C G A A T G G T G C ; and 2202, C C G C T C A C A A T T C C A C A C A . PCR was performed in a 25 u,l reaction containing 1 ul of phage template, 10 pmol of -22 and 2202 primers, 0.1 m M of each dNTP, 1.5 m M of MgCl , and 0.25 U of Taq 2  polymerase (Pharmacia). The PCR cycling conditions were 95 °C for 3 min, followed by 35 cycles of 95 °C for 45 sec, 61 °C for 1 min, and 72 °C for 1 min. 5 u.1 of the PCR products were separated in a 2% agarose gel in order to check quality before further purification.  32  2.3.2 Purification of lacl template for sequencing 20 |il of the amplified 1.2 kb product of the lacl gene was ethanol precipitated, washed in 70% ethanol, and then resuspended in 20 u.1 d H 0 . This D N A was further 2  purified using QIAquick D N A purification kit (Qiagen) (Andrew et ah, 1996). 5 |il of the -40 ul end product was separated on a 2% agarose gel to check quality and the remainder of the D N A was used for sequence analysis.  2.3.3 Sequencing of lacl mutants Templates obtained from randomly selected lacl mutants were sequenced using primers that spanned the lacl gene (Andrew et al, 1996) with an ABI 388 D N A sequencing instrument (Applied Biosystems). The first 59 amino acids of the lacl protein make up the D N A binding domain and over 50% of the lacl mutations occur within this region (Gu et al, 1994). Therefore, the sequencing strategy employed involved initial sequencing with the -11 forward and 487 reverse primers to amplify an -500 bp product spanning this region (the numbers of these primers indicate their nucleotide positions within the lacl gene shown in Figure 2.2). Subsequent sequencing reactions with the forward primers 455 and 798, and reverse primers 899 and 1201 were performed only when the -11 and 487 primers failed to reveal mutations. All mutations were confirmed with a minimum of two sequencing runs with at least two different primers.  2.3.4 Sequence analysis for mutation detection The alignment of the consensus and mutant lacl sequences and detection of mutations was performed using the 'mutalign' script that utilizes the UNIX-based sequence analysis package SEQNCE (Delaney Software Ltd.). A n examle containing a G—>T base substitution is shown on Figure 2.3.  33  2.3.5 Statistical analysis for the comparison of mutational spectra The distribution of different types of mutations within sample groups were compared by Chi-square analysis (%2-test). Significant differences were determined at a 95% level of confidence (p<0.05).  34  Figure 2.2 Sequence of the lacl gene The sequence and translated region of the lacl gene are shown on the top. Nucleotide 1 is the starting position for transcription. Below, the position and orientation of various PCR primers used for sequencing are presented.  35  -40  -20  20  1  GAC ACC A T C GAA TGG TGC AAA ACC TTT CGC GGT A T G GCA TGA T A G CGC CCG GAA GAG AGT CAA T T C AGG GTG 40  60  80  GTG AAT GTG AAA CCA GTA ACG T T A T A C GAT GTC GCA GAG TAT GCC GGT G T C TCT TAT CAG A C C GTT T C C CGC ^ Va I 100  L y s  Pro  Vat  7h r  Leu  Tyr 120  Asp  Va I  Ala  Glu  Ty r  Ala  G I y 140  Vat  Set  Ty r  Gin  T rt r  Vat 160  Se r  Ar g  GTG GTG A A C CAG GCC AGC CAC GTT TCT GCG AAA ACG CGG GAA A A A GTG GAA GCG GCG A T G GCG GAG C T G AAT ^  Va I  Vat  As n  Gin  Ala 180  Se r  His  Vat  Se r  Ala  L y s  Th r 200  Ar g  Glu  Ly s  Va I  Glu  Ala Ala 220  Met  Ala  Glu  Leu  As n  T A C A T T CCC AAC CGC GTG GCA CAA CAA C T G GCG GGC AAA CAG TCG T T G C T G ATT GGC GTT GCC A C C T C C AGT ^  Ty r 240  lie  Pro  A s n  Ar g  Va I  Ala  Gin 260  Gin  Leu  Ala  G I y  L y s  Gin 280  Sa r  Leu  Leu  lie  G I y  Va I  Ala 300  Th r  Se r  Se r  C T G GCC C T G CAC GCG CCG TCG CAA ATT GTC GCG GCG ATT AAA TCT CGC GCC GAT CAA C T G GGT GCC AGC GTG ^  Ala  Leu  Leu  His 320  Ala  Pro  Ser  Gin  lie  Vat Ala 340  Ala  lie  L y s  Ser  Arg  Ala 360  Asp  Gin  Leu  G Iy  Ala  S e r  Vat 380  GTG GTG T C G A T G GTA GAA CGA AGC GGC GTC GAA GCC TGT AAA GCG GCG GTG CAC AAT C T T C T C GCG CAA CGC ^  Va I  Vat  S e r  Mat  Vat  Glu Ar g 400  S e r  G I y  Va I  Glu  Ala  Cys 420  L y s  Ala  Ala  Va I  H I s  A s n  Leu 440  Leu  Ala  Gin  Ar g  GTC AGT GGG C T G A T C ATT AAC TAT CCG C T G GAT GAC CAG GAT GCC ATT GCT GTG GAA GCT GCC T G C ACT AAT  y Va I  Ser  G I y  Leu  lie  lie  A s n  Tyr  460  Pro 480  Leu  Asp  Asp  Gin  Asp  Ala  lie 500  Ala  Vat  Glu  Ala  Ala  Cys  Th r  A s n  520  CTT CCG GCG T T A T T T CTT GAT GTC TCT GAC CAG ACA CCC A T C A A C AGT A T T ATT T T C T C C CAT GAA GAC GGT ^  l/s I  Pro  Ala  Lou  Ph e 540  Leu  Asp  Vat  S e r  Asp  Gin  Th r 560  Pro  lie  As n  S e r  lie  lie 580  Ph e  Se r  His  Glu  Asp  G I y  ACG CGA C T G GGC GTG GAG CAT CTG GTC GCA T T G GGT CAC CAG CAA A T C GCG C T G T T A GCG GGC CCA T T A AGT ^  Thr 600  Ar g  Leu  G Iy  Va I  Glu  His  Leu 620  Va I  Ala  Leu  G Iy  His  Gin  Gin  lie  Ala  Leu  Leu  Ala  G Iy 660  640  Pro  Leu  Se r  TCT GTC T C G GCG CGT C T G CGT C T G GCT GGC TGG CAT AAA TAT C T C ACT CGC AAT CAA A T T CAG CCG A T A GCG ^  Sac  Va I  S e r  Ala 680  Ar g  Leu  Ar g  Leu  Ala  G I y Tr p 700  His  L y s  Tyr  Leu  Th r  Ar g 720  As n  Gin  lie  Gin  Pro  lie  Ala 740  GAA CGG GAA GGC GAC TGG AGT GCC A T G T C C GGT TTT CAA C A A A C C A T G CAA A T G C T G AAT GAG GGC A T C GTT ^  G I u  Ar g  Glu  G I y  Asp  Tr p Se r 760  Ala  Met  Se r  G Iy  Ph e  Gin 780  Gin  Th r  Mat  Gin  Met  Leu  As n 800  Glu  G Iy  lie  Va I  CCC ACT GCG A T G C T G GTT GCC AAC GAT CAG A T G GCG C T G GGC GCA A T G CGC GCC A T T A C C GAG T C C GGG C T G  y Pr  o  Th r Ala 820  Met  Leu  Vat  Ala  As n  Asp 840  Gin  Met  Ala  Leu  G Iy  Ala  Met 860  Ar g  Ala  lie  Th r  Glu  Ser  G Iy  L a u  880  CGC GTT GGT GCG GAT A T C TCG GTA GTG GGA T A C GAC GAT A C C GAA GAC AGC T C A TGT TAT A T C CCG CCG T T A ^  Ar g  Va I  G Iy  Ala  Asp 900  lie  S e r  Va I  Va I  G Iy  Tyr  Asp . 920  Asp  Th r  Glu  Asp  Se r  Se r Cys 940  Tyr  lie  Pro  Pro  Leu  A C C A C C A T C A A A CAG GAT T T T CGC C T G C T G GGG CAA A C C AGC GTG GAC CGC T T G C T G C A A C T C TCT CAG GGC ^  Thr 960  Thr  lie  L y s  Gin  Asp  Ph e  Ar g 980  Leu  Leu  G Iy  Gin  Th r  Ser Va I 1000  Asp  Ar g  Leu  Leu  Gin  Leu 1020  S e r  Gin  G I y  CAG GCG G T G AAG GGC AAT CAG CTG T T G CCC CTC T C A CTG G T G A A A AGA AAA A C C A C C CTG GCG C C C AAT ACG ^  G In  Ala  Va I  L y s 1040  G Iy  As n  Gin  Leu  Leu  Pro  Va I  Se r  Leu  Va I  L y $  Ar g  1060  L y s 1080  Thr  Th r  Leu  Ala  Pro  As n  Th r 1100  CAA A C C GCC TCT C C C CGC GCG T T G GCC GAT TCA T T A A T G CAG CTG GCA CGA CAG CTT T C C CGA CTG GAA AGC ^  G I n  Th r  Ala  Se r  Pro  Ar g Ale 1120  Leu  Ala  Asp  S e r  Leu  Mel 1140  Gin  Leu  Ala  Ar g  Gin  Vat  Se r 1160  Ar g  Leu  Glu  Sar  GGG CAG TGA GCG CAA CGC AAT T A A TGT GAG T T A GCT CAC T C A T T A GGC A C C CCA GGC T T T ACA C T T TAT GCT ^  G I y  G I n 1180  1200  1220  T C C GGC T C G TAT GTT GTG TGG AAT TGT GAG CGG ATA ACA A T T T C A C A C A  -11  primer  455  487 primer ^  primer ^  j  lacl  36  ^  899 primer ^  1 2 U  1  primer t  Figure 2.3 Sequence alignment of a mutant with the wild type lacl gene A n alignment is shown between the consensus wild type lacl sequence, and two sequences from the n6a8 lacl mutant. The middle region of the mutant lacl gene was sequenced with the 455 forward and the 899 reverse primers. There is a G—»T transversion in the fourth line falling at nucleotide 588.  all sequences match (there is no mutation) *  mutation (mutants are identical, but different from consensus)  x  sequencing error (one mutant sequence matches consensus)  X  one non-matching sequence (possible mutation, second sequence is needed)  37  consens n6a8 . 8 9 9 _ n6a8.455  CTCGCGCAACGCGTCAGTGGGCTGATCATTAACTATCCGCTGGATGACCAGGA-TGCCAT TTAACTATCCGGTGGATGACCAAGAATGCCAT xxxxxxxxxxxXxxxxxxxxxxXxx  xxxxxx  consens n 6 a 8 . 8 9 9_ n6a8.455  TGCT-GTGGAAGCTGCCTGCACTAATGTTCCGGCGTTATTTCTTGATGTCTCTC TGCTTGTGAAAGCTGCCTGCACTAAATGTCCGGCGTTATTTCTTGATGTC CGTTATTTCTTGATGTCTCTGACCAGA xxxx xxxXxxxxxxxxxxxxxxxxXXXxxxxx--  consens n6a8 . 899_ n6a8.4 5 5  CACCCATCAACAGTATTATTTTCTCCCAT CACCCATCAACAGTATTATTTTCTCCCAT-AAGACGGTACGCGACTGGGCGTGGAGCATC CACCCATCAACAGTATTATTTTCTCCCATGAAGA  consens n 6 a 8 . 8 9 9_ n6a8.4 5 5  TGGTCGCATTGGGTCACCAGCAAATCGCGCTGTTAGCGGGCCCATTAAGTTCTGTCTC TGGTCGCATTGGGTCACCAGCAAATCGCGCTGTTAGCGGTC TGGTCGCATTGGGTCACCAGCAAATCGCGCTGTTAGCGGTCCCATTAAGTTCTGTCTCGG  consens n6a8.899. n6a8.455  CGCGTCTGCGTCTGGCTGGCTGGCATAAATATCTCACTCGCAATCAAATTGAGCCGATAG CGCGTCTGCGTCTGGCTGGCTGGCATAAATATCTCACTCGCAATCAAATTCAGCCGATAG CGCGTCTGCXTCTGGCTGGCTGGCATAAATATCTCACTCGCAATCAAATTCAXCCGATAG -x-x-  consens n6a8.899_ n6a8.455  CGGAACGGGAAGGCGACTGGAGTGCCATGTCCGGTTTTCAACAAAC^ CGGAACGGGAAGGCGACTGGAGTGCCATGTCCGGTTTTCAAC^ CGGAACGGGAAGGCGACTGGAGTGCCATGTCCXGTTTTCAA x  consens n6a8.899. n6a8.455  ATGAGGGCATCGTTCCCACTGCGATGCTGGTTGCCAACGATCAGATGGCGCTGGGCGCAA ATGAGGGCATCGTTCCCACTGCGATGCTGGTTGCCAACGATCAGATGGCGCTG^ ATGAGGGGATCGTTCCCACTGCXATGCTGGTTGCCA^ x x x  consens n6a8.899_ n6a8.455  TGCGCGCCATTACCGAGTCCGGGCTGCGCGTTGGTGCGGATATCTCGGTAGTGGGATACG TGCGCGCCATTACCGAGTCCGGGCTGCGCGTTGGTGCGGATATCTCGGTAGTGGGATACG TGCGCGCCATTACCGAATTCXGGCTGCGCGTTGGTGCAGATATCTCGGTAGTGGGATACG x-x-x X  consens n6a8.899. n6a8.455  ACGATACCGAAGAC^GCTCATGTTATATCCCGCCGTTAACCACCATCAAACAGGATTTTC ACGATACCGAAGACAGCTCATGTTATATCCCGCCG ACGATACC XXXXXXXXXXXXXXXXXXXXXXXXXXX  38  Chapter 3 Thymic tumors of MSH2-deficient hosts exhibit dramatic increases in genomic instability  3.1 Introduction As the spontaneous mutation rate of human cells appears to be insufficient to account for the multiple mutations that are required for the evolution of most malignancies (Loeb, 1991) it has been proposed that tumor evolution requires a mutator phenotype, such as would result from disruption of the genes encoding D N A repair proteins (Loeb and Christians, 1996). In keeping with this prediction, highly penetrant mutations in human mismatch repair (MMR) genes (primarily the mutS hMSH2, and the mutL homologs hMLHl,  hPMS2 and hPMSl)  homolog  have been found to be  responsible for most cases of the familial cancer syndrome, hereditary non-polyposis colorectal cancer (HNPCC) (Kinzler and Vogelstein, 1996). Affected individuals inherit a germline mutation in one allele and in association with the loss of the second allele develop malignancies, predominantly of the proximal colon, endometrium and ovary. Interestingly, mice with homozygous germline deficiencies of specific MMR genes only generate a limited spectrum of tumor types with varying frequencies and latencies (Prolla et al, 1996; Reitmair et al, 1996a; Reitmair et al, 1996b; Reitmair et al, 1995). Msh2~ ~ mice, for example, develop neoplasms with a high frequency, but the spectrum /  is dominated by thymic lymphomas (Reitmair et al, 1995). This suggests that the increased mutation rate afforded by MMR-deficiency may be necessary, but not always sufficient for malignant transformation to occur. Tumors and cell lines lacking MMR almost invariably demonstrate microsatellite instability and malignant lines generally exhibit greatly elevated mutation rates of specific indicator genes (Bhattacharyya et al, 1995; Bhattacharyya et al, 1994; Dunlop, 1996; Eshleman et al, 1995; Eshleman et al, 1996; Glaab and Tindall, 1997; Kat et al, 1993; Malkhosyan et al, 1996; Ohzeki et al, 1997; Phear et al, 1996). Such studies have 39  limitations, however, as MMR deficient tumor lines may contain uncharacterized mutations able to potentiate genomic instability, and also as mutation rates can be greatly influenced by cell culture conditions (Richards et al, 1997). Other than by the semiquantitative method of measuring instability at microsatellites, little is known about gene-specific mutant frequencies (or spectra) in primary MMR deficient tumors. Mice with induced deficiencies of specific M M R components provide a unique opportunity for quantifying genomic instability, both in normal tissues and spontaneously arising tumors. To this end, shuttle phage-based mutation detection systems (Glazer et al, 1986; Mirsalis et al., 1995) have been used to investigate tumor specific mutant frequencies. Thus, adenocarcinomas arising in bi-transgenic mice carrying a lacl reporter and the polyomavirus middle-T oncogene failed to show an increase in mutant frequency (Jakubczak et al., 1996). Similarly, of eight thymic lymphomas in p53''' Big Blue lacl transgenic mice (Buettner et al, 1996; Sands et al, 1995) only one demonstrated a 2.3-fold lacl mutant frequency increase as compared to normal thymus. In contrast, we have found that thymic lymphomas arising within Mshl ' 1  hosts  uniformly demonstrate remarkable increases in lacl gene mutant frequency as compared to normal Mshl''' thymi.  3.2 Results  3.2.1 Mutant frequency in Msh2''' thymic lymphomas Mutant frequencies were determined for six thymic lymphomas, which arose spontaneously in 3-4 month old MshT'' mice (Table 3.1-1). Tumor specific frequencies were elevated as compared with those of normal thymi from 3-4 week old MshT'' animals (t-test, p=0.0039). The latter already demonstrate an ~15-fold elevation as compared to 3.1*10", the MF from thymi of D N A repair-proficient animals (Andrew et 5  al, 1997) (Table 3.1-II) (Figure 3.1). When compared to MshT'', and to MshT'  +  40  thymi, the  Table 3.1 Spontaneous mutant frequencies for Mshl' ' thymic lymphomas and normal 1  tissues Spontaneous lacl mutant frequencies for (I) Mshl' ' 1  thymic lymphomas, (II)  normal thymi (young mice), (III) Mshl' ' normal thymi (older mice), (IV) Mshl' ' 1  1  brains from mice with thymic lymphomas.  41  Mshl'  1  normal  ill  Tissue  Animal  Age (days)  Sex  Total PFU  a  Number of mutants  Mutant ftequency (*10 ) 152.0 323.3 208.4 279.7 817.0 525.8 5  Thymic lymphoma  A B C D E •F  115 121 139 132 123 81  F F F F F M  211 759 447 581 1729 1105  138,820 234,760 214,460 207,740 211,620 210,140  Mean  Tissue  Animal  Age (days)  Sex  Total PFU  Number of mutants  G H I  19 19 26  F F M  295,020 300,460 281,640  166 89 155  Mean  266.8+99.6  Mutant frequency (*10' ) 56.3 29.6 55.0  Mutant frequency*(100%-clonality) 25.6 26.3 39.7  47.0 ± 12.3  30.5 + 6.5  m Tissue  Animal  Age (days)  Sex  Total PFU  Number of mutants  Mutant frequency (*1Q- ) 58.9 41.7 5  Normal thymus  J K  132 103  F M  185,100 249,180  109 104  Mean  Tissue  Animal  50.3±8.6  Age (days)  Sex  Total PFU  Number of mutants  Mutant frequency (*10") 40.4 44.1 5  Normal brain  E F  123 81  F M  220,080 215,300  89 95  Mean a b  42.25±1.85  PFU = Plaque Forming Units See Table 3.2 for clonality  42  b  384.4+226.1  5  Normal thymus  Mutant frequency*(100%-clonality) 139.3 242.5 189.5 256.4 449.4 323.6  Figure 3.1 Spontaneous mutant frequencies for Mshl' ' thymic lymphomas and 1  normal tissues Raw lacl mutant frequencies of individual control and tumor tissues determined from plating data (gray bars) and the corresponding mutant frequencies after correction for clonality (black bars).  43  o o  o o  o o  o o  o o  o o  o o  o o  (9-0 U) Aouenbaji )ue)n|/\|  44  o o  average mutant frequency increase in the tumors was 8.2-, and -120-fold, respectively. There was variation in the mutant frequency observed amongst the tumors (3.2- to 17.4fold over the mean value of normal thymi). It was important to establish whether the elevations resulted from clonal expansions of cells harboring specific lacl gene mutations. To establish the approximate level of clonality for each of the tumors, as well as the mutation spectrum, a total of 100 lacl mutants from Mshl' '  thymic lymphomas,  1  and 58 lacl mutants from the normal thymi of 3-4 week old Msh2~ ~ mice were analyzed /  (Table 3.2) (Figure 3.2). From the sequencing data (Table 3.2) it was possible to establish the approximate level of clonality for each sample (Table 3.1). A total of 19 different mutations were observed to occur more than once among the various thymic lymphomas and control thymi. In three cases (G-»A at 93; frameshift at an (A)5 repeat at 135-139; G->A at 180), the sites corresponded to putative 'hot spots' previously identified in lacl genes recovered from normal Mshl' ' tissues (Andrew et al., 1997). In 1  spite of the fact that a number of the recurrent mutations likely arose independently of each other, mutations observed more than once per sample were eliminated to arrive at the corrected frequencies. To estimate the contribution of clonality to the mutant frequency data, 11 to 24 lacl mutants were sequenced per tumor (Table 3.2). After correction for clonality, a comparison of the mutant frequencies of thymic lymphomas and normal Msh2' ' /  thymi demonstrated an 8.6-fold (compared to 8.2-fold prior to  correction) elevation in the tumors when the mean values were compared (t-test, p=0.00064) (Table 3.1). In addition, after correction for clonality, the frequency variation observed between the different tumors was greatly reduced (Figure 3.1).  3.2.2 Mutant frequency of Msh2''~ normal tissues Table 3.1-1 and 3.1-II show the mutant frequencies of thymic lymphomas from 34 month old Msh2' ' /  mice as compared to data from Mshl' ' 1  45  normal thymi obtained at  Table 3.2 lacl mutation spectrum of MshT ' thymic lymphomas, and normal thymi. 1  lacl mutation spectrum for (I) MshT ' thymic lymphomas, (II) MshT ' normal thymi. 1  1  46  3£  00 £  ON OO ^  *  co CN  lu  CO  CN OO OO CN ~ ^  00  O  Ov  co  u a  w  N  O — Ifl  - 1 CN O — . —. O "  rn "  J5  a  S  B  0  u o  O  o rr-  l-  o  .2  <u ^ X5  VO C l  es  o o  1 *  l i s  —  —c OO IT) —1 CN  0  CJ  O  Tj  O  O  O  73 w 2 a c  3 C  cn • ~ £ O « U • S T> tn 3  'C o 3 C  s « J-  0  cj •o  CD  to  cn «  —  ^ 00 — co — CO TJ-  <-i VO  m in m  rn  10 CN  «  C O VO  « a  E -  CN  —  —  in  c  c O  C3  3  •o c  C 0)  E  a o Cn U  >•  O  H U H U  —1  t TT T  a  c  8  <c q < a  C/5  Q. D  ~  TJ-  o<o » „ 3U H U  CN  T  GO  CN  TTT g o a << i. CO  hUO DU <H <H  CB  IH  U  «N r~ = CN - ~ <N ^  2 ON  XI  E  o  CN  3  ca o  T3 C 3 O  H  a  Cu  U en  tn  e o H cj  2< a •3 t T 5 H 2o< U  55 >  vo cr> s ^ ^ M -  !C -  8 o H  a-  •2 < a •s T T  CN  CO  H  S  o VO VO  a  1^  CO  o  So  O VO  CN-  ~  Ov  .5 co  VO  CN  vo  ^ -4  CN  cn  c  ov  -a c  O  H  3  E  C  tn  l l  1 cn  47  W3  2 g.2 ^— C  C3  C3  1.2  E  o is  c _  S 3 .CO CS  3 -2  u  o  H  ^  CN —  c  <fflUQWfc  U  O 150  U  c  C  <u a, <u "g  C U  <n  CJ T3  u.  tn  cj ft  E « 3 c II  ll S  | l  C  1  T3  g  cj  CJ <n XI  5  c  C  Figure 3.2 lacl mutation spectrum of Mshl' ' thymic lymphomas, and normal thymi. 1  Mutation spectra of the lacl gene mutants rescued from normal Mshl''' thymi and Mshl''' thymic lymphomas are shown. The patterned bars represent the numbers of transitions falling at CpG sites.  48  49  3-4 weeks of age. To determine whether the elevated frequencies of the tumors might simply be a function of age, two normal thymi were isolated from 3.5 and 4 month old Mshl' ' 1  mice. The frequency data obtained (Table 3.1-III) (Figure 3.1), however, was not  significantly different (t-test, p=0.8) from that of normal 3-4 week old Mshl' ' 1  thymi.  Thus, the increased mutant frequencies of the thymic tumors were not attributable to age-related effects. Were the lacl mutant frequencies of the lymphomas solely a property of the tumors, or did lymphomas tend to arise in hosts having elevated mutational frequencies in all tissues? Brain, in contrast to most other tissues (such as liver, kidney, spleen), demonstrates little if any histological evidence of infiltrating lymphoma cells. The brains of two MshT'' mice (E and F) whose lymphomas exhibited the highest elevations in mutant frequency (Table 3.1-IV) (Figure 3.1) were analyzed. Mutant frequencies for the two brains were considerably lower than for the corresponding lymphomas in these same animals (t-test, p=0.0053), and were not significantly different from those of normal MshT'' brains or thymi (Andrew et ah, 1997) (a comparison of the combined E and F brain mutant frequencies with those of 3-4 week old, and 3-4 month old thymi by t-test was p=0.68, and p=0.58, respectively). Thus, the highly elevated lacl mutant frequencies appear to be not only intrinsic to, but also characteristic of the Mshl' ' thymic lymphomas. 1  3.2.3 Mutation spectrum of Mshl' " thymic tumors 1  After correcting for clonality, the numbers of transitions (including those falling at C p G sites), transversions, frameshifts (+1 or -1), insertions and deletions (>1 nucleotide) were compared between the tumors and the normal tissues (Table 3.2). A comparison of the pooled lacl mutation spectra of normal MshT'' thymi and the tumors (Figure 3.2) did not show any significant difference (%2 test, p>0.5). Mutations within individual mice did show variability, raising the possibility that lacl mutation spectrum might differ somewhat from tumor to tumor. However, as a relatively low number of 50  lacl mutants were obtained from each tumor, the apparent individual differences in spectrum may not be significant. In keeping with the paucity of repetitive sequences in the lacl gene, transitions dominated the mutation spectrum in both groups (Table 3.2). The percentage of transitions falling at CpG sites (potentially resulting from the deamination of 5-methylcytosine) were similar in normal and malignant tissues, accounting for -25% of transitions in both. The percentage of transitions falling at CpG sites was lower than observed previously, where they accounted for >50% of total transitions in normal Msh2  +/+  liver (Andrew et al, 1996), Msh2  +/+  and Mshl''' small  intestine and brain (Andrew et al., 1997), and the combined spectrum of testis, ovary and spleen in the lacl genes of Big Blue® mice (Kohler et al., 1991). Also, in Big Blue® thymi, >50% transitions fell at C p G sites (Buettner et al., 1996). The decreased percentage of transitions at CpG sites in the Mshl' ' thymi and lymphomas may be due 1  to lower levels of CpG methylation of the BC-1 lacl transgene in this tissue. Frameshifts (+/-1) were the second most common mutation type in both. Frameshifts (-60% for tumors and -80% for normal thymi) were observed at mono- or dinucleotide repeats of 3 or more units. The proportion of frameshift mutations was modestly elevated in the tumors (21% versus 15% in the Msh2~'' thymi). Transversion mutations varied considerably between animals; however, as the numbers were low, the significance of this observation is questionable (Table 3.2). Insertions or deletions (>1 bp) were rare in both normal and tumor DNA, with only 1 such lacl mutation found in each.  3.2.4 Multiple mutations in single lacl genes There were five examples of lacl genes having more than one mutation among the 77 independent tumor mutants (Table 3.3). In addition, mouse F yielded a complex lacl mutation C i i G G G A A A A A G T - > C i i G G A A A A A A T that could have resulted from 3  3  one -1 frameshift and one transition, or from one -1 frameshift and two transitions. In contrast, in the 39 lacl genes sequenced from normal tissues of BC-11 Mshl' ' mice, in this 1  51  Table 3.3 Multiple mutations occurring in single lacl genes from Mshl' ' 1  lymphomas and an Mshl' ' 1  thymic  normal thymus.  Multiple mutations in five lacl genes were found in Mshl'''  thymic lymphomas from  animals A, B and E; and in one Mshl''' normal thymus G. A n additional, more complex multiple mutation from a thymic lymphoma is described in the text.  52  Tissue  Animal  Thymic lymphoma  A  Thymus  # Mutations per lacl gene 2  Type of mutation (coding sequence) T->A T->C  Site of mutation  4  T-*C T->C T-*C T->C  451 547 549 552  B  2  A->G AA  320 350-352  E  2  C->G A->C  755 858  2  T->C T->C  87 162  2  C->T G->A  161 222  G  53  703 726  study and previously (Andrew et al., 1996; Andrew et al, 1997) we found only one mutant harboring more than one mutation (Table 3.3), thus the frequency of multiple mutations is significantly higher in the thymic lymphomas (%-test, p<0.05). In several 2  lacl genes recovered from the tumors clustering of mutations was observed. This was highlighted by the lacl mutant having four T->C transitions at positions 451, 547, 549, 552.  3.3 Discussion D N A repair genes, such as MSH2,  have been described as 'caretaker' genes,  responsible for the maintenance of genomic integrity. The inactivation of such genes results in increased genetic instability, which in turn leads to an increased rate of mutation in 'gatekeeper' genes that regulate cell proliferation and death (Kinzler and Vogelstein, 1997). The adenomatous polyposis coli (APC) and BAX genes, for example, belong to the latter category (Kinzler and Vogelstein, 1996; Rampino et al, 1997). In mice, Mshl  deficiency appears to play a role in accelerating intestinal tumorigenesis, an  effect due in part to APC mutations in Min ~ mice (Reitmair et al., 1996b). With regard to +/  lymphomagenesis in Mshl' ' mice, are the increased levels of genomic instability within 1  the normal thymi (Andrew et al., 1997) sufficient to account for the increased incidence of these tumors, or is a further mutational rate increase necessary for this process? As all Mshl' ' 1  lymphomas examined demonstrated significant increases in mutant  frequency as compared to normal Mshl' ' 1  thymi, it suggested that a superimposed  mutator phenotype might have been required to facilitate either tumor genesis or progression. Alternatively, the increases in mutant frequency might be simply indicative of a 'mitotic clock', registering cell divisions as a function of lacl gene mutations. Given that the MshT' thymic lymphomas demonstrate histological evidence of apoptosis, including a classical 'starry sky' appearance due to cells containing phagocytosed apoptotic cell bodies, it is difficult to estimate the number of tumor cell 54  divisions that have transpired by the time an animal is moribund. Thymic lymphomas in p53' ' mice have also been studied using Big Blue mice containing a lacl transgene f  similar to ours (Buettner et al, 1996; Sands et al, 1995). Assuming that similar numbers of cell divisions occur in the y53~'~ thymic lymphomas, it is remarkable that with one exception these tumors failed to show an increase in lacl gene mutant frequency. It is clear that proliferating Mshl''' cells should acquire mutations at a greater rate than MMR-proficient cells. However, if the lacl mutant frequency increases in the  Mshl ' 1  tumors were simply a function of errors accumulated during cell division, then some consistent evidence of induction of lacl mutant frequency would have been anticipated in all of the p53 " lymphomas. Perhaps the single y53~'~ lymphoma showing a 2.3-fold v  induction of lacl mutation might reveal evidence of MMR deficiency. The comparison of the pooled BC-1 lacl mutations obtained from the Mshl''' lymphomas and the Mshl' ' thymi (Figure 3.2) did not reveal any gross differences in 1  spectrum, except for the trend towards increased frameshifts in the tumors. A greater number of mutants would have to be characterized in order to unveil tumor specific differences in mutation spectrum or in the position of the mutations on the lacl gene. The findings, however, do suggest that the mutagenic mechanisms responsible for the normal thymic lacl mutation spectrum may be similar to those operating within the lymphomas. Thus, except for a possible increase in frameshifts, the pooled lacl mutation spectra of the lymphomas did not suggest the existence of novel, tumor-specific mutagenic mechanisms such as were observed in solid tumor xenografts and in hypoxic cells in culture (Reynolds et al, 1996). Perhaps lymphomas are better vascularized than solid tumors, and thus undergo less hypoxia during their growth. The spectrum of the lymphomas mirrored that of the Mshl'''  thymi, whose spectrum is undoubtedly  dominated by errors arising during D N A synthesis. Thus, the Mshl''' tumors exhibited a mutator phenotype more like that observed in MSH1''' cells grown under restrictive conditions (Richards et al, 1997). 55  The non-random distributions of mutations, as seen within several of the lacl genes, are unlikely the result of a generalized increase in mutation rate. Indeed, treatment of Mshl''' mice with an alkylating agent led to 5- to 8-fold induction of mutant frequency that were not accompanied by the isolation of multiple mutations within the same gene (38 lacl mutants characterized) (Andrew et al, 1998). Clustered mutations have also been observed following in vitro D N A synthesis across templates with oxidative lesions (Feig and Loeb, 1994), as well as in adenine phosphoribosyl transferase gene mutants in both MMR proficient and deficient colorectal cancer cell lines (Harwood et al, 1991; Meuth, 1996). Interestingly, tumor suppressor genes with multiple mutations were recovered from the colorectal cancers of two patients with H N P C C (Lazar et al, 1994). As suggested in these reports, the presence of clustered mutations is highly suggestive of a focal mutagenic mechanism, such as an error-prone D N A polymerase involved in translesion D N A synthesis across damaged templates or in patch-repair. A n additional mutator mechanism may be at work within the cells of the Mshl''' tumors. The combination of a greatly increased mutant frequency and the absence of any major shift in spectrum (as compared to normal MshT'' thymi) raises the possibility that an alteration in D N A polymerase fidelity may be a possible factor in tumor development. Furthermore, an error-prone repair polymerase could account for the clustering of mutations observed in some of the lacl mutants rescued from the MshT'' lymphomas (Table 3.3). Error-prone polymerases might arise from mutations in polymerase genes (or their associated subunits) (Kunkel et al., 1997), mutations of genes regulating the accuracy of mRNA translation, such as tRNA gene mutations responsible for the mutA and mutC mutator phenotypes (Slupska et al., 1996), or from in vivo microenvironmental abnormalities that generate tumor stress, including inadequate oxygenation or deficiencies of growth factors and nutrients (Loeb, 1997; Richards et ah, 1997). 56  Chapter 4 Tumors arising in D N A mismatch repair-deficient mice show a wide variation in mutant frequency  4.1 Introduction As tumor initiation and progression require a variety of specific genetic changes, at a pace exceeding that provided by the spontaneous mutation rate, it has been hypothesized that mutator phenotypes must be a common feature of neoplasia (Loeb, 1991). This hypothesis has now been supported by the finding of increased rates of genomic instability (for example, instability at repetitive sequences, or chromosomal aberrations) within a great variety of tumor types (Loeb and Christians, 1996). The high frequency of silent, unselected mutations in genes such as p53 also provides evidence of an increased mutation rate of cancer cells (Strauss, 1998). Genomic instability which is commonly exhibited by tumors can be linked to changes in processes that regulate the fidelity of D N A polymerases or the effectiveness of D N A repair pathways (Hoffmann and Cazaux, 1998). Mutations in one of four human D N A mismatch repair (MMR) genes (the mutS ortholog hMSH2 and the mutL orthologs hMLHl,  hPMS2 and  hPMSl),  for example, are commonly found in the autosomal dominant cancer syndrome, hereditary non-polyposis colorectal cancer. This syndrome is characterized not only by early onset colon carcinomas, but also by malignant tumors of the endometrium, stomach, upper urinary tract, small intestine, and ovary (Prolla, 1998). Tumors arising within individuals with this syndrome, who are, in most cases, heterozygous for a germline mutation in a mismatch repair gene, exhibit inactivation of the second allele of the same repair gene (Toft and Arends, 1998). Homozygous germline inactivation of a MMR gene has recently been reported in two families. In these cases,  hMLHl-deficient  children developed hematological malignancies and neurofibromatosis type 1-like features at an early age (Ricciardone et al., 1999; Wang et al., 1999a). Cells deficient in  57  specific MMR components may show not only microsatellite instability, but also exhibit dramatic increases in the frequency of base substitutions in tumor lines (Pro 11a, 1998). MMR-deficient mice generated via gene targeting provide model systems for exploring the consequences of MMR deficiency. In contrast to human H N P C C , mice lacking one allele of the Mshl gene tend not to experience an increased rate of tumor development within their lifetime. Homozygous Mshl''' mice, on the other hand, exhibit an  increased incidence of various types of tumors, including lymphomas  (predominantly of the thymus), small intestinal adenomas and carcinomas, and squamous cell tumors (de Wind et al, 1995; Reitmair et al, 1995; Toft and Arends, 1998). Such model systems clearly demonstrate that M M R deficiency is associated with carcinogenesis; however, it is interesting to note that tumors arise only within a restricted subset of tissues, and with varying latencies, despite of a global lack of MMR activity (de Wind et al, 1995; Reitmair et al, 1995; Toft and Arends, 1998). D N A obtained from the tissues of mice lacking Mshl  and transgenic for the lacl  mutational reporter gene, demonstrate an increase (~10-15-fold) in spontaneous mutant frequency as compared to Mshl heterozygotes, or wild-type mice (Andrew et al., 1997). Interestingly, Mshl'''  murine thymic lymphomas revealed greatly elevated mutational  frequencies compared to normal MshT'' tissues, as well as mutation clusters within lacl genes recovered from these tumors (Baross-Francis et al., 1998). Similar to Msfo2-deficiency, mice having a targeted disruption of Pmsl  also  demonstrate an elevated rate of tumorigenesis, consisting primarily of thymic lymphomas and miscellaneous sarcomas (Prolla et al., 1998). Interestingly, and unlike Mshl  deficient mice, a lack of PMS2 does not appear to lead to intestinal epithelial  tumor formation (Prolla, 1998; Prolla et al, 1998). Also unlike Ms/i2-deficient mice, Pmsl' '' males are sterile due to a defect in chromosome pairing during meiosis (Baker et al., 1995). Such findings demonstrate that MMR genes can play roles in processes that are unrelated to MMR. Similar to Mshl''',  Pmsl' ' 1  58  mice demonstrate elevated mutant  frequencies in all tissues examined, as compared with Pms2 ~, or wild-type mice, as +/  determined using either the supF (Narayanan et ah, 1997), or lacl (Andrew et al, 2000) transgenic reporter systems. To establish whether the lacl gene hypermutation previously observed in Mshl''' lymphomas (Baross-Francis et al, 1998) would also be seen in lymphomas of other MMR-deficient mice, lymphomas of Msh2 ~Pms2~ ~, and Msh2~ ~Pms2~ ~ double mutant +/  /  /  /  mice were examined using the lacl system. The lacl gene mutant frequencies obtained were similar to those reported previously for Mshl ' 1  lymphomas (Baross-Francis et al.,  1998). In contrast, three of four advanced non-thymic tumors arising within Mshl''  mice  failed to show lacl gene mutant frequency elevations, suggesting that hypermutation is not a consistent feature of tumors arising in mismatch repair deficient mice, even within the same genotype.  4.2 Results  4.2.1 Mutant frequency of thymic lymphomas from PMS2 deficient and  Mshl'Pmsl' ' 1  double knock-out mice Mutant frequencies obtained from thymic lymphomas of two mice, and one Pms2' 'MshT '/lacl l  Pms2' 'Msh2' '/lacl /  /  mouse (Table 4.1-III, Figure 4.1) were within the same  1  range as the lacl mutant frequencies we previously reported for six Msh.2' ' thymic 1  lymphomas (Baross-Francis et al., 1998). Thus, the thymic lymphoma mutant frequencies of mouse S {Pms2' 'Msh2 '/lacT), /  +/  and Q and R {Pms2' 'Msh2' '/lacT) /  /  were  elevated 6.0-, 4.1-, and 17.3-fold, respectively, as compared to brain tissue from the corresponding host (Table 4.1-III, Table 4.1-IV). As lymphoma cell infiltrates in the mouse brain are either minimal or absent by histological examination (Baross-Francis et al., 1998), this tissue was selected as a control for 'background' lacl mutant frequencies unique to each animal. Brain mutant frequencies are similar to those of other tissues 59  Table 4.1 Spontaneous lacl mutant frequencies for MSH2 and PMS2 deficient tumors and normal tissues. Spontaneous lacl mutant frequencies for (I) Non-thymic tumors deficient for MSH2 or PMS2 (II) Brains deficient for MSH2 or PMS2 (III) MshT'-Pmsl'- and thymic lymphomas (IV) Mshl' 'Pms2' ' and Msh2 ~Pms2' ~ brains. l  1  +/  60  /  MshT'-Pmsl 1  (I) Non-thymic tumors deficient for MSH2 or PMS2 Genotype Number of Mutant AnimaP mutants/PFU" frequency Mshl Pmsl (*10") +/+ 34/101,110 34 -/L(127F) +/+ 28/158,000 -/18 M (193 F) +/42/180,140 23 -/N (83 F). +/852/146,283 580 -/P (373 M) 5  (II) Brains deficient for MSH2 or PMS2 Animal Genotype Number of mutants/PFU Mshl Pmsl +/+ 39/108,390 -/L(127 F) +/+ 24/119,524 -/M(193F) +/14/32,900 -/N(83F) +/2/14,880 -/P (373 M)  /  5  5  d  (lU) MshT'Pmsl' and MshT Pmsl'- thymic lymphomas Genotype Number of Mutant Animal frequency mutants/PFU (*10") Mshl Pmsl -/-/150 157/102,640 Q (90 M) 491/101,240 480 -/-/R (68 M) +/-/135/113,300 120 S (351 F) 1  Mutant frequency *(100%-clonality)  5  Mean  250±160  (IV) Mshl'Pmsl''and MshT'-Pmsl'' brains Animal Genotype Number of mutants/PFU Pmsl Mshl -/-/1/2,693 Q (90 M) -/5/17,740 -/R (68 M) -/23/112,680 +/S (351 F)  Mutant frequency (*10 ) 37 28 20  Mean  5  28±7  Animal name (age in days, sex) PFU = Plaque Forming Units Clonality based on sequencing data "Mean + standard deviation a  b  c  61  Tumor type  c  Mutant frequency (*1Q-) 36 20 43 13 28±12  Mean  Mutant frequency *(100%-clonality) (*10' ) 34 18 23 180  (  *  1 0  -5)  140 480 90 240±180  Diffuse lymphoma Squamous cell carcinoma Osteogenic sarcoma Diffuse lymphoma  Figure 4.1 lacl mutant frequencies of D N A mismatch repair-deficient tumors and normal brains. Spontaneous lacl mutant frequencies of D N A mismatch repair-deficient tumors (gray bars are based on plating data and black bars show mutant frequencies after correction for clonality), and for corresponding brains (white bars). Each cluster of three bars represent one mouse. (The tumor types are: DL - diffuse lymphoma; SSC - squamous cell carcinoma; OS - osteogenic sarcoma; TL - thymic lymphoma.)  62  in Mshl ' 1  mice, including thymus (Andrew et al., 1997; Baross-Francis et al., 1998). To  estimate the potential effect of clonality on the mutant frequencies of these thymic lymphomas, -10 (9-11) randomly selected mutants were characterized for each tumor (data not shown). Table 4.1-1 and Table 4.1-III show that after correction for clonality mutant frequencies remained elevated. These findings clearly mirror the results obtained in our previous study of Mshl' ' 1  thymic lymphomas, where lacl gene mutant  frequency elevations in the 3.2- to 17.4-fold range were seen (Baross-Francis et al., 1998). Including the three thymic lymphomas in the current study, all nine consecutive thymic lymphomas analyzed to date show elevated  lacl  gene mutant  frequencies  ('hypermutation').  4.2.2 Mutant frequency of MMR deficient non-thymic tumors We next sought to establish whether the lacl gene 'hypermutation' observed in the mismatch repair deficient thymic lymphomas would prove to be a common feature of all tumors in these mice. To evaluate this possibility, lacl mutant frequencies were determined for four non-thymic tumors obtained from Mshl' ' mice (Table 4.1-1, Figure 1  4.1). Interestingly, three out of four non-thymic tumors (L, a -2 cm diffuse extra-thymic 3  lymphoblastic lymphoma, likely of B-cell origin, found on the shoulder; M , an -1.5 cm  3  squamous cell carcinoma arising from the skin at the groin; N , an -1.5 cm osteogenic 3  sarcoma arising from the lower femur on one of the hind legs), failed to show lacl mutant frequency elevations when compared to the corresponding brains of each affected animal (Table 4.1-1, Table 4.1-II, Figure 4.1). Indeed, the frequencies of these three tumors were within the same range as normal Mshl' ' 1  or Pmsl'' tissues (Andrew et  al., 1997; Andrew et al., 2000; Baross-Francis et al., 1998). Tumor P (a diffuse extra-thymic lymphoma, likely of B-cell origin, found on the wall of small intestine), in contrast, exhibited an -45-fold lacl mutant frequency increase compared to brain from the same host (Table 4.1-1, Table 4.1-II, Figure 4.1), suggesting that peripheral lymphomas may be 64  heterogeneous with respect to the 'hypermutator' phenotype. After correction for clonality, the increase seen in tumor P was still ~ 13-fold, a value comparable to that of the thymic lymphomas (Baross-Francis et al, 1998). Thus, elevated lacl  mutant  frequencies are not a feature of every Mshl''' tumor, but instead appear to depend on tumor type, or possibly subtype (in the case of the peripheral lymphomas).  4.2.3 Mutational clusters in single lacl genes are specific to thymic lymphomas Multiple mutations in single lacl genes appeared in the thymic lymphomas of mice Q, R and S with a higher frequency than in normal tissues. Out of 32 mutants sequenced from these tumors, 28 mutations were independent (data not shown), 3 of which contained mutational clusters (Table 4.2). So far only one such cluster was found among lacl mutants from 67 Pmsl' ' and 39 Mshl' ' normal thymi, (published in (Andrew 1  1  et al., 2000; Baross-Francis et al., 1998)). Thus the frequency of multiple mutations is significantly higher in the thymic lymphomas (x -test, p<0.05). Among the 42 2  independent mutants (out of 49 sequenced, data not shown) from the non-thymic tumors, there were no examples of multiple mutations within the lacl gene.  4.3 Discussion Although the numbers of mice belonging to each genotype in this study were relatively small, the results show that deficiencies of Mshl, Pmsl, or both of these mismatch repair genes, lead to similar lacl mutant frequencies in thymic tumors and in normal tissues. In keeping with current models of MSH2 and PMS2 functioning within the D N A mismatch repair pathway (Kolodner and Marsischky, 1999), brain tissue of the double mutant {Pmsl' ' and MshT ') 1  1  mice Q and R, revealed lacl mutant frequencies that were  within the range of those encountered in single-mutant mice (Andrew et al, 1997; Andrew et al, 2000; Baross-Francis et al, 1998).  65  Table 4.2 Multiple mutations occurring within single lacl genes rescued from Mshl' ' 1  Pmsl' ', and Msh2 'Pms2''' thymic lymphomas. 1  +l  Multiple mutations  were  l y m p h o m a s o f Mshl'''Pmsl'''  f o u n d w i t h i n s i n g l e lacl  genes rescued  m i c e Q a n d R, a n d Mshl^'Prnsl' ' 1  66  m o u s e S.  from  thymic  Type of mutation (coding sequence) T->C T->C  Site of mutation  Q  # Mutations per lacl gene 2  R  3  T->C T->C T->C  141 171 183  S  2  T->A T->C  526 780  Tissue  Animal  Mshl'Pmsl'thymic lymphoma  Mshl*''Pmsl'thymic lymphoma  67  54 62  Spontaneous lacl mutant frequency elevations of 5- to 15-fold are typical of MshT ''/lacl mouse tissues as compared to control mice (Andrew et al, 1997), and similar differences are seen when VmsT ' and control mice are compared (Andrew et al., 2000). 1  Interestingly, both MshT '  and VmsT ' thymic lymphomas demonstrated additional  1  striking increases  1  in lacl  mutations, perhaps consistent with some type of  'hypermutator' mechanism (Baross-Francis et al, 1998). In contrast, the non-thymic tumors analyzed, despite being of a size and degree of progression (metastatic deposits in multiple organs) comparable to the lymphomas (data not shown), revealed lacl mutant frequencies that were within the range of those observed in normal D N A mismatch repair-deficient tissues (Andrew et al, 1997; Baross-Francis et al, 1998). What factor(s) is responsible for the dramatic increases in mutant frequency seen in the MMRdeficient lymphomas? One possibility is that the lacl mutation rate within the thymic tumors is similar to that of normal tissues, and that the lacl genes simply accumulate mutations as a function of the number of cell divisions within the tumor. As there is no method for accurately quantifying the number of cell divisions required to generate a normal tissue or a tumor, in vivo mutation rates cannot be determined with any level of confidence. However, higher numbers of cell divisions do not necessarily lead to increased lacl mutations, as an increased lacl mutant frequency was not observed in all the D N A mismatch repair-deficient tumors we analyzed. It is still plausible, however, that the mismatch repair-deficient thymic lymphomas undergo much greater levels of proliferation and apoptosis during their growth than the non-thymic tumors (L, M , N), thus accounting for their elevated mutant frequencies. On the other hand, when lacl mutant frequencies of p53" thymic lymphomas were examined, only one out of eight A  thymic tumors showed an increase over the control tissue (Buettner et al, 1996; Sands et al, 1995). Mitotic counts performed in four of these thymic lymphomas, including one of which showed an increased mutant frequency, revealed no apparent correlation between the mitotic index and lacl mutant frequency (Buettner et al, 1996). This finding 68  supports our contention that cell proliferation may not be the only mechanism responsible for the increased mutant frequencies observed in the nine thymic lymphomas we have analyzed thus far. The finding of multiple, often non-randomly clustered lacl mutations in a number of the phage rescued from the previously examined thymic lymphomas of mismatch repair-deficient mice (Baross-Francis et al., 1998), and in the current study (Table 4.2, tumors Q, R and S), suggests a mutational mechanism beyond that simply afforded by increased numbers of cell divisions. Such a mutator could arise in various ways, such as: alterations in nucleotide pools due to mutations in synthetic pathways or specific nutrient deficiencies, D N A polymerase gene mutations, or defects in other D N A damage recognition and repair components (Hoffmann and Cazaux, 1998). Whether the hypermutation seen in the thymic tumors provides any advantage with respect to tumorigenesis and progression, or whether it merely represents a gratuitous mutational process non-essential for either of these processes remains to be established. It is tempting to speculate, however, that activation of such a mutator mechanism (perhaps stemming from mutation of a specific gene caused by the absence of D N A mismatch repair), as an early event within a developing thymocyte might promote the increased rate of tumorigenesis observed in this tissue.  69  Chapter 5 Elevated mutant frequencies and increased C:G—>T:A transitions in Mlhl' ' 1  versus Pmsl' ' murine small intestinal epithelial cells 1  5.1 Introduction  The DNA mismatch repair (MMR) system, responsible for the recognition and repair of single-base mispairs, small insertions and deletions, which result from replication errors or physical/chemical damage to DNA, and the prevention of homeologous recombination, was initially described in E . coli (Kolodner, 1995; Modrich and Lahue, 1996). In bacteria, initial recognition of mismatches arising during DNA replication is carried out by the MutS protein, which in turn recruits MutL and then MutH, the latter nicking the DNA at the nearest hemimethylated GATC site. The nascent strand is excised by exonuclease activity which is followed by patch repair (Kolodner, 1995; Marra and Schar, 1999). Eukaryotes have a system similar to that of bacteria, and a number of MutS and MutL orthologs have been identified in both yeast and mammals. The orthologs of MutS, MSH2, MSH6, and MSH3 participate in primary recognition, with MSH2 functioning as an obligate partner in two heterodimers: MutSa, composed of MSH2 and MSH6, primarily binds to single base loop and base/base mispairs; and MutS(3, made up of MSH2 and MSH3, recognizes insertion/deletion loops (IDLs) (Jiricny, 1998; Kolodner and Marsischky, 1999; Marra and Schar, 1999). Yeast MutLa, which appears to be involved in the repair of both mismatches and IDLs, is a MLH1 and PMSI heterodimer, whereas mammalian MutLa is composed of MLH1 and PMS2 (PMS2 being the closest mammalian ortholog to yeast PMSI) (Guerrette et al., 1999; Jiricny, 1998; Kolodner and Marsischky, 1999; Marra and Schar, 1999; Prolla et al, 1994). In yeast a second complex containing MLH1 and MLH3 (MLH3 being the closest ortholog of mammalian PMSI), MutLP, also participates in the repair of a subset of frameshift errors (Flores-Rozas and Kolodner, 1998). In humans, the presence of an  70  hMLHl/hPMSl-containing MutLB complex has recently been demonstrated (Raschle et al, 1999), but no biochemical function has been ascribed to this heterodimer thus far. Mutations in one of five different mismatch repair genes, primarily hMSH2 hMLHl,  and to a lesser extent hPMS2, hPMSl  and hMSH6,  and  account for most cases of the  autosomal dominant familial cancer syndrome, hereditary nonpolyposis colorectal cancer (HNPCC) (Kolodner, 1995; Prolla, 1998; Toft and Arends, 1998; Wang et al., 1999b). In most instances affected individuals inherit a germline mutation on one allele and associated with a loss of heterozygosity develop predominantly colon, endometrial, ovarian tumors, as well as malignancies of the stomach, pancreas, small intestine, skin, breast and urinary tract (Prolla, 1998; Toft and Arends, 1998; Wang et al., 1999b). Homozygous germline inactivation of a MMR gene was reported recently in two families. In both cases, h M L H 2-deficient children developed hematological malignancies and neurofibromatosis type 1-like features at an early age (Ricciardone et al., 1999; Wang et al., 1999a). MMR genes can also be functionally inactivated via transcriptional silencing, as shown for hMLHl  in various human cancers (Herman et al.,  1998; Leung et al, 1999; Wang et al., 1999a; Wheeler et al, 1999). Mouse models of MMR deficiency have been developed via gene targeting, thus enabling study of the roles of various MMR genes in mutagenesis and carcinogenesis (Heyer et ah, 1999). Mice with homozygous deficiencies in the MutL orthologs Mlhl and Pmsl generally exhibit similar increases in genome instability and both are prone to malignancy, with lymphomas being the predominant tumor type (Baker et al, 1995; Baker et al, 1996; Narayanan et al, 1997; Prolla et al, 1998). In contrast, an increased predisposition to malignancy has not been evident in Pmsl''' mice (Prolla et al, 1998). Interestingly, while Mlhl'''  mice develop small intestinal tumors, consisting of both  adenomas and adenocarcinomas, Pmsl''' mice do not develop tumors at this site. The discordant tumor spectrum of the Mlhl''  and Pmsl'' mice has suggested a differential  role for these two proteins in tumor formation, likely as a result of differences in 71  genomic instability. In keeping with this, Yao et al. (Yao et al., 1999) recently demonstrated that mice deficient in Mlhl exhibited a greater degree of instability at repetitive sequences than mice lacking Pms2. To enable a quantitative and qualitative assessment of mutations in the mice deficient for the two components of MutLa, MlhV ' and Pms2 ~ mice were crossed with 1  +/  the BC-1 line (Andrew et al., 1996) which carries a transgenic lacl mutational reporter gene, to obtain Mlhl'''/lacl* and Pmsl '/lacl* mice, lacl mutant frequencies (MF) and 1  spectra were determined for purified small intestinal epithelial cell D N A , a site showing differential susceptibility to cancer in Mlhl''' and Pmsl'' mice (Prolla et al., 1998). Mlhl''' mice showed a 1.5-fold increase in MF as compared to Pmsl'' hosts. Sequence analysis of the lacl mutants revealed that this difference was primarily due to an ~2-fold increase in C:G—>T:A changes in the Mlhl''' mouse small intestinal D N A . This increase in point mutations may in part account for the discordance with respect to small intestinal neoplasia in these mice. Furthermore, the results provide additional evidence for differential roles of MLH1 and PMS2 in the mammalian MMR process.  5.2 Results  5.2.1 Mutant frequencies of Mlhl' ' 1  and Pmsl ' 1  small intestinal epithelial cells  lacl mutational frequencies were determined from small intestinal epithelial cell D N A of 5 MlhT''/lacr  and 5 Pmsl 'llacT 1  mice (Table 5.1-1 and -II, Figure 5.1). The Mlhl'  '' epithelial cells demonstrated an ~ 1.5-fold elevation in MF as compared to the Pmsl'' mice (34 x 10"^ versus 23 x 10"^), a significant difference as demonstrated by t-test (p<0.03).  72  Table 5.1 Spontaneous lacl mutant frequencies for small intestinal epithelial cells of Mlhl' ', Pmsl' ' and wild type mice. 1  1  lacl mutant frequencies are shown for small intestinal epithelial cells of (I) Mlhl' ' 1  (II) Pmsl ' 1  mice (III) Wild type littermates of Mlhl' ' 1  Pmsl ' mice. 1  73  mice  mice (IV) Wild type littermates of  Name Ml M2 M3 M4 M5  Age (days) 57 43 36 57 69  Sex M F M F M  Total PFU 226,960 203,660 221,280 207,500 228,320  a  Number of mutants 71 59 67 81 90  PI P2 P3 P4 P5  Age (days) 45 45 45 39 61  Sex F M M F F  Total PFU 208,200 228,240 205,900 217,760 216,160  Number of mutants 57 38 38 40 76  am M6 M7 M8  Age (days) 66 45 68  Sex F F F  Total PFU 203,400 222,140 211,060  Number of mutants 4 5 3  (JV)  P6 P7 P8  Mutant frequency (*10") 27 17 18 18 35 5  Mutant frequency (*10") 2.0 2.2 1.4 5  1.9±0.4  Mean  Name  b  23±8  Mean  Name  5  34±5  Mean  Name  Mutant frequency (*10 ) 31 29 30 39 39  Age (days) 45 60 60  Sex M F M  Total PFU 239,820 232,240 240,020  Number of mutants 4 4 5  Mutant frequency (*10 ) 1.7 1.7 2.1 5  1.8+0.2  Mean PFU = Plaque Forming Units "Mean + standard deviation a  74  Figure 5.1 lacl mutant frequencies for small intestinal epithelial cells of Mlhl' ', Pms2~ 1  '" and wild type mice. Spontaneous lacl mutant frequencies (MF) within small intestinal epithelial cell D N A obtained from Mlhl''' (gray bars) and Pmsl''' (black bars) mice. Mlhl  +/+  lacl M F s obtained from the wild-type littermates of Mlhl,  and PmsT  are  and Pmsl deficient mice,  respectively (open bars). Each bar represents M F from an individual mouse.  75  /+  76  5.2.2 Mutant frequencies of small intestinal epithelial cells from wild type mice To minimize the possibility that this increase was influenced by variations in genetic background, mutational frequencies were also obtained for 3 wild-type littermates of each of the Mlhl  and Pmsl lines (Table 5.1-III and -IV, Figure 5.1). The  MMR deficient mice all showed a >10-fold increase in MF as compared to their wildtype littermates: on average Mlhl' ' 1  mice demonstrated an 18-fold, and Pmsl' ' mice a 131  fold, elevation in MF. The average MF of the wild-type littermates of Mlhl' ' 1  mice were  not significantly different from those of the Pmsl' ' wild-type littermates (1.9 x 10"^, and 1  1.8 x 10"5, respectively; t-test p>0.93). Thus, the observed MF difference of the and Pmsl' ' 1  Mlhl' ' 1  mice was unlikely the result of variations in genetic background.  Furthermore, as the mice analyzed were of a similar ages (1-2 months old, see Table 5.1), it indicated that the Mlhl;Pmsl  genotypes were responsible for the observed MF  differences in these two knockout lines.  5.2.3 Mutational spectra from Mlhl' ' and Pmsl' ' small intestinal epithelial cells 1  1  The mutational spectra of MMR-deficient small intestinal epithelial cells were determined by sequencing of 12-15 lacl mutants per mouse. In total, 74 Mlhl' ' 1  and 71  Pmsl''' lacl mutants were characterized (Table 5.2). To guard against the possibility that recurrent lacl mutants may have been clonal in origin, mutations seen more than once in the same animal were eliminated from the analysis. The resultant corrected mutational spectra are shown in Table 5.2. It should be noted that recurrent lacl mutations may be due to independent mutations occurring at D N A 'hotspots'. Thus, for example, a -1 deletion of an A at positions 135-139 was seen in 9 of the 10 mice, G—>A transitions at positions 56 and 180 were each seen in 4 mice, and C—>T transitions at position 42 of the lacl gene were seen in 4 different mice. Nevertheless, as the percentage of recurrent  77  Table 5.2 Spontaneous lacl mutation spectra from small intestinal epithelial cells of Mlhl' ' and Pmsl' ' mice 1  1  The raw numbers of mutations are shown as a result of sequence analysis. The numbers of independent mutations after eliminating identical mutations that are potentially results of clonal expansion are shown as well.  78  Pms2' ~ mice independent # independent mutations mutations (%) 44 26 15 25 3 5  Transitions CG->TA CpG CG^TA nonCpG TA->CG  42 25 9  Mlhl'' mice # independent mutations 41 24 9  8  8  12  11  8  14  Transversions CG->AT CG->GC AT->CG AT—>TA  10 7 1 0 2  9 6 1 0 2  14 9 2 0 3  11 3 0 4 4  11 3 0 4 4  19 5 0 7 7  Deletions/ Insertions +1 nucleotide -1 nucleotide Ins >1 nucleotide Del >1 nucleotide  22 2 15 1 4  15 2 9 1 3  23 3 14 1 5  21 2 23 0 2  22 2 18 0 2  37 3 j; 0 5  Total number of mutations  74  65  100  71  59  100  Type of mutation # mutations  /  independent mutations (%) 63 37 14  # mutations  79  33 19 3  mutations was relatively low, being 12% in the Mlhl''  and 17% in the Pmsl''  mice, the  mutational spectra were not greatly altered following correction for clonality. Base transitions predominated in lacl mutants isolated from Mlhl' ' 1  and Pmsl'''  mice, followed by insertion/deletion mutations, and transversions (Table 5.2). However, when the numbers of lacl mutants within these various categories are compared, a significant difference between Mlhl' ' 1  and Pmsl' ' small intestinal cells was 1  evident (%2 test, p « 0 . 0 5 for both the original spectra and after correction for clonality). In order to find the type of mutation responsible for this difference, the distribution of Mlhl' ' 1  and Pmsl' '  mutations as a percentage of the total number was compared. As  1  shown in Table 5.2, base transitions accounted for 63% of the Mlhl' ' 1  and 44% of the  Pmsl' ' intestinal cell lacl mutations. Within the transitions, interestingly, C:G—>T:A 1  mutations predominated over T:A—>C:G changes. Amongst C:G—»T:A transitions, mutations were seen at CpG sites and non-CpG sites, with the ratio of mutations at CpG-to-non-CpG  sites being somewhat higher in the Pmsl'''  animals (Table 5.2).  Transversions accounted for the least number of lacl mutations, with the percentages of these being 14% in the Mlhl'''  and 19% in the Pmsl''' animals. Deletions/insertions were  represented by frameshifts of 1, 2, or 4 nucleotides, with deletions predominating (>80%) in both knockout lines. More than 90% of the changes were single nucleotide deletions within mononucleotide repeats of 3-5 nucleotides (data not shown) and these occurred in both the Mlhl'''  and Pmsl''' mice. The number of expansions was uniformly  low in all the MMR deficient mice, and there was no significant difference between the contraction and expansion mutations in the two lines (%2 test, p>0.05). Although the percentage of deletion/insertion mutations was higher in the Pmsl' , than in the Mlhl' ' 1  animals (being 37% and 23%, respectively), a simple comparison of  percentages may not be the best way to compare the mutational spectrum of the mice owing to the finding that the lacl mutational frequency was -1.5-fold higher in the Mlhl' '' than in the Pmsl' ' mice. Thus, for every 100 lacl mutations in Pmsl' ' mice there would 1  1  80  be approximately 150 lacl mutations in Mlhl'''  mice. To allow a more accurate  comparison between the absolute numbers of the types of mutations, the hypothetical distributions of 150 Mlhl'''  and 100 Pmsl''' lacl mutations were calculated based on the  spectra of independent mutations (obtained from Table 5.2). These are presented in Figure 5.2. It was evident that the major difference between Mlhl'''  and Pmsl''' mice was  in the C:G—>T:A transition mutations whose numbers more than doubled in the Mlhl' ' 1  mice (76 vs 30). If these mutations are not included, there is no longer any significant difference between the distributions of the remaining mutations (%2 test, p > 0 . 2 7 » 0 . 0 5 ) . The T:A—>C:G transitions (18 vs 14), the transversions (21 vs 19) and insertion/deletion mutations (34 vs 37), for Mlhl'''  and Pmsl''' mice, respectively, did not change as much  as the C:G—>T:A events. We therefore conclude that the mutational frequency increase seen in the Mlhl'''  animals is primarily the result of an elevated number of C:G—>T:A  transitions.  5.3 Discussion The mammalian MLH1 and PMS2 proteins participate in the mismatch repair pathway as a heterodimer. However, MLH1 and PMS2 deficient mice show a different phenotype regarding their tumor spectrum, specifically, in their predisposition to malignancies of the small intestine (Prolla et al., 1998). To determine whether this might be a consequence of the differential role of the two proteins in D N A MMR, we assessed mutant frequencies and spectra within the small intestinal epithelial cell D N A obtained from Mlhl'''  and Pmsl''' mice employing a lacl transgenic forward mutation detection  system. The lacl gene can be inactivated by a very wide range of mutations, including a great variety of transitions, transversions and frameshifts, and to date, more than 600 sites of inactivating mutations have been described (de Boer and Glickman, 1998). Using this system to assess lacl mutations in Mlhl' ' 1  shown  that  although  and Pmsl' ' small intestinal cells, we have  MLH1/PMS2 81  1  may  both  be  required  for  Figure 5.2 Predicted distributions of the mutational spectra of 150 Mlhl' ' and 100 1  Pmsl' ' lacl mutants obtained from small intestinal epithelial cell D N A . 1  The calculation was based on the spectra of independent mutations shown in Table 5.2.  82  83  the repair of base substitutions and insertions/deletions, MLH1 may be involved in a PMS2-independent repair pathway particularly towards D N A lesions that result in C:G—>T:A transition mutations. Recently, a study of microsatellite instabilities and frameshift mutations in monoand dinucleotide repeats was carried out in Mlhl' ' 1  and Pmsl' ' mice (Yao et al, 1999). 1  This study revealed a significant difference between the expansion and contraction mutations within these repeats, which resulted in elevated frameshift mutant frequencies in Mlhl''' mice (Yao et al., 1999). With the lacl reporter system, however, we did not find a similar difference in expansions and contractions. Although different tissues were assessed in these two studies (colon and skin versus small intestinal epithelial cells in this study), the observed difference is more likely a reflection of the mutational assay system employed. Mutational frequencies and spectra depend not only on the specific reporter genes employed, but also such variables as sequence context, D N A methylation, chromosomal location and transcriptional activity (Dogliotti, 1996). The lacl gene, for example, lacks longer mononucleotide repeats such as those present within the supF reporter gene or microsatellite sequences used by Yao et al (Yao et al., 1999). Indeed, as lacl gene mononucleotide repeats are generally <6 nucleotides in length, it is possible that differences in the ratio of expansions-tocontractions in Mlhl''  and Pmsl'' mice are not observable with the lacl reporter system.  On the other hand, the supF reporter gene employed by Yao et al. is not efficient for assessing single base substitutions, which are more readily accessible using the lacl system. Interestingly, and consistent with our findings, the spectra of mutations in hMlhl  and hPmsl deficient human tumor cell lines using hypoxanthine-guanine  phosphoribosyltransferase (hprt) as the reporter, demonstrated a reduction in C:G->T:A transitions in cells lacking hPMS2, relative to the h M L H l deficient cells (Kato et al., 1998; Ohzeki et al, 1997).  84  The major causes of C:G—»T:A transitions in mammals include the generation of G:T (or C:A) mismatches during D N A replication, recombination events, or deamination of 5-methylC at C p G sites (Marra and Schar, 1999). Non-methylated cytosines can also undergo deamination to uracil, which can potentially lead to C:G—>T:A changes. In Mlhl'''  mice, for example, C:G—>T:A transitions were increased at  non-CpG (over CpG) sites (Table 5.2), suggesting an impaired recognition/repair of G:T mismatches arising as a result of D N A polymerase errors, rather than from deaminations of methylC residues. Thus, as MMR proteins are thought to follow the D N A polymerase replication complexes (Jiricny, 1998; Kolodner and Marsischky, 1999), we speculate that MLH1 and PMS2 proteins may have differential roles in the repair of G:T or C:A mismatches arising as a consequence of D N A polymerase-induced misincorporations. The differences between Mlhl'''  and Pmsl'''  mice with respect to mutant  frequencies and mutation spectra suggest that MLH1 may associate with other MutL orthologs, such as PMSI or MLH3, during MMR. It was recently shown, for example, that h M L H l dimerizes with hPMSl to yield the complex hMutLB (Raschle et al, 1999). Although a specific biochemical function for hMutLB has not yet been demonstrated in vitro (Raschle et al, 1999), it remains possible that this complex participates in some aspect of M M R in vivo. This is consistent with the finding of a hPMSl  mutation in an  HNPCC kindred, which raises the possibility that hPMSl does play an important role in the maintenance of genome stability (Prolla, 1998; Toft and Arends, 1998; Wang et al, 1999b). A n additional mammalian MutL  ortholog, MLH3,  was recently cloned from  humans and mice (Lipkin et al, 2000). Sequence alignments show that hMLH3 is more closely related to yeast MLH3  than is hPMSl.  hMLH3 was shown to interact with  h M L H l , and cells transfected with a dominant negative mutant of hMLH3 protein exhibited microsatellite instability (Lipkin et al, 2000). MLHLPMS2 (MutLa) appears to associate with both MutSa and MutSB during MMR, and it is possible that MLHT.PMS1 85  (MutLB) and/or the M L H 1 / M L H 3 complexes may similarly be capable of interacting with MutScc. Our results suggest that either one or both of the latter two MutL complexes are involved in the repair of some fraction of the D N A lesions that are manifested by the excess of C:G—»T:A transitions in Mlhl' ' 1  mice (as compared to Pmsl' ' 1  mice). Yeast MLH3, which forms a functional heterodimer with MLH1, has been shown to play a role in the repair of frameshifts (Flores-Rozas and Kolodner, 1998). While this complex may also have a role in the repair of mismatches, it is possible that the yeast and mammalian MutL homologs will differ with respect to their specificities, similar to the differences seen when MutS and MutL homologs in E. coli and S. cerevisiae are compared (Yang et al., 1999). It would be of interest to assess lacl gene mutations in mice with combined deficiencies of Pmsl and Pmsl, or Pmsl and Mlh3. This might shed light on which complex is primarily responsible for the correction of G:T mispairs that arise in Pmsl''' small intestinal cells. It would also be of interest to determine whether either of these double mutants develop intestinal tumors with a frequency similar to that of Mlhl''' mice. The elevation of C:G->T:A transitions in the Mlhl'''  mice may facilitate small  intestinal tumorigenesis through the inactivation of tumor suppressors or the activation of proto-oncogenes. Interestingly, a high increase of frameshifts does not seem to be critical to tumorigenesis in MMR-deficient hosts, since older Msh6''' mice still appear to be prone to intestinal tumors despite lacking high the levels of microsatellite instability seen in Mlhl'''  mice (Edelmann et al., 1997). A candidate for inactivation via C:G—>T:A  transition mutations is the adenomatous polyposis coli (Ape) gene, which frequently undergoes loss-of-function mutations in colorectal cancer (Kinzler and Vogelstein, 1996). Ape has been shown to be inactivated by C:G-»T:A transitions (Huang et al., 1996; Lazar et al., 1994; Olschwang et al., 1997). The p53 and K-ras genes are also often mutated in intestinal cancer, with many of the key pro-oncogenic changes in these genes resulting from C:G-»T:A mutations (Lazar et al, 1994; Olschwang et al, 1997). Loss of 86  the tumor supressor gene PTEN,  has also been associated with intestinal tumor  formation in PTEN ~ mice (Di Cristofano et al, 1998; Suzuki et al, 1998), and PTEN +/  in  human tumors can be inactivated by various C:G->T:A transitions (Lynch et al, 1997). These selected examples highlight some of the key growth control genes that might represent targets for the increased level of transition mutations that we see in intestinal DNA.  87  Mlhl' ' 1  Chapter 6 Discussion  6.1 Summary of thesis The thesis objective was to study the evolution of mutational load within D N A mismatch repair-deficient murine tumors, as well as to explore the potential functional differences between the Mlhl and Pmsl MMR genes in small intestinal mutagenesis and tumorigenesis. In Chapter 3, lacl mutational frequency and spectrum from Mshl' ' murine thymic 1  lymphomas, the most common tumors arising in these animals, were presented. These tumors exhibit on average above 8-fold or greater elevations in MF as compared to the MF in normal thymi of Mshl''  mice. The increased MF was in addition to the ~ 15-fold  increase in MF in normal tissues of MMR-deficient animals as compared to the MF in wild type animals. In addition, in the thymic lymphomas lacl genes were recovered that contained multiple mutations at a significantly higher frequency than in normal tissues. In Chaper 4, mutational frequencies and spectra were determined from additional thymic tumors deficient for PMS2, and both PMS2 and MSH2, as well as some non-thymic cancers. In the thymic lymphomas increased mutational frequencies and multiple mutations similar to the Mshl' ' 1  thymic lymphomas were again evident;  however, three out of four non-thymic MMR-deficient tumors did not show an elevated MF at lacl. These results suggest that while hypermutation may not be necessary for induction of tumorigenesis, the increased MF may account for the higher rate of tumor susceptibility in the thymus. Furthermore, the elevated mutant frequencies and multiple mutations in the thymic lymphomas suggest the presence of a lymphomaspecific secondary mutator mechanism. In Chapter 5, it was shown that small intestinal epithelial cells of Mlhl''  mice  demonstrated significantly higher lacl mutant frequencies and increased C:G->T:A transitions as compared to small intestinal epithelial cells of Pmsl'' 88  mice. This finding  may provide an explanation for the differential susceptibility of Mlhl''' and Pmsl''' mice to intestinal tumors. The difference also suggests that other heterodimers containing MLH1, such as MLH1:PMS1 or MLH1:MLH3 may be involved in a PMS2-independent repair of D N A lesions that result in CG—>TA transitions. The overall conclusion from the work presented is that deficiency for various D N A mismatch repair genes leads to increased mutational load, which in turn increases the susceptibility to tumor formation. Within MMR-deficient thymic tumors, components of other systems responsible for maintaining genomic integrity may also be altered, leading to further increases in mutational load and hence, accelerated tumorigenesis.  6.2 D N A mismatch repair-deficiency leads to tumorigenesis via subsequent mutations in key oncogenes and tumor suppressors Loss of D N A mismatch repair leads to cancer formation indirectly. A generalized, genome-wide increase in point mutations includes increased levels of base substitutions and small insertions/deletions that activate proto-oncogenes or inactivate tumor suppressors. Several studies have been aimed at finding mutations in cancerrelated genes in mismatch repair-deficient tumors; however, the list is far from complete. Microsatellite instability, a feature of most MMR-deficient tumors^ can lead to alterations in the length of repetitive mononucleotide runs occurring in the coding regions of cancer-associated genes. These include genes that regulate colonic epithelial cell proliferation, such as adenomatous polyposis coli (APC) and T-cell factor 4 (TCF-4); regulate cell growth such as insulin-like growth factor receptor II (IGF-IIR)  and  transforming growth factor beta receptor II (TGF-RRIT); promote apoptosis such as BAX and caspase-5; and MMR genes hMSH3 and hMSH6 (Duval et al, 1999; Loeb and Loeb, 2000; Olschwang et al, 1997; Prolla, 1998; Schwartz et al, 1999). Other cancer genes, such as the tumor suppressor p53, and the oncogenes K-ras, N-ras and (5-catenin involved in 89  cell-cycle checkpoint and cell proliferation, are primarily mutated by base substitution in MMR deficient tumors (Kinzler and Vogelstein, 1996; Lazar et al, 1994; Miyaki et al, 1999; Olschwang et al, 1997). Mutations in several other genes such as mutated in colorectal cancer (MCC), deleted in colorectal cancer (DCC), and PTEN have been linked to malignancies of the colon; thus, these are also candidates for inactivation by loss of MMR (Heyer et al, 1999). It is expected that many cancer genes, especially tumor suppressors, will be found to have been inactivated via several different types of mutations including base substitutions, insertions and deletions. One good example is the APC gene, in which all these types of mutations have been reported (Huang et al, 1996; Kinzler and Vogelstein, 1996; Lazar et al, 1994; Olschwang et al, 1997). However, many of the cancer genes containing mononucleotide repeats were assessed only for frameshift errors in MMR-deficient tumors; thus based on these findings the importance of base substitutions in their inactivation may have been underestimated. Further, more thorough sequence analysis of these genes will undoubtedly lead to a more complete picture of potential inactivating mutations associated with MMR-deficiency.  6.3 Loss of other roles of M M R proteins may also contribute to carcinogenesis This thesis has focused on the roles of D N A mismatch repair components in maintaining genomic integrity, and how increased rates of mutations appear to be associated with increased susceptibility to tumor formation. Several other studies have suggested additional functions for MMR proteins other than in D N A mismatch repair in various cellular processes that may also be relevant to carcinogenesis. In yeast, mismatch repair proteins are involved in the repair of branched D N A structures. During heteroduplex D N A formation in meiotic recombination, mismatches are generated. A pathway involving the MutS homologs MSH2, MSH3, MSH6, and the MutL homologs MLH1 and PMSI similar to postreplicative MMR appears to repair small mismatches and IDLs. Another pathway requiring MSH2, RAD1 and RADIO 90  repairs larger loops. Two other MutS homologs MSH4 and MSH5, that do not function in postreplicative MMR, and the MutL homologs MLH1 and potentially MLH3 have a later role in recombination in the resolution of crossovers (Kirkpatrick, 1999; Wang et al, 1999c). Consistent with the yeast studies, Mlhl''' and Msh5'' males and females, and male Pmsl''' mice are sterile, confirming the roles of these mammalian genes in meiosis; however, this defect is not apparent in Mshl''', Msh3''' and Msh6''' animals (Baker et al, 1995; Baker et al, 1996; de Vries et al, 1999). MMR proteins also function in the prevention of homeologous recombination by recognizing mismatches between divergent sequences and presumably resolving the recombination complex; however, the exact mechanism remains to be elucidated. Thus, MMR deficiency increases the rates of homeologous recombination in bacteria, and somatic recombination in yeast and mammalian cells (Jiricny, 1998; Kolodner and Marsischky, 1999; Nicholson et al, 2000). This way, inactivation of MMR genes also potentially increases genetic instability in the form of chromosomal rearrangements, mutations which may be highly relevant to tumorigenesis or progression. MutS- or MutL-deficient E. coli are defective in transcription-coupled nucleotide excision repair of ultraviolet photoproducts. Similar observations were found in MMRdeficient human cell lines, which suggests that some MMR genes may play a role in nucleotide excision repair, a pathway that is responsible for correcting a variety of D N A lesions (Umar and Kunkel, 1996). M M R proteins have a role in the recognition of a subset of chemically induced D N A adducts by a mechanism that is not well understood. Thus, mismatch repairdeficient cell lines are highly resistant to several DNA-damaging agents, including alkylating agents and cisplatin, which is commonly used in cancer chemotherapy. MMR-deficient cells also demonstrate increased level of survival following ionizing radiation, suggesting that MMR may play a role in the processing of radiation-induced damaged bases (Fritzell et al, 1997; Marra and Schar, 1999; Prolla, 1998). These D N A 91  lesions may be normally recognized by MMR proteins, such that following abortive attempts at repair may mediate cell cycle arrest and/or cell death of normal cells. A number of recent studies have implicated that MMR proteins promote programmed cell death (Hickman and Samson, 1999; Toft et al, 1999; Wu et al, 1999; Zhang et al, 1999), which means that loss of apoptosis as a result of MMR deficiency may be an important contributing factor to cancer predisposition, as well as resistance to chemotherapeutic agents.  6.4 Potential future directions The finding of remarkable increases of genome instability in MMR-deficient murine thymic lymphomas in Chapters 3 and 4 raises several questions that require further investigation. It has been suggested that an increased rate of cell division is likely to contribute to the mutant frequency elevations in these tumors, but to what extent this is true, remains to be established. Unfortunately, to date there is no method to quantify accurately the number of cell divisions that generate a tissue or tumor in vivo. If such techniques became available, it would be of interest to determine the mutation rates in the various tumors and normal tissues. If increased mutation rate is required for tumorigenesis, it is likely to be an early event in tumor progression. The tumors examined in Chapters 3 and 4 in this thesis were all in late stages, as the animals were already moribound. It would be interesting to examine the changes in mutant frequency during different stages of tumor development to find out how it correlates with tumor growth and how early the secondary MF increases in the thymic lymphomas took place. To date there is no satisfying explanation for the MMR-deficient tumor spectra in humans or mice. The current model of eukaryotic MMR is mostly based on yeast studies, and the mammalian orthologs of known MMR genes are very similar to the yeast genes. However, even though the interacting enzymes and pathways seem to be 92  similar, there may be differences in the tissue-specific roles of mammalian MMR proteins or yet unknown other genes involved in the pathway, which may account for the differential susceptibility of various tissues to tumor formation. Examination of potential differential expression of MMR genes in various tissues and detailed mutational studies may provide some answers to these quesions. The presence of a lymphoma-specific secondary mutator mechanism in MSH2 and PMS2 deficient mice was also suggested as a possible explanation for the M F increases compared to normal tissues. There are several possibilities for the secondary mutator as there are several other genes involved in the maintenance of genome stability that could be inactivated due to MMR deficiency. Deregulation of nucleotide synthesis can lead to an imbalance of dNTP pools, which can affect the fidelity of D N A replication and be a source of mutagenesis (Hoffmann and Cazaux, 1998). Measuring the amounts of dNTPs or anabolic enzymes involved in nucleotide synthesis in the MMR- tissues and tumors may enable evaluation of this possibility. Altered expression or mutation of D N A polymerases could also lead to increased error rates during replication or D N A repair. There are great variations in the fidelity of mammalian D N A polymerases,  and there is increasing evidence  for their  interchangeability during cellular processes. Overexpression of error-prone polymerases, found in some human cancers, such as D N A polymerase B or D N A polymerase  for example, may lead to increased mutagenesis. Mutations in D N A  polymerases B and 8 that may render them error-prone have also been found in human tumors (Hoffmann and Cazaux, 1998; Loeb and Loeb, 2000). In the MMR-deficient murine tumors analysis of the expression level of D N A polymerases and sequence analysis of the polymerase genes may provide an answer whether these were involved in generating the high mutant frequencies in the thymic lymphomas. There are ongoing experiments in the laboratory to overexpress wild-type and dominant negative mutant D N A polymerases in transgenic mice. It will be interesting to see if they lead to a cancer 93  predisposition phenotype and whether the tumors will exhibit elevated MF and clustered mutations similar to the MMR deficient thymic tumors. Analysis of mutations in mice deficient for MLH1 and PMS2 in Chapter 4 led to a proposal that the mammalian PMSI and/or MLH3 that have recently been shown to interact with MLH1 (Lipkin et al., 2000; Raschle et al., 1999) may also participate in the mammalian M M R pathway in MutL-like heterodimers. To test this hypothesis, mutational studies need to be done using mice deficient in PMSI and MLH3. If there is partial redundancy between PMS2 and PMSI or MLH3, then mice deficient for PMS2 and PMSI or PMS2 and MLH3 should exhibit a phenotype similar to MLH1 deficient animals.  94  References  Andrew, S.E., McKinnon, M., Cheng, B.S., Francis, A., Penney, J., Reitmair, A . H . , Mak, T.W. and Jirik, F.R. (1998) Tissues of MSH2-deficient mice demonstrate hypermutability on exposure to a D N A methylating agent. Proc Natl Acad Sci USA, 95,1126-1130.  Andrew, S.E., Pownall, S., Fox, J., Hsiao, L., Hambleton, J., Penney, J.E., Kohler, S.W. and Jirik, F.R. (1996) A novel lacl transgenic mutation-detection system and its application to establish baseline mutation frequencies in the scid mouse. Mutat Res, 357, 57-66.  Andrew, S.E., Reitmair, A.H., Fox, J., Hsiao, L., Francis, A., McKinnon, M . , Mak, T.W. and Jirik, F.R. (1997) Base transitions dominate the mutational spectrum of a transgenic reporter gene in MSH2 deficient mice. Oncogene, 15,123-129.  Andrew, S.E., Xiaoxin, S.X., Baross-Francis, A., Narayanan, L., Milhausen, K., Liskay, R.M., Jirik, F.R. and Glazer, P.M. (2000) Mutagenesis in PMS2- and MSH2-deficient mice indicates differential protection from transversions and frameshifts. Carcinogenesis, 21, 1291-1295.  Baker, S.M., Bronner, C.E., Zhang, L., Plug, A.W., Robatzek, M . , Warren, G., Elliott, E.A., Yu, J., Ashley, T., Arnheim, N . and et al. (1995) Male mice defective in the D N A mismatch repair gene PMS2 exhibit abnormal chromosome synapsis in meiosis. Cell, 82, 309-319.  Baker, S.M., Plug, A.W., Prolla, T.A., Bronner, C.E., Harris, A . C , Yao, X., Christie, D.M., Monell, C , Arnheim, N., Bradley, A., Ashley, T. and Liskay, R.M. (1996) Involvement of 95  mouse Mlhl in D N A mismatch repair and meiotic crossing over [see comments]. Nat Genet, 13,336-342.  Ban, C , Junop, M . and Yang, W. (1999) Transformation of MutL by ATP binding and hydrolysis: a switch in D N A mismatch repair. Cell, 97,85-97.  Baross-Francis, A., Andrew, S.E., Penney, J.E. and Jirik, F.R. (1998) Tumors of D N A mismatch repair-deficient hosts exhibit dramatic increases in genomic instability. Proc Natl Acad Sci USA, 95,8739-8743.  Baross-Francis, A., Milhausen, M.K., Andrew, S.E., Jevon, G. and Jirik, F.R. (2000) Tumors arising in D N A mismatch repair-deficient mice show a wide variation in mutation frequency as assessed by a transgenic reporter gene. Carcinogenesis, 21, 12591262.  Bellacosa, A . , Cicchillitti, L . , Schepis, F., Riccio, A . , Yeung, A.T., Matsumoto, Y., Golemis, E.A., Genuardi, M . and Neri, G. (1999) MED1, a novel human methyl-CpGbinding endonuclease, interacts with D N A mismatch repair protein M L H l . Proc Natl Acad Sci USA, 96,3969-3974.  Bhattacharyya, N.P., Ganesh, A., Phear, G., Richards, B., Skandalis, A . and Meuth, M . (1995) Molecular analysis of mutations in mutator colorectal carcinoma cell lines. Hum Mol Genet, 4,2057-2064.  Bhattacharyya, N.P., Skandalis, A., Ganesh, A . , Groden, J. and Meuth, M . (1994) Mutator phenotypes in human colorectal carcinoma cell lines. Proc Natl Acad Sci 91, 6319-6323. 96  USA,  The Big Blue Website. 11 October, 1999, revision date. Maintained by J. de Boer at the Center for Environmental Health of the University of Victoria, BC. [Online] http://eden.ceh.uvic.ca [June 27,2000, last date accessed.]  Buermeyer, A.B., Deschenes, S.M., Baker, S.M. and Liskay, R.M. (1999) Mammalian D N A mismatch repair. Annu Rev Genet, 33,533-564.  Buettner, V.L., Hill, K.A., Nishino, H., Schaid, D.J., Frisk, C S . and Sommer, S.S. (1996) Increased mutation frequency and altered spectrum in one of four thymic lymphomas derived from tumor prone p53/Big Blue double transgenic mice. Oncogene, 1 3 , 24072413.  Culligan, K.M., Meyer-Gauen, G., Lyons-Weiler, J. and Hays, J.B. (2000) Evolutionary origin, diversification and specialization of eukaryotic MutS homolog mismatch repair proteins [In Process Citation]. Nucleic Acids Res, 28,463-471.  Curia, M.C., Palmirotta, R., Aceto, G., Messerini, L., Veri, M.C., Crognale, S., Valanzano, R., Ficari, F., Fracasso, P., Stigliano, V., Tonelli, F., Casale, V., Guadagni, F., Battista, P., Mariani-Costantini, R. and Cama, A. (1999) Unbalanced germ-line expression of h M L H l and hMSH2 alleles in hereditary nonpolyposis colorectal cancer. Cancer Res, 59, 3570-3575.  de Boer, J.G. and Glickman, B.W. (1998) The lacl gene as a target for mutation in transgenic rodents and Escherichia coli. Genetics, 148,1441-1451.  97  de Vries, S.S., Baart, E.B., Dekker, M., Siezen, A., de Rooij, D.G., de Boer, P. and te Riele, H. (1999) Mouse MutS-like protein Msh5 is required for proper chromosome synapsis in male and female meiosis. Genes Dev, 13,523-531.  de Wind, N . , Dekker, M . , Berns, A., Radman, M . and te Riele, H . (1995) Inactivation of the mouse Msh2 gene results in mismatch repair deficiency, methylation tolerance, hyperrecombination, and predisposition to cancer. Cell, 82,321-330.  Di Cristofano, A., Pesce, B., Cordon-Cardo, C. and Pandolfi, P.P. (1998) Pten is essential for embryonic development and tumour suppression. Nat Genet, 19,348-355.  Dogliotti, E. (1996) Mutational spectra: from model systems to cancer-related genes. Carcinogenesis, 17,2113-2118.  Dunlop, M.G. (1996) Mutator genes and mosaicism in colorectal cancer. Curr Opin Genet Dev, 6, 76-81.  Duval, A., Gayet, J., Zhou, X.P., Iacopetta, B., Thomas, G. and Hamelin, R. (1999) Frequent frameshift mutations of the TCF-4 gene in colorectal cancers with microsatellite instability. Cancer Res, 59,4213-4215.  Edelmann, W., Cohen, P.E., Kane, M . , Lau, K., Morrow, B., Bennett, S., Umar, A., Kunkel, T., Cattoretti, G., Chaganti, R., Pollard, J.W., Kolodner, R.D. and Kucherlapati, R. (1996) Meiotic pachytene arrest in MLHl-deficient mice. Cell, 85,1125-1134.  Edelmann, W., Umar, A., Yang, K., Heyer, J., Kucherlapati, M . , Lia, M . , Kneitz, B., Avdievich, E., Fan, K., Wong, E., Crouse, G., Kunkel, T., Lipkin, M . , Kolodner, R.D. and 98  Kucherlapati, R. (2000) The D N A mismatch repair genes Msh3 and Msh6 cooperate in intestinal tumor suppression [In Process Citation]. Cancer Res, 60, 803-807.  Edelmann, W., Yang, K., Umar, A., Heyer, J., Lau, K., Fan, K., Liedtke, W., Cohen, P.E., Kane, M.F., Lipford, J.R., Yu, N., Crouse, G.F., Pollard, J.W., Kunkel, T., Lipkin, M., Kolodner, R. and Kucherlapati, R. (1997) Mutation in the mismatch repair gene Msh6 causes cancer susceptibility. Cell, 91,467-477.  Eshleman, J.R., Lang, E.Z., Bowerfind, G.K., Parsons, R., Vogelstein, B., Willson, J.K., Veigl, M.L., Sedwick, W.D. and Markowitz, S.D. (1995) Increased mutation rate at the hprt locus accompanies microsatellite instability in colon cancer. Oncogene, 10, 33-37.  Eshleman, J.R., Markowitz, S.D., Donover, P.S., Lang, E.Z., Lutterbaugh, J.D., Li, G.M., Longley, M . , Modrich, P., Veigl, M . L . and Sedwick, W . D . (1996) Diverse hypermutability of multiple expressed sequence motifs present in a cancer with microsatellite instability. Oncogene, 12,1425-1432.  Feig, D.I. and Loeb, L . A . (1994) Oxygen radical induced mutagenesis is D N A polymerase specific. / Mol Biol, 235,33-41.  Fishel, R. (1998) Mismatch repair, molecular switches, and signal transduction. Genes Dev, 12,2096-2101.  Flores-Rozas, H . and Kolodner, R.D. (1998) The Saccharomyces cerevisiae MLH3 gene functions in MSH3-dependent suppression of frameshift mutations. Proc Natl Acad Sci U S A, 95,12404-12409.  99  Fritzell, J.A., Narayanan, L., Baker, S.M., Bronner, C.E., Andrew, S.E., Prolla, T.A., Bradley, A., Jirik, F.R., Liskay, R.M. and Glazer, P.M. (1997) Role of D N A mismatch repair in the cytotoxicity of ionizing radiation. Cancer Res, 57, 5143-5147.  Glaab, W.E. and Tindall, K.R. (1997) Mutation rate at the hprt locus in human cancer cell lines with specific mismatch repair-gene defects. Carcinogenesis, 18,1-8.  Glazer, P.M., Sarkar, S.N. and Summers, W.C. (1986) Detection and analysis of U V induced mutations in mammalian cell D N A using a lambda phage shuttle vector. Proc Natl Acad Sci USA, 83,1041-1044.  Gu, M . , Ahmed, A., Wei, C., Gorelick, N . and Glickman, B.W. (1994) Development of a lambda-based complementation assay for the preliminary localization of lacl mutants from the Big Blue mouse: implications for a DNA-sequencing strategy. Mutat Res, 307, 533-540.  Guerrette, S., Acharya, S. and Fishel, R. (1999) The interaction of the human MutL homologues in hereditary nonpolyposis colon cancer. / Biol Chem, 274,6336-6341.  Harwood, J., Tachibana, A . and Meuth, M . (1991) Multiple dispersed spontaneous mutations: a novel pathway of mutation in a malignant human cell line. Mol Cell Biol, 11,3163-3170.  Herman, J.G., Umar, A., Polyak, K., Graff, J.R., Ahuja, N . , Issa, J.P., Markowitz, S., Willson, J.K., Hamilton, S.R., Kinzler, K.W., Kane, M.F., Kolodner, R.D., Vogelstein, B., Kunkel, T.A. and Baylin, S.B. (1998) Incidence and functional consequences of h M L H l  100  promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci USA, 9 5 , 68706875.  Heyer, J., Yang, K., Lipkin, M . , Edelmann, W. and Kucherlapati, R. (1999) Mouse models for colorectal cancer. Oncogene, 18,5325-5333.  Hickman, M.J. and Samson, L.D. (1999) Role of D N A mismatch repair and p53 in signaling induction of apoptosis by alkylating agents. Proc Natl Acad Sci USA,  96,  10764-10769.  Hoffmann, J.S. and Cazaux, C. (1998) D N A synthesis, mismatch repair and cancer (Review). Int J Oncol, 12,377-382.  Huang, J., Papadopoulos, N., McKinley, A.J., Farrington, S.M., Curtis, L.J., Wyllie, A . H . , Zheng, S., Willson, J.K., Markowitz, S.D., Morin, P., Kinzler, K.W., Vogelstein, B. and Dunlop, M . G . (1996) A P C mutations in colorectal tumors with mismatch repair deficiency. Proc Natl Acad Sci USA,  93,9049-9054.  Jakubczak, J.L., Merlino, G., French, J.E., Muller, W.J., Paul, B., Adhya, S. and Garges, S. (1996) Analysis of genetic instability during mammary tumor progression using a novel selection-based assay for in vivo mutations in a bacteriophage lambda transgene target. Proc Natl Acad Sci USA,  93,9073-9078.  Jiricny, J. (1998) Eukaryotic mismatch repair: an update. Mutat Res, 409,107-121.  101  Johnson, R.E., Kovvali, G.K., Prakash, L. and Prakash, S. (1996) Requirement of the yeast MSH3 and MSH6 genes for MSH2-dependent genomic stability. / Biol Chem, 271, 7285-7288.  Kat, A., Thilly, W.G., Fang, W.H., Longley, M.J., Li, G.M. and Modrich, P. (1993) A n alkylation-tolerant, mutator human cell line is deficient in strand- specific mismatch repair. Proc Natl Acad Sci USA, 90, 6424-6428.  Kato, T., Yatagai, F., Glickman, B.W., Tachibana, A. and Ikenaga, M . (1998) Specificity of mutations in the PMS2-deficient human tumor cell line H E C - l - A . Mutat Res, 422, 279283.  Kinzler, K.W. and Vogelstein, B. (1996) Lessons from hereditary colorectal cancer. Cell, 87,159-170.  Kinzler, K.W. and Vogelstein, B. (1997) Cancer-susceptibility genes. Gatekeepers and caretakers [news; comment]. Nature, 386,761, 763.  Kirkpatrick, D.T. (1999) Roles of the D N A mismatch repair and nucleotide excision repair proteins during meiosis. Cell Mol Life Sci, 55,437-449.  Kohler, S.W., Provost, G.S., Fieck, A., Kretz, P.L., Bullock, W.O., Sorge, J.A., Putman, D.L. and Short, J.M. (1991) Spectra of spontaneous and mutagen-induced mutations in the lacl gene in transgenic mice. Proc Natl Acad Sci USA, 88, 7958-7962.  102  Kohler, S.W., Provost, G.S., Kretz, P.L., Fieck, A., Sorge, J.A. and Short, J.M. (1990) The use of transgenic mice for short-term, in vivo mutagenicity testing. Genet Anal Tech Appl, 7,212-218..  Kolodner, R. (1996) Biochemistry and genetics of eukaryotic mismatch repair. Genes Dev, 10,1433-1442.  Kolodner, R.D. (1995) Mismatch repair: mechanisms and relationship to cancer susceptibility. Trends Biochem Sci, 20,397-401.  Kolodner, R.D. and Marsischky, G.T. (1999) Eukaryotic D N A mismatch repair. Curr Opin Genet Dev, 9,89-96.  Kunkel, T.A., Resnick, M.A. and Gordenin, D.A. (1997) Mutator specificity and disease: looking over the FENce [comment]. Cell, 88,155-158.  Lazar, V., Grandjouan, S., Bognel, C., Couturier, D., Rougier, P., Bellet, D. and Bressacde Paillerets, B. (1994) Accumulation of multiple mutations in tumour suppressor genes during colorectal tumorigenesis in F1NPCC patients. Hum Mol Genet, 3,2257-2260.  Leung, S.Y., Yuen, S.T., Chung, L.P., Chu, K.M., Chan, A.S. and Ho, J.C. (1999) h M L H l promoter methylation and lack of h M L H l expression in sporadic gastric carcinomas with high-frequency microsatellite instability. Cancer Res, 59,159-164.  Lindahl, T., Karran, P. and Wood, R.D. (1997) D N A excision repair pathways. Curr Opin Genet Dev, 7,158-169.  103  Lipkin, S.M., Wang, V., Jacoby, R., Banerjee-Basu, S., Baxevanis, A.D., Lynch, H.T., Elliott, R.M. and Collins, F.S. (2000) MLH3: a D N A mismatch repair gene associated with mammalian microsatellite instability. Nat Genet, 24,27-35.  Loeb, K.R. and Loeb, L . A . (2000) Significance of multiple mutations in cancer. Carcinogenesis, 21,379-385.  Loeb, L.A. (1991) Mutator phenotype may be required for multistage carcinogenesis. Cancer Res, 51,3075-3079.  Loeb, L . A . (1997) Transient expression of a mutator phenotype in cancer cells [comment]. Science, 277,1449-1450.  Loeb, L.A. and Christians, F.C. (1996) Multiple mutations in human cancers. Mutat Res, 350,279-286.  Lynch, E.D., Ostermeyer, E.A., Lee, M.K., Arena, J.F., Ji, H . , Dann, J., Swisshelm, K., Suchard, D., MacLeod, P.M., Kvinnsland, S., Gjertsen, B.T., Heimdal, K., Lubs, H . , Moller, P. and King, M.C. (1997) Inherited mutations in PTEN that are associated with breast cancer, cowden disease, and juvenile polyposis. Am } Hum Genet, 61,1254-1260.  Malkhosyan, S., McCarty, A., Sawai, H . and Perucho, M . (1996) Differences in the spectrum of spontaneous mutations in the hprt gene between tumor cells of the microsatellite mutator phenotype. Mutat Res, 316,249-259.  Marra, G. and Schar, P. (1999) Recognition of D N A alterations by the mismatch repair system. Biochem J, 338,1-13. 104  Marsischky, G.T., Filosi, N . , Kane, M.F. and Kolodner, R. (1996) Redundancy of Saccharomyces cerevisiae MSH3 and MSH6 in MSH2-dependent mismatch repair. Genes Dev, 10,407-420.  Meuth, M . (1996) Patterns of mutation in cancer cells. Cancer Surv, 28,33-46.  Mirsalis, J . C , Monforte, J.A. and Winegar, R.A. (1995) Transgenic animal models for detection of in vivo mutations. Annu Rev Pharmacol Toxicol, 35,145-164.  Miyaki, M., Iijima, T., Kimura, J., Yasuno, M., Mori, T., Hayashi, Y., Koike, M . , Shitara, N., Iwama, T. and Kuroki, T. (1999) Frequent mutation of beta-catenin and APC genes in primary colorectal tumors from patients with hereditary nonpolyposis colorectal cancer. Cancer Res, 59,4506-4509.  Modrich, P. and Lahue, R. (1996) Mismatch repair in replication fidelity, genetic recombination, and cancer biology. Annu Rev Biochem, 65,101-133.  Narayanan, L., Fritzell, J.A., Baker, S.M., Liskay, R.M. and Glazer, P.M. (1997) Elevated levels of mutation in multiple tissues of mice deficient in the D N A mismatch repair gene Pms2. Proc Natl Acad Sci USA, 94,3122-3127.  Nicholson, A., Hendrix, M., Jinks-Robertson, S. and Crouse, G.F. (2000) Regulation of mitotic homeologous recombination in yeast. Functions of mismatch repair and nucleotide excision repair genes. Genetics, 154,133-146.  105  Ohzeki, S., Tachibana, A., Tatsumi, K. and Kato, T. (1997) Spectra of spontaneous mutations at the hprt locus in colorectal carcinoma cell lines defective in mismatch repair. Carcinogenesis, 18,1127-1133.  Olschwang, S., Hamelin, R., Laurent-Puig, P., Thuille, B., De Rycke, Y., Li, Y.J., Muzeau, F., Girodet, J., Salmon, R.J. and Thomas, G. (1997) Alternative genetic pathways in colorectal carcinogenesis. Proc Natl Acad Sci USA, 94,12122-12127.  Phear, G., Bhattacharyya, N.P. and Meuth, M . (1996) Loss of heterozygosity and base substitution at the APRT locus in mismatch-repair-proficient and -deficient colorectal carcinoma cell lines. Mol Cell Biol, 16,6516-6523.  Prolla, T.A. (1998) D N A mismatch repair and cancer. Curr Opin Cell Biol, 10,311-316.  Prolla, T.A., Abuin, A. and Bradley, A. (1996) D N A mismatch repair deficient mice in cancer research. Semin Cancer Biol, 7,241-247.  Prolla, T.A., Baker, S.M., Harris, A.C., Tsao, J.L., Yao, X., Bronner, C.E., Zheng, B., Gordon, M . , Reneker, J., Arnheim, N., Shibata, D., Bradley, A . and Liskay, R.M. (1998) Tumour susceptibility and spontaneous mutation in mice deficient in M l h l , Pmsl and Pms2 D N A mismatch repair. Nat Genet, 18,276-279.  Prolla, T.A., Pang, Q., Alani, E., Kolodner, R.D. and Liskay, R.M. (1994) MLH1, PMSI, and MSH2 interactions during the initiation of D N A mismatch repair in yeast. Science, 265,1091-1093.  106  Rampino, N., Yamamoto, H., Ionov, Y., Li, Y., Sawai, H., Reed, J.C. and Perucho, M . (1997) Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science, 2 7 5 , 967-969.  Raschle, M., Marra, G., Nystrom-Lahti, M., Schar, P. and Jiricny, J. (1999) Identification of hMutLbeta, a heterodimer of h M L H l and hPMSl. / Biol Chem, 274,32368-32375.  Reitmair, A . H . , Cai, J . C , Bjerknes, M., Redston, M., Cheng, H . , Pind, M.T., Hay, K., Mitri, A., Bapat, B.V., Mak, T.W. and Gallinger, S. (1996a) MSH2 deficiency contributes to accelerated APC-mediated intestinal tumorigenesis. Cancer Res, 56,2922-2926.  Reitmair, A . H . , Redston, M., Cai, J.C, Chuang, T . C , Bjerknes, M., Cheng, H . , Hay, K., Gallinger, S., Bapat, B. and Mak, T.W. (1996b) Spontaneous intestinal carcinomas and skin neoplasms in Msh2-deficient mice. Cancer Res, 5 6 , 3842-3849.  Reitmair, A . H . , Schmits, R., Ewel, A., Bapat, B., Redston, M., Mitri, A., Waterhouse, P., Mittrucker, H.W., Wakeham, A., Liu, B. and et al. (1995) MSH2 deficient mice are viable and susceptible to lymphoid tumours. Nat Genet, 11, 64-70.  Reynolds, T.Y., Rockwell, S. and Glazer, P.M. (1996) Genetic instability induced by the tumor microenvironment. Cancer Res, 56,5754-5757.  Ricciardone, M . D . , Ozcelik, T., Cevher, B., Ozdag, H . , Tuncer, M . , Gurgey, A., Uzunalimoglu, O., Cetinkaya, H., Tanyeli, A., Erken, E. and Ozturk, M . (1999) Human MLH1 deficiency predisposes to hematological malignancy and neurofibromatosis type 1. Cancer Res, 59,290-293.  107  Richards, B., Zhang, H . , Phear, G. and Meuth, M . (1997) Conditional mutator phenotypes in hMSH2-deficient tumor cell lines [see comments]. Science, 277,1523-1526.  Sands, A.T., Suraokar, M.B., Sanchez, A., Marth, J.E., Donehower, L.A. and Bradley, A. (1995) p53 deficiency does not affect the accumulation of point mutations in a transgene target. Proc Natl Acad Sci USA, 92,8517-8521.  Schmutte, C. and Fishel, R. (1999) Genomic instability: first step to carcinogenesis. Anticancer Res, 19,4665-4696.  Schwartz, S., Jr., Yamamoto, H., Navarro, M., Maestro, M., Reventos, J. and Perucho, M . (1999) Frameshift mutations at mononucleotide repeats in caspase-5 and other target genes in endometrial and gastrointestinal cancer of the microsatellite mutator phenotype. Cancer Res, 59,2995-3002.  Shibata, D., Navidi, W., Salovaara, R., Li, Z . H . and Aaltonen, L.A. (1996) Somatic microsatellite mutations as molecular tumor clocks. Nat Med, 2, 676-681.  Simpkins, S.B., Bocker, T., Swisher, E.M., Mutch, D.G., Gersell, D.J., Kovatich, A.J., Palazzo, J.P., Fishel, R. and Goodfellow, P.J. (1999) MLH1 promoter methylation and gene silencing is the primary cause of microsatellite instability in sporadic endometrial cancers. Hum Mol Genet, 8, 661-666.  Slupska, M . M . , Baikalov, C , Lloyd, R. and Miller, J.H. (1996) Mutator tRNAs are encoded by the Escherichia coli mutator genes mutA and mutC: a novel pathway for mutagenesis. Proc Natl Acad Sci USA, 93, 4380-4385.  108  Strauss, B.S. (1998) Hypermutability in carcinogenesis. Genetics, 148,1619-1626.  Suzuki, A., de la Pompa, J.L., Stambolic, V., Elia, A.J., Sasaki, T., del Barco Barrantes, I., Ho, A., Wakeham, A., Itie, A., Khoo, W., Fukumoto, M . and Mak, T.W. (1998) High cancer susceptibility and embryonic lethality associated with mutation of the PTEN tumor suppressor gene in mice. Curr Biol, 8,1169-1178.  Toft, N.J. and Arends, M.J. (1998) D N A mismatch repair and colorectal cancer. / Pathol, 185, 123-129.  Toft, N.J., Winton, D.J., Kelly, J., Howard, L.A., Dekker, M., te Riele, H., Arends, M.J., Wyllie, A . H . , Margison, G.P. and Clarke, A.R. (1999) Msh2 status modulates both apoptosis and mutation frequency in the murine small intestine. Proc Natl Acad Sci U S A, 96, 3911-3915.  Umar, A . and Kunkel, T.A. (1996) DNA-replication fidelity, mismatch repair and genome instability in cancer cells. Eur } Biochem, 238,297-307.  Umar, A., Risinger, J.L, Glaab, W.E., Tindall, K.R., Barrett, J.C. and Kunkel, T.A. (1998) Functional overlap in mismatch repair by human MSH3 and MSH6. Genetics, 148,16371646.  Wang, Q., Lasset, C , Desseigne, F., Frappaz, D., Bergeron, C , Navarro, C , Ruano, E. and Puisieux, A . (1999a) Neurofibromatosis and early onset of cancers in h M L H l deficient children. Cancer Res, 59, 294-297.  109  Wang, Q., Lasset, C , Desseigne, F., Saurin, J . C , Maugard, C , Navarro, C , Ruano, E., Descos, L., Trillet-Lenoir, V., Bosset, J.F. and Puisieux, A. (1999b) Prevalence of germline mutations of h M L H l , hMSH2, hPMSl, hPMS2, and hMSH6 genes in 75 French kindreds with nonpolyposis colorectal cancer. Hum Genet, 105, 79-85.  Wang, T.F., Kleckner, N . and Hunter, N . (1999c) Functional specificity of MutL homologs in yeast: evidence for three Mlhl-based heterocomplexes with distinct roles during meiosis in recombination and mismatch correction. Proc Natl Acad Sci USA, 96, 13914-13919.  Weber, J.L. and May, P.E. (1989) Abundant class of human D N A polymorphisms which can be typed using the polymerase chain reaction. Am } Hum Genet, 44,388-396.  Wheeler, J.M., Beck, N.E., Kim, H . C , Tomlinson, LP., Mortensen, N.J. and Bodmer, W.F. (1999) Mechanisms of inactivation of mismatch repair genes in human colorectal cancer cell lines: the predominant role of h M L H l . Proc Natl Acad Sci USA, 96,10296-10301.  Wu, J., G u , L . , Wang, H . , Geacintov, N.E. and L i , G.M. (1999) Mismatch repair processing of carcinogen-DNA adducts triggers apoptosis. Mol Cell Biol, 19, 8292-8301.  Yang, Y., Karthikeyan, R., Mack, S.E., Vonarx, E.J. and Kunz, B.A. (1999) Analysis of yeast pmsl, msh2, and mlhl mutators points to differences in mismatch correction efficiencies between prokaryotic and eukaryotic cells. Mol Gen Genet, 261, 777-787.  Yao, X., Buermeyer, A.B., Narayanan, L., Tran, D., Baker, S.M., Prolla, T.A., Glazer, P.M., Liskay, R.M. and Arnheim, N . (1999) Different mutator phenotypes in M l h l versus Pms2-deficient mice. Proc Natl Acad Sci USA, 96, 6850-6855. 110  Zhang, H., Richards, B., Wilson, T., Lloyd, M . , Cranston, A., Thorburn, A., Fishel, R. and Meuth, M . (1999) Apoptosis induced by overexpression of hMSH2 or h M L H l . Cancer Res, 59,3021-3027.  Ill  

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.831.1-0089720/manifest

Comment

Related Items