Molecular Basis

In 1993, three independent groups reported a novel type of genomic instability in a subset of colorectal tumors.20-22 This instability was termed microsatellite instability (MSI) and was characterized by expansions or contractions in the number of tandem repeats at numerous microsatellite loci in the DNA. In these studies, MSI was identified in both a subset of sporadic colon tumors and nearly all tumors derived from patients diagnosed with HNPCC. Micro-

satellites are tandem repeats of a simple DNA sequence (1-6 base pairs [bp] in length) that are abundant and randomly distributed throughout the genome. They comprise approximately 3% of the human genome and are found, on average, every 2kb.23 These tandem repeats are thought to arise by slippage events during DNA replication. Slippage events have been frequent enough during human evolution that there are usually multiple alleles at any given microsatellite sequence.

The molecular basis for the observed instability and its role in HNPCC was subsequently elucidated in a remarkably short period of time due to two additional critical observations. First, investigators working in the field of yeast genetics recognized that the pattern of MSI observed in the colorectal tumors was similar to that seen in bacteria and yeast with defects in DNA mismatch repair (MMR).24 This finding suggested that the genes involved in human MMR might be responsible for the MSI observed in a subset of CRC. The second important observation was the discovery of linkage in a subset of HNPCC families to chromosome 2p.25 Linkage analysis, along with the clues provided by the yeast MMR work, subsequently led to the cloning of a human MutS homolog (hMSH2) and the identification of germline mutations in the same subset of HNPCC kindreds with linkage to chromosome 2p.26,27 These critical observations set the stage for the tremendous amount of work accomplished over the past decade to gain a molecular understanding of HNPCC.

MMR is one of several mechanisms involved in the correction of mutations that occur as a result of exogenous or endogenous mutagens or misincorporation during DNA replication. This system is highly conserved through evolution, and much of what is known about MMR is based on work performed in bacteria and yeast (reviewed in Reference 28). MMR requires the concerted action of several proteins. In humans, at least six different proteins are directly involved in MMR. Deformities in the DNA helix caused by single base pair mismatches or insertion or deletion loop mutations are recognized and bound by a heterodimer called MutS consisting of hMSH2 and either hMSH6 or hMSH3. Each of these heterodimers has specific but partially redundant recognition specificity for the two different types of mutations. hMSH2/hMSH3 preferentially binds to insertion/deletion loops, while hMSH2/hMSH6 preferentially recognizes single base pair mismatches. A second heterodimer called MutL, composed of hMLHl and either hPMS2 or hMLH3 (or possibly hPMSl), is responsible for differentiating between the template and newly synthesized strand (destined for repair) and for coordinating the interplay between the recognition complex and other proteins necessary for MMR. The hMLH1/hPMS2 heterodimer provides the bulk of this function, while hMLH1/hMLH3 plays a minor role. Following mismatch recognition and assembly of the repair proteins, the error-containing strand is excised, resynthesized, and relegated to complete the repair process (reviewed in Reference 29).

With the identification of the human MMR genes, it has become apparent that a significant fraction of clinically defined HNPCC is due to alterations in these genes. In mutation carriers, MMR function is phenotypically normal, although every somatic cell carries one inactivated and one normal copy of the MMR gene mutated in the specific HNPCC family. However, cells in individuals who are heterozygous for a mutation in an MMR gene are susceptible to loss of the wild-type allele through large-scale chromosomal losses or deletions manifesting as loss of heterozygosity [LOH] or other somatic mutations within the gene. Inactivation of both alleles leads to the loss of functional MMR and a hypermutable phenotype characterized by a large increase in single-base changes and insertion or deletion mutations at microsatellite loci and other loci.30,31 This phenotype is observed in the laboratory as MSI, which is an outcome of defects in the repair process, not of increased replication errors. The resulting hypermutable cell, however, is also susceptible to the accumulation of mutations in cellular protooncogenes and tumor suppressor genes, conferring a growth advantage that results in clonal expansion and drives the oncogenic process. It is important to note that the vast majority of microsatellites do not occur within critical regions of genes. Therefore, instability at these noncoding loci does not appear to have a phenotypic effect. However, MSI provides a very useful phenotypic marker for identifying those tumors with defective MMR.

Since the initial cloning of hMSH2 and the discovery of germline mutations in a subset of families with HNPCC, the human homologues of most, if not all, MMR genes have been cloned and characterized. In addition to hMSH2, germline mutations have been identified in four other MMR genes in families with HNPCC: hMLHl, hMSH6, hPMS1,and hPMS2.32-35 Mutations also have been observed in hMLH3, but the putative pathogenic role of these mutations remains to be determined.36

Hundreds of germiline mutations have been reported and are cataloged in two databases (http://www.insight-group. org/ and http://www.med.mun.ca/MMRvariants/). Germline mutations in hMSH2 and hMLHl account for approximately 40% each of the reported mutations in families with defective MMR as the underlying cause of disease. This likely reflects experimental data indicating that the protein products of these genes are indispensable for MMR function.29 Nearly 10% of germline mutations occur in the hMSH6 gene. It has been reported that hMSH6 mutations often occur in HNPCC families characterized by a less-typical clinical presentation, including a later onset of cancer, a relatively higher occurrence of endometrial cancer, and a lower degree of MSI in tumor tissue.37-39 A smaller number of families have been reported that carry hPMS2 mutations.

Mutation analysis of MMR genes has provided estimates of the proportion of families that meet clinical criteria for HNPCC and also have identifiable germline mutations.40,41 Depending on the techniques used and the number of genes examined, between 40% and 90% of families that fit the Amsterdam criteria have identifiable germline muta-tions.42 If the less-restrictive Amsterdam II criteria are utilized, this proportion decreases to 5% to 50%.43,44 There are several possible explanations for this finding. The manner in which the population of HNPCC families under study is ascertained plays an important role in the proportion of families that are found to be positive for mutations in MMR genes. Large, well-defined families with many affected individuals are much more likely to harbor recognizable mutations than are families that barely meet the minimum clinical criteria. A portion of undetected mutations in these families also may be due to technical issues related to the mutation-screening methods, which detect only certain types of mutations and generally have limited mutation-detection rates that range from 60% to 90%. In addition, most laboratories have traditionally examined only the hMSH2 and hMLHl genes for mutations, thus biasing against families that carry mutations in other MMR genes. Also, families that fit the clinical criteria for HNPCC may do so because of environmentally induced aggregates or may occur by chance as a result of the high frequency of CRC in the general population. Finally, another portion of these clinically defined HNPCC families may be due to yet undiscovered genes that predispose to HNPCC. These unidentified genes may or may not be involved in MMR. In support of this hypothesis, linkage analyses suggest that HNPCC families exist that do not involve known loci.45,46 More important, a significant fraction of HNPCC families do not have evidence of defective MMR when tumors from family members are tested.17 Families which fit the Amsterdam criteria I for HNPCC but do not have defective MMR have recently been designated as "familial colorecteal cancer type X" to distinguish them from families with hereditary MMR deficiency. These families have a lower incidence of colorectal cancer and may not have any increased incidence of other cancers.42

Because of the heterogeneity noted above, it is now possible to make a distinction between HNPCC and a hereditary defect in an MMR gene which is now termed Lynch syndrome.43 HNPCC is a clinical diagnosis primarily based on family history and other clinical information. As discussed above, current data indicate that there is a subset of families with HNPCC in which tumors do not exhibit any phenotypic evidence of defects in MMR (MSI or the absence of protein expression), and germline mutations in MMR genes have not been identified. Therefore, it is likely that other genes not involved in MMR are responsible for the diagnosis of some patients with HNPCC. Conversely, not all patients with germline defects in MMR genes meet the clinical criteria for HNPCC.17 Thus, Lynch syndrome is a genetic diagnosis based on the finding of defective MMR in the tumor and a germline mutation in one of the MMR genes. Not all patients with clinically defined HNPCC will have Lynch syndrome, and not all patients with Lynch syndrome will fulfill the clinical criteria for HNPCC. Lynch syndrome is likely to be a relatively homogenous diagnosis despite the locus heterogeneity, whereas HNPCC is more heterogeneous and includes both Lynch syndrome and other hereditary genetic causes of HNPCC. This important distinction has implications for the identification of patients with CRC who should undergo tumor screening and potentially germline testing (see Strategy for HNPCC testing, below). This distinction also has important implications as it relates to the natural history of the disease.42

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment