A subgroup of microsatellite stable colorectal cancers has elevated mutation rates and different responses to alkylating and oxidising agents.

An early step in the carcinogenesis of hereditary non-polyposis colorectal cancer (HNPCC) and some sporadic colorectal cancers (CRCs) is the acquisition of a 'mutator phenotype' resulting from defects in DNA mismatch repair (MMR) genes, which normally maintain genomic stability. This mutator phenotype causes an approximately 100-1000-fold increase in base substitutions and small insertion/deletion mutations thereby driving carcinogenesis. It also causes genome-wide microsatellite instability (MSI) due to the inability to repair mutations within these small, hard to replicate, repetitive DNA elements. In contrast, less is known about the role of mutator phenotypes in microsatellite stable (MSS) CRC. In this report, we have measured the mutation rates in 11 MSS CRC cell lines to obtain an estimate of the prevalence of mutator phenotypes in MSS carcinogenesis. Of the 11 cell lines, three of them (27%) possess spontaneous hypoxanthine phosphoribosyltransferase mutation rates approximately 10-100-fold above background. When challenged with alkylating and oxidising agents, the degree of survival and apoptotic responses are different, indicating that these cell lines may represent more than one mutator phenotype. These data demonstrate that a significant portion of MSS CRC cell lines has increased mutation rates and that this may play a role in MSS CRC carcinogenesis.

Human MutY: gene structure, protein functions and interactions, and role in carcinogenesis.

Faithful maintenance of the genome is crucial to the individual and the species. Oxidative DNA damage, such as 8-oxo-7,8-dihydroguanine (8-oxoG), poses a major threat to genomic integrity. 8-OxoG can mispair with 2'-deoxycytidine 5'-triphosphate or with 2'-deoxyadenosine triphosphate during DNA replication, forming C*8-oxoG and A*8-oxoG mispairs. Human MutY is responsible for recognition and removal of the inappropriately inserted adenine in an A*8-oxoG mispair. If unrepaired, the A*8-oxoG mispairs can result in deleterious C:G to A:T transversions. Human MutY functions in a postreplication repair pathway and is targeted to the newly synthesized daughter strand of DNA for removal of the adenine base. The human MutY protein is targeted to both the mitochondria and the nucleus and associates with the proliferating cell nuclear antigen, apurinic/ apyrimidinic endonuclease 1, replication protein A and mutS homolog 6 proteins. Mutations in the human MutY gene and defective activity of the human MutY protein have been detected in cancer. A direct correlation between defective A*8-oxoG repair and increased levels of genomic 8-oxoG has now been established.

An early lesion in hepatic carcinogenesis: loss of heterozygosity in human cirrhotic livers and dysplastic nodules at the 1p36-p34 region.

Loss of heterozygosity (LOH) of chromosome 1 has been suggested, by karyotyping, to be an initial episode in human hepatocarcinogenesis. However, this alteration has not yet been investigated in cirrhotic nodules (CNs) or dysplastic nodules (DNs). In an initial study from explanted or resected cirrhotic livers, LOH in 1p36-p32 was examined in 31 hepatocellular carcinomas (HCCs), 25 low-grade dysplastic nodules (LGDNs), and 24 high-grade dysplastic nodules (HGDNs). In HCCs, LOH was detected most frequently at loci D1S2843 (1p36.1) (28.6%), D1S513 (1p34.3) (29.2%), and MYCL1 (1p34.1) (28.6%). In HGDN and LGDN, LOH incidences at D1S513 were 11.1% and 13.6%, respectively. To further refine those results and to determine sequential relationships among CN, DN, and HCC, LOH was next studied in an additional 53 HCCs, 56 HGDNs, 30 LGDNs, and 215 CNs from 11 explanted human cirrhotic livers, including 30 "nodule-in-nodule" lesions. Seven markers between D1S2843 (1p36.12) and MYCL1 (1p34.1), and 1 each at D1S484 (1q24.1), IGF2R-3 (6q26), and TP53 (17p13.1) were used. LOH at D1S2843 and D1S513 was detected in HCCs (20.4% and 23.5%, respectively), HGDNs (7.7% and 18.5%), LGDNs (13.6% and 6.9%), and CNs surrounding either HCCs or DNs (7.4% and 8.3%). These results demonstrate that LOH at D1S2843 and D1S513 are early events in human liver carcinogenesis. Data from CN surrounding either HCCs or DN, and also nodule-in-nodule lesions, provide evidence supporting a CN-->DN-->HCC progression. Different deletion patterns from multiple HCCs and DNs suggest independent origins for carcinogenesis in the same individual.

Microsatellite instability is infrequent in medullary breast cancer.

Microsatellite instability (MSI), characterized by contraction or expansion in microsatellite length or short tandem repeats compared with germline lengths, is found in 85% to 90% of colon cancer arising in hereditary nonpolyposis colorectal cancer families. These cancers commonly have characteristic histologic appearances, including medullary features with intense lymphoid infiltrates. In pancreatic cancer, a rare medullary histologic subtype more often demonstrates MSI than the more common adenocarcinoma subtype. We hypothesized that the medullary histologic pattern might correlate with MSI in additional tumor types and analyzed 8 cases of typical and atypical medullary carcinoma of the breast. Tumor and normal DNA was extracted from paraffinized tissue blocks of tumor and histologically uninvolved axillary lymph nodes, respectively. We analyzed the tumors for instability in 5 primary (BAT25, BAT26, D17S250, D5S346, D2S123) and 3 alternative (BAT40, D18S55, D18S58) microsatellites recommended at the National Cancer Institute--sponsored conference for diagnosis of MSI in colorectal cancer. All 8 tumors were microsatellite stable at the 8 loci, suggesting that MSI is not commonly associated with medullary or atypical medullary breast carcinoma, in contrast with the reported association with medullary tumors of the colon and pancreas.

Glycophosphatidylinositol-anchored protein deficiency as a marker of mutator phenotypes in cancer.

Phosphatidylinositol glycan-A (PIGA) is a gene that encodes an element required for the first step in glycosylphosphatidylinositol (GPI) anchor assembly. Because PIGA is X-located, a single mutation is sufficient to abolish cell surface GPI-anchored protein expression. In this study, we investigated whether mutation of the PIGA gene could be exploited to identify mutator (Mut) phenotypes in cancer. We examined eight Mut colon cancer lines and four non-Mut colon cancers as controls. In every case, flow cytometric analyses of cells sorted for low fluorescence after staining for GPI-linked CD59 and CD55 revealed negative peaks in the Mut lines but not in the controls. Single cell cloning of purged and sorted GPI-anchor- HCT116 cells and sequencing of the PIGA gene in each clone uniformly showed mutations. Pretreatment of the Mut lines with anti-CD55 or anti-CD59 antibodies and complement or with the GPI-anchor-reactive bacterial toxin aerolysin enriched for the GPI-anchor- populations. Expansion of purged GPI-anchor+ cells in the Mut lines and analyses using aerolysin in conjunction with flow cytometry yielded PIGA gene mutation frequencies of 10(-5) to 10(-4), values similar to the mutation frequencies of the hprt gene. This novel approach allows for the detection of as yet undescribed repair or replication defects and in addition to its considerably greater ease of use than existing techniques and in principle would not require the production of cell lines.

Detection of microsatellite instability.

In 1993, three groups independently discovered that the lengths of microsatellites in tumors could vary from the normally constant pattern defined at birth (5-5) (see review in ref. 4). This discovery has been designated either microsatellite instability (MSI) or replication errors (RER). A recent international consensus conference convened by the National Cancer Institute defined MSI/RER as "a change in length due to either insertion or deletion of repeating units, in a microsatellite within a tumor when compared to normal tissue" (5). Microsatellites are regions of repetitive DNA in which the repeating unit is small, varying in length from 1 to 6 nucleotides, and in which the number of repeating units in a microsatellite can vary from 10-60 (6-7). Because microsatellite lengths generally vary from person to person, they have received widespread use in forensics, gene mapping, parentage testing.

Cytotoxicity and mutagenicity of frameshift-inducing agent ICR191 in mismatch repair-deficient colon cancer cells.

BACKGROUND: Deficiency of DNA mismatch repair is a common feature of cancers exhibiting instability of microsatellite DNA sequences. Cancers with microsatellite instability are recognizable by their high rate of spontaneous frameshift mutations within microsatellite sequences, their resistance to killing by cytotoxic agents, and their localization to specific tissues, e.g., the proximal colon and stomach. We hypothesized that the mismatch repair deficiency of these cancers would make them vulnerable to environmental or chemical frameshift-inducing agents. This study was undertaken to test whether exogenous frameshift-inducing agents selectively induce mutations in mismatch repair-deficient cells of mutagen-exposed tissues like the colon and whether cytotoxic doses of these agents would preferentially kill those cells. METHODS: Cytotoxicity of the acridine mutagen 6-chloro-9-[3-(2-chloroethylamino)propylamino]-2-methoxy-acridine (ICR191), a DNA frameshift inducer, was determined in the mismatch repair-deficient human colon carcinoma cell line HCT116 versus the repair-reconstituted derivative HCT116+C3. Vulnerability to the mutagenic effects of ICR191 was determined by transfection of HCT116 or HCT116+C3 cells with a frameshift reporter vector, followed by treatment of the cells with ICR191. Alternatively, the reporter vector was reacted ex vivo with ICR191, and the derivatized vector was then transfected into HCT116 or HCT116+C3 cells. RESULTS: ICR191 proved to be fivefold to 10-fold more potent in inducing mutations in mismatch repair-deficient HCT116 cells than in mismatch repair-proficient HCT116+C3 cells. Moreover, at cytotoxic doses of ICR191, repair-deficient HCT116 cells proved to be fivefold more vulnerable to killing than did HCT116+C3 cells. CONCLUSIONS: Frameshift-inducing mutagens can selectively induce mutations in mismatch repair-deficient cells versus mismatch repair-proficient cells. Environmental exposures may, therefore, favor development of cancers with microsatellite instability in tissues like the gut. Frameshift-inducing agents can, however, also preferentially kill mismatch repair-deficient cancer cells and, thus, may be promising as model therapeutic compounds.

Detection of microsatellite instability by fluorescence multiplex polymerase chain reaction.

We have created a clinical molecular diagnostic assay to test for microsatellite instability (MSI) at multiple loci simultaneously in paraffin-embedded surgical pathology colon resection specimens. This fluorescent multiplex polymerase chain reaction (PCR) assay analyzes the five primary microsatellite loci recommended at the 1997 National Cancer Institute-sponsored conference on MSI for the identification of MSI or replication errors in colorectal cancer: Bat-25, Bat-26, D2S123, D5S346, and D17S250. Amplicon detection is accomplished by capillary electrophoresis using the ABI 310 Genetic Analyzer. Assay validation compared 18 specimens previously assessed by radioactive PCR and polyacrylamide gel electrophoresis detection to results generated by the reported assay. Germline and tumor DNA samples were amplified in separate multiplex PCR reactions, sized in separate capillary electrophoresis runs, and compared directly to identify novel length alleles in tumor tissue. A concordance of 100% between the two modalities was achieved. The multiplex assay routinely detected a subpopulation of 10% tumor alleles in the presence of 90% normal alleles. A novel statistical model was generated that corroborates the validity of using results generated by analysis of five independent microsatellites to achieve a single overall MSI diagnosis. The assay presented is superior to standard radioactive monoplex PCR, polyacrylamide gel electrophoretic analysis, primarily due to the multiplex PCR format.

Pseudo-spikes are common in histologically benign lymphoid tissues.

T cell receptor gene rearrangement is a classic marker of T cell clonality and is a useful adjunct in the diagnosis of T cell lymphomas and leukemias. Rearranged V-J gene segments amplified by polymerase chain reaction (PCR) are traditionally analyzed by polyacrylamide gel electrophoresis. We and others have analyzed TCR-gamma PCR products using capillary gel electrophoresis, which produces single nucleotide resolution and provides improved diagnostic sensitivity over conventional methods. However, with this marked increase in resolution and sensitivity, it is necessary to re-define normal variation of TCR-gamma gene rearrangement in control tissues to allow appropriate interpretation of monoclonality if present. Using DNA capillary gel electrophoresis, we examined the spectrum of normal patterns for TCR-gamma in a variety of T-cell-rich, histologically benign tissue types, including spleen, lymph node, tonsil, and blood, and compared this with the patterns in T cell lymphoma samples. We defined relative peak heights as h1/h2, where h1 represents the peak height of the largest peak above the normally distributed population, and h2 represents the peak height of the normally distributed curve. We found spikes in almost 20% of histologically benign samples with relative peak heights that were more than 0.5 and up to 1.5. We designated these as pseudo-spikes, because they may be mistaken for monoclonal spikes. In contrast, the relative peak height of the T cell lymphoma samples that showed clonal rearrangement was much higher than that of the pseudo-spikes, being at least 2 in 11/11 and at least 3 in 10/11 cases. Our data suggest that peaks with relative height of at least 3 represent a true clonal population in diagnostic samples. Peaks with relative heights of less than 1.5 may be insignificant, while peaks with relative heights between 1.5 to 3 may warrant further evaluation. Although capillary gel electrophoresis is superior in assessing T cell clonality, caution must be exercised when interpreting results, because pseudo-spikes appear to be common in benign tissues with lymphoid populations and are not necessarily indicative of clonal malignant T cell population.

Use of single nucleotide polymorphisms (SNP) and real-time polymerase chain reaction for bone marrow engraftment analysis.

Allogeneic bone marrow transplant engraftment assays use polymorphisms in the human genome to determine the relative percentages of donor and recipient cells present in the recipient. We describe a novel posttransplant assay approach using single nucleotide polymorphisms (SNPs), the most common type of polymorphism in humans. Using samples of defined genotype, we used real-time polymerase chain reaction (PCR) and allele-specific fluorescent TaqMan probes to assay a SNP of the cytochrome P450 CYP2C9 gene. Standard curves of chimeric mixes showed a linear relationship between the ratio of two alleles and the ratio of their respective fluorophore emission, except for mixes with a low percentage (< 5%) of the less common allele. We validated the SNP real-time PCR assay by comparing it to Southern hybridization analysis, analyzing DNA mixes in a blinded fashion with both methods. The correlation between the two methods was high. We have produced a statistical model that varies allele frequency to predict how many SNPs would be required to produce a functional SNP panel. Additional development will be necessary to produce such a panel of highly informative SNPs for clinical use. A real-time PCR SNP assay may ultimately provide more accurate quantification and shortened turnaround time compared to current post-engraftment assays.

Leukemia arising out of paroxysmal nocturnal hemoglobinuria.

In paroxysmal nocturnal hemoglobinuria (PNH), one or more hematopoietic stem cells that are defective in GPI anchor assembly as a result of mutation in the PIG-A gene preferentially expand in the bone marrow and give rise to peripheral blood elements that are deficient in GPI anchored protein expression. According to current concepts, 5-15% of PNH patients develop leukocyte dyscrasias which invariably are acute myelogenous leukemia (AML). In this review, the literature from 1962 to the present is analyzed regarding the type of leukocyte dyscrasia, incidence, and cytogenetic features of the abnormal cells that have been reported. Among a total of 119 cases that are well-documented, 104 myeloid dyscrasias involving several categories in addition to AML, as well as 15 lymphoid dyscrasias are described. Of 1,760 patients in 15 series that contain 20 or more patients, 16 (1%) are reported as having developed "acute leukemia." However, of 288 listed as having died, 13 (5%) are recorded as having had "acute leukemia." In 32 of the patients with hematological dyscrasias where karyotypes were analyzed, 7 were found to be normal and 25 found to harbor various alterations with the +8 abnormality present in 8. In 5 of 7 instances evidence indicates that the dyscratic cell arises from the PNH clone. Processes potentially involved in the evolution of the dyscratic cells from PNH clones are discussed.

Improved detection of human immunodeficiency virus type 1 variants by analysis of replicate amplification reactions: relevance to studies of human immunodeficiency virus type 1 vertical transmission.

BACKGROUND: Human immunodeficiency virus type 1 (HIV-1)-infected individuals typically harbor mixtures of HIV-1 variants. For HIV-1 transmission studies, methods used for genotypic analysis should reliably detect variant mixtures. Such studies typically analyze complementary DNAs (cDNAs) from a single polymerase chain reaction (PCR) amplification. This approach may fail to detect variant mixtures in some samples because of analytic bias. METHODS: To evaluate the impact of analytic bias on the detection of HIV-1 variants, we analyzed samples from a mother and infant known to contain both subtypes A and D HIV-1. The env third variable region of HIV-1 gp120 (V3 region) was amplified and cloned in five replicate experiments using a single plasma sample from each individual. Ten cDNAs from each experiment were analyzed. RESULTS: The subtype mixture was detected in only four of 10 amplification experiments (three of five for the mother and one of five for the infant). Sequencing of uncloned PCR products showed that a single subtype, either A or D, was preferentially amplified in each experiment. However, the subtype mixture was detected for each sample by analyzing the five replicate experiments as a group. CONCLUSIONS: This shows that mixtures of HIV-1 variants may be more readily detected when replicate amplification reactions are analyzed. This approach may be useful for characterizing HIV-1 variants for studies of HIV-1 transmission.

A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer.

In December 1997, the National Cancer Institute sponsored "The International Workshop on Microsatellite Instability and RER Phenotypes in Cancer Detection and Familial Predisposition," to review and unify the field. The following recommendations were endorsed at the workshop. (a) The form of genomic instability associated with defective DNA mismatch repair in tumors is to be called microsatellite instability (MSI). (b) A panel of five microsatellites has been validated and is recommended as a reference panel for future research in the field. Tumors may be characterized on the basis of: high-frequency MSI (MSI-H), if two or more of the five markers show instability (i.e., have insertion/deletion mutations), and low-frequency MSI (MSI-L), if only one of the five markers shows instability. The distinction between microsatellite stable (MSS) and low frequency MSI (MSI-L) can only be accomplished if a greater number of markers is utilized. (c) A unique clinical and pathological phenotype is identified for the MSI-H tumors, which comprise approximately 15% of colorectal cancers, whereas MSI-L and MSS tumors appear to be phenotypically similar. MSI-H colorectal tumors are found predominantly in the proximal colon, have unique histopathological features, and are associated with a less aggressive clinical course than are stage-matched MSI-L or MSS tumors. Preclinical models suggest the possibility that these tumors may be resistant to the cytotoxicity induced by certain chemotherapeutic agents. The implications for MSI-L are not yet clear. (d) MSI can be measured in fresh or fixed tumor specimens equally well; microdissection of pathological specimens is recommended to enrich for neoplastic tissue; and normal tissue is required to document the presence of MSI. (e) The "Bethesda guidelines," which were developed in 1996 to assist in the selection of tumors for microsatellite analysis, are endorsed. (f) The spectrum of microsatellite alterations in noncolonic tumors was reviewed, and it was concluded that the above recommendations apply only to colorectal neoplasms. (g) A research agenda was recommended.

Accumulated clonal genetic alterations in familial and sporadic colorectal carcinomas with widespread instability in microsatellite sequences.

A subset of hereditary and sporadic colorectal carcinomas is defined by microsatellite instability (MSI), but the spectra of gene mutations have not been characterized extensively. Thirty-nine hereditary nonpolyposis colorectal cancer syndrome carcinomas (HNPCCa) and 57 sporadic right-sided colonic carcinomas (SRSCCa) were evaluated. Of HNPCCa, 95% (37/39) were MSI-positive as contrasted with 31% (18/57) of SRSCCa (P < 0.000001), but instability tended to be more widespread in SRSCCa (P = 0.08). Absence of nuclear hMSH2 mismatch repair gene product by immunohistochemistry was associated with germline hMSH2 mutation (P = 0.0007). The prevalence of K-ras proto-oncogene mutations was similar in HNPCCa and SRSCCa (30% (11/37) and 30% (16/54)), but no HNPCCa from patients with germline hMSH2 mutation had codon 13 mutation (P = 0.02), and two other HNPCCa had multiple K-ras mutations attributable to subclones. 18q allelic deletion and p53 gene product overexpression were inversely related to MSI (P = 0.0004 and P = 0.0001, respectively). Frameshift mutation of the transforming growth factor beta type II receptor gene was frequent in all MSI-positive cancers (85%, 46/54), but mutation of the E2F-4 transcription factor gene was more common in HNPCCa of patients with germline hMSH2 mutation than in those with germline bMLH1 mutation (100% (8/8) versus 40% (2/5), P = 0.04), and mutation of the Bax proapoptotic gene was more frequent in HNPCCa than in MSI-positive SRSCCa (55% (17/31) versus 13% (2/15), P = 0.01). The most common combination of mutations occurred in only 23% (8/35) of evaluable MSI-positive cancers. Our findings suggest that the accumulation of specific genetic alterations in MSI-positive colorectal cancers is markedly heterogeneous, because the occurrence of some mutations (eg, ras, E2F-4, and Bax genes), but not others (eg, transforming growth factor beta type II receptor gene), depends on the underlying basis of the mismatch repair deficiency. This genetic heterogeneity may contribute to the heterogeneous clinical and pathological features of MSI-positive cancers.

Chromosome number and structure both are markedly stable in RER colorectal cancers and are not destabilized by mutation of p53.

Fourteen colorectal cancer cell lines, categorized according to the presence or absence of microsatellite instability, were further analysed for chromosomal stability by karyotyping. NonRER (microsatellite stable) cell lines typically displayed highly aberrant karyotypes with alterations not only of chromosome number but also of chromosome structure including chromosomal deletions, inversions, and translocations. RER (microsatellite unstable) cell lines, in contrast, displayed significantly fewer alterations of chromosome number. Moreover, RER cell lines also displayed significantly fewer cytogenetically evident alterations of chromosome structure. Compared to NonRER colon cancers, RER colon cancers are significantly less likely to have undergone chromosomal gain, loss, or breakage. Characterization of p53 gene status by gene sequencing was performed in an attempt to determine if p53 gene status correlated with the chromosomal stability of the RER cancers. Gene mutations in p53 were present in all of the NonRER colon cancers. However, p53 gene mutations were also found present in four of nine of the RER colon cancers. Unexpectedly, RER colon cancers bearing mutant p53 demonstrated the same stability of chromosome number, and the same stability of chromosome structure, as the RER colon cancers with wild-type p53. Therefore, in RER colon cancers specific p53 independent mechanisms actively maintain the stability of both chromosome number and structure.

Increased transversions in a novel mutator colon cancer cell line.

We describe a novel mutator phenotype in the Vaco411 colon cancer cell line which increases the spontaneous mutation rate 10-100-fold over background. This mutator results primarily in transversion base substitutions which are found infrequently in repair competent cells. Of the four possible types of transversions, only three were principally recovered. Spontaneous mutations recovered also included transitions and large deletions, but very few frameshifts were recovered. When compared to known mismatch repair defective colon cancer mutators, the distribution of mutations in Vaco411 is significantly different. Consistent with this difference, Vaco411 extracts are proficient in assays of mismatch repair. The Vaco411 mutator appears to be novel, and is not an obvious human homologue of any of the previously characterized bacterial or yeast transversion phenotypes. Several hypotheses by which this mutator may produce transversions are presented.

Wild-type p53 protein potentiates cytotoxicity of therapeutic agents in human colon cancer cells.

Wild-type p53 is induced by DNA damage. In different cell types, this induction is suggested either to facilitate DNA repair by inducing a cell cycle pause or to potentiate cell death via apoptosis. Wild-type p53 in different cell types has similarly been associated with either enhancement of or increased resistance to the cytotoxicity of many cancer therapeutic agents. We have constructed a colorectal cancer cell line bearing, in addition to endogenous mutant p53 alleles, an exogenous wild-type p53 allele that is under the regulatable control of the lac repressor. Induction of wild-type p53 by isopropyl-beta-thiogalactopyranoside in these cells induces a reversible growth arrest but does not induce cell death. However, we find that the induction of wild-type p53 powerfully potentiates the cytotoxicity of both irradiation and 5-fluorouracil, two agents that are used clinically in the treatment of colorectal cancer. We also find that induction of wild-type p53 potentiates the cytotoxicity of topotecan, a member of the camptothecin family of drugs that also has clinical activity against colon cancer. These findings suggest that the common loss of wild-type p53 in many colorectal cancers may play a role in the clinical resistance of these tumors to anticancer agents. Although some cancer cells may not be directly killed by p53 gene therapy, our findings suggest that genetic alteration of some cancers to induce wild-type p53 may increase their sensitivity to cytotoxic gene therapy.

Diverse hypermutability of multiple expressed sequence motifs present in a cancer with microsatellite instability.

Colon cancer and an increasing number of other cancers have been found to exhibit instability of DNA microsatellite sequences. Such tumors have been designated as replication errors (RER) tumors. However, as microsatellites are only rarely found within coding regions of the genome, instability of these sequences cannot directly contribute to carcinogenesis. Recently, we have shown RER colon cancers also demonstrate a marked 100-fold increase in mutation rates measured within an expressed gene, hprt, suggesting the mutator phenotype in these tumors extends beyond microsatellite sequences. To determine whether the RER phenotype indeed destabilizes non-repetitive DNA sequences we have sequenced hprt gene mutations recovered from the RER colon cancer cell line RKO. Greater than 10% of hprt mutants proved to be a single 3 bp deletion located in a nonrepetitive ATTAT sequence motif. Additionally, 1-4 bp deletions or insertions were found to be randomly located throughout the hprt gene. Lastly, one third of hprt mutations proved to be transitions or transversions. The microsatellite instability demonstrated in RKO is thus a global mutator phenotype which destabilizes DNA sequences both inside and outside of repetitive sequence elements and which augments base substitutions as well as frameshifts. These findings extend the characteristics of mutations associated with RER tumors and suggest additional mechanisms by which mutator phenotypes may alter oncogenes and tumor suppressor genes.