Mutated gene-specific phenotypes of dinucleotide repeat instability in human colorectal carcinoma cell lines deficient in DNA mismatch repair.

Mutations in DNA mismatch repair (MMR) genes in hereditary non-polyposis colon cancer (HNPCC) patients revealed the importance of MMR deficiency as a risk for carcinogenesis. Since diverse mutations occur in several MMR genes, the instability of repeat sequences dispersed in the genome, which are also governed by the MMR system, is a well used marker. However, the relationship between repeat sequence instability and MMR gene mutation in human cells has not been well defined mainly because precise systems to analyse repeat sequences have not been available. Using our newly developed system, we analysed alteration of dinucleotide repeats in human cell lines which harbour mutations in MMR genes. Among 24 subclones of DLD-1 cells (hMSH6-) only one had a dinucleotide repeat alteration in only one microsatellite locus, while LoVo cells (hMSH2-/hMSH6-) exhibited marked dinucleotide repeat instability (DRI). HCT116 cells, a hMLH1-mutant, showed an ultimate DRI phenotype. Interestingly, SW48 cells lacking hMLH1 expression also demonstrated DRI, albeit the extent of diversity being significantly lower than HCT116. These data suggest that the DRI phenotype in human cells is highly dependent on mutated MMR genes and on forms of mutation. The results of DRI analyses used to detect MMR-deficiency should be interpreted with caution.     

Hypersensitivity in DNA mismatch repair-deficient colon carcinoma cells to DNA polymerase reaction inhibitors

We studied the cytotoxic effects of various DNA replication inhibitors on MMR-deficient and -proficient colon carcinoma cell lines. DNA polymerase (pol) inhibitors including aphidicolin and gemcitabine, and hydroxyurea were more toxic (1.7 to 2.8-fold) to hMLH1-deficient HCT116 than to hMLH1-proficient HCT116+ch3. Similarly, pol inhibitors were more toxic to hMSH2-deficient LoVo than to hMSH2-proficient LoVo+ch2. In contrast, DNA topoisomerase I inhibitors, such as CPT-11, SN-38, and topotecan, were more toxic to MMR-proficient cells. Our results suggest that MMR-deficient colon carcinoma cells are hypersensitive to inhibitors of the pol reaction.     

Antisense inhibition of hMLH1 is not sufficient for loss of DNA mismatch repair function in the HCT116+chromosome 3 cell line.

We have reported that transfer of chromosome 3 (Chr3) containing a single wild-type copy of the hMLH1 gene into HCT116 colon cancer cells, a cell line deficient in DNA mismatch repair (MMR) activity attributable to inactivating hMLH1 mutations, corrects all of the aspects of the MMR repair-deficient phenotype. We inhibited the expression of the wild-type hMLH1 gene using antisense RNA in HCT116+Chr3 cells to determine if this would result in reversion to the MMR-deficient phenotype. Despite profound inhibition of hMLH1 expression, DNA MMR activity and alkylation sensitivity were not impaired in the antisense-transfected HCT116+Chr3 cells. Additionally, arrest of the cell cycle at the G2 phase with alkylation damage occurs in these cells, a phenotype associated with MMR proficiency. These results indicate that even with a reduction in the expression of hMLH1 protein below the limits of detection by Western blotting, DNA MMR activity remained fully functional (by direct DNA MMR activity assay). We would speculate that hMLH1 is expressed in substantially greater abundance than would be minimally necessary for DNA MMR and that minor reductions in the expression of this protein would not be sufficient to permit DNA MMR dysfunction. Alternatively, Chr3 may contain a second hMLH1 homologue that might overlap with the function of hMLH1.     

Different effects of methotrexate on DNA mismatch repair proficient and deficient cells

Antifolates exert their antiproliferative activity through the inhibition of dihydrofolate reductase and, as a consequence, of thymidylate synthesis, thereby inducing nucleotide misincorporation and impairment of DNA synthesis. We investigated the processes involved in the repair of antifolate-induced damage and their relationship with cell death. Since misincorporated bases may be removed by DNA mismatch repair (MMR), the study was carried out on the MMR-proficient human cell lines HeLa and HCT116+chr3, and, in parallel, on the MMR-deficient cell lines HeLa cell-clone12, defective in the protein hPMS2, and HCT116, with an inactive hMLH1. After treatment with methotrexate (MTX), we observed that DNA repair synthesis occurs independently of the cellular MMR function. Clear signs of apoptosis such as nuclear shrinkage, chromatin condensation and degradation, DNA laddering, and poly (ADP-ribose) polymerase (PARP) proteolysis, were visible in both MMR(+) and MMR(-) cells. Remarkably, cell viability was lower and the apoptotic process was triggered more efficiently in the MMR-competent cells.     

A novel mechanism for aspirin-mediated growth inhibition of human colon cancer cells.

PURPOSE: The molecular mechanisms by which aspirin and other nonsteroidal anti-inflammatory drugs exert chemopreventative effects in colon cancer are unclear and complex. Current investigations focus on the chemopreventive properties of nonsteroidal anti-inflammatory drugs, independent of their ability to inhibit cyclooxygenase (COX) activity, and presumably, identification of non-COX pathways will suggest new targets for clinical use. It was demonstrated recently that aspirin results in reduced microsatellite instability in colorectal cancer cells. We hypothesized that aspirin treatment might alter expression of DNA mismatch repair (MMR) proteins, representing another potential non-COX mechanism for its action. EXPERIMENTAL DESIGN: In this study, we have examined the effects of aspirin on the cellular growth rates, MMR protein levels, cell cycle analysis and apoptosis in MMR-deficient (HCT116) and MMR-proficient (HCT116+chr3 and SW480) human colon cancer cell lines. RESULTS: We found that treatment with aspirin inhibited the growth of these three cancer cell lines. In HCT116+chr3 cells, treatment with 1 mM of aspirin increased expression of the hMLH1 and hPMS2 proteins by 2.5-fold and 2-fold, respectively, and increased expression of the hMSH2 and hMSH6 proteins by 2-3-fold. For SW480 cells, treatment with 1 and 5 mM of aspirin increased expression of the hMLH1 and hPMS2 proteins by 2-4-fold and 3-5-fold, respectively, and increased expression of the hMSH2 and hMSH6 proteins by 3-7-fold. For all three of the cell lines, treatment with 1 and 2.5 mM of aspirin induced apoptosis at 48 and 72 h. Aspirin induced G(0)/G(1) cell cycle arrest in HCT116 cells. CONCLUSIONS: We conclude that aspirin acts through COX-independent mechanisms by resulting in an increase in MMR protein expression and subsequent apoptosis, which might serve as an additional means of growth inhibition in aspirin-treated human colon cancer cells.     

Human colon cancer cells surviving high doses of cisplatin or oxaliplatin in vitro are not defective in DNA mismatch repair proteins

PURPOSE: Alterations in the DNA mismatch repair (MMR) proteins have been associated with an increased resistance of many cancer cell lines to cisplatin. The aim of this work was to investigate whether defects in DNA MMR proteins are involved in the survival of human colorectal cancer cells in the presence of high concentrations of cisplatin and oxaliplatin, a diaminocyclohexane (DACH) platinum compound whose adducts are not recognized by the MMR system. METHODS: Six unselected human colon cancer cell lines (HT29, HCT15, HCT116, Caco2, SW480 and SW620) were treated with a single 3-h exposure to cisplatin or oxaliplatin at suprapharmacological concentrations, ranging from 50 to 200 microg/ml. The microsatellite stability and the expression of MMR proteins in the parental cell lines and in the drug-selected subpopulations were studied. RESULTS: Most cells underwent apoptosis in the days following the cisplatin or oxaliplatin treatment, but some colonies expanded 3 to 4 weeks after, suggesting the presence of innately resistant cells in the six parental cell lines. Microsatellite instability (MIN), which reflects genetic defects in the DNA MMR system, was detected only in the HCT116 parental cell line and its drug-selected counterparts, due to a known mutation in the hMLH1 gene. No acquired MIN was observed in the other cisplatin-selected sublines derived from the HT29, HCT15, Caco2, SW480 or SW620 parental cells. In the same way, Western blot analysis showed that expression of the DNA MMR proteins hMLH1, hPMS1, hPMS2, hMSH2 and hMSH6 did not differ between the parental and the drug-surviving cells. CONCLUSIONS: These results indicate that high-level resistance of human colon cancer cells to high doses of cisplatin and oxaliplatin does not seem to be related to acquired defects in the DNA MMR proteins.     

The mismatch repair protein, hMLH1, mediates 5-substituted halogenated thymidine analogue cytotoxicity, DNA incorporation, and radiosensitization in human colon cancer cells

Deficiency in DNA mismatch repair (MMR) is found in some hereditary (hereditary nonpolyposis colorectal cancer) and sporadic colon cancers as well as other common solid cancers. MMR deficiency has recently been shown to impart cellular resistance to multiple chemical agents, many of which are commonly used in cancer chemotherapy. It is therefore of interest to find an approach that selectively targets cells that have lost the ability to perform MMR. In this study, we examine the response of MMR-proficient (hMLH1+) and MMR-deficient (hMLH1-) colon carcinoma cell lines to the halogenated thymidine (dThd) analogues iododeoxyuridine (IdUrd) and bromodeoxyuridine (BrdUrd) before and after irradiation. These dThd analogues are used clinically as experimental sensitizing agents in radioresistant human cancers, and there is a direct correlation between the levels of dThd analogue DNA incorporation and tumor radiosensitization. In contrast to the well-characterized, marked increase in cytotoxicity (> 1 log cell kill) found with 6-thioguanine exposures in HCT116/3-6 (hMLH1+) cells compared to HCT116 (hMLH1-) cells, we found only modest cytotoxicity (10-20% cell kill) in both cell lines when treated with IdUrd or BrdUrd for 1 population doubling. Upon further analysis, the levels of halogenated dThd analogues in DNA were significantly lower (two to three times lower) in HCT116/3-6 cells than in HCT116 cells, and similar results were found in Mlh1+/+ spontaneously immortalized murine embryonic fibroblasts and fibroblasts from Mlh1 knockout mice. As a result of the higher levels of the dThd analogue in DNA, there was an increase in radiation sensitivity in HCT116 cells but not in HCT116/3-6 cells after pretreatment with IdUrd or BrdUrd when compared to treatment with radiation alone. Additionally, we found no differences in the cellular metabolic pathways for dThd analogue DNA incorporation because the enzyme activities of dThd kinase and thymidylate synthase, as well as the levels of triphosphate pools, were similar in HCT116 and HCT116/3-6 cells. These data suggest that the hMLH1 protein may participate in the recognition and subsequent removal of halogenated dThd analogues from DNA. Consequently, whereas MMR-deficient cells and tumor xenografts have shown intrinsic resistance to a large number of chemotherapeutic agents, the 5-halogenated dThd analogues appear to selectively target such cells for potential enhanced radiation sensitivity.     

Effect of H(2)O(2) on cell cycle and survival in DNA mismatch repair-deficient and -proficient cell lines

Patients who develop tumors with Lynch syndrome, which is caused by mutational inactivation of the DNA mismatch repair (MMR) system, have a relatively favorable prognosis compared to patients who develop sporadic tumors. Paradoxically, DNA MMR-deficient cells are resistant to many chemotherapeutic agents, and are capable of bypassing the G2/M checkpoint in vitro. Colon cancers that develop in the setting of Lynch syndrome show an abundant recruitment of immune cells into tumor tissues, which might be expected to increase oxyradical formation, and make the tumor cells more vulnerable to cell death. We examined the chemosensitivity and cell cycle response to oxidative stress in several MMR-deficient (HCT116, SW48, and DLD1) and -proficient (CaCo2, SW480, and HT29) colorectal cancer cell lines. H(2)O(2) induced a G2/M cell cycle arrest in both MMR deficient and proficient cell lines, however MMR-deficient cell lines were more sensitive to H(2)O(2) toxicity, and the response was more prolonged in MMR-deficient cells. Interestingly, human MutL-homologue (hMLH1-)defective HCT116 and hMLH1-restored HCT116+ch3 cell lines responded to H(2)O(2) with the same degree of G2/M arrest. The survival response of HCT116+ch3 was nearly identical to that of hMLH1-defective HCT116+ch2, although better than the response observed in HCT116 cells. In conclusion, greater cellular sensitivity and G2/M arrest in response to oxidative stress in MMR-deficient colorectal cancer cells could be one of the reasons for the more favorable prognosis seen in patients with Lynch syndrome. However, this sensitivity appears not to be a direct result of a deficient MMR function, but is more likely attributable to spectrum of target gene mutations that occurs in MMR-deficient tumors.     

Role of hMLH1 promoter hypermethylation in drug resistance to 5-fluorouracil in colorectal cancer cell lines

Loss of DNA mismatch repair (MMR) occurs in 10-15% of sporadic colorectal cancer, is usually caused by hMLH1 hypermethylation, and has been shown to confer resistance to various chemotherapeutic reagents, including 5-fluorouracil (5-FU). We tested the hypothesis that demethylation of the hMLH1 promoter in hypermethylated colorectal cancer cells would restore MMR proficiency and drug sensitivity to 5-FU. We used the MMR-deficient cell lines SW48, HCT116, HCT116+chr2 and the -proficient cell line HCT116+chr3. After treatment with the demethylating agent 5-Aza-2'-deoxycytidine (5 aza-dC), hMLH1 mRNA and protein expression were determined by RT-PCR and immunoblots. The methylation status for hMLH1 was investigated by methylation-specific PCR. Cells were subsequently treated with 5-FU and the growth characteristics ascertained by clonogenic assays. hMLH1 hypermethylation was reverted in SW48 cells 24 hr after treatment with 5 aza-dC and was accompanied by hMLH1 mRNA and protein reexpression. While 5 aza-dC alone did not affect the growth of SW48 cells, all other cell lines responded with a pronounced growth inhibition. 5-FU treatment strongly reduced the colony formation of HCT116+chr3 cells. These effects were significantly less in the MMR-deficient cells. Combined treatment of SW48 cells resulted in a similar growth pattern as seen in 5-FU only treated HCT116+chr3 cells. We demonstrate that in vitro resistance to 5-FU can be overcome by reexpression of hMLH1 protein through 5 aza-dC-induced demethylation in hypermethylated cell lines. Induction of the expression of methylated tumor suppressor or MMR genes could have a significant impact on the development of future chemotherapy strategies.     

Selective sensitivity to carboxyamidotriazole by human tumor cell lines with DNA mismatch repair deficiency

We have previously reported that high-dose nifedipine had a selective antiproliferative effect on colon cancer cell lines deficient in DNA mismatch repair (MMR). We hypothesized that carboxyamidotriazole (CAI), a calcium channel blocker, would also have a selective inhibitory effect on colon cancer cell lines with DNA MMR deficiency. In addition, we speculated that this effect may also be seen in cell lines deficient in DNA MMR derived from other tumor types. Fourteen human cancer cell lines with and without DNA MMR derived from carcinomas of the colon, bladder, ovary and prostate were treated with CAI, vehicle or control drugs (nifedipine and 5-flurouracil). The effect of treatment on growth inhibition, invasion, apoptosis and cell cycle progression was assessed. Selective sensitivity to CAI was observed in all cancer cell lines deficient in MMR. Compared with the MMR-proficient cells, the matched deficient cells were significantly more sensitive to the growth inhibitory effect of CAI and nifedipine, but less sensitive to 5-flurouracil. CAI significantly inhibited the invasive ability of MMR-deficient cancer cells compared to 5-flurouracil. CAI induced more apoptosis but similar level of G(2)/M arrest in MMR (hMLH1- or hMSH6-)-deficient colon cancer cells than MMR-proficient counterparts. CAI selectively inhibits proliferation and invasion in MMR-deficient human cancer cell lines. The antitumor effect is at least partly explained by G2/M cell cycle arrest and induction of apoptosis. These findings may have clinical implications directing clinical trials in selectively targeted patients with DNA MMR tumors.     

Hypoxia-induced enrichment and mutagenesis of cells that have lost DNA mismatch repair

Loss of DNA mismatch repair (MMR) increases the risk of spontaneous mutations. We sought to determine whether there was an interaction between hypoxia and MMR deficiency that might contribute to the phenomenon of tumor progression. Human colon carcinoma HCT116+ch2 (MMR-deficient) and HCT116+ch3 (MMR-proficient) sublines were exposed for varying periods of time to an environment of <0.1% O2 and pH as low as 6.1. When a population containing 5% MMR-deficient cells and 95% MMR-proficient cells was subjected to hypoxia for 72 h, the MMR-deficient cells were enriched by a factor of 2-fold in the surviving population, whereas no enrichment was detected in cells maintained under aerobic conditions. The potential of hypoxia to destabilize the genome was determined by measuring the frequency of clones in the surviving population resistant to very high concentrations of 6-thioguanine or cisplatin. A 72-h exposure to hypoxia did not increase the frequency of resistant clones in the MMR-proficient cells but produced a 7.8-fold increase in 6-thioguanine-resistant clones and a 2.5-fold increase in cisplatin-resistant clones in the MMR-deficient cells. Loss of MMR increased the frequency of mutations in a reporter vector sensitive to frameshift mutations in a microsatellite sequence. Exposure to hypoxia for a time period as short as 48 h further increased the number of mutations in both cell types, but the absolute number of mutants was higher in the MMR-deficient cells. These results indicate that hypoxia and its accompanying low pH enrich for MMR-deficient cells and that loss of MMR renders human colon carcinoma cells hypersensitive to the ability of hypoxia to induce microsatellite instability and generate highly drug-resistant clones in the surviving population.     

Role of the hMLH1 DNA mismatch repair protein in fluoropyrimidine-mediated cell death and cell cycle responses.

DNA mismatch repair (MMR) is an efficient system for the detection and repair of mismatched and unpaired bases in DNA. Deficiencies in MMR are commonly found in both hereditary and sporadic colorectal cancers, as well as in cancers of other tissues. Because fluorinated thymidine analogues (which through their actions might generate lesions recognizable by MMR) are widely used in the treatment of colorectal cancer, we investigated the role of MMR in cellular responses to 5-fluorouracil and 5-fluoro-2'-deoxyuridine (FdUrd). Human MLH1(-) and MMR-deficient HCT116 colon cancer cells were 18-fold more resistant to 7.5 microM 5-fluorouracil (continuous treatment) and 17-fold more resistant to 7.5 microM FdUrd in clonogenic survival assays compared with genetically matched, MLH1(+) and MMR-proficient HCT116 3-6 cells. Likewise, murine MLH1(-) and MMR-deficient CT-5 cells were 3-fold more resistant to a 2-h pulse of 10 microM FdUrd than their MLH1(+) and MMR-proficient ME-10 counterparts. Decreased cytotoxicity in MMR-deficient cells after treatment with various methylating agents and other base analogues has been well reported and is believed to reflect a tolerance to DNA damage. Synchronized HCT116 3-6 cells treated with a low dose of FdUrd had a 2-fold greater G(2) cell cycle arrest compared with MMR-deficient HCT116 cells, and asynchronous ME-10 cells demonstrated a 4-fold greater G(2) arrest after FdUrd treatment compared with CT-5 cells. Enhanced G(2) arrest in MMR-proficient cells in response to other agents has been reported and is believed to allow time for DNA repair. G(2) cell cycle arrest as determined by propidium iodide staining was not a result of mitotic arrest, but rather a true G(2) arrest, as indicated by elevated cyclin B1 levels and a lack of staining with mitotic protein monoclonal antibody 2. Additionally, p53 and GADD45 levels were induced in FdUrd-treated HCT116 3-6 cells. DNA double-strand break (DSB) formation was 2-fold higher in MMR-proficient HCT116 3-6 cells after FdUrd treatment, as determined by pulsed-field gel electrophoresis. The formation of DSBs was not the result of enhanced apoptosis in MMR-proficient cells. FdUrd-mediated cytotoxicity was caused by DNA-directed and not RNA-directed effects, because administration of excess thymidine (and not uridine) prevented cytotoxicity, cell cycle arrest, and DSB formation. hMLH1-dependent responses to fluoropyrimidine treatment, which may involve the action of p53 and the formation of DSBs, clearly have clinical relevance for the use of this class of drugs in the treatment of tumors with MMR deficiencies.     

DNA mismatch repair genes hMLH1, hMSH2, and hMSH6 are not inactivated in bronchioloalveolar carcinomas of the lung.

BACKGROUND: Defective DNA mismatch repair (MMR) appears to be rare in nonsmall cell carcinomas of the lung. Defective DNA MMR results from genetic or epigenetic alterations that inactivate the DNA MMR genes hMLH1 or hMSH2, and rarely hMSH6. The loss of normal DNA MMR is thought to promote tumorigenesis by accelerating the accumulation of mutations in oncogenes and tumor suppressor genes. Inactivation of hMLH1, hMSH2, and hMSH6 is observed as a loss of expression of these proteins by immunohistochemistry. Bronchioloalveolar carcinoma is a subtype of adenocarcinoma with distinctive clinical and pathologic features. MATERIALS AND METHODS. An immunohistochemical study was performed on paraffin embedded sections of 33 bronchioloalveolar carcinomas (20 nonmucinous and 13 mucinous) for hmlh1, hmsh2, and hmsh6 proteins. RESULTS All the tumors showed normal expression of hmlh1, hmsh2, and hmsh6. CONCLUSIONS: These findings suggest that defective DNA MMR due to inactivation of hMLH1, hMSH2, or hMSH6 does not play a significant role in the pathogenesis of bronchioloalveolar carcinomas.     

Mismatch repair and microsatellite instability in esophageal cancer cells

Using in vitro mismatch repair (MMR) assay, we have identified 3 of 22 esophageal cancer cell lines exhibiting reduced MMR activity. By means of gel-shift assay, decreased binding ability to GT mismatch and CA loop was observed in these 3 cell lines. However, we could not find any mutations in the hMSH2, hMSH3 and hMSH6 genes, the protein products of which exhibit mismatch binding activity in human cells. In addition, when using antibodies against 5 MMR-related proteins (hMSH2, hMSH3, hMSH6, hPMS2 and hMLH1), no aberrant expression was detected in any of them. When we examined 9 microsatellite loci in endogenous genomic DNA, these 3 esophageal cancer cell lines, deficient in MMR, did not exhibit microsatellite instability. However, when we examined the repetitious sequence on exogenous plasmid DNA which was introduced into these 3 esophageal cancer cells, the results suggested that MMR deficiency in esophageal cancer cells could result in moderate instability of the exogenous sequence.     

Combined mismatch and nucleotide excision repair defects in a human cell line: mismatch repair processes methylation but not UV- or ionizing radiation-induced DNA damage.

Interaction between long patch mismatch repair (MMR) and persistent DNA O6-methylguanine or 6-thioguanine (6-TG) is implicated in the cytotoxicity of methylating agents and 6-TG, respectively. Human cells with defective MMR tolerate DNA methylation damage and are cross-resistant to 6-TG. To determine whether MMR contributes to the lethal effects of persistent UV-induced DNA lesions, MMR deficiency was introduced into nucleotide excision repair (NER)-defective XP12RO cells. The doubly repair-defective cells, designated XP12ROB4, did not express detectable hMSH2 protein. They had the mutator phenotype, N-methyl-N-nitrosourea and 6-TG resistance typical of MMR-defective cells. Active MMR was not required for the cytotoxicity of UV light, and the hMSH2 defect did not detectably alter the survival of XP12ROB4. The level of spontaneous or UV-induced SCE was also similar in XP12RO and XP12ROB4, indicating that hMSH2 is not required for this recombination process. The combined deficiency in MMR and NER did not confer a significant degree of tolerance to ionizing radiation, and the survival of XP12RO and XP12ROB4 after gamma-radiation was similar. Although it recognizes and processes some persistent damaged or modified DNA base pairs, MMR is unlikely to serve as a general sensor of DNA damage.     

Differential radiosensitization in DNA mismatch repair-proficient and -deficient human colon cancer xenografts with 5-iodo-2-pyrimidinone-2'-deoxyribose.

PURPOSE: 5-iodo-2-pyrimidinone-2'-deoxyribose (IPdR) is a pyrimidinone nucleoside prodrug of 5-iododeoxyuridine (IUdR) under investigation as an orally administered radiosensitizer. We previously reported that the mismatch repair (MMR) proteins (both hMSH2 and hMLH1) impact on the extent (percentage) of IUdR-DNA incorporation and subsequent in vitro IUdR-mediated radiosensitization in human tumor cell lines. In this study, we used oral IPdR to assess in vivo radiosensitization in MMR-proficient (MMR+) and -deficient (MMR-) human colon cancer xenografts. EXPERIMENTAL DESIGN: We tested whether oral IPdR treatment (1 g/kg/d for 14 days) can result in differential IUdR incorporation in tumor cell DNA and subsequent radiosensitization after a short course (every day for 4 days) of fractionated radiation therapy, by using athymic nude mice with an isogenic pair of human colon cancer xenografts, HCT116 (MMR-, hMLH1-) and HCT116/3-6 (MMR+, hMLH1+). A tumor regrowth assay was used to assess radiosensitization. Systemic toxicity was assessed by daily body weights and by percentage of IUdR-DNA incorporation in normal bone marrow and intestine. RESULTS: After a 14-day once-daily IPdR treatment by gastric gavage, significantly higher IUdR-DNA incorporation was found in HCT116 (MMR-) tumor xenografts compared with HCT116/3-6 (MMR+) tumor xenografts. Using a tumor regrowth assay after the 14-day drug treatment and a 4-day radiation therapy course (days 11-14 of IPdR), we found substantial radiosensitization in both HCT116 and HCT116/3-6 tumor xenografts. However, the sensitizer enhancement ratio (SER) was substantially higher in HCT116 (MMR-) tumor xenografts (1.48 at 2 Gy per fraction, 1.41 at 4 Gy per fraction), compared with HCT116/3-6 (MMR+) tumor xenografts (1.21 at 2 Gy per fraction, 1.20 at 4 Gy per fraction). No substantial systemic toxicity was found in the treatment groups. CONCLUSIONS: These results suggest that IPdR-mediated radiosensitization can be an effective in vivo approach to treat "drug-resistant" MMR-deficient tumors as well as MMR-proficient tumors.