[CANCER RESEARCH 55, 3721-3725. September l, 1W5] Advances in Brief Evidence for a Connection between the Mismatch Repair System and the G2 Cell Cycle Checkpoint1 Mary T. Hawn, Asad Umar, John M. Carethers, Giancarlo Marra, Thomas A. Kunkel, C. Richard Boland,2 and Minoru Koi Division of Gastroenterology, Department of Internal Medicine, University of Michigan Medical School, Ann Arbor, Michigan 48109-0586 IM. T. H., J. M. €.,G. M., C. R. B., M. K.I; Gastroenterology Section, Veteran's Affairs Medical Center, Ann Arbor, MI 48105-2399 [M. T. H., J. M. C., G. M., C. K. B., M. K.j; and Laboratories of Molecular Carcinogenesis JM. K.j and Molecular Genetics ¡A.U., 7".A. K.j, National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709 Abstract instability has been reported to occur in approximately 10-15% of sporadic colonie neoplasias (2, 3), and has also been linked to a defect The human colon tumor cell line HCT116 is deficient in wild-type in mismatch repair in sporadic colon and endometrial tumor cell lines liMIJII. is defective in mismatch repair (MMR), exhibits microsatellite instability, and is tolerant to JV-methylWV'-nitro-JV-nitrosoguanidine (7-9). Previous work with the human colon cancer cell line HCT116, (MNNG). Transferring a normal copy of li.MI.III on chromosome 3 into which has no normal hMLHl gene product, has shown that this line is the cell line restores MMR activity, stabilizes microsatellite loci, and deficient in MMR (8, 9), exhibits microsatellite instability, and is increases the sensitivity of the cell to MNNG. Previous studies in other cell tolerant to the methylating agent MNNG (7). These studies suggested lines tolerant to alkylating agents such as MNNG or JV-methylnitrosourea that inactivation of both copies of the hMLHl gene leads to complete have shown cross-tolerance to 6-thioguanine (6TG), leading to a hypoth loss of mismatch repair activity, and that such a defect is responsible esis that tolerance to MNNG or 6TG may be the result of MMR deficiency. for the microsatellite instability found in HNPCC and some sporadic To test this hypothesis, we studied the effects of 6TG on the MNNG- tumors. Consistent with this hypothesis, we previously demonstrated tolerant, MMR-deficient IK I IK. cell line and its MNNG-sensitive, that the introduction of one copy of the wild-type hMLHl gene on MMR-proficient, MNNG-tolerant, and MMR-deficient derivatives. Con chromosome 3 into HCT116 cells (HCT116+chr3) restored mismatch tinuous exposure to low doses of 6TG (0.31-1.25 ug/ml) had no apparent effect on colony-forming ability (CFA) in MNNG-tolerant, MMR-deficient repair activity and lowered mutation frequency at a microsatellite cells, whereas MNNG-sensitive, MMR-proficient cells exhibited a dose- locus (7). Tolerance to alkylating agents such as MNNG has been dependent decrease in CFA. Growth kinetics and cell cycle analysis re observed in cell lines (10, 11) that have also been shown to be vealed that the growth of 6TG-treated HCT116+chr3 cells was arrested at deficient in MMR and demonstrate microsatellite instability (9, 12, G2 after exposure to low dose of 6TG. In contrast, the same exposure to 13). Previously, we demonstrated that transfer of chromosome 3 into 6TG did not induce G2 arrest but rather a G, delay in IK"l'I 16 and HCT116 cells diminished the tolerance of the cells to MNNG toxicity. HCT116+chr2. To obtain further evidence for the role of MMR on 6TG After MNNG treatment, these cells exhibit prolonged growth arrest at and MNNG toxicity, we isolated an MNNG-resistant revertan! clone, M2, G2, followed by eventual cell death (7), suggesting that the mismatch from the MNNG-sensitive, MMR-proficient HCT116+chr3 cell line and repair system is involved in mediating MNNG toxicity. characterized the MMR activity, hMLHl status, and 6TG response. The Alkylating agent-resistant cells have been shown to be cross-toler results showed that M2 cells lost MMR activity as well as the previously ant to the guanine base analogue 6TG (11, 14, 15). O''-methylguanine, introduced normal hMLHl gene. Restoration of the CFA of M2 and an the major methylation product from MNNG adduci formation (12), absence of G2 arrest were observed after treatment with low doses of 6TG. has a similar molecular volume to 6TG (16). Both 6TG and 0"- These results suggest that the mismatch repair system interacts with the G2 checkpoint in response to 6TG or MNNG-induced DNA lesions. The methylguanine are unable to form stable base pairs with either of the results further suggest that any agent that induces DNA mispairs will pyrimidines (16). Several investigators have hypothesized that toler cause G2 arrest in MMR-proficient cells but not in MMR-deficient cells. ance to 6TG and O6-methylguanine may be the result of a defect in the MMR system (11, 13, 15). Introduction To test this hypothesis, we examined whether MMR-deficient, HNPCC3 tumors exhibit a high rate of mutation at microsatellite MNNG-tolerant HCT116 cells were also tolerant to low doses of 6TG and whether MMR-proficient, MNNG-sensitive HCT116 containing sequences (1-3). This phenotype has been linked to an inherited the transferred chromosome 3 (HCT116+chr3-6) were sensitive to the germline mutation in one of the MMR genes, hMLHl, hMSH2, same doses of 6TG. In addition, we isolated and characterized an hPMSl, or HPMS2, in which a subsequent somatic mutation in the MNNG-resistant revertant clone from HCT116+chr3-6 cells. Using wild-type alíeleoccurs before tumor formation (4-6). Microsatellite these HCT116 cell line derivatives, we then addressed whether the MMR system mediates 6TG toxicity by examining the colony-form Received 5/31/95; accepted 7/21/95. The costs of publication of this article were defrayed in part by the payment of page ing ability and cell cycle progression after treatment with 6TG. charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Materials and Methods 1This work was partly supported by NIH Grant CA39233, the University of Michigan Cancer Center Grant CA46592, the Johnson Family Fund for Familial Colorectal Cancer, Chemical Reagents. 6TG and SAG (Sigma Chemical Co., St. Louis. MO) National Cancer Institute Grant R25 CA57716, the James Lind Scholarship, and the were dissolved in 0.1 N NaOH and stored at -10°C. MNNG (Aldrich Chemical National Institute of Environmental Health Sciences. 2 To whom requests for reprints should be addressed, at University of Michigan Co., Milwaukee, WI) was dissolved in DMSO (Sigma) and stored at -10°C. Medical Center, Department of Internal Medicine, 4410 Keesge III, 200 Zina Pitcher Cell Culture. The human colon cancer cell line HCT116 was maintained in Place, Ann Arbor, MI 48109-0586. IMDM (GIBCO-BRL, Grand Island, NY) containing 10% FBS. HCT116 + hu 1The abbreviations used are: HNPCC, hereditary nonpolyposis colon cancer; MMR, mismatch repair; MNNG,-/V-methyl-N'-nitro-W-nitrosoguanidine; 6TG, 6-thioguanine; man chromosome 2, clone 1 (HCT116+chr2-l), HCT116 + human chromo SAG, 8-azaguanine; IMDM, Iscove's modified Dulbecco's medium; FBS, fetal bovine some 2, clone 3 (HCT116+chr2-3), HCT116 + human chromosome 3, clone 5 serum; hprl, hypoxanthine phosphoribosyl transferase; HAT, hypoxanthine aminopterin (HCT116+chr3-5), and HCT116 + human chromosome 3, clone 6 thymidine; SSCP, single-strand conformational polymorphism. (HCT116+chr3-6) represent cell lines modified by the transfer of a single 3721 Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1995 American Association for Cancer Research. MISMATCH RKPAIR AND G, CELL CYCLE CHECKPOINT normal human chromosome as described previously (7). These cell cultures exclusion. A nuclear prep was obtained as follows: cells were washed in PBS, were maintained in IMDM containing 10% FBS and G418 (400 /J.g/ml; resuspended (10s cells/ml) in nuclear lysis buffer [10 mM Tris (pH 7.5) 0.32 M GIBCO-BRL). The MNNG-resistant revertan! cell line derived from sucrose-3 mM MgCl2-2 mM CaClr().2% (v/v) NP40] and incubated on ice for HCT116+chr3-6, HCT116+chr3-6-M2, was maintained in IMDM containing 10 min. The tubes were centrifuged at 3000 rpm for 10 min. The nuclei were 10% FBS and G418. HAT medium (Sigma) was dissolved in IMDM contain then washed with nuclear lysis buffer (4 X H)5 cells/ml) without NP40 and ing 10% FBS at a final concentration of KM) /AM hypoxanthine, 0.4 JAM centrifuged for 10 min. Nuclei were stained with propidium iodine (5 /Ag/ml) aminopterin, and In JAMthymidine. in resuspension buffer [O.I MTris (pH 7.5)-0.15 M NaCl-1 mM CaCM).5 HIM 6TG Cytotoxicity Assay. Exponentially growing cell lines were plated in MgClrO.()l% (v/v) NP40] for 2-6 h. Cell cycle analysis was performed, and duplicate at densities of \(i2 to IO4 on 6-cm plates and IO5 to 10* on 10-cm percent of cells in each phase was determined using a Coulter Epics C flow plates. After allowing for attachment to the plate for 18-24 h, the medium was cytometer (Hialeah, FL). replaced with fresh medium containing 6TG (0.31-10 ng/ml). Cells were Mismatch Repair Activity. The efficiency of cytoplasmic extracts in maintained in 6TG-containing medium for 10 days, changing the medium repairing DNA mismatches was performed as follows. Cell-free extracts were every 3 days. The plates were washed with PBS, fixed with methanol, and prepared as described previously (10). Procedures for mismatch repair have stained with 3% Giemsa. Colonies with greater than 50 growing cells were been described (9).