Cockayne's syndrome

ABSTRACT

An ongoing study in our laboratories is to examine the relationship of DNA repair defects to human cancer. Our underlying hypothesis has been that human tumors may arise that lack interesting DNA repair pathways if these pathways are important in preventing cancer. In this study, we found that the UV-irradiated adenoviruses showed hypersensitivity when assayed on monolayers of certain human colon tumor cell lines, including three that are reported to have defects in long patch DNA mismatch repair genes and one with no reported defect in mismatch repair. The survival curves showed two components. The first sensitive component was characteristic of 77-95% of the infections depending upon the cell line and the experiment and had an average slope indicating 4.8fold hypersensitivity to UV. The average of the second-- component slopes indicated that the remainder of the infections was accompanied by near-normal repair. Although the value of the first component indicated that the colon tumor lines supported the growth of UV-damaged adenoviruses poorly, the cell lines themselves showed the same post-UV colony-forming ability as did normal human fibroblasts, and their ability to support the growth of N-methyl-N'-nitro-N-nitrosoguanidine-- damaged adenoviruses was normal, Le. it parallelled their ability to repair O^sup 6^-methylguanine in vitro. We previously observed two-component survival curves when assaying UV-irradiated adenovirus on monolayers of all of seven strains of fibroblasts from Cockayne's syndrome patients. By contrast, single-component curves have been obtained using 21 strains of normal human fibroblasts and seven other tumor lines. We interpret the two-com

ponent survival curves in terms of the defective transcription-coupled repair of UV-induced DNA damage that is characteristic both of Cockayne's and certain colon tumor cell lines. In addition, four mismatch repair-- deficient colon tumor lines were resistant to killing by elevated levels of dG.

Abbreviations: AdMLP, adenovirus major late promoter; CS, Cockayne syndrome; D^sub 37^, dose that reduces survival to 37% on the straight line portion of the curve; dG, deoxyguanosine; DMEM, Dulbecco's modified Eagle medium; DMEM-dil, DMEM diluted with fetal bovine serum and antibiotics; ERCC, excision repair cross complementing; GTBP, G:T DNA mismatch binding protein; hMLH, human homologue of the yeast mutH gene; hMSH, a human homologue of a yeast Mut S gene; hPMS, a human homologue of a yeast postmeiotic segregation gene; Mer, deficient in ability to support the plaquing by MNNG-treated adenoviruses; Mer, as proficient as normal human fibroblasts to support plaquing by MNNG-treated adenovirus; Mfd, mutation frequency decline; MGMT, O^sup 6^-methylguanine-DNA methyltransferase; MNNG, N-methyl-N'-nitro-N-nitrosoguanidine-; MNU, methylnitrosourea nt, nucleotide; PBS, phosphate-buffered saline; RH, relative humidity; 6TG, 6-thioguanine; TCR, transcription-- coupled repair; XP, xeroderma pigmentosum.

INTRODUCTION

The plaque-forming ability of adenoviruses that are damaged prior to infection by photons or chemicals often depends upon the DNA repair capacity of the host cells used for the assay. For example, UV-damaged adenoviruses produce far more plaques when assayed on monolayers of normal human fibroblasts than when assayed on monolayers of fibroblasts from patients with xeroderma pigmentosum (XP) sec that are defective in repair of UV-damaged DNA (1-3). Similarly, N-methyl-N'-nitro-N-nitrosoguanidine- (MNNG)- or MNU-- treated adenoviruses produce many more plaques on monolayers of human fibroblasts, tumor or SV40-transformed cell lines that are able to repair O6-methylguanine than on such lines that are deficient in such repair (4,5).

When adenoviruses are damaged either by UV or MNNG, the shapes of the viral survival curves obtained after assay on normal fibroblasts are linear (or show slight shoulders) on plots of log survival versus linear dose. Normal repair of UV-damaged adenovirus likely depends upon cellular capacity to perform preferential repair of the transcribed DNA strand (transcription-coupled repair or TCR) because the survival of UV-irradiated adenovirus was reduced in Cockayne's syndrome (CS) cell strains (6) that are defective in TCR (7,8). When assayed on any of seven strains of fibroblasts from persons with CS, the UV-treated virus showed uncharacteristic two-component survival curves (6). The initial, rapidly inactivated component (slope not determined) was followed by a second more resistant component (D^sub 37^ = 209 +/- 45 [SD] J/m^sup 2^) that extrapolated back to 23.4 +/- 12.3 (SD)% of the infected cells. Cloning studies showed that the sensitive and resistant portions of the curves did not arise from heterogeneity within either the cellular or viral populations, and survival curves of plaque-forming ability on 10 control fibroblast strains showed a single component having a D^sub 37^ = 234 +/- 40 [SD] J/m^sup 2^.

Since 1981, when we published our results obtained with CS cell strains, much has been learned about the underlying biology of the syndrome. The ERCC6 or CSB gene that complements the CS complementation group B defect (9) is involved in preferential repair of the transcribed DNA strand (7-9). The CSB gene encodes seven domains that are conserved among helicases involved in maintaining transcription regulation, chromosome stability and DNA repair. It was thought that CSB may act in TCR much as the Mfd (mutation frequency decline) protein does in Escherichia coli, i.e. to dissociate RNA polymerase complexes stalled at DNA lesions in a damaged transcribed strand (10), thereby facilitating repair in that strand. However, no helicase activity has been detected and, while able to bind to RNA polymerase II, CSB does not release the stalled RNA polymerase from its template ( 11 ). Like the CSB product, both the ERCC2 and ERCC3 products, subunits of the basal transcription factor TFIIH (12,13), play roles in repairing UV-damaged adenoviruses: ERCC2 is defective in XP complementation group D (12) and ERCC3 is defective in a cell strain from an XP group B patient who was also diagnosed with CS (13). The product of the gene mutant in CS complementation group A lines, CSA, interacts both with the CSB protein and with p44, a subunit of TFIIH (14), suggesting that a CS complex is sometimes physically coupled to TFIIH. Extracts from cells of both groups afford reduced levels of RNA polymerase II-dependent transcription in vitro (15). Addition of CSB protein to a reconstituted RNA polymerase II-dependent in vitro transcription system enhanced the rate of elongation from the adenovirus major late promoter (AdMLP) about three-fold (16), suggesting a direct role of CSB in transcription. However, CS cells do not grow slowly in vitro nor does adenovirus 5 produce small plaques on monolayers of CS (A or B) cells (Day, unpublished observations). Thus either basal transcription is not rate limiting to cell cycling or adenoviral growth or, under in vitro conditions, CSB may be one of many promoter-specific transcriptional activators (operating on the AdMLP at least) rather than a basal initiation factor. The function of CSA is not yet known.

Long patch DNA base mismatch repair has also been implicated in preferential repair of UV damage in the transcribed DNA strand (TCR). In support of this connection, both the LoVo and HCT-116 cell lines used in this study are deficient in TCR (17) as well as in long patch mismatch repair (18). Like TCR, long patch DNA mismatch repair involves cellular discrimination of DNA strand identity. In E. coli, long patch mismatch repair is directed by the state of methylation of GATC sites to errors caused by faulty DNA replication. Newly replicated daughter strands containing the errors are first identified by their transient lack of methylation at GATC sites, then removed and replaced during long patch mismatch repair (19). (In human cells the mechanism for strand discrimination is not known, but a single-strand DNA break directs removal of that strand.) Mismatch repair in E. coli begins by the ATP-dependent binding of the mutS gene product to the DNA base mismatch, followed by the incorporation into the complex of the mutL and mutH products and the concomitant conversion of ATP to ADP. The mutH subunit incises the nonmethylated daughter strand at the hemimethylated GATC site either 5' or 3' to the mismatch. Exonuclease activity (5' --> 3' or 3' -->5') and helicase II activity produce a single-stranded gap by removing daughter-strand DNA between the incised GATC site and 50-100 nucleotides (nt) past the mismatched base pair (20). To complete repair, DNA polymerase III and ligase fill and seal the gap (19). There are several mutS homologs in humans cells, hMSH2 (21,22), hMSH3 (23,24) and GTBP (or hMSH6) (25,26), whose gene products form heterodimers (hMSH2:GTBP (25,26) or hMSH2:hMSH3 (27,28)). The human hMLH1, hPMS1 and hPMS2 genes are homologs of the E. coli mutL and yeast PMSI and PMS2 postmeiotic segregation genes (29-34). Human tumors mutant in the human hMSH2 (21,22), hMSH3 (35,36), hMLH1 (29,30), hPMS1, hPMS2 (31) and GTBP (37) genes have been identified primarily from patients with hereditary nonpolyposis colon cancer, and cell lines mutant in hMSH2, hMLH1 and GTBP lack DNA mismatch repair (18,38,39). Here we report that UV-irradiated adenoviruses show two-- component survival, much as in CS fibroblasts, when infecting three cell lines, LoVo, DLD- 1 and HCT- 116, having defects in hMSH2, GTBP and hMLH1, respectively. In addition, another colon carcinoma cell line SW620 with no identified mismatch repair defect also gives rise to a two-- component inactivation curve.

MATERIALS AND METHODS

Cell culture and media. Cell strains CRL1199 (XP11BE; XP complementation group B), CRL1204 (XP10BE; XP complementation group C) and CRL1223 (XP12BE; XP complementation group A) were obtained from the American Type Culture Collection, Rockville, MD. Cockayne syndrome cell strains GM1856 (CS complementation group A), as well as GM 1629 and GM 1428 (complementation group B), were obtained from the NIGMS Human Genetic Mutant Cell Repository, Camden, NJ. The CS strain Andre was from a biopsy supplied by Dr. Peter Nissley, NIH, Bethesda, Maryland. The KD normal fibroblast strain was established in our laboratory from the lip of a 30 year-old female (2). The M049 line is a cell line established in our laboratory from a human malignant glioma. The SW620, HCT-116, DLD-1, HCT15 and LoVo are colon carcinoma cell lines from the American Type Culture Collection. The DLD-1 and HCT]5 lines are described (40) and are likely from the same patient (41). The first three were kind gifts of Dr. Mark Meuth. These are listed in Table 1 along with summarized results of DNA base mismatch repair-related studies obtained with them. The A427 and A549 human lung carcinoma cell lines (42) were obtained over 20 years ago from the program of Dr. Walter Nelson-Rees supported by the NCI Viral Oncology Program and under the auspices of the Office of Naval Research and the Regents of the University of California. The A1235 line is a human malignant glioma cell line kindly supplied by Dr. S. A. Aaronson, NIH, Bethesda, MD. Cells were cultured in two types of culture medium. The first was 900 mL Dulbecco's modified Eagle medium (DMEM, Gibco cat. no. 4301600EH) plus 90 mL fetal bovine serum plus 9 mL antibiotic mixture (Gibco, cat. no. 600-5140AG; containing 10000 units penicillin plus 10 mg streptomycin per mL). The second, F12 for UV survival curves, contained 900 mL F12 (Gibco cat. no. 21700) plus 90 mL fetal bovine serum plus 9 mL antibiotic mixture. The F12 for the deoxyguanosine (dG) experiments contained 900 mL F12-HAT variation (Irvine Scientific cat. no. 9447; lacking folic acid, hypoxanthine and thymidine). Cells were cultured at 36-37C, 80-90% relative humidity (RH) and either 8-9% CO2 (DMEM) or 5% CO2 (F12). The DMEM-dil contained 900 mL DMEM plus 9 mL fetal bovine serum plus 9 mL antibiotic mixture.

Adenovirus plaque assay, UV irradiation and MNNG-treatment.

Adenovirus 5, freshly plaque-purified on monolayers of the A427 cell line, was propagated on monolayers of the A549 cell line and purified by banding in CsCI as described (43,44). Purified preparations were stored as 50 (mu)L aliquots in liquid nitrogen in 70% 0.01 M tris-HCI, pH 8.0 and 30% glycerol. For inactivation with UV irradiation, viruses were diluted 1:1000 in phosphate-buffered saline (PBS) and distributed in 2 mL portions into 60 mm culture dishes and irradiated under an unfiltered germicidal bulb at room temperature at ~1 W/m^sup 2^ for periods as long as 20 min. For inactivation with MNNG, viruses were diluted 1:1000 in 0.3 M tris-HCl, pH 9.0, and distributed in 0.9 mL portions into 5-1.5 mL eppendorf tubes. To each tube, 0.1 mL of an MNNG stock (prepared at 0, 2.5, 5.0, 7.5 and 10 mg MNNG/mL ethanol) was added and the viruses treated at 37 deg C for 30 min. N-acetyl-L-cysteine (0.1 mL of 0.5 M stock, pH 7.0 in water) was added, the tubes were incubated at room temperature for 15 min to inactivate unreacted MNNG and the treated viruses were diluted 1:10 in DMEM-dil. After UV or MNNG treatment, virus samples were further diluted in 10-fold steps in DMEM-- dil and assayed for plaque formation. In the assay, 0.3 mL aliquots of the diluted viruses were adsorbed to duplicate monolayers of cells (propagated in DMEM in 60 mm culture dishes) for 2 h. The medium was aspirated from the monolayers and the monolayers were overlayed with 5 mL of a DMEM-based nutrient agar (43) that confined viral growth to the areas of the initially infected cells. Incubation was at 36.5-37 deg C, 8-9% CO2 and 75-80% RH. Another 8 mL agar was added to each plate on the day following infection, and on days 12-19 the plates were overlayed with 4 mL agar containing 0.005% neutral red. Plaques, visualized the next day as clear areas against a red background of stained metabolically active cells were then enumerated. In several experiments, F12 was substituted for DMEM in the preparation of agar. When this was done, the cell monolayers to be infected were also propagated in F12. In certain experiments, prior to infection with adenoviruses, cell monolayers were irradiated. In this case, medium was aspirated, the cells were washed once with PBS, the PBS was aspirated and the culture dishes were irradiated in groups of four. As soon as possible, 0.3 mL of diluted viruses was pipetted onto each plate. The remainder of the assay followed the outline described previously.

Colony-forming ability. Stock cultures of each cell line, maintained in F12, were subcultured to 100 mm culture dishes in F12 medium at densities from 100 to 10000 per dish. On the day following, the medium was aspirated from the plates that were then washed once with PBS. After removing the PBS, the cells were subjected to UV irradiation under the conditions described above using three replicate plates at each cell density and at each UV dose. Fresh F12 medium (10-20 mL) was added to each plate and the plates were incubated at 36.5 deg C, 85% RH. The medium was replaced with 10 mL fresh medium after 1 week (in the case of 10 mL/plate) or 2 weeks (in the case of 20 mL/plate). (For certain cell lines, notably SW620, 20 mL medium was supplied for a period of 2 weeks in an effort to minimize formation of satellite colonies.) Colonies were enumerated after 3 weeks of culture by staining with crystal violet. Several experiments were carried out in DMEM. In this case cells were cultured in DMEM at 36.5 deg C, 8-9% CO2 and 85% RH both for a week prior to UV irradiation (and during colony formation afterward).

O^sup 6^methylguanine-DNA methyltransferase (MGMT) assay. The MGMT content of the various cell lines was assayed as described previously (45). Briefly, cells propagated in DMEM were harvested after trypsinization and washed with PBS. Cell-free extracts (about 10 mg protein/mL) were prepared by the method of Manley et al. (46). A synthetic 45 bp DNA, in which replacement of the 5' G in a 5'-CCGG-3' site by O^sup 6^methylguanine blocked the endonuclease activity of HpaII, was used as a test substrate for MGMT activity. It was prepared 5'-^sup 32^P-labeled in the O^sup 6^-methylguanine-containing strand. After incubation of extract with substrate, repurification of the DNA, and HpaI treatment to digest any repaired DNA (with demethylated O^sup 6^-methylguanine), reaction products were subjected to electrophoresis through 12% denaturing polyacrylamide gels. Quantitation of band intensities was performed after scanning autoradiograms with an LKB Ultroscan XL laser densitometer.

RESULTS

Survival of UV-irradiated adenovirus

We obtained a series of UV survival curves with the purified adenovirus 5 preparation. In each experiment, the virus was diluted into PBS, irradiated and assayed for remaining plaque formation on monolayers of human colon carcinoma cell lines (Table 1) as well as control cell lines and cell strains (lines and strains as described in the Materials and Methods and in Table 1). Representative curves for seven strains are shown in Fig. 1. (A summary of the results of all the plaque assay experiments is presented in Table 2.) In Fig. 1, the control strains and KD (open squares), GM38 (open diamonds) and A1235 (closed circles) lines afforded one-component survival curves, within error of measurement (Table 2), as we have observed previously for the survival of UV-irradiated adenoviruses assayed on monolayers of 21 normal human fibroblast strains and seven human tumor cell lines (2,4,6,47-49). The curves obtained for DLD-1 (open circles; altered in GTBP) and LoVo (closed triangles; altered in hMSH2) showed multicomponent behavior (referred to from here onward as two-component for convenience). A third colon carcinoma cell line SW620 (closed squares), that behaves as though wild type with respect to DNA mismatch repair, also showed pronounced two-component behavior. Survival on monolayers of an XP complementation group C cell strain, XP1OBE (open triangles), is shown for comparison.

We performed two large replicate experiments to measure the two inactivation slopes (k^sub 1^ and k^sub 2^) on LoVo, DLD-1, SW620 and HCT-116 monolayers, as well as the percentage of the population giving rise to the second component (Fig. 2). The inactivation slopes measured on the control (KD, PF1, GM38 and A1235) monolayers were very similar, averaging 0.0063 +/- 0.0002. (The percentage of the population calculated to be involved in the single-component inactivation observed on the KD, PF-1, GM38 and A1235 control lines was 98 +/- 2.) By comparison, in the case of the LoVo, DLD-1, SW620 and HCT-116 colon carcinoma cell lines, we judge (by extrapolation to zero dose) that between 5.5 and 23% (average 13 +/- 7%) of the infectious events were accompanied by a level of repair indistinguishable from that of normal: the average of their more resistant components was 0.0058 +/- 0.0006. The remaining 77-95% of the infections in the LoVo, DLD-1, SW620 and HCT-116 carcinoma lines yielded an average inactivation slope k^sub 1^ = 0.028 +/0.009, 4.8-fold steeper than the resistant component.

To test whether the two-component behavior was a phenomenon caused by two populations of cells having different capacities to support the growth of UV-irradiated adenoviruses, we prepared two clones of DLD-1, DLD-1 Cl.1 and DLD-1 Cl.2. The use of cultures of both clones as monolayers for assaying the survival of UV-irradiated adenovirus gave rise to two components (see Table 2). This supports the contention that the two-component behavior is not a phenomenon caused by two populations of cells having different capacities to support the growth of UV-irradiated adenoviruses.

We wondered whether the agar itself might cause the two-- component behavior. Experiments in which DLD-1 monolayers infected with viruses to be assayed were not overlayed with directly with agar, but rather with DMEM containing 10% fetal calf serum for 24 h prior to being overlayed with agar, also gave rise to two-component curves (data not shown). No progeny adenovirus 5 are liberated from the cells during this time and repair is presumably complete at 24 h because repair is no longer inhibitable by hydroxyurea (Day, unpublished results). Similarly, experiments in which the overlaying agar lacked components (tryptose phosphate broth, MgCI2) that are not present in DMEM did not increase survival of UV-treated virus in DLD-1 cells (data not shown).

Survival of UV-irradiated cell lines

We analyzed post-UV colony-forming ability of the cell lines because the average 4.8-fold difference in sensitivities of two components observed in Fig. 2 suggested that a significant percentage of the cells (averaging 87%) might be defective in DNA repair and possibly sensitive to killing by UV radiation. Stocks of cell lines growing in F12 medium were subcultured in F12 at densities of 100, 1000 and 10000 per 100 mm culture dish and allowed to attach overnight. These were irradiated and cultured 2-3 weeks to form colonies. Figure 3 shows that the curves describing the post-- UV survival of the DLD-1, LoVo, HCT15, HCT-116 and SW620 colon carcinoma cell lines were not detectably different from those of the control KD fibroblast strain or the A1235 malignant glioma cell line. All of these are significantly more resistant than XP10BE, XP11BE or XP12BE that belong to XP complementation groups C, B and A, respectively. The XP strains serve as controls that simply ensure that cells known to be defective in repairing UV damage were indeed detectably more sensitive to UV than the KD normal fibroblasts. All curves were performed twice, each with three replicate plates per datum point. Table 3 presents an analysis of the curves and associated errors of measurement.

We reasoned that the different media used might have caused LoVo and DLD-1 to give rise to UV-sensitive behavior in the virus assay, in which DMEM was used, and to UV-resistant behavior in the colony-forming assay in which F12 was used. We found (Fig. 4), however, that the UV survival curves obtained in DMEM (solid symbols) both with three lines defective in mismatch repair and in the control KD fibroblast strain were not experimentally distinguishable from those obtained in F12. Table 4 presents an analysis of the curves as multitarget survival curves along with estimates of error of measurement.

Survival of UV-irradiated viruses in UV-irradiated cells

Again seeking to explain disparate sensitivities shown to UV by LoVo, DLD-1, HCT-116 and SW620 in the virus assay and in the colony-forming assays, we tested whether UV irradiating the DLD-1 cell monolayers would increase their ability to support the growth of UV-irradiated viruses. Monolayers of DLD-1 were washed once with PBS and irradiated with 0, 1, 3, 10, 30 and 100 J/m^sup 2^ of UV immediately prior to assaying the survival of adenoviruses that had received 400 J/m2 of UV (Fig. 5). The survival of plaque-- forming ability increased up to 3-fold at the dose of 10 J/ m^sup 2^, but not nearly enough to account for the 50-fold difference in virus survival noted between normal and DLD-1 cells (Fig. 2). The monolayers irradiated with 30 and 100 J/ m^sup 2^ of UV failed to take up neutral red (died), making measurements of plaque-forming ability impossible. To test the possibility that we did not allow sufficient time for UV induction of repair processes, we repeated the 0 and 10 J/m^sup 2^ points after 18 hr (or 0 h) of post-UV incubation. Irradiating the cells with UV again produced very little change in virus survival in cells incubated either 0 h or 18 h after irradiation.

Sensitivity of cell lines to dG

Cockayne's syndrome strains and lines have been reported to be hypersensitive to killing by dG (50). Because UVirradiated adenovirus showed two-component survival when infecting either CS fibroblasts (6) or the colon carcinoma cell lines (this report), we tested whether the colon carcinoma lines might also show sensitivity to inactivation of colony-forming ability by dG. Cell lines and strains were cultured for 2 weeks in F12 medium lacking hypoxanthine, folic acid and thymidine but containing 0, 10, 25, 50, 75 and 100 FM dG, after which colonies were enumerated. Figure 6A shows that the DLD-1, LoVo, HCT15 and HCT-116 cells lines were more resistant, not more sensitive, to killing by dG than the KD or GM38 normal human fibroblasts, the A1235 malignant glioma cell line and the SW620 line, which is proficient in long patch DNA mismatch repair (51). We investigated further and performed some experiments with CS fibroblast strains, GM1629, GM1428, GM1856 and Andre, as well as DLD-I and the two clones of DLD-I (Fig. 6B). Unexpectedly, the sensitivities of four strains of fibroblasts from CS patients were not different from those of the normal fibroblasts within experimental error. But the two clones of DLD-1 were as resistant as the uncloned DLD-1 cell line. We reasoned that defective HPRT genes in the DLD-I, LoVo, HCT 15 and HCT-116 lines might have rendered them resistant to dG, in which case they would be resistant to 6-thioguanine (6TG). We cultured the four mismatch repair-deficient cell lines in F12 medium lacking hypoxanthine, folic acid and thymidine but containing 30 FM 6TG (the concentration routinely used to select for mutants in the HPRT gene by many laboratories including ours). All four cell lines were sensitively killed by 30 FM 6TG (data not shown). Thus, it is unlikely that a lack of hypoxanthine phosphoribosyl transferase from the long patch mismatch repair-deficient cells causes their elevated survival in the presence of high concentrations of dG.

Ability of the cells to repair virus damaged by MNNG

To determine whether MNNG-treated adenovirus, in addition to UV-treated virus, might show two-component survival on the lines, we treated the virus with MNNG and assayed its survival on monolayers of DLD-1, SW620, HCT116 and a known Mer- line A1235 (5). Lines defective in supporting the growth of MNNG-treated virus, but not non-- damaged virus, are known as Mer- lines (5). The survival curves (Fig. 7) were as expected for Mer+ and Mer- cell lines, that are proficient and deficient in the ability to repair O^sup 6^-methylguanine (5) and contain or lack MGMT (52), respectively. To determine whether the cell lines contained the levels of MGMT expected on the basis of their Mer phenotype, we tested their ability to demethylate O^sup 6^-methylguanine contained in a synthetic 45 bp substrate. The results showed that Mert cell lines contained MGMT, and Mer cell lines lacked it, much as we found previously (52). Among Mer+ lines, HT29, a Mer+ control, contained 330 fmol MGMT/mg protein; LoVo, 400; HCT-116, 66; DLD1, 82 and HCT15, 200. The Mer lines A172 and SW620 contained 0 fmol MGMT/mg protein within experimental error. That is, the low survival of MNNG-treated adenovirus on SW620 monolayers can be explained by the lack of MGMT in this line. Thus, whatever genetic alteration is responsible for defective repair of UV-damaged adenovirus in Lovo, HCT-116, DLD-1, HCT-15 and SW620 does not also cause aberrant repair of MNNG-damaged adenovirus.

DISCUSSION

We found that three human colon carcinoma cell lines, DLD-1, LoVo and HCT-116 (previously reported to have mutant GTBP, hMSH2 and hMLH1 genes, respectively, and to lack long patch mismatch repair) as well as one colon tumor cell line SW620 (likely wild type in such repair) fail to restore infectivity to UV-damaged adenoviruses normally. Yet, they show normal colony formation after being UV irradiated. This normal survival does not appear to be caused by the UV induction of repressed DNA repair processes as judged by the fact that UV irradiating the cells did not increase the survival of the UV-treated adenovirus. The percent survival did increase (the absolute survival did not) but not nearly enough to explain the lack of virus repair in the repair-defective colon carcinoma cell lines. Whether infection of the cells was performed immediately after UV or 18 h after UV made no experimentally detectable difference in the UV enhancement in survival of the UV-treated virus. Table 2 shows that the standard errors of the percentages of the infected population that gave rise to the resistant components were greater (relative to the corresponding resistant fraction of the cells) in the cell lines giving rise to two-- component survival curves than in the cell lines giving rise to one-component curves. We suggest that the greater variation is not artifactual, but that it is caused by actual experiment-to-experiment variation in the percentage of the infected cellular population that gave rise to the second component. Supporting the idea that such fluctuations in DNA repair may exist among the cellular population is the fact that Richards et al. (53) determined that hMSH2-deficient lines show elevated spontaneous mutation frequencies only when maintained at high cell densities. Our monolayers were plated at ~75% confluency and allowed to grow for 2-4 days prior to infection. The biphasic viral response to UV observed in the colon tumor cell lines is unlikely to be caused by inhomogeneity in the virus population because the virus preparation was freshly cloned and because no evidence of high resistance of a second component was observed when the virus was assayed on monolayers of the XP10BE fibroblast strain.

The average D^sub 37^ (=1/K^sub 1^) of the sensitive components of the viral inactivation curve as presented in Table 2 is 37 +/15 J/m^sup 2^. This is in the range of values previously obtained (1,2) with four XP complementation group C fibroblast strains (range 31-78 J/m^sup 2^), outside the range of seven group A and D strains (range 7.4-15 J/m^sup 2^) and within experimental error of an XP group B strain (25 J/m^sup 2^). While providing an estimate of the magnitude of the repair defect of the colon tumor lines that was detected by the adenovirus assay, the significance of this comparison to interpreting the two-component viral survival curves is not clear.

Changing the post-UV colony-forming medium from F12 to DMEM, the medium used in the plaque assay, had no effect on post-UV cellular reproductive survival. We judge this to mean that the DMEM used for the plaque assay does not, by itself, cause reduced survival. The result that SW620, believed to be wild type with respect to mismatch repair, was deficient in supporting the growth of UV-treated adenovirus indicates that the inability of DLD-I, LoVo and HCT-116 to support the growth of UV-treated adenoviruses normally may not result from a lack of mismatch repair function per se but rather from a lack of another process requiring mismatch repair gene products. That is, the loss of long patch DNA mismatch repair may be one of several mechanisms whereby cells may lose the ability to repair UVtreated adenovirus normally. As noted, CS cells, all of which give rise to two-component adenovirus survival curves similar to those observed here, are defective in preferential repair of the transcribed DNA strand (8) and have mutations in CSB or CSA (9,14). The observation that both LoVo and HCT-116 also lack TCR (17) supports the idea that the two-- component adenovirus survival curves are more generally related to defective host-cell TCR.

How might loss of TCR be involved in generating two-- component adenoviral survival curves? Our results lead us to conclude that two classes of events must befall the infecting UV-treated virions, one class linked to the small amount of DNA repair associated with the first component and the other to the nearly normal repair associated with the second component. Clearly any model in which the presence of a TCR factor would increase the repairing population from 5-23% to 100% would account for our data. Two simple explanations involve consideration of (1) the cell cycle and (2) the biochemistry of adenovirus infection. First, cell cycle parameters may determine whether or not the virus will survive the UV damage: UV-treated cells may undergo an obligatory survival-enhancing event while they cycle in preparation for colony formation, explaining their normal survival after UV. But virus survival may have a low probability of being enhanced if the stage of the virus infection that is susceptible to cellular survival-enhancing biochemistry occurs at an unfavourable time of the cell cycle. Secondly, the virus biochemistry may not favor post-UV survival in the absence of TCR factors. A UV-treated virion has no ongoing transcription, so it is not clear how adenovirus DNA would be susceptible to TCR. To make a plaque, the immediate early EIA gene of an infecting virion must be transcribed and the ElA products must stimulate transcription of EIA and the early genes EIB, E2A, E2B, E3 and E4 (54). Most of the transcripts must be expressed: while the E3 region is not required for plaque formation, the rest are. The long primary E2B transcript encodes the adenovirus DNA synthetic machinery including the adenovirus DNA polymerase. One might envision the following scenario: general transcription factors initiate RNA synthesis at one of the adenovirus early promoters, and during elongation, the RNA polymerase encounters a UV photoproduct. In the absence of TCR, the RNA polymerase is permanently stalled, the infection is terminated and no plaque forms. This kind of event would contribute to the sensitive component of the survival curve. However, should loading of a DNA repair complex onto the same promoter occur before loading of the RNA polymerase, then the adenoviral DNA would be repaired and viral infection could proceed. Such events would contribute to the resistant component of the survival curve. In the presence of normal TCR, the function of TCR factors, then, would be to coordinate the loading/unloading of repair and transcription complexes onto/from damaged DNA templates. This possibility is certainly in the realm of the hypotheses that account for TCR function in human cells.

The transcription factor TFIIH is a link between transcription and nucleotide excision repair. TFIIH is comprised of 9 proteins, including the XPB and XPD DNA nucleotide excision repair gene products (12,13). Core TFIIH (from yeast) is reported to exist in at least two forms, (1) as the transcriptionally active holo-TFIIH and (2) as a complex with other nucleotide excision repair proteins (termed a nucleotide excision repairosome and transcriptionally inactive) (55). Different experiments show that about 15% of RNA polymerase II is in association with a complex that contains the CSB protein (56), encouraging the idea that a specific protein complex could be loaded onto the DNA to enhance repair of genes undergoing transcription. In vitro, TFIIH is lost from transcriptionally active RNA polymerase II complexes after the transcript is ~30 nt in length, raising the question of whether repair occurs prior to 30 nt and/or after the 30 nt stretch (57). In a rad7 Delta background, the RAD26 TCR factor is not required for repair of DNA damage within the first -50 nt of DNA that encodes the transcript (58). However, the factor is required to repair the DNA encoding the remainder of the transcript, suggesting the possibility that TCR provides a TFIIH-related repair function to a transcriptionally active RNA polymerase II past ~50 nt, although it is not known whether TFIIH is released from actively transcribing RNA polymerase II complexes in vivo. Thus, a TFIIH-CSB-containing repair factor that loads onto the E1A promoter site or exchanges with transcriptionally active TFIIH to enhance repair is a possibility.

How are mismatch repair proteins involved in assisting virus recovery? The GTBP, hMSH2 and hMLH1 proteins are believed to coordinate the repair of errors in the newly synthesized DNA strand during semiconservative DNA replication (19). They bind to DNA base mismatches and so may comprise part of a damage recognition DNA repair complex during TCR. These proteins may bind to sites of DNA damage in UV-treated cells and, when a transcribing complex encounters the complexed site, may assist in removing transcriptional subunits in preparation for repair (somewhat as envisioned for the Mfd protein (12) or CSB (9)) and/or to recruit/establish an active repair complex at the damaged site.

Why adenovirus should show UV sensitivity in both CS strains and tumor lines lacking long patch mismatch repair, while when the cells are UV irradiated, only the CS strains show sensitivity, is an interesting question. One interpretation for the UV sensitivity of CS cells would be that binding of long patch mismatch repair factors to DNA damage in the absence of CS factors would have some probability of being lethal, while lack of the long patch mismatch repair factors in the presence of the CS factors would not lead to lethality.

The fact that four mismatch repair-defective cell lines are resistant to dG bears further study. Because of the tight correlation between sensitivity to dG and defective mismatch repair, it is tempting to link the two. One could imagine that normal cells might be killed by a DNA mismatch repair response to a (large?) number of errors caused by forced insertion of dGMP at the replication point. Such a response would be lacking in the mismatch repair-defective lines, and they would be resistant to dG but probably mutagenized by it. Our preliminary work shows that while dG is mutagenic in HCT-116, it is not in DLD-1, so the possible connection between long patch DNA mismatch repair and resistance to dG must be viewed with caution. We were unable to confirm the finding that cell lines from CS patients are sensitive to killing by dG (50). The greatest difference in the disparate data sets is that normal fibroblasts in Squires et al. (SO) show greater resistance than our normal lines; their CS lines show the same resistance as both our normal and our CS lines.

The reason for the difference in the two sets of data may be very interesting.

Acknowledgements-We thank Dr. Mark Meuth for stimulating our interest in cell lines defective in long patch DNA mismatch repair. We thank Dr. Roseline Godbout for helpful discussions and encouragement. This study was supported by NCI (US) grant CA49936 and MRC Canada grants MA-12300 and MT-14355 to R.S.D. The authors dedicate this paper to the memory of Professor Raymond Latarjet who pioneered many areas of research, including that represented by our paper, the use of viral probes for studying radiation response. His life and research remain an inspiration to us all.

REFERENCES

1. Day, R. S., III (1974) Cellular reactivation of ultraviolet irradiated human adenovirus 2 in normal and xeroderma pigmentosum fibroblasts. Photochem. Photobiol. 19, 9-14. 2. Day, R. S., III (1974) Studies on repair of adenovirus 2 by human fibroblasts using normal, xeroderma pigmentosum, and xeroderma pigmentosum heterozygous strains. Cancer Res. 34, 1965-1970.

3. Day, R. S., III (1975) Xeroderma pigmentosum variants have

decreased repair of UV-damaged DNA. Nature 253, 748-749. 4. Day, R. S., III and C. H. J. Ziolkowski (1979) Human brain tumor cell strains with deficient host-cell reactivation of N-- methyl-N'-nitro-N-nitrosoguanidine--damaged adenovirus 5. Nature 279, 797-799.

5. Day, R. S., Ill, C. H. J. Ziolkowski, D. A. Scudiero, S. A. Meyer, A. Lubiniecki, A. Girardi, S. M. Galloway and G. D. Bynum (1980) Defective repair of alkylated DNA by human tumor and SV40 transformed human cell strains. Nature 288, 724-727.

6. Day, R. S., III, C. Ziolkowski and M. DiMattina (1981) Decreased host cell reactivation of UV-irradiated adenovirus 5 by fibroblasts from Cockayne's syndrome patients. Photochem. Photobiol. 34, 603-607.

7. Venema, J., L. H. F. Mullenders, A. T. Natarajan, A. A. van Zeeland and L. V. Mayne (1990) The genetic defect in Cockayne's syndrome is associated with a defect in repair of UVinduced DNA damage in transcriptionally active DNA. Proc. Natl. Acad. Sci. USA 87, 47044711.

8. van Hoffen, A., A. T. Natarajan, L. V. Mayne, A. A. van Zeeland, L. H. F. Mullenders and J. Venema (1993) Deficient repair of the transcribed strand of active genes in Cockayne's syndrome. Nucleic Acids Res. 21, 5890-5895. 9. Troelstra, C., A. van Gool, J. de Wit, W. Vermeulen, D. Bootsma and J. H. J. Hoeijmakers (1992) ERCC6, a member of a subfamily of putative helicases, is involved in Cockayne's syndrome and preferential repair of active genes. Cell 71, 939-953. 10. Selby, C. P. and A. Sancar (1993) Molecular mechanism of transcription-repair coupling. Science 260, 53-58. 11. Selby, C. P. and A. Sancar (1997) Human transcription-repair coupling factor CSB/ERCC6 is a DNA-stimulated ATPase but is not a helicase and does not disrupt the ternary transcription complex of stalled RNA polymerase II. J. Biol. Chem. 272, 1885-1890.

12. Schaeffer, L., V. Moncollin, R. Roy, A. Staub, M. Mezzina, A. Sarasin, G. Weeda, J. H. J. Hoeijmakers and J.-M. Egly (1994) The ERCC2/DNA repair protein is associated with the class II BTF2/TFIIH transcription factor. EMBO J. 13, 2388-2392. 13. Schaeffer, L., R. Roy, S. Humbert, V. Moncollin, W. Vermeulen, J. H. J. Hoeijmakers, P. Chambon and J.-M. Egly. (1993) DNA repair helicase: a component of BTF2 (TFIIH) basic transcription factor. Science 260, 58-63.

14. Henning, K. A., L. Li, N. Iyer, L. D. McDaniel, M. S. Reagan, R. Legerski, R. A. Schultz, M. Stefanini, A. R. Lehmann, L. V. Mayne and E. C. Friedberg. (1995) The Cockayne syndrome A gene encodes a WD repeat protein that interacts with CSB protein and a subunit of RNA polymerase II TFIIH. Cell 82, 555564.

15. Dianov, G. L., J. F. Houle, N. Iyer, V. A. Bohr and E. C. Friedberg (1997) Reduced RNA polymerase II transcription in ex

tracts of Cockayne syndrome and xeroderma pigmentosum/ Cockayne syndrome cells. Nucleic Acids Res. 25, 3636-3642. 16. Selby, C. P. and A. Sancar (1997) Cockayne syndrome group B protein enhances elongation by RNA polymerase II. Proc. Natl. Acad. Sci. USA 94, 11205-11209. 17. Mellon, I., D. K. Fajpal, M. Koi, C. R. Bolland and G. N. Champe (1996) Transcription-coupled repair deficiency and mutations in human mismatch repair genes. Science 272, 557-560. 18. Umar, A., J. C. Boyer, D. C. Thomas, D. C. Nguyen, J. I. Risinger, J. Boyd, Y. Ionov, M. Perucho and T. A. Kunkel (1994) Defective mismatch repair in extracts of colorectal and endometrial cancer cell lines exhibiting microsatellite instability. J. Biol. Chem. 269, 14367-14370.

19. Modrich, P. and R. Lahue (1996) Mismatch repair in replication fidelity, genetic recombination, and cancer biology. Annu. Rev. Biochem. 65, 101-133.

20. Grilley, M., J. Griffith and P. Modrich (1993) Bidirectional excision in methyl-directed mismatch repair. J. Biol. Chem. 268, 1183011837.

21. Fishel, R., M. K. Lescoe, M. R. S. Rao, N. G. Copeland, N. A. Jenkins, J. Garber, M. Kane and R. Kolodner (1993) The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 75, 1027-1038. 22. Leach, F. S., N. C. Nicolaides, N. Papadopoulos, B. Liu, J. Jen, R. Parsons, P. Peltomaki, P. Sistonen, L. A. Aaltonen, M. Nystrom-Lahti, X.-Y. Guan, J. Zhang, P. S. Meltzer, J.-W. Yu, F.T. Kao, D. J. Chen, K. M. Cerosaletti, R. E. K. Fournier, S. Todd, T. Lewis, R. J. Leach, S. L. Naylor, J. Weissenbach, J.P. Mecklin, H. Jarvinen, G. M. Petersen, S. R. Hamilton, J. Green, J. Jass, P. Watson, H. T. Lynch, J. M. Trent, A. de la Chapelle, K. W. Kinzler and B. Vogelstein (1993) Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell 75, 1215-1225.

23. Fujii, H. and T. Shimada (1989) Isolation and characterization of cDNA clones derived from the divergently transcribed gene in the region upstream from the dihydrofolate reductase gene. J. Biol. Chem. 264, 10057-10064.

24. New, L., K. Liu and G. F. Crouse (1993) The yeast gene MSH3 defines a new class of eukaryotic MutS homologues. Mol. Gen. Genet. 239, 97-108.

25. Drummond, J. T.. G.-M. Li, M. J. Longley and P. Modrich (1995) Isolation of an hMSH2-p160 heterodimer that restores DNA mismatch repair to tumor cells. Science 268, 1909-1912. 26. Palombo, F., P. Gallinari, I. Iaccarino, T. Lettieri, M. Hughes, A. D'Arrigo, O. Truong, J. J. Hsuan and J. Jiricny (1995) GTBP, a 160-kilodalton protein essential for mismatch-binding activity in human cells. Science 268, 1912-1914.

27. Palombo, F., I. Iaccarino, E. Nakajima, M. Ikejima, T. Shimada and J. Jiricny (1996) hMut(Beta), a heterodimer of hMSH2 and hMSH3, binds to insertion/deletion loops in DNA. Curr. Biol. 6, 1181-1184.

28. Acharya, S., T. Wilson, S. Gradia, M. F. Kane, S. Guerrette, G. T. Marsischky, R. Kolodner and R. Fishel (1996) hMSH2 forms specific mispair-binding complexes with hMSH3 and hMSH6. Proc. Natl. Acad. Sci. USA 93, 13629-13634. 29. Bronner, C. E., S. M. Baker, P. T. Morrison, G. Warren, L. G. Smith, M. K. Lescoe, M. Kane, C. Earabino, J. Lipford, A. Lindblom, P. Tannergard, R. J. Bollag, A. R. Godwin, D. C. Ward, M. Nordenskjold, R. Fishel, R. Kolodner and R. M. Liskay (1994) Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary nonpolyposis colon cancer. Nature 368, 258-261.

30. Papadopoulos. N., N. C. Nicolaides, Y.-F. Wei, S. M. Ruben, K. C. Carter, C. A. Rosen, W. A. Haseltine, R. D. Fleichmann, C. M. Fraser, M. D. Adams, J. C. Venter, S. R. Hamilton, G. M. Petersen, P. Watson, H. T. Lynch, P. Peltomaki, J.-P. Mecklin, A. de la Chapelle, K. W. Kinzler and B. Vogelstein (1994) Mutation of a mutL homolog in hereditary colon cancer. Science 263, 1625-1629.

Nicolaides, N. C., N. Papadopoulos, B. Liu, Y.-F. wel, K. ts Carter, S. M. Ruben, C. A. Rosen, W. A. Haseltine, R. D. Fleichmann, C. M. Fraser, M. D. Adams, J. C. Venter, M. G. Dunlop, S. R. Hamilton, G. M. Petersen, A. de la Chapelle, B. Vogelstein and K. W. Kinzler (1994) Mutations of two PMS

homologues in hereditary nonpolyposis colon cancer. Nature 371, 75-80.

32. Williamson, M. S., J. C. Game and S. Fogel (1985) Meiotic gene conversion mutants in Saccharomyces cerevisiae. I. Isolation and characterization of PMSI-I and PMSI-2. Genetics 110, 609646.

33. Kramer, W., B. Kramer, M. S. Williamson and S. Fogel (1989) Cloning and nucleotide sequence of DNA mismatch repair gene PMSI from Saccharomyces cerevisiae: homology of PMSI to procaryotic mutL and hexB. J. Bacteriol. 171, 5339-5346. 34. Kolodner, R. (1996) Biochemistry and genetics of eukaryotic

mismatch repair. Genes & Dev. 10, 1433-1442. 35. Risinger, J. I., A. Umar, J. Boyd, A. Berchuck, T. A. Kunkel and J. C. Barrett (1996) Mutation of MSH3 in endometrial cancer and evidence for its functional role in heteroduplex repair. Nature Genet. 14, 102-105.

36. Akiyama, Y., N. Tsubouchi and Y. Yuasa (1997) Frequent somatic mutations of hMSH3 with reference to microsatellite instability in hereditary nonpolyposis colorectal cancers. Biochem. Biophys. Res. Commun. 236, 248-252.

37. Akiyama, Y., H. Sato, T. Yamada, H. Nagasaki, A. Tsuchiya, R. Abe and Y. Yuasa (1997) Germ-line mutation of the hMSH6/ GTBP gene in an atypical hereditary nonpolyposis colorectal cancer kindred. Cancer Res. 57, 3920-3923. 38. Boyer, J. C., A. Umar, J. I. Risinger, J. R. Lipford, M. Kane, S. Yin, J. C. Barrett, R. D. Kolodner and T. A. Kunkel (1995) Microsatellite instability, mismatch repair deficiency, and genetic defects in human cancer cell lines. Cancer Res. SS, 60636070.

39. Papadopoulos, N., N. C. Nicolaides, B. Liu, R. Parsons, C. Lengauer, F. Palombo, A. D'Arrigo, S. Markowitz, J. K. V. Wilson, K. W. Kinzler, J. Jiricny and B. Vogelstein (1995) Mutations of GTBP in genetically unstable cells. Science 268, 1915-1917. 40. Dexter, D. L., J. A. Barbarosa and P. Calabresi (1979) N,N-- dimethylformamide-induced alteration of cell culture characteristics and loss of tumorigenicity in cultured human colon carcinoma cells. Cancer Res. 39, 1020-1025. 41. da Costa, L. T., B. Liu, W. S. El-Deiry, S. R. Hamilton, K. W. Kinzler, B. Vogelstein, S. Markowitz, J. K. V. Willson, A. de la Chapelle, K. M. Downey and A. G. So (1995) Polymerase a variants in RER colorectal tumours. Nature Genetics 9, 10-11. 42. Giard, D. J., S. A. Aaronson, G. J. Todaro, P. Arnstein, J. H. Kersey, H. Dosik and W. P. Parks (1973) In vitro cultivation of human tumors: establishment of cell lines derived from a series of solid tumors. J. Natl. Cancer Inst. 51, 1417-1423. 43. Day, R. S., III, L. H. Honore and K. D. Dobler (1991) Effect of passaging on Mer phenotype of human fetal cell cultures. Mutat. Res. 254, 153-160.

44. Green, M. and W. S. M. Wold (1979) Human adenoviruses: growth, purification, and transfection assay. Methods Enzymol. 58, 425435.

45. Sibghat-Ullah and R. S. Day, III (1992) Incision at O"-methylguanine: thymine mispairs in DNA by extracts of human cells. Biochemistry 31, 7998-8008.

46. Manley, J. L., A. Fire, A. Cano, P. A. Sharp and M. L. Gefter (1980) DNA-dependent transcription of adenovirus genes in a

soluble whole cell extract. Proc. Natl. Acad. Sci. USA 77, 38553859.

47. Day, R. S., III and C. H. J. Ziolkowski (1984) Human adenoviruses as indicators of inhibition of DNA repair. In DNA Repair and Its Inhibition (Edited by A. Collins, R. Johnson and C. S. Downes), pp. 69-78. IRL Press, Oxford.

48. Day, R. S., III and C. H. J. Ziolkowski (1982) Differential effects of hydroxyurea on the survival of UV- and MNNG-treated adenovirus 5. Mutat. Res. 94, 257-262.

49. Day, R. S., III, C. H. J. Ziolkowski, D. A. Scudiero, S. A. Meyer and M. R. Mattern (1980) Human tumor cell strains defective in the repair of alkylation damage. Carcinogenesis 1, 21-32. 50. Squires, S., D. J. Oates, S. D. Bouffler and R. T. Johnson (1992) Cockayne's syndrome fibroblasts are characterized by hypersensitivity to deoxyguanosine and abnormal DNA precursor pool metabolism in response to deoxyguanosine or ultraviolet light. Somatic Cell Mol. Genet. 18, 387401.

51. Battacharyya, N. P., A. Skandalis, A. Ganesh, J. Groden and M. Meuth (1994) Mutator phenotypes in human colorectal carcinoma cell lines. Proc. Natl. Acad. Sci. USA 91, 6319-6323. 52. Yarosh, D. B., M. Rice, R. S. Day, III, R. S. Foote and S. Mitra (1984) O6-methylguanine-DNA methyltransferase in human cells. Mutat. Res. 131, 27-36.

53. Richards, B., H. Zhang, G. Phear and M. Meuth (1997) Conditional mutator phenotypes in hMSH2-deficient tumor cell lines. Science 277, 1523-1526.

54. Berk, A. J. (1986) Adenovirus promoters and EIA transactiva

tion. Annu. Rev. Genet. 20, 45-79.

55. Svejstrup, J. Q., Z. Wang, W. J. Feaver, X. Wu, D. A. Bushnell, T. F. Donahue, E. C. Friedberg and R. D. Kornberg (1995) Different forms of TFIIH for transcription and DNA repair: holo-TFIIH and a nucleotide excision repairosome. Cell 80, 2128.

56. van Gool, A. J., E. Citterio, S. Rademakers, R. van Os, W. Vermeulen, A. Constantinou, J.-M. Egly, D. Bootsma and J. H. J. Hoeijmakers (1997) The Cockayne syndrome B protein, involved in transcription-coupled DNA repair, resides in an RNA polymerase II-containing complex. EMBO J. 16, 5955-5965. 57. Zawel, L., K. P. Kumar and D. Reinberg (1995) Recycling of the general transcription factors during RNA polymerase II transcription. Genes & Dev. 9, 1479-1490.

58. Tijsterman, M., R. A. Verhage, P. van de Putte, J. G. Tasseron-- de Jong and J. Brower (1997) Transitions in the coupling of transcription and nucleotide excision repair within RNA polymerase II-transcribed genes of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 94, 8027-8032.

59. Liu, B., N. C. Nicolaides, S. Markowitz, J. K. V. Willson, R. E. Parsons, J. Jen, N. Papadopoulos, R. Peltomaki, A. de la Chapelle, S. R. Hamilton, K. W. Kinzler and B. Vogelstein (1995) Mismatch repair gene defects in sporadic colorectal cancers with microsatellite instability. Nature Genet. 9, 48-55. 60. Parsons, R., G.-M. Li, M. J. Longley, W.-H. Fang, N. Papadopoulos, J. Jen, A. de la Chapelle, K. W. Kinzler, B. Vogelstein and P. Modrich (1993) Hypermutability and mismatch repair deficiency in RER+ tumor cells. Cell 75, 1227-1236. 61. Branch, P., R. Hampson and P. Karran (1995) DNA mismatch binding defects. DNA damage tolerance, and mutator phenotypes in human colorectal carcinoma cell lines. Cancer Res. SS, 2304-2309.

Rufus S. Day, III*1, Aghdass Rasouli-Nia1,2, James Meservy^l, Sibghat-Ullah Lari^^1, Kelly Dobler^^1, Shigeru Tsunoda1,3, Junji Miyakoshi3, Hiraku Takebe3 and David Murray2 'Molecular Oncology Program and 2Radiobiology Program, Cross Cancer Institute, Edmonton, Alberta, Canada, and 3Faculty of Medicine, Department of Experimental Radiology, Kyoto, Japan

This paper is dedicated to the memory of Dr. Raymond Latarjet (1911-1998).

*To whom correspondence should be addressed at: Biological Sciences, University College of the Cariboo, Box 3010, 900 McGill Road, Kamloops, British Columbia V2C 5N3, Canada. Fax: 250828-5450; e-mail: rday@cariboo.bc.ca

^Present address: Baylor College of Medicine, Department of Biochemistry, One Baylor Plaza, Houston, TX 77030-3498, USA.

^^Present address: Biological & Medical Research Department (MBC 03), King Faisal Specialist Hospital & Research Centre, P.O. Box 3354, Riyadh 11211, Saudi Arabia.

Copyright American Society of Photobiology Aug 1999
Provided by ProQuest Information and Learning Company. All rights Reserved