The CDKN2 (MISI) gene is located at 9p21; its product, p 16. inhibits the cyclin D/CDK4 complex that phosphorylates pRb, thus negatively regulating cell cycle progression IM. Serrano et al., Nature i l.oml.i. 366: 704, 1994; A. Kamb et al, Science (Washington DC), 264: 436, 1994; T. Nobori et al.. Nature (Lond.), 368: 753,1994]. CDKN2 mutations are more common in cultured human uroepithelial cells (HUC) than in uncultured bladder cancers. We examined the status of CDKN2/pl6 in early and late passage (P) cultures of HUC. HUC immortalization was not accompanied by pl6 loss, even in cells with a hemizygous 9p21-pter deletion, but late passage cultures with a pl6 loss showed decreased generation time. Thus, the data do not indicate that CDKN2 is a candidate for a chromosome 9 senescence gene but suggest that pl6 loss may confer a growth advantage in vitro. Significant differences in pl6 levels were observed among HUC cell lines, but no CDKN2 mutations were detected. However, an inverse correlation between elevated pl6 and loss of pRb function was observed (P < 10 Y Ten samples with normal pRb showed low or undetectable p 16 levels, while seven samples with known pRb alterations showed abundant pl6 but nevertheless grew vigorously in culture. These results support the hypothesis that pl6 mediated cell cycle inhibition, as well as pl6 regulation, occurs via pRb dependent pathway(s).

Metabolic activation of iY-hydroxy-4-acetylaminobiphcnyl (7V-OHAABP) and Ã-V-acetoxy-4-acetylaminobiphenyl (A'-OAc-AABP), the proximate carcinogenic metabolites of the human bladder carcinogen 4-aminobiphenyl (ABP), was examined in human uroepithelial cells (HL'C). Bioconversion was studied by incubating HUC cultures with I'Hj/V-OAc-AABP or |'H|/V-OH-AABP. Three Ã3rgano-soluble me tabolites, A'-OH-AABP, 4-acetylaminobiphenyl (AABP), and ABP were identified in ethyl acetate extracts from cultures exposed to A'-OAcAABP. Similarly, AABP and ABP were characterized as the major metabolites from cultures treated with /V-OH-AABP. Incubation of A'OAc-AABP with HUC microsomes in vitro yielded primarily the Odeacetylation product A'-OH-AABP. The HUC microsomes also cata lyzed the A'-deacetylation of A'-OAc-l'TlAABP, Af-OH-rqAABP, and I'HIAABP. The Oand A-deacetylase activities for A'-OAc-AABP were 55.9 and 38.2 nmol/mg/min, respectively. These Oand A'-deacetylase activities were both blocked by paraoxon. Incubation of ['HjA'-OAcAABP or ('H^-OH-AABP with HUC microsomes and tRNA or DNA showed that 23.0 and 8.0 nmol of A'-OAc-AABP and 74.5 and 25.2 pmol of A'-OH-AABP were bound per mg protein/mg RNA or DNA, respec tively. In comparison, the acetyl CoA-dependent HUC cytosol-mediated bindings of l'H^-OH-ASP to RNA and DNA were 801 and 447 pmol/ mg nucleic acid/mg protein. The HUC microsome-mediated bindings of A'-OAc-AABP and A'-OH-AABP to nucleic acids were inhibited by paraoxon, whereas the cytosol-mediated binding of A'-OH-ABP was insensitive to paraoxon inhibition. Chromatography of the DNA hydrolysate obtained from the in vitro incubation of ['HJA'-OAc-AABP or j'H] A'-OH-AABP with HUC microsomes showed A'-(deoxyguanosine-8-yl)4-aminobiphenyl as the major adduct, based on comparison with authentic synthetic standard. These results show that human uroepithelia contain microsomal acetyl transferases that are capable of converting the proxi mate metabolites A'-OAc-AABP and A'-OH-AABP of the human bladder carcinogen ABP, to reactive electrophiles that bind to DNA. The occur rence of these acetyl transferases in the target organ of the human bladder carcinogen ABP suggests that metabolic activation of some proximate metabolites of ABP could occur directly in HUC and could play a pivotal role in susceptibility to aryl-amine/acetamide induced human bladder cancers.

pSV2Neo, a plasmid that contains the wild-type simian virus 40 (SV40) origin of replication (ori), is widely used in mammalian cell transfection experiments. We observed that pSV2Neo transforms two nontumorigenic SV40-immortalized human uroepithelial cell lines (SV-HUC and CK/SV-HUC2) to G418 resistance (G418r) at a frequency lower than that at which it transforms SV-HUC tumorigenic derivatives (T-SV-HUC). Transient expression studies with the chloramphenicol transferase assay showed that these differences could not be explained by differences in Neo gene expression. However, when we replaced the SV40 ori in pSV2Neo with a replication-defective ori to generate G13.1Neo and G13.1'Neo, the G418r transformation frequency of the SV40-immortalized cell lines was elevated. Because SV40 T antigen stimulates replication at its ori, we tested plasmid replication in these transfected cell lines. The immortalized cell lines that showed low G418r transformation frequencies after transfection with pSV2Neo showed high levels of plasmid replication, while the T-SV-HUC that showed high G418r transformation frequencies failed to replicate pSV2Neo. To determine whether differences in the status of the T-antigen gene contributed to the phenomenon, we characterized the T-antigen gene in these cell lines. The results showed that the T-SV-HUC had sustained mutations in the T-antigen gene that would interfere with the ability of the T antigen to stimulate replication at its ori. Most T-SV-HUC contained a super-T-antigen replication-defective ori that apparently resulted from the partial duplication of SV40 early genes, but one T-SV-HUC had a point mutation in the ori DNA-binding domain of the T-antigen gene. These results correlate with the high G418r transformation frequencies with pSV2Neo in T-SV-HUC compared with SV-HUC and CK/SV-HUC2. Furthermore, these results suggest that alterations in SV40 T antigen may be important in stabilizing human cells immortalized by SV40 genes that contain the wild-type SV40 ori, thus contributing to tumorigenic transformation. This is the first report of a super T antigen occurring in human SV40-transformed cells.

The global genomic repair of DNA adducts was examined in human papillary transitional cell carcinoma (TCC) cell lines after exposure to N:-hydroxy-4-acetylaminobiphenyl (N-OH-AABP), the proximate carcinogenic metabolite of the human bladder carcinogen 4-aminobiphenyl (ABP). (32)P-post-labeling analysis of TCC cultures exposed to N-OH-AABP revealed a major adduct, identified as the 3',5'-bisphosphate derivative of N-(deoxyguanosin-8-yl)-4-aminobiphenyl (dG-C8-ABP). The amount of adduct formation in TCC10 was dependent upon the dose and the duration of exposure and ranged between 1 and 5 adducts/10(7) nucleotides. To test if p53 regulates repair of the dG-C8-ABP adduct in genomic DNA, an isogeneic set of cell lines was obtained by infection of the TCC10 cultures with a retroviral construct expressing a trans-dominant mutant of p53, namely a Val-->Ala mutation at codon 143. The TDM143-TCC10 line expressing the mutant form of p53 was selected. The rate of repair of dG-C8-ABP was compared between TCC10 and TDM143-TCC10 cultures after treatment with 15 microM N-OH-AABP. The rate of disappearance of the adduct was monitored over a period of time after chemical treatment. (32)P-post-labeling analysis of dG-C8-ABP in parental TCC10 showed its rapid removal, the majority of adducts disappearing within 48 h. In contrast to TCC10, TDM143-TCC10 was relatively slower in removal of dG-C8-ABP. After 24 h DNA repair TDM143-TCC10 showed an approximately 3-fold greater amount of dG-C8-ABP compared with TCC10. These results imply that p53 plays a role in the repair of ABP adducts and that in p53 null cells the unrepaired DNA damage could cause accumulation of mutations, which might contribute to increased genomic instability and neoplastic progression.

An in vitro/in vivo transformation system has been developed as a model for bladder tumorigenesis. SV40-immortalized human uroepithelial cells are exposed to putative carcinogens and then implanted into athymic nude mice to testfortumorigenesis.Studieswith4-aminobiphenyl(4-ABP)demonstratedthat one cell line, SV-HUC-PC, was sensitive to chemical-induced transformation and another line, SV-HUC-BC, was refractory.Weare currently testing this system as a model to identify occupational carcinogens and develop biomarkers of exposure and effects of exposure. As part of this study, we examined P450- dependent metabolism, glutathione transferase, and the effects of chemicals on deoxyribonucleic acid(DNA)synthesisandrepair inSV-HUC-PC andSV-HUCBC. Activities for CYP1A1/1A2, CYP3A, and CYP2B1/2B2 were estimated by determining o -dealkylation of ethoxy-, benzoxy-, and pentoxy-resorufin, respectively. Coumarin hydroxylase and p -nitrophenol hydroxylase were used to estimate CYP2A and CYP2E1, respectively. SV-HUC-PC microsomes had fivefold greaterCYP1A1/1A2activityandtwofoldhigherCYP3AactivitythanSV-HUCBC. CYP2B1/2B2 and CYP2A activities and glutathione transferase were not different between the two cell lines. DNA synthesis and repair, by BrdU incorporation, was not different between the two lines when N-methyl-N-nitroN-nitrosoguanidine (MNNG) or other reactive metabolites were tested; however, SV-HUC-PC was more sensitive to n -nitrosodimethylamine, 4-ABP, and 4,4-methylene bis (2-chloroaniline) (MOCA). The data demonstrate that, while these cells have retained form-specific P450 activities, SV-HUC-PC has greater CYP1A1/1A2 and CYP3A activities.

Human bladder cancers are heterogeneous. For example, at first presentation papillary transitional cell carcinomas (TCCs) are typically superficial and often multifocal. Papillary TCCs frequently recur, but most never progress to invasive TCC. In contrast, bladder carcinoma in situ (CIS) usually presents as a solitary flat lesion and, if left untreated, invariably progresses to invasive TCC. Some TCC are already invasive at the time of presentation. Squamous cell carcinoma (SCC) tends to present at a later stage than most TCCs and has a relatively aggressive clinical course. Multiple genetic alterations have been identified in invasive human bladder cancers and are present in different combinations and in different frequencies in the different manifestations of bladder cancer described above. A high percentage ( approximately 67%) of superficial papillary TCCs show early losses involving chromosome 9q, while very few show either a TP53 or a CDKN2/16 mutation. Thus, loss of 9q may be the earliest event in initiation of papillary TCC. In contrast, bladder CIS and SCC show relatively low percentages of 9q loss. However, approximately 65% of bladder CIS contain a TP53 alteration and approximately 67% of bladder SCC contain a CDKN2/p16 alteration. Mutations in these two tumor suppressor genes have powerful implications for loss of genome stability and cell growth regulation, respectively, consistent with the aggressive phenotypes of these cancers. Thus, these data suggest a model of bladder cancer pathogenesis in which the predominant genetic alteration may be the "initiating event" in cancer pathogenesis and may play a role in determining the biological potential of the tumor.

Inactivation of both the pRb (pRb-cyclin D1/cyclin-dependent kinase 4/6-p16) and p53 (p53-p21(WAF1)-p14(ARF)) pathways is thought to be essential for immortalization in vitro and malignant transformation in vivo. We identified different combinations of pRb and p53 pathway alterations in 12 invasive transitional cell carcinomas (TCCs) and addressed the functional significance of the different combinations observed. Results showed four combinations of alterations including -pRb/-p53 (ie., pRb inactivated in the pRb pathway and p53 inactivated in the p53 pathway; four TCCs), -p16/-p53 (four TCCs), -p16/-p21(WAF1) (one TCC), and -p16/ -p14(ARF) (two TCCs). These groups include two new combinations (ie., -p16/-p53 and -p16/-p21(WAF1)) not reported previously for TCCs. An alteration in the key components of the p53 pathway was not detected in one invasive TCC that had inactivated p16. Note that all four TCCs with inactivated pRb had mutant p53; thus, the combinations of -pRb/ -p21(WAF1) and -pRb/-p14(ARF) were not observed. Only two of eight TCCs with altered p16 had concomitant p14(ARF) loss, demonstrating that simultaneous inactivation of these two 9p21INK4a tumor suppressor genes is not obligatory. To determine the biological phenotypes of TCCs with different combinations of pRb and p53 pathway alterations, their downstream responses to gamma radiation were studied in vitro. As expected, none of eight TCCs with mutant p53 responded to gamma radiation by elevation of p53, p21(WAF1), or mdm2 or by cell cycle arrest. Only two of four TCCs with wild-type p53 and wild-type pRb (the combination of -p16/-p14(ARF)) showed normal downstream responses to gamma radiation and underwent cell cycle arrest. Two TCCs with wild-type pRb and wild-type p53 (the combination of -pl6/-p21(WAF1) and one TCC with -p16) failed to show cell cycle arrest in response to radiation. This was attributed to the absence of p21(WAF1) in one TCC. In summary, these data support a model of invasive bladder cancer pathogenesis in which both the pRb and p53 pathways are usually inactivated and the biology of the tumor is impacted by the mechanism of their inactivations.

The cell cycle regulatory genes p16/CDKN2 and RB are frequently deleted in prostate cancers. In this study, we examined the role of alterations in p16 and pRb during growth, senescence, and immortalization in vitro of human prostate epithelial cells (HPECs). HPECs are established from normal prostate tissues and cultured on collagen-coated dishes. Our results show that p16 is reproducibly elevated at senescence in HPECs. HPECs are immortalized using human papilloma virus 16 E6 and/or E7 as molecular tools to inactivate p53 and/or pRb, respectively. Immortalization occurs infrequently in this system and only after a latent period during which additional genetic/epigenetic changes are thought to occur. Notably, all of the E6-immortalized HPEC lines but none of the E7 lines show inactivation of p16/CDKN2 (by deletion, methylation, or mutation) in association with immortalization. In contrast, E7 lines, in which pRb function is abrogated by E7 binding, retain the high levels of p16 observed at senescence. Thus, all lines show either a p16 or pRb inactivation. Analysis of six independent lines from metastatic prostate cancers reveals a similar loss of either p16 or pRb. Comparative genomic hybridization of HPECs shows that gains of chromosomes 5q, 8q, and 20 are nonrandomly associated with bypassing senescence (probability = 0.95). These results suggest that high levels of the cyclin-dependent kinase inhibitor p16 mediate senescence G1 arrest in HPECs and that bypassing this block by a p16/pRb pathway alteration is required for immortalization in vitro and possibly tumorigenesis in vivo. Our results further indicate that inactivation of the p16/pRb pathway alone is not sufficient to immortalize HPECs and that additional genetic alterations are required for this process.

PURPOSE:Normal human uroepithelial cells (HUCs) proliferate rapidly in culture during early passage and then spontaneously undergo replicative senescence. We previously reported that the cyclin D1-CDK4/6 inhibitor, p16INK4a, is elevated at senescence in HUCs. Hence, we proposed that p16INK4a may play a critical role in mediating senescence in this cell type. In the current study, we further characterized the senescent state in HUCs. We also tested the possible roles of changes in other cell cycle proteins, including p53, p21WAF1, pRb, and cyclin D1 in HUC senescence. METHODS:Normal HUCs cultured from explants of ureteral mucosa were used for these studies. Senescence associated-beta-galactosidase activity (SA-beta-gal) was used to identify cells in senescence. Flow cytometric analysis was used to determine changes in cell cycle distribution at senescence. Response of cells to serum stimulation was determined by Northern analysis of c-fos. Western analysis was used to assess changes in p53, p21WAF, p16INK4a, cyclin D1 and plasminogen activator inhibitor-1 (PAI-1) levels at senescence. RESULTS:beta-gal-positive HUCs were blocked at G1/S in senescence and failed to show c-fos induction in response to serum stimulation. As previously reported, senescent HUCs also showed elevated p16INK4a. However, unlike human fibroblasts, neither p53 nor p21WAF1 elevation accompanied HUCs senescence. PAI-1 levels were also not elevated in HUC senescence. CONCLUSION:These findings support a model in which elevation of p16INK4a, but not p53 or p21WAF1 plays a critical role in HUC replicative senescence. These findings elucidate the tumor suppressor mechanism of p16INK4a and the frequent loss of either p16INK4a or pRb in invasive human bladder tumors.

Gain of 20q has been observed in many cancer types, including bladder cancers. However, the biological significance of low-copy-number 20q gain in human cancer pathogenesis has not yet been defined. We reported that immortalization of human uroepithelial cells (HUC) transformed with human papillomavirus 16 (HPV 16) E7 is associated with single-copy 20q gain (P = 2 × 10-7). We also observed 20q13.2 amplification in some cell lines, but only after 20 passages. Thus, we hypothesized that low-copy gain of 20q gene(s) contributes in a dominant way to bypassing HUC senescence. To test this hypothesis, we fused precrisis E7-transformed HUCs (pcE7s) with three independent immortal E7-HUCs that acquired a single-copy 20q gain at immortalization. In one of these lines, a single-copy gain of 20q and a 10p12.1–pter loss were the only cytogenetic alterations. Immortal cell hybrids were obtained with all three crosses. Southern analysis for unique HPV16 insertion sites, as well as fluorescence in situ hybridization (FISH) with whole chromosome 20 painting probes (WCP20) for marker chromosomes in the immortal clones, confirmed the hybrid and independent nature of representative immortal clones. In contrast, when we used the same protocol, no immortal somatic cell hybrids were obtained when HPV16 E6 immortal HUC (E6-HUC) that showed 3p and 9p losses, but no 20q gain, were fused with precrisis E6-transformed HUC (pcE6s). This latter observation is consistent with many results demonstrating that recessive changes are required for cell immortalization. Therefore, the new results reported herein for the first time demonstrate that dominant changes can contribute to bypassing senescence, and that such genes may be located on 20q. Genes Chromosomes Cancer 26:304–311, 1999. © 1999 Wiley-Liss, Inc.

A role for the Mut L homologue-1 (MLH1) protein, a necessary component of DNA mismatch repair (MMR), in G2-M cell cycle checkpoint arrest after 6-thioguanine (6-TG) exposure was suggested previously. A potential role for MLH1 in G1 arrest and/or G1-S transition after damage was, however, not discounted. We report that MLH1-deficient human colon carcinoma (HCT116) cells showed decreased survival and a concomitant deficiency in G2-M cell cycle checkpoint arrest after ionizing radiation (IR) compared with genetically matched, MMR-corrected human colon carcinoma (HCT116 3-6) cells. Similar responses were noted between murine MLH1 knockout compared to wild-type primary embryonic fibroblasts. MMR-deficient HCT116 cells or embryonic fibroblasts from MLH1 knockout mice also demonstrated classic DNA damage tolerance responses after 6-TG exposure. Interestingly, an enhanced p53 protein induction response was observed in HCT116 3-6 (MLH1+) compared with HCT116 (MLH1-) cells after IR or 6-TG. Retroviral vector-mediated expression of the E6 protein did not, however, affect the enhanced G2-M cell cycle arrest observed in HCT116 3-6 compared with MLH1-deficient HCT116 cells. A role for MLH1 in G2-M cell cycle checkpoint control, without alteration in G1, after IR was also suggested by similar S-phase progression between irradiated MLH1-deficient and MLH1-proficient human or murine cells. Introduction of a nocodazole-induced G2-M block, which corrected the MLH1-mediated G2-M arrest deficiency in HCT116 cells, clearly demonstrated that HCT116 and HCT116 3-6 cells did not differ in G1 arrest or G1-S cell cycle transition after IR. Thus, our data indicate that MLH1 does not play a major role in G1 cell cycle transition or arrest. We also show that human MLH1 and MSH2 steady-state protein levels did not vary with damage or cell cycle changes caused by IR or 6-TG. MLH1-mediated G2-M cell cycle delay (caused by either MMR proofreading of DNA lesions or by a direct function of the MLH1 protein in cell cycle arrest) may be important for DNA damage detection and repair prior to chromosome segregation to eliminate carcinogenic lesions (possibly brought on by misrepair) in daughter cells.

This paper concerns the statistical analysis of certain binary data arising in molecular studies of cancer. In allelic-loss experiments, tumour cell genomes are analysed at informative molecular marker loci to identify deleted chromosomal regions. The resulting binary data are used to infer properties of putative suppressor genes, genes involved in normal cell cycling. Various factors can complicate this inference, including background loss of heterozygosity, spatial (that is, within chromosome) dependence of the binary responses, non-informativeness of markers, covariates such as protein levels or tumour histology, heterogeneity of cells within tumours, and measurement error. We focus on the first three factors, discussing methods for statistical inference that separate background loss from significant loss. We outline the extension to other inferences, such as comparison questions and the relationship to covariates. Using characteristic features of tumourigenesis, we present a framework for the stochastic modelling of allelic-loss data, and build models within this framework; in particular, we propose a simple model that has chromosome breaks at locations of a Poisson process, and preferential selection cells with inactivated suppressor genes. We illustrate these methods on allelic-loss data from induced rat mammary tumours and human bladder cancers.