15-Lipoxygenase-2 gene regulation by its product 15-(S)-hydroxyeicosatetraenoic acid through a negative feedback mechanism that involves peroxisome proliferator-activated receptor gamma.

An inverse relationship exists between the expression of 15-lipoxygenase-2 (15-LOX-2) and peroxisome proliferator-activated receptor gamma (PPARgamma) in normal prostate epithelial cells (PrECs) compared with their expression in prostate carcinoma cells (PC-3). The reason for this difference, however, is unknown. We hypothesized that this inverse expression partly involves the 15-LOX-2 promoter and 15-S-hydroxyeicosatetraenoic acid (15-(S)-HETE), a product of 15-LOX-2 that binds to PPARgamma. We identified an active steroid nuclear receptor half-site present in the 15-LOX-2 promoter fragment F-5 (-618/+177) that can interact with PPARgamma. After forced expression of wild-type PPARgamma, 15-(S)-HETE (1 microM) decreased F-5 reporter activity in PrECs whereas forced expression of 15-LOX-2 resulted in 15-(S)-HETE production which enhanced F-5 activity in PC-3. In contrast, the expression of dominant-negative PPARgamma reversed the transcriptional activation of F-5 by enhancing it 202-fold in PrEC or suppressing it in PC-3; the effect in PC-3 was positively increased 150-fold in the presence of 15-(S)-HETE (1 microM). Peroxisome proliferator-activated receptor gamma interacted with 15-LOX-2 promoter sequences in pulldown experiments using biotinylated 15-LOX-2 (-560/-596 bp) oligonucleotides. In gelshift analyses PPARgamma and orphan receptor RORalpha were shown to interact with the F-5 fragment in PC-3 cells. These data suggest that crosstalk mechanisms exist between the 15-LOX-2 gene and PPARgamma to counterbalance expression and help explain the inverse relationship of these genes in normal versus cancer cells.

Retinoic acid receptor beta expression and growth inhibition of gynecologic cancer cells by the synthetic retinoid N-(4-hydroxyphenyl) retinamide.

BACKGROUND: The synthetic retinoid N-(4-hydroxyphenyl) retinamide (4HPR) can inhibit the growth of tumor cells. Preliminary results from a clinical trial suggest that 4HPR may reduce ovarian cancer incidence. We examined the growth-inhibitory effects of 4HPR on gynecologic cancer cell lines in vitro and the role of retinoid receptors in modulating this effect. METHODS: Twelve human gynecologic cancer cell lines (the ovarian cell lines--A224, AD10, UCI 101, UCI 107, SKOV3, 222, CP70, ML3B, and ML5; the cervical cell lines--HT3 and ME180; and the endometrial cell line--Hec 1A were tested for sensitivity to 4HPR (by assaying cell proliferation rates). Gel electrophoretic analysis of DNA fragmentation was used to measure programmed cell death (apoptosis). Specific retinoid receptor (retinoic acid receptor [RAR] and retinoid X receptor) messenger RNA (mRNA) levels were measured by northern blot hybridization. AD10 cells were stably transfected with human RARbeta complementary DNA, and the effect of 4HPR on cell proliferation was examined. RESULTS: 4HPR inhibited the growth of all 12 cell lines, but to varying degrees; IC50 values (i.e., concentrations that inhibit proliferation by 50%) ranged from 0.3 to 9 microM. Following 4HPR treatment, ovarian cancer cells that were sensitive to 4HPR (222, CP70, and UCI 101; IC50 <3 microM) contained higher levels of RARbeta transcripts than more resistant cells (AD10, ME180, Hec 1A, and A224; IC50 > or =3 microM) (2.8-fold; two-sided P = .006). Anchorage-independent growth of transfected AD10 cells expressing high levels of RARbeta was totally abolished, even in the absence of 4HPR; transfectants expressing low levels of RARbeta exhibited lower levels of anchorage-independent growth and grew more slowly in the presence of 4HPR than control untransfected AD10 cells. CONCLUSION: 4HPR inhibited the proliferation of ovarian cancer cells in vitro; RARbeta expression appeared to be associated with this effect.

Induction of programmed cell death in Kaposi's sarcoma cells by preparations of human chorionic gonadotropin.

BACKGROUND: Isolation of the first neoplastic acquired immunodeficiency syndrome-related Kaposi's sarcoma (KS) cell line (KS Y-1) has furthered understanding of the pathogenesis of KS. Studies with KS Y-1 cells have indicated that inhibition of KS cell proliferation occurs in early pregnancy in mice and after treatment with certain commercial preparations of human chorionic gonadotropin (hCG, a pregnancy hormone purified from urine). The activity of the commercial preparations has been attributed to an hCG-associated factor(s) (HAF). While several clinical benefits of HAF are clearly evident, the basis for its anti-KS properties remains unknown. We investigated the apoptosis-inducing effects of HAF and the expression of apoptosis-related proteins in KS cells. METHODS: KS Y-1 and KS SLK cells were treated with clinical-grade crude preparations of hCG, recombinant hCG, or urine fractions exhibiting anti-KS activity and then examined for features of apoptosis. Levels of proteins associated with apoptosis were monitored by western blot analysis, and cell DNA content was assessed by flow cytometry. Tumors induced in mice by inoculation of KS Y-1 cells were treated with preparations of hCG, and the tumors were examined for cell morphology and also for DNA fragmentation by use of the terminal deoxynucleotidyl transferase-mediated digoxigenin-deoxyuridine triphosphate nick-end-labeling (TUNEL) assay. RESULTS: The HAF present in some preparations of hCG and in urine fractions has the ability to induce apoptosis in KS cells in vitro and in vivo. HAF-triggered apoptosis was preceded by increased levels of the apoptosis-related proteins c-Myc and c-Rel and cell accumulation in Go/G1 phase of the cell cycle. KS Y-1 cells transfected with a c-Myc complementary DNA showed elevated rates of apoptosis. CONCLUSION: The anti-KS activity of HAF appears to induce apoptosis. Such activity suggests a role for HAF in pregnancy-related regulation of cell death.

Differential expression of cyclooxygenase-2 and its regulation by tumor necrosis factor-alpha in normal and malignant prostate cells.

Cyclooxygenase (COX)-2 expression is elevated in some malignancies; however, information is scarce regarding COX-2 contributions to the development of prostate cancer and its regulation by inflammatory cytokines. The present study compared and contrasted the expression levels and subcellular distribution patterns of COX-1 and COX-2 in normal prostate [prostate epithelial cell (PrEC), prostate smooth muscle (PrSM), and prostate stromal (PrSt)] primary cell cultures and prostatic carcinoma cell lines (PC-3, LNCaP, and DU145). The basal COX-2 mRNA and protein levels were high in normal PrEC and low in tumor cells, unlike many other normal cells and tumor cells. Because COX-2 levels were low in prostate smooth muscle cells, prostate stromal cells, and tumor cells, we also examined whether COX-1 and COX-2 gene expression was elevated in response to tumor necrosis factor-alpha (TNF-alpha), a strong inducer of COX-2 expression. Northern blot analysis and reverse transcription-PCR demonstrated different patterns and kinetics of expression for COX-1 and COX-2 among normal cells and tumor cells in response to TNF-alpha. In particular, COX-2 protein levels increased, and the subcellular distribution formed a distinct perinuclear ring in the normal cells at 4 h after TNF-alpha exposure. The COX-2 protein levels also increased in cancer cells, but the subcellular distribution was less organized; COX-2 protein appeared diffuse in some cells and accumulated as focal deposits in the cytoplasm of other cells. TNF-alpha induction of COX-2 and prostaglandin E2 correlated inversely with induction of apoptosis. We conclude that COX-2 expression may be important to PrEC cell function. Although it is low in stromal and tumor cells, COX-2 expression is induced by TNF-alpha in these cells, and this responsiveness may play an important role in prostate cancer progression.

Regulation of urokinase-type plasminogen activator expression by the v-mos oncogene.

We undertook a study to determine if the serine-threonine kinase-encoding v-mos oncogene regulated the expression of the urokinase-type plasminogen activator. An expression vector encoding v-mos, but not a kinase-inactive mutant, stimulated urokinase promoter activity in CAT assays employing a squamous cell carcinoma cell line. The induction of urokinase promoter activity by v-mos was mediated, in part, via an increased AP-1 activity since (a) mutation of 2 AP-1 binding sites (at -1967 and -1885), or the co-expression of a transactivation domain-lacking c-jun mutant reduced the induction of the urokinase promoter by v-mos and (b) expression of v-mos increased the activity of a CAT reporter driven by three AP-1 tandem repeats. The stimulation of the urokinase promoter by v-mos was partially countered by co-expression of an ERK1/ERK2-inactivating phosphatase. Western blotting and zymographic analysis indicated that v-mos-transformed NIH3T3 cells (MSV NIH-3T3) secreted more urokinase compared with NIH3T3 cells and this was associated with a higher level of activated ERK1 and ERK2. Expression of a catalytically-inactive MAPKK mutant reduced the activity of a urokinase promoter-driven CAT reporter in the MSV NIH-3T3 cells. In conclusion, the data herein indicate that urokinase expression is regulated by v-mos through a MAPKK-dependent signaling pathway.

cJun overexpression in MCF-7 breast cancer cells produces a tumorigenic, invasive and hormone resistant phenotype.

We have previously demonstrated decreased Jun/AP-1 activity in the breast cancer cell line MCF-7 when compared to normal or immortalized mammary epithelial cells. In this paper, we overexpress Jun in MCF-7 cells (MCF7Jun) and demonstrate that it results in diverse biologic and biochemical changes, which mimic those seen clinically in breast cancer. Overexpression of Jun causes significant alterations in the composition of AP-1, decreased junB and increased fra-1 expression and results in an increased biologic aggressiveness. MCF7Jun cells exhibit increased cellular motility, increased expression of a matrix degrading enzyme MMP-9, increased in vitro chemoinvasion and tumor formation in nude mice in the absence of exogenous estrogens. Furthermore, MCF7Jun cells are unresponsive to the growth stimulating effects of estrogen and growth inhibitory effects of tamoxifen. Analysis of the estrogen receptor (ER) expression and activity showed that the MCF7Jun cells have no detectable ER. MCF-7 cells overexpressing mutant forms of cJun were responsive to the growth stimulatory effects of estrogen indicating that full-length cJun is required to acquire the estrogen-independent phenotype in breast cancer cells.

Selenium effects on prostate cell growth.

Epidemiological and clinical data suggest that selenium may prevent prostate cancer, but the biological effects of selenium on normal or malignant prostate cells are not well known. We evaluated the effects of sodium selenite (Na2SeO3) or l-selenomethionine (SeMet) on monolayer and anchorage-independent growth in a series of normal primary prostate cultures (epithelial, stromal, and smooth muscle) and prostate cancer cell lines (LNCaP, PC-3, and DU145). We observed differential, dose-dependent growth inhibition and apoptosis within prostate cancer cells (compared with normal prostate cells) treated with 1-500 microM of Na2SeO3 or SeMet. Na2SeO3 more potently inhibited growth at any given concentration. The androgen-responsive LNCaP cells were the most sensitive to selenium growth suppression (IC50s at 72 h for Na2SeO3 and SeMet were 0.2 and 1.0 microM, respectively). Growth of the primary prostate cells virtually was not suppressed (IC50s at 72 h for Na2SeO3 and SeMet were 22-38 and >500 microM, respectively). We also observed that DNA condensation and DNA fragmentation (terminal deoxynucleotidyltransferase dUTP nick end labeling/fluorescence-activated cell sorting) were elevated in selenium-treated cells and that activated caspase-3 colocalized with terminal deoxynucleotidyltransferase dUTP nick end labeling-stained cells by immunofluorescence. Higher basal poly(ADP-ribose) polymerase (PARP) expression levels and PARP cleavage (a substrate for caspase-3) were observed during apoptosis in tumor cells, compared with normal cells. Selective tumor cell death was associated with an increase in sub-G0-G1 cells after propidium iodide staining and fluorescence-activated cell sorting analysis. SeMet caused an increase in arrest in the G2-M phase of the cell cycle selectively in cancer cells. Inhibition of cancer cell growth by SeMet was associated with phosphorylation of P-Tyr15-p34/cdc2, which caused growth arrest in the G2-M phase. Anchorage-independent growth of prostate cancer cells in soft agar was sensitive to selenium. Our results suggest that Na2SeO3 is the more potent inducer of apoptosis in normal and cancer prostate cells. Our SeMet results involving PARP and G2-M cell-cycle arrest (cited above) indicate that SeMet selectively induces apoptosis in cancer but not primary cells of the human prostate. Our overall findings are relevant to the molecular mechanisms of selenium actions on prostate carcinogenesis and help demonstrate the selective, dose-dependent effects of selenium (especially SeMet) on prostate cancer cell death and growth inhibition.

Effects of novel phenylretinamides on cell growth and apoptosis in bladder cancer.

Superficial bladder cancer is a major target for chemoprevention. Retinoids are important modulators of epithelial differentiation and proliferation and are effective in the treatment and prevention of several epithelial cancers. One class of compounds, the retinamides, is structurally similar to other retinoids but have the added feature of being potent apoptosis inducers. Among these, fenretinide (N-[4-hydroxyphenyl]retinamide), or 4HPR, has promise for bladder cancer chemoprevention and is currently under Phase III study in this setting. In addition to 4HPR, there are several new structurally related phenylretinamides bearing hydroxyl, carboxyl, or methoxyl residues on carbons 2, 3, and 4 of the terminal phenylamine ring [designated N-(2-hydroxyphenyl)retinamide, N-(3-hydroxyphenyl)retin amide, N-(2-carboxyphenyl)retin- amide, N-(3-carboxyphenyl)retin amide, N-(4-carboxy- phenyl)retinamide, and N-(4-methoxyphenyl)retinamide, respectively]. The objective of this study was to compare the growth inhibitory and apoptotic effects of these phenylretinamides with 4HPR in human bladder transitional cell cancer-derived cell lines of varying histological grade (RT4, grade 1; UM-UC9 and UM-UC10, grade 3; and UM-UC14, grade 4) by cell counting, cell cycle fluorescence-activated cell sorter analysis and a dual stain apoptosis assay. All of the seven phenylretinamides reduced cell number, altered the cell cycle distribution, and induced apoptosis when administered at a concentration of 10 microM, which is within the pharmacologically achievable range. Although the relative potencies of the phenylretinamides varied depending on the cell line, N-(3-hydroxy phenyl)retin- amide was the most active with significantly greater growth inhibition than 4HPR in all of the four cell lines. These in vitro findings warrant further study of these novel phenylretinamides, which may have potential as preventive or therapeutic agents in transitional cell cancer.

Interobserver reproducibility of the Lagios nuclear grading system for ductal carcinoma in situ.

Several studies have shown an association between high nuclear grade or necrosis of ductal carcinoma in situ (DCIS) lesions and the risk of local disease recurrence in patients with DCIS treated surgically with less than mastectomy. Although criteria for separating low from high nuclear grade lesions have been published, no information exists regarding interobserver reproducibility (IR). To assess IR in the classification of DCIS, six surgical pathologists from four institutions used the Lagios grading system to grade 125 DCIS lesions. Before meeting to evaluate the cases, a training set of 12 glass slides, including cases chosen to present conflicting cues for classification, was mailed to the participants with a written criteria summary. This was followed by a working session in which criteria were reviewed and agreed on. The pathologists then graded the lesions independently. The area of interest was marked on each slide before grading. After initial grading, the pathologists met again to resolve discrepant lesion classifications. A complete agreement among raters was achieved in 43 (35%) cases, with five of six raters agreeing in another 45 (36%) cases. In no case did two raters differ by more than one grade. The pairwise kappa agreement values ranged from fair to substantial (0.30 to 0.61). Generalized kappa value indicated moderate agreement (0.46, standard error = 0.02). Kappa statistics for the distinction between grades 1 and 2 and 2 and 3 were 0.29 and 0.48, respectively, (standard error = 0.02). Only one of the six raters differed significantly in scoring. With adherence to specific criteria, IR in the classification of DCIS cases can be obtained in most cases. Although these pathologists made a few grading system modifications, further refinements are needed, especially if grading will influence future therapy.

The molecular biology of lung cancer: application to early detection and prevention.

With the overall incidence of lung cancer increasing and little progress being made in effective treatment, new approaches to this disease, such as early detection and prevention, are desperately needed. Rapid advances in molecular oncology over the past decade have provided us with new insights into the pathogenesis of lung cancer. Characterization of molecular changes in growth-regulating genes in lung cancer cells has been a first step toward identification of molecular markers of early lung cancer. In addition, we have made great strides in our understanding of the process of tumor promotion in lung cancer by the identification of specific substances as tumor promoters and the biochemical pathways by which they function. These insights will provide us with a beginning for the development of screening assays, early detection tests, and chemopreventive agents, hopefully allowing us to make an impact on lung cancer mortality.

Regulation of nuclear oncogenes expressed in lung cancer cell lines.

Lung cancer is a major cause of mortality in the United States and accounts for the majority of all cancer deaths in both men and women. It is hoped that through broadening our understanding of the mechanisms involved in transformation of bronchial epithelial cells we will be able to improve methods of diagnosis and treatment of this disease, with the ultimate goal of reducing on lung cancer mortality. A knowledge of the molecular mechanisms involved in processes such as cell division and differentiation is paramount to this task, because it is known that aberrant responses to growth factors or cytokines found in the normal cellular milieu can lead to abnormal cell growth and/or transformation. Signals initiated at the cell membrane by tumor promoters, growth factors, or cytokines are transduced from the cell membrane to the nucleus and are, in part, mediated centrally by transcription factors encoded by nuclear protooncogenes. The transcription factors myc, jun, and fos have been characterized in both normal and transformed lung epithelial cells through detailed studies using cell lines. In this manuscript, we review what is known about the expression and regulation of these nuclear protooncogenes in normal and malignant epithelial cells of the lung, and their role in the development of lung cancer.

Sensory control of the initiation of hatching in chicks: effects of a local anesthetic injected into the neck.

Previous work shows that folding a posthatching chick into the hatching position results in the re-initiation of hatching. Furthermore, bending the neck to the right or left serves as a selective signal for turning on hatching behavior. The present study addresses the issue of whether sensory receptors located in the neck provide this signal. Three groups of chicks were folded into the hatching position and placed in glass eggs. In the experimental group, sensory input from the neck was eliminated with a local anesthetic, lidocaine. In these chicks, hatching was initiated only after a long latency, correlated with the time at which the anesthetic wore off. In the two control groups, in which saline was injected into the neck or lidocaine was injected into the thigh, the latency was much shorter. Therefore sensory receptors located in the neck appear to provide input that serves as a selective signal for initiating hatching.

Neural control of limb coordination. I. Comparison of hatching and walking motor output patterns in normal and deafferented chicks.

Previous work has shown that the neural circuits underlying the leg movements of walking and hatching coexist in post-hatching chicks (Bekoff and Kauer, 1984). In the present study, quantitative analysis of leg EMGs shows that there are some similarities, but also significant differences, in the motor output patterns of walking and hatching. This study examines the effect of removing sensory feedback from the legs on the production of the distinctive leg motor patterns. The temporal characteristics and interlimb coordination of hatching and walking are little affected. However, major changes in intralimb motor output patterns are seen when compared to records from normal chicks. These changes fall into one of 2 categories. Some parameters show similar changes in both behaviors after deafferentation (e.g., increases in flexor burst durations and cycle period). This suggests that certain features of sensory input from the legs normally modulate the hatching and walking pattern-generating circuitry in similar ways. Other parameters show convergence. That is, these aspects of the 2 intralimb motor patterns become more similar to each other after removal of sensory input. This is consistent with the hypothesis that some feature of sensory input from the legs normally modulates one set of multiuse intralimb circuitry to produce different output patterns. In general, the walking pattern becomes more like hatching after deafferentation, rather than the reverse, which suggests that the hatching pattern is a more basic one. The maintenance of some residual differences in intralimb motor patterns after leg deafferentation suggests that other sources of modulation must also be involved, or that there are some additional elements of circuitry that are called into play during the normal production of walking and hatching.

Autoantibody-associated cross-reactive idiotype-bearing human B lymphocytes: distribution and characterization, including Ig VH gene and CD5 antigen expression.

Monoclonal antibodies (MoAbs) specific for autoantibody associated cross-reactive idiotypes (CRIs) frequently recognize the Igs of neoplastic B cells in patients with chronic lymphocytic leukemia (CLL) and/or Waldenstrom's macroglobulinemia. Very little is known regarding the normal B cells expressing CRIs (CRI-positive B cells). Using a variety of MoAbs against CRIs we investigated the distribution and topographic localization of CRI-positive B cells in normal adult human lymphoid tissues. We found that CRI-positive B cells represent a significant B-cell subpopulation expressing surface IgM (greater than 90%), IgG (approximately 5%), or IgA (approximately 2%). CRI-positive B cells are homogeneously distributed throughout all lymphoid tissues, accounting for 10% to 15% of all B lymphocytes, with the exception of the thymus, in which they represent the predominant B cell population. Immunophenotypic studies showed (1) that a small subpopulation (3.7% +/- 0.8%) of CRI-positive B cells are activated in vivo, based on CD25 and CD38 antigen expression; and (2) that approximately 50% of CRI-positive B cells express the 67-Kd pan-T-lymphocyte CD5 antigen, suggesting that the CRI-positive B-cell subset and the recently described CD5-positive B-cell subset are closely related. This hypothesis is supported by the fact that CRI-positive B cells produce oligo or polyreactive Igs, which are a characteristic feature of CD5-positive B cells, and also by the fact that both B-cell subpopulations appear to use similar and restricted Ig VH gene family members.

Retinoids in the chemoprevention of bladder cancer.

Bladder cancer is strongly related to tobacco use and is estimated to cause 54,500 new cancer cases and 11,700 deaths in the United States in 1998. Approximately two thirds of new US cases will be superficial tumors, predominantly low-grade papillary. After standard therapeutic resection (with or without intravesical therapy), the superficial bladder tumor recurrence rate is 30% to 70% within 12 months of resection. Morbidity is substantial, with frequent cystoscopy, recurrence, resections, and possible cystectomy for progression to invasive cancers. Therefore, new approaches, including chemoprevention, are needed. Data suggest that bladder carcinogenesis is a multi-step, multifocal (field effect) process, possibly involving the spread of premalignant clones--all of which are prerequisites for effective chemopreventive approaches. To date, retinoids are the best-studied chemopreventive agents in this site, achieving mixed clinical results (with 13-cis-retinoic acid and etretinate) in superficial bladder tumors. This review includes the epidemiology and biology of bladder carcinogenesis in addition to preclinical and clinical retinoid data, and focuses on the most promising avenue of current retinoid chemoprevention in the bladder: the potent apoptosis-inducing retinoid fenretinide (4-HPR), which currently is in three phase III trials.

Dominant-negative mutants of cJun inhibit AP-1 activity through multiple mechanisms and with different potencies.

We have previously described a dominant-negative mutant of cJun that lacks the transactivation domain (TAD) of cJun and prevents AP-1-mediated transcriptional activation by quenching endogenous Jun or Fos proteins. We now report the development of a panel of cJun mutants that have inactivating mutations in the TAD, DNA-binding domain (DBD), or leucine zipper domain. These mutants are all unable to activate transcription, but only TAD and DBD mutants function in a dominant-negative fashion by inhibiting both cJun-induced transcriptional activation and transformation induced by the tumor promoter 12-O-tetradecanoylphorbol-13-acetate in ras-transfected rat embryo cells. Although the TAD and DBD mutants both function as transdominant inhibitors, they work through different mechanisms and with different inhibitory potencies. The DBD mutants, which function by inhibiting DNA binding, are relatively weak inhibitors, whereas the TAD mutants inhibit by quenching and are much more potent. Dimerization assays demonstrate that mutations in the DBD decrease the dimerization affinity of these mutants with cJun. These results demonstrate that the most potent dominant-negative mutants of cJun are proteins that have intact DBDs and quench the activity of the endogenous transcription factors.

Molecular characterization of adenocarcinoma of the cervix.

In an attempt to characterize the molecular alterations of cervical adenocarcinoma, we analyzed 32 paraffin-embedded specimens for the presence of K-ras mutations, p53 overexpression, p16 and Rb protein expression, and the presence of HPV 16 and 18 DNA. Overall 25/32 (78%) of the tumors displayed an abnormality in at least one of these analyses. K-ras mutations were detected by PCR amplification and RFLP analysis in 3 tumors, including 2 at codon 12 and 1 at codon 61. p53 overexpression determined by immunohistochemistry was demonstrated with > 80% of tumor nuclei staining in 4 cases, 10-15% of nuclei staining in 3 cases, and < 1% of nuclei staining in 5 cases. The pattern of staining was diffuse in 6 cases, focal in 1 case, and scattered in 5 cases. Analysis of p16 protein expression in 23 specimens revealed 1 tumor with abnormal staining, while Rb protein expression was determined to be normal in all 25 tumors tested. HPV DNA, detected by PCR with type-specific primers, was found in 16 tumors (50%), including 7 (22%) with HPV 16 and 9 (28%) with HPV 18. There was no correlation among these abnormalities except that the presence of HPV and strong p53 overexpression (> 80% tumor nuclei staining) were mutually exclusive events. Clinical correlation demonstrated that p53 overexpression involving the majority of tumor cell nuclei is characteristic of advanced stage disease, while HPV positivity and activated ras genes are associated with early stage disease.