Frequent microsatellite alterations at chromosomes 9p21 and 3p14 in oral premalignant lesions and their value in cancer risk assessment.

To better understand genetic alterations in oral premalignant lesions, we examined 84 oral leukoplakia samples from 37 patients who had been enrolled in a chemoprevention trial. The samples were analyzed for two microsatellite markers located at chromosomes 9p21 and 3p14. Loss of heterozygosity (LOH) at either or both loci was identified in 19 of the 37 (51%) patients. Of these 19 patients, seven (37%) have developed head and neck squamous cell carcinoma (HNSCC) while only one of 18 (6%) of patients without LOH developed HNSCC. Our data suggest that clonal genetic alterations are common in oral premalignant lesions; that multiple genetic alterations have already occurred in oral premalignant lesions, allowing at least a focal clonal expansion; and that losses of the 9p21 and 3p14 regions may be related to early processes of tumorigenesis in HNSCC. These genetic alterations in premalignant tissues may serve as markers for cancer risk assessment.

Chromosome breakage at a major fragile site associated with P-glycoprotein gene amplification in multidrug-resistant CHO cells.

Recent studies of several drug-resistant Chinese hamster cell lines suggested that a breakage-fusion-bridge mechanism is frequently involved in the amplification of drug resistance genes. These observations underscore the importance of chromosome breakage in the initiation of DNA amplification in mammalian cells. However, the mechanism of this breakage is unknown. Here, we propose that the site of chromosome breakage consistent with the initial event of P-glycoprotein (P-gp) gene amplification via the breakage-fusion-bridge cycle in three independently established multidrug-resistant CHO cells was located at 1q31. This site is a major chromosome fragile site that can be induced by methotrexate and aphidicolin treatments. Pretreatments of CHO cells with methotrexate or aphidicolin enhanced the frequencies of resistance to vinca alkaloid and amplification of the P-gp gene. These observations suggest that chromosome fragile sites play a pivotal role in DNA amplification in mammalian cells. Our data are also consistent with the hypothesis that gene amplification can be initiated by stress-induced chromosome breakage that is independent of modes of action of cytotoxic agents. Drug-resistant variants may arise by their growth advantage due to overproduction of cellular target molecules via gene amplification.

Sequential increases in proliferating cell nuclear antigen expression in head and neck tumorigenesis: a potential biomarker.

BACKGROUND: Proliferating cell nuclear antigen (PCNA) is a 36-kd nuclear protein whose expression is associated with DNA synthesis and cell proliferation. Tumorigenesis in head and neck squamous cell carcinoma is proposed to be a multistep process; dysregulation of proliferation is a potential marker of this process. PURPOSE: PCNA dysregulation was analyzed in squamous cell carcinoma tissue samples containing premalignant lesions (hyperplasia and/or dysplasia) and in adjacent normal epithelium to better understand proliferative changes during head and neck tumor development. METHODS: Immunohistochemical analysis was performed on formalin-fixed, paraffin-embedded sections by using a monoclonal anti-PCNA antibody. PCNA expression was analyzed in 33 head and neck squamous cell carcinomas and in their adjacent premalignant lesions from different sites and compared with that in the control samples, which had never been exposed to first-hand tobacco smoke. PCNA expression was assessed by semiquantitative scoring (scale 0-3) in three epithelial layers (basal, parabasal, and superficial). The labeling index and the weighted mean index of PCNA expression were calculated. RESULTS: Normal epithelium adjacent to the tumor had much more proliferative activity than the controls: The weighted mean index of PCNA expression was four-fold higher in the basal layer and sixfold higher in the parabasal layer. PCNA expression increased as tissues progressed from adjacent normal epithelium to hyperplasia (P < .001), hyperplasia to dysplasia (P < .001), and dysplasia to squamous cell carcinoma (P = .065); the total increase in PCNA expression ranged from fourfold to 10-fold from adjacent normal epithelium to squamous cell carcinoma. PCNA expression was higher in the parabasal than in the basal layer at all premalignant stages (23 of 25 samples in adjacent normal epithelium, 12 of 13 in hyperplasia, and 17 of 22 in dysplasia). As the tissue progressed from normal through premalignant stages to squamous cell carcinomas, we observed not only incremental increases in the labeling index, but also incremental increases in PCNA expression per labeled cells. CONCLUSIONS: These results indicate that PCNA could be a useful biomarker for multistep carcinogenesis in head and neck cancer and may serve as an intermediate end point in chemopreventive trials.

Phenotype and genotype of advanced premalignant head and neck lesions after chemopreventive therapy.

BACKGROUND: The goal of chemoprevention is to reduce the risk of cancer development by reversing or blocking the tumorigenic process through the use of pharmacologic or natural agents. To determine the potential role of genetic alterations in assessing cancer risk and in evaluating the efficacy of chemopreventive agents, we studied 22 patients with advanced premalignant lesions of the head and neck who were part of a prospective cancer prevention trial that is investigating a regimen of 13-cis-retinoic acid, interferon alfa, and alpha-tocopherol administered for 12 months or until disease progression. METHODS: We used polymerase chain reaction analysis of microsatellite DNA sequences in cells from precancerous lesions to determine the frequencies of genetic alterations--namely, loss of heterozygosity (LOH) and microsatellite instability--at chromosomal loci that are commonly deleted in head and neck cancer. RESULTS: Prior to treatment, 17 (81%) of 21, eight (44%) of 18, and eight (42%) of 19 patients who were informative (i.e., heterozygous) at chromosomes 9p21, 3p14, and 17p13, respectively, exhibited LOH in at least one of their lesion biopsy specimens. Among nine patients who exhibited LOH at chromosome 9p21 in pretreatment biopsy specimens and who had completed at least 5 months of therapy, the genetic loss persisted in eight--including three of the four patients who exhibited complete histologic responses (i.e., no evidence of dysplasia in their biopsy specimens). IMPLICATION: Our data suggest that clinical and histologic assessments of the response to chemopreventive agents may be insufficient to determine their efficacy and that critical genetic alterations could be used as independent biomarkers to augment the ability to evaluate the efficacy of such agents.

Long-term impact of smoking on lung epithelial proliferation in current and former smokers.

BACKGROUND: Lung cancer risk remains elevated for many years after quitting smoking. To assess using proliferation indices in bronchial tissues as an intermediate endpoint biomarker in lung cancer chemoprevention trials, we determined the relationship between the extent, intensity, and cessation of tobacco smoking and proliferative changes in bronchial epithelial biopsy specimens. METHODS: Bronchial biopsy specimens were obtained from up to six epithelial sites in 120 current smokers (median pack-years, 42) and 207 former smokers (median pack-years, 40; median quit-years, 8.1). Sections from the paraffin-embedded specimens were stained with hematoxylin--eosin to determine the metaplasia index and with an antibody to Ki-67 to determine the proliferative (labeling) index for the basal and parabasal (Ki-67 PLI) layers. All statistical tests were two-sided. RESULTS: Biopsy sites with metaplasia had statistically significantly higher Ki-67-labeling indices than those without metaplasia (P<.001) in both current and former smokers. Increased proliferation was observed in multiple biopsy sites, with the average Ki-67 PLI of the subject strongly correlating with the metaplasia index (r =.72 for current smokers; P<.001), even in sites without metaplasia (r =.23 for current smokers; P<.001). In current smokers, the Ki-67 PLI was associated with the number of packs smoked/day (P =.02) but not with smoking years or pack-years. In subjects who had quit smoking, the Ki-67 PLI dropped statistically significantly within 1 year (P =.008) but remained detectable for more than 20 years, even in the absence of squamous metaplasia. CONCLUSION: Smoking appears to elicit a dose-related proliferative response in the bronchial epithelia of active smokers. Although the proliferative response decreased gradually in former smokers, a subset of individuals had detectable proliferation for many years and may benefit from targeted chemoprevention. Bronchial epithelial proliferation, measured by Ki-67, may provide a useful biomarker in the assessment of lung cancer risk and in the response to chemopreventive interventions.

p53 expressions: predicting recurrence and second primary tumors in head and neck squamous cell carcinoma.

BACKGROUND: The survival rate for head and neck squamous cell carcinoma remains poor despite therapeutic advances over the last two decades. For patients with disease confined to the head and neck, there are two major and biologically distinct patterns of treatment failures after definitive therapy: recurrence of primary disease and development of second primary tumors. Understanding the biological basis of patterns of treatment failure after definitive therapy is needed to guide the development of adjuvant treatment and strategies to prevent second primary tumors. PURPOSE: To determine whether expression of the p53 protein has prognostic significance and/or is associated with patterns of treatment failure, we examined protein expression in primary tumor specimens of patients with head and neck squamous cell carcinoma. METHODS: Immunohistochemical analysis with a monoclonal antibody (DO7) specific for p53 protein was used to detect expression of the protein in formalin-fixed, paraffin-embedded tumor samples from 69 head and neck cancer patients treated with definitive local therapy (surgery and/or radiotherapy) between January 1980 and October 1983 at The University of Texas M. D. Anderson Cancer Center. We quantitated p53 protein expression and assessed its association with duration of patient survival, patterns of treatment failure (recurrence of primary tumor and development of second primary tumor), and other clinical parameters. All reported P values resulted from two-sided statistical tests. RESULTS: We found detectable levels of p53 protein expression in the tumor cell nuclei of 41 of 69 patients. Thirty-six (52%) of 69 patients whose tumors exhibited p53 protein expression in greater than or equal to 10% of the cell nuclei were grouped as p53 positive, and 33 (48%) of 69 patients whose tumors exhibited less than 10% nuclear expression were groups as p53 negative. The clinical characteristics of the patients in the p53-positive, and p53-negative groups were well balanced. Overall survival was significantly lower, and the times to tumor recurrence, to second primary tumors, and to any treatment failure were significantly shorter in the p53-positive group that in the p53-negative group (P=.0002, P=.047, P=.003, and P=.0009, respectively), mainly because the p53 positivity was associated with earlier development of tumor recurrence and second primary tumors. The rate of second primary tumor development per person per year was also significantly higher in the p53-positive group that in the p53-negative group. By use of multivariate analysis according to the Cox regression model, p53 expression status was identified as the most significant predictor of overall survival duration (P=.007), time to tumor recurrence (P=.053), time to second primary tumors (P=.035), and time to any treatment failure (P=.004). CONCLUSIONS: The expression of p53 protein in primary head and neck squamous cell carcinoma was significantly predictive of shorter survival because of its association with earlier development of both tumor recurrence and second primary tumors. Thus, p53 expression may be a valuable marker for identifying individuals at high risk of developing a recurrence of primary disease and second primary tumors who may benefit from adjuvant therapy and chemoprevention after definitive local therapy.

Effects of N-(4-hydroxyphenyl)retinamide on hTERT expression in the bronchial epithelium of cigarette smokers.

BACKGROUND: Telomerase activation plays a critical role in tumorigenesis. To determine the role of telomerase in early lung carcinogenesis and as a potential biomarker in chemoprevention trials, we analyzed the expression of the human telomerase reverse transcriptase catalytic subunit (hTERT) in bronchial biopsy specimens from cigarette smokers who were enrolled in a randomized, double-blinded, placebo-controlled chemoprevention trial of N-(4-hydroxyphenyl)retinamide (4-HPR). METHODS: We obtained biopsy specimens from six predetermined sites in the bronchial tree from the 57 participants, before treatment and 6 months after treatment with 4-HPR or placebo. We used in situ hybridization to examine hTERT messenger RNA (mRNA) expression in 266 pretreatment (baseline) and post-treatment site-paired biopsy specimens from 27 patients in the 4-HPR-treated group and from 30 patients in the placebo-treated group. All statistical tests were two-sided. RESULTS: At baseline, 62.4% (95% confidence interval [CI] = 53.9% to 71%) of the biopsy specimens obtained from the group treated with 4-HPR and 65.2% (95% CI = 57.4% to 73.1%) of the biopsy specimens obtained from the placebo-treated group expressed hTERT mRNA. After 6 months, 45.6% (95% CI = 36.9% to 54.3%) of the biopsy specimens obtained from the 4-HPR-treated group and 68.1% (95% CI = 60.4% to 75.8%) of the biopsy specimens obtained from the placebo-treated group expressed hTERT mRNA. The reduction in hTERT expression observed between the two treatment groups over time was statistically significant (P =.01) when we used the biopsy site as the unit of analysis, but not when we used the individual as the unit of analysis (P =.37). CONCLUSIONS: Telomerase is frequently reactivated in the lungs of cigarette smokers. The modulation of hTERT expression in 4-HPR-treated smokers suggests that a novel molecular mechanism underlies the potential chemopreventive properties of 4-HPR. hTERT expression is a promising potential biomarker for risk assessment and for the evaluation of the efficacy of chemopreventive agents in lung carcinogenesis.

Clonal genetic alterations in the lungs of current and former smokers.

BACKGROUND AND PURPOSE: Genetic damage has been identified at multiple chromosomal sites (i.e., loci) in lung cancer cells. We questioned whether similar damage could be detected in the bronchial epithelial cells of chronic smokers who do not have this disease. METHODS: Biopsy specimens from six different bronchial regions were obtained from 54 chronic smokers (40 current smokers and 14 former smokers). The presence of squamous metaplasia and dysplasia (abnormal histologic changes) in the specimens was documented by examination of hematoxylin-eosin-stained sections, and a metaplasia index ([number of biopsy specimens with metaplasia/total number of biopsy specimens] x 100%) was calculated for each subject. Loss of heterozygosity (i.e., loss of DNA sequences from one member of a chromosome pair) involving microsatellite DNA at three specific loci-chromosome 3p14, chromosome 9p21, and chromosome 17p13-was evaluated by means of the polymerase chain reaction. Fisher's exact test and logistic regression analysis were used to assess the data. Reported P values are two-sided. RESULTS: Data on microsatellite DNA status at chromosomes 3p14, 9p21, and 17p13 were available for 54, 50, and 44 subjects, respectively. The numbers of individuals who were actually informative (i.e., able to be evaluated for a loss of heterozygosity) at the three loci were 36 (67%), 37 (74%), and 34 (77%), respectively. DNA losses were detected in 27 (75%), 21 (57%), and six (18%) of the informative subjects at chromosomes 3p14, 9p21, and 17p13, respectively. Fifty-one subjects were informative for at least one of the three loci, and 39 (76%) exhibited a loss of heterozygosity. Forty-two subjects were informative for at least two of the loci, and 13 (31%) exhibited losses at a minimum of two loci. Loss of heterozygosity at chromosome 3p14 was more frequent in current smokers (22 [88%] of 25 informative) than in former smokers (five [45%] of 11 informative) (P = .01) and in subjects with a metaplasia index greater than or equal to 15% (21 [91%] of 23 informative) than in subjects with a metaplasia index of less than 15% (six [46%] of 13 informative) (P = .003). In five informative individuals among nine tested nonsmokers, a loss of heterozygosity was detected in only one subject at chromosome 3p14 (P = .03), and no losses were detected at chromosome 9p21 (P = .05). CONCLUSIONS: Genetic alterations at chromosomal sites containing putative tumor-suppressor genes (i.e., 3p14 and the FHIT gene, 9p21 and the p16 gene [also known as CDKN2], and 17p13 and the p53 gene [also known as TP53]) occur frequently in the histologically normal or minimally altered bronchial epithelium of chronic smokers.

Heterogeneity in chromosome damage and repair rates after bleomycin in ataxia telangiectasia cells.

Cells derived from patients with ataxia telangiectasia (AT) are known to be exceptionally sensitive to ionizing radiation and chemotherapeutic agents such as bleomycin (BLM), neocarzinostatin, and etoposide. This increased sensitivity is manifested by high chromosome aberration frequencies after treatment. In order to probe the underlying basis for this phenomenon, the technique of premature chromosome condensation was used to determine whether the increased chromosome damage observed after bleomycin treatment is due to increased initial chromosome damage or to a decreased capability of these cells to repair chromosome damage. Five AT cell lines were brought to quiescence and treated with BLM, and initial chromosome damage and repair rates were determined in the G1 prematurely condensed chromosomes. AT cells exhibited increased aberration frequencies compared to normal human fibroblasts immediately after BLM treatment. The five AT cell lines were heterogeneous in the fast component of chromosome break repair, varying from a nearly normal fast repair component in one cell line to a nearly defunct fast repair component in two other AT cell lines. Thus, while the AT cell lines were heterogeneous in the basis for their chromosome breakage sensitivity, all AT cell lines tested showed increased residual chromosome damage after BLM treatment while still in the quiescent phase.

Mitomycin-induced chromatid breaks in HeLa cells: a consequence of incomplete DNA replication.

The formation of chromosome aberrations induced by alkylating agents such as mitomycin C has been shown to require the passage of the treated cell through S phase. However, the exact mechanisms by which mitomycin C-induced DNA lesions are translated into chromosome aberrations during S phase are not known. The purpose of these studies was to better understand the molecular basis of chromosome aberration formation after mitomycin C treatment. The morphology of metaphases of cells treated in G1 phase with mitomycin C resembled that of prematurely condensed chromosomes of S-phase cells. Consequently we postulated that chromosome aberrations resulted from cells reaching mitosis without completing DNA replication. This was tested by treating HeLa cells in G1 phase with mitomycin C and then analyzing these cells at mitosis for residual DNA damage and DNA content. Utilizing the DNA alkaline elution assay for DNA damage, we showed that HeLa cells progress through S phase into mitosis with intact DNA-DNA interstrand cross-links. These cross-links, originally induced into parental DNA, were associated equally with parental and newly replicated DNA at the time the cells reached mitosis. This suggests that recombinational events had taken place during the DNA replication process. Cells that were treated in G1 phase and allowed to proceed to mitosis in the presence of bromodeoxyuridine to density label newly replicated DNA were analyzed with cesium chloride density sedimentation. Unreplicated DNA was present in the mitotic cells of the treated populations but not in the untreated control cells. Further, flow cytometric measurements, made under hypotonic conditions in order to reduce chromatin condensation effects, demonstrated that the mitotic cells from the mitomycin C-treated populations contained 10-20% less DNA than untreated mitotic controls. These results indicate that chromosome breaks induced by mitomycin C are the result of cells reaching mitosis without having fully completed DNA replication.

Kinetics and extent of repair of bleomycin-induced chromosome damage in quiescent normal human fibroblasts and human mononuclear blood cells.

The kinetics of chromosome damage repair after bleomycin treatment was studied in quiescent mononuclear human blood cells and human fibroblasts using the technique of premature chromosome condensation. Quiescent cells were treated with bleomycin for 30 min, washed free of drug, and fused with mitotic HeLa cells after various repair times. The amount of chromosome damage remaining was assayed in the G1 prematurely condensed chromosomes. The chromosome repair kinetics profile exhibited fast- and slow-repair components. The fast chromosome repair component was apparent within 2 hr after bleomycin treatment, with a significant amount of repair having occurred within 30 min. The absolute rate of chromosome repair then significantly slowed beyond 2 hr after treatment. The rate of chromosome repair was slightly dependent on dose with a higher rate observed at a higher dose. Interestingly, while quiescent fibroblasts were more sensitive to bleomycin than were mononuclear blood cells, both cell populations exhibited similar repair kinetics. These results suggest the following: (a) quiescent human fibroblasts and mononuclear blood cells show similar chromosome repair kinetics after bleomycin treatment; (b) there exist both fast- and slow-repair components after bleomycin treatment; and (c) the rate of chromosome repair is dependent on the degree of initial damage.

Direct and indirect clastogenic activity of anthracenedione in Chinese hamster ovary cells.

The purpose of this study was to characterize the clastogenic activity of 1,4-dihydroxy-5,8-bis [[(2-[(2-hydroxyethyl)amino] ethyl)amino]]-9, 10-anthracenedione (NSC 301739), an antitumor compound now under clinical investigation. Chromosome damage in Chinese hamster ovary cells in G2 phase was assayed directly by the technique of premature chromosome condensation, and this damage was compared with the aberration levels determined when the G2 cells attained metaphase. 1,4-Dihydroxy-5,8-bis [[(2-[(2-hydroxyethyl)amino]ethyl)amino]]-9, 10-anthracenedione was observed to slow the progression of cells to mitosis and induce chromatid gaps, breaks, and exchanges directly in interphase cells. The frequency of gaps, breaks, and complete exchanges observed at metaphase were similar to those observed in G2 prematurely condensed chromosomes; however, the frequency of incomplete exchanges was increased in mitotic preparations. The additional exchanges appeared to result from chromosome stickiness occurring during chromosome condensation for metaphase. The chromosome attachments were strong and resulted in persistent chromosome bridges during anaphase. These results suggest that 1,4-dihydroxy-5,8-bis[[(2-[(2-hydroxyethyl)amino]ethyl)amino]]-9, 10-anthracenedione induces chromosome damage through both direct and indirect mechanisms.

Induction and repair of DNA and chromosome damage by neocarzinostatin in quiescent normal human fibroblasts.

In an attempt to understand better the molecular basis of chromosome aberration formation, neocarzinostatin (NCS)-induced damage and repair were compared at the chromosome and DNA levels. Chromosome damage and repair in quiescent normal human fibroblasts (PA2 and DET 550) were assayed by the premature chromosome condensation technique in which the G1 prematurely condensed chromosomes are condensed and easily enumerated. DNA damage was monitored by neutral DNA elution. NCS induced chromosome breaks directly in the G1 cells; thus, its clastogenic action does not require passage of cells through S phase. The dose-response curve for NCS-induced damage suggested a two-hit-type event in the formation of chromosome breaks. The half-time of chromosome repair was approximately 4.5 hr, and all but one of the chromosome breaks were repaired by 46 hr. Neither 1-beta-D-arabinofuranosylcytosine nor cycloheximide blocked the repair of either DNA or chromosome damage induced by NCS in quiescent human cells.

The nature of adriamycin-induced cytotoxicity in Chinese hamster cells as revealed by premature chromosome condensation.

The short- and long-term effects of adriamycin treatment on the cell cycle kinetics and chromosome structure of Chinese hamster ovary cells were investigated. Adriamycin treatment, either pulse or continuous, did not delay the progression of G1 cells into S phase but did prolong the duration of S and G2 phases. This effect was dose dependent, and the prolongation of G2 period was greater than that of S. The frequency of chromosomal aberrations induced by adriamycin was studied in the mitotic as well as in the prematurely condensed chromosomes (PCC) of the treated cells. The types of aberrations observed were chromosome stickiness, chromatid exchanges, breaks, and gaps. The frequency of aberrations continued to increase as cells progressed to mitosis even after the removal of drug from the medium. At prolonged intervals (24 to 72 hr) after a brief or extended treatment, the labeling indices of the treated populations decreased, indicating an eventual block in the cell cycle progression. Visualization of chromosomes of these blocked interphase cells by the PCC method revealed that normal G2 PCC were totally absent. We observed that the PCC of these blocked cells were abnormal in the following respects: (a) some of the G2 PCC were extensively damaged; (b) many of the S PCC failed to incorporate [3H]thymidine; (c) a significant fraction of these PCC were poorly condensed and hence it was difficult to determine whether they were in G1 or G2. These results indicate that there is a good correlation between damage to the genome and the inhibition of cell cycle progression after adriamycin treatment.

Induction of DNA strand breaks and cross-links by 2,5-diaziridinyl-3,6-bis(carboethoxyamino)-1,4-benzoquinone in Chinese hamster ovary cells.

The DNA-damaging effects of 2,5-diaziridinyl-3,6-bis(carboethoxyamino)-1,4-benzoquinone (AZQ) in Chinese hamster ovary cells were investigated. As determined by alkaline elution, DNA strand breaks were observed in cells treated with 50 microM AZQ for 2 hr. The single-strand break frequency was 31.3 +/- 5.3 (S.D.) rad equivalents. Strand breaks could also be detected at lower drug concentration if proteinase K treatment was included before DNA elution. In comparison, DNA cross-links were apparent in cells treated with as low as 6.25 microM AZQ. The cross-linking frequencies were 39.7, 124.3, 230.3, and 625.1 rad-equivalents for 6.25, 12.5, 25, and 50 microM AZQ, respectively. Both DNA-DNA and DNA-protein cross-links in AZQ-treated cells were revealed by the proteinase K assay. The DNA strand breaks induced by AZQ were rapidly rejoined within 1 hr after drug removal. DNA interstrand cross-links increased within the first 6 hr of postincubation and then slightly decreased by 12 hr, and most of the cross-links disappeared after cells were allowed to recover for 24 hr. DNA-protein cross-links were immediately formed during the drug treatment period and were gradually decreased after drug removal.

Aromatic DNA adducts in adjacent tissues of breast cancer patients: clues to breast cancer etiology.

The etiology of the majority of human breast cancers is unknown. Environmental factors have long been suspected to play a role, but no specific causative agent has been identified. If the hypothesis that environmental carcinogen exposure contributes to human breast cancer is true, carcinogen-DNA adducts would be expected to be present in human breast tissues. To address this possibility, aromatic DNA adducts were measured in 87 surgical specimens of normal human breast tissues from 87 breast cancer patients undergoing mastectomy using the nuclease P1-enhanced version of the 32P postlabeling assay. Breast tissue samples from 29 noncancer patients undergoing reduction mammoplasty served as controls. Whereas aromatic DNA adducts were detected in all tissue samples examined, the total adduct levels in cancer patients were significantly higher than that in noncancer controls [mean +/- SEM, 97.4 +/- 23.4/10(9) nucleotides (range, 3.8-1737.1) versus 18.1 +/- 11.6/10(9) nucleotides (range, 5.6-56.7), respectively; P < 0.01, t test and Mann-Whitney test]. This difference was not affected by the age distribution of the two groups. The typical smoking-related DNA adduct pattern (i.e., a diagonal radioactive zone) was observed in 29 of 87 tissues (17 of 17 current smokers, 5 of 8 former smokers, 4 of 52 nonsmokers, and 3 of 10 patients with unknown smoking status) and in 2 of 10 control tissues. It was of interest that a benzo(a)pyrene (BP)-like DNA adduct was observed in 36 normal adjacent breast tissues (41%), 27 of which were from nonsmokers. Levels of this BP-like adduct were extremely high (> 100/10(9) nucleotides) in 5 patients (4 nonsmokers and 1 smoker) and moderately high (> 10/10(9) nucleotides) in 13 other patients (8 nonsmokers and 5 smokers). One patient exhibited this adduct at a level of 1500/10(9) nucleotides, which is comparable to the highest level of total adducts reported in human tissues related to carcinogen exposure (e.g., cigarette smoking). In contrast, this adduct was absent (< 1/10(9) nucleotides) in all of the control tissues. Cochromatography and rechromatography analysis of DNA samples from human breast tissues and from MCF-7 cells treated with BP revealed that this adduct could be generated by BP exposure but is not the major BP 7,8-diol-9,10-epoxide-deoxyguanine adduct detected previously in animal tissues and human mammary epithelial cells. These findings support the hypothesis that environmental carcinogen exposure, in addition to cigarette smoking, may be associated with the etiology of human breast cancer.

Dysregulation of epidermal growth factor receptor expression in premalignant lesions during head and neck tumorigenesis.

The development of head and neck cancer, believed to result from field cancerization and a multistep process of tumorigenesis, is often associated with an accumulation of genotypic and phenotypic alterations. The phenotypic changes could be the result of dysregulation of growth control genes such as epidermal growth factor receptor (EGFR). With the goal of identifying a potential biomarker of the multistep process of tumorigensis, we studied specimens of 36 head and neck squamous cell carcinomas from 5 different sites that contained normal epithelia and/or premalignant lesions adjacent to the tumors. Almost all of the individuals from whom these specimens were obtained had been exposed to first-hand smoking and/or alcohol consumption. Using a monoclonal anti-EGFR antibody for immunohistochemical analysis on paraffin-embedded sections with attached 886 cells for internal control, the levels of EGFR expression were assessed by image analysis. The relative staining intensity of EGFR in normal epithelia adjacent to tumors was 2-fold higher than that in normal control epithelium (P = 0.021), suggesting that, even in histologically normal epithelium, EGFR was already up-regulated in tissues surrounding tumors. These findings supported the theory of field cancerization in head and neck tumorigenesis. As tissue progressed from normal tissue adjacent to tumor to hyperplasia and to dysplasia, EGFR expression remained elevated. However, in the step from dysplasia to squamous cell carcinoma, EGFR expression was further and dramatically up-regulated (P = 0.01). Therefore, these results indicate that EGFR dysregulation happens in two steps, the moderate up-regulation of EGFR expression in normal epithelium adjacent to tumor and the further up-regulation of EGFR expression in the change from dysplasia to squamous cell carcinoma. In summary, the studies presented here indicate that EGFR dysregulation might be a useful marker for identifying individuals at risk of tumor development and an intermediate end point in chemoprevention trials.

Effect of cell proliferation and chromatin conformation on intercalator-induced, protein-associated DNA cleavage in human brain tumor cells and human fibroblasts.

Antineoplastic intercalating agents such as 4'-(9-acridinylamino)methanesulfon-m-anisidide (m-AMSA) stabilize a cleavable complex between topoisomerase II and DNA. The production of protein-associated DNA cleavage in whole cells exposed to m-AMSA is thought to represent the cellular correlate of this topoisomerase II-mediated reaction. Protein-associated DNA cleavage can be quantified in mammalian cells by using alkaline elution technology. In an attempt to understand the impact of phenotypic and biochemical cellular characteristics on protein-associated DNA cleavage, we quantified m-AMSA-induced DNA cleavage in quiescent or proliferative normal human fibroblasts (cell strain 1508) and human glioblastoma cells (line T98G) as well as in asynchronously proliferating HeLa cells. The magnitude of DNA cleavage in quiescent fibroblasts and quiescent glioblastoma cells was identical and low relative to that observed in the HeLa cells. The magnitude of DNA cleavage was enhanced in both cell types following proliferation. This enhancement was greater in the glioblastoma cells than in the fibroblasts. These results were not due to alterations in cellular m-AMSA uptake. Chromatin was more elongated (open) in the quiescent glioblastoma cells than in the quiescent fibroblasts (as visualized by using the premature chromosome condensation assay), suggesting chromatin accessibility to drug per se may not be a critical determinant of the magnitude of m-AMSA-induced DNA cleavage. The onset of the enhanced m-AMSA-induced DNA cleavability that accompanied proliferation closely followed the formation of regions of localized chromatin decondensation, a late G1 event, and coincided with the onset of enhanced thymidine uptake, a marker for the onset of S phase. m-AMSA-induced cytotoxicity was also enhanced in proliferating compared with quiescent cells. The major finding of this study is that the cellular target for m-AMSA, putatively topoisomerase II, is more susceptible to drug action in proliferating cells than in quiescent cells. Effects of chromatin conformation or cellular phenotype upon topoisomerase II-mediated events such as m-AMSA-induced DNA cleavage are less certain.

Model for the formation of double minutes from prematurely condensed chromosomes of replicating micronuclei in drug-treated Chinese hamster ovary cells undergoing DNA amplification.

Double minutes (DM) have been associated with gene amplification in drug-resistant cells and tumor cells. However, the mechanisms by which DM are formed have not been elucidated. We present here a model to describe a possible mechanism of DM formation based on the observations made in two independent early drug-selected multidrug-resistant cell lines and from in vitro somatic cell fusion experiments between synchronized S- and M-phase cells. The multidrug-resistant cell lines contain both DM and amplified mdr (P-glycoprotein) gene. Cytogenetic analyses of cells at early stages of selection revealed the presence of a number of micronuclei in a subpopulation of these cells. These micronuclei were often asynchronous in their progression through the cell cycle. As a result, premature condensation of micronuclear chromatin was often observed in metaphase plates. The pulverized chromatin pattern seen in certain instances of S-phase prematurely condensed chromosomes displays a striking resemblance to DM structures. These DM-like structures are linked by replicating DNA as revealed by DNA labeling experiments. Somatic cell hybrids between S- and M-phase cells when grown in vitro demonstrated that S-phase prematurely condensed chromatin indeed gives rise to extra chromosomal structures in the successive cell generations. It is hypothesized that distinct DM-like structures may arise from the partially replicated and prematurely condensed S-phase chromosomes following their liberation as extra chromosomal entities after replication and/or recombination in the succeeding division cycle(s). The enrichment for DM containing specific genes in drug-resistant cells may result from the subsequent drug selections.

In vitro induction of benzo(a)pyrene diol epoxide-DNA adducts in peripheral lymphocytes as a susceptibility marker for human lung cancer.

Given the same exposure, DNA adduct profiles can be considered as a phenotypic marker for carcinogen metabolism and DNA repair, which may reflect individual susceptibility to chemical carcinogenesis. Based on this notion, we have established a straightforward assay that measures induced DNA adducts in peripheral lymphocytes exposed in vitro to a model carcinogen, benzo(a)pyrene diol epoxide (BPDE) by 32P-postlabeling. To test the hypothesis that the levels of induced DNA adducts are a predictor for cancer risk, we conducted a pilot study of 21 lung cancer patients and 41 healthy frequency-matched controls. We found that the peripheral lymphocytes of cancer patients tended to accumulate higher levels of BPDE-DNA adducts than controls did (mean +/- SE of relative adduct labeling x 10(7) value; 59.6 +/- 12.0 versus 39.4 +/- 6.1 for cases and controls, respectively; P = 0.09). Using the tertile relative adduct labeling value of controls (10 adducts/10(7) nucleotides) as the cutoff point, 18 of 21 cases and 23 of 41 controls distributed above this level (odds ratio, 4.7; 95% confidence interval, 1.2-18.5). In logistic regression analysis, the level of induced adduct was an independent risk factor (odds ratio, 6.4; 95% confidence interval, 1.3-29.4) after adjustment for the potential confounding factors, i.e., age, sex, ethnicity, and smoking. Stratified analyses showed that greater differences in DNA adduct levels induced by BPDE between cases and controls were observed in individuals younger than 65 years and in nonsmokers. Despite the small sample size, the significant association between the level of BPDE-induced DNA adducts and risk for lung cancer suggests that this assay is a promising method for further investigations.