The human prostate consists of the peripheral zone (PZ), transition zone (TZ) and central zone (CZ), with cancer occurring most often in the PZ. To investigate the hypothesis that the presence of polycyclic aromatic hydrocarbon (PAH)-DNA adducts might explain the elevated PZ susceptibility, matched tissue sets consisting of PZ and TZ were examined from men undergoing radical retropubic prostatectomy. Quantitative gene expression analysis of 27 individuals showed that levels of CYP1B1 and NAT1 were consistently higher in the PZ compared to TZ. Matched tissue sets from 19 individuals were analyzed for PAH-DNA adduct formation by immunohistochemistry (IHC), using antiserum elicited against DNA modified with r7,t8-dihydroxy-t-9,10-oxy-7,8,9,10-tetrahydro-benzo[a]pyrene (BPDE), and semi-quantitation of adducts using the Automated Cellular Imaging System (ACIS). PAH-DNA adduct levels were highest in glandular epithelial cells, and comparison of PZ and TZ showed no significant differences between zones. Although expression of some activating and/or detoxifying enzymes may be higher in the PZ, PAH-DNA adduct levels were similar in both zones, suggesting that factors other than PAH-DNA adducts may be responsible for promotion of tumor formation in the human prostate. Epidemiological studies have shown that, among women with cancer-associated human papillomoa virus (HPV) infection, smoking may increase the cervical cancer risk an additional 2- to 4-fold. To elucidate the contribution of smoking, we examined human cervical paraffin-embedded sections for the presence of PAH-DNA adducts using immunohistochemistry (IHC) by automated cellular imaging system (ACIS). The study, involving 142 coded samples of human cervix, has shown a broad range (25-191 adducts/108 nucleotides) for PAH-DNA adduct formation. There was, however, no association between smoking and PAH-DNA adduct levels, or between degree of cervical neoplasia and PAH-DNA adduct levels. The data support the contention that tobacco-associated cervical cancer in HPV-infected women has multifactorial origins. We are performing biomarker studies in esophageal samples from Linxian, China where the esophageal/stomach cancer mortality rate is 20%, high levels of polycyclic aromatic hydrocarbons (PAHs) have been measured in ingested food, and residents excrete high levels of 1-hydroxypyrene. Using IHC with ACIS for semi-quantitation of PAH-DNA adducts, we are investigating the potential risk for esophageal cancer conferred by the presence of esophageal PAH-DNA adducts. The nested case-control study design provides a powerful epidemiologic framework for this study. In addition, to model this human exposure in the mouse, we are breeding mice defective in DNA repair (XPA -/-) and p53 (+/-). These mice are susceptible to esophageal tumorigenesis upon PAH ingestion, and we are feeding them a carcinogenic concentration of BP in the diet. We are planning to investigate BP-DNA adduct formation in the esophagus, and to determine whether or not the addition of chlorophyllin (CHLN) to the BP-diet will reduce esophageal DNA adduct levels. If successful, the mouse experiments will elucidate the feasibility of using CHLN as a chemopreventive agent in the Linxian population. Normal human mammary epithelial cell (NHMEC) strains, cultured from human breast tissue obtained at reduction mammoplasty, provide a relevant model for investigation of human interindividual differences in carcinogen metabolism and DNA-damage response. In this study, cells of NHMEC strain M98013 were exposed to BP (4 microM BP) and CHLN (5 microM). Expression of the BP-metabolizing genes CYP1A1 and CYP1B1 was determined by microarray and real time (RT)-PCR, BPdG adducts were measured by BPDE-DNA chemiluminescence immunoassay (CIA) and EROD assay was used to measure CYP1A1 and CYP1B1 enzymes. CYP1A1 and CYP1B1 expression levels were increased in cells exposed to BP and decreased 70-85% in cells incubated with CHLN either with BP, before BP, or before and with BP exposure. Formation of BPdG adducts was reduced up to 85% in cells treated with CHLN, and the extent of adduct reduction correlated directly with the inhibition of CYP1A1 and CYP1B1 expression (p less than 0.01). The data suggest that CHLN may act not only by altering CYP1A1 and CYP1B1 expression levels but also by lowering the effective dose of carcinogen. To further elucidate the formation of BP-DNA adducts and understand PAH metabolism, 15 NHMEC strains and MCF-7 (human breast tumor) cells were exposed to 4 M BP. Cells were assayed for CYP1B1 and NQO1 transcripts by quantitative real-time PCR (qRT-PCR) and BP-DNA adducts by BPDE-DNA CIA. BP-DNA adduct levels (adducts/108 nucleotides) were 0.2-10.6 in NHMECs and 790 in MCF-7 cells exposed to BP for 12h. In unexposed cells, CYP1B1 abundance (transcripts/ng RNA [tpn]) was 569-3,452 in NHMECs, and 12,897 in MCF-7 cells. After 12h of BP exposure, CYP1B1 abundance increased 7- to 8-fold in NHMECs and 2.8-fold in MCF-7 cells. In unexposed cells, NQO1 gene abundance was 9,004-22,530 tpn in NHMECs, and 4,962 tpn in MCF-7 cells. After 12h of BP exposure, NQO1 abundance increased 1.1- to 1.4-fold in NHMECs, and 2.7-fold in MCF-7 cells. CYP1A1 and 1B1 enzyme activities were measured by EROD assay, and NQO1 enzyme activity was measured by dicumorol assay, in NHMEC strain M98016 and in MCF-7 cells. In unexposed MCF-7 cells, EROD activity was 4 fold higher than that found in the NHMECs. In BP-exposed NHMECs and MCF-7 cells EROD activity was increased 1.6- to 5-fold. NQO1 levels were essentially unchanged with BP exposure. The data indicate that the high BP-DNA adduct level observed in MCF-7 cells appears to be driven by high levels of CYP1B1 expression and enzyme activity, and low levels of NQO1 enzyme activity. It is possible that normal human breast cells are protected from BP-DNA damage, and possibly mutagenesis, by their high capacity for detoxification and comparatively low levels of activation. The potential formation of tamoxifen (TAM)-DNA adducts in human endometrium is a controversial topic of interest, as TAM-exposed women are at risk for endometrial cancer. Modeling of chronic human TAM exposures in female monkeys revealed TAM-DNA adduct formation in ovary and uterus, but our current efforts to measure TAM-DNA adducts in human endometrium have been inconclusive. Hypothesizing that TAM-induced gene expression changes may provide insight into breast tissue response, TAM-exposed cultured normal human mammary epithelial cell (NHMEC) strains, derived from reduction mammoplasty tissue taken from 3 individuals, were subjected to microarray. We found up-regulation of the same immune response genes participating in the STAT1/JAK-interferon pathway in all 3 NHMEC strains, suggesting a novel mechanism of TAM activity in the breast. In a second project we are using IHC/ACIS, with antiserum elicited against TAM-modified DNA, to investigate TAM-DNA adduct formation in human reproductive organ tissues taken from women at hysterectomy.
Showing the most recent 10 out of 36 publications
