Our studies of polycyclic aromatic hydrocarbon (PAH) carcinogenesis comprise cultured normal human mammary epithelial cells (NHMECs), MCF7 breast cancer cells, cancer-susceptible mice lacking the DNA repair gene XPA and heterozygous for the tumor promoter p53, and human subjects. PAHs, including the human carcinogen benzo(a)pyrene (BP), are considered to be activated to DNA binding species by metabolic enzymes which include Cytochrome P450 1A1 (CYP1A1), CYP1B1, and NAD(P)H:Quinone Oxidoreductase 1 (NQO1). Expression of these enzymes is typically evaluated by studies in which the data are presented as fold-change. However, because fold-change does not reveal the number of RNA transcripts in either basal or induced cells we have quantified RNA transcript levels (RNA copies/ng cDNA or RNA cpn) for basal and BP-induced CYP1A1, CYP1B1, and NQO1 genes. These genes are important in the formation of r7, t8, t9trihydroxyc10(N2-deoxyguanosyl)7,8,9,10-tetrahydrobenzo[a]pyrene (BPdG), the major BP-DNA adduct associated with cancer induction. In this study we used 16 strains of normal human mammary epithelial cells (NHMECs), cultured from reduction mammoplasty tissue obtained from 16 healthy women, with MCF-7 breast cancer cells as positive controls. Before BP exposure, basal RNA cpn values were 56 to 836 for CYP1A1, 336 to 5587 for CYP1B1, and 5943 to 40112 for NQO1. All cells were exposed to 4.0 micro M BP for 12h and after exposure RNA cpn values were 251 to 13234 for CYP1A1, 4133 to 57077 for CYP1B1, and 4456 to 55887 for NQO1. BPdG levels, measured by chemiluminescence immunoassay (CIA) varied from 0.85 to 15.8 adducts/108 nucleotides in the NHMECs and were 790 adducts/108 nucleotides in the MCF-7 cells. A strong linear association (p=0.0015) was observed between BPdG level and BP-induced CYP1A1 RNA cpn level in the NHMECs, but BP-induced levels of RNA cpn for CYP1B1 and NQO1 were not significantly correlated with BPdG level in these cells. Western Blots of 4 NHMEC strains, chosen for variability in BPdG level, showed a linear correlation between BPdG and CYP1A1 protein, but no correlation between CYP1B1 and NQO1. The data showed a complex relationship between expression of metabolic enzymes and BPdG formation in NHMECs exposed to BP. In Hepatic P450 Reductase Null (HRN) mice, P450 oxidoreductase (Por) is deleted specifically in hepatocytes, resulting in the loss of essentially all hepatic CYP450 function. HRN mice were used here to evaluate the contribution of CYP450 enzymes in the formation of BP-induced hepatic DNA adducts in mice. Our hypothesis was that the HRN mice should have fewer hepatic BPdG adducts than the WT mice. However, treatment of HRN mice with a single i.p. or oral dose of BP (12.5 or 125 mg/kg body weight) resulted in higher BPdG adduct levels in HRN mouse liver (up to 10-fold), compared to wild-type (WT) mouse liver, suggesting that in mouse liver P450s are more important for BP detoxification than BP-DNA adduct formation (activation). Cellular localization of BPdG adducts by immunohistochemistry confirmed that HRN mice have a higher capacity for formation of BP-induced DNA adducts in liver than their WT counterparts. In addition, the data showed that it is not likely that the metabolic pathways in HRN and WT mice involve generation of different reactive species. However, increased protein expression of cytochrome b5 in hepatic microsomes of HRN, relative to WT mice, suggest that cytochrome b5 may modulate the P450-mediated bioactivation of BP in HRN mice, partially substituting the function of Por. Overall the data show that multiple metabolic pathways may contribute to BPdG formation, and the dominance of one or another may depend on the species and cell type. We have evaluated DNA damage (DNA adduct formation) after feeding BP to wild type (WT) and cancer susceptible Xpa(-/-)p53(+/-) mice deficient in nucleotide excision repair and haploinsufficient for the tumor suppressor p53. DNA damage was evaluated by: high performance liquid chromatography/electrospray ionization tandem mass spectrometry (HPLC/ES-MS/MS), which measures BPdG;and a CIA, using anti-r7,t8-dihydroxy-t-9,10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE)?DNA antiserum, which measures both BPdG and the other stable BP-DNA adducts. When mice were fed 100 ppm BP for 28 days, BP-induced DNA damage measured in esophagus, liver, and lung, was typically higher in Xpa(-/-)p53(+/-) mice, compared to WT mice. This result is consistent with the previously-observed increased tumor susceptibility of Xpa(-/-)p53(+/-) mice. BPdG, the major DNA adduct associated with tumorigenicity, was the primary DNA adduct formed in esophagus (a target tissue in the mouse), while total BP-DNA adducts predominated in higher levels in the liver (a non-target tissue in the mouse). In an attempt to lower BP-induced DNA damage, we fed the WT and Xpa(-/-)p53(+/-) mice 0.3% chlorophyllin (CHL) in the BP-containing diet for 28 days. The addition of CHL resulted in an increase of BP-DNA adducts in esophagus, liver and lung of WT mice, a lowering of BPdG in esophagi of WT mice and livers of Xpa(-/-)p53(+/-) mice, and increased BPdG in livers of WT mice. Therefore, the addition of CHL to a BP-containing diet showed a lack of consistent chemoprotective effect, indicating that oral CHL administration may not reduce PAH-DNA adduct levels consistently in human organs. Tamoxifen (TAM) is a selective estrogen receptor modulator used worldwide for adjuvant therapy and chemoprevention of breast cancer. Women receiving TAM have an increased risk of endometrial and myometrial cancer, which may be due to genotoxicity or receptor-related mechanisms. Our TAM studies comprise experiments in NHMECs, immortalized human endometrial stromal cells, two species of non-hum,an primates and tissues taken from human patients. Controversy has surrounded the issue of whether or not TAM-DNA adducts form in humans, and we have examined this question in uterine tissues of aging Erythrocebus patas (patas) and Macaca fascicularis (macaque) monkeys given oral TAM dosing, as well as in endometrial and myometrial samples from women given TAM therapy. Adducts were determined by TAM-DNA chemiluminescence immunoassay (CIA) using an antiserum elicited against DNA modified with (E)-alpha-(deoxy-guanosin-N2-yl)-tamoxifen (dG-TAM). Of 5 female patas, 2 were unexposed and 3 were given oral dosing of 1.7 mg TAM/kg bw/day for 3 months. Macaques were either treated for 4 months with 1.3 mg TAM/kg bw/day (n=4) or unexposed (n=6). Normal and tumor uterine samples from women who received 20 mg/day (n=8) were analyzed along with samples (n=8) from unexposed women. We found 35 adducts / 108 nucleotides in the patas monkeys, 17 adducts / 108 nucleotides in the macaques, and 3 adducts /108 nucleotides in women. These data demonstrate that TAM-DNA adducts are formed in monkey uterus as well as human endometrium, myometrium and endometrial tumor, and suggest that TAM-DNA adduct formation may contribute to human endometrial cancer.
Showing the most recent 10 out of 18 publications
