This research effort has concentrated on mechanisms of interaction between chemical carcinogens, some of which are commonly-used drugs, and DNA. Topics under study include both the extent of DNA adduct formation and persistence, and the biological consequences of DNA damage in cultured cells, animal models and human tissues. Information on DNA adduct processing in nuclear and mitochondrial DNA are correlated with specific effects of exposure, including tumorigenesis, clinical response, specific toxicities and functional alterations in target organs and organelles. We are particularly interested in searching out themes that are common to both animal models and human subjects, with the intention of applying the knowledge gained to either enhance or reduce a specific effect in humans. The work encompasses issues relevant to both human environmental/occupational exposure and human clinical response. The compounds of intensive investigation include the antiretroviral nucleoside analogs zidovudine (AZT) and lamivudine (3TC) as well as the chemotherapeutic agents cisplatin and tamoxifen (TAM), and the polycyclic aromatic hydrocarbons (PAHs), which are environmental pollutants. Drug combinations that include one or two nucleoside analogs have become the standard of care for individuals infected with the Human Immunodeficiency Virus (HIV), and are given to both adults and to HIV-infected pregnant women to prevent vertical viral transmission. Genotoxicity and mitochondrial toxicity have been modeled in fetal Erythrocebus patas monkeys taken at term after in utero exposure to AZT or AZT plus 3TC, the doses and duration of which were chosen to mimic human exposures. Fetuses exposed to both drugs sustained twice the burden of DNA damage as fetuses exposed to AZT alone, as well as significant telomere shortening in virtually all organs. Mitochondrial toxicities in the fetal patas monkeys exposed in utero to the AZT/3TC combination included mitochondrial morphological damage by transmission electron microscopy (EM), changes in oxidative phosphorylation (OXPHOS) enzyme assays and depletion of mitochondrial (mt)DNA. The toxicity was more severe with AZT/3TC, compared to AZT alone. Genotoxicity and mitochondrial toxicity are also being followed during the first year of life in infant patas monkeys exposed in utero to human-equivalent nucleoside analog drug combinations. Infant monkeys are assessed by echocardiogram, and monthly muscle biopsies that provide tissues for EM, cytochemistry, mtDNA quantitation and measurement of OXPHOS enzymes. Lactic acid levels, levels and persistence of drug incorporated into blood cell DNA, and chromosomal aberrations in bone marrow are also being examined. Drug combinations under investigation include AZT plus 3TC, AZT plus ddI, D4T plus 3TC and D4T plus ddI. Two indicators of mitochondrial toxicity, mitochondrial morphology (by EM) and mtDNA quantity, were examined in umbilical cord and cord blood leukocytes taken from HIV-1-uninfected infants born to either HIV-1-infected women receiving Combivir (AZT plus 3TC) therapy during pregnancy (n=10), or HIV-1-uninfected women (n=9). By EM, moderate to severe mitochondrial morphological damage was observed in umbilical cords from 6 Combivir-exposed infants and none of the unexposed infants. In addition, mtDNA depletion of at least 35% was found in umbilical cord and cord blood of Combivir-exposed infants. The data suggest that, regardless of an unremarkable clinical presentation, some Combivir-exposed children sustain mitochondrial damage at birth. We have examined changes in gene expression induced by chronic AZT exposure in cultured human cervical epithelioid carcinoma (HeLa) cells. Gene expression was analyzed by NCI cDNA microarray and mitochondrial microarray. Up-regulation of several nuclear-encoded elements involved in mitochondrial function was found by cDNA microarray. The elements included 3 subunits of Cytochrome C, Superoxide dismutase, ATP synthase and 5 subunits of NADH ubiquinone oxidoreductase. Expression of genes (11 polypeptides) encoded in mitochondrial DNA in the same cells was examined by a mitochondrial microarray. Mitochondrial polypeptides upregulated in cells chronically-exposed to AZT included 5 subunits of NADH ubiquinone oxidoreductase, Cytochrome b, 3 subunits of Cytochrome oxidase and ATP synthase. The data suggest that cells exposed chronically to AZT experience an up-regulation of OXPHOS in an attempt to compensate for drug-DNA damage and inhibition of the mitochondrial polymerase . The microarray data are consistent with the known toxicity of the drug. In Linxian, China the esophageal/stomach cancer mortality rate is 20% and high levels of PAHs have been measured in ingested food. A semi-quantitative immunohistochemical staining method (ACIS) for PAH-DNA adducts has been used successfully to examine PAH-DNA adducts in esophageal biopsies taken in 1995. A second pilot study employed biopsies taken from 5 Linxian residents in 1985, and the ACIS was used to examine PAH-DNA adduct formation in the superficial squamous layers and basal layers of squamous epithelium. Four of the 5 samples had staining levels of 22.2-87.8 AUs/1000 nuclei in the basal layer, and 1.1-20.2 AUs/1000 nuclei in the superficial squamous layer. In parallel sections stained with BPDE-DNA-absorbed antiserum, the values for AUs/1000 nuclei were reduced to mean background levels of 0.50 for the basal layer and 0.07 for the squamous layer, indicating that the observed staining was specific for PAH-DNA adducts. Studies in the literature have found an association between ingestion of well-cooked meat and colon adenoma incidence, and demonstrated the presence of PAH's in meats cooked at high-temperature. A highly-sensitive chemiluminescence immunoassay (CIA) has been used to measure PAH-DNA adducts in DNA from human lymphocytes of colon adenoma patients and controls, to investigate potential PAH exposures in the etiology of colon cancer. The assays are nearly complete but final results await sample de-coding. The increase in endometrial cancer risk in women receiving tamoxifen (TAM) may be due to a hormonal or a genotoxic mechanism. To investigate a possible genotoxic mechanism human endometrial DNA samples were assayed by TAM-DNA CIA, and about 10% of human endometrial DNA samples from TAM-exposed women appeared to be positive with mean values of 1 adduct in 109 nucleotides. Currently the study comprises less than 40 samples and the results are still inconclusive. However, organ-DNA samples from 3 retired breeder cynomolgus monkeys exposed orally to 2 mg TAM/kg body weight/day for 28 days have been examined for TAM-DNA adduct formation using three different methods. Measurable levels of TAM-DNA adducts were found in multiple organs, including uterus, and values from the different methods correlated well. The monkey study suggests that humans may also form TAM-DNA adducts. A study was designed to explore the potential contribution to DNA repair by Waf1/Cip1 using p21Waf1/Cip1 null (KO), heterozygous (HET) and wild type (WT) cultured mouse keratinocytes. Cells were exposed to to 5M cisplatin for 8, 24, 48 and 72 hr and examined for cytotoxicity, cisplatin-DNA adduct formation, proliferation rate and apoptosis. The conclusions are as follows: a) WT cells have normal DNA repair capacity, low proliferation (15% at 24 hr) and undergo normal apoptosis (10% at 24-72 hr), so the major mechanisms for adduct removal are DNA repair and apoptosis; b) null cells have high proliferation (45% at 24 hr) but low apoptosis (1.5%) and presumably a weak DNA repair capacity, as DNA adduct levels in these cells were similar to those in the WT cells; c) HET cells had low DNA adduct levels, presumably because there was some DNA repair occurring, while the proliferation rate was moderate (24% at 24 hr) and the apoptosis rate was low (1.2%).
Showing the most recent 10 out of 36 publications
