Induction by 7,12-dimethylbenz(a)anthracene of molecular and biochemical alterations in transformed human mammary epithelial stem cells, and protection by N-acetylcysteine.

Several lines of evidence suggest that stem cells are major targets for carcinogens. A normal human breast epithelial cell type was previously shown to possess stem cell characteristics. Further cell lines were derived following sequential transfection with SV40 large T-antigen (immortal, non-tumorigenic M13SV1 cells), exposure to X-rays (weakly tumorigenic M13SV1R2 cells), and ectopic expression of c-erbB2/neu (highly tumorigenic M13SV1R2N1 cells). We evaluated some characteristics of these cells and their susceptibility to the breast carcinogen 7,12-dimethylbenz(a)anthracene (DMBA). Compared to M13SV1 cells, the two untreated tumorigenic cell lines displayed higher levels of connexin 43 expression and NF-kappaB nuclear translocation, and a higher frequency of fhit loss. The baseline nuclear translocation of AP-1 and pCREB was particularly evident in M13SV1R2N1 cells and was further enhanced by DMBA treatment, indicating an interaction between c-erbB2/neu and DMBA-induced signalling. Treatment with DMBA did neither affect the baseline fhit loss nor p53 mutation, whereas it increased NF-kappaB nuclear translocation, the proportion of apoptotic cells, and the levels of connexin 43, common 4977-bp mitochondrial DNA deletion, and bulky adducts to nuclear DNA. DMBA-treated M13SV1 cells underwent significant oxidative DNA damage and exhibited the highest DNA adduct levels, while they had the lowest apoptotic rate. Co-treatment of cells with N-acetylcysteine (NAC) attenuated DMBA-induced toxicity and DNA alterations, particularly in M13SV1 cells. Thus, the immortal cell line derived from the normal human adult breast stem cell without further tumorigenic progression is the most susceptible both to DMBA-related alterations and to the protective effects of NAC.

Modulation of multigene expression and proteome profiles by chemopreventive agents.

Analysis of transcriptome and proteome profiles by microarray technologies provides a formidable, new tool in cancer chemoprevention research. An ideal chemopreventive agent should not excessively alter per se the basal make-up of multigene expression and protein synthesis and should at the same time be able to attenuate alterations induced by risk factors. In order to validate this working hypothesis, we previously performed a series of studies in animal models using the thiol N-acetyl-l-cysteine (NAC) and the nonsteroidal antiinflammatory drug sulindac. We report herein the results of new studies evaluating modulation of DNA adduct levels and expression of 4858 genes in lung and liver of Sprague-Dawley rats, either unexposed or exposed to environmental cigarette smoke (ECS). The tested chemopreventive agents included NAC, oltipraz (OPZ), 5,6-benzoflavone (5,6-BF), phenethyl isothiocyanate (PEITC), and indole 3-carbinol (I3C). Combinations of OPZ with NAC and of PEITC with I3C were also assayed. Excepting OPZ, all treatments inhibited by at least 50% the formation of bulky DNA adducts in the lung of ECS-exposed rats. Hierarchical cluster analysis and principal component analysis allowed us to classify the agents according to their influence on basal gene expression and their ability to attenuate ECS-induced transcriptome alterations. PEITC and I3C were the most effective but the least safe agents. 5,6-BF displayed intermediate patterns. OPZ was poorly effective in lung and considerably altered the basal gene expression in liver. NAC had a medium efficacy and was the safest agent, as also supported by the analysis of 518 proteins in rat lung.

Chemoprevention of genome, transcriptome, and proteome alterations induced by cigarette smoke in rat lung.

Post-genomic methodologies have provided novel tools for evaluating safety and efficacy of cancer chemopreventive agents. We exposed rats to environmental cigarette smoke (ECS) for 28 days, with or without oral administration of N-acetylcysteine (NAC). As assessed by 32P-postlabelling, ECS caused a 10-fold increase of DNA adduct levels, which were significantly reduced by NAC. Of 518 proteins tested by antibody microarray, ECS stimulated 56 activities involved in stress response, protein removal, cell replication, apoptosis, phagocytosis, and immune response. NAC alone did not change the amounts of any protein, whereas it significantly decreased the amounts of 6 ECS-induced proteins. The intensity of expression of 278 related genes, assessed by cDNA microarray, was significantly correlated with protein amounts. These observed molecular alterations, which can be attenuated by NAC, represent in part adaptive responses and in part reflect mechanisms contributing to the pathogenesis of smoke-related diseases, including lung cancer, asthma, chronic bronchitis, and emphysema.

Gene expression in the lung of p53 mutant mice exposed to cigarette smoke.

We showed previously that p53 mutations play a role in cigarette smoke-related carcinogenesis not only in humans but also in A/J mice. In fact, (UL53-3 x A/J)F(1) mice, carrying a dominant-negative germ-line p53 mutation, responded to exposure to environmental cigarette smoke more efficiently than their wild-type (wt) littermate controls in terms of molecular alterations, cytogenetic damage, and lung tumor yield. To clarify the mechanisms involved, we analyzed by cDNA array the expression of 1,185 cancer-related genes in the lung of the same mice. Neither environmental cigarette smoke nor the p53 status affected the expression of the p53 gene, but the p53 mutation strikingly increased the basal levels of p53 nuclear protein in the lung. Environmental cigarette smoke increased p53 protein levels in wt mice only. The p53 mutation enhanced the expression of positive cell cycle regulators in sham-exposed mice, which suggests a physiologic protective role of p53. In environmental cigarette smoke-exposed mice, the p53 mutation resulted in a lack of induction of proapoptotic genes and in overexpression of genes involved in cell proliferation, signal transduction, angiogenesis, inflammation, and immune response. Mutant mice and wt mice reacted to environmental cigarette smoke in a similar manner regarding genes involved in metabolism of xenobiotics, multidrug resistance, and protein repair. Irrespective of the p53 status, environmental cigarette smoke poorly affected the expression of oncogenes, tumor suppressor genes, and DNA repair genes. Taken together, these findings may explain the increased susceptibility of p53 mutant mice to smoke-related alterations of intermediate biomarkers and lung carcinogenesis.

Proteomic analysis as related to transcriptome data in the lung of chromium(VI)-treated rats.

Assessing the parallelism between transcriptome data and proteome data represents one of the major challenges of post-genomic research. We evaluated the levels of 380 proteins in lung S12 fractions from Sprague-Dawley rats by antibody microarrays. Approximately half of these proteins were detectable under physiological conditions. There was a poor parallelism between mRNA and protein levels for a subset of 84 coinciding or related activities, whose gene expression had previously been investigated by cDNA array. The proportion of detectable proteins was almost twice as high as the proportion of transcriptionally active genes, which reflects the longer half-life of proteins compared to mRNA. Following the local stimulus provided by a short-term and high-dose exposure to sodium dichromate by the intra-tracheal route, 64 additional proteins were detectable in lung S12 fractions, and the correlation between gene expression and protein levels became significant. Sixteen proteins were increased more than twice following chromium(VI) administration. They included ten activities involved in the positive regulation of the cell cycle, three proteins involved in stress response and protein repair, two pro-apoptotic activities, and one protein involved in lipoprotein catabolism. An increase of P53 protein was detected by Western blot in lung nuclear fractions. Post-genomic analyses, highlighting the stimulation of defence mechanisms triggered by DNA damage, contribute to explain the previously reported discrepancy between the ability of chromium(VI) to induce oxidative stress and genotoxic damage in the lung and its failure to induce lung tumors under comparable experimental conditions. The proteome analysis showed a prominent role of apoptosis, counter-balanced by a positive regulation of the cell cycle aimed at replacing lost cells. In conclusion, our results suggest that, under basal conditions, mRNA undergoes a selective inactivation and post-transcriptional regulation resulting in de-coupling of transcriptome data and proteome data. However, this parallelism is re-established when the cell undergoes genotoxic damage.

Systemic genotoxic effects produced by light, and synergism with cigarette smoke in the respiratory tract of hairless mice.

No information is available on the interaction between cigarette smoke, the most important man-made carcinogen, and light, the most widespread natural carcinogen. In order to clarify this issue, SKH-1 hairless mice were exposed to environmental smoke and/or to the light emitted by sunlight-simulating halogen quartz bulbs. After 28 days, intermediate biomarkers were evaluated in skin, respiratory tract, bone marrow and peripheral blood. The results showed that, individually, the light produced extensive alterations not only in the skin but even at a systemic level, as shown by formation of bulky DNA adducts in both lung and bone marrow and induction of cytogenetic damage in bone marrow and peripheral blood erythrocytes. Smoke damaged the respiratory tract and produced significant alterations in the skin as well as an evident cytogenetic damage in both bone marrow and peripheral blood. Interestingly, as compared with exposure to smoke only, alternate daily cycles of exposure to both light and smoke significantly increased malondialdehyde concentrations and DNA adduct levels in lung and the frequency of micronuclei in pulmonary alveolar macrophages. The oral administration of sulindac, a non-steroidal anti-inflammatory drug, attenuated several biomarker alterations due to the combined exposure of mice to light and smoke. In conclusion, the light induces a systemic genotoxic damage, which is presumably due to the UV-mediated formation in the skin of long-lived derivatives, such as aldehydes. This damage may mechanistically be involved in light-related hematopoietic malignancies. In addition, the light displayed an insofar unsuspected synergism with smoke in the induction of DNA damage in the respiratory tract.

Modulation of cigarette smoke-related end-points in mutagenesis and carcinogenesis.

The epidemic of lung cancer and the increase of other tumours and chronic degenerative diseases associated with tobacco smoking have represented one of the most dramatic catastrophes of the 20th century. The control of this plague is one of the major challenges of preventive medicine for the next decades. The imperative goal is to refrain from smoking. However, chemoprevention by dietary and/or pharmacological agents provides a complementary strategy, which can be targeted not only to current smokers but also to former smokers and passive smokers. This article summarises the results of studies performed in our laboratories during the last 10 years, and provides new data generated in vitro, in experimental animals and in humans. We compared the ability of 63 putative chemopreventive agents to inhibit the bacterial mutagenicity of mainstream cigarette smoke. Modulation by ethanol and the mechanisms involved were also investigated both in vitro and in vivo. Several studies evaluated the effects of dietary chemopreventive agents towards smoke-related intermediate biomarkers in various cells, tissues and organs of rodents. The investigated end-points included metabolic parameters, adducts to haemoglobin, bulky adducts to nuclear DNA, oxidative DNA damage, adducts to mitochondrial DNA, apoptosis, cytogenetic damage in alveolar macrophages, bone marrow and peripheral blood erytrocytes, proliferation markers, and histopathological alterations. The agents tested in vivo included N-acetyl-L-cysteine, 1,2-dithiole-3-thione, oltipraz, phenethyl isothiocyanate, 5,6-benzoflavone, and sulindac. We started applying multigene expression analysis to chemoprevention research, and postulated that an optimal agent should not excessively alter per se the physiological background of gene expression but should be able to attenuate the alterations produced by cigarette smoke or other carcinogens. We are working to develop an animal model for the induction of lung tumours following exposure to cigarette smoke. The most encouraging results were so far obtained in models using A/J mice and Swiss albino mice. The same smoke-related biomarkers used in animal studies can conveniently be applied to human chemoprevention studies. We participated in trials evaluating the effects of N-acetyl-L-cysteine and oltipraz in smokers from Italy, The Netherlands, and the People's Republic of China. We are trying to develop a pharmacogenomic approach, e.g. based on genetic metabolic polymorphisms, aimed at predicting not only the risk of developing cancer but also the individual responsiveness to chemopreventive agents.