Using Salmonid Microarrays to Understand the Dietary Modulation of Car
http://www.100md.com
《毒物学科学杂志》
Department of Environmental and Occupational Health Sciences, University of Washington, Seattle, Washington 98105
ABSTRACT
The highlighted article in this issue by Tilton et al. (2005a) is an innovative approach to evaluate the modulation of estrogen receptor (ER) and aryl hydrocarbon (Ah)-receptor pathways as mechanisms underlying indole-3-carbinol (I3C) tumor promotion in rainbow trout (Onchorhynchus mykiss). I3C and its major in vivo component 3,3'-diindolylmethane (DIM) are potent tumor promoters that appear to target both of the aforementioned receptor pathways. However, the relative importance of I3C modulation of ER and AhR-dependent pathways in the promotion of rainbow trout hepatocarcinogenesis has not been established. Previously, researchers within this group reported that I3C promotes aflatoxin B1 (AFB1)-induced trout hepatocarcinogenesis post-initiation at low concentrations in the diet that induce the expression of vitellogenin, a downstream marker for the activation of ER-dependent pathways in fish. Furthermore, the promotional effects of I3C on AFB1 hepatocarcinogenesis in rainbow trout occur at concentrations that differentially induce vitellogenin, but not CYP1A expression. Interestingly, higher I3C concentrations induce the expression of both CYP1A and vitellogenin. Thus, the relative induction of vitellogenin and CYP1A expression, which are respective markers for activation of fish ER and AhR-mediated gene expression, suggest that these pathways may be important for tumor promotion by dietary I3C in trout. Understanding the complexities of I3C-mediated tumor promotion is essential from several perspectives. For example, there is the obvious need to increase our basic understanding of dietary modulation of carcinogenesis. In addition, I3C also exhibits significant antioxidant and cancer chemoprotective effects under certain experimental conditions and in certain models which have led to its recent marketing as a dietary supplement, as well as its development as a possible chemopreventive agent in humans.
Key Words: indole-3-carbinol; 3,3'-diindolylmethane; tumor promotion; aflatoxin B1; rainbow trout; toxicogenomics.
The utility of rainbow trout as an experimental model to study the mechanisms of dietary carcinogenesis has been firmly established via an impressive array of studies conducted at Oregon State University over the past 40 years, and largely through the NIEHS-funded Marine and Freshwater Biomedical Center. In this regard, Drs. Sinnhuber and Bailey recognized early on the promise of the rainbow trout as a model of exquisite sensitivity for cancer studies, subsequently publishing a number of reports arising from studies designed to reduce the risk of aflatoxin B1 (AFB1) liver cancer. The strengths of the trout model reside in its sensitivity to several classes of carcinogens, including AFB1, and well-described tumor pathology (Williams et al., 2003). In addition, trout exhibit very low spontaneous tumor backgrounds over the typical 9–12 month period it requires to produce tumors. Relative to rodents, trout have a low cost associated with husbandry for large-scale cancer studies. The ability to rear and maintain large numbers of clonally-derived trout have enabled Bailey and colleagues to design and conduct large-scale tumor studies requiring thousands of animals and that address statistically challenging dose-response questions. The development of clonal trout lines and triploid individuals has lent new approaches to the study of cancer genetics in this experimental model. An example of the experimental utility of the trout as a cancer model is evidenced by a report that identified indole-3-carbinol (I3C) as a tumor promoter for AFB1-initiated liver cancer at doses near those recommended for dietary supplementation in humans (Oganesian et al., 1999). A subsequent report of the largest animal tumor study ever conducted, which included 42,000 trout and which addressed dibenzo[a,l]pyrene (DBP) carcinogenesis to a 1:5,000 incidence, resulted in data predicting that a DBP dose producing 1 in 106 cancers is 1,000-fold higher than predicted by the conservative linear model (Williams et al., 2003). When similar studies are confirmed with other carcinogens, including those with genotoxic and potentially nongenotoxic mechanisms of action and with other cellular targets, the rainbow trout model has the ability to greatly impact the utilization of animal tumor data in the context of human risk assessment.
However, as the field of environmental carcinogenesis has moved to address a more mechanistic understanding of the actions of modulators of carcinogenesis, the lack of good genetic maps, genome sequences, and DNA-based reagents that are becoming more widely available for mammalian species has hindered applications of aquatic models to studies involving molecular mechanisms of chemical carcinogenesis. In contrast, the sequencing of whole genomes in other species has greatly accelerated development of tools such as microarray-based gene expression profiling that make it possible to profile simultaneously the expression of all the genes expressed in a particular organism or tissue. Application of this technology has been used to identify clusters of genes that respond differently to chemical treatment or between tissues and disease states. Phenotypic anchoring of gene expression profiles to a specific pathophysiology has allowed the investigator to uncover molecular pathways and cellular processes affected by chemical exposures and of importance during the process of carcinogenesis.
Recent technical advances in the field of fish genomics have provided a basis to build upon the established strength of the rainbow trout model. As a result, investigators in aquatic toxicology and environmental carcinogenesis are now employing microarray technology to examine gene expression patterns and gene classes of functional importance in environmental toxicology and carcinogenesis. Currently, at least two salmonid microarray platforms show considerable promise for studies in salmonids. A custom rainbow trout 70-mer oligonucleotide array containing 1400 genes pertinent to several disciplines including immunology, toxicology stress physiology, and endocrinology has been developed at Oregon State University (OSUrbt ver. 2.0) and has been applied to studies to address global molecular changes in the context of carcinogenesis in trout (Tilton et al., 2005b). In Canada, the Genomics Research on Atlantic Salmon Project (GRASP) project funded by Genome British Columbia (BC), the province of BC and Genome Canada, has led to the production of salmonid microarrays, including a recent 16,000-gene platform (GRASP 16k v 2.0). The GRASP platform has been largely selected from Atlantic salmon and rainbow trout expressed sequence tagged (EST) databases, but includes other salmonid express sequence tags as well (http://web.uvic.ca/cbr/grasp). Evaluation of the GRASP arrays has demonstrated their utility in a variety of experimental settings and across salmonids, largely because the majority of the elements on the arrays are 3' ESTs that are highly variable in structure relative to other genes within salmonid species, while also exhibiting over 95% homology across salmonids, thus minimizing cross-reactivity to nontarget genes (Rise et al., 2004).
The primary compound addressed by Tilton et al. (2005a) I3C, is a naturally occurring glucosinolate hydrolysis product found in several vegetables of the Brassica family, including cauliflower, broccoli, and cabbage, that has been touted for its cancer chemoprotective properties. In this regard, cruciferous vegetables and some of their specific compounds have been shown to modulate carcinogenesis in animals and humans. Among these agents is glucobrassican, which is hydrolyzed by endogenous plant enzymes to release I3C. This compound has been widely studied and has been shown to have chemopreventive effects against development of chemically induced tumors in several species and using a variety of initiation agents including AFB1 (Bailey et al., 1987), polycyclic aromatic hydrocarbons (Grubbs et al., 1995) nitroso-compounds (Grubbs et al., 1995), and heterocyclic aromatic amines (Mori et al., 1999). In addition, there is evidence to indicate that dietary I3C prevents the development of estrogen-enhanced cancers including breast (Grubbs et al., 1995) and cervical cancers (Yuan et al., 1999) in animals. This broad range of I3C cancer chemoprotection in various species, disease states, and organ sites, as well as its potent antioxidant activities (Shertzer and Senft, 2000), has supported studies targeting the development of I3C as a cancer chemopreventative agent, and also its consumer marketing as a dietary supplement.
However, despite its demonstrated chemoprotective effects, dietary I3C has paradoxically been found to promote tumor formation in multiple organs in rodents (Kim et al., 1997) and trout (Oganesian et al., 1999). In the case of AFB1-induced liver cancer in rainbow trout, the chemopreventive properties of I3C occur when the compound is administered in the diet concurrent, or prior to exposure to carcinogenic agents, thus appearing to blocking the initiation of DNA injury (Dashwood et al., 1991). In contrast, the tumor promoting effects of I3C occur appears to involve prolonged exposure when the compound is administered after the chemical initiation of DNA injury (Dashwood et al., 1991). Furthermore, it is possible that the promotional potency of I3C is at least as great as its potency as an anti-initiating agent (Oganesian et al., 1999). Because it is a relatively unstable molecule and undergoes oligomerization in the gut to several compounds, including the major acid-condensation product 3,3'-diindolylmethane (DIM), it is not been clear as to what extent the chemoprotective or tumor-promoting effects observed with I3C are mediated by conversion to DIM. Both I3C and DIM alter cell cycle progression, proliferation, and apoptosis and act as anti-estrogens in certain cell models, suggesting that these chemicals may also target other stages of carcinogenesis. I3C is generally considered to be an anti-estrogen and abrogates estradiol-mediated cellular and biochemical effects in estradiol-responsive cells and tissues, possibly via inhibition of transcription of E2-responsive genes through competition for coactivators or increasing the rate of ER degradation (Ashok et al., 2002). In contrast, studies with the DIM have shown it to have estrogenic activity by ligand-independent activation of ER in breast cancer cells (Riby et al., 2000) and via ER-dependent mechanisms in trout (Shilling et al., 2001). Clearly, the nature of the stimulatory and inhibitory effects of I3C and its acid-condensation products on estrogen-responsive gene transcription under conditions relevant to human cancer risk has not yet been clearly established. The fact that I3C is now being promoted as a dietary supplement, despite the fact that it is a potent tumor promoter under certain conditions, raises important questions on the molecular mechanisms of action of these agents.
With this background, we can turn to the highlighted work in which microarray technology and global gene network analysis was used to study commonality of chemical-mediated transcriptional effects to better understand the complex mechanisms of action of I3C and its major in vivo component DIM. The remarkably similar transcriptional responses, based on correlation analysis of trout liver gene expression profiles and on hierarchical clustering of gene responses by I3C, DIM, and estrogen (approximately 90% commonality among E2 and DIM gene expression profiles, and approximately 70% for E2 and DIM), indicate that these compounds may promote hepatocarcinogenesis via estrogenic mechanisms in trout liver. This hypothesis is supported by the observed upregulation of transcripts encoding vitellogenic liver proteins, which are among the most sensitive markers for estrogenic responses in fish species. Also indicative of an estrogenic mechanism of tumor promotion was the consistent downregulation of genes involved in redox regulation and lipid, glucose, and retinol homeostasis by the treatments. Similar responses have been reported by others in rats treated with 17-ethinylestradiol, a potent estrogen and tumor promoter (Stahlberg et al., 2005). The observed downregulation of genes important for angiogenesis, immune responses, and the formation of extracellular matrix in liver observed by the authors provides further insight as to the mechanisms of tumor promotion by I3C, which are also consistent with exposure to estrogens (Stahlberg et al., 2005). Interestingly, these data indicate a possible dual role for estrogens, simultaneously stimulating tumor growth, but also inhibition of tumor invasion. The fact that the vitellogenic response in DIM-exposed trout was greater than that observed for I3C, but with far less potency than observed for E2, is also noteworthy. Coupled with the observation that DIM was a more potent inducer of transcriptional effects than I3C in vivo, this further supports the hypothesis that DIM is the active in vivo component of I3C. These data may also partially explain the fact that I3C has both estrogenic and anti-estrogenic activities.
In addition to the contributions from the highlighted study on the nature of I3C-mediated tumor promotion, the data provided by Tilton et al. (2005a) using a model estrogen suggests similar regulatory processes in certain conserved genes across phyla. In this regard, it is important to consider that relative to the rodent literature, we know considerably less about gene regulation in most fish species, especially in regards to transcriptional control (e.g., role of cis- and trans- acting transcription factors) of fish genes involved in xenobiotic metabolism, apoptosis, DNA repair, and immune function. Thus, despite the wealth of valuable data generated from these experiments, it should be noted that there could be subsets of trout genes in molecular pathways whose expression was not differently regulated by I3C, DIM, or E2, but are nonetheless important with regards to mechanisms of action of these agents. Clearly, this issue is of relevance to microarray studies in all biological systems. Another challenge for microarray studies involving trout and other aquatic organisms is the pressing need to more efficiently translate the information on individual genes into knowledge of biological processes and pathways. Tilton et al. (2005a) as well as researchers in our laboratory and other aquatic groups, have begun to approach analysis of gene expression patterns based upon function using web-based ontology databases, allowing assigning of putative homolog descriptions followed by hierarchial clustering of gene expression profiles. However, the relative lack of information on most fish genomes (with the exceptions of zebrafish and Fugu) renders this a tedious process. Ultimately, visualization of biological annotation with expression data and allowing the encompassing thousands of genes in an interactive view would greatly enhance the power of these studies. In this regard, organizations including the Gene Ontology (GO) consortium and the GenMAPP organization have developed databases to visualize metabolic and signaling processes in rodents and humans. The ability to rapidly map biological pathways in rainbow trout would greatly improve our understanding of molecular processes affected by gene expression changes and would further enhance the utility of this already valuable model organism.
Lastly, it is worthwhile to discuss the implications of this work with regards to the possible premature marketing of I3C by some manufacturers as a dietary supplement. The highlighted study and other studies on I3C tumor promotion raise a cautionary note regarding the arbitrary use of this compound as a dietary supplement. Clearly, I3C and DIM have potent biochemical activities and affect a host of cellular processes important in physiological homeostasis. The seemingly dichotomous effects of these phytochemicals involving both potent chemoprotective and also tumor promoting capabilities warrant the continued careful and thoughtful investigation of the mechanisms of action of these agents in different species. Ultimately, the identification of the molecular pathways through which phytochemicals mediate receptor pathways should facilitate the assessment of the risk posed to humans by these compounds. Clearly, the culmination of 40 years of research dedicated to using the rainbow trout model in carcinogenesis research at Oregon State University has enriched our understanding of cancer pathogenesis, contributed to our prediction of risks associated with chemical carcinogenesis, and helped us better understand the role of phytochemicals in preventing and treating disease.
REFERENCES
Ashok, B. T., Chen, Y. G., Liu, X., Garikapaty, V. P., Seplowitz, R., Tschorn, J., Roy, K., Mittelman, A., and Tiwari, R. K. (2002). Multiple molecular targets of indole-3-carbinol, a chemopreventive anti-estrogen in breast cancer. Eur. J. Cancer Prev. 11(Suppl. 2), S86–S93.
Bailey, G., Selivonchick, D., and Hendricks, J. (1987). Initiation, promotion, and inhibition of carcinogenesis in rainbow trout. Environ. Health Perspect. 71, 147–153.
Dashwood, R. H., Fong, A. T., Williams, D. E., Hendricks, J. D., and Bailey, G. S. (1991). Promotion of aflatoxin B1 carcinogenesis by the natural tumor modulator indole-3-carbinol: Influence of dose, duration, and intermittent exposure on indole-3-carbinol promotional potency. Cancer Res. 51, 2362–2365.
Grubbs, C. J., Steele, V. E., Casebolt, T., Juliana, M. M., Eto, I., Whitaker, L. M., Dragnev, K. H., Kelloff, G. J., and Lubet, R. L. (1995). Chemoprevention of chemically-induced mammary carcinogenesis by indole-3-carbinol. Anticancer Res. 15, 709–716.
Kim, D. J., Han, B. S., Ahn, B., Hasegawa, R., Shirai, T., Ito, N., and Tsuda, H. (1997). Enhancement by indole-3-carbinol of liver and thyroid gland neoplastic development in a rat medium-term multiorgan carcinogenesis model. Carcinogenesis 18, 377–381.
Mori, H., Sugie, S., Rahman, W., and Suzui, N. (1999). Chemoprevention of 2-amino-1-methyl-6-phenylimidazo [4,5-b]pyridine-induced mammary carcinogenesis in rats. Cancer Lett. 143, 195–198.
Oganesian, A., Hendricks, J. D., Pereira, C. B., Orner, G. A., Bailey, G. S., and Williams, D. E. (1999). Potency of dietary indole-3-carbinol as a promoter of aflatoxin B1-initiated hepatocarcinogenesis: Results from a 9000 animal tumor study. Carcinogenesis 20, 453–458.
Riby, J. E., Chang, G. H., Firestone, G. L., and Bjeldanes, L. F. (2000). Ligand-independent activation of estrogen receptor function by 3,3'-diindolylmethane in human breast cancer cells. Biochem. Pharmacol. 60, 167–177.
Rise, M. L., von Schalburg, K. R., Brown, G. D., Mawer, M. A., Devlin, R. H., Kuipers, N., Busby, M., Beetz-Sargent, M., Alberto, R., Gibbs, A. R., et al. (2004). Development and application of a salmonid EST database and cDNA microarray: Data mining and interspecific hybridization characteristics. Genome Res. 14, 478–490.
Shertzer, H. G., and Senft, A. P. (2000). The micronutrient indole-3-carbinol: Implications for disease and chemoprevention. Drug Metabol. Drug Interact. 17, 159–188.
Shilling, A. D., Carlson, D. B., Katchamart, S., and Williams, D. E. (2001). 3,3'-Diindolylmethane, a major condensation product of indole-3-carbinol, is a potent estrogen in the rainbow trout. Toxicol. Appl. Pharmacol. 170, 191–200.
Stahlberg, N., Merino, R., Hernandez, L. H., Fernandez-Perez, L., Sandelin, A., Engstrom, P., Tollet-Egnell, P., Lenhard, B., and Flores-Morales, A. (2005). Exploring hepatic hormone actions using a compilation of gene expression profiles. BMC Physiol. 5, 8.
Tilton, S. C., Givan, S. A., Pereira, C. B., Bailey, G. S., and Williams, D. E. (2005a). Toxicogenomic profiling of the hepatic tumor promoters indole-3-carbinol, 17-estradiol and -naphthoflavone in rainbow trout. ToxSci, 10.1093/toxsci/kfi341. Advance Access publication September 28, 2005.
Tilton, S. C., Gerwick, L. G., Hendricks, J. D., Rosato, C. S., Corley-Smith, G., Givan, S. A., Bailey, G. S., Bayne, C. J., and Williams, D. E. (2005b). Use of a rainbow trout oligonucleotide microarray to determine transcriptional patterns in aflatoxin B1-induced hepatocellular carcinoma compared to adjacent liver. Toxicol. Sci. 88, 319–330.
Williams, D. E., Bailey, G. S., Reddy, A., Hendricks, J. D., Oganesian, A., Orner, G. A., Pereira, C. B., and Swenberg, J. A. (2003). The rainbow trout (Oncorhynchus mykiss) tumor model: Recent applications in low-dose exposures to tumor initiators and promoters. Toxicol. Pathol. 31(Suppl.), 58–61.
Yuan, F., Chen, D. Z., Liu, K., Sepkovic, D. W., Bradlow, H. L., and Auborn, K. (1999). Anti-estrogenic activities of indole-3-carbinol in cervical cells: Implication for prevention of cervical cancer. Anticancer Res. 19, 1673–1680.(Evan P. Gallagher)
ABSTRACT
The highlighted article in this issue by Tilton et al. (2005a) is an innovative approach to evaluate the modulation of estrogen receptor (ER) and aryl hydrocarbon (Ah)-receptor pathways as mechanisms underlying indole-3-carbinol (I3C) tumor promotion in rainbow trout (Onchorhynchus mykiss). I3C and its major in vivo component 3,3'-diindolylmethane (DIM) are potent tumor promoters that appear to target both of the aforementioned receptor pathways. However, the relative importance of I3C modulation of ER and AhR-dependent pathways in the promotion of rainbow trout hepatocarcinogenesis has not been established. Previously, researchers within this group reported that I3C promotes aflatoxin B1 (AFB1)-induced trout hepatocarcinogenesis post-initiation at low concentrations in the diet that induce the expression of vitellogenin, a downstream marker for the activation of ER-dependent pathways in fish. Furthermore, the promotional effects of I3C on AFB1 hepatocarcinogenesis in rainbow trout occur at concentrations that differentially induce vitellogenin, but not CYP1A expression. Interestingly, higher I3C concentrations induce the expression of both CYP1A and vitellogenin. Thus, the relative induction of vitellogenin and CYP1A expression, which are respective markers for activation of fish ER and AhR-mediated gene expression, suggest that these pathways may be important for tumor promotion by dietary I3C in trout. Understanding the complexities of I3C-mediated tumor promotion is essential from several perspectives. For example, there is the obvious need to increase our basic understanding of dietary modulation of carcinogenesis. In addition, I3C also exhibits significant antioxidant and cancer chemoprotective effects under certain experimental conditions and in certain models which have led to its recent marketing as a dietary supplement, as well as its development as a possible chemopreventive agent in humans.
Key Words: indole-3-carbinol; 3,3'-diindolylmethane; tumor promotion; aflatoxin B1; rainbow trout; toxicogenomics.
The utility of rainbow trout as an experimental model to study the mechanisms of dietary carcinogenesis has been firmly established via an impressive array of studies conducted at Oregon State University over the past 40 years, and largely through the NIEHS-funded Marine and Freshwater Biomedical Center. In this regard, Drs. Sinnhuber and Bailey recognized early on the promise of the rainbow trout as a model of exquisite sensitivity for cancer studies, subsequently publishing a number of reports arising from studies designed to reduce the risk of aflatoxin B1 (AFB1) liver cancer. The strengths of the trout model reside in its sensitivity to several classes of carcinogens, including AFB1, and well-described tumor pathology (Williams et al., 2003). In addition, trout exhibit very low spontaneous tumor backgrounds over the typical 9–12 month period it requires to produce tumors. Relative to rodents, trout have a low cost associated with husbandry for large-scale cancer studies. The ability to rear and maintain large numbers of clonally-derived trout have enabled Bailey and colleagues to design and conduct large-scale tumor studies requiring thousands of animals and that address statistically challenging dose-response questions. The development of clonal trout lines and triploid individuals has lent new approaches to the study of cancer genetics in this experimental model. An example of the experimental utility of the trout as a cancer model is evidenced by a report that identified indole-3-carbinol (I3C) as a tumor promoter for AFB1-initiated liver cancer at doses near those recommended for dietary supplementation in humans (Oganesian et al., 1999). A subsequent report of the largest animal tumor study ever conducted, which included 42,000 trout and which addressed dibenzo[a,l]pyrene (DBP) carcinogenesis to a 1:5,000 incidence, resulted in data predicting that a DBP dose producing 1 in 106 cancers is 1,000-fold higher than predicted by the conservative linear model (Williams et al., 2003). When similar studies are confirmed with other carcinogens, including those with genotoxic and potentially nongenotoxic mechanisms of action and with other cellular targets, the rainbow trout model has the ability to greatly impact the utilization of animal tumor data in the context of human risk assessment.
However, as the field of environmental carcinogenesis has moved to address a more mechanistic understanding of the actions of modulators of carcinogenesis, the lack of good genetic maps, genome sequences, and DNA-based reagents that are becoming more widely available for mammalian species has hindered applications of aquatic models to studies involving molecular mechanisms of chemical carcinogenesis. In contrast, the sequencing of whole genomes in other species has greatly accelerated development of tools such as microarray-based gene expression profiling that make it possible to profile simultaneously the expression of all the genes expressed in a particular organism or tissue. Application of this technology has been used to identify clusters of genes that respond differently to chemical treatment or between tissues and disease states. Phenotypic anchoring of gene expression profiles to a specific pathophysiology has allowed the investigator to uncover molecular pathways and cellular processes affected by chemical exposures and of importance during the process of carcinogenesis.
Recent technical advances in the field of fish genomics have provided a basis to build upon the established strength of the rainbow trout model. As a result, investigators in aquatic toxicology and environmental carcinogenesis are now employing microarray technology to examine gene expression patterns and gene classes of functional importance in environmental toxicology and carcinogenesis. Currently, at least two salmonid microarray platforms show considerable promise for studies in salmonids. A custom rainbow trout 70-mer oligonucleotide array containing 1400 genes pertinent to several disciplines including immunology, toxicology stress physiology, and endocrinology has been developed at Oregon State University (OSUrbt ver. 2.0) and has been applied to studies to address global molecular changes in the context of carcinogenesis in trout (Tilton et al., 2005b). In Canada, the Genomics Research on Atlantic Salmon Project (GRASP) project funded by Genome British Columbia (BC), the province of BC and Genome Canada, has led to the production of salmonid microarrays, including a recent 16,000-gene platform (GRASP 16k v 2.0). The GRASP platform has been largely selected from Atlantic salmon and rainbow trout expressed sequence tagged (EST) databases, but includes other salmonid express sequence tags as well (http://web.uvic.ca/cbr/grasp). Evaluation of the GRASP arrays has demonstrated their utility in a variety of experimental settings and across salmonids, largely because the majority of the elements on the arrays are 3' ESTs that are highly variable in structure relative to other genes within salmonid species, while also exhibiting over 95% homology across salmonids, thus minimizing cross-reactivity to nontarget genes (Rise et al., 2004).
The primary compound addressed by Tilton et al. (2005a) I3C, is a naturally occurring glucosinolate hydrolysis product found in several vegetables of the Brassica family, including cauliflower, broccoli, and cabbage, that has been touted for its cancer chemoprotective properties. In this regard, cruciferous vegetables and some of their specific compounds have been shown to modulate carcinogenesis in animals and humans. Among these agents is glucobrassican, which is hydrolyzed by endogenous plant enzymes to release I3C. This compound has been widely studied and has been shown to have chemopreventive effects against development of chemically induced tumors in several species and using a variety of initiation agents including AFB1 (Bailey et al., 1987), polycyclic aromatic hydrocarbons (Grubbs et al., 1995) nitroso-compounds (Grubbs et al., 1995), and heterocyclic aromatic amines (Mori et al., 1999). In addition, there is evidence to indicate that dietary I3C prevents the development of estrogen-enhanced cancers including breast (Grubbs et al., 1995) and cervical cancers (Yuan et al., 1999) in animals. This broad range of I3C cancer chemoprotection in various species, disease states, and organ sites, as well as its potent antioxidant activities (Shertzer and Senft, 2000), has supported studies targeting the development of I3C as a cancer chemopreventative agent, and also its consumer marketing as a dietary supplement.
However, despite its demonstrated chemoprotective effects, dietary I3C has paradoxically been found to promote tumor formation in multiple organs in rodents (Kim et al., 1997) and trout (Oganesian et al., 1999). In the case of AFB1-induced liver cancer in rainbow trout, the chemopreventive properties of I3C occur when the compound is administered in the diet concurrent, or prior to exposure to carcinogenic agents, thus appearing to blocking the initiation of DNA injury (Dashwood et al., 1991). In contrast, the tumor promoting effects of I3C occur appears to involve prolonged exposure when the compound is administered after the chemical initiation of DNA injury (Dashwood et al., 1991). Furthermore, it is possible that the promotional potency of I3C is at least as great as its potency as an anti-initiating agent (Oganesian et al., 1999). Because it is a relatively unstable molecule and undergoes oligomerization in the gut to several compounds, including the major acid-condensation product 3,3'-diindolylmethane (DIM), it is not been clear as to what extent the chemoprotective or tumor-promoting effects observed with I3C are mediated by conversion to DIM. Both I3C and DIM alter cell cycle progression, proliferation, and apoptosis and act as anti-estrogens in certain cell models, suggesting that these chemicals may also target other stages of carcinogenesis. I3C is generally considered to be an anti-estrogen and abrogates estradiol-mediated cellular and biochemical effects in estradiol-responsive cells and tissues, possibly via inhibition of transcription of E2-responsive genes through competition for coactivators or increasing the rate of ER degradation (Ashok et al., 2002). In contrast, studies with the DIM have shown it to have estrogenic activity by ligand-independent activation of ER in breast cancer cells (Riby et al., 2000) and via ER-dependent mechanisms in trout (Shilling et al., 2001). Clearly, the nature of the stimulatory and inhibitory effects of I3C and its acid-condensation products on estrogen-responsive gene transcription under conditions relevant to human cancer risk has not yet been clearly established. The fact that I3C is now being promoted as a dietary supplement, despite the fact that it is a potent tumor promoter under certain conditions, raises important questions on the molecular mechanisms of action of these agents.
With this background, we can turn to the highlighted work in which microarray technology and global gene network analysis was used to study commonality of chemical-mediated transcriptional effects to better understand the complex mechanisms of action of I3C and its major in vivo component DIM. The remarkably similar transcriptional responses, based on correlation analysis of trout liver gene expression profiles and on hierarchical clustering of gene responses by I3C, DIM, and estrogen (approximately 90% commonality among E2 and DIM gene expression profiles, and approximately 70% for E2 and DIM), indicate that these compounds may promote hepatocarcinogenesis via estrogenic mechanisms in trout liver. This hypothesis is supported by the observed upregulation of transcripts encoding vitellogenic liver proteins, which are among the most sensitive markers for estrogenic responses in fish species. Also indicative of an estrogenic mechanism of tumor promotion was the consistent downregulation of genes involved in redox regulation and lipid, glucose, and retinol homeostasis by the treatments. Similar responses have been reported by others in rats treated with 17-ethinylestradiol, a potent estrogen and tumor promoter (Stahlberg et al., 2005). The observed downregulation of genes important for angiogenesis, immune responses, and the formation of extracellular matrix in liver observed by the authors provides further insight as to the mechanisms of tumor promotion by I3C, which are also consistent with exposure to estrogens (Stahlberg et al., 2005). Interestingly, these data indicate a possible dual role for estrogens, simultaneously stimulating tumor growth, but also inhibition of tumor invasion. The fact that the vitellogenic response in DIM-exposed trout was greater than that observed for I3C, but with far less potency than observed for E2, is also noteworthy. Coupled with the observation that DIM was a more potent inducer of transcriptional effects than I3C in vivo, this further supports the hypothesis that DIM is the active in vivo component of I3C. These data may also partially explain the fact that I3C has both estrogenic and anti-estrogenic activities.
In addition to the contributions from the highlighted study on the nature of I3C-mediated tumor promotion, the data provided by Tilton et al. (2005a) using a model estrogen suggests similar regulatory processes in certain conserved genes across phyla. In this regard, it is important to consider that relative to the rodent literature, we know considerably less about gene regulation in most fish species, especially in regards to transcriptional control (e.g., role of cis- and trans- acting transcription factors) of fish genes involved in xenobiotic metabolism, apoptosis, DNA repair, and immune function. Thus, despite the wealth of valuable data generated from these experiments, it should be noted that there could be subsets of trout genes in molecular pathways whose expression was not differently regulated by I3C, DIM, or E2, but are nonetheless important with regards to mechanisms of action of these agents. Clearly, this issue is of relevance to microarray studies in all biological systems. Another challenge for microarray studies involving trout and other aquatic organisms is the pressing need to more efficiently translate the information on individual genes into knowledge of biological processes and pathways. Tilton et al. (2005a) as well as researchers in our laboratory and other aquatic groups, have begun to approach analysis of gene expression patterns based upon function using web-based ontology databases, allowing assigning of putative homolog descriptions followed by hierarchial clustering of gene expression profiles. However, the relative lack of information on most fish genomes (with the exceptions of zebrafish and Fugu) renders this a tedious process. Ultimately, visualization of biological annotation with expression data and allowing the encompassing thousands of genes in an interactive view would greatly enhance the power of these studies. In this regard, organizations including the Gene Ontology (GO) consortium and the GenMAPP organization have developed databases to visualize metabolic and signaling processes in rodents and humans. The ability to rapidly map biological pathways in rainbow trout would greatly improve our understanding of molecular processes affected by gene expression changes and would further enhance the utility of this already valuable model organism.
Lastly, it is worthwhile to discuss the implications of this work with regards to the possible premature marketing of I3C by some manufacturers as a dietary supplement. The highlighted study and other studies on I3C tumor promotion raise a cautionary note regarding the arbitrary use of this compound as a dietary supplement. Clearly, I3C and DIM have potent biochemical activities and affect a host of cellular processes important in physiological homeostasis. The seemingly dichotomous effects of these phytochemicals involving both potent chemoprotective and also tumor promoting capabilities warrant the continued careful and thoughtful investigation of the mechanisms of action of these agents in different species. Ultimately, the identification of the molecular pathways through which phytochemicals mediate receptor pathways should facilitate the assessment of the risk posed to humans by these compounds. Clearly, the culmination of 40 years of research dedicated to using the rainbow trout model in carcinogenesis research at Oregon State University has enriched our understanding of cancer pathogenesis, contributed to our prediction of risks associated with chemical carcinogenesis, and helped us better understand the role of phytochemicals in preventing and treating disease.
REFERENCES
Ashok, B. T., Chen, Y. G., Liu, X., Garikapaty, V. P., Seplowitz, R., Tschorn, J., Roy, K., Mittelman, A., and Tiwari, R. K. (2002). Multiple molecular targets of indole-3-carbinol, a chemopreventive anti-estrogen in breast cancer. Eur. J. Cancer Prev. 11(Suppl. 2), S86–S93.
Bailey, G., Selivonchick, D., and Hendricks, J. (1987). Initiation, promotion, and inhibition of carcinogenesis in rainbow trout. Environ. Health Perspect. 71, 147–153.
Dashwood, R. H., Fong, A. T., Williams, D. E., Hendricks, J. D., and Bailey, G. S. (1991). Promotion of aflatoxin B1 carcinogenesis by the natural tumor modulator indole-3-carbinol: Influence of dose, duration, and intermittent exposure on indole-3-carbinol promotional potency. Cancer Res. 51, 2362–2365.
Grubbs, C. J., Steele, V. E., Casebolt, T., Juliana, M. M., Eto, I., Whitaker, L. M., Dragnev, K. H., Kelloff, G. J., and Lubet, R. L. (1995). Chemoprevention of chemically-induced mammary carcinogenesis by indole-3-carbinol. Anticancer Res. 15, 709–716.
Kim, D. J., Han, B. S., Ahn, B., Hasegawa, R., Shirai, T., Ito, N., and Tsuda, H. (1997). Enhancement by indole-3-carbinol of liver and thyroid gland neoplastic development in a rat medium-term multiorgan carcinogenesis model. Carcinogenesis 18, 377–381.
Mori, H., Sugie, S., Rahman, W., and Suzui, N. (1999). Chemoprevention of 2-amino-1-methyl-6-phenylimidazo [4,5-b]pyridine-induced mammary carcinogenesis in rats. Cancer Lett. 143, 195–198.
Oganesian, A., Hendricks, J. D., Pereira, C. B., Orner, G. A., Bailey, G. S., and Williams, D. E. (1999). Potency of dietary indole-3-carbinol as a promoter of aflatoxin B1-initiated hepatocarcinogenesis: Results from a 9000 animal tumor study. Carcinogenesis 20, 453–458.
Riby, J. E., Chang, G. H., Firestone, G. L., and Bjeldanes, L. F. (2000). Ligand-independent activation of estrogen receptor function by 3,3'-diindolylmethane in human breast cancer cells. Biochem. Pharmacol. 60, 167–177.
Rise, M. L., von Schalburg, K. R., Brown, G. D., Mawer, M. A., Devlin, R. H., Kuipers, N., Busby, M., Beetz-Sargent, M., Alberto, R., Gibbs, A. R., et al. (2004). Development and application of a salmonid EST database and cDNA microarray: Data mining and interspecific hybridization characteristics. Genome Res. 14, 478–490.
Shertzer, H. G., and Senft, A. P. (2000). The micronutrient indole-3-carbinol: Implications for disease and chemoprevention. Drug Metabol. Drug Interact. 17, 159–188.
Shilling, A. D., Carlson, D. B., Katchamart, S., and Williams, D. E. (2001). 3,3'-Diindolylmethane, a major condensation product of indole-3-carbinol, is a potent estrogen in the rainbow trout. Toxicol. Appl. Pharmacol. 170, 191–200.
Stahlberg, N., Merino, R., Hernandez, L. H., Fernandez-Perez, L., Sandelin, A., Engstrom, P., Tollet-Egnell, P., Lenhard, B., and Flores-Morales, A. (2005). Exploring hepatic hormone actions using a compilation of gene expression profiles. BMC Physiol. 5, 8.
Tilton, S. C., Givan, S. A., Pereira, C. B., Bailey, G. S., and Williams, D. E. (2005a). Toxicogenomic profiling of the hepatic tumor promoters indole-3-carbinol, 17-estradiol and -naphthoflavone in rainbow trout. ToxSci, 10.1093/toxsci/kfi341. Advance Access publication September 28, 2005.
Tilton, S. C., Gerwick, L. G., Hendricks, J. D., Rosato, C. S., Corley-Smith, G., Givan, S. A., Bailey, G. S., Bayne, C. J., and Williams, D. E. (2005b). Use of a rainbow trout oligonucleotide microarray to determine transcriptional patterns in aflatoxin B1-induced hepatocellular carcinoma compared to adjacent liver. Toxicol. Sci. 88, 319–330.
Williams, D. E., Bailey, G. S., Reddy, A., Hendricks, J. D., Oganesian, A., Orner, G. A., Pereira, C. B., and Swenberg, J. A. (2003). The rainbow trout (Oncorhynchus mykiss) tumor model: Recent applications in low-dose exposures to tumor initiators and promoters. Toxicol. Pathol. 31(Suppl.), 58–61.
Yuan, F., Chen, D. Z., Liu, K., Sepkovic, D. W., Bradlow, H. L., and Auborn, K. (1999). Anti-estrogenic activities of indole-3-carbinol in cervical cells: Implication for prevention of cervical cancer. Anticancer Res. 19, 1673–1680.(Evan P. Gallagher)