ToxSci Advance Access originally published online on June 27, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Toxicological Sciences 74, 233-234 (2003)
Copyright © 2003 by the Society of Toxicology
TOXICOLOGICAL HIGHLIGHT |
The Tg.AC Mouse Model Passes Test by Failing to Respond
Cancer Biology Group, National Center for Toxicogenomics, NIEHS, Research Triangle Park, North Carolina 27709
ABSTRACT
This issue highlights two companion articles, "Selection of Drugs to Test the Specificity of the Tg.AC Assay by Screening for Induction of the gadd153 Promoter in Vitro" by Karol L. Thompson and Frank D. Sistare (pp. 260–270) and "Evaluation of the Tg.AC Assay: Specificity Testing with Three Noncarcinogenic Pharmaceuticals that Induce Selected Stress Gene Promoters in Vitro and the Inhibitory Effects of Solvent Components" by Karol L. Thompson, Barry A. Rosenzweig, James L. Weaver, Jun Zhang, Karl K. Lin, and Frank D. Sistare (pp. 271–278). These articles outline a unique paradigm for selecting and assaying chemicals for carcinogenic activity using the short-term in vivo Tg.AC mouse model. This Tg.AC mouse model was the primary focus of the work; however, the strategy employed could be exploited for the other alternative mouse models.
Short-term in vivo assays have gained prominence in the past 5 years as adjuncts and potential replacements for the mouse 2-year bioassay (Tennant et al., 1999
). As alternatives to the 2-year bioassay, the transgenic mouse models have demonstrated several advantages. For example, the assay length, 6 months, is significantly shorter than the 2 years required for rodent bioassays. Transgenic assays require the use of fewer animals than do bioassays. Additionally, because the transgenic assays are genetically modified in specific genes and pathways, mechanistic understanding of the tumorigenic response can be obtained. One disadvantage of the assays, however, is the sensitivity and specificity of the individual transgenic mouse models. The International Life Sciences Institutes (ILSI) have performed an in-depth evaluation of the alternative models for carcinogenicity testing (Popp, 2001; Robinson and MacDonald, 2001); Thompson et al.’s work complements those ILSI consortial efforts to evaluate the specificity of the alternative in vivo mouse models.
The intent of Thompson et al. in their work is to provide a greater understanding of the predictive value of the Tg.AC short-term in vivo assay for carcinogenicity testing. The Tg.AC model is a transgenic mouse model with the unique ability to mount a tumorigenic response within 6 months in skin paint assays when dosed topically with nongenotoxic carcinogens. Incomplete mechanistic understanding of the transcriptional activation of the zetaglobin-promoted v-Ha ras transgene of Tg.AC has led to concerns about the specificity of its tumorigenic response, including concerns that (in the Tg.AC assay) potential false positive responses could result from nongenotoxic drugs signaling through normal proliferative pathways via receptors to activate the oncogenic v-Ha-ras transgene. To evaluate this response, Thompson and Sistare used a battery of in vitro reporter gene assays, previously shown to have the highest correlative value to tumor induction in Tg.AC assay, to select the appropriate nongenotoxic carcinogen to further test in Tg.AC (Thompson et al., 2000). The reporter gene assays consisted of three promoters driving the green fluorescent protein (GFP), chloramphenicol acetyltransferase (CAT), or luciferase coding regions. The promoters employed were derived from the gadd153 gene, the fetal zetaglobin gene, and the human fos proto-oncogene. Stable lines derived from transfection were used to provide a more homogeneous response and allow for high-throughput screening.
To challenge the specificity of the Tg.AC assay, a list of candidate compounds was compiled from an FDA in-house database, which contained rat and mouse 2-year bioassay results for each compound. Ninety-nine drugs that tested negative in the 2-year bioassays for rats and mice were chosen, with an added priority to drugs with pharmacological actions different from those chosen by the ILSI working group in its evaluative study of Alternative Methods for Carcinogenicity testing. Of the chosen set of 99 drugs, 18 were antihypertensive-related, 12 were nonsteroidal anti-inflammatory (NSAIDs), and 11 were allergic and pulmonary disorder-related. Psychoactive, antiviral, antiepileptic, and diabetic drugs constituted the remainder of the set.
Dose ranges for the compounds were determined, and each was assayed in the battery of reporter gene assays. Not surprisingly, the in vitro assays were not predictive of carcinogenic activity. Of the drugs tested, 40% were positive in two of the three reporter assays, while 20% of the drugs tested positive in the gadd153 promoter induction assay. Selection of the candidates for in vivo testing was primarily based on the most robust response in the gadd153 promoter assay. The gadd153 promoter response demonstrated a large dynamic range, with 100-fold induction ratios with some chemicals, and displayed a discriminative response to DNA damage as well as cellular stress and toxicity (Luethy and Holbrook, 1992). Ultimately, Thompson and Sistare selected three compounds from the list of 99. Amiloride, dipyridamole, and pyrimethamine were selected as the candidate compounds to subsequently test in the Tg.AC in vivo assay.
Scientists involved in toxicological research will appreciate the other two practical considerations discussed in the chemical selection process. Tg.AC uses the skin papillomas as the "reporter phenotype": a positive response is indicated when papillomas develop on the dorsum following topical treatment with a compound, therefore making topical applicability a prerequisite for any chemical selected for the Tg.AC assay. Thompson et al. investigated the effects of dimethyl sulfoxide (DMSO) and ethanol as solvents in Tg.AC skin paint studies. Considerations for an appropriate solvent include wetting, surface tension, and solubility of the compound to allow the delivery of the appropriate dose.
The final aspects of the study, found in the Thompson et al. articles, were the results of the 26-week skin paint study conducted in hemizygous Tg.AC mice with three carefully selected drugs at two doses. The data from the Thompson et al. study indicated that no skin papillomas were observed in mice treated for 26 weeks with amiloride, dipyridamole, or pyrimethamine. The researchers conclude that the Tg.AC model passed the test, giving the correct negative response for all three nongenotoxic carcinogens selected.
In summary, these two articles underscore the potential uses of transgenic models in chemical carcinogenesis testing, thus strengthening the utility of the Tg.AC mouse.
NOTES
1 For correspondence via e-mail: cannon1{at}niehs.nih.gov. 
REFERENCES
Luethy, J. D., and Holbrook, N. J. (1992). Activation of the Gadd153 promoter by genotoxic agents: A rapid and specific response to DNA damage. Cancer Res. 52, 5–10.
Popp, J. A. (2001). Criteria for the evaluation of studies in transgenic models. Toxicol. Pathol. 29(Suppl.), 20–23.
Robinson, D. E., and MacDonald, J. (2001). Background and framework for ILSI’s collaborative evaluation program on alternative models for carcinogenicity assessment. Toxicol. Pathol. 29(Suppl.), 13–19.
Tennant, R. W., Stasiewicz, S., Mennear, J. H., French, J. E., and Spalding, J. W. (1999). Genetically altered mouse models for identifying carcinogens. IARC Sci. Publ. 146, 123–150.
Thompson, K. L. Rosenzweig, B. A., and Sistare, F. D. (2000). Evaluation of in vivo reporter gene induction assays for use in a rapid prescreen for compound selection to test specificity in the Tg. AC mouse short-term carcinogenicity assay. Toxicol. Sci. 57, 43–53.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||