ors have not been characterized, although it is well documented that transcription by these highly related receptors is disrupted by exposure to iAs in the low micromolar range. Because iAs can bind to vicinal thiols which are found in the steroid binding domain and the DNA binding domain of steroid hormone receptors, two possible hypotheses are that iAs might inhibit transcription by interfering with either steroid binding to receptor or receptor binding to DNA. In fact Simons et al showed that in an in vitro system 7 mM iAs was effective in inhibiting Dex binding to GR. However, this concentration of iAs is not an achievable intracellular concentration because it would be toxic and result in cell death. In our analyses by order RO4929097 ICP-MS of intracellular As+3 
we found that treatment of cells with 8 mM iAs for 15 minutes to 1 hour, resulted in only 0.1 to 1.5 mM intranuclear As+3 with a correlation between the length of time of treatment and the amount of As+3 detected. Note that these are the concentrations used in the in vitro system to determine whether iAs is acting directly or indirectly on proteins at the promoter. Treatment of 1470.2 cells with higher iAs concentrations than 8 mM induces a stress response and results in cell death. Others have shown that GR does in 21150909 fact translocate from the cytoplasm to the nucleus in the presence of iAs at similar concentrations to those used in this study. This transition is ligand dependent and it is shown here that GR binds to a GRE equally well in cells that have been treated with iAs at levels that inhibit transcription. Thus it appears that netiher iAs-mediated inhibition of steroid hormone binding to GR or GR binding to the promoter are likely mechanisms underlying inhibition of transcription by iAs. Because both CARM1 and one of the p160 coactivators are essential to transcriptional regulation at all steroid hormone receptorregulated promoters, identification of CARM1 and GRIP1 as players in iAs-mediated transcriptional inhibition raises the possibility that iAs may repress transcription by other steroid hormone receptors similarly. In support of this, we have found that iAs-mediated inhibition of an estrogen-responsive promoter in MCF7 breast cancer cells is functionally related to iAs effects on CARM1 and on SRC3/AIB1 a GRIP1 homologue that interacts with ERs. While this manuscript was in preparation it was reported that TIF2, the human homologue of GRIP1/SRC2, does not bind to an AR-regulated promoter in LNCaP cells 24 hours after treatment with arsenic trioxide and an androgen. It was also shown that AR was no longer at the promoter 24 hours after ATO treatment however, and p160 coactivators interact with hormone-responsive promoters via interaction with steroid receptors so TIF2 would not be expected to be there. No direct evidence to support iAsmediated disruption of the AR/TIF2 interaction is shown. Our data shows that iAs inhibits CARM1 but not GRIP1/SRC2 interaction with the GR-activated MMTV promoter as early as 30 minutes following treatment with Dex and iAs in the form of iAs-mediated Inhibition of Histone Modification The apparent level of H3K18ac and H3R17me in response to iAs in these experiments was below basal levels. It is possible that 19615387 a demethylase or a histone deacetylase is inappropriately associated with the promoter. Notably, nickel and chromium, two genotoxic metals, repress transcription rapidly and do so partially through mechanisms that affect histone modification. Ho
