By performing primer-specific RT-PCR analyses, three laboratories including ours have found that exons I.3 and PII are the two major exon Is present in aromatase mRNAs isolated from breast tumors. These results suggest that promoters I.3 and 11 are the major promoters directing aromatase expression in breast tumors. The characterization of transcription factors that interact with the two elements near promoters I.3 and II, i.e., S1 and CREaro, helps us better understand the mechanism of the switch of promoter usage between normal breast tissue and cancer tissue. The positions of the two regulatory regions were mapped by DNase I footprinting and DNA deletion analyses. We applied the yeast one-hybrid approach to screen a human breast tissue hybrid cDNA expression library for genes encoding the proteins binding to these regions. Our results suggest that in normal breast tissue, the function of promoters I.3 and II is suppressed through the binding of EAR-2, COUP-TFI, and RAR gamma to S1, and through the binding of Snail/Slug proteins to their binding site that quenches the CREaro activity. In cancer tissue, the expression levels of EAR-2, COUP-TFI, EAR gamma, Snail, and Slug decrease, and aromatase expression is then up-regulated through the binding of ERR alpha to S1 and the binding of CREB I or related factors to CREaro. In a separate study, we found that estrogen could up-regulate aromatase expression in breast cancer cells by a non-genomic action of ER alpha via cross-talk with growth factor-mediated pathways. Our preliminary results suggest that protein kinase C delta participates in this ER alpha-growth factor mediated regulation. To further understand the regulatory mechanism, we have recently initiated an in vivo footprinting analysis of the -260/+76 bp region of promoter I.3. The experiments were conducted with both MCF-7 and MDA-MB-231 breast cancer cells. Our results revealed several footprinted sites. Five regions (sites 1-5) were then selected for functional analysis through DNA site-directed mutagenesis experiments. This analysis has also confirmed the promoter I.3 TATA site and Snail/Slug binding site. These mutants showed higher luciferase activity when compared to the wild-type, indicating that the proteins binding to these sites were acting as repressors. By reviewing findings from our laboratory and other laboratories, a detailed mechanism for the transcriptional regulation of aromatase expression in breast cancer tissue is summarized and discussed. (c) 2005 Elsevier Ltd. All rights reserved.