Identification and Characterization of ABCG2 Enhancer Regions

The mitoxantrone resistance protein (MXR; BCRP; ABCG2), referred to here as MXR, is an efflux membrane transporter expressed apically in several tissues with a broad range of both exogenous and endogenous substrates. The transport activity, tissue distribution and cellular localization of MXR suggest that it plays a pivotal role in endogenous substrate disposition as well as the protection and detoxification of the body from xenobiotics. Overexpression of MXR is associated with drug resistance to a variety of anticancer drugs, and it has been linked with decreased disease-free survival in several cancers. An individual’s susceptibility to certain drug-induced side effects has also been linked to MXR expression or nonsynonymous single nucleotide polymorphisms (SNPs) in the MXR gene ABCG2. Therefore, it is important to not only characterize the functional effects of ABCG2 SNPs, but to also look at other mechanisms which alter MXR expression and function. We hypothesize that there are regulatory regions surrounding ABCG2, and our aim for this study was to identify and characterize regions surrounding ABCG2 that increase gene expression. We computationally analyzed a ~300 kb stretch of DNA (chr4:89130400-89439035, hg18), which includes the ABCG2 gene, in order to identify putative ABCG2 enhancers and then tested these regions for in vitro and in vivo enhancer activity. Genomic regions with potential enhancer capabilities were identified as having high evolutionary conservation determined by comparative genomics (ECR and Vista browsers) and/or the presence of specialized transcription factor binding sites (TFBS) identified with computational methods (rVista, Transfac, Cister). These regions were cloned from human DNA and inserted into the enhancer assay vector pGL4.23. The constructs were transiently transfected into liver (HepG2), kidney (HEK293T), intestine (HCT116) and breast (MCF-7) cell lines, and the luciferase gene activity was measured as a surrogate for the enhancer capabilities of the genomic region.  We tested 29 regions and identified seven separate regions (250-1000 bp in length) surrounding ABCG2 exhibiting significant enhancer activity in vitro. Global and cell-specific enhancers were identified, with at least three enhancer regions in each of the cell lines. HEK293T, MCF-7 and HepG2 cell lines each had two cell-specific enhancer regions.  An additional region showed enhancer activity in both the HepG2 and HCT116 cell lines; two global enhancer regions showed activity in all four cell lines. Studies are ongoing utilizing the mouse tail vein injection assay to test these regions for in vivo enhancer activity in the liver. Our study provides evidence that there are genomic regions surrounding ABCG2 capable of increasing gene expression, possibly in a tissue-specific manner. The results from this research will inform future clinical investigations examining how interindividual variation in MXR expression and function contributes to differences in drug response. This work was supported by NIH grant GM61390 and a predoctoral fellowship from the American Foundation for Pharmaceutical Education.