Meta-analysis of human cancer microarrays reveals GATA3 is integral to the estrogen receptor alpha pathway

Background: The transcription factor GATA3 has recently been shown to be necessary for mammary gland morphogenesis and luminal cell differentiation. There is also an increasing body of data linking GATA3 to the estrogen receptor alpha (ER alpha) pathway. Among these it was shown that GATA3 associates with the promoter of the ER alpha gene and ER alpha can reciprocally associate with the GATA3 gene. GATA3 has also been directly implicated in a differentiated phenotype in mouse models of mammary tumourigenesis. The purpose of our study was to compare coexpressed genes, by meta-analysis, of GATA3 and relate these to a similar analysis for ER alpha to determine the depth of overlap. Results: We have used a newly described method of meta-analysis of multiple cancer studies within the Oncomine database, focusing here predominantly upon breast cancer studies. We demonstrate that ER alpha and GATA3 reciprocally have the highest overlap with one another. Furthermore, we show that when both coexpression meta-analysis lists for ER alpha and GATA3 are compared there is a significant overlap between both and, like ER alpha, GATA3 coexpresses with ER alpha pathway partners such as pS2 (TFF1), TFF3, FOXA1, BCL2, ERBB4, XBP1, NRIP1, IL6ST, keratin 18(KRT18) and cyclin D1 (CCND1). Moreover, as these data are derived from human tumour samples this adds credence to previous cell-culture or murine based studies. Conclusion: GATA3 is hypothesized to be integral to the ER alpha pathway given the following: (1) The large overlap of coexpressed genes as seen by meta-analysis, between GATA3 and ER alpha, (2) The highest coexpressing gene for GATA3 was ER alpha and vice-versa, (3) GATA3, like ER alpha, coexpresses with many well-known ER alpha pathway partners such as pS2.