Analysis of SAFB1 and SAFB2 Knockout Mice Reveals Non-Redundant Functions of the Two Estrogen Receptor Co-Repressor Paralogs.

Abstract

Abstract The Scaffold Attachment Factor B1 (SAFB1) and B2 (SAFB2) have been shown to be involved in chromatin organization, transcriptional regulation, and RNA processing. The paralogs SAFB1 and SAFB2 share 74% similarity at the amino acid level, with up to 98% similarity in some functional domains. These functional domains include DNA and RNA-binding domains, and a C-terminal repression domain. We have previously shown that SAFB1 and SAFB2 function as estrogen receptor (ER) corepressors – they directly bind to ER, and repress its transcriptional activity. There is also evidence that both proteins play a role in breast cancer; their shared chromosomal locus on chromosome 19p13 displays high rates of LOH, and protein loss was associated with worse survival of breast cancer patients.To understand the roles of SAFB1 and SAFB2 in development and function of hormone responsive tissues, we generated gene-specific knockout (KO) mouse models. Deletion of SAFB1 resulted in a high degree of embryonic and perinatal lethality. Surviving SAFB1 KO mice displayed severe growth retardation associated with low serum IGF-I levels, male infertility and female subfertility. In contrast, SAFB2 KO mice were born at the expected Mendelian ratio and did not show any obvious defects in growth or fertility. Initial gross pathology analysis has not yet revealed any significant defects.Since the SAFB proteins are known to play a role in estrogen response, we analyzed mammary gland development in the two mouse models. Young virgin SAFB1 KO mice showed delayed mammary gland development with a significantly reduced number of terminal end buds and decreased outgrowth compared to wild type (WT) controls. This was likely a result of delayed puberty, due to low IGF-I levels, since at the four month time point, mammary gland growth was restored in the virgin SAFB1 KO mice to that seen in WT controls. Interestingly, the KO glands exhibited increased alveolar development and side branching. To measure proliferation directly, we performed mammary gland transplantation experiments, thereby excluding secondary effects due to systemic defects. These studies showed a significant increase in proliferation in SAFB1 KO glands compared to WT control glands. In contrast to the SAFB1-null mice, we did not observe any obvious mammary gland defects in pubertal or adult SAFB2-null mice. However, mammary glands from aged virgin SAFB2 mice (1.5 years old) showed extensive side branching and precocious development of alveolar buds resembling glands of late pregnant mice. This was associated with increased proliferation of alveolar cells in the SAFB2 KO glands.In summary, genetic ablation of the ER co-repressors SAFB1 and SAFB2 results in defects in mammary gland development. In general, the two mouse models have very different phenotypes, revealing diverse and non-redundant functions of SAFB1 and SAFB2, findings that were unexpected based on their high sequence similarity. We are currently performing additional studies to finalize characterization of the in vivo phenotypes, focusing on hormone responsive tissues, and also to understand the mechanism underlying the observed phenotypes. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 2162.