Repressive links

The regulation of gene transcription relies on the concerted efforts of several proteins, including those that modify the structure of the chromatin. Jan-Hermen Dannenberg, Gregory David, Ron DePinho and colleagues looked at the transcriptional networks regulated by the co-repressor mSIN3A in both normal and neoplastic cells using combined genetic, biochemical and computational approaches.

mSIN3A interacts with histone deacetylases (HDACs) and numerous transcription factors to regulate diverse signalling pathways and biological processes. To further investigate the function of this protein, Dannenberg and co-authors engineered mice that harboured a conditional mSin3A-knockout allele with embedded Lox-P sites (mSin3A^L) that delete the gene in vivo on crossing to mice expressing Cre recombinase, or in vitro on exposure to Cre-encoding retroviruses.

mSin3A-null mice are embryonic lethal, indicating a crucial function for mSIN3A in normal development. As early lethality hampered in-depth dissection of mSIN3A function, additional studies were conducted in mSin3A^L/L mouse embryonic fibroblasts (MEFs) that were depleted of mSIN3A following exposure to the Cre-encoding retrovirus. Loss of mSIN3A initiates unscheduled DNA synthesis, triggering an S-phase checkpoint that results in profound growth arrest at G2/M and increased apoptosis — data that might explain the lethal phenotype observed in the null embryos.

As mSIN3A–HDAC has been implicated in regulatory modifications of the p53 tumour-suppressor protein, the investigators assessed the impact of combined deletion of mSIN3A and p53 on these cellular phenotypes. Working with MEFs as well as lymphomas and sarcomas derived from Trp53 ^−/− mice carrying conditional mSin3A alleles, loss of p53 was shown to not be enough to overcome mSIN3A-mediated growth arrest or apoptosis. The authors also showed that loss of both p53 and http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=gene&cmd=Retrieve&dopt=Graphics&list_uids=19645 tumour-suppressors failed to alter the lethal outcome.

So what genes trigger the growth arrest in mSIN3A-deficient cells? To address this question, Dannenberg and colleagues conducted a time course to map the changes in gene expression after mSIN3A deletion. They showed that genes involved in numerous crucial processes were altered, including DNA replication, cell-cycle regulation, apoptosis and mitochondrial metabolism. Computational analysis of the mSIN3A transcriptome, using a knowledge-based database, identified several well-known genes as nodal points, including the Myc –Mad, E2f and Trp53 transcriptional networks.

Loss of mSIN3A expression also changed the expression of many components of the mSIN3A–HDAC complex, consistent with the ability of mSIN3A to regulate its own transcription and that of other members of this complex. In addition, mSIN3A showed novel links to DNA-repair and peroxisome-proliferator-activated-receptor networks. These findings were verified using in silico promoter analyses and complemented by chromatin-immunoprecipitation assays that also revealed links to the signal transducer and activator of transcription (STAT) network, and the nucleosome remodelling factor FALZ. However, the authors additionally note that not all MYC, E2F and p53 target genes are upregulated on loss of mSIN3A, indicating that mSIN3A might only regulate a subset of these targets and/or that mSIN3A regulates these genes depending on the physiological conditions or the tissue type.

This integrated approach has produced novel insights into the diverse functions of mSIN3A and has confirmed the importance of mSIN3A in cancer-related pathways and processes. No doubt this approach will be useful to address the function of other components of multi-protein transcriptional complexes.