Deacetylase inhibitors repress STAT5-mediated transcription by interfering with bromodomain and extra-terminal (BET) protein function.
Bottom Line: We showed previously that the deacetylase inhibitor trichostatin A (TSA) inhibits STAT5-mediated transcription by preventing recruitment of the transcriptional machinery at a step following STAT5 binding to DNA.The mechanism and factors involved in this inhibition remain unknown.Instead, they induce a rapid increase in global histone acetylation apparently resulting in the delocalization of the bromodomain and extra-terminal (BET) protein Brd2 and of the Brd2-associated factor TBP to hyperacetylated chromatin.
Affiliation: Stat5 Signaling Research Group, Institute of Immunology, University of Regensburg, 93053 Regensburg, Germany.Show MeSH
Related in: MedlinePlus
Mentions: Lysine to glutamine (K>Q) and lysine to arginine (K>R) substitutions that mimic the acetylated or unmodified lysine respectively (70) were then introduced into STAT5A-1*6. Lysine residues that were proposed to be critical for the regulation of STAT5 but also of STAT3 at equivalent positions (8–10,53–56,60)—namely K84, K359, K384, K675, K681, K689, K696 and K700—were mutated either individually or in combination (Figure 2). Transcriptional activity of the STAT5A-1*6 lysine mutants was assessed upon transfection in Ba/F3 cells maintained in IL-3-free medium, hence in conditions where endogenous wild-type STAT5 is inactive. Expression (protein level) and activation (phosphorylation) of the respective STAT5A-1*6 mutants were verified by western blot, and their transcriptional activity was evaluated by measuring expression of the STAT5 target gene Cis by quantitative RT-PCR. Neither single nor multiple lysine mutations impaired STAT5A-1*6 transcriptional activity (Figure 3A and B). Of note, STAT5A-1*6 proteins mutated at lysine 675 (K675Q, K675R as well as 5xQ and 5xR) were poorly detectable in western blot, likely due to protein instability, and thus their activity could not be readily assessed. Nevertheless, these results indicate that acetylation at the investigated lysines, notably at K359, K689 or K696 previously reported to be acetylated and important for STAT5 activity (8,10,56), is not required for STAT5A-1*6 activity in Ba/F3 cells. These experiments therefore indicate that the impaired STAT5 transcriptional activity induced by deacetylase inhibitors is unlikely to be due to a change in STAT5 acetylation, at least at the investigated lysines. This proposition is further supported by the important finding that the transcriptional activity of STAT5A-1*6 3xQ/R proteins, mutated simultaneously at K689, K696 and K700, remained sensitive to TSA (Figure 3C). Of note, the observation that the STAT5A-1*6 K696Q/R mutants are as active as the non-mutated STAT5A-1*6 protein indicates that not only acetylation but also SUMOylation at K696 is not required for STAT5A-1*6 transcriptional activity in Ba/F3 cells.
Affiliation: Stat5 Signaling Research Group, Institute of Immunology, University of Regensburg, 93053 Regensburg, Germany.