Final Report Summary - ACTINONSRF (MAL: an actin-regulated SRF transcriptional coactivator) ACTINonSRF focussed predominantly on the MRTFs, transcriptional coactivators of the SRF transcription factor. We had previously shown that G-actin controls MRTF activity by binding to a regulatory domain containing multiple G-actin-binding RPEL motifs. ACTINonSRF aimed to elucidate the role of G-actin in control of MRTF activity and target gene expression, and to assess whether other RPEL proteins are controlled by G-actin. ACTINonSRF demonstrated a general principle for G-actin regulation of protein function: competitive binding with other effector proteins, including importinαβ, PP1 catalytic subunits, and Rho family GTPases. G-actin also controls other biochemical functions including MRTF-SRF interaction, RNA polII phosphorylation, and membrane targeting, but here the mechanisms remain unknown. We determined structures of trivalent and pentavalent G-actin/MRTF-A complexes (Mouilleron et al, 2011), showing that the pentavalent complex effectively prevents access of importinαβ to an NLS within the RPEL domain. We showed that G-actin functionally cooperates with at least six Crm1-dependent NES located throughout MRTF-A to promote its nuclear export (Panayiotou et al, 2016). Phosphorylation acts both positively and negatively to control MRTF-A activity (Panayiotou et al, 2016) ERK-mediated RPEL domain phosphorylation activates MRTF-A, while phosphorylation of the N-terminal NES inhibits MRTF-A by promoting nuclear export. We mapped over 30 phosphorylation sites, of which 25 are targets for proline-directed kinases.ACTINonSRF conducted the first ChIP-seq analysis of MRTF genomic targets (Esnault et al 2014) revealing hundreds of new MRTF-SRF target genes, playing a major role in cell adhesion / contractility and transcriptional regulation. SRF is responsible for the overwhelming majority of MRTF genomic targeting. This analysis also demonstrated novel roles for MRTF-SRF signalling in circadian and Myc-dependent gene expression, which we characterised in collaborations (Gerber et al, 2013; Wiese et al, 2015). We found that the TCF family of SRF cofactors act as general inhibitors of MRTF-SRF signalling, by competing with the MRTFs for SRF, thereby controlling fibroblast pro-invasive and contractile behaviour (Gualdrini et al, 2016). Much of the bioinformatic analysis in this study was developed in parallel with our analysis of chromatin events in TCF-SRF gene expression (Esnault et al, 2017).Elevated YAP-TEAD activity has previously been shown to potentiate contractile and pro-invasive behaviour of cancer-associated fibroblasts (CAFs). We showed that . MRTF-SRF activity is also elevated in CAFs, and acts in a similar manner. Interestingly, MRTF-SRF and YAP-TEAD signalling exhibit mutual dependence as a result of regulating expression of each other's signalling pathways (Foster et al, 2018, in press). We also provided evidence that MRTF-SRF signalling to cytoskeletal genes is required for colonisation of the bone marrow by hematopoietic stem cells (Costello et al 2015). On the mechanistic level, G-actin also plays a novel role in the regulation of MRTF transcriptional activation. It controls MRTF interaction with SRF, and under resting conditions, disrupts phosphorylation of the RNA PolII CTD at MRTF-SRF target genes.Ongoing studies aim to elucidate the biochemistry of these phenomena.ACTINonSRF also characterised other RPEL-containing proteins. We demonstrated that G-actin controls nuclear shuttling of Phactr1, a member of the Phactr family of PP1-binding proteins, and formation of the Phactr1-PP1 complex, which controls cytoskeletal dynamics (Wiezlak et al, 2012). Study of Phactr1/G-actin complexes revealed how individual G-actin/RPEL units interact to assemble oligomeric complexes (Mouilleron et al, 2012). In ongoing work we showed that the Phactr1-PP1 complexes is holoenzyme, and have identified candidate G-actin responsive substrates, setting the stage for a complete analysis of these proteins.Finally, our improved understanding of the G-actin/RPEL interaction allowed identification of two new RPEL protein families (ArhGAP12, ArhGAP32 proteins). Structural and biochemical studies show that G-actin controls their GAP activity by inhibiting their binding to Rho-family GTPases (in preparation).
