Fig 2: MEF2C binds to promoters and enhancers, which is reduced by Notch signaling.(A) Color-coded centered heatmaps indicating ChIP-Seq peak binding intensities (number of reads in a –1.0 kb to +1.0 kb region relative to the summit of the peak) for MEF2C, Bromodomain 4 (BRD4), and histone H3 with epigenetic trimethylation (H3K4me3) or acetylation marks (H3K27ac) for LOUCY parental cells (blue) or LOUCY parental cells cultured on OP9-DL1 stromal cells for 24 hours (green). Centered heatmaps for 5 kb regions upstream of transcription start site (uTSS), GeneHancer-defined enhancers (Enh), 5 kb regions upstream of long noncoding areas (defined by deepbase and lncipedia [uLnc]), or other regions (Supplemental Figure 3). (B) The number of upregulated genes (from collapsed probe sets; red circle) and downregulated genes (blue circle) with a log2 fold change > 0.6 (with P < 0.05 and q < 0.1) 24 hours after induction of MEF2C expression are displayed (see also Figure 2, C and D). The smaller black circles indicate the number of genes having MEF2C binding sites within the gene body or within 10 kb flanking regions. (C) Motifs as identified by MEME-ChIP found within 50 bp upstream or downstream regions relative to the MEF2C ChIP-Seq peak summits (101 bp window) are indicated including their expectancy values (E value), the number of sites identified, and their relative locations in these windows.

Fig 3: MEF2C blocks T cell development.(A) MEF2C-eGFP transgenic mice: loxP-flanked stop cassette followed by MEF2C cDNA, IRES, and eGFP following exon 1 of the Rosa26 locus (top). Cre recombinase activity removes the stop cassette and activates expression of MEF2C and eGFP from the Rosa26 promoter (bottom). SA, splice acceptor site; arrows, genotyping primers. (B) Genotype of spleens of mice with WT or MEF2C-eGFP transgene (tg) alleles, in the absence or the presence of Lck-Cre. (C) Western blot of endogenous or induced MEF2C and a-tubulin protein levels in total thymocytes from mice of different genotypes. (D) Flow cytometry histograms of eGFP expression in lymph node or thymus cells from duplicate MEF2C-eGFP hetero- or homozygous mice with or without Lck-Cre as indicated by the gray, light green, and dark green histograms. (E) Absolute cell numbers (mean ± SD) for cells from MEF2C-eGFP control (no Cre, open circles) and MEF2C-eGFP/Lck-Cre mice (blue filled circles) at 10–16 and 20–34 weeks of age. Populations are CD4/CD8 DN, DP, CD4 SP, and CD8 SP. DN are also gated on lineage– (CD3, B220, CD11b, Ly6G/C, Ter-119) cells to further divide the immature thymocyte populations into DN1-DN4. DN1, CD44+CD25–; DN2, CD44+CD25+; DN3, CD44–CD25+; DN4, CD44–CD25–. (F) Total CD45+ cell numbers (mean ± SD) as in E in the thymus (left panel) or inguinal lymph nodes (right panel). (G) Total CD19+ cell numbers (mean ± SD) as in E in the BM (left panel) or thymus (right panel). *P < 0.05, **P < 0.01 by Mann-Whitney U test.

Fig 4: MEF2C blocks T cell lineage differentiation in murine and human cocultures.(A) Mouse differentiation assay: lineage-depleted (CD3, B220, CD11b, Ly6G/C, 7-4, Ter-119) BM cells from MEF2C-eGFPtg/tg mice were exposed to Tat-Cre before a coculture experiment (day 0) on OP9-DL1 or OP9 stromal cells to monitor T and B cell differentiation, respectively. (B) Human differentiation assay: human CD34+ cells from healthy umbilical cord blood donors transduced with BFP control (left) or MEF2C-t2a-BFP (right) lentiviruses and cocultured on OP9-DL1 stromal cells for 2 or 5 days. (C) Transduction as in B, but at day 12 after splitting the coculture. CD45+ cell numbers of a representative example of 3 independent coculture experiments are displayed. For transduced cells, the BFP+ percentage is indicated.

Fig 5: SIK inhibitors block MEF2C function.(A) Effects of the compounds from the Selleck Epigenetic Compound Library and various additional inhibitors (1 µM) were screened for their potential to induce a CD3e+ phenotype in LOUCY cells on DLL4-coated plates. As controls, media, non-DLL4–stimulated cells or MEF2C-overexpressing (LOUCY_MEF2C-BFP) cells (+dox) have been indicated in blue. Normalized CD3e fluorescence intensities have been plotted for duplicate experiments. (B) SIK kinase inhibitors YKL-06-062 and HG-9-91-01; dasatinib, which has SIK off-target activity (12 nM, 48 nM, and 180 nM for SIK1, SIK2, or SIK3, respectively) (MedChemExpress or Proteomicsdb.org), and imatinib were tested to induce a CD3e+ phenotype in LOUCY_MEF2C-BFP cells when cultured on DLL4-coated plates in the absence of dox (except when indicated). (C) Western blot of MEF2C, P-MEF2C (Ser222), and BCL2 in noninduced (no dox) or MEF2C-induced (+dox) LOUCY_MEF2C-BFP cells that were incubated in the presence of increasing concentrations of inhibitors HG-9-91-01 for 4 days. Band intensity ratios for various proteins relatively to ß-actin control levels is indicated, with the –dox condition set at the value = 1. (D) Cell viability of noninduced (no dox) or MEF2C-induced (+dox) LOUCY_MEF2C-BFP after a 4-day exposure to a serial dilution of prednisolone combined with a serial dilution of HG-9-91-01 at the concentrations indicated. ZIP synergy scores are calculated from an n = 3 per condition. ZIP scores < –10 are indicated with a dash-dot line and a blue color. ZIP scores > 10 are indicated by a dashed line.