Histone lysine methyltransferases MLL3 and MLL4 direct gene expression to produce platelets efficiently

Abstract

Circulating blood platelets are responsible for maintaining hemostasis. They are released into blood vessels from mature megakaryocytes. Although several transcription factors have been reported to orchestrate the transcriptional programs required for platelet production, how chromatin regulators control these processes is still poorly understood. MLL3 and MLL4 are the main lysine methyltransferases responsible for the deposition of H3K4me1 histone marks at enhancers. MLL3 and MLL4 typically form complexes with other co-factors, such as PTIP. Recently, we showed that loss of PTIP leads to decreased platelet numbers in mice. Here, we find that, although MLL3/4 double deficiency does not alter megakaryopoiesis and endomitosis, the final step of megakaryocyte maturation is affected due to an abnormal cytoskeleton and demarcation membrane system. MLL3/4 double-deficient mice develop macrothrombocytopenia; platelets are preactive and pro-apoptotic, leading to their rapid clearance from the circulation. Increased megakaryopoeisis in the bone marrow and spleen cannot compensate for these abnormalities. Mechanistically, the expression of genes responsible for normal megakaryocyte function and platelet production is altered in MLL3/4-deficient megakaryocytes, partly due to impaired enhancer functions associated with these genes. Our findings provide insights into the epigenetic programs that are important for platelet biogenesis.

Fig. 1. Determination of platelet counts in PTIP- and MLL3/4 SET domain-deficient mice.

Platelet counts (PLT) and mean platelet volume (MPV) in the peripheral blood of Ptip ^f/f;Pf4Cre mice, PLT p = <0.0001 and MPV p = <0.0001 (n = 10) (a), Mll3Set ^f/f;Pf4Cre mice (n = 8) (b), Mll4Set ^f/f;Pf4Cre mice (n = 6) (c) and Mll3Set ^f/f;Mll4Set ^f/f;Pf4Cre mice, PLT p = < 0.0001 and MPV p = <0.0021 (n = 8) (d). Values are mean ± SD. P values were determined by student t test (unpaired, two-sided), ns=not significant. e Transmission electron microscopy performed on washed platelets from Wt;Pf4Cre, Ptip ^f/f;Pf4Cre and DKO;Pf4Cre mice. A α-granule, D dense granule, OCS open canalicular system. Source data are provided as a Source Data file.

Fig. 2. Platelet recovery and lifespan in the peripheral blood of Ptip^f/f;Pf4Cre and DKO;Pf4Cre mice.

a Design for platelet depletion and regeneration assay. b, c Platelet regeneration in the peripheral blood of Ptip ^f/f;Pf4Cre mice and DKO;Pf4Cre mice (n = 3). d Design for platelet lifespan assay in vivo. e, f Determination of platelet lifespan in Ptip ^f/f;Pf4Cre mice and DKO;Pf4Cre mice (n = 3). g Analysis of platelet apoptosis in Ptip ^f/f;Pf4Cre mice and DKO;Pf4Cre mice, p = 0.0240 (n = 3). h Analysis of platelet pre-activation in Ptip ^f/f;Pf4Cre mice and DKO;Pf4Cre mice, p = 0.0161 (n = 3). i Binding of Jon/A in epinephrine stimulated platelets, p = 0.0150 (n = 3). Values are mean ± SD. P values were determined by Student’s t-test (unpaired, two-sided), ns not significant, * for p < 0.05, ** for p < 0.01, *** for p < 0.001). Source data are provided as a Source Data file.

Fig. 3. Megakaryocyte development profiling in Ptip^f/f;Pf4Cre and DKO;Pf4Cre mice.

a Total number of MKs, immature MKs, and mature MKs in femur bone marrow and spleen of Ptip^f/f;Pf4Cre mice, determined by flow cytometry (Supplementary fig. 8a) (n = 5). b Total number of MKs, immature MKs, and mature MKs in femur bone marrow and spleen of DKO;Pf4Cre mice, determined by flow cytometry (n = 5). Values are mean ± SD. P values were determined by Student’s t-test (unpaired, two-sided), ns not significant. c Representative H&E staining for bone marrow and spleens from Wt;Pf4Cre, Ptip^f/f;Pf4Cre and DKO;Pf4Cre mice (megakaryocytes are highlighted by dashed circles). Source data are provided as a Source Data file.

Fig. 4. Cytoskeleton structure of megakaryocytes derived from fetal livers of Ptip^f/f;Pf4Cre and DKO;Pf4Cre mice.

a Representative histogram (Supplementary fig. 8b) showing the ploidy of megakaryocytes in the bone marrow and spleens from Wt;Pf4Cre, Ptip^f/f;Pf4Cre and DKO;Pf4Cre mice. b Wt;Pf4Cre, Ptip^f/f;Pf4Cre and DKO;Pf4Cre fetal liver cells were cultured in the presence of TPO for 4 days and megakaryocytes were obtained by sedimentation (n = 3). Values are mean ± SD. P values were determined by Student’s t-test (unpaired, two-sided), ns=not significant. Round megakaryocytes were cultured for 1 more day and the megakaryocytes, proplatelets, and platelets were separated by sedimentation and centrifugation as described in Supplementary fig. 4c. Values are mean ± SD. P values were determined by Student’s t-test (unpaired, two-sided), ns not significant. On day 5, numbers of megakaryocytes, proplatelets and platelets normalized per 100,000 round megakaryocytes on day 4 were calculated. c Wt;Pf4Cre, Ptip^f/f;Pf4Cre and DKO;Pf4Cre megakaryocytes were stained for immunofluorescence confocal microscopy with antiαtubulin antibodies and DAPI. Scale bar, 25 µm. d Western blot for α-tubulin and GAPDH in Wt;Pf4Cre, Ptipf/f;Pf4Cre and DKO;Pf4Cre megakaryocytes. e The diameter of randomly selected cells (n = 8) was analyzed using ImageJ software. Values are mean ± SD. P values were determined by Student’s t-test (unpaired, two-sided), ns not significant. f Bone marrow HSCs from Wt;Pf4Cre, Ptip^f/f;Pf4Cre, and DKO;Pf4Cre mice were treated with 10 ng/mL TPO and 15 U/mL heparin for 4 days. Total megakaryocytes were stained with BV421 conjugated anti-CD41 antibodies. Platelet-releasing megakaryocytes are indicated by white arrowheads. g The percentage of platelet-releasing megakaryocytes in total CD41+ cells were analyzed (n = 5). Values represent the mean ± SD. P values were determined by Student’s t-test (unpaired, two-sided), ns not significant. h Representative transmission electron microscopy images of fetal liver-derived megakaryocytes from Wt;Pf4Cre, Ptip^f/f;Pf4Cre, and DKO;Pf4Cre embryos. Nucleus/cytoplasm (N/C) ratio is shown for each representative megakaryocyte. N nucleus, DMS demarcation membrane system, Pro-PLT proplatelet. Source data are provided as a Source Data file.

Fig. 5. In situ platelet-forming activities in Ptip^f/f;Pf4Cre and DKO;Pf4Cre megakaryocytes.

Bone marrow from Wt;Pf4Cre, Ptip^f/f;Pf4Cre, and DKO;Pf4Cre mice was fixed and decalcified for transmission electron microscopy. a–c Representative images of WT, PTIP KO, and DKO megakaryocytes adjacent to the vascular sinusoid are shown. Green dashed lines indicate megakaryocytes and red dashed lines show the edge of vessels, scale bar, 5 µm. d–f Selected areas in the WT, PTIP KO, and DKO megakaryocytes at higher magnification showing the intracellular membrane activities (areas indicated by blue dashed boxes in a–c) scale bar, 1 µm. g–i Proplatelet and platelet forming areas in the WT, PTIP KO and DKO megakaryocytes (areas indicated by blue dashed boxes in d–f), scale bar, 0.5 µm. j–l Other representative areas in the WT, PTIP KO, and DKO megakaryocytes showing differential membrane activities, scale bar, 1 µm (n = 3). Images match the orange squares in a-c. N nucleus, DMS demarcation membrane system, ER endoplasmic reticulum, A α-granule.

Fig. 6. Alteration of gene expression in Ptip^f/f;Pf4Cre and DKO;Pf4Cre megakaryocytes.

Fetal liver-derived Wt;Pf4Cre, Ptip^f/f;Pf4Cre, and DKO;Pf4Cre megakaryocytes were collected and processed for RNA-seq (n = 3). a Box plot of log2 expression relative to the average in Wt;Pf4Cre, Ptip^f/f;Pf4Cre, and DKO;Pf4Cre megakaryocytes. Genes downregulated and upregulated in DKO;Pf4Cre compared to Wt;Pf4Cre are presented on the left and right, respectively. Each box represents the interquartile range (IQR; 25th to 75th percentile), the center line indicates the median, and whiskers extend to the minimum and maximum values within 1.5 × IQR. Data beyond this range is not shown. b Correlation analysis for the genes with significant expression changes in Ptip^f/f;Pf4Cre and DKO;Pf4Cre megakaryocytes. Purple dots indicate genes with significant expression changes in both PTIP KO and DKO MKs, red dots indicate genes with significant expression changes in DKO MKs only, and blue dots indicate genes with significant expression changes in PTIP KO MKs only. c Venn diagrams showing the overlap of genes with significant expression changes between the indicated comparison groups. Red circles indicate the number of genes that were commonly up- or down-regulated between DKO MKs and WT MKs. Blue circles indicate the number of genes that were commonly up or down regulated between PTIP KO MKs and WT MKs. The P value of the overlap (by hypergeometric test) is shown under each Venn diagram. d Heatmap showing the relative expression of genes commonly down-regulated and up-regulated in both Ptip^f/f;Pf4Cre and DKO;Pf4Cre megakaryocytes when compared to Wt;Pf4Cre. e Megakaryocyte- and platelet-specific gene signatures regulated by PTIP or MLL3/4. Gene set enrichment analysis (GSEA) normalized enrichment score (NES) and false discovery rate (FDR) are indicated, with FDR < 0.05 highlighted in red. f GSEA enrichment plots (score curves) showing individual gene sets that regulate megakaryocyte development and differentiation, comparing DKO and WT megakaryocytes. Net enrichment score and FDR value of each gene set is presented in Fig. 6e.

Fig. 7. MLL3/4 regulate platelet-specific genes at the enhancer regions.

Fetal liver-derived Wt;Pf4Cre, Ptip^f/f;Pf4Cre and DKO;Pf4Cre megakaryocytes were collected, processed and incubated with anti-H3K4me1, anti-H3K27ac, and anti-H3K4me3 antibodies for CUT&RUN analysis (n = 3). a Bar plots showing the numbers of peaks with decreased (left) and increased (right) signal for the indicated histone marks in Ptip^f/f;Pf4Cre and DKO;Pf4Cre megakaryocytes compared to Wt;Pf4Cre megakaryocytes. b Correlation analysis for the significantly altered peaks for H3K4me1 and H3K27ac histone marks in DKO megakaryocytes. Purple dots indicate the significantly altered peaks for both H3K4me1 and H3K27ac, red dots indicate the altered peaks for H3K4me1 only, and blue dots indicate the altered peaks for H3K27ac only. c Metaplots showing the signal distribution of H3K4me1 (left), H3K27ac (middle), and H3K4me3 (right) over the peaks that potentially regulate a portion of the platelet-specific genes with altered expression identified from the GSEA analysis. d CUT& RUN signal tracks showing the enrichment of H3K4me1, H3K27ac, and H3K4me3 marks at the Adra2a gene locus in WT, PTIP KO, and DKO megakaryocytes. The green box indicates the promoter region, and the blue box indicates a neighboring enhancer region. e ChIP-qPCR was performed to test the binding of MLL3/4 to the enhancer region (blue box in (d) n = 3), as well as the binding of PTIP to the promoter region (green box in (d) n = 3). f Design of primers used for the analysis of promoter-enhancer interaction at Adra2a gene locus. P promoter, E enhancer, NC negative control. g 3C-qPCR showing the relative promoter-enhancer interaction at the Adra2a gene locus, normalized to the DNA input, using primers targeting the promoter region (n = 3). P-E: interaction between promoter and enhancer, P-NC: interactions between promoter and negative control region. Values represent the mean ± SD. P values were determined by Student’s t-test (unpaired, two-sided). Source data are provided as a Source Data file.