Fig 2: Tet2- and Tet3-mediated DNA demethylation is required for the activation of Blimp1-NFATc1 axis AScatter plot of fold changes in RNA-seq data and MeDIP-seq data. The log2 fold changes in RNA-seq expression between wild-type control and Tet2Rank –/–; Tet3aRank +/– BMMs stimulated with RANKL for 2 days are shown on the x-axis, and log2 fold changes in MeDIP-seq expression (> 0) are shown on the y-axis. The names of significant genes are labeled (FDR of differential expression analysis < 0.01, p-value of differential methylated region analysis < 0.01).BMeDIP-qPCR analysis to validate regions that are hypermethylated in Tet2Rank –/–; Tet3aRank +/– BMMs stimulated with RANKL for 2 days. Data denote mean ± s.e.m. *P < 0.05; **P < 0.01 (n = 3–5 biological replicates; t-test).CGene expression of TET target candidates in BMMs stimulated with RANKL for 2 days. Data denote mean ± s.e.m. **P < 0.01 (n = 4–7 biological replicates; t-test).DGenome browser view of the expression and the methylation site distribution in the Prdm1 gene locus of control and Tet2Rank –/–; Tet3aRank +/– BMMs stimulated with RANKL for 2 days.EBisulfate sequencing analysis of selected region of the second intron of the Prdm1 gene in control and Tet2Rank –/–; Tet3aRank +/– BMMs stimulated with RANKL for 2 days.FEffect of retroviral NFATc1 expression on osteoclastogenesis of control and Tet2Rank –/–; Tet3aRank +/– BMMs stimulated with RANKL for 3 days. TRAP-stained cells (left panel) and the number of TRAP-positive cells with more than three nuclei (right). Scale bars, 100 µm. Data denote mean ± s.e.m. **P < 0.01 (n = 3 biological replicates; ANOVA).GBisulfate sequencing analysis of the selected region of the second intron of the Prdm1 gene in BMMs stimulated with RANKL and culture under 5% and 2% oxygen for 3 days.HEffect of retroviral NFATc1 expression on osteoclastogenesis of BMMs stimulated with RANKL and culture under 5% and 2% oxygen for 5 days. Control and Tet2Rank –/–; Tet3aRank +/– BMMs stimulated with RANKL for 3 days. TRAP-stained cells (left panel) and the number of TRAP-positive cells with more than three nuclei (right). Scale bars, 100 µm. Data denote mean ± s.e.m. **P < 0.01 (n = 3 biological replicates; ANOVA).

Fig 3: Loss of HDAC1 prevents PB differentiation by perturbing the gene regulatory network that specifies PB versus B cell fate. (A) Experimental scheme of co-transduction for activated B cells transduced with combinations of retroviral particles expressing Hdac1-targeting sgRNA (BFP+) and Hdac2 or Bach2-targeting sgRNA (mCherry+). (B) Western blots for HDAC1 and HDAC2 proteins showing gene KO efficiencies of sgRNAs targeting Hdac1 and Hdac2 genes. SgRosa26-1 was used as a negative control. Beta-Tubulin used as a loading control. (C) FACS analysis of one technical repeat experiment at day 6 post transduction showing frequencies of CD138+ PBs in WT (GFP-, top) and Cas9 (GFP+, bottom) cells. (D) Scatter plot showing the ratio of GFP+ (Cas9)/GFP- (WT) PBs corresponding to the experiment shown in (C) plus two independent repeat experiments. Each dot corresponds to one independent biological repeat experiment averaged over two technical replicates. (E) FACS analysis of one representative technical replicate of three independent experiments showing frequencies of GFP+ B cells in Hdac1 KO (sgHdac1-1, blue), Hdac2 KO (sgHdac2-2, black), or Hdac1 and Hdac2 double KO (dKO) (sgHdac1-1 and sgHdac2-2, red) cell populations in a single culture well, at day 2 and 6 post co-transduction. (F) Western blots showing expression levels of HDAC1, BLIMP1, IRF4, BACH2 and PAX5 transcription factors in sorted CD138- activated B cells (aB) and CD138+ PBs (PB) treated with indicated sgRNAs. Beta-Actin was used as a loading control. (G) Pre-gating (left) on Hdac1 KO (sgHdac1-1, blue), Bach2 KO (sgBach2-1, black), or Hdac1 and Bach2 dKO (sgHdac1-1 and sgBach2-1, red) cell populations present in a single culture well for the FACS plot (middle) showing frequencies of CD138+ PBs in WT (GFP-, top) and Cas9 (GFP+, bottom) cells in each pre-gated cell population. Scatter plot (right) showing the ratio of GFP+/GFP- PBs corresponding to the experiment shown on the middle (FACS analysis; one technical replicate of three) plus independent biological repeat experiments with different sgRNA combinations (n=3 independent experiments averaged over three technical replicates each; horizontal line indicates median).