Fig 2: PRDM10 binding is associated with transcriptional activation of target genes.a Volcano plots of RNA-seq data from Prdm10?/? vs. Prdm10F/F mESCs at indicated time-points post-deletion, with Padj < 0.05, fold-change (FC) > 2 used as the cut off to define genes with significant changes in expression. Red: genes directly bound by PRDM10; green: genes not bound by PRDM10 (left panels). Of these genes, PRDM10 targets are predominantly downregulated (right). b Volcano plot of RNA-seq data from Prdm10?/? (KO) vs. Prdm10+/+ and Prdm10?/+ (CTL) 8-cell stage embryos (left). 28 genes bound by PRDM10 showed significant expression changes (Padj < 0.05, fold-change > 2); of these, all were downregulated in Prdm10-null embryos (right). c Venn diagram depicting the overlap of bound and downregulated genes in RNA-seq datasets for embryos (8-cell) and mESCs (Day 2 and/or Day 4 post-induction), with a total of 18 genes identified as top candidates. d Heatmap showing all 18 bound and downregulated genes from Fig. 3c, categorized by their respective gene knockout phenotypes. Asterisks indicate genes directly involved in translation, Eif3b and Eef1d. Color scale: Z-score for row-normalized expression values, scaled separately for Day 2 p.i mESCs, Day 4 p.i mESCs, and embryos. Pre-implant: preimplantation lethality; post-implant: postimplantation lethality; undefined: developmental timing of embryonic lethality unknown.

Fig 3: PRDM10 is a sequence-specific zinc finger transcription factor.a Genomic feature annotation of PRDM10 binding sites (n = 528) reveals strong enrichment in proximal gene promoters. b Distribution of PRDM10 peaks relative to gene transcriptional start sites (TSS) showing that most peaks are located within 500 bp of a TSS. c Sequence motif identified as highly enriched in PRDM10 binding sites (P = 10–257). d Density histogram showing localization of motif relative to PRDM10 peak centers. e Sequence conservation profiles for motifs detected within PRDM10 peaks (blue) vs. control regions (orange). Control: all genomic regions within ±1 kb of gene TSS. Y-axis: phyloP vertebrate conservation score. Shaded regions: 25–75% percentile of conservation scores. f PRDM10 motif validation by luciferase reporter assay. HEK293T cells were transfected with reporter constructs containing either the canonical (WT) or mutated (MUT) motif sequence, together with Prdm10 expression plasmid or vector control. PRDM10 stimulates transcriptional activation only in the presence of the canonical motif. n = 3 samples. g Binding of labeled probe (10 nM) containing WT vs. MUT motif in the presence of PRDM10441–880 protein (125–1000 nM), assessed by gel-shift assay. Open arrowheads: bound probe; solid arrowheads: free probe. h Competition assay showing specificity of PRDM10 interaction with its motif. PRDM10441-880 binding to 10 nM labeled WT probe is diminished in the presence of 20- to 320-fold molar excess of unlabeled WT probe; competition with the MUT probe has no effect. i Schematic of Prdm10 full-length and mutant expression constructs tested in reporter assays (left). Expression of PRDM10FL, PRDM10N-880, and PRDM10441-C selectively activates the WT motif reporter; n = 3 samples (right). j Western blot of FLAG-tagged PRDM10 constructs from whole cell lysates of transfected HEK293T cells. Arrowheads: proteins of interest. Loading control: ß-actin. Reporter activity expressed as relative luminescence units (RLU) of firefly luminescence normalized to Renilla control (f and i). Representative data shown from one out of two (g and h) or three (f and i) independent experiments. Mean ± s.d. ***P < 0.001, ****P < 0.0001, n.s., not significant; two-tailed unpaired Student’s t test (f and i).

Fig 4: EIF3B promotes global translation downstream of PRDM10.a Immunoblotting detection of FLAG-tagged EIF3B protein expression in Prdm10F/F; CreERT2 mESCs transformed with pJ549-Eif3b or vector. b qRT-PCR validation of Eif3b transcript in Prdm10F/F; CreERT2 mESCs transformed with pJ549-Eif3b or vector, analyzed 6 days post-induction. CDS-specific primers were used to detect both endogenous and overexpressed Eif3b transcript; n = 3 biological replicates. c Growth curve analysis of Prdm10F/F and Prdm10?/? mESCs in the presence of empty vector or exogenous Eif3b, showing partial phenotypic rescue by Eif3b overexpression. Cells were passaged at constant density every 2 days and counted daily up to Day 8 post-induction; n = 3. Representative data shown from one of three independent experiments. d Average population doubling times over the course of the growth assay for Prdm10F/F and Prdm10?/? mESCs expressing vector control or exogenous Eif3b. Each point represents data from one independent experiment; n = 3. e Representative polysome profiles of Prdm10F/F and Prdm10?/? mESCs analyzed 5 days post-induction (left). A significantly reduced polysome:monosome (P/M) ratio indicates impaired global translation in Prdm10?/? mESCs (right); n = 5 biological replicates, across two independent experiments. f Representative polysome profiles of Prdm10F/F and Prdm10?/? mESCs expressing Eif3b or vector, analyzed at Day 5 post-induction (left). Eif3b overexpression restores global translation in Prdm10?/? mESCs to levels comparable to Prdm10F/F controls, as quantified by P/M ratio (right); n = 5 biological replicates across two independent experiments. Gene expression presented relative to Ubb (b). Data presented as mean ± s.d.; **P < 0.01, ****P < 0.0001, n.s., not significant; one-way ANOVA with Tukey’s multiple comparisons test (b, d and f); two-way ANOVA with Tukey’s multiple comparisons test (c).

Fig 5: PRDM10 transcriptionally regulates Eif3b expression.a ChIP-seq signal tracks showing PRDM10 occupancy at the Eif3b promoter in EtOH- vs. OHT-treated Prdm10F/F; CreERT2 mESCs, analyzed at 2 days post-induction. Vertical axis: fold-change over input. b PRDM10 enrichment at the Eif3b promoter is detected as two subpeaks, denoted P1 and P2 (left). PRDM10-dependent activation of luciferase reporters containing Eif3b promoter sequences derived from subpeak P1 or P2 (right); n = 2 samples. Representative data shown from one out of three independent experiments. c qRT-PCR quantification of changes in Eif3b transcript at indicated time-points after Prdm10 deletion; n = 2–3 biological replicates. d Western blot analysis of EIF3B protein levels in Prdm10F/F; CreERT2 mESCs at indicated time-points after treatment with EtOH (E) or 4-OHT (O). e qRT-PCR analysis showing restoration of Eif3b transcript levels in Prdm10?/? mESCs by exogenous Prdm10. Data shown at Day 4 post-induction; n = 3. Gene expression shown relative to Ubb (c and e). Error bars denote mean ± s.d.; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, n.s., not significant; two-tailed unpaired Student’s t test (b, c and e).