Tait J et al. (2026), Catalytic activity of KMT5B promotes ciliogenes...

XB-ART-61802

Life Sci Alliance 2026 Jun 07;96:. doi: 10.26508/lsa.202503455.

Show Gene links Show Anatomy links

Catalytic activity of KMT5B promotes ciliogenesis without affecting global chromatin accessibility.

Tait J, Marthen C, Hoelscher B, Straub T, Hsam O, Rupp RA.

???displayArticle.abstract???
Multiciliated cells (MCCs) are specialized cells found in the brain, reproductive and respiratory tracts of mammals, and the epidermis of tadpole-stage Xenopus embryos. KMT5B and KMT5C are histone methyltransferases that deposit the dimethyl mark on histone 4 lysine 20 (H4K20). We previously showed that KMT5B/C double knockdown down-regulates H4K20me2 levels in bulk chromatin, as well as transcription of ciliary genes. MCCs of embryos lacking both enzymes, or only KMT5B, have depleted cilia. Here, we separate the function of KMT5B in multiciliogenesis and show that single knockdown of KMT5B, not KMT5C, leads to aberrant transcription and down-regulation of ciliary genes. This phenotype is rescued by catalytically active PHF8, an H4K20me1 demethylase, whereas hormone-inducible multicilin (MCI), master regulator of cilia, has no effect. Notably, the expression of key transcription factors of ciliogenesis is unaffected by KMT5B depletion, which dominates the transcriptional response to ectopic multicilin. Finally, ATAC-seq in animal caps shows KMT5B knockdown results in few differentially accessible peaks and does not compact chromatin at ciliary genes. This suggests KMT5B regulates MCCs via an alternative pathway to the canonical MCI-driven programme.

???displayArticle.pubmedLink??? 41946567
???displayArticle.pmcLink??? PMC13057974
???displayArticle.link??? Life Sci Alliance

???attribute.lit??? ???displayArticles.show???

	Figure 1. Knockdown of KMT5B, but not KMT5C, leads to phenotypic and transcriptional changes.(A) 4-channel confocal images depicting single knockdowns of KMT5B or KMT5C in Xenopus laevis embryos. Cilia are magenta (acetylated α-tubulin), basal bodies are green (hyls1-GFP), actin meshwork and cell borders are red (phalloidin), and DNA is blue (DAPI). Embryos are injected in one of two ventroanimal blastomeres at the eight-cell stage, giving rise to mosaic embryos in which uninjected (labelled wt in white) and knockdown (white asterisks) multiciliated cells can be visualized in the same field of view. (B, C) Venn diagrams showing the number of genes that are up-regulated (B) or down-regulated (C) upon knockdown of KMT5B or KMT5C in Xenopus laevis animal caps by RNA-seq compared with control morphants. (D, E) MA plot showing log2 fold change (x-axis) and log10 mean counts (y-axis) in KMT5B (D) or KMT5C (E) knockdown animal caps by RNA-seq.
	Figure S1. Knockdown of KMT5C does not lead to a change in cell size.(A) 4-channel confocal images depicting control morphant or KMT5C single knockdown Xenopus laevis embryos. Cilia are magenta (acetylated α-tubulin), basal bodies are green (hyls1-GFP), actin meshwork and cell borders are red (phalloidin), and DNA is blue (DAPI). Embryos are injected in one of two ventroanimal blastomeres at the eight-cell stage. (A′) Enlarged inlays from Panel (A). Scale bars = 10 μm.
	Figure S2. Representative images of three KMT5B knockdown multiciliated cells.Top images represent orthogonal views showing clumped BBs near cell nucleus; bottom row shows flattened Z-stacks. Cell in (A, B, C) shows that a neighbouring, uninjected multiciliated cell is fully ciliated, whereas the basal bodies of the KMT5B morphant cell are still located deep. Cilia are magenta (acetylated α-tubulin), basal bodies are green (hyls1-GFP), actin meshwork and cell borders are red (phalloidin), and DNA is blue (DAPI). Embryos are injected in one of two ventroanimal blastomeres at the eight-cell stage. Scale bars = 10 μm.
	Figure 2. KMT5B regulates ciliogenesis through an alternative pathway.(A, B) GO analysis of genes that are up-regulated (A) or down-regulated (B) upon KMT5B knockdown. Up-regulated genes largely relate to mRNA processing and macromolecule localization, whereas down-regulated genes relate to cilia and microtubules. (C) Log2 fold change of multiciliogenic regulators upon KMT5B knockdown. None of the major known ciliogenic regulators are significantly misregulated based on RNA-seq results. (D) Log2 fold change of E2F genes upon KMT5B knockdown. Only the L and S copies of cell cycle–related e2f1 are down-regulated, whereas tfdp1.S is up-regulated.
	Figure 3. Catalytic activity of PHF8 rescues the ciliogenic phenotype.(A) Representative immunocytochemistry images from tailbud-stage embryos injected with KMT5B Mo or CoMo and mRNA luciferase or human PHF8 variants (489 wt/489 c.i.). Scale bars = 1 mm (whole embryo) or 200 μm (inserts) and n = 3 biological replicates. (B) Quantification of (A). Percentage of embryos either affected by morpholino injections (H1Mo) or rescued by PHF8 mRNA injection. Wt PHF8 mRNA (489 Wt) rescues significantly, whereas catalytically inactive PHF8 mRNA (489 c.i.) shows no significant difference when compared to H1Mo. Significance is indicated by asterisks; P < 0.05, *P < 0.01, ns, not significant. N = 4 biological replicates; the yellow dashed line represents regions of ß-galactosidase staining, indicating injected regions.
	Figure 4. KMT5B knockdown drives transcriptional and phenotypic changes even in the presence of MCI-hGR.(A) Four-channel confocal microscopy images showing Xenopus laevis embryos injected with control morpholino (CoMo) + MCI-hGR or KMT5B morpholino + MCI-hGR. Cilia are magenta (acetylated α-tubulin), basal bodies are green (hyls-GFP), actin meshwork and cell borders are red (phalloidin), and DNA is blue (DAPI). Embryos are injected in one of two ventroanimal blastomeres at the eight-cell stage, giving rise to mosaic embryos in which wt and knockdown multiciliated cells can be visualized in the same field of view. Uninjected cells are labelled in white as “wt,” whereas injected cells are indicated with an asterisk (*). Scale bars = 10 μm. (B) Cluster analysis heatmap of top responding genes across all conditions and replicates. Genes group into two main clusters: genes down-regulated or up-regulated upon KMT5B knockdown (in the absence or presence of MCI-hGR). Group A includes replicates from the single knockdown RNA-seq experiment, and Group B includes replicates from the MCI-hGR containing RNA-seq experiment, and samples are batch-corrected between experiments. (C) MA plot showing all genes (light grey), significantly misregulated genes (dark grey), ciliary genes (yellow), and significantly misregulated genes (red). (D, E) GO analysis showing the up-regulated (D) and down-regulated (E) gene categories upon MCI-hGR overexpression. The down-regulated gene categories are largely related to ciliogenesis and microtubule assembly.
	Figure S3. Overexpression of MCI leads to up-regulation of multiciliogenic genes.(A) Protocol for multicilin overexpression. Embryos are injected in both cells of the two-cell stage with the morpholino and mRNA of interest. They are allowed to develop until the blastula stage (NF8-9), at which point animal cap explants are dissected. At the late blastula stage (NF11), animal caps are induced with 10 μM dexamethasone. Animal cap explants are harvested at the neurula stage (NF16) and further processed for RNA-seq. Simultaneously, sibling embryos are injected on one cell of the two-cell stage with constructs of interest and allowed to develop until the late blastula stage (NF11). Embryos are then induced with 10 μM dexamethasone. At the tailbud stage (NF28), embryos are fixed for whole-mount immunocytochemistry. (B) Representative images of quality control embryos assessed to exhibit both a KMT5B knockdown phenotype and a multicilin overexpression phenotype. Embryos were stained with anti-acetylated α-tubulin for ciliary tufts. (C) Quantification of (B).
	Figure S4. Overexpression of MCI leads to up-regulation of multiciliogenic genes.(A, B) MA plot (A) and volcano plot (B) showing the change in gene expression in control morpholino + MCI-hGR–overexpressing animal caps. (C, D) GO analysis showing the up-regulated (C) and down-regulated (D) gene categories upon MCI-hGR overexpression. The up-regulated gene categories are largely related to ciliogenesis and microtubule assembly.
	Figure S5. Genes upregulated upon MCI overexpression are overlapping but distinct from genes downregulated by KMT5B knockdown.(A) UpSet plot detailing the number of overlapping and non-overlapping gene numbers comparing KMT5B knockdown and MCI-hGR overexpression. (B) Bubble plot of gene ontology (GO) analysis of genes that were down-regulated upon KMT5B knockdown and up-regulated upon MCI- hGR overexpression.
	Figure S6. Comparing gastrocoel roof plate (GRP) ciliary length and prevalence between KMT5B morphant and control morphant embryos.(A, B) Scanning electron micrographs of the inner explant surface of a CoMo (A)- or KMT5B-Mo (B)–injected explant. Magnification 480×. Scale bar in (A) = 10 μm. Panels (C, D) show ImageJ-assisted assignment of cell borders and ciliary position within a 320 × 320 μm large area from the centre of the SEM images shown above. (E, F) Comparative analysis of ciliary length, posterior ciliary length, and number of GRP cilia between KMT5B and control morphant embryos. No significant difference in any comparison (P < 0.05). N = 4 explants/condition, total cell numbers: 97/153 cells ciliated in the CoMo condition; 89/118 cells ciliated in the KMT5B-Mo condition.
	Figure 5. Depletion of KMT5B leads to increased chromatin accessibility.(A) MA plot showing the distribution of peaks in KMT5B knockdown animal caps. Significantly changing peaks are shown in red (P < 0.05). Most of the peaks are increasing rather than decreasing, indicating a mild chromatin opening effect of KMT5B knockdown in Xenopus laevis animal caps. (B) Distribution of peaks across genomic features in all peaks and changing peaks. N = 3 biol. replicates. (C) Representative browser tracks of the most significantly down-regulated ciliogenic genes from the KMT5B knockdown RNA-seq experiment. (D) Representative browser tracks of the most significantly changing peaks between the KMT5B knockdown and control animal caps.
	Figure S7. KMT5B knockdown does not lead to decreased accessibility of key ciliogenic regulators.Representative ATAC-seq browser tracks of key ciliogenic transcription factors comparing KMT5B knockdown (red) against control knockdown (grey).