Xap5 is a nuclear protein essential for primary ciliogenesis.

a Immunofluorescence analysis of endogenous Xap5 (green) in NIH/3T3 cells. Nuclei were counterstained with DAPI (blue). Scale bar, 10 μm. b Validation of the Xap5 nuclear localization signal (NLS). Representative images of NIH/3T3 cells expressing either Flag-tagged wild-type Xap5 (WT) or an NLS-deleted mutant (ΔNLS) (green). Scale bar, 10 μm. c Western blot analysis confirming the complete depletion of Xap5 protein in the Xap5 knockout (KO) clonal cell line. Gapdh serves as a loading control. The blot shown is representative of three independent biological replicates. d Schematic illustrating the time-course assay for primary cilia assembly induced by serum starvation. CS, calf serum. e Representative immunofluorescence images of cilia from WT and Xap5 KO cells at the indicated time points. Cilia were co-stained for acetylated tubulin (red) and Arl13b (green). Nuclei were counterstained with DAPI (blue). Scale bar, 5 μm. f Quantification of primary cilia length in WT and Xap5 KO cells over a 48-h time course of serum starvation. Data are presented as means ± SEM (n = 3 independent experiments). g Quantification of the percentage of ciliated cells as in f. Data are presented as mean ± SEM (n = 3 independent experiments). All statistical significance between WT and KO at each time point was determined by multiple two-tailed unpaired t-tests. n.s., not significant; **P < 0.01; ***P < 0.001. Source data are provided as a Source Data file.

Xap5 forms a nuclear complex with its cofactor Nono.

a Representative immunofluorescence images showing the predominantly nuclear localization of Nono in NIH/3T3 cells. Top: Endogenous Nono detected with a specific anti-Nono antibody (green). Bottom: Ectopically expressed Nono-GFP fusion protein (green). Nuclei were counterstained with DAPI (blue). Scale bars, 10 μm. b Endogenous co-immunoprecipitation (Co-IP) of Xap5 and Nono from NIH/3T3 cell lysates. Left: Immunoprecipitation (IP) with an anti-Xap5 antibody specifically co-precipitated endogenous Nono. Right: Conversely, IP with an anti-Nono antibody co-precipitated endogenous Xap5. Control IgG was used as a negative control for the IP. The resulting precipitates (IP) and input lysates were analyzed by western blotting (IB) with the indicated antibodies. Data are representative of three independent experiments. Source data are provided as a Source Data file.

Loss of Nono phenocopies Xap5 depletion, impairing primary cilia elongation.

a Western blot analysis confirming the complete depletion of Nono protein in Nono KO cell lysates compared to WT. The blot shown is representative of three independent biological replicates. Gapdh serves as a loading control. b Representative immunofluorescence images of primary cilia in WT and Nono KO cells after 48 h of serum starvation. Cilia were co-stained for acetyl-tubulin (red) and Arl13b (green). Nuclei were counterstained with DAPI (blue). Scale bar, 5 μm. c Quantification of primary cilia length in WT and Nono KO cells over a 48-h time course of serum starvation. Data are presented as mean ± SEM (n = 3 independent experiments). d Quantification of the percentage of ciliated cells as in c. Data are presented as mean ± SEM (n = 3 independent experiments). All statistical significance between WT and KO at each time point was determined by multiple two-tailed unpaired t-tests. n.s., not significant; **P < 0.01; ***P < 0.001. Source data are provided as a Source Data file.

Xap5 and Nono co-regulate a ciliogenic transcriptional network.

All analyses were performed on NIH/3T3 cells following 24 h of serum starvation. a, b Representative enriched Gene Ontology (GO) terms (Biological Process) for genes downregulated in Xap5 KO (a) and Nono KO (b) cells. c Venn diagram illustrating the significant overlap between genes downregulated in Xap5 KO cells, Nono KO cells, and a curated list of cilia-associated genes (see also Supplementary Data 3). d, e qPCR validation of shared, downregulated ciliary target genes in Xap5 KO (d) and Nono KO (e) cells. Data are presented as mean ± SEM from three independent experiments. P values were determined by two-tailed unpaired t-test. *P < 0.05, **P < 0.01, ***P < 0.001. f Representative western blots showing reduced protein levels of Sox5 and Sox9 in Xap5 KO (left) and Nono KO (right) cells compared to their respective WT controls. Gapdh serves as a loading control. Blots are representative of three independent experiments. Source data are provided as a Source Data file.

Xap5 and Nono directly co-occupy the regulatory regions of ciliary genes.

a, b Genome-wide binding profiles of Xap5 (a) and Nono (b) in NIH/3T3 cells determined by CUT&Tag. Top panels show the average enrichment profiles over gene bodies (from transcription start site, TSS, to transcription end site, TES). Bottom panels show heatmaps of binding signals centered on the TSS (±3 kb). c, d Venn diagrams illustrating the significant overlap between genes bound by Xap5 (c) or Nono (d) and the respective sets of differentially expressed genes identified upon their knockout. e Comparison of de novo DNA-binding motifs derived from Xap5 and Nono CUT&Tag peaks in this study, alongside the Xap5 motif from our previous work in testicular germ cells28. The high degree of similarity indicates a conserved DNA-binding preference across different cellular contexts. f Venn diagram showing the overlap between direct Xap5 target genes (defined as genes that are both bound by Xap5 and downregulated in Xap5 KO), direct Nono target genes, and the curated ciliary gene set (from Supplementary Data 3). g Direct co-occupancy of Xap5 and Nono at the Sox5 and Sox9 gene loci. Left: Integrative Genomics Viewer (IGV) tracks showing CUT&Tag signal enrichment. Right: CUT&Tag-qPCR validating the co-enrichment of Xap5 and Nono at the Sox5 and Sox9 promoters. Red dashed box: Xap5 and Nono binding regions. Data are presented as mean ± SEM from three independent experiments. P values were determined by two-tailed unpaired t-test. ***P < 0.001. Source data are provided as a Source Data file.

Downstream effector Sox5 is essential for primary ciliogenesis.

a Western blot analysis confirming the complete depletion of Sox5 protein in Sox5 KO cell lysates compared to WT. The blot shown is representative of three independent biological replicates. Gapdh serves as a loading control. b Representative immunofluorescence images of primary cilia in WT and Sox5 KO cells after serum starvation. Cilia were co-stained for acetylated tubulin (red) and Arl13b (green). Nuclei were counterstained with DAPI (blue). Scale bar, 5 μm. c Quantification of primary cilia length in WT and Sox5 KO cells over a 48-h time course of serum starvation. Data are presented as mean ± SEM (n = 3 independent experiments). d Quantification of the percentage of ciliated cells as in c. Data are presented as mean ± SEM from three independent experiments. P values were determined by two-tailed unpaired t-test. n.s., not significant; ***P < 0.001. Source data are provided as a Source Data file.

Sox5 executes a ciliogenic transcriptional program downstream of the Xap5-Nono complex.

a Gene Set Enrichment Analysis (GSEA) plot for the “cilium organization” pathway, performed on all ranked differentially expressed genes from the Sox5 KO transcriptome. b Representative enriched GO terms (Biological Process) for genes significantly downregulated (fold change ≥ 1.5) in Sox5 KO cells. c Venn diagram illustrating the significant overlap among genes downregulated in Xap5 KO, Nono KO, and Sox5 KO cells, identifying a common set of co-regulated genes. d qPCR validation confirming the reduced expression of shared downstream target genes in Sox5 KO cells compared to WT. Data are presented as mean ± SEM from three independent experiments. P values were determined by two-tailed unpaired t-test. *P < 0.05, **P < 0.01. Source data are provided as a Source Data file.

Design and validation of NLS-mutant and CRISPR/Cas9-mediated Xap5 KO alleles.

a Amino acid sequences of the wild-type nuclear localization signal (NLS) in Xap5 and the NLS-deleted mutant (ΔNLS) generated for this study. Key basic residues targeted for mutation are highlighted in red. b Validation of CRISPR/Cas9-mediated knockout of Xap5. Top: Schematic of the wild-type genomic locus showing the target sites for the two guide RNAs (sgRNA1 and sgRNA2, red) and their adjacent PAM sequences (blue). Bottom: Sanger sequencing chromatograms from a selected knockout clone revealing two distinct mutant alleles (allele-1 and allele-2). Both alleles share an identical mutation at the sgRNA2 target site but harbor different frameshift-inducing mutations at the sgRNA1 site, consistent with independent non-homologous end joining (NHEJ) repair events at this locus.

Identification of Xap5-interacting nuclear proteins by mass spectrometry.

Table showing selected high-confidence nuclear proteins identified by immunoprecipitation (IP) of endogenous Xap5 from NIH/3T3 cell lysates followed by mass spectrometry (IP-MS). Notable hits include core histones (e.g., Hist1h4a, H2afx) and key RNA-binding proteins such as Nono and Sfpq, suggesting Xap5 may function in complexes involved in chromatin regulation and RNA processing.

Validation of Nono KO and its impact on nuclear architecture.

a Sanger sequencing chromatogram validating the CRISPR/Cas9-mediated KO of the Nono gene in a selected NIH/3T3 clone. The WT genomic sequence, sgRNA3 target site (red), and PAM sequence (blue) are indicated above. The resulting KO allele harbors a frameshift-inducing indel at the target site, consistent with successful gene disruption. b Quantification of nuclear area (μm²) in WT and Nono KO cells over a 48-h time course of serum starvation. Data are presented as the mean ± SEM (n = 3, independent biological replicates). Statistical significance between WT and KO cells at each time point was determined by multiple two-tailed unpaired t-tests. ***P < 0.001. Source data are provided as a Source Data file.

Transcriptomic profiling of Xap5 and Nono KO cells.

All analyses were performed on cells following 24 h of serum starvation. a Volcano plot of differentially expressed genes in Xap5 KO (n=3 biological replicates) versus WT (n=2) cells. Using cutoffs of absolute fold change ≥ 1.5 and a false discovery rate (FDR) < 0.01 (highlighted dots), we identified 1,116 upregulated and 1,071 downregulated genes. b Volcano plot for Nono KO (n=3) versus WT (n=2) cells, analyzed using the same conditions and cutoffs as in a. This analysis identified 1,425 upregulated and 1,042 downregulated genes. c, d Representative enriched Gene Ontology (GO) terms for Cellular Component and Molecular Function categories among genes downregulated in Xap5 KO (c) and Nono KO (d) cells. Cilia- and transcription-related terms are prominent in both datasets, reinforcing their dual roles in ciliogenesis and nuclear function.

Genomic distribution and metagene profiles of Xap5 and Nono CUT&Tag peaks.

a, b Pie charts showing the genomic distribution of CUT&Tag peaks for Xap5 (a) and Nono (b). A substantial fraction of peaks for both Xap5 and Nono are located in promoter regions (defined as ≤3 kb upstream of a TSS). c, d Metagene plots (top) and heatmaps (bottom) showing the enrichment of Xap5 (c) and Nono (d) CUT&Tag signals centered on transcription start sites (TSS ±3 kb). The strong signal enrichment at the TSS is consistent with their roles as direct transcriptional regulators.

Validation of Sox5 KO alleles by Sanger sequencing.

Sanger sequencing chromatograms confirming CRISPR/Cas9-mediated deletion of Sox5. The wild-type genomic sequence, positions of the sgRNA4 (antisense strand) and sgRNA5 target sites (red), and the PAM sequences (blue) are indicated. Two independent Sox5 KO alleles (KO allele-1 and KO allele-2) from the same clone display distinct mutations between the sgRNA cut sites. KO allele-1 harbors distinct frameshift-inducing mutations at each of the two sgRNA target sites, while KO allele-2 contains a large deletion spanning the entire region between the sgRNA sites. These results confirm successful biallelic disruption of the Sox5 locus.

Transcriptomic analysis of Sox5 KO cells.

All analyses were performed on cells following 24 h of serum starvation. a Volcano plot of differentially expressed genes in Sox5 KO (n=3 biological replicates) versus WT (n=2) cells. Using cutoffs of absolute fold change ≥ 1.5 and FDR < 0.01 (highlighted dots), we identified 2,124 upregulated and 2,105 downregulated genes. b Representative enriched GO terms for Cellular Component and Molecular Function categories among genes downregulated in Sox5 KO cells. The significant enrichment of cilia-related terms reinforces the link between Sox5 and ciliogenesis. c Venn diagram illustrating the overlap between the curated ciliary gene set (from Supplementary Data 3) and genes either upregulated or downregulated in Sox5 KO cells. The analysis reveals that a total of 182 ciliary genes were dysregulated, consisting of 126 downregulated and 56 upregulated genes, indicating a broad impact of Sox5 on the ciliary transcriptome.

Primers used in the study.

sgRNAs used in the study.