Fig 1.
miR expression profiling and QTL analysis reveal upstream candidate genes of miR-29a.
(A) Expression of miR-29a in hippocampi of 54 CC mice strains. (B) LOD score plot of miR-29a expression using Gene Miner. The QTL was detected on chromosome, Chr 7 (red) with a significant score above 12.5. (C) Gene ontology (GO) analysis significantly represented for the candidate genes that were enriched in modules related to both regulation of chromatin structure and neuronal activities. Bar graphs indicate the statistical significance of the enrichment, as -log10 (adj p-value; cut-off level for significance, *p< < 0.05, adjusted by Benjamini-Hochberg correction). (D) The effectiveness of candidate gene knockdown using siRNAs was assessed through qPCR, comparing expression levels to those of the scrambled negative control. N= = 3 experiments, mean ± SD, ****p< < 0.0001, One-way Anova. (E) Bar graph illustrating change in miR-29a expression, determined through RNAi and qPCR. Data from three individual experiments are shown here represented as mean ± SD.
Fig 2.
Screening of upstream candidate genes using neuronal differentiation in mNPCs.
(A-B) Role of candidate genes in neuronal differentiation. Representative images (A) and quantification (B) of βIII-tubulin positive cells using immunostaining in differentiated mNPCs transfected with individual siRNAs and scrambled siRNA negative control. N = 4 experiments. mean ± SD, **p < 0.01. One-way Anova. Image scale bars = 10 μm. (C-D) Representative histograms (C) and quantification (D) of βIII-tubulin detection using flow cytometry in differentiated mNPCs transfected with siPsmd13, siNap1l4 and siRNA of non-targeting control. N = 3 experiments. mean ± SD, *p < 0.05. One-way Anova. (E-F) miR-29a alter neuronal differentiation of mNPCs. Representative images (E) and quantification (F) of βIII-tubulin positive cells using immunostaining in differentiated mNPCs transfected with miR-29a inhibitor and miR negative control. N = 3 experiments. mean ± SD, *p < 0.05. Unpaired T-test. Image scale bars = 10 μm. (G-H) Representative histograms (G) and quantification (H) of βIII-tubulin detection using flow cytometry in differentiated mNPCs transfected with miR-29a inhibitor and inhibitor control. N = 3 experiments. mean ± SD, **p < 0.01. Unpaired T-test. (I-J) Psmd13 acts through miR-29a to alter neuronal differentiation. Representative images (I) and quantification (J) of βIII-tubulin positive cells using immunostaining in differentiated mNPCs transfected with siRNA non-targeting control, siPsmd13 plus mimic control, siPsmd13 plus miR-29a mimics and miR-29a mimics. N = 3 experiments. mean ± SD, **p < 0.01, *p < 0.05. One-way Anova. Image scale bars = 10 μm. (K-L) Representative histograms (K) and quantification (L) of βIII-tubulin detection using flow cytometry in differentiated mNPCs transfected with control plus siRNA non-targeting control, siPsmd13 plus mimic control siPsmd13 plus miR-29a mimics and miR-29a mimics. N = 3 experiments. mean ± SD, *p < 0.05. One-way Anova.
Fig 3.
Psmd13 associates with Dicer and regulate miR-29a expression in mNPCs.
(A) Quantitative analysis of the expression fold change of miR-29a (by Taqman PCR), in siControl and siPsmd13 undifferentiated mNPCs. (B) Quantitative analysis of the expression fold change of miR-29a (by Taqman PCR), in siControl and siPsmd13 differentiated mNPCs. (C) Quantitative analysis of the expression fold change of Dicer in siControl and siPsmd13 mNPCs in undifferentiated mNPCs. (D) Quantitative analysis of the expression fold change of Dicer in siControl and siPsmd13 mNPCs in differentiated mNPCs. N = 3 experiments. mean ± SD, **p < 0.01. Unpaired T-test. (E) Western blotting was conducted using specific antibodies to detect Dicer and Actin proteins in extracts from control and Psmd13-depleted mouse neural progenitor cells under undifferentiated conditions. (F) Western blotting was conducted using specific antibodies to detect Dicer and Actin proteins in extracts from control and Psmd13-depleted mouse neural progenitor cells under differentiated conditions. (G) Dual luciferase reporter assay in mNPCs for miR-29a and mutated miR-29a sites. Values for Firefly Luciferase luminescence normalized to Renilla luminescence. Fold change relative to control in undifferentiated mNPCs. (H) Dual luciferase reporter assay in mNPCs for miR-29a. Values for Firefly Luciferase luminescence normalized to Renilla luminescence. Fold change relative to control in differentiated mNPCs. N = 3 experiments. mean ± SD, *p < 0.05., ns – not significant. Unpaired T-test. (I) Co-IP of Dicer was performed with corresponding antibody followed by western blotting to detect endogenous Psmd13 proteins in undifferentiated mNPCs. Input and IgG antibody was used as controls for the experiment. (J) Co-IP of Psmd13 was performed with corresponding antibody followed by western blotting to detect endogenous Dicer proteins in undifferentiated mNPCs. Input and IgG antibody was used as controls for the experiment. (K) Co-IP of Dicer was performed with corresponding antibody followed by western blotting to detect endogenous Psmd13 proteins in differentiated mNPCs. Input and IgG antibody was used as controls for the experiment. (L) Co-IP of Psmd13 was performed with corresponding antibody followed by western blotting to detect endogenous Dicer proteins in differentiated mNPCs. Input and IgG antibody was used as controls for the experiment.
Fig 4.
Psmd13 depletion impacts global miR regulation in mNPCs.
(A) Box plots of read distributions for significantly differentially bound sites in the undifferentiated mNPCs between Psmd13-depleted and control groups. (B) Box plots of read distributions for significantly differentially bound sites in the differentiated mNPCs between Psmd13-depleted and control groups. (C) Volcano plot of differentially expressed miRs (padj < 0.05). Log2 fold change between differentiated and undifferentiated mNPCs is plotted on the x-axis and the − log10 of the padj value is plotted on the y-axis. (D) The top 20 enriched GO terms of molecular function in the undifferentiated state. Circle size indicates the number of genes enriched in each term. Color saturation represents the significance level. (E) The top 20 enriched GO terms of molecular function in the differentiated state. Circle size indicates the number of genes enriched in each term. Color saturation represents the significance level.
Fig 5.
Psmd13 dependency for Dicer binding at miR-29a locus in mNPCs.
(A) ChIP‐seq density heatmaps of Psmd13 and Dicer binding, and ChIP seq of different histone marks in mNPCs are shown within the extended gene regions (−5 kb of TSS to +5 kb of TSS) of total genome‐wide genes. (B) Density map showing the global binding profiles of Dicer and Psmd13 ChIP signal. The x axis represents the distance to the TSS. The y axis represents per base read coverage. (C) Overlap of the genes annotated between Psmd13 and Dicer displayed as Venn diagram. The P‐value was calculated using Fisher’s exact test. (D) Transcription factor binding motifs enriched in the Dicer and Psmd13 ChIP-seq peak dataset identified by MEME-ChIP. (E) The top 20 enriched GO terms of molecular function. Circle size indicates the number of genes enriched in each term. Color saturation represents the significance level. (F) Individual ChIP analysis showing chromatin co-localization of Psmd13 and Dicer at the miR-29a genomic loci in mNPCs (N = 3, data are shown as mean ± SD). The binding was confirmed against a control region on the miR-29a locus. (G) ChIP-seq profiles showing Dicer binding at Psmd13 loci in Psmd13-depleted undifferentiated and differentiated mNPCs, as well as in WT mNPCs. Peaks are visualized in IGV, with undifferentiated samples shown in brown, differentiated in green, and WT in magenta, all exhibiting enrichment over the Input control. WT = wild- type. (H) Individual ChIP analysis comparing chromatin binding of Dicer at miR-29a genomic loci in control and Psmd13-depleted undifferentiated mNPCs (N = 3, data are shown as mean ± SD). Unpaired t test was used for statistical analysis (**p < 0.01). (I) Individual ChIP analysis comparing chromatin binding of Dicer at miR-29a genomic loci in control and Psmd13-depleted differentiated mNPCs (N = 3, data are shown as mean ± SD). Unpaired t test was used for statistical analysis (****p < 0.0001). (J) Fold changes of Dicer binding versus signal intensity between undifferentiated and differentiated mNPCs are visualized as MA plot. Pink represents differentially bound peaks (FDR < 0.05). The x axis values (“log concentration”) represent logarithmically transformed, normalized counts, averaged for all samples, for each site. The y axis values represent log 2 (fold change) values.
Fig 6.
Association of proteasome inhibition with Dicer levels and miR-29a expression in mNPCs.
(A) Western blot showing PSMD13 protein levels in mNPCs following treatment with 5 μM MG132, 0.2 μM Capzimin, and DMSO control for the specified durations. B-actin was used as a loading control. (B) Quantitative analysis of the expression fold change of Psmd13 mRNA in MG132 treated mNPCs in a time-dependant experiment (normalised to control). N = 3 experiments. mean ± SD, ****p < 0.0001. One-way Anova. (C) Western blot showing the levels of Dicer protein after treatment with 5 μM MG132, 0.2 μM Capzimin and DMSO control in mNPCs for the indicated time duration. B-actin was used as a loading control. (D) Quantitative analysis of the expression fold change of Psmd13 mRNA in MG132 treated mNPCs in a time-dependant experiment (normalised to control). N = 3 experiments. mean ± SD, **p < 0.01, ****p < 0.0001. One-way Anova. (E) Quantitative analysis of the expression fold change of miR-29a (by Taqman PCR), in MG132 treated mNPCs in a time-dependant experiment (normalised to control). N = 3 experiments. mean ± SD, ****p < 0.0001. One-way Anova. (F-G) Representative images (F) and quantification (G) of βIII-tubulin positive cells using immunostaining in differentiated mNPCs treated with MG132, Psmd13-depleted and control mNPCs. N = 3 experiments. mean ± SD, **p < 0.01, *p < 0.05. One-way Anova. Image scale bars = 10 μm. (H) ChIP-seq profiles showing Dicer binding at Psmd13 loci in MG132 treated and untreated mNPCs. Peaks are visualized in IGV, with MG132 shown in blue and untreated in green, all exhibiting enrichment over the Input control. (I) Individual ChIP analysis comparing chromatin binding of Dicer at miR-29a genomic loci in MG132, Psmd13-depleted and control mNPCs (N = 3, data are shown as mean ± SD). (J) Fold changes of Dicer binding between MG132 and untreated mNPCs are visualized as MA plot. Pink represents differentially bound peaks (FDR < 0.05). The x axis values (“log concentration”) represent logarithmically transformed, normalized counts, averaged for all samples, for each site. The y axis values represent log 2 (fold change) values. (K) Box plots of read distributions for significantly differentially bound sites in the MG132 and untreated mNPCs. (L) Heatmaps showing the enrichment of Dicer reads between MG132 and untreated mNPCs in a 1500 bp window centered around the TSS. Scale is as indicated in the signal.
Fig 7.
Psmd13-Dicer interaction modulates miR-29a expression and neuronal differentiation in mNPCs.