1. Developmental Biology
  2. Genetics and Genomics

PRMT1-SFPQ regulates intron retention to control matrix gene expression during craniofacial development

  1. Julia Raulino Lima
  2. Nicha Ungvijanpunya
  3. Qing Chen
  4. Hoang Quoc Hai Pham
  5. Tal Rosen
  6. Greg Park
  7. Mohammadreza Vantankhah
  8. Steven Yen
  9. Yang Chai
  10. Amy E Merrill
  11. Zhaoyang Liu
  12. Jian-Fu Chen
  13. Yanzhong Yang
  14. Weiqun Peng  Is a corresponding author
  15. Jian Xu  Is a corresponding author
  1. Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, United States
  2. Department of Orthodontics, Faculty of Dentistry, Chulalongkorn University, Thailand
  3. Department of Physics, The George Washington University, United States
  4. Department of Cancer Genetics and Epigenetics, City of Hope, United States
  5. Children’s Hospital Los Angeles, United States
https://doi.org/10.7554/eLife.101386.3
Share this article
Cite this article
  1. Julia Raulino Lima
  2. Nicha Ungvijanpunya
  3. Qing Chen
  4. Hoang Quoc Hai Pham
  5. Tal Rosen
  6. Greg Park
  7. Mohammadreza Vantankhah
  8. Steven Yen
  9. Yang Chai
  10. Amy E Merrill
  11. Zhaoyang Liu
  12. Jian-Fu Chen
  13. Yanzhong Yang
  14. Weiqun Peng
  15. Jian Xu
(2026)
PRMT1-SFPQ regulates intron retention to control matrix gene expression during craniofacial development
eLife 13:RP101386.
https://doi.org/10.7554/eLife.101386.3
  2. Download BibTeX
  3. Download .RIS

eLife Assessment

This important work establishes a connection between PRMT1 and SFPQ by identifying common phenotypes downstream of their inactivation. In the resubmission, authors now include NMD as a contributor to aberrant gene expression underpinning craniofacial development. The complementary experiments help strengthen some solid conclusions. This paper describes an interesting mechanism for the regulation of RNA levels, which is of interest to the readers of eLife.

https://doi.org/10.7554/eLife.101386.3.sa0

Abstract

Spliceosomopathies, which are a group of disorders caused by defects in the splicing machinery, frequently affect the craniofacial skeleton and limb, but the molecular mechanism underlying this tissue-specific sensitivity remains unclear. Splicing factors and small nuclear ribonucleoproteins (snRNPs) are core components of splicing machinery, and splicing factors are further controlled by post-translational modifications, among which arginine methylation is one of the most prevalent. We determined the splicing mechanisms in the cranial neural crest cells (CNCCs), a multipotent developmental population that gives rise to the majority of the craniofacial skeleton, and focused on an upstream regulator of splicing proteins, protein arginine methyltransferase 1 (PRMT1). PRMT1 is the highest expressing arginine methyltransferase in CNCCs, and its role in craniofacial development is evident from our earlier investigation, where CNCC-specific Prmt1 deletion caused cleft palate and mandibular hypoplasia. PRMT1 catalyzes arginine methylation of splicing factors to modify protein localization, expression, and activity. In the present study, we uncover roles of PRMT1 in the regulation of intron retention, a type of alternative splicing where introns are retained in the mature mRNA. CNCCs from the mandibular primordium of Prmt1-deficient embryos demonstrated an increase in the percentage of intron-retaining mRNA of matrix genes, which triggered nonsense-mediated decay (NMD), causing a reduction in matrix mRNA abundance. We further identified SFPQ as a substrate of PRMT1 that depends on PRMT1 for arginine methylation and protein expression in the developing craniofacial structures. Depletion of SFPQ in CNCCs phenocopied PRMT1 deletion whereby matrix, Wnt signaling components, and neuronal gene transcripts contained higher IR and exhibited lower expression. We further recognized gene length as a common feature among SFPQ-regulated genes in CNCCs. Altogether, these findings demonstrate that the PRMT1-SFPQ pathway modulates matrix gene expression via IR-triggered NMD in CNCCs during craniofacial development.

Introduction

Craniofacial abnormalities affecting bone formation in the skull and face are the most common birth defects in infants. Proper formation of these structures involves coordination of proliferation, migration, and differentiation of cranial neural crest cells (CNCCs) (Martik and Bronner, 2017; Plein et al., 2015). CNCCs are a transient population of progenitor cells that populate the first and second pharyngeal arches and give rise to the facial skeleton including the maxilla, mandible, and palates, and the anterior skull (Chai et al., 2000; Jiang et al., 2002; Martik and Bronner, 2017). Dysregulation of transcription factors, chromatin remodelers, RNA regulatory proteins, and signaling molecules have been implicated in impaired neural crest development that results in craniofacial defects (Bélanger et al., 2018; Martik and Bronner, 2017; Plein et al., 2015; Strobl-Mazzulla et al., 2012). Spliceosomopathies, which are a group of disorders caused by defects in the splicing machinery, frequently affect the craniofacial skeleton and limb. However, the mechanisms behind the sensitivity of these tissues to spliceosomal defects remain incompletely understood. Splicing factors and small nuclear ribonucleoproteins (snRNPs) are core components of splicing machinery, which control pre-mRNA splicing and mRNA maturation through the process of alternative splicing (AS). AS includes seven types of events: exon skipping, mutually exclusive exons, alternative 5′ and 3′-splice sites, alternative promoters, alternative polyadenylation, and intron retention (IR; Hooper et al., 2020). IR, where introns are retained within the mature mRNA sequence, is abundant in plants and fungi (Wang et al., 2008). Although initially overlooked in mammals due to challenges in analysis, recent technical advances have unveiled its significant roles in various aspects, including cell differentiation and stress response, especially in neuronal, immune, and cancer cells (Braunschweig et al., 2014; Marquez et al., 2012; Monteuuis et al., 2019; Wong and Schmitz, 2022).

In exploring the splicing mechanisms in CNCCs, we focused on upstream regulators for splicing factors. The activity of splicing factors is tightly controlled by post-translational modifications, among which methylation is the most abundant, surpassing phosphorylation (Ruta et al., 2021; Thandapani et al., 2013). Splicing regulator methylation mostly occurs on arginine motifs, catalyzed by the protein arginine methyltransferase (PRMT) family of enzymes. These methylation events determine their protein stability, subcellular localization, and splicing activity, shaping the splicing product (Blanc and Richard, 2017; Liu and Dreyfuss, 1995; Snijders et al., 2015). PRMT1 is the most abundant PRMT in most cell types. It specifically catalyzes methylation of arginine (R) residues within RG/RGG/GAR repeats, which are motifs enriched in RNA-binding proteins (RBPs), particularly splicing regulators. This results in asymmetric dimethylation of arginine (ADMA) (Smith et al., 2004; Thandapani et al., 2013). The importance of PRMT1 in craniofacial development is evident from our earlier investigation in which genetic deletion of Prmt1 in CNCCs leads to cleft palate and mandibular hypoplasia (Gou et al., 2018a; Gou et al., 2018b). In the present study, we uncovered a previously unrecognized role of PRMT1 in regulating IR in CNCCs. The mandibular primordium of Prmt1-deficient embryos demonstrated increased retention of introns in mature mRNAs that encode bone and cartilage matrix components. These retained introns contain premature termination codons (PTCs) that trigger nonsense-mediated decay (NMD) to degrade mRNA and reduce gene expression. We further identified SFPQ, EWSR1, and TAF15 as downstream substrates of PRMT1 in the developing craniofacial structures, which depend on PRMT1 for arginine methylation. SFPQ further depends on PRMT1 for protein stability. Depletion of SFPQ in CNCCs partially phenocopied PRMT1 deletion, causing elevated retention of introns in mRNA of matrix, neuronal, and Wnt signaling pathway genes. Matrix and Wnt pathway genes susceptible to perturbation of the PRMT1-SFPQ pathway are further characterized as long genes with a median length of 100 kb. Their retained introns trigger NMD to reduce gene expression. Together, these findings demonstrate that the PRMT1-SFPQ pathway modulates CNCC gene expression via control of splicing. Deficiency of this pathway leads to aberrant IR, which induces mRNA decay that downregulates matrix expression in CNCCs during craniofacial development.

Results

CNCCs from the embryonic mandibular process display abundant IR, which was further elevated by loss of Prmt1

PRMT1 is the most abundant enzyme of the arginine methyltransferase family in CNCCs, with an expression level ~10- to 100-folds higher than other PRMTs (Figure 1A). We previously demonstrated critical roles for PRMT1 in CNCCs during embryogenesis using genetic deletion, whereby neural crest-specific deletion of Prmt1 caused cleft palate and shorter mandibles in mouse (Gou et al., 2018a; Gou et al., 2018b). Shorter mandibles, or mandibular hypoplasia, is a phenotype shared by most spliceosomopathies, a group of conditions caused by splicing defects (Griffin and Saint-Jeannet, 2020; Merkuri and Fish, 2019). To investigate the molecular mechanisms underlying PRMT1’s role in mandible development and splicing regulation, we conducted a genome-wide analysis of transcriptional and splicing alterations within the embryonic mandibles following Prmt1 deletion. For this purpose, CNCCs, the cell lineage that gives rise to bone and cartilage in the mandible, were isolated from mandibular processes of control (Wnt1-Cre; Rosa26LSLtdTomato) or Prmt1 deficient (Prmt1 CKO, Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato) embryos at embryonic day 13.5 (E13.5) for mRNA extraction and sequencing (Figure 1B—source data 1). The ensuing comparative analysis using rMATS (Wang et al., 2024) unveiled changes across five types of AS events (Figure 1C), echoing our earlier finding where PRMT1 regulates exon usage in epicardial cells during cardiac morphogenesis (Jackson-Weaver et al., 2020). In contrast to embryonic epicardial cells, we noted a higher abundance of IR changes in mandibular CNCCs (Figure 1C). The IR changes were exemplified by the increase of IR in Pex12, Mmp23, and Ecm1 (Figure 1D, F), and the decrease of retained intronic expression in Tbx1 (Figure 1E, F). To comprehensively analyze IR at a gene-specific level, we employed IRI (intron reads index), which is an algorithm that quantifies the ratio of normalized read counts between intronic and exonic regions (Ni et al., 2016; Sun et al., 2023; Tian et al., 2020). IRI analysis demonstrated that Prmt1 deficiency did not cause a global shift in IR levels (Figure 1C—source data 3) but changed IR in genes of multiple biological pathways (Figure 1C—source data 2). We noted protein binding, metal ion binding, and cartilage development among genes that present increased IR, while mitochondrion and metabolic pathways were identified among genes with decreased IR (Figure 1C—source data 1). These findings identify IR as a prevalent phenomenon in CNCCs during craniofacial development, and the deletion of Prmt1 further enhanced IR in genes involved in diverse biological processes.

Figure 1
Download asset Open asset
CNCC-specific deletion of Prmt1 elevates intron retention in the embryonic mandibular process.

(A) Expression levels of PRMT1–9 mRNAs in primary isolated cranial neural crest cells (CNCCs) from Wnt1-Cre; Rosa26LSLtdTomato mouse embryo heads at E13.5 and E15.5. TPM, transcript per million. (B) Diagram illustrating the isolation of CNCCs from embryonic mandibles, followed by poly(A)+mRNA isolation and sequencing (n = 4 in control and Prmt1 CKO group). (C) Prmt1 deletion in CNCC caused changes in alternative splicing (AS). Changes in AS events were analyzed by rMATS using RNA sequencing data and significant changes in each type of AS were shown in a stacked bar chart. SE, skipped exon. IR, intron retention. MXE, mutually exclusive exons. A5SS, alternative 5′ splice site. A3SS, alternative 3′ splice site. (D, E) Intron retention was prevalent in CNCCs and altered by Prmt1 deletion. Track view of genes demonstrating intronic and exonic expression with blue boxes indicating exons and blue lines indicating introns. Intron retention was elevated by Prmt1 deletion in Pex12, Mmp23, and Ecm1 (D) and reduced in Tbx1 (E). (F) Quantification of intron expression in Pex12, Mmp23, Ecm1, and Tbx1 by the percentage of intron-retaining mRNAs in each gene, calculated from IRI analysis based on RNA-seq data. *p < 0.05 Prmt1 CKO vs. Control. Control: Wnt1-Cre; Rosa26LSLtdTomato. Prmt1 CKO: Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato.

Figure 1—source data 1

CNCCs labeled by Tdtomato in mouse embryo.

Sagittal sections of Wnt-Cre; Rosa26LSLtdTomato. whole mouse embryo at E13.5 showing the CNCCs (red, labeled by Tdtomato) and nuclei (blue, labeled by DAPI). The boxed fields showed higher magnification of the craniofacial structures including the mesenchyme, where CNCCs are located, and epithelium, labeled by the asterisks. Scale bars = 1.0 mm, 500 μm, and 100 μm, respectively.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig1-data1-v1.pdf
Figure 1—source data 2

GO analysis of genes with differentially regulated intron retention in Prmt1 CKO mandibles.

Excel file attached.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig1-data2-v1.xlsx
Figure 1—source data 3

Prmt1 deletion in CNCCs did not cause a global change in intron retention.

Whole genome IRI value distribution in the control (Cont, pink) and Prmt1 CKO mutant (Mut, blue) embryos was plotted. There is no significant difference between control and mutate intron retention distribution.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig1-data3-v1.pdf

Neural crest-specific deletion of Prmt1 caused a significant reduction of matrix gene expression in the developing mandibles

The main documented role for IR is to reduce expression of the gene harboring retained introns via NMD, the cytosolic RNA surveillance system (Schmitz et al., 2017). When intron-retaining mRNA transcripts are loaded onto the translation machinery, introns that contain PTCs within their open reading frames trigger NMD, leading to mRNA transcript degradation. Reduction in mRNA abundance leads to decreased protein expression for genes harboring these introns (Wong et al., 2016). The intron-retaining transcripts illustrated in Figure 1, Pex12, Mmp23, and Ecm1 demonstrated increased intronic expression and decreased exonic expression, supporting this negative correlation (Figure 1D, F). To systematically analyze genes that are differentially downregulated, we compared the transcription profile between control and Prmt1 CKO mandibular CNCCs and revealed downregulation of 303 genes and concurrent upregulation of 160 genes by Prmt1 deletion (Figure 2A, B). Upon pathway analysis, glycosaminoglycan (GAG) degradation (35 out of 303 genes) and extracellular matrix (ECM) (33 out of 303 genes) emerged as top pathways among downregulated genes (Figure 2C). These GAG degradation and ECM genes encode matrix proteins pivotal to osteogenic/chondrogenic differentiation and the formation of bone and cartilage matrix. Ingenuity Pathway Analysis of downregulated genes further suggested a disruption in bone, cartilage, and connective tissue development within Prmt1 CKO mutant mandibles (Figure 2D). In addition, Prmt1 deficiency upregulated genes in cytokine-mediated signaling pathway and p53 signaling, which are consistent with previous reports that Prmt1 deletion resulted in cytokine production in oral epithelium and p53 accumulation in embryonic epicardium (Jackson-Weaver et al., 2020; Zhang et al., 2018). Prmt1 deletion also upregulated genes involved in adult behavior, postsynaptic membrane organization, and regulated exocytosis, which echoes findings on PRMT1 function in the neuronal lineages and cancer (Chen et al., 2021; Hashimoto et al., 2021a; Hashimoto et al., 2021b; Hashimoto et al., 2022; Xu and Richard, 2021; Figure 2E).

Figure 2
Download asset Open asset
CNCC-specific Prmt1 deletion reduced matrix gene expression in the developing mandibles.

(A) Volcano plot illustrating upregulation of 160 and downregulation of 303 genes in the mandibular primordium of Prmt1-deficient embryos at E13.5, compared to control mandibles. (B) Heatmap showing differential gene expression between control and Prmt1-deficient mandibles. (C) GO analysis of pathway enrichment in downregulated genes demonstrating glycosaminoglycan (GAG) degradation and extracellular matrix (ECM) organization as the top pathways. (D) Ingenuity Pathway Analysis suggesting connective tissue, bone, and cartilage development as affected biological processes based on downregulated genes. (E) GO analysis of pathway enrichment in upregulated genes demonstrating adult behavior and cytokine-mediated signaling and p53 signal transduction as top pathways. Control: Wnt1-Cre; Rosa26LSLtdTomato (n = 4). Prmt1 CKO: Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato (n = 4).

Deletion of Prmt1 in CNCCs elevated IR in mRNAs that encode ECM and GAG degradation enzymes

To determine whether downregulation of matrix genes in Prmt1 CKO embryos is regulated by IR-mediated mechanisms, we first focused on ECM genes downregulated in Prmt1-deficient mandibles and investigated their IR events. The findings are presented in a scatter plot where the red line denotes unchanged IR. A prominent elevation of IR within ECM genes was revealed by the fact that most ECM data points landed above the red line (Figure 3A). ECM genes with the most significant IR increase were labeled and highlighted in red. The IR increase was also exemplified by Adamts16 and Cthrc1 (Figure 3B, C), where track view of the two genes illustrated increased IR and reduced exonic expression (Figure 3Ba, b). To validate the reproducibility of IR elevation in ECM transcripts, we collected four additional embryos from each group, isolated mature mRNA using poly(A)+ magnetic beads, and designed primers for intronic and exonic regions of additional ECM genes with RT-PCR analysis. In line with our initial observations, intronic expression increased significantly, coupled with decreased exonic expression in Dcn, Tnn, Lox, Loxl1, Matn2, Col14a1, Adamts12, Adamts2, Adam12, and Scara5 of the Prmt1 CKO CNCCs as compared to control (Figure 3D). The decrease in exonic expression, representing mRNA abundance, led to a decline in protein expression, as demonstrated by immunostaining for LOXL1 and FBLN5 in CNCCs (Figure 3E–H). These findings demonstrate that ECM transcripts in Prmt1 CKO mandibles exhibited IR elevation and reduction in mRNA expression, leading to lower ECM protein production. Next, we analyzed the IR events in GAG degradation genes, which is the top downregulated gene cluster in the Prmt1 CKO group and clinically associated with craniofacial birth defects (Mizumoto and Yamada, 2021; Paganini et al., 2020; Schwartz and Domowicz, 2002). Similar to ECM genes, GAG degradation genes showed IR elevation in the scatter plot where most data points landed above the red line (Figure 3I). The track view of St6galnac3 and Galn11 further illustrated the increased IR and lower mRNA abundance (Figure 3J, K). Taken together, ECM and GAG degradation genes that are downregulated in Prmt1 CKO embryos demonstrate elevated IR.

Figure 3
Download asset Open asset
PRMT1 regulates intron retention in ECM and GAG degradation genes.

(A–C) Intron retention increased in the majority of ECM gene transcripts that were downregulated in Prmt1-deficient embryos, as illustrated by a scatter plot based on intron retention index (IRI). The red line delineates unchanged levels of intron retention. Genes with the top differential IR were represented by red dots and labeled. ECM gene transcripts Adamts16 and Cthrc1 demonstrating higher intron retention in Prmt1-deleted embryos were illustrated by track view in B and quantified for intronic (left) and exonic (right) expression as shown in C. (D) Higher intron retention and lower mRNA abundance of ECM genes were validated in additional embryo samples by RT-PCR. Primers that span the intronic or intron–exon junction region were used to assess intronic expression. Primers that span the exonic region were used to examine exonic expression that indicated mRNA abundance. Reduced expression of LOXL1 and FBLN5 was examined at the protein level by immunostaining (E, G) and quantified (F, H). Ee and Ef illustrated the plane of section and the region of analysis for E and G. (I–K) Intron retention increased in the majority of GAG degradation gene transcripts that were downregulated in Prmt1 deficiency, as indicated by a scatter plot based on IRI. The red line defines where intron retention is unchanged. Genes were represented by black dots and labeled in red. GAG degradation genes St6galnac3 and Galnt11 demonstrating higher intron retention in Prmt1-deleted embryos were illustrated by track view (J) and quantified for intronic (left) and exonic (right) expression (K). *p < 0.05 Prmt1 CKO vs. Control. Control: Wnt1-Cre; Rosa26LSLtdTomato. Prmt1 CKO: Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato. Scale bar: (E, G) 100 µm.

IR-triggered NMD acts as a physiological and stress-induced mechanism for mRNA decay in CNCCs

We inspected the retained introns in these ECM and GAG degradation transcripts and identified PTCs within their open reading frames, suggesting PTC-triggered NMD as a potential mechanism for transcript decay (Figure 4—source data 1; Wong et al., 2016). To test this hypothesis, we inhibited the NMD machinery using a chemical inhibitor NMDI14, which disrupts the interaction between UPF1 and SMG7 to block NMD complex formation and activity (Martin et al., 2014). To this end, Wnt1-Cre; Prmt1fl/+; Rosa26LSLtdTomato were bred with Prmt1fl/fl; Rosa26LSLtdTomato mice to generate Prmt1 CKO (Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato) and Prmt1 heterozygous embryos as littermate control (Prmt1 Het, Wnt1-Cre; Prmt1fl Rosa26LSLtdTomato). Primary CNCCs from E13.5 Prmt1 CKO and Prmt1 Het were plated and treated with NMDI14 or DMSO as control, followed by poly(A)+ mRNA extraction and RT-PCR analysis using primers that span the intronic or exonic regions. We first validated the inhibition of NMD activity by NMDI14 through analysis of glutathione peroxidase 1 (Gpx1) intron 1, which is a well-documented substrate for NMD-mediated transcript decay (Sun et al., 2000). We further demonstrated that NMDI14 caused a significant increase of Adamts2 intron 19 and Alpl intron 5 expression in the control CNCCs from Prmt1 Het (Figure 4Aa), suggesting that inhibition of NMD blocked IR-triggered mRNA decay, thereby causing accumulation of introns. A more prominent accumulation of introns was induced by NMDI14 in the Prmt1 CKO CNCCs, accompanied by significant increase of mRNA abundance indicated by higher exonic expression levels (Figure 4Ab). These data suggest that Prmt1 deficiency invoked IR-induced NMD for mRNA decay which reduced mRNA abundance.

Figure 4
Download asset Open asset
IR-triggered NMD functions as a basal and stress-responsive mechanism for mRNA decay in CNCCs.

(Aa, Ab) Treatment with the NMD inhibitor NMDI14 (NMDI) led to accumulation of intron-retaining (intron) transcripts of Gpx1, Adamts2, and Alpl in CNCCs from Control (Prmt1 Het) and Prmt1 CKO embryos, analyzed by RT-PCR. Gpx1 serves as a positive control to validate NMD inhibition by NMDI14. (Ba–Bf) NMDI14 caused accumulation of intron-retaining (intron) and total mRNAs (exon) of Adamts2, Alpl, Eln, Matn2, Loxl1, and Bgn in CNCCs. CNCCs were isolated from E13.5 Wnt1-Cre; Rosa26LSLtdTomato and analyzed by RT-PCR. Primers that span the intronic or intron–exon junction region were used to assess intronic expression. Primers that span the exonic region were used to examine exonic expression that indicated mRNA abundance. (Ca, Cd) Intron retention of matrix transcripts Adamts2 and Fbln5 was detected in four independent control embryos by RNA-seq, as illustrated by track views. (Cb–Cf) NMDI14 treatment caused accumulation of intron-retaining Adamts2 and Fbln5 transcripts. CNCCs isolated from E13.5 Wnt1-Cre; Rosa26LSLtdTomato embryos were treated by DMSO or NMDI14 (NMDI), followed by mRNA extraction and assessment with semi-quantitative PCR. Primers (indicated by the red arrows) were designed to span regions (red line) of intron 19 of Adamts2 or intron 7 of Fbln5. *p < 0.05 NMDI vs. DMSO of the same group.

Figure 4—source data 1

Number of premature termination codons (PTCs) in the retained introns of ECM and GAG degradation transcripts of Prmt1 CKO group.

Excel file attached.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig4-data1-v1.xlsx
Figure 4—source data 2

PDF file containing original gel blot for Figure 4C, indicating the relevant genes and bands.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig4-data2-v1.zip
Figure 4—source data 3

Original files for gel blot analysis displayed in Figure 4C.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig4-data3-v1.zip

We noted significant accumulation of retained introns in control CNCCs from Prmt1 Het (Figure 4Aa), suggesting a possibility that IR-induced NMD is a basal or physiological mechanism to regulate mRNA levels in CNCCs. To assess this possibility, we isolated CNCCs from control Wnt1-Cre; Rosa26LSLtdTomato embryos at E13.5 and treated cells with NMDI14 or DMSO, followed by poly(A)+ mRNA extraction and RT-PCR analysis. Multiple ECM genes were assessed for the role of NMD, which revealed significant accumulation of retained introns within Adamts2, Alpl, Eln, Matn2, Loxl1, and Bgn, accompanied by increased mRNA abundance following NMDI14 treatment (Figure 4Ba–f). To strengthen these findings, we used traditional semi-quantitative PCR to detect a longer span (600–1000 bp) of the intronic region. In the NMDI14-treated group, accumulation of Adamts2 intron 19 and Fbln5 intron 7 were revealed by higher intensity bands compared to DMSO group (Figure 4Ca-f and Figure 4—source data 2 and 3). Overall, these findings demonstrate that the NMD process is responsible for the decay of ECM transcripts. More importantly, we defined IR-triggered NMD as a physiological mechanism in CNCCs to control ECM mRNA abundance and gene expression during craniofacial development. Upon molecular stress caused by Prmt1 deletion, NMD is further exploited to disrupt ECM expression.

PRMT1 methylates splicing factors SFPQ, EWSR1, and TAF15 in CNCCs

Next, we investigated the molecular mechanism by which Prmt1 deletion disrupts IR. PRMT1 is a methyltransferase that catalyzes methylation of arginine (R) residues within RG/RGG/GAR repeats. These repeats are highly enriched in RBPs, particularly splicing regulators (Thandapani et al., 2013). PRMT1 generates asymmetric dimethylation (ADMA) on arginine to regulate the turnover, subcellular localization, and activity of splicing regulators (Smith et al., 2004). We therefore hypothesized that PRMT1 regulates the retention of introns via methylation of splicing regulators. Earlier studies by Graham et al. and our team have characterized the landscape of arginine methylation governed by PRMT1 and identified a cohort of RNA processing proteins exhibiting PRMT1-dependent methylation (Hartel et al., 2019). Among these, six splicing regulators are highly expressed in embryonic CNCCs: SFPQ, SRSF1, EWSR1, TAF15, TRA2B, and G3BP1 (Figure 5A). Given that pre-mRNA splicing occurs in the nucleus following transcription, we examined the subcellular localization of these splicing factors in control CNCCs from E13.5 mouse embryos. SFPQ, EWSR1, and TAF15 predominantly localized to the nucleus (Figure 5Ba–l). TRA2B was distributed in both the nucleus and cytoplasm (Figure 5Bm–p). While SRSF1 and G3BP1 were primarily expressed in the cytoplasm (Figure 5Bq–x). These observations were corroborated by quantitative analysis of the nuclear to cytoplasmic ratio of their expression (Figure 5C), suggesting nuclear proteins SFPQ, EWSR1, TAF15, and TRA2B as prime candidates to mediate PRMT1-controlled splicing activity. All four splicing factors are documented substrates for PRMT1-catalyzed methylation in Hela, Jeg3, or HEK293 epithelial cells (Hartel et al., 2019; Jobert et al., 2009; Li et al., 2018; Pahlich et al., 2005; Snijders et al., 2015). To determine whether PRMT1 is responsible for methylation of these splicing factors in the craniofacial structures, we employed proximity ligation assay (PLA), a technique designed for in situ detection of protein modifications. To this end, E13.5 embryo sections were probed with anti-SFPQ antibody and anti-ADMA antibody that detects methyl-arginine using the PLA kit, which revealed the presence of methylated SFPQ as distinct green punctate staining. Within the craniofacial complex, SFPQ methylation was observed in the mandibular processes at E13.5 (Figure 5Da–c, E). In Prmt1 CKO embryos, the signal of methyl-SFPQ markedly decreased within the CNCC-derived mesenchyme region (Figure 5Dd–f, E). The same analysis was conducted in the maxillary regions, which also demonstrated robust SFPQ methylation in the control group and a dramatic reduction in Prmt1-deleted embryos (Figure 5Fa–f, G). This CNCC-specific deletion of PRMT1 only reduced SFPQ methylation in the CNCCs, as non-CNCC lineages within the epithelium continued to display methyl-SFPQ signals (Figure 5—source data 1). To detect EWSR1 methylation, E13.5 embryo sections were probed with anti-EWSR1 antibody and anti-ADMA antibody using PLA kit. A low level of EWSR1 methylation was observed in both mandibular and maxilla processes at E13.5 (Figure 5H, I). Using the same approach, we also detected TRA2B methylation in mandibular and maxilla processes at E13.5 (Figure 5J, K), but the levels of EWSR1 and TRA2B methylation were much lower compared to SFPQ. In contrast to the low levels of methylation in craniofacial structures, the embryonic abdominal tissue showed prominent methylation of EWSR1 and TRA2B, validating the PLA techniques and EWSR1/TRA2B antibodies in efficient detection of their methylation (Figure 5—source data 1). We further determined whether methylation of EWSR1 and TRA2B depends on PRMT1 by comparing methylation signal between control and Prmt1 deficient embryos and detected a significant reduction of methyl-EWSR1 and methyl-TRA2B within the CNCC-derived mesenchyme region in Prmt1 CKO samples (Figure 5I–K). TAF15 methylation was not assessed due to lack of a compatible antibody. Collectively, these data indicate that SFPQ, EWSR1, and TRA2B are methylated by PRMT1 in the CNCCs at E13.5 within the craniofacial complex, with SFPQ showing the most prominent methylation and in a PRMT1-dependent manner.

Figure 5
Download asset Open asset
PRMT1 methylates SFPQ, EWSR1, and TRA2B in CNCCs.

(A) Expression levels of splicing factors SFPQ, SRSF1, EWSR1, TAF15, TRA2B, HnRNPA1, WDR70, and G3BP1 in primary isolated CNCCs from Wnt1-Cre; Rosa26LSLtdTomato mouse embryonic heads at E13.5 and E15.5. TPM, transcript per million. (B, C) Subcellular localization of splicing factors in CNCCs by immunostaining using sagittal sections of the mandibular process from wild-type mouse embryos, which revealed nuclear expression of SFPQ (a–d), EWSR1 (e–h), and TAF15 (i–l); even distribution of TRA2B between nucleus and cytosolic compartment (m–p), and cytoplasmic expression of SRSF1 (q–t) and G3BP1 (u–x). The subcellular distribution was quantified in C and presented as nuclear to cytosolic signal ratio (Nuc/Cyto Ratio). SFPQ methylation diminished in the mandibular (D, E) and maxillary (F, G) processes of Prmt1 deficient embryos at E13.5. (H–K) Reduction of EWSR1 and TRA2B methylation was observed in the mandibular processes of Prmt1 deficient embryos at E13.5. Methylation was detected by proximity ligation assay (PLA). Green puncta indicated methyl-SFPQ, TRA2B, or EWSR1. Nuclei were counterstained with DAPI (blue). Representative images are shown for Control (Da–Dc; Fa–Fc; Ha–Hc; Ja–Jc) and Prmt1 CKO (Dd–Df; Fd–Ff; Hd–Hf; Jd–Jf). Higher magnification views in (Dc, Df, Fc, Ff, Hc, Hf, Jc, Jf) illustrate methyl-SFPQ, TRA2B, and EWSR1 (green puncta) in the nuclei, as indicated by white arrows. (E, G, I, and K) showed quantification of PLA puncta normalized to cell number in four biological replicates, presented as mean ± SEM. *p < 0.05 Prmt1 CKO vs. Control. Control: Wnt1-Cre; Rosa26LSLtdTomato. Prmt1 CKO: Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato. Scale bar = 100 µm in B, D, F, H, and J except in enlarged panels, where scale bar = 25 µm (Bd, Bh, Bl, Bp, Bt, Bx, Dc, Df, Fc, Ff, Hc, Hf, Jc, Jf).

Figure 5—source data 1

SFPQ, EWSR1, TAF15, and TRA2B methylation in control and Prmt1 CKO embryos.

(A) SFPQ methylation signal remained robust in the epithelial region of craniofacial structures in both control (Aa, Ab) and CNCC-specific Prmt1 deletion (Ac, Ad) embryos. (B) EWSR1 methylation signal was robust in the abdominal region of control (Ba, Bb) and Prmt1 CKO (Bc, Bd) embryos. (C) TRA2B methylation signal was robust in the abdominal region of control (Ca, Cb) and Prmt1 CKO (Cc, Cd) embryos. Control: Wnt1-Cre; Rosa26LSLtdTomato. Prmt1 CKO: Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato. Scale bars = 25 µm in A, B, C.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig5-data1-v1.pdf

PRMT1-catalyzed methylation protects SFPQ from proteasomal degradation

PRMT1-induced methylation controls the activity of splicing factor by modifying their stability, subcellular distribution, or RNA binding affinity (Rho et al., 2007). We first focused on SFPQ and examined whether SFPQ protein expression or subcellular localization was disturbed by Prmt1 deletion. In control embryos, SFPQ exhibited prominent expression in CNCCs and mainly localized in the nucleus within the mandibular and maxillary processes (Figures 5Ba–d, C–6Aa–c, Ca–c). In Prmt1-deficient CNCCs, SFPQ protein expression level declined significantly in the maxilla and mandible (Figure 6A–D). A corroborating western blot (WB) analysis using tissues from the craniofacial structures, including facial and anterior skull regions of embryonic heads, confirmed this decline in SFPQ expression (Figure 6E,F and Figure 6—source data 2 and 3). These data suggest two possible mechanisms: PRMT1 promotes SFPQ expression or protects SFPQ from degradation. We first analyzed SFPQ mRNA levels in control and Prmt1-deficient CNCCs, which showed no discernable difference (Figure 6G), suggesting that loss of PRMT1 did not reduce SFPQ mRNA expression. Next, we tested whether PRMT1 regulates SFPQ degradation and assessed the proteasome-mediated degradation by treating CNCCs with MG132, a proteasome inhibitor. MG132 treatment in control CNCCs from Prmt1 Het embryos did not alter the expression of SFPQ, but in the Prmt1 CKO CNCCs, MG132 caused a significant restoration of SFPQ protein (Figure 6H, I), suggesting that SFPQ protein in Prmt1-deficient CNCCs was reduced by proteasomal degradation. To test the relationship between arginine methylation and SFPQ degradation, we assessed SFPQ methylation in these MG132-treated CNCCs. SFPQ proteins rescued by MG132 in Prmt1-deficient CNCCs was not methylated (Figure 6J, K), suggesting that PRMT1-catalyzed arginine methylation may protect SFPQ from proteasomal degradation in CNCCs.

Figure 6
Download asset Open asset
PRMT1 depletion reduces SFPQ protein levels via proteasomal degradation.

SFPQ protein level was significantly reduced in the mandibular (A, B) and maxillary (C, D) processes of Prmt1 deficient embryos. SFPQ protein was detected by immunostaining and quantified in B and D. SFPQ protein levels declined dramatically in the Prmt1 deficient embryonic head, as detected by western blotting (E) and quantified with ImageJ (F). (G) PRMT1 deletion did not alter SFPQ mRNA levels in CNCC. (H–K) SFPQ protein accumulated in Prmt1-deficient CNCCs upon MG132 treatment, and the accumulated SFPQ protein remained un-methylated. CNCCs isolated from Prmt1 CKO embryos and littermate controls (Prmt1 heterozygous) were treated with MG132. SFPQ protein was detected by immunostaining in H and quantified in I. SFPQ methylation was detected by PLA in J and quantified in K.* p < 0.05 Prmt1 CKO Control. Control, Wnt1-Cre; Rosa26LSLtdTomato. Prmt1 CKO, Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato. Scale bar = 100 µm in A and C. Scale bar = 25 µm in H and J.

Figure 6—source data 1

SFPQ, EWSR1, TAF15, and TRA2B protein expression and subcellular localization in control and Prmt1 CKO embryos.

EWSR1, TAF15, and TRA2B protein expression and subcellular localization were not altered in the mandibular processes of Prmt1 deficient embryos. The level of EWSR1 (A), TAF15 (C), and TRA2B (E) protein was detected by immunostaining in the embryonic mandible of control and Prmt1 deficient embryos. The protein expression level was quantified in B, D, and F. Subcellular localization was quantified in G–I. Control: Wnt1-Cre; Rosa26LSLtdTomato. Prmt1 CKO: Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato. Scale bars = 100 μm in Aa–e, Ca–e, Ea–e. Scale bars = 25 μm in Ac, Af, Cc, Cf, Ec, Ef.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig6-data1-v1.pdf
Figure 6—source data 2

PDF file containing original western blots for Figure 6E, indicating the relevant bands and treatments.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig6-data2-v1.zip
Figure 6—source data 3

Original files for western blot analysis displayed in Figure 6E.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig6-data3-v1.zip

We further examined whether protein expression of EWSR1, TAF15, and TRA2B was disturbed by Prmt1 deletion in the craniofacial complex. Unlike SFPQ, EWSR1, TAF15, and TRA2B showed a similar level of protein expression in Prmt1 CKO and control embryos (Figure 6—source data 1). These findings were supported by WB analysis using tissues from the craniofacial structures (Figure 6—source data 2 and 3). The subcellular distribution of EWSR1, TAF15, and TRA2B was not altered either (Figure 6—source data 1). In summary, loss of Prmt1 reduced SFPQ protein expression via proteasomal degradation, whereas the protein expression and subcellular localization of EWSR1, TAF15, and TRA2B were not altered by Prmt1 deficiency.

SFPQ regulates the splicing of matrix, Wnt signaling, and neuronal genes in CNCCs

We further investigated the role of SFPQ in CNCCs. Primary CNCCs were isolated from control (Wnt1-Cre; Rosa26LSLtdTomato) embryos at E13.5 and enriched by Td+ signal using cell sorting. Within a day, Td+ CNCCs were transfected with control siRNA or two independent SFPQ-targeting siRNAs to deplete SFPQ, and then poly(A)+ mRNA was purified followed by sequencing (Figure 7A). To determine whether these primary isolated CNCCs retained neural crest cell characteristics after siRNA-mediated knockdown, the expression of CNCC markers Twist1, Sox10, Msx1, Snai2, and Tfap2a was examined (Achilleos and Trainor, 2012; Ishii et al., 2012). The CNCC marker expression in transfected CNCCs after isolation is comparable to fresh CNCCs from E13.5 or E15.5 embryos (Figure 7—source data 1), validating CNCC identity and supporting this method as a feasible approach for mechanistic study of neural crest cell biology using developmental stage-specific embryos. Subsequently, using sequencing data from these siRNA-transfected CNCCs which exhibit about 50% reduction of Sfpq (Figure 7A), we conducted bioinformatic analyses of differentially expressed genes (DEGs) and differentially retained introns (IRI) between control and Sfpq-depleted CNCCs. First, we confirmed that the transcriptomic and intronic changes induced by the two independent siRNAs targeting SFPQ are predominantly overlapping, as illustrated by scatter plot analysis of the DEG and IRI data (Figure 7Ba, b). Next, we focused on changes in retained introns caused by Sfpq depletion and revealed higher IR in 397 (siSFPQ #1 vs. siCont) and 388 (siSFPQ #2 vs. siCont) genes and lower IR in 129 (siSFPQ #1 vs. siCont) and 151 (siSFPQ #2 vs. siCont) genes (Figure 7Ca, b). Genes with higher IR also bear significant overlap with downregulated genes. Among 269 (siSFPQ #1 vs. siCont) and 232 (siSFPQ #2 vs. siCont) genes downregulated by Sfpq depletion, 81 (siSFPQ #1 vs. siCont) and 60 (siSFPQ #2 vs. siCont) genes overlapped with elevated IR, respectively (Figure 7Cc, d). Sfpq depletion also led to downregulation of 269 (siSFPQ #1 vs. siCont) and 232 (siSFPQ #2 vs. siCont) genes, among which 81 (siSFPQ #1 vs. siCont) and 60 (siSFPQ #2 vs. siCont) genes overlapped with elevated IR, respectively (Figure 7Cc, d). Genes showing higher IR in Sfpq depleted CNCCs encompassed ‘proteoglycan metabolic process’ and ‘proteoglycan biosynthesis process’, which included GAG degradation genes, and ‘connective tissue development’, which includes ECM genes (Figure 7Da, b). We also analyzed genes that were differentially expressed in control vs. Sfpq-depleted CNCCs (Figure 7E). GO analysis of differentially downregulated genes by SFPQ-specific siRNAs further revealed ECM and extracellular structure as the top pathways suppressed by SFPQ depletion, suggesting that IR-mediated regulation of ECM is a key functional consequence following Sfpq deficiency (Figure 7F). SFPQ has demonstrated direct roles in transcriptional regulation, where it facilitates RNA polymerase II (Pol II) recruitment and transcriptional termination in neuronal lineages (Takeuchi et al., 2018). To determine whether SFPQ directly regulates the transcription of matrix genes, we performed Cleavage Under Target and Tagmentation (CUT&Tag) with antibody against Rbp1, the largest subunit of Pol II, in ST2 cells, which is a mesenchymal cell line used for mechanistic studies of osteogenic differentiation (Canales et al., 2023; Ishida et al., 2021; Li et al., 2019; Mizukami et al., 2023; Pregizer et al., 2007; Seong et al., 2023; Strauss et al., 2019; Tu et al., 2007; Yang et al., 2019). ST2 cells were transfected with control or SFPQ-targeting siRNAs and collected for CUT&Tag followed by analyses of Pol II binding at the promoter region (3 kb ±TSS) and gene body of matrix genes. Pol II recruitment was similar between control and SFPQ-depleted groups with marginal differences (Figure 7G, H), suggesting that transcriptional repression cannot fully account for the downregulation of matrix gene in SFPQ depleted groups.

Figure 7
Download asset Open asset
PRMT1-SFPQ pathway regulates matrix genes in CNCCs.

(A) SFPQ knockdown caused around 50% reduction in Sfpq expression in CNCCs. CNCCs were transfected with two independent siRNAs targeting SFPQ, or control siRNA, followed by poly(A)+ mRNA extraction and RNA sequencing. *p < 0.01 siSFPQ vs. siControl. (B) The two independent siRNAs targeting SFPQ caused similar transcriptomic and intronic changes, shown by scatter plot of DEG (Ba) and IRI (Bb). DEG, differentially expressed genes. IRI, intron retention index. (C) SFPQ depletion altered intron retention in CNCCs. Changes of IR events in genes showing increased (Up, yellow color) or decreased (Down, orange color) IR were illustrated by stacked bar graphs. Pie chart demonstrated differentially regulated genes, with shaded areas among downregulated genes (orange) highlighting their overlap with IR elevated genes (red). (D) GO analysis of genes with elevated IR following SFPQ depletion. (E, F) Heatmap and GO analysis of SFPQ-regulated genes in CNCC. (G, H) Pol II CUT&Tag analysis in ST2 cells transfected with control or SFPQ siRNAs showing Pol II recruitment in downregulated and upregulated genes (G), promoter regions (Ha), and gene body (Hb). p-value was indicated at the top. siControl, control siRNA. siSFPQ#1 and siSFPQ#2, two independent SFPQ siRNAs.

Figure 7—source data 1

CNCC marker expression in siRNA transfected CNCCs compared to fresh isolated CNCCs from mouse embryos.

Bar chart depicting the CNCC’s marker expression in three different conditions. Each condition is presented as mean ± SEM.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig7-data1-v1.pdf

More intriguingly, Wnt signaling emerged as the top pathway regulated by SFPQ in CNCCs (Figure 7Da, b). Sfpq depletion increased the percentage of intron-retaining mRNAs of genes in this cluster and reduced total mRNA expression (Figure 8A). The Wnt signaling components-regulated by SFPQ encode both positive and negative regulators of Wnt signaling and encompass both canonical and non-canonical Wnt signal regulators. Additionally, neuronal genes constitute a big fraction among SFPQ-regulated introns as ‘axon guidance’, ‘axonogenesis’, ‘neuron projection development’, and ‘learning or memory’ emerged as the top biological processes among differentially upregulated introns and SFPQ-regulated genes, and ‘regulation of membrane potential’, ‘learning or memory’, ‘cognition’, and ‘locomotory behavior’ denotes differentially downregulated genes (Figure 7F). In CNCCs, Sfpq depletion increased the percentage of intron-retaining mRNAs of these neuronal genes and reduced total mRNA expression (Figure 8B). Altogether, these findings demonstrate that depletion of SFPQ enhanced IR and reduced mRNA expression in matrix, Wnt signaling components, and neuronal genes.

Figure 8
Download asset Open asset
SFPQ regulates intron retention of Wnt signaling and neuronal genes in CNCCs.

SFPQ depletion in CNCCs elevated intron retention and decreased mRNA abundance of Wnt signaling components (A) and neuronal genes (B). The levels of mRNA abundance (left) and intron retention (right) were illustrated by a two-sided bar graph. TPM, transcripts per million. *p < 0.01 siSFPQ vs. siControl. siControl, control siRNA. siSFPQ#1 and siSFPQ#2, two independent SFPQ siRNAs.

SFPQ depletion reduces matrix and Wnt signaling gene expression via IR-induced NMD

In Sfpq-depleted CNCCs, retained introns also contained PTCs within their open reading frames, suggesting the involvement of PTC-triggered NMD (Figure 9—source data 1). To investigate the role of IR-induced NMD, we utilized ST2 cells. SFPQ knockdown in ST2 cells caused an increase in retained introns and a decrease in mRNA expression for Col4a2, St6galnac3, and Ptk7 (Figure 9A). To determine whether Col4a2, St6galnac3, and Ptk7 mRNA were degraded through IR-induced NMD, we inhibited NMD with NMDI14 (Martin et al., 2014). NMDI14 treatment caused accumulation of intron 4 of Col4a2, intron 1 of St6galnac3, and intron 1 of Ptk7 (Figure 9B), indicating that the decay of intron-retaining transcripts in the Sfpq-depleted cells depends on the NMD machinery. To assess the impact of NMD on the abundance of these mRNAs, we designed PCR primers to detect the exonic regions and observed that inhibition of NMD restored and increased the abundance of these mRNAs in Sfpq-depleted ST2 cells (Figure 9C). These data indicate that depletion of Sfpq causes aberrant IR that triggers mRNA decay through NMD. We further noted that NMDI14 induced accumulation of intron 4 of Col4a2 and intron 1 of St6galnac3 in the control group (Figure 9B), accompanied by accumulation of their transcripts (Figure 9C), suggesting that NMD is also a basal mechanism to regulate gene expression in ST2 cells, a function similar to that in embryonic CNCCs.

Figure 9
Download asset Open asset
SFPQ depletion reduces long gene expression via intron retention triggered NMD.

(A) SFPQ depletion promoted intron retention and reduced mRNA abundance of Col4a2, St6galnac3, and Ptk7 in ST2 cells. Bar chart showing RT-PCR analysis of intronic and exonic expression. (B, C) NMD inhibitor NMDI14 caused the accumulation of retained introns and total mRNAs of Col4a2, St6galnac3, and Ptk7 in ST2 cells. Bar chart showing RT-PCR analysis of intronic (B) and exonic expression (C) in DMSO or NMDI-treated cells. NMDI14-mediated inhibition of NMD was validated using Gpx1 as a positive control. *p < 0.05 siSFPQ vs. siControl. #p < 0.05 NMDI vs. DMSO treatment of the same group. (D) SFPQ binding peaks were mapped to retained intron 1 but not spliced intron 6 of Ptk7. Peak distribution from published Sfpq CLIP-seq data using E13.5 brain (top) and track view of RNA-seq data using siControl or siSFPQ-transfected CNCCs (bottom) for Ptk7, with the retained Ptk7 intron 1 (red box) and spliced intron 6 (green box) highlighted. (E) SFPQ binding peaks were significantly enriched in retained intron regions within CNCCs. Violin plot displaying the density of SFPQ binding peaks in introns with elevated retention compared to introns with reduced retention or no change. p-value was calculated using Mann–Whitney U test and indicated at the top. (F) SFPQ binding peaks were preferentially enriched in genes with higher intron retention when compared to genes with no IR change. p = 0.07 using Fisher’s exact test. (G) SFPQ-regulated genes were significantly longer than average. Violin plot displaying the distribution of length for genes showing increased intron retention (IR Up), decreased intron retention (IR Down), or unchanged intron retention (No Change). Median length of the IR Up group was highlighted with solid blue line. Median length of the IR Down and No Change groups was highlighted with dotted blue line. p-value was calculated using Mann–Whitney U test and indicated at the top.

Figure 9—source data 1

Number of premature termination codons (PTCs) in the retained introns of matrix transcripts of Sfpq knockdown group.

Excel file attached.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig9-data1-v1.xlsx

Disturbance of SFPQ activity impaired splicing of long genes and long introns

SFPQ predominantly binds to intronic regions and regulates splicing (Hosokawa et al., 2019). We used published CLIP-seq dataset from embryonic mouse brain to assess whether these aberrantly retained introns are SFPQ targets (Hosokawa et al., 2019). As illustrated in Ptk7 sequence, the retained intron 1 exhibited a much higher density of SFPQ binding peaks when compared to the constitutively spliced intron 6 within the same gene (Figure 9D). We further conducted analysis for all introns at the genomic level by comparing SFPQ binding peaks among introns with higher retention to introns with unaltered or lower retention. The comparison demonstrated that introns with elevated retention in Sfpq-depleted CNCCs showed significantly higher enrichment of SFPQ binding peaks (Figure 9E). We also analyzed SFPQ CLIP-seq peaks at the gene level, comparing genes with elevated IR to genes with no IR changes and showed that genes containing elevated IR in Sfpq-depleted CNCCs exhibited a trend toward higher enrichment of SFPQ binding peaks based on this embryonic brain dataset (Figure 9F). These data suggest that introns with aberrant IR elevation in Sfpq-depleted CNCCs are splicing targets bound by SFPQ.

To understand how SFPQ targets these matrix, Wnt and neuronal genes and introns, we examined molecular characteristics shared among SFPQ-regulated genes and recognized that they are long genes (>100 kd in length) and retain long introns (>10 kd in length). Of note, Wnt pathway gene Wwox and matrix genes St6galnac3 are 913 and 526 kb in length, and their retained introns are 639 and 215 kb in length, respectively, in contrast to a median size of 0.6–2.4 kb for introns in the mouse genome (Hong et al., 2006). In the human and mouse genome, long genes mostly occur because of long introns (Breschi et al., 2017; Lopes et al., 2021). We then plotted the length of SFPQ-regulated genes against genes across the genome and demonstrated that the genes with increased IR in Sfpq-depleted CNCCs have a median length of ~100 kb. This is much longer than gene length in the control groups, where the medial length is around 10 kb (Figure 9G). These findings suggest that SFPQ-regulated splicing facilitates regulation of long gene expression during craniofacial development.

SFPQ, EWSR1, TAF15, and TRA2B regulate distinct transcriptional and splicing events

To assess genes and IR co-regulated by SFPQ and PRMT1, we cross-analyzed genes with IR in both Sfpq-depleted CNCCs and Prmt1-deficient CNCCs. In Sfpq-depleted CNCCs, 179 genes exhibited significantly higher IR in both siRNA groups. In Prmt1-deficient CNCCs, 773 genes exhibited significantly higher IR in all embryos analyzed (Figure 10A). The cross-analysis showed that SFPQ-regulated IR bears partial (64 out of 773) overlap with PRMT1-regulated introns, exemplified by matrix genes Col4a2, Adam12, Ntn1, App, St6galnac3, Galnt11, Galnt10, and Asph (Figure 10A–E). Wnt signaling genes Wwox and Dkks, and neuronal genes that regulate membrane potential (38 out of 303 genes) were also identified among genes with decreased expression in Prmt1 CKO CNCCs, suggesting the regulation of these genes by the PRMT1-SFPQ pathway. The overlap may be under-represented because SFPQ was reduced to 50% by the depletion, but the partial overlay does suggest a notion that PRMT1 regulates IR through multiple splicing factors.

Figure 10
Download asset Open asset
SFPQ, EWSR1, TAF15, and TRA2B regulate distinct transcriptional and splicing programs.

(A) Genes with increased IR events in SFPQ-depleted CNCCs demonstrated 8.28% (64 out of 773) overlap with Prmt1-deficient CNCCs. (B, E) Matrix genes Col4a2, Adam12, Ntn1, App, St6Galnac3, Galnt10, and Asph were regulated by both Prmt1 deletion and SFPQ depletion. Bar graph showing elevated IR (B, C) and reduced mRNA abundance (D, E) in CNCCs. TPM, transcripts per million. *p < 0.05 siSFPQ vs. siControl. *p < 0.05 Prmt1 CKO vs. Control. Control: Wnt1-Cre; Rosa26LSLtdTomato. Prmt1 CKO: Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato. (F–S) ST2 cells were transfected with control or SFPQ, EWSR1, TAF15, TRA2B siRNAs and mature mRNA was extracted for sequencing. GSEA demonstrated enrichment of ECM genes among downregulated genes upon depletion of SFPQ, EWSR1, TRA2B, and TAF15 (F–I). Pie charts showed the percentage of genes with increased (dark gray) or decreased IRI (light gray) upon SFPQ, EWSR1, TAF15, and TRA2B deletion (J–M). rMATS analysis showed widespread and significant splicing changes following depletion of SFPQ, EWSR1, TAF15, and TRA2B (N). Postn mRNA expression was decreased in all four depletion groups (O). Exon skipping events of Postn were noted in all four depletion groups, illustrated by Sashimi plots (P) and quantified by bar graphs based on rMATS analysis (Q–S). *p < 0.05 siRNAs vs. siControl. siControl, control siRNA. siSFPQ#1 and siSFPQ#2, two independent SFPQ siRNAs.

Figure 10—source data 1

Transcriptional landscape altered by the knockdown of EWSR1, TAF15, and TRA2B.

ST2 cells were transfected with control siRNA or siRNAs targeting EWSR1, TAF15, or TRA2B, followed by mRNA isolation, sequencing, and bioinformatic analysis for GO analysis of upregulated (6A, 6C, 6E) and downregulated (6B, 6D, 6F) genes.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig10-data1-v1.pdf
Figure 10—source data 2

SFPQ motifs in the vicinity of differential alternative splicing events of CNCCs.

rMAPS2 output pages depicting the spatial distribution of SFPQ motifs for mutually exclusive exons (MXE) (A), exon skipping (SE) (B), intron retention (RI) (C), alternative 5′ splice site (A5SS) (D), and alternative 3′ splice site (A3SS) (E) events. The results demonstrate high motif scores paired with low p-values for SFPQ, underscoring its significant role in alternative splicing regulation. The red line represents the enriched motif for enhanced exons, the blue line represents the enriched motif for silenced exons, and the black line represents the motif density for background (nonregulated) exons. Solid lines represent the peak quality Motif score (peak height) as scaled on the left. The dotted lines represent the negative log10 (p-value) as scaled on the right. The green box indicates the cassette exon.

https://cdn.elifesciences.org/articles/101386/elife-101386-fig10-data2-v1.pdf

We further characterized gene expression and IR regulated by SFPQ, EWSR1, TAF15, and TRA2B using ST2 cells. To this end, cells were transfected with control siRNA or two independent siRNAs targeting each splicing factor, and then poly(A)+ mRNA was purified followed by sequencing. GSEA analysis demonstrated that depletion of SFPQ, EWSR1, TRA2B, and TAF15 each caused downregulation of ECM genes (Figure 10F–I), suggesting that the four splicing factors all promote ECM gene expression. Global analysis for DEGs further showed that these splicing regulators control distinct biological functions (Figure 10—source data 1). We also analyzed IR following their knockdown. siSFPQ caused differential IR in 514 genes, with increased IR in 171 genes and decreased IR in 370 genes (Figure 10J—source data 1). siEWSR1 caused increased IR in 58 genes and decreased IR in 318 genes (Figure 10K), where genes with elevated IR belonged to diverse biological functions (Figure 10—source data 1). In contrast, siTAF15 and siTRA2B caused predominantly decrease in IR with IR increase in <3% of genes with differential IR (Figure 10L, M—source data 1). These findings suggest that the roles of EWSR1, TAF15, and TRA2B are distinct from SFPQ in the regulation of IR. Bioinformatic analysis for EWSR1, TAF15, and TRA2B did not reveal discernible overlap between ECM genes and IR changes, suggesting that they regulate ECM gene expression through IR-independent mechanisms in ST2 cells. Examination of other types of AS events by rMATS analysis (multivariant analysis of transcript splicing, MATS) revealed many splicing changes (Figure 10N, Supplementary file 1). For example, the ECM gene Postn is known to exhibit exon skipping at exon 17 and 21, which affects Postn expression and activity that alter mandibular morphogenesis (Ishihara et al., 2023). We observed that the mRNA expression of Postn is reduced in all four splicing factor depletion groups and exon skipping of Postn showed a trend toward enhancement by the depletion, as illustrated by Sashimi plots and quantified from rMATS analysis (Figure 10O–S).

To identify additional splicing regulators accountable for IR changes in Prmt1 deficient embryos, we conducted in silico analysis using RNA-seq data of mandibular CNCCs from control and Prmt1 CKO embryos by performing rMAPS2 (RNA map analysis) subsequent to rMATS (Hwang et al., 2020; Park et al., 2016; Wang et al., 2024). A list of RBPs was identified with highly significant binding scores in the differentially expressed genes, suggesting them as potential downstream mediators for PRMT1-regulated splicing changes. This in silico approach predicted additional splicing regulators that may function downstream of PRMT1 in elevating IR, including SRSF9, Lin28A, PABPN1, ESRP2, TARDBP, KHDRBS1, and FXR1 (Figure 10, Supplementary file 2). SFPQ is also among the RBPs with top occurrences, with SFPQ binding motifs depicted in the vicinity of AS events (Figure 10—source data 2, Supplementary file 3). Taken together, PRMT1-regulated splicing in CNCCs is mediated by a plexus of splicing factors including SFPQ.

Discussion

In this study, we revealed prevalent IR within CNCCs during embryonic development and demonstrated that the PRMT1-SFPQ pathway regulates the splicing of matrix and Wnt pathway genes during craniofacial development. Genetic deletion of Prmt1 or depletion of Sfpq in CNCCs caused aberrant IR that triggers NMD to degrade intron-retaining transcripts. To our knowledge, this study is the first to characterize the functional significance of IR-induced NMD in CNCCs and during craniofacial development. A main feature shared by SFPQ-regulated genes in CNCCs is gene length, with a median length of 100 kb, in contrast to the median length of ~10 kb across the genome. Introns exhibiting high retention also tend to be long introns, for example, intron 4 of Col4a2 which displayed dramatically increased retention in Sfpq deficiency boasts a length of 45 kb. Long genes with long introns are linked to neuronal function, where retained introns have been proposed to facilitate a rapid response to external stimuli, allowing a time frame shorter than that required for de novo transcription (Mauger et al., 2016; Ni et al., 2016). Genes over 100 kb in length require two to several hours for transcription based on a rate of 1–2 kb/min, which poses a challenge for the fast tempo of embryogenesis. This intron-regulated mechanism may represent a physiological strategy that enables rapid production of matrix proteins when they are required for craniofacial morphogenesis. This post-transcriptional regulation allows cells to control the rate of transcript production and to achieve rapid protein translation without waiting for the time period required for transcription. Long genes regulated by this mechanism are highly enriched in matrix formation, responsible for the production of building blocks for bone, cartilage, and connective tissues. Since spliceosomopathies preferentially affect the craniofacial skeleton and limb, IR-regulated matrix expression provides mechanistic insights for the higher susceptibility of these tissue types to spliceosome dysfunction.

Many of the PRMT1-SFPQ pathway regulated matrix and Wnt signaling genes are associated with congenital defects. For example, in our study, the long gene Wwox exhibited increased retention of long introns (introns 3 and 4) and decreased expression upon Prmt1 or Sfpq deficiency. In human patients, pathogenic variants of WWOX with large deletions within the long introns have been associated with epileptic encephalopathy syndrome manifesting shared facial phenotype (Dvinge and Bradley, 2015). The deletion within these long introns spanning the SFPQ binding region may affect SFPQ recognition and subsequent splicing of WWOX transcripts. Variants of another SFPQ target, Ptk7, are associated with neural tube defects and scoliosis (Berger et al., 2017). SFPQ itself is also linked to neurodegenerative diseases including amyotrophic lateral sclerosis, frontotemporal dementia, and Alzheimer’s disease, and was recently identified as a genetic regulator in CNCCs through an integrated genomic analysis (Feng et al., 2021). Besides SFPQ-regulated targets, splicing factors we identified in the unbiased in silico analysis, such as Rbfox2 and HuR (Figure 10, Supplementary file 2), are also associated with craniofacial defects in human patients and mouse studies (Cibi et al., 2019; Glessner et al., 2014; Homsy et al., 2015; McKean et al., 2016; Verma et al., 2022). We propose that IR-induced downregulation acts as a causal mechanism for dysregulated gene expression that contributes to craniofacial deformity.

IR-triggered NMD in controlling the expression of matrix genes in CNCCs presented a distinct mechanism in CNCCs. NMD is the RNA turnover pathway to degrade aberrant RNAs, driven by a protein complex composed of SMG and UPF proteins (Han et al., 2018; Tan et al., 2022). IR-triggered NMD in cancer has both promotive and suppressive roles, and NMD inhibitors have been tested for cancer therapy and recently cancer immunotherapy (Tan et al., 2022). During embryonic development, the functional significance of NMD machinery is implied by human genetic findings and mouse genetic models. SMG9 mutation in human patients causes malformation in the face, hand, heart, and brain (Shaheen et al., 2016). Smg6, Upf1, Upf2, and Upf3a knockout mice die at early embryonic stages (E5.5–E9.5), and Smg1 gene trap mutant mice die at E12.5, all presenting severe developmental defects (Han et al., 2018). In this study, upon Prmt1 or Sfpq deficiency, IR-induced NMD adopted a pathogenic role, degrading aberrant intron-retaining transcripts to reduce gene expression. We also demonstrated a physiological role for IR-triggered NMD that tunes matrix gene expression during embryogenesis, as inhibition of NMD using chemical inhibitors led to accumulation of transcripts such as Alpl, Eln, and Loxl1.

In mammalian cells, PRMT1 methylates over forty splicing factors to modulate protein stability, subcellular localization, thereby influencing their accessibility to mRNA (Thandapani et al., 2013; Wu et al., 2022). PRMT1-catalyzed arginine methylation also affects their RNA binding affinity and specificity, altering their activity and the splicing product (Fedoriw et al., 2019; Fong et al., 2019). Therefore, it is not surprising that the disruption of this methyltransferase leads to splicing defects. We recognized SFPQ as a key PRMT1 substrate in CNCCs and demonstrated abundant SFPQ methylation in the mandibular and maxillary primordium. Our previous work also revealed SFPQ methylation on arginine (Arg) 7 and 9 depends on PRMT1 (Hartel et al., 2019). This aligns with previous work demonstrating that PRMT1 is responsible for the association of SFPQ with mRNA in mRNP complexes in mammalian cells (Snijders et al., 2015). Our findings on altered neuronal genes in Sfpq-depleted CNCCs also echoed previous reports in which SFPQ regulates many neuronal genes involved in axon extension, branching, viability, and synaptogenesis via direct binding to their intronic regions (Cosker et al., 2016; Ruskin et al., 1988). Loss of SFPQ function compromises splicing patterns, especially the accurate splicing of long introns in embryonic brain and AML samples (Luisier et al., 2018; Taylor et al., 2022; Thomas-Jinu et al., 2017). Findings in our study further revealed previously unappreciated roles of SFPQ in the regulation of matrix genes through IR, which bear functional significance in bone and cartilage matrix deposition and craniofacial skeleton formation. Manipulation of SFPQ expression (knockdown with 50% efficiency) demonstrated that SFPQ shares around a partial overlap with PRMT1 in IR-regulated neural crest gene expression. Given that PRMT1 methylates over 40 splicing factors in mammals and many of these splicing factors are abundantly expressed in embryonic CNCCs, we expect that PRMT1-regulated IR will be mediated by multiple splicing regulators besides SFPQ. This notion is further supported by the identification of additional splicing regulators via in silico analysis. Additionally, PRMT1-governed splicing will be mediated by multiple RBPs encompassing all types of AS, as suggested by PRMT1-mediated methylation of EWSR1 and TRA2B, and the distinct splicing footprints among EWSR1, TAF15, TRA2B, and SFPQ.

Altogether, these findings demonstrate that the PRMT1-SFPQ pathway regulates IR in CNCCs and suggests IR-triggered NMD as a mechanism that controls matrix and Wnt signaling gene expression during craniofacial morphogenesis.

Materials and methods

Key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Gene (M. musculus)Prmt1GenBankNC_000080.7N/A
Strain, strain background (M. musculus)Prmt1fl/+Yu et al., 2009N/AN/A
Strain, strain background (M. musculus)Wnt1-CreJackson Laboratory009107N/A
Strain, strain background (M. musculus)R26RtdTomatoJackson Laboratory007914N/A
Cell line (M. musculus)ST2Cell Engineering DivisionRCB0224N/A
Transfected construct (M. musculus)siRNA to SFPQQIAGENSI0578384816 µl
Transfected construct (M. musculus)siRNA to SFPQQIAGENSI057838766 µl
Biological sample (M. musculus)Primary CNCCThis paperN/AN/A
Biological sample (M. musculus)Mouse embryosThis paperN/AN/A
AntibodyGoat polyclonal anti-rabbit Alexa Fluor 488InvitrogenA110701/500
AntibodyRabbit monoclonal mix anti-AsymmetricCell Signaling135221/100
AntibodyGoat polyclonal anti-SFPQEverest BiotechEB095231/100
AntibodyRabbit polyclonal anti-SFPQCell Signaling23020S1/100
AntibodyPRMT1 antibodyCell Signaling Technology24491/100
AntibodySFPQ antibodyEverest BiotechEB095231/100
AntibodySFPQ antibodyCell Signaling Technology230201/100
AntibodyAsymmetric di-methyl arginine antibody (ADMA)Cell Signaling Technology135221/100
AntibodyTRA2B antibodyGeneTexGTX1147521/100
AntibodyTAF15 antibodyThermo Fisher Scientific8TA-2B101/100
AntibodyG3BP1 antibodyThermo Fisher ScientificMA5-574061/100
AntibodySRSF1 antibodyThermo Fisher ScientificMA5635181/100
AntibodyEWSR1 antibodyAbcamab1332881/100
AntibodyGAPDH antibodyCell Signaling Technology971661/5000
Commercial assay or kitDuolink in Situ kitMillipore SigmaDUO92101-1KTN/A
Chemical compound, drugNMDI14MedChemExpressHY-11137450 µg/ml
Chemical compound, drugLipofectamine RNAiMAXThermo Fisher137785009 µl
Chemical compound, drug Antigen Unmasking SolutionsVector LaboratoriesH-3300–250N/A
Software, algorithmFigures-Illustrator CC 2017.1.1Adobehttp://www.adobe.com/cn/N/A
Software, algorithm Image Analysis-CellProfiler2Kamentsky et al., 2011https://cellprofiler.orgN/A
Software, algorithmRstudioR Foundation for Statistical Computinghttp://www.rstudio.com/N/A
Software, algorithmAcquisition- analysis-Keyence BZ-X800KEYENCEhttps://www.keyence.comN/A
Software, algorithmAcquisition- analysis-Leica DMI 3000BLeicahttp://www.leica-microsystems.comN/A
Software, algorithmFastQC v0.11.8Babraham Institutehttps://www.bioinformatics.babraham.ac.uk/projects/fastqc/N/A
Software, algorithmTrim GaloreBabraham Institutehttps://www.bioinformatics.babraham.ac.uk/projects/trim_galore/N/A
Software, algorithmhisat2 (v2.1.0)Kim et al., 2019PMID:31375807 PMCID:PMC7605509N/A
Software, algorithmsamtools (v1.7)Li et al., 2009PMID:19505943 PMCID:PMC2723002N/A
Software, algorithmhtseq-count (v1.99.2)Anders et al., 2015PMID:25260700 PMCID:PMC4287950N/A
Software, algorithmIRToolsPeng, 2018https://github.com/WeiqunPengLab/IRTools/N/A
Software, algorithmEdgeRRobinson et al., 2010PMID:19910308
PMCID:PMC2796818		N/A
Software, algorithmDAVIDSherman et al., 2022https://david.ncifcrf.gov/tools.jspN/A
Software, algorithmMetascapeZhou et al., 2019https://metascape.org/gp/index.html#/main/step1N/A
Software, algorithmrMATS
version 4.0 (turbo)
Xinglabhttps://github.com/Xinglab/rmats-turbo
RRID:SCR_023485		N/A
Software, algorithmrMAPS2rMAPS2https://rmaps.cecsresearch.org/N/A
Chemical compound, drugHuman/Mouse/Rat BMP-2Gibco12002C250UGN/A
Chemical compound, drugMG-132MedChem ExpressHY-13259N/A
Chemical compound, drugDMSOSigma-AldrichD4540N/A
Commercial assay or kitPeroxidase AffiniPure Goat Anti-Mouse IgG, light chain specificJackson ImmunoResearch115-035-174N/A
Commercial assay or kitMaxima H Minus cDNA Synthesis Master MixThermo Fisher ScientificM1681N/A
Commercial assay or kitPowerUp SYBR Green Master Mix for qPCRApplied BiosystemsA25742N/A
Chemical compound, drugFormalinMillipore SigmaHT501128N/A
Chemical compound, drugRIPA Lysis and Extraction BufferThermo Fisher Scientific89900N/A
Commercial assay or kitPierce ECL Western Blotting SubstrateThermo Fisher Scientific32106N/A
Commercial assay or kitDynaMag-2 MagnetInvitrogen12321DN/A
Commercial assay or kitNEBNext Poly(A) mRNA Magnetic Isolation ModuleNew England BiolabsE7490LN/A

Animals

Prmt1fl/fl mice were generously provided by Dr. Stéphane Richard (McGill University). Wnt1-Cre; Prmt1fl/fl;Rosa26LSLtdTomato and Wnt1-Cre;Rosa26LSLtdTomato mice were obtained by mating Wnt1-Cre, Rosa26LSLtdTomato mice (Jax #009107 and #007914) with Prmt1fl/fl mice. Animals were genotyped using established protocols (Chai et al., 2000; Yu et al., 2009), and all care and experiments followed USC’s IACUC protocols.

Proximity ligation assay

Request a detailed protocol

PLA was performed using a Duolink In Situ kit (Millipore Sigma Cat# DUO92101-1KT), SFPQ antibody (Everest Biotech Cat# EB09523), and pan-methyl arginine antibody, ADMA (Cell Signaling Technology Cat# 13522) to detect methyl-SFPQ. Tissue sections were blocked with the Duolink blocking solution in a humidity chamber for 1 hr at 37°C before incubating with the primary antibodies (diluted 1:50 in Duolink PLA probe diluent) overnight at 4°C. The next day, tissue sections were incubated with the Goat-PLUS and Rabbit-MINUS PLA probes for 1 hr at 37°C, followed by ligase and amplification solution. After the final washes, tissue sections were mounted with Duolink In Situ mounting media with DAPI to counterstain nuclei. Images were visualized under the confocal microscope. The number of PLA signals that appear as punctuated green signals in the nuclei of the cells, and the quantification was performed with CellProfile software.

Tissue processing and immunofluorescence staining

Request a detailed protocol

E13.5 control and Wnt1-Cre; Prmt1fl/fl mice embryos were fixed in 4% PFA overnight, dehydrated, embedded in O.C.T., sectioned at 8 µm thickness in sagittal orientation, and mounted on glass slides. The samples were submitted to antigen retrieval using Antigen Unmasking Solutions (Vector Laboratories Cat# H-3300-250) and washed and incubated in 0.5% Triton X-100 in PBS for 20 min for permeabilization. The samples were briefly washed with PBS, blocked with 10% goat serum for 1 hr, and then incubated with primary antibody SFPQ (Cell Signaling Technology Cat#23020, 1:100) at 4°C overnight in a moisture chamber. After this, the samples were blocked with a secondary antibody (1:400, goat anti-rabbit IgG (H+L) Alexa Fluor 488 Cat# A-11008) for 1 hr followed by DAPI diluted 1:1000 in PBS for 10 min. Finally, the samples were mounted with mounting media (Electron Microscopy Science Cat# 1798510) and covered with cover glass for imaging using Keyence BZ-X800 and Leica DMI 3000B microscopes and quantification with CellProfile software.

Western blotting

Request a detailed protocol

Heads of E13.5 control and Wnt1-Cre; Prmt1fl/fl mice were dissected, followed by removal of the brain. The tissue was homogenized with a pestle followed by lysis with RIPA Lysis and Extraction Buffer. The total protein concentration was determined by comparison with BSA standards. Twenty micrograms of total protein from each sample were loaded into each well of a 10% polyacrylamide gel. WB analysis was carried out as previously described (Zhang et al., 2018). Antibodies against PRMT1 (Cell Signaling, Cat# 2449; 1:1000), SFPQ (Cell Signaling Technology Cat#23020, 1:1000), and GAPDH (Cell Signaling, Cat# 97166; 1:5000) were used for WB. The samples were analyzed three times in independent blots and the bands were quantified by optical densitometry. The analysis was performed using ImageJ digital imaging processing software (ImageJ 1.48v, National Institutes of Health, Bethesda, MD, USA). The expression of each analyzed protein was normalized with GAPDH.

NCC collection and FACS analysis

Request a detailed protocol

The heads of E13.5 Wnt1-Cre; Rosa26LSLtdTomato and Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato mice were dissected and placed in a sterile tube containing Hanks’ Balanced Salt Solution (Thermo Fisher Scientific Cat# 14175079). Following centrifugation, the supernatant was discarded, and the tissue was subjected to TrypLE (Thermo Fisher Scientific Cat# 12605010) incubation for 5–15 min at 37°C on a rotator. The resulting dissociated tissue was neutralized by the addition of fetal bovine serum (FBS) and then filtered through a 40-μm strainer to obtain a single-cell suspension. After centrifugation, cells were resuspended in an appropriate volume of serum-free medium for FACS analysis. Sorting was performed based on tdTomato fluorescence (Excitation/emission: 554/582 nm), and td-positive cells were collected into separate tubes containing DMEM (Genesee Scientific Cat# 25-501) supplemented with 20% FBS.

Primary CNCC culture, ST2 cells culture, and siRNA transfection

Request a detailed protocol

CNCCs were cultured in DMEM supplemented with 20% FBS in an incubator until they attach to the bottom. ST2 cells, purchased from RIKEN Cell Bank RCB0224, authenticated with STR profiling and tested for mycoplasma contamination every 6 months, were cultured in RPMI 1640 supplemented with 10% FBS. Then, reverse transfection with control siRNA (QIAGEN Cat# 1027310), SFPQ siRNA #1 (QIAGEN Cat# SI05783848), and SFPQ siRNA #2 (QIAGEN Cat# SI05783876) at 40 nM using Lipofectamine RNAiMax transfection reagent (Invitrogen) for siRNA delivery were performed. 48 hr later, total RNA was isolated using TRIzol reagent (Invitrogen) following the manufacturer’s protocols followed by mRNA isolation using NEBNext High-Input Poly(A) mRNA Magnetic Isolation Module (NEBNext Cat# E3370S).

mRNA isolation and sequencing

Request a detailed protocol

Poly(A) mRNA isolation was extracted from total RNA by using the NEBNext High-Input Poly(A) mRNA Magnetic Isolation Module (NEBNext Cat# E3370S). Five sets of primary isolated CNCCs from control (Wnt1-Cre; Rosa26LSLtdTomato) or Prmt1 CKO (Wnt1-Cre; Prmt1fl/fl; Rosa26LSLtdTomato), and three sets of isolated CNCCs transfected with siControl or siSFPQs were sequenced at 40 million reads sequencing depth and 150 bp paired-end sequencing.

Bioinformatic analysis of differential gene expression (DEGs) and differential IR

Request a detailed protocol

The sequencing quality of RNA-seq libraries was assessed by FastQC v0.11.8 (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Low-quality bases and adapters were trimmed using Trim Galore (v0.6.7, here). The reads were mapped to mouse genome mm10 using hisat2 (v2.1.0) (Kim et al., 2019). The mapped sam files from hisat2 were converted to bam files which were then further turned to sorted bam files by samtools (v1.7) (Li et al., 2009). Mapped reads were then processed by htseq-count (v1.99.2) to calculate the read count of all genes (Anders et al., 2015). The sorted bam files were then given to IRTools (https://github.com/WeiqunPengLab/IRTools/ copy archived at WeiqunPengLab, 2026) to calculate the IR of each gene and the read count of genes’ constitutive intronic regions (CIRs) and constitutive exonic regions (CERs). The expression level of a gene was expressed as TPM (Transcripts Per Kilobase Million) value. EdgeR was used to identify differentially expressed genes (DEGs) by requiring ≥1.5-fold expression changes and adjusted p-value <0.05 (Robinson et al., 2010). The intron retention index (IRI) of a gene is defined as the ratio of the overall read density of its CIRs to the overall read density of its CERs. Genes with 0 < IRI < 1 and CER expression greater than 1 were selected to identify differential IR genes. Student’s t-test was used to identify differential IR genes by requiring ≥1.5-fold log2(IR) changes and p-value <0.05. Gene Ontology analysis was done by DAVID (https://david.ncifcrf.gov/tools.jsp) and Metascape (https://metascape.org/gp/index.html#/main/step1) (Sherman et al., 2022; Zhou et al., 2019). The RNA-seq data are deposited to the Gene Expression Omnibus (GEO) database with accession number GSE266474.

rMATS and rMAPS analysis

Request a detailed protocol

rMATS (https://github.com/Xinglab/rmats-turbo copy archived at Kutschera et al., 2026) and rMAPS2 (http://rmaps.cecsresearch.org/) were used to find potential RBPs that contribute to differential splicing events (Hwang et al., 2020; Wang et al., 2024). rMATS was first used to identify differential AS events between wild type and Prmt1 CKO samples by requiring FDR <0.05. The output of rMATs were then fed to rMAPS2 to identify significant RBPs contributing to each type of splicing events.

Enrichment analysis of SFPQ binding

Request a detailed protocol

To assess the SFPQ binding of the genes whose IR is regulated by SFPQ, we downloaded SFPQ CLIP-seq peak data from GEO (GSE96080). The CLIP-seq peaks shared between the replicates were filtered by p < 0.01 and log(FC) >1 to obtain the significant peaks for downstream analysis. Next, we defined the SFPQ-regulated IR up and down gene sets by combining those from two SFPQ knockdown conditions, respectively. We constructed the control gene set by selecting those with no significant change in IR (IR difference <0.0005, 0.9 < IR FC < 1.1, and p > 0.5) in either SFPQ knockdown conditions. For each gene set, the percentage of genes with introns overlapping SFPQ CLIP-seq peaks was calculated. Furthermore, for each gene, we evaluated the SFPQ peak density as the number of SFPQ peaks on the gene divided by the gene length.

Semi-quantitative and quantitative PCR

Request a detailed protocol

For both experiments, Control vs. Wnt1-Cre; Prmt1fl/fl mandibles, siRNA-transfected CNCCs and ST2, mRNAs were quantified by real-time PCR with IQ Sybr Green Supermix (Bio-Rad) and normalized against Gapdh mRNA levels. Relative changes in expression were calculated using the ΔΔCt method. ST2 cells are purchased from RIKEN, authenticated with STR profiling, and tested for mycoplasma contamination every 6 months. Primer sequences are listed in Supplementary file 4.

Cleavage Under Target and Tagmentation

Request a detailed protocol

CUT&Tag was adapted from in Epicypher DIY protocol and Kaya-Okur, H.S., Janssens, D.H., Henikoff, J.G. et al. Efficient low-cost chromatin profiling with CUT&Tag. Nat Protoc 15, 3264–3283 (2020). https://doi.org/10.1038/s41596-020-0373-x. Before extracted nuclei, 10 µl of ConA beads was resuspended twice in Bead Activation Buffer (20 mM HEPES, pH 7.9; 10 mM KCl; 1 mM CaCl2; 1 mM MnCl2). 100,000 cells were washed once with PBS and nuclei were extracted by resuspending cells in nuclear extraction (NE) buffer (20 mM HEPES-KOH, pH 7.9, 20 mM KCl, 0.1% Triton X-100, 20% Glycerol, 0.5 mM Spermidine, and protease inhibitor) and incubating for 10 min on ice. Nuclei were collected by centrifuge at 600 × g for 3 min and resuspended in 100 µl of NE buffer per 100,000 nuclei. 100 µl of nuclei were transferred to PCR tube containing 10 µl of activated ConA beads, gently vortexed, and incubated at room temp for 10 min. The supernatant was discarded and 50 µl of Antibody buffer (20 mM HEPES, pH 7.5; 150 mM NaCl; 0.5 mM Spermidine; 0.01% Digitonin; 2 mM ETDA; and protease inhibitor) was added. An antibody against RNA Pol subunit 1 (Rbp1 CTD (4H8) mouse mAbm Cell Signal Tech.) was added and incubated on a nutator at 4°C overnight. Beads were then incubated in 50 µl of 10 µg/ml of secondary antibody made in Digitonin150 buffer (20 mM HEPES, pH 7.5; 150 mM NaCl; 0.5 mM Spermidine; 0.01% Digitonin; and protease inhibitor) at room temperature for 30 min on a nutator. Wash twice with Digitonin150 buffer before resuspending beads in pAG-Tn5 diluted in Digitonin300 buffer (20 mM HEPES, pH 7.5; 300 mM NaCl; 0.5 mM Spermidine; 0.01% Digitonin; and protease inhibitor) and incubating for 1 hr at room temperature on a nutator. Wash twice with Digitonin300 buffer then start the tagmentation reaction by resuspending beads in Tagmentation buffer (Digitonin300 supplemented with 10 mM MgCl2) and incubated at 37°C in thermocycler. Beads were washed once with 50 µl TAPS Buffer (10 mM TAPS pH 8.5; 0.2 mM EDTA) then incubated in 50 µl of Release Buffer (10 mM TAPS pH 8.5; 0.1% SDS; 0.4 mg/µl Proteinase K; 15 mM EDTA) at 55°C for 1 hr; 75°C for 20 min; and then cool down to 20°C. Tagmented DNA were purified with 1.8x AMPureXP beads and eluted with 20 µl of DNase/RNase free H2O. The library was prepared with dual index primers and NEBNext High-Fidelity 2X PCR master mix (M0541) and amplified for 14 cycles. Library DNA was purified with 1.3x AMPureXP beads and eluted with 15 µl 0.1× TE buffer. Sequencing run was performed on the NovaSeq X Series platform with 10B reagent kit (300 Cycle). NovaSeq Control Software 1.2.2.48004 was used for sequencing and Illumina BCL Convert v4.2.7 to convert base call (BCL) files into FASTQ files.

Statistical analysis

Request a detailed protocol

Two-tailed Student’s t-tests or Fisher’s exact tests were applied for statistical analysis. For all graphs, error bars represent standard deviations. A p-value of <0.05 was considered statistically significant.

Data availability

RNA sequence data were deposited in GEO (accession number GSE266474). All data generated or analyzed during this study are included in the manuscript and supporting files; source data files have been provided for all figures.

The following data sets were generated
    1. Lima JR
    2. Chen Q
    3. Peng W
    4. Xu J
    (2026) NCBI Gene Expression Omnibus
    ID GSE266474. PRMT1-SFPQ regulates intron retention to control matrix gene expression during craniofacial development.
    https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE266474

References

Article and author information

Author details

  1. Julia Raulino Lima

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing - review and editing
    Contributed equally with
    Nicha Ungvijanpunya and Qing Chen
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-6824-4904

  2. Nicha Ungvijanpunya

    1. Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    2. Department of Orthodontics, Faculty of Dentistry, Chulalongkorn University, Bangkok, Thailand
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology
    Contributed equally with
    Julia Raulino Lima and Qing Chen
    Competing interests
    No competing interests declared

  3. Qing Chen

    Department of Physics, The George Washington University, Washington, DC, United States
    Contribution
    Resources, Data curation, Software, Formal analysis, Writing – original draft
    Contributed equally with
    Julia Raulino Lima and Nicha Ungvijanpunya
    Competing interests
    No competing interests declared

  4. Hoang Quoc Hai Pham

    Department of Cancer Genetics and Epigenetics, City of Hope, Duarte, United States
    Contribution
    Data curation, Formal analysis, Investigation
    Competing interests
    No competing interests declared

  5. Tal Rosen

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Formal analysis, Funding acquisition, Validation, Investigation
    Competing interests
    No competing interests declared

  6. Greg Park

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Formal analysis, Validation, Investigation
    Competing interests
    No competing interests declared

  7. Mohammadreza Vantankhah

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Formal analysis, Visualization, Methodology
    Competing interests
    No competing interests declared

  8. Steven Yen

    1. Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    2. Children’s Hospital Los Angeles, Los Angeles, United States
    Contribution
    Formal analysis, Methodology
    Competing interests
    No competing interests declared

  9. Yang Chai

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Resources, Funding acquisition, Project administration
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-2477-7247

  10. Amy E Merrill

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Resources, Funding acquisition, Project administration
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-3660-1575

  11. Zhaoyang Liu

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Formal analysis, Funding acquisition
    Competing interests
    No competing interests declared

  12. Jian-Fu Chen

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Resources, Funding acquisition, Project administration
    Competing interests
    No competing interests declared

  13. Yanzhong Yang

    Department of Cancer Genetics and Epigenetics, City of Hope, Duarte, United States
    Contribution
    Formal analysis, Supervision, Investigation
    Competing interests
    No competing interests declared

  14. Weiqun Peng

    Department of Physics, The George Washington University, Washington, DC, United States
    Contribution
    Conceptualization, Resources, Data curation, Software, Supervision, Investigation, Writing – original draft
    For correspondence
    wpeng@gwu.edu
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-5521-9091

  15. Jian Xu

    Center for Craniofacial Molecular Biology, Department of Biomedical Sciences, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, United States
    Contribution
    Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing – original draft, Writing - review and editing
    For correspondence
    xujian@usc.edu
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-8162-889X

Funding

National Institute of Dental and Craniofacial Research (R01DE028943)

  • Jian Xu

National Institute of Dental and Craniofacial Research (U01DE022937)

  • Steven Yen

Anandamahidol foundation (scholarship)

  • Nicha Ungvijanpunya

National Institute of Arthritis and Musculoskeletal and Skin Diseases (R01AR083966)

  • Zhaoyang Liu

National Institute of Dental and Craniofacial Research (T90)

  • Tal Rosen

The funders had no role in study design, data collection, and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank the Flow Cytometry Facility of USC Stem Cell at the University of Southern California. This research was supported by NIH NIDCR R01DE028943 (to JX), NIDCR U01DE022937 (to SY), and NIAM R01AR083966 (to ZL), NIDCR T90 grant (to TR), and Anandamahidol Foundation Scholarship (to NU).

Ethics

This study was performed in strict accordance with the institute guidelines for the ethical and humane use of animals for biomedical research at the University of Southern California. The University's Institutional Animal Care and Use Committee makes specific recommendations regarding experimental protocols to ensure ethical and humane treatment of all vertebrate animals involved in research experimentation. All of the animals were handled according to approved institutional animal care and use committee (IACUC) protocols (#20959).

Version history

  1. Preprint posted:
  2. Sent for peer review:
  3. Reviewed Preprint version 1:
  4. Reviewed Preprint version 2:
  5. Version of Record published:

Cite all versions

You can cite all versions using the DOI https://doi.org/10.7554/eLife.101386. This DOI represents all versions, and will always resolve to the latest one.

Copyright

© 2024, Lima, Ungvijanpunya, Chen et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,114
    views
  • 50
    downloads
  • 0
    citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Julia Raulino Lima
  2. Nicha Ungvijanpunya
  3. Qing Chen
  4. Hoang Quoc Hai Pham
  5. Tal Rosen
  6. Greg Park
  7. Mohammadreza Vantankhah
  8. Steven Yen
  9. Yang Chai
  10. Amy E Merrill
  11. Zhaoyang Liu
  12. Jian-Fu Chen
  13. Yanzhong Yang
  14. Weiqun Peng
  15. Jian Xu
(2026)
PRMT1-SFPQ regulates intron retention to control matrix gene expression during craniofacial development
eLife 13:RP101386.
https://doi.org/10.7554/eLife.101386.3

Share this article

https://doi.org/10.7554/eLife.101386

Categories and tags

Research organism