The alternative splicing program of differentiated smooth muscle cells involves concerted non-productive splicing of post-transcriptional regulators

Abstract

Alternative splicing (AS) is a key component of gene expression programs that drive cellular differentiation. Smooth muscle cells (SMCs) are important in the function of a number of physiological systems; however, investigation of SMC AS has been restricted to a handful of events. We profiled transcriptome changes in mouse de-differentiating SMCs and observed changes in hundreds of AS events. Exons included in differentiated cells were characterized by particularly weak splice sites and by upstream binding sites for Polypyrimidine Tract Binding protein (PTBP1). Consistent with this, knockdown experiments showed that that PTBP1 represses many smooth muscle specific exons. We also observed coordinated splicing changes predicted to downregulate the expression of core components of U1 and U2 snRNPs, splicing regulators and other post-transcriptional factors in differentiated cells. The levels of cognate proteins were lower or similar in differentiated compared to undifferentiated cells. However, levels of snRNAs did not follow the expression of splicing proteins, and in the case of U1 snRNP we saw reciprocal changes in the levels of U1 snRNA and U1 snRNP proteins. Our results suggest that the AS program in differentiated SMCs is orchestrated by the combined influence of auxiliary RNA binding proteins, such as PTBP1, along with altered activity and stoichiometry of the core splicing machinery.

Figure 1.

Splice sensitive array profiling of dedifferentiating smooth muscle cells (SMCs). (A) Overall experimental design. RNA was isolated from differentiated contractile SMCs of mouse aorta and bladder, and from SMCs that had been cultured for 7 days. RNA was used to prepare target for hybridization to splice-sensitive microarrays, followed by prediction of transcript level and splicing changes by ASPIRE. (B) qRT-PCR analysis of changes in marker gene expression between differentiated (dark) and proliferative (light) aorta (red) and bladder (blue). (C) Distribution of regulated ASE types between differentiated and proliferative aorta (top left) and bladder (top right) SMCs, and of all ASEs represented on the array (bottom). For this analysis events annotated on the array as ‘promoter’ and ‘bleeding promoter’ were combined, as were ‘terminal’ and ‘bleeding terminal’. (D) Venn diagram comparing direction of changes in CE splicing in aorta and bladder SMCs. The vast majority of CEs are coregulated in the two tissues.

Figure 2.

Validation of cassette and mutually exclusive splicing events. Selected cassette exons (CEs) with increased inclusion in differentiated (A) or proliferative cells (B) and mutually exclusive exon events (C) were validated by RT-PCR for both aorta and bladder. Values shown are mean ± sd (n = 3) of percentage of exon inclusion. In panel C, the % inclusion of the ‘differentiated exon is indicated by the red asterisk; for Actn1 the differentiated exon is smaller; for Itga7 and Tpm1 the differentiated exon amplicon is larger. For every primer pair, a no RT control and a no template control were run in parallel; no signal was detected in any of the reactions (not shown). D = differentiated, P = proliferative phenotype.

Figure 3.

Properties of SMC regulated CEs and retained introns. (A) Schematic of the experimental sets of exons and introns. Top: CEs. Differentiated exons represented in dark blue, proliferative exons in green. Bottom: regulated intron retention (IR reg), more retained in differentiated cells. (B) Length distributions of CEs. (C) Length distributions of introns. (D) Intron GC content (%). (E) Splice site strength (Maximum Entropy). Dn = donor or 5′ splice site, Ac—acceptor or 3′ splice site. (F) AG dinucleotide exclusion zone (AGEZ) (G) Polypyrimidine tract score associated with best predicted branch point. Datasets analyzed in the figure: regulated CEs in proliferative (PCE) and differentiated (DCE) cells, control CEs unregulated in SMCs (CE ctrl), constitutive introns (CIs), retained introns regulated in SMCs (IR reg), non-regulated retained introns (IR ctrl). DCE us—intron upstream of differentiated CE, DCE ds—intron downstream of DCE, PCE us and PCE ds—introns upstream and downstream respectively of PCE, CE ctrl us and ds—introns upstream and downstream respectively of control CEs unregulated in SMCs. Statistical tests described in ‘Materials and Methods’ section, and significant differences are described in accompanying text. The vertical dashed line in each panel represents the median value from the CI set.

Figure 4.

Sequence motif enrichments reveal a role for PTB in regulating SMC CEs. (A) Motif enrichment analysis. The diagram indicates k-mer and RNA-compete motifs enriched in the indicated regions associated with differentiated (top) or proliferative (bottom) exons. All motifs were significantly enriched (P < 0.01). The five motifs marked with an asterisk also passed a FDR test < 0.05. Numbers adjacent to motifs indicate log₂ fold enrichment. (B) PTBP1 and PTBP2 were knockdown in rat PAC1 cells and its effects assessed in SMC CE from Figure 2. (C) Effect of PTBP1/2 knockdown in rat PAC1 cells on Atp2b4 minigene, containing the regulated exon with 225 nt of upstream intron and 275 nt of downstream intron, cloned into an GFP exon trapping vector. (D) Effect of overexpression of STAR family proteins on Atp2b4 minigene in rat PAC1 cells. Histograms show mean and standard deviation of the mean of at least three samples. Statistically significance was calculated using Student's t-test, and is shown *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 5.

IR in splicing factor genes in differentiated SMCs. (A) Distribution of dIrank values of CE (left) and IR (right) events. dIrank indicates the degree of confidence in the ASPIRE predicted change in inclusion. Positive values indicate increased inclusion in differentiated cells (D versus P comparisons, aorta left, bladder right) or in differentiated bladder compared with differentiated aorta (middle). CEs show an even distribution of positive and negative dIrank values in all comparisons, while IR events are skewed toward positive values in D versus P comparisons. (B–D) qRT-PCR validation of IR events. Top, schematic representation of the event (in gray) with arrows showing primer position, and red ‘stop signs’ denoting premature termination codons (PTCs). Bottom, plots shown are average values of relative fold change for IR or intron spliced (IS) normalized against the geometric mean of 5 genes not changing in the microarray (see ‘Materials and Methods’ section), except for CLK1 that was normalized against its own gene expression. Dark red represents Differentiated (D) Aorta; light red Proliferative (P) Aorta. Error bars represent standard deviation of the mean (n = 3). (B) simple IR events with higher retention in D samples. (C) Srsf1 intron 5, showing higher retention in P samples. (D) Dual IR in Rbm3. Statistically significance was calculated using Student's t-test, and is shown *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6.

Other non-productive processing of splicing factors and regulators. (A) ‘Poison’ CE inclusion events. (B) Combined non-productive processing by internal polyA site selection and poison CE inclusion in Snrnp70. (C) Internal polyA site selection. Top, schematic representation of the regulated event (in gray) with arrows showing primer position and red ‘stop signs’ denoting PTCs. Bottom, histograms showing average values of relative fold change for each primer pair normalized against the geometric mean of five genes not changing in the microarray, except Sf3b3 that was normalized against its own gene expression. Dark red represent Differentiated (D) samples, light red Proliferative (P) samples. Error bars represent standard deviation of the mean (n = 3). Exon inclusion (Inc), exon skipping (skp), alternative 3′ end proximal (Prox), alternative 3′end distal (Dist). For Snrnp70 gene Alternative 3′ end (Alt3′), CE inclusion (CE Inc) or CE skipping (Prod). Statistically significance was calculated using Student's t-test, and is shown *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 7.

Divergent levels of U1 snRNP proteins and U1 snRNA during phenotypic modulation. (A) Western blots for different splicing factors and RNA binding proteins in mouse aorta and bladder samples, comparing differentiated (D) and proliferative (P) samples. Acta2 and Myh-11 antibodies are markers of smooth muscle differentiation. Rpb1, Tubulin and Gapdh are loading controls. snRNA expression levels in mouse (B) and rat PAC1 cells (C) measured by qPCR. Primers for snRNA levels were normalized against the same set of genes as in Figures 5 and 6. Error bars represent standard deviation of the mean (n = 3). Statistically significance was calculated using Student's t-test, and is shown *P < 0.05. (D) RT-PCR for Snrp70 (qPCR with primers as in Figure 6) and Actn1 and Tpm1 (with primers as in Figure 2) in P and D Rat PAC1 cells. Error bars represent mean and standard deviation, (n = 3). (E) Left panels: RNA-FISH for U1 snRNA in rat PAC1 D and P cells. Right panels show DAPI staining for nuclei. (F) Immunofluorescence in rat PAC1 cells for SNRNP70 and U1C. Sm Actin is a marker of SMC differentiation.