Alternative splicing of the Snap23 microexon is regulated by MBNL, QKI, and RBFOX2 in a tissue-specific manner and is altered in striated muscle diseases

Figures

• 

  Figure 1.. Alternative splicing of the Snap23 microexon is developmentally regulated in striated muscle tissues. (A) Schematic of the murine Snap23 gene locus. PCR primers (arrows) bind the constitutive exons (blue) flanking the alternative microexon (exon 6, gold). (B) Schematic of the two SNAP23 protein isoforms produced by alternative splicing of the microexon. Protein domains were identified using PROSITE (expasy) and UniProt. (C-D) Different mouse tissues at postnatal day 4.5 (P4.5) and adulthood (3–4 months old) were evaluated by reverse transcription PCR (RT-PCR) (C) and western blot assays (D). The percent spliced-in (PSI) values were calculated by densitometry. Results are shown as the mean ± SEM, *p ≤ 0.05 versus P4.5, unpaired t-test with Welch’s correction, n = 3–4 independent replicates. aa: amino acids; AE: alternative exon; bp: base pairs; CR: cysteine-rich; gastroc: gastrocnemius.

• 

  Figure 2.. The Snap23 microexon is mis-spliced in mouse models of heart and skeletal muscle diseases. (A-B) Timeline of experiments using left anterior descending (LAD) ligation (A) and transverse aortic constriction (TAC) (B) surgeries in mice. (C-D) Alternative splicing of the Snap23 microexon in LAD (C) and TAC (D) models of heart disease was evaluated by RT-PCR assays. (E) Scatterplot correlating microexon inclusion (percent spliced-in, PSI) with cardiac ejection fraction after TAC surgery. (F) Timeline of the barium chloride (BaCl2) injection performed in mice. Red = healthy muscle, pink = regenerating muscle, gray = injured muscle. (G) Alternative splicing of the Snap23 microexon 4, 14 and 41 days after BaCl2 injection was evaluated by RT-PCR assays. (H) Scatterplots correlating inclusion of the Snap23 microexon (PSI) with skeletal muscle weights after BaCl2 injection. PSI values were calculated by densitometry. Results are shown as the mean ± SEM, *p ≤ 0.05 versus the respective control condition (basal, sham, vehicle), ordinary one-way ANOVA with Tukey’s multiple comparisons test, n = 3–4 independent replicates for LAD (A), unpaired t-test with Welch’s correction, n = 2–4…

• 

  Figure 3.. The Snap23 developmental splicing transition is recapitulated during C2C12 cell differentiation. (A) Schematic of C2C12 myoblast differentiation into myotubes. (B-C) Alternative splicing of the Snap23 microexon during C2C12 cell differentiation was evaluated by RT-PCR (B) and western blot (C) assays. The percent spliced-in (PSI) values were calculated by densitometry. (D) Total SNAP23 protein levels were evaluated by western blot assays and quantified by densitometry. Results are shown as the mean ± SEM, *p ≤ 0.05 versus UD, ordinary one-way ANOVA with Dunnett’s multiple comparisons test, n = 3 independent replicates. bp: base paris; D: differentiation day; FBS: foetal bovine serum; HS: horse serum; MYH3: myosin heavy chain 3; ns: not significant, UD: undifferentiated.

• 

  Figure 4.. Strength of the splice donor site determines the level of Snap23 microexon inclusion. (A) Schematic of the murine Snap23 gene locus annotated with the splice donor site and splice acceptor site sequences. (B) Schematics of the empty, wild-type, and edited Snap23 minigenes expressed in undifferentiated C2C12 cells. Arrows indicate the location of the primers used to evaluate alternative splicing of the Snap23 microexon in the minigenes. (C) Alternative splicing of the three minigene constructs in myoblasts was evaluated by RT-PCR assays. The percent spliced-in (PSI) values were calculated by densitometry. bp: base pairs; RSV: Rous sarcoma virus; SA: splice acceptor; SD: splice donor.

• 

  Figure 5.. MBNL, QKI, and RBFOX2 are positive splicing regulators of the Snap23 microexon. (A) Differences in inclusion of the Snap23 microexon (ΔPSI) were plotted for each RBP knockdown. RNA was harvested from either undifferentiated myoblasts (si-Ptbp1, si-Ptbp2, si-Ptbp1 + si-Ptbp2) or differentiated myotubes (si-Elavl1, si-Celf1, si-Celf2, si-Celf1 + si-Celf2, si-Mbnl1, si-Mbnl2, si-Mbnl1 + si-Mbnl2, si-Qki, si-Rbfox2, si-Qki + si-Mbnl1 + si-Mbnl2, si-Qki + si-Rbfox2). The ΔPSI values were calculated as follows: PSI (si-RBP) – PSI (si-control). (B-D) Alternative splicing of the Snap23 microexon following depletion of MBNL1 + MBNL2 (B), QKI (C), or RBFOX2 (D) was evaluated by RT-PCR assays. The percent spliced-in (PSI) values were calculated by densitometry. (E) Schematic of the putative MBNL, QKI, and RBFOX2 binding motifs identified in the upstream and downstream intronic regions flanking the alternatively spliced Snap23 microexon. (F-G) Alternative splicing of the Snap23 microexon following depletion of QKI + MBNL1 + MBNL2 (F) or QKI + RBFOX2 (G) was evaluated by RT-PCR assays. PSI values were calculated by densitometry. Results are shown as the mean ± SEM, *p ≤ 0.05 versus mock and…

• 

  Figure 6.. QKI and RBFOX2 bind their putative motifs with high affinity in vitro. (A) Schematic of the fluorescence polarization (FP) assay. (B-C) Binding curves for Snap23 oligonucleotides incubated with purified QKI (B) or RBFOX2 (C) proteins. Results are shown as mean ± SEM, nonlinear regression (curve fit), 4-parameter logistical binding model with least squares fit, n = 4 independent replicates.

• 

  Figure 7.. Putative QKI and RBFOX2 motifs are necessary for Snap23 microexon alternative splicing regulation. (A) Schematic of the wild-type Snap23 minigene expressed in C2C12 cells. (B) Alternative splicing of the Snap23 minigene in wild-type myotubes (differentiated for 4 days) after deletion of the putative RBP motifs was evaluated by RT-PCR assays. (C) Alternative splicing of the Snap23 minigene in myotubes depleted of MBNL1 + MBNL2 (differentiated for 4 days) after deletion of the putative RBP motifs was evaluated by RT-PCR assays. The percent spliced-in (PSI) values were calculated by densitometry. Results are shown as the mean ± SEM, *p ≤ 0.05 versus wild-type, one-way ANOVA with Dunnett’s multiple comparisons test, n = 3–4 independent replicates. bp: base pairs; RSV: Rous sarcoma virus; SA: splice acceptor; SD: splice donor.

• 

  Figure 8.. Blocking endogenous QKI and RBFOX2 motifs reduces Snap23 microexon inclusion. (A) Schematic of the RBP motifs blocked using morpholino (MO) antisense oligonucleotides in C2C12 cells. (B) Alternative splicing of the Snap23 microexon in myotubes (differentiated for 4 days) after putative RBP motifs were blocked with the respective MOs was evaluated by RT-PCR assays. The percent spliced-in (PSI) values were calculated by densitometry. Results are shown as the mean ± SEM, *p ≤ 0.05 versus control MO, one-way ANOVA with Dunnett’s multiple comparisons test, n = 3–4 independent replicates. bp: base pairs.