Alternative polyadenylation mediates microRNA regulation of muscle stem cell function

Abstract

Pax3, a key myogenic regulator, is transiently expressed during activation of adult muscle stem cells, or satellite cells (SCs), and is also expressed in a subset of quiescent SCs (QSCs), but only in specific muscles. The mechanisms regulating these variations in expression are not well understood. Here we show that Pax3 levels are regulated by miR-206, a miRNA with a previously demonstrated role in myogenic differentiation. In most QSCs and activated SCs, miR-206 expression suppresses Pax3 expression. Paradoxically, QSCs that express high levels of Pax3 also express high levels of miR-206. In these QSCs, Pax3 transcripts are subject to alternative polyadenylation, resulting in transcripts with shorter 3' untranslated regions (3'UTRs) that render them resistant to regulation by miR-206. Similar alternate polyadenylation of the Pax3 transcript also occurs in myogenic progenitors during development. Our findings may reflect a general role of alternative polyadenylation in circumventing miRNA-mediated regulation of stem cell function.

Figure 1. miR-206 is highly expressed in QSCs and regulates Pax3 transcript in SCs

Quantitative RT-PCR analysis of Pax3 mRNA (A) or miR-206 (B) levels in QSCs sorted from uninjured muscle and from myoblasts (Mb) sorted from injured muscle 3.5 days after BaCl₂ injection. (C) Quantitative analysis of mRNA levels of Pax3 and Cyclophilin B (Pipb) in primary myoblast cultures treated with miR -1 (black) or miR-206 (white) in growth medium. (D) Quantitative analysis of Pax3 mRNA in primary myoblast cultures treated with anti-miR-1 (black) or anti-miR-206 (white), then cultured in differentiation medium for 1 or 2 days.(E) Luciferase reporter assays showing the long form of Pax3 3′UTR repression by miR-206 in 293 cells. Luciferase constructs and miR-206-expressing plasmid were co-transfected in 293 cells, and luciferase activity was measured 48 hours post-transfection. Mutation of both target sites is necessary to abolish the repression of luciferase activity by miR-206 (m1+m2). (F) Luciferase reporter assays showing the long form of Pax3 3′UTR repression by miR-206 in C2C12 cells after differentiation. After transfection with luciferase constructs, C2C12 cells were cultured in differentiation medium for 48 hours to allow endogenous miR-206 upregulation. Mutation of both target sites is necessary to abolish the repression of luciferase activity by miR-206 (m1+m2). Pax3 murine long 3′UTR was appended to the luciferase ORF (Luc). The different luciferase constructs are indicated on the left of the graphs (WT, m1, m2, m1+m2). miR-206 complementary sites (206₁and 206₂) (vertical line) and mutated sites (cross) are indicated.(G) Competitive inhibition of miR-206 using the Pax3 3′UTR construct. Immunoblot analysis of Pax3 protein level in satellite cell-derived myoblasts, 48 hours after transfection with luciferase constructs containing either wild-type or mutated miR-206 target sites, then cultured in differentiation medium. Repression of Pax3 transcript by miR-206 was rescued by overexpressing wild-type Pax3 3′UTR construct, which acts as a competitive inhibitor (* p<0.05; ** p<0.001; N.S. - not significant; n=3). See also figure S1.

Figure 2. miR-206 regulates Pax3-mediated proliferation and myogenic lineage progression during adult myogenesis ex vivo and in vivo

(A) Low magnification (10x) images of single fibers treated with control miRNA, anti-miR-206, miR-206 or Pax3 siRNA, then stained for Syn4 (green) and DAPI (blue) 3 days after isolation (Bar: 40 μm). (B) Number ofSyn4^+ve cells counted on single fiber explants cultured for 3 days. Fibers were treated with control, anti-miR-206 with control siRNA, or anti-miR-206 with Pax3 siRNA. Quantitative analysis of MyoG (C) or MyoD (D) expression in Syn4^+ve SCs per fiber in single fiber explants treated with control miRNA, miR-206, anti-miR-206, or Pax3 siRNA and cultured for 3 days. (E) Quantitative analysis of MyoD expression inSyn4^+ve SCs per fiber in single fiber explants treated with anti -miR-206 either with Control or Pax3 siRNA. (F) Western blot analysis of QSCs from limb muscle (except extensor digitorum longus (EDL)) in mice injected with control antagomirs (Control) or a nti-miR-206 antagomirs (Antagomir-206). (G) Number of Syn4^+ve cells counted on EDL single fiber explants from mice injected with control antagomir (Control) or anti-miR-206 antagomir (Antagomir -206) and cultured for 3 days. (H) Quantitative analysis of MyoD expression in Syn4^+ve SCs per fiber on single fiber explants (EDL) from mice injected with control antagomirs (Control) or anti-miR-206 antagomirs (Antagomir -206) and cultured for 3 days (Line indicates mean; * p<0.01; ** p<0.0001;N.S. -not significant; n=51). See also Figure S2.

Figure 3. Pax3 mRNA is not susceptible to miR-206 regulation in diaphragm SCs

(A) Western blot analysis of Pax3 protein level (left) and quantitative RT-PCR analysis of Pax3 mRNA level (right) in QSCs from limb (L) and diaphragm (D) muscles (n=3). (B) Number of Syn4^+ve cells counted on single fiber explants from limb and from diaphragm cultured for 3 days. Fibers were treated with control or Pax3 siRNA (Line indicates mean, n=51). (C) Quantitative RT-PCR analysis of miR-206 levelin QSCs from limb (L) and diaphragm (D) muscles (n=3). (D) Number of Syn4^+ve cells counted on single fiber explants from limb (EDL) and from diaphragm cultured for 3 days. Fibers were treated with control, miR-206 or anti-miR-206 (Line indicates mean;*p<0.05, ** p<0.0001; n=51). See also Figure S3.

Figure 4. Differential polyadenylation of Pax3 mRNA 3′UTR in limb and diaphragm QSCs and in embryonic limb progenitors

(A) Graphical representation indicating the positions of the putative alternative polyadenylation sites (PAS₁, PAS₂, PAS₃, PAS₄) and putative miR-206 targeted sites (206₁, 206₂) in the Pax3 3′UTR.(B) DNA sequencing of 3′UTRs of three different isoforms of Pax3 detected by 3′RACE. PAS consensus sequence (PAS) and polyadenylation tail region (A(n)) are indicated. (C) Quantitative RT-PCR analysis of Pax3 transcripts bearing short and long 3′UTRs in limb (L) and in diaphragm (D) QSCs. (D) Ratios of long and short Pax3 mRNA isoforms in limb buds at E10.5 and E11.5. In (C) and (D), the proportions of the short form in L and D were compared. (E) Quantitative RT-PCR analysis of Pax3 mRNA levels in E10.5 myogenic progenitors treated with control or miR-206. Pax3 siRNA was used as a positive control (**p<0.001; N.S. - not significant; n=3). See also Figure S4.