HIF-1α Directly Controls WNT7A Expression During Myogenesis

Abstract

Herein we unveil that Hypoxia-inducible factor-1α (HIF-1α) directly regulates WNT7A expression during myogenesis. In fact, chromatin immunoprecipitation (ChiP) and site-directed mutagenesis experiments revealed two distinct hypoxia response elements (HREs) that are specific HIF-1α binding sites on the WNT7A promoter. Remarkably, a pharmacological activation of HIF-1α induced WNT7A expression and enhanced muscle differentiation. On the other hand, silencing of WNT7A using CRISPR/Cas9 genome editing blocked the effects of HIF-1α activation on myogenesis. Finally, treatment with prolyl hydroxylases (PHDs) inhibitors improved muscle regeneration in vitro and in vivo in a cardiotoxin (CTX)-induced muscle injury mouse model, paving the way for further studies to test its efficacy on acute and chronic muscular pathologies.

Keywords: FG-4592; Hypoxia-inducible factor-1α; Prolyl-hydroxylases; WNT7a; hypertrophy; myogenesis.

FIGURE 1

Assessment of hypoxia inducible factor-1α (HIF-1α) binding to the WNT7A promoter after 24h at 1% O₂. (A) Schematic representation of pNL1.1 plasmid containing the WNT7A promoter upstream of the luciferase gene reporter. Under normoxia, HIF-1α is degraded and cannot bind the HRE sequences on the WNT7A promoter, inhibiting the luminescence signal; under hypoxia, HIF-1α is stabilized and can bind the WNT7A promoter, inducing the expression of the luciferase gene. (B) Quantification of pNL1.1-WNT7A promoter activity by luciferase assay in murine myoblasts under normoxia and hypoxia (1% O₂; n = 11). (C) Schematic representation of the seven binding sequences designed within the WNT7A promoter and used for qPCR analysis after chromatin immunoprecipitation (ChiP). (D) qPCR analysis of HIF-1α binding on Seq.1 and Seq.4 of the WNT7A promoter (n = 3). (E) Schematic representation of the deletions of the HRE binding sequences identified in Seq. 1 (Deletion 1.1) and Seq. 4 (Deletion 4.1, 4.2, and 4.3) inserted in the pNL1.1-WNT7A promoter construct using site-directed mutagenesis experiments. (F) Quantification of mutated pNL1.1-WNT7A promoter activity by luciferase assay (n = 5 for mutation 1.1, n = 3 for mutation 4.1, 4.2 and 4.3). Data information: all data represent mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Ordinary one-way ANOVA).

FIGURE 2

Pharmacological activation of hypoxia inducible factor-1α (HIF-1α) with prolyl-hydroxylases (PHDs) inhibitors and its effects on WNT7A regulation. (A) Schematic representation of the pcDNA3 construct containing the oxygen dependent domain (ODD)-luciferase reporter gene. Under normoxia, the proline residues on the ODD sequence are hydroxylated by the PHDs, and the complex degraded by the proteasome, eventually inhibiting the generation of the luminescence signal; the pharmacological inhibition of the PHDs causes the stabilization of the ODD, inducing the production of luminescence. (B) IC₅₀ quantification of IOX2 or FG-4592 by luciferase assay with the ODD-pcDNA3 construct (n = 3). (C) Western blot analysis and relative quantification of HIF-1α nuclear localization in murine myoblasts treated with IOX2 or FG-4592 for 24 h. The nuclear marker Lamin A/C was used as the housekeeper (n = 3). (D) qPCR analysis of HIF-1α target genes, VEGF and PHD2, upon PHDs inhibition with IOX2 or FG-4592 (n = 3). (E) Schematic representation of pNL1.1 plasmid containing the WNT7A promoter upstream of the luciferase gene reporter under normoxia and upon IOX2 or FG-4592 treatment. (F) Quantification of pNL1.1-WNT7A promoter activity by luciferase assay of murine myoblasts treated with IOX2 or FG-4592 (n = 4). (G) Western blot analysis and relative quantification of Wnt7a accumulation in murine myoblasts treated with IOX2 or FG-4592 for 24 h. EE1a was used as the housekeeper (n = 3). Data information: all data represent mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 (Ordinary one-way ANOVA).

FIGURE 3

Effects of hypoxia inducible factor-1α (HIF-1α) pharmacological activation on skeletal muscle differentiation. (A) Immunofluorescence staining of myosin heavy chain (MHC; green) in proliferating (MHC NEG. CTRL) and differentiated murine myoblasts treated with IOX2 or FG-4592 for 24 h in growth medium (GM) and then switched to differentiation medium (DM) without the prolyl-hydroxylases (PHDs) inhibitors for 7 days. Nuclei were stained with Hoechst 33342. Magnification is 200×. (B) Quantification of the fusion index, as the ratio between MHC-positive nuclei and the total number of nuclei, and of the differentiation index, as the ratio between myotubes area and MHC-positive nuclei at the end of differentiation (n = 3). Data information: all data represent mean ± SD. ***p < 0.001, ****p < 0.0001 (Ordinary one-way ANOVA).

FIGURE 4

Modulation of myogenic differentiation markers upon pharmacological hypoxia. (A) Schematic representation of myogenic marker during skeletal muscle differentiation. (B) Western blot analysis and relative quantification of MyoD nuclear localization of murine myoblasts treated for 24 h with IOX2 or FG-4592. The nuclear marker Lamin A/C was used as the housekeeper (n = 4). (C) qPCR analysis of MyoR upon prolyl-hydroxylases (PHDs) inhibitors treatment (n = 4 for IOX2 and n = 3 for FG-4592). (D,E) qPCR analyses of MYOGENIN at 3 days (D, n = 5 for IOX2 and n = 3 for FG-4592) and of myosin heavy chain (MHC) at 7 days of differentiation (E, n = 6 for IOX2 and n = 3 for FG-4592), in murine myoblasts pre-treated with IOX2 or FG-4592. Data information: all data represent mean ± SD. *p < 0.05, ****p < 0.0001 (Ordinary one-way ANOVA).

FIGURE 5

Characterization of WNT7A inhibition using CRISPR/Cas9 model in C2C12. (A) Graphical representation of the mouse WNT7A-targeted locus. Boxes and lines indicate exons and introns, respectively. The oligo represents sgRNA target sequence and Protospacer Adjacent Motifs (PAM) site is marked in red. (B) Sanger DNA sequencing were conducted on PCR products amplified from the genomic WNT7A loci of C2C12. (C,D) Genomic PCR analysis (C) and quantification of WNT7A inhibition (D). RPLI was used as the housekeeper (n = 3). (E,F) Western Blot analysis (E) and quantification (F) of WNT7A expression in wild type (WT) and WNT7A-silenced murine myoblasts (WNT7A-KO; n = 3). (G) Cell viability analyzed by RealTime-Glo^TM kits MT Cell Viability Assay of wild type and WNT7A-silenced murine myoblasts (n = 3). Data information: all data represent mean ± SD. *p < 0.05, ****p < 0.0001 (Student’s t-test).

FIGURE 6

Modulation of hypoxia inducible factor-1α (HIF-1α) pathway induced by WNT7A silencing upon physical and pharmacological hypoxia. (A,B) Western blot analysis (A) and relative quantification (B) of HIF-1α nuclear localization in wild type (WT) and WNT7A-silenced murine myoblasts (WNT7A-KO) treated under normoxia, hypoxia, FG-4592 or IOX2. The nuclear marker Lamin A/C was used as the housekeeper (n = 4 for Normoxia, n = 3 for Hypoxia, IOX2 and FG-4592). (C) qPCR analysis of HIF-1α target genes, VEGF, PHD2, and PHD3 in WT and WNT7A-silenced murine myoblasts (WNT7A-KO) upon physical and pharmacological hypoxia (n = 4 for Normoxia, n = 3 for Hypoxia, IOX2 and FG-4592). Data information: all data represent mean ± SD. For western blot analysis of HIF-1α *p < 0.05 (Student’s t-test). For qPCR analysis of HIF-1α target genes *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (ordinary one-way ANOVA).

FIGURE 7

Effects of WNT7A inhibition on skeletal muscle differentiation upon physical and pharmacological hypoxia. (A) Negative control of myosin heavy chain (MHC) staining in proliferation wild type (WT) and WNT7A-silenced murine myoblasts (WNT7A-KO). (B) Schematic representation of MHC staining at 7 days of the differentiation process. (C,E,G, I) Immunofluorescence staining of MHC (green) in WT and WNT7A-silenced murine myoblasts (WNT7A-KO) treated under normoxia (C), hypoxia (E), FG-4592 (G), or IOX2 (I) for 24 h in GM and then switched to differentiation medium (DM) without the prolyl-hydroxylases (PHDs) inhibitors for 7 days. Nuclei were stained with Hoechst 33342. Magnification is 200× (n = 3). (D,F,H,J) Quantification of the fusion index, as the ratio between MHC-positive nuclei and the total number of nuclei, and of the differentiation index, as the ratio between myotubes area and MHC-positive nuclei at the end of differentiation (n = 3). Data information: all data represent mean ± SD. **p < 0.01, ***p < 0.001, ****p < 0.0001 (Student’s t-test).

FIGURE 8

Modulation of MyoD nuclear translocation and myosin heavy chain (MHC) expression in wild type (WT) and WNT7A-silenced murine myoblasts upon physical and pharmacological hypoxia. (A,B) Western blot analysis (A) and relative quantification (B) of MyoD nuclear localization were tested after 24 h preconditioning under physical or chemical hypoxia in WT and WNT7A-KO. The nuclear marker Lamin A/C was used as the housekeeper (n = 3). (C,D) Western blot analysis (C) and relative quantification (D) of MHC were measured at the end of the skeletal muscle differentiation in WT and WNT7A-KO. The early endosomal antigen 1a (EE1a) was used as the housekeeper (n = 3). Data information: all data represent mean ± SD. *p < 0.05, **p < 0.01, ****p < 0.0001 (Student’s t-test).

FIGURE 9

Effects of FG-4592 treatment on muscle regeneration. (A) Schematic illustration of the experimental procedure: 24 h before cardiotoxin (CTX) injury of tibialis anterior (TA) muscle, mice were injected by i.p. with FG-4592 (10 mg/kg). Mice were sacrificed 7 days after CTX-induced injury, and TA muscles were collected. (B) Immunohistochemical detection of hematoxylin/eosin in FG-4592 pre-treated mice, compared to controls. Scale bars = 40 μM (n = 7 mice for each group). (C) Immunofluorescence staining of laminin (green) in saline and FG-4,592 pre-treated mice. Nuclei were stained with Hoechst 33,342. Scale bars = 40 μM (n = 7 mice for each group). (D,E) Quantification of CSA (D) and minimal Feret diameter (E) averages in FG-4592 or saline pre-treated mice. (F,G) Representation of CSA (F) and minimal Feret diameter (G) distribution in saline or FG-4592 pre-treated mice. Data information: all data represent mean ± SD. *p < 0.05 (Student’s t-test).

FIGURE 10

Assessment of activation of hypoxia inducible factor-1α (HIF-1α) pathway in FG-4,592 or saline treated mice upon cardiotoxin CTX-induced muscle injury. (A,B) Western blot analysis (A) and relative quantification (B) of HIF-1α and Wnt7a accumulation in saline and FG-4592 pre-treated mice 7 days after CTX-induced muscle injury (n = 7). (C) Immunofluorescence staining of CD31 (green) on tibial anterior muscle sections of saline and FG-4592 pre-treated mice 7 days after CTX-induced muscle injury (Scale bars = 40 μM). (D and E) Quantification of the total number of capillaries (D), normalized for the total number of muscle fibers, and the density of capillaries (E), calculated as the total number of capillaries on the area of the fibers, in saline or FG-4592 pre-treated mice (n = 7 for each group). Data information: all data represent mean ± SD. p > 0.05, n.s., *p < 0.05 (Student’s t-test).

FIGURE 11

Schematic representation of hypoxia inducible factor-1α (HIF-1α) binding to the WNT7A promoter. Under normoxia, HIF-1α is hydroxylated by the active prolyl-hydroxylases (PHDs) and sent to proteasomal degradation. Under hypoxia or upon PHDs inhibition, HIF-1α cannot be hydroxylated, and it translocates to the cell nucleus, where it forms a transcriptional complex with its α-subunit and CBP/p300, and binds to specific HRE sequences on the WNT7A promoter region.