Hepatocyte Growth Factor Regulates the miR-206-HDAC4 Cascade to Control Neurogenic Muscle Atrophy following Surgical Denervation in Mice

Abstract

Hepatocyte growth factor (HGF) has been well characterized for its roles in the migration of muscle progenitors during embryogenesis and the differentiation of muscle stem cells, but its function in adult neurogenic muscle atrophic conditions is poorly understood. Here we investigated whether HGF/c-met signaling has any effects on muscle-atrophic conditions. It was found that HGF expression was upregulated in skeletal muscle tissue following surgical denervation and in hSOD1-G93A transgenic mice showing severe muscle loss. Pharmacological inhibition of the c-met receptor decreased the expression level of pri-miR-206, enhanced that of HDAC4 and atrogenes, and resulted in increased muscle atrophy. In C2C12 cells, HGF inhibited phosphorylation of Smad3 and relieved TGF-β-mediated suppression of miR-206 expression via JNK. When extra HGF was exogenously provided through intramuscular injection of plasmid DNA expressing HGF, the extent of muscle atrophy was reduced, and the levels of all affected biochemical markers were changed accordingly, including those of primary and mature miR-206, HDAC4, and various atrogenes. Taken together, our finding suggested that HGF might play an important role in regard to neurogenic muscle atrophy and that HGF might be used as a platform to develop therapeutic agents for neuromuscular disorders.

Keywords: HDAC4; Smad3; hepatocyte growth factor; miR-206; neurogenic muscle atrophy.

Graphical abstract

Figure 1

Expression Kinetics of HGF in Denervated Muscle (A) Expression kinetics of HGF protein after denervation. The muscle was isolated 3, 7, 10, and 14 days after denervation, and total proteins were analyzed by ELISA to measure the protein level of HGF. ***p < 0.001, ****p < 0.0001 versus control muscle (unpaired Student’s t test), n = 4 per group. (B) Change in the RNA level of HGF after denervation. RNAs were prepared from TA muscles 3 days after denervation, followed by qRT-PCR. The values were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). **p < 0.01 (unpaired student’s t test). (C) Expression kinetics of c-met and phosphorylated c-met proteins in denervated TA muscle. Muscle was isolated on days 3 and 7, and total proteins were prepared, followed by western blot using specific antibodies to total or phosphorylated c-met. Each lane represents a sample from an individual mouse. Two representative results are shown. Two independent experiments were performed (n = 4), and similar results were obtained. (D) Comparison of the HGF protein level in TA muscles between wild-type (WT) and 150-day-old hSOD1-G93A transgenic mice. The TA muscle was isolated, and total proteins were analyzed by ELISA to measure the protein level of HGF. ****p < 0.0001 versus WT mice (unpaired Student’s t test), n = 6 per group. All data are represented as mean ± SEM.

Figure 2

Effect of the c-met Inhibitor PHA-665752 on Muscle Atrophy in the Sciatic Nerve Transection Model After sciatic nerve transection, mice were injected i.p. with 20 mg/kg of PHA-665752 on a daily basis until sacrifice. (A) Effect on muscle weight. The graph on the left shows actual weight, whereas, on the right, muscle mass was normalized with the initial weight of mice. Den, denervation; PHA, PHA-665752; ns, not significant. *p < 0.05 (one-way ANOVA), n = 5 per group. (B) The effect on the cross-sectional area (CSA) of TA muscle was analyzed 10 days after denervation. At least 300 muscle fiber areas were counted per sample. Mean CSA is indicated in the graph. **p < 0.05, ***p < 0.01 (one-way ANOVA). Scale bars, 200 μm. (C) Effect on the expression of MuRF1 and Atrogin-1. The RNA level of two genes was determined by real-time qRT-PCR using TA muscles isolated 3 days after denervation. *p < 0.05 (one-way ANOVA), n = 4 per group. All data are represented as mean ± SEM.

Figure 3

Effect of the c-met Inhibitor PHA-665752 on the miR-206-HDAC4 Cascade After denervation by sciatic nerve transection, mice were injected i.p. with 20 mg/kg of PHA-665752 on a daily basis until sacrifice. Three days later, TA muscles were prepared, and total RNAs and proteins were isolated, followed by qRT-PCR or western blot. (A) Effect on HDAC4 RNA. n = 4 per group. (B) Effect on HDAC4 and total and phosphorylated Smad3 protein. This presents two representative results from two independent experiments, with the total number of mice being 4. The graph shows the result of quantification of HDAC4 protein. Values were normalized to GAPDH. ***p < 0.001 (unpaired student’s t test). (C) Effect on the miR-206 primary transcript. *p < 0.05 (one-way ANOVA), n = 4 per group. ND, not detected. All data are represented as mean ± SEM. See also Figure S1.

Figure 4

Effect of Recombinant HGF Protein on miR-206 and Smad3 in C2C12 Cells C2C12 cells were plated and then cultured in differentiation medium in the presence or absence of recombinant TGF-β and HGF proteins. Total RNAs and proteins were prepared and analyzed for miR-206 and Smad3 by qRT-PCR and western blot, respectively. For the western blot, two independent experiments were performed, and one representative result is shown. The graph displays the result of the protein band quantification. (A) Effect on the pri-miR-206 transcript. Values were normalized to GAPDH. *p < 0.05, **p < 0.01 (unpaired Student’s t test), n = 3 per group. (B) Effect on mature miR-206. Values were normalized to miR-103a-3p. *p < 0.05, (unpaired Student’s t test), n = 3 per group. (C) Effect of HGF on Smad3 phosphorylation. The graph shows the result of protein band quantification. Values were normalized to total Smad3. *p < 0.05, **p < 0.01 (unpaired Student’s t test). (D) Effect of various chemical inhibitors on HGF-mediated suppression of phosphorylated Smad3. Values were normalized to total Smad3. *p < 0.05 (unpaired Student’s t test). (E) Effect of the JNK inhibitor on HGF-mediated regulation of pri-miR-206 transcript expression. Values were normalized to GAPDH. *p < 0.05, (unpaired Student’s t test), n = 3 per group. All data are represented as mean ± SEM. See also Figure S2.

Figure 5

Effect of HGF Overexpression by Intramuscular Injection of an HGF-Expressing Plasmid on Muscle Atrophy pCK-HGF-X7 was injected i.m. at the time of sciatic nerve transection, followed by one repeat injection 7 days later. TA muscles were prepared at appropriate time points. (A) Effect on TA muscle weight. Representative TA muscles from 14 days after denervation are shown. *p < 0.05 versus the Den+pCK group (one-way ANOVA), n = 4 per group. Scale bars, 1 mm. (B) Effect on the cross-sectional area of TA muscles. TA muscles were analyzed 10 days after denervation. At least 300 muscle fiber areas were counted per sample. Mean CSA is indicated in the graph. ***p < 0.001, ****p < 0.0001 (one-way ANOVA), n = 4 per group. Scale bars, 100 μm. All data are represented as mean ± SEM. See also Figure S3.

Figure 6

Effect of HGF Overexpression by Intramuscular Injection of an HGF-Expressing Plasmid on the miR-206-HDAC4 Cascade pCK-HGF-X7 was administered i.m. at the time of sciatic nerve transection. Three days after denervation, the TA muscle was isolated, and total RNAs and proteins were analyzed by qRT-PCR and western blot. (A) Effect on the expression of MuRF1 and Atrogin-1. *p < 0.05, **p < 0.01 (one-way ANOVA), n = 4 per group. (B) Effect on HDAC4 RNA. (C) Effect on HDAC4 protein. For the western blot, two representative results are shown. Two independent experiments were performed (n = 4). Values were normalized to GAPDH for both RNA and protein analysis. **p < 0.01 (one-way ANOVA). (D) Effect on the pri-miR-206 transcript. Values were normalized to GAPDH. **p < 0.01 (one-way ANOVA), n = 4 per group. (E) Effect on mature miR-206. Values were normalized to miR-103a-3p. *p < 0.05 (unpaired Student’s t test). n = 3 per group. All data are represented as mean ± SEM.