Hepatocyte Growth Factor (HGF) Promotes Peripheral Nerve Regeneration by Activating Repair Schwann Cells

Abstract

During the peripheral nerve regeneration process, a variety of neurotrophic factors play roles in nerve repair by acting on neuronal or non-neuronal cells. In this report, we investigated the role(s) of hepatocyte growth factor (HGF) and its receptor, c-met, in peripheral nerve regeneration. When mice were subjected to sciatic nerve injury, the HGF protein level was highly increased at the injured and distal sites. The level of both total and phosphorylated c-met was also highly upregulated, but almost exclusively in Schwann cells (SCs) distal from the injury site. When mice were treated with a c-met inhibitor, PHA-665752, myelin thickness and axon regrowth were decreased indicating that re-myelination was hindered. HGF promoted the migration and proliferation of cultured SCs, and also induced the expression of various genes such as GDNF and LIF, presumably by activating ERK pathways. Furthermore, exogenous supply of HGF around the injury site, by intramuscular injection of a plasmid DNA expressing human HGF, enhanced the myelin thickness and axon diameter in injured nerves. Taken together, our results indicate that HGF and c-met play important roles in Schwann cell-mediated nerve repair, and also that HGF gene transfer may provide a useful tool for treating peripheral neuropathy.

Figure 1

Increased expression of HGF in the nerve injury site. (A) Time kinetics of HGF expression in the sciatic nerve after nerve injury. Following nerve crush, total ipsilateral sciatic nerves were isolated at appropriate time points, and total proteins were analyzed by ELISA to measure the HGF protein level. (B) Spatial distribution of HGF expression in the injured sciatic nerve. The level of the HGF protein was measured in three different areas of the sciatic nerve (proximal, injury and distal sites) at 4 d.p.i by ELISA. ***p < 0.001, ****p < 0.0001, n.s = not significant (C) High level expression of HGF in the injured site of sciatic nerve. Ipsilateral nerves were isolated at 4 d.p.i and subjected to immunohistochemical assay using an antibody to HGF (red). Scale bar = 100 μm (D) HGF expression in DRG and thigh muscles detected at 4 d.p.i. The protein level of HGF was measured in thigh muscles at 0, and 4 d.p.i by ELISA. n = 3 for each group. n.s = not significant. Values represent the mean ± S.E.M.

Figure 2

Expression and activation of c-met. (A) c-met expression in different areas of the injured nerve. After nerve crush, proximal (P), injury (I), and distal (D) regions were isolated at different time points followed by Western blot using an antibody to phosphorylated or total c-met. GAPDH was used as a loading control. Full-length blots are presented in Supplementary Fig. S4 (B) The bar graph shows the expression level of total c-met expression at crush 4 d.p.i. Full-length blots from three independent experiments are presented in Supplementary Figs S4A,B and S5. **P < 0.01. Values represent the mean ± S.E.M. (C) Immunohistochemical analysis of c-met receptor. Injured sciatic nerves were isolated at 4 d.p.i.

Figure 3

Identification of cell types expressing c-met. The distal sites of injured sciatic nerves were analyzed for various cell markers by immunohistochemistry assay using antibodies to CNPase and GFAP for SCs, CD11b for macrophages, CD31 for endothelial cells (all green), and phosphorylated c-met (red). The phosphorylated c-met was mainly merged with SCs marker (white arrow). Injured sciatic nerves were prepared at crush 4 d.p.i. Nuclei were counterstained with Hoechst (blue). n = 3 for each group. Scale bar = 20 μm.

Figure 4

Nerve regeneration was inhibited by c-met inhibitor, PHA-665752. After sciatic nerve crush, mice were intraperitoneally injected with 20 mg/kg of PHA-665752 daily until sacrifice, followed by TEM and IHC. (A) Effects of PHA-665752 on c-met expression. Sciatic nerves were isolated at 4 d.p.i, and total proteins were prepared from three different areas followed by Western blot using specific antibodies to phosphorylated and total c-met. (n = 4 for each group). Full-length blots are presented in Supplementary Fig. S5. (B) Effects of PHA-665752 on HGF expression in the injured nerve. Injured nerves were isolated at 4 d.p.i and analyzed by ELISA. n = 3 for each group. (C) Effects of PHA-665752 on re-myelination as measured by TEM. Electron micrographs of sciatic nerves showed the cross section of sciatic nerve at 14 d.p.i, 4 mm from injury site. The white arrows indicate apoptotic cells. n = 3 for each group. Scale bar, 5 μm. (D) The scatter plot shows the distribution of myelin thickness versus axon diameter in PHA-995752 treated and untreated groups. The bar graphs show g-ratio value and the distribution of g-ratio. 250~300 axons, n = 3 for each group. *p < 0.05, ***p < 0.001. (E) Effects on axon diameter. The graph shows the distribution of axon diameter. (F) Effects on the number of re-myelinated axons. The number of axons was calculated from TEM photos. n = 3 for each group. ***p < 0.001. In all these experiments, values represent the mean ± S.E.M.

Figure 5

Effects of HGF on migration and proliferation of Schwann cells. Primary Schwann cells isolated from rat sciatic nerves were treated with recombinant human HGF protein to measure the effects on migration and proliferation. (A) Expression of c-met receptor in primary SCs. Scale bar = 25 μm. Full-length blots are presented in Supplementary Fig. S6. The samples derived from the same experiment and blots were processed in parallel. (B) Effects of HGF on SC migration. Primary SCs were treated with 10 ng/ml and 25 ng/ml of recombinant HGF protein, and their migration was analyzed using Boyden chambers. ****p < 0.0001. Values represent the mean ± S.E.M. of three independent experiments. (C) Effect of HGF on proliferation of SCs. SCs were treated with 5, 10, 25, 50, and 100 ng/ml HGF and SC proliferations were analyzed by WST-1 assay. (D) SCs proliferation was analyzed by immunocytochemistry assay, using an antibody to Ki67 (Red). Scale bar = 50 μm. (E) Signaling cascades activated by HGF treatment in SCs. Primary Schwann cells isolated from rat sciatic nerves were treated with recombinant human HGF protein. Total proteins were isolated from SCs treated with HGF at appropriate times and analyzed by Western blot using antibodies to respective proteins. Full-length blots are presented in Supplementary Fig. S7. The samples derived from the same experiment and blots were processed in parallel.

Figure 6

Effects of ERK and AKT inhibitors in primary SCs. Primary SCs were treated with 25 ng/ml of recombinant HGF protein in the presence of 10 μM of U0126 or 10 μM of AKTi. Total RNAs were prepared at 1 hr and subjected to quantitative RT-PCR. Effects on Cell migration and proliferation were measured by Boyden chamber and WST-1 assays, respectively. (A) Effects of U0126 and AKTi on the RNA level of Egr-1, c-Fos, GDNF and Lif. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. control. ^#p < 0.05, ^##p < 0.01, ^###p < 0.001, ^####p < 0.0001 vs. HGF only treated group. (B) Effect on SC migration. **p < 0.01, ***p < 0.001, n.s = not significant. (C) Effect on SC proliferation was measured by WST-1 assay. *p < 0.05 vs. control. ^#p < 0.05 vs. HGF only treated group. (D) Effects of PHA-665752 on gene expression in injured nerve was measured by quantitative RT-PCR. Total RNAs were isolated from sciatic nerve at crush 4 d.p.i, *p < 0.001, **p < 0.01, ***p < 0.001. Values represent the mean ± S.E.M.

Figure 7

Overexpression of HGF facilitates nerve regeneration. pCK-HGFX7 was i.m. injected to the thigh muscle around the injured sciatic neuron at the time of nerve crush surgery. pCK lacking the HGF sequence was used as a control. (A) Analysis of sciatic nerves by TEM. Sciatic nerves were isolated at 7, 14, and 28 d.p.i (4 mm distal from injury site). n = 3 for each groups. Scale bar, 5 μm. (B) The graph shows scatter plots of g-ratio in pCK-HGFX7 and pCK control groups at 28 d.p.i. The bar graphs show g-ratio value and the distribution of g-ratio. *p < 0.05, ****p < 0.0001. (C) The graph presents the distribution of axon diameter. (300~400 axons, n = 3 for each group). (D) Effects of HGF on number of myelinated axons in injured nerves. n = 3 for each group. n.s = not significant. Values represent the mean ± S.E.M.