Functional regeneration of the transected recurrent laryngeal nerve using a collagen scaffold loaded with laminin and laminin-binding BDNF and GDNF

Abstract

Recurrent laryngeal nerve (RLN) injury remains a challenge due to the lack of effective treatments. In this study, we established a new drug delivery system consisting of a tube of Heal-All Oral Cavity Repair Membrane loaded with laminin and neurotrophic factors and tested its ability to promote functional recovery following RLN injury. We created recombinant fusion proteins consisting of brain-derived neurotrophic factor (BDNF) and glial cell line-derived neurotrophic factor (GDNF) fused to laminin-binding domains (LBDs) in order to prevent neurotrophin diffusion. LBD-BDNF, LBD-GDNF, and laminin were injected into a collagen tube that was fitted to the ends of the transected RLN in rats. Functional recovery was assessed 4, 8, and 12 weeks after injury. Although vocal fold movement was not restored until 12 weeks after injury, animals treated with the collagen tube loaded with laminin, LBD-BDNF and LBD-GDNF showed improved recovery in vocalisation, arytenoid cartilage angles, compound muscle action potentials and regenerated fibre area compared to animals treated by autologous nerve grafting (p < 0.05). These results demonstrate the drug delivery system induced nerve regeneration following RLN transection that was superior to that induced by autologus nerve grafting. It may have potential applications in nerve regeneration of RLN transection injury.

Figure 1. Structure of Heal-All Oral Cavity Repair Membrane.

The basement membrane (a) and porous surface (c) were taken in kind. SEM images displayed the microstructure of the basement membrane surface (b) and porous surface (d).

Figure 2. Effect of recombinant proteins on cell survival in PC 12 cells by CCK-8 assay.

(a) More living cells were found in the neurotrophic factors with NtA, when the same quality recombinant proteins were provided. (b) More living cells were found in only LBD-GDNF added than only LBD-BDNF (P < 0.05). Whatever ratios of LBD-BDNF and LBD-GDNF mixed are preferable than single usage of LBD-BDNF or LBD-GDNF (P < 0.05), and the most living cells were found when the ratio of LBD-BDNF to LBD-GDNF was 2:3 (P < 0.05). The ratio of 0 or 10 indicates cells were cultured in the mentioned conditions without LBD-BDNF or LBD-GDNF. n = 6. *P < 0.05.

Figure 3. Vocalisation spectras, spectral area and amplitude percentages of different groups were shown.

Vocalisation spectras were shown of the Sham (a), ANG (b), CS (c), CS+LN (d), CS+LN+LBD-BDNF (e), CS+LN+LBD-GDNF (f) and CS+LN+LBD-BDNF+LBD-GDNF (g) groups at 12 weeks after surgery. The Sham group exhibits a high amplitude with wide and continuous waveforms. (h) Spectral area percentages of different groups. The spectral area, calculated as a percentage of that in the Sham group increased over time. The spectral area percentages measured at 12 weeks after surgery were 100.0 ± 7.0%, 67.0 ± 8.2%, 54.4 ± 16.0%, 60.7 ± 11.3%, 65.1 ± 3.5%, 65.9 ± 4.6%, and 71.7 ± 9.0% of all the above group orders, respectively. The spectral area percentage of the CS+LN+LBD-BDNF group was much higher than that of the CS group at week 4, 8 and 12 week (P < 0.05), but much lower than that of CS+LN+LBD-BDNF+LBD-GDNF and Sham groups (P < 0.05). At week 12, there was no significant difference between the CS+LN+LBD-BDNF group and CS+LN+LBD-GDNF or ANG group (P > 0.05). The spectral area percentage of the CS+LN+LBD-BDNF+LBD-GDNF group showed no significant difference compared to that of the ANG group at 4 and 8 weeks (P > 0.05), while the ANG group showed significant lower compared to the CS+LN+LBD-BDNF+LBD-GDNF group at 12 weeks after surgery (P < 0.05). (i) The amplitude percentages of different groups were shown. The spectral amplitude, calculated as a percentage of that in the Sham group increased with time. The amplitude percentages of the ANG, CS+LN+LBD-BDNF, CS+LN+LBD-GDNF and CS+LN+LBD-BDNF+LBD-GDNF groups were not significantly different from that of the Sham group at 12 weeks after surgery (P > 0.05). From week 4 to week 12, the amplitude percentage of the CS+LN+LBD-BDNF+LBD-GDNF group was not significantly different from that of the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group (P > 0.05). n = 30. *P < 0.05.

Figure 4

(a) Open and (b) closed angles between arytenoid cartilages during movement of the vocal fold. The arytenoid cartilages’ angles of the arytenoid cartilages improved over time. (a) The open angles measured 4, 8 and 12 weeks after surgery were 28.79 ± 4.63°, 31.97 ± 2.78°, 35.52 ± 3.78° of the CS+LN+LBD-BDNF+LBD-GDNF group, respectively. The open angle of the CS+LN+LBD-BDNF+LBD-GDNF group showed no significant difference compared to that of the ANG group at 4 and 8 weeks after surgery (P > 0.05); while the CS+LN+LBD-BDNF+LBD-GDNF group showed significant improvements compared to the ANG group at 12 weeks (P < 0.05). (b) The closed angles at week 4, 8 and 12 were 8.35 ± 0.46°, 8.11 ± 0.87°, 6.88 ± 0.51° of the CS+LN+LBD-BDNF+LBD-GDNF group, respectively. Twelve weeks after surgery, the CS+LN+LBD-BDNF+LBD-GDNF group showed markedly smaller closed angle compared to the ANG group (P < 0.05), although the angle remained worse than that of the Sham group (P < 0.05). n = 30. *P < 0.05.

Figure 5. The latency and amplitude of the right PCA and TA muscles in the above groups.

In all experimental groups, response latencies decreased and amplitudes increased from 4 to 12 weeks after surgery, although none of the groups showed responses at week 12 that were comparable to those of the Sham group (p < 0.05). (a) The latencies of the right PCA were significantly restored in the CS+LN+LBD-BDNF+LBD-GDNF group compared to those of the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group from week 4 to week 12 (P < 0.05). The latencies were shorter in the ANG group than in the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group from week 4 to week 8 (P < 0.05) and the latencies of the CS+LN+LBD-BDNF+LBD-GDNF group were shorter than those of the ANG group at week 12 (P < 0.05). (b) The amplitudes of the right PCA were significantly restored in the ANG group compared to those of the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group from week 4 to week 8 (P < 0.05). The amplitudes were much higher in the CS+LN+LBD-BDNF+LBD-GDNF group than in the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group at all time points after surgery (P < 0.05). (c) The latencies of the right TA were significantly restored in the CS+LN+LBD-BDNF+LBD-GDNF group compared to those of the CS+LN+LBD-GDNF group from week 4 to week 12 (P < 0.05). The latencies of the CS+LN+LBD-BDNF group showed significant differences compared to those of the CS+LN+LBD-BDNF+LBD-GDNF group, except for week 8. The latencies were shorter in the ANG group than in the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group from week 4 to week 8 (P < 0.05). Further, the differences were not significant between the ANG group and the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group. (d) The amplitudes of the right TA were significantly restored in the ANG group compared to those of the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group from week 4 to week 8 (P < 0.05). The amplitudes were much higher in the CS+LN+LBD-BDNF+LBD-GDNF group than in the CS+LN+LBD-GDNF or CS+LN+LBD-GDNF group at all time points (P < 0.05). n = 30. *P < 0.05.

Figure 6. Detection of BDNF and GDNF expression in the larynx.

(a,c) Detection of BDNF expression in the laryngeal by Western blotting analysis. (b,d) Detection of GDNF expression in the laryngeal by Western blotting analysis. (c) Expression of BDNF tended to increase from 4 to 12 weeks after surgery, except for the CS+LN+LBD-BDNF and CS+LN+LBD-BDNF+LBD-GDNF groups. BDNF expression was markedly higher in the CS+LN+LBD-BDNF, CS+LN+LBD-GDNF or CS+LN+LBD-BDNF+LBD-GDNF group than in the Sham group at all time points (P < 0.05). While BDNF expression in the CS+LN+LBD-BDNF group was higher than in the CS+LN+LBD-BDNF+LBD-GDNF group at week 4 (P < 0.05), BDNF expression in the CS+LN+LBD-BDNF+LBD-GDNF group was higher than in the CS+LN+LBD-BDNF group at weeks 8 and 12 (P < 0.05). (d) Expression of GDNF tended to increase from 4 to 12 weeks after surgery, except for the CS+LN+LBD-GDNF and CS+LN+LBD-BDNF+LBD-GDNF groups. GDNF expression was markedly higher in the CS+LN+LBD-BDNF, CS+LN+LBD-GDNF or CS+LN+LBD-BDNF+LBD-GDNF group than in the Sham group at all time points (P < 0.05). GDNF expression in the CS+LN+LBD-GDNF group was higher than in the CS+LN+LBD-BDNF+LBD-GDNF group at week 4 (P < 0.05), while this pattern was reversed at weeks 8 and 12 (P < 0.05). n = 30. *P < 0.05.

Figure 7

BDNF (a) and GDNF (b) mRNA expression in the right recurrent laryngeal nerve by real-time PCR. (a) BDNF mRNA expression increased over time in all experimental groups. Although mRNA expression in the CS+LN+LBD-BDNF group was markedly lower than in the ANG group at weeks 4 and 8 (P < 0.05), this pattern was reversed at week 12 (P < 0.05). Furthermore, BDNF expression was markedly lower in the CS+LN+LBD-BDNF group than in the CS+LN+LBD-BDNF+LBD-GDNF group at all time points (P < 0.05). (b) GDNF mRNA expression increased over time in all experimental groups. GDNF mRNA was higher in the CS+LN+LBD-GDNF and CS+LN+LBD-BDNF+LBD-GDNF groups than in the ANG group at week 12 (P < 0.05). Furthermore, GDNF expression was much lower in the CS+LN+LBD-GDNF group than in the CS+LN+LBD-BDNF+LBD-GDNF group at all time points after surgery (P < 0.05). n = 15. *P < 0.05.

Figure 8. Transmission electron microscopy of the middle segment of the regenerated nerve at 12 weeks after surgery.

(a), the Sham group; (b), the ANG group; (c), the CS group; (d), the CS+LN group; (e), the CS+LN+LBD-BDNF group; (f), the CS+LN+LBD-GDNF group; (g), the CS+LN+LBD-BDNF+LBD-GDNF group. Myelinated nerve fibre cross sectional area (h) and myelin sheath thickness (i) were quantitatively evaluated and statistically analyzed. (h) Myelinated nerve fibre cross sectional area of all experimental groups increased with time. The area was higher in the CS+LN+LBD-BDNF+LBD-GDNF group than in the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group at all time points (P < 0.05), and higher than in the ANG group at week 12 (P < 0.05). (i) Myelin sheath thickness of all experimental groups increased with time. The thickness was thicker in the CS+LN+LBD-BDNF+LBD-GDNF group than in the CS+LN+LBD-BDNF or CS+LN+LBD-GDNF group (P < 0.05), and there was a negligible increase in the ANG group compared to the CS+LN+LBD-BDNF+LBD-GDNF group at all time points (P < 0.05). Scale bar = 2 μm. *P < 0.05.