MLCK and phosphor-MLC expression increased in sensory neurons after peripheral axotomy.

(A-D) Sciatic nerve axotomy was performed 3 days before and L4-L5 DRG tissues were collected for Western blotting or qPCR to examine the expression level of MLCK and phosphor-MLC (p-MLC). Representative images (A) and quantification of Western blots (B, D) demonstrate that expression levels of MLCK and p-MLC increase significantly in adult DRGs following sciatic nerve transection. (N=3, ** p < 0.01; *** p < 0.001).

(A) (C) L4-L5 DRGs were collected for qRT-PCR, 1, 3 and 7 days after axotomy, respectively. The results reveal that mRNA expression of MLCK in the DRG neurons also increased after axotomy after 1, 3 and 7 days, although the difference at 3 days was the most significant. (N=3, *p < 0.05; ** p < 0.01).

(E-F) Sections of L4-L5 DRGs from naive and sciatic nerve axotomized mice were stained with antibodies against MLCK, Tuj-1 or p-MLC. Immunohistochemistry images indicate that MLCK (E) and p-MLC (F) expression were elevated in the sensory neurons following sciatic nerve axotomy. Scale bar: 100μm.

Suppression of MLCK activity inhibits axon growth in adult DRG neurons.

(A-E) DRG neurons from adult mice were cultured and treated with DMSO or 10.0 µM ML-7. After 3 days, neurons were lysed for Western blotting or staining with Tuj1 antibody. Western blot images (A) and quantification of p-MLC (B) indicate that ML-7 treatment successfully inhibits MLCK activity. Quantification of axon length indicated that inhibition of MLCK activity with ML-7 dramatically blocks adult sensory neuronal axon growth in vitro (D, E), and reduces the numbers of neurons with axons (E). Scale bar: 100μm (N=3, *p < 0.05; *** p < 0.001).

(F-I) Adult DRG neurons were isolated and electroporated with scramble siRNA (Ctrl) or MLCK siRNA (siMLCK) mixed with GFP. Tuj1 (Red) antibody was used to stain neurons 3 days after transfection. A representative image (F) and quantification of Western blots for MLCK (G) indicate that siMLCK significantly suppressed MLCK and p-MLC expression (N=3, *p < 0.05). Quantification of mean axon length (H, I) indicates that transfection of siMLCK inhibits adult DRG neuronal axon growth. Scale bar: 100μm. (N=3, *** p < 0.001).

(J, K) L4/L5 DRGs of adult mice were electroporated with scramble siRNA (Ctrl) or siMLCK combined with GFP, respectively. Two days later, sciatic nerves were crushed and then regenerated axons traced after a further three days. Representative images (J) and measurement of regenerating axonal length (K) indicated that decreased MLCK expression with siRNA inhibits sensory axon regeneration in vivo. Crush sites are marked using red arrow. Regenerating axon tips are marked with yellow arrow (N=5, * p < 0.05). Scale bar: 1mm.

Knockdown of MLCP protein promotes axon growth from adult peripheral sensory neurons.

(A-D) Adult DRG neurons were isolated and electroporated with scramble siRNA (Ctrl) or MYPT1 siRNA (siMYPT1) mixed with GFP. A representative Western blot image (A) and quantification of protein band (B) indicate that transfection of siMYPT1 in cultured DRG neurons significantly increases p-MLC expression (N=3, ** p < 0.01). Quantification of axon length (C, D) indicated siMYPT1 transfection in the adult sensory neurons promotes its axon growth on the permissive substrate PDL or inhibitory substrates, such as CSPGs and myelin. (N=3, ***p < 0.001).

(E, F) L4/L5 DRGs of adult mice were electroporated with scramble siRNA (Ctrl) or siMYPT1, respectively. Two days later, sciatic nerves were crushed and then regenerated axons traced after a further three days. Representative images (E) and measurement of regenerating axonal length (F) indicated that siMYPT1 promotes sensory axon regeneration in vivo. Crush sites are marked using red arrow. Regenerating axon tips are marked with yellow arrow (N=5, * p < 0.05). Scale bar: 1mm.

Inhibition of MLCP activity promotes axon growth from developing embryonic neurons.

(A, B) Cortical neurons were isolated from embryonic day 14.5 embryos. Neurons were treated with DMSO, ML-7 or PDBu, or transfected with ctrl siRNA, siMLCK or siMYPT1. The p-MLC expression levels were examined using Western blot analysis after 3 days of culture.

(C-F) coritcal neurons were treated with DMSO, ML-7 or PDBu or transfected with ctrl siRNA, siMLCK or siMYPT1. After three days, cells were stained with Tuj1 to visualize axons. Representative images (C, E) and quantification (D, F) of axonal length demonstrate ML-7 or siMLCK inhibits embryonic cortical neuronal axon growth. In contrast, inhibition of MLCP activity with PDBu or siMYPT1 promoted embryonic cortical neuronal axon growth (N=3, ***p < 0.001). Scale bar: 100 μm.

(G-J) Hippocampal neurons were isolated from embryonic day 18.5 embryos. Neurons were treated with DMSO, ML-7 or PDBu or transfected with scramble siRNA, siMLCK or siMYPT1. Similarly, Representative images (G, I) and quantification (H, J) of axonal length demonstrate that ML-7 or siMLCK inhibits hippocampal neuronal axon growth. In contrast, suppression of MLCP with PDBu or siMYPT1 promoted hippocampal neuronal axon growth (N=3, ***p < 0.001). Scale bar: 100 μm.

Inhibition of MLCP activity promotes CNS axon regeneration in vivo.

(A-C) Local application of a gelatin sponge soaked with PBS (Ctrl) or 50.0 μM PDBu on the site of the lesion (ON), or microinjection of PDBu (2.0 μl, 50.0 μM) into the vitreous body (ivit) were performed after optic nerve crush injury. Three day later, CTB-labeled regenerating axons were quantified. Representative images of optic nerve sagittal sections (A) and quantification of CTB-positive fibers (B) at the point distance 100 μm and 200 μm from the injury site indicate that only local administration of PDBu at the site of the lesion promotes the axon growth. (C) The longest axons per section of group were measured and quantified.

(D-H) T8 whole spinal cord crush injury was performed in adult mice then treated with PBS or 50.0 μM PDBu by intrathecal injection once every 4 days. BDA was injected into the right sensorimotor cortex 2 weeks after spinal cord injury to label the regenerating cortical spinal tract (CST) axons. After 2 weeks, sagittal sections of the spinal cord around the crush site were created, and the distance between the tip of the axons and injury site was measured. Representative image of Western blots (D) and its quantification (E) demonstrates that intrathecal injection of PDBu increases p-MLC level, it means that PDBu sufficiently inhibits MLCP activity. Representative image of spinal cord sagittal section (F) and the measurement of distance from BDA-positive axon tips to the injury site demonstrate that intrathecal injections of PDBu prevent injury-induced axon retraction. Arrow indicates BDA-positive axons, dash line indicates lesion site. (H) Mean BBB scores evaluation indicated that there was no difference between PDBu treatment and the PBS control group. (N=5, ***p < 0.001).

MLCK regulates axon regeneration independent of myosin II activity.

(A, B) Adult L4-L5 DRG neurons were cultured in vitro for 3 days with the addition of the pharmacological inhibitors indicated. Representative images (A) and quantification of axon length (B) demonstrate that, although myosin II inhibitor blebbistatin markedly promoted sensory neuronal axon growth, however, it did not rescue the inhibitory effect of ML-7 on sensory axon growth (N=3, ***p < 0.001). (C, D) The blebbistatin administration did not influence MLC phosphorylation levels in cultured DRG neurons, representative image of Western blots (C) and its quantification (D) (N=3, ns: no significant).

Inhibition of MLCK blocks transcription-independent axon growth of adult sensory neurons.

(A) Schematic figure of the culture-and-replate model. Adult sensory neurons were cultured for 3 days then replated to allow growth of new axons for one day. During this process, ML-7 was added prior to or after replating.

(B, C) Representative images (B) and quantification of axon length (C) indicate that treatment with ML-7 after replating significantly inhibits adult sensory axon growth. However, treatment with ML-7 prior to replating has no effect on adult sensory neural axon growth (N=3, ***p < 0.001).

(D, E) Mice underwent sciatic nerve axotomy 7 days previously. L4-L5 DRGs were harvested from naive or axotomized mice and cultured for 1 day with ML-7 treatment. Neurons were stained with Tuj1. Representative images (D) and quantification of axon length (E) indicate that blockade of MLCK activity with ML-7 inhibits axotomy-induced axon growth of adult sensory neurons (N=3, ***p < 0.001).

MLCK and MLCP activity regulate F-actin distribution of the growth cone.

(A, B) Embryonic E14.5 cortical neurons were isolated and treated with DMSO, PDBu or ML-7. After 3 days of culture, neurons were stained with Tuj1 antibody (Green) to visualize tubulin and phalloidin (Red) to visualize F-actin. Representative images (A), quantification of the size of the growth cone (B), and F-actin content (C) indicate that ML-7 treatment increases the growth cone size and F-actin content, and PDBu treatment reduces the growth cone size and the F-actin content. Scale bar: 1μm (N=3, ***p < 0.001).

ML-7 treatment inhibits axon growth in a dose-dependent manner.

The DRG neurons from adult mice were cultured and treated with a series of doses of ML-7. After 3 days culture, neurons were stained with Tuj1 antibody and axonal lengths quantified.

PDBu treatment promotes axon growth from adult peripheral sensory neurons.

(A, B) Representative Western blot image (A) and quantification of protein band (B) indicates that administration of PDBu, the MLCP pharmacological inhibitor, in cultured DRG neurons increases p-MLC expression (N=3, ** p < 0.01).

(C, D) DRG neurons were cultured on permissive (PDL) or inhibitory substrates (CSPGs or myelin) in the presence or absence of PDBu, as indicated. Neurons were stained with Tuj1. Representative images (C) and quantification of axon length (D) show that increased p-MLC expression with PDBu addition, which promotes adult sensory neuronal axon growth on either permissive substrate PDL or inhibitory substrates, such as CSPGs and myelin. (N=3, ***p < 0.001).

Optic nerve crush injury model

Optic nerve was crushed behind the eyeball with a custom-modified jeweler’s 5# forceps for 1 second. Two weeks later, 2.0 µl Alexa Fluor 488-conjugated CTB (green) was injected intravitreally to label regenerating optic nerve axons. Immunofluorescent staining of GFAP staining (red) was also performed in the longitudinal section of crushed optic nerves.

Whole spinal cord crush injury model

Whole spinal cord was crushed at T8 level with a custom-modified jeweler’s 5# forceps for 1 second. Two weeks later, a 1.6 μl aliquot of 10% BDA was injected into the sensorimotor cortex to label regenerating cortical spinal tract axons, and another two weeks later, the spinal cord was dissected out. Immunofluorescent staining of BDA (red) and GFAP staining (green) was performed in the sagittal section of spinal cord.

Intrathecal PDBu injection prevented axonal retraction bulb formation.

Representative images of spinal cord sagittal section (A) and quantification of retraction bulb formation (B). White triangle indicates retraction bulb (N=3, *** p < 0.001).