ROCK2 is a major regulator of axonal degeneration, neuronal death and axonal regeneration in the CNS - PubMed (original) (raw)

ROCK2 is a major regulator of axonal degeneration, neuronal death and axonal regeneration in the CNS

J C Koch et al. Cell Death Dis. 2014.

Abstract

The Rho/ROCK/LIMK pathway is central for the mediation of repulsive environmental signals in the central nervous system. Several studies using pharmacological Rho-associated protein kinase (ROCK) inhibitors have shown positive effects on neurite regeneration and suggest additional pro-survival effects in neurons. However, as none of these drugs is completely target specific, it remains unclear how these effects are mediated and whether ROCK is really the most relevant target of the pathway. To answer these questions, we generated adeno-associated viral vectors to specifically downregulate ROCK2 and LIM domain kinase (LIMK)-1 in rat retinal ganglion cells (RGCs) in vitro and in vivo. We show here that specific knockdown of ROCK2 and LIMK1 equally enhanced neurite outgrowth of RGCs on inhibitory substrates and both induced substantial neuronal regeneration over distances of more than 5 mm after rat optic nerve crush (ONC) in vivo. However, only knockdown of ROCK2 but not LIMK1 increased survival of RGCs after optic nerve axotomy. Moreover, knockdown of ROCK2 attenuated axonal degeneration of the proximal axon after ONC assessed by in vivo live imaging. Mechanistically, we demonstrate here that knockdown of ROCK2 resulted in decreased intraneuronal activity of calpain and caspase 3, whereas levels of pAkt and collapsin response mediator protein 2 and autophagic flux were increased. Taken together, our data characterize ROCK2 as a specific therapeutic target in neurodegenerative diseases and demonstrate new downstream effects of ROCK2 including axonal degeneration, apoptosis and autophagy.

PubMed Disclaimer

Figures

Figure 1

Figure 1

AAV.shRNA-mediated knockdown of ROCK2 and LIMK1 in vitro and in vivo. (a) Vector maps of the AAV used to knockdown ROCK2, LIMK1 and EGFP as control. The respective shRNAs are expressed under the control of an H1 promoter. The fluorophore dsRed is expressed under the control of a synapsin promoter. ITR, AAV-2 inverted terminal repeat; Int, intron; SV40-pA, SV40 polyadenylation site. (b) Immunoblots for ROCK2 and LIMK1 of primary RGCs transduced with AAV.ROCK2-shRNA and AAV.LIMK1-shRNA compared with AAV.EGFP-shRNA, respectively. The quantifications of the band intensities normalized to loading control show a >50% knockdown efficacy for both AAV-shRNAs (_n_=3; bars represent means±S.E.M. **P<0.005 according to Student's _t_-test). (c) RT-PCR of RNA extracts from primary RGCs transduced with AAV.EGFP-shRNA (black), AAV.ROCK2-shRNA (dark gray) and AAV.LIMK1-shRNA (light gray) on DIV 7. Exemplary amplification results for each primer assessed by SYBR green fluorescence intensities (y axis) per PCR cycle (x axis) are shown in the four upper graphs. Quantifications of relative mRNA expression levels normalized to GAPDH mRNA expression are shown at the bottom. (d and e) Immunohistochemistry of transversal retina sections 2 weeks after intravitreal injection of the given AAV. The ROCK2 staining (d; left column) shows a strong signal in the RGCs transduced with AAV.EGFP-shRNA (top row) and reflects the subcellular localization of ROCK2 close to the membrane. In contrast, the RGCs transduced with AAV.ROCK2-shRNA (d; bottom row) show a clearly diminished ROCK2 signal. Correspondingly, the LIMK1 staining (e; left column) shows a stronger signal in the RGCs transduced with AAV.EGFP-shRNA (e; top row) as compared with AAV.LIMK1-shRNA (bottom row). DsRed (middle columns) corresponds to virally transduced RGCs. White arrows highlight representative RGCs transduced with the given AAV (i.e. showing dsRed fluorescence) and their corresponding ROCK2 or LIMK1 staining. Scale bar: 100 _μ_m

Figure 2

Figure 2

Knockdown of ROCK2 and LIMK1 rescues the inhibitory effects of CSPG on neurite outgrowth in RGCs. (a and b) Representative photos of RGCs transduced with the AAV given on the left side on DIV 5 (phase contrast on the left, dsRed fluorescence on the right representing the virally transduced cells). Neurites of the RGCs transduced with AAV.EGFP-shRNA and without treatment are much shorter on the non-permissive substrate CSPG (b) as compared with Laminin (a). In contrast, RGCs transduced with AAV.ROCK2-shRNA and AAV.LIMK1-shRNA are not affected by CSPG with regards to their neurite growth. Scale bar: 100 _μ_m. (c) Quantification of the mean neurite length per RGC transduced with the given AAV or untreated and plated either on laminin or CSPG (_n_=3 RGC cultures; bars represent means±S.E.M.; **P<0.005, ***P<0.0005, according to one-way analysis of variance (ANOVA) followed by Dunnett's post-hoc test). (d) Quantification of the mean cell number per view field of RGCs transduced with the given AAV or untreated and plated either on laminin or CSPG. There was no significant difference between the groups (_n_=3 RGC cultures; bars represent means±S.E.M.)

Figure 3

Figure 3

Knockdown of ROCK2 and LIMK1 increases axonal regeneration after optic nerve crush. (a) Schematic drawing of experimental procedures. AAVs were injected intravitreally 14 days before crush lesion of the optic nerve. Twenty-eight days after crush, the animals were killed and the number of GAP43-positive axons was quantified on optic nerve sections at different distances from the crush site. (b) Representative photos of optic nerves transduced with the AAV given on top and immunostained for GAP43 at different distances distal from the crush site (marked on the scale bar at the right side). For AAV.ROCK2-shRNA and AAV.LIMK1-shRNA, single GAP43-positve axons (white arrows) could be followed over distances over 4500 _μ_m distal from the crush site. Scale bar: 100 _μ_m. (c) Quantification of the number of GAP43-positive axons at different distances from the crush site. Transduction with AAV.ROCK2-shRNA and AAV.LIMK1-shRNA significantly increased the number of axons at almost all analyzed positions along the nerve (_n_=10 optic nerves per group; bars represent means±S.E.M.; *P<0.05; **P<0.005 according to one-way ANOVA followed by Dunnett's post-hoc test). (d) Representative overview photos of the area 500 _μ_m adjacent to the crush site of optic nerves transduced with the AAV given on top and immunostained for GAP43. Transduction with AAV.ROCK2-shRNA increased the number of GAP43-positve axons on both sides of the crush. Scale bar: 100 _μ_m. (e) Quantification of the number of GAP43-positive axons at 100 _μ_m proximal to the crush site shows a significant increase only for the optic nerves transduced with AAV.ROCK2-shRNA (_n_=10 optic nerves per group; bars represent means±S.E.M.; **P<0.005, according to one-way ANOVA followed by Dunnett's post-hoc test)

Figure 4

Figure 4

Knockdown of ROCK2 but not LIMK1 increases RGC survival after optic nerve axotomy. (a) Schematic drawing of experimental procedures. AAVs were injected intravitreally 14 days before axotomy. Five days before axotomy, 2 _μ_l 5% FluoroGold were injected stereotactically in the colliculus superior on both sides in order to label the RGCs. Fourteen days after transection of the optic nerve, the number of retinal ganglion cells was counted on retina flat mounts. (b) Representative photos of the central retina (optic nerve head on top) showing RGCs transduced with the given AAV in red and FluoroGold-positive cells in yellow. Top row: × 10 magnification, scale bar: 400 _μ_m; bottom row: × 20 magnification, scale bar: 200 _μ_m. (c) Quantification of the number of FluoroGold-positive RGCs at 2 weeks after axotomy in the central region of the retina (0–1 mm around the optic nerve head) given in percent of AAV.EGFP-shRNA control. Transduction with AAV.ROCK2-shRNA but not AAV.LIMK1-shRNA significantly increased the number of surviving RGCs compared with control (_n_=5 retinas per group; bars represent means±S.E.M.; **P<0.005; ns: not significant, according to one-way analysis of variance followed by Dunnett's post-hoc test)

Figure 5

Figure 5

Knockdown of ROCK2 attenuates degeneration of the proximal axon following optic nerve crush assessed by in vivo live imaging. (a) Schematic drawing of experimental procedures. AAVs were intravitreally injected 14–21 days before optic nerve crush. Before and over 6 h after optic nerve crush an in vivo live imaging of the optic nerve was performed to monitor axonal degeneration in the area 500 _μ_m proximal and distal to the crush site. (b) Representative composite overview photos from an in vivo live imaging of an optic nerve transduced with AAV.ROCK2-shRNA at 5 min (top row, ‘0 h') and 6 h (bottom row, ‘6 h') after optic nerve crush. The crush suture (white arrow) can be seen on the right side on the photos from the proximal side of the crush (left column) and on the left side on the photos from the distal side of the crush (right column). A dashed box at both time points highlights two representative axons on each side of the crush. The axons on the proximal side (box ‘1') appear stable even at 6 h after crush, whereas the axons on the distal side (box ‘2') are already fragmented at this time point. Scale bar: 100 _μ_m. (c) Representative time row photos of single axons (composite pictures from z-stacks) from an in vivo live imaging of the optic nerve at the time points given on top (in hours after crush). The axons transduced with AAV.EGFP-shRNA (top row) show the typical kinetics of acute axonal degeneration and fragment within 6 h on both proximal and distal side of the crush. In contrast, axons transduced with AAV.ROCK2-shRNA (bottom row) on the proximal side of the crush show only little signs of degeneration like bulb formation (black arrows) but do not fragment within the 6 h time period. On the distal side, however, this axon stabilizing effect was not as pronounced as some axons also stayed stable (such as the axon on the left side of the pictures; marked with *), whereas others degenerated with normal kinetics of acute axonal degeneration (such as the axon on the right side). (d) Graphs show the quantification of the axonal integrity ratio (AIR), which is defined as the sum length of axonal fragments divided by the total axon length before fragmentation. On the proximal side, the AIR was significantly higher at later time points in the AAV.ROCK2-shRNA transduced axons reflecting their greater stability. On the distal side, the standard deviations of the AIR in the AAV.ROCK2-shRNA group were large reflecting the varying kinetics of single axons (as depicted in c), but taken together there were no significant differences between the two groups (_n_=8 axons on both sides of the crush from four optic nerves per group; error bars represent S.E.M.; *P<0.05, according to Student's _t_-test)

Figure 6

Figure 6

ROCK2 downregulation modulates apoptosis, cell survival and autophagy. Immunoblotting was performed of protein lysates from primary RGC cultures on DIV 9. Photos of representative western blots are shown on the left side, the quantification of band intensities normalized to tubulin or GAPDH as loading control and relative to AAV.EGFP-shRNA (EGFP) are depicted on the right side (_n_=3 independent RGC cultures; bars represent means±S.E.M.; *P<0.05, **P<0.005, ***P<0.0005, according to one-way analysis of variance followed by Dunnett's post-hoc test). (a and b) Calpain activity (corresponding to the 145 kDa breakdown product (BDP) of spectrin) as well as caspase-3 activity (corresponding to the 120 kDa BDP of spectrin) are significantly decreased in RGCs transduced with AAV.ROCK2-shRNA (ROCK2) compared with AAV.EGFP-shRNA (EGFP) and AAV.LIMK1-shRNA (LIMK1). (c and d) Decreased caspase-3 activity in the RGCs transduced with AAV.ROCK2-shRNA was confirmed by immunoblotting for cleaved caspase-3. (e and f) Detection of the pro-survival factors pAkt and Bcl2 showed increased pAkt levels in the RGCs transduced with AAV.ROCK2-shRNA, whereas Bcl2 levels did not differ significantly among the groups. (g and h) Levels of phospho-PTEN were not altered in the RGCs transduced with AAV.ROCK2-shRNA. However, protein levels of the two CRMP2 splicing variants CRMP2A and CRMP2B were almost doubled in the ROCK2-shRNA group as compared with EGFP-shRNA control. (i and j) ROCK2-shRNA increased LC3-II levels by more than two-fold compared with EGFP-shRNA, whereas p62-levels were decreased. Addition of 10 nM bafilomycin for 4 h before lysis further enhanced LC3-II levels and this effect was stronger in the ROCK2-shRNA group as compared with control. (k and l) Expression levels of the main upstream regulator of autophagy mTOR, its phosphorylated form p-mTOR and the downstream target p-S6 were not significantly changed by AAV.ROCK2-shRNA expression

Figure 7

Figure 7

Schematic drawing of the effects mediated by ROCK2 downregulation observed in the present study. For details refer to the Discussion

Similar articles

Cited by

References

    1. Liu K, Tedeschi A, Park KK, He Z. Neuronal intrinsic mechanisms of axon regeneration. Annu Rev Neurosci. 2011;34:131–152. - PubMed
    1. Mueller BK, Mack H, Teusch N. Rho kinase, a promising drug target for neurological disorders. Nat Rev Drug Discov. 2005;4:387–398. - PubMed
    1. Tonges L, Koch JC, Bahr M, Lingor P. ROCKing regeneration: Rho kinase inhibition as molecular target for neurorestoration. Front Mol Neurosci. 2011;4:39. - PMC - PubMed
    1. Hashimoto R, Nakamura Y, Kosako H, Amano M, Kaibuchi K, Inagaki M, et al. Distribution of Rho-kinase in the bovine brain. Biochem Biophys Res Commun. 1999;263:575–579. - PubMed
    1. Duffy P, Schmandke A, Schmandke A, Sigworth J, Narumiya S, Cafferty WB, et al. Rho-associated kinase II (ROCKII) limits axonal growth after trauma within the adult mouse spinal cord. J Neurosci. 2009;29:15266–15276. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources