Disrupted autophagy after spinal cord injury is associated with ER stress and neuronal cell death

Abstract

Autophagy is a catabolic mechanism facilitating degradation of cytoplasmic proteins and organelles in a lysosome-dependent manner. Autophagy flux is necessary for normal neuronal homeostasis and its dysfunction contributes to neuronal cell death in several neurodegenerative diseases. Elevated autophagy has been reported after spinal cord injury (SCI); however, its mechanism, cell type specificity and relationship to cell death are unknown. Using a rat model of contusive SCI, we observed accumulation of LC3-II-positive autophagosomes starting at posttrauma day 1. This was accompanied by a pronounced accumulation of autophagy substrate protein p62, indicating that early elevation of autophagy markers reflected disrupted autophagosome degradation. Levels of lysosomal protease cathepsin D and numbers of cathepsin-D-positive lysosomes were also decreased at this time, suggesting that lysosomal damage may contribute to the observed defect in autophagy flux. Normalization of p62 levels started by day 7 after SCI, and was associated with increased cathepsin D levels. At day 1 after SCI, accumulation of autophagosomes was pronounced in ventral horn motor neurons and dorsal column oligodendrocytes and microglia. In motor neurons, disruption of autophagy strongly correlated with evidence of endoplasmic reticulum (ER) stress. As autophagy is thought to protect against ER stress, its disruption after SCI could contribute to ER-stress-induced neuronal apoptosis. Consistently, motor neurons showing disrupted autophagy co-expressed ER-stress-associated initiator caspase 12 and cleaved executioner caspase 3. Together, these findings indicate that SCI causes lysosomal dysfunction that contributes to autophagy disruption and associated ER-stress-induced neuronal apoptosis.

Figure 1

Accumulation of autophagosomes after SCI is due to impaired autophagy flux. (a) Representative western blots of autophagy markers LC3, Beclin 1, VPS34, phospho-ULK1 (P-ULK1), ATG12-ATG5 conjugate, p62 and ubiquitinated proteins in lysates from spinal cord of sham and SCI rats. Each lane represents an individual animal. (b–h) Quantification of western blot data from a. Data represent mean±S.D. normalized to corresponding sham; n≥4; *P<0.05, **P<0.01, ***P<0.001 by one-way ANOVA, followed by two-tailed t-test. (i) Representative images ( × 20) of IHC staining against LC3 (green) and p62 (red) in VH of gray matter of sham and SCI animals. Co-localization of LC3 and p62 is apparent at day 1 after SCI. Scale bar is 20 μm. (j and k) Quantification of LC3-positive and double-positive LC3⁺/p62⁺ cells in sham and SCI animals normalized to the total area imaged. (l) Representative high magnification images ( × 60) of LC3-positive cells in VH of gray matter from sham and day 1 SCI animals. Accumulation of LC3-positive puncta corresponding to autophagosomes is apparent after SCI. Scale bar is 10 μm. (m) Quantification of LC3-positive puncta in VH of gray matter normalized to the total number of imaged cells with neuronal morphology. All IHC data represent mean±S.E.; n≥4; *P<0.05 by one-way ANOVA, followed by post hoc

Figure 2

SCI leads to lysosomal dysfunction. (a) Representative western blots of lysosomal proteins LAMP1 and CTSD from spinal cord of sham and SCI animals. (b and c) Quantification of western blot data from a. Data represent mean±S.D. normalized to corresponding sham; n≥4; ***P<0.001 by one-way ANOVA, followed by two-tailed t-test. (d) Representative high magnification images ( × 60) of LC3 (green) and CTSD (red) staining in VH of gray matter from sham and day 1 SCI animals. Decrease in number and intensity of CTSD puncta (lysosomes) and increase in LC3 puncta (autophagosomes) is apparent after SCI. Scale bar is 10 μm. (e) Close-up of the indicated area (white box) from CTSD staining in (d). Arrows point to representative CTSD-positive lysosomes. Scale bar is 1 μm. (f–h) Quantification of data from (d): (f) CTSD-positive puncta (lysosomes), (g) LC3-positive puncta (autophagosomes) and LC3⁺/CTSD⁺ double-positive puncta (autolysosomes) in VH of gray matter normalized to the total number of imaged cells with neuronal morphology. (h) Decrease in fraction of autolysosomes (LC3⁺/CTSD⁺) as compared with all autophagosomes (LC3⁺) at day 1 after SCI. All IHC data represent mean±S.E.; n≥3; *P<0.05, **P<0.01 by one-way ANOVA, followed by post hoc

Figure 3

Autophagosomes accumulate in neurons of the VH gray matter at day 1 after SCI. (a) Representative images of IHC staining for LC3 (green) and neuronal marker NeuN (red) in VH of gray matter from sham and SCI animals. Stronger co-localization between NeuN and LC3 is apparent at day 1 after SCI. Scale bar is 20 μm. (b and c) Quantification of neurons (NeuN+) co-localizing with LC3 in ventral (b) and dorsal (c) horn of gray matter from sham and SCI animals normalized to total area imaged. Data are normalized to total area imaged and represent mean±S.E.; n≥4; *P<0.05, by one-way ANOVA, followed by post hoc

Figure 4

Autophagosomes accumulate in microglia and oligodendrocytes of the dorsal white matter adjacent to injury after SCI. (a) Representative images of IHC staining for LC3 (green) and activated microglia marker CD11B (red) in dorsal white matter of sham and SCI animals. Increased co-localization between LC3 and CD11B is apparent at day 1 after SCI. (b and c) Quantification of microglia (CD11B⁺) co-localizing with LC3 in dorsal (b) and ventral (c) white matter from sham and SCI animals. (d) Representative images of IHC staining for LC3 (green) and oligodendrocytes marker CC1 (red) in dorsal white matter of sham and SCI animals. Increased co-localization between LC3 and CC1 is apparent after SCI. All scale bars are 20 μm. (e and f) Quantification of oligodendrocytes (CC1⁺) co-localizing with LC3 in dorsal (e) and ventral (f) white matter from sham and SCI animals. Data are normalized to total area imaged and represent mean±S.E.; n≥4; *P<0.05, **P<0.01, ***P<0.001 by one-way ANOVA, followed by post hoc

Figure 5

Impaired autophagy flux is associated with neuronal cell death after SCI. (a) Representative western blots of cell death (spectrin, caspase 12 (CASP12)) and ER stress (GRP78, CHOP, ATF4) markers from spinal cord of sham and SCI animals. (b–f) Quantification of western blot data from a. Data represent mean±S.D. normalized to corresponding sham; n≥4; *P<0.05, **P<0.01, ***P<0.001 by one-way ANOVA, followed by two-tailed t-test. (g) Representative images of IHC staining for cleaved caspase 3 (C-CASP3, green) and p62 (red) in VH of gray matter from sham and SCI animals. Co-localization between cleaved CASP3 and p62 is apparent at day 1 after SCI. (h and i) Quantification of cleaved CASP3-positive cells and double-positive C-CASP3⁺/p62⁺ cells in ventral (f) and dorsal (g) horns of gray matter from sham and SCI animals. (j) Representative images of IHC staining for p75 (green) and p62 (red) in VH of gray matter from sham and SCI animals. Co-localization is apparent at day 1 after SCI. All scale bars are 20 μm. (k and l) Quantification of p75-positive and double-positive p75⁺/p62⁺ cells in ventral (i) and dorsal (j) horns of gray matter from sham and SCI animals. All IHC data are normalized to total area imaged and represent mean±S.E.; n≥4; *P<0.05, **P<0.01, ***P<0.001 by one-way ANOVA, followed by post hoc

Figure 6

Impaired autophagy flux is associated with ER stress and ER-stress-induced neuronal apoptosis after SCI. (a) Representative images of IHC staining for ER-stress-associated caspase 12 (CASP12, green) and p62 (red) in VH of the gray matter from sham and SCI animals. Co-localization of CASP12 and p62 is apparent at day 1 after SCI. (b and c) Quantification of CASP12-positive and CASP12⁺/p62⁺ double-positive cells in ventral (b) and dorsal (c) horns of gray matter from sham and SCI animals. (d) Representative images of IHC staining for ER stress marker GRP78 (green) and p62 (red) in VH of gray matter from sham and SCI animals. Co-localization of GRP78 and p62 is apparent at 1 day after SCI. All scale bars are 20 μm. (e and f) Quantification of GRP78-positive and GRP78⁺/p62⁺ double-positive cells in ventral (e) and dorsal (f) horns of gray matter from sham and SCI animals. Data are normalized to total area imaged and represent mean±S.E.; n≥4; *P<0.05, **P<0.01, ***P<0.001 by one-way ANOVA, followed by post hoc