Protein disulfide isomerase ERp57 protects early muscle denervation in experimental ALS

Amyotrophic lateral sclerosis (ALS) is a progressive fatal neurodegenerative disease that affects motoneurons. Mutations in superoxide dismutase 1 (SOD1) have been described as a causative genetic factor for ALS. Mice overexpressing ALS-linked mutant SOD1 develop ALS symptoms accompanied by histopathological alterations and protein aggregation. The protein disulfide isomerase family member ERp57 is one of the main up-regulated proteins in tissue of ALS patients and mutant SOD1 mice, whereas point mutations in ERp57 were described as possible risk factors to develop the disease. ERp57 catalyzes disulfide bond formation and isomerization in the endoplasmic reticulum (ER), constituting a central component of protein quality control mechanisms. However, the actual contribution of ERp57 to ALS pathogenesis remained to be defined. Here, we studied the consequences of overexpressing ERp57 in experimental ALS using mutant SOD1 mice. Double transgenic SOD1G93A/ERp57WT animals presented delayed deterioration of electrophysiological activity and maintained muscle innervation compared to single transgenic SOD1G93A littermates at early-symptomatic stage, along with improved motor performance without affecting survival. The overexpression of ERp57 reduced mutant SOD1 aggregation, but only at disease end-stage, dissociating its role as an anti-aggregation factor from the protection of neuromuscular junctions. Instead, proteomic analysis revealed that the neuroprotective effects of ERp57 overexpression correlated with increased levels of synaptic and actin cytoskeleton proteins in the spinal cord. Taken together, our results suggest that ERp57 operates as a disease modifier at early stages by maintaining motoneuron connectivity.


Materials and Methods
Phenotypic characterization. Disease progression of SOD1 G93A ALS model was followed using body weight measurements, wire hang test, rotarod test, and clinical score analysis. Body weight was measured once a week starting at 5 weeks of age until end point.
Rotarod test was conducted by placing the mouse in a rotating cylinder under constant acceleration protocol from 4 to 40 rpm in 2 min. Latency to fall from the cylinder to the base platform was recorded as measure of motor performance. This protocol assures the fall of mice due to limb problems and not from stamina draining. One week before rotarod test, mice were trained for 5 days by walking on the cylinder at a constant speed of 4 rpm for 1 minute followed by 10 rpm for another minute. Rotarod test was performed in a single session of 3 trials once a week starting at six weeks of age. Mice that failed learning the test during the training period were excluded from analysis.
Clinical score was determined by assessing four disease parameters quantified according to an arbitrary scale: 1) Hindlimb clasping: under healthy conditions mice fully extend their hindlimbs when hold by the tail, while symptomatic mice are unable to extend the hindlimbs; 2) Kyphosis: curvature of the spine; 3) Absence of grooming: observed as dirty fur and eyes; 4) Paralysis: starting as difficulty to move hindlimbs during cage walking. For each parameter, the score was assigned depending on the severity: 0: absence of sign; 1: mild sign; 3: moderate sign; 5: severe sign. Clinical score was measured once a week starting at 5 weeks of age and its value corresponded to the sum of each parameter. Measurements of clinical score were performed by an observer blinded to the genotype to avoid bias.
Lumbar spinal cord histological analysis. For total motoneuron count, 16 serial sections from lumbar spinal cord per animal were used assuring 3.2 mm coverage of the lumbar segment of the spinal cord from L5 to L2 (one section every 200 m). For gliosis analysis, 4 serial sections from lumbar spinal cord per animal were used assuring 3.2 mm coverage of the lumbar segment of the spinal cord from L5 to L2 (one section every 800 m).
For total motoneuron count, anti-ChAT immunohistochemistry (IHQ) was performed by treating spinal cord sections with 3% H 2 O 2 -10% MeOH in TBS for 15 min at RT to inhibit endogenous peroxidases. After one wash step with TBS, epitope retrieval was performed using citrate buffer (10 mM sodium citrate, 0.05% Tween 20, pH 6.0) for 5 min at 95°C. After one wash with TBS, spinal cord cryosections were mounted on super frost slides (VWR International). Sections were blocked using 5% donkey serum diluted in 0.2% Triton X-100 in TBS (blocking buffer) for 1 h at room temperature (RT) and then incubated with 1:250 anti-ChAT antibody (MerckMillipore, AB144P) in blocking buffer for 48 h at RT. Sections were then washed with TBS six times for 15 min each and incubated with 1:2,000 anti-goat HRP-conjugated antibody (MerckMillipore, AP180P) in blocking buffer for 3 h at RT and then washed with TBS six times for 15 min each. Chromogenic reaction was performed using DAB kit (Vector Lab) following manufacturer's instructions. ChAT positive cells were manually counted using bright field microscopy.
Sections were blocked using 1% BSA diluted in 0.025% Triton X-100 in PBS for 2 h at RT and then incubated with 1:500 anti-GFAP or 1:500 anti-Iba1 antibodies in 1% BSA in PBS overnight at 4°C. Sections were then washed four times for 5 min each and incubated with 1:1,000 anti-rabbit Alexa-488 conjugated secondary antibody (Molecular Probes), and 1:5,000 Hoechst 33,342 for nuclear staining, in 1% BSA in PBS for 2 h at RT. After four washes in PBS, sections were covered with coverslips using Fluoromount-G (Thermo Fisher Scientific) as mounting medium. Confocal microscopy (Nikon eclipse C2+) was used to obtain micrographs of both ventral horns per section. Alexa-488 staining area in ventral horn was quantified using ImageJ (NIH, Bethesda, Maryland).
Automatic background subtraction was performed using ImageJ custom macro for all GFAP or Iba1 set of images. This macro was set up using symptomatic SOD1 G93A and non-Tg technical control L5 sections. Non-transgenic (Non-Tg) animals were plotted as control (n = 6-15 female mice per genotype). d Body weight measurements. Statistical analysis was performed using twoway ANOVA with Tukey's multiple comparison test. Mean  S.E. is shown (n = 6-14 male mice per genotype; n = 7-16 female mice per genotype). e Rotarod test performance. Statistical analysis was performed using two-way ANOVA with Tukey's multiple comparison test. Mean  S.E. is shown (n = 6-12 male mice per genotype; n = 7-14 female mice per genotype). f Clinical score analysis. Statistical analysis was performed using two-way ANOVA with Tukey's multiple comparison test. Mean  S.E. is shown (n = 6-14 male mice per genotype; n = 7-16 female mice per genotype).     Each box shows minimum, maximum and median values for each genotype. Differences in protein levels are shown relative to non-transgenic littermates (non-Tg). Gene Ontology term (GO-term) corresponds to biological process category assigned to each hit. Statistical analysis was performed using multiple t-test with two-stage step-up method using Benjamini, Krieger and Yekutieli approach with a False Discovery Rate of 5%. Asterisk (*) indicates hits with q-value ≤ 0.05 compared to non-Tg group (n = 3-4 animals per genotype).