Decreased Levels of Foldase and Chaperone Proteins Are Associated with an Early-Onset Amyotrophic Lateral Sclerosis

Abstract

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterized by a progressive upper and lower motor neuron degeneration. One of the peculiar clinical characteristics of ALS is the wide distribution in age of onset, which is probably caused by different combinations of intrinsic and exogenous factors. We investigated whether these modifying factors are converging into common pathogenic pathways leading either to an early or a late disease onset. This would imply the identification of phenotypic biomarkers, that can distinguish the two populations of ALS patients, and of relevant pathways to consider in a therapeutic intervention. Toward this aim a differential proteomic analysis was performed in peripheral blood mononuclear cells (PBMC) from a group of 16 ALS patients with an age of onset ≤55 years and a group of 16 ALS patients with an age of onset ≥75 years, and matched healthy controls. We identified 43 differentially expressed proteins in the two groups of patients. Gene ontology analysis revealed that there was a significant enrichment in annotations associated with protein folding and response to stress. We next validated a selected number of proteins belonging to this functional group in 85 patients and 83 age- and sex-matched healthy controls using immunoassays. The results of the validation study confirmed that there was a decreased level of peptidyl-prolyl cis-trans isomerase A (also known as cyclophilin A), heat shock protein HSP 90-alpha, 78 kDa glucose-regulated protein (also known as BiP) and protein deglycase DJ-1 in PBMC of ALS patients with an early onset. Similar results were obtained in PBMC and spinal cord from two SOD1^G93A mouse models with an early and late disease onset. This study suggests that a different ability to upregulate proteins involved in proteostasis, such as foldase and chaperone proteins, may be at the basis of a different susceptibility to ALS, putting forward the development of therapeutic approaches aiming at boosting the protein quality control system.

Keywords: biomarkers; chaperone; foldase; protein folding; response to stress.

FIGURE 1

Identification of phenotypic biomarkers in PBMC of early and late ALS patients. (A) Scheme of the strategy to identify phenotypic protein biomarkers in PBMC of patients early onset ALS (EA) and late-onset ALS (LA) and matched controls, early (EC), and late (LC). 2D-DIGE analysis was done with 16 pooled samples for each experimental group and spot volume values for EA and LA were normalized to EC and LC, respectively (nEA, nLA). Spots were considered differential if fold change (nLA/nEA) was ≤0.8 or ≥1.3 and identified by mass spectrometry (MS, Supplementary Table S1). Differential proteins were classified on the basis of GO annotations and a selected number of proteins associated with “protein folding and response to stress” biological function was validated by dot blot immunoassays. (B) Representative Sypro-Ruby stained 2D gel of the PBMC proteome. The numbered spots correspond to the differential proteins listed in Table 3 and Supplementary Table S1.

FIGURE 2

Validation of candidate phenotypic biomarkers in PBMC of early and late ALS patients. (A–H) PPIA, HSP90, GRP78, ERp57, DJ-1, HSC70, and TDP-43 were analyzed by dot blot immunoassays in PBMC samples from an independent set of ALS patients (n = 85), n = 38 EA and n = 47 LA, and matched controls (n = 83), n = 35 EC and n = 48 LC. Immunoreactivity was normalized to protein loading, as assessed by Ponceau Red staining, and then to the mean values of matched controls (A–F,H). We found that PPIA, HSP90, GRP78, and DJ-1 were significantly lower in early ALS than in late ALS (A–C,E) and that TDP-43 was higher in both early and late ALS compared to matched controls (G). ^∗p < 0.05, by Student’s t-test.

FIGURE 3

Analysis of phenotypic biomarkers in PBMC of SOD1^G93A mouse models with early (129Sv) and late (C57) disease onset. (A–F) PPIA, HSP90, GRP78, ERp57, DJ-1, and TDP-43 were analyzed by dot blot immunoassays in PBMC samples (n = 5 per group) from 129Sv SOD1^G93A and C57 SOD1^G93A (G93A) mice at disease onset, respectively, at 14 and 17 weeks of age, and matched non-transgenic controls (Ntg). Immunoreactivity was normalized to protein loading, as assessed by Ponceau Red staining, and then to the mean values of matched controls. We found that PPIA, HSP90 were significantly lower in early ALS than in late ALS (A,B) similarly to human samples. ^∗p < 0.05, by Student’s t-test.

FIGURE 4

Analysis of phenotypic biomarkers in lumbar spinal cord of SOD1^G93A mouse models with early (129Sv) and late (C57) disease onset. (A–F) PPIA, HSP90, GRP78, ERp57, DJ-1, and TDP-43 were analyzed by dot blot immunoassays in lumbar spinal cord samples (n = 5 per group) from 129Sv SOD1^G93A and C57 SOD1^G93A (G93A) mice at disease onset, respectively, at 14 and 17 weeks of age, and matched non-transgenic controls (Ntg). Immunoreactivity was normalized to protein loading, as assessed by Ponceau Red staining, and then to the mean values of matched controls. We found that PPIA, HSP90, GRP78, and DJ-1 were significantly lower in early ALS than in late ALS (A–C,E) similarly to PBMC from patients. ^∗p < 0.05, by Student’s t-test.