Analysis of the mitochondrial proteome in multiple sclerosis cortex

Abstract

Mitochondrial dysfunction has been proposed to play a role in the neuropathology of multiple sclerosis (MS). Previously, we reported significant alterations in the transcription of nuclear-encoded electron transport chain genes in MS and confirmed translational alterations for components of Complexes I and III that resulted in reductions in their activity. To more thoroughly and efficiently elucidate potential alterations in the expression of mitochondrial and related proteins, we have characterized the mitochondrial proteome in postmortem MS and control cortex using Surface-Enhanced Laser Desorption Ionization Time of Flight Mass Spectrometry (SELDI-TOF-MS). Using principal component analysis (PCA) and hierarchical clustering techniques we were able to analyze the differential patterns of SELDI-TOF spectra to reveal clusters of peaks which distinguished MS from control samples. Four proteins in particular were responsible for distinguishing disease from control. Peptide fingerprint mapping unambiguously identified these differentially expressed proteins. Three proteins identified are involved in respiration including cytochrome c oxidase subunit 5b (COX5b), the brain specific isozyme of creatine kinase, and hemoglobin β-chain. The fourth protein identified was myelin basic protein (MBP). We then investigated whether these alterations were consistent in the experimental autoimmune encephalomyelitis (EAE) mouse model of MS. We found that MBP was similarly altered in EAE but the respiratory proteins were not. These data indicate that while the EAE mouse model may mimic aspects of MS neuropathology which result from inflammatory demyelinating events, there is another distinct mechanism involved in mitochondrial dysfunction in gray matter in MS which is not modeled in EAE.

Figure 1. Multivariate analysis of cohort 1, fractions 3 and 6 and cohort 2 fraction 6 using principal component analysis and hierarchical clustering

Multivariate analyses including hierarchical clustering and principal component score scatter plots show the segregation of MS and control samples. Panel A: Cohort 1 fraction 3 analysis. Panel B: Cohort 1 Fraction 6 analysis. Panel C: Cohort 2 fraction 6 analysis. Peaks identified by peptide fingerprint mapping are indicated with asterisks.

Figure 1. Multivariate analysis of cohort 1, fractions 3 and 6 and cohort 2 fraction 6 using principal component analysis and hierarchical clustering

Multivariate analyses including hierarchical clustering and principal component score scatter plots show the segregation of MS and control samples. Panel A: Cohort 1 fraction 3 analysis. Panel B: Cohort 1 Fraction 6 analysis. Panel C: Cohort 2 fraction 6 analysis. Peaks identified by peptide fingerprint mapping are indicated with asterisks.

Figure 2. Proteomic differential expression of identified proteins

SELDI-TOF mass spectra showing the differential expression of a set of peaks. Panel A: The 16 kDa peak was identified as hemoglobin β chain, Panel B: The 42kDa peak was identified as Creatine Kinase type B, Panel C: The 10.6kDa peak was identified as COX5b, Panel D: The 16kDa peak was identified as MBP. Mass spectra of mitochondrial samples outlined in red are control samples and those outlined in blue are MS samples.

Figure 3. Confirmation of the identity of differentially expressed proteins

A. Mitochondrially enriched protein solutions were incubated with an antibody to COX5b covalently bound to beads. Aliquots analyzed by SELDI-TOF-MS before and after incubation confirm the identity of the mass spectral peak at 10.6 kDa as COX5b. The red box highlights the region of interest for these spectra. B. Representative western blot demonstrating the relative purity of the cellular fractionation. Western blots were performed on cytoplasmic (cyto) and mitochondrial (mito) fractions isolated from MS and control cortex, run side by side and blotted with an antibody to the neuron specific protein, neurofilament (NF), and an antibody to the mitochondrial encoded COX2 protein. Multiple NF immunoreactive proteins are denoted by arrows. C. Representative western blots show increased MBP, hemoglobin β (Hbb), and creatine kinase (CKB) and decreased COX5b in mitochondrial fractions isolated from MS cortex. Increased expression of MBP was also observed in the EAE mouse brain while CKB, Hbb, and COX5b were not similarly altered in EAE and MS. To control for protein loading, western blots were reprobed with antibodies to either Complex II (CII) or COX2. D. Representative western blots of hemoglobin β (Hbb) and creatine kinase B (CKB) in cytoplasmic fractions isolated from MS and control cortex and also from EAE and control brains or cortex with GAPDH as a loading control. E. Quantitation was done for MBP, COX5b, Hbb, and CKB expression for MS and control samples and for EAE cortex. Data is expressed as percent of control for MS and EAE samples and densitometry was standardized to either CII or COX2 for mitochondrial fractions or to GAPDH for cytoplasmic fractions. Error bars represent SEM. * p≤ 0.05

Figure 4. Increased immunofluorescent hemoglobin staining in MS parietal cortex pyramidal cell layer

A. Confocal images taken at 20X of hemoglobin staining in control and MS postmortem tissue. Micrographs are from control (left column) and MS (right column) postmortem tissue sections stained for hemoglobin (Hbb) (top row) and neurofilament (NF) (middle row). Hemoglobin can be seen to be localized in a number of neurons, denoted by white arrows, including pyramidal cells, colocalized with the neurofilament stain. The last row displays the 2 channels overlayed with the organization of the pyramidal cell layers clearly visible. B. Images are z-projected multichannel confocal micrographs from gray matter from control and MS parietal cortex. Control (row 1) and MS (row 2) tissue sections were stained for neurofilament (column 1) and hemoglobin (column 2), and image stacks acquired from the pyramidal cell layer using a 60X oil objective and identical acquisition parameters. Note increased hemoglobin staining intensity in the MS sample in agreement with the reported SELDI-TOF-MS results. The third column depicts the hemoglobin (red fluorescence) and neurofilament (green fluorescence) data overlayed, and clearly displays hemoglobin fluorescence within pyramidal cell somas. The scale bar represents 100 µm.

Figure 4. Increased immunofluorescent hemoglobin staining in MS parietal cortex pyramidal cell layer

A. Confocal images taken at 20X of hemoglobin staining in control and MS postmortem tissue. Micrographs are from control (left column) and MS (right column) postmortem tissue sections stained for hemoglobin (Hbb) (top row) and neurofilament (NF) (middle row). Hemoglobin can be seen to be localized in a number of neurons, denoted by white arrows, including pyramidal cells, colocalized with the neurofilament stain. The last row displays the 2 channels overlayed with the organization of the pyramidal cell layers clearly visible. B. Images are z-projected multichannel confocal micrographs from gray matter from control and MS parietal cortex. Control (row 1) and MS (row 2) tissue sections were stained for neurofilament (column 1) and hemoglobin (column 2), and image stacks acquired from the pyramidal cell layer using a 60X oil objective and identical acquisition parameters. Note increased hemoglobin staining intensity in the MS sample in agreement with the reported SELDI-TOF-MS results. The third column depicts the hemoglobin (red fluorescence) and neurofilament (green fluorescence) data overlayed, and clearly displays hemoglobin fluorescence within pyramidal cell somas. The scale bar represents 100 µm.