Profiling of the Peripheral Blood Mononuclear Cells Proteome by Shotgun Proteomics Identifies Alterations of Immune System Components, Proteolytic Balance, Autophagy, and Mitochondrial Metabolism in Glaucoma Subjects

Abstract

Glaucoma is a chronic optic neuropathy and is the second cause of irreversible blindness worldwide. Although the pathogenesis of the disease is not fully understood, the death of retinal ganglion cells and degeneration of the optic nerve are likely promoted by a combination of local and systemic factors. Growing attention has been paid to nonintraocular pressure risk factors, including mechanisms of inflammation and neuroinflammation. Phenotypical and molecular alterations of circulating immune cells, in particular, lymphocyte subsets, have been documented in murine models of glaucoma and in human subjects. Very recently, oxygen consumption rate and nicotinamide adenine dinucleotide levels of human peripheral blood mononuclear cells (PBMC) have been proposed as biomarkers of disease progression, thus suggesting that immune cells of glaucoma subjects present severe molecular and metabolic alterations. In this framework, this pilot study aimed to be the first to characterize global proteome perturbations of PBMC of patients with primary open-angle glaucoma (POAG) compared to nonglaucomatous controls (control) by shotgun proteomics. The approach identified >4,500 proteins and a total of 435 differentially expressed proteins between POAG and control subjects. Clustering and rationalization of proteomic data sets and immunodetection of selected proteins by Western blotting highlighted significant alterations of immune system compartments (i.e., complement factors, regulators of immune functions, and lymphocyte activation) and pathways serving key roles for immune system such as proteolysis (i.e., matrix metalloproteinases and their inhibitors), autophagy (i.e., beclin-1 and LC3B), cell proliferation (Bcl2), mitochondrial (i.e., sirtuin), and energetic/redox metabolism (i.e., NADK). Based on these findings, this proteomic study suggests that circulating immune cells suffer from heterogeneous alterations of central pathways involved in cell metabolism and homeostasis. Larger, properly designed studies are required to confirm specifically how immune cellular alterations may be involved in the pathogenesis of both neuroinflammation and glaucomatous disease.

Figure 1

Schematic representation of the study design and experimental workflow.

Figure 2

(A) Christmas tree plot: protein intensities were log₂ transformed in the POAG/control ratio and plotted against the abundance weights, a parameter which identifies the log₂ (intensities of POAG proteins + intensities of control proteins); (B) violin plots reporting the coefficient of variation (CV) of proteins identified in POAG and control PBMCs before any further normalization; (C) contamination of blood proteins in PBMC proteome was calculated; (D) principal component analysis (PCA) of POAG and control proteomes. The centroid for both experimental groups is indicated by a blue triangle (POAG) and a green dot (control).

Figure 3

(A) Venn diagram showing the proteins identified in POAG and control experimental groups; (B) volcano plot showing the upregulated (red) and downregulated (turquoise) proteins in the POAG/control ratio. Dashed lines set the threshold used for labeling a protein as differentially represented. Filters were set as 0.5 ≤ log₂FC ≤ 0.5 and p ≤ 0.05 after Benjamini–Hochberg correction. This last value is graphed as −log₁₀(p-value_adj) along the y axis. FC: fold change.

Figure 4

(A) Molecular function (MF) and (B) biological processes (BP) terms found enriched by submitting proteins upregulated in POAG PBMCs to GO. GeneRatio was calculated and data filtered for p ≤ 0.05; (C) nodes and edges found as enriched by the KEGG pathway (p ≤ 0.05).

Figure 5

(A) Molecular function (MF) and (B) biological processes (BP) terms found to be enriched by submitting proteins downregulated in POAG (labeled as enriched in control PBMCs). GeneRatio was calculated and data filtered for p ≤ 0.05. GO term “adaptive response based on somatic recombination of immune receptors built from immunoglobulin superfamily domains” is replaced by “somatic hypermutations” to optimize image resolution.

Figure 6

Western blotting panel analysis of selected proteins of n = 26 POAG and n = 26 control subjects. A representative blot is shown. Dashed lines in the TIMP-3 and MMP-9 blots indicate that the original image was cropped. Uncropped images are provided in the Figure S4. Total proteins as stained by Ponceau S were used for normalization (see the Materials and Methods for further details). The box plots show the median and min max values of protein intensities as fold change with respect to nonglaucomatous control subjects. Unpaired τ test was used for statistical analysis after having verified a normal distribution of data (Shapiro–Wilk test); *p ≤ 0.05; **p ≤ 0.01. The box plot and statistical analysis are not provided for TIMP-3, MMP-9, and LC3B as the proteins turned out to be below the detection limit for one of the two experimental groups. Regarding LC3B, note that the proteins migrate as a doublet band (16–14 kDa). LC3B–II (14 kDa), which corresponds to LC3B–I (16 kDa) conjugated with phosphatidyl ethanolamine (PE), migrates faster due to its higher hydrophobicity. Lipidation of LC3B–I occurs during autophagy activation.

Figure 7

Plots showing the relationship between intensities of Bcl2, beclin-1 and NADK determined by Western blotting analysis and main ocular, demographic, and systemic parameters for control (A) and POAG (B) subjects, namely, age, number (N) of systemic diseases, therapy with ACE inhibitors, β (B)-blockers or oral acetazolamide, mean deviation (dB) of visual field loss, number (N) of ocular hypotensive medications (OHMs), mean IOP (mmHg), glaucoma surgery. The intervals of confidence are reported.