Proteomic analysis of infiltrating neutrophils from rheumatoid arthritis synovial fluid and their contribution to protein carbamylation

Abstract

Introduction: Carbamylated proteins and dysregulated neutrophils are implicated in rheumatoid arthritis (RA) pathogenesis. Herein, we characterized the neutrophils present in RA synovial fluid (SF) using proteomic techniques and evaluated their contribution to protein carbamylation.

Methods: RA-SF neutrophil proteomic profile and SF proteome signature were investigated using high-resolution mass spectrometry. Carbamylated proteins and the degree of protein carbamylation were evaluated by mass spectrometric analysis. ELISA and chemiluminescence kits were used to examine myeloperoxidase (MPO) activity, and hydrogen peroxide (H₂O₂) generation.

Results: SF neutrophils exhibited a shift in proteomic cargo with up-regulated proteins involved in defense responses, neutrophil degranulation, and reactive oxygen species metabolic processes, while proteins down-regulated were associated with megakaryocyte differentiation, leukocyte migration, and integrin-mediated signaling pathway. Elevated levels of neutrophil-derived proteins were detected in RA-SF. In addition, we specifically identified many carbamylated proteins and observed an increased frequency of protein carbamylation in RA-SF samples. Functionally, neutrophils from RA-SF showed a significantly increased level of MPO release and HH₂O₂ generation. Moreover, MPO activity was higher in RA-SF than in autologous blood samples, which correlated well with the degree of protein carbamylation in RA-SF.

Discussion: Synovial neutrophils were found to be activated and increased releasing protein cargo, including MPO and ROS, into the synovial fluid. Presence of many carbamylated proteins in RA-SF and an increased MPO activity showed a strong correlation to the degree of protein carbamylation, suggesting neutrophil-derived MPO in promoting generation of aberrantly carbamylated proteins.

Keywords: carbamylation; myeloperoxidase; neutrophil; rheumatoid arthritis; synovial fluid.

Figure 1

Differential protein expression in neutrophils from RA-SF versus blood. (A) Volcano plots illustrating differential protein abundance between neutrophils from RA-SF versus blood. Vertical and gray lines indicate the limit where proteins passed a significance value. (B) Interaction network of significant proteins from neutrophils from RA-SF versus autologous blood, developed using the online platform STRING. (C, D) Histogram showing log2 fold changes of individual protein clusters indicated in (B, E, F) Gene set enrichment analysis of the proteome of neutrophils from RA-SF versus blood. Normalized enrichment scores (NES) were displayed in different sizes to reveal the FDR-adjusted p-values using different colors to indicate the protein ratio within the indicated pathway.

Figure 2

Analysis of neutrophil-derived proteins in RA synovial fluid. (A) Comparison of number of proteins identified in RA-SF versus autologous or HC plasma. (B) Gene Ontology analysis of cellular component categorization of unique proteins identified in RA-SF samples. The y-axis and x-axis show the number of identified proteins and the False Discovery Rate (FDR) p-value, respectively. (C) Comparison of the abundance of neutrophil-derived protein expression in RA-SF and plasma samples from autologous RA and HC. Protein abundance was estimated by log10 values of absolute iBAQ intensities. (D) Coverage analysis of significantly upregulated neutrophil proteins in synovial fluid. (E) Peptide-spectrum match (PSM) analysis of significantly upregulated neutrophil proteins in synovial fluid. (n = 12/group, **p < 0.01, unpaired T-test). (DEFA1, defensin1; MMP3, stromelysin-1; LTF, lactotransferrin; ELANE, elastase; LCN2, neutrophil gelatinase-associated lipocalin; CRTAC1, cartilage acidic protein 1; COMP, cartilage oligomeric matrix protein; CTSG, neutrophil serine proteases cathepsin G; ACAN, aggrecan core protein; PADI4, protein-arginine deiminase type-4).

Figure 3

Detection of carbamylated peptides and quantification of albumin carbamylation. (A) A typical extracted ion chromatogram of albumin K236 peptide with and without carbamylation. (B) MS/MS spectra of albumin K236 peptide containing either unmodified or carbamylated K236 residue. The black arrow depicts specific neutral loss (NL) of isocyanic acid from b3, b4, and y7 fragmented ion; the red K depicts carbamylation. (C) Estimate of the relative abundance of albumin carbamylation sites. The percentage was calculated as the intensities of carbamylated peptides divided by the intensities of extracted ion chromatograms of parent peptide, independent of modifications. Values are expressed as median + 95% confidence interval (n = 20, *, p < 0.05, **, p < 0.01).

Figure 4

Evaluating the correlation between albumin carbamylation and MPO activity in RA-SF. (A, B) Detection of MPO activity and ROS production by different subsets of unstimulated (A) or fMLP-stimulated (B) neutrophils. Cell supernatants were collected after 2 h of culture (n = 12) and used for analyzing MPO activity. ROS production was analyzed immediately by adding an assay mixture as described in the materials and methods. (C) Detection of MPO activity in freshly collected RA-SF and plasma samples from RA and HC. Values in (A-C) are expressed as median + 95% confidence interval. (D-F) Assessment of correlation between the frequency of albumin carbamylation and MPO activity (D), MPO activity and neutrophil count (E), and the frequency of albumin carbamylation and neutrophil count in RA-SF (F). Spearman correlation coefficients were calculated to detect potential correlations. *, p < 0.05; **, p < 0.01; NE: neutrophils.