Table 1.

Proteomic analysis of saliva, salivary gland tissue and/or blood from patients with pSS and healthy control (HC) subjects and/or non-pSS, sicca control subjects.

Author (year)	No. of subjects	Materials and methods	Findings
Ryu OH, et al., 2006 (30)	41 pSS 15 non-pSS, sicca 5 HC	PS 2-DGE, SELDI-TOF-MS, ELISA	SELDI-TOF-MS of 10-200 kDa peaks revealed 8 peaks with >2-fold changes in the SS group that differed from non-SS (p< 0.005). Peaks of 11.8, 12.0, 14.3, 80.6 and 83.7 kDa were increased, while 17.3, 25.4, and 35.4 kDa peaks were decreased in SS samples. 2D-DIGE identified significant ↑ of β2-microglobulin, lactoferrin, immunoglobulin (Ig) kappa light chain, polymeric Ig receptor, lysozyme C and cystatin C in all stages of SS. Two presumed proline-rich proteins, amylase and carbonic anhydrase VI, were ↓ in the patient group.
Hu S, et al., 2007 (23)	10 pSS 10 HC	WS, PS, sublingual/sub-mandibular saliva MALDI-TOF-MS, 2-DGE, LC-MS/MS, RT-PCR, immunoblotting	Sixteen salivary proteins were ↓ and 25 were ↑ in pSS patients compared with HC. In pSS, the levels of 16 peptides (10 ↑ and 6 ↓) differed significantly from those of HC. WS contained more informative peptides and proteins than gland-specific saliva. Moreover, 27 mRNAs were significantly up-regulated (≥3-fold change) in saliva from pSS. In pSS, 19 of 27 overexpressed genes were interferon-inducible or related to lymphocyte filtration and antigen presentation, known to be involved in the pathogenesis of pSS.
Giusti L, et al., 2007 (31)	12 pSS 12 HC	UWS 2-DGE	The WS protein pattern differed between pSS patients and healthy controls, particularly carbonic anhydrase VI. A comparison of WS protein profile of pSS patients compared with the one obtained from HC revealed a set of differentially expressed proteins. These proteins were related to acute and chronic inflammation while some others were involved in oxidative stress injury.
Hjelmervik T, et al., 2009 (45)	6 pSS 6 non-SS controls	MSG tissue 2-DGE, LC-ESI-MS/MS	A total of 522 proteins were identified in pSS and non-SS controls, out of which 158 in pSS only, 91 in non‐SS controls only and 273 in both groups. Heat shock proteins, mucins, carbonic anhydrases, enolase, vimentin and cyclophilin B were among the proteins identified. The differences in the proteomes of MSG tissue from pSS patients and non‐SS controls were mainly related to ribosomal proteins, immunity and stress. Alpha‐defensin‐1 and calmodulin were among six proteins exclusively identified in pSS patients. Overall, proteins related to matrix, metabolism and stress were ↓, while IgG and serum albumin were ↑ in pSS.
Wei P, et al., 2013 (33)	12 pSS 13 HC	UWS MALDI-TOF-MS	The study investigated differences in low molecular weight salivary proteins (1-10 kDa) in pSS patients compared to HC. Seven m/z (mass-to-charge) (m/z 1068.1, 1196.2, 1738.4, 3375.3, 3429.3, 3449.7 and 3490.6 ratio peaks with significant differences between the 2 groups with 5 peptides being up-regulated and 2 down-regulated in the pSS patients compared to HC subjects. In the validation phase, 4 out of 5 pSS patients were diagnosed as pSS, and 4 of the 5 healthy controls were diagnosed as HC.
Gallo A, et al., 2013 (34)	82 pSS 17 non-pSS sicca 29 HC	UWS SELDI-TOF-MS, 2DE, MALDI-TOF-MS, immunohistochemistry	Gross cystic disease fluid protein-15(GCDFP-15)/prolactin-inducible protein (PIP) is a secretory acinar glycoprotein of 14 KDa. Proteomic analysis found that a putative peak of 16547 m/z, identified as the GCDFP-15/PIP protein, was among the best independent biomarkers for pSS to discriminate between patients and HC with a sensitivity of 96% and a specificity of 70%, with a global cross validated error of 29%. The intensity of GCDFP-15/PIP was significantly lower in pSS patients than in non-pSS sicca subjects and HC. GCDFP-15/PIP expression also correlated with salivary flow rate and focus score. Immunohistochemistry confirmed that GCDFP-15/PIP staining was faint in mucus acini, but no obvious difference were in serous acinar cells. RT-PCR showed that GCDFP-15/PIP mRNA was significantly lower in pSS patients than in non-pSS sicca patients and HC, supporting the hypothesis that the reduction of GCDFP-15/PIP is related to a decreased protein synthesis.
Deutsch O, et al., 2015 (35)	18 pSS 18 HC	UWS Affinity and immunodepletion, 2-DGE, LC-MS/MS	The use of depletion strategy before proteomics analysis increased identification ability by 3-fold. Overall, 79 biomarker candidates were identified. Proteins with the most pronounced fold changes were related to SS serum or tissue factors. Bioinformatics analyses of proteins with a >3-fold increase in SS patients showed calcium-binding proteins, defense-response proteins, proteins involved in apoptotic regulation, stress-response proteins and cell motion-related proteins. Preliminary validation by western blotting of profilin and CA-I indicated similar expression profile trends to those identified by quantitative MS.
Nishikawa A, et al., 2016 (52)	82 pSS (30 + 52) 30 HC	Serum High-throughput proteomic analysis, ELISA	82 proteins, identified as pSS-associated, were differentially expressed between 30 pSS patient and 30 HC (57 were ↑, 25 were ↓). Nine were identified as disease activity-associated biomarkers and among these, five proteins (CXCL13, TNF-R2, CD48, BAFF, and PD-L2) were validated as candidate biomarkers by ELISA in a separate pSS-validation cohort. CXCL13 exhibited the most significant correlation with the lymphadenopathy, glandular, and pulmonary domains of the ESSDAI. CXCL13, TNF-R2 and CD48 exhibited a positive correlation with the biological domain of the ESSDAI. TNF-R2 exhibited the most negative correlation with uptake in the submandibular gland.
Dalaleu N, et al., 2016 (37)	48 pSS	UWS 187-plex capture antibody-based assay	The study identified disease-phenotype driven biomarker signatures. Hyposalivation was associated with significant alteration in 22 out of 119 reliably detectable biomarkers. Thereof, a 4-plex signature allowed accurate prediction of salivary gland function for >80% of the cases. With respect to histopathological features, the most distinct profiles were identified in conjunction with ectopic germinal centers. Pregnancy-associated plasma protein A, thrombospondin 1 and peptide YY could recapitulate the presence or absence of tertiary lymphoid organization for 93.8% of the patients. Functional annotation of alterations associated with hyposalivation identified the IL1 system as a dominant pro-inflammatory component, while changes observed in context with ectopic lymphoid organization revealed specific shifts in chemotactic profiles and altered regulation of apoptotic processes.
Chaudhury NMA, et al., 2016 (38)	25 pSS 35 HC	UWS Western blot, LC-MS/MS	The concentration of mucins MUC5B and MUC7 were similar between patients and controls, but a comparison of protein Western blotting and glycan staining identified a ↓ in mucin glycosylation in SS, particularly on MUC7. Analysis of O-glycans released from MUC7 by β-elimination revealed an increase in core 1 sulfation, but an even larger reduction in sialylation resulting in a global decline of charged glycans. This was primarily due to the loss of the extended core 2 disialylated structure, with and without fucosylation. A reduction in the extended, fucosylated core 2 disialylated structure on MUC7, residual mucosal wetness, and whole saliva flow rate appeared to have a negative and cumulative effect on the perception of oral dryness.
Aqrawi LA, et al., 2017 (25)	27 pSS 32 HC	SWS, EV, tears LC-MS	SWS: ↑ of neutrophil gelatinase-associated lipocalin (LCN2, innate immunity), calmodulin (CALM, cell signalling) and calmodulin-5 (CALM-5, wound repair), granulins (GRN, wound repair) and epididymal secretory protein-1 (ESP1, cholesterol homeostasis within the endosome and/or lysosome) in pSS. SWS EV: ↑ adipocyte plasma membrane-associated protein (APMAP, adipocyte differentiation), guanine nucleotide-binding protein subunit alpha-13 (GNA-13, cell signalling), WD repeat-containing protein-1 (WDR1, involved in the disassembly of actin filaments), tyrosine-protein phosphatase non-receptor type substrate-1 (SIRPA, regulates NK cells and dendritic cell inhibition) and lymphocyte-specific protein 1 (LSP1, activation of innate immune system) in pSS.
Hall SC, et al., 2017 (40)	15 pSS 15 non-pSS sicca 14 HC	USW, PS iTRAQ, Lectin affinity capture-MS	The iTRAQ analyses revealed up- and down-regulation of numerous proteins that could be involved in the disease process (e.g., histones) or attempts to mitigate the ensuing damage (e.g., bactericidal/permeability increasing fold containing family (BPIF) members). An immunoblot approach confirmed the pSS-associated ↑ of β2-microglobulin (in PS) and ↓ of carbonic anhydrase VI (in WS) and BPIFB2 (in PS). Beyond the proteome, the N-glycosites of pSS and HC samples were profiled. They were enriched for glycopeptides using lectins Aleuria aurantia and wheat germ agglutinin, which recognize fucose and sialic acid/N-acetyl glucosamine, respectively. MS analyses showed that pSS is associated with increased N-glycosylation of numerous salivary glycoproteins in PS and WS.
Tasaki S, et al., 2017 (54)	30 pSS 30 HC	Whole blood transcriptomes, serum proteomes and peripheral immuno-phenotyping SOMAmer-based capture array	Identification of SS gene signatures (SGS) dysregulated in widespread omics layers, including epigenomes, mRNAs and proteins. SGS predominantly involved the interferon signature and ADAMs substrates. Besides, SGS was significantly overlapped with SS-associated genes indicated by a genome-wide association study and expression trait loci analyses. Combining the molecular signatures with immunophenotypic profiles revealed that cytotoxic CD8 T cells were associated with SGS. Activation of SGS in cytotoxic CD8 T cells isolated from patients with pSS was observed.
Bodewes ILA, et al., 2019 (57)	63 pSS (including 22 with fatigue and 23 without fatigue 20 HC	Serum SOMAscan, ELISA	pSS vs. HC: 58 proteins were ↑ and 46 were ↓. IFN-pos. and IFN-neg. pSS patients: ↑ IgG levels, higher frequency of anti-SSA and -SSB and ↓ levels of C3 complement in IFN-pos. pSS patients. 22 fatigued vs. 23 non-fatigued pSS patients: 14 serum proteins were ↑ including SNAP-25, ENO1, UCHL1 and 2 serum proteins were ↓ in fatigued. IL36a and several complement factors were upregulated in fatigued pSS patients compared to non-fatigued.
Aqrawi LA, et al., 2019 (21)	10 pSS 15 non-pSS, sicca 10 HC	SWS, tears LC-MS	SWS from pSS vs. non-pSS sicca: ↑ of peptidyl-prolyl cis-trans isomerase FKB1A (T-cell regulation, upregulation of NF-kappa-B signaling), CD44 antigen (FOXP3 expression and regulatory T-cell suppresion), β-2 microglobulin (B2MG, innate immunity). SWS from pSS vs. HC: ↑ Secreted Ly-6/uPAR-related protein 1 ( Ach-receptor activity, cell migration and proliferation), B2MG, Clusterin (Innate immunity, modulates NF-kappa-B activity and TNF production). SWS-EV from pSS vs. non-pSS sicca: CD44, Major vault protein (IFN-mediated signaling, MAP kinase activity, neutrophil degranulation), Neutrophil gelatinase-associated lipocalin (Innate immunity, tumor-associated antigen, cell adhesion). SWS-EV from pSS vs. HC: Ficolin-1 (Innate immunity), CD44, Annexin A4 (NF-kappa-B, apoptosis, IL-8 secretion).
Qiao L, et al., 2019 (58)	26 pSS with NMOSD 34 pSS without NMOSM 30 NMOSD	Serum 2-DGE, MALDI-TOF/MS, ELISA	Nine proteins were significantly differently expressed between pSS with NMOSD and pSS without NMOSD). Serum levels of clusterin and complement factor H (CFH) were further verified by ELISA. Serum clusterin was higher in NMOSD with pSS than without (298.33 ± 184.52 vs. 173.49 ± 63.03 ng/ml, p < 0.01), while the levels of CFH were lower in pSS patients with NMOSD than without.
Cecchettini A, et al., 2019 (43)	20 pSS 20 HC	UWS Nano-HPLC-SWATH-MS	PSS patients were stratified in three subgroups according to focus score in MSG tissue and unstimulated salivary flow. 203 proteins were differentially expressed in pSS patients compared to HC with evident differences in the expression of normal constituents of the human salivary proteome (i.e. prolactin-inducible protein, proline-rich proteins, and cystatins) and several mediators of inflammatory processes. 63 proteins were shared and specifically modulated in the three subsets of pSS patients converging on several inflammatory pathways. Among them S100A protein appeared of particular interest merging on IL-12 signaling and being significantly influenced by either salivary flow impairment or intensity of immune cell infiltration in the tissue.
Sembler-Møller, et al., 2020 (22)	24 pSS 16 non-pSS sicca	SWS, plasma, MSGB LC-MS	1013 proteins were detected in SWS, 219 proteins in plasma and 3166 in MSGT. SWS: 40 proteins differed significantly between the two groups, 34 of which were ↑ in the pSS group. In pSS, proteins involved in immunoinflammatory processes were ↑, whereas proteins related to salivary secretion were ↓. The combination of neutrophil elastase, calreticulin and TRIM-29 revealed the best performance in distinguishing pSS from non-pSS with an AUC of 0.97.
Huang S, et al., 2020 (60)	8 pSS 10 HC	PBMC Immobilized metal affinity chromatography (IMAC)	787 proteins were identified as differentially expressed proteins, and 175 phosphosites on 123 proteins were identified as differentially phosphorylated proteins. Using module and hub protein analyses, 16 modules for the proteins were identified, 2 clusters for the phosphoproteins and selected the top 10 hub proteins (UBA52, MAPK1, HSPA8, ACTB, ACLY, APP, ACTN1, ACTN4, VWF, and TGFB1). Finally, 22 motifs were identified using motif analysis of the phosphosites and found 17 newly identified motifs, while 6 motifs were experimentally verified for known protein kinases.
Tian Q, et al., 2020 (61)	133 pSS 108 HC	PBMC ELISA, Immunofluorescence staining, tranfection assay, and IFN-alfa treatment.	The ↑ of PARP-9 and CXCL10 as well as their co-localization was confirmed in pSS. PARP-9 levels in LSGs ↑ with ↑ focus scores in patients with pSS. PARP-9 and DTX3L were present in the infiltrating mononuclear cells from salivary glands in female NOD/LtJ mouse models. Ingenuity Pathway Analysis networks of differentially expressed proteins demonstrated that PARP-9, STAT1, and IFN-induced protein with tetratricopeptide repeats 1 (IFIT-1) participated in the IFN-related pathway. Furthermore, PARP-9 could ↑ the expression of IFIT1 and CXCL10 in B cells. Moreover, PARP-9 and CXCL10 could be induced by IFNα in B cells.
Schmidt A, et al., 2020 (62)	45 pSS, 30 controls	Plasma Hybrid immunoaffinity targeted LC-MS	The aim was to quantify plasma levels of the chemokine CXCL10, which has been associated with many immunological disorders including pSS. The hybrid approach enabled sensitive, specific, and simultaneous quantification of total, full-length (active) CXCL10 1−77 and DPP4-truncated (inactive) CXCL10 3−77 in human plasma down to the low pg/mL level, reaching ELISA sensitivities. The ratio of CXCL10 1−77 to truncated CXCL10 3−77 was ↑ in patients with pSS and provided the highest correlation with pSS disease activity.

Arrow up: upregulated.

Arrow down: downregulated.