Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Apr 20.
Published in final edited form as: Arthritis Rheum. 2007 Nov;56(11):3588–3600. doi: 10.1002/art.22954

Salivary Proteomic and Genomic Biomarkers for Primary Sjögren’s Syndrome

Shen Hu 1, Jianghua Wang 1, Jiska Meijer 2, Sonya Ieong 1, Yongming Xie 1, Tianwei Yu 1, Hui Zhou 1, Sharon Henry 1, Arjan Vissink 2, Justin Pijpe 2, Cees Kallenberg 2, David Elashoff 1, Joseph A Loo 1, David T Wong 1
PMCID: PMC2856841  NIHMSID: NIHMS191796  PMID: 17968930

Abstract

Objective

To identify a panel of protein and messenger RNA (mRNA) biomarkers in human whole saliva (WS) that may be used in the detection of primary Sjögren’s syndrome (SS).

Methods

Mass spectrometry and expression microarray profiling were used to identify candidate protein and mRNA biomarkers of primary SS in WS samples. Validation of the discovered mRNA and protein biomarkers was also demonstrated using real-time quantitative polymerase chain reaction and immunoblotting techniques.

Results

Sixteen WS proteins were found to be down-regulated and 25 WS proteins were found to be up-regulated in primary SS patients compared with matched healthy control subjects. These proteins reflected the damage of glandular cells and inflammation of the oral cavity system in patients with primary SS. In addition, 16 WS peptides (10 up-regulated and 6 down-regulated in primary SS) were found at significantly different levels (P <0.05) in primary SS patients and controls. Using stringent criteria (3-fold change; P <0.0005), 27 mRNA in saliva samples were found to be significantly up-regulated in the primary SS patients. Strikingly, 19 of 27 genes that were found to be overex-pressed were interferon-inducible or were related to lymphocyte filtration and antigen presentation known to be involved in the pathogenesis of primary SS.

Conclusion

Our preliminary study has indicated that WS from patients with primary SS contains molecular signatures that reflect damaged glandular cells and an activated immune response in this autoimmune disease. These candidate proteomic and genomic biomarkers may improve the clinical detection of primary SS once they have been further validated. We also found that WS contains more informative proteins, peptides, and mRNA, as compared with gland-specific saliva, that can be used in generating candidate biomarkers for the detection of primary SS.

Sjögren’s syndrome (SS), which was first described in 1933 by the Swedish physician Henrik Sjögren (1), is a chronic autoimmune disorder clinically characterized by a dry mouth (xerostomia) and dry eyes (keratoconjunctivitis sicca). The disease primarily affects women, with a ratio of 9:1 over the occurrence in men. While SS affects up to 4 million Americans, about half of the cases are primary SS. Primary SS occurs alone, whereas secondary SS presents in connection with another autoimmune disease, such as rheumatoid arthritis or systemic lupus erythematosus (SLE). Histologically, SS is characterized by infiltration of exocrine gland tissues by predominantly CD4 T lymphocytes. At the molecular level, glandular epithelial cells express high levels of HLA–DR, which has led to the speculation that these cells are presenting antigen (viral antigen or autoantigen) to the invading T cells. Cytokine production follows, with interferon (IFN) and interleukin-2 (IL-2) being especially important. There is also evidence of B cell activation with autoantibody production and an increase in B cell malignancy. SS patients exhibit a 40-fold increased risk of developing lymphoma.

SS is a complex disease that can go undiagnosed for several months to years. Although the underlying immune-mediated glandular destruction is thought to develop slowly over several years, a long delay from the start of symptoms to the final diagnosis has been frequently reported. SS presumably involves the interplay of genetic and environmental factors. To date, few of these factors are well understood. As a result, there is a lack of early diagnostic markers, and diagnosis usually lags symptom onset by years. A new international consensus for the diagnosis of SS requires objective signs and symptoms of dryness, including a characteristic appearance of a biopsy sample from a minor or major salivary gland and/or the presence of autoantibody such as anti-SSA (2–4). However, establishing the diagnosis of primary SS has been difficult in light of its nonspecific symptoms (dry eyes and mouth) and the lack of both sensitive and specific biomarkers, either body fluid– or tissue-based, for its detection. It is widely believed that developing molecular biomarkers for the early diagnosis of primary SS will improve the application of systematic therapies and the setting of criteria with which to monitor therapies and assess prognosis (e.g., lymphoma development).

Saliva is the product of 3 pairs of major salivary glands (the parotid, submandibular, and sublingual glands) and multiple minor salivary glands that lie beneath the oral mucosa. Human saliva contains many informative proteins that can be used for the detection of diseases. Saliva is an attractive diagnostic fluid because testing of saliva provides several key advantages, including low cost, noninvasiveness, and easy sample collection and processing. This biologic fluid has been used for the survey of general health and for the diagnosis of diseases in humans, such as human immunodeficiency virus, periodontal diseases, and autoimmune diseases (5–8). Our laboratory is active in the comprehensive analysis of the saliva proteome (for more information, see www.hspp.ucla.edu), thus providing the technologies and expertise to contrast proteomic constituents in primary SS with those in control saliva (9–11). Thus far, we have identified over 1,000 proteins in whole saliva (WS). In addition, we have recently identified and cataloged ~3,000 messenger RNAs (mRNA) in human WS (12). These studies have provided a solid foundation for the discovery of biomarkers in the saliva of patients with primary SS. We have previously demonstrated proteome- and genome-wide approaches to harnessing saliva protein and mRNA signatures for the detection of oral cancer in humans (13,14).

There have been continuous efforts in the search for biomarkers in human serum or saliva for the diagnosis of primary SS. Some gene products were found at elevated levels in SS patient sera or saliva, including β2-microglobulin (β2m), soluble IL-2 receptor, IL-6, anti-Ro/SSA, anti-La/SSB, and anti–α-fodrin autoantibodies (15–20). However, none of them individually is sensitive or specific enough to use for the confirmative diagnosis of SS (15). Therefore, it is crucial to use emerging proteome- and genome-wide approaches to discover a wide spectrum of informative and discriminatory biomarkers that can be combined to improve the sensitivity and specificity for the detection of primary SS.

PATIENTS AND METHODS

Patient cohort

Because sample quality is critical for clinical proteomics studies, a standardized procedure, in strict accordance with the American–European Consensus Group Criteria for SS (2), was used for the identification and recruitment of primary SS patients for this study. A diagnostic evaluation of SS was performed in all patients and included assessments of subjective complaints of oral and ocular dryness, sialometry (unstimulated WS), sialography, histopathology of salivary gland tissue, serology (SSA and SSB antibodies), eye tests (rose bengal staining and Schirmer’s test) according to the American–European classification criteria for SS (2), and screening for extraglandular manifestations. Three of the primary SS patients were being treated with hydroxychloroquine, and 1 patient was being treated with prednisolone. Eight patients had a focus score of >1 on examination of parotid gland biopsy tissue.

The enrolled primary SS patients and healthy control subjects were well matched for age, sex, and ethnicity. The mean ±SD age was 37.2 ± 9.8 years in the primary SS patients (n = 10) and 37.0 ± 10.6 years in the healthy control subjects (n = 10). All subjects enrolled in this study were Caucasian women, since primary SS mainly affects women. All of the enrolled control subjects were negative for serum anti-SSA/ SSB antibodies, and none of them reported any sicca symptoms, including oral and ocular dryness.

Saliva sample collection

Samples of WS and saliva from the parotid and submandibular/sublingual glands were collected from each primary SS patient and control subject for comparative analysis. Saliva sample collection was performed at the University Medical Center Groningen, using our standardized saliva collection protocols. Subjects were asked to refrain from eating, drinking, smoking, or performing oral hygiene procedures for at least 1 hour prior to the collection. Samples were collected in the morning, at least 2 hours after eating and rinsing the mouth with water, according to established protocols (21,22). WS was stimulated by chewing paraffin and was collected over a period of 15 minutes. Glandular saliva specimens from individual parotid glands and, simultaneously, from the submandibular/sublingual glands were collected into Lashley cups (placed over the orifices of the Stenson’s duct) and by syringe aspiration (from the orifices of the Warton’s duct, located anteriorly in the floor of the mouth), respectively.

After collection, the saliva samples were immediately mixed with protease inhibitors (Sigma, St. Louis, MO) to ensure preservation of the integrity of the proteins and then centrifuged at 2,600g for 15 minutes at 4°C. The supernatant was removed from the pellet, immediately aliquoted, and stored at −80°C. All samples were kept on ice during the process. Two patients who had very low submandibular/ sublingual gland salivary flow rates (0.03 ml/minute) did not produce enough submandibular/sublingual gland saliva for this study.

Sample preparation for proteomic analysis

The saliva samples were precipitated overnight at −20°C with cold ethanol. Following centrifugation at 14,000g for 20 minutes, the supernatants were collected and dried with a speed vacuum for use in the peptide biomarker study. The pellet was then washed once with cold ethanol and collected for assay of total protein using a 2-D Quant kit (Amersham, Piscataway, NJ). We pooled saliva samples according to the total protein content from all patients with primary SS and those from all control subjects. However, both the patients and controls were analyzed individually for the peptide profiling experiment.

Matrix-assisted laser desorption ionization–time-of-flight mass spectrometry (MALDI-TOF-MS)

Profiling of saliva peptides in 10 primary SS patients and 10 matched control subjects was performed using a MALDI-TOF-MS system (Applied Biosystems, Foster City, CA). The peptide fraction from each patient (n = 10) and control (n = 10) sample was dissolved in 10 µl of 50% acetonitrile (ACN)/0.1% trifluoroacetic acid (TFA). The sample was mixed with α-cyano-4-hydroxycinnamic acid (10 mg/ml in 50% ACN/0.1% TFA) at a ratio of 1:2, and 1 µl of the mixture was spotted on the MALDI plate for measurement. Three measurements were performed for each sample, and the signals were averaged for subsequent data analysis.

In order to achieve an accurate comparison of the MALDI-TOF-MS data between the patient and control groups, baseline correction and Gaussian smoothing were initially performed to eliminate broad artifacts and noise spikes. Afterward, peak alignment was undertaken to ensure accurate alignment of the mass/charge (m/z) values across the set of spectra, and peak normalization was performed against the total peak intensity. These steps ensured comparability of the MALDI-TOF-MS spectra among all subjects. Subsequent statistical analysis (t-test) was used to reveal peptides that were present at significantly different levels in the primary SS patients as compared with the control subjects.

Two-dimensional gel electrophoresis

Saliva samples from the 10 primary SS patients and from the 10 control subjects were equally pooled according to the total protein content and then precipitated using the same procedures described above. The pellet was washed once with cold ethanol and then resuspended in rehydration buffer. A total of 100 µg of proteins was loaded onto each gel for the 2-D gel separation procedure. Isoelectric focusing was performed using immobilized pH gradient strips (11 cm long, with an isoelectric point [pI] of 3–10 nonlinear) on a Protean isoelectric focusing cell (Bio-Rad, Hercules, CA), and sodium dodecyl sulfate– polyacrylamide gel electrophoresis was performed in 8–16% precast Criterion gels on a Criterion Dodeca Cell (Bio-Rad). Fluorescent SYPRO Ruby stain (Invitrogen, Carlsbad, CA) was used to visualize the protein spots.

The gel images were acquired and analyzed using PDQuest software (Bio-Rad). The images were initially processed through transformation, filtering, automated spot detection, normalization, and matching. The 2-D gel image was transformed to adjust the intensity of the protein spot and filtered to remove small noise features without affecting the protein spot. The images were then normalized based on the total density of the gel image. The 2-D gel images of the primary SS patients (master gel) and the control subjects were used as a “match set” for automated detection of the protein spots on the gel. Within the match set, the detected spots were reviewed manually, and the relative protein levels in the patient sample compared with the control sample were summarized.

Liquid chromatography tandem mass spectrometry (LC-MS/MS) and database searching

Protein spots showing differential protein levels were excised by a spot-excision robot (Proteome Works; Bio-Rad) and deposited into 96-well plates. Proteins in each gel spot were reduced with dithiothreitol, alkylated with iodoacetamide, and then digested overnight at 37°C with 10 ng of trypsin. After digestion, the peptides were extracted and stored at −80°C prior to LC-MS/MS analysis.

LC-MS/MS analysis of peptides was performed using an LC Packings Nano-LC system (Dionex, Sunnyvale, CA) with a nanoelectrospray interface (Protana, Odense, Denmark) and a quadrupole time-of-flight (Q-TOF) mass spectrometer (QSTAR XL; Applied Biosystems). A New Objective PicoTip tip (internal diameter 8 mm; New Objective, Woburn, MA) was used for spraying, with the voltage set at 1,850V for online MS and MS/MS analyses. The samples were first loaded onto an LC Packings PepMap C18 precolumn (300 µm × 1 mm; particle size 5 µm) and then injected onto an LC Packings PepMap C18 column (75 µm × 150 mm; particle size 5 µm) (both from Dionex) for nano-LC separation at a flow rate of 250 nl/minute. The eluents used for LC were 1) 0.1% formic acid and 2) 95% ACN/0.1% formic acid, and a 1%/minute gradient was used for the separation.

The acquired MS/MS data were searched against the International Protein Index (IPI) human protein database (available at http://www.ebi.ac.uk/IPI/IPIhelp.html) using the Mascot (Matrix Science, Boston, MA) database search engine. Positive protein identification was based on standard Mascot criteria for statistical analysis of LC-MS/MS data.

Immunoblotting

Western blot analysis of α-enolase was performed on the same set of saliva samples (10 primary SS and 10 control samples). Proteins were separated on 12% NuPAGE gels (Invitrogen) at 150V and then transferred to a polyvinylidene difluoride membrane (Bio-Rad) using an Invitrogen blot transfer cell. After saturating with 5% milk in Tris buffered saline–Tween buffer (overnight at 4°C), the blots were sequentially incubated for 2 hours at room temperature with polyclonal goat α-enolase primary antibody and horseradish peroxidase–conjugated anti-goat IgG secondary antibody (Santa Cruz Biotechnology, Santa Cruz, CA). The bands were detected by enhanced chemiluminescence (Amersham) and quantified using Quantity One software (Bio-Rad).

Profiling of salivary mRNA by high-density oligonucleotide microarray analysis

Samples of stimulated parotid gland saliva or WS from 10 primary SS patients and 8 matched controls were preserved in RNAlater reagent (Qiagen, Valencia, CA) at a 1:1 ratio and then frozen at −80°C. Total salivary RNA was isolated from 560 µl of RNAlater-preserved saliva (280 µl of parotid gland saliva/WS and 280 µl of RNAlater) using a viral RNA mini kit (Qiagen) as described previously (12). Isolated total RNA was treated with 2 rounds of recombinant DNase I (Ambion, Austin, TX) digestion, and the RNA concentration was measured with a NanoDropND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE). The salivary RNA quality was examined by real-time reverse transcription–polymerase chain reaction (RT-PCR) analysis for expression of the salivary internal reference gene transcripts S100 calcium-binding protein A8 and annexin A2 (data not shown).

For microarray study, total salivary RNA was subjected to 2 rounds of T7-based RNA linear amplification (10). One microliter (200 ng/µl) of poly(dI-dC) (Amersham) was added to 11 µl of the salivary RNA sample, and 2 rounds of first-strand and second-strand complementary DNA (cDNA) synthesis were performed with a RiboAmp HS RNA amplification kit (Arcturus, Mountain View, CA) according to the manufacturer’s instructions. After purification, the cDNA were in vitro transcribed to RNA and then biotinylated with GeneChip Expression 3′-Amplification Reagents for in vitro transcription labeling (Affymetrix, Santa Clara, CA). The labeled RNA was purified with the reagents provided with the RiboAmp HS RNA Amplification kit. The quality and quantity of amplified RNA were determined by spectrophotometry, with optical densities at 260/280 nm >1.9 for all samples.

Biotinylated RNA samples (15 µg each) were subsequently fragmented, and the quality of the fragmented RNA was assessed using an Agilent 2100 Bioanalyzer (Agilent, Palo Alto, CA). The Affymetrix human genome U133 Plus 2.0 array, which contains > 54,000 probe sets representing >47,000 transcripts and variants, including ~38,500 well-characterized human genes, was applied to salivary mRNA profiling. Fragmented RNA were hybridized overnight to the microarrays. After a high-stringency wash to remove the unbound probes, the hybridized chips were stained and scanned according to the manufacturer’s standard expression protocol. The scanned images were read with the Affymetrix microarray Robust Multiarray Average (RMA) software (23). We deposited the microarray data we obtained into a Minimum Information About a Microarray Experiment (MIAME)−compliant database (available at http://www.mged.org/workgroups/MIAME/miame.html); the accession number is GSE7451.

Statistical analysis for the mRNA study

The expression microarrays were scanned, and the fluorescence intensity was measured using Microarray Suite 5.0 software (Affymetrix). The arrays were then imported into the statistical software R (24). After quantile normalization and RMA background correction, the RMA expression index was computed in R using the Bioconductor routine (25). Since most human RNAs are not present in saliva (12), we used the present/absent call generated by the Affymetrix Microarray Suite 5.0 software to exclude probe sets that were assigned an “absent” call in most (>75%) of the samples. Principal components analysis was performed to assess the information contained in the data to separate primary SS and control cases. Student’s 2-tailed t-test was used for comparison of the average gene expression signal intensity between samples from the SS patients (n = 10) and controls (n = 8). P values were adjusted with the Benjamini and Hochberg false discovery rate (FDR) criterion (26). Fold ratios between SS and control samples were calculated for the transcripts. For the further validation study using real-time quantitative PCR, we applied stringent criteria: an alpha level of 0.001 for the t-test, which corresponded to a 5% FDR based on the data, and a fold ratio of 3. For functional analysis using MAPPFinder (27), we applied an alpha level of 0.01, which corresponded to an 8% FDR, and a fold ratio of 2, to obtain a larger list of genes.

Real-time quantitative RT-PCR

The biomarker candidates generated by microarray profiling were validated by real-time quantitative RT-PCR on the same set of samples used for the microarray analysis. All primers used for quantitative PCR were designed with the Primer3 program and synthesized by Sigma. Total RNA was reverse-transcribed using reverse transcriptase and gene-specific primers. One microliter of total RNA was used in a 20-µl volume of cDNA synthesis reaction and then subjected to the following thermal cycling conditions: 25°C for 10 minutes, 42°C for 45 minutes, and 95°C for 5 minutes. Three microliters of cDNA was used as template for each 20-µl PCR, which contained forward primer (200 nM), reverse primer (200 nM), and 10 µl of 2× SYBR Green PCR Master Mix (Applied Biosystems). PCRs were performed in a 96-well plate on the Bio-Rad iCycler or IQ5 instrument (95°C for 3 minutes followed by 50 cycles of 95°C for 30 seconds, 62°C for 30 seconds, and 72°C for 30 seconds). All PCRs were performed in duplicate for all candidate mRNA.

The specificity of the PCR was confirmed according to the melting curve of each gene, and the average threshold cycle (Ct) was examined. The relative expression of the candidate genes was calculated according to the 2(− ΔCt) method, where ΔCt = Ct in primary SS patients − Ct in controls. The expression ratio ([primary SS patients/controls] = 2[− ΔCt]) is shown as the fold change (28).

Pathway analysis

PathwayArchitect software, version 1.1.0 (Stratagene, La Jolla, CA) was used to investigate the functional pathways presented by the differentially expressed genes.

RESULTS

Salivary flow rate and total salivary protein and mRNA contents in primary SS patients

Patients with primary SS who had been carefully diagnosed and monitored were enrolled in this study. All 10 patients were positive for anti-SSA/Ro antibodies, and 9 of them were also positive for anti-SSB/La antibodies. Their mean ± SD IgG level was 23.4 ± 7.4 gm/liter, and their mean ± SD IgM rheumatoid factor level was 136.3 ± 99.6 kIU/liter. These patients exhibited significantly lower (~50%) salivary flow rates than did the age-, sex-, and ethnicity-matched healthy control subjects. The mean ± SD stimulated salivary flow rates in the 10 primary SS patients were 0.13 ± 0.12 ml/minute for the parotid glands (per gland), 0.32 ± 0.38 ml/minute for the submandibular/sublingual glands, and 0.61 ± 0.23 ml/ minute for WS. These rates in the 10 control subjects were 0.21 ± 0.07 ml/minute for the parotid glands (per gland), 0.78 ± 0.36 ml/minute for the submandibular/ sublingual glands, and 1.03 ± 0.31 ml/minute for WS. Due to the low volume of saliva obtained from the primary SS patients, the salivary proteins were equally pooled for the 10 primary SS patients and separately for the 10 control subjects for the 2-DE analyses.

On average, the mean ± SD total protein concentrations in the controls were determined to be 1.26 ± 0.40 mg/ml in submandibular/sublingual gland saliva (n = 8 subjects), 0.93 ± 0.38 mg/ml in parotid gland saliva (n = 10 subjects), and 0.95 ± 0.52 mg/ml in WS (n=10 subjects). The total protein concentrations in the primary SS patients were 1.45 ± 0.49 mg/ml in submandibular/sublingual gland saliva (n=8 patients), 1.40 ± 0.56 mg/ml in parotid gland saliva (n = 10 patients), and 1.38 ± 0.37 mg/ml in WS (n=10 patients). There were consistently higher concentrations of proteins in the SS patients (WS, submandibular/sublingual gland saliva, and parotid gland saliva) than in the matched healthy control subjects. In addition, saliva from the primary SS patients appeared to contain a higher concentration of total RNA than did that from the matched controls. In parotid gland saliva, the mean ± SD RNA concentration was determined to be 5.8 ± 3.1 µg/ml in the primary SS patients and 3.6 ± 1.5 µg/ml in the controls (P = 0.05). In WS, the average RNA concentration was 10.9 ± 5.4 µg/ml for primary SS patients and 6.6 ± 3.6 µg/ml for matched controls (P = 0.057).

Discovery of candidate peptide markers for primary SS

The expression of 16 WS peptides was found to be significantly different (P = 0.0046–0.0441) in primary SS patients (n = 10) and controls (n = 10). Ten of the 16 peptides were overexpressed (m/z 1,107, 1,224, 1,333, 1,380, 1,451, 1,471, 1,680, 1,767, 1,818, and 2,039) and 6 were underexpressed (m/z 2,534, 2,915, 2,953, 3,311, 3,930, and 4,187) in the primary SS patients. The peptide with an m/z of 1,451 exhibited the highest up-regulation (25.9-fold) in primary SS patients (results not shown). We also compared the native peptide patterns in saliva from the parotid and submandibular/ sublingual glands between primary SS patients and control subjects (results not shown). WS was found to contain more informative peptides than did gland-specific (parotid or submandibular/sublingual) saliva. On average, 53 MALDI peaks were observed in WS from the 10 primary SS patients, with only 24 peaks and 26 peaks detectable in saliva from their parotid and submandibular/sublingual glands, respectively.

Findings of 2-DE of WS proteins from primary SS patients and matched control subjects

Figure 1 presents the 2-DE patterns of the proteins in pooled WS samples from 10 primary SS patients and 10 control subjects. A number of proteins were found to be differentially expressed between the patient and control groups. By performing the PDQuest analysis and normalizing the protein spot signals, the relative levels of these proteins were quantified. The differentially expressed proteins (Figure 1, spots 1–42) were excised and subsequently identified using in-gel tryptic digestion and LC-Q-TOF-MS. Pooled parotid and submandibular/ sublingual gland saliva from primary SS patients and control subjects was also analyzed by 2-DE (results not shown). WS was again found to be more informative than parotid or submandibular/sublingual gland saliva for generating candidate protein biomarkers for the detection of primary SS. A total of 325 protein spots were detected by 2-DE analysis of WS, whereas 232 and 267 spots were detected by 2-DE analysis of parotid and submandibular/sublingual gland saliva, respectively.

Figure 1.

Figure 1

Open in a new tab

Comparative analysis of proteins in whole saliva (WS) samples from patients with primary Sjögren’s syndrome (pSS) and age-, sex-, and ethnicity-matched control subjects, as determined by 2-dimensional gel electrophoresis (2-DE) and liquid chromatography–quadrupole time-of-flight mass spectrometry (LC-Q-TOF-MS). Shown are the 2-DE patterns of proteins in pooled WS from 10 control subjects and 10 primary SS patients. A total of 100 µg of total proteins from each pooled sample was used for the 2-D gel separation. The differentially expressed proteins (spots 1–42; see Table 1 for the complete list) were identified using in-gel tryptic digestion and LC-Q-TOF-MS.

LC-Q-TOF-MS identification of proteins at altered expression levels

The differentially expressed WS proteins identified by LC-Q-TOF-MS and Mascot database searching, as well as their theoretical isoelectric point (pI), relative molecular mass (Mr), IPI accession number, the number of peptides matched, and ratios of expression levels between the primary SS patient and matched control groups are shown in Table 1.

Table 1.

Salivary proteins differentially expressed between primary SS patients and healthy control subjects, as identified by LC-MS/MS and Mascot database searching*

Spot
no.		 Accession
no.		 Protein name Mascot
score		 No. of peptides
matched		 pI Mr Expression ratio
(patients:controls)
1 IPI00295105 Carbonic anhydrase VI 163 4 6.65 35,343 0.22
2 IPI00295105 Carbonic anhydrase VI 114 5 6.65 35,343 0.35
3 IPI00295105 Carbonic anhydrase VI 78 2 6.65 35,343 0.29
4 IPI00004573 Polymeric immunoglobulin receptor 235 5 5.58 83,262 0.48
5 IPI00004573 Polymeric immunoglobulin receptor 293 7 5.58 83,262 0.39
6 IPI00004573 Polymeric immunoglobulin receptor 182 4 5.58 83,262 0.56
7 IPI00019038 Lysozyme C 103 2 9.38 16,526 0.21
8 IPI00022974 Prolactin-inducible protein 147 3 8.26 16,562 0.52
9 IPI00009650 Von Ebner’s gland protein 239 4 5.39 19,238 0.32
10 IPI00032293 Cystatin C 153 3 9.0 15,789 0.43
11 IPI00013382 Cystatin SN 152 3 6.82 16,361 0.46
12 IPI00013382 Cystatin SN 130 3 6.82 16,361 0.61
13 IPI00002851 Cystatin D 50 1 6.70 16,070 0.56
14 IPI00032294 Cystatin S 166 3 4.95 16,214 0.65
IPI00013382 Cystatin SA 208 4 4.85 16,445
15 IPI00007047 Calgranulin A 104 2 6.51 10,828 0.53
16 IPI00007047 Calgranulin A 79 2 6.51 10,828 Absent in primary SS
17 IPI00027462 Calgranulin B 126 4 5.71 13,234 1.05
18 IPI00219806 Psoriasin 133 4 6.28 11,464 1.44
19 IPI00410714 Hemoglobin α1-globin chain 157 5 7.96 15,292 Absent in control
20 IPI00218816 Hemoglobin β-chain 48 1 6.75 15,988 2.73
21 IPI00218816 Hemoglobin β-chain 51 1 6.75 15,988 7.58
22 IPI00007797 Fatty acid binding protein, epidermal 211 6 6.60 15,155 3.21
23 IPI00472762 IGHG1 protein 333 14 8.33 50,822
IPI00472610 Hypothetical protein 363 14 7.50 52,633 22.64
IPI00430840 Ig γ1-chain C-region 333 14 7.48 54,866
24 IPI00472610 IGHM protein 260 11 7.50 53,270 Absent in control
IPI00550718 Ig γ1-chain C-region 257 11 8.46 53,331
25 IPI00465248 α-enolase 409 12 6.99 47,139 4.37
26 IPI00300786 Salivary α-amylase, fragment 241 5 5.73 57,731 3.41
27 IPI00300786 Salivary α-amylase, fragment 230 4 5.73 57,731 2.19
28 IPI00300786 Salivary α-amylase, fragment 375 7 5.73 57,731 31.53
29 IPI00300786 Salivary α-amylase, fragment 260 5 5.73 57,731 2.57
30 IPI00300786 Salivary α-amylase, fragment 171 5 5.73 57,731 2.50
31 IPI00300786 Salivary α-amylase, fragment 194 4 5.73 57,731 11.92
32 IPI00300786 Salivary α-amylase, fragment 149 4 5.73 57,731 1.57
33 IPI00300786 Salivary α-amylase, fragment 148 4 5.73 57,731 4.03
34 IPI00549682 Fructose-bisphosphate aldolase A 218 4 8.75 52,306 2.59
35 IPI00332161 Ig γ1-chain C-region 138 5 8.46 36,083 2.54
36 IPI00215983 Carbonic anhydrase I 119 4 6.59 28,852 7.4
37 IPI00218414 Carbonic anhydrase II 98 2 8.67 31,337 2.11
38 IPI00013885 Caspase 14 172 5 5.44 27,662 3.32
39 IPI00419424 Ig κ-chain C-region 263 7 5.82 27,313 1.79
40 IPI00004656 β2-microglobulin 62 2 6.06 13,706 2.21
41 IPI00021439 Actin 461 11 5.29 41,710 3.18
42 IPI00022434 Serum albumin, fragment 492 10 5.41 69,321 Absent in control
Open in a new tab
*

Liquid chromatography mass spectrometry/mass spectrometry (LC-MS/MS) analysis and Mascot database searching were performed to identify the proteins. Shown are the theoretical isoelectric point and molecular mass of the protein precursors, as well as the ratio of protein levels in patients with primary Sjögren’s syndrome (SS) and matched control subjects, as detected by 2-dimensional gel electrophoresis.

Figure 2A depicts the tandem MS spectrum of a double-charged tryptic peptide (m/z 450.3). The precursor ion was well fragmented to yield sufficient structural information for confident identification of the peptide sequence TIAPALVSK, which originated from α-enolase. Mascot database searching indicated that 12 peptides were matched to this protein, resulting in a sequence coverage of 31%. Validation of α-enolase was also performed by Western blotting of the same set of samples used for the 2-DE study (Figure 2B). An equal amount of total proteins from each sample was used for immunoblotting of α-enolase and actin. Both α-enolase and actin were found to be up-regulated in WS from primary SS patients, which is consistent with the 2-DE results (Table 1). P values were calculated to be 0.006 for α-enolase without actin normalization and 0.037 with actin normalization for comparisons between the primary SS patient and healthy control groups.

Figure 2.

Figure 2

Open in a new tab

Analysis of α-enolase by electrospray ionization tandem mass spectrometry (ESI-MS/MS) and immunoblotting. A, ESI-MS/MS spectrum of the tryptic peptide TIAPALVSK (mass/charge [m/z] 450.3 atomic mass units [amu]) from α-enolase. This protein was found to be overexpressed in whole saliva from patients with primary Sjögren’s syndrome (pSS), as determined by 2-dimensional gel electrophoresis. B, Immunoblotting of whole saliva from 10 patients with primary SS and 10 age-, sex-, and ethnicity-matched control subjects for α-enolase and actin. An equal amount of proteins from each sample was used for the immunoblots.

Identification of candidate genomic markers of primary SS in saliva samples

For all the arrays, the mean ± SD percentage of genes present was 13.2 ± 2.9%. This is similar to the finding in our previous study (12) and indicates consistency of the techniques used for sample preparation. Microarray profiling indicated that WS contains >10 times more informative mRNA than does parotid gland saliva. A total of 328 mRNA had a >2-fold change in WS from primary SS patients, while only 21 mRNA had a >2-fold change in parotid gland saliva from these patients. Therefore, we focused on the discovery and validation of WS candidate mRNA biomarkers using microarray and real-time quantitative RT-PCR strategies.

Gene expression profiles of individual WS samples from 10 primary SS patients and 8 controls were compared. After filtering the transcripts by the criteria of being “present” in >25% of the samples, a total of 6,413 transcripts were retained for further analysis. This number is consistent with our previous results, showing that only a small number of RNAs are present in saliva (12). Principal components analysis indicated that the information contained in the data could well segregate control subjects and primary SS patients (Figure 3). We then performed statistical testing and fold change analysis to identify differentially expressed genes. Only a few mRNA were found at significantly lower levels in primary SS patients as compared with the controls when using a threshold of >2-fold change and a significance level of P < 0.01 (FDR 0.08). Yet, by the same criteria, 162 genes showed significant up-regulation in samples from patients with primary SS.

Figure 3.

Figure 3

Open in a new tab

Principal components analysis of the gene expression data in patients with primary Sjögren’s syndrome (SS) and in age-, sex-, and ethnicity-matched control subjects. Results of the principal components (PC1 and PC2) analysis suggest that the gene expression data we obtained segregated the 8 control subjects (blue symbols) from the 10 primary SS patients (red symbols).

Pathway analysis indicated that 37 genes were involved in the IFNα pathway, and most of them have been reported to be IFNα or IFNβ inducible (29,30). These results suggest that activation of IFN pathways is involved in the pathogenesis of primary SS and that the related information is reflected in the saliva. To facilitate biomarker discovery, we narrowed the number of candidate biomarkers by using more stringent threshold criteria of P < 0.001 (FDR 0.05) and 3-fold change. Based on these criteria, we found 27 genes that were highly overexpressed in samples from primary SS patients. These genes are sufficiently informative for segregating the primary SS patients from the control subjects (Figure 4).

Figure 4.

Figure 4

Open in a new tab

Heat map of 27 mRNA that were significantly up-regulated in patients with primary Sjögren’s syndrome (SS) as compared with the age-, sex-, and ethnicity-matched control subjects, as determined by microarray profiling analysis. Control and SS patient numbers are shown at the bottom.

Among the top 27 genes, 13 were validated by real-time quantitative RT-PCR. Eleven of the 13 genes were found to be significantly up-regulated in primary SS patients (>10-fold change), including the IFNα-inducible protein G1P2, which showed an ~500-fold change in primary SS patients. Table 2 shows the average Ct values of these genes in primary SS patients and control subjects, as well as the quantitative PCR fold change in comparison with that of microarray profiling.

Table 2.

Real-time quantitative RT-PCR validation of 13 genes selected from the top 27 genes found to be differentially expressed in primary SS patients and healthy control subjects*

Average Ct
Gene Healthy
controls		 Primary SS
patients		 ΔCt
(controls/patients)		 Quantitative RT-PCR,
fold change (2[−ΔCt])		 P, by
t-test		 Microarray,
fold change
GIP2 44.5 ±1.9 35.5 ± 2.1 9.0 495.5 <0.001 15.76
B2M 45.0 ± 2.1 38.8 ± 3.4 6.2 72.1 <0.001 8.67
IFIT2 41.1 ± 2.0 35.9 ± 2.6 5.1 35.5 <0.001 12.19
BTG2 38.5 ± 5.3 33.5 ± 2.0 5.0 32.4   0.01 3.22
IFIT3 43.8 ± 0.5 39.1 ± 2.4 4.7 25.3 <0.001 122.82
MNDA 37.3 ± 1.2 33.7 ± 2.1 3.7 12.7 <0.001 8.67
FCGR3B 40.6 ±1.5 36.9 ± 2.2 3.6 12.5 <0.001 25.32
TXNIP 39.2 ± 2.1 35.6 ± 3.2 3.6 11.7   0.01 3.42
IL18 45.3 ± 2.1 41.8 ± 2.5 3.5 11.5   0.01 6.12
HLAB 36.4 ± 2.7 32.9 ± 2.0 3.5 11.2   0.01 4.34
EGR1 37.4 ± 2.4 33.9 ± 2.0 3.4 10.3   0.01 7.20
COP1 40.5 ±1.5 38.7 ± 3.3 1.8 3.4   0.18 7.62
TNSF 39.6 ± 0.4 38.9 ± 2.9 0.7 1.6   0.95 8.03
Open in a new tab
*

All real-time quantitative reverse transcription–polymerase chain reaction (RT-PCR) analyses were performed in duplicate. See Patients and Methods for calculations of the fold change (primary Sjögren’s syndrome [SS] patients/healthy controls) and threshold cycle (Ct) data.

DISCUSSION

Although saliva has been extensively explored as a source of information that can be used in the diagnosis of primary SS, most of the previously published studies mainly examined individual components of the saliva. High-throughput profiling techniques, such as proteomics and expression microarray analysis, enable us to explore salivary proteins and mRNA in a global manner and may therefore provide new and deeper insights that may lead to the discovery of salivary biomarkers for primary SS. Recently, surface-enhanced laser desorption ionization time-of-flight mass spectrometry and differential gel electrophoresis have been used to identify very promising candidate biomarkers of SS in tears and in parotid gland saliva (31,32). It was found that the proteomic profile of parotid gland saliva from SS patients is a mixture of increased inflammatory proteins and decreased acinar proteins as compared with the profile in non-SS controls (32).

In order to determine which oral fluid compartment is more informative for the discovery of biomarkers that can be used to detect primary SS, we used both proteomic and microarray approaches to profile peptides, proteins, and mRNA in WS, parotid gland saliva, and submandibular/sublingual gland saliva from each study subject. WS as a fluid includes secretions from 3 major salivary glands, numerous minor salivary glands, and gingival fluid, as well as cell debris. There has therefore been concern about the complex background in WS for discovery of disease biomarkers, whereas parotid gland saliva, if collected carefully, may contain more specific biomarkers for primary SS. Yet, there are no published reports of any advantage of using gland-specific saliva versus WS in terms of the diagnostic potential for primary SS. The findings of our study allow us to conclude that WS is more informative than glandular saliva for generating biomarkers to be used for the detection of primary SS.

Microarray profiling indicated that WS from primary SS patients contained 328 mRNA with 2-fold change in expression, whereas the parotid gland saliva from primary SS patients contained only 21 mRNA with a >2-fold change in expression. Similarly, findings of the MALDI-TOF-MS and 2-DE analyses suggested that WS from primary SS patients has more informative proteomic components than does parotid or submandibular/ sublingual gland saliva. Since the salivary flow rate varies from person to person, the peptide or protein composition among different individuals could be affected by the very low salivary flow rate of the parotid and submandibular/sublingual glands. With regard to the low flow rate of glandular saliva, as well as the additional skill set and clinical time necessary to collect gland-specific saliva, WS may be a more appropriate clinical diagnostic fluid for the discovery and detection of biomarkers of primary SS.

The panel of candidate peptide/protein markers for primary SS is completely distinct from the panel we obtained for oral cancer (13). This suggests that the panels of discriminatory salivary proteomic components are likely to be different for different diseases. The majority of underexpressed proteins found in WS from primary SS patients are secretory proteins, including 3 glycoforms of carbonic anhydrase VI (Figure 1B, spots 1–3), cystatins, lysozyme C, polymeric immunoglobulin receptor (pIgR), calgranulin A, prolactin-inducible protein, and von Ebner gland protein. This suggests that the level of secretory proteins in WS from primary SS patients may be directly affected by injury to salivary glandular cells. Several of these down-regulated proteins in the WS of primary SS patients, including pIgR, lysozyme C, and cystatin C, were found up-regulated in the parotid gland saliva of primary SS patients in a previously published study (32). This may be factual, as evidenced by our comparative analysis of parotid gland salivary proteins in primary SS patients and control subjects (results not shown). For example, in our 2-DE study, pIgR was also found to be up-regulated in the pooled parotid gland saliva of primary SS patients as compared with the matched control subjects (results not shown). A future study of salivary proteins from the parotid gland versus WS in the same primary SS patients would be of interest to the primary SS research community.

Two glycolysis enzymes, fructose-bisphosphate aldolase A and α-enolase, were found at elevated levels in the WS of primary SS patients. Fructose-bisphosphate aldolase A plays a central role in glucose metabolism, catalyzing either net cleavage or synthesis during glycolysis or gluconeogenesis. Alpha-enolase is a multifunctional glycosis enzyme involved in various processes, such as growth control, hypoxia tolerance, and allergic responses. Previously, α-enolase was identified as an autoantigen in Hashimoto encephalopathy, which is an autoimmune disease associated with Hashimoto thyroiditis (33). Alpha-enolase was also found as an autoantigen in lymphocytic hypophysitis, and serum autoantibodies directed against α-enolase were detected in patients with lymphocytic hypophysitis as well as in patients with other autoimmune diseases. Excessive production of autoantibodies, which are generated as a consequence of uptake of enolase by antigen-presenting cells and subsequent B cell activation, can potentially initiate tissue injury as a result of immune complex deposition (34,35). Overexpressed proteins in WS from patients with primary SS also included psoriasin, fatty acid binding protein, carbonic anhydrases I and II, salivary amylase fragments, caspase 14, β2m, hemoglobin (β and α1 global chains), and immunoglobulins. The elevated level of caspase 14 protein and caspases 1 and 4 RNA in primary SS patients also suggested an interesting role of apoptosis in the pathogenesis of primary SS.

Our study clearly demonstrates that primary SS–related gene expression signatures are present in saliva and they are able to differentiate primary SS patients from control subjects. To the best of our knowledge, this is the first study on the discovery of candidate salivary mRNA markers for the detection of primary SS. We identified 162 differentially expressed genes in the saliva of primary SS patients, as compared with a reported 35 and 424, respectively, identified in 2 studies of microarray profiling of minor salivary gland biopsy tissues (36,37). One of the important findings of this study is that the 37 up-regulated genes in the saliva of primary SS patients were involved in the IFN pathway. This further confirmed the findings from previous tissue-based studies and demonstrated that the IFN-inducible gene signature associated with primary SS is reflected in patients’ saliva (36–39). Beyond the IFN-inducible genes, the class I major histocompatibility complex is another major group of up-regulated genes found to be common to salivary gland and WS from patients with primary SS (36,37). Other genes reported to be of particular interest in the pathogenesis of primary SS (37) that were found to be overexpressed in saliva are proteasome subunit β type 9, guanylate binding protein 2, IFN-induced protein 44, and IFNα-inducible protein G1P2, and β2m. These common genes found in saliva and minor salivary gland tissue from patients with primary SS support our hypothesis that saliva harbors the biomarkers for primary SS.

The mechanism of IFN pathway activation in the pathogenesis of primary SS may be more complicated. Activation of IFN pathways (both type I and type II) in primary SS suggests the involvement of viral infection in its pathogenesis. Immune complexes consisting of autoantibodies and DNA- or RNA-containing autoantigens derived from apoptotic or necrotic cells are also able to induce the production of type I IFN. However, IFN itself is not among the genes we found to be overexpressed in the saliva of the primary SS patients. On the other hand, low-dose IFNα has been reported to be effective in the treatment of some patients with primary SS. A single-blind controlled trial showed that IFNα therapy significantly improved salivary gland dysfunction in SS patients (40). Serial labial salivary gland biopsy in 9 patients responding to IFNα therapy showed a significant decrease (P < 0.02) in lymphocytic infiltration and a significant increase (P = 0.004) in the proportion of intact salivary gland tissue after IFNα treatment (41).

Type I IFN pathway dysregulation, however, has been reported in such distinct diseases as SLE, dermato-myositis, psoriasis, and SS (36), indicating that the consequences of activation of this pathway are likely to be tissue type–dependent and, from a therapeutic point of view, that local immune modulation (e.g., direct infusion into salivary glands) may be more efficient than systemic interference. An initial viral infection–induced type I IFN production in salivary glands, with prolonged activation triggered by autoantibodies from nucleic acid–containing immune complexes, has been proposed as the mechanism of primary SS (42). More importantly, activation of this IFN pathway may provide potential therapeutic targets for primary SS, and saliva may be used to monitor the response to the IFN-related target modulation.

One of the up-regulated genes seen in the saliva of patients with primary SS is β2m, which is also regulated by IFN. Significantly elevated levels of β2m have previously been detected in saliva from patients with primary SS (43). The concentration of salivary (but not serum) β2m was highly related to the salivary gland biopsy focus score (43). The value of salivary β2m protein as a biomarker for primary SS has been evaluated, and it has been suggested that determination of β2m levels in the saliva could be used as a noninvasive measurement for confirmation of the diagnosis of SS (44). Interestingly, but not surprisingly, we found that both the mRNA and protein levels of β2m are concor-dantly overexpressed in the saliva of patients with primary SS.

From the top 27 mRNA found to be overexpressed in WS from primary SS patients, as revealed by microarray profiling, we were able to validate 11 of the genes; expression of the other 16 genes was too low for quantitative PCR assessment. The most overexpressed mRNA was found to be G1P2, which has a function in cell signaling and has been reported to be up-regulated at the mRNA level in minor salivary glands from patients with primary SS (37). There were discrepancies with regard to the fold change as determined by the quantitative PCR and the microarray studies.

There are many factors that may contribute to the observed discrepancies, including the procedures unique to the microarray analysis, such as nonspecific and/or cross-hybridization of labeled targets to array probes, as well as those unique to real-time quantitative RT-PCR, such as amplification biases (45). Also, the increased distance between the location of the PCR primers and the microarray probes on a given gene was found to decrease the correlation between the 2 methods (46). In our study, the amplified RNA used for microarray assay and the unamplified RNA used for the real-time quantitative RT-PCR validation studies can introduce variances in the fold change between the 2 methods. Furthermore, we do not expect there to be perfect correlation between the fold change as determined by quantitative PCR and by microarray analyses, since there is considerable variability in the fold change statistic, especially in the case of genes that are near the limit of detection by quantitative PCR. For genes with expression levels that are too low for the quantitative PCR techniques in current use, it is still possible that they may be validated when the technology improves. Nevertheless, these 11 highly expressed genes, once they are further validated in a new and independent patient cohort, may be used in the clinical detection of primary SS.

There was little correlation between the protein and mRNA markers identified. This has been observed for biologic systems when efforts were made to correlate the gene expression at both the protein and mRNA levels (47,48). In a previous correlation analysis of the human saliva proteome and transcriptome, we demonstrated that complementary validation (e.g., Western blotting for protein or RT-PCR for mRNA) is required in the conduct of RNA–protein correlation studies of individual genes after initial mass spectrometry and expression microarray profiling (49). If mutual validation is performed, there may be higher correlations between the protein and mRNA candidate markers in saliva identified in the present study. Nevertheless, the discrepancy we found may suggest that the combination of both mRNA and protein markers is important for improving the detection of primary SS.

As a result of this preliminary study, a number of promising salivary protein and mRNA candidates that are characteristic of primary SS have been identified. Many of these candidate biomarkers have not previously been associated with SS and, in combination, they may eventually be validated as specific biomarkers of primary SS, thus improving the clinical diagnosis of primary SS. Ideally, the biomarkers would be very specific for primary SS and would discriminate primary SS from other autoimmune diseases of a similar immunopathologic background. Future studies will include new primary SS patients as well as patients with other autoimmune diseases as control groups, aiming to validate the candidate genes either through the use of real-time quantitative RT-PCR for mRNA or immunoassays for proteins. Absolute quantification will provide a cutoff value for each biomarker selected, and combination of the mRNA and protein markers will allow the eventual construction of a multimarker prediction model that can be used as an adjunct to the current diagnostic criteria for the clinical diagnosis of primary SS.

Acknowledgments

Supported by USPHS grants R01-DE-17593 and U01-DE-16275.

Footnotes

AUTHOR CONTRIBUTIONS

Dr. Wong had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Hu, Wang, Yu, Vissink, Pijpe, Loo, Wong.

Acquisition of data. Hu, Wang, Meijer, Ieong, Xie, Henry, Vissink, Pijpe, Kallenberg.

Analysis and interpretation of data. Hu, Wang, Meijer, Yu, Vissink, Pijpe, Wong.

Manuscript preparation. Hu, Wang, Meijer, Yu, Vissink, Pijpe, Kallenberg, Loo, Wong.

Statistical analysis. Yu, Zhou, Elashoff.

REFERENCES

  • 1.Sjogren H. Zur Kenntnis der Keratoconjunctivitis sicca. Acta Ophthalmol. 1933;11 Suppl 2:1–151. doi: 10.1111/j.1755-3768.1971.tb08678.x. [DOI] [PubMed] [Google Scholar]
  • 2.Vitali C, Bombardieri S, Jonsson R, Moutsopoulos HM, Alexander EL, Carsons SE, et al. and the European Study Group on Classification Criteria for Sjögren’s Syndrome Classification criteria for Sjögren’s syndrome: a revised version of the European criteria proposed by the American-European Consensus Group. Ann Rheum Dis. 2002;61:554–558. doi: 10.1136/ard.61.6.554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Fox R. Sjögren’s syndrome. Lancet. 2005;366:321–331. doi: 10.1016/S0140-6736(05)66990-5. [DOI] [PubMed] [Google Scholar]
  • 4.Pijpe J, Kalk WW, van der Wal JE, Vissink A, Kluin PM, Roodenburg JL, et al. Parotid gland biopsy compared with labial biopsy in the diagnosis of patients with primary Sjögren’s syndrome. Rheumatology (Oxford) 2007;46:335–341. doi: 10.1093/rheumatology/kel266. [DOI] [PubMed] [Google Scholar]
  • 5.Malamud D. Oral diagnostic testing for detecting human immunodeficiency virus-1 antibodies: a technology whose time has come. Am J Med. 1997;102:9–14. doi: 10.1016/s0002-9343(97)00032-6. [DOI] [PubMed] [Google Scholar]
  • 6.Kaufman E, Lamster IB. Analysis of saliva for periodontal diagnosis: a review [review] J Clin Periodontol. 2000;27:453–465. doi: 10.1034/j.1600-051x.2000.027007453.x. [DOI] [PubMed] [Google Scholar]
  • 7.Kalk WW, Vissink A, Stegenga B, Bootsma H, Nieuw Amerongen AV, Kallenberg CG. Sialometry and sialochemistry: a non-invasive approach for diagnosing Sjögren’s syndrome. Ann Rheum Dis. 2002;61:137–144. doi: 10.1136/ard.61.2.137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Pijpe J, Kalk WW, Bootsma H, Spijkervet FK, Kallenberg CG, Vissink A. Progression of salivary gland dysfunction in patients with Sjögren’s syndrome. Ann Rheum Dis. 2007;66:107–112. doi: 10.1136/ard.2006.052647. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Hu S, Xie Y, Ramachandran P, Ogorzalek Loo RR, Li Y, Loo JA, et al. Large-scale identification of proteins in human salivary proteome by liquid chromatography/mass spectrometry and two-dimensional gel electrophoresis-mass spectrometry. Proteomics. 2005;5:1714–1728. doi: 10.1002/pmic.200401037. [DOI] [PubMed] [Google Scholar]
  • 10.Hu S, Denny P, Denny P, Xie Y, Loo JA, Wolinsky LE, et al. Differentially expressed protein markers in human submandibular and sublingual secretions. Int J Oncol. 2004;25:1423–1430. [PubMed] [Google Scholar]
  • 11.Ramachandran P, Boontheung P, Xie Y, Sondej M, Wong DT, Loo JA. Identification of N-linked glycoproteins in human saliva by glycoprotein capture and mass spectrometry. J Proteome Res. 2006;5:1493–1503. doi: 10.1021/pr050492k. [DOI] [PubMed] [Google Scholar]
  • 12.Li Y, Zhou X, St John MA, Wong DT. RNA profiling of cell-free saliva using microarray technology. J Dent Res. 2004;83:199–203. doi: 10.1177/154405910408300303. [DOI] [PubMed] [Google Scholar]
  • 13.Hu S, Yu T, Xie Y, Yang Y, Li Y, Tsung S, et al. Discovery of oral fluid biomarkers for human oral cancer using mass spectrometry. Cancer Genomics Proteomics. 2007;4:55–64. [PubMed] [Google Scholar]
  • 14.Li Y, St John MA, Zhou X, Kim Y, Sinha U, Jordan RC, et al. Salivary transcriptome diagnostics for oral cancer detection. Clin Cancer Res. 2004;10:8442–8450. doi: 10.1158/1078-0432.CCR-04-1167. [DOI] [PubMed] [Google Scholar]
  • 15.Castro J, Jimenez-Alonso J, Sabio JM, Rivera-Civico F, Martin-Armada M, Rodriguez MA, et al. for the Grupo Lupus Virgen de las Nieves Salivary and serum β2-microglobulin and γ-glutamyl-transferase in patients with primary Sjögren syndrome and Sjögren syndrome secondary to systemic lupus erythematosus. Clin Chim Acta. 2003;334:225–231. doi: 10.1016/s0009-8981(03)00162-1. [DOI] [PubMed] [Google Scholar]
  • 16.Tishler M, Yaron I, Shirazi I, Levartovsky D, Yaron M. Salivary and serum soluble interleukin-2 receptor in primary Sjögren’s syndrome. Arch Oral Biol. 1999;44:305–308. doi: 10.1016/s0003-9969(98)00114-9. [DOI] [PubMed] [Google Scholar]
  • 17.Tishler M, Yaron I, Shirazi I, Yossipov Y, Yaron M. Increased salivary interleukin-6 levels in patients with primary Sjögren’s syndrome. Rheumatol Int. 1999;18:125–127. doi: 10.1007/s002960050070. [DOI] [PubMed] [Google Scholar]
  • 18.Sfriso P, Ostuni P, Botsios C, Andretta M, Oliviero F, Punzi L, et al. Serum and salivary neopterin and interferon-γ in primary Sjögren’s syndrome: correlation with clinical, laboratory and histopathologic features. Scand J Rheumatol. 2003;32:74–78. doi: 10.1080/03009740310000067. [DOI] [PubMed] [Google Scholar]
  • 19.Ben-Chetrit EF, Rubinow A. Anti-SSA/Ro and anti-SSB/La antibodies in serum and saliva of patients with Sjögren’s syndrome. Clin Rheumatol. 1993;12:471–474. doi: 10.1007/BF02231773. [DOI] [PubMed] [Google Scholar]
  • 20.Witte T. Antifodrin antibodies in Sjögren’s syndrome: a review. Ann N Y Acad Sci. 1995;1051:235–239. doi: 10.1196/annals.1361.064. [DOI] [PubMed] [Google Scholar]
  • 21.Burlage FR, Pijpe J, Coppes RP, Hemels ME, Meertens H, Canrinus A, et al. Variability of flow rate when collecting stimulated human parotid saliva. Eur J Oral Sci. 2005;113:386–390. doi: 10.1111/j.1600-0722.2005.00241.x. [DOI] [PubMed] [Google Scholar]
  • 22.Navazesh M. Methods for collecting saliva. Ann N Y Acad Sci. 1993;694:72–77. doi: 10.1111/j.1749-6632.1993.tb18343.x. [DOI] [PubMed] [Google Scholar]
  • 23.Irizarry RA, Bolstad BM, Collin F, Cope LM, Hobbs B, Speed TP. Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res. 2003;31:e15. doi: 10.1093/nar/gng015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method. Methods. 2001;25:402–408. doi: 10.1006/meth.2001.1262. [DOI] [PubMed] [Google Scholar]
  • 25.R Development Core Team. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2006. [Google Scholar]
  • 26.Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 2004;5:R80. doi: 10.1186/gb-2004-5-10-r80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B. 1995;57:289–300. [Google Scholar]
  • 28.Doniger SW, Salomonis N, Dahlquist KD, Vranizan K, Lawlor SC, Conklin BR. MAPPFinder: using gene ontology and GenMAPP to create a global gene-expression profile from microarray data. Genome Biol. 2003;4:R7. doi: 10.1186/gb-2003-4-1-r7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Der SD, Zhou A, Williams BR, Silverman RH. Identification of genes differentially regulated by interferon α, β, or γ using oligonucleotide arrays. Proc Natl Acad Sci U S A. 1998;95:15623–15628. doi: 10.1073/pnas.95.26.15623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Sanda C, Weitzel P, Tsukahara T, Schaley J, Edenberg HJ, Stephens MA, et al. Differential gene induction by type I and type II interferons and their combination. J Interferon Cytokine Res. 2006;26:462–472. doi: 10.1089/jir.2006.26.462. [DOI] [PubMed] [Google Scholar]
  • 31.Tomosugi N, Kitagawa K, Takahashi N, Sugai S, Ishikawa I. Diagnostic potential of tear proteomic patterns in Sjögren’s syndrome. J Proteome Res. 2005;4:820–825. doi: 10.1021/pr0497576. [DOI] [PubMed] [Google Scholar]
  • 32.Ryu OH, Atkinson JC, Hoehn GT, Illei GG, Hart TC. Rheumatology (Oxford) Vol. 45. 2006. Identification of parotid salivary biomarkers in Sjögren’s syndrome by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry and two-dimensional difference gel electrophoresis; pp. 1077–1086. [DOI] [PubMed] [Google Scholar]
  • 33.Ochi H, Horiuchi I, Araki N, Toda T, Araki T, Sato K, et al. Proteomic analysis of human brain identifies α-enolase as a novel autoantigen in Hashimoto’s encephalopathy. FEBS Lett. 2002;528:197–202. doi: 10.1016/s0014-5793(02)03307-0. [DOI] [PubMed] [Google Scholar]
  • 34.O’Dwyer DT, Smith AI, Matthew ML, Andronicos NM, Ranson M, Robinson PJ, et al. Identification of the 49-kDa autoantigen associated with lymphocytic hypophysitis as α-enolase. J Clin Endocrinol Metab. 2002;87:752–757. doi: 10.1210/jcem.87.2.8205. [DOI] [PubMed] [Google Scholar]
  • 35.Pratesi F, Moscato S, Sabbatini A, Chimenti D, Bombardieri S, Migliorini P. Autoantibodies specific for α-enolase in systemic autoimmune disorders. J Rheumatol. 2000;27:109–115. [PubMed] [Google Scholar]
  • 36.Gottenberg JE, Cagnard N, Lucchesi C, Letourneur F, Mistou S, Lazure T, et al. Activation of IFN pathways and plasmacytoid dendritic cell recruitment in target organs of primary Sjögren’s syndrome. Proc Natl Acad Sci U S A. 2006;103:2770–2775. doi: 10.1073/pnas.0510837103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Hjelmervik TO, Petersen K, Jonassen I, Jonsson R, Bolstad AI. Gene expression profiling of minor salivary glands clearly distinguishes primary Sjögren’s syndrome patients from healthy control subjects. Arthritis Rheum. 2005;52:1534–1544. doi: 10.1002/art.21006. [DOI] [PubMed] [Google Scholar]
  • 38.Bave U, Nordmark G, Lovgren T, Ronnelid J, Cajander S, Eloranta ML, et al. Activation of the type I interferon system in primary Sjögren’s syndrome: a possible etiopathogenic mechanism. Arthritis Rheum. 2005;52:1185–1195. doi: 10.1002/art.20998. [DOI] [PubMed] [Google Scholar]
  • 39.Baechler EC, Batliwalla FM, Reed AM, Peterson EJ, Gaffney PM, Moser KL, et al. Gene expression profiling in human autoimmunity. Immunol Rev. 2006;210:120–137. doi: 10.1111/j.0105-2896.2006.00367.x. [DOI] [PubMed] [Google Scholar]
  • 40.Shiozawa S, Tanaka Y, Shiozawa K. Single-blinded controlled trial of low-dose oral IFN-α for the treatment of xerostomia in patients with Sjögren’s syndrome. J Interferon Cytokine Res. 1998;18:255–262. doi: 10.1089/jir.1998.18.255. [DOI] [PubMed] [Google Scholar]
  • 41.Ferraccioli GF, Salaffi F, De Vita S, Casatta L, Avellini C, Carotti M, et al. Interferon α-2 (IFNα2) increases lacrimal and salivary function in Sjögren’s syndrome patients: preliminary results of an open pilot trial versus OH-chloroquine. Clin Exp Rheumatol. 1996;14:367–371. [PubMed] [Google Scholar]
  • 42.Nordmark G, Alm GV, Ronnblom L. Mechanisms of disease: primary Sjögren’s syndrome and the type I interferon system. Nat Clin Pract Rheumatol. 2006;2:262–269. doi: 10.1038/ncprheum0173. [DOI] [PubMed] [Google Scholar]
  • 43.Swaak AJ, Visch LL, Zonneveld A. Diagnostic significance of salivary levels of β2-microglobulin in Sjögren’s syndrome. Clin Rheumatol. 1988;7:28–34. doi: 10.1007/BF02284053. [DOI] [PubMed] [Google Scholar]
  • 44.Maddali Bongi S, Campana G, D’Agata A, Palermo C, Bianucci G. The diagnosis value of β2-microglobulin and immunoglobulins in primary Sjögren’s syndrome. Clin Rheumatol. 1995;14:151–156. doi: 10.1007/BF02214934. [DOI] [PubMed] [Google Scholar]
  • 45.Chuaqui RF, Bonner RF, Best CJ, Gillespie JW, Flaig J, Hewitt SM, et al. Post-analysis follow-up and validation of microarray experiments. Nat Genet. 2002;32 Suppl:509–514. doi: 10.1038/ng1034. [DOI] [PubMed] [Google Scholar]
  • 46.Etienne W, Meyer MH, Peppers J, Meyer RA., Jr Comparison of mRNA gene expression by RT-PCR and DNA microarray. Biotechniques. 2004;36:618–626. doi: 10.2144/04364ST02. [DOI] [PubMed] [Google Scholar]
  • 47.Gygi SP, Rochon Y, Franza BR, Aebersold R. Correlation between protein and mRNA abundance in yeast. Mol Cell Biol. 1999;19:1720–1730. doi: 10.1128/mcb.19.3.1720. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Baliga NS, Pan M, Goo YA, Yi EC, Goodlett DR, Dimitrov K, et al. Coordinate regulation of energy transduction modules in Halobacterium sp. analyzed by a global systems approach. Proc Natl Acad Sci U S A. 2002;99:14913–14918. doi: 10.1073/pnas.192558999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Hu S, Li Y, Wang J, Xie Y, Tjon K, Wolinsky L, et al. Human saliva proteome and transcriptome. J Dent Res. 2006;85:1129–1133. doi: 10.1177/154405910608501212. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES