Metabolic Profile of Whole Unstimulated Saliva in Patients with Sjögren’s Syndrome

“

TL;DR

This study used ¹H-NMR metabolomics to compare the salivary metabolite profiles of 7 female patients with primary Sjögren's Syndrome (pSS) and 6 matched healthy controls. Methylamine was significantly higher in pSS saliva (p < 0.05), while proline and macromolecule signals were higher in controls, though most other metabolite differences did not reach statistical significance. The findings suggest NMR-based salivary metabolomics may be a promising screening tool for pSS, but larger studies are needed to confirm diagnostic utility.

Full summary

4,490 CHARS

**Background:** Primary Sjögren's Syndrome (pSS) is a chronic autoimmune disease characterized by exocrine gland dysfunction, particularly xerostomia and keratoconjunctivitis sicca. Diagnosis remains challenging as no single gold-standard test exists. Salivary metabolomics using Nuclear Magnetic Resonance (NMR) spectroscopy offers a non-invasive, reproducible approach to identify potential biomarkers. Previous NMR studies on stimulated saliva identified metabolites such as choline, taurine, alanine, and glycine as significantly elevated in pSS. This study aimed to expand the NMR metabolite database for pSS using unstimulated whole saliva and compare the metabolome of pSS patients with healthy controls.

**Methods:** This single-center case-control study (Ethical Committee Code: 1183/2018/SPER/AOUMO) selected 7 female pSS patients (mean age 65.4) from a cohort of 60 confirmed diagnoses, after applying strict exclusion criteria (head/neck radiotherapy, HIV/HCV, lymphoma, sarcoidosis, graft-versus-host disease, anticholinergic drugs, oral lesions, active infections, diabetes, smoking, oncological diseases). Six sex- and age-matched healthy controls were recruited. Unstimulated whole saliva was collected between 9:00-12:00 AM under standardized conditions (no food/drink/smoking 1 hour prior, mouth rinse with water). Samples were frozen in liquid nitrogen, stored at -80°C, then thawed, centrifuged at 15,000×g for 10 min at 4°C, and buffered with trisodium phosphate (pH 7.45). ¹H-NMR metabolomics was performed blindly using a Bruker Avance III HD 600 MHz spectrometer with CPMG pulse sequence. 2D experiments (COSY, TOCSY, HSQC) aided metabolite identification. Approximately 50 metabolites were detected. Multivariate analysis (PCA, PLS-DA) was performed on binned and normalized spectra (0.8-8.5 ppm). Deconvoluted areas of 30 selected metabolite signals were analyzed by t-test and fold-change.

**Key Results:** Visual comparison of average spectra suggested higher levels of 4-hydroxyphenylacetic acid, phenylacetic acid, δ-valerolactam, taurine, choline, ethanolamine, trimethylamine, methylamine, 5-aminopentanoic acid, and butyric acid in pSS samples, and higher formic acid, monosaccharides, macromolecules, proline, and lactic acid in controls. However, the t-test confirmed statistically significant differences (p < 0.05) only for methylamine (higher in pSS, fold-change 2.3, p = 0.01 for absolute areas; p = 0.05 for normalized areas), proline (higher in HCs), and residual macromolecule signals (4.2-4.5 ppm, higher in HCs). PCA showed partial clustering in the PC2/PC3 score plot (19.6% and 12.2% variance explained). PLS-DA improved separation between groups. Fold-change analysis highlighted metabolites such as phloretic acid (FC 4.0, p = 0.14), δ-valerolactam (FC 3.2, p = 0.38), trimethylamine (FC 3.0, p = 0.08), and 2-methylbutanoic acid (FC 2.9, p = 0.12) as elevated in pSS, though not statistically significant. The pSS group metabolome was characterized by higher amines (methylamines, putrescine, sarcosine), organic acids (propanoic, butanoic, pentanoic, phenylacetic, phenylpropanoic series), and 5-aminopentanoic acid/δ-valerolactam. The HC group showed higher lactic, succinic, pyruvic, and formic acids, and amino acids.

**Clinical Implications:** This preliminary study demonstrates that ¹H-NMR metabolomics of unstimulated whole saliva can distinguish pSS patients from healthy controls, with methylamine emerging as a statistically significant differentiating metabolite (p < 0.05). The reduction in fucose and certain amino acids (glycine, tyrosine) in pSS saliva aligns with prior GC-MS findings by Kageyama et al., though contrasts with some NMR studies on stimulated saliva. The elevated methylamine may reflect altered bacterial metabolism due to changes in salivary composition and buffer capacity. The study's main limitation is the very small sample size (7 pSS, 6 controls), which limited statistical power and prevented clear clustering in PCA. Only PLS-DA provided reasonable group separation. Additionally, potential microbial contamination from dental plaque may have influenced metabolites of bacterial origin (acetate, propionate, trimethylamine). Despite these limitations, the results support the potential of NMR-based salivary metabolomics as a non-invasive screening and monitoring tool for pSS. Larger studies are required to validate these preliminary findings and establish diagnostic and prognostic biomarkers.