Potential role of oral microbiota and autoantibodies against either aquaporin-5 or aquaporin-1 in the pathogenesis of Sjögrens syndrome

Abstract

Background Sjögrens syndrome (SS) is an autoimmune disorder that primarily targets the salivary and lacrimal glands, leading to the dryness of the mouth and eyes. Aquaporins (AQPs) are small hydrophobic transmembrane water channel proteins that are widely expressed in animals and plants as well as in bacteria and archaea. The expressions of mRNA and proteins for AQP5 and AQP1 have been shown in human salivary glands. AQP5 is expressed at the apical membrane of acinar cells and intercalated ducts in the lacrimal and salivary glands, while AQP1 expression is confined to the capillary endothelial and myoepithelial cells. AQP5 plays an important role in the production of saliva. AQP5-deficient mice secrete hypertonic saliva with substantially reduced volume. AQPs, porins, or glycerol transport proteins of some oral bacteria have a high degree of homology with human AQP5 and AQP1. B cell epitope analysis showed the presence of evolutionary conserved linear epitopes in the extracellular loops of humans AQP5/1 and the bacterial porins, suggesting a possibility that antibodies produced against the bacterial porins may cross react with the extracellular loops of human AQP5/1 and block the entry of water molecules into the channel. Growing evidence also suggests that dysbiosis of microbiota is associated with pathogenesis of several autoimmune diseases. However, the role of oral microbiota in the etiopathogenesis of SS in not known. The purpose of this study was to investigate the potential role of autoantibodies against human AQP5 and AQP1 and oral microbiota in the pathogenesis of SS. Material and methods Frozen sections of mouse submandibular salivary glands, Chinese hamster ovary (CHO) cells overexpressing green fluorescent protein (GFP) alone or human AQP5-GFP as a fusion protein, and Madin-darby canine kidney cells (MDCK) overexpressing AQP5 or AQP1 were used in the indirect immunofluorescence assay (IIFA) to detect anti-AQP5/1 autoantibodies in the sera from patients with primary SS. The lysates of human embryonic kidney (HEK)-293 cells overexpressing the AQP5/1-GFP fusion protein or GFP alone were used for immunoprecipitation. Synthetic peptides corresponding to the extracellular loops of human-AQP5 were used to map the epitopes of AQP5. The communities of oral bacteria from healthy control (HC), patients with dry mouth due to medication as sicca control (SC) and SS patients were analyzed by pyrosequencing. Human salivary gland (HSG) cells were infected with selected oral bacteria to examine the bacterial invasion and expression of cytokines and chemokines. C57BL/6 mice were randomly divided into four groups (n=5 per group)

Sham, Fn, Pm and Fn-Pm groups. Fn, Pm, Fn-Pm groups were sublingually inoculated six times with 2x109 cells of F. nucleatum (Fn), P. melaninogenica (Pm), and Fn plus Pm together, respectively, The pilocarpine stimulated salivary flow rate of each mice was measured at two weeks interval. Mouse salivary glands collected for histopathology and mRNA extraction to analyze the expression of cytokines/chemokines. Sera were used to detect anti-AQP5 autoantibodies. Results Serum IgG from the SS patients, but not from the control subjects, stained acinar cells in the mouse salivary glands, the signals of which colocalized with those of AQP5-specific antibodies. Serum IgG from the SS patients also selectively stained AQP5-GFP expressed in CHO cells. However, both the control and SS sera immunoprecipitated the AQP5-GFP in protein lysate, suggesting that autoantibodies againstAQP5 were also present in the control sera. The screening of 53 control and 112 SS serum samples by IIFA using the AQP5-expressing MDCK cells revealed the presence of significantly higher levels of anti-AQP5 IgG in the SS samples than in the control samples with sensitivity of 0.73 and a specificity of 0.68. Anti-AQP1 autoantibodies were also detected only in the 16.07% SS samples. Furthermore, the presence of anti-AQP5 autoantibodies, but not that of anti-AQP1 autoantibodies, was associated with low resting salivary flow in SS patients. Synthetic peptides corresponding to the extracellular loops of AQP5 significantly inhibited the binding of anti-AQP5 autoantibodies in the SS samples to MDCK-AQP5 cells. Pyrosequencing data revealed drastic changes in the bacterial communities associated with the dryness of mouth. Although the bacterial communities of SC and SS were similar to each other, the relative abundance of species/phylotypes were different between SC and SS. Logistic regression analysis revealed that Pm was associated with increased SS risk (OR 5.396 per 1% increase, CI 95% 1.793-16.238, p=0.003). Pm and Fn were chosen for in-vitro and in-vivo experiments because Pm has a porin with high homology to AQP5 and Fn has high invasive ability among the several oral species with known invasive capabilities. HSG cells produced the high levels of proinflammatory cytokine IL-6 and C-X-C motif chemokine ligand (CXCL)-10 in response to Fn but not to Pm in a dose dependent manner. In addition, sublingual inoculation of Fn and Pm into C57BL/6 mice produced sicca symptoms with mild recruitment of T cells to the salivary gland tissues in a few mice. Conclusion In conclusion, autoantibodies to human AQP5 and altered oral microbiota may contribute to the etiopathogenesis of SS. Anti-AQP5 autoantibodies detected in the sera from SS patients may be a novel biomarker of SS and may provide new insight into the pathogenesis of SS.

Keywords: Sjögrens syndrome, autoantibodies, AQP5, AQP1, Pyrosequencing analysis