Type I and II interferon signatures in Sjogren’s syndrome: Contributions in distinct clinical
phenotypes and Sjogren’s related lymphomagenesis
Adrianos Nezos1, Fotini Gravani2, Anna Tasidou3, Michael Voulgarelis4, Haralampos M. Moutsopoulos4, Mary K. Crow5 and Clio P. Mavragani1,4
1Department of Physiology, School of Medicine, National University of Athens, Athens, Greece, 2Department of Rheumatology, General Hospital of Athens “G.Gennimatas”, 3Department of Hemopathology, Evangelismos Hospital, Athens, Greece, 4Department of Pathophysiology, School of Medicine, National University of Athens, Athens, Greece, 5Mary Kirkland Center for Lupus Research, Hospital for Special Surgery, Weill Medical College of Cornell University, New York, NY, USA.
Both type I and II interferons (IFNs) have been implicated in the pathogenesis of Sjogren’s syndrome (SS). We aimed to explore the contribution of type I and II IFN signatures in the generation of distinct SS clinical phenotypes including non-Hodgkin’s lymphoma (NHL) development, a major SS complication.
MATERIALS & METHODS
Transcript levels of 3 interferon inducible genes (IFIGs) preferentially induced by type I IFN, 2 IFIGs preferentially induced by IFNγ, and IFNα and IFNγ genes were quantitated by Real-time polymerase chain reaction in:
* Peripheral blood (PB) from: SS patients (n=31), SS patients complicated by lymphoma (SSL, n=13) and healthy donors (HD, n=30)
* Minor salivary gland tissues (MSG) derived from 31 SS patients, 10 SSL patients and 17 sicca controls (SC) (Demographics and clinical data of study participants in Table 1).
•In PB and MSG tissues, overexpression of both type I and type II IFIGs was observed in SS patients versus HC and SC, respectively, with a predominance of type I IFN signature in PB and a type II IFN signature in MSG tissues (Figures 1 and 2).
• In MSG tissues derived from SSL patients we observed lower IFNα, but higher IFNγ and type II IFIG transcripts compared to both SS and SC. In ROC curve analysis, IFNγ/IFNα ratio in MSG tissues showed the best discrimination for lymphoma development, with an area under the curve of 0.88 (95% CI:0.72-1.00, p-value: 0.001) (Figure 3).
• IFNα mRNA expression in MSG tissues was also associated with the mRNA expression of the apoptotic genes TNF-related apoptosis-inducing ligand (TRAIL), tripartite motif containing 21 (TRIM21)/ribonucleoprotein Ro52 and tumor protein 53 (Figure 4).
• SS patients with salivary gland enlargement, lymphopenia, anti-Ro/SSA antibodies and hypergammaglobulinemia exhibited higher type I IFN scores in PB compared to their counterparts without those features; hypergammaglobulinemia was associated with increased type II IFN scores in PB (Figure 5).
• In MSG tissues, both type I IFN score and IFNα transcripts were higher in patients with arthralgias; IFNγ mRNA levels were increased in patients with Raynaud’s phenomenon, purpura and low C4 levels (Figure 6).
•Both type I and II IFN scores were significantly associated with PB and MSG B cell activating factor (BAFF) transcript levels (Figure 7).
Discrete expression patterns of type I and II IFN signatures might be related to distinct clinical SS features and SS related lymphomagenesis. IFNγ/IFNα mRNA ratio in diagnostic salivary gland biopsies is proposed as a novel histopathological biomarker for the prediction of in situ lymphoma development in the setting of SS.