Dissertation committee:
Μιχαήλ Κουτσιλιέρης, Καθηγητής, Ιατρική Σχολή Αθηνών, ΕΚΠΑ
Δημήτριος Μπούμπας, Καθηγητής, Ιατρική Σχολή Αθηνών, ΕΚΠΑ
Χρηστος Κόνσουλας, Αναπληρωτής Καθηγητής, Ιατρική Σχολή Αθηνών, ΕΚΠΑ
Κλειώ Μαυραγάνη, Επίκουρη Καθηγήτρια, Ιατρική Σχολή Αθηνών, ΕΚΠΑ
Ελένη Κοτσιφάκη, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή Αθηνών, ΕΚΠΑ
Αθανάσιος Αρμακόλας, Επίκουρος Καθηγητής, Ιατρική Σχολή Αθηνών, ΕΚΠΑ
Αναστάσιος Φιλίππου, Επίκουρος Καθηγητής, Ιατρική Σχολή Αθηνών, ΕΚΠΑ
Summary:
Sjögren syndrome (SS or autoimmune epithelitis) is a chronic autoimmune disease characterized by a lymphocytic infiltration and consequent progressive destruction of the exocrine glands (mainly salivary and lacrimal glands) leading to the loss of secretory function and mucosal dryness (xerostomia and xerophtalmia). SS is divided in primary (pSS) and secondary (sSS) when associated with other autoimmune diseases such as RA.
A growing body of evidence suggests the pivotal role of Interferon (IFN) in the pathogenesis of many autoimmune diseases including pSS. IFNs are ubiquitous cytokines expressed by all mononuclear blood cells (especially plasmacytoid dendritic cells – pDCs) in response to mostly viral infections. IFNs are divided in 3 major types (known as type I, type II, and type III). In total, IFNs promote or inhibit the expression of more than 300 genes that are implicated in various physiological and pathological processes of innate and adaptive immunity.
Genetic expression and microarray studies in both peripheral blood mononuclear cells and salivary glands of pSS patients provided evidence supporting the crucial role of IFN in pSS. Single nucleotide polymorphisms (SNPs) in genes implicated in IFN pathway (IRF5, STAT4, TNPO3) have been strongly associated with increased risk of pSS development. Additionally, histopathological data from salivary glands of pSS patients revealed the increased presence of pDCs (main producers of type I IFN). Inappropriately increased expression of endogenous nucleic acids either as a result of defective methylation or other unknown mechanisms was shown to activate type I IFN system in the salivary glands of pSS patients via the activation of TLR-7/-9 and other cytoplasmic receptors.
SS is a systemic disease with up to 40% of patients showing evidence of diverse clinical features from multiple affected organs (extraglandular manifestations). Interestingly the majority of pSS patients presents psychiatric comorbidities such as severe and unexplained fatigue, depression, anxiety disorder and sleep impairment. Moreover, comparably to RA and SLE, recent studies underscored the strong association between SS and premature atherosclerosis as well as the heightened cardiovascular risk in pSS patients.
Based on robust data demonstrating that exogenous administration of type I IFN in patients with hepatitis C and other diseases can bring out or deteriorate fatigue and other psychiatric manifestations as well as the recently observed association of type I IFN with subclinical atherosclerosis we aimed to explore the possible association between type I IFN induced gene expression in PBMCs and both the psychiatric comorbidities (anxiety disorder, depression, fatigue and sleep impairment) and the presence of premature atherosclerosis in the pSS patients of our cohort.
In a cohort of 106 pSS patients, fatigue, depression, anxiety disorder, personality features and impaired sleep were assessed using the FACIT-F, Zung Depression Scale, Stae-Trait Anxiety Inventory, Eysenck Personality Questionnaire Scale and Athens insomnia Scale respectively. In peripheral blood samples, quantitative reverse transcriptase-polymerase chain reaction technique was adopted in order to measure the levels of indoleamine 2,3-dioxygenase (IDO-1) as well as other IFN inducible genes of type I IFN (IFIT-1, MX-1 and IFI-44) and of type II IFN (MIG-1, GBP-1). IDO-1 converts tryptophan, the precursor molecule of serotonin (5-Hydroxytryptamine), to kinurenine thus leading to decreased serotonin levels in the brain which is known to play a crucial role in the pathogenesis of psychiatric manifestations. Additionally, kinurenine metabolism products are thought to negatively affect central nervous system function.
Given that type I IFN promotes the differentiation and activation of B-cells via BAFF gene induction, we then measured serum BAFF levels using ELISA as well as multiple autoantibodies against non-specific and tissue-specific autoantigens. The latter included the autoantibodies against adrenal autoantigen 21-hydroxylase (anti-21 (OH)), based on data from a previous study that associated these autoantibodies with dampened cortisol production which might contribute to the presence of fatigue in pSS patients.
Fatigue- defined as FACIT-F <30- was detected in 32 out of 106 patients with pSS. In contrast to our original hypothesis, a negative correlation was detected between fatigue and type I IFN levels. Though a definite explanation for the current observation is not readily available, we presume that type I IFN inhibition of IL-1 system might be responsable for fatigue’s decreased prevalence in pSS patients with high type I IFN levels as shown in a recent study where pharmacological inhibition of IL-1 receptor lead to significant improvement in fatigue levels in pSS patients.
Based on the hypothesis that B-cell activation is induced by type I IFN stimulated genes (including BAFF), we next explored whether biomarkers of B cell activation, including autoantibodies, BAFF serum levels and hyperglobulinemia, are associated with fatigue in pSS patients. No significant correlation was observed between fatigue and positive ANA titers, RF, anti-SSA/Ro, anti-SSB/La, anti-TPO, anti-TG, AMA and hyperglobulinemia. Moreover, FACIT-F score levels were independent of the number of autoantibodies detected in the serum of pSS patients [ANA≥1/320, RF, anti-Ro/SSA, anti-La/SSB, AMA, anti-TPO, anti-Tg, anti-21(OH)]. Accordingly, serum BAFF (sBAFF) levels showed no association with fatigue (mean ± SD: 1,137 ± 637 in fatigued vs 986 ± 367 pg/ml in non-fatigued pSS patients, p=0.58).
Additionally, the presence of anti-21(OH) antibodies was not associated with fatigue (9.5% of fatigued pSS patients vs 8.2% of non-fatigued, p=1.000), thus not supporting the hypothesis that subclinical adrenal insufficiency might be involved in pSS-associated fatigue. Taken together these data indicate that B-cell activation as explored through multiple biomarkers has no association with fatigue in patients with pSS.
However, multivariate analysis in our study cohort revealed that neuroticism [OR=6.9 (95%CI: 1.7-28.0)], depression [OR= 3.0 (95%CI: 0.8-11.0)] and fibromyalgia [OR= 5.5 (95%CI: 1.1-27.7)] are independent risk factors for fatigue and should, therefore, be taken into account and addressed in everyday clinical practice with active collaboration between rheumatologists and mental health professionals. Further studies are warranted in order to investigate a potential contribution of these and other biological pathways in pSS-associated fatigue.
A second aspect of our study was focused in a subgroup of 59 pSS patients from our cohort who underwent ultrasound scan with Doppler technique of the carotid and femoral arteries in order to assess the presence premature/subclinical atherosclerosis. Subclinical atherosclerosis was defined by the presence of plaque and/or arterial wall thickening (defined as intima media thickness (IMT) score >0.90 mm) in carotid and femoral arteries as determined by ultrasound.
We initially chose 31 age-/sex-matched pSS patients, to investigate the association between the levels of type I and type II IFN inducible genes expression and the presence of atherosclerotic plaque and of increased IMT levels. No statistically significant difference was observed between increased type I IFN levels and atherosclerotic plaque (p=0.479) as well as hightened IMT levels (p=0.715). Similar results were noted for type II IFN signature. However BAFF mRNA expression levels were significantly associated with plaque formation (p=0.03) in these patients indirectly supporting the hypothesis implicating type I IFN system activation in pSS-related atherosclerosis.
Given that sleep impairment and psychological disorders represent common comorbidities in pSS patients as well as significant risk factors for atherosclerosis in the general population, we then explored the possible association between psychological comorbidities and atherosclerosis in the 59 pSS patients from our cohort. 41 out of 59 (69.5%) of these pSS patients had evidence of atherosclerotic plaque and increased IMT levels as determined by ultrasound. In the univariate analysis, age and high triglyceride serum levels were associated both with atherosclerotic plaque and high IMT levels. Hypertension was significantly associated only with IMT levels. Whilst in the univariate analysis we observed a strong association of atherosclerotic plaque with state anxiety and impaired sleep, in the multivariate analysis sleep impairment alone emerged as an idependent risk factor for atherosclerotic plaque in pSS patients (OR = 4.2, 95% CI =1.1-15.6, p = 0.03).
This is the first time a strong association between sleep impairment and subclinical atherosclerosis was observed in pSS patients. Provided these results are confirmed in larger studies, they will lead to the inclusion of sleep impairment in the list of risk factors for atherosclerosis in pSS patients.
In conclusion, our study showed inverse correlation of type I IFN system activation with fatigue and positive association with premature atherosclerosis through BAFF activation in the setting of pSS. However it also provided new and significant associations of these comorbidities with the psychiatric/psychological profile of these patients.
