Expression and function of KLRG1 and IL1R8 in systemic lupus erythematosus

Background: Systemic Lupus Erythematosus (SLE) is a multifactorial autoimmune disease, characterized by several immunological alterations and different clinical phenotypes1. KLRG1 is a transmembrane protein expressed in humans on NK cells and on CD4+ and CD8 + T cells. KLRG1 inhibits the cytotoxic activity of NK cells against tissues expressing its principal ligands: E-cadherin, N-cadherin and R-cadherin. Moreover, it has been shown that KLRG1 has an inhibitory effect on CD8+ T cells proliferative capacity and also their effector functions2. Thus, KLRG1-mediated signal might have a role in preventing autoimmunity, increasing the activation threshold of NK cells and T cells. Recently, KLRG1 gene emerged as a disease susceptibility gene for Systemic Lupus Erythematosus in four different ethnic groups3. Interleukin -1R like receptors (ILRs) and Toll Like Receptors (TLRs) are key receptors of innate immunity, inflammation and also adaptive immune responses. IL1R8, also known as SIGIRR and TIR8, is a member of ILRs family, it is ubiquitously expressed and it is a regulatory molecule with an inhibitory function towards other ILRs and TLRs receptors4. Uncontrolled or deregulated activation of ILRs- or TLRs- dependent inflammatory and immune responses can cause tissue damage and acute or chronic inflammatory disorders. SLE mice models IL1R8 knock-out develop massive lymphoproliferations, serositis, glomerulonephritis, and increased autoantibodies production. In humans there are only a few data, showing a relationship between IL1R8 and SLE pathogenesis. Objective: The principal aim of this project was to characterize KLRG1 and IL1R8 expression on NK cells, NKT cells, CD56+ CD3bright cells, CD4+ and CD8+ T cells (naïve, central memory, effector memory and effector) from SLE patients compared to healthy subjects (HS). Expression of these receptors have been investigated through flow cytometry analysis after isolation of PBMCs from peripheral blood. Possible correlations of these results with clinical data, including SLE disease activity score (SLEDAI-2K), were also investigated. As secondary aims, functional studies about KLRG1-mediated signaling on NK cells in SLE have been performed. Results and methods: Blood samples were obtained from SLE patients and HS. Peripheral blood mononuclear cells (PBMCs) were obtained by density gradient centrifugation (Lympholyte-H; Cedarlane Laboratories, Hornby, Ontario, Canada), and phenotypic characterization was performed. All experiments were conducted by flow cytometry. For the statistical analysis Mann-Whitney U test and Spearman’s rank correlation test were used appropriately. Eighteen patients (18F, median age 38 years IQR 17, median disease duration 7.5 years IQR 12.75) affected by SLE according to the 1997 ACR criteria, and twelve healthy subjects (12F, median age 34 years, IQR 18.5) were enrolled for the evaluation of KLRG1 expression on NK cells. Eighteen SLE patients (18F, median age 38 years IQR 22.5, median disease duration 7 years IQR 8) and fourteen healthy subjects (14F, median age 31.5 years, IQR 23) were enrolled for the characterization of IL1R8. Two lupus patients with active disease (SLEDAI-2K=6 and SLEDAI-2K=14) were enrolled for the in vitro study about hydroxychloroquine (HCQ) influence on KLRG1 expression. Preliminary in vitro functional studies about KLRG1-mediated signaling on NK cells have been also performed. All patients were enrolled from Sapienza Lupus Cohort. Disease activity was measured by SLEDAI-2K. KLRG1 was expressed in a lower amount on total NK cells from SLE patients compared to HS (p=0.0003); analysis of CD56dim and CD56bright populations showed less KLRG1 in SLE patients compared to HS (p=0.0017 and p=0.0079, respectively). KLRG1 was also expressed in a lower amount on NKT cells and CD56+ CD3bright cells from SLE patients compared to HS (p=0.0048 and p=0.0016). Examination of CD4+ and CD8+ T cell subsets showed less quantity of KLRG1 in SLE patients compared to HS on CD4+ naïve (p=0.0009), effector memory (p=0.04), effector (p=0.029) and also on CD8+ naïve (p=0.0027) and effector memory T cells (p=0.0185). The lower levels of KLRG1 on NK cells inversely correlated with the SLEDAI-2K (p=0.034, r=-0.5) and were inversely associated with arthritis (p=0.0017, r=-0.4), hematological disorders (p=0.0016, r=-0.6) and anti-dsDNA (p=0.01, r=-0.5). A direct association between KLRG1 expression on NK cells and the use of HCQ was found (p=0.0002). Analysis of the two main subpopulations of NK cells showed an inverse association between KLRG1 on CD56dim cells and arthritis (p<0.0001, r=-0.4), hematological disorders (p=0.0009, r=-0.6), the SLEDAI-2K (p=0.01, r=-0.5), and the presence of anti-dsDNA (p=0.009, r=-0.4). KLRG1 on CD56dim was also directly associated with the use of HCQ (p=0.0005). KLRG1 on CD56bright cells was also inversely associated with arthritis (p=0.01, r=-0.26) and directly associated with the use of HCQ (p=0.0064). Moreover, inverse associations were also found between KLRG1 on CD4+ naïve, effector memory, effector T cells and arthritis (p<0.0001 r=-0.5, p<0.0001 r=-0.3 and p<0.0001 r=-0.4 respectively) and between CD8+ naïve T cells and arthritis (p>0.0001, r=-0.5) and anti-dsDNA (p=0.009, r=-0.4) and between CD8+effector memory T cells and arthritis (P<0.0001, r=-0.29). There was no statistically significant difference about IL1R8 expression on NK cells subpopulations, on NKT cells and on CD56+CD3bright cells and T cells subpopulations between SLE patients and HS, despite in this last group there was a tendency to a higher expression in SLE patients compared to HS. PBMCs of three lupus patients with active disease, who were not taking HCQ, have been treated with 2 different doses of HCQ. After 24 hours, treatment with HCQ increased KLRG1 expression on NK cells in a dose-dependent fashion, even though without reaching a statistically significant difference. Preliminary in vitro functional studies in two HS and one SLE patient, confirmed inhibition of NK cells IFN and CD107a production on KLRG1+ cells when treated with E-cadherin, in the two healthy subjects. In the lupus patient compared to the two healthy subjects, E-cadherin does not seem to be able to interfere with CD107a mobilization but it can reduce IFN production. More experiments, extended to a larger population, are needed to evaluate if there is a difference about the grade of this inhibition between SLE patients and HS. Conclusions: This study explored for the first time a possible involvement of KLRG1 receptor in Systemic Lupus Erythematosus pathogenesis. SLE patients appear to express less amount of this receptor compared to healthy subjects on their lymphocytes, especially on NK cells population. KLRG1 expression on NK cells inversely associates with the SLEDAI-2K and directly associates with the use of HCQ. Moreover, after in vitro treatment with HCQ, KLRG1 levels on NK cells tends to increase. In healthy subjects, KLRG1 binding to its ligand E-cadherin inhibits NK cells activity. This inhibition seems to be altered in SLE, especially for the NK cells degranulation capacity. These results suggest a possible role of KLRG1 in the pathogenesis of this disease. However, more experiments, extended to larger cohorts of patients, are necessary to better understand its function. IL1R8 is a well-known receptor for its role as a negative regulator of immune responses and inflammation. In this study no differences emerged about its expression between SLE patients and healthy subjects. More studies, including larger populations are necessary to better understand if this receptor might actually have a role in this disease.