CXCL9 and its Receptor CXCR3, an Important Link Between Inflammation and Cardiovascular Risks in RA Patients

Cardiovascular disease (CVD) is the most common cause of mortality in rheumatoid arthritis (RA), and Inflammation has a decisive role in its pathogenesis. CXCL9 contributes to multi aspects of inflammatory reactions associated with the pathogenesis of CVD. In the current study, we evaluated the association of plasma CXCL9 and CXCR3 gene expression with Cardiovascular risk factors in RA patients for the first time. Thirty newly diagnosed, 30 on-treatment RA patients, and 30 healthy subjects were recruited in this study. The plasma concentration of CXCL9 and CXCR3 gene expression were measured using ELISA and Real-Time PCR, respectively. The CVD risk was evaluated using Framingham Risk Score (FRS) and Systematic Coronary Risk Evaluation (SCORE). The plasma levels of CXCL9 were significantly higher in the newly diagnosed and on-treatment RA patients compared to the control group (P < 0.0001 and P < 0.001, respectively). Also, The CXCR3 gene expression was strongly elevated in newly diagnosed and on-treatment patients (P < 0.001 and P < 0.01, respectively). The CXCL9 and CXCR3 were significantly associated with RA disease activity (P = 0.0005, r = 0.436; P = 0.0002, r = 0.463, respectively). The FRS was remarkably higher in newly diagnosed and on-treatment patients (P = 0.014 and P = 0.035, respectively). The CXCR3 gene expression significantly correlated with age, systolic blood pressure, FRS, and SCORE (P = 0.020, r = 0.298; P = 0.006, r = 0.346; P = 0.006, r = 0.349; P = 0.007, r = 0.341, respectively). The CXCL9 plasma concentration had a significant negative correlation with plasma HDL and LDL levels (P = 0.033, r = -0.275; P = 0.021, r = -0.296, respectively). CXCL9 and CXCR3 correlates with different variables of CVD in RA.


INTRODUCTION
Rheumatoid arthritis (RA) is a chronic progressive autoimmune disease with the highest occurrence among autoimmune rheumatic musculoskeletal conditions and is associated with multiple aspects of inflammation [1][2][3].The prevalence of rheumatoid arthritis varies globally, with a higher prevalence in industrialized countries and generally affecting 0.1-2% of the adult world's population [4,5].Rheumatoid arthritis is mainly characterized by synovial inflammation, cartilage, and bone invasion, and ultimately destruction and deformation of joints, especially in the hands and feet [1,6,7].
Systemic inflammation in rheumatoid arthritis can also affect any of the tissues and organs of the body [6,8].Among variable extra-articular manifestations of RA, cardiovascular disease (CVD) has particular importance [8].The risk of CVD in the RA population is 1.5-2 times more than the healthy subjects, and roughly 50% of deaths in the RA population are attributed to cardiovascular diseases [8][9][10][11].CVD encompasses a series of disorders related to the heart and circulatory system, including heart failure, atherosclerosis, cerebrovascular, and several other cardiac abnormalities.Ischemic heart diseases following atherosclerosis and congestive heart failure are the main clinical manifestations of cardiovascular diseases in the RA population [8,11,12].
CVD risk calculators designed for the general population, including the Systematic Coronary Risk Evaluation (SCORE) calculator, the Framingham Risk Score (FRS), and the Reynolds Risk Score, are not suitable for assessing CVD risk in RA individuals and often underestimate the CVD risk in this population.Most calculation models use blood pressure, lipid levels, smoking status, age, and gender to predict the risk of cardiovascular disorders and do not include the RA-related risk factors for the development of CVD, including inflammatory mediators and medication used for RA treatment.Despite many efforts that have been made to develop consensus CVD risk calculators for the RA population, promising results have not been achieved.The new EULAR (European Alliance of Associations for Rheumatology) recommendations for CVD risk management further accentuate the importance of designing a customized model for RA patients [11,13].
Several factors underlie the pathogenesis of CVD in RA patients, including traditional risk factors (smoking, BMI (body mass index), degree of physical activity, hypertension, hyperlipidemia, and diabetes mellitus) and risk factors related to RA, such as chronic inflammation, components of the adaptive immune system, post-translational modification events, autoantibodies, pro-inflammatory cytokine production, chemokine gene expression, oxidative stress, and adverse effect of some medications prescribed for RA management [14][15][16].The inflammaging (an age-related inflammatory condition), which exists in early pre-clinical RA, amplifies the pathophysiological events that lead to CVD in these patients and may further accelerate the early onset of CVD in the RA population [14,17].
Cytokines and chemokines exert an indispensable role in the pathogenesis of RA.CXC chemokines [18][19][20] contribute to both RA pathogenesis and variable aspect of cardiovascular diseases, including atherosclerosis, hypertension, and myocardial infarction [21,22].Using deep-learning methods on blood immune biomarkers (in the Stanford 1000 Immunomes Project), recently Nazish Sayed and colleagues introduced CXCL9 (Chemokine (C-X-C motif) ligand 9) chemokine as the most pivotal contributor to agerelated chronic inflammation (iAge), cardiac senescence, and inferior vascular performance [17].CXCL9 is produced following the stimulation of endothelial cells by IFN-γ (Interferon-gamma), and it promotes the atherosclerotic reaction by recruiting T cells through the CXCR3 (C-X-C Motif Chemokine Receptor 3) receptor, which is expressed by endothelial cells [22].The CXCL9 and its receptor CXCR3, as indicators of CVD pathology, increase in hypertension, myocardial infarction (MI), and left ventricular dysfunction (LVD) [23].In addition, it has recently been elucidated that CXCL9 is produced by the senescent endothelium, which regulates vascular function and cellular aging through CXCR3 [17,24].
Considering the presence of inflammaging and high prevalence of cardiovascular disorders in RA, and the established role of CXCL9 and its receptor CXCR3 in age-related chronic inflammation (iAge), cardiac senescence, poor vascular function, atherosclerosis, LVD, and MI, in this study, for the first time, we evaluated the association of plasma levels of CXCL9 and CXCR3 gene expression in the peripheral blood of newly-diagnosed and on-treatment RA patients with the traditional CVD risk factor (age, gender, BMI, smoking, blood pressure, lipid profile, and diabetes).

Study Population
The current study included 60 patients with RA, which were divided into two categories: 30 newly diagnosed (patients who did not receive any medication and with disease onset < 24 weeks) and 30 on-treatment patients with disease duration of 1 to 12 years, referring to Imam Reza Hospital, Kermanshah University of Medical Sciences (KUMS), between April 2022 and October 2022, and as well as 30 healthy subjects after matching for age and sex.The members participating in the study with any previous history of systemic rheumatic diseases other than RA, chronic diseases (cardiovascular, pulmonary, kidney, liver, etc.), and autoimmune, inflammatory, metabolic, and infectious diseases as well as pregnant women were excluded from our study and also patients or controls being treated with anti-lipid drugs, including statins, were excluded from the present investigation.The diagnosis of RA was performed by expert rheumatologists based on the classification criteria of the American College of rheumatology/The European Alliance of Associations for Rheumatology 2010 (ACR/EULAR 2010).Data regarding age, sex, weight, and relevant medical history were collected into a predesigned form, and the entire participant signed informed consent.This study followed the Declaration of Helsinki and was conducted with approval from the Ethics Committee of KUMS (IR.KUMS.MED.REC.1401.017).Demographic information and Disease modified antirheumatoid drugs (DMARD) dosage in study groups were shown in Table 1.

Measurement of the Plasma Levels of CXCL9
We performed an enzyme-linked immunosorbent assay (ELISA) to detect the plasma levels of CXCL9 (Zell-Bio GmbH, Germany, Cat.NO: ZB-10049C-H9648) based on the protocols of the ELISA kit.

Quantitative Polymerase Chain Reaction (qPCR)
Total RNA was extracted from peripheral blood samples with an RNX PLUS kit (SinaClon, Tehran Iran) according to the manufacturer's recommendations.RNA concentration were evaluated using NanoDrop 2000 UV-Vis Spectrophotometer (Thermo Scientific, USA).Reverse-transcribed complementary DNA was synthesized using the cDNA synthesis kit (PARSTOUS, Iran).
Primers were designed by online websites (UCSC, Oligocalc, and Oligoanalyzer) and were as follows: CXCR3 forward 5 / -TCC ACC TAG CTG TAG CAG ACAC-3 / and reverse 5 / -TCC TGC GTA GAA GTT GAT GTTG-3 / ; GAPDH (as a housekeeping gene): forward 5 / GAA ACC TGC CAA GTA TGA TG-3 / and reverse 5 / -AGG AAA TGA GCT TGA CAA AG-3 / .Real-time PCR analysis was accomplished using the Light cycler 96 device (Roche Applied Science, Penzberg, Germany) and using SYBR green as a detector color.The total volume of each real-time PCR reaction was 15 μl and included the following materials: 7.5 μl of the master mix of SYBR Green (Ampliqon), 1 μl of cDNA, 0.5 μl of each of the forward and reverse primers, and 5.5 μl sterilized distilled water.The temperature, time, and the number of cycles of each real-time PCR reaction included the following: 1. a pre-incubation cycle with a temperature of 95 °C for 30 s (s), 2. forty cycles of 2-step amplifcation including 95 °C for 5 s and 60 °C for 30 s; 3. a melting cycle of 95 °C for 5 s, 60° C for 60 s, and 95 °C for 1 s; 4. a cooling cycle with a temperature of 50 °C for 30 s.The relative gene expression for each sample was calculated by the Pfaffl method (Ratio = (E target ) ΔCt target (control−sample) / (E Ref ) ΔCt Ref(control−sample) ) [25].

Measurement of Fasting Blood Sugar (FBS) and Lipid Profile
Six milliliters of peripheral blood samples were collected in ethylene diamine tetra-acetate (EDTA) tubes after 12 h of fasting for measurement of Fasting blood sugar (FBS), and lipid profile.FBS and lipid profile (including total cholesterol, high-density lipid (HDL), low-density lipid (LDL) cholesterol, and triglyceride) were measured by glucose oxidase-peroxidase method (Biosystems, Barcelona, Spain) and via enzymatic reactions using commercial kits according to manufacturer's instruction (Biosystem, Barcelona Spain), respectively, and results were read using fully automated 7020 chemistry analyzer (Hitachi, Tokyo, Japan).

Calculation of Systematic Coronary Risk Evaluation (SCORE) and Framingham Risk Score (FRS) Assessment
Using SCORE and FRS algorithms, which incorporate plasma lipid levels, blood pressure, smoking, age, and sex, we calculated CVD risks in our study population.The SCORE algorithm was created and approved in 2003 from 12 European cohorts to evaluate the 10-year risk of CVD mortality, which includes fatal myocardial infarction.The FRS was derived and internally ratified in the USA cohort to estimate the 10-year risk of CVD and includes cerebrovascular and peripheral vascular disease events and heart failure [27][28][29][30].According to the EULAR recommendations, in the current study, CVD risk estimation performed using a 1.5 multiplication factor to the risk estimate by the SCORE and FRS calculator for RA patients [13].

Statistical Analysis
Statistical analysis was conducted with SPSS software version 24.0 (SPSS, Chicago, IL, USA) and drawing of the graph was carried out by the software GraphPad Prisms ® 6.0(GraphPad Software, La Jolla, California, USA).The distribution of normality of the groups was determined by 1-sample Kolmogorov-Smirnov (K-S) test and also one-way ANOVA test was used to compare between three groups.The correlation between two variables was determined using the Spearman and Pearson correlation.P value indicated as statistically significant at the level of < 0.05.Results were presented as mean ± standard deviation (SD).

The Serum Levels of FBS, Lipids (LDL, HDL, TG, and Cholesterol) and CXCL9
The mean serum concentration of FBS, LDL, HDL, TG, cholesterol, and CXCL9 in three groups are given in Table 2.

The Comparison of FBS, Lipid Profile and CXCL9 in Patients (New Case + On-treatment) and Control Group
The BPS (systolic blood pressure), BPD (diastolic blood pressure) and serum concentration of FBS, LDL, cholesterol, and TG were not significantly different between the new cases and on-treatment of RA patients and healthy subjects (P = .593,P = .191,P = .093,P = .105,P = .427,P = .241,respectively).While the mean serum levels of HDL and CXCL9 were substantially different between the three groups (P < 0.001 and P < 0.001).Also, FRS was different between the three groups, but, to an interesting, SCORE didn't have any difference between the three groups of the study population (P = 0.029, P = 0.078) (Fig. 1).

The Gene Expression of CXCR3 in the Study Groups
The gene expression of CXCR3 was elevated significantly in new-case and on-treatment of RA population compared to healthy subjects (p < 0.001, p < 0.01) (Fig. 1d).

DISCUSSION
The inflammatory process involved in the pathogenesis of CVD, the most prevalent extra-articular manifestation of RA, remains unidentified.In the current investigation, for the first time, we evaluated the correlation between the concentration of plasma CXCL9 and its receptor, CXCR3, gene expression in peripheral blood leukocytes with conventional CVD risk factors in newly diagnosed and on-treatment RA patients [11,13].
Compared to healthy subjects, the plasma levels of CXCL9 were significantly higher in newly diagnosed and on-treatment RA patients in our study (P < 0.0001 and P < 0.001, respectively).Also, the plasma concentration of CXCL9 was remarkably lower in on-treatment patients compared with the newly diagnosed RA patients (P < 0.001).Our results were parallel to previous investigations, which showed elevated levels of CXCL9 in RA patients [31,32].
In contrast to previous studies, which could not find a significant positive correlation between the plasma levels of CXCL9 and RA clinical Parameters, including disease activity score-28 (DAS-28), we found a remarkable association between CXCL9 and DAS-28 (P = 0.0005, r = 0.436) [31,32].In RA, the disease continuum has four main stages, including the preclinical stage, transition step, early RA, and established RA [33].The elevated levels of CXCL9 in newly diagnosed and on-treatment patients and the remarkable association of CXCL9 with DAS-28 show that CXCL9 mediates the RA pathogenesis  in early inflammatory events and established RA stage through the mechanisms has not been fully elucidated yet [31,34].
CXCL9 exerts its inflammatory role by binding to its receptor CXCR3, which also serves as the receptor for CXCL10 and CXCL11 [35].Using Real-Time PCR, we evaluated the gene expression of CXCR3 in the peripheral blood of both RA and control groups.The gene expression of CXCR3 was significantly higher in the newly diagnosed and on-treatment RA patients compared to the control group (P < 0.001 and P < 0.01, respectively), but we could not find meaningful differences between the two groups of RA patients.CXCR3 is expressed on the Th1 cell surface and has a crucial role in Th1 cell attraction to RA synovium through its interaction with CXCL9 and CXCL10, the chemokines produced by fibroblastlike synoviocytes (FLS) and leukocytes in synovial tissue [36].The assessment of the correlation between CXCR3 gene expression and RA severity showed a remarkable association between CXCR3 expression by peripheral leukocytes of RA patients and DAS-28 (P = 0.0002, r = 0.463).This finding further clarified the importance of CXCR3 gene expression as a potential indicator of disease activity in RA patients.
In the following, we evaluate the association of well-established CVD risk prediction algorithms, including Framingham Risk Score (FRS) and Systematic Coronary Risk Evaluation (SCORE), as well as conventional CVD risk factors, including hypertension, lipid profile, diabetes, age, smoking and sex, with CXCL9 plasma levels and CXCR3 gene expression in peripheral blood leukocytes.
The mean estimated score with Framingham risk score in the newly diagnosed, on-treatment, and control group was 9.96 ± 11.02, 7.20 ± 6.16, and 4.71 ± 6.07 respectively, with statistically significant difference between newly diagnosed and on-treatment RA patients with the control group (P = 0.014 and P = 0.035, respectively).In our study, the mean estimated risk score with the SCORE calculator in the newly diagnosed, on-treatment, and control groups were 11.53 ± 12.9, 8.83 ± 6.77, and 6.13 ± 6.60, respectively, which was not statistically significant among patient groups and control.Our finding was following Wagan et al., which showed FRS significantly differs between RA and the control group [37].
Although most CVD risk evaluation algorithms predict the risk of atherosclerosis, FRS includes broader CVD outcomes, including peripheral vascular disease and cerebrovascular events, compared with the SCORE [11].The higher FRS in RA patients in our study may reflect a higher risk of peripheral vascular disease in RA.In the following regarding the role of CXCL9 and CXCR3 in the pathogenesis of CVD and poor vascular performance, we evaluated the correlation between plasma level of CXCL9 and CXCR3 gene expression with FRS and SCORE [17].In contrast to CXCL9, CXCR3 gene expression showed a significant association with FRS and SCORE (P = 0.006; r = 0.349; P = 0.007; r = 0.341, respectively).The CXCR3 receptor gene expression has an association with the variable aspect of CVD in the normal population [22,38].To the best of our knowledge, our study showed the association of CXCR3 with the CVD risk algorithm, including FRS and SCORE, in RA patients for the first time.The elevated gene expression of CXCR3 by peripheral blood leukocytes in RA patients may make them more sensitive to increased plasma levels of CXCL9, CXCL10, and CXCL11, which may promote their migration to the vascular inflammatory site and accelerate the atherosclerotic reaction and other CVD complications [39].
In the following, we evaluate the correlation between the plasma level of CXCL9 and CXCR3 gene expression with traditional CVD risk factors, including hypertension, lipid profile, diabetes, and age.In our  study, the plasma HDL was significantly increased in on-treatment patients.In inflammatory settings such as RA, HDL has pro-inflammatory and pro-thrombotic properties that can enhance atherogenesis and increase CV risk [40,41].Increased HDL levels in our ontreatment patients can be attributed to DMARDs, including methotrexate treatment which has been documented in previous studies [42].The plasma HDL and LDL levels showed a significant negative correlation with plasma levels of CXCL9 in our RA patients (P = 0.033, r = -0.275;P = 0.021, r = -0.296respectively), which may indicate the possible effect of CXCL9 in dyslipidemia in RA inflammatory setting.The wellestablished effect of cytokines and chemokines in lipid metabolism in RA and other autoimmune diseases further supports our finding [42,43].
In the following, we found a significant correlation between CXCR3 gene expression and systolic blood pressure in RA patients (P = 0.006).The association of CXCR3 and its ligands, including CXCL9, CXCL10, and CXCL11, with hypertension and left ventricular dysfunction (LVD) has been documented in the previous investigation [23].It is worth mentioning that LVD is a prevalent condition in RA patients [44].In contrast to a recent study conducted with Sayed et al., we could not find a significant association between plasma CXCL9 and aging in our RA patients and healthy subjects, but there was a meaningful association between CXCR3 gene expression and age (P = 0.020, r = 0.298) in our RA population.Rheumatoid arthritis accompanies various aspects of immunosenescence and aging, and it can be hypothesized that CXCR3 may involve in the inflammatory events related to aging in RA patients.
In our study, we should consider the possible impact of treatment on our results.
All newly diagnosed patients in our study did not recently receive any therapy for rheumatoid arthritis.Regarding the last EULAR recommendation updated in 2022, the treatment of RA should begin with DMARDs, and methotrexate (MTX) should be part of the first treatment strategy [45].NSAIDS and COX2i may be used by our patients for the management of their pain before the diagnosis of RA.There is not ample evidence that confirms the direct effect of NSAIDs and COX2i on the CXCL9/CXCR3 axis in inflammatory settings such as RA, but we cannot exclude the possible impact of NSAIDs and COX2i on CXCL9/CXCR3 axis, which we consider it as our study limitation.Also, there are not enough studies which show the association of MTX and HCQ therapy on the CXCL9/CXCR3 axis in RA patients.previous data showed that HCQ treatment in patients with osteoarthritis does not affect the serum levels of CXCL9 [46].Furthermore, the study conducted by Kotrych et al. didn't show any correlation between CXCL9 gene polymorphism and response to MTX therapy.In contrast, the recent investigation by Brynedal et al. reported the inhibition of CXCL9 production in patients who responded to MTX therapy in RA patients.Further study is warranted to determine the effect of MTX and HCQ on the CXCL9/ CXCR3 axis [47,48].
In RA, inflammatory markers, such as CRP, ESR, and disease severity, significantly increase the risk of cardiovascular disorders, consequently, the exacerbation of the inflammatory state in RA may magnitude the risk of CVD development.Furthermore, Pro-inflammatory mediators, released in the systemic circulation, can directly impose damage to the vessel walls' endothelium and leads to the aggravation of this condition in RA patients [49].
In RA patients, lowering the disease activity with effective DMARD therapy can reduce the risk of cardiovascular disease.But it is worth noting that some drugs can also have harmful effects on the vasculature and increase the risk of cardiovascular disease, possibly by disrupting the mechanisms related to vascular repair.Therefore, the treatments can act as a double-edged sword in the management of cardiovascular disease in RA because they can have opposite effects in restoring the balance between endothelial damage and vascular repair.Furthermore, treatment with some drugs may paradoxically affect traditional cardiovascular risk factors, such as blood pressure and lipid levels.For example, NSAIDs increase the risk of cardiovascular events and stroke, possibly through increased blood pressure, coagulation, and oxidative stress.In addition, glucocorticoids, due to their steroid nature, have harmful effects on lipid metabolism and glucose homeostasis and enhance oxidative molecule species.On the other hand, Methotrexate probably reduces the risk of cardiovascular disease and death in patients with RA by dampening inflammation and cytokine production.Hydroxychloroquine is also associated with the normalization of lipid and glucose profiles with antithrombotic effects and protection against the development of atherosclerosis [16,49,50].

CONCLUSION
In conclusion, our study showed the association between plasma CXCL9 and CXCR3 gene expression with RA disease activity, lipid profile changes, and blood pressure in RA patients.

Fig. 1
Fig.1Comparing the plasma levels of CXCL9 and HDL, gene expression of CXCR3, and FRS among three groups.

Fig. 2
Fig. 2 Association between plasma levels of CXCL9 with the DAS-28, HDL, and LDL.

Fig. 3
Fig. 3 Association between gene expression of CXCR3 with different variables.
P = 0.647 CXCL9 and its Receptor CXCR3, an Important Link Between

Table 1
The Demographic Information, DMARD Dosage, Clinical, Inflammatory and Serological Markers in all Studied Groups

Table 2
The Mean BMI, BP and Plasma Levels of FBS, TG, LDL, HDL, Cholesterol and CXCL9 Data are Mean ± SEM BMI Body Mass Index, BPS Systolic Blood Pressure, BPD Diastolic Blood Pressure, FBS Fasting Blood Sugar, HDL High Density Lipoprotein, LDL Low Density Lipoprotein, TG triglyceride, CXCL9 Chemokine (C-X-C motif) ligand 9

Table 5
Analysis of CXCL9 and CXCR3 in the Newly Diagnosed Patients, On-treatment Patients, and Control Group, According to the FRS and SCORE Risk Stratification Into Low, Moderate, and High-risk Groups for FRS, and Low-moderate, High, and Very High-risk Groups for SCORE