A novel Link of Serum IL-39 Levels in Patients with Rheumatoid Arthritis

Rheumatoid arthritis (RA) is an inflammatory condition causing joint pain and stiffness, with often debilitating and life-limiting consequences. Recently, a new B-cell secreted cytokine, IL-39, was identified in mice. The most up-to-date research indicates that although IL-39 is expressed in murine models of lupus and has a role in mediating the inflammatory response in this context, there is no solid, replicated evidence of the existence of IL-39 in humans. This study aimed to clarify the existence and role of IL-39 in the human body and to elucidate whether it plays a role in rheumatoid arthritis. Accordingly, serum samples were collected from 66 patients with rheumatoid arthritis who were under therapy and from 66 healthy controls attending the Baghdad Teaching Hospital Rheumatology Unit. Enzyme-linked immunosorbent assay (ELISA) was used to detect serum levels of IL-39. Our results showed that mean ages of RA patients were 46.48 ± 10.17 years, and for healthy controls 44.97 ± 11.658 years. The results revealed that serum IL-39 levels were significantly lower in RA patients ( p = 0.016) (4.95 ± 1.1001) compared to healthy controls (5.55 ± 1.762). Corresponding sensitivity of IL-39 was 56.1%and specificity was 60.6% at cutoff values of ≤ 4.99 ng/ml. In Conclusion, IL-39 is found in humans but is downregulated in rheumatoid arthritis patients. This suggests either that IL-39 can have both pro-or anti-inflammatory functions based on the underlying disease or that the role of IL-39 is masked by the effects of treatment.


Introduction
Among the different types of inflammatory arthritis, a group of conditions characterised by an autoimmune-mediated process that leads to damage and swelling of the joints known as rheumatoid arthritis (RA), is the most common.The main symptom of RA is a progressive, life-limiting pain and stiffness of the joints, mainly affecting the hands, wrists, fingers, and feet, although other joints can also be affected [1,2].These symptoms can have significant negative consequences on the overall quality of life of the affected individuals.There are several risk factors for rheumatoid arthritis including gender, with females being more affected than males; ethnicity; lifestyle habits such as smoking; and genetic factors such as possession of the HLA-DRB1 gene [3].Overall, the pathophysiology of RA is mediated by autoimmune and inflammatory-related processes that result in an erosion of the bone and cartilage of the joints [4].With a worldwide prevalence of 2% and 1% in the United Kingdom, RA affects millions of people each year, causing a significant human and economic burden on society and public health systems all over the world [5].Based on this, a large amount of scientific research efforts has been dedicated to identify certain biomarkers of this disease, both for the purpose of diagnosis as well as prognosis prediction.Specifically, cytokines, small proteins involved in signalling in the immune system, and with proven functions in the pathophysiology of rheumatoid arthritis have received particular attention.Cytokines, which are produced by immune cells such as T-cells and B-cells, infiltrate the liquid surrounding and lubricate the joints, also known as synovial liquid, and mediate the erosion of bone and cartilage, which in turn causes inflammation and pain [6].Previous research has identified several cytokines with important functions in the pathophysiology of arthritis, including cytokines such as TNFα, IL-1, and IL-17 [7,8].Indeed, such research has identified that some of these cytokines have great potential as not only diagnostic and predictive tools but even as therapeutic targets.
More recently, IL-39 was identified as a new member of the IL-12 family of cytokines in mice, specifically in a mouse model of systemic lupus erythematosus, an autoimmune disease in which the person's immune system attacks several bodily tissues, including lungs, heart, kidneys, skin and other organs [9,10].IL-39 is secreted by B-cells and is composed of two subunits, IL-23p19 and Ebi3 [9].Furthermore, the researchers described that IL-39 is having a role in the mediation of inflammation in the mouse model of lupus and thus highlighted its potential role as a therapeutic target for this condition [9].However, since this initial study, little to no research has been carried out on IL-39.This is mainly due to the fact that more recent publications have indicated that IL-39 has a murine-specific expression and is thus not expressed in humans.Indeed, with the exception of a publication from 2017 reporting the presence of IL-39 in the serum of patients with acute coronary syndrome [11], no other publication since has found any evidence of IL-39 expression from human cells [12,13].The findings that IL-39 is involved in the pathophysiology of lupus in mice, which is an autoimmune and inflammatory condition, suggest it could have interesting functions in RA.However, research into any potential link between IL-39 expression and RA is hampered by the lack of clarity around its expression from human cells with literature yielding contradicting results.Thus, while recent research has identified several cytokines such as IL-40 and IL-17 as having potential as diagnostic and predictive tools for rheumatoid arthritis, the usefulness of IL-39 in this aspect remains to be elucidated.The present study focuses on confirming the presence of IL-39 in humans and aims to understand whether it is specifically expressed in the context of RA or not.

Patients and Controls
From November 2021 to January 2022, blood samples were collected from 66 RA patients (55 females, 11 male; age range 23-69 years) and 66 healthy controls (49 females, 17 males; age range 22-69 years) attending the Baghdad Teaching Hospital Rheumatology Unit.Exclusion criteria included RA with comorbidities with other connective tissue diseases, seronegative spondyloarthritis, malignancy, pregnancy, or patient refusal to participate.Inclusion criteria included patients with RA who had been diagnosed according to the 2010 ACR/EULAR classification criteria [14], any gender, age >18yrs, and an active disease state.
For each patient, an information sheet was completed under the supervision of the rheumatologist.The information included gender, age, disease duration, disease activity score-28 (DAS-28), and type of current therapy.The following laboratory tests were also included: rheumatoid factors (RFs), C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), haemoglobin (Hb), white blood cell (WBC) count, and anti-cyclic citrullinated peptide (ACCP) antibodies.The study objectives and procedures were provided to all participants and formal consent was obtained before the onset of the study.The study was approved by the scientific ethics committee of the College of Medicine, University of Baghdad.The immunological tests were performed in the Medical Research Unit at the College of Medicine, Al-Nahrain University.
RA patients were under therapy, with five different protocols followed for different treatments.Patients were either treated with: 1) methotrexate (MTX; a single oral weekly dose of 5 -15 mg); 2) etanercept (single weekly subcutaneous dose of 25 mg) or adalimumab (20 mg); 3) corticosteroids (oral methylprednisolone: single daily dose of 50 mg); or 4) disease-modifying anti-rheumatic drugs (DMARDs), which included leflunomide (daily oral dose of 5 mg), sulfasalazine (daily oral dose of 500 -1000 mg), azathioprine (daily oral dose of 50 mg), or hydroxyl chloroquine (daily oral dose of 200 -500 mg).In a further group of patients, the therapy included a combination of the four protocols (combined group).

Immune assays
The serum was extracted by centrifugation at 1000-3000 rpm for 10 minutes after blood samples were taken in gel tubes, following which it was frozen at -20°C.Antibodies to IL-39 were detected in human serum using an enzyme-linked immune sorbent assay (ELISA) (Sun Long Biotech Company, China), as directed by the manufacturer.A plate reader was used to measure the absorbance at 450 nm.

Statistical analysis
Statistical analysis was performed by using the SPSS Statistical package (Version 21; SPSS, IBM) and Microsoft Office Excel (2010) for drawing the figures except for the receiver operating characteristic (ROC) curve.
Independent samples of students (t-test) were performed for comparisons of quantitative variables between studied groups (Age / Year & serum IL -39 ng/ml level).Normally distributed data is expressed as (Mean±SD) with Pearson chi-square test ( χ 2 ) for comparisons of qualitative variables between studied groups (i.e., Age groups / Year, gender}& a binomial Z-test was performed for comparison of anti-CCP, RF, CRP assay, duration / Year, and smoking & treatment data.The validity of ELISA test was estimated with a ROC curve, cutoff value, area under curve (AUC), sensitivity (%), specificity (%), positive predictive value % (NPV), negative predictive value % (NPV), and accuracy.The statistical significance threshold (Pvalue) was accepted at P<0.05 & P<0.01.

Results and Discussion
Rheumatoid arthritis is a complex disease with inflammatory and autoimmune pathophysiology.Based on this, a potential avenue of research for therapeutic development involves the use of agents able to downregulate the immune response, and with reducing the progressive erosion of joints, and with it alleviate inflammation and pain.In this context, significant focus has been placed on cytokines, small proteins that are tightly linked to the inflammatory and autoimmune processes underlying RA, the usefulness of which for diagnostic, prognostic, and therapeutic purposes has already been described [7,15].

IL-39 serum levels
The mean study (see Table 4) confirmed significant differences (p = 0.016, p < 0.05) for IL -39 levels, which was decreased in the sera of RA patients (4.95 ± 1.1001 ng/ml) compared to healthy controls (5.55 ± 1.762 ng/ml).As shown by the ELISA results, the expression of IL-39 was detected in the serum of both healthy and RA patients.This is a significant result since the studies by both Bridgewood [12]and Ecoeur et al. [13] failed to detect IL-39 in human cells.This, however, raises the question as to what the source of such contradicting results is.Therefore, a closer look at the methodological protocols and samples should be carried out, taking a closer look at the sensitivity and specificity of the immunoassays used to detect IL-39.Notably, rather than displaying increased levels of IL-39, RA patients had significantly decreased levels of IL-39.Although further research would be required to determine the cause of this, such results hint toward IL-39 having an anti-inflammatory rather than inflammatory role in RA.This would be similar to the anti-inflammatory function that has been reported for IL-10 in a number of diseases, including some of inflammatory and/or autoimmune nature, such as inflammatory bowel disease (IBD), colitis-associated cancer (CAC) and systemic lupus erythematosus, for example [16,17].In the context of arthritis, reduced levels of IL-10 are associated with more severe disease, as exemplified by Il-10 KO murine models [18].If confirmed this would be in stark contrast with the results obtained by Wang et al. (2016) [13] which showed that IL-39 mediates the inflammatory process in lupus as well as those of Luo et al. (2017) [7] which also linked IL-39 with inflammation and cardiac disease.It is however possible that the baseline levels of IL-39 in RA patients were masked by the treatments that the patients were undergoing (e.g., steroids, biologics and DMARDs).Indeed, while the distribution of IL-39 levels across the different treatments was analysed, we did not specifically look at differences between the levels of IL-39 between rheumatoid arthritis patients with and without treatment.This, therefore, presents both a limitation as well as a future course of research.Other limitations of this study include the small number of patients, which limits the statistical power of the analysis, as well as the lack of functional analysis of IL-39.Further, a more indepth analysis of treatment-induced changes in IL-39 levels would have been necessary to determine the link between IL-39 and inflammation.Furthermore, Table 5 shows non-significant differences (p > 0.05) between the mean of level IL -39 in the sera of RA patients when considered for all parameters across the study.when selecting the patient sample.In this study, the distribution of patient demographics was non-significant, and the distribution of other key parameters among the RA population was highly significant including disease duration, smoking habit, CRP, steroid use, and biologics.A more equal distribution of patient-related parameters may allow for a more stringent analysis of the results.Finally, we looked at the validity of using IL-39 as a diagnostic tool for RA.Overall, the results showed poor performance of IL-39 as a diagnostic marker, with specificity and sensitivity of around 60% which is well under the threshold to be considered adequate for clinical use [20].

ROC curve analysis:
Valuation of serum IL -39 tests was performed.Low validity was observed on the basis of: cut-off value {≤ 4.99} ng/ml, area under curve (AUC) {0.622}, sensitivity (True positive %) {56.1 %}, specificity (True negative %) {60.6%}, positive predictive value (PPV) {58.7%}, negative predictive value (NPV) {58%}, and finally, accuracy {58.34%}, with significant difference (P=0.016,P<0.05) as shown in Figure 1.Although this finding does not rule out the potential role of IL-39 in the pathology of rheumatoid arthritis it does indicate that it might have limited usefulness in a diagnostic context.Given however the contradicting results present in the literature, and the small number of patients included in this study, there are enough indications to suggest repetition of this work.Further, the lack of a diagnostic utility does not exclude the possibility that, if proved to have anti-inflammatory properties like other cytokines, like IL-10, IL-39 could have potential as a therapeutic option for this condition.If on the other hand it was confirmed to have pro-inflammatory functions, its utility as a marker of disease activity and severity should be further investigated.

Conclusion
The results from this study provide another layer of evidence of the existence of IL-39 in humans.This together with the one previous study indicating the expression of IL-39 in cardiac patients, suggests that IL-39 may indeed be expressed in humans.However, given the contradicting evidence, further work is necessary to support these findings.Further, given the fact that IL-39 was shown to mediate inflammation in the context of lupus in mice and cardiac disease in humans, the present result showing it was actually downregulated in RA patients is potentially contradictory.This suggests either that IL-39 can have both pro-or antiinflammatory functions based on the underlying disease or that the role of IL-39 in RArelated inflammation is being masked by the effects of treatment.Future studies should aim to corroborate the existence of IL-39 in humans as well as the reduced levels of this cytokine in the context of RA.Following this, potential avenues for future work would include taking a closer look at the exact function of IL-39 in this patient group as well as understanding whether it has any prognostic value and/or therapeutic value.

ETHICAL CLEARANCE
This study was approved by The Scientific Ethics Committee of the College of Medicine, University of Baghdad.It was also approved by Iraq's Ministry of Health, and The Ministry of Education and Scientific Research.

Figure 1 :
Figure 1: Validity tests of IL -39 ng/ml by using ROC test

Table 1 :
Demographics distributions according to studied groups

Table 3 :
Distributions of RA patients' age groups/year according to their gender.

Table 4 :
Mean distributions of immunological assay IL -39 ng/ml among studied groups

Table 5 :
Mean distributions of immunological assay IL -39 ng/ml among RA patients.