Serum MicroRNA Signatures as Diagnostic Tools that Distinguish Rheumatoid and Psoriatic Arthritis

Inflammatory Arthritis (IA) constitutes a diverse range of autoimmune conditions sharing characteristic features including soft tissue swelling, joint inflammation and subsequent bone and cartilage destruction of a debilitating nature. Unfortunately, the prevalence of IA in Ireland is 1-2%, of which Rheumatoid Arthritis (RA) and Psoriatic Arthritis (PsA) encompass two of the most common forms. While common pathogenic mechanisms are involved in driving inflammation in both arthropathies there are significant differences, including the presence/absence of autoantibodies, synovial vascularity, presence of skin psoriasis, immune cell infiltrates, molecular signalling, the pattern of periarticular inflammation, bone erosion and new bone formation at the entheseal complex of peripheral and spinal joints. These differences may explain distinct clinical manifestations of the two diseases, and more importantly, explain different responses to specific therapies impacting on disease outcomes and prognoses. While there are common treatments that are utilised as first line treatment for RA and PsA, including methotrexate and tumor necrosis factor inhibitors (TNFi), their response to other biologic therapies differs, with Rituximab and anti-IL-6R only utilised for treatment of RA, while anti-IL-17A and anti- IL-12/IL23 are utilised for PsA. Furthermore, the multifaceted nature of synovial inflammation, joint destruction, enthesitis, axial disease, and skin manifestations often makes diagnosis, prediction of disease progression and prediction of response difficult. However, currently there are few biomarkers that distinguish disease pathotype or response to therapy. Thus there is a need for the identification of noninvasive, convenient, and reliable biomarkers for the diagnosis and prediction of treatment response outcomes in autoimmune disease.

‘MicroRNA (miRNA)’ which are small endogenous regulatory RNA molecules have emerged as potential therapeutic targets and biomarkers in autoimmunity. MiRNA bind to complementary sequences on messenger RNA (mRNA), and they function generally by suppressing the translation of target proteins (Figure 1A). In the context of Inflammatory Arthritis, their effects generally down regulate proteins involved in keeping the immune response in-check, thus once unchecked this allows for an uncontrolled immune response which leads to cartilage study hasdemonstratedthe potentialfor serummiRNAs to beinstrumentalin a diagnosticcapacity in aclinical setting”degradation and bone erosions. Over the last 10 years numerous studies have demonstrated that altered miRNA expression in Inflammatory Arthritis influences immune cell regulation, enhances pro-inflammatory signaling pathways, and leads to the overproduction of proinflammatory proteins that are critical to driving the pathogenic mechanisms involved in joint destruction and subsequent joint disability. However, more recent studies have identified that, in addition to their localization within the cell, miRNAs are also present in extracellular fluids such as serum, plasma, and synovial fluid, and through the circulation can be transported to specific cell types or tissues to carry out their function. Interestingly, it is now known that circulatory miRNAs are more stable than cellular miRNAs and are emerging as potential non-invasive biomarkers for disease.

Therefore, in this study we examined the expression profile of miRNAs, specifically focusing on a defined immunology miRNA panel, to identify a potential circulatory miRNA signature that could distinguish RA from PsA, in addition to evaluating the potential implication for disease pathogenesis. To do this we utilised the clinically amenable FirePlex Biofluids miRNA assay. While previous approaches employed to detect circulating miRNA relied on laborious isolation techniques and required high sample volume to obtain sufficient miRNA and thus limited the potential use in the clinical setting, the FirePlex platform allows for miRNA detection in very small volumes of blood (~20µl) that can be measured by automated flow cytometric analysis, thus allows for robust, reproducible results that can easily be assessed in clinical labs on hospital sites.

We obtained serum samples from patients with active RA and PsA, in addition to healthy controls (HC) and performed an immunological miRNA analysis using the FirePlex miRNA Immunology-V2 panel (FirePlex Bioworks Inc.). We identified a miRNA signature of seven miRNA (miR-126-3p, miR-29b-3p, miR-22-3p, miR-223-3p, miR- 320a, let-7g-5e, and let-7g-5p) that were significantly elevated in RA serum compared to both PsA patients and HC. Further computational bioinformatic analysis demonstrated that the miRNA expression distribution was associated with a dominant skew towards 3 specific miRNA in RA vs PsA: miR-29b-3p, miR-22- 3p and miR-223-3p. In contrast, we identified three miRNA that were significantly elevated in PsA compared to RA which included miR-203a-3p, miR-185-5p, and miR-151a-5p. To examine sensitivity/specificity of miRNA to either RA or PsA we performed Receiver Operating Characteristic (ROC) curves, that demonstrated high sensitivity and specificity of each miRNA signature to either that of RA or PsA.

As highlighted above, miRNA regulate their function by targeting different proteins. Therefore, we next utilised computational analysis (DIANA miRPATH tool) to identify the specific genes and inflammatory pathways targeted by the identified miRNA signatures, followed by STRING software analysis which allowed the identification of target gene interactions. Functional pathway analysis of the RA miRNA signature (miR- 29b-3p, miR-22-3p and miR- 223-3p) showed significant association with pathways that regulate key pathogenic mechanisms involved in synovial inflammation. Specifically, the miRNA signature was involved in mediating a diverse number of cellular processes including metabolism, cell growth, invasion, and proliferation, all of which contribute to the uncontrolled proliferation and activation of stromal cells within the inflamed joint, particularly for RA which is known to have a more aggressive synovial invasive layer compared to that of PsA. While few studies have examined these miRNA in RA, studies have shown that increased levels have been associated with RA disease activity and development of RA in susceptible individuals demonstrating positivity for ACPA antibodies. Furthermore, the RA miRNA have also been associated with regulation of matrix metalloproteinases, invasive signalling pathways, and chondrocyte dysfunction, key mechanisms associated with cartilage breakdown. Interestingly, the miRNA signature associated with PsA regulate angiogenic function and specific signalling pathways associated with blood vessel migration, leukocyte adhesion and trans-endothelial cell migration, a feature more pronounced in PsA. Indeed, many of the pathways associated with the PsA miRNA are also associated with cancer, and specifically involved in driving angiogenic processes. In particular, studies have shown that miR-151a-5p is associated with the Notch signalling pathway, which we know is increased in the PsA joint and in psoriasis skin lesions and is critically involved in the regulation of the pathogenic dysfunctional blood vessel pattern observed in PsA. Furthermore, miR-185-5p is associated with high levels of the VEGF receptor, which activates the Notch signalling pathway in endothelial cells that line the blood vessels. The PsA miRNA were also associated with the Hippo signalling pathway which again target genes associated with angiogenic growth factor regulation and cellular adhesion. Interestingly, the distinct synovial blood vessel pattern observed in the PsA joint (elongated, tortuous and dilated) is similar to the dysfunctional blood vessel patterns observed in malignant tumors. This is in stark contrast to RA where the blood vessel pattern is defined by straight, regular branching synovial vessels. The implications for these different blood vessel patterns are still unclear but must be involved in differential recruitment of immune cells to the joint, in addition to differential levels of nutrients entering the synovial cavity.

In summary, this study has demonstrated the potential for serum miRNAs to be instrumental in a diagnostic capacity in a clinical setting. The benefit of a serum miRNA signature as a biomarker utilising this method is that it is a minimally invasive approach requiring a minimal amount of blood that would be sufficient for diagnosis using the FirePlex assay. This assay has clinical potential as it is a highly reproducible test which requires only minimal serum that can simultaneously detect a panel of selected miRNAs, without the need for laborious RNA purification and amplification steps which can lead to lack of reproducibility in the clinical setting. This platform would also be an efficient, convenient, and cost-effective approach. The serum miRNA signatures identified in this study discriminated Rheumatoid Arthritis (RA) and Psoriatic Arthritis (PsA), and importantly from that of healthy subjects (Figure 1B). However, further investigations are required in larger multicentre patient cohorts for validation. If they prove to be consistent, it would be interesting to examine in future approaches if the circulatory miRNA signatures could be useful not only in the monitoring of disease progression but also as pharmacodynamic biomarkers of treatment response. In conclusion these findings are valuable, and the identification of reliable biomarkers that may facilitate pathotype diagnosis would facilitate interventions with more targeted therapeutic approaches, thus alleviating symptoms, and potentially inducing earlier disease remission whilst ultimately enhancing patient quality of life.

Acknowledgements: These studies would be impossible to perform without patient involvement, therefore we would like to thank all patients who contributed to this study.

Read the full magazine: hospitalprofessionalnews.ie/subscribe