Background:VEGF-A is a key regulator of rheumatoid (RA) and osteoarthritis (OA). During articular inflammation in OA and RA there is increased synovial angiogenesis and upregulation of angiogenic growth factors such as VEGF-A. VEGF-A comprises two splice variant families, VEGF-Axxxa and VEGF-Axxxb (xxx represents the number of amino acids, from 121 to 206), resulting from alternative splice site selection in exon 8. This site selection is controlled by Serine/Arginine Rich Splicing Factor Kinase 1 (SRPK1), which phophorylates Serine/Arginine Rich Splicing Factor 1 (SRSF1), inducing it to translocate to the nucleus. In most normal tissues, VEGF-Axxxb isoforms predominate, with anti-nociceptive and anti-angiogenic functions. I contrast, in pathological conditions such as inflammation and solid tumours, VEGF-Axxxa isoforms predominate, with pro-nociceptive and pro-angiogenic functions. VEGF-A has been proposed as a therapeutic target in RA and OA. To date, there are no published data on the functionally distinct VEGF-A splice variants in either RA or OA.Objectives:To determine the patterns of, and relationships between, VEGF-A, SRPK1, and SRSF1 expression and activation and synovial inflammation in human RA and OA.Methods:The study was approved by the Nottingham Research Ethics Committee 1 (05/Q2403/24) and Derby Research Ethics Committee 1 (11/H0405/2). Tissues were selected from age- and sex-matched cases in the University of Nottingham joint tissue repository. Post-mortem (PM) samples of healthy knee synovium (n=14, no history of arthritis or knee pain in the 12 months prior to death, no significant articular or synovial pathology) and arthroplasty-derived synovium samples from OA (n=35) or RA (n=14) patients were compared. OA samples were selected to represent the variety of inflammation levels, from low to high grade (0-3, Haywood et al., 2003). 8um thick sections were stained for SRSF1, SRPK1, total VEGF-A and VEGFxxxb via immunohistochemistry. Expression was estimated as fractional area, relative staining intensity (VEGF-Axxxb), and SRSF1 activation quantified by the degree of nuclear localisation. Statistical analyses were performed using Kruskal-Wallis followed by Dunn’s tests and Spearman’s rank correlations.Results:SRPK1 expression was similar across all conditions. SRSF1 showed significantly higher expression in the OA tissue compared to PM (H(2)= 11.29, p=0.002; OA median=0.2, IQR(0.15, 0.28); PM median=0.09, IQR(0.07, 0.16)), and significantly higher nuclear localisation (indicating activation) in RA vs. OA, and in both RA and OA vs PM (H(2)=37.65, p<0.0001 RA cf. PM; p=0,007 OA cf. PM; RA median=89, IQR(83, 93); OA median=36.1, IQR(29, 42); PM median=19.8, IQR(14,21)). Nuclear SRSF1 was significantly correlated with inflammation score (r= 0.52, p<0.05). Total VEGF-A expression was significantly increased in RA compared to PM and OA (H(2)=23.3, p<0.001 RA cf. PM; RA median=0.4, IQR(0.37,0.59); PM median=0.18, IQR(0.15,0.2)) and was also correlated with the severity of inflammation (r=0.47 p<0.05). VEGF-Axxxb showed no changed in expression in OA or RA, although VEGF-Axxxb staining intensity was significant higher in RA samples, compared to controls (H(2)=7.2 p=0.02; RA median=2.3(1, 4); PM median=0.9 (0.7, 1.4)).Conclusion:Increased levels of SRSF1 activation, and the association of total VEGF-A expression with inflammation score, support the hypothesis that there is activation of alternative splicing in inflamed synovium in RA and OA. Targeting this pathway could be a novel therapeutic strategy in OA & RA.
Read full abstract