ABUNDANT PLACENTA GROWTH FACTOR (PLGF) EXPRESSION IN THE INFLAMMATORY JOINT

Background: Placenta growth factor (PlGF) binds only to the vascular endothelial growth factor receptor-I (VEGF-RI/Flt-1) and not VEGF-RII. Inhibition of Flt-1 has been shown to be protective against joint destruction by reducing synovial inflammation and neovasculatization in mouse models of arthritis. Two of the four PlGF isoforms (PlGF-2 and -4) have heparin binding sites resulting in their being sequestered to the extra-cellular matrix; the other isoforms (PlGF-1 and -3) are soluble. When measuring VEGF and PlGF, their bioavailability must be taken into account, due to the presence of their natural antagonist soluble Flt-1 (sFlt-1), a Flt-1 splice variant.Objectives: To determine (1) serum and synovial fluid (SF) levels of PlGF, VEGF and sFlt-1, (2) synovial tissue expression of the PlGF and Flt-1 isoforms, and (3) establish their cellular source.Methods: PlGF and sFlt-1 levels were measured by ELISA in serum and matching SF samples obtained from patients with rheumatoid arthritis (RA) n=18, spondyloarthropathy (SpA) n=20 [comprising of psoriatic arthritis (PsA), seronegative arthritis and ankylosing spondylitis] and osteoarthritis (OA) n=11. Synovial biopsy was obtained at knee arthroscopy. RNA was extracted from synovial tissue (n=14), synoviocytes (PsA n=2) and peripheral blood neutrophils (PsA n=2) and monocytes (PsA n=1). PlGF, Flt-1 and sFlt-1 mRNA were measured by RT-PCR. Mann-Whitney U statistical analysis was performed.Results: PlGF was significantly higher in SF when compared to serum (p=0.001) [mean pg/ml ± s.e.m., RA 3195±568 vs 81±45, SpA 2856±309 vs 11±7 and OA 516±108 vs 95±74]. VEGF was higher in SF when compared to serum, but only significant in the OA group (p=0.0001) [mean pg/ml ± s.e.m., RA 170±36 vs 78±16, SpA 79±24 vs 47±10 and OA 102±15 vs 26±8]. sFlt-1 was higher in SF when compared to serum, this was significant in the RA and SpA groups (p=0.04) [mean pg/ml ± s.e.m., RA 3276±1460 vs 94±7, SpA 596±186 vs 92±8 and OA 122±55 vs 95±14]. SF PlGF, VEGF and sFlt-1 was higher in the RA and SpA groups when compared to OA (p=0.001, p=0.03 and p=0.03 respectively). PlGF concentration was considerably higher than VEGF in SF (p=0.0001). VEGF was present in 78% and PlGF in 39% of total serum samples.Both PlGF (343.8 pg/mg) and sFlt-1 (164.9 pg/mg) were spontaneously released from synovial biopsies, suggesting that synovial tissue is a source of the PlGF and sFlt-1 found in SF. PlGF-1, -2 and -4 mRNA was detected in synovial tissue, synoviocytes and neutrophils. Monocytes express mRNA for PlGF-1, -2 and -3. The soluble isoform PlGF-1 was the most abundant isoform in the synovial tissue and synoviocytes. Both Flt-1 and sFlt-1 mRNA was detected in synovial tissue, neutrophils and monocytes.Conclusion: The higher frequency of detection of VEGF in serum is in keeping with its role in vascular homeostasis, whilst PlGF is detected less frequently consistent with its pathological function. High levels of PlGF in comparison to VEGF in the arthritic joint suggests it may play a significant role in the promotion of monocyte migration and angiogenesis in the arthritic joint. Therefore, Flt-1 and PlGF may represent novel therapeutic targets allowing down-regulation of both angiogenesis and inflammatory cell migration in human inflammatory arthritis.Citation: Ann Rheum Dis, volume 66, supplement II, year 2007, page 141Session: Cytokines and inflammatory mediators