Synovial membrane complement gene expression after anterior cruciate ligament transection

Abstract

Purpose: Complement activation has been described as a pathomechanism of osteoarthritis (OA). It features an abnormally high expression of complement activator genes (i.e. complement C1q, C4 and C2 on the classical pathway of activation and complement factor B and D on the alternative pathway; encoded by C1QA, C1QB, C1QC, C4A, C2, CFB and CFD) in synovial membranes of affected patients. Both pathways lead to complement C3 (C3) activation and deposition of membrane attack complex (MAC, consisting of complement components C5b-C9, encoded by C5, C6, C7, C8A, C8B, C8G and C9) on chondrocytes, affecting their gene expression (i.e. expression of matrix metalloproteinases). Complement mediated OA was observed in mouse models of meniscectomy. However, the role of complement activation after ACL injury, a promising model for posttraumatic OA, is unclear. The aim of this study was to investigate if ACL injury models of OA feature complement activation and therefore are suited to test complement-targeting disease modifying interventions. Specifically, we assessed the changes in gene expression of complement effectors and inhibitors, which are elicited in the synovial membranes early after ACL transection in a porcine model. Methods: 24 adolescent Yucatan minipigs received unilateral ACL transection after IACUC approval. Synovial tissue was collected after 1, 5, 9 and 14 days (each n = 6). Whole transcriptome sequencing was used to quantify mRNA expression in the synovium. In order to control for injury-related changes in gene expression, healthy control tissue was obtained from 6 additional untreated animals (n = 12). Individual cDNA libraries were constructed with Illumina TruSeq Kit, multiplexed (8 per lane) and sequenced on multiple lanes of an Illumina HiSeq 2000. Raw reads were mapped to the pig genome (Susscr3) and differential gene expression was calculated with the edgeR subroutine package. Results: mRNA expression of most complement activators, MAC components and their inhibitors was significantly upregulated 1 day after ACL transection. However, they returned to control levels within 14 days. Specifically, the expression of all assessed classical pathway activators, as well as the classical pathway inhibitors Plasma protease C1 inhibitor (SERPING1) and C4b-binding protein (C4BPA) was significantly increased at day 1 (see table). The increased expression of almost all classical pathway activators and inhibitors returned to control levels within 5 days post-transection. Expression of the alternative pathway activator CFD was significantly elevated at all observed time points (11.1, 7.6, 5.3 and 22.2-fold for day 1, 5, 9 and 14, p = 5.05E-18, 6.74E-11, 9.10E-09 and 1.52E-18, respectively), while the expression of the inhibitor complement factor H (encoded by CFH) was increased 1, 5 and 9 days post-transection, before returning to control levels at day 14 (15.9, 3.7, 5.0 and 1.6-fold, p = 3.12E-20, 3.98E-05, 1.07E-06 and .266, respectively). In contrast, the more upstream alternative pathway activator CFB was significantly downregulated at day 9, before returning to control levels at day 14 (0.2 and 1.6-fold, p = 1.86E-07 and .296, respectively). Expression levels of C3 and all MAC components were significantly elevated at day 1 (p <1E-10, see table), whereas the change in CD59 mRNA levels was insignificant. While the abundance of transcripts encoding MAC components and CD59 drop to insignificant values thereafter, expression of C3 remained significantly greater than control at days 5 and 9 (3.5 and 3.1-fold, p = 8.05E-22 and 2.15E-05, respectively). Conclusions: Most complement activators, MAC components and inhibitors were significantly upregulated in the synovium after ACL transection. Although further characterization of the complement activation in human knee joints after ACL injury is needed, the porcine model may be a relevant preclinical model for the testing of therapeutic candidates for blocking the complement response of the joint after injury.Tabled 1Synovial gene expression at day 1 after transection (as fold-change compared to healthy controls)Complement activatorsClassical pathway: C1QA (8.9-fold, p = 8.50E-15), C1QB (6.8, p = 2.85E-11), C1QC, (8.2, p = 6.78E-14), C4A (11.6, p = 1.25E-16) and C2 (6.3, p = 5.18E-11); Alternative pathway: CFD (11.1-fold, p = 5.05E-18), CFB (0.6, p = 0.185)C3 and MAC componentsC3 (16.9-fold, p = 8.05E-22), C5 (2.4, p = 0.010), C6 (30.0, p = 0.019), C7 (7.0, p = 4.21E-10), C8G (55.0, p = 3.16E-22), C8B (1.7, p = 0.881), C8A (30.6, p = 0.887), C9 (5.9, p = 4.55E-10)Complement inhibitorsClassical pathway: SERPING1 (4.4-fold, p = 2.29 E-07), C4BPA (6.3-fold, p = 2.26E-11); Alternative pathway: CFH (15.9, p = 3.12E-20); MAC inhibitor: CD59 (1.9-fold, p = 0.031) Open table in a new tab