Gremlin-1 and BMP-4 Overexpressed in Osteoarthritis Drive an Osteochondral-Remodeling Program in Osteoblasts and Hypertrophic Chondrocytes.

Osteoarthritis (OA) is a whole joint disease characterized by an important remodeling of the osteochondral junction. It includes cartilage mineralization due to chondrocyte hypertrophic differentiation and bone sclerosis. Here, we investigated whether gremlin-1 (Grem-1) and its BMP partners could be involved in the remodeling events of the osteochondral junction in OA. We found that Grem-1, BMP-2, and BMP-4 immunostaining was detected in chondrocytes from the deep layer of cartilage and in subchondral bone of knee OA patients, and was positively correlated with cartilage damage. ELISA assays showed that bone released more Grem-1 and BMP-4 than cartilage, which released more BMP-2. In vitro experiments evidenced that compression stimulated the expression and the release of Grem-1 and BMP-4 by osteoblasts. Grem-1 was also overexpressed during the prehypertrophic to hypertrophic differentiation of murine articular chondrocytes. Recombinant Grem-1 stimulated Mmp-3 and Mmp-13 expression in murine chondrocytes and osteoblasts, whereas recombinant BMP-4 stimulated the expression of genes associated with angiogenesis (Angptl4 and osteoclastogenesis (Rankl and Ccl2). In conclusion, Grem-1 and BMP-4, whose expression at the osteochondral junction increased with OA progression, may favor the pathological remodeling of the osteochondral junction by inducing a catabolic and tissue remodeling program in hypertrophic chondrocytes and osteoblasts.

Knee and hip intra-articular adipose tissues (IAATs) compared with autologous subcutaneous adipose tissue: a specific phenotype for a central player in osteoarthritis.

OBJECTIVES: Compared with subcutaneous adipose tissue (SCAT), infrapatellar fat pad (IFP), the main knee intra-articular adipose tissue (IAAT), has an inflammatory phenotype in patients with osteoarthritis (OA). We phenotyped suprapatellar fat pad (SPFP) and hip acetabular fat pad (AFP), two other IAATs, to determinate the unique signature of IAATs compared with SCAT. METHODS: IFP, SPFP, AFP and autologous SCAT were obtained from patients with OA during total knee (n=38) or hip replacement (n=5). Fibrosis and adipocyte area were analysed by histology and vascularisation, leucocyte and mast cell infiltration were analysed by immunohistochemistry for von Willebrand factor, leucocytes and tryptase, respectively. Secretion of interleukin (IL)-6, IL-8 and prostaglandin E2 (PGE2) was assessed by ELISA. The mRNA expression of adipocyte-associated genes (ATGL, LPL, PPAR-γ, FABP4 and CD36) and developmental genes (SFRP2, HoxC9 and EN1) was determined. The inflammatory response of isolated fibroblast-like synoviocytes (FLS) to autologous IFP and SPFP conditioned media was examined. RESULTS: Fibrosis, vascularisation and leucocyte and mast cell infiltration were greater in IAATs than SCAT, and levels of IL-6, IL-8 and PGE2 were greater in all IAATs than SCAT. IFP and SPFP induced a similar inflammatory response to FLS. Adipocyte area was smaller in IAATs than SCAT. Adipocyte-associated and developmental genes showed a similar gene expression pattern in all IAATs, different from SCAT. CONCLUSIONS: IFP but also SPFP and AFP (gathered under the term 'IAAT') may play a deleterious role in OA by affecting joint homeostasis because of their inflammatory phenotype and their close interaction with synovium in the same functional unit.

Induction of an inflammatory and prodegradative phenotype in autologous fibroblast-like synoviocytes by the infrapatellar fat pad from patients with knee osteoarthritis.

OBJECTIVE: The infrapatellar fat pad (IFP) of the knee joint has an inflammatory phenotype in osteoarthritis (OA). Its close proximity to the synovial membrane suggests that the IFP could be involved in the induction of OA synovitis. This study was undertaken to investigate the response of fibroblast-like synoviocytes (FLS) to autologous IFP and subcutaneous adipose tissue (SCAT) from patients with severe knee OA. METHODS: Samples of IFP, SCAT, and autologous synovial membrane tissue close to the IFP were harvested during surgery from 28 patients with end-stage knee OA. FLS from 14 patients were stimulated with autologous IFP- or SCAT-conditioned medium, and levels of messenger RNA (mRNA) expression and protein release of interleukin-6 (IL-6), IL-8, secretory phospholipase A2 (sPLA2 ), cytosolic PLA2 , cyclooxygenase 2 (COX-2), microsomal prostaglandin E synthase, prostaglandin E2 (PGE2 ), and matrix metalloproteinases (MMPs) 1, 3, 9, and 13 were evaluated. Both IFP- and SCAT-conditioned medium were evaluated by enzyme-linked immunosorbent assay for secretion of IL-6, soluble IL-6 receptor (sIL-6R), IL-8, tumor necrosis factor α (TNFα), PGE2 , IL-1ß, and interferon-γ. In addition, OA FLS were treated with PGE2 receptor antagonists to evaluate the contribution of IFP-derived PGE2 to the inflammatory response of FLS to the IFP. RESULTS: Stimulation of OA FLS with IFP-conditioned medium induced the mRNA expression and protein release of IL-6, IL-8, sPLA2 , COX-2, PGE2 , and MMPs 1, 3, 9, and 13. The extent of stimulation was consistently stronger with IFP-conditioned medium than with SCAT-conditioned medium. Moreover, secretion of IL-6, sIL-6R, IL-8, TNFα, and PGE2 was greater in IFP-conditioned medium than in SCAT-conditioned medium, especially PGE2 , whose secretion was 75-fold stronger in IFP-conditioned medium (P < 0.0001). PGE2 receptor antagonists dose-dependently inhibited the release of IL-6, IL-8, and PGE2 by IFP-stimulated FLS. CONCLUSION: This study showed that the IFP has a potential role in the induction of synovial inflammation in patients with severe knee OA. Furthermore, secretion of PGE2 by the IFP may be involved in the OA inflammatory process.

Restriction of apolipoprotein A-IV gene expression to the intestine villus depends on a hormone-responsive element and parallels differential expression of the hepatic nuclear factor 4alpha and gamma isoforms.

The apoA-I/C-III/A-IV gene cluster, like most intestine-specific genes, displays a specific pattern of expression along the intestinal cephalocaudal and crypt-to-villus axes. We have shown that this specific pattern of expression requires the distal apoA-IV promoter and the apoC-III enhancer. Using a new set of transgenic mice, we demonstrate here that the restriction of apoA-IV gene transcription to villus enterocytes requires a hormone-responsive element (HRE) located within the apoA-IV distal promoter. We showed, using nuclear extracts from villus or crypt epithelial cells, that this HRE bound the transcription factor hepatic nuclear factor 4 (HNF-4). We also found that the HNF-4gamma isoform was produced only in the villus, whereas the HNF-4alpha isoform was produced along the entire length of the crypt-to-villus axis. Our results demonstrate that the HRE of the distal apoA-IV promoter is responsible for the restriction of gene expression to villus epithelial cells and that this HRE binds HNF-4 isoforms. The in vivo observation of parallel gradients for apoA-IV and HNF-4gamma gene expression raises questions concerning whether this transcription factor plays a specific role in the control of enterocyte differentiation.