Liraglutide, a glucagon-like peptide 1 receptor agonist, exerts analgesic, anti-inflammatory and anti-degradative actions in osteoarthritis

Abstract

Osteoarthritis (OA) is a common disabling disease worldwide, with no effective and safe disease-modifying drugs (DMOAD) in the market. However, studies suggest that drugs, such as liraglutide, which possess strong potential in decreasing low-grade systemic inflammation may be effective in treating OA. Therefore, the aim of this study was to examine the anti-inflammatory, analgesic, and anti-degradative effects in OA using in vitro and in vivo experiments. The results showed that intra-articular injection of liraglutide alleviated pain-related behavior in in vivo sodium monoiodoacetate OA mouse model, which was probably driven by the GLP-1R-mediated anti-inflammatory activity of liraglutide. Moreover, liraglutide treatment significantly decreased IL-6, PGE2 and nitric oxide secretion, and the expression of inflammatory genes in vitro in chondrocytes and macrophages in a dose-dependent manner. Additionally, liraglutide shifted polarized macrophage phenotype in vitro from the pro-inflammatory M1 phenotype to the M2 anti-inflammatory phenotype. Furthermore, liraglutide exerted anti-catabolic activity by significantly decreasing the activities of metalloproteinases and aggrecanases, a family of catabolic enzymes involved in cartilage breakdown in vitro. Overall, the findings of this study showed that liraglutide ameliorated OA-associated pain, possess anti-inflammatory and analgesic properties, and could constitute a novel therapeutic candidate for OA treatment.

Conflict of interest statement

Dr. Berenbaum reports personal fees from AstraZeneca, Boehringer, Bone Therapeutics, CellProthera, Expanscience, Galapagos, Gilead, Grunenthal, GSK, Eli Lilly, Merck Sereno, MSD, Nordic Pharma, Nordic Bioscience, Novartis, Pfizer, Roche, Sandoz, Sanofi, Servier, UCB, Peptinov, 4P Pharma, grants from TRB Chemedica; In addition. Dr Berenbaum is the CEO of 4Moving Biotech and chair of the scientific advisory board of 4P Pharma. Dr. Rattenbach is the CEO of 4P Pharma and the chairwoman of 4Moving Biotech. 4Moving Biotech and Sorbonne University own 2 patents for method of use of GLP1 analogs in the treatment of osteoarthritis (PCT/FR2013/051998 and PCT/IB2018//059100).

© 2022. The Author(s).

Figures

Figure 1

Expression of GLP-1 receptor in OA human and non-OA mouse knee joint. (a) Immunohistochemical staining of human OA knee cartilage sections was performed to determine the presence of GLP-1R (Mankin score: 3/14, scale bar = 100 µm or 50 µm). (b) Immunohistochemical staining of human OA synovial membrane sections was performed to determine the presence of GLP-1R (si: subintima, i: intima, v: blood vessel, scale bar = 100 µm or 50 µm). (c) Immunohistochemical staining of non-OA mice knee joint sagittal sections to determine the presence of GLP-1R (ac: articular cartilage, m: meniscus, js: joint space, T: tibia, F: femur, bm: bone marrow, scale bar = 100 µm or 50 µm). (d) Immunohistochemical staining of non-OA mice synovial membrane sagittal sections to determine the presence of GLP-1R (sm: synovial membrane, js: joint section, m: meniscus, scale bar = 100 µm or 50 µm). Control experiment was performed without primary antibody incubation. Arrows indicate example of cells positive for GLP-1R staining.

Figure 2

Liraglutide displayed analgesic effect in MIA mice models of OA. Mice knee joints were intra-articularly (IA) injected with 0.75 mg of MIA or saline on day 1. For the short-term study (a,b), treatments (liraglutide, dexamethasone, or vehicle) were injected IA on day 3 and inflammation pain sensitivity was determined by the von Frey test on day 2 (for randomization), 7, and 10 (n = 15–19 per group). For the long-term study (c), treatments were administered on days 8, 15, and 22, and von Frey tests were performed on day 7 (for randomization), 14, 21, and 28 (n = 9–10 per group). (a) Paw withdrawal threshold was assessed by von Frey filament stimulation on days 2, 7, and 10. (b) The efficacy rate of liraglutide in the MIA short-term study was analyzed using GraphPad Prism 9.0 and the EC50 value was determined at day 10 from the von Frey calculated values. (c) Paw withdrawal threshold was assessed by von Frey filament stimulation on days 0, 7, 14, 21, and 28. Arrows indicate the treatment with IA administration. Statistical analysis: Mean ± SEM. Mann–Whitney test with sequential strategy, **p < 0.01, ***p < 0.001, ****p < 0.0001 versus MIA control.

Figure 3

Liraglutide displayed anti-inflammatory effect in vivo in short-term MIA mice model of OA. Mice knee joints were intra-articularly (IA) injected with 0.75 mg of MIA or saline on day 1. Treatments (liraglutide, dexamethasone, or vehicle) were IA injected on day 3 (n = 15–19 per group). At the end of the study, on day 11, the mice were euthanized, and the right knee joint was recovered for histological analyses. (a) Representative photographs of sagittal sections of the paws of mice IA injected with 0.75 mg of MIA to induce inflammation and subsequently IA treated with vehicle, liraglutide or dexamethasone (positive control). (b) Histogram representing the total synovitis score calculated from Krenn et al., synovitis score system (n = 8–9 per group). (c) Based on the results of the von Frey test on day 10 and the synovitis score obtained for each animal, a correlation curve between these two parameters was calculated using GraphPad Prism 9.0 (n = 8–9 per group). Statistical analysis: Mean ± SEM. Mann–Whitney test with sequential strategy, **p < 0.01, ****p < 0.0001 versus MIA control. Simple linear regression, ****p < 0.0001.

Figure 4

Anti-inflammatory effects of liraglutide in murine primary chondrocytes. Murine primary chondrocytes were stimulated with IL-1β (2 ng/mL) and co-treated with 10 doses of liraglutide (6.6, 13.3, 26.6, 53.1, 106.3, 212.5, 425, 850 nM, 1.7 and 3.4 µM) for 24 h (n = 4). (a) The inhibition rate of liraglutide in murine primary chondrocytes was analyzed using GraphPad Prism 9.0, and the IC50 values were determined for NO, PGE2, and IL-6, which were detected in the culture supernatant. (b) Relative mRNA expression of iNos, Cox2, and Tnf-α in murine primary chondrocytes co-treated for 24 h with 2 ng/mL IL-1β and liraglutide (13.3 nM, 53.1 nM, and 1 700 nM, shown in red circles in (a)) or vehicle (n = 4). Statistical analysis: Mean ± SEM, Mann–Whitney test with sequential strategy, *p < 0.05, versus stimulated control (IL-1β alone).

Figure 5

Anti-inflammatory effects of liraglutide in murine macrophages. Murine macrophage RAW 264.7 cells were stimulated with 100 ng/mL of LPS and co-treated with 10 doses of liraglutide (6.6, 13.3, 26.6, 53.1, 106.3, 212.5, 425, 850 nM, 1.7 and 3.4 µM) for 24 h (n = 4). (a) The inhibition rate of liraglutide on RAW 264.7 cell line was analyzed by GraphPad Prism 9.0, and the IC50 values were determined for NO, PGE2, and IL-6 detected in the culture supernatant. (b) Relative mRNA expression of Il-6, Cox2, Tnf-α in RAW 264.7 cell line co-treated with LPS and liraglutide (13.3 nM, 53.1 nM, and 1 700 nM, shown in red circle in graphs (a)) or vehicle for 24 h (n = 4). (c) Relative mRNA expression of Mcp-1, Cd38, and Erg-2 in RAW 264.7 cell line treated with LPS and liraglutide (13.3 nM, 53.1 nM, and 1 700 nM, shown in red circle in (a)) or vehicle for 24 h (n = 4). Statistical analysis: Mean ± SEM, Mann–Whitney test with sequential strategy, *p < 0.05, versus stimulated control (LPS alone).

Figure 6

Anti-inflammatory effects of liraglutide are mediated by the GLP-1 receptor pathway. Primary cultured murine articular chondrocytes were incubated with 2 ng/mL IL-1β (a) and RAW 264.7, which were incubated with 100 ng/mL LPS (b) and co-treated with 50 nM of liraglutide or co-treated with three doses of exendin fragment 9–39 (25 nM, 50 nM, and 100 nM) for 24 h (n = 6). Nitrite, PGE2, and IL-6 concentration of the supernatant was determined. Statistical analysis: Mean ± SEM, Mann–Whitney test with sequential strategy, **p < 0.01, versus stimulated control (IL-1β or LPS) and $p < 0.05, $$p < 0.01, versus IL-1β + liraglutide or LPS + liraglutide group.

Figure 7

Anti-catabolic effects of liraglutide on murine primary chondrocytes. Murine primary chondrocytes were co-treated with 2 ng/mL of IL-1β and 10 doses of liraglutide (6.6, 13.3, 26.6, 53.1, 106.3, 212.5, 425, 850 nM, 1.7 and 3.4 µM) for 24 h (n = 4). (a) The inhibition rate of liraglutide in murine primary chondrocytes was analyzed using GraphPad Prism 9.0 and the IC50 values were determined. The expression of MMP-3, MMP-13, and GAG (rGAG) was detected in the cell supernatant. (b) Relative mRNA expression of Adamts4, Adamts5, Mmp-3, and Mmp-13 in murine primary chondrocytes co-treated with IL-1β and liraglutide (13.3 nM, 53.1 nM, and 1 700 nM) or vehicle for 24 h (n = 4). (c) Murine primary chondrocytes were stimulated with 2 ng/mL of IL-1β and co-treated with 50 nM of liraglutide for 72 h (n = 3). The concentration of glycosaminoglycans was determined in the culture supernatant using the GAG assay. Statistical analysis: Mean ± SEM, Mann–Whitney test with sequential strategy, *p < 0.05, **p < 0.01, versus stimulated control (IL-1β alone). ns non-statistical.