Background: Articular cartilage is an avascular tissue with limited oxygen supply. In a hypoxic environment, the metabolism of chondrocytes therefore relies mainly on glycolysis, which depends on the supply of glucose by membrane transporters. GLUT1 is the most abundantly expressed glucose transporter in chondrocytes.

Objectives: In this study, we examined the impact of GLUT1 inhibition on cartilage and chondrocyte metabolism in a murine model of osteoarthritis (OA).

Methods: Transcriptomic analyses (RNA-seq) were used to identify differentially expressed genes in murine primary articular chondrocytes harvested in 1% (hypoxia) or 21% O2 (normoxia). GLUT1 expression was assessed in human and murine (meniscectomy [MNX] model) OA cartilage samples. 12-week male adult mice undergoing MNX or Sham surgery received intra-articular injections of STF31, a pharmacological GLUT1 inhibitor, or DMSO. OA histopathological features (OARSI score, microCT analysis, synovial inflammation) were compared between groups 4 and 8 weeks after MNX. Murine chondrocytes and human OA cartilage explants were treated with inflammatory cytokines (IL-1β 1ng/ml, TNFα 25 ng/ml) in the presence or absence of STF31, in hypoxic (1% O2) or normoxic (21% O2) conditions. The effect of STF31 was determined in vitro by qPCR (mRNA expression of catabolic [Mmp3, Mmp13] and anabolic [Col2a1, Acan] markers), and ex vivo by histology (Safranin O staining) and immunohistochemistry (MMP13 labelling). The effect of GLUT1 inhibition on glycolysis and mitochondrial respiration (extracellular acidification rate [ECAR], oxygen consumption rate [OCR]) in chondrocytes stimulated by IL-1β or TNFα was examined by real-time metabolic analysis (Seahorse).

Results: In 21% O2 chondrocytes, we observed a decreased expression of several clusters of genes related to glycolysis as compared to 1% O2 chondrocytes, and a decreased expression of Slc2a1 (GLUT1-encoding gene; log2[fold change] = -1,27; adjusted p = 8,51x10^7). GLUT1 expression was decreased in OA vs. healthy cartilage, both in human and murine samples. In the MNX model of OA, STF31-treated mice had increased scores of cartilage degradation (mean ± SD OARSI score = 9.0 ± 2.2 vs 5.1 ± 2.0, p=0.0023) and synovial inflammation (mean ± SD synovitis score = 3.5 ± 0.3 vs 2.5 ± 0.4, p=0.0008), and decreased osteophyte volumes (mean volume = 60,7x10^4 μm3 vs 19,3 x10^4 μm3, p=0.0054) compared with control mice. STF31 aggravated the pro-catabolic and anti-anabolic effects of IL-1β and TNFα in murine articular chondrocytes and increased extra-cellular matrix degradation in human osteoarthritic cartilage explants. The Seahorse study showed that pro-inflammatory cytokines increased glycolysis in chondrocytes, an effect that was partially inhibited by STF31.

Conclusion: Our results suggest that the increase in glycolysis observed in chondrocytes subjected to pro-inflammatory cytokines is a protective response. Inhibition of GLUT1 weakens the chondrocyte and aggravates murine OA. The loss of GLUT1 expression could therefore contribute to the disruption of chondrocyte metabolism and cartilage damage in OA.

REFERENCES: NIL.

Acknowledgements: NIL.

Disclosure of Interests: None declared.