Background: Rheumatoid arthritis (RA) is a systemic autoimmune disease resulting in chronic inflammation and joint damage. Synovial joints exist under relatively low oxygen tension, and further reductions in oxygen availability during RA create a hypoxic cellular environment that contributes to joint pathogenesis. In this cellular response, hypoxia-inducible factor (HIF) plays an important role under the regulation of oxygen-sensing prolyl hydroxylases (PHD1–3). These PHDs also display HIF-independent functions involving the regulation of NF-kB, and might impact the inflammatory response in RA.

Objectives: To investigate the role of hypoxia and the exact contribution of the individual PHDs to RA pathogenesis.

Methods: We first investigated the effect of environmental hypoxia (10% vs 21% O ₂ ) on disease progression in a passive transfer arthritis model (collagen antibody-induced arthritis). In order to investigate the role of the individual PHD oxygen sensors, we then subjected mice with germline deficiencies of Phd1, Phd2 or Phd3 to the same arthritis model. A human synovial single-cell RNA-seq dataset was then probed to find the cell type that contributed the most to this phenotype. Based on these results, mice with myeloid-specific deletion of Phd1 were then generated and subjected to the arthritis model. Finally, gene expression of chemoattractants was measured in synovium from diseased mice and cultured human macrophages under hypoxic conditions.

Results: Mice housed in hypoxia (10% O ₂ ) versus normoxia (21% O ₂ ) presented lower clinical arthritis scores and reduced inflammation and damage in the knee joint. Phd2 ^± and Phd3 ^-/- mice showed similar clinical symptoms compared to controls, whereas Phd1 deletion protected against arthritis symptoms based on clinical scoring and histopathology. Based on human single-cell data, we could show that macrophages are among the main cell types with a robust hypoxic transcriptional signature in RA patients. Myeloid-specific deletion of Phd1 resulted in similar protection against arthritis compared to whole-body Phd1 deficiency, with a reduced expression of chemoattractants (e.g., CCL2, CCL3 and CCL4) in both synovium from diseased mice and cultured human macrophages under hypoxic conditions.

Conclusions: Our data indicate that the oxygen sensor PHD1 is a critical regulator of arthritis development and a potential target for the treatment of arthritis. We show that these protective effects are macrophage-driven and the resultant of reduced macrophage infiltration in the inflamed joint, rather than depending on an M1-to-M2 macrophage polarization.

REFERENCES: NIL.

Acknowledgments: NIL.

Disclosure of Interests: None declared.