Do G-CSF and Neutrophils Contribute to the Pathophysiology of Rheumatoid Arthritis? — ASN Events

Do G-CSF and Neutrophils Contribute to the Pathophysiology of Rheumatoid Arthritis? (#254)

Gabrielle Goldberg 1 , Simon Chatfield 1 2 , Jane Murphy 1 , Ee Shan Pang 1 , Yunshun Chen 1 , Gordon Smyth 1 , Milica Ng 3 , Clare O'Neill 2 , Samantha Busfield 3 , Arna Andrews 3 , Ian Wicks 1 2
  1. Inflammation Division, Walter and Eliza Hall Institute, Parkville, Victoria, Australia
  2. Rheumatology Unit, Royal Melbourne Hospital, Parkville, Victoria, Australia
  3. CSL Limited, Bio21 Institute, Parkville, Victoria, Australia
Rheumatoid arthritis (RA) is characterised by a persistent, but poorly understood interplay between innate and adaptive immunity. Neutrophils are the predominant cell type in inflamed RA synovial fluid (SF). Granulocyte colony‑stimulating factor (G-CSF) is a key regulator of neutrophil production, function and survival. In this study peripheral blood (PB) neutrophil phenotype and function were analysed according to disease activity (DAS28 scores and other clinical parameters) in RA patients (n=50-60). We found significant correlations between disease activity and PB neutrophil percentage (r=0.26, p≤0.05), as well as between neutrophil activation state (CD62L, CD11b) and the expression of receptors for factors regulating neutrophil production and function. For example, CD62L vs G-CSF-R (r=0.58, p≤0.0001); CD62L vs CD35 (r=-0.37, p≤0.01); CD62L vs CXCR2 (r=0.42, p≤0.01); CD62L vs CXCR1 (r=0.46, p≤0.001). To further explore the role of neutrophils and G-CSF in RA, transcriptional profiling using RNASeq was performed comparing: PB neutrophils isolated from healthy donors (HD) and RA patients (n=5); neutrophils isolated from paired PB and SF samples of RA patients (n=3); neutrophils or white blood cells from HDs stimulated with G-CSF in vitro (n=4). 194 genes were differentially expressed (DE) in RA neutrophils (logFC≥1; adjusted p≤0.05) when compared to HDs. Over 50 of those genes were also differentially expressed when neutrophils were stimulated with G-CSF in vitro.  There were 1724 DE genes (logFC≥1; adjusted p≤0.05) when comparing PB and SF neutrophils from RA patients. Bioinformatic interrogation using Ingenuity Pathway Analysis software demonstrated that G-CSF was a likely regulator (p=9.83x10-21) of these differences. These data provide evidence that neutrophils and G-CSF contribute to the pathogenesis of RA. G-CSF may therefore represent a potential therapeutic target in the treatment of RA and other inflammatory diseases where there is a pathogenic contribution from neutrophils.