Caught in a Trap? Proteomic Analysis of Neutrophil Extracellular Traps in Rheumatoid Arthritis and Systemic Lupus Erythematosus

Neutrophil Extracellular Traps (NETs) are implicated in the development of auto-immunity in diseases such as rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) through the externalization of intracellular neoepitopes e.g., dsDNA and nuclear proteins in SLE and citrullinated peptides in RA. The aim of this work was to use quantitative proteomics to identify and measure NET proteins produced by neutrophils from healthy controls, and from patients with RA and SLE to determine if NETs can be differentially-generated to expose different sets of neoepitopes. Ultra-pure neutrophils (>99%) from healthy individuals (n = 3) and patients with RA or SLE (n = 6 each) were incubated ± PMA (50 nM, PKC super-activator) or A23187 (3.8 μM, calcium ionophore) for 4 h. NETs were liberated by nuclease digestion and concentrated onto Strataclean beads prior to on-bead digestion with trypsin. Data-dependent LC-MS/MS analyses were conducted on a QExactive HF quadrupole-Orbitrap mass spectrometer, and label-free protein quantification was carried out using Progenesis QI. PMA-induced NETs were decorated with annexins, azurocidin and histone H3, whereas A23187-induced NETs were decorated with granule proteins including CAMP/LL37, CRISP3, lipocalin and MMP8, histones H1.0, H1.4, and H1.5, interleukin-8, protein-arginine deiminase-4 (PADI4), and α-enolase. Four proteins were significantly different between PMA-NETs from RA and SLE neutrophils (p < 0.05): RNASE2 was higher in RA, whereas MPO, leukocyte elastase inhibitor and thymidine phosphorylase were higher in SLE. For A23187-NETs, six NET proteins were higher in RA (p < 0.05), including CAMP/LL37, CRISP3, interleukin-8, MMP8; Thirteen proteins were higher in SLE, including histones H1.0, H2B, and H4. This work provides the first, direct comparison of NOX2-dependent (PMA) and NOX2-independent (A23187) NETs using quantitative proteomics, and the first direct comparison of RA and SLE NETs using quantitative proteomics. We show that it is the nature of the stimulant rather than neutrophil physiology that determines NET protein profiles in disease, since stimulation of NETosis in either a NOX2-dependent or a NOX2-independent manner generates broadly similar NET proteins irrespective of the disease background. We also use our proteomics pipeline to identify an extensive range of post-translationally modified proteins in RA and SLE, including histones and granule proteins, many of which are known targets of auto-antibodies in each disease.