AUTHOR=Wasén Caroline , Erlandsson Malin C. , Bossios Apostolos , Ekerljung Linda , Malmhäll Carina , Töyrä Silfverswärd Sofia , Pullerits Rille , Lundbäck Bo , Bokarewa Maria I. TITLE=Smoking Is Associated With Low Levels of Soluble PD-L1 in Rheumatoid Arthritis JOURNAL=Frontiers in Immunology VOLUME=9 YEAR=2018 URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2018.01677 DOI=10.3389/fimmu.2018.01677 ISSN=1664-3224 ABSTRACT=Background

Smoking is a risk factor for developing rheumatoid arthritis (RA), but the mechanism remains uncertain. We previously demonstrated that smoking lowers the T cell activation threshold by limiting programmed death protein 1 (PD-1) expression.

Aim

To investigate how smoking influence the levels of soluble PD-1 ligand (sPD-L1).

Method

Serum levels of sPD-L1 were measured in 246 RA patients and in 168 healthy subjects. The analysis was done with respect to inflammation, smoking, treatments, and autoantibody status. The effect of therapeutic TNF-inhibiting antibodies (TNFi) on sPD-L1 was studied in 16 RA patients at their first infliximab infusion. The expression of Fcγ-receptor (FcγR) subclass IIB and IIIA was analyzed with quantitative polymerase chain reaction in peripheral blood mononuclear cells (PBMCs) from 12 RA patients and 15 healthy controls, and in healthy PBMC exposed to IgG containing antibodies to cyclic citrullinated peptides (aCCP).

Results

The negative association between smoking and sPD-L1 in RA patients was established by multiple logistic regression (OR = 0.52, p = 0.038). Other covariates in the regression model were serum levels of IL-1β representing inflammation (OR = 1.6, p = 0.0076) and aCCP positivity (OR = 1.9, p = 0.047). First infliximab infusion repressed sPD-L1 (p = 0.023) in patients, and low levels of sPD-L1 were found in patients with early RA treated with TNFi (p = 0.018). Treatment with TNFi was associated with higher sPD-L1 in patients with long disease duration (p = 0.041) and restored levels in smokers. In vitro exposure to aCCP+ IgG suppressed sPD-L1 (p = 0.036), but aCCP+ patients with long disease duration had higher sPD-L1 (p = 0.016). High ratio of the inhibitory FcγR subclass IIB over the stimulatory IIIA resulted in low sPD-L1 release (p = 0.029). Smoking was associated with a higher FcγR IIB/IIIA ratio (p = 0.00062) and lower levels of sPD-L1 (p = 0.013).

Conclusion

In RA, serum sPD-L1 was related to systemic inflammation and aCCP positivity. Smoking altered the expression of FcγRs and limited sPD-L1 in RA patients, permitting inappropriate T cell responses. Differential regulation of sPD-L1 during the early and late RA may indicate transposition from acute to chronic inflammation.