Abstract: Objective： To detect the levels of serum C-C chemokine ligand 19 (CCL19) in patients with systemic lupus erythematosus (SLE) and to evaluate the correlation between CCL19 expression and clinical features and laboratory parameters, trying to reveal the possible role of CCL19 in the pathogenesis of systemic lupus erythematosus. Methods: The levels of serum CCL19 were measured by enzyme linked immunosorbent assay (ELISA) in 90 patients with SLE and 30 healthy controls. These SLE patients included 75 patients who received treatment with glucocorticoids and disease-modifying anti-rheumatic drug (DMARD) and 15 patients without therapy. The frequencies of peripheral blood B cells and the B cell subsets were assessed in the patients with SLE by flow cytometry. The correlation between the clinical data, laboratory parameters, B cell subset frequencies and serum CCL19 levels were analyzed. Indepen-dent samples t test, paired t test, Pearson and Spearman correlation were used for statistical analyses. Results: The levels of CCL19 were markedly higher in the SLE patients without therapy and the patients with therapy than in the health controls［(596.25±409.19) ng/L and (422.90±395.84) ng/L vs. (157.79±125.23) ng/L, all P<0.001］. Serum CCL19 levels in the SLE patients without therapy were higher than the SLE patients who accepted glucocorticoids and DMARD treatment (P<0.05). The levels of serum CCL19 were positively correlated with anti-double stranded deoxyribonucleic acid (dsDNA), anti-nucleosome antibody (AnuA), IgA, IgG and IgM (r=0.38, P=0.007; r=0.332, P=0.029; r=0.519, P=0.007; r=0.461, P=0.018, respectively). Serum CCL19 levels in the SLE patients with photosensitivity, arthritis and secondary Sjgren’s syndrome were higher than the SLE patients without photosensitivity, arthritis and secondary Sj-gren’s syndrome, respectively ［(562.25±399.12) ng/L, (565.6±435.24) ng/L and (694.9±531.02) ng/L vs. (394.7±281.42) ng/L, (385.90±325.33) ng/L and (424.8±305.46) ng/L, all P<0.05］. The levels of serum CCL19 were positively correlated with the percentage of CD27-B cells and CD27-IgD-double-negative memory B cells (r=0.519, P=0.007; r=0.461, P=0.018, respectively). However, the levels of serum CCL19 were negatively correlated with the percentage of CD27+ memory B cells and CD27+IgD- switched memory B cells (r=－0.433, P=0.027; r=－0.616, P=0.001, respectively). Conclusion: The increased serum CCL19 levels in SLE patients were associated with the production of autoantibodies, and CCL19 might be involved in the pathogenesis of SLE by disturbing the homeostasis of B cell subsets.

Key words: Lupus erythematosus, systemic, C-C chemokine ligand 19, B cell, Autoantibody