GILZ Regulates the Expression of Pro-Inflammatory Cytokines and Protects Against End-Organ Damage in a Model of Lupus

Abstract

Glucocorticoid-induced leucine zipper (GILZ) mimics many of the anti-inflammatory effects of glucocorticoids, suggesting it as a point of therapeutic intervention that could bypass GC adverse effects. We previously reported that GILZ down-regulation is a feature of human SLE, and loss of GILZ permits the development of autoantibodies and lupus-like autoimmunity in mice. To further query the contribution of GILZ to protection against autoimmune inflammation, we studied the development of the lupus phenotype in Lyn-deficient (Lyn^-/-) mice in which GILZ expression was genetically ablated. In Lyn^-/- mice, splenomegaly, glomerulonephritis, anti-dsDNA antibody titres and cytokine expression were exacerbated by GILZ deficiency, while other autoantibody titres and glomerular immune complex deposition were unaffected. Likewise, in patients with SLE, GILZ was inversely correlated with IL23A, and in SLE patients not taking glucocorticoids, GILZ was also inversely correlated with BAFF and IL18. This suggests that at the onset of autoimmunity, GILZ protects against tissue injury by modulating pro-inflammatory pathways, downstream of antibodies, to regulate the cycle of inflammation in SLE.

Keywords: GILZ; IL-23; glomerulonephritis; glucocorticoid; systemic lupus erythematosus.

Figure 1

Effect of GILZ deficiency on development of splenomegaly in Lyn-deficient mice. (A) Spleen weights (mg) in male littermate mice according to age (days). Curves were fit to a one-phase decay model with robust regression. (B) Spleen weights in mice dichotomized according to the two age ranges 70-84 days (10-12 weeks) and 245-315 days (35-45 weeks). Kruskal-Wallis test followed by Dunn’s correction for multiple comparisons results are shown. *P<0.05, **P<0.01, ***P<0.005, ****P<0.0001. NS, not significant.

Figure 2

Effect of GILZ deficiency on glomerulonephritis in Lyn-deficient mice. (A) Renal injury was assessed by measurement of segmental necrosis (glomeruli affected in %) and glomerular crescents (glomeruli affected in %) in mice older than 200 days of age. (B) Representative images of glomerular lesions stained with periodic acid-Schiff (PAS) stain are shown, with segmental necrosis indicated with black arrowheads and glomerular crescents with white arrowheads. (C) Indicative differences in kidney size between mouse strains. Statistical significance was determined using Kruskal-Wallis test with Dunn’s correction for multiple comparisons (n=5-8). All data shown are mean ± SEM. NS, not significant.

Figure 3

Autoantibodies reactive to specified extractable nuclear antigens (ENA) in mouse sera. Detection of IgG F(ab)2 autoantibodies against extractable nuclear antigens (ENA) detected by flow cytometry of serum from (A) young mice (<150 days) or (B) older (>200 days). Bars show median fluorescence intensity ± SEM normalized to positive control ENA beads. P-values were derived from Student’s t-test comparing Lyn^-/- with GILZ/Lyn^-/- mice. All data are presented as mean ± SEM (n = 6-24).

Figure 4

Immunofluorescence staining of immune complex deposition in glomeruli. (A) Immune complex (IC) deposition in glomeruli measured by immunofluorescence (IF) staining of complement component C3 (green) and immunoglobulin G (IgG; red). Representative images are shown from mice around 100 days of age. (B) IC deposition in mice young than 150 days or (C) older than 200 days was scored by a blinded expert using a scale of 0–3. (n=4–7). Statistical significance was determined using Kruskal-Wallis test with Dunn’s correction for multiple comparisons. All data are mean ± SEM. NS, not significant.

Figure 5

Effect of GILZ deficiency on serum cytokine expression in Lyn-deficient mice. Measurement of serum cytokines (pg/mL) including BAFF, IFNγ, IL-10, IL-1α, IL-1β, IL-18, IL-12p70, IL-6, IL-23A and IL-17A by Luminex in (A) young (< 200 days), and (B) aged mice (> 200 days). P-values were derived from Student’s t-test comparing Lyn^-/- with GILZ/Lyn^-/- mice. All data are presented as mean ± SEM (n = 4-20).

Figure 6

Correlations between GILZ and cytokine mRNA expression in patients with SLE. (Panels A, C, D) Publicly available microarray data was obtained from PBMCs of healthy controls and patients with SLE enrolled in the tabalumab trial at baseline (GSE88884). Relative mRNA expression levels of IFNG, IL12A, IL6, IL1B, IL10, IL17A, IL23A, BAFF and IL18 were mined from the dataset and their correlations with GILZ were determined. Panels depict the degree to which GILZ expression is correlated with individual cytokines based on Spearman rho values, including all SLE patients (n=1,760; panel A), or limited to patients not taking glucocorticoids (GC; n=460, panel C), or patients taking GC (n=1293 included, panel D). (B) Distributions of the relative expression (RE) of GILZ isoform 1 in the healthy subjects (n = 60) and SLE patients divided into those not taking (n = 460), and those taking (n = 1293), glucocorticoids. The P-value was derived by Kruskal-Wallis test followed by Dunn’s correction for multiple comparisons.