Growth Differentiation Factor 15 Ameliorates Anti-Glomerular Basement Membrane Glomerulonephritis in Mice

Abstract

Growth differentiation factor 15 (GDF15) is a member of the transforming growth factor-β (TGF-β) cytokine family and an inflammation-associated protein. Here, we investigated the role of GDF15 in murine anti-glomerular basement membrane (GBM) glomerulonephritis. Glomerulonephritis induction in mice induced systemic expression of GDF15. Moreover, we demonstrate the protective effects for GDF15, as GDF15-deficient mice exhibited increased proteinuria with an aggravated crescent formation and mesangial expansion in anti-GBM nephritis. Herein, GDF15 was required for the regulation of T-cell chemotactic chemokines in the kidney. In addition, we found the upregulation of the CXCR3 receptor in activated T-cells in GDF15-deficient mice. These data indicate that CXCL10/CXCR3-dependent-signaling promotes the infiltration of T cells into the organ during acute inflammation controlled by GDF15. Together, these results reveal a novel mechanism limiting the migration of lymphocytes to the site of inflammation during glomerulonephritis.

Keywords: T cells; chemokines; glomerulonephritis; inflammation; innate immunity.

Figure 1

Evaluation of the anti-GBM model and expression of GDF15. (A) We used the commercially available GBM antiserum that was raised in sheep against rat GBM. We first examined its nephritogenic potential in C57BL/6 mice by assessing albuminuria 7, 14, and 21 days after a single intravenous injection of antiserum in pre-immunized mice, as well as in mice without pre-immunization (gray bar, 14 days). (n = 5, one-way ANOVA). (B) Total RNA isolated from kidneys of saline- or antiserum-injected C57BL/6 mice underwent quantitative real-time RT-PCR analysis and revealed significantly higher expression of Gdf15 in treated mice. (C) Serum GDF15 level was significantly increased in antiserum-injected C57BL/6 mice (n = 12, Student’s t-Test). Data are mean ± SEM. * p < 0.05; ** p < 0.01.

Figure 2

Systemic inflammation, kidney function, and histopathology of anti-GBM nephritis. (A) Sera were obtained from wild type or GDF15-deficient C57BL/6 mice on day 14 after saline or antiserum (anti-GBM) injection. Cytokine levels were quantified by flow cytometry (n = 15–17, one-way ANOVA). (B) Renal function parameter (n = 15–17, one-way ANOVA). (C) Serum IgG levels (n = 15–17, one-way ANOVA) and immunohistochemistry staining for IgG on kidney sections were quantified. (D) Kidneys from WT or KO mice were paraffin-embedded, stained with Periodic acid-Schiff (PAS) reagent, and quantified to assess tubular casts formation and tubular injury score (n = 8 mice per group, one-way ANOVA). Representative images of renal sections (original magnification 400×). Data are mean ± SEM. * p < 0.05; ** p < 0.01; *** p < 0.001.

Figure 3

Kidney histopathology of anti-GBM nephritis. Sections were obtained from wild type or GDF15-deficient C57BL/6 mice on day 14 after saline or antiserum (anti-GBM) injection. Kidneys from WT or KO mice were paraffin-embedded and stained either with (A) PAS, (B) anti-CD31 antibody, (C) anti-ki67 antibody and (D) anti-WT1 antibody, and quantified (n = 8 mice per group, one-way ANOVA). Representative images of renal sections (original magnification 400×). Data are mean ± SEM. * p < 0.05; ** p < 0.01.

Figure 4

Kidney inflammation in wild type and Gdf15 KO mice with anti-GBM nephritis. (A) Kidney sections were stained with anti- CD3, Ly6G, or Mac2 antibodies and quantified by counting, as indicated on graphs and in material and methods (n = 9–15 mice per group, one-way ANOVA). (B) Heat map depicting kidney expression of pre-selected genes of wild type and GDF15-deficient mice upon anti-GBM serum treatment. (C) Gene expression levels in kidneys were quantified by real-time PCR. Data are shown as means of the ratio of the specific mRNA vs. that of Gapdh mRNA (n = 6–8 samples per group, Student’s t-Test). Data are mean ± SEM. * p < 0.05; ** p < 0.01; *** p < 0.001 versus control mice.

Figure 5

Stimulation of splenic T cells from WT and Gdf15 KO mice with or without LPS ex vivo. (A) T cells were isolated from WT and Gdf15 KO mice and stimulated with or without LPS for 24 h. The percentage of CD4+CD8- and CD8+CD4- T cells was quantified by flow cytometry (gating strategy, n = 4–5 per group, one-way ANOVA). (B,C) Mean fluorescence intensity (MFI) of the surface markers CD28, CD154, CD62L, CD2, and CD11a/CD18 on LPS-stimulated or untreated CD4+CD8- T cells (B) and CD8+CD4- T cells (C) (n = 4–5 per group, one-way ANOVA). Data are mean ± SD. * p < 0.05.

Figure 6

Quantification of T cell migration and CXCR3 expression. (A) T cells from the spleens of WT and GDF15-deficient mice were isolated and migration assays were performed. The percentage of migrated T cells towards the chemokines CXCL10, CXCL1, and CCL5 was quantified after 24 h (n = 4 per group, one-way ANOVA, * p < 0.05). (B) Cxcr3 expression levels in LPS-stimulated or untreated T cells were quantified by real-time PCR. Data are shown as means of the ratio of the Cxcr3 mRNA vs. Gapdh mRNA (n = 4 per group, one-way ANOVA, * p < 0.05). (C) CXCR3 protein expression levels in LPS-activated T cells were quantified by Western blot. The histogram shows the densitometry of two independent experiments; no statistical analysis was performed. Data are mean ± SEM.

Figure 7

Gene expression analysis of GDF15, CXCL10, and CXCR3 genes in (A) glomerular and (B) tubular compartment of manually microdissected kidney biopsies from patients with RPGN. Values are expressed as a log2-fold change compared to controls (living donors, LD). All represented genes are significantly changed (q < 0.05). (A) Glomerular expression single hybridization (LD: n = 18, RPGN: n = 23), (B) Tubular compartment expression (LD: n = 18, RPGN: n = 21). Red represents upregulation and blue represents downregulation of the transcript.