Interferon Regulatory Factor-5 in Resident Macrophage Promotes Polycystic Kidney Disease

Abstract

Background: Autosomal dominant polycystic kidney disease is caused by genetic mutations in PKD1 or PKD2. Macrophages and their associated inflammatory cytokines promote cyst progression; however, transcription factors within macrophages that control cytokine production and cystic disease are unknown.

Methods: In these studies, we used conditional Pkd1 mice to test the hypothesis that macrophage-localized interferon regulatory factor-5 (IRF5), a transcription factor associated with production of cyst-promoting cytokines (TNFα, IL-6), is required for accelerated cyst progression in a unilateral nephrectomy (1K) model. Analyses of quantitative real-time PCR (qRT-PCR) and flow-cytometry data 3 weeks post nephrectomy, a time point before the onset of severe cystogenesis, indicate an accumulation of inflammatory infiltrating and resident macrophages in 1K Pkd1 mice compared with controls. qRT-PCR data from FACS cells at this time demonstrate that macrophages from 1K Pkd1 mice have increased expression of Irf5 compared with controls. To determine the importance of macrophage-localized Irf5 in cyst progression, we injected scrambled or IRF5 antisense oligonucleotide (ASO) in 1K Pkd1 mice and analyzed the effect on macrophage numbers, cytokine production, and renal cystogenesis 6 weeks post nephrectomy.

Results: Analyses of qRT-PCR and IRF5 ASO treatment significantly reduced macrophage numbers, Irf5 expression in resident-but not infiltrating-macrophages, and the severity of cystic disease. In addition, IRF5 ASO treatment in 1K Pkd1 mice reduced Il6 expression in resident macrophages, which was correlated with reduced STAT3 phosphorylation and downstream p-STAT3 target gene expression.

Conclusions: These data suggest that Irf5 promotes inflammatory cytokine production in resident macrophages resulting in accelerated cystogenesis.

Graphical abstract

Figure 1.

1K Pkd1 mice have enhanced levels of macrophage chemoattractant cytokines and increased macrophage numbers compared with control kidneys. (A) Representative histology image of 1K and 2K kidneys taken from Pkd1 mice 3 weeks after unilateral nephrectomy (UNx). There are dilated tubules and mild cyst formation in 1K Pkd1 kidneys compared with 2K kidneys (magnification 20×). Quantification of cystic index is shown. (B) Number of kidney macrophages is shown as a percentage of total cells in 2K compared with 1K for both Pkd1 and control mice. There was a significant increase in the number of kidney infiltrating and resident macrophages in 1K Pkd1 kidneys compared with both 1K and 2K control kidneys. There was also an increased number of resident macrophages in 1K Pkd1 kidneys compared with 2K Pkd1 kidneys. (C) Kidney cytokine mRNA levels are shown as fold change compared with 2K control kidney. Csf1, Ccl2, Il6, Il1b, and Tnfa mRNA levels were significantly higher in 1K Pkd1 mice compared with both 1K control and 2K control mice. Csf1, Ccl2, Il1b, and Il6 levels were higher in 1K Pkd1 compared with 2K Pkd1 mice. ****P<0.0001, ***P<0.001, **P<0.01, *P<0.05, t test or one-way ANOVA with multiple comparisons with Tukey test. 1K, 1 kidney; 2K, 2 kidney.

Figure 2.

Irf5 expression is increased in kidneys isolated from Pkd1 mice and patients with autosomal dominant polycystic kidney disease (ADPKD). (A) Quantitative real-time PCR (qRT-PCR) analysis of whole-kidney Irf5 expression in control and Pkd1 kidneys 3 weeks post-UNx or sham surgery. Our data indicate that 1K Pkd1 kidneys had significantly increased levels of Irf5 expression compared with control (both 1K and 2K, ****P<0.0001) and 2K Pkd1 kidneys (***P<0.001, ****P<0.0001). (B) qRT-PCR analysis of whole-kidney IRF5 expression in tissue harvested from patients with ADPKD and controls. There is increased kidney IRF5 levels in patients with ADPKD compared with control patients (**P<0.01). (C) qRT-PCR analysis of Irf5 expression in flow-sorted infiltrating macrophages, resident macrophages, lotus tetragonolobus agglutinin (LTA⁺) proximal tubule epithelium, and dolichos biflorus agglutinin+ (DBA⁺) collecting duct epithelium from wild-type and Pkd1 mice kidneys (3 weeks post-UNx). Each dot represents an individual mouse kidney (N=4–5 mice per group). Irf5 was detected both in infiltrating and resident macrophage, but was not detected in proximal tubule (LTA⁺) and collecting duct (DBA⁺) epithelium in wild-type or 1K Pkd1 mice. Irf5 levels in infiltrating and resident macrophage from 1K Pkd1 mice kidneys were significantly higher compared with wild-type kidneys (one-way ANOVA). GAPDH, glyceraldehyde 3-phosphate dehydrogenase; Inf Mac, infiltrating macrophages; PKD, polycystic kidney disease; Res Mac, resident macrophages.

Figure 3.

Treatment of 1K Pkd1 mice with IRF5 antisense oligonucleotide (ASO) significantly reduces macrophage accumulation, proinflammatory gene expression, and renal cystic disease. (A) Treatment protocol for mice receiving IRF5 or scrambled (Scr) ASO. (B) Whole-kidney Irf5 mRNA levels shown as fold change compared with control cre- kidney. There was significant suppression of kidney Irf5 mRNA levels in 1K Pkd1 IRF5 ASO-treated mice compared with 1K Pkd1 scramble ASO-treated and control (cre-) mice (one-way ANOVA). **P<0.001, ***P<0.001, ****P<0.0001. (C) Number of kidney macrophages shown as percentage of total kidney cells in 1K Pkd1 IRF5 ASO- compared with Pkd1 scrambled ASO-treated mice. There was a significant reduction in number of infiltrating macrophages and resident macrophages after 6 weeks of treatment with IRF5 ASO compared with scramble ASO treatment (*P<0.05). (D) Representative confocal images of 1K Pkd1 IRF5 or scrambled ASO-treated mice stained with F4/80 (white), LTA (green), and Hoescht (dark blue). Our data indicate that F4/80⁺ macrophages accumulate in cystic regions of 1K Pkd1 ASO-treated mice. The number of these cells was reduced upon IRF5 ASO treatment. N=3 mice per group. (E) Kidney cytokine mRNA levels shown as fold change compared to with ASO-treated group. There were significant reductions in Ccl2, Csf1, Cx3cl1, Il6, and Il1b mRNA levels after 6 weeks of IRF5 ASO treatment. There were no differences in mRNA levels for Tnfa (P=0.07) and Il12 (P=0.53) between IRF5 and scramble ASO treatment groups (t test). *P<0.05, **P<0.01. (F) Kidney histology from 1K Pkd1 mice treated with IRF5 ASO or scramble ASO. Quantification of the cystic index (percentage cystic area/total kidney area) and cystic number is shown as mean±SEM in the associated graphs. **P<0.05; t test or one-way ANOVA with multiple comparisons with Tukey test; N=8–11 mice per group. Sac, euthanized.

Figure 4.

Treatment of 1K Pkd1 mice with IRF5 ASO reduces Irf5 and Il6 expression in resident, but not infiltrating, macrophages. (A) qRT-PCR analysis of Irf5 expression in flow-sorted cells from 1K Pkd1 mice treated with IRF5 or scrambled ASO for 6 weeks. mRNA levels for Irf5 were significantly lower in IRF5 ASO-treated kidney resident macrophages compared with scramble ASO-treated resident macrophages (**P<0.01, two way ANOVA). There were no differences in level of Irf5 in infiltrating macrophages upon IRF5 ASO treatment. Irf5 was not detected in either epithelial population. (B) Il6 mRNA levels are significantly lower in resident macrophages isolated from 1K Pkd1 IRF5 ASO mice compared with 1K Pkd1 scramble ASO-treated mice (*P<0.05, two way ANOVA). There were no differences in Il6 expression in infiltrating macrophages, LTA⁺ epithelium, or DBA⁺ epithelium between IRF5 ASO- and scramble ASO-treated mice. Data are shown as the mean±SEM. (C) Tnfa mRNA expression was not different between treatment groups. Data are shown as the mean±SEM.

Figure 5.

1K Pkd1 IRF5 ASO-treated mice have reduced STAT3 phosphorylation and expression of p-STAT3 target genes compared with scrambled ASO-treated mice. (A) Representative confocal images depicting p-STAT3 (white) expression in kidney tissue harvested from 1K Pkd1 mice treated with IRF5 or scrambled ASO. There is increased p-STAT3 staining in cystic tubules in scramble ASO- compared with IRF5 ASO-treated mouse kidneys (40× objective). (B) qRT-PCR of whole-kidney tissue isolated from 1K Pkd1 mice that were treated with IRF5 or scramble ASO shown as fold difference. Expression of Socs3 and C-myc were significantly reduced in the IRF5-treated group (**P<0.01, ***P<0.001, t test). There was no difference in kidney Cyclin D1 levels in IRF5 ASO treatment compared with scramble treatment. Expression of Socs3, C-myc, and Cyclin D1 (Ccnd1) are shown as the mean±SEM.

Figure 6.

Proposed mechanism of IRF5-dependent cyst growth. Unilateral nephrectomy causes renal hypertrophy–induced injury in Pkd1 mice resulting in production of macrophage chemoattractant cytokines in kidney epithelium (i.e., Ccl2 and Csf1) and an accumulation of Irf5-expressing inflammatory macrophages. These Irf5-expressing macrophages release proinflammatory cytokines (IL-6) stimulating STAT3 phosphorylation and cyst growth in mice lacking Pkd1. Suppressing Irf5 expression with ASO results in decreased cytokine production, reduced p-STAT3 in the epithelium, and slowed cyst growth.