Myeloid-Derived Suppressor Cells Induce Podocyte Injury Through Increasing Reactive Oxygen Species in Lupus Nephritis

Abstract

The expansion of myeloid-derived suppressor cells (MDSCs) has been documented in murine models and patients with lupus nephritis (LN), but the exact role of MDSCs in this process remains largely unknown. In this study, we investigated whether MDSCs are involved in the process of podocyte injury in the development of LN. In toll-like receptor-7 (TLR-7) agonist imiquimod-induced lupus mice, we found the severe podocyte injury in glomeruli of lupus mice and significant expansion of MDSCs in spleens and kidneys of lupus mice. The function of TLR-7 activated MDSCs was enhanced including the increased generation of reactive oxygen species (ROS) in vivo and in vitro. Moreover, the ROS production of MDSCs induced podocyte injury through activating the p-38MAPK and NF-kB signaling. Furthermore, we verified that podocyte injury was indeed correlated with expansion of MDSCs and their ROS secretion in LN of pristane-induced lupus mice. These findings first indicate that the podocyte injury in LN was associated with the increased MDSCs in kidney and MDSCs may be a promising therapeutic target of LN in the future.

Keywords: lupus nephritis; myeloid-derived suppressor cells; podocytes; reactive oxygen species; toll-like receptor-7.

Figure 1

Podocyte injury in IMQ-induced lupus mice. C57BL/6 mice were treated with IMQ or PBS for 10 weeks, sacrificed to spleens and kidneys for analysis. (A) Marked splenomegaly in IMQ-treated mice, compared with wild mice. (B) PAS staining of kidneys histologic differences in wild mice and IMQ mice. (C) Proteinuria in wild mice and IMQ mice was determined using Mouse Albumin ELISA Quantitation Set. (D,E) Immunofluorescence of glomerular Wilms’ tumor protein (WT-1) (D) and nephrin (E). (F) Assessment of podocyte foot processes by transmission electron microscopy. (G,H) Expression of nephrin and WT-1 in wild mice and IMQ-treated mice was determined by quantitative real-time RT-PCR. Data represent the mean scores ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Figure 2

Expansion of myeloid-derived suppressor cells (MDSCs) in IMQ-treated mice. (A,B) The frequency of renal (A) and splenic (B) MDSCs in wild mice and IMQ-treated mice were determined by FACS, right panel is the statistics for the results of flow cytometer. (C,D) Correlation analysis between mRNA of Wilms’ tumor protein (WT-1) (C), nephrin (D), and the frequency of MDSCs in IMQ-treated mice. Each symbol represents an individual mouse (WT-1: n = 16, r² = 0.5797, P = 0.0006; nephrin: n = 16, r² = 0.5168, P = 0.0014). (E) Expression of P47phox, inducible nitric oxide synthase, and arginase-1 in MDSCs purified from kidneys of wild mice and IMQ-treated mice was measured by quantitative real-time RT-PCR (qRT-PCR). (F) Expression of functional molecules from splenic MDSCs of wild mice and IMQ-treated mice was determined by qRT-PCR. Data represent the mean scores ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Figure 3

Toll-like receptor-7 (TLR-7)-activated myeloid-derived suppressor cells (MDSCs) induce podocyte injury. BM-derived MDSCs were co-cultured with mouse podocytes at ratio of 1:1 using transwell co-culture systems. (A) Representative light photomicrographs of podocytes after 48 h co-culture with MDSCs. (B,C) Expression of Wilms’ tumor protein (WT-1) (B) and nephrin (C) in podocytes was determined by immunofluorescence after 48 h co-culture with MDSCs. (D,E) Expression of nephrin and WT-1 in podocytes was determined by quantitative real-time RT-PCR (qRT-PCR) after 24 h co-culture with MDSCs (D) and Western Blot for nephrin and WT-1 after 48 h co-culture with MDSCs (E). (F) Expression of desmin and vascular endothelial growth factor in podocytes was determined by qRT-PCR after 24 h co-culture with MDSCs. (G) Podocyte apoptosis was assessed with Annexin V and propidium iodide by flow cytometry after 48 h co-culture with MDSCs. (H) Expression of IL-1β, IL-6, and TNF-α in podocytes was measured by qRT-PCR after 24 h co-culture with MDSCs. Data represent the mean scores ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Figure 4

Toll-like receptor-7 (TLR-7) activation induces myeloid-derived suppressor cells (MDSCs) to generate reactive oxygen species (ROS). (A) Expression of gp91^phox and p47^phox in MDSCs was measured by quantitative real-time RT-PCR (qRT-PCR) after 12 h treatment with R848 (10, 50, and 100 ng/ml); MDSCs treated with R848 for 24 h, loaded with DCFDA. Fluorescence intensity of DCFDA was measured by FACS. (B) Expression of arginase-1 in MDSCs was measured by QPCR after 12 h treatment with R848. (C) Expression of inducible nitric oxide synthase (iNOS) in MDSCs was measured by qRT-PCR after 12 h treatment with R848; NO in the supernatant of MDSCs was determined using NO detection kit after 24 or 48 h treatment with R848. (D) Expression of Wilms’ tumor protein (WT-1), nephrin, IL-1β, IL-6, and TNF-α in podocytes was measured by qRT-PCR after 24 h co-culture with treated R848 with MDSCs pretreated with N-acetyl-l-cysteine (NAC) for 1 h. (E) Expression of WT-1, nephrin, IL-1β, IL-6, and TNF-α in podocytes was measured by qRT-PCR after 24 h co-culture with treated R848 with MDSCs pretreated with NAC for 1 h. BM-derived MDSCs were pretreated with ROS inhibitors N-acetyl-l-cysteine (NAC, 5 mM) or iNOS inhibitors L-NMMA (0.5 mM) for 1 h, and co-cultured with mouse podocytes at ratio of 1:1 using transwell co-culture systems. Data represent the mean scores ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Figure 5

Myeloid-derived suppressor cells (MDSCs) induce podocyte injury by reactive oxygen species (ROS). BM-derived MDSCs were pretreated with ROS inhibitors N-acetyl-l-cysteine (5 mM) or inducible nitric oxide synthase inhibitors L-NMMA (0.5 mM) for 1 h, and co-cultured with mouse podocytes at ratio of 1:1 using transwell co-culture systems. (A) Representative light photomicrographs of podocytes after 48 h co-culture with MDSCs. (B) Expression of desmin in podocytes was measured by quantitative real-time RT-PCR after 24 h co-culture with MDSCs. (C,D) Podocyte apoptosis was assessed with annexin V by flow cytometry after 48 h co-culture with MDSCs (C) and the statistic results of podocyte apoptosis (D). Data represent the mean scores ± SEM. (E) The protein levels of p-p38, p-38, p-p65, and p-65 were assessed by western blot and normalized to GAPDH. *P ≤ 0.05, ** P ≤ 0.01, ***P ≤ 0.001.

Figure 6

Podocyte injury and expansion of myeloid-derived suppressor cells (MDSCs) in pristane-induced lupus mice. (A) PAS staining in kidneys of mice treated with or without pristane for 7 months. (B,C) Immunofluorescence of glomerular nephrin (B) and WT-1 (C) in kidney samples from BALB/c mice treated with PBS or pristane. (D) Assessment of podocyte foot processes by transmission electron microscopy. Boxed area in the above panel is shown at higher magnification at below. Original magnification 6,000× (up) and 25,000 (down). (E) Expression of nephrin and WT-1 was determined by quantitative real-time RT-PCR (qRT-PCR) in kidney from PBS or pristine-treated BALB/c mice. (F) Immunofluorescence of glomerular CD11b and GR-1, represent MDSCs, in PBS or pristine-treated BALB/c mice. (G) Expression of toll-like receptor-7 in renal MDSCs of PBS or pristine-treated BALB/c mice was determined by FACS. (H) The differences of reactive oxygen species in glomerulus were detected by DHE staining between wild mice and pristane-treated mice. (I) Expression of P47^phox in MDSCs purified from kidneys of PBS or pristine-treated BALB/c mice was measured by qRT-PCR. Data represent the mean scores ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.