Renal dendritic cells ameliorate nephrotoxic acute kidney injury

Abstract

Inflammation contributes to the pathogenesis of acute kidney injury. Dendritic cells (DCs) are immune sentinels with the ability to induce immunity or tolerance, but whether they mediate acute kidney injury is unknown. Here, we studied the distribution of DCs within the kidney and the role of DCs in cisplatin-induced acute kidney injury using a mouse model in which DCs express both green fluorescence protein and the diphtheria toxin receptor. DCs were present throughout the tubulointerstitium but not in glomeruli. We used diphtheria toxin to deplete DCs to study their functional significance in cisplatin nephrotoxicity. Mice depleted of DCs before or coincident with cisplatin treatment but not at later stages experienced more severe renal dysfunction, tubular injury, neutrophil infiltration and greater mortality than nondepleted mice. We used bone marrow chimeric mice to confirm that the depletion of CD11c-expressing hematopoietic cells was responsible for the enhanced renal injury. Finally, mixed bone marrow chimeras demonstrated that the worsening of cisplatin nephrotoxicity in DC-depleted mice was not a result of the dying or dead DCs themselves. After cisplatin treatment, expression of MHC class II decreased and expression of inducible co-stimulator ligand increased on renal DCs. These data demonstrate that resident DCs reduce cisplatin nephrotoxicity and its associated inflammation.

Figure 1.

DCs in the kidney are shown. (A) Flow cytometry of renal CD11c⁺ MHCII⁺ DCs from WT and CD11c-DTRtg mice for GFP expression. (B through F) Confocal microscopy of kidney sections of WT (B) and CD11c-DTRtg mice (C through F) for GFP-positive DC localization. (D) Magnification of boxed area in C showing GFP-positive cells between renal tubules. (E and F) Absence of GFP-positive cells (DCs) in glomerulus. Arrows point to GFP-positive DCs. Nuclei are stained with DAPI.

Figure 2.

DC depletion in CD11c-DTRtg mice after DT treatment is shown. (A and B) Histogram of GFP expression by splenic (A) and renal (B) CD11c⁺ MHCII⁺ DCs in WT and CD11c-DTRtg mice and their depletion at 24 h after DT treatment. (C and D) Flow cytometric quantification of spleen (C) and kidney (D) CD11c⁺ MHCII⁺ DCs obtained from CD11c-DTRtg mice at various time points after DT injection. *P < 0.05 versus 0 h group (n = 4 to 12).

Figure 3.

DC depletion increases susceptibility to cisplatin nephrotoxicity. WT and CD11c-DTRtg mice were treated with either DT or DT and cisplatin. (A and B) Blood collected at various time points with respect to cisplatin injection was analyzed for BUN (A) and serum creatinine (B) as a measure of renal function. **P < 0.01 versus all other groups; ⁺P < 0.05 versus all other groups (n = 9 to 17).

Figure 4.

Effect of DC depletion on renal morphology in cisplatin nephrotoxicity is shown. (A through D) WT mice (A and C) and CD11c-DTRtg mice (B and D) were treated with DT (A and B) or DT and cisplatin (C and D). Cisplatin was injected 24 h after DT injection (C and D). Kidneys were harvested at 48 h after cisplatin injection and stained with periodic acid-Schiff. (E) Tubular injury scoring in the renal cortex of periodic acid-Schiff–stained kidney sections harvested at 24 and 48 h after cisplatin injection. **P < 0.01 versus CD11c-DTRtg mice (DT + Cis); ⁺P < 0.05 versus WT mice (DT + Cis) (n = 4 to 5).

Figure 5.

Effect of DC depletion on renal leukocyte infiltration in cisplatin nephrotoxicity is shown. WT and CD11c-DTRtg mice treated with DT or DT and cisplatin were killed at 24 h, and single-cell kidney suspensions were analyzed by flow cytometry gating on CD45⁺ leukocytes for total numbers of CD4⁺ and CD8⁺ T cells, CD11b⁻PDCA-1⁺ plasmacytoid DCs, CD3⁻NK1.1⁺ NK cells, CD11c⁻B220⁺ B cells, and F4/80⁻Ly-6G⁺ neutrophils per kidney. *P < 0.05 versus all other groups; ⁺P < 0.05 versus WT mice (DT) (n = 4 to 5).

Figure 6.

Effect of DC depletion on renal neutrophil infiltration and localization in cisplatin nephrotoxicity is shown. (A through D) WT mice (A and C) and CD11c-DTRtg mice (B and D) were treated with DT (A and B) or DT and cisplatin (C and D). Cisplatin was injected 24 h after DT treatment (C and D). Kidneys harvested 48 h after cisplatin injection were stained for neutrophils. (E) Enumeration of neutrophils in renal cortex of kidney sections harvested at 24 and 48 h after cisplatin treatment. **P < 0.01 versus CD11c-DTRtg mice (DT + Cis); ⁺P < 0.05 versus WT mice (DT + Cis) (n = 4 to 5).

Figure 7.

Kinetic analysis of renal DC role in cisplatin nephrotoxicity is shown. WT and CD11c-DTRtg mice were treated with DT at −24, −1, or 24 h with respect to cisplatin injection. (A and B) Blood collected at 48 h after cisplatin injection was analyzed for BUN (A) and serum creatinine (B). (C and D) WT and CD11c-DTRtg mice were administered an injection of cisplatin followed by DT 24 h later. Blood collected at various time points with respect to cisplatin injection was measured for BUN (C) and serum creatinine (D). **P < 0.01 versus WT mice (n = 7 to 10).

Figure 8.

DC-mediated protection is independent of DT effect on nonhematopoietic cells. CD11c-DTRtg to WT chimera mice were administered an injection of saline, DT, cisplatin, or DT and cisplatin. (A and B) Blood collected at various time points with respect to cisplatin injection was analyzed for BUN (A) and serum creatinine (B). (C) Survival rate was determined in CD11c-DTRtg to WT chimera after the administration of cisplatin or DT and cisplatin (n = 8 to 13). ⁺P < 0.05 versus saline and DT; *P < 0.05 versus all other groups.

Figure 9.

Effect of dead cells on DC-mediated protection in cisplatin nephrotoxicity is shown. (A) Histogram of blood leukocytes obtained from WT(CD45.1) to WT(CD45.2), WT(CD45.1):CD11c-DTRtg(CD45.2) to WT(CD45.2), and CD11c-DTRtg(CD45.2) to WT(CD45.2) chimera to demonstrate the efficiency of donor bone marrow replenishment and the mixed chimerism of the WT(CD45.1):CD11c-DTRtg(CD45.2) to WT(CD45.2) mice. (B and C) Blood collected from WT(CD45.1) to WT(CD45.2) and WT(CD45.1):DTRtg(CD45.2) to WT(CD45.2) chimera mice that were administered an injection of DT and cisplatin at various time points with respect to cisplatin injection was measured for BUN (B) and serum creatinine (C). n = 6 to 7.

Figure 10.

Renal infiltration of inflammatory DCs, neutrophils, and monocytes in cisplatin nephrotoxicity is shown. (A and B) WT mice that were administered an injection of saline or cisplatin were killed at 48 or 72 h, and single-cell renal suspensions were analyzed by flow cytometry gating on CD45⁺ leukocytes for the expression of Gr-1 versus CD11b and CD11c (A) and Ly-6G versus 7/4 (B). (C) Absolute number of Gr-1⁺CD11b⁺CD11c⁻ leukocytes and Gr-1⁺CD11b⁺CD11c⁺ inflammatory DCs. (D) Absolute number of 7/4^hiLy-6G⁻ monocytes and 7/4^hiLy-6G⁺ neutrophils. *P < 0.05 versus 72 h cisplatin; ⁺P < 0.05 versus 48 and 72 h cisplatin (n = 3 to 4).

Figure 11.

Renal DC expression of surface markers in response to cisplatin treatment is shown. (A through C) WT mice treated with saline or cisplatin were killed at 24 h, and renal DCs (CD45⁺CD11c⁺NK1.1⁻PDCA⁻) were analyzed for the expression of MHCI, MHCII, CD40, CD80, and CD86 (A) and ICOS-L (B and C) by flow cytometry. *P < 0.05 versus saline (n = 3 to 5).