Endogenous Toll-Like Receptor 9 Regulates AKI by Promoting Regulatory T Cell Recruitment

Abstract

Toll-like receptor 9 (TLR9) enhances proinflammatory responses, but whether it can act in a regulatory capacity remains to be established. In experimental murine AKI induced by cisplatin, Tlr9^−/− mice developed enhanced renal injury and exhibited fewer intrarenal regulatory T cells (Tregs) compared with genetically intact mice. A series of reconstitution and depletion studies defined a role for TLR9 in maintaining Treg-mediated homeostasis in cisplatin-induced AKI. When Rag1^−/− mice were reconstituted with nonregulatory CD25⁻ splenocytes from wild-type (WT) or Tlr9^−/− mice, AKI was similarly enhanced. However, when Rag1^−/− mice were reconstituted with CD4⁺CD25⁺ regulatory cells, WT CD4⁺CD25⁺ cells were more renoprotective and localized to the kidney more efficiently than Tlr9^−/− CD4⁺CD25⁺ cells. In Treg-depleted Foxp3^DTR mice, reconstitution with naive WT CD4⁺CD25⁺ cells resulted in less severe AKI than did reconstitution with Tlr9^−/− Tregs. Tlr9^−/− mice were not deficient in CD4⁺CD25⁺ cells, and WT and TLR9-deficient Tregs had similar suppressive function ex vivo. However, expression of adhesion molecules important in Treg trafficking was reduced on peripheral CD4⁺CD25⁺ cells from Tlr9^−/− mice. In conclusion, we identified a pathway by which TLR9 promotes renal Treg accumulation in AKI.

Following the recognition of the proinflammatory nature of TLR9 >10 years ago,¹ several studies have demonstrated that TLR9 promotes the development of autoimmune conditions, including experimental autoimmune encephalomyelitis and arthritis.^2,3 We have demonstrated that TLR9 ligation drives renal autoimmunity and vasculitis.^4,5 The identification of TLR9 as a proinflammatory target in autoimmune diseases has led to the development of TLR9 inhibitors to limit organ injury.⁶ However, evidence indicates that under some circumstances TLR9 mediates a regulatory effect.^7,8 TLR9-deficient MRL/lpr lupus-prone mice develop worse autoimmunity and lupus nephritis, potentially through enhanced effector responses and impaired regulatory T cell (Treg) suppressive function.^9–11 However, whether TLR9 regulates innate immunity has not been examined.

AKI is a major cause of morbidity and mortality,¹² with approximately 20% of cases occurring as a consequence of nephrotoxic drugs.¹³ Although cisplatin is a commonly used chemotherapeutic agent, its use is often compromised because of dose-dependent nephrotoxicity.¹⁴ Experimentally, cisplatin nephrotoxicity is characterized by an early influx of leukocytes that drive renal injury. Despite the innate nature of this model, Tregs have been found to limit inflammation and injury in cisplatin-induced AKI.^14,15

To investigate the role of TLR9 in AKI, we administered cisplatin to C57BL/6 wild-type (WT) and Tlr9^−/− mice. After 72 hours, mice developed AKI, with elevated serum urea and acute tubular necrosis. Compared with WT mice, Tlr9^−/− mice developed heightened renal functional and histologic injury (Figure 1A), demonstrating that TLR9 is protective in cisplatin-induced AKI. In experimental lupus, diabetes mellitus, and atherosclerosis, exaggerated injury in Tlr9^−/− mice has been attributed to enhanced adaptive autoimmunity.^7–9 However, the current model is one of short-term, drug-induced injury where innate immunity plays a significant role. Studies have shown pathogenic roles for neutrophils, neutrophil attracting chemokines, and CD4⁺ T cells in cisplatin-induced AKI.^16–18 Therefore, we examined renal neutrophil recruitment in WT and Tlr9^−/− mice. Neutrophil recruitment was increased in Tlr9^−/− mice (Figure 1B), as was renal expression of two key neutrophil chemoattractants, CXCL1 and CXCL2 (Figure 1C). However, renal interstitial CD4⁺ T cell recruitment did not differ in WT and Tlr9^−/− mice (Figure 1B). Evidence also indicates that Tregs play protective roles in this model, via inhibitory effects on innate immune cells.¹⁵ Therefore, we examined Treg recruitment in WT and Tlr9^−/− mice. Twenty-four hours after cisplatin administration, the proportion of renal CD4⁺CD25⁺ T cells was significantly reduced in Tlr9^−/− mice relative to WT mice (Figure 1D), together with reduced Foxp3 expression (Figure 1E). This finding suggests that TLR9 promotes Treg recruitment to the kidney in this model.

Figure 1: Acute renal injury and inflammation are enhanced in the absence of TLR9. (A) Seventy-two hours after cisplatin administration, renal function as measured by serum urea and renal histologic injury as assessed by tubular injury score on periodic acid-Schiff–stained sections were significantly higher in Tlr9^−/− mice compared with WT mice. Representative periodic acid-Schiff–stained images show greater disruption of the renal tubular architecture with increased tubular necrosis, cast formation, and tubular cell debris in Tlr9 ^−/− kidneys compared with WT mice (n=10 per group). Scale bar, 50 μm. (B) At 24 hours after injury, inflammation was assessed by quantification of neutrophils and CD4⁺ T cells. Compared with WT mice, interstitial neutrophils were increased in Tlr9 ^−/− kidneys, whereas CD4⁺ T cell numbers were similar (n=10 per group). Representative photomicrographs showing neutrophils and CD4⁺ T cells in WT and Tlr9 ^−/− kidneys. Arrows indicate leukocyte populations. Scale bar, 50 μm. (C) At 24 hours after cisplatin-induced AKI, kidney mRNA expression of key neutrophil chemoattractants CXCL1 and CXCL2 was increased in Tlr9 ^−/− mice compared with WT mice (n=9 per group). (D) Representative flow cytometry plots showing CD4⁺CD25⁺ cells in WT and Tlr9 ^−/− kidneys 24 hours after cisplatin injury gated on CD45⁺ population. The proportion of CD4⁺CD25⁺ cells in Tlr9 ^−/− kidneys was reduced compared with WT mice (n=6–7 per group). (E) At 24 hours after injury, kidney Foxp3 mRNA expression was higher in WT mice compared with Tlr9 ^−/− mice (n=9 per group). All bar graphs represent mean±SEM (unpaired t test). NS, not significant. *P<0.05; **P<0.01; ***P<0.001.

Increased injury in Tlr9^−/− mice in various inflammatory models has been postulated to stem from effects of TLR9 on the function of both effector T cells and Tregs.^{10,11,19–21} To assess the role of TLR9 in effector and regulatory T cells in cisplatin-induced AKI, we performed a series of reconstitution and depletion experiments, identifying the respective populations based on CD25 expression. First, to assess the possibility that the increased injury observed in Tlr9^−/− mice stemmed from enhanced function of effector T cells, we reconstituted Rag1^−/− mice with naive CD25⁻ splenocytes (nonregulatory cells) from WT and Tlr9^−/− mice and then induced cisplatin nephrotoxicity. Compared with unreconstituted Rag1^−/− mice injected with cisplatin (dotted line in Figure 2A), renal injury was similarly enhanced in cisplatin-injected Rag1^−/− mice reconstituted with CD25⁻ cells from WT or Tlr9^−/− mice (Figure 2A, Supplemental Figure 1). These findings are consistent with WT and Tlr9^−/− effector T cells having equivalent functional capacity in this setting. This finding was supported by analysis of WT and Tlr9^−/− mice depleted of CD25⁺ (regulatory) cells before cisplatin administration. Although injury was accelerated in mice depleted of CD25⁺ cells, requiring termination of the experiments after only 56 hours, renal functional and histologic injury did not differ in WT or Tlr9^−/− mice (Figure 2B, Supplemental Figure 1). Together these findings indicate that increased renal injury in Tlr9^−/− mice was not due to increased effector cell function.

Figure 2: CD4⁺CD25⁺ T cells are defective in Tlr9 ^−/− mice. (A) CD25⁻ (nonregulatory) splenocytes from naive WT and Tlr9 ^−/− mice were transferred into Rag1 ^−/− mice and cisplatin was injected 3 days later. Renal function was measured by serum urea and renal histologic injury was assessed by tubular injury score on periodic acid-Schiff–stained sections after a further 72 hours. Compared with cisplatin-induced renal injury in unreconstituted Rag1 ^−/− mice (shown by the dotted line), injury was more severe in Rag1 ^−/− mice reconstituted with CD25^- nonregulatory cells, but WT and Tlr9 ^−/− nonregulatory cells were similar in their disease-promoting effects (n=9 per group). (B) CD25⁺ cells were depleted in WT and Tlr9 ^−/− mice using a monoclonal anti-CD25 antibody before cisplatin administration. Experiments ended early at 56 hours because of accelerated disease. No difference in renal functional or structural injury was detected between the groups (n=10 per group). (C) Kidneys of Rag1 ^−/− mice reconstituted with WT CD4⁺CD25⁺ regulatory cells underwent less injury than those of Rag1 ^−/− mice reconstituted with Tlr9 ^−/− CD4⁺CD25⁺ regulatory cells and injected with saline, as evidenced by reduced serum urea and renal histologic injury (n=4–7 per group). (D) Endogenous Foxp3 cells were depleted by diphtheria toxin injections in Foxp3^DTR mice and mice then underwent reconstitution with CD4⁺CD25⁺ cells (2×10⁶) from WT or Tlr9 ^−/− mice. Cisplatin was administered 24 hours later and injury measured after a further 72 hours. Compared with Foxp3^DTR mice receiving WT CD4⁺CD25⁺ cells, AKI was more severe in Foxp3^DTR mice receiving Tlr9 ^−/− CD4⁺CD25⁺ cells. Renal injury was also greater in Foxp3^DTR mice that received DT compared with Foxp3^DTR mice that were not treated with DT (n=4–8 per group). All bar graphs represent mean±SEM (unpaired t test and one-way ANOVA). NS, not significant. *P<0.05; **P<0.01; ***P<0.005.

In vitro work in human cells has suggested that Treg and TLR9 activation occurs in tandem, and that TLR9 activation can regulate Treg suppressive activity.²² In addition, increased injury in Tlr9^−/− MRL/lpr lupus-prone mice has been associated with impaired Treg function.¹¹ These findings led us to hypothesize that increased cisplatin-induced AKI in Tlr9^−/− mice stemmed from impaired regulatory capacity. To test this hypothesis, we reconstituted Rag1^−/− mice with CD4⁺CD25⁺ regulatory cells before cisplatin administration. Given that Rag1^−/− mice are protected from cisplatin-induced AKI, the dose was increased (25 µg/g) to induce more disease and allow scope to demonstrate suppression.²³ Compared with mice reconstituted with CD4⁺CD25⁺ splenocytes from WT mice, renal injury was significantly greater after reconstitution with CD4⁺CD25⁺ splenocytes from Tlr9^−/− mice and saline injection (Figure 2C, Supplemental Figure 1), indicating that CD4⁺CD25⁺ cells require TLR9 to optimally regulate AKI. To more definitively test whether TLR9 intrinsic to Tregs regulates cisplatin-induced AKI, we depleted endogenous Tregs in Foxp3^DTR mice via diphtheria toxin injections and reconstituted them with CD4⁺CD25⁺ T cells from WT or Tlr9^−/− mice, before cisplatin administration.

Compared with mice receiving WT CD4⁺CD25⁺ T cells, functional and histologic renal injury was enhanced in mice receiving Tlr9^−/− CD4⁺CD25⁺ T cells, demonstrating that CD4⁺CD25⁺ T cells require TLR9 for optimal regulatory function in this model (Figure 2D, Supplemental Figure 1). One possible explanation for this finding is that regulatory function is impaired in Tlr9^−/− CD4⁺CD25⁺ Tregs. To address this issue, we compared suppressive capacity of WT and Tlr9^−/− Tregs ex vivo using a T cell proliferation assay. Tlr9^−/− and WT Tregs were equally effective at suppressing CD4⁺ T cell proliferation (Figure 3A), indicating that TLR9 does not modulate the suppressive activity of Tregs in these conditions. Gene expression data also showed that Tlr9^−/− CD4⁺CD25⁺ Tregs are not deficient in making mRNA for IL-10 (mean±SEM, 1.08±0.26 versus 3.10±0.73 relative quantification; P=0.04) and TGF-β (1.02±0.11 versus 1.7410±0.45 relative quantification; P=NS). Recently, Treg-generated adenosine has shown to regulate innate immunity in renal ischemia-reperfusion injury.^24,25 However, we found no difference in the cell surface expression of CD73, PD-1, and CTLA-4 on renal CD4⁺CD25⁺ cells in WT and Tlr9^−/− mice 24 hours after cisplatin (Figure 3B). This finding further highlights that Tlr9^−/− CD4⁺CD25⁺ cells do not have reduced functional capacity.

Figure 3: TLR9 does not affect CD4⁺CD25⁺ T cell proportions and ex vivo suppressive function. (A) Naive CD4⁺CD25⁺ splenocytes were sorted from WT and Tlr9 ^−/− mice and cocultured with CFSE-labeled WT CD4⁺CD25⁻ cells (T effectors [Teffs]) stimulated with precoated anti-CD3 and anti-CD28 for 72 hours. Suppressive function of CD4⁺CD25⁺ cells was measured by proliferation (percentage) of Teff CFSE dilution. Cells were plated at the Teff:Treg ratios indicated on the graph. US, unstimulated. (B) The expression of CD73, PD-1, and CTLA-4 on CD4⁺CD25⁺ cells in the kidney were assessed by flow cytometry 24 hours after cisplatin administration (n=5 per group). (C) Flow cytometry was used to determine the proportion and number of CD4⁺CD25⁺ splenocytes and the proportion and number of Foxp3⁺ cells relative to the total of CD4⁺CD25⁺ populations in spleens from WT and Tlr9 ^−/− mice. Data are shown for untreated (naive) animals and 24 hours after cisplatin administration (n=5 per group). (D) The proportions of necrotic (Annexin V⁻7AAD⁺), late apoptotic (Annexin V⁺7AAD⁺), and early apoptotic (Annexin V⁺7AAD⁻) CD4⁺CD25⁺ cells in the spleen and kidney were assessed by flow cytometry 24 hours after cisplatin administration (n=6–8 per group). All bar graphs represent mean±SEM (unpaired t test). *P<0.05; **P<0.01.

We next examined whether the heightened injury seen in Tlr9^−/− mice reflected their reduced presence in the periphery. We quantified both CD4⁺CD25⁺ T cells and Foxp3⁺CD4⁺CD25⁺ T cells in spleens from naive mice and 24 hours after cisplatin administration. Tlr9^−/− mice displayed increased numbers of CD4⁺CD25⁺ T cells and Foxp3⁺CD4⁺CD25⁺ Tregs in both naive animals and after cisplatin (Figure 3B). However, the proportion of Tregs relative to the total number of splenocytes did not differ between the strains (Figure 3C). There was also no difference in Treg numbers in the peripheral blood of naive WT and Tlr9^−/− mice (data not shown). These findings indicate that Tlr9^−/− Tregs are not defective by virtue of being less abundant. We also assessed whether the rate of apoptosis was altered in Tlr9^−/− Tregs, using flow cytometry to examine Annexin V/7AAD staining on splenic and renal Tregs 24 hours after cisplatin administration. Quantitation of early apoptotic, late apoptotic, and necrotic CD4⁺CD25⁺ T cell populations revealed no significant differences between WT and Tlr9^−/− cells (Figure 3D), indicating that altered apoptosis did not contribute to reduced Treg capacity in Tlr9^−/− mice in this model.

To exert renoprotective effects in AKI, Tregs must undergo recruitment to the inflamed kidney.²⁶ Therefore, we performed additional experiments to examine the ability of Tlr9^−/− Tregs to undergo recruitment to the kidney. We previously observed fewer Tregs in kidneys of Tlr9^−/− mice after cisplatin administration (Figure 1D), suggesting that the absence of TLR9 reduced the capacity of Tregs to undergo recruitment. To address this possibility specifically, we quantified Foxp3⁺ cell numbers in the kidneys of Rag1^−/− mice reconstituted with CD4⁺CD25⁺ cells from WT or Tlr9^−/− mice. There were fewer Foxp3⁺ cells in the kidneys of Rag1^−/− mice reconstituted with Tlr9^−/− CD4⁺CD25⁺ cells (Figure 4A), highlighting that in the absence of TLR9, Tregs have impaired recruitment to the kidney during AKI.

Figure 4: CD4⁺CD25⁺ T cells require endogenous TLR9 for optimal recruitment to the kidney after acute injury. (A) The number of Foxp3⁺ T cells recruited to the kidney following transfer of CD4⁺CD25⁺ cells into Rag1 ^−/− mice receiving cisplatin (shown in Figure 2C), as quantified via Foxp3⁺ staining on formalin-fixed kidney sections. Tlr9 ^−/− CD4⁺CD25⁺ cells were recruited to the kidney with reduced efficiency relative to WT CD4⁺CD25⁺ cells. Representative photomicrographs showing an increased number of Foxp3⁺ cells in kidneys of Rag1 ^−/− mice reconstituted with WT CD4⁺CD25⁺ cells compared with Tlr9 ^−/− cells (n=5–7 per group). Arrow indicates Foxp3⁺ staining in the kidney. Scale bar, 50 μm. (B) Comparison of expression of CD11a, CD44, and the α ₄ integrin on CD4⁺CD25⁺ cells in blood of naive WT and Tlr9 ^−/− mice, as determined by flow cytometry. The FACS histograms shown were generated by concatenation of at least seven individual observations for CD11a, CD44, and α ₄ integrin on WT (red histogram) and Tlr9 ^−/− (blue histogram) CD4⁺CD25⁺ cells. Gray-shaded histogram represents fluorescence minus one control. For CD11a and CD44, dashed lines define CD11a^hi and CD44^hi populations, respectively, with the graphs showing group data from the individual mice. For α ₄ integrin, mean fluorescence intensity data are shown (n=7–8 per group). All dot plots represent mean±SEM (unpaired t test). NS, not significant. *P<0.05; **P<0.01.

Results from other studies suggest that when Tregs enter the periphery they become CD44^hi and upregulate other homing receptors; these changes are important in facilitating their migration to nonlymphoid sites.^27,28 Therefore, we investigated the expression of adhesion molecules on circulating Tregs from naive WT and Tlr9^−/− mice. We focused on molecules of potential importance in Treg trafficking: CD11a, CD44, and the α₄ integrin. In WT mice, approximately 15% of Tregs expressed CD11a at high levels, while in Tlr9^−/− mice, the proportion of Tregs in this population was significantly reduced (Figure 4B). Similarly, for CD44, in WT mice 16% of Tregs were CD44^hi, while for Tlr9^−/− Tregs this figure was only about 10% (Figure 4B). In contrast, α₄ integrin expression did not differ between WT and Tlr9^−/− Tregs (Figure 4B). These data raise the possibility that changes in adhesion molecule expression on Tlr9^−/− Tregs underlie their reduced recruitment to the kidney. Moreover, these alterations identify a potential mechanism to explain their reduced capacity to regulate AKI.

In addition to adhesion molecules, chemokine receptors also allow Tregs to enter inflammatory sites. However, we found that downregulation of chemokine receptor gene or protein expression (CCR4, CCR6, CCR7, and CXCR3) does not account for the difference in regulatory capacity we observed in Tlr9^−/− mice (Supplemental Table 1). This was further supported with an in vitro chemotaxis assay toward the CCR4 ligand CCL22 that failed to show defective migration capacity of Tlr9^−/− CD4⁺CD25⁺ cells (Supplemental Figure 2).

Experimental cisplatin-induced AKI is increased in the absence of TLR9. This does not reflect upregulated effector cell responses but rather the effect of impaired recruitment of Tregs to the injured kidney. This is distinct from previous studies showing a regulatory function of TLR9 related to altered effector cell function.^7–10,19 We observed subtle but significant reductions in adhesion molecule expression on CD4⁺CD25⁺ T cells from Tlr9^−/− mice, providing a mechanism for their impaired recruitment to the kidney. The current study demonstrates for the first time an immunomodulatory role for TLR9 in AKI by enhancing the capacity of Tregs to migrate to damaged tissue in a model of common and relevant kidney disease.

Concise Methods

Mouse Model of Cisplatin-Induced AKI

Male C57BL/6 WT mice (8–10 weeks old) were purchased from Monash University Animal Services (Melbourne, Victoria, Australia). Rag1^−/− mice were bred at the Walter and Eliza Hall Institute (Melbourne, Australia). Tlr9^−/− were originally from S. Akira, and Foxp3^DTR mice were kindly provided by Dr. Tim Sparwasser and Dr. Katharina Lahl.^1,29 Studies adhered to the National Health and Medical Research Council of Australia guidelines for animal experimentation. Cisplatin (20–25 mg/kg; Sigma-Aldrich, St. Louis, MO) was injected intraperitoneally for all experiments, and mice were culled 24 or 72 hours later.

Reconstitution and Depletion Studies

For reconstitution studies, CD25⁻ splenocytes were isolated from WT and Tlr9^−/− mice by magnetic activated cell sorting (MACS) and negative selection, while CD4⁺CD25⁺ cells were isolated from WT and Tlr9^−/− mice by MACS and positive selection, both according to the manufacturer’s protocol (Miltenyi Biotec, Bergisch Gladbach, Germany). CD4⁺CD25⁺ cells were also purified by enriching the CD4+ T cells from spleen and lymph nodes of WT and Tlr9^−/− mice by MACS using the L3T4 microbead antibody (Miltenyi Biotec), labeling with rat anti-mouse CD25 (PE, PC61; BD Biosciences) and rat anti-mouse CD4 (FITC, GK1.5; BD Biosciences), and then sorting the double-positive population on the Mo-Flo XDP cell sorter (Beckman Coulter, Botany, New South Wales, Australia). For CD25⁻ reconstitution experiments, Rag1^−/− mice received 1×10⁷ CD25⁻ splenocytes intravenously, while for CD4⁺CD25⁺ reconstitutions, mice received 1×10⁶ CD4⁺CD25⁺ cells. Three days later, mice received cisplatin and experiments typically ended after 72 hours.

Two approaches were used in Treg depletion studies. For depletion of CD25⁺ cells, WT and Tlr9^−/− mice were administered anti-CD25 (PC61, 1 mg intraperitoneally). Mice were injected with cisplatin 24 hours later. In studies to deplete Foxp3⁺ Tregs, Tregs were depleted from Foxp3^DTR mice using a previously published protocol.²⁹ Diphtheria toxin was administered (1 μg intraperitoneally) twice, 3 days apart. Twenty-four hours later, mice were reconstituted with 2×10⁶ CD4⁺CD25⁺ cells from WT or Tlr9^−/− mice. Cisplatin was administered 24 hours later and experiments ended a further 3 days later.

Renal Injury

Kidney sections were fixed in buffered formalin for 24 hours, processed, and embedded in paraffin wax. Tubular injury was assessed on periodic acid-Schiff–stained sections, and scoring was performed as previously described.³⁰ The tubular injury score, determined by assessing tubular epithelial cell loss, tubular necrosis, accumulation of cellular debris, and tubular cast formation, was scored according to the percentage of affected tubules under high-power microscopy. The percentage of tubules affected was assigned a score: 0, normal; 1, ≤10%–25%; 2, 26%–50%; 3, 51%–75%; and 4, >75%. Serum urea was measured in serum samples collected at the end of the experiments using an auto-analyzer. Values for serum urea are recorded in mmol/L.

Real-Time PCR

RNA was extracted from whole kidney in Trizol (Life Technologies, Carlsbad, CA). RNA (500 ng) was treated with 1 unit of amplification-grade DNase I (Life Technologies), primed with random primers (Applied Biosystems, Foster City, CA) and reverse-transcribed using the High-Capacity cDNA reverse transcription kit (Applied Biosystems). Gene-specific oligonucleotide primers were designed using the Primer 3 software (Whitehead Institute for Biomedical Research, Cambridge, MA) and synthesized by Invitrogen as previously described.³¹ Values were standardized to the housekeeping gene 18S and expressed as a fold difference relative to WT mice receiving cisplatin.

CD4⁺ T Cell, Neutrophil, and Foxp3 Staining

Kidney sections were fixed in periodate lysine paraformaldehyde for 4 hours, then cryoprotected with 20% sucrose solution and frozen in liquid nitrogen. Tissue sections were cut and a three-layered immunoperoxidase technique was used to stain for T cells and neutrophils, as previously described.^5,32 Foxp3 was detected on formalin-fixed, paraffin-embedded 3-μm sections. Primary antibodies used were anti-mouse CD4 (GK1.5; American Type Culture Collection, Manassas, VA), anti–Gr-1 (RB6–8C5; DNAX, Palo Alto, CA) for neutrophils, and anti-mouse Foxp3 (FJK-16s; eBiosciences, San Diego, CA). The secondary antibody used was rabbit anti-rat biotin (BD Bioscience, North Ryde, Australia). To enumerate positive-stained cells, 10–20 consecutive interstitial sections were assessed per animal and results are expressed as cells per high-power field.

Leukocyte Isolation

Renal cell isolation was performed as previously described.³³ Briefly, kidneys were digested with 5 mg/ml collagenase D (Roche Diagnostics, Indianapolis, IN) and 100 μg/ml DNase I (Roche Diagnostics) in HBBS (Sigma-Aldrich) for 30 minutes at 37°C. Cells were filtered, erythrocytes lysed, and the CD45⁺ leukocyte population isolated by MACS using mouse CD45 microbeads (Miltenyi Biotec).

Spleens were minced and filtered through a 70-μM filter (BD Bioscience), and erythrocytes were lysed (NH₄Cl). Blood (1 ml) was collected by cardiac puncture and then the erythrocytes were lysed (NH₄Cl) for 10 minutes at room temperature.

Flow Cytometry

For surface staining, cells were stained for 20 minutes at 4°C with the following directly conjugated antibodies from BD Biosciences: anti-mouse CD25 (APC-Cy7, PC61), anti-mouse CD44 (APC, IM7), anti-mouse CD152 (CTLA-4, PE, UC10–4F10–11); BioLegend (San Diego, CA): anti-mouse CD45 (Pacific blue, 30-F11), anti-mouse CD11a (AF488, M17/4), anti-mouse CD194 (CCR4, APC, 2G12), anti-mouse CD196 (CCR6, APC, 29–2L17), and anti-mouse CD197 (CCR7, APC, 4B12); eBiosciences: anti-mouse CD4 (PE, GK1.5), anti-mouse CD73 (PE, Ty/11.8), anti-mouse PD-1 (PE, J43), and anti-mouse CD183 (CXCR3, APC, CXCR3–173); and anti-mouse CD49d (PS/2) is home-grown and conjugated with Alexa Fluor 488 (Molecular Probes, Eugene, OR). For intracellular Foxp3 staining, cells were surface stained as described, fixed and permeabilized for 60 minutes at 4°C with fixation/permeabilization solution (eBioscience), washed, and then stained with anti-mouse Foxp3 (eFlour 450, FJK-16s; eBiosciences) for 30 minutes at 4°C in permeabilization buffer (eBioscience). Apoptotic cell death and necrosis was analyzed by FITC Annexin V and 7AAD staining (BioLegend) as per the manufacturer’s instructions. All samples were acquired on the FACSCanto II machine using FACSDiva software (BD Biosciences), and flow cytometric data were analyzed using FlowJo (TreeStar, Inc., Palo Alto, CA).

Treg Cell Suppression Assay

CD4⁺ T cells were isolated from spleens of WT and Tlr9^−/− mice by MACS using mouse CD4 (L3T4) microbeads (Miltenyi Biotec), as previously described.³⁴ To purify CD4⁺CD25⁻ T cells and CD4⁺CD25⁺ Tregs, the enriched CD4⁺ T cells were incubated for 20 minutes at 4°C with rat anti-mouse CD25 (PE, PC61; BD Biosciences) and rat anti-mouse CD4 (APC, RM4–4; BD Biosciences) and sorted on the Mo-Flo XDP cell sorter (Beckman Coulter, Botany, NSW, Australia). Purified CD4⁺CD25⁻ T cells from WT mice were labeled with 10 μM CFSE (Life Technologies) in PBS supplemented with 0.1% BSA for 10 minutes at 37°C, and then washed in complete medium (RPMI containing 10% FCS, 2 mM L-Glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, 50 μM 2-ME). Carboxyfluorescein succinimidyl ester (CFSE)–labeled CD4⁺CD25⁻ T cells (10⁵/well) were co-cultured with WT or Tlr9^−/− CD4⁺CD25⁺ Tregs at the indicated ratios in 96-well round bottom plates precoated with anti-mouse CD3 (2.5 μg/ml; eBioscience) and anti-mouse CD28 (5 μg/ml; BioLegend, San Diego, CA) for 72 hours at 37°C, 5% CO₂. Cells were then harvested and CFSE dilution analyzed on the FACSCanto II machine using FACSDiva software (BD Biosciences), and flow cytometric data were analyzed using FlowJo (TreeStar, Inc.).

Statistical Analyses

A t test was used for analysis of two groups, and a one-way ANOVA (Tukey post hoc test) was used for more than two groups using GraphPad Prism software, version 6.0 (GraphPad Software, Inc., San Diego, CA). A P value <0.05 was considered to represent a statistically significant difference. Data are expressed as mean±SEM.

Disclosures

None.

The authors would like to thank Dr. Tim Sparwasser (The Centre for Experimental and Clinical Infection Research, Hannover, Germany) and Dr. Katharina Lahl (Technical University of Munich, Munich, Germany) for providing Foxp3^DTR mice.

This work was supported by funding from the National Health and Medical Research Council (NHMRC) of Australia (grant ID 1024289). M.J.H. is an NHMRC Senior Research Fellow (grant ID 1042775).