MicroRNA-155 Drives T_H17 Immune Response and Tissue Injury in Experimental Crescentic GN

Abstract

CD4⁺ T cells play a pivotal role in the pathogenesis of autoimmune disease, including human and experimental crescentic GN. Micro-RNAs (miRs) have emerged as important regulators of immune cell development, but the impact of miRs on the regulation of the CD4⁺ T cell immune response remains to be fully clarified. Here, we report that miR-155 expression is upregulated in the kidneys of patients with ANCA-associated crescentic GN and a murine model of crescentic GN (nephrotoxic nephritis). To elucidate the potential role of miR-155 in T cell-mediated inflammation, nephritis was induced in miR-155^−/− and wild-type mice. The systemic and renal nephritogenic T_H17 immune response decreased markedly in nephritic miR-155^−/− mice. Consistent with this finding, miR-155–deficient mice developed less severe nephritis, with reduced histologic and functional injury. Adoptive transfer of miR-155^−/− and wild-type CD4⁺ T cells into nephritic recombination activating gene 1-deficient (Rag-1^−/−) mice showed the T cell-intrinsic importance of miR-155 for the stability of pathogenic T_H17 immunity. These findings indicate that miR-155 drives the T_H17 immune response and tissue injury in experimental crescentic GN and show that miR-155 is a potential therapeutic target in T_H17-mediated diseases.

In the last decade, it has become clear that the CD4⁺ T helper cell-driven immune response significantly contributes to renal tissue injury in human and experimental crescentic GN.^1–3 CD4⁺ T cells can be classified according to their cytokine expression profile into four major subsets, namely T_H1, T_H2, T_H17, and regulatory T cells (Tregs).^4–7 Recent studies have highlighted the pivotal pathogenic role of the T_H1 and T_H17 immune responses in crescentic GN,^8,9 including the identification and characterization of IFN-γ– and IL-17A–producing CD4⁺ T cells in nephritic kidneys of mice and humans, as well as evidence for the contribution of IFN-γ and IL-17A/IL-23 to renal tissue injury in GN.^10–13 However, the precise regulation of systemic and renal CD4⁺ T cell immunity in crescentic GN remains to be elucidated.

Micro-RNAs (miRNAs) are small, noncoding RNAs that bind to mRNA and mediate mRNA cleavage, translational repression, or mRNA destabilization.¹⁴ miRNAs are essential for animals, and Dicer knockout mice, which are deficient in the enzyme for final miRNA processing, are not viable because of the lack of mature miRNAs.¹⁵ As fine-tuning regulators of gene expression, miRNAs are involved in various cellular processes and have emerged as important regulators of immune cell development. Mice deficient in T cell miRNAs have vital T helper cells with an impaired capacity to proliferate.¹⁶ Moreover, individual miRNAs have been implicated in T cell function, especially in influencing T_H1 cells,¹⁷ T_H17 cells,^18,19 and Treg cells.²⁰ Recently, miRNA-146 (miR-146) and miR-155 have been shown to be increased in both the urine and kidney of patients with IgA nephropathy.²¹ In addition, miR-193a might play a unique role in irreversible podocyte injury in human FSGS²²; however, the functional role of miRNAs in autoimmune-mediated kidney disease is largely unknown.

The aim of the present study was to elucidate the role of miRNAs in human and experimental crescentic GN. Therefore, we (1) assessed the expression profile of miRNAs in human and experimental crescentic GN, (2) studied the potential functional role of miRNAs in nephrotoxic nephritis with a specific focus on CD4⁺ T cell immune response, and (3) evaluated the potential mechanisms of miRNA-mediated regulatory processes in murine crescentic GN.

Results

miR-155 Is Abundantly Expressed in Human and Experimental Crescentic GN

As a first step, we investigated which miRNAs were abundantly expressed in the kidney in human ANCA-associated GN, the most common form of human crescentic GN. To address this question, we analyzed miRNA expression profiles in renal tissue from paraffin-embedded biopsies of patients with ANCA-associated GN compared with control biopsies from living and deceased kidney donors (Figure 1A). Of the miRNAs tested, the most abundantly regulated was miR-155, which was upregulated 2.9-fold. Although quantitative PCR analysis confirmed significant upregulation of mature miR-155 (3.2±0.42), it showed no regulation of the miRNAs let-7d and miR-301a (Figure 1B). Of note, the amount of renal miR-155 expression was inversely correlated to the GFR in these patients (Figure 1C) (P=0.05). We performed in situ hybridizations to localize renal miR-155. Figure 1D shows glomerular miR-155 expression in ANCA-GN patients but not healthy controls. Consecutive serial staining for the leukocyte marker CD45 suggests that miR-155 in ANCA-GN is predominantly expressed by infiltrating cells (Figure 1E).

Figure 1.

miR-155 is upregulated in the kidneys of patients with ANCA-associated and experimental GN. (A) Renal tissue from paraffin-embedded kidney biopsies from patients with ANCA-associated GN or control kidneys was analyzed for RNA expression using Affymetrix RNA Array Human U133 Puls 2.0. Corrected P values>0.05 were considered significant, and fold exchange from controls is given. (B) Quantitative PCR from renal tissue of patients with ANCA-associated GN (n=10) and controls (n=13) for miR-155, miR-301a, and let-7d in relation to RNU48. (C) Renal miR-155 expression levels were correlated to the renal function (GFR). (D) In situ hybridization of cryosections from renal tissue from ANCA-GN and control patients. Detection was performed with digoxigenin-labeled probes. (E) Confocal microscopic images of immunohistochemistry of consecutive renal tissue sections stained for CD45 and ToproIII. (F) miR-155 expression was quantified in the course of murine nephrotoxic nephritis in splenocytes and renal CD4⁺ cells. Bars represent mean ± SEM. **P<0.01.

To analyze the potential role of miR-155 in experimental crescentic GN, we induced nephrotoxic nephritis and measured systemic and renal miR-155 expression by quantitative PCR. miR-155 was upregulated in splenocytes on day 5 and FACS-sorted renal CD4⁺ T cells from day 10 on, with a maximum expression level on day 20 (Figure 1F).

miR-155 Contributes to Renal Tissue Injury in Experimental GN

To find out if miR-155 has a functional role in GN, we induced nephrotoxic nephritis (NTN) in miR-155^−/− and wild-type mice.²³ Figure 2A provides exemplified periodic acid–Schiff (PAS) staining of renal cortex from nephritic miR-155–deficient and wild-type mice, showing reduced crescent formation in miR-155^−/−. Glomerular crescent formation and tubulointerstitial injury were significantly reduced in nephritic miR-155–deficient mice compared with wild-type animals (Figure 2B). In addition, serum creatinine was significantly reduced, but the reduction in albuminuria did not reach statistical significance (Figure 2, C and D).

Figure 2.

miR-155^−/− mice are less susceptible to NTN. (A) PAS staining of renal cortex from miR-155–deficient and wild-type mice 10 days after NTN induction. (B) Crescent formation and tubulointerstitial damage in the different groups (n=10–16 per NTN group). (C) Albumin-to-creatinine ratio as measured by murine albumin-specific ELISA and urinary creatinine on day 8 on NTN induction. (D) Serum creatinine analyzed on day 10 after NTN induction. (E) Renal infiltration of inflammatory cells was quantified for CD3⁺, MAC2⁺, and F4/80⁺. Bars represent mean ± SEM. Con, control. *P<0.05; **P<0.01.

Infiltration of tubulointerstitial T cells and glomerular Mac2⁺ macrophages did not differ between nephritic wild-type and miR-155–deficient mice (Figure 2E), whereas tubulointerstitial F4/80⁺ macrophages/dendritic cells were significantly reduced (Figure 2E). We, therefore, analyzed infiltrating renal macrophages and dendritic cells in nephritic wild-type and knockout mice in more detail by FACS. However, no major phenotypic differences were found between these two groups with respect to MHCII expression, a marker for antigen presentation, and the activation marker Ly6C (data not shown).

miR-155 Drives the T_H17 Immune Response in Crescentic GN

To elucidate the pathogenic mechanisms of miR-155, we analyzed the T helper cell responses in nephritic miR-155–deficient and wild-type mice in more detail. Renal PCR analyses for T_H1, T_H2, T_H17, and Treg cytokines and transcription factor expression revealed a selective reduction in the T_H17 cytokine IL-17 (Figure 3A). We also stained renal T cells for the presence of intracellular cytokine production or FoxP3 expression. FACS dot plots in Figure 3B and the quantification in Figure 3C show a significant difference specific to IL-17–producing T_H17 cells, with a reduction of more than 50% in miR-155–deficient compared with wild-type mice. In supernatants from splenocyte cultures restimulated with sheep IgG, IL-17 was significantly reduced in miR-155^−/− cells, whereas IFN-γ was detected at higher levels (Figure 3D).

Figure 3.

miR-155 promotes the T_H17 immune response in experimental GN. (A) Renal cortex was analyzed by quantitative PCR for expression of IL-17, IFN-γ, FoxP3, and IL-4 (n=5–6 per NTN group). (B) FACS analysis for intracellular cytokines (IL-17, IFN-γ, and IL-4) in renal T cells (gated on CD45⁺CD3⁺CD4⁺ cells). (C) Quantification of intracellular (IL-17, IFN-γ, and IL-4) and intranuclear (FoxP3) FACS analysis from renal T cells. (D) IL-17 and IFN-γ measurement by ELISA from supernatants of cultured splenocytes from nephritic miR-155–deficient, wild-type, and control mice after stimulation with sheep IgG. (E) Histologic quantification of neutrophils in renal cortex on staining for granulocyte-differentiation antigen (Gr-1)-positive cells. (F) Chemokine expression as measured by quantitative PCR from renal cortex. (G) Crescent formation in miR-155–deficient and wild-type mice after NTN and treatment with anti–IL-17 antibody or isotype control. Animals were treated on days −1, 3, and 7. (H) Mice were treated with miR-155–specific inhibitors (antagomir-155) or nonbinding controls (scramblemir) before NTN induction. On day 10 after NTN, renal CD4⁺ T cells were analyzed for the intracellular cytokines IL-17 and IFN-γ. (I) Quantification of IL-17– and IFN-γ–positive cells. Bars represent mean ± SEM. *P<0.05; **P<0.01; *** P<0.001.

Neutrophil infiltration has been shown to be one of the effector mechanisms of the T_H17 immune response,²⁴ and accordingly, GR-1–positive neutrophils were markedly reduced in miR-155–deficient mice (Figure 3E). In line with a reduced renal T_H17 immune response, the T_H17-associated chemokine (C-C motif) ligand 20 (CCL20) was significantly reduced in the renal cortex of miR-155^−/− mice, which was measured by quantitative PCR. In contrast, no significant reduction was found in CCL2 and CCL5 (Figure 3F).

The attenuated severity of GN in miR-155–deficient mice might be because of the impaired T_H17 immune response. To prove this hypothesis, we treated nephritic wild-type and knockout mice with a neutralizing anti–IL-17 antibody.¹³ Renal PAS staining revealed almost identical histologic tissue injury with respect to glomerular crescent formation in the anti–IL-17 antibody-treated wild-type and miR-155^−/− groups 10 days after induction of nephritis (Figure 3G). Of note, anti–IL-17 treatment reduced glomerular crescent formation in nephritic wild-type but not miR-155–deficient mice. This result indicates that differences in IL-17/ T_H17 immunity might directly contribute to the ameliorated course of crescentic GN in miR-155–deficient mice.

To investigate if therapeutic administration of miRNA inhibitors is able to block the T_H17 immune response, we treated wild-type mice with an miR-155–specific antagomir (antagomir-155) or a nontargeting control (scramblemir) before NTN induction.²⁵ In nephritic mice, antagomir-155 reduced renal T_H17 cells (0.83%±0.02) compared with scramblemir-treated mice (4.76%±2.5). These findings underscore the importance of miR-155 in renal T_H17 immunity.

T Cell-Intrinsic Role of miR-155

To find out whether miR-155 drives the T_H17 immune response in cis by a T cell-intrinsic effect or in trans by other miR-155–expressing cells (for example, by dendritic cells or B cells), we transferred miR-155–deficient CD4⁺ T cells into T and B cell-deficient Rag-1^−/− mice and induced nephrotoxic nephritis. As indicated by crescent formation and tubulointerstitial injury score, miR-155–deficient T cells induced less renal tissue injury in nephritic recipients than wild-type T cells (Figure 4, A and B). Albuminuria and creatinine were also reduced but did not reach statistical significance (Figure 4, C and D). Furthermore, intracellular cytokine staining of renal T cells showed a diminished IL-17 production in CD4⁺ T cells in transferred miR-155^−/− cells (Figure 4E) and a marked decrease in renal IL-17 mRNA expression (Figure 4F). In contrast, the T_H1 immune response was not reduced after transfer of miR-155^−/− or wild-type cells (Figure 4, E and F).

Figure 4.

T cell-intrinsic effect of miR-155 on IL-17/T_H17 immune response in NTN. CD4⁺ splenocytes from miR-155–deficient and wild-type mice were purified using MACS sorting and transferred into Rag-1–deficient mice. Subsequently, NTN was induced. Depicted are (A) crescent formation, (B) tubulointerstitial injury, (C) albuminuria, and (D) serum creatinine levels assessed on day 10 after NTN induction (n=7 per NTN group). (E) FACS analysis of intracellular cytokine staining for IL-17 and IFN-γ of renal CD45⁺CD3⁺CD4⁺ T cells. (F) IL-17 and IFN-γ measurement by quantitative PCR from renal cortex of nephritic RAG-1–deficient mice. Controls are Rag-1–deficient mice without NTN and cell transfer. Bars represent mean ± SEM. *P<0.05; **P<0.01; *** P<0.001.

miR-155–Deficient T Cells Show Instability in T_H17 Phenotype

We next aimed at investigating how miR-155 influences the T_H17 immune response. We, therefore, polarized CD4⁺ splenocytes from miR-155^−/− and wild-type mice into the T_H17 and T_H1 phenotypes. This process resulted in polarization of 30%–40% into the T_H17 phenotype and 30%–50% into the T_H1 phenotype under T_H17- and T_H1-polarizing conditions, respectively. Unexpectedly, T_H17 polarization was not reduced in miR-155–deficient cells (Figure 5, A and B).

Figure 5.

Stability of T_H17 cells depends on miR-155. (A) FACS analysis of polarized T cells from miR-155–deficient and wild-type mice. Splenocytes were cultured under T_H0-, T_H1-, and T_H17-polarizing conditions. After 3 days of in vitro stimulation, cells were stained for intracellular cytokines. (B) Quantification of intracellular cytokine staining from FACS analysis (n=5 per group). (C) Congenic CD4⁺ splenocytes from CD45.2⁺ miR-155–deficient and CD45.1⁺ wild-type mice were polarized. Subsequently, competitive transfer of miR-155–deficient and wild-type cells into Rag-1–deficient mice was performed. On day 10 after NTN induction, mice were euthanized, and intracellular IL-17 and IFN-γ were measured by FACS analysis of CD4⁺ T cells from knockout and wild-type (n=8) mice. Bars represent mean ± SEM. *P<0.05.

Because the polarization did not differ between the groups, we hypothesized that the difference might to be caused by the stability of the T_H17 phenotype. We, therefore, transferred congenically marked wild-type and miR-155^−/− T_H17-polarized T cells into nephritic Rag-1^−/− mice at a 1:1 ratio. On day 10 after NTN induction, mice were euthanized, and renal T cells were analyzed by flow cytometry for intracellular cytokines. The renal CD4⁺ cell population was gated for CD45.2⁺ (miR-155^−/−) and CD45.1⁺ (wild type) and assessed for IL-17 positivity. The result differed significantly between the two groups (Figure 5C) and showed that 26.0% of wild-type cells and only 16.2% of miR-155^−/− cells were able to maintain the T_H17 phenotype. Thus, renal T cells showed reduced T_H17 stability in miR-155^−/− cells compared with congenic wild-type cells in the same kidney.

Role of miR-155 for the Humoral Immune Response in Experimental GN

Because miR-155 has been implicated in the formation of germinal centers,²³ we aimed to investigate the impact of miR-155 on the humoral immune response in our system. Therefore, we analyzed the mouse anti-sheep IgG-specific antibody responses in nephritic miR-155^−/− and wild-type mice. Indeed, miR-155^−/− mice had reduced serum titers of anti-sheep IgG antibodies, which were assessed by ELISA (Supplemental Figure 1A). In line with this finding, semiquantitative scoring of glomerular mouse IgG deposition was decreased in nephritic miR-155–deficient mice (Supplemental Figure 1C). In contrast, sheep IgG and C3 deposition were similar between nephritic wild-type and miR-155^−/− mice (Supplemental Figure 1, B and D).

To further analyze the role of miR-155 for B cells in experimental GN, we transferred sorted miR-155^−/− B cells or wild-type B cells combined with sorted wild-type T cells into Rag-1^−/− mice and induced NTN. As shown in representative PAS staining (Figure 6A) and quantification of glomerular crescents and tubulointerstitial injury score (Figure 6B), the tissue damage was similar between both groups. In addition, albuminuria and kidney function were not significantly altered (Figure 6C). However, mice that received miR-155–deficient B cells had reduced plasma IgM and IgG antibody levels (Figure 6D). Furthermore, the systemic and renal antigen-specific antibody response to sheep IgG was reduced, which was measured by semiquantitative scoring of glomerular mouse IgG depositions and plasma levels of mouse anti-sheep antibodies (Figure 6, E and F).

Figure 6.

Role of B cell miR-155 in experimental GN. Sorted B cells from miR-155–deficient and wild-type mice were transferred with wild-type CD4⁺ T cells into Rag-1^−/− recipients, and subsequently, NTN was induced. Controls are untouched Rag-1^−/− mice (n=7–10 per nephritic group). (A) PAS staining of renal cortex forms the groups as indicated. Depicted are (B) quantification of crescents and tubulointerstitial damage as well as (C) urinary albumin-to-creatinine ratio and serum creatinine. (D) Total IgM and IgG were measured by ELISA. (E) Glomerular antibody depositions of sheep IgG and mouse IgG were assessed by scoring of renal immunohistological staining. (F) Antigen-specific antibody titers from serum were measured by ELISA. (G) Renal cortex was analyzed by quantitative PCR for expression of IL-17, IFN-γ, FoxP3, and IL-4. (H) Quantification of intracellular cytokines (IL-17 and IFN-γ) by FACS analysis from renal CD4⁺ T cells. Bars represent mean ± SEM. Tub Int, tubulointerstitial damage. **P<0.01; *** P<0.001.

We also evaluated the impact of miR-155 in B cells on the T_H17 and T_H1 immune response. Quantitative PCR analysis and intracellular cytokine staining revealed no significant differences between the two nephritic groups (Figure 6, G and H), indicating that miR-155 in B cells has no major impact on the CD4⁺ T cell immune response in experimental GN.

Taken together, these results underscore the importance of miR-155 for antibody generation but argue against a major role of miR-155 in B cells for the clinical course of nephrotoxic nephritis.

Discussion

Here, we show that renal miR-155 expression is upregulated in patients with ANCA-associated GN and a murine model of crescentic GN. We provide a comprehensive analysis of the T cell immune response in nephritic miR-155–deficient mice and document the importance of miR-155 in systemic and renal T_H17 immune responses and consecutive renal tissue injury.

CD4⁺ T cells play a central role in the immunopathogenesis of crescentic GN. Studies by our group and others from recent years have underscored the importance of systemic and renal T_H1 and T_H17 immunity for renal tissue injury^11,13,26 and Treg cells in the protection against an overwhelming nephritogenic immune response.^27–29 However, the regulation of the T cell immune response in GN is still incompletely characterized.

miRNAs have recently been implicated in the regulation of T cell development and function.^16,30 It has been shown that miRNAs are of functional relevance in autoimmune disease, such as experimental autoimmune encephalitis, the murine model of multiple sclerosis.^18,31 Recently, the role of miRNAs in the kidney under homeostatic and pathologic conditions has emerged as a field of study of its own,^32,33 and although pilot studies have been undertaken to analyze the expression profile of miRNAs in patients with IgA nephropathy^21,34 and lupus nephritis,^35,36 their functional role in renal autoimmune disease is still largely unknown.³⁷ The first evidence for a functional role of miRNA in glomerular disease was derived from an elegant study showing that overexpression of miR-193a in mice resulted in podocyte injury and subsequent FSGS.²² Moreover, we found that mice with miRNA-deficient CD4⁺ T cells (CD4-cre × Dicer^fl/fl mice)^15,38 developed less severe GN (Supplemental Figure 2).

Because the expression profile of miRNAs in human crescentic GN is not defined, we analyzed miRNA expression in renal biopsies from patients with ANCA-associated GN. Gene expression array data and quantitative PCR for miRNA expression identified miR-155 as abundantly upregulated. Renal miR-155 expression colocalized with infiltrating CD45⁺ leukocytes, which is in line with a recent publication indicating that miR-155 is predominantly expressed by lymphocytes in human atopic dermatitis.³⁹ Of note, miR-155 expression was inversely correlated to renal function (GFR) in these ANCA-GN patients. This finding alone, however, is not enough to prove a pathogenic relevance of miR-155, because the correlation might solely reflect the amount of infiltrating (miR-155–expressing) lymphocytes, a well established risk factor in ANCA-associated GN.⁴⁰

To get more insight into the role of miR-155 in the immunopathogenesis of crescentic GN, loss- or gain-of-function experiments are most important. Consequently, we used the model of nephrotoxic nephritis and showed that miR-155^−/− mice developed less severe nephritis in terms of histologic and functional parameters. These data provide direct evidence for the pathogenic role of miR-155 in experimental crescentic GN.

In line with our results, miR-155 has also been shown to be of functional relevance in autoimmunity (e.g., experimental autoimmune encephalitis and autoimmune arthritis) as a result of alterations in immune cells, including T cells.^41–43 Moreover, miR-155 has been implicated in the regulation of distinct T helper cell subsets, often with conflicting results.^17,41,42,44 Here, we provide a comprehensive analysis of the renal and systemic T cell immune response and clearly show that the T_H17 immune response is significantly reduced in nephritic miR-155^−/− mice, whereas T_H1, T_H2, and Tregs are not altered. In line with this result, we found that miR-155 inhibition with antagomirs specifically reduced renal T_H17 immune response in nephritic mice. Moreover, transfer experiments of CD4⁺ T cells revealed a T cell-intrinsic effect of miR-155 on the T_H17 immune response. Unexpectedly and in contrast to recent publications,^43,45 in vitro polarization of naïve T cells into the T_H17 phenotype did not depend on miR-155. Using a well established protocol,⁴⁶ we achieved a T_H17 in vitro polarization rate of 30%–40%, which is in striking contrast to studies by O’Connell et al.⁴³ and Murugaiyan et al.⁴⁵ (3%–5%) and might, thus, explain the difference observed. Modification of the T_H17 polarization protocol (i.e., ± IL-23) did not reveal a difference between knockout and wild-type cells (data not shown). As we have shown in competitive transfer experiments, the reduced T_H17 immune response in miR-155–deficient cells is most likely attributable to an altered stability of the T_H17 phenotype.

In addition to the observed effect on T_H17 immunity, nephritic miR-155^−/− mice had a defect in the generation of antigen-specific IgG antibodies by unclear mechanisms. This finding is in line with recent reports^23,43,44,47 and suggests that therapeutic targeting of miR-155 might be beneficial in both humoral and T_H17 cell-mediated autoimmune diseases. Our miR-155^−/− B cell transfer experiments confirm the importance of miR-155 for antibody generation but argue against a major influence of miR-155 in B cells on the clinical course of the nephritis. However, we cannot rule out completely that miR-155 deficiency might modify the clinical course of the nephritis by affecting additional pathways beyond the T_H17 immune response.

One limitation of the present study is the lack of a direct mechanistic link between miR-155 and T_H17 cell immunity. In silico analysis using the miRANDA search algorithm has revealed negative regulators of the T_H17 immune response, namely suppressor of cytokine signaling 1 (SOCS-1), Src homology 2 domain-containing inositol-5-phosphatase 1 (SHIP-1), and E26 avian leukemia oncogene 1 (ETS-1), as potential miR-155 targets that might contribute to an impaired T_H17 immune response (data not shown). Indeed, these molecules have been shown (SOCS-1 and SHIP-1) or proposed (ETS-1) to be targets of miR-155.^19,20,41 However, the impact of miR-155–putative targets on T_H17 stability in vivo has yet to be clarified. For technical reasons, it is currently very difficult to isolate IL-17–producing T cells from nephritic mice for a direct comparison of the expression profile of potential miR-155 targets in wild-type and miR-155–deficient T_H17 cells. IL-17 reporter/fate reporter mice might help to overcome this problem in the future.^46,48

In summary, our study shows for the first time the functional importance of miRNAs for T cell-mediated crescentic GN. We provide evidence for upregulation of miR-155 in human and experimental GN. Moreover, miR-155 deficiency ameliorates experimental GN by limiting T_H17 immunity. Thus, miR-155 might be an intriguing therapeutic target in T_H17 cell-driven autoimmune diseases.

Concise Methods

Animals

All mice in this study were on C57BL/6 background. miR-155–deficient, Rag-1–deficient, and CD45.1-congenic mice were purchased from The Jackson Laboratory (Bar Harbor, ME). CD4-cre mice were from C. Wilson,³⁸ and conditional Dicer knockout mice were from C. Tabin.¹⁵ All animals were kept at the animal facility of the University Medical Center Hamburg-Eppendorf.

Induction of NTN and Cell Transfer

NTN was induced in 8- to 10-week-old male mice by intraperotineal injection of 0.5 ml nephrotoxic sheep serum as previously described.⁴⁹ Mice were euthanized on day 10. Blocking experiments were performed by subcutaneous injection of either 0.5 mg anti–IL-17 antibody or isotype control into each animal on days −1, 3, and 7 on NTN induction.¹³ In cell transfer experiments, Magnetic Activated Cell Sorting (MACS) of CD4⁺ spleen cells (Mouse CD4⁺ Cell Isolation Kit II; Miltenyi) and B cells (Mouse B Cell Isolation Kit; Miltenyi) was performed, and 2–5×10⁶ cells were intravenously injected into Rag-1^−/− mice. NTN was induced 24 hours later. In miR-155 inhibition experiments, scramblemir (5′-GACUCCACUCUUCUAGAAUAAC-3′) and miR-155–specific antagomir (5′-CCCCUAUCACAAUUAGCAUUAA-3′) were given at a dose of 80 mg/kg on 3 consecutive days²⁵ and subsequently, NTN was induced. miRNA inhibitors were synthesized by Dharmacon, Lafayette, CO.

Real-Time RT-PCR Analysis

Total RNA of the renal cortex was prepared according to standard laboratory methods. For miRNA quantification, RNA was prepared with the mirVana miRNA Isolation Kit and TaqMan-based detection primers (Life Technologies, Darmstadt, Germany). Real-time PCR was performed on a Step One Plus Real-Time PCR System (Applied Biosystems, Foster City, CA). All samples were run in duplicate and normalized to 18S rRNA or small RNAs as indicated to account for small variabilities in RNA and cDNA content.

Morphologic Examinations

Light microscopy and immunohistochemistry were performed by routine procedures. Crescent formation was assessed in 30 glomeruli per mouse in a blinded fashion in PAS-stained paraffin sections. As a measure of tubulointerstitial injury, the interstitial area was estimated by point counting four independent areas of renal cortex per mouse in low magnification fields (×200) as previously described.⁴⁹ Paraffin-embedded sections (2 µm) were stained with an antibody directed against the neutrophil marker GR-1 (Ly6 G/C; NIMP-R14; Hycult Biotech, The Netherlands), CD3 (A0452; Dako, Germany), F4/80 (BM8; BMA, Germany), MAC-2 (M3/38; Cedarlane, Canada), and sheep IgG or mouse IgG (Jackson Immunoresearch Laboratories), and they were developed with a polymer-based secondary antibody alkaline phosphatase kit (POLAP; Zytomed, Berlin, Germany) or the Vectastain ABC-AP Kit (Vector Laboratories, Burlingame, CA). MAC-2⁺ cells in 30 glomerular cross-sections and F4/80⁺, GR1⁺, and CD3⁺ cells in 30 tubulointerstitial high-powered fields (magnification, ×400) per kidney were counted by light microscopy in a blinded fashion. Glomerular sheep or mouse IgG deposition was scored from zero to three in 30 glomeruli per mouse as previously described.⁴⁹

Assessment of the Humoral Nephritogenic Immune Responses

Mouse anti-sheep IgG antibody titers were measured by ELISA using sera collected 10 days after induction of nephritis as previously described.⁴⁹ IgM and IgG titers were measured by Ready-Set-Go ELISA according to the manufacturer’s instructions (eBioscience, San Diego, CA).

In Situ Hybridization and Immunohistology in Cryosections

For in situ hybridization, cryosections from kidney biopsies of healthy controls and patients with ANCA-GN were labeled with miR-155–specific, digoxigenin-labeled locked nucleic acid (LNA) detection probes according to standard protocols (Exiqon, Vedbaek, Denmark). Cryosections were stained by indirect immunofluorescence for the leukocyte marker CD45 (HI30; Biolegend, San Diego, CA) using FITC-conjugated streptavidin (Vector, Burlingame, CA). Nuclei were stained with 1:1000 TO-PRO-3 (Molecular Probes) for 5 minutes at room temperature. Visualization was performed by confocal microscopy with an LSM 510 meta microscope (Zeiss, Jena, Germany) using the LSM software.

Splenocyte Cultures and T Cell Polarization

For splenocyte cultures, a total of 4×10⁶ cells/ml was incubated for 60 hours in the presence of sheep IgG. Supernatants were collected, and IL-17 and IFN-γ were measured by ELISA according to the manufacturer’s instructions (BioLegend, San Diego, CA).

For polarization of naïve CD4⁺ cells, splenocytes were isolated from naïve wild-type or miR-155^−/− mice using MACS CD4⁺ T Cell Isolation Kit II (Milteny Biotec, Bergisch-Gladbach, Germany). A total of 5×10⁵ cells/well was plated in anti-CD3 antibody-coated 96-well plates (eBioscience, San Diego, CA) and incubated for 65 hours with soluble anti-CD28. For T_H0 conditions, anti–IFN-γ was added, and for T_H1-polarizing conditions, anti–IL-4 and recombinant IL-12 were used. T_H17 polarization was achieved in the presence of anti–IL-4, anti–IFN-γ, IL-6, TGF-β, IL-1β, and IL-23 (BioLegend).

Flow Cytometry

The following fluorochrome-conjugated antibodies were used for FACS analysis: CD45 (30-F11), CD45.1 (A20), CD45.2 (104), CD3 (17A2), CD4 (GK1.5), CD8 (53–6.7), CD11b (M1/70), CD11c (HL3), and CD19 (6D5; BD Biosciences, Franklin Lakes, NJ; Biolegend, San Diego, CA; and eBioscience, San Diego, CA). Staining of intracellular IL-17A (TC11–18H10.1), IL-4 (11B11), and IFN-γ (XMG1.2) was performed as recently described.⁴⁹ In brief, cells were activated by incubation at 37°C and 5% CO₂ for 4 hours with phorbol 12-myristate 13-acetate (50 ng/ml; Sigma) and ionomycin (1 µg/ml; Calbiochem-Merck, Darmstadt, Germany) in X-VIVO medium (Lonza AG, Walkersville, MD). After 30 minutes of incubation, Brefeldin A (10 µg/ml; Sigma) was added. Dead cell staining was performed (LIVE/DEAD Fixable Read Dead Stain Kit; Invitrogen), and samples were acquired on a Becton & Dickinson LSRII System using the Diva software. Data analysis was performed with the FlowJo software (Tree Star Inc., Ashland, OR).

Cell Sorting

FACS sorting was performed on a Becton & Dickinson Aria III System. In case of MACS sorting, cells were labeled according to the manufacturer’s protocol (Miltenyi, Bergisch-Gladbach, Germany). The sorting result was verified by flow cytometry.

RNA Isolation and mRNA Array in Human Patients

Gene expression was profiled using Affymetrix RNA Array Human U133 Plus 2.0 arrays (Affymetrix, Santa Clara, CA) from paraffin-embedded renal specimens of patients with ANCA-GN or controls. Total RNA was isolated with the High Pure FFPE RNA Micro Kit (Roche, Grenzach-Wyhlen, Germany). Analysis of human tissue was approved by the local ethics committee (PV3162).

Statistical Analyses

Differences between two individual experimental groups (of normally distributed values) were compared by a two-tailed t test. In the case of multiple comparisons, one-way ANOVA with Bonferroni post-test for multiple comparisons was used. Array data were corrected for multiple testing by bootstrapping, and corrected P values<0.05 were considered statistically significant.

Disclosures

None.