IL-17 Receptor C Signaling Controls CD4⁺ T_H17 Immune Responses and Tissue Injury in Immune-Mediated Kidney Diseases

Abstract

Background 

IL-17A–producing CD4⁺ T helper (T_H17) cells play a critical role in autoimmune and chronic inflammatory diseases, such as crescentic GN. The proinflammatory effects of IL-17 are mediated by the activation of the IL-17RA/IL-17RC complex. Although the expression of these receptors on epithelial and endothelial cells is well characterized, the IL-17 receptor expression pattern and function on hematopoietic cells, e.g., CD4⁺ T cell subsets, remains to be elucidated.

Methods 

Crescentic GN (nephrotoxic nephritis) was induced in IL-17A, IFNγ, and Foxp3 triple-reporter mice for sorting of renal CD4⁺ T cell subsets and subsequent single-cell RNA sequencing. Moreover, we generated T_H17 cell–specific IL-17RA and IL-17RC gene–deficient mice and studied the functional role of IL-17 signaling in T_H17 cells in crescentic GN, imiquimod-induced psoriasis, and in the CD4⁺CD45RB^high T cell transfer colitis model.

Results 

We identified a specific expression of the IL-17 receptor A/C complex on CD4⁺ T_H17 cells. Single-cell RNA sequencing of T_H17 cells revealed the activation of the IL-17 receptor signaling pathway in experimental crescentic GN. Disruption of the IL-17RC signaling pathway in CD4⁺ T cells and, most importantly, specifically in CD4⁺ T_H17 cells, potentiates the IL-17 cytokine response and results in an accelerated course of experimental crescentic GN. Comparable results were observed in experimental models of psoriasis and colitis.

Conclusions 

Our findings indicate that IL-17 receptor C signaling has a previously unrecognized function in the regulation of CD4⁺ T_H17 cells and in the control of organ-specific autoimmunity and might provide new insights into the development of more efficient anti-T_H17 treatment strategies.

CD4⁺ T cells are key drivers of autoimmune and chronic inflammatory diseases. Their effector functions are largely mediated through the release of proinflammatory or regulatory cytokines. On the basis of their cytokine and transcription factor expression patterns, CD4⁺ T cells can be divided into functionally distinct subsets, such as T helper 1 (T_H1), T_H2, T_H17, follicular T_H, and regulatory T cells, and less well-characterized T_H9 and T_H22 cells.¹ In 2005, the general concept that only IFNγ-expressing T_H1 cells drive tissue injury in autoimmunity was challenged by the discovery of T_H17 cells, a highly pathogenic IL-17–producing CD4⁺ effector T cell subset.^2,3

T_H17 cells are characterized by the key transcription factors RORγt and STAT3; the production of cytokines IL-17A, IL-17F, IL-22, and GM-CSF; and by high CCR6 expression.^456–7 The current understanding of the pathogenic mechanisms in tissue-specific immunity and the pathways that lead to organ damage has notably advanced through the identification of the T_H17/IL-17 pathway.⁸ Numerous experimental and human studies showed that IL-17A-producing CD4⁺ T_H17 cells drive tissue injury in immune-mediated kidney^910–11 and skin diseases.¹² Furthermore, the T_H17/IL-17 axis plays a critical role in the immune regulation of inflammatory intestinal diseases.¹³ Despite these advances, the biologic functions and signaling pathways of IL-17 cytokines and their receptors remain to be fully elucidated.

The IL-17 family consists of six members (IL-17A to IL-17F), of which IL-17A and IL-17F are, by far, the best characterized.¹⁴ Their biologic effects are facilitated by binding to heterodimeric receptors of the IL-17 receptor family (IL-17RA to IL-17RE). Except for IL-17RB, all other receptor complexes contain the ubiquitously expressed subunit IL-17RA and a second, ligand-specific receptor. IL-17A and IL-17F bind to the same receptor complex, comprising IL-17RA and IL-17RC subunits. It is assumed that the IL-17RA/IL-17RC complex is predominantly expressed by epithelial and endothelial tissue cells and, thus, mediates the tissue-specific effects of these cytokines.¹⁵

In experiments leading to this study, we observed that the clinical course of crescentic GN (cGN) in mice completely lacking IL-17 cytokine signaling (IL-17RA–deficient mice) was almost comparable to that of wild-type animals. This was an unexpected finding that contrasts somewhat with other publications, including those from our own laboratory, which have shown that the IL-23/IL-17 axis plays a key role in kidney tissue damage in cGN.^910–11 Further analyses revealed increased production of T_H17-associated cytokines, such as IL-17A, IL-17F, IL-22, and GM-CSF, in these animals. Interestingly, we also found that T_H17 cells themselves highly express the IL-17RA/IL-17RC complex. Taken together, this leads to our hypothesis that IL-17 receptor signaling in T_H17 cells may limit the pathogenicity of the IL-23/IL-17 pathway in a negative feedback loop.

Therefore, in this study, we assessed the activation and expression pattern of IL-17 receptor subunits in CD4⁺ T cells and generated cell-specific IL-17 receptor knockout mice to define the effect of IL-17RA and IL-17RC signaling on the cellular immune response and the clinical outcome in experimental cGN.

Methods

Animals

Il17ra^−/− mice were obtained from Amgen; Il17rc^−/−, Il17rc^flox/flox, and Rag1^−/− mice were obtained from The Jackson Laboratory. Moreover, we successfully created Il17ra^flox/flox mice (in cooperation with the transgenic core facility of the University Medical Center Hamburg-Eppendorf, Dr. Hermans-Borgmeyer) and used recently generated IL-17A, IFNγ, and Foxp3 triple-reporter mice.¹⁶ In addition, Il17ra^flox/flox and Il17rc^flox/flox mice were successfully crossed to Il17a-Cre mice.¹⁷ All mice were on the C57BL/6J background. Age-matched C57BL/6J wild-type controls were bred in our facility. All mice were raised under specific pathogen-free conditions. All animal experiments were performed according to national and institutional animal care and ethical guidelines and were approved by the local authorities.

Human Tissue Analysis

All studies were approved by the Ethik-Kommission der Ärztekammer Hamburg, the local ethics committee of the chamber of physicians in Hamburg (registration numbers PV 5026 and PV 5822) and were conducted in accordance with the ethical principles of the Declaration of Helsinki.

Induction of Immune-Mediated Kidney, Skin, and Gut Diseases in Mice

cGN was induced in 8- to 12-week-old male mice by intraperitoneal injection of 2.5 mg of nephrotoxic sheep serum per gram of body weight.¹⁸ Urine samples were collected by housing the mice in metabolic cages for 3–5 hours. Urinary albumin excretion was determined by standard ELISA technology (Mice-Albumin Kit; Bethyl), and urinary creatinine and BUN levels were measured using standard laboratory methods. For IL-17A targeting, a neutralizing mouse antibody to IL-17A (clone MM17F3) was used at 500 μg per mouse. The antibody was injected intraperitoneally on days −1, 2, and 5 after induction of cGN.

For the induction of imiquimod (IMQ)-induced, psoriasis-like skin inflammation, one ear was treated daily with 5 mg IMQ cream (5% IMQ; MEDA Pharma, Bad Homburg, Germany).¹⁹ To induce T cell transfer colitis, 1 × 10⁴ CD4⁺CD45RB^high T cells were sorted from the spleen of mice and intravenously injected into Rag1^−/− mice.²⁰

Morphologic Analyses

Immunohistochemistry was performed using routine laboratory methods. In a blinded fashion, glomerular crescent formation and tubulointerstitial injury were assessed in sections stained with Periodic acid–Schiff (PAS) as described.²¹ Paraffin-embedded sections were stained with antibodies directed against GR-1 (Ly6 G/C, NIMP-R14; Hycult Biotech, Uden, The Netherlands), sheep IgG, or mouse IgG (both Jackson Immunoresearch Laboratories, West Grove, PA). GR-1⁺ cells in ten high-power fields per kidney were counted in a blinded manner.²²

Real-Time RT-PCR Analyses

Total RNA of the kidney was prepared according to standard laboratory methods. Real-time PCR was performed for 40 cycles on a StepOnePlus Real-Time PCR system (Applied Biosystems, Foster City, CA) as previously described.²² All samples were run in duplicate and normalized to 18S rRNA.

Isolation of Leukocytes from Murine Tissues

The leukocytes from murine kidneys, skin, and gut were isolated as previously described.²² Briefly, kidneys were digested for 45 minutes at 37°C by adding 0.4 mg/ml collagenase D (Roche, Mannheim, Germany) and 0.01 mg/mL DNase I (Roche) to RPMI 1640 medium (Life Technologies, Karlsruhe, Germany) supplemented with 10% heat-inactivated FCS (Gibco, Eggenstein, Germany). Subsequently, kidneys were finely minced using the gentleMACS Dissociator (Miltenyi Biotec, Teterow, Germany). Single-cell suspensions were separated using Percoll density gradient centrifugation.

Assessment of the Humoral Immune Responses

Mouse anti-sheep IgG antibody titers were determined in sera by ELISA. Glomerular mouse and sheep IgG deposition was assessed by quantification of immunohistochemical staining, as previously described.²³

Flow Cytometry

Cells were stained with fluorochrome-labeled antibodies directed against CD45, CD3, CD4, CD8, γδTCR, NK1.1, IL-17A, IL-17F, IL-22, IFNγ, CD11b, and Ly6G (Biolegend, BD Biosciences, eBioscience, or R&D Systems), as previously described.²⁴ Flow cytometry measurements were performed using the BD FACS LSR II. Data were analyzed by using the FlowJo software (Tree Star).

Hydrodynamic Gene Transfer for the Induction of IL-17A Overexpression

The principle of hydrodynamic gene transfer, which allows for nonviral introduction of plasmid DNA into hepatocytes, was described in detail elsewhere.²⁵ Briefly, 10 μg of either IL-17A or control vector in sodium chloride solution was administered by tail vein injection in a volume corresponding to 10% of mouse body weight.

Cell Transfer in Rag1^−/− Mice

CD4⁺ T cells were isolated by magnetic activated cell sorting using the Mouse CD4⁺ Cell Isolation Kit II (Miltenyi Biotec). A total of 0.5 × 10⁶–1.0 × 10⁶ wild-type, Il17ra^−/−, or Il17rc^−/− CD4⁺ T cells were intravenously injected (ratio 1:1) into Rag1^−/− mice. NTN was induced 8 days after cell transfer.

Single-Cell RNA Sequencing of Renal CD4⁺ T Cells

FACS-sorted cells were subjected to droplet-based single-cell sequencing using the 10x Chromium platform, as described previously (3′ version 2; 10x Genomics, Pleasanton, CA).^19,26 We used the Cell Ranger software pipeline (version 4.0.0; command, cellranger count; 10x Genomics) to demultiplex cellular barcodes and align reads to the mouse reference genome (refdata-gex-mm10-2020-A, reference 2020-A). We generated a feature-barcode matrix, containing 7641 cell barcodes and 1949 median features per barcode. The R package Seurat (version 3.1.4) was used for further analysis.²⁷ To remove low-quality cells or doublets, cells with <1000 or >4000 genes and cells with >10% mitochondrial genes were excluded. Single-cell RNA-sequencing data were normalized by applying the Seurat LogNormalize method with default parameters, scaled our Seurat object, and regressed out the total number of UMIs, the percentage of mitochondrial genes, and the G2M/S score of the cell cycle. The cell cycle scores were calculated by using CellCycleScoring and the cell cycle makers.²⁸

To reduce dimensionality, we detected highly variable genes (function, FindVariableFeatures; selection.method, “vst”; nfeatures, 2000) and performed the principal component analysis. We visually (method, Elbowplot) selected the principal components 1–20 to calculate the KNN graph (method, Find Neighbors), which then computed the cell clusters (method, FindClusters; resolution, 0.6). Uniform manifold approximation and projection was used to visualize clustering results. We removed clusters 9, 10, and 11, because these clusters contained only cells in G2M or S phase. To identify differentially expressed genes, we calculated the significant differential expressed (DE) genes (method, FindAllMarkers; logfc.threshold, 0.1) and named the clusters according to the top DE genes.

The intracellular IL-17 signaling score was generated by using the Seurat AddModuleScore and the following genes which are annotated to the KEGG-pathway “mmu04657”: Il25, Il17ra, Il17rb, Tradd, Fadd, Casp3, Casp8, Traf3ip2, Traf6, Nfkb1, Rela, Fos, Fosb, Jun, Jund, Fosl1, Il4, Il5, Il13, Ccl17, Ccl11, Il17rc, Traf3, Anapc5, Tnfaip3, Hsp90aa1, Hsp90ab1, Hsp90b1, Tbk1, Tab2, Tab3, Map3k7, Ikbkg, Chuk, Ikbkb, Nfkbia, Mapk14, Mapk11, Mapk12, Mapk13, Mapk8, Mapk9, Mapk10, Mapk1, Mapk3, Mapk4, Mapk6, Mapk7, Mapk15, Cebpb, Traf4, Ikbke, Usp25, Traf5, Traf2, Srsf1, Elavl1, Gsk3b, Cxcl1, Cxcl2, Cxcl3, Cxcl5, Cxcl10, Ccl2, Ccl12, Ccl7, Ccl20, Il6, Tnf, Ptgs2, Csf3, Csf2, Defb4, Muc5ac, Muc5b, S100a8, S100a9, Lcn2, Mmp1a, Mmp1b, Mmp3, Mmp9, Mmp13, Il17re, Il1b, and Ifng. However, Il17a, Il17b, Il17c, Il17d, Il17e, and Il17f were removed from the KEGG pathway.²⁹

The single-cell gene expression count and metadata tables containing clustering and quality control metrics for each cell are available at FigShare: https://figshare.com/s/128c57c0a7343b86ab7a for the filtered matrix and annotation, and https://figshare.com/s/8c1c1fd9ec93926e330c for complete cellranger files. All raw data were uploaded to the Sequence Read Archive (SRA) via the Gene Expression Omnibus (SRA BioProject, PRJNA722296; accession number, SRR14246980; reviewer link, https://dataview.ncbi.nlm.nih.gov/object/PRJNA722296?reviewer=dqi5ts48vlqit6tphit69 k2366).

Statistical Analysis

The results are shown as the mean±SEM when presented as a bar graph, or as single data points with the mean in a scatter dot plot. Differences between two individual experimental groups were compared using a two-tailed t test. P<0.05 was considered to be statistically significant.

Results

IL-17RA Deficiency Does Not Protect from Renal Injury in Experimental cGN

Several groups have recently demonstrated the importance of the T_H17/IL-17 pathway in driving tissue injury in experimental models of cGN and, potentially, in their human counterparts.^910–11 From a theoretic point of view, targeting or deleting IL-17RA is likely to lead to complete protection against T_H17/IL-17–driven tissue injury, because it will totally block the biologic function of all IL-17 cytokines. To test this hypothesis, we induced cGN (nephrotoxic nephritis) in IL-17RA^−/− and wild-type mice. Unexpectedly, the severity of the disease at day 8 with respect to glomerular crescent formation, tubulointerstitial injury (Figure 1, A and B), BUN level, and albuminuria (Figure 1C) was not reduced in nephritic IL-17RA^−/− mice, as compared with wild-type animals. Almost identical results were obtained when mice were analyzed 20 days after the induction of cGN (Supplemental Figure 1).

Figure 1.: IL-17RA deficiency does not protect from renal injury in experimental cGN but promotes T _H 17 immune responses. (A) Representative photographs of PAS-stained kidney sections from control, nephritic wild-type, and nephritic IL-17RA^−/− mice at day 8 after induction of cGN. Original magnification, ×400. (B) Quantification of glomerular crescent formation and tubulointerstitial damage in control, nephritic wild-type, and nephritic IL-17RA^−/− mice. (C) BUN levels and albumin-creatinine ratio. (D) Representative FACS plots of intracellular cytokine staining for IL-17A in renal T cells (gated on singlets, live, CD45⁺, and CD3⁺ cells). (E) Quantification of intracellular cytokine FACS analysis for renal and splenic CD4⁺ T cells for IL-17A and (F) for IL-17F, IL-22, and IFNγ in CD4⁺ T cells in the kidney. Symbols represent individual data points with the mean as a bar. **P<0.01, ****P<0.0001.

IL-17RA/IL-17A Signaling Controls the CD4⁺ T_H17 Response in Health and Disease

To understand the immunologic mechanisms that led to this unanticipated clinical outcome in nephritic IL-17RA^−/− mice, we investigated the renal and systemic immune responses in these animals. Flow cytometry studies showed a significantly enhanced renal and systemic (spleen) T_H17 response, whereas the T_H1 cell response in the kidney and the spleen was not affected (Figure 1, D–F). Even under noninflammatory conditions, a significantly boosted CD4⁺ T_H17 response was detectable in the absence of IL-17RA signaling (Supplemental Figure 2A). In contrast, non-T_H17–associated cytokines, such as IFNγ, did not show significant differences. Almost identical results were obtained when analyzing gut or skin CD4⁺ T_H17 cells under homeostatic or inflammatory conditions, such as IMQ-induced, psoriasis-like inflammation¹⁹ and Citrobacter rodentium–induced colitis,²² respectively (Supplemental Figure 2, B–G).

To study the humoral immune response in nephritic wild-type and IL-17RA^−/− mice, glomerular sheep and mouse IgG deposition was semiquantitatively scored, and serum titers of anti-sheep IgG, IgG1, and IgG2a antibodies were assessed by ELISA. As shown in Supplemental Figure 3, no differences between nephritic wild-type and knockout mice could be detected.

CD4⁺ T_H17 Cells Highly Express the IL-17RA/IL-17RC Complex

To test whether IL-17 cytokines might have direct effects on CD4⁺ T cells, we analyzed the IL-17 receptor expression pattern of CD4⁺ T cells. Hence, we induced cGN in IL-17A/IFNγ/Foxp3 triple-reporter mice¹⁶ for in vivo cell sorting of vital and unstimulated renal CD4⁺ T cell subsets at day 8. Subsequent mRNA expression analysis of sorted T cells revealed predominant expression of IL-17RC and IL-17RE by T_H17 cells (CD4⁺, IL-17A⁺, IFNγ⁻, Foxp3⁻) and a high level of IL-17RB expression by Foxp3⁺ regulatory T cells (CD4⁺, IL-17A⁻, IFNγ⁻, Foxp3⁺), whereas IL-17RA was ubiquitously expressed by all CD4⁺ T cell subsets (Figure 2A). Moreover, we sorted human CD4⁺ T_H1 cells (CXCR3⁺, CCR6⁻) and T_H17 cells (CXCR3⁻, CCR6⁺) from healthy kidney tissue from tumor nephrectomies and assessed the IL-17 receptor mRNA expression by RT-PCR. In line with the murine results, human T_H17 cells demonstrated higher IL-17RC expression levels (Figure 2B). The analysis of CD4⁺ T cells from the gut and skin of mice after induction of colitis or psoriasis, respectively (Supplemental Figure 4, A–C), showed identical results, demonstrating a similar CD4⁺ T cell subset–specific IL-17 receptor expression pattern in different species and tissues.

Figure 2.: IL-17 receptor expression and activation on CD4 ⁺ T _H 17 cells. (A) IL-17 receptor mRNA expression pattern of FACS-sorted renal CD4⁺ T cells from nephritic triple IL-17A/INFγ/FoxP3 reporter mice. (B) IL-17RA and IL-17RC expression of sorted human renal CCR6⁻, CXCR3⁺ CD4⁺ T_H1 cells and CCR6⁺, CXCR3⁻ CD4⁺ T_H17 cells. (C) Systemic IL-17A overexpression was induced by injection of an IL-17A expression plasmid. Controls received an empty plasmid. (D) Serum IL-17A levels were measured by electrochemiluminescence immunoassay analysis, and renal T_H17 responses were analyzed by FACS at day 8 after cGN induction. Data are presented as bar graphs with the mean±SEM. *P<0.05, ***P<0.001. eGFP, enhanced green fluorescent protein; mRFP, monomeric red fluorescent protein.

Systemic IL-17A Overexpression Reduces the Renal CD4⁺ T_H17 Response in cGN

Having shown that global disruption of IL-17 receptor signaling boosted renal T_H17 response in GN, we next asked whether the systemic activation of IL-17 receptor signaling could have the opposite effect. Therefore, we induced overexpression of IL-17A in the liver by hydrodynamic gene transfer using minicircle vectors encoding IL-17A 2 days before induction of experimental GN (Figure 2C). Mice treated with the IL-17A vector exhibited increased IL-17A plasma levels (Figure 2D) and significantly lower frequencies of CD4⁺ T_H17 cells in the inflamed kidney at day 8 (Figure 2D).

IL-17 Receptor Pathways Are Active in Renal T_H17 Cells

We next examined whether the specific IL-17RA/IL-17RC expression profile observed in CD4⁺ T_H17 cells resulted in IL-17 signaling activation in these cells. Therefore, we took advantage of IL-17A fate reporter mice (Il17a^Cre×R26^eYFP), in which T cells that had produced IL-17A constitutively express eYFP. This allowed the sorting of active T_H17 and also ex-T_H17 cells from the nephritic kidney. We then performed single-cell RNA sequencing of these FACS-sorted renal eYFP⁺ cells (Figure 3A). As expected, eYFP⁺ CD4⁺ emerged as a heterogeneous population (Figure 3B), consisting of IL-17A⁺ cells and IL-17A⁻ ex-T_H17 cells, reflecting the high plasticity of these cells. Further clustering revealed three clusters with high IL-17A expression levels (C1, C5, and C6), a cluster with Foxp3-expressing regulatory T cells (C3), and also some cells that display a T_H1-like phenotype (C8) (Figure 3, B–D). To get insight regarding which cells were potentially influenced by IL-17 signaling, we calculated a score on the basis of the genes annotated in the KEGG IL17-signaling pathway (mmu04657).²⁹ We removed all IL-17 cytokines from this pathway analysis to include only expression of intracellular downstream signaling genes. IL-17A–producing cell clusters had a significantly higher IL-17 signaling score than the ex-T_H17 cluster (Figure 3E). To analyze which genes of the IL-17 signaling score were highly expressed in IL-17–expressing cells, we calculated the DE genes in cluster C1 and annotated the log fold change to the pathway (Figure 3F). These data show the activation of IL-17 receptor signaling in IL-17A–producing CD4⁺ T_H17 cells.

Figure 3.: Single-cell RNA-sequencing analysis of sorted renal CD4 ⁺ T _H 17 cells show intracellular IL-17 signaling pathway activation. (A) Experimental setup and sorting strategy of fate-mapped renal CD4⁺ T_H17 cells from the cGN kidney. (B) Dimensional reduction by uniform manifold approximation and projection (UMAP) and unsupervised clustering of renal T_H17 cells. (C) IL-17A and marker gene expression in the different clusters of renal T_H17 cells. (D) Expression of indicated genes in the UMAP space. (E) UMAP plot indicating which cells have activated intracellular IL-17 signaling (left) and comparison of IL-17^high clusters (C1, C5, C6) versus IL-17^low clusters (C0, C2, C3, C4, C7, C8). (F) Expression of genes annotated to intracellular IL-17 signaling. ****P<0.0001. scRNA-seq, single-cell RNA sequencing; YFP, yellow fluorescent protein.

IL-17 Receptor Signaling Deficiency in CD4⁺ T Cells Stimulates IL-17 Cytokine Expression

To directly investigate the role of IL-17 receptor signaling in T cells, we sorted CD4⁺ T cells from the spleens of CD45.1 wild-type and CD45.2 IL-17RA^−/− mice and transferred them into Rag1^−/− mice (competitive adoptive transfer, 1:1 ratio), which are deficient in both T and B cells (Figure 4A). Next, we induced experimental GN and analyzed the origin of renal T_H17 cells at day 8 using the congenic markers CD45.1 and CD45.2. We found that IL-17RA deficiency on CD4⁺ T cells resulted in a significantly increased abundance of renal CD4⁺ T_H17 cell as compared with the transferred wild-type cells (Figure 4B).

Figure 4.: IL-17RA signaling in CD4 ⁺ T _H 17 cells limits the pathogenic T _H 17 immune response in cGN. (A) Experimental setup and (B) representative FACS plot and quantification of renal IL-17A⁺ T cells from nephritic Rag1^−/− mice repopulated with CD45.1 wild-type (WT) and CD45.2 IL-17RA^−/− CD4⁺ T cells. (C) Generation of IL-17A–Cre×IL-17RA^fl/fl mice. A scheme of the targeting strategy for the Il17ra allele, resulting in deletion of exon 3. FRT sites are represented by black lines and loxP sites by red lines. (D) Representative PCR analysis of IL-17RA exon 3 in CCR6⁺ CD4⁺ T_H17 FACS-sorted cells from kidneys of IL-17A–Cre×IL-17RA^wt/wt and IL-17A–Cre×IL-17RA^fl/fl mice. (E) Representative photographs of PAS-stained kidney sections and (F) quantification of glomerular crescent formation and tubulointerstitial damage of nephritic IL-17A–Cre×IL-17RA^wt/wt and IL-17A–Cre×IL-17RA^fl/fl mice. (G) BUN and albumin-creatinine ratio determined in the aforementioned groups. (H) Quantification of FACS-analyzed renal IL-17A⁺ T cells. Symbols represent individual data points with the mean as a horizontal line. *P<0.05, **P<0.01, ****P<0.0001.

IL-17RA Signaling in T_H17 Cells Regulates the T_H17 Immune Response in cGN

We then aimed to specifically study the functional role of IL-17RA in CD4⁺ T_H17 cells. To that end, we generated IL-17RA^flox/flox mice and crossed them to IL-17A Cre animals (Figure 4, C and D). IL-17A Cre IL-17RA^fl/fl mice were viable, fertile, and showed no obvious developmental abnormalities compared with IL-17A Cre IL-17RA^wt/wt mice (data not shown). Subsequently, we tested whether the course of cGN was influenced by ablation of IL-17RA in CD4⁺ T_H17 cells. We observed that the glomerular damage was aggravated in nephritic IL-17A Cre IL-17RA^fl/fl mice (Figure 4, E and F). In addition, albuminuria was slightly, but not significantly, higher in these animals, whereas the BUN level was not affected (Figure 4G). Finally, disruption of IL-17 signaling in CD4⁺ T_H17 cells potentiated the production of IL-17A in the nephritic kidneys (Figure 4H). Thus, our findings provide the first evidence that IL-17RA signaling in CD4⁺ T_H17 cells controls T_H17 cell responses and immune-mediated kidney disease.

Activation of the IL-17RA/IL-17RC Complex in CD4⁺ T Cells Controls the T_H17 Cell Response

IL-17RC is the obligate coreceptor of IL-17RA for signaling induced by IL-17A and IL-17F. To analyze whether IL-17RC deficiency mimics the effects of IL-17RA deficiency observed in immune-mediated glomerular disease, we compared the clinical course of experimental cGN between wild-type and IL-17RC–deficient mice. Evaluation of PAS-stained kidney sections revealed an increased glomerular crescent formation in IL-17RC^−/− mice (Figure 5, A and B), whereas tubulointerstitial injury, the BUN level, and the albumin-creatinine ratio was comparable between the two groups (Figure 5, B and C). Flow-cytometric analysis showed that the CD4⁺ T_H17 response with respect to IL-17A, IL-17F, and IL-22 production was upregulated in the kidneys and spleens of nephritic IL-17RC^−/− mice, whereas the renal T_H1 response remained unchanged (Figure 5, D–F).

Figure 5.: Activation of the IL-17RC controls the T _H 17 response in cGN. (A) Representative photographs of PAS-stained kidney sections from wild-type and IL-17RC^−/− mice at day 8 after induction of cGN. Original magnification, ×400. (B) Quantification of glomerular crescent formation and tubulointerstitial damage. (C) BUN levels and albumin-creatinine ratio. (D) Representative FACS plots of intracellular cytokine staining for IL-17A in renal T cells (gated on singlets, live, CD45⁺, and CD3⁺cells). (E) Quantification of intracellular cytokine FACS analysis for renal and splenic CD4⁺ T cells for IL-17A and (F) for IL-17F, IL-22, and IFNγ in CD4⁺ T cells in the kidney. (G) Representative FACS plot and quantification of renal IL-17A⁺ T cells from nephritic Rag1^−/− mice repopulated with CD45.1 wild-type and CD45.2 IL-17RC^−/− CD4⁺ T cells. Symbols represent individual data points with the mean as a bar. *P<0.05, **P<0.01.

Next, we isolated CD4⁺ T cells from the spleens of CD45.1 wild-type and CD45.2 IL-17RC^−/− mice and transferred them into Rag1^−/− mice (competitive adoptive transfer, 1:1 ratio). Eight days after induction of cGN, we assessed the IL-17A production of the transferred cells in the inflamed kidneys. This analysis revealed that a higher fraction of renal T_H17 cells were IL-17RC–deficient T cells when compared with wild-type T cells (Figure 5G).

Disruption of IL-17RC Signaling in T_H17 Cells Promotes the T_H17 Immune Response and Subsequent Tissue Injury in cGN

To study the precise role of IL-17RC specifically on CD4⁺ T_H17 cells, we generated IL-17A Cre IL-17RC^flox/flox animals, by breeding IL-17RC^flox/flox with IL-17A Cre mice, and induced cGN. The comparison of the renal tissue damage in nephritic IL-17A Cre IL-17RC^flox/flox and IL-17A Cre IL-17RC^wt/wt animals 8 days after disease induction revealed a significantly aggravated course of GN in terms of glomerular crescent formation and tubulointerstitial injury in mice that lack IL-17RC signaling in T_H17 cells (Figure 6, A and B). In line with these structural alterations, the BUN level was increased in IL-17A Cre IL-17RC^flox/flox mice as compared with the IL-17A Cre IL-17RC^wt/wt group (Figure 6C). Renal T cell immunoprofiling by flow cytometry revealed that the frequencies of IL-17A–, IL-17F–, and IL-22–producing CD4⁺ T cells were increased in nephritic IL-17A Cre IL-17RC^flox/flox mice, whereas the T_H1 response was not affected (Figure 6D). Of note, the disruption of IL-17RC signaling in T_H17 cells resulted in more pronounced effects on the clinical course of the cGN compared with the targeted deletion of IL-17RA in T_H17 cells (Figure 4, E–G). Further tissue analysis revealed that the uncontrolled T_H17 response in IL-17RC T cell–deficient animals resulted in an increased renal expression of the neutrophil attractant chemokines CXCL1 and CXCL5 (Figure 6E), which was accompanied by the infiltration of pathogenic neutrophils (Figure 6F). Therefore, we provide a potential mechanistic link between the dysregulated negative feedback loop in T_H17 cells and aggravated renal tissue injury. At the later stage of GN, at day 20, the T_H17 immune response was less dominant compared with the earlier time point. Of note, animals with IL-17RC deficiency in IL-17A–producing T cells developed more severe GN with respect to glomerular crescent formation, tubulointerstitial injury, and neutrophil recruitment (Supplemental Figure 5).

Figure 6.: IL-17RC activation in CD4 ⁺ T _H 17 cells limits the pathogenic T _H 17 immune response in cGN. (A) Representative photographs of PAS-stained kidney sections and (B) quantification of glomerular crescent formation and tubulointerstitial damage of IL-17A–Cre×IL-17RC^wt/wt and IL-17A–Cre×IL-17RC^fl/fl mice 8 days after cGN induction. (C) BUN and albumin-creatinine ratio determined in the aforementioned groups. (D) Quantification of intracellular cytokine FACS analysis for renal CD4⁺ T cells for IL-17A, IL-17F, IL-22, and IFNγ. (E) Renal chemokine expression assessed by RT-PCR. (F) Representative photographs and quantification of renal GR-1⁺ neutrophil infiltration. (G) cGN was induced in IL-17A–Cre×IL-17RC^wt/wt and IL-17A–Cre×IL-17RC^fl/fl mice and at days −1, 2, and 5 of cGN induction; IL-17A was neutralized with mAb. (H) Representative PAS staining of renal tissue section of the respective groups 8 days after diseases induction. (I) Quantification of glomerular crescents, tubulointerstitial injury, and (J) BUN level and albuminuria in the two groups. Data are presented as individual data points with the mean as a horizontal line. *P<0.05, **P<0.01.

Aggravated Course of cGN in Mice Lacking IL-17RC Is Due to an Enhanced T_H17 Response

To test whether the aggravated course of cGN was a consequence of the enhanced T_H17 cell response in IL-17A Cre IL-17RC^flox/flox animals, we used an IL-17A neutralization approach. Therefore, cGN was induced in IL-17A Cre IL-17RC^wt/wt and IL-17A Cre IL-17RC^flox/flox animals in the presence of anti–IL-17A mAb (Figure 6G). Blockade of IL-17A was associated with reduced levels of glomerular crescent formation, as described previously.²⁴ Most importantly, neutralization of IL-17A abolished the difference in crescent formation and BUN levels between mice lacking IL-17RC signaling in T_H17 cells and the control group (Figure 6, H–J), indicating that the exacerbated cGN in these mice is a consequence of the overwhelming T_H17 response.

Direct IL-17RC Signaling in CD4⁺ T Cells Controls the Development of IMQ-Induced Psoriasis and T Cell Transfer Colitis

Finally, we tested the generalizability of our findings and analyzed the function of IL-17RC signaling in models of immune-mediated skin and gut disease. IMQ-induced skin inflammation, an IL-17–dependent mouse model for psoriasis,³⁰ was induced in IL-17A Cre IL-17RC^wt/wt and IL-17A Cre IL-17RC^flox/flox mice. Mice were monitored and scored daily for the development of psoriasis-like skin inflammation in terms of ear desquamation, erythema, and thickness (Figure 7A). In addition, we assessed a total psoriasis score, which represents the mean of the sum of the mentioned clinical scores (Figure 7B). As shown in Figure 7, A and B, in the representative phenotypic presentation of mouse ears (Figure 7C), and hematoxylin and eosin staining of ear sections (Figure 7D), the disruption of IL-17RC signaling in T_H17 cells significantly amplifies the development of psoriasis-like skin lesions and the recruitment of neutrophils to the skin (Figure 7E). Of note, in the presence of a neutralizing anti–IL-17A antibody, these differences were no longer detectable (Figure 7, F and G), suggesting the aggravated skin inflammation in mice lacking IL-17RC in T_H17 cells is a consequence of an uncontrolled IL-17A production.

Figure 7.: IL-17RC activation in T cells limits tissue injury in experimental models of psoriasis and colitis. (A–E) IMQ-induced psoriasis was elicited in IL-17A–Cre×IL-17RC^wt/wt and IL-17A–Cre×IL-17RC^fl/fl mice (n=10 per group). (A) Desquamation, erythema, and ear thickness, and (B) the combined psoriasis score (zero to four) were determined in the course of the disease in all groups as indicated. (C) Representative photographs and (D) hematoxylin and eosin (HE) staining of ears of these mice, and (E) representative photographs and quantification of GR-1⁺ neutrophil infiltration. (F and G) Effect of the application of a neutralizing anti–IL-17A antibody on (F) desquamation, erythema, and ear thickness, and (G) the combined psoriasis score in the indicated groups (n=8–10 per group). (H–J) Experimental colitis was induced by the intravenous transfer of CD4⁺CD45RB^high T cells sorted from the spleen of IL-17A–Cre×IL-17RC^wt/wt and IL-17A–Cre×IL-17RC^fl/fl mice into Rag1 ^{-/ -} animals (n=5 per group). (H) Representative HE staining of the colon, (I) colonoscopy pictures, and (J) the total colitis score of these mice 4 weeks after the T cell transfer. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. wt, wild type.

In addition, we induced an experimental colitis model, by the intravenous transfer of CD4⁺CD45RB^high T cells sorted from the spleen of IL-17A Cre and IL-17A Cre IL-17RC^flox/flox mice into Rag1^-/- animals. The CD4⁺CD45RB^high T cell population is capable of inducing chronic intestinal inflammation, resembling aspects of human inflammatory bowel disease.²⁰ After 4 weeks, colonoscopy was performed in a blinded fashion and the disease severity was scored on the basis of the following parameters: granularity of mucosal surface, stool consistency, vascular pattern, translucency of the colon, and fibrin visibility (zero to three points for each), resulting in a total score between zero (healthy) and 15 (severe colitis).³¹ Representative hematoxylin and eosin staining of gut sections (Figure 7H), colonoscopy pictures (Figure 7I), and the total colitis score demonstrated an aggravated course of the disease in mice receiving IL-17RC^−/− T cells than the corresponding Rag1^-/- mice that received wild-type T cells (Figure 7J), indicating IL-17RC controls CD4⁺ T cells pathogenicity in experimental colitis.

Discussion

Here, we report that the IL-17RA/IL-17RC complex is expressed by T_H17 cells. Disruption of this IL-17 receptor signaling pathway in CD4⁺ T cells and, most importantly, in CD4⁺ T_H17 cells potentiates the T_H17 cell response and results in an accelerated course of experimental cGN. Our findings indicate that IL-17 receptor signaling in CD4⁺ T cells controls the T_H17 response via the IL-17RA/IL-17RC complex through a negative feedback loop.

The cytokines of the IL-17 family provide protection from invading bacterial and fungal pathogens. However, they also have a downside in that they significantly contribute to the organ damage that occurs in autoimmune and chronic inflammatory diseases. Basic and translational research and clinical trials in this field have substantially advanced our understanding of autoimmunity and introduced the T_H17/IL-17 axis as a new target for the treatment of psoriasis, psoriatic arthritis, ankylosing spondylitis, and, potentially, other autoimmune conditions.⁸

We and others recently demonstrated that the T_H17/IL-17 pathway plays a critical role in autoimmune kidney diseases, such as cGN. These studies included the identification and characterization of CCR6⁺ IL-17–producing T cells in murine kidneys in experimental models of cGN and in patients with ANCA-associated GN.^{19,22,23,32,33} Moreover, functional studies in models of proliferative GN and cGN presented direct evidence of the contribution of IL-17A, IL-17C, IL-17F, STAT3, IL-23p19, IL-23 receptor, and RORγt to kidney injury in this disease group.^{18,34353637383940–41} However, it is still unclear which of the different members of the T_H17/IL-17 axis might represent the most effective therapeutic target in immune-mediated (kidney) diseases. Because targeting or neutralization of the IL-17RA is likely to lead to complete blocking of the biologic effects of all IL-17 cytokines, this mechanism might represent an attractive therapeutic approach.

To test if IL-17 receptor blockade is beneficial in autoimmune kidney disease, we induced a well-characterized mouse model of cGN in IL-17RA–deficient mice. This model is induced by injection of nephrotoxic sheep serum directed against the glomerular basement membrane. This provoked a T_H17 (and a T_H1) response directed against the planted antigen, which resulted in the T_H17/IL-17A–driven formation of glomerular crescents, tubulointerstitial damage, and loss of renal function which resembles several aspects of human cGN.^4243–44 Unexpectedly, analyses 8 and 20 days after disease induction showed that IL-17RA deficiency was not protective against kidney damage in experimental cGN. This finding is in some contrast to the study by Ramani et al.⁴⁵ that showed that renal pathology was significantly decreased in IL-17RA^−/− mice 14 days after GN induction. In a more recent study, Ghali et al.⁴⁶ analyzed five clinical/histologic parameters (glomerular crescents, segmental necrosis, interstitial injury, serum urea, and proteinuria) in IL-17RA–deficient and wild-type animals 21 days after GN induction. Remarkably, only glomerular crescent formation was reduced in IL-17RA–deficient GN animals, whereas proteinuria was increased.⁴⁶ Intriguingly, analysis of the renal and systemic immune responses in IL-17RA^−/− mice revealed that CD4⁺ T cells displayed a markedly potentiated production of IL-17A and IL-17F, and of the T_H17-associated proinflammatory cytokines IL-22, TNFα, and GM-CSF.^4748–49 This suggests that disruption of IL-17 signaling in T cells may augment the “pathogenicity” of T_H17 cells and thereby reverse the clinical benefit that might result from loss of IL-17 receptor activation in renal target cells. In addition, a recent publication indicates that IL-17A and IL-17F also form complexes with IL-17RC homodimers, suggesting the possibility of IL-17RA–independent IL-17 signaling pathways.⁵⁰ Taken together, this might explain why the lack of IL-17RA is not as effective as expected in protecting against kidney damage in cGN.

It was previously reported that the complete absence of IL-17 signaling in IL-17RA^−/− mice resulted in an augmented IL-17 cytokine response. This finding was explained by ligand-dependent regulation of T_H17 cells and IL-17R–dependent clearance of the cytokines.^51,52 More recently, however, a study showed that the in vivo t_1/2 of injected recombinant IL-17A was almost the same in the serum of both wild-type and IL-17RA^−/− mice, indicating augmented IL-17 responses in these mice were not primarily a consequence of reduced receptor-mediated clearance of ligands.⁵³ Furthermore, the authors demonstrated that IL-17RA signaling in enteric epithelial cells is critical for the defense against segmented filamentous bacteria (SFB), which promotes CD4⁺ T_H17 cell development in the gut.⁵³ These results indicated that enteric IL-17R signaling controlled SFB growth in the gut and, thereby, constrained intestinal T_H17 cell development. Accordingly, we observed higher SFB colonization in the gut of IL-17RA–deficient mice (data not shown). Yet, these findings do not sufficiently explain the expansion of T_H17 cells at extraintestinal sites and the potentially higher susceptibility to autoimmune inflammation.

Our results indicate that the expression and activation of the IL-17RA/IL-17RC complex on T_H17 CD4⁺ T cells intrinsically limits T_H17 cell pathogenicity (in terms of cytokine production). Indeed, our CD4⁺ T cell transfer experiments into Rag1^−/− mice and, most importantly, the conditional deletion of IL-17RA and IL-17RC in T_H17 cells provide direct evidence for this previously unknown regulatory pathway. In particular, the disruption of IL-17RC signaling in T_H17 cells resulted in a markedly (IL-17A–driven) aggravated tissue inflammation. In contrast, the clinical course of the GN in IL-17RC global knockouts was comparable with that of wild-type animals. These profound differences, despite an augmented T_H17 immune response in both groups, is a consequence of the different tissue response to the proinflammatory cytokines IL-17A and IL-17F. In mice with a deficiency in IL-17RC in T_H17 cells, all other cells, such as renal epithelial and endothelial cells, are able to respond to these cytokines. This resulted in augmented renal expression of the chemokines CXCL1 and CXCL5, which was followed by the infiltration of pathogenic neutrophils,⁵⁴ providing a potential mechanistic link between the negative IL-17RC feedback loop in T_H17 cells and aggravated renal tissue injury (Figure 8). In contrast, IL-17A and IL-17F cannot exert proinflammatory effects in global IL-17RC–deficient animals. The observed effects of IL-17RC deletion were more pronounced than the deletion of IL-17RA, most likely as a consequence of the disruption of the IL-17RA/IL-17RE complex (activated by IL-17C), which affects different signaling pathways and might have opposite effects on T_H17 cell function.^18,55

Figure 8.: IL-17RC signaling limits CD4 ⁺ T _H 17 pathogenicity in experimental cGN via a negative feedback loop. This diagram illustrates the potential role of the pathway in renal inflammation. The IL-17RA/IL-17RC complex is highly expressed in renal CD4⁺ T_H17 cells and the receptor signaling pathway is activated in these cells in experimental cGN. Disruption of the IL-17RC signaling pathway potentiates the production of IL-17A and IL-17F, and other T_H17-associated cytokines, such as IL-22 and GM-CSF (not shown), in CD4⁺ T_H17 cells. These proinflammatory cytokines induce the expression and production of chemokines, such as CXCL1 and CXCL5, in the inflamed kidney. This is accompanied by the recruitment of pathogenic CXCR2-expressing neutrophils, ultimately leading to an exacerbation of experimental cGN, and thus provide a potential mechanistic link between the dysregulated IL-17RC feedback loop in T_H17 cells and aggravated renal tissue injury.

The molecular mechanisms that connect the IL-17RA/IL-17RC signaling pathway in CD4⁺ T_H17 cells with reduced pathogenicity remains to be fully elucidated. In preliminary experiments, however, we were able to show that activation of the IL-17A/IL-17RC pathway on CD4⁺ T cells resulted in increased expression of the transcription factor SOCS3 (data not shown). This is of interest, because SOCS3 is a known negative regulator of the T_H17 pathway,⁵ and it has been reported that IL-17 signaling in natural killer cells suppresses their antitumor and antiviral activity via upregulation of SOCS3.⁵⁶ Moreover, during the preparation of the manuscript, a sophisticated study reported that, in a model of autoimmune uveitis, T_H17 cells were negatively regulated by their own signature cytokine IL-17A.⁵⁷In vitro studies revealed a T_H17 cell–intrinsic autocrine loop triggered by binding of IL-17A to its receptor, leading to an expression of IL-24 via activation of the transcription factor NF-κB and the induction of SOCS3, which repressed the T_H17 cytokine program.⁵⁷ However, our data derived from the conditional deletion of IL-17RA and, most strikingly, of IL-17RC in T_H17 cells go beyond their findings and demonstrated the in vivo function of IL-17 receptor signaling for the control of the pathogenicity of the T_H17 response on the cellular level for the first time in immune-mediated kidney, skin, and gut diseases. This highlights the general importance of our finding.

In conclusion, our study shows that IL-17RC signaling in CD4⁺ T cells constrains the T_H17 immune response. This indicates that disruption of IL-17 signaling boosts the “pathogenicity” of T_H17 cells by the enhanced production of IL-17A, IL-17F, IL-22, and GM-CSF, which might reverse the clinical benefits that results from loss of IL-17 receptor activation in target cells. As a consequence, we observed an aggravated course of experimental GN in mice lacking IL-17RC signaling in CD4⁺ T cells. This discovery might have implications for the pathogenesis of immune-mediated diseases and may explain the sometimes unexpected clinical effects of IL-17 receptor targeting in experimental and human autoimmune diseases.^58,59

Disclosures

R.A. Flavell reports serving as an advisor to GlaxoSmithKline and Zai Lab. N. Gagliani reports receiving honoraria from Novartis and research funding from Roche. S. Huber reports having consultancy agreements with Abbvie, Falk, Ferring, and Janssen Cilag; and having other interests in/relationships with American Association of Immunology, DACED, Deutsche Gesellschaft für Innere Medizin, and Deutsche Gesellschaft für Verdauungs-und Stoffwechselkrankheiten. T.B. Huber reports receiving research funding from Amicus Therapeutics and Fresenius Medical Care; having consultancy agreements with, and receiving honoraria from, AstraZeneca, Bayer, Boehringer-Ingelheim, DaVita, Deerfield, Fresenius Medical Care, Goldfinch Bio, Mantrabio, Novartis, and Retrophin; and serving as a scientific advisor for, or member of, Kidney International (on the journal editorial board) and Nature Reviews Nephrology (on the journal advisory board). All remaining authors have nothing to disclose.

Funding

This study was supported by Deutsche Forschungsgemeinschaft grant SFB 1192 (to U. Panzer and C.F. Krebs).