MHC class II in renal tubules plays an essential role in renal fibrosis

Abstract

Immunomodulation is considered a potential therapeutic approach for chronic kidney disease (CKD). Although it has been previously reported that CD4⁺ T cells contribute to the development of renal fibrosis, the role of MHC class II (MHCII) in the development of renal fibrosis remains largely unknown. The present study reports that the expression of MHCII molecules in renal cortical tubules is upregulated in mouse renal fibrosis models generated by unilateral ureter obstruction (UUO) and folic acid (FA). Proximal tubule epithelial cells (PTECs) are functional antigen-presenting cells that promote the proliferation of CD4⁺ T cells in an MHCII-dependent manner. PTECs from mice with renal fibrosis had a stronger ability to induce T cell proliferation and cytokine production than control cells. Global or renal tubule-specific ablation of H2-Ab1 significantly alleviated renal fibrosis following UUO or FA treatment. Renal expression of profibrotic genes showed a consistent reduction in H2-Ab1 gene-deficient mouse lines. Moreover, there was a marked increase in renal tissue CD4⁺ T cells after UUO or FA treatment and a significant decrease following renal tubule-specific ablation of H2-Ab1. Furthermore, renal tubule-specific H2-Ab1 gene knockout mice exhibited higher proportions of regulatory T cells (Tregs) and lower proportions of Th2 cells in the UUO- or FA-treated kidneys. Finally, Immunohistochemistry (IHC) studies showed increased renal expression of MHCII and the profibrotic gene α smooth muscle actin (α-SMA) in CKD patients. Together, our human and mouse data demonstrate that renal tubular MHCII plays an important role in the pathogenesis of renal fibrosis.

Introduction

Chronic kidney disease (CKD) is currently a severe public health issue. Its prevalence is estimated at 10.4% in men and 11.8% in women of the worldwide general population [1]. CKD is a pathological condition related to progressive renal fibrosis and functional parenchymal loss, subsequently leading to end-stage renal disease [2, 3]. Renal fibrosis is a common characteristic of CKD and involves glomerular sclerosis and interstitial fibrosis [4]. To date, the precise mechanisms underlying renal fibrosis are still unclear. Currently, AngII-converting enzyme inhibitors and/or AngII type 1 receptor blockers (ARBs) represent the mainstays of treatment of CKD but have limited clinical efficacy [5]. Therefore, the mechanisms involved in CKD warrant further investigation, and new therapies are urgently needed.

A large body of evidence demonstrates that reactive oxygen species (ROS) and advanced glycation end-products (AGEs) and pathological states such as hypertension, hyperglycemia, hypoxia, and proteinuria are involved in the pathogenesis of renal fibrosis [6]. Recently, many studies have identified the immune system as an important contributor to the pathogenesis of renal fibrosis. Multiple immunological cell types, including dendritic cells (DCs), macrophages and T lymphocytes, are activated in injured renal tissues, suggesting that immune dysregulation may contribute to the development of CKD [6–8]. Major histocompatibility complex class II (MHCII) molecules are highly expressed in antigen-presenting cells (APCs) such as DCs, macrophages, endothelial cells, thymic epithelial cells, and B cells. The physiological function of MHCII is presenting peptide antigens to CD4⁺ T cells to subsequently allow activation of the adaptive immune response. MHCII molecules are expressed in cultured renal tubular cells, where their expression is markedly amplified by lymphokine interferon-gamma (IFN-γ) [9, 10]. The function of MHCII in renal tubular cells is unknown. Although it has been previously reported that CD4⁺ T cells contribute to the development of renal fibrosis after UUO [11], the role of MHCII in renal fibrosis remains largely uncharacterized.

In the present study, we report that MHCII expression is present in renal cortical tubules and upregulated in the kidneys of mice with UUO or FA. Mice with global or renal tubule-specific MHCII deficiency exhibited a marked attenuation of renal fibrosis and a significant reduction in profibrotic gene expression in the kidneys. The underlying mechanism may be associated with higher proportions of regulatory T cells (Tregs) and lower proportions of Th2 cells in the kidneys of mice with renal tubule-specific MHCII gene knockout. Research on human subjects also showed increased renal expression of MHCII and α-SMA in CKD patients. Therefore, our findings demonstrate a critical role of renal tubular MHCII in the pathogenesis of renal fibrosis and suggest that MHCII may act as a potential therapeutic target for the treatment of CKD.

Materials and methods

Animal studies

C57BL/6J (JAX#000664), B6.H2-Ab1^−/−(JAX#003584), B6.H2-Ab1^flox/flox (JAX#013181), B6.Ksp-cre (JAX#012237), and OT-II transgenic mice (JAX#004194) were purchased from the Jackson Laboratory. B6.H2-Ab1^flox/flox mice were crossed with B6.Ksp-cre mice to create kidney-specific H2-Ab1 KO mice (kH2-Ab1^−/−). Male WT, H2-Ab1^−/− and kH2-Ab1^−/− mice (8–10 weeks old) were utilized and appropriated by the Animal Experimentation Committee of Shenzhen University. UUO- and FA-induced mouse models of renal fibrosis were induced as described previously [12, 13]. Briefly, the left ureter of the mice was ligated under general anesthesia. Sham-operated mice were used as controls. Fibrosis was induced by a single dose, 250 mg/kg, of FA by intraperitoneal injection (F7876; Sigma, USA) in 0.3 M sodium bicarbonate (vehicle). Mice injected with vehicle were used as the controls. After 7 days of UUO or 28 days of FA, the mice were sacrificed, and their kidneys were removed for examination by light microscopy, immunohistochemical analysis, real-time PCR, and Western blotting analyses.

Renal morphology and function

Kidney samples were fixed in 4% paraformaldehyde overnight at 4 °C, dehydrated in an ascending ethanol series, embedded in paraffin and sectioned at 4 μm. Kidney sections were stained with Masson trichrome reagent. The plasma concentration of urea nitrogen was assayed with an ELISA quantitation kit (EIABUN; Invitrogen, USA), and the plasma creatinine concentration was measured with an ELISA quantitation kit (DICT-500; BioAssay Systems, USA) according to the manufacturer’s instructions.

Immunohistochemical analysis

Kidneys were fixed in 4% paraformaldehyde overnight at 4 °C. After the samples were blocked for 20 min with blocking solution (1% BSA), the sections at 4 μm were incubated with primary antibodies against H2-Ab1 (ab180779; Abcam, UK), α-SMA (ab5694; Abcam, UK), and CD4 (ab183685; Abcam, UK) overnight at 4 °C. Images were obtained using light microscopy (Olympus BX-50; Olympus Optical, Tokyo, Japan). Twenty views per slide were randomly selected and measured with a digital camera at ×200 magnification.

Quantitative real-time PCR

Total RNA was extracted from the kidneys. Two micrograms of each RNA sample was reverse-transcribed, and the resulting cDNA samples were then used as templates for quantitative PCR. Real-time PCR analysis involved the use of SYBR Green (Invitrogen, USA) and primers from Applied Biosystems according to the manufacturer’s instructions. PCRs were carried out at 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 60 °C for 30 s and 72 °C for 30 s with a final extension at 72 °C for 5 min. Peptidylprolyl isomerase A was used as an internal control.

Western blot analysis

Renal tissues were lysed in lysis buffer containing 50 mM Tris (pH 7.5), 1% Nonidet P-40, 0.1% sodium dodecyl sulfate, 150 mM NaCl, 0.5% sodium deoxycholate, PhosSTOP (Roche), and protease inhibitors (Roche). Protein samples (40 μg) were transferred to nitrocellulose membranes. The membrane was washed and blocked in PBST (1 × PBS with 0.02% Tween-20) supplemented with 5% skim milk powder for 1 h at room temperature with gentle shaking and then incubated overnight (4 °C) with the following primary antibodies: anti-H2-Ab1 antibody (ab63567; Abcam, UK), anti-fibronectin antibody (ab2413; Abcam, UK), anti-α-SMA antibody (ab5694; Abcam, UK), or anti-collagen IV (PA1-28534; Invitrogen, USA). The membrane was washed three times for 30 min in PBST and then incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. After being washed three times, the membrane was then transferred to ECL Reagent (Bio–Rad). Images were obtained by a protein analysis system (GE Healthcare).

Preparation of spleen and renal single-cell suspensions for flow cytometry

Spleen and renal single-cell suspensions were prepared as described previously [14, 15]. Briefly, mice were subjected to perfusion with saline to remove intravascular leukocytes. Spleens were harvested, and single-cell suspensions were prepared by processing the spleen with a 40 μm mesh. Kidneys were minced into fragments of 1 mm³ and digested in a mixture of 1 mg/ml collagenase IV (Worthington, USA) and 100 μg/ml DNase I (Roche, USA) in DMEM at 37 °C for 36 min. The digested tissues were then passed sequentially through 100 μm and 40 μm meshes. The cell suspension was centrifuged, and red blood cells in the resulting pellet were lysed using red blood cell lysis buffer (Sigma). Cells were incubated with Near-IR fluorescent reactive dye (Invitrogen) at room temperature for 15 min, incubated with CD16/32 antibody (eBioscience, 14-0161-85, 1:200 dilution) for another 15 min and stained with detection antibodies for 30 min at room temperature. Fluorochrome-conjugated antibodies directed against the following mouse antigens were used for analysis by flow cytometry: BioLegend: CD45 (Cat#103128, 1:200 dilution), CD3 (Cat#100214, 1:200 dilution), CD8a (Cat#100712, 1:200 dilution)), IFN-γ (Cat#505810, 1:100 dilution), IL-4 (Cat#504118, 1:100 dilution), and IL-17A (Cat#506904, 1:100 dilution). Invitrogen: CD4 (Cat#11-0042-81, 1:200 dilution) and FOXP3 (Cat#17-5773-80, 1:200 dilution). PMA/ionomycin (Cat#2030421, 100X, Dakewe, China) and monensin (Cat#420701, 1000X, BioLegend) were added to the cell culture media and incubated for 4 h for cellular cytokine induction. After staining with cell-surface antibodies such as CD45, CD3, CD4, and CD8, the membranes of cells were fixed and permeabilized, and then, the cells were stained with intracellular cytokine antibodies such as anti-IFN-γ, anti-IL-4, or anti-IL-17A. FOXP3 staining was performed using a staining kit (Cat#00-5523-00, Invitrogen) according to the manufacturer’s protocol. Analyses were performed on a BD LSRII Flow Cytometer (BD Biosciences) and analyzed with FlowJo software (Treestar).

Proximal tubule epithelial cell (PTEC)-T cell coculture and antigen-specific T cell activation assays

Cultures of primary mouse PTECs were performed as previously described [16]. PTECs from the kidneys of mice with UUO were isolated and plated at 5 × 10³ cells per well in 96-well U bottom plates. PTECs were cultured in medium in the presence of 10 μg/ml anti-MHCII (Cat#M5/114, Bioxcell) or absence (IgG) of MHCII-neutralizing antibodies. PTECs were pulsed with medium alone or OVA peptide (OVA_323-339, Abbiotec) at 10 μg/ml for 2 h at 37 °C. CD4⁺ T cells were isolated from the spleens of OT-II mice using the MojoSort^™ Mouse CD4 Naïve T Cell Isolation Kit (Cat#480040, BioLegend). CD4⁺ T cells were labeled with 2 μM carboxyfluorescein succinimidyl ester (CFSE) (CellTrace kits, Molecular Probes). PTECs were washed and resuspended in 100 μl of DMEM/F-12 medium. A total of 5 × 10⁴ CFSE-labeled CD4⁺ T cells per well were added to antigen-pulsed PTECs at a 1:1 ratio in 200 μl of media. After 3 days, T cells were stained for flow cytometry, and CFSE dilution was examined in CD3⁺ CD4⁺ lymphocytes [17]. T-cell cytokines in the media were measured by ELISAs.

Human kidney samples

Human kidney tissue samples were obtained from six patients who underwent nephrectomy due to malignancies as a control and eight patients who underwent renal biopsy for CKD and were confirmed to have IgA nephropathy or diabetic nephropathy in the Department of Nephrology of the First Affiliated Hospital of Shenzhen University between 2015 and 2019. Experienced pathologists evaluated the samples for the presence of renal fibrosis. The study involving human samples was approved by the Ethics Committee for the Clinical Investigation of Shenzhen University and complied with all relevant ethical regulations and the guidelines of the Declaration of Helsinki. Prior to inclusion in the study, all patients provided informed consent for the collection and use of their kidney tissues for research.

Statistical analysis

Data are shown as the mean ± SEM using GraphPad Prism software (GraphPad Software Inc., CA, USA). Analysis involved ANOVA and Student’s t test. One-way ANOVA was applied followed by a Bonferroni test to compare the data relating to baseline characteristics and histological analysis. The results were considered significant at two-tailed p < 0.05.

Results

The expression of MHCII genes is upregulated in renal cortical tubules following UUO or FA treatment

Identification of pivotal immunological processes involved in the initiation and progression of renal fibrosis was completed through data mining of a GEO database (accession# GSE79443) in NCBI [18]. Bioinformatics analysis showed that 19 of 38 enhanced pathways shared multiple MHC genes (Supplementary Table 1, marked in red). Among them, eight genes are involved in MHCII antigen presentation. Importantly, the expression of Class II transactivator (Ciita), which regulates γ-interferon-activated transcription of MHCII genes, was also increased following UUO (Supplementary Table 2). Real-time PCR analysis confirmed that the mRNA levels of six MHCII genes (H2-Ab1, H2-Eb1, CD74, H2-DMa, H2-DMb1, H2-DMb2) and Ciita were increased 5–8-fold in the kidneys after UUO (Supplementary Fig. 1).

To further elucidate the expression of MHCII in the kidney, we determined the intrarenal localization of the MHCII molecule H2-Ab1 in the mouse kidney. H2-Ab1 mRNA expression was high in the renal cortex, followed by the outer medulla (OM) and inner medulla (IM) (Fig. 1A). Similarly, H2-Ab1 protein levels were greater in the renal cortex than in the OM and IM (Fig. 1B). Immunohistochemistry (IHC) studies demonstrated that the H2-Ab1 protein is primarily localized in cortical renal tubules with low expression in the glomeruli and the medulla (Fig. 1C). By using renal tubule-specific markers, we further found that H2-Ab1 was expressed in almost all segments of renal tubules, with the highest expression in the proximal tubules (Supplementary Fig. 2A). H2-Ab1 was also expressed in thick ascending limbs (TALs) of Henle’s loop, distal convoluted tubules and medullary collecting ducts (Supplementary Fig. 2B–D). The results from immunofluorescence staining also confirmed that H2-Ab1 was expressed in almost all segments of renal tubules (Supplementary Fig. 3A–D). These results demonstrate that H2-Ab1 is constitutively expressed in renal cortical tubules and suggest that MHCII may play an important role in immunological regulation in the kidney. Supporting this notion, in a classic renal fibrosis model, H2-Ab1 mRNA and protein levels were markedly elevated in the kidneys of the UUO-treated mice (Fig. 1D, E). Induced H2-Ab1 expression was mainly restricted to renal cortical tubules (Fig. 1F). Moreover, we used another murine renal fibrosis model: folic acid (FA)-induced nephropathy. The H2-Ab1 mRNA and protein levels were significantly increased in the kidneys of the FA-treated mice (Fig. 1G, H). Induced H2-Ab1 expression was mainly expressed in renal cortical tubules (Fig. 1I). Collectively, our findings suggest that MHCII expression is upregulated in renal cortical tubules during renal fibrosis and may play a critical role in the pathogenesis of renal fibrosis in mice.

Fig. 1.

Renal expression of H2-Ab1 is induced in the UUO and FA models. A Real-time PCR assays showing H2-Ab1 mRNA levels primarily present in the renal cortex, with lower levels in the outer medulla (OM) and inner medulla (IM). Data are presented as the mean ± SEM. **P < 0.01 vs. cortex, n = 4–5. B Western blot assays showing H2-Ab1 protein levels primarily present in the renal cortex, with lower levels in the OM and IM. Representative Western blots are presented. C Intrarenal localization of the H2-Ab1 protein in the renal cortex and medulla as assessed by immunohistochemistry. H2-Ab1 immunoreactivity was observed in the renal cortex and medulla. H2-Ab1 signals (arrowheads) were mainly located in the cortical tubules. Representative immunostaining for H2-Ab1: (1) Cortex-anti-H2-Ab1; (2) Medulla-anti-H2-Ab1; (3) Cortex-isotype-IgG; (4) Medulla-isotype-IgG (bar = 100 μm). D Real-time PCR assays showing that H2-Ab1 mRNA expression was increased in the obstructed kidney after 7 days of UUO. **P < 0.01 vs. the control, n = 5–6. E Western blotting assay showing that renal H2-Ab1 protein expression was upregulated in the obstructed kidney. Representative Western blots are presented. F Representative immunohistochemical staining for increased H2-Ab1 expression in the obstructed kidney (bar = 50 μm). **P < 0.01 vs. the control, n = 5–6. G Real-time PCR assay showing that H2-Ab1 mRNA expression was increased in the fibrotic kidney after 28 days of FA treatment. **P < 0.01 vs. the control, n = 5–6. H Western blotting assay showing that renal H2-Ab1 protein expression was upregulated in the fibrotic kidney. Representative Western blots are presented. I Representative immunohistochemical staining for increased H2-Ab1 expression in the fibrotic kidney (bar = 50 μm). Data are the mean ± SEM. **P < 0.01 vs. the control, n = 5–6

Renal PTECs are functional antigen-presenting cells that induce antigen-specific CD4⁺ T cell proliferation and cytokine production

Although renal PTECs express MHCII molecules, it remains generally unclear whether PTECs are functional APCs. Thus, we measured the capacity of renal PTECs to trigger antigen-specific T-cell proliferation. We cocultured OVA-stimulated PTECs from the control or UUO-induced mice with CD4⁺ T cells from OVA-specific OT-II transgenic mice. UUO-induced renal fibrosis significantly enhanced the ability of PTECs to induce T cell proliferation (Fig. 2A, B). This phenomenon could be blocked by MHCII-neutralizing antibodies, suggesting that this effect was MHCII-dependent (Fig. 2A, B). Moreover, T-cell cytokines in the cocultures were measured by ELISAs. Renal PTECs from the UUO-induced mice promoted the accumulation of IL-2 in the media, which was attenuated by MHCII-neutralizing antibodies (Fig. 2C). Overall, these data indicated that PTECs can act as functional APCs and that PTECs from mice with renal fibrosis may have an enhanced antigen presentation ability.

Fig. 2.

Renal proximal tubule epithelial cells (PTECs) trigger MHCII-restricted antigen-dependent CD4⁺ T cell proliferation and cytokine production. PTECs from the kidneys of the mice with UUO in the presence (anti-MHCII) or absence (IgG) of MHCII-neutralizing antibodies were cultured and grown in 96-well U bottom plates for 3 days. Adherent cells were pulsed with OVA peptide (OVA_323-339) for 2 h before CFSE-labeled CD4⁺ T cells from OT-II mice were added. After 3 days, CFSE dilution was examined by flow cytometry to quantify proliferating CD4⁺ T cells in cocultures. A Representative histograms demonstrate antigen-induced proliferation of CFSE-labeled OT-II CD4⁺ T cells in cocultures containing OVA peptide-pulsed PTECs from the kidneys of the mice with UUO in the presence or absence of MHCII-neutralizing antibodies. B Quantitation of OT-II CD4⁺ T cell proliferation in cocultures. C IL-2 production in cocultures containing OVA peptide-stimulated renal PTECs from the UUO-induced mice with or without MHCII-neutralizing antibodies. Data are the mean ± SEM. ^*P < 0.05, ^**P < 0.01 vs.^. the control; ^#P < 0.05, ^##P < 0.01 vs. the UUO group, n = 3–4

Global H2-Ab1 gene deletion attenuated UUO- or FA-induced renal fibrosis

Since H2-Ab1-deficient C57BL/6 mice cannot form functional MHCII complexes to activate CD4⁺ T cells [19, 20], global H2-Ab1 gene knockout mice (H2-Ab1^−/−) were utilized to determine the role of MHCII in renal fibrosis. As expected, no H2-Ab1 expression was observed in the kidneys of the H2-Ab1^−/− mice (Fig. 3D). UUO-induced extracellular matrix deposition in the tubulointerstitial areas was significantly attenuated in the H2-Ab1^−/− mice, with profibrotic gene and α-SMA expression markedly reduced (Fig. 3A–F), indicating that MHCII contributes to the pathogenesis of renal fibrosis following UUO.

Fig. 3.

H2-Ab1 gene deficiency attenuates UUO-induced renal fibrosis in mice. A Masson trichrome staining demonstrating intensive fibrosis in tubulointerstitial areas (arrowheads) in mice with both genotypes after UUO. However, the H2-Ab1^−/− mice exhibited much less renal fibrosis than the wild-type (WT) mice. Representative images: (1) WT; (2) WT + UUO; (3) H2-Ab1^−/−; (4) H2-Ab1^−/−+UUO (bar = 100 μm). B Semiquantitative analysis of (A) showing significantly increased collagen staining in the kidneys of the WT mice with UUO, which was markedly attenuated in the H2-Ab1^−/− mice. C Real-time PCR assay indicating that the mRNA expression of collagen I, collagen IV, fibronectin, Actα and TGFβ was markedly increased in the kidneys of the UUO-treated WT mice and was significantly attenuated in the H2-Ab1^−/− mice. D Western blotting assay showing that H2-Ab1^−/− deficiency resulted in a marked decrease in UUO-induced protein expression of fibronectin, α-SMA and collagen IV. E Representative immunostaining for α-SMA by immunohistochemistry: (1) WT; (2) WT + UUO; (3) H2-Ab1^−/−; (4) H2-Ab1^−/−+UUO (bar = 50 μm). F Semiquantification of α-SMA protein immunoreactivity. α-SMA protein expression was significantly lower in the H2-Ab1^−/− mice than in the WT mice after UUO. Data are the mean ± SEM. ^*P < 0.05, ^**P < 0.01 vs. WT; ^#P < 0.05 vs. WT + UUO, n = 6–8

In tandem, another mouse renal fibrosis model, FA-induced nephropathy, was utilized to confirm the role of MHCII in renal fibrosis. Similar to that in the UUO-induced mice, extracellular matrix deposition in the tubulointerstitial areas was also markedly attenuated in the FA-treated H2-Ab1^−/− mice, with profibrotic gene and α-SMA expression significantly decreased (Supplementary Fig. 4A–E). In addition, the levels of plasma blood urea nitrogen (BUN) and creatinine in the H2-Ab1^−/− mice were markedly lower than those in the WT mice after FA treatment (Supplementary Fig. 4F). Both the WT mice and the H2-Ab1^−/− mice showed significant proteinuria following FA treatment, and more severe proteinuria was observed in the WT mice than in the H2-Ab1^−/− mice (Supplementary Fig. 4G), indicating that ablation of H2-Ab1 significantly improved renal function in the FA-induced mice. These findings further indicate that MHCII contributes to the pathogenesis of renal fibrosis.

Renal tubule-specific H2-Ab1 deficiency ameliorated renal fibrosis following UUO or FA treatment

To further determine the specific role of renal tubule MHCII in the pathogenesis of UUO-induced renal fibrosis, we generated renal tubule-specific MHCII knockout mice (kH2-Ab1^−/−) by crossing a H2-Ab1^flox/flox mouse with a KSP-cre mouse (Supplementary Fig. 5A, B). The KSP-cre transgenic mice expressed Cre recombinase under the control of the mouse cadherin 16 promoter in renal tubules, especially collecting ducts, loops of Henle and distal tubules [21]. As expected, the expression of renal tubule H2-Ab1 was markedly reduced in the kidneys of the kH2-Ab1^−/− mice (Supplementary Fig. 5C, D). Moreover, to test whether the kH2-Ab1^−/− mice showed changes only in tubule epithelial cells, we isolated renal tubules and glomeruli from the WT and kH2-Ab1^−/− mice to measure the mRNA levels of H2-Ab1 by real-time PCR. As shown in Supplementary Fig. 5E, there was a dramatic decrease in the H2-Ab1 mRNA levels in the renal tubules of the kH2-Ab1^−/− mice compared to the WT mice, but no change in the H2-Ab1 mRNA levels was observed in the glomeruli of the WT and kH2-Ab1^−/− mice. The results from immunofluorescence staining also showed that the H2-Ab1 expression levels were significantly reduced in almost all segments of renal tubules of the kH2-Ab1^−/− mice compared to the WT mice (Supplementary Fig. 3A–H). These data indicated that the mice with renal tubule-specific MHCII knockout were successfully generated. Masson trichrome staining displayed substantial fibrosis in the tubulointerstitial areas of the WT mice after UUO that was markedly attenuated in the kH2-Ab1^−/− mice (Fig. 4A, B). Consistently, renal expression of profibrotic genes, including TGFβ, collagen I, collagen IV and fibronectin, was markedly decreased in the kH2-Ab1^−/− mice compared with the WT mice following UUO (Fig. 4C). In addition, the expression of α-SMA, a myofibroblast marker, was markedly reduced in the kidneys of the kH2-Ab1^−/− mice (Fig. 4C–F). Altogether, these results demonstrate that renal tubule-specific H2-Ab1 deficiency attenuated renal fibrosis in the UUO model, supporting a critical role of renal MHCII in UUO-induced renal fibrosis.

Fig. 4.

Renal tubule-specific H2-Ab1 gene deficiency attenuates UUO-induced renal fibrosis in mice. A Masson trichrome staining indicates substantial fibrosis (arrowheads) in the tubulointerstitial areas of the wild-type (WT) mice after UUO. However, renal histological changes were markedly attenuated in the kH2-Ab1^−/− mice following UUO. Representative images: (1) WT; (2) WT + UUO; (3) kH2-Ab1^−/−; (4) kH2-Ab1^−/−+UUO (bar = 100 μm). B Semiquantitative analysis of (A) showing significantly increased collagen staining in the kidneys of the WT mice with UUO, which was markedly reduced in the kH2-Ab1^−/− mice. C Real-time PCR assay demonstrating that the mRNA expression of collagen I, collagen IV, fibronectin, Actα and TGFβ was markedly increased in the kidneys of the UUO-treated WT mice and was significantly attenuated in the kH2-Ab1^−/− mice. D Western blotting assays showed that the kH2-Ab1^−/− mice had markedly decreased UUO-induced protein expression of fibronectin, a-SMA, and collagen IV. E Representative immunostaining for α-SMA by immunohistochemistry: (1) WT; (2) WT + UUO; (3) kH2-Ab1^−/−; (4) kH2-Ab1^−/−+UUO (bar = 50 μm). F Semiquantification of α-SMA protein immunoreactivity. α-SMA protein expression was markedly lower in the kH2-Ab1^−/− mice than in the WT mice after UUO. Data are the mean ± SEM. ^*P < 0.05, ^**P < 0.01 vs. WT; ^#P < 0.05 vs. WT + UUO, n = 6–8

Similar to that in the UUO-induced mice, FA-induced extracellular matrix deposition in the tubulointerstitial areas was also significantly attenuated in the kH2-Ab1^−/− mice (Supplementary Fig. 6A, B). Renal profibrotic genes and α-SMA expression were significantly decreased in the kH2-Ab1^−/− mice compared with the WT mice following FA (Supplementary Fig. 6C–E). In addition, the levels of plasma BUN and creatinine in the kH2-Ab1^−/− mice were markedly lower than those in the WT mice after FA treatment (Supplementary Fig. 6F). Renal tubule-specific ablation of H2-Ab1 markedly decreased the levels of proteinuria in the FA-treated mice compared to the WT mice (Supplementary Fig. 6G), indicating that renal tubule-specific H2-Ab1 deficiency markedly improved renal function in the FA-induced mice. Thus, these findings further showed that renal tubule-specific H2-Ab1 deficiency attenuated mouse renal fibrosis, supporting an important role of renal MHCII in renal fibrosis.

Renal tubule-specific H2-Ab1 deficiency resulted in higher proportions of Tregs and lower proportions of Th2 cells in the UUO- or FA-treated mouse kidneys

Since MHCII presents peptide antigens to CD4⁺ T cells to subsequently allow activation of the adaptive immune response, CD4⁺ T cell populations in kidneys of different groups were measured by flow cytometry. Renal cells were first gated for cell populations (FSC-A vs. SSC-A), single cells (FSC-A vs. FSC-H), live cells, and leukocytes (CD45⁺), which were then gated for T cells (CD3⁺) and CD4⁺ T cells (Supplementary Fig. 7). The percentages of renal CD4⁺ T cells were markedly increased in the UUO-treated mice and were significantly decreased in the kH2-Ab1^−/− mice compared with the WT mice following UUO (Fig. 5A), suggesting that renal CD4⁺ T cells may contribute to the pathogenesis of renal fibrosis. To investigate which subtypes of renal CD4⁺ T cells are associated with renal fibrosis, we analyzed the alteration of renal CD4⁺ T cell subtypes in mouse renal fibrosis models. Isotype controls for flow cytometry were used to distinguish the different CD4⁺ T cell subtypes (Supplementary Fig. 8A–D). Our data showed that the percentages of Tregs, FOXP3-positive CD4⁺ T cells (Fig. 5B), Th1 cells, IFN-r-positive CD4⁺ T cells (Fig. 5C), Th2 cells, IL-4-positive CD4⁺ T cells (Fig. 5D), Th17 cells, and IL-17A-positive CD4⁺ T cells (Fig. 5E) were markedly increased in the UUO-treated kidneys. Then, we found that mice with renal tubule-specific H2-Ab1 gene knockout exhibited higher proportions of Tregs (Fig. 5B) and lower proportions of Th2 cells (Fig. 5D) in the UUO-treated kidneys. Moreover, there were no obvious changes in renal Th1 (Fig. 5C) and Th17 cell percentages (Fig. 5E) between the WT and kH2-Ab1^−/− mice following UUO. Immunohistochemistry (IHC) studies further confirmed that CD4 expression in the kidney was markedly increased by UUO treatment and was significantly decreased with renal tubule-specific ablation of H2-Ab1 (Fig. 5F). Real-time PCR data also showed that the mRNA expression of FOXP3 (a marker gene for Treg cells), T-bet (a marker gene for Th1 cells), Gata-3 (a marker gene for Th2 cells), and RORγt (a marker gene for Th17 cells) was markedly increased in the kidneys of the UUO-treated WT mice. The mRNA expression of FOXP3 was further upregulated, but the mRNA expression of Gata-3 was significantly decreased in the kH2-Ab1^−/− mice after UUO (Fig. 5G). We also measured the percentages of renal CD8⁺ T cells in different groups. There were no changes in renal CD8⁺ T cell percentages (Supplementary Fig. 9A) between the WT and kH2-Ab1^−/− mice following UUO. In addition, the percentages of CD4⁺ T cells were markedly increased in the spleens of the WT mice with UUO; however, there were no obvious changes in the spleens of the kH2-Ab1^−/− mice with UUO (Supplementary Fig. 10A), suggesting that CD4⁺ T cells in the kidney rather than in the spleen may contribute to the pathogenesis of renal fibrosis. There were no changes in CD8⁺ T cell percentages among the different groups (Supplementary Fig. 10B).

Fig. 5.

Analysis of CD4⁺ T cells and their subtypes in the kidneys of the WT and kH2-Ab1⁻/⁻ mice after UUO. A FACS analysis of changes in CD4⁺ T cells in the kidneys of the WT and kH2-Ab1^−/− mice after UUO. The percentages of renal CD4⁺ T cells in the WT mice were markedly increased following UUO and were significantly decreased in the kH2-Ab1^−/− mice. B The percentages of Tregs, FOXP3-positive CD4⁺ T cells, were markedly upregulated in the UUO-treated WT mouse kidneys and were further elevated in the kH2-Ab1^−/− mice. C The percentages of renal Th1 cells and IFN-r-positive CD4⁺ T cells were increased in the WT mice with UUO; however, there were no obvious changes in the kH2-Ab1^−/− mice with UUO. D The percentages of renal Th2 cells and IL-4-positive CD4⁺ T cells in the WT mice were markedly increased following UUO and were significantly decreased in the kH2-Ab1^−/− mice. E The percentages of renal Th17 cells and IL-17A-positive CD4⁺ T cells were increased in the WT mice with UUO; however, no obvious changes were observed in the kH2-Ab1^−/− mice with UUO. F Representative immunostaining for CD4 by immunohistochemistry: (1) WT; (2) WT + UUO; (3) kH2-Ab1^−/−; (4) kH2-Ab1^−/−+UUO (bar = 50 μm). A marked increase in CD4 expression in the kidneys of the mice after UUO treatment was significantly decreased with renal tubule-specific ablation of H2-Ab1. G Real-time PCR assay demonstrating that the mRNA expression of FOXP3, T-bet, Gata-3, and RORγt was markedly increased in the kidneys of the UUO-treated WT mice. The mRNA expression of FOXP3 was further upregulated in the kH2-Ab1^−/− mice after UUO, but the mRNA expression of Gata-3 was significantly decreased in the kH2-Ab1^−/− mice after UUO. Data are the mean ± SEM. ^*P < 0.05, ^**P < 0.01 vs. WT; ^#P < 0.05 vs. WT + UUO, n = 6–8

Similar to that of the UUO-induced mice, the percentages of renal CD4⁺ T cells were markedly increased in the FA-treated mice and were significantly decreased in the kH2-Ab1^−/− mice compared with the WT mice following FA treatment (Supplementary Fig. 11A). The percentages of Tregs, FOXP3-positive CD4⁺ T cells (Supplementary Fig. 11B), Th1 cells, IFN-r-positive CD4⁺ T cells (Supplementary Fig. 11C), Th2 cells, IL-4-positive CD4⁺ T cells (Supplementary Fig. 11D), Th17 cells, and IL-17A-positive CD4⁺ T cells (Supplementary Fig. 11E) were markedly increased in the FA-treated kidneys. Mice with renal tubule-specific H2-Ab1 gene knockout exhibited higher proportions of Tregs (Supplementary Fig. 11B) and lower proportions of Th2 cells (Supplementary Fig. 11D) in the FA-treated kidneys. Moreover, there were no obvious changes in renal Th1 (Supplementary Fig. 11C) and Th17 cell percentages (Supplementary Fig. 11E) between the WT and kH2-Ab1^−/− mice following FA treatment. Moreover, the results from IHC studies showed that renal CD4 expression was significantly upregulated following FA treatment and was markedly reduced with renal tubule-specific ablation of H2-Ab1 (Supplementary Fig. 11F). Real-time PCR data also demonstrated that the mRNA expression of FOXP3, T-bet, Gata-3, and RORγt was markedly increased in the kidneys of the FA-treated WT mice. The mRNA expression of FOXP3 was markedly upregulated, while the mRNA expression of Gata-3 was notably decreased in the kH2-Ab1^−/− mice after FA treatment (Supplementary Fig. 11G). Similarly, there were no changes in renal CD8⁺ T cell percentages (Supplementary Fig. 9B) between the WT and kH2-Ab1^−/− mice following FA. The percentages of CD4⁺ T cells were significantly increased in the spleens of the WT mice with FA; however, there were no obvious changes in the spleens of the kH2-Ab1^−/− mice with FA (Supplementary Fig. 10C), indicating that CD4⁺ T cells in the kidney other than in the spleen may contribute to the pathogenesis of renal fibrosis. There were no changes in the CD8⁺ T cell percentages among the different groups (Supplementary Fig. 10D).

Increased renal expression of MHCII and the profibrotic gene α-SMA in CKD patients

To determine whether MHCII expression is associated with the human kidney in CKD, we determined the renal expression of MHCII and the profibrotic gene α-SMA in the kidneys of CKD patients. Masson trichrome staining showed substantial fibrosis in the tubulointerstitial areas of the CKD patients (Fig. 6A-B). Immunohistochemistry (IHC) studies showed that MHCII protein levels were markedly elevated in the kidneys of the CKD patients (Fig. 6A, C), and induced MHCII expression was mainly restricted to renal cortical tubules (Fig. 6A). Moreover, renal expression of α-SMA, a profibrotic marker, was significantly induced in renal cortical tubules of the CKD patients (Fig. 6A, D). These clinical data demonstrated that renal tubular MHCII plays an important role in the pathogenesis of renal fibrosis.

Fig. 6.

Increased renal expression of MHCII and the profibrotic gene α-SMA in CKD patients. A Masson trichrome staining demonstrating intensive fibrosis and immunohistochemistry staining showing marked increases in MHCII and α-SMA immunoreactivities in tubulointerstitial areas (arrowheads) in the CKD patients. Representative images: (1) Masson staining in healthy kidneys; (2) Masson staining in CKD patient kidneys; (3) MHCII staining in healthy kidneys; (4) MHCII staining in CKD patient kidneys; (5) α-SMA staining in healthy kidneys; (6) α-SMA staining in CKD patient kidneys (bar = 100 μm). B Semiquantitative analysis of (A) showing significantly increased extracellular matrix protein deposition in the kidneys of the CKD patients. C Semiquantification of MHCII protein immunoreactivity. MHCII protein expression was significantly elevated in the kidneys of the CKD patients. D Semiquantification of α-SMA protein immunoreactivity. α-SMA protein expression was markedly increased in the kidneys of the CKD patients. Data are the mean ± SEM. ^**P < 0.01 vs. the controls, N = 6–8

Discussion

Chronic kidney disease (CKD) has become a major public health concern [22]. The 2016 revision of the Global Burden of Disease (GBD) Study reported that 753 million patients suffer from CKD worldwide [23]. Increasing evidence has demonstrated that immune dysfunction may play a pivotal role in the pathogenesis of CKD [6–8]. In this study, we showed that the MHCII molecule is constitutively expressed in renal cortical tubules. Its expression was significantly upregulated in the kidneys of mice with UUO or FA treatment. Global or renal tubule-specific deletion of H2-Ab1 markedly attenuated UUO- or FA-induced renal fibrosis with reduced profibrotic gene expression. The underlying mechanism may be associated with higher proportions of Tregs and lower proportions of Th2 cells in the kidneys of the mice with renal tubule-specific MHCII gene knockout. The clinical data also showed increased renal expression of MHCII in CKD patients. These findings demonstrate that MHCII may play an important role in renal immune regulation and contribute to the pathogenesis of renal fibrosis.

The presentation of peptide antigens to CD4⁺ T cells by MHCII molecules is critical for the initiation of an effective immune response. An early study in a UUO murine model provided preliminary evidence that CD4⁺ T cells promote interstitial fibrosis in the obstructed kidney via a mechanism that allows myofibroblast accumulation and matrix deposition, suggesting that CD4⁺ T cells may contribute to the pathogenesis of renal fibrosis [11]. CD4⁺ T cells, also known as T helper cells (Th cells), have a few functionally distinct subsets, including Th1, Th2, Th17 and regulatory T cells (Tregs). In general, Th2 cells contribute to profibrotic outcomes by secreting IL-4 and IL-13, whereas Th1 cells exert an antifibrotic effect via the production of IFN-γ [24–26]. Since reconstitution of mice with Th2 cells results in more severe renal fibrosis than that with Th1 cells, Th2 cells are thought to contribute to the development of renal fibrosis [27]. Th17 cells are characterized by the production of IL-17. To date, the role of IL-17A in UUO-induced renal fibrosis remains controversial. It has been reported that IL-17-deficient mice exhibit reduced renal fibrosis via the suppression of RANTES-mediated inflammatory cell infiltration [28]. Despite the effects of IL-17 in other inflammation models, neither IL-17A deficiency nor treatment with an IL-17A blocking antibody attenuated renal ischemia reperfusion injury and progression to CKD [29]. Tregs are also involved in renal fibrosis, since their abundance was markedly increased in UUO-induced mouse kidneys [30, 31]. Recently, single-cell RNA sequencing revealed a protective role for Tregs during kidney injury and fibrosis. Tregs are not only sufficient to protect the kidney from initial damage but also to prevent the development of late fibrosis [15]. Therefore, the subtypes of CD4⁺ T cells may play distinct roles in renal immune regulation and renal fibrosis. Our data indicated that the percentages of renal CD4⁺ T cells were markedly increased in the UUO- or FA-treated mice, with a significant decrease in the kH2-Ab1^−/− mice following UUO or FA, suggesting that MHCII plays an important role in the regulation of CD4⁺ T cells in the kidney during renal fibrosis. Furthermore, alterations in renal CD4⁺ T cell subtypes were measured. We found that mice with renal tubule-specific H2-Ab1 gene knockout exhibited expanded proportions of Tregs and decreased proportions of Th2 cells in the UUO- or FA-treated kidneys, with no obvious changes in renal Th1 and Th17 cell percentages, which suggested that expanded proportions of renal Tregs and decreased proportions of renal Th2 cells may account for attenuated renal fibrosis in the kH2-Ab1^−/− mice following UUO or FA. Increasing evidence has shown that Tregs induce tissue protection. Further studies are needed to confirm that the increased renal Tregs are responsible for the reduced renal fibrosis in the kH2-Ab1^−/− mice. Moreover, our data showed that there were no changes in renal CD8⁺ T cell percentages between the WT and kH2-Ab1^−/− mice following UUO or FA, indicating that renal CD4⁺ T cells rather than CD8⁺ T cells may contribute to the pathogenesis of renal fibrosis. In addition to distinguishing the roles of CD4⁺ T cells in the kidney from those in the spleen, we measured the changes in splenic CD4⁺ T cells among the different groups. Although the percentages of CD4⁺ T cells were markedly increased in the spleens of the WT mice with UUO or FA, there were no obvious changes in the spleens of the kH2-Ab1^−/− mice with UUO or FA. Since the expression of renal tubule H2-Ab1 was markedly reduced in the kidneys of the kH2-Ab1^−/− mice with no changes in the spleen, the data further suggested that CD4⁺ T cells in the kidney but not the spleen may contribute to the pathogenesis of renal fibrosis.

In addition, the percentages of CD4⁺ T cells were markedly increased in the kidneys and spleens of the WT mice with UUO or FA and were significantly decreased in the kidneys and spleens of the H2-Ab1^−/− mice following UUO or FA (Supplementary Fig. 12A-D). As shown in Fig. 5A and Supplementary Fig. 11A, the percentages of renal CD4⁺ T cells were markedly increased in the UUO- or FA-treated WT mice and significantly decreased in the kH2-Ab1^−/− mice following UUO or FA. No obvious changes in the spleens of the kH2-Ab1^−/− mice with UUO or FA were observed compared with those of the WT mice following UUO or FA (Supplementary Fig. 10A, 10C). Thus, these results suggested a direct role of kidney MHCII in renal fibrosis. Notably, a dramatic decrease was observed in the number of CD4⁺ T cells in the thymus, spleen, and lymph nodes of the global H2-Ab1^−/− mice from the Jackson Laboratory (#003584) because of impaired development of CD4⁺ T cells in H2-Ab1-deficient mice [19]. As shown in Supplementary Fig. 12, there were very low proportions of CD4⁺ T cells in the spleen and kidney of the global H2-Ab1^−/− mice. This result indicated the important role of MHCII in sustaining CD4⁺ T cells.

Increasing evidence has shown that acute kidney injury (AKI) may contribute to the development and progression of CKD [32–34]. FA is a classic inducer in the AKI model, and transition to CKD is observed after 28 days of FA injection. CD4⁺ T cells promoted AKI [35]. However, the role of MHCII in AKI remains unclear. Renal damage, including renal tubular dilatation, cast formation and accumulation in the lumens, was clearly observed at 48 h after FA injection (Supplementary Fig. 13A). Renal tubule-specific ablation of H2-Ab1 markedly decreased the degree of renal tubular damage in the mice subjected to FA (Supplementary Fig. 13A, B). The levels of BUN and creatinine were significantly lower in the FA-induced kH2-Ab1^−/− mice than in the WT mice (Supplementary Fig. 13C, D), indicating less severe acute renal injury in the kH2-Ab1^−/− mice than in the WT mice. These data suggested that renal tubule-specific ablation of H2-Ab1 may improve FA-induced interstitial fibrosis by reducing FA-induced acute renal injury.

Since MHCII-mediated antigen presentation provides a key signal to activate CD4⁺ T cells, upregulation of MHCII expression in renal tubules may be a key mechanism of the dysregulation of CD4⁺ T cells in renal fibrosis. Professional APCs expressing MHCII molecules include DCs, macrophages, and B lymphocytes. Extensive evidence has identified several other cell types expressing MHCII that are capable of antigen presentation, which are regarded as atypical APCs [36]. Atypical APCs consist of mast cells, basophils, eosinophils, neutrophils, innate lymphoid cells, endothelial cells, and epithelial cells. Among epithelial cells, intestinal epithelial cells, airway epithelial cells and keratinocytes can act as APCs [36]. Antigen presentation by PTECs can induce CD4⁺ T cell activation, proliferation, and inflammatory cytokine production [9, 10]. Recently, it has been reported that mouse and human renal cortical epithelial cells, especially proximal tubule epithelial cells (PTECs), express MHCII [37]. However, it remains largely unclear whether PTECs are functional APCs and whether renal MHCII contributes to renal fibrosis. In the present study, we found that MHCII is expressed in renal cortical tubules, where its expression can be induced following UUO or FA treatment, and may play an essential role in the pathology of immune-mediated renal fibrosis. Moreover, UUO-induced renal fibrosis significantly enhanced the ability of PTECs to induce T cell proliferation, which was blocked by MHCII-neutralizing antibodies, suggesting that this effect was MHCII-dependent. Compared to the controls, PTECs from the UUO-induced mice expressed higher levels of H2-Ab1 expression (Supplementary Fig. 14A, B), indicating that upregulation of MHCII expression in PTECs from the UUO-induced mice may partially account for the elevated ability of PTECs to induce T cell proliferation and activation.

The pathological role of MHCII in renal fibrosis was supported by our findings that global deletion of the MHCII molecule H2-Ab1 significantly attenuated renal fibrosis in the mice with UUO or FA. More importantly, renal tubule-specific ablation of the H2-Ab1 gene also resulted in a marked reduction in tubulointerstitial fibrosis following UUO or FA. The beneficial effect of H2-Ab1 gene knockout is associated with a significant decrease in profibrotic gene expression and myofibroblast activation. Our findings indicate that MHCII-expressing renal cortical tubules may function as APCs and contribute to renal fibrosis in UUO- and FA-induced models.

In summary, this study provides evidence that H2-Ab1 expression in renal cortical tubules is notably induced in murine fibrotic kidney models. Global and renal tubule-specific deletion of H2-Ab1 markedly attenuated renal fibrosis following UUO or FA treatment. The underlying mechanism may be due to expanded proportions of Tregs and decreased proportions of Th2 cells in the kidney. In addition, clinical data confirmed the induction of renal expression of MHCII in CKD patients. Therefore, renal tubular MHCII plays a critical role in the pathogenesis of renal fibrosis and may represent an attractive therapeutic target for the treatment of CKD.

Author contributions

YZ, YG, and TD designed the study. YZ, ZL, and CL generated most of the data. RC, JW, YZ, TC, and JS generated some of the data. YZ, ZL, and CL analyzed the data. YZ drafted the paper. ZH, ZH, BL, and XZ performed the critical review. YZ, YG, and TD edited the paper. All authors approved the final version of the paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41423-021-00763-z.