TNF-α in T lymphocytes attenuates renal injury and fibrosis during nephrotoxic nephritis

Abstract

Nephrotoxic serum nephritis (NTN) models immune-mediated human glomerulonephritis and culminates in kidney inflammation and fibrosis, a process regulated by T lymphocytes. TNF-α is a key proinflammatory cytokine that contributes to diverse forms of renal injury. Therefore, we posited that TNF-α from T lymphocytes may contribute to NTN pathogenesis. Here, mice with T cell-specific deletion of TNF-α (TNF TKO) and wild-type (WT) control mice were subjected to the NTN model. At 14 days after NTN, kidney injury and fibrosis were increased in kidneys from TNF TKO mice compared with WT mice. PD1⁺CD4⁺ T cell numbers and mRNA levels of IL-17A were elevated in NTN kidneys of TNF TKO mice, suggesting that augmented local T helper 17 lymphocyte responses in the TNF TKO kidney may exaggerate renal injury and fibrosis. In turn, we found increased accumulation of neutrophils in TNF TKO kidneys during NTN. We conclude that TNF-α production in T lymphocytes mitigates NTN-induced kidney injury and fibrosis by inhibiting renal T helper 17 lymphocyte responses and infiltration of neutrophils.

INTRODUCTION

Nephrotoxic serum nephritis (NTN) produces severe kidney inflammation and injury, which is orchestrated by cells of the innate and adaptive immune systems (1). While NTN is widely used as a model of crescentic glomerulonephritis, renal pathology in NTN mice varies depending on the pathogenic antibody injected and can include glomerular cell proliferation and matrix deposition, tubular damage, and inflammatory cell infiltration. Later stages of NTN can manifest with interstitial fibrosis (29). While contributions of the immune system to glomerular pathology have been well documented, the role of inflammatory cells in the progression to tubulointerstitial fibrosis has received less scrutiny.

Human kidney biopsies reveal renal T cell accumulation at the initiation of NTN (4, 28, 35). Both CD4⁺ T cells and CD8⁺ conventional T cells contribute to the pathogenesis of experimental NTN, while regulatory T cells are protective (20). CD8⁺ T cells exacerbate podocyte injury via antigen-dependent cytotoxic effects (7). CD4⁺ T cells differentiate into activated subsets that secrete pathogenic, proinflammatory cytokines (3). T helper (Th)1 lymphocytes specific to antigens in the glomerular basement membrane (GBM) can induce severe glomerulonephritis (39). Th17 lymphocytes, another subset of CD4⁺ T effector cells, infiltrate the glomerulus and tubulointerstitium via a chemokine (C-C motif) receptor 6-dependent mechanism (36). Th17 lymphocytes then recruit neutrophils by stimulating the production of chemokine (C-X-C motif) ligand 5 in the kidney epithelium (10). In turn, transfer of GBM antigen-specific Th17 lymphocytes to lymphocyte-deficient recipients can induce marked albuminuria early during NTN (32). Nevertheless, mechanisms through which T cell subsets regulate the severity of NTN require elucidation.

TNF-α is a key proinflammatory cytokine produced by both Th1 and Th17 lymphocytes (8, 25, 33). We therefore posited that T cell-derived TNF-α contributes to renal inflammation and fibrogenesis during NTN. We tested this hypothesis using conditional mutant mice that harbor genetic deletion of TNF-α within infiltrating T cells. However, we were surprised to discover that T cell-derived TNF-α limits T lymphocyte-mediated glomerulosclerosis, tubular damage, interstitial fibrosis, and inflammation.

METHODS

Animals.

Floxed TNF mice (12) from the C57BL/6 background were backcrossed for at least six generations to the 129SVE strain and then intercrossed with 129SVE CD4-Cre mice to generate CD4⁻Cre⁺ TNF^flox/flox (TNF TKO) mice and wild-type (WT) Cre⁻ littermates for our experiments. Animals had free access to standard rodent chow and water. Eight- to twelve-week-old male mice were used for experiments. Genotyping methods have been previously described elsewhere (42). Mice were bred and maintained in Association for Assessment and Accreditation of Laboratory Animal Care-accredited animal facilities at the Durham Veterans Affairs Medical Center according to National Institutes of Health guidelines. All animal experiments were approved by the Durham Veterans Affairs Medical Center Institutional Animal Care and Use Committee and conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

NTN model.

Experimental mice were injected intraperitoneally with sheep IgG (250 μg/20 g) in 100 μL sterile PBS containing complete Freund’s adjuvant (100 μL/20 g) at day 0. At day 5, mice received a tail vein injection of sheep anti-mouse GBM serum (Probetex). Mice were euthanized after 14 days for blood, urine, and kidney tissue collection.

Assessment of renal function.

Serum blood urea nitrogen (BUN) levels were measured in individual samples with a urea nitrogen colorimetric detection kit (Invitrogen). Urinary concentrations of albumin and creatinine were measured in individual samples with an ELISA kit (Exocell).

Histological analyses.

Kidney tissues were fixed overnight with 10% neutral-buffered formalin and embedded in paraffin. Sections (5 μm) were stained with periodic acid-Schiff reagent at the Duke University Medical Center Department of Pathology. The percentage of glomerular matrix deposition was quantitated as the average matrix percentage for all glomeruli in each individual mouse, and these average values were used for analysis between groups. The tubular injury score was quantitated by evaluation of morphological changes, including tubular necrosis, dilation, cast formation, epithelial swelling, and vacuolar degeneration, and scored from 0−5 as follows: 0 = no morphological deformities, 1 ≤ 10%, 2 = 10–25%, 3 = 25–50%, 4 = 50–75%, and 5 ≥ 75%. More than 10 contiguous fields were examined in each sample, and averaged values were collected for comparison between the two groups of mice. Immunofluorescent staining was performed as previously described (42) with anti-sheep IgG-FITC antibody (catalog no. F5137, Sigma). Immunohistochemical staining with anti-collagen type I (COL-1) antibody was performed to visualize the deposition and location of interstitial collagen, as previously described (42). The observer was blinded to the experimental conditions.

Western blot analysis.

Kidney tissues were homogenized in RIPA buffer, and the concentration of protein was quantitated using the DC protein assay kit (Bio-Rad). Equal amounts of protein were subjected to electrophoresis with bis-Tris gels and then transferred to PVDF membranes. The blots were blocked and then incubated with anti-COL-1 antibody (SouthernBiotech), anti-fibronectin (FN) antibody (Abcam), anti-α-smooth muscle actin (α-SMA) antibody (Sigma-Aldrich), and anti-GAPDH antibody (catalog no. 2118, Cell Signaling Technology) overnight at 4°C. The blots were then washed and incubated with secondary antibodies for 1 h at room temperature. Target bands were detected using ECL solution and quantified by ImageJ analysis.

Real-time PCR analyses of mRNA expression.

Total mRNA was isolated with an RNeasy Mini Kit according to the manufacturer’s instructions (Qiagen). After mRNA quantitation, cDNA was synthesized using the High-Capacity cDNA Reverse Transcription kit (ThermoFisher). Gene expression levels were determined by a real-time PCR system (model 7900HT, Applied Biosystems) using TaqMan probes for Serpine1 [plasminogen activator inhibitor 1 (PAI-1)], Lcn2 [neutrophil gelatinase-associated lipocalin (NGAL)], COL-1a1, transforming growth factor (TGF)-β₁, FN, TNF-α, and SYBR green primers for IL-17A, IL-17F, IL-21, and IL-22.

Flow cytometry.

At 14 days after NTN, kidneys were harvested and minced into single cell suspensions, as previously described (41). The single cell suspensions were blocked and incubated with the following fluorescent antibodies: anti-CD11b-FITC (clone M1/70), anti-CD45-brilliant violet 510 (clone 30-F11), anti-lymphocyte antigen 6 complex (Ly6C)-phycoerythrin-cyanine 7 (clone HK1.4), anti-Ly6G-phycoerythrin-cyanine 5.5 (clone 1A8), anti-CD11c-phycoerythrin (clone N418), anti-major histocompatibility complex class II-brilliant violet 645 (clone M5/114), anti-CD24-phycoerythrin-cyanine 7 (clone M1/69), anti-CD64-allophycocyanin (clone X54-5/7.1), anti-CD3-brilliant violet 421 (clone145-2C11), anti-CD8-phycoerythrin-cyanine 5.5 (clone 53-6.7), anti-CD4-phosphatidylethanolamine (clone RM4-5), anti- TNF-α-allophycocyanin (clone MP6-XT22), anti-IL17A-allophycocyanin (clone eBio17B7), anti-Foxp3-phyceoerythrin (clone FJK-16s), and anti-PD1-FITC (clone J43) for 30 min at 4°C. During intracellular staining, single cell suspensions of splenocytes were stimulated with anti-CD3/anti-CD28 antibodies, treated with a protein transport inhibitor (Invitrogen), and then subjected to antibody staining and fixation. Before analysis, 20 μL of Count Bright absolute counting beads (catalog no. C36950, Invitrogen) were added to cells, and samples were analyzed on a flow cytometer (model LSRII, BD). Total cell numbers were obtained using the enumeration formula, as described in the manufacturer’s instructions. Total cell numbers per whole kidney are presented. Data were analyzed using FlowJo 10.2.

Statistical analyses.

Values are means ± SE. Comparisons between two groups were assessed using an unpaired Student’s t test. For nonnormally distributed values, a Mann-Whitney test was used. All statistical analyses were calculated using GraphPad Prism software.

RESULTS

Generation of mice with specific deletion of TNF-α in T lymphocytes.

To examine the functions of TNF-α in T lymphocytes during NTN, we bred CD4-Cre mice with a TNF flox line harboring loxP sites on either side of the coding region for TNF-α. During our experiments, we subjected CD4⁻Cre⁺ TNF^flox/flox (TNF TKO) mice and CD4⁻Cre⁻ TNF^flox/flox (WT) littermates to experimental NTN. First, to verify the deletion of TNF-α mRNA in T lymphocytes from TNF TKO mice, we stimulated single cell suspensions of splenocytes with antibodies to CD3/CD28 and measured intracellular TNF-α protein levels via flow cytometry. Compared with WT littermates, TNF-α production in CD4⁺ and CD8⁺ splenic T cells of TNF TKO mice was completely abrogated (Fig. 1A). At the mRNA level, splenic T cells from TNF TKO mice exhibited >95% deletion of TNF-α compared with WT control mice (Fig. 1B). Otherwise, WT and TNF TKO mice were phenotypically normal at baseline and had similar body weights (27.1 ± 0.6 and 26.9 ± 0.5 g, P = 0.86) and renal function as measured by BUN levels (17.1 ± 1.1 and 17.5 ± 1.3 mg/dl, P = 0.85).

Fig. 1.

Generation of mice with specific deletion of TNF-α in T lymphocytes (TNF TKO mice). A: single cell suspensions of splenocytes from wild-type (WT) and TNF TKO mice were stimulated and stained with anti-CD3/anti-CD28, and TNF-α protein levels in CD4⁺ and CD8⁺ T cells were determined by flow cytometry. SSC, side scatter. B: RT-PCR analysis of TNFα mRNA expression in stimulated CD4⁺ T lymphocytes. Values are means ± SE (n = 3 for each group). *P < 0.05.

T cell-derived TNF-α preserves renal function during NTN.

We next evaluated whether T cell-derived TNF-α impacts the severity of NTN. At 14 days after NTN induction, glomerular matrix deposition was greater in kidneys from TNF TKO mice compared with WT mice (Fig. 2, A and B), despite similar levels of sheep IgG deposition in the glomerulus (1.0 ± 0.1 and 1.1 ± 0.2 arbitrary units, P = 0.5; Fig. 2A). Similarly, BUN levels were significantly higher and urinary albumin-to-creatinine ratios were higher in TNF TKO mice than in WT mice at 14 days after NTN (Fig. 2, C and D).

Fig. 2.

T cell-derived TNF-α preserves renal function during nephrotoxic nephritis (NTN). A: at 14 days after NTN, kidney sections were collected from mice with specific deletion of TNF-α in T lymphocytes (TNF TKO mice) and wild-type (WT) control mice for periodic acid-Schiff (PAS) staining and immunofluorescent staining for sheep IgG. B: glomerular matrix deposition at 14 days after NTN in kidneys from TNF TKO and WT mice. C and D: blood urea nitrogen (BUN) and urinary albumin-to-creatinine (Cr) ratio at 14 days after NTN in kidneys from TNF TKO and WT mice. Values are means ± SE; n = 12 WT and 11 TNF TKO mice. *P < 0.05.

TNF-α in T cells limits tubular damage during NTN.

To examine the effects of T cell-derived TNF-α on renal tubular architecture during NTN, we computed renal tubular injury scores in the two groups. At 14 days after NTN, TNF TKO mice developed more severe tubular injury than WT mice (Fig. 3A). Similarly, mRNA levels of PAI-1 and NGAL were significantly increased in NTN kidneys from TNF TKO mice compared with WT mice (Fig. 3, B and C). These data indicate that TNF-α deficiency in T lymphocytes augments NTN-induced kidney damage.

Fig. 3.

TNF-α in T cells limits tubular damage during nephrotoxic nephritis (NTN). A: kidney sections from mice with specific deletion of TNF-α in T lymphocytes (TNF TKO mice) and wild-type (WT) control mice were scored for tubular injury on a scale of 1–5 (0 = no morphological deformities, 1 ≤ 10%, 2 = 10–25%, 3 = 25–50%, 4 = 50–75%, and 5 ≥ 75%) by an observer blinded to experimental groups. B and C: RT-PCR determination of RNA levels of plasminogen activator inhibitor 1 (PAI-1) and neutrophil gelatinase-associated lipocalin (NGAL) at 14 days after NTN. Values are means ± SE; n = 12 WT and 11 TNF TKO mice. *P < 0.05.

TNF-α deficiency in T lymphocytes exacerbates kidney fibrosis during NTN.

To evaluate whether TNF-α deficiency in T lymphocytes impacts fibrogenesis during NTN, immunohistochemistry staining for COL-1 was performed on kidney sections. Tubulointerstitial collagen accumulation was observed in the injured kidneys of both TNF TKO and WT mice (Fig. 4A). At 14 days after NTN, mRNA levels for COL-1, TGF-β, and FN were significantly augmented in kidneys from TNF TKO mice compared with WT mice (Fig. 4, B–D). To confirm these findings at the protein level, we performed Western blot analysis for matrix proteins in baseline control and injured kidneys (Fig. 4E). At 14 days after NTN induction, protein levels of COL-1 and α-SMA were significantly increased in kidneys from TNF TKO mice relative to WT mice (Fig. 4, F and G).

Fig. 4.

TNF-α deficiency in T lymphocytes exacerbates kidney fibrosis during nephrotoxic nephritis (NTN). A: at 14 days after NTN, kidney sections from mice with specific deletion of TNF-α in T lymphocytes (TNF TKO mice) and wild-type (WT) control mice were collected for collagen type I (COL-1) staining. B–D: quantitative PCR determination of renal mRNA levels for COL-1, transforming growth factor (TGF)-β, and fibronectin (FN) at 14 days after NTN. E: Western blot analysis of protein levels of fibrosis markers [COL-1 and α-smooth muscle actin (α-SMA)] in baseline control and NTN kidneys. F and G: quantitation of Western blot results (E) for COL-1 and α-SMA. Values are means ± SE; n = 12 WT and 11 TNF TKO mice. *P < 0.05.

PD1⁺CD4⁺ T lymphocyte infiltration in NTN kidneys.

As T cells play a key role in NTN pathogenesis, we posited that TNF-α deficiency may impact the differentiation of T cell subsets within the injured kidney. At 14 days after NTN, we quantitated the numbers of T cell subsets by flow cytometry (Fig. 5A). The numbers of both CD4⁺ and CD8⁺ T cells were increased in the kidneys of TNF TKO mice compared with WT mice (Fig. 5, B and C). Recently, PD1⁺CD4⁺ Th17 cells were found to be an important CD4⁺ T cell subset expressing the profibrotic cytokine TGF-β (5). At 14 days after NTN, the number of PD1⁺CD4⁺ T cells was higher in kidneys from TNF TKO mice compared with WT mice (Fig. 5D), suggesting that TNF-α deficiency in T lymphocytes may favor accumulation of these PD1⁺CD4⁺ T cells in the kidney.

Fig. 5.

PD1⁺CD4⁺ T lymphocyte infiltration in nephrotoxic nephritis (NTN) kidneys from mice with specific deletion of TNF-α in T lymphocytes (TNF TKO mice) and wild-type (WT) control mice. A: fluorescence flow cytometry of single cell suspensions from NTN kidneys from WT and TNF TKO mice. SSC, side scatter. B and C: numbers of CD4⁺CD8⁻ and CD4⁻CD8⁺ T cells (n = 12 WT and 11 TNF TKO mice). D: results from a separate flow experiment using NTN kidneys to determine numbers of PD1⁺CD4⁺ T cells (n = 5 WT and 5 TNF TKO mice). Values are means ± SE. *P < 0.05.

T cell-derived TNF-α constrains Th17 responses during NTN.

Given the critical role of Th17 lymphocytes in the pathogenesis of NTN, we measured renal mRNA levels for Th17 cytokines at 14 days after NTN. IL-17A mRNA levels were significantly upregulated in NTN kidneys from TNF TKO mice compared with WT mice (Fig. 6A). In contrast, mRNA levels for other Th17 cytokines, including IL-17F, IL-21, and IL-22, were similar in the two groups (Fig. 6, B–D). We further found increased infiltration of CD4⁺IL-17A⁺ T cells in nephritic kidneys from TNF TKO mice compared with WT mice (Fig. 6, E and F). However, the accumulation of CD4⁺Foxp3⁺ T cells in the NTN kidneys was similar between WT and TNF TKO mice (Fig. 6, G and H). Thus, TNF-α deficiency in T lymphocytes augments renal expression of IL-17A, a key driver of NTN pathogenesis.

Fig. 6.

T cell-derived TNF-α constrains T helper (Th)17 cell responses during nephrotoxic nephritis (NTN). A–D: at 14 days after NTN, kidneys were harvested from mice with specific deletion of TNF-α in T lymphocytes (TNF TKO mice) and WT control mice, and RT-PCR was performed to determine mRNA levels for IL-17A, IL-17F, IL-21, and IL-22. E and F: representative intracellular staining images of CD4⁺IL-17A⁺ T cells and total number of CD4⁺IL17A⁺ T cells in kidneys from WT and TNF TKO mice at 14 days after NTN. G and H: representative intracellular staining images of CD4⁺Foxp3⁺ T cells and total number of CD4⁺Foxp3⁺ T cells in kidneys from WT and TNF TKO mice at 14 days after NTN. Values are means ± SE; n = 12 WT and 11 TNF TKO mice. *P < 0.05.

TNF-α deficiency in T lymphocytes exacerbates infiltration of neutrophils during NTN.

Myeloid cell subsets, including macrophages, dendritic cells, and neutrophils, have each been implicated in the pathogenesis of NTN (6, 10, 13, 14). We therefore quantitated the numbers of myeloid subsets in the kidney by flow cytometry at 14 days after NTN (Fig. 7A). Based on our gating strategy, the numbers of macrophages and dendritic cells were similar between kidneys from TNF TKO and WT mice (Fig. 7, B and C). However, the number of neutrophils was significantly increased in kidneys from TNF TKO mice compared with WT mice (Fig. 7D).

Fig. 7.

TNF-α deficiency in T lymphocytes exacerbates infiltration of neutrophils during nephrotoxic nephritis (NTN). A: representative flow cytometry gating strategy on single cell suspensions from kidneys from mice with specific deletion of TNF-α in T lymphocytes (TNF TKO mice) and wild-type (WT) control mice at 14 days after NTN. SSC, side scatter. B–D: total numbers of macrophages (Macs), dendritic cells (DCs), and neutrophils (PMNs) in kidneys from WT and TNF TKO mice. Values are means ± SE; n = 10 WT and 7 TNF TKO mice. *P < 0.05.

DISCUSSION

Th cells regulate the pathogenesis of not only infectious but also autoimmune diseases (21). Th1 and Th2 subsets were first identified by Mosmann and Coffman based on their activities and secreted cytokines (24). While the Th1-Th2 balance as a dichotomy has been widely used in T cell biology, the roles of Th1 and Th2 cells in NTN are clearly tempered by the activities of the proinflammatory Th17 subset (30). In the present study, we were surprised to find that TNF-α deficiency in T lymphocytes augments kidney injury and fibrosis by promoting Th17 responses. First, the extent of kidney injury and fibrosis in TNF TKO mice exceeds that in WT mice. Second, TNF-α deficiency in T lymphocytes permits increased IL-17A expression and renal accumulation of PD1⁺CD4⁺ T lymphocytes. Finally, T lymphocyte-specific TNF-α deletion augments neutrophil accumulation in the kidney, an effect that is augmented by Th17 activity. Thus, our data establish that TNF-α in T lymphocytes protects against kidney injury and fibrosis and exacerbates Th17 responses during experimental nephritis.

T lymphocyte infiltration into the renal parenchyma is commonly observed in diverse types of sterile kidney diseases (9, 22). Clinical biopsies have shown increased accumulation of CD4⁺ and CD8⁺ T cells in both the glomerulus and tubulointerstitium during glomerulonephritis (4, 28). Whereas CD8⁺ T cells can cause podocyte injury via direct cytotoxic effects (13), autoreactive CD4⁺ T cells regulate the pathogenesis of NTN by augmenting myeloid cell infiltration (15). Accordingly, transfer of antigen-specific T cells alone is sufficient to initiate experimental NTN (39). In an earlier study (38), we found that angiotensin type 1 receptors on T cells attenuate renal fibrosis by inhibiting Th1 differentiation and accumulation of Th1-associated proinflammatory cells. We therefore posited that T cell-derived TNF-α, as a cytokine regulated by angiotensin II, would exacerbate NTN. However, we found that CD4 Cre-mediated TNF-α deletion in T cells leads to more severe glomerular and tubular damage. Moreover, compared with kidneys from WT control mice, expression of fibrosis markers is enhanced and collagen deposition is augmented in kidneys from nephrotic TNF TKO mice. We also found increased infiltration of both CD4⁺ and CD8⁺ T cells in TNF TKO kidneys during NTN. Thus, contrary to our hypothesis, T cell-derived TNF-α protects the kidney from NTN-mediated renal injury.

TNF-α has been implicated in a wide range of kidney diseases and is produced by a diverse set of cell lineages, including Th cells, macrophages, mesangial cells, and renal epithelial cells (11, 17, 27). Global TNF-α deletion or blockade attenuates glomerular and tubular injury and renal inflammation in experimental NTN (18, 19, 34). TNF-α produced by kidney cells and infiltrating macrophages instigates tubular injury (37, 40), so the protective effect of T cell-derived TNF-α in our model is surprising and led us to examine renal accumulation of T cell subsets relevant to NTN pathogenesis. The Th17 CD4 subset and Th17-specific cytokines play unique roles in the pathogenesis of kidney disease, and previous studies have reported that TNF-α promotes Th17 subset differentiation and Th17 responses in nonrenal tissues (26, 31, 43). However, we found that both IL-17A mRNA levels and numbers of CD4⁺IL-17A⁺ T cells were significantly increased in kidneys from TNF TKO mice. Although this finding is surprising, the enhanced accumulation of T cells in NTN kidneys from the TNF TKO cohort is consistent with previous findings from Bertrand et al. (2), who demonstrated that TNF-α signaling promotes cell death of activated T lymphocytes in damaged tissues. PD1 ligands have been reported to be protective in acute kidney injury, as PD1 ligand blockade exacerbated renal function decline, inflammation, and tubular injury during ischemic acute kidney injury (16). However, Celada et al. (5) demonstrated that PD1⁺CD4⁺ T cells promote tissue fibrosis by augmenting the production of TGF-β and IL-17A. Similarly, we found that renal fibrosis and accumulation of PD1⁺CD4⁺ T cells were significantly increased in NTN kidneys of TNF TKO animals compared with WT animals. In addition, IL-17A can activate proinflammatory signaling cascades such as the NF-κB pathway (23) and augment neutrophil infiltration (32), highlighting IL-17A-dependent networks of inflammation.

Th1 cytokines promote the differentiation and renal infiltration of myeloid cell lineages, which can further regulate renal inflammation and fibrosis (6). Dendritic cell activation is necessary for antigen presentation to activate T cells that drive progression of experimental glomerular injury (13, 14). However, in our experiments, the accumulation of macrophages and dendritic cells was similar in NTN kidneys from TNF TKO and WT cohorts. Neutrophils are recruited into the renal tubulointerstitium by Th17-induced chemokine expression in renal tubular epithelial cells (10), and, at 14 days after NTN, we found increased neutrophil accumulation in kidneys from TNF TKO mice compared with WT mice. Collectively, these data suggest that TNF-α deficiency in T lymphocytes aggravates kidney injury and fibrosis by promoting local Th17 responses. These effects may be specific to the kidney, as T cell-derived TNF-α promotes, rather than limits, inflammation in the injured brain (20a).

In summary, our study demonstrates a paradoxical role for TNF-α in T lymphocytes to inhibit glomerular and tubular injury and kidney fibrosis. In our model, TNF-α in T cells constrains renal expression of Th17-derived cytokines and Th17-mediated myeloid cell responses. Our study highlights tissue-specific actions of TNF-α that require careful consideration and further study to maximize the benefits of targeting TNF-α in renal disease without inducing pathogenic off-target effects.

GRANTS

This work was supported by National Institutes of Health Grants DK-118019 and HL-128355 and Veterans Health Administration, Office of Research and Development, Biomedical Laboratory Research and Development Grant BX000893. Y. Wen was in receipt of American Heart Association Predoctoral Fellowship 18PRE34030402.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Y.W. and S.D.C. conceived and designed research; Y.W., N.P.R., J.-d.Z., T.R., X.L., J.R., and S.A.N. performed experiments; Y.W. analyzed data; Y.W. and S.D.C. interpreted results of experiments; Y.W. prepared figures; Y.W. drafted manuscript; Y.W., J.R.P., and S.D.C. edited and revised manuscript; Y.W., N.P.R., J.-d.Z., T.R., X.L., J.R., J.R.P., S.A.N., and S.D.C. approved final version of manuscript.