If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Activation of the JUN amino-terminal kinase (JNK) pathway is prominent in most forms of acute and progressive tubulointerstitial damage, including acute renal ischemia/reperfusion injury (IRI). Two forms of JNK, JNK1 and JNK2, are expressed in the kidney. Systemic administration of pan-JNK inhibitors suppresses renal IRI; however, the contribution of JNK1 versus JNK2, and the specific role of JNK activation in the proximal tubule in IRI, remains unknown. These questions were addressed in rat and mouse models of acute bilateral renal IRI. Administration of the JNK inhibitor, CC-930, substantially reduced the severity of renal failure, tubular damage, and inflammation at 24 hours in a rat IRI model. Additionally, Jnk1−/− mice, but not Jnk2−/− mice, were shown to be significantly protected against acute renal failure, tubular damage, and inflammation in the IRI model. Furthermore, mice with conditional Jnk1 deletion in the proximal tubule also showed considerable protection from IRI-induced renal failure, tubular damage, and inflammation. Finally, primary cultures of Jnk1−/−, but not Jnk2−/−, tubular epithelial cells were protected from oxidant-induced cell death, in association with preventing phosphorylation of proteins (receptor interacting serine/threonine kinase 3 and mixed lineage kinase domain-like pseudokinase) in the necroptosis pathway. In conclusion, JNK1, but not JNK2, plays a specific role in IRI-induced cell death in the proximal tubule, leading to acute renal failure.
Renal ischemia/reperfusion injury (IRI) is a common cause of acute kidney injury (AKI), which involves a partial or complete cessation of blood flow to the kidney with subsequent restoration of the circulation and re-oxygenation. This can occur in various settings, including kidney transplantation, cardiopulmonary bypass surgery, and severe hypotension associated with major blood loss or sepsis.
Ischemia results in a mismatch between oxygen supply and demand in local tissues, whereas re-oxygenation leads to excess production of oxygen free radicals. The ensuing hypoxia, oxidative stress, and impaired removal of waste result in tubular injury, inflammation, and, in cases of severe injury, cell death.
The pathogenesis of renal IRI is complex and involves direct tubular epithelial cell damage through hypoxia/re-oxygenation and mitochondrial damage, as well as indirect tubular damage via the recruitment and activation of neutrophils and monocyte/macrophages.
JNK is extremely sensitive to activation by reactive oxygen species, osmotic stress, and proinflammatory cytokines. Multiple JNK isoforms are encoded by three genes (Jnk1, Jnk2, and Jnk3), of which JNK1 and JNK2 are widely expressed, including in cells of the kidney.
or whether only one JNK isoform is involved. In addition, these studies do not define whether JNK signaling in the proximal tubule is primarily responsible for inducing acute renal failure following IRI, or whether JNK signaling in other cell types may also be involved.
This study addressed two important mechanistic questions regarding the role of JNK signaling in IRI. First, is there redundancy between JNK1 and JNK2 in this response? Second, is JNK activation within the proximal tubule sufficient to induce tubular cell damage, inflammation, and acute renal failure?
Materials and Methods
The following primary antibodies were used: rabbit antibodies against phosphorylated JUN Ser63, cleaved caspase 3, phosphorylated receptor interacting serine/threonine kinase 3 (RIP3) Thr231/Ser232 and phosphorylated mixed lineage kinase domain-like pseudokinase (MLKL) Ser345 (Cell Signaling, San Diego, CA), rabbit anti–γ-glutaryl transferase, rat anti-mouse neutrophils (Ly6G), mouse anti-rat neutrophils (RP1), and mouse anti–α-tubulin (Abcam, Melbourne, VIC, Australia). CC-930 is a pan-JNK inhibitor and was provided by Celgene Corp. (San Diego, CA).
Outbred male Sprague-Dawley rats and inbred male C57BL/6J mice were obtained from the Monash Animal Research Platform (Clayton, VIC, Australia). Mapk8/Jnk1−/− and Mapk9/Jnk2−/− mice on the C57BL/6J background were imported from Jackson Laboratories (Bar Harbor, ME), and validation of the gene knockout is presented in Supplemental Figure S1. Jnk1f/f and γ-glutamyltransferase (γGT)–Cre recombinase expressing mice have been described previously,
and were backcrossed at least 12 times onto the C57BL/6J background. Animal studies were approved by the Monash Medical Center Animal Ethics Committee and performed according to the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes.
Rats underwent bilateral IRI, as previously described.
Male rats were anesthetized with ketamine and xylazine. A heating blanket connected to a rectal thermometer was used to maintain body temperature at 37°C. Following a midline abdominal incision, both renal pedicles were clamped using non-traumatic vascular clamps for 25 minutes, during which the abdomen was temporarily sutured. Clamps were removed, and kidney reperfusion was confirmed visually. The abdomen was sutured in two layers, and saline was provided by s.c. injection. Analgesia involved s.c. injection of 0.05 mg/kg buprenorphine and 2 to 3 drops of bupivacaine onto the sutures at the end of surgery. Male mice underwent the same bilateral IRI surgical procedure, with a 17-minute (Jnk1−/− study) or a 19-minute (Jnk2−/− and Jnk1f/fγGT-Cre study) ischemic time.
Groups of 9 or 10 rats were treated with 60 mg/kg b.i.d. CC-930 or vehicle alone (0.5% carboxymethylcellulose and 0.25% Tween-20) by oral gavage, starting 1 hour before surgery and continuing until animals were sacrificed 24 hours after reperfusion. Sham control animals underwent the same surgical procedure, except that the renal pedicles were not clamped. Serum creatinine and blood urea nitrogen levels were measured using a Dupont ARL Analyser (Wilmington, DE) by the Department of Biochemistry, Monash Health.
Analysis of Kidney Damage
Tubular damage was assessed in the outer medulla on periodic acid–Schiff–stained sections (2 μm thick) of formalin-fixed tissue. The percentage of tubular cross-sections exhibiting damage was scored under high power (×400): damage was characterized as loss of the brush border, nuclear loss, and sloughing of cells into the lumen. The average number of tubular cross-sections scored in the rat model was 1293 per case, and the average number was 1155 per case in the mouse model. In the rat model, cell death was assessed in sections (4 μm thick) of formalin-fixed tissue by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining with the ApopTag Peroxidase In Situ Apoptosis Detection Kit (Millipore-Chemicon, Ryde, NSW, Australia). The number of TUNEL+ tubular cells in the outer medulla was counted in high-power (×400) fields. All scoring was performed on blinded slides (F.Y.M., K.G.).
Immunostaining for phosphorylated JUN (p-JUN), cleaved caspase 3, and γGT was performed on formalin-fixed sections (4 μm thick) using antigen retrieval with 0.1 mol/L sodium citrate, pH 6.0, and a three-layer avidin-biotin peroxidase complex staining method, as previously described.
Blots were incubated with antibodies to phosphorylated JUN, phosphorylated RIP3, phosphorylated MLKL, or α-tubulin antibodies, which were detected with goat anti-rabbit Alexa Fluor 680 or donkey anti-mouse IRDye 800 secondary antibodies (Thermo Fisher Scientific, Scoresby, VIC, Australia) using the Odyssey Infrared Image Detecting System (LI-COR, Lincoln, NE). Densitometry analysis used ImageJ software version 1.51r (NIH, Bethesda, MD; https://imagej.nih.gov/ij).
RNA was extracted from frozen kidney samples with the Ambion RiboPure Kit (Thermo Fisher Scientific) and reverse transcribed with random primers using the SuperScript III First-Strand Synthesis System (Thermo Fisher Scientific). The PCR was performed using a StepOne Real-Time PCR system (Thermo Fisher Scientific) using TaqMan probes. The primer/probes for Nos2 and Tnf have been described previously,
and the other primer/probes were purchased from Thermo Fisher Scientific. All amplicons were normalized against 18S in the same reaction, and the relative amount of mRNA was determined using the comparative cycle threshold (ΔCt) method.
Primary Tubular Epithelial Cell Culture
Tubular cells were isolated from the kidney of normal Jnk1−/−, Jnk2−/−, and wild-type (WT) mice and cultured on collagen-coated plates, as previously described.
To measure cell death, cells were starved in 1% fetal calf serum for 18 hours, and then varying concentrations of hydrogen peroxide were added for 24 hours. Cells then were analyzed using the Cell Death Detection ELISA Kit (Roche, Mannheim, Germany), with results normalized to the DNA content in cell lysates using a Quant-iT DNA Assay Kit (Molecular Probes, Eugene, OR) and expressed as the ratio of OD/DNA content.
Data are shown as means ± SD. Data were analyzed by one-way analysis of variance with the Tukey multiple comparison test using GraphPad Prism 8.0 software (GraphPad Software, San Diego, CA).
Pharmacologic JNK Inhibition Prevents IRI-Induced AKI
Vehicle-treated rats developed severe acute kidney injury at 24 hours after bilateral renal IRI based on a marked increase in serum creatinine and blood urea nitrogen levels (Figure 1, A and B ). This was associated with pronounced histologic damage, including extensive tubular necrosis (Figure 1, C–E). Tubular damage was also evident by up-regulation of the tubular damage marker, Kim1/Hacvr1 (Figure 2E), and considerable tubular cell death, as shown by TUNEL staining (Figure 2, A, B and D). Vehicle-treated IRI showed a substantial neutrophil infiltrate based on immunostaining and increased Elane mRNA levels (Supplemental Figure S2, A and B, and Figure 2F), as well as a significant infiltrate of macrophages (Supplemental Figure S2, D and E) with increased mRNA levels of Cd68 and the macrophage activation marker, Nos2 (Figure 2, G and H). There was also up-regulation of the inflammatory cytokine, Tnf (Figure 2I).
Because CC-930 does not prevent JNK phosphorylation,
phosphorylation of JUN at Ser63 was examined as a readout of JNK activity. Only occasional cells exhibited p-JUN staining in the sham control. By contrast, many tubular epithelial cells exhibited p-JUN staining in the vehicle-treated IRI group (Figure 3, A and B ). By comparison, CC-930 treatment substantially reduced the number of p-JUN+ cells, as shown by immunostaining and Western blot analysis (Figure 3, C–E). CC-930 treatment provided substantial protection against acute kidney injury, as shown by much lower serum creatinine and blood urea nitrogen levels (Figure 1, A and B), less histologic damage (Figure 1, C and F), reduced Kim1/Havcr1 mRNA levels (Figure 2E), and fewer TUNEL+ cells (Figure 2, C and D). In addition, CC-930 treatment significantly reduced the neutrophil and macrophage infiltration (Supplemental Figure S2, C and F) and expression of Nos2 and Tnf (Figure 2, E–I).
Jnk1−/− Mice Are Protected from IRI-Induced AKI
Because CC-930 inhibited virtually all JNK activity in the IRI model, a genetic approach was used to examine the relative contribution of JNK1 versus JNK2. The response of WT and Jnk1−/− mice was compared in a 24-hour model of bilateral IRI-induced acute kidney injury. WT mice developed acute kidney injury 24 hours after IRI, as shown by increased levels of serum creatinine and blood urea nitrogen, histologic tubular damage, and tubular cell death, as identified by TUNEL and cleaved caspase 3 staining (Figure 4, A–E, G, H, J, K, and M–O ). This tubular damage was associated with a substantial neutrophil infiltrate in the outer medulla and inner cortex (Figure 5, A and B ).
In contrast, Jnk1−/− mice showed significant protection from acute kidney injury in the renal IRI model based on lower levels of serum creatinine and blood urea nitrogen (Figure 4, A and B). Jnk-1−/− mice also showed less severe histologic tubular damage as well as a reduction in tubular cell death identified by TUNEL and cleaved caspase 3 staining (Figure 4, F, I, and L–O). Furthermore, Jnk1−/− mice exhibited less neutrophil infiltration (Figure 5, C and D).
Jnk2−/− Mice Are Not Protected from IRI-Induced AKI
Wild-type and Jnk2−/− mice were examined in a 24-hour model of bilateral IRI. Compared with the marked increase in JUN phosphorylation seen in the WT IRI group by Western blot analysis, Jnk2−/− mice showed an approximately 50% reduction in JUN phosphorylation (Figure 6, C and D ). However, Jnk2−/− mice were not protected from the development of severe kidney injury following IRI based on serum creatinine and blood urea nitrogen levels (Figure 7, A and B ), tubular damage, and Kim1/Havcr1 levels (Figure 7, C, E, and F). Similarly, Jnk2−/− mice were not protected from neutrophil or macrophage infiltration (Supplemental Figure S3, A–C and E–G, and Figure 8, A and B ) or the up-regulation of Ccl2, Nos2, Tnf, or Il6 mRNA levels (Figure 8, C–F).
Jnk1PTJnk2−/− Mice Are Protected from IRI-Induced AKI
The protection of Jnk1−/− mice in the renal IRI model could be due to JNK1 facilitating tubular cell necrosis and/or JNK1 in neutrophils and macrophages facilitating their recruitment and activation in the injured kidney. To address the first possibility, mice lacking Jnk1 in proximal tubules (Jnk1f/fγGTCre mice, termed Jnk1PT) were generated on the Jnk2−/− background (note, the Jnk2−/− background was employed before it was known that JNK2 plays no role in the IRI model).
First, a pilot study was performed to assess JNK signaling at 30 minutes after reperfusion, a time of maximal JNK activation and before the loss of tubular morphology.
Jnk1f/fJnk2−/− mice showed a strong induction of JUN Ser63 staining at 30 minutes in most tubular segments, including γGT stained proximal tubules, in the inner cortex and outer medulla (Figure 6A). By contrast, Jnk1PTJnk2−/− mice showed a major reduction in p-JUN Ser63 staining in γGT+ proximal tubules at 30 minutes after reperfusion, although staining in other tubular segments was still apparent (Figure 6B). Western blot analysis of kidneys showed an almost complete inhibition of JUN phosphorylation in Jnk1PTJnk2−/− mice (Figure 7, Figure 6 C and D).
Compared with the severe renal dysfunction seen in WT and Jnk2−/− mice at 24 hours after IRI, Jnk1PTJnk2−/− mice showed a marked reduction in serum creatinine and blood urea levels (Figure 7, A and B). In addition, Jnk1PTJnk2−/− mice showed less histologic tubular damage and a reduction in Kim1/Havcr1 mRNA levels (Figure 7, D–F). Furthermore, Jnk1PTJnk2−/− mice showed a reduction in neutrophil and macrophage infiltration (Supplemental Figure S3, D and H, and Figure 8, A and B) and a reduction in Ccl2, Nos2, and Il6 mRNA levels; however, the reduction in Tnf mRNA levels seen compared with WT IRI mice did not reach statistical significance compared with the Jnk2−/− IRI group (Figure 8, C–F).
JNK1 Activates the RIP3/MLKL Pathway of Necroptosis in Tubular Epithelial Cells
The role of JNK1 and JNK2 in oxidant-induced cell death was investigated in primary cultures of tubular epithelial cells isolated from Jnk1−/−, Jnk2−/−, and WT mice. Tubular cells from WT and Jnk2−/− mice showed an equivalent dose-dependent increase in cell death in response to hydrogen peroxide stimulation, whereas Jnk1−/− tubular cells were significantly protected against cell death (Figure 9A). CC-930 treatment significantly reduced hydrogen peroxide–induced cell death in both WT and Jnk2−/− cells (Figure 9, B and D), whereas no further protection against cell death was seen in Jnk1−/− cells (Figure 9C). Components of the necroptosis pathway were examined at 4 hours after hydrogen peroxide stimulation in cultured tubular epithelial cells. Both WT and Jnk2−/− cells showed hydrogen peroxide–induced phosphorylation of RIP3 and MLKL, but this was absent in Jnk1−/− cells (Figure 10).
This study has demonstrated that pharmacologic inhibition of JNK signaling provides substantial protection against IRI-induced acute renal failure. Genetic studies identified that it is the JNK1 enzyme within the proximal tubule that plays a key role in the development of acute renal failure following IRI.
The demonstration that CC-930 provides substantial protection against acute renal failure following IRI provides significant new information. Wang et al
showed that treatment with the JNK inhibitor SP600125 provides significant protection against acute renal failure in a rat bilateral IRI model. However, SP600125 is not selective for JNK as it inhibits 22 other kinases with equal or greater potency than JNK isoforms.
Therefore, this finding was replicated in the more challenging situation of severe IRI and CC-930 JNK inhibitor was used as it has greater potency and a considerably longer half-life compared with CC-401.
In the current study, the rat IRI model induced a 5.4-fold increase in serum creatinine at 24 hours after IRI, and CC-930 treatment reduced serum creatinine levels by 60%. Furthermore, CC-930 treatment was effective in inhibiting JNK signaling, reducing tubular damage and necrosis, and reducing leukocyte infiltration and inflammation. These findings support the potential for a pharmacologic approach to preventing anticipated renal IRI-induced acute renal failure.
Previous studies have shown a considerable redundancy between JNK1 and JNK2 signaling in terms of JUN phosphorylation in tubular cells in the obstructed mouse kidney.
In the current study, both JNK1 and JNK2 made a substantial contribution to JNK signaling in the kidney following IRI. Genetic deletion of Jnk2 reduced JUN phosphorylation by approximately 50%, and then combined Jnk2 deletion and Jnk1 deletion in the proximal tubule reduced JUN phosphorylation by approximately 90%. However, global Jnk1, but not Jnk2, deletion provided substantial protection against IRI-induced acute renal failure. This finding was supported by in vitro studies showing protection of Jnk1−/−, but not Jnk2−/−, tubular cells from reactive oxygen species–induced cell death.
The finding that JNK1, and not JNK2, mediates IRI-induced acute renal failure is consistent with some, but not other, studies of nonrenal IRI. Jnk2−/− mice show protection in a model of liver IRI.
Studies using CC-930 treatment and global Jnk1−/− mice indicate that systemic blockade of JNK not only reduces tubular cell necrosis, but also reduces neutrophil and macrophage infiltration, macrophage activation, and general markers of inflammation. The question is whether JNK signaling is important in multiple cell types in renal IRI. Although JNK signaling, as assessed by JUN phosphorylation, occurred predominantly in tubular epithelial cells in the IRI kidney, it does not preclude the importance of JNK signaling in infiltrating leukocytes because activation of this pathway may be transient and therefore difficult to detect by immunostaining. For example, JNK signaling is important for inducing a proinflammatory macrophage phenotype and renal injury in experimental glomerulonephritis,
Jnk1 deletion in the proximal tubule was effective in reducing tubular cell damage and death as well as reducing neutrophil and macrophage infiltration, macrophage activation, and inflammatory markers. Thus, preventing the release of danger-associated molecular patterns from injured and dying tubular cells may be of greater importance in driving the leukocyte infiltration and activation than JNK activation per se in these cells, although studies of JNK deletion in myeloid cells to confirm this have yet to be performed.
Given that Jnk1−/−, but not Jnk2−/−, mice were protected in the renal IRI model, it follows that there is a specific JNK1 target involved in the induction of tubular cell death. A central role for the necroptosis pathway in mediating IRI-induced acute kidney injury has been shown by using mice lacking the key enzymes RIP3 or MLKL and the use of the RIP1 inhibitor Necrostatin-1.
In the in vitro studies, deletion of Jnk1, but not Jnk2, in tubular cells prevented activation (phosphorylation) of RIP3 and MLKL at 4 hours after exposure of cells to reactive oxygen species. This indicates that JNK1 signaling is an early and essential component of the necroptosis pathway in tubular cells exposed to oxidative stress. Studies in other cell types and other animal models have shown that JNK and RIP3/MLKL act in an amplification loop to promote necroptotic cell death.
However, this is the first study to demonstrate that JNK1 is required for RIP3 and MLKL activation in a necroptotic pathway. Further studies are required to identify the intermediate steps between JNK1 activation and RIP3 activation. One limitation of these in vitro studies is that they focus on oxidative stress–induced cell death without examining ischemia, an important cause of cell death in the IRI model.
In conclusion, the current study demonstrates that JNK1, but not JNK2, plays a critical role in acute renal failure following IRI. In particular, JNK1 activation in the proximal tubule promotes tubular cell damage and necrosis, with cell culture studies identifying a role for JNK1 in activation of the RIP3/MLKL pathway of necroptosis. Finally, pharmacologic blockade of JNK signaling has therapeutic potential for prevention of anticipated renal ischemia/reperfusion injury.
We thank Prof. Eric Neilson (Vanberbilt University Medical Center, Nashville, TN) for kindly providing γ-glutamyltransferase–Cre mice; and Celgene for providing CC-930 for these studies.
Characterization of Jnk1−/− and Jnk2−/− mice. A and B: Real-time PCR analysis of kidney tissue from groups of wild-type (WT), Jnk1−/−, and Jnk2−/− mice. A:Jnk1 mRNA was detected in WT and Jnk2−/− mice, but not in Jnk1−/− mice. B:Jnk2 mRNA was detected in WT and Jnk1−/− mice. A weak signal for Jnk2 mRNA was detected in Jnk2−/− mice. The TaqMan PCR probe for Jnk2 mRNA is located downstream of the deleted exons, which may allow detection of a low-abundance, truncated transcript. C: Western blot analysis of mouse kidney using an antibody against both JNK1 and JNK2 (JNK1/2; Santa Cruz Biotechnology, Dallas, TX; SC571). Compared with WT kidney, there is a clear reduction in the JNK1/2 signal in both Jnk1−/− and Jnk2−/− mice. Loading was assessed by detection of TUBA1A/tubulin. D: Western blot analysis for JNK2 (Cell Signaling antibody CS9258) detects high levels of multiple JNK2 transcripts in WT and Jnk1−/− kidneys, but only weak bands in Jnk2−/− kidneys (blot is deliberately overexposed to show these bands), which presumably reflects weak cross-reactivity with JNK1. n = 6 WT, Jnk1−/−, and Jnk2−/− mice (A and B). N.D., not detected.