The protective effect of H151, a novel STING inhibitor, in renal ischemia-reperfusion-induced acute kidney injury

Abstract

Renal ischemia-reperfusion (RIR)-induced acute kidney injury (AKI) is a common renal functional disorder with high morbidity and mortality. Stimulator of interferon (IFN) genes (STING) is the cytosolic DNA-activated signaling pathway that mediates inflammation and injury. Our recent study showed that extracellular cold-inducible RNA-binding protein (eCIRP), a newly identified damage-associated molecular pattern, activates STING and exacerbates hemorrhagic shock. H151 is a small molecule that selectively binds to STING and inhibits STING-mediated activity. We hypothesized that H151 attenuates eCIRP-induced STING activation in vitro and inhibits RIR-induced AKI in vivo. In vitro, renal tubular epithelial cells incubated with eCIRP showed increased levels of IFN-β, STING pathway downstream cytokine, IL-6, tumor necrosis factor-α, and neutrophil gelatinase-associated lipocalin, whereas coincubation with eCIRP and H151 diminished those increases in a dose-dependent manner. In vivo, 24 h after bilateral renal ischemia-reperfusion, glomerular filtration rate was decreased in RIR-vehicle-treated mice, whereas glomerular filtration rate was unchanged in RIR-H151-treated mice. In contrast to sham, serum blood urea nitrogen, creatinine, and neutrophil gelatinase-associated lipocalin were increased in RIR-vehicle, but in RIR-H151, these levels were significantly decreased from RIR-vehicle. In contrast to sham, kidney IFN-β mRNA, histological injury score, and TUNEL staining were also increased in RIR-vehicle, but in RIR-H151, these levels were significantly decreased from RIR-vehicle. Importantly, in contrast to sham, in a 10-day survival study, survival decreased to 25% in RIR-vehicle, but RIR-H151 had a survival of 63%. In conclusion, H151 inhibits eCIRP-induced STING activation in renal tubular epithelial cells. Therefore, STING inhibition by H151 can be a promising therapeutic intervention for RIR-induced AKI.

NEW & NOTEWORTHY Renal ischemia-reperfusion (RIR)-induced acute kidney injury (AKI) is a common renal functional disorder with a high morbidity and mortality rate. Stimulator of interferon genes (STING) is the cytosolic DNA-activated signaling pathway responsible for mediating inflammation and injury. Extracellular cold-inducible RNA-binding protein (eCIRP) activates STING and exacerbates hemorrhagic shock. H151, a novel STING inhibitor, attenuated eCIRP-induced STING activation in vitro and inhibited RIR-induced AKI. H151 shows promise as a therapeutic intervention for RIR-induced AKI.

INTRODUCTION

Renal ischemia-reperfusion (RIR) injury is characterized by vascular dysfunction, renal tubular damage, and renal inflammation. RIR injury often occurs as a sequelae in various pathologies, such as infection, trauma, hemorrhage, or iatrogenic events that suddenly compromise blood flow to the kidneys (1, 2). After the initial ischemic event, restoration of blood flow or reperfusion occurs. First, reperfusion allows blood flow to the ischemic tissue, promoting aerobic metabolism and subsequent recovery from ischemia. On the other hand, reperfusion also leads to neutrophil accumulation, oxygen-free radical generation, and proinflammatory cytokine activation, which with repeated insult can progress to acute kidney injury (AKI). AKI is a common and detrimental complication among hospitalized patients. According to clinical observations, 10–15% of hospitalized patients suffer from AKI, and among patients with AKI in intensive care units, mortality is up to 50% or higher (3, 4). Despite enormous efforts in preclinical investigations, there are no effective therapeutics to prevent RIR-induced AKI.

Recently, it has been shown that the cytosolic DNA-sensing adaptor stimulator of interferon (IFN) genes (STING) plays an important role in cisplatin-induced AKI (5). STING is the critical signaling component in the cytosolic DNA-induced signaling pathway that detects cytosolic DNA of pathogen or self-origin (6). STING is expressed mainly in hematopoietic cells such as macrophages, dendritic cells, natural killer cells, and T cells and is generally localized to the endoplasmic reticulum. Upon introduction of pathogen-derived double-stranded DNA or self-DNA, cGMP-AMP synthase (cGAS) is activated and the STING ligand cyclic di-GMP-AMP (cGAMP) is produced (7). By cGAMP ligand binding, STING is activated, translocated to the Golgi apparatus, and then recruits TANK-binding kinase 1 (TBK1). TBK1 then auto-phosphorylates and activates IFN regulatory factor 3 (IRF3). Activated IRF3 promotes type I IFN production. Activated STING also causes NF-κB phosphorylation by activating IκB kinase (IKK), leading to the transcription of proinflammatory cytokines (7).

Our recent study showed that extracellular cold-inducible RNA binding protein (eCIRP) activates the STING pathway and exacerbates hemorrhagic shock-induced lung injury in mice (8). eCIRP was identified as a damage-associated molecular pattern in various injury conditions including RIR injury (9, 10). eCIRP-induced STING activation led to type 1 IFN production in lung tissues and macrophages in vitro. Furthermore, in contrast to wild-type mice, we demonstrated that administration of eCIRP in mice deficient in STING (STING^−/− mice) showed decreased expression of type 1 IFN and downstream STING signaling proteins including TBK1 (8). H151 is a potent small molecule that selectively and irreversibly binds to STING and inhibits STING palmitoylation and clustering, which is critical for STING relocation and downstream protein activation (11). Since H151 potently binds human and murine STING, testing the compound on mouse models provides valuable knowledge for its potential development for clinical application. Renal tubular epithelial cells (RTECs) are the most vulnerable cells affected during RIR (5). Therefore, we hypothesized that blocking eCIRP-induced STING activation and subsequent type 1 IFN production inhibits inflammation in these cells. We further hypothesized that targeting the STING pathway could be a potential therapeutic strategy for RIR-induced AKI. We, therefore, investigated the impact of H151 in eCIRP-induced STING activation in RTECs in vitro and the effect of H151 in RIR-induced AKI in mice.

MATERIALS AND METHODS

STING Inhibitor H151

H151 was purchased from InvivoGen (Cat. No. inh-h151) as a lyophilized powder. The powder was dissolved in sterilized DMSO to make a 10 mg/mL stock solution. The stock solution was then diluted with 10% Tween 80 in PBS before its use in mice.

Primary RTEC Isolation and Treatment

Primary RTECs were isolated as described in a previous study (12). The renal cortex was collected, minced, and digested in 37°C for 5 min. After the undigested tissue was removed, the digestion was stopped by adding 10 mL of culture media. The cell suspension was centrifuged at 50 g for 5 min to collect the pellet containing tubular cells, and the pellet was washed for a second time to collect purified RTECs. The cell pellet was then resuspended in culture media and seeded onto a collagen-coated culture dish. Isolated RTECs were then coincubated with 1 µg/mL eCIRP and H151 at a series of concentrations for 24 h. After incubation, the cells and supernatants were collected for further analysis.

Experimental Animals

Male C57BL6 mice aged from 8 to 12-wk old were purchased from Charles River Laboratories. Mice were housed under a standard 22°C 12/12-h light-dark cycle and fed standard Purina mouse chow and water ad libitum. All animals were allowed to acclimate for 1 wk after arrival before experimentation. All animal experiments were approved by the Institutional Animal Care and Use Committee of The Feinstein Institutes for Medical Research.

Animal Model of RIR

Before surgery, animals were randomly divided into sham, RIR-vehicle, and RIR-H151 treatment groups. RIR was performed as previously described (10, 13). Anesthesia was induced by inhalation of 3−4% isoflurane with an O₂ flow rate of 0.5–1 L/min and maintained at 1−2% isoflurane for the duration of the surgery and ischemia period. Under anesthesia, mice were secured onto the surgical plane in a supine position. With a midline laparotomy incision, the bilateral kidney hila were exposed, and the renal pedicles were carefully dissected. Microvascular clamps were placed across the renal artery and vein bilaterally to induce ischemia. During ischemia, mice were placed on a heating pad to maintain body temperature. If ischemia was successfully induced, the insulted kidneys became homogenously dark purple. After 30 min of ischemia, the microvascular clamps were removed. Still under anesthesia and immediately after the initiation of reperfusion, 10 mg/kg body wt H151 in 10% Tween 80 in PBS was given intraperitoneally, and the mouse abdomen was closed with nylon sutures. The dosage was applied according to previous in vivo research on H151 (11, 14, 15). The vehicle group received equivalent amounts of 10% Tween 80 in PBS at the beginning of reperfusion. In the sham group, under anesthesia, similar procedures to those done in the vehicle and treatment groups were performed except for the microvascular clamping. Mice were hydrated with a subcutaneous injection of normal saline. Upon recovery from anesthesia, mice were moved to their home cages and monitored twice a day until the end of the experiment. At 24 h after reperfusion, blood was collected under anesthesia and the mice were euthanized thereafter. The right kidneys were harvested, flash frozen in liquid nitrogen, and stored at −80°C for future analysis. The left kidneys were fixed in 10% formalin and processed for histological examination. Whole blood samples were spun down at 3,000 g for 10 min. Serum was collected and stored at −80°C for later use.

Transdermal Measurement of Renal Function

Transdermal glomerular filtration rate (GFR) measurement was performed as previously described (16, 17). At 22 h after RIR or sham operation, mice were lightly anesthetized by isoflurane inhalation, and the fur on the flank of each mouse was removed using an animal shaver followed by depilatory cream. A transdermal mini GFR monitor (Cat. No. TDM-MD004, MediBeacon) was secured on the flank after the exposed skin was carefully cleaned. After the animal was anesthetized with the monitor attached for 3 min to collect baseline information, FITC-sinistrin at 7.5 mg/100 g body wt was administered by retroorbital injection. Each mouse was then allowed to recover from anesthesia, and the device was kept on the mouse for at least 1.5 h to collect sufficient data for analysis. No blood samples were collected from these mice as the GFR measurement is noninvasive. After data collection, the mice were euthanized. The collected information was processed and evaluated with MB_lab and MB_Studio software from MediBeacon to obtain GFR data.

Determination of Blood Urea Nitrogen and Creatinine Levels

Serum blood urea nitrogen (BUN) and creatinine levels were measured on the same day of blood collection before being stored at −80°C. The measurement was carried out by colorimetric enzymatic assays using commercially available kits (Pointe Scientific).

Real-Time Quantitative PCR

The frozen kidney tissue was crushed into a fine powder, and total RNA was isolated using TRIzol reagent (Invitrogen). cDNA was synthesized from total RNA using murine leukemia virus reverse transcriptase (Applied Biosystems). PCR was performed in 20 µL reaction mixture consisting of 10 µL Power SYBR Green PCR Master Mix (Applied Biosystems), 0.25 µL each of target gene forward and reverse primers, 7.5 µL of diethyl pyrocarbonate (DEPC)-treated water, and 2 µL of cDNA. PCRs were carried out and analyzed by a StepOnePlus real-time PCR machine (Applied Biosystems). The relative expression level of the target mRNA was calculated by the 2^–ΔΔCt method (where C_t is threshold cycle), and β-actin mRNA was used as the internal control for normalization during analysis. The results are presented as fold changes relative to the corresponding RNA level in the sham group. The primers for quantitative PCR were as follows: IFN-β, forward 5′- TGACGGAGAAGATGCAGAAG-3′ and reverse 5′- ACCCAGTGCTGGAGAAATTG-3′; interleukin (IL)-6, forward 5′- CCGGAGAGGAGACTTCACAG-3′ and reverse 5′- CAGAATTGCCATTGCACAAC-3′; neutrophil gelatinase-associated lipocalin (NGAL), forward 5′- CTCAGAACTTGATCCCTGCC-3′ and reverse 5′- TCCTTGAGGCCCAGAGACTT-3′; keratinocyte chemoattractant (KC), forward 5′- GCTGGGATTCACCTCAAGAA-3′ and reverse 5′- ACAGGTGCCATCAGAGCAGT-3′; macrophage inflammatory protein (MIP)-2, forward 5′- CATCCAGAGCTTGAGTGTGA-3′ and reverse 5′- CTTTGGTTCTTCCGTTGAGG-3′; kidney injury molecule (KIM)-1, forward 5′- TGCTGCTACTGCTCCTTGTG-3′ and reverse 5′- GGGCCACTGGTACTCATTCT-3′; and β-actin, forward 5′- CGTGAAAAGATGACCCAGATCA-3′ and reverse 5′- TGGTACGACCAGAGGCATACAG-3′.

Enzyme-Link Immunosorbent Assay

Serum levels of NGAL, the early cell injury indicator, and kidney tissue protein levels of proinflammatory cytokine IL-6 were measured by ELISA kits (Bio-techne for NGAL and BD Biosciences for IL-6) according to the manufacturers’ instructions. To measure plasma levels of NGAL, the plasma was diluted in normal saline before being seeded on the ELISA plate. For protein extraction, frozen kidney tissue was crushed into powder in liquid nitrogen, and protein was then extracted from this powder using RIPA lysis buffer that contained proteinase and phosphatase inhibitors. Protein lysate concentrations were then determined by Bradford protein assay reagent (Bio-Rad, Hercules, CA); 100 µg protein was then used in the kidney tissue ELISA for IL-6 measurement.

Renal Histopathological Evaluation

After being fixed in formalin, the harvested kidneys were embedded in paraffin. The tissue was then sectioned into 5-µm slices and stained with hematoxylin and eosin (H&E). Kidney injury was evaluated through light microscopy observation under blinded conditions. As previously reported in the literature (18), renal tubular cells suffering from RIR injury present with a loss of the tubular brush border, renal tubular disruption, and cast formation. The severity of kidney injury was assessed based on previous research (10) and scored in the following five categories: tubular cell injury, tubular cell detachment, cast formation, tubular dilation, and interstitial congestion or bleeding. Each category was then given a score from 0 to 5, depending on the percentage observed in the slide. Percent differences were scored as follows: 0 (absent), 1 (>0−10%), 2 (>10−25%), 3 (>25−50%), 4 (>50−75%), or 5 (>75%) with a maximum combined score of 25. For each sample, two cross sections were then stained with H&E, and six random fields of each cross section were scored. The final score for the samples was calculated as the average score from all six evaluations.

TUNEL Assay

Kidney cellular apoptosis was identified by an in situ cell death detection kit (Cat. No. 11684795910, Roche Diagnostics). Paraffin-embedded kidney tissue was sectioned into 5-µm slices and fixed onto slides. Tissue sections were then deparaffinized in xylene, washed and hydrated in a graded ethanol series, and incubated with proteinase K for 20 min at room temperature for permeabilization. Afterward, tissue sections were washed with Tris buffered saline and incubated with a premixed TUNEL enzyme and fluorescence-labeled nucleotide solution for 30 min at 37°C. Tissue sections that were incubated with only fluorescence-labeled nucleotide solution served as the negative control for each sample. TUNEL fluorescence from at least four random fields per sample was examined and recorded under a fluorescent microscope (Nikon Eclipse Ti-S). TUNEL-positive cells were then counted using ImageJ software.

Survival Study

Since we did not observe any mortality after 30 min of ischemia, we used a slightly more severe model of RIR for the survival study. Mice underwent bilateral ischemia for 34 min followed by reperfusion. At the time of reperfusion, mice were then randomly given either vehicle or H151 compound intraperitoneally. In the following 10 days, mice were checked twice daily, and their survival status was recorded. Mice that succumbed to death and moribund mice that fulfilled the criteria for early euthanasia were included in the survival analysis. In terms of the criteria for early euthanasia in the survival study, the humane end points were defined as the point at which death is imminent or suffering is irreversible. In the presence of two or more of the following criteria, mice were euthanized. The humane end points were as follows: 1) weight loss > 20%, 2) minimal or no response to stimuli, 3) Grimace score of 2, 4) body condition score ≤ 2, and 5) labored/agonal breathing.

Statistical Analysis

Statistical analysis was carried out using GraphPad Prism (GraphPad Software). Data are presented as means ± SE. All tested parameters were compared by one-way ANOVA except for the survival study. Post hoc analysis was conducted using a Student Newman–Keul’s test. The one-way ANOVA depicted the overall significance among the three groups, and the Student Newman–Keul test tested the difference between the two groups. When there was an overall significance of P < 0.05 via ANOVA, then only the post hoc analysis was done to compare the data between the two groups. A log-rank (Mantel-Cox) test was used for the survival study analysis. This test was used to compare two Kaplan-Meier curves, one curve pertaining to the vehicle group versus survival time and the second curve pertaining to the treatment group versus survival time. Both curves were then analyzed with a two-tailed test for a critical value of P < 0.05.

RESULTS

H151 Reduced eCIRP-Induced STING Activation and Inflammation in RTECs

To determine the effect of H151 on eCIRP-induced STING activation, in vitro experiments were conducted in primary RTECs. RTECs were coincubated with eCIRP and H151, and IFN-β, the downstream protein in the STING pathway, was measured to quantify the progression of STING-induced inflammation. After eCIRP treatment for 24 h, both IFN-β mRNA and protein expression were increased by 1.5- and 3-fold, respectively, in contrast to nontreated cells, but these increases were significantly attenuated by H151 (Fig. 1, A and B). To understand the impact of H151 on eCIRP-triggered cell inflammation and injury, we measured the expression of IL-6, tumor necrosis factor (TNF)-α, and NGAL from the cell supernatant. Although expression of the proinflammatory cytokines IL-6 and TNF-α increased by 3.5- and 3-fold, respectively, in the eCIRP treatment group, expression of these proteins was significantly decreased in the eCIRP and H151 coincubation group from the eCIRP-alone group (Fig. 1, C and D). Furthermore, mRNA expression of one of the initial kidney cell injury indicators, NGAL, increased in the eCIRP-alone group, but NGAL was decreased in the eCIRP and H151 coincubation group relative to the eCIRP-alone group (Fig. 1E). Together, these findings illustrated that H151 effectively inhibited eCIRP-induced STING activation and alleviated cell damage in vitro.

Figure 1.

H151 inhibited extracellular cold-inducible RNA-binding protein (eCIRP)-induced STING activation in renal tubular epithelial cells (RTECs). RTECs were coincubated with eCIRP (1 μg/mL) and various doses of H151 for 24 h. Both cells and the supernatant were then harvested for analysis. A: mRNA levels of interferon (IFN)-β were measured by quantitative RT-PCR and normalized to β-actin (n = 5–6/group). B: IFN-β levels in the RTEC supernatant were determined by ELISA (n = 13/group). Interleukin (IL)-6 (C) and tumor necrosis factor (TNF)-α (D) levels in the RTEC supernatant were determined by ELISA (n = 8/group). E: mRNA levels of neutrophil gelatinase-associated lipocalin (NGAL) were determined by quantitative RT-PCR and normalized to β-actin (n = 6/group). *P < 0.05 vs. control; #P < 0.05 vs. the eCIRP-treated group.

H151 Improved Renal Function After RIR Injury

RIR leads to tissue damage especially in glomerular and renal tubular cells, resulting in kidney dysfunction. As the primary indicator to quantify renal function, GFR in RIR mice was measured. In contrast to sham mice, GFR significantly dropped by 52% in RIR-vehicle mice. However, GFR was either unchanged or near the sham level in RIR mice that received H151 treatment (Fig. 2A). To verify the effect of H151 on renal function restoration, serum levels of BUN, creatinine, and NGAL were measured. BUN and creatinine increased by 4.5- and 3-fold, respectively, in RIR-vehicle mice, whereas in RIR-H151 mice, these levels effectively decreased by 49% and 53%, respectively, relative to RIR-vehicle mice (Fig. 2, B and C). NGAL levels soared up to 67-fold in RIR-vehicle mice, whereas with RIR-H151 treatment, the NGAL level dropped by 25% from RIR-vehicle mice (Fig. 2D).

Figure 2.

H151 improved renal function. A: at 24 h after renal ischemia-reperfusion, mice were subjected to transcutaneous glomerular filtration rate (GFR) measurement (n = 5–8/group). Serum collected at 24 h after renal ischemia-reperfusion was analyzed by specific assay kits for blood urea nitrogen (BUN; B), creatinine (C), and neutrophil gelatinase-associated lipocalin (NGAL; D) levels (n = 6–8/group). *P < 0.05 vs. sham; #P < 0.05 vs. the vehicle (Veh) group.

H151 Attenuated Renal Injury and Inflammation and STING Activation After RIR Injury

As specific and sensitive kidney injury biomarkers, NGAL and KIM-1 were evaluated from kidney tissues. In contrast to sham mice, there were 1,426- and 465-fold increases in NGAL and KIM-1 mRNA, respectively, in RIR-vehicle mice. In the RIR-H151 treatment group, mRNA expressions of NGAL and KIM-1 significantly decreased by 55% and 58%, respectively, from RIR-vehicle (Fig. 3, A and B). To assess the effect of H151 on RIR-induced renal inflammation, we assessed MIP-2, KC, and IL-6 mRNA levels and IL-6 protein levels. Compared with sham mice, MIP-2 and KC mRNA were increased by 18- and 43-fold, respectively, in RIR-vehicle mice. In the RIR-H151 treatment group, MIP-2 and KC expression levels decreased by 67% and 70%, respectively, relative to RIR-vehicle mice (Fig. 3, C and D). Compared with sham mice, renal IL-6 mRNA and protein were markedly increased by 29- and 1.65-fold, respectively, in RIR-vehicle mice. In the RIR-H151 treatment group, these values were significantly decreased by 95% and 30%, respectively, from RIR-vehicle mice (Fig. 3, E and F). Compared with sham mice, renal IFN-β mRNA was increased by 6.5-fold in RIR-vehicle mice. In the RIR-H151 treatment group, IFN-β mRNA was significantly decreased by 63% from RIR-vehicle mice (Fig. 3G). Next, we performed histological analysis of kidney tissue after staining with H&E, and the tissue damage was scored. Unlike sham mice, RIR-vehicle mice showed increased tubular cell detachment, tubular dilation, cast formation, and tubular cell injury. In the RIR-H151 treatment group, these parameters were markedly reduced relative to the RIR-vehicle group (Fig. 4A). As expected, a significantly higher score was calculated in RIR-vehicle mice relative to sham mice, whereas the score decreased by 35% in RIR-H151 mice relative to RIR-vehicle mice (Fig. 4B). Apoptosis after RIR injury was also assessed using TUNEL staining of kidney tissue sections. Green fluorescence indicative of TUNEL staining was increased in RIR-vehicle mice in contrast to sham mice. This TUNEL staining was markedly reduced in the RIR-H151 treatment group relative to the RIR-vehicle treatment group (Fig. 4C). Relative to the sham group, the RIR-vehicle treatment group showed a 102-fold increase in TUNEL-positive cells, whereas in the RIR-H151 treatment group, the count was decreased by 87% relative to the RIR-vehicle treatment group (Fig. 4D). Collectively, these data demonstrated the effectiveness of H151 on attenuating the short-term impact of RIR-induced AKI.

Figure 3.

H151 reduced renal injury and inflammation. mRNA levels of neutrophil gelatinase-associated lipocalin (NGAL; A) and kidney injury molecule (KIM)-1 (B) were measured by quantitative RT-PCR and normalized to β-actin (n = 7–8/group). mRNA levels of macrophage inflammatory protein (MIP)-2 (C) and keratinocyte chemoattractant (KC; D) were measured by quantitative RT-PCR and normalized to β-actin (n = 7–8/group). *P < 0.05 vs. sham; #P < 0.05 vs. the vehicle (Veh) group. E: mRNA levels of interleukin (IL)-6 were measured by quantitative RT-PCR and normalized to β-actin (n = 7–8/group). F: IL-6 levels in kidney tissue were determined by ELISA (n = 7–11/group). G: mRNA levels of interferon (IFN)-β were measured by quantitative RT-PCR and normalized to β-actin (n = 5/group). *P < 0.05 vs. sham; #P < 0.05 vs. the Veh group.

Figure 4.

H151 improved renal histology and attenuated apoptosis. A and B: kidney injury was assessed by histological analysis. Tissue from each group was stained with hematoxylin and eosin, observed under light microscopy, and scored at ×200 magnification (n = 6/group). C and D: kidney cell apoptosis was evaluated by TUNEL staining (n = 6/group). TUNEL-positive cells were quantified at ×200 magnification by ImageJ software. *P < 0.05 vs. sham; #P < 0.05 vs. the vehicle (Veh) group.

H151 Improved Survival After RIR Injury

Since H151 attenuated renal injury and improved renal function was observed within 24 h, we examined its effect in a 10-day survival study to explore its long-term impact in RIR-induced AKI. Mice underwent bilateral ischemia for 34 min followed by reperfusion and H151 treatment. They were then monitored for 10 days. RIR mice treated with vehicle had a survival rate of 32%, whereas mice treated with H151 had a significantly increased survival rate of 63% (Fig. 5). Of note, sham mice were not included in the survival study because those mice were less likely to die within the 10-day observation period. These findings further confirmed the protective role of H151 in RIR-induced AKI.

Figure 5.

H151 improved survival in renal ischemia-reperfusion-induced acute kidney injury. After bilateral kidney ischemia for 34 min and subsequent reperfusion, mice were treated with either vehicle or H151 and observed for 10 days, and survival rates were plotted (n = 19/group). *P < 0.05 vs. the vehicle group.

DISCUSSION

In the present study, we showed that eCIRP-induced IFN-β mRNA and protein expression in RTECs were dose dependently inhibited by H151 treatment. We also demonstrated that eCIRP-induced proinflammatory cytokines IL-6 and TNF-α, as well as the injury indicator NGAL, were attenuated by H151 in a dose-dependent manner. These findings suggest the involvement of eCIRP and the STING pathway in these cells. We have previously shown that eCIRP activates the cGAS-STING pathway in hemorrhagic shock (8). eCIRP is released into the circulation during trauma hemorrhage, and the released eCIRP induces inflammation in macrophages via the Toll-like receptor 4 (TLR4) signaling pathway (9). A subsequent study demonstrated that eCIRP released during trauma hemorrhage induces mitochondrial (mt)DNA fragmentation via TLR4 signaling (19). Besides exogenous DNA and nuclear DNA, DNA from damaged mitochondria also acts as a potent activator of the STING pathway (20). Mitochondrial dysfunction is considered as an early event in AKI, and it contributes to renal tubular injury. Both animal and human studies have shown that mtDNA released from damaged mitochondria is indeed a marker for renal dysfunction and disease progression in AKI (21–23). A recent study indicated that mitochondrial damage leads to escape of mtDNA into the cytosol of renal cells, which activates the cytosolic cGAS-STING DNA sensing pathway (24). Mitochondrial stress can also release mtDNA into the cytoplasm, which has been shown to activate the cGAS-STING pathway and induce inflammation and tissue damage (5, 6, 25). A recent study reported that receptor-interacting serine/threonine-protein kinase 3 translocation from the cytosol to mitochondria causes the release of mtDNA and subsequently activates the cGAS-STING pathway in RIR injury (26). In RTECs of AKI mice, the mitochondrial membrane potential decreases and mtDNA gets released into the cytoplasm (5). RTECs are considered the most vulnerable cells during RIR (5). Therefore, the mechanism by which eCIRP causes STING activation could very well involve RIR-induced mitochondrial stress in RTECs and the subsequent release of mtDNA into the cytoplasm, which, in turn, activates the STING pathway leading to subsequent inflammation. Additional studies are needed to confirm this, however.

In the present study, we also examined the in vivo effect of a novel STING inhibitor, H151, on a 24-h mouse model of RIR-induced AKI. Our study indicated that while RIR caused a significant decrease in GFR, renal function was either unchanged or remained near normal levels with H151 treatment. GFR and serum levels of creatinine and BUN are a few of the accepted clinical measurements for evaluating renal function and monitoring kidney disease progression (27, 28). Compared with the RIR-vehicle group, the surrogate kidney injury markers present in the circulation such as BUN, creatinine, and NGAL were significantly attenuated after RIR-H151 treatment. In addition to the systemic levels, relative to those in RIR-vehicle group, kidney mRNA expression of NGAL and KIM-1 was also markedly decreased with RIR-H151 treatment, suggesting that H151 attenuated local injury in the kidneys. Likewise, proinflammatory chemokines (MIP-2 and KC) and the proinflammatory cytokine IL-6 from the kidneys were also attenuated with RIR-H151 treatment relative to the RIR-vehicle group, suggesting protection from renal inflammation. As a result of RIR-induced inflammation, renal tubular cells generally become seriously injured and undergo apoptosis (29, 30). Relative to the RIR-vehicle group, RIR-H151 treatment also improved histology and significantly attenuated apoptotic cell death. These data clearly demonstrated that H151 protects the kidneys at least for 24 h from RIR-induced AKI. AKI could persist not only for hours but even for several days. Thus, it was important to determine the long-term effect of H151 in RIR-induced AKI. Therefore, we performed a 10-day survival study and demonstrated that H151 greatly improved the survival rate of mice with RIR injury from 32% to 63%. Thus, blockade of STING activity with H151 in vivo ameliorated renal inflammation, injury, and dysfunction and improved survival in RIR. Thus, our study showed the role of eCIRP signaling in the cGAS-STING pathway to promote inflammation and injury in RIR-induced AKI and that the attenuation of eCIRP-induced STING activation in RTECs by H151 contributes in part to the protective effect of H151 in RIR-induced AKI.

H151 is a selective small-molecule antagonist of STING that was identified based on a human cell screening system (11). In fact, another STING inhibitor, C-176, has shown beneficial effects in AKI by reducing renal inflammation, tubular injury, and renal dysfunction (5). C-176 is a covalent antagonist of STING that had the same covalent site as H151. However, C-176 binds to mouse STING with a high affinity but not to human STING, which limits its potential for clinical applications. With a recent study that showed that H151 treatment attenuating cisplatin-induced AKI in mice (14) and the results of our present study in RIR-induced AKI, we can suggest the possibility of H151 for becoming a therapeutic intervention for AKI.

Additional experiments including the H151-only treatment group to obtain cell viability data and to determine whether eCIRP effects do or do not involve TLR4 could have strengthened the cell culture experiments. Due to the low yield of cells and the challenges involved in primary culture isolation, we were not able to include those experiments in our study. However, others have shown that 1 µM H151 showed no significant increase in cytokine or chemokine levels in THP-1 cells (31). Also, we have previously shown that treatment with recombinant murine CIRP (rmCIRP) in TLR4-deficient (TLR4^−/−) macrophages lost the response for TNF-α production. In addition, rmCIRP injected into wild-type mice showed increases in serum cytokine levels (TNF-α, IL-6, and high mobility group box 1) and organ injury markers (aspartate transaminase and alanine transaminase), whereas these deleterious effects were not observed in rmCIRP-treated TLR4^−/− mice (9). Recent studies in our laboratory identified triggering receptor expressed on myeloid cells 1 (TREM-1) as a receptor also for eCIRP (32). These prior studies implied that eCIRP signaling in RTECs could be mediated via the TLR4 or TREM-1 receptor. However, additional experiments are required for confirmation.

Our research provides solid evidence to support the protective role of H151 on eCIRP-induced STING activation in RTECs in vitro and RIR-induced AKI in vivo; however, there are several limitations to our study. First, a study by Lee et al. (33) has shown that volatile anesthetics such as isoflurane resulted in lower plasma creatinine and reduced renal necrosis after RIR injury compared with injectable anesthetic such as pentobarbital or ketamine, suggesting that isoflurane confers protection against RIR injury. These protective effects by isoflurane were independent of the effects of isoflurane on blood flow or blood pressure (34). Since we used isoflurane anesthesia and did not measure any hemodynamic parameters, we consider the use of isoflurane as the anesthetic agent as a major limitation of our study. However, since our study objective was to investigate the effect of H151 on RIR-induced inflammation and apoptosis, our results clearly demonstrated that H151 protected the kidneys from I/R injury. Second, RIR injury can result in multiorgan system inflammation and injury. It would have been interesting to examine the extrarenal effects of H151. Since we did not collect any additional organs and therefore could not examine these effects, we consider this as a limitation of our study.

Third, we did not analyze the time course of STING activation in the RIR mice model, and we did not use additional time points including delayed treatment in our study. Since a substantial increase in either CIRP mRNA expression in the kidneys or eCIRP levels in the serum was only observed after 5 h post-RIR (10), factors other than eCIRP could cause mtDNA release during RIR and cause STING activation. Therefore, the benefit observed by H151 may not be limited to downregulation of eCIRP signaling. Thus, it is unlikely that delayed treatment with H151 could be more effective than treatment given at the time of reperfusion. Also, additional time points beyond 24 h used in our study could address the role of the immune cells such as inflammatory cell infiltration in this RIR-induced AKI. Fourth, we used only a single dose in our study. Most parameters tested in our study showed more than 50% inhibition with H151 treatment compared with the vehicle-treated group of mice. Furthermore, 63% of RIR mice treated with H151 survived the 10-day observational study, whereas there was only 25% survival with vehicle treatment. It is highly possible that either increasing or multiple dosing of H151 could have resulted in better protection. Furthermore, although we have not conducted any half-life studies for this compound, it has been previously reported in the literature that after intraperitoneal administration of H151, systemic levels were detected at 30 min, and that they reached time 0 levels by 2 h, indicating a relatively short plasma half-life for the compound (11). Mechanistically, H151 covalently targets STING, and therefore it is likely that the tissue distribution half-life of H151 could be longer than the plasma half-life. Since our study is only a proof-of-concept study, we used only a single dose of H151 and treatment only at the time of reperfusion based on our previous study in intestinal ischemia-reperfusion injury (15). Additional treatment protocols including absorption, distribution, metabolism, and elimination studies are required to optimize the efficacy of H151 for the development of this compound as a therapeutic for RIR-induced AKI.

In addition, as a small molecule, there is a chance that H151 may hit some unknown off-targets even though it has been reported to selectively bind a single cysteine residue to prevent STING-mediated activity (11). Studies involving STING knockout mice in vivo could confirm the specificity of H151 on STING-dependent kidney injury in AKI. In this regard, we have previously shown that STING knockout mice were beneficial in a trauma hemorrhage model (8). Therefore, the specificity of H151 can be confirmed using STING knockout mice in the future. Also, our study was performed on mouse AKI models and mouse-originated RTECs. Even though H151 is a special antagonist for human and murine STING (11), there could be a differential response to H151 treatment in different species. Additional research is needed to verify the potential pharmaceutical application. Furthermore, our research was performed on male animals; however, there is sexual dimorphism in the development of AKI (35, 36). Whether the conclusion of the male research is applicable to both sexes still needs further exploration.

Although we did not address the issue of eCIRP release in this study, we have previously demonstrated that eCIRP is released from immune cells such as macrophages exposed to hypoxia and reperfusion (9). During RIR-induced AKI, ischemia-reperfusion injury could increase the release eCIRP from immune cells and causes mitochondrial stress in RTECs presumably via the TLR or TREM-1 signaling pathway and triggers mtDNA release into the cytoplasm and activates the STING pathway. STING activation then leads to the release of IFN-β, proinflammatory cytokines (IL-6), and chemokines (MIP-2 and KC) promoting renal dysfunction and injury, which ultimately results in AKI.

Perspectives and Significance

STING is the cytosolic DNA-activated signaling pathway that mediates inflammation and injury. Our recent study showed that eCIRP activates STING and exacerbates hemorrhagic shock-induced lung injury. eCIRP is a damage-associated molecular pattern identified in various injury conditions including RIR injury. H151 is a small molecule that selectively binds to STING and inhibits STING-mediated activity. H151 inhibited eCIRP-induced STING activation in RTECs. H151 administration in vivo effectively attenuates renal dysfunction or even restores kidney function, but the inhibition may not be exclusive to eCIRP-induced STING activation in RTECs. Nevertheless, H151 could be developed as a novel therapeutic for RIR-induced AKI.

DATA AVAILABILITY

Data will be made available upon reasonable request.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grant R01HL076179 (to P.W.).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

M.B., A.J., and P.W. conceived and designed research conceived and designed research; Z.H. and F.Z. performed experiments; Z.H., F.Z., and A.J. analyzed data; Z.H., A.J., and P.W. interpreted results of experiments; Z.H., F.Z., and A.J. prepared figures; Z.H. and A.J. drafted manuscript; M.B., A.J., and P.W. edited and revised manuscript; Z.H., F.Z., M.B., A.J., and P.W. approved final version of manuscript.