NAD+ ameliorates endotoxin‐induced acute kidney injury in a sirtuin1–dependent manner via GSK‐3β/Nrf2 signalling pathway

Abstract Acute kidney injury (AKI) is a substantial worldwide public health concern with no specific and effective therapies in clinic. NAD+ is a pivotal determinant of cellular energy metabolism involved in the progression of AKI; however, its mechanism in kidney injury remains poorly understood. Sirtuin 1 (SIRT1) is an NAD+‐dependent deacetylase associated with renal protection and acute stress resistance. In this study, we have investigated the role of NAD+ in AKI and the potential mechanism(s) involved in its renoprotective effect. NAD+ was notably decreased and negatively correlated with kidney dysfunction in AKI, restoring NAD+ with NMN significantly ameliorates LPS‐induced oxidative stress and apoptosis and attenuates renal damage. We also found that the protection of NAD+ is associated with SIRT1 expressions and performs in a SIRT1‐dependent manner. Inhibition of SIRT1 blunted the protective effect of NAD+ and up‐regulated the activity of glycogen synthase kinase‐3β (GSK‐3β) that was concomitant with mitigated Nrf2 nuclear accumulation, thereby exacerbates AKI. These findings suggest that NAD+/SIRT1/GSK‐3β/Nrf2 axis is an important mechanism that can protect against AKI which might be a potential therapeutic target for the treatment of AKI.

also an urgent need to discover new renoprotective therapies for improving clinical outcomes in patients with AKI.
NAD + (nicotinamide adenine dinucleotide) is a ubiquitous hydride acceptor that plays an essential role in energy metabolism and adaptive stress responses. 8,9 The kidney is both among the highest mitochondria and NAD + containing organs, yet is also highly susceptible to NAD + depletion. 10,11 Oxidative stresses and mitochondrial damage are drivers of sepsis-induced AKI, followed by cellular NAD + depletion and NAD + synthesis reduction. 10,12,13 Accumulation studies suggest that elevating intracellular NAD + levels, including regulation of critical enzymes in NAD + biosynthesis and exogenous supplementation of its precursors, are potential therapeutic strategies for renal diseases in model organisms ranging from mice to humans. [14][15][16][17] However, the mechanisms by which effectiveness remains incompletely understood.
Sirtuin 1 (SIRT1) is an NAD + -consuming deacetylase associated with cytoprotective effects exerted by the inhibition of inflammasome activation, deacetylation of downstream proteins (such as p53, nuclear factor-κB (NF-κB), forkhead box protein (Fox) and peroxisome proliferator-activated receptorγ co-activator-1α (PGC-1α)) and induction catalase activity, which lead to anti-inflammatory, anti-apoptotic and antioxidant effects. [18][19][20] Studies have demonstrated that SIRT1 activation attenuates renal oxidative stress and apoptosis and improves the decline of renal functions. [21][22][23] Therefore, SIRT1 might be a therapeutic agent for the treatment of renal disorders. It has been well documented that nicotinamide mononucleotide (NMN), an NAD + precursor, rescues age-related susceptibility to cisplatin-induced AKI depended on SIRT1 which contributes to the resistance of aging. 19,24 Thus, it seemed reasonable to hypothesize that the effect of NAD + on LPS-induced AKI is associated with SIRT1.
In this study, we examined that the role of NAD + /SIRT1 in sepsis-induced AKI. Our study suggests that NAD + supplementation contributed to rescued the renal damage and its deficits increased vulnerability to sepsis-induced AKI, which performed in a SIRT1dependent manner. Furthermore, the mechanism by which NAD + / SIRT1 alleviates the sepsis-induced AKI involves the regulation of glycogen synthase kinase-3β (GSK-3β) / nuclear factor erythroid 2-related factor 2 (Nrf2) signalling pathway. The study, therefore, sheds light on the mechanisms underlying the renoprotective capacity of NAD + and identifies NAD + /SIRT1/GSK-3β/Nrf2 pathway as a potential therapeutic target, which might be used to prevent or ameliorate sepsis-induced AKI.

| Animals
Wild-type C57BL/6 mice (aged 6~8 weeks, weighting 20~22 g, male) were purchased from Vital River. NMNAT1 (the critical enzyme for NAD + salvage biosynthetic pathway) conditional-knockout ( −/− ) mice on a C57BL/6 background (NMNAT1 flox R26 CreERT2 ) were purchased by Beijing Biocytogen Co., Ltd, in which the NMNAT1 gene was induced by tamoxifen (Sigma) dissolved in corn oil at a concentration of 10 mg/ml, which injected intraperitoneally once a day for a total of five consecutive days and waited for one week. All mice were housed at 25°C with 12 h light and dark photocycle, and they were free access to food and water ad libitum throughout the study period. The study was performed in accordance with the legislation of laboratory animals and approved by the Animal Care and Use Committee of the Tianjin Nankai Hospital (Approval No. NKYY-DWLL-2019-012, Tianjin, China).

| Experimental protocol
To evaluate the effect of NAD + on LPS-induced AKI, wild-type and NMNAT1 −/− mice were assigned to four groups, respectively: control (CON); LPS-treated mice (LPS); LPS plus NMN-treated mice (LPS + NMN) and NMN-treated mice (NMN). All mice were anaesthetized with 2%~3% isoflurane (R510-22, RWD Life Science) inhalation for induction and 1.5% isoflurane for maintenance. Wild-type and NMNAT1 −/− mice were used in the LPS-induced AKI model by caudal vein injection of LPS (E.coli-L2630, Sigma) 15 mg/kg diluted in 2 ml saline as described previously. 25 In addition, the mice were intraperitoneally injected with 500 mg/kg NMN (M3501, Sigma, USA) for 7 consecutive days prior to LPS treatment. To explore the effect of SIRT1 on LPS-induced AKI and the involved GSK-3β/Nrf2 signalling pathway, wild-type C57BL/6 mice were divided into four groups: control (CON); LPS-treated mice (LPS); LPS plus SIRT1 selective agonist (SRT-1720)-treated mice (LPS + SRT-1720) and LPS plus SIRT1 specific inhibitor-treated mice (LPS + EX-527). Furthermore, to determine the effect of NAD + on AKI depends on SIRT1, LPS plus NMN-treated mice with or without EX-527: LPS + NMN group; LPS + NMN + EX-527 group, mice were injected with SIRT1 agonist SRT-1720 (30 mg/kg, SC0267, Beyotime) or SIRT1 inhibitor EX-527 (5 mg/kg, SC0281, Beyotime) intraperitoneally before LPS injection as described before. 26,27 Twelve hours after LPS treatment, all the mice were sacrificed under deep anaesthesia via cervical dislocation. Blood samples were collected through cardiac puncture and centrifuged 3000 g for 10 min at 4℃, and the isolated serum was stored at −80℃. Kidney tissues were fixed in 4% paraformaldehyde for pathological analyses, and remaining kidney tissues were stored at −80℃ for further analyses.

| Haematoxylin and eosin staining for renal pathology
To evaluate renal histopathology changes, kidney tissue was immediately collected and fixed overnight with 4% paraformaldehyde, embedded in paraffin, and subsequently sectioned at a 4 μm thickness for further haematoxylin and eosin staining according to standard procedures. A minimum of 10 fields on each slide were examined and scored for pathological injury by a pathologist in a blinded fashion. The tubular injury score from 0 to 4 was given: 0, normal; 1, mild injury (5%~25% of tubules were damaged); 2, moderate injury (25%~50% of tubules were damaged); 3, severe injury (50%~70% tubules were damaged and 4, almost all tubules in the field were damaged), which was based on a standard previously reported. 28,29 Images were acquired with Leica DM4000B microscopy (Leica).

| Assessment of kidney function
Blood urea nitrogen (BUN) and serum creatinine (Scr) were assayed as indicators of renal functions, and kidney injury molecule-1 (KIM-1) and neutrophil-gelatinase-associated lipocalin (NGAL) have been identified as renal injury biomarkers to indicate AKI. 30 The serum collected was measured for BUN and Scr levels using AutoAnalyzer (Roche Diagnostics) and KIM-1 and NGAL levels using enzyme-linked immunosorbent assay (enzyme-linked Biotechnology Co., Ltd).

| NAD + measurement
NAD + concentrations from the kidney were measured with an NAD/ NADH Assay Kit (KA1657, Abnova) according to the previous report. 31 The assay is based on the lactate dehydrogenase cycling reaction, in which the formed NADH reduces a formazan (MTT) reagent. Frozen kidney tissue (20 mg) was prepared for homogenizing samples and transferred into a 1.5 ml Eppendorf tube with either 100 μl NAD or NADH extraction buffer for NAD and NADH determination respectively. The extracts were heated at 60℃ for 5 min, followed by 20 μl assay buffer and 100μl the opposite extraction buffer to neutralize the solutions. After centrifugation at 14,000 g for 5 min, the supernatant was obtained to determine the concentrations of NAD and NADH. The NAD + content was calculated as the total NAD content minus the NADH content.

| Measurement of mitochondrial DNA content
Mitochondrial DNA (mtDNA) is tightly associated with mitochondrial enzyme activity and ATP production and is therefore indicated as a biomarker of mitochondrial function. 32 The mtDNA concentrations were measured by RT-PCR as previously described. 33 Briefly, total RNA from kidneys was isolated by a RNeasy Mini Kit

| Detection of mitochondrial ROS
A mitochondria isolation kit (Thermo Scientific) was used for mitochondria fraction preparation. Mitochondrial ROS was measured by for 30 min at room temperature, then, washed three times with PBS and collected in a glass slide containing PBS. The fluorescence signal was measured, respectively, at excitation and emission wavelength of 488 nm and 520 nm by a microplate reader. The fluorescent images were assessed using Image Pro-Plus software and the results for CON (WT) group were expressed as a relative reference.

| Oxygen consumption rate
The mitochondrial respiration was examined by analysing the mi- Observations were performed on a JEM-1230 transmission electron microscope (JEOL, Japan) at 80 kV.

| TUNEL assay
The cell death was evaluated by a TdT-mediated dUTP nick-endlabelling (TUNEL) in situ cell death detection kit (KGA7072, keyGEN BioTECH) that followed the manufacturer's protocols. Detection of the apoptotic cells exhibiting green fluorescence was quantified by Leica DM4000B microscopy and represented as percentages of TUNEL-positive cells out of total cells in the fields.

| Immunostaining
Preparation of paraffin tissue sections was performed as described The relative mRNA levels for the specific genes were normalized to GAPDH mRNA and calculated by the 2 −ΔΔC t method.

| Western blot analysis
The Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime) was used for cytoplasmic (for total Nrf2) and nuclear proteins (for nuclear Nrf2 and Histone H3) extraction. Total proteins were extracted from kidneys using a total protein isolation kit (Thermo Fischer Scientific) and its concentrations were determined by the BCA protein assay kit (Sigma). Equal amounts of protein were separated on 10% SDS-PAGE gel and then were transferred to a PVDF membrane

| Statistical analysis
All values were expressed as mean ±SD and Student's t test or one-

| NAD + supplementation ameliorates LPSinduced AKI whereas its deficiency aggravates kidney injury
Substantial evidence indicates that NAD + is a ubiquitous coenzyme that functions as a guardian against oxidative stress. 35

| NAD + attenuates LPS-driven apoptosis in renal tissues.
Apoptosis is well-described as a feature of LPS-induced AKI. 37 To investigate the effects of NAD + on LPS-triggered renal apoptosis, we performed terminal deoxynucleotidyltransferase-mediated dUTP nick-end-labelling (TUNEL) staining on different conditions. We found more TUNEL-positive cells in LPS-treated wild-type and NMNAT1 −/− mice relative to controls, respectively, and fewer TUNEL-positive cells were shown in NMN supplementation (Figure 2A

| NAD + protected against mitochondrial oxidative damage and improved mitochondrial morphology in LPS-induced AKI
Oxidative stress is confirmed as a major hallmark of LPS-induced AKI and is characterized by the accumulation of reactive oxygen species (ROS), impaired antioxidant capability as well as mitochondrial dysfunction. 38,39 As shown in Figure 3, exposure to LPS both in WT and KO mice showed a striking increase in the levels of mitochondrial ROS and inducible nitric oxide synthase (iNOS) contents while a marked reduction of mtDNA and OCR, which were significantly suppressed by pre-processing with NMN. Correspondingly, the normal nuclear and mitochondrial morphology were shown in control groups, whereas LPS caused mitochondria vacuolization and nuclear karyopyknosis, and NMN pretreatment dramatically improved the ultrastructural damage.
In addition, compared with CON (WT) group, mtDNA and OCR

| The protection of NAD + is associated with SIRT1 expression
SIRT1 is a highly conserved NAD + -dependent deacetylase known to have protective effects against a variety of renal diseases. 22,23,40 To elucidate whether the protection of NAD + in LPS-induced AKI is positively associated with the expression of SIRT1, we firstly observed the changes of SIRT1 in different concentrations regimes of NAD + .
As shown in Figure 4, an apparent reduction of SIRT1 expression was found in kidneys from NMNAT1 −/− mice relative to wild-type. LPS treatment remarkably decreased the mRNA and protein expressions of SIRT1 both in wild-type and NMNAT1 −/− mice, the results were consistent with the previous study. 41,42 Additionally, the expressions of SIRT1 were restored by NMN supplementation, confirming that SIRT1 levels were responsive to intracellular NAD + concentration.
Briefly, the tendency of SIRT1 was entirely similar for NAD + , and it was reasonable to believe that these two events interact which indicated the protection of NAD + is correlated with SIRT1 expression.

| SIRT1 protects against LPS-induced renal dysfunction and apoptosis
To directly observe the role of SIRT1 in the LPS-driven AKI model,

| The effect of NAD + on the kidney depends on SIRT1
To explore whether the protective effect of NAD + on sepsis-induced AKI depends on SIRT1, studies were performed to examine the effect of NMN on SIRT1 inhibitor (EX-527) under LPS stimulation.
As shown in Figure 6, the therapeutic effect of NMN on AKI was substantially diminished by EX-527, which exhibited significantly higher levels of BUN, Scr, KIM-1 and NGAL with severe renal damage ( Figure 6A-F). Consistently, the cellular apoptosis assay also supported that the NAD + rescued the LPS-induced AKI in a SIRT1dependent manner and down-regulation of SIRT1 exacerbated the kidney injury ( Figure 6G-H).

| SIRT1 modulates the GSK-3β/Nrf2 signalling pathway is LPS-induced AKI
To further investigate the underlying specifically regulation of SIRT1 on sepsis-induced AKI, we sought to explore molecules that mediate this progress. Glycogen synthase kinase-3β (GSK-3β) is known to be ubiquitously expressed in kidneys and as a potential target for relieving LPS-elicited inflammation and apoptosis. 40,43 The activity of GSK-3β is inhibited by phosphorylation at the serine 9 residue and represents as the value of 1-(p-GSK-3β/GSK-3β). 43 Burgeoning studies reported that GSK-3β is crucial for the regulation of NF-E2related factor (Nrf2) and its inactivation up-regulates the nuclear accumulation of Nrf2 for the antioxidant response. 44

| DISCUSS ION
Previous studies have demonstrated that NAD + precursors, such as nicotinamide mononucleotide (NMN), nicotinamide riboside (NR) and nicotinamide (NAM) attenuate acute renal injury by reducing oxidative stress, inflammation and improving renal functions. 15,24,45,46 These manners ultimately modulate the concentrations of NAD + and exert the critical role in protecting the kidney from injury. 46 This study was based on supplementation of NMN or knockdown the key biosynthetic enzyme NMNAT1 to regulate bi-directionally cellular NAD + levels and identified the NAD + -consuming deacetylase, SIRT1, as an important factor associated with the capacity of NAD + during LPS-induced AKI. Here, we showed that the AKI had decreased NAD + and SIRT1, and more significantly in NMNAT1 −/− mice, whereas the supplementation of NMN restored the contents of NAD + and SIRT1 and rescued LPS-induced acute renal damage.
Activation of SIRT1 greatly facilitated the renal protective effect of NAD + and inhibition of SIRT1 considerably weakened, suggesting that the protective action of NAD + relied on SIRT1. Thus, our study provided the rationale for the therapeutic target of NAD + for LPSinduced AKI.
Reduction of NAD + or NAD + biosynthetic impairment has been identified in acute kidney injury, 14,15,46 although the mechanisms have not been fully elucidated. As we know, there are three major pathways contributing to NAD + synthesis in which salvage pathway accounts for most of the NAD + in mammals. 17,47 Briefly, NAD + is converted from NMN by three isoforms of NMNATs, and NMNAT1 is ubiquitously expressed with highest enzyme activity. 17,48 Thus, deficit of NMNAT1 could contribute to the insufficient production of NAD + and administration of NMN effectively enhances total cellular NAD + levels. Studies have revealed that renal tubular cells are abundant with mitochondria which highly rely on the NAD homeostasis and are the primary target organelle in AKI. 29 Mitochondrial dysfunction is not only a common pathologically consequence but also a pathogenic factor, resulting in a vicious cycle that drives the aggravation of AKI, 49 GSK-3β is highly expressed in kidneys and is critical for modulating the self-defensive response after oxidative stress by promoting Nrf2 nuclear exit and degradation, thus regulating the Nrf2-mediated antioxidant response. 44,53-55 As a key redox-sensitive player in oxidative stress, GSK-3β kinase activity may be amplified by inhibiting phosphorylation at serine 9, which promotes GSK-3β accumulation and impairs Nrf2-mediated antioxidant response, driving the exacerbation of oxidative injury. 56,57 Considering the fact that SIRT1 signalling has been associated with cellular antioxidant defense, 21 we hypothesized that the SIRT1 protective effect is mediated by GSK-3β/Nrf2. In agreement, this study showed that GSK-3β was observably overactive following LPS-induced AKI, concomitant with a self-defensive increase of the Nrf2 nuclear accumulation. The activity of GSK-3β was inhibited by SRT-1720, the agonist of SIRT1, and the nuclear accumulation of Nrf2 was apparently increased, while EX-527 exerts the opposite effects. These results indicated that GSK-3β/Nrf2 may be involved in the process of SIRT1 attenuating AKI.
Although we examined that NAD + levels were negatively correlated with the severity of AKI and the protective effect of NAD + relied on SIRT1, but not yet been validated clinically.
Therefore, the effectiveness and mechanisms of NAD + in protecting against AKI remain to be further discovered by a clinical trial in the future.
In conclusion, the study demonstrated that LPS-induced renal damage is negatively associated with NAD + contents and the expression of SIRT1, and restoring NAD + markedly attenuated AKI possibly by ameliorating mitochondrial dysfunction and apoptosis.
We also identified that the renoprotective effects of NAD + are exerted in a SIRT1-dependent manner by regulation of the GSK-3β/ Each result was replicated in three independent experiments, and the data are presented as mean ±SD (n = 3). *Significance compared with CON group (*p < 0.05, **p < 0.01). #Significance compared with LPS group ( # p < 0.05, ## p < 0.01) Nrf2 pathways. This study theoretically provides evidence for the protective effect of NAD + in sepsis-associated AKI, which may offer a viable approach for AKI therapy.

CO N FLI C T O F I NTE R E S T
The authors declare no competing interests. Project administration (equal); Supervision (equal).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.