NAD⁺-Boosting Ameliorates Heme Protein–Mediated Acute Kidney Injury

Visual Abstract

Abstract

Key Points

• Understanding how heme proteins/heme induce AKI and whether NAD⁺-boosting ameliorates heme protein–mediated AKI is clinically relevant.

• Heme proteins/heme directly decrease kidney NAD⁺ content and NAD⁺-boosting is markedly protective in heme protein–mediated AKI.

• These findings provide new insights in heme protein–mediated AKI and support interest in NAD⁺-boosting as a therapeutic strategy in AKI.

Background

Understanding how heme proteins and heme induce AKI is clinically relevant from numerous perspectives, and, in this regard, a widely used model of heme protein–mediated AKI (HP-AKI) involves the glycerol model in mice. Using this model, we have previously demonstrated that NAD⁺ content is decreased in HP-AKI. Because there is significant current interest in NAD⁺-boosting as a strategy in clinical AKI, we examined the effect of NAD⁺-boosting in this model.

Methods

NAD⁺-boosting was achieved by the administration of nicotinamide mononucleotide (NMN) in the glycerol model of HP-AKI. The effect of NMN, compared with vehicle, was examined in mice with HP-AKI and with sham AKI at day 1 after HP-AKI was induced.

Results

The administration of NMN preserved kidney NAD⁺ content in HP-AKI, and compared with the administration of vehicle in HP-AKI, NMN improved glomerular filtration markers; reduced histologic injury as assessed by tubular necrosis, dilation, and cast formation and by assessment of dystrophic calcification; reduced expression of kidney injury marker 1, a sensitive marker of AKI; preserved mitochondrial ultrastructure and increased expression of proteins that promote mitochondrial integrity; reduced apoptosis as assessed by terminal deoxynucleotidyl transferase–mediated digoxigenin-deoxyuridine nick-end labeling staining and genes and protein expression that contribute to apoptosis, while increasing expression of an anti-apoptotic gene; reduced expression of several renal injury-related genes; and mitigated the severity of the senescence phenotype, as assessed by multiple markers. Finally, the administration of myoglobin or heme in vivo diminished kidney NAD⁺ content in mice with intact kidneys.

Conclusions

These findings demonstrate the remarkable protective effects of NAD⁺-boosting by NMN in HP-AKI as revealed by a multitude of markers relevant to AKI. The reduction in NAD⁺ content in the kidney in HP-AKI may reflect the direct effects of heme proteins and/or heme. We suggest that these findings support current interest in NAD⁺-boosting as a therapeutic strategy in clinical AKI.

Introduction

Understanding how heme proteins and heme cause AKI is important for several reasons. First, some 7%–10% of cases of AKI arise from rhabdomyolysis and the attendant exposure of the kidney to myoglobin.¹ Second, sepsis-associated AKI accounts for 25%–50% of cases of AKI,² and, as is now well recognized, free extracellular hemoglobin from subclinical hemolysis during sepsis is commonly implicated in sepsis-associated AKI^3,4; less commonly, hemoglobin causes AKI as a consequence of red blood cell injury incurred by cardiopulmonary bypass and other extracorporeal devices, by thrombotic thrombocytopenic purpura/hemolytic uremic syndrome, and in assorted hemolytic conditions.^3,4 Third, even in conditions wherein exposure of the kidney to myoglobin or hemoglobin does not overtly occur, heme contributes to AKI. Specifically, in conditions such as ischemia-reperfusion injury⁵ and nephrotoxic cisplatin-induced AKI,⁶ intrarenal content of heme is increased and is implicated in AKI because of the pro-oxidant, proinflammatory, proapoptotic, and prosenescent effects of heme^4,7; increased renal content of free heme arises from destabilized/degraded cytochrome p450 and other intracellular heme proteins.^4,7

The glycerol model of heme protein-mediated AKI (HP-AKI), induced by the intramuscular administration of hypertonic glycerol, exposes the kidney to both myoglobin and hemoglobin, both of which contribute to AKI. This model has been long and widely used in the study of HP-AKI because it has provided mechanistic insights regarding the role of vasoconstriction induced by heme proteins⁸; mechanisms whereby heme proteins induce tubular injury⁹; the complicity of ACE2 in enabling renal uptake of heme¹⁰; the role of macrophages and inflammation¹¹ and the involvement of macrophage extracellular traps¹² in causing such injury; the rapid appearance of a senescence phenotype in HP-AKI¹³; tubular dysfunction in AKI¹⁴; adaptive protective responses such as renal induction of heme oxygenase-1¹⁵ and ferritin¹⁶; and therapeutic strategies including those based on the administration of antioxidants¹⁷ and agents that reduce isoprostane synthesis.¹⁸

In our prior studies using this model, we demonstrate that the kidney content of NAD⁺ is decreased.¹⁹ In models of ischemia reperfusion injury,²⁰ cisplatin-induced AKI,²¹ and sepsis-induced AKI,²² boosting NAD⁺ content has been shown to exert beneficial effects on AKI. Considerable interest surrounds the disruption of NAD⁺ synthesis that occurs in the acutely injured kidney, and, indeed, clinical trials are currently underway testing the efficacy of boosting NAD⁺ in human AKI.^23–25

Our prior observations demonstrating diminished kidney content of NAD⁺ in HP-AKI¹⁹ led us to conclude that this finding “…validates the investigation of novel, clinically relevant therapies such as NAD⁺-boosting agents… in HP-AKI.¹⁹”

The present study examined whether augmenting NAD⁺ content in the kidney protects against HP-AKI and, if so, how does such protection occur.

Methods

In Vivo Studies

All studies were approved by the Institutional Animal Care and Use Committee of Mayo Clinic and performed in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. We used male C57BL6/J mice (10–15 weeks old) purchased from Jackson Lab (Bar Harbor, ME). In all studies, mice were age-matched and weight-matched for random assignment to control or experimental groups, accordingly.

HP-AKI Model and NAD⁺ Boosting by Nicotinamide Mononucleotide

Mice were subjected to the glycerol model of HP-AKI as we described previously.^13,15 Briefly, after overnight dehydration (approximately 18 hours), mice were given an intramuscular injection of glycerol (50% in water, 6 ml/kg) under ketamine and xylazine anesthesia (IP, 90 and 10 mg/kg for ketamine and xylazine, respectively). The NAD⁺ precursor nicotinamide mononucleotide (NMN; 500 mg/kg body wt, IP, catalog no. A315149, AmBeed, Arlington Hts., IL) was administered daily starting 2 days before administration of glycerol and continuing until day 1. NMN was prepared as a 50 mg/ml solution in isotonic saline containing tris buffer (50 mM, pH 7.0). One day after glycerol injection, mice were killed, and kidney tissue and plasma were collected for analysis as outlined below. Kidney function was assessed by the measurement of serum creatinine and BUN levels as previously described.¹³

Administration of Myoglobin or Hemin

Myoglobin (250 mg/100 g body wt, intravenous, catalog no. M0630, Sigma Aldrich, St. Louis, MO) was infused through the tail vein in C57BL/6J mice, after 16–18 hours of dehydration. In additional studies, hemin (hemin ferriprotoporphyrin IX chloride, 50 µmol/kg, IP) was administered to C57BL/6J mice at 6 and 24 hours before kidneys were harvested for gene expression studies.^10,13

Histological Studies

Histologic examination was performed on formalin-fixed, paraffin-embedded kidney sections stained with hematoxylin and eosin. Semiquantitative scoring was used to assess histologic alterations including necrosis, cast formation, tubule dilation, corticomedullary injury, and extension of injury into the cortex.^10,13

Immunofluorescence Staining

Immunofluorescence staining was performed on 8 μm sections of formalin-fixed, paraffin-embedded kidney tissues as previously described.^10,19 Briefly, slides were deparaffinized, and antigen retrieval with acidic citrate buffer (pH 6.0) was performed. Slides were blocked (5% normal donkey serum, 5% BSA in 0.1% Triton X-100, PBS) for 2 hours and incubated with primary antibodies, lamin B1 and kidney injury marker 1 (KIM-1) at 4°C overnight. After washing, secondary antibody incubation was performed for 2 hours (next day at room temperature). Secondary antibodies used were Donkey anti-Rabbit AF555 and Donkey anti-Goat AF555 (catalog nos. A32794, and A32816, ThermoFisher Scientific). Washes using PBS+0.05% Triton X were performed between the incubations. Microscope images (Zeiss LSM980) were acquired using a 20× lens (NA 0.3), and images were prepared using Photoshop. All exposure levels were identical between groups. Supplemental Information provides information for antibodies used for these studies in tabular form.

Terminal Deoxynucleotidyl Transferase–Mediated Digoxigenin-Deoxyuridine Nick-End Labeling Staining

Terminal deoxynucleotidyl transferase–mediated digoxigenin-deoxyuridine nick-end labeling (TUNEL) staining was performed on 8 μm sections of formalin-fixed, paraffin-embedded kidney tissues using a Cell Signaling Kit (#64936, Cell Signaling, Danvers, MA), according to the manufacturer's instructions. Briefly, slides were deparaffinized and antigen retrieval with acidic citrate buffer (pH 6.0) was performed. Slides were preincubated with TUNEL equilibration buffer, and then, TUNEL reaction mix was added and incubated for 2 hours at 37°C. DAPI staining was performed after the incubation and slides mounted using VECTASHIELD Antifade mounting medium (Vector Laboratories, Newark, CA). Microscope images (Zeiss LSM980) were acquired using a 20× lens (NA 0.3) and images were prepared using Photoshop. All exposure levels were identical between groups.

Transmission Electron Microscopy

These studies were performed as previously described.¹⁹ Kidney tissue pieces were fixed a minimum of 12 hours in fixative (4% paraformaldehyde+1% glutaraldehyde in PBS, pH 7.2). After fixation, tissue was washed with 0.1 M phosphate buffer, stained with 2% osmium tetroxide, washed in H₂O, stained with 2% aqueous uranyl acetate, washed in H₂O, dehydrated through a graded series of ethanol and acetone, and embedded in Spurr resin (EMS, Hatfield, PA). After a 24 hour polymerization at 60°C, 0.1 μm ultrathin sections were prepared and post-stained with lead citrate. Micrographs were acquired using a JEOL 1400 Plus transmission electron microscope (JEOL, Inc., Peabody, MA) operating at 80 kV and equipped with a NanoSprint12 camera (AMT, Inc., Woburn, MA).

Gene and Protein Expression

Kidney mRNA and protein expression was assessed as previously described.^10,13 Briefly, two-step real-time RT-PCR using a Transcriptor First Strand cDNA synthesis kit (Roche, Indianapolis, IN) and TaqMan Gene Expression assays (ThermoFisher Scientific) was used for mRNA quantitation. Kidney protein expression was assessed by Western analysis with separation on 12% Tris-HCl gels (BioRad Life Science, Hercules, CA). Primary antibodies for nuclear respiratory factor 1 (NRF1), phospho-AMP-activated protein kinase-α (p-AMPK-α), poly(ADP-ribose) polymerase (PARP), cleaved PARP (c-PARP), and cleaved caspase-3 (c-Caspase-3) were used in overnight incubations at 4°C. Normalization for densitometric analysis of Western blots was performed by assessment of total protein staining using amido black (Supplemental Figure 1). In addition, the Supplemental Data also provide information for TaqMan Gene Expression assays and antibodies used for these studies in tabular form.

NAD⁺ Content

Measurement of NAD⁺ content in whole kidney lysates was performed as in our previous study, using an enzyme cycling reaction, performed for these studies with a fluorometric assay kit (catalog no. ab176723; Abcam) according to the manufacturer's instructions.¹⁹ Briefly, frozen kidney tissues were homogenized in ten volumes of lysis buffer (2.5% TCA) followed by centrifugation at 12,000×g for 3 minutes at 4°C. After a 20-fold dilution in tris buffer (90 mM, pH 7.8), NAD⁺ levels were assayed, and content was normalized to tissue wet weight.

Statistical Analyses

Data are expressed as mean±SEM and considered statistically significant for P < 0.05. The Student t test was used for parametric data, and the Mann–Whitney U test was used for nonparametric data. The Fisher exact test was used to analyze the effect of NMN on dystrophic calcification.

Results

Administration of NMN Protects against HP-AKI and Preserves NAD⁺ Content

NMN-treated mice compared with vehicle-treated mice subjected to HP-AKI demonstrated significantly lower serum creatinine and BUN at 1 day after HP-AKI (Figure 1, A and B). Furthermore, we confirmed that whole kidney NAD⁺ content was preserved in the NMN-treated mice compared with vehicle-treated mice at 1 day after HP-AKI (Figure 1C).

Figure 1.

Administration of NMN protects against HP-AKI and preserves NAD⁺ content. Serum creatinine (A) and BUN (B) were significantly lower in NMN-treated versus vehicle-treated mice at day 1 after HP-AKI. (C) Significantly higher renal NAD⁺ content was observed in NMN-treated versus vehicle-treated mice at day 1 after HP-AKI. n=5 each in sham groups, n=7 and n=8 in vehicle-treated and NMN-treated HP-AKI groups, respectively; **P < 0.01 and ****P < 0.0001, vehicle-treated versus NMN-treated HP-AKI group. HP-AKI, heme protein-mediated AKI; NMN, nicotinamide mononucleotide.

Administration of NMN Reduces Renal Histologic Injury in HP-AKI

Histological analysis of the kidney demonstrated that NMN-treated mice exhibited markedly reduced tubular injury compared with vehicle-treated mice at 1 day after HP-AKI. Specifically, significantly lower cast formation, tubular necrosis, and tubular dilation were observed in NMN-treated mice (Table 1); in addition, there was less cortico-medullary injury (the site where histologic injury starts in AKI) and less extension of histologic injury into the cortex in NMN-treated mice (Table 1). NMN treatment also prevented dystrophic renal calcification, the latter representing severe histologic damage including cell death. In the vehicle-treated HP-AKI group, four of seven mice exhibited dystrophic renal calcification, whereas none of the eight mice in the NMN-treated HP-AKI group showed any evidence of dystrophic calcification (P < 0.05). These histologic findings are displayed in representative kidney cortex sections of vehicle-treated HP-AKI mice (Figure 2Ai) compared with NMN-treated HP-AKI mice (Figure 2Aii) at day 1. In addition, increased injury was also observed in the corticomedullary junction of vehicle-treated HP-AKI mice (Figure 2A, iii and v) compared with NMN-treated HP-AKI mice (Figure 2A, iv and vi) at day 1.

Table 1.

Histologic scoring of renal injury in vehicle-treated and nicotinamide mononucleotide-treated mice at day 1 after HP-AKI

Renal Injury Criterion	HP-AKI		P Value
	Vehicle	NMN
Cast formation	2.4±0.2	1.0±0.3	0.0082
Necrosis	3.1±0.1	0.9±0.3	0.0009
Tubular dilation	3.3±0.2	1.4±0.3	0.0009
Cortico-medullary injury	4.0±0.0	1.4±0.4	0.001
Extension into cortex	3.9±0.1	0.9±0.4	0.0011

HP-AKI, heme protein-mediated AKI; NMN, nicotinamide mononucleotide.

Figure 2.

Administration of NMN reduces renal histologic injury in HP-AKI. (A) Representative renal histological sections of the cortex (i), corticomedullary junction lower magnification (iii) and higher magnification (v) in vehicle-treated mice, and of the cortex (ii) and corticomedullary junction lower magnification (iv) and higher magnification (vi) in NMN-treated mice at day 1 after HP-AKI. (B) Immunofluorescence staining for KIM-1 in vehicle-treated and NMN-treated mice kidneys at day 1 after HP-AKI. Note the reduced KIM-1 staining in HP-AKI mice kidneys after NMN treatment (right panel). Scale bars=50 µm. KIM-1, kidney injury marker 1.

Immunofluorescence studies at day 1 after HP-AKI assessing the expression of KIM-1 (a marker of AKI) demonstrated prominent KIM-1 expression in the kidney in the vehicle-treated (Figure 2B, left panel), HP-AKI group, but not in the NMN-treated group at day 1 after HP-AKI (Figure 2B, right panel). KIM-1 expression was not observed in the vehicle-treated and NMN-treated sham mice kidneys (data not shown).

Administration of NMN Preserves Mitochondrial Ultrastructure in HP-AKI

To evaluate the effect of NMN treatment on mitochondrial integrity after HP-AKI, mitochondrial ultrastructure was examined in the proximal tubules using transmission electron microscopy at day 1 after HP-AKI. In both sham vehicle-treated (Figure 3A, i and iv) and sham NMN-treated kidneys (not shown), mitochondria in proximal tubules appeared elongated and structurally intact. By contrast, proximal tubules from vehicle-treated HP-AKI mice exhibited marked mitochondrial damage characterized by rounded, oval-shaped organelles with irregular and disrupted cristae (Figure 3A, ii and v). Notably, mitochondrial morphology in NMN-treated HP-AKI kidneys was well preserved, showing more elongated and intact profiles consistent with improved structural integrity (Figure 3A, iii and vi). Parenthetically, other aspects of cellular ultrastructure benefited from NMN, Vehicle-treated HP-AKI mice demonstrated marked brush border membrane (BBM) destruction, with only remnants of microvilli visible (Figure 3A viii) compared with sham vehicle-treated (Figure 3A vii); notably, the BBM was remarkably preserved in NMN-treated HP-AKI mice (Figure 3A ix).

Figure 3.

Administration of NMN preserves mitochondrial ultrastructure in HP-AKI. (A) Representative TEM images of proximal tubule epithelial cell mitochondria at day 1 after HP-AKI. Elongated mitochondria were observed in vehicle-treated sham kidneys (i and iv), while vehicle-treated HP-AKI kidneys displayed rounded, misshapen mitochondria with disrupted cristae (ii and v). However, NMN-treated HP-AKI kidneys exhibited relatively preserved, elongated mitochondrial morphology (iii and vi). Vehicle-treated HP-AKI kidneys (viii) compared with sham (vii) demonstrated almost complete loss of BBM in proximal tubules, with only remnants of microvilli, while BBM is relatively preserved in NMN-treated HP-AKI kidneys (ix). Scale bar=1 µm. (B and C) Western blot analysis of NRF1 protein (B) and p-AMPK-α (C) in the vehicle-treated and NMN-treated mice kidneys at day 1 after HP-AKI or sham. n=4 each in sham groups, n=7 and n=8 in vehicle-treated and NMN-treated HP-AKI group, respectively. Data are presented as mean±SEM; *P < 0.05, vehicle-treated versus NMN-treated HP-AKI group. BBM, brush border membrane; NRF1, nuclear respiratory factor 1; p-AMPK-α, phospho-AMP-activated protein kinase-α; TEM, transmission electron microscopy.

Quantitative morphometric analysis of more than 300 mitochondria from five to eight samples per group revealed a significant reduction in the mean mitochondrial aspect ratio from mitochondria of proximal tubules (major axis/minor axis) in the vehicle-treated HP-AKI group, indicating increased mitochondrial fragmentation (Table 2). By contrast, in NMN-treated, HP-AKI mice, the mitochondrial aspect ratio was preserved (Table 2).

Table 2.

Morphometric analysis of mitochondrial structure

Mean Aspect Ratio (Major Axis/Minor Axis)
Sham		HP-AKI
Vehicle	NMN	Vehicle	NMN
3.33±0.3	3.18±0.3	1.69±0.1	3.13±0.2a

Analysis revealed a significant reduction in the mean aspect ratio in vehicle-treated HP-AKI kidneys, which is restored toward normal with NMN treatment. >300 mitochondria were analyzed in each group. No significant difference in the mean aspect ratio was observed within sham groups (vehicle-treated versus NMN-treated mice). n=5 in each sham group and n=6 and n=8 in vehicle-treated HP-AKI and NMN-treated HP-AKI mice, respectively. HP-AKI, heme protein-mediated AKI; NMN, nicotinamide mononucleotide.

^aP = 0.004 NMN-treated HP-AKI compared with vehicle-treated HP-AKI.

Relevant to preservation of mitochondrial integrity in NMN-treated compared with vehicle-treated mice after HP-AKI is significantly greater expression of proteins involved in mitochondrial biogenesis and fusion (e.g., NRF1, Figure 3B) and proteins that serve as energy sensors and promoters of mitochondrial integrity (e.g., p-AMPK-α, Figure 3C).

Administration of NMN Protects against Apoptosis in HP-AKI

To evaluate the effect of NMN on apoptosis in the kidney after HP-AKI, we assessed the expression of apoptosis and its related genes and proteins. NMN-treated mice compared with vehicle-treated mice, after HP-AKI, demonstrated markedly reduced TUNEL staining (Figure 4A, upper right versus upper left panel) and markedly reduced expression of c-PARP (Figure 4B) and c-Caspase-3 (Figure 4C). Corroborating these findings in NMN-treated mice compared with vehicle-treated mice, after HP-AKI, and as shown in Table 3, are increased expression of BCL-2 (an anti-apoptotic gene), reduced expression of RIPK-3 (a mediator of necroptosis and inflammation), and a lower Bax/BCL-2 ratio (an index for the severity of apoptosis).

Figure 4.

Administration of NMN protects against apoptosis in HP-AKI. (A) Representative images from vehicle-treated HP-AKI kidneys at day 1 after HP-AKI revealed many TUNEL-positive nuclei especially in the tubular epithelial cells, indicating extensive apoptosis (left, upper panel). By contrast, kidneys from NMN-treated HP-AKI mice exhibit markedly fewer TUNEL-positive staining, reflecting attenuation of apoptotic cell death (right, upper panel). (B) Western blot analysis of total PARP and c-PARP protein in the vehicle-treated and NMN-treated mice kidneys at day 1 after HP-AKI or sham. (C) Western blot analysis of c-Caspase-3 protein in the vehicle-treated and NMN-treated mice kidneys at day 1 after HP-AKI or sham. n=4 each in sham group, n=7 and n=8 in vehicle-treated, and NMN-treated HP-AKI group, respectively; data are presented as mean± SEM; **P < 0.01 and ****P < 0.0001, vehicle-treated versus NMN-treated HP-AKI group. c-Caspase-3, cleaved Caspase3; c-PARP, cleaved-poly(ADP-ribose) polymerase; PARP, poly(ADP-ribose) polymerase; TUNEL, terminal deoxynucleotidyl transferase–mediated digoxigenin-deoxyuridine nick-end labeling.

Table 3.

Apoptosis-related genes and ratio (at day 1 after sham or HP-AKI in vehicle-treated and NMN-treated mice kidneys)

Gene	Sham		HP-AKI		P Value
	Vehicle	NMN	Vehicle	NMN
BCL-2	3.0±0.1	3.1±0.1	2.5±0.2	3.1±0.1	0.0207
RIPK-3	2.1±0.2	2.8±0.3	19.4±1.5	8.8±1.7	0.0004
Bax/BCL-2 ratio	1.0±0.1	1.1±0.1	2.4±0.1	1.9±0.2	0.037

Realtime RT-PCR analysis of BCL-2 and RIPK-3 mRNA expression normalized for 18S rRNA expression and Bax/BCL-2 ratio in sham (vehicle-treated and NMN-treated) and HP-AKI (vehicle-treated and NMN-treated) at day 1 after HP-AKI. Data are presented as mean±SEM; n=5 in each sham group and n=7 and n=8 in vehicle-treated and NMN-treated HP-AKI groups, respectively. P values are from HP-AKI-sham when compared with HP-AKI-NMN-treated group. HP-AKI, heme protein-mediated AKI; NMN, nicotinamide mononucleotide.

Administration of NMN Reduces Injury-Related Gene Expression

We assessed the effects of NMN on representative, injury-related gene expression in HP-AKI that, depending on the species, can serve as a promoter of, or a protector against, AKI. As shown in Figure 5, NMN reduced the HP-AKI-induced elevation in the proinflammatory cytokine (IL-6) and the procoagulant, proinflammatory species (PAI-1). Countervailing anti-inflammatory responses that mitigate the severity of AKI and are proportionately induced depending upon the severity of AKI include the cytoprotectant heme oxygenase-1 (HO-1) and its upstream prosurvival transcription factor hypoxia-inducible factor 1-alpha (HIF-1α)²⁶; expression of these genes may be used as a readout of the severity of AKI. In both instances, NMN lessened the induction of HO-1 and HIF-1α that occurs after HP-AKI (Figure 5).

Figure 5.

Administration of NMN reduces injury-related gene expression. Inflammatory gene expression: (A) IL-6 mRNA and (B) PAI-1 mRNA expression in the vehicle-treated and NMN-treated mice kidneys at day 1 after HP-AKI or sham. Anti-inflammatory gene expression: (C) HO-1 mRNA and (D) HIF-1α mRNA expression in the vehicle-treated and NMN-treated mice kidneys at day 1 after HP-AKI or sham. n=5 in each sham group, n=7 and n=8 in vehicle-treated and NMN-treated HP-AKI groups, respectively. **P < 0.01 and ***P < 0.001, vehicle-treated versus NMN-treated HP-AKI group.

Administration of NMN Mitigates the Severity of the Senescent Phenotype in HP-AKI

We recently demonstrated that soon after HP-AKI there is a senescent phenotype, including among other indices, reduced expression of lamin B1 (a classic finding in cells undergoing senescence). In the present studies, immunofluorescence of lamin B1 revealed reduced lamin B1 staining in proximal tubular epithelial cells of vehicle-treated HP-AKI kidneys, indicative of nuclear envelope damage as part of the senescence process; lamin B1 expression, however, was preserved in NMN-treated, HP-AKI mice (Figure 6A). In addition, administration of NMN blunted the induction of clusterin (another senescence marker) that occurs after HP-AKI (Figure 6, B and C), and the induction of MMP3 and MMP7 (Figure 6, D and E), the latter representing proteolytic genes that are integral components of the senescence associated secretory phenotype.

Figure 6.

Administration of NMN mitigates the severity of the senescent phenotype in HP-AKI. (A) Immunofluorescence staining of lamin B1 showed reduced perinuclear lamin B1 expression in tubular epithelial cells of vehicle-treated HP-AKI mice (middle panel) compared with vehicle-treated sham (left panel), consistent with nuclear envelope injury and cellular senescence. By contrast, lamin B1 staining is relatively preserved in NMN-treated HP-AKI kidneys, with perinuclear localization comparable with vehicle-treated shams (right panel). Scale bar=50 µm. (B) Western blot analysis of secretory clusterin (α-chain) protein in the vehicle-treated and NMN-treated mice kidneys at day 1 after HP-AKI or sham. n=4 in each sham group, n=8 and n=7 in vehicle-treated and NMN-treated HP-AKI groups, respectively. (C) Clusterin mRNA, (D) MMP3 mRNA, and (E) MMP7 mRNA expression in the vehicle-treated and NMN-treated mice kidneys at day 1 after HP-AKI or sham. n=5 in each sham group, n=7 and n=8 in vehicle-treated and NMN-treated HP-AKI groups, respectively. Data are presented as mean±SEM; ***P < 0.001, **P < 0.01, and *P < 0.05, vehicle-treated versus NMN-treated HP-AKI group.

Myoglobin or Heme, Administered In Vivo, Diminishes NAD⁺ Content in Mice with Intact Kidneys

In HP-AKI, kidney content of heme proteins, such as myoglobin, as well as heme is increased, both of which are critical drivers of HP-AKI. We thus questioned whether the reduction in NAD⁺ content observed in HP-AKI may reflect the effects of these entities. We thus administered myoglobin or heme to mice with intact (unstressed, without HP-AKI) kidneys. As demonstrated in Table 4, the administration of either myoglobin or heme reduced kidney content in NAD⁺ in mice with intact kidneys.

Table 4.

Effect of myoglobin or heme administration in vivo on renal NAD⁺ content (pmol/mg wet weight) at day 1

	Vehicle	Myoglobin	P Value
NAD+ Content	88.9±1.9	49.4±7.5	0.0022

	Vehicle	Hemin	P Value
NAD+ Content	96.6±2.2	83.1±1.5	0.0022

Data are presented as mean±SEM; n=4 in all groups.

Discussion

To the best of our knowledge, this study is the first to demonstrate the following combination of findings: First, reduction in NAD⁺ content prominently occurs in the kidney in HP-AKI; second, restoration in NAD⁺ content attends the administration of NMN; and third, as revealed by an array of injury-related indices, such administration of NMN markedly reduces the severity of HP-AKI.

Administration of NMN protected against the fall in GFR in HP-AKI as revealed by two filtration markers, namely serum creatinine and BUN. These functional effects were accompanied by less histologic injury. HP-AKI, as for most forms of AKI, usually starts in the cortico-medullary region of the kidney and extends into the cortex. Such initial injury and its extension into the cortex were both significantly reduced in HP-AKI when NAD⁺ was boosted. These changes were accompanied by less cast formation, tubular dilatation, and cell necrosis. Furthermore, dystrophic calcification, a process reflecting the deposition of calcium in severely injured and necrotic tissue, was prominent in the kidney in the majority of vehicle-treated, HP-AKI mice, but nonexistent in the kidney when NMN was administered to mice with HP-AKI. Similarly, KIM-1, a marker of AKI, was abundantly expressed in mice with HP-AKI, whereas such expression was absent in NMN-treated HP-AKI mice.

Our prior studies demonstrate that mitochondria are major targets in HP-AKI.^9,19 For example, within hours of HP-AKI, mitochondrial respiration is decreased,⁹ as is mitogenesis, and the dynamic balance between mitochondrial fusion and fission is decidedly tilted toward mitochondrial fission.¹⁹ We thus examined the effect of NMN in HP-AKI, focusing on mitochondrial integrity. We demonstrate that significant ultrastructural derangements occur in mitochondria, and such derangements in mitochondrial integrity are markedly reduced by NMN as determined both by qualitative assessment and quantitative indices, with NMN preserving the mean mitochondrial aspect ratio. In HP-AKI, NMN also protected against the reduction that occurs in NRF1, the latter representing a driver of mitogenesis, and in phospho-AMPKα, a promoter of mitochondrial integrity.²⁷ In this regard, studies examining the effect of NMN on measures of mitochondrial function such as mitochondrial respiration and ATP content would be of interest.

Mitochondrial injury in AKI can instigate apoptosis in which caspase-3 is one of the main effectors of apoptosis.²⁷ We first demonstrate that apoptosis, as revealed by TUNEL staining, is prominent in HP-AKI, and such staining was markedly attenuated by NMN. c-Caspase-3 was upregulated in HP-AKI, and this heightened expression of c-Caspase-3 was strikingly reduced by NMN. We also assessed another marker of cell death, c-PARP, the latter representing a valid index of cell death.²⁸ NMN prevented the markedly increased expression of c-PARP that occurs in HP-AKI. Interestingly, PARP-1 may also link mitochondrial and nuclear processes in orchestrating cell death because PARP-1 activation provokes the release of apoptosis-inducing factor (AIF) from mitochondria²⁸; AIF can induce nuclear DNA fragmentation.²⁸ Validation of such speculation would require studies of the effects of PARP inhibitors on AIF expression, DNA fragmentation, and apoptosis.

AKI is associated with marked inflammatory processes, which, in turn, mount anti-inflammatory responses.²⁹ We thus examined expression of representative genes that reflect such responses. Our prior studies demonstrate that HP-AKI is attended with marked induction of IL-6 and PAI-1 as proinflammatory processes and induction of HO-1 as an anti-inflammatory response²⁶; the degree of expression of such anti-inflammatory responses often parallels the severity of the proinflammatory processes. We demonstrate that the induction of IL-6 and PAI-1 in HP-AKI was reduced by NMN. One of the transcription factors upstream of HO-1 is HIF-1α, a transcription factor that is protective in AKI.³⁰ NMN reduced expression of both HO-1 and HIF-1α that occurs in HP-AKI.

Our recent studies demonstrate the significance of a senescence phenotype in HP-AKI.¹³ This phenotype was again observed in the present study as evidenced by the reduced expression of lamin B1, the striking induction of clusterin, and the induction of MMP3 and MMP7, degradative enzymes that are salient components of the senescence associated secretory phenotype. NMN mitigated all these aspects of the senescence phenotype. Clusterin, a molecular chaperone that binds to denatured proteins, is induced after HP-AKI, as shown in our prior studies.³¹ Clusterin is secreted by human cells undergoing senescence,³² and recent novel observations by Ferenbach and colleagues have emphasized the tight correlation between clusterin production, as reflected by its urinary excretion, with the expression of p21^Cip1 by Ki67-negative cells undergoing senescence.^33,34 As shown in our present studies, NMN significantly reduced the induction of clusterin, as assessed by expression of both mRNA and protein.

Studies in vivo in which we administered myoglobin or heme to mice with intact kidneys demonstrate that either led to a reduction in NAD⁺ content in the kidney. These findings identify heme proteins and heme as novel determinants of kidney NAD⁺ content. We thus suggest that in HP-AKI, the reduction in NAD⁺ content that occurs may reflect the abundant intrarenal presence of myoglobin and/or heme in HP-AKI. Which one or combination of the three synthetic pathways that are known to generate NAD⁺²³ are compromised is beyond the scope of this study. NAD⁺ is also regenerated from NADH by mitochondria at complex 1.²³ We speculate that two other processes may thus be relevant to reduced NAD⁺ content in HP-AKI. First, it is possible that the severe mitochondrial injury we have observed in HP-AKI may lead to impaired regeneration of NAD⁺ from NADH. Second, activation of PARP is associated with increased consumption of NAD⁺.²³ It is possible that the reduction in NAD⁺ may reflect such PARP activation. Studies of PARP activity would be of interest. Whether PARP activation contributes to decreased NAD⁺ content in HP-AKI would require studies of the effect of PARP inhibitors on kidney NAD⁺ content in this model.

A recent prior study demonstrated that vitamin B3 is protective in the glycerol model of AKI.³⁵ B3 may be converted to NAD⁺ through the Salvage Pathway.²³ It is possible that in this prior study restoration of NAD may be the underlying mechanism for the protective effects of B3. However, this prior study examined neither the NAD⁺ content of the kidney in HP-AKI nor the effect of vitamin B3 on NAD⁺ content in HP-AKI.

Our study demonstrates that multiple processes may contribute to the beneficial effects of NMN in HP-AKI. These pathways and processes are highly interconnected and bidirectional in AKI²⁹—wherein one process begets others, which, in turn, feed back to exacerbate the instigating process—it is thus not possible to specify the proximate site at which NMN exerts its beneficial effects.

We emphasize that our findings, demonstrating the remarkable protective effects of NMN when administered before the instigation of HP-AKI, is a preventive strategy that speaks to the adverse consequence of decreased NAD⁺ content that occurs in HP-AKI. The translation of such findings into a potential therapeutic strategy would require assessing whether the administration of NMN after the instigation of HP-AKI is also protective. Such studies would require the administration of different doses of NMN and varying severity of HP-AKI. These clinically important studies are beyond the scope of the present work and are the focus of future investigation.

In summary, we demonstrate that HP-AKI is attended by significant reduction in NAD⁺ content in the kidney; that the administration of NMN effectively restores levels of NAD⁺ in the kidney; and that such restoration is attended by significantly reduced severity of HP-AKI, as assessed by multiple indices. We conclude that these findings provide new evidence regarding the significance of kidney NAD⁺ content in safeguarding the kidney against HP-AKI and in supporting the groundswell of interest in boosting kidney NAD⁺ content as a therapeutic approach in clinical AKI.

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/KN9/B379.

Author Contributions

Conceptualization: Anthony J. Croatt, Luis A. Juncos, Karl A. Nath, Raman Deep Singh.

Data curation: Allan W. Ackerman, Trace A. Christensen, Anthony J. Croatt, Joseph P. Grande, Karl A. Nath, Raman Deep Singh.

Formal analysis: Allan W. Ackerman, Trace A. Christensen, Anthony J. Croatt, Joseph P. Grande, Karl A. Nath, Raman Deep Singh.

Funding acquisition: Karl A. Nath.

Investigation: Anthony J. Croatt, Karl A. Nath, Raman Deep Singh.

Methodology: Allan W. Ackerman, Anthony J. Croatt, Joseph P. Grande, Raman Deep Singh.

Project administration: Anthony J. Croatt, Karl A. Nath, Raman Deep Singh.

Resources: Anthony J. Croatt, Karl A. Nath, Raman Deep Singh.

Software: Allan W. Ackerman, Anthony J. Croatt, Joseph P. Grande, Raman Deep Singh.

Supervision: Anthony J. Croatt, Karl A. Nath.

Validation: Allan W. Ackerman, Trace A. Christensen, Anthony J. Croatt, Joseph P. Grande, Luis A. Juncos, Karl A. Nath, Raman Deep Singh.

Visualization: Anthony J. Croatt, Karl A. Nath, Raman Deep Singh.

Writing – original draft: Allan W. Ackerman, Trace A. Christensen, Anthony J. Croatt, Luis A. Juncos, Karl A. Nath, Raman Deep Singh.

Writing – review & editing: Anthony J. Croatt, Joseph P. Grande, Luis A. Juncos, Karl A. Nath, Raman Deep Singh.

Funding

K.A. Nath: National Institute of Diabetes and Digestive and Kidney Diseases (RO1 DK133401).

Declarative Statements

All animal experiments were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals or an equivalent standard that meets or exceeds the ethical and welfare requirements outlined in the NIH Guide. All protocols were approved by the appropriate institutional animal care and use committee.

Data Availability Statements

Original data generated for the study will be made available upon reasonable request to the corresponding author. Data Type: Image Data and Observational Data.