Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Nov 14.
Published in final edited form as: Clin Sci (Lond). 2026 Apr 15;140(4):589–605. doi: 10.1042/CS20258807

Macula densa-specific NOS1 knockout determines susceptibility to ischemic acute kidney injury

Minghua Li 1, Jessica Bahena Lopez 2, Runze Ni 1, Jin Wei 3, Kiran Dhanireddy 4,5, Vijay Subramanian 4,5, Lei Wang 1,5, Kristof Williams 1, Nathan Hall 1, Nadezhda N Zheleznova 1, Yan Lu 6, Liying Fu 7, Enyin Lai 5, Jie Zhang 3, James A McCormick 2, Ruisheng Liu 1,5
PMCID: PMC13764694  NIHMSID: NIHMS2219196  PMID: 41837645

Abstract

Acute kidney injury (AKI) causes rapid loss of renal function with high morbidity and mortality. Our previous research showed that neuronal nitric oxide synthase (NOS1) mediates nitric oxide-dependent dilation of the afferent arteriole, inhibiting tubuloglomerular feedback (TGF), which critically regulates single nephron glomerular filtration rate (GFR). This mechanism could affect how vulnerable the organ is to ischemic injury, but this has not been directly tested. In the present study, we generated inducible macula densa (MD)-specific NOS1 knockout mice (NKCC2-Cre-NOS1 flox/flox) using a new inducible NKCC2-Cre line crossed with NOS1 flox mice. AKI was induced by 18-min bilateral renal pedicle clamping at 37°C, followed by 48-h reperfusion. Kidney injury was assessed by measuring GFR, plasma creatinine, histology, cytokines, apoptosis, fibrotic factors, and proteomics. Complete NOS1 deletion in MD cells was confirmed through NKCC2/NOS1 immunofluorescence. MD NOS1 deletion enhanced the TGF response, which was further amplified following AKI compared with control (Cre −/−). Following AKI, compared with controls (Cre −/−), NOS1 knockouts showed a significantly lower GFR (236 ± 66 to 24 ± 22 μl/min) and higher plasma creatinine levels, alongside more severe tubular damage evidenced by H&E staining. Cytokine array analysis showed chemokines such as MCP-1 and CXCL1 and macrophage marker CD68 were significantly increased; Western blot analysis showed cleaved caspase-3 levels were significantly increased, indicating enhanced apoptosis. Additionally, fibrosis markers TIMP-1, collagen-3, and α-SMA were significantly up-regulated at both mRNA and protein levels. We further observed increased hypoxia marker hypoxia-inducible factor-1α in MD-specific NOS1 knockout mice (Cre +/−). Global label-free proteomic profiling with targeted validation identified genotype-dependent responses involving haptoglobin, Tacstd2, and Cyp20a1, linking NOS1 deficiency to exaggerated inflammatory, fibrotic, and metabolic pathways. These findings highlight a novel role of MD NOS1 and TGF in AKI pathophysiology, suggesting NOS1 targeting as a therapeutic strategy. Haptoglobin was identified as a downstream plasma signal of NOS1-dependent AKI responses, indicating potential translational value.

Introduction

Acute kidney injury (AKI) is a common and serious complication among hospitalized patients. Nearly one in five admissions involves AKI, and in many cases, it may progress to chronic kidney disease [1,2]. Although this is a significant health concern, effective treatments still remain limited. One primary reason for this is the incomplete understanding of why kidneys are particularly susceptible to ischemic stress. The kidney’s capacity to regulate blood flow and oxygen delivery contributes to this vulnerability.

Nitric oxide (NO) plays a key role in kidney response to stress. It functions by dilating blood vessels, thereby regulating renal blood flow and filtration [3–5]. In the kidney, all three known nitric oxide synthase (NOS) isoforms can produce NO [6]. Endothelial NOS (NOS3) is found in the lining of blood vessels [7]. Inducible NOS (NOS2) appears in mesangial cells and immune cells [8]. Neuronal NOS (NOS1), meanwhile, is present in certain tubule cells. Within the juxtaglomerular apparatus, macula densa (MD) cells express especially high levels of NOS1 [6,7]. NO produced by MD NOS1 diffuses to nearby afferent arterioles, or via the cGMP signal pathway, causing vasodilation and blunting tubuloglomerular feedback (TGF). Inhibiting MD NOS1 enhances TGF, resulting in greater afferent arteriole constriction. NO from the MD NOS1 helps to stabilize GFR [9–11]. Adjusting NOS activity in MD cells can affect afferent arteriole tone and glomerular pressure [12,13]. Additionally, MD NOS1 controls renin release—its inhibition or deletion is usually considered to lower renin levels under various conditions [14,15]. Because NO is so important for kidney blood flow, a deficiency of MD NOS1 may increase susceptibility to ischemic injury. Without the vasodilating and TGF-dampening effects of NO from MD, excessive constriction of the afferent arterioles may occur, reducing blood flow and oxygen delivery to nephrons during stress. Studies indicate that lower NO worsens ischemic AKI. For example, removing endothelial NOS in mice leads to loss of small blood vessels and increased damage after kidney injury [16], and blocking all NOS before kidney ischemia raises vascular resistance, reduces blood flow, and increases mortality rates [17]. These findings suggest that an active NO system plays a protective role against acute ischemia. We have long understood the importance of endothelial NO, yet the specific influence of MD-derived NOS1 on the risk of AKI remains less certain. Interestingly, MD NOS1 could be especially vital for managing blood flow in particular regions of the kidney through TGF. When this regulation is lost, ischemia may worsen—especially in areas where oxygen is already scarce.

The hypoxia-inducible factor (HIF) pathway is another important component of ischemia–reperfusion injury-induced acute kidney injury (IRI-AKI). When oxygen drops, hypoxia-inducible factor-1α (HIF-1α) is up-regulated and activates genes that help cells survive in low-oxygen conditions. However, if HIF-1α stays active too long, it increases production of factors that cause fibrosis and chronic kidney disease [18,19]. We propose that eliminating NOS1 from MD cells exacerbates the AKI by permitting excessive TGF activity, resulting in increased vasoconstriction and hypoxia. This sustains HIF-1α activation and drives fibrosis.

Modern molecular profiling methods provide new opportunities to dissect AKI pathways. Proteomic approaches can survey thousands of proteins simultaneously, revealing candidate effectors and potential biomarkers of injury. Such studies have highlighted acute-phase proteins, regulators of metabolism, and fibrosis-associated mediators in human and experimental AKI [20–22]. Integrating proteomics with physiological models can therefore connect functional outcomes with molecular signatures and point toward translational readouts.

In the present study, we tested the hypothesis that deletion of MD NOS1 reduces the kidney’s ability to adapt to ischemia, leading to more hypoxia, cell death, and fibrosis. Using a mouse with an inducible, MD-specific NOS1 knockout (MD-NOS1KO), we looked at kidney function, tissue changes, and molecular signals after ischemia–reperfusion injury. We also used global proteomic analysis to find proteins whose AKI responses depended on genotype and then confirmed three candidates: haptoglobin (Hp), Tacstd2, and Cyp20a1. These results show that MD NOS1 helps protect against ischemic kidney injury and reveal potential molecular markers that could be useful for future research.

Materials and methods

Animals

All animal studies were approved by the Institutional Animal Care and Use Committees at the University of South Florida, Morsani College of Medicine (protocol 12005R), and Oregon Health & Science University (protocol IP000286). Animals were deeply anesthetized by isoflurane inhalation. Kidneys were harvested, and animals were killed by exsanguination under anesthesia as a terminal procedure.

Generation of animal models

Development of tamoxifen inducible NKCC2-Cre mice strain

To establish a temporally controllable MD-NOS1KO model, we first developed a tamoxifen-inducible NKCC2-Cre mice strain (B6(Cg)-NKCC2<tm1(P2A-iCreERT2), Slc12a1-Cre-ERT2). The self-cleaving P2A-iCreERT2 cassette was inserted upstream of the Slc12a1 stop codon in exon 27. To guide better integration to the targeted allele, the cassette included a 1.5 kb homology arm (Figure 1A). Cas9/sgRNA plasmid construction was completed by Biocytogen Boston Corp., and the sequences were further confirmed by DNA sequencing. The cleavage efficiency of the CRISPR-Cas9 system was confirmed by the Universal CRISPR Activity Assay (Supplementary Figure S1A). Then, tamoxifen-inducible NKCC2-cre strain was created by zygote microinjection of targeting vectors into C57BL/6J background mice.

Figure 1. Generation and characterization of NKCC2-Cre mice.

Figure 1.

Generation and characterization of NKCC2-Cre mice. (A) Schematic representation of the targeting strategy for generating tamoxifen-inducible NKCC2-Cre mice. The P2A-iCreERT2 cassette was inserted upstream of the Slc12a1 stop codon in exon 27 with 1.5 kb homology arms to facilitate targeted integration. (B) Immunoblots and densitometric quantification of total and phosphorylated NKCC2, NCC, WNK4, and SPAK in kidney homogenates from controls (Cre −/−) and NKCC2-Cre (Slc12a1-Cre-ERT2) female mice (Cre −/−, n = 8; Cre +/−, n = 8). The complete dataset including males is shown in Supplemental Figure S1C. Protein abundance was normalized to Cre −/− females (set to 1.00 ± SD). Numbers above each blot represent mean ± SD, with P-values derived from unpaired Student’s t-tests. (C) Urinary sodium excretion (UNaV) and urine volume (UVol) at baseline and following furosemide administration in Cre −/− and Cre +/− mice. Data are paired for individual animals. Statistical comparisons were performed using paired Student’s t-tests; ns = not significant, ** = P <0.01, *** = P <0.001, **** = P <0.0001. (D) Representative kidney sections stained for NKCC2 (red), eGFP (green), and DAPI (blue) in NKCC2-Cre mice and control littermates following tamoxifen induction. Merged images show colocalization of eGFP with NKCC2-positive thick ascending limb (TAL) segments. Scale bar = 20 μm. (E) Western blot confirmation of NOS1 deletion in MD-NOS1KO mice. NOS1 protein expression was significantly lower following tamoxifen administration in cortical tissue lysates. GAPDH served as loading control. (F) Quantitative RT-PCR analysis of NOS isoform specificity. Only NOS1 mRNA levels were significantly reduced in MD-NOS1KO mice (Cre +/−), while NOS2 and NOS3 remained unchanged, confirming selective NOS1 deletion. Data normalized to GAPDH and presented as fold change relative to Cre −/− controls. **P <0.01; ns = not significant. (G) Representative kidney sections stained for NOS1 (red), NKCC2 (green), and DAPI (blue) in MD-NOS1KO mice (Cre +/−).

Development of tamoxifen inducible macula densa-specific NOS1 knocout mice strain

MD-NOS1KO mouse model was generated by crossing Slc12a1-Cre-ERT2 mice with NOS1 flox/flox (B6.129S4-Nos1tm2Plh/J; Stock#: 008519; Jackson Lab, U.S.A.). Complete deletion of NOS1 from the MD cells was confirmed by immunofluorescence double staining of NKCC2 and NOS1 (Supplementary Figure S1D).

Induction of ischemia-reperfusion injury-induced acute kidney injury

For the introduction of IRI-AKI, we used similar methods as we previously described [23–26] in male C57BL/6 mice (10–12 weeks, 25 ± 3 g) by clamping bilateral renal pedicles for 30 min at 36.8–37.2°C body temperature. Mice recovered with free access to food and water and were killed at 48 h after IRI. Controls (Cre −/−) were Cre-negative littermates (NKCC2-CreERT2−/−; NOS1flox/flox) that received identical tamoxifen treatment.

Measurement of tubuloglomerular feedback

TGF was measured with the micro-perfusion in vitro method in isolated perfused juxtaglomerular apparatus as described in Supplementary Methods.

Glomerular filtration rate measurement

Measurement of glomerular filtration rate (GFR) was performed as we previously described [27]. Briefly, GFR was measured in conscious mice by transdermal FITC-sinistrin clearance (MediBeacon GmbH, Mannheim, Germany) at two time points: baseline immediately before induction of IRI-AKI and post-injury at 48 h after IRI-AKI. For tracer administration, mice were briefly anesthetized with 3% isoflurane, and FITC-sinistrin (5.6 mg per 100 g body weight; 28 mg ml−1 stock) was injected into the retro-orbital venous sinus. And, 5 min later, the flank was shaved, and a disposable transdermal fluorescence sensor was secured to the skin. Anesthesia was discontinued, allowing the mouse to regain full consciousness while the device continuously recorded fluorescence decay. The monitor automatically terminated the recording once the signal reached background levels. Elimination kinetics were analyzed with MediBeacon Studio software to calculate GFR (ml min−1 kg−1).

Proteomic analysis

Kidney cortical tissue from control (CRE −/−) and KO mice (Cre +/−) with or without AKI was processed for mass spectrometry-based proteomics as described in Supplementary Methods.

Detailed protocols for standard biochemical assays (Western blotting, PCR, etc.) are in the Supplementary Methods.

Statistics

All data are presented as mean ± standard deviation (SD). GraphPad Prism 9 (GraphPad Software, San Diego, CA) was used for all analyses. Normality was assessed by the Shapiro–Wilk test. For comparisons between two groups, unpaired two-tailed Student’s t-tests were applied when data passed normality; otherwise, the Mann–Whitney U test was used. For experiments involving three or more groups, one-way ANOVA followed by Tukey’s multiple comparisons post hoc test was done for normally distributed datasets and Kruskal–Wallis with Dunn’s post hoc test for non-parametric data. P-value <0.05 was considered statistically significant. Two-way ANOVA with Sidak’s multiple comparisons test was applied when appropriate. Statistical significance was defined as P <0.05.

Results

Characterization of tamoxifen-inducible NKCC2-Cre mice strain

We assessed NKCC2 protein expression and activity using Western blot analysis and a furosemide response test. Western blot analysis of kidney lysates from both males and females revealed a significant reduction in total NKCC2 protein abundance in NKCC2-Cre mice compared with controls (Cre −/−). Phosphorylated NKCC2 levels did not significantly differ in NKCC2-Cre mice (Figure 1B, Supplementary Figure S1C, and Supplementary Tables S3 and S4). These findings are consistent with a previous study reporting reduced NKCC2 protein levels in mice expressing TAL-specific Cre recombinase targeted to the NKCC2 gene [28].

Expression of total NCC, phosphorylated NCC, total WNK4, phosphorylated WNK4, total SPAK, and phosphorylated SPAK remained unchanged between kidneys from male and female controls (Cre −/−) and NKCC2-Cre mice (Figure 1B, Supplementary Figure S1C, Supplementary Tables S3 and S4). This indicates that the observed effects of the transgene are specific to the NKCC2 locus, and the effects on NKCC2 expression did not lead to compensatory effects of the distal convoluted tubule immediately downstream of the TAL.

To evaluate NKCC2 activity in vivo, we performed a furosemide response test by collecting urine from mice placed in metabolic cages over a 4-h period. Despite the modest reduction in NKCC2 protein abundance, changes in sodium excretion and urinary volume in response to furosemide treatment were comparable between NKCC2-Cre and control (Cre −/−) mice (Figure 1C). These results suggest that NKCC2 activity remains intact in NKCC2-Cre mice. Plasma electrolyte levels and renal function parameters did not significantly differ between NKCC2-Cre mice and control (Cre −/−) mice, as summarized in Table 1 and Supplementary Table S1.

Table 1.

Blood values

Variable plasma n = 15 (F = 11, M = 4) Control n = 11 (F = 8, M = 3) NKCC2-Cre (+/−) P-value
Na+ (mM/l) 140.8 ± 1.8 140.7 ± 1.7 0.943
K+ (mM/l) 4.0 ± 0.4 4.1 ± 0.3 0.695
Cl− (mM/l) 111.5 ± 1.5 110.3 ± 1.6 0.073
iCa2+ (mM/l) 1.24 ± 0.05 1.21 ± 0.05 0.150
TCO2 (mM/l) 21.7 ± 1.8 22.4 ± 1.8 0.334
BUN (mg/dl) 21.8 ± 2.6 21.0 ± 2.8 0.427
Hct %PCV 35.9 ± 1.7 37 ± 1.7 0.115
Hb (g/dl) 12.2 ± 0.6 12.5 ± 0.6 0.140
Anion Gap (mM/l) 12.9 ± 1.6 12.3 ± 1.7 0.395

Data are presented as means ± SD, with sample sizes of n = 15 for the control group and n = 11 for the NKCC2-Cre (Slc12a1-Cre-ERT2) group. Values were compared between genotypes using a paired, two-tailed t-test. *P <0.05 was considered statistically significant. PCV refers to packed cell volume. F refers to Females and M to Males.

NKCC2-Cre mice display TAL-specific Cre recombinase activity

To evaluate the TAL-specificity, leakiness, and inducibility of Cre-ERT2 in NKCC2-Cre mice, we crossed them with INTACT (Isolating Nuclei TAgged in specific Cell Types) mice (https://www.jax.org/strain/021039). This transgenic line promotes nuclear translocation of Cre-ERT2 and drives GFP expression in Cre-ERT2-expressing cells upon tamoxifen treatment. The resulting NKCC2-Cre/INTACT mice were administered tamoxifen for 5 consecutive days to promote sfGFP expression at the inner nuclear membrane of TAL cells. To confirm the localization of GFP expression, we performed immunofluorescence using an anti-GFP antibody to enhance the GFP signal alongside an NKCC2 antibody to label TAL cells. GFP expression was observed to colocalize entirely with NKCC2 staining in tamoxifen-treated mice (Figure 1D, wide section view at Supplementary Figure S2B), demonstrating that Cre-ERT2 activity in NKCC2-Cre mice is specific to TAL cells.

To quantify recombination inducibility, we measured the proportion of NKCC2-positive cells that co-expressed GFP. Immunofluorescence images were collected from three NKCC2-Cre/INTACT mice treated with tamoxifen. The Cre recombination efficiency, calculated as the percentage of GFP+/NKCC2+ cells relative to total NKCC2+ cells, was 99.5% ± 0.2% after tamoxifen induction (Supplementary Figure S2B).

To assess potential Cre-ERT2 leakiness, the same experimental approach was applied to three vehicle-treated mice. In the absence of tamoxifen, only 1.2% ± 0.4% of TAL cells expressed GFP, indicating minimal basal Cre-ERT2 activity without tamoxifen induction (Supplementary Figure S2B).

Characterization of macula densa-specific NOS1 knockout mouse model

Since the major site of expression of NOS1 in the TAL is the MD [29–31], our approach generated an MD-NOS1KO mouse line. Mice heterozygous for the Slc12a1-Cre-ERT2 allele and homozygous for the NOS1 flox allele were identified by PCR genotyping (Supplementary Figure S1B). The primer used for genotyping is listed in Supplementary Table S2. To confirm the deletion is specific to NOS1, we compared the mRNA expression levels of NOS1 against NOS2 and NOS3 in cortex tissue samples using the qRT-PCR method. As shown in Figure 1F, only the NOS1 mRNA was significantly reduced compared with control (Cre −/−) mice, and the mRNA levels of NOS2 and NOS3 remained unchanged. Western blot analysis confirmed the protein expression of NOS1 was successfully deleted upon tamoxifen administration (Figure 1E). Immunofluorescence double staining for NKCC2 and NOS1 confirmed complete deletion of NOS1 from MD cells (Figure 1G, high magnification; see also Supplementary Figure S1D).

Macula densa-specific deletion of NOS1 intensified acute kidney injury after renal ischemia reperfusion

To determine the effects of MD-specific deletion of NOS1 on AKI, we induced AKI using the 30-min renal pedicle clamp method as described in our previous study [25]. Histopathologic evaluation by H&E staining showed clear genotype-dependent differences in the severity and distribution of ischemia–reperfusion (IR) injury (Figure 2A and Supplementary Figure S1F). At baseline, kidneys from both control (Cre −/−) and MD-NOS1KO mice displayed preserved tubular architecture with intact brush borders, closely apposed tubules, and an absence of interstitial edema, intratubular casts, or fibrosis. After 30 min of IRI, control (Cre −/−) kidneys developed the characteristic patchy injury pattern largely confined to the outer stripe of the outer medulla. Lesions consisted of focal tubular epithelial necrosis, scattered granular casts, and mild interstitial edema, yielding a composite tubular injury score of 9.0 ± 3.1. In contrast, MD-NOS1KO kidneys showed a more extensive injury that involved both cortex and medulla. Tubules were often collapsed, and their lumina were filled with proteinaceous or granular casts. Epithelial necrosis was more widespread, with sloughing of epithelial cells into the tubular lumen, accompanied by interstitial edema and focal inflammatory infiltrates. The composite tubular injury score in MD-NOS1KO mice (Cre +/−) was markedly higher at 20.2 ± 2.8 (Figure 2A,B).

Figure 2. MD-specific NOS1 deletion exacerbates AKI following ischemia-reperfusion.

Figure 2.

MD-specific NOS1 deletion exacerbates AKI following ischemia-reperfusion. (A) Representative H&E-stained kidney sections showing histopathological changes at 24 h after 30-min renal pedicle clamping. MD-NOS1KO mice (Cre +/−) exhibited more extensive tubular injury involving both cortex and medulla compared with Cre −/− mice, where damage was primarily confined to the outer stripe of the outer medulla. (B) Quantitative histopathological scoring of tubular injury. Composite injury scores were significantly higher in MD-NOS1KO mice (Cre +/−) (20.2 ± 2.8) compared with Cre −/− mice (9.0 ± 3.1) following IRI-AKI. Data presented as mean ± SEM. ***P <0.001; ****P <0.0001. (C) In vivo measurements of TGF response. MD-NOS1KO mice (Cre +/−) showed enhanced TGF responses compared with Cre −/− mice both at baseline and after AKI, with more pronounced augmentation in the inducible deletion model. Data presented as mm Hg change. *P <0.05; ns = not significant. (D–G) Functional assessment of kidney injury. (D) GFR measurements showed a dramatic decline in MD-NOS1KO mice (Cre +/−) (236 ± 66 to 24 ± 22 μl/min) compared with Cre −/− mice following IRI-AKI. (E) Plasma creatinine levels were significantly elevated in MD-NOS1KO mice (Cre +/−) after IRI-AKI. (F,G) mRNA expression of kidney injury markers KIM-1 and NGAL was both significantly increased in MD-NOS1KO mice (Cre +/−) compared with Cre −/− mice following IRI-AKI. Data normalized to GAPDH and presented as fold change relative to controls. *P <0.05; **P <0.01; ***P <0.001; ns = not significant.

As we previously reported, deletion of MD NOS1 enhances TGF response. However, the augmentation of the TGF response is more evident in the inducible NOS1 deletion model used in the present study compared with the non-inducible NOS1 deletion used in the previous studies. This difference indicates the significance of long-term compensatory response (Figure 2C).

For further evaluations of the kidney injury difference by different genotypes in response to IRI-AKI, we measured the GFR. As shown in Figure 2D, MD-specific deletion of NOS1 caused a dramatic GFR decline (236 ± 66 to 24 ± 22 μl/min) compared with the control (Cre −/−) genotype. The plasma creatinine levels were also significantly higher in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice (P <0.05) in response to IRI-AKI (Figure 2E). Additionally, we measured the typical kidney injury makers [32–34] KIM-1 (kidney injury molecule-1) and NGAL (neutrophil gelatinase-associated lipocalin). The mRNA levels of KIM-1 and NGAL were both significantly increased in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice (P <0.01) in response to IRI-AKI (Figure 2F,G).

IRI-AKI induced programed cell death exacerbated in MD-NOS1KO mice

To determine whether the more severe structural injury in MD-NOS1KO mice (Cre +/−) involved programed cell death, we evaluated apoptosis markers in kidney tissue upon IRI-AKI. For evaluating and quantifying apoptosis in kidney tissue, we used TUNEL assay (terminal deoxynucleotidyl transferase dUTP Nick End Labeling). The results showed TUNEL-positive cells significantly increased in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice in response to IRI-AKI (Figure 3A,B). To explore whether the transcriptional profile tended to be pro-apoptotic, we measured TP53 mRNA and the Bax/Bcl2 ratio by qRT-PCR. As shown in Figure 4B, TP53 mRNA was significantly increased in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice in response to IRI-AKI. And the Bax/Bcl-2 expression ratio was also significantly increased in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice (Figure 4C). To determine whether the pro-apoptotic proteins also correlated with transcriptional changes, we measured pro-apoptotic cleaved form of poly(ADP-ribose) polymerase 1 (PARP1) and caspase-3. The results showed that the cleaved form of PARP1 and caspase-3 were dramatically increased in the MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice in response to IRI-AKI (Figure 4A).

Figure 3. Enhanced apoptosis in MD-NOS1KO mice (Cre +/−) following IRI-AKI.

Figure 3.

Enhanced apoptosis in MD-NOS1KO mice (Cre +/−) following IRI-AKI. (A) Representative TUNEL staining images showing apoptotic cell death in kidney sections from different experimental groups. Positive control demonstrated complete DNA fragmentation, while MD-NOS1KO mice (Cre +/−) exhibited substantially more TUNEL-positive cells compared with Cre −/− mice following IRI-AKI. Green fluorescence indicates fragmented DNA; blue indicates nuclear staining (DAPI). Scale bar = 2 μm. (B) Quantification of TUNEL-positive cells across experimental groups. MD-NOS1KO mice (Cre +/−) showed approximately a 2.5-fold increase in apoptotic cells compared with Cre −/− mice after IRI-AKI, with minimal apoptosis observed in control conditions.

Figure 4. Molecular mechanisms of enhanced injury in MD-NOS1KO mice (Cre +/−).

Figure 4.

Molecular mechanisms of enhanced injury in MD-NOS1KO mice (Cre +/−). (A–C) Pro-apoptotic signaling analysis. (A) Western blot analysis and quantification of cleaved caspase-3 and cleaved PARP-1 demonstrated increased apoptotic protein activation in MD-NOS1KO mice (Cre +/−) following IRI-AKI. GAPDH served as loading control. (B) TP53 mRNA levels were significantly elevated in MD-NOS1KO mice (Cre +/−) following IRI-AKI. (C) Bax/Bcl-2 ratio showed a pro-apoptotic shift specifically in MD-NOS1KO mice (Cre +/−) after injury. (D) Cytokine mRNA expression analysis revealed unexpected down-regulation of inflammatory markers TNF-α and IL-1β in MD-NOS1KO mice (Cre +/−), while IL-6 remained unchanged following IRI-AKI. (E) Chemokine expression patterns showed significant up-regulation of MCP-1 and CXCL1, but not MIP-2, in MD-NOS1KO mice (Cre +/−) compared with Cre −/− mice following IRI-AKI. (F) Macrophage marker analysis demonstrated differential immune cell infiltration patterns. CD68 (bone marrow-derived monocytes) significantly increased while F4/80 (resident macrophages) decreased in MD-NOS1KO mice (Cre +/−), following IRI-AKI. All qRT-PCR data normalized to GAPDH and presented as fold change relative to control conditions. *P <0.05; **P <0.01; ***P <0.001; ns = not significant.

Limited cytokine production and active chemokine signaling in MD-NOS1KO mice

To determine whether cytokine production changes may contribute to more severe injury in MD-NOS1KO mice (Cre +/−) upon IRI-AKI, we measured the cytokine profile using a commercially available cytokine array kit. The results showed the 10 most commonly expressed cytokines remain unchanged or even tend to decrease in MD-NOS1KO mice (Cre +/−) compared with the control (Cre −/−) mice in response to IRI-AKI (Supplementary Figure S2A). To confirm these surprising results, we also measured mRNA levels of TNF-alpha, Interleukin-1 beta (IL-1 β), and Interleukin-1 (IL-6). As shown in Figure 4D, the mRNA levels of TNF-alpha and IL-1 β were significantly decreased, and IL-6 remained unchanged in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice in response to IRI-AKI, which correlated well with cytokine array results. These interesting results lead us to measure possible changes of chemokines in response to IRI-AKI, so we measured the mRNA levels of MCP-1 (monocyte chemoattractant protein-1, CCl2), CXCL1 and MIP-2 (macrophage inflammatory protein, CXCL2) using qRT-PCR. As shown in Figure 4E, both MCP-1 and CXCL1 were dramatically increased in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice in response to IRI-AKI, though MIP-2 remained unchanged. To determine whether these chemokine changes translated into altered immune cell influx to the kidney, we measured two different macrophage markers, CD68 and F4/80. The marker of bone-marrow-derived monocytes, CD68 was significantly increased in the MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice in response to IRI-AKI. In contrast, the resident macrophage marker, F4/80, significantly down-regulated in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice in response to IRI-AKI (Figure 4F). To characterize macrophage polarization, we examined M2 markers CD206 and YM-1 by qPCR (Supplementary Figure S2C). CD206 expression increased after IRI-AKI in both genotypes. YM-1 was not detected in control mice but was modestly induced in control (CRE −/−) and markedly elevated in MD-NOS1KO mice (Cre +/−) following AKI (P <0.01). These findings, combined with reduced M1 cytokines (TNF-α, IL-1β), suggest that NOS1 deletion promotes M2 macrophage polarization after ischemic injury.

HIF-1α induced pro-fibrotic factor activation in MD-NOS1KO mice

To determine whether MD-specific deletion of NOS1 can accelerate the increased chemokine burst into profibrotic remodeling, we measured fibrogenic transcripts, including the marker of myofibroblast activation, α-SMA (alpha-smooth muscle actin); marker of interstitial collagen deposition, collagen-3; and inhibitor of matrix metalloproteinases, TIMP-1 (tissue inhibitor of metalloproteinase 1) [35–37]. As shown in Figure 5A, α-SMA, collagen-3, and TIMP-1 dramatically increased (more than 20-fold) in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) in response to IRI-AKI (Figure 5A). Along with mRNA levels of these molecules, Western blot analysis showed that the protein expression of α-SMA, collagen-3, and TIMP-1 is also increased in similar patterns of transcriptome increase in MD-NOS1KO mice (Cre +/−) (Figure 5B).

Figure 5. HIF-1α-mediated pro-fibrotic signaling activation in MD-NOS1KO mice (Cre +/−).

Figure 5.

HIF-1α-mediated pro-fibrotic signaling activation in MD-NOS1KO mice (Cre +/−). (A) Quantitative RT-PCR analysis of fibrogenic transcripts showing dramatic up-regulation (>20-fold) of α-SMA, collagen-3, and TIMP-1 in MD-NOS1KO mice (Cre +/−) compared with Cre −/− mice following IRI-AKI. Data normalized to GAPDH and presented as fold change relative to controls. (B) Western blot analysis confirming protein-level increases in fibrotic markers α-SMA, collagen-3, and TIMP-1 in MD-NOS1KO mice (Cre +/−), correlating with transcriptional changes. GAPDH served as loading control. (C) Representative immunohistochemistry images for HIF-1α expression in kidney sections. MD-NOS1KO mice (Cre +/−) showed substantially increased HIF-1α-positive staining compared with Cre −/− mice following IRI-AKI, while control conditions showed minimal expression. (D) Western blot analysis of HIF-1α and its regulator PHD2. HIF-1α protein levels were significantly increased in MD-NOS1KO mice (Cre +/−) following IRI-AKI, while PHD2 expression remained unchanged, indicating a PHD2-independent mechanism of HIF-1α stabilization. GAPDH served as loading control. *P <0.05; ***P <0.001.

Since HIF-1α plays a crucial role in hypoxia-induced fibrogenesis [19,38], we evaluated whether MD-specific NOS1 deletion affects HIF-1α. Immunohistochemistry (IHC) analysis of kidney tissue samples showed HIF-1α antibody-positive signaling increased in MD-NOS1KO mice (Cre +/−) compared with control (Cre −/−) mice in response to IRI-AKI (Figure 5C). Western blot analysis of HIF-1α expression correlated well with IHC results (Figure 5D). Densitometric quantification confirmed that HIF-1α protein levels were significantly elevated in MD-NOS1KO mice (Cre +/−) following ischemia reperfusion injury compared with AKI controls (Cre −/−) (2.15 ± 0.37 versus 0.74 ± 0.07, P <0.001; Supplementary Figure S1E). To test whether MD-specific NOS1 deletion also affects the principal negative regulator of HIF-1α stability, PHD2, we further analyzed the PHD2 protein expression using Western blot. To our surprise, PHD2 protein expression remained unchanged across all groups (P >0.05; Supplementary Figure S1E), showing a slight, non-significant increase in MD-NOS1KO mice (Cre +/−) compared with controls (Cre −/−) mice following IRI-AKI. This finding indicates a PHD2-independent mechanism for HIF-1α accumulation in MD-NOS1KO mice (Cre +/−) after IRI-AKI (Figure 5D).

Proteomic analysis and validation reveal genotype-dependent molecular responses to AKI in MD-NOS1KO mice

To further investigate downstream pathways influenced by MD-NOS1, we performed unbiased proteomic profiling of renal cortex from controls (Cre −/−) and KO mice (Cre +/−) with or without AKI. As shown in volcano plots (Figure 6A, Supplementary Figure S3A) and heatmap plots (Figure 6B), two-way ANOVA identified four proteins with significant genotype × treatment interactions (FDR <0.05), indicating that their AKI responses were dependent on NOS1 status. These candidates included Hp, histidine-rich glycoprotein (HRG), Tacstd2 (TROP2), and cytochrome P450 family member Cyp20a1.

Figure 6. Proteomic validation and functional markers linked to renal performance.

Figure 6.

Proteomic validation and functional markers linked to renal performance. (A) Volcano plot of proteomic changes showing significant genotype × treatment interactions. Four proteins displayed genotype-dependent responses to AKI (FDR <0.05): Hp, HRG, Tacstd2 (TROP2), and cytochrome P450 family member Cyp20a1. (B) Heatmap of normalized protein abundances illustrating distinct clustering by genotype and treatment. (C-E) Functional validation linking proteomic markers to measured GFR in mice. (C) Plasma Hp strongly correlated with GFR (R2 = 0.76; P <0.0001). (D) Tacstd2 mRNA levels correlated with GFR (R2 = 0.75; P <0.0001). (E) Cyp20a1 mRNA similarly associated with GFR (R2 = 0.60; P <0.0001). Each point represents one animal (circles = Cre −/−, triangles = MD-NOS1KO); solid lines show fitted regressions with shaded 95% CIs. These relationships link molecular signatures to renal function, reinforcing the genotype-specific vulnerability of MD-NOS1KO mice (Cre +/−) to ischemic AKI.

Targeted assays confirmed three of these interaction effects. Plasma Hp concentrations were markedly higher in KO compared with control (Cre −/−) mice and rose disproportionately in KO following AKI, yielding a highly significant interaction (F(1,16) = 28.7, P <0.0001; Figure 6C). Renal Tacstd2 mRNA levels were selectively induced by AKI in KO but not control (Cre −/−) mice (interaction P <0.01; Figure 6D). Similarly, Cyp20a1 transcripts showed a significant interaction with greater induction in KO-AKI compared with other groups (interaction P <0.05; Figure 6E). These results corroborate the proteomic discovery and demonstrate that MD NOS1 deletion reprograms the molecular response to ischemic injury across inflammatory, epithelial repair/fibrosis, and metabolic pathways.

In addition to the four proteins with significant genotype × treatment interactions, we examined pathway-level changes in the proteomic dataset. In KO mice (Cre +/−), the AKI versus Sham comparison revealed strong enrichment of metabolic and catabolic programs, including fatty acid, amino acid, and organic acid metabolism, together with wound healing and coagulation pathways (Supplementary Figure S3B). By contrast, in the KO versus control (Cre −/−) comparison under AKI conditions, enriched categories highlighted angiogenesis, vascular development, endocytosis, acute inflammatory response, and the acute-phase response (Supplementary Figure S3C). These analyses indicate that deletion of MD NOS1 reprograms the response to ischemia, coupling altered metabolic adaptation with activation of inflammatory and vascular pathways.

Molecular markers and functional relevance to human kidney injury

To connect molecular signatures with kidney function, we first evaluated the key proteomic markers in our mouse model. Our regression analyses revealed a strong association between plasma Hp levels and GFR (R2 = 0.76; P <0.0001). Similarly, Tacstd2 and Cyp20a1 mRNA levels showed close relationships with GFR (R2 = 0.75 and 0.60; P <0.0001 for both). These findings point to a tight link between these molecular markers and renal function (Figure 6C–E). Notably, these patterns reflect what we observe in the KO-AKI phenotype, grounding the molecular response firmly within a functional framework.

Turning to clinical relevance, we next looked at NOS1 expression in human kidneys using single-nucleus RNA-seq data from the Kidney Precision Medicine Project (KPMP). Focusing on MD/TAL cells, we found that both NOS1 expression and the proportion of NOS1+ cells dropped in AKI patients compared with healthy donors (Figure 7A–D). To assess clinical relevance, we analyzed KPMP single-nucleus RNA-seq data. Within MD/TAL cells, both NOS1 expression and the fraction of NOS1+ cells were lower in AKI compared with healthy kidneys (Figure 7A–D). These findings suggest that NOS1-linked mechanisms observed in mice extend to human kidney injury contexts, underscoring conserved MD stress responses in AKI.

Figure 7. Cross-species validation of NOS1 loss in AKI.

Figure 7.

Cross-species validation of NOS1 loss in AKI. (A) Mean NOS1 expression across kidney cell types from KPMP single-nucleus RNA-seq (n = 304,989 cells, 80 patients). MD exhibits highest expression, followed by cortical TAL (C-TAL), while adaptive/injured TAL (aTAL1) shows markedly reduced NOS1. (B) Percentage of cells expressing NOS1 (>0 counts) decreases in injury-associated states, indicating loss of NO-producing cell populations. (C) Distribution of TAL cell subtypes in the KPMP dataset. C-TAL represents the predominant ‘healthy’ TAL population (n = 19,237), while aTAL1 (n = 8822) represents an adaptive/maladaptive/repairing state with reduced NOS1. (D) Violin plots show NOS1 expression distribution across TAL cell states. MD cells (purple) show the highest expression (mean 1.38, 64.2% positive). C-TAL cells (blue, mean 0.556, 35.5% positive) represent the baseline TAL state. aTAL1 cells (red, mean 0.160, 12.0% positive) represent an injury-associated state with a 71% reduction in NOS1 compared with C-TAL (P <2.2 × 10−16, Mann-Whitney test). Box plots (white) show median and quartiles.

Discussion

In the present study, we established a novel MD-NOS1KO model and revealed that loss of MD-derived NO dramatically worsens ischemia-reperfusion injury. Mice lacking the MD-NOS1 gene showed much worse damage to their kidney tubules and developed fibrogenic factors compared with normal mice. Functionally, GFR significantly declined in MD-NOS1KO mice (Cre +/−) after IR, and classic injury markers KIM-1 and NGAL were markedly higher. These data suggest that MD NOS1 activity is reno-protective in AKI, consistent with prior reports that NO donors improve outcomes in renal IR. For example, Garcia-Criado et al. showed that giving an NO donor at reperfusion improved renal function and blunted the surge of TNF-α and IL-1α during AKI [39]. By contrast, the absence of MD NOS1 leaves the kidney exposed to uncontrolled hypoxic insult and inflammation, driving cell death and dysfunction. And MD NOS1 expression significantly reduced following kidney IRI. The reduced MD NOS1 promotes enhanced TGF responses, which contributes to the enhanced vascular response of the afferent arterioles and decreased tubular blood flow. Deletion of the MD NOS1 further enhances TGF response following kidney IRI. These changes exacerbate AKI due to more severe constriction of the afferent arterioles and reduction of the blood supply to the tubules.

Previously, we generated a conventional MD-NOS1KO line with a non-inducible NKCC2 cre line. To minimize the potential compensatory mechanisms that may be activated during development, we developed a new inducible NKCC2 cre line in the present study to examine the significance of the MD NOS1 and TGF mechanisms in the development of IRI-AKI.

Mechanistically, our findings implicate a hypoxia–HIF-1α–fibrosis axis activated by MD NOS1 deletion. Loss of NOS1-derived NO likely reduces renal medullary oxygenation; indeed, we observed a strong increase in HIF-1α levels in MD-NOS1KO kidneys after IR. HIF-1α is a master regulator of the hypoxic response, and its activation is well known to promote fibrogenesis. Higgins et al. showed that epithelial HIF-1α drives epithelial-mesenchymal transition, collagen deposition, and interstitial fibrosis after injury [19]. In the present study, MD-NOS1KO kidneys exhibited massive up-regulation of profibrotic genes (α-SMA, collagen-3, and TIMP-1). Indeed, NO can inhibit mitochondrial O2 consumption and redistribute O2 to prolyl hydroxylases, limiting HIF-1α stabilization [40]. In our present model, the absence of MD NO would remove this brake on HIF-1α, consistent with the observed HIF-1α rise and downstream fibrogenic signaling. Thus, loss of NOS1 creates an unchecked hypoxic signal that drives a profibrotic program, linking MD physiology directly to maladaptive AKI-to-fibrosis progression.

Another notable finding was enhanced apoptosis in MD-NOS1KO kidneys. TUNEL staining and raised p53 and Bax/Bcl-2 ratios indicate that NOS1 deletion amplifies programmed cell death after IR. This aligns with previous evidence that NO can protect tubular epithelia from death signals. In obstructive kidney injury, for example, NOS inhibition markedly increased tubular apoptosis [41]. In our present study, the absence of NOS1 may deprive cells of cGMP-dependent survival signals or allow pro-apoptotic pathways (like oxidative stress) to dominate. The increase in cleaved PARP and caspase-3 in MD-NOS1KO kidneys underscores that apoptosis is a key component of the worsened injury.

Interestingly, cytokine versus chemokine signaling was dysregulated in MD-NOS1KO mice (Cre +/−). Despite more tissue damage, MD-NOS1KO mice (Cre +/−) had lower TNF-α and IL-1β transcripts than controls (Cre −/−) after IR, while MCP-1 (CCL2) and CXCL1 were dramatically elevated. MCP-1 is a potent monocyte/macrophage chemoattractant, and its up-regulation usually correlates with injury severity [42]. Our data show MCP-1 (and CD68+ cell infiltration) were greatly increased in MD-NOS1KO kidneys, suggesting that monocyte-driven inflammation is dominant. The paradoxically low TNF-α/IL-1β might reflect altered signaling in the absence of NOS1: MD-derived NO may normally amplify some inflammatory cytokine responses, like its role in macrophages [43], so its loss blunts those pathways. Regardless, the net effect in MD-NOS1KO kidneys is a ‘chemokine surge’ that summons inflammatory cells. This is likely maladaptive: chronic MCP-1 signaling has been linked to fibrosis, as it drives macrophages to release TGF-β and collagen-promoting factors. The shift from resident F4/80+ macrophages (decreased) to infiltrating CD68+ monocytes (increased) in MD-NOS1KO suggests a pro-fibrotic inflammatory milieu.

Our proteomic and validation studies provide an additional molecular dimension to the pathophysiological role of MD-NOS1 in AKI. Analysis of interaction between genotype (NOS1 KO versus Sham) and treatment (with AKI versus without AKI) in significantly differentially expressed proteins shows Hp predominantly increased in both proteomics and plasma ELISA. Hp is an acute-phase protein that eliminates free hemoglobin and rescues the oxidative injury, and its selective induction in KO-AKI suggests that NOS1 deficiency might amplify systemic inflammatory and oxidative stress pathways.

Importantly, Hp is readily measurable in plasma, raising the possibility that it could serve as a translational biomarker of NOS1-dependent AKI susceptibility. Tacstd2 (TROP2), a marker of epithelial regeneration and fibrosis, was selectively up-regulated in KO-AKI, aligning with our histological evidence of enhanced fibrotic remodeling. Cyp20a1 resides in proximal tubules of kidney; our data shows it specifically increased in KO-AKI. This means AKI caused hypoxia can induce an alteration of metabolic and mitochondrial stress responses. These findings demonstrate that MD-NOS1 deletion not only exacerbates histological injury but also reprograms downstream molecular responses such as inflammation, epithelial repair/fibrosis, and metabolic adaptation. These molecular signatures provide strong evidence of our proposed hypoxia–HIF-1α–fibrosis axis and provide potential translational readouts linking MD physiology to systemic biomarkers. Across species, NOS1-associated molecular programs converge on renal functional integrity. In mice, Hp, Tacstd2, and Cyp20a1 track precisely with measured GFR. In human kidney tissue, NOS1 expression within MD/TAL is reduced during AKI, paralleling the murine phenotype. These results point to conserved pathways linking MD NOS1 loss with kidney injury susceptibility, while recognizing that human data remain associative.

The cross-talk between MD NOS1 and the renin-angiotensin system likely contributes to the exaggerated AKI phenotype in our model. NOS1-derived NO is a well-established modulator of TGF, and its loss would be expected to leave efferent arteriolar constriction unopposed [44,45]. Moreover, angiotensin II enhances TGF-mediated vasoconstriction through calcium-dependent mechanisms, potentially compounding the hemodynamic consequences of NOS1 deletion [46]. Sympathetic nervous system activation, which occurs during ischemia-reperfusion injury and promotes both renin release and renal vasoconstriction, may further amplify these effects [47]. One limitation is that while the MD-mediated TGF response is central to the AKI phenotype, the influence of NOS1 on renin-angiotensin and sympathetic nerve system activity may also be relevant. Human eGFR data are limited to a single time point, and the number of donors is modest; causality cannot be determined, and covariates are restricted. And renal blood flow and vascular resistance were not directly measured in the present study. However, the enhanced TGF response observed in MD-NOS1KO mice (Cre +/−) would be expected to increase afferent arteriolar constriction and reduce glomerular perfusion. This additional hemodynamic burden during reperfusion, when oxygen delivery is already compromised, could exacerbate medullary hypoxia and worsen tubular injury. Our observation of elevated HIF-1α expression in knockout kidneys supports this interpretation. The potential contribution of altered renal hemodynamics to AKI severity in this model warrants further investigation. Future studies using laser Doppler flowmetry or contrast-enhanced imaging could directly quantify how these hemodynamic changes contribute to injury outcomes. Our study used only male mice for AKI experiments, which represents a limitation. Both sexes were included for initial characterization of the NKCC2-CreERT2 line to confirm comparable Cre activity and transporter function. However, sex hormones are known to modulate renal hemodynamics and ischemic injury susceptibility, with estrogen generally conferring protection [48]. Whether MD NOS1 deletion differentially affects AKI severity in female mice remains an important question for future investigation.

Taken together, our findings position MD-NOS1 and NO-mediated TGF as critical modulators of AKI severity. In normal mice, MD NOS1-derived NO buffers TGF and preserves microcirculatory oxygenation by maintaining normal blood flow; its loss renders kidneys prone to severe hypoxic injury. The result is a cascade of uncontrolled inflammation, cell death, and fibrotic activation. Importantly, the present study is the first to show that targeting the JGA/MD segment alters AKI outcomes via a HIF-1α–driven mechanism. The potent profibrotic response in MD-NOS1KO resembles the maladaptive AKI-to-CKD transition, highlighting MD NOS1 as a potential gatekeeper of renal repair [49].

Supplementary Material

1

Clinical perspectives.

  • Background: AKI remains a major cause of morbidity and mortality, and its pathophysiology is not fully understood. NO from MD NOS1 plays a critical role in regulating TGF and renal oxygen delivery, but its contribution to ischemic kidney injury has not been directly tested.

  • Results summary: Using a newly generated MD-NOS1KO mouse model, we show that NOS1 deletion markedly worsens ischemia–reperfusion–induced AKI, with greater tubular damage, apoptosis, and fibrotic activation. Proteomic and transcriptomic profiling identified functional markers (Hp, Tacstd2, Cyp20a1) tightly linked to GFR, and human kidney single-cell data revealed reduced NOS1 expression in AKI compared with healthy tissue.

  • Significance to human health: These findings establish MD NOS1 as a key determinant of renal resilience to ischemic stress. The identified molecular markers and loss of NOS1 expression in human AKI highlight a translational pathway connecting local tubuloglomerular regulation to global kidney injury mechanisms, suggesting new therapeutic and biomarker directions.

Acknowledgement

We especially thank Elizabeth Wood at the USF proteomic core for the excellent work.

Funding

The present work was supported by National Institutes of Health [Grants DK134000, DK134028, and DK138092 to R.L. and DK098141 and DK132066 to J.A.M.]. J.B.L. was supported by a Ben J. Lipps Research Fellowship from KidneyCure.

Abbreviations

α-SMA

alpha-smooth muscle actin

AKI

acute kidney injury

GFR

glomerular filtration rate

HIF

hypoxia-inducible factor

HIF-1α

hypoxia-inducible factor-1α

HRG

histidine-rich glycoprotein

IHC

immunohistochemistry

IL-1β

interleukin-1 beta

IRI-AKI

ischemia–reperfusion injury-induced acute kidney injury

KIM-1

kidney injury molecule-1

KPMP

Kidney Precision Medicine Project

MCP-1

monocyte chemoattractant protein-1

MD-NOS1KO

MD-specific NOS1 knockout

NGAL

neutrophil gelatinase-associated lipocalin

NO

nitric oxide

NOS

nitric oxide synthase

PARP1

poly (ADP-ribose) polymerase 1

SD

standard deviation

TAL

thick ascending limb

TGF

tubuloglomerular feedback

TIMP-1

tissue inhibitor of metalloproteinase 1

Footnotes

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Data Availability

All raw data generated in the present study (qPCR, ELISA, Western blot, histology, and figure source files) have been deposited in Mendeley Data and are available without restriction at https://doi.org/10.17632/y2c5xrrchv.1. Additional datasets and analysis scripts are available from the corresponding author upon reasonable request. Public human single-nucleus RNA-seq data were obtained from the KPMP repository and GEO accession GSE201575. Additional datasets and analysis scripts are available from the corresponding author upon reasonable request.

References

  • 1.Sawhney S, Marks A, Fluck N, Levin A, Prescott G and Black C. (2017) Intermediate and long-term outcomes of survivors of acute kidney injury episodes: a large population-based cohort study. Am. J. Kidney Dis. 69, 18–28, 10.1053/j.ajkd.2016.05.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Wang HE, Muntner P, Chertow GM and Warnock DG. (2012) Acute kidney injury and mortality in hospitalized patients. Am. J. Nephrol. 35, 349–355, 10.1159/000337487 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Carlstrom M. (2021) Nitric oxide signalling in kidney regulation and cardiometabolic health. Nat. Rev. Nephrol. 17, 575–590, 10.1038/s41581-021-00429-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Raij L. (2008) Nitric oxide and cardiovascular and renal effects. Osteoarthritis Cartilage 16, S21–S26, 10.1016/S1063-4584(08)60009-6 [DOI] [PubMed] [Google Scholar]
  • 5.Lee J. (2008) Nitric oxide in the kidney : its physiological role and pathophysiological implications. Electrolyte Blood Press. 6, 27–34, 10.5049/EBP.2008.6.1.27 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Herrera M and Garvin JL. (2005) Recent advances in the regulation of nitric oxide in the kidney. Hypertension 45, 1062–1067, 10.1161/01.HYP.0000159760.88697.1e [DOI] [PubMed] [Google Scholar]
  • 7.Forbes MS, Thornhill BA, Park MH and Chevalier RL. (2007) Lack of endothelial nitric-oxide synthase leads to progressive focal renal injury. Am. J. Pathol. 170, 87–99, 10.2353/ajpath.2007.060610 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Oliveira F, Assreuy J and Sordi R. (2022) The role of nitric oxide in sepsis-associated kidney injury. Biosci. Rep. 42, 10.1042/BSR20220093 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Ren YL, Garvin JL and Carretero OA. (2000) Role of macula densa nitric oxide and cGMP in the regulation of tubuloglomerular feedback. Kidney Int. 58, 2053–2060, 10.1111/j.1523-1755.2000.00377.x [DOI] [PubMed] [Google Scholar]
  • 10.Thorup C, Erik A and Persson G. (1996) Macula densa derived nitric oxide in regulation of glomerular capillary pressure. Kidney Int. 49, 430–436, 10.1038/ki.1996.62 [DOI] [PubMed] [Google Scholar]
  • 11.Zhang J, Wei J, Jiang S, Xu L, Wang L, Cheng F. et al. (2019) Macula densa SGLT1-NOS1-tubuloglomerular feedback pathway, a new mechanism for glomerular hyperfiltration during hyperglycemia. J. Am. Soc. Nephrol. 30, 578–593, 10.1681/ASN.2018080844 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Persson AE, Brown R, Liu R and Ollerstam A. (2002) Nitric oxide modulates and adenosine mediates the tubuloglomerular feedback mechanism. Acta Physiol. Scand. 176, 91–94, 10.1046/j.1365-201X.2002.01026.x [DOI] [PubMed] [Google Scholar]
  • 13.Welch WJ, Wilcox CS and Thomson SC. (1999) Nitric oxide and tubuloglomerular feedback. Semin. Nephrol. 19, 251–262 [PubMed] [Google Scholar]
  • 14.Bosse HM, Bohm R, Resch S and Bachmann S. (1995) Parallel regulation of constitutive NO synthase and renin at JGA of rat kidney under various stimuli. Am. J. Physiol. 269, F793–F805, 10.1152/ajprenal.1995.269.6.F793 [DOI] [PubMed] [Google Scholar]
  • 15.Liu C, Wang X, Parris C, Pang Q, Naeem MU and Wang L. (2023) Macula densa nitric oxide synthase 1 controls renin release and renin-dependent blood pressure changes. Discov. Med. 35, 525–532, 10.24976/Discov.Med.202335177.53 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Nakayama T, Sato W, Kosugi T, Zhang L, Campbell-Thompson M, Yoshimura A. et al. (2009) Endothelial injury due to eNOS deficiency accelerates the progression of chronic renal disease in the mouse. Am. J. Physiol. Renal. Physiol. 296, F317–F327, 10.1152/ajprenal.90450.2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Atanasova I, Burke TJ, McMurtry IF and Schrier RW. (1995) Nitric oxide synthase inhibition and acute renal ischemia: effect on systemic hemodynamics and mortality. Ren. Fail. 17, 389–403, 10.3109/08860229509037604 [DOI] [PubMed] [Google Scholar]
  • 18.Bondi CD, Rush BM, Hartman HL, Wang J, Al-Bataineh MM, Hughey RP. et al. (2022) Suppression of NRF2 activity by HIF-1alpha promotes fibrosis after ischemic acute kidney injury. Antioxidants (Basel) 11, 1810, 10.3390/antiox11091810 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Higgins DF, Kimura K, Bernhardt WM, Shrimanker N, Akai Y, Hohenstein B. et al. (2007) Hypoxia promotes fibrogenesis in vivo via HIF-1 stimulation of epithelial-to-mesenchymal transition. J. Clin. Invest. 117, 3810–3820, 10.1172/JCI30487 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Chavali K, Coker H, Youngblood E and Karaduta O. (2024) Proteogenomics in nephrology: a new frontier in nephrological research. Curr. Issues Mol. Biol. 46, 4595–4608, 10.3390/cimb46050279 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Noren A, Boi R, Pullerits R, Molne J, Ebefors K, Friman S. et al. (2025) Proteomic analysis of human kidney biopsies unveils emerging acute kidney injury very early after liver graft reperfusion. J. Transl. Med. 23, 658, 10.1186/s12967-025-06695-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Pejchinovski I, Turkkan S and Pejchinovski M. (2023) Recent advances of proteomics in management of acute kidney injury. Diagnostics (Basel) 13, 2648, 10.3390/diagnostics13162648 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Chen W, Wang L, Liang P, Mast J, Mathis C, Liu CY. et al. (2022) Reducing ischemic kidney injury through application of a synchronization modulation electric field to maintain Na(+)/K(+)-ATPase functions. Sci. Transl. Med. 14, eabj4906, 10.1126/scitranslmed.abj4906 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Wei J, Wang Y, Zhang J, Wang L, Fu L, Cha BJ. et al. (2019) A mouse model of renal ischemia-reperfusion injury solely induced by cold ischemia. Am. J. Physiol. Renal. Physiol. 317, F616–F622, 10.1152/ajprenal.00533.2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Wei J, Zhang J, Wang L, Jiang S, Fu L, Buggs J. et al. (2019) New mouse model of chronic kidney disease transitioned from ischemic acute kidney injury. Am. J. Physiol. Renal. Physiol. 317, F286–F295, 10.1152/ajprenal.00021.2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Zhang J, Wang X, Wei J, Wang L, Jiang S, Xu L. et al. (2020) A two-stage bilateral ischemia-reperfusion injury-induced AKI to CKD transition model in mice. Am. J. Physiol. Renal. Physiol. 319, F304–F311, 10.1152/ajprenal.00017.2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Zhang J, Wang X, Cui Y, Jiang S, Wei J, Chan J. et al. (2021) Knockout of macula densa neuronal nitric oxide synthase increases blood pressure in db/db Mice. Hypertension 78, 1760–1770, 10.1161/HYPERTENSIONAHA.121.17643 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Bourqui L, Winter DV, Odermatt A, Loffing-Cueni D and Loffing J. (2023) A novel mouse model for an inducible gene modification in the renal thick ascending limb. Am. J. Physiol. Renal. Physiol. 324, F446–F460, 10.1152/ajprenal.00250.2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Tojo A, Madsen KM and Wilcox CS. (1995) Expression of immunoreactive nitric oxide synthase isoforms in rat kidney. Effects of dietary salt and losartan. Jpn. Heart J. 36, 389–398, 10.1536/ihj.36.389 [DOI] [PubMed] [Google Scholar]
  • 30.Bachmann S, Bosse HM and Mundel P. (1995) Topography of nitric oxide synthesis by localizing constitutive NO synthases in mammalian kidney. Am. J. Physiol. 268, F885–F898, 10.1152/ajprenal.1995.268.5.F885 [DOI] [PubMed] [Google Scholar]
  • 31.Lu Y, Wei J, Stec DE, Roman RJ, Ge Y, Cheng L. et al. (2016) Macula densa nitric oxide synthase 1β protects against salt-sensitive hypertension. J. Am. Soc. Nephrol. 27, 2346–2356, 10.1681/ASN.2015050515 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Khalil R, Elghadban H, Abdelsalam M and Tawfik M. (2020) Kidney injury molecule-1: a potential marker of renal recovery after laparoscopic sleeve gastrectomy. Kidney Res. Clin. Pract. 39, 162–171, 10.23876/j.krcp.19.109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Ostermann M, Legrand M, Meersch M, Srisawat N, Zarbock A and Kellum JA. (2024) Biomarkers in acute kidney injury. Ann. Intensive Care 14, 145, 10.1186/s13613-024-01360-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Verbrugge FH, Dupont M, Shao Z, Shrestha K, Singh D, Finucan M. et al. (2013) Novel urinary biomarkers in detecting acute kidney injury, persistent renal impairment, and all-cause mortality following decongestive therapy in acute decompensated heart failure. J. Card. Fail. 19, 621–628, 10.1016/j.cardfail.2013.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Genovese F, Akhgar A, Lim SS, Farris AB, Battle M, Cobb J. et al. (2021) Collagen type iii and vi remodeling biomarkers are associated with kidney fibrosis in lupus nephritis. Kidney360 2, 1473–1481, 10.34067/KID.0001132021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Ina K, Kitamura H, Tatsukawa S and Fujikura Y. (2011) Significance of alpha-SMA in myofibroblasts emerging in renal tubulointerstitial fibrosis. Histol. Histopathol. 26, 855–866 [DOI] [PubMed] [Google Scholar]
  • 37.Tan RJ and Liu Y. (2024) Matrix metalloproteinases in kidney homeostasis and diseases: an update. Am. J. Physiol. Renal. Physiol. 327, F967–F984, 10.1152/ajprenal.00179.2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Xiong A and Liu Y. (2017) Targeting hypoxia inducible factors-1alpha as a novel therapy in fibrosis. Front. Pharmacol. 8, 326, 10.3389/fphar.2017.00326 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Garcia-Criado FJ, Eleno N, Santos-Benito F, Valdunciel JJ, Reverte M, Lozano-Sanchez FS. et al. (1998) Protective effect of exogenous nitric oxide on the renal function and inflammatory response in a model of ischemia-reperfusion. Transplantation 66, 982–990, 10.1097/00007890-199810270-00003 [DOI] [PubMed] [Google Scholar]
  • 40.Olson N and van der Vliet A. (2011) Interactions between nitric oxide and hypoxia-inducible factor signaling pathways in inflammatory disease. Nitric Oxide 25, 125–137, 10.1016/j.niox.2010.12.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Miyajima A, Chen J, Poppas DP, Vaughan ED Jr. and Felsen D. (2001) Role of nitric oxide in renal tubular apoptosis of unilateral ureteral obstruction. Kidney Int. 59, 1290–1303, 10.1046/j.1523-1755.2001.0590041290.x [DOI] [PubMed] [Google Scholar]
  • 42.Liu Y, Xu K, Xiang Y, Ma B, Li H, Li Y. et al. (2023) Role of MCP-1 as an inflammatory biomarker in nephropathy. Front. Immunol. 14, 1303076, 10.3389/fimmu.2023.1303076 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Baig MS, Zaichick SV, Mao M, de Abreu AL, Bakhshi FR, Hart PC. et al. (2015) NOS1-derived nitric oxide promotes NF-kappaB transcriptional activity through inhibition of suppressor of cytokine signaling-1. J. Exp. Med. 212, 1725–1738, 10.1084/jem.20140654 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Ollerstam A and Persson AE. (2002) Macula densa neuronal nitric oxide synthase. Cardiovasc. Res. 56, 189–196, 10.1016/S0008-6363(02)00536-9 [DOI] [PubMed] [Google Scholar]
  • 45.Schnermann J. (2015) Concurrent activation of multiple vasoactive signaling pathways in vasoconstriction caused by tubuloglomerular feedback: a quantitative assessment. Annu. Rev. Physiol. 77, 301–322, 10.1146/annurev-physiol-021014-071829 [DOI] [PubMed] [Google Scholar]
  • 46.Lai EY, Patzak A, Persson AE and Carlstrom M. (2009) Angiotensin II enhances the afferent arteriolar response to adenosine through increases in cytosolic calcium. Acta. Physiol. (Oxf.) 196, 435–445, 10.1111/j.1748-1716.2009.01956.x [DOI] [PubMed] [Google Scholar]
  • 47.Li Q, Deng Y, Liu L, Zhang C, Cai Y, Zhang T. et al. (2021) Sympathetic denervation ameliorates renal fibrosis via inhibition of cellular senescence. Front. Immunol. 12, 823935, 10.3389/fimmu.2021.823935 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Neugarten J and Golestaneh L. (2019) Influence of sex on the progression of chronic kidney disease. Mayo Clin. Proc. 94, 1339–1356, 10.1016/j.mayocp.2018.12.024 [DOI] [PubMed] [Google Scholar]
  • 49.Li M, Bahena Lopez J, Ni R, Wei J, Dhanireddy K, Subramanian V. et al. (2025) Data for: macula densa–specific nos1 knockout determines susceptibility to ischemic acute kidney injury. Mendeley Data, 10.17632/y2c5xrrchv.1 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1

Data Availability Statement

All raw data generated in the present study (qPCR, ELISA, Western blot, histology, and figure source files) have been deposited in Mendeley Data and are available without restriction at https://doi.org/10.17632/y2c5xrrchv.1. Additional datasets and analysis scripts are available from the corresponding author upon reasonable request. Public human single-nucleus RNA-seq data were obtained from the KPMP repository and GEO accession GSE201575. Additional datasets and analysis scripts are available from the corresponding author upon reasonable request.

RESOURCES