Inhibition of PGK1 ameliorates acute kidney injury through inactivating the PKM2/ALOX12/ferroptosis pathway in a study with male mice

Zhu, Xue-Xue; Meng, Xin-Yu; Zhang, Ao-Yuan; Zhao, Chen-Yang; Wei, Yi-Ting; Wen, Yuan-Yuan; Su, Jia-Bao; Fu, Xiao; Chen, Guo; Xu, An-Jing; Wang, Meng-Yuan; Ji, Le-Ming; Bao, Neng; Bao, Zheng-Yang; Li, Na; Lu, Qing-Bo; Sun, Hai-Jian

doi:10.1038/s41467-025-64480-1

Download PDF

Article
Open access
Published: 24 October 2025

Inhibition of PGK1 ameliorates acute kidney injury through inactivating the PKM2/ALOX12/ferroptosis pathway in a study with male mice

Xue-Xue Zhu¹^na1,
Xin-Yu Meng¹^na1,
Ao-Yuan Zhang¹^na1,
Chen-Yang Zhao¹^na1,
Yi-Ting Wei¹,
Yuan-Yuan Wen¹,
Jia-Bao Su²,
Xiao Fu¹,
Guo Chen¹,
An-Jing Xu¹,
Meng-Yuan Wang¹,
Le-Ming Ji¹,
Neng Bao³,
Zheng-Yang Bao ORCID: orcid.org/0000-0002-6953-4954⁴,
Na Li⁴,
Qing-Bo Lu^1,5^na1 &
…
Hai-Jian Sun ORCID: orcid.org/0000-0002-7791-5335^1,5,6

Nature Communications volume 16, Article number: 9436 (2025) Cite this article

2763 Accesses
15 Altmetric
Metrics details

Subjects

Abstract

Despite decades of studies into the mechanisms underlying acute kidney injury (AKI), the effective clinical therapies are still limited. Here we report that the development of AKI was attenuated by conditional knockout of phosphoglycerate kinase 1 (PGK1) in renal tubular epithelial cells (RTECs) and aggravated by specific overexpression of PGK1 and administration of 3-phosphoglycerate (3-PG) in male mice. PGK1 interacted with pyruvate kinase M2 (PKM2), inducing PKM2 phosphorylation, promoting the formation of the PKM2 dimer and the subsequent nuclear translocation of PKM2. In the nucleus, the interaction of PKM2 with Pknox1 potentiated the enrichment of Pknox1 within ALOX12 promoters, which resulted in ALOX12-mediated ferroptosis in RTECs. Our drug screening experiments identified L7DG as a potential PGK1 inhibitor which exhibited protective effects against ischemic/reperfusion (I/R)-induced renal injury. Overall, we establish that genetic and pharmacological inhibition of PGK1 might be proposed as an avenue for managing AKI.

Introduction

Acute kidney injury (AKI), which is characterized by a sharp decline in renal function, remains a health burden due to its high morbidity and mortality^1,2. Severe AKI episodes can lead to abnormal cellular repair and fibrosis, eventually resulting in chronic kidney disease (CKD) and end-stage renal disease (ESRD)^3,4. The primary causes of AKI include ischemia/reperfusion (I/R) injury, nephrotoxic insults, and sepsis⁵. Major pathophysiological characteristics of AKI are programmed cell death and chronic inflammation^6,7. Reduction of AKI is a critical clinical issue that has raised considerable concern among physicians and scientists^8,9. Regrettably, there is a lack of definitive treatment options for AKI. Hence, a better understanding of the molecular mechanisms underlying the pathogenesis of AKI could provide insights into therapeutic targets for AKI.

As a critical enzyme in glycolysis, phosphoglycerate kinase 1 (PGK1) is responsible for catalyzing the conversion of 1,3-bisphosphoglycerate to 3-phosphoglycerate (3-PG)^10,11. In addition to its role in glycolysis, PGK1 also functions as a polymerase α cofactor protein, thus influencing the tricarboxylic acid cycle, DNA replication, and repair¹⁰. Post-translational modifications such as methylation, phosphorylation, O-GlcNAcylation, and acetylation can activate PGK1, particularly in cancers^{10,12,13,14,15,16,17}. Elevated PGK1 levels are linked to tumorigenesis, progression, and resistance to chemoradiotherapy¹⁰. PGK1 is highly expressed in the renal system¹⁸, while upregulation of PGK1 is positively correlated with the development and progression of clear cell renal cell carcinoma^19,20. Suppression of metabolic enzyme activity of PGK1 has recently been reported to inhibit the proliferation of renal clear cell carcinoma²¹. Thus, emerging evidence suggests that PGK1 may play a role in renal disease. However, the actual role of PGK1 in AKI pathogenesis is still unresolved. In this regard, the precise mechanisms through which PGK1 influences renal pathology should be elucidated via further research to explore its potential as a therapeutic target in AKI.

Here, we discover consistent upregulation of PGK1 in cellular models, murine models, and human biopsies of AKI, which correlates with renal injury. Consequently, we hypothesized that PGK1 may regulate the pathophysiological processes of AKI. In vivo experiments were conducted to test this hypothesis using PGK1 conditional knockout or overexpression mice and in vitro studies with PGK1-overexpressed or PGK1-silenced renal tubular epithelial cells (RTECs). We also identified the downstream mechanisms of PGK1 using RNA sequencing, co-immunoprecipitation, and mass spectrometry analyses. Additionally, we evaluated PGK1-targeted drug therapy in a mouse model of AKI. Our data revealed that the interaction of PGK1 with PKM2 contributed to renal injury by inducing phosphorylation, dimer formation, and nuclear translocation of PKM2. The nuclear PKM2 recruited the transcription factor pknox1 to the promoters of Alox12, resulting in the upregulation of Alox12 and following ferroptosis in AKI. Targeting PGK1 could provide a promising strategy to prevent and treat AKI.

Results

Upregulation of PGK1 and 3-PG in mouse models and human biopsies induced by AKI

We measured levels of PGK1 and 3-PG in I/R, cisplatin, or lipopolysaccharide (LPS)-induced AKI mouse models as well as in human AKI patient biopsies to assess the possible role of PGK1 in AKI. We observed higher protein and mRNA levels of PGK1 in the renal tissues in mice after renal I/R surgery via immunoblotting and RT-PCR techniques (Fig. 1a, b). Also, we observed higher expression of PGK1 in renal tubules of I/R-induced mice, as evidenced by immunofluorescence double staining and immunohistochemistry (Fig. 1c). In parallel, we found that the contents of 3-PG in both serum and kidneys were substantially elevated in I/R-induced AKI mice (Fig. 1d, e), along with higher PGK1 activity (Fig. 1f). The upregulations of PGK1 and 3-PG were also detected in other models of AKI mice subjected to intraperitoneal injection cisplatin (Fig. 1g–l) or LPS (Fig. 1m–r). The translational and transcriptional levels of PGK1 were higher in the renal tissues from AKI patients than those in healthy controls (Fig. 1s–t). With the aid of immunofluorescence staining and immunohistochemistry, we found that PGK1 was abundantly expressed in the kidneys of AKI patients, with the positive signals mostly in the renal tubules (Fig. 1u). We found that serum levels of 3-PG tended to be higher in patients suffering from AKI (Fig. 1v), which was positively associated with AKI-related markers, blood urea nitrogen (BUN), and serum creatinine (Scr) (Fig. 1w, x). In support of this observation, the protein expression of PGK1 displayed a positive correlation with BUN and Scr levels, respectively, in AKI patients (Supplementary Fig. S1a, b). Collectively, these results indicate a positive association between PGK1/3-PG levels and AKI severity. Furthermore, the correlation between PGK1 protein expression and BUN/Scr levels reinforced the idea that PGK1/3-PG may play a role in AKI-induced renal injury. These findings suggest that PGK1/3-PG could serve as potential biomarkers for AKI severity, and it is possible that targeting PGK1-mediated metabolic pathways might be a viable therapeutic strategy for mitigating AKI progression.

To compare differences in expression levels of PGK1 in several human kidney cells, we explored the KIT renal single-cell proteomics database (http://humphreyslab.com/SingleCell/) to compare the differences in PGK1 expression levels in kidneys of human. Using retrieval analysis, we discovered wide expression of PGK1 in several inherent kidney cells, with a higher expression level of the protein in proximal RTECs (Supplementary Fig. S2a-d). Consistently, we ascertained major expression of PGK1 in RTECs (Supplementary Fig. S2e-f) using HUMAN PROTEIN ATLAS, which is an online database that aims to map the entire human proteins in organs, tissues, and cells. In compliance with this, double immunofluorescence staining demonstrated clearer expression of PGK1 in proximal tubules (marked as aquaporin 1, AQP1), collective tubes (marked as aquaporin 3, AQP3), and distal tubules (marked as calbindin-D28k, CD28K), with higher expression in the proximal tubules (Supplementary Fig. S3a–c). Therefore, in vitro human HK-2 cells or primary RTECs were used to explore the role of PGK1 in AKI. Our results also disclosed that the mRNA level of PGK1 was induced in primary RTECs with respect to different stimuli, including hypoxia/reoxygenation (H/R), cisplatin, and LPS (Supplementary Fig. S4a), which were consistent with animal results.

RTECs-specific knockout of PGK1 or PGK1 inactivation protected against AKI

The function of PGK1 in AKI was determined through the construction of RTECs-specific deletion of PGK1 in mice to exhibit PGK1's importance in RTECs (Supplementary Fig. S5a, b). The mice with tamoxifen-inducible RTECs-specific knockout of PGK1 were established by crossing PGK1^flox/flox mice with Cdh16-Cre mice^22,23. We confirmed the deficiency of PGK1 at the levels of genomic DNA (Supplementary Fig. S5c, d, Table S1). Western blotting and RT-PCR showed that approximately 80% of PGK1 was deleted from the kidneys (Supplementary Fig. S5e, f). Both double immunofluorescence staining (Supplementary Fig. S3a) and immunohistochemistry (Supplementary Fig. S3d) showed lower expression of PGK1 in the proximal tubules of PGK1^{flox/flox/Cdh16Cre} (PGK1^CKO) mice when compared with PGK1^flox/flox mice.

I/R-induced AKI mice exhibited higher levels of BUN, Scr, and cystatin C, whereas RTCEs-specific knockout of PGK1 attenuated these changes (Supplementary Fig. S6a, Fig. 2a–c). HE staining showed that I/R-induced obvious tubular injury, such as tubular cell necrosis, renal tubule dilation, and protein cast formation in renal tubules, when compared with sham surgery mice (Fig. 2d, f). Strikingly, these histological lesions were attenuated in mice with renal depletion of PGK1 (Fig. 2d, f). Oxidative stress, renal cell apoptosis, and inflammation are driving forces for the development and progression of AKI^24,25,26. As pro-inflammatory cytokines, interleukin-1β (IL-1β), IL-6, and tumor necrosis factor α (TNF-α) are involved in the initiation and progression of inflammation and can directly induce cell apoptosis, which contributes to AKI²⁷. IL-10 plays a protective role in AKI by limiting the extent of renal inflammation and tissue damage²⁸. The 12-lipoxygenase (12-Lox) contributes to the production of bioactive lipid mediators that lead to oxidative stress, thus contributing to kidney cell damage²⁹. Ras-related C3 botulinum toxin substrate 1 (Rac-1) activation promotes the production of reactive oxygen species (ROS) and enhances inflammatory responses, contributing to tissue injury in AKI²⁵. NADPH oxidase 2 (Nox2) is a key enzyme involved in the generation of ROS, which are highly reactive molecules that play a critical role in cellular signaling and homeostasis³⁰. Meanwhile, excessive ROS production from Nox2 can lead to cellular damage and is implicated in various diseases, including AKI³¹. In AKI, excessive activation of Bcl-2-associated X protein (Bax) leads to mitochondrial dysfunction and cell death, leading to renal tubular cell injury³². B-cell lymphoma 2 (Bcl-2) has been established to act as a principal regulator of cell survival by preventing excessive cell apoptosis in AKI³³. Thus, we examined whether the improvement of AKI-induced renal injury by RTECs-specific knockout of PGK1 was related to the reduction of oxidative damage, apoptosis, and inflammation. Both DHE oxidation and immunofluorescence staining of Nox2 demonstrated that conditional knockout of PGK1 attenuated excessive production of ROS in the kidneys (Fig. 2e–g). TUNEL analysis showed that loss of PGK1 prevented I/R-induced renal cell apoptosis (Fig. 2f, h). Based on the immunofluorescence analysis, we showed that knockout of PGK1 reduced F4/80⁺ macrophage infiltration (Fig. 2f, i) and inhibited the protein expression of IL-1β (Fig. 2f, j) in AKI kidneys. In addition, PGK1 deletion reduced the mRNA changes in IL-1β, IL-6, TNF-α, 12-Lox, and Rac-1 genes in I/R-induced AKI mice (Fig. 2k, l, n–p), while rescuing IL-10 mRNA levels (Fig. 2m). The suppressive effects of PGK1 ablation on inflammation and ROS production were further confirmed by the measurement of inflammatory markers (IL-1β, IL-6, IL-10, TNF-α) and oxidative damage markers (8-iso-PGF2α and MDA) (Supplementary Fig. S6b–g). In addition, PGK1 deficiency in RTECs lowered serum and renal 3-PG levels in I/R-induced mice (Supplementary Fig. S6h, i). Western blotting analysis demonstrated that the PGK1^CKO mice had lower protein expression of NGAL and Bax, as well as higher protein expression of Bcl-2, than those in PGK1^flox/flox mice that underwent renal I/R injury (Fig. 2q-t). In keeping with this observation, PGK1-deficient mice had lower renal injury in response to cisplatin (Supplementary Figs. S7 and 8). Besides, similar protective effects were observed through pharmacological inhibition of PGK1 by its inhibitor CBR-4701^34,35,36 in I/R-induced wild-type (WT) mice (Supplementary Fig. S9), as evidenced by measurement of serological indicators (BUN, Scr, and cystatin C), pathological staining, DHE oxidation, TUNEL staining, F4/80 staining, and MDA content (Supplementary Fig. S9). These results collectively imply that PGK1 inactivation provided protection against I/R- and cisplatin-induced renal injury.

**Fig. 2: RTECs-specific knockout of PGK1 protected against I/R-induced AKI.**

RTECs-specific overexpression of PGK1 or 3-PG infusion worsened I/R- or cisplatin-induced AKI

To determine the potential of PGK1 overexpression in AKI mice, we overexpressed PGK1 in mice using an AAV9-packaged PGK1-overexpression plasmid under the control of a Ksp-cadherin promoter^37,38,39, thus allowing the RTECs-specific overexpression of PGK1. Western blotting results showed that the PGK1 protein level increased approximately 2-fold (Supplementary Fig. S10a). Both immunohistochemistry (Supplementary Fig. S10b) and double immunofluorescence staining (Supplementary Fig. S10c) showed higher expression of PGK1 in the proximal tubules of mice with RTECs-specific overexpression of PGK1. Contrary to what was observed in mice with RTECs-specific knockout of PGK1, we found that RTECs-specific overexpression (Supplementary Fig. S11a) further upregulated serum BUN, Scr, and cystatin C levels in I/R-induced AKI mice (Supplementary Fig. S11b-d). PGK1-overexpression mice exhibited more prominent histological injury in the kidneys, as shown by H&E staining (Supplementary Fig. S11e, g). Further, RTECs-specific overexpression of PGK1 further aggravated I/R-induced ROS production (Supplementary Fig. S11f–h), apoptosis (Supplementary Fig. S11g, i), and inflammation (Supplementary Fig. S11g, j, k). The mRNA levels of pro-inflammatory indices (IL-1β, IL-6, and TNF-α) and pro-oxidative indicators (12-Lox and Rac-1) were further enhanced by RTECs-specific overexpression of PGK1 (Supplementary Fig. S11l, m, o–q). Conversely, PGK1 overexpression overtly attenuated the mRNA level of an anti-inflammatory maker, IL-10 (Supplementary Fig. S11n). In compliance with the mRNA data, ELISA results showed that PGK1 overexpression further deteriorated inflammation and ROS production in AKI mice induced by I/R insult (Supplementary Fig. S12a–f). After overexpression of PGK1, the levels of 3-PG in the kidneys and serum almost approached those of AKI mice (Supplementary Fig. S12g, h). In addition, overexpression of PGK1 further facilitated the protein abundance of NGAL in mice after renal I/R injury (Supplementary Fig. S13a, b). In line with the TUNEL analysis, the protein expression of Bax was upregulated, whereas the protein expression of Bcl-2 was downregulated in I/R-induce mice with PGK1 overexpression (Supplementary Fig. S13a–d).

Compared with the sham-operated group, mice subjected to I/R exhibited renal dysfunction, as evidenced by elevated circulating levels of BUN, Scr, and cystatin C (Fig. 3a–c). Notably, treatment with 3-PG (Supplementary Fig. S14a) further exacerbated these elevations relative to the I/R group (Fig. 3a–c). Histopathological analysis revealed that 3-PG administration aggravated I/R-induced renal damage (Fig. 3d, f), ROS production (Fig. 3e–g), apoptosis (Fig. 3f, h), and inflammatory cell infiltration (Fig. 3f, i, j). Consistent with these findings, 3-PG treatment increased renal mRNA expression of pro-inflammatory cytokines and oxidative stress markers, including IL-1β, IL-6, TNF-α, 12-Lox, and Rac-1, while further suppressing the anti-inflammatory cytokine IL-10 (Fig. 3k–p). Western blot analysis showed that 3-PG enhanced the protein expression of NGAL and Bax, while reducing the expression of the anti-apoptotic protein Bcl-2 in I/R kidneys (Fig. 3q–t). In addition, 3-PG pretreatment upregulated the circulating levels of IL-1β, IL-6, and TNF-α but downregulated the circulating level of IL-10 in I/R mice (Supplementary Fig. S14b–e). The levels of ROS were further enhanced in renal tissues of I/R mice after treatment with 3-PG, as indicated by the levels of 8-iso-PGF2α and MDA (Supplementary Fig. S14f. g). Moreover, after intraperitoneal injection of 3-PG, the levels of 3-PG slightly increased in the kidneys and serum, which may be related to its metabolism in the body (Supplementary Fig. S14h, i). Interestingly, 3-PG infusion had no effect on PGK1 activity and PGK1 protein overexpression in I/R mice (Supplementary Fig. S14j, k). Similarly, these results were recapitulated in cisplatin-induced AKI mice treated with 3-PG (Supplementary Fig. S15). These results indicate that PGK1 and its metabolite 3-PG are important drivers for AKI pathogenesis.

**Fig. 3: 3-PG infusion worsened I/R-induced AKI.**

Effects of the PGK1/3-PG axis in RTEC injury induced by H/R or cisplatin in vitro

Given the abundant expression of PGK1 in RTECs, we subsequently investigated the potential role of the PGK1/3-PG axis in HK-2 cells' response to H/R or cisplatin. CCK-8 assays revealed that knockdown of PGK1 rescued the decline in cell viability induced by H/R (Supplementary Fig. S16a). This finding was corroborated by PI staining, which confirmed that PGK1 knockdown reduced the apoptosis of HK-2 cells (Supplementary Fig. S16b, c). Additionally, PGK1 ablation inhibited H/R-induced ROS overproduction, as demonstrated by DHE oxidation (Supplementary Fig. S16b, d) and the measurement of 8-iso-PGF2α (Supplementary Fig. S16e) and MDA content (Supplementary Fig. S17a). We also examined DHE oxidation with adequate controls, confirming that there was an increase in ROS production in HK-2 cells after incubation with hydrogen peroxide (H₂O₂) (Supplementary Fig. S17b). Besides, the results showed that N-acetylcysteine (NAC) alleviated H/R-induced ROS accumulation, whereas H₂O₂ exacerbated H/R-triggered ROS generation (Supplementary Fig. S17c). Furthermore, H/R stimulation increased the mRNA expression of oxidative stress markers, including 12-Lox and Rac-1, which was reduced by PGK1 knockdown (Supplementary Fig. S16f, g). Also, we assessed the levels of pro-inflammatory cytokines, such as IL-1β, IL-6, TNF-α, and the anti-inflammatory cytokine IL-10. Results indicated that PGK1 knockdown decreased the protein and mRNA levels of IL-1β, IL-6, and TNF-α, while increasing the protein and mRNA levels of IL-10 (Supplementary Fig. S16h–o). Moreover, immunoblotting confirmed the efficiency of PGK1 knockdown (Supplementary Fig. S16p). The protein expression of IL-1β and cleaved-caspase-3 was elevated in H/R-induced cells, and this upregulation was reduced by PGK1 knockdown (Supplementary Fig. S16p). These results were reproduced in HK-2 cells induced by cisplatin (Supplementary Fig. S18). Contrary to the result of PGK1 knockdown in primary RTECs, overexpression of PGK1 potentiated the effects of H/R on apoptosis, ROS production, and inflammatory response (Supplementary Fig. S19). Importantly, pretreatment with the PGK1-produced metabolite, 3-PG, aggravated cell apoptosis, ROS production, and inflammation in primary RTECs challenged by H/R (Supplementary Fig. S20). These data suggest that the activation of the PGK1/3-PG metabolic axis contributes to RTEC injury in response to AKI.

A shift from aerobic to anaerobic metabolism plays a crucial role in regulating ATP production and oxidative stress, particularly in the context of PGK1 inhibition. To address this, we examined the effects of PGK1 downregulation on mitochondrial oxygen consumption rate (OCR) and cellular glycolysis in HK-2 cells. Analysis of the mitochondrial OCR revealed a reduction in both basal and maximal OCR in H/R-triggered HK-2 cells (Supplementary Fig. S21a–c). This suppression was prevented following PGK siRNA transfection, which restored mitochondrial respiration (Supplementary Fig. S21b, c). Exposure of H/R induced a Warburg-like metabolic shift, as evidenced by elevated glycolytic ATP production, which was prevented by downregulation of PGK1 (Supplementary Fig. S21d). To further validate this, we analyzed glycolytic function (Supplementary Fig. S21e) and found that the basal glycolysis (Supplementary Fig. S21f), reserve (Supplementary Fig. S21g), and maximal glycolytic capacities (Supplementary Fig. S21h) were all increased in H/R-exposed cells. Moreover, the XF ATP rate index was decreased (Supplementary Fig. S21i), indicating an imbalance in energy production after H/R stimulation. H/R treatment also led to a higher proportion of ATP being generated via glycolysis (Supplementary Fig. S21j), with a corresponding reduction in ATP production from mitochondrial oxidative phosphorylation (Supplementary Fig. S21k). Nonetheless, these changes were eliminated by PGK1 downregulation (Supplementary Fig. S21e–k). These results indicated that PGK1 impairs oxidative phosphorylation and increases glycolytic metabolism, and this could potentially reduce overall ATP availability in HK-2 cells. Since PGK1 is a key glycolytic enzyme, alterations in its expression could potentially affect broader cellular metabolic processes. We thus performed an exhaustive analysis of all possible metabolic disruptions in primary RTCEs due to PGK1 manipulation using untargeted metabolomics. Preliminary analysis indicates that 46 metabolites were upregulated, while 6 metabolites were obviously downregulated in RTECs after knocking down PGK1 (Supplementary Fig. S22a, b). These upregulated metabolites mainly include cis-7-Hexadecenoic acid, Palmitoleate, 8-POHPA, 9-PAHSA, FA 17:1, and 8-Methylheptadecanoic acid, to name a few (Supplementary Fig. S22c). The downregulated metabolites mainly included Isodeoxycholic acid, N1-Acetylspermidine, and C17-Sphinganine (Supplementary Fig. S22c). KEGG enrichment was performed on the signaling pathways that were affected by differential metabolites, and it was found that PGK1 knockdown mainly affected the metabolic pathways and amid biosynthesis of secondary metabolites (Supplementary Fig. S23). Conversely, PGK1 overexpression changed a host of metabolites in RTECs, including 44 upregulated metabolites and 6 downregulated metabolites (Supplementary Fig. S24a, b). These upregulated metabolites mainly included Methylthioadenosine, Uridine 5′-Diphospho-N-Acetylgalactosamine, Uridine 5′-Diphosphoglucuronic Acid, and Lumichrome, to name a few (Supplementary Fig. S24c). The downregulated metabolites mainly included 4-Methylnonanoic acid and PC 30:0 (Supplementary Fig. S24c). Importantly, PGK1 overexpression primarily affected the metabolic pathways, amino sugar and nucleotide sugar metabolism, as well as biosynthesis of nucleotide sugars (Supplementary Fig. S25). These results indicate that PGK1 acts as a central metabolic regulator in RTECs, with knockdown favoring fatty acid accumulation and lipid metabolism shifts, while overexpression promotes nucleotide biosynthesis and cell proliferation-associated pathways. Future research should focus on the mechanistic underpinnings of PGK1’s metabolic regulation, its interaction with key signaling pathways, and its potential as a therapeutic target. A deeper understanding of PGK1’s role in renal metabolism could pave the way for innovative treatments aimed at restoring metabolic homeostasis in kidney diseases.

RNA sequencing revealing that PGK1 activation leads to ferroptosis in AKI by upregulating ALOX12

We next employed RNA sequencing to elucidate the molecular mechanisms by which PGK1 contributes to AKI. Compared with the kidneys that were subjected to I/R injury, RTECs-specific PGK1 knockout resulted in alterations in the expression of numerous genes, with 54 genes being upregulated and 103 genes being downregulated (Fig. 4a–c). From our RNA-sequencing data, we observed the 10 most downregulated genes in PGK1^CKO mice, including 12-lipoxygenase (ALOX12), ENTPD1, TMC4, KRTCAP3, FAM20A, MIR497HG, CCDC81, NEK5, SLC26A7, and SPRN. RT-PCR analysis revealed that the mRNA levels of ALOX12, ENTPD1, and MIR497HG were upregulated in H/R-exposed RTECs, which were prevented when PGK1 was silenced (Supplementary Fig. S26). Among these three genes, the mRNA level of ALOX12 tended to be much higher in RTECs upon exposure to H/R (Supplementary Fig. S26). KEGG pathway analysis revealed that PGK1 knockout influenced several signaling pathways in the kidneys, including the iron ion binding pathway (Fig. 4d), indicating that PGK1 might play a necessary role in the development of AKI by regulating the ferroptotic response. Among the altered genes, ALOX12 attracted our particular attention since ALOX12 has been shown to be involved in the propagation of ferroptosis through the direct oxidation of membrane polyunsaturated fatty acids (PUFAs), thereby contributing to the buildup of toxic lipid peroxides^40,41. As expected, the protein expression of ALOX12 was markedly upregulated in the kidneys from mice that were subjected to I/R (Fig. 4e, g), and in H/R-stimulated HK-2 cells (Fig. 4f, h). Prussian and FerroOrange staining showed a higher abundance of Fe²⁺ in I/R-induced kidneys, but this was counteracted by RTECs-specific knockout of PGK1 (Fig. 4i, j). The absence of PGK1 prevented excessive lipid peroxide in I/R-induced kidneys (Fig. 4i, k, and Supplementary Fig. S27a-b). The content of Fe²⁺ was higher in the kidneys from mice subjected to I/R, which was attenuated by RTECs-specific knockout of PGK1 (Supplementary Fig. S27c). The mRNA level of PTGS2, a marker of ferroptosis, was remarkably upregulated in kidneys from I/R mice, whereas PGK1 knockout abolished this upregulation (Fig. 4l). Importantly, RTECs-specific knockout of PGK1 impeded the upregulation of ALOX12 protein in I/R-induced kidneys (Fig. 4m, n). It has been established that p53 can induce ferroptosis by regulating ALOX12 to promote iron metabolism dysfunction and lipid peroxidation, revealing that ALOX12-mediated ferroptosis is critical for p53-dependent tumor suppression⁴². The p53/ALOX12 pathway is required for CX-5461 to trigger ferroptosis in B cells, thus attenuating lupus⁴³. Therefore, we examined whether PGK1 contributed to ferroptosis in AKI by regulating the p53/ALOX12 signaling pathway. Results showed that RTECs-specific knockout of PGK1 prevented the upregulated p53 mRNA level in I/R-induced kidneys (Supplementary Fig. S27d). Intriguingly, the mRNA levels of ACSL4 and GPX4, GPX4 activity, and GSH activity in I/R-induced kidneys were not affected by RTECs-specific knockout of PGK1 (Supplementary Fig. S27e–h). Furthermore, downregulation of PGK1 suppressed H/R-triggered Fe²⁺ accumulation, lipid peroxide, ferroptosis, and p53 and ALOX12 upregulations in H/R-induced HK-2 cells (Supplementary Fig. S28a–h). However, silencing PGK1 had no effect on ACSL4 and GPX4 mRNA levels, GPX4 activity, and GSH activity in HK-2 cells (Supplementary Fig. S28i–l). These findings indicate that PGK1 accelerates ferroptosis by activating ALOX12-mediated ferroptosis rather than ACSL4-induced ferroptosis in the context of AKI.

**Fig. 4: RNA sequencing reveals that PGK1 activation leads to ferroptosis in AKI by upregulating ALOX12.**

Importantly, HK-2 cells were pre-treated with an iron chelator, Deferoxamine (DFO), or an iron overload inducer, Ferric ammonium citrate (FAC), and then stained with FerroOrange and Liperfluo probes (Supplementary Fig. S29a, b). These control experiments confirmed that the observed Fe²⁺ accumulation was consistent with ferroptosis induction in this study (Supplementary Fig. S29a, b). HK-2 cells were treated with FAC to induce iron overload, while FAC pretreatment blocked the effects of PGK1 siRNA on H/R-induced cell apoptosis and inflammation (Supplementary Fig. S30a–e), hinting that the anti-inflammatory effects of PGK1 inhibition may involve a ferroptosis-dependent mechanism. In accordance with this, downregulation of PGK1 blocked ALOX12-mediated ferroptosis in cisplatin-induced kidneys (Supplementary Fig. S31) or primary RTECs (Supplementary Fig. S32).

In addition, we have performed additional experiments using ferroptosis inhibitors Liproxstatin-1 (in vivo) and Ferrostatin-1 (in vitro) in the context of PGK1 overexpression^44,45,46. Our results showed that both inhibitors attenuated the effects of PGK1 overexpression on renal injury in response to H/R (Supplementary Fig. S33) or I/R (Supplementary Fig. S34), further confirming that PGK1 promotes acute kidney injury, at least in part, through the induction of ferroptosis. Similar results were also observed in I/R-induced mice treated with the ALOX12 inhibitor ML355^29,47 (Supplementary Fig. S35) or HK-2 cells with a deficiency of ALOX12 (Supplementary Fig. S36). Contrary to what we observed in AKI mice with PGK1 deficiency, 3-PG administration aggravated I/R-elicited ferroptosis and lipid peroxidation in kidneys (Supplementary Fig. S37a–e).

The nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway plays a critical role in AKI pathophysiology, particularly in inflammation, oxidative stress, and cell death⁴⁸. The mitogen-activated protein kinase (MAPK) signaling is another critical regulator of AKI that induces mitochondrial dysfunction, ROS production, and cell death, consisting of three major kinase cascades, ERK (extracellular signal-regulated kinase), JNK (c-Jun N-terminal kinase), and p38⁴⁹. Therefore, it is worth studying whether 3-PG contributes to AKI by regulating the NF-κB and MAPK signaling pathways. 3-PG treatment increased p65 phosphorylation in I/R-exposed kidneys, suggesting that 3-PG induced activation of the NF-κB pathway in AKI settings (Supplementary Fig. S37f, j). However, 3-PG treatment had no effect on the phosphorylation of JNK, p38, and ERK in kidneys under I/R conditions (Supplementary Fig. S37f, h–j). Overall, these results imply that PGK1 is critically involved in AKI pathogenesis by upregulating the p53/ALOX12 axis-regulated ferroptosis, a special signaling pathway distinct from the classical ferroptosis pathway.

Co-IP MS disclosing that the interaction of PGK1 with PKM2 induces dimerization and nuclear translocation of PKM2

It is interesting to know how PGK1 affects the transcription of ALOX12 in RTECs. We thus screened proteins that may interact with PGK1 using a Co-IP MS approach. From the identified thousands of proteins, we can see that the interaction between pyruvate kinase M2 (PKM2) and PGK1 had the best binding score (Table S2, Fig. 5a). In support, the direct interaction of PGK1 with PKM2 was ascertained by confocal staining (Fig. 5b), co-IP assay (Fig. 5c, d), and GST co-IP assay (Fig. 5e, f). According to the prediction results from molecular docking, PGK1 and PKM2 may form a protein interaction interface with a binding score of −212.04 (Supplementary Fig. S38a, b). To explore the potential fragments participating in this interaction, we mapped the possible binding sites of PGK1 and PKM2 by producing a variety of truncated mutants. Our results indicated that the 221–376 fragment of PGK1 and the 329–531 fragment of PKM2 might be responsible for the formation of the PGK1/PKM2 complex (Supplementary Fig. S38c, d). These results provide evidence for the direct binding of PGK1 and PKM2 in RTECs. Similar to PGK1, PKM2 expression at the protein level was higher in I/R-induced kidneys and H/R-exposed RTECs (Supplementary Fig. S39a–d). Also, the interaction of PGK1 with PKM2 was obviously enhanced in I/R-induced kidneys and H/R-exposed RTECs (Supplementary Fig. S39e–h). Similar results were detected in cisplatin-induced kidneys or HK-2 cells (Supplementary Fig. S39i–p).

**Fig. 5: Interaction of PGK1 with PKM2 induces dimerization and nuclear translocation of PKM2.**

PKM2 is known for its ability to exist in multiple oligomeric states, including dimeric and tetrameric forms⁵⁰. Phosphorylation of PKM2 is a key post-translational modification that influences its dimerization and subsequent functional roles in the cell⁵¹. Specific phosphorylation events, such as those occurring at tyrosine residue 105 (Y105), have been shown to promote the dimeric state of PKM2⁵². The dimeric form of phosphorylated PKM2 is associated with non-metabolic functions, particularly in the nucleus, where it acts as a coactivator for transcription factors such as hypoxia-inducible factor 1α (HIF-1α) and β-catenin^53,54. This nuclear role of PKM2 is linked to the regulation of gene expression involved in cell cycle progression, survival, and metabolic reprogramming⁵⁵. Notably, emerging evidence demonstrates that PKM2 participates in AKI progression through modulating metabolism regulation, macrophage polarization, podocyte injury, and fibroblast activation^56,57. Given the function of PGK1 as a protein kinase^{58,59,60,61,62,63,64} and the phosphorylation modification of PKM2 in cellular signaling transduction, we therefore proposed that PGK1 might function as a kinase to regulate PKM2 via phosphorylation modification. As anticipated, the total protein, phosphorylation, dimeric forms, and nuclear abundance of PKM2 were elevated in I/R-induced kidneys, which were dramatically reduced by RTECs-specific knockout of PGK1 (Fig. 5g–l). Immunofluorescence results showed that H/R promoted PKM2 nuclear translocation (Fig. 6a). In addition, silencing PGK1 prevented H/R-elicited PKM2 phosphorylation, dimeric formation, and nuclear translocation in HK-2 cells (Supplementary Fig. S40a–f). More importantly, these results were replicated in cisplatin-exposed kidneys (Supplementary Fig. S41a–f) or HK-2 cells (Supplementary Fig. S41g–l). These findings demonstrate that PGK1 interacts with PKM2, thus resulting in the phosphorylation, dimeric formation, and nuclear translocation of PKM2 during the development of AKI.

Accumulated PKM2 in the nucleus interacted with Pknox1, increasing the DNA binding of Pknox1 to ALOX12 promoters

We next used four databases, such as GTRD, TFDB, PROMO, and JASPER, to predict potential transcription factors that may be involved in regulating ALOX12 expression (Supplementary Fig. S42a). A total of 14 transcription factors were observed in all four databases using a Vein chart (Supplementary Fig. S42a). After transfection of these 14 transcription factor overexpression plasmids, Pknox1 was found to increase the RLU of PGK1 by approximately 4-fold (Supplementary Fig. S42b). Besides, the DNA binding of Pknox1 to ALOX12 promoters was obviously higher in I/R-induced kidneys and H/R-exposed RTECs (Supplementary Fig. S42c). The protein expression of Pknox1 was downregulated in I/R-induced kidneys from PGK1^CKO mice (Supplementary Fig. S42d). By contrast, the protein expression of Pknox1 in I/R-induced kidneys was further upregulated after RTECs-specific PGK1 overexpression (Supplementary Fig. S42e). PGK1 deficiency reduced the binding of Pknox1 to ALOX12 promoters in I/R mice and H/R-exposed HK-2 cells (Fig. 6b, c). Pknox1 and PKM2 siRNA3 were selected for our cellular experiments since Pknox1 and PKM2 siRNA3 displayed the strongest capability to inhibit the mRNA and protein levels of Pknox1 in HK-2 cells (Supplementary Fig. S43a–d). Of note, knockdown of Pknox1 restored the cell viability (Fig. 6d), reduced cell apoptosis (Fig. 6e, f), and lowered the Fe²⁺ accumulation and lipid peroxide in H/R-induced HK-2 cells, as evidenced by FerroOrange and Liperfluo accumulation (Fig. 6e, g, h). Additionally, knockdown of Pknox1 (Supplementary Fig. S44a) weakened the effects of H/R on the expression of ALOX12 and PTGS2 (Fig. 6i–k, Supplementary Fig. S44b). Although p53 knockdown (Supplementary Fig. S44c) reduced H/R-induced ALOX12 expression, the effect was less pronounced than that observed with Pknox1 knockdown (Supplementary Fig. S44d). As such, Pknox1 might mainly mediate the upregulation of ALOX12 and the following ferroptosis in AKI.

Here too, it is deserved to investigate how Pknox1 contributes to the upregulation of ALOX12 by linking PGK1-mediated PKM2 nuclear translocation. We speculated that accumulated PKM2 in the nucleus may interact with Pknox1 and promote the enrichment of Pknox1 within ALOX12 promoters, leading to transcription activation of ALOX12 in the context of AKI. The interaction of PKM2 with Pknox1 was obviously enhanced in I/R-induced kidneys (Supplementary Fig. S45a) or H/R-exposed HK-2 cells (Supplementary Fig. S45b). Knockdown of PGK1 and PKM2 inhibited the RLU of ALOX12 (Fig. 6l) and reduced the binding of Pknox1 to the ALOX12 promoters (Fig. 6m), followed by a decrease in the transcription and translation levels of Pknox1 in H/R-stimulated HK-2 cells (Fig. 6n–p). More importantly, downregulation of PKM2 also hindered H/R-triggered cell apoptosis, lipid peroxidation, and ferroptosis in primary RTECs, along with lowered protein expression of ALOX12 (Supplementary Fig. S46). Therefore, it was concluded that PGK1 induced PKM2 phosphorylation and nuclear translocation, whereby the nuclear PKM2 interacted with Pknox1, leading to the enrichment of Pknox1 within the ALOX12 promoters, and inducing subsequent ALOX12-mediated ferroptosis in AKI.

PGK1 is negatively regulated by miR-3185 in AKI

PGK1 is upregulated in the context of AKI, a finding that underscores its potential importance in the pathophysiology of AKI. MicroRNAs (miRNAs) are a class of endogenous, non-coding RNAs found in eukaryotic organisms, typically 20-25 nucleotides in length. Once processed, these miRNAs are incorporated into the RNA-induced silencing complex (RISC), where they recognize target mRNAs via base-pairing. Depending on the degree of complementarity, miRNAs guide the RNA-induced silencing complex (RISC) to either degrade the target mRNA or inhibit its translation. Using four predictive databases, including TargetScan, miRDB, miRwalk, and miRTarBase, we identified numerous miRNAs potentially involved in regulating PGK1 expression. Venn diagram analysis revealed that only miR-3185 was commonly predicted across all four databases (Supplementary Fig. S47a). This suggests that miR-3185 may play a pivotal role in the regulation of PGK1 expression. Notably, compared to healthy individuals, miR-3185 levels were reduced in the peripheral blood of AKI patients (Supplementary Fig. S47b). Similar reductions in miR-3185 expression were observed in kidneys subjected to I/R injury (Supplementary Fig. S47c), and in primary RTECs subjected to H/R (Supplementary Fig. S47d), as well as in kidneys and RTECs treated with cisplatin (Supplementary Fig. S47e, f). Knockdown of miR-3185 elevated PGK1 protein expression (Supplementary Fig. S47g, h). Conversely, overexpression of miR-3185 restored cell viability (Supplementary Fig. S47i), while reducing apoptosis and ROS production in primary RTECs exposed to H/R (Supplementary Fig. S47j–l), along with lower protein expression of PGK1 (Supplementary Fig. S47m, n). Dual-luciferase reporter assay revealed that miR-3185 directly downregulated the transcriptional activity of PGK1 in RTECs (Supplementary Fig. S47o, p). These findings highlight the critical regulatory role of miR-3185 in modulating PGK1 expression and its potential therapeutic implications in mitigating kidney damage during AKI.

Luteolin-7-diglucuronide (L7DG) acts as an inhibitor of PGK1 that ameliorates AKI in vitro and in vivo

In light of the therapeutic potential of inhibiting PGK1 during the treatment of AKI, screening for effective PGK1 inhibitors holds clinical value. Thus, we conducted a virtual screening that targets the ADP-binding site of PGK1 to identify small-molecule compounds with strong binding affinity for the protein. The virtual screening was performed using Schrödinger Maestro 11.4 software, which was applied to compounds from four libraries provided by MedChemExpress (MCE) (Table S3). These libraries included HY-L001P (MCE Bioactive Compound Library Plus, containing 20,100 compounds), HY-L009 (Kinase Inhibitor Library, containing 2,800 compounds), HY-L007 (Immunology/Inflammation Compound Library, containing 4,400 compounds), and HY-L901 (50K Diversity Library, containing 50,000 compounds). Through this computational approach, we aimed to identify small molecules with the potential to inhibit PGK1 effectively, thus offering avenues for the treatment of AKI.

From the top 13 compounds in terms of binding energy, we further found that L7DG had the strongest ability to inhibit the RLU of PGK1 (Table S4). Molecular docking showed L7DG with PGK1 had the strongest direct binding, with −13.024 kcal/mol as the binding energy (Supplementary Fig. S48a, b, Table S4). Furthermore, the cellular thermal shift assay (CETSA) and drug affinity responsive target stability (DARTS) assay showed that L7DG could directly bind to PGK1 (Supplementary Fig. S48c–f). Similar to PGK1 siRNA, L7DG pretreatment prevented apoptosis (Supplementary Fig. S49a–c), ROS production (Supplementary Fig. S49b, d–g), and inflammation (Supplementary Fig. S49h–o) in HK-2 cells in response to H/R. The higher Pknox1 mRNA level in H/R-exposed HK-2 cells was suppressed by L7DG pretreatment (Supplementary Fig. S49p). In conformity with this, the protein expression of PKM2, P-PKM2, PGK1, IL-1β, and cleaved-caspase 3 was lifted in H/R-exposed HK-2 cells, while pretreatment with L7DG curbed these upregulations (Supplementary Fig. S49q–v). Administration of L7DG prevented H/R-induced upregulations of Fe²⁺ and MDA in HK-2 cells (Supplementary Fig. S49w, x). Next, we examined whether L7DG can attenuate AKI by inhibiting PGK1 (Supplementary Fig. S50a). In support of cellular results, administration of L7DG restored renal function (Fig. 7A–C), improved renal pathological structure (Fig. 7d, e), reduced renal ROS overproduction (Figs. 7d, 7f, g, and Supplementary Fig. S50b, c), inhibited renal cell apoptosis (Fig. 7D, H) and inflammation (Figs. 7d, 7i–p, and Supplementary Fig. S50d–g), lessened renal ferroptosis (Supplementary Fig. S50h) and 3-PG contents (Supplementary Fig. S50i) in AKI mice through suppression of the mRNA level of Pknox1 (Fig. 7q) and downregulation of the protein expression of PKM2, P-PKM2, PGK1, and NGAL (Fig. 7r–v). Overall, L7DG might block PGK1 to exert therapeutic effects on AKI.

**Fig. 7: L7DG acts as an inhibitor of PGK1 that ameliorates AKI.**

Discussion

PGK1 is an enzyme that catalyzes the reversible conversion of 1,3-bisphosphoglycerate to 3-PG during glycolysis. PGK1 and 3-PG are critical components of the glycolytic pathway, playing roles in cellular energy production and metabolic regulation. 3-PG is a key metabolite in the glycolytic pathway, and its levels can influence various cellular processes, including energy production, biosynthesis, and redox homeostasis. However, there is no evidence for the engagement of the PGK1/3-PG metabolic axis in AKI. In the present study, the upregulation of PGK1 and the increased levels of 3-PG have been observed in various models of AKI, including I/R- and cisplatin-induced kidney injury. This upregulation is associated with increased ROS production, inflammation, and apoptosis in RTECs, thereby contributing to the progression of kidney damage. These findings suggest that PGK1 and 3-PG may act as mediators of ROS production, inflammation, and apoptosis, which are hallmarks of AKI. Therefore, our results indicated that PGK1 and its metabolite 3-PG are key drivers in AKI pathogenesis (Supplementary Fig. S51).

In this study, we demonstrated that both PGK1 and 3-PG were critically implicated in the pathogenesis of AKI. In addition, 3-PG accumulation might serve as a regulatory signal, possibly influencing various pathways that exacerbate kidney damage. However, the exact molecular mechanisms by which PGK1 and 3-PG contribute to ROS production, inflammation, and apoptosis in RTECs remain unclear. Furthermore, longitudinal studies in AKI patients could help determine whether PGK1 expression or 3-PG levels are associated with disease progression or patient outcomes. Such studies could validate the PGK1/3-PG axis as a biomarker for AKI severity or prognosis. Interestingly, our results showed that PGK1 overexpression induced an increase in tubular injury even without an AKI model, compared to the pharmacological administration of 3-PG. Under baseline conditions, PGK1 overexpression could disrupt normal cellular metabolism, leading to altered energy homeostasis and oxidative stress, which are known contributors to kidney injury. Overexpression of PGK1 could shift the metabolic flux, potentially causing an imbalance between energy production and consumption in renal tubular cells. This might lead to metabolic stress, mitochondrial dysfunction, and increased oxidative stress, all of which can contribute to kidney injury. 3-PG is a highly regulated intermediate. When injected exogenously, it is likely to be metabolized and cleared by the body through various pathways, limiting its potential for accumulating to levels that could cause significant damage. This could explain why 3-PG infusion did not have a pronounced effect on kidney injury. In contrast, overexpression of PGK1 results in a continuous buildup of 3-PG in the kidney, which may lead to metabolic overload and injury over time. Overexpression of PGK1 directly impacts the activity of the enzyme, leading to a feedback loop of glycolytic dysregulation and increased oxidative stress. In contrast, 3-PG injection does not influence PGK1 activity and is simply one of the metabolites produced during glycolysis. Thus, the pathological effects of PGK1 overexpression are more likely due to the persistent activation of PGK1 and its downstream metabolic consequences, rather than just the accumulation of 3-PG itself. This underscores the complex and context-dependent role of PGK1 and its metabolites in kidney injury and suggests that chronic metabolic dysregulation, rather than the mere presence of specific metabolites, is a key driver of renal damage. Further studies are needed to dissect the precise molecular mechanisms through which PGK1 exacerbates kidney injury and to explore potential therapeutic strategies for mitigating its effects.

Regarding the positive relationship between PGK1/3-PG levels and AKI severity, we agree that it is crucial to investigate whether there is a dose-response relationship between PGK1/3-PG expression and the severity of AKI. While our current data suggest a correlation between elevated PGK1/3-PG levels and increased injury markers, the exact dose-response relationship has not yet been fully explored in our present study. Future studies are recommended to perform dose-response experiments by modulating PGK1 expression and administering exogenous 3-PG at varying concentrations in RTECs and animal models of AKI. In this study, we also found that PGK1 impaired oxidative phosphorylation and increased glycolytic metabolism, and this could potentially reduce overall ATP availability in HK-2 cells. This could drive cells to upregulate anaerobic glycolysis to compensate for energy loss. A shift from aerobic to anaerobic metabolism impacts ATP production, mitochondrial ROS levels, and oxidative stress. In the context of PGK1 inhibition and nuclear PKM2 translocation, this metabolic shift may serve as a compensatory mechanism to maintain energy homeostasis while mitigating oxidative damage. Meanwhile, nuclear PKM2 translocation is known to influence metabolic reprogramming by promoting glycolytic flux, enhancing lactate production, and reducing mitochondrial oxidative phosphorylation. Further investigation is needed to delineate how these metabolic alterations induced by PGK1 influence cellular function and survival under different physiological and pathological conditions, especially in AKI.

As an important metabolic enzyme, we found that PGK1 plays a crucial role in metabolic regulation in RTECs using metabolomics analysis. The changes in metabolite levels upon PGK1 knockdown and overexpression suggest that PGK1 influences multiple metabolic pathways, particularly lipid metabolism, nucleotide metabolism, and secondary metabolite biosynthesis. This supports the idea that PGK1 plays a broad role in regulating energy homeostasis and lipid metabolism, and its knockdown may trigger metabolic compensatory mechanisms. PGK1 knockdown leads to lipid accumulation and a shift toward fatty acid metabolism, possibly due to decreased glycolytic activity. PGK1 overexpression enhances nucleotide metabolism and biosynthetic pathways, suggesting that PGK1 promotes anabolic metabolism. Thus, the specific balance of PGK1 should be considered for targeted therapeutic interventions in renal diseases. Further studies should explore how PGK1 modulation affects kidney disease progression and potential therapeutic targets.

Alox12 is an enzyme that plays a role in lipid metabolism, specifically in the oxidation of PUFAs⁶⁵. Its role in ferroptosis, a regulated form of cell death characterized by iron-dependent lipid peroxidation, has garnered increasing attention. Alox12 has been shown to be involved in the propagation of ferroptosis through the direct oxidation of membrane PUFAs, contributing to the buildup of toxic lipid peroxides^66,67,68. Our RNA sequencing identified changes in gene expression following RTECs-specific PGK1 knockout in I/R-induced kidneys, with notable downregulation of Alox12, a gene involved in ferroptosis. PGK1 knockout reduced ferroptosis markers such as lipid peroxides and Fe²⁺ accumulation, while also decreasing Alox12 expression, both in I/R- and cisplatin-induced models of AKI. The results suggest that targeting the PGK1/Alox12 axis could be a therapeutic strategy for preventing or mitigating ferroptosis in AKI. Inhibiting PGK1 or Alox12 might reduce lipid peroxidation and protect RTECs from ferroptotic cell death, thereby preserving kidney function in AKI patients. Notwithstanding, further research is needed to delineate the precise molecular mechanisms by which PGK1 regulates Alox12 expression and activity. This could involve investigating the signaling pathways or transcription factors that are activated by PGK1 and that control Alox12 expression. It would also be important to explore whether PGK1 influences other enzymes or proteins involved in ferroptosis, such as GPX4, SLC7A11, or ACSL4, to gain a more comprehensive understanding of how PGK1 orchestrates ferroptosis in RTECs. Further research could focus on developing non-invasive biomarkers that reflect changes in the PGK1/Alox12 axis, such as urinary or plasma levels of lipid peroxides or iron-related metabolites.

Our present results suggest that the PGK1/3-PG axis could have non-metabolic roles, such as influencing signaling pathways related to cell survival and death. This raises the question of whether PGK1 might interact with other proteins or pathways in RTECs to mediate its effects on kidney injury. Co-IP MS results revealed that PGK1 interacted with PKM2, an enzyme that plays a critical role in glycolysis and is known for its ability to exist in multiple oligomeric states, including dimeric and tetrameric forms. Notably, PGK1 can function as a protein kinase, contributing to various signaling pathways and regulating cellular functions such as apoptosis, inflammation, and ROS overproduction. For example, PGK1-mediated Beclin1 S30 phosphorylation is involved in glutamine deprivation- and hypoxia-induced autophagy and brain tumorigenesis⁶⁴. Our results disclosed that PGK1 was shown to phosphorylate PKM2, promoting its dimeric form, which translocated to the nucleus and contributed to the transcriptional regulation of genes involved in AKI progression. Our further investigation revealed that nuclear PKM2 interacted with Pknox1, a transcription factor that is potentially bound to the Alox12 promoters. The interaction of PKM2 with Pknox1 was enhanced in I/R-induced kidneys and H/R-treated RTECs. PGK1 or PKM2 knockdown reduced Pknox1 binding to the Alox12 promoter, which decreased Alox12 expression and subsequent ferroptosis. These findings highlight that the interaction of PGK1 with PKM2 leads to the activation of signaling pathways that promote the transcription of genes involved in ferroptosis, such as Alox12, thereby exacerbating kidney injury during AKI. In this study, we discovered that 3-PG infusion and PGK1 overexpression exhibited similar effects in AKI. This suggests the possibility that 3-PG itself may contribute to ferroptosis, potentially through an indirect mechanism or via crosstalk with lipid peroxidation pathways. We found that 3-PG treatment increased ferroptosis markers, indicating that 3-PG promotes ferroptosis in AKI, akin to PGK1 overexpression. These results suggested the dual roles of PGK1 in ferroptosis through both metabolic and non-metabolic pathways. Moreover, we found that 3-PG treatment increased p65 phosphorylation in I/R-exposed kidneys, suggesting that 3-PG induced activation of the NF-κB pathway in AKI settings. The results suggest that 3-PG itself contributes to ferroptosis, rather than being a mere byproduct of PGK1 activity. This raises important mechanistic questions regarding whether 3-PG serves as a metabolic signal that exacerbates ROS production and lipid peroxidation, thereby promoting ferroptosis independently of PGK1’s direct action on ALOX12. This deserves further study. NF-κB is a central regulator of the inflammatory response, while its activation often correlates with tissue injury in AKI. The phosphorylation of p65 could promote the expression of pro-inflammatory cytokines, which may exacerbate ROS production and thus contribute to ferroptosis. NF-κB activation could also trigger pathways that indirectly affect lipid peroxidation, further exacerbating ferroptosis in the kidney during AKI. Thus, the observation that 3-PG treatment increased p65 phosphorylation in I/R-exposed kidneys supports the hypothesis that PGK1 may be linking ferroptosis with inflammation via 3-PG-induced NF-κB signaling.

Our findings indicate that miR-3185 negatively regulates PGK1 expression, with reduced levels of miR-3185 correlating with increased PGK1 expression in AKI models. The study suggests that restoring miR-3185 levels could mitigate kidney damage by downregulating PGK1, thereby reducing apoptosis and ROS production in RTECs. The dual-luciferase reporter assay provides direct evidence of this regulatory interaction, confirming that miR-3185 binds to the PGK1 mRNA and reduces its transcriptional activity. The observed reduction in miR-3185 levels in AKI models, including in both peripheral blood and kidney tissues, suggests that miR-3185 downregulation is a consistent feature of AKI. This study contributes to the growing body of evidence that miRNAs play crucial roles in the regulation of key genes that are involved in AKI pathophysiology. The identification of miR-3185 as a critical regulator of PGK1 highlights the importance of miRNA-mediated gene regulation in kidney injury and suggests that other miRNAs may similarly modulate additional genes involved in AKI. Consistent downregulation of miR-3185 in peripheral blood and kidney tissues of AKI patients suggests that miR-3185 could serve as a biomarker for AKI diagnosis or prognosis. Further research should explore the feasibility of using miR-3185 levels in blood or urine as a non-invasive biomarker for early detection of AKI or for monitoring disease progression.

Ultimately, our study utilized virtual screening to identify compounds with high binding affinity for the ADP-binding site of PGK1, ultimately selecting L7DG on the basis of its strong binding energy and ability to inhibit PGK1 activity. This compound was further validated in vitro and in vivo, demonstrating its capacity to mitigate key pathological features of AKI, including apoptosis, ROS production, and inflammation in RTECs and AKI mouse models. By inhibiting PGK1, L7DG was able to reduce apoptosis, ROS production, and inflammation in RTECs exposed to H/R, which are key mechanisms driving kidney injury in AKI. The compound’s ability to improve renal function and pathological structure in AKI mouse models further supports its potential for clinical application. Further research should delve into the precise molecular mechanisms by which L7DG inhibits PGK1 and how this inhibition translates into the observed protective effects in AKI. It would also be valuable to explore whether L7DG has any off-target effects or interacts with other enzymes or pathways in the kidney, which could influence its overall efficacy and safety profile. Overall, these results demonstrated that the metabolic axis involving PGK1 and 3-PG is therefore critical in mediating the cellular responses to AKI. The mechanistic link between PGK1 and PKM2, which leads to PKM2 phosphorylation, dimer formation, nuclear translocation, and subsequent activation of Alox12-mediated ferroptosis, provides valuable insight into how PGK1 contributes to cell death and tissue damage in AKI. This pathway may represent a potential target for therapeutic intervention of AKI.

Our results showed that PGK1 expression was substantially downregulated in mice with RTEC-specific knockout of PGK1 under the control of the mouse cadherin 16 (Cdh16 or Ksp-cadherin) promoter. It is noted that this recombination efficiency may not be uniform across all proximal tubule cells. Cre recombinase expression follows the expression of the endogenous gene and is detected in the epithelial cells of developing nephrons, ureteric bud, mesonephric tubules, Wolffian duct, and Müllerian duct. In the adult mouse, expression is limited to the renal tubules, especially the collecting ducts, loops of Henle, and distal tubules. The expression of Cre recombinase in proximal tubule cells is not as robust under the ksp Cre promoter. The residual PGK1 expression in some proximal tubule cells could partially account for any observed phenotypic variability. Future experiments should be conducted to address this limitation, such as the usage of alternative Cre lines with more specific or robust expression in proximal tubule cells (e.g., PEPCK-Cre or SGLT2-Cre models). Untargeted metabolomics results indicated that PGK1 might affect cellular metabolism in RTECs, and it remains to be answered whether any unintended disruptions occur that could influence the interpretation of PGK1 as a therapeutic target for AKI. It should be acknowledged that these unintended metabolic disruptions might limit the therapeutic applicability of PGK1. To balance this, we hope to propose that the pharmacological inhibition of PGK1, which could offer reversible modulation of its activity, may provide a more targeted therapeutic approach in our future studies.

Methods

Human renal biopsies and blood samples

This study aimed to investigate the role of PGK1 in the pathogenesis of AKI and to elucidate the underlying mechanisms. To achieve these objectives, we conducted a comprehensive analysis using human kidney tissues, multiple mouse models of AKI, and in vitro models of renal tubular cell injury. All human procedures were reviewed and approved by the Medical Ethics Committee of the Affiliated Hospital of Jiangnan University (LS2024240) and followed the tenets of the Declaration of Helsinki. All age- and sex-matched participants were informed of the usage of their feces and blood, and written informed consent was obtained. AKI was diagnosed based on the Kidney Disease: Improving Global Outcomes (KDIGO) criteria, which include a serum creatinine increase of 0.3 mg/dL (≥26.5 µM) within 48 h or an increase to the baseline (≥1.5 times) within 7 days, and/or a reduction in output of urine for 6–12 h to <0.5 mL/kg/h. Regarding exclusion criteria, individuals who had cancer of the prostate and small cell lung, as well as infectious diseases, sepsis, allergic dermatitis, enteritis, and arthritis were excluded. The Department of Nephrology, an Affiliated Hospital of Jiangnan University, supplied the renal biopsies, which were carried out as part of routine clinical diagnostics. We collected control samples from kidney poles of healthy individuals (without renal disease) that were undergoing tumor nephrectomies. Prior to the analysis, we collected the entire sera and stored them at −80 °C. Before the fresh renal tissues were used, they were fixed in formalin and embedded in paraffin. Tables S5–S8 detail the characteristics of AKI patients and healthy volunteers. The correlation of serum 3-PG with BUN or Scr was examined using Pearson’s analysis.

Animals

The Animal Research Ethics Committee of Jiangnan University (Ethical code: JN.No20230415t0081030[160]) approved the animal experiments included in this paper. The experiments were carried out in the preclinical laboratory at Jiangnan University. Under controlled conditions with a consistent 12-h light/dark cycle, as well as regulated temperature and humidity, we housed the mice in hygienic cages with unrestricted access to water and standard food. C57BL/6J Cdh16-CreERT2 mice (Strain NO. T007046) and C57BL/6JGpt-Pgk1^em1Cflox/Gpt mice (Strain NO. T065236) were generated by CRISPR/Cas9 strategies and purchased from GemPharmatech (Nanjing, China). Crossing of PGK1^flox/flox mice with Cdh16-Cre mice was performed to produce mice with tamoxifen-inducible PGK1 deletion specific to renal tubular epithelial cells. On the note, we crossed all mice on the background of C57BL/6 for at least three generations. Using specific primers, we confirmed the genotype of PGK1^flox/flox and PGK1 conditional knockout (PGK1^CKO) mice with the PCR technique.

Eight-week-old PGK1^flox/flox and PGK1^CKO mice underwent renal ischemia-reperfusion injury (IRI) surgery. The surgery procedure was performed by making an incision at the rib-spine junction to access the kidneys after we had utilized isoflurane (1.5% for maintenance, 2% for induction) to anesthetize the mice. Renal pedicles were clamped with microaneurysm clips for 30 min. After clip removal, the mice were placed in a 37 °C incubator until they were fully conscious. Prior to confirmation of the anesthesia via tail pinching, we re-anesthetized the mice with isoflurane (5%) after 24 h, before cervical dislocation was performed to sacrifice them. Next, we collected and stored samples of blood and kidneys accordingly at −80 °C. To obtain serum, we centrifuged the samples for 20 min at 800 × g. Prior to histological analysis, we fixed the remaining renal tissues in 4% paraformaldehyde (PFA).

Cisplatin (20 mg/kg) dissolved in saline and injected into eight-week-old C57BL/6J mice to induce AKI mice, while we injected the control mice with the same saline amount. On the other hand, eight-week-old C57BL/6J mice received an intraperitoneal injection of lipopolysaccharides (LPS, Sigma–Aldrich, Cat#L2630) at a dose of 10 mg/kg to induce AKI mice. Eight-week-old PGK1^flox/flox mice and PGK1^CKO mice were subjected to renal IRI surgery (IRIS) and cisplatin treatment, respectively. After 3 days of cisplatin treatment and 24 h of LPS treatment, we re-anesthetized the mice with isoflurane (5%), prior to confirmation of anesthesia with a tail pinch, and performance of cervical dislocation to sacrifice them. We collected and stored blood and kidney samples before performing centrifugation and histological analysis as described above.

To ascertain specific overexpression of PGK1 in renal tubular epithelial cells, we subjected 6-week-old mice to intra-renal pelvic injection of recombinant adeno-associated virus (rAAV) 9 vectors encoding PGK1 or empty vectors (1 × 10¹⁰ viral particles/mouse) under the guidance of Ksp-cadherin promoters^38,69. After 6 weeks, the mice underwent renal I/R surgery to determine the role of PGK1 overexpression (OE) in AKI. The kidneys and serum were collected after 24 h for I/R surgery.

To determine the role of 3-PG in AKI, mice were intraperitoneally injected with 3-PG (30 mg/kg) for 2 consecutive weeks before undergoing renal I/R surgery or cisplatin treatment. Finally, we examined the potential renal protection of a PGK1 inhibitor, L7DG, in a mouse model of AKI induced by renal I/R surgery. Mice received daily intraperitoneal injections of L7DG (10 mg/kg)^70,71 for 3 consecutive days, and then subjected to renal I/R injury. The blood samples were gathered at specific intervals for biochemical assays after renal I/R injury for 24 h. Following the euthanization of the mice, the kidneys were harvested and stored under appropriate conditions for subsequent experiments.

Cell culture, hypoxia/reoxygenation (H/R) model, cisplatin model, and transfection

HK-2 cells were cultured in DMEM/F12 medium supplemented with 10% FBS in a culture incubator (ThermoScientific, USA) at 37 °C with 5% CO₂. A cell H/R model was used to mimic renal I/R injury in vitro. In brief, HK-2 cells were cultured in an anaerobic condition (5% CO₂, 94% N₂, and 1% O₂) with glucose deprivation at 37 °C for 12 h, followed by reoxygenation in a normoxic incubator for 2 h. According to the manufacturer’s instructions, HK-2 cells were seeded in 6-well plates until 80% confluence. Subsequently, cells were transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) with miR-3185 mimics (miR-3185 mimic, 100 nM), mimics control (mock, 100 nM), miR-3185 inhibitor (miR-3185 INH, 100 nM) or inhibitor control (NC INH, 100 nM), siPGK1 (100 nM), siPknox1 (100 nM), siPKM2 (100 nM), siNC (100 nM), PGK1-overexpression plasmid (PGK1 OE, 0.5 μg), and empty vector plasmid (vector, 0.5 μg) for 48 h. After transfection for 24 h, the cells were subjected to normoxic or hypoxic conditions for 12 h, then to normoxia or reoxygenation for 2 h. Then, the cells were collected for biochemical analysis. For the cisplatin model, the cells were stimulated with cisplatin (20 μM) for an additional 24 h after transfection with knockdown or overexpression sequences of target genes for 24 h. Then, the cells were collected for biochemical analysis. HK-2 cells were pre-treated with L7DG (10 μM)⁷² for 24 h, followed by hypoxia for 12 h and reoxygenation for 2 h. Then, the cells were collected for biochemical analysis.

Primary tubular epithelial cells (RTECs) were isolated from mice^73,74. In short, the kidney cortex tissue was harvested and digested with collagenase (3 mg/mL) at 37 °C for 30 min, and the digestion was terminated with FBS. The cell suspension was then filtered through meshes, centrifuged for 20 min at 13,685 × g, and resuspended in primary cell culture medium, a 50:50 mixture of DMEM and F12 supplemented with epidermal growth factor (EGF), ITS, and penicillin/streptomycin (1%). All sequences were synthesized by GenePharma (Suzhou, China). The targeting sequences were as follows:

siPGK1 sense-5′-GGAUGUCUAUGUCAAUGAUGC-3′,

antisense-5′-GCAUCAUUGACAUAGACAUCC-3′′

siPknox1 sense-5′-CCCAGCUUCAGUUACAGUUTT-3′,

antisense-5′-AACUGUAACUGAAGCUGGGTT-3′

siPKM2 sense-5′-CCAUAAUCGUCCUCACCAATT-3′,

antisense-5′-UUGGUGAGGACGAUUAUGGTT-3′

sip53 sense-5′-GAGGUUGGCUCUGACUGUA-3′;

antisense-5′-UACAGUCAGAGCCAACCUC-3′

siNC sense-5′-UUCUCCGAACGUGUCACGU-3′,

antisense-5′-ACGUGACACGUUCGGAGAAdTdT-3′

miRNA-3185-mimics: AGAAGAAGGCGGUCGGUCUGCGG,

GCAGACCGACCGCCUUCUUCUUU′

miRNA-3185-inhibitor: CCGCAGACCGACCGCCUUCUUCU

U6-R AACGCTTCACGAATTTGCGT,

U6-F CTCGCTTCGGCAGCACA

The pcDNA3.1 empty vector and pcDNA3.1-PGK1 plasmids were obtained from Genepharma (Suzhou, China).

Virtual screening of small-molecule inhibitors that target the ADP-binding site of human PGK1

The work by Cliff et al. has reported the fully closed conformation of Human PGK1 in a complex with ADP, 3-PG, and magnesium trifluoride, which is critical for its catalytic activity⁷⁵. Building on this foundation, our study focuses on the ADP-binding pocket of Human PGK1, particularly the surrounding amino acids (ALA214, LYS219, GLY312, ASN336), for the virtual screening of small molecules with strong binding affinities to the target protein. This research utilizes Schrödinger Maestro 11.4 software for virtual screening and PyMol for 3D visualization. The three-dimensional structure of Human PGK1 (PDB ID: 2WZB) was downloaded from the RCSB PDB database. The protein was prepared using the Protein Preparation Wizard module, which involved hydrogenation, removal of water molecules, and deletion of extraneous ions, while retaining the ADP molecule. Energy minimization was then performed using the OPLS2005 force field with an RMSD of 0.30 Å. A grid file was generated centering on the ADP-binding pocket, with a box size of 20 Å × 20 Å × 20 Å. The compound structures from the HY-L001P (MCE Bioactive Compound Library Plus, 20.1K compounds), HY-L009 (Kinase Inhibitor Library, 2.8K compounds), HY-L007 (Immunology/Inflammation Compound Library, 4.4K compounds), and HY-L901 (50K Diversity Library, 50K compounds) were preprocessed using LigPrep, including hydrogenation and energy optimization, with outputs in 3D structures for virtual screening. Virtual screening was conducted using the Screening Workflow module. Prepared compounds were imported, and molecular docking was performed using the Glide module, which matches the receptor and ligand molecules based on geometric and energy compatibility. Initially, high-throughput virtual screening (HTVS) was conducted on compounds from the MCE Bioactive Compound Library Plus. The top 15% scoring molecules were then subjected to a second round of screening using the standard precision (SP) mode, followed by a third round using the extra precision (XP) mode. Similarly, SP mode was applied to the Kinase Inhibitor Library, and the top 15% were further screened using XP mode. The Immunology/Inflammation Compound Library underwent SP screening, with the top 10% further refined using XP mode. For the 50K Diversity Library, HTVS was followed by SP and XP screenings, with the top 10% of compounds being selected at each stage. The final selection involved the manual review of the binding affinity and structural properties of the top 200 compounds from each library: MCE Bioactive Compound Library Plus, Kinase Inhibitor Library, Immunology/Inflammation Compound Library, and 50K Diversity Library.

Histology

Kidney samples were collected and fixed in 4% paraformaldehyde. Subsequently, the specimens were dehydrated through a series of ascending alcohol concentrations, cleared in xylene, embedded in paraffin, and cut into slices with 5-μm thickness. Post-deparaffinization and hydration, the renal sections were stained with a hematoxylin and eosin (H&E) staining kit (Nanjing Jiancheng Bioengineering, Nanjing, China). The histology of kidney tissues was photographed using a digital slice scanner (3DHISTECH, Hungary).

Immunohistochemical staining

The renal tissues were embedded in paraffin after fixing with 4% PFA and subsequently cut into 5-μm-thick sections. The paraffin sections were subjected to baking at 60 °C for 2 h, before we performed deparaffinization, and antigen repair for 15 min. After heat-induced epitope retrieval, the sections were blocked with 5% BSA for 1 h, and then treated with the primary antibodies against PGK1 (#17811-1-AP, Proteintech) at 4 °C overnight. The next day, HPR-linked goat anti-rabbit antibodies were exposed at room temperature, and then the slices were visualized utilizing 3,3-diaminobenzidine (DAB, MCE, Shanghai, China) staining. The images were obtained under a microscope (Nikon, Japan).

Measurement of blood urea nitrogen (BUN) and serum creatinine (Scr)

The serum concentrations of BUN and Scr were determined using commercially available assay kits (Jiancheng Institute of Biotechnology, Nanjing, China) according to the vendors’ recommended procedures. The OD values were read by a microplate reader (Winooski, VT, USA).

TUNEL staining

To evaluate cell apoptosis in the kidney, we performed TUNEL staining using the One-Step TUNEL Apoptosis Assay Kit (Abbkine, Wuhan, China). After deparaffinization, hydration, and antigen retrieval of renal sections, the slices were dyed with a TUNEL reaction mixture at 37 °C for 1 h before we incubated the nuclei for 10 min with DAPI. The percentage of TUNEL-positive cells was imaged and counted using a fluorescence microscope (Zeiss, Germany).

Fe²⁺ content

The free iron levels in the renal tissues and HK-2 cells were evaluated using the FerroOrange dye probe (Dojindo, Japan)^17,76,77. The tissues and cells were stained in a dark environment at 37 °C for a duration of 45 min. Subsequent observation and detection were performed using a fluorescence microscope (Zeiss, Germany), with the quantification of the dye being conducted using the automatic particle analysis feature within ImageJ software (version 1.53t).

Propidium iodide (PI) staining

The apoptosis in HK-2 cells was measured using PI dye (Solarbio, Beijing, China). In brief, the treated cells were incubated with the PI solution in a dim environment at a temperature of 37 °C for 30 min. Subsequently, the cell nuclei were stained with Hoechst dye under the same temperature conditions for a period of 10 min. PI-positive cells were observed and visualized using a fluorescence microscope (Zeiss, Germany).

Dihydroethidium (DHE) staining

The DHE dye (Beyotime, Shanghai, China) was used for the detection and quantification of ROS release in renal tissues and HK-2 cells⁷⁸. The fixed renal sections or HK-2 cells were incubated with the 10 μM DHE working solution under dark conditions at a temperature of 37 °C for 30 min. Subsequently, the fluorescence emitted by the oxidized DHE was captured and recorded using a fluorescence microscope (Zeiss, Germany), and the staining intensity was quantified utilizing ImageJ software (version 1.53t).

Measurement of 3-PG

The contents of 3-PG were measured with commercial ELISA kits according to the manufacturer’s instructions (Shanghai Fantai Biotechnology Co., Ltd). In brief, the standard wells and sample wells were prepared, and a set of calibration standards was constructed. After adding the stop solution (50 μL) to each well, the optical density was measured at 450 nm using a microtiter plate reader for 15 min. The calibration standards were assayed at the same time as the samples, thus producing a standard curve of optical density versus 3-PG contents. The concentrations of 3-PG in each sample were then determined by comparing the optical density of the sample to the standard curve.

Determination of PGK1 activity

The renal tissues or collected cells underwent lysis on ice in extraction solution, and the lysis solutions were centrifuged at 10,000 × g at 4 °C for 10 min, and the supernatants were collected on ice for further testing (BC2255, Solarbio, Beijing, China). The spectrophotometer reader was preheated for more than 30 min, and the wavelength was adjusted to 340 nm. After adding the working buffer in the blank and testing tubes, the absorbance value was measured at 340 nm for 10 s after thorough mixing (A1). Then, the plate was quickly placed in a water bath or incubator at 37 °C for 5 min. The absorbance value was then determined again at 340 nm (A2). The activity of PGK1 was determined via the difference between A2 and A1 and normalized to the protein contents in each sample.

Prussian blue staining

The iron accumulation in renal tissues was assessed using a Prussian blue staining kit (Solarbio, Beijing, China)^79,80. Renal sections were deparaffinized with xylene, followed by hydration through a gradient ethanol series and rinsing with distilled water. The sections were then stained with Prussian blue for 15 min, rinsed thoroughly with distilled water, and counterstained with hematoxylin for 1 min to stain the cell nuclei. After a final rinse in distilled water for 30 s, the sections were dehydrated through a gradient ethanol series, treated with xylene for clearing, and mounted with neutral gum. The stained sections were observed under a light microscope.

Measurement of the iron ion

Cellular iron (Fe²⁺) levels were measured using an iron assay kit (MAK025, Sigma, USA)⁸¹. The collected samples were centrifuged at 10,000 × g for 15 min to collect the supernatant. A total of 100 µL aliquot of the supernatant was incubated with 5 µL of iron buffer for 30 min, followed by the addition of 100 µL of iron probe. After 1 h of incubation, the absorbance at 593 nm was measured immediately.

Measurement of mitochondrial oxygen consumption rate (OCR) and cellular glycolysis

Mitochondrial oxygen consumption rate (OCR) was measured using the Seahorse XF24 Analyzer (Seahorse Bioscience, Cedar Creek, TX)^82,83. HK-2 cells were subjected to H/R with or without PGK1 siRNA. Prior to the assay, cells were approximately 95% confluent. Cells were washed twice with assay medium. After probe calibration, the probe plate was replaced with the cell plate, and the protocol was optimized to separately measure OCR. The Cell Mito Stress Test Kit (Seahorse Bioscience) was used to assess mitochondrial function by sequential injection of the following compounds: oligomycin (1 μM), carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP, 1 μM), and a combination of rotenone (1 μM) and antimycin A (1 μM). Data were analyzed using Seahorse XF24 Wave software. Extracellular acidification rate (ECAR), an indicator of glycolytic activity, was also measured using the Seahorse XF24 Analyzer^82,83. The Glycolytic Stress Test Kit (Seahorse Bioscience) was employed. Glucose (2 mg/ml), oligomycin (1 μM), and 2-deoxy-D-glucose (100 mM) were sequentially added during the assay. ECAR data were processed using Seahorse XF24 Wave software.

Real-time assay of the rate of ATP production

To evaluate real-time ATP production from both glycolysis and mitochondrial respiration, the XF Real-Time ATP Rate Assay Kit (Seahorse Bioscience) was used in conjunction with the Seahorse XF24 Analyzer (Agilent Technologies, formerly Seahorse Bioscience)⁸³. This assay also allowed for calculation of the XF ATP Rate Index, a parameter reflecting the Warburg phenotype. Briefly, cells were seeded in Seahorse XF24 cell culture microplates at a density optimized for each cell type (typically ~4–6 × 10⁴ cells/well) and allowed to adhere overnight in complete growth medium. Prior to the assay, cells were washed and incubated for 1 h at 37 °C in a non-CO₂ incubator with XF Base Medium supplemented with 10 mM glucose, 1 mM pyruvate, and 2 mM glutamine (pH adjusted to 7.4). The XF24 Analyzer measured the oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) simultaneously under basal conditions and following the sequential injection of oligomycin (1.5 μM) to inhibit ATP synthase and rotenone/antimycin A (0.5 μM each) to block mitochondrial respiration completely. The Seahorse Wave software was used to calculate mitochondrial ATP production rate derived from OCR changes after oligomycin injection, glycolytic ATP production rate derived from ECAR changes, corrected for non-glycolytic acidification, XF ATP Rate Index calculated as the ratio of mitochondrial to glycolytic ATP production rates, providing a quantitative measure of metabolic phenotype. All results were normalized to cell number or total protein content to ensure comparability across samples.

Enzyme-linked immunosorbent assay (ELISA)

The levels of interleukin-1β (IL-1β), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), cystatin C, and 8-isoprostane prostaglandin F2α (8-iso-PGF2α)^25,84,85 were measured using ELISA kits (BOSTER, Wuhan, China). All procedures are executed according to the manufacturer’s instructions.

Quantitative real-time-PCR (RT-PCR)

Total RNA was extracted from HK-2 cells and renal tissues utilizing the FreeZol reagent (Vazyme, China), adhering to the manufacturer’s protocols. Thereafter, each sample containing 1 μg RNA was reverse transcribed using the Hifair® III 1st Strand cDNA Synthesis SuperMix (Yeasen, Shanghai, China). Real-time PCR was quantified employing SYBR Green PCR master mix (Yeasen, Shanghai, China) on the Applied Biosystems QuantStudio 3 (Thermo Fisher, USA). The relative mRNA levels were determined using the 2^-∆∆Ct method, normalized to β-actin. Gene-specific primer sequences are listed in Tables S9 and S10.

Detection of lipid peroxidation

The collected HK-2 cells or kidney sections were incubated with 4′,6-Diamidino-2′-phenylindole (DAPI, D8200, Solarbio, Beijing, China) staining solution for 10 min at 37 °C. Subsequently, samples were dyed using Liperfluo working solution (L248, Dojindo, Japan) in PBS for 30 min⁸⁶. The fluorescent signals were observed and imaged under a fluorescence microscope (Axio Vert A1, Zeiss, Oberkochen, Baden-Württemberg, Germany). Additionally, malondialdehyde (MDA, Jiancheng Institute of Biotechnology, Nanjing, China) levels were evaluated using the corresponding kits following the manufacturer’s instructions⁸⁷.

Western blotting

Kidneys and HK-2 cells were lysed using Western and IP Lysis Buffer (Beyotime, Shanghai, China) containing phosphatase and protease inhibitors (Beyotime, Shanghai, China). Protein concentrations were determined using a BCA Protein Assay Kit (Abbkine, Wuhan, China). Protein samples (20 μg) were resolved using 8%–-12% sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) gels and then transferred onto polyvinylidene difluoride (PVDF, Millipore Darmstadt, Germany) membranes. The membranes were blocked with 5% non-fat milk (Beyotime, Shanghai, China) for 1 h at room temperature. After that, the blots were incubated overnight at 4 °C with the primary antibodies against anti-PGK1 (#17811-1-AP, Proteintech, Rosemont, USA), anti-β-actin (#BM0627, Boster, Wuhan, China), anti-NGAL (#26991-1-AP, Proteintech, Rosemont, USA), anti-Bax (#50599-2-lg, Proteintech, Rosemont, USA), anti-Bcl-2 (#A00040-2, Boster, Wuhan, China), anti-ALOX12 (#MA02275, BOSTER, Wuhan, China), anti-PKM2 (#T2100, Abcam, Waltham, MA, USA), anti-p-PKM2 (#PA5-37684, Invitrogen, Carlsbad, CA, USA), anti-LaminB1 (#12987-1-AP, Proteintech, Rosemont, USA), anti-IL-1β (#16806-1-AP, Proteintech, Rosemont, USA), anti-cleaved-caspase 3 (#9661, Cell Signaling Technology, Danvers, MA, USA). After washing with TBST, the membranes were incubated with secondary anti-rabbit IgG (SA00001-2, Proteintech, Rosemont, USA) or anti-mouse IgG (SA00001-1, Proteintech, Rosemont, USA) for 1 h. Western blot of nuclear PKM2 was performed against total protein. Enhanced chemiluminescence solution (ECL, Bio-sharp, Shanghai, China) was used to visualize the membranes, while ImageJ software was utilized to analyze the densities.

Blue native PAGE (BN-PAGE) analysis

The collected protein and BN-PAGE were conducted to analyze dimeric and tetrameric PKM2. We extracted the proteins using a BN-PAGE lysis buffer containing protease and phosphatase inhibitors. The samples were incubated on ice for 30 min, and then the supernatant was transferred into tubes after centrifugation for 10 min at 16,000 × g. After mixing with the BN-PAGE sample buffer, BN-PAGE was carried out on equal amounts of protein at 4 °C and 100 V, followed by transfer onto PVDF membranes. Subsequent steps were the same as Western blotting protocols⁵⁰.

Immunofluorescence staining

The kidney sections were fixed in 4% paraformaldehyde, embedded in paraffin, and then cut into 5-μm slices. The stimulated HK-2 cells or renal sections were permeabilized with 0.1% Triton X-100 (Beyotime Biotechnology, P0096) for 20 min, followed by blocking with 5% BSA at 37 °C for 1 h. After washing with PBS thrice, the samples were incubated overnight at 4 °C with the following primary antibodies, including antibodies against PGK1 (#17811-1-AP, Proteintech, Rosemont, USA), F4/80 (#29414-1-AP, Proteintech, Rosemont, USA), IL-1β (#16806-1-AP, Proteintech, Rosemont, USA), Nox2 (#19013-1-AP, Proteintech, Rosemont, USA), and PKM2 (#T2100, Abcam, Waltham, MA, USA), and Pknox1 (#10614-1-AP, Proteintech, Rosemont, USA). On the second day, the cells or paraffin sections were incubated with the appropriate fluorescence-coupled secondary antibodies for 1 h. The nuclei were then counterstained with DAPI. Fluorescence intensity was photographed and measured using a fluorescence microscope (Zeiss, Germany).

Cell counting Kit‑8 (CCK‑8) assay

HK-2 cells were seeded in 96-well plates overnight. After stimulation, each well was added 10 μL CCK-8 solution (BS350A, Biosharp, Anhui, China), and incubated for 1 h at 37 °C. Absorbance was read at 450 nm using a microplate instrument (Synergy H4, BioTek, Vermont, USA).

Nuclear/cytoplasmic extraction

On the account of the protocol of the manufacturer, we utilized the Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime Biotechnology, Shanghai, China) to extract the nuclear and cytoplasmic proteins from cells. Briefly, cell pellets were incubated with cytoplasmic extraction reagent I (CER I). After vigorous vortexing for 15 s and incubation on ice for 10 min, the cytoplasmic supernatant was extracted and collected using cytoplasmic extraction reagent II (CER II). Then, the pellets were incubated in an ice-cold nucleoprotein extraction reagent (NER). The samples were placed on ice and vortexed every 10 min for 15 s for a period of 40 min. After centrifugation at 16,000 × g for 10 min, the nuclear supernatant was obtained.

GPX4 activity

The measurement of GPX4 activity was conducted using a commercial kit (E-BC-K883-M, Elabscience, Wuhan, China). The determination principle of this reagent kit is that the product generated by GPX4 catalyzing the substrate consumes a reducing agent. The reducing agent has a maximum absorbance at 340 nm. GPX4 inhibitor is added to the system, and GPX4-specific enzyme activity is calculated by measuring non-specific enzyme activity and total enzyme activity. In short, the kidneys or cells were homogenized and centrifuged at 10,000 × g for 10 min at 4 °C, and the supernatant was obtained to undergo protein concentration determination. A total of 20 μL samples were added into the corresponding enzyme-linked immunosorbent assay well, followed by incubation with the reaction working solution. The optical density of each well was calculated at a wavelength of 340 nm on an enzyme-linked immunosorbent assay reader. The amount of enzyme required to catalyze the substrate to produce 1 μM of product per gram of protein per minute is one activity unit at 25 °C.

RNA sequencing

A comprehensive gene expression analysis was conducted by OE Biotech Co., Ltd. (Shanghai, China). Total RNA was extracted using Trizol reagent (Invitrogen, CA, USA), and cDNA samples were sequenced employing the Novaseq 6000 sequencing platform from Illumina. The raw sequencing data were filtered and analyzed in reference to the Mus musculus genome and gene annotations retrieved from the National Center for Biotechnology Information (NCBI) database. Differential gene expression was assessed, and the resultant data were utilized to generate heatmaps and perform Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. For the GO and KEGG enrichment analysis, a gene set was deemed enriched with a P-value < 0.05 and a fold change >2 or <0.05 threshold. Under accession code PRJNA1161211, we have submitted raw RNA sequencing data to the Sequencing Read Archive (SRA).

Drug affinity responsive target stability (DARTS)

The lysates from HK-2 cells were mixed with 10 μM L7DG or isopycnic PBS for 1 h at room temperature. Then, each sample was digested using Pronase E (P861439, Macklin, Shanghai, China) for 30 min, followed by mixing with 2× SDS loading buffer and boiling for 10 min. After these treatments, the protein stability was present using Western blotting.

Cellular thermal shift assay (CETSA)

Collected HK-2 cells were lysed by Western and IP lysis buffer (P0013, Beyotime, Shanghai, China) with protease inhibitors. Simply, lysates (5 mg/mL) were incubated with 10 μM L7DG or PBS and heated for 5 min at gradient temperature (45 °C, 50 °C, 55 °C, 60 °C, 65 °C, 70 °C, and 75 °C). Following centrifugation at 13,000 × g for 15 min at 4 °C, the supernatant was collected and detected by Western blotting.

Detection of LC-MS/MS

To identify PGK1-interacting proteins in renal tissue, co-immunoprecipitation (Co-IP) followed by liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis was performed. Renal tissues from normal mice (n = 2) were homogenized in ice-cold RIPA lysis buffer (Beyotime, supplemented with protease and phosphatase inhibitors). Protein lysates (500 µg per sample) were incubated overnight at 4 °C with anti-PGK1 antibody, followed by incubation with protein A/G magnetic beads (Thermo Fisher Scientific) for 2 h at 4 °C. Negative control immunoprecipitations were performed using normal rabbit IgG. The bead-bound complexes were washed thoroughly and eluted in loading buffer, then subjected to SDS-PAGE. Gels were stained with Coomassie Brilliant Blue, and each lane was cut into ~10 gel bands for in-gel digestion. Gel pieces were destained and subjected to reduction with 10 mM dithiothreitol (DTT) at 56 °C for 30 min, followed by alkylation with 55 mM iodoacetamide in the dark for 30 min at room temperature. After sequential dehydration and rehydration steps, proteins were digested overnight at 37 °C with sequencing-grade trypsin (Promega). Peptides were extracted with 50% acetonitrile/0.1% formic acid, concentrated by vacuum centrifugation, and reconstituted in 0.1% formic acid for LC-MS/MS analysis. Peptide samples were analyzed using an EASY-nLC 1200 ultra-high-performance liquid chromatography system coupled with an Orbitrap Q Exactive Plus mass spectrometer (Thermo Fisher Scientific). Peptides were separated on a C18 reversed-phase column (75 μm × 25 cm, 2 μm particle size) using a linear gradient from 5% to 30% buffer B (0.1% formic acid in acetonitrile) over 90 min at a flow rate of 300 nL/min. Mass spectra were acquired in data-dependent acquisition (DDA) mode. Full MS scans were collected over a range of 350–1600 m/z at a resolution of 70,000 (m/z 200), with an automatic gain control (AGC) target of 3e6 and a maximum injection time of 50 ms. The top 20 most intense ions were selected for fragmentation via high-energy collision dissociation (HCD) with a normalized collision energy of 27%. MS/MS spectra were recorded at a resolution of 17,500, with dynamic exclusion set to 30 s. Raw MS data were processed using Proteome Discoverer v2.4 (Thermo Fisher Scientific) with the Sequest HT search engine against the UniProt mouse reference proteome (2023_01 release). Search parameters included: enzyme: Trypsin (max missed cleavages = 2), fixed modification: Carbamidomethylation (C); variable modifications: Oxidation (M), Acetylation (N-terminus); precursor mass tolerance: 10 ppm; fragment mass tolerance: 0.02 Da. Peptide-spectrum matches (PSMs) were validated using the Percolator algorithm with a strict false discovery rate (FDR) threshold of <1% at the peptide and protein level. Only proteins identified with at least two unique peptides were considered for further analysis. Label-free quantification was based on the intensity of precursor ions. Proteins significantly enriched in PGK1-IP samples were identified using a two-sided Student’s t-test, followed by Benjamini–Hochberg correction for multiple testing (FDR < 0.05).

All LC-MS/MS raw data generated from Co-IP samples were analyzed using Proteome Discoverer software (version 2.4, Thermo Fisher Scientific) with the integrated Sequest HT search engine for peptide and protein identification. Peptides were searched against the UniProtKB/Swiss-Prot Mus musculus (mouse) reference proteome database (release 2023_01), which contains approximately 17,000 reviewed protein entries. A reversed decoy database was generated automatically within the search workflow for accurate false discovery rate (FDR) estimation. Peptide-spectrum match (PSM) validation was performed using the Percolator algorithm, which calculates posterior error probabilities and q-values. PSMs and protein identifications were filtered to a false discovery rate of <1% at both the peptide and protein levels. A minimum of two unique peptides per protein was required for confident identification. The strict parsimony principle was applied, with shared peptides assigned to the most likely protein group (i.e., the leading protein). Not specifically evaluated in this study, as no targeted PTM analysis was performed. Label-free quantification (LFQ) was based on MS1 precursor ion intensities. Peptide intensities were normalized using total ion current (TIC) across samples to correct for sample loading variability. All high-confidence protein identifications from PGK1 co-immunoprecipitates were used for downstream statistical analyses, including differential enrichment in the PGK1-IP sample. The hydrogen deuterium exchange mass spectrometry (HDX-MS) methods and data reporting comply with the community standards described by Masson et al.⁸⁸, and include the HDX summary data as a Table S11.

Co-immunoprecipitation (Co-IP)

The protocol of Co-IP in this study was conducted^89,90. In short, 20 μL of Protein A/G Magnetic Beads (MedChemExpress) were washed thrice with PBST. They were then combined with 2 μg of the required primary antibodies or the specified polyclonal control IgG. The bead-antibody mixture was rotated at room temperature for 2 h. The cells or tissues were lysed using a buffer containing (Beyotime, Shanghai, China), 1 mM DTT (Beyotime, Shanghai, China), and a protease inhibitor cocktail. After centrifugation at 4 °C and 12,000 × g, the supernatant containing proteins was added to the magnetic bead-antibody complex and incubated for 2 h. Subsequently, the immunoprecipitated complexes underwent three washes with lysis buffer, were eluted in sample buffer by boiling for SDS-PAGE, and were subjected to immunoblot analysis.

Plasmid constructs and co-IP assay

Full-length PGK1 and PKM2 sequences were cloned into pcDNA3.1-Flag and pcDNA3.1-hemagglutinin (HA) vectors, generating pcDNA3.1-Flag-PGK1 and pcDNA3.1-HA-PKM2 constructs. Similarly, pcDNA3.1-HA-PGK1 and pcDNA3.1-Flag-PKM2 plasmids were constructed by cloning PGK1 and PKM2 cDNA into pcDNA3.1-HA and pcDNA3.1-Flag vectors, respectively. Plasmids encoding pcDNA5-HA-GST-PGK1 and pcDNA5-HA-GST-PKM2 were generated by cloning the corresponding cDNA into pcDNA5-HA-GST vectors. Additionally, PGK1 fragments (residues 1–123, 1–171, 1–220, and 221–376) were cloned into pcDNA5-Flag and psi-Flag vectors, while PKM2 fragments (residues 1–120, 1–328, 121–433, and 329–532) were cloned into pcDNA5-Flag and psi-Flag vectors. Following transfection of the specified plasmids, 500 μg of protein was incubated with antibody-coupled resin for 2 h. The protein-antibody complexes were then eluted with 50 μl of elution buffer, followed by mixing and washing. The eluted proteins were analyzed by immunoblotting using the respective Tag antibodies.

Luciferase reporter assay

PCR amplified the full-length PGK1 or Pknox1 promoter region from −2995 bp to the transcription start site, and a sequence of PGK1 or Pknox1 promoters was cloned into the pGL3 luciferase vector (Promega). Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA) facilitated the transfection of the PGK1 or Pknox1 promoter luciferase vectors into HK-2 cells. Following a 48-h transfection, luciferase activities were gauged using the Dual-luciferase reporter assay system as per the manufacturer’s protocols (Promega). The firefly luciferase activity was normalized against renilla luciferase activity to account for transfection efficiency, presenting the data as relative luciferase activities.

Chromatin immunoprecipitation (ChIP) assay

A ChIP assay was conducted to identify DNA sequences bound by Pknox1 at Alox12 promoters. A ChIP kit (Ab185913, Abcam, Cambridge, UK) was employed to perform a ChIP experiment, which followed the manufacturer’s protocol. We crosslinked the cells for 10 min with 1% formaldehyde, followed by quenching with glycine (125 mM). Next, we isolated fragments of DNA that are linked with binding proteins after the separation of chromatin. Either Pknox1 antibody or preimmune IgG (CST, 2729, 1:200) was used to subject portions of each chromatin lysate to overnight immunoprecipitation at 4 °C to capture the protein-DNA complexes. Then, reverse cross-linking released DNA before the digestion of proteins. Afterward, we utilized RT-qPCR to discover precipitated genomic DNA using primers that were specific to Pknox1-binding sites within the Alox12 promoters, with non-precipitated genomic DNA serving as an input control. Afterward, we calculated the normalized enrichment value by subtracting the IP relative value from the input relative value.

Structure-based protein interaction between PGK1 and PKM2

For structure-based protein interaction interface analysis between PGK1 and PKM2, we retrieved detailed protein information from the UniProt database, including experimentally determined structure files and sequence data. We prioritized widely validated and accepted data to ensure accuracy in our analyses. Protein structures were preprocessed using UCSF Chimera 1.17.1 software, including hydrogenation, charge calculation, and energy minimization. Molecular docking experiments were then performed using ClusPro, an energy-based tool for predicting protein-protein interactions. This allowed us to explore and identify potential interaction interfaces between PGK1 and PKM2. After docking, we visualized the results with Maestro 13.5 software, enabling detailed analysis and interpretation of the interactions. We focused particularly on interaction interfaces that could reveal functional and mechanistic insights.

Metabolomics

The metabolomic profiling in cells was conducted using a Vanquish UHPLC system coupled to an Orbitrap Q Exactive™ HF or HF-X mass spectrometer (Thermo Fisher Scientific, Germany) by OEbiotech (Shanghai, China). Each sample was extracted in 500 μL of 80% methanol–water, vortexed, chilled on ice for 5 min, and centrifuged at 15,000 × g at 4 °C for 20 min. The supernatant was diluted with MS-grade water to adjust the methanol content to 53%, followed by a second centrifugation under the same conditions. The final supernatant was used for analysis. Chromatographic separation was performed using a 2.1 × 100 mm, 1.9 μm Hypersil GOLD column (Waters, Germany). In positive mode, the mobile phases were 0.1% formic acid (A) and methanol (B); in negative mode, 5 mM ammonium acetate at pH 9.0 (A) and methanol (B). The gradient ran from 2% to 98% B over 10.1 min, held for 1.9 min, at a flow rate of 0.2 mL/min and 40 °C column temperature. Injection volume was 2 μL. Mass spectrometry was operated in both positive and negative ESI modes. Key settings included: spray voltage 3.5 kV, sheath gas 35 psi, auxiliary gas 10 L/min, capillary temperature 320 °C, S-lens RF level 60, and auxiliary gas heater at 350 °C. MS/MS spectra were acquired in data-dependent acquisition mode. Raw data were processed using Compound Discoverer 3.3 (Thermo Fisher Scientific) for peak alignment, detection, and quantification. Processing parameters included 5 ppm mass tolerance, 30% intensity tolerance, and normalization to the first QC sample. Molecular formulas were predicted from ion patterns and matched against the mzCloud (https://www.mzcloud.org/), mzVault, and Mass List databases for identification. To correct for variation, peak intensities were normalized to total ion current. For non-normal distributions, normalization was performed relative to QC values. Metabolites with CVs >30% in QC samples were excluded from downstream analysis. Identified metabolites were annotated using the KEGG, HMDB, and LIPID MAPS databases. Our metabolomics data and methods comply with the community requirements^91,92.

Screening of miRNAs that potentially regulated PGK1

In order to screen candidate miRNAs that regulate PGK1, we used four databases, including TargetScan, miRDB, miRWalk, and miRTarBase, to screen miRNAs that regulate PGK1, while we performed intersection analysis on the results of the four databases using the Venn diagram.

Statistical analysis

Sample sizes for all animal and human experiments in this study were determined based on prior published studies investigating AKI models^93,94, as well as power calculations where applicable. For most in vivo experiments (e.g., Western blot, qPCR, immunohistochemistry, enzyme activity assays), a group size of n = 4–6 per condition was selected to allow detection of biologically meaningful differences with an estimated effect size (Cohen’s d) of ≥1.5, α = 0.05, and power (1–β) of 0.8. These parameters were calculated using G*Power version 3.1 software⁹⁵. For all in vivo experiments, a sample size of six animals per group (n = 6) was used. This number was chosen based on prior studies reporting similar effect sizes in AKI models⁷, and is generally sufficient to detect biologically relevant differences (Cohen’s d > 1.2) with at least 80% statistical power at a significance level of α = 0.05. The selected group size also reflects ethical considerations to minimize animal use while ensuring reproducibility, in accordance with the 3Rs principle. For human serum and tissue comparisons, sample size was based on availability and prior studies showing that such cohort sizes are sufficient for detecting correlations (r ≥ 0.5) with statistical power >0.9 under Pearson’s test⁹⁶. The results were presented as the mean ± SEM. Data analysis and plots were conducted by Prism 8 (GraphPad Software, San Diego, USA). Statistical differences between two groups were analyzed using an unpaired Student’s t-test, while comparisons among multiple groups were performed using ANOVA/Bonferroni test. Statistical significance is denoted by P-values, with thresholds set at less than 0.05.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The RNA sequencing data used in this study are available in the Sequencing Read Archive (SRA) under accession number PRJNA1161211. The mass spectrometry proteomics data generated in this study have been deposited in the ProteomeXchange Consortium under accession number PXD067914 via the iProX partner repository (https://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD067914). The mass spectrometry metabolomics data generated in this study have been deposited in the National Genomics Data Center under accession number OMIX011682. The remaining data that support the findings of this study are available within the article, supplementary information, or Source Data file. Source data are provided with this paper.

References

Ronco, C., Bellomo, R. & Kellum, J. A. Acute kidney injury. Lancet 394, 1949–1964 (2019).
Article CAS PubMed Google Scholar
Bellomo, R. et al. Acute kidney injury in sepsis. Intensive Care Med. 43, 816–828 (2017).
Article CAS PubMed Google Scholar
Gao, L., Zhong, X., Jin, J., Li, J. & Meng, X. M. Potential targeted therapy and diagnosis based on novel insight into growth factors, receptors, and downstream effectors in acute kidney injury and acute kidney injury-chronic kidney disease progression. Signal Transduct. Target. Ther. 5, 9 (2020).
Article CAS PubMed PubMed Central Google Scholar
Medeiros, M. C. et al. Salivary microbiome changes distinguish response to chemoradiotherapy in patients with oral cancer. Microbiome 11, 268 (2023).
Article CAS PubMed PubMed Central Google Scholar
Linkermann, A. Nonapoptotic cell death in acute kidney injury and transplantation. Kidney Int. 89, 46–57 (2016).
Article PubMed Google Scholar
Allison, S. J. Acute kidney injury: AIMing to enhance debris clearance and improve outcomes in AKI. Nat. Rev. Nephrol. 12, 123 (2016).
Article PubMed Google Scholar
Wang, J. N. & Wang, F. Inhibition of METTL3 attenuates renal injury and inflammation by alleviating TAB3 m6A modifications via IGF2BP2-dependent mechanisms. Sci. Transl. Med. 14, eabk2709 (2022).
Article CAS PubMed Google Scholar
Yang, L. et al. Acute kidney injury in China: a cross-sectional survey. Lancet 386, 1465–1471 (2015).
Article PubMed Google Scholar
Xu, X., Nie, S., Xu, H. & Liu, B. Detecting neonatal AKI by serum cystatin C. J. Am. Soc. Nephrol. 34, 1253–1263 (2023).
Article PubMed PubMed Central Google Scholar
Duncan, L., Shay, C. & Teng, Y. PGK1: an essential player in modulating tumor metabolism. Methods Mol. Biol. 2343, 57–70 (2022).
Article CAS PubMed Google Scholar
Wu, Y. Q. & Zhang, C. S. Low glucose metabolite 3-phosphoglycerate switches PHGDH from serine synthesis to p53 activation to control cell fate. Cell Res. 33, 835–850 (2023).
Article CAS PubMed PubMed Central Google Scholar
Shao, F. et al. Associations of PGK1 promoter hypomethylation and PGK1-mediated PDHK1 phosphorylation with cancer stage and prognosis: a TCGA pan-cancer analysis. Cancer Commun. 39, 54 (2019).
Article Google Scholar
Qian, X., Li, X. & Lu, Z. Protein kinase activity of the glycolytic enzyme PGK1 regulates autophagy to promote tumorigenesis. Autophagy 13, 1246–1247 (2017).
Article CAS PubMed PubMed Central Google Scholar
Li, X. et al. Mitochondria-translocated PGK1 functions as a protein kinase to coordinate glycolysis and the TCA cycle in tumorigenesis. Mol. Cell 61, 705–719 (2016).
Article CAS PubMed PubMed Central Google Scholar
Hu, H. et al. Acetylation of PGK1 promotes liver cancer cell proliferation and tumorigenesis. Hepatology 65, 515–528 (2017).
Article CAS PubMed Google Scholar
Nie, H. et al. O-GlcNAcylation of PGK1 coordinates glycolysis and TCA cycle to promote tumor growth. Nat. Commun. 11, 36 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Yan, J. et al. Fucoxanthin alleviated myocardial ischemia and reperfusion injury through inhibition of ferroptosis via the NRF2 signaling pathway. Food Funct. 14, 10052–10068 (2023).
Article CAS PubMed Google Scholar
Cui, X., Zhou, J., Qiu, J., Johnson, M. R. & Mrug, M. Validation of endogenous internal real-time PCR controls in renal tissues. Am. J. Nephrol. 30, 413–417 (2009).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Y., Chen, M., Liu, M. & Xu, Y. Glycolysis-related genes serve as potential prognostic biomarkers in clear cell renal cell carcinoma. Oxid. Med. Cell. Longev. 2021, 6699808 (2021).
Article PubMed PubMed Central Google Scholar
He, Y. et al. PGK1 contributes to tumorigenesis and sorafenib resistance of renal clear cell carcinoma via activating CXCR4/ERK signaling pathway and accelerating glycolysis. Cell Death Dis. 13, 118 (2022).
Article CAS PubMed PubMed Central Google Scholar
He, Y. et al. Novel inhibitors targeting the PGK1 metabolic enzyme in glycolysis exhibit effective antitumor activity against kidney renal clear cell carcinoma in vitro and in vivo. Eur. J. Med. Chem. 267, 116209 (2024).
Article CAS PubMed Google Scholar
Xiao, J. J. et al. Regulator of calcineurin 1 deletion attenuates mitochondrial dysfunction and apoptosis in acute kidney injury through JNK/Mff signaling pathway. Cell Death Dis. 13, 774 (2022).
Article CAS PubMed PubMed Central Google Scholar
Kai, T. et al. Kidney-specific knockout of Sav1 in the mouse promotes hyperproliferation of renal tubular epithelium through suppression of the Hippo pathway. J. Pathol. 239, 97–108 (2016).
Article CAS PubMed Google Scholar
Zhu, X. X. et al. Cichoric acid ameliorates sepsis-induced acute kidney injury by inhibiting M1 macrophage polarization. Eur. J. Pharmacol. 976, 176696 (2024).
Article CAS PubMed Google Scholar
Lu, Q. B. et al. Salusin-β mediates tubular cell apoptosis in acute kidney injury: Involvement of the PKC/ROS signaling pathway. Redox Biol. 30, 101411 (2020).
Article CAS PubMed Google Scholar
Sun, H. J. & Leng, B. Polysulfide and hydrogen sulfide ameliorate cisplatin-induced nephrotoxicity and renal inflammation through persulfidating STAT3 and IKKβ. Int. J. Mol. Sci. 21, 7805 (2020).
Article CAS PubMed PubMed Central Google Scholar
Yang, X. et al. Total alkaloids of Aconitum carmichaelii Debx alleviate cisplatin-induced acute renal injury by inhibiting inflammation and oxidative stress related to gut microbiota metabolism. Phytomedicine 135, 156128 (2024).
Article CAS PubMed Google Scholar
Chen, W. J. et al. Butyrate modulates gut microbiota and anti-inflammatory response in attenuating cisplatin-induced kidney injury. Biomed. Pharmacother 181, 117689 (2024).
Article CAS PubMed Google Scholar
Guo, S. et al. Baicalein alleviates cisplatin-induced acute kidney injury by inhibiting ALOX12-dependent ferroptosis. Phytomedicine 130, 155757 (2024).
Article CAS PubMed Google Scholar
Wu, H. et al. Breaking the vicious loop between inflammation, oxidative stress and coagulation, a novel anti-thrombus insight of nattokinase by inhibiting LPS-induced inflammation and oxidative stress. Redox Biol. 32, 101500 (2020).
Article CAS PubMed PubMed Central Google Scholar
Chen, H. C., Hou, H. Y., Sung, J. M. & Shieh, C. C. Deletion of NADPH oxidase 2 attenuates cisplatin-induced acute kidney injury through reducing ROS-induced proximal tubular cell injury and inflammation. Front. Med. 10, 1097671 (2023).
Article Google Scholar
Abdelnaser, M., Attya, M. E., El-Rehany, M. A. & Fathy, M. Clemastine mitigates sepsis-induced acute kidney injury in rats; the role of α-Klotho/TLR-4/MYD-88/NF-κB/ Caspase-3/ p-P38 MAPK signaling pathways. Arch. Biochem. Biophys. 763, 110229 (2025).
Article CAS PubMed Google Scholar
Li, H. et al. miR-16-5p aggravates sepsis-associated acute kidney injury by inducing apoptosis. Ren. Fail. 46, 2322688 (2024).
Article PubMed PubMed Central Google Scholar
Duan, A. et al. The antiarthritic effect of CBR-470-1 in hypoxic environment is to increase the level of NOD-like receptor family pyrin domain containing 3 ubiquitination by decreasing phosphoglycerate kinase 1 activity. Clin. Transl. Med. 15, e70118 (2025).
Article CAS PubMed Google Scholar
Chen, X. et al. Copper homeostasis and copper-induced cell death in the pathogenesis of cardiovascular disease and therapeutic strategies. Cell Death Dis. 14, 105 (2023).
Article PubMed PubMed Central Google Scholar
Zheng, J. et al. PGK1 inhibitor CBR-470-1 protects neuronal cells from MPP+. Aging 12, 13388–13399 (2020).
Article CAS PubMed PubMed Central Google Scholar
Cebotaru, L., Cebotaru, V., Wang, H., Arend, L. J. & Guggino, W. B. STIM1fl/fl Ksp-Cre mouse has impaired renal water balance. Cell. Physiol. Biochem. 39, 172–182 (2016).
Article CAS PubMed Google Scholar
Schievenbusch, S. et al. Combined paracrine and endocrine AAV9 mediated expression of hepatocyte growth factor for the treatment of renal fibrosis. Mol. Ther. 18, 1302–1309 (2010).
Article CAS PubMed PubMed Central Google Scholar
Tiwari, S. et al. Impaired sodium excretion and increased blood pressure in mice with targeted deletion of renal epithelial insulin receptor. Proc. Natl. Acad. Sci. USA 105, 6469–6474 (2008).
Article ADS CAS PubMed PubMed Central Google Scholar
Contursi, A. et al. Biology and pharmacology of platelet-type 12-lipoxygenase in platelets, cancer cells, and their crosstalk. Biochem. Pharmacol. 205, 115252 (2022).
Article CAS PubMed Google Scholar
Tang, K. et al. Convergence of eicosanoid and integrin biology: 12-lipoxygenase seeks a partner. Mol. Cancer 14, 111 (2015).
Article PubMed PubMed Central Google Scholar
Chu, B. et al. ALOX12 is required for p53-mediated tumour suppression through a distinct ferroptosis pathway. Nat. Cell Biol. 21, 579–591 (2019).
Article CAS PubMed PubMed Central Google Scholar
Wu, Y. et al. CX-5461 ameliorates disease in lupus-prone mice by triggering B-cell ferroptosis via p53-SLC7A11-ALOX12 pathway. Free Radic. Biol. Med. 223, 325–340 (2024).
Article CAS PubMed Google Scholar
Zhao, L. et al. Cigarette smoke extract induces ferroptosis in human retinal pigment epithelial cells. Heliyon 10, e38151 (2024).
Article CAS PubMed PubMed Central Google Scholar
Friedmann Angeli, J. P. et al. Inactivation of the ferroptosis regulator Gpx4 triggers acute renal failure in mice. Nat. Cell Biol. 16, 1180–1191 (2014).
Article CAS PubMed Google Scholar
Shi, Z. et al. Liproxstatin-1 alleviated ischemia/reperfusion-induced acute kidney injury via inhibiting ferroptosis. Antioxidants 13, 182 (2024).
Article CAS PubMed PubMed Central Google Scholar
Gao, S., Hu, J. & Li, Y. Targeting of the Alox12-12-HETE in blast crisis chronic myeloid leukemia inhibits leukemia stem/progenitor cell function. Cancer Manag. Res. 12, 12509–12517 (2020).
Article CAS PubMed PubMed Central Google Scholar
Li, Z. et al. The STING antagonist SN-011 ameliorates cisplatin induced acute kidney injury via suppression of STING/NF-κB-mediated inflammation. Int. Immunopharmacol. 146, 113876 (2025).
Article CAS PubMed Google Scholar
Zaghlool, S. S., Abdelaal, N., El-Shoura, E. A. M., Mahmoud, N. I. & Ahmed, Y. M. Restoring glomerular filtration rate by sulforaphane modulates ERK1/2/JNK/p38MAPK, IRF3/iNOS, Nrf2/HO-1 signaling pathways against folic acid-induced acute renal injury in rats. Int. Immunopharmacol. 123, 110777 (2023).
Article CAS PubMed Google Scholar
Wei, Y. et al. Pyridoxine induces glutathione synthesis via PKM2-mediated Nrf2 transactivation and confers neuroprotection. Nat. Commun. 11, 941 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Xie, W. et al. Pyruvate kinase M2 regulates mitochondrial homeostasis in cisplatin-induced acute kidney injury. Cell Death Dis. 14, 663 (2023).
Article CAS PubMed PubMed Central Google Scholar
Zhou, Z. et al. Oncogenic kinase-induced PKM2 tyrosine 105 phosphorylation converts nononcogenic PKM2 to a tumor promoter and induces cancer stem-like cells. Cancer Res. 78, 2248–2261 (2018).
Article CAS PubMed PubMed Central Google Scholar
Salani, B. et al. IGF1 regulates PKM2 function through Akt phosphorylation. Cell Cycle 14, 1559–1567 (2015).
Article CAS PubMed PubMed Central Google Scholar
Wu, H. et al. PKM2 depletion induces the compensation of glutaminolysis through β-catenin/c-Myc pathway in tumor cells. Cell. Signal. 26, 2397–2405 (2014).
Article CAS PubMed Google Scholar
Wang, S., Yang, S. & Ping, Z. Age-induced accumulation of succinate promotes cardiac fibrogenesis. Circ. Res. 1, 1–23 (2023).
Google Scholar
Chen, D. Q., Han, J., Liu, H., Feng, K. & Li, P. Targeting pyruvate kinase M2 for the treatment of kidney disease. Front. Pharmacol. 15, 1376252 (2024).
Article CAS PubMed PubMed Central Google Scholar
Ye, Y. et al. Pyruvate kinase M2 mediates fibroblast proliferation to promote tubular epithelial cell survival in acute kidney injury. FASEB J. 35, e21706 (2021).
Article CAS PubMed Google Scholar
Zhu, G. et al. Glycolytic enzyme PGK1 promotes M1 macrophage polarization and induces pyroptosis of acute lung injury via regulation of NLRP3. Respir. Res. 25, 291 (2024).
Article CAS PubMed PubMed Central Google Scholar
Zhu, Y. et al. TMEM52B ioforms P18 and P20 differentially promote the oncogenesis and metastasis of nasopharyngeal carcinoma. Adv. Sci. 11, e2402457 (2024).
Article Google Scholar
Lu, Y. et al. Inhibition of phosphoglycerate kinase 1 attenuates autoimmune myocarditis by reprogramming CD4+ T cell metabolism. Cardiovasc. Res. 119, 1377–1389 (2023).
Article CAS PubMed Google Scholar
Wu, W. et al. Targeted delivery of PD-L1-derived phosphorylation-mimicking peptides by engineered biomimetic nanovesicles to enhance osteosarcoma treatment. Adv. Health Mater. 11, e2200955 (2022).
Article Google Scholar
Zhang, T. et al. PGK1 represses autophagy-mediated cell death to promote the proliferation of liver cancer cells by phosphorylating PRAS40. Cell Death Dis. 13, 68 (2022).
Article PubMed PubMed Central Google Scholar
Feng, X. et al. Hypoxia-induced acetylation of PAK1 enhances autophagy and promotes brain tumorigenesis via phosphorylating ATG5. Autophagy 17, 723–742 (2021).
Article CAS PubMed Google Scholar
Qian, X. et al. Phosphoglycerate kinase 1 phosphorylates beclin1 to induce autophagy. Mol. Cell 65, 917–931.e916 (2017).
Article CAS PubMed PubMed Central Google Scholar
Körner, A. et al. Inhibition of Δ24-dehydrocholesterol reductase activates pro-resolving lipid mediator biosynthesis and inflammation resolution. Proc. Natl. Acad. Sci. USA 116, 20623–20634 (2019).
Article ADS PubMed PubMed Central Google Scholar
Fan, Z. & Xin, P. Ferroptosis is crucial for cisplatin induced sertoli cell injury via N6-methyladenosine dependent manner. World J. Mens Health 42, 865–880 (2024).
Article PubMed PubMed Central Google Scholar
Li, C. et al. Inhibition of ALOX12-12-HETE alleviates lung ischemia-reperfusion injury by reducing endothelial ferroptosis-mediated neutrophil extracellular trap formation. Research 7, 0473 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
Yao, Z. et al. The involvement of IRP2-induced ferroptosis through the p53-SLC7A11-ALOX12 pathway in Parkinson’s disease. Free Radic. Biol. Med. 222, 386–396 (2024).
Article CAS PubMed Google Scholar
Saito, S. & Ohno, S. I. rAAV6-mediated miR-29b delivery suppresses renal fibrosis. Clin. Exp. Nephrol. 23, 1345–1356 (2019).
Article CAS PubMed Google Scholar
Ning, B. B. et al. Luteolin-7-diglucuronide attenuates isoproterenol-induced myocardial injury and fibrosis in mice. Acta Pharmacol. Sin. 38, 331–341 (2017).
Article CAS PubMed PubMed Central Google Scholar
Tang, B. X. et al. Luteolin-7-diglucuronide, a novel PTP1B inhibitor, ameliorates hepatic stellate cell activation and liver fibrosis in mice. Acta Pharmacol. Sin. 46, 122–133 (2025).
Article CAS PubMed Google Scholar
Hennebelle, T., Sahpaz, S., Gressier, B., Joseph, H. & Bailleul, F. Antioxidant and neurosedative properties of polyphenols and iridoids from Lippia alba. Phytother. Res. 22, 256–258 (2008).
Article CAS PubMed Google Scholar
Xia, Z. et al. C/EBPα aggravates renal fibrosis in CKD through the NOX4-ROS-apoptosis pathway in tubular epithelial cells. Biochim. Biophys. Acta Mol. Basis Dis. 1870, 167039 (2024).
Article CAS PubMed Google Scholar
Du, R. et al. RTN3 deficiency exacerbates cisplatin-induced acute kidney injury through the disruption of mitochondrial stability. Mitochondrion 75, 101851 (2024).
Article CAS PubMed Google Scholar
Cliff, M. J. et al. Transition state analogue structures of human phosphoglycerate kinase establish the importance of charge balance in catalysis. J. Am. Chem. Soc. 132, 6507–6516 (2010).
Article ADS CAS PubMed Google Scholar
Huang, G. et al. Salidroside sensitizes Triple-negative breast cancer to ferroptosis by SCD1-mediated lipogenesis and NCOA4-mediated ferritinophagy. J. Adv. Res. 74, 589–607 (2024).
Article PubMed PubMed Central Google Scholar
Yang, R. et al. Polyphyllin I induced ferroptosis to suppress the progression of hepatocellular carcinoma through activation of the mitochondrial dysfunction via Nrf2/HO-1/GPX4 axis. Phytomedicine 122, 155135 (2024).
Article CAS PubMed Google Scholar
Murphy, M. P., Bayir, H., Belousov, V. & Chang, C. J. Guidelines for measuring reactive oxygen species and oxidative damage in cells and in vivo. Nat. Metab. 4, 651–662 (2022).
Article PubMed PubMed Central Google Scholar
Wang, K. et al. Lycopene ameliorates polycystic ovary syndrome in rats by inhibiting ovarian ferroptosis through activation of the AMPK/Nrf2 pathway. J. Biochem. Mol. Toxicol. 39, e70158 (2025).
Article CAS PubMed Google Scholar
Won, J. P., Lee, H. G., Yoon, H. J. & Seo, H. G. Biochanin A-mediated anti-ferroptosis is associated with reduction of septic kidney injury. Life Sci. 358, 123124 (2024).
Article CAS PubMed Google Scholar
Zhou, Y. et al. Naringenin inhibits ferroptosis in renal tubular epithelial cells of diabetic nephropathy through SIRT1/FOXO3a signaling pathway. Drug Dev. Res. 86, e70044 (2025).
Article CAS PubMed Google Scholar
Sun, X. et al. Endothelin-1 induces a glycolytic switch in pulmonary arterial endothelial cells via the mitochondrial translocation of endothelial nitric oxide synthase. Am. J. Respir. Cell Mol. Biol. 50, 1084–1095 (2014).
Article PubMed PubMed Central Google Scholar
Sun, X. et al. Mitochondrial fission produces a Warburg effect via the oxidative inhibition of prolyl hydroxylase domain-2. Redox Biol. 81, 103529 (2025).
Article CAS PubMed PubMed Central Google Scholar
Zhu, X. X. et al. Gut microbiome and metabolites mediate the benefits of caloric restriction in mice after acute kidney injury. Redox Biol. 77, 103373 (2024).
Article CAS PubMed PubMed Central Google Scholar
Guo, J., Wang, J. & Feng, J. Aspirin resistance mediated by oxidative stress-induced 8-Isoprostaglandin F2. J. Clin. Pharm. Ther. 44, 823–828 (2019).
Article PubMed Google Scholar
Zou, Y. et al. crVDAC3 alleviates ferroptosis by impeding HSPB1 ubiquitination and confers trastuzumab deruxtecan resistance in HER2-low breast cancer. Drug Resist. Updat. 77, 101126 (2024).
Article CAS PubMed Google Scholar
Zheng, F. et al. Crucial roles of asprosin in cisplatin-induced ferroptosis and acute kidney injury. Free Radic. Biol. Med. 227, 296–311 (2025).
Article CAS PubMed Google Scholar
Masson, G. R. & Burke, J. E. Recommendations for performing, interpreting and reporting hydrogen deuterium exchange mass spectrometry (HDX-MS) experiments. Nat. Methods 16, 595–602 (2019).
Article CAS PubMed PubMed Central Google Scholar
Lu, Q. B. et al. Disrupted cardiac fibroblast BCAA catabolism contributes to diabetic cardiomyopathy via a periostin/NAP1L2/SIRT3 axis. Cell Mol. Biol. Lett. 28, 93 (2023).
Article CAS PubMed PubMed Central Google Scholar
Sun, H. J. et al. Chicoric acid ameliorates sepsis-induced cardiomyopathy via regulating macrophage metabolism reprogramming. Phytomedicine 123, 155175 (2023).
Article PubMed Google Scholar
Alseekh, S. & Aharoni, A. Mass spectrometry-based metabolomics: a guide for annotation, quantification and best reporting practices. Nat. Methods 18, 747–756 (2021).
Article CAS PubMed PubMed Central Google Scholar
Kirwan, J. A. et al. Quality assurance and quality control reporting in untargeted metabolic phenotyping: mQACC recommendations for analytical quality management. Metabolomics 18, 70 (2022).
Article CAS PubMed PubMed Central Google Scholar
Wang, J. N., Suo, X. G. & Yu, J. T. NAT10 exacerbates acute renal inflammation by enhancing N4-acetylcytidine modification of the CCL2/CXCL1 axis. Proc. Natl. Acad. Sci. USA 122, e2418409122 (2025).
Article CAS PubMed PubMed Central Google Scholar
Li, C. et al. Genetic and pharmacological inhibition of GRPR protects against acute kidney injury via attenuating renal inflammation and necroptosis. Mol. Ther. 31, 2734–2754 (2023).
Article CAS PubMed PubMed Central Google Scholar
Faul, F., Erdfelder, E., Buchner, A. & Lang, A. G. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav. Res. Methods 41, 1149–1160 (2009).
Article PubMed Google Scholar
Yu, J. T. et al. Insulin-like growth factor binding protein 7 promotes acute kidney injury by alleviating poly ADP ribose polymerase 1 degradation. Kidney Int. 102, 828–844 (2022).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

This work was funded and supported by the National Natural Science Foundation of China (82300414, 82570420, 82500439, 82370364, and 82170424), Jiangsu Natural Science Foundation (BK20240204, BK20231049, BE2020634 and BK20191138), Clinical Research and Translational Medicine Research Program of Affiliated Hospital of Jiangnan University (LCYJ202306, LCYJ202226), Wuxi Science and Technology Development Fund Project “Light of the Taihu Lake” (K20221028), Wuxi Municipal Health Commission Youth Project (Q202226), Top Talent Support Program for young and middle-aged people of Wuxi Health Committee (HB2023045, BJ2020049, BJ2023029), high-level introduction of talents and scientific research start-up funds of JNU (1286010241230530), the Fundamental Research Funds for the Central Universities (JUSRP124036), Medical Discipline Program of Wuxi Health Commission, the Science and Technology Projects of Wuxi City (M202207, BJ2023029), Chinese Postdoctoral Science Foundation (2023M731344), Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX24_2651).

Author information

These authors contributed equally: Xue-Xue Zhu, Xin-Yu Meng, Ao-Yuan Zhang, Chen-Yang Zhao, Qing-Bo Lu.

Authors and Affiliations

MOE Medical Basic Research Innovation Center for Gut Microbiota and Chronic Diseases, School of Medicine, Jiangnan University, Wuxi, China
Xue-Xue Zhu, Xin-Yu Meng, Ao-Yuan Zhang, Chen-Yang Zhao, Yi-Ting Wei, Yuan-Yuan Wen, Xiao Fu, Guo Chen, An-Jing Xu, Meng-Yuan Wang, Le-Ming Ji, Qing-Bo Lu & Hai-Jian Sun
Department of Anesthesiology, Affiliated Hospital of Jiangnan University, Jiangnan University, Wuxi, China
Jia-Bao Su
Department of Nephrology, Affiliated Hospital of Jiangnan University, Jiangnan University, Wuxi, China
Neng Bao
Research Institute for Reproductive Health and Genetic Diseases, Wuxi Maternity and Child Health Care Hospital, Wuxi, China
Zheng-Yang Bao & Na Li
Department of Endocrinology, Affiliated Hospital of Jiangnan University, Jiangnan University, Wuxi, China
Qing-Bo Lu & Hai-Jian Sun
State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing, China
Hai-Jian Sun

Authors

Xue-Xue Zhu
View author publications
Search author on:PubMed Google Scholar
Xin-Yu Meng
View author publications
Search author on:PubMed Google Scholar
Ao-Yuan Zhang
View author publications
Search author on:PubMed Google Scholar
Chen-Yang Zhao
View author publications
Search author on:PubMed Google Scholar
Yi-Ting Wei
View author publications
Search author on:PubMed Google Scholar
Yuan-Yuan Wen
View author publications
Search author on:PubMed Google Scholar
Jia-Bao Su
View author publications
Search author on:PubMed Google Scholar
Xiao Fu
View author publications
Search author on:PubMed Google Scholar
Guo Chen
View author publications
Search author on:PubMed Google Scholar
An-Jing Xu
View author publications
Search author on:PubMed Google Scholar
Meng-Yuan Wang
View author publications
Search author on:PubMed Google Scholar
Le-Ming Ji
View author publications
Search author on:PubMed Google Scholar
Neng Bao
View author publications
Search author on:PubMed Google Scholar
Zheng-Yang Bao
View author publications
Search author on:PubMed Google Scholar
Na Li
View author publications
Search author on:PubMed Google Scholar
Qing-Bo Lu
View author publications
Search author on:PubMed Google Scholar
Hai-Jian Sun
View author publications
Search author on:PubMed Google Scholar

Contributions

Xue-Xue Zhu: Conceptualization, Data curation, Project administration, Investigation, and writing - original draft. Xin-Yu Meng: Data curation, Investigation, Methodology, Software. Ao-yuan Zhang and Chen-yang Zhao: Investigation, Methodology, Visualization. Yi-Ting Wei, Yuan-Yuan Wen, and Jia-Bao Su: Investigation, Formal analysis. Xiao Fu, Guo Chen, An-Jing Xu, Meng-Yuan Wang, and Le-Ming Ji: Software, Validation. Neng Bao: Methodology, Visualization. Zheng-Yang Bao, Na Li, Qing-Bo Lu, and Hai-Jian Sun: Conceptualization, Funding acquisition; Investigation, Project administration; Resources, Writing - review & editing. All data were generated in-house, and no paper mill was used. All authors agree to be accountable for all aspects of work, ensuring integrity and accuracy.

Corresponding authors

Correspondence to Zheng-Yang Bao, Na Li or Hai-Jian Sun.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Transparent Peer Review file

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Zhu, XX., Meng, XY., Zhang, AY. et al. Inhibition of PGK1 ameliorates acute kidney injury through inactivating the PKM2/ALOX12/ferroptosis pathway in a study with male mice. Nat Commun 16, 9436 (2025). https://doi.org/10.1038/s41467-025-64480-1

Download citation

Received: 10 October 2024
Accepted: 15 September 2025
Published: 24 October 2025
Version of record: 24 October 2025
DOI: https://doi.org/10.1038/s41467-025-64480-1