Dapagliflozin May Attenuate Podocyte Injury in Diabetic Kidney Disease through Inhibition of Lactate-Mediated H4K12 Lactylation

Abstract

Introduction

Podocyte injury is central to diabetic kidney disease (DKD). Metabolic reprogramming, particularly lactate accumulation, may drive podocyte injury via epigenetic mechanisms, yet this process remains poorly understood. This study aimed to clarify lactate’s role in podocyte injury and explore the protective effects of dapagliflozin.

Methods

We retrospectively analyzed plasma lactate levels in healthy individuals, patients with diabetes, and DKD patients and examined correlations with proteinuria. Subsequently, db/db mice and cultured podocytes were used to explore mechanisms of lactate-induced podocyte injury and potential protective effects of dapagliflozin in vivo and in vitro.

Results

Plasma lactate levels were analyzed in 66 healthy controls, 150 patients with diabetes, and 152 patients with DKD. Lactate concentrations were significantly higher in DKD patients compared with both diabetic patients and healthy controls. In DKD patients, lactate levels were positively correlated with serum creatinine and albuminuria. Logistic regression analysis identified hypertension, plasma lactate, white blood cell count, glycated hemoglobin, and serum creatinine as independent risk factors for DKD. Notably, dapagliflozin treatment significantly reduced plasma lactate levels in DKD patients. In diabetic mice and high-glucose-stimulated podocytes, marked lactate accumulation accompanied by increased histone H4 lysine 12 lactylation (H4K12la) was observed, indicating lactate-driven epigenetic regulation. In contrast, dapagliflozin treatment reduced lactate production and H4K12la levels, attenuated proteinuria, and ameliorated podocyte injury.

Conclusions

Lactate contributes to podocyte injury in DKD at least in part through promoting H4K12 lactylation. Dapagliflozin exerts renoprotective effects by attenuating this lactate-mediated epigenetic pathway, highlighting a potential therapeutic mechanism beyond glucose control.

Introduction

Diabetic kidney disease (DKD) is one of the most prevalent and disabling microvascular complications of diabetes mellitus (DM) and remains a leading cause of end-stage renal disease [1]. Its hallmark clinical manifestations include persistent proteinuria and a progressive decline in glomerular filtration rate (GFR) [2]. Despite substantial advances in the prevention and management of diabetes, the global burden of DKD continues to rise, particularly in developing countries and regions with rapidly aging populations [3].

Emerging evidence indicates that metabolic reprogramming plays a critical role in maintaining energy homeostasis and redistributing metabolic substrates, thereby contributing to the onset and progression of DKD [4, 5]. Under hyperglycemic conditions, glucose metabolism shifts from oxidative phosphorylation toward aerobic glycolysis, resulting in excessive lactate production [6]. Accumulated lactate has been shown to induce mitochondrial dysfunction, activate pro-inflammatory signaling, and promote extracellular matrix deposition, thereby exacerbating renal injury [7, 8]. Notably, recent discoveries have revealed that lactate is not merely a metabolic byproduct but also functions as a signaling molecule that mediates histone and non-histone protein lactylation, thereby regulating gene transcription and cellular functions [5, 9]. These findings suggest that lactate acts as a pivotal metabolic and epigenetic mediator in the pathophysiology of DKD.

Podocytes are highly specialized, terminally differentiated epithelial cells that form the outermost layer of the glomerular filtration barrier. Their interdigitating foot processes, connected by the slit diaphragm, are essential for maintaining selective filtration. Disruption of this structure, along with foot process effacement, contributes critically to the pathogenesis of proteinuria in DKD [10]. Maintaining podocyte energy homeostasis is thus vital for preserving glomerular integrity. Under hyperglycemic conditions, podocyte glucose metabolism becomes dysregulated, leading to lactate accumulation. Elevated lactate has been shown to promote epithelial-mesenchymal transition through histone H3 lysine 14 (H3K14) lactylation, contributing to podocyte injury [11]. Moreover, lactate induces non-histone lactylation of leucyl-tRNA synthetase 1, which suppresses autophagy and promotes podocyte apoptosis, ultimately aggravating proteinuria [12]. Collectively, these observations suggest that reducing lactate accumulation may represent a potential therapeutic strategy to alleviate podocyte injury in DKD.

Sodium-glucose cotransporter 2 (SGLT2) inhibitors have emerged as widely used glucose-lowering agents in recent years [13]. Since the approval of dapagliflozin in 2012, this drug class has become integral to the management of type 2 diabetes and its complications [14]. Beyond glycemic control, SGLT2 inhibition decreases proximal tubular reabsorption of sodium and glucose, enhances sodium chloride delivery to the macula densa, restores tubuloglomerular feedback, and induces afferent arteriolar vasoconstriction. These changes lower intraglomerular pressure, attenuate hyperfiltration, and reduce urinary protein excretion. At the podocyte level, SGLT2 inhibitors have been reported to preserve podocyte integrity, inhibit apoptosis, enhance autophagy, and mitigate inflammatory responses, although the underlying molecular mechanisms remain incompletely defined [15–17].

Evidence from nonrenal tissues provides important mechanistic clues. In diabetic cardiomyocytes, lactate accumulation can induce macrophage histone H4 lysine 12 lactylation (H4K12la), leading to oxidative stress and myocardial injury [18]. Furthermore, dapagliflozin has been shown to improve glucose metabolism and reduce epicardial lactate production in patients with coronary artery disease, thereby limiting inflammatory infiltration associated with epicardial adipose tissue expansion [19]. These findings raise the possibility that SGLT2 inhibitors may exert podocyte-protective effects by modulating lactate metabolism and lactate-derived epigenetic modifications; however, direct experimental evidence supporting this mechanism is still lacking.

In this context, the present study aimed to (1) determine whether plasma lactate levels are elevated in patients with DKD and examine their association with DKD progression and (2) explore whether SGLT2 inhibitors alleviate podocyte injury by regulating lactate metabolism and histone H4K12la within podocytes. Clarifying these mechanisms may not only deepen our understanding of DKD pathophysiology but also provide a novel mechanistic rationale for the renoprotective effects of SGLT2 inhibitors.

Methods

The research workflow is shown in Figure 1. First, we aimed to determine whether plasma lactate levels are significantly elevated in patients with DKD and whether lactate serves as an independent risk factor for DKD progression. Second, we evaluated whether SGLT2 inhibitors can reduce the risk of hyperlactatemia in patients with DKD. Finally, through in vivo animal models and in vitro podocyte experiments, we investigated whether SGLT2 inhibitors alleviate podocyte injury in DKD by suppressing lactate-mediated H4K12la.

Fig. 1.

Research flowchart.

Human Study

We retrospectively enrolled 66 healthy controls, 150 patients with diabetes, and 152 patients with DKD at the First Affiliated Hospital of Guangzhou Medical University between January 2020 and April 2024. Baseline demographic and clinical data, including age, sex, vital signs, body mass index (BMI), laboratory test results, comorbidities, and medication history, were obtained from the hospital’s electronic medical records.

Participants were categorized into three groups according to the following inclusion criteria: (1) Healthy control group: adults aged ≥18 years without DM, impaired glucose tolerance, impaired fasting glucose, metabolic syndrome, or chronic kidney disease and without comorbidities known to significantly affect lactate metabolism, such as severe uncontrolled hypertension, infectious diseases, or malignancy. (2) DM group: adults with type 2 diabetes mellitus (T2DM) meeting at least one of the following criteria: a diagnosis based on the 1999 World Health Organization (WHO) criteria; hemoglobin A1c (HbA1c) ≥6.5% according to the 2011 WHO recommendation; a previously documented diagnosis of T2DM; or current use of glucose-lowering therapy. Participants in this group had no proteinuria and no chronic kidney disease. (3) DKD group: adults with T2DM, as defined above, who were diagnosed with DKD according to the KDIGO 2022 Clinical Practice Guideline for Diabetes Management in Chronic Kidney Disease.

The exclusion criteria were as follows: (1) Type 1 DM, gestational diabetes, other specific forms of diabetes, or acute diabetic complications; (2) a confirmed diagnosis of kidney disease unrelated to diabetes, such as membranous nephropathy, IgA nephropathy, lupus nephritis, or antineutrophil cytoplasmic anti-body-associated vasculitis; (3) acute kidney injury within the preceding 3 months or current renal replacement therapy; (4) recent use of medications known to affect glucose metabolism (e.g., systemic corticosteroids, calcineurin inhibitors) or nephrotoxic agents; (5) active infectious, hematologic, autoimmune, or malignant disease; severe psychiatric disorders; or acute stress conditions (e.g., recent surgery or trauma); (6) symptomatic chronic heart failure, acute cardiovascular events within the past 3 months (e.g., hospitalization for heart failure, myocardial infarction, or stroke), or severe respiratory failure; (7) active hepatitis, tuberculosis, syphilis, or other communicable diseases; (8) pregnancy or lactation; and (9) incomplete clinical data.

This study was approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou Medical University (Approval No. ES-2024-K091-02, Guangzhou, China). The requirement for informed consent was waived due to the retrospective nature of the study. All procedures were conducted in accordance with the Declaration of Helsinki and the ethical standards of the Human Research Ethics Committee of the First Affiliated Hospital of Guangzhou Medical University. The study was designed and reported in accordance with the STROBE Statement for observational studies.

Animal Study

All animal procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and were approved by the Animal Care and Use Committee of the First Affiliated Hospital of Bengbu Medical University (Approval No. 2025735). The study was designed and reported in accordance with the ARRIVE Guidelines.

Eight-week-old male C57BL/KsJ db/db mice and their heterozygous non-diabetic littermates (db/m) were obtained from Hangzhou Ziyuan Laboratory Animal Technology Co., Ltd. The db/db mice carry a mutation in the leptin receptor gene and spontaneously develop obesity, hyperglycemia, and diabetic kidney injury and are widely used as a model of type 2 diabetes, whereas db/m mice do not develop diabetes and serve as non-diabetic controls. Following a 2-week acclimatization period, the mice were randomly assigned to four groups (n = 6 per group): db/m, db/db, low-dose dapagliflozin (db/db + 4 mg/kg/day), and high-dose dapagliflozin (db/db + 8 mg/kg/day). DAPA was administered by oral gavage once daily for 10 weeks. The selected doses and treatment duration were based on previous studies demonstrating that long-term administration of DAPA at mg/kg-level doses effectively ameliorates renal injury in db/db mice [20, 21].

At 21 weeks of age, all mice were euthanized. Serum and urine samples were collected, and kidneys were harvested for subsequent analyses.

Measurement of Lactate Concentrations

Renal tissues or cultured podocytes were homogenized in Lactate Assay Buffer and centrifuged at 12,000 rpm for 10 min at 4°C. The resulting supernatants were collected, and lactate levels were quantified using an L-Lactate Assay Kit (BC2235, Solarbio), according to the manufacturer’s instructions.

Mouse Urine Albumin/Creatinine Detection

Urinary albumin in mice was quantified using an ELISA kit (Bethyl Laboratories) according to the manufacturer’s instructions. Urinary creatinine was measured with an enzymatic ELISA kit (Crystal Chem) following the manufacturer’s protocol. The degree of albuminuria was expressed as the albumin-to-creatinine ratio (mg/g).

Kidney Histological Evaluation

Mouse kidneys were fixed in 4% paraformaldehyde and embedded in paraffin. The tissues were sectioned at a thickness of 4 μm and stained with periodic acid-Schiff (PAS). Histological alterations in the kidney were assessed based on PAS staining following previously established methods [22]. Quantification of the glomerular area and mesangial expansion index was performed using Image Pro-Plus 6.0 software (Media Cybernetics).

Transmission Electron Microscopy

Transmission electron microscopy was performed to examine the thickness of the glomerular basement membrane (GBM) and the morphology of podocyte foot processes, following previously reported procedures [22, 23].

Cell Culture and Treatment

Conditionally immortalized mouse podocytes (MPC-5, Wuhan Pricella Biotechnology Co., Ltd., Cat. No. CL-0855) were cultured as previously described [22, 24]. Cells were maintained at 33°C in proliferation medium. To induce differentiation, podocytes were reseeded and cultured at 37°C for 10∼14 days. Differentiated podocytes were then exposed to high glucose (HG, 30 mm) or an osmotic control (mannitol, 30 mm) for 48 h, a standard duration for inducing podocyte injury in vitro [22, 25]. Based on previous studies showing that concentrations below 10 μm are well tolerated in podocytes [26] and renal tubular epithelial cells [27], cells were co-incubated with 4 μm (low-dose dapagliflozin) or 8 μm (high-dose dapagliflozin) dapagliflozin during HG stimulation. All experiments were independently repeated at least three times.

Transfection of siRNA

Differentiated podocytes were transfected with lactate dehydrogenase A (LDHA) – targeting siRNA (50 nm) or a nonspecific control siRNA (Ribobio) using Lipofectamine 2000 (Invitrogen, Thermo Fisher Scientific) according to the manufacturer’s instructions. Following transfection, cells were exposed to high glucose (HG, 30 mmol/L) for 48 h. Podocytes cultured under normal glucose conditions (5.3 mmol/L) served as the control group, and an osmolality-matched control group was established by adding mannitol to normal glucose medium (final concentration 30 mmol/L).

Immunofluorescence Staining

Frozen kidney sections (5 μm) were fixed with 4% paraformaldehyde, blocked with 5% bovine serum albumin, and incubated overnight at 4°C with primary antibodies against WT-1 (ab89901, Abcam, 1:200) and H4K12la (PTM-1411RM, PTM-BIO, 1:200). After washing with PBS, sections were incubated with species-appropriate fluorescent secondary antibodies (1:500) for 1 h at room temperature in the dark. Nuclei were counterstained with DAPI. Images were acquired using a confocal microscope.

Western Blot Analysis

Protein was extracted from the kidney cortex, and Western blot analysis was performed as previously described [22, 28]. The antibodies used were anti-synaptopodin (sc-515842; Santa Cruz Biotechnology, 1:1,000), anti-podocin (P0372; Sigma-Aldrich, 1:1,000), anti-H4K12la (PTM-1411RM; PTM-BIO, 1:1,000), anti-histone H4 (ab51997; Abcam, 1:2,000), and anti-glyceraldehyde-3-phosphate dehydrogenase (5174S; Cell Signaling Technology, 1:5,000). Quantification of protein bands was performed using the ImageJ program (National Institutes of Health, Bethesda, MD). Expression levels of synaptopodin and podocin were normalized to glyceraldehyde-3-phosphate dehydrogenase, whereas H4K12la levels were normalized to histone H4.

Statistical Analysis

Statistical analyses were performed using SPSS software (version 25.0; IBM Corp., Armonk, NY, USA) and GraphPad Prism (version 9.5; GraphPad Software, San Diego, CA, USA). Data with a normal distribution are presented as the mean ± standard deviation (SD). Differences between two groups were analyzed using an independent samples t test, whereas comparisons among multiple groups were assessed using one-way analysis of variance followed by the least significant difference post hoc test when overall significance was observed. Non-normally distributed data are expressed as the median (interquartile range), and comparisons between two or more groups were conducted using the Mann-Whitney U test or Kruskal-Wallis test, respectively. Categorical variables are presented as percentages (%), and group differences were evaluated using the chi-square test. Binary logistic regression analysis was applied to identify factors associated with hyperlactatemia and diabetic DKD. A p value <0.05 was considered statistically significant.

Results

Elevated Plasma Lactate Level in DKD Patients

The clinical characteristics of 66 healthy controls, 150 patients with diabetes, and 152 patients with DKD are summarized in Table 1. Age, prevalence of hypertension, and plasma lactate levels progressively increased from the healthy control group to the diabetes and DKD groups (p < 0.05), consistent with the findings reported by Jiang et al. [29]. Although previous studies have suggested that metformin may elevate lactate levels [30], metformin use did not significantly differ between the diabetes and DKD groups in this research (p = 0.132). Compared with the other groups, DKD patients exhibited higher systolic blood pressure, longer diabetes duration, and a greater prevalence of smoking (p < 0.01). In addition, both DKD and diabetic patients showed higher BMI and a greater incidence of dyslipidemia compared with healthy controls (p < 0.05). Several laboratory parameters also differed significantly among the three groups – including white blood cell (WBC) count, hemoglobin, glycosylated hemoglobin (HbA1c), blood urea nitrogen (BUN), serum creatinine (SCr), triglycerides (TG), high-density lipoprotein cholesterol (HDLC), lactate dehydrogenase (LDH), albumin (ALB), aspartate aminotransferase (AST), alanine aminotransferase (ALT), eGFR, urinary protein, and albuminuria – with most comparisons showing statistical significance (p < 0.05).

Table 1.

Comparisons of baseline characteristics in all participants

Variables	Control (n = 66)	Diabetes (n = 150)	DKD (n = 152)	p value
Age, years	54.18±11.90	57.61±9.98**	64.02±12.48**,a	<0.001
Sex, n (%)	​	​	​	0.080
Female	37 (56.1)	75 (50.0)	62 (40.8)**,a	​
Male	29 (43.9)	75 (50.0)	90 (59.2)**,a	​
Smoking, n (%)	​	​	​	0.007
Yes	23 (34.8)	51 (34.0)	77 (50.7)**,a	​
No	43 (65.2)	99 (66.0)	75 (49.3)**,a	​
Hypertension, n (%)	​	​	​	<0.001
Yes	12 (18.2)	53 (35.3)**	115 (75.7)**,a	​
No	54 (81.8)	97 (64.7)**	37 (24.3)**,a	​
Hyperlipidemia, n (%)	​	​	​	0.013
Yes	28 (42.4)	96 (64.0)**	88 (57.9)**	​
No	38 (57.6)	54 (36.0)**	64 (42.1)**	​
MET, n (%)	​	​	​	0.132
Yes	NA	66 (43.0)	54 (35.5)	​
No	NA	84 (56.0)	98 (64.5)	​
BMI, kg/m2	21.96 [20.20, 25.80]	23.88 [21.52, 26.67]**	24.09 [22.05, 26.62]**	0.007
SBP, mm Hg	123 [113, 133]	128 [115, 138]	140 [126, 155]**,a	<0.001
DBP, mm Hg	76 [70, 87]	80 [73, 88]	79 [69, 87]	0.223
Duration of diabetes, years	NA	5 [1, 12]	11 [5, 18]a	<0.001
Plasma lactate, mmol/L	1.44 [1.24, 1.70]	1.88 [1.45, 2.33]**	1.91 [1.53, 2.38]**,a	<0.001
WBC, 109/L	5.99 [5.06, 7.13]	6.63 [5.80, 8.14]**	7.55 [6.40, 8.66]**,a	<0.001
HB, g/L	129 [120, 136]	133 [125, 145]	126 [110, 138]a	<0.001
PLT, 109/L	223 [192, 261]	234 [196, 270]	237 [189, 279]	0.423
HbA1c, %	5.54 [5.38, 5.82]	8.07 [6.74, 9.82]**	8.50 [7.12, 10.01]**	<0.001
BUN, mmol/L	5.15 [4.30, 5.90]	5.75 [4.70, 7.00]*	8.60 [6.45, 11.85]**,a	<0.001
SCr, µmol/L	69.7 [63.7, 76.3]	69.4 [61.4, 78.1]	108.2 [84.8, 149.9]**,a	<0.001
CHOL, mmol/L	4.90±1.03	5.12±1.23	5.04±1.83	0.605
TG, mmol/L	0.81 [1.08, 1.39]	1.45 [0.97, 2.00]**	1.64 [1.22, 2.59]**,b	<0.001
HDLC, mmol/L	1.26 [1.13, 1.45]	1.17 [1.03, 1.36]*	1.17 [0.99, 1.39]	0.027
LDLC, mmol/L	3.01 [2.70, 3.61]	3.30 [2.72, 3.88]	3.13 [2.40, 4.04]	0.381
LDH, U/L	164.7 [154.7, 185.0]	175.2 [155.4, 195.8]	193.0 [168.6, 226.6]**,a	<0.001
ALB, g/L	38.8 [37.0, 40.6]	40.2 [37.9, 42.2]	37.1 [34.4, 40.3]*,a	<0.001
AST, U/L	19.4 [16.1, 23.5]	20.3 [17.7, 26.8]	20.0 [16.3, 25.0]a	0.021
ALT, U/L	15.1 [12.5, 22.0]	19.4 [14.2, 27.6]**	16.6 [11.9, 25.9]a	0.001
eGFR, mL/min/1.73 m2	93.33±13.81	93.12±14.10	52.95±27.18**,a	<0.001
Urine protein, g/24 h	0.08 [0.06, 0.11]	0.08 [0.06, 0.10]	0.52 [0.21, 1.38]**,a	<0.001
Albuminuria, mg/24 h	5.98 [2.26, 10.89]	6.05 [4.06, 10.80]	302.00 [90.21, 771.08]**,a	<0.001

DKD, diabetic kidney disease; MET, metformin; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; WBC, white blood cell; HB, hemoglobin; PLT, platelet; HbA1c, glycosylated hemoglobin; BUN, blood urea nitrogen; SCr, serum creatinine; CHOL, cholesterol; TG, triglycerides; HDLC, high-density lipoprotein cholesterol; LDLC, low-density lipoprotein cholesterol; LDH, lactate dehydrogenase; ALB, albumin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; eGFR, estimated glomerular filtration rate.

*p < 0.05, **p < 0.01 vs. control; ^ap < 0.01, ^bp < 0.05 vs. diabetes.

Plasma Lactate Levels Were Positively Correlated with SCr and Albuminuria

Considering previous evidence that metformin can elevate lactate production, whereas SGLT2 inhibitors, angiotensin-converting enzyme inhibitors (ACEIs), and angiotensin II receptor blockers (ARBs) may reduce lactate levels [30, 31], we further analyzed correlations after excluding patients receiving these medications. In Spearman rank correlation analyses (Fig. 2a, b), plasma lactate was positively associated with serum creatinine (r = 0.185, p = 0.016) and albuminuria (r = 0.203, p = 0.008). Moreover, lactate levels showed positive correlations with BMI, WBC count, LDH, and HbA1c (all p < 0.05; online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000552052).

Fig. 2.

Lactate levels were positively correlated with SCr and albuminuria. Logistic regression analysis identified hypertension, plasma lactate, WBC, HbA1c, and SCr as independent risk factors for DKD. a Plasma lactate is positively correlated with SCr (p = 0.016). b Plasma lactate is positively correlated with albuminuria (p = 0.007). c Multivariate analyses of risk factors for DKD. Hypertension, plasma lactate, WBC, HbA1c, and SCr are independent risk factors for DKD (p < 0.05). Model adjusted for age, sex, smoking, hypertension, hyperlipidemia, BMI, plasma lactate, WBC, hemoglobin, HbA1c, SCr, ALB, and MET use. Data are expressed as OR (95% CI), unless stated otherwise. SCr, serum creatinine; OR, odds ratio; CI, confidence interval; BMI, body mass index; WBC, white blood cell; HB, hemoglobin; HbA1c, glycosylated hemoglobin; ALB, albumin; MET, metformin.

Logistic Regression Analysis Identified Hypertension, Plasma Lactate, WBC Count, Glycated Hemoglobin, and SCr as Independent Risk Factors for DKD

Because proteinuria is a hallmark feature of DKD and may confound other associations, proteinuria-related variables were excluded from the logistic regression analysis. Univariable binary logistic regression revealed that age, sex, smoking, hypertension, systolic blood pressure (SBP), plasma lactate, WBC count, hemoglobin, BUN, SCr, HbA1c, triglycerides, LDH, albumin, and eGFR were significantly associated with DKD (p < 0.05; online suppl. Table 2). In multivariable models adjusting for potential confounders, hypertension, plasma lactate (p < 0.01), WBC count, HbA1c, and SCr remained independent predictors of DKD (p < 0.05; Fig. 2c). Notably, the positive association between elevated lactate levels and DKD persisted even after adjusting for metformin use.

Dapagliflozin Is Associated with Reductions in Plasma Lactate and Albuminuria in Patients with DKD

SGLT2 inhibitors effectively reduce proteinuria through multiple complementary mechanisms. By increasing sodium delivery to the macula densa, they restore tubuloglomerular feedback and lower intraglomerular pressure, thereby attenuating glomerular hyperfiltration [13, 17, 32]. Additionally, SGLT2 inhibition alleviates oxidative stress, inflammation, and mitochondrial dysfunction in podocytes, preserving slit diaphragm integrity and reducing albumin leakage [16, 26, 33]. However, whether SGLT2 inhibitors affect lactate/lactation modification remains to be further investigated. To minimize confounding effects from medications known to influence lactate metabolism, patients receiving ACEIs, ARBs, or metformin were excluded (the distribution of these medications among DKD patients is shown in online suppl. Table 3). Compared with the non-SGLT2 group, patients treated with SGLT2 inhibitors exhibited a significantly lower incidence of hyperlactatemia (40.0% vs. 16.0%, p = 0.041). Consistently, plasma lactate levels were reduced in the SGLT2 group (2.07 ± 0.73 vs. 1.60 ± 0.55 mmol/L, p = 0.008), accompanied by lower lactate dehydrogenase (p = 0.011) and albuminuria (p = 0.033). No significant differences were observed in other baseline clinical characteristics (p > 0.05; online suppl. Table 4). To further assess the association between SGLT2 inhibitor use and hyperlactatemia, we conducted binary logistic regression with hyperlactatemia as the dependent variable. After adjusting for potential confounders, SGLT2 inhibitor use remained independently associated with a reduced risk of hyperlactatemia (p = 0.034), whereas white blood cell count showed a positive association (p = 0.030). Collectively, these findings indicate that SGLT2 inhibitor use is associated with approximately an 87% reduction in the odds of hyperlactatemia among patients with DKD (Table 2). However, the underlying mechanisms by which SGLT2 inhibitors may modulate lactate metabolism and thereby ameliorate proteinuria warrant further investigation.

Table 2.

Univariate and multivariate binary logistic regression analyses for hyperlactatemia in DKD patients

Variables	Univariate analysis		Multivariate analysis
	OR (95% CI)	p value	OR (95% CI)	p value
SGLT2i use	0.258 (0.075, 0.890)	0.032	0.132 (0.020, 0.856)	0.034
Age, years	0.980 (0.944, 1.017)	0.278	0.981 (0.919, 1.047)	0.570
Sex	2.015 (0.623, 6.513)	0.242	2.999 (0.392, 22.966)	0.290
Smoking	2.000 (0.677, 5.906)	0.201	1.349 (0.231, 7.886)	0.740
Hypertension	1.167 (0.373, 3.647)	0.791	2.477 (0.469, 13.077)	0.285
Hyperlipidemia	1.655 (0.508, 5.397)	0.403	0.535 (0.088, 3.267)	0.498
BMI, kg/m2	1.192 (1.002, 1.418)	0.048	1.203 (0.944, 1.533)	0.136
DM course, years	0.945 (0.882, 1.013)	0.113	0.942 (0.829, 1.070)	0.357
WBC, 109/L	1.342 (1.009, 1.785)	0.044	1.652 (1.051, 2.595)	0.030
HbA1c (%)	0.990 (0.794, 1.235)	0.931	1.107 (0.766, 1.600)	0.588
LDH, U/L	1.007 (0.999, 1.014)	0.083	1.002 (0.992, 1.012)	0.688
ALB, g/L	1.089 (0.973, 1.220)	0.138	1.079 (0.909, 1.280)	0.386
eGFR, mL/min/1.73 m2	1.002 (0.985, 1.018)	0.856	0.977 (0.940, 1.014)	0.222
Albuminuria, mg/24 h	1.000 (1.000, 1.000)	0.737	0.999 (0.999, 1.000)	0.073

SGLT2i, sodium-dependent glucose transporter 2 inhibitor; BMI, body mass index; DM, diabetes mellitus; WBC, white blood cell; HbA1c, glycosylated hemoglobin; LDH, lactate dehydrogenase; ALB, albumin; eGFR, estimated glomerular filtration rate.

Dapagliflozin May Attenuate Podocyte Injury in DKD through Inhibition of Lactate-Mediated H4K12 Lactylation

Lactate-mediated lysine lactylation is a recently identified post-translational modification driven by elevated intracellular lactate [4, 34]. Increased lactate levels in the kidney, particularly under diabetic or hypoxic conditions, can enhance histone and non-histone lactylation and thereby alter gene expression and protein function [11, 12]. In podocytes, excessive lactylation may disrupt cytoskeletal organization and slit-diaphragm protein stability, leading to impaired filtration barrier integrity and the development of proteinuria [5, 34].

Lactate-mediated histone H4K12la has been reported in several cell types, including macrophages [35], microglia [36], and cancer cells [37], where it modulates gene transcription and cellular stress responses. However, whether H4K12la occurs in podocytes has not yet been clarified. Thus, we sought to determine whether elevated lactate contributes to podocyte injury in DKD by inducing H4K12la. Given that SGLT2 inhibition reduces glycolytic flux and lactate production, we further propose that dapagliflozin may protect against podocyte injury in DKD, at least in part, by limiting lactate accumulation and downstream lactylation modifications.

Building on our previous observation that SGLT2 inhibitors may ameliorate proteinuria through modulation of lactate metabolism [38, 39], we administered dapagliflozin at low (4 mg/kg) and high (8 mg/kg) doses to db/db mice for 10 weeks. Dapagliflozin treatment markedly reduced blood glucose, urinary albumin to creatinine ratio (UACR), and renal cortical lactate levels in a dose-dependent manner (Fig. 3a–c). PAS staining showed pronounced mesangial expansion in db/db glomeruli compared with db/m controls, which was substantially attenuated by dapagliflozin (Fig. 3d, e). Ultrastructural analysis further demonstrated that dapagliflozin significantly ameliorated hyperglycemia-induced GBM thickening and podocyte foot process effacement (Fig. 3f–h). Consistently, Western blot analysis revealed that the expression of podocyte-specific markers, podocin and synaptopodin, was markedly restored following dapagliflozin treatment compared with untreated db/db mice, supporting its podocyte-protective effects (Fig. 3i–k).

Fig. 3.

Dapagliflozin reduces renal lactate levels and alleviates podocyte injury in diabetic mice. a Blood glucose. b Urinary albumin excretion (mg/g creatinine). Urine samples were collected in a metabolic cage at 21 weeks of age for each mouse. c Renal lactate concentrations. d, e The glomerular alterations, as revealed by PAS staining in mice (n = 6). Bar = 50 μm. Blue arrows indicate a representative site used for the measurement of mesangial expansion (mesangial matrix index). f Representative transmission electron microscopy images of glomeruli. Blue arrows indicate a representative measurement site for glomerular GBM thickness, while red arrows indicate a representative measurement site for podocyte foot process width (n = 5). Bar = 2 μm. g Quantitative analysis of GBM thickness. h Quantitative analysis of podocyte foot process width. i–k Glomerular podocyte-specific markers expression, including synaptopodin and podocin (n = 3). The main bands at expected molecular weights were quantified; additional weaker bands were not consistently regulated. *p < 0.05, **p < 0.01. DAPA-L: db/db + 4 mg/kg/day, DAPA-H: db/db + 8 mg/kg/day. GBM, glomerular basement membrane.

To further elucidate the mechanism by which dapagliflozin ameliorates podocyte injury in DKD through reducing lactate accumulation, we examined changes in podocyte H4K12la. Immunofluorescence staining using the podocyte nuclear marker Wilms tumor 1 (WT1) demonstrated clear colocalization of H4K12la and WT1 within glomerular podocytes. Compared with db/m controls, db/db mice displayed a reduced number of WT1-positive podocytes along with markedly elevated H4K12la signals (Fig. 4a). We next evaluated nuclear H4K12la levels in dapagliflozin-treated db/db mice using Western blot analysis. Dapagliflozin treatment significantly reversed the high glucose-induced increase in nuclear H4K12la expression, supporting its inhibitory effect on podocyte lactylation (Fig. 4b, c).

Fig. 4.

Dapagliflozin inhibits H4K12la expression in podocytes of diabetic mice and in podocytes stimulated by high glucose. a Representative glomerular immunofluorescence staining of H4k12la. WT-1 was used as a podocyte marker. WT-1, green; H4K12la, red; merge, yellow. Bar = 25 μm (n = 6). b, c The levels of H4K12la were analyzed in the renal cortex of the db/db diabetic mice and their control group using Western blot analysis (n = 3). *p < 0.05, **p < 0.01. DAPA-L: db/db + 4 mg/kg/day, DAPA-H: db/db + 8 mg/kg/day. d LDHA was silenced by LDHA-siRNA. The mRNA expression level of LDHA was determined by RT-qPCR analysis. e Lactate levels in podocytes. f–i The protein levels of synaptopodin and podocin were assessed; the main bands at expected molecular weights were quantified, and additional weaker bands were not consistently regulated. H4K12la bands were clear and specific. Data were from 3 independent experiments. *p < 0.05, **p < 0.01. Normal glucose (NG), 5.3 mmol/L; high glucose (HG), 30 mmol/L; mannitol (MA), 30 mmol/L worked as osmolarity control. HG + DAPA-L: cells were co-incubated with 4 μm dapagliflozin during HG stimulation for 48 h. HG + DAPA-H: cells were co-incubated with 8 μm dapagliflozin during HG stimulation for 48 h. WT-1, Wilms tumor 1; H4K12la, histone H4 lysine lactylation; LDHA, lactate dehydrogenase A.

Previous studies have demonstrated that inhibiting LDHA expression can reduce lactate production and lactate-derived protein lactylation [40, 41]. Therefore, we further employed siRNA to knock down LDHA in order to validate the inhibitory effect of dapagliflozin on lactate-mediated H4K12 lactylation. As shown in Figure 4d–h, the expression of podocyte-specific markers, podocin and synaptopodin, was markedly restored following either dapagliflozin treatment or siLDHA transfection compared with the untreated high-glucose group, accompanied by a pronounced reduction in H4K12la levels.

Collectively, these findings suggest that hyperglycemia-driven lactate accumulation is accompanied by increased podocyte H4K12la and subsequent podocyte injury, which contributes to proteinuria in DKD. Therefore, dapagliflozin may mitigate podocyte injury in DKD, at least in part, by suppressing lactate-mediated H4K12la.

Discussion

Podocyte injury is a central pathological event driving the onset and progression of DKD. Persistent hyperglycemia leads to metabolic, hemodynamic, and inflammatory disturbances that collectively disrupt podocyte structure and function. Podocytes are terminally differentiated epithelial cells that maintain glomerular filtration barrier integrity through their interdigitating foot processes and slit diaphragms. Due to the absence of proliferative capacity, podocyte loss is irreversible and directly contributes to glomerulosclerosis and progressive albuminuria [2, 42].

Persistent hyperglycemia triggers metabolic stress characterized by increased glucose flux, accumulation of advanced glycation end products, and enhanced glycolysis [43]. These metabolic alterations induce mitochondrial dysfunction and excessive reactive oxygen species production, leading to oxidative damage and podocyte apoptosis. Hyperglycemia also activates mTOR signaling and impairs autophagy, which contributes to podocyte hypertrophy, cytoskeletal disorganization, and foot process effacement. In parallel, hemodynamic stress driven by renin-angiotensin-aldosterone system activation further disrupts slit diaphragm integrity. However, the precise molecular mechanisms linking these metabolic and hemodynamic disturbances to podocyte structural and functional deterioration remain to be fully elucidated.

Recent studies suggest that lactate accumulation and lactate-derived histone lysine lactylation act as metabolic-epigenetic regulators in DKD [5, 34]. Elevated intracellular lactate under hyperglycemic conditions may induce aberrant lactylation of chromatin-associated proteins, thereby reprogramming transcriptional networks that govern cytoskeletal organization, inflammatory activation, and cell survival [11]. However, these findings have been predominantly reported in renal tubular cells [4, 5], and evidence in podocytes remains limited. Nevertheless, the lactate/lactylation axis may represent a critical mechanistic link between metabolic reprogramming and podocyte injury [12, 34], highlighting a promising therapeutic target for DKD.

In this study, we found that plasma lactate concentrations were significantly elevated in patients with DKD compared with diabetic patients without kidney involvement and healthy controls. In DKD patients, lactate levels were positively correlated with serum creatinine and albuminuria. Moreover, logistic regression analysis identified hypertension, plasma lactate, WBC count, HbA1c, and serum creatinine as independent risk factors for DKD.

Lactate homeostasis plays a critical role in systemic metabolic equilibrium, with the kidney serving as a major organ for lactate handling and metabolism [44]. Under chronic hyperglycemic conditions, patients with DKD commonly exhibit metabolic reprogramming characterized by impaired glucose utilization, aberrant activation of aerobic glycolysis, and accelerated glucose turnover [45]. Enhanced glycolytic flux consequently increases lactate production, which may further aggravate renal injury [5, 46]. Consistent with these findings, substantial lactate accumulation has been documented in DKD animal models and podocyte/tubular in vitro systems, where it is closely associated with renal dysfunction and proteinuria. Multi-omics analyses further support these observations, linking elevated lactate levels to progressive kidney injury in DKD [12, 47].

SGLT2 inhibitors effectively reduce proteinuria through multiple complementary mechanisms [32]. By increasing sodium delivery to the macula densa, they restore tubuloglomerular feedback and lower intraglomerular pressure, thereby attenuating glomerular hyperfiltration [13, 48, 49]. Additionally, SGLT2 inhibition alleviates oxidative stress, inflammation, and mitochondrial dysfunction in podocytes, preserving slit diaphragm integrity and reducing albumin leakage [33, 49, 50]. However, whether SGLT2 inhibitors affect lactate/lactation modification remains to be further investigated. In this study, to minimize confounding effects from medications known to influence lactate metabolism, patients receiving ACEIs, ARBs, or metformin were excluded. Compared with the non-SGLT2 group, patients treated with SGLT2 inhibitors (dapagliflozin) exhibited a significantly lower incidence of hyperlactatemia. Consistently, plasma lactate levels were reduced in the SGLT2 group, accompanied by lower lactate dehydrogenase and albuminuria. To further assess the association between SGLT2 inhibitor use and hyperlactatemia, we conducted binary logistic regression with hyperlactatemia as the dependent variable. After adjusting for potential confounders, SGLT2 inhibitor use remained independently associated with a reduced risk of hyperlactatemia.

Based on these findings, we speculated that dapagliflozin may alleviate podocyte injury and consequent proteinuria in DKD by reducing lactate accumulation and lactate-derived protein lactylation (e.g., H4K12 lactylation). To test this hypothesis, we conducted experiments in db/db mice and in cultured podocytes exposed to high-glucose conditions. Our results demonstrated that dapagliflozin treatment markedly reduced blood glucose, UACR, and renal cortical lactate levels in a dose-dependent manner. PAS staining revealed pronounced mesangial expansion in db/db glomeruli compared with db/m controls, which was substantially attenuated by dapagliflozin. Ultrastructural examination further confirmed that dapagliflozin significantly ameliorated hyperglycemia-induced GBM thickening and podocyte foot process effacement. Consistent with these morphological improvements, Western blot analysis demonstrated a robust restoration of podocyte-specific markers, including podocin and synaptopodin, following dapagliflozin treatment, supporting its podocyte-protective effects.

Immunofluorescence staining using the podocyte nuclear marker WT1 revealed clear colocalization of H4K12la with WT1-positive podocytes. Compared with db/m controls, db/db mice exhibited a reduced number of WT1-positive podocytes accompanied by markedly elevated H4K12la signals. We next examined nuclear H4K12la expression in vivo and in vitro using Western blotting. Dapagliflozin treatment significantly reversed the high glucose-induced increase in nuclear H4K12la levels both in db/db mouse kidneys and in cultured podocytes, indicating its inhibitory effect on podocyte lactylation.

Collectively, these findings suggest that dapagliflozin ameliorates podocyte injury in DKD at least in part by reducing lactate accumulation and suppressing lactate-derived histone H4K12la. Consistently, our clinical analysis showed that SGLT2 inhibitor use was associated with a lower risk of hyperlactatemia in patients with DKD. However, this study has several limitations. The sample size was relatively small, and the mechanistic evidence remains largely associative rather than causative. In addition, lactate levels and H4K12la protein expression were mainly evaluated in renal cortical tissue, which may not fully resolve cell type-specific alterations within the glomerulus. Although we focused on H4K12 lactylation, other lactylation sites and downstream transcriptional targets were not systematically investigated. Further studies using genetic or pharmacological approaches to specifically modulate lactate production and histone lactylation in podocytes are warranted to validate and extend our findings.

Statement of Ethics

This study protocol was reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou Medical University (Approval No. ES-2024-K091-02, Guangzhou, China). The requirement for written informed consent was waived by the Ethics Committee due to the retrospective nature of the study. All procedures involving human participants were conducted in accordance with the ethical standards of the World Medical Association Declaration of Helsinki. All animal experiments were approved by the Animal Care and Use Committee of the First Affiliated Hospital of Bengbu Medical University (Approval No. 2025735) and were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health.

Conflict of Interest Statement

The authors declare that they have no competing interests.

Funding Sources

This study was financially supported by the Guangdong Basic and Applied Basic Research Foundation (Nos. 2025A1515010471 and 2023A1515012474), the Natural Science Project of Bengbu Medical University (No. 2024byzd069), the Chengdu Medical Research Project (No. 202303053935), and the Guangzhou City Science and Technology Project (No. 2023A03J0344). The funding sources were used to support the experimental work of this study. The funding bodies had no role in study design, data collection, data analysis, data interpretation, manuscript preparation, or the decision to publish.

Author Contributions

Caixia Cui: writing – original draft, validation, and formal analysis. Ningying Fu and Shaoguo Tao: methodology, investigation, and validation. Jiasi Chen and Yuqi Shang: project administration and conceptualization. Xuan Peng and Qian Zhao: software and data curation. Shanshan Wen: investigation and data curation. Haiyan Kang: visualization. Hongbo Peng: validation and software. Jie Xiao: writing – review and editing. Yonghua Peng: writing – review and editing, supervision, and resources. Guibao Ke: writing – review and editing, investigation, and conceptualization. All authors reviewed the manuscript.

Funding Statement

This study was financially supported by the Guangdong Basic and Applied Basic Research Foundation (Nos. 2025A1515010471 and 2023A1515012474), the Natural Science Project of Bengbu Medical University (No. 2024byzd069), the Chengdu Medical Research Project (No. 202303053935), and the Guangzhou City Science and Technology Project (No. 2023A03J0344). The funding sources were used to support the experimental work of this study. The funding bodies had no role in study design, data collection, data analysis, data interpretation, manuscript preparation, or the decision to publish.

Data Availability Statement

The data that support the findings of this study are not publicly available due to privacy reasons but are available from the corresponding author upon request.