Acetate ameliorates membranous nephropathy by modulating the immune response and podocyte function

Abstract

Background

Membranous nephropathy (MN) is an antibody-mediated podocytopathy in which subepithelial immune deposits and complement activation lead to podocyte injury and proteinuria. Short-chain fatty acids (SCFAs) are microbiota-derived metabolites with immunomodulatory properties, but their therapeutic potential and mechanisms in MN remain incompletely defined.

Materials and methods

We evaluated acetate, propionate, and butyrate in a passive Heymann nephritis (PHN) rat model (n = 6/group). Rats received 150 mM sodium acetate, propionate, or butyrate in drinking water from day 0 to day 42. Proteinuria and biochemical indices were monitored; glomerular immune deposition and ultrastructure were assessed by immunofluorescence and transmission electron microscopy. Antigen-specific antibodies were quantified by ELISA. Glomerular expression of synaptopodin and WT1 was detected by immunohistochemistry. Splenic CD4⁺ T-cell subsets were analyzed by flow cytometry. In vitro, conditionally immortalized human podocytes were exposed to plasma from patients with primary MN, with or without acetate and the GPR43 antagonist GLPG0974, migration, synaptopodin expression, reactive oxygen species (ROS), apoptosis, and viability were assessed, and SCFA receptor expression was examined.

Results

All three SCFAs significantly reduced proteinuria in PHN rats compared with disease controls and ameliorated ultrastructural injury, including GBM thickening and podocyte foot-process effacement. Acetate and propionate reduced subepithelial immune-complex deposits. Acetate selectively decreased antigen-specific anti-sheep IgG responses without altering total IgG. PHN was associated with decreased Tregs and increased Th1/Th17 polarization; acetate increased the Treg proportion in splenocytes. Acetate protected podocyte cytoskeletal integrity and viability in vivo. Acetate acted through GPR43, as the GPR43 antagonist GLPG0974 partially or completely blocked acetate’s effects on ROS reduction, migration improvement, and synaptopodin mRNA upregulation. Acetate treatment reduced HDAC activity in both renal cortex and podocytes while increasing HAT activity. Acetate mitigated primary MN plasma–induced dysfunction by improving migration and synaptopodin signal, reducing ROS, limiting apoptosis, and restoring cell viability.

Conclusion

Acetate ameliorates experimental MN through coordinated immunomodulatory effects and amelioration of podocyte foot process effacement. These findings support targeting acetate pathways as a mechanistically plausible adjunct strategy with low potential toxicity for MN.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12882-026-05175-7.

Introduction

Primary membranous nephropathy (MN) is a major cause of adult nephrotic syndrome and a leading immune-mediated glomerular disease worldwide. It is characterized by subepithelial immune-complex deposition along the glomerular basement membrane (GBM), complement activation, and podocyte injury, resulting in proteinuria and progressive loss of kidney function in a subset of patients [1]. In most cases, MN is an organ-specific autoimmune disease driven by circulating autoantibodies against podocyte antigens. The most prominent target antigen is the M-type phospholipase A2 receptor (PLA2R) [1, 2]. Current management integrates optimized supportive care with immunosuppression for patients at moderate to high risk of progression, as recommended by the KDIGO 2021 guideline [3]. Therapeutic options include B-cell depletion with rituximab, calcineurin inhibitors, and glucocorticoid–cyclophosphamide–based regimens. Although rituximab can induce durable remissions and has demonstrated efficacy in randomized trials, a substantial proportion of patients remain refractory or relapse, and broad immunosuppression is associated with infections and other toxicities [4, 5]. These unmet needs motivate the search for adjunctive or alternative approaches that can rebalance immune responses while preserving podocyte integrity.

An expanding body of work implicates the gut–kidney axis in immune and metabolic regulation relevant to kidney disease [6]. In chronic kidney disease, gut dysbiosis is associated with reduced production of beneficial microbial metabolites, including short-chain fatty acids (SCFAs) [7]. Importantly, dysbiosis has also been reported in primary MN, with altered microbial composition and decreased fecal organic acids/SCFAs in patient cohorts, suggesting that impaired SCFA generation may accompany—or contribute to—disease activity [8, 9].

SCFAs—mainly acetate, propionate, and butyrate—are produced by bacterial fermentation of dietary fiber and signal through multiple mechanisms. In addition to serving as energy substrates, SCFAs engage metabolite-sensing receptors (including FFAR2/GPR43 and FFAR3/GPR41, and related receptors expressed in renal and immune compartments) and act as histone deacetylase (HDAC) inhibitors to shape transcriptional programs in immune cells [6, 10]. These pathways influence dendritic cell maturation, effector T-cell polarization, and regulatory T-cell (Treg) differentiation and function, promoting IL-10 and Foxp3-dependent immune tolerance in appropriate contexts [11]. Given that MN is sustained by antibody-producing B cells and T-cell–dependent immune dysregulation [12], SCFAs represent an attractive, mechanistically grounded strategy to modulate pathogenic immunity without broadly suppressing host defenses.

Beyond systemic immune effects, SCFAs may directly protect glomerular cells. In experimental proteinuric kidney disease, SCFA improved proteinuria and preserved podocyte structure through combined epigenetic and GPR109A-dependent mechanisms, and GPR109A was reported to be highly expressed on podocytes [13]. These findings raise the possibility that acetate supplementation could benefit MN by simultaneously dampening pathogenic immune responses and enhancing podocyte resilience.

In this study, we investigated whether SCFAs ameliorate MN by modulating immune responses, and specifically whether acetate affects podocyte function. Using a passive Heymann nephritis model and complementary in vitro assays, we evaluated effects of SCFAs on proteinuria, histopathology, humoral immune activity, and inflammatory pathways, and tested direct actions of acetate on podocyte cytoskeletal integrity and injury signaling. These findings support targeting acetate pathways—via dietary or microbiota-based strategies—as a potential adjunctive approach for MN.

Materials and methods

Induction of passive Heymann nephritis and SCFA treatment in rats

Animals and disease induction

Five-week-old Sprague–Dawley rats were randomly assigned to five groups (n = 6 per group). Healthy control group: Received a single tail vein injection of normal saline (0.4 mL/100 g body weight) and received drinking water with 150 mM sodium chloride added to match the sodium concentration in SCFA-treated groups [11]. Disease control group (PHN): Received a single tail vein injection of sheep anti-rat Fx1A serum (PTX-002 S, Probetex, Inc., USA) (0.4 mL/100 g body weight) and received sodium-matched drinking water. Acetate-treatment, propionate-treatment and butyrate-treatment group: Received sheep anti-rat Fx1A antiserum as above and received 150 mM sodium acetate, propionate or butyrate (S8750, P1880, 303410; all from Sigma‒Aldrich, St. Louis, MO, USA) in drinking water ad libitum from day 0 to day 42. Day 0 is defined as the day of immunization. The rat dose was approximately 6 mmol/kg/day (Supplementary Table 3).

The Sprague-Dawley rats used in this study were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. For passive Heymann nephritis, 24-hour urine samples were collected in metabolic cages, and blood was obtained by angular venipuncture before immunization and weekly thereafter. Rats were anesthetized by intraperitoneal injection of 5% pentobarbital sodium (1 mL/kg body weight).

After the rats lost consciousness, we opened the abdominal cavity and collected blood from the abdominal aorta using a venous puncture needle. The rats were subsequently euthanized via exsanguination. All rats were euthanized at Day 42 after immunization, and kidney tissues, spleens, and blood samples were harvested.

All animal procedures were performed in accordance with institutional guidelines for the care and use of laboratory animals and were approved by the Experimental Animal Ethics Committee of Peking University First Hospital (Beijing, China).

Assessment of kidney injury

Urinary protein, serum creatinine, and blood urea nitrogen were measured using an automatic biochemical analyzer (Unicel DxC 600 Synchron, Beckman Coulter, USA). Renal cortical tissue samples were dissected into approximate 1 mm³ pieces and fixed in glutaraldehyde-based electron microscopy fixative for 2 h at room temperature. Ultrathin sections were examined by transmission electron microscopy. For each tissue specimen, 20 representative fields were randomly captured at ×13,000 magnification. Average number of electron-dense deposits beneath the basement membrane: Measure the total length along the basement membrane and count the number of electron-dense deposits attached to this segment of basement membrane. Calculate as number/1 µm of basement membrane. The thickness of GBM was measured at regular intervals along the GBM to avoid tilting the slice area. The width of the basement membrane was measured at regular intervals for each image, and at 20 points of thickness per image were measured for each image to obtain the average GBM thickness in micrometers. Foot process width is the width of a single foot process measured along the glomerular basement membrane. The foot process width of all glomeruli was measured for each image. The average width of each animal was calculated. The results are expressed in micrometers. For immunofluorescence staining, frozen kidney Sect.  (5 μm) were fixed in cold acetone and incubated with fluorescein isothiocyanate (FITC)-conjugated goat anti-rat IgG antibody (1:50; Abcam, Cambridge, MA, USA) and Alexa Fluor 488-conjugated rabbit anti-sheep antibody (1:150). ImageJ was used to measure the average fluorescence intensity of least 30 randomly selected glomerular cross-section normalized to the glomerular area.

Enzyme-linked immunosorbent assays

Serum antibodies against Anti-Rat Fx1A Serum were quantified by ELISA. Anti-Rat Fx1A Serum diluted 1:300 in carbonate buffer (pH = 9.6) was used to coat 96-well plates overnight. Serum samples were diluted 1:100 in PBST (PBS with 0.1% Tween-20) and incubated at 37 °C for 1 h. Rabbit anti-rat IgG conjugated to alkaline phosphatase (1:5000; A6066, Sigma) was added and incubated at 37 °C for 30 min. p-Nitrophenyl phosphate (1 mg/mL; 4876, Sigma) dissolved in substrate buffer (1 mol/L diethanolamine, 0.5 mol/L MgCl₂, pH = 9.8) was used for color development, and absorbance was read at 405 nm (Bio-Rad, Tokyo, Japan).

Human podocyte culture and in vitro stimulation

Conditionally immortalized human podocytes (kindly provided by Prof. Jochen Reiser, Rush University Medical Center, Chicago, IL, USA) were used in this study. The cells were cultured at 33℃ under permissible conditions using RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin and 100 µg/mL streptomycin to promote cell proliferation. To induce differentiation, cells were transferred to 37 ℃ and cultured for 10–14 days [14]. Differentiated podocytes in six-well plates were synchronized in RPMI 1640 for 6 h and then stimulated with RPMI 1640 containing: (1) 5% plasma from MN patients [15] plus saline, (2) 5% MN plasma supplemented with 100 µM acetate [11, 16, 17], (3) 5% plasma from MN patients + 100 µM acetate + 0.1 µM GLPG0974 (GLPG), antagonist of GPR43 (HY-12940, MCE) or 5% plasma from MN patients + 100 µM acetate + 1 µM Trichostatin A (TSA, HY-15144, MCE), (4) 5% plasma from healthy controls. All media were filtered through 0.22 μm filters.

Scratch-wound migration assay

Among the three SCFAs (acetate, propionate, and butyrate) administered in vivo, acetate showed the most consistent protective effects on renal function and histopathology. Therefore, acetate was selected as the representative SCFA for subsequent in vitro mechanistic studies. Podocytes were treated with 5% MN plasma and control saline, 5% MN plasma and 100 µM acetate, 5% healthy plasma and control saline in each group. A single cross-shaped scratch was made in each well using a sterile 200-µl pipette tip. After stimulation as described above, wound areas were imaged by light microscope at 0 and 8 h and 16 h and 24 h. Ten positions along each scratch were randomly selected and the wound width was measured at each time point. Migration distance at time x was calculated as the mean reduction in scratch width from baseline (0 h) to time x across the 10 positions.

Immunofluorescence staining of SCFA receptors and Synaptopodin

Immunofluorescence colocalization was performed to assess the expression of the SCFA receptors GPR41 (PA5-75521, Thermo Fisher Scientific), GPR43 (bs-13536R, Bioss), GPR109A (DF4890, Affinity Biosciences, China) and Synaptopodin (ab224491, Abcam) in podocytes. Cells were fixed with 4% paraformaldehyde and blocked with 3% BSA. Rabbit anti-human GPR41, GPR43 and GPR109A polyclonal antibodies (each 1:200) and Rabbit anti-human Synaptopodin monoclonal antibody (1:200) were applied overnight at 4 °C. After washing, goat anti-rabbit IgG (H + L) secondary antibodies were incubated in the dark at 37 °C for 30 min. Nuclei were counterstained with DAPI. Images were acquired using a fluorescence microscope (Leica, Germany). Fluorescence intensity was quantified using ImageJ software in at least 20 randomly selected fields per condition. The integrated density was normalized to the number of nuclei (DAPI) to obtain the mean fluorescence intensity per cell.

Immunohistochemistry of rat renal sections

Formaldehyde-fixed paraffin-embedded renal sections of rats were dewaxed in xylene ethanol and rehydrated through graded ethanol. Antigen retrieval was performed by heating the slide tris EDTA buffer (0.01 M, pH = 9.0) for 4 min. The immunostainings of Synaptopodin (PSH10-45, HuABIO) and Wilms tumor 1 (SC06-41, HuABIO). The primary antibodies were detected with the rabbit secondary antibodies (PV6001, ZSGB-Bio). Hematoxylin staining was performed after immunohistochemical staining. Stained samples were examined by ImageJ (Fiji Is Just ImageJ, version 2.17.0); as the mean optical density (Integrated optical density/Relative area.). All the and at least 20 glomeruli per renal section at × 400 were observed blindly for quantitative assessments of immunohistochemical staining.

Intracellular reactive oxygen species measurements

After 24 h of stimulation, podocytes were harvested by trypsinization and collected by centrifugation. Cells were resuspended in serum-free medium containing DCFH-DA (1:1000, D6883, Sigma-Aldrich) and incubated at 37 °C for 20 min. Intracellular ROS was quantified by flow cytometry using the FITC channel.

Flow-cytometric assessment of podocyte apoptosis

Stimulated podocytes were harvested by trypsinization. Aliquots (1 × 10⁵ cells in 100 µL per tube) were stained with 5 µL PE-conjugated Annexin V and 5 µL 7-AAD (FXP027-100, 4 A BIOTECH, China) for 15 min at room temperature in the dark. After adding 400 µL of 1× Annexin V binding buffer, samples were analyzed by flow cytometry within 1 h.

CCK-8 assay

Podocytes were seeded in 96-well plates. After 24 h of stimulation, 10 µL of CCK-8 reagent (FXP132-100, 4 A BIOTECH, China) was added to each well and incubated at 37 °C for 1–4 h. Absorbance was measured at 450 nm using a microplate reader. Cell viability (%) = [(As-Ab)/(Ac-Ab)] × 100%; Inhibition rate (%) = [(Ac-As)/(Ac-Ab)] × 100%, where “As” is the absorbance of the test well, “Ac” is the absorbance of the control well, and “Ab” is the absorbance of the blank well.

Flow cytometry of rat splenic lymphocyte subsets

Splenocytes were isolated from rat spleens. After red blood cell lysis, cells were resuspended in L500 lymphocyte culture medium and stained for flow cytometry. Th1 cells were identified using anti-CD3 (201403, Biolegend), anti-CD4 (201520, Biolegend), and anti-IFN-γ antibodies (507810, Biolegend); Th17 cells using anti-CD3, anti-CD4, and anti-IL-17 A antibodies (506914‌, Biolegend); and Tregs using anti-CD4, anti-CD25, and anti-Foxp3 antibodies (320012, ‌Biolegend).

Quantitative real-time PCR

Total RNA was extracted from podocytes using TRIzol (15596018, Thermo Fisher Scientific). cDNA was synthesized using the High-Capacity cDNA Reverse Transcription Kit (4368814, Thermo Fisher Scientific). Quantitative PCR was performed in triplicate using PowerUp™ SYBR Green Master Mix system (A25742, Thermo Fisher Scientific). mRNA levels of Synaptopodin, NF-κB, PLA2R, PPAR-γ, IL-6, IL-10 and TNF-α were calculated using the 2^−ΔΔCT method with β-actin as the reference gene. Primers sequences are provided in Supplementary Table 1.

Histone Acetyltransferase (HAT) and HDAC activity detection

After differentiation of podocytes, stimulate with 5% MN patient plasma. After treatment with 100 µM acetate or 1 µM Trichostatin A (TSA) and for 24, 48–72 h, cells lysates were collected. HDAC activity in renal cortical tissue protein extracts and podocytes was detected with colorimetric HDAC Activity Assay Kit (ab1432, Abcam, UK) according to manufacturer’s protocol. The effects of 1 µM TSA on HDAC activity were also measured after treatment. The absorbance of the chromophore was measured at 405 nm using a microplate reader for the quantification of relative HDAC activity. HAT enzymatic activity in podocyte protein extracts was determined using a Histone Acetyltransferase Activity Detection Kit (EPI001, Sigma-Aldrich, USA) in accordance with the manufacturer’s standard protocol. After sufficient incubation, the absorbance value of each reaction system was detected at 440 nm using a microplate spectrophotometer for quantitative analysis.

Statistical analysis

Statistical analyses were performed using SPSS version 13.0 (SPSS Inc., Chicago, IL). Data distribution was assessed using the Shapiro–Wilk test. Normally distributed data are presented as mean ± SEM, whereas non-normally distributed data are presented as median (interquartile range). For comparisons between two groups, the Student’s t test or Mann–Whitney U test was used, as appropriate. For comparisons among multiple groups, one-way ANOVA or the Kruskal–Wallis test was used for continuous variables, and the chi-square test or Fisher’s exact test was used for categorical variables. All tests were two-sided, and P < 0.05 was considered statistically significant.

Results

Validation of the PHN rat model and baseline IgG deposition

To establish the passive Heymann nephritis (PHN) model, rats in the disease-control and treatment groups received the same dose of Anti-Rat Fx1A Serum, whereas healthy controls received an equivalent volume of normal saline.

Immunofluorescence staining confirmed robust glomerular deposition of sheep IgG in all PHN groups. The intensity of sheep IgG deposition did not differ among the disease-control, acetate-, propionate-, and butyrate-treatment groups, indicating comparable heterologous antibody binding across PHN groups. PHN induction elicited a host anti-sheep IgG response, which contributes to immune complexes formation. Quantification of glomerular sheep IgG fluorescence intensity demonstrated that anti-sheep IgG levels were significantly elevated in disease-control group and in all SCFA-treated groups compared with healthy controls throughout follow-up (Fig. 1K).

Fig. 1.

Heterologous antibody deposition in glomeruli and host anti-sheep IgG responses in the PHN model. (A–E) Representative immunofluorescence images showing glomerular deposition of anti-Fx1A sheep IgG in rats from each group after induction of passive Heymann nephritis (PHN). (F–J) Representative immunofluorescence images showing glomerular deposition of rat IgG in each group after PHN induction. (K) Quantification of glomerular anti-Fx1A sheep IgG fluorescence intensity. (L) Quantification of glomerular rat IgG fluorescence intensity. (M) Time-course of circulating anti-sheep anti-Fx1A serum IgG levels from week 1 to week 6. The animals received the treatment daily through drinking water ad libitum from day 0 and were euthanized on day 42 post-modeling. * shows the p-value of acetate treatment group and disease control. Data are presented as mean ± SEM, N = 6. (*P < 0.05, **P < 0.01, ***P < 0.001)

When constructing the model, the four groups of rats received comparable amounts of antibody serum, and the levels of anti-sheep IgG that effectively entered the rats and deposited in the renal glomeruli showed no statistically significant differences, indicating consistent severity of the disease model across the groups. In the disease control group, granular deposition of rat IgG along the capillary walls was observed in the rat kidneys, and its intensity was significantly higher than that in the negative control group (2.60 ± 0.30 vs. 10.58 ± 1.28, P < 0.001) (Fig. 1L). However, the deposition of rat IgG was decreased in the acetate- and butyrate- treatment groups compared with disease control group (4.44 ± 0.61 vs. 10.58 ± 1.28, P < 0.05; 5.92 ± 0.48 vs. 10.58 ± 1.28, P < 0.05) (Fig. 1L).

SCFAs reduce circulating anti-Fx1A serum antibody levels

In the disease-control group, anti-Fx1A serum antibody levels rose progressively, peaking at week 5 (3.04 ± 0.14 vs. healthy control: 0.63 ± 0.10, P < 0.05) (Fig. 1M), and remaining stable at week 6. Acetate significantly reduced anti-Fx1A serum antibody levels at week 5 compared with disease controls (2.19 ± 0.63 vs. 3.04 ± 0.14, P < 0.05), whereas propionate and butyrate showed non-significant downward trend (Fig. 1M).

SCFAs effectively attenuate renal injury in PHN rats

PHN rats developed proteinuria beginning in week 2, which progressed over time. By week 6, urinary protein excretion in the disease-control group was significantly higher than that in healthy controls (urinary protein-to-creatinine ratio [UPCR]: 2.66 ± 2.13 vs. 0.20 ± 0.03, P < 0.001; Fig. 2D). Serum creatinine and blood urea nitrogen (BUN) remained within the normal range and were comparable between the disease-control and healthy-control group (all P > 0.05, Fig. 2E, F).

Fig. 2.

SCFAs attenuate proteinuria and ultrastructural glomerular injury in the PHN rat model. (A) Time course of UPCR in each group. SCFA treatment (initiated at week 0) was associated with persistently lower urinary protein levels compared with the disease control group. (B–C) Longitudinal changes in serum creatinine and blood urea nitrogen (BUN) in each group. (D) UPCR at week 6 across groups. (E–F) Serum creatinine and BUN at week 6. (G–K) Representative transmission electron microscopy images of glomeruli from each group, showing subepithelial electron-dense deposits, podocyte foot process effacement, and glomerular basement membrane (GBM) thickening in disease controls, with attenuation in SCFA-treated rats. (L) Quantification of mean podocyte foot process width. (M) Quantification of subepithelial electron-dense deposits (number per high-power field). (N) Quantification of GBM thickness. Data are presented as mean ± SEM, (N = 6, *P < 0.05, **P < 0.01, ***P < 0.001). Statistical analyses were performed as described in the Methods

SCFA administration significantly reduced proteinuria in PHN rats. Compared with the disease-control group, UPCR was lower in the acetate (0.75 ± 0.28 vs. 2.66 ± 2.13; P < 0.05), propionate (2.24 ± 1.47 vs. 2.66 ± 2.13; P < 0.05), and butyrate group (1.25 ± 0.98 vs. 2.66 ± 2.13; P < 0.05) (Fig. 2D).

SCFAs attenuate glomerular ultrastructural injury in PHN rats

Consistent with the ultrastructural features of human MN, PHN rats exhibited subepithelial immune complex deposits, podocyte foot process effacement, and GBM thickening. These parameters were quantified by transmission electron microscopy (Fig. 2G-N).

Subepithelial immune complex deposits. The burden of deposits was substantially increased in the disease-control group compared with healthy controls (0.61 ± 0.07 vs. 0.01 ± 0.01, P < 0.01). Acetate, propionate and butyrate significantly reduced deposits (acetate: 0.30 ± 0.07 vs. 0.61 ± 0.07, P < 0.01; propionate: 0.42 ± 0.10 vs. 0.61 ± 0.07, P < 0.05; butyrate: 0.46 ± 0.09 vs. 0.61 ± 0.07, P < 0.05). (Fig. 2M).

Podocyte foot process effacement. Mean foot process width was increased in the disease-control group compared with healthy controls (0.74 ± 0.33 vs. 0.34 ± 0.02 μm, P < 0.05). Foot process width was significantly reduced by acetate and butyrate (acetate: 0.41 ± 0.06 vs. 0.74 ± 0.33 μm, P < 0.05; butyrate: 0.49 ± 0.04 vs. 0.74 ± 0.33 μm, P < 0.05), whereas propionate produced a non-significant numerical decrease (0.49 ± 0.04 vs. 0.74 ± 0.33 μm, P > 0.05) (Fig. 2L).

GBM thickening. GBM thickness was markedly increased in the disease-control group compared with healthy controls (0.52 ± 0.12 vs. 0.17 ± 0.02 μm, P < 0.01). All three SCFAs significantly attenuated GBM thickening (acetate: 0.26 ± 0.02; propionate: 0.35 ± 0.04; butyrate: 0.35 ± 0.05; each vs. 0.56 ± 0.11 μm, P < 0.01) (Fig. 2N).

T-cell differentiation is skewed toward a proinflammatory phenotype in PHN and partially restored by acetate

Flow cytometry of splenocytes revealed a reduction in regulatory T cells (Tregs; CD3⁺CD4⁺CD25⁺Foxp3⁺) among CD3⁺CD4⁺ T cells in the disease-control group compared with healthy controls (9.1 ± 1.2% vs. 11.1 ± 0.8%, P < 0.01, Fig. 3A). In parallel, Th17 (1.7 ± 0.5% vs. 0.9 ± 0.2%, P < 0.01, Fig. 3C) and Th1 frequencies (9.7 ± 1.8% vs. 5.8 ± 1.4%, P < 0.01, Fig. 3B) were increased in the disease-control group.

Fig. 3.

SCFAs modulate splenic CD4⁺ T-cell subsets in PHN rats. Flow cytometry was used to quantify the frequencies of regulatory T cells (Tregs) (A), Th1 (B), and Th17 (C) among total CD4⁺ T cells in splenocytes. The Th17/Treg (D) and Th1/Treg (E) ratios were also calculated. Compared with disease controls, SCFA treatment decreased the proportion of Th17 cells (A), increased the proportion of Tregs (C), and decreased the Th17/Treg ratio (D). Data are presented as mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). Statistical analyses were performed as described in the Methods

Acetate significantly increased the Treg proportion compared with disease controls (11.7 ± 1.0% vs. 9.1 ± 1.2%, P < 0.01) (Fig. 3A). The ratio of Th1/Treg cells and the ratio of Th17/Treg cells decreased in the acetate treatment group slightly compared with disease control, but showed no significant differences (P > 0,05) (Fig. 3D and E).

SCFAs protect podocytes by protecting cytoskeleton in PHN rats

As shown in Fig. 4, synaptopodin expression was significantly lower in the Disease Control group compared with the Healthy Control group (0.123 ± 0.004 vs. 0.170 ± 0.008, P < 0.01). In contrast, Acetate and propionate treatment significantly increased synaptopodin expression compared to the Disease Control group (Acetate: 0.171 ± 0.008 vs. 0.123 ± 0.004, P < 0.01, Propionate: 0.160 ± 0.019 vs. 0.123 ± 0.004, P < 0.01). no significant difference was observed between the butyrate treatment group and the Disease Control group (P > 0.05) (Fig. 4K).

Fig. 4.

SCFAs protect synaptopodin and WT1 in the glomeruli of PHN rats model. Immunohistochemical staining of representative glomeruli of synaptopodin (A-E) and WT-1 (F-J) in each group of rats. (K) The mean optical density of synaptopodin in SCFA treatment group, healthy control and disease control group, respectively. (L) The mean optical density of WT-1 in SCFA treatment group, healthy control and disease control group, respectively. Data are presented as mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). Statistical analyses were performed as described in the Methods

WT1 expression was significantly lower in the Disease Control group compared with the Healthy Control group (0.062 ± 0.001 vs. 0.071 ± 0.003, P < 0.05). In contrast, Acetate treatment significantly increased WT1 expression compared to the Disease Control group (0.069 ± 0.001 vs. 0.062 ± 0.001, P < 0.01). However, no significant difference was observed between the propionate/butyrate treatment group and the Disease Control group (P > 0.05) (Fig. 4L).

Acetate mitigates MN plasma–induced podocyte dysfunction

Consistent with cytoskeletal injury, synaptopodin signal intensity decreased after 48 h of MN plasma exposure compared with healthy controls (6.25 ± 0.32 vs. 7.58 ± 1.19, P < 0.05). Acetate co-treatment restored synaptopodin signal (7.98 ± 0.87 vs. 6.25 ± 0.32, P < 0.01) (Fig. 5B).

Fig. 5.

Acetate preserves synaptopodin expression and improves podocyte viability under injurious stimulation. (A)Representative immunofluorescence images of synaptopodin in differentiated podocytes exposed to the indicated stimuli. (B) Quantification of synaptopodin fluorescence intensity across groups. (C) Podocyte metabolic activity/viability after 24 h stimulation, assessed by a cell activity assay. (D) Representative flow-cytometry gating strategy used to define the live-cell population. (E) Quantification of podocyte apoptosis by flow cytometry, shown as the percentage of live cells under each condition. Data are presented as mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). Statistical analyses were performed as described in the Methods

CCK8 assay further confirmed decreased cell viability after 24 h of MN plasma stimulation (0.84 ± 0.11 vs. 1.50 ± 0.32, P < 0.001). Acetate increased viability in a dose-dependent manner at 1, 10, 50, and 100 µM (1.12 ± 0.20, P < 0.05; 1.22 ± 0.13, P < 0.001; 1.30 ± 0.08, P < 0.01; and 1.23 ± 0.16, P < 0.001, each vs. 0.84 ± 0.11 MN plasma alone, respectively) (Fig. 5C).

Annexin V/7-AAD staining showed that MN plasma reduced the proportion of live podocytes at 48 h compared with healthy controls (16.8 ± 1.44 vs. 21.4 ± 1.55%, P < 0.01), whereas acetate increased podocyte viability (21.6 ± 2.95% vs. 16.8 ± 1.44%, P < 0.05) (Fig. 5E).

Expression of SCFA receptors on human podocytes

Immunofluorescence staining demonstrated expression of the SCFA receptors GPR41, GPR43, and GPR109A on differentiated human podocytes, supporting the possibility of direct SCFA signaling in podocytes (Supplementary Fig. 1).

Acetate activates GPR43 to protect podocyte cytoskeleton

Acetate is more selective for GPR43 [18], to determine whether GPR43 is involved in acetate protection of podocytes, we compared the effects of acetate alone or in combination with GPR43 antagonists.

Reactive oxygen species (ROS) levels increased following 48 h of MN plasma exposure (7.85 ± 0.35 vs. 3.71 ± 0.24, P < 0.01). Acetate treatment reduced ROS levels (5.47 ± 0.10 vs. 7.85 ± 0.35, P < 0.01). However, the addition of GLPG partially blocked this effect (6.80 ± 0.20 vs. 5.47 ± 0.10, P < 0.05), though ROS levels remained significantly higher than those in healthy controls (P < 0.01) (Fig. 6A).

Fig. 6.

The effects of acetate alone or in combination with GLPG in inhibit HDAC and protect podocyte cytoskeleton. Podocytes were treated with 5% MN plasma and control saline, 5% MN plasma and 100 µM acetate, 5% MN plasma + 100 µM acetate + 0.1 µM GLPG0974, 5% healthy plasma and control saline in each group. (A) Intracellular reactive oxygen species (ROS) levels measured under the indicated conditions. (B) Acetate improves podocyte migration under MN plasma stimulation. Scratch-wound migration assay showing podocyte wound closure under indicated stimuli. Time-course plot summarizing migration distance at 8, 16, and 24 h. (C-E) Acetate promotes a protective transcriptional program in podocytes exposed to primary MN plasma. Podocyte cells were treated After 48 h, mRNA expression was quantified by RT–qPCR. Acetate increased transcripts encoding synaptopodin (C), WT-1 (E), while reducing expression of NF-κB (D). (F) HDAC activity in renal cortex of healthy group acetate treatment group and disease control group rats. (G-H) Acetate, and TSA inhibited HDACs activity in podocytes. (H) HDAC activity were inhibited in podocytes by treatment of acetate. The results were normalized using the control as 100%. (G) The OD value of HAT activity in podocytes by treatment of acetate. Data are presented as mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). Statistical analyses were performed as described in the Methods

Acetate regulates podocyte migration ability through GPR43

Podocyte migration was impaired by stimulation with plasma from patients with MN (Supplementary Fig. 5). At 16 h, migration distance was significantly reduced in the MN plasma group compared with healthy controls (75.9 ± 10.8 vs. 311.4 ± 22.2 μm, P < 0.001). Co-treatment with 100 µM acetate significantly improved migration compared with MN plasma alone (123.2 ± 5.4 vs. 75.9 ± 10.8 μm, P < 0.001). The addition of GLPG abolished this protective effect, as the 5% MN+Acetate+GLPG group remained significantly lower than the 5% MN+Acetate group (76.3 ± 3.6 vs. 123.2 ± 5.4 μm, P < 0.001) (Fig. 6B). At 24 h, MN plasma significantly reduced migration distance relative to healthy controls (140.6 ± 15.3 vs. 399.2 ± 28.8 μm, P < 0.001). Acetate treatment significantly increased migration compared with MN plasma alone (302.6 ± 12.2 vs. 140.6 ± 15.3 μm, P < 0.001). The GLPG-treated group showed intermediate migration distance, which was significantly lower than the acetate treatment group (213.1 ± 15.8 vs. 302.6 ± 12.2 μm, P < 0.001) (Fig. 6B).

Acetate alters podocyte cytoskeleton proteins and cytokines at the mRNA level

After 48 h of plasma stimulation in MN patients, the mRNA expression of synaptopodin, a cytoskeletal protein in podocytes, was significantly downregulated compared with the healthy control group (0.0105 ± 0.0015 vs. 0.0244 ± 0.0020, P < 0.001). Acetate treatment significantly upregulated synaptopodin expression compared with the disease group (0.0169 ± 0.0017 vs. 0.0105 ± 0.0015, P < 0.01). The 5% MN+Acetate+GLPG group showed no significant difference compared to the disease group (P > 0.05), and remained significantly lower than the acetate treatment group (0.0105 ± 0.0009 vs. 0.0169 ± 0.0017, P < 0.01) (Fig. 6C).

After 48 h of plasma stimulation in MN patients, acetate treatment significantly upregulated WT-1 mRNA expression compared with the disease group (0.00340 ± 0.00051 vs. 0.00032 ± 0.00012, P < 0.01), restoring expression to levels comparable to those in healthy controls (0.00340 ± 0.00051 vs. 0.00365 ± 0.00039, P > 0.05) (Fig. 6E).

NF-κB mRNA expression was significantly upregulated compared with healthy controls (0.473 ± 0.030 vs. 0.079 ± 0.010, P < 0.001). Acetate treatment significantly reduced NF-κB expression compared with the disease group (0.244 ± 0.022 vs. 0.473 ± 0.030, P < 0.05). The addition of GLPG completely blocked this effect, as the 5% MN+Acetate+GLPG group showed no significant difference from the disease group (P > 0.05) and remained significantly higher than the acetate treatment group (0.489 ± 0.027 vs. 0.244 ± 0.022, P < 0.01) (Fig. 6D).

MN plasma increased the mRNA expression level of PLA2R of podocytes at 48 h compared with healthy controls (0.116 ± 0.028 vs. 0.008 ± 0.003, P < 0.01). The expression level of PLA2R protein after acetate treatment decreased compared with the MN disease group (0.041 ± 0.014 vs. 0.116 ± 0.028, P < 0.05) (Supplementary Fig. 6C). Acetate increased transcripts encoding PPAR-γ (P > 0.05) and IL-10 (0.002 ± 0.001 vs. 0.021 ± 0.016, P < 0.05), while reducing expression of, IL-6 (P > 0.05) and TNF-α mRNA (0.146 ± 0.089 vs. 0.507 ± 0.302, P < 0.001) (Supplementary Fig. 6B-G).

Acetate as HDAC inhibitor regulates podocytes

HDAC activity in the renal cortex was significantly reduced by acetate treatment compared with the disease control group (0.305 ± 0.017 vs. 0.405 ± 0.022 a.u., P < 0.01), returning to levels comparable to those in healthy controls (P > 0.05) (Fig. 6F).

Acetate and TSA treatment significantly inhibited HDAC activity in podocytes compared with controls (normalized to 100%) at 24 h (66.92 ± 2.98% and 43.85 ± 1.49%, P < 0.001), 48 h (54.23 ± 2.44% and 39.33 ± 0.99%, P < 0.001), and 72 h (40.38 ± 2.30% and 22.31 ± 1.41%, P < 0.001) (Fig. 6H).

Acetate and TSA treatment significantly increased HAT activity in podocytes compared with controls at 24 h (0.225 ± 0.008 vs. 0.058 ± 0.000, P < 0.001; 0.345 ± 0.025 vs. 0.058 ± 0.000, P < 0.001), 48 h (0.244 ± 0.013 vs. 0.058 ± 0.000, P < 0.001; 0.371 ± 0.008 vs. 0.058 ± 0.000, P < 0.001), and 72 h (0.317 ± 0.039 vs. 0.067 ± 0.007, P < 0.01; 0.427 ± 0.020 vs. 0.067 ± 0.007, P < 0.001) (Fig. 6G).

Acetate regulates B cell proliferation in vitro

We stimulated PBMC cells of patients with 10 µg/ml PLA2R in vitro and treated them with 100 µL of acetate. After 6 days, the number of plasma cells and memory B cells in PBMCs was detected. Compared with the normal control group, PLA2R stimulated the proliferation of plasma cells and memory B cells in PBMCs (4.92 ± 0.64 vs. 3.64 ± 0.35, P < 0.001; 60.05 ± 3.07 vs. 52.42 ± 1.80, P < 0.001, Fig. 7A and B). After treatment with acetate, plasma cell proliferation in PBMCs was significantly reduced compared with the disease control group (56.85 ± 2.74 vs. 60.05 ± 3.07, P < 0.05, Fig. 7B). Acetate can reduce the production of plasma cells stimulated by PLA2R.

Fig. 7.

Acetate regulates B cell proliferation in vitro. PBMC cells of patients were stimulated with 10ug/ml PLA2R in vitro and treated with 100 µL of acetate. After 6 days, the number of memory B cells (A) and plasma cells (B) in PBMCs was detected. Data are presented as mean ± SEM, N = 10 (*P < 0.05, **P < 0.01, ***P < 0.001)

Discussion

In this study, we assessed the major SCFAs—acetate, propionate, and butyrate. The results showed that they could alleviate renal pathological changes and reduce proteinuria of passive Heymann nephritis, with acetate showing the most significant effect. Our findings support a protective role for acetate in experimental MN and provide mechanistic context for the emerging gut–kidney axis in glomerular autoimmunity. A notable feature of acetate biology is its potential for dual action: modulation of immune programs that shape autoantibody responses and direct support of tissue resilience in target organs. Acetate reduced specific IgG levels, thereby reducing deposits in the glomerulus. They restored the decrease of podocyte cytoskeleton protein, podocyte migration and cell activity induced by MN plasma. This concept is particularly relevant to MN, where the clinical course reflects an interplay between persistent immunologic activity (often tracked by anti-PLA2R titers in primary MN) and the capacity of podocytes and the filtration barrier to recover after injury [1, 3].

Although MN is often framed as an antibody-mediated disease, cellular immune programs influence the generation and maintenance of pathogenic humoral responses and the inflammatory milieu that amplifies glomerular injury [1, 6]. SCFAs are canonical microbial metabolites generated by fermentation of dietary fiber and can shape immunity via receptor-mediated signaling and epigenetic mechanisms [6, 10, 11]. In addition to acting locally in the gut, SCFAs reach the systemic circulation and may alter immune cell differentiation, cytokine production, and migratory behavior [6, 7].

Prior work demonstrated that acetate can promote Treg homeostasis and expand regulatory programs, while also modulating effector differentiation depending on context and concentration [10, 11]. Mechanistically, SCFAs can inhibit histone deacetylases and influence the mTOR–S6K axis, providing a direct link between metabolite availability and immune transcriptional programs [10]. SCFA-sensing G-protein–coupled receptors expressed on immune cells and other tissues provide an additional layer of signaling that may integrate microbial metabolism with host physiology [6, 19].

Interpreted in this framework, our findings are consistent with the hypothesis that acetate supplementation favors immune recalibration toward regulation rather than broad immunosuppression. Such an effect could be clinically valuable in MN, where a reduction in immunologic activity is a prerequisite for durable remission, but preservation of protective immunity is also important, especially in older patients and those with comorbidities.

Clinical observations increasingly report dysbiosis in primary MN, including altered microbial community structure and changes in predicted metabolic functions related to SCFA pathways [8, 9]. Across chronic kidney disease more broadly, reductions in SCFA pools and depletion of SCFA-producing taxa have been associated with heightened systemic inflammation, suggesting that impaired SCFA signaling may contribute to a permissive inflammatory environment [7]. Together with renal and vascular expression of SCFA receptors, these observations provide a plausible context in which an SCFA-deficient or dysregulated state could amplify immune activation relevant to MN⁶. Importantly, observational studies cannot establish causality, and MN itself as well as immunosuppressive therapy may alter the microbiome and metabolome. An important next step will be to integrate microbiome profiling and targeted metabolomics with longitudinal immune phenotyping and clinical outcomes, to determine whether SCFA-related signatures track with disease activity and treatment response.

Beyond immune regulation, acetate directly protect glomerular intrinsic cells. Podocyte integrity depends on cytoskeletal stability, and synaptopodin is a key actin-associated protein that supports stress-fiber organization and resistance to injury [20]. Direct podocyte protection is mechanistically appealing in MN, because podocyte recovery may lag behind immunologic remission and may be influenced by ongoing metabolic and oxidative stress [1, 2]. Furthermore, although acetate treatment did not significantly reduce glomerular complement deposition, its protective effects in vivo may be partly mediated through alternative mechanisms, such as systemic immune cell modulation. Our in vitro experiments using cultured human podocytes provide supportive, hypothesis-generating evidence that acetate may have direct effects on this cell type.

Experimental work in proteinuric kidney disease demonstrates that SCFAs can protect the kidney through both epigenetic effects and metabolite-sensing receptors. Acetate exhibits a selective affinity for GPR43 and serves as a key energy source for immune cells, particularly T cells, thereby participating in immunometabolism [21]. Propionate strongly stimulates both GPR41 and GPR43, while acetate and butyrate selectively target GPR43 and GPR41, respectively [22], butyrate improves proteinuria via HDAC inhibition and GPR109A signaling, with GPR109A expressed in podocytes [13]. Acetate may reduce ROS and improve migration in podocytes potentially through GPR43 or HDAC inhibition [23]. In podocytes, the overexpression and deficiency of RHOA, CDC42, and RAC1 activities are associated with podocyte dysfunction, leading to the disappearance of foot processes and the disruption of filtration barriers [24]– [28]. Our study provides correlative evidence suggesting a potential pathway by which acetate might influence MN, involving SCFA receptor expression, modulation of inflammatory transcripts NF-κB,, reduction of ROS, and protection of podocyte integrity markers.

SCFA suppresses arthritis by raising 5-HIAA (serotonin metabolite) to activate AhR (a Breg marker), thereby supporting Breg function while inhibiting GC B cells and plasmablast differentiation [29]. In addition to immune regulation and direct cytoskeleton protection, SCFAs may also contribute to stabilizing the podocyte cell membrane environment. Recent studies have shown that in MN, antigen bound by antibodies can be shed from the podocyte surface as a protective mechanism [30]. SCFAs may stabilize podocyte membranes in MN by supporting phospholipid biosynthesis, as antigen shedding from podocytes is a protective mechanism in MN [31].

Gut microbiota enhances SCFA production, upregulating PPAR-γ and downregulating PPAR-α/β/δ [32]. SCFAs, via GPR43, promote Th1-derived IL-10 [33], reduce TNF-α/IL-6/MDA, increase SOD/GSH-Px, inhibit apoptosis, and suppress the TNF-α/NF-κB/IL-6 pathway [34].

While these preclinical findings are promising, it is important to emphasize that this study was conducted in an experimental model and using in vitro systems. Further studies in additional animal models, followed by carefully designed clinical trials, will be necessary to determine whether SCFA-based approaches have therapeutic utility in human MN. We acknowledge that while our in vitro experiments used disease-relevant stimuli to demonstrate the anti-inflammatory effects of acetate on podocytes, they did not identify the specific cytokines (such as IL-6 or TNF-α) that are directly targeted. Future studies employing single cytokine challenges are warranted to elucidate the precise signaling pathways involved in acetate-mediated podocyte protection.

There are a few limitations about our study. First, we acknowledge a critical limitation of our experimental model. The passive Heymann nephritis (PHN) model is induced by injection of a heterologous anti-serum, which initiates a rapid, pre-formed immune complex deposition. Unlike primary human MN, this model does not recapitulate the breakdown of immune tolerance or the spontaneous generation of pathogenic autoantibodies. Consequently, the immunomodulatory effects we observed, such as the reduction in anti-sheep IgG and the shift in T-cell subsets, occur within the context of an ongoing, established adaptive response to a foreign antigen rather than representing true prevention of autoimmunity. Our findings should therefore be interpreted as demonstrating the therapeutic potential of acetate in an established model of antibody-mediated podocyte injury, not as evidence for its role in preventing disease initiation. Secondly, the in vitro culture system used in the present study does not fully recapitulate the complex glomerular microenvironment in vivo. The mechanisms by which acetate modulate immune-podocyte crosstalk or interact with other glomerular cell types remain to be explored. Future studies using co-culture systems or kidney organoids are warranted to bridge the gap between our cellular observations and the whole-animal phenotype. We demonstrate that acetate modulates T-cell subsets in the context of PHN, we did not include a control group receiving acetate alone without disease induction. However, a previous human dose‑titration study demonstrated that propionic acid alone at 1,000 mg/day for 14 days significantly increased Treg cell frequencies compared with baseline (p < 0.0001) [35]. These findings confirm that SCFAs alone are sufficient to expand Treg cells in humans, suggesting that the immune changes observed in our study are at least partially mediated by SCFAs rather than other confounders. Nevertheless, future studies with a standalone SCFA control arm are warranted for more definitive causal attribution. Moreover, systematic time-course validation and more rigorous dose-response assessments—using at least two or more acetate concentrations and concomitant measurement of acetate levels in kidney tissue—are required to extend our findings. Nevertheless, we acknowledge a significant technical limitation: we did not measure acetate concentrations in plasma or renal tissue. The dose we used (150 mM in drinking water, approximating 6 mmol/kg/day) is comparable to doses previously reported to be effective in other rodent models of immune-mediated kidney disease [11]. However, without direct measurement, it remains unclear whether the observed effects are mediated by a physiological or supraphysiological exposure. Future studies should incorporate detailed pharmacokinetic and dose-response analyses, including measurement of acetate levels in target tissues, to establish the therapeutic window and confirm the physiological relevance of the intervention.

In summary, our findings support acetate as mechanistically plausible modulators of the gut–immune–podocyte axis in MN. By promoting regulatory immune programs and enhancing podocyte stress resistance, acetate represents promising adjunct candidates that warrant further mechanistic and translational evaluation.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

MN

Membranous nephropathy

SCFAs

Short-chain fatty acids

PHN

Heymann nephritis

ROS

Reactive oxygen species

Treg

Regulatory T

GPR

G protein-coupled receptor;‌ WT1,Wilms Tumor 1

GBM

Glomerular basement membrane

PLA2R

Phospholipase A2 receptor

HDAC

Histone deacetylase

PHN

Passive Heymann nephritis

TNF-α

Tumor necrosis factorα

Th1

T helper 1

Th17

T helper 17

TSA

Trichostatin A

HAT

Histone Acetyltransferase

PBMCs

Peripheral blood mononuclear cells

IFN-γ‌

‌Interferon-gamma

PPAR-γ

Peroxisome proliferator-activated receptor gamma

‌PCR

Polymerase Chain Reaction

ELISA

Enzyme-Linked Immunosorbent Assay

Author contributions

Jing Liu (First Author): Conceptualization, methodology, software, investigation, data curation, formal analysis, writing - original draft; Yan-yan Cui: Methodology, Data Curation, Formal Analysis; Kai Zheng: Formal Analysis, Writing - Original Draft, software; Xiao-yu Jia: Visualization, supervision, funding acquisition, writing - review & editing; Zhao Cui (Corresponding Author): Resources, supervision, funding acquisition, writing - review & editing; Ming-hui Zhao: Supervision, Funding Acquisition, Writing - Review & Editing.

Funding

This work was supported by the National Natural Science Foundation of China (82325009), The Science and Technology Project of Suzhou Municipal Health Commission (QNXMM2025026), Suzhou Science and Technology Development Project (Science and Technology Project - Medical and Health Innovation) (SYWD2025188).

Data availability

All the data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.

Declarations

Ethics approval and consent to participate

This study strictly adheres to the Helsinki 9 Declaration and was approved by the Ethics Committee of Peking University First Hospital, approval number: J202055 and 2017[1280]. All human plasma samples were collected after obtaining written informed consent from participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.