Endothelial Ffar4 protects against diabetic kidney disease by potentiating endogenous retinoic acid metabolism

Abstract

Background

Diabetic kidney disease (DKD) occurs in up to 40% of individuals with diabetes and remains the primary cause of kidney failure worldwide, and a complex interaction of genetic and environmental dietary factors may be involved. Free fatty acid receptor 4 (Ffar4) may serve as a link between the genetic and dietary aspects of DKD progression; however, its role in DKD remains unclear.

Methods

Ffar4-mediated DKD protection was evaluated using comprehensive genetic models. In addition, the effects of Ffar4 on glomerular inflammation and endothelial injury in mice were evaluated in vivo and in vitro, and the regulation of the Aldh1a1 gene by Ffar4 to maintain endogenous retinoic acid (RA) metabolic balance and related signaling pathways in the glomeruli was investigated.

Results

We found that Ffar4 expression was decreased in diabetes and was associated with renal complications. Conventional and endothelial-specific Ffar4 knockout exacerbated DKD, whereas endothelial-specific Ffar4 overexpression improved renal function. Mechanistically, Ffar4 regulated endogenous RA metabolism in the glomeruli through the Atf4–Aldh1a1 pathway. RA supplementation partially reversed DKD progression in endothelial-specific Ffar4 knockout mice.

Conclusions

Taken together, these findings revealed a novel role of Ffar4 in potentiating endogenous RA production and delaying the progression of DKD-related multi-dysfunction.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s11658-026-00883-2.

Introduction

Diabetes kidney disease (DKD) represents the leading cause of renal failure, affecting approximately 40% of patients with diabetes [1, 2]. This condition not only diminishes quality of life but also imposes significant economic burdens. The rising prevalence of DKD and limited therapeutic options render it a persistent medical challenge [3–6].

Traditionally, the primary aetiology of this renal complication has been attributed to the interaction between haemodynamic and metabolic factors. However, current evidence indicates that diabetic kidney damage severity cannot be completely explained by metabolically driven increases in systemic and intraglomerular pressure [7, 8]. It remains unclear why certain diabetic patients progress to uraemia, while others exhibit no overt renal impairment. This heterogeneity likely stems from complex genetic susceptibility–environment interactions.

Endothelial cells constitute the majority of glomerular cells, accounting for more than 50% of the population [9]. Maintaining renal endothelial homeostasis is crucial for preserving the structure and function of the glomerulus, sustaining an anti-inflammatory and anti-thrombotic state, and preventing renal fibrosis [10–13]. Increasing data suggest that endothelial dysfunction plays a role in driving DKD pathogenesis as an early event [11, 14]. However, the molecular pathways underlying DKD progression in glomerular endothelial cells (GECs) remain incompletely defined.

G protein coupled receptors (GPCRs) are extensively studied and represent promising therapeutic targets [15]. The free fatty acid receptors (FFARs) family-including Ffar1 (GPR40), Ffar2 (GPR43), Ffar3 (GPR41), and Ffar4 (GPR120), which play important roles in regulating biological processes central to glucose and lipid metabolism [15]. At present, Ffar4 primarily serves as a receptor of long-chain fatty acids (LCFA) in the FFAR family, which may bridge the gap between genetic susceptibility and environmental factors in diabetic kidney complications [16–18]. Widely expressed across tissues, Ffar4 is an emerging anti-diabetic drug target. Evidence suggests tissue-specific biological functions for Ffar4, including anti-inflammatory effects and metabolic improvement [17–21]. A detailed description of the physiological functions of Ffar4 in different tissues is essential and may provide possible mechanisms underlying the contradictory findings of ω-3 polyunsaturated fatty acids (PUFA) trials [22, 23]. Although, previous studies have predominantly focused on elucidating the role of Ffar4 in acute kidney injury (AKI) and unilateral ureteral obstruction (UUO) [20, 24, 25], the specific role of Ffar4 in DKD remains unclear.

In this work, we found that Ffar4 expression was decreased in patients with DKD and mice. Endothelial Ffar4 knockout displayed accelerated DKD progression, whereas endothelial-specific overexpression of Ffar4 improved kidney function. Mechanistically, we found that endothelial Ffar4 regulated endogenous RA metabolism through the Atf4–Aldh1a1 pathway.

Materials and methods

Details for the methods are given in the Supplementary Materials and Methods [26–31].

Human participants

We obtained peripheral kidney tissue samples from a total of 10 patients with T2DKD at the Nephrology Department of Jiangnan University Medical Center (JUMC). All participants with T2DKD fulfilled the diagnostic criteria set by the American Diabetes Association (Supplementary Table S1). The kidneys in the control group originated from non-diabetic patients who underwent renal surgery for chronic kidney cancer at JUMC, with samples collected from tumour-free regions. Exclusion criteria included individuals under the age of 18, those with other kidney disorders, pregnant women, infected individuals and those with genetic diseases. The study was authorized by the Ethics Committee of JUMC(Ethics Review No. Y-125), and all participants gave consent.

Animal studies

All the animal procedures adhered to the Guide for Care and Use of Laboratory Animals of the School of Medicine, Jiangnan University, and were approved by the Animal Ethics Committee of Jiangnan University (No: JN20211215c0500630).

Generation of Ffar4-conditioned knockout mice

Total Ffar4-knockout (KO, RRID: MGI: 7,256,540) mice were obtained from Shanghai Bioraylab. Villin-Cre mice were obtained from Shanghai Biomodel Organism. Floxed Tek-Cre and Cdh16-Cre mice were constructed commercially by the Nanjing Biomedical Research Institute of Nanjing University. Floxed Ffar4 (fl/fl, RRID: MGI: 7,256,541) and Ffar4-overexpressing transgenic mice (cag/cag, RRID: MGI: 7,256,542) were constructed commercially by Shanghai Biomodel Organism and Nanjing Biomedical Research Institute of Nanjing University, respectively. The fl/fl mice were mated with Villin-Cre mice to generate gut-specific Ffar4-knockout mice. The fl/fl mice were mated with Tek-Cre mice to generate vascular endothelial-specific Ffar4 knockout mice. The fl/fl mice were mated with Cdh16-Cre mice to generate kidney tubular Ffar4-specific knockout mice. The cag/cag mice were crossed with Tek-Cre mice to obtain vascular endothelial-specific Ffar4-overexpressing transgenic mice.

Cell line

This study utilized four endothelial cell lines: EKO, EOE, CRE (all derived from transgenic mice) and human umbilical vein endothelial cells (HUVECs). The EKO, EOE and CRE cell lines were extracted from the primary glomerular vascular endothelial cells of EKO mice (Tek-Cre; Ffar4 Loxp/Loxp), EOE mice (Tek-Cre; Ffar4 cag/cag) and CRE mice (Tek-Cre). HUVECs (CRL-1730) were maintained in-house, originally sourced from ATCC (Cat# BFN607200285). All cell lines were cultured in cultured in endothelial cell medium (Gibco, M200500) with 10% fetal bovine serum (FBS) at 37 °C in a 5% CO₂ atmosphere.

Statistical analysis

All the data were analysed by GraphPad Prism 8.0. P values < 0.05 were considered significant. The data are presented as the mean ± standard error of the mean (SEM). Statistical significance was determined using two-tailed, unpaired Student’s t tests for two group comparisons or one-way analysis of variance (ANOVA) followed by Tukey’s post hoc tests for multiple comparisons.

Results

Decreased Ffar4 expression is associated with DKD progression

To elucidate the role of Ffar4 in DKD, we assessed Ffar4 expression in established DKD models (STZ-induced diabetic and db/db mice). Renal Ffar4 expression was significantly decreased in both models (Fig. 1A, B). Concurrently, inflammation marker TNFR1 and vascular endothelial injury indicator VCAM1 [32, 33] increased, while vascular homeostasis regulator VE-cadherin decreased (Fig. 1C–E). To further establish the clinical relevance of Ffar4, we analysed DKD data (https://www.ukbiobank.ac.uk; ID: finn − b − DM_NEPHROPATHY) about the quantitative trait locus data of Ffar4 and performed two-sample Mendelian randomization (MR) analysis (Fig. 1F, G). Mutations in Ffar4 significantly increased the incidence of diabetic kidney disease within the population. These findings suggest that Ffar4 may play a crucial role in the pathophysiology of DKD.

Fig. 1.

Decreased Ffar4 expression is associated with DN progression. A, Expression of Ffar4 in the kidney of db/db mice (n = 6). B, Expression of Ffar4 in the kidney of STZ-induced DKD model mice (n = 6). C, Immunoblotting of TNFR1, VE cadherin and VCAM1 in kidney tissue (n = 3). D, Statistical analysis of TNFR1, VE cadherin and VCAM1 in STZ-induced DKD model mice kidney tissue (n = 3). E, Statistical analysis of TNFR1, VE cadherin and VCAM1 in db/db mice kidney tissue (n = 3). F, This plot was used to visualize the effect of each SNP. The horizontal axis represents the exposure effect, and the vertical axis represents the outcome effect. The slopes of the lines represent the causal effect of each method. G, Leave-one-out analyses to evaluate whether any single instrumental variable drove the causal effect. The data are presented as means ± SEMs; Two-tailed unpaired Student’s t test (A–B, D–E); *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Systemic Ffar4 deletion exacerbates DKD progression

To clarify whether Ffar4 deficiency affects DKD progression, we generated systemic Ffar4 knockout mice (Fig. 2A). Given that > 90% of diabetic patients have type 2 diabetes with persistent metabolic dysregulation [34], we established a T2DKD model using HFD + low-dose STZ [35] (Fig. 2B). Ffar4^−/−-DKD mice displayed increased baseline fasting glucose and impaired glucose tolerance versus Ffar4^+/+-DKD controls (Fig. 2C). Moreover, Ffar4^−/−-DKD mice showed elevated 24-h urinary protein, serum creatinine, urinary protein/creatinine ratios and inflammatory cytokines (TNFα, IL-1β, and IL-6) (Fig. 2D, E). DKD is characterized by a progressive form of glomerulopathy. Histopathological analysis revealed exacerbated diabetic lesions in Ffar4^−/−-DKD mice, including pronounced glomerular dilatation and increased glomerular/interstitial fibrosis (Fig. 2F–H). Collectively, systemic Ffar4 deletion exacerbates DKD pathogenesis.

Fig. 2.

Systemic Ffar4 deletion exacerbates DKD-induced renal dysfunction and inflammation. A, CRISPR-Cas9 gene editing strategy for total Ffar4^−/− mice. B, Schematic for DKD or control group modelling. C, Fasting blood glucose (FBG, mmol/L) and the area under the curve (AUC) for both the intraperitoneal glucose tolerance test (ITT) and oral glucose tolerance test (GTT) of mice (n = 5). D, 24-h urinary protein (24H UPE, mg/d), serum creatinine (CRE, μmol/L) and urinary albumin/creatinine ratio (UACR, μg/mg) of the mice (n = 5). E, Expression of inflammatory cytokines (IL-1β,IL-6 and TNFα) in kidney tissue (n = 5). F, Representative images of haematoxylin and eosin (H&E)-stained sections (400 × , scale bar = 100 μm). G, Representative images of Masson-stained sections (400 × , scale bar = 100 μm). H, The glomerular area (μm²) and fibrosis area of renal interstitium (%) and glomerular (%) in these groups (n = 5). The data are presented as means ± SEMs; Two-tailed unpaired Student’s t test (A, C–E, H), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Endothelial loss of Ffar4 accelerates DKD progression

Ffar4 is widely expressed, notably in intestinal tissues [36]. Analysis of healthy mouse kidney single-cell data revealed high Ffar4 expression in renal endothelial and tubular epithelial cells (Fig. 3A). We thus generated endothelial-specific Ffar4 (Tek-Cre; Ffar4 Loxp/Loxp, (Fig. 3B, C, abbreviated EKO) and tubule-specific Ffar4 (Cdh16-Cre; Ffar4 Loxp/Loxp, (Supplementary Fig. S1A, abbreviated TKO) knockouts. Magnetic bead-sorted CD31⁺ endothelial cells confirmed significantly reduced Ffar4 expression in EKO mice (Fig. 3C).

Fig. 3.

Endothelial Ffar4 mediates DKD-induced renal dysfunction and inflammation. A, Single-cell sequencing plot of the kidney interactive transcriptome (Visualization available at http://humphreyslab.com/SingleCell/, Healthy Mouse Dataset: Wu et al., JASN 2019; and RBK RID: 14-4KBC). B, CRISPR-Cas9 gene editing strategy for endothelial-specific Ffar4 knockout mice (EKO). C, qPCR of Ffar4 in the renal endothelial cells sorted by magnetic beads (n = 5). D, Fasting blood glucose (FBG, mmol/L) and the area under the curve (AUC) for both the intraperitoneal glucose tolerance test (ITT) and oral glucose tolerance test (GTT) of EKO mice (n = 5). E, 24-h urinary protein (24H UPE, mg/d), serum creatinine (CRE, μmol/L), and urinary albumin/creatinine ratio (UACR, μg/mg) of the mice (n = 5). F, Representative images of haematoxylin and eosin (H&E)-stained sections and images of Masson-stained sections (400 × , scale bar = 100 μm). G, The fibrosis area of renal glomerular (%), glomerular area (μm²) and fibrosis area of renal interstitium (%) and glomerular (%) in these groups (n = 5). H, Immunofluorescent staining of VCAM1 in the CRE-DKD and EKO-DKD groups. Scale bar, 10 μm. (n = 12) per group. I, Statistical analysis of VCAM1 level in the CRE-DKD and EKO-DKD groups (n = 12). The data are presented as the means ± SEMs. Two-tailed unpaired Student’s t test (C–E, G, I). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

After DKD model induction, EKO-DKD mice showed severely impaired glucose metabolism (elevated FBG, impaired GTT/ITT; (Fig. 3D) and renal dysfunction (Fig. 3E) were observed. Histopathology revealed exacerbated glomerular dilatation and glomerular/interstitial fibrosis (Fig. 3F, G). Immunofluorescence analysis showed elevated VCAM1 in EKO-DKD mice (Fig. 3H, I), indicating endothelial Ffar4 ablation amplifies inflammatory responses and renal injury. These phenotypes were absent in TKO-DKD mice (Supplementary Fig. S1B–G). Thus, endothelial Ffar4 critically regulates renal inflammation and pathology in DKD.

Endothelial Ffar4 overexpression represses DKD-related renal dysfunction and inflammation

To comprehensively investigate the role of endothelial Ffar4 in DKD-related renal dysfunction, we generated endothelial-specific Ffar4 overexpression mice (Tek-Cre; Ffar4 cag/cag, (Fig. 4A, B, abbreviated EOE). After model induction, alongside EOE-DKD mice, exhibited reduced fasting blood glucose and improved glucose/insulin tolerance (Fig. 4C). Renal function improved (decreased 24-h urinary protein, serum creatinine, protein/creatinine ratios; (Fig. 4D), with attenuated glomerular dilation and fibrosis (Fig. 4E, F). EOE-DKD mice showed reduced endothelial inflammatory proteins (Fig. 4G, H). These results demonstrate endothelial Ffar4 overexpression mitigates DKD-related renal dysfunction and inflammation.

Fig. 4.

Endothelial Ffar4 overexpression represses DKD-induced renal dysfunction and inflammation. A, Transgenic strategy for endothelial-specific Ffar4 expression. Endothelial-specific Ffar4 overexpressing mice (EOE) were obtained by mating with Tek-Cre mice, which removes the STOP fragment. B, qPCR of Ffar4 in the renal endothelial cells sorted by magnetic beads (n = 5). C, Fasting blood glucose (FBG, mmol/L) and the area under the curve (AUC) for both the intraperitoneal glucose tolerance test (ITT) and oral glucose tolerance test (GTT) of mice (n = 5). D, 24-h urinary protein (24H UPE, mg/d), serum creatinine (CRE, μmol/L) and urinary albumin/creatinine ratio (UACR, μg/mg) of the mice (n = 5). E, Representative images of HE-stained sections and images of Masson-stained Sects. (400 × , scale bar = 100 μm). F, The fibrosis area of renal glomerular (%), glomerular area (μm²) and fibrosis area of renal interstitium (%) in these groups (n = 5). G, Statistical analysis of VCAM1 level in the CRE-DKD and EOE-DKD groups (n = 12). H, Immunofluorescent staining of VCAM1 in the CRE-DKD and EOE-DKD groups. Scale bar, 10 μm. (n = 12) per group. The data are presented as means ± SEMs. Two-tailed unpaired Student’s t test (B, C–D, F–G), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Transcriptomic analysis indicates that Aldh1a1 is a potential downstream regulator of Ffar4 in DKD

The glomerular endothelium comprises the predominant compartment of the glomerulus and critically regulates the sieving properties of the glomerular filtration barrier [9, 10, 37]. To elucidate the molecular mechanisms underlying Ffar4 function in DKD, we performed transcriptomic analysis of primary glomerular endothelium isolated from EKO-DKD and EOE-DKD mice compared with their respective controls (Fig. 5A). In EKO-DKD glomerular endothelium, 148 genes were upregulated and 177 downregulated, while EOE-DKD glomerular endothelium exhibited 293 upregulated and 141 downregulated genes. Gene Ontology (GO) analysis revealed significant differences in cellular signalling pathways between Ffar4-knockout/overexpressing mice and controls (Fig. 5B, C). The RA pathway and its key gene, aldehyde dehydrogenase 1 family member A1 (Aldh1a1), were identified. qPCR confirmed significantly reduced Aldh1a1 expression in endothelial cells from EKO mice and increased expression in EOE mice (Fig. 5D).

Fig. 5.

Transcriptome analysis suggests that Aldh1a1 is a potential target in Ffar4-regulated DKD. A, Schematic of the methods used for glomerular endothelium isolation, RNA extraction, RNA-seq and data analysis. B, Volcano plot and heatmap showing the distribution of upregulated (red) and downregulated (blue) genes. GO enrichment analysis of downregulated (blue) and upregulated (red) genes in the glomeruli of CRE-DKD and EKO-DKD mice. C, Volcano plot and heatmap showing the distribution of upregulated (red) and downregulated (blue) genes. GO enrichment analysis of downregulated (blue) and upregulated (red) genes in the glomeruli of CRE-DKD and EOE-DKD mice. D, qPCR of Aldh1a1 in the glomeruli endothelial cells (n = 6). E, Immunohistochemical quantification of Aldh1a1/E-selectin/CD68/α-SMA expression in glomeruli. F, Representative immunohistochemistry images of Aldh1a1/E-selectin/CD68/α-SMA in human kidney biopsy samples from the nondiabetic and DKD groups (400 × , scale bar = 100 μm); The data are presented as means ± SEMs. One-way analysis of variance (D), Two-tailed unpaired Student’s t test (E), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Aldh1a1, a major rate-limiting enzyme in endogenous RA metabolism, influences gluconeogenesis and lipid metabolism [38, 39]. Endothelial Aldh1a1 was significantly inversely correlated with key DKD indicators (inflammatory factors, FBG, UACR) in mice (Supplementary Fig. S2A, B). Immunohistochemistry of human DKD biopsy samples showed markedly decreased glomerular Aldh1a1 expression alongside increased E-selectin, CD68 and α-SMA levels (Fig. 5E, F). Clinical databases further associated serum Aldh1a1 with renal function (Supplementary Fig. S2C, D). Glomerular Aldh1a1 protein levels inversely correlated with E-selectin, CD68 and α-SMA (Supplementary Fig. S2E). Together, these findings demonstrate that Aldh1a1 may play a pivotal downstream regulator of Ffar4 in mediating DKD progression.

Enhancement of the Aldh1a1–RA axis protects against the exacerbation of DKD induced by Ffar4 deletion

To further define the Ffar4–Aldh1a1 regulatory axis, we isolated primary glomerular vascular endothelial cells from EOE, EKO and control (CRE) mice (Supplementary Fig. S3A–D). qPCR analysis revealed a significant reduction in Aldh1a1 expression in EKO endothelial cells (Fig. 6A). Importantly, Aldh1a1 transfection attenuated inflammatory responses in Ffar4-KO ECs under high glucose/palmitate (HG-Pal) stimulation (Fig. 6A, B). Considering the pivotal role of Aldh1a1 in promoting RA synthesis, primary endothelial cells were treated with RA. RA treatment dose-dependently reduced HG-Pal induced inflammatory cytokine expression (Fig. 6C). As Aldh1a enzymes catalyze retinal-to-RA conversion [40] (Fig. 6D), we hypothesized altered retinal/RA levels in EKO-DKD mice. Liquid chromatography–mass spectrometry (LC–MS) analysis of serum and renal cortex showed decreased RA but increased retinal versus controls (Fig. 6E). Conversely, EOE-DKD mice exhibited increased RA and decreased retinal (Fig. 6F).

Fig. 6.

RA supplementation alleviates Ffar4 deletion-induced DKD exacerbation A, Expression of inflammatory cytokines (IL-1β, IL-6, IL-18 and Aldh1a1) in primary endothelial cells in the CRE-con, EKO-pcAlah1a1 and EKO-con groups after HG-Pal stimulation for 24 h (n = 3). B, Expression of endothelial activation markers (VCAM1, and E-selectin) in primary endothelial cells in the EKO-pcAlah1a1 and EKO-con groups after HG-Pal stimulation for 24 h (n = 6). C, Expression of inflammatory cytokines (IL-1β, IL-6, TGFβ and Rarα) in EKO group primary endothelial cells after HG-Pal stimulation for 24 h (n = 4). D, Schematic representation of retinol, retinaldehyde and RA metabolism. E, Fresh serum and renal cortical tissue samples were collected from CRE-DKD and EKO-DKD mice for targeted LC‒MS analysis (n = 4) (Serum, ng/mL. Renal cortical tissue protein, ng/mg). F, Fresh serum and renal cortical tissue samples were collected from CRE-DKD and EOE-DKD mice for targeted LC‒MS analysis (n = 4) (Serum, ng/mL. Renal cortical tissue protein, ng/mg). G, 24-h urinary protein (24H UPE, mg/d), serum creatinine (CRE, μmol/L) and urinary albumin/creatinine ratio (UACR, μg/mg) in the CRE-DKD, EKO-RA Low (RA, 1mg/kg boby weight), EKO-RA High (RA, 5mg/kg boby weight) and EKO-DKD groups (n = 4). H, Statistical analysis of VCAM1 level in the CRE-DKD, EKO-RA High and EKO-DKD groups (n = 12). I, Immunofluorescent staining of VCAM1 in the CRE-DKD, EKO-RA High and EKO-DKD groups. Scale bar, 10 μm. (n = 12) per group. The data are presented as means ± SEMs; Two-tailed unpaired Student’s t test (B, E–F), One-way analysis of variance (A, C, G-H), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

To further verify whether RA mitigates Ffar4 deletion-induced DKD progression, we performed rescue experiments in RA-treated EKO mice. RA supplementation normalized fasting blood glucose and glucose tolerance (Supplementary Fig. S4A). Critically, RA administration dose-dependently reduced 24-h urinary protein and creatinine, preserving kidney function (Fig. 6G). These improvements were further supported by histopathological staining (Supplementary Fig. S4B–D). High-dose RA also significantly decreased endothelial inflammatory protein expression (Fig. 6H, I). These results indicated that exogenous RA ameliorates DKD progression in EKO mice.

Inhibition of the Aldh1a1–RA signalling pathway blocks the alleviation of DKD induced by Ffar4 overexpression

To further substantiate the role of the Aldh1a1–RA axis as a downstream regulator of Ffar4 in DKD, we isolated primary endothelial cells from EOE mice and silenced Aldh1a1 using siRNA. Aldh1a1 silencing significantly increased inflammatory factors and endothelial activation markers under HG-Pal stimulation (Fig. 7A, B). RA exerts its pleiotropic effects by directly activating RAR receptors (RARα/β/γ) or by serving as a heterodimeric partner for RXR-mediated signaling. We assessed their expression in ECs post-RA intervention. RARα exhibited the highest induction at both protein and mRNA levels (Fig. 7C). Pharmacological inhibition of RARα and FFAR4 in EOE mice reversed the protective effects of Ffar4 overexpression against DKD in vitro (Fig. 7D and Supplementary Fig. S5A–B) and in vivo (Fig. 7E–G and Supplementary Fig. S6A–E). These results indicated that Ffar4 mediated the protective effect against renal dysfunction and inflammation via the Aldh1a1–RA–Rarα axis.

Fig. 7.

Alah1a1-generated RA attenuates endothelial inflammation through RARα. A, Expression of inflammatory cytokines (IL-1β, IL-6, IL-10 and Aldh1a1) in primary endothelial cells in the CRE-con, EOE-siAlah1a1 and EOE-con groups after HG-Pal stimulation for 24 h (n = 4). B, Expression of endothelial activation markers (VCAM1, and E-selectin) in primary endothelial cells in the EOE-siAlah1a1 and EOE-con groups after HG-Pal stimulation for 24 h (n = 6). C, Expression of RA receptors in CRE-con primary endothelial cells after RA stimulation for 24 h (n = 4). D, Expression of inflammatory cytokines (IL-1β, IL-6, TGFβ and Rarα) in CRE-inhibitor (Rarα inhibitor RO 41–5235, 1 μmol) and CRE-con groups primary endothelial cells after HG-Pal stimulation for 24 h (n = 4). E, 24-h urinary protein (24H UPE, mg/d), serum creatinine (CRE, μmol/L) and urinary albumin/creatinine ratio (UACR, μg/mg) of the mice (n = 4). F, Statistical analysis of VCAM1 level in the CRE-DKD, EOE Inhibitor and EOE-DKD groups (n = 12). G, Immunofluorescent staining of VCAM1 in the CRE-DKD, EOE Inhibitor and EOE-DKD groups. Scale bar, 10 μm. n = 12 per group. The data are presented as means ± SEMs; One-way analysis of variance (A, D) Two-tailed unpaired Student’s t test (B. C), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Ffar4 regulates Aldh1a1 transcription via Atf4

Transcriptomic analysis coupled with transcription factor prediction revealed Atf4 as a potential reciprocal regulator of Aldh1a1 and β-arrestin2 (Fig. 8A). Given that Ffar4 recruits β-arrestin2 to drive receptor internalization and dampen inflammation [41], we investigated the relationship between Atf4 and β-arrestin2. Co-immunoprecipitation (Co-IP) confirmed their endogenous interaction in various endothelial cell lines (Fig. 8B and Supplementary Fig S7A). Furthermore, Atf4 knockdown markedly reduced Aldh1a1 in endothelial cells under HG-Pal stimulation (Fig. 8C), and dual-luciferase assays demonstrated direct Atf4 binding to the Aldh1a1 promoter (Fig. 8D, E). CUT and RUN analysis of Atf4–DNA interactions (using JASPAR-predicted binding sites; (Fig. 8F, G) validated direct ATF4 binding at three high-score loci (positions 292–305, 117–130, 872–885 bp) in the Aldh1a1 promoter, confirmed by qPCR and agarose gel electrophoresis (Fig. 8H). After the β-Arrestin2 protein was inhibited, this binding effect was significantly downregulated (Supplementary Fig. S7B–C). Fluorescence colocalization revealed predominant glomerular Aldh1a1 expression in endothelia (Fig. 8I, J). Collectively, these results demonstrate that Ffar4 enhances Aldh1a1 expression by activating the transcription factor Atf4, enabling precise regulation of RA synthesis (Fig. 8K).

Fig. 8.

Ffar4 regulates Aldh1a1 transcription via Atf4. A, Jaspar, PROMO and endothelial transcriptome analyses were employed for predicting the reciprocal TF of Aldh1a1. B, A Co-IP assay was used to detect the interaction between endogenous Atf4 and β-arrestin2. C, Expression of Atf4 and Aldh1a1 in primary endothelial cells in the CRE-con, EOE-siAtf4 and EOE-con groups after HG-Pal stimulation for 24 h (n = 3). D, Schematic diagram of plasmid construction. E, Dual-luciferase reporter assay of the Aldh1a1 promoter. F, Atf4 binding sites in the Aldh1a1 promoter sequence were analysed using the JASPAR database. G, JASPAR database analysis of Atf4-binding sites on the Aldh1a1 promoter. H, DNA agarose gel electrophoresis and qPCR analysis of Aldh1a1 promoter DNA using the CUT and RUN assay and qPCR with different primers (n = 4). I, Immunofluorescent staining of VCAM1, Nephrin and Aldh1a1. Scale bar, 10 μm. n = 4 per group. J, Colocalization analysis of VCAM1, Nephrin and Aldh1a1 in kidney tissue (n = 4). K, Schematic of the Ffar4–Aldh1a1 pathway. The data are presented as means. One-way analysis of variance (C), Two-tailed unpaired Student’s t test (E, H) *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001Renal endothelial transcriptomic dataRenal endothelial transcriptomic data

Discussion

In this study, we observed a reduction in Ffar4 expression in individuals with diabetes, which was correlated with an increased risk of renal complications. Conventional and endothelial-specific knockout of Ffar4 exacerbated DKD-related kidney damage, while the endothelial-specific overexpression of Ffar4 protected renal function, establishing endothelial Ffar4 as a critical regulator of DKD pathogenesis. Mechanistically, Ffar4 governs endogenous RA synthesis via the Atf4–Aldh1a1 axis, modulating glomerular inflammation (Fig. 9).

Fig. 9.

Schematic of the mechanism by which endothelial Ffar4 negatively regulates the progression of diabetic kidney disease. Ffar4 recruits β-arrestin2 and binds to the transcription factor Atf4, leading to its activation of the promoter sequence of Aldh1a1. Increased Aldh1a1 expression promotes endogenous RA production and activates the Rarα signalling pathway to protect against endothelial dysfunction and reduce inflammation, thereby maintain the renal function. (Illustration created with BioRender.com)

Ffar4 exhibits pleiotropic, tissue-specific roles. It regulates SirT3 to mitigate senescence in cisplatin-induced AKI [20], and its agonism improves hyperinsulinemia and inflammation in metabolic dysfunction [15, 42–44]. Conversely, Ffar4 upregulation exacerbates colitis [45], and its agonists show limited efficacy in autoimmune contexts [46], likely owing to tissue-specific signaling biases and lack of targeted ligands [24, 25, 47, 48]. Dietary LCFAs further complicate signaling by differentially engaging G-proteins or β-arrestins [49]. Here, systematic tissue-specific Ffar4 knockout models reveal a distinct renoprotective role for endothelial Ffar4 in DKD.

RA exerts multiple effects in vivo, including promoting angiogenesis and stimulating GLP-1 release in the gut [50]. The kidney plays a crucial role in the process of retinol metabolism and RA synthesis [40, 51]. Aldh1a variants associated with severe inflammation [52], and macrophage Aldh1a1 overexpression dampens inflammation via RA receptor signaling. In human diabetic nephropathy biopsies, Aldh1a1 reduction is inversely correlated with inflammatory markers. Diabetes-induced suppression of endothelial Ffar4 signaling contributes to impaired local RA synthesis within the vascular endothelium, which may subsequently influence systemic RA homeostasis. In this context, restoring endothelial RA production via Ffar4 activation represents a spatially precise means of replenishing RA levels, potentially offering a targeted therapeutic strategy with reduced risk of systemic side effects. Our research demonstrates that Ffar4 modulates RA synthesis, primarily activating the Rarα-dependent pathway while potentially engaging additional nuclear receptors through RXR heterodimerization, highlighting its therapeutic potential for metabolic disorders.

Atf4, a key metabolic regulator, maintains homeostasis and counters stress [53–56]. We show β-Arrestin2, which is downstream of Ffar4, that interacts with Atf4, enhancing Aldh1a1 transcription [41, 49]. This Ffar4–Atf4–Aldh1a1 axis improves the vascular microenvironment, suggesting tissue-specific Ffar4 agonism could prevent DKD.

In summary, we establish Ffar4 as a gatekeeper of DKD progression via Atf4–Aldh1a1-directed RA synthesis. Targeting this axis presents novel therapeutic opportunities for DKD management.

Abbreviations

DKD

Diabetes kidney disease

GECs

Glomerular endothelial cells

GPCRs

G protein-coupled receptors

FFARs

Free fatty acid receptors

LCFAs

Long-chain fatty acids

STZ

Streptozotocin

VCAM1

Vascular cell adhesion molecule 1

MR

Mendelian randomization

FBG

Fasting blood glucose

Aldh1a1

Aldehyde dehydrogenase 1 family member A1

RA

Retinoic acid

NR

Nuclear receptor

RAR

Retinoic acid receptor

Atf4

Activating transcription factor 4

CUT&RUN

Cleavage under targets and release using nuclease

ITT

Intraperitoneal glucose tolerance test

GTT

Oral glucose tolerance test

24H UPE

24-H urinary protein

CRE

Serum creatinine

UACR

Urinary albumin/creatinine ratio

EKO

Endothelial-specific Ffar4 knockout mice

EOE

Endothelial-specific Ffar4 overexpressing mice

TKO

Kidney tubule-specific Ffar4 KO mice

TNFR1

Tumor necrosis factor receptor 1

HG-Pal

High sugar and high palmitic acid

Author contributions

S.Z. and J.L. designed the experiments; X.L. and X.Y. provided a lot of technical support and suggestions for modifications; J.L., T.Z., W.W. and Z.W. conducted the experiments; J.L., W.W. and S.C. analysed the data; J.L., Y.Q.C. and S.Z. wrote the paper. All authors read and approved the final manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Data availability

GWAS date were derived from the following resources available in the public domain: (UK Biobank ID: finn − b − DM_NEPHROPATHY, eQTLGen database and eqtl-a-ENSG00000186188). Single-cell sequencing plot of the kidney interactive transcriptome were derived from the following resources available in the public domain:(Visualization available at Kidney Interactive Tranomics database, Healthy Mouse Dataset: Wu et al., JASN 2019; and RBK RID: 14-4KBC). RNA-seq data have been deposited in the CNCB (China National Center for Bioinformation) under accession number CRA032309. All data generated or analysed during this study are included in this published article. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The project was conducted in compliance with the guidelines outlined in the Declaration of Helsinki and the Basel Declaration. Studies involving human tissues were approved by the Ethics Committee of Jiangnan University Medical Center (permission number: Ethics Review No. Y-125; date issued: 15 January 2021). The study was performed in accordance with the principles of the Declaration of Helsinki. The animal experiments were performed in accordance with the Basel Declaration and were approved by the Animal Ethics Committee of Jiangnan University (permission number: JN20211215c0500630; date issued: 15 December 2021). The ethics committee follows the guidelines of the International Council for Laboratory Animal Science (ICLAS) to ensure the ethical compliance of the experiments.

Consent for publication

The manuscript has been approved by all the authors.

Competing interests

The authors declare no competing interests.