Skip to main content
Download PDF
Download ePub

SETD7 drives diabetic endothelial dysfunction through FBXO45-mediated GPX4 ubiquitylation

Cardiovascular Diabetology volume 24, Article number: 178 (2025) Cite this article

Abstract

Background

Vasculopathy is the most prevalent complication of diabetes. Endothelial damage, a primary contributor to hyperglycemic vascular complications, impacts macro- and micro-vasculatures, causing functional impairment of multiple organs. SETD7 was initially identified as a transcriptional activator based on its ability to methylate histone 3 lysine 4. However, its function in the context of diabetic endothelial dysfunction remains poorly understood. This study aims to elucidate the involvement and underlying mechanisms of SETD7 in diabetic endothelial dysfunction.

Methods

SETD7 knockout mice were generated to investigate the effects of SETD7 on Streptozotocin (STZ)-induced hyperglycemia and vascular endothelial injury. Endothelial-specific SETD7 interruption adeno-associated virus (AAV) system was utilized to investigate the effects of SETD7 on diabetic vascular endothelial injury in BKS-DB(Lepr) KO/KO (db/db) mice. In vitro manipulation of SETD7 activation or knockdown was conducted to assess its regulation on the lipid peroxidation, oxidative stress, and cell function of primary rat aortic endothelial cells (RAECs) under high glucose conditions.

Results

Our study revealed that knockout and endothelial deficiency of SETD7 partially restored damaged vascular function and attenuated the inflammatory response caused by high glucose in both STZ-induced and db/db mice. Moreover, SETD7 activation aggravated oxidative stress injury and resulted in profound dysfunction through Glutathione Peroxidase 4 (GPX4)-mediated lipid peroxidation in RAECs. Mechanistically, SETD7 deficiency reduced p53 mono-methylation and blocked FBXO45 transcription, thereby inhibiting the protein degradation of GPX4 and subsequent lipid peroxidation as well as oxidative stress.

Conclusions

In summary, our study demonstrates that SETD7-p53-FBXO45-GPX4 is involved in high glucose-induced oxidative stress injury and exacerbated endothelial dysfunction, which offering great significance for mitigating hyperglycemia-induced endothelial damage.

Graphic abstract

Introduction

Diabetes is a severe chronic metabolic disease characterized by persistent hyperglycemic condition accompanied by vascular complications [1]. Endothelial cells (ECs) play a crucial role in maintaining vascular endothelial homeostasis through regulating vascular morphology, function, and metabolism [2]. Therefore, high glucose-induced dysfunction of the vascular endothelium serves as an initiating factor and hallmark for a wide range of diabetic vascular complications [3,4,5].

Hyperglycemia-induced vascular endothelial dysfunction involves a complex regulatory process, including defective endothelial repair, inability of angiogenesis, oxidative stress, inflammation, metabolic abnormalities [6, 7]. Typically, elevated intracellular glucose levels lead to endothelial metabolic disorder and subsequent overproduction of reactive oxygen species (ROS) [8]. Excessive ROS is thought to activate nuclear factor-κB (NF-κB) signaling pathway, resulting in release of pro-inflammatory cytokines such as vascular cell adhesion molecule-1 (VCAM-1), intracellular adhesion molecule-1 (ICAM-1), and interleukin (IL)-1β, IL-6 [9]. Additionally, the abnormal oxidative environment in diabetes destroys endothelial function by causing nitric oxide (NO) deficiency due to dysfunctional endothelial nitric oxide synthase (eNOS) [10].

GPX4 (Glutathione Peroxidase 4), a pivotal antioxidant enzyme, utilizes reduced glutathione (GSH) to convert lipid hydroperoxides into their corresponding alcohols and free hydrogen peroxide to water, thereby preserving the integrity of cellular components [11]. Inhibition of GPX4 has been associated with increased generation of ROS and suppressed expression of eNOS, resulting in endothelial dysfunction in atherosclerosis [12, 13]. Interestingly, high glucose triggered a ubiquitination-mediated protein degradation of GPX4 in retinal vascular endothelial cells [14]. This observation highlights the vulnerability of GPX4 to hyperglycemic conditions, leading to a decreased antioxidative capacity and heightened endothelial dysfunction.

SETD7 belongs to the SET (Su (var), Enhancer of zeste, Trithorax) family of histone methyltransferases. SETD7 is traditionally recognized for its role in catalyzing mono- and di-methylate histone3 lysine 4 (H3K4), thereby promoting the transcriptionally favorable open chromatin conformation [15]. However, SETD7 has also been implicated in modifying various non-histone proteins [16]. SETD7 was observed to be upregulated in peripheral blood mononuclear cells of type 2 diabetes patients and correlated the levels with fasting plasma glucose [17, 18]. Specifically, in vascular endothelial cells, glucose spikes have been linked to SETD7 activation and the subsequent chromatin modification via H3K4me1, which act through the NF-κB pathway to mediate endothelial inflammation [19].

This study aimed to explore the role of endothelial SETD7 in a high glucose environment by generating SETD7 deficient mice (Setd7−/−) and using adeno-associated virus (AAV)-based conditional knockdown system specific to endothelial cells. In the present study, we observed a predominant upregulation of SETD7 expression in the aorta of diabetic mice. Additionally, SETD7 deficiency alleviated the diabetic endothelial dysfunction, and attenuated the oxidative stress and inflammatory response in aortic tissue and endothelial cells. Our findings revealed that aberrant SETD7 activation drives oxidative stress injury and leads to severe dysfunction by mediating GPX4 ubiquitination degradation in endothelial cells. Hence, SETD7 represents a new intervention target to prevent endothelial dysfunction in diabetes.

Methods

Animals protocols

The animal studies were conducted following the ethical guidelines of the Institutional Ethics Committee of Fudan University, China. SETD7 deficient mice (Setd7−/−) were generated using CRISPR/Cas9 to knockout the Setd7 gene (Shanghai Model Organisms Center, Inc.). The control wild type mice (WT) for all experiments in this study were littermates. To induce diabetes, both male WT or Setd7−/− mice were intraperitoneally injected with low-dose of Streptozocin (STZ, 50 mg/kg) daily for five consecutive days.

Additionally, diabetic BKS-DB(Lepr) KO/KO (db/db) mice and BKS-DB(Lepr) WT/WT (WT/WT) were acquired from Gem Pharmatech (Jiangsu, China). To generate mice with Setd7 conditional deletion in ECs, we used adeno-associated virus (AAV) vectors carrying short hairpin RNA against murine Setd7 mRNA under the control of the murine endothelial-specific TIE (TEK receptor tyrosine kinase) promoter (HBAAV2/9-TIE-mir30-m-Setd7-ZsGreen, AAV-shSetd7) [20], designed and synthesized by Hanbio (Shanghai, China). At eight weeks of age, the male db/db mice were divided into two groups. One group received AAV-shSetd7, while the other group was injected with a negative control (HBAAV2/9-TIE-ZsGreen, AAV-Ctr). Both groups received the treatments via tail-vein injections at a dosage of 1 × 1012 vg. Control group mice were injected with AAV-Ctr accordingly. All mice were given free access to food and water, and all animals’ blood glucose levels were regularly monitored over the subsequent 14 weeks.

Vasoreactivity study in wire myograph

After collection from mice, aortae were immediately immersed in ice-cold Krebs-Ringer solution (Krebs buffer) at pH 7.4 to cleanse off fat and connective tissues. Arterial segments were cut into 2 mm rings and mounted on stainless steel wires within a myograph system (Danish Myo Technology, Denmark), maintained at 37 °C and aerated with 95% O2/5% CO2. The arterial rings underwent a 30 min equilibration, followed by 3 mN basal tension. After a subsequent 1 h of equilibration, rings were firstly contracted by 60 mM KCl until their contraction to KCl reached the maximum. For relaxation studies, arterial rings were washed to the baseline, then contracted by 1 µmol/L phenylephrine (PE) to a plateau. Endothelium-dependent relaxations (EDRs) induced by acetylcholine (Ach) were recorded.

Rat aortic endothelial cells (RAECs) isolation and culture

Aortas excised from euthanized mice were immediately submerged in ice-cold PBS, cleaned to remove excess fat and lateral vessels, then digested with type II collagenase at 37 °C for 1 h. Post-digestion, the aortas were sectioned into 2 mm segments and cultured in endothelial cell medium (ECM) enhanced with 10% FBS, with medium changes every 2–3 days for one week. Following this, endothelial cells that detached were collected by centrifugation, cultured in low glucose DMEM with 10% FBS, 1 mg/mL glucose, and 1% penicillin-streptomycin until they reached confluency. Cells were exposed to different glucose concentrations (5.5 mM, 11 mM, or 22 mM), and mannitol was added to adjust the corresponding osmotic pressure.

Transwell assay and wound healing assay

In the Transwell assay, 4 × 104 RAECs in serum-free medium were placed into a Transwell chamber (Corning Life Sciences, USA) above a well containing the same medium with 10% FBS. After 48 h, cells from the upper chamber were gently removed, and those adhered to the membrane were fixed with methanol and stained with hematoxylin. The migration cells were captured using a microscope and quantified by dissolving and detecting the absorbance value.

For the wound healing assay, 5 × 105 RAECs were seeded into six-well plates and incubated overnight. A straight scratch was then made, floating cells were removed, and the wells were washed three times with PBS. The cells were subsequently cultured in 2% FBS medium. Images taken at the same position at 0-, 12-, and 24-h post-scratch were used to assess cell migration by measuring the area healed relative to the initial wound at 0 h.

ROS detection and JC-10 staining

To evaluate ROS levels, RAECs were plated in twelve-well plates and allowed to adhere overnight. After this, cells were incubated with 2,7-dichlorodi-hydro fluorescein diacetate (0.5 µM, DCFH-DA, Yeasen, China), diluted in 500 µL of serum-free medium. This mixture was incubated at 37 °C for 10 min in darkness, and were washed with PBS. The ROS levels were then assessed using fluorescence microscopy to visualize the fluorescence.

To evaluate mitochondrial membrane potential, RAECs were seeded in twelve-well plates and incubated with JC-10 staining solution (5 µg/mL) (Beyotime, China) of 500 µL serum-free medium, incubated at 37 °C for 20 min in the dark, and washed with PBS. Relative amounts of mitochondrial JC-10 monomers or aggregates were measured by fluorescence microscopy using 488 nm and 561 nm lasers. The ratio of green/red fluorescence intensity of JC-10 was normalized to access the loss of mitochondrial membrane potential.

Malondialdehyde (MDA), glutathione peroxidase (GSH-Px), and superoxide dismutase (SOD) detection

The levels of MDA were quantitatively measured using MDA assay kit (Abcam, Cambridge, UK). The treated cells were ultrasonically broken in MDA lysis buffer with BHT, and centrifuged at 13,000 g for 10 min to collect supernatant. Developer VII/TBA reagent were added and incubated at 95 °C water bath for 60 min. The absorbance at 532 nm was detected to quantitatively determine the intracellular MDA content.

The levels of GSH-Px and SOD activity were quantitatively measured using GSH-Px and SOD assay kit (Nanjing Jiancheng Bioengineering Institute, China). The treated cells were ultrasonically broken in RIPA lysis buffer, and the corresponding detection solutions were added according to the manufacturer’s instructions. The subsequent absorbance measurements at 412 nm and 450 nm enabled the quantitative evaluation of GSH-Px and SOD activity.

Aortic ring sprouting assay

Segments of thoracic aorta were meticulously isolated and sectioned into uniform rings approximately 1 mm in width. Each ring was placed in a 96-well plate with 40 µL of Matrigel to anchor the ring, followed by an additional 40 µL of Matrigel to fully encapsulate it. To each well, 100 µL of complete extracellular matrix (ECM) medium were added in the presence and absence of high glucose (22 mM). The culture medium was refreshed every 3 days. Sprouting from the aortic rings was monitored and documented using microscopy from 7 to 10 days.

Dual-luciferase reporter assay

HEK293T cells were plated in 6-well plates and transfected with p53 or pGL6-CMV-Luc-FBXO45 plasmid, using Lipofectamine 2000 liposomes (ThermoFisher Scientific, MA, USA) as per the manufacturer’s guidelines, pGL6-CMV-Luc served as negative control. 48 h after transfection, luciferase reporter gene activity was assessed with a dual luciferase assay kit (Beyotime, Shanghai, China). The transfected cells were lysed and centrifuged to collect supernatant. Take 100 µL sample lysis to add 100 µL of firefly luciferase detection reagent, and then measure the relative light unit (RLU). After determine firefly luciferase, add 100 µL the renilla luciferase assay detection reagent, to determine the RLU. The RLU value determined by renilla luciferase was used as the internal reference.

Statistical analysis

All experiments were conducted a minimum of three times. Data are presented as mean ± SEM. Statistical differences were assessed with the unpaired 2-tailed Student’s t test for two experimental groups and one-way ANOVA for multiple groups. A p value of less than 0.05 was considered statistically significant. Statistical analyses were performed using GraphPad Prism.

Results

Identification of high glucose-regulated SETD7 in diabetic vessels

To investigate the involvement of SETD7 in diabetic stress, STZ-induced mice and db/db mice were utilized as diabetic murine models. Notably, an upregulation of SETD7 expression was observed in aorta from both STZ-induced mice and db/db mice (Fig. 1A). To investigate the endothelial injury under hyperglycemic conditions in vitro, RAECs were incubated with 5.5 mM glucose plus 16.5 mM mannitol (osmotic control, Control), or 22 mM D-glucose (high glucose, HG) for 72 h. Transwell and wound healing assays were employed to assess the effect of high glucose on cell migration, revealing a significant reduction in the migratory capacity of RAECs upon exposure to HG (Fig. 1B, C). In line with the in vivo observations, Setd7 mRNA expression was increased in HG-induced cells (Fig. 1D), accompanied by a time- and concentration-dependent upregulation of its protein levels following HG exposure (Fig. 1E, F). Furthermore, immunofluorescence (IF) double staining provided additional evidence that HG enhanced the expression of SETD7 and inflammatory marker COX-2 (Fig. 1G). Together, these results demonstrate that SETD7 expression is upregulated in diabetic aorta and HG-induced RAECs, indicating the potential involvement of SETD7 in the pathogenesis of diabetic endothelial injury.

Fig. 1
figure 1

Identification of high glucose-regulated SETD7 in diabetic vessels. A The protein expression of SETD7 in aortic tissues from STZ-induced mice (n = 5) and db/db mice (n = 5). B Representative images of wound healing assay photographed at 0, 12, and 24 h of culture (100 ×). C Representative images of the Transwell assay photographed after 36 h of culture (200 ×). D The mRNA level of Setd7 in high glucose-induced RAECs (33 mM D-glucose for 24 h). E The protein expression of SETD7 in RAECs treated with or without high glucose concentrations (11 mM, 22 mM) for 72 h. F The protein expression of SETD7 in RAECs treated with or without 22 mM high glucose for 36–72 h. G Representative images of immunofluorescence staining of SETD7 and COX-2 in RAECs treated with or without high glucose. Scale bars, 50 μm. Statistical significances were calculated using (A–D) two tail student’s t test, (E, F) one-way ANOVA, Tukey’s multiple comparisons tests. Data are expressed as the mean ± SEM, statistical analysis revealed a significant difference with *p < 0.05, **p < 0.01, and ***p < 0.001

Full size image

SETD7 deficiency alleviates STZ-induced endothelial impairment

To investigate the relevance of SETD7 deficiency in maintaining normal vasodilator function and diabetes-induced endothelial dysfunction, Setd7−/− and WT mice were induced by STZ. Firstly, the protein expression of SETD7 was decreased in aorta from knockout diabetic mice compared to the WT diabetic mice, as determined by western blot and IF analyses (Fig. 2A, B). Subsequently, we conduced vasoreactivity study using wire myograph to detect the acetylcholine (Ach)-induced endothelium-dependent relaxations (EDRs) and observed impaired EDRs in STZ-induced mice thoracic aorta, in contrast, Setd7−/− mice exhibited preserved endothelium-dependent vasorelaxation (Fig. 2C, D). Meanwhile, IF stanning of CD31 revealed hyperglycemia-induced aortic de-endothelialization was dramatically increased in WT mice, while SETD7 deficiency partly recovered damaged endothelium (Supplementary Fig. 1A). Accordingly, knockdown of SETD7 reversed the down-regulation of eNOS, VE-cadherin, and VEGF-α in aorta with endothelial dysfunction (Fig. 2E), and decreased the up-regulation of vascular inflammation (VCAM-1, iNOS, COX-2, IL-6, and IL-1β) (Supplementary Fig. 1B). To further elucidate the functional role of SETD7 in regulating ECs function, we examined the vessel sprouting using aortic rings from Setd7−/− mice and their WT littermates. As shown in Fig. 2F, high glucose treatment significantly impaired the aortic ring sprouting and this impairment was mitigated in aortic rings obtained from SETD7 deficiency mice. Collectively, these results suggest that the persistent activation of SETD7 under high glucose environment continues to aggravate vascular endothelial cell damage, whereas knockout of SETD7 partially alleviates the damage.

Fig. 2
figure 2

SETD7 deficiency alleviates STZ-induced endothelial impairment. WT and Setd7−/− mice were administrated with 50 mg/kg/d STZ or sodium citrate by intraperitoneal injection for five consecutive days. A The protein expression of SETD7 in aortic tissues of WT and Setd7−/− mice (n = 4). B Representative images of immunofluorescence SETD7-stained thoracic aorta sections isolated from WT and Setd7−/− mice. Scale bars, 50 μm. C Representative traces of acetylcholine (Ach)-induced endothelium-dependent relaxations (EDRs) in aortae from WT and Setd7−/− mice. D Concentration-response curves of Ach-induced EDRs in aortae from WT and Setd7−/− mice (n = 6). E The protein expression of eNOS, VE-cadherin, and VEGF-α in aortic tissues of WT and Setd7−/− mice (n = 4). F Representative images of the new vessel sprouting from the isolated thoracic aortic rings of WT and Setd7−/− mice in the presence and absence of high glucose (HG, 22 mM) for 7–10 days (n = 3). The new vessel sprouting was photographed and quantitated at 7, and 10 days of culture (40 ×). Statistical significances were calculated using (A) one-way ANOVA, (D-F) two-way ANOVA, Tukey’s multiple comparisons tests. Data are expressed as the mean ± SEM, statistical analysis revealed a significant difference with **p < 0.01 and ***p < 0.001, #p < 0.05, ##p < 0.01, ##p < 0.01, and ###p < 0.001

Full size image

Conditionally endothelial SETD7-deficient alleviates endothelial function in Db/db mice

To further investigate the role of endothelial SETD7 in endothelial dysfunction, we have generated SETD7 endothelial specific AAV2/9 system using a construct containing TIE promoter fused with murine Setd7 shRNA (AAV-shSetd7). As depicted in Fig. 3A, AAV-shSetd7 decreased the expression of SETD7 in aortic endothelial cells compared to db/db mice that received an AAV negative Control (AAV-Ctr). Conversely, the expression of SETD7 in aortic smooth muscle cells did not show obvious changes (Supplementary Fig. 1C). EDRs, which was determined in the presence of Ach, exhibited reduced levels in aortas of db/db mice as compared to non-diabetic mice, however, the administration of AAV-shSetd7 significantly improved EDRs in aortas of diabetic mice (Fig. 3B, C). CD31 IF staining also exhibited a repaired endodermal integrity (Supplementary Fig. 1D). Moreover, specific deficiency of SETD7 in endothelial cells led to recovery of eNOS and VEGF-α expression levels (Fig. 3D) and reduction of vascular inflammation markers (iNOS, COX-2, IL-6, and IL-1β) (Supplementary Fig. 1E). Consistently, vessel sprouting assay, using aortic rings from WT mice treated or untreated with AAV-shSetd7, showed that SETD7 endothelial-specific knockdown redeemed the impaired aortic ring sprouting under high glucose conditions (Fig. 3E).

Fig. 3
figure 3

Conditionally endothelial SETD7-deficient alleviates endothelial function in db/db mice. Diabetic BKS-DB(Lepr) KO/KO (db/db) mice and BKS-DB(Lepr) WT/WT (WT/WT) were administrated with AAV-shSetd7 or AAV-Ctr by tail-vein injection to specifically knockdown endothelial SETD7. A Representative images of immunofluorescence SETD7-stained thoracic aorta sections isolated from WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr. Scale bars, 50 μm. B Representative traces of Ach-induced EDRs in aortae from WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr. C Concentration-response curves of Ach-induced EDRs in aortae from WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr (n = 6). D The protein expression of eNOS and VEGF-α in aortic tissues of WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr (n = 4). E Representative images of the new vessel sprouting from the isolated thoracic aortic rings of WT mice received AAV-shSetd7 or AAV-Ctr in the presence and absence of high glucose (HG, 22 mM) for 10 days (n = 3). The new vessel sprouting was photographed and quantitated at 7, and 10 days of culture (40 ×). F Representative images of H&E-stained and fibrinogen-stained retinal cross-sections of WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr. Scale bars, 50 μm. G Representative images of the capillaries in trypsin-digested retinas and quantification of pericytes to endothelial cells ratio (PE/EC, n = 5). Scale bars, 20 μm. H The mRNA level of Setd7, Vegfa, and Aqp4 in retinas from WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr. Statistical significances were calculated using (G) one-way ANOVA, (C–E, H) two-way ANOVA, Tukey’s multiple comparisons tests. Data are expressed as the mean ± SEM, statistical analysis revealed a significant difference with ***p < 0.001, ##p < 0.01, ###p < 0.001, ns = not significant

Full size image

Diabetic retinopathy (DR) is another diabetic-induced microangiopathy, which is characterized by retinal vascular dysfunction, such as vascular leakage, retinal oedema and microaneurysms [21]. Figure 3F illustrated a representative Haematoxylin and eosin (H&E) staining map of the retina, revealing that AAV-shSetd7 intervention significantly reduced the ratio of retinal positive areas to whole positive areas. Furthermore, targeted knockdown of endothelial-specific SETD7 diminished retinal leakage as evidenced by fibrinogen staining (Fig. 3F). According to examination of the retinal vascular pavement, AAV-shSetd7 considerably decreased acellular capillary formation and increased the pericytes to endothelial cells ratio (Fig. 3G). VEGF-α and AQP4 play a critical role in vascular permeability. Compared with diabetic group, our results demonstrated that Setd7 knockdown significantly decreased the levels of Vegfa and Aqp4 expression (Fig. 3H). Taken together, these results suggest that endothelial SETD7 deficiency confers protection in maintaining vascular function of diabetic mice.

SETD7 regulates HG-induced endothelial injury in vitro

To illuminate the role of SETD7 in endothelial dysfunctions in vitro, we assessed the impact of SETD7 silencing in HG-treated ECs. Silencing SETD7 resulted in a remarkable alteration in SETD7 expression level, as well as endothelial function markers (eNOS, VE-cadherin, and VEGF-α) (Fig. 4A) and inflammation markers (VCAM-1, ICAM-1, iNOS, COX-2, IL-6, and IL-1β) (Supplementary Fig. 2A). Furthermore, mRNA levels of Setd7, Vcam1 and Enos were also altered in ECs where SETD7 was silenced (Fig. 4B). Similarly, IF staining indicated that SETD7 knockdown resulted in reduced expressions of both SETD7 and COX-2 (Fig. 4C). Then, the impact of inhibiting SETD7 on HG-induced endothelial injury was investigated. Prior to exposure to HG, RAECs were pretreated with or without sinefungin, a SETD7 inhibitor. In line with the observed outcomes related to SETD7 deficiency, sinefungin significantly protected against the less of eNOS, VE-cadherin, and VEGF-α expression and reduced the elevated levels of inflammatory mediators (VCAM-1, ICAM-1, iNOS, COX-2, IL-6, and IL-1β) (Supplementary Fig. 2B, C).

Fig. 4
figure 4

SETD7 regulates HG-induced endothelial injury in vitro. RAECs were cultured in different glucose concentration conditions (Control: 5.5 mM glucose + 16.5 mM mannitol, HG: 22 mM glucose) with or without siScr or siSetd7 treatment. A The protein expression of SETD7, eNOS, VE-cadherin, and VEGF-α in siScr and siSetd7 RAECs with or without HG stimulation. B The mRNA level of Setd7, Vcam1, and Enos in siScr and siSetd7 RAECs with or without HG stimulation. C Representative images of immunofluorescence staining of SETD7 and COX-2 in siScr and siSetd7 RAECs with or without HG stimulation. Scale bars, 50 μm. D Heatmap showing differentially expressed genes (DEGs) in siScr and siSetd7 RAECs with or without HG stimulation. E Venn diagram showing the identified proteins in Control with siScr vs. HG with siScr (blue, 378), HG with Setd7 vs. HG with siScr (purple, 300). A total of 78 genes were identified in three groups. F GO-based enrichment analysis of 78 DEGs in terms of the biological processes. G Representative images of wound healing assay photographed at 0, 12, and 24 h of culture (100 ×). H Representative images of the Transwell assay photographed after 36 h of culture (200 ×). RAECs were transfected with vector (Vector) or Setd7 overexpression plasmid (Setd7oe). I The protein expression of Myc-SETD7, eNOS, VE-cadherin, and VEGF-α in RAECs transfected with Vector or Setd7oe. J Representative images of wound healing assay photographed at 0, 12, and 24 h of culture (100 ×). K Representative images of the Transwell assay photographed after 36 h of culture (200 ×). Statistical significances were calculated using (A, B, G, I and J) two-way ANOVA, Tukey’s multiple comparisons tests, (H) one-way ANOVA, (K) two tail student’s t test. Data are expressed as the mean ± SEM, statistical analysis revealed a significant difference with *p < 0.05, **p < 0.01, ***p < 0.001, ###p < 0.001, ns = not significant

Full size image

To further investigate SETD7 mediation on the diabetic endothelial dysfunction, RAECs were subjected to RNA sequencing analysis (RNA-seq). The PCA and heat map revealed distinct differences between the groups, with samples within each biological repeat clustering together, indicating good reproducibility within the groups (Supplementary Fig. 2D, Fig. 4D). The Venn diagram of different groups of differential genes displayed a shared set of 78 differentially expressed genes (DEGs) among all three groups (Fig. 4E). KEGG enrichment analysis of these DEGs showed that differential gene enrichment was in the cell motility pathway (Fig. 4F). Thus, consistent with the endothelial cell damage caused by hyperglycemia in vivo, Transwell and wound healing assays confirmed that RAECs cultured in high glucose media exhibited impaired cell migration ability, which was significantly restored upon SETD7 silencing (Fig. 4G, H).

We subsequently investigated the potential of SETD7 to drive endothelial cell injury. Lentiviral-mediated overexpression of SETD7 was utilized to enhance its expression in RAECs. SETD7 upregulation significantly elevated the expression levels of the inflammation markers (VCAM-1, ICAM-1, iNOS, COX-2, IL-6, and IL-1β) (Supplementary Fig. 2E, F) and decreased indicators of endothelial function (eNOS, VE-cadherin, and VEGF-α) (Fig. 4I). Additionally, activation of SETD7 had detrimental effects on cell migration as detected by Transwell and wound healing assays (Fig. 4J, K). Collectively, these results suggest that blocking SETD7 suppresses the exacerbation of endothelial dysfunction and concomitant inflammation induced by high glucose. Conversely, upregulation of SETD7 alone is sufficient to induce diabetes-associated endothelial dysfunction and inflammation in RAECs.

Loss of endothelial SETD7 reverses GPX4-mediated lipid peroxidation and oxidative stress

Oxidative stress is a key factor in endothelial dysfunction under hyperglycemia, and impaired mitochondrial function in endothelial cells leads to the accumulation of intracellular ROS [8]. HG significantly increased intracellular ROS levels in RAECs, in contrast, siSetd7 notably reduced the production of ROS (Fig. 5A). Next, mitochondrial depolarization was monitored by JC-10 staining. The results demonstrated that siSetd7 led to a recovery in mitochondrial membrane potential as indicated by the decrease in green fluorescence intensity (Fig. 5B). To further evaluate the altered oxidative status of RAECs, we examined malondialdehyde (MDA) content, glutathione peroxidase (GSH-Px) and superoxide dismutase (SOD) activity. HG-incubated RAECs exhibited elevated MDA levels and decreased GSH-Px and SOD activity which were reversed upon SETD7 knockdown (Fig. 5C), suggesting the involvement of SETD7 in diminishing antioxidant capacity and enhancing peroxidative capacity in endothelial cells with hyperglycemia.

Fig. 5
figure 5

Loss of endothelial SETD7 reverses GPX4-mediated lipid peroxidation and oxidative stress. RAECs were cultured in different glucose concentration conditions (Control: 5.5 mM glucose + 16.5 mM mannitol, HG: 22 mM glucose) with or without siScr or siSetd7 treatment. A Representative images showing ROS production. Scale bars, 100 μm. B Representative images of membrane potential detected by JC-10 fluorescence staining. Scale bars, 100 μm. C MDA level, GSH-Px activity, and SOD activity of siScr and siSetd7 RAECs with or without HG stimulation. D The protein expression of GPX4, NOX4, and xCT in siScr and siSetd7 RAECs with or without HG stimulation. E Representative images of immunofluorescence staining of GPX4 in siScr and siSetd7 RAECs with or without HG stimulation. Scale bars, 50 μm. RAECs were transfected with vector (Vector) or Setd7 overexpression plasmid (Setd7oe). F: Representative images of ROS production and JC-10 fluorescence staining. Scale bars, 100 μm. G MDA level and SOD activity of RAECs transfected with Vector or Setd7oe. H The protein expression of GPX4 and NOX4 in RAECs transfected with Vector or Setd7oe. Statistical significances were calculated using (A–C) one-way ANOVA, (D) two-way ANOVA, Tukey’s multiple comparisons tests, (F–H) two tail student’s t test. Data are expressed as the mean ± SEM, statistical analysis revealed a significant difference with **p < 0.01, ***p < 0.001, #p < 0.05, ##p < 0.01, ###p < 0.001, ns = not significant

Full size image

Considering the established role of GPX4 in mitigating oxidative stress, we further verify that SETD7 modulates endothelial dysfunction potentially via GPX4-mediated lipid peroxidation and oxidative stress. The GPX4 expression was assessed in HG-induced RAECs, revealing that high glucose exposure downregulated GPX4 expression and upregulated NOX4 expression (Supplementary Fig. 3A, B). However, siSetd7 treatment repaired the decreased GPX4 expression and the increased NOX4 expression (Fig. 5D, E). GPX4 plays a master role in blocking ferroptosis [22], we subsequently explored whether ferroptosis could be implicated in SETD7-mediated endothelial dysfunction. In this study, high glucose conditions did not significantly induce cell death as measured by CCK8 assays. Moreover, silencing SETD7 had no impact on cell viability (Supplementary Fig. 3C). Furthermore, Fer-1 (a ferroptosis inhibitor) had no effect on restoring GPX4 levels (Supplementary Fig. 3D), showing the notion that SETD7 mediated endothelial dysfunction under high glucose conditions through ferroptosis was disproven.

Inhibiting SETD7 also resulted in a reduction of the intracellular ROS levels in HG-induced RAECs, as well as a significant reversal of the decreased expression of GPX4 (Supplementary Fig. 3E, F). Conversely, overexpression of SETD7 increased the production of ROS and mitochondrial depolarization, monitored through ROS production and JC-10 staining (Fig. 5F). Similarly, activation of SETD7 resulted in higher MDA levels and lower SOD activity (Fig. 5G). GPX4 expression downregulation and NOX4 upregulation were also observed in cells overexpressing SETD7 (Fig. 5H).

Administration of RSL3, a ferroptosis agonist, induces ferroptosis by inactivating GPX4 [14]. Here, we used a low dose of RSL3 (500 nM) to avoid inducing cell death while inhibiting GPX4 activity. RSL3 treatment effectively blocked the expression of GPX4, eNOS and VEGF-α, but concurrently promoted vascular inflammation markers (iNOS, COX-2, IL-6, and IL-1β). Nevertheless, these effects were restrained by siSetd7 (Supplementary Fig. 4A). Additionally, siSetd7 enhanced the cell migration ability when co-incubated with RSL3 (Supplementary Fig. 4B, C). Notably, oxidative stress and lipid peroxidation levels, along with MDA content, GSH-Px and SOD activity measurement demonstrated that the impact of RSL3 stimulation was partially inhibited by siSetd7 intervention (Supplementary Fig. 4D-F). In the presence of HG, RSL3 stimulation partly reversed the protective effect of siSetd7 on oxidative stress and inflammation (Supplementary Fig. 4G). These findings suggest that SETD7 modulates endothelial dysfunction through GPX4-mediated lipid peroxidation and oxidative stress, highlighting the central role of GPX4 in the progression.

FBXO45 interacts with GPX4 to promote GPX4 ubiquitination

FBXO45 contains a conserved F-box domain and a SPRY domain, which facilitate substrate recruitment to the ubiquitin ligase complex for initiating protein degradation [28]. Firstly, high glucose levels were found to increase both mRNA and protein expressions of FBXO45 in RAECs, while upregulated expression of FBXO45 was also observed in the aorta of STZ-induced mice and db/db mice (Supplementary Fig. 5A-C). SETD7 was identified to promote FBXO45 transcription and protein expression, as detected by RT-qPCR, western blot and IF assay (Supplementary Fig. 5D-H). We further verified the involvement of FBXO45 in endothelial dysfunctions in vitro. Silencing FBXO45 displayed a protective effect on endothelial function. Treatment with siFbxo45 effectively alleviated migration damage in HG-induced cells as measured by wound healing and Transwell assays (Supplementary Fig. 6A, B). Additionally, deletion of FBXO45 attenuated oxidative stress and lipid peroxidation levels, as demonstrated through several indicators: reduced ROS and JC-10 staining positivity, lower MDA level, along with increased activities of GSH-Px and SOD in HG-induced RAECs (Supplementary Fig. 6C, D). Meanwhile, siFbxo45 suppressed vascular inflammation markers (iNOS, COX-2, IL-6, and IL-1β) and enhanced the expressions of eNOS, VE-cadherin and VEGF-α (Supplementary Fig. 6E, F). The mRNA level determination revealed reduced Nos2 and Nox4 expression in HG-induced cells treated with siFbxo45 (Supplementary Fig. 6G). These results suggest that FBXO45 is involved in SETD7 regulated endothelial dysfunctions and silencing FBXO45 displays a protective effect on endothelial function.

Importantly, overexpression of FBXO45 resulted in a decrease in GPX4 protein levels without affecting its mRNA levels (Supplementary Fig. 7A, B). Meanwhile, siFbxo45 did not affect the mRNA levels of GPX4 in HG-induced RAECs (Fig. 6A), but western blot and IF assays suggested that siFbxo45 up-regulated the protein levels of GPX4 (Fig. 6B, C). We hypothesized that FBXO45 induces ubiquitination of GPX4. Firstly, co-immunoprecipitation (Co-IP) analysis demonstrated an interaction between FBXO45 and GPX4 using the HG-induced RAECs and HEK293T cells co-transfected with FBXO45 and GPX4 plasmid (Fig. 6D, E). Next, we investigated whether knockdown of FBXO45 enhances the stability of GPX4 protein by employing Cycloheximide (CHX), a protein synthesis inhibitor. The HEK293T cells were transiently transfected with GPX4 expression plasmid and FBXO45 expression plasmid, followed by treatment with 200 µg/mL CHX for different times to inhibit protein synthesis in the presence or absence of siFbxo45. The degradation rates of the existing GPX4 protein were measured by western blot analysis. The results showed that overexpression of FBXO45 strengthened GPX4 degradation, while siFbxo45 recruited the reservation of GPX4 expression (Fig. 6F). To further evaluate the relationship between regulation of GPX4 protein stability by FBXO45 and the proteasome system, we used MG132, a 26 S proteasome inhibitor. Remarkably, FBXO45-mediated GPX4 downregulation was blocked by pretreatment with the MG132, whereas no such effect was observed with 3-MA (an autophagy inhibitor) or NH4Cl (a lysosome inhibitor) in RAECs (Supplementary Fig. 7C).

Fig. 6
figure 6

FBXO45 interacts with GPX4 to promote GPX4 ubiquitination. RAECs were cultured in different glucose concentration conditions (Control: 5.5 mM glucose + 16.5 mM mannitol, HG: 22 mM glucose) with or without siScr or siFbxo45 treatment. A The mRNA level of Fbxo45 and Gpx4 in siScr and siFbxo45 RAECs with or without HG stimulation. B Representative images of immunofluorescence staining of GPX4 in siScr and siFbxo45 RAECs with or without HG stimulation. Scale bars, 50 μm. C The protein expression of FBXO45 and GPX4 in siScr and siFbxo45 RAECs with or without HG stimulation. D Co-IP assay of GPX4 and FBXO45 with indicated FBXO45 antibodies in RAECs protein lysate. E Co-IP assay of the GPX4 protein by an FBXO45 antibody in HEK293T cells transfected with pcDNA3.1-HA-GPX4 and pcDNA3.1-Flag-FBXO45. PcDNA3.1-vector was used as a negative control. F Degradation of the GPX4 protein was measured after the treatment of 200 µg/mL CHX at the indicated time points in HEK293T cells, which transfected with Flag-FBXO45 expression plasmids and siFbxo45. G Analysis of GPX4 ubiquitination was performed by Co-IP using an anti-GPX4 antibody, followed by immunoblot with anti-Ub antibody and anti-GPX4 antibody in HEK293T cells transfected with the indicated constructs with or without MG132 (10 µM for 10 h). H Analysis of GPX4 ubiquitination was performed by Co-IP using an anti-GPX4 antibody, followed by immunoblot with anti-Ub antibody and anti-GPX4 antibody in HEK293T cells transfected with the indicated constructs with or without siFbxo45 at the presence of MG132 (10 µM for 10 h). Statistical significances were calculated using (A, C, F) two-way ANOVA, Tukey’s multiple comparisons tests. Data are expressed as the mean ± SEM, statistical analysis revealed a significant difference with **p < 0.01, ***p < 0.001, #p < 0.05, ##p < 0.01, ###p < 0.001, ns = not significant

Full size image

Ubiquitination of GPX4 has been observed in various diabetic conditions [14]. Our study also observed ubiquitin-mediated proteolysis (Supplementary Fig. 7D). To investigate the impact of FBXO45 on GPX4 ubiquitination, we conducted ubiquitination assays. High glucose elevated the overall ubiquitination level of RAECs (Supplementary Fig. 7E). Additionally, HEK293T cells transfected with His-Ub exhibited pronounced degradation of GPX4 (Supplementary Fig. 7F). Next, we performed ubiquitination assays in HEK293T induced with exogenous HA-GPX4 and His-Ub, in the presence or absence of Flag-FBXO45 and/or MG132. The cell extracts were immunoprecipitated using an anti-GPX4 antibody followed by immunoblot analysis with an anti-Ubiquitin antibody. GPX4-ubiquitination was enhanced by MG132 treatment (Fig. 6G, lane 3), and the existence of FBXO45 obviously increased the ubiquitination of GPX4 without MG132 treatment (Fig. 6G, lane 4); furthermore, GPX4-ubiquitination was further elevated when FBXO45 was present along with MG132 treatment (Fig. 6G, lane 5). Subsequently, the Co-IP assay was performed on HEK293T cells that were induced with HA-GPX4, His-Ub, and MG132 in the presence or absence of Flag-FBXO45 and/or siFbox45. Consistently, GPX4-ubiquitination levels were lowered with siFbox45 neutralized the FBXO45 induction (Fig. 6H, lane 4 and 5). The above bioinformatics analysis results validate the involvement of FBXO45 in protein ubiquitination modification process of GPX4.

SETD7-mediated methylation of p53 promotes the transcriptional activation of FBXO45

Next, we further investigated the regulatory mechanisms of SETD7 on FBXO45 transcription. SETD7 has been shown to directly methylate the p53 protein, leading to p53 stabilization and nuclear translocation, thereby enhancing transactivation of target genes involved in ROS response [23]. Meanwhile, a database search discovered that only four factors were identified as both FBXO45 transcriptional regulator and predicted interactor with SETD7, among which p53 was included (Supplementary Fig. 8A). Thus, we assume that SETD7 facilitates the transcriptional activation of FBXO45 via p53 nuclear translocation. GSEA analysis of the siSetd7-treated RAECs revealed a potential association between SETD7 and Tp53 activity regulation (Supplementary Fig. 8B). In this case, we verified that SETD7 clearly affected the protein expression of p53 (Fig. 7A). Co-IP assay identified that SETD7 directly impacted the overall methylation level of p53 (Supplementary Fig. 8C). In vitro methylation experiment demonstrated that SETD7 was responsible for the mono-methylation of p53 (monop53) (Fig. 7B). It is also known SETD7 may regulate p53 activation through MDM2, an E3 ubiquitin ligase known to negatively regulate p53 [24]. However, our findings indicated that SETD7 failed to regulate MDM2 (Supplementary Fig. 8D), suggesting its primary role in mediating p53 activation and nuclear translocation. Nucleus isolation and IF assay exhibited that siSetd7 reduced the p53 nuclear location in HG-induced RAECs (Supplementary Fig. 8E, Fig. 7C). The extracted nuclear proteins were subjected to Co-IP assay, which further confirmed p53 mono-methylation and its nuclear translocation were regulated by SETD7 under high glucose exposure (Fig. 7D).

Fig. 7
figure 7

SETD7-mediated methylation of p53 promotes the transcriptional activation of FBXO45. RAECs were cultured in different glucose concentration conditions (Control: 5.5 mM glucose + 16.5 mM mannitol, HG: 22 mM glucose) with or without siScr, siSetd7, or siTp53 treatment. A The protein expression of p53 in siScr and siSetd7 RAECs with or without HG stimulation. B In vitro methyltransferase assays. Recombinant HA-p53 protein was incubated with the recombinant His-SETD7 and S-Adenosyl methionine (SAM) in a mixture of methylase buffer with or without preincubation of SETD7 inhibitor sinefungin. C Representative images of immunofluorescence staining of p53 in siScr and siSetd7 RAECs with or without HG stimulation. Scale bars, 50 μm. D Co-immunoprecipitation (Co-IP) assay of mono-methylated p53 with indicated p53 antibody in RAECs nuclear protein. E Dual-luciferase reporter assay of p53 activate Fbxo45 transcription. F The mRNA level of Fbxo45 and Nox4 in siScr and siSetd7 RAECs with or without HG stimulation. G The protein expression of p53, FBXO45, and GPX4 in siScr and siTp53 RAECs with or without HG stimulation. H Representative images of immunofluorescence staining of GPX4 in siScr and siTp53 RAECs with or without HG stimulation. Scale bars, 50 μm. I Chip assay of the enrichment of p53 in the Fbxo45 promoter region in siScr and siSetd7 RAECs with or without HG stimulation. Statistical significances were calculated using (A, E) one-way ANOVA, (F, G, and I) two-way ANOVA, Tukey’s multiple comparisons tests. Data are expressed as the mean ± SEM, statistical analysis revealed a significant difference with **p < 0.01, ***p < 0.001, ###p < 0.001, ns = not significant

Full size image

To provide evidence for the transcriptional activation mechanism between p53 and FBXO45, we performed a dual-luciferase reporter assay to confirm their interaction (Fig. 7E). Furthermore, the mRNA levels and protein expressions of FBXO45 were validated to be influenced by p53 through RT-qPCR, western blotting, and IF assay (Fig. 7F-H). The binding site of p53 on the FBXO45 gene promoter was analyzed using JASPAR (Supplementary Fig. 8F), which was further confirmed by chip assay. Chip-PCR results demonstrated that SETD7 primarily regulated the enrichment of p53 on the 0–500 bp promoter region of Fbxo45 in high glucose-induced ECs, thereby triggering its transcription (Supplementary Fig. 8G, Fig. 7I). These findings illustrate that SETD7-mediated methylation of p53 promotes the transcriptional activation of FBXO45, thereby facilitating GPX4 loss under diabetic conditions.

Involvement of SETD7-p53-FBXO45-GPX4 axis in diabetic mice endothelial dysfunction

From in vitro experiments, we proposed the involvement of SETD7-p53-FBXO45-GPX4 axis in regulating endothelial dysfunction. In animal experiments, we further observed a decrease in the upregulation of p53, FBXO45 and NOX4, along with an increase in the downregulation of GPX4 in aorta from STZ-induced Setd7−/− mice compared with STZ-induced WT mice (Fig. 8A). Immunohistochemistry (IHC) and IF analysis also confirmed these results (Fig. 8B, C). Similarly, in db/db mice, SETD7 endothelial-specific deficiency reduced the upregulation of p53, FBXO45 and NOX4 expression. It also inhibited the decrease of GPX4 (Fig. 8D). IHC and IF analysis yielded consistent findings with those obtained from the western blot (Fig. 8E, F). These results suggest that the SETD7-p53-FBXO45-GPX4 axis links the lipid peroxidation and oxidative stress and endothelial dysfunction in diabetic mice.

Fig. 8
figure 8

Involvement of SETD7-p53-FBXO45-GPX4 axis in diabetic mice endothelial dysfunction. A The protein expression of p53, FBXO45, GPX4, and NOX4 in aortic tissues of WT and Setd7−/− mice induced by STZ (50 mg/kg/d) for five consecutive days (n = 4). B Representative images of immunohistochemical FBXO45-stained thoracic aorta sections isolated from WT and Setd7−/− mice. Scale bars, 50 μm. C Representative images of immunofluorescence GPX4-stained thoracic aorta sections isolated from WT and Setd7−/− mice. Scale bars, 50 μm. D The protein expression of p53, FBXO45, GPX4, and NOX4 in aortic tissues of WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr (n = 4). E Representative images of immunohistochemical FBXO45-stained thoracic aorta sections isolated from WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr. Scale bars, 50 μm. F Representative images of immunofluorescence GPX4-stained thoracic aorta sections isolated from WT/WT and db/db mice received AAV-shSetd7 or AAV-Ctr. Scale bars, 50 μm. Statistical significances were calculated using (A and D) two-way ANOVA, Tukey’s multiple comparisons tests. Data are expressed as the mean ± SEM, statistical analysis revealed a significant difference with **p < 0.01, ***p < 0.001, #p < 0.05, ###p < 0.001

Full size image

Discussion.

In this study, we discovered that hyperglycemia induces SETD7 to impair endothelial function through the activation of lipid peroxidation and oxidative stress in diabetes. Knockdown of endothelial SETD7 in vivo limits oxidative stress and the impairment of endothelial barriers. Mechanistically, SETD7 medicates methylation of p53 to promote the transcription of FBXO45, which results in ubiquitination and degradation of GPX4 and subsequent severe dysfunction. Combining the results from in vitro and in vivo experiments, our research highlights the involvement of the SETD7-p53-FBXO45-GPX4 axis under hyperglycemia and its adverse role in high glucose-induced oxidative stress injury, offering novel insights into endothelial dysfunction associated with hyperglycemia.

Endothelial cells are more susceptible to cellular damage induced by hyperglycemia compared to other cell types, including fatty acid oxidation, decrease in nitric oxide levels, oxidative stress, inflammatory activation, and impaired barrier function [25]. Epigenetic changes caused by transient hyperglycemia can be stored as cellular memory and progressively lead to irreversible cell damage [26]. In fact, vascular endothelial cells exposed to hyperglycemia have been reported to continue experiencing increase oxidative stress and inflammation even after blood glucose normalization [27]. In this study, we attempted to investigate the role of SETD7, a lysine methyltransferase, in endothelial damage caused by hyperglycemia. Initially, activation of SETD7 exacerbates endothelial cell dysfunction exposed to high-glucose environment, which is consistent with its function as a metabolite-sensitive sensor for hyperglycemia and its association with proinflammatory gene expression [19]. Conversely, SETD7 knockout mice preserved endothelium-dependent vasorelaxation and alleviated vascular endothelial cell damage during prolonged hyperglycemic stimulation. Correspondingly, we detected the endothelial function of the aorta from db/db mice with endothelial-specific knockdown of SETD7, further confirming that the endothelial-knockdown of SETD7 alleviates the potential endothelial cell dysfunction induced by high glucose. Meanwhile, SETD7 deficiency protected microangiopathy associated with diabetic retinopathy, as evidenced by modifications in retinal edema, retinal leakage, and pericytes cells loss.

In larger arteries, eNOS-derived NO is considered one of the most crucial local regulators of vasodilation [28]. Meanwhile, a recent study linked SETD7 with defects of diabetic angiogenesis that SETD7 upregulation was associated with the high glucose-induced impaired endothelial cell migration [29]. Thus, we detected the expression of eNOS and VE-cadherin or VEGF-α (angiogenesis markers) in aorta and the endothelial cells, confirming that the knockout of SETD7 alleviates potential endothelial cell dysfunction and high glucose-induced cell migration damage. However, our findings propose a novel mechanism by which SETD7 regulates endothelial function by participation in oxidative stress. To date, the essential role of antioxidants in restoring endothelium-dependent vascular function has been emphasized by various evidence, leading to several therapeutic avenues aimed at improving endothelial function [30]. Indeed, our study also demonstrated the presence of ROS production in hyperglycemia-damaged endothelial cells. Loss of SETD7 protected cells from oxidative stress, as detected by decreased ROS production, reduced mitochondrial depolarization and MDA content, as well as recovered GSH-Px and SOD activity. GPX4 serves as a crucial antioxidant peroxidase; its dysfunction in preventing lipid peroxidation can cause various tissue damages. We noticed the involvement of GPX4 in SETD7-mediated endothelial dysfunction. Results suggested that SETD7 contributes to GPX4 ablation under high glucose induction but not through ferroptosis.

Our results further suggested that high glucose induced the ubiquitylation of GPX4, which inspire us to explore the ubiquitylation and proteasome-dependent degradation of GPX4. The ubiquitin-proteasome system (UPS) is the major mechanism for intracellular protein degradation, and several E3 Ub ligases have shown to regulate ROS production by targeting critical enzymes, including GPX4 [31]. FBXO45 (F-box/SPRY) domain-containing protein is an atypical F-box protein that can recruit substrates to the ubiquitin ligase complex, triggering protein degradation through both its F-box and SPRY domains [32, 33]. The expression of FBXO45 was regulated by SETD7, and deletion of FBXO45 also alleviated endothelial oxidative stress injury and the inflammatory response. Intriguingly, our results have provided further support for the ability of FBXO45 to bind GPX4 and promote its ubiquitylation and proteasome-dependent degradation. Thus, FBXO45 may serve as a crucial player in SETD7-mediated GPX4 downregulation.

The previous results showed that SETD7 regulated not only the protein expression but also the mRNA level of FBXO45. Meanwhile, the SETD7 lysine methyltransferase not only methylates H3K4, which is implicated in gene-activating transcriptional events, but also methylates lysine residues of nonhistone proteins, including several transcriptional factors [16]. We found that p53 may act as the intermediate between SETD7 and FBXO45. Researches have demonstrated that SETD7 promotes p53 stabilization and activation of p53 target genes by directly methylation p53 in response to ROS [23, 34]. Additionally, high glucose induces an increase in p53 expression levels leading to premature cellular senescence [35], and increased p53 expression is closely related to impaired endothelium-dependent vasodilation in diabetes [36, 37]. Our study demonstrated that silencing SETD7 reduces the expression of p53. Subsequently, SETD7 regulates p53 mono-methylation and nuclear translocation, thereby enhancing the transactivation of target genes in response to ROS and diabetic stimuli. Meanwhile, p53 also activates the transcription of FBXO45 through binding to its 0–500 bp promoter region under hyperglycemia. Although reports suggest that p53 sensitizes cells for ferroptosis by repressing xCT, indirectly suppressing GPX4 [38], our study demonstrates that SETD7-mediated GPX4 inhibition is independent of xCT and ferroptosis.

This study identified the presence of SETD7-p53-FBXO45-GPX4 in high glucose-induced endothelial cells. However, no evidence of ferroptosis was observed in this study, nor did it demonstrate the typical inhibitory role of GPX4 in ferroptosis or the extensively studied regulatory function of p53-xCT. These findings suggest that the regulatory role of this pathway in high glucose-induced endothelial cell damage may be specific to certain cellular contexts. Considering that SETD7 is also expressed in other cell types, such as smooth muscle cells, immune cells, and adipocytes, its expression and function in these cell types may involve additional mediators.

In summary, our study demonstrates that SETD7 activation exacerbates endothelial dysfunction in a high-glucose environment. This suggests a novel mechanism whereby SETD7 enhances GPX4 ubiquitylation through the activation of p53 and the promotion of FBXO45 transcription during hyperglycemia, ultimately leading to GPX4 degradation and the occurrence of lipid peroxidation. Therefore, targeting SETD7 may hold therapeutic potential for mitigating endothelial cell damage in diabetes.

Data availability

No datasets were generated or analysed during the current study.

References

  1. Niu F, Liu W, Ren Y, Tian Y, Shi W, Li M, et al. β-cell neogenesis: a rising star to rescue diabetes mellitus. J Adv Res. 2024;62:71–89. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jare.2023.10.008.

    Article  PubMed  Google Scholar 

  2. Zhou HL, Jiang XZ, Ventikos Y. Role of blood flow in endothelial functionality: a review. Front Cell Dev Biol. 2023;11:1259280. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fcell.2023.1259280.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Shi Y, Vanhoutte PM. Macro- and microvascular endothelial dysfunction in diabetes. J Diabetes. 2017;9:434–49. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/1753-0407.12521.

    Article  CAS  PubMed  Google Scholar 

  4. Kaur R, Kaur M, Singh J. Endothelial dysfunction and platelet hyperactivity in type 2 diabetes mellitus: molecular insights and therapeutic strategies. Cardiovasc Diabetol. 2018;17:121. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12933-018-0763-3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. La Sala L, Prattichizzo F, Ceriello A. The link between diabetes and atherosclerosis. Eur J Prev Cardiol. 2019;26:15–24. https://doiorg.publicaciones.saludcastillayleon.es/10.1177/2047487319878373.

    Article  PubMed  Google Scholar 

  6. Xue C, Chen K, Gao Z, Bao T, Dong L, Zhao L, et al. Common mechanisms underlying diabetic vascular complications: focus on the interaction of metabolic disorders, immuno-inflammation, and endothelial dysfunction. Cell Commun Signal. 2023;21:298. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12964-022-01016-w.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Niu C, Chen Z, Kim KT, Sun J, Xue M, Chen G, et al. Metformin alleviates hyperglycemia-induced endothelial impairment by downregulating autophagy via the Hedgehog pathway. Autophagy. 2019;15:843–70. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/15548627.2019.1569913.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. An Y, Xu B-T, Wan S-R, Ma X-M, Long Y, Xu Y, Jiang Z-Z. The role of oxidative stress in diabetes mellitus-induced vascular endothelial dysfunction. Cardiovasc Diabetol. 2023;22:237. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12933-023-01965-7.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Tang S-C, Liao P-Y, Hung S-J, Ge J-S, Chen S-M, Lai J-C, et al. Topical application of glycolic acid suppresses the UVB induced IL-6, IL-8, MCP-1 and COX-2 inflammation by modulating NF-κB signaling pathway in keratinocytes and mice skin. J Dermatol Sci. 2017;86:238–48. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jdermsci.2017.03.004.

    Article  CAS  PubMed  Google Scholar 

  10. Huang S, Chen G, Sun J, Chen Y, Wang N, Dong Y, et al. Histone deacetylase 3 Inhibition alleviates type 2 diabetes mellitus-induced endothelial dysfunction via Nrf2. Cell Commun Signal. 2021;19:35. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12964-020-00681-z.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Miao Y, Chen Y, Xue F, Liu K, Zhu B, Gao J, et al. Contribution of ferroptosis and GPX4’s dual functions to osteoarthritis progression. EBioMedicine. 2022;76:103847. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.ebiom.2022.103847.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Bai T, Li M, Liu Y, Qiao Z, Wang Z. Inhibition of ferroptosis alleviates atherosclerosis through attenuating lipid peroxidation and endothelial dysfunction in mouse aortic endothelial cell. Free Radic Biol Med. 2020;160:92–102. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.freeradbiomed.2020.07.026.

    Article  CAS  PubMed  Google Scholar 

  13. Malekmohammad K, Sewell RDE, Rafieian-Kopaei M. Mechanisms of medicinal plant activity on nitric oxide (NO) bioavailability as prospective treatments for atherosclerosis. Curr Pharm Des. 2020;26:2591–601. https://doiorg.publicaciones.saludcastillayleon.es/10.2174/1381612826666200318152049.

    Article  CAS  PubMed  Google Scholar 

  14. Zhang J, Qiu Q, Wang H, Chen C, Luo D. TRIM46 contributes to high glucose-induced ferroptosis and cell growth Inhibition in human retinal capillary endothelial cells by facilitating GPX4 ubiquitination. Exp Cell Res. 2021;407:112800. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.yexcr.2021.112800.

    Article  CAS  PubMed  Google Scholar 

  15. Jetton TL, Flores-Bringas P, Leahy JL, Gupta D. SetD7 (Set7/9) is a novel target of PPARγ that promotes the adaptive pancreatic β-cell glycemic response. J Biol Chem. 2021;297:101250. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jbc.2021.101250.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Wang J, Tan S, Zhang Y, Xu J, Li Y, Cheng Q, et al. Set7/9 aggravates ischemic brain injury via enhancing glutamine metabolism in a blocking Sirt5 manner. Cell Death Differ. 2024;31:511–23. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41418-024-01264-y.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Paneni F, Costantino S, Battista R, Castello L, Capretti G, Chiandotto S, et al. Adverse epigenetic signatures by histone methyltransferase Set7 contribute to vascular dysfunction in patients with type 2 diabetes mellitus. Circ Cardiovasc Genet. 2015;8:150–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/CIRCGENETICS.114.000671.

    Article  CAS  PubMed  Google Scholar 

  18. Paneni F, Costantino S, Castello L, Battista R, Capretti G, Chiandotto S, et al. Targeting prolyl-isomerase Pin1 prevents mitochondrial oxidative stress and vascular dysfunction: insights in patients with diabetes. Eur Heart J. 2015;36:817–28. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/eurheartj/ehu179.

    Article  CAS  PubMed  Google Scholar 

  19. Okabe J, Orlowski C, Balcerczyk A, Tikellis C, Thomas MC, Cooper ME, El-Osta A. Distinguishing hyperglycemic changes by Set7 in vascular endothelial cells. Circ Res. 2012;110:1067–76. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/CIRCRESAHA.112.266171.

    Article  CAS  PubMed  Google Scholar 

  20. Tian Z, Ning H, Wang X, Wang Y, Han T, Sun C. Endothelial autophagy promotes atheroprotective communication between endothelial and smooth muscle cells via exosome-mediated delivery of miR-204-5p. Arterioscler Thromb Vasc Biol. 2024;44:1813–32. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/ATVBAHA.123.319993.

    Article  CAS  PubMed  Google Scholar 

  21. Luo L, Cai Y, Jiang Y, Gong Y, Cai C, Lai D, et al. Pipecolic acid mitigates ferroptosis in diabetic retinopathy by regulating GPX4-YAP signaling. Biomed Pharmacother. 2023;169:115895. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.biopha.2023.115895.

    Article  CAS  PubMed  Google Scholar 

  22. Bersuker K, Hendricks JM, Li Z, Magtanong L, Ford B, Tang PH, et al. The CoQ oxidoreductase FSP1 acts parallel to GPX4 to inhibit ferroptosis. Nature. 2019;575:688–92. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41586-019-1705-2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Daks A, Shuvalov O, Fedorova O, Parfenyev S, Simon H-U, Barlev NA. Methyltransferase Set7/9 as a multifaceted regulator of ROS response. Int J Biol Sci. 2023;19:2304–18. https://doiorg.publicaciones.saludcastillayleon.es/10.7150/ijbs.83158.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Lezina L, Aksenova V, Fedorova O, Malikova D, Shuvalov O, Antonov AV, et al. KMT Set7/9 affects genotoxic stress response via the Mdm2 axis. Oncotarget. 2015;6:25843–55. https://doiorg.publicaciones.saludcastillayleon.es/10.18632/oncotarget.4584.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Knapp M, Tu X, Wu R. Vascular endothelial dysfunction, a major mediator in diabetic cardiomyopathy. Acta Pharmacol Sin. 2019;40:1–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41401-018-0042-6.

    Article  CAS  PubMed  Google Scholar 

  26. Yamazaki T, Mimura I, Tanaka T, Nangaku M. Treatment of diabetic kidney disease: current and future. Diabetes Metab J. 2021;45:11–26. https://doiorg.publicaciones.saludcastillayleon.es/10.4093/dmj.2020.0217.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Miller RG, Costacou T. Cardiovascular disease in adults with type 1 diabetes: looking beyond glycemic control. Curr Cardiol Rep. 2022;24:1467–75. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s11886-022-01763-9.

    Article  PubMed  Google Scholar 

  28. Chen Q, Hao H, Guo Z, Zuo Y, Cheng CK, Zhang C-L, et al. Pien tze Huang (PZH) protects endothelial function in diabetic mice. Life Sci. 2024;349:122723. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.lfs.2024.122723.

    Article  CAS  PubMed  Google Scholar 

  29. Mohammed SA, Gorica E, Albiero M, Karsai G, Mengozzi A, Caravaggi CM et al. Targeting SETD7 rescues Diabetes-induced impairment of angiogenic response by transcriptional repression of semaphorin 3G. Diabetes. 2025. https://doiorg.publicaciones.saludcastillayleon.es/10.2337/db24-0997

  30. Zhang X, Leng S, Liu X, Hu X, Liu Y, Li X, et al. Ion channel Piezo1 activation aggravates the endothelial dysfunction under a high glucose environment. Cardiovasc Diabetol. 2024;23:150. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12933-024-02238-7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Sheng X, Xia Z, Yang H, Hu R. The ubiquitin codes in cellular stress responses. Protein Cell. 2024;15:157–90. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/procel/pwad045.

    Article  CAS  PubMed  Google Scholar 

  32. Wu L, Yu K, Chen K, Zhu X, Yang Z, Wang Q, et al. Fbxo45 facilitates pancreatic carcinoma progression by targeting USP49 for ubiquitination and degradation. Cell Death Dis. 2022;13:231. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41419-022-04675-2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Díaz VM, de Herreros AG. F-box proteins: keeping the epithelial-to-mesenchymal transition (EMT) in check. Semin Cancer Biol. 2016;36:71–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.semcancer.2015.10.003.

    Article  CAS  PubMed  Google Scholar 

  34. Campaner S, Spreafico F, Burgold T, Doni M, Rosato U, Amati B, Testa G. The methyltransferase Set7/9 (Setd7) is dispensable for the p53-mediated DNA damage response in vivo. Mol Cell. 2011;43:681–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.molcel.2011.08.007.

    Article  CAS  PubMed  Google Scholar 

  35. Chen T, Ma C, Fan G, Liu H, Lin X, Li J, et al. SIRT3 protects endothelial cells from high glucose-induced senescence and dysfunction via the p53 pathway. Life Sci. 2021;264:118724. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.lfs.2020.118724.

    Article  CAS  PubMed  Google Scholar 

  36. Wu Y, Lee S, Bobadilla S, Duan SZ, Liu X. High glucose-induced p53 phosphorylation contributes to impairment of endothelial antioxidant system. Biochim Biophys Acta Mol Basis Dis. 2017;1863:2355–62. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.bbadis.2017.06.022.

    Article  CAS  PubMed  Google Scholar 

  37. Yokoyama M, Shimizu I, Nagasawa A, Yoshida Y, Katsuumi G, Wakasugi T, et al. p53 plays a crucial role in endothelial dysfunction associated with hyperglycemia and ischemia. J Mol Cell Cardiol. 2019;129:105–17. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.yjmcc.2019.02.010.

    Article  CAS  PubMed  Google Scholar 

  38. Le Jiang, Kon N, Li T, Wang S-J, Su T, Hibshoosh H, et al. Ferroptosis as a p53-mediated activity during tumour suppression. Nature. 2015;520:57–62. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nature14344.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

The working model of this study was drawn using Biorender (www.biorender.com).

Funding

This work was supported by Construction of TCM discipline in Pudong New Area (YC-2023-0202, YC-2023-0607).

Author information

Authors and Affiliations

  1. Phenome Research Center of TCM, Department of Traditional Chinese Medicine, Shanghai Pudong Hospital, Pharmacophenomics Laboratory, Human Phenome Institute, Fudan University, 825, Zhangheng Road, Pudong New District, Shanghai, 201203, China

    Wen Zhong, Ruoxue Chen, Jialin Zhao, Yuyu Zhang, Jintao He, Huibin Wang, Feng Zhu, Chunxiang Fan & Xinhua Liu

Authors
  1. Wen Zhong
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Ruoxue Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Jialin Zhao
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Yuyu Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Jintao He
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Huibin Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Feng Zhu
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Chunxiang Fan
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Xinhua Liu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

ZW performed the experiments, analyzed the data, and wrote the manuscript. ZJL analyzed RNA-seq data. CRX, ZYY, HJT, ZF, and WHB contributed to the manuscript. LXH and FCX designed the experiments, supervised the study, and revised the manuscript.

Corresponding authors

Correspondence to Chunxiang Fan or Xinhua Liu.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

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

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

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

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhong, W., Chen, R., Zhao, J. et al. SETD7 drives diabetic endothelial dysfunction through FBXO45-mediated GPX4 ubiquitylation. Cardiovasc Diabetol 24, 178 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12933-025-02740-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12933-025-02740-6

Keywords

Cardiovascular Diabetology

ISSN: 1475-2840