Endothelial cystathionine-γ-lyase (CTH) regulates body weight and insulin resistance in mice fed a high-fat diet

Abstract

Hydrogen sulfide (H₂S) has been reported to play a critical role in vascular function and metabolism, yet specific molecular mechanisms remain largely unknown. Given this importance and the role of endothelium in metabolic health, we hypothesized that endothelial cystathionine-γ-lyase deficiency (ecCTH-KO) could modulate metabolic function influencing insulin sensitivity and weight gain. To test this, we examined body weight, adipose tissue inflammation, glucose metabolism, and insulin signaling in ecCTH-KO and wild-type (WT) mice under a high-fat diet (HFD) (45% fat). ecCTH-KO mice exhibited significantly greater weight gain than WT mice despite similar food intake, with increased epididymal (eWAT) and mesenteric white adipose tissue (mWAT), but unchanged brown adipose tissue (BAT) mass. Importantly, reduced UCP1 expression was observed in eWAT suggesting decreased energy expenditure, while MCP-1 and IL-6 expression was elevated in mWAT, indicating augmented adipose-specific inflammation. Metabolically, ecCTH-KO mice displayed significantly reduced sulfane sulfur levels in adipose tissue and skeletal muscle along with impaired glucose tolerance and insulin resistance, confirmed by increased HOMA-IR values and reduced insulin-stimulated AKT phosphorylation in skeletal muscle and liver. Lastly, insulin dependent GLUT4 translocation was also impaired in skeletal muscle. These findings demonstrate that endothelial CTH deficiency significantly contributes to weight gain and insulin resistance by increasing adiposity and inflammation with concomitant disruption of skeletal muscle insulin signaling, revealing a crucial and previously unknown role of vascular sulfane sulfur in weight gain and glucose metabolism in skeletal muscle.

Graphical abstract

Highlights

• •Endothelial CTH-KO mice gain more weight than wild type mice on a high-fat diet with increased epididymal (eWAT) and mesenteric (mWAT) white adipose tissue mass.
• •Endothelial CTH-KO showed selective reduced sulfane sulfur levels in skeletal muscle and white adipose tissue (WAT) compared to other sulfide metabolites.
• •Endothelial CTH-KO mice showed reduced UCP1 expression in eWAT suggesting decreased energy expenditure while MCP-1 and IL-6 levels were elevated in mWAT.
• •Endothelial CTH-KO mice showed impaired glucose tolerance, increased HOMA-IR, and insulin resistance involving impaired AKT signaling and GLUT4 membrane translocation.

1. Introduction

Obesity, defined by excessive fat accumulation and weight gain that exceeds the standard body weight for a given height, is increasingly recognized as a global health crisis. Its prevalence has surged in recent decades, with the World Health Organization (WHO) reporting that more than 650 million adults worldwide are obese. Obesity is closely linked to a variety of metabolic disorders and complications, including Type 2 diabetes, cardiovascular diseases, hypertension, hyperlipidemia, chronic kidney disease, sleep apnea, and cancer [[1], [2], [3], [4]]. Epidemiological studies have demonstrated that obesity significantly reduces life expectancy, with individuals with obesity having a 4-9 year shorter life expectancy compared to their healthy counterparts [5]. Current treatment strategies for obesity include medical and surgical interventions, but these remain inadequate for the growing number of affected individuals. Until recently, anti-obesity medications have often failed to provide long-term success and suffer from poor patient adherence [6,7]. Bariatric surgery, on the other hand, has shown more significant weight loss outcomes but also carries potential risks [8]. Despite the emergence of more effective new-generation anti-obesity medications during the past couple of years, there remains the need for novel therapeutic approaches targeting the underlying mechanisms of obesity and its complications.

One such mechanism involves endothelial dysfunction, which has been increasingly implicated in the pathogenesis of insulin resistance and metabolic disorders. Endothelial cells line the blood vessels and play a crucial role in regulating vascular tone, blood flow, and the delivery of nutrients, including glucose, to tissues. Recent studies have highlighted that microvascular dysfunction, particularly in skeletal muscle, impairs glucose delivery and diminishes insulin-mediated glucose uptake, leading to insulin resistance [9]. Endothelial dysfunction has been identified as a key player in the progression of metabolic diseases, as it contributes to impaired glucose metabolism and contributes to obesity [10]. Furthermore, endothelial dysfunction is often accompanied by inflammation and oxidative stress, which exacerbates metabolic dysfunction, including insulin resistance and adiposity.

Hydrogen sulfide (H₂S) is a gasotransmitter and signaling molecule that has garnered attention for its role in regulating endothelial function. Synthesized by transsulfuration pathway enzymes such as cystathionine γ-lyase (CTH), H₂S plays a critical role in vascular homeostasis, regulating vasodilation, endothelial cell proliferation, inflammatory response, and modulation of nitric oxide levels [11]. Impaired insulin signaling in endothelial cells disrupts glucose delivery to skeletal muscle due to reduced nitric oxide (NO)-mediated vasodilation, contributing to insulin resistance [10]. Impairment of capillary density and insulin sensitivity in skeletal muscle has been implicated by high fat diet (HFD) mediated upregulation of FOXO protein in endothelial cells [12]. Moreover, a recent study indicated that obesity diminishes insulin transcytosis across skeletal muscle capillaries, linking structural endothelial changes to decreased insulin action [13]. Furthermore, endothelial nitric oxide synthase (eNOS) and the PI3K-AKT pathway regulate insulin-mediated vasodilation and glucose uptake in muscle, highlighting the endothelium's critical metabolic role [14]. Previous research has shown that inhibition of CTH in endothelial cells leads to endothelial dysfunction, especially under conditions of hyperglycemia, suggesting that H₂S may play a key role in maintaining endothelial integrity and function [15]. Despite these insights, one critical metabolic pathway remains poorly understood that being endothelial-specific regulation of skeletal muscle metabolic and insulin signaling through generation of H₂S metabolites through mechanisms such as CTH.

Here our study addresses this knowledge gap by investigating the effects of endothelial-specific CTH knockout (ecCTH-KO) on body weight and glucose metabolism in high fat diet fed mice. We hypothesized that diminished endothelial-derived CTH dependent sulfide production leads to insulin resistance and disrupted glucose metabolism in skeletal muscle, contributing to increased body weight.

2. Materials and methods

2.1. Animal experiments

Eight-week-old male C57BL/6J mice (WT) and endothelial-specific CTH-KO (ecCTH KO) mice with a C57BL/6 background developed in our lab were used in this study. we developed vascular cell-specific Cth mutant mice using C57BL/6NTac- Cth^{tm1a(EUCOMM)Hmgu}/Ieg strain crossed with B6.129S4- Gt(ROSA)26Sor^tm2(FLP*^)Sor/J Flp deleter line and then backcrossed to the B6.129-Tg(Cdh5-cre)^1Spe/J (VE-Cad Cre) driver line to obtain endothelial cell CTH-KO (Sup Fig. 1). In this study, we used only male mice to avoid the confounding effects of estrogen. Previous studies have shown that estrogen enhances insulin sensitivity [16], which could influence the interpretation of our results. The mice were housed in cages under a 12-h light/dark cycle at a controlled temperature of 68°- 74 °F, with four mice per cage and free access to food and water and fed a 45% kcal high-fat diet (HFD) (3.85 kcal/g; D12451, Research Diets, Inc., NJ, USA) or a chow control diet (D12450H, Research Diets, Inc., NJ, USA) until the end of experiments. Body weight and food intake were measured weekly; at the age of 16 weeks, blood samples and tissue were collected and frozen in liquid nitrogen until analysis. Feed efficiency was calculated by dividing weekly body weight gain by weekly energy intake. In each treatment group 5-11 mice were used. All animal procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee (IACUC) of LSU Health Sciences Center–Shreveport. Animal care and experimental protocols adhered to National Institutes of Health guidelines.

Fig. 1.

Effect of ecCTH-KO on body weight and energy intake.

Eight-week-old mice were fed a high-fat diet for 8 weeks. (A) Weekly body weight gain (Δbody weight) from day 0 to the indicated week; (B) Δ body weight at the 8-week time point; (C) average energy intake from day 0 to 8 weeks; (D) weekly cumulative energy intake; (E) weekly energy intake for each week; and (F) average feed efficiency from day 0 to 8 weeks Data are expressed as mean ± SEM. Statistical significance is indicated by *p < 0.05, **p < 0.005, ***p < 0.001, ****p < 0.0001 (Tukey's multiple comparisons test or Student's t-test). N = 5-11.

2.2. Glucose tolerance test and insulin tolerance test

An intraperitoneal glucose tolerance test (GTT) was performed after an overnight fast. Blood glucose levels were measured using a glucometer (AlphaTRAK, NJ, USA) at 0, 30, 60, and 120 min following intraperitoneal injection of glucose (1 g/kg body weight). An insulin tolerance test (ITT) was conducted after a 4-h fast, with blood glucose levels measured at 0, 15, 30, 60, and 90 min after intraperitoneal injection of insulin (2 U/kg body weight; Sigma-Aldrich, St. Louis, MO, USA).

2.3. Measurement of plasma parameters

Blood samples were collected after overnight fasting. The plasma total cholesterol levels, plasma triglyceride levels, and plasma free fatty acid levels were assessed using a Cayman chemical assay kit and blood insulin levels were assessed using a Crystal chemical assay kit (Elk Grove Village, IL, USA) according to the manufacturer's protocols. Homeostatic model assessment of insulin resistance (HOMA-IR) was calculated as previously described [17].

2.4. Measurement of sulfide metabolites by monobromobimane (MBB) derivatization method

Free H₂S, acid labile, and bound sulfane sulfur levels were quantified using the monobromobimane derivatization method followed by reversed-phase high-performance liquid chromatography (RP-HPLC), as we've previously described [18]. Briefly, tissue samples were collected and immediately stabilized in vials containing 100 mM Tris–HCl buffer (pH 9.5) supplemented with 0.1 mM diethylenetriaminepentaacetic acid (DTPA) at a 1:5 (v/v) ratio and snap-frozen until analysis. Samples were derivatized with MBB and separated using an RP-HPLC system equipped with an Agilent Eclipse XDB-C18 column (5 μm, 80 Å, 4.6 × 250 mm). Sulfide–dibimane adducts were detected by fluorescence with excitation at 390 nm and emission at 475 nm.

2.5. Measurement of sulfane sulfur and hydrogen sulfide by fluorescence method

Free hydrogen sulfide and sulfane sulfur were measured using the fluorescent probes SF7 and SSP4, respectively, as we've previously described [19]. SSP4 specifically detects polysulfides (sulfane sulfur), whereas SF7 selectively detects free hydrogen sulfide. Samples were diluted 1:10 in HBSS and incubated either with 15 μM Sulfane Sulfur Probe 4 (SSP4; Dojindo Laboratories, Japan) in the presence of 100 μM hexadecyltrimethylammonium bromide (CTAB; G-Bioscience, St. Louis, MO, USA) for 30 min, or with Sulfidefluor-7 acetoxymethyl ester (SF7; Cayman Chemical, Ann Arbor, MI, USA) for 30 min. Fluorescence intensity was measured at 515 nm (λ_ex = 482 nm) using a fluorophotometer (TECAN, Männedorf, Switzerland).

2.6. Membrane protein extraction

Following a 6 h fast, mice were administered insulin intraperitoneally (100 IU/kg body weight). Gastrocnemius skeletal muscle tissues were harvested 15 min after insulin injection. Plasma membrane proteins were isolated using a subcellular protein fractionation kit Mem ™ Plus (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions. Isolated fractions were stored at −80 °C until further analysis.

2.7. Western blot analysis

After a 6 h fast, mice were injected intraperitoneally with insulin (100 IU/kg body weight). Liver and skeletal muscle tissues were harvested 15 min after insulin administration. Samples were treated with lysis buffer (Cellytic, Sigma-Aldrich) and then added to sample buffer and heated to 100 °C for 5 min and stored at −80 °C until analysis. 20 μg of total protein were separated by sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) and transferred onto a polyvinylidene difluoride (PVDF) membrane. The membrane was blocked with 5% skim milk in Tris-buffered saline with 0.1% Tween 20 (TBST) for 1 h at room temperature, followed by incubation with primary antibodies against phospho-AKT (Ser473), AKT, GLUT4, eNOS, Phospho-eNOS and GAPDH (Cell Signaling Technology, Danvers, MA, USA) overnight at 4 °C. Membranes were fixed in 0.5% skim milk with species-appropriate secondary antibodies for 1 h. Chemiluminescence was measured using an analyzer (ChemiDoc Imaging System; Bio-Rad, Hercules, CA, USA). Densitometry analysis was performed using ImageJ software [20].

2.8. Real-time PCR

Mesenteric white adipose tissue (mWAT), epididymal white adipose tissue (eWAT), and brown adipose tissue (BAT) were harvested from 16-week-old mice and immediately processed for RNA isolation. Total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. RNA concentration and purity were assessed spectrophotometrically. Complementary DNA (cDNA) was synthesized from equal amounts of RNA using a reverse transcription kit (Bio-Rad). Quantitative real-time PCR was performed using SYBR Green Master Mix (Bio-Rad) on a CFX96 Touch Real-Time PCR Detection System (Bio-Rad). Gene expression levels were normalized to the appropriate housekeeping gene, and relative expression was calculated using the ΔΔCt method. Primer sequences used for amplification are listed in Supplemental Table 1.

2.9. Vascular immunohistochemical staining

Vascular immunohistochemical staining was performed using rabbit anti-CSE (1:500) (Cell Signaling Technology, Danvers, MA, USA) and rat anti-mouse CD31 (1:500) (BD Pharmingen) followed by nuclear counterstain with 4′,6-diamidine-2′-phenylindole dihydrochloride (DAPI) and imaged using epifluorescent microscopy as we've previously described [21,22].

2.10. Statistical analysis

Data are presented as means ± SEM. Statistical significance was determined using GraphPad prism 9 (GraphPad Software, San Diego, CA, USA). Either Student's t-test or Tukey's multiple comparisons test were used. A p value < 0.05 was considered significant.

3. Results

3.1. ecCTH-KO mice increased body weight with high fat diet

We developed constitutive ecCTH-KO mice in C57BL/6J background by generating a CTH^tm1c conditional mutant allele obtained from the European Mutant Mouse Archive (EMMA). Tm1a mice were first crossed with Flp recombinase-expressing mice to produce the tm1c allele, which was then bred with VE-Cadherin Cre⁺ driver mice to generate the endothelial-specific ecCTH^Δ/Δ mutant allele (Sup Fig. 1A). Analysis of skeletal muscle revealed a significant reduction in CTH mRNA expression in ecCTH-KO mice. Correspondingly, bound sulfide and acid-labile sulfide levels measured by MBB method were significantly decreased in skeletal muscle (Sup. Fig. 1B). Vascular co-staining of CTH and CD31 in the aorta confirmed efficient and endothelial-specific deletion of CTH (Sup. Fig. 1C). As skeletal muscle contains micro vessels, the observed reduction of CTH in muscle tissue reflects loss of endothelial CTH rather than muscle cell–intrinsic expression. To determine whether ecCTH-KO mice gained weight, body weight was measured weekly. Under the high-fat diet (HFD), both WT and ecCTH-KO mice exhibited significant weight gain compared to those on a control chow diet. However, ecCTH-KO mice gained significantly more weight than WT mice on HFD (Fig. 1 A &B). In contrast, under the Chow diet, there was no significant difference in weight gain between WT and ecCTH-KO mice (Supp Fig. 2B). Total body weight was significantly increased in ecCTH-KO mice compared with WT mice under high-fat diet conditions, whereas no significant difference was observed between genotypes when fed a chow diet (Supp Fig. 2A). At the start of the experiment, the baseline body weights of WT and ecCTH-KO mice were comparable, measuring 25.3 ± 0.42 g and 24.3 ± 0.55 g, respectively. After 8 weeks of high-fat diet HFD feeding, both groups gained weight; however, ecCTH-KO mice exhibited significantly greater body weight, reaching 35.6 ± 1.20 g, compared to 32.9 ± 0.73 g in WT controls (Supp. Table 2). This differential weight gain was further confirmed by time-course analysis, which showed a consistent elevated amount of body weight in ecCTH-KO mice throughout the study period. To determine whether the observed weight gain was due to increased caloric intake, food consumption was carefully monitored. Average energy intake over the 8-week period was similar between WT and ecCTH-KO mice (Fig. 1C), and no significant differences were found in either cumulative weekly intake (Fig. 1D) or week-by-week caloric intake. These findings suggest that increased weight gain in ecCTH-KO mice is not attributable to hyperphagia.

Interestingly, despite similar food consumption, ecCTH-KO mice demonstrated significantly higher feed efficiency during the observation period (Fig. 1F), indicating a greater conversion of consumed energy into body mass. This suggests that endothelial CTH deficiency may impair energy expenditure, possibly through mechanisms affecting mitochondrial function, or inflammatory responses, thus promoting increased fat accumulation despite unchanged caloric intake.

3.2. ecCTH-KO mice showed decreased polysulfide levels in skeletal muscle and white adipose tissue and increased cholesterol profile

To explore the relationship between weight gain and alterations in sulfide metabolism, we next determined hydrogen sulfide (H₂S) and polysulfides in various tissues. Specifically, we used SF7-AM fluorescence probe to detect free H₂S and SSP4 fluorescence probe to measure per- and polysulfide (sulfane sulfur) species (Fig. 2). In ecCTH-KO mice, fluorescence intensity of the SF7 probe remained unchanged across key metabolic tissues—including plasma (Fig. 2 A), liver (Fig. 2B), skeletal muscle (Fig. 2C, mesenteric white adipose tissue (mWAT) (Fig. 2D), brown adipose tissue (BAT) (Fig. 2E), and epididymal adipose tissue (eWAT) (Fig. 2F)—compared to wild-type (WT) controls. Unchanged SF7 fluorescence intensity indicated that there was no change in free hydrogen sulfide in those tissues. In contrast, SSP4 fluorescence, which reflects sulfane sulfur content, showed tissue-specific reductions. While plasma (Fig. 2A), liver (Fig. 2B), and brown adipose tissue (BAT) (Fig. 2E) exhibited comparable SSP4 fluorescence between WT and ecCTH-KO mice, a marked decrease in SSP4 signal was observed in skeletal muscle (Fig. 2C), mWAT (Fig. 2D), and epididymal white adipose tissue (eWAT) (Fig. 2F) of ecCTH-KO mice. To further investigate metabolic alterations, we analyzed circulating lipid profiles. Plasma levels of total cholesterol, triglycerides, and free fatty acids were significantly elevated in ecCTH-KO mice compared to WT controls (Supp Fig. 3), indicating systemic dyslipidemia. These data reinforce the notion that disruption of endothelial CTH-derived sulfide metabolism contributes to metabolic dysfunction.

Fig. 2.

Sulfide metabolites in HFD-fed WT and ecCTH-KO mice. Eight-week-old mice were fed HFD for 8 weeks. Fluorescence-based detection of sulfide metabolites was performed at the 8-week time point. SSP4 fluorescence reflects sulfane sulfur levels, and SF7 fluorescence indicates free hydrogen sulfide. Measurements were conducted in (A) plasma, (B) liver, (C) skeletal muscle, (D) mesenteric white adipose tissue (mWAT), (E) brown adipose tissue (BAT), and (F) epididymal white adipose tissue (eWAT). Data are expressed as mean ± SEM. Statistical significance is indicated by *p < 0.05, **p < 0.005, ***p < 0.001, ****p < 0.0001. N = 3-7.

3.3. ecCTH-KO mice gained body weight by increasing fat mass

To further understand whether the observed increase in body weight in ecCTH-KO mice on a high-fat diet (HFD) was primarily due to increased fat accumulation, we analyzed the weights of the liver, and multiple adipose tissue depots, including epididymal white adipose tissue (eWAT), mesenteric white adipose tissue (mWAT), and brown adipose tissue (BAT) (Fig. 3A–F). While liver weights were comparable between wild-type (WT) and ecCTH-KO mice (Fig. 3B), hepatic triglyceride (TG) content was significantly higher in ecCTH-KO mice (Fig. 3C). This suggests that lipid accumulation in the liver increased. The most prominent differences were seen in white adipose depots: both eWAT and mWAT weights were significantly elevated in ecCTH-KO mice compared to WT (Fig. 3D–E), whereas BAT weight remained unaffected (Fig. 3F). These findings suggest selective expansion of white adipose tissue as a major contributor to the increased body weight in ecCTH-KO mice, reinforcing the role of endothelial CTH in regulating fat storage. Given that uncoupling protein 1 (UCP1) is a critical regulator of thermogenesis and energy dissipation, especially in brown and beige adipocytes, we next examined UCP1 mRNA expression across adipose tissues. Interestingly, UCP1 expression was significantly reduced in the eWAT of ecCTH-KO mice (Fig. 3H), but not in mWAT (Fig. 3H) or BAT (Fig. 3I). To evaluate whether mitochondrial biogenesis was also affected, we assessed the expression of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α), a master regulator of mitochondrial function. Importantly, PGC-1α levels were not significantly different in eWAT, mWAT, or BAT between ecCTH-KO and WT mice (Fig. 3J, K, and 3L), suggesting that mitochondrial biogenesis per se was not impaired.

Fig. 3.

Increased fat mass in ecCTH-KO mice. Eight-week old mice were fed HFD for eight weeks.(A) Representative body composition images of WT and ecCTH-KO mice; (B) liver weight; (C) hepatic triglyceride (TG) content; (D–F) fat mass of mesenteric white adipose tissue (mWAT), epididymal white adipose tissue (eWAT), and brown adipose tissue (BAT); (G–I) mRNA expression of UCP1 in mWAT, eWAT, and BAT; (J–L) mRNA expression of PGC-1α in mWAT, eWAT, and BAT. Data are expressed as mean ± SEM. Statistical significance is indicated by *p < 0.05, **p < 0.005, ***p < 0.001, ****p < 0.0001. N = 5-6.

3.4. ecCTH-KO mice showed an increased inflammatory phenotype in WAT

Given the well-established link between inflammation in white adipose tissue (WAT) [23] and the established link between H₂S and inflammation [24], we sought to investigate the inflammatory profile of different adipose depots in ecCTH-KO mice. We focused on the expression of key pro-inflammatory cytokines—monocyte chemoattractant protein-1 (MCP-1), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-α)—at the mRNA level in mWAT, eWAT, and BAT (Fig. 4). Our results revealed a significant increase in MCP-1 (Fig. 4 A) and IL-6 (Fig. 4D) expression specifically in the mWAT of ecCTH-KO mice, pointing to a heightened inflammatory state within this visceral fat depot. MCP-1 is known to recruit monocytes/macrophages to adipose tissue, while IL-6 is a key mediator of systemic inflammation and metabolic dysregulation. The elevation of both markers in mWAT suggests that endothelial H₂S deficiency promotes a local inflammatory milieu that could disrupt insulin signaling and exacerbate metabolic dysfunction. Interestingly, in eWAT, no significant alterations in MCP-1 or IL-6 expression were observed (Fig. 4B–G), suggesting that the inflammatory response in ecCTH-KO mice may be depot-specific rather than systemic across all WAT. Furthermore, BAT, which typically plays a protective role against obesity through thermogenesis, showed a significant change in MCP1expression, but no change in IL-6 expression (Fig. 4C–F). TNF-α, another well-known pro-inflammatory cytokine implicated in insulin resistance, did not exhibit any significant changes in mRNA expression in mWAT, eWAT, or BAT (Fig G-I). This observation implies that the inflammation in ecCTH-KO mice is not uniformly mediated by all cytokines but may be driven primarily by MCP-1 and IL-6 in mWAT. Collectively, these findings demonstrate that the inflammatory response associated with endothelial CTH deficiency is localized primarily to mesenteric fat with secondary impact on BAT and characterized by selective cytokine expression. This depot-specific inflammatory response may contribute to metabolic dysfunction and obesity-associated complications in ecCTH-KO mice.

Fig. 4.

Inflammatory cytokine mRNA expression in HFD-fed WT and ecCTH-KO mice.

MCP1, IL-6, and TNF-α mRNA expression were measured by real-time PCR: (A–C) MCP1 mRNA expression in mesenteric white adipose tissue (mWAT), epididymal white adipose tissue (eWAT), and brown adipose tissue (BAT), respectively; (D–F) IL-6 mRNA expression in mWAT, eWAT, and BAT, respectively; and (G–I) TNF-α mRNA expression in mWAT, eWAT, and BAT, respectively. Data are expressed as mean ± SEM. Statistical significance is indicated by *p < 0.05, **p < 0.005, ***p < 0.001, ****p < 0.0001. N = 5-6.

3.5. ecCTH deficiency impairs glucose homeostasis via impaired AKT-dependent GLUT4 translocation

Given that ecCTH-KO mice exhibited increased body weight and an enhanced inflammatory response, we further examined their glucose metabolism responses to assess potential impairments in glucose tolerance and insulin sensitivity. We performed a glucose tolerance test (GTT) to evaluate glucose tolerance. Baseline blood glucose or insulin levels showed no significant differences between WT and ecCTH-KO (Supp fig. 3 D) mice. However, following glucose administration, blood glucose levels in ecCTH-KO mice were significantly elevated at 30 min post-injection and remained higher than those in WT mice at 120 min. Area under the curve (AUC) analysis further confirmed a significant increase in glucose intolerance in ecCTH-KO mice compared to WT controls (Fig. 5A). Next, we performed an insulin tolerance test (ITT) and found ecCTH-KO mice exhibited significantly higher blood glucose levels and AUC values, indicative of exacerbated insulin resistance (Fig. 5B). To further assess insulin sensitivity, we measured fasting blood glucose levels and calculated the homeostatic model assessment for insulin resistance (HOMA-IR). HOMA-IR values were significantly elevated in ecCTH-KO mice, further confirming insulin resistance (Fig. 5B). To investigate the molecular mechanisms underlying insulin resistance in ecCTH-KO mice, we examined key insulin signaling pathways in skeletal muscle and liver. Insulin is known to activate AKT and stimulate phosphorylation of endothelial nitric oxide synthase (eNOS) activation promoting nitric oxide–mediated vasodilation and enhanced perfusion of insulin-sensitive tissues such as skeletal muscle [25]. In addition, AKT activation is essential for the translocation of GLUT4-containing vesicles from intracellular compartments to the plasma membrane, thereby facilitating glucose uptake [26]. Consistent with impaired insulin signaling, we observed that insulin-stimulated phosphorylation of AKT (p-AKT) was significantly reduced in both liver (Fig. 5C) and skeletal muscle (Fig. 5D) of ecCTH-KO mice compared with WT controls. Similarly, insulin-induced phosphorylation of eNOS (peNOS) was markedly reduced in skeletal muscle of ecCTH-KO mice (Supp. Fig. 4). We next assessed GLUT4 distribution because GLUT4 is the principal glucose transporter responsible for insulin-mediated glucose uptake in skeletal muscle and its membrane translocation is regulated by AKT [27]. Total GLUT4 protein levels in skeletal cytosol were comparable between WT and ecCTH-KO mice (Fig. 5E). However, membrane-associated GLUT4 was significantly decreased in skeletal muscle from insulin-stimulated ecCTH-KO mice (Fig. 5F), indicating defective GLUT4 translocation rather than altered expression. Collectively, these findings demonstrate that endothelial CTH deficiency impairs insulin signaling by reducing AKT activation, leading to diminished GLUT4 translocation to the plasma membrane and reduced glucose entry into skeletal muscle cells.

Fig. 5.

Glucose homeostasis and insulin signaling in liver and skeletal muscle of HFD-fed mice. (A) Glucose tolerance assessed by intraperitoneal glucose tolerance test (GTT) with corresponding area under the curve (AUC); (B) insulin sensitivity determined by intraperitoneal insulin tolerance test (ITT), ITT AUC, and HOMA-IR index; (C–G) insulin signaling pathway analysis by western blotting: (C) phosphorylated and total AKT in liver, (D) phosphorylated and total AKT in gastrocnemius skeletal muscle, (E) GLUT4 protein levels in the cytosolic fraction, and (F) in the plasma membrane of skeletal muscle. Western blotting samples were collected 15 min after insulin injection. Data are expressed as mean ± SEM. Statistical significance is indicated by *p < 0.05, **p < 0.005, ***p < 0.001, ****p < 0.0001. N = 3-8.

4. Discussion

This study revealed an unexpected role of endothelial CTH in regulating adipose and skeletal muscle metabolism under a high-fat diet. We observed specific reduction in sulfane sulfur levels in skeletal muscle, mWAT, and eWAT of ecCTH-KO mice. We also observed increased fat mass accumulation and elevated levels of inflammatory cytokines in WAT, along with higher plasma free fatty acid concentrations. Such alterations are known to contribute to impaired glucose tolerance and insulin resistance which was also observed. Collectively, our findings identify endothelial sulfide metabolism as an important mediator and possible future therapeutic target in regulating glucose metabolism and development of obesity.

Skeletal muscle function is heavily influenced by its surrounding microvasculature, which ensures adequate oxygen and nutrient supply. Microvascular endothelial cells regulate blood flow and contribute to metabolic balance [[28], [29], [30], [31]]. Alterations in vascular function, as seen in conditions such as Type 2 diabetes, are closely linked to skeletal muscle dysfunction and metabolic disturbances [[32], [33], [34]]. Studies have shown that vascular damage or dysfunction impairs skeletal muscle microcirculation, contributing to decreased muscle function and increased adiposity [35]. Recently, a study showed that endothelial cells directly regulate muscle mass by releasing the Notch receptor ligand delta-like 4 (Dll4), which activates Notch signaling pathways leading to muscle atrophy [36]. Furthermore, endothelial dysfunction may result in reduced capillary density and perfusion, leading to impaired oxygen and nutrient transport and, consequently, metabolic imbalance [34,37]. Endothelial dysfunction is not only associated with muscle atrophy but also with obesity and insulin resistance. Reduced skeletal muscle capillary density has been linked to insulin resistance, as capillaries serve as conduits for insulin delivery from pancreatic β-cells to skeletal muscle. Defects in endothelial insulin signaling impair insulin transport in skeletal muscle, reducing interstitial insulin concentrations and consequently decreasing insulin-mediated glucose uptake [10,38]. Studies from obese human cohorts have demonstrated a reduction in the arterial-to-lymph insulin gradient during hyper insulinemic-euglycemic clamp studies, supporting the hypothesis that endothelial dysfunction contributes to insulin resistance [39]. While numerous studies have shown that sulfide species regulate endothelial function and that endothelial dysfunction is involved in insulin resistance, there has been no knowledge regarding the link between endothelial sulfide production and insulin resistance in skeletal muscle and obesity. Our results provide clear evidence that deficiency of endothelial sulfide production in ecCTH-KO is associated with insulin resistance in skeletal muscle in the HFD model.

Several previous studies have demonstrated an association between CTH and insulin signaling. A recent study demonstrated that elevated levels of glucose, palmitate, and oleate significantly suppress both hydrogen sulfide (H₂S) production and CTH expression in mouse C2C12 myoblasts. Skeletal muscle from db/db mice exhibits reduced CTH expression, which can be restored through administration of sodium hydrosulfide (NaHS), an H₂S donor [40]. Further evidence suggests that the CTH plays a regulatory role in glucose transporter 4 (GLUT4) expression in cultured muscle cells. Specifically, CTH knockdown in C2C12 myotubes results in diminished GLUT4 expression and impaired glucose uptake [41]. Moreover, supplementation with l-cysteine, the metabolic precursor of H₂S, has been shown to enhance GLUT4 translocation to the plasma membrane and promote glucose uptake in C2C12 cells [42]. In vivo studies have shown that treatment with GYY4137, a slow-releasing H₂S donor, improves insulin resistance in high-fat diet-fed mice [43]. Likewise, administration of Na₂S has been reported to enhance insulin-stimulated glucose utilization by promoting activation of the insulin signaling pathway [44]. A recent study has shown that insulin signaling in skeletal muscle is impaired in muscle-specific CTH knockout mice [45]. However, none of these studies have linked endothelial CTH/sulfide signaling with skeletal muscle insulin signaling. Our study provides intriguing insights into the role of endothelial-derived reactive sulfur species in regulating skeletal muscle insulin signaling and glucose metabolism that require further study.

We have previously demonstrated that endothelial-dependent sulfide production critically regulates cardiomyocyte function [46]. In the present study, our data demonstrate that endothelial-specific deletion of CTH leads to impaired glucose tolerance and insulin resistance, which are mechanistically linked to defective AKT signaling and reduced GLUT4 translocation in skeletal muscle. These findings uncover a previously unrecognized role for endothelial CTH-derived sulfide signaling in coordinating vascular–metabolic crosstalk during insulin action. Insulin signaling requires coordinated activation of intracellular pathways and adequate tissue perfusion. Insulin stimulates AKT phosphorylation and activates endothelial nitric oxide synthase (eNOS) [25]. Activation of eNOS promotes nitric oxide–mediated vasodilation and enhancing delivery of insulin and glucose to peripheral tissues, particularly skeletal muscle. Insulin-depending phosphorylation of eNOS was significantly diminished in ecCTH-KO skeletal muscle, suggesting impaired endothelial responsiveness and potentially reduced microvascular perfusion of insulin-sensitive tissues. AKT activation drives vesicle-dependent translocation of GLUT4 to the plasma membrane [26]. In ecCTH-KO mice, insulin-stimulated AKT phosphorylation was markedly blunted in both liver and skeletal muscle, indicating systemic defects in insulin signaling. Importantly, despite normal total GLUT4 expression, membrane-associated GLUT4 was significantly decreased in skeletal muscle of insulin-stimulated ecCTH-KO mice. This finding indicates that endothelial CTH deficiency disrupts glucose uptake primarily by impairing GLUT4 vesicle translocation rather than by altering GLUT4 abundance. Because AKT activation is a key upstream regulator of GLUT4 trafficking, the reduced AKT phosphorylation observed in ecCTH-KO mice provides a direct mechanistic link between endothelial CTH loss and defective skeletal muscle insulin signaling. Mechanistically, these findings suggest that endothelial-derived sulfane sulfur species facilitate insulin signaling through AKT activation, thereby enabling efficient GLUT4 trafficking to the plasma membrane and glucose uptake in skeletal muscle. Collectively, our data identify endothelial CTH/sulfane sulfur signaling as a previously unrecognized regulator of skeletal muscle insulin responsiveness and systemic glucose homeostasis.

Using ecCTH-KO mice, we demonstrate that sulfane sulfur rather than hydrogen sulfide is associated with impaired insulin signaling in skeletal muscle. Most previous studies have focused on the role of free H₂S in insulin signaling and obesity. However, our findings reveal an important role for sulfane sulfur modulating insulin signaling in skeletal muscle. Importantly, many physiological effects previously attributed to H₂S are now known to be mediated through its conversion to sulfane sulfur, making sulfane sulfur the true bioactive mediator in numerous cellular processes [47]. Recent scientific advances have highlighted the superior biological properties of sulfane sulfur over free hydrogen sulfide. Sulfane sulfur species, including persulfides (R–SSH), polysulfides, and protein-bound sulfur, exhibit several advantages over hydrogen sulfide (H₂S) in biological systems. They act as more stable, bioavailable reservoirs of sulfur compared to volatile and rapidly cleared H₂S. This stability enables prolonged biological activity and controlled H₂S release when needed [48]. Furthermore, sulfane sulfur species exhibit greater nucleophilicity and, in their protonated forms, electrophilicity than H₂S, enabling more selective and potent interactions with protein thiols and other biological targets [49,50]. Sulfane sulfur also plays a direct role in post-translational modifications, particularly in S-sulfhydration of protein cysteine residues. These modifications often protect cysteines from irreversible oxidative damage while fine-tuning redox-sensitive signaling networks [51], and demonstrate superior redox-regulatory capacity by modulating oxidative stress through reversible thiol modifications [52]. Future studies will be necessary to define precise signaling pathways through which sulfane sulfur–mediated redox regulation influences insulin signaling.

Obesity and dysfunctional insulin signaling are closely interconnected. In our study, we observed that ecCTH deletion significantly reduced sulfane sulfur levels in WAT, particularly in mWAT and eWAT depots in HFD-fed mice. Correspondingly, both mWAT and eWAT weights were markedly increased. These findings align with previous reports indicating a role for CTH in adipose tissue function. Previous research has shown that high-fat diet (HFD) feeding results in decreased H₂S bioavailability and altered expression of sulfur-metabolizing enzymes in adipose tissue [53,54] and liver [55]. Genetic deletion of sulfur-producing enzymes has been associated with lipid accumulation in liver [56] under HFD conditions, while supplementation with H₂S donors has been shown to promote lipid metabolism and reduce fatty liver development [57]. However, none of the previous studies have explored inflammation in WAT as a consequence of sulfide deficiency leading to obesity. In our current report, we showed that sulfane sulfur deficiency is associated with inflammation in WAT. It is well established that inflammation in WAT, particularly in visceral depots, contributes significantly to obesity and insulin resistance. Visceral adipose tissue (VAT) is recognized as the most metabolically active and pro-inflammatory fat depot [58]. It secretes elevated levels of cytokines such as TNF-α, IL-6, and MCP-1 and is heavily infiltrated by M1 macrophages, which sustain chronic low-grade inflammation. This inflammatory state is further exacerbated by adipocyte hypertrophy and hypoxia, promoting insulin resistance and metabolic dysfunction [59]. In line with these previous studies, we found that inflammation is higher in mWAT (i.e. visceral adipose tissue) as evidence by higher expression of IL-6 and MCP1. Interestingly, we also observed increased MCP-1 expression in BAT of HFD-fed ecCTH-KO mice, despite no measurable reduction in sulfane sulfur levels in BAT. This raises the question of how BAT becomes inflamed in the absence of local sulfane sulfur deficiency. One plausible explanation is that sulfane sulfur deficiency in WAT may initiate inflammation that secondarily affects BAT. Crosstalk between WAT and BAT is well documented and it has been clearly demonstrated that WAT and BAT can interact through cytokines [60]. Infiltrating macrophages and other immune cells in WAT impair the ability of precursor cells to differentiate into thermogenically active beige adipocytes by secreting pro-inflammatory cytokines and creating an inflammatory microenvironment [61]. Furthermore, WAT inflammation may exert a secondary negative impact on BAT, adding a novel layer to the understanding of adipose tissue crosstalk and the systemic consequences of sulfur metabolism dysregulation. Future studies will be needed to better understand mechanisms of ecCTH dependent sulfide production on BAT inflammation.

A recent study demonstrated that conditional or constitutive absence of CTH renders mice extremely sensitive to dietary cysteine removal [62]. In their model, CTH-KO mice subjected to a cysteine-free diet lost approximately 30% of their body weight within one week, a phenotype that was rapidly reversible upon cysteine supplementation. This dramatic weight loss was attributed to cysteine deficiency-induced activation of the integrated stress response (ISR) pathway (GCN2 → eIF2α → ATF4) and oxidative stress responses, resulting in elevated levels of GDF15 and FGF21, which collectively contributed to anorexia and the observed weight loss. In contrast, our study revealed that endothelial cell–specific CTH knockout mice exhibited increased body weight when maintained on a high-fat diet. A key distinction between the two studies lies in the dietary intervention: the previous work employed a cysteine-restricted diet, whereas our high-fat diet contained cysteine. Moreover, while cysteine restriction led to reduced food intake in their model, we did not observe any difference in food intake between wild-type and ecCTH-KO mice on a high-fat diet. This discrepancy suggests that complete dietary cysteine deficiency may trigger a shutdown of protein synthesis, leading to anorexia and pronounced weight loss, whereas in our model, ecCTH deletion primarily caused endothelial dysfunction, impaired insulin signaling in skeletal muscle, and inflammation in white adipose tissue, which contributed to increased weight gain. While our study provides new insights into the metabolic consequences of endothelial CTH deletion under a high-fat diet, it remains unknown how ecCTH-KO mice would respond to a cysteine-restricted diet. Future studies combining endothelial-specific CTH deletion with dietary cysteine restriction could help delineate the relative contributions of cysteine metabolism in endothelial cells to systemic energy homeostasis and stress responses. Lastly, a previous study by Guo et al. demonstrated that global CSE deficiency exacerbates high-fat diet–induced obesity, insulin resistance, and glucose intolerance, primarily through hepatic mechanisms involving FoxO1-dependent gluconeogenesis [63]. Consistent with these findings, our study shows that endothelial cell–specific CTH deficiency promotes greater body weight gain, increased fat mass, dyslipidemia, glucose intolerance, and insulin resistance under HFD conditions, despite unchanged food intake. However, our results extend the prior work by identifying a critical endothelial contribution to systemic metabolic regulation, characterized by reduced polysulfide (sulfane sulfur) rather than H2S levels in skeletal muscle and white adipose tissue, depot-specific adipose inflammation (notably increased MCP-1 and IL-6 in mesenteric WAT), impaired insulin signaling (reduced p-AKT, peNOS, and GLUT4 translocation), and altered thermogenic regulation (reduced UCP1 in eWAT). While the Gao et al. study focused mainly on liver-centric mechanisms of insulin resistance, our findings highlight a vascular–endothelial sulfide axis.

Our study has several limitations. First, indirect calorimetry was not performed; therefore, we cannot precisely define the contribution of whole-body energy expenditure or substrate utilization to the observed metabolic phenotype. Future studies incorporating indirect calorimetry will strengthen mechanistic insights. Second, wild type controls were not littermates of ecCTH-KO mice. Although animals shared the same backcrossed genetic background and housing conditions, potential effects of genetic drift, early-life upbringing, microbiome composition, or genotype-specific housing, and parental effects cannot be fully excluded. To minimize genetic variability, our laboratory routinely employs speed congenic breeding with microsatellite marker genotyping. Third, glucose and insulin tolerance tests were dosed by total body weight rather than lean mass, which may influence absolute metabolic readouts, particularly in obese mice [64]. Fourth, the high-fat diet exposure was a shorter time of 8 weeks, modeling early metabolic stress rather than advanced obesity. Longer-term high-fat feeding (e.g., 16 weeks) may produce more pronounced metabolic phenotypes. This may reveal additional genotype-dependent effects and will be examined in future studies. Nevertheless, consistent relative differences across experimental groups and reproducibility across independent experiments support the robustness of our conclusions.

In conclusion, our findings provide unique evidence that endothelial CTH and sulfane sulfur deficiency leads to dysregulation of skeletal muscle insulin signaling and inflammation in white adipose tissue (WAT), ultimately contributing to metabolic dysfunction. These findings suggest that targeting endothelial CTH activity and production of sulfane sulfur may offer an important new therapeutic strategy for improving insulin sensitivity and preventing metabolic diseases including obesity.

Duality of interest

The corresponding author is a co-editor in chief of the journal but was not involved in the handling or peer review of the manuscript.

CRediT authorship contribution statement

Takashi Yagi: Data curation, Investigation, Methodology, Writing – original draft. Shafiul Alam: Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. Saima I. Niti: Methodology. Sibile Pardue: Investigation, Methodology. Xinggui Shen: Investigation, Methodology, Writing – review & editing. John D. Glawe: Investigation, Methodology. Gopi K. Kolluru: Conceptualization, Methodology, Writing – review & editing. Tomohiro Tanaka: Conceptualization, Methodology, Validation, Writing – review & editing. Christopher G. Kevil: Conceptualization, Supervision, Writing – original draft, Writing – review & editing.

Declaration of competing interest

The author is an Editorial Board Member/Editor-in-Chief/Associate Editor/Guest Editor for this journal and was not involved in the editorial review or the decision to publish this article.

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: The authors have filed intellectual property regarding the role of endothelial CSE in regulating diet induced obesity.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

No data was used for the research described in the article.