Targeted Disruption of Selenocysteine Lyase in Brown Adipocytes Controls Glutathione Peroxidase 1 and 4 Expression in Males

Abstract

Brown adipocytes (BA), the predominant cell type in brown adipose tissue (BAT), are essential for adaptive thermogenesis in response to environmental temperature or diets that impact energy expenditure. BAT thermogenic activation is regulated by thyroid hormones (TH) and norepinephrine, with TH activation carried out by the selenoprotein class of deiodinases, making BAT reliant on the micronutrient selenium (Se). Se is utilized to synthesize selenocysteine (Sec), incorporated into selenoproteins. The enzyme Sec lyase (Scly) decomposes Sec to provide selenide for selenoprotein synthesis. Besides deiodinases, glutathione peroxidases (GPXs) are critical selenoproteins for antioxidant defense and redox balance in BAT. Whole-body Scly knockout (KO) mice exhibit obesity, glucose intolerance, fatty liver, and BAT whitening. However, the specific tissue where Scly loss drives this phenotype remains unclear. As BAT regulates energy expenditure and glucose metabolism, we hypothesized either hepatocyte or BA Scly deletion drives the observed phenotype in Scly KO mice. We generated hepatocyte-specific and BA-specific Scly KO mice and assessed metabolic and thermogenic outcomes. Hepatocyte-specific Scly KO mice showed no metabolic phenotype, suggesting hepatic Scly is dispensable. BA-specific Scly KO mice exhibited normal glucose and insulin tolerance. Under Se-deficiency, male BA-specific Scly KO mice recovered body temperature after initial cold-induced thermogenic impairment. Only male mice showed reduced expression of GPX1 and GPX4 in BAT across diets, without TH impairments. These findings demonstrate that Scly in BA supports local Se homeostasis and selenoprotein expression in a sex- and Se-dependent manner, with its loss leading to transient thermoregulatory impairment, contributing to the Scly KO phenotype.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12011-025-04904-7.

Introduction

An imbalance between energy intake and energy expenditure, specifically when intake exceeds energy expenditure, leads to excess energy being stored as lipids, resulting in dyslipidemia or increased adiposity depending on the types of fats. Over time, this imbalance may progress to obesity and elevate the risks of metabolic disorders, such as insulin resistance, impaired glucose tolerance, and type 2 diabetes [1, 2]. Brown adipocytes (BA) are the primary cells in brown adipose tissue (BAT), which is a specialized fat depot that dissipates fat-stored energy by generating heat to help maintain core body temperature, serving as a thermogenic organ. BAT is essential to non-shivering adaptive thermogenesis, i.e., the ability to regulate heat production in response to changes in the environmental temperature or to hypercaloric diets [3], thereby playing an important role in regulating energy balance [4, 5]. BAT is a mitochondria-rich tissue, and its thermogenic function is mediated by the unique expression of uncoupling protein 1 (UCP1), a proton pump localized in the inner mitochondrial membrane, in BA, making it the primary site for adaptive thermogenesis [6, 7]. The activation of UCP1 catalyzes the proton leak across the mitochondrial membrane that uncouples oxidative phosphorylation from ATP synthesis in the respiratory chain. This proton leak results in reduced ATP synthesis and the generation of heat [7–9].

Selenium (Se), an essential micronutrient participating in glucose and lipid metabolic homeostasis, is heavily involved in BAT adaptive thermogenesis [10]. Se is a critical component in the synthesis of selenoproteins, including iodothyronine deiodinases (DIOs), a family of selenoproteins that regulate the activation and inactivation of thyroid hormones (TH). By removing an iodine, iodothyronine deiodinase type 2 (DIO2) converts thyroxine (T4) into the active TH triiodothyronine (T3) within BA and is particularly necessary for the adaptive thermogenesis process [11]. Mice without DIO2 exhibit impaired BAT development and adaptive thermogenesis when exposed to cold temperatures or when subjected to caloric overload [12]. Other selenoproteins that participate in regulating BAT function by modulating redox homeostasis and antioxidant defense include glutathione peroxidase GPX1, critical to reduce hydrogen peroxides, and GPX4 which reduces lipid peroxides [13, 14]. Se is also critical for the synthesis of fully loaded selenoprotein P (SELENOP), which functions primarily as a Se transport protein supplying Se to other tissues, while also contributing to redox balance through enzymatic action. When SELENOP is deficient in BAT, mice display impaired thermogenesis [15].

Maintenance of selenoprotein production in active tissues like BAT relies on the regulation of Se homeostasis. Obtained through diet in both organic and inorganic forms, Se is converted into selenide, utilized for the production of the Se-containing amino acid selenocysteine (Sec). Efficient recycling of Se in cells, particularly under Se-deficient conditions, depends on the enzyme selenocysteine lyase (Scly), which can decompose Sec from metabolized organic Se forms or possibly from selenoprotein degradation to produce selenide [16]. Selenide is then delivered to selenophosphate synthetase 2 (SEPHS2) for selenophosphate biosynthesis, an intermediate essential for the synthesis of selenoproteins [17]. Thus, Scly-mediated Sec decomposition is important for sustaining selenoprotein synthesis and consequently redox homeostasis and energy metabolism in the BAT. Disruption of the Scly gene in mice (Scly KO) induces metabolic syndrome traits including obesity, dyslipidemia, glucose intolerance, hyperinsulinemia, and hepatic steatosis. These effects were exacerbated under normocaloric, Se-deficient diets in which their weight gain and hypercholesterolemia aggravated, accompanied by worsening of the fatty liver. Notably, the severity of these traits is sex-dependent, with females displaying milder phenotypes compared to male mice [18–20].

Given the characterized role of Scly in Se metabolism and the essential function of selenoproteins in BAT thermogenesis, we tested the hypothesis that the targeted loss of Scly either in hepatocytes or in BA would drive the metabolic dysfunction, leading to diminished BAT thermogenic capacity and response. We generated a mouse model with targeted deletion of Scly in hepatocytes or in BA. BA-Scly KO mice were fed with adequate and deficient levels of Se and exposed to cold (4℃) to activate BAT thermogenesis. Male mice displayed a normal core body temperature response to cold, yet reduced expression of selenoproteins GPX1 and GPX4, highlighting a specific role of Scly in the selenoprotein synthesis regulation. The liver-specific Scly KO had no remarkable impact on mouse phenotype nor on its selenoprotein synthesis. Neither model exhibited overt signs of metabolic dysfunction.

Methods

Chemicals and antibodies

All chemicals used in experiments were purchased from Sigma-Aldrich (St. Louis, MO) or Fisher Scientific (Waltham, MA), unless specified. Antibodies used in this study, their vendors and RRIDs are presented in Supplemental Table 1.

Animals and serum collection

Animal procedures were approved by the University of Hawaii Institutional Animal Care and Use Committee (IACUC) protocols #17–2521 and #17–2616. Mice were used for experiments in minimal numbers as calculated by a power analysis tool from the University of British Columbia, Canada, where mu1 = 0.97, mu2 = 0.95, with sigma = 0.12 and 80% power to provide significant results, and were based on previous experiments assessing metabolic phenotypes and selenoprotein expression in mice using specialized Se diets [18]. Hepatocyte-specific Scly KO mice (Alb-Cre-Scly KO) and BA-specific Scly KO mice (Ucp1-Cre-Scly KO) were generated by crossing Alb-Cre or Ucp1-Cre heterozygous (The Jackson Laboratory; Bar Harbor, ME) mice, respectively, with Scly-floxed mice, as previously described [21]. Scly-floxed mice lacking Cre served as the control group. Mice were group-housed, on a 12 h/12 h day/light cycle, kept at room temperature (23℃) or thermoneutrality (30℃) inside a rodent incubator (Powers Scientific, Inc., Pipersville, PA), and both sexes were used in experiments. Mice were fed a standard chow diet containing adequate Se levels from a Mineral Mix (0.2 ppm of sodium selenite) or a customized low sodium selenite (0.08 ppm) diet (Inotiv, Inc., Teklad Laboratory Animal Diets; West Lafayette, IN) for 8 weeks starting at weaning age. Mice were anesthetized via intraperitoneal injection of avertin (250 mg/kg), followed by cardiac puncture with a 26½- gauge needle to collect blood and euthanize the mice. BAT and liver were extracted and snap-frozen in liquid nitrogen. Blood was allowed to clot at room temperature for 15–30 min before centrifugation at 2000 x g for 10 min. Serum was carefully separated and stored at −80℃ until analysis.

Cold exposure experiment

Mice were provided with minimal bedding and food, then subjected to a six-hour exposure at 4 ℃ while remaining in their cage placed on a receiver platform. Internal core temperatures were continuously monitored and recorded during cold exposure using the VitalView system program (Version 5.1; Starr Life Sciences Corp., Oakmont, PA) using implanted thermoprobes (Starr Life Sciences Corp.). The thermoprobes were implanted via survival surgery in the posterior region of the peritoneal cavity. For the surgery, mice were anesthetized with 5% isoflurane in oxygen (1 L/min), and after confirmation of anesthesia by the absence of voluntary movement, the abdominal area was sanitized with betadine and 70% isopropyl alcohol (3 alternating scrubs), covered with a sterile drape prior to surgical incision to access the intraperitoneal cavity, and a sterilized thermoprobe was tied securely to the back skin with a 7 − 0 silk suture. After surgical incision closure, mice were monitored at least once daily, and single-housed throughout the recovery and experimental periods. Following a one-week recovery period, the cold exposure was carried out. Mice were euthanized via CO₂ asphyxiation or avertin plus cardiac puncture after six hours of cold exposure, and serum and tissues were extracted and snap-frozen in liquid nitrogen for molecular analysis.

Glucose tolerance test (GTT) and insulin tolerance test (ITT)

GTT and ITT were performed as previously described [18]. Briefly, mice were fasted overnight prior to the assessment. For GTT, mice received a single intraperitoneal injection of glucose (Sigma-Aldrich) at a dose of 1.0 g/kg body weight, prepared as a 10% (w/v) solution. For ITT, mice received a single intraperitoneal injection of human insulin (Humulin R, Eli Lilly; Indianapolis, IN) at a dose of 0.75 units/kg body weight prepared as a 0.1 U/mL solution. Blood glucose levels were measured immediately before injection (baseline) and at 30, 60, 80, and 120 min post-injection using a glucometer (OneTouch Ultra 2 m; Malvern, PA) and test strips (Ultra Test Strips Blue; Malvern, PA). Blood was obtained via tail nick using a sterile blade.

Triiodothyronine (T3) and Thyroxine (T4) Enzyme-Linked Immunosorbent Assays (ELISA)

Total T3 and T4 levels in mouse serum were measured using commercially available ELISA kits Total T3 (Abnova Corporation, Catalog No: KA0925; Taipei, Taiwan) and T4 (AccuDiag, Catalog No: 3149-15; Woodland Hills, CA). After ELISA was carried out, absorbance at 450 nm was measured using a SpectraMax Plate reader Mini AFL system (Molecular Devices; San Jose, CA) and analyzed with SoftMax Pro software (Version 7.1.2.1; Molecular Devices; RRID: SCR_014240). The ELISA kits were stored at 4℃ and all procedures were performed according to the manufacturers’ protocol.

Western blot

BAT and liver proteins were extracted using a 1X RIPA lysis buffer (Thermo Fisher; Waltham, MA) containing protease and phosphatase inhibitors (Cell Signaling; Danvers, MA). 10–20 µg of total protein were loaded into Criterion 4–20% SDS-PAGE gels (BioRad, Hercules, CA) and transferred to an Immobilon-FL^® membrane (Millipore-Sigma; Burlington, MA) overnight in a Criterion (BioRad) wet assembly of tris-glycine buffer with 9% methanol. Primary antibodies were incubated overnight at 4 °C with rotation while fluorescent-labeled secondary antibodies were incubated for 1 h at room temperature with rotation. Blots were imaged using a Li-Cor Odyssey Fc infrared imager (Li-Cor; Lincoln, NE) and bands analyzed with the ImageStudio software (Li-Cor). Primary and secondary antibodies used in the study and their concentrations are listed in Supplementary Table 1.

RNA extraction and real-time qPCR

One µg of total RNA extracted from the BAT using E.Z.N.A. Total RNA kit (Omega Biotek, Catalog No: R6834-02; Norcross, GA) was reverse transcribed using the High-Capacity kit (Applied Biosystems - ThermoFisher Scientific;). 10 ng of cDNA were used in real-time qPCR reactions for selected genes using specific primers, listed in Supplementary Table 2. qPCR reactions were carried out according to MIQE guidelines [22], using the PerfeCTa SYBR Green SuperMix (Quantabio; Beverly, MA) and amplified on a LightCycler 490 II (Roche; Indianapolis, IN). Gene expression was calculated using the ∆^Ct method after appropriate melting curves were generated. The Ct values were normalized to the geometric mean of the housekeeping genes 18s, Actb (beta-actin), and Gapdh, and results were plotted as fold change.

Selenium content measurement

Approximately half of each mouse BAT maintained on a chow diet was transferred to microcentrifuge tubes, flash-frozen on dry ice, and used to quantify total Se using inductively coupled plasma triple quadrupole mass spectrometry (ICP-QQQ-MS) at Brooks Applied Labs (Seattle, WA) using certified Se standards and reference materials. For the Se content in the serum, approximately 4–10 mg of BAT from each mouse maintained on a Se-deficient diet were digested in 10% nitric acid. Samples were diluted 1:5 or 1:10 in ultrapure water (Millipore Sigma, Burlington, MA) depending on tissue mass and homogenate consistency. An internal standard of 1 µL of 1 ppm gallium was added to each sample. 10 µL sample were pipetted onto siliconized quartz discs and heated at 70–80℃ to dry the sample onto the disc. These quartz discs were then read and analyzed for Se concentration using S4 T-STAR™ Total Reflection X-ray Fluorescence (TXRF; Bruker, Billerica, MA) using a molybdenum tube excitation at 50 kV/600 µA with an analysis time of 1000 s per sample. Trace element concentrations were calculated relative to the gallium internal standard and multiplied by the dilution factor.

Statistical analysis

Results were analyzed using GraphPad Prism 10 software (GraphPad, La Jolla, CA., RRID: SCR_002798). All data were expressed as the mean ± SEM. Normality was assessed using the D’Agostino and Pearson omnibus test. For the data that passed the normality test, the statistical significance between two groups was determined by Student’s t-test with Welch’s correction, one-way ANOVA with Tukey’s post hoc test for multiple groups with one variable, and two-way ANOVA for multiple groups and two variables. The null hypothesis was rejected for α < 0.05 with the two-tailed t-test, one-way, and two-way ANOVAs.

Results

BA-specific Scly Deletion Does not Affect Body Weight and Glucose Metabolism

We successfully generated BA-specific and liver-specific Scly KO mouse models. The breeding strategy and genotyping confirmation for each model are illustrated in Supplemental Figure S1. We determined whether BA-specific loss of Scly affects glucose and insulin responses in both control (Scly floxed) and Ucp1-Cre-Scly KO mice. Body weights were recorded weekly, and no significant differences were observed between control or Ucp1-Cre-Scly KO mice of either sex on chow (Fig. 1A-D). There were also no significant differences in BAT and liver weights between male and female mice groups (Fig. 1E-H). Effects on glucose metabolism were measured with a GTT. Ucp1-Cre-Scly KO mice on chow maintained their glucose tolerance (Fig. 1I) as shown by the comparable area under the curve (AUC) analysis (Fig. 1J). We next examined thyroid hormone levels and adaptive thermogenic responses by measuring core body temperatures during acute cold exposure.

Fig. 1.

Loss of Scly in the BAT does not affect weight gain and glucose tolerance under Se-adequate diet. (A, C) Body weights of male and female control and Ucp1-Cre-Scly KO mice over the course of 12 weeks on a chow diet; n = 4 for male and female mice groups. (B, D) Area under the curve (AUC) analysis of body weights. (E, F) Male BAT and liver weights after euthanasia; control mice n = 4, Ucp1-Cre-Scly KO mice n = 6 (G, H) Female BAT and liver weights; control mice n = 4, Ucp1-Cre-Scly KO n = 4. (I) Glucose tolerance test (GTT) in male control and Ucp1-Cre-Scly KO mice; control mice n = 4 and Ucp1-Cre-Scly KO mice n = 6 (J) AUC for GTT of males

Thyroid Hormone Levels and Thermogenic Capacity Were not Affected by BA-specific Deletion of Scly

Given the role of TH in thermogenesis regulation, serum T3 and T4 levels were measured. In male Ucp1-Cre-Scly KO mice, T3 levels were significantly elevated following cold exposure compared to control mice, while T4 levels remained unchanged (Fig. 2A, B). In females, no differences were observed in either T3 or T4 levels between genotypes (Fig. 2C, D). Similarly, mRNA expression of Dio2, a thyroid hormone-activating enzyme, were similar between control or Ucp1-Cre-Scly KO mice in both sexes (Fig. 1C, F). Male and female control and Ucp1-Cre-Scly KO mice housed under chow diet and room temperature, maintained core body temperatures during cold exposure (Fig. 1G - J). Expression of Pparg, a key regulator of adipocyte maturation and lipogenesis [23], was also unchanged in both male and female Ucp1-Cre Scly KO compared to control mice (Fig. 1. K). In line with these findings, protein levels of the thermogenic marker UCP1 were unaffected by the BA-specific deletion of Scly in either sex (Fig. 1. L, M).

Fig. 2.

Loss of Scly in BAT does not alter thyroid hormone levels or affect thermoregulation under Se-adequate diet. (A, B, D, E) Serum triiodothyronine (T3) and thyroxine (T4) levels in male and female control and Ucp1-Cre-Scly KO mice; male control mice n = 7 and Ucp1-Cre-Scly n = 8, female control mice n = 10 and Ucp1-Cre-Scly KO n = 8. (C, F, K) Dio2 and Pparg mRNA levels in male and female control and Ucp1-Cre-Scly KO mice on a chow diet; male control mice n = 4 and Ucp1-Cre-Scly n = 6, female control and Ucp1-Cre-Scly KO n = 5. (G, I) Core body temperatures of male and female mice during cold exposure at 4℃ for 6 h; male control and Ucp1-Cre-Scly KO n = 5, female control mice n = 3 and Ucp1-Cre-Scly KO mice n = 6. (H, J) AUC for body temperature measurements. (L) Western blot of UCP1 protein in BAT of male and female control and Ucp1-Cre-Scly KO mice; male mice groups n = 4, control female mice n = 3 and Ucp1-Cre-Scly KO mice n = 4. (M) Quantification of UCP1 protein in BAT of male and female mice. Values are mean ± SEM *p < 0.05 in (A)

Sex-specific reduction in GPX1 and GPX4 protein expression despite preserved Se levels and selenoprotein mRNA levels under Se-adequate conditions

To assess whether Scly deletion in BAT alters total Se content, we quantified total Se content in BAT and serum. Under a Se-adequate diet, Se levels in both BAT and serum were similar between control and Ucp1-Cre-Scly KO mice of either sex (Fig. 3A, B, F, G). We next evaluated gene expression of key BAT selenoproteins (Gpx1, Gpx4, and Selenop) [15, 24]. In male Ucp1-Cre-Scly KO mice, no significant changes in gene expression were observed compared to control mice (Fig. 3C-E). However, in female Ucp1-Cre-Scly KO mice, Gpx4 mRNA expression was upregulated by 50%, while the expression of all other genes remained unchanged (Fig. 3H-J). Although mRNA levels were mostly unaffected, protein expression analysis revealed that male Ucp1-Cre-Scly KO mice exhibited significantly lower levels of GPX1 and GPX4 compared to control mice, whereas UCP1, GPX3, SEPHS2, and TXNRD1 protein levels were maintained (Fig. 3K, L). In contrast, female Ucp1-Cre-Scly KO mice displayed no significant differences in GPX1 and GPX4 protein expression (Fig. 3M, N).

Fig. 3.

Loss of Scly in mouse BAT impacts in selenoprotein gene and protein expression. (A, F) Se content in BAT of male and female mice; n = 5 for both male and female groups. (B, G) Se content in serum of male and female mice; male control mice n = 4 and Ucp1-Cre-Scly n = 6, female control mice n = 4 and Ucp1-Cre-Scly KO n = 5. (C-E, H-J)) Gpx1, Gpx4, and Selenop mRNA levels in male and female control and Ucp1-Cre-Scly KO mice on a chow diet; male control mice n = 4 and Ucp1-Cre-Scly n = 6, female control and Ucp1-Cre-Scly KO n = 5. (K, M) Western blot of GPX1 and GPX4 protein in BAT of male and female control and Ucp1-Cre-Scly KO mice; male mice groups n = 4, control female mice n = 3 and Ucp1-Cre-Scly KO mice n = 4. (L, N) Quantification of GPX1 and GPX4 protein in BAT of male and female mice. Values are mean ± SEM *p < 0.05 and **p < 0.01 by t-test with a 95% confidence in (I) and (L)

Loss of Scly in BA combined with a Se-deficient diet does not impact metabolic and thermogenic function but selectively reduces Dio2 and Pparg expression in male mice

Similar to observations on a chow diet, male and female Ucp1-Cre-Scly KO mice maintained normal metabolic profiles under a Se-deficient diet. No significant differences were observed in body weight (Fig. 4A-D), BAT and liver weights (Fig. 4E-H), glucose tolerance (Fig. 4I-L), or insulin sensitivity (Fig. 4M-P) between control and Ucp1-Cre Scly KO mice. Serum levels of T3 and T4 were also unchanged in both male and female Ucp1-Cre Scly KO mice compared to controls (Fig. 5A, B, D, E). However, Dio2 mRNA levels were reduced by approximately 50% in male Ucp1-Cre Scly KO mice (Fig. 5C), while remaining unaffected in females. Despite this, male Ucp1-Cre Scly KO mice exhibited core body temperatures similar to controls during cold exposure (Fig. 5G, H; p = 0.4261), and female Ucp1-Cre Scly KO mice maintained thermogenic responses comparable to the controls, particularly during the first 60 min of cold exposure (Fig. 5I, J). Pparg mRNA levels were also significantly reduced in male Ucp1-Cre Scly KO mice but unchanged in females. Notably, UCP1 protein expression remained unaffected by BA-specific Scly deletion in either sex under Se-deficient conditions (Fig. 5L, M).

Fig. 4.

Metabolic assessment of Se-deficient mouse BAT after targeted loss of Scly. (A, C) Percentage of body weight change at indicated time points upon feeding a low-Se diet (0.08 ppm); male and female mice groups n = 4 (B, D) Final body weight (g) at 8 weeks, when euthanasia was performed; male mice groups n = 7, female mice control n = 5 and Ucp1-Cre-Scly KO mice n = 6. (E, F) Male BAT and liver weights after euthanasia; control mice n = 10, Ucp1-Cre-Scly KO mice n = 7 (G, H) Female BAT and liver weights; control mice n = 6, Ucp1-Cre-Scly KO n = 10. (I, K) GTT for male and female mice on a low-Se diet; male control mice n = 10 and Ucp1-Cre-Scly KO n = 7, female mice control n = 6 and Ucp1-Cre-Scly KO mice n = 10. (J, L) AUC for GTT of male and female mice, respectively. (M, O) ITT for male and female mice; male mice group n = 10, female mice control n = 6 and Ucp1-Cre-Scly KO mice n = 10. (N, P) AUC for ITT of male and female mice. Values are mean ± SEM *p < 0.05, **p < 0.01, and ***p < 0.001 by t-test with a 95% confidence in (A) and (C)

Fig. 5.

Ucp1-Cre Scly KO did not significantly alter thyroid hormone levels or thermogenic capacity in mice housed at thermoneutrality and fed a Se-deficient diet. (A, B, D, E) Serum triiodothyronine (T3) and thyroxine (T4) levels in male and female control and Ucp1-Cre-Scly KO mice; male control mice n = 13 and Ucp1-Cre-Scly n = 14, female control mice n = 10 and Ucp1-Cre-Scly KO n = 14. (C, F, K)) Dio2 and Pparg mRNA levels in male and female control and Ucp1-Cre-Scly KO mice on a low Se diet; male control mice n = 7 and Ucp1-Cre Scly KO mice n = 4, female mice groups n = 6. (G, I) Core body temperatures of male and female control and Ucp1-Cre Scly KO mice on a low Se diet exposed to the cold at 4℃ for 6 h; male control mice n = 3 and Ucp1-Cre-Scly KO n = 4, female mice control n = 7 and Ucp1-Cre-Scly KO mice n = 4. (H, J) AUC for body temperature of the male (p = 0.4261) and female mice. (L) Western blot of UCP1 protein in BAT of male and female control and Ucp1-Cre-Scly KO mice; male mice groups n = 4, female mice groups n = 6. (M) Quantification of UCP1 protein in BAT of male and female mice. Values are mean ± SEM *p < 0.05 in (C) and (K)

Loss of Scly in BA leads to differences in selenoprotein expression under Se-deficiency according to sex

Under the Se-deficient diet, BAT and serum Se levels of both male and female Ucp1-Cre-Scly KO mice remained unaffected (Fig. 6A, B, F, G). However, male Ucp1-Cre Scly KO showed approximately a 75% reduction in Gpx1 transcript levels and about a 50% reduction in Gpx4 and Selenop compared to controls (Fig. 6C-E), whereas female Ucp1-Cre Scly KO mice maintained expression of selenoprotein genes (Gpx1, Gpx4, and Selenop) (Fig. 6H-J). These findings suggest that BA-specific Scly deletion leads to sex-specific transcriptional alterations under Se-deficient conditions, with male mice displaying greater sensitivity.

Fig. 6.

Loss of Scly in BA alters selenoprotein mRNA levels and expression under Se deficiency according to sex. (A, F) Se content in BAT of male and female mice; male control mice n = 6 and Ucp1-Cre Scly KO mice n = 5, female mice groups n = 4. (B, G) Se content in serum of male and female mice; male control mice n = 6 and Ucp1-Cre Scly KO mice n = 4, female mice groups n = 5. (C-E, H-J) Gpx1, Gpx4, and Selenop mRNA levels in male and female control and Ucp1-Cre-Scly KO mice on a low Se diet; male control mice n = 7 and Ucp1-Cre Scly KO mice n = 4, female mice groups n = 6. (K, M) Western blot of GPX1 and GPX4 protein in BAT of male and female control and Ucp1-Cre-Scly KO mice; male mice groups n = 4, female mice groups n = 6. (L, N) Quantification of GPX1 and GPX4 protein in BAT of male and female mice. Values are mean ± SEM *p < 0.05, **p < 0.01, and ***p < 0.001 by t-test with a 95% confidence in (C) and (H)

To determine whether these transcriptional changes translated to altered protein expression, we assessed GPX1 and GPX4 selenoprotein expression. Male Ucp1-Cre Scly KO mice showed significantly reduced protein levels of both GPX1 and GPX4, while protein levels of GPX3, SEPHS2, and TXNRD1 remained unchanged compared to control mice (Fig. 6K, L). In contrast, protein expression of GPX1 and GPX4 in female Ucp1-Cre Scly KO remained unchanged compared to controls; yet these particular controls had a widespread protein quantification which can be explained by natural variability (Fig. 6M, N). Together these findings indicate that loss of Scly in BAT under Se-deficient and thermoneutral conditions leads to a sex-specific alteration in selenoprotein expression, with male Ucp1-Cre-Scly KO mice exhibiting decreased GPX1 and GPX4 at both the transcript and protein levels, while female mice maintain selenoprotein expression despite Scly deletion in BAT.

Hepatocyte-specific Deletion of Scly Does not Lead To Obesity or Alter Selenoprotein Expression

To determine whether deletion of Scly in the liver could be the driver of the obesity phenotype originally observed in the whole-body Scly KO mice, we assessed metabolic parameters in Alb-Cre-Scly KO mice. No significant differences in body, liver, and BAT weights between control and Alb-Cre-Scly KO mice in either sex was found (Fig. 7A-F). Both male and female Alb-Cre-Scly KO mice also maintained normal glucose tolerance (Fig. 7G, H). Additionally, the loss of Scly in the liver did not affect the protein levels of the selenoproteins GPX1 and GPX4 in this tissue (Fig. 7I-L). These findings suggest that the hepatocyte-specific loss of Scly does not contribute to the systemic metabolic dysfunction or impaired selenoprotein expression.

Fig. 7.

Hepatocyte-specific loss of Scly does not result in obesity or alter selenoprotein expression. (A, B)) Body weights (g) of male and female control and Alb-Cre-Scly KO mice; control male mice n = 5 and Alb-Cre-Scly KO mice n = 4, female mice n = 5 for both groups. (C, D) Male liver and BAT weights at time of euthanasia; control male mice n = 5 and Alb-Cre-Scly KO mice n = 4. (E, F) Female liver and BAT weights; control female mice n = 4 and Alb-Cre-Scly KO mice n = 3. (G, H) GTT for male and female mice; male mice groups n = 2, control female mice n = 3 and Alb-Cre-Scly KO mice n = 4. (I, K) Western blot of GPX1, and GPX4 in liver of male and female control and Alb-Cre-Scly KO mice. (J, L) Quantification of GPX1, and GPX4 protein in liver of male and female mice; control male mice n = 4, Alb-Cre-Scly KO mice n = 3, control female mice n = 3 and Alb-Cre-Scly KO mice n = 4. Values are mean ± SEM ****p < 0.0001 by t-test with a 95% confidence in (K, M).

Discussion

This study investigated the role of the Sec-decomposing enzyme Scly in BA and in hepatocytes for sustaining selenoprotein synthesis, metabolic homeostasis and, in the BAT, adaptive thermogenic response. Given the essential role of selenoproteins in BAT adaptive thermogenesis, we hypothesized that the targeted loss of Scly in BA of mice would lead to metabolic dysfunction and impair selenoprotein expression and consequently compromise BAT thermogenic capacity. However, the targeted deletion of Scly in BA did not alter systemic metabolic parameters, including body weight, BAT or liver weights, glucose tolerance, or insulin sensitivity, even under Se-deficient conditions. These findings contrast from the phenotype observed in whole-body Scly KO mice, which exhibit obesity, glucose intolerance, and hepatic steatosis [18]. The absence of a metabolic phenotype in the Ucp1-Cre Scly KO mouse model indicates that Scly loss in BA alone is not sufficient to drive systemic metabolic dysfunction. The findings, along with the Alb-Cre Scly KO model, further supports that a single tissue may not be driving this phenotype, as these mice also maintained glucose tolerance, suggesting instead that the phenotype of the whole-body Scly KO results from combined disruption of Scly across multiple tissues, rather than a single one driving the phenotype. In the case of targeted disruption of the Scly gene in specific tissues, other major metabolic regulatory organs including the liver, pancreas, skeletal muscle, and kidney [25–27] may compensate for the local Scly loss.

Under Se-deficient conditions, male Ucp1-Cre-Scly KO mice exhibited significantly reduced expression of key selenoprotein genes (Gpx1, Gpx4, Dio2, and Selenop) and proteins (GPX1, GPX4, and DIO2), and the adipogenic marker Pparg. Despite alterations in selenoprotein mRNA expression, core body temperature during cold exposure was maintained, indicating preserved thermogenic capacity. This dissociation suggests that reduced expression of key selenoproteins does not necessarily impair BAT thermogenesis function under acute thermal stress and highlights a role for Scly in regulating selenoprotein expression in BAT rather than directly influencing thermogenic function. Notably, while Dio2 expression was reduced in male Ucp1-Cre-Scly KO mice, circulating T3 and T4 remained unchanged, indicating that Scly deletion in BA does not affect T3 and T4 availability to BAT. Yet, since Dio2 mediates conversion of T4 to active T3 within BA, its downregulation may reflect altered local TH signaling [28]. Local TH activation is crucial for UCP1 expression and consequently adaptive thermogenesis, and its impairment despite normal serum TH levels could suggest reduced thermogenic capacity or function under prolonged stress [29, 30]. Despite unaffected BAT and serum Se levels, Selenop expression was significantly reduced in male Ucp1-Cre Scly KO mice. As the primary Se transport protein essential to supply Se to BAT, this downregulation suggests either impaired Se uptake or redirection of Se from selenoprotein synthesis to other metabolic pathways [31]. Selenop was prioritized due to its well-established role in maintaining Se homeostasis in BAT, unlike other selenoproteins whose roles in BAT remain less defined [15]. The concurrent downregulation of Gpx1 and Gpx4, both essential for reducing hydrogen and lipid peroxides, may compromise primary antioxidant defenses [32]. Notably, the adipocyte regulator Pparg expression was also reduced, possibly reflecting a shift in lipid handling within BA. Nevertheless, this alteration was not associated with impaired thermogenic capacity from acute cold exposure.

We further observed changes in selenoprotein expression in thermoneutral conditions (30℃), where BAT activation is minimized by reducing thermal stress and allows detection of basal metabolic reactions to maintain core body temperature [33]. Male Ucp1-Cre Scly KO mice continued to display reduced GPX1 and GPX4 expression under both chow and on a Se-deficient diet, suggesting that Scly deletion impairs selenoproteins synthesis even in the absence of thermogenic demand. These findings reinforce that the redox imbalance observed in male mice is independent of thermogenic activation and may show an impairment in Se recycling within BAT. The inefficient recycling of Se in BAT for the synthesis of selenoproteins, particularly GPX1 and GPX4 which are involved in adaptive thermogenesis, may potentially lead to increased oxidative stress within this tissue, yet is insufficient to curb heat production. In contrast, maintenance of selenoprotein expression of GPX1, GPX4, and UCP1 and preserved thermogenic ability during acute cold exposure observed in the female Ucp1-Cre Scly KO mice suggests the presence of compensatory mechanisms, potentially involving estrogen signaling or enhanced mitochondrial function that maintain redox homeostasis in females despite the loss of Scly.

Sex-specific effects were evident in our animal model, particularly in the regulation of GPX1 and GPX4 selenoproteins, and thermogenic gene and protein expressions. While male Ucp1-Cre Scly KO mice displayed significantly reduced levels of Gpx1, Gpx4, Dio2, and Pparg, female Ucp1-Cre Scly KO mice maintained transcript and protein levels. These molecular differences were accompanied by preserved thermogenic capacity in both sexes during acute cold exposure, but still suggest BAT is more vulnerable to oxidative stress involving glutathione responses under Se deficiency. Previous studies have shown that female mice possess greater BAT mass and a higher threshold for BAT activation in response to cold, contributing to their resilience against disrupted Se metabolism [32]. These differences may stem from sex-specific variation in adipose tissue distribution, BAT activity, and hormonal influence on thermogenesis [25, 34]. Female rodents store more subcutaneous fat, while males predominantly accumulate visceral fat [26]; female mice also have larger BAT depots, which are associated with slower declines in thermogenic activity [34]. Functionally, female mice exhibit more efficient adaptive thermogenesis, characterized by greater mitochondrial recruitment and higher Ucp1 levels compared to males [27, 34]. Additionally, sex hormones, particularly estrogen and estrogen-related receptors (ERRs) play a critical role in thermogenic capacity. ERRs are expressed largely in the female BAT and regulate genes involved in oxidative and thermogenic responses [35]. Male mice, by contrast, express lower levels of estrogen receptors in BAT, and studies have shown that loss of ERRs diminishes BAT thermogenic function and leads to hypothermia during cold exposure [35]. Estrogen also facilitates the BA proliferation and differentiation, including Ucp1 expression through the mediation of estrogen receptor-⍺ (ER⍺). Supporting this, the BAT-specific removal of ER⍺ in female mice leads to reduced expression of Ucp1 in BAT, lower core body temperature, and altered BAT metabolism [36]. Thus, differences in adipose tissue distribution and sex hormones likely contribute to the observed sex-specific outcomes in Ucp1-Cre Scly KO mice. This suggests that Scly plays a more pivotal role in regulating GPX1 and GPX4 selenoproteins in male BAT, while females may leverage hormone-dependent compensatory mechanisms and efficient BAT function to maintain thermogenic capacity and selenoprotein levels despite the loss of Scly. The effect was specific to GPX1 and GPX4, as protein levels for GPX3, SEPHS2, and TXNRD1 in male BAT remained unchanged.

In conclusion, our study shows that the loss of Scly in brown adipocytes or hepatocytes does not lead to dysfunction in energy metabolism but reduces expression of key antioxidant selenoproteins, GPX1 and GPX4, potentially diminishing the capacity to respond to oxidative stress via glutathione dependent pathways in BAT. The target deletion of Scly also reinforced Se -dependent sex differences in adaptive thermogenesis during cold exposure. These findings refine our understanding of the role of Se in BA and BAT adaptive thermogenesis, emphasizing Scly as a key regulator of localized selenoprotein synthesis in BAT, rather than a contributor to systemic metabolic regulation.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

L.A.S., B.K.S., and P.J.D.S. conceived the project and designed the experiments. L.A.S., B.K.S., P.J.D.S., P.T., N.A., S.M.S., G.L.R.,and K.A.H., performed mouse experiments. P.J.D.S., B.K.S., A.G.S., S.M.S., and V.Y. performed western blotting, qPCR, and TXRF. L.A.S. performed ELISA Assays. P.J.D.S., B.K.S., and L.A.S. wrote the paper that was edited and reviewed by all authors. L.A.S. and B.K.S. supervised the project.

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.