Sex-specific Effects of α2δ-1 in the Ventromedial Hypothalamus of Female Mice Controlling Glucose and Lipid Balance

Abstract

The thrombospondin receptor alpha2delta-1 (α2δ-1) plays essential roles promoting the activity of SF1 neurons in the ventromedial hypothalamus (VMH) and mediating glucose and lipid metabolism in male mice. Its role in the VMH of female mice remains to be defined, especially considering that this hypothalamic region is sexually dimorphic. We found that α2δ-1 depletion in SF1 neurons differentially affects glucose and lipid balance control and sympathetic tone in females compared to males. Mutant females show a modest increase in relative body weight gain when fed a high-fat diet (HFD) and normal energy expenditure, indicating that α2δ-1 is not a critical regulator of energy balance in females, similar to males. However, diminished α2δ-1 function in the VMH leads to enhanced glycemic control in females fed a chow diet, in contrast to the glucose intolerance reported previously in mutant males. Interestingly, the effects of α2δ-1 on glucose balance in females are influenced by diet. Accordingly, females but not males lacking α2δ-1 exhibit diminished glycemic control as well as susceptibility to hepatic steatosis when fed a HFD. Increased hepatic sympathetic tone and CD36 mRNA expression and reduced adiponectin levels underlie these diet-induced metabolic alterations in mutant females. The results indicate that α2δ-1 in VMH SF1 neurons critically regulates metabolic function through sexually dimorphic mechanisms. These findings are clinically relevant since metabolic alterations have been reported as a side effect in human patients prescribed gabapentinoid drugs, known to inhibit α2δ-1 function, for the treatment of seizure disorders, neuropathic pain, and anxiety disorders.

The ventromedial hypothalamus (VMH) plays a well-established role in appetite and body weight regulation (1-3) as well as in glucose homeostasis (4-8) and lipid metabolism (9, 10). These effects are facilitated, in part, via regulation of sympathetic output to metabolic organs in the periphery (11, 12). The underlying mechanisms remain poorly understood. We reported a previously unrecognized and essential role of alpha2delta-1 (α2δ-1) in suppressing appetite and facilitating glycemic control, acting downstream of brain-derived neurotrophic factor (BDNF) in the VMH (13). α2δ-1 is an auxiliary subunit of voltage-gated calcium channels, facilitating calcium currents and neurotransmitter release (14). Additionally, it serves as a receptor for thrombospondins to induce excitatory synapse assembly through calcium-channel independent mechanisms (15). α2δ-1 is expressed in brain regions involved in metabolic control, with notably high expression in the VMH (16, 17).

Mice with central BDNF depletion exhibit dramatic obesity, hyperglycemia, dyslipidemia, and decreased excitatory tone in the VMH as well as a significant reduction in cell-surface expression of α2δ-1 in the VMH (13). Rescuing this deficit broadly in the VMH via viral gene delivery, significantly mitigated hyperphagia and body weight gain and normalized glycemic control in BDNF mutant mice. Moreover, a more discrete and selective deletion of α2δ-1 in steroidogenic factor-1 (SF1) neurons, which in the brain are exclusive to the VMH, elicited significant reductions in neuronal activity and sympathetic output without affecting calcium currents, ultimately resulting in glucose intolerance and alterations in lipid metabolism (18). In total, these studies support a chief role of α2δ-1 as increasing VMH neuronal excitability and activity in a calcium channel-independent manner to facilitate energy and glucose balance control. However, a caveat of these previous investigations is that they were performed exclusively in males. Considering that the VMH is a sexually dimorphic region of the brain, it is important to assess whether α2δ-1 plays similar roles in females.

Here, we examined the effects of depleting α2δ-1 from SF1 neurons in female mice. We found that in contrast to males with similar deficits in α2δ-1, mutant females exhibit normal sympathetic output and lipolysis in white adipose tissue and enhanced glycemic control under chow conditions when compared to female controls. However, mutant females display glucose intolerance and increased susceptibility to hepatic steatosis when challenged with a high-fat diet (HFD) compared with controls. The collective findings indicate that the effects of α2δ-1 in the VMH influencing metabolic function are sexually dimorphic. These findings are clinically relevant since metabolic alterations have been reported as a side effect in human patients prescribed gabapentinoid drugs, known to inhibit α2δ-1 function (19), for the treatment of seizure disorders, neuropathic pain, and anxiety disorders (20, 21).

Materials and Methods

Animals

All procedures were approved by the Institutional Animal Care and Use Committee at Tufts University and conducted in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals guidelines. Mice with specific α2δ-1 depletion in SF1-positive neurons of the VMH were generated by breeding homozygous floxed α2δ-1 mice (18) with SF1-Cre transgenic mice obtained from the Jackson Laboratory (Stock No: 012462; Tg(Nr5a1-cre)7Lowl/J). Experiments described herein used age-matched littermate mutant (α2δ-1^{2L/2L:SF1-Cre}) and control (α2δ-1^2L/2L) mice to reduce genetic background differences. Females from the ages of 15 to 20 weeks were used for all of the experiments. They were housed in proximity to males to promote regular estrous cyclicity and they had a 14-hour light/10-hour dark cycle (lights on from 6:00 am to 8:00 pm) and at a temperature of 72^oF.

α2δ-1 immunofluorescence

Mice were deeply anesthetized and transcardially perfused with 10 mL of cold saline followed by 30 mL of 4% paraformaldehyde (PFA), pH 7.2. Brains were immediately removed, post-fixed in 4% PFA for 4 hours at 4^oC, cryoprotected in a 30% sucrose solution and frozen in mounting media until 30 μm-thick coronal sections were obtained using a Leica CM1900 Cryostat. Free-floating brain sections were blocked using 10% normal goat serum (NGS, 005-000-121, Jackson ImmunoResearch Laboratories, Inc.) (22) and 0.25% Triton X-100 for 90 minutes at room temperature. Brain sections were incubated overnight at 4°C with mouse anti-α2δ-1 (D219, Sigma, 1:200) (23) followed by incubations with fluorescence-conjugated secondary antibodies at room temperature for 1 hour. A Nikon A1R microscope configured for confocal microscopy was used to obtain images of labeled brain sections. Scanning parameters for the laser power, gain, and offset were determined using the control group and then applied to all images throughout the experiment.

Food intake and body weight measurements

Mice were individually housed with unrestricted access to water and a premeasured amount of food. Body weight and food intake measurements were monitored beginning at 6 weeks of age and measured weekly at the same time of day. The amount of grams of food consumed each week was then used to calculate caloric intake (kcal). Animals were fed either a chow diet (CD; Harlan Teklad 2918; 3.1 kcal/g with 18% from fat) or, at 8 weeks of age, switched to a 58% high-fat diet with sucrose (HFD; Research Diets D12331; 5.56 kcal/g) for up to 18 weeks.

Energy expenditure

Energy expenditure was assessed by indirect calorimetry using OxyletPro Physiocage metabolic chambers, a LE405 Gas Analyzer to sample air from individual cages, and Metabolism v2.2 software to integrate gas samples and compute energy expenditure (Panlab/Harvard Apparatus). Mice were housed in individual metabolic assessment chambers containing normal bedding and unrestricted access to food and water. The flow rate of fresh air into each cage was set according to individual animal’s body weight (e.g., a 34 g mouse would set a flow rate of 0.34 L/min). Each cage was sampled every 15 minutes for 3-minute increments. Data were discarded for the first day that animals were in the chamber. Measurements for the remaining 3 test days were analyzed and the data, averaged across days, are reported as average oxygen uptake and carbon dioxide production.

Locomotor activity

Mice were individually housed in their home cages (standard 15 × 24 cm plastic cages) and transported to the testing room (14-hour light/dark cycle) for 6 days of acclimation to the testing room. The home cage was situated such that the Smart Frame Activity System photobeam frame surrounded it. During testing, locomotor activity, as measured by beam breaks, was recorded continuously using MotorMontitor software (Hamilton/Kinder). Data collected from the first day were discarded. The remaining 5 days of data were binned in hourly increments and averaged across test days.

Glucose tolerance tests

For the glucose tolerance test (GTT), animals were fasted overnight for 16 hours and then, using a scalpel, a small cut was made in the distal portion of the tail to enable the collection of a fasting baseline (0) blood glucose measurement. Following this baseline measurement, 1.5 g/kg D-glucose was administered intraperitoneally (i.p.). Subsequent blood samples were collected at 15, 30, 60, and 120 minutes post–glucose injection. The GTT was performed at 20 weeks of age. Two measurements were collected at each time-point and then averaged. Data for fasting blood glucose measurements, blood glucose time-course over 120 minutes, and area under the curve (AUC) are presented.

Tissue collection and measurements

Mice were anesthetized using isofluorane and following cervical dislocation, tissues were dissected, flash frozen in liquid nitrogen, and stored at −80°C until further analysis. Tissues collected included liver, perigonadal white adipose tissue (WAT), brown adipose tissue (BAT), skeletal muscle (SM, soleus and gastrocnemius muscle), brain and serum. All tissue collections were performed from 8 to 11 am. Serum insulin (Rat/Mouse Insulin ELISA kit, Millipore Sigma) (24), leptin (Mouse Leptin ELISA kit, Millipore Sigma) (25), adiponectin (Mouse Adiponectin ELISA kit, Millipore Sigma) (26), glycerol (Glycerol Assay kit, Millipore Sigma), and NEFA (Free Fatty Acid quantification kit, Millipore Sigma) levels were measured using commercially available kits and used according to the manufacturer’s instructions. Triglyceride and total cholesterol concentrations in liver, WAT, and serum samples were determined by the Vanderbilt Hormone Assay and Analytical Services Core (Vanderbilt University, Nashville, TN). Norepinephrine content in liver, WAT, BAT, SM, and serum samples was analyzed using high-performance liquid chromatography (HPLC) methods by the Neurochemistry Core Laboratory at the Vanderbilt Brain Institute (Vanderbilt University School of Medicine, Nashville, TN).

Quantitative RT-PCR analysis

Total RNA was extracted from liver tissue using the RNeasy lipid tissue mini kit (Qiagen, Valencia, CA) kit. Following cDNA synthesis, real-time PCR amplification was performed using a MX-3000P Stratagene cycler and 2X SYBR green PCR master mix (Qiagen, Valencia, CA). A two-step protocol was used: 95°C for 10 minutes and 45 cycles with 95°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds. The following primers were used for the analysis: DGAT1, TCCGCCTCTGGGCATTC (forward) and GAATCGGCCCACAATCCA (reverse); DGAT2, AGTGGCAATGCTATCATCATCGT (forward) and AAGGAATAAGTGGGAACCAGATCA (reverse); CD36, CCGAGGACCACACTGTGTC (forward) and AACCCCA CAAGAGTTCTTTCAAA (reverse); SCD1, TTCTTGCGATA CACTCTGGTGC (forward) and CGGGATTGAATGTTC TTGTCGT (reverse); TNFα, CTCCAGGCGGTGCCTATG (forward) and GGGCCATAGAACTGATGAGAGG (reverse); SREBP‐1c, GCAGCCACCATCTAGCCTG (forward) and CAGCA GTGAGTCTGCCTTGAT (reverse); Fas, GGAGGTGGTG ATAGCCGGTAT (forward) and TGGGTAATCCATAGA GCCCAG (reverse); PPARγ, CACAAGAGCTGACCCAATGGT (forward) and GATCGCACTTTGGTATTCTTGGA (reverse).

Western blot analysis

Protein lysates were generated by homogenization of tissues with a polytron homogenizer in T-Per protein extraction reagent (Thermo Fisher) containing protease and phosphatase inhibitors. Western blot analysis of samples was conducted using standard methods. Blot images were acquired using a Fuji film LAS-4000 Image reader and densitometry performed using ImageStudioLite. The following primary antibodies were used: mouse anti-α2δ-1 (1:2000, Sigma, St. Louis, MO) (23), mouse anti-β tubulin (1:10,000 Sigma, St. Louis, MO) (27), rabbit anti-HSL (28), anti-pHSL_S660 (29) or anti-pHSL_S565 (30)(1:1000; Cell Signal Technology, Danvers, MA) and mouse anti-β-actin (1:10,000; Thermo Fisher) (31).

Statistical analysis

Analytical software used to perform the statistical analysis included GraphPad Prism and SAS (Statistical Analysis Software). Repeated measures ANOVA, using Bonferroni’s method to adjust for multiple comparisons, was performed to analyze weekly food intake and body weight measurements as well as the time course data for the glucose tolerance tests. Two-way ANOVAs and Bonferroni’s multiple comparisons tests were used to analyze norepinephrine, triglycerides, total cholesterol, serum leptin, serum adiponectin, serum glycerol, serum nonesterified free fatty acids, hepatic transcript, and hormone-sensitive lipase (HSL) concentrations. Student unpaired t tests were performed to analyze group comparisons for locomotor activity, energy expenditure, AUC, and serum insulin. Comparisons were determined by be statistically significant when P < 0.05. All values are depicted as mean ± standard error of the mean (SEM).

Results

Α2 δ -1 in the VMH of female mice is not a critical regulator of energy balance

We showed previously that α2 δ -1 depletion in VMH SF1 neurons of male mice does not have an effect on feeding, energy expenditure or body weight when fed a chow (CD) or high-fat diet (HFD), negating a required role in the regulation of energy balance (18). Because the VMH is sexually dimorphic (32), we investigated whether α2 δ -1 is also dispensable in mutant females. For this, we crossed floxed α2 δ -1 mice with SF1-cre mice to generate α2 δ -1^{2L/2L:SF1-cre} mutant and α2 δ -1^2L/2L control females. Figs. 1A and 1B show that α2 δ -1 is depleted in the VMH of α2 δ -1^{2L/2L:SF1-cre} as previously reported for mutant males (18). Because SF1 is also expressed in the adrenal glands, cre-mediated recombination is expected to occur in this tissue. However, as we reported previously, α2δ-1 is not expressed in wild-type adrenal glands (18) and thus, its depletion in this region is not a confounding factor.

Figure 1.

α2δ-1 deletion in SF1 neurons in female mice does not affect energy balance regulation. (A) Representative brain sections from female α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L mice immunolabeled with anti-α2δ-1. Scale bar = 50 µM. (B) Western blot analysis of α2δ-1 protein content in the VMH of female α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L mice (n = 3–4). *P = 0.005 unpaired t test. Body weights (C) and food intake (D) of female α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L mice fed a chow diet (n = 9-11). Body weights: ANOVA: Genotype, F (1,16) = 2.9, P = 0.1; Time, F (12, 192) = 31.8, P < 0.0001; * P < 0.05. Body weights (E) and food intake (F) of female α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L mice fed HFD (n = 9-11). Body weights ANOVA: Genotype, F (1, 18) = 1.6, P = 0.2; Time, F (4.6, 82) =108.7, P < 0.0001; * P < 0.05; Food intake: Genotype and time, F (12, 175) = 2.1, P = 0.02. (G) Body composition in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L females fed a CD. (H) Body composition in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L females fed a HFD. (I) Locomotor activity of female α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L mice (n = 4-9) fed a chow diet. (J) Energy expenditure in females expressed as average VO2 (left) and VCO2 (right) (n = 3-6).

ANOVA analysis did not show a significant effect of genotype on body weight in mutant females fed CD (Fig. 1C). Moreover, no significant changes in CD intake were observed (Fig. 1D). Total body weights in α2 δ -1^{2L/2L:SF1-cre} females compared with α2 δ -1^2L/2L controls were not significantly different when challenged with a HFD starting at 8 weeks of age (Fig. 1E). When we calculated percentage body weight gain following transition from a CD to a HFD, we found that there was a significant effect of genotype (P = 0.003) and a significant interaction of genotype and time (P < 0.0001); however, this effect was modest, with mutants exhibiting a 38% weight gain compared with 28% in controls (data not shown). Measurements of HFD intake revealed a significant interaction of genotype and time (P = 0.02) (Fig. 1F), with mutants exhibiting a mild increase in food intake. Finally, measurements of lean and fat mass in females fed a CD (Fig. 1G) or a HFD (Fig. 1H) did not reveal any significant differences in mutants compared with controls.

Locomotor activity under CD conditions was normal in mutant females (Fig. 1I). Furthermore, energy expenditure was not affected in α2δ-1^{2L/2L:SF1-Cre} mutants fed a CD or HFD (Fig. 1J). In aggregate, the findings suggest that α2δ-1 expression in SF1 neurons is not required for the regulation of energy expenditure in female mice. These findings indicate that VMH α2δ-1 does not play a critical role in energy balance regulation in females.

α2δ-1^{2L/2L:SF1-cre} mutant females exhibit enhanced and diminished glycemic control under chow and HFD conditions, respectively

The VMH plays requisite roles in the regulation of glucose balance (4-8). We asked whether α2δ-1 in SF1 neurons is essential for those effects in female mice. α2δ-1^{2L/2L:SF1-Cre} females fed a CD had a significant increase in glucose tolerance compared with α2δ-1^2L/2L controls fed a similar diet (Fig. 2A and 2B). Accordingly, there was a significant effect of genotype (P = 0.0006) in the GTT (2A) and a trend (P = 0.09) towards a significant reduction in AUC in mutants compared with controls (Fig. 2B). Consistent with their enhanced glycemic control, α2δ-1^{2L/2L:SF1-Cre} females fed a CD showed a significant (P = 0.04) reduction in fasted levels of circulating insulin (Fig. 2C), suggesting increased insulin sensitivity. These findings in females are opposite to our previous observations in α2δ-1^{2L/2L:SF1-Cre} mutant males fed a similar diet, which display glucose intolerance (18) (Table 1).

Figure 2.

α2δ-1^{2L/2L:SF1-cre} mutant females exhibit enhanced glycemic control under chow conditions and increased glucose intolerance following chronic administration of a HFD. (A) Time course for glucose tolerance test (GTT) in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L females on CD (n = 8-10 per group). ANOVA: Genotype, F (1, 15) = 6, P = 0.03 (genotype); Time, F (2.0, 29.9) = 41.9 and P < 0.0001. * P = 0.05; **, P = 0.01. (B) Area under the curve (AUC) for GTT under CD conditions (n = 8-10). Unpaired, two-tailed t test: P = 0.09. (C) Fasting serum insulin levels in α2δ-1^2L/2L and α2δ-1^{2L/2L:SF1-Cre} females (n = 8-9) on CD. Unpaired, two-tailed t test: * P = 0.04. (D) Time course for GTT in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L females on HFD (n = 9-11 per group). ANOVA: Genotype, F (1, 90) = 5.5, P = 0.02; Time, F (4, 90), P < 0.001 * P = 0.05. (E) AUC for GTT under HFD conditions (n = 9-11 per group). (F) Fasting serum insulin levels in α2δ-1^2L/2L and α2δ-1^{2L/2L:SF1-Cre} females (n = 7-11) on HFD. (G) Time course for GTT in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L males on HFD (n = 10-11 per group). (H) AUC for GTT under HFD conditions in males (n = 10-11 per group).

Table 1.

Effects of Depleting α2δ-1 in the VMH of Female and Male Mice: Metabolic Parameters in Female and Male α2δ-1^{2L/2L:SF1-Cre} Mice Compared With Their Sex- and Diet-Matched Controls

Metabolic Parameter	α2δ-1-12L/2L:SF1-cre Females	α2δ-1-12L/2L:SF1-cre Males
Food Intake (CD)	Normal	Normal*
Food Intake (HFD)	Mild increase	Normal*
Body Weight (CD)	Normal	Normal*
Body Weight (HFD)	10% increase in relative weight gain	Normal*
Energy Expenditure	Normal	Normal*
Fat and Lean Mass	Normal	Normal*
Glucose Tolerance (CD)	Improved	Reduced*
Glucose Tolerance (HFD)	Reduced	Normal
Fasted Insulin Levels (CD)	Reduced (44% of control levels)	Normal*
Fasted Insulin Levels (HFD)	Normal	Normal*
Liver TG (HFD)	Increased (120% of control levels)	Normal
Lipolysis in WAT	Normal	Elevated*
Serum Glycerol levels	Normal	Elevated (315% of control levels) *
Activated HSL levels in WAT	Normal	Elevated *
Serum NE Content (CD)	Normal	Reduced (27% of control levels)*
WAT NE Content (CD)	Normal	Reduced (55% of control levels)*
Liver NE Content (HFD)	Increased (450% of control levels)	Normal

*Previously reported in (18)

We tested the effects of dietary challenge on glucose homeostasis in α2δ-1^{2L/2L:SF1-Cre} females. In contrast to our findings in females fed a CD, female mutants subjected to a HFD displayed mild glucose intolerance compared with α2δ-1^2L/2L females fed a similar diet (Fig. 2D and 2E). Accordingly, there was a main effect of genotype (P = 0.02) in the GTT (Fig. 2D). Furthermore, female mutants fed a HFD had similar circulating levels of insulin to those of diet-matched controls (Fig. 2F). Examination of α2δ-1^{2L/2L:SF1-Cre} and control males fed a HFD did not reveal significant differences in the GTT or AUC (Fig. 2G and 2H) (Table 1). In total, the data indicate that depletion of α2δ-1 in SF1 neurons produces sex-specific effects on glycemic control. Moreover, they suggest that perturbing VMH α2δ-1 function has a bidirectional effect on glucose balance influenced by diet in females but not in males.

Depletion of α2δ-1 in SF1 neurons in females increases susceptibility to diet-induced hepatic steatosis

Lipid metabolism is greatly influenced by neuronal activity in the VMH (9, 10). We asked whether α2δ-1 depletion in this hypothalamic region affected lipid balance control in female mice. For this, we measured levels of triglycerides (TG) and cholesterol in serum, WAT, and liver of fed α2δ-1^{2L/2L:SF1-Cre} mutants and α2δ-1^2L/2L controls. There was a significant effect of diet on TG levels in serum (P = 0.003), WAT (P = 0.007), and liver (P = 0.03) (Fig. 3A, 3B and 3C). Notably, there was a significant effect of genotype (P = 0.05) and a trend (P = 0.07) towards a significant interaction of diet and genotype on TG content in the liver. Indeed, whereas hepatic content of TG significantly increased as a consequence of HFD administration in mutant females, it did not in controls. Accordingly, α2δ-1^{2L/2L:SF1-Cre} females fed a HFD exhibited a significant increase in TG levels in the liver compared with their α2δ-1^2L/2L counterparts (Fig. 3C). We reported previously that TG content in serum, WAT, and liver of α2δ-1^{2L/2L:SF1-Cre} males fed a CD was normal (18). Here, we investigated whether the effects of dietary challenge observed in mutant females were also present in mutant males. α2δ-1^{2L/2L:SF1-Cre} males fed a HFD contained normal and decreased levels of TG in serum and WAT, respectively (Fig. 3D and E). However, TG content in liver was not significantly different in α2δ-1^{2L/2L:SF1-Cre} males compared with α2δ-1^2L/2L controls (Fig. 3F) (Table 1). The results indicate that α2δ-1 depletion in SF1 neurons results in susceptibility to diet-induced hepatic steatosis in females.

Figure 3.

Depletion of α2δ-1 in SF1 neurons in females elicits deficits in lipid balance control. Levels of triglycerides in serum (A), perigonadal white adipose tissue (WAT) (B) and liver (C) of α2δ-1^2L/2L and α2δ-1^{2L/2L:SF1-Cre} (n = 4-5 per group) females. ANOVA, Serum: Diet, F (1,12) = 13.8, P = 0.003; WAT: Diet, F (1,12) = 10.7, P = 0.007; Liver: Diet, F (1,12) = 13.7, P = 0.03 (diet) and Genotype, F (1,12) = 4.7, P = 0.05. Levels of triglycerides in serum (D), perigonadal WAT (E) and liver (F) of α2δ-1^2L/2L and α2δ-1^{2L/2L:SF1-Cre} males (n = 8-9 per group) fed a HFD. Levels of cholesterol in serum (G), WAT (H) and liver (I) of α2δ-1^2L/2L females (n = 4-5 per group). ANOVA, Serum: Genotype, F (1, 12) = 6.9, P = 0.02; WAT: Genotype, F (1, 12) = 5.6, P = 0.04 (genotype) and genotype x diet, F (1,12) = 4.9, P = 0.05. * P < 0.05.

Levels of cholesterol were also differentially affected by diet in α2δ-1^{2L/2L:SF1-Cre} females compared with α2δ-1^2L/2L controls. There was a significant effect of genotype (P = 0.02) on circulating levels of cholesterol (Fig. 3G). In WAT, there was a significant effect of genotype (P = 0.04) and a significant interaction of genotype and diet (P = 0.05) on cholesterol levels. Accordingly, WAT cholesterol levels were significantly higher in control HFD females compared with HFD mutants (Fig. 3H). There were no significant differences in cholesterol content in liver under either diet (Fig. 3I). Mutant males fed a CD exhibited normal and decreased levels of cholesterol in serum and WAT, respectively, as reported previously (18). However, cholesterol content in WAT, serum and liver were normal following administration of a HFD (data not shown).

Next, we interrogated mechanisms driving lipid accumulation in the livers of α2δ-1^{2L/2L:SF1-Cre} females fed a HFD. For this, we measured hepatic transcript content of factors involved in lipogenesis and TG synthesis, including Fas, SREBP-1c, DGAT1, DGAT2, CD36, SCD1, PPARγ and TNFα in control and mutant females fed CD or HFD. Quantitative RT-PCR analysis revealed that expression of Fas, SREBP-1c, DGAT1, DGAT2, SCD1, PPARγ, and TNFα was comparable in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L females fed similar diets. In contrast, whereas levels of the fatty acid translocase CD36 were similar in controls and mutants fed a CD, they were significantly elevated in mutants fed a HFD compared with controls fed a similar diet (Fig. 4). This finding suggests that CD36-mediated increases in hepatic fatty acid uptake contribute to susceptibility to diet-induced liver steatosis in α2δ-1^{2L/2L:SF1-Cre} females.

Figure 4.

Increased expression of CD36 mRNA in the liver of α2δ-1 ^{2L/2L:SF1-Cre} females fed a HFD. Quantitative RT-PCR analysis of transcript levels in liver of factors involved in lipogenesis and TG synthesis in α2δ-1^2L/2L and α2δ-1^{2L/2L:SF1-Cre} females (n = 6 per group). CD36, ANOVA: Genotype: F (1, 17) = 4.4, P = 0.05; Diet, F (1, 17) = 31.1, P < 0.001; Genotype × Diet: F (1, 17) = 9.2, P = 0.007. * P = 0.003.

Normal lipolysis but diminished adiponectin secretion by WAT in α2δ-1^{2L/2L:SF1-Cre} females fed a HFD

During lipolysis, triglycerides are hydrolyzed and free fatty acids and glycerol are released from WAT into the circulation. This process is augmented during conditions of negative energy balance to provide the animal with substrates for oxidative metabolism and ATP production in peripheral tissues. Disinhibited lipolysis can result in increased hepatic lipid accumulation and reduced insulin sensitivity (33, 34). We reported previously that α2δ-1^{2L/2L:SF1-Cre} mutant males fed a CD exhibited an elevation in lipolysis as indicated by increased levels of the activated form of the lipolytic enzyme, hormone sensitive lipase (pHSL_S660) in WAT and a 315% elevation in circulating levels of glycerol (18). We investigated whether this was also the case in female mutants, contributing to the metabolic alterations that they exhibit. Measurements of total protein levels of HSL and its activated (pHSL₆₆₀) and inhibited forms (pHSL₅₆₅) in WAT were performed in the fed state, similar to measurements of TG content in liver and WAT. We found that levels of HSL, pHSL₆₆₀, and pHSL₅₆₅ were similar in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L females administered similar diets (Fig. 5A, 5B and 5C). Consistent with these findings, whereas there was a significant effect of diet on circulating levels of glycerol and free fatty acids (FFA), there was no significant effect of genotype, indicating that lipolysis in WAT was not affected in mutant females (Fig. 5D and 5E) (Table 1).

Figure 5.

Normal lipolysis but diminished adiponectin secretion by WAT in α2δ-1 ^{2L/2L:SF1-Cre} females fed a HFD. Representative blot (A) and quantification of Western blot analysis of HSL (B), pHSL660 (C) and pHSL565 (D) protein content in WAT samples obtained from fed α2δ-1^2L/2L (C) and α2δ-1^{2L/2L:SF1-Cre} females (M) fed CD or HFD (n = 6). Serum levels of FFA (E), Glycerol (F) and adiponectin (G) in α2δ-1^2L/2L and α2δ-1^{2L/2L:SF1-Cre} females fed CD or HFD (n = 6). ANOVA, Adiponectin: Diet, F (1, 19) = 8.3, P = 0.01 and Genotype, F (1, 19) = 4.0, P = 0.05. * P = 0.008.

Next, we asked whether adiponectin levels were altered in α2δ-1^{2L/2L:SF1-Cre} females, as this adipokine counters excess lipid storage in the liver (35, 36). Indeed, we found that whereas levels of adiponectin were similar in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L females fed a CD, they were significantly reduced in mutant females fed HFD compared with diet-matched controls (Fig. 5F). Serum levels of leptin were also measured and found to be comparable in α2δ-1^{2L/2L:SF1-Cre} and α2δ-1^2L/2L females fed a CD or HFD and only influenced by diet (data not shown).

In total, the results indicate that, unlike male α2δ-1^{2L/2L:SF1-Cre} mice, mutant females exhibit normal levels of lipolysis. Furthermore, they suggest that diminished circulating levels of adiponectin in α2δ-1^{2L/2L:SF1-Cre} females could underlie their susceptibility to diet-induced hepatic steatosis.

Hepatic sympathetic tone in α2δ-1^{2L/2L:SF1-Cre} females is significantly elevated by HFD administration.

The VMH regulates peripheral glucose and lipid metabolism through the control of sympathetic output to metabolic organs (5, 10, 37-39). We found previously that sympathetic tone was significantly reduced in α2δ-1^{2L/2L:SF1-Cre} males fed a CD as indicated by 45% and 63% reductions in norepinephrine (NE) content in WAT and serum, respectively (18). Because SF1 neurons critically regulate sympathetic tone in the periphery, we asked whether alterations in this branch of the autonomic system in α2δ-1^{2L/2L:SF1-Cre} females might contribute to the metabolic alterations that they exhibit. There was a significant effect of diet on NE content in WAT. However, levels of NE in serum, WAT, BAT, and smooth muscle were normal in mutant females compared with diet-matched controls (Fig. 6A-6D). In contrast, levels of NE in liver were significantly and dramatically elevated in α2δ-1^{2L/2L:SF1-Cre} females compared with α2δ-1^2L/2L controls fed a HFD (Fig. 6E). This effect of HFD on hepatic sympathetic outflow was not observed in mutant males (Fig. 6F) (Table 1), which also contained normal levels of NE in serum, WAT, BAT, and smooth muscle (data not shown). The data indicate that α2δ-1 depletion in the VMH exerts sexually dimorphic effects on sympathetic tone, diminishing sympathetic function in males but not in females fed a CD and increasing hepatic sympathetic outflow in females but not in males fed a HFD. Furthermore, considering previous findings showing that increased sympathetic tone onto the liver elicits hepatic steatosis (40), the increase in hepatic sympathetic outflow in α2δ-1^{2L/2L:SF1-Cre} females fed a HFD likely contributes to their susceptibility to TG accumulation.

Figure 6.

Increased hepatic sympathetic output in α2δ-1 ^{2L/2L:SF1-Cre} females fed a HFD. Norepinephrine content in serum (A), white adipose tissue (WAT) (B), brown adipose tissue (BAT) (C), skeletal muscle (SM) (D) and liver (E) of fed α2δ-1^2L/2L and α2δ-1^{2L/2L:SF1-Cre} (n = 7-8) females fed a chow diet (CD) or high-fat diet (HFD). ANOVA, WAT: Diet, F (1, 28) = 9.6, P = 0.004 (diet); Liver: Genotype, F (1, 26) = 5.7, P = 0.02. * P = 0.03. (F) Norepinephrine content in liver of α2δ-1^2L/2L and α2δ-1^{2L/2L:SF1-Cre} males fed a HFD (n = 7-8).

Discussion

Here, we show that effects of α2δ-1 in the VMH influencing glucose and lipid balance and sympathetic outflow are sexually dimorphic. In females, α2δ-1 depletion results in improved glycemic control under CD conditions and in mild body weight gain, glucose intolerance, elevated sympathetic tone, and TG accumulation in the liver when fed a HFD. These bidirectional effects influenced by diet, suggest an important role of α2δ-1 coordinating adaptive mechanisms during challenging dietary conditions. They are also clinically relevant considering that alterations in metabolic function have been reported in individuals treated with gabapentinoid drugs, which inhibit α2δ-1 (19-21).

Depleting α2δ-1 in SF1 neurons had very mild effects on body weight of females and only when challenged with a HFD, indicating that α2δ-1 in those cells does not play a critical role in energy balance regulation, similar to males. However, sex-specific effects of α2δ-1 in the VMH were evident in mechanisms controlling glucose homeostasis. We reported previously that α2δ-1^{2L/2L:SF1-Cre} males fed a CD are glucose intolerant relative to littermate controls, indicating an essential and body weight-independent role of α2δ-1 facilitating glycemic control. In contrast, α2δ-1^{2L/2L:SF1-Cre} females fed a CD exhibited a significant increase in glucose tolerance and reduced fasted levels of serum insulin, suggesting enhanced insulin sensitivity. Diet influenced the consequences of depleting α2δ-1 in the female VMH. Accordingly, mutant females fed a HFD exhibited a mild reduction in glycemic control compared with α2δ-1^2L/2L females fed a similar diet. This effect was absent in mutant males fed a HFD, which performed similarly to control males in the GTT. Because there were no significant effects of genotype on total body weights of α2δ-1^2L/2L females fed a CD or a HFD, it is unlikely that observed alterations in glucose balance are driven by changes in body weight. Because SF1 is also expressed in the pituitary and we used the SF1-cre line of mice, it is also important to consider that altered α2δ-1 function in this region might contribute to the observed metabolic alterations. In total, the data suggest that α2δ-1 facilitates adaptive cellular mechanisms mediating metabolic flexibility in female mice.

Lipid balance control is also differentially influenced by VMH α2δ-1 in females versus males. Accordingly, α2δ-1^{2L/2L:SF1-Cre} females fed a CD did not exhibit the elevation in lipolysis in WAT observed previously in α2δ-1^{2L/2L:SF1-Cre} males administered the same diet (18). However, mutant females but not males had increased susceptibility to hepatic steatosis when fed a HFD. Decreased cholesterol content in WAT also suggests that lipid balance during dietary challenge is influenced by VMH α2δ-1 in female mice. This is a significant finding considering that dyslipidemia often precedes alterations in glycemic control, which are also exhibited by α2δ-1^{2L/2L:SF1-Cre} females fed a HFD. For example, previous studies showed that abnormal cholesterol levels greatly impact adipocyte metabolic activity, which can ultimately have deleterious effects on insulin sensitivity (41). Furthermore, we found that circulating levels of adiponectin were significantly reduced in α2δ-1^{2L/2L:SF1-Cre} females fed a HFD, suggesting impaired WAT function. Because adiponectin was reported to facilitate metabolic flexibility in mice, this deficit likely contributes to metabolic dysfunction in mutant females fed a HFD. Specifically, HFD-induced hepatic fat accumulation and reductions in insulin sensitivity were prevented in transgenic mice with elevated levels of circulating adiponectin (35, 36).

Identified elevations in hepatic sympathetic outflow are expected to contribute to the susceptibility to diet-induced liver steatosis observed in α2δ-1^{2L/2L:SF1-Cre} females. In support, Hurr and Morgan reported recently that diet-induced liver steatosis was associated with an increase in the firing rate of hepatic sympathetic fibers in mice and increased transcript content of the fatty acid translocase CD36 in the liver (40). Importantly, hepatic sympathetic denervation mitigated free fatty acid uptake and TG accumulation in livers of mice fed a HFD and this was mediated, in part, by blunting diet-induced elevations in CD36 mRNA expression in the liver. This is particularly relevant as a great proportion of hepatic TG is derived from FFA uptake from the circulation by cell membrane transporters (42) and as α2δ-1^{2L/2L:SF1-Cre} females fed a HFD also exhibited elevated expression of hepatic CD36. The collective evidence suggests that α2δ-1 in the VMH of female mice is a critical component of central mechanisms regulating hepatic sympathetic outflow and that in its absence, hepatic sympathetic overactivity leads to increased CD36 expression and TG accumulation in the liver. It is also important to note that unlike mutant females, sympathetic tone to the livers of α2δ-1^{2L/2L:SF1-Cre} males fed a HFD was similar to that of male controls. Sex-specific effects on sympathetic tone following depletion of VMH α2δ-1 were also evident in animals fed a CD. Whereas NE content in serum and WAT was dramatically reduced in α2δ-1^{2L/2L:SF1-Cre} males (18), it was normal in females as demonstrated here.

Altered SF1 neuronal activity is a plausible central mechanism driving metabolic alterations in female mutant mice. In support, activation of SF1 neurons via DREADD technology increases insulin sensitivity and glucose uptake in the periphery (5). α2δ-1 serves both as a voltage-gated calcium channel subunit and as a receptor for thrombospondins (TSPs), facilitating excitatory synapse assembly (14, 15). Eroglu et al (15) showed that cortical neurons in transgenic mice overexpressing α2δ-1 contained a higher density of excitatory synapses and elevated frequency of miniature EPSCs compared with wild-type mice (15). In contrast, inhibiting α2δ-1 via administration of gabapentin blunted the effects of thrombospondins inducing excitatory synaptogenesis. Finally, we showed that depleting α2δ-1 reduces the excitatory tone and activity of SF1 neurons without affecting calcium currents in male mice (18), implicating actions of α2δ-1 facilitating excitatory synapse assembly.

It remains unclear why α2δ-1 depletion results in sex-specific effects on glucose and lipid balance and sympathetic output in the periphery. One possibility is that α2δ-1 couples to different and opposing signaling cascades in females versus males, differentially affecting neuronal activity. However, considering that there are no examples in the literature indicating that α2δ-1 inhibits excitatory transmission and that there is evidence that it does not affect inhibitory transmission (15), this scenario is unlikely. Alternatively, there might be sex differences in the pattern of α2δ-1 expression or in the requirement of this thrombospondin receptor within distinct SF1⁺ neuronal subpopulations. This is important considering that diverse leptin-induced, leptin-inhibited and insulin-inhibited SF1 neuronal subpopulations have been identified in the VMH, indicating functional heterogeneity within this cell population (43). Future investigations outside the scope of the current study should examine this possibility. Sex-specific effects might also arise from α2δ-1 influencing effects of estrogen on VMH neuronal activity and energy and glucose balance control in females but not in males. This possibility is worth considering as there is a higher density of estrogen receptors in female versus male VMH (44) and that estrogen signaling via ERα in the VMH is protective against obesity and glucose intolerance in females but not in males (45). Finally, potential interactions between sex and genetic background should also be considered. Indeed, mice studied in our investigations were in a mixed C57Bl6 and FVB/NJ background, and females in both of those backgrounds exhibit resistance to diet-induced obesity and diet-induced alterations in glycemic control compared with males (46, 47).

In summary, our studies revealed female-specific and bimodal effects of VMH α2δ-1 impacting glucose and lipid balance control and sympathetic tone. The collective findings illustrate the importance of assessing sex-specific mechanisms regulating metabolic function and inform pathological mechanisms underlying metabolic disturbances in individuals administered the anti-epileptic and antinociceptive drugs gabapentin and pregabalin, which bind and inhibit α2δ-1 (20, 21).

Glossary

Abbreviations

AUC

area under the curve

BAT

brown adipose tissue

BDNF

brain-derived neurotrophic factor

CD

chow diet

GTT

glucose tolerance test

HFD

high fat diet

HSL

hormone-sensitive lipase

SF1

steroidogenic factor-1

SM

skeletal muscle

TG

triglyceride

VMH

ventromedial hypothalamus

WAT

white adipose tissue

Additional Information

Disclosure Summary: The authors declare no conflicting financial interest.

Data Availability: Data sets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.