Role of the ventromedial hypothalamic Steroidogenic Factor 1/ Adrenal 4
Binding Protein neurons in the regulation of whole body energy and
glucose metabolism in mice.
Eulalia Coutinho
Department of Physiological Sciences,
School of Life Science,
SOKENDAI (The Graduate University for Advanced Studies)
September 2015
INTRODUCTION:
The hypothalamus is the brain center which communicates with the periphery to control autonomic pathways which regulate energy intake, expenditure and storage. It is made up of different nuclei: arcuate hypothalamus (ARC), dorsomedial hypothalamus (DMH), lateral hypothalamus (LH), paraventricular hypothalamus (PVH) and the ventromedial hypothalamus (VMH), etc. The VMH was the first hypothalamic region found to play a crucial role in body weight regulation and energy homeostasis. The VMH came to be known as the 'satiety center' because studies showed that VMH lesions resulted in hyperphagia and obesity. Electrical stimulation of the VMH suppressed feeding behavior and decreased food intake in rats. Researchers also found that electrical or chemical stimulation of the VMH increased glucose uptake in peripheral tissues like skeletal muscle and brown adipose tissue. The VMH has glucosensing neurons which have been well studied. VMH neurons also have receptors for hormones like leptin and insulin and sense and integrate these hormones to regulate food intake and metabolism. Microinjection of leptin into the VMH increased glucose uptake by the skeletal muscle and brown adipose tissue (BAT) through the sympathetic nervous system.
The VMH consists of different neuron populations characterized by the expression of either steroidogenic factor-1 (SF-1), brain derived neurotrophic factor (BDNF), pituitary adenylate cyclase activating polypeptide (PACAP) or cerebellin 1 (Cbln1). SF-1 neurons are concentrated in the dorsomedial part of the VMH and play a key role in the development of VMH structure as well as expression of other genes. Its indispensable role in the development of the VMH and the effect of its knockout on body weight regulation has shed light on the importance of SF-1 neurons in energy homeostasis. Most of the studies demonstrating the importance of SF-1 neurons in the regulation of body weight, food intake and glucose metabolism deleted a key receptor or signaling molecule from SF-1 neurons. However, deletion of a key receptor or signaling molecule could interfere with the normal development of the VMH and its neuronal circuit, which in turn, could be responsible for the
reported results. Also, the effect of specific activation or inhibition of SF-1 neurons is unknown. Therefore, in my research, I examined the effect of activation or inhibition of VMH specific SF-1 neurons on energy and glucose metabolism. I used the DREADD system, to activate or inhibit SF-1 neurons.
MATERIALS AND METHODS:
SF-1 Cre-recombinase transgenic (SF-1 Cre Tg) mice were obtained from The Jackson Laboratory [STOCK Tg(Nr5a1-cre)7Lowl/J]. Bilateral steel cannualae were implanted to twelve-fourteen weeks old SF-1 Cre Tg mice targeting the VMH. The stereotaxic coordinates obtained from the Atlas by Franklin and Paxinos, were AP 1.32 (1.32 mm anterior to the bregma), H 5.72 (5.72 mm below the surface of the skull above bregma) and L 0.3 (0.3 mm lateral to bregma on either sides). Surgery for catheter implantation into the jugular vein and carotid artery was carried out 3 days before day of clamp. Catheters were flushed daily after surgery to prevent clotting in the catheters. Two weeks after VMH cannulation, DREADD (hM3Dq and hM4Di) AAV vectors (500 nl) were microinjected to both sides at the rate of 50 nl/ min. To activate the DREADD, clozapine-n-oxide (CNO) (Enzo Life Sciences, New York, USA) was injected into AAV infected mice intraperitoneally. On another day, same mice were injected saline and used as a control, except for hyperinsulinemic-euglycemic clamp and 2-[14C] Deoxyglucose uptake experiments. For these experiments, wild-type mice obtained from SF-1 Cre Tg breeding and injected with saline was used as control mice.
Food intake was measured for 3 hours after overnight fasting and during the start of dark period. Total energy expenditure and respiratory quotient (RQ) were measured by indirect calorimetry with use of gas analysis consisting CO2 and O2 mass spectrometric analyzer (Arco-2000, Arco system, Chiba, Japan). Locomotor activity (cm/min) was simultaneously measured with force plate system that was set below the cage (Actracer-2000, Arco system). Energy expenditure and the amount of carbohydrate or fat oxidized were calculated from O2 consumption (VO2) and CO2 production (VCO2) as described previously. Equations for those calculations were as follows:
Energy expenditure (cal/min) = 3.816 × VO2 + 1.231 × VCO2
Carbohydrate oxidation (cal/min) = [(4.51 × VCO2) - (3.18 × VO2)] × 4.1 Lipid oxidation (cal/min) = [1.67 (VO2 - VCO2)] ×9.3
Mice were acclimated to the special cages for 3 days prior to measurement. During this period, body weight and food intake were measured to ensure acclimatization. On the day of the experiment, saline or CNO was injected and VO2, VCO2 and locomotor activity were measured every minute. The data were averaged for every consecutive 5 minutes and summarized for 1 hour.
For insulin tolerance tests (ITT), food was removed 1 hour before the start of the test. At time (t) = -30 (minutes), saline or clozapine-n-oxide (CNO) (Enzo Life Sciences) was injected into mice expressing DREADD (hM3Dq or hM4Di) in the VMH, intraperitoneally. At t = 0, insulin (1U/kg, Sigma-Aldrich Japan, Tokyo, Japan) was injected intraperitoneally. Blood glucose was measured at t
= -30, 0, 15, 30, 60 and 90 using One Touch glucometer (Life Scan Japan, Johnson & Johnson, Tokyo, Japan).
The hyperinsulinemic-euglycemic clamp protocol was performed in conscious and unrestrained mice. Insulin (2.5mU/kg/min) was infused to induce hyperinsulinemia. In order to maintain the euglycemia, 30% glucose was infused. For assessment of glucose uptake, 2-[14C] Deoxyglucose (2DG) was injected intravenously. An arterial blood sample (40 µl) was collected at t = 0, 5, 15, 25, 35, 45 after injection of 2DG. Immediately after collection of the final blood sample, mice were killed with an overdose of pentobarbital sodium, and the soleus, red and white part of the gastrocnemius, inguinal white adipose tissue (Inguinal WAT), BAT, spleen, heart, and liver were rapidly dissected and immediately dropped in liquid N2 for immunoblot analysis and 2DG uptake measurement.
For immunoblot analysis, tissues were homogenized on ice in 20mM Tris (pH 7.4), 5mM EDTA (pH8.0), 10mM sodium pyrophosphate, 1% Triton X, phosphatase inhibitor (Calbiochem, San Diego, USA) and proteinase inhibitor (Sigma-Aldrich Japan). The homogenates were centrifuged, and the resulting supernatants (20µg of protein) were fractionated by SDS-PAGE. Immunoblot analysis was then carried out with specific antibodies. Immune complexes were visualized with horseradish peroxidase–conjugated secondary antibodies (Santa Cruz Biotechnology, Santa Cruz, USA) and
enhanced chemiluminescence reagents (GE Healthcare, Tokyo, Japan). Protein bands were quantified using CS analyzer software (Atto Corporation, Tokyo, Japan).
Total RNA was isolated from each tissue with the use of Trizol (Life Technologies Japan, Tokyo, Japan), and portions of the RNA (400 ng) were subjected to reverse transcription with an oligo dT primer and avian myeloblastosis virus reverse transcriptase (Takara, Shiga, Japan). The resulting cDNA was measured and was subjected to qPCR. For qPCR, cDNA was amplified using SYBR Green PCR Mix and dye (Takara) with an ABI 7500 real-time PCR system. Data were normalized by the amount of 36B4.
All animal experiments were performed in accordance with institutional guidelines for the care and handling of experimental animals and they were approved by the Institutional Animal Care and Use Committee (IACUC) of the National Institutes of Natural Sciences (NINS).
Data are as means ± standard error of the mean (SEM). Statistical comparisons among multiple groups and between 2 groups were performed by analysis of variance (ANOVA) followed by Tukey's HSD post hoc test, and Student’s t test, respectively. A P value of <0.05 was considered statistically significant.
RESULTS
SF-1 Cre Tg mice expressed cre recombinase in the VMH, pituitary gland, adrenal glands and gonads. I examined posthumous brain sections from DREADD AAV infected mice and they showed expression of DREADD-mCherry only in the VMH. CNO injection into hM3Dq expressing SF-1 Cre Tg (hM3Dq) mice activated SF-1 neurons, while CNO injection into hM4Di expressing SF-1 Cre Tg (hM4Di) mice inhibited SF-1 neurons.
Since the VMH and its SF-1 neurons have been suggested to play a role in food intake regulation, I measured 3 hours food intake after overnight fasting in hM3Dq and hM4di mice. CNO injection to hM3Dq mice decreased the 3 hours food intake compared to control saline injection. CNO injection to hM4Di mice didn’t alter its food intake. I also measured the food intake for the first 3 hours of dark period. CNO decreased food intake during the dark period in hM3Dq mice. However, there was no change in food intake in hM4Di mice. Both these results indicate that activation of SF-1 neurons decreased food intake, while inhibition of SF-1 neurons did not alter food intake.
Next, I measured energy expenditure, respiratory quotient (RQ) and locomotor activity using the Arco system. Prior to the start of experiment, I kept the mice in the metabolic cages to acclimatize them to the cages. On the day of measurement, I injected saline or CNO at 9:00 into hM3Dq mice and measured energy expenditure, RQ and locomotor activity for 5 hours post injection. RQ, which is the ratio of the volume of CO2 produced by the volume of O2 consumed was also measured and calculated. CNO injection to hM3Dq mice increased the energy expenditure compared to the control saline injection. CNO injection showed a tendency to decrease locomotor activity compared to saline injection. CNO injection into hM3Dq mice decreased the RQ significantly. From the VO2 and VCO2, I calculated the energy from carbohydrate oxidation or fat oxidation using formulas mentioned in materials and methods. CNO injection increased fat oxidation whereas there was no change in carbohydrate oxidation. I also measured energy expenditure in hM4Di mice during the dark period.
CNO injection to hM4Di mice did not alter energy expenditure. There was also no change in locomotor activity in hM4Di mice after CNO. These results suggest that increased energy expenditure on activation of SF-1 neurons was due to increased fatty acid oxidation and independent of locomotor activity. Conversely, inhibition of SF-1 neurons didn’t alter energy expenditure or locomotor activity.
Furthermore, CNO injection to hM3Dq mice increased insulin sensitivity and improved insulin tolerance during ITT. In contrast, injection of CNO into hM4Di mice impaired insulin tolerance during ITT. During hyperinsulinemic-euglycemic clamp, SF-1 neuronal activation significantly increased glucose infusion rate required to maintain euglycemia. In accordance, activation of SF-1 neurons also enhanced insulin-induced suppression of gene expression of rate-limiting gluconeogenic enzymes like glucose 6 phosphatase (G6Pase) and phosphoenol pyruvate carboxykinase (PEPCK) in the liver significantly. Activation of SF-1 neurons showed a tendency to increase glucose uptake by the red type skeletal muscle under basal insulin levels. However under hyperinsulinemia, SF-1 neuronal activation greatly increased insulin-stimulated glucose uptake in red type skeletal muscle. I found an increase in phosphorylation of Akt (a key molecule in insulin signaling) in the soleus muscle of hM3Dq mice, further demonstrating an increase in insulin sensitivity in the muscle under hyperinsulinemia.
DISCUSSION
The occurrence of metabolic diseases like obesity and diabetes has been on the rise globally, becoming a serious health problem worldwide. Imbalances in whole body energy and glucose homeostasis are closely linked to the trend of increased obesity and diabetes. Therefore, understanding the central- peripheral pathways and molecular and cellular mechanisms underlying energy balance, glucose homeostasis, and insulin sensitivity is critical for developing new strategies for the prevention and treatment of metabolic diseases including obesity and diabetes.
In my study, using a pharmacogenetic method – DREADD system, I found that activation of SF-1 neurons decreased food intake, whereas inhibition of SF-1 neurons didn’t affect food intake. I also found that activation of SF-1 neurons increased energy expenditure. I found that activation of SF- 1 neurons decreased RQ, suggesting an increase in fat oxidation. Consistently, activation of SF-1 neurons increased fat oxidation, while carbohydrate oxidation was unchanged. In the case of inhibition of SF-1 neurons, O2 consumption, locomotor activity was unchanged.
Furthermore, activation of SF-1 neurons improved insulin tolerance during ITT. In contrast, inhibition of SF-1 neurons impaired insulin tolerance during ITT. Activation of SF-1 neurons also enhanced GIR to keep plasma glucose level and insulin-stimulated glucose uptake by red type skeletal muscle and suppression of hepatic gluconeogenic genes’ expression during hyperinsulinemic- euglycemic clamp.
In summary, my data suggest that SF-1 neurons in the VMH play key roles in the regulation of food intake, energy expenditure, peripheral insulin sensitivity and glucose metabolism. These results support the idea that SF-1 neurons in the VMH are a crucial neuronal population for whole-body metabolism and suggest that through its activity, it regulates energy balance and glucose homeostasis.
