Control of Physiologic Glucose Homeostasis via the Hypothalamic Modulation of Gluconeogenic Substrate Availability

The brain augments glucose production during fasting, but the mechanisms are poorly understood. Here, we show that Cckbr-expressing neurons in the ventromedial hypothalamic nucleus (VMNCckbr cells) prevent low blood glucose during fasting through sympathetic nervous system (SNS)-mediated augmentation of adipose tissue lipolysis and substrate release. Activating VMNCckbr neurons mobilized gluconeogenic substrates without altering glycogenolysis or gluconeogenic enzyme expression. Silencing these cells (CckbrTetTox animals) reduced fasting blood glucose, impaired lipolysis, and decreased circulating glycerol (but not other gluconeogenic substrates) despite normal insulin, counterregulatory hormones, liver glycogen, and liver gluconeogenic gene expression. Furthermore, β3-adrenergic adipose tissue stimulation in CckbrTetTox animals restored lipolysis and blood glucose. Hence, VMNCckbr neurons impact blood glucose not by controlling islet or liver physiology, but rather by mobilizing gluconeogenic substrates. These findings establish a central role for hypothalamic and SNS signaling during normal glucose homeostasis and highlight the importance of gluconeogenic substrate mobilization during physiologic fasting.


Abstract:
The brain augments glucose production during fasting, but the mechanisms are poorly understood.Here, we show that Cckbr-expressing neurons in the ventromedial hypothalamic nucleus (VMN Cckbr cells) prevent low blood glucose during fasting through sympathetic nervous system (SNS)-mediated augmentation of adipose tissue lipolysis and substrate release.Activating VMN Cckbr neurons mobilized gluconeogenic substrates without altering glycogenolysis or gluconeogenic enzyme expression.Silencing these cells (Cckbr TetTox animals) reduced fasting blood glucose, impaired lipolysis, and decreased circulating glycerol (but not other gluconeogenic substrates) despite normal insulin, counterregulatory hormones, liver glycogen, and liver gluconeogenic gene expression.Furthermore, b3-adrenergic adipose tissue stimulation in Cckbr TetTox animals restored lipolysis and blood glucose.Hence, VMN Cckbr neurons impact blood glucose not by controlling islet or liver physiology, but rather by mobilizing gluconeogenic substrates.These findings establish a central role for hypothalamic and SNS signaling during normal glucose homeostasis and highlight the importance of gluconeogenic substrate mobilization during physiologic fasting.
Mammals maintain blood glucose within a tight physiologic range by integrating environmental, nutrient, and hormonal signals across multiple organ systems.While the actions of pancreatic islet hormones on liver, adipose tissue, and muscle plays crucial roles in prandial and fasting glucose homeostasis 1 , the central nervous system (CNS) maintains blood glucose during times when glucose must be generated from stored metabolic fuels 2 , such as during inter-meal intervals (i.e., physiologic fasting) or in response to allostatic stressors such as hypoglycemia [3][4][5][6] .Some CNS circuits, such as those in the hypothalamic arcuate nucleus (ARC), modulate islet hormone secretion and hepatic insulin sensitivity [7][8][9][10] and others, including Lepr-expressing neurons of the ventromedial nucleus of the hypothalamus (VMN; VMN Lepr neurons), tend to decrease blood glucose by promoting glucose uptake and metabolism [11][12][13][14] .Hence, the CNS contributes to control of glucose homeostasis at least in part by regulating islet hormone secretion and augmenting glucose disposal into metabolically active tissues.
We recently found that VMN neurons defined by expression of cholecystokinin b receptor (Cckbr; VMN Cckbr neurons, which are largely distinct from VMN Lepr neurons) contribute to the maintenance of blood glucose during physiologic fasting, in addition to driving glucose production during the counter-regulatory response (CRR) to hypoglycemia 15 .In contrast to previously described neuronal populations, however, silencing VMN Cckbr neurons lowers blood glucose during physiologic fasting without altering islet hormone secretion or glucose disposal, but rather by controlling glucose production 15 .Thus, VMN Cckbr neurons must control blood glucose by modulating sympathoadrenal (i.e., epinephrine/norepinephrine and glucocorticoid) function, liver gluconeogenic capacity, and/or the liberation of gluconeogenic substrates from tissue storage depots.We previously showed that VMN Cckbr neuron activation augments sympathoadrenal output, but it remains unclear whether these neurons also alter gluconeogenic capacity and/or other aspects of liver physiology.
Here, we identify the previously unknown mechanisms by which VMN Cckbr neurons control glucose homeostasis during physiologic fasting.The acute activation of VMN Cckbr neurons, as during allostatic challenges like hypoglycemia, increases sympathoadrenal output and promotes the mobilization of several gluconeogenic substrates to rapidly increase blood glucose without altering islet or liver physiology.
Similarly, VMN Cckbr neurons maintain fasting glucose concentrations without altering circulating hormones or liver physiology, but instead support sympathetic outflow to white adipose tissue (WAT), promoting lipolysis to provide a continuous source of glycerol to fuel glucose production.These data reveal a previously unrecognized islet hormone-and glucose uptake-independent gluconeogenic substrate-driven mechanism by which the CNS controls fasting blood glucose.We previously found that activating VMN Cckbr neurons, like the CRR, promotes the release of corticosterone and catecholamines without altering islet hormones 15 .Because these counterregulatory mediators act on liver, muscle, and fat to promote the release of gluconeogenic substrates, we measured plasma pyruvate, lactate, amino acids, and glycerol following the optogenetic stimulation of VMN Cckbr neurons.We found that VMN Cckbr neuron activation increased circulating pyruvate, lactate, and glycerol, although amino acids did not change (Fig 2A).

Activating VMN
VMN Cckbr neuron signaling is necessary for appropriate glucose mobilization during starvation, insulinopenic diabetes, and CRR-conditions in which the utilization of lipid stores provides an alternative source of energy via ketone bodies 15,17 .We thus predicted that activating VMN Cckbr neurons would increase ketone body production.Indeed, VMN Cckbr neuron stimulation robustly increased blood ketones (Fig 2B).Treating mice with a β3-adrenergic receptor antagonist (SR-59230A), which primarily blocks adrenergic signaling to adipose tissue, abrogated ketone production but not glucose mobilization/hyperglycemia (Fig 2C).These findings demonstrate that VMN Cckbr neuron activation promotes ketone production and that this ketogenesis (but not the accompanying glucose production) requires adipose tissue lipolysis, consistent with the idea that fatty acids are needed for ketone production but that gluconeogenic substrates other than glycerol (e.g., pyruvate, lactate, and/or amino acids) that derive from nonadipose tissue sources suffice to support glucose production in response to optogenetic VMN Cckbr neuron activation.
Silencing VMN Cckbr neurons impairs gluconeogenic substrate delivery from adipose tissues Tetanus toxin-mediated silencing of VMN Cckbr neurons (Cckbr TetTox mice) reduces blood glucose across the diurnal cycle, with the largest difference occurring at the onset of the light period, when food intake rapidly drops (Fig 3A).This suggests that the activity of these neurons contributes not only to the response to allostatic stimuli 15 , but also to the homeostatic control of blood glucose during the early stages of fasting.We thus sought to establish the mechanisms by which physiologic action of VMN Cckbr neurons modulates fasting blood glucose.
We measured islet and adrenal hormones following a 4-hour fast during the early light phase (Fig 3B ), revealing unchanged insulin, glucagon, and corticosterone in Cckbr TetTox compared to control mice (despite Cckbr TetTox animals having lower blood glucose levels).The finding of unchanged islet hormones in Cckbr TetTox animals is consistent with the failure of optogenetically stimulating VMN Cckbr neurons to alter insulin or glucagon 15 .Although VMN Cckbr neuron activation increases corticosterone 15 , the normal circulating corticosterone of Cckbr TetTox animals suggests that processes other than those mediated by VMN Cckbr neurons suffice to maintain baseline corticosterone concentrations during physiologic fasting.
Despite the chronic nature of VMN Cckbr neuron silencing in Cckbr TetTox animals, which should provide adequate opportunity for any alterations in gene expression to manifest, the expression of gluconeogenic genes in the liver remained unchanged in Cckbr TetTox animals compared to controls (Fig 3C ), as did liver glycogen content (Fig 3D).Thus, the decreased blood glucose of Cckbr TetTox animals during physiologic fasting does not result from changes in circulating hormones, liver glycogen, or gluconeogenic gene expression.
To determine whether decreased blood glucose in Cckbr TetTox mice might result from impaired sensitivity to glucagon and/or the inability to mobilize glycogen stores, we monitored blood glucose in Cckbr TetTox mice and controls following the injection of glucagon, which rapidly mobilizes hepatic glycogen and increases hepatic gluconeogenic capacity over the longer term.We observed similarly robust initial glucose excursions in Cckbr TetTox mice and controls following glucagon injection, although blood glucose levels decreased more rapidly in Cckbr TetTox mice (Fig 3E).
These data are consistent with normal short-term glucagon-stimulated glucose mobilization in Cckbr TetTox mice, despite a potentially impaired long-term gluconeogenic response.
Overall, these findings suggest that the low blood glucose of Cckbr TetTox mice does not result from alterations in islet hormones, corticosterone, or the glycogenolytic response to glucagon.Furthermore, any alteration in gluconeogenesis in the Cckbr TetTox animals does not arise from changes in liver gluconeogenic gene expression.Hence, given the robust mobilization of gluconeogenic substrates by VMN Cckbr neuron activation, we tested the hypothesis that low gluconeogenic substrate availability underlies the low fasting blood glucose observed in Cckbr TetTox animals.Because increasing SNS activation of WAT in vivo increases lipolysis and plasma glycerol concentrations, we predicted that this increase in gluconeogenic substrate availability would also increase glucose production.Indeed, treating mice with CL-316,243 similarly increased blood glucose in Cckbr TetTox and control animals (Fig 5C-D).Hence, restoring adipose tissue lipolysis alleviates the glucose production defect caused by silencing VMN Cckbr neurons.This suggests that VMN Cckbr neurons regulate SNS outflow to WAT, promoting lipolysis to provide glycerol that fuels the gluconeogenesis needed to maintain normoglycemia during in the early stages of physiologic fasting.
To understand the cause of decreased adipose tissue lipolysis in Cckbr TetTox mice, we collected and analyzed gonadal and subcutaneous WAT (gWAT and psWAT, respectively) depots from Cckbr TetTox mice 10 -12 weeks after viral injection.While we found no significant difference in adipose depot mass or adipocyte size between Cckbr TetTox and control GFP mice in either WAT depot (Fig 6A -B), adipose depot mass tended to be increased in Cckbr TetTox mice.As Cckbr TetTox WAT depot size is not lower than controls, the observed decrease in lipolysis in these animals does not result from decreased lipid stores.The variance in adipocyte size differed significantly between Cckbr TetTox mice and controls for both depots (F = 8.29, DFn = 7, DFd = 5, p = 0.03 for gWAT and F = 18.39,DFn = 7, DFd = 5, p = 0.005 for psWAT) (Fig 6A -B), however, consistent with altered adipose tissue remodeling due to impaired lipolysis in the Cckbr TetTox adipose tissue 22 .This finding, in combination with our observation that pharmacologic activation of SNS signaling restores WAT lipolysis in Ccbkr TetTox mice suggests that the reduction in glycerol and lipolysis results from impaired SNS outflow to WAT, rather than an intrinsic defect in WAT function.

Discussion
Our findings demonstrate that VMN Cckbr neurons do not alter glycogenolysis, gluconeogenic gene expression, or pancreatic hormones, but rather maintain appropriate blood glucose by mobilizing gluconeogenic substrates.While artificial activation of VMN Cckbr cells increases the availability of multiple gluconeogenic substrates (as during allostatic responses), during the early stages of physiologic fasting VMN Cckbr neurons mainly promote SNS-mediated WAT lipolysis to provide glycerol as a gluconeogenic substrate.In addition to revealing that CNS-directed control of gluconeogenic substrate mobilization plays a crucial role in determining blood glucose concentrations during fasting, these findings demonstrate that distinct CNS circuits control the individual processes (e.g., pancreatic hormone secretion, liver physiology, and substrate mobilization) that contribute to glucose production and fasting glucose homeostasis.
In contrast, we found that activating the VMN Cckbr subset of VMN neurons promotes glucose production primarily via the sympathoadrenal mobilization of gluconeogenic substrates.Hence the mobilization of gluconeogenic substrates by the CNS suffices to promote glucose production, and this process is separable from other aspects of CNS-mediated glucose mobilization during allostatic-like responses.While activating VMN Cckbr neurons increases glycerol, pyruvate, and lactate, it does not alter circulating amino acid concentrations.This suggests either that long-term activation is required to mobilize amino acids or that amino acid conversion to glucose is sufficiently rapid that our methods are unable to detect underlying changes in amino acid mobilization.Alternatively, a distinct circuit may regulate protein degradation and amino acid mobilization.
Globally interfering with VMN function (e.g., by deleting VMN vGLUT2, thereby blocking glutamatergic signaling by VMN neurons) mediates effects opposite to those of VMN activation, impairing glucagon secretion, glycogenolysis, and hepatic transcriptional responses to hypoglycemia and during fasting 26 .While silencing the VMN Cckbr subset of VMN neurons decreases fasting blood glucose, it does not alter pancreatic hormone secretion, glycogen responsiveness/glycogenolysis, or long-term hepatic gluconeogenic gene expression.Rather, Cckbr TetTox mice exhibit decreased glycerol production and a normal hyperglycemic response to exogenous substrates, suggesting that within the VMN, distinct cell types may control different components of the glycemic response to fasting.
While hepatic gluconeogenesis uses multiple substrates (e.g., amino acids and TCA cycle-derived lactate and pyruvate, in addition to glycerol), hormonal signaling and the availability of individual substrates dictates the relative importance of each under a distinct conditions.Our data demonstrate that during early physiologic fasting, glycerol represents a key gluconeogenic substrate for the maintenance of adequate blood glucose.Indeed, other recent studies have demonstrated the importance of glycerol for hepatic gluconeogenesis under physiologic conditions [27][28][29] .
The VMN controls sympathetic outflow to multiple metabolic target organs, including islets 8 , liver, and BAT 11 .Our finding that VMN Cckbr neurons support lipolysis are consistent with previous reports that manipulating VMN neuronal activity alters FFA mobilization and lipid metabolism [30][31][32] .However, to date, retrograde, polysynaptic viralbased tracing studies from WAT have identified few VMN neurons (which appear only after prolonged infection periods 33 ).While other hypothalamic sites (such as the PVN and DMH, which lie downstream from the VMN) connect more directly to WAT, VMN Cckbr neurons do not project to these regions 15,34 .Thus, the downstream neuroanatomic pathways through which VMN Cckbr neurons mediate WAT lipolysis remain to be defined.
Integrating these findings into a comprehensive model of overall glucose homeostasis, VMN Cckbr neurons (and the CNS in general) play a crucial role in maintaining adequate blood glucose during physiologic fasting and allostatic responses, while CNS effects are less prominent during feeding.As insulin decreases during fasting and hormonal and nutrient signals drive liver physiology toward gluconeogenesis, the brain increases lipolysis through activation of the SNS to provide glycerol for gluconeogenesis, while continued low level insulin action prevents unrestrained catabolism in WAT and skeletal muscle.
In response to strong allostatic challenges, such as hypoglycemia, the CNS coordinates the CRR to promote glucagon release and suppress insulin secretion, which augments glycogenolysis and gluconeogenic capacity.Simultaneously, adrenal glucocorticoid secretion and direct tissue SNS output mobilize gluconeogenic substrates to prevent low blood glucose 35,36 .Whereas direct glucose sensing by glucagonsecreting islet alpha cells contribute to glucagon secretion, intact CNS circuit signaling (including through the VMN) 1,25,26 is necessary for adequate glucose mobilization.
VMN Cckbr neurons contribute to the sympathoadrenal responses that provide the substrate to drive gluconeogenesis (and ketogenesis) during allostatic responses, but do not control glucagon secretion 15 ; thus, additional circuits that contribute to other aspects of the CRR remain to be identified.
In type 1 diabetes, the absolute lack of insulin increases hepatic gluconeogenic capacity and derepresses lipolysis and proteolysis, promoting unrestrained release of substrates for gluconeogenesis (and ketogenesis) and raising blood glucose.Because silencing VMN Cckbr neurons decreases blood glucose in models of type 1 diabetes 15 , CNS-mediated processes also contribute to the substrate mobilization that contributes to the hyperglycemia of insulinopenic diabetes.
The signals that modulate VMN Cckbr neuronal activity (and other CNS neurons that control glucose production) during physiologic fasting, allostatic responses, and insulinopenic diabetes remain unknown and thus represent important future research directions.To understand how dysregulation of VMN Cckbr neurons and similar CNS systems might contribute to the pathophysiology of type 2 diabetes, it will also be important to understand the regulation of these neurons in diabetic states, as well as in obesity.Even absent a comprehensive understanding of their regulation, however, VMN Cckbr neurons could represent a pharmacologic target by which to improve glycemic control in all forms of diabetes.

Animals
Mice were bred in our colony in the Unit for Laboratory Animal Medicine at the University of Michigan; these mice and the procedures performed were approved by the University of Michigan Committee on the Use and Care of Animals (Protocol# 00011066) and in accordance with Association for the Assessment and Approval of Laboratory Animal Care and National Institutes of Health guidelines.Mice were provided with food and water ad libitum (except as noted below) in temperaturecontrolled rooms on a 12-hour light-dark cycle.For all studies, animals were processed in the order of their ear tag number, which was randomly assigned at the time of tailing (before genotyping).ARRIVE guidelines were followed; animals were group-housed, unless otherwise noted.
Cckbr Cre mice were previously described 15 .Prior to experiments, mice were genotyped via PCR across the genomic region of interest.Animals were processed and studied in the order of their randomly assigned ear tag number and investigators were blinded to genotype and/or treatment for all studies.All experiments were conducted with approximately equal number of males and females unless otherwise indicated.

Reagents
Viral reagents AAV-DIO-ChR2-eYFP, AAV-DIO-GFP, and AAV-DIO-TT were as previously described 15 and were packaged by the University of Michigan Viral Vector Core (University of Michigan, Ann Arbor, Michigan, USA).

Stereotaxic injections
Eight-to 12-week-old animals were anesthetized with 1.5% to 2% isoflurane in preparation for craniotomy.After exposing the skull, bregma and lambda were leveled, a hole was drilled, the tip of a pulled glass pipette was inserted at the coordinates of our target, and the contents of the pipette were released at approximately 25 nL/min.For the VMN, 100 -200 nL virus was injected at anteroposterior (AP) -0.750, mediolateral (ML) ±0.35, and dorsoventral (DV) -5.60.After 3 minutes to allow the virus to diffuse into the brain, the pipette was raised slowly from the hole in the skull, the hole in the skull was filled with bone wax, and the skin was closed with surgical sutures (Henry Schein).Analgesics were administered prophylactically to all mice to prevent postsurgical pain.Mice were allowed 4 weeks to recover from surgery before experimental manipulation.Analysis of fluorescent reporter expression was used to confirm proper targeting of the brain region in all experiments.Cases lacking expression in the region of interest were omitted from analyses.

Optogenetics
Optogenetic fibers were placed 0.5 mm above the injection site for the VMN (AP: -0.75 mm, ML:0.35 mm, and DV: -5.10 mm from bregma) and were anchored above the skull with Metabond.Mice were allowed at least 4 weeks to recover from surgery before stimulation experiments.Laser stimulation (473nm, MBL-F-473, Opto Engine) was performed at 5-ms pulses, 40 pulses per second for a total of 30 -60 minutes with an approximate irradiance 21 mW/mm 2 .Parameters were chosen to match previous optogenetic VMN stimulation that examined glucose mobilization 15,25 .Mice were fasted for 2 hours in the early light period prior to optogenetic stimulation.Blood was obtained from the tail vein and blood glucose was measured with Accu-Chek blood glucose meter (Accu-Chek Aviva Plus, Roche Diagnostics) for data from Fig 1A, and all glucose measurements in Figure 2. Blood glucose for figure 1C was measured in 10uL whole blood lysed in 1mL RUO Glucose/Lactate Hemolyzing Solution (EKF Diagnostics) using a Biosen C-Line Glucose and Lactate Analyzer (EKF Diagnostics).Blood for gluconeogenic substrate and hormone measurement was collected in EDTA-coated microtubes.

Hormone and metabolite measurements
Insulin and glucagon were measured according to manufacturer's instructions (Crystal Chem Ultra High Sensitive Mouse Insulin ELISA, cat: 90080; Mercordia Glucagon ELISA, cat: 10-1281-01).Corticosterone was measured in 20μL of serum by liquid chromatography-tandem mass spectrometry as previously described 37 .Plasma pyruvate, amino acids, glycerol, and non-esterified fatty acids were measured with a commercial colorimetric kit according to manufacturer's instructions as follows: pyruvate (Sigma Aldrich, MAK071, 10μL), amino acids (Abcam, ab65347, 5μL), glycerol (Sigma Aldrich, MAK117, 10 μL), NEFA (Sigma Aldrich MAK044, 5 μL).All assays were performed in duplicate and compared to the provided standard curve.Lactate was measured in 10uL whole blood lysed in 1mL RUO Glucose/Lactate Hemolyzing Solution (EKF Diagnostics) using a Biosen C-Line Glucose and Lactate Analyzer (EKF Diagnostics).Plasma ketone levels were measured using 1.5μL of tail blood on Precision Xtra Blood Ketone Test Strips and a handheld Precision Xtra Blood Glucose and Ketone Meter (Abbott Labs).For liver glycogen measurements, liver was collected following euthanasia with isoflurane, frozen on dry ice and stored at -80C.For liver homogenization, 10 -20mg of frozen liver was homogenized in 100uL sterile water, boiled for 5 minutes, and centrifuged at 13,000rpm for 5 minutes.The supernatant was then diluted 1:1 with sterile water and 20μL was used for glycogen measurement (Glycogen Assay Kit, Cat: MAK016, Sigma).

Perfusion and IHC to confirm viral targeting
To confirm viral targeting for all mice, brains were collected and processed as previously described 17 .Briefly, mice were anesthetized with isoflurane before transcardial perfusion with PBS followed by 10% formalin.Brains were removed and placed into 10% formalin overnight, followed by 30% sucrose for at least 36 hours.
Brains were cut into 30-μm sections on a freezing microtome in 4 series and stored in anti-freeze solution (25% ethylene glycol, 25% glycerol).Sections were washed with PBS and then pretreated for 1 hour in blocking solution (PBS containing 0.1% triton, 3% normal donkey serum) and then incubated overnight in blocking solution containing chicken anti-GFP (1:1000, GFP-1020, Aves).The following day, sections were washed in PBS, treated with blocking solution containing anti-chicken Alexa Fluor 488 fluorescent secondary antibody (1:200), and washed again in PBS.Sections were mounted onto slides and cover-slipped with Fluoromount-G (Southern Biotech).Slides were imaged using an Olympus Bx53 microscope.

Glycerol Appearance Assay
Glycerol appearance assays were performed with the assistance of the University of Michigan Animal Phenotyping Core.Under anesthesia, mice were catheterized with silicon tubing.After surgery, mice were housed individually, and their body weight was monitored daily.Mice that lost more than 10% of their presurgery body weight were removed from the study.On the day of data collection, mice were fasted for 2.5 hours starting 3 hours after lights-on.[2-3 H]glycerol was infused using a boluscontinuous infusion protocol of 2μCi in 2min + 0.1μCi/min for a 90-minute equilibration period.After the equilibration period, [2-3 H]glycerol was infused at a steady state and plasma collected for blood glucose, plasma glycerol and [ 3 H] plasma measurements as previously described 38 .Thirty minutes after initial plasma collection, β3-agonist CL-316,243 (1.5mg/kg, i.v.Sigma) was injected and plasma glucose, glycerol, and plasma [ 3 H] radioactivity were measured at baseline and after 10 minutes.

Continuous Glucose Monitoring
Continuous glucose monitors were surgically implanted, and data was collected with assistance from the University of Michigan Physiology Phenotyping Core.Briefly, under isoflurane anesthesia mice were shaved on the ventral neck and a midline skin incision from chin to mid-sternum was performed, and the salivary glands were retracted.The left common carotid artery was dissected and distally ligated and proximally occluded with silk sutures.A bent 25-gauge needle was inserted into the common carotid and the glucose sensor tip of a HD-XG mouse glucose telemetry implant (Data Sciences International, St. Paul, MN) was inserted and then advanced to the aortic arch and secured with tissue adhesive and a silk suture.The body of the device was rotated caudally and inserted into a subcutaneous pocket that was created over the right ventral flank.The glands were reapproximated and the skin incision closed with 2-3 staples.The animal was returned to its housing cage and placed on a heating pad until it was mobile.
Following surgical recovery, singly housed mice were placed on a receiver plate.
To calibrate the glucose telemetry probe, a tail vein blood sample was taken to measure Statistical Analysis Sample sizes were not predetermined using statistical tests but were similar to previous studies with similar approaches 15 .Figure 1 and Figure 2 C-D were analyzed using a paired Student's t test.Figure 2A, Figure 3 B-E, Figure 4, Figure 5, and Figure 6 data were analyzed by unpaired Student's t test.Data were omitted if the injection missed or the animal was sick or injured at the time of the experiment (loss of >10% BW).Continuous glucose monitoring was separated into 4-hr circadian windows and each 4hr window was separately analyzed by 2-way ANOVA.Significance was set at a P value of 0.05 or less.All data analysis was performed using GraphPad Prism software (GraphPad Software).
Cckbr neurons mobilizes gluconeogenic and ketogenic substrates To understand the mechanisms by which VMN Cckbr neuron activation increases blood glucose, we injected AAV DIO-ChR2-GFP into the VMN of Cckbr Cre mice and placed a fiber optic cannula above the injection site (Cckbr ChR2 animals), permitting the activation of VMN Cckbr cells by blue light.Consistent with our previous findings, optogenetic activation of VMN Cckbr neurons rapidly mobilized glucose (Fig 1A); this stimulus produced no change in liver glycogen content after 30 minutes, however (Fig 1B).Furthermore, although pretreatment of the Cckbr ChR2 animals with the glycogenolysis inhibitor CP91,149 tended to decrease blood glucose in the absence of optogenetic stimulation as previously reported 16 , this treatment did not interfere with robust glucose mobilization in response to blue light (Fig 1C).Hence, liver glycogenolysis contributes minimally, if at all, to glucose mobilization during the activation of VMN Cckbr neurons.We also measured the expression of key gluconeogenic genes in the liver following the activation of VMN Cckbr neurons (Fig 1D).Consistent with the rapid onset and peak of glucose mobilization during VMN Cckbr neuron activation (before transcriptionally mediated changes could take effect), the expression of Ppargc1a (Pgc1α) and the gluconeogenic genes Pck (Pepck) and G6pc (G6pase) were unchanged by optogenetic stimulation (Fig 1D).
We measured plasma concentrations of lactate, pyruvate, amino acids, and glycerol in Cckbr TetTox animals following a 4-hour fast.While lactate, pyruvate, and amino acids were not different between Cckbr TetTox mice and controls, Cckbr TetTox mice exhibited plasma glycerol levels 38% lower than those in control mice (Fig 4A).To determine whether this alteration in glycerol might reflect a decrease in adipose tissue lipolysis, we also examined plasma nonesterified fatty acids (NEFA, Fig 4B), revealing decreased NEFA in Cckbr TetTox mice.Because alterations in production or utilization can each impact the concentration of circulating metabolites, we infused radiolabeled glycerol in these mice to directly examine the rate of glycerol appearance in the circulation 18 , revealing decreased glycerol production in the Cckbr TetTox mice compared to controls (Fig 4C).While decreased adipose tissue lipolysis could theoretically decrease blood glucose by increasing glucose oxidation (e.g., in skeletal muscle) to compensate for lost fatty acid oxidation, we previously showed that VMN Cckbr neurons promote glucose production without altering glucose disposal/utilization 15 .Hence, these findings suggest that lower baseline adipose tissue lipolysis in Cckbr TetTox mice may reduce blood glucose by decreasing glycerol-mediated gluconeogenesis.If the impairment in glucose production in Cckbr TetTox mice results from limitations in gluconeogenic substrate delivery in the face of otherwise normal liver gluconeogenic capacity, Cckbr TetTox mice should exhibit normal glycemic responses to gluconeogenic substrate infusion.We thus performed pyruvate (Fig 4D) and glycerol (Fig 4E) challenges in Cckbr TetTox mice, revealing similar glucose excursions in Cckbr TetTox mice and controls in both cases.Thus, Cckbr TetTox mice exhibit normal hepatic gluconeogenic responses to substrate infusion, including in response to glycerol.Together, these findings led us to test whether the decreased blood glucose in Cckbr TetTox mice results from decreased gluconeogenesis due to limitations in substrate availability that stem from impaired adipose tissue lipolysis.Impaired SNS outflow to WAT decreases lipolysis and glycerol-dependent gluconeogenesis in Cckbr TetTox mice.SNS outflow to WAT drives lipolysis through activation of β3-adrenergic receptors, expression of which is predominantly restricted to adipose tissue in mice 19-21 .To determine whether decreased SNS outflow to WAT might underlie the decreased lipolysis and low circulating glycerol observed during physiologic fasting in Cckbr TetTox mice, we treated animals with the β3-adrenergic receptor agonist CL-316,243 to stimulate β3-adrenergic receptors, bypassing the neural/SNS control of lipolysis.While Cckbr TetTox mice exhibited decreased plasma glycerol and in vivo lipolysis following a 4hour fast, CL-316,243 robustly increased both glycerol and lipolysis in these animals (Fig 5A-B).The resulting CL-316,243-stimulated plasma glycerol concentrations and appearance rates did not differ between control and Cckbr TetTox mice (Fig 5A-B).Hence, direct SNS stimulation of adipose tissue equalizes lipolysis and glycerol availability in Cckbr TetTox and control mice.

PhysiologyFigure 1 :Figure 2 :Figure 3 :Figure 4 :
Figure 1: Activating VMN Cckbr neurons mobilizes glucose without altering glycogen or gluconeogenic gene expression.Cckbr Cre mice were injected with AAV-DIO-ChR2-eYFP unliterally into the VMN followed by optogenetic fiber placement to permit optogenetic activation of VMN Cckbr neurons.Mice were fasted for 2 hours during the early light cycle prior to testing.Shown is (A) the glycemic response to light or no light control stimulation (n = 19) and (B) liver glycogen content from tissue collected after 30 minutes of light or control stimulation (n = 5 no light, 8 light).(C) A separate cohort of mice were treated with CP91,149, a glycogenolysis inhibitor, or vehicle at the start of the fasting period and glucose was measured before and 30 minutes after optogenetic stimulation (n = 5); the right-hand panel shows the percent change for light-stimulated vs no light conditions for each treatment.(D) mRNA expression of hepatic gluconeogenic genes was quantified following 30 minutes control or light delivery (n = 6 light off, 8 light on).Data are plotted as mean SEM.*p< 0.05, **p< 0.01, ****p<0.0001,by paired Student's t-test (A, C) and unpaired Student's t-test (B, C(t=0 comparison), D).