Abstract Context Hypoglycemia, one of the major factors limiting optimal glycemic control in insulin-treated patients with diabetes, elicits a brain response to restore normoglycemia by activating counterregulation. Animal data indicate that local release of norepinephrine (NE) in the hypothalamus is important for triggering hypoglycemia-induced counterregulatory (CR) hormonal responses. Objective To examine the potential role of brain noradrenergic (NA) activation in humans during hypoglycemia. Design A hyperinsulinemic-hypoglycemic clamp was performed in conjunction with positron emission tomographic imaging. Participants Nine lean healthy volunteers were studied during the hyperinsulinemic-hypoglycemic clamp. Design Participants received intravenous injections of (S,S)-[11C]O-methylreboxetine ([11C]MRB), a highly selective NE transporter (NET) ligand, at baseline and during hypoglycemia. Results Hypoglycemia increased plasma epinephrine, glucagon, cortisol, and growth hormone and decreased [11C]MRB binding potential (BPND) by 24% ± 12% in the raphe nucleus (P < 0.01). In contrast, changes in [11C]MRB BPND in the hypothalamus positively correlated with increments in epinephrine and glucagon levels and negatively correlated with glucose infusion rate (all P < 0.05). Furthermore, in rat hypothalamus studies, hypoglycemia induced NET translocation from the cytosol to the plasma membrane. Conclusions Insulin-induced hypoglycemia initiated a complex brain NA response in humans. Raphe nuclei, a region involved in regulating autonomic output, motor activity, and hunger, had increased NA activity, whereas the hypothalamus showed a NET-binding pattern that was associated with the individual’s CR response magnitude. These findings suggest that NA output most likely is important for modulating brain responses to hypoglycemia in humans. Hypoglycemia is a common complication observed in insulin-treated patients with diabetes (1). The body responds to hypoglycemia by promoting a counterregulatory (CR) response, which involves the activation of glucose-sensing cells in peripheral tissues and the central nervous system (CNS) (2). In particular, the CNS plays a central role in coordinating this CR response to hypoglycemia by stimulating the release of CR hormones (glucagon, epinephrine, cortisol, and growth hormone) and initiating symptom responses, including activation of the sympathetic autonomic nervous system (palpitations, tremors, sweating) and induction of neuroglucopenic symptoms (hunger, dizziness, confusion) (3). Together, these integrated responses lead to an increase in endogenous glucose production, a decrease in peripheral glucose utilization, hypoglycemia awareness, and stimulation of food-seeking behavior and food intake. Animal studies have provided evidence that systemic hypoglycemia triggers changes in specific glucose-sensing neurons in the CNS that stimulate the release of CR hormones and promote endogenous glucose production, which serves as a primary defense mechanism (4, 5). In particular, local glucoprivation of glucose sensing neurons in the hypothalamus can elicit the CR response (6, 7). Hypothalamic activity is modulated by changes in local neurotransmitter release, including γ-aminobutyric acid, glutamate, serotonin, and norepinephrine (NE) (2). With regard to the latter, noradrenergic (NA) activity in the hypothalamus is increased by both systemic (8) and local (9, 10) hypoglycemia and appears to be particularly important in mounting a CR response. Moreover, activity of hypothalamic tyrosine hydroxylase, the rate-limiting enzyme of NE synthesis, increases in response to both hypoglycemia and neuroglycopenia (11). This provides additional evidence that NA neurons play an important role in regulating glucose metabolism. In keeping with this view, local injection of NE into the hypothalamus has been reported to increase blood glucose levels and promote food intake in rodents (12). However, these observations have been limited to animals. Thus, the extent to which central activation of the CNS NA system promotes the CR response to hypoglycemia in humans remains to be determined. Of note, the cell bodies for the central NA system are grouped in nuclei in the medulla and pons (13) and project axons to most CNS regions, including the cortex, amygdala, hypothalamus, hippocampus, thalamus, brain stem (including the raphe nuclei), cerebellum, and spinal cord (14). When these neurons are activated, they release NE from the terminal axon into the synaptic cleft, which will bind to postsynaptic adrenergic receptors and complete NA neurotransmission. The NE released is cleared from the synaptic cleft by the presynaptic NE transporter (NET), which represents the major mechanism for terminating NE signaling (15). By regulating NE concentration in the synapse, NET modulates NA neurotransmission (15) and serves as an indicator of NA system activity. Selective ligands for assessing NET binding are now available for use in positron emission tomography (PET) experiments to evaluate NET in vivo in the human brain (16, 17). In particular, (S,S)-[(11)C]O-methylreboxetine ([11C]MRB), a highly selective NET ligand, has been used for human brain studies. MRB (an analog of reboxetine that has been approved in Europe as an antidepressant) has a high affinity for NET (18, 19), and the radiotracer [11C]MRB is capable of helping estimate NET availability through determination of its binding potential (BPND). Therefore, the current study was designed to evaluate the CNS NA system in the human brain during hypoglycemia by using a hypoglycemic-hyperinsulinemic clamp technique in conjunction with PET imaging after a bolus injection of the radioligand [11C]MRB. We postulated that hypoglycemia would activate NA neurons, thereby stimulating NE release in the synaptic cleft and in turn competing with [11C]MRB binding to NET, eventually lowering specific binding of [11C]MRB in NET-rich regions of the brain. In addition, we also probed the properties of the hypothalamic NET-specific response in a rat model. NET protein levels in hypothalamic tissue (cytosolic and membrane fractions) were determined under both baseline and hypoglycemia conditions. Materials and Methods Participants Nine healthy volunteers participated in this study [five men and four women; age, 34 ± 4 years; hemoglobin A1c, 5.1% ± 0.2% (32 ± 0.2 mmol/mol); body mass index, 23.4 ± 1.0 kg/m2]. The volunteers came to the Hospital Research Unit at the Yale New Haven Hospital in New Haven, Connecticut, for a screening visit, including a medical history and physical examination performed by one of the study physicians and a urine toxicology screen. Participants with a history of any major medical or psychiatric disease were excluded. Participants who qualified were invited to participate in the PET–hypoglycemic clamp study. The Human Investigation Committee and Radiation Safety Committee at Yale University approved the study. All participants provided written informed consent before participating in the study. This research study is registered at ClinicalTrials.gov: NCT02056249. PET-hypoglycemia study Participants arrived at the PET center at 7 am after an 8-hour overnight fast. An intra-arterial line was placed in the wrist of one arm for blood draws (tracer kinetics, glucose, and hormones) and two intravenous (IV) access sites were obtained in the contralateral arm: one for a bolus infusion of the radiotracer and a second for infusion of insulin and dextrose. The participants were scanned twice: at baseline and during the hypoglycemic-hyperinsulinemic clamp (2 mU/kg per min) (Fig. 1). Before the baseline scan, the participants received an IV bolus injection of [11C]MRB (639 ± 126 MBq) with injected mass of 1.97 ± 0.78 µg and specific activity of 292 ± 121 MBq/nmol. PET acquisition of the brain was obtained for the subsequent 2 hours, starting at the beginning of the [11C]MRB bolus, with a High Resolution Research Tomograph (Siemens/CTI, Knoxville, TN) PET scanner (with an intrinsic resolution of ~3 mm full width half maximum). The second PET scan was started ~2 hours after the baseline scan was completed. A primed-continuous IV infusion of insulin was given at the start of the 6-minute PET transmission scan, which was followed by the second [11C]MRB bolus (644 ± 85 MBq; injected mass, 1.88 ± 0.85 µg; and specific activity = 372 ± 318 MBq/nmol), administered 259 ± 32 minutes after the first bolus injection. The insulin infusion continued throughout the total duration of the PET scan (~120 minutes). Plasma glucose levels were allowed to drop freely to 60 mg/dL, at which time a variable IV dextrose infusion was started, and glucose levels were kept at ~55 mg/dL by adjusting the glucose infusion rate (GIR) based on the plasma glucose levels. Plasma glucose was measured every 5 minutes throughout the study. Insulin and CR hormones (glucagon, catecholamines, cortisol, and growth hormone) were obtained throughout the hyperinsulinemic-hypoglycemic clamp. Figure 1. View largeDownload slide (a) Study design scheme: hyperinsulinemic hypoglycemic clamp combined with PET imaging. At time −120 minutes and at time 0, the participants received an IV bolus infusion of [11C]MRB, followed by a 2-hour-long PET examination. After the second [11C]MRB bolus infusion (time 0), the hyperinsulinemic-hypoglycemic clamp was initiated, which consisted of a primed-constant IV insulin infusion (2 mU/kg/min). An IV variable dextrose infusion was given to maintain plasma glucose levels at target range (50 to 55 mg/dL). (b) Plasma glucose levels and (c) GIR during the hyperinsulinemic-hypoglycemic clamp. The values represent the average values ± standard error of the mean (SE). Figure 1. View largeDownload slide (a) Study design scheme: hyperinsulinemic hypoglycemic clamp combined with PET imaging. At time −120 minutes and at time 0, the participants received an IV bolus infusion of [11C]MRB, followed by a 2-hour-long PET examination. After the second [11C]MRB bolus infusion (time 0), the hyperinsulinemic-hypoglycemic clamp was initiated, which consisted of a primed-constant IV insulin infusion (2 mU/kg/min). An IV variable dextrose infusion was given to maintain plasma glucose levels at target range (50 to 55 mg/dL). (b) Plasma glucose levels and (c) GIR during the hyperinsulinemic-hypoglycemic clamp. The values represent the average values ± standard error of the mean (SE). MRI A 3-Tesla MRI (Siemens) examination of the brain was performed in each participant for anatomical coregistration with the PET imaging studies (20). PET imaging The synthetic procedures for [11C]MRB have been described previously (16, 21). During the PET scans, participants wore a rigid optical tracking tool, attached to the head with a swim cap, to record head motion with an infrared detector (Vicra, NDI Systems, Waterloo, ON, Canada). A 6-minute transmission scan was acquired after initiation of head-motion recording and just before [11C]MRB injection. List mode data were acquired for 120 minutes and reconstructed with all corrections (attenuation, normalization, scatter, randoms, dead-time, and motion) by using the MOLAR algorithm (22), and a second step of motion correction was applied, as previously described . The BPND values, an index of the number of available binding sites [for review, see Innis et al. (23)], were calculated by using the multilinear reference tissue model (24), with optimizations to reduce noise in parametric images (24) with a reference region (the caudate nucleus) that has lowest levels of NET expression (19). MRI-based region-of-interest definition The regions of interest [ROIs; predetermined NET-rich areas of the brain (25–28), including the locus ceruleus, raphe nuclei, thalamus, and hypothalamus] were defined in template space (29) and were applied to the PET images by using the participant's MR image as an intermediate step, as previously described (16, 17). The paracentral cortical lobe, thalamus, and caudate ROIs were taken from the automated anatomical labeling template (30). The locus ceruleus and raphe nuclei ROIs were drawn on average PET images resliced in template space in previous studies using tracers with high uptake in each of these structures (16). The hypothalamus ROI was manually drawn on the template MRI (31). Biochemical analysis The plasma glucose concentration was determined by the glucose oxidase method (YSI Inc., Yellow Springs, OH). Plasma insulin and glucagon (Millipore, St. Charles, MO), growth hormone (MP Biomedical, Irvine, CA), and cortisol (Diagnostic Products Corp., Los Angeles, CA) were measured by double-antibody radioimmunoassay, and plasma catecholamines were measured by high-performance liquid chromatography (ESA, Chelmsford, MA). Statistical analysis Statistical analyses were performed by using SPSS software, version 19.0 (IBM, Armonk, NY). All values represent the mean ± standard error of the mean. A paired t test was performed to compare hormone levels and the [11C]MRB BPND from the preselected ROIs at baseline and during the hypoglycemic clamp. Pearson correlations were performed between the BPND from these brain areas and hormones levels. Methods for animal studies Animals Studies were performed in male Sprague-Dawley rats (weight ~280 to 300 g; Charles River Labs, Raleigh, NC) that were individually housed in the Yale Animal Resources Center in temperature- (22°C to 23°C) and humidity-controlled rooms. The animals had free access to rat chow (Harlan Teklad, Indianapolis, IN) and water. The animals were acclimatized to handling and a 12-hour light cycle (lights on between 0700 hours and 1900 hours) for 1 week before experimental manipulation. Principles of laboratory animal care were followed, and experimental protocols were approved by the Institutional Animal Care & Use Committee at Yale University. Hypoglycemia Hypoglycemia was induced in the animals with a single intraperitoneal injection of regular human insulin (10 U/kg; Eli Lily, Indianapolis, IN). Blood glucose was monitored every 30 minutes from a tail nick. One hour after insulin administration, when plasma glucose levels had reached ~30 to 40 mg/dL, the animals were euthanized with an overdose of sodium pentobarbital; the brains were rapidly harvested and frozen on dry ice. A second group of animals received a saline injection and were euthanized under similar conditions. Four animals were studied in each set of experiments. Western blots The brains were coronally sectioned on a cryostat, and frozen micropunches were taken through the ventromedial hypothalamus (VMH). The membrane and cytosolic fractions were separated by using a membrane protein extraction kit (ProteoExtract Native Membrane Protein Extraction Kit, Calbiochem®, Merck KGaA, Darmstadt, Germany), and the membrane fraction was confirmed by the presence of cadherin, a protein that is expressed exclusively in membranes. Ten micrograms of protein was loaded onto a 4% to 20% gradient gel and the samples were run at 85 V for 1 to 1.5 hours at room temperature. The protein was subsequently transferred onto a nitrocellulose membrane overnight at 25 V at 4°C. The next day, the membranes were blocked with 5% milk in tris-buffered saline–Tween 20 for 1 hour at room temperature. After blocking, the membrane was incubated in the primary antibody against the NET (Millipore AB2234) overnight at a 1:1000 dilution or β-actin (1:5000). After labeling of the NET, the membranes were stripped and labeled with the β-actin antibody. After several washes, the membrane was incubated in the secondary anti-rabbit antibody (1:5000) or anti-mouse (1:20,000) for 1 hour. After washing, chemiluminescent reagent was applied to the membrane for 2 minutes before the membrane was exposed to film. Relative optical density was quantified by using Scion Image software (Scion Corp., Frederick, MD) and expressed as a ratio against the loading control. Results Hypoglycemic-hyperinsulinemic clamp Human study participants had normal fasting plasma glucose (88 ± 2 mg/dL) and hemoglobin A1c (5.1% ± 0.2%) levels. Before the start of the hypoglycemic-hyperinsulinemic clamp study, plasma insulin levels were 10 ± 2 mU/mL. After initiation of the insulin infusion, glucose levels declined into the hypoglycemic range; mean glucose levels during the last 30 minutes of the PET-hypoglycemic clamp were 57 ± 1 mg/dL (Fig. 1b). The GIR required to keep glucose within the target range averaged 3.5 ± 0.6 mg/kg/min (Fig. 1c). As shown in Table 1, all hormones, except for NE, significantly increased during hypoglycemia (P < 0.05). Plasma epinephrine at the end of the hypoglycemic clamp (120-minute time point) rose nearly ninefold (P < 0.005). Table 1. Hormone Concentrations During the Hypoglycemic-Hyperinsulinemic Clamp Hormone Baseline End of Study Insulin, mU/mL 10 ± 2 104 ± 12a Epinephrine, pg/mL 69 ± 13 483 ± 99a Norepinephrine, pg/mL 377 ± 44 405 ± 45 Glucagon, pg/mL 66 ± 6 96 ± 13b Cortisol, μg/dL 15 ± 5 26 ± 4b Growth hormone, ng/mL 9 ± 4 30 ± 9b Hormone Baseline End of Study Insulin, mU/mL 10 ± 2 104 ± 12a Epinephrine, pg/mL 69 ± 13 483 ± 99a Norepinephrine, pg/mL 377 ± 44 405 ± 45 Glucagon, pg/mL 66 ± 6 96 ± 13b Cortisol, μg/dL 15 ± 5 26 ± 4b Growth hormone, ng/mL 9 ± 4 30 ± 9b Data are expressed as mean ± standard error of the mean. Paired t test analysis comparing baseline vs end of the hyperinsulinemic-hypoglycemic clamp study (average, 90 to 120 minutes). a P < 0.005. b P < 0.05. View Large Table 1. Hormone Concentrations During the Hypoglycemic-Hyperinsulinemic Clamp Hormone Baseline End of Study Insulin, mU/mL 10 ± 2 104 ± 12a Epinephrine, pg/mL 69 ± 13 483 ± 99a Norepinephrine, pg/mL 377 ± 44 405 ± 45 Glucagon, pg/mL 66 ± 6 96 ± 13b Cortisol, μg/dL 15 ± 5 26 ± 4b Growth hormone, ng/mL 9 ± 4 30 ± 9b Hormone Baseline End of Study Insulin, mU/mL 10 ± 2 104 ± 12a Epinephrine, pg/mL 69 ± 13 483 ± 99a Norepinephrine, pg/mL 377 ± 44 405 ± 45 Glucagon, pg/mL 66 ± 6 96 ± 13b Cortisol, μg/dL 15 ± 5 26 ± 4b Growth hormone, ng/mL 9 ± 4 30 ± 9b Data are expressed as mean ± standard error of the mean. Paired t test analysis comparing baseline vs end of the hyperinsulinemic-hypoglycemic clamp study (average, 90 to 120 minutes). a P < 0.005. b P < 0.05. View Large PET imaging The average injected dose of [11C]MRB per scan was 642 ± 104 MBq (0.58 to 3.47 µg). [11C]MRB BPND was measured in NET-rich areas of the brain, including locus ceruleus, raphe nuclei, thalamus, hypothalamus, and paracentral lobule (Fig. 2a). During hypoglycemia, in comparison with the baseline scan, [11C]MRB BPND decreased by 24% ± 12% in the raphe nuclei (P < 0.01). No statistically significant changes in MRB binding were observed in the other NET-rich areas measured with PET imaging (Fig. 2b). Figure 2. View largeDownload slide (a) PET image at baseline and during hypoglycemia. The first column represents the MRI template corresponding to the anatomical sagittal (first row), coronal (second row), and axial (third row) slices of the PET images. The second and third columns represent, respectively, the mean [11C] MRB PET image at baseline and during the hyperinsulinemic-hypoglycemic clamp. (b) [11C]MRB BPND in the NE-rich areas of the brain identified with the PET scan at baseline and during the hyperinsulinemic-hypoglycemic clamp study. The caudate served as a reference region. Data are expressed as mean ± standard error of the mean. *P < 0.01 (statistically significant difference between baseline and the hyperinsulinemic-hypoglycemic clamp study). Hypothal, hypothalamus; LC, locus ceruleus; PL, paracentral lobule; RN, raphe nuclei. Figure 2. View largeDownload slide (a) PET image at baseline and during hypoglycemia. The first column represents the MRI template corresponding to the anatomical sagittal (first row), coronal (second row), and axial (third row) slices of the PET images. The second and third columns represent, respectively, the mean [11C] MRB PET image at baseline and during the hyperinsulinemic-hypoglycemic clamp. (b) [11C]MRB BPND in the NE-rich areas of the brain identified with the PET scan at baseline and during the hyperinsulinemic-hypoglycemic clamp study. The caudate served as a reference region. Data are expressed as mean ± standard error of the mean. *P < 0.01 (statistically significant difference between baseline and the hyperinsulinemic-hypoglycemic clamp study). Hypothal, hypothalamus; LC, locus ceruleus; PL, paracentral lobule; RN, raphe nuclei. Correlations The hypothalamus was the only NET-rich brain region in which changes in [11C]MRB BPND positively correlated with changes in epinephrine (r = 0.723; P = 0.028) and glucagon levels (r = 0.670; P = 0.048), and inversely correlated with GIR (r = −0.805; P = 0.009) (Fig. 3). In contrast, the changes in [11C]MRB BPND in the raphe nuclei during the two PET scans (hypoglycemic-clamp scan BPND minus baseline scan BPND) did not correlate with GIR or with increments in the CR hormones glucagon, epinephrine, growth hormone, and cortisol; although the binding response in the raphe nuclei negatively correlated with changes in NE levels (hypoglycemia at 90 to 120 minutes minus baseline levels, r = −0.678; P = 0.045). Figure 3. View largeDownload slide Scatterplots correlating the changes in [11C]MRB BPND in the hypothalamus (hypoglycemic-clamp scan BPND minus baseline scan BPND) with (a) average GIR in the last 30 minutes of the hypoglycemic clamp (90 to 120 minutes) (P = 0.009), (b) change in epinephrine levels (average, 90 to 120 minutes of the hypoglycemic clamp minus baseline; P = 0.028), and (c) change in glucagon levels (average, 90 to 120 minutes minus baseline; P = 0.048). Figure 3. View largeDownload slide Scatterplots correlating the changes in [11C]MRB BPND in the hypothalamus (hypoglycemic-clamp scan BPND minus baseline scan BPND) with (a) average GIR in the last 30 minutes of the hypoglycemic clamp (90 to 120 minutes) (P = 0.009), (b) change in epinephrine levels (average, 90 to 120 minutes of the hypoglycemic clamp minus baseline; P = 0.028), and (c) change in glucagon levels (average, 90 to 120 minutes minus baseline; P = 0.048). To further explore these correlations between the changes in NET-binding in the hypothalamus and CR responses, we determined that the changes in [11C]MRB BPND in this region, in contrast to the raphe, were centered around zero for the average CR response, with no absolute changes in [11C]MRB BPND during hypoglycemia and baseline. On the basis of these findings, we divided the study participants into two groups according to their changes in [11C]MRB BPND in the hypothalamus (hypoglycemia minus baseline scan). Figure 4 shows that participants (n = 5) with a decrease in NET binding (increased NA activity) had a low peripheral CR hormonal response (epinephrine and glucagon) and a correspondingly elevated GIR. In contrast, when NET binding increased (low NA activity), the participants (n = 4) showed more robust CR hormonal responses and a correspondingly low requirement for exogenous glucose (P < 0.05 for the comparison of the groups by increase vs decrease in [11C]MRB BPND in the hypothalamus). There were no statistically significant differences in age, body mass index, hemoglobin A1c, glucose, and insulin levels between the two groups (all P = not significant), indicating that NA activity in the hypothalamus during insulin-induced hypoglycemia was not regulated directly by blood glucose levels but by the magnitude of the CR hormonal response. Figure 4. View largeDownload slide Participants were divided into two groups based on the changes in [11C]MRB BPND in the hypothalamus (hypoglycemia baseline): increase/unchanged (n = 4) and decrease (n = 5). (a) The change in hypothalamus was calculated by subtracting [11C]MRB BPND during the hyperinsulinemic-hypoglycemic clamp minus baseline. (b) Plasma glucose levels, (c) GIR, (d) plasma glucagon levels, (e) and plasma epinephrine levels during the hyperinsulinemic-hypoglycemic clamp are shown as divided into groups by the changes in [11C]MRB BPND in the hypothalamus. *P < 0.05 (statistically significant difference between the increase and decrease groups). Data are expressed as mean ± standard error of the mean (SE). Figure 4. View largeDownload slide Participants were divided into two groups based on the changes in [11C]MRB BPND in the hypothalamus (hypoglycemia baseline): increase/unchanged (n = 4) and decrease (n = 5). (a) The change in hypothalamus was calculated by subtracting [11C]MRB BPND during the hyperinsulinemic-hypoglycemic clamp minus baseline. (b) Plasma glucose levels, (c) GIR, (d) plasma glucagon levels, (e) and plasma epinephrine levels during the hyperinsulinemic-hypoglycemic clamp are shown as divided into groups by the changes in [11C]MRB BPND in the hypothalamus. *P < 0.05 (statistically significant difference between the increase and decrease groups). Data are expressed as mean ± standard error of the mean (SE). Animal studies Cytosolic and plasma membrane NET protein levels in the VMH-projecting NA neurons of normal hypoglycemia-naive rats were compared under baseline (euglycemic) conditions and after an acute bout of hypoglycemia. Under euglycemic conditions, there was no significant difference between NET protein levels in the cytosolic and membrane fractions. However, after the induction of hypoglycemia, NET protein levels in the membrane fraction were significantly higher than those in the cytosolic fraction (Fig. 5), suggesting that hypoglycemia may have induced translocation of NET from the cytosolic pool to the plasma membrane. Figure 5. View largeDownload slide (Top) Representative immunoblot showing cytosolic (C) and membrane (M) NET protein levels in the VMH. (Bottom) Densitometric analysis of immunoblots for NET in the VMH expressed as a ratio of NET to β-actin under baseline conditions (black bar) and in response to hypoglycemia (striped bar). *P < 0.05 (statistically significant difference comparing cytosolic vs membrane protein levels during hypoglycemia). R.O.D., relative optical density. Figure 5. View largeDownload slide (Top) Representative immunoblot showing cytosolic (C) and membrane (M) NET protein levels in the VMH. (Bottom) Densitometric analysis of immunoblots for NET in the VMH expressed as a ratio of NET to β-actin under baseline conditions (black bar) and in response to hypoglycemia (striped bar). *P < 0.05 (statistically significant difference comparing cytosolic vs membrane protein levels during hypoglycemia). R.O.D., relative optical density. Discussion The current study demonstrates that the PET radioligand [11C]MRB has the ability to identify the NET-rich areas of the human brain (25–28)—the locus ceruleus, raphe nuclei, thalamus, hypothalamus, and paracentral lobule—both at baseline and during hypoglycemia. Of particular interest, these human experiments demonstrate that NET binding in the raphe nuclei was reduced by 24% during hypoglycemia, thus providing evidence that NA system activation in this specific CNS region likely contributes to the in vivo CR response to hypoglycemia in humans. It is also noteworthy that it was not the raphe nuclei but instead the NET-binding changes in the hypothalamus that was associated with the magnitude of both the peripheral CR hormonal and metabolic responses to hypoglycemia. Moreover, individuals with poor CR responses showed the greatest NA activation in the hypothalamus. To evaluate the CNS NA system response in vivo during hypoglycemia in humans, we used [11C]MRB, a highly selective NET-ligand radiotracer for PET imaging. NET is a transmembrane transporter that plays an important role in regulating NA neurotransmission (15). This radiotracer is capable of imaging NET availability in living systems (16, 17). In our study, [11C]MRB-PET imaging not only identified the NET-rich areas of the brain but also provided a means to measure acute changes in NET occupancy. This tracer thus holds promise as a clinical tool to evaluate the CNS NA system in humans. Unexpectedly, we identified the raphe nuclei as being responsive to hypoglycemia. The raphe nuclei are located in the brainstem and receive inputs from the locus coeruleus via NA neurons. The raphe nuclei consist predominantly of serotonergic neurons that project to the entire CNS and serve as the main source of serotonin for the brain (32). These neurons are involved in modulating pain sensation, motor activity, sleep-wake cycles, mood disorders, circadian rhythm, and food intake (33–35). Although not widely studied, insulin-induced hypoglycemia in cats increases activity in NA neurons in the locus ceruleus (36) and decreases serotoninergic neuronal activity in medullary (37), but not in dorsal (38), raphe nuclei. In these animal studies, insulin-induced hypoglycemia promoted a CR hormonal response, which was accompanied by a decrease in neuronal firing in the medullary raphe nuclei and a decrease in muscle tone (37). In humans, fluoxetine, a selective serotonin reuptake inhibitor increases autonomic as well as motor CR responses in healthy participants and in patients with type 1 diabetes mellitus (39, 40). Although MRB PET imaging cannot determine the effect of hypoglycemia within distinct raphe nuclei, these findings suggest that increased NA activity in raphe nuclei may promote increased autonomic output, decreased motor activity, increased stress response, and hunger, all of which may be important for recognizing hypoglycemia in humans. The significant decrement observed in [11C]MRB BPND in the raphe nuclei during hypoglycemia could be due to reduced NET availability in the presynaptic membrane of the NA neuron and/or increased competition caused by endogenously released NE. The [11C]MRB tracer binds to NET and at baseline [11C]MRB BPND corresponds to NET availability in the presynaptic membrane of the NA neuron. On the basis of a prior study showing that increased NET occupancy results in decreased [11C]MRB-binding (17), we postulate that during hypoglycemia, activated NA neurons release NE in the synapse, thereby effectively competing with the [11C]MRB ligand for NET binding and in turn decreasing [11C]MRB BPND. The rodent studies demonstrating greater membrane NET levels compared with cytosolic levels following the induction of hypoglycemia supports the latter conclusion. During hypoglycemia, activated NA neurons released NE in the synapse. NET is then translocated to the plasma membrane to take up and repackage NE for release. This increase in NET trafficking is consistent with the observed decrease in [11C]MRB BPND in our human studies. However, further studies will be needed to precisely define how hypoglycemia affects NET and [11C]MRB BPND. Furthermore, we also cannot exclude a direct effect of insulin on NET. Long-term insulin treatment has been reported to decrease NET messenger RNA levels in the locus ceruleus of the rat (41). Thus, human studies using the euglycemic-hyperinsulinemic clamp technique will be required to differentiate the effects of hypoglycemia and hyperinsulinemia per se on the central NA system. As expected, hypoglycemia induced the release of CR hormones in the periphery. The individual changes in NET binding in the hypothalamus during hypoglycemia positively correlated with hormonal changes as well as negatively correlated with GIR (Fig. 3). These results highlight that NET binding in the hypothalamus appears titrated, rather than a simple uniform directional response (as observed in the raphe nuclei). When participants successfully mounted a large hormonal CR response to hypoglycemia, [11C]MRB BPND in the hypothalamus was unchanged or increased (consistent with low local NA activation). In contrast, participants who appeared to have an unsuccessful CR glucose response (low CR hormones and high GIR) presented with decreased NET binding (i.e., elevated NA activation) (Figs. 3 and 4). These results raise the question of whether the NA system could be a mechanism involved in regulating hypothalamic activity in response to the magnitude of the CR hormonal response. The hypothalamus has been shown in animal studies to be the central coordinator of the CR response to hypoglycemia (2), which may, at least in part, be mediated by local release of NE (8–10) from NA neurons. However, in our human studies, [11C]MRB BPND in the hypothalamus was not statistically different when we compared the hypoglycemic-clamp and baseline scans, because of CR responders and nonresponders. Animal studies have demonstrated that systemic insulin-induced hypoglycemia increases extracellular NE in the VMH and paraventricular nucleus, but not in the lateral hypothalamus nucleus (8). [11C]MRB PET imaging determines NET binding by anatomical brain regions. Therefore, we cannot exclude that distinct changes in NA neuronal activity within specific hypothalamic nuclei may have occurred. In addition, although the [11C]MRB tracer can accurately determine changes in NET occupancy under different conditions (17), the low signal-to-noise ratio observed with this tracer (28) may have limited our ability in identifying minor changes in NET binding. However, despite that, we were able to correctly localize the NET-rich regions of the brain and that hypoglycemia alters NA activity in the raphe nuclei. Another potential limitation of this tracer is that it requires ~2 hours of imaging acquisition, limiting our ability to evaluate the time course of NET-binding throughout the hypoglycemic clamps, which could have helped further establish the interplay between the CNS NA system and CR. In summary, [11C]MRB, a selective NET-ligand radiotracer for PET imaging, was shown to be an effective tool for measuring the complex brain NA system activity induced by acute hypoglycemia in humans. We showed that humans experiencing hypoglycemia had increased NA activation in the raphe nuclei, a region that has the capacity to influence the individual’s autonomic and symptomatic responses to hypoglycemia. In contrast, hypothalamic NET-binding changes under hypoglycemia revealed a pattern across a wide range of individual CR responses. The largest trigger for NA activation in healthy humans was observed in those with poor CR responses. Taken together, these findings suggest that although the CNS NA system is activated, the hypothalamus may titrate its response in healthy humans. Thus, a hypoglycemic clamp combined with a PET imaging approach using [11C]MRB may provide insights into the potential role of the CNS NA system in the development of hypoglycemia unawareness in insulin-treated patients with diabetes. Abbreviations: Abbreviations: [11C]MRB (S,S)-[11C]O-methylreboxetine BPND binding potential CNS central nervous system CR counterregulatory GIR glucose infusion rate IV intravenous MRB methylreboxetine NA noradrenergic NE norepinephrine NET norepinephrine transporter PET positron emission tomography ROI region of interest VMH ventromedial hypothalamus Acknowledgments We thank Christian Schmidt, Ralph Jacob, Mikhail Smolgovsky, Irene Chernyak, and Codruta Todeasa for their technical assistance, the staff of the Yale PET Center and the Hospital Research Unit at the Yale New Haven Hospital, and the volunteers who participated in this study. Financial Support: This work was supported by in part by National Institute of Diabetes and Digestive and Kidney Diseases (grants R01 DK20495 and T32 DK 07058; R.S.S.), the Diabetes Research Center (grant P30 DK045735; R.S.S.), and the Yale Center for Clinical Investigation supported by the Clinical and Translational Science Award (grant UL1 TR001863 from the National Center for Advancing Translational Science; R.S.S.). Clinical Trial Information: ClinicalTrials.gov no. NCT02056249 (registered 3 February 2014). Disclosure Summary: The authors have nothing to disclose. References 1. The Diabetes Control and Complications Trial Research Group . Hypoglycemia in the Diabetes Control and Complications Trial . Diabetes . 1997 ; 46 ( 2 ): 271 – 286 . CrossRef Search ADS PubMed 2. Chan O , Sherwin R . Influence of VMH fuel sensing on hypoglycemic responses . Trends Endocrinol Metab . 2013 ; 24 ( 12 ): 616 – 624 . Google Scholar CrossRef Search ADS PubMed 3. Mitrakou A , Ryan C , Veneman T , Mokan M , Jenssen T , Kiss I , Durrant J , Cryer P , Gerich J . Hierarchy of glycemic thresholds for counterregulatory hormone secretion, symptoms, and cerebral dysfunction . Am J Physiol . 1991 ; 260 ( 1 Pt 1 ): E67 – E74 . Google Scholar PubMed 4. Biggers DW , Myers SR , Neal D , Stinson R , Cooper NB , Jaspan JB , Williams PE , Cherrington AD , Frizzell RT . Role of brain in counterregulation of insulin-induced hypoglycemia in dogs . Diabetes . 1989 ; 38 ( 1 ): 7 – 16 . Google Scholar CrossRef Search ADS PubMed 5. Borg WP , During MJ , Sherwin RS , Borg MA , Brines ML , Shulman GI . Ventromedial hypothalamic lesions in rats suppress counterregulatory responses to hypoglycemia . J Clin Invest . 1994 ; 93 ( 4 ): 1677 – 1682 . Google Scholar CrossRef Search ADS PubMed 6. Borg WP , Sherwin RS , During MJ , Borg MA , Shulman GI . Local ventromedial hypothalamus glucopenia triggers counterregulatory hormone release . Diabetes . 1995 ; 44 ( 2 ): 180 – 184 . Google Scholar CrossRef Search ADS PubMed 7. Borg MA , Sherwin RS , Borg WP , Tamborlane WV , Shulman GI . Local ventromedial hypothalamus glucose perfusion blocks counterregulation during systemic hypoglycemia in awake rats . J Clin Invest . 1997 ; 99 ( 2 ): 361 – 365 . Google Scholar CrossRef Search ADS PubMed 8. Beverly JL , De Vries MG , Bouman SD , Arseneau LM . Noradrenergic and GABAergic systems in the medial hypothalamus are activated during hypoglycemia . Am J Physiol Regul Integr Comp Physiol . 2001 ; 280 ( 2 ): R563 – R569 . Google Scholar CrossRef Search ADS PubMed 9. Beverly JL , de Vries MG , Beverly MF , Arseneau LM . Norepinephrine mediates glucoprivic-induced increase in GABA in the ventromedial hypothalamus of rats . Am J Physiol Regul Integr Comp Physiol . 2000 ; 279 ( 3 ): R990 – R996 . Google Scholar CrossRef Search ADS PubMed 10. de Vries MG , Lawson MA , Beverly JL . Hypoglycemia-induced noradrenergic activation in the VMH is a result of decreased ambient glucose . Am J Physiol Regul Integr Comp Physiol . 2005 ; 289 ( 4 ): R977 – R981 . Google Scholar CrossRef Search ADS PubMed 11. Figlewicz DP , Van Dijk G , Wilkinson CW , Gronbeck P , Higgins M , Zavosh A . Effects of repetitive hypoglycemia on neuroendocrine response and brain tyrosine hydroxylase activity in the rat . Stress . 2002 ; 5 ( 3 ): 217 – 226 . Google Scholar CrossRef Search ADS PubMed 12. Chafetz MD , Parko K , Diaz S , Leibowitz SF . Relationships between medial hypothalamic alpha 2-receptor binding, norepinephrine, and circulating glucose . Brain Res . 1986 ; 384 ( 2 ): 404 – 408 . Google Scholar CrossRef Search ADS PubMed 13. Livett BG . Histochemical visualization of peripheral and central adrenergic neurones . Br Med Bull . 1973 ; 29 ( 2 ): 93 – 99 . Google Scholar CrossRef Search ADS PubMed 14. Szabadi E . Functional neuroanatomy of the central noradrenergic system [ published correction appears in J Psychopharmacol. 2013;27(10):964 ]. J Psychopharmacol . 2013 ; 27 ( 8 ): 659 – 693 . Google Scholar CrossRef Search ADS PubMed 15. Mandela P , Ordway GA . The norepinephrine transporter and its regulation . J Neurochem . 2006 ; 97 ( 2 ): 310 – 333 . Google Scholar CrossRef Search ADS PubMed 16. Ding YS , Singhal T , Planeta-Wilson B , Gallezot JD , Nabulsi N , Labaree D , Ropchan J , Henry S , Williams W , Carson RE , Neumeister A , Malison RT . PET imaging of the effects of age and cocaine on the norepinephrine transporter in the human brain using (S,S)-[(11)C]O-methylreboxetine and HRRT . Synapse . 2010 ; 64 ( 1 ): 30 – 38 . Google Scholar CrossRef Search ADS PubMed 17. Hannestad J , Gallezot JD , Planeta-Wilson B , Lin SF , Williams WA , van Dyck CH , Malison RT , Carson RE , Ding YS . Clinically relevant doses of methylphenidate significantly occupy norepinephrine transporters in humans in vivo . Biol Psychiatry . 2010 ; 68 ( 9 ): 854 – 860 . Google Scholar CrossRef Search ADS PubMed 18. Lin SF , Fan X , Yeckel CW , Weinzimmer D , Mulnix T , Gallezot JD , Carson RE , Sherwin RS , Ding YS . Ex vivo and in vivo evaluation of the norepinephrine transporter ligand [11C]MRB for brown adipose tissue imaging . Nucl Med Biol . 2012 ; 39 ( 7 ): 1081 – 1086 . Google Scholar CrossRef Search ADS PubMed 19. Gallezot JD , Weinzimmer D , Nabulsi N , Lin SF , Fowles K , Sandiego C , McCarthy TJ , Maguire RP , Carson RE , Ding YS . Evaluation of [(11)C]MRB for assessment of occupancy of norepinephrine transporters: Studies with atomoxetine in non-human primates . Neuroimage . 2011 ; 56 ( 1 ): 268 – 279 . Google Scholar CrossRef Search ADS PubMed 20. Naganawa M , Jacobsen LK , Zheng M-Q , Lin S-F , Banerjee A , Byon W , Weinzimmer D , Tomasi G , Nabulsi N , Grimwood S , Badura LL , Carson RE , McCarthy TJ , Huang Y . Evaluation of the agonist PET radioligand [11C]GR103545 to image kappa opioid receptor in humans: kinetic model selection, test-retest reproducibility and receptor occupancy by the antagonist PF-04455242 . Neuroimage . 2014 ; 99 : 69 – 79 . Google Scholar CrossRef Search ADS PubMed 21. Ding YS , Lin KS , Garza V , Carter P , Alexoff D , Logan J , Shea C , Xu Y , King P . Evaluation of a new norepinephrine transporter PET ligand in baboons, both in brain and peripheral organs . Synapse . 2003 ; 50 ( 4 ): 345 – 352 . Google Scholar CrossRef Search ADS PubMed 22. Carson RE , Barker W , Liow J-S , Johnson C . Design of a motion-compensation OSEM list-mode algorithm for resolution-recovery reconstruction for the HRRT. Nuclear Science Symposium Conference Record, 2003 IEEE 2003 ;5: 3281 – 3285 . 23. Innis RB , Cunningham VJ , Delforge J , Fujita M , Gjedde A , Gunn RN , Holden J , Houle S , Huang S-C , Ichise M , Iida H , Ito H , Kimura Y , Koeppe RA , Knudsen GM , Knuuti J , Lammertsma AA , Laruelle M , Logan J , Maguire RP , Mintun MA , Morris ED , Parsey R , Price JC , Slifstein M , Sossi V , Suhara T , Votaw JR , Wong DF , Carson RE . Consensus nomenclature for in vivo imaging of reversibly binding radioligands . J Cereb Blood Flow Metab . 2007 ; 27 ( 9 ): 1533 – 1539 . Google Scholar CrossRef Search ADS PubMed 24. Ichise M , Liow J-S , Lu J-Q , Takano A , Model K , Toyama H , Suhara T , Suzuki K , Innis RB , Carson RE . Linearized reference tissue parametric imaging methods: application to [11C]DASB positron emission tomography studies of the serotonin transporter in human brain . J Cereb Blood Flow Metab . 2003 ; 23 ( 9 ): 1096 – 1112 . Google Scholar CrossRef Search ADS PubMed 25. Gross-Isseroff R , Israeli M , Biegon A . Autoradiographic analysis of [3H]desmethylimipramine binding in the human brain postmortem . Brain Res . 1988 ; 456 ( 1 ): 120 – 126 . Google Scholar CrossRef Search ADS PubMed 26. Donnan GA , Kaczmarczyk SJ , Paxinos G , Chilco PJ , Kalnins RM , Woodhouse DG , Mendelsohn FA . Distribution of catecholamine uptake sites in human brain as determined by quantitative [3H] mazindol autoradiography . J Comp Neurol . 1991 ; 304 ( 3 ): 419 – 434 . Google Scholar CrossRef Search ADS PubMed 27. Ordway GA , Stockmeier CA , Cason GW , Klimek V . Pharmacology and distribution of norepinephrine transporters in the human locus coeruleus and raphe nuclei . J Neurosci . 1997 ; 17 ( 5 ): 1710 – 1719 . Google Scholar CrossRef Search ADS PubMed 28. Logan J , Wang GJ , Telang F , Fowler JS , Alexoff D , Zabroski J , Jayne M , Hubbard B , King P , Carter P , Shea C , Xu Y , Muench L , Schlyer D , Learned-Coughlin S , Cosson V , Volkow ND , Ding YS . Imaging the norepinephrine transporter in humans with (S,S)-[11C]O-methyl reboxetine and PET: problems and progress . Nucl Med Biol . 2007 ; 34 ( 6 ): 667 – 679 . Google Scholar CrossRef Search ADS PubMed 29. Collins DL , Zijdenbos AP , Kollokian V , Sled JG , Kabani NJ , Holmes CJ , Evans AC . Design and construction of a realistic digital brain phantom . IEEE Trans Med Imaging . 1998 ; 17 ( 3 ): 463 – 468 . Google Scholar CrossRef Search ADS PubMed 30. Tzourio-Mazoyer N , Landeau B , Papathanassiou D , Crivello F , Etard O , Delcroix N , Mazoyer B , Joliot M . Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain . Neuroimage . 2002 ; 15 ( 1 ): 273 – 289 . Google Scholar CrossRef Search ADS PubMed 31. Page KA , Arora J , Qiu M , Relwani R , Constable RT , Sherwin RS . Small decrements in systemic glucose provoke increases in hypothalamic blood flow prior to the release of counterregulatory hormones . Diabetes . 2009 ; 58 ( 2 ): 448 – 452 . Google Scholar CrossRef Search ADS PubMed 32. Mancall EL . Brain stem . In: Mancall EL , Brock DJ , eds. Gray's Clinical Neuroanatomy: The Anatomic Basis for Clinical Neuroscience . Philadelphia : Saunders Elsevier ; 2011 : 157 – 184 . 33. Hornung JP . The human raphe nuclei and the serotonergic system . J Chem Neuroanat . 2003 ; 26 ( 4 ): 331 – 343 . Google Scholar CrossRef Search ADS PubMed 34. Deurveilher S , Semba K . Reciprocal connections between the suprachiasmatic nucleus and the midbrain raphe nuclei: a putative role in the circadian control of behavioral states. In: Monti JM , Pandi-Perumal SR , Jacobs BL , Nutt DJ , eds. Serotonin and Sleep: Molecular, Functional and Clinical Aspects . Basel, Switzerland : Birkhäuser ; 2008 . https://doi.org/10.1007/978-3-7643-8561-3_4. 35. Carlini VP , Varas MM , Cragnolini AB , Schiöth HB , Scimonelli TN , de Barioglio SR . Differential role of the hippocampus, amygdala, and dorsal raphe nucleus in regulating feeding, memory, and anxiety-like behavioral responses to ghrelin . Biochem Biophys Res Commun . 2004 ; 313 ( 3 ): 635 – 641 . Google Scholar CrossRef Search ADS PubMed 36. Morilak DA , Fornal CA , Jacobs BL . Effects of physiological manipulations on locus coeruleus neuronal activity in freely moving cats. III. Glucoregulatory challenge . Brain Res . 1987 ; 422 ( 1 ): 32 – 39 . Google Scholar CrossRef Search ADS PubMed 37. Martín-Cora FJ , Fornal CA , Metzler CW , Jacobs BL . Insulin-induced hypoglycemia decreases single-unit activity of serotonergic medullary raphe neurons in freely moving cats: relationship to sympathetic and motor output . Eur J Neurosci . 2002 ; 16 ( 4 ): 722 – 734 . Google Scholar CrossRef Search ADS PubMed 38. Fornal CA , Litto WJ , Morilak DA , Jacobs BL . Single-unit responses of serotonergic neurons to glucose and insulin administration in behaving cats . Am J Physiol . 1989 ; 257 ( 6 Pt 2 ): R1345 – R1353 . Google Scholar PubMed 39. Briscoe VJ , Ertl AC , Tate DB , Dawling S , Davis SN . Effects of a selective serotonin reuptake inhibitor, fluoxetine, on counterregulatory responses to hypoglycemia in healthy individuals . Diabetes . 2008 ; 57 ( 9 ): 2453 – 2460 . Google Scholar CrossRef Search ADS PubMed 40. Briscoe VJ , Ertl AC , Tate DB , Davis SN . Effects of the selective serotonin reuptake inhibitor fluoxetine on counterregulatory responses to hypoglycemia in individuals with type 1 diabetes . Diabetes . 2008 ; 57 ( 12 ): 3315 – 3322 . Google Scholar CrossRef Search ADS PubMed 41. Figlewicz DP , Szot P , Israel PA , Payne C , Dorsa DM . Insulin reduces norepinephrine transporter mRNA in vivo in rat locus coeruleus . Brain Res . 1993 ; 602 ( 1 ): 161 – 164 . Google Scholar CrossRef Search ADS PubMed Copyright © 2018 Endocrine Society
Journal of Clinical Endocrinology and Metabolism – Oxford University Press
Published: Mar 23, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera