| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology, Diabetes and Metabolism, Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts 02215
Address all correspondence and requests for reprints to: Barbara B. Kahn, M.D., Division of Endocrinology, Diabetes and Metabolism, Beth Israel Deaconess Medical Center, 99 Brookline Avenue, Boston, Massachusetts 02215. E-mail: bkahn{at}bidmc.harvard.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1- and 2-AMPK activities were increased 30–50% in arcuate and ventromedial/dorsomedial hypothalamus but not in other hypothalamic regions, hindbrain, or cortex. Increased AMPK activity persisted in arcuate hypothalamus at 60 min after 2DG injection when serum glucagon and corticosterone levels were increased 2.5- to 3.4-fold. When 2DG was injected icv daily for 4 d, hypothalamic 1- and 2-AMPK responses were markedly blunted in arcuate hypothalamus, and 1-AMPK was also blunted in mediobasal hypothalamus 10 min after 2DG on d 4. Both AMPK isoforms were activated normally in arcuate hypothalamus at 60 min. Counterregulatory hormone responses were impaired by recurrent neuroglucopenia and were partially restored by icv injection of 5-aminoimidazole-4-carboxamide 1-ß-D-ribofuranoside, an AMPK activator, before 2DG. Glycogen content increased 2-fold in hypothalamus after recurrent neuroglucopenia, suggesting that glycogen supercompensation could be involved in down-regulating the AMPK glucose-sensing pathway in HAAF. Thus, activation of hypothalamic AMPK may be important for the full counterregulatory hormone response to neuroglucopenia. Furthermore, impaired or delayed AMPK activation in specific hypothalamic regions may play a critical role in the etiology of HAAF. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several studies have established the involvement of the glucose transporter GLUT2 (8, 9), glucokinase (10), and KATP channels (11, 12) in the glucose-sensing mechanisms in the hypothalamus and the HB. However, their potential role in HAAF has not been elucidated. Gruetter and colleagues (13) recently proposed that brain glycogen metabolism could be involved in HAAF. Indeed, this study demonstrated that a single episode of insulin-induced hypoglycemia depletes brain glycogen stores. However, brain glycogen content rebounds to levels that exceed the pre-hypoglycemic concentrations several hours after the hypoglycemia. Thus, it was postulated that brain glycogen serves as an energy store during hypoglycemia and that glycogen supercompensation may participate in HAAF (14). However, this hypothesis has not been proven.
The AMP-activated protein kinase (AMPK) is an evolutionarily conserved enzyme that senses the energy status of cells and regulates fuel availability. Recently, AMPK has been proposed to act as a nutrient and glucose sensor in the hypothalamus (15, 16). AMPK is a heterotrimeric protein consisting of catalytic - and regulatory ß- and 
-subunits that is activated allosterically by increases in the AMP/ATP ratio as well as by phosphorylation on Thr172 by upstream kinases (16). Recently, two upstream AMPK kinases have been identified, LKB1 and Ca2+/calmodulin-dependent protein kinase kinase (CaMKK) (17, 18). AMPK is expressed throughout the brain including in several areas controlling food intake and neuroendocrine function (i.e. hypothalamus and HB).
Several studies have demonstrated that hypothalamic AMPK is regulated by blood glucose levels. Peripheral or central hyperglycemia inhibits AMPK in several hypothalamic nuclei, the arcuate nucleus (ARC), the mediobasal hypothalamus (MH), the paraventricular nucleus (PVN), and the lateral hypothalamus (LH) (19, 20), whereas hypoglycemia induced by insulin or 2DG administration activates AMPK (15, 19). Inhibition of hypothalamic AMPK activation by compound C or dominant negative AMPK adenoviruses impaired secretion of counterregulatory hormones, suggesting that AMPK activation during insulin-induced hypoglycemia may be required for a normal counterregulatory hormone response (15). Also, hypothalamic infusion of the pharmacological activator of AMPK, 5-aminoimidazole-4-carboxamide 1-ß-D-ribofuranoside (AICAR), amplifies the counterregulatory response in a rat model of HAAF induced by repetitive insulin-induced hypoglycemia (21). However, impairment of hypothalamic AMPK activation during HAAF has not been demonstrated yet. Altogether, these data demonstrate that in addition to sensing hexose deprivation in lower organisms, AMPK also acts as a glucose sensor in hypothalamus in mammals and is involved in the subsequent counterregulatory responses to hypoglycemia.
The primary goal of the present study was to assess whether HAAF is associated with impaired activation of AMPK in the hypothalamus and HB. We first investigated the normal pattern of AMPK activation in different brain loci after a single episode of 2DG-induced neuroglucopenia. Then, using repetitive central 2DG injection, we created and validated a HAAF rat model in which we demonstrated impaired activation of AMPK by 2DG in several hypothalamic nuclei. We finally investigated whether AICAR could reverse the HAAF and whether hypothalamic glycogen content was altered by repetitive neuroglucopenia.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Lateral ventricle cannulation
Rats were implanted with a 26-gauge stainless steel cannula (Plastics One, Roanoke, VA) aimed at the right lateral ventricle. Under pentobarbital (60 mg/kg body weight) anesthesia, rats were placed in a Kopf stereotaxic apparatus. Using a sterile technique, the dorsal surface of the skull was exposed, and a hole was drilled to yield an implantation site 0.9 mm caudal to the bregma, 1.5 mm lateral to the midline, and 3.4 mm ventral to the dura according to the atlas of Paxinos and Watson (22). Subsequently, the guide cannula was implanted. To prevent cannula occlusion, a stainless steel stylet of equal length was inserted before cannula implantation. The cannula was secured in place with cranioplast cement and four screws anchored into the skull around the cannula. Only rats with cerebrospinal fluid spontaneously rising after obturator removal were used for intracerebroventricular (icv) injections. The rats were allowed to recover for 10 d after the surgery and were handled and habituated to the testing environment and procedures every day to minimize stress.
The icv injections
The icv injections of 7 mg (40 µmol) of 2DG, at a speed of 10 µl in 3 min using microdialysis pumps, were made using a 31-gauge injector (equal length of the cannula). The injector was kept in place for an additional minute before it was removed and replaced by the dummy cannula. The dose of 2DG was determined from previously published studies (23, 24). The dose of AICAR (Toronto Research Chemical, North York, Ontario, Canada; 500 nmol) was also determined from a previous study (15) and was injected using the same protocol as 2DG.
Experiment 1: time course effect of an icv 2DG injection on counterregulatory response and AMPK activity in hypothalamus
In this experiment, fed rats were injected at 1200 h with saline or 2DG (n = 7–13 in each group). Rats were then killed by decapitation 10 or 60 min after the icv injection. Trunk blood was collected in tubes containing aprotinin (0.1 mg) for serum separation. After centrifugation, serum was stored at –40 C until assayed for corticosterone and glucagon. Immediately after the decapitation, hypothalamic nuclei, HB, and cortex (Cx) were dissected as described (20) and frozen in liquid nitrogen for AMPK activity assay. The dissection of the HB was performed as follows. After the brain was removed from the skull, the cerebellum was removed to expose the fourth ventricle, the HB, and the spinal cord. A first cut was performed at the anterior part of the fourth ventricle (
11 mm posterior to bregma) and a second cut was performed at the posterior part of the fourth ventricle (14 mm posterior to bregma). The mediodorsal part was dissected from this coronal section (2 mm thick in the dorsoventral axis and 4 mm wide).
Experiment 2: effect of recurrent icv 2DG injections on hypothalamic AMPK activity
In this experiment, rats were assigned to three different groups and were injected icv on 4 consecutive days with either saline or 2DG. Repetitive saline (R-Saline, n = 15) rats were injected every day with saline, R-Saline/2DG rats (n = 16) were injected with saline on the first 3 d and 2DG on d 4, and repetitive 2DG (R-2DG, n = 16) rats were injected each day with 2DG. At 1200 h (T0) in the afternoon, blood glucose levels were checked from the tail vein (One-touch Ultra glucometer), and animals were injected icv with saline or 2DG. After injection, rats were placed back into their cages without food for 1 h. One hour after the icv injection (T60), blood glucose levels were checked again, a preweighed quantity of food was introduced in the cage, and food intake was measured over a period of 1 h. This treatment was conducted during 3 consecutive days during which body weight and 24-h food intake were also recorded. During these 3 consecutive days, R-Saline and R-Saline/2DG groups were considered as the same group because they were both receiving icv saline injection. On d 4, blood glucose levels were checked and rats received the last icv injection of saline or 2DG and were killed by decapitation (using a guillotine) either 10 or 60 min after the icv injection. Trunk blood and brain areas were collected as described in experiment 1.
Experiment 3: effect of AICAR on counterregulation after recurrent icv 2DG injections
In this experiment, rats were assigned to two different groups. All rats were injected on 4 consecutive days with 2DG (as described in experiment 2). On d 4, one group received an icv injection of AICAR (500 nmol in 5 µl) (R-2DG/AICAR, n = 5), whereas the other group received a saline injection (R-2DG/Saline, n = 7) 5 min before the last 2DG injection. Rats were then killed 60 min after the 2DG injection as described in experiment 1.
Experiment 4: effect of recurrent icv 2DG injections on brain glycogen levels.
In this experiment, rats were assigned to two different groups and were injected on 3 consecutive days with saline or 2DG (as described in experiment 2). On d 4, rats were killed 20 h after the last icv injection as described in experiment 1. The whole hypothalamus and a part of the Cx were rapidly dissected on ice and frozen in liquid nitrogen. The glycogen content of those tissues was determined by modifications of a procedure described by Chan and Exton (25). Samples were solubilized with 0.5 N KOH at 95 C. Glycogen was precipitated with 3 vol of ethanol and 0.1 vol of Na2SO4 (6%) at –80 C. The precipitate was washed with 70% ethanol and digested with amylo--1,4--1,6-glucosidase (EC 3.2.1.3) in acetate buffer (pH 4.9) during 60 min at 37 C. The amount of released glucose was determined using the glucose oxidase method (Thermo Electron). Glycogen from rabbit liver (Sigma Chemical Co., St. Louis, MO) was used as a standard.
AMPK activity assays
The 1- and 2-AMPK activity was measured in the different brain areas by immunoprecipitation from lysates (50 µg protein) with specific antibodies against the 1 and 2 catalytic subunits (generous gift from Dr. D. Carling) bound to protein-G/Sepharose beads. Kinase activity was measured using synthetic SAMS peptide and [-32P]ATP (20).
Western blot analysis
Tissue lysates were prepared as described previously. The total amount of 1- and 2-AMPK protein in hypothalamic nuclei was determined using 10% SDS acrylamide gels using antibodies to 1- and 2-AMPK (Santa Cruz Biotechnology, Santa Cruz, CA) (20). Detection of 1- and 2-AMPK was performed on independent blots. The same blots were reblotted with CaMKKß antibody (Santa Cruz Biotechnology). The specificity of the CaMKKß antibody was verified using brain lysates from CaMKKß knockout mice (generous gift from Dr. A. Means). Chemiluminescence (Western Lightning; PerkinElmer, Wellesley, MA) was quantified by laser densitometry within the linear range (Gene Snap).
Hormone assays
Serum glucagon concentration was measured on duplicate samples using RIA (Linco Research, St Charles, MO). Serum concentrations of corticosterone and insulin were determined on duplicate samples by an ELISA kit (IDS, Boldon, UK).
Data analysis
All data are presented as means ± SEM. Significance is set at P < 0.05. Glycemia time-course analysis was performed using repeated-measures ANOVA. Comparisons of mean serum corticosterone, glucagon, and AMPK activity after a single 2DG injection were made by unpaired t test. Food intake data analysis was performed using two-way ANOVA. Analyses of glycemia, serum corticosterone, and glucagon, AMPK activity, and AMPK protein levels after repetitive 2DG injection were made by one-way ANOVA with Bonferroni’s post test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Serum levels of corticosterone and glucagon in rats injected with 2DG were not different from saline-injected rats before the onset of hyperglycemia (10 min after icv 2DG). However, they were increased 2.5- to 3.4-fold in 2DG-injected rats compared with saline-injected rats at 60 min after icv 2DG when hyperglycemia was maximal (Fig. 1, B and C). Thus, the present model shows the expected metabolic responses of rats exposed acutely to central 2DG (23, 24, 26).
|
1- and 2-AMPK activity were measured in several hypothalamic areas, HB and Cx, during the same time course. 2DG-induced neuroglucopenia induced a rapid and strong activation of both 1- and 2-AMPK in the ARC and the MH, including the ventro- and dorsomedial hypothalamus (VMH and DMH) 10 min after the injection, whereas there was no change of 1- or 2-AMPK activity in the PVN, LH, HB, and Cx at the same time point (Fig. 2, A and B). We eliminated the possibility that the activation of both 1- and 2-AMPK 10 min after the injection was related to a hyperosmotic effect by injecting the same concentration of the nontransported glucose isomer L-glucose icv. The L-glucose injection had no effect on 1- or 2-AMPK activity in any brain area we studied (data not shown). At 60 min after the 2DG injection, both 1- and 2-AMPK were still activated in the ARC. However, 2-AMPK activity was reduced by 26% and 22% in the MH and PVN, respectively, compared with saline control, whereas 1-AMPK activity in the same nuclei was normal. Both 1- and 2-AMPK activity were unchanged in the LH, HB, and Cx 60 min after icv 2DG.
|
).
|
). On d 2, there was a marked reduction in the food intake response to 2DG in the R-2DG group compared with d 1, and on d 3, the 2DG effect on food intake was no longer statistically significant. Consistent with the food intake data, 2DG induced a 2.3-fold increase in glycemia 60 min after the injection on d 1 in the R-2DG group, and the response was decreased by 75% on d 2, 3, and 4 (Fig. 4A). However, a small 2DG effect remained on d 2, 3, and 4 in the R-2DG group compared with R-Saline or R-Saline/2DG groups (25% increase) (Fig. 4A). On d 4, icv 2DG induced hyperglycemia (T60) in the R-Saline/2DG group, the rats that had received daily saline injections and 2DG only on d 4. The response was similar to that observed in the R-2DG group on d 1. There was no difference in the basal glycemia (T0) between the three groups on d 1, 2, 3, and 4, suggesting that there was no chronic alteration of whole-body glucose homeostasis. In agreement with the reduced food intake response to 2DG, at the behavioral level, the rats showed reduced signs of fatigue, i.e. increased overall physical activity, on d 2, 3, and 4 in response to 2DG compared with d 1.
|
|
, B and C). However, 60 min after the injection, serum corticosterone and glucagon levels were decreased markedly in the R-2DG group compared with the R-Saline/2DG (Fig. 4, B and C). This was consistent with the markedly decreased blood glucose levels in response to 2DG observed in the R-2DG group on d 4 compared with R-saline/2DG (Fig. 4A). Taken together, these data demonstrate impaired 2DG-induced food intake and counterregulatory hormone response after repetitive icv 2DG injection and therefore validate our HAAF model.
Repetitive 2DG-induced neuroglucopenia impairs AMPK activation
Hypothalamic AMPK activity was measured at the two different time points on d 4 in our HAAF model. Repetitive saline injection did not alter the ability of 2DG to activate 1- and 2-AMPK in the ARC and MH at 10 min in the R-Saline/2DG group compared with R-saline (Fig. 5, A and B) because AMPK activation is similar to the data in Fig. 2A. However, 10 min after the 2DG injection, 1- and 2-AMPK activation in the ARC as well as 1-AMPK activation in the MH were impaired in the R-2DG group compared with R-Saline/2DG (Fig. 5, A and B). 2-AMPK activity in the ARC and 1-AMPK activity in the ARC and MH remained at baseline levels, i.e. equivalent to R-Saline in the R-2DG group. Sixty minutes after icv 2DG, 1- and 2-AMPK were activated in the ARC of R-Saline/2DG and R-2DG groups compared with R-Saline (Fig. 5, A and B). However, in the R-2DG group (Fig. 5A), at 60 min after 2DG injection, we did not observe a decrease in 2-AMPK activity in the MH and PVN of the R-2DG group compared with the R-Saline group, whereas 2-AMPK was decreased in MH and PVN of R-Saline/2DG rats.
|
1- and/or 2-AMPK protein, these protein levels were measured by Western blotting of the same lysates. There was no change in the total amount of either 1- or 2-AMPK protein in the hypothalamic nuclei (ARC, MH, and PVN) of the R-2DG group compared with R-Saline or R-Saline/2DG (data not shown). Protein levels of one of the upstream AMPK kinases, CaMKKß, were also not changed in these hypothalamic nuclei among the three groups (Fig. 6). Data are shown only for the nuclei in which we observed changes in 1- and/or 2-AMPK activity. Neither 1- nor 2-AMPK activity was changed in LH, HB, or Cx in the R-2DG group at 10 or 60 min after the injection compared with R-Saline or R-Saline/2DG (data not shown).
|
, A and B) were similar to the levels in R-2DG rats (Fig. 4, A and B). The icv AICAR injection resulted in a tendency for increased blood glucose at 60 min after the 2DG injection (135 ± 4 mg/dl, R-2DG/Saline, vs. 160 ± 20 mg/dl, R-2DG/AICAR; P < 0.13). AICAR injection in rats that received repetitive 2DG increased plasma corticosterone 2-fold (P < 0.05) compared with rats that received repetitive 2DG followed by saline (Fig. 7A) and tended to increase plasma glucagon by 40% (P < 0.06) (Fig. 7B). The small effect on glycemia agrees with published results showing that 1 h after icv AICAR injection (without preceding 2DG), peripheral glycemia is unchanged, but serum glucagon and corticosterone levels are increased (15). It is possible that at another time point after AICAR injection, the increase in counterregulatory hormones would be even greater. These results indicate that central AICAR injection in the setting of neuroglucopenia is sufficient to at least partially restore the counterregulatory hormone response.
|
Repetitive 2DG-induced neuroglucopenia increases brain glycogen content
To determine whether glycogen supercompensation (13) is involved in the down-regulation of the AMPK response with repetitive neuroglucopenia, hypothalamic and cortical glycogen content was measured after 3 consecutive days of icv 2DG injection. As a control, we measured glycogen content in the hypothalamus and Cx dissected immediately upon killing or 10 min after killing (induced by CO2) to deplete glycogen stores. Hypothalamic and cortical glycogen content in the rats in which the dissection was done 10 min after killing was approximately 0.03 µg/mg wet tissue compared with 0.21–0.25 µg/mg wet tissue when the brain was dissected immediately after killing (data not shown). Twenty hours after the last 2DG injection, glycogen content in the whole hypothalamus and Cx was increased by 69 and 153%, respectively, in the R-2DG group compared with saline-injected rats (Fig. 8).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Consistent with previous reports (23, 24), we found that icv administration of 2DG induces a counterregulatory response with increasing serum levels of glucagon and corticosterone, resulting in a 2.5-fold increase in blood glucose levels. An advantage of using icv 2DG compared with insulin-induced hypoglycemia is that the 2DG effects are acutely localized to the brain so that we can study signaling in the hypothalamus in the absence of other potentially confounding systemic effects that could modify the effects in the brain. We also found that 1- and 2-AMPK activities are quickly increased selectively in the ARC and MH (containing the VMH and DMH) at 10 min after 2DG injection, and this activation precedes the appearance of any glucose or hormonal response in the periphery. 2DG increases the AMP/ATP ratio in a neuronal cell line (29), and this is a plausible mechanism for the 2DG effect on hypothalamic AMPK activity.
Similar to our findings, Han et al. (15) also found activation of AMPK in the ARC and VMH in response to insulin-induced hypoglycemia. In addition, they found AMPK activation in the PVN, which we did not observe. This difference may result from differences in the methods by which hypoglycemia was induced. The absence of a short-term effect of 2DG in the PVN in our study cannot be related to a lack of 2DG penetration into that area because the PVN, like the ARC, is periventricular. However, because the PVN receives neuronal projections from the ARC and the HB, we speculate that modulation of neural activity in the ARC and/or HB induced by 2DG might obscure a direct effect of 2DG on AMPK activity in PVN neurons. In addition, despite the presence of glucose-sensing neurons in the HB, AMPK activity was not changed by 2DG in this area. This could be related to the fact that a large HB area was collected including the nucleus of the solitary tractus, the dorsal motor nucleus of the vagus, the C3 catecholamine cell group, and the area postrema, all of which contain glucose-sensitive neurons. We cannot rule out that nuclei-specific changes of AMPK activity were obscured by combining these nuclei. Nevertheless, the results of our model and that of Han et al. (15) are more similar than different, and both studies are consistent with a third model showing that in rats with recurrent insulin-induced hypoglycemia, AICAR injection into the VMH augments the dampened counterregulatory hormone response (21). Thus, all three studies support the notion that the VMH is one of the important sites in which AMPK activation promotes the counterregulatory hormone response to neuroglucopenia.
We also demonstrate isoform-specific and time-dependent AMPK regulation after 2DG administration. The 1- and 2-AMPK in the ARC remain activated 60 min after 2DG administration, whereas specifically 2-AMPK, but not 1-AMPK, activity is decreased in the MH and PVN at this same time point. One potential explanation for this isoform specificity is that the hyperglycemia that develops after 2DG administration may directly suppress 2-AMPK activity without changing 1-AMPK activity as has been previously shown after ip or icv glucose injection (19, 20). Similarly, elements of the counterregulatory response (e.g. catecholamines, glucagon, fatty acids, and glucocorticoids) may selectively affect one AMPK isoform. Catecholamines, glucagon, and fatty acids have all been shown to activate AMPK activity in peripheral tissues, although isoform specificity has not been investigated. However, in support of isoform-specific effects of hormones and substrates, in addition to high glucose, insulin and leptin also suppress only 2- and not 1-AMPK activity in the hypothalamus (20). Our results reveal significant complexity in the regulation of AMPK activity in terms of differential activation or inhibition of distinct AMPK isoforms with different time courses in different hypothalamic nuclei. This complexity could be attributed to secondary changes in neuronal activity induced by 2DG as well as direct or indirect neurohumoral feedback from the periphery. The mechanisms underlying the isoform-, time-, and nuclei-specific regulation of AMPK are not yet known.
We have developed and validated a model in which repeated icv 2DG administration replicates the key features of HAAF including diminished food intake and an impaired counterregulatory response after neuroglucopenia (3, 4, 5, 24). Although plasma epinephrine levels were not determined in this study, several studies have demonstrated impaired epinephrine response after repetitive glucopenia induced by insulin or 2DG (4, 21, 24). A key finding in our study is that HAAF induced by repetitive neuroglucopenia is associated with impaired activation of 2-AMPK in the ARC and impaired activation of both 1- and 2-AMPK in the MH 10 min after induction of neuroglucopenia on the fourth day. However, at 60 min after induction of neuroglucopenia, 1- and 2-AMPK activity in the ARC increased to a similar extent in the R-2DG and R-Saline/2DG groups, indicating a delay rather than a complete loss of AMPK activation in the ARC. Interestingly, we did not observe suppression of 2-AMPK activity in the MH and PVN after repeated 2DG administration in contrast to the suppression observed in these nuclei 60 min after a single 2DG injection. This observation is consistent with the possibility that suppression of AMPK activity in these nuclei may require the presence of an intact counterregulatory response and the resultant hyperglycemia.
It is important to note that the changes of AMPK activity were not related to any changes of 1- or 2-AMPK protein levels or the upstream AMPK kinase CaMKKß protein expression in the same nuclei. Recent data of McCrimmon et al. (21) showed increased mRNA levels of 1- and 2-AMPK in the hypothalamus of rats after repetitive insulin-induced hypoglycemia, but they did not report the protein levels of these catalytic subunits. Our data clearly show increased kinase activity of AMPK without a change in the amount of the catalytic subunit proteins.
To determine whether activation of AMPK is required for the counterregulatory hormone response, we injected AICAR before 2DG on the fourth day of repetitive 2DG. Administration of icv AICAR is sufficient to partially reverse the impairment in the counterregulatory response in this HAAF model. Together, these findings suggest that activation of the hypothalamic AMPK pathway is critical for the brisk and full counterregulatory hormone response to hypoglycemia and that failure to activate hypothalamic AMPK results in HAAF. However, the interpretation of our data is limited by the fact that the central AICAR effect might also involve AMPK-independent actions because AICAR has been demonstrated to modulate excitatory glutamatergic neurotransmission (27).
To our knowledge, this is the first report identifying altered responses of the AMPK pathway in association with HAAF. McCrimmon et al. (21) recently showed that injection of AICAR into the VMH amplifies the counterregulatory hormone response during a hypoglycemic clamp performed in a rat model of HAAF. However, neither impairment of hypothalamic AMPK activation by recurrent hypoglycemia nor the effect of AICAR on AMPK activity was investigated in that study.
Repetitive insulin-induced hypoglycemia or 2DG-induced glucopenia attenuates hypoglycemia-induced c-Fos expression, a readout of neuronal activation, in the hypothalamus and HB (3, 4, 5) These data strongly suggest that recurrent hypoglycemia alters the ability of neurons to sense glucopenia and are consistent with the general observation that repeated hypoglycemia shifts the glycemic thresholds for counterregulation to lower plasma glucose levels (1, 7). Our studies now demonstrate that hypothalamic AMPK activation in response to glucoprivation is impaired or delayed with HAAF, and this delay may be a critical component in the down-regulation of the counterregulatory hormone response to neuroglucopenia. Despite the absence of AMPK regulation in the PVN and HB of our HAAF model, we cannot rule out that AMPK signaling and/or other mechanisms are impaired in those areas and may participate in HAAF (3, 4). In line with this, Han et al. (15) demonstrated that dominant negative AMPK expression in the PVN inhibits hypoglycemia-induced secretion of corticosterone, suggesting a critical role for AMPK in the PVN.
The mechanism by which repetitive 2DG administration results in impaired hypothalamic AMPK activation remains uncertain. One recent hypothesis to explain the HAAF phenomenon is based on the observation that after a single episode of hypoglycemia, brain glycogen levels rebound to levels higher than the pre-hypoglycemic concentrations (13, 14). Gruetter and colleagues (14) have proposed that this glycogen supercompensation could provide an expanded source of glycolytic fuel (e.g. lactate), delaying neuroglucopenia in the setting of systemic hypoglycemia and delaying the onset of the counterregulatory response. In agreement with this, either lactate or glucose perfusion into the VMH during systemic hypoglycemia or VMH glucopenia decreases counterregulatory responses (30, 31, 32). Brain glycogen levels have not previously been reported in HAAF. In line with Gruetter’s hypothesis, we now show that hypothalamic and cortical glycogen content are increased after repetitive 2DG injections and remain elevated for at least 20 h after the last 2DG injection. This strong and sustained effect on brain glycogen suggests that glycogen supercompensation may be an important component of HAAF. Interestingly, the ß-regulatory subunit of AMPK has a glycogen-binding domain (33), and high cellular glycogen content represses activation of AMPK, presumably by binding to this domain (34, 35). However, whether high glycogen content could inhibit AMPK activity in the hypothalamus and not in the Cx is not known. However, our current and previous data (20, 36) show that AMPK activity in the Cx is not regulated by the same metabolic and hormonal changes that regulate hypothalamic AMPK activity. For example, cortical AMPK activity does not act as a nutrient sensor, even in response to changes in glucose concentrations (20). Determining whether the impairment in hypothalamic AMPK activity in this HAAF model is related to glycogen accumulation will be of great interest.
In summary, 2DG-induced neuroglucopenia results in dynamic changes of hypothalamic 1- and 2-AMPK activities in a nuclei-specific fashion, and repetitive 2DG-induced glucopenia resulting in HAAF disrupts the modulation of AMPK by 2DG. The impaired counterregulatory response induced by recurrent glucopenia was partially restored by AICAR. Finally, we demonstrated that hypothalamic glycogen content was increased by recurrent glucopenia. Together, these data point to a critical role for AMPK and hypothalamic glycogen metabolism in the etiology of HAAF. These findings suggest that drugs that activate AMPK in glucose-sensing hypothalamic regions may have therapeutic potential for individuals who suffer from HAAF.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Current address for T.A.: Montreal Diabetes Research Center, Montreal, Quebec H1W4A4.
Disclosure Statement: The authors have nothing to disclose.
First Published Online December 21, 2006
1 T.A. and J.K. contributed equally to this work. 
Abbreviations: AICAR, 5-Aminoimidazole-4-carboxamide 1-ß-D-ribofuranoside; AMPK, AMP-activated protein kinase; ARC, arcuate nucleus; CaMKK, Ca2+/calmodulin-dependent protein kinase kinase; Cx, cortex; 2DG, 2-deoxy-D-glucose; DMH, dorsomedial hypothalamus; HAAF, hypoglycemia-associated autonomic failure; HB, hindbrain; icv, intracerebroventricular; LH, lateral hypothalamus; MH, mediobasal hypothalamus; PVN, paraventricular nucleus; R-2DG, repetitive 2DG; R-Saline, repetitive saline; VMH, ventromedial hypothalamus.
Received August 1, 2006.
Accepted for publication December 11, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-lipoic acid mediated by suppression of hypothalamic AMP-activated protein kinase. Nat Med 10:727–733[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  G. Oz, A. Kumar, J. P. Rao, C. T. Kodl, L. Chow, L. E. Eberly, and E. R. Seaquist Human Brain Glycogen Metabolism During and After Hypoglycemia Diabetes, September 1, 2009; 58(9): 1978 - 1985. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. McCrimmon Glucose Sensing During Hypoglycemia: Lessons From the Lab Diabetes Care, August 1, 2009; 32(8): 1357 - 1363. [Full Text] [PDF]
|
|
|
|
 |
|
 G. R. Steinberg and B. E. Kemp AMPK in Health and Disease Physiol Rev, July 1, 2009; 89(3): 1025 - 1078. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. T. Borer, E. Wuorinen, K. Ku, and C. Burant Appetite Responds to Changes in Meal Content, Whereas Ghrelin, Leptin, and Insulin Track Changes in Energy Availability J. Clin. Endocrinol. Metab., July 1, 2009; 94(7): 2290 - 2298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Fan, Y. Ding, S. Brown, L. Zhou, M. Shaw, M. C. Vella, H. Cheng, E. C. McNay, R. S. Sherwin, and R. J. McCrimmon Hypothalamic AMP-activated protein kinase activation with AICAR amplifies counterregulatory responses to hypoglycemia in a rodent model of type 1 diabetes Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2009; 296(6): R1702 - R1708. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Sherwin Bringing Light to the Dark Side of Insulin: A Journey Across the Blood-Brain Barrier Diabetes, September 1, 2008; 57(9): 2259 - 2268. [Full Text] [PDF]
|
|
|
|
 |
|
 R. I. Herzog, O. Chan, S. Yu, J. Dziura, E. C. McNay, and R. S. Sherwin Effect of Acute and Recurrent Hypoglycemia on Changes in Brain Glycogen Concentration Endocrinology, April 1, 2008; 149(4): 1499 - 1504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Marty, M. Dallaporta, and B. Thorens Brain Glucose Sensing, Counterregulation, and Energy Homeostasis Physiology, August 1, 2007; 22(4): 241 - 251. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
