- Email: [email protected]
Dorsomedial hindbrain participation in glucoprivic feeding response to 2DG but not 2DG-induced hyperglycemia or activation of the HPA axis
Brain Research 801 Ž1998. 21–28
Research report
Dorsomedial hindbrain participation in glucoprivic feeding response to 2DG but not 2DG-induced hyperglycemia or activation of the HPA axis Brenda K. Edmonds, Gaylen L. Edwards
)
Department of Physiology and Pharmacology, College of Veterinary Medicine, The UniÕersity of Georgia, Athens, GA 30602, USA Accepted 12 May 1998
Abstract 2-Deoxy-D-glucose Ž2DG. is a glucose analogue that inhibits intracellular utilization of glucose and produces a characteristic behavioral response known as glucoprivic feeding. The area postrema ŽAP. is a caudal hindbrain structure shown previously to be involved in 2DG-induced glucoprivic feeding. In addition, peripheral administration of 2DG is known to elicit activation of both the hypothalamic-pituitary-adrenal ŽHPA. axis and the sympathoadrenomedullary system. The neural substrates for these neuroendocrine and neural responses to 2DG are not known although they may also involve the AP. The possible role of the AP in 2DG-induced feeding, activation of the HPA axis and hyperglycemia was investigated in Sprague–Dawley rats with lesions centered on the area postrema ŽAPX. and sham-operated ŽSHM. rats administered 2DG Ž200 mgrkg. or physiological saline Ž1 mlrkg.. Peripheral administration of 2DG evoked a feeding response in SHM rats that was abolished in APX animals. Interestingly, 2DG administered at this dose produced a significant increase in plasma corticosterone and plasma glucose in both SHM and APX rats for up to 4 h after drug treatment. Collectively, these findings suggest that the AP is involved in the behavioral Žfeeding. response to peripheral administration of 2DG, but does not appear to be a common neural substrate for the neuroendocrine ŽHPA axis. and sympathoadrenal Žhyperglycemic. responses to this agent. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Area postrema; Sympathoadrenomedullary; Hypothalamic-pituitary-adrenal axis; Glucoprivic feeding; 2-deoxy-D-glucose; Corticosterone
1. Introduction 2-Deoxy-D-glucose Ž2DG. is a glucose analogue that decreases intracellular glucose utilization by competing with D-glucose for the enzyme phosphoglucoisomerase w20x. This pharmacological blockade of glucose utilization by 2DG has been shown to induce feeding when the drug is administered into either the cerebroventricles w8,19x or peripheral circulation w9,18x. Caudal hindbrain structures, in particular the area postrema ŽAP., have been suggested to play a role in mediating 2DG-induced glucoprivic feeding w1,13,19,21x. Contreras et al. Ž1982. provided support for the idea that the AP is involved in short-term feeding responses to 2DG as glucoprivic feeding in AP-lesioned ŽAPX. rats was significantly attenuated, as compared to SHM animals, when consumption was measured in a 2-h test w1x. Studies aimed at determining the role of the AP in ) Corresponding author. [email protected]
Fax:
q 1-706-542-3015;
E-mail:
0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 5 2 8 - 9
more long-term Ž6 h. feeding tests have been less conclusive. Hyde and Miselis Ž1983. reported that feeding in APX rats was not significantly different from control animals over a 6-h feeding test w13x, whereas Ritter and Taylor reported that APX animals consumed significantly less than sham-operated controls during the 6-h test period following 2DG administration w21x. In the present study, feeding in response to 2DG administration was measured in animals with lesions centered on the AP as well as in SHM control animals during a 6-h trial. In addition to its effects on feeding behavior, peripheral administration of 2DG elicits both neural and neuroendocrine responses in animals. Peripheral administration of 2DG has been shown to activate the hypothalamic-pituitary-adrenal ŽHPA. axis, as indicated by significantly elevated plasma corticosterone concentrations in 2DG-treated animals as compared to saline-treated controls w30x. The neural substrates for 2DG-induced activation of the HPA axis are not known although they may involve hindbrain structures including the AP.
22
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
The paraventricular nuclei ŽPVN. of the hypothalamus, in particular the parvocellular neurosecretory cells of the PVN that contain corticotropin releasing factor ŽCRF., receive noradrenergic and adrenergic terminals from hindbrain nuclei w2,17x. Studies using expression of c-Fos, the protein product of the proto-oncogene c-fos as a marker for neuronal activation, provided evidence for 2DG-induced activation of hindbrain neurons located in the AP, nucleus of the solitary tract ŽNTS., ventrolateral medulla ŽVLM., and the parabrachial nuclei ŽPBN. w22x. Activation of these hindbrain structures by 2DG may result in subsequent stimulation of the HPA axis. It has not been determined if the AP is a common neural substrate for both the behavioral Žfeeding. and neuroendocrine ŽHPA axis. response to peripheral 2DG administration. Our lab previously reported involvement of the AP in both behavioral and neuroendocrine responses to peripheral administration of the chemical agent paraquat w5x. Paraquat is an herbicide that has been demonstrated to act in the AP to condition taste aversions w4x. The effects of 2DG on neuronal systems, such as the sympathoadrenomedullary system, are well documented w14,24x. Peripheral administration of 2DG was shown to produce marked increases in blood glucose, a response involving activation of both the sympathoadrenomedullary system and sympathetic nerve terminals. 2-Deoxyglucose-induced hyperglycemia was attenuated in animals whose adrenal glands had been surgically removed Žadrenalectomized. and in animals who had received peripheral injection of 6-hydroxydopamine, an agent that selectively destroys sympathetic nerve terminals Žsympathectomized. w14x. The neural substrates for 2DG-induced hyperglycemia are not known, but may involve hindbrain catecholamine-containing neurons. 2-Deoxy-glucose-induced activation of hindbrain catecholaminergic neurons located in the AP, NTS, VLM may mediate the hyperglycemic response to 2DG via pathways from these hindbrain cell groups to sympathetic preganglionic neurons of the spinal cord w3x. Neurons located in the rostral portion of the ventrolateral medulla have been shown to innervate sympathetic neurons of the intermediolateral cell column w27x. Likewise, 2DG-induced hyperglycemia may result from activation of hindbrain catecholaminergic neurons that project to the PVN of the hypothalamus, and subsequently activate sympathetic pathways via projections from the PVN to sympathetic preganglionic neurons w10,27,28x. Neuronal efferents arising from the AP project to other hindbrain nuclei, including the NTS, PBN, and VLM, such that the AP may be involved in mediating 2DG-induced responses of the sympathetic nervous system. The goals of this study were to confirm previous studies suggesting a role for the AP in glucoprivic feeding in response to 2DG and to determine if the AP is a common neural substrate for both the neuroendocrine ŽHPA axis. and neuronal Žsympathetic. responses to 2DG.
2. Materials and methods 2.1. Animals and housing A total of 34 male Sprague–Dawley rats ŽHarlan, Indianapolis, IN., approximately 250 g initial weight, were used in these studies. They were housed individually in hanging wire mesh cages. The cages were maintained in a temperature and humidity controlled environment with a 12 h lightr12 h dark cycle. The animals were given free access to food ŽPurina Rat Chow pellets. and water, and weighed once daily between 0800 and 1100 h for a 5-day acclimation period. The animals were randomly assigned to two groups, APX Ž n s 16. and SHM Ž n s 18.. Behavioral testing and blood sampling were conducted during the light phase of the lightrdark cycle. 2.2. Surgical lesioning Lesions of the AP were performed under ketamineacepromazine-xylazine Ž50 mgrkg ketamine, 3.3 mgrkg acepromazine, 3.3 mgrkg xylazine. general anesthesia using aseptic techniques according to a previously described technique w7x. Briefly, this involved aspiration of the AP Žafter visualization. via a surgical approach through the foramen magnum. Small, 30-gauge, stainless steel tubing was used for aspiration to minimize damage to areas surrounding the AP. Sham-operated controls Ž n s 18. were treated in a similar manner except the AP was touched with a cotton swab instead of surgically lesioned. The animals were allowed to recover for 15 to 30 days following surgery. 2.3. BehaÕioral (feeding) tests This initial study determined whether or not feeding would occur in response to peripheral administration of 2DG. A total of 34 animals Ž n s 18 in the SHM group and n s 16 in the APX group. were randomly assigned to one of two treatment groups, 2DG- or physiological saline solution ŽPSS.-treated, such that a total of four test groups were formed: SHM, 2DG-treated ŽSHM–DG.; SHM, saline-treated ŽSHM–PSS.; APX, 2DG-treated ŽAPX–DG.; and APX, saline injected ŽAPX–PSS.. Animals were allowed a brief 30-min pretest period where they were given access to three pellets of fresh, preweighed food and unlimited water, while additional sources of food were withheld. Immediately following the pretest, the animals received an injection of either 2DG Ž200 mgrkg,s.c., Sigma, St. Louis, MO. or PSS Ž1 mlrkg, s.c.., which served as vehicle for 2DG. The animals had continuous access to the fresh food and water, and food consumption was measured for the next 6 h. Intake was recorded at 30, 60, 120, 240, and 360 min, respectively after 2DG or PSS injection.
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
23
Following a 3-day recovery period, the food intake study was repeated using a cross-over design, such that animals within either surgery group ŽSHM or APX. who previously received 2DG now received PSS, and those previously receiving PSS, now received 2DG. This allowed each animal to serve as its own saline-injected control for 2DG treatment.
3-day recovery, the animals were treated with 2DG or PSS in a cross-over manner and tail blood collected over a 4-h period. Plasma glucose concentrations were measured by the enzymatic Žglucose oxidase. determination using commercial kits ŽSigma, St. Louis, MO..
2.4. Plasma corticosterone determination
Following all studies, each animal was decapitated and the brain removed and rapidly frozen on dry ice. The brainstem was immersion fixed in 10% buffered formalin with 30% sucrose for at least 5 days. Cryosectioning was performed to produce 30 micron sections mounted on gelatin coated slides, stained with Cresyl violet, and coverslipped. Histological examination of the slides was performed to assess completeness of the lesion and evaluate possible damage to adjacent structures. The extent of lesion was determined by comparing sections of the caudal medulla from SHM animals with an intact AP with sections of the caudal medulla from APX rats. Measuring the amount of tissue remaining in the NTS medial to the solitary tract allowed us to estimate the amount of NTS removed with the AP.
For this study, a total of 10 animals Ž n s 5rsurgery group, SHM or APX. used previously in the food intake study were randomly selected and assigned to two treatment groups: one group received 2DG at 200 mgrkg dissolved in PSS Žs.c.., the other group received an equivalent volume of physiological saline ŽPSS. at 1 mlrkg Žs.c.. Thus, there were a total of four test groups including SHM–DG, SHM–PSS, APX–DG, and APX–PSS. Blood was collected from the tail vein at four time points Ž0, 60, 120, and 240 min.. Immediately after the 0 min collection, animals received an injection of either 2DG or PSS, depending on the treatment group. Food, but not water, was withheld during the 4-h testing period. These time points were selected based on previous studies showing that plasma corticosterone levels are maximal within the first 3 h following activation of the HPA axis by a variety of chemical stimuli w15,26x. Following a 3-day recovery period, the animals were treated in a cross-over manner such that animals previously receiving 2DG received PSS, and those previously receiving PSS received 2DG. This allowed each animal to serve as its own control for 2DG treatment. Plasma corticosterone concentrations were determined by radioimmunoassay using commercial kits ŽICN Biomedical, Irvine, CA.. All samples were assayed in duplicate and any sample with a coefficient of variation greater than 10% was reassayed. Internal control samples provided with each kit were within the appropriate control range. 2.5. Plasma glucose determination To determine if peripheral 2DG administration would induce hyperglycemia in SHM and APX animals, a total of 24 animals Ž n s 12rsurgery group, SHM or APX. used previously in the behavioral testing study were randomly assigned to two treatment groups. Thus, four test groups Ž n s 6rgroup. were established for the initial test. These groups were: SHM–DG, SHM–PSS, APX–DG, and APX–PSS. Tail blood was collected at four time points Ž0, 60, 120, and 240 min. as described for the plasma corticosterone study. Food, but not water, was withheld during the 4-h testing period. Administration of either 2DG Ž200 mgrkg, s.c.. or PSS Ž1 mlrkg, s.c.. occurred immediately following the 0 min collection. These time points were selected based on previous studies demonstrating that the level of plasma glucose reaches a peak level within 60 min after peripheral 2DG administration w18,20x. Following a
2.6. Histology
2.7. Statistics For the feeding behavior testing, cumulative food consumption over the 360-min testing period was expressed as mean " S.E.M. and comparisons were made using two-way analysis of variance ŽANOVA. with time as a repeated measure. Likewise, for plasma corticosterone and glucose determinations, plasma corticosterone and glucose concentrations over the 240-min testing period were expressed as mean " S.E.M. and comparisons were made using ANOVA with time as a repeated measure. Statistical comparisons of cumulative food intake, plasma corticosterone, and plasma glucose for individual time points Ž30 to 360 min for food intake, 0 to 240 min for plasma corticosterone and glucose. between surgery groups ŽSHM vs. APX. were performed using an unpaired Student’s t-tests when ANOVA indicated a significant difference. Comparisons of cumulative food intake, plasma corticosterone, and plasma glucose for individual time points between treatment groups Ž2DG vs. PSS. and within surgery groups and were performed using paired Student’s t-tests when ANOVA indicated a significant difference.
3. Results 3.1. Feeding behaÕior testing Feeding behavior was examined in a total of 34 Ž16 APX and 18 SHM. animals. Mean cumulative consumption was determined at each individual time point Ž30, 60, 120, 240, and 360 min. after 2DG or PSS treatment ŽFig. 1.. Consumption amounts recorded for each time point
24
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
Fig. 1. Cumulative average consumption of food by APX Ž n s 16. and SHM Ž n s 18. rats over a 6-h period following 2DG Ž200 mgrkg. or an equivalent volume of physiological saline Ž1 mlrkg. administered at time 0 min. Values are mean " S.E.M. Cumulative intake was significantly lower for SHM–PSS Ž p - 0.05. than for SHM–DG at 120, 240 and 360 min after treatment. Cumulative intake was significantly lower for APX–DG Ž p - 0.005. than for SHM–DG rats for all time periods.
included all food consumed at prior time points within the 4-h testing period. Mean cumulative consumption for SHM–PSS animals was significantly lower Ž p - 0.05. than for SHM–DG animals for the overall 360-min testing period and for the individual time points, 120, 240 and 360 min, after treatment. Mean cumulative consumption for APX–DG animals was significantly lower Ž p - 0.005. than SHM–DG animals for the overall 360 min and at each of the individual time points within the 6-h test period. When comparing food consumption within the APX surgery group, APX–PSS treated animals were not significantly different Ž p ) 0.05. from APX–DG treated animals for the overall 360 min Additionally, the APX–PSS treated animals consumed significantly less over the 6 h trial than the SHM–PSS treated rats Ž p - 0.01.. 3.2. Plasma corticosterone concentrations Administration of 2DG Ž200 mgrkg., as compared to PSS Ž1 mlrkg., elevated plasma corticosterone concentrations in both SHM and APX rats. Plasma corticosterone concentrations were significantly elevated in SHM–DG treated as compared to SHM–PSS treated animals Ž p 0.001., and in APX–DG treated as compared to APX–PSS treated animals Ž p - 0.001. for the overall 240-min testing period ŽFig. 2.. Baseline plasma corticosterone levels, measured at time point 0, were significantly different Ž p - 0.001. between the two surgery groups and this difference remained at each subsequent time point Ž60, 120, and 240 min.. However, the two groups had a similar
plasma corticosterone profile for the overall 240-min testing period. Following 2DG treatment, plasma corticosterone levels for both SHM and APX animals rose sharply and peaked at 60 min and then declined toward baseline levels by 240 min. By comparison, plasma corticosterone levels in SHM–PSS rats remained relatively constant throughout the 240-min testing period and did not differ significantly Ž p ) 0.05. from baseline levels, while corticosterone levels in APX–PSS animals declined slightly from baseline levels by 60 min after treatment and then remained constant for the remainder of the 240-min period. 3.3. Plasma glucose determinations Administration of 2DG, as compared to PSS, changed the plasma glucose profile for both SHM and APX rats ŽFig. 3.. Following treatment with 2DG, plasma glucose levels for both surgery groups ŽSHM and APX. rose sharply and peaked at 60 min and then declined toward baseline by 240 min. Plasma glucose levels within treatment groups ŽSHM–DG vs. APX–DG; SHM–PSS vs. APX–PSS. were not statistically different Ž p ) 0.05. for the overall 240-min testing period or at any time point within the 4-h test period. By contrast, plasma glucose levels for SHM and APX animals administered saline remained near baseline levels throughout the 240-min testing period. These animals ŽSHM–PSS and APX–PSS. had significantly lower Ž p 0.001. plasma glucose levels than either group of 2DGtreated animals ŽSHM–DG and APX–DG. for the overall 240-min or for any time point during the 4-h test period.
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
25
Fig. 2. Plasma corticosterone concentration over a 4-h period in APX Ž n s 5. and SHM Ž n s 5. rats receiving either 2DG Ž200 mgrkg. or an equivalent volume of physiological saline Ž1 mlrkg. at time 0 min. Values are mean " S.E.M. Plasma corticosterone concentration was significantly elevated Ž p - 0.001. in SHM–DG as compared to SHM–PSS animals, and in APX–DG as compared to APX–PSS rats over the 240-min period. Plasma corticosterone concentrations were significantly Ž p - 0.001. elevated at time 0 min for APX–DG as compared to SHM–DG animals, although the temporal profiles of plasma corticosterone were similar for the two groups.
Fig. 3. Plasma glucose concentration over a 4-h period in APX and SHM rats receiving either 2DG Ž200 mgrkg. or an equivalent volume of physiological saline Ž1 mlrkg. at time 0 min. Values are mean " S.E.M. Ž n s 12.. Plasma glucose concentration was not significantly different within treatment groups ŽSHM–DG vs. APX–DG, SHM–PSS vs. APX–PSS., although concentrations for 2DG-treated animals ŽSHM–DG and APX–DG. were significantly higher Ž p - 0.001. than saline-treated ŽSHM–PSS and APX–PSS. rats over the 240-min period.
26
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
3.4. Histology Histological examination revealed that the AP was completely removed in all APX rats. There was variable damage to the surrounding NTS in APX rats, but this
damage was limited to the immediately adjacent NS. Measurement of the remnant NTS indicated that the damage did not extend more than 65 mm from the border of the AP ŽFig. 4B.. Additionally, the damage to the NTS appeared to be to NTS adjacent to the rostral border of the AP with
Fig. 4. Photomicrographs of representative cross-sections made through the caudal brainstem just caudal to obex at the level of the APrNTS. ŽA. Intact Žnot lesioned. ŽB. AP lesion. Note that the entire AP is removed in the APX rats with minimal damage to the adjacent NTS; XII, dorsal motor nucleus of the vagus nerve.
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
limited damage to the caudal NTS ventral to the AP. No damage to the AP or surrounding NTS was observed in SHM animals ŽFig. 4A..
4. Discussion These studies confirm previous studies that indicate that 2DG-induced feeding is significantly attenuated in rats with lesions centered on the AP, as compared to SHM controls w1,13,21x. Our data reveal that this attenuation was present at 30 min after drug treatment and persisted for the remainder of the 6-h test. Interestingly, the hyperglycemic response to 2DG and the activation of the HPA axis induced by 2DG remained intact in rats with lesions centered on the AP. These findings suggest the possibility that different neural substrates may underlie the behavioral, neuroendocrine and sympathoadrenal responses to 2DG. Studies aimed at examining the central neural pathways for glucoprivic feeding have provided evidence that the caudal hindbrain, and specifically the APrNTS, is the site of the presumed receptor cells for glucoprivic feeding in response to 2DG w19,21,22x. Lesion studies indicated that rostral projections of APrNTS neurons to the lateral parabrachial nucleus ŽLPBN. w23x and the PVN w23,25x were not required for glucoprivic feeding because feeding was not impaired by LPBN lesions or lesions of the entire PVN of the hypothalamus. However, lesions of the central nucleus of the amygdala ŽCNA. did impair glucoprivic feeding w23,29x. This subnucleus of the amygdala is known to be interconnected with both the LBPN and the APrNTS w11,12x. The fact that lesions of the PVN did not impair 2DG-induced feeding suggested neural circuits involving the PVN are not essential for the initiation of feeding in response to 2DG. Administration of 2DG to SHM and APX rats resulted in significantly elevated plasma corticosterone concentrations as compared to saline-treated controls, indicating activation of the HPA axis. Previous studies have reported increased plasma corticosterone levels in response to 2DG administration w30x. The neural substrates for 2DG-induced activation of the HPA axis are not known, but based on our findings, did not appear to involve the AP. Our finding of no significant difference in plasma corticosterone levels between SHM–DG and APX–DG animals suggests that the AP is not involved in 2DG-induced activation of this neuroendocrine axis. It is interesting to note that basal corticosterone levels were elevated in animals with a lesion of the AP. We have noted a similar elevation in our earlier reports. The mechanism underlying this increase is unclear. One possibility is the activity of neuropeptide Y ŽNPY. in the arcuate-paraventricular pathway in the hypothalamus. Previous reports indicate that increased NPY release in the paraventricular nucleus results in enhanced corticosterone release w16x. We
27
have recently reported elevated NPY mRNA in the arcuate nucleus and elevated NPY-immunoreactivity in the paraventricular nucleus of rats with lesions of the AP w6x. Thus, the heightened NPY activity in this pathway could account for the difference in basal corticosterone between SHM and APX animals. The hyperglycemia following 2DG administration is thought to be a response mediated by catecholamine release from the sympathoadrenomedullary system and sympathetic nerve terminals w14x. Recent studies provided strong evidence that the glucoreceptors mediating the sympathoadrenal hyperglycemic responses to central Ži.c.v.. administration of antimetabolites were located in the caudal hindbrain w19x. Our study investigated the potential role for the AP in 2DG-induced hyperglycemia. Our observation that SHM–DG treated animals had significantly elevated plasma glucose levels, as compared to their PSStreated controls, was consistent with previous studies that peripheral administration of 2DG evoked hyperglycemia w9,20x. A similar pattern in plasma glucose levels was observed in rats with lesions centered on the AP. Lesioned animals receiving 2DG had significantly elevated plasma glucose levels as compared to saline treated APX rats. Saline treatment did not alter plasma glucose levels and plasma glucose levels after saline treatment were similar in both the SHM and APX groups. Collectively, these findings suggest that the AP is not a neural substrate for 2DG-induced hyperglycemia. Activation of the sympathetic nervous system, and specifically the sympathoadrenomedullary system, by peripheral administration of 2DG does not require the AP although may likely involve other hindbrain structures such as the NTS, which project to the ventrolateral medulla and result in activation of neurons in the intermediolateral cell column of the spinal cord. In summary, we observed that peripheral administration of 2DG evoked short-term behavioral responses Žglucoprivic feeding. as well as activation of neuroendocrine ŽHPA axis., and neuronal Žhyperglycemia. systems in the rat. The neural pathways mediating these effects have not been elucidated; however, we found that AP lesions impaired 2DG-induced feeding, but not activation of the HPA axis or the hyperglycemia that occurs subsequent to a glucoprivic challenge, suggesting that there are distinct neural pathways for the behavioral, neuroendocrine and sympathoadrenal responses to 2DG. These findings for 2DG contrast with our previous reports for the chemical agent paraquat w5x. When administered peripherally, paraquat Ž25 mmolrkg, s.c.. elicited a characteristic behavioral response, conditioned taste aversion, and produced a significant increase in body weight. Additionally, paraquat administration was found to activate the HPA axis and evoked hyperglycemia in intact animals Žunpublished observation.. By contrast, in APX animals these responses were attenuated, suggesting that the AP is a possible common neural substrate for the behavioral, neuroendocrine and sympathoadrenal responses to
28
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
paraquat. In conclusion, the neural circuits involved in mediating the behavioral and neurological responses to chemical agents are complex and appear to be distinct for the chemical agent involved.
w13x
w14x
Acknowledgements This work was supported by a research grant from the National Institutes of Health ŽDK 42533. and funds from the Department of Physiology and Pharmacology, College of Veterinary Medicine, The University of Georgia.
References w1x R.J. Contreras, E. Fox, M.L. Drugovich, Area postrema lesions produce feeding deficits in the rat: effects of preoperative dieting and 2-deoxy-glucose, Physiol. Behav. 29 Ž1982. 875–884. w2x E.T. Cunningham, M.C. Bohn, P.E. Sawchenko, Organization of adrenergic inputs to the paraventricular and supraoptic nuclei of the hypothalamus in the rat, J. Comp. Neurol. 292 Ž1990. 651–667. w3x E.T. Cunningham, R.R. Miselis, P.E. Sawchenko, The relationship of efferent projections from the area postrema to vagal motor and brainstem catecholamine-containing cell groups: an axonal transport and immunohistochemical study in the rat, Neuroscience 58 Ž1994. 635–648. w4x M.S. Dey, R.I. Krieger, R.C. Ritter, Paraquat-induced, dose dependent conditioned taste aversions and weight loss mediated by the area postrema, Toxicol. Appl. Pharmacol. 87 Ž1987. 212–221. w5x B.K. Edmonds, G.L. Edwards, The area postrema is involved in paraquat-induced conditioned aversion behavior and neuroendocrine activation of the hypothalamic-pituitary-adrenal axis, Brain Res. 712 Ž1996. 127–133. w6x G.L. Edwards, B.D. White, B. He, R.G. Dean, R.J. Martin, Elevated hypothalamic neuropeptide Y levels in rats with dorsomedial hindbrain lesions, Brain Res. 755 Ž1997. 84–90. w7x G.L. Edwards, T.G. Beltz, J.D. Power, A.K. Johnson, Rapid-onset ‘need-free’ sodium appetite after lesions of the dorsomedial medulla, Am. J. Physiol. 264 Ž1993. R1242–R1247. w8x R.M. Engeset, R.C. Ritter, Intracerebroventricular 2-DG causes feeding in absence of other signs of glucoprivation, Brain Res. 202 Ž1980. 229–233. w9x A.N. Epstein, S. Nicolaidis, R.R. Miselis, The glucoprivic control of food intake and the glucostatic theory of feeding behavior, in: G.J. Morgan, F.L. Calarescu ŽEds.., Neural Integration of Physiological Mechanisms and Behavior, University Press, Toronto, Canada, 1975, pp. 148–168. w10x L.A. Fisher, M.R. Brown, Central regulation of stress responses: regulation of the autonomic nervous system and visceral function by corticotropin releasing factor-41, Bull. Clin. Endocr. Metab. 5 Ž1991. 35–50. w11x H. Herbert, M.M. Moga, C.B. Saper, Connections of the parabrachial nucleus with the nucleus of the solitary tract and the medullary reticular formation in the rat, J. Comp. Neurol. 293 Ž1990. 540–580. w12x H. Herbert, C.B. Saper, Cholecystokinin-, galanin-, and corticotropin-releasing factor-like immunoreactive projections from the
w15x
w16x w17x
w18x
w19x
w20x
w21x
w22x
w23x
w24x w25x
w26x
w27x
w28x
w29x
w30x
nucleus of the solitary tract to the parabrachial nucleus in the rat, J. Comp. Neurol. 293 Ž1990. 581–598. T.A. Hyde, R.R. Miselis, Effects of area postremarcaudal medial nucleus of the solitary tract lesions on food intake and body weight, Am. J. Physiol. 244 Ž1983. R577–R587. S. Karlsson, B. Ahren, Contribution of adrenergic nerves and the adrenals to 2-deoxy-glucose-induced insulin and glucagon secretion in the mouse, Int. J. Pancreatology 10 Ž3–4. Ž1991. 207–215. J. Lachuer, S. Gailler, B. Barbagli, M. Buda, M. Tappaz, Differential early time course of activation of brainstem catecholaminergic groups in response to various stresses, Neuroendocrinology 53 Ž1991. 589–596. J. Lehmann, Neuropeptide Y: an overview, Drug Dev. Res. 19 Ž1990. 329–351. P.M. Plotsky, E.T. Cunningham, E.P. Widmaier, Catecholaminergic modulation of corticotropin-releasing factor and adrenocorticotropin secretion, Endocr. Rev. 10 Ž1989. 437–458. R.C. Ritter, M. Roelke, M. Neville, Glucoprivic feeding behavior in absence of other signs of glucoprivation, Am. J. Physiol. 234 Ž1978. E617–E621. R.C. Ritter, P.G. Slusser, S. Stone, Glucoreceptors controlling feeding and blood glucose: location in the hindbrain, Science 213 Ž1981. 451–453. S. Ritter, Glucoprivation and glucoprivic control of food intake. in: R.C. Ritter, S. Ritter, C.D. Barnes ŽEds.., Feeding Behavior: Neural and Humoral Controls, Academic Press, Orlando, FL, 1986, pp. 271–313. S. Ritter, J.S. Taylor, Vagal sensory neurons are required for lipoprivic but not glucoprivic feeding in rats, Am. J. Physiol. 258 Ž1990. R1395–R1401. S. Ritter, N.Y. Calingasan, T.T. Dinh, J.S. Taylor, Expression of c-Fos protein is induced in specific neurons by metabolic inhibitors that increase food intake, Neurosci. Abstr. 17 Ž1991. 192. S. Ritter, N.Y. Calingasan, B. Hutton, T.T. Dinh, Cooperation of vagal and central neural systems in monitoring metabolic events controlling feeding behavior. in: S. Ritter, R.C. Ritter, C.D. Barnes ŽEds.., Neuroanatomy and Physiology of Abdominal Vagal Afferents, CRC Press, Boca Raton, FL, 1992, pp. 249–277. A. Scheurink, S. Ritter, Sympathoadrenal responses to glucoprivation and lipoprivation in rats, Physiol. Behav. 53 Ž1993. 995–1000. G. Shor-Posner, A.P. Azar, S. Insinga, S.F. Leibowitz, Deficits in the control of food intake after hypothalamic paraventricular lesions, Physiol. Behav. 35 Ž1985. 883–890. W.P. Smotherman, J.W. Hennessey, S. Levine, Plasma corticosterone levels during recovery from LiCl produced taste aversions, Behav. Biol. 16 Ž1976. 401–412. A.M. Strack, W.B. Sawyer, J.H. Hughes, K.B. Platt, A.D. Loewy, A general pattern of CNS innervation of the sympathetic outflow demonstrated by transneuronal pseudorabies viral infections, Brain Res. 491 Ž1989. 156–162. A.M. Strack, W.B. Sawyer, K.B. Platt, A.D. Loewy, CNS cell groups regulating outflow to adrenal gland as revealed by transneuronal cell body labeling with pseudorabies virus, Brain Res. 491 Ž1989. 274–296. M.G. Tordoff, P.J. Geiselman, C.V. Grijalva, S.W. Keifer, D. Novin, Amygdaloid lesions impair ingestive responses to 2-deoxyD-glucose but not insulin, Am. J. Physiol. 242 Ž1982. R129–R135. J. Weidenfeld, A.P. Corcos, A. Wohlman, S. Feldman, Characterization of the 2-deoxy-glucose effect on the adrenocortical axis, Endocrinology 134 Ž1994. 1924–1931.
Research report
Dorsomedial hindbrain participation in glucoprivic feeding response to 2DG but not 2DG-induced hyperglycemia or activation of the HPA axis Brenda K. Edmonds, Gaylen L. Edwards
)
Department of Physiology and Pharmacology, College of Veterinary Medicine, The UniÕersity of Georgia, Athens, GA 30602, USA Accepted 12 May 1998
Abstract 2-Deoxy-D-glucose Ž2DG. is a glucose analogue that inhibits intracellular utilization of glucose and produces a characteristic behavioral response known as glucoprivic feeding. The area postrema ŽAP. is a caudal hindbrain structure shown previously to be involved in 2DG-induced glucoprivic feeding. In addition, peripheral administration of 2DG is known to elicit activation of both the hypothalamic-pituitary-adrenal ŽHPA. axis and the sympathoadrenomedullary system. The neural substrates for these neuroendocrine and neural responses to 2DG are not known although they may also involve the AP. The possible role of the AP in 2DG-induced feeding, activation of the HPA axis and hyperglycemia was investigated in Sprague–Dawley rats with lesions centered on the area postrema ŽAPX. and sham-operated ŽSHM. rats administered 2DG Ž200 mgrkg. or physiological saline Ž1 mlrkg.. Peripheral administration of 2DG evoked a feeding response in SHM rats that was abolished in APX animals. Interestingly, 2DG administered at this dose produced a significant increase in plasma corticosterone and plasma glucose in both SHM and APX rats for up to 4 h after drug treatment. Collectively, these findings suggest that the AP is involved in the behavioral Žfeeding. response to peripheral administration of 2DG, but does not appear to be a common neural substrate for the neuroendocrine ŽHPA axis. and sympathoadrenal Žhyperglycemic. responses to this agent. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Area postrema; Sympathoadrenomedullary; Hypothalamic-pituitary-adrenal axis; Glucoprivic feeding; 2-deoxy-D-glucose; Corticosterone
1. Introduction 2-Deoxy-D-glucose Ž2DG. is a glucose analogue that decreases intracellular glucose utilization by competing with D-glucose for the enzyme phosphoglucoisomerase w20x. This pharmacological blockade of glucose utilization by 2DG has been shown to induce feeding when the drug is administered into either the cerebroventricles w8,19x or peripheral circulation w9,18x. Caudal hindbrain structures, in particular the area postrema ŽAP., have been suggested to play a role in mediating 2DG-induced glucoprivic feeding w1,13,19,21x. Contreras et al. Ž1982. provided support for the idea that the AP is involved in short-term feeding responses to 2DG as glucoprivic feeding in AP-lesioned ŽAPX. rats was significantly attenuated, as compared to SHM animals, when consumption was measured in a 2-h test w1x. Studies aimed at determining the role of the AP in ) Corresponding author. [email protected]
Fax:
q 1-706-542-3015;
E-mail:
0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 5 2 8 - 9
more long-term Ž6 h. feeding tests have been less conclusive. Hyde and Miselis Ž1983. reported that feeding in APX rats was not significantly different from control animals over a 6-h feeding test w13x, whereas Ritter and Taylor reported that APX animals consumed significantly less than sham-operated controls during the 6-h test period following 2DG administration w21x. In the present study, feeding in response to 2DG administration was measured in animals with lesions centered on the AP as well as in SHM control animals during a 6-h trial. In addition to its effects on feeding behavior, peripheral administration of 2DG elicits both neural and neuroendocrine responses in animals. Peripheral administration of 2DG has been shown to activate the hypothalamic-pituitary-adrenal ŽHPA. axis, as indicated by significantly elevated plasma corticosterone concentrations in 2DG-treated animals as compared to saline-treated controls w30x. The neural substrates for 2DG-induced activation of the HPA axis are not known although they may involve hindbrain structures including the AP.
22
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
The paraventricular nuclei ŽPVN. of the hypothalamus, in particular the parvocellular neurosecretory cells of the PVN that contain corticotropin releasing factor ŽCRF., receive noradrenergic and adrenergic terminals from hindbrain nuclei w2,17x. Studies using expression of c-Fos, the protein product of the proto-oncogene c-fos as a marker for neuronal activation, provided evidence for 2DG-induced activation of hindbrain neurons located in the AP, nucleus of the solitary tract ŽNTS., ventrolateral medulla ŽVLM., and the parabrachial nuclei ŽPBN. w22x. Activation of these hindbrain structures by 2DG may result in subsequent stimulation of the HPA axis. It has not been determined if the AP is a common neural substrate for both the behavioral Žfeeding. and neuroendocrine ŽHPA axis. response to peripheral 2DG administration. Our lab previously reported involvement of the AP in both behavioral and neuroendocrine responses to peripheral administration of the chemical agent paraquat w5x. Paraquat is an herbicide that has been demonstrated to act in the AP to condition taste aversions w4x. The effects of 2DG on neuronal systems, such as the sympathoadrenomedullary system, are well documented w14,24x. Peripheral administration of 2DG was shown to produce marked increases in blood glucose, a response involving activation of both the sympathoadrenomedullary system and sympathetic nerve terminals. 2-Deoxyglucose-induced hyperglycemia was attenuated in animals whose adrenal glands had been surgically removed Žadrenalectomized. and in animals who had received peripheral injection of 6-hydroxydopamine, an agent that selectively destroys sympathetic nerve terminals Žsympathectomized. w14x. The neural substrates for 2DG-induced hyperglycemia are not known, but may involve hindbrain catecholamine-containing neurons. 2-Deoxy-glucose-induced activation of hindbrain catecholaminergic neurons located in the AP, NTS, VLM may mediate the hyperglycemic response to 2DG via pathways from these hindbrain cell groups to sympathetic preganglionic neurons of the spinal cord w3x. Neurons located in the rostral portion of the ventrolateral medulla have been shown to innervate sympathetic neurons of the intermediolateral cell column w27x. Likewise, 2DG-induced hyperglycemia may result from activation of hindbrain catecholaminergic neurons that project to the PVN of the hypothalamus, and subsequently activate sympathetic pathways via projections from the PVN to sympathetic preganglionic neurons w10,27,28x. Neuronal efferents arising from the AP project to other hindbrain nuclei, including the NTS, PBN, and VLM, such that the AP may be involved in mediating 2DG-induced responses of the sympathetic nervous system. The goals of this study were to confirm previous studies suggesting a role for the AP in glucoprivic feeding in response to 2DG and to determine if the AP is a common neural substrate for both the neuroendocrine ŽHPA axis. and neuronal Žsympathetic. responses to 2DG.
2. Materials and methods 2.1. Animals and housing A total of 34 male Sprague–Dawley rats ŽHarlan, Indianapolis, IN., approximately 250 g initial weight, were used in these studies. They were housed individually in hanging wire mesh cages. The cages were maintained in a temperature and humidity controlled environment with a 12 h lightr12 h dark cycle. The animals were given free access to food ŽPurina Rat Chow pellets. and water, and weighed once daily between 0800 and 1100 h for a 5-day acclimation period. The animals were randomly assigned to two groups, APX Ž n s 16. and SHM Ž n s 18.. Behavioral testing and blood sampling were conducted during the light phase of the lightrdark cycle. 2.2. Surgical lesioning Lesions of the AP were performed under ketamineacepromazine-xylazine Ž50 mgrkg ketamine, 3.3 mgrkg acepromazine, 3.3 mgrkg xylazine. general anesthesia using aseptic techniques according to a previously described technique w7x. Briefly, this involved aspiration of the AP Žafter visualization. via a surgical approach through the foramen magnum. Small, 30-gauge, stainless steel tubing was used for aspiration to minimize damage to areas surrounding the AP. Sham-operated controls Ž n s 18. were treated in a similar manner except the AP was touched with a cotton swab instead of surgically lesioned. The animals were allowed to recover for 15 to 30 days following surgery. 2.3. BehaÕioral (feeding) tests This initial study determined whether or not feeding would occur in response to peripheral administration of 2DG. A total of 34 animals Ž n s 18 in the SHM group and n s 16 in the APX group. were randomly assigned to one of two treatment groups, 2DG- or physiological saline solution ŽPSS.-treated, such that a total of four test groups were formed: SHM, 2DG-treated ŽSHM–DG.; SHM, saline-treated ŽSHM–PSS.; APX, 2DG-treated ŽAPX–DG.; and APX, saline injected ŽAPX–PSS.. Animals were allowed a brief 30-min pretest period where they were given access to three pellets of fresh, preweighed food and unlimited water, while additional sources of food were withheld. Immediately following the pretest, the animals received an injection of either 2DG Ž200 mgrkg,s.c., Sigma, St. Louis, MO. or PSS Ž1 mlrkg, s.c.., which served as vehicle for 2DG. The animals had continuous access to the fresh food and water, and food consumption was measured for the next 6 h. Intake was recorded at 30, 60, 120, 240, and 360 min, respectively after 2DG or PSS injection.
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
23
Following a 3-day recovery period, the food intake study was repeated using a cross-over design, such that animals within either surgery group ŽSHM or APX. who previously received 2DG now received PSS, and those previously receiving PSS, now received 2DG. This allowed each animal to serve as its own saline-injected control for 2DG treatment.
3-day recovery, the animals were treated with 2DG or PSS in a cross-over manner and tail blood collected over a 4-h period. Plasma glucose concentrations were measured by the enzymatic Žglucose oxidase. determination using commercial kits ŽSigma, St. Louis, MO..
2.4. Plasma corticosterone determination
Following all studies, each animal was decapitated and the brain removed and rapidly frozen on dry ice. The brainstem was immersion fixed in 10% buffered formalin with 30% sucrose for at least 5 days. Cryosectioning was performed to produce 30 micron sections mounted on gelatin coated slides, stained with Cresyl violet, and coverslipped. Histological examination of the slides was performed to assess completeness of the lesion and evaluate possible damage to adjacent structures. The extent of lesion was determined by comparing sections of the caudal medulla from SHM animals with an intact AP with sections of the caudal medulla from APX rats. Measuring the amount of tissue remaining in the NTS medial to the solitary tract allowed us to estimate the amount of NTS removed with the AP.
For this study, a total of 10 animals Ž n s 5rsurgery group, SHM or APX. used previously in the food intake study were randomly selected and assigned to two treatment groups: one group received 2DG at 200 mgrkg dissolved in PSS Žs.c.., the other group received an equivalent volume of physiological saline ŽPSS. at 1 mlrkg Žs.c.. Thus, there were a total of four test groups including SHM–DG, SHM–PSS, APX–DG, and APX–PSS. Blood was collected from the tail vein at four time points Ž0, 60, 120, and 240 min.. Immediately after the 0 min collection, animals received an injection of either 2DG or PSS, depending on the treatment group. Food, but not water, was withheld during the 4-h testing period. These time points were selected based on previous studies showing that plasma corticosterone levels are maximal within the first 3 h following activation of the HPA axis by a variety of chemical stimuli w15,26x. Following a 3-day recovery period, the animals were treated in a cross-over manner such that animals previously receiving 2DG received PSS, and those previously receiving PSS received 2DG. This allowed each animal to serve as its own control for 2DG treatment. Plasma corticosterone concentrations were determined by radioimmunoassay using commercial kits ŽICN Biomedical, Irvine, CA.. All samples were assayed in duplicate and any sample with a coefficient of variation greater than 10% was reassayed. Internal control samples provided with each kit were within the appropriate control range. 2.5. Plasma glucose determination To determine if peripheral 2DG administration would induce hyperglycemia in SHM and APX animals, a total of 24 animals Ž n s 12rsurgery group, SHM or APX. used previously in the behavioral testing study were randomly assigned to two treatment groups. Thus, four test groups Ž n s 6rgroup. were established for the initial test. These groups were: SHM–DG, SHM–PSS, APX–DG, and APX–PSS. Tail blood was collected at four time points Ž0, 60, 120, and 240 min. as described for the plasma corticosterone study. Food, but not water, was withheld during the 4-h testing period. Administration of either 2DG Ž200 mgrkg, s.c.. or PSS Ž1 mlrkg, s.c.. occurred immediately following the 0 min collection. These time points were selected based on previous studies demonstrating that the level of plasma glucose reaches a peak level within 60 min after peripheral 2DG administration w18,20x. Following a
2.6. Histology
2.7. Statistics For the feeding behavior testing, cumulative food consumption over the 360-min testing period was expressed as mean " S.E.M. and comparisons were made using two-way analysis of variance ŽANOVA. with time as a repeated measure. Likewise, for plasma corticosterone and glucose determinations, plasma corticosterone and glucose concentrations over the 240-min testing period were expressed as mean " S.E.M. and comparisons were made using ANOVA with time as a repeated measure. Statistical comparisons of cumulative food intake, plasma corticosterone, and plasma glucose for individual time points Ž30 to 360 min for food intake, 0 to 240 min for plasma corticosterone and glucose. between surgery groups ŽSHM vs. APX. were performed using an unpaired Student’s t-tests when ANOVA indicated a significant difference. Comparisons of cumulative food intake, plasma corticosterone, and plasma glucose for individual time points between treatment groups Ž2DG vs. PSS. and within surgery groups and were performed using paired Student’s t-tests when ANOVA indicated a significant difference.
3. Results 3.1. Feeding behaÕior testing Feeding behavior was examined in a total of 34 Ž16 APX and 18 SHM. animals. Mean cumulative consumption was determined at each individual time point Ž30, 60, 120, 240, and 360 min. after 2DG or PSS treatment ŽFig. 1.. Consumption amounts recorded for each time point
24
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
Fig. 1. Cumulative average consumption of food by APX Ž n s 16. and SHM Ž n s 18. rats over a 6-h period following 2DG Ž200 mgrkg. or an equivalent volume of physiological saline Ž1 mlrkg. administered at time 0 min. Values are mean " S.E.M. Cumulative intake was significantly lower for SHM–PSS Ž p - 0.05. than for SHM–DG at 120, 240 and 360 min after treatment. Cumulative intake was significantly lower for APX–DG Ž p - 0.005. than for SHM–DG rats for all time periods.
included all food consumed at prior time points within the 4-h testing period. Mean cumulative consumption for SHM–PSS animals was significantly lower Ž p - 0.05. than for SHM–DG animals for the overall 360-min testing period and for the individual time points, 120, 240 and 360 min, after treatment. Mean cumulative consumption for APX–DG animals was significantly lower Ž p - 0.005. than SHM–DG animals for the overall 360 min and at each of the individual time points within the 6-h test period. When comparing food consumption within the APX surgery group, APX–PSS treated animals were not significantly different Ž p ) 0.05. from APX–DG treated animals for the overall 360 min Additionally, the APX–PSS treated animals consumed significantly less over the 6 h trial than the SHM–PSS treated rats Ž p - 0.01.. 3.2. Plasma corticosterone concentrations Administration of 2DG Ž200 mgrkg., as compared to PSS Ž1 mlrkg., elevated plasma corticosterone concentrations in both SHM and APX rats. Plasma corticosterone concentrations were significantly elevated in SHM–DG treated as compared to SHM–PSS treated animals Ž p 0.001., and in APX–DG treated as compared to APX–PSS treated animals Ž p - 0.001. for the overall 240-min testing period ŽFig. 2.. Baseline plasma corticosterone levels, measured at time point 0, were significantly different Ž p - 0.001. between the two surgery groups and this difference remained at each subsequent time point Ž60, 120, and 240 min.. However, the two groups had a similar
plasma corticosterone profile for the overall 240-min testing period. Following 2DG treatment, plasma corticosterone levels for both SHM and APX animals rose sharply and peaked at 60 min and then declined toward baseline levels by 240 min. By comparison, plasma corticosterone levels in SHM–PSS rats remained relatively constant throughout the 240-min testing period and did not differ significantly Ž p ) 0.05. from baseline levels, while corticosterone levels in APX–PSS animals declined slightly from baseline levels by 60 min after treatment and then remained constant for the remainder of the 240-min period. 3.3. Plasma glucose determinations Administration of 2DG, as compared to PSS, changed the plasma glucose profile for both SHM and APX rats ŽFig. 3.. Following treatment with 2DG, plasma glucose levels for both surgery groups ŽSHM and APX. rose sharply and peaked at 60 min and then declined toward baseline by 240 min. Plasma glucose levels within treatment groups ŽSHM–DG vs. APX–DG; SHM–PSS vs. APX–PSS. were not statistically different Ž p ) 0.05. for the overall 240-min testing period or at any time point within the 4-h test period. By contrast, plasma glucose levels for SHM and APX animals administered saline remained near baseline levels throughout the 240-min testing period. These animals ŽSHM–PSS and APX–PSS. had significantly lower Ž p 0.001. plasma glucose levels than either group of 2DGtreated animals ŽSHM–DG and APX–DG. for the overall 240-min or for any time point during the 4-h test period.
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
25
Fig. 2. Plasma corticosterone concentration over a 4-h period in APX Ž n s 5. and SHM Ž n s 5. rats receiving either 2DG Ž200 mgrkg. or an equivalent volume of physiological saline Ž1 mlrkg. at time 0 min. Values are mean " S.E.M. Plasma corticosterone concentration was significantly elevated Ž p - 0.001. in SHM–DG as compared to SHM–PSS animals, and in APX–DG as compared to APX–PSS rats over the 240-min period. Plasma corticosterone concentrations were significantly Ž p - 0.001. elevated at time 0 min for APX–DG as compared to SHM–DG animals, although the temporal profiles of plasma corticosterone were similar for the two groups.
Fig. 3. Plasma glucose concentration over a 4-h period in APX and SHM rats receiving either 2DG Ž200 mgrkg. or an equivalent volume of physiological saline Ž1 mlrkg. at time 0 min. Values are mean " S.E.M. Ž n s 12.. Plasma glucose concentration was not significantly different within treatment groups ŽSHM–DG vs. APX–DG, SHM–PSS vs. APX–PSS., although concentrations for 2DG-treated animals ŽSHM–DG and APX–DG. were significantly higher Ž p - 0.001. than saline-treated ŽSHM–PSS and APX–PSS. rats over the 240-min period.
26
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
3.4. Histology Histological examination revealed that the AP was completely removed in all APX rats. There was variable damage to the surrounding NTS in APX rats, but this
damage was limited to the immediately adjacent NS. Measurement of the remnant NTS indicated that the damage did not extend more than 65 mm from the border of the AP ŽFig. 4B.. Additionally, the damage to the NTS appeared to be to NTS adjacent to the rostral border of the AP with
Fig. 4. Photomicrographs of representative cross-sections made through the caudal brainstem just caudal to obex at the level of the APrNTS. ŽA. Intact Žnot lesioned. ŽB. AP lesion. Note that the entire AP is removed in the APX rats with minimal damage to the adjacent NTS; XII, dorsal motor nucleus of the vagus nerve.
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
limited damage to the caudal NTS ventral to the AP. No damage to the AP or surrounding NTS was observed in SHM animals ŽFig. 4A..
4. Discussion These studies confirm previous studies that indicate that 2DG-induced feeding is significantly attenuated in rats with lesions centered on the AP, as compared to SHM controls w1,13,21x. Our data reveal that this attenuation was present at 30 min after drug treatment and persisted for the remainder of the 6-h test. Interestingly, the hyperglycemic response to 2DG and the activation of the HPA axis induced by 2DG remained intact in rats with lesions centered on the AP. These findings suggest the possibility that different neural substrates may underlie the behavioral, neuroendocrine and sympathoadrenal responses to 2DG. Studies aimed at examining the central neural pathways for glucoprivic feeding have provided evidence that the caudal hindbrain, and specifically the APrNTS, is the site of the presumed receptor cells for glucoprivic feeding in response to 2DG w19,21,22x. Lesion studies indicated that rostral projections of APrNTS neurons to the lateral parabrachial nucleus ŽLPBN. w23x and the PVN w23,25x were not required for glucoprivic feeding because feeding was not impaired by LPBN lesions or lesions of the entire PVN of the hypothalamus. However, lesions of the central nucleus of the amygdala ŽCNA. did impair glucoprivic feeding w23,29x. This subnucleus of the amygdala is known to be interconnected with both the LBPN and the APrNTS w11,12x. The fact that lesions of the PVN did not impair 2DG-induced feeding suggested neural circuits involving the PVN are not essential for the initiation of feeding in response to 2DG. Administration of 2DG to SHM and APX rats resulted in significantly elevated plasma corticosterone concentrations as compared to saline-treated controls, indicating activation of the HPA axis. Previous studies have reported increased plasma corticosterone levels in response to 2DG administration w30x. The neural substrates for 2DG-induced activation of the HPA axis are not known, but based on our findings, did not appear to involve the AP. Our finding of no significant difference in plasma corticosterone levels between SHM–DG and APX–DG animals suggests that the AP is not involved in 2DG-induced activation of this neuroendocrine axis. It is interesting to note that basal corticosterone levels were elevated in animals with a lesion of the AP. We have noted a similar elevation in our earlier reports. The mechanism underlying this increase is unclear. One possibility is the activity of neuropeptide Y ŽNPY. in the arcuate-paraventricular pathway in the hypothalamus. Previous reports indicate that increased NPY release in the paraventricular nucleus results in enhanced corticosterone release w16x. We
27
have recently reported elevated NPY mRNA in the arcuate nucleus and elevated NPY-immunoreactivity in the paraventricular nucleus of rats with lesions of the AP w6x. Thus, the heightened NPY activity in this pathway could account for the difference in basal corticosterone between SHM and APX animals. The hyperglycemia following 2DG administration is thought to be a response mediated by catecholamine release from the sympathoadrenomedullary system and sympathetic nerve terminals w14x. Recent studies provided strong evidence that the glucoreceptors mediating the sympathoadrenal hyperglycemic responses to central Ži.c.v.. administration of antimetabolites were located in the caudal hindbrain w19x. Our study investigated the potential role for the AP in 2DG-induced hyperglycemia. Our observation that SHM–DG treated animals had significantly elevated plasma glucose levels, as compared to their PSStreated controls, was consistent with previous studies that peripheral administration of 2DG evoked hyperglycemia w9,20x. A similar pattern in plasma glucose levels was observed in rats with lesions centered on the AP. Lesioned animals receiving 2DG had significantly elevated plasma glucose levels as compared to saline treated APX rats. Saline treatment did not alter plasma glucose levels and plasma glucose levels after saline treatment were similar in both the SHM and APX groups. Collectively, these findings suggest that the AP is not a neural substrate for 2DG-induced hyperglycemia. Activation of the sympathetic nervous system, and specifically the sympathoadrenomedullary system, by peripheral administration of 2DG does not require the AP although may likely involve other hindbrain structures such as the NTS, which project to the ventrolateral medulla and result in activation of neurons in the intermediolateral cell column of the spinal cord. In summary, we observed that peripheral administration of 2DG evoked short-term behavioral responses Žglucoprivic feeding. as well as activation of neuroendocrine ŽHPA axis., and neuronal Žhyperglycemia. systems in the rat. The neural pathways mediating these effects have not been elucidated; however, we found that AP lesions impaired 2DG-induced feeding, but not activation of the HPA axis or the hyperglycemia that occurs subsequent to a glucoprivic challenge, suggesting that there are distinct neural pathways for the behavioral, neuroendocrine and sympathoadrenal responses to 2DG. These findings for 2DG contrast with our previous reports for the chemical agent paraquat w5x. When administered peripherally, paraquat Ž25 mmolrkg, s.c.. elicited a characteristic behavioral response, conditioned taste aversion, and produced a significant increase in body weight. Additionally, paraquat administration was found to activate the HPA axis and evoked hyperglycemia in intact animals Žunpublished observation.. By contrast, in APX animals these responses were attenuated, suggesting that the AP is a possible common neural substrate for the behavioral, neuroendocrine and sympathoadrenal responses to
28
B.K. Edmonds, G.L. Edwardsr Brain Research 801 (1998) 21–28
paraquat. In conclusion, the neural circuits involved in mediating the behavioral and neurological responses to chemical agents are complex and appear to be distinct for the chemical agent involved.
w13x
w14x
Acknowledgements This work was supported by a research grant from the National Institutes of Health ŽDK 42533. and funds from the Department of Physiology and Pharmacology, College of Veterinary Medicine, The University of Georgia.
References w1x R.J. Contreras, E. Fox, M.L. Drugovich, Area postrema lesions produce feeding deficits in the rat: effects of preoperative dieting and 2-deoxy-glucose, Physiol. Behav. 29 Ž1982. 875–884. w2x E.T. Cunningham, M.C. Bohn, P.E. Sawchenko, Organization of adrenergic inputs to the paraventricular and supraoptic nuclei of the hypothalamus in the rat, J. Comp. Neurol. 292 Ž1990. 651–667. w3x E.T. Cunningham, R.R. Miselis, P.E. Sawchenko, The relationship of efferent projections from the area postrema to vagal motor and brainstem catecholamine-containing cell groups: an axonal transport and immunohistochemical study in the rat, Neuroscience 58 Ž1994. 635–648. w4x M.S. Dey, R.I. Krieger, R.C. Ritter, Paraquat-induced, dose dependent conditioned taste aversions and weight loss mediated by the area postrema, Toxicol. Appl. Pharmacol. 87 Ž1987. 212–221. w5x B.K. Edmonds, G.L. Edwards, The area postrema is involved in paraquat-induced conditioned aversion behavior and neuroendocrine activation of the hypothalamic-pituitary-adrenal axis, Brain Res. 712 Ž1996. 127–133. w6x G.L. Edwards, B.D. White, B. He, R.G. Dean, R.J. Martin, Elevated hypothalamic neuropeptide Y levels in rats with dorsomedial hindbrain lesions, Brain Res. 755 Ž1997. 84–90. w7x G.L. Edwards, T.G. Beltz, J.D. Power, A.K. Johnson, Rapid-onset ‘need-free’ sodium appetite after lesions of the dorsomedial medulla, Am. J. Physiol. 264 Ž1993. R1242–R1247. w8x R.M. Engeset, R.C. Ritter, Intracerebroventricular 2-DG causes feeding in absence of other signs of glucoprivation, Brain Res. 202 Ž1980. 229–233. w9x A.N. Epstein, S. Nicolaidis, R.R. Miselis, The glucoprivic control of food intake and the glucostatic theory of feeding behavior, in: G.J. Morgan, F.L. Calarescu ŽEds.., Neural Integration of Physiological Mechanisms and Behavior, University Press, Toronto, Canada, 1975, pp. 148–168. w10x L.A. Fisher, M.R. Brown, Central regulation of stress responses: regulation of the autonomic nervous system and visceral function by corticotropin releasing factor-41, Bull. Clin. Endocr. Metab. 5 Ž1991. 35–50. w11x H. Herbert, M.M. Moga, C.B. Saper, Connections of the parabrachial nucleus with the nucleus of the solitary tract and the medullary reticular formation in the rat, J. Comp. Neurol. 293 Ž1990. 540–580. w12x H. Herbert, C.B. Saper, Cholecystokinin-, galanin-, and corticotropin-releasing factor-like immunoreactive projections from the
w15x
w16x w17x
w18x
w19x
w20x
w21x
w22x
w23x
w24x w25x
w26x
w27x
w28x
w29x
w30x
nucleus of the solitary tract to the parabrachial nucleus in the rat, J. Comp. Neurol. 293 Ž1990. 581–598. T.A. Hyde, R.R. Miselis, Effects of area postremarcaudal medial nucleus of the solitary tract lesions on food intake and body weight, Am. J. Physiol. 244 Ž1983. R577–R587. S. Karlsson, B. Ahren, Contribution of adrenergic nerves and the adrenals to 2-deoxy-glucose-induced insulin and glucagon secretion in the mouse, Int. J. Pancreatology 10 Ž3–4. Ž1991. 207–215. J. Lachuer, S. Gailler, B. Barbagli, M. Buda, M. Tappaz, Differential early time course of activation of brainstem catecholaminergic groups in response to various stresses, Neuroendocrinology 53 Ž1991. 589–596. J. Lehmann, Neuropeptide Y: an overview, Drug Dev. Res. 19 Ž1990. 329–351. P.M. Plotsky, E.T. Cunningham, E.P. Widmaier, Catecholaminergic modulation of corticotropin-releasing factor and adrenocorticotropin secretion, Endocr. Rev. 10 Ž1989. 437–458. R.C. Ritter, M. Roelke, M. Neville, Glucoprivic feeding behavior in absence of other signs of glucoprivation, Am. J. Physiol. 234 Ž1978. E617–E621. R.C. Ritter, P.G. Slusser, S. Stone, Glucoreceptors controlling feeding and blood glucose: location in the hindbrain, Science 213 Ž1981. 451–453. S. Ritter, Glucoprivation and glucoprivic control of food intake. in: R.C. Ritter, S. Ritter, C.D. Barnes ŽEds.., Feeding Behavior: Neural and Humoral Controls, Academic Press, Orlando, FL, 1986, pp. 271–313. S. Ritter, J.S. Taylor, Vagal sensory neurons are required for lipoprivic but not glucoprivic feeding in rats, Am. J. Physiol. 258 Ž1990. R1395–R1401. S. Ritter, N.Y. Calingasan, T.T. Dinh, J.S. Taylor, Expression of c-Fos protein is induced in specific neurons by metabolic inhibitors that increase food intake, Neurosci. Abstr. 17 Ž1991. 192. S. Ritter, N.Y. Calingasan, B. Hutton, T.T. Dinh, Cooperation of vagal and central neural systems in monitoring metabolic events controlling feeding behavior. in: S. Ritter, R.C. Ritter, C.D. Barnes ŽEds.., Neuroanatomy and Physiology of Abdominal Vagal Afferents, CRC Press, Boca Raton, FL, 1992, pp. 249–277. A. Scheurink, S. Ritter, Sympathoadrenal responses to glucoprivation and lipoprivation in rats, Physiol. Behav. 53 Ž1993. 995–1000. G. Shor-Posner, A.P. Azar, S. Insinga, S.F. Leibowitz, Deficits in the control of food intake after hypothalamic paraventricular lesions, Physiol. Behav. 35 Ž1985. 883–890. W.P. Smotherman, J.W. Hennessey, S. Levine, Plasma corticosterone levels during recovery from LiCl produced taste aversions, Behav. Biol. 16 Ž1976. 401–412. A.M. Strack, W.B. Sawyer, J.H. Hughes, K.B. Platt, A.D. Loewy, A general pattern of CNS innervation of the sympathetic outflow demonstrated by transneuronal pseudorabies viral infections, Brain Res. 491 Ž1989. 156–162. A.M. Strack, W.B. Sawyer, K.B. Platt, A.D. Loewy, CNS cell groups regulating outflow to adrenal gland as revealed by transneuronal cell body labeling with pseudorabies virus, Brain Res. 491 Ž1989. 274–296. M.G. Tordoff, P.J. Geiselman, C.V. Grijalva, S.W. Keifer, D. Novin, Amygdaloid lesions impair ingestive responses to 2-deoxyD-glucose but not insulin, Am. J. Physiol. 242 Ž1982. R129–R135. J. Weidenfeld, A.P. Corcos, A. Wohlman, S. Feldman, Characterization of the 2-deoxy-glucose effect on the adrenocortical axis, Endocrinology 134 Ž1994. 1924–1931.