- Email: [email protected]
Hypothalamic lesions increase levels of neuropeptide Y mRNA in the arcuate nucleus of mice
Neuroscience Letters, 165 (1994) 13-17 © 1994 Elsevier Science Ireland Ltd. All rights reserved 0304-3940/94/$ 07.00
13
NSL 10075
Hypothalamic lesions increase levels of neuropeptide Y mRNA in the arcuate nucleus of mice John K. Young ~'*, James C. McKenzie", Linda S. Brady b, Miles Herkenham b "Department of Anatomy, Howard UniversiO,. 520 W Street NW, Washington, DC 20059, USA hSeetion on Functional Neuroanatomy, Clinical Neuroendocrinology Branch, National Institute ~f Mental Health, Bethesda, MD 20892, U~5'A (Received 17 March 1993: Revised version received 2 July 1993: Accepted 27 September 1993)
Key words." Hypothalamus; Neuropeptide Y; Lesion; Goldthioglucose: In situ hybridization: mRNA A recent study demonstrated that hypothalamic lesions induced by goldthioglucose (GTG) in mice produced an increase in neuronal imm unoreactivity for neuropeptide Y (NPY) in the hypothalamic arcuate nucleus. Since NPY is a potent stimulator of feeding, this increase represented a potential explanation for the hyperphagia seen after GTG lesions. To examine whether or not this increase in NPY immunoreactivity was ~lccompanied by an increase in the mRNA for NPY, in situ hybridization histochemistry tbr NPY mRNA in control and in lesioned mice was performed. A 47% increase in NPY mRNA levels in the arcuate nucleus was observed in lesioned mice compared with controls, suggesting that an increased expression of the gene for NPY contributes to elevations in hypothalamic NPY after lesioning. This elevation in NPY may, in turn. relate to mechanisms provoking hyperphagia.
Goldthioglucose (GTG), a toxic analogue of glucose, has long been known to induce hypothalamic damage and hyperphagia in rodents. At least three aspects of this ability of GTG nevertheless remain controversial: (1) by what means does GTG provoke damage and why is hypothalamic damage restricted to the area bordering the dorsal margin of the arcuate nucleus? (2) what types of cells are affected by GTG lesions? and (3) how does GTG damage induce hyperphagia? As regards the first question, there is evidence for a role of glia in the localization of GTG damage. Hemecontaining glia are numerous in the region damaged by GTG and drugs that impair glial function diminish effects of GTG [29]; release of heine from these glia after uptake of the gold in the GTG molecule could explain the neuronal death and gliosis within the lesion [21, 29]. Neurons immunoreactive for tyrosine hydroxylase (TH) are adjacent to the focus of damage and are one type of cell probably damaged [32]. Also, TH+ neurons are in intimate contact with the heine-rich glia that may mediate effects of GTG [31]. Neurons in more ventral portions of the arcuate nucleus do not appear damaged, but may undergo functional or morphological changes. One such functional change that could relate to the appearance of hyperphagia is an enhanced immunoreac*Corresponding author.
tivity for neuropeptide Y (NPY) after a GTG lesion [32]. NPY is a well-known stimulator of feeding in mice and rats [3, 17]. Also, the hyperphagia resulting from fasting or experimental diabetes is associated with an increased NPY production in the arcuate nucleus [3, 4, 25]. Genetically hyperphagic Zucker rats have an elevated production of hypothalamic NPY [23]. Finally, an agent that produces hypophagia - cobalt protoporphyrin, which appears to disrupt the function of heine proteins in the paraventricular nucleus- also decreases the ability of NPY to stimulate feeding [9]. An increase in neuronal NPY immunoreactivity after hypothalamic lesions thus suggests a possible mechanism for lesion-induced hyperphagia. However, such an increase could merely reflect a decreased transport of NPY away from cell somas or a decreased degradation of NPY after a lesion [32]. This study was performed to determine if changes in the mRNA of NPY in arcuate cells accompanies the increase in the NPY peptide after a lesion. Twelve 2-month-old male mice (CD1 strain) were obtained from Charles River. CD1 mice were employed because CD1 mice display consistent alterations in hypothalamic NPY in response to challenges like lasting or refeeding [4]. Mice were housed in group cages at 24°C and given free access to Purina chow pellets and water. The time course of the experiment was designed with
14
reference to data on the onset and duration oF GTGinduced hyperphagia. Food intake after GTG is actually lowered for the first 6 days, perhaps due to generalized toxic effects of GTG and to malaisse [6, 28]. Intake then rises to reach a level - 100% above normal by day 12 after GTG and then gradually falls to 25% above normal by day 18 [6]. The only other study to measure intake after GTG in mice reported that food intake was still significantly elevated (by 14%) even 10 wk after lesioning [2]. This time course is similar to VMH lesion-induced hyperphagia and obesity in rats+ which is maximal up until 10 wk but still persistant, if reduced, for many weeks later on. The reasons for the decline in hyperphagia after lesioning are uncertain; feeding-inhibitory signals from the gastrointestinal tract after chronic overeating [11] or a denervation hypersensitivity of hypothalamic neurons deafferented by the lesion may be involved. This experiment, therefore, was designed to examine hypothalamic NPY mRNA at a time point when lesioninduced hyperphagia would be maximal. On day 1, six mice were injected i.p. 0.35 mg/g GTG (30 mg/ml saline) at 11:00: the remaining mice were given saline injections [32]. Mice were sacrificed on day 13 via a lethal injection of chloral hydrate, unfixed brains were removed, blocked by making coronal slices just anterior and posterior to the hypothalamus, quickly frozen in powdered dry ice and wrapped in aluminum foil for storage at -40°C. Over the following week, frozen 15-/,tm-thick sections were cut coronally through the midportion of the arcuate nucleus, thaw-mounted onto gelatin-coated slides, dried and stored at -40 ° C. For in situ hybridization, a synthetic oligonucleotide probe directed against human/rat NPY bases 171-218 [3] was labeled with [~-35S]dATP (spec. act. 1000 Ci/mmol; New England Nuclear) at the 3' end using terminal deoxynucleotidyl transferase (25 U//A; BoehringerMannheim Biochemicals) and tailing buffer (Bethesda Research Laboratory). The high degree of homology of the NPY gene in humans, rats and many other vertebrate species permits the detection of NPY peptide and mRNA with probes directed against the human molecules in brain and peripheral tissues of mice [4, 8]. Slide-mounted sections were processed for in situ hybridization histochemistry as previously described [3], apposed to film (Hyperfilm-flmax, Amersham) for 5 days and developed (DI9, KQdak) for 5 snin at 20°~. •Autoradiographic film images of brain sections and standards of known radioactivity (American Radiolabeled Chemicals) were quantified as previously desct;ibed to convert light transmittance to disintegrations/rain (dpm)/mg tissue [3]. Statistical significance between control and experimental o~,~,n~ wa~ determined bv one-way ANOVA followed
Fig. 1. A: arcuate nucleus of a control mouse, stained for iron and counterstained with thionin. B: arcuate nucleus of a GTG-lesioned mouse (stained as in A). Note iron deposits (arrows) at the lateral margins of the nucleus and smaller overall size of the nucleus compared with A ( 1 4 0 x ) .
by Dunnett's two-tailed t test. To examine the cellular localization of probe label, slides were dipped in NTB-2 nuclear track emulsion for autoradiography, developed and counterstained with Cresyl violet [3]. To confirm the presence of lesions in the GTG group, additional sections from both groups were stained for iron to detect iron released from damaged blood vessels and counterstained with Toluidine blue [32]. Also, crosssectional areas of the arcuate nuclei from lesioned and control brains were measured by projecting an image of Cresyl violet-stained sections onto paper at a magnification of 44 x, tracing the outlines of the arcuate nuclei bilaterally and measuring the traced areas using a digitizing tablet, an Apple lie computer and BioQuant software for digitizing morphometry [29]. No mortality after GTG was observed. Analysis of the slides showed the presence of lesions in all five GTG brains examined, as indicated by iron deposited during
15
Fig. 2. A: autoradiograph of the hypothalamus of a control mouse after in situ hybridization for NPY m R N A , viewed using dark-field microscopy ( 120 x ). The signal value obtained by film densitometry for this mouse q1727.5 d p m / m g protein) was close to the mean [or this group. B: bright-field view of cells labeled for NPY m R N A in the section in A (500 x ). C: autoradiograph of a G T G mouse after in situ hybridization for NPY m R N A . The value obtained by tilm densitometry f\3r this mouse (3014.2 d p m / m g proteinl was close to the overall mean for this group. D: bright-ticld view of cells labeled lk~r N PY m R N A in tile section in C. Note lhe higher density of silver grains in this section compared with B.
lesion formation [32]. One brain from a G T G - t r e a t e d mouse was accidentally thawed during an unscheduled cryostat defrosting cycle and was discarded. The arcuate nuclei of the lesioned group had a mean cross-sectional area that was 42% smaller than that of controls: 0.17 _+ 0.07 square ram, G T G group, vs. 0.29 + 0.07 square mm, control group, .? + S.D., (P < 0.025, t test). This reduction in size appeared mainly due to a decrease in the dorsal area of the nucleus (Fig. I). Collapse of the lesion into a cell dense scar had been accompanied by a distorted ventricular profile in the previous study of brains fixed by perfusion [32]. Here, little overall distortion of hypothalamic anatomy was apparent, suggesting that such distortions had previously resulted from adhesions between the two sides of the hypothalamus that caused focal resistance to the dilatation of the ventricle during perfusion. Film densitometry showed an increased expression of the N P Y gene in GTG-lesioned mice (.,7:_+ S.D.): con-
trols = 1839 + 400 (n = 6), G T G = 2721 _+ 583 dpm/mg tissue (n = 5), amounting to a 47.9% increase over control levels, which is comparable to effects of food restriction [3, 4]. ANOVA indicated this difference is statistically significant at a level of 0.0157 (F~.~ = 8.824). Dunnet's two-tailed test confirmed significance at a level of P = 0.05. Counts of labeled cells after dipping in NTB-2 emulsion revealed similar numbers of N P Y + cells/nucleus in both groups: control = 40.8 + 9.6 cells/nucleus, G T G = 41.8 + 16.1 cells/nucleus (.i: _+ S.D.) (N.S., no significant difference, t test). Most labeled cells were found in the ventral division of the arcuate nucleus in both groups: control mice had a few N P Y + cells near the dorsal margin of the nucleus, whereas in the G T G mice, these scattered cells were absent, probably due to the lesion-induced damage to the dorsal arcuate area (Fig. 2). The general distribution of labeled cells was identical to other reports of NPY immunoreactivity or m R N A probe la-
16 beling [4, 5, 10, 15, 18, 19, 32], supporting the validity of the probe used in the present study and in a previous study [3]. The density of silver grains over cells appeared higher in the G T G group, which likely accounted for the higher values determined by film densitometry (Fig. 2). Several reasons can be proposed to account for this lesion-induced increase in N P Y m R N A . Lesions produce hyperinsulinemia and decreased gonadal function [24, 27] that could perhaps affect hypothalamic levels of NPY. However, hyperinsulinemia and castration decrease, rather than increase, N P Y production [22, 25], so lesion-induced endocrine changes distinct from these must be sought to explain changes in N P Y m R N A . Another explanation is that the lesion disrupted an inhibitory afferent input to N P Y neurons. T H + neurons in the lateral portions of the arcuate nucleus innervate N P Y + neurons and inhibit them [15, 18, 19, 26, 32]. Destruction of this input by G T G is thus one plausible explanation for effects of G T G upon N P Y and food intake. Lesions, of course, could also alter other inputs or cause morphological alterations in the N P Y neurons themselves. An elevation in N P Y m R N A could result from enhanced transcription or an altered posttranscriptional m R N A stability [14]. Finally, Magni and Barnea [14] have evidence for a role for glia in the control of hypothalamic neuropeptides, which may be relevent to both the normal control of feeding and to lesion-induced hyperphagia. This study does not prove a causal relationship between an elevation in N P Y m R N A and hyperphagia. One approach to such a p r o o f would be to measure intake and N P Y m R N A at varying times after a lesion. Such a study, however, would be correlative rather than conclusive and could be complicated by changes in N P Y m R N A in the hours after measuring intake and also by the difficulty in estimating daily intake and food spillage in such small animals [2, 28]. Another approach would be to see if agents that depress N P Y also depress G T G induced hyperphagia. Estrogen, for example, depresses N P Y production [reviewed in 30] and, while not blocking the lesioning ability of G T G , does restrain feeding in GTG-lesioned mice [2, 28]. Adrenalectomy may or may not depress N P Y in nonfasted rats (5 vs. 13) but does appear to depress the elevation in N P Y induced by fasting [20]. Adrenalectomy also prevents the appearance of hyperphagia in GTG-lesioned mice [7]. The ability of adrenalectomy to exert these effects m a y relate to glucocorticoid-response elements upstream of the N P Y gene [16]. Alternatively, the general metabolic state of a cell may affect N P Y m R N A transcription or stability since agents that depress NPY, like hyperinsulinemia, adrenalectomy or estrogen, all also enhance glucose uptake in nenrnl tissues [1. 12, 25. 281. More investigation of these
topics will be needed to determine if there is a relationship between lesion-induced hyperphagia and hypothalamic NPY. Supported by a H o w a r d University Faculty Research Support Grant. 1 Bishop, J. and Simpkins, J.W., Role of estrogens in peripher~il and cerebral glucose utilization, Rev. Neurosci., 3 (1992) 121 137. 2 Blaustein, J.D., Gentry, R.T., Roy, E.J. and Wade, G.N., Effectsof ovariectomy and estradiol on body weight and food intake in goldthioglucose-treatedmice,Physiol. Behav., 17(1976) 1027 t030. 3 Brady, L.S., Smith, M.A., Gold, P.W. and Herkenham, M., Altered expression of hypothalamic neuropeptide mRNAs in food-restricted and food-deprived rats, Neuroendocrinotogy, 52 (1990) 441-447. 4 Chua, Jr., S.C., Brown, A.W., Kim, J., Hennesey, K.L., Leibel, R.L. and Hirsch, J., Food deprivation and hypothalamic neuropeptide gene expression: effects of strain background and the diabetes mutation, Mol. Brain. Res., 11 (1991) 291-299. 5 Dean, R.G. and White, B.D., Neuropeptide Y expression in rat brain: effects of adrenalectomy, Neurosci. Lett., 114 (1990) 339 344. 6 Debons, A.F., Siclari, E., Das, K.C. and Fuhr, B., Gold thioglucose-induced hypothalamic damage, hyperphagia and obesity: dependence on the adrenal gland, Endocrinology, 110 (1982) 20242029. 7 Debons, A.F., Zurek, Tse, C.S. and Abrahamson, S., Central nervous system control of hyperphagia in hypothalamic obesity: dependence on adrenal corticoids, Endocrinology, 1t 8 (1986) 16781681. 8 Ericsson, A., Schalling, M., Mclntyre, K.R., Landsberg, J.M., Larhammar, D., Seroogi, K., Hokfelt, T. and Persson, H., Detection of neuropeptide Y and its mRNA in megakaryocytes:enhanced levels in certain autoimmune mice, Proc. Natl. Acad. Sci. USA, 84 (1987) 5585 5589. 9 Galbraith, R.A., Chua, Jr., S.C. and Kappas, A., Hypothalamic mechanisms for cobalt protoporphyrin-induced hypophagia and weight loss: inhibition of the feeding response to NPY, Mol. Brain Res., 15 (1992) 298-302. 10 Gehlert, D.R., Chronwall, B.M., Schafer, M.P. and O'Donohue, T.L., Localization of neuropeptide Y messenger ribonucleic acid in rat and mouse brain by in situ hybridization, Synapse, 1 (1987) 25 -31. 11 Hallonquist, J.D. and Brandes, J.S., Ventromedial hypothalamic lesions and weight gain in rats absenceof a Static phase, Physiol. Behav., 27 (1981) 709-713. 12 Kadekaro, M., Ito, M. and Gross, RM., Loc'a~cerebral glucose utilization is increased in acutely adrenalectomized rats, Neuroendocrinology, 47 (1988) 329-334. 13 Larsen, RJ., Mikkelsen, J.D., Jessop, D.S., Lightman, S.L. and Chowdrey, H.S., Neuropeptide Y mRNA and immunoreacti~ityin hypothalamic neuroendocrine neurons: effects of adrenal~ctomy and chronic osmotic stimulation, J. Neurosci., 13 (t 993) 1138211'47. 14 Magni, R and Barnea, A., Forskolin and phorbol ester stimul~ttion of neuropeptide Y (NPY) production and secretion by aggregating fetal brain cells in culture: evidence for regulation of NPY biosynthesis at transcriptional and posttranscriptional levels, Endocrinology, 130 (1992) 976-984. 15 Meister, B., Ceccatelli, S., Hokfelt, T., Anden, N.-E., Anden, M. and Theodorsson, E., Neurotransmitter, neuropeptides, and bind-
17
16
17
18 19
20
21 22
23
ing sites in the rat mediobasal hypothatamus: effects of monosodium glutamate lesions, Exp. Brain Res., 76 (1989) 343 368. Misaki, N., Higuchi, H., Yamagata, K. and Miki, N., Identification of glucocorticoid responsive elements (GRE's) at far upstream of rat NPY gene, Neurochem. Int., 21 (1992) 185 189. Morley, J.E., Hernandez, E.N. and Flood, J.F., Neuropeptide Y increases |\)od intake in mice, Am. J. Physiol., 253 (1987) R516 R522. Pelletier, G., Ultrastructural localization of neuropeptide Y in the hypothalamus, Ann. N.Y. Acad. Sci., 611 (1990) 232 246. Pelletier, G. and Simard, J., Dopaminergic regulation of preproneuropeptide Y levels in the rat arcuate nucleus, Neurosci. Lett., 127 (1991) 96 98. Ponsalle, P., Srivastava, L.S., Uht, R.M. and White, J.D., Glucocorticoids are required for food deprivation-induced increases in hypothalamic neuropeptide Y expression. J. Neuroendocr.. 4 ( 1992t 585 591. Regan, R.F. and Panter, S.S., Neurotoxicity of hemoglobin in cortical cell culture, Neurosci. Lett., 153 (1993) 219 222. Sahu, A., Kalra, S.E, Crowley, W.R., O'Donohue, T.L. and Kalra, ES., Neuropeptide Y levels in microdissected regions of the bypothalamus and in vitro release in response to KC1 and prostaglandin E2. Effects of castration, Endocrinology, 120 (1987) 1831 1837. Sanacora, G., Kershaw, M., Finkelstein, J.A. and White, J.D., Increased hypothalamic content of preproneuropeptide Y messenger ribonucleic acid in genetically obese Zucker rats and its regulation by food deprivation, Endocrinology, 127 (1990) 730 737.
24 Scallet, A.C. and Olney, J.W., Components of hypothalamic obesity: bipiperidyl mustard lesions add hyperphagia to monosodium glutamate induced hyperinsulinemia, Brain Res., 374 (1986) 380 384. 25 Schwartz, M.W., Figlewicz, D.E, Baskin, D.G., Woods, S.C. and Porte, Jr., D., Insulin in the brain: a hormonal regulator of energy balance, Endocr. Rev., 13 (1992) 387 414. 26 Smialowska, M. and Leguto, B., Haloperidol-induced increase in neuropeptide Y immunoreactivity in the locus coeruleus of the rat brain, Neuroscience, 47 (1992) 351 355. 27 Wright, P. and Turner, C., Sex differences in body weight following gonadectomy and goldthioglucose injections in mice, Physiol. Behav.,ll(1973) 155 159. 28 Young, J.K., Nance, D.M. and Gorski, R.A., Effects of estrogen upon hypothalamic vulnerability to goldthioglucose in mice, Brain Res. Bull., 3 (1978) 231 235. 29 Young, J.K., The glial drug methionine sulfoximine reduces goldthioglucose lesions in mice, Brain Res. Bull., 22 (1989) 929 936. 30 Young, J.K., Estrogen and the etiology of anorexia nervosa, Neurosci. Biobehav. Rev., 15 (199l) 327 331. 31 Young, J.K., McKenzie, J.C. and Baker, J.H., Association of ironcontaining astrocytes with dopaminergic neurons of the arcuate nucleus, J. Neurosci. Res., 25 (1990) 204 213. 32 Young, J.K., Hypothalamic lesions increase neuronal immunoreactivity for neuropeptide Y, Brain Res. Bull., 29 (1992) 375 380.
13
NSL 10075
Hypothalamic lesions increase levels of neuropeptide Y mRNA in the arcuate nucleus of mice John K. Young ~'*, James C. McKenzie", Linda S. Brady b, Miles Herkenham b "Department of Anatomy, Howard UniversiO,. 520 W Street NW, Washington, DC 20059, USA hSeetion on Functional Neuroanatomy, Clinical Neuroendocrinology Branch, National Institute ~f Mental Health, Bethesda, MD 20892, U~5'A (Received 17 March 1993: Revised version received 2 July 1993: Accepted 27 September 1993)
Key words." Hypothalamus; Neuropeptide Y; Lesion; Goldthioglucose: In situ hybridization: mRNA A recent study demonstrated that hypothalamic lesions induced by goldthioglucose (GTG) in mice produced an increase in neuronal imm unoreactivity for neuropeptide Y (NPY) in the hypothalamic arcuate nucleus. Since NPY is a potent stimulator of feeding, this increase represented a potential explanation for the hyperphagia seen after GTG lesions. To examine whether or not this increase in NPY immunoreactivity was ~lccompanied by an increase in the mRNA for NPY, in situ hybridization histochemistry tbr NPY mRNA in control and in lesioned mice was performed. A 47% increase in NPY mRNA levels in the arcuate nucleus was observed in lesioned mice compared with controls, suggesting that an increased expression of the gene for NPY contributes to elevations in hypothalamic NPY after lesioning. This elevation in NPY may, in turn. relate to mechanisms provoking hyperphagia.
Goldthioglucose (GTG), a toxic analogue of glucose, has long been known to induce hypothalamic damage and hyperphagia in rodents. At least three aspects of this ability of GTG nevertheless remain controversial: (1) by what means does GTG provoke damage and why is hypothalamic damage restricted to the area bordering the dorsal margin of the arcuate nucleus? (2) what types of cells are affected by GTG lesions? and (3) how does GTG damage induce hyperphagia? As regards the first question, there is evidence for a role of glia in the localization of GTG damage. Hemecontaining glia are numerous in the region damaged by GTG and drugs that impair glial function diminish effects of GTG [29]; release of heine from these glia after uptake of the gold in the GTG molecule could explain the neuronal death and gliosis within the lesion [21, 29]. Neurons immunoreactive for tyrosine hydroxylase (TH) are adjacent to the focus of damage and are one type of cell probably damaged [32]. Also, TH+ neurons are in intimate contact with the heine-rich glia that may mediate effects of GTG [31]. Neurons in more ventral portions of the arcuate nucleus do not appear damaged, but may undergo functional or morphological changes. One such functional change that could relate to the appearance of hyperphagia is an enhanced immunoreac*Corresponding author.
tivity for neuropeptide Y (NPY) after a GTG lesion [32]. NPY is a well-known stimulator of feeding in mice and rats [3, 17]. Also, the hyperphagia resulting from fasting or experimental diabetes is associated with an increased NPY production in the arcuate nucleus [3, 4, 25]. Genetically hyperphagic Zucker rats have an elevated production of hypothalamic NPY [23]. Finally, an agent that produces hypophagia - cobalt protoporphyrin, which appears to disrupt the function of heine proteins in the paraventricular nucleus- also decreases the ability of NPY to stimulate feeding [9]. An increase in neuronal NPY immunoreactivity after hypothalamic lesions thus suggests a possible mechanism for lesion-induced hyperphagia. However, such an increase could merely reflect a decreased transport of NPY away from cell somas or a decreased degradation of NPY after a lesion [32]. This study was performed to determine if changes in the mRNA of NPY in arcuate cells accompanies the increase in the NPY peptide after a lesion. Twelve 2-month-old male mice (CD1 strain) were obtained from Charles River. CD1 mice were employed because CD1 mice display consistent alterations in hypothalamic NPY in response to challenges like lasting or refeeding [4]. Mice were housed in group cages at 24°C and given free access to Purina chow pellets and water. The time course of the experiment was designed with
14
reference to data on the onset and duration oF GTGinduced hyperphagia. Food intake after GTG is actually lowered for the first 6 days, perhaps due to generalized toxic effects of GTG and to malaisse [6, 28]. Intake then rises to reach a level - 100% above normal by day 12 after GTG and then gradually falls to 25% above normal by day 18 [6]. The only other study to measure intake after GTG in mice reported that food intake was still significantly elevated (by 14%) even 10 wk after lesioning [2]. This time course is similar to VMH lesion-induced hyperphagia and obesity in rats+ which is maximal up until 10 wk but still persistant, if reduced, for many weeks later on. The reasons for the decline in hyperphagia after lesioning are uncertain; feeding-inhibitory signals from the gastrointestinal tract after chronic overeating [11] or a denervation hypersensitivity of hypothalamic neurons deafferented by the lesion may be involved. This experiment, therefore, was designed to examine hypothalamic NPY mRNA at a time point when lesioninduced hyperphagia would be maximal. On day 1, six mice were injected i.p. 0.35 mg/g GTG (30 mg/ml saline) at 11:00: the remaining mice were given saline injections [32]. Mice were sacrificed on day 13 via a lethal injection of chloral hydrate, unfixed brains were removed, blocked by making coronal slices just anterior and posterior to the hypothalamus, quickly frozen in powdered dry ice and wrapped in aluminum foil for storage at -40°C. Over the following week, frozen 15-/,tm-thick sections were cut coronally through the midportion of the arcuate nucleus, thaw-mounted onto gelatin-coated slides, dried and stored at -40 ° C. For in situ hybridization, a synthetic oligonucleotide probe directed against human/rat NPY bases 171-218 [3] was labeled with [~-35S]dATP (spec. act. 1000 Ci/mmol; New England Nuclear) at the 3' end using terminal deoxynucleotidyl transferase (25 U//A; BoehringerMannheim Biochemicals) and tailing buffer (Bethesda Research Laboratory). The high degree of homology of the NPY gene in humans, rats and many other vertebrate species permits the detection of NPY peptide and mRNA with probes directed against the human molecules in brain and peripheral tissues of mice [4, 8]. Slide-mounted sections were processed for in situ hybridization histochemistry as previously described [3], apposed to film (Hyperfilm-flmax, Amersham) for 5 days and developed (DI9, KQdak) for 5 snin at 20°~. •Autoradiographic film images of brain sections and standards of known radioactivity (American Radiolabeled Chemicals) were quantified as previously desct;ibed to convert light transmittance to disintegrations/rain (dpm)/mg tissue [3]. Statistical significance between control and experimental o~,~,n~ wa~ determined bv one-way ANOVA followed
Fig. 1. A: arcuate nucleus of a control mouse, stained for iron and counterstained with thionin. B: arcuate nucleus of a GTG-lesioned mouse (stained as in A). Note iron deposits (arrows) at the lateral margins of the nucleus and smaller overall size of the nucleus compared with A ( 1 4 0 x ) .
by Dunnett's two-tailed t test. To examine the cellular localization of probe label, slides were dipped in NTB-2 nuclear track emulsion for autoradiography, developed and counterstained with Cresyl violet [3]. To confirm the presence of lesions in the GTG group, additional sections from both groups were stained for iron to detect iron released from damaged blood vessels and counterstained with Toluidine blue [32]. Also, crosssectional areas of the arcuate nuclei from lesioned and control brains were measured by projecting an image of Cresyl violet-stained sections onto paper at a magnification of 44 x, tracing the outlines of the arcuate nuclei bilaterally and measuring the traced areas using a digitizing tablet, an Apple lie computer and BioQuant software for digitizing morphometry [29]. No mortality after GTG was observed. Analysis of the slides showed the presence of lesions in all five GTG brains examined, as indicated by iron deposited during
15
Fig. 2. A: autoradiograph of the hypothalamus of a control mouse after in situ hybridization for NPY m R N A , viewed using dark-field microscopy ( 120 x ). The signal value obtained by film densitometry for this mouse q1727.5 d p m / m g protein) was close to the mean [or this group. B: bright-field view of cells labeled for NPY m R N A in the section in A (500 x ). C: autoradiograph of a G T G mouse after in situ hybridization for NPY m R N A . The value obtained by tilm densitometry f\3r this mouse (3014.2 d p m / m g proteinl was close to the overall mean for this group. D: bright-ticld view of cells labeled lk~r N PY m R N A in tile section in C. Note lhe higher density of silver grains in this section compared with B.
lesion formation [32]. One brain from a G T G - t r e a t e d mouse was accidentally thawed during an unscheduled cryostat defrosting cycle and was discarded. The arcuate nuclei of the lesioned group had a mean cross-sectional area that was 42% smaller than that of controls: 0.17 _+ 0.07 square ram, G T G group, vs. 0.29 + 0.07 square mm, control group, .? + S.D., (P < 0.025, t test). This reduction in size appeared mainly due to a decrease in the dorsal area of the nucleus (Fig. I). Collapse of the lesion into a cell dense scar had been accompanied by a distorted ventricular profile in the previous study of brains fixed by perfusion [32]. Here, little overall distortion of hypothalamic anatomy was apparent, suggesting that such distortions had previously resulted from adhesions between the two sides of the hypothalamus that caused focal resistance to the dilatation of the ventricle during perfusion. Film densitometry showed an increased expression of the N P Y gene in GTG-lesioned mice (.,7:_+ S.D.): con-
trols = 1839 + 400 (n = 6), G T G = 2721 _+ 583 dpm/mg tissue (n = 5), amounting to a 47.9% increase over control levels, which is comparable to effects of food restriction [3, 4]. ANOVA indicated this difference is statistically significant at a level of 0.0157 (F~.~ = 8.824). Dunnet's two-tailed test confirmed significance at a level of P = 0.05. Counts of labeled cells after dipping in NTB-2 emulsion revealed similar numbers of N P Y + cells/nucleus in both groups: control = 40.8 + 9.6 cells/nucleus, G T G = 41.8 + 16.1 cells/nucleus (.i: _+ S.D.) (N.S., no significant difference, t test). Most labeled cells were found in the ventral division of the arcuate nucleus in both groups: control mice had a few N P Y + cells near the dorsal margin of the nucleus, whereas in the G T G mice, these scattered cells were absent, probably due to the lesion-induced damage to the dorsal arcuate area (Fig. 2). The general distribution of labeled cells was identical to other reports of NPY immunoreactivity or m R N A probe la-
16 beling [4, 5, 10, 15, 18, 19, 32], supporting the validity of the probe used in the present study and in a previous study [3]. The density of silver grains over cells appeared higher in the G T G group, which likely accounted for the higher values determined by film densitometry (Fig. 2). Several reasons can be proposed to account for this lesion-induced increase in N P Y m R N A . Lesions produce hyperinsulinemia and decreased gonadal function [24, 27] that could perhaps affect hypothalamic levels of NPY. However, hyperinsulinemia and castration decrease, rather than increase, N P Y production [22, 25], so lesion-induced endocrine changes distinct from these must be sought to explain changes in N P Y m R N A . Another explanation is that the lesion disrupted an inhibitory afferent input to N P Y neurons. T H + neurons in the lateral portions of the arcuate nucleus innervate N P Y + neurons and inhibit them [15, 18, 19, 26, 32]. Destruction of this input by G T G is thus one plausible explanation for effects of G T G upon N P Y and food intake. Lesions, of course, could also alter other inputs or cause morphological alterations in the N P Y neurons themselves. An elevation in N P Y m R N A could result from enhanced transcription or an altered posttranscriptional m R N A stability [14]. Finally, Magni and Barnea [14] have evidence for a role for glia in the control of hypothalamic neuropeptides, which may be relevent to both the normal control of feeding and to lesion-induced hyperphagia. This study does not prove a causal relationship between an elevation in N P Y m R N A and hyperphagia. One approach to such a p r o o f would be to measure intake and N P Y m R N A at varying times after a lesion. Such a study, however, would be correlative rather than conclusive and could be complicated by changes in N P Y m R N A in the hours after measuring intake and also by the difficulty in estimating daily intake and food spillage in such small animals [2, 28]. Another approach would be to see if agents that depress N P Y also depress G T G induced hyperphagia. Estrogen, for example, depresses N P Y production [reviewed in 30] and, while not blocking the lesioning ability of G T G , does restrain feeding in GTG-lesioned mice [2, 28]. Adrenalectomy may or may not depress N P Y in nonfasted rats (5 vs. 13) but does appear to depress the elevation in N P Y induced by fasting [20]. Adrenalectomy also prevents the appearance of hyperphagia in GTG-lesioned mice [7]. The ability of adrenalectomy to exert these effects m a y relate to glucocorticoid-response elements upstream of the N P Y gene [16]. Alternatively, the general metabolic state of a cell may affect N P Y m R N A transcription or stability since agents that depress NPY, like hyperinsulinemia, adrenalectomy or estrogen, all also enhance glucose uptake in nenrnl tissues [1. 12, 25. 281. More investigation of these
topics will be needed to determine if there is a relationship between lesion-induced hyperphagia and hypothalamic NPY. Supported by a H o w a r d University Faculty Research Support Grant. 1 Bishop, J. and Simpkins, J.W., Role of estrogens in peripher~il and cerebral glucose utilization, Rev. Neurosci., 3 (1992) 121 137. 2 Blaustein, J.D., Gentry, R.T., Roy, E.J. and Wade, G.N., Effectsof ovariectomy and estradiol on body weight and food intake in goldthioglucose-treatedmice,Physiol. Behav., 17(1976) 1027 t030. 3 Brady, L.S., Smith, M.A., Gold, P.W. and Herkenham, M., Altered expression of hypothalamic neuropeptide mRNAs in food-restricted and food-deprived rats, Neuroendocrinotogy, 52 (1990) 441-447. 4 Chua, Jr., S.C., Brown, A.W., Kim, J., Hennesey, K.L., Leibel, R.L. and Hirsch, J., Food deprivation and hypothalamic neuropeptide gene expression: effects of strain background and the diabetes mutation, Mol. Brain. Res., 11 (1991) 291-299. 5 Dean, R.G. and White, B.D., Neuropeptide Y expression in rat brain: effects of adrenalectomy, Neurosci. Lett., 114 (1990) 339 344. 6 Debons, A.F., Siclari, E., Das, K.C. and Fuhr, B., Gold thioglucose-induced hypothalamic damage, hyperphagia and obesity: dependence on the adrenal gland, Endocrinology, 110 (1982) 20242029. 7 Debons, A.F., Zurek, Tse, C.S. and Abrahamson, S., Central nervous system control of hyperphagia in hypothalamic obesity: dependence on adrenal corticoids, Endocrinology, 1t 8 (1986) 16781681. 8 Ericsson, A., Schalling, M., Mclntyre, K.R., Landsberg, J.M., Larhammar, D., Seroogi, K., Hokfelt, T. and Persson, H., Detection of neuropeptide Y and its mRNA in megakaryocytes:enhanced levels in certain autoimmune mice, Proc. Natl. Acad. Sci. USA, 84 (1987) 5585 5589. 9 Galbraith, R.A., Chua, Jr., S.C. and Kappas, A., Hypothalamic mechanisms for cobalt protoporphyrin-induced hypophagia and weight loss: inhibition of the feeding response to NPY, Mol. Brain Res., 15 (1992) 298-302. 10 Gehlert, D.R., Chronwall, B.M., Schafer, M.P. and O'Donohue, T.L., Localization of neuropeptide Y messenger ribonucleic acid in rat and mouse brain by in situ hybridization, Synapse, 1 (1987) 25 -31. 11 Hallonquist, J.D. and Brandes, J.S., Ventromedial hypothalamic lesions and weight gain in rats absenceof a Static phase, Physiol. Behav., 27 (1981) 709-713. 12 Kadekaro, M., Ito, M. and Gross, RM., Loc'a~cerebral glucose utilization is increased in acutely adrenalectomized rats, Neuroendocrinology, 47 (1988) 329-334. 13 Larsen, RJ., Mikkelsen, J.D., Jessop, D.S., Lightman, S.L. and Chowdrey, H.S., Neuropeptide Y mRNA and immunoreacti~ityin hypothalamic neuroendocrine neurons: effects of adrenal~ctomy and chronic osmotic stimulation, J. Neurosci., 13 (t 993) 1138211'47. 14 Magni, R and Barnea, A., Forskolin and phorbol ester stimul~ttion of neuropeptide Y (NPY) production and secretion by aggregating fetal brain cells in culture: evidence for regulation of NPY biosynthesis at transcriptional and posttranscriptional levels, Endocrinology, 130 (1992) 976-984. 15 Meister, B., Ceccatelli, S., Hokfelt, T., Anden, N.-E., Anden, M. and Theodorsson, E., Neurotransmitter, neuropeptides, and bind-
17
16
17
18 19
20
21 22
23
ing sites in the rat mediobasal hypothatamus: effects of monosodium glutamate lesions, Exp. Brain Res., 76 (1989) 343 368. Misaki, N., Higuchi, H., Yamagata, K. and Miki, N., Identification of glucocorticoid responsive elements (GRE's) at far upstream of rat NPY gene, Neurochem. Int., 21 (1992) 185 189. Morley, J.E., Hernandez, E.N. and Flood, J.F., Neuropeptide Y increases |\)od intake in mice, Am. J. Physiol., 253 (1987) R516 R522. Pelletier, G., Ultrastructural localization of neuropeptide Y in the hypothalamus, Ann. N.Y. Acad. Sci., 611 (1990) 232 246. Pelletier, G. and Simard, J., Dopaminergic regulation of preproneuropeptide Y levels in the rat arcuate nucleus, Neurosci. Lett., 127 (1991) 96 98. Ponsalle, P., Srivastava, L.S., Uht, R.M. and White, J.D., Glucocorticoids are required for food deprivation-induced increases in hypothalamic neuropeptide Y expression. J. Neuroendocr.. 4 ( 1992t 585 591. Regan, R.F. and Panter, S.S., Neurotoxicity of hemoglobin in cortical cell culture, Neurosci. Lett., 153 (1993) 219 222. Sahu, A., Kalra, S.E, Crowley, W.R., O'Donohue, T.L. and Kalra, ES., Neuropeptide Y levels in microdissected regions of the bypothalamus and in vitro release in response to KC1 and prostaglandin E2. Effects of castration, Endocrinology, 120 (1987) 1831 1837. Sanacora, G., Kershaw, M., Finkelstein, J.A. and White, J.D., Increased hypothalamic content of preproneuropeptide Y messenger ribonucleic acid in genetically obese Zucker rats and its regulation by food deprivation, Endocrinology, 127 (1990) 730 737.
24 Scallet, A.C. and Olney, J.W., Components of hypothalamic obesity: bipiperidyl mustard lesions add hyperphagia to monosodium glutamate induced hyperinsulinemia, Brain Res., 374 (1986) 380 384. 25 Schwartz, M.W., Figlewicz, D.E, Baskin, D.G., Woods, S.C. and Porte, Jr., D., Insulin in the brain: a hormonal regulator of energy balance, Endocr. Rev., 13 (1992) 387 414. 26 Smialowska, M. and Leguto, B., Haloperidol-induced increase in neuropeptide Y immunoreactivity in the locus coeruleus of the rat brain, Neuroscience, 47 (1992) 351 355. 27 Wright, P. and Turner, C., Sex differences in body weight following gonadectomy and goldthioglucose injections in mice, Physiol. Behav.,ll(1973) 155 159. 28 Young, J.K., Nance, D.M. and Gorski, R.A., Effects of estrogen upon hypothalamic vulnerability to goldthioglucose in mice, Brain Res. Bull., 3 (1978) 231 235. 29 Young, J.K., The glial drug methionine sulfoximine reduces goldthioglucose lesions in mice, Brain Res. Bull., 22 (1989) 929 936. 30 Young, J.K., Estrogen and the etiology of anorexia nervosa, Neurosci. Biobehav. Rev., 15 (199l) 327 331. 31 Young, J.K., McKenzie, J.C. and Baker, J.H., Association of ironcontaining astrocytes with dopaminergic neurons of the arcuate nucleus, J. Neurosci. Res., 25 (1990) 204 213. 32 Young, J.K., Hypothalamic lesions increase neuronal immunoreactivity for neuropeptide Y, Brain Res. Bull., 29 (1992) 375 380.