Metabolic effects of galanin injections into the paraventricular nucleus of the hypothalamus

Peptides,Vol. 13, pp. 323-327, 1992

0196-9781/92 $5.00 + .00 Copyright© 1992 PergamonPressLtd.

Printed in the USA.

Metabolic Effects of Galanin Injections Into the Paraventricular Nucleus of the Hypothalamus JOSI~ A. M E N I ~ N D E Z , * D A L E M. A T R E N S * A N D S A R A H F. L E I B O W I T Z t

*Department of Psychology, University of Sydney, N S W 2006, Australia -[Rockefeller University, New York, N Y 10021 R e c e i v e d 18 J u n e 1991 MENI~NDEZ, J. A., D. M. ATRENS AND S. F. LEIBOWlTZ. Metabolic effects ofgalanin injections into the paraventricular nucleus of the hypothalamus. PEPTIDES 13(2) 323-327, 1992.--The metabolic effects of single injections of galanin into the paraventricular nucleus of the hypothalamus (PVN) were investigated in an open-circuit calorimeter. Wistar rats were tested, with no food available during the tests. In the dose range of 0.03-0.3 nmol, galanin produced a very short-latency (approximately 2 minutes) and short-lasting (approximately 15 minutes) reduction in energy expenditure. Since the same doses had no effect on respiratory quotient or locomotor activity, the metabolic effect is not secondary to changes in energy substrate utilization or locomotor activity. This antithermogenic effect complements the eating stimulatory action of PVN galanin, and together these phenomena suggest a role for galanin as an anabolic neuropeptide. The similarity of galanin's effects to those of norepinephrine, with which it coexists in PVN nerve endings, further suggests the involvement of this amine and the PVN alpha2-noradrenergic system in galanin's mechanism of action. Galanin Paraventricular hypothalamus PVN Substrate utilization Activity Energy balance

Energy expenditure Indirect calorimetry

Thermogenesis Rat

Respiratory quotient

METHOD

THE 29 amino acid peptide, galanin is widely distributed in the central nervous system. The paraventricular nucleus of the hypothalamus (PVN) has particularly rich galanin innervation, and it is a primary site for galanin's stimulatory action on eating behavior (2,5-8,10,11,20). Despite its clearly demonstrated stimulation of eating effect, no information is yet available as to whether the metabolic aspects of energy balance are also affected by galanin injection into the PVN. That is, it is not clear whether galanin also affects thermogenesis and energy substrate utilization, nor has the possible involvement of locomotor activity changes in these effects been elucidated. The possibility of metabolic effects is suggested by galanin's coexistence in the PVN with the amine norepinephrine (12). Norepinephrine has been shown to have pronounced metabolic effects, in addition to potentiating food intake (18). Moreover, galanin's eating effects are possibly mediated, at least in part, through the corelease of PVN norepinephrine (6,10,20). The likelihood of galanin producing effects on the metabolic aspects of energy balance is also suggested by galanin's differential increase of carbohydrate and fat consumption, depending upon the period of the diurnal cycle (20,21). The present study was designed to elucidate the effects of PVN injections of galanin on energy expenditure and energy substrate utilization, as well as on locomotor activity. It provides the first evidence for galanin's involvement in the regulation of the metabolic aspects of energy balance.

Subjects Eleven Wistar rats obtained from the University of Sydney breeding farm were used. They weighed between 300 and 400 grams at the time of surgery. The rats were individually housed in clear acrylic cages, with food (Allied Rat & Mouse Kubes, Sydney) and tap water provided ad lib. The colony room was maintained at 22 +- 2°C, with a 14/10 hour light/dark cycle. Each rat was handled daily for one week before surgery and for two weeks after in order to minimize the stress of human contact.

Surgery and Histology The rats were anesthetized with 1 ml/kg Ketalar (100 mg/ ml ketamine hydrochloride; Parke-Davis Pty. Ltd.) and 0.1 ml Rompun (20 mg/ml xylazine hydrochloride; Bayer Australia Ltd.), both injected intramuscularly. They were placed in a stereotaxic apparatus and implanted with a single, stainless steel, 22-gauge guide cannula fitted with a dummy cannula (Plastics One, USA). The coordinates relative to bregma were: posterior 1.8, lateral 1.8, and ventral 7.2, with the cannulae implanted at an angle of 10 degrees offthe midline (17). The placement of the tip of the cannula was 1 m m dorsal to the PVN, and the injector cannula extended 1 m m beyond the tip of the guide cannula. At the conclusion of testing, the rats were given a lethal dose of Nembutal, after which their brains were removed for histological analysis. The brains were frozen to - 1 2 ° C , sectioned at

323

324 40 um, and stained with toluidine blue. Cannulae placements were determined microscopically with reference to the atlas of Paxinos and Watson (17).

Apparatus Respiratory quotient (RQ) and energy expenditure (EE) were calculated after recording oxygen (02) consumption and carbon dioxide (CO2) production on an open-circuit calorimeter. Two clear acrylic cylindrical chambers with stainless steel grid floors and a volume of 6.28 liters each were used. One was used for testing the rats and the other as reference standard for calibration of the atmospheric air. Compressed atmospheric air at a flow rate of 1600 ml/min and a pressure of 8 kPa above atmospheric was continuously drawn through both chambers. A system of solenoids allowed the air leaving one of the chambers to be split and directed for analysis, while the air from the other chamber was exhausted to the atmosphere. A sample of 110 ml/min was directed through a Perma Pure (Toms River, N J) permeation drier (model PD750-12PP), a CD-3A COz analyzer, and a S-3A O2 analyzer (Applied Electrochemistry, USA). The rest of the air was exhausted to the room. The analyzers were calibrated daily with primary gravimetric standards (Commonwealth Industrial Gases, Sydney). Motor activity was recorded by placing the testing chamber on an electronic balance (Mettler PE-2000) and using the unintegrated signal from the strain gauge. The reliability and validity of this method has been demonstrated in other studies (1316,18). A Z-80 based, S-100 bus microcomputer system controlled and monitored the calorimeter. The computer provided minuteby-minute records of air flow, CO2 production, O2 consumption, and activity counts. The following calculations were made: EE (kJ) = moles O2 (364 + 113 RQ); RQ = vol. CO2 produced/ vol. O2 consumed (1,3,4). Energy expenditure was expressed in joules/gram to account for different body weights.

Experimental Procedure Each rat was habituated to the metabolic apparatus by running 60-minute tests before and after surgery. This procedure also provided baseline data on the metabolic and activity parameters, and it allowed for the determination of any effect produced by the surgical procedure itself. The rats were also habituated to the injection procedure by introducing the injector cannula in place on two separate occasions before any experimentation. The experiment itself began approximately two weeks after surgery. The test sessions were conducted in the light phase of the cycle (between 10:00 a.m. and 4:00 p.m.). The rats were given a 30-minute period inside the testing chamber before each treatment. They were then removed and injected in counterbalanced order with 0.5 ul of either sterile saline (NaC1 0.9%) or one of five doses ofgalanin (Peninsula Labs Inc., USA): 0.01, 0.03, 0.1, 0.3, and 1 nmol, dissolved in sterile saline. These doses were selected on the basis ofgalanin's effect on feeding behavior (7). The dummy cannula was removed, and the injections were performed over a one-minute period through a 28-gauge injector cannula (Plastics One) which projected 1 mm beyond the guide cannula. The rats were unrestrained during the injection procedure. After the injection, the injector cannula was removed and the dummy cannula resecured. The rats were then placed in the testing chamber, and respiratory exchange and activity were monitored for 60 minutes. Following this, the rats were

MENI~NDEZ, ATRENS AND LEIBOWITZ returned to their home cages. At least seven days elapsed between successive injections to the same rat. Each recording session in the metabolic apparatus was preceded and followed by a 5-minute analysis of air leaving the calibration chamber, in order to assess any drift in the analyzers. Food was not available during the test sessions.

Data Analysis The data from each galanin treatment were compared with those of the saline treatment by a two-way analysis of variance with repeated measures on the two factors, time and treatment. RESULTS NO statistical differences were found between the presurgery, postsurgery, and saline injection data for any of the three parameters studied (energy expenditure, respiratory quotient, and activity). These data rule out the possibility of any surgery- or saline-induced metabolic effects. They also eliminate the possibility of any confounding effect clue to the volume injected or any cannula-induced damage to the PVN. The possibility of confounding due to cannula-induced tissue damage was also reduced by the histological analysis, which showed minimal incidental damage. Energy expenditure was significantly reduced by the 0.03, 0.1, and 0.3 nmol doses of galanin during the first few minutes of testing, though with different time patterns (Fig. 1). The effect of the three doses was observable within two minutes of the initiation of testing, F(I,10) -- 9.70, 6.62, and 8.93, p < 0.05. The 0.03 and 0.1 nmol effects were still observable at five, F(I, 10) = 12.59 and 9.24, p < 0.05, and ten minutes, F(1,10) = 10.16 and 11.57, p < 0.05. From fifteen minutes on, the energy expenditure values for the different galanin doses converged and, though still below the saline values, the differences were not statistically significant (Fig. 1). The highest and lowest doses of galanin ( 1.0 and 0.01 nmol) had no effect on energy expenditure. Neither respiratory quotient nor locomotor activity was affected by the galanin injections (Figs. 2 and 3). The tendency of energy expenditure values (Fig. 1), and to a lesser degree respiratory quotient values (Fig. 2), to start at high levels and decrease over time has been repeatedly found in this paradigm and has been ascribed to the stress of the handling and injection procedures (13-16,18). They are accompanied by a similar pattern in locomotor activity values (Fig. 3), which may certainly explain the initially high energy expenditure data. However, the facts that activity was not affected by galanin and that the peptide-induced reduction in energy expenditure occurred when its values were at their highest level support the suggestion that galanin exerts a primary inhibition of thermogenesis that is not secondary to changes in locomotor activity. DISCUSSION Despite the clear effects of PVN galanin on feeding (2,68,10,20), little is known about its metabolic effects. The likelihood of such effects is suggested by findings that PVN galanin preferentially enhances carbohydrate and fat ingestion (20). Moreover, there is some evidence that these eating effects may be mediated, in part, through the PVN noradrenergic system (10,20), which is known to participate in the regulation of energy metabolism (l 8). The present study provides the first evidence that PVN galanin reduces energy expenditure without affecting locomotor activity. This suggests that galanin inhibits thermogenesis, and that this

PVN G A L A N I N A N D E N E R G Y BALANCE

325

0.46

N! 0.44

E

E

R 0.42 G

O saline

Y

~ 1 nmol

E X p 0.40 E

<~ 0.3 nmol • 0.1 nmoI

N D

• 0.03 nmol

I 0.38 T U R E 0.36

• 0.01 nmol

i

I 0.34

~ 2

I

I 4

I 6

I

I ! 8 10 TIME AFTER INJECTION (rain)

I

I 14

12

,

FIG. 1. Mean energy expenditure (joules/gram) over a 15-minute period immediately after the injection of either saline or one of five doses of galanin. For clarity of illustration, the remainder of the test session (16-60 minutes, where no effect was detected) is not displayed, and S.E.M. values (which are in the range of 0.004-0.05) are not included.

effect is not related to any locomotor activity-induced energy production. Galanin injections into the PVN reduced energy expenditure in the dose range 0.03-0.3 nmol, which is similar to the doses previously demonstrated to stimulate feeding (7). The effect ofgalanin on energy expenditure was detected within two minutes of the injection, a latency which is similar to or even shorter than the latency for stimulating feeding (7). The reduction in energy expenditure was a relatively brief phenomenon, lasting less than

15 minutes, compared to the 20-30-minute feeding response (7). The lack of effect of the 0.01 nmol dose is also similar to the feeding study (7), but the lack of effect of the 1.0 nrnol dose is inconsistent with its maximal feeding effect (7). It should be noted here that two problems may be confounding the metabolic data. One is related to the repeated injections (six in the present study) of solutions in the same rat, and the other is related to the stress effect of handling which coincides

0.94

R E S P I R 0.92 A T O R Y

I £ ) saline 1 nmol <~ 0.3 nmol

• 0.1 nmol

Q U O 0.90 T I E N T

0.88

I i 0.03 nmol • 0.01 nmol

I 2

J

I 4

I

I 6

I

I i 8 10 TIME AFTER INJECTION (min)

I

i 12

i

i 14

I

FIG. 2. Mean respiratory quotient over a 15-minute period immediately after the injection of either saline or one of five doses of galanin. For easy comparison with Fig. 1, the remainder of the test session (16-60 minutes) is not displayed, and for clarity of illustration, S.E.M. values (which are in the range of 0.005-0.026) are not included.

326

MENI~NDEZ, ATRENS AND LEIBOWITZ

25°l r 2OO

T

y

,s, saline 150

£r 1 nmol O- 0.3 nmol •~ 0.1 nmol

C O 100 U N T S

• - 0.03 nmol • 0.01 rlmol

50

I 2

I

I 4

I

! 6

I

i

i

I

8 10 TIME AFTER INJECTION (min)

I

I 12

i

I

I

14

FIG. 3. Mean locomotor activity (activity counts) over a 15-minute period immediately after the injection of either saline or one of five doses of galanin. For easy comparison with Fig. 1, the remainder of the test session (16-60 minutes) is not displayed, and for clarity of illustration, S.E.M. values (which are in the range 1.7-52.2) are not included.

with the timing of the galanin effect. In order to overcome the first problem, a control experiment was done in which a different group of six similarly PVN-implanted rats was injected six times with one of the effective doses ofgalanin (0.03 nmol). The energy expenditure data obtained, and the time course of the effect, were entirely similar to the results obtained in the main experiment. The second problem can be confidently ignored if the experimental design is carefully considered. The counterbalanced order of injection, which includes the five doses of galanin and the vehicle, makes the stress effect constant across all rats and treatments. The additional observation that galanin did not have significant effects on respiratory quotient suggests that this peptide has no impact on energy substrate utilization (1,4). The baseline values (around 0.90) obtained in the present experiment reflect the mixed catabolism of carbohydrates, fats, and, to a lesser degree, proteins (1,4) at the time of the test (light phase of the cycle). Acute injections ofgalanin into the PVN produce, therefore, a very short-latency and short-lasting shift of the rat's metabolism towards a state of energy conservation. This anabolic state is characterized by a reduction in thermogenesis, with no appreciable change in energy substrate utilization. The latency, duration, and magnitude of this effect are very similar to those reported for PVN injections of norepinephrine (18) (see below). This galanin-induced anabolic state is further enhanced by the stimulation of eating (7,8). The combination of these metabolic and behavioral effects makes galanin a strong anabolic force when injected into the PVN. Galanin's mechanism of action remains to be established. Evidence suggests that it might produce its eating effects, at least in part, by activating the PVN alpha2-noradrenergic system, in a manner similar to norepinephrine (9,10,18,20). This suggestion is supported by the demonstration that galanin and norepinephrine are stored together within the same terminals of the PVN (12), by the similarity of their feeding and macronutrient

selection effects (6,11,20,22), and also by the evidence that galanin's stimulatory feeding effect is antagonized by alpha-adrenergic receptor blocking agents and norepinephrine synthesis inhibitors (6,10,20). The effects reported here add further support for this hypothesis. Injections of norepinephrine into the PVN produce a short-latency and short-lasting reduction in energy expenditure, without affecting respiratory quotient (I 8). This similarity suggests that the galanin-induced reduction of energy expenditure may be mediated through the activation of PVN alpha2 receptors, either by itself or by the corelease of norepinephrine. However, the differences between norepinephrine and galanin, particularly in relation to their effects on the endocrine system ( 11,21), should also be considered. Whereas neuropeptide Y, like galanin, potentiates feeding after PVN injections (19), any similarity between these peptides appears to be limited to the stimulation of feeding, since PVN injections of neuropeptide Y and galanin have clearly different metabolic as well as endocrine effects. In particular, neuropeptide Y greatly affects the respiratory quotient but not energy expenditure, and this effect has a very long latency and duration (16). Both the nature of this effect and its temporal characteristics are very different from the effect of galanin reported here. The evidence suggests, therefore, that neuropeptide Y and galanin, though anabolic in nature, have different roles in the PVN. While neuropeptide Y is a primary modulator of long-term energy substrate utilization, galanin is a primary modulator of shortterm thermogenesis. This adds to the already known differences between galanin and neuropeptide Y in relation to norepinephfine and endocrine factors such as corticosterone (10,11). In conclusion, the evidence provided here makes it possible to include galanin in the list ofPVN anabolic neurotransmitters, together with norepinephrine (18) and neuropeptide Y (16). These agents can be further subdivided into antithermogenic agents (galanin and norepinephrine) and inducers of fat deposition (neuropeptide Y). It seems likely that, though there are

PVN G A L A N I N A N D E N E R G Y B A L A N C E

327

functional interrelationships in the action of these three neurotransmitters, most (in the case of neuropeptide Y) or at least part (in the case of galanin) of their effects are through effector systems independent of norepinephrine.

ACKNOWLEDGEMENTS This research was supported by grants from the Australian Research Council to Dale M. Atrens, and by Grant MH 43422 to Sarah F. Leibowitz.

REFERENCES 1. Bray, G. A. Nutrient balance and obesity: An approach to control of food intake in humans. Med. Clin. North Am. 73:29--45; 1989. 2. Crawley, J. N.; Austin, M. C.; Fiske, S. M.; Martin, B.; Consolo, S.; Berthold, M.; Langel, U.; Fisone, G.; Bartfai, T. Activity of centrally administered galanin fragments on stimulation of feeding behavior and on galanin receptor binding in the rat hypothalamus. J. Neurosci. Res. 10:3695-3700; 1990. 3. Gleeson, M.; Brown, J. F.; Waring, J. J.; Stock, M. J. The effects of physical exercise on metabolic rate and dietary-induced thermogenesis. Br, J. Nutr. 47:173-181; 1982. 4. Kleiber, M. The fire of life; an introduction to animal energetics. New York: Robert R. Krieger; 1975. 5. Koenig, J. I.; Hooi, S. C.; Maiter, D. M.; Martin, J. B.; Gabriel, S. M.; Strauss, R. M.; Kaplan, L. M. On the interactions of galanin within the hypothalamic-pituitary axis of the rat. In: Hokfelt, T.; Bartfai, T., eds. Galanin and the hypothalamic-pituitary axis. New York: Macmillan Press (in press). 6. Kyrkouli, S. E.; Stanley, B. G.; Hutchinson, R.; Seirafi, R. D.; Leibowitz, S. F. Peptide-amine interactions in the hypothalamic paraventricular nucleus: Analysis of galanin and neuropeptide Y in relation to feeding. Brain Res. 521:185-191; 1990. 7. Kyrkouli, S. E.; Stanley, B. G.; Leibowitz, S. F. Galanin: Stimulation of feeding induced by medial hypothalamic injection of this novel peptide. Eur. J. Pharmacol. 122:159-160; 1986. 8. Kyrkouli, S. E.; Stanley, B. G.; Seirafi, R. D.; Leibowitz, S. F. Stimulation of feeding by galanin: Anatomical localization and behavioral specificity of this peptide's effects in the brain. Peptides 11:9951001; 1990. 9. Leibowitz, S. F. Hypothalamic paraventricular nucleus: Interaction between alpha2-noradrenergic system and circulating hormones and nutrients in relation to energy balance. Neurosci. Biobehav. Rev. 12:101-109; 1988. 10. Leibowitz, S. F. Hypothalamic neuropeptide Y and galanin: Functional studies of coexistence with monoamines. In: Mutt, V.; Hokfelt, T.; Fuxe, K., eds. Nobel Symposium XIV: Neuropeptide Y. New York: Raven Press; 1989:267-281.

11. Leibowitz, S. F. Hypothalamic galanin in relation to feeding behavior and endocrine systems. In: Hokfelt, T.; Bartfai, T., eds. Galanin: A new multifunctional peptide in the neuro-endocrine system. New York: Macmillan Press (in press). 12. Levin, M. C.; Sawchenko, P. E.; Howe, P. R.; Bloom, S. R.; Pollack, J. M. Organization of galanin-immunoreactive inputs to the paraventricular nucleus with special reference to their relationship to catecholamingeric afferents. J. Comp. Neurol. 261:562-582; 1987. 13. McGregor, I. S.; Menrndez, J. A.; Atrens, D. M, Metabolic effects of neuropeptide Y injected into the sulcal prefrontal cortex. Brain Res. Bull. 24:363-367; 1990. 14. Menrndez, J. A.; Atrens, D. M. Insulin increases energy expenditure and respiratory quotient in the rat. Pharmacol. Biochem. Behav. 34:765-768; 1989. 15. Menrndez, J. A.; Atrens, D. M. Insulin and the paraventricular hypothalamus: Modulation of energy balance. Brain Res. 555:193201; 1991. 16. Menrndez, J. A.; McGregor, I. S.; Healey, P. A.; Atrens, D. M.; Leibowitz, S, F. Metabolic effects of neuropeptide Y injections into the paraventricular nucleus of the hypothalamus. Brain Res. 516: 8-14; 1990. 17. Paxinos, G.; Watson, C. The rat brain in stereotaxic coordinates, 2nd edition. Sydney: Academic Press; 1986. 18. Siviy, S. M.; Kritikos, A.; Atrens, D. M.; Shepherd, A. Effects of norepinephrine infused in the paraventricular hypothalamus on energy expenditure in the rat. Brain Res. 487:79-88; 1989. 19. Stanley, B. G.; Leibowitz, S. F. Neuropeptide Y: Stimulation of feeding and drinking by injection into the paraventricular nucleus. Life Sci. 35:2635-2642; 1984. 20. Tempel, D. L.; Leibowitz, K. J.; Leibowitz, S. F. Effects of PVN galanin on macronutrient selection. Peptides 9:309-314; 1988. 21. Tempel, D. L.; Leibowitz, S. F. Galanin inhibits insulin and corticosterone release after injection into the PVN. Brain Res. 536:353357; 1990. 22. Tempel, D. L.; Leibowitz, S. F. Diurnal variations in the feeding responses to norepinephrine, neuropeptide Y and galanin in the PVN. Brain Res. Bull. 25:821-825; 1990.