- Email: [email protected]
Periaqueductal gray lesions do not affect grooming, induced electrically in the hypothalamic paraventricular area in the rat
Behavim~ral Brain Research, 59 (1993) 95-101 © 1993 Elsevier Science B.V. All rights reserved. 0166-4328/93/$06.00
95
BBR 01537
Periaqueductal gray lesions do not affect grooming, induced electrically in the hypothalamic paraventricular area in the rat Annemoon M.M. Van Erp a'**, Menno R. Kruk a'*, Wout Meelis a, Jan G. Veening b ~Ethopharmacology Group, Sylvius Laboratory, PO Box 9503, 2300 RA University of Leiden, Leiden (The Netherlands), b Department (?/'Anatomy and Embo,ology, UniversiO, of Nfjmegen, N(jrnegen (The Netherlands) (Received 28 June 1993) (Revised version received 15 August 1993) (Accepted 9 September 1993)
Key words: Electrical stimulation; Grooming; Hypothalamus; Lesion; Paraventricular nucleus; Periaqueductal gray
Electrical stimulation inducing behavioral responses from the hypothalamus seems to activate systems involved in the execution phase of the behaviour rather than in the introductory or decision-making phase. However, the pathways involved are not fully understood. Projections originating from hypothalamic areas involved in specific behavioral responses are rather complex. The periaqueductal gray (PAG) has been proposed to be an essential output station of hypothalamic behavioral mechanisms. Here we report that lesions of the periaqueductal gray area have no effect on grooming responses evoked by electrical stimulation of the hypothalamic paraventricular area. Neither threshold current intensities needed to evoke grooming, nor latencies were affected 7 or 14 days after lesioning. The lesions caused severe behavioural deficits. Animals did not drink or eat spontaneously, had problems with motor coordination and sometimes showed strong defensive reactions upon touch. However, their grooming responses induced by hypothalamic stimulation were not changed. The PAG may have a modulatory role on grooming behaviour; however, this modulatory effect apparently is overruled during electrical stimulation of the hypothalamus.
INTRODUCTION
Electrical stimulation of specific areas in the hypothalamus evokes complex behavioural patterns. Depending on the exact location of the electrode, different responses are evoked, such as attack behaviour or selfgrooming. Self-grooming can be evoked in the paraventricular nucleus (PVH) and dorsal hypothalamic area (DHA) 13"2~'22"2~ Several descending and ascending pathways have been described that may be involved in the execution of these behavioural responses. One particular pathway runs from the hypothalamus through the periaqueductal gray area (PAG) and the adjoining parts of the brain stem ~s. The PAG seems to be involved in aggressive and defensive behaviour 2"3'~7 and has been shown to be crucial for ACTH-induced grooming 25. However, Mos and coworkers x8 showed that PAG lesions only slightly and transiently reduce aggression evoked by hypothalamic stimulation and aggression in a territorial setting (resident/intruder * Corresponding author. Fax: (31) (71) 276292. ** Present address: Dept. of Psychopharmacology, Tufts University, Medford, USA.
paradigm). Therefore, it seemed worthwhile to investigate the effect of P A G lesions on hypothalamically induced grooming behaviour, to study the role of the PAG in the execution of self-grooming.
MATERIALS A N D M E T H O D S
Surgery Experimental procedures have been described by Kruk etal. 12 and Mos etal. ~s. Fifteen male rats (Wistar/Harlan, Zeist) weighing 400-500 g were implanted bilaterally with bipolar electrodes (150 #m) aimed at the co-ordinates AP 7.40 mm ML 0.50 mm & DV 1.90 mm according to the atlas of Paxinos and Watson ~9. For details about the electrodes and connectors used see Kruk et al. I'~-. During surgery a piece of india rubber was placed on the skull behind the electrodes and embedded with a thin layer of dental cement, to allow a second operation to implant a pair of lesioning electrodes in the PAG. After surgery rats were housed individually in Macrolon cages in lownoise rooms at 22 ° C and 750Jjo relative humidity. Food and water were available ad libitum and an inverted
96 day/night cycle was installed, with lights on from 15.00 h to 03.00 h.
Behavioural testing and electrode selection Behavioural testing was started 1 week after surgery. Animals were placed in perspex stimulation cages (diameter 40 cm, height 60 cm, closed foam polypropylene floor). Rats were stimulated with trains of 40 Hz biphasic square-wave pulses with a phase duration of 0.2 ms and a phase interval of 12.5 ms. At the first test train duration was 30 s with 60 s intervals; at subsequent tests train duration was decreased at least to 15 s (and usually to 10 s). Threshold current intensity needed to evoke a grooming response was determined using the up-and-down method of Dixon and Mood 7 as modified by Wetheril135. Animals were stimulated only when they were not grooming spontaneously'. If the? did groom, stimulation was postponed until they had stopped. Subsequent threshold determinations on consecutive days were used to assess the stability of baseline thresholds. Electrode sites from which no stable thresholds on 3 subsequent days could be obtained, as well as sites at which no grooming within 15 s upon start of stimulation was evoked were excluded from the experiment. In animals with two grooming-evoking electrodes the electrode with lowest threshold current intensity for grooming was included in the data analysis. PA G lesions Thirteen animals with stable baseline thresholds for grooming were implanted bilaterally with two monopolar electrodes aimed at the PAG at the co-ordinates AP 3.00 mm, ML 0.60 ram, DV 4.20 mm according to Paxinos and Watson ~9. The teflon-coated silver electrodes (diameter 0.33 mm) with 0.5 mm bare tips were checked carefully for current leakage before implantation. After hardening of the dental cement the electrodes were connected to a Grass LM4 radiofrequency lesion maker, while the animal was still under anesthesia (for detailed description of the procedure, see Mos et al.~S). Lesions were made by passing a 100 kHz current for 20 s between the PAG-electrodes. During the 20 s the current was kept at approximately 20 mA (at a voltage of 40-80 Volt). The animals were allowed to recover for at least 1 week after surgery. All animals received water, applied orally via a syringe, and wet mesh the first day after surgery, because they failed to drink and eat by themselves. If necessary, this was continued until the animals had recovered and started to regain weight. The behavioural and postural consequences of the lesion were monitored during the post-
operative recovery period. Specml attention was paid to the reaction of the rats upon approach and touch by the experimenters hand during handling.
Postlesion behavioural testing Lesion effects on stimulation-induced behaviour were determined 7 and 14 day~ after surgery. Lesion effects were assessed as changes in threshold current required to evoke grooming. An increase in threshold means that the lesion inhibits grooming; a decrease would indicate a facilitation of grooming. Latencies of stimulationevoked grooming were measured during thresholds determination. A two-way ANOVA was used to detect lesion effects on hypothalamic grooming thresholds and latencies, compared with the last three threshold determinations before lesioning. Histology After completion of the experiment the rats were perfused with physiological saline followed by a 4'3o Ibrmaldehyde solution. Twenty-#m sections of the brain were stained with Luxol fast blue and Cresyl violet. The electrode tips were localized precisely and drawn on a detailed cytoarchitectonic atlas of the rat hypothalamus ~9. Lesions were drawn at plates of the atlas of Paxinos and Watson, Lesions were indicated to be complete if the PAG had disappeared completely at the levels AP 2.70 mm and[or 2.96 mm. so that we could be sure that all hypothatamic fibres descending through the PAG were interrupted. AnimNs with incomplete lesions were excluded from data analysis.
RESULTS
Basal grooming response to electrical stimulation At 19 of 30 electrodes that were implanted in the hypothalamic paraventricular area. reliable thresholds for grooming were obtained. At these electrodes grooming could be evoked within 15 s after onset of stimulation with stable thresholds, the range of these thresholds being 14.4 to 83.3 #A. The exact localization of electrode tips in the hypothalanms is shown in Fig. l. All grooming evoking electrodes were located in or close to the paraventricular nucleus: electrodes at which no reliable thresholds for grooming could be determined or no grooming at all were located at greater distances from the PVH (data not shown). This is in concordance with previous results t3"2~. Electrodes in rats that died during or shortly after lesioning had a similar anatomical distribution as electrodes m rats that were u s e d t o determine postlesion thresholds for grooming (Fig. 1).
97
,,
r ~ , ~
ITe lY
Fig. 1. Localization of electrode sites in the hypothalamic paraventricular area at which reliable grooming thresholds were determined within 15 s after onset of stimulation. Sites are drawn on adjacent sections (150 #m) of a detailed cyto-architectonic atlas of the hypothalamus *9. Left panel: anterior part; right panel: posterior part. Each symbol represents one electrode tip. Circles: sites in rats that were used for postlesion threshold determinations. Triangles: sites in rats that did not survive lesioning. AHA, anterior hypothalamic area; ARH, arcuate nucleus: BST, bed nucleus of stria terminalis; DHA, dorsal hypothalamic area; fx, fornix; PVH, paraventricular hypothalamic nucleus; SCN, suprachismatic nucleus; THAL, thalamus; vlll, third ventricle; VMH, ventromedial hypothalamic nucleus.
grooming thresholds or latencies (Fig. 2) 7 or 14 days after lesioning (thresholds: P = 0.4874 F = 0.891, dr= 4; latencies: P=0.8965 F=0.266, d f = 4 ; two-way ANOVA), as measured at 6 electrodes in 6 animals with complete lesions. Additional data from 4 other electrodes in the same animals gave similar results. In Fig. 2 pooled data for these 10 electrodes are presented (thresholds: P=0.4202 F = 1.000, d r = 4 ; latency: P = 0.2065 F = 1.558, dr= 4). Although some animals had small motor problems and would topple over more often while grooming, this apparently did not affect their ability to respond properly upon stimulation. They started to groom at the same latency and at the same current threshold intensity as before lesioning. We could detect no changes in stimulation-evoked grooming nor in spontaneous grooming patterns as performed in between stimulation trials.
Effect of PA G lesions Periaqueductal gray lesions had a quite severe effect on the behaviour of the rats. Two rats died during lesioning and 2 rats a few days after lesioning. The first day after lesioning all rats failed to drink and eat, and in some rats this would continue for several days. Rats received water intra-orally, wet mesh and supplementary saturated glucose-solution until their weight stabilized. Weight loss was rather extensive, from 50-80 g (in rats with starting weight of 400 g) or even 100 g (starting weight more than 500 g). Several animals had a problem in maintaining motor control and would topple over while grooming. The behavioural deficits disappeared slowly; after 7 days most animals behaved more or less normally, although a few still had small motor problems. PAG lesions had no effect on hypothalamic
grooming thresholds
grooming latencies
8O
15 [~
'~
So
"~ m
~o
e-
2O
prelesion - 3
12
A
prelesion - 2 9 [~
O
_=
prelesion - 1
6 3 0
post lesion day 7 post lesion day 14
Fig. 2. Left: effect of PAG lesion on thresholds for hypothalamic grooming. Bars represent pooled data of 10 electrodes in 6 rats with complete lesions (see text). Open bars: three last thresholds determined before lesioning. Hatched bars: thresholds at day 7 and day 14 after lesioning. Right: effect of PAG lesion on grooming latencies. Latencies were determined during threshold determinations. We observed no significant results between treatments (two-way ANOVA).
98 TABLE 1 S i z e a n d extension o f P A G lesion,s at several anterior-postertor levels o f the brain according io P a x i n o s a n d Watson ~
Plates 3.80 to 1.36 m m AP to the interaural point. Symbols: T = P A G totally destro 3 ed: P = P A G partly destroxed: - no lesion visible. Rats 8 and 9 were excluded because no stable thresholds for grooming were obtained. Rats 1, 3. 5 and 11 died during <~r shorttv after lesioning, Rat
('~ttl+Ilt'tll
Plate 3.80
3.40
3.20
2.96
_.~ +,(I
~.,~ ~,~"
7¢}
7n
1
T
Y
T
T~
T~
1
1
P
2
-
P
P
P
1 ~
P
P
P
4
-
P
1TM
I *
I
P
P
6 7
P
P T
T I-
T* I "
"f * 1~
I P
P P
P
10
T
~I
T
T ~
T ~
l
P
P
12 13 14 15
P P -
P P T P
P P F P
P P T'* P
P P 1~ P
P P P P
P P P
died
incomplete incomplete P P
incomplete
* Indicates complete lesions at plate 2.70 or 2.96.
A
cTx
~
B
C
L
9
Fig. 3. A: drawing of a moderate but complete P A G lesion {hatched area) in plates of the atlas of Paxinos and Watson 1~. Secuons of rat 14. drawn on plates 3.70, 2.70 and 1.70 mm interaural. B: same as ~. Moderate to large P A G lesion: rat 7 which had problems with motor coordination. Lesion extends more ventrally, including red nuclei. C: same as A. Very large P A G lesion: rat 1 wttich died shortly after lesiomng. Lesion includes red nuclei and parts of the medial superior colliculi, resulting in gliosis and cell death in the substantia nigra pars reticulata (stippled area). Aq, cerebral aqueduct: CTX. cortex: P A G , periaqueductal gray: RN. red nucleus: SC, superior colficulus: SN, substantia nigra: VTA. ventral tegrnental area.
99
Histoh)gical verification of lesion size The lesions appeared to be moderate to large in size. They extended from plate AP 1.70 mm to 3.80 mm in the atlas of Paxinos and Watson 19 (Table I). Moderate lesions were confined to the periaqueductal gray area, but sometimes the PAG was not destroyed completely at anterior levels plates 2.70 mm and 2.96 ram. An example of a complete, moderate lesion is given in Fig. 3A. Moderate to large lesions appeared to have partly destroyed the red nuclei (Fig. 3B). Not surprisingly, these animals had problems with maintaining motor control. One animal that died shortly after surgery appeared to have a very large lesion, including parts of the colliculi superiori, thalamus and red nuclei (Fig. 3C). In this animal, the damage to the superior colliculi may have been the cause of cell death and gliosis in the substantia nigra, pars reticulata. This was not observed in any of the other lesioned animals. After careful histological control, 6 animals were used in the data analysis of the effects of PAG lesions upon hypothalamically induced grooming.
DISCUSSION
Our study shows that lesions of the periaqueductal gray area have no effect on thresholds for hypothalamically evoked self-grooming behaviour. This is in line with the effect of PAG lesions on hypothalamic attack responses. Mos and coworkers is showed that PAG lesions only slightly and transiently reduce thresholds for hypothalamic attack. In general, hypothalamic responses may consist of either changes in the introductory phase of behaviour, e.g. after ventromedial hypothalamic stimulation 32"33 or consist of a clearcut response, often of a compulsive nature, e.g. attack or grooming responses ~L-~4"2s. Animals interrupt ongoing behaviour immediately and show the particular response elicited within seconds after onset of stimulation. Behaviourally it seems that a system involved in the execution phase of the behaviour, rather than the introductory or decision making phase, is activated l~ Which pathways are involved in these hypothalamic responses? Fibres descend towards the lower parts of the brainstem either via the ventral tegmental area and the central tegmental field, or along the midline via the PAG 15'2°'34. In addition, ascending fibres and fibres with a thalamic destination may play a role ls'2°, as well as other ascending projections to the forebrain. Interestingly, hypothalamic projections to the PAG show a topographical organization 4's'~6-2°,24,34 and the fibres originating from the hypothalamic attack area (caudal, lateral PAG) show a distribution that is very different
from fibres originating from the hypothalamic grooming area (dorsal and lateral PAG) 2°. Spruijt and coworkers 25 showed that the PAG is a prerequisite for ACTHinduced grooming. PAG lesions diminished the response to i.c.v, injections of ACTHl_24. However, grooming did not disappear completely, but was reduced to baseline level. There are several possible explanations for these seemingly contradictory results. Recent experiments, in which grooming was induced by direct injection of neuropeptides into the PVH 29, suggest that peptides such as ACTH and c~-MSH prolong a grooming response that was initiated by other factors, e.g. handling of the rat or exposure to a novel environment. In contrast, a peptide such as oxytocin seems to initiate grooming 3°. Apparently, separate systems exist in the central nervous system that regulate the initiation of grooming, its continuation and probably the interruption of this behaviour. We hypothesise that ACTH, for which binding sites are present in the PAG 26, serves to modulate grooming that has been initiated by other factors outside the PAG. This could explain the fact that after PAG lesions grooming is still performed at baseline level. The modulatory effect of the PAG on grooming has been confirmed recently by Van Wimersma Greidanus, who showed that PAG lesions attenuate ACTH- but not oxytocin-induced grooming 3~. Nevertheless, our results suggest that the PAG is not the major destination station involved in hypothalamically evoked grooming responses. A recent study by Roeling 23 showed that one of the many efferent pathways originating from the PVH that is possibly involved in the execution of grooming, is the connection descending via lateral hypothalamus, ventral tegmental area (VTA) and central tegmental field. Interestingly, the VTA is one of the areas of the brain that has been reported to be involved in grooming behaviour: injection of neuropeptides, such as oxytocin, CCK m and ~-MSH 27, into the VTA induces selfgrooming behaviour. The possible involvement of the VTA in the control of hypothalamic responses deserves special attention in future studies, although a similar study using VTA lesions may prove difficult to perform because of severe effects on autonomic functions. Lesions of the periaqueductal gray resulted in severe behavioural deficits. Some of these effects may have been caused by lesions of tissue surrounding the PAG, especially in animals with large lesions in which ascending serotonergic and/or noradrenergic pathways may have been disrupted. Animals didn't eat or drink (also reported by Mos et al. lS): they seemed to have "forgotten" how to do it, because they would drink readily when helped and showed interest in wet mesh after they had been drinking. It is interesting to note
100
that severn animals showed moderate to strong defensive reactions upon touch, despite the intensive handling period before surgery. They didn't try to bite, but upon touch by the experimenters hand tried to escape by running and jumping and sometimes they screamed when picked up. This observation is at odds with reports that PAG lesions attenuate defensive behaviour 3'6, for which we have no explanation. Others have reported that PAG lesions do not change defensive reactions induced by other means either, e.g. inferior colliculus stimulation ~. In addition, some animals showed reflex-like face washing movements upon stimulation of the mouth region with a wet syringe during drinking sessions. These movements were quite strong and compulsive, performed in a position in which a rat normally would never groom (restrained, lying on its back). These movements were not just attempts to remove the syringe, but included the repetitive "front paw over ear" movements that are characteristic for face washing. This is an important observation, showing that the animals are still responding to sensory stimuli. Apparently the sensory input pathways are intact, as well as the motor components of the grooming behaviour. In spite of the fact that they have a problem in maintaining their body posture, they still groom, even while lying on their back. We could not correlate the nature of observed behavioural deficits (defensive behaviour, face washing upon stimulation of mouth region) to the size or extension of the lesions. These clear deficits, however, did not affect the hypothNamically evoked grooming response, neither inhibitory nor facilitatory. ACKNOWLEDGEMENTS
The investigations were supported by the Foundation for Biological Research (BION), which is subsidized by the Netherlands Organization for Scientific Research (NWO). REFERENCES 1 Bagri, A., Di Scala, G. and Sandner, G., Wild running elicited by microinjections of bicuculline or morphine into the inferior colliculus of rats: lack of effect of periaqueductal gray lesions, Pharmacol. Biochem. Behav., 41 (1992) 727-732. 2 Bandler, R., Brain mechanisms of aggression as revealed by electrical and chemical stimulation: suggestions of a central role for the midbrain central grey region, Prog. Psychobiol. Physiol. P.~3'ehol., 13 (1988) 67-154. 3 Bandler, R. and Depaulis, A., Elicitation of intraspecific defense reactions in the rat from midbrain periaqueductal grey by microinjection of kainic acid, without neurotoxic effects, Neurosci. Lett., 88 (1988) 291-296.
4 Behbehani. M.M.. Park. M.R. and Clement. M.E.. Interactions between the lateral hypothalamus and the periaqueductal gray. J. Neurosci.. 8 (1988) 2780-2787. 5 Beitz. A.J.. The organizauon ot ~afferent projecuons to the midbrain periaqueductal gray of the r~t. Veuroscience. " ~/ 982) 133159. 6 Depaulis. A. and Bandler. R. tkds.L Ihe Midbruin Periaqueductai Gray Matter. Functional. Anatomical and Neurochemical Organiza~ non Plenum. New York, 199l. Dixon. W.J and Mood, A.M_ ~ method for obtaining and analyzing sensitivity data. J. Am, St¢lt. Assoc.. 43 (19481 109--126 Geeraedts, L . M . G . Nieuwenhu~s~ R, and Veening, J.G.. The medial forebrain bundle of the rat III. Cytoarchitecture of the rostral (telencephalic) part of the MFB bed nucleus_ L Comp. Neurot. 294 (19901 507-536. 9 Geeraedts. L.M.G.. Nieuwenhuvs. R. and Veening, ,1.(J., ]'he medial forebrain bundle of the rzlt IV C3toarchitecture of the caudal (lateral hypothalamic) part of the MFB bed nucleus, J Comp. Neurol. 294 (1990) 537-5~-~8 10 Kaltwasser. M.Th. and Crawler, J,N.. Oxytocin and cholecystokinin induce grooming behavior in the ventral tegmentum of the rat. Brain Res_ 426 (19871 1-7 I1 Kruk. M.R.. Ethology and pharmacology ot hypothalamic aggression in the rat. Neurosci. Bh~behav. Rev.. 15 (19911 527538 12 Kruk. M.R.. Van der Poel. '\.M and De Vos-Frcrichs. I.P.. The induction of aggressive behaviour b) electric:d stimulation in the hypothalamus of male rats, Beh~rt'imo'. 70 119791 292-322. 13 Lammers. J.H.C.M.. Kruk. M . R . Meelis. W. and Van der Pool. A.M.. Hypothalamic subs trates rot brain stimulation-induced grooming, digging and circling in the rat, Brain Rea.. 418 (19871 1-19. 14 Lammers. J.H.C.M., Kruk. M.R.. Meelis. W. and Van der Poel. A.M.. Hypothalamic substrates for brain stimulation-induced attack, teeth-chattering and social grooming in the rat. Brain Res.. 449 ( 19881 311-327. 15 kuiten. P.G.M.. Ter Horst. G . J , Karst. H. and Stefl'ens. A.B.. The course of paraventricular hypothalamic efferents to autonomic structures m medulla and spinal cord. Brain Re.~. 329 t 19851 374-378. 16 Morrell. J.I.. Greenberger, L.M. and PfaflL D.W., Hypothalamic. other diencephalic, and telencepha/ic neurons that projecl to the dorsat midbrain. J. Comp. Neurol.. 201 ( 1981 ) 589-620. 17 Mos. J.. Kruk. M.R.. Van der Poel. A.M. and Meelis. W.. Aggressive behaviour induced bv eleelrical stimulation in the midbrain central gray of male rz~ts. Aggress. Behav.. ,~ (t982) 261-284 I~; Mos, J., Lammers, J.H.C.M.. Van der Poel. A.M.. Bermond. B.. Meelis. W and Kruk. M.R.. Effects of midbrain central gra3 lesions on spontaneous and electrically induced aggression in the rat. Aggress. Behav.. 9 (19831 133-155. 19 Paxinos, G. and Watson. C.. The Rat Brain in Stereotaxi( Coordinates. 2rid. edn. Academic Press. Sydney, 1986. 20 Roeling, T.A.P.. The Neuronal Substrate ol Hypothalamk'all3 Induced Behaviour #t the Rat. A Behavioral and Anatomical Study. Thesis, Nijmegen. 1993. 21 Roeling, T.A.P.. Van Erp, A.M.M.. Meelis. W.. Kruk. M.R. and Veening, J.G.. Behavioural effect~ of N M D A injected into the hypothalamic paraventricular nucleus of the rat. Bra#~ Res.. 550 19911 220-224. 22 Roeling, T.A.P.. Veening, J.G., Kruk. M.R. and Nieuwenhuys, R.. Grc~ming bebaviour elicited b 5 kainic acid evoked cell body
101
23
24
25
26
27
28
29
stimulation in the hypothalamus of the rat, Neurosci. Res. Cornmum, 6 (1990) 11-118. Roeling, T.A.P., Veening, J.G,, Peters, J.P.W., Vermelis, M.E.J. and Nieuwenbuys, R., Efferent connections of the hypothatamic "'grooming area" in the rat, Neuroscience, 56 (1993) 56. Semenenku, F.M. and Lmnb, B.M., Projections of anterior hypothalamic neurones to the dorsal and ventral periaqueductal grey in the rat, Brain Re.s., 582 (1992) 237-245. Spruijt, B.M., Cools, A.R. and Gispen, W.H., The periaqueductal gra3.: a prerequisite for ACTH-induced excessive grooming, Behav. Brain Res.. 20 (1986) 19-25. Tatro, .I.B., Melanotropin receptors in the brain are differentially distributed and recognize both corticotropin and :~-melanocyte stimulating homaone, Brui~l Res.. 536 (1990) 124-132. Torre, E. and Celis, M.E., Cholinergic mediation in the ventral tegmental area of ~.-melanotropin induced excessive grooming: changes in the dopamine activity in the nucleus accumbens and caudatc putamen, L([;" Sci., 42 (1988) 1651-1657. Van Erp, A.M.M., Kruk, M.R., Willekens-Bramer, D.C., Bressers, W.M.A., Roeling, T.A.P., Veening, J.G. and Spruijt, B.M., Grooming induced by intrahypothalamic injection of A C I H in the rat: comparison with grooming induced by intrahypothalamic electrical stimulation and i.c.v, injection of ACTH, Brain Res.. 538 (1991) 203-210. Van Erp, A.M.M.. Kruk, M.R,, Van Oers, H.J.J. and Hemmers,
30
31
32 33
34
35
N.M., Differential effect of ACTHl_24 and ~-MSH in PVH induced grooming, Brain Res.. 603 (1993) 296-301. Van Erp, A.M.M., Kruk, M.R., Semple, D.M. and Verbeet, D.W.P., Initiation of grooming in resting rats by i.c.v, oxytocin but not by ACTH I _,4 and :~-MSH, Brain Re,v.. 607 (1993) 108112. Van Wimersma Greidanus, Tj.B. and Maigret, C., Differential role of the periaqueductal gray in the grooming inducing effect of neuropeptides, Neurosci. Res. ()mml., 11 (1992) 145-153. Veening, J.G., Behavioural Function.~ qf the VMH in the Rat: An Ethological Approach, Thesis, University of Groningen, 1975. Veening, J.G., Brain and behaviour: morphological and functional aspects of the hypothalamus in thc rat, Eur. J. Morphol., 30 (1992) 53-66. Veening, J.G., Buma, P., Ter Horst, G.J., Roeling, T.A.P., Luiten, P.G.M. and Nieuwenhuys, R., Hypothalamic projections to the PAG in the rat: topographical, immuno-electromnicroscopical and functional aspects. In A. Depaulis and R. Bandler (Eds.), The Midbra#7 Periaqueductal Gray Matter. Plenum, New York, 1991, pp. 387-415. Wetherill, G.B., Sequential estimation of points on quantal response curves. In G. Barrie and G.B. Wetherill (Eds.), Sequential Methods #7 Statistics, Methuen and Co., London, 1966, pp. 162-179.
95
BBR 01537
Periaqueductal gray lesions do not affect grooming, induced electrically in the hypothalamic paraventricular area in the rat Annemoon M.M. Van Erp a'**, Menno R. Kruk a'*, Wout Meelis a, Jan G. Veening b ~Ethopharmacology Group, Sylvius Laboratory, PO Box 9503, 2300 RA University of Leiden, Leiden (The Netherlands), b Department (?/'Anatomy and Embo,ology, UniversiO, of Nfjmegen, N(jrnegen (The Netherlands) (Received 28 June 1993) (Revised version received 15 August 1993) (Accepted 9 September 1993)
Key words: Electrical stimulation; Grooming; Hypothalamus; Lesion; Paraventricular nucleus; Periaqueductal gray
Electrical stimulation inducing behavioral responses from the hypothalamus seems to activate systems involved in the execution phase of the behaviour rather than in the introductory or decision-making phase. However, the pathways involved are not fully understood. Projections originating from hypothalamic areas involved in specific behavioral responses are rather complex. The periaqueductal gray (PAG) has been proposed to be an essential output station of hypothalamic behavioral mechanisms. Here we report that lesions of the periaqueductal gray area have no effect on grooming responses evoked by electrical stimulation of the hypothalamic paraventricular area. Neither threshold current intensities needed to evoke grooming, nor latencies were affected 7 or 14 days after lesioning. The lesions caused severe behavioural deficits. Animals did not drink or eat spontaneously, had problems with motor coordination and sometimes showed strong defensive reactions upon touch. However, their grooming responses induced by hypothalamic stimulation were not changed. The PAG may have a modulatory role on grooming behaviour; however, this modulatory effect apparently is overruled during electrical stimulation of the hypothalamus.
INTRODUCTION
Electrical stimulation of specific areas in the hypothalamus evokes complex behavioural patterns. Depending on the exact location of the electrode, different responses are evoked, such as attack behaviour or selfgrooming. Self-grooming can be evoked in the paraventricular nucleus (PVH) and dorsal hypothalamic area (DHA) 13"2~'22"2~ Several descending and ascending pathways have been described that may be involved in the execution of these behavioural responses. One particular pathway runs from the hypothalamus through the periaqueductal gray area (PAG) and the adjoining parts of the brain stem ~s. The PAG seems to be involved in aggressive and defensive behaviour 2"3'~7 and has been shown to be crucial for ACTH-induced grooming 25. However, Mos and coworkers x8 showed that PAG lesions only slightly and transiently reduce aggression evoked by hypothalamic stimulation and aggression in a territorial setting (resident/intruder * Corresponding author. Fax: (31) (71) 276292. ** Present address: Dept. of Psychopharmacology, Tufts University, Medford, USA.
paradigm). Therefore, it seemed worthwhile to investigate the effect of P A G lesions on hypothalamically induced grooming behaviour, to study the role of the PAG in the execution of self-grooming.
MATERIALS A N D M E T H O D S
Surgery Experimental procedures have been described by Kruk etal. 12 and Mos etal. ~s. Fifteen male rats (Wistar/Harlan, Zeist) weighing 400-500 g were implanted bilaterally with bipolar electrodes (150 #m) aimed at the co-ordinates AP 7.40 mm ML 0.50 mm & DV 1.90 mm according to the atlas of Paxinos and Watson ~9. For details about the electrodes and connectors used see Kruk et al. I'~-. During surgery a piece of india rubber was placed on the skull behind the electrodes and embedded with a thin layer of dental cement, to allow a second operation to implant a pair of lesioning electrodes in the PAG. After surgery rats were housed individually in Macrolon cages in lownoise rooms at 22 ° C and 750Jjo relative humidity. Food and water were available ad libitum and an inverted
96 day/night cycle was installed, with lights on from 15.00 h to 03.00 h.
Behavioural testing and electrode selection Behavioural testing was started 1 week after surgery. Animals were placed in perspex stimulation cages (diameter 40 cm, height 60 cm, closed foam polypropylene floor). Rats were stimulated with trains of 40 Hz biphasic square-wave pulses with a phase duration of 0.2 ms and a phase interval of 12.5 ms. At the first test train duration was 30 s with 60 s intervals; at subsequent tests train duration was decreased at least to 15 s (and usually to 10 s). Threshold current intensity needed to evoke a grooming response was determined using the up-and-down method of Dixon and Mood 7 as modified by Wetheril135. Animals were stimulated only when they were not grooming spontaneously'. If the? did groom, stimulation was postponed until they had stopped. Subsequent threshold determinations on consecutive days were used to assess the stability of baseline thresholds. Electrode sites from which no stable thresholds on 3 subsequent days could be obtained, as well as sites at which no grooming within 15 s upon start of stimulation was evoked were excluded from the experiment. In animals with two grooming-evoking electrodes the electrode with lowest threshold current intensity for grooming was included in the data analysis. PA G lesions Thirteen animals with stable baseline thresholds for grooming were implanted bilaterally with two monopolar electrodes aimed at the PAG at the co-ordinates AP 3.00 mm, ML 0.60 ram, DV 4.20 mm according to Paxinos and Watson ~9. The teflon-coated silver electrodes (diameter 0.33 mm) with 0.5 mm bare tips were checked carefully for current leakage before implantation. After hardening of the dental cement the electrodes were connected to a Grass LM4 radiofrequency lesion maker, while the animal was still under anesthesia (for detailed description of the procedure, see Mos et al.~S). Lesions were made by passing a 100 kHz current for 20 s between the PAG-electrodes. During the 20 s the current was kept at approximately 20 mA (at a voltage of 40-80 Volt). The animals were allowed to recover for at least 1 week after surgery. All animals received water, applied orally via a syringe, and wet mesh the first day after surgery, because they failed to drink and eat by themselves. If necessary, this was continued until the animals had recovered and started to regain weight. The behavioural and postural consequences of the lesion were monitored during the post-
operative recovery period. Specml attention was paid to the reaction of the rats upon approach and touch by the experimenters hand during handling.
Postlesion behavioural testing Lesion effects on stimulation-induced behaviour were determined 7 and 14 day~ after surgery. Lesion effects were assessed as changes in threshold current required to evoke grooming. An increase in threshold means that the lesion inhibits grooming; a decrease would indicate a facilitation of grooming. Latencies of stimulationevoked grooming were measured during thresholds determination. A two-way ANOVA was used to detect lesion effects on hypothalamic grooming thresholds and latencies, compared with the last three threshold determinations before lesioning. Histology After completion of the experiment the rats were perfused with physiological saline followed by a 4'3o Ibrmaldehyde solution. Twenty-#m sections of the brain were stained with Luxol fast blue and Cresyl violet. The electrode tips were localized precisely and drawn on a detailed cytoarchitectonic atlas of the rat hypothalamus ~9. Lesions were drawn at plates of the atlas of Paxinos and Watson, Lesions were indicated to be complete if the PAG had disappeared completely at the levels AP 2.70 mm and[or 2.96 mm. so that we could be sure that all hypothatamic fibres descending through the PAG were interrupted. AnimNs with incomplete lesions were excluded from data analysis.
RESULTS
Basal grooming response to electrical stimulation At 19 of 30 electrodes that were implanted in the hypothalamic paraventricular area. reliable thresholds for grooming were obtained. At these electrodes grooming could be evoked within 15 s after onset of stimulation with stable thresholds, the range of these thresholds being 14.4 to 83.3 #A. The exact localization of electrode tips in the hypothalanms is shown in Fig. l. All grooming evoking electrodes were located in or close to the paraventricular nucleus: electrodes at which no reliable thresholds for grooming could be determined or no grooming at all were located at greater distances from the PVH (data not shown). This is in concordance with previous results t3"2~. Electrodes in rats that died during or shortly after lesioning had a similar anatomical distribution as electrodes m rats that were u s e d t o determine postlesion thresholds for grooming (Fig. 1).
97
,,
r ~ , ~
ITe lY
Fig. 1. Localization of electrode sites in the hypothalamic paraventricular area at which reliable grooming thresholds were determined within 15 s after onset of stimulation. Sites are drawn on adjacent sections (150 #m) of a detailed cyto-architectonic atlas of the hypothalamus *9. Left panel: anterior part; right panel: posterior part. Each symbol represents one electrode tip. Circles: sites in rats that were used for postlesion threshold determinations. Triangles: sites in rats that did not survive lesioning. AHA, anterior hypothalamic area; ARH, arcuate nucleus: BST, bed nucleus of stria terminalis; DHA, dorsal hypothalamic area; fx, fornix; PVH, paraventricular hypothalamic nucleus; SCN, suprachismatic nucleus; THAL, thalamus; vlll, third ventricle; VMH, ventromedial hypothalamic nucleus.
grooming thresholds or latencies (Fig. 2) 7 or 14 days after lesioning (thresholds: P = 0.4874 F = 0.891, dr= 4; latencies: P=0.8965 F=0.266, d f = 4 ; two-way ANOVA), as measured at 6 electrodes in 6 animals with complete lesions. Additional data from 4 other electrodes in the same animals gave similar results. In Fig. 2 pooled data for these 10 electrodes are presented (thresholds: P=0.4202 F = 1.000, d r = 4 ; latency: P = 0.2065 F = 1.558, dr= 4). Although some animals had small motor problems and would topple over more often while grooming, this apparently did not affect their ability to respond properly upon stimulation. They started to groom at the same latency and at the same current threshold intensity as before lesioning. We could detect no changes in stimulation-evoked grooming nor in spontaneous grooming patterns as performed in between stimulation trials.
Effect of PA G lesions Periaqueductal gray lesions had a quite severe effect on the behaviour of the rats. Two rats died during lesioning and 2 rats a few days after lesioning. The first day after lesioning all rats failed to drink and eat, and in some rats this would continue for several days. Rats received water intra-orally, wet mesh and supplementary saturated glucose-solution until their weight stabilized. Weight loss was rather extensive, from 50-80 g (in rats with starting weight of 400 g) or even 100 g (starting weight more than 500 g). Several animals had a problem in maintaining motor control and would topple over while grooming. The behavioural deficits disappeared slowly; after 7 days most animals behaved more or less normally, although a few still had small motor problems. PAG lesions had no effect on hypothalamic
grooming thresholds
grooming latencies
8O
15 [~
'~
So
"~ m
~o
e-
2O
prelesion - 3
12
A
prelesion - 2 9 [~
O
_=
prelesion - 1
6 3 0
post lesion day 7 post lesion day 14
Fig. 2. Left: effect of PAG lesion on thresholds for hypothalamic grooming. Bars represent pooled data of 10 electrodes in 6 rats with complete lesions (see text). Open bars: three last thresholds determined before lesioning. Hatched bars: thresholds at day 7 and day 14 after lesioning. Right: effect of PAG lesion on grooming latencies. Latencies were determined during threshold determinations. We observed no significant results between treatments (two-way ANOVA).
98 TABLE 1 S i z e a n d extension o f P A G lesion,s at several anterior-postertor levels o f the brain according io P a x i n o s a n d Watson ~
Plates 3.80 to 1.36 m m AP to the interaural point. Symbols: T = P A G totally destro 3 ed: P = P A G partly destroxed: - no lesion visible. Rats 8 and 9 were excluded because no stable thresholds for grooming were obtained. Rats 1, 3. 5 and 11 died during <~r shorttv after lesioning, Rat
('~ttl+Ilt'tll
Plate 3.80
3.40
3.20
2.96
_.~ +,(I
~.,~ ~,~"
7¢}
7n
1
T
Y
T
T~
T~
1
1
P
2
-
P
P
P
1 ~
P
P
P
4
-
P
1TM
I *
I
P
P
6 7
P
P T
T I-
T* I "
"f * 1~
I P
P P
P
10
T
~I
T
T ~
T ~
l
P
P
12 13 14 15
P P -
P P T P
P P F P
P P T'* P
P P 1~ P
P P P P
P P P
died
incomplete incomplete P P
incomplete
* Indicates complete lesions at plate 2.70 or 2.96.
A
cTx
~
B
C
L
9
Fig. 3. A: drawing of a moderate but complete P A G lesion {hatched area) in plates of the atlas of Paxinos and Watson 1~. Secuons of rat 14. drawn on plates 3.70, 2.70 and 1.70 mm interaural. B: same as ~. Moderate to large P A G lesion: rat 7 which had problems with motor coordination. Lesion extends more ventrally, including red nuclei. C: same as A. Very large P A G lesion: rat 1 wttich died shortly after lesiomng. Lesion includes red nuclei and parts of the medial superior colliculi, resulting in gliosis and cell death in the substantia nigra pars reticulata (stippled area). Aq, cerebral aqueduct: CTX. cortex: P A G , periaqueductal gray: RN. red nucleus: SC, superior colficulus: SN, substantia nigra: VTA. ventral tegrnental area.
99
Histoh)gical verification of lesion size The lesions appeared to be moderate to large in size. They extended from plate AP 1.70 mm to 3.80 mm in the atlas of Paxinos and Watson 19 (Table I). Moderate lesions were confined to the periaqueductal gray area, but sometimes the PAG was not destroyed completely at anterior levels plates 2.70 mm and 2.96 ram. An example of a complete, moderate lesion is given in Fig. 3A. Moderate to large lesions appeared to have partly destroyed the red nuclei (Fig. 3B). Not surprisingly, these animals had problems with maintaining motor control. One animal that died shortly after surgery appeared to have a very large lesion, including parts of the colliculi superiori, thalamus and red nuclei (Fig. 3C). In this animal, the damage to the superior colliculi may have been the cause of cell death and gliosis in the substantia nigra, pars reticulata. This was not observed in any of the other lesioned animals. After careful histological control, 6 animals were used in the data analysis of the effects of PAG lesions upon hypothalamically induced grooming.
DISCUSSION
Our study shows that lesions of the periaqueductal gray area have no effect on thresholds for hypothalamically evoked self-grooming behaviour. This is in line with the effect of PAG lesions on hypothalamic attack responses. Mos and coworkers is showed that PAG lesions only slightly and transiently reduce thresholds for hypothalamic attack. In general, hypothalamic responses may consist of either changes in the introductory phase of behaviour, e.g. after ventromedial hypothalamic stimulation 32"33 or consist of a clearcut response, often of a compulsive nature, e.g. attack or grooming responses ~L-~4"2s. Animals interrupt ongoing behaviour immediately and show the particular response elicited within seconds after onset of stimulation. Behaviourally it seems that a system involved in the execution phase of the behaviour, rather than the introductory or decision making phase, is activated l~ Which pathways are involved in these hypothalamic responses? Fibres descend towards the lower parts of the brainstem either via the ventral tegmental area and the central tegmental field, or along the midline via the PAG 15'2°'34. In addition, ascending fibres and fibres with a thalamic destination may play a role ls'2°, as well as other ascending projections to the forebrain. Interestingly, hypothalamic projections to the PAG show a topographical organization 4's'~6-2°,24,34 and the fibres originating from the hypothalamic attack area (caudal, lateral PAG) show a distribution that is very different
from fibres originating from the hypothalamic grooming area (dorsal and lateral PAG) 2°. Spruijt and coworkers 25 showed that the PAG is a prerequisite for ACTHinduced grooming. PAG lesions diminished the response to i.c.v, injections of ACTHl_24. However, grooming did not disappear completely, but was reduced to baseline level. There are several possible explanations for these seemingly contradictory results. Recent experiments, in which grooming was induced by direct injection of neuropeptides into the PVH 29, suggest that peptides such as ACTH and c~-MSH prolong a grooming response that was initiated by other factors, e.g. handling of the rat or exposure to a novel environment. In contrast, a peptide such as oxytocin seems to initiate grooming 3°. Apparently, separate systems exist in the central nervous system that regulate the initiation of grooming, its continuation and probably the interruption of this behaviour. We hypothesise that ACTH, for which binding sites are present in the PAG 26, serves to modulate grooming that has been initiated by other factors outside the PAG. This could explain the fact that after PAG lesions grooming is still performed at baseline level. The modulatory effect of the PAG on grooming has been confirmed recently by Van Wimersma Greidanus, who showed that PAG lesions attenuate ACTH- but not oxytocin-induced grooming 3~. Nevertheless, our results suggest that the PAG is not the major destination station involved in hypothalamically evoked grooming responses. A recent study by Roeling 23 showed that one of the many efferent pathways originating from the PVH that is possibly involved in the execution of grooming, is the connection descending via lateral hypothalamus, ventral tegmental area (VTA) and central tegmental field. Interestingly, the VTA is one of the areas of the brain that has been reported to be involved in grooming behaviour: injection of neuropeptides, such as oxytocin, CCK m and ~-MSH 27, into the VTA induces selfgrooming behaviour. The possible involvement of the VTA in the control of hypothalamic responses deserves special attention in future studies, although a similar study using VTA lesions may prove difficult to perform because of severe effects on autonomic functions. Lesions of the periaqueductal gray resulted in severe behavioural deficits. Some of these effects may have been caused by lesions of tissue surrounding the PAG, especially in animals with large lesions in which ascending serotonergic and/or noradrenergic pathways may have been disrupted. Animals didn't eat or drink (also reported by Mos et al. lS): they seemed to have "forgotten" how to do it, because they would drink readily when helped and showed interest in wet mesh after they had been drinking. It is interesting to note
100
that severn animals showed moderate to strong defensive reactions upon touch, despite the intensive handling period before surgery. They didn't try to bite, but upon touch by the experimenters hand tried to escape by running and jumping and sometimes they screamed when picked up. This observation is at odds with reports that PAG lesions attenuate defensive behaviour 3'6, for which we have no explanation. Others have reported that PAG lesions do not change defensive reactions induced by other means either, e.g. inferior colliculus stimulation ~. In addition, some animals showed reflex-like face washing movements upon stimulation of the mouth region with a wet syringe during drinking sessions. These movements were quite strong and compulsive, performed in a position in which a rat normally would never groom (restrained, lying on its back). These movements were not just attempts to remove the syringe, but included the repetitive "front paw over ear" movements that are characteristic for face washing. This is an important observation, showing that the animals are still responding to sensory stimuli. Apparently the sensory input pathways are intact, as well as the motor components of the grooming behaviour. In spite of the fact that they have a problem in maintaining their body posture, they still groom, even while lying on their back. We could not correlate the nature of observed behavioural deficits (defensive behaviour, face washing upon stimulation of mouth region) to the size or extension of the lesions. These clear deficits, however, did not affect the hypothNamically evoked grooming response, neither inhibitory nor facilitatory. ACKNOWLEDGEMENTS
The investigations were supported by the Foundation for Biological Research (BION), which is subsidized by the Netherlands Organization for Scientific Research (NWO). REFERENCES 1 Bagri, A., Di Scala, G. and Sandner, G., Wild running elicited by microinjections of bicuculline or morphine into the inferior colliculus of rats: lack of effect of periaqueductal gray lesions, Pharmacol. Biochem. Behav., 41 (1992) 727-732. 2 Bandler, R., Brain mechanisms of aggression as revealed by electrical and chemical stimulation: suggestions of a central role for the midbrain central grey region, Prog. Psychobiol. Physiol. P.~3'ehol., 13 (1988) 67-154. 3 Bandler, R. and Depaulis, A., Elicitation of intraspecific defense reactions in the rat from midbrain periaqueductal grey by microinjection of kainic acid, without neurotoxic effects, Neurosci. Lett., 88 (1988) 291-296.
4 Behbehani. M.M.. Park. M.R. and Clement. M.E.. Interactions between the lateral hypothalamus and the periaqueductal gray. J. Neurosci.. 8 (1988) 2780-2787. 5 Beitz. A.J.. The organizauon ot ~afferent projecuons to the midbrain periaqueductal gray of the r~t. Veuroscience. " ~/ 982) 133159. 6 Depaulis. A. and Bandler. R. tkds.L Ihe Midbruin Periaqueductai Gray Matter. Functional. Anatomical and Neurochemical Organiza~ non Plenum. New York, 199l. Dixon. W.J and Mood, A.M_ ~ method for obtaining and analyzing sensitivity data. J. Am, St¢lt. Assoc.. 43 (19481 109--126 Geeraedts, L . M . G . Nieuwenhu~s~ R, and Veening, J.G.. The medial forebrain bundle of the rat III. Cytoarchitecture of the rostral (telencephalic) part of the MFB bed nucleus_ L Comp. Neurot. 294 (19901 507-536. 9 Geeraedts. L.M.G.. Nieuwenhuvs. R. and Veening, ,1.(J., ]'he medial forebrain bundle of the rzlt IV C3toarchitecture of the caudal (lateral hypothalamic) part of the MFB bed nucleus, J Comp. Neurol. 294 (1990) 537-5~-~8 10 Kaltwasser. M.Th. and Crawler, J,N.. Oxytocin and cholecystokinin induce grooming behavior in the ventral tegmentum of the rat. Brain Res_ 426 (19871 1-7 I1 Kruk. M.R.. Ethology and pharmacology ot hypothalamic aggression in the rat. Neurosci. Bh~behav. Rev.. 15 (19911 527538 12 Kruk. M.R.. Van der Poel. '\.M and De Vos-Frcrichs. I.P.. The induction of aggressive behaviour b) electric:d stimulation in the hypothalamus of male rats, Beh~rt'imo'. 70 119791 292-322. 13 Lammers. J.H.C.M.. Kruk. M . R . Meelis. W. and Van der Pool. A.M.. Hypothalamic subs trates rot brain stimulation-induced grooming, digging and circling in the rat, Brain Rea.. 418 (19871 1-19. 14 Lammers. J.H.C.M., Kruk. M.R.. Meelis. W. and Van der Poel. A.M.. Hypothalamic substrates for brain stimulation-induced attack, teeth-chattering and social grooming in the rat. Brain Res.. 449 ( 19881 311-327. 15 kuiten. P.G.M.. Ter Horst. G . J , Karst. H. and Stefl'ens. A.B.. The course of paraventricular hypothalamic efferents to autonomic structures m medulla and spinal cord. Brain Re.~. 329 t 19851 374-378. 16 Morrell. J.I.. Greenberger, L.M. and PfaflL D.W., Hypothalamic. other diencephalic, and telencepha/ic neurons that projecl to the dorsat midbrain. J. Comp. Neurol.. 201 ( 1981 ) 589-620. 17 Mos. J.. Kruk. M.R.. Van der Poel. A.M. and Meelis. W.. Aggressive behaviour induced bv eleelrical stimulation in the midbrain central gray of male rz~ts. Aggress. Behav.. ,~ (t982) 261-284 I~; Mos, J., Lammers, J.H.C.M.. Van der Poel. A.M.. Bermond. B.. Meelis. W and Kruk. M.R.. Effects of midbrain central gra3 lesions on spontaneous and electrically induced aggression in the rat. Aggress. Behav.. 9 (19831 133-155. 19 Paxinos, G. and Watson. C.. The Rat Brain in Stereotaxi( Coordinates. 2rid. edn. Academic Press. Sydney, 1986. 20 Roeling, T.A.P.. The Neuronal Substrate ol Hypothalamk'all3 Induced Behaviour #t the Rat. A Behavioral and Anatomical Study. Thesis, Nijmegen. 1993. 21 Roeling, T.A.P.. Van Erp, A.M.M.. Meelis. W.. Kruk. M.R. and Veening, J.G.. Behavioural effect~ of N M D A injected into the hypothalamic paraventricular nucleus of the rat. Bra#~ Res.. 550 19911 220-224. 22 Roeling, T.A.P.. Veening, J.G., Kruk. M.R. and Nieuwenhuys, R.. Grc~ming bebaviour elicited b 5 kainic acid evoked cell body
101
23
24
25
26
27
28
29
stimulation in the hypothalamus of the rat, Neurosci. Res. Cornmum, 6 (1990) 11-118. Roeling, T.A.P., Veening, J.G,, Peters, J.P.W., Vermelis, M.E.J. and Nieuwenbuys, R., Efferent connections of the hypothatamic "'grooming area" in the rat, Neuroscience, 56 (1993) 56. Semenenku, F.M. and Lmnb, B.M., Projections of anterior hypothalamic neurones to the dorsal and ventral periaqueductal grey in the rat, Brain Re.s., 582 (1992) 237-245. Spruijt, B.M., Cools, A.R. and Gispen, W.H., The periaqueductal gra3.: a prerequisite for ACTH-induced excessive grooming, Behav. Brain Res.. 20 (1986) 19-25. Tatro, .I.B., Melanotropin receptors in the brain are differentially distributed and recognize both corticotropin and :~-melanocyte stimulating homaone, Brui~l Res.. 536 (1990) 124-132. Torre, E. and Celis, M.E., Cholinergic mediation in the ventral tegmental area of ~.-melanotropin induced excessive grooming: changes in the dopamine activity in the nucleus accumbens and caudatc putamen, L([;" Sci., 42 (1988) 1651-1657. Van Erp, A.M.M., Kruk, M.R., Willekens-Bramer, D.C., Bressers, W.M.A., Roeling, T.A.P., Veening, J.G. and Spruijt, B.M., Grooming induced by intrahypothalamic injection of A C I H in the rat: comparison with grooming induced by intrahypothalamic electrical stimulation and i.c.v, injection of ACTH, Brain Res.. 538 (1991) 203-210. Van Erp, A.M.M.. Kruk, M.R,, Van Oers, H.J.J. and Hemmers,
30
31
32 33
34
35
N.M., Differential effect of ACTHl_24 and ~-MSH in PVH induced grooming, Brain Res.. 603 (1993) 296-301. Van Erp, A.M.M., Kruk, M.R., Semple, D.M. and Verbeet, D.W.P., Initiation of grooming in resting rats by i.c.v, oxytocin but not by ACTH I _,4 and :~-MSH, Brain Re,v.. 607 (1993) 108112. Van Wimersma Greidanus, Tj.B. and Maigret, C., Differential role of the periaqueductal gray in the grooming inducing effect of neuropeptides, Neurosci. Res. ()mml., 11 (1992) 145-153. Veening, J.G., Behavioural Function.~ qf the VMH in the Rat: An Ethological Approach, Thesis, University of Groningen, 1975. Veening, J.G., Brain and behaviour: morphological and functional aspects of the hypothalamus in thc rat, Eur. J. Morphol., 30 (1992) 53-66. Veening, J.G., Buma, P., Ter Horst, G.J., Roeling, T.A.P., Luiten, P.G.M. and Nieuwenhuys, R., Hypothalamic projections to the PAG in the rat: topographical, immuno-electromnicroscopical and functional aspects. In A. Depaulis and R. Bandler (Eds.), The Midbra#7 Periaqueductal Gray Matter. Plenum, New York, 1991, pp. 387-415. Wetherill, G.B., Sequential estimation of points on quantal response curves. In G. Barrie and G.B. Wetherill (Eds.), Sequential Methods #7 Statistics, Methuen and Co., London, 1966, pp. 162-179.