Brain Research, 106 (1976) 57-69 ~) Elsevier ScientificPublishing Company, Amsterdam - Printed in The Netherlands
57
H Y P O T H A L A M I C A N D E X T R A H Y P O T H A L A M I C SUBSTRATES OF PRED A T O R Y ATTACK. SUPPRESSION A N D T H E I N F L U E N C E OF H U N G E R *
ROBERT E. ADAMEC
Psychology Department, Dalhousie University, Halifax, Nova Scotia (Canada) (Accepted September 15th, 1975)
SUMMARY
Electrical stimulation of medial hypothalamic and ventromedial hypothalamic areas of the cat brain stops the initiation of spontaneous predatory attack in cats, confirming similar evidence of other investigators. Furthermore, a new attack suppressing area, the mammillary bodies, was uncovered. Facilitation of predatory attack by hunger raised the electrical threshold for attack in the mammillary bodies. In addition, baseline levels of neural activity in attack suppressing brain areas prior to any brain stimulation were found to decrease when the cats were hungry and killing was facilitated and neural activity increased when the cats were on ad lib. feeding. These data support the hypothesis that modulation of excitability of neural systems functioning to suppress attack is involved in facilitation of attack behavior by hunger.
INTRODUCTION
A number of midline thalamic, hypothalamic and midbrain areas have been implicated in suppression of feline predatory attack 4,5,1a. The present investigation confirms these findings and extends the neural substrate of attack suppression to include the mammillary bodies. Furthermore the effects of hunger on the effectiveness of behavioral suppression produced by direct electrical activation of the inhibitory neural substrate was investigated. The hypothesis tested was that a motivational state which facilitates attack 1, 6-18 should alter the behavioral suppression produced by excitation of this inhibitory neural system. * Part of a doctoral dissertation, McGill University, 1974.
58 METHODS
Subjects Eleven adult male cats were used. Four spontaneously attacked and killed rats within 120 sec. Seven did not kill rats within a 10 rain exposure period. Predatory disposition of all cats had been tested repeatedly (a minimum of 3 times per week.) for 4 months and was found to be stable prior to the present experiment.
Surgical procedure All cats were implanted with a bipolar twisted platinum-10 ~ iridium electrodes (0.007 in. in diameter with a 0.003 in. coating of teflon) using sterile stereotaxic procedures (Kopf stereotaxic instrument). Cats were anesthetized with a combination of Nembutal (30 mg/kg) and acepromazine maleat (Atravet). Electrodes were aimed at the ventromedial hypothatamus, midtine hypothalamus, and mammillary bodies using the atlas of Jasper and Ajmone Marsan. Postoperative care included daily subcutaneous injections of 30 ml of 10 ~o dextrose and 90 ml of amino acid solution until spontaneous eating resumed. Four consecutive daily injections of 100 mg of chloramphenicol succinate (Mycinol) and 100 #g of cyanocobalamin (Vitamin B12) were also given. Following 4 weeks of recovery, postoperative behavioral measures were taken and no change in behavior as a result of surgery was found.
General method of testing All testing was carried out in the same box (Fig. 1). Cats were adapted to sitting in the box with stimulating and recording cables attached for a minimum of 8 h prior to any testing. Two days prior to electrical brain stimulation baseline measures of predatory attack behavior against mice were taken in 4 consecutive predatory tests. Tests were separated by 5 min intervals. Two days later adequate intensities for inhibitory effects of stimulation of the hypothalamic points on preying were determined. On the next day these effects were systematically examined in 4 consecutive tests given in the same manner as were the behavioral baseline tests. Another 4 consecutive tests with mice in the absence of brain stimulation were given two days later. If a cat did not kilt mice spontaneously, he was induced to do so with 48 h of food deprivation. In these cases, both baseline measures and test response with brain stimulation were carried out while the cats were hungry. Measures of predatory behavior were latency to bite prey (sec), latency to kill, timed from the first bite (sec) and the proportion of time near the prey spent biting the prey. Rate at which the prey were eaten after a kill was also recorded as weight of prey consumed/sec of time spent eating (g/sec). It should be noted that the attack behavior of any cat in this situation did not differ qualitatively or quantitatively from the baseline taken for 4 months prior to this experiment.
Recording procedure Prior to any brain stimulation 2 min of baseline multiple unit activity was differ-
59 /f~..
to EEG&MS
/ .
prey box Fig. 1. Pictured is the testing box in which stimulating and recording procedures were done. One side of the box was a sliding plexiglass door through which the behavior of the cat was observed. During brain stimulation, prey were introduced by partially opening the door and placing the prey on the floor in the visual field of the cat. Recording of electrical activity was done in the same box with the prey absent.
entially recorded from all brain areas while cats were satiated, then food deprived for 48 h, then resatiated for 48 h. Activity was integrated with Ballantyne mean square meters and displayed on model 7 Grass polygraph for later analysis. Furthermore, in two cats records of EEG and multiple unit activity were taken following stimulation. There was no evidence of seizure activity in these records. Moreover, throughout the experiment there was no behavioral evidence of seizures during stimulation such as immobility, or facial or limb clonus.
Procedure for determining stimulation intensities which inhibited attack The first day of stimulation of the brain involved finding current levels, if any, which would produce inhibition of predatory attack. Stimulation was switched on for 10 sec and then live prey introduced to the cat in the testing box. Cats were initially stimulated with 62 Hz train of biphasic constant current pulses (1 msec pulse width) of 0.4 mA peak-to-peak intensity generated by a Grass $8 stimulator and then given live prey during the stimulation. If no behavioral effects were observed, the intensity was doubled on successive tests until inhibition of killing or adverse motor or affective displays (including seizures) were observed. If such adverse effects were observed at the initial stimulating intensity, the current was lowered. If forced motor movements occurred before or during an apparent inhibition of attack, the neural area stimulated was not considered to be inhibitory to killing and testing was discontinued. All tests with stimulation were separated by a 5 rain period beginning at the termination of stimulation. At least two stimulations yielding attack suppression were required before a current intensity was considered inhibitory to attack.
Behavioral testing procedure Two days after current levels which blocked the killing of a mouse were estab-
60 lished, systematic behavioral observations of the effects of stimulation were conducted. The previously established inhibitory current was switched on 10 sec before the introduction of prey and left on for a maximum of 40 sec, if no biting attack occurred .... or a maximum of time which exceeded the maximum baseline latency to k i l l - - i f a biting attack occurred. Thus there were a total of 6 demonstrations of attack suppression on two separate days. All behavioral measures were taken both during stimulation, and, for most animals, immediately following its cessation. These 'post-stimulation' behavioral measures were taken in order to determine whether any effects of the stimulation persisted after its termination. If no attack upon the prey occurred after offset of the brain stimulation, a maximum of 5 additional minutes of exposure to the prey was allowed before its removal. This procedure was repeated with 4 different prey, using the same stimulating current. Each successive test was begun 5 rain after the prey used in the previous test were either devoured entirely or removed (after 2 rain). Following retesting without brain stimulation, the effects of hunger on stimulation-induced suppression of preying were examined in some cats. Cats were deprived of food for 48 h and then retested for the influence of brain stimulation on attack. If the stimulation no longer suppressed attack, the current was raised in 0.4 mA peakto-peak steps until suppression of attack was again observed. Two days of ad lib. feeding followed. The current level necessary to inhibit preying was again determined by first stimulating the cats at the intensity which was originally found to inhibit preying, and adjusting the level if necessary. If a cat were already deprived of food at the initiation of testing for the effects of hunger, he was deprived for an additional 24 h.
RESULTS
Behavioral effects of brain stimulation The effects of stimulation on the various behavioral indices were analyzed for each animal separately. The 4 prestimulation behavioral tests were compared (median test) with the tests made during brain stimulation and with tests immediately following stimulation. The results of these comparisons appear in Table 1. The location of electrodes tested is listed in the table and also appears in Fig. 2. The analysis reveals that stimulation of diverse areas of the brain, including areas outside of the hypothalamus, inhibit predatory attack. Stimulation of the ventromedial, anterior medial and posterior hypothalamus will suppress muricidal attack, as well as stimulation of the fornix, mammillary bodies and mammillary peduncle. In addition, sites which yielded inhibition of mouse killing were also tested for inhibition of rat killing in all cats which were killing rats at the time of testing. These sites, namely the ventromedial hypothalamus and mammillary bodies, also inhibited predatory attack upon rats. It is apparent from the table that all behavioral measures were affected by brain
61 TABLE I
Cat
A.
Electrode placement
Stimulating current (mA)
Latency to bite
Latency to kill
% biting
Rate eating prey
T A B L E O F B R A I N SITES Y I E L D I N G I N H I B I T I O N O F M O U S E A N D R A T K I L L I N G I N R A T K I L L E R S
Balem
Right ventromedial hypothalamus
0.08
Pre Stim Post
0.8 30.0* 9.7 NS
10.8 94.3 0.6 30.0* 0.0" 0.0" ll.8 NS 97.7 NS 0.5 NS
Govinda Left mammillary bodies 2.10
Pre Stim Post Pre Stim Post
1.0 30.0* 2.6 NS 1.0 5.9*
12.6 30.0* 8.8 NS 9.6 23.6*
1.5 30.0* 4.9 NS
12.5 85.5 1.0 30.0* 0.0" 0.0" 15.7 NS 81.8 NS 1.5 NS
Fornix
2.10
Lupus
Bilaterally in mamillary bodies
1.20
Pre Stim Post
Malbob
Right posterior hypothalamus
2.00
Pre Stim Post
B.
0.8 4.8*
21.2 68.7*
98.0 1.3 0.0" 0.0" 93.0 NS 1.4 NS 98.8 1.0 79.5 NS 1.9 NS
99.3 87.3*
0.3 0.3 NS
T A B L E O F B R A I N SITES Y I E L D I N G I N H I B I T I O N O F M O U S E K I L L I N G I N N O N - R A T - K I L L E R S
Balthaar Right medial hypothala- 0.4 mus at the level of anterior hypothalamus
Pre Stim Post
1.8 30.0* 10.5"
22.0 30.0* 27.8 NS
Bubastis Left mammillary bodies 0.4
Pre Stim Post
2.4 30.0* 20.3*
7.8 86.0 0.6 30.0* 0.0" 0.0" 16.1 NS 78.0 NS 0.5 NS
Hegel
Pre Stim Post
2.0 30.0* --
21.0 30.0*
22.6 73.8 0.4 30.0* 9.8* 0.0" 41.0 NS 45.5 NS 0.4 NS
Left nucleus mammillary peduncle
Heideger Left mammillary peduncle
2.7
Pre Stim Post
6.1 27.0* 99.1"
Jaganti
Nucleus interpeduncularis
1.00
Pre Stim Post
1.6 33.1 * 23.2*
Strider
Right mammillary body 0.4
Pre Stim Post
7.9 30.0* 57.8*
Yamaha
Right mammillary body 2.4
Pre Stim Post
5.8 30.0* 117.4"
23.3 40.0* 100.2"
82.0 0.1 0.0" 0.0" 53.0 NS 0.1 NS
85.0 0.0"
92.0 10.5" 29.0*
0.6 0.0"
0.3 0.0" 0.5 NS
16.9 93.8 0.2 30.0* 0.0" 0.0" 55.5 NS 27.8 NS 0.3 NS 16.5 30.0* 233.5*
99.5 0.0" 24.0*
0.6 0.0" 0.1 NS
62 (continued table 1) C,
FREQUENCY OF DELETERIOUS AFTEREFFECTS OI: INHIBITORY STIMULATION OF THE BRAIN
Group
N 7~sted
Number showing deleterious aftereffects
Z e df
Rat-killers Non-rat-killers
3 6
0 6
5.50*
I
Stimulating current needed to produce inhibition of preying Mean peak-to-peak intensity(mA) Mann-Whitney U
Rat-killers Non-rat-killers * P **
P
3 5
1.13 1.25
6
0.014.
57
H Y P O T H A L A M I C A N D E X T R A H Y P O T H A L A M I C SUBSTRATES OF PRED A T O R Y ATTACK. SUPPRESSION A N D T H E I N F L U E N C E OF H U N G E R *
ROBERT E. ADAMEC
Psychology Department, Dalhousie University, Halifax, Nova Scotia (Canada) (Accepted September 15th, 1975)
SUMMARY
Electrical stimulation of medial hypothalamic and ventromedial hypothalamic areas of the cat brain stops the initiation of spontaneous predatory attack in cats, confirming similar evidence of other investigators. Furthermore, a new attack suppressing area, the mammillary bodies, was uncovered. Facilitation of predatory attack by hunger raised the electrical threshold for attack in the mammillary bodies. In addition, baseline levels of neural activity in attack suppressing brain areas prior to any brain stimulation were found to decrease when the cats were hungry and killing was facilitated and neural activity increased when the cats were on ad lib. feeding. These data support the hypothesis that modulation of excitability of neural systems functioning to suppress attack is involved in facilitation of attack behavior by hunger.
INTRODUCTION
A number of midline thalamic, hypothalamic and midbrain areas have been implicated in suppression of feline predatory attack 4,5,1a. The present investigation confirms these findings and extends the neural substrate of attack suppression to include the mammillary bodies. Furthermore the effects of hunger on the effectiveness of behavioral suppression produced by direct electrical activation of the inhibitory neural substrate was investigated. The hypothesis tested was that a motivational state which facilitates attack 1, 6-18 should alter the behavioral suppression produced by excitation of this inhibitory neural system. * Part of a doctoral dissertation, McGill University, 1974.
58 METHODS
Subjects Eleven adult male cats were used. Four spontaneously attacked and killed rats within 120 sec. Seven did not kill rats within a 10 rain exposure period. Predatory disposition of all cats had been tested repeatedly (a minimum of 3 times per week.) for 4 months and was found to be stable prior to the present experiment.
Surgical procedure All cats were implanted with a bipolar twisted platinum-10 ~ iridium electrodes (0.007 in. in diameter with a 0.003 in. coating of teflon) using sterile stereotaxic procedures (Kopf stereotaxic instrument). Cats were anesthetized with a combination of Nembutal (30 mg/kg) and acepromazine maleat (Atravet). Electrodes were aimed at the ventromedial hypothatamus, midtine hypothalamus, and mammillary bodies using the atlas of Jasper and Ajmone Marsan. Postoperative care included daily subcutaneous injections of 30 ml of 10 ~o dextrose and 90 ml of amino acid solution until spontaneous eating resumed. Four consecutive daily injections of 100 mg of chloramphenicol succinate (Mycinol) and 100 #g of cyanocobalamin (Vitamin B12) were also given. Following 4 weeks of recovery, postoperative behavioral measures were taken and no change in behavior as a result of surgery was found.
General method of testing All testing was carried out in the same box (Fig. 1). Cats were adapted to sitting in the box with stimulating and recording cables attached for a minimum of 8 h prior to any testing. Two days prior to electrical brain stimulation baseline measures of predatory attack behavior against mice were taken in 4 consecutive predatory tests. Tests were separated by 5 min intervals. Two days later adequate intensities for inhibitory effects of stimulation of the hypothalamic points on preying were determined. On the next day these effects were systematically examined in 4 consecutive tests given in the same manner as were the behavioral baseline tests. Another 4 consecutive tests with mice in the absence of brain stimulation were given two days later. If a cat did not kilt mice spontaneously, he was induced to do so with 48 h of food deprivation. In these cases, both baseline measures and test response with brain stimulation were carried out while the cats were hungry. Measures of predatory behavior were latency to bite prey (sec), latency to kill, timed from the first bite (sec) and the proportion of time near the prey spent biting the prey. Rate at which the prey were eaten after a kill was also recorded as weight of prey consumed/sec of time spent eating (g/sec). It should be noted that the attack behavior of any cat in this situation did not differ qualitatively or quantitatively from the baseline taken for 4 months prior to this experiment.
Recording procedure Prior to any brain stimulation 2 min of baseline multiple unit activity was differ-
59 /f~..
to EEG&MS
/ .
prey box Fig. 1. Pictured is the testing box in which stimulating and recording procedures were done. One side of the box was a sliding plexiglass door through which the behavior of the cat was observed. During brain stimulation, prey were introduced by partially opening the door and placing the prey on the floor in the visual field of the cat. Recording of electrical activity was done in the same box with the prey absent.
entially recorded from all brain areas while cats were satiated, then food deprived for 48 h, then resatiated for 48 h. Activity was integrated with Ballantyne mean square meters and displayed on model 7 Grass polygraph for later analysis. Furthermore, in two cats records of EEG and multiple unit activity were taken following stimulation. There was no evidence of seizure activity in these records. Moreover, throughout the experiment there was no behavioral evidence of seizures during stimulation such as immobility, or facial or limb clonus.
Procedure for determining stimulation intensities which inhibited attack The first day of stimulation of the brain involved finding current levels, if any, which would produce inhibition of predatory attack. Stimulation was switched on for 10 sec and then live prey introduced to the cat in the testing box. Cats were initially stimulated with 62 Hz train of biphasic constant current pulses (1 msec pulse width) of 0.4 mA peak-to-peak intensity generated by a Grass $8 stimulator and then given live prey during the stimulation. If no behavioral effects were observed, the intensity was doubled on successive tests until inhibition of killing or adverse motor or affective displays (including seizures) were observed. If such adverse effects were observed at the initial stimulating intensity, the current was lowered. If forced motor movements occurred before or during an apparent inhibition of attack, the neural area stimulated was not considered to be inhibitory to killing and testing was discontinued. All tests with stimulation were separated by a 5 rain period beginning at the termination of stimulation. At least two stimulations yielding attack suppression were required before a current intensity was considered inhibitory to attack.
Behavioral testing procedure Two days after current levels which blocked the killing of a mouse were estab-
60 lished, systematic behavioral observations of the effects of stimulation were conducted. The previously established inhibitory current was switched on 10 sec before the introduction of prey and left on for a maximum of 40 sec, if no biting attack occurred .... or a maximum of time which exceeded the maximum baseline latency to k i l l - - i f a biting attack occurred. Thus there were a total of 6 demonstrations of attack suppression on two separate days. All behavioral measures were taken both during stimulation, and, for most animals, immediately following its cessation. These 'post-stimulation' behavioral measures were taken in order to determine whether any effects of the stimulation persisted after its termination. If no attack upon the prey occurred after offset of the brain stimulation, a maximum of 5 additional minutes of exposure to the prey was allowed before its removal. This procedure was repeated with 4 different prey, using the same stimulating current. Each successive test was begun 5 rain after the prey used in the previous test were either devoured entirely or removed (after 2 rain). Following retesting without brain stimulation, the effects of hunger on stimulation-induced suppression of preying were examined in some cats. Cats were deprived of food for 48 h and then retested for the influence of brain stimulation on attack. If the stimulation no longer suppressed attack, the current was raised in 0.4 mA peakto-peak steps until suppression of attack was again observed. Two days of ad lib. feeding followed. The current level necessary to inhibit preying was again determined by first stimulating the cats at the intensity which was originally found to inhibit preying, and adjusting the level if necessary. If a cat were already deprived of food at the initiation of testing for the effects of hunger, he was deprived for an additional 24 h.
RESULTS
Behavioral effects of brain stimulation The effects of stimulation on the various behavioral indices were analyzed for each animal separately. The 4 prestimulation behavioral tests were compared (median test) with the tests made during brain stimulation and with tests immediately following stimulation. The results of these comparisons appear in Table 1. The location of electrodes tested is listed in the table and also appears in Fig. 2. The analysis reveals that stimulation of diverse areas of the brain, including areas outside of the hypothalamus, inhibit predatory attack. Stimulation of the ventromedial, anterior medial and posterior hypothalamus will suppress muricidal attack, as well as stimulation of the fornix, mammillary bodies and mammillary peduncle. In addition, sites which yielded inhibition of mouse killing were also tested for inhibition of rat killing in all cats which were killing rats at the time of testing. These sites, namely the ventromedial hypothalamus and mammillary bodies, also inhibited predatory attack upon rats. It is apparent from the table that all behavioral measures were affected by brain
61 TABLE I
Cat
A.
Electrode placement
Stimulating current (mA)
Latency to bite
Latency to kill
% biting
Rate eating prey
T A B L E O F B R A I N SITES Y I E L D I N G I N H I B I T I O N O F M O U S E A N D R A T K I L L I N G I N R A T K I L L E R S
Balem
Right ventromedial hypothalamus
0.08
Pre Stim Post
0.8 30.0* 9.7 NS
10.8 94.3 0.6 30.0* 0.0" 0.0" ll.8 NS 97.7 NS 0.5 NS
Govinda Left mammillary bodies 2.10
Pre Stim Post Pre Stim Post
1.0 30.0* 2.6 NS 1.0 5.9*
12.6 30.0* 8.8 NS 9.6 23.6*
1.5 30.0* 4.9 NS
12.5 85.5 1.0 30.0* 0.0" 0.0" 15.7 NS 81.8 NS 1.5 NS
Fornix
2.10
Lupus
Bilaterally in mamillary bodies
1.20
Pre Stim Post
Malbob
Right posterior hypothalamus
2.00
Pre Stim Post
B.
0.8 4.8*
21.2 68.7*
98.0 1.3 0.0" 0.0" 93.0 NS 1.4 NS 98.8 1.0 79.5 NS 1.9 NS
99.3 87.3*
0.3 0.3 NS
T A B L E O F B R A I N SITES Y I E L D I N G I N H I B I T I O N O F M O U S E K I L L I N G I N N O N - R A T - K I L L E R S
Balthaar Right medial hypothala- 0.4 mus at the level of anterior hypothalamus
Pre Stim Post
1.8 30.0* 10.5"
22.0 30.0* 27.8 NS
Bubastis Left mammillary bodies 0.4
Pre Stim Post
2.4 30.0* 20.3*
7.8 86.0 0.6 30.0* 0.0" 0.0" 16.1 NS 78.0 NS 0.5 NS
Hegel
Pre Stim Post
2.0 30.0* --
21.0 30.0*
22.6 73.8 0.4 30.0* 9.8* 0.0" 41.0 NS 45.5 NS 0.4 NS
Left nucleus mammillary peduncle
Heideger Left mammillary peduncle
2.7
Pre Stim Post
6.1 27.0* 99.1"
Jaganti
Nucleus interpeduncularis
1.00
Pre Stim Post
1.6 33.1 * 23.2*
Strider
Right mammillary body 0.4
Pre Stim Post
7.9 30.0* 57.8*
Yamaha
Right mammillary body 2.4
Pre Stim Post
5.8 30.0* 117.4"
23.3 40.0* 100.2"
82.0 0.1 0.0" 0.0" 53.0 NS 0.1 NS
85.0 0.0"
92.0 10.5" 29.0*
0.6 0.0"
0.3 0.0" 0.5 NS
16.9 93.8 0.2 30.0* 0.0" 0.0" 55.5 NS 27.8 NS 0.3 NS 16.5 30.0* 233.5*
99.5 0.0" 24.0*
0.6 0.0" 0.1 NS
62 (continued table 1) C,
FREQUENCY OF DELETERIOUS AFTEREFFECTS OI: INHIBITORY STIMULATION OF THE BRAIN
Group
N 7~sted
Number showing deleterious aftereffects
Z e df
Rat-killers Non-rat-killers
3 6
0 6
5.50*
I
Stimulating current needed to produce inhibition of preying Mean peak-to-peak intensity(mA) Mann-Whitney U
Rat-killers Non-rat-killers * P **
P
3 5
1.13 1.25
6
0.014.
