- Email: [email protected]
Intermale social aggression: Suppression by medial preoptic area lesions
Physiology & Behavior, Vol. 38, pp. 16%173. Copyright©PergamonPress Ltd., 1986. Printedin the U.S.A.
0031-9384/86$3.00 + .00
Intermale Social Aggression: Suppression by Medial Preoptic Area Lesions I D. J. A L B E R T , M. L. W A L S H , B. B. G O R Z A L K A , S. M E N D E L S O N A N D C. Z A L Y S
Psychology Department, University o f British Columbia, Vancouver, B.C. V6T I W 5 R e c e i v e d 24 M a r c h 1986 ALBERT, D. J., M. L. WALSH, B. B. GORZALKA, S. MENDELSON AND C. ZALYS. lntermale social aggression: Suppression by medial preoptic area lesions. PHYSIOL BEHAV 38(2) 16%173, 1986.--The intermale social aggressive behavior of male rats cohabiting with a female rat was quantitatively scored weekly in response to the introduction of an unfamiliar intruding male. Resident male rats whose aggressiveness toward an intruder reached a criterion level were subjected to either sham lesions or bilateral lesions in the region of the medial preoptic area. The lesioned rats continued to exhibit levels of piloerection and lateral attack that were not significantly lower than those of sham-lesioned animals, However, the lesioned animals did emit significantly fewer bites and spent significantly less time in the "on-top" position than did sham-lesioned animals. The lesioned animals also displayed significantly less sexual behavior than the shamlesioned animals but were not different in terms of defensiveness toward the experimenter. It is suggested that bilateral lesions in the region of the medial preoptic area cause a decrease in the intensity of intermale social aggression but do not prevent external stimuli from eliciting the aggression. Aggression
Defensiveness
Intermale aggression
Medialpreoptic area
I N T E R M A L E social aggression is the form of aggression in which adult males of a species fight with one another to establish dominance relationships. In rats, it is readily distinguished from defensiveness or predatory aggression by the characteristic display of piloerection and lateral attack [2, 3, 8]. A recent experiment with rats indicates that intermale social aggression is suppressed by lesions in the medial preoptic area [7]. The suppression appears relatively specific in that mouse killing and defensiveness are unchanged by the lesion. Sexual behavior is also suppressed by the preoptic area lesions but not by damage to the regions which maximally suppress aggressive behavior [7]. An important limitation of this experiment for understanding the neural mechanisms controlling intermale social aggression is that the aggression was observed outside of the animal's normal living area and the male rats studied were relatively inexperienced aggressors. These procedures would be expected to produce a relatively low level of aggression [4] that might be unusually susceptible to disturbance by lesions. Unfortunately, there was no quantitative description of the aggressive behavior prior to, or following the lesion. The best laboratory situation available for fostering intermale social aggression is a mixed-sex living group where one or more males reside with one or more females [18]. Introduction of an unfamiliar male at weekly intervals will usually result in an increasingly intense attack by a resident male (alpha male). Attack by the resident male will include the piloerection and lateral attack characteristic of intermale social aggression.
Sexual behavior
The present experiment reexamined the effect of medial preoptic area lesions on intermale social aggression in rats. Quantitative observations of intermale social aggression were made in mixed-sex living groups prior to and following electrolytic lesions. Defensiveness toward the experimenter and sexual behavior toward a receptive female by alpha males were also observed.
METHOD The subjects were adult male and female hooded rats derived from stock obtained from Charles River Canada. They were initially housed in groups of two males and one female. Thirteen of 18 colonies were eventually reduced to a male/female pair. In some cases this happened as a result of severe fighting between the resident males, causing severe injury. In other cases it was because the two males were equally aggressive and interfered with one another's attacks on intruders. The groups of animals were housed in plywood cages 50x60x20 cm high with grey sides, a hardware cloth top, a Plexiglas front, and San-i-cel on the floor. Illumination was a normal 12/12 hr light/dark cycle.
Preoperative Behavioral Testing Prior to any surgical manipulations, each group of animals had a male intruder (slightly smaller than the resident males) introduced into their cage for a 15 min period one day per
~Supported by grants from the Natural Sciences and Engineering Research Council of Canada.
169
ALBERT El' AI,.
170 ~ASELINE
BA.~ELINE
POSTOPERATIVE
/I
423,, !
42' 36'
/ /
3O-
I
30.
9, i 24
24-
~
[~SHAM LL
18-
POA
12POA
,l~y
°°^
WEEK5
WEEK5
WEEK5 BASELINE
l
\/
BASELINE
PO5TOPERATIVE I00
g0-
72-
8O
}
/
SXAM
38o. 6o.
bO-
\\
%E( ~
k\
-
•~SHAM
54.
lI ki~-m . . . . 36-
la"
;
0
0
i WEEK5
o
WEEKS
;
WEEK5
FIG. I. Mean bites, lateral attack frequency, lateral attack duration, on-top, piloerection, and composite aggression scores both pre- and postoperatively for lesioned (open squares) and sham-lesioned rats (solid squares). Vertical lines represent the standard error of the mean. Asterisks indicate days on which lesioned and sham-lesioned groups differed significantly (p<0.05).
week during the last third of the light cycle. The intruders were injected with Diazepam (0.5 mg/animal) 15 min prior to being introduced into a colony in order to minimize their defensiveness. Two observers recorded the aggressive behavior emitted by each resident male(s). The aggressive behaviors noted were lateral attacks, lateral attack duration, bites, on-top, and piloerection. In a lateral attack, the attacking animal moves into a position in which its body is parallel to that of the target animal and then pushes against the target animal with its hind leg. Bites were recorded only when a bite was actually observed. On-top was recorded when the attacking resident animal assumed a position over the prone victim who was lying on its side or back. For each observation session, a composite aggression score was computed for each resident male using the following formula: composite aggression = (number of lateral attacks) + 0.2 (duration of lateral attacks) + (number of bites) + 0.2 (duration of on-top) + (piloerection). Piloerection was scored as 4 if it endured for the entire session; 2 if it lasted for
only a part of a test session, and 0 if it did not occur. This method of calculating a composite aggression score was derived in a previous experiment in such a way as to give equal weight to each behavioral measure and a slightly lower weight to piloerection [4]. When the composite aggression score for an alpha male was greater than 30 for 3 weeks in succession, the alpha male was subjected to either bilateral lesions of the medial preoptic area or to sham lesions. Animals were chosen randomly for the lesion or sham-lesion groups except that the groups were balanced as much as possible in terms of the preoperative composite aggression scores.
Surgery Electrolytic lesions of the medial preoptic area were made with the aid of a Kopf stereotaxic instrument, sodium pentobarbital anesthesia, a stainless steel electrode, and standard surgical procedures. The electrode coordinates were: 2.1
AGGRESSION
171
FIG. 2. Schematic representations of large (left) and small (right) lesions in the region of the medial preoptic area. Both caused large declines in aggression. Only the large lesion caused a complete suppression of sexual behavior.
mm anterior to bregma, 0.7 mm lateral to the midline, and 8.3 mm ventral to the cortical surface (2.0 mA, 10 sec, mouthbar 5.0 mm dorsal to the interaural line). Sham-lesioned animals were subjected to the same surgical procedure except that the electrode was not lowered into the brain. Following surgery both the alpha male and its subordinate cagemate were placed in individual cages for 2 days and then returned to the colony group cage.
Postoperative Behavioral Testing Postoperative aggression tests identical to those done preoperatively began 7 days following surgery. Testing continued every 7 days for 4 weeks. Following the completion of all four tests of aggressive behavior, each resident male rat was tested for sexual behavior toward an ovariectomized female displaying behavioral estrus. Estrus was induced by subcutaneous injection of 10 /zg estradiol benzoate (Steraloids) 48 hr prior to testing and 500 ~.g progesterone (Steraioids) 4 hr prior to testing. The sexual behavior of each male was observed in a glass cylinder (29 cm dia. ; 45 cm high; San-i-cel on floor) beginning 4 to 6 hr after the onset of the dark cycle. The lesioned and sham-lesioned male rats were scored for mounts with pelvic thrusting, intromissions, and ejaculations. The tests of sexual behavior lasted 30 min if no sexual behavior occurred or for a maximum of an additional 30 rain after the last mount was observed in the initial 30-min period [20]. Tests of sexual behavior were conducted for each male on three occasions, 7 days apart. Defensiveness toward the experimenter was assessed following the tests of sexual behavior. Each animal was tested individually in a large gray box (60×60x60 cm high) with San-i-cel on the floor. The animal was presented with a series of stimuli (pencil in front of the nose, tap on the back,
gloved hand in front of the nose, poke in the side with a blunt rod, grasping by the tail, and grasping around the abdomen) and its response to each stimulus scored on a scale of 0 to 3, 0 being little or no response and 3 being a highly aggressive response. F o r details of this procedure see Albert and Richmond [1] or Albert, Walsh and Longley [5].
Histology Following behavioral testing, the lesioned animals were killed. Their brains were removed and placed in formolsaline. They were later sectioned in a cryostat and then stained with cresyl violet.
RESULTS Pre- and postoperative data for each measure of social aggression toward an intruder are shown in Fig. 1. Preoperatively, there were no significant differences between the lesion (N= 10) and sham-lesion (N=8) groups in terms of lateral attacks, bites, on-top, piloerection, or composite aggression score (all t ' s < l . 2 , all p's>0.20). Following medial preoptic area lesions, the amount of aggressive behavior by the lesioned alpha male rats dropped below that of the sham-lesioned animals in some but not all dimensions of aggressive behavior (Fig. 1). The decrease in aggression is visible in the lower composite aggression scores of the rats with preoptic area lesions on each of the test days, F(1,112)= 10.5, p <0.001; all t ' s > 2 . 2 , all p ' s < 0 . 0 5 . In terms of the individual indices of aggressiveness, the number of bites on the intruder by lesioned alpha males was significantly lower than that of sham-lesioned animals on three of the four postoperative test days, F(1,112)=6.2, p<0.01; all t ' s > 2 . 0 all p ' s < 0 . 0 5 , and the amount of on-top
172
AI,BERT U f AL.
time was significantly lower on the first, third, and fourth test days, F(1,112)=8.1, p<0.01; all t's>2.4, all p's<0.05. Lateral attack frequency and duration scores also declined postoperatively in alpha male rats with lesions but analysis of variance indicated that these changes were not statistically reliable, F(I,112)=0.8, p>0.20; F(I,112)-2.75, p=0.10, respectively. Individual paired comparisons confirmed the results of the overall analysis (Fig. 1). The frequency of piloerection also did not differ between groups postoperatively on any of the test days (all)(2's<2.8, all p's>0.05). Sexual and Defensive Behavior The proportion of lesioned animals that mounted, intromitted, and ejaculated on each of the three test trials was 2/9, 2/9, and 0/9, respectively. Each of these proportions is significantly smaller than the corresponding proportions for sham-lesioned animals (6/6, 6/6, 4/6) (Fisher's Exact Probability Test, all p's<0.05). One lesioned and two shamlesioned animals were not tested for sexual behavior. Defensiveness toward the experimenter was not altered by lesions in the medial preoptic area, The mean defensiveness score of the lesioned animals was 3.9 (SEM_+ 1.0) while that of the sham-lesioned control group was 3.1 (_+1.1) (t=0.5, p>0.20). Interaction Between Behavioral Measures Of the three animals showing a complete absence of sexual behavior, all displayed postoperative declines in aggressive behavior of 50% or more. However, of the 5 animals displaying intromission on at least one trial, 4 also exhibited declines in aggressive behavior of 50% or more. Correlation coefficients for the relationship between postoperative composite aggression scores and each of the measures of sexual behavior (mounts, intromissions, and ejaculations) were all small and nonsignificant. Defensiveness toward the experimenter was in the normal range for lesioned animals and, accordingly, cannot be related to either the change in sexual or intermale social aggressive behavior. Histology The lesions destroyed substantial portions of the preoptic area ventral to the crossover of the anterior commissure. They began slightly anterior to or at the level of the crossover of the commissure and ended at the level of the optic chiasm. The suppression of intermale social aggression occurred in animals with the smaller as well as the larger lesions (Fig. 2). In half the animals, lesions began slightly anterior to the crossover of the anterior commissure while in the other half, lesions began under the crossover. Both lesions suppressed intermale social aggression. Sexual behavior was highly sensitive to damage in the region of the medial preoptic area ventral to the anterior commissure. As with aggressive behavior, the present experiments did not associate the decline in sexual behavior with damage to any specific part of this region. DISCUSSION
Lesions in the medial preoptic area caused a suppression
of intermale social aggression. This observation supports those made previously by Bermond [7]. As Bermond reported, the suppression occurred without a concomitant increase or decrease in defensive aggression toward the experimenter. In the present findings there is no clear relationship between the specific parts of the medial preoptic area damaged and the disturbance of intermale social aggression. This contrasts with Bermond's evidence that damage in the anterior region of the medial preoptic area interfered with this behavior significantly more than posterior damage [7]. Unfortunately, we cannot evaluate the discrepancy between the two sets of results because Bermond does not report quantitative data. The decline in aggressive behavior caused by the lesions in the medial preoptic region was not the result of a failure of the intruder to elicit aggression or of an inability of the alpha male to perform the behaviors associated with intermale social aggression. The lesions produced a decline in the intensity rather than an elimination ofintermale social aggression. Following the lesion, most alpha males continued to display piloerection and lateral attack in response to the intruder. However, when the lateral attacks occurred, they were shorter in duration and of lower intensity so that they much less often culminated in biting or on-top behavior. Accordingly, it was the bite and on-top behaviors that were most visibly attenuated by the lesions. The suppression of male sexual behavior by lesions in the medial preoptic area is consistent with previous findings [7, 9, 13, 17l. There are, in addition, observations that testosterone implants into this region induce male sexual behavior in castrated male rats [10, 14, 16]. Bermond [7] has argued that the neural circuitry modulating intermale social aggression is more anteriorly located than the region modulating sexual behavior. There was not enough anterior/posterior variability in the lesions of the present experiment to confirm this anatomical dissociation. However, our results are in agreement with the findings of Bermond to the extent that some lesions suppressed one behavior more than the other, so that the postoperative levels of sexual and aggressive behaviors were not significantly correlated (see, for example, Fig. 2). Testicular testosterone is necessary for intermale social aggression [4, 12, 15]. For testosterone to influence behavior, testosterone or a related substance must enter the central nervous system and influence neural activity. Because studies have shown that there are androgen binding sites in the medial preoptic area [19], it is reasonable to suppose that this may be an area were androgen acts to influence intermale social aggression. Surprisingly, no specific evidence associates the androgen binding cells in the medial preoptic area with the regulation of aggressive behavior. A study using implants of testosterone pellets into the medial preoptic area of castrated rats found only a small increase in biting associated with the implants [6]. Since androgen binding sites are found not only in the medial preoptic region but also in the medial hypothalamus and lateral septum among other areas [19,21], it is obviously premature to strongly infer that the androgen receptors in the medial preoptic region are the route by which testicular testosterone influences intermale social aggression. Investigations of male sexual behavior have also been unable to establish that androgen receptors in the medial preoptic area are the exclusive route by which testosterone influences male sexual behavior: testosterone implants in the anterior and posterior hypothalamus as well as the medial
AGGRESSION
173
preoptic area h a v e each been able to elicit male sexual beh a v i o r in castrated rats [11]. To summarize, lesions of the medial preoptic area suppress intermale social aggression. This suppression takes the form o f a d e c r e a s e in the intensity of the aggression rather than preventing its elicitation. While attention has been fo-
cused on the medial preoptic area because o f the importance of testosterone in intermale social aggression and the abundance of testosterone binding sites on the cells in this region, direct e v i d e n c e for the importance o f such cells on the modulation of intermale social aggression remains to be obtained.
REFERENCES
1. Albert, D. J. and S. E. Richmond. Septal hyperreactivity: A comparison of lesions within and adjacent to the septum, Physiol Behav 15: 33%347, 1975. 2. Albert, D. J. and M. L. Walsh. The inhibitory modulation of agonistic behavior in the rat brain: A review. Neurosci Biobehav Rev 6: 125-143, 1982. 3. Albert, D. J. and M. L. Walsh. Neural systems and the inhibitory modulation of agonistic behavior: Comparison of mammalian species. Neurosci Biobehav Rev 8: 5-24, 1984. 4. Albert, D. J., M. L. Walsh, B. B. Gorzalka, Y. Siemens and H. Louie. Testosterone removal in rats results in a decrease in social aggression and a loss of social dominance. Physiol Behav 36: 401-407, 1986. 5. Albert, D. J., M. L. Walsh and W. Longley. Group rearing abolishes hyperdefensiveness induced in weanling rats by lateral septal or medial accumbens lesions but not by medial hypothalamic lesions. Behav Neural Biol 44: 101-109, 1985. 6. Bean, N. J. and R. Conner. Central hormonal replacement and home-cage dominance in castrated rats. Horm Behav l l : 100109, 1978. 7. Bermond, B. Effects of medial preoptic hypothalamus anterior lesions on three kinds of behavior in the rat: Intermale aggressive, male-sexual, and mouse-killing behavior. Aggress Behav 8: 335-354, 1982. 8. Blanchard, R. J. and D. C. Blanchard. Aggressive behavior in the rat. Behav Biol 21: 197-224, 1977. 9. Chen, J. J. and D. K. Bliss. Effects of sequential preoptic and mammillary lesions on male rat sexual behavior. J Comp Physiol Psychol 87: 841-847, 1974. 10. Davidson, J. M. Activation of the male rat's sexual behavior by intracerebral implantation of androgen. Endocrinology 79: 783-794, 1966. 11. Davidson, J. M. Hormones and sexual behavior in the male. In: Neuroendoerinology, edited by D. T. Krieger and J. C. Hughes. Sunderland, MA: Sinauer Associates, Inc., 1980, pp. 232-238.
12. DeBold, J. F. and K. A. Miczek. Aggression persists after ovariectomy in female rats. Horm Behav 18: 177-190, 1984. 13. Heimer, L. and K. Larsson. Impairment of mating behavior in male rats following lesions in the preoptic-anterior hypothalamic continuum. Brain Res 3: 248-263, 1966/1967. 14. Kierniesky, N. C. and A. A. Gerall. Effects of testosterone propionate implants in the brain on the sexual behavior and peripheral tissue of the male rat. Physiol Behav 11: 633-640, 1973. 15. Koolhaas, J. M., T. Schuurman and P. R. Wiepkema. The organization of intraspecific agonistic behaviour in the rat. Prog Neurobiol 15: 247-268, 1980. ~. 16. Lisk, R. D. Neural localization for androgen activation of copulatory behavior in the male rat. Endocrinology 80: 754-761, 1%7. 17. Lisk, R. D. Copulatory activity of the male rat following placement of preoptic-anterior hypothalamic lesions. Exp Brain Res 5: 306-313, 1968. 18. Lore, R., M. Nikoletseas and L. Takahashi. Colony aggression in laboratory rats: A review and some recommendations. Aggress Behav 10: 59-71, 1984. 19. McEwen, B. S. Gonadal steroid receptors in neuroendocrine tissues. In: Receptors and Hormone Action, Vol H, edited by B. W. O'Malley and L. Birnbaumer. New York: Academic Press, 1978, pp. 353-400. 20. Mendelson, S. D. and B. B. Gorzalka. Serotonin antagonist pirenperone inhibits sexual behavior in the male rat: Attenuation by quipazine. Pharmacol Biochem Behav 22: 565-571, 1985. 21. Sar, M. and W. E. Stumpf. Autoradiographic localization of radioactivity in the rat brain after injection of 1,2-Htestosterone. Endocrinology 92: 251-256, 1973.
0031-9384/86$3.00 + .00
Intermale Social Aggression: Suppression by Medial Preoptic Area Lesions I D. J. A L B E R T , M. L. W A L S H , B. B. G O R Z A L K A , S. M E N D E L S O N A N D C. Z A L Y S
Psychology Department, University o f British Columbia, Vancouver, B.C. V6T I W 5 R e c e i v e d 24 M a r c h 1986 ALBERT, D. J., M. L. WALSH, B. B. GORZALKA, S. MENDELSON AND C. ZALYS. lntermale social aggression: Suppression by medial preoptic area lesions. PHYSIOL BEHAV 38(2) 16%173, 1986.--The intermale social aggressive behavior of male rats cohabiting with a female rat was quantitatively scored weekly in response to the introduction of an unfamiliar intruding male. Resident male rats whose aggressiveness toward an intruder reached a criterion level were subjected to either sham lesions or bilateral lesions in the region of the medial preoptic area. The lesioned rats continued to exhibit levels of piloerection and lateral attack that were not significantly lower than those of sham-lesioned animals, However, the lesioned animals did emit significantly fewer bites and spent significantly less time in the "on-top" position than did sham-lesioned animals. The lesioned animals also displayed significantly less sexual behavior than the shamlesioned animals but were not different in terms of defensiveness toward the experimenter. It is suggested that bilateral lesions in the region of the medial preoptic area cause a decrease in the intensity of intermale social aggression but do not prevent external stimuli from eliciting the aggression. Aggression
Defensiveness
Intermale aggression
Medialpreoptic area
I N T E R M A L E social aggression is the form of aggression in which adult males of a species fight with one another to establish dominance relationships. In rats, it is readily distinguished from defensiveness or predatory aggression by the characteristic display of piloerection and lateral attack [2, 3, 8]. A recent experiment with rats indicates that intermale social aggression is suppressed by lesions in the medial preoptic area [7]. The suppression appears relatively specific in that mouse killing and defensiveness are unchanged by the lesion. Sexual behavior is also suppressed by the preoptic area lesions but not by damage to the regions which maximally suppress aggressive behavior [7]. An important limitation of this experiment for understanding the neural mechanisms controlling intermale social aggression is that the aggression was observed outside of the animal's normal living area and the male rats studied were relatively inexperienced aggressors. These procedures would be expected to produce a relatively low level of aggression [4] that might be unusually susceptible to disturbance by lesions. Unfortunately, there was no quantitative description of the aggressive behavior prior to, or following the lesion. The best laboratory situation available for fostering intermale social aggression is a mixed-sex living group where one or more males reside with one or more females [18]. Introduction of an unfamiliar male at weekly intervals will usually result in an increasingly intense attack by a resident male (alpha male). Attack by the resident male will include the piloerection and lateral attack characteristic of intermale social aggression.
Sexual behavior
The present experiment reexamined the effect of medial preoptic area lesions on intermale social aggression in rats. Quantitative observations of intermale social aggression were made in mixed-sex living groups prior to and following electrolytic lesions. Defensiveness toward the experimenter and sexual behavior toward a receptive female by alpha males were also observed.
METHOD The subjects were adult male and female hooded rats derived from stock obtained from Charles River Canada. They were initially housed in groups of two males and one female. Thirteen of 18 colonies were eventually reduced to a male/female pair. In some cases this happened as a result of severe fighting between the resident males, causing severe injury. In other cases it was because the two males were equally aggressive and interfered with one another's attacks on intruders. The groups of animals were housed in plywood cages 50x60x20 cm high with grey sides, a hardware cloth top, a Plexiglas front, and San-i-cel on the floor. Illumination was a normal 12/12 hr light/dark cycle.
Preoperative Behavioral Testing Prior to any surgical manipulations, each group of animals had a male intruder (slightly smaller than the resident males) introduced into their cage for a 15 min period one day per
~Supported by grants from the Natural Sciences and Engineering Research Council of Canada.
169
ALBERT El' AI,.
170 ~ASELINE
BA.~ELINE
POSTOPERATIVE
/I
423,, !
42' 36'
/ /
3O-
I
30.
9, i 24
24-
~
[~SHAM LL
18-
POA
12POA
,l~y
°°^
WEEK5
WEEK5
WEEK5 BASELINE
l
\/
BASELINE
PO5TOPERATIVE I00
g0-
72-
8O
}
/
SXAM
38o. 6o.
bO-
\\
%E( ~
k\
-
•~SHAM
54.
lI ki~-m . . . . 36-
la"
;
0
0
i WEEK5
o
WEEKS
;
WEEK5
FIG. I. Mean bites, lateral attack frequency, lateral attack duration, on-top, piloerection, and composite aggression scores both pre- and postoperatively for lesioned (open squares) and sham-lesioned rats (solid squares). Vertical lines represent the standard error of the mean. Asterisks indicate days on which lesioned and sham-lesioned groups differed significantly (p<0.05).
week during the last third of the light cycle. The intruders were injected with Diazepam (0.5 mg/animal) 15 min prior to being introduced into a colony in order to minimize their defensiveness. Two observers recorded the aggressive behavior emitted by each resident male(s). The aggressive behaviors noted were lateral attacks, lateral attack duration, bites, on-top, and piloerection. In a lateral attack, the attacking animal moves into a position in which its body is parallel to that of the target animal and then pushes against the target animal with its hind leg. Bites were recorded only when a bite was actually observed. On-top was recorded when the attacking resident animal assumed a position over the prone victim who was lying on its side or back. For each observation session, a composite aggression score was computed for each resident male using the following formula: composite aggression = (number of lateral attacks) + 0.2 (duration of lateral attacks) + (number of bites) + 0.2 (duration of on-top) + (piloerection). Piloerection was scored as 4 if it endured for the entire session; 2 if it lasted for
only a part of a test session, and 0 if it did not occur. This method of calculating a composite aggression score was derived in a previous experiment in such a way as to give equal weight to each behavioral measure and a slightly lower weight to piloerection [4]. When the composite aggression score for an alpha male was greater than 30 for 3 weeks in succession, the alpha male was subjected to either bilateral lesions of the medial preoptic area or to sham lesions. Animals were chosen randomly for the lesion or sham-lesion groups except that the groups were balanced as much as possible in terms of the preoperative composite aggression scores.
Surgery Electrolytic lesions of the medial preoptic area were made with the aid of a Kopf stereotaxic instrument, sodium pentobarbital anesthesia, a stainless steel electrode, and standard surgical procedures. The electrode coordinates were: 2.1
AGGRESSION
171
FIG. 2. Schematic representations of large (left) and small (right) lesions in the region of the medial preoptic area. Both caused large declines in aggression. Only the large lesion caused a complete suppression of sexual behavior.
mm anterior to bregma, 0.7 mm lateral to the midline, and 8.3 mm ventral to the cortical surface (2.0 mA, 10 sec, mouthbar 5.0 mm dorsal to the interaural line). Sham-lesioned animals were subjected to the same surgical procedure except that the electrode was not lowered into the brain. Following surgery both the alpha male and its subordinate cagemate were placed in individual cages for 2 days and then returned to the colony group cage.
Postoperative Behavioral Testing Postoperative aggression tests identical to those done preoperatively began 7 days following surgery. Testing continued every 7 days for 4 weeks. Following the completion of all four tests of aggressive behavior, each resident male rat was tested for sexual behavior toward an ovariectomized female displaying behavioral estrus. Estrus was induced by subcutaneous injection of 10 /zg estradiol benzoate (Steraloids) 48 hr prior to testing and 500 ~.g progesterone (Steraioids) 4 hr prior to testing. The sexual behavior of each male was observed in a glass cylinder (29 cm dia. ; 45 cm high; San-i-cel on floor) beginning 4 to 6 hr after the onset of the dark cycle. The lesioned and sham-lesioned male rats were scored for mounts with pelvic thrusting, intromissions, and ejaculations. The tests of sexual behavior lasted 30 min if no sexual behavior occurred or for a maximum of an additional 30 rain after the last mount was observed in the initial 30-min period [20]. Tests of sexual behavior were conducted for each male on three occasions, 7 days apart. Defensiveness toward the experimenter was assessed following the tests of sexual behavior. Each animal was tested individually in a large gray box (60×60x60 cm high) with San-i-cel on the floor. The animal was presented with a series of stimuli (pencil in front of the nose, tap on the back,
gloved hand in front of the nose, poke in the side with a blunt rod, grasping by the tail, and grasping around the abdomen) and its response to each stimulus scored on a scale of 0 to 3, 0 being little or no response and 3 being a highly aggressive response. F o r details of this procedure see Albert and Richmond [1] or Albert, Walsh and Longley [5].
Histology Following behavioral testing, the lesioned animals were killed. Their brains were removed and placed in formolsaline. They were later sectioned in a cryostat and then stained with cresyl violet.
RESULTS Pre- and postoperative data for each measure of social aggression toward an intruder are shown in Fig. 1. Preoperatively, there were no significant differences between the lesion (N= 10) and sham-lesion (N=8) groups in terms of lateral attacks, bites, on-top, piloerection, or composite aggression score (all t ' s < l . 2 , all p's>0.20). Following medial preoptic area lesions, the amount of aggressive behavior by the lesioned alpha male rats dropped below that of the sham-lesioned animals in some but not all dimensions of aggressive behavior (Fig. 1). The decrease in aggression is visible in the lower composite aggression scores of the rats with preoptic area lesions on each of the test days, F(1,112)= 10.5, p <0.001; all t ' s > 2 . 2 , all p ' s < 0 . 0 5 . In terms of the individual indices of aggressiveness, the number of bites on the intruder by lesioned alpha males was significantly lower than that of sham-lesioned animals on three of the four postoperative test days, F(1,112)=6.2, p<0.01; all t ' s > 2 . 0 all p ' s < 0 . 0 5 , and the amount of on-top
172
AI,BERT U f AL.
time was significantly lower on the first, third, and fourth test days, F(1,112)=8.1, p<0.01; all t's>2.4, all p's<0.05. Lateral attack frequency and duration scores also declined postoperatively in alpha male rats with lesions but analysis of variance indicated that these changes were not statistically reliable, F(I,112)=0.8, p>0.20; F(I,112)-2.75, p=0.10, respectively. Individual paired comparisons confirmed the results of the overall analysis (Fig. 1). The frequency of piloerection also did not differ between groups postoperatively on any of the test days (all)(2's<2.8, all p's>0.05). Sexual and Defensive Behavior The proportion of lesioned animals that mounted, intromitted, and ejaculated on each of the three test trials was 2/9, 2/9, and 0/9, respectively. Each of these proportions is significantly smaller than the corresponding proportions for sham-lesioned animals (6/6, 6/6, 4/6) (Fisher's Exact Probability Test, all p's<0.05). One lesioned and two shamlesioned animals were not tested for sexual behavior. Defensiveness toward the experimenter was not altered by lesions in the medial preoptic area, The mean defensiveness score of the lesioned animals was 3.9 (SEM_+ 1.0) while that of the sham-lesioned control group was 3.1 (_+1.1) (t=0.5, p>0.20). Interaction Between Behavioral Measures Of the three animals showing a complete absence of sexual behavior, all displayed postoperative declines in aggressive behavior of 50% or more. However, of the 5 animals displaying intromission on at least one trial, 4 also exhibited declines in aggressive behavior of 50% or more. Correlation coefficients for the relationship between postoperative composite aggression scores and each of the measures of sexual behavior (mounts, intromissions, and ejaculations) were all small and nonsignificant. Defensiveness toward the experimenter was in the normal range for lesioned animals and, accordingly, cannot be related to either the change in sexual or intermale social aggressive behavior. Histology The lesions destroyed substantial portions of the preoptic area ventral to the crossover of the anterior commissure. They began slightly anterior to or at the level of the crossover of the commissure and ended at the level of the optic chiasm. The suppression of intermale social aggression occurred in animals with the smaller as well as the larger lesions (Fig. 2). In half the animals, lesions began slightly anterior to the crossover of the anterior commissure while in the other half, lesions began under the crossover. Both lesions suppressed intermale social aggression. Sexual behavior was highly sensitive to damage in the region of the medial preoptic area ventral to the anterior commissure. As with aggressive behavior, the present experiments did not associate the decline in sexual behavior with damage to any specific part of this region. DISCUSSION
Lesions in the medial preoptic area caused a suppression
of intermale social aggression. This observation supports those made previously by Bermond [7]. As Bermond reported, the suppression occurred without a concomitant increase or decrease in defensive aggression toward the experimenter. In the present findings there is no clear relationship between the specific parts of the medial preoptic area damaged and the disturbance of intermale social aggression. This contrasts with Bermond's evidence that damage in the anterior region of the medial preoptic area interfered with this behavior significantly more than posterior damage [7]. Unfortunately, we cannot evaluate the discrepancy between the two sets of results because Bermond does not report quantitative data. The decline in aggressive behavior caused by the lesions in the medial preoptic region was not the result of a failure of the intruder to elicit aggression or of an inability of the alpha male to perform the behaviors associated with intermale social aggression. The lesions produced a decline in the intensity rather than an elimination ofintermale social aggression. Following the lesion, most alpha males continued to display piloerection and lateral attack in response to the intruder. However, when the lateral attacks occurred, they were shorter in duration and of lower intensity so that they much less often culminated in biting or on-top behavior. Accordingly, it was the bite and on-top behaviors that were most visibly attenuated by the lesions. The suppression of male sexual behavior by lesions in the medial preoptic area is consistent with previous findings [7, 9, 13, 17l. There are, in addition, observations that testosterone implants into this region induce male sexual behavior in castrated male rats [10, 14, 16]. Bermond [7] has argued that the neural circuitry modulating intermale social aggression is more anteriorly located than the region modulating sexual behavior. There was not enough anterior/posterior variability in the lesions of the present experiment to confirm this anatomical dissociation. However, our results are in agreement with the findings of Bermond to the extent that some lesions suppressed one behavior more than the other, so that the postoperative levels of sexual and aggressive behaviors were not significantly correlated (see, for example, Fig. 2). Testicular testosterone is necessary for intermale social aggression [4, 12, 15]. For testosterone to influence behavior, testosterone or a related substance must enter the central nervous system and influence neural activity. Because studies have shown that there are androgen binding sites in the medial preoptic area [19], it is reasonable to suppose that this may be an area were androgen acts to influence intermale social aggression. Surprisingly, no specific evidence associates the androgen binding cells in the medial preoptic area with the regulation of aggressive behavior. A study using implants of testosterone pellets into the medial preoptic area of castrated rats found only a small increase in biting associated with the implants [6]. Since androgen binding sites are found not only in the medial preoptic region but also in the medial hypothalamus and lateral septum among other areas [19,21], it is obviously premature to strongly infer that the androgen receptors in the medial preoptic region are the route by which testicular testosterone influences intermale social aggression. Investigations of male sexual behavior have also been unable to establish that androgen receptors in the medial preoptic area are the exclusive route by which testosterone influences male sexual behavior: testosterone implants in the anterior and posterior hypothalamus as well as the medial
AGGRESSION
173
preoptic area h a v e each been able to elicit male sexual beh a v i o r in castrated rats [11]. To summarize, lesions of the medial preoptic area suppress intermale social aggression. This suppression takes the form o f a d e c r e a s e in the intensity of the aggression rather than preventing its elicitation. While attention has been fo-
cused on the medial preoptic area because o f the importance of testosterone in intermale social aggression and the abundance of testosterone binding sites on the cells in this region, direct e v i d e n c e for the importance o f such cells on the modulation of intermale social aggression remains to be obtained.
REFERENCES
1. Albert, D. J. and S. E. Richmond. Septal hyperreactivity: A comparison of lesions within and adjacent to the septum, Physiol Behav 15: 33%347, 1975. 2. Albert, D. J. and M. L. Walsh. The inhibitory modulation of agonistic behavior in the rat brain: A review. Neurosci Biobehav Rev 6: 125-143, 1982. 3. Albert, D. J. and M. L. Walsh. Neural systems and the inhibitory modulation of agonistic behavior: Comparison of mammalian species. Neurosci Biobehav Rev 8: 5-24, 1984. 4. Albert, D. J., M. L. Walsh, B. B. Gorzalka, Y. Siemens and H. Louie. Testosterone removal in rats results in a decrease in social aggression and a loss of social dominance. Physiol Behav 36: 401-407, 1986. 5. Albert, D. J., M. L. Walsh and W. Longley. Group rearing abolishes hyperdefensiveness induced in weanling rats by lateral septal or medial accumbens lesions but not by medial hypothalamic lesions. Behav Neural Biol 44: 101-109, 1985. 6. Bean, N. J. and R. Conner. Central hormonal replacement and home-cage dominance in castrated rats. Horm Behav l l : 100109, 1978. 7. Bermond, B. Effects of medial preoptic hypothalamus anterior lesions on three kinds of behavior in the rat: Intermale aggressive, male-sexual, and mouse-killing behavior. Aggress Behav 8: 335-354, 1982. 8. Blanchard, R. J. and D. C. Blanchard. Aggressive behavior in the rat. Behav Biol 21: 197-224, 1977. 9. Chen, J. J. and D. K. Bliss. Effects of sequential preoptic and mammillary lesions on male rat sexual behavior. J Comp Physiol Psychol 87: 841-847, 1974. 10. Davidson, J. M. Activation of the male rat's sexual behavior by intracerebral implantation of androgen. Endocrinology 79: 783-794, 1966. 11. Davidson, J. M. Hormones and sexual behavior in the male. In: Neuroendoerinology, edited by D. T. Krieger and J. C. Hughes. Sunderland, MA: Sinauer Associates, Inc., 1980, pp. 232-238.
12. DeBold, J. F. and K. A. Miczek. Aggression persists after ovariectomy in female rats. Horm Behav 18: 177-190, 1984. 13. Heimer, L. and K. Larsson. Impairment of mating behavior in male rats following lesions in the preoptic-anterior hypothalamic continuum. Brain Res 3: 248-263, 1966/1967. 14. Kierniesky, N. C. and A. A. Gerall. Effects of testosterone propionate implants in the brain on the sexual behavior and peripheral tissue of the male rat. Physiol Behav 11: 633-640, 1973. 15. Koolhaas, J. M., T. Schuurman and P. R. Wiepkema. The organization of intraspecific agonistic behaviour in the rat. Prog Neurobiol 15: 247-268, 1980. ~. 16. Lisk, R. D. Neural localization for androgen activation of copulatory behavior in the male rat. Endocrinology 80: 754-761, 1%7. 17. Lisk, R. D. Copulatory activity of the male rat following placement of preoptic-anterior hypothalamic lesions. Exp Brain Res 5: 306-313, 1968. 18. Lore, R., M. Nikoletseas and L. Takahashi. Colony aggression in laboratory rats: A review and some recommendations. Aggress Behav 10: 59-71, 1984. 19. McEwen, B. S. Gonadal steroid receptors in neuroendocrine tissues. In: Receptors and Hormone Action, Vol H, edited by B. W. O'Malley and L. Birnbaumer. New York: Academic Press, 1978, pp. 353-400. 20. Mendelson, S. D. and B. B. Gorzalka. Serotonin antagonist pirenperone inhibits sexual behavior in the male rat: Attenuation by quipazine. Pharmacol Biochem Behav 22: 565-571, 1985. 21. Sar, M. and W. E. Stumpf. Autoradiographic localization of radioactivity in the rat brain after injection of 1,2-Htestosterone. Endocrinology 92: 251-256, 1973.