Steroid-specific regulation of agonistic responding in the anterior hypothalamus of male hamsters

Physiology& Behavior,Vol. 50, pp. 793-799. ©Pergamon Press plc, 1991. Printed in the U.S.A.

0031-9384/91 $3.00 + .00

Steroid-Specific Regulation of Agonistic Responding in the Anterior Hypothalamus of Male Hamsters D I A N E M. H A Y D E N - H I X S O N * A N D C R A I G F. F E R R I S t

*Department of Psychiatry, Box 3859 DUMC, Duke University Medical Center, Durham, NC 27710 "pDepartment of Physiology, University of Massachusetts Medical Center, Worcester, MA 01655 R e c e i v e d 4 January 1991 HAYDEN-HIXSON, D. M. AND C. F. FERRIS. Steroid-specific regulation of agonistic responding in the anterior hypothalamus of male hamsters. PHYSIOL BEHAV 50(4) 793-799, 1991.--The agonistic behaviors of adult male golden hamsters (N = 108 dyads) were examined 5 min after stereotaxic microinjection of adrenal and gonadal steroids into the anterior hypothalamus. Flank marks, attacks, bites, and retreats were scored over a 15 min test period during which steroid-injected animals were paired in a neutral arena with vehicle-injected conspecifics. Animals microinjected with either 10 - 6 M cortisol or 10 - 6 M 13-estradiol displayed significantly (p<0.05) higher levels of flank marking than other steroid-treated animals. Animals microinjected with 10 -6 M cortisol displayed significantly higher levels of aggression than their opponents. In contrast, the behavior of the vehicle-injected animals paired with 10 - 6 M cortisol-treated opponents was characterized by submissive responding. This profile of the 10 - 6 M cortisol treatment, i.e., promoting aggression in a steroid-treated animal while eliciting submission from its vehicle-treated opponent, was not observed in pairs in which steroid-injected animals were treated with equimolar concentrations of testosterone, dihydrotestosterone, progesterone, 13-estradiol, or deoxycorticosterone. These findings suggest steroids exert immediate effects on agonistic responding in the anterior hypothalamus of male hamsters. The immediate action(s) of cortisol appear to include facilitating aggression and flank marking, while the immediate action(s) of 13-estradiol appears to be confined to the communicative aspect of agonistic responding in this species. Cortisol Deoxycorticosterone Submission Flank marking

Dihydrotestosterone

Estrogen

CONSIDERABLE evidence suggests gonadal (4, 12, 26, 30, 44) and adrenal (3, 15, 23, 25) steroids influence agonistic responding, and the consequences of competitive interactions can, in turn, influence the basal release of gonadal and adrenal steroids (5, 17, 25, 36). In male golden hamsters, correlational data indicate that bidirectional interactions between hormones and behavior play an important role in regulating agonistic responding. For example, gonadal (4, 10, 12, 26, 46) or adrenal (3,24) glandectomized male hamsters are typically less aggressive than intact or steroid-replaced glandectomized males under testing conditions that normally maximize spontaneous aggression in this species (23,24). Likewise, male hamsters exposed to social conflict tests reportedly establish dominant/subordinate relationships in which the subordinate animal exhibits significant increases in the circulating levels of cortisol and corticosterone only after acute exposures (16,17), but significant decreases in the circulating levels of testosterone after repeated exposures (17). It is widely accepted that steroid hormones modulate behavior through indirect actions on neurotransmission in the central nervous system (CNS) (2,28). Furthermore, it is assumed that the retention and biological activity of hormones in specific areas of the brain relate, at least in part, to the densities of steroid receptors in those areas (27, 31, 32, 34). According to the classic model of steroid hormone action, steroids released into the

Progesterone

Testosterone

Aggression

peripheral circulation reach neurons in the brain, where they bind to intracellular receptors that, once activated, interact with the genome to regulate transcription and protein synthesis. Hence, "classic" regulatory actions of steroids on neurotransmission are slow in onset and long in duration. However, it is also recognized that steroids have direct effects on CNS excitability that are too rapid, i.e., of short latency and duration, to be consistent with the temporal delays normally associated with genome-mediated mechanisms of action (2, 14, 28, 31, 38). Rapid effects of steroid hormones on neuronal excitability are reported in vivo and in vitro following exogenous steroid administration [see (14) and (38) for reviews]. While it is generally assumed that endogenously released steroid also exerts direct effects on membrane excitability, the functional significance of these potential rapid actions of steroids to ongoing behavior is not established (2,38). Dramatic, and relatively rapid, increases in adrenal steroid release occur during brief agonistic interactions in a number of species, including hamsters (16,17), mice (5,40), guinea pigs (35), and baboons (36). Therefore, it is conceivable that endogenously released adrenal steroids could influence brain activity in ways that have immediate functional significance to the ongoing agonistic behavior of an animal. To test this hypothesis, the aggressive, submissive and flank marking behaviors of adult male golden hamsters were exam793

71)4

HAYDEN-HIXS()N AND EERRIS

ined immediately fbllowing the direct administration of gonadal and adrenal steroids into the anterior hypothalamus. The anterior hypothalamus was chosen because it is an area where overlapping gonadal (7. 11, 20. 33, 45) and adrenal (15,421 steroidresponsive circuits appear to exist in hamsters. The anterior hypothalamus is also an area implicated in the central regulation of agonistic responding in a number of species, including hamsters (9, 13, 15, 29, 43), rats (1, 21, 22), and cats (41). Briefly, the behavioral effects of an acute microinjection of one of two doses (10 6 M and 1 0 3 Mh of one of six steroids ([3-estradiol, testosterone, dibydrotestosterone, progesterone, deoxycorticosterone, and cortisol) were quantified during a 15 rain paired encounter with an unfamiliar male conspecific. METHOD

Subjects, Housing and Screening Adult male golden hamsters (90-140 g), purchased from Harlan Sprague-Dawley (Indianapolis, IN), were singly housed in Plexiglas cages (24 × 22 × 20), maintained on a reverse 14:10 light:dark cycle (lights off at 0800 h), and provided food and water ad lib for at least two weeks prior to, and throughout all nontesting phases of the experiments. Animals were weighed on the day of surgery and on the day of behavioral testing.

Screening Animals were screened for aggressiveness by placing a larger stimulus male hamster (110-160 g) of predetermined dominant status (an animal that had consistently subordinated weightmatched conspecifics during paired encounters) into their home cage. Animals that exhibited either flight behavior or submissive posturing when approached by the stimulus male were designated "submissive." Animals that either initiated attacks or engaged in roll-biting defensive behaviors after being attacked by the stimulus male were designated "aggressive." Only animals designated "aggressive" by this screening procedure were used in the following study.

Steroid Microinjections Surgery was performed between 1200 and 1700 h under sodium pentobarbital anesthesia (50 mg/kg). Hamsters were stereotaxically implanted with a 26-gauge guide cannula aimed at the left anterior hypothalamic area [8 ° medially angled from perpendicular; 1.7 mm lateral of the midsagittal suture: 4.5 mm ventral from dura; incisor bar level with the interaural line; 1.0-1.2 mm anterior of Bregma]. The cannula was secured in place with a dental cement mold about the cannula and two 7.5 mm stainless steel wound clips fitted into burr holes on the skull. Microinjections were made with a 33-gauge needle attached to a ! ml Hamilton syringe with PE 20 tubing. Injections were given in a volume of 100 nl to unanesthetized hamsters six days after surgery. Micromolar ( 1 0 - 6 M: ~32---4 pg W/V) and millimolar (10 3 M: ~32___4 ng w/v) doses were used because cortisol, the principle circulating glucocorticoid in hamsters (37), has biphasic effects on agonistic responding at these concentrations. After acute microinjection in the anterior hypothalamus, 10 - 6 M cortisol treatments induce, and 10 - 3 M treatments suppress overt aggression (15). (Note: As all solutions were prepared using formula weights, 100 nl injection of 10 - 6 M cortisol contained ~-36 pg, [3-estradiol ~27 pg, etc.)

Behavioral Testing All agonistic encounters paired a vehicle-injected (control) male with a weight-matched steroid-injected (experimental) male.

Steroid-injected animals received either 10 " M ~t i(! ' M ~t,-lutions of testosterone (1713-hydroxy-3-oxo-4 androstene, # T 1500, Sigma), dihydrotestosterone (5c~-androstan-17[3-ol-3-one, #A-8380, Sigma), 13-estradiol ( 1,3,5[ 10]-estratriene-3, 17[3-diol, #E-8875, Sigma), progesterone /4-pregnene-3,20-dione, #P0103, Sigma), 11 -deoxycorticosterone (4-pregnen-21 -ol-3,20-dione, #D-6875, Sigma), or cortisol (hydrocortisone. #H-4001, Sigma) dissolved in 100% ethanol and diluted 1:9 with (1.9~,~ NaC1. Vehicle-injected animals (N = 108~ were given 1~1~ ethanol in 0.9~k NaCI. Each animal received only one injection. Eight control/experimental pairs were tested for 1(1 " M [3-estradiol or progesterone: and for I(t ~ M deoxycorticosterone, dihydrotestosterone, or [3-estradiol. Nine pairs were tested for 10 ~ M cortisol, testosterone, dihydrotestosterone, or deoxycorticosterone. Ten pairs were tested for 10 ~ M testosterone or progesterone. Twelve pairs were tested for 10 " M cortisol. Testing was done under dim red illumination between 0800 and 1200 b. Five minutes postinjection, each steroid-treated animal was paired with an unfamiliar vehicle-treated opponent in a neutral Plexiglas cage (24 × 22 × 20 cm~.

Behavioral Analyses and Scoring Criteria Attacks, bites, retreats, and flank marks were concurrently scored for each animal. The level of agonistic behavior of each animal was quantified as an aggression index, calculated as the sum of the attacks and bites, minus the retreats. The member of each dyad with the higher combined aggression index and flank marking score was operationally defined as dominant, its opponent subordinate. Animals with positive aggression indices at the end of a 15 rain pairing were operationally defined as aggressive, and animals with negative aggression indices, submissive. l~ Attacks. An attack was scored when ever an animal made physical contact with its opponent after stopping an intervening behavior, i.e., grooming, digging, or flank marking, to approach, follow, or chase the other animal. 2) Bites. A bite was scored only when an animal opened its mouth, pushed its head into the other animal's flank, abdomen, etc, and elicited an audible vocalization from its opponent. It should be noted that no attempt was made to distinguish between offensive/initiated and defensive/retaliatory attacks and bites in this scoring procedure. 3) Retreats. A retreat was scored when an animal: a) fled from its opponent, or attempted to escape from the testing cage, b) exhibited a full submissive on-back posture (limbs extended laterally, eyes closed), or c) responded with an upright submissive posture (stationary, head averted, rear limb elevated, ears and tail erect or extended over the back) in response to the approach, attack, or bite of its opponent. 4) Flank marks. Flank marking is a highly stereotyped form of scent marking in male and female hamsters (8, 18, 19, 46). A flank mark was scored whenever an animal rubbed a pigmented sebaceous gland, located in the dorsolateral flank area, against vertical surfaces of the testing cage.

Statistical Analyses The dependent variables in all statistical analyses were either individual aggression index and flank marking scores, or pairwise aggression index and flank marking difference scores. Pairwise difference scores were used to identify differential treatment effects on the polarity of the interactions within dyads. A difference score was calculated for each pair as the steroid-treated animal's score minus the vehicle-treated animal's score. (Note: Using this convention, dyads with highly polarized dominating-

AGGRESSION IN MALE HAMSTERS

795

FIG. 1. Representative coronal section through the anterior hypothalamus at the level of the steroid microinjections. Microinjection site is indicated by an arrow. Abbreviations: AH = anterior hypothalamus; SCN = suprachiasmatic nucleus; OC = optic chiasm.

subordinating relationships have more positive or negative difference scores than dyads containing equally aggressive or nonaggressive animals; and groups with chance distributions of aggressive and submissive steroid-treated animals have mean difference scores of zero.) Results were expressed as mean_+ SEM. Paired t-tests were used to identify significant differences in the behaviors of vehicle- and steroid-treated opponents within groups. One-factor analyses of variance (ANOVAs), followed by Fisher PLSD post hoc tests of multiple comparisons (Statview 512, Brain Power Inc., 1986) were used to identify significant differences between treatment groups.

Histological Analyses At the completion of behavioral testing, animals were sacrificed by decapitation. Their brains were removed from the skull and stored in Perfix (Fisher Scientific) at 4°C for at least 24 h before sectioning. Microinjection sites were histologically verified from 100 Ixm coronal sections, cut on a vibratome (Technical Products Series 1000), mounted on subbed slides, stained with 4% thionin, dehydrated, coverslipped with permount (Fisher Scientific), and analyzed using a light microscope. RESULTS Of the twelve steroid treatments examined, only 10 - 6 M cortisol and 10 -3 deoxycorticosterone (DOC) induced high lev-

els of aggression in the steroid-treated animals after acute microinjection in the anterior hypothalamus (see Fig. 1 for the location of a typical microinjection site). In contrast, 10 - 6 M cortisol and 10 - 6 M [3-estradiol (13-E2) induced flank marking in the steroid-treated animal, while 10 - 6 M testosterone (T) and 10 -3 M cortisol treatments induced flank marking in the vehicle-treated opponent of the steroid-treated animal.

Aggression and Submission Several notable aggression-promoting or aggression-inhibiting trends in the effects of the steroid treatments were observed. More vehicle-treated animals were dominant than steroid-treated animals in the case of 10 - 6 M progesterone (P4:6/8 pairs), 10 - 6 M DOC (6/9 pairs), and 10 -3 M cortisol (9/12 pairs). More steroid-treated animals were aggressive than vehicle-treated animals in the case of 10 6 M cortisol (8/9 pairs) and 10 - 3 M DOC (7/8 pairs). However, significant (p<0.05) differences in aggression between the steroid-treated animals and their vehicletreated opponents were only observed in the case of the 10 - 6 M cortisol treatment, t(8)= - 3 . 6 0 7 , p = 0 . 0 0 6 9 (see Table 1). Significant aggression index F-ratios were observed in the ANOVA comparisons of vehicle-treated males, F ( l l , 1 0 7 ) = 2.133, p = 0.0248, steroid-treated males, F(11,107) = 3.283, p = 0.0008, and the pairwise aggression index difference scores, F(I 1,107)= 3.281, p = 0.0008. Post hoc comparisons for the vehicle-treated groups showed that aggression indices were significantly (Fishers PLSD<0.05) lower in vehicle-treated animals

796

H A Y D E N - H I X S O N AND FERRIS

TABLE 1 AGGRESSION INDEX PAIRED t-TEST COMPARISONS OF STEROID/VEHICLE DYADS Group

Vehicle AI*

Steroid AI*

.~f

t Value

1.11 +3.30 7.70_+ 4.10

3.11 _+4.50 4.80-+4.71

8 9

0.263 0.345

NS NS

10 - 6 M 10 3 M

9.00 + 6.57 6.00_+2.48

- 0.44 _+3.65 5.38_+2.39

8 7

0.952 0.188

NS NS

[3-Estradiol 10 - 6 M

- 1.50_+5.16

5.63_+6.95

7

0.592

NS

3.13_+3.60

6.88_+2.75

7

-0.646

NS

11.63_+5.11 5.50_+3.71

-2.25_+4.69 10.30_+3.11

7 9

1.602 -0.793

NS NS

8 7

0.521 - 1.639

NS NS

8 10

- 3.607 0.937

0.0069 NS

Testosterone 10 -6 M 10 ~ M

Prob

Dihydrotestosterone

10 3 M

Progesterone 10 6 M 10 3 M Deoxycorticosterone l0

9.11+4.32 -0.50+2.46

6M

10 3 M

4.11_+5.98 18.63_+ 10.25

Cortisol

10-6 M 10 3 M

- 10.78 _+5.14 7.83_+2.83

35.00 + 9.63 3.33_+2.78

*Aggression index (AI) values are means_+ SEM. NS = not significant.

paired with 10 - 6 M cortisol-treated males than vehicle-treated animals paired with all 10 6 M steroid treatment groups except [3-E2, and all 10 - 3 M steroid treatment groups except DOC. Aggression indices were also significantly lower in vehicletreated animals paired with 10 6 M [3-E2-treated males than vehicle-treated animals paired with 10 - 6 M P4-treated males. Post hoc comparisons o f the steroid-treated groups showed that aggression indices were significantly higher in animals treated with 10 - 6 M cortisol than all other steroids; and significantly higher in 10 - 3 M DOC-treated males than animals treated with 10 - 6 M P4 or DHT, and 1 0 - ~ M cortisol. Post hoc comparisons for the aggression index difference scores were identical to those al-

1

-10" -20 -

L0-6MDOSE WI t0-3MDOSE

•

-30" .40"

T

DHT

E2

P4

DOC

F

STEROID TREATMENT

FIG. 2. Aggression index (AI) difference scores for the 10 6 M and 10 - 3 M steroid treatments. Values are means---SEMs. Abbreviations: T = testosterone; DHT = dihydrotestosterone; E2 = [3-estradiol; P4 = progesterone; DOC = deoxycorticosterone; F = cortisol.

ready reported for the steroid-treated animals (see above and Fig. 2~.

Flank Marking Vehicle-treated animals flank marked at higher levels than their opponents only in the case o f the 10 - 3 M cortisol (9/12 pairs) treatment group. Steroid-treated animals flank marked at higher levels than their opponents only in the case o f the 10 - 6 M cortisol (9/9 pairs) and 13-E2 (6/8 pairs) treatment groups. Significant paired t-values were not observed for any o f these effects (see Table 2). Although flank marking F-ratios for the vehicle-treated groups, F ( l I , 1 0 7 1 = 1.751, p = 0 . 0 7 3 7 , and the computed flank marking difference scores, F ( 1 1 , 1 0 7 ) = 1.391, p = 0 . 1 8 9 7 , were not significant, the F-ratio for the steroid-treated animals was significant, F(11,107) = 2.044, p = 0.0321. Post hoc comparisons showed that flank marking levels were significantly higher in vehicletreated males paired with 10 6 M T- or 10 - 3 M cortisol-treated males than in vehicle-treated males paired with 10 - ~ M P4- or DOC-, and 10 - 3 M T-, DHT-, or DOC-treated males. Post hoc comparisons showed that flank marking levels were significantly higher in males treated with 10 - 6 M cortisol than males treated with any 10 - 6 M steroid except T or [3-E2, and any 10 - 3 M steroid except P4. In addition, males treated with 10 ~ M [3-E~ flank marked significantly more than males treated with 10 - 6 M DOC or any 1 0 - 3 M steroid except P4 o r DOG. Significantly higher flank marking difference scores were observed for the 10 6 M cortisol treatments than all 10 - 6 M treatments except T and [3-E 2, and all 10 s M treatments except P4 or DOC (see Fig. 3). DISCUSSION

The present data show that acute microinjections o f 10 e M cortisol and 10 3 M deoxycorticosterone (DOC) into the ante-

AGGRESSION IN MALE HAMSTERS

797

TABLE 2 FLANK MARKING PAIRED t-TEST COMPARISONS OF STEROID/VEHICLE DYADS

Group

Vehicle FM*

Steroid FM*

df

t Value

Prob

4.78___2.89 0.20 ±0.13

7.56---4.93 0.50 ±0.31

8 9

-0.431 -0.896

NS NS

M

2.67 ± 1.34 1.25 ±0.65

4.00 + 1.34 0.63 ---0.50

8 7

-0.676 0.919

NS NS

13-Estradiol 10-6 M 10 3 M

3.16 ---0.93 1.63 ±0.98

10.25 - 3.90 1.63 ± 1.03

7 7

- 1.771 0.000

NS NS

Progesterone 10 - 6 M 10 -3 M

0.63 ---0.26 1.80 ± 0.83

1.63 ± 1.21 6.40 ± 2.99

7 9

-0.837 - 1.397

NS NS

0.11 ±0.11 0.13±0.13

0.22---0.15 2.13±1.23

8 7

-0.555 -1.578

NS NS

1.44± 1.44 4.17 -+ 1.22

12.89---7.03 1.50 ± 1.23

8 11

- 1.596 2.053

NS 0.0646

Testosterone 10 - 6 M

10 -3 M Dihydrotestosterone 10 - 6 M 10 - 3

Deoxycorticosterone 10 - 6 M

10 3 M Cortisol 10 - 6 M 10- 3 M

*Flank mark (FM) values are means ± SEM. NS = not significant.

rior hypothalamus induced aggression in animals during paired encounters with vehicle-treated conspecifics. However, only animals microinjected with 10 - 6 M cortisol displayed significantly elevated aggression and flank marking. This particular display of agonistic and communicative behavior by cortisol-treated animals was accompanied by an equally significant show of submissive behavior by their vehicle-treated opponents. While DOCtreated animals showed heightened aggression, their vehicletreated opponents showed no inclination toward submissive responding. Hence, cortisol microinjection appears to affect agonistic responding, and possibly the physiochemistry of scent marking of the recipient in such a manner that it affects the perception and behavior of a potential combatant. Likewise, [3-estradiol (13-E2) microinjections into the anterior hypothalamus 20"

r~ ~e

16" 12"

7 ee

.~

4'

o. -4" -S"

• []

IO-6MDOSE 1o-3MDOSE

-12 T

DHT

E2

P4

DOC

F

STEROID TREATMENT

FIG. 3. Flank marking (FM) difference scores for the 10 - 6 M and 10 - 3 M steroid treatments. Values are means ___SEMs. Abbreviations: same as Fig. 2.

induced flank marking in animals during paired encounters with vehicle-treated opponents. However, the behavioral interactions following estradiol treatment were not accompanied by any significant change in aggression by either member of a pair. The results of this study clearly show that cortisol and estradiol exert immediate actions that affect the normal topology, i.e., form and intensity, of agonistic responding in this species. In fact, these results suggest that at least two, possibly distinct, hormone-sensitive mechanisms interact with the neural correlates of aggressive and communicative behaviors in the anterior hypothalamus of male golden hamsters. The form and intensity of intermale agonistic responding in male hamsters is well defined. Either high levels of overt aggression accompanied by low levels of flank marking, or high levels of flank marking accompanied by low levels of aggression, appear to correlate with dominant social status during agonistic encounters between male hamsters (8, 18, 19, 46). Specifically, the relative frequencies of overt aggression and flank marking change over the course of consecutive daily pairings between the same two male hamsters. During the initial encounter between two unfamiliar males, overt aggression is the predominant agonistic response. Over successive daily encounters between the same two animals, the frequency of flank marking increases, and the frequency of overt aggression decreases in the more dominant animal. Typically, by the fifth consecutive pairing, flank marking becomes the predominant response, with absolute levels being higher in dominant than subordinate males (8). Therefore, a behavioral profile consisting of high levels of aggression and flank marking, similar to that observed in animals treated with 10 - 6 M cortisol, is more characteristic of the agonistic responding of the more aggressive/dominant male in its second or third, than first, pairing with the same conspecific (8). This finding suggests very rapid increases in circulating glucocorticoids, like those that occur during social conflict in this

7~s

H~x,'~I)EN-HIXSON AND [aERRIS

species (16, 17~, could have immediate, and functionally significant effects on ongoing behaviors. Specifically, the results of this study suggest glucocorticoids can have rapid central actions that promote a profile of agonistic behavior that maximizes the likelihood of an animal successfully subjugating an opponent, while simultaneously minimizing the potential risk of serious injury to itself. If this is the case, it appears glucocorticoids have rapid modulatory actions that promote adaptive behavioral responding in this species. In this context, it is also worth noting that the vehicle-treated opponents of the 10 ~ M cortisol-treated animals exhibited response profiles characteristic of the behavior of a submissive animal repeatedly subordinated by its opponent. Because all vehicleand steroid-treated animals exhibited comparable levels of aggression during the initial screening procedure, these findings suggest male hamsters with more rapid adrenocortical responses may have a competitive advantage in pairings with conspecifics of comparable social status. However, the biphasic effects of acute cortisol microinjections in the anterior hypothalamus on agonistic responding suggest upper and lower "critical concentration" thresholds for aggression-promoting actions of glucocorticoids may exist. Correlational evidence suggests endogenously released cortisol promotes aggression in male hamsters. Specifically, the highest diurnal levels of aggression coincide with the time of peak basal glucocorticoid release in this species (23). While there is no evidence that changes in circulating cortisol levels affect endogenous cortisol levels in the anterior hypothalamus of hamsters, glucocorticoid levels in the anterior hypothalamus of rats increase approximately 2 fold after ACTH treatment, and decrease 34 fold following adrenalectomy (27). Like cortisol, animals treated with 10 6 M, [3-estradiol exhibited flank-marking profiles more characteristic of the behavior of a dominant animal during its third or fourth, than first encounter with the same opponent, i.e., high levels of flank marking. Unlike cortisol, [3-estradiol-treated animals were not highly aggressive, and their vehicle-treated opponents did not exhibit response profiles typical of subordinate animals after three or four repeated pairings. In fact, the form and intensity of aggressive responding in both animals in the vehicle/[3-estradiol treatment pairs were very similar. The observed behaviorally selective effect of [3-estradiol suggests the immediate stimulatory effects of cortisol and [3-estradiol in this study probably involved different mechanisms of action. Alternatively, it is equally possible that these two steroids normally have very different doseresponse effects on a common mechanism. However, evidence suggests that lower levels of estradiol probably would not be

more effective. Female hamslers are reportedly less responslxc to behavioral effects of estradiol implants than female rats 120!, and male hamsters even less responsive to estradiol than female

hamsters (6). While not notable in its own right, the observed trend toward biphasic effects with the two doses of progesterone used in this study is interesting for other reasons. Progesterone implants inhibit aggression in the ventromedial hypothalamus, but not the anterior hypothalamic area of ovariectomized female hamsters (29,43). If progesterone exhibits similar site-specific actions in intact male hamsters, then progesterone microinjections may also have more pronounced effects on behavioral regulation in the ventromedial hypothalamus than anterior hypothalamus. Acute peripheral P~ administration inhibits aggression in both male (10! and female (l(/, 30, 43) hamsters. Hence, it seems unlikely sexdependent effects of acute anterior or ventromedial hypothalamic Pa administration on aggression would exist. In conclusion, the purpose of this study was to determine whether steroid hormones exert immediate effects on agonistic responding in male golden hamsters after being directly administered into the anterior hypothalamus, an area of the brain implicated in the central regulation of agonistic behavior I I, 9, 13, 15, 21, 22, 41 ). The findings of this study suggest that adrenal and gonadal hormones exert immediate, behaviorally selective, and steroid-specific effects on agonistic responding when their local concentrations increase in the anterior hypothalamic area of male golden hamsters. While considerable effort has been made to examine the functional significance of long-term changes in the peripheral levels of specific adrenal and gonadal hormones as they relate to behavior, this study is among the first to examine the specificity and possible functional significance of very rapid changes in the levels of circulating glucocorticoids that typically occur during exposure to relatively mild environmental stressors like social competition. It is important to note that these findings may have limited generalizability. This study focused on hormone sensitivity in one very specific area of the brain, on a fairly restricted class of behaviors, using a very limited number of steroid concentrations. In spite of these limitations, the findings of this study clearly demonstrate that the potential for other immediate, and functionally significant concomitants of acutely administered steroids in neurobehavioral regulation. ACKNOWLEDGEMENTS Preliminary findings of this research were presented at the 20th Annual meeting of the Society for Neuroscience, Abstract No. 308.16. This work was supported by NIH grant NS 23557 (C.F.F.i and a grant from the Harry, Guggenheim Foundation /C.F.F.I.

REFERENCES I. Albert, D. J.; Walsh, M. L. Neural systems and the inhibitory modulation of agonistic behavior: A comparison of mammalian species. Neurosci. Biobehav. Rev. 8:5-24; 1984. 2. Blaustein, J. D.; Olster, D. H. Gonadal steroid hormone receptors and social behaviors. In: Balthazart, J., ed. Advances in comparative and environmental physiology, vol. 3. Berlin: Springer-Verlag; 1989:31-104. 3. Brain, P. F. Effects of hormones of the pituitary adrenal axis on behavior. In: Brown, K.; Cooper, S. J., eds. Chemical influences on behaviour. New York: Academic Press; 1979:331-372. 4. Brain, P. F. Pituitary-gonadal influences on social aggression. In: Svare, B. B., ed. Hormones and aggressive behavior. New York: Plenum Press; 1983:3-25. 5. Bronson, F. H. Establishment of social rank among grouped male mice: Relative effects on circulating FSH, LH, and corticosterone. Physiol. Behav. 10:947-951; 1973. 6. Ciaccio, L. A.; Lisk, R. D. Effect of hormone priming on retention

of 3H-oestradiol by males and females. Nature 236:82-83; 1972. 7. Doherty, P. C.; Sheridan, P. J. Uptake and retention of androgen in neurons of the brain of the golden hamster. Brain Res. 219:327334; 1981. 8. Ferris, C. F.; Axelson, J. F.: Shinto, L. J.; Albers, H. E. Scent marking and the maintenance of dominant/subordinate status in male golden hamsters. Physiol. Behav. 40:661-664; 1987. 9. Ferris, C. F.; Potegal, M. Vasopressin receptor blockade in the anterior hypothalamus suppresses aggression in hamsters. Physiol. Behav. 44:235-239; 1988. 10. Fraile, I. G.; MeEwen, B. S.; Pfaff, D. W. Progesterone inhibition of aggressive behavior in hamsters. Physiol. Behav. 39:225-229; 1987. 11. Fraile, I. G.; Pfaff, D. W.; McEwen, B. S. Progestin receptors with and without estrogen induction in male and female hamster brain. Neuroendocrinology 45:487-491 ; 1987. 12. Grelk, D. F.; Papson, B. A.; Cole, J. E.; Rowe, F. A. The influ-

A G G R E S S I O N IN M A L E H A M S T E R S

13.

14.

15.

16.

17.

18.

19. 20.

21.

22.

23. 24.

25.

26.

27.

ence of caging conditions and hormone treatments on fighting in male and female hamsters. Horm. Behav. 5:355-366; 1974. Hammond, M. A., Rowe, F. A. Medial preoptic and anterior hypothalamic lesions: Influences on aggressive behavior in female hamsters. Physiol. Behav. 17:507-513; 1976. Harrison, N. L.; Majewska, M. D.; Meyers, D. E. R.; Barker, J. L. Rapid actions of steroids on CNS neurons. In: Lakoski, J. M.; Perez-Pollo, J. R.; Rassin, D. K., eds. Neural control of reproductive function. New York: Alan R. Liss;1989:137-166. Hayden-Hixson, D. M.; Ferris, C. F. Cortisol exert site-, context-, and dose-dependent effects on agonistic behaviors of male golden hamsters. J. Neuroendocrinol.; in press. Huhman, K. L.; Bunnell, B. N.; Mougey, E. H.; Meyerhoff, J. L. Effects of social conflict on POMC-derived peptides and glucocorticoids in male golden hamsters. Physiol. Behav. 47:949-956; 1990. Huhman, K. L.; Moore, T. O.; Ferris, C. F.; Mougey, E. H.; Meyerhoff, J. L. Acute and repeated exposure to social conflict in male golden hamsters: Increases in plasma POMC-peptides and cortisol and decreases in plasma testosterone. Physiol. Behav.; submitted. Johnston, R. E. Scent marking by male golden hamsters (Mesocricetus auratus): I. Effects of odors and social encounters. Z. Tierpsychol. 37:75-98; 1975. Johnston, R. E. Scent marking by male hamsters. III. Behavior in a semi-natural environment. Z. Tierpsychol. 37:213-221; 1975. Krieger, M. S.; Morrell, J. I.; Pfaff, D. W. Autoradiographic localization of estradiol-concentrating cells in the female hamster brain. Neuroendocrinology 22:193-205; 1976. Lammers, J. H. C. M.; Kruk, M. R.; Meelis, W.; Van der Poel, A. M. Hypothalamic substrates for brain stimulation-induced patterns of locomotion and escape jumps in the rat. Brain Res. 449: 294-310; 1988. Lammers, J. H. C. M.; Kruk, M. R.; Meelis, W.; Van der Poel, A. M. Hypothalamic substrates for brain stimulation-induced attack, teeth chattering and social grooming in the rat. Brain Res. 449:311327; 1988. Landau, I. T. Light-dark rhythms in aggressive behavior of the golden hamster. Physiol. Behav. 14:767-774; 1975. Landau, I. T. Effects of adrenalectomy on rhythmic and non-rhythmic aggressive behavior in the male golden hamster. Physiol. Behav. 14:775-780; 1975. Leshner, A. I. The hormonal responses to competition and their behavioral significance. In: Svare, B. B., ed. Hormones and aggressive behavior. New York: Plenum Press; 1983:393-404. Lisk, R. D.; Nachtigall, M. J. Estrogen regulation of agonistic and proceptive responses in the golden hamster. Horm. Behav. 22:3548; 1988. Magarifios, A. M.; Ferrini, M.; De Nicola, A. F. Corticosteroid receptors and glucocorticoid content in microdissected brain regions: Correlative aspects. Neuroendocrinology 50:673-678; 1989.

799

28. McEwen, B. S.; De Kloet, E. R.; Rostene, W. Adrenal steroid receptors and actions in the nervous system. Physiol. Rev. 66:11211188; 1986. 29. Meisel, R. L.; Fraile, I. G.; Pfaff, D. W. Hypothalamic sites of progestin action on aggression and sexual behavior in female Syrian hamsters. Physiol. Behav. 47:219-223; 1990. 30. Meisel, R. L.; Sterner, M. R.; Diekman, M. A. Differential hormonal control of aggression and sexual behavior in female Syrian hamsters. Horm. Behav. 22:453-466; 1988. 31. Meyer, J. S. Biochemical effects of corticosteroids on neural tissues. Physiol. Rev. 65:946-1020; 1985. 32. Mooradian, A. D.; Morley, J. E.; Korenman, S. G. Biological actions of androgens. Endocr. Rev. 8:1-28; 1987. 33. Munn, A. R., III; Sar, M.; Stumpf, W. E. Topographic distribution of progestin target ceils in hamster brain and pituitary after injection of [3H]R5020. Brain Res. 274:1-10; 1983. 34. Rories, C.; Spelsberg, T. C. Ovarian steroid action on gene expression: Mechanisms and models. Annu. Rev. Physiol. 51:653-681; 1989. 35. Sachser, N. Short-term responses of plasma norepinephrine, epinephrine, glucocorticoid and testosterone titers to social and nonsocial stressors in male guinea pigs of different social status. Physiol. Behav. 39:11-20; 1987. 36. Sapolsky, R. M. The endocrine stress-response and social status in the wild baboon. Horm. Behav. 16:279-292; 1982. 37. Schindler, W. J.; Knigge, K. M. Adrenal cortical secretion by the golden hamster. Endocrinology 65:739-747; 1959. 38. Schumacher, M.; McEwen, B. S. Steroid and barbituate modulation of the GABAa receptor: Possible mechanisms. Mol. Neurobiol. 3:275-304; 1989. 39. Schur, L. W. Social structure and plasma corticosterone level in female albino mice. Physiol. Behav. 40:689-693; 1987. 40. Siegel, A.; Pott, C. B. Neural substrates of aggression and flight in the cat. Prog. Neurobiol. 31:261-283; 1988. 41. Sutanto, W.; De Kloet, E. R. Species-specificity of corticosteroid receptors in hamster and rat brains. Endocrinology 121:1405-1411 ; 1987. 42. Takahashi, L. K.; Lisk, R. D. Diencephalic sites of progestin action for inhibiting aggression and facilitating sexual receptivity in estrogen-primed golden hamsters. Endocrinology 116:2393-2399; 1985. 43. Van de Poll, N. E.; Van Zanten, S.; De Jonge, F. H. Effects of testosterone, estrogen, and dihydrotestosterone upon aggressive and sexual behavior of female rats. Horm. Behav. 20:418--431; 1986. 44. Vito, C. C.; DeBold, J. F.; Fox, T. O. Androgen and estrogen receptors in adult hamster brain. Brain Res. 264:132-137; 1983. 45. Whitsett, J. M. The development of aggressive and marking behavior in intact and castrated male hamsters. Horm. Behav. 6:47-57; 1975.