Psychopharmacology (1994) 116 : 491-498

Psychopharmacology © Springer-Verlag 1994

Defeat engenders pentylenetetrazole-appropriate responding in rats: antagonism by midazolam J.A. Vivian, E.M. Weerts, K.A. Miczek Department of Psychology, Tufts University, Medford, MA 02155, USA Received : 29 January 1994 / Final version: 30 March 1994

Abstract. Defeat and the threat of defeat by an aggressive conspecific is stressful and may engender an anxietyor fear-like state in animals; the present experiment investigated whether defeat generalized to the discriminative stimulus properties of P T Z and how benzodiazepine receptors were involved in this generalization. Separate groups of male Long-Evans rats (Rattus norvegicus) were trained to discriminate 20 mg/kg pentylenetetrazole (PTZ) or 0.4 mg/kg midazolam (MDZ) from saline in a two-choice drug-discrimination task. After establishing stimulus control, PTZ- and MDZ-trained rats were exposed to an aggressive conspecific which resulted in defeat, as defined by the display of defensive and submissive postures as well as audible and ultrasonic vocalizations. Administration of saline after defeat resulted in greater than 80 % P T Z lever selection in 15 out of 25 PTZtrained rats; this effect was attenuated through pretreatment with M D Z (1 mg/kg). Furthermore, short-term defeat substitution for the PTZ discriminative stimulus was not accompanied by long-term changes in the postdefeat generalization curves for PTZ and M D Z when compared to pre-defeat generalization curves. N o r did defeat alter the antagonism of PTZ by diazepam (2.5 mg/kg) or M D Z by flumazenil (10 mg/kg). In order further to characterize the necessary features for defeat substitution for the PTZ discriminative stimulus, exposure to a threatening conspecific was also attempted by PTZtrained rats protected from physical contact with a wire mesh cage. In these tests, saline continued to engender greater than 50 % PTZ lever responding in 15 of 25 rats. These results suggest that an anxiety-like state is induced during defeat and exposure to a threatening conspecific in most rats; this state, as well as the PTZ discriminative stimulus, can be reversed by benzodiazepine receptor agonists. In contrast, short-term defeat substitution for the P T Z discriminative stimulus does not appear to be

*Present address: Division of Behavioral Biology, Johns Hopkins University School of Medicine, Hopkins Bayview Research Campus, Baltimore, MD 21224, USA Correspondence to." K.A. Miczek, Tufts University, Research Building, 490 Boston Avenue, Medfbrd, MA 02155, USA

related to long-term alterations in the benzodiazepine receptor.

Key words:

Aggression - Anxiety - Anxiolytic Benzodiazepine Defeat Defense Drug discrimination

Drug discrimination procedures have proven useful in the study of a variety of anxiolytic and anxiogenic compounds such as diazepam (Ator and Griffiths 1989), pentobarbital (DeYry and Slangen 1986) and pentylenetetrazole (PTZ; Lal and Shearman 1982). PTZ is reported to produce anxiety in humans (Rodin and Calhoun 1970). In animal studies, drug discrimination procedures using P T Z as a discriminative stimulus have been applied to the study of various anxiolytic drugs including the benzodiazepine receptor agonists chlordiazepoxide and diazepam, barbiturates and meprobamate (Lal and Shearman 1982). There is also evidence for "anxiogenic" effects of PTZ, as the benzodiazepine receptor inverse agonists fi-CCE, fi-CCM and Ro15-4513 substituted for PTZ (Lal and Emmett-Oglesby 1983; Harris et al. 1987; Ator et al. 1989). Also, rats which were trained to discriminate PTZ and were subsequently withdrawn from diazepam, morphine, alcohol, nicotine or cocaine responded on the PTZ-appropriate lever, suggesting that certain forms of distress such as withdrawal can cross-generalize to a PTZinduced drug effect (Emmett-Oglesby et aI. 1983, 1984; Harris et al. 1986; Lal et al. 1988; Wood et al. 1989). The validity of PTZ as an "anxiogenic" discriminative stimulus relies on the pharmacological validity of compounds which substitute for or antagonize the PTZ cue. It has proven difficult, however, to separate the proconvulsant activity of P T Z from its anxiety-like effects (Andrews and Stephens 1990). Inferences based on this method would be strengthened if non-drug, endogenous fear- or anxiety-like states substituted for the PTZ discriminative stimulus. Defeat during an agonistic encounter is a behavioral manipulation that may induce an anxiety-like state as it

492 has been s h o w n to alter physiological correlates indicative o f stress a n d heightened " e m o t i o n a l i t y " . I n c r e a s e d h e a r t rate, t e m p e r a t u r e a n d c o r t i c o s t e r o n e have been d e m o n s t r a t e d after e x p o s u r e to an a t t a c k i n g conspecific ( F o k k e m a et al. 1988; M i c z e k et al. 1991; T o r n a t z k y a n d M i c z e k 1993). It m a y be possible to evaluate p r o c e d u r e s p u r p o r t i n g to e n g e n d e r a n x i e t y - such as aggressive b e h a v i o r - b y e x p o s i n g P T Z - t r a i n e d subjects to these p r o cedures followed b y d r u g d i s c r i m i n a t i o n sessions. In fact, P T Z d i s c r i m i n a t i n g rats injected with saline r e s p o n d e d on the P T Z lever after e x p o s u r e to a p r e d a t o r such as a cat ( G a u v i n a n d H o l l o w a y 1991). I m m e d i a t e l y after defeat in a social c o n f r o n t a t i o n , p a r t i a l g e n e r a l i z a t i o n to the P T Z discriminative stimulus was o b s e r v e d as 9 o u t o f 22 rats r e s p o n d e d on the P T Z lever (Vellucci et al. 1988). The s u b s t i t u t i o n o f socially significant stressors for the P T Z discriminative stimulus p r o v i d e s v a l i d a t i o n for m e t h o d s e m p l o y i n g p r e d a t o r y a n d agonistic e n c o u n t e r s in the assessment o f anxiety-like states a n d c o m p o u n d s t h a t are effective in its t r e a t m e n t . D r u g d i s c r i m i n a t i o n m e t h o d s m a y also p r o v i d e inform a t i o n a b o u t l o n g - t e r m r e c e p t o r changes due to e n d o g e nous processes t h a t are altered by e t h o l o g i c a l l y valid challenges. I f a n x i e t y - p r o d u c i n g p r o c e d u r e s such as d e f e a t cause l o n g - t e r m changes at the G A B A a - b e n z o d i a z e p i n e r e c e p t o r c o m p l e x o r in the activity o f e n d o g e n o u s anxiogenic substances ( A l h o et al. 1985), d e t e r m i n i n g generalization o r a n t a g o n i s m curves before a n d after the defeat m a y reveal the i m p a c t o f such p u t a t i v e e n d o g e n o u s processes. U s i n g a d r u g d i s c r i m i n a t i o n p r o c e d u r e , the overall objective o f the c u r r e n t e x p e r i m e n t was to c h a r a c t e r i z e defeat in the i n t r u d e r c o n f r o n t i n g a resident ( M i c z e k 1979) using P T Z - or m i d a z o l a m ( M D Z - ) t r a i n e d rats. Defeat in a social c o n f r o n t a t i o n c o m p r i s e s m a j o r e n d o c r i n e a n d c a r d i o v a s c u l a r stress responses as well as C N S changes indicative o f " f e a r " and " a n x i e t y " ( M i c z e k et al. 1991). A f t e r establishing M D Z a n d P T Z stimulus control, rats were t h r e a t e n e d a n d a t t a c k e d resulting in defeat b y an aggressive conspecific. I m m e d i a t e l y after defeat, P T Z - t r a i n e d rats were tested for s u b s t i t u t i o n o f the d e f e a t - i n d u c e d state for P T Z lever selection. I n a d d i tion, a n t a g o n i s m o f the d e f e a t - i n d u c e d anxiety-like state using the b e n z o d i a z e p i n e r e c e p t o r a g o n i s t M D Z was also assessed. Lastly, g e n e r a l i z a t i o n a n d a n t a g o n i s m o f P T Z a n d M D Z were c o m p a r e d before a n d after d e f e a t to evaluate c h a n g e s in the r e c e p t o r - m e d i a t e d d i s c r i m i n a t i v e stimulus effects.

Materials and Methods Subjects Thirty-six male hooded Long-Evans rats (Charles River Laboratories, Wilmington, Moss.) weighing 350 375 g at the beginning of the experiment were housed individually in standard stainless steel hanging cages with wire mesh flooring. They were maintained at 80 85% of their free-feeding weights with Purina rodent chow pellets given immediately after the daily experimental session. Additional male and female hooded Long-Evans rats (males: 550-650 g, females: 250 350 g) were pair-housed in 45.7 × 45.7 x

7 l. 1 cm stainless steel cages with sawdust bedding; males were selected for their consistent display of aggressive behavior and were used as stimulus residents in agonistic encounters. All animals were housed in an environmentally controlled vivarium (12/12-h light/dark cycle, 21_+1° C, 30-40% humidity) and had free access to water. Pair-housed rats had free access to Purina rodent chow and all experimental sessions were performed during the light cycle. All animals were maintained in accordance with USDA regulations and the experimental protocol was approved by the Tufts University Animal Care and Use Committee. Apparau~s Three standard experimental chambers for operant conditioning (BRS/LVE, Laurel, Md.) were used. Mounted on one walt of each chamber were two levers 5 cm above the floor, 22 cm apart and a dipper positioned between the levers for presentation of 0.01 ml milk (50% condensed milk, 50% water). Each chamber was placed in a light-proof, fan-ventilated and sound-attenuating enclosure illuminated with a 7.5 W white light. An Apple II computer connected to an interface (Med Associates, East Fairfield, Vt.) controlled the scheduling of events and recorded data. A lever press defined a response. Procedure Training. One half of the rats (n = 18) were trained to lever press on the left lever, the other half were trained on the right lever. Rats were initially trained to respond for milk reinforcement under a fixed ratio 10 requirement (FR10) and then adapted to saline injections given before the session (ca, five sessions). Finally, rats were administered PTZ (n=27; 20 mg/kg training dose) o1" MDZ 0l = 9; 0.4 m J k g training dose) 15 or 5 min prior to the experimental session, respectively. During discrimination sessions both levers were presented and reinforcement was contingent upon completion of the schedule (FR10) and the drug administered. Responding on the drug-inappropriate lever was recorded but without consequence. Alter the rats demonstrated a stable response rate and selected the appropriate lever before the first reinforcement in nine of ten consecutive sessions (i.e., 90% correct in 90% of the preceding sessions), generalization tests began. Generalization and antagonism. Generalization tests were performed using a cumulative dosing procedure which allowed for the assessment of a complete dose-effect curve within a single session (Wenger 1980). After administration of a test dose of PTZ, rats were returned to their home cage for a 15-min injection-test interval and then placed into the operant chamber. These sessions were terminated after ten responses on either lever without reinforcement (extinction tests) or 10 min. Rats were removed from the operant chambers after completion of the response contingency, immediately injected with the next dose of PTZ and returned to their home cage. Each of the four doses was administered at ca. 15-min intervals. In additional experiments, diazepam (chosen for its longer duration of action over the cumulative dosing period in comparison to MDZ; Greenblatt et al. 1981) was administered 30 rain prior to the first extinction test in PTZ-trained rats. Similarly, MDZ-trained rats were administered MDZ doses at 5-rain injection-test intervals using the cumulative dosing procedure; in antagonism tests, flumazenil was administered 15 min prior to the first extinction test. Generalization and antagonism tests were performed before and after defeat experience (see below). Time course. The time course of the discriminative stimulus effects of PTZ was assessed after administration of the training dose of PTZ. These extinction sessions were performed at 5, 15, 30, 45, 60, 90, 120, 150 and t 80 min after administration of PTZ and were terminated after ten lever presses on either lever.

493

Social housing. Generalization tests with a subset of PTZ (n = 9) and M D Z (n = 9) rats were perfbrmed after a 2-week interval in which rats were pair-housed with a same sex conspecific. Subsequently, rats were returned to isolated housing for 2 weeks before a third generalization curve was determined. Defeat substitution. PTZ-trained rats serving as "intruders" were placed in the home cage of a male conspecific ("resident"). These resident-intruder encounters invariably led to attacks by the resident with definitive signs of defense and submission exhibited by the intruder (i.e., upright postures, escapes, crouch postures and ultrasonic vocalizations); they were terminated after the display of 5 uninterrupted seconds of the supine posture. Intruder rats were removed, injected with saline, and then placed into a protective cage within the resident's cage for 15 min. After this interval, rats were placed into the operant chamber and allowed to complete ten responses in 10 rain. These defeat experiments in saline-injected intruders were performed seven times. The first three defeats were in successive experimental sessions, each separated by 1 day. Subsequently, it was necessary to re-establish pre-defeat response rates. After successful retraining (ca. 10 15 sessions), generalization and antagonism studies followed by further defeat substitutions were performed. Finally, in an attempt to antagonize the discriminative stimulus resulting fiom defeat, intruder rats (n = 14) were injected with saline and midazolam during the sixth exposure to the aggressive opponent. M D Z 1 mg/kg was chosen to antagonize the discriminative stimulus associated with attack exposure due to its quick onset of action and ability to reverse the PTZ discriminative stimulus (Vellucci et al. 1988). Finally, saline was administered during the seventh defeat test. MDZ-trained rats were also exposed to three defeat experiences separated by 24 h but were returned to their home cage. To maintain stimulus control, two lever training sessions with M D Z or saline were performed 4-6 h before being exposed to the aggressive resident. Five to eight days after the third defeat, flumazenil antagonism of the M D Z discrimination was studied.

Threat of defeat substitution. These tests differed from the previously described "defeat" tests in that the intruder was not physically attacked by the resident. PTZ-trained rats were placed into the protective cage within the resident's home cage until the production of the first ultrasonic vocalization (US). Subsequently, rats were administered saline and returned to the protective cage for 15 min. At the end of the exposure to the threatening resident, rats were removed from the protective cage, placed into the operant chamber and allowed to complete ten responses on either lever in 10 min. The experimental protocol is depicted in Fig. 1. Drugs, Pentylenetetrazole (Sigma; cumulative doses of 2.5, 5, 10, 20 and 40 mg/ml dissolved in 0.9% physiological saline) was administered

Schedule

intraperitoneally 15 min prior to the experimental session. Diazepam (Hoffman-LaRoche; 2.5 mg/ml suspended in a solution containing 85% distilled water, 14% propylene glycol and 1% Tween 80) was administered subcutaneously 30 rain prior to the experimental session. In PTZ-trained rats, midazolam (HoffmanLaRoche; 1 m J m l dissolved in 0.9% physiological saline) was administered subcutaneously 15 min prior to the experimental session. In MDZ-trained rats, midazotam (cumulative doses of 0.03, 0.1, 0.3, 1, 3 mg/ml dissolved in 0.9% physiological saline) was administered subcutaneously 5 min prior to the experimental session. Flumazenil (Hoffman-LaRoche; 10 mg/ml suspended in a solution containing 85% distilled water, 14% propylene glycol and 1% Tween 80) was administered intraperitoneally 15 rain prior to the experimental session. All drugs were administered in a volume of 1 ml/kg body weight.

Data analysis Discrimination data were analyzed as the mean percentage of drug lever responding in relation to total responses. Fifty percent effective doses (ED50) and 95% confidence intervals were estimated from first order regression equations; non-overlapping confidence interval comparisons were accepted as statistically significant. For the comparison of midazolam antagonism during defeat, a one-way (drug) repeated measures ANOVA with pairwise contrasts was performed on the mean percentage of drug lever responding. Alpha was 0.05, two-tailed.

Results P T Z and M D Z discrimination A p p r o x i m a t e l y 50 sessions (range: 1 6 - 1 0 8 sessions) w e r e required until stimulus control was demonstrated (90% c o r r e c t , n i n e o u t o f ten c o n s e c u t i v e sessions) in P T Z t r a i n e d rats. U s i n g the 10 sessions p r e c e d i n g the first gene r a l i z a t i o n test, l e v e r s e l e c t i o n ( M e a n + S E M ) w a s 94 + 1% P T Z - a p p r o p r i a t e at a r a t e o f 32.2 + 1 r e s p o n s e s p e r m i n u t e a n d 93 +_ 2 % s a l i n e - a p p r o p r i a t e at a r a t e o f 44.1 _+ 1.1 r e s p o n s e s p e r m i n u t e . M D Z - t r a i n e d rats d e m o n s t r a t e d a c c u r a t e d i s c r i m i n a t i o n a f t e r ca. 37 sessions (range: 18-56 sessions). L e v e r s e l e c t i o n w a s 96_+ 1% MDZ-appropriate at a r a t e o f 45.7 + 3 r e s p o n s e s p e r m i n u t e a n d 96_+ 1% s a l i n e - a p p r o p r i a t e at a r a t e o f 47.3 + 2 r e s p o n s e s p e r m i n u t e . T w o P T Z rats failed c o n sistently to d i s c r i m i n a t e P T Z a n d w e r e n o t i n c l u d e d in s u b s e q u e n t analyses. After administration of the training dose of PTZ (20 m g / k g ) , P T Z lever s e l e c t i o n p e a k e d (ca. 90%) b e t w e e n 15 a n d 60 m i n ; s u b s e q u e n t l y , P T Z l e v e r s e l e c t i o n d i m i n i s h e d g r a d u a l l y o v e r t h e r e m a i n i n g 2 h.

Discrimination Generalization Antagonism

TimeCotrcse

Defeat Threat Generalization Antagonism

FR10

FRIO

FRt0

FR10 FR10

FRt0

FRIO

FRIO

Drug SAL vs. PTZ Dose(mg/kg) 20

PTZ 2.5,5, 10,20

DZP (2.5 mg/kg)+ PTZ PTZ 5, 10,20,40 20

SAL

SAL

PTZ 2.5,5,10,20

DZP (2.5 mg/kg)+ Frz 5,10,20,40

Drug SAL vs. MDZ Dose (mg/kg) 0.4

MDZ 0.03, 0.1, 0.3

FLZ (10 mg/kg)+ MDZ MDZ 0.3, I, 3 N/A

N/A N/A

N/A N/A

MDZ N/A

FLZ (10 mg/kg) + MDZ 0.3, 1, 3

Session Dur

10 rain

10 rain

10 min 10 rain I0 min

30->10 rain

Fig. 1. Experimental protocol

10 min

10 min

494 Table 1. Effects of defeat on PTZ lever selection

Subject 1

2 3 4b 5b 6 7 8b 9a,b l0 b 11 12b 13b 14 15 16a'b 17 18 19~ 20 21 22 23 a'b 24 25

Defeat 1 Lever responses PTZ Total

% PTZ

Defeat 2 Lever responses PTZ Total

10 10 10 l0 10 10 10 10 10 10 t0 10 10 10 10 10 10 10 5 4 3 3 2 1 0

100 100 100 100 100 100 100 91 91 91 91 83 83 83 83 77 77 56 33 29 23 23 17 9 0

10 t0 I0 10 10 10 i0 2 0 10 6 10 10 10 10 2 10 3 0 10 t0 5 0 3 0

10 10 10 10 10 10 I0 11 11 11 11 I2 12 10 12 13 13 i8 15 14 13 13 12 11 10

12 10 11 10 10 11 14 12 0 10 16 14 11 15 16 3 10 13 0 10 10 15 2 13 10

% PTZ

Defeat 3 ......... Lever responses PTZ Total

83 100 9t

10 t0 10

100

t

1

100 91 71 17

9 10 4 1

10 11 14 1

-

100 35 71 91 67 63 67 100 23 -

100 I00 33 0 23 0

% PTZ

17t2 t6

1

1

8 2 1 3 10 t0 6 10 0

I6 12 1 3 15 16 6 11 10

10

10

10 I0 10 8 10 10

10 I3 11 9 17 10

59 83 63 100 90 91 29 100 100 50 17 100 100 67 63 100 91 0 100 i00 77 91 89 59 I00

-

a denotes failure to complete one FR10 schedule a~er defeat 2 b denotes failure to complete one F R I 0 schedule after defeat 3

Pre-dejeat generalization and antagonism Individual and social housing. Housing condition did not alter the PTZ or MDZ dose-effect functions. Cumulative doses of PTZ dose-dependently increased drug-appropriate responding (Fig. 2 - h o u s i n g data not shown); the ED50s (Mean_+95% confidence interval) were 9.96 _+1.23 mg/kg and 9.53_+1.56 mg/kg during individual and social housing conditions, respectively. Similarly, MDZ dosedependently increased drug-appropriate responding (Fig. 3 - housing data not shown); the EDs0 was 0.15 _+ 0.01 mg/kg during individual housing versus 0.14 +_ 0.01 mg/kg during social housing.

Diazepam antagonism of PTZ. Diazepam produced a twofold rightward shiR in the PTZ dose-effect function. In the presence of 2.5 mg/kg diazepam, ascending cumulative doses of PTZ dose-dependently increased PTZ lever responding; the EDs0 (Mean + 95%CI) was 21.63 + 3.29 mg/kg (Fig.2).

Flumazenil antagonism of MDZ. Flumazenil produced a tenfold rightward shift in the MDZ dose-effect function. In the presence of 10 mg/kg flumazenil, MDZ dose-dependently increased MDZ-appropriate responding; the EDs0 was 1.42+0.38 mg/kg (F'ig. 3). Defeat substitution. After introduction into the cage of an aggressive conspecific ("resident"), PTZ- and MDZtrained rats ("intruder") were pursued and attacked resulting in the display of submissive behavior; these behaviors included ultrasonic vocalizations, supine and

crouch postures. Intruder rats received ca. seven bites in 100 s across all defeat encounters. Immediately after defeat exposure, PTZ-trained rats were placed into the operant chambers and often remained in a crouch posture for the first minutes of the experimental session. After the first defeat, saline occa-

1 O0

(pre-defeat)

.f~

o PTZ • +DZP

/

o Prz (po~t-def~ot) • +DzP

80

~

/ ,7

1

_

:

,~/"

,,7 /7"

",,7 /

60

/

//

40

//

/:

20 /

t

L

S

P

.

2.5

// 0

.

.

.

5

,:

.

.

.

.

.

10

.

20

40

Dose ( m g / k g )

Fig. 2. Lej? panel, The percentage of PTZ lever presses after saline and PTZ for the ten sessions immediately prior to the first generalization test. Right panel The effect of PTZ on the percentage of PTZ lever presses pre- (circles) and post-defeat (squares). Open symbols represent PTZ generalization; filled symbols represent DZP (2.5 mg/kg) antagonism of PTZ. Error bars denote 1 SEM

495 100

o MDZ(pre-defeat) ~R • +FLZ(pre) / • +FLZ(post) /

80

/

I

I

100

I

I

/,I 80

'~

t~

//

60

(o c9

60 L_

N 40 a

>

// "

Dq

40 !

0_

20

20 I

0

(7 S M

......... 0.03

V 0.1

0.3

1

3

Dose ( m g / k g )

PRE-SAL

MDZ

POST-SAL

Fig. 3. Left panel. The percentage of MDZ lever presses after saline and MDZ for the ten sessions immediately prior to the first generalization test. Right panel. The effect of MDZ on the percentage of MDZ lever presses pre- (circles) and post-defeat (squares). Open symbols represent MDZ generalization;fitledsymboIs represent FLZ (10 mg/kg) antagonism of MDZ. Error bars denote 1 SEM

Fig. 4. The percentage of PTZ lever presses immediately after defeat. Open bars represent PTZ lever selection during saline-defeat sessions which preceded and followed MDZ (1 mg/kg)-defeat sessions, the filled bar represents PTZ lever selection during MDZ-defeat sessions; (see text). Asterisks indicate significant differences (P<0.05) from saline-defeat sessions, error bars denote 1 SEM

sioned PTZ-appropriate responses in most rats, i.e. 15 of 25 rats completed the session with over 80% trI'Z-lever responses, while only 3 of 25 completed saline-lever responses (Table 1). In subsequent repetitions of the defeat exposure, the overall distribution of lever presses was decreased (i.e., extinction) and the number of subjects failing to complete the reinforcement contingency within the allotted 10 min increased. In 14 subjects, midazolam (1 mg/kg) was given before saline in an attempt to antagonize the defeat-induced PTZ lever selection. In the six subjects that completed these antagonism sessions (compared with the most recent defeat and saline only sessions), PTZ lever selection returned to saline-appropriate responses [F(2,10)=82.32, P<0.01; Fig. 4].

Post-defeat generalization and antagonism

1~reat of defeat substitution. In an attempt to explore the minimal social defeat conditions that were required for PTZ lever selection, PTZ-trained rats with previous defeat experience were placed into the home cage of the aggressive resident but within a protective wire mesh cage. During these encounters, intruder rats oriented toward the resident and often remained in a crouch posture while they emitted ultrasonic vocalizations. Saline was administered after the first ultrasonic vocalization and the latency to vocalize across all threat exposures was 159 s. Similar to defeat exposure, many threatened rats responded on the PTZ-appropriate lever (Table 2). Specifically, exposure to the threat o f attack engendered greater than 80% PTZ-appropriate responding in 7 out of 25 rats, while 9 rats selected the saline-appropriate lever; the correlation between PTZ and saline-appropriate responses emitted in the defeat and threat exposures was r = 0.13. Still, PTZ lever selection among the remaining rats was higher than 50% (i.e., partial generalization).

Defeat did not alter the discriminative stimulus properties of PTZ, PTZ in the presence of DZP, or M D Z in the presence of F L Z (PTZ: Fig. 2; MDZ: Fig. 3). Cumulative doses of PTZ dose-dependently increased drug-appropriate responding; the EDs0 (mean+95%CI) was 9.54+l.61 mg/kg (we-defeat: 9.96+1.23 mg/kg). In the presence o f 2.5 mg/kg diazepam, cumulative doses of PTZ dosedependently increased PTZ lever responding; the EDs0 was 19.69+2.17 mg/kg (pre-defeat: 21.63+3.29 mg/kg). In midazolam-trained rats, pretreatment with flumazenil continued to produce a tenfold rightward shift in the M D Z dose-effect function. In the presence of 10 mg/kg flumazenil, M D Z dose-dependently increased MDZappropriate responding; the EDs0 was 1.74_+0.47 mg/kg (pre-defeat: 1.42+0.47 mg/kg).

Discussion

Exposure to defeat engendered PTZ-appropriate responding in a MDZ-reversible manner, but did not alter the discriminative stimulus properties of either PTZ or MDZ. These results suggest that an anxiety-like state is induced during defeat and the threat of defeat in many rats which is similar to the "anxiogenic" effects that are thought to occur following PTZ administration (Lal and EmmettOglesby 1983). Increased benzodiazepine binding and m R N A for cortical GABAA has been demonstrated after severe defeat in mice (Miller et al. 1987; Kang et al. 1991). However, the short-term substitution of the defeat experience for the P T Z cue was not paralleled with changes in the receptor-mediated discriminative stimulus effects of PTZ nor M D Z obtained after the defeat; this latter result argues against the proposition that defeat produces longterm alterations in GABAa-benzodiazepine receptor

496 Table 2. Effects of threat of defeat on PTZ lever selection

............... ;i:i~reat 1 .................. Threat 2 Lever responses Lever responses Subject PTZ Total % PTZ PTZ Total 1b 10 12 83 . . . . . . . 2 .... 3 2a 0 0 10 10 3 t0 12 83 t0 17 4 0 10 0 0 10 5 10 10 100 10 10 6 10 14 71 10 13 7 I0 12 83 10 16 8 10 19 53 1 11 9 10 16 63 10 l0 10 10 14 71 0 10 11 0 10 0 0 10 t2 I 1I 9 6 16 13 9 I9 47 10 15 14 l0 19 53 10 15 15 1 11 9 1 11 16 10 10 100 10 14 17 l0 15 67 I0 lI 18 10 14 71 6 16 19 10 10 100 3 13 20 I0 1i 91 10 15 21 10 14 71 10 lI 22 0 10 0 3 13 23 1 11 9 0 10 24 1 11 9 3 13 25 0 10 0 0 10 a denotes failure to complete 1 FR10 schedule after threat of defeat 1 b denotes failure to complete 1 FR10 schedule after threat of defeat 2

mechanisms as revealed through drug discrimination techniques. Ethologically relevant manipulations such as exposure to a predator or to an aggressive conspecific induced interoceptive states similar to the discriminative stimuli produced by PTZ (Vellucci et al. 1988; Gauvin and Holloway 1991). An endogenous ligand that acts via the GABAAbenzodiazepine receptor complex has been suggested and would act as an "anxiety factor" (Costa et al. 1979; Alho et al. 1985); this compound may facilitate adaptive defensive responses in animals exposed to a threatening situation. In the present experiment, exposure to a threatening and attacking conspecific resulting in defeat generalized completely to the PTZ discriminative stimulus in 15 out of 25 animals, suggesting that endogenous anxiety-like processes could comprise a component of the defeat experience. Previously, substitution of the defeat experience for the PTZ discriminative stimulus was demonstrated in 9 out of 22 rats while rats responding on the saline lever did so at a much slower rate and with a greater number of responses on the PTZ-appropriate lever (Vellucci et al. 1988). The higher proportion of rats selecting the PTZ lever in the present experiment was most likely due to the increased aggression displayed by the residents (i.e., greater number of attack bites, offensive threats and aggressive postures) in the agonistic encounters. Because defeat provides a non-drug manipulation which substituted for the PTZ discriminative stimulus, the present evidence substantiates the suggestion that an anxiety-like state is induced by PTZ (Lal and EmmettOgelsby 1983). This interpretation is further strengthened by the observation that the administration of M D Z after defeat significantly decreased selection of the PTZ lever

% PTZ 66 100 59 0 100 77 63 9 I00 0 0 38 67 67 9 7t 91 38 23 67 91 23 0 23 0

in favor of the saline lever in all rats completing the FR requirement. M D Z antagonism of the defeat-induced generalization to PTZ provides further evidence for the induction of a receptor mediated anxiety-like state due to defeat. Exposure to the threat of an aggressive defeat was employed in an effort to more thoroughly characterize the determinants of defeat-induced PTZ generalization. Evidence for stress- and anxiety-related reactions during these threat encounters has been accumulating; for example, physiological indicators such as pituitary-adrenal activity, heart rate and blood pressure, as well as the production of ultrasonic vocalizations concurrent with submissive and defensive postures, were increased while exploratory behavior was decreased (Miczek et al. 1991; Heinrichs et al. 1992; Vivian and Miczek 1993b; Meehan et al. 1994; Tornatzky and Miczek 1994). Furthermore, compounds effective in the treatment of anxiety such as diazepam, gepirone and clonidine were found to ameliorate some of these changes (Vivian and Miczek 1993a; Tornatzky and Miczek 1994). Still, the underlying visual, auditory and olfactory factors responsible for inducing anxietyqike states in threatening situations are not completely understood. In the present experiment, greater than 50 % PTZ lever selection was observed in 15 out of 25 rats after threat exposure, with 7 out of 25 rats emitting greater than 80% PTZ-appropriate responses. Although the decreased FFZ lever selection observed in threat exposures may also be influenced by habituation to the threat stimuli, these results suggest that stimuli present within the resident's cage activated endogenous anxiety-like processes similar to those observed during defeat. Finally, because there was minimal tactile contact

497 between the rats, the proposition that pain or related injury provided the discriminative stimulus responsible for P T Z lever selection proves less tenable. Considering the importance of the GABAA-benzodiazepine receptor complex in influencing anxiety (Haefely 1990), it was anticipated that changes in endogenous stress- or anxiety-like processes activated as a result of social housing or defeat might be detected as shifts in the generalization and antagonism dose-effect curves obtained before and after social housing or defeat. However, the generalization to the P T Z and M D Z discriminative stimulus was unaltered by exposure to a nonaggressive conspecific during social housing. Similarly, exposure to a threatening and attacking conspecific resulting in defeat failed to modify the discriminative stimulus properties of P T Z or M D Z and argues against long-term alterations in GABAa-benzodiazepine receptor mechanisms due to defeat. Prior experience to pharmacological and behavioral challenges modified the behavioral effects of squirrel monkeys and rats to subsequent challenges (Glowa and Barrett 1983; Vivian and Miczek 1991). Still, alterations in the generalization curves of discriminative stimuli have been difficult to demonstrate (Miczek 1991). In the current experiment, post-defeat P T Z generalization and antagonism curves were obtained from 5 to 25 sessions after the last defeat (due to the necessity to maintain reasonable response rates). Interestingly, this loss of stimulus control was not paralleled in the M D Z - t r a i n e d rats which received operant sessions not paired with an exposure to defeat. Although exposure to defeat did not alter post-defeat P T Z and M D Z antagonism curves when compared to pre-defeat curves, changes in the timing of the defeat and operant sessions m a y have influenced operant performance. In the present experiments, the discriminative stimulus produced with 0.4 mg/kg M D Z exerted control within ca. 37 sessions. Ascending doses of M D Z engendered a greater percentage of M D Z lever responses and a 10fold rightward shift in M D Z ' s discriminative stimulus effects was observed after pretreatment with flumazenil. Taken together, these latter results are consistent with the acquisition of M D Z stimulus control and the mediation of its effects at the benzodiazepine receptor site (Garcha et al. 1985; Evans and Johanson 1989). The discriminative stimulus associated with 20 mg/kg P T Z exerted control within ca. 50 sessions. Ascending doses of P T Z produced an increasingly greater percentage of P T Z lever choices and the P T Z discriminative stimulus was maintained for the duration of the cumulative dosing period (i.e., 60 rain). At no time during P T Z generalization or antagonism tests was seizure-related behavior noted; these results are consistent with previous reports concerning P T Z discrimination including the acquisition and time course of P T Z discrimination (Lal and Emmett-Oglesby 1983). A functional, non-benzodiazepine receptor mediated antagonism o f P T Z ' s discriminative stimulus effects was demonstrated as the benzodiazepine receptor agonist diazepam, but not the antagonist flumazenil (up to 80 mg/kg - data not shown), produced a rightward shift in P T Z ' s dose-effect function. In fact, although P T Z modifies GABAergic neurotrans-

mission, it does not do so via the benzodiazepine receptor as the P T Z binding site is distinct from the benzodiazepine site (Haefely 1988); although flumazenil binds to the benzodiazepine receptor, it exerts little or no effect on G A B A (Haefely 1985). Thus, diazepam antagonism of the P T Z discriminative stimulus m a y be related to the functionally opposite effects of the two compounds at the GABAa-benzodiazepine receptor complex. In conclusion, exposure to a threatening and attacking conspecific resulting in defeat produces an anxietylike state which can be reversed by the anxiolytic midazolam. It is possible that the intensity and timing of the social experience and other "anxiogenic" stimuli, such as odor cues from a previous defeat, contribute to the magnitude of generalization to the P T Z discriminative stimulus. Future studies need to determine the necessary features important in the generalization from the P T Z discriminative stimulus to a non-drug endogenous anxiety-like state.

Acknowledgements. This research was supported by USPHS research grants AA 05233 and DA 02632. The constructive comments by Dr. J.O. Valentine are gratefully acknowledged.

References A1ho H, Costa E, Ferrero P, Fujimoto M, Cosenza-Murphy D, Guidotti A (1985) Diazepam-binding inhibitor: a neuropeptide located in selected neuronal populations of rat brain. Science 229 : 179 182 Andrews JS, Stephens DN (1990) Drug discrimination models in anxiety and depression. Pharmacol Ther 47 : 267-280 Ator NA, Griffiths RR (1989) Differential generalization to pentobarbital in rats trained to discriminate lorazepam, chlor-' diazepoxide, diazepam, or triazolam. Psychopharmacology 98 : 20-30 Ator NA, Cook JM, Griffiths RR (1989) Drug discrimination in pentylenetetrazot-trained baboons: generalization to buspirone and b-carboline-3-carboxylic acid ethyl ester but not lorazepam or pentobarbital. Drug Dev Res 16 : 257-267 Costa E, Rodbard D, Pert CB (1979) Is the benzodiazepine receptor coupled to a chloride ion channel? Nature 277:315 317 DeVry J, Slangen J L (1986) Effects of training dose on discrimination and cross-generalization of chlordiazepoxide, pentoba rbital and ethanol in the rat. Psychopharmacology 88 : 341-345 Emmett-Oglesby MW, Spencer D, Lewis M, Elmesallamy F, Lal H (1983) Anxiogenic aspects of diazepam withdrawal can be detected in animals. Eur J Pharmacol 92:127-130 Emmett-Oglesby MW, Harris CM, Lane JD, Lal H (1984) Withdrawal from morphine generalizes to a pentylenetetrazol stimulus. Neuropeptides 5 : 37-40 Evans SM, Johanson CE (1989) Discriminative stimulus properties of midazolam in the pigeon. J Pharmacol Exp Ther 248 : 29-38 Fokkema DS, Smit K, van der Gugten J, Koolhaas JM (i988) A coherent pattern among social behavior, blood pressure, corticosterone and catecholamine measures in individual male rats. Physiol Behav 42 : 485-489 Garcha HS, Rose IC, Stolerman IP (1985) Midazolam cue in rats: generalization tests with anxiolytic and other drugs. Psychopharmacology 87 : 233237 Gauvin DV, Holloway FA (1991) Cross-generalization between an ecologically relevant stimulus and a pentylenetetrazol-discriminative cue. PharmacoI Biochem Behav 39 : 521-523 Glowa JR, Barrett JE (1983) Drug history modifies the behavioral effects of pentobarbital. Science 220 : 333 335

498 Greenblatt DJ, Shader RI, Divoli M, Harmatz JS (1981) Benzodiazepines: a summary of pharmacokinetic properties. Br J Clin Pharmacol 11 : 1ls-16s Haefely W (1985) Pharmacology of benzodiazepine antagonists. Pharmacopsychiatry 18 : 163-166 Haefely W (1988) Benzodiazepines. Int Anesthesiol Clin 26: 26~272 Haefely W (1990) The GABAA-benzodiazepine receptor: biology and pharmacology. In: Burrows GD, Roth M, Noyes R (eds) Handbook of anxiety, Vol. 3: The neurobiology of anxiety. Elsevier, Amsterdam, pp 165-188 Harris CM, Emmett-Oglesby MW, Robinson NG, Lal H (1986) Withdrawal from chronic nicotine substitutes partially for the interoceptive stimulus produced by pentylenetetrazole (PTZ). Psychopharmacology 90 : 85-89 Harris CM, Benjamin D, Lal H (1987) Anxiety-like subjective effects of ethanol antagonist Ro15-4513 demonstrated in pentylenetetrazole discrimination. Neuropharmacology 26:1545---1547 Heinrichs SC, Merto Pich E, Miczek KA, Britton KT, Koob GF (1992) Corticotropin-releasing factor antagonist reduces emotionality in socially defeated rats via direct netucotropic action. Brain Res 581 : 190-197 Kang 1, Thompson ML, Heller J, Miller LG (1991) Persistent elevation in GABAA receptor subunit mRNAs following social stress. Brain Res Bull 26:809-812 Lal H, Emmett-Oglesby MW (1983) Behavioral analogues of anxiety-Animal models. Neuropharmacol 22 : 1423 1441 Lal H, Shearman GT (1982) Attenuation of chemically induced anxiogenic stimuli as a novel method for evaluating anxiolytic drugs: a comparison of clobazam with other benzodiazepines. Drug Dev Res 1 : 127-134 Lal H, Harris CM, Benjamin D, Springfield AC, Bhadra S, EmmettOglesby MW (1988) Characterization of pentytenetetrazot-like interoceptive stimulus produced by ethanol withdrawal. J Pharmacol Exp Ther 24:508-518 Meehan WP, Tornatzky W, Miczek KA (1994) Blood pressure via telemetry during social confrontations in rats: effects of clonidine. Physiol Behav (submitted) Miczek KA (1979) A new test for aggression in rats without aversire stimulation: differential effects of d-amphetamine and cocaine. Psychopharmacology 60 : 253-259

Miczek KA (1991) Tolerance to the analgesic, but not discriminative stimulus effects of morphine after brief social defeat in rats. Psychopharmacology 104:181 186 Miczek KA, Thompson ML, Tornatzky W (1991) Subordinate animals: behavioral and physiological adaptations and opioid tolerance. In: Brown M, Koob G, Rivier C (eds) Neurobiology of stress. Marcel Dekker, New York, pp 323-357 Miller LG, Thompson MI., Greenblatt D J, Deutsch SI, Shader RI (1987) Rapid increase in brain benzodiazepine receptor binding following defeat stress in mice. Brain Res 414 : 395 400 Rodin EA, Calhoun HD (1970) Metrazol in a °'normal" volunteer population. J Nerv Ment, Dis 150 : 438-450 Tornatzky W, Miczek KA (1993) Long-term impairment of autonomic circadian rhythms after brief intermittent social stress. Physiol Behav 53 : 983.993 Tornatzky W, Miczek KA (1994) Behavioral and autonomic responses to intermittent social stress: differential protection by clonidine and metroprolol. Psychopharmacotogy (in press) Vellucci SV, Martin PJ, Everitt BJ (1988) The discriminative stimulus produced by pentylenetetrazoh effects of systemic anxiolytics and anxiogenics, aggressive defeat and midazolam or muscimol infused into the amygdala. J Psychopharmacol 2: 8093 Vivian JA, Miczek KA (t991) Ultrasounds during morphine withdrawal in rats. Psychopharmacology 104:187-193 Vivian JA, Miczek KA (1993a) Diazepam and gepirone selectively attenuate either 20 32 kHz or 32-64 kHz ultrasonic vocalizations during aggressive encounters. Psychopharmacology I12: 66-73 Vivian JA, Miczek KA (1993b) Morphine attenuates ultrasonic vocalization during agonistic encounters in adult male rats. Psychopharmacology 111 : 367--375 Wenger GR (t980) Cumulative dose-response curves in behavioral pharmacology. Pharmacol Biochem Behav 13 : 647 651 Wood DM, Laraby PR, Lal H (1989) A pentylenetetrazoMike stimulus during cocaine withdrawal: blockade by diazepam and not haloperidol. Drug Dev Res 16 : 269-276