Psychopharmacology (1998) 139 : 364–375

© Springer-Verlag 1998

O R I G I NA L I N V E S T I G AT I O N

Je¤rey A. Vivian · Klaus A. Miczek

Effects of l and d opioid agonists and antagonists on affective vocal and reflexive pain responses during social stress in rats

Received: 23 September 1997 / Final version: 15 January 1998

Abstract The present experiments evaluated the inßuence of intraventricular l and d opioid receptors on a¤ective vocal and reßexive responses to aversive stimuli in socially inexperienced, as well as defensive and submissive responses in defeated, adult male LongEvans rats. Defeat stress consisted of: (1) an aggressive confrontation in which the experimental intruder rat exhibited escape, defensive and submissive behaviors [i.e., upright, supine postures and ultrasonic vocalizations (USV)], and subsequently, (2) protection from the resident stimulus rat with a wire mesh screen for 10–20 min. Defeat stress was immediately followed by an experimental session with tactile startle (20 psi). The l opioid receptor agonists morphine (0.1–0.6 µg ICV) and [D-Ala2-N-Me-Phe4-Gly5-ol]-enkephalin (DAMGO; 0.01–0.3 µg ICV), and the d opioid receptor agonist [D-Pen2,5]-enkephalin (DPDPE; 10–100 µg ICV) dose-dependently decreased startle-induced USV and increased tail-ßick latencies in socially inexperienced and defeated rats. Of greater interest, morphine, DAMGO and DPDPE increased the occurrence of the submissive crouch posture, and defeated rats were more

sensitive than socially inexperienced rats to the startleinduced USV-suppressive and antinociceptive e¤ects of morphine and DPDPE. The antinociceptive e¤ects of DAMGO were likewise obtained at lower doses in defeated rats. Finally, the USV-suppressive e¤ects of morphine and DAMGO were reversed with the l receptor antagonist naltrexone (0.1 mg / kg IP), but the USV-suppressive e¤ects produced by DPDPE were not reversed with the d receptor antagonist naltrindole (1 mg / kg IP). These results conÞrm l, but not d opioid receptor activation as signiÞcant in a¤ective vocal, passive-submissive behavior, as well as reßexive antinociception. Furthermore, similar to previous studies with restraint and electric shock stress, the facilitation of l opioid e¤ects on vocal responses and antinociception is consistent with the proposal that defeat stress activated endogenous opioid mechanisms. Key words A¤ect · Aggression · Antinociception · Opioids · Rat · Stress · Vocalization

Introduction J.A. Vivian1 · K.A. Miczek Department of Psychology, Tufts University, Medford, Massachusetts, USA J.A. Vivian · K.A. Miczek Department of Psychiatry, Tufts University, Boston, Massachusetts, USA J.A. Vivian · K.A. Miczek Department of Pharmacology, Tufts University, Boston, Massachusetts, USA K.A. Miczek (*) Research Building, 490 Boston Avenue, Medford, MA 02155, USA e-mail : [email protected], Fax : +1-617-627-3939 Present address : Department of Pharmacology, University of Michigan Medical School, 1301 MSRB III, Ann Arbor, MI 48109-0632, USA 1

Social conßict has been proposed to alter a¤ective expressions which are modulated by endogenous opioid mechanisms (Panksepp et al. 1980), which in turn, modulate many aspects of the behavioral and reßexive responses in defensive and submissive rodents. In mice, defeat in an aggressive encounter induced an analgesia which was naloxone-reversible and morphine crosstolerant (Miczek et al. 1982), and these Þndings were conÞrmed in several mouse strains (Rodgers and Hendrie 1983; Külling et al. 1988; Siegfried and Frischknecht 1988; Thompson et al. 1988). In rats, naloxone reversed the antinociceptive response induced by exposure to a cat (Lichtman and Fanselow 1990), as well as the behavioral immobility (“freezing”) observed after exposure to shock (Fanselow and Bolles 1979; Fanselow and Baackes 1982).

365

Like their role in antinociception, opioid mechanisms play an integral role in the defensive and submissive repertoire of rodents. When defeated, rats refrain from locomotor and exploratory behaviors (Nabeshima et al. 1985; Miczek et al. 1990,1991a), and freezing or crouching is one of the Þrst behavioral responses exhibited by the submissive rat (but see Archer 1973). Morphine dose-dependently increased the duration of the crouch posture, decreased locomotor (walking and rearing) and grooming behavior (Vivian and Miczek 1993a). These behavioral e¤ects were naltrexone-reversible and were demonstrated in female as well as male rats (Haney and Miczek 1994). Three lines of evidence for a heightened a¤ective response during defeat have been demonstrated. (1) In rats, submissive responses are correlated with autonomic responses evident during reactions to stress. Activation of the hypothalamic-pituitary-adrenal (HPA) axis resulting in immediate increases in core temperature and blood pressure, as well as increases in circulating levels of ACTH, corticosterone and b-endorphin were observed in submissive intruders during agonistic confrontations (Schuurman 1980; Fokkema and Koolhaas 1985; Tornatzky and Miczek 1993, 1994; Meehan et al. 1995; reviewed in Miczek et al. 1991a). (2) In rats defeated by an aggressive conspeciÞc and treated with saline, pentylenetetrazole (PTZ) lever selection was greater than 80 % (Vivian et al. 1994) and reversed after midazolam. PTZ lever selection was not restricted to defeat-related behavior but was observed in other defensive contexts such as after exposure to a cat (Gauvin and Holloway 1991). Additionally, defeat immediately followed by plus maze exposure decreased open arm entries and time spent on the open arms (Heinrichs et al. 1992; Merlo Pich et al. 1993). In each of these contexts, defeat was interpreted to increase anxiety-like behaviors in rats. These results, using methods sensitive to the e¤ects of anxiolytic drugs, are consistent with the proposal that changes in a¤ective state are a component of defensive responding. Finally, (3) it has been proposed that ultrasonic vocalizations (USV) may serve as an indicator of a¤ective state (Cuomo et al. 1988; Miczek et al. 1991b, 1995), and USV are emitted in distinctive contexts, comprising maternal separation (Insel and Winslow 1991), aggression (Sales 1972), exposure to a predator (Shepherd et al. 1992), electric shock (van der Poel et al. 1989), startling stimuli (Kaltwasser 1990; Vivian and Miczek 1994) and drug withdrawal states (Vivian and Miczek 1991; Miczek and Vivian 1993; Barros and Miczek 1996; Mutschler and Miczek, 1998a,b). Not surprisingly, USV are sensitive to the e¤ects of opioids, benzodiazepines and serotonin anxiolytics (Vivian and Miczek 1993a,b; Haney and Miczek 1994; Vivian et al. 1997). In rat pups, l and d opioids suppressed separation-induced USV at doses which often reduced locomotor behavior (Gardner 1985; Barr 1992). USVsuppressive e¤ects were also demonstrated in adult rats

exposed to shock (Tonoue et al. 1986). In defeated female rats, the d opioid receptor agonist DPDPE decreased USV at doses that selectively increased crouch postures and antinociception (Haney and Miczek 1995). The current experiment focuses on opioid modulation of a¤ective, defensive, submissive and reßexive responding including vocal behavior, postures and antinociception in response to aggressive and startling stimuli. SpeciÞc objectives of the current experiment are : (1) to determine if USV emitted in startling and aggressive contexts are sensitive to l and d opioid agonist and antagonist manipulations, and (2) to determine the role of l and d opioid mechanisms in the augmented antinociception demonstrated after defeat, i.e., do l and d opioid receptor agonists produce antinociception at lower doses in defeated rats?

Materials and methods Subjects Experimentally naive male hooded Long-Evans rats (Charles River, Wilmington, Mass., USA) weighing 280–425 g were housed individually in clear polycarbonate cages (48 × 27 × 20 cm3) with stainless steel lids and sawdust bedding. Additional male Long-Evans rats weighing 500–650 g served as stimulus males (Residents) for aggressive encounters and were housed with female Long-Evans rats (250–400 g) in 46 × 46 × 71 cm3 stainless steel cages with sawdust bedding. All rats had free access to Purina rodent chow pellets and water in an environmentally controlled vivarium (12 / 12 h light / dark cycle, 21 ± 1°C, 30–40 % humidity).

Apparatus Startle Startle stimuli and responses were controlled using an SR-LAB startle response system (San Diego Instruments) connected to an IBMcompatible computer. Acoustic stimuli were presented through a speaker located approximately 15 cm above a clear acrylic cylinder (L : 20 cm, D : 8 cm) containing the subject, while tactile stimuli were presented through a 0.4 cm copper tube positioned approximately 3 cm above the subject. Responses to startle stimuli were measured with an accelerometer located on the underside of the cylinder which converted the rat’s movement into voltage changes. The startle apparatus was enclosed in a sound attenuating chamber.

Audio recordings USV were detected with a 0.64 cm condenser microphone (Bruel and Kjaer 4135) connected to a preampliÞer (Bruel and Kjaer 2633), bandpass Þlter (Krohn-Hite 3550) and measuring ampliÞer (Bruel and Kjaer 2610) which provided a ßat frequency response between 20 and 100 kHz. During startle sessions, the ampliÞer output was monitored with : (1) an oscilloscope (Goldstar OS-9020 A), and (2) a customdesigned MacIntosh-based automated signal-detection system. This hardware / software system digitized and frequency Þltered (bandpass : 15–40 kHz) ampliÞer output in order to determine the onset

366 and o¤set of each sound pulse to provide the rate and duration of USV during a startle session. During defeat sessions, the ampliÞer output was monitored with an oscilloscope and recorded with an instrumentation recorder (Hewlett-Packard 3968 A). Audio signals were recorded onto tapes (Maxell UD35–90N); USV were analyzed via tape playback at 1 / 4 speed through an ampliÞer into headphones (Sony MDR 103). A trained listener depressed keys in response to audible signals heard through the headphones and concurrently displayed on a spectrum analyzer (Tektronix 5L4N). USV rate and duration were summarized with customized software (Princeton Economics).

Video recordings During aggressive encounters, the experimental setting was illuminated with an infrared light source. Behavior was recorded with an infrared sensitive camera (Canon Ci20R) in conjunction with a VHS video recorder (Zenith VR3300) and time code generator (Skotel TCF-80N). The rate and duration of each behavioral event were analyzed on a monitor (NEC PM1971A) by observers trained to depress keys speciÞc to each event (Miczek 1982). The behavioral events included social, motor and exploratory behaviors : walk, rear, inactivity, ano-genital contact and nasal orientation; defensive behaviors : escape and upright postures; submissive behaviors : crouch and supine postures. Behavioral events followed the operational deÞnitions of Miczek (1979).

Locomotor behavior Locomotor activity of the subjects was measured within the home cage with an infrared motion detector (Columbus Instruments Opto-Varimex Mini). Fifteen infrared beams detected subject movement (deÞned as the interruption of a new beam).

Antinociception The response latencies to a thermal pain stimulus were determined with the tail-ßick method (d’Amour and Smith 1941). The intensity of the thermal stimulus was calibrated to provide a baseline response latency of 2 s (range : 1.5–2.6 s) and a maximum latency of 6 s was allowed to prevent tissue damage (Dewey and Harris 1975).

surface of the rat] were presented with an average interstimulus interval (ISI) of 30 s (range : 25–35 s). During the session (5 min habituation, 9 min startle presentation), a continuous 60 dB background noise was present. Startle amplitude was deÞned as the maximal accelerometer voltage (Vmax) measured during a 200-ms recording window concurrent with startle presentation.

Matching test Tail-ßick responses were determined, followed immediately by a motor test in which rats were returned to their home cage, which was placed into the infrared motion detector. After a 5-min habituation period, locomotor activity was measured for 5 min. Subsequently, rats were placed into the startle chamber. The purpose of these matching sessions was twofold : to assign the rats to groups with similar startle amplitudes (Vmax) and to determine the presence of USV. Rats failing to vocalize were excluded from the remainder of the experiment.

Surgery Within 48 h of the matching sessions, rats were anesthetized with ketamine (100 mg / kg IP), xylazine HCl (8 mg / kg IM) and placed into a stereotaxic frame (Kopf Instruments); the incisor bar was set to [3.3 mm (from intra-aural zero). Unilateral guide cannulae (23 ga, 8 mm) were positioned above one of the lateral ventricles (0.8 mm posterior to bregma, 1.6 mm lateral from midline, 1.5 mm ventral from dura; Paxinos and Watson 1986) and cemented into place. An occluder (30 ga, 8 mm) was inserted into the guide cannula to maintain patency. Rats were allowed a minimum of 1 week recovery from surgery before experimentation. During this time, subjects were handled daily and evaluated for accurate cannula placement by intraventricular infusions of angiotensin II (25 ng / 5 µl; Appelbaum and Holtzman 1985). Rats which failed to drink 5 ml of water in 15 min were discontinued from further study.

Startle test

Rats were exposed to matching startle stimuli (see below), surgically implanted with cannulae within 72 h, allowed to recover for at least 1 week, and subsequently exposed to tail-ßick and startle stimuli. Rats were divided into two groups : socially inexperienced and defeated. Socially inexperienced rats were exposed to tail-ßick and startle stimuli and a within-subject design was employed (i.e., all rats received all the treatments in 48-h intervals in a randomized design). Defeated rats received similar tail-ßick and startle stimuli as socially inexperienced rats; in addition, they were defeated in an agonistic encounter with an aggressive conspeciÞc (see below). In defeated subjects, a between-subjects design was employed; rats were administered an opioid at only one dose (Fig. 1).

After drug infusion in socially inexperienced rats, the rat was returned to the home cage for a motor test in which ambulatory measurements were taken during the last 5 min of the injectionstartle test interval. Subsequently, the rat was placed into the startle chamber. Tail-ßick responses were obtained prior to drug infusion and startle testing (after an appropriate injection-test interval had elapsed). For the purpose of the social defeat stress manipulation, the experimental rat (Intruder) was placed into the 46 × 46 × 71 cm3 home cage of an aggressive stimulus male rat (Resident) until deÞnitive signs of submission were observed (Attack Exposure). Submissive signs included audible squeals and USV, upright, crouch and supine postures. Immediately after the supine posture was displayed for 5 uninterrupted seconds, the intruder was placed within a 18 × 18 × 33 cm3 protective cage for 15–25 min (Threat of Attack Exposure), depending on the drug administered. During the last 5 min of this interval, video and audio recordings were taken for 5 and 2 min, respectively. Subsequently, the rat was placed into the startle chamber. Tail-ßick responses were obtained prior to drug infusions and startle testing.

Startle stimuli and measurement

Histology

After a 5-min habituation period, 18 tactile startle stimuli [50 ms burst of air (20 psi) directed through a copper tube to the dorsal

Rats were injected with cresyl violet through the cannula into the lateral ventricle and overdosed with sodium pentobarbital

Procedure

367 Fig. 1 Upper panel : timeline of events for the experiment (all components). Lower panel : timeline of events for a test session. TF Tail-ßick, ICV inj. ICV injection. See text for details

(50 mg / kg IP). The brains were removed, cut coronally and examined for staining throughout the ventricular system. In the event that dye was not detected, the subject was excluded from the analysis.

Drug administration Morphine sulfate (NIDA; 0.1, 0.3, 0.6, 1, 3, 6 µg ICV dissolved in 0.9 % sodium chloride), [D-Ala2-N-Me-Phe4-Gly5-ol]-enkephalin (DAMGO; Sigma; 0.01, 0.03, 0.06, 0.1, 0.3, 1, 3 µg ICV dissolved in 0.9 % sodium chloride) and [D-Pen2,5]-enkephalin (DPDPE; Sigma; 10, 30, 60, 100 µg ICV dissolved in distilled water) were administered 25, 15, and 25 min prior to the startle test, respectively. Microinfusions were administered through 30 ga stainless steel injectors extending 0.5–1.5 mm beyond the tip of the guide cannula in a volume of 5 µl, and at a rate of 1 µl per 10 s, except for DPDPE (100 µg) which was administered in a volume of 10 µl. Injector cannulae were left in place for 1 min after the termination of the infusion. In antagonism studies, naltrexone hydrochloride (NIDA; 0.1 mg / kg IP dissolved in 0.9 % sodium chloride) or naltrindole hydrochloride (Sigma; 1 mg / kg IP dissolved in distilled water) were administered 5 and 20 min prior to agonist administration, respectively.

Data analysis In socially inexperienced rats, mean USV rate and duration (per min), startle amplitude (Vmax), and locomotor activity were analyzed with individual one factor (Dose) within-subjects ANOVAs. Tail-ßick latencies were converted to percent maximum possible e¤ect ( % MPE) using the formula : % MPE = 100 × (post-drug latency – baseline latency) / (maximum latency – baseline latency) %MPEs were analyzed with a one factor (Dose) within-subjects ANOVA. When signiÞcant e¤ects were present, post-hoc contrasts were performed. In defeated subjects, startle-induced USV rate and duration (per min), startle amplitude (Vmax) and tail-ßick %MPEs were analyzed with individual one-factor (Dose) between-subjects ANOVAs. Data from the audio and video recordings were summarized and the mean rate and duration of the behavioral measures were analyzed with

individual one factor (Dose) between-subjects ANOVAs. When signiÞcant e¤ects were detected, post-hoc Dunnett analyses were performed. In some cases, audio and video defeat data did not meet the assumptions necessary for ANOVA. In these cases, the nonparametric Kruskal-Wallis and Wilcoxon tests were used. Alpha was 0.05, two-tailed. In order to compare opioid e¤ects in naive and defeated rats, ED50s (the dose at which a 50 % response was observed) and 95 % conÞdence limits were calculated for startle-induced USV and %MPE. Data were converted into probits with the LitchÞeld and Wilcoxon correction (Tallarida and Murray 1981).

Results Opioid e¤ects in socially inexperienced rats Rats readily emitted 0.3–3.5 s, 20–32 kHz USV in response to 20 psi tactile startle stimuli. USV were rarely emitted prior to the Þrst startle stimulus, but often began during the Þrst three startle stimuli. In control conditions, startle-induced USV were emitted at a rate of 30.8 ± 3.5 calls per minute (mean ± SEM) with a total duration of 24.5 ± 2.3 s per minute. The e¤ects of the tested opioids on the rate of USV paralleled those on duration; only rate measures will be portrayed. Morphine Morphine dose-dependently decreased the rate of startle-induced USV, and increased the latency to remove the tail from a thermal stimulus, beginning at the 1 µg dose [F(4,40) = 14.67, P < 0.05; F(4,40) = 85.41, P < 0.05, respectively; Figs 2,3]. Morphine produced no systematic e¤ects on the startle reßex, but decreased locomotor activity at 6 µg [F(4,40)=4.81, P < 0.05; Table 1A].

368

Fig. 2 Upper panel : the e¤ects of morphine on startle-induced 20–32 kHz USV in socially inexperienced (open circles) and socially defeated ( Þlled circles) rats. Middle panel : the e¤ects of DAMGO on startle-induced 20–32 kHz USV in socially inexperienced (open circles) and socially defeated ( Þlled circles) rats. Lower panel : the e¤ects of DPDPE on startle-induced 20–32 kHz USV in socially inexperienced (open circles) and socially defeated ( Þlled circles) rats. For all Þgures, lines depict Þrst order regression equations, error bars denote 1 SEM and asterisks indicate signiÞcant di¤erences (P < 0.05) from control

Fig. 3 Upper panel : the e¤ects of morphine on tail-ßick latencies in socially inexperienced (open circles) and socially defeated ( Þlled circles) rats. Middle panel : the e¤ects of DAMGO on tail-ßick latencies in socially inexperienced (open circles) and socially defeated ( Þlled circles) rats. Lower panel : the e¤ects of DPDPE on tail-ßick latencies in socially inexperienced (open circles) and socially defeated ( Þlled circles) rats. For all Þgures, lines depict Þrst order regression equations, error bars denote 1 SEM, and asterisks indicate signiÞcant di¤erences (P < 0.05) from control

DAMGO

DPDPE

DAMGO dose-dependently decreased the rate of startle-induced USV, beginning at the 0.06 µg dose [F(4,28) = 10.81, P < 0.05]. DAMGO dose-dependently increased tail-ßick latencies, beginning at the 0.03 µg dose [F(4,28) = 68.93, P < 0.05; Figs 2,3]. DAMGO produced systematic e¤ects neither on locomotor activity nor on the startle reßex (Table 1B). Pretreatment with the opioid antagonist naltrexone (0.1 mg / kg) reduced the potency of DAMGO to suppress startle-induced USV and to produce antinociception. In the presence of naltrexone, DAMGO dose-dependently decreased startle-induced USV, and increased tail-ßick latencies, at 1 µg and beginning at 0.3 µg, respectively [F(4,36) = 4.39, P < 0.05; F(4,36) = 41.47, P < 0.05, respectively].

DPDPE dose-dependently decreased the rate of startle-induced USV, beginning at the 60 µg dose [F(3,21) = 12.74, P < 0.05]. DPDPE dose-dependently increased tail-ßick latencies, beginning at 30 µg [F(3,21) = 74.36, P < 0.05; Figs. 2, 3]. DPDPE produced biphasic e¤ects on locomotor activity. At the lowest dose tested (30 µg), DPDPE increased ambulatory behavior (173 % of control), while the highest dose resulted in little locomotion; at this dose, rats were often cataleptic [F(3,21) = 6.33, P < 0.05]. DPDPE produced no systematic e¤ects on the startle reßex (Table 1C). Pretreatment with the selective d antagonist naltrindole (1 mg / kg) had no e¤ect on the baseline level of startle-induced USV but enhanced the USV-suppressive e¤ects produced by DPDPE. After pretreat-

369 Table 1 E¤ects of l and d opioids on startle reßexes and locomotor behavior in socially inexperienced rats

(Dose µg ICV) A. Morphine Startle Locomotion

0 1000 ± 292 193 ± 32

0.3 892 ± 212 196 ± 24

1 764 ± 163 119 ± 32

3 608 ± 85 112 ± 37

6 922 ± 120 55 ± 29*

B. DAMGO Startle Locomotion

0 971 ± 161 254 ± 39

0.03 850 ± 124 388 ± 148

0.06 784 ± 83 205 ± 33

0.1 818 ± 99 230 ± 91

0.3 978 ± 116 328 ± 92

– –

+Naltrexone (0.1 mg / kg IP) Startle 772 ± 45 Locomotion 232 ± 41

– –

– –

750 ± 88 223 ± 39

733 ± 63 181 ± 29

651 ± 88 775 ± 89 159 ± 51 126 ± 66

C. DPDPE Startle Locomotion

– –

30 689 ± 87 372 ± 85

60 540 ± 118 142 ± 60

100 595 ± 128 22 ± 11*

730 ± 151 383 ± 68*

601 ± 82 598 ± 131 458 ± 123* 166 ± 59

0 705 ± 112 215 ± 42

+Naltrindole (1 mg / kg IP) Startle 866 ± 130 Locomotion 198 ± 26

10

861 ± 121 517 ± 61*

0.6

– –

1

Units : Startle, Vmax; Locomotion, beams / min. Values indicate the mean ± 1 SEM. Asterisks indicate signiÞcant di¤erences (P < 0.05) from control

ment with naltrindole, DPDPE dose-dependently decreased startle-induced USV, beginning at the 30 µg dose [F(4,48) = 13.51, P < 0.05]. Conversely, the antinociceptive e¤ects produced by DPDPE were reversed by naltrindole : only a 29 % increase in tailßick latencies was observed at the highest dose tested [100 µg; F(4,48) = 8.30, P < 0.05]. In the presence of naltrindole, DPDPE increased locomotor behavior at the 10,30 and 60 µg doses [F(4,36) = 5.26, P < 0.05]. Opioid e¤ects in defeated rats Stimulus and response characteristics of the attack Morphine, DAMGO and DPDPE administered to the intruder immediately prior to the Attack Encounter had no signiÞcant e¤ects on the stimulus resident’s attack behavior. The latency to attack (48.3 ± 5.3 s), frequency of bites (4.4 ± 0.3 bites per min of the encounter) and exposure duration (95.4 ± 7.4 s) were consistent across drug and dose conditions. Similarly, drug treatment condition did not alter the intruder rats’ behaviors. Intruder rats engaged in defensive (11.2 ± 1.3 s per min of the encounter) and submissive (23.3 ± 2.2 s per min of the encounter) behaviors more than 50 % of the time during Attack Encounters. There were two exceptions: the highest dose of DAMGO (0.3 µg) increased the duration of crouch postures [F(4,32) = 17.76, P < 0.05], and the highest dose of DPDPE (60 µg) induced a cataleptic-like reaction [F(3,33) = 9.19, P < 0.05]. Finally, audible squeals, 20–32 kHz and 32–64 kHz defeat-induced USV were also emitted and were not systematically or di¤erentially a¤ected by treatment condition. Morphine. Morphine decreased the rate of defeatinduced 20–32 kHz and 32–64 kHz USV [1 µg : v2(4) =

9.74, P < 0.05; 0.3–1 µg : v2(4) = 16.68, P < 0.05, respectively]. Morphine dose-dependently increased the duration of the crouch posture, beginning at the 0.6 µg dose [F(4,40) = 2.98, P < 0.05]. In contrast, the duration of time spent in exploratory and nonsocial behaviors, particularly walking and inactivity, was decreased [F(4,40) = 2.70, P < 0.05; F(4,40) = 4.05, P < 0.05, respectively; Table 2A]. DAMGO. DAMGO appeared to decrease (maximum e¤ect : 48 % reduction) the rate of defeat-induced USV (NS), and increased the crouch posture (NS); in general, DAMGO produced inconsistent e¤ects on motor, defensive, social and nonsocial behaviors (Table 2B). DPDPE. DPDPE decreased the rate of defeat-induced 20–32 kHz and 32–64 kHz USV at 30 and 60 µg [v2(3) = 17.92, P < 0.05; v2(3) = 11.12, P < 0.05, respectively]. At low doses (10, 30 µg), the behavioral e¤ects produced by DPDPE alone and after naltrindole pretreatment were inconsistent. At high doses (60, 100 µg), DPDPE produced a cataleptic-like response which was even more prominent following naltrindole administration. In general, DPDPE decreased most behaviors at doses higher than 30 µg, including rearing, anogenital and nasal contact, inactivity and autogrooming; concurrent with these decreases were increases in crouch and / or lie postures (Table 2C). Behavior after the threat of attack Defeat experience immediately prior to the acoustic startle session markedly increased the rate of 20–32 kHz USV during startle sessions. In control conditions, defeated subjects vocalized almost twice as much as socially inexperienced subjects [F(1,43) = 35.65,

370 Table 2 E¤ects of defeated rats

l and d opioids on ultrasonic vocalizations, locomotor, defensive, submissive behavior and startle reßexes in socially (Dose µg ICV)

A. Morphine Defeat-induced 20–32 kHz USV Defeat-induced 32–64 kHz USV Startle Walk Crouch

0 0.1 0.3 0.6 1 32.3 ± 8.2 38.4 ± 8.3 34.4 ± 8.6 23.4 ± 6.2 9.9 ± 4.9* 22.0 ± 7.0 48.1 ± 24.4 5.8 ± 2.4* 1.1 ± 0.7* 0.13 ± 0.1* 655 ± 126 704 ± 78 594 ± 106 723 ± 88 963 ± 169 9.4 ± 1.9 7.9 ± 2.1 7.8 ± 2.4 2.3 ± 1.2* 3.7 ± 0.7* 24.4 ± 6.9 28.5 ± 6.9 34.8 ± 6.7 47.9 ± 5.4* 47.7 ± 3.5*

B. DAMGO Defeat-induced 20–32 kHz USV Defeat-induced 32–64 kHz USV Startle Walk Crouch

0 21.7 ± 9.9 7.7 ± 5.8 794 ± 66 8.4 ± 2.6 18.4 ± 7.9

0.01 30.2 ± 9.2 16.4 ± 8.5 684 ± 76 11.5 ± 2.7 21.4 ± 6.6

0.03 0.06 22.6 ± 6.3 12.6 ± 8.5 0.7 ± 0.2 3.3 ± 2.3 1291 ± 175* 699 ± 66 6.3 ± 1.9 11.3 ± 2.8 29.2 ± 4.0 34.3 ± 5.8

0.1 11.8 ± 7.2 1.2 ± 0.8 890 ± 162 8.9 ± 1.7 38.1 ± 6.5

– – – – –

+Naltrexone (0.1 mg / kg IP) Defeat-induced 20–32 kHz USV Defeat-induced 32–64 kHz USV Startle Walk Crouch

24.7 ± 6.8 11.3 ± 6.8 837 ± 169 11.0 ± 2.6 23.3 ± 8.8

– – – – –

– – – – –

– – – – –

22.8 ± 5.5 11.5 ± 9.4 683 ± 51 7.6 ± 2.8 31.2 ± 6.2

C. DPDPE Defeat-induced 20–32 kHz USV Defeat-induced 32–64 kHz USV Startle Walk Crouch

0 32.3 ± 8.2 22.0 ± 7.0 655 ± 126 9.4 ± 2.0 24.4 ± 6.1

10 30 31.9 ± 10.7 3.8 ± 3.7* 44.4 ± 20.4 1.6 ± 1.4* 1005 ± 138 537 ± 70 9.9 ± 1.9 10.2 ± 3.7 29.9 ±3.3 29.5 ± 7.1

60 1.1 ± 1.1* 0.4 ± 0.2* 753 ± 57 11.0 ± 4.3 29.0 ± 8.2

+ Naltrindole (1 mg/kg IP) Defeat-induced 20–32 kHz USV Defeat-induced 32–64 kHz USV Startle Walk Crouch

23.6 ± 9.6 11.1 ± 5.8 691 ± 98 9.0 ± 2.5 20.0 ± 4.5

– – – – –

0 ± 0* 0 ± 0* 0.9 ± 0.9* 0.2 ± 0.2* 601 ± 109 779 ± 115 11.5 ± 3.8 3.3 ± 2.6 27.8 ± 6.6 47.6 ± 8.4

– – – – –

5.4 ± 4.4* 7.5 ± 6.3 888 ± 171 15.2 ± 3.7 21.1 ± 8.2

0.3

– – – – –

1

– – – – –

3

2.5 ± 2.5* 0 ± 0* 0 ± 0* 0.3 ± 0.2* 661 ± 127 605 ± 83 12.5 ± 3.5 10.0 ± 5.0 34.3 ± 8.1 37.3 ± 9.4

100 – – – – –

Units : USV, calls / min; Startle, Vmax, Locomotion, beams / min; Walk, Crouch, s / min. Values indicate the mean ± 1 SEM. Asterisks indicate signiÞcant di¤erences (P < 0.05) from control

P < 0.05]. This increase in the rate of startle-induced USV was observed in the habituation period, in which subjects began to vocalize prior to the Þrst startle stimulus. Moreover, individual call durations were decreased from 0.8 s / call in socially inexperienced subjects to 0.58 s / call in defeated subjects. Finally, defeat experience had no e¤ect on the baseline tail-ßick or startle reßexes. Morphine. Morphine dose-dependently decreased the rate of startle-induced 20–32 kHz USV, increased tailßick latencies, beginning at the 0.3 µg dose [F(4,40) = 25.27, P < 0.05; F(4,40) = 6.67, P < 0.05; Figs 2,3]. Across the range of doses tested, morphine produced no systematic e¤ects on the startle reßex. DAMGO. DAMGO dose-dependently decreased startle-induced USV, beginning at the 0.01 µg dose [F(4,33) = 14.51, P < 0.05]. DAMGO dose-dependently increased tail-ßick latencies, beginning at the 0.03 µg dose [F(4,33) = 40.72, P < 0.05; Figs. 2,3]. DAMGO was largely ine¤ective in altering the startle reßex, although there was a modest increase in startle

amplitude at the 0.03 µg dose [F(4,33) = 4.16, P < 0.05; Table 2B]. Pretreatment with naltrexone (0.1 mg / kg) increased the rate of startle-induced 20–32 kHz USV [F(1,12) = 11.12, P < 0.05] and reduced the potency of DAMGO to suppress USV and the tail-ßick reßex. Subsequent to naltrexone administration, DAMGO dose-dependently decreased startle-induced USV, beginning at the 1 µg dose [F(3,29) = 18.97, P < 0.05]. In the presence of naltrexone, DAMGO dose-dependently decreased tail-ßick latencies, beginning at the 0.3 µg dose [F(3,29) = 58.40, P < 0.05]; DAMGO did not alter the startle reßex. DPDPE. DPDPE dose-dependently decreased the rate of startle-induced USV and increased tail-ßick latencies, beginning at the 30 and 60 µg doses, respectively [F(3,33) = 13.49, P < 0.05; F(3,33) = 84.68, P < 0.05, respectively]. Similar to morphine and DAMGO e¤ects on startle amplitude, DPDPE produced no systematic e¤ects on the startle reßex. Pretreatment with naltrindole did not alter the startle-induced USV rate-suppressive e¤ects, and reversed

371 Table 3 Comparison of the attenuation of startle-induced 20–32 kHz USV and tail-ßick antinociception in socially inexperienced and defeated rats

Agonist

Measure

Socially inexperienced ED50 (95 % CL)

Morphine DAMGO DPDPE

Socially defeated Range

ED50 (95 % CL)

USV Antinociception

1.49 (0.68–3.26) 2.26 (1.39–3.66)

USV Antinociception

0.06 (0.04–0.1) –

0.06–0.1

0.11 (0.06–0.20) 0.025 (0.001–0.058)*

USV Antinociception

84.6 (62.9–113.8) –

60

37.8(27.7–51.7)* –

Range

0.78 (0.52–1.18)* 1.05 (0.65–1.70)*

30–60

ED50s are expressed in µg. When ED50s were not estimable, the range column indicates the dose range at which an 84 % or greater suppression of the measurement was observed. Asterisks indicate signiÞcant di¤erences (P < 0.05) from socially inexperienced rats

the antinociceptive e¤ects, produced by DPDPE. In the presence of naltrindole, DPDPE dose-dependently decreased startle-induced USV and increased tailßick latencies, beginning at the 30 and 60 µg doses, respectively [F(3,29) = 25.53, P < 0.05; F(3,29) = 14.17, P < 0.05, respectively]. Comparison of opioid e¤ects in socially inexperienced and defeated rats Defeat augmented the antinociceptive e¤ects of all the opioid agonists. In defeated rats, lower doses of morphine, DAMGO and DPDPE were necessary to suppress the tail-ßick reßex in comparison to socially inexperienced rats. Interestingly, defeat did not produce symmetrical changes on the e¤ects of opioid agonists on startle-induced USV: while morphine and DPDPE suppression of startle-induced USV were enhanced in defeated rats, DAMGO e¤ects in defeated rats remained unchanged (Figs. 1, 2, Table 3).

Discussion The results of the current experiment support the proposal that opioid mechanisms are involved in a¤ective, defensive, submissive and reßexive responding (Miczek et al. 1991a, 1995). The activation of l, but not d, opioid receptors selectively decreased defeat- and startleinduced USV, increased the occurrence of crouch postures while restricting the behavioral repertoire, and increased the latency to respond to painful thermal stimuli during aggressive contexts in male rats. Defeatinduced augmentation of opioid suppressive e¤ects on the tail-ßick reßex were evident : defeated rats were more sensitive to the antinociceptive e¤ects of morphine, DAMGO and DPDPE. Similarly, although the e¤ects produced by DAMGO may have been too shortlived, defeated subjects were also more sensitive to the

suppressive e¤ects of morphine and DPDPE on vocal responses elicited in the startle chamber. Opioid mediation of a¤ective vocal responses Naloxone reversed the suppression of startle-induced 20–32 kHz and defeat-induced 20–32 kHz and 32–64 kHz USV produced by morphine and DAMGO; this further implicates l opioid receptors in a¤ective vocal behavior (Vivian and Miczek 1993a; Haney and Miczek 1994). In addition to their USV-suppressive e¤ects in neonatal (Carden et al. 1991) and adult rats (Haney and Miczek 1994), l opioid receptor agonists decreased separation-induced vocalizations in neonatal chicks, guinea pigs, dogs (for review see Panksepp et al. 1980), a¤ective defense vocalizations in the cat (Brutus et al. 1988, 1989), and separation-induced vocalizations in juvenile rhesus monkeys (Kalin et al. 1988). Additionally, naloxone, although less selective for the l receptor than DAMGO, has been shown to increase shock-induced vocalizations in adult guinea pigs (Herman and Panksepp 1981) and isolation calls in adult squirrel monkeys (Harris and Newman 1988). Investigations of d opioid e¤ects on vocalizations are less frequent (but see Shaikh et al. 1991; Siegel et al. 1995), and evidence is equivocal as to their involvement in USV. d opioid receptor agonists, including DPDPE, met- and leu-enkephalin, decreased separation-induced and shock-elicited USV in neonatal and adult rats, respectively (Tonoue et al. 1986; Kehoe et al. 1990). Nonetheless, as shown previously in female rats, defeat-induced 20–32 kHz and 32–64 kHz USV, and startle-induced 20–32 kHz USV were not very sensitive to the e¤ects of DPDPE, since very high doses were necessary for their suppression (Haney and Miczek 1995). Even if the e¤ects of DPDPE were mediated through the d opioid receptor, behavioral selectivity was not evident, as a cataleptic-like behavior was prominent at the highest doses employed. Perhaps the most persuasive argument against d opioid mediation

372

of the suppression of USV was the lack of naltrindole antagonism. Naltrindole has been well established as a d selective antagonist (Portoghese 1993), and its e¤ectiveness was demonstrated by its reinstatement of the tail-ßick reßex in the current experiment. Opioid mediation of tail-ßick antinociception Antinociception is the sine qua non of opioid pharmacology and the tail-ßick reßex has proven to be quite sensitive to the e¤ects of l and d opioids. In the current experiment, morphine DAMGO and DPDPE produced antinociceptive e¤ects; these latter e¤ects were reversed following the administration of naltrexone and naltrindole, respectively. These e¤ects are consistent with the many demonstrations which implicate l and d opioids and their e¤ects on centrifugally mediated nociception, including tests involving thermal pain stimuli (for reviews see Yaksh 1993; Porreca and Burks 1993). Opioid mediation of locomotor, defensive and submissive behavior Opioid mediation of submissive behavior has been proposed and supported with the results from the present experiment (Miczek et al. 1982, 1986, 1991a; Rodgers and Randall 1988; Siegfried et al. 1990). SpeciÞcally, morphine and DAMGO produced doserelated increases in crouch behavior concurrent with a restriction of other behaviors, and implicates the involvement of l receptor systems in submissive behavior. Previously, morphine and DPDPE decreased locomotor behavior and increased crouch posture (Vivian and Miczek 1993a; Haney and Miczek 1994,1995), yet the cataleptic response produced by higher doses of DPDPE makes it di¦cult to discern d receptor involvement in submissive behavior. Augmentation of opioid e¤ects on vocal responses and antinociception As a result of being exposed to defeat, the startleinduced 20–32 kHz USV-suppressive dose-response e¤ects of morphine and DPDPE were shifted to the left. Similarly, defeated rats were more sensitive to the antinociceptive e¤ects of morphine, DAMGO and DPDPE. These results are consistent with the proposal that stress can augment many of the e¤ects produced by morphine and DAMGO. Restraint- or cold exposure-stressed rats were more sensitive to the antinociception and hyperthermia produced by morphine and DAMGO (Appelbaum and Holtzman 1984,1985,1986; Calcagnetti et al. 1992). Although DAMGO e¤ects on startle-induced USV did not di¤er between defeated and naive rats, this result may be solely due to the short

duration of action of DAMGO (i.e., maximal e¤ects within 20 min; Calcagnetti et al. 1992). Additionally, the rate of startle-induced USV may have inßuenced the opioid e¤ects. Defeated rats were vocalizing at a greater rate than inexperienced rats, and dissimilar drug e¤ects, including opioid e¤ects, have been demonstrated depending on the rate and topography of the ongoing behavior (i.e. rate dependency; Dews 1955; McMillan and Morse 1967). Nonetheless, an attractive possibility for the augmentation of opioid e¤ects observed in the present experiment is that defeat stress induced the release of endogenous opioid peptides, particularly l opioid peptides, which potentiated the e¤ects of the exogenously administered opioids (Miczek et al. 1982). The suggestion of an endogenous opioid release due to speciÞc types of environmental stress conÞrms previous interpretations. Naloxone-reversible and morphine cross-tolerant stress induced analgesia after footshock (Akil et al. 1976; Drugan et al. 1981) and forced swims (Bodnar 1993) in rats are robust. These e¤ects are correlated with decreased opioid binding (Christie et al. 1981; Seeger et al. 1984; Nabeshima et al. 1985). In addition to footshock stress and cold water swim stressors, electroconvulsive shock (Hong et al. 1979), 90 min water deprivation (Stein et al. 1992), exposure to formalin and thermal stimuli (Kuraishi et al. 1984), and isolation stress (Petkov et al. 1985) have been found to increase measures indicative of endogenous opioid activity, including decreased opioid binding and increased enkephalin immunoreactivity and enkephalin mRNA in rats (Olson et al. 1993). With respect to defeat stress, an opioid-mediated small elevation in tail-ßick latencies was reported for intruders immediately after attack by a lactating rat (Rodgers et al. 1983), and increased plasma b-endorphin immunoreactivity has been demonstrated in submissive rats (Dijkstra et al. 1992). Interestingly, an increased sensitivity to the antinociceptive e¤ects produced by morphine followed by a longer lasting decreased sensitivity to morphine has been reported in rats (Miczek et al. 1991a). Finally, defeat increased metenkephalin production in the periaqueductal gray (PAG) in rats (Kream et al., submitted). These results provide the basis for the proposal that opioid peptides are released during speciÞc stressful events, in this case, exposure to an attacking and threatening conspeciÞc which leads to defeat. Evidence for speciÞc opioid receptor e¤ects and their mediation of stress-induced or -augmented analgesia is accumulating. Decreased DAMGO binding in the septum (Stein et al. 1992), and decreased diprenorphine binding in PAG, reticular formation and midline thalamic nuclei (Seeger et al. 1984) indicated l receptor involvement in intermittent footshock stress in rats. Similarly, forced swims decreased diprenorphine binding in PAG, reticular formation and thalamus in rats (Seeger et al. 1984). d receptor involvement in stressinduced and -augmented analgesia has also been

373

demonstrated. Increased met- and leu-enkephalin levels were demonstrated in the hypothalamus, nucleus accumbens, septum and amygdala after electroconvulsive shock in rats (Hong et al. 1979), and decreased levels were observed after a 21-day regimen of footshock stress (McGivern et al. 1983). Decreased d receptor number (Bmax) and a¦nity (Kd) for met-enkephalin in frontal cortex, hippocampus and PAG was also reported after social isolation in rats (Petkov et al. 1985). The d opioid receptor antagonist 16-Me cyprenorphine (M80) reversed conditional antinociception by 75 % in the formalin test (Fanselow et al. 1989) and naltrindole reversed tail-shock antinociception by 65 % in the tail-ßick test in rats (Watkins et al. 1992). Additionally, prolonged cold water swim stress induced analgesia was attenuated by naltrindole and ICI 154,129 in rats (Kamei et al. 1993; Hart et al. 1983). Conclusion The suppression of USV and the tail-ßick reßex, and the increase in crouch postures by morphine and DAMGO conÞrms the inßuence of l opioid systems on a¤ective vocal expressions, submissive behavior and antinociceptive reßexes in adult rats (van der Poel et al. 1989; Vivian and Miczek 1993a; Haney and Miczek 1994). Morphine and DAMGO suppressed startle-induced 20–32 kHz USV at doses which were roughly comparable to those which suppress the tailßick reßex. However, these naltrexone-reversible e¤ects on USV and nociception were not related to an overall behavioral suppression, as locomotor behavior was only modestly a¤ected during DAMGO treatment, while the startle reßex remained unchanged. On the other hand, d opioid involvement in USV remains equivocal due to the high doses of DPDPE necessary to produce its e¤ects, the lack of naltrindole antagonism, and the manifestation of behavioral side-e¤ects at higher doses (see also Haney and Miczek 1995). Finally, defeated rats were more sensitive than socially inexperienced rats to the suppressive e¤ects produced by morphine, DAMGO and DPDPE on a¤ective vocalizations and reßexive tail-ßick nociception; these latter results are consistent with the release of endogenous opioid peptides during defeat.

References Akil H, Madden J, Patrick JD, Barchas JD (1976) Stress-induced increase in endogenous opiate peptides : Concurrent analgesia and its partial reversal by naloxone. In : Kosterlitz HW (ed) Opiates and endogenous opioid peptides. North Holland, Amsterdam, pp 63–70 Appelbaum BD, Holtzman SF (1984) Characterization of stressinduced potentiation of opioid e¤ects in the rat. J Pharmacol Exp Ther 231 : 555–565

Appelbaum BD, Holtzman SF (1985) Stress-induced changes in the analgesic and thermic e¤ects of morphine administered centrally. Brain Res 358 : 303–308 Appelbaum BD, Holtzman SF (1986) Stress-induced changes in the analgesic and thermic e¤ects of opioid peptides in the rat. Brain Res 377 : 330–336 Archer J (1973) Tests for emotionality in rats and mice : a review. Anim Behav 21 : 205–235 Barr G (1992) Behavioral e¤ects of opiates during development. In: Miller M (ed) Development of the central nervous system: e¤ects of alcohol and opiates. Wiley Liss, New York, pp 221–254 Barros HMT, Miczek KA (1996) Withdrawal from oral cocaine in rats : ultrasonic vocalizations and tactile startle. Psychopharmacology 125 : 379–384 Bodnar RJ (1993) Measurement of stress-induced analgesia. In : Conn PM (ed) Methods in neurosciences, vol 14. Academic Press, San Diego, pp 281–293 Brutus M, Zuabi S, Siegel A (1988) E¤ects of D-Ala2-Met5enkephalinamide microinjections placed into the bed nucleus of the stria terminalis upon a¤ective defense behavior in the cat. Brain Res 473 : 147–152 Brutus M, Zuabi S, Siegel A (1989) Microinjections of D-Ala2-Met5enkephalinamide placed into nucleus accumbens suppress hypothalamically elicited hissing in the cat. Exp Neurol 104 : 55–61 Calcagnetti DJ, StaÞnsky JL, Crisp T (1992) A single stress exposure potentiates analgesia induced by intrathecally administered DAGO. Brain Res 592 : 305–309 Carden SE, Barr GA, Hofer MA (1991) Di¤erential e¤ects of speciÞc opioid receptor agonists on rat pup isolation calls. Dev Brain Res 62 : 7–22 Christie MJ, Chesher GB, Bird KD (1981) The correlation between swim-stress induced antinociception and (3H)-leu-enkephalin binding to brain homogenates in mice. Pharmacol Biochem Behav 15 : 853–857 Cuomo V, Cagiano R, DeSalvia MA, Maselli MA, Renna G, Racagni G (1988) Ultrasonic vocalization in response to unavoidable aversive stimuli in rats : e¤ects of benzodiazepines. Life Sci 43 : 485– 491 d’Amour F, Smith D (1941) A method for determining loss of pain sensation. J Pharmacol Exp Ther 72 : 74–79 Dewey WL, Harris LS (1975) The tail-ßick test. In : Ehrenpress S, Neidle A (eds) Methods in narcotic research. Dekker, New York, pp 101–109 Dews PB (1955) Studies on behavior I. Di¤erential sensitivity to pentobarbital of pecking performance in pigeons depending on the schedule of reward. J Pharmacol Exp Ther 113 : 393– 401 Dijkstra H, Tilders FJH, Hiehle MA, Smelik PG (1992) Hormonal reactions to Þghting in rat colonies : prolactin rises during defence, not during o¤ense. Physiol Behav 51 : 961–968 Drugan RC, Grau JW, Maier SF, Madden J, Barchas JD (1981) Cross tolerance between morphine and the long-term analgesic reaction to inescapable shock. Pharmacol Biochem Behav 14 : 677–682 Fanselow MS, Baackes MP (1982) Conditioned fear-induced opiate analgesia on the formalin test : evidence for two aversive motivational systems. Learn Motivat 13 : 200–221 Fanselow MS, Bolles RC (1979) Naloxone and shock-elicited freezing in the rat. J Comp Physiol Psychol 93 : 736–744 Fanselow MS, Calcagnetti DJ, Helmstetter FJ (1989) Delta opioid antagonist, 15-Me cyprenorphine, selectively attenuates conditioned fear- and DPDPE-induced analgesia on the formalin test. Pharmacol Biochem Behav 32 : 469–473 Fokkema DS, Koolhaas JM (1985) Acute and conditioned blood pressure changes in relation to social and psychosocial stimuli in rats. Physiol Behav 34 : 33–38 Gardner CR (1985) Distress vocalizations in rat pups. A simple screening method for anxiolytic drugs. J Pharmacol Meth 14 : 181–187

374 Gauvin DV, Holloway FA (1991) Cross-generalization between an ecologically relevant stimulus and a pentylenetetrazol-discriminative cue. Pharmacol Biochem Behav 39 : 521–523 Haney M, Miczek KA (1994) Ultrasound emitted by female rats during agonistic interactions : e¤ects of morphine and naltrexone. Psychopharmacology 114 : 441– 448 Haney M, Miczek KA (1995) Delta opioid receptors : reßexive, defensive and vocal a¤ective responses in female rats. Psychopharmacology 121 : 204–212 Harris JC, Newman JD (1988) Combined opiate / adrenergic receptor blockade enhances squirrel monkey vocalization. Pharmacol Biochem Behav 31 : 223–226 Hart SL, Slusarczyk H, Smith TW (1983) The involvement of opioid delta-receptors in stress induced antinociception in mice. Eur J Pharmacol 95 : 283–285 Heinrichs SC, Merlo Pich E, Miczek KA, Britton KT, Koob GF (1992) Corticotropin-releasing factor antagonist reduces emotionality in socially defeated rats via direct neurotropic action. Brain Res 581 : 190–197 Herman BH, Panksepp J (1981) Ascending endorphin inhibition of distress vocalization. Science 211 : 1060–1062 Hong JS, Gillin JC, Yang HYT, Costa E (1979) Repeated electroconvulsive shocks and the brain content of endorphins. Brain Res 177 : 273–278 Insel TR, Winslow JT (1991) Rat pup ultrasonic vocalizations : an ethologically relevant behaviour responsive to anxiolytics. In : Olivier B, Mos J, Slangen JL (eds) Animal models in psychopharmacology. Birkhauser, Basel, pp 15–36 Kalin NH, Shelton SE, Barksdale CM (1988) Opiate modulation of separation-induced distress in non-human primates. Brain Res 440 : 285–292 Kaltwasser MT (1990) Startle-inducing acoustic stimuli evoke ultrasonic vocalization in the rat. Physiol Behav 48 : 13–17 Kamei J, Hitosugi H, Misawa M, Nagase H, Kasuya Y (1993) Delta-opioid receptor-mediated forced swimming stressinduced antinociception in the formalin test. Psychopharmacology 113 : 15–18 Kehoe P, Boylan C, Shoemaker W (1990) Di¤erential e¤ects of speciÞc endogenous opioid systems on a¤ective behaviors in neonatal rats. Soc Neurosci Abstr 16 : 211 Külling P, Frischknecht HR, Pasi A, Waser PG, Siegfried B (1988) Social conßict-induced changes in nociception and beta-endorphin-like immunoreactivity in pituitary and discrete brain areas of C57BL / 6 and DBA / 2 mice. Brain Res 450 : 237–246 Kuraishi Y, Sugimoto S, Hamada T, Kayanoki Y, Takagi H (1984) Noxious stimuli and met-enkephalin release from nucleus reticularis gigantocellularis. Brain Res Bull 12 : 123–127 Lichtman AH, Fanselow MS (1990) Cats produce analgesia in rats on the tail-ßick test : naltrexone sensitivity is determined by the nociceptive test stimulus. Brain Res 533 : 91–94 McGivern RF, Mousa S, Couri D, Berntson GC (1983) Prolonged intermittent footshock stress decreases met and leu enkephalin levels in brain with concomitant decreases in pain threshold. Life Sci 33 : 47–54 McMillan DE, Morse WH (1967) Some e¤ects of morphine and morphine antagonists on schedule-controlled behavior. J Pharmacol Exp Ther 157 : 175–184 Meehan WP, Tornatzky W, Miczek KA (1995) Blood pressure via telemetry during social confrontations in rats : e¤ects of clonidine. Physiol Behav 58 : 81–88 Merlo Pich E, Heinrichs SC, Rivier C, Miczek KA, Fisher DA, Koob GF (1993) Blockade of pituitary-adrenal axis activation induced by peripheral immunoneutralization of corticotropinreleasing factor does not a¤ect the behavioral response to social defeat stress in rats. Psychoneuroendocrinology 18 : 495–507 Miczek KA (1979) A new test for aggression in rats without aversive stimulation : di¤erential e¤ects of d-amphetamine and cocaine. Psychopharmacology 60 : 253–259

Miczek KA (1982) Ethological analysis of drug action on aggression, defense and defeat. In : Spiegelstein MY, Levy A (eds) Behavioral models and the analysis of drug action. Elsevier, Amsterdam, pp 225–239 Miczek KA, Vivian JA (1993) Automatic quantiÞcation of withdrawal from 5-day diazepam in rats : ultrasonic distress vocalizations and hyperreßexia to acoustic startle stimuli. Psychopharmacology 110 : 379–382 Miczek KA, Thompson ML, Shuster L (1982) Opioid-like analgesia in defeated mice. Science 215 : 1520–1522 Miczek KA, Thompson ML, Shuster L (1986) Analgesia following defeat in an aggressive encounter : development of tolerance and changes in opioid receptors. In : Kelly DD (ed) Stressinduced analgesia, vol 467. New York Academy of Science, New York, pp 14–29 Miczek KA, Thompson ML, Tornatzky W (1990) Short and long term physiological and neurochemical adaptations to social conßict. In : Puglisi-Allegra S, Oliviero A (eds) Psychobiology of Stress. Kluwer, Amsterdam, pp 15–30 Miczek KA, Thompson ML, Tornatzky W (1991a) Subordinate animals : behavioral and physiological adaptations and opioid tolerance. In : Brown M, Koob G, Rivier C (eds) Neurobiology of stress. Dekker, New York, pp 323–357 Miczek KA, Tornatzky W, Vivian JA (1991b) Ethology and neuropharmacology : rodent ultrasounds. In : Olivier B, Mos J, Slangen JL (eds) Animal Models in Psychopharmacology. Birkhäuser, Basel, pp 409–427 Miczek KA, Weerts EM, Vivian JA, Barros HM (1995) Aggression, anxiety and vocalizations in animals : GABAA and 5–HT anxiolytics. Psychopharmacology 121 : 38–56 Mutschler NH, Miczek KA (1998a) Withdrawal from IV cocaine “binges” in rats : ultrasonic distress calls and startle. Psychopharmacology (in press) Mutschler NH, Miczek KA (1998b) Withdrawal from a selfadministered or non-contingent cocaine binge : di¤erences in ultrasonic distress calls in rats. Psychopharmacology (in press) Nabeshima T, Matsuno K, Kamei H, Noda Y, Kameama T (1985) Electric footshock-induced changes in behavior and opioid receptor function. Pharmacol Biochem Behav 23 : 769–775 Olson GA, Olson RD, Kastin AJ (1993) Endogenous opiates : 1992. Peptides 14 : 1339–1378 Panksepp J, Herman BH, Vilberg T, Bishop T, DeEskinazi FG (1980) Endogenous opioids and social behavior. Neurosci Biobehav Rev 4 : 473– 487 Paxinos G, Watson C (1986) The rat brain in stereotaxic coordinates. Academic Press, Sydney Petkov VV, Konstantinov E, Grachovska T (1985) Changes in brain opiate receptors in rats with isolation syndrome. Pharmacol Res Commun 17 : 575–584 Porreca F, Burks TF (1993) Supraspinal opioid receptors in antinociception. In : Herz A (ed) Opioids II. Springer, Berlin, pp 21–51 Portoghese PS (1993) Selective nonpeptide opioid antagonists. In : Herz A (ed), Opioids I. Springer, Berlin, pp 279–304 Rodgers RJ, Hendrie CA (1983) Social conßict activates statusdependent endogenous analgesic and hyperalgesic mechanisms in male mice : e¤ects of naloxone on nociception and behaviour. Physiol Behav 30 : 775–780 Rodgers RJ, Randall JI (1988) Environmentally induced analgesia : situational factors, mechanisms and signiÞcance. In : Rodgers RJ, Cooper SJ (eds) Endorphins, opiates and behavioural processes. Wiley, Chichester, pp 107–142 Rodgers RJ, Hendrie CA, Water AJ (1983) Naloxone partially antagonizes post-encounter analgesia and enhances defensive responding in male rats exposed to attack from lactating conspeciÞcs. Physiol Behav 30 : 781–786 Sales GD (1972) Ultrasound and aggressive behaviour in rats and other small mammals. Anim Behav 20 : 88–100

375 Schuurman T (1980) Hormonal correlates of agonistic behavior in adult male rats. In: McConnel PS, Boer GJ, Romijn HJ, van der Poll NE (eds) Progress in brain research: adaptive capabilities of the nervous system, vol 53. Elsevier, Amsterdam, pp 415–420 Seeger TF, Sforzo GA, Pert CB, Pert A (1984) In vivo autoradiography : visualization of stress-induced changes in opiate receptor occupancy in the rat brain. Brain Res 305 : 303–311 Shaikh MB, Lu CL, Siegel A (1991) A¤ective defense behavior elicited from the feline midbrain periaqueductal gray is regulated by µ and d opioid receptors. Brain Res 557 : 344–348 Shepherd JC, Blanchard DC, Weiss SM, Rodgers RJ, Blanchard RJ (1992) Morphine attenuates antipredator ultrasonic vocalizations in mixed-sex rat colonies. Pharmacol Biochem Behav 41 : 551–558 Siegel A, Schubert K, Shaikh MG (1995) Neurochemical mechanisms underlying amygdaloid modulation of aggressive behavior in the cat. Aggress Behav 21 : 49–62 Siegfried B, Frischknecht HR (1988) Naltrexone-reversible pain suppression in the isolated attacking mouse. Behav Neural Biol 50 : 354–360 Siegfried B, Frischknecht HR, Nunes de Souza RL (1990) An ethological model for the study of activation and interaction of pain, memory and defensive systems in the attacked mouse. Role of endogenous opioids. Neurosci Biobehav Rev 14 : 481–490 Stein EA, Hiller JM, Simon EJ (1992) E¤ects of stress on opioid receptor binding in the rat central nervous system. Neuroscience 51 : 683–690 Tallarida RJ, Murray RB (1981) Manual of pharmacological calculations with computer programs. Springer, New York Thompson ML, Miczek KA, Noda K, Shuster L, Kumar MSA (1988) Analgesia in defeated mice : evidence for mediation via central rather than pituitary or adrenal endogenous opioid peptides. Pharmacol Biochem Behav 29 : 451–456 Tonoue T, Ashida Y, Makino H, Hata H (1986) Inhibition of shock-elicited ultrasonic vocalization by opioid peptides in the rat : A psychotropic e¤ect. Psychoneuroendocrinology 12 : 177–184

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 : di¤erential protection by clonidine and metropolol. Psychopharmacology 116 : 346–356 van der Poel AM, Noach EJK, Miczek KA (1989) Temporal patteming of ultrasonic distress calls in the adult rat: e¤ects of morphine and benzodiazepine. Psychopharmacology 97 : 147–148 Vivian JA, Miczek KA (1991) Ultrasounds during morphine withdrawal in rats. Psychopharmacology 104 : 187–193 Vivian JA, Miczek KA (1993a) Morphine attenuates ultrasonic vocalizations during agonistic encounters in adult rats. Psychopharmacology 111 : 367–375 Vivian JA, Miczek KA (1993b) Diazepam and gepirone selectively attenuate either 20–32 kHz or 32–64 kHz ultrasonic vocalizations during aggressive encounters. Psychopharmacology 112 : 66–73 Vivian JA, Miczek KA (1994) Diazepam withdrawal : e¤ects of diazepam and gepirone on acoustic startle-induced 22 kHz ultrasonic vocalizations. Psychopharmacology 114 : 101–108 Vivian JA, Weerts EM, Miczek KA (1994) Defeat engenders pentylenetetrazole-appropriate responding in rats : antagonism by midazolam. Psychopharmacology 116 : 491–498 Vivian JA, Barros HMT, Manitiu A, Miczek KA (1997) Ultrasonic vocalizations in rat pups : modulation at the c-aminobutyric acidA receptor complex and the neurosteroid recognition site. J Pharmacol Exp Ther 282 : 318–325 Watkins LR, Wiertelak EP, Maier SR (1992) Delta opiate receptors mediate tail-shock induced antinociception at supraspinal levels. Brain Res 582 : 10–21 Yaksh TL (1993) The spinal action of opioids. In : Herz A (ed) Opioids II. Springer, Berlin, pp 53–90