Approach to a social stranger is associated with low central nervous system serotonergic responsivity in female cynomolgus monkeys (Macaca fascicularis).код для вставкиСкачать
American Journal of Primatology 61:187–194 (2003) RESEARCH ARTICLE Approach to a Social Stranger is Associated With Low Central Nervous System Serotonergic Responsivity in Female Cynomolgus Monkeys (Macaca fascicularis) STEPHEN B. MANUCK1, JAY R. KAPLAN2n, BETH A. RYMESKI3, LYNN A. FAIRBANKS4, and MARK E. WILSON5 1 Behavioral Physiology Laboratory, Department of Psychology, University of Pittsburgh, Pittsburgh, PA 2 Comparative Medicine Clinical Research Center, Wake Forest University School of Medicine, Winston-Salem, NC 3 Department of Anthropology, Wake Forest University, Winston-Salem, NC 4 Neuropsychiatric Institute, University of California-Los Angeles, Los Angeles, CA 5 Yerkes National Primate Research Center, Emory University, Atlanta, GA It is widely hypothesized that individual differences in central nervous system (CNS) serotonergic activity underlie dimensional variation in ‘‘impulsive’’ vs. ‘‘inhibited’’ social behavior in both humans and nonhuman primates. To assess relative impulsivity in a social context, a behavioral challenge involving animals’ exposure to a social stranger (termed the ‘‘Intruder Challenge’’) was recently validated in adolescent and adult male vervet monkeys (Cercopithecus aethiops sabaeus). Among these animals, monkeys that quickly approached the intruder were found to have lower cerebrospinal fluid (CSF) concentrations of the serotonin (5-HT) metabolite, 5-hydroxyindoleacetic acid, than less impulsive animals. In the present study we extended these observations to determine whether approach to a social stranger, as operationalized by the Intruder Challenge, is similarly associated with diminished CNS serotonergic function in female cynomolgus monkeys (Macaca fascicularis). Study animals were 25 adult monkeys that had been housed for 2 years in stable social groups. In each animal, the rise in plasma prolactin concentration induced by acute administration of the 5-HT agonist, fenfluramine, was used to assess ‘‘net’’ central serotonergic responsivity. When exposed later to an unfamiliar female of the same species in a catch-cage placed for 20 min within the subjects’ home enclosure, monkeys that approached to within 1 m of the intruder (median latency to approach ¼ 3 min) were found to have significantly smaller prolactin responses to fenfluramine (diminished serotonergic Contract grant sponsor: NIH; Contract grant numbers: HL 40962 and HL 65137 (SBM); HL 45666 (JRK); HD 38917 (MEW). n Correspondence to: Jay R. Kaplan, Ph.D., Department of Pathology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157-1040. E-mail: firstname.lastname@example.org Received 15 May 2003; revision accepted 15 August 2003 DOI 10.1002/ajp.10118 Published online in Wiley InterScience (www.interscience.wiley.com). r 2003 Wiley-Liss, Inc. 188 / Manuck et al. responsivity) compared to ‘‘inhibited’’ animals that failed to approach the intruder (t ¼ 2.9, df ¼ 23, Po0.009; rpb ¼ –0.51). Neither approach behavior nor the animals’ fenfluramine-induced prolactin responses covaried significantly with nondirected expressions of arousal (or anxiety) or with aggressive behaviors exhibited during testing. We conclude that in female cynomolgus monkeys, social impulsivity (vs. inhibition) correlates inversely with individual differences in CNS serotonergic activity, as assessed by neuroendocrine challenge. Am. J. Primatol. 61:187–194, 2003. r 2003 Wiley-Liss, Inc. Key words: impulsivity; serotonin; fenfluramine INTRODUCTION People imprisoned for impulse-related crimes, or who suffer from psychiatric disorders in which problems of impulse control figure prominently (e.g., borderline and antisocial personality disorders, and suicidal depression) tend to show altered central nervous system (CNS) serotonergic activity, as indicated (indirectly) by a low cerebrospinal fluid (CSF) concentration of the serotonin (5-HT) metabolite, 5-hydroxyindoleacetic acid (5-HIAA), or a blunted neuroendocrine response to pharmacologic agents that enhance serotonergic neurotransmission [e.g., Coccaro et al., 1989; Linnoila et al., 1983; Mann et al., 1995; Virkkunen et al., 1994]. Otherwise normal individuals who report on a standard personality questionnaire that they are highly impulsive also exhibit diminished central serotonergic responsivity, compared to persons who describe themselves as less impulsive [Manuck et al., 1998]. Analogous findings have been reported for nonhuman primates, in which lower cisternal CSF 5-HIAA concentrations in rhesus monkeys (Macaca mulatta) are associated with several naturally occurring behaviors of a putatively impulsive nature. These include such acts as leaping long distances in the forest canopy, risking injury by escalating aggressive encounters to fights of heightened severity, and (among adolescent males) emigrating from natal groups at an early age, when the animals are more susceptible to attack and predation [Higley et al., 1992, 1996, 1997; Kaplan et al., 1995; Mehlman et al., 1994, 1995]. Although problems associated with impulsivity often occur in a social context, standardized assessments of impulsivity rarely incorporate social cues to elicit impulsive behavior. To address this limitation, a behavioral challenge employing a salient social stimulus–an unfamiliar ‘‘intruder’’–was recently developed and validated in a large sample of adolescent and adult male vervet monkeys (Cercopithecus aethiops sabaeus) [Fairbanks, 2001]. Animals that responded impulsively to the appearance of a strange animal (intruder), as indexed by latency to approach and frequency of antagonistic interactions, were also found to have lower CSF 5-HIAA levels than less impulsive monkeys [Fairbanks et al., 2001]. Here we extend these observations to determine whether social impulsivity (operationalized by the Intruder Challenge) is similarly associated with diminished central serotonergic activity in female cynomolgus monkeys (Macaca fascicularis). In the present study we employed a standard neuropharmacologic challenge – the rise in plasma prolactin concentration induced by acute administration of the 5-HT agonist, fenfluramine – to assess in vivo CNS serotonergic responsivity (Botchin et al., 1993; Coccaro et al., 1989; Yetham & Steiner, 1993). Approach to a Stranger and Serotonin / 189 MATERIALS AND METHODS Subjects The study animals were 25 female cynomolgus monkeys that had been imported as adults from Indonesia (Institut Pertanian Bogor, Bogor, Indonesia). Although their exact ages were unknown, all of the animals had radiographic evidence of epiphyseal closure and were therefore probably 49 years of age. At the time the behavioral observations described in this report were conducted, the monkeys had been housed for 24 months in five five-member isosexual social groups of stable membership (these animals comprised the untreated control condition of an experimental study designed to evaluate estrogenic effects on bone metabolism and atherosclerosis). The animals were fed monkey chow, and groups were housed in pens measuring 2 3.2 2.5 m (with outdoor exposure). The animals’ social ranks were evaluated once each month. These determinations were based on the observation of dyadic fight episodes and were defined by the direction of fight outcomes (recorded as unambiguous fight gestures expressed by one of the participants). The animal that defeated all others in her social group was assigned rank 1, the animal that defeated all but the highest-ranking monkey was assigned rank 2, and so forth. Dominance ranks were fully transitive (as is typical of this species) and were stable over the course of study. Intruder Challenge The procedures of the Intruder Challenge employed in this study were based on the protocol described by Fairbanks et al. . At the start of each challenge, either two animals (those of social ranks 2 and 4) or three animals (of ranks 1, 3, and 5) were left in their home enclosure, and the remaining monkeys were removed to an adjacent housing area. A tarp-covered catch-cage, measuring 50 43 36 cm and containing an unfamiliar female of the same species, was then placed in the home enclosure. The tarp was removed and the behaviors of each animal were scored (by one observer per subject) as present or absent in 1min intervals over a 20-min test session. Recorded behaviors included those denoting ‘‘Approach’’ (sitting within 1 m of the intruder’s cage, touch, sniff) and ‘‘Aggression’’ (lunge, smack, or threaten, either as expressed toward the intruder or redirected to a cagemate), and nondirected expressions of arousal or ‘‘Anxiety’’ (yawn, scratch, body-shake). ‘‘Latency’’ to approach the intruder was measured as time elapsed (min) until the subject first moved within 1 m of the intruder’s cage. Testing proceeded until all 25 monkeys were evaluated, but not more than two challenges were conducted on a single day. Interobserver reliabilities for behavioral observations made on this set of animals were 40.90. Assessment of CNS Serotonergic Responsivity Fenfluramine increases serotonergic neurotransmission by causing a release of serotonin stores, reuptake inhibition, and possible activation of postsynaptic serotonin receptors [Borroni et al., 1983; Coccaro et al., 1989]. Stimulation of hypothalamic serotonin receptors promotes the pituitary release of prolactin (PRL), so the relative increase in circulating PRL concentration induced by fenfluramine provides an index of ‘‘net’’ serotonergic responsivity in the hypothalamic-pituitary axis [Yetham & Steiner, 1993]. The fenfluramine challenge employed here was based on the protocol developed by Botchin et al.  and was administered 4 months prior to the behavioral observations 190 / Manuck et al. described above. At 0830 hr on each of 2 test days scheduled 1 week apart, all animals in a social group were placed in a squeeze cage and given an intramuscular (IM) injection of either sterile saline or 4 mg/kg fenfluramine hydrochloride. The monkeys were then returned to the group home cage. One hour later, they were returned to the squeeze cage and anesthetized with ketamine hydrochloride (10 mg/kg, IM). Blood samples were collected by venipuncture 10–20 min after the animals were anesthetized, and then transferred to tubes containing EDTA and frozen at –201C for later determination of plasma PRL levels. In two of the five social groups (selected randomly), saline was injected in the first week, followed by fenfluramine in the second week. For the remaining groups, the order of injections was reversed. The animals were fasted for 12 hr prior to each procedure. PRL was measured by enzyme-linked immunoabsorbent assay (ELISA), using a commercially available kit for the assay of human PRL (Diagnostic Systems Laboratory, Webster, TX). The assay uses 25 ml of serum, and has a sensitivity of 0.52 IU/dL. Serial dilution of monkey serum (6, 12, and 25 ml) provided estimates of PRL parallel to the standard curve and, when corrected for the dilution factor, yielded comparable estimates of PRL (32.8, 36.4, and 34.8 IU/ dL, respectively). The intra-assay coefficient of variation (CV) was 8.3% at 5.5 IU/ dL (n ¼ 10), and the interassay CVs were 14.7% (n ¼ 14) and 6.6% at 1.95 and 12.3 IU/dL, respectively. Following the administration of fenfluramine, PRL concentrations averaged 54.8 (SE ¼ 4.8) IU/dL among the 25 study animals, and reflected a significant rise over corresponding saline-day PRL values (M ¼ 6.4 [1.8] IU/dL; Wilcoxon T ¼ 4.3, Po0.0001). Each monkey’s PRL response to fenfluramine was expressed as the difference in PRL concentration between the drug- and salineday injections. Across all animals, the fenfluramine-stimulated rise in PRL (DPRL[fen]) correlated strongly with PRL levels following drug administration (r ¼ 0.93, df ¼ 23, Po0.0001), but not with ‘‘baseline’’ (saline-day) PRL values (r ¼ 0.26, df ¼ 23, n.s.). The mean PRL change also did not differ by order of saline and fenfluramine injection (fenfluramine first vs. second: M ¼ 49.2 [7.2] vs. 47.8 [4.8] IU/dL; t ¼ 0.2, df ¼ 23, n.s.). RESULTS Of the 25 study animals, only 11 (44%) actually approached within 1 m of the intruder at any time during the 20-min challenge, and those that did approach tended to do so very quickly. Six approached within the initial 3 min, and an additional four approached within 8 min of first being exposed to the intruder; the median latency to approach among the 11 animals that did so was 3 min. Due to the discontinuity of distribution in approach behavior, these data were treated on analysis as a nominal variable; for convenience, animals that approached the intruder were designated as ‘‘impulsive’’ and those that did not were identified as ‘‘inhibited.’’ With respect to central serotonergic activity, the fenfluramineinduced PRL responses of impulsive monkeys were significantly smaller than those of inhibited animals (DPRL[fen]: M ¼ 37.0 [SE ¼ 6.2) and 57.2 [3.9] IU/dL, t ¼ 2.9, df ¼ 23, Po0.009). A correlation of social impulsivity (impulsive vs. inhibited) with the animals’ PRL responses to fenfluramine yielded a point-biserial coefficient (rpb) of –0.51 (df ¼ 23, Po0.009). PRL concentrations following fenfluramine administration covaried similarly with social impulsivity (rpb ¼ –0.50, df ¼ 23, Po0.01), whereas baseline (saline-day) PRL values did not (rpb ¼ –0.22, df ¼ 23, n.s.). Lastly, animals’ PRL responses (DPRL[fen]) did not Approach to a Stranger and Serotonin / 191 correlate significantly with approach latency (in min) among the 11 monkeys that actually approached the intruder (rs ¼ 0.14, df ¼ 9, n.s.). Aggressive behaviors during the Intruder Challenge also varied among animals, but were not expressed at all in 11 of the 25 monkeys. Among the remaining 14 animals (56%), the number of recorded aggressive acts varied from one to 26, with a median value of 4.0. Interestingly, whether or not the monkeys expressed any aggressive behavior during the challenge was unrelated to whether or not they approached the intruder (w21o1, n.s.). Moreover, the PRL response to fenfluramine in animals that exhibited aggression did not differ significantly from that in monkeys that showed no aggressive behavior (DPRL[fen]: M ¼ 54.7 [7.7] and 54.8 [5.3] IU/dL, t ¼ 0.01, df ¼ 23, n.s.), and among the 14 monkeys that displayed aggression, the PRL response did not covary significantly with the number of aggressive acts observed (rs ¼ –0.37, df ¼ 12, n.s.). Behaviors expressive of nondirected arousal or anxiety (e.g., yawn and scratch) were common and varied continuously from zero to 29 such acts over the 20-min challenge (M ¼ 11.6 [1.9]; median ¼ 10). These behaviors were exhibited with equal frequency by monkeys that did or did not approach the intruder (M ¼ 10.9 [2.8] and 12.1 [2.2], respectively, t ¼ 0.35, df ¼ 23, n.s.; rpb ¼ –0.07, df ¼ 23, n.s.), and among those that did or did not exhibit any aggressive behavior during the challenge (M ¼ 10.5 [2.19] and 12.9 [2.71], respectively; t ¼ 0.70, df ¼ 23, n.s.; rpb ¼ – 0.14, df ¼ 23, n.s.). Also, anxiety did not covary with the animals’ PRL responses to fenfluramine (r ¼ 0.11, df ¼ 23, n.s.). Finally, social rank (ranks 1–5) did not correlate significantly with whether or not animals approached the intruder (rpb ¼ 0.0, df ¼ 23, n.s.), displayed aggression during the challenge (rpb ¼ –0.22, df ¼ 23, n.s.), or exhibited behaviors indicative of anxiety (r ¼ –0.02, df ¼ 23, n.s.). Rank did not covary with the animals’ PRL response to fenfluramine (r ¼ 0.12, df ¼ 23, ns). DISCUSSION Social impulsivity in response to a stranger (as assessed by the Intruder Challenge) was found in this study to be associated with diminished CNS serotonergic responsivity among 25 adult female cynomolgus macaques. Specifically, impulsive monkeys (those that readily approached a novel animal (intruder)) showed a blunted PRL response to the 5-HT agonist, fenfluramine, compared to socially ‘‘inhibited’’ animals, which failed to approach the intruder at any time during the challenge. This finding closely parallels the observations of Fairbanks et al. , who reported that similarly assessed impulsivity in adolescent and adult male vervet monkeys covaried inversely with cisternal CSF 5-HIAA concentrations, an alternate measure of central serotonergic activity. Nonetheless, the monkeys observed here appeared to be more inhibited in their approach behavior than male vervets: 450% of the animals tested by Fairbanks et al.  approached the intruder within 1 min, while only two of 25 female cynomolgus monkeys in the current study approached that quickly. It is noteworthy, though, that the median latency to approach among animals that did so in this study (impulsives) was still relatively brief (3 min) and that in both studies behaviors indicative of generalized arousal or anxiety did not correlate significantly with either approach behavior or indices of serotonergic function. Fairbanks et al.  identified two correlated components of animals’ behavioral responses to the Intruder Challenge among vervet monkeys: latency to approach the intruder, and aggressive interactions (threats and assertive displays). Although the latency and aggression components both covaried 192 / Manuck et al. inversely with serotonergic function (CSF 5-HIAA) in vervets, the association between approach latency and the animals’ 5-HIAA concentrations remained significant after the correlated variation in aggressive behavior was adjusted for statistically. Moreover, because the aggression component did not similarly predict 5-HIAA levels after adjustment for the animals’ approach behavior, Fairbanks et al.  proposed that dimensional variation in ‘‘impulsivity vs. inhibition’’ comprises the key behavioral correlate of central serotonergic function (at least as assessed by the Intruder Challenge). In this view, aggressiveness (unlike impulsivity) is not a direct consequence of low serotonergic activity, but rather a category of behavior that uninhibited, impulsive individuals are simply more likely to express, depending on circumstance and motivation. For instance, among the male vervets observed by Fairbanks et al. , animals approaching an intruder within 1 min (the most impulsive monkeys) were also more likely to make threatening gestures toward the stranger, as compared to animals that approached the intruder more hesitantly or not at all, yet over 40% of the latter group were still characterized as highly aggressive (viz., having aggression ‘‘scores’’ above the sample median). Our results are consistent with those of Fairbanks et al. , inasmuch as social impulsivity covaried inversely with monkeys’ fenfluramine-induced PRL responses and did so independently of variation in the animals’ aggressive behavior. However, in contrast to the previous study, the likelihood that animals would engage in any aggressive acts during the challenge was unrelated to whether or not they approached the intruder, and was not itself associated with diminished serotonergic responsivity. Because the current experimental animals were female cynomolgus monkeys, and those in the previous investigation were male vervet monkeys, it is conceivable that in the present study impulsivity was independent of aggression, and aggressive behavior was unrelated to central serotonergic activity due to a difference in either sex or species. Lending some plausibility to a primary sex difference is the fact that we previously found fenfluramine-stimulated PRL responses in male cynomolgus monkeys (n ¼ 75) to be attenuated in animals that showed the highest rates of ‘‘overt’’ aggression when housed socially with other monkeys [Botchin et al., 1993], and, in laboratory testing (n ¼ 40), in monkeys that displayed the greatest number of aggressive gestures when they viewed slides of threatening content [Kyes et al., 1995]. Although a preliminary study of eight socially-housed female cynomolgus monkeys suggested that rates of aggressive behavior in these animals may also correlate inversely with PRL response to fenfluramine [Shively et al., 1995], this association was not replicated in a subsequent investigation involving a larger number of females (n ¼ 40) [Shively, 1998] (Shively, personal communication). Obviously, though, only a direct comparison of males and females of both species and under identical testing conditions can adequately address whether the covariation of approach (impulsive) and aggressive behaviors and the relationship between aggression and serotonergic function differ significantly by sex and/or between cynomolgus and vervet monkeys. Finally, in agreement with Fairbanks et al. , we view an uninhibited approach to a social stranger to be indicative of social impulsivity, which in conjunction with a blunted PRL response to fenfluramine is consistent with the broader human literature linking CNS serotonergic activity to psychopathologies of impulse control and normative (trait) variation in impulsivity [e.g., Coccaro et al., 1989; Linnoila et al., 1983; Virkkunen et al., 1994; Manuck et al., 1998]. This interpretation is also consistent with naturalistic observations in rhesus monkeys, which show low cisternal CSF 5-HIAA concentrations Approach to a Stranger and Serotonin / 193 to be associated with impulsive acts expressed in both social and nonsocial contexts [Higley et al., 1992, 1997; Mehlman et al., 1994]. Nonetheless, in the absence of other, correlated indices of impulsive behavior in the present study, alternative interpretations (such as social attraction) of an animal’s rapid approach to a social stranger during the Intruder Challenge cannot be fully excluded. ACKNOWLEDGMENTS The authors acknowledge the valuable assistance of Dewayne Cairnes and Melissa Ayers in collecting the data reported in this paper. REFERENCES Borroni E, Ceci A, Garattini S, Mennini R. 1983. Differences between D-fenfluramine and D-norfenfluramine in serotonin presynaptic mechanisms. J Neurochem 40: 891–903. Botchin MB, Kaplan JR, Manuck SB, Mann JJ. 1993. Low versus high prolactin response to fenfluramine challenge: marker of behavioral differences in adult male cynomolgus monkeys. Neuropsychopharmacology 9:93–99. Coccaro EF, Siever LJ, Klar HM, Maurer G, Cochrane K, Cooper TB, Mohs RC, Davis KL. 1989. Serotonergic studies in patients with affective and personality disorders. Arch Gen Psychiatry 46:587–599. Fairbanks LA. 2001. Individual differences in response to a stranger: social impulsivity as a dimension of temperament in vervet monkeys (Cercopithecus aethiops sabaeus). J Comp Psychol 115:22–28. Fairbanks LA, Melega WP, Jorgensen MJ, Kaplan JR, McGuire MT. 2001. Social impulsivity inversely associated with CSF 5-HIAA and fluoxetine exposure in vervet monkeys. Neuropsychopharmacology 24:370–378. Higley JD, Mehlman PT, Taub DM, Higley SB, Suomi SJ, Linnoila M, Vickers JH. 1992. Cerebrospinal fluid monoamine and adrenal correlates of aggression in free-ranging rhesus monkeys. Arch Gen Psychiatry 49:436–441. Higley JD, Mehlman PT, Higley SB, Fernald B, Vickers J, Lindell SG, Taub DM, Suomi SJ, Linnoila M. 1996. Excessive mortality in young free-ranging male nonhuman primates with low cerebrospinal fluid 5-hydroxyindoleacetic acid concentrations. Arch Gen Psychiatry 53: 537–542. Higley JD, Mehlman PT, Poland RE, Taub DM, Vickers J, Suomi SJ, Linnoila M. 1997. CSF testosterone and 5-HIAA correlate with different types of aggressive behaviors. Biol Psychiatry 40:1067–1082. Kaplan JR, Fontenot MB, Berard J, Manuck SB, Mann JJ. 1995. Delayed dispersal and elevated monoaminergic activity in freeranging rhesus monkeys. Am J Primatol 35:229–234. Kyes RC, Botchin MB, Kaplan JR, Manuck SB, Mann JJ. 1995. Aggression and brain serotonergic responsivity: response to slides in male macaques. Physiol Behav 57: 205–208. Linnoila M, Virkkunen M, Sheinin M, Nuutila A, Pimon R, Goodwin FK. 1983. Low cerebrospinal fluid 5-hydroxyindoleacetic acid concentration differentiates impulsive from nonimpulsive violent behavior. Life Sci 33:2609–2614. Mann JJ, McBride PA, Malone KM, DeMeo MD, Keilp J. 1995. Blunted serotonergic responsivity in depressed patients. Neuropsychopharmacology 13:53–64. Manuck SB, Flory JD, McCaffery JM, Matthews KA, Mann JJ, Muldoon MF. 1998. Aggression, impulsivity, and central nervous system serotonergic responsivity in a nonpatient sample. Neuropsychopharmacology 19:287–299. Mehlman PT, Higley JD, Faucher I, Lilly AA, Taub DM, Vickers J, Suomi SJ, Linnoila M. 1994. Low CSF 5-HIAA concentrations and severe aggression and impaired impulsive control in nonhuman primates. Am J Psychiatry 151:1485–1491. Mehlman PT, Higley JD, Faucher I, Lilly AA, Taub DM, Vickers J, Suomi SJ, Linnoila M. 1995. Correlation of CSF 5-HIAA concentration with sociality and the timing of emigration in free-ranging primates. Am J Psychiatry 152:907–913. Shively CA, Fontenot MB, Kaplan JR. 1995. Social status, behavior, and central serotonergic responsivity in female cynomolgus monkeys. Am J Psychiatry 37:333–339. 194 / Manuck et al. Shively CA. 1998. Social subordination stress, behavior, and central monoaminergic function in female cynomolgus monkeys. Biol Psychiatry 44:882–891. Virkkunen M, Rawlings R, Tokola R, Poland RE, Guidotti A, Nemeroff C, Bissette G, Kalogeras K, Karonen S-L, Linnoila M. 1994. CSF biochemistries, glucose metabo- lism, and diurnal activity rhythms in alcoholic, violent offenders, fire setters, and health volunteers. Arch Gen Psychiatry 51:20–27. Yetham LN, Steiner M. 1993. Neuroendocrine probes of serotonergic function: a critical review. Life Sci 53:447–463.