BEHAVIOURAL BRAIN RESEARCH
Behavioural Brain Research 78 (1996) 175 182
Behavioral changes in Anolis carolinensis following injection with fluoxetine
A. Wallace D e c k e l 1
University of Connecticut Health Center, MC 2103, 263 Farmington Avenue, Farmington, CT06030, USA
Received 31 March 1995; revised 31 August 1995; accepted 31 October 1995
Eight adult male lizards of the genus and species Anolis carolinensis were used in this experiment. In order to induce aggressive responding, animals were caged separately and daily underwent pairing with another male, during which aggressive responses and changes in skin color were measured. After obtaining a baseline measure of aggressive responding, animals were injected either with fluoxetine or vehicle-controls in a cross-over design. Subjects were then exposed to five more days of (non drug) pairing with the intruder male, after which they underwent a second trial with fluoxetine/vehicle. Finally, two post-drug paired-trials were obtained. Fluoxetine injection significantly reduced the aggressive responding in the males while causing the postorbital eyespot to significantly darken. Subjects also showed increased aggressivity and skin-color reactivity subsequent to the two drug trials, although it is unclear if the fluoxetine, or non-specific factors of the injection paradigm, accounted for these changes. These results suggest that serotonergic CNS systems tonically regulate aggression in Anolis carolinesis, similar to that seen in many other species. They further suggest that eyespot-darkening and aggressive responding can be pharmacologically dissociated, implicating serotonin in the regulation of this phenomenon.
Keywords: Anolis carolinensis; Aggression; Fluoxetine; Lizard; Behavior; Brain; Serotonin
A number of laboratories have reported on the use of the lizard species Anolis carolinensis, in studies examining aggression and aggressive behaviors [ 1 - 1 1 ] . Anolis c. has a number of clearly identifiable aggressive responses that frequently occur both in the wild and in captivity [ 1-5]. Behaviors include dewlapping (i.e., forced extension of the colored throat fan in the male), head bobbing, changes in body posture, and threatened and actual biting. One common aggressive response, the assertion display [2,5], consists of a series of deep and shallow head nods which is typically accentuated by pushups and brief extensions, 5-12 times, of the dewlap. It is commonly seen in aggressive encounters among males during which territorial defense or pecking-order is defended. While the neuroanatomical regulation of these responses is poorly understood, past work [6,10] has
1Fax: (203) 679-1316. Tel.: (203) 679-3365. E-mail: Deckel@psychiatry.uchc.edu.
0166-4328/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved
indicated that lesions of the paleostriatum, but not the amygdala (i.e., the basal portion of the dorsal ventricular ridge), will reduce aggressive responding in Anolis c. One of the unique aspects of Anolis c. is the apparent split brain properties of its central nervous system. Aside from a small anterior commisure and a vestigial interhemispheric hippocampal (dorsal-pallial) commisure that interconnects regions of the dorsal cortex [12,13], there appears to be no homologue to the mammalian corpus callosum. As the retinal fibers from each eye project exclusive to the contralateral lateral geniculate/optic tectum in Anolis c. [-13], this presumably results in a functionally split brain, with each hemisphere receiving information from one eye and having little, if any, information available to it from the other eye/hemisphere. Work by Greenberg et al.  further demonstrated that the brain of Anolis c. is not functionally integrated. These authors demonstrated that animals with amygdala lesions failed to respond to certain forms of social stimulation when the contralateral, but not ipsilateraL eye was patched. From this, Greenberg and
A. WallaceDeckell/Behavioural Brain Research78 (1996) 175-182
colleagues suggested that the cortex ipsilateral to the patched eye did not have access to the visual input available to the contralateral cortex. Recent work  has further suggested that unpatched, male Anolis c. have a preference for use of their left eye when engaged in aggressive interactions with other males. No preference for eye use, however, was seen in non-aggressive interactions. These findings suggest that the preferred eye during aggressive encounters between male Anolis c. lizards is the left, and that neural control of aggression in free-ranging anoles may thus be lateralized to the right hemisphere. The above work, however, fails to address which neurotransmitter systems in Anolis c. are involved in the regulation of aggression. In other systems where behavior has been well studied, such as the rodent or monkey, the neural mechanisms that underlie aggression remain poorly understood. However, recent work has implicated serotonin in the tonic inhibition of aggression across many different species, including primate [ 14], rodents [15-18] and non-rodent carnivores . Clinical and preclinical research, as well, has implicated CNS serotonergic systems in the control of aggression in both adults [20-23] and children [24,25]. These findings suggest that serotonin is a likely candidate for tonic regulation of aggression in Anolis c., as well. Because of its split-brain neuroanatomy and apparent preference for its left eye/right hemisphere in aggressive interaction, this in an animal with a measurable and easily provoked aggressive response, Anolis c. appears to be an underappreciated but otherwise excellent candidate for studying the neural circuitry of aggression. In the current experiment, it was hypothesized that the serotonergic system of the Anolis c. is tonically inhibitory to the expression of aggressive behaviors. Specifically, this experiment tested the ability of the serotonin reuptake inhibitor, fluoxetine, to inhibit aggression in this species. It further examined the effect that fluoxetine had on peripherally related aspects of the aggressive response in Anolis, such as changes in skin color and eyespot darkening.
the animals to burrow underground during the evening. All subjects were watered daily and maintained on an ad lib diet of meal worms and/or crickets, three times per week. While in their home cages, animals were socially restricted and not allowed to see any other Anolis.
2.2. General methods
After at least three days of social isolation, aggressive episodes were measured in all subjects once daily in the morning via the following procedure. An intruding male (the same male was used for all subjects throughout the course of this experiment) was placed into the subject's cage for 5 min. The time to the first aggressive response by the subject, the number and type of aggressive responses, and skin coloration both prior to and following the 5-min exposure were then measured. All subjects underwent daily 5-min pairings with the intruder male until they reliably began to show aggressive responses. Following this pre-trial preparation, subjects were tested behaviorally according to the following schedule: (1) pre-drug trials: 6 daily trials prior to the first drug injection to establish a baseline of aggressive responses; (2) drug trial 1: injection of either 0.3 mg fluoxetine in 0.03 ml H20 (this dose, in pilot studies, was found to be well tolerated by the subjects and demonstrate behavioral effects) or 0.03 ml H20 15 min prior to the start of the trial, followed the next day by reversal of the drug condition in a cross-over design; (3) betweendrug trials: 5 daily trials of non-drug; (4) drug trial 2, as described above; and (5) post-drug trials: 2 non-drug trials.
2.3. Aggressive responses
2. Methods 2.1. Subjects
Eight adult male Anolis carolinensis lizards were obtained from commercial sources and used as subjects for this experiment. Animals had an average snout to tail-tip length of 180.5 mm. All animals were maintained on a 16-h light/8-h dark cycle, kept warm with continuous exposure to a 75 watt lamp, and housed in a room whose daily ambient temperature was 70 degrees Fahrenheit. Animals were kept individually in clear plastic cages measuring 5" x 5" x 8". The bottoms of the cages were covered with pine shavings which allowed
Based on the work of a number of authors [1-7] aggressive actions were rated in ascending order of aggressivity. These included: (1) head bobbing (i.e., a series of rapid head bobs unaccompanied by any other behaviors), (2) dewlapping (i.e, a series of rapid body bobs, often with accompanying tail-lifting movements but unaccompanied by other behaviors; Fig. lb), (3) whole body bobbing (i.e., a series of rapid body bobs, often with accompanying tail-lifting movements but unaccompanied by other behaviors), (4) head bobbing plus dewlapping, (5) aggressive locomotion towards the other animal, (6) threatened bite, as evidenced by an open mouth on the part of the aggressing animal (Fig. lc), and (7) actual bite.
Skin color was measured prior to the introduction of the intruding male, and following the completion of the five minute trial. Skin coloration was graded on the fivepoint scale used by Hennig et al. . This included: (1)
A. WallaceDeckell/BehaviouralBrain Research 78 (1996) 175-182
foreign male into the subject's cage, animals were injected either with placebo (0.3 ml H20) or fluoxetine (0.3 mg fluoxetine in 0.3 ml H20) i.p. Subjects then were allowed to remain undisturbed in their cage until the intruder male was introduced. Twenty-four hours later, conditions were reversed, such that those animals that had received fluoxetine the previous day now received placebo, and vice versa. 2.6. Post-orbital eyespot darkening In Anolis c. the postorbital eyespot is a small patch of skin surrounding the postorbital region of the eye which becomes blackened during challenge. During resting conditions, it is not visible or noticeable. The eyespot will often darken when the animal is behaving aggressively or is acutely stressed (Fig. lb), and is believed to reflect adrenergic activation . In preliminary work, we found that postorbital darkening begins, on the average, between 30-60 s following introduction of the intruder male. Once it appears, the darkened spot is maintained throughout the aggressive period. We first became aware during drug trial 1 that many of the fluoxetine-injected subjects showed a significant darkening of their eye spot beginning approximately 5 rain following injection, and that none of the water-injected controls showed this response. As a result, during drug trial 2 postorbital eyespot darkening was rated as either being 'present' or 'absent' 15 min after injection, prior to the start of the behavioral trial.
Fig. 1. Aggressive responding in Anolis c. Three consecutive photographs taken a few minutes apart of a male Anolis c., shortly before (a) and during (b, c) exposure to an intruding male. The postorbital eyespot (indicated by arrows) has become prominentduring the onset of aggressive responding. (a) Appearance of non-aggressive male. (b) Post-orbital eyespotand dewlappingresponse(arrows).(c) 'Threatened' bite response.
maximally light (i.e., green), (2) light with speckling, (3) midway between light and dark, (4) mostly dark with some light speckling, and (5) maximum dark (i.e., brown). Change in skin color, obtained by subtracting the scoring for the final coloration from the initial scoring, was then obtained for each subject during each trial. 2.5. Drug trials Two drug trials examining the effect of the 5-HT uptake inhibitor fluoxetine (gift of Eli Lilly), separated by seven days, were conducted using a crossover design. Preliminary work determined that 0.3 mg/animal fluoxetine (i.e., approximately 60 mg/kg), i.p., was well tolerated by the animal, left the animal responsive, alert and awake without any observable sedation, and had the observable pharmacological effect of causing the eye spot (i.e., an area of skin posterior to the eye) to appreciably darken following injection. The drug trials were conducted in the following manner. Fifteen minutes prior to introduction of the
3.1. Drug trials 3.1.1. General The subjects tolerated the water and fluoxetine injections with no observable distress. Following injection with fluoxetine, animals appeared alert, awake, and in many cases visually tracked the intruder male when it was placed into their cage. 3.1.2. Aggressive responses during the drug trials Seven of the eight fluoxetine-injected animals during drug trial 1 showed no aggressive responses. Conversely, seven of the eight vehicle-injected controls responded aggressively to the intruding male during drug trial 1. On drug trial 2, three of the eight fluoxetine-injected, and one vehicle-injected, showed no aggressive responding. Four of the five responding fluoxetine-injected animals, however, had under five total aggressive responses, while controls averaged 18.1 aggressive responses. Table 1 shows the mean and standard deviations of the aggressive responses in fluoxetine- and vehicle-injected subjects. Analysis of variance (ANOVA) during the initial drug trial found that the fluoxetine-injected group
178 Table 1 Aggressive responses during drug trials
A. Wallace Deckell/Behavioural Brain Research 78 (1996) 175-182
Drug trial 1 water 1. Time to first response (s) 2. Head-bobbing only 3. Dewlapping only 4. Head-bobbing plus dewlapping (assertation) 5. Aggressive movement 6. Threatened bites 7. Actual bites 8. Total 114.1 (102) 0.62 (1.06) 0.25 (0.7) 9.62 (7.8) 0 0 0 11.0 (0.7) fluoxetine 269 (86)*** 0 0 0.62 (1.8)** 0 0 0 0.62 (1.8)***
Drug trial 2 water 85.2 (90) 0 0.25 (0.7) 14.1 (8.7) 0.62 (0.92) 0.12 (.35) 0 18.1 (11.9) fluoxetine 153.3 (133) 0 0 3.0 (3.4)*** 0 0 0 3.0 (3.4)***
Means and standard deviations of aggessive responses between fluoxetine- and control-injected subjects during drug trial 1 and drug trial 2. * P<0.05; ** P<0.01; *** P<0.005, fluoxetine vs H20.
had significantly fewer head bobbing plus dewlapping assertion responses (F(1,14) = 10.22; P = 0.006), took significantly longer to demonstrate the first aggressive response (F(1,14)=10.86; P=0.005), and made fewer total aggressive responses (F(1,14)=10.96; P=0.005) than vehicle-injected animals. During drug trial 2, a similar pattern of responding was seen. Both the number of head bobbing plus dewlapping (assertion responses: F(1,14) = 11.38; P = 0.005) and the total number of aggressive responses (F(1,14) = 11.80; P=0.004) were significantly reduced in the fluoxetineinjected animals in comparison to vehicle-injected controis. These results are also shown in Table 1.
3.1.3. Changes in skin color during the drug trials
No significant drug-induced changes were detected in skin color during either drug trial 1 or 2. Animals were uniformly maximally green following injection with either vehicle or fluoxetine by the second drug trial, and showed only minimal darkening at any point during the initial drug trials.
responding, i.e., if the aggressivity of the subjects changed on the non-drug trial days that followed the fluoxetine trials. Initial ANOVA analyses found no differences in responding within the three time periods (i.e., within pre-drug, between-drug or post-drug groups). For example, subjects were no more active on trial 1 of the between-drug group than they were on trial 5. As a result, total number of responses within each period were pooled for analysis, and the average number and type of aggressive responses were compared between the pre-drug, between-drug and post-drug groups. ANOVA found that change in the 'head bobbing plus dewlapping' assertion were significantly different (P=0.005) on repeated measures analyses. Similarly, the score for total aggressive responses showed a nearly significant trend (P=0.075) across time. Post-hoc Tukey t-test analysis found the post-drug trial to show significantly more assertion challenges (i.e., head-bobbing and dewlap) than either the pre-drug or between-drug groups, as shown in Table 2. The above analysis indicates that subjects in the post-drug condition had significantly more assertion
Table 2 Aggressive responses during pre-drug, between-drug, and postdrug trials Pre-drug 1. Time to first response (s) 2. Head-bobbing only 3. Dewlapping only 4. Head-bobbing plus dewlapping 5. Aggressive movement 6. Threatened bites 7. Actual bites 67.3 (78) 2.06 (3.8) 0.06 (0.24) 11.37 (7.9) 0.25 (0.48) 0.10 (0.37) 0.19 (0.67) Between-drug 71.6 (87) 1.17 (3.5) 0.08 (0.27) 15.35 (8.8) 0.31 (0.77) 0.18 (0.68) 0.10 (0.38) Post-drug 48.4 (31) 0.12 (.34) 0 18.8 (8.9)*+ 0.12 (0.34) 0 0.06 (0.25)
3.1.4. Changes in postorbital eyespot darkening following injection
Changes in the postorbital eyespot color were measured during drug trial 2. Seven out of eight of the fluoxetine-injected subjects showed a significant darkening of the eyespot by the end of the 15-min waiting period following injection, while all eight control-injected animals showed no change in eyespot 15 min post injection (chi-square: P < 0.05). Darkening was apparent as early as 1 min following injection, and generally appeared by 5-7 min post-injection.
3.2. Non-drug trials 3.2.1. Changes in aggressivity between pre-drug, between drug, and post-drug trials
The second series of analyses examined what effect the drug injections had on the chronic rate of aggressive
Means and standard deviations of aggessive responses during trials prior to the drug trials, between the two drug trials, and following the completion of the drug trials. * P<0.05 post-drug vs. pre-drug; + P < 0.05 post-drug vs. between-drug.
A. Wallace DeckelX/Behavioural Brain Research 78 (1996) 175-182
responses than in the pre-drug and/or between-drug conditions. To determine if this effect was simply a sensitization effect that increased with increasing exposure to the intruder male, a second analysis was run to see if the number of aggressive responses varied as a function of the number of trials completed. No significant differences were seen as a function of number of trials either for the pre-drug, between-drug, or post-drug conditions. That is, subjects were not significantly more aggressive on the fifth day of the pre-drug condition than on the first, on the fifth day of between-drug than the first, or on the second day of post-drug than the first. Thus the subjects did not appear to be more aggressive after repeated exposure to the intruder per se, but rather showed increased aggression across drug conditions.
3.2.2. Changes in skin coloration between pre-drug, between-drug, and post-drug trials
Both the skin coloration at the end of the aggressive pairing (F(2, 100)=6.14; P=0.003), and the change in coloration from the beginning to end of the pairing (F(2,100)-3.71; P=0.028), showed significant changes on ANOVA. Post-hoc Tukey t-tests found that subjects in the post-drug condition lightened (i.e., became significantly more green) over the course of the aggression trial in comparison to the pre-drug or between-drug conditions, while initial coloration at the start of the aggression trials did not significantly vary over the course of the experiment (Table 3).
The current findings suggest that acute fluoxetine administration significantly decreases aggression in the lizard Anolis carolinensis. Fluoxetine-injected animals showed a drastic reduction in their aggressive responses in comparison to their behavior when injected with water. There was no obvious evidence of sedation,
Table 3 Changes in skin color during pre-drug, between-drug, and postdrug trials Pre-drug 1. Initial skin color 2. Final skin color 3. Change in skin color 1.27 (0.61) 1.35 (0.70) Between-drug 1.18 (0.60) 1.0 (0) 0.18 (0.60) Post-drug 1.43
0.37 (0.80)+ *
Means and standard deviations of initial skin color, final skin color, and color changes during trials prior to the drug trials, between the two drug trials, and following the completion of the drug trials. * P<0.05 post-drug vs. pre-drug; + P<0.05 post-drug vs. betweendrug; ** P<0.05 pre-drug vs. post-drug; + + P<0.05 pre-drug vs. between-drug.
discomfort, rigidity, or other obvious behavioral artifacts that could account for the differences. Indeed, in most cases the fluoxetine-injected subjects behaved normally following the introduction of the male intruder into their cage, visually tracking the movements of the intruder, orienting towards him, etc. Despite the outwards appearance of normality in their behavioral repertoire, fluoxetine-injected animals failed to show the expected aggressive responses. A large body of animal experimentation has linked brain serotonergic systems to aggressivity. Rats with 5,7-dihydroxytryptamine lesions of dorsal and median raphe serotonergic nuclei demonstrate a modest reduction in offensive aggressive behaviors when intruder rats are placed in their cages . Administration of eltoprazine, a nonspecific postsynaptic 5-HT 1 agonist, to these rats also reduces offensive behaviors, while not effecting other social and nonsocial behaviors . These results have been extended to other non-selective 5-HT 1 agonists, including RU24969 and TFMPP, which, like eltoprazine, selectively reduce offensive aggressive behaviors [15,16]. In mice, selective agonists of 5-HT~A receptors, including 8-OH DPAT, flesinoxan, gepirone, buspirone, isapirone, and BMY 14802, also reduce isolation-induced aggressivity . Conversely, in rats, 5-HT1A specific agonists such as buspirone, isapirone and 8-OH-DPAT, while decreasing aggression in territorial and maternal aggression [ 15,18], also cause a decrease in social interest and activity . Administrations of 5-HT 2 antagonists in paradigms using intruder-introduced aggression also have been found to blunt aggression , while benzodiazepines and traditional antidepressants and antipsychotics only reduced aggressive behaviors at debilitating and sedating dosages . 5-HT augmenting drugs also decrease footshock-induced aggression in paired rats . Specifically, 5-hydroxytryptophan (the precursor of 5HT), fluoxetine (a selective inhibitor of 5-HT reuptake), and quipazine (a 5-HT receptor agonist) attenuate footshock-induced aggression (FIA), while the 5-HT depleting drug, p-chlorophenylalanine, and the 5-HT 2 antagonist, ketanserin, increased aggression in response to FIA . Predatory killing behaviors, both in rats and some strains of mice, are blunted after administration of 5-HT enhancing drugs . Conversely, defensive aggression, i.e., aggression only displayed during periods of external threat, appears not to be associated with 5-HT brain systems [ 19]. While the relationship between 5-HT~A receptors and the mediation of isolation-induced aggression in mice has been demonstrated , no effects on aggression are seen after administration of citalopram (a 5-HT uptake inhibitor) [16,28] or paroxetine , while sertraline, fluoxetine, femoxetine and fluvoxamine showed weak anti-aggressive effects in mice 1-28]. In addition, no effects are seen when the 5-HT 2 antagonists ritanserin or DOI are administered to aggressive isolated
A. WallaceDeckell/BehaviouralBrainResearch78 (1996) 175-182
viors. Although the work cited above suggests that, in other species, it is particularly the 5-HTIA serotonin receptor subtype that is involved in inhibition of aggression, the current experiment leaves unclear which of the many serotonin receptors may be involved in this regulation in Anolis carolinensis. Further work to clarify this issue is underway in our laboratory. Beyond the effects of fluoxetine in the acute-injection paradigm, animals chronically behaved differently in the pre-drug, between-drug, and post-drug conditions. The number of assertion responses significantly increased in the post-drug group. Changes in the aggressivity in the post-drug group appears not to be a simple sensitization response to increasing number of pairings with the agonistic male. Aggression did not increase in the later trials of the pre-drug condition in comparison to the earliest trials, nor were changes in aggressivity seen between early vs. later trials in the between-drug or postdrug conditions. Rather, subjects appeared to become more aggressive in response to some aspect of the injection procedures. Unfortunately, the current experimental design does not allow for clarification as to which aspect of the drug injections, such as handling, injecting, or the fluoxetine itself, lead to the behavioral changes over time. However, these findings raise the possibility that animals were becoming increasingly aggressive once the fluoxetine plasma levels, and CNS levels of serotonin, were manipulated. While further experimentation will be required to more fully and adequately assess this phenomena, these results raise the possibility not only that acute increase of serotonin levels decrease aggression, but that withdrawal from serotonin-enhancing agents may lead to 'rebound' aggression. Skin color at the completion of the 5-min behavioral trial in the pre-drug condition was significantly darker than skin color in the between-drug or post-drug conditions. Additionally, post-drug subjects became significantly 'greener' over the course of the 5-min behavioral trial, suggesting that skin color 'reactivity' was greater in the post-drug condition than in the other conditions. These findings are similar to an earlier report , in which it was found that aggression in Anolis carolinensis typically occurred in animals whose skin color 'lightens' significantly, but was reduced as the skin color darkened during pairing with an intruder male. Others as well [9,40] have reported that brown-skinned Anoles tended to lose significantly more often in agonistic encounters with other males. It is known that increases in catecholamines as well as adrenal corticosteroids lead to a change of the skin pigmentation in Anolis c. MSH (melanocytestimulating hormone) and ACTH (adrenal corticotropin hormone) are both released in acute stress, and darken body color, in Anolis c. [9,35-39]. Catecholamines act to darken chromatophores  via direct action at the level of the chromatophore, and indirectly affect skin
mice except at very high dosages [ 16]. Fenfluramine, a 5-HT releasing compound, inhibited murine aggression dose-dependently, while depletion of central stores of 5-HT by p-chlorophenylalanine attenuated this effect significantly . Thus while both the mouse and rat literature support a role of 5-HT1A receptors in the regulation of aggression, it is unclear what role, if any, other 5-HT systems play in the regulation of aggression in the mouse. In rhesus monkeys, a significant negative correlation between aggression and cerebrospinal fluid 5-HIAA has been found, implicating an inhibiting effect of 5-HT systems on aggressive behavior in primates [ 14]. Young male rhesus macaques with low CSF 5-HIAA concentrations are more likely to be involved in chases and physical assaults, and to be more at risk for experiencing physical wounding . Low concentrations of 5-HIAA also correlate with risk-taking in these monkeys, such as taking longer leaps in the forest canopy . In adult male vervet monkeys, augmentation of 5-HT systems by administration of tryptophan or fluoxetine, or decreases in 5-HT levels via administration of drugs that reduced the 5-HT function (i.e., fenfluramine or cyproheptadine) did not effect dominance acquisition . Raleigh et al.  suggest that these results support the conceptual distinction between acquired dominance status and aggression, and suggest that serotonergic mechanisms may mediate the latter, but not former, behavioral repertoire. In many other animal models, 5-HT systems have been linked to the inhibition of aggression. Silver foxes that have been tamed to human contact, in comparison to undomesticated and aggressive foxes, have higher serotonin (5-HT) and tryptophan hydroxylase levels in the midbrain and hypothalamus, a reduced density (Bmax) of 5-HTIA receptors in the hypothalamus, and a higher 5-hydroxyindole acetic acid (5-HIAA) content in midbrain, hypothalamus and hippocampus . Depletion of 5-HT brain levels by p-chlorophenylalanine has been linked to social aggression in chicks . Cichlid fish given intracerebral injections of 5-HT demonstrated reduced aggression . In clinical studies, as well, tendencies towards impulsivity and aggression have been linked to a variety of different 5-HT measurements, such as decreases in 5-HT receptors, decreased 5-HIAA CSF concentrations, etc. [20-22,34]. In youths, physical aggression has also been linked to abnormal serotonin activity . Thus from rodents to humans, serotonin has been found to have an inhibitory role in the modulation of aggression. The current experiment extends these findings to lizards. Specifically, fluoxetine reduced the number of social assertion responses in the injected subjects. As controls did not exhibit biting or threatened-biting behaviors post-injection with water, apparently in response to the stress of injection and handling, it is unclear what effect, if any, fluoxetine had on this level of aggressive beha-
A. Wallace Deckell/Behavioural Brain Research 78 (1996) 175 182
darkening by interacting with MSH via B2-receptor activation [9,37,38], presumably in cortical regions of the CNS. Thus control of skin color changes appears both to be under direct control of adrenal output, and to be regulated via MSH-related central nervous system mechanisms. The current experiment suggests that the reactivity of skin color changed in the post-drug as opposed to predrug condition, with subjects showing a significant 'greening' in their coloration over the course of the 5-min pairing with the intruding male. The reason for this change is not clear, and may involve injection, handling, or be a function of exposure to the intruder. However, given the acute effects of the fluoxetine in darkening the postorbital eyespot, presumably as a result of flooding the post-synaptic cleft with serotonin (5-HT), it may be that the chronic skin color changes resulted from alteration of central 5-HT receptor numbers. These reasoning remains speculative, however, and further work will be required to better elucidate this mechanism. The presence of the darkened 'eyespot' in seven of the eight fluoxetine-injected subjects offered external confirmation of the pharmacological effect of the fluoxetine upon the injected subject. All of the animals with the postorbital eyespot darkening during the first injection with fluoxetine failed to demonstrate any aggressive responses during their pairing with the intruder male, demonstrating the dissociability of aggression with postorbital darkening. These results are surprising in light of a recent report by Summers and Greenberg , in which postorbital darkening was found to significantly predict outcome in aggressive encounters between male Anolis c. Those animals showing the quickest darkening were more aggressive and dominated the animal with the slower eyespot darkening . Earlier work by Hadley and Goldman 1-36] reported that the eyespot, unlike the rest of the skin of Anolis, possesses only B2-adrenergic receptors and does not have direct sympathetic innervation. Citing this, Summers and Greenberg concluded that postorbital darkening likely reflected greater catecholamine activation that, in turn, leads to the greater aggressivity of the animal with the quickest eyespot darkening. However, the presence of the darkened postorbital eyespot in the current experiment, in the absence of obvious arousal or aggressivity of the subject, suggests that the control of the eyespot may not simply be under the control of circulating catecholamines, as Summers and Greenberg [-9] suggested. Rather, the current findings suggest that autonomic arousal and eyespot darkening may be dissociable under certain circumstances, and suggests that the eyespot may be at least under the partial control of serotonergic influences. Whether these influences are local or central is not clear from the current experimental findings. In summary, results from the current experiment suggest that: (1) acute increase of 5-HT levels in the
CNS of Anolis carolinensis is accompanied by a sharp decrease in aggressive behaviors; (2) post-orbital eyespot darkening can be dissociated from a general arousal response in Anolis carolinensis, and may be under direct control, at least in part, of CNS serotonergic systems; and (3) increased aggression and skin-color reactivity was seen over the course of this experiment, possibly as a consequence of fluoxetine injection, although further experimentation will be required to rule-out nonspecific factors such as handling, etc.
[-1] Carpenter, C.C., Ritualistic social behaviors in lizards. In N. Greenberg and P.D. MacLean (Eds.), Behavior and Neurology of Lizards, NIMH, Bethesda, MD, 1978, pp. 253-267.  Greenberg, N., A neuroethological study of display behavior in the lizard Anolis carolinensis (Reptilia, Lacertilia, Iguanidae), Am. Zool., 17 (1977) 191-201.  Jenssen, T.A., Evolution of anoline lizard display behavior, Am. Zool., 17 (1977) 203-215. [-4] Jenssen, T.A., Display diversity in anoline lizards and problems of interpretation. In N. Greenberg and P.D. MacLean (Eds.), Behavior and Neurology of Lizards, NIMH, Bethesda MD, 1978, pp. 269-285. [-5] Greenberg, B. and Noble, G.K., Social behavior of the american chameleon (Anolis carolinensis), Physiol. Zool., 17 (19441 392-439.  Greenberg, N., MacLean, P.D. and Ferguson, J.L., Role of the paleostriatum in species-typical display behavior of the lizard (Anolis carolinensis), Brain Res., 172 (1979) 229-241.  Deckel, A.W., Laterality of aggressive responses in Anolis, J. Exp. Zool., 272 (1995) 194. [-8] Hennig, C.W., Dunlap, W.P. and Harston, C.T., Tonic immobility and skin color in Anoles: effects of serotonin precursors and metabolites, Physiol. Behav., 22 (1979) 1079 1088.  Summers, C.H. and Greenberg, N., Somatic correlates of adrenergic activity during aggression in the lizard, Anolis carolinensis, Horm. Behav., 28 (1994) 29-40. [-10] Greenberg, N., Scott, M. and Crews, D., Role of the amygdala in the reproductive and aggressive behavior of the lizard, Anolis carolinensis, Physiol. Behav., 32 (1984) 147-151. [-11] Evans, L.T., A study of social hierarchy in the lizard Anolis carolinensis, J. Genet. Psychol., 48 (1936) 88-111.  Armstrong, J.A., Gamble, H.J. and Goldby, F., Observations on the olfactory apparatus and the telencephalon of Anolis, a microsmatic lizard, J. Anat., 87 (1953) 288 307. [,13] Butler, A.B. and Northcutt, R.G., Retinal projections in Iguana iguana and Anolis carolinensis, Brain Res., 26 (1971) 1- 13.  Higley, J.D., Mehlman, P.T., Taub, D.M. and Higley, S.B., Cerebrospinal fluid monoamine and adrenal correlates of aggression in free-ranging rhesus monkeys, Arch. Gen. Psychiatry, 49 (1992) 436-441.  Olivier, B. and Mos, J, Rodent models of aggressive behavior and serotonergic drugs, Progr. Neuropsychopharmacol. Biol. Psychiatry, 16 (1992) 847-870.  Sanchez, C., Arnt, J., Hynel, J. and Moltzen, E.K., The role of serotonergic mechanisms in inhibition of isolation-induced aggression in male mice, Psychopharmacology, 110 (1993): 53-59.  Sijbesma, H., Schipper, J., de Kloet, E.R. and Mos, J., Postsynaptic 5-HT-sub-1 receptors and offensive aggression in rats: A combined behavioural and autoradiographic study with eltoprazine, Pharmacol. Biochem. Behav., 38 (1991) 447-458. [,18] White, S.M., Kucharik, R.F. and Moyer, J.A., Effects of serotoner-
A. Wallace Deckell/Behavioural Brain Research 78 (1996) 175-182
gic agents on isolation-induced aggression, Pharmacol. Biochem. Behav., 39 (1991) 729-736. [ 19] Nikulina, E.M., Neural control of predatory aggression in wild and domesticated animals, Neurosci. Biobehav. Rev., 15 (1991) 545-547. 1-20] Csernansky, J.G. and Sheline, Y.I., Abnormalities of serotonin metabolism and nonpsychotic psychiatric disorders, Ann. Clin. Psychiatry, 5 (1993) 275-281.  Linnoila, V.M. and Virkkunen, M., Aggression, suicidality, and serotonin, J. Clin. Psychiatry, 53 (1992) 46-51.  Pihl, R.O., Peterson, R.O and Jordan, B., Alcohol, serotonin, and aggression, Alcohol Health Res. World, 17 (1993) 113-116. 1,23] Siever, L. and Trestman, R.L., The serotonin system and aggressive personality disorder, Int. Clin. Psychopharmacol., 8 (1993) 33-39. 1-24] Lahey, B.B., Hart, E.L., Pliszka, S. and Applegate, B., Neuropysiological correlates of conduct disorder: A rationale and a review of research, J. Clin. Child Psychol., 22 (1993) 141 153.  Zubieta, J.K. and Alessi, N.E., Is there a role of serotonin in the disruptive behavior disorders? A literature review, J. Child Adolescent Psychopharmacol., 3 (1993) 11-35.  Schipper, J., de Kloet, E.R., Mos, J., van Aken, H. and Olivier, B., Postsynaptic 5-HTD~ receptors and offensive aggression in rats: A combined behavioural and autoradiographic study with eltroprazine, Pharmacol. Biochem. Behav., 38 (1991) 447 458.  Datla, K.P., Mitra, S.K. and Bhattacharya, S.K., Serotonergic modulation of footshock induced aggression in paired rats, Indian J. Biol., 29 (1991) 631-635.  Sanchez, C. and Hyttel, J., Isolation-induced aggression in mice: effects of 5-hydroxytryptamine uptake inhibitors and involvement of postsynaptic 5-HYla receptors, Eur. J. Pharmacol., 264 (1994) 241-247.  Mehlman, P.T., Higley, J.D., Faucher, I., Lilly, A.A., Taub, D.M., Vickers, J., Suomi, S.J. and Linnoila, M., Low CSF 5-HIAA con-
centrations and severe aggression and impaired impulse control in nonhuman primates, Am. J. Psychiatry, 151 (1994) 1485-1491. 1,30] Raleigh, M.J., McGuire, M.T., Brammer, G.L., Pollack, D.B. and Yuwiler, A., Serotonergic mechanisms promote dominance acquisition in adult male vervet monkeys, Brain Res., 559 (1991) 181-190. [ 31 ] Popova, N.K., Voitendo, N.N., Kulikov, A.V. and Avgustinovich, D.F., Evidence for the involvement of central serotonin in mechanism of domestication of silver foxes, Pharmacol. Biochem. Behav., 40 (1991) 751-756.  Buchanan, C.P., Shrier, E.M. and Hill, W.L., Time-dependent effects of PCPA on social aggression in chicks, Pharmacol. Biochem. Behav., 49 (1994) 484-488. 1-33] Munro, A.D., Effects of melatonin, serotonin, and naloxone on aggression in isolated cichlid fish (Aequidens pulcher), J. Pineal Res., 3 (1986) 257-262. 1-34] Van Praag, H.M., Serotonergic dysfunction and aggression control, Psych. Med., 21 (1991) 15 19.  Greenberg, N., Chen, T. and Vaughan, G., Circulating melanotropic levels are altered by acute and chronic social stress in lizards, Neurosci. Abstr., 12 (1986) 834. 1-36] Hadley, M.E. and Goldman, J.M., Physiological color change in reptiles, Am. Zool., 9 (1969) 489-504.  Goldman, J.M. and Hadley, M.E., Evidence for separate receptors for melanocyte-stimulating hormone and catecholamine regulation of cyclic AMP in the control of melanophore responses, Br. J. Pharmacol., 39 (1970) 160-166. 1,38] Vaughan, G.L. and Greenberg, N., Propranolol, a beta-adrenergic antagonist, retards response to MSH in the skin of Anolis carolinensis, Photochem. Photobiol., 46 (1987) 109-144.  Kleinholz, L.H., Studies in reptilian colour changes, II. The pituitary and adrenal glands in the regulation of the melanophores of Anolis carolinensis, J. Ex. Biol., 15 (1938) 474-491. 1-40] Greenberg, N. Chen, T. and Crews, D., Social status, gonadal state, and the adrenal stress response in the lizard, Anolis carolinensis, Horm. Behav., 18 (1990) 1 11.