Self-Injurious Behavior Is Decreased by Cyproterone Acetate in Adult Male Rhesus (Macaca mulatta)

Hormones and Behavior 35, 195–203 (1999) Article ID hbeh.1999.1513, available online at http://www.idealibrary.com on

Self-Injurious Behavior Is Decreased by Cyproterone Acetate in Adult Male Rhesus (Macaca mulatta) G. G. Eaton,* J. M. Worlein,* S. T. Kelley,† S. Vijayaraghavan,* D. L. Hess,* ,‡ M. K. Axthelm,§ and C. L. Bethea* ,‡ ,¶ *Division of Reproductive Science, ¶Division of Neuroscience, †Division of Laboratory Animal Medicine, and §Division of Pathobiology, Oregon Regional Primate Research Center, Beaverton, Oregon 97006; and ‡Department of Physiology and Pharmacology, Oregon Health Sciences University, Portland, Oregon 97201 Received March 3, 1998; revised December 11, 1998; accepted January 4, 1998

Self-injurious behavior (SIB) presents a serious problem in laboratory macaques that cannot be socially housed for scientific reasons and among institutionalized children and adults where it is often associated with different forms of brain dysfunction. We have experienced limited success in reducing SIB in macaques by enhancing their environment with enrichment devices. Psychotropic drugs also help, but problems are associated with their use. Because sexual and aggressive behavioral problems in men have been treated with progestational drugs, we tested the efficacy of cyproterone acetate (CA, 5–10 mg/kg/week) on reducing SIB in 8 singly housed, adult male rhesus macaques. The main findings were: (1) SIB and other atypical behaviors were significantly reduced during CA treatment; (2) serum testosterone was significantly reduced during CA treatment; (3) cerebral spinal fluid (CSF) levels of 5HIAA and HVA, metabolites of serotonin and dopamine, respectively, declined significantly during CA treatment; (4) the duration of SIB positively correlated with levels of 5HIAA in CSF; but (5) sperm counts were not reduced during treatment. Thus, CA was a partially effective treatment (3 months) for adult male macaques whose behavioral problems include SIB. In summary, CA reduced SIB, overall aggression, serum testosterone, CSF 5HIAA, and CSF HVA. We hypothesized that the progestin activity of CA represses the hypothalamic gonadal axis and decreases testosterone, which in turn decreases SIB. In addition, we speculate that the decrease in 5HIAA and HVA in CSF may have been caused by progestins decreasing the activity of MAO. Therefore, the reduction of SIB may also be related to an increase in the availability of active monoamines in the CNS. © 1999 Academic Press Key Words: progestin; androgen; testosterone; cyproterone acetate; self-injurious behavior (SIB); sperm; males; serotonin; Macaca mulatta. 0018-506X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

Self-injurious behavior (SIB, i.e., hitting or biting various body parts) is a serious problem among significant numbers of institutionalized children and adults. It also presents a serious problem in laboratory macaques that cannot be socially housed for scientific reasons (Kraemer et al., 1997). SIB in macaques has been associated with conditions of social deprivation during early development (Harlow et al., 1971; Erwin, Deni, 1979; McKinney et al., 1973; Anderson, Chamove, 1981), and frustration has been suggested as one of the proximal causes (Chamove et al., 1984). There is evidence that environmental enrichment, especially social housing, can reduce SIB (Goosen and Ribbens, 1980; Chamove et al., 1984; De Monte et al., 1992). The neurobiological basis of SIB is unknown but a decrease in striatal dopamine and the involvement of D1 dopamine receptors as well as increased turnover of serotonin and induction of tachykinin have been documented in rodent models of SIB (Sivam, 1996; Lara-Lemus et al., 1997; Breese et al., 1990). Drugs such as benzodiazepines and phenothiazines help in severe cases with macaques (personal observations). However, diazepam must be administered several times a day and fluphenazine has serious side effects (McKinney et al., 1980). Paroxitine, a selective serotonin uptake inhibitor had no lasting effect in an open clinical trial (Davanzo et al., 1998). Nonetheless, in a recent study of SIB in macaques, the administration of tryptophan significantly decreased SIB (Weld et al., 1998). Tryptophan administration elevates serotonin in the central nervous system (CNS) (Chamberlain et al., 1987). Blocking degradation of serotonin with monoamine oxidase (MAO) inhibitors would also el-

195

196

evate CNS levels of serotonin as well as, the other biogenic amines (Von Korff, 1979), but MAO inhibitors have not been tested for relief of SIB. The synthetic steroids, medroxyprogesterone acetate (MPA) and cyproterone acetate (CA), have been used to suppress abnormal sexual and aggressive behavior in men (Cooper, 1986; Thibaut et al., 1991; Thibaut and Colonna, 1992). Treatment of men with CA decreases testosterone, LH, FSH, and eventually sperm counts (Jeffcoate et al., 1980; Meriggiola et al., 1998) with a concomitant decrease in libido (Whalen, 1984). Thus, these compounds have been referred to as antiandrogenic, but their mechanism of action is not through blockade of androgen receptors. Cellular and molecular studies indicate that these steroids have strong progestin activity and moderate (MPA) to weak (CA) androgen agonist activity (Bardin and Catterall, 1981; Poulin et al., 1991; Steinsapir et al., 1991; Berrevoets et al., 1993; Warriar et al., 1994; Miyamoto et al., 1998; Rana et al., 1998; Luthy et al., 1988; Kuil and Mulder, 1996; Imai et al., 1990). Data are scarce and somewhat inconsistent on the effects of CA and MPA on the behavior of nonhuman primates. MPA has been reported to reduce aggression and serum testosterone levels in a male chimpanzee (Pan troglodytes) (Orkin, 1993). In male cynomolgus monkeys (M. fascicularis), MPA decreased plasma testosterone and sexual behavior (Zumpe and Michael, 1988), but the effect on male–male aggression was equivocal because control stimulus males increased their rates of aggression (Zumpe et al., 1991). CA slightly increased the frequency of sexual behavior in male stump-tailed macaques (M. arctoides) tested with females, while serum testosterone was significantly reduced (Slob et al., 1983). We hypothesized that the frequency of SIB would be reduced in male macaques by increasing progestin levels and/or by the concomitant reduction in serum testosterone, previously correlated with aggressive behavior in macaques (Eaton et al., 1973). In addition, we speculated that CA may decrease MAO activity, thus elevating CNS levels of biogenic amines. Therefore, this study determined the effect of CA treatment on the frequency of SIB and other atypical behaviors in singly housed adult male rhesus. In addition, we assessed sperm counts after prolonged suppression of gonadal hormones. Finally, the concentrations of the metabolites of serotonin and dopamine, 5 hydroxyindole acetic acid (5HIAA), and homovanillic acid (HVA), respectively, were examined in cerebral spinal fluid (CSF) and correlated with changes in behavior.

Eaton et al.

MATERIALS AND METHODS Subjects Eight adult male rhesus monkeys from 8 to 11 years of age were used in the study. These animals were all participating in a retrovirus study, so they were individually housed. Five of the eight were antibodypositive, and three were virus-positive for type D simian retrovirus. All had a record of SIB and/or other atypical behaviors serious enough to be entered into their permanent medical histories in the previous 2 years. All animals were weighed monthly, and their health was monitored by a veterinarian who is Board Certified in Laboratory Animal Medicine. Observations were conducted between October 1995 and July 1996. Design Each animal served as its’ own control in an ABA paradigm. Behavioral and physiological data were collected for a 14-week baseline period (baseline 1 and baseline 2 were divided into two 7-week blocks for trend analysis purposes), then CA was administered weekly (5 mg/kg, s.c. injections) for 8 weeks (low dose) under the same data collection regimen. After the first 8 weeks, injections were increased to twice a week (total 5 10 mg/kg) for 4 weeks (high dose). Treatment was then stopped, and baseline data were collected for an additional 5 weeks after the withdrawal of CA (Withdrawal). Behavioral Observations Fifteen-minute focal observations were conducted on each male, 2 times a week, with a standardized protocol that has been previously used to demonstrate that an environmental enrichment device reduced abnormal behavior in singly housed rhesus (Eaton et al., 1993). The study males were habituated to the observer for 3 weeks and then randomly observed once during a morning and once during an afternoon time period each week. Observations were conducted by the author (G.G.E.) with a laptop computer and software program (The Observer, Version 2.0, (c) 1989, Noldus Information Technology) that recorded the frequency and duration of each behavior. The behaviors recorded were as follows: (1) aggressive behaviors—self bite, self hit, bite toy, threaten observer, threaten reflection, threaten away, display; (2) atypical behaviors—stereotyped movements, saluting, self-

197

SIB in Macaques

clasping, self-orality, floating limb, hair pluck, nail bite, cage lick; (3) typical behaviors— environmental exploration, nonfocused activity, passive, scratch, selfgroom, yawn. Definitions for atypical behaviors follow Erwin and Deni (1979) and Capitanio (1986). Blood Samples Each male was sedated (ketamine HCl, i.m., 10 mg/ kg) and bled approximately once a month (10 ml, femoral vein) for testosterone determinations and for chemical analysis of nitrogen metabolism (Cooper, 1986; Kaneko, 1989). Serum testosterone radioimmunoassays were conducted in the ORPRC P30 Hormone Assay Laboratory. Testosterone data from one male was dropped after he was terminated for another experiment before this study was completed. The data are from 7 males over 9 time periods (one blood draw during baseline 1; two blood draws during baseline 2; two blood draws during the low dose treatment period; one blood draw during the high dose treatment period; and three blood draws during withdrawal period).

HPLC for Biogenic Amines with Electrochemical Detection CSF samples (500 ml) were purified through 10,000 MW exclusion columns and diluted (1:1) in HPLC mobile phase (101 mM sodium acetate trihydrate, 67 mM citric acid monohydrate, 130 mM EDTA, 320 mM 1-octane-sulfonic acid, 4% methanol, pH 4.0). The compounds were separated through a C18 reverse phase column (Keystone Sci., Inc., Bellefonte, PA; ODS Hypersil, 100 3 4.4 mm, 3-mm particle size). The eluent was passed through an electrochemical detector (Waters 460, Waters Scientific Inc., Milford, MA) set at 10.60 V and 0.2 nA. The concentration of the calibration standards equaled 5 pg/ml. The average retention times for 5HIAA and HVA were 9.7 and 15.1 min, respectively. Standard preparations were analyzed at increasing injection volumes and the measurement of 5HIAA was linear between 10 and 500 pg. The measurement of HVA was linear between 50 and 500 pg. The concentrations of 5HIAA and HVA in CSF were quantitated by comparing the area under the peak of the unknown sample with the linear curve generated with increasing concentrations of standard.

Sperm Analysis Four randomly selected males were electroejaculated (penile band) at approximately 1-month intervals for sperm analysis (one sample obtained during baseline 1; two samples obtained during baseline 2; two samples obtained during low dose treatment period; one sample obtained during the high dose treatment period; and one sample obtained during the withdrawal period). Sperm quality (count, motility, morphology, percentage live) was assessed with a computer assisted motility analysis system that measured both head and flagella motion (Stephens et al., 1988). One male was dropped from this part of the study after it was discovered that he had abnormally high numbers of dead sperm before drug treatment. Cerebral Spinal Fluid (CSF) Samples CSF fluid (1 ml) for neurotransmitter determinations was drawn monthly from five of the animals during sedation. A 20-g spinal needle (3.5 inches, Becton-Dickinson, Franklin Lakes, NJ) was inserted into the cisterna magna and CSF was collected into sterile tubes. The CSF was centrifuged to remove cell debris and stored at 280°C until amine analysis by HPLC. A note was made if any sample was contaminated with blood.

Statistical Analysis Statistical analyses of all behavioral and physiological measures were performed with a repeated measures ANOVA. Posthoc pairwise comparisons were conducted with Student–Newman–Keuls test (SYSTAT, Inc., Evanston, IL and INSTAT, Graph Pad, San Diego, CA). In addition, linear regression analysis was performed on the total number of seconds an animal engaged in SIB during the observation period versus the concentrations of serum testosterone, CSF 5HIAA, and CSF HVA samples obtained at the midpoint of each observation period. The severity of SIB differed between animals, hence the total number of seconds in each observation period was converted to the percentage of Baseline 1 for each individual (relative seconds). Two CSF samples were contaminated with blood and reliable measurements of biogenic amines could not be obtained. Therefore, the mean of the available samples at those time points was substituted for the missing points in order to perform the repeated measures ANOVA and linear regression analysis. Finally, linear regression analysis was performed on the dose of CA versus the levels of serum testosterone, 5HIAA, and HVA and also performed on the levels of 5HIAA versus HVA.

198

RESULTS The frequency of self-biting was significantly reduced during CA treatment (Fig. 1A). Self-biting then increased to predrug, baseline levels after cessation of drug treatment. Other measures of aggressive behavior such as self-hit, threats to the males’ own reflections in the sides of their cages, threats toward the observer, and threats toward other male macaques in the room followed the same general pattern as selfbiting. A composite measure of total aggression (percentage of observation time) by the males is shown in Fig. 1B. Aggression was significantly reduced during CA, and total aggression returned to predrug baseline levels following cessation of drug treatment. In addition to SIB, other atypical behaviors (a composite measure including stereotypy, salute, selfclutch, self-oral, cage lick, hair pluck, and nail bite) declined significantly in total duration (percentage of observation time) across the study. However, unlike SIB, total atypical behavior did not return to predrug levels after CA was withdrawn, but stayed as low as it had been during drug treatment (Fig. 1C). Serum testosterone levels are presented in Fig. 2 with the frequency of yawning. The frequency of yawning dropped significantly from about four per hour during baseline periods 1 and 2, to one per hour during treatment and then returned to baseline levels of five per hour when CA was withdrawn. Serum testosterone paralleled yawning and dropped significantly from a high of 10.05 6 3.16 ng/ml during baseline 1 to a low of 1.67 6 0.26 ng/ml during the high dose treatment period. Testosterone returned to 4.01 6 0.49 ng/ml during withdrawal. Sperm counts did not drop with CA treatment (n 5 4 monkeys). The mean sperm count equaled 2.60 6 0.51 during baseline 1; 2.47 6 0.18 during baseline 2; 2.41 6 0.79 during low dose periods; 3.40 6 0.38 during the high dose period; and 3.17 6 1.32 during withdrawal. Sperm motility also did not vary across treatment, nor did sperm morphology. There was a significant reduction in the concentration of 5HIAA and HVA in the CSF during high dose treatment (Fig. 3). The reduction in 5HIAA and HVA was associated with the reduction in selfbiting and overall aggressive behavior, but SIB significantly decreased during the low dose of CA, whereas the CSF metabolites decreased only during the high dose of CA. However, considerable individual variation contributed to the mean values. Following normalization of the duration of SIB, there was a significant linear correlation between

Eaton et al.

the relative seconds (percentage of baseline) that animals engaged in SIB versus the concentration of 5HIAA in CSF (P , 0.042; r 5 0.4257). However, there was no correlation between relative seconds of SIB and HVA even though there was a highly significant correlation between CSF concentrations of 5HIAA and HVA (P , 0.0001; r 5 0.9065). Finally, there was a significant correlation between the relative seconds of SIB and serum testosterone (P , 0.021; r 5 0.4274), the dose of CA administered and serum testosterone (P , 0.0001; r 5 20.67), and between the dose of CA and the concentration of 5HIAA in the CSF (P , 0.0073; r 5 20.5555), but not between the dose of CA and HVA in the CSF.

DISCUSSION Our results suggest that SIB is modulated by administration of a progestin and/or the concomitant decrease in endogenous testosterone. The frequency of self-biting and other aggressive behaviors dropped significantly from pretreatment baseline levels with CA treatment and then increased significantly when CA was withdrawn and serum testosterone increased. Atypical behaviors also decreased significantly across the treatment period, but did not increase following withdrawal, suggesting that habituation to the observer occurred with atypical behavior but not with SIB. Hence, SIB may be qualitatively different from other types of abnormal behavior. The frequency of yawning and incidence of aggressive behaviors in this study declined along with testosterone levels and then returned to baseline levels when CA was withdrawn and testosterone levels increased. Earlier data showed that yawning in male rhesus is an accurate behavioral indication of serum testosterone and that aggression is dependent upon androgen levels (Eaton et al., 1973). Rhesus males engage in SIB in our colony more frequently than do females, further indicating that SIB is augmented by androgens. Male rhesus monkeys housed indoors continue to exhibit seasonal variation in testosterone with a peak during the breeding season (Oct, Nov, Dec) and a nadir in April, May, and June (Michael and Bonsall, 1977). Thus, some of the variation in testosterone during baseline 1 may be due to this seasonality. However, it is unlikely that the increase in testosterone that occurred during the withdrawal of CA (May, June) was due to seasonal effects. The relative contribution of the progestin addition and testosterone decline on SIB and aggression is of

SIB in Macaques

199

interest. In other words, is SIB solely dependent upon testosterone which declined due to the progestin treatment? Would block of androgen activity with a pure antagonist such as flutamide reduce SIB to the same extent? Or does the progestin act on neural centers involved in aggression? This could be tested by treatment with a pure progestin agonist and testosterone replacement. For example, CA blocked male sexual behavior in sheep even when testosterone was present in a manner identical to progesterone (Fabre-Nys, 1979). Sperm counts did not decline in the macaques treated with CA in this study. This differs somewhat from the results of clinical studies employing CA for contraceptive purposes (Meriggiola et al., 1998). In men treated with CA, azoospermia was achieved in approximately 9 weeks. However, the men treated for contraceptive study received CA daily at a significantly higher dose. Psychiatric studies indicate that the serotonin neural system is dysfunctional in violent criminal populations (Coccaro et al., 1996). In addition, several studies have reported lower levels of 5HIAA in the CSF of impulsively aggressive humans (Brown and Linnoila, 1990) and monkeys (Higley et al., 1992; Mehlman et al., 1994). Indeed, elevation of serotonin with tryptophan administration reduced SIB in macaques (Weld et al., 1998). However, as recently reviewed (Kraemer et al., 1997), environmental causes of SIB do not reliably produce reductions in brain serotonin and experimental reduction of sero-

FIG. 1. (A) Self-bite. Mean frequency per hour of self-biting by 8 adult male rhesus monkeys. Fifteen-minute behavioral observations were conducted on each male twice a week. Baseline 1 and 2 were 7 weeks each with no drug treatment. Low dose was 8 weeks of 5 mg/kg/week/male of cyproterone acetate. High dose was 4 weeks of 10 mg/kg/week/male of cyproterone acetate. Withdrawal was 5 weeks of no drug treatment. There was a significant difference between treatment periods (Anova df 5 4, 28; F 5 3.00; P 5 0.035). The high dose treatment mean was significantly lower than baseline

2 (pairwise comparison, df 5 7, t 5 3.25, P 5 0.015). (B) Aggression. Percentage of observation time of a composite score of aggression by 8 adult male rhesus monkeys. Fifteen-minute behavioral observations were conducted on each male twice a week. Aggression included self-bite, threatening the observer, threatening their own reflection in the cage wall, undirected threatening, self-hit, biting the enrichment toy, and displays. Treatment was as described in (A). There was a significant difference between the treatment groups (Anova df 5 4, 28; F 5 4.04; P 5 0.010). The high dose treatment mean was significantly lower than baselines 1 and 2 (pairwise comparison, df 5 7; baseline 1 vs high dose t 5 2.90, P 5 0.023; baseline 2 vs high dose t 5 3.83, P 5 0.007). The low dose treatment mean was also significantly lower than baseline 2 (pairwise comparison df 5 7, t 5 2.71, P 5 0.030). (C) Abnormal behavior. Percentage of observation time of a composite score of atypical behavior by 8 adult male rhesus monkeys. Fifteen-minute behavioral observations were conducted on each male twice a week. Atypical behavior included stereotypy, salute, self-clutch, self-oral, cage lick, hair pluck, and nail bite. Treatment was as described in (A). There was significant decline in atypical behavior across the study (Anova df 5 4, 28; F 5 3.19; P 5 0.028).

200

FIG. 2. Testosterone and yawn. Mean frequency per hour of yawning in 8 adult male rhesus monkeys, and mean serum levels of testosterone in 7 of the same males. Treatment was as described in the legend to Fig. 1A. There was a significant difference in yawning between treatment periods (Anova, df 5 4, 28; F 5 4.915; P 5 0.004). The high and low dose treatment means were significantly lower than baseline 2 and withdrawal (pairwise comparison, all df 5 7, baseline 2 vs low dose, t 5 3.55, P 5 0.009; baseline 2 vs high dose, t 5 2.67, P 5 0.032; withdrawal vs low dose, t 5 23.63, P 5 0.008; Withdrawal vs High Dose, t 5 23.46, P 5 0.011). There was a significant difference in testosterone between treatment periods as well. (Anova, df 5 8, 48; F 5 3.79; P 5 0.002).

tonin does not reliably promote SIB. Therefore, both serotonin and testosterone may contribute to the neuropathology of SIB. In rodents, neonatal treatment with the neurotoxin 6-hydroxydopamine produces SIB which is characterized by a large reduction in striatal dopamine and increased turnover of serotonin (Sivam, 1996; Breese et al., 1990). Group means of 5HIAA and HVA exhibited a significant decline during the high dose treatment period with a modest rebound after drug withdrawal. Linear regression analysis revealed a positive correlation between the duration of SIB and CSF concentrations of 5HIAA, but not HVA, even though 5HIAA and HVA levels were highly correlated with one another. That is, shorter periods of SIB were associated with lower levels of 5HIAA. Because 5HIAA is a metabolite of serotonin and elevations in CNS serotonin decrease SIB (Weld et al., 1998), it follows that the decrease in 5HIAA (and SIB) in this study may reflect an increase in CNS

Eaton et al.

serotonin. However, the interpretation of the concentration of monoamines in CSF is fraught with caveats. Usually the level of 5HIAA reflects the overall level of serotonin neural function, wherein lower synthesis and release of serotonin leads to lower levels of 5HIAA in CSF (Murphy, 1990). In contrast, previous data from the coauthor’s laboratory (C.L.B.) found that 5HIAA decreased in female macaques during treatment with progesterone (Schutzer et al., 1997) concomitantly with changes in gene expression, which indicate increased serotonin production (Pecins-Thompson et al., 1996, 1998; Pecins-Thompson and Bethea, 1998). Together these data suggest that the level of 5HIAA in CSF may also be dependent on serotonin degradation by MAO and, moreover, that MAO activity is decreased by progestins. This reasoning is further supported by reports that MAO inhibitors increase the concentration of serotonin and decrease the concentration of 5HIAA in rat brain and in human plasma (McKenna et al., 1992; Juorio et al., 1986; Celada et al., 1992). The observation that SIB correlated with 5HIAA, but not HVA, in spite of a significant correlation between 5HIAA and HVA requires comment. A similar pattern was observed in that the dose of CA correlated with 5HIAA, but not HVA. While this could be a statistical effect due to the higher variability of HVA, it is also important to consider possible physiological explanations. MAO occurs in two isoforms, MAO-A and MAO-B (Bach et al., 1988; Westlund et al., 1985). Serotonin is the preferred substrate for MAO-A, whereas catecholamines are preferentially degraded by MAO-B (Glover et al., 1977; Arai et al., 1997). If progestins have a more specific effect on MAO-A than MAO-B, then a closer dose–response relationship might be predicted for 5HIAA than for HVA (Chevillard et al., 1981). In conclusion, treatment of adult male rhesus monkeys with a synthetic steroid exhibiting strong progestin and weak androgen activities, CA, reduced serum testosterone and SIB. The progestin activity of CA is believed to cause the decline in testosterone, which in turn is thought to be related to the decline in SIB. However, the possibility that progestins may act on neural centers involved in aggression cannot be discounted. The concentration of 5HIAA, a metabolite of serotonin, was positively correlated with the relative length of time that animals engaged in SIB. Moreover, CSF levels of 5HIAA and HVA declined during treatment with CA in a manner observed with MAO inhibitors.

SIB in Macaques

201

FIG. 3. Concentration of 5HIAA (A) and HVA (B) in the CSF of SIB monkeys over the course of the study. CSF samples were obtained during each period as indicated on the abscissa and subjected to HPLC analysis for biogenic amines. Serotonin was at the limit of detection of the assay. Norepinephrine and epinephrine were easily detectable but no statistical trends were present. (A) 5HIAA. *High dose is different from baseline 2 and low dose 2 at P , 0.05; withdrawal is different from low dose 1 at P , 0.05 (Anova and post hoc pairwise comparison). **High dose is different from baseline 1 and low dose 1 at P , 0.01 (Anova and post hoc pairwise comparison). (B) HVA. *High Dose is different from baseline 1, baseline 2, and low dose 1 at P , 0.05 (Anova and post hoc pairwise comparison).

These observations have led to the hypothesis that progestins decrease serum testosterone as well as the activity of MAO and that the subsequent behav-

ioral modifications may be related to an increase in the availability of active neurotransmitters in the extracellular space.

202

ACKNOWLEDGMENTS We thank Berlex Laboratories, Inc., Wayne, NJ, for generously supplying the cyproterone acetate. We are grateful to Dr. Susan Iliff for drawing the CSF samples, to Mr. William Schutzer for technical assistance with HPLC analysis of the biogenic amines, and to Mr. Kevin Grund for collection of the sperm samples. This work was supported by NIH Grants HD17269 to C.L.B., P30 Population Center Grant HD18185, and RR00163 for the operation of the ORPRC.

REFERENCES Anderson, J. R., and Chamove, A. S. (1981). Self-aggressive behavior in monkeys. Curr. Psychol. Rev. 1, 139 –158. Arai, R., Kimura, H., and Maeda, T. (1997). Preferential localization of monoamine oxidase type A activity in neurons of the locus coeruleus and type B activity in neurons of the dorsal raphe nucleus of the rat: A detailed enzyme histochemical study. Brain Res. 745, 352–356. Bach, A. W. J., Lan, N. C., Johnson, S. L., Abell, C. W., Bembenek, M. E., Kwan, S.-W., Seeburg, P. H., and Shih, J. C. (1988). cDNA cloning of human liver monoamine oxidase A and B: Molecular basis of differences in enzymatic properties. Proc. Natl. Acad. Sci. USA 85, 4934 – 4938. Bardin, C. W., and Catterall, J. F. (1981). Testosterone: A major determinant of extragenital sexual dimorphism. Science 211, 1285– 1294. Berrevoets, C. A., Veldscholte, J., and Mulder, E. (1993). Effects of antiandrogens on transformation and transcription activation of wild-type and mutated (LNCaP) androgen receptors. J. Steroid Biochem. Mol. Biol. 46, 731–736. Breese, G. R., Criswell, H. E., and Mueller, R. A. (1990). Evidence that lack of brain dopamine during development can increase the susceptibility for aggression and self-injurious behavior by influencing D1-dopamine receptor function. Prog. Neuropsychopharmacol. Biol. Psychiatry 14S, S65–S80. Brown, G. L., and Linnoila, M. I. (1990). CSF serotonin metabolite (5HIAA) studies in depression, impulsivity and violence. J. Clin. Psychiatry 51, 31– 41. Capitanio, J. P. (1986). Behavioral pathology. In G. Mitchell, J. Erwin (Eds.), Comparative Primate Biology, Part A: Behavior, Conservation, and Ecology, pp. 411– 454. A. R. Liss, New York. Celada, P., Perez, J., Alvarez, E., and Artigas, F. (1992). Monoamine oxidase inhibitors phenelzine and brofaromine increase plasma serotonin and decrease 5-hydroxyindoleacetic acid in patients with major depression: Relationship to clinical improvement. J. Clin. Psychopharmacol. 12, 309 –315. Chamberlain, B., Ervin, F. R., Phil, R. O., and Young, S. N. (1987). The effect of raising and lowering tryptophan levels on aggression in vervet monkeys. Pharmacol. Biochem. Behav. 28, 503–510. Chamove, A. S., Anderson, J. R., and Nash, V. J. (1984). Social and environmental influences on self-aggression in monkeys. Primates 25, 319 –325. Chevillard, C., Barden, N., and Saavedra, J. M. (1981). Estradiol treatment decreases type A and increases type B monoamine oxidase in specific brain stem areas and cerebellum of ovariectomized rats. Brain Res. 222, 177–181. Coccaro, E. F., Berman, M. E., Lavpisso, R. J., and Hauger, R. L. (1996). Relationship of prolactin response to d-fenfluramine to

Eaton et al.

behavioral and questionnaire assessments of aggression in personality-disordered men. Biol. Psychiatry 40, 157–164. Cooper, A. J. (1986). Progestins in the treatment of male sex offenders: a review. Can. J. Psychiatry 31, 73–79. Davanzo, P. A., Beline, T. R., Widawski, M. H., and King, B. H. (1998). Paroxetine treatment of aggression and self-injury in persons with mental retardation. Am. J. Ment. Retard. 102, 427– 437. De Monte, M., Anderson, J. R., and Charbonnier, H. (1992). Selfaggression in stumptail macaques: effects of frustration and social partners. Primates 33, 115–120. Eaton, G. G., Goy, R. W., and Phoenix, C. H. (1973). Effects of testosterone treatment in adulthood on sexual behavior of female pseudohermaphrodite rhesus monkeys. Nature 242, 119 –120. Eaton, G. G., Kelley, S. T., and Iliff-Sizemore, S. A. (1993). Rawhide “chew-bones” reduce abnormal behavior in individually housed adult rhesus macaques. Am. J. Primatol. 30, 308. Erwin, J. M., and Deni, R. (1979). Strangers in a strange land: Abnormal behaviors or abnormal environment? In J. M. Erwin, T. L. Maple, G. Mitchell (Eds.), Captivity and Behavior, pp. 1–28. Van Nostrade Reinhold, New York. Fabre-Nys, C. (1979). Effect of progesterone on male and female sex behavior induced by steroid hormones in adult ovariectomized ewes. C. R. Seances Acad. Sci. D 289, 1319 –1322. Glover, V., Sandler, M., Owen, F., and Riley, G. J. (1977). Dopamine is a monoamine oxidase B substrate in man. Nature 265, (5589), 80 – 81. Goosen, C., and Ribbens, L. G. (1980). Autoaggression and tactile communication in pairs of adult stumptailed macaques (Macaca arctoides). Behavior 73, 155–174. Harlow, H. F., Harlow, M. K., and Suomi, S. J. (1971). From thought to therapy: lessions from a primate laboratory. Am. Sci. 59, 538 – 549. Higley, J. D., Mehlman, P. T., Taub, D. M., Higley, S. B., Suomi, S. J., Linnoila, M., and Vickers, J. H. (1992). Cerebrospinal fluid monoamine and adrenal correlates of aggression in free-ranging rhesus monkeys. Arch. Gen. Psychiatry 49, 436 – 441. Imai, K., Watanabe, K., Takahashi, O., Shimizu, N., Nakata, S., Kawashima, K., Suzuki, T., and Yamanaka, H. (1990). A study of the androgenic activity of anti-androgen. Nippon Naibunpi Gakkai Zasshi 66, 597– 606. Jeffcoate, W. J., Matthews, R. W., Edwards, C. R., Field, L. H., and Besser, G. M. (1980). The effect of cyproterone acetate on serum testosterone, LH, FSH, and prolactin in male sexual offenders. Clin. Endocrinol. (Oxford) 13, 189 –195. Juorio, A. V., Greenshaw, A. J., and Boulton, A. A. (1986). Effects of acute and chronic phenelzine on regional monoamine metabolism in rats and its potentiation by deuterium substitution. NaunynSchmeidebergs Arch. Pharmacol. 333, 240 –245. Kaneko, J. J. (1989). Clinical Biochemistry of Domestic Animals. Academic Press, San Diego. Kraemer, G. W., Schmidt, D. E., and Ebert, M. H. (1997). The behavioral neurobiology of self-injurious behavior in rhesus monkeys. Current concepts and relations to impulsive behavior in humans. Ann. N.Y. Acad. Sci. 836, 12–38. Kuil, C. W., and Mulder, E. (1996). Deoxyribonucleic acid-binding ability of androgen receptors in whole cells: Implications for the actions of androgens and antiandrogens. Mol. Biol. 137, 1870 – 1877. Lara-Lemus, A., Perez de la Mora, M., Mendez-France, J., PalomeroRivero, M., and Drucker-Colin, R. (1997). Effects of REM sleep deprivation on the d-amphetamine-induced self-mutilating behavior. Brain Res. 770, 60 – 64.

SIB in Macaques

Luthy, I. A., Begin, D. J., and Labrie, F. (1988). Androgenic activity of synthetic progestins and spironolactone in androgen-sensitive mouse mammary carcinoma (Shionogi) cells in culture. J. Steroid Biochem. 31, 845– 852. McKenna, K. F., Baker, G. B., Coutts, R. T., and Greenshaw, A. J. (1992). Chronic administration of the antidepressant-antipanic drug phenelzine and its N-acetylated analogue: Effects on monoamine oxidase, biogenic amines and alpha 2-adrenoreceptor function. J. Pharm. Sci. 81, 832– 835. McKinney, W. T., Young, L. D., Suomi, S. J., and Davis, J. M. (1973). Chlorpromazine treatment of disturbed monkeys. Arch. Gen. Psychiatry 29, 490 – 494. McKinney, W. T., Moran, E. C., and Kraemer, G. W. (1980). Longterm chlorpromazine in rhesus monkeys: Production of dyskinesias and changes in social behavior. Psychopharmacology 72, 35–39. Mehlman, P. T., Higley, J. D., Fuacher, I., Lilly, A. A., Taub, D. M., Vickers, J., Suomi, S. J., and Linnoila, M. (1994). Low CSF 5HIAA concentrations and severe aggression and impaired impulse control in nonhuman primates. Am. J. Psychiatry 151, 1485–1491. Meriggiola, M. C., Bremner, W. J., Costantino, A., Di Cintio, G., and Flamigni, C. (1998). Low dose of cyproterone acetate and testosterone enanthate for contraception in men. Hum. Reprod. 13, 1225– 1229. Michael, R. P., and Bonsall, R. W. (1977). A 3-year study of an annual rhythm in plasma androgen levels in male rhesus monkeys (macaca mulatta) in a constant laboratory environment. J. Reprod. Fert. 49, 129 –131. Miyamoto, H., Yeh, S., Wilding, G., and Chang, C. (1998). Promotion of agonist activity of antiandrogens by the androgen receptor coactivator, ARA70, in human prostate cancer DU145 cells. Proc. Natl. Acad. Sci. USA 95, 7379 –7384. Murphy, D. L. (1990). Peripheral indices of central serotonin function in humans. Ann. N.Y. Acad. Sci. 600, 282–295. Orkin, J. L. (1993). Use of medroxyprogesterone acetate to reduce aggression in a male chimpanzee (Pan troglodytes). Lab. Anim. Sci. 43, 260 –261. Pecins-Thompson, M., Brown, N. A., Kohama, S. G., and Bethea, C. L. (1996). Ovarian steroid regulation of tryptophan hydroxylase mRNA expression in rhesus macaques. J. Neurosci. 16, 7021– 7029. Pecins-Thompson, M., and Bethea, C. L. (1998). Ovarian steroid regulation of 5HT1A autoreceptor messenger ribonucleic acid expression in the dorsal raphe of rhesus macaques. Neuroscience 89, 267–277. Pecins-Thompson, M., Brown, N. A., and Bethea, C. L. (1998). Regulation of serotonin re-uptake transporter mRNA expression by ovarian steroids in rhesus macaques. Mol. Brain Res. 53, 120 –129. Poulin, R., Baker, D., Poirier, D., and Labrie, F. (1991). Multiple actions of synthetic ‘progestins’ on the growth of ZR-75-1 human breast cancer cells: An in vitro model for the simultaneous assay of androgen, progestin, estrogen, and glucocorticoid agonistic and antagonistic activities of steroids. Breast Cancer Res. Treat. 17, 197–210.

203

Rana, S., Bisht, D., and Chakraborti, P. K. (1998). Activation of rat androgen receptor by androgenic ligands is unaffected by antiandrogens in Saccharomyces cerevisiae. Gene 209, 247–254. Schutzer, W. E., Widmann, A. A., and Bethea, C. L. (1997). Effect of estrogen (E) and progesterone (P) on CSF concentrations of biogenic amines in pigtail macaques. Soc. Neurosci. 18, 2039. [Abstract No. 793] Sivam, S. P. (1996). Dopamine, serotonin and tachykinin in selfinjurious behavior. Life Sci. 58, 2367–2375. Slob, A. K., Schenck, P. E., and Nieuwenhuijsen, K. (1983). Effects of cyproterone acetate on social and sexual behavior in adult male laboratory housed stump-tailed macaques (Macaca arctoides). In H. D. Steklis and A. S. Kling (Eds.), Hormones, Drugs and Social Behavior in Primates, pp. 175–204. Spectrum, New York. Steinsapir, J., Mora, G., and Muldoon, T. G. (1991). Effects of steroidal and nonsteroidal antiandrogens on the androgen binding properties of the rat ventral prostate androgen receptor. Biochim. Biophys. Acta 1094, 103–112. Stephens, D. T., Hickman, R., and Hoskins, D. D. (1988). Description, validation, and performance characteristics of a new computer-automated sperm motility analysis system. Biol. Reprod. 38, 577–586. Thibaut, F., Kuhn, J. M., and Colonna, L. (1991). A possible antiaggressive effect of cyproterone acetate. Br. J. Psychiatry 159, 298 – 299. Thibaut, F., and Colonna, L. (1992). Cyproterone acetate in the treatment of aggression. Am. J. Psychiatry 149, 411. Von Korff, R. W. (1979). Monoamine Oxidase: Structure, Function and Altered Functions. Academic Press, New York. Warriar, N., Page, N., Koutsilieris, M., and Govindan, M. V. (1994). Antiandrogens inhibit human androgen receptor-dependent gene transcription activation in the human prostate cancer cells LNCaP. Prostate 24, 176 –186. Weld, K. P., Mench, J. A., Woodward, R. A., Bolesta, M. S., Suomi, S. J., and Higley, J. D. (1998). Effect of tryptophan treatment on self-biting and central nervous system serotonin metabolism in rhesus monkeys (Macaca mulatta). Neuropsychopharmacology 19, 314 –321. Westlund, K. N., Denney, R. M., Kochersperger, L. M., Rose, M., and Abell, C. W. (1985). Distinct monoamine oxidase A and B populations in primate brain. Science 230, 181–182. Whalen, R. E. (1984). Multiple actions of steroids and their antgonists. Arch. Sex. Behav. 13, 497–502. Zumpe, D., and Michael, R. P. (1988). Effects of medroxyprogesterone acetate on plasma testosterone and sexual behavior in male cynomolgus monkeys (Macaca fascicularis). Physiol. Behav. 42, 343– 349. Zumpe, D., Bonsall, R. W., Kutner, M. H., and Michael, R. P. (1991). Medroxyprogesterone acetate, aggression and sexual behavior in male cynomolgus monkeys (Macaca fascicularis). Horm. Behav. 25, 394 – 409.