Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cocaine induces larger physiological responses in female than male rats. This laboratory has shown that cocaine increased serum ACTH more in female than male rats (Kuhn and Francis, 1996). We have also shown that cocaine stimulated locomotion more in female rats (Bowman and Kuhn, 1996; Walker et al., 2000a). Several reports indicate that acquisition, maintenance, and reinstatement of cocaine self-administration are also greater in female than male rats (Roberts et al., 1989; Lynch and Carroll, 1999, 2000). Fast scan cyclic voltammetry has recently been used to show that basal dopamine release and uptake are 30 to 40% greater in the striatum of female rats (Walker et al., 2000b) and that cocaine increased dopamine overflow more in female striatum (Walker and Kuhn, 1997). Pharmacokinetic differences do not account for these sexually dimorphic behavioral, neurochemical, and neuroendocrine responses to cocaine in rats (Bowman et al., 1999).

Patterns of cocaine use in humans also exhibit sex differences. More men than women in the United States use cocaine (U.S. Department of Health and Human Services, 1993) although certain aspects of cocaine taking and addiction are more severe in women. Prevalence of cocaine dependence was actually shown to be higher in adolescent females than adolescent males (Kandel et al., 1997). Two reports of clinical treatment of cocaine abuse reported more severe cocaine use patterns in females at the time of intake (Kosten et al., 1993; Robbins et al., 1999). McCance-Katz et al. (1999) reported that women consumed cocaine by more addictive routes (as did Kosten et al., 1993) and progressed to dependence more rapidly than men. Audio and videotapes of cocaine preparation and use elicited more self-reported cocaine craving in female users (Robbins et al., 1999). Significantly more female than male crack users reported emergency room visits following crack use (Dudish and Hatsukami, 1996). Scant evidence exists describing the impact of circulating levels of gonadal hormones on cocaine responses in humans. Women administered cocaine intranasally had greater peak plasma cocaine levels in their follicular than luteal menstrual phase although subjective drug effects were the same in both phases (Lukas et al., 1996). Intravenous cocaine administration produced equivalent cocaine concentrations in follicular and luteal phase women, however (Mendelson et al., 1999). Because so little is known about how gonadal hormones influence cocaine effects, the present studies were designed to test the hypothesis that fluctuating ovarian hormone levels would affect neuroendocrine responses to cocaine in rats.

Neuroendocrine function clearly influences behavioral effects of stimulants (Goeders, 1997). Marinelli et al. (1997) have shown that adrenalectomy attenuated cocaine-stimulated locomotion and corticosterone replacement restored it, suggesting that plasma corticosterone influences the behavioral response. Corticosterone seems to mediate, at least partially, the stress-induced sensitization of locomotor effects of cocaine and amphetamine (Deroche et al., 1992; Rouge-Pont et al., 1995). Plasma corticosterone also plays a role in cocaine reinforcement. Corticosterone has been reported to influence acquisition (Mantsch et al., 1998), maintenance (Deroche et al., 1997), and reinstatement (Piazza et al., 1994; Mantsch and Goeders, 1999) of cocaine self-administration. Thus, the enhanced HPA axis effects of cocaine in female rats could contribute to enhanced behavioral reactivity.

Conversely, cocaine influences neuroendocrine function. Cocaine disrupts reproductive function in female humans, monkeys, and rats. Menstrual cycle disruptions in women and monkeys include amenorrhea, luteal phase dysfunction, and anovulation (Mello and Mendelson, 1997). Rats developed irregular estrous cycles after 7 days of repeated cocaine injections (King et al., 1990, 1993). The mechanism of these disruptions may be through cocaine's direct stimulation of luteinizing hormone release (Mello et al., 1990a, 1993; Mendelson et al., 1992) or indirect through putative changes in ovarian hormone concentrations that could alter feedback control of gonadotropin release. Cocaine has been reported to increase plasma levels of progesterone in rats (Quinones-Jenab et al., 2000). One purpose of the present study was to investigate cocaine effects on ovarian steroids at each phase of the estrous cycle as a possible mediator of its reproductive function perturbations.

The present studies postulated that estrous cycle phase and therefore ovarian hormones would modulate pituitary ACTH and adrenal corticosterone responses to cocaine. Such effects would have implications for sex- and menstrual cycle-related differences in human cocaine use. Another thrust of this work was to determine whether cocaine alters plasma levels of ovarian steroids because such an effect would have repercussions for control of gonadotropin secretion. Thus, this work has investigated cocaine effects and reciprocal interactions of the HPA and hypothalamo-pituitary-gonadal axes.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subjects. Adult male and female Sprague-Dawley rats were purchased from Charles River Laboratories (Raleigh, NC). Female rats were adrenalectomized, ovariectomized, or sham operated by the supplier before shipment. Rats were segregated by sex and were housed in plastic cages under a 12-h light/dark cycle with lights on at 6:00 AM. Food and water were provided ad libitum. Animal care was in accordance with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health publication 865-23, Bethesda, MD) and approved by the Institutional Animal Care and Use Committee.

Experimental Rationale and Protocol. On the day before the experiment all animals were moved from the main vivarium to local animal quarters, weighed, and left in place for the experiment. Injections for all experiments began between 7:30 and 8:30 AM. Thirty minutes after injection all rats were decapitated rapidly after removal from the home cage and trunk blood was collected and allowed to clot on ice. Sera were then isolated by centrifugation (3000g for 15 min) and frozen at 80°C until assayed.

Estrous Cycle Monitoring. Estrous cycle stage was monitored by analysis of cell types in vaginal lavages. Daily vaginal lavages of cycling and ovariectomized female rats were collected in the morning for at least eight consecutive days and allowed to dry on microscope slides. Slides were then fixed with ethanol and stained with toluidine blue. Identification of cell types was made microscopically according to published methods (Long and Evans, 1922). Only rats that could be reliably staged for estrous cycle stage were used. Two ovariectomized females were disqualified from the study because histological analysis showed evidence of estrous cycling (Cooper et al., 1993).

Hormone Measurements. The serum concentrations of ACTH, corticosterone, estradiol, and progesterone were measured by radioimmunoassay (RIA) in all rats. Testosterone concentrations were assayed only in male rats. The RIA for each hormone used kits from Diagnostic Products Corporation (Los Angeles, CA). Double-antibody kits were used for ACTH and estradiol. Inter- and intra-assay coefficients of variation for all the hormone assays were determined and ranged from 4.2 to 10.0%.

Drugs. Cocaine hydrochloride was provided by Research Triangle Institute through an arrangement with National Institute on Drug Abuse. Cocaine solutions were prepared the day of the experiment in physiological saline. Cocaine and the saline vehicle were administered i.p. at a volume of 1 ml/kg.

Data Analysis and Statistics. RIA standard curves were analyzed and unknown hormone concentrations were determined by nonlinear regression using Prism 3.02 (GraphPad Software, Inc., San Diego, CA). Two-factor ANOVA was used to determine effects of estrous cycle and cocaine and their interactions on hormone concentrations. To determine the main effect of estrous cycle only the data from the four estrous stages were used. To determine the effect of sex, data from the estrous stages were combined and compared with those from males using a two-factor ANOVA [sex by dose (cocaine or saline)]. To determine the effect of ovariectomy, data from the estrous stages were combined and compared with those from ovariectomized females using a two-factor ANOVA [surgical status (sham or ovariectomy) by dose (cocaine or saline)].

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Estrous Cycle, Ovariectomy, and Cocaine on HPA-Axis and Ovarian Hormones

ACTH. The effects of saline or cocaine injection (15 mg/kg) on serum concentrations of ACTH in cycling and ovariectomized females, and intact male rats are shown in Fig. 1A. The range of cocaine effects on ACTH was from more than a 5-fold stimulation compared with saline in proestrous females to 2-fold in males and ovariectomized females. When data from all cycling females were summed and compared with males, ACTH concentrations were significantly greater in females than males [F(1,114) = 4.04, p < 0.05]. Cocaine increased ACTH in both sexes [F(1,114) = 14.33, p < 0.001]. Estrous cycle significantly affected serum ACTH in saline- and cocaine-treated rats together [F(3,85) = 6.0, p < 0.001]. Cocaine stimulated ACTH secretion most in proestrous females {cycle by treatment interaction [F(3,85) = 3.6, p < 0.02]}. ACTH concentrations were less in ovariectomized than all sham-operated, cycling females together [F(1,106) = 3.98, p < 0.05]. The average ACTH concentrations in all cycling female and ovariectomized rats administered cocaine were 161 ± 22 versus 89 ± 17 pg/ml, respectively.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Sex differences, estrous cycle, and ovariectomy effects on cocaine-stimulated ACTH (A) and corticosterone (B) concentrations in rat serum. Intact males, ovariectomized females, and females at each estrous cycle stage were injected with the saline vehicle or 15 mg/kg cocaine i.p. 30 min before blood collection (N = 8-15 rats/group). di 1, di 2, and pro are diestrus 1 and 2 and proestrus, respectively. ANOVA indicated main effects of sex, ovariectomy, and estrous cycle (under Results). The inset graph depicts the  of each group in B (cocaine  saline). Significantly different than diestrous 2 and estrous females. *Significantly different than all other groups. #Significantly different than all other groups except diestrous 2 and estrous rats treated with cocaine.

Corticosterone. Figure 1B shows the effects of saline or cocaine injection (15 mg/kg) on serum concentrations of corticosterone in the same rats as in Fig. 1A. Cocaine administration increased serum corticosterone in all groups of rats [F(1,135) = 103, p < 0.001]. Cocaine increased corticosterone more in cycling females than males [sex and sex by cocaine (p values < 0.007)]. Likewise, ovariectomy attenuated cocaine stimulation of corticosterone secretion [F(1,107) = 10.2, p < 0.002]. Estrous cycle significantly affected serum corticosterone in cycling females (p < 0.001). Unlike its effect on ACTH, cocaine increased corticosterone similarly at each cycle stage.

Estradiol. Serum concentrations of the ovarian hormones estradiol and progesterone were also determined in the rats from Fig. 1. As expected a profound surge in serum estradiol was observed on the morning of proestrus (Fig. 2A). Estradiol concentrations varied across the phases of the estrous cycle [F(3,82) = 80.3, p < 0.001]. Cocaine did not affect estradiol concentrations in cycling female rats (p = 0.19). Estradiol concentrations were significantly greater in intact females relative to both ovariectomized females and males (21 ± 3, 3 ± 1, 6 ± 1 pg/ml, respectively, p values < 0.001).

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Estradiol and progesterone concentrations in serum of cycling female, ovariectomized, and intact male rats. Estradiol (A) and progesterone (B) were determined in the same samples in Fig. 1. Estradiol varied across the estrous cycle, was significantly greatest during proestrus, and was unaffected by cocaine. Progesterone also varied across the cycle but was higher in cocaine-treated rats. The inset graph depicts the  of each group in B (cocaine  saline). *Greater than diestrous 1, male and ovariectomized rats, p < 0.05.

Progesterone. Cocaine dramatically increased progesterone concentrations in cycling female rats and modestly in ovariectomized females and intact males (Fig. 2B). Progesterone concentrations were highest during diestrus in saline-injected rats and cocaine stimulated progesterone secretion most in proestrous females. Estrous cycle significantly affected progesterone concentrations [F(3,85) = 4.6, p < 0.005] and cocaine increased progesterone concentrations in cycling females [F(1,85) = 32.8, p < 0.001]. Progesterone concentrations were higher in cycling females than males (p < 0.001) and ovariectomy significantly decreased cocaine-stimulated progesterone secretion (p < 0.001). Cocaine significantly increased progesterone concentrations in the ovariectomized females and males (p values < 0.01 by t tests).

Cocaine Dose-Response Effects on HPA-Axis Hormones and Testosterone in Male Rats. The neuroendocrine effects of cocaine were further investigated in male rats. Table 1 indicates that 30 min postinjection none of the three cocaine doses altered serum testosterone or estradiol. Cocaine increased serum progesterone in males in a dose-related manner [F(3,97) = 11.4, p < 0.001]. Progesterone values in males injected with 10 or 20 mg/kg cocaine were greater that those of the saline control and 5-mg/kg groups (p < 0.05). Cocaine increased serum corticosterone in a similar dose-related manner [F(3,103) = 4.1, p < 0.009], except that only 20 mg/kg cocaine induced significant increases above those of the vehicle and 5-mg/kg groups (p < 0.05).

Effect of Adrenalectomy

ACTH. Because the preceding experiment showed that cocaine-stimulated progesterone secretion was greatest during proestrus, all rats in this experiment were injected on the morning of proestrus to look for attenuation of this cocaine effect. Figure 3A shows that ACTH concentrations were markedly elevated in adrenalectomized rats. The loss of corticosterone negative feedback on CRF neurons in the hypothalamus led to 14-fold greater serum ACTH concentrations in adrenalectomized relative to sham-operated females following saline injection. Thus, surgery significantly affected ACTH (p < 0.001). Cocaine increased serum ACTH in shams [F(1,18) = 7.86, p < 0.02]. Cocaine did not increase ACTH in the adrenalectomized females, probably because secretion was already either maximal or near-maximal.

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Effects of adrenalectomy and cocaine on the HPA axis and ovarian hormones. Adrenalectomized (ADX) and sham-adrenalectomized female rats at proestrus were injected with saline or 15 mg/kg cocaine i.p. and killed 30 min later (N = 8-10 rats/group). *Different than sham females given cocaine, p < 0.05; #all adrenalectomized females different from all sham females, p < 0.05.

Corticosterone. Figure 3B shows that serum corticosterone was significantly lower in adrenalectomized female rats [F(1,34) = 51.9, p < 0.001]. Cocaine significantly increased serum corticosterone overall [F(1,34) = 12.8, p < 0.002]. Cocaine did not affect corticosterone in adrenalectomized rats as indicated by a significant interaction of surgery and dose [F(1,34) = 12.8, p < 0.002]. As expected, the adrenals mediate the effect of cocaine on serum corticosterone.

Progesterone. Figure 3C indicates that cocaine increased progesterone in sham but not in adrenalectomized females. Cocaine significantly increased serum progesterone [F(1,36) = 5.33, p < 0.03]. Unlike corticosterone, the adrenals are not the only source of progesterone and thus adrenalectomy did not significantly alter basal serum progesterone concentrations (p = 0.19).

Estradiol. Figure 3D shows that estradiol concentrations were significantly decreased in the sera of adrenalectomized females on the morning of proestrus [F(1,36) = 13.0, p < 0.001]. In contrast to the effect on progesterone, cocaine did not affect estradiol concentrations (p = 0.10).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present studies show that the HPA response to cocaine is strongly influenced by activational effects of ovarian hormones. Estrous cycle stage influenced ACTH responses to cocaine and ovariectomy attenuated serum concentrations of ACTH and corticosterone following cocaine. These studies also revealed an unexpected finding that cocaine stimulated secretion of progesterone in both male and female rats. Cocaine-stimulated progesterone secretion was estrogen responsive because it was greatest during the high-estrogen state of proestrus and was attenuated by ovariectomy. The source of cocaine-stimulated progesterone secretion was shown to be the adrenal gland because adrenalectomy completely abolished cocaine-stimulated progesterone secretion.

Sex, Cycle, and Ovariectomy Effects on ACTH and Corticosterone. The presently reported sex differences in cocaine-stimulated ACTH and corticosterone are similar to previous reports of sex differences in basal HPA activity (Lesniewska et al., 1990; Chisari et al., 1995) and cocaine-stimulated ACTH secretion (Kuhn and Francis, 1996). In rats, HPA axis responses to many stimuli, including stressors and drugs, are sexually dimorphic with females exhibiting greater responses than males (Le Mevel et al., 1979; Aloisi et al., 1994; Handa et al., 1994a; Rivier, 1994; Ogilvie and Rivier, 1997). However, the endocrine basis of these differences is not always clear.

Mediation of Cocaine-Stimulated Progesterone Secretion. Cocaine-stimulated progesterone secretion was an unexpected finding of this study. This effect was more robust in intact than ovariectomized females or males. Cocaine-stimulated progesterone secretion (or  progesterone) was greater when estradiol peaked at proestrus than when estradiol was low at diestrus 1, in ovariectomized and male rats. These results implicate ovarian hormones in this phenomenon. However, the presence of cocaine-stimulated progesterone secretion in ovariectomized and male rats led us to investigate the adrenal as the site of origin.