Adolescent cocaine and injection stress effects on the estrous cycle

Physiology & Behavior 70 (2000) 417 ± 424

Adolescent cocaine and injection stress effects on the estrous cycle D.K. Raapa,*, B. Morinb, C.N. Medicib, R.F. Smithb a

Department of Psychology, University of Alaska-Fairbanks, PO Box 756480, Fairbanks, AK 99775-6480, USA b Department of Psychology, George Mason University, 4400 University Drive, Fairfax, VA 22030, USA Received 5 October 1999; received in revised form 22 February 2000; accepted 22 March 2000

Abstract Chronic cocaine exposure during critical periods of development induces short- and long-term effects. During the pubertal period, the hypothalamic ± pituitary ± gonadal (HPG) axis undergoes many dynamic changes. The present study investigated whether chronic periadolescent cocaine alters reproductive maturity in the rat. Sixty female Long ± Evans hooded rats were randomly assigned to one of three conditions (20 mg cocaine/kg/day, saline injected and uninjected), for dosing from postnatal day 21 (P21) through P60. Several indicators of reproductive maturation and functioning were assessed during and following treatment. Cocaine exposure had no effect on the onset of puberty or on the date of first ovulation. The number of proestrus ± estrus transitions was significantly lower in cocaine-exposed females compared to uninjected females, but not compared to saline-injected controls. This reduction was observed during exposure to cocaine, as well as after the cessation of injections. During the dosing period, cocaine-exposed rats also exhibited a greater number of cycles that had no clear P ± E transition than did UN subjects; this effect disappeared once injections stopped. These alterations suggest immediate, and possibly persisting, alterations in the control of ovulation after chronic cocaine exposure throughout adolescence. Interestingly, during the injection period, the saline-injected females had a significantly greater number of diestrus days compared to uninjected and cocaine-injected animals, as well as a lower proportion of regular 4- and 5-day cycles. These differences disappeared once injections stopped. These results suggest a stress-induced irregularity of the estrous cycle, possibly attenuated by cocaine and recoverable after exposure. The present findings indicate that the HPG axis is susceptible to short-term, and possibly to long-term, alterations induced by cocaine exposure throughout the adolescent period. D 2000 Elsevier Science Inc. All rights reserved. Keywords: Female; Puberty; Ovulation; HPG axis; Drug abuse; Development

1. Introduction Early cocaine exposure in humans and in rats produces effects on neurobehavioral development that extend well beyond the immediate exposure period. Prenatal exposure to cocaine induces functional, cognitive, morphological, and neurochemical alterations in offspring, many of which persist into adulthood [1 ± 7]. Because many aspects of central nervous system (CNS) development continue postnatally, especially in the rat, it is not surprising that cocaine exposure during the immediate and later postnatal period also produces long-lasting detrimental effects [8,9]. Few studies have investigated the effects of cocaine during childhood and adolescence, despite reports of high usage levels among human adolescents [10,11]. The peria-

* Corresponding author. Tel.: +1-907-474-5781. E-mail address: [email protected] (D.K. Raap).

dolescent period is marked by significant CNS growth and maturation. Both functional and morphological neuronal changes continue to occur throughout the pubertal period in rats (see Ref. [12] for extensive review). For example, puberty is marked by continued dendritic elaboration, an increase in receptor content and number of spiny cells, and continued development of synaptic connections [12]. These developmental changes during the pubertal period are also well documented in humans [13,14]. Plasticity of the CNS during the peripubertal period is evident in extensive animal research on environmental manipulation. For example, rats reared from weaning (i.e., approximately P22) through approximately P60 in environmentally complex housing exhibit a variety of neuromorphological changes compared to animals reared in an impoverished environment (e.g., [15,16]). Few reports, however, have examined long-term alterations after chronic ``internal'' manipulations (e.g., drug exposure) during puberty. It is clear that periadolescent administration of abusable

0031-9384/00/$ ± see front matter D 2000 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 1 - 9 3 8 4 ( 0 0 ) 0 0 2 8 7 - 0

418

D.K. Raap et al. / Physiology & Behavior 70 (2000) 417±424

drugs may induce some neurochemical changes similar to those seen in adults (e.g., Ref. [17]). Data from our laboratory indicate that chronic cocaine injections (20 mg/ kg/day) from P21± P60 in rats induces alterations in weight gain, open-field activity, and a complex learning task [18]. This may also be true of other abusable drugs, as Levin [19] has reported long-term changes in behavior after adolescent nicotine administration, and we [20] reported long-term behavioral changes after adolescent alcohol. Sustained behavioral changes after periadolescent drug administration suggest that such exposure may induce lasting changes in the structure and/or neurochemistry of the CNS. Such changes may also be related to differential susceptibility of the adolescent CNS to abusable drugs, as the work of Swartzwelder et al. [21 ± 23] and others have shown. During puberty, a complex pattern of changes occurs within the CNS, leading to the development of mature reproductive competence. Puberty marks the final and crucial stage in the development of a system regulated by complex interactions between aminergic neurotransmitter systems (e.g., dopamine, norepinephrine), gonadal steroids (e.g., androgens, estrogen, progesterone), and hypothalamic ± pituitary hormones (luteinizing hormone, luteinizing hormone-releasing hormone, follicle stimulating hormone) [24]. Without the coordinated activity of each of the components of the hypothalamic ± pituitary ± gonadal (HPG) axis during puberty, alterations in later reproductive functioning are highly probable [25]. Although the initial morphology and connectivity of the HPG axis occurs prenatally, the peripubertal period marks a highly sensitive time in the maturation of this system. Peripubertal changes in morphology and function of existing networks render the female reproductively competent (see Ref. [12] for review). Cocaine can affect components of the complex cascade required for reproductive maturation and adult functioning. For example, changes in LHRH secretion are important in the onset of puberty [26], and the normal stimulatory effect of NE on LHRH release is attenuated by the administration of cocaine [27]. In mature animals, cocaine severely disrupts the regularity of the estrous cycle, such as by decreasing ovulation rate in adult female rats [28]. King et al. [28] hypothesized that cocaine induces female reproductive dysfunction, in part, by interrupting the pulsatile release of LHRH from the hypothalamus. Cocores et al. [29] reported that chronic cocaine abuse in humans was associated with menstrual cycle irregularities, and a similar trend was noted in rhesus monkeys [30]. Cocaine's potential to interfere with normal development of the HPG axis is also shown by studies of exposure during the prenatal period. Prenatal exposure to cocaine has been shown to attenuate DA activity during the neonatal period [7] and alter DA receptor binding [7,31]. Prenatal cocaine affects neonatal prolactin (PRL) and GH levels [32], and LH regulation in adult offspring [33]. Vathy et al. [34] reported

reduced sexual behaviors in adult female offspring prenatally exposed from G11± G18 to 20 mg/kg cocaine. Other reports have indicated an increase in basal LH levels, a decrease in scent-marking behavior, and decreased sexual behavior as indicated by a higher latency to initiate a mount and lower numbers of intromissions among adult male offspring prenatally exposed to 20 mg/kg cocaine from G15 ± G20 [4]. Cutler et al. [35] reported that prenatal dosing with 215 mg/kg/day resulted in reduced anogenital distance in male offspring. In addition to these effects of prenatal cocaine on HPG development, acute cocaine may affect puberty onset in mice [36]. Reproductive maturation in the female is a complex cascade of events, many aspects of which are potentially vulnerable to cocaine. As an initial assessment of whether periadolescent cocaine disrupts the achievement of reproductive maturity, several measures of maturity and the achievement of normal estrous cyclicity was examined. This assessment of cyclicity was continued for a comparable period of time following dosing to determine whether cocaine effects seen during dosing persist or abate following the cessation of exposure. 2. Materials and methods 2.1. Subjects Female offspring of 20 female Long ± Evans hooded rats (Harlan Laboratories, Indianapolis, IN) bred in our laboratory were used as subjects. Offspring remained with their biological dams in breeding boxes with hardwood chip bedding until they were weaned on postnatal day 21 (P21). Animals were maintained on a 12:12 light/dark cycle (lights on 07:00 ±19:00 hours) in a temperature-regulated room (72 ± 78°F). Food and water were available ad lib at all times. 2.2. Procedures 2.2.1. Weaning On P21, offspring were eartagged and transferred to individual hanging wire-mesh cages. One female from each of the 20 litters was then randomly assigned to each of three conditions Ð uninjected, saline injected, or 20 mg/kg cocaine. Thus, there were 20 subjects per Table 1 Mean ( ‹ SEM) age of puberty onset, age of first ovulation, and number of days the first ovulation was delayed from onset (n = 20/group)

Uninjected Saline

Age of puberty onset (PND) 36.6 ‹ 0.6 Age of first ovulation (PND) 37.2 ‹ 0.6 Days delay: first ovulation from onset 0.6 ‹ 0.3 No significant differences.

36.0 ‹ 0.5 37.4 ‹ 0.7 1.4 ‹ 0.5

Cocaine 37.1 ‹ 0.6 38.2 ‹ 0.7 1.1 ‹ 0.4

D.K. Raap et al. / Physiology & Behavior 70 (2000) 417±424

419

Fig. 1. Mean proportion of regular 4- and 5-day cycles by condition for each age (n = 20 per condition for all figures). Asterisk denotes significant different from uninjected controls.

group, and each subject had one sibling in each of the other two conditions.

dose has been shown to have no significant effect on body weight in periadolescent females [18].

2.2.2. Cocaine dosing Solutions in a strength of 10 mg cocaine HCl (NIDA Drug Supply System)/3 ml 0.9% saline were prepared daily immediately prior to dosing. Subcutaneous (s.c.) injections at a dose of 20 mg/kg/day cocaine or an equal volume of isotonic saline (0.9% NaCl solution) were administered once daily between 09:00 and 12:00 hours from P21 ±P60 with a 1/2-in., 26-gauge needle under the skin of the flank, alternating flanks each day to minimize skin necrosis. Uninjected animals were handled daily to equate their handling with the injected animals. The uninjected and saline-injected offspring served as the only control groups; pair-fed/injected controls were not used because a 20-mg/kg

2.2.3. Onset of puberty Each animal was assessed daily (10:00± 12:00 hours) beginning at P25 for the date of vaginal opening, designated as the postnatal day of onset [12]. 2.2.4. First ovulation Once vaginal opening occurred, each animal was assessed by vaginal smear cytology using light microscopy for indications of the first preovulatory surge of gonadotropins (i.e., for the first presence of nonleukocytic cells). First ovulation was expressed as both age (i.e., postnatal day of occurrence) and as delay from onset (i.e., number of days from onset).

Fig. 2. Mean number of days spent in diestrus for each age by condition. Asterisk denotes significant difference from uninjected and cocaine dosed animals, which do not differ from each other.

420

D.K. Raap et al. / Physiology & Behavior 70 (2000) 417±424

Fig. 3. Mean number of proestrus ± estrus transitions by condition for each. Asterisk denotes significant difference from uninjected controls.

2.2.5. Estrous cyclicity The estrous cycle of each female offspring was monitored daily (09:00± 11:00 hours) from P41 through P80; hence, monitoring included 20 days during injections (P41 ± P60) and 20 days postinjection (P61 ±P80). The stage of cyclicity was determined, using light microscopy, by the relative presence of cornified cells (C cells), intermediate cells (N cells), and/or leukocytes (L cells) in vaginal smears. Cycles were then coded using a slight modification of the protocol described in detail by LeFevre and McClintock [37]. Briefly, samples containing only C and/or N cells were coded with an E; all leukocytic samples were coded with an L. Regular 4- and 5-day cycles included cycles containing only 2 E days and 2 or 3 L days. An ``orderly index'' was obtained by comparing the number of regular 4- and 5-day cycles to the total number of 4- and 5-day cycles within each of the two time periods (P41 ±P60 and P61 ± P80). Secondly, ``frequency of ovulation'' measures were obtained by de-

termining (1) the number of ``obvious'' (i.e., the occurrence of 2 consecutive E days where the first E day contained a majority of N cells and the second E day contained a majority of C cells) proestrus ± estrus transitions within each observation period; and (2) the number of times an assumption of ovulation had to be made that was not ``obvious'' (i.e., at the onset of each cycle, defined by a transition from L to E). The latter differs from the former: The ``assumptions'' are the total number of cycles in the observation period minus the number of obvious transitions. 2.3. Data analysis Effects of treatment on body weight (at each of P30, P40, P50, P60, P70, and P80), onset of puberty (expressed as postnatal day of vaginal opening), first ovulation age (expressed as postnatal day at which first nonleukocytic sample appeared), and first ovulation delay from onset (expressed

Fig. 4. Mean number of proestrus ± estrus transition assumptions that were not obvious, by condition. Asterisk indicates significant differences from uninjected controls.

D.K. Raap et al. / Physiology & Behavior 70 (2000) 417±424

421

Table 2 Mean ( ‹ SEM) body weight during the dosing period (n = 20/group) Postnatal age (days) Dose group

21

30

40

50

60

70

80

Uninjected Saline Cocaine

42.2 ‹ 0.67 42.0 ‹ 0.57 41.7 ‹ 0.56

80.5 ‹ 1.20 79.3 ‹ 1.2 77.3 ‹ 1.2

141.3 ‹ 2.2 138.7 ‹ 2.1 135.0 ‹ 1.8

184.8 ‹ 3.1 178.9 ‹ 2.8 174.2 ‹ 2.5

212.1 ‹ 3.1 207.7 ‹ 2.7 202.0 ‹ 3.7

235.9 ‹ 3.3 235.8 ‹ 3.8 233.4 ‹ 3.7

249.0 ‹ 2.8 249.7 ‹ 5.0 250.9 ‹ 4.4

No significant differences.

as number of days separating onset and first ovulation) were analyzed using one-way analyses of variance (ANOVAs) followed by planned comparisons. Estrous cyclicity-orderly index and estrous cyclicityfrequency of ovulation were analyzed using two-way [dose  (3)  age (2)] ANOVAs followed by simple effects analyses and planned comparisons. 3. Results 3.1. Puberty onset and first ovulation Means and standard errors of the mean (SEM) for onset of puberty, first ovulation age and first ovulation delay from onset are presented in Table 1. One-way ANOVAs revealed no significant differences between groups for onset of puberty, for first ovulation age, or for first ovulation delay from onset. 3.2. Estrous cycle-orderly index Saline animals had a significantly lower proportion of regular cycles during P41 ± P60 (i.e., during exposure) compared to uninjected animals; this effect was limited to the actual exposure period, F(2, 57) = 3.93, p < 0.05 (see Fig. 1). Scheffe analysis indicated that saline animals had a lower proportion than UN subjects. The mean proportion of ``regular'' cycles in the cocaine-injected animals was not significantly different from either the uninjected animals or the saline-injected animals at either time. Furthermore, saline animals spent significantly more days in diestrus than uninjected or cocaine-treated animals during the injection period, F(2, 57) = 4.50, p < 0.05); this difference disappeared once injections stopped (Fig. 2). 3.3. Estrous cycle-frequency of ovulation Cocaine-treated females had a significantly lower number of P ± E transitions compared to uninjected females both during the injection period and after injections stopped, F(2, 57) = 5.54, p < 0.01 (see Fig. 3), but did not differ from the saline-injected controls. Saline-treated animals did not differ from uninjected controls at either time. Upon further examination of the daily records, welldefined P± E transitions were not apparent in certain animals

that were otherwise exhibiting signs of cycling. In such instances, it was inferred, or ``assumed,'' that ovulation was occurring despite the lack of a clear P ± E transition. Hence, an assessment was made of group and age differences in the number of times an assumed ovulation occurred (i.e., a cycle occurred, but with no clear P± E transition). ANOVA, F(2, 57) = 3.50, p < 0.05, revealed that during exposure, cocaineinjected animals required significantly more such assumptions that were not obvious than the saline-treated rats (Fig. 4). After exposure (i.e., P61± P80), no significant group differences were present. 3.4. Body weight Table 2 shows the mean body weights and SEM for P21, P30, P40, P50, P60, P70, and P80. An ANOVA at each age revealed no significant differences (F < 1). At all ages, cocaine-treated rats were within 5.5% of the uninjected group, and were usually within 3%. 4. Discussion Cocaine injection during adolescence produced alterations in estrous cyclicity that persisted after injections stopped, while saline injections induced transient alterations that differed from the cocaine effects. The cocaine-induced reduction in proestrus ±estrus transitions and increase in the number of cycles lacking a clear P ±E transition indicate a disruption in ovulatory control. In contrast, saline injections induced short-term disruptions in estrous cyclicity, as indicated by longer periods of diestrus and fewer regular cycles. Neither cocaine exposure nor saline injections from P21± P60 had an effect on the onset of puberty, the age of first ovulation, or the delay of the first ovulation from the onset of puberty. 4.1. Effects of saline injections Saline injections induced a different pattern of disruptions in the estrous cycle than that induced by cocaine exposure. Saline-injected controls exhibited significant alterations in two orderly index measures of estrous cyclicity while receiving injections. First, saline-injected animals spent more time than the uninjected controls in the diestrus phase of the cycle during the injection period. Secondly,

422

D.K. Raap et al. / Physiology & Behavior 70 (2000) 417±424

there was a lower proportion of regular 4- and 5-day cycles among saline-injected animals compared to uninjected controls. These abnormalities did not persist past the injection period and were not observed in cocaine-treated animals. Chronic exposure to stress has been shown to cause an increase in basal levels of PRL in rats [38]. The role of PRL in the estrous cycle is to enhance ovarian responsiveness to gonadotropins [39]. However, hyperprolactinemia has been shown to induce reproductive failure; animals were observed to remain in diestrus [39]). Injection stress in the present study may have induced sufficient elevations in PRL to temporarily cause the noted irregularities, but not high enough to induce long-term alterations. 4.2. Differences in effects of cocaine and saline Interestingly, cocaine injections did not have a significant effect on regularity measures compared to the uninjected controls, as the saline injections did. The cocainetreated animals had a nonsignificantly higher proportion of regular cycles and spent significantly fewer days in diestrus than the saline animals, indicating an apparent cocaineinduced attenuation of the saline injection effects. Cocaine effects were thus not the same as the presumably stressrelated effects of a needle injection. The pattern of the apparent attenuation of the injection-induced alterations is consistent with findings in many reports of cocaine attenuated behavioral and neuroendocrine stress responses. For example, it has been shown that repeated cocaine exposure attenuates the rise in corticosterone levels and behavioral alterations induced by stress [40]. Again, a PRL-involved mechanism might be involved with the cocaine-attenuated findings in the current study. Administration of a dopamine agonist has been shown to inhibit the release of PRL [41]. Because cocaine administration enhances the effect of dopamine (see Ref. [40] for a review), levels of PRL are thereby likely depleted. 4.3. Effects of cocaine exposure 4.3.1. Estrous cycle As assessed by the number of proestrus ± estrus (P ±E) transitions, ovulation was altered by cocaine injection during adolescent exposure and after the cessation of injections. Female rats exposed to 20 mg/kg cocaine from weaning through puberty experienced significantly fewer P ±E transitions than uninjected control females both during the injection period (P41 ± P60) and after exposure (P61 ±P80). As COC ±SAL differences did not approach significance in the P61± P80 time frame, we cannot conclude that cocaine results in persisting effects not induced by injection stress itself. However, it is important to note that the pattern of effects induced by cocaine is strikingly different than the pattern of effects induced by saline injection. Indeed, the significant effects of injection stress on cyclicity and the number of days spent in diestrus

during dosing indicates that the injection stress is a manipulation in its own right, not just a control. The pattern of effects in the SAL group did not mimic the pattern in the COC group; indeed, during dosing, there were significant differences between them (see above). Further, if we compare only the COC and SAL groups, we would conclude that COC animals spend significant fewer days in diestrus during dosing than do controls. That is demonstrably incorrect, however, as animals not subjected to injection stress during dosing spent as many days in diestrus as COC animals. Thus, the COC ± SAL comparison is not appropriate as a sole comparison. If we ignore the SAL controls, however, we run the risk of attributing to COC dosing an effect that is partly or mostly due to injection stress. It is clear that during the dosing period, real COC effects exist, as COC subjects differ on at least one measure from both UN and SAL groups. Further, the COC ± UN difference in P± E transitions in the P61 ±P80 period is consistent with COC ± UN differences observed during dosing, while the trend toward a SAL ± UN difference in the later time period is very different than the types of SAL ±UN differences observed during dosing. It is clear that animals injected with cocaine from P21 ± P60 differ from UN animals during the P61 ±P80 period (while SAL animals do not), but it appears that the combination of injection stress and cocaine induces these effects, which persist past the dosing period, while injection stress alone does not have a significant persisting effect. Exploratory analyses revealed that chronic cocaine exposure from P21± P60 also had adverse effects on the number of times an assumption of ovulation had to be made. It is well accepted that a clear P ± E transition in the 4- or 5-day estrous cycle of a rat is an indication of ovulation. There were several cocaine animals that showed signs of cycling (e.g., transitions in and out of diestrus), but did not show a clear P ± E transition. Quantitative analyses revealed that an assumption of ovulation had to be made significantly more times among the females who were exposed to cocaine than among the SAL females. This effect was present during dosing, but was not significant during the postinjection period. These effects are consistent with those reported by Potter et al. [30] during cocaine infusions (4 mg/kg/day) in the rhesus monkey. A majority of cocaine-treated animals exhibited abnormal cycle times (e.g., both short and long cycles were reported) and ovulation failure as determined by laparoscopy. Our results suggest abnormalities that persist past the dosing period; similar investigations have not been conducted in adult monkeys. There are several possible levels of neuroendocrine control that may underlie the cocaine-induced alterations. Mechanisms may include disruption of noradrenergic innervation of the ovary or the hypothalamus [27], dopaminergic [7], or serotonergic [14,42] innervation of the hypothalamus, or hormonal control mechanisms [27].

D.K. Raap et al. / Physiology & Behavior 70 (2000) 417±424

423

4.3.2. Body weight Other studies [18] report a male-limited reduction in adolescent weight gain in animals receiving 20 mg/kg/day cocaine. No statistically significant reduction was seen in the present study, although the cocaine dosed group had a mean body weight as much as 5.5% lower than uninjected subjects (at P50 only). Research has indicated that significant weight loss can adversely affect cyclic reproductive function in rats (King TS, personal communication, August 14, 1995; [43]) and in humans [13]. However, it is highly unlikely that a transient, slight, and temporary weight reduction, which remained within the range of normal, resulted in the cocaine-induced alterations reported here.

rats, primates, and humans. Because the fundamental mechanisms involved in the development and functioning of the CNS and its regulation of reproductive maturity are conserved across species, an animal model provides a means for developing experimental procedures to control for many of the confounds and ethical issues involved in human drugabuse research. Our findings indicate that periadolescent cocaine induces immediate estrous cycle abnormalities, and suggests the potential for persistence of these irregularities. Given the incidence of adolescent substance abuse, further research is needed to define the mechanisms underlying cocaine-induced alterations in reproductive competence.

4.4. Other measures

Acknowledgments

Exposure to cocaine from weaning through puberty had no apparent effect on the timing of the onset of puberty or the timing of the first ovulation. The assessment of these two measures, however, was based solely upon qualitative indicators. The opening of the vaginal canal in female rats indicates only that estrogen levels reached a high enough level for full canalization [12]. Secondly, evaluating the types of cells shed from the vaginal epithelium (i.e., daily vaginal smears) to assess when the first ovulation occurred gives an indication only of whether the hormonal levels were substantial enough to initiate the first preovulatory surge of gonadotropins. These two measures give no quantitative evaluation of the underlying hormonal levels or control mechanisms, only an indication that the dosing regimen used did not sufficiently alter HPG axis control of the end points measured.

This work was submitted by the first author, in partial fulfillment of the requirements of the PhD at George Mason University. This work was supported by NIDA Grant DA09686-01 to RFS.

5. Conclusions Cocaine injections from weaning through puberty and adolescence resulted in immediate and persisting alterations in estrous cyclicity and indicators of ovulation in the female rat, compared to uninjected controls. Injection stress alone appears to be associated with transient irregularities in the estrous cycle that are not evident once the stress is eliminated. The potential for cocaine to interfere with reproductive functioning, and for this interference to persist past the exposure period, has serious implications. Most aspects of reproductive and sexual functioning are controlled either directly or indirectly by the HPG axis. Any alteration in the development or regulation of this system could potentially alter behavioral and/or physiological functioning well into adulthood. Clinical studies have noted menstrual irregularities among adult women who regularly use cocaine, but no research has investigated the potential for long-term consequences of chronic adolescent cocaine use in humans. The findings from the current study add to a growing body of literature indicating developmental consequences of early cocaine exposure in several species, including mice,

References [1] Heyser CJ, Spear NE, Spear LP. Effects of prenatal exposure to cocaine on conditional discrimination learning in adult rats. Behav Neurosci 1992;106:837 ± 45. [2] Hutchings D, Fico TA, Dow-Edwards DL. Prenatal cocaine: maternal toxicity, fetal effects and locomotor activity in rat offspring. Neurotoxicol Teratol 1989;11:65 ± 89. [3] Johns JM, Means MJ, Anderson DR, Means LW, McMillen BA. Prenatal exposure to cocaine: II. Effects on open-field activity and cognitive behavior in Sprague ± Dawley rats. Neurotoxicol Teratol 1992;14:343 ± 9. [4] Raum WJ, McGivern RF, Peterson MA, Shryne JH, Gorski RA. Prenatal inhibition of hypothalamic sex steroid uptake by cocaine: effects on neurobehavioral sexual differentiation in male rats. Dev Brain Res 1990;53:230 ± 6. [5] Smith RF, Mattran KM, Kurkjian MF, Kurtz SL. Alterations in offspring behavior induced by chronic prenatal cocaine dosing. Neurotoxicol Teratol 1989;11:35 ± 8. [6] Sobrian SK, Burton LE, Robinson NL, Ashe WK, James H, Stokes LM, Turner LM. Neurobehavioral and immunological effects of prenatal cocaine exposure in rat. Pharmacol Biochem Behav 1990; 35:617 ± 29. [7] Spear LP, Kirstein CL, Frambes NA. Cocaine effects on the developing central nervous system: behavioral psychopharmacological, and neurochemical studies. In: Hutchings DE, editor. Prenatal abuse of licit and illicit drugs.. Ann N Y Acad Sci, 526. New York: The New York Academy of Sciences, 1989. pp. 280 ± 9. [8] Barron S, Irvine J. Effects of neonatal cocaine exposure on two measures of balance and coordination. Neurotoxicol Teratol 1994;16: 89 ± 94. [9] Dow-Edwards DL, Freed-Malen LA, Hughes HE. Long-term alterations in brain function following cocaine administration during the preweanling period. Dev Brain Res 1993;72:309 ± 13. [10] National Institute on Drug Abuse. National household survey on drug abuse: population estimates 1992. Washington, DC: Superintendent of Documents, US Government Printing Office, 1993. [11] Rouse BA. Trends in cocaine use in the general population. In: Schober SE, Schade CE, editors. The epidemiology of cocaine use and abuse. Natl Inst Drug Abuse Res Monogr, 110. Washing-

424

[12] [13] [14] [15]

[16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27]

D.K. Raap et al. / Physiology & Behavior 70 (2000) 417±424 ton, DC: Superintendent of Documents, US Government Printing Office, 1991. pp. 5 ± 18. Ojeda SR, Urbanski HF. Puberty in the rat. In: Knobil E, Neill JD, editors. The physiology of reproduction, 2nd ed. New York: Raven Press, 1994. pp. 363 ± 410. Goldfein A, Monroe SE. Ovaries. In: Greenspan FS, Baxter JD, editors. Basic and clinical endocrinology 4th ed. Norwalk, CT: Appleton and Lange, 1994. pp. 419 ± 70. Styne DM. Puberty. In: Greenspan FS, Baxter JD, editors. Basic and clinical endocrinology, 4th ed. Norwalk, CT: Appleton and Lange, 1994. pp. 501 ± 24. Greenough WT. Experience effects on the developing and the mature brain: dendritic branching and synaptogenesis. In: Krasnegor NA, Blass EM, Hofer MA, Snotherman WP, editors. Perinatal development: a psychobiological perspective. New York: Academic Press, 1987. pp. 195 ± 219. Greenough WT, West RW, DeVoogd TJ. Subsynaptic plate perforations: changes with age and experience in the rat. Science 1978; 202:1096 ± 8. Philpot RM, Kirstein CL. The effects of repeated alcohol exposure on the neurochemistry of the periadolescent nucleus accumbens septi. NeuroReport 1998;9:1359 ± 63. Smith RF, Medici CN, Raap DK. Enduring behavioral effects of weaning-through-puberty cocaine dosing in the rat. Psychobiology 1999;27:432 ± 7. Levin ED. Persisting effects of chronic adolescent nicotine administration on radial arm maze learning and response to nicotine challenges. Neurotoxicol Teratol 1999;21:338. Siciliano D, Smith RF, Morin L. Juvenile alcohol consumption in the rat induces behavioral changes evident later in adulthood. Soc Neurosci Abstr 1997;23:1864. Swartzwelder JS, Richardson RC, Markwiese-Foerch B, Wilson WA, Little PJ. Developmental differences in the acquisition of tolerance to alcohol. Alcohol 1998;15:311 ± 4. Swartzwelder HS, Wilson WA, Tayyeb MI. Age-dependent inhibition of long-term potentiation in immature vs. mature hippocampus. Alcohol: Clin Exp Res 1995;19:1480 ± 5. Swartzwelder HS, Wilson WA, Tayyeb MI. Differential sensitivity of NMDA receptor-mediated synaptic potentials to ethanol in immature vs. mature hippocampus. Alcohol: Clin Exp Res 1995;19:320 ± 3. Kalra SP, Kalra PS. Neural regulation of luteinizing hormone secretion in the rat. Endocr Rev 1983;4:311 ± 51. Ojeda SR, Dissen GA, Junier MP. Neurotrophic factors and female sexual development. Front Neuroendocrinol 1992;13:120 ± 62. Urbanski HF, Ojeda SR. Activation of luteinizing hormone-releasing hormone release advances the onset of female puberty. Neuroendocrinology 1987;46:273 ± 6. King TS, Schenken RS, Kang IS, Javors MA, Riehl RM. Cocaine disrupts estrous cyclicity and alters the reproductive neuroendocrine axis in the rat. Neuroendocrinology 1990;51:15 ± 22.

[28] King TS, Canez MS, Gaskill S, Javors MA, Schenken RS. Chronic cocaine disruption of estrous cyclicity in the rat: dose-dependent effects. J Pharmacol Exp Ther 1993;264:29 ± 34. [29] Cocores JA, Dackis CA, Gold MS. Sexual dysfunction secondary to cocaine abuse in two patients. J Clin Psychiatry 1986;47:384 ± 5. [30] Potter DA, Luther MF, Eddy CA, Siler-Khodr TM, King TS, Schenken RS. Effects of follicular-phase cocaine administration on menstrual and ovarian cyclicity in rhesus monkeys. Am J Obstet Gynecol 1998;178:118 ± 25. [31] Meyer JS, Shearman LP, Collins LM, Maguire RL. Cocaine binding sites in fetal rat brain: implications for prenatal cocaine action. Psychopharmacology (Berlin) 1993;112:445 ± 51. [32] Smith RF, Royall GD, Coss M. Prenatal cocaine produces dose-dependent suppression of prolactin and growth hormone in neonatal rats. Physiol Behav 1995;58:619 ± 23. [33] Sheppard DK, Smith RF, Chandhoke V, Medici CN. Prenatal cocaine alters basal and estrogen-primed LH levels in female rats. Soc Neurosci Abstr 1994;20:598. [34] Vathy I, Katay L, Mini KN. Sexually dimorphic effects of prenatal cocaine on adult sexual behavior and brain catecholamines in rats. Dev Brain Res 1993;73:115 ± 22. [35] Cutler AR, Wilkerson AE, Gingras JL, Levin ED. Prenatal cocaine and/or nicotine exposure in rats: preliminary findings on long-term cognitive outcome and genital development at birth. Neurotoxicol Teratol 1996;18:635 ± 43. [36] Chen C-J, Vandenbergh JG. Cocaine attenuates puberty acceleration in female house mice. Pharmacol Biochem Behav 1993;44:281 ± 5. [37] LeFevre J, McClintock MK. Reproductive senescence in female rats: a longitudinal study of individual differences in estrous cycles and behavior. Biol Reprod 1988;38:780 ± 9. [38] Van de Kar LD, Richardson-Morton KD, Rittenhouse PA. Stress: neuroendocrine and pharmacological mechanisms. In: Jasmin G, Cantin M, editors. Stress revisited. Methods Achiev Exp Pathol, 14. Basel: Karger, 1991. pp. 133 ± 73. [39] Advis JP, Ojeda SR. Hyperprolactinemia-induced precocious puberty in the female rat: ovarian site of action. Endocrinology 1978;103: 924 ± 35. [40] Levy AD, Rittenhouse PA, Li Q, Yracheta J, Kunimoto K, Van de Kar LD. Influence of repeated cocaine exposure on the endocrine and behavioral responses to stress in rats. Psychopharmacology (Berlin) 1994;113:547 ± 54. [41] Ben-Jonathan N. Dopamine: a prolactin inhibiting hormone. Endocr Rev 1985;6:564 ± 89. [42] Levy AD, Baumann MH, Van de Kar LD. Monoaminergic regulation of neuroendocrine function and its modification by cocaine. Front Neuroendocrinol 1994;15:85 ± 156. [43] Cagampang FRA, Maeda K-I, Yokohama A, Ota K. Effect of food deprivation on the pusatile LH release in the cycling and ovariectomized rat. Horm Metab Res 1990;22:269 ± 72.