Estradiol: a key biological substrate mediating the response to cocaine in female rats

Abstract

A consistent finding in drug abuse research is that males and females show differences in their response to drugs of abuse. In women, increased plasma estradiol is associated with increased vulnerability to the psychostimulant and reinforcing effects of drugs of abuse. Our laboratory has focused on the role of estradiol in modulating the response to cocaine. We have seen that ovariectomy increases the locomotor response to a single cocaine injection, whereas estradiol exacerbates the locomotor response to repeated cocaine administration. Cocaine-induced sensitization of brain activity, as measured by fMRI, is also dependent on plasma estradiol. Moreover, we observed that although all ovariectomized rats show conditioned place preference to cocaine, it is more robust in ovariectomized rats with estradiol. Opioid receptors are enriched in brain regions associated with pleasure and reward. We find that in females, the effectiveness of kappa opioid agonists in decreasing the locomotor response to repeated cocaine varies with plasma estradiol. We also find that estradiol regulates the density of mu opioid receptors in brains areas associated with reward. These data hint that in females, estradiol modulates the behavioral effects of cocaine by regulating mu and kappa opioid signaling in mesocorticolimbic brain structures. Identifying the mechanisms that mediate differences in vulnerability to drugs of abuse may lead to effective therapeutic strategies for the treatment and prevention of addiction and relapse. We encourage health practitioners treating persons addicted to drugs to consider gender differences in response to particular pharmacotherapies, as well the sex steroid milieu of the patient.

Figure 1. Sex differences in the locomotor response to cocaine of adult rats

Female rats show higher cocaine-induced horizontal activity than male rats at day 1 (A) and at day 13 (B) (* p<0.05). Data presented as the means ± S.E.M.; n=30–69 per group. Methods: Rats were habituated to the LMA chamber (AccuScan™ Instruments, Columbus, OH) for 60 min one day prior to experiments. During each testing session, LMA is recorded for 90 min: 30 min prior to injection and 60 min after saline or cocaine (15 mg/kg) injection. For our repeated cocaine experiments, rats are injected for 5 consecutive days, and on day 13, with saline or cocaine (15 mg/kg, ip). Locomotor activity was recorded on days 1, 5 and 13. Data were analyzed using a Repeated Measures Anova with sex as the categorical factor.

Figure 2. Locomotor response to cocaine of male and female prepubertal rats

No sex differences were observed in cocaine-induced horizontal activity among prepubertal rats at day 1 (A) or at day 13 (B) (p>0.05). Data expressed as means ± S.E.M.; n=30–69 per group. Methods: Pregnant rats were housed in pairs. After birth, each litter was culled to 10 pups, 5 males and 5 females. Each dam was housed individually with its litter and left undisturbed. At weaning (day 21), pups were housed in groups of 5 of the same sex. Behavioral testing began on day 22 and ended on day 35. Females were examined daily for puberty acquisition by checking for vaginal opening. The protocol followed for sensitization was the same as described in Fig.1. Data were analyzed using a Repeated Measures Anova with sex as the categorical factor.

Figure 3. Blood oxygen level dependent signal in response to acute cocaine in male and female rats

No sex differences were observed in cocaine-induced blood oxygen level dependent signal (BOLD) activation in the dorsal striatum (dSTR), nucleus accumbens (NAcc), and medial prefrontal cortex (mPFC) of gonadally intact male and female rats (p>0.05). Data expressed as the mean number of positive BOLD voxels ± S.E.M. Methods: Rats were acclimated to the functional magnetic resonance imager (fMRI) for 3 days. On day 4, animals were placed in the fMRI and anatomical images were obtained. Cocaine was administered into the lateral cerebral ventricle (20ug/10ul) and functional images collected for 10 min. Data were analyzed using Students’s T-test with sex as the categorical factor

Figure 4. Cocaine-induced locomotor activity in ovariectomized rats with and without estradiol treatment

Estradiol decreased cocaine-induced LMA in OVX rats (p<0.05). (A) Time course of horizontal activity comparing OVX and OVX-EB rats receiving saline or cocaine (15mg/kg) injections. (B) Total horizontal activity showing that OVX-EB rats have reduced cocaine-induced LMA compared to OVX rats. Data expressed as mean ± S.E.M.; n=30–69 per group. Methods: The protocol to measure LMA is the same as described in Figure 1. (A) Data were analyzed with a two-way ANOVA using hormone (OVX vs. OVX-EB) and treatment (saline or cocaine) as independent variables and total horizontal activity as the dependent variable. (B) Data were analyzed using a Repeated Measures Anova with hormone and drug as the independent variables, Tukey test was used for posthoc comparisons (OVX+Coc vs. OVX-EB+Coc, * p>0.05).

Figure 5. Blood oxygen level dependent (BOLD) signal response to acute and repeated cocaine in ovariectomized females without and with estradiol

Left: Anatomical and functional overlay images show selected regions of positive BOLD changes (day 1 vs day 13) following cocaine administration. Scale bar hue to the right indicates threshold level for BOLD signal changes. Right: Percent BOLD signal change (mean ± S.E.M.) OVX-EB rats show lower BOLD signal to a single cocaine injection than OVX rats. However, increases in BOLD signal after repeated cocaine administration were observed only in OVX-EB rats (p < 0.05). Data were analyzed with a two-way ANOVA (significance at p < 0.05), with a Bonferroni posthoc test. Symbol * denotes difference between OVX-EB acute vs repeated treatments and ** difference between OVX and OVX-EB. Data included in this figure have been adapted from a previous publication (Febo et al., 2005, J Neurosci 25:1132)

Fig 6. Locomotor activity in ovariectomized rats with and without estradiol treatment after repeated cocaine administration

Estradiol (EB) enhanced cocaine-induced behavioral sensitization in ovariectomized (OVX) rats. Time course of horizontal activity comparing Day 1 and Day 5 of OVX (A) and OVX-EB (B) rats receiving saline or cocaine (15mg/kg) injections. OVX-EB rats show a higher percent change in cocaine-induced sensitization (C) (*p<0.05). Data were expressed as means ± S.E.M. Methods: The protocol to measure LMA is the same as described in Figure 1. Data in A and B were analyzed with a Repeated Measures ANOVA; Data in C were analyzed with a one-way ANOVA using hormone as the independent variable and per cent change in locomotor activity as the dependent variable.

Fig 7. Cocaine-induced conditioned place preference in ovariectomized rats with and without estradiol

The time spent in the chamber associated with cocaine was increased in all rats treated previously with cocaine, however, the increase was higher in ovariectomized rats treated with estradiol. Values expressed as means ± SEM, n=8–16 per group. Methods: During 4 days, rats were injected twice daily, with saline or with saline or cocaine (15 mg/kg; 4 hrs between injections). They were confined to the chamber they were injected for 30 min. The day prior to injections (pre-conditioning) and the day after the last injection (post-conditioning), rats were allowed to roam freely between the 2 chambers. The time spent in each chamber was recorded. Data were analyzed with a Repeated Measures ANOVA: Pre vs post-conditioning: * and + indicate significantly different (p<0.05).

Figure 8. Cocaine-induced horizontal activity in OVX and OVX-EB rats pretreated with the kappa opioid agonist U-69593

U-69593 decreased cocaine-induced horizontal activity on day 5 in OVX-EB rats. Data are presented as means ± SEM; n=8–19 per group. Methods: For 5 days, OVX and OVX-EB rats were injected with U-69593 (0.32mg/kg, sc) or with vehicle (25% propylene glycol) followed 15 min later by a saline or cocaine (15mg/kg, ip) injection. Locomotor activity was recorded as described in Fig 1. Approximately 8 OVX and 8 OVX-EB rats were analyzed per group. Data were analyzed by a Two Way ANOVA *** p<0.05 with hormone and kappa ligand used as the independent variables. Data included in this figure have been adapted from a previous publication (Puig-Ramos et al., 2008, Behav Neurosci 122: 151).

Fig 9. Mu opioid receptor density of ovariectomized rats with and without estradiol treatment

Mu opioid receptor density (MOPr) density was lower in the nucleus accumbens of OVX-EB compared to OVX rats. (A) Representative autoradiograms showing [3H]DAMGO MOPr density in OVX and OVX-EB rats. Scale bar shows color for different MOPr densities (femtomol per gram wet tissue weight). (B) Sampling sites within the subregions of the nucleus accumbens. DM, dorsomedial; DL, dorsolateral; VM, ventromedial. C) [3H]DAMGO densities for the three brain subregions in OVX and OVX-EB rats. Data were presented as the means ± SEM and analyzed with Student’s T-test, (* p<0.05). Values represented in panel C were obtained by averaging 3–4 different anterior-posterior sections per animal, n=8–19 per group.