Nandrolone alters the behavioral response to cocaine as well as striatal and cortical dopamine receptors of prepubertal male rats

Abstract

Nandrolone, is an anabolic androgenic steroid (AAS) used by adolescents and young adults. Supraphysiologic doses of AAS are correlated with dysfunctions in anxiety and reward. This study examined whether exposure to nandrolone before puberty altered anxiety- and addictive-like behaviors. Dopamine type 2 receptors (D2DR) in the nucleus accumbens (NAc) and medial prefrontal cortex (mPFC) were also analyzed. Beginning on day 28 and ending on day 37, male rats received a daily injection of nandrolone decanoate (20 mg/kg) and were subsequently evaluated for anxiety. Their locomotor response (sensitization) and preference (conditioned place preference (CPP) to cocaine (15 mg/kg) were also assessed. Nandrolone reduced anxiety and ambulation, accelerated the development of sensitization to cocaine, and reduced CPP to cocaine by 27%. Expression of D2DR in the NAc and the PFC of males was increased by nandrolone, whereas treatment with cocaine reduced accumbal D2DR. We hypothesize that nandrolone accelerated the development of the neural circuitry that participates in behavioral sensitization and reduced the rewarding properties of cocaine. The observed increase in accumbal D2DR may have potentially mediated the reduction in anxiety and ambulation and hastened the maturation of the neural circuitry responsible for the sensitized response to cocaine.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-17890-6.

Introduction

Anabolic-androgenic steroids (AAS), derivatives of the gonadal hormone testosterone (T), were developed in the 1950 s to treat men with hypogonadism and delayed puberty¹. The term androgenic comes from their masculinizing properties, anabolic from their promotion of metabolic processes, such as enhanced protein synthesis and erythropoiesis². AAS provide athletes with an edge in training and competition, by increasing muscle strength and endurance and promoting competitive behaviors³. Before the 1990 s, the use of AAS was circumscribed mainly to athletes during training and before competitions⁴. However, during the last two decades, young men have been using AAS all year round to improve their physical appearance^4,5.The use of AAS as self-medication is also on the rise among the female-to-male transgender population⁶.

In the USA, about 3–4 million people have used AAS; of those, approximately 1 million have developed dependence⁷. The worldwide prevalence of AAS use is 3.3%, which is higher for men (6.4%) than for women (1.6%)⁸. AAS users report administering doses that can be more than 100 times the physiological dose⁹. This is of significant concern because AAS can have deleterious side effects, particularly on the cardiovascular system. AAS can also cause hepatic toxicity, decrease fertility, alter secondary sexual characteristics^10,11 and increase the prevalence of psychiatric disorders^12–14. Some symptoms are the result of exposure to AAS (mania, aggression, risk-taking behaviors, irritability), whereas others result from AAS withdrawal (depression, loss of libido, suicidal thoughts, hypersomnia)^15,16. There is considerable variability in the display and range of these symptoms. Fortunately, in very few people, these symptoms are disabling.

Androgens, including AAS, have rewarding properties^17–19 and may contribute to the development of substance abuse and dependency to other drugs^7,20. They also render brain substrates of the reward system more susceptible to the rewarding effects of several drugs of abuse such as cocaine, amphetamine, alcohol and opioids^21,22. Approximately 32% of subjects using AAS will develop a dependency to AAS⁷, this risk is higher for women and adolescents of both sexes^23,24. Several studies report a higher prevalence of drug abuse, particularly cocaine, in AAS users than in the general population^25,26. Owing to their widespread non-medical use and associated adverse effects, anabolic-androgenic steroids (AAS) were reclassified as Schedule III controlled substances by the U.S. Drug Enforcement Administration (DEA) in 2004. Nonetheless, AAS do not have a direct psychoactive effect and although evidence indicates AAS have addictive potential, there is currently no specific diagnostic criteria for AAS substance use disorder.

Among the AAS, nandrolone (19-nortestosterone), in its long-lasting ester form of nandrolone decanoate (ND), is the AAS most widely used worldwide²⁷. The androgenic activity of this compound is lower than that of dihydrotestosterone (DHT); in contrast, its anabolic properties are higher than those of T²⁸making it attractive for abuse by male and female athletes. Nandrolone has a higher affinity for the androgen receptor (AR) than T and is less susceptible to degradation by the 17 beta-hydroxysteroid dehydrogenase enzyme, which increases its appeal. Furthermore, although nandrolone and T can be reduced to DHT in target tissue containing the enzyme 5 alpha-reductase, the binding of the enzyme to nandrolone is weaker than that to T^29,30. Also, the binding of nandrolone to AR is weaker than that of DHT. This explains the more potent effects of nandrolone compared to T on target tissues without 5 alpha reductase activity and the weaker effect on tissues with a high 5 alpha reductase activity, resulting in greater anabolic/androgenic properties^31,32.

Adolescents, compared to children and adults, show the highest incidence of risk-taking behavior and experimentation with drugs of abuse³³. The use of AAS by adolescents is alarming because they can increase the behavioral response to other drugs of abuse (cross-sensitize) such as fenproporex and cocaine^34–36. Cocaine, one of the main drugs used by people who abuse AAS³⁷ shares many of the harmful side effects on the cardiovascular system³⁸. It also promotes risk-taking behavior³⁹ aggravating the detrimental health effects of AAS.

Drugs of abuse exert their behavioral and addictive effects by acting on regions of the brain associated with decision-making, motivation, and reward, such as the prefrontal cortex (PFC) and nucleus accumbens (NAc)⁴⁰. Dopaminergic receptors in these brain areas participate in regulating addictive behaviors^41,42. Decreased D2-like dopamine receptor (D2DR) expression is associated with increased novelty-seeking and risk-taking behaviors, traits associated with addiction^43,44. Manipulating striatal D2DR can alter the response to drugs of abuse, such as cocaine. Mice lacking striatal presynaptic D2DR show increased sensitivity to the locomotor activating effects of cocaine⁴⁵ while rats with increased D2DR sensitivity show enhanced cocaine self-administration⁴⁶.

Insults during adolescence affect brain development and interfere with the maturation of higher brain functions such as learning and memory. Our hypothesis was that AAS during adolescence reorganizes dopaminergic circuitry resulting in learning and memory deficits and increased risk taking and addictive behaviors. This study investigated whether exposure to nandrolone before puberty affected anxiety-like behaviors, as well as the behavioral response to cocaine. The nucleus accumbens and prefrontal cortex were studied to determine whether nandrolone and/or cocaine induce changes in the D2DR population of these brain substrates. We used the rat as our animal model since they are susceptible to AAS, their physiological responses are similar to that of humans, their brain undergoes synaptic remodeling with sex steroids, and they exhibit risk taking behaviors similar to humans. They also are one of the lowest steps in the evolutionary scale of laboratory animals.

Materials and methods

Animals

All the experiments were designed to minimize the number of animals used for each experiment. The significance level (alpha) was set at 0.05 and the power at 0.80, which means there was an 80% chance of detecting a true effect, while maintaining a 5% probability of a Type I error. The behavioral sensitization protocol was used to determine the sample size.

Pregnant Sprague Dawley rats were purchased from Charles Rivers Laboratories (Willmington, MA, USA). Dams were housed in pairs, with water and Purina^® rat chow provided ad libitum. They were kept in a temperature and humidity-controlled room, in a light-dark cycle with lights off at 5 PM (12 L:12D). After parturition, pups were cross-fostered, and each dam was housed with their litter of 8–10 pups. The litter was half male and half female. The experimenter did not keep track of the origin of each pup. The day of weaning (day 23), animals were separated from the dam and males and females housed separately in groups of 2–3 per cage. Although male and female progeny were included in the study, this manuscript only reports results obtained from males. We are currently preparing a manuscript discussing the data obtained with the female population. The experimental protocol was designed and approved prior to the initiation of the study. All animal experiments were revised and approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Puerto Rico Medical Sciences Campus (Protocol# 1140215) and adhere to USDA, NIH, AAALAC and ARRIVE guidelines. All the methods used are in accordance with NIH, AAALAC and ARRIVE guidelines and were followed as described in the UPR MSC IACUC-approved protocol (#1140215).

Animals were monitored daily for erratic behavior, such as a reduction in 15–20% body weight in a week, or the inability to eat or drink for a period of 24 h. No animals in the study displayed any of the above behaviors, which were the humane endpoints established for this study.

After the behavioral tests were concluded, animals were euthanized in a separate room by decapitation in a guillotine followed by rapid freezing of the brain in dry ice. Euthanasia was carried out by decapitation, a method acceptable by the American Veterinary Medical Association when its use is required by the experimental design and approved by IACUC (AVMA Guidelines, 2020). For the determination of dopaminergic receptors this method is necessary to avoid artifacts that may occur following stress or administration of an anesthetic⁴⁷. Personnel performing decapitations were properly trained and had experience in this procedure.

Drugs and chemicals

The AAS used in this study was 4-estren-17beta-ol-3-one decanoate (nandrolone decanoate) (Steraloids, Inc., Newport, RI, USA), dissolved in sesame oil. Sesame oil was used as vehicle because of its stability against oxidation and its common use as a solvent for steroids hormones. Although it contains minute amounts of phytosterols, this is true for other commonly used solvents such as peanut and corn oils^48,49. Nonethless all of our control comparisons were with sesame oil-treated rats to avoid any confound that may be attributed to the use of sesame oil. Nandrolone was administered subcutaneously (s.c.) at a dose of 20 mg/kg/day. Doses between 500 and 2000 mg/week have been reported by users of AAS⁵⁰. The dose of 20 mg/kg/day in rats is equivalent to a human dose of 1350 mg/wk in a 60 kg man, similar to the supraphysiological doses used by AAS users^50–52. Cocaine-HCl (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in 0.9% sterile saline and administered intraperitoneally (i.p.) at a dose of 15 mg/kg. The dose of 15 mg/kg of cocaine has been used extensively in our laboratory and has been proven to be effective in inducing behavioral sensitization and Conditioned Place Preference (CPP) to cocaine. We have found that higher doses (30 mg/kg) can induce tolerance, and lower doses are not as effective in a context-free behavioral sensitization paradigm^36,53.

Nandrolone treatment

On postnatal day 28 (PN-28), rats were weighed and randomly distributed into two groups that for 10 consecutive days received daily injections of nandrolone (ND) (20 mg/kg) or of sesame oil (Oil). Days 28 to 37 are within the conservative range for adolescence that spans days 28 to 42⁵⁴ in rats. We used two cohorts of rats of 40 individuals each for a total of 80 rats. To minimize potential confounders and variability when conducting behavioral studies, each session had representatives of each control and experimental group.

The first cohort of rats was treated first with nandrolone (n = 20 oil, n = 20 nandrolone) and then at day 39, all 40 rats were tested in the open field. This cohort was further subdivided into the following groups to be tested for behavioral sensitization: saline/oil (n = 10); cocaine/oil (n = 10), saline/nandrolone (n = 10) and cocaine nandrolone (n = 10). The test took place from days 40 to 62. They were euthanized on day 63.

Our second cohort (n = 40) was also treated with nandrolone (n = 20 oil, n = 20 nandrolone) and then on day 38, all 40 rats were tested in the Elevated Plus Maze (EPM). From days 40 to 53 this cohort was further subdivided into the following groups: saline/oil (n = 10), cocaine/oil (n = 10), saline/nandrolone (n = 10), cocaine/nandrolone (n = 10) and tested for conditioned place preference (CPP) to cocaine. This last group was euthanized on day 54.

Elevated plus maze

The EPM is a frequently used paradigm to measure anxiety-related behaviors⁵⁵. Rats previously exposed to anxiogenic drugs decrease the time spent in the open arms of the maze, while anxiolytic drugs increase the time spent in open arms⁵⁶.

Our testing apparatus consisted of a plus-shaped custom-made apparatus with two 50 cm open arms and two 50 cm enclosed arms without a roof. The apparatus stands at a height of 70 cm from the ground. The open arms had a 1 cm ledge; the floors were lined with rugged plastic to avoid slipping. An infrared video camera was placed in the center above the maze. The camera was connected to a computer containing the ANY-Maze^TM software. At the beginning of the test, the rats were placed at the junction of the open and closed arms and the video tracking system was activated. The software automatically recorded the number of entries into the open and closed arms, as well as the time spent in each arm. The entry into an arm was defined as the time point when more than 95% of the rat is in the arm. This was considered to be time zero. The test ended after 5 min. The amount of time spent in the closed and open arms and the number of entries into the open and closed arms were measured. The more time spent in closed arms, the higher the anxiety.

Open field test

The OFT is an assay of locomotor activity that is used to measure anxiety, exploratory and risk-taking behavior, and thigmotaxis^57–59. Anxiolytic drugs increase the time rats spend in the center area⁶⁰. Locomotor activity chambers from Versamax™ were used to measure Open Field Behavior. These chambers are made of clear acrylic (42 cm × 42 cm × 30 cm) with 16 equally spaced (2.5 cm) infrared beams across the length and width of the cage at a height of 2 cm (horizontal beams). An additional set of 16 infrared beams were located at the height of 10 cm (stereotyped activity). All beams were connected to a Data Analyzer that sent information to a personal computer.

The animals were placed in the activity chambers and allowed to roam freely for 10 min. Breaking of infrared beams determined the position of the rats in the activity chamber. The amount of time spent in the center of the chamber versus in the periphery was compared, as well as the total distance traveled. Animals that spent less time in the center of the chamber were classified as more anxious and as showing less risk-taking behavior when compared to their counterpart controls (oil treated).

Locomotor activity

Horizontal and stereotyped activity was measured with an automated animal activity cage system (Versamax™; AccuScan Instruments, Columbus, Ohio) (see description above). This system differentiates between horizontal, stereotyped, or rearing activity based on sequential breaking of different horizontal beams (Horizontal), the same beams (Stereotypies) or vertical beams (Rearing). Stereotyped behavior refers to repetitive motor responses of seemingly unknown function that arise in certain contexts, such as after psychostimulant administration^61,62.

Activity was measured in an isolated room with low illumination. Animals were habituated to the chamber for 1 h, 1 day prior to injections (Day 39). On days 40, 44, 52, and 62, rats were placed for 30 min in the chambers, and basal locomotor activity was recorded. The animals then received a saline or cocaine injection and locomotor activity was recorded for 60 additional minutes. On days 41–43, animals received a daily injection of 0.9% saline or cocaine (15 mg/kg) in their home cages. During days 45–51 and 53–61, animals remained undisturbed in their home cages (Fig. 1). Animals were sacrificed the day after the last behavioral test at 63 days of age.

Fig. 1.

Behavioral sensitization (A) and CPP (B) protocols. A Five days after rats were weaned from their mother, they received a daily oil or nandrolone injection for ten consecutive days (PN-28 to 37). For the behavioral sensitization experiments, rats were habituated to the locomotor activity chamber for 1 h (2 days post-nandrolone treatment, PN-39). The data obtained from the first ten minutes of habituation (Day 39 and Day 40 was used as the data for the open field test (Fig. 2B and C). From days 40 through 44, on day 52, and on day 62, rats were injected with saline or with cocaine (15 mg/kg). Rats were injected in the locomotor activity chamber (days 40,44, 52, and 62) or in their home cage (days 41, 42, and 43) (see the context of injection). From days 45 through 51 and from days 53 through day 61, rats remained undisturbed in their home cages. B. Five days after rats were weaned from their mother, they received a daily oil or nandrolone injection for ten consecutive days (PN-28 to 37). On day 38, a second group of rats were tested in the elevated plus maze (EPM). These same rats were used for the CPP experiments. To determine the rat’s preference for a particular CPP chamber, rats were allowed to roam freely through both chambers for 3 days (Days 40–42: pre-conditioning). The amount of time spent in each chamber was calculated to determine which side it preferred. For the following 10 days (days 43–52, conditioning days), rats were injected daily, alternating between saline and cocaine (15 mg/kg) injections. Rats received cocaine in the non-preferred chamber and saline in the preferred chamber. Saline animals received saline in both chambers. After the injection, rats were confined for 30 min in the chamber where they received the injection. On the last day (day 53: post-conditioning day), the animals were placed in the activity chamber and allowed to roam freely between the two chambers for 30 min. The amount of time spent in each chamber was recorded and compared to that spent in pre-conditioning.

Conditioned place preference

Cocaine-induced CPP was measured using Versamax™ activity chambers (described above). Each chamber was divided into 2 smaller chambers. For the pre- and post-conditioning sessions, the chambers were separated by an acrylic wall that had an opening; during the conditioning phase, the wall was replaced by a solid acrylic wall that separated the two chambers. Each chamber had different visual and tactile cues. During preconditioning, animals were placed in the CPP apparatus for 3 consecutive days and allowed to roam between both chambers for 15 min. The amount of time spent in each chamber was recorded. The conditioning phase consisted of alternating injections of saline and cocaine for 10 days. Saline was injected in the preferred chamber, and cocaine was injected in the non-preferred chamber with 24 h of separation between injections. The rats were confined for 30 min to the chamber where they received the injection. During postconditioning, the animals were placed in the activity chamber and allowed to roam between the 2 chambers for 15 min (Fig. 1). The time spent in each chamber during pre- and post-conditioning was compared. Rats that showed a significant increase in the time spent in the chamber where they received cocaine displayed conditioned place preference. This method has been validated by many laboratories^63,64.

Western blots

Western blots were used to quantify D2DR levels in mPFC and NAc. The protein concentration of the samples was determined using the BioRad Protein Assay method (BioRad Laboratories, Hercules, CA, USA). The molecular weight standards used for all Western blots were the BioRad Precision Plus Protein™ All Blue Standards (#1610373). For our gels, 20 µg of protein were mixed in SDS/-mercaptoethanol, vortexed and heated at 95 °C for 7 min prior to separation by 10% SDS-PAGE (BioRad Laboratories). Following electrophoresis, proteins were transferred to a 0.2 μm nitrocellulose membrane using Trans-Blot Turbo (BioRad Laboratories). Nonspecific binding to the membrane was blocked by incubation in Odyssey blocking buffer for 60 min at room temperature. This was followed by overnight incubation at 4 °C with a D2 receptor antibody (1:200; Santa Cruz Biotech, Santa Cruz, CA, USA, #sc-5303) and a B-Actin antibody (1:2500; Abcam, MA, #ab8227) dissolved in Odyssey Blocking Buffer. The D2DR receptor antibody used was a mouse monoclonal antibody raised against amino acids 1–50 that has been validated by various researchers^65–67. The next day, the membranes were washed 3X in TRIS buffered saline and polysorbate 20 (PBS-T). After washing, the membranes were incubated for one hour in IRDye 680RD goat anti-rabbit (1:15000; LI-COR, Lincoln, NE, USA, #926-68071) and IRDye 800CW goat anti-mouse (1:15000; LICOR, Lincoln, NE, USA, #926-32210). Proteins were detected using the Odyssey CLx infrared imaging system (excitation/emission filters at 700 nm/800 nm range, LI-COR Biosciences, Lincoln, NE, USA). The optical density of the D2 receptors of each sample was obtained using Odyssey software (LI-COR Biosciences), normalized against the background, and then corrected against their own levels of B-Actin. The Western blots images were not subjected to image editing or processing by editing software with the exception of cropping the images for publication purposes. We have included the replicates of the Western blots as part of the Supplementary Information.

Statistical analyses

All data were analyzed using GraphPad Prism version 9.00 for Windows (GraphPad Software, San Diego, California, USA). An unpaired t-test was used to compare two groups, a two-way ANOVA was used to compare more than two groups, and repeated measures MANOVA was used to analyze repeated measures.

To evaluate the effects of nandrolone and cocaine on behavioral sensitization, a linear mixed-effects model was applied to average locomotor activity measured between 30 and 60 min across three exposure days. The model included nandrolone and cocaine as between-subject factors, day as a within-subject factor, and subject ID as a random effect to account for repeated measures. Fixed effects and their interactions were tested, including the three-way interaction between nandrolone, cocaine, and day. Statistical significance was assessed using Wald z-tests, and parameter estimates were used to interpret main effects and interaction terms related to sensitization patterns. In addition, the time course of each group was analyzed separately using repeated measures (RM) ANOVA with days (40, 44, 52, and 62) and minutes (35–90) as repeated factors. Tukey multiple comparisons were used for post hoc analysis to compare locomotor and stereotyped activity over time.

CPP was analyzed by repeated measures MANOVA, using pre- and postconditioning as repeated factors. A two-way ANOVA was used to compare the expression of D2DR in the PFC and the NAc between the groups. The results of the statistical analysis are included as supplemental material. Data are presented as the mean ± standard error of the mean (SEM). A p-value of less than 0.05 (p < 0.05) was considered statistically significant.

Results

Open field and basal locomotor activity

Rats treated with nandrolone spent more time in the center of the open field and ambulated less than oil-treated rats (Fig. 2A and B, respectively). Data were analyzed with Student’s T test—Fig. 2A: t = 2.361, df = 36, p = 0.0264; Fig. 2B: t = 5.174, df = 36, p < 0.0001. The decrease in distance traveled corresponds to the first 10 min of habituation (sensitization protocol-Day 39). A decrease in total horizontal activity was also observed during the first 10 min in the habituation portion on day 40 of the sensitization protocol (Fig. 2C; Student’s T-test: t = 2.152, df = 18, p < 0.0453). This difference was not observed on subsequent test days (days 44, 52, and 62).

Fig. 2.

Open field activity of prepubertal rats exposed to nandrolone. Prepubertal males were injected daily with nandrolone (20 mg/kg) or sesame oil from PN 28–37. On days 39 and 40, rats were tested for open-field behavior (see Fig. 1A). Rats were placed in the activity cage, and the time spent in the center of the field, as well as the distance traveled, was recorded for 10 min using the Accuscan Versamax monitoring system. Rats treated with nandrolone (20 mg/kg) spent more time in the center of the open field (A) and traveled less distance on days 0 (B) and 1 (C) than oil-treated rats. Data are presented as mean ± SEM (n = 10) and were analyzed with a Student’s t-test. Asterisks represent a significant difference compared to the oil group.

Elevated plus maze

Nandrolone administration decreased the time spent in the closed arms of the EPM compared to rats that received oil (Fig. 3A: Student T-test: t = 2.623, df = 18, p = 0.0173). Additionally, nandrolone decreased the total number of arms entered, which is considered an indication of decreased locomotor activity (Fig. 3B: Student’s T-test: t = 3.28, df = 18, p = 0.0041).

Fig. 3.

Results from the EPM test of prepubertal rats treated with nandrolone. Male rats were injected daily with nandrolone (20 mg/kg) or sesame oil from PN 28–37 and tested in the elevated plus maze on day 38. Rats were placed at the junction between the open and closed arms, and behavioral activity was recorded for 5 min using the ANY-MazeÔ tracking software. Rats treated with nandrolone spent more time in the open arms (A), an indication of decreased anxiety. In addition, rats that received nandrolone had fewer total entries than oil-treated rats (B), an indication of reduced ambulation. Data are presented as mean ± SEM (n = 10) and analyzed with a Student t-test. Asterisks represent a significant difference compared to oil-treated rats.

Behavioral sensitization

To assess the influence of nandrolone on cocaine-induced behavioral sensitization, a mixed-effects analysis was conducted using average activity levels from 30 to 60 min across repeated exposure days. Cocaine significantly increased locomotor activity compared to controls (β = 1156.97, p = 0.004), with a progressive enhancement across days, as shown by significant interactions on Day 13 (β = 1650.71, p = 0.004) and Day 23 (β = 2698.76, p < 0.001), indicating robust sensitization (see Table 1 of supplementary material).

Nandrolone alone did not significantly affect activity (β = − 142.06, p = 0.657), but a significant three-way interaction with cocaine and Day 5 (β = 2003.57, p = 0.013) showed that nandrolone enhanced cocaine-induced activity early in the exposure period, i.e. they displayed behavioral sensitization (Fig. 4D and E) (see Table 1 of supplementary material). This effect was not significant on later days (Day 13: p = 0.399; Day 23: p = 0.899), suggesting that nandrolone may accelerate the onset of sensitization to cocaine without affecting its long-term magnitude. Of the 12 cocaine-induced locomotor activity time points measured during the 60 min after injection, 8 were higher in rats who received nandrolone-cocaine (Fig. 4C vs. 4D). This pattern was maintained on day 52: 10 of 12 cocaine-induced locomotor activity time points for rats treated with nandrolone-cocaine were higher than those on day 40 (Fig. 4D).

Fig. 4.

Cocaine-induced locomotor activity of rats exposed to nandrolone during postnatal days 28–37. Cocaine-induced locomotor activity of prepubertal rats injected daily from PN 28–37 with nandrolone (20 mg/kg) or sesame oil and tested for behavioral sensitization to cocaine from PN 40–62. A and B Timecourse of locomotor activity of saline (A) and nandrolone (B) treated males. No differences between oil and nandrolone-treated males were observed. C. A significant increase in cocaine-induced locomotor activity was observed when comparing the timecourses of day 40 vs. day 52 and day 40 vs. day 62 in Oil-treated males. Thus, these males displayed behavioral sensitization by day 52. D. A significant increase in cocaine-induced locomotor activity was observed when comparing the timecourses of day 40 vs. day 44 vs. day 52 and day 62 in nandrolone-treated males. These results suggest that nandrolone accelerates the maturation of the neural circuitry that regulates behavioral sensitization. Data are presented as mean ± SEM and analyzed with Repeated Measures ANOVA using Tukey’s multiple comparison for post-hoc analysis. E. Rats that received nandrolone show a robust increase in cocaine-induced locomotor activity by day 44, an increase that is maintained after two withdrawal periods. In contrast, it is at day 52 that oil-treated rats show a difference that is further increased by day 62. Data are presented as mean ± SEM (n = 10). Repeated Measures Two-Way ANOVA: F_(3,27) = 4.173, p = 0.0150. Oil-Coc: Day 40 vs. Day 44, p = 0.9968; Day 40 vs. Day 52, p = 0.0288; Day 40 vs. Day 62, p = 0.0001. ND-Coc: Day 40 vs. Day 44, p = 0.0011; Day 40 vs. Day 52, p = < 0.0001; Day 40 vs. Day 62, p = 0.0002. Oil-Coc vs. ND- Coc: Day 40, p = 0.8684; Day 44, p = 0.0002; Day 52, p = 0.0230; Day 62 p = 0.9209. P < 0.05 was considered statistically significant on Day 40 vs. Day 44, Day 52, and Day 62. (*): significantly different within groups; (#): significantly different between groups (Oil-Cocaine vs. Nandrolone-Cocaine).

In comparison, oil-cocaine-treated rats did not show differences in cocaine-induced locomotor activity between days 40 and 44 in any of the 12 measured time points and showed differences in only 5 of the 12 time points comparing day 40 with day 52 (Fig. 4C), reiterating that rats treated with nandrolone-cocaine displayed behavioral sensitization earlier (day 44) than oil-cocaine treated rats (day 52) (Fig. 4D: Two-way RM ANOVA, days, F (2.45, 22.08) = 13.03; p < 0.0001). No significant differences were observed between days in oil-saline and nandrolone-saline groups (Fig. 4A-B). Similar results were obtained with stereotyped activity (Fig. 5A-E: Two-way RM ANOVA, Days, F(2.58, 23.23) = 10.03, p = 0.0003). Additionally, the cocaine-induced locomotor activity of the rats treated with nandrolone was greater than that of the rats treated with oil at 9 of the 12 time points measured on day 44.

Fig. 5.

Cocaine-induced stereotyped activity of rats exposed to nandrolone during postnatal days 28–37. Cocaine-induced stereotyped activity of prepubertal rats injected daily from PN 28–37 with nandrolone (20 mg/kg) or sesame oil and tested for behavioral sensitization to cocaine from PN 40–62. A and B. Timecourse of stereotyped activity of saline (A) and nandrolone (B) treated males. No differences between oil and nandrolone-treated males were observed. C. A significant increase in cocaine-induced stereotyped activity was observed when comparing the timecourses of day 40 vs. day 52 and day 40 vs. day 62 in Oil-treated males. Thus, saline males displayed behavioral sensitization by day 52. D. A significant increase in cocaine-induced locomotor activity was observed when comparing the timecourses of day 40 vs. day 44 vs. day 52 and day 62 in nandrolone-treated males. As we observed with total horizontal activity, these results indicate that nandrolone accelerates the development of behavioral sensitization. Data are presented as mean ± SEM (n = 10) and analyzed with Repeated Measures of Two-Way ANOVA using Tukey’s multiple comparisons for post hoc analysis. E. Rats that received nandrolone show a robust increase in cocaine-induced stereotyped activity by day 44, an increase that is maintained after two withdrawal periods. In contrast, it is at day 52 that oil-treated rats show a difference that is further increased by day 62. Data are presented as mean ± SEM. Repeated Measures Two-Way ANOVA: F_(3,27) = 6.746, p = 0.0015. Oil-Coc: Day 40 vs. Day 44, p = 0.5042; Day 40 vs. Day 52, p = 0.015; Day 40 vs. Day 62, p < 0.0001. ND-Coc: Day 40 vs. Day 44, p < 0.0001; Day 40 vs. 52, p = < 0.0001; Day 40 vs. Day 62, p < 0.0002. Oil-Coc vs. ND- Coc: Day 40, p = 0.9224; Day 44, p < 0.0001; Day 52, p = 0.0281; Day 62 p = 0.9999. P < 0.05 was considered statistically significant on Day 40 vs. Day 44, Day 52, and Day 62. (*): significantly different within groups; (#): significantly different between groups (Oil-Cocaine vs. Nandrolone-Cocaine).

Western blot analysis of D2DR in mPFC (sensitization brains)

Our results indicate that cocaine decreased D2DR in mPFC (Fig. 6A and C). This effect was not altered by preexposure to nandrolone (Fig. 6A and C): One-way ANOVA, Tukey multiple comparisons, F (3, 12) = 110.5, < 0.0001; Oil-Sal vs. ND-Sal, p = 0.1559; Oil-Sal vs. Oil-Coc, p < 0.0001; Oil-Sal vs. ND-Coc, p < 0.0001; ND-Sal vs. Oil-Coc, p < 0.0001; ND-Sal vs. ND-Coc, p < 0.0001; Oil-Coc vs. ND-Coc, p = 0.9909.

Fig. 6.

D2DR in the mPFC and NAc of rats exposed to nandrolone during postnatal days 28–37 and tested for behavioral sensitization. A and C: Nandrolone did not affect D2DR expression in the mPFC. However, a decrease in D2DR expression in the mPFC was observed in animals treated with cocaine. B and D: In contrast, nandrolone treatment increased D2DR expression in the NAc and, similar to what was observed in the PFC, cocaine reduced D2DR in the NAc. Data are presented as mean ± SEM (n = 4). Representative western blots of D2DR and B-Actin in the mPFC (A) and NAc (B) of rats exposed to nandrolone during postnatal days 28–37,  tested for behavioral sensitization and sacrificed on PN 63. Data were analyzed using One-Way ANOVA. Western blots images were cropped for publishing purposes; replicates are included as Supplementary Material.

Western blot analysis of D2DR in NAc (sensitization brains)

Rats treated with nandrolone showed an increase in D2DR expression in NAc (Fig. 6B and D). Cocaine treatment decreased the expression of D2DR in this brain area; this decrease was more pronounced in animals that received nandrolone (Fig. 6B and D: One-way ANOVA, Tukey’s multiple comparisons, F(3, 12) = 84.90, < 0.0001; Oil-Sal vs. ND-Sal, p = 0.0002; Oil-Sal vs. Oil-Coc, p = 0.0339; Oil-Sal vs. ND-Coc, p < 0.0001; ND-Sal vs. Oil-Coc, p < 0.0001; ND-Sal vs. ND-Coc, p < 0.0001; Oil-Coc vs. ND-Coc, p = 0.0002.

Conditioned place preference

All animals showed CPP to cocaine. However, nandrolone decreased the time spent in the chamber associated with cocaine (Fig. 7). Rats treated with oil-cocaine showed more robust CPP to cocaine than rats treated with nandrolone-cocaine (Fig. 7). During postconditioning, oil-treated rats spent 70% of their time on the side associated with cocaine vs. 31% during the preconditioning phase. On the contrary, rats treated with nandrolone spent 48% of their time during post-conditioning in the chamber associated with the drug versus 30% during preconditioning. Data were analyzed with a two-way ANOVA, with pre- and post-conditioning as repeated measures = F(1, 62) = 28.22, p < 0.0001, and cocaine as independent factor. Cocaine effect: F_(3,62) = 21.97, p < 0.0001; Oil-Coc pre vs. postconditioning, p < 0.0001; ND-Coc pre vs. postconditioning, p = 0.0397; Oil-Coc post vs. ND-Coc postconditioning, p = 0.0035.

Fig. 7.

Conditioned Place Preference to cocaine of rats exposed to nandrolone during postnatal days 28–37. Male rats were injected daily from PN 28–37 with nandrolone (20 mg/kg) or sesame oil and tested for CPP to cocaine from PN 40–53. Nandrolone-treated males showed a decrease in the time spent in the chamber associated with cocaine compared to oil-treated males. During the postconditioning test, nandrolone-treated rats spent 48% of their time in the chamber associated with cocaine compared to 70% spent by oil-treated males. Rats injected with saline did not show a change in the time spent in the chamber where they were injected with saline. Although oil and nandrolone-treated males were conditioned to cocaine, conditioning was more robust in Oil-treated males. Data are presented as mean ± SEM (n = 8) and analyzed by a Two Way ANOVA (See supplementary material for detailed statistical analysis).

Western blot analysis for D2DR in mPFC (CPP brains)

Rats that received cocaine had decreased D2DR in PFC compared to saline-treated rats (Fig. 8A and C). However, the brains of the males treated with nandrolone used in the CPP experiments had more D2DR in the PFC than the oil-treated rats (Fig. 8A and C). The difference between these two experiments is that the animals used for the CPP experiments were killed on day 53, and those used for the sensitization experiments were sacrificed 11 days later, on day 64. Additionally, for the sensitization studies, rats received 7 cocaine injections (days 40–44, day 52, and day 62), while, for the CPP experiments, rats received 5 cocaine injections (days 43, 45, 47, 49, and 51). Figure 8C: One-way ANOVA, Tukey multiple comparisons, F(3, 12) = 28.78, < 0.0001; Oil-Sal vs. ND-Sal, p = 0.1752; Oil-Sal vs. ND-Coc, p = 0.0013; Oil-Sal vs. ND-Coc, p = 0.0007; ND-Sal vs. Oil-Coc, p < 0.0001; ND-Sal vs. ND-Coc, p < 0.0001; Oil-Coc vs. ND-Coc, p = 0.9825.

Fig. 8.

D2DR in the mPFC and NAc of rats exposed to nandrolone during postnatal days 28–37 and tested for CPP. Representative western blots of D2DR and B-Actin in the mPFC (A) and NAc (B) of rats exposed to nandrolone during postnatal days 28–37, tested for CPP and sacrificed on PN 53. B and D: Nandrolone increased D2DR expression in the mPFC and NAc of prepubertal rats. Fig. 6B and D: In contrast, cocaine decreased D2DR expression in both brain areas (mPFC and NAc), similar to what was observed in brains obtained from animals that were tested for behavioral sensitization and killed at a later age, day 64 (Fig. 6B and D). Data are presented as mean ± SEM (n = 4) and analyzed using One-Way ANOVA. Western blots images were cropped for publishing purposes; replicates are included as Supplementary Material.

Western blot analysis for D2DR in NAc (CPP brains)

The results obtained from the brains of animals sensitized to cocaine (Fig. 8B and D) are similar to those obtained with rats that were conditioned to cocaine; rats treated with nandrolone showed increased expression of D2DR in NAc. In contrast, cocaine treatment decreased the expression of D2DR in this area of the brain. Figure 8D: One-way ANOVA, Tukey’s multiple comparisons, F (3, 12) = 45.87, < 0.0001; Oil-Sal vs. ND-Sal, p = 0.0316; Oil-Sal vs. Oil-Coc, p < 0.0001; Oil-Sal vs. ND-Coc, p = 0.0001; ND-Sal vs. Oil-Coc, p < 0.0001; ND-Sal vs. ND-Coc, p < 0.0001; Oil-Coc vs. ND-Coc, p = 0.6828.

Discussion

Summary of results

Prepubertal males treated with nandrolone (PN 28–37) showed reduced anxiety and basal locomotor activity compared to oil-treated males. In contrast, nandrolone increased risk-taking behavior, augmented the locomotor response to cocaine after 5 injections, and accelerated the development of cocaine sensitization in young animals (i.e., they required less cocaine injections than oil-treated males). Nandrolone also attenuated the CPP to cocaine. Changes in D2DR in NAc may partially mediate these behavioral changes: an increase in D2DR was observed in the NAc of males treated with nandrolone, while those treated with cocaine or with nandrolone and cocaine had less D2DR in NAc and PFC.

Anxiety and risk-taking behaviors

The use of AAS is associated with adverse effects on emotion, cognition, and rewarding behaviors⁶⁸. Most of the effects of AAS on anxiety appear after cessation of use, which may be at the end of an AAS cycle or after longer withdrawal periods⁶⁹. Additionally, people with symptoms of depression or withdrawal are more likely to revert to the use anabolic steroids or other drugs of abuse⁷⁰.

Children born to mothers with polycystic ovarian syndrome, which results in increased androgen exposure during fetal development, have a higher prevalence of psychiatric disorders, such as depression and anxiety⁷¹. Additional studies in rodents confirm that exposure to excessive androgens during prenatal development increases the number of offspring that subsequently display anxiety behaviors⁷². In addition, several psychiatric disorders like anxiety, depression, mood disorders, psychosis, and substance abuse tend to manifest themselves close to puberty⁷³.

Very few studies have examined the effects of AAS administered prepubertally on emotional behaviors. Our study found that preexposure to nandrolone decreased anxiety-like behaviors. Nandrolone-treated rats showed a decrease in the time spent in the closed arms of an EPM and an increase in the time spent in the center of an open field compared to oil-treated rats. An increase in time spent in the center of an open field is also associated with increased risk-taking behavior, as rats that venture more into the center of an open field are at greater risk of being detected by predators.

Previous studies show that the administration of nandrolone (15 mg/kg) to adult rats decreased anxiety, as measured in the EPM⁷⁴. Decreased anxiety following androgen administration has also been reported by others^75–78. However, it is not unequivocally established whether AAS have anxiolytic properties since several studies with adult male rats⁷⁹ or with adolescent rats tested as adults^80,81 report that AAS exert anxiogenic effects.

The difference between studies investigating the effect of AAS on anxiety can vary depending on the type of androgen administered, the age of exposure, the duration of treatment, and the dose administered. These last studies used mainly an aromatizable form of androgen, and although the doses used were smaller (5 mg/kg vs. 20 mg/kg), as well as the total dose administered (60, 150, or 175 mg/kg vs. 200 mg/kg) the duration of treatment was longer (30–35 days vs. 10 days). Additionally, our current study is unique in that nandrolone was administered prior to puberty (PN 28–37), and the effects on anxiety and risk-taking behaviors were evaluated the day after the last injection (days 38 and 39) and not after a withdrawal period.

A correlation between high-risk behaviors in adolescents and testosterone has also been established. For example, a positive correlation was found between salivary testosterone and nucleus accumbens activation in response to a monetary reward in adolescent boys aged 10 to 16⁸². Not only do testosterone levels correlate positively with receiving a financial reward, but AAS users also engage more frequently in other risky behaviors, such as gambling.

Locomotor activity

Rats treated with nandrolone showed a decrease in the distance traveled in an open field and a reduction in the total number of arm entries when tested in an EPM, indicative of lower ambulation. Lower locomotor activity of males treated with nandrolone was also evident during the first 10 min of habituation (day 39) and on day 40 of the sensitization trial (Fig. 3C). This effect of nandrolone on locomotor activity appears to be related to novelty-induced exploration since it was not observed after further testing (days 44, 52 and 62).

Most studies agree that androgens decrease ambulation. In a similar study, prepubertal males treated with nandrolone (15 mg/kg) for 30 days showed reduced locomotor activity in an open field and EPM⁸³. In adults, nandrolone is reported to decrease activity in the running wheel⁸⁴in an open field^79,85 and in the EPM⁸⁶. Nonetheless, there are some studies that fail to find an effect⁸⁷ and others that report a decrease in cocaine-induced locomotor activity⁸⁸ after nandrolone withdrawal.

The mechanism by which nandrolone decreases locomotor activity is unclear. It has previously been reported that testosterone administration to prepubertal rats modulates the expression of enzymes that participate in dopaminergic metabolism in brain regions, such as the substantia nigra, that regulate locomotion and affect⁸⁹. Others report that the addictive properties of AAS may be exerted through the endogenous opioid system that in turn stimulate dopaminergic centers in the brain⁹⁰.

AAS and cocaine

Several studies indicate that AAS users are more likely to develop a dependency on other drugs of abuse^7,91. These results must be interpreted with caution since, in some cases, AAS users abused other drugs before using AAS. In addition, studies in adult rodents investigating the role of androgens in modulating the response to drugs of abuse have provided conflicting reports.

Prior studies from our laboratory and of others indicate that removal of the primary source of androgens by gonadectomy increases the response to psychostimulants^36,92–95although not all studies have obtained similar results^96–101. Furthermore, testosterone administration to adult males is reported to decrease the acute locomotor response to cocaine and amphetamine in gonadectomized^36,95 and gonadally intact drug-naive males^92,102. Indeed, our laboratory has found that testosterone is necessary for adult male rats to develop and express sensitization to cocaine³⁶.

Previous studies indicate that prepubertal male rats do not become sensitized to cocaine at this early age^103–105. In fact, our control rats did not show sensitization until after a withdrawal period and re-exposure to cocaine. Oil-cocaine rats showed sensitization when they were 52 days of age. At this time, they had received 5 daily cocaine injections, a 7-day withdrawal period, and a cocaine challenge. Sensitization in these oil-treated animals became more robust after a second withdrawal period and re-exposure to cocaine when they were 62 days of age. On the contrary, we found that exposure to nandrolone during days 28–37 accelerated the development of sensitization, which was evident after 5 cocaine injections, when they were 44 days of age.

We also observed that cocaine-induced hyperactivity was maintained for a longer period of time in nandrolone-treated males (day 52). It is not clear whether this effect can be attributed to changes in cocaine metabolism. Cocaine and nandrolone are metabolized mainly in the liver, cocaine by esterases and cytochrome P450 enzymes, while nandrolone metabolism occurs mainly by 5α-reductase and 3α- and 3β-hydroxysteroid dehydrogenase enzymes. Therefore, it is not surprising that pharmacokinetic interactions have been observed between the compounds. Synergistic effects on the cardiovascular system¹⁰⁶ and seizures¹⁰⁷as well as altered dopaminergic and serotonergic outflow⁸⁸have been reported. Unfortunately, although areas of the mesocorticolimbic pathway are rich in androgen receptors¹⁰⁸very few studies have investigated if androgens alter the metabolism of cocaine^109,110. Another possibility is that exposure to androgens during the prepubertal period accelerates the maturation of some components of the mesocorticolimbic circuitry essential for the display of behavioral sensitization. For example, the cytochrome P450 system is involved in the metabolism of androgens and cocaine¹¹¹ therefore, both drugs can interact pharmacokinetically and affect the individual response of each drug. Furthermore, treatment with the antiandrogen flutamide decreases plasma levels of cocaine and its main metabolites, suggesting that testosterone may enhance the effects of cocaine¹¹²as we have previously reported³⁶.

Our data also indicate that prior exposure to AAS increased the behavioral response to cocaine, a process known as cross-sensitization. Cross-sensitization occurs with previous exposure between the same drug and between different drugs of abuse¹¹³. The mechanisms that mediate cross-sensitization are not fully understood. Evidence suggests that dopamine is the substrate involved in this process¹¹⁴. Studies found that androgens like testosterone induce cross-sensitization to cocaine in prepubertal but not in adult male rats³⁵.

One scenario may be that AAS promotes cross-sensitization by activating corticotropin-releasing hormone (CRH) gene expression. The gene that codes for CRH contains androgen and estrogen response elements that modulate expression of CRH^115,116. Indeed, we have recently reported that altered corticotropin-releasing hormone receptor 1 (CRF-R1) sensitivity may lead to the observed DA hyperresponsiveness observed in socially isolated adolescent rats¹¹⁷. In addition, increases in dopamine, particularly in the medio-striatal brain region, is implicated in processing reward value, as well in mediating stereotyped behavior that results from psychostimulant administration¹¹⁸.

CPP

All rats tested between days 40 and 53 days developed CPP to cocaine. In addition, nandrolone-cocaine treated rats spent less time in the cocaine-associated chamber during the postconditioning day compared to oil-cocaine treated rats. The increase in time spent in the chamber associated with cocaine was 124% in oil-cocaine rats compared to 57% in those pretreated with nandrolone. Thus, although both groups of rats showed CPP to cocaine, CPP was lower in rats that previously received nandrolone.

Our studies agree with those of others that report a decrease in CPP to other drugs of abuse, such as cannabinoids¹¹⁹ and opioids¹²⁰after AAS administration. Furthermore, exposure to nandrolone (15 mg/kg) during PN 40–53 decreased sucrose consumption 15 days after withdrawal⁷⁷. However, there are studies that indicate that withdrawal from nandrolone increases cannabinoids¹²¹ and alcohol¹²² intake. Therefore, there is no consensus on whether prior exposure to nandrolone results in an enhancement or an aversion to other drugs of abuse or rewarding stimuli such as sucrose. However, there is evidence that the addictive properties of AAS can be exerted through the endogenous opioid system, which in turn stimulates the dopaminergic centers of the brain⁹⁰.

The CPP paradigm involves several cognition components, such as acquisition, retrieval, and extinction of spatial and contextual memories^123–125. Several lines of evidence link the use of AAS with cognitive dysfunction¹²⁶ and altered decision-making when studied in paradigms such as the operant discounting task^126–129.

We cannot discard the possibility that during the CPP test, the reduced time spent in the chamber associated with cocaine is due to a decrease in motivation. However, the increased sensitized locomotor response to cocaine displayed on day 44 by animals that received nandrolone evokes uncertainty about this possibility. Sensitization has been defined as the successive increase in locomotor hyperactivity elicited by repeated administration of psychostimulants. It involves neuroadaptations in the mesocorticolimbic system that contribute to changes in the motivational circuitry underlying craving and relapse^130–132. Since many investigators relate sensitization to increased motivation, and animals treated with nandrolone showed increased sensitization, we are inclined to favor changes in the cognitive and non-rewarding aspects of CPP¹⁷.

The effects of testosterone on cognition ranges from neuroprotective to inducing severe executive dysfunction depending on the dose¹³⁴. Very high or very low plasma concentrations of testosterone (like those found at the tails of a normal distribution curve) are associated with impaired cognitive function^{126,135–137}. For example, patients with prostate cancer treated with androgen deprivation therapy¹³⁶ or elderly men with low testosterone¹³⁷perform poorer on visuomotor tasks. Similarly, men that have abused AAS for several years show deficits when tested on several cognitive tasks, particularly those related to visuospatial memory¹³⁸. Data in animals yield similar results. Performance on the Morris Water Maze, a hippocampal-dependent spatial learning and memory test is impaired in AAS-treated rats^135,139.

Additionally, the rewarding properties of AAS are evident in previous studies that find that rodents will self-administer AAS orally and intracranially^17,19an effect that disappears with the administration of an androgen receptor antagonist¹³³.

AAS and D2DR

After the conclusion of our behavioral studies, we investigated whether the D2DR receptors in NAc and mPFC were affected by prior treatment with nandrolone, cocaine, or both. Separate groups of animals were used for the above-mentioned experiments. The rats used for CPP received 5 injections of cocaine (15 mg/kg), one every other day for 10 days, and were killed 24 h after the last injection of cocaine. Rats used for the sensitization experiments received 7 cocaine injections (15 mg/kg) during a 23-day period: 5 daily injections, a 7-day withdrawal period, 1 challenge injection, a 9-day withdrawal period, a second challenge injection, and were euthanized 24 h after the last cocaine injection. Thus, the sensitization group received two additional cocaine injections and underwent two drug-free periods of 7 and 9 days prior to euthanasia.

D2DR in the NAc of rats used for the sensitization experiments

Rats treated with nandrolone (days 28–37) had a higher concentration of D2DR in the NAc on the day of euthanasia (days 54 and 63) compared to oil-saline rats. In contrast, all the groups that received cocaine had lower levels of D2DR compared to oil saline rats. Interestingly, groups that were injected with nandrolone (days 28–37) and used for sensitization experiments (euthanized on day 63) had the lowest levels of D2DR in the NAc of all groups.

Behavioral sensitization has two phases: initiation and expression. The initiation comprises rapid neural effects that induce behavioral sensitization; the expression has long-term consequences¹¹⁴. Initiation of behavioral sensitization does not require activation of dopamine receptors¹⁴⁰however, its expression does^62,141,142. Most studies agree that the NAc, although not essential for the development of locomotor sensitization to cocaine, is necessary for its expression¹⁴³. Unfortunately, the role of each dopaminergic receptor subtype in the process of sensitization is still not entirely clear.

Some studies report that D2DR in the NAc does not affect cocaine-induced locomotor activity or behavioral sensitization^41,62,144. However, other studies attest to D2DR modulation of cocaine-induced sensitization¹⁴³. For example, blocking D2DR receptors in the NAc significantly decreased cocaine-induced locomotor activity^145–147 and abolished cocaine-induced sensitization¹⁴⁶. Furthermore, deletion of D2DR in medium spiny neurons (MSN) of the NAc in mice results in decreased cocaine-induced locomotor activity¹¹³. In contrast, experiments using the conditional mutant mice “autodrd2KO”, which is characterized by a lack of D2DR auto receptors (those in dopamine neurons and terminals), indicate that sensitivity to cocaine is enhanced in these animals⁴⁵. Mice lacking D2DR auto receptors also show enhanced CPP as well as hyperlocomotion (without altering dopamine transporters function)⁴⁵.

These data argue that the process by which D2DRs in the NAc modulate the response to cocaine can vary, depending on whether these receptors are in dopaminergic terminals or GABAergic MSN in the NAc. This may partially explain several conflicting reports. It is possible that the observed decrease in D2DR in the NAc in the current study occurred mainly in dopaminergic terminals. Repeated exposure to cocaine during sensitization would transiently result in increased dopamine (DA), which, in turn, would induce down-regulation of D2DR. D2DRs have a higher affinity for DA than D1DR^148,149. This would result in less dopaminergic autoinhibition and greater, or extended, DA release, resulting in an exacerbated locomotor response to cocaine, as we observed in the current study. It is also possible that DA, by modulating AMPA trafficking, contributes to the enhancement of sensitization¹⁵⁰. However, more studies are needed to determine the cell type or terminal where the decrease in D2DR occurred.

Additionally, nandrolone treatment increased D2DR in NAc. This group of animals showed the highest locomotor activity in response to cocaine and developed sensitization earlier than oil-treated rats (after 5 cocaine injections). These data are in agreement with previous studies that show that D2DR availability predicts future drug seeking behavior⁹⁹.

Androgens modulate D2DR

Androgens can modulate dopamine receptors in the mesocorticolimbic circuitry^89,151–153. Previous research found that administering nandrolone for two weeks at low doses (1 and 5 mg/kg) increased D2DR in the NAc core and shell of male rats but found that a higher dose (15 mg/kg) had no effect¹⁵⁴. The authors of this last study did not state the age or weight of the rats. In our study, the rats were 28 days old when they received the first injection of nandrolone, and the dose used was 20 mg/kg. Thus, we cannot determine whether the difference between these two studies is dose-related or age-related.

Nandrolone can also alter DA metabolism¹⁵⁵. A decrease in the activity of the dopamine-metabolizing enzymes monoamine oxidase A and B has been reported after nandrolone treatment^156,157. Furthermore, levels of tyrosine hydroxylase in the substantia nigra increase prior to puberty, coinciding with the increase in testosterone⁸⁹.

The Akil group, among others, have selectively bred rats according to their initial response to a novel environment and classified them as high responders (HR) and low responders (LR)^158–161. These two lines also show differences in behavioral traits relevant to addiction, with HR displaying a greater amount of drug taken, persistence of drug-seeking, and drug-induced locomotor activity¹⁶². They also differ in the neural substrates that regulate addictive behaviors. Like the males treated with nandrolone in this study, HRs have higher D2DR in the NAc core compared to LR¹⁵⁸. HRs also have higher levels of fibroblast growth factor 2 (FGF2) and lower D1DR levels in NAc. Interestingly, previous studies show that testosterone increases plasma levels of FGF2 and of Insulin Growth Factor¹⁶³ and mRNA expression of FGF2 in vitro.¹⁶⁴ Dysregulation of neurotrophic factors like FGF2 is involved in increased vulnerability to drugs^132,136. Indeed, drugs such as amphetamines and cocaine modulate the expression of FGF2^161,165. This may explain the synergistic effect of nandrolone and cocaine in decreasing D2DR in the NAc.

Prefrontal cortex

The mPFC plays an important role in addictive behaviors such as decision-making, memory retrieval, and cocaine-seeking behaviors and is necessary for the induction of sensitization to cocaine^166–169. It contains a distinct population of glutamatergic pyramidal neurons that project to the striatum and other subcortical regions that express D1DR and D2DR¹⁷⁰. Dopaminergic modulation of glutamatergic function contributes to reward, salience, attention, and working memory^171–174 Recent evidence indicates that DA modulates ensemble activity, facilitating and strengthening information processing in the PFC¹⁷³.

The maturation of the PFC circuitry continues after puberty, and dopaminergic innervation from the VTA to the PFC increases gradually until day 60¹⁷⁵. At approximately this age, males have acquired adult testosterone plasma levels and the ability to display male sexual behavior¹⁷⁶. An increase in dendritic spine density after androgen or estrogen administration suggests that gonadal steroids play a role in PFC function¹⁷⁷. Interestingly, on day 40, an increase in D2DR receptors in the PFC is observed, coinciding with the activation of the HPG axis in male rats¹⁷⁸. It is possible that in our current study, nandrolone administration from day 28 to 37 accelerated the maturation of the motivation circuitry in the PFC, which in turn could be responsible for the expression of sensitization at a younger age.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

J.A.F.A. and C.R.Q. participated in designing the experiments, data collection, analysis, writing and editing the paper. I.G.S.M., A.M.M., A.G.S., E.U.P.C. and R.J.T.R. participated in data collection. A.C.S. designed the study, oversaw project, data analysis, discussion and wrote the paper. All authors reviewed the manuscript.

Data availability

All relevant data concerning the study are contained within this paper.

Declarations

Competing interests

The authors declare no competing interests.