Nonassociative processes and place preferences conditioned by testosterone - Term paper, research paper & custom essay writing service

Home

Nonassociative processes and place preferences conditioned by testosterone

Nonassociative processes and place preferences conditioned by testosterone.

by Barbara J. Caldarone , Howard S. Stock , Glenn C. Abrahamsen , Michael L. Boechler , Bruce B. Svare , Robert A. Rosellini

Anabolic-androgenic steroids (AAS), the synthetic variants of the primary masculinizing androgen testosterone, are abused by growing numbers of individuals ranging from adolescents seeking to improve their appearance to athletes attempting to elevate their performance. Many anecdotal reports, survey data, and case studies have cited that AAS users report experiencing euphoria following steroid administration (for reviews see Brower, 1992; Kashin & Kleber, 1989). Controlled studies, however, have not been conducted to determine if AAS themselves can act as primary positive reinforcers (i.e., produce euphoria) or whether they act as secondary reinforcers, via improved physical appearance. Double blind studies are difficult to interpret because subjects report being able to distinguish AAS from placebos (Haupt & Rovere, 1984). However, subjects' ability to detect AAS in itself suggests that AAS can produce a subjective state that is discriminable in humans.

Few studies using laboratory animals have been conducted examining discriminative states produced by androgens. Although one study found testosterone propionate effective in conditioned taste aversion (CTA) learning in female mice (Peeters, Smets, & Broekkamp, 1992), two separate studies from our laboratory, one utilizing testosterone propionate in male and female mice (Miele, Rosellini, & Svare, 1988), the other testosterone cypionate and nandrolone in male mice (Ganesan, Rosellini, & Svare, 1993), demonstrated that these androgens were ineffective in producing a CTA. Research using ovarian steroids, however, has shown that these hormones can produce subjectively discriminable states that can serve as a conditioned (CS) or unconditioned stimulus (US). For example, progesterone has been shown to act as a CS in a state dependent learning paradigm (Stewart, Krebs, & Kaczender, 1967). In this study, females were trained to make one type of shock avoidance response when administered progesterone and another type of response when given saline. Similarly, estrogens are known to function as a US capable of producing robust, dose dependent CTAs (de Beun, Jansen, Smeets, Niesing, Slangen, & van de Poll, 1991; de Beun, Peeters, & Broekkamp, 1993; Miele et al., 1988; Peeters et al., 1992).

Several years ago, we undertook a set of studies to test the hypothesis that AAS produce positive reinforcing effects that can function as positive reinforcers. In this series of experiments, we chose to employ a conditioned place preference (CPP) paradigm (Svare, 1990). This paradigm involves the temporal pairing of contextual stimuli, which function as a CS, with the presumed euphoric effects of the drug, which act as a US. These tests have been used successfully to assess the positive reinforcing properties of other drugs of abuse (e.g., Bozarth, 1987; Mucha & Iversen, 1984; Reid, Hunter, Beaman, & Hubbell, 1985; Reid, Marglin, Mattie, & Hubbell, 1989; van der Kooy, 1987). Although operant conditioning techniques, such as intravenous self-administration, have also been successfully utilized in this context (Weeks & Collins, 1987), these techniques are not well suited for use with AAS because steroids are not soluble in water or in low ethanol concentrations.

Since the undertaking of our CPP experiments examining the potential positive reinforcing properties of AAS, two studies have been published examining the efficacy of testosterone in producing a CPP (Alexander, Packard, & Hines, 1994; de Beun, Jansen, Slangen, & van de Poll, 1992). Specifically, de Beun et al. (1992) reported achieving a CPP with a 1 mg/kg dose of testosterone in male, but not female rats and Alexander et at. (1994) reported achieving a CPP with 800 and 1200 [[micro]gram]/kg testosterone-hydroxypropyI-B-cyclodextrin inclusion complex. Our present series of studies, which utilized control groups not included by de Beun et al. (1992) or Alexander et al. (1994), question the associative nature of the CPP previously reported.

Experiment 1

The purpose of our initial experiment was to evaluate testosterone's capacity to produce a CPP. In order to evaluate the effectiveness of our place conditioning paradigm, in addition to testosterone, we utilized morphine as a conditioning agent, a substance that has been demonstrated in numerous reports to produce a reliable CPP (Mucha & Iversen, 1984; Reid et al., 1989; van der Kooy, 1987). Comparisons were made between the relative capacity of these two drugs in producing a place preference.

Method

Subjects

Subjects were experimentally naive male Sprague-Dawley rats obtained from Taconic Farms of Germantown, New York. At approximately 80 days of age, they arrived at our laboratory where they were individually housed under conditions of ad libitum food and water for 1 week prior to the beginning of the experiment. All procedures were conducted during the light phase of a 12-hr light-dark cycle.

Drugs

Morphine sulfate (Sigma Chemical Co., St. Louis, MO) was dissolved in 0.9% NaCl at a concentration of 2 mg/ml. Testosterone (Sigma Chemical Co., St. Louis, MO) was dissolved in sesame oil (Fisher Scientific Company, Fair Lawn, NJ) at a concentration of 10 mg/ml.

Apparatus

Four conditioning chambers were used throughout these studies. These chambers were closely modeled after those employed by Reid et al. (1989). Three walls of the chambers were constructed of aluminum and the ceiling and front wall of clear Plexiglas. Each chamber measured 57.15 x 27.9 x 20.3 cm. The walls on the left side of each chamber consisted of 2.54-cm wide horizontal black and white stripes, and those on the right side were uniform gray. The floor consisted of wire mesh (1.27 cm) that was placed at a 45 [degrees] angle to the wall on the left side and at 90 [degrees] to the wall on the right side of the chamber. Each chamber could be divided into two equal halves by means of a Plexiglas partition painted with the appropriate stimulus for each side. The chambers were dimly illuminated by means of two 40-W lights (Sylvania, Model #40A15/FAN/RP) each of which was located 5.08 cm above and behind the rear corner of each side of the chamber. Each chamber was suspended in a sound- and light-attenuating container by means of an axle system that allowed the chamber to pivot and therefore tip when the animal placed the majority of its weight on one side. The location of the animal in the chamber was monitored by a microswitch system. Control of the apparatus and data collection were accomplished by a Zenith XT computer.

Procedure

This study consisted of four phases: (1) habituation to the laboratory and apparatus, (2) baseline preference assessment, (3) place preference conditioning, and (4) place preference test. Two cycles of conditioning and testing were conducted for each experiment. Data are reported for the second test day.

Habituation. After a 1-week acclimation to the laboratory, the animals were transported to the experimental room for 2 days and were allowed to explore the conditioning chambers for 30 min per day. During each of these periods, the animals had access to both sides of the chambers.

Baseline. On the 3rd day, each animal was again given access to each side of the chamber for 30 min and the amount of time that the animal spent on each side was recorded. Subjects were balanced across the experimental conditions based on baseline side preference (gray vs. striped).

Conditioning. This phase of the study was modeled after the conditioning procedures employed by Reid et al. (1989). The day after assessment of baseline preference, each animal was given a subcutaneous injection of either morphine (8 mg/kg), testosterone (1 mg), or vehicle (saline or oil) immediately prior to placement in the side of the chamber that the animal was to be conditioned (side of putative conditioning). Because animals tended to show a marked preference for either one side of the chamber or the other on the baseline day, all animals were conditioned against their original preference. The conditioning phase consisted of two 4-day cycles. Each cycle consisted of three consecutive pairings of drug exposure with a specific side of the chamber and one pairing of the alternate side of the chamber with a vehicle injection. To investigate the possibility that the order of drug/vehicle presentation might influence any CPP observed with morphine or testosterone, two additional groups of animals were used for each drug. These procedures yielded four groups. One group received three consecutive pairings of morphine with placement into one side of the chamber followed on the 4th day by an injection of saline and placement into the other side of the chamber (MMMS, n = 11). A second group received saline on the 1st day of the cycle followed by 3 days of morphine (SMMM, n = 12). The other two groups received identical treatments with the exception that they received testosterone and the oil vehicle (TTTO, n = 12; OTTT, n = 12).

Place preference test. On the day following the last day of pairing, the animal's preference for each side of the chamber was assessed by allowing it access to both sides for 30 min.

Results

As seen in Figure 1, groups (morphine or testosterone) did not differ in the amount of time spent on the least preferred side of the chamber on the baseline day. On the test day, animals that received either morphine or testosterone paired with a specific side of the chamber increased the amount of time spent on the side of the chamber that had been paired with the drug (side of putative conditioning). No differences were seen in the amount of time spent on the side of putative conditioning for morphine or testosterone paired animals. Analysis of variance (ANOVA) conducted on the amount of time spent by each animal on the side of putative conditioning as a function of order of conditioning (drug-vehicle vs. vehicle-drug), drug type (morphine vs. testosterone), and days (baseline to test) failed to reveal a significant main effect or interaction of order of conditioning. Therefore, data were collapsed across the order variable for further analyses. Analysis of baseline to test data showed a significant effect of days, F(1, 43) = 48.23, p [less than] .001, confirming that animals exposed to morphine or testosterone spent an increased amount of time on the drug paired side on the test relative to the baseline day. No significant main effect or interactions with drug type was found.

Discussion

Our findings that animals increased the amount of time spent on the side of a chamber paired with the affective consequences of morphine replicates numerous other studies reporting a similar effect of morphine (Mucha & Iversen, 1984; Reid et al., 1989; van der Kooy, 1987) and most closely replicates the findings of Reid et al. (1989) after which our apparatus, conditioning, and test procedures were modeled.

Our results differ from Reid et al. (1989) in one respect. They report no differential baseline preference and therefore randomly condition animals to either side of the chamber. In our procedure, we observed a baseline preference and in order to provide a more conservative assessment of any potential CPP, conditioned against the animal's baseline preference. This difference, however, does not obviate the demonstration that in our laboratory morphine does increase the amount of time spent on the side of the chamber paired with the presumed positive reinforcing properties of the drug and therefore is indicative of a CPP.

Our establishment of a CPP with morphine cannot be attributed to an effect of novelty, as defined by the side in which the animals have spent the least amount of time during conditioning. Scoles and Siegel (1986) have found that rats tend to move to the side of the conditioning chamber in which they have spent the least amount of time. Because our procedure consisted of three pairings of the drug with one side and one pairing of the vehicle with the alternate side, the novel side of the chamber is that where the animal received the vehicle. Thus, any such tendency in our study for animals exploring the novel side should have been evidenced by the animals spending more time on the vehicle-exposed side of the chamber. This did not occur in our experiment because animals spent more time on the side of the chamber paired with the drug.

The absence of an effect of order of conditioning in our study also demonstrates that establishment of a CPP with morphine cannot be attributed to novelty. If the novel side of the chamber is defined by the side opposite to which the animal was exposed on the last day of conditioning (the day prior to the place preference test), it would be expected that vehicle-drug (SMMM) animals would spend more time on the alternate as opposed to the putative conditioned side of the chamber. Because order of conditioning did not influence the amount of time spent on the putative conditioned side on the test day, the CPP established with morphine cannot be attributed to novelty.

Most importantly these data indicate that although animals increased the amount of time spent on the side of the chamber paired with either morphine or testosterone, the increase was not differential across drug type. That is, testosterone was as effective as morphine in increasing the amount of time animals spent on the side of the chamber paired with the drug. This result suggests that testosterone may have positive reinforcing properties and thus be an effective agent for establishing a CPP.

Experiment 2

Although the results of Experiment 1 suggested that testosterone, like morphine, is capable of establishing a CPP, additional experimentation was necessary before we confidently concluded that the increase in time spent on the drug-exposed side of the chamber specifically resulted from pairing the side of the chamber with testosterone. To assess whether the results of Experiment 1 resulted from an association between the putative positive reinforcing state produced by testosterone with the environmental stimuli, we replicated the basic conditioning procedures of the first experiment. In addition, a control group was added to the experiment that received drug administration and chamber placement in a temporal manner which should preclude any associative conditioning effects. Specifically, this control group received the drug in the home cage environment 3 hours following removal from the conditioning chamber. These home cage controls allowed us to assess whether the apparent CPP observed in Experiment 1 was caused by the pairing of the stimuli in the specific side of the chamber with the presumed positive reinforcing effect of the drug. Based on Pavlovian conditioning principles (Macintosh, 1974; Pavlov, 1927; Rescorla, 1969), the increased preference for contextual stimuli paired with a drug should be a function of the temporal pairing of the contextual stimuli with the presumed positive reinforcing properties of the drug. If the apparent CPP produced by testosterone is associative in nature, then animals that received testosterone before placement into the conditioning chamber would be expected to increase the amount of time spent on the side of putative conditioning on the test day relative to their own baseline. Alternatively, animals that received testosterone 3 hours after removal from the chamber would not be expected to show this increase.

Method

All procedures were identical to those used in Experiment 1 with the exception that only one order of conditioning, drug-vehicle, was used.(1) As in Experiment 1, two experimental groups were administered either 8 mg/kg of morphine (morphine-paired, n = 9) or 1 mg of testosterone (testosterone-paired, n = 9) and immediately placed into the side of putative conditioning. Two control groups received equivalent exposure to the drug, vehicle, and chamber. In these control groups, however, injections were administered three hours following removal from the chamber. These two home cage control groups (morphine-home cage, n = 9; testosterone-home cage, n = 9) therefore served as controls for possible nonassociative effects of drug exposure.

Results

As seen in the left panel of Figure 2, morphine-paired and morphine-home cage animals did not differ in the amount of time spent on the side of putative conditioning on the baseline day. However, animals that received morphine paired with the side of putative conditioning showed a larger increase in the amount of time spent on that side from baseline to the test day as compared to their respective home cage controls. ANOVA revealed a significant Conditioning Type (paired vs. home cage) x Days interaction, F(1, 16) = 6.21, p = .024. The source of this interaction was confirmed by additional analyses which showed that although animals were not significantly different on the baseline day, morphine-paired animals spent significantly more time on the side of putative conditioning on the test day relative to the home cage controls, F(1, 16) = 9.41, p = .007. These results indicate that morphine administration paired with a specific context resulted in a CPP, as compared to the home cage controls.

The right panel of Figure 2 shows that animals that were administered testosterone and immediately placed into the conditioning chamber (testosterone-paired) and animals that were administered testosterone 3 hours after removal from the conditioning chambers (testosterone-home cage) tended to spend more time on the side of putative conditioning on the test day relative to baseline. ANOVA conducted as a function of conditioning type and days revealed a marginal effect of days, F(1, 16) = 3.33, p = .087, suggesting that animals tended to increase the amount of time spent on the putative conditioned side on the test relative to baseline days. Additionally, a significant main effect of conditioning type, F(1, 16) = 4.72, p = .045, was found indicating that the testosterone-home cage group overall spent more time on the side of putative conditioning than the testosterone-paired animals. Most importantly, animals exposed to testosterone did not show a significant Conditioning Type x Days interaction, F(1, 16) = .63, p = .438, suggesting that animals that received testosterone paired with the side of putative conditioning did not significantly increase the amount of time spent on that side compared to their respective home cage control.

Discussion

The CPP observed with morphine replicates previous work (Mucha & Iversen, 1984; Reid et al., 1989; van der Kooy, 1987) indicating that the morphine place preference is caused by the association of the positive reinforcing effects of morphine with the contextual stimuli on the putative conditioned side of the test chamber. Furthermore, animals that received morphine paired with the contextual stimuli spent more time on the putative conditioned side of the chamber than animals that received equivalent morphine and chamber exposure in a manner that prevented the formation of an association between these events. Thus, the results of this study with morphine, when considered in combination with those of Experiment 1, indicate that our CPP procedures are effective in conditioning a place preference with morphine both on a within- and between-subject basis.

The capacity of testosterone to condition a place preference is, however, questioned by the results of the present experiment. Although the results of Experiment 1, which employed a within-subject procedure, suggested that testosterone was effective in producing a CPP, the present between-subject assessment raises questions about the associative nature of this effect. Whereas both testosterone-paired and testosterone-home cage groups tended to increase the amount of time spent on the side of putative conditioning on the test relative to baseline, overall testosterone-paired animals spent less time on that side compared to home cage controls. This indicates that the pairing of testosterone with the conditioning context is not necessary for the observance of the increase in the amount of time spent on the side of putative conditioning and strongly questions the associative basis of the effect observed in Experiment 1.

However, before it can be confidently concluded that testosterone cannot produce a CPP, it must be acknowledged that the supraphysiological dose used in Experiments 1 and 2 may have produced residual positive reinforcing effects. Steroid hormones, such as testosterone, have either rapid, short duration effects on neuronal activity, or slow onset, long duration genomic effects, which may persist for hours or days (McEwen, Coirini, & Schumacher, 1990). If these residual effects were present on the day after drug administration, they could have become associated with the stimuli that were present while the animal was in the conditioning chamber. Because order of conditioning included 3 days of testosterone followed by 1 day of vehicle exposure, both testosterone-paired and testosterone-home cage animals would have received 2 days of residual testosterone exposure on the putative conditioned side and 1 day of residual testosterone on the alternate side. If the residual testosterone exposure did, in fact, produce a positive reinforcing state, these animals would have received conditioning from the residual drug effects for 2 days on the putative conditioning side and 1 day of conditioning on the alternate side, which according to general theories of learning (Macintosh, 1974; Pavlov, 1927; Rescorla, 1969) could produce some conditioning.

Experiment 3

The results of Experiment 2 questioned whether testosterone was effective in establishing a CPP. Before drawing any strong negative conclusions about testosterone's capacity to produce a CPP, we examined the possibility that at smaller doses, testosterone could be effective in establishing a CPP. Therefore, a third experiment was conducted that replicated the design of Experiment 2, using two lower doses of testosterone. The dosage was lowered for three reasons. First, to investigate the possibility that long duration effects of testosterone produced conditioning from residual drug effects, lower doses were used to minimize any conditioning that may have occurred because of the residual effects of testosterone. However, if residual conditioning did occur, a separation in magnitude of conditioning would be expected between the two doses. Second, physiological doses were utilized in an attempt to minimize any nonassociative effects that may have been caused by the supraphysiological dose used in Experiments 1 and 2. Third, the supraphysiological dose used in Experiments 1 and 2 may have had both positive and negative reinforcing effects. Therefore, the dose was lowered in an attempt to unmask the positive reinforcing effects by removing any potential negative reinforcing effects caused by the high dose.

Method

The procedures used for this study were identical to those of Experiment 2, with two exceptions. The first was that one half of the animals received a 10[[micro]gram] injection of testosterone whereas the other half received a 100[[micro]gram] injection. The second difference was that all animals received four testosterone injections per conditioning cycle and one vehicle injection. Because the dose of testosterone was lowered, we chose to increase the number of drug-context pairings from the three employed in the above experiments to four, in an attempt to maximize the possibility of observing a CPP. Basic principles of Pavlovian conditioning suggest that increasing the number of pairings of the CS with the US should strengthen the formation of the association (Macintosh, 1974; Pavlov, 1927; Rescorla, 1969).

For two of the groups in this study, testosterone was paired with placement in a specific side of the chamber (Group 10[[micro]gram]-paired, n = 13, 100[[micro]gram]-paired, n = 13) and one injection of oil vehicle was given prior to placement on the alternate side of the chamber. The other two groups (Group 10[[micro]gram]-home cage, n = 13; 100[[micro]gram]-home cage, n = 13) received equivalent exposure to the drug, the vehicle, and the chamber. However, testosterone and vehicle injections were administered 3 hours following placement in a specific side of the chamber.

Results and Discussion

As seen in Figure 3, although groups did not differ on baseline day, animals that received 10 and 100[[micro]gram] of testosterone paired with the conditioning environment (10[[micro]gram]-paired and 100[[micro]gram]-paired) showed an increase in the amount of time spent on the side of putative conditioning on the test day. Furthermore, animals that were administered testosterone 3 hours after removal from the conditioning environment (10[[micro]gram]-home cage and 100[[micro]gram]-home cage) showed a similar increase. ANOVA conducted as a function of dose (10 or 100[[micro]gram]), conditioning type, and days showed a significant effect of days, F(1,48) = 12.79, p = .001, confirming the general pattern suggested above, that all groups increased the amount of time spent on the putative conditioned side of the chamber on the test relative to the baseline day. No other significant main effects or interactions were observed. This pattern of statistical significance indicated that although animals increased the amount of time spent on the side of putative conditioning on the test relative to baseline, the increase was not specific to either the dose of testosterone administered or most importantly, to receiving testosterone paired with the side of putative conditioning. Thus, in animals that received a 10 and 100[[micro]gram] dose of testosterone, no evidence of a CPP was observed.

It is also unlikely that the increase in the amount of time spent on the side of putative conditioning for paired and home cage animals was caused by conditioning from the residual effects of testosterone. The physiological doses used in this study should have minimized any potentially positive reinforcing effects of testosterone present on the day after administration. The increase in time spent on the side of putative conditioning, however, does not appear to be differential across doses ranging from 1 mg (Experiments 1 and 2) to 10[[micro]gram], suggesting that testosterone does not produce positive reinforcing consequences via a long term mechanism.

General Discussion

The results of the present studies demonstrate that in our laboratory and under the present experimental conditions, morphine produces a CPP that is evidenced both on a within-subject (Experiment 1) and between-subject (Experiment 2) basis. These studies replicate numerous reports in the literature demonstrating morphine's effectiveness in producing a positive reinforcing state that can become associated with contextual stimuli present during drug exposure (Mucha & Iversen, 1984; Reid et al., 1989; van der Kooy, 1987).

The results of our experiments, however, provide a more ambiguous answer to whether testosterone can function as an effective agent in conditioning a place preference. In Experiment 1, animals given testosterone appeared to display a CPP on a within-subject basis, because they exhibited a significant increase in the amount of time spent on the testosterone-exposed side of the chamber on the place preference test relative to their own baseline. Experiment 2, however, showed that this increase was not specific to receiving the hormone paired with the conditioning context. In this experiment a home cage control group was added in which animals were administered testosterone in the home cage 3 hours after removal from the conditioning context, which should preclude the formation of an association between context and drug exposure. These home cage control animals, however, showed increases in the amount of time spent on the side of putative conditioning, in a manner similar to animals that received testosterone paired with the conditioning context. The results of this experiment, which showed no evidence that testosterone can produce a CPP, suggest that testosterone does not have positive reinforcing properties. Additionally, in Experiment 3, two lower doses of testosterone showed no evidence of establishing a CPP relative to home cage controls after two conditioning cycles. Interestingly, in this experiment, no CPP was established even though the number of conditioning trials was increased from three to four, which according to Pavlovian conditioning principles should increase the likelihood of obtaining a place preference (Macintosh, 1974; Pavlov, 1927; Rescorla, 1969). Thus, based on these studies, we suggest that it is unlikely that testosterone can produce a positive reinforcing state that, when paired with external stimuli, can produce a CPP.

The results of the present studies suggest that testosterone does not condition a place preference, because general increases were observed in time spent on the putative conditioned side of the chamber in both animals exposed to testosterone paired with the chamber and in the home cage environment. Although it could be proposed that these general increases are caused by a simple change in preference from the baseline to test day, this explanation is unlikely. In the present studies, animals received 2 days of habituation to the conditioning chambers before the baseline test and side preferences were recorded for each day. Examination of the habituation and baseline data revealed that animals, on the first habituation day, tended to prefer one side and that side preference increased on the subsequent habitation and baseline day. Furthermore, Bozarth (1987) found that animals showed stable preconditioning preference for 20 days of repeated testing and animals that received saline during conditioning trials failed to show reliable shifts in preference, again demonstrating that initial side preferences remained stable in the absence of drug exposure.

An alternate explanation for these general increases may be due to a nonassociative, long term effect of testosterone in both the paired and home cage testosterone-exposed animals, perhaps via an attentional mechanism. For example, in male chicks, exposure to testosterone has been shown to induce a long term increase in the ability to sustain attention on a visual stimulus (Andrew, 1972). In these studies, testosterone was shown to increase "persistence" in search of a particular type of food in a particular environment. Thus, it is conceivable that in our CPP paradigm, testosterone may have increased a "preference" for a particular side of the chamber in a manner that did not involve associative learning.

This issue raises the question involving the associative nature of the place preference reported with testosterone by de Beun et al. (1992) and Alexander et al. (1994). In agreement with our studies, these authors reported results that are consistent with the general increases seen in the amount of time spent on the side of putative conditioning following testosterone exposure. These authors utilized control groups that received only context-vehicle exposure but did not employ testosterone exposed home cage control groups, as was done in the present experiments. Therefore, the possibility remains that the CPP reported in those studies may have been caused by a nonassociative effect of testosterone.

However, methodological differences between the place conditioning procedures used in the present studies and those used by de Beun et al. (1992) and Alexander et al. (1994) may explain why we found no evidence that testosterone can condition a place preference. Some of these differences include strain of rat [Sprague-Dawley vs. Wistar (de Beun et al., 1992)]; form of testosterone [testosterone vs. testosterone-hydroxypropyl-B-Cyclodextrin inclusion complex, a recently developed form of testosterone (Alexander et al., 1994)]; number of drug-vehicle pairings [two conditioning cycles of three testosterone and one vehicle pairing vs. one conditioning cycle of four testosterone and four vehicle pairings (de Beun et al., 1992) or one conditioning cycle of six testosterone and six vehicle pairings (Alexander et al., 1994)]; and side of the chamber to which animals were conditioned [least preferred vs. random (de Beun et al., 1992 and Alexander et al., 1994)]. Time before placement into the conditioning chamber following testosterone exposure was also different between our studies and de Beun et al. (1992). De Beun et al. (1992) exposed animals to the conditioning chamber 60 minutes following testosterone injection, whereas in the present studies, animals were placed into the chamber immediately following injection. In an additional exploratory experiment conducted in our laboratory, however, comparisons between animals injected 60 minutes before chamber placement and home cage controls again revealed no differences in testosterone-paired and testosterone-exposed home cage controls in amount of time spent on the putative conditioned side.

In addition to the variables listed above, our experiments differ from de Beun et al. (1992) in another potentially important respect. Our studies were conducted using intact animals, as opposed to the de Beun et al. (1992) experiment that utilized castrated animals. It is conceivable that removal of testicular hormones through castration could have produced an aversive state in the animals in the de Beun et al. (1992) study. Subsequent exposure to testosterone may have relieved this aversive condition, returning animals to their precastration state. In support of this hypothesis, castration has been shown to decrease concentration of dopamine in the nucleus accumbens, a neurochemical system known to be critical for reward (Koob & Goeders, 1989; Stellar & Rice, 1989). Furthermore, this reduction in dopamine was prevented by treatment with testosterone (Mitchell & Stewart, 1989). Additional support for this replacement hypothesis comes from the fact that de Beun et al. (1992) failed to observe a CPP in female animals. Thus, the proposed CPP reported by de Beun et al. (1992) may have resulted from the removal of an aversive condition and not from the production of a positive reinforcing state. Similar results have been observed in studies with morphine. Using a CPP paradigm, Bechara and van der Kooy (1992) found that the rewarding effects of morphine in dependent animals may be due to alleviating the aversiveness of withdrawal symptoms.

Although the combination of results observed in the present experiment suggest that testosterone may not be an effective agent for the establishment of a conditioned place preference, we must emphasize that in these initial investigations we manipulated only the most obvious and a rather limited set of variables that should have influenced the conditioning of a place preference with testosterone. As has been the case with other chemical agents such as ethanol (Reid et al., 1985), it is possible that modified parameters, such as preexposure to the US, would be effective in producing a CPP with testosterone. Furthermore, other techniques for measuring the rewarding capacity of drugs, such as brain stimulation may be a fruitful area of investigation. One study, however, failed to find an effect of the AAS methandrostenolone (Dianabol) on stimulation of the median forebrain bundle (Clark, Lindenfeld, & Gibbons, 1993). Clearly, additional variables will need to be investigated before a firm conclusion can be reached concerning the putative positive reinforcing properties of AAS.

In conclusion, the present series of experiments question the efficacy of the CPP paradigm in studying the potential rewarding properties of AAS. The results of the present studies indicated that our conditioning techniques were effective in producing a reliable CPP with morphine, replicating numerous reports in the literature (Mucha & Iversen, 1984; Reid et al., 1989; van der Kooy, 1987). However, using identical conditioning techniques that were effective in establishing a CPP with morphine, we found no evidence that several doses of testosterone could condition a place preference. In agreement with the previous studies reporting a CPP with testosterone (Alexander et al., 1994; de Beun et al., 1992), we have demonstrated that animals which received testosterone paired with a specific side of the chamber showed a general increase in the amount of time spent on the paired side. However, in contrast to de Beun et al. (1992) and Alexander et al. (1994), we have shown that this increase is not specific to receiving testosterone paired with environmental stimuli, because animals that received testosterone in the home cage, exhibited a similar increase. Therefore, our data suggest that testosterone (a) produces no positive reinforcing consequences or (b) may act via a yet unspecified nonassociative mechanism (e.g., persistence) that would occlude the observance of a CPP. Finally, these studies emphasize the importance of the inclusion of appropriate control groups before confident conclusions can be drawn concerning the capacity of AAS to produce euphoric effects.

1 Although Experiment 1 revealed no significant differences in order of conditioning for animals receiving testosterone, drug-vehicle (TTTO) animals tended to show a smaller increase than vehicle-drug (OTTT) animals in the amount of time spent on the side of putative conditioning on the test day relative to baseline. We chose to employ the TTTO order of conditioning in the present study to provide the more conservative assessment of any potential CPP observed.

References

ALEXANDER, G. M., PACKARD, M. G., & HINES M. (1994). Testosterone has rewarding affective properties in male rats: Implications for the biological basis of sexual motivation. Behavioral Neuroscience, 108, 424-428.

ANDREW, R. J. (1972). Changes in search behavior in male and female chicks following different doses of testosterone. Animal Behaviour, 20, 741-750.

BECHARA, A., & VAN DER KOOY, D. (1992). A single brain stem substrate mediates the motivational effects of both opiates and food in nondeprived rats but not in deprived rats. Behavioral Neuroscience, 106, 351-363.

BOZARTH, M. A. (1987). Conditioned place preference: A parametric analysis using systemic heroin injections. In M. A. Bozarth (Ed.), Methods of assessing the reinforcing properties of abused drugs (pp. 241-273). New York: Springer-Verlag.

BROWER, K. J. (1992). Addictive potential of anabolic steroids. Psychiatric Annals, 22, 30-34.

CLARK, A. S., LINDENFELD, R. C., & GIBBONS, C. H. (1993). Anabolic steroids and brain reward. Society for Neuroscience Abstract, 19, 1460.

DE BEUN, R., JANSEN, E., SLANGEN, J. L., & VAN DE POLL, N. E. (1992). Testosterone as appetitive and discriminative stimulus in rats. Physiology and Behavior, 52, 629-643.

DE BEUN, R., JANSEN, E., SMEETS, M. A., NIESING, J., SLANGEN, J. L., & VAN DE POLL, N. E. (1991). Estradiol-induced conditioned taste-aversion and place aversion in rats: Sex- and dose-dependent effects. Physiology and Behavior, 50, 995-1000.

DE BEUN, R., PEETERS, B. W. M. M., & BROEKKAMP, C. L. E. (1993). Stimulus characterization of estradiol applying a crossfamiliarization taste aversion procedure in female mice. Physiology and Behavior, 53, 715-719.

GANESAN, R., ROSELLINI, R. A., & SVARE, B. (1993). Investigation of anabolic steroids in two taste aversion paradigms. Appetite, 20, 1-11.

HAUPT, H. A., & ROVERE, G. D. (1984). Anabolic steroids: A review of the literature. American Journal of Sports Medicine, 12:469-484.

KASHIN, K. B., & KLEBER, H. D. (1989). Hooked on hormones? An anabolic steroid addiction hypothesis. JAMA-Journal of the Medical Association, 262, 3166-3170.

KOOB, G. F., & GOEDERS, N. E. (1989). Neuroanatomical substrates of drug self-administration. In J. M. Liebman & S. J. Cooper (Eds.), The neuropharmacological basis of reward (pp. 214-263). Oxford: Clarendon Press.

MACINTOSH, N. J. (1974). The psychology of animal learning. London: Academic Press.

MCEWEN, B. S., COIRINI, H., & SCHUMACHER, M. (1990). Steroid effects on neuronal activity: When is the genome involved? In Steroids and Neuronal Activity (Ciba Foundation Symposium 153) (pp. 3-!2). New York: John Wiley & Sons.

MIELE, J., ROSELLINI, R. A., & SVARE, B. (1988). Estradiol benzoate can function as an unconditioned stimulus in a conditioned taste aversion paradigm. Hormones and Behavior, 22, 116-130.

MITCHELL, J. B., & STEWART, J. (1989). Effects of castration, steroid replacement, and sexual experience on mesolimbic dopamine and sexual behaviors in the male rat. Brain Research, 491, 116-127.

MUCHA, R. F., & IVERSEN, S. D. (1984). Reinforcing properties of morphine and naloxone revealed by conditioned place preferences: A procedural examination. Psychopharmacology, 82, 241-247.

PAVLOV, I. P. (1927). Conditioned reflexes. Oxford: Oxford University Press.

PEETERS, B. W. M. M., SMETS, R. J. M., & BROEKKAMP, C. L. E. (1992). Sex steroids possess distinct stimulus properties in female and male mice. Brain Research Bulletin, 28, 319-321.

REID, L. D., HUNTER, G. A., BEAMAN, C. M., & HUBBELL, C. L. (1985). Toward understanding ethanol's capacity to be reinforcing: A conditioned place preference following injections of ethanol. Pharmacology, Biochemistry, and Behavior, 22, 483-487.

REID, L. D., MARGLIN, S. H., MAGGIE, M. E., & HUBBELL, C. L. (1989). Measuring morphine's capacity to establish a place preference. Pharmacology, Biochemistry, and Behavior, 33, 765-775.

RESCORLA, R. A. (1969). Pavlovian conditioned inhibition. Psychological Bulletin, 72, 77-94.

SCOLES, M. T., & SIEGEL, S. (1986). A potential role of saline trials in morphine-induced place preference conditioning. Pharmacology, Biochemistry, and Behavior, 25, 1169-1173.

STELLAR, J. R., & RICE, M. B. (1989). Pharmacological basis of intracranial self-stimulation reward. In J. M. Liebman & S. J. Cooper (Eds.), The neuropharmacological basis of reward (pp. 14-65). Oxford: Clarendon Press.

STEWART, J., KREBS, W. H., & KACZENDER, E. (1967). State-dependent learning produced with steroids. Nature, 216, 1223-1224.

SVARE, B. (1990). Anabolic steroids and behavior: A preclinical research prospectus. In G. C. Lin & L. Erinoff (Eds.), Anabolic Steroid Abuse, NIDA Research Monograph, 102, 224-241.

VAN DER KOOY, D. (1987). Place conditioning: A simple and effective method for assessing the motivational properties of drugs. In M. A. Bozarth (Ed.), Methods of assessing the reinforcing properties of abused drugs (pp. 229-240). New York: Springer-Verlag.

WEEKS, J. R., & COLLINS, R. J. (1987). Screening for drug reinforcement using intravenous self-administration in the rat. In M. A. Bozarth (Ed.), Methods of assessing the reinforcing properties of abused drugs (pp. 188-199). New York: Springer-Verlag.

< Prev		Next >

[ Back ]

Our benefits

Free extras

Main Menu

Service features