After natural menopause in women, androstenedione becomes the primary hormone secreted by the residual follicle deplete ovaries. Two independent studies, in rodents that had undergone ovarian follicular depletion, found that higher serum androstenedione levels correlated with increased working memory errors. This led to the hypothesis that androstenedione impairs memory. The current study directly tested this hypothesis, examining the cognitive effects of androstenedione administration in a rodent model. Middle-aged ovariectomized rats received vehicle or one of two doses of androstenedione (4 or 8 mg/kg daily). Rats were tested on a spatial working and reference memory maze battery including the water radial arm maze, Morris maze, and delay-match-to-sample task. Results showed that androstenedione at the highest dose impaired reference memory and working memory, including ability to maintain performance as memory demand was elevated. The latter was true for both high temporal demand memory retention of one item of spatial information, as well as the ability to handle multiple items of spatial working memory information. Glutamic acid decarboxylase (GAD) levels were measured in multiple brain regions to determine whether the gamma-aminobutyric acid (GABA) system mediates androstenedione's cognitive impairments. Results showed that higher entorhinal cortex GAD levels were correlated with poorer Morris maze performance, regardless of androstenedione treatment. These findings suggest that androstenedione, the main hormone produced by the follicle deplete ovary, is detrimental to spatial learning, reference memory, and working memory, and that spatial reference memory performance might be related to the GABAergic system.

The capability of cocaine-associated stimuli in eliciting craving in human addicts, even after extended periods of abstinence, is modeled in animals using cue reinstatement of extinguished cocaine-seeking behavior. This study aimed to examine brain activation in response to cocaine cues in this model apart from activation produced by test novelty using a novel cue control. Rats trained to self-administer cocaine paired with either an oscillating light or tone cue underwent daily extinction training and were then tested for reinstatement of extinguished cocaine-seeking behavior elicited by response-contingent presentations of either their assigned cocaine-paired cue or the alternate, novel cue. Additional controls received saline infusions and cue presentations yoked to a cocaine-trained rat. Brains were harvested for Fos immunohistochemistry immediately after the 90-min reinstatement test. Surprisingly, conditioned and novel cues both reinstated responding to a similar degree; however magnitude of reinstatement did vary by cue modality with the greatest reinstatement to the light cues. In most brain regions, Fos expression was enhanced in rats with a history of cocaine training regardless of cue type with the exception of the Cg1 region of the anterior cingulate cortex, which was sensitive to test cue modality. Also Fos expression within the dorsomedial caudate-putamen was correlated with responding in the novel, but not conditioned, cue groups. In subsequent experiments, we observed a similar pattern of reinstatement in rats trained and tested for sucrose-seeking behavior, whereas rats trained and tested with the cues only reinstated to a novel light and tone, but not a familiar cue. The results suggest that novel cues reinstate responding to a similar extent as conditioned cues regardless of whether animals have a history of operant-delivered drug or a natural reinforcer. Furthermore, similar brain circuits as those involved in cocaine-seeking behavior are activated by novel cues, suggesting converging processes exist to drive conditioned and novel reinforcement seeking.

Nicotine is thought to underlie the reinforcing and dependence-producing effects of tobacco-containing products. Nicotine supports self-administration in rodents, although measures of its reinforcing effects are often confounded by procedures that are used to facilitate acquisition, such as food restriction, prior reinforcement training, or response-contingent co-delivery of a naturally reinforcing light. This study examined whether rats acquire nicotine self-administration in the absence of these facilitators. A new mathematical modeling procedure was used to define the criterion for acquisition and to determine dose-dependent differences in rate and asymptote levels of intake. Rats were trained across 20 daily 2-h sessions occurring 6 days/week in chambers equipped with active and inactive levers. Each active lever press resulted in nicotine reinforcement (0, 0.015, 0.03, 0.06 mg/kg, IV) and retraction of both levers for a 20-s time out, whereas inactive lever presses had no consequences. Acquisition was defined by the best fit of a logistic function (i.e., S-shaped) versus a constant function (i.e., flat line) for reinforcers obtained across sessions using a corrected Akaike information criterion (AICc) as a model selection tool. The results showed an inverted-U shaped function for dose in relation to the percentage of animals that acquired nicotine self-administration, with 46% acquiring at 0.015 mg/kg, 73% at 0.03 mg/kg, and 58% at 0.06 mg/kg. All saline rats failed to acquire as expected. For rats that acquired nicotine self-administration, multiple model comparisons demonstrated that the asymptote (highest number of reinforcers/session) and half learning point (h; session during which half the assymptote had been achieved) were justified as free parameters of the reinforcers/session function, indicating that these parameters vary with nicotine dose. Asymptote exhibited an inverted U-shaped function across doses and half learning point exhibited a negative relationship to dose (i.e., the higher the dose the fewer sessions to reach h). These findings suggest that some rats acquire nicotine self-administration without using procedures that confound measures of acquisition rate. Furthermore, the modeling approach provides a new way of defining acquisition of drug self-administration that takes advantage of using all data generated from individual subjects and is less arbitrary than some criteria that are currently used.