Annemarie M. Baars

Standard housed rats are more sensitive to rewards than enriched housed rats as reflected by their anticipatory behaviour.

The present study was designed to investigate the effects of potentially stressful standard housing conditions for laboratory rats on the sensitivity to rewards as reflected by their anticipatory behaviour for sucrose. This anticipatory response is evoked in a conditioning paradigm in which a sucrose reward is repeatedly announced by a stimulus. The underlying neurocircuitry of this anticipatory response in expectation of a reward involves mesolimbic dopaminergic systems of which it is known that they can be sensitised by stressors. The results show that the anticipatory response for the sucrose reward is stronger in the standard housed animals which indicates that these animals are more sensitive to the reward than the enriched animals. From this, it is concluded that standard housed rats are stressed which is likely to be caused by deprivation of the ability to satisfy behavioural needs in these impoverished housing conditions.

Roles of the Basolateral Amygdala and Hippocampus in Social Recognition in Rats

Lesions of the amygdala or hippocampus have a large impact on social behavior of rats. In this study we investigated whether a social recognition test was also affected by those lesions. An NMDA-induced lesion of the basolateral amygdala did not impair the ability to distinguish a familiar from an unfamiliar juvenile rat. It was argued that the cortico-medial amygdala may be more important for social recognition than the basolateral amygdala. Fimbria-transected rats could no longer distinguish a familiar from an unfamiliar juvenile. Moreover, during all encounters they spent less time investigating the juvenile. The precise nature of this deficit, especially the reason for the overall reduced social investigation time, could not be specified with the classical procedure of the social recognition test.

Mu-opioid receptor knockout mice show diminished food-anticipatory activity

We have previously suggested that during or prior to activation of anticipatory behaviour to a coming reward, µ‐opioid receptors are activated. To test this hypothesis schedule induced food‐anticipatory activity in µ‐opioid receptor knockout mice was measured using running wheels. We hypothesized that µ‐knockout mice show little food‐anticipatory activity. In wildtype mice we observed that food‐anticipatory activity increased proportional to reduced food intake levels during daily scheduled food access, and thus reflects the animals physiological need for food. µ‐Knockout mice do not adjust their schedule induced running wheel behaviour prior to and during feeding time in the same way as wildtype mice; rather than showing more running wheel activity before than during feeding, they showed an equal amount of activity before and during feeding. As food‐anticipatory activity is dependent on the mesolimbic dopamine system and µ‐opioid receptors regulate dopaminergic activity, these data suggest a change in the dopamine systems activity in µ‐knockout mice. As we observed that µ‐knockout mice tended to show a stronger locomotor activity response than wildtype mice to the indirect dopamine agonist d‐amphetamine, it appears that the dopaminergic system per se is intact and sensitive to activation. We found no differences in the expression of pro‐opiomelanocortin, a precursor of endogenous endorphin, in the arcuate nucleus between µ‐knockout mice and wildtype mice during restricted feeding, showing that the µ‐opioid receptor does not regulate endogenous endorphin levels. These data overall suggest a role for µ‐opioid receptors in adapting reward related behaviour to the requirements of the environment.

Decision-making performance is related to levels of anxiety and differential recruitment of frontostriatal areas in male rats

In humans, high levels of anxiety are associated with poor performance in the Iowa Gambling Task (IGT). The IGT measures decision-making under conditions of uncertainty. In this study, we investigated the association between anxiety and decision-making in rats. Rats were screened for anxiety on the elevated plus maze (EPM) and subsequently tested in a rat analogue of the IGT (r-IGT). We explored the role of frontostriatal areas related to r-IGT performance using c-fos immunohistochemistry following the last training-session. High levels of anxiety were associated with poor r-IGT performance: high anxious rats made fewer choices for the advantageous option and collected fewer sucrose pellets in the r-IGT than low anxious rats. Analysis of win-stay/lose-shift behaviour of choices for the advantageous option revealed that good performing-low anxious subjects showed an increase in win-stays and a decrease in lose-shifts across trial blocks while poor performing-high anxious subjects did not. Furthermore, decision-making performance and, indirectly, anxiety levels were related to neural activity in parts of the medial prefrontal cortex, that is prelimbic and infralimbic cortex, and in parts of the striatum, that is nucleus accumbens shell and core. These data suggest a similar frontostriatal circuitry underlying affective decision-making in humans and rats.

Transient Inactivation of the Medial Prefrontal Cortex Affects Both Anxiety and Decision-Making in Male Wistar Rats

In both humans and rats high levels of anxiety impair decision-making in the Iowa gambling task (IGT) in male subjects. Expression of the immediate early gene c-fos as marker of neural activity in rat studies indicated a role of the medial prefrontal cortex (prelimbic and infralimbic region; mPFC) in mediating the relationship between anxiety and decision-making. To delineate this relationship further and assess the underlying neurobiology in more detail, we inactivated in the present study the mPFC in male rats using a mixture of the GABA-receptor agonists muscimol and baclofen. Rats were exposed to the elevated plus maze (EPM) to measure effects on anxiety and to the rodent version of the IGT (r-IGT). Inactivation led to increased levels of anxiety on the EPM, while not affecting general activity. The effect in the r-IGT (trials 61–120) was dependent on levels of performance prior to inactivation (trial 41–60): inactivation of the mPFC hampered task performance in rats, which already showed a preference for the advantageous option, but not in rats which were still choosing in a random manner. These data suggest that the mPFC becomes more strongly involved as rats have learned task-contingencies, i.e., choose for the best long-term option. Furthermore they suggest, along with the data of our earlier study, that both anxiety and decision-making in rats are mediated through a neural circuitry including at least the mPFC. The data are discussed in relation to recent data of rodent studies on the neural circuitry underlying decision-making.

Male and female Wistar rats differ in decision-making performance in a rodent version of the Iowa Gambling Task

The Iowa Gambling Task (IGT) measures decision-making processes by simulating real-life decisions involving reward, punishment, and uncertainty of outcomes. In humans, men show more choices for the advantageous option than women. Here, we investigated sex differences in a rat model of the IGT (r-IGT). In our r-IGT mildly food-deprived rats learn to differentiate a long-term advantageous arm from a long-term disadvantageous arm differing in frequency and amount of sugar pellets as well as unpalatable but not uneatable quinine-treated sugar pellets. We also used a T-maze discrimination procedure in which rats learn to differentiate a high from a low reward arm to further explore sex differences in reward-related decision-making. In line with human data, male rats showed a stronger task progression of choices for the advantageous option than female rats. Furthermore, male rats showed more win-stay and less lose-shift behaviour in the advantageous arm as the task progressed than female rats. Whilst both male and female rats had a stronger preference for the high over the small reward arm in the T-maze, males increased this preference over sessions, whilst females did not. These data are discussed in relation to sex differences in processing rewards and punishments.

Announced rewards counteract the impairment of anticipatory behaviour in socially stressed rats.

It is known that stress can influence the sensitivity to rewarding stimuli. Previous observations revealed that socially stressed rats do not display an appetitive behavioural response in anticipation of a reward. A previous study showed that this insensitivity to rewards (anhedonia) could be restored by chronic administration of an antidepressant. Several lines of evidence exist for the role of dopamine in the mechanism of action of antidepressant treatments concerning their therapeutic effect on anhedonia. Therefore, it was hypothesized that regular activation of the reward system, that involves mesolimbic dopaminergic systems, could counteract the effect of social stress on reward-sensitivity. For this, it was investigated whether a treatment of regular reward announcements could prevent the development of anhedonia. This was confirmed by the result that socially stressed rats that received this treatment were able to display anticipatory behaviour which is characterized by increased activity after presentation of a stimulus that was previously associated with a sucrose reward. Surprisingly, a non-treated socially stressed group, that did not show an anticipatory response for sucrose, did display anticipatory behaviour for another type of reward (enriched cage). It seems that, although one might assume the existence of an anhedonic state based upon the absence of anticipatory activity towards a sucrose reward, this assumption cannot be generalised to other types of reward. It will be discussed whether this might be caused by the highly rewarding properties of the enriched cage which probably has a therapeutical efficacy of its own.

Time-dependent effects of corticosterone on reward-based decision-making in a rodent model of the Iowa Gambling Task.

Corticosteroid hormones, released after stress, are known to change neuronal activity in two time-domains: within minutes via non-genomic pathways and with a delay of >1 h through pathways involving transcriptional regulation. Recent evidence in rodents and humans indicates that these two modes of corticosteroid action differently affect cognitive tasks. Here, we investigated whether reward-based decision-making, in a rat model of the Iowa Gambling Task (rIGT), is also differently altered by rapid versus delayed actions of corticosterone. We targeted the rapid and delayed time domain by injecting corticosterone (CORT, 1 mg/kg, s.c.) at 30 min (rapid) or 180 min (delayed) respectively prior to behavioural testing, during the final 3 days of the behavioural paradigm. In saline treated rats, the number of visits to the disadvantageous arm decreased over trial blocks, whilst this was attenuated when CORT was administered 30 min before testing. This attenuation was associated with a significantly increased c-Fos expression in the lateral orbitofrontal cortex and insular cortex, and a trend for an increase in the infralimbic cortex. The rapid corticosteroid effect contrasted with treatment 180 min before testing, where the number of visits to the disadvantageous arm as well as c-Fos labelling was not affected. These findings indicate that rapid corticosteroid actions impair reward-based decision-making.

Nociception and Conditioned Fear in Rats: Strains Matter

When using rats in pain research, strain-related differences in outcomes of tests for pain and nociception are acknowledged. However, very little is known about the specific characteristics of these strain differences. In this study four phylogenetically distant inbred rat strains, i.e. Wistar Kyoto (WKY), Fawn Hooded (FH), Brown Norway (BN) and Lewis (LE), were investigated in different tests related to pain and nociception. During Pavlovian fear conditioning, the LE and WKY showed a significantly longer duration of freezing behaviour than the FH and BN. Additionally, differences in c-Fos expression in subregions of the prefrontal cortex and amygdala between rat strains during retrieval and expression of conditioned fear were found. For example, the BN did not show recruitment of the basolateral amygdala, whereas the WKY, FH and LE did. During the hot plate test, the WKY and LE showed a lower thermal threshold compared to the BN and FH. In a follow-up experiment, the two most contrasting strains regarding behaviour during the hot plate test and Pavlovian fear conditioning (i.e. FH and WKY) were selected and the hot plate test, Von Frey test and somatosensory-evoked potential (SEP) were investigated. During the Von Frey test, the WKY showed a lower mechanical threshold compared to the FH. When measuring the SEP, the FH appeared to be less reactive to increasing stimulus intensities when considering both peak amplitudes and latencies. Altogether, the combined results indicate various differences between rat strains in Pavlovian fear conditioning, nociception related behaviours and nociceptive processing. These findings demonstrate the necessity of using multiple rat strains when using tests including noxious stimuli and suggest that the choice of rat strains should be considered. When selecting a strain for a particular study it should be considered how this strain behaves during the tests used in that study.

Individual Variation in Alcohol Intake Predicts Reinforcement, Motivation, and Compulsive Alcohol Use in Rats

BACKGROUND Alcohol is one of the most commonly used psychoactive substances. Prolonged alcohol use can result in alcohol use disorder (AUD), characterized by excessive and compulsive alcohol consumption. Importantly, however, the development of AUD only happens in a minority of individuals who consume alcohol. To understand the individual vulnerability for AUD, models that capture both the individual variability in alcohol consumption and the transition from casual to compulsive alcohol use are essential. METHODS Individual variability in voluntary alcohol intake and the preference for alcohol were assessed under continuous alcohol access (CAA) and intermittent-every-other-day alcohol access (IAA) schedules in the home cage using outbred Lister Hooded rats. Subsequently, the reinforcing properties of alcohol were tested in an operant setting. In subsequent experiments, we performed a quinine adulteration experiment to assess inflexible alcohol consumption and blood alcohol levels (BALs) were assessed after voluntary alcohol consumption. RESULTS We found marked individual differences in alcohol consumption and preference under both access schedules, whereby subgroups of high- and low-alcohol-drinking rats (HD and LD) could be identified. HD with IAA increased their alcohol intake over days in the first month, whereas LD did not. Moreover, when alcohol access time was extended from 7 to 24 h/d for rats with IAA, alcohol intake profoundly increased in HD with IAA, whereas LD with IAA maintained low levels of alcohol intake. Furthermore, HD earned more alcohol than LD under both fixed ratio and progressive ratio schedules of reinforcement. We further found that HD continued their intake of a quinine-adulterated alcohol solution to a larger extent than LD and HD showed higher BALs after 30 minutes of alcohol consumption. CONCLUSIONS These profound individual differences in alcohol intake, reinforcement, motivation, and AUD-like behavior provide a promising tool to unravel the neurobehavioral underpinnings of individual vulnerability for AUD.