Kenneth D. Carr

Chronic food restriction: enhancing effects on drug reward and striatal cell signaling.

Chronic food restriction (FR) increases behavioral sensitivity to drugs of abuse in animal models and is associated with binge eating, which shares comorbidity with drug abuse, in clinical populations. Behavioral, biochemical and molecular studies conducted in this laboratory to elucidate the functional and mechanistic bases of these phenomena are briefly reviewed. Results obtained to date indicate that FR increases the reward magnitude and locomotor-activating effects of abused drugs, and direct dopamine (DA) receptor agonists, as a result of neuroadaptations rather than changes in drug disposition. Changes in striatal DA dynamics, and postsynaptic cell signaling and gene expression in response to D-1 DA receptor stimulation have been observed. Of particular interest is an upregulation of NMDA receptor-dependent MAP kinase and CaM Kinase II signaling, CREB phosphorylation, and immediate-early and neuropeptide gene expression in nucleus accumbens (NAc) which may facilitate reward-related learning, but also play a role in the genesis of maladaptive goal-directed behaviors. Covariation of altered drug reward sensitivity with body weight loss and recovery suggests a triggering role for one of the endocrine adiposity hormones. However, neither acute nor chronic central infusions of leptin or the melanocortin 3/4 receptor agonist, MTII, have attenuated d-amphetamine reward or locomotor activation in FR rats. Interestingly, chronic intracerebroventricular leptin infusion in ad libitum fed (AL) rats produced a sustained decrease in food intake and body weight that was accompanied by a reversible potentiation of rewarding and locomotor-activating effects of d-amphetamine. This raises the interesting possibility that rapid progressive weight loss is sufficient to increase behavioral sensitivity to drugs of abuse. Whether weight loss produced by leptin infusion produces the same neuroadaptations as experimenter-imposed FR, and whether any of the observed neuroadaptations are necessary for expression of increased behavioral responsiveness to acute drug challenge remain to be investigated.

Evidence of increased dopamine receptor signaling in food-restricted rats.

It is well established that chronic food restriction enhances sensitivity to the rewarding and motor-activating effects of abused drugs. However, neuroadaptations underlying these behavioral effects have not been characterized. The purpose of the present study was to explore the possibility that food restriction produces increased dopamine (DA) receptor function that is evident in behavior, signal transduction, and immediate early gene expression. In the first two experiments, rats received intracerebroventricular (i.c.v.) injections of the D1 DA receptor agonist SKF-82958, and the D2/3 DA receptor agonist quinpirole. Both agonists produced greater motor-activating effects in food-restricted than ad libitum-fed rats. In addition, Fos-immunostaining induced by SKF-82958 in caudate-putamen (CPu) and nucleus accumbens (Nac) was greater in food-restricted than ad libitum-fed rats, as was staining induced by quinpirole in globus pallidus and ventral pallidum. In the next two experiments, neuronal membranes prepared from CPu and Nac were exposed to SKF-82958 and quinpirole. Despite the documented involvement of cyclic AMP (cAMP) signaling in D1 DA receptor-mediated c-fos induction, stimulation of adenylyl cyclase (AC) activity by SKF-82958 in CPu and Nac did not differ between groups. Food restriction did, however, decrease AC stimulation by the direct enzyme stimulant, forskolin, but not NaF or MnCl(2), suggesting a shift in AC expression to a less catalytically efficient isoform. Finally, food restriction increased quinpirole-stimulated [(35)S]guanosine triphosphate-gammaS binding in CPu, suggesting that increased functional coupling between D2 DA receptors and G(i) may account for the augmented behavioral and pallidal c-Fos responses to quinpirole. Results of this study support the hypothesis that food restriction leads to neuroadaptations at the level of postsynaptic D1 and D2 receptor-bearing cells which, in turn, mediate augmented behavioral and transcriptional responses to DA. The signaling pathways mediating these augmented responses remain to be fully elucidated.

Feeding, drug abuse, and the sensitization of reward by metabolic need

The incentive-motivating effects of external stimuli are dependent, in part, upon the internal need state of the organism. The increased rewarding efficacy of food as a function of energy deficit, for example, has obvious adaptive value. The enhancement of food reward extends, however, to drugs of abuse and electrical brain stimulation, probably due to a shared neural substrate. Research reviewed in this paper uses lateral hypothalamic electrical stimulation to probe the sensitivity of the brain reward system and investigate mechanisms through which metabolic need, induced by chronic food restriction and streptozotocin-induced diabetes, sensitizes this system. Results indicate that sensitivity to rewarding brain stimulation varies inversely with declining body weight. The effect is not mimicked by pharmacological glucoprivation or lipoprivation in ad libitum fed animals; sensitization appears to depend on persistent metabolic need or adipose depletion. While the literature suggests elevated plasma corticosterone as a peripheral trigger of reward sensitization, sensitization was not reversed by meal-induced or pharmacological suppression of plasma corticosterone. Centrally, reward sensitization is mediated by opioid receptors, since the effect is reversed by intracerebroventricular (i.c.v.) infusion of naltrexone, TCTAP (μ antagonist) and nor-binaltorphimine (κ antagonist). The fact that these same treatments, as well as i.c.v. infusion of dynorphin A antiserum, block the feeding response to lateral hypothalamic stimulation suggests that feeding and reward sensitization are mediated by a common opioid mechanism. Using in vitro autoradiography, radioimmunoassays and a solution hybridization mRNA assay, brain regional μ and κ opioid receptor binding, levels of prodynorphin-derived peptides, and prodynorphin mRNA, respectively, were measured in food-restricted and diabetic rats. Changes that could plausibly be involved in reward sensitization are discussed, with emphasis on the increased dynorphin A1–8 and prodynorphin mRNA levels in lateral hypothalamic neurons that innervate the pontine parabrachial nucleus, where μ binding decreased and κ binding increased. Finally, the possible linkage between metabolic need and activation of a brain opioid mechanism is discussed, as is evidence supporting the relevance of these findings to drug abuse.

Neuroanatomical patterns of Fos-like immunoreactivity induced by a palatable meal and meal-paired environment in saline- and naltrexone-treated rats

Opioid antagonists block the positive hedonic response to food taste and are potent inhibitors of palatability-driven feeding. However, the specific brain regions within which opioid peptide secretion contributes to the maintenance of palatability-driven feeding have not been clearly established. In the present study, c-Fos immunohistochemistry was used to identify regions rostral to the hindbrain that display cellular activation in response to a palatable meal and the meal-paired environment. Further, it was determined whether any of the cellular responses could be prevented by pretreating animals with naltrexone. Twenty brain regions known to be involved in gustation, appetite and reward functions were examined. Ingestion of the palatable meal (3.0 g of 30% shortening, 20% sucrose and 50% powdered Purina rat chow) increased Fos-like immunoreactivity (FLI) in lateral hypothalamus (LH), ventral tegmentum (VTA) and medial preoptic area (MPOA), and decreased FLI in the habenula (Hab). The meal-paired environment increased FLI in the VTA and nucleus accumbens shell (NAC shell). Naltrexone (1.0 mg/kg, i.p.) did not block consumption of the small meal but did prevent all of the distinctive increases in FLI induced by the meal and meal-paired environment. Since naltrexone, alone, increased FLI in VTA, NAC shell, central amygdala (ceA) and laterodorsal bed nucleus of the stria terminalis (BSTLD), the blunting of ingestion reward by naltrexone may result from direct or transsynaptic activating effects on opponent neuronal activity within this highly interconnected set of structures that mediate and modulate reward.

Chronic food restriction and weight loss produce opioid facilitation of perifornical hypothalamic self-stimulation

Electrical stimulation frequency thresholds for lateral hypothalamic (LH) self-stimulation were monitored throughout a 3 week period of food restriction and a subsequent 3 week period of re-feeding. Rats with electrodes placed in the perifornical LH were sensitive to this dietary manipulation as evidenced by a high positive correlation between body weight and self-stimulation threshold. Rats with electrodes in the zona incerta/subincertal region or ventral hypothalamus displayed little or no change in threshold. Lateral ventricular injection of naltrexone (200.0 nM) reversed the decline in threshold that was otherwise present during food restriction in rats with perifornical placements. Naltrexone had no effect on thresholds of rats with placements outside the perifornical region. These findings suggest that food restriction and weight loss activate an opioid mechanism that facilitates perifornical LH self-stimulation. The documented association of perifornical LH with the phenomenon of stimulation-induced feeding, and the reciprocal connections between this region and gustatory structures, supports the hypothesis that facilitation of self-stimulation by food restriction is related to the natural phenomenon of positive alliesthesia (i.e. the hunger-dependency of food reward).

Regulation of α and β components of noradrenergic cyclic AMP response in cortical slices

The cyclic AMP response to catecholamines in the rat cerebral cortex is mediated by both β- and α-adrenoceptors. The β-receptors cause a direct activation of adenylate cyclase whereas the alpha α-receptors play a modulatory role and act by potentiating the response to β stimulation. The present study investigated whether the functions of these two types of cyclic AMP-linked receptors are regulated differently by various physiological factors known to affect adrenoceptor function. It was found that treatments that affect central noradrenergic neuronal function including repeated administration of desmethylimipramine or lesion of central noradrenergic pathways produced selective changes in the cAMP response to β-receptor stimulation whereas treatments that affect adrenocortical function including ACTH of corticosterone administration and hypophysectomy or adrenalectomy produced selective changes in the potentiation response to α-receptor stimulation. The change in the α potentiation effect caused by corticosterone was found to be abolished in the presence of prazosin indicating that the hormone affects α1-adrenoceptor function. The results support the hypothesis that the β response in the cortex is under the control of the noradrenergic system while the α potentiation response is under the control of the adrenocortical system.

Insulin enhances striatal dopamine release by activating cholinergic interneurons and thereby signals reward

Insulin activates insulin receptors (InsRs) in the hypothalamus to signal satiety after a meal. However, the rising incidence of obesity, which results in chronically elevated insulin levels, implies that insulin may also act in brain centres that regulate motivation and reward. We report here that insulin can amplify action potential-dependent dopamine (DA) release in the nucleus accumbens (NAc) and caudate–putamen through an indirect mechanism that involves striatal cholinergic interneurons that express InsRs. Furthermore, two different chronic diet manipulations in rats, food restriction (FR) and an obesogenic (OB) diet, oppositely alter the sensitivity of striatal DA release to insulin, with enhanced responsiveness in FR, but loss of responsiveness in OB. Behavioural studies show that intact insulin levels in the NAc shell are necessary for acquisition of preference for the flavour of a paired glucose solution. Together, these data imply that striatal insulin signalling enhances DA release to influence food choices.

Curve-shift analysis of self-stimulation in food-restricted rats: Relationship between daily meal, plasma corticosterone and reward sensitization

Chronic food restriction lowers the threshold for lateral hypothalamic electrical self-stimulation (LHSS). This effect has previously been interpreted to reflect a sensitization of reward. In the present study a curve-shift method was used to explicitly differentiate effects of food restriction on brain stimulation rewarding efficacy and performance. Food restriction consistently shifted rate-frequency curves to the left, lowering the M-50 and Theta-0 parameters of rewarding efficacy. Asymptotic rates of reinforcement and slopes of rate-frequency functions were unaffected, confirming that food restriction does not facilitate LHSS by enhancing performance. In this and previous studies, LHSS in food-restricted rats was measured in the period immediately preceding the daily meal when hunger (i.e., period since last meal) and plasma corticosterone are at peak levels. In the light of evidence that corticosterone may regulate sensitivity of the mesolimbic dopamine pathway and account for the sensitizing effect of stress on psychomotor effects of opiates and stimulants, LHSS and corticosterone were measured in the immediate pre-and post-meal periods. While all food-restricted rats displayed elevated corticosterone levels in the pre-meal period and generally displayed a decline to control levels in the post-meal period, the sensitization of reward was not reversed in the post-meal period. These results indicate that chronic food restriction produces a sensitization of reward that does not depend upon the acute state of hunger that precedes the daily meal and does not vary with dynamic changes in plasma corticosterone level.

Stereotypies elicited by injection of N-propylnorapomorphine into striatal subregions and nucleus accumbens

Injection of the dopamine (DA) agonist R-(-)-N-n-propylnorapomorphine (NPA; 5-40 micrograms) into anterior ventral striatal sites (either lateral (VL) or medial (VM) elicited dose-dependent oral and sniffing stereotypies of rapid onset, long duration and high intensity. In contrast, injection into anterior dorsolateral (DL) or posterior ventral (lateral (PL) or medial (PM] sites produced little oral and moderate sniffing behavior of slower onset, shorter duration and low intensity. Injection into the dorsomedial (DM) striatum produced intermediate effects. Intra-accumbens NPA elicited weak oral activity and moderate sniffing which was similar in onset, duration and intensity to the least sensitive striatal sites (DL, PM and PL). In other experiments, DA receptors were inactivated with the irreversible blocking agent N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline (EEDQ; 6 mg/kg) and behavioral recovery was monitored by challenge with 20 micrograms NPA into the VL or the nucleus accumbens (NA) at various times after EEDQ. Sniffing behavior recovered rapidly (normal by day 4 in both regions), whereas oral activity required 8 (NA) and 12 days (VL) to return to control levels. The results are discussed in terms of a possible topographic distribution of behavior in the striatum. Alternatively, heterogeneity of DA receptor density may account for these findings.

Medial thalamic injection of opioid agonists: μ-agonist increases while κ-agonist decreases stimulus thresholds for pain and reward

Selective agonists for mu- and kappa-opioid receptor types were infused, bilaterally, into the intralaminar central lateral nucleus of the rat. Subcataleptic doses of the mu-agonist, DAGO (0.25 and 1.0 microgram), elevated tailshock threshold for eliciting pain vocalization and motor responses. The hyperalgesic effect of U50,488 is not likely to be the result of antagonist action at a mu 2-isoreceptor; the general mu-antagonist, naloxone, and its less lipophilic quaternary analogue, both failed to produce a significant reduction in pain thresholds. Paralleling their effects on pain, DAGO and U50,488 elevated and reduced, respectively, lateral hypothalamic electrical stimulation threshold for positive reinforcement. These results suggest that medial thalamic opioid mechanisms are not exclusively involved in pain modulation but may generally regulate the responsiveness of the organism to motivating stimuli. Moreover, mu- and kappa-receptors may mediate opposite behavioral effects of opioid peptides.