Using rat ultrasonic vocalization to study the neurobiology of emotion: from basic science to the development of novel therapeutics for affective disorders.

The use of ultrasonic vocalizations as an experimental tool for studying emotional states in rodents has led to an increased understanding of the basic science of affect as well as the development of novel diagnostics and therapeutics for the treatment of affective disorders. At the behavioral level, the rules that govern the generation of affective 'feeling' states are similar to those of the psychophysics of sensory perception. Emotions are elicited primarily in response to active social stimuli. A linear increase in affective response requires a logarithmic increase in stimulation and habituation of a given affective response allows for transition across the cycle of emotional/affective states (approach→consummatory phase→avoidance). At the neuronal level, the coordinated expression of affective responses in the medial prefrontal cortex is orchestrated by rhythmic activity, which is initiated and maintained by a variety of short-term and long-term synaptic plasticity processes. An objective measure of affective states may emerge from these psychophysical and neuronal properties of emotion. Enhancing synaptic plasticity with pharmacological agents that modulate NMDA receptor activity as well as IGFI receptor activity may have therapeutic potential for the treatment of affective disorders.

The long-lasting antidepressant effects of rapastinel (GLYX-13) are associated with a metaplasticity process in the medial prefrontal cortex and hippocampus.

Rapastinel (GLYX-13) is an N-methyl-d-aspartate receptor (NMDAR) modulator that has characteristics of a glycine site partial agonist. Rapastinel is a robust cognitive enhancer and facilitates hippocampal long-term potentiation (LTP) of synaptic transmission in slices. In human clinical trials, rapastinel has been shown to produce marked antidepressant properties that last for at least one week following a single dose. The long-lasting antidepressant effect of a single dose of rapastinel (3mg/kg IV) was assessed in rats using the Porsolt, open field and ultrasonic vocalization assays. Cognitive enhancement was examined using the Morris water maze, positive emotional learning, and contextual fear extinction tests. LTP was assessed in hippocampal slices. Dendritic spine morphology was measured in the dentate gyrus and the medial prefrontal cortex. Significant antidepressant-like or cognitive enhancing effects were observed that lasted for at least one week in each model. Rapastinel facilitated LTP 1day-2weeks but not 4weeks post-dosing. Biweekly dosing with rapastinel sustained this effect for at least 8weeks. A single dose of rapastinel increased the proportion of whole-cell NMDAR current contributed by NR2B-containing NMDARs in the hippocampus 1week post-dosing, that returned to baseline by 4weeks post-dosing. The NMDAR antagonist 3-(2-carboxypiperazin-4-yl)propyl-1-phosphonic acid (CPP) blocked the antidepressant-like effect of rapastinel 1week post dosing. A single injection of rapastinel also increased mature spine density in both brain regions 24h post-dosing. These data demonstrate that rapastinel produces its long-lasting antidepressant effects via triggering NMDAR-dependent processes that lead to increased sensitivity to LTP that persist for up to two weeks. These data also suggest that these processes led to the alterations in dendritic spine morphologies associated with the maintenance of long-term changes in synaptic plasticity associated with learning and memory.

Selective breeding for 50 kHz ultrasonic vocalization emission produces alterations in the ontogeny and regulation of rough-and-tumble play.

Ultrasonic vocalizations (USVs) are emitted by rodents and can signal either negative or positive affective states in social and nonsocial contexts. Our recent work has utilized selective breeding based upon the emission of 50 kHz USVs in response to standard cross species hand play-namely experimenters 'tickling' rats. Previous work has shown that high-tickle responsive animals (i.e., rats emitting abundant 50 kHz USVs) are gregarious and express enhanced positive emotional behaviors relative to animals exhibiting low 50 kHz USVs. The present study extends this work by examining the developmental profile of play behavior and the suppression of play behavior by predator (cat) odor in juvenile high-line and low-line animals. Results support dissociations in key play measures between these groups, with high-line animals emitting more dorsal contacts during play and low-line animals emitting more pinning behavior. For cat-odor induced play suppression, we found that high-line animals exhibit elevated suppression of play for a prolonged period compared to low-line rats. In contrast, low-line animals returned to normal levels of play just 1 day post-predator odor experience. These findings support the idea that emotional arousal may differ between these selectively bred groups, and extends previous work by demonstrating a possible influence of altered emotional learning and conditioning in these phenotypically different animals. One possibility is that high-line animals exhibit enhanced associative learning abilities leading to stronger negative contextual conditioning. These findings suggest that selection for positive or negative social-emotional phenotypes may also segregate genes that control emotional learning abilities in unanticipated ways.

Positive emotional learning is regulated in the medial prefrontal cortex by GluN2B-containing NMDA receptors.

In rats, hedonic ultrasonic vocalizations (USVs) is a validated model of positive affect and is best elicited by rough-and-tumble play. Here we report that modulation of GluN2B-containing NMDA receptors (NMDAR) in the medial prefrontal cortex (MPFC) is involved in positive emotional learning. Rough and tumble play increased both GluN1 and GluN2B NMDAR subunit mRNA and protein levels in the frontal cortex. GLYX-13, a GluN2B-preferring, NMDAR glycine-site partial agonist (1 mg/kg, i.v.) significantly increased positive emotional learning whereas the GluN2B receptor-specific antagonist, ifenprodil (10 mg/kg, i.p.), inhibited positive emotional learning. Animals selectively bred for low rates of hedonic USVs were returned to wild-type levels of positive emotional learning following GLYX-13 treatment. MPFC microinjections of GLYX-13 (0.1-10 µg/side) significantly increased rates of positive emotional learning. Thus GluN2B-containing NMDARs may be involved in positive emotional learning in the MPFC by similar mechanisms as spatial/temporal learning in the hippocampus.

Regional changes in gene expression after limbic kindling.

Repeated electrical stimulation results in development of seizures and a permanent increase in seizure susceptibility (kindling). The permanence of kindling suggests that chronic changes in gene expression are involved. Kindling at different sites produces specific effects on interictal behaviors such as spatial cognition and anxiety, suggesting that causal changes in gene expression might be restricted to the stimulated site. We employed focused microarray analysis to characterize changes in gene expression associated with amygdaloid and hippocampal kindling. Male Long-Evans rats received 1 s trains of electrical stimulation to either the amygdala or hippocampus once daily until five generalized seizures had been kindled. Yoked control rats carried electrodes but were not stimulated. Rats were euthanized 14 days after the last seizures, both amygdala and hippocampus dissected, and transcriptome profiles compared. Of the 1,200 rat brain-associated genes evaluated, 39 genes exhibited statistically significant expression differences between the kindled and non-kindled amygdala and 106 genes exhibited statistically significant differences between the kindled and non-kindled hippocampus. In the amygdala, subsequent ontological analyses using the GOMiner algorithm demonstrated significant enrichment in categories related to cytoskeletal reorganization and cation transport, as well as in gene families related to synaptic transmission and neurogenesis. In the hippocampus, significant enrichment in gene expression within categories related to cytoskeletal reorganization and cation transport was similarly observed. Furthermore, unique to the hippocampus, enrichment in transcription factor activity and GTPase-mediated signal transduction was identified. Overall, these data identify specific and unique neurochemical pathways chronically altered following kindling in the two sites, and provide a platform for defining the molecular basis for the differential behaviors observed in the interictal period.

Uncovering the molecular basis of positive affect using rough-and-tumble play in rats: a role for insulin-like growth factor I.

Positive emotional states have been shown to confer resilience to depression and anxiety in humans, but the molecular mechanisms underlying these effects have not yet been elucidated. In laboratory rats, positive emotional states can be measured by 50-kHz ultrasonic vocalizations (hedonic USVs), which are maximally elicited by juvenile rough-and-tumble play behavior. Using a focused microarray platform, insulin-like growth factor I (IGFI) extracellular signaling genes were found to be upregulated by hedonic rough-and-tumble play but not depressogenic social defeat. Administration of IGFI into the lateral ventricle increased rates of hedonic USVs in an IGFI receptor (IGFIR)-dependent manner. Lateral ventricle infusions of an siRNA specific to the IGFIR decreased rates of hedonic 50-kHz USVs. These results show that IGFI plays a functional role in the generation of positive affective states and that IGFI-dependent signaling is a potential therapeutic target for the treatment of depression and anxiety.

Rats selectively bred for low levels of 50 kHz ultrasonic vocalizations exhibit alterations in early social motivation.

In rats, the rates of 50 kHz ultrasonic vocalizations (USVs) can be used as a selective breeding phenotype and variations in this phenotype can be an indicator of affective states. The 50 kHz USV is elicited by rewarding stimuli (e.g., food, sexual behavior) and therefore can express a positive affective state. Conversely, the 22 kHz USV is elicited by aversive stimuli (e.g., presence of a predator, social defeat) indicating a negative affective state. In the present study, we tested the effect of selectively breeding for 50 kHz USVs on a variety of maternal social/emotional behaviors in young rat pups (PND 10-12). These measures consisted of an assessment of isolation calls and conditioned odor preference paradigm. Results indicate that animals selected for low levels of 50 kHz USVs show the greatest alterations in social behaviors compared to the control animals. The low line animals had an increase in isolation calls tested during place preference conditioning and a decrease in 50 kHz ultrasonic calls in all conditions. These same low line animals failed to show a typical preference for a maternally-associated odor during the place preference test. The different social behaviors of the high line animals did not consistently vary from those of the control group. These results have important implications for the study of genetic and epigenetic mechanisms underlying emotional states, and possibly contribute to the research underlying the emotional changes in developmental disorders such as autistic spectrum disorder by providing a novel animal model that displays communication deficits that are interdependent with significant social behavioral impairments.

Modeling depression: social dominance-submission gene expression patterns in rat neocortex.

Gene expression profiles in the cortex of adult Long-Evans rats as a function of a stressful social loss and victory in inter-male fighting encounters were examined. This social dominance and subordination model has been postulated to simulate early changes in the onset of depression in the losers. Microarrays were fabricated containing 45mer oligonucleotides spotted in quadruplicate and representing 1178 brain-associated genes. Dynamic range, discrimination power, accuracy and reproducibility were determined with standard mRNA "spiking" studies. Gene expression profiles in dominant and subordinate animals were compared using a "universal" reference design [Churchill GA (2002) Fundamentals of experimental design for cDNA microarrays. Nat Genet 32 (Suppl):490-495]. Data were analyzed by significance analysis of microarrays using rank scores [Tusher VG, Tibshirani R, Chu G (2001) Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA 98:5116-5121; van de Wiel MA (2004) Significance analysis of microarrays using rank scores. Kwantitatieve Methoden 71:25-37]. Ontological analyses were then performed using the GOMiner algorithm [Zeeberg BR, Feng W, Wang G, Wang MD, Fojo AT, Sunshine M, Narasimhan S, Kane DW, Reinhold WC, Lababidi S, Bussey KJ, Riss J, Barrett JC, Weinstein JN (2003) GoMiner: a resource for biological interpretation of genomic and proteomic data. Genome Biol 4(4):R28]. And finally, genes of special interest were further studied using quantitative reverse transcriptase polymerase chain reaction. Twenty-two transcripts were statistically significantly differentially expressed in the neocortex between dominant and subordinate animals. Ontological analyses revealed that significant gene changes were clustered primarily into functional neurochemical pathways associated with protein biosynthesis and cytoskeletal dynamics. The most robust of these were the increased expression of interleukin-18, heat shock protein 27, beta3-tubulin, ribosome-associated membrane protein 4 in subordinate animals. Interleukin-18 has been found to be over-expressed in human depression and panic disorder as well as other physiological stress paradigms [Takeuchi M, Okura T, Mori T, Akita K, Ohta T, Ikeda M, Ikegami H, Kurimoto M (1999) Intracellular production of interleukin-18 in human epithelial-like cell lines is enhanced by hyperosmotic stress in vitro. Cell Tissue Res 297(3):467-473] and heat shock proteins have been shown to be involved in the pathogenesis of many neurodegenerative and psychiatric disorders [Iwamoto K, Kakiuchi C, Bundo M, Ikeda K, Kato T (2004) Molecular characterization of bipolar disorder by comparing gene expression profiles of postmortem brains of major mental disorders. Mol Psychiatry 9(4):406-416; Pongrac JL, Middleton FA, Peng L, Lewis DA, Levitt P, Mirnics K (2004) Heat shock protein 12A shows reduced expression in the prefrontal cortex of subjects with schizophrenia. Biol Psychiatry 56(12):943-950]. Thus, the gene expression changes that we have observed here are consistent with and extend the observations found in the clinical literature and link them to the animal model used here thereby reinforcing its use to better understand the genesis of depression and identify novel therapeutic targets for its treatment.

Nucleus accumbens amphetamine microinjections unconditionally elicit 50-kHz ultrasonic vocalizations in rats.

The authors have hypothesized that, in adult rats, 50-kHz ultrasonic vocalizations (USVs) index a state characterized by high arousal and expectations of reward. This study was conducted to investigate whether dopamine agonism of the nucleus accumbens (NAcc) could evoke such an appetitive state, by examining the effects of NAcc amphetamine (AMPH) microinjections on USVs. Intra-NAcc AMPH injections (0.3, 1.0, 3.0, 10.0 microg unilaterally) produced robust, dose-dependent increases in 50-kHz USVs, which could not be accounted for by concomitant increases in locomotor activity (LA). However, AMPH injections into dorsal control caudate putamen sites produced a modest, dose-dependent increase in LA without significant increases in 50-kHz USVs. These findings indicate that NAcc AMPH microinjections selectively evoke 50-kHz USVs in rats, supporting the notion that dopamine elevations in the NAcc may unconditionally elicit a state of reward anticipation.

Evaluation of rat ultrasonic vocalizations as predictors of the conditioned aversive effects of drugs.

RATIONALE: Since cues that predict aversive outcomes can elicit both avoidance and 20 kHz ultrasonic vocalizations (USVs) in adult rats, 20 kHz USVs may also index the conditioned aversive effects of drugs. OBJECTIVE: We evaluated whether exposure to compartments associated with drugs with aversive effects would selectively increase 20 but not 50 kHz USVs in rats. METHOD: Rats were injected with naloxone (NAL) or lithium chloride (LiCl) and placed in one compartment or with saline (VEH) and placed in another compartment for three 50-min conditioning sessions. 20 kHz USVs, 50 kHz USVs, and time spent in each chamber were recorded during subsequent 15-min testing sessions during which rats had access to both compartments (expt 1) or were confined to the drug- or VEH-paired compartment (expt 2). RESULTS: In expt 1, animals conditioned either with NAL (0.3 and 3.0 mg/kg) or LiCl (10 and 30 mg/kg) emitted increased 20 kHz USVs in the drug-paired compartment, relative to VEH-conditioned controls. Conditioning with high doses of both drugs also increased conditioned place aversion and decreased emission of 50 kHz USVs. In expt 2, restriction of animals to the compartment paired with high doses of NAL and LiCl also increased emission of 20 kHz USVs and decreased 50 kHz USVs, relative to VEH-conditioned controls. CONCLUSIONS: In rats, cues associated with drugs with aversive effects increase 20 kHz USVs and decrease 50 kHz USVs, suggesting that USVs may provide a useful model for predicting the conditioned aversive effects of drugs.

Tickling induces reward in adolescent rats.

In adolescent rats, 50-kHz vocalizations are most evident during tickling and rough-and-tumble play. The following experiments evaluated whether 50-kHz vocalizations reflect positive social affect by determining (1) if tickling is a rewarding event, (2) if social or isolate housing conditions differentially influence the response (since housing condition has been found to effect the reward magnitude of social encounters), and (3) if drugs that work on mu-opiate receptors, which has been hypothesized to control positive social affect, modulate tickling. Tickling was positively reinforcing as demonstrated by elevated operant behavior, conditioned place preference, and approach measures. A significant negative correlation between vocalization rate and approach latency measures was found. Social housing reduced tickle-induced vocalizations and approach speeds compared to isolate housing. Naloxone (1 mg/kg) increased vocalization in the socially housed rats and decreased it in isolated Subjects (Ss). These findings suggest that tickling can be used to induce positive social affect in rodents, and that it is modulated by endogenous opioids.

50-kHz chirping (laughter?) in response to conditioned and unconditioned tickle-induced reward in rats: effects of social housing and genetic variables.

In these studies the incidence of conditioned and unconditioned 50-kHz ultrasonic vocalizations (USVs) in young rats was measured in response to rewarding manual tickling by an experimenter. We found that isolate-housed animals vocalize much more then socially housed ones, and when their housing conditions are reversed, they gradually shift their vocalization tendencies. Isolate-housed animals also show quicker acquisition of instrumental tasks for tickling, and exhibit less avoidance of tickling as compared to socially housed Ss. Isolate-housed animals also show rapid acquisition of 50-kHz USVs to a conditioned stimulus that predicts tickle reward, while socially housed animals do not. We successfully bred for high and low vocalization rates in response to tickling within four generations. The high tickle line showed quicker acquisition of an instrumental task for, as well as less avoidance of, tickling as compared to the random and low tickle lines. They also played more. Lastly, we found that the glutamate antagonist MK-801 can reduce tickle-induced 50-kHz USVs, but is resistant to opioid, dopamine and cholinergic stimulant and blocking agents. Overall, these results suggest that tickle evoked 50-kHz USVs may be a useful behavioral marker of positive social affect in rats. Difficulties with such concepts are also discussed.

Anticipation of rewarding electrical brain stimulation evokes ultrasonic vocalization in rats.

Adult rats emit increased rates of 50-kHz ultrasonic vocalizations (USVs) before receiving social and pharmacological rewards. This study sought to determine whether anticipation of rewarding electrical stimulation of the brain (ESB) would also elicit these vocalizations. In Experiments 1 and 2, rats showed increased 50-kHz USVs before receiving experimenter-delivered ventral tegmental area (VTA) and lateral hypothalamic (LH) ESB on a fixed time 20-s schedule. In Experiments 3 and 4, rats increased their rate of 50-kHz USVs in response to cues that predicted the opportunity to self-stimulate the VTA or LH. Interestingly, unexpected termination of either type of ESB evoked 20-kHz, rather than 50-kHz, USVs. In Experiment 5, a cue that predicted daily 1-hr feeding sessions increased 50-kHz USVs, whereas a cue that predicted footshock decreased 50-kHz USVs. These effects could not be explained simply by changes in locomotor activity or general arousal. Together, these findings support the hypothesis that short 50-kHz USVs may selectively index a state of reward anticipation in rats.

Paradoxical effects of serotonin and opioids in pemoline-induced self-injurious behavior.

Self-injurious behavior (SIB) is a symptom of various psychiatric disorders with differing etiologies. Although no generally effective pharmacological treatment of SIB is available, subsets of individuals exhibiting SIB have been found to respond to opioid antagonists and selective serotonin reuptake inhibitors (SSRIs). The present study evaluated the efficacy of these two treatments in the pemoline-induced model of self-biting behavior (SBB) in rats. Using a factorial design, adult rats receiving daily pemoline at 100 mg/kg or the peanut oil vehicle were pretreated with either distilled water vehicle (1 cc/kg), naltrexone (1 mg/kg), or paroxetine (1 mg/kg). Each day, animals were rated on the severity of SBB and also periodically behavioral changes were evaluated using various other outcome measures. Paroxetine significantly increased the severity of SBB induced by pemoline, while naltrexone only marginally increased the SBB. These results were not expected and suggest that further studies into the role of serotonin agonists and antagonists are needed in evaluating this model.

High-frequency ultrasonic vocalizations index conditioned pharmacological reward in rats.

We have proposed that short (<0.5 s), high-frequency (approximately 50 kHz) ultrasonic vocalizations ("50-kHz USVs") index a positive affective state in adult rats, because they occur prior to rewarding social interactions (i.e., rough-and-tumble play, sex). To evaluate this hypothesis in the case of nonsocial stimuli, we examined whether rats would make increased 50-kHz USVs in places associated with the administration of rewarding pharmacological compounds [i.e., amphetamine (AMPH) and morphine (MORPH)]. In Experiment 1, rats made a greater percentage of 50-kHz USVs on the AMPH-paired side of a two-compartment chamber than on the vehicle-paired side, even after statistical correction for place preference. In Experiment 2, rats made a higher percentage of 50-kHz USVs on the MORPH-paired side than on the vehicle-paired side, despite nonsignificant place preference. These findings support the hypothesis that 50-kHz USVs mark a positive affective state in rats and introduce a novel and rapid marker of pharmacological reward.

Anticipation of play elicits high-frequency ultrasonic vocalizations in young rats.

The authors provide initial documentation that juvenile rats emit short, high-frequency ultrasonic vocalizations (high USVs, approximately 55 kHz) during rough-and-tumble play. In an observational study, they further observe that these vocalizations both correlate with and predict appetitive components of the play behavioral repertoire. Additional experiments characterized eliciting conditions for high USVs. Without prior play exposure, rats separated by a screen vocalized less than playing rats, but after only 1 play session, separated rats vocalized more than playing rats. This findings suggested that high USVs were linked to a motivational state rather than specific play behaviors or general activity. Furthermore, individual rats vocalized more in a chamber associated with play than in a habituated control chamber. Finally, congruent and incongruent motivational manipulations modulated vocalization expression. Although play deprivation enhanced high USVs, an arousing but aversive stimulus (bright light) reduced them. Taken together, these findings suggest that high USVs may index an appetitive motivation to play in juvenile rats.