BNST PKCδ neurons are activated by specific aversive conditions to promote anxiety-like behavior

The bed nucleus of the stria terminalis (BNST) is a critical mediator of stress responses and anxiety-like behaviors. Neurons expressing protein kinase C delta (BNSTPKCδ) are an abundant but understudied subpopulation implicated in inhibiting feeding, but which have conflicting reports about their role in anxiety-like behaviors. We have previously shown that expression of PKCδ is dynamically regulated by stress and that BNSTPKCδ cells are recruited during bouts of active stress coping. Here, we first show that in vivo activation of this population is mildly aversive. This aversion was insensitive to prior restraint stress exposure. Further investigation revealed that unlike other BNST subpopulations, BNSTPKCδ cells do not exhibit increased cfos expression following restraint stress. Ex vivo current clamp recordings also indicate they are resistant to firing. To elucidate their afferent control, we next used rabies tracing with whole-brain imaging and channelrhodopsin-assisted circuit mapping, finding that BNSTPKCδ cells receive abundant input from affective, arousal, and sensory regions including the basolateral amygdala (BLA) paraventricular thalamus (PVT) and central amygdala PKCδ-expressing cells (CeAPKCδ). Given these findings, we used in vivo optogenetics and fiber photometry to further examine BNSTPKCδ cells in the context of stress and anxiety-like behavior. We found that BNSTPKCδ cell activity is associated with increased anxiety-like behavior in the elevated plus maze, increases following footshock, and unlike other BNST subpopulations, does not desensitize to repeated stress exposure. Taken together, we propose a model in which BNSTPKCδ cells may serve as threat detectors, integrating exteroceptive and interoceptive information to inform stress coping behaviors.

High fat diet blunts stress-induced hypophagia and activation of Glp1r dorsal lateral septum neurons in male but not in female mice

OBJECTIVE

While stress typically reduces caloric intake (hypophagia) in chow-fed rodents, presentation of palatable, high calorie substances during stress can increase caloric consumption (i.e. "comfort feeding") and promote obesity. However, little is known about how obesity itself affects feeding behavior in response to stress and the mechanisms that can influence stress-associated feeding in the context of obesity.

METHODS

We assessed food intake and other metabolic parameters in lean and obese male and female mice following acute restraint stress. We also measured real-time activity of glucagon-like peptide-1 (Glp1) receptor (Glp1r)-expressing neurons in the dorsal lateral septum (dLS) during stress in lean and obese mice using fiber photometry. Glp1r activation in various brain regions, including the dLS, promotes hypophagia in response to stress. Finally, we used inhibitory Designer Receptors Activated Exclusively by Designer Drugs (DREADDs) to test whether activation of Glp1r-expressing neurons in the LS is required for stress-induced hypophagia.

RESULTS

Lean male mice display the expected hypophagic response following acute restraint stress, but obese male mice are resistant to this acute stress-induced hypophagia. Glp1r-positive neurons in the dLS are robustly activated during acute restraint stress in lean but not in obese male mice. This raises the possibility that activation of dLS Glp1r neurons during restraint stress contributes to subsequent hypophagia. Supporting this, we show that chemogenetic inhibition of LS Glp1r neurons attenuates acute restraint stress hypophagia in male mice. Surprisingly, we show that both lean and obese female mice are resistant to acute restraint stress-induced hypophagia and activation of dLS Glp1r neurons.

CONCLUSIONS

These results suggest that dLS Glp1r neurons contribute to the hypophagic response to acute restraint stress in male mice, but not in female mice, and that obesity disrupts this response in male mice. Broadly, these findings show sexually dimorphic mechanisms and feeding behaviors in lean vs. obese mice in response to acute stress.

The impact of intermittent exercise on mouse ethanol drinking and abstinence-associated affective behavior and physiology

BACKGROUND

Negative emotional states are associated with the initiation and maintenance of alcohol use and drive relapse to drinking during withdrawal and protracted abstinence. Physical exercise is correlated with decreased negative affective symptoms, although a direct relationship between drinking patterns and exercise level has not been fully elucidated.

METHODS

We incorporated intermittent running wheel access into a chronic continuous access, two-bottle choice alcohol drinking model in female C57BL/6J mice. Wheel access was granted intermittently once mice established a preference for alcohol over water. After 6 weeks, alcohol was removed (forced abstinence) and mice were given continuous access to unlocked or locked wheels. Negative affect-like behavior, home cage behavior, and metabolic activity were measured during protracted abstinence.

RESULTS

Wheel access shifted drinking patterns in the mice, increasing drinking when the wheel was locked, and decreasing drinking when unlocked. Moreover, alcohol preference and consumption were strongly negatively correlated with the amount of running. An assessment of negative affect-like behavior in abstinence via the novelty suppressed feeding and saccharin preference tests (SPT) showed that unlimited wheel access mitigated abstinence-induced latency increases. Mice in abstinence also spent more time sleeping during the active dark cycle than control mice, providing additional evidence for abstinence-induced anhedonia- and depression-like behavior. Furthermore, running wheel access in abstinence decreased dark cycle sleep to comparable alcohol- and wheel-naïve mice. Given the positive impact of exercise and the negative impact of alcohol on metabolic health, we compared metabolic phenotypes of alcohol-abstinent mice with and without wheel access. Wheel access increased energy expenditure, carbon dioxide production, and oxygen consumption, providing a potential metabolic mechanism through which wheel access improves affective state.

CONCLUSIONS

This study suggests that including exercise in AUD treatment regimens has the potential to reduce drinking, improve affective state during abstinence and could serve as a non-pharmacological approach to prevent the development of an AUD in high-risk individuals.

Cocaine Augments Dopamine-Mediated Inhibition of Neuronal Activity in the Dorsal Bed Nucleus of the Stria Terminalis

The dorsal region of the bed nucleus of the stria terminalis (dBNST) receives substantial dopaminergic input which overlaps with norepinephrine input implicated in stress responses. Using ex vivo fast scan cyclic voltammetry in male C57BL6 mouse brain slices, we demonstrate that electrically stimulated dBNST catecholamine signals are of substantially lower magnitude and have slower uptake rates compared with caudate signals. Dopamine terminal autoreceptor activation inhibited roughly half of the catecholamine transient, and noradrenergic autoreceptor activation produced an ∼30% inhibition. Dopamine transporter blockade with either cocaine or GBR12909 significantly augmented catecholamine signal duration. We optogenetically targeted dopamine terminals in the dBNST of transgenic (TH:Cre) mice of either sex and, using ex vivo whole-cell electrophysiology, we demonstrate that optically stimulated dopamine release induces slow outward membrane currents and an associated hyperpolarization response in a subset of dBNST neurons. These cellular responses had a similar temporal profile to dopamine release, were significantly reduced by the D2/D3 receptor antagonist raclopride, and were potentiated by cocaine. Using in vivo fiber photometry in male C57BL/6 mice during training sessions for cocaine conditioned place preference, we show that acute cocaine administration results in a significant inhibition of calcium transient activity in dBNST neurons compared with saline administration. These data provide evidence for a mechanism of dopamine-mediated cellular inhibition in the dBNST and demonstrate that cocaine augments this inhibition while also decreasing net activity in the dBNST in a drug reinforcement paradigm.SIGNIFICANCE STATEMENT The dorsal bed nucleus of the stria terminalis (dBNST) is a region highly implicated in mediating stress responses; however, the dBNST also receives dopaminergic inputs from classically defined drug reward pathways. Here we used various techniques to demonstrate that dopamine signaling within the dBNST region has inhibitory effects on population activity. We show that cocaine, an abused psychostimulant, augments both catecholamine release and dopamine-mediated cellular inhibition in this region. We also demonstrate that cocaine administration reduces population activity in the dBNST, in vivo Together, these data support a mechanism of dopamine-mediated inhibition within the dBNST, providing a means by which drug-induced elevations in dopamine signaling may inhibit dBNST activity to promote drug reward.

Delineation of an insula-BNST circuit engaged by struggling behavior that regulates avoidance in mice

Active responses to stressors involve motor planning, execution, and feedback. Here we identify an insular cortex to BNST (insula→BNST) circuit recruited during restraint stress-induced active struggling that modulates affective behavior. We demonstrate that activity in this circuit tightly follows struggling behavioral events and that the size of the fluorescent sensor transient reports the duration of the struggle event, an effect that fades with repeated exposure to the homotypic stressor. Struggle events are associated with enhanced glutamatergic- and decreased GABAergic signaling in the insular cortex, indicating the involvement of a larger circuit. We delineate the afferent network for this pathway, identifying substantial input from motor- and premotor cortex, somatosensory cortex, and the amygdala. To begin to dissect these incoming signals, we examine the motor cortex input, and show that the cells projecting from motor regions to insular cortex are engaged shortly before struggle event onset. This study thus demonstrates a role for the insula→BNST pathway in monitoring struggling activity and regulating affective behavior.

Anxiety during abstinence from alcohol: A systematic review of rodent and human evidence for the anterior insula's role in the abstinence network

Alcohol Use Disorder (AUD) is a chronic, relapsing disease that impacts almost a third of Americans. Despite effective treatments for attaining sobriety, the majority of patients relapse within a year, making relapse a substantial barrier to long-term treatment success. A major factor contributing to relapse is heightened negative affect that results from the combination of abstinence-related increases in stress-reactivity and decreases in reward sensitivity. Substantial research has contributed to the understanding of reward-related changes in AUD. However, less is known about anxiety during abstinence, a critical component of understanding addiction as anxiety during abstinence can trigger relapse. Most of what we know about abstinence-related negative affect comes from rodent studies which have identified key brain regions responsible for abstinence-related behaviors. This abstinence network is composed of brain regions that make up the extended amygdala: the nucleus accumbens (NAcc), the central nucleus of the amygdala (CeA), and the bed nucleus of the stria terminalis (BNST). More recently, emerging evidence from rodent and human studies suggests a fourth brain region, the anterior insula, might be part of the abstinence network. Here, we review current rodent and human literature on the extended amygdala's role in alcohol abstinence and anxiety, present evidence for the anterior insula's role in the abstinence network, and provide future directions for research to further elucidate the neural underpinnings of abstinence in humans. A better understanding of the abstinence network is critical toward understanding and possibly preventing relapse in AUD.

Glutathione reductase deficiency alters lung development and hyperoxic responses in neonatal mice

Cellular antioxidants protect against hyperoxic lung injury. The role of the glutathione (GSH) system in lung development and bronchopulmonary dysplasia (BPD) pathogenesis has not been systematically investigated. The current study utilized GSH reductase-deficient (Gsr-KO) neonatal mice to test the hypothesis that early disruption of the GSH system negatively impacts lung development and hyperoxic responses. Lungs from wild-type (Gsr-WT) and Gsr-KO mice were analyzed for histopathology, developmental markers, redox indices, and transcriptome profiling at different developmental stages following exposure to room air or hyperoxia (85% O2) for up to 14 d. Lungs from Gsr-KO mice exhibited alveolar epithelial dysplasia in the embryonic and neonatal periods with relatively normal lung architecture in adulthood. GSH and its oxidized form (GSSG) were 50-70% lower at E19-PND14 in Gsr-KO lungs than in age-matched Gsr-WT. Differential gene expression between Gsr-WT and Gsr-KO lungs was analyzed at discrete developmental stages. Gsr-KO lungs exhibited downregulated cell cycle and DNA damage checkpoint genes at E19, as well as lung lipid metabolism and surfactant genes at PND5. In addition to abnormal baseline lung morphometry, Gsr-KO mice displayed a blunted response to hyperoxia. Hyperoxia caused a more robust upregulation of the lung thioredoxin system in Gsr-KO compared to Gsr-WT. Gsr-dependent, hyperoxia-responsive genes were highly associated with abnormal cytoskeleton, skeletal-muscular function, and tissue morphology at PND5. Overall, our data in Gsr-KO mice implicate the GSH system as a key regulator of lung development, cellular differentiation, and hyperoxic responses in neonatal mice.

Integrated human pseudoislet system and microfluidic platform demonstrate differences in GPCR signaling in islet cells

Pancreatic islets secrete insulin from β cells and glucagon from α cells, and dysregulated secretion of these hormones is a central component of diabetes. Thus, an improved understanding of the pathways governing coordinated β and α cell hormone secretion will provide insight into islet dysfunction in diabetes. However, the 3D multicellular islet architecture, essential for coordinated islet function, presents experimental challenges for mechanistic studies of intracellular signaling pathways in primary islet cells. Here, we developed an integrated approach to study the function of primary human islet cells using genetically modified pseudoislets that resemble native islets across multiple parameters. Further, we developed a microperifusion system that allowed synchronous acquisition of GCaMP6f biosensor signal and hormone secretory profiles. We demonstrate the utility of this experimental approach by studying the effects of Gi and Gq GPCR pathways on insulin and glucagon secretion by expressing the designer receptors exclusively activated by designer drugs (DREADDs) hM4Di or hM3Dq. Activation of Gi signaling reduced insulin and glucagon secretion, while activation of Gq signaling stimulated glucagon secretion but had both stimulatory and inhibitory effects on insulin secretion, which occur through changes in intracellular Ca2+. The experimental approach of combining pseudoislets with a microfluidic system allowed the coregistration of intracellular signaling dynamics and hormone secretion and demonstrated differences in GPCR signaling pathways between human β and α cells.

Endocannabinoid control of the insular-bed nucleus of the stria terminalis circuit regulates negative affective behavior associated with alcohol abstinence

Negative affect is a core symptom domain associated with an array of neurological and psychiatric disorders and is only partially targeted by current therapies, highlighting the need for better, more targeted treatment options. This study focuses on negative affective symptoms associated with prolonged alcohol abstinence, one of the leading causes of relapse. Using a mouse model of chronic alcohol consumption followed by forced abstinence (CDFA), prolonged alcohol abstinence increased c-fos expression and spontaneous glutamatergic neurotransmission in the dorsal bed nucleus of the stria terminalis (dBNST), a region heavily implicated in negative affect in both humans and rodents. Further, pharmacologically enhancing endogenous cannabinoids (eCB) with JZL184 prevents abstinence-induced increases in dBNST neuronal activity, underscoring the therapeutic potential of drugs targeting the brain's eCB system. Next, we used a channelrhodopsin-assisted mapping strategy to identify excitatory inputs to the dBNST that could contribute to CDFA-induced negative affect. We identified the insular cortex (insula), a region involved in regulating interoception, as a dense, functional, eCB-sensitive input to the dBNST. Using a chemogenetic strategy to locally mimic eCB signaling, we demonstrate that the insula strongly influences the CDFA behavioral phenotype and dBNST neuronal activity. Lastly, we used an anterograde strategy for transynaptic targeting of Cre expression in combination with a Gq-DREADD to selectively recruit dBNST neurons receiving insula projections. Chemogenetic recruitment of these neurons mimicked behavioral and c-fos responses observed in CDFA. Collectively, this study supports a role for the insula-BNST neural circuit in negative affective disturbances and highlights the therapeutic potential of the eCB system for treating negative affective disorders.

Stimulation of the dorsal periaqueductal gray enhances spontaneous recovery of a conditioned taste aversion

Due to its relevance to clinical practice, extinction of learned fears has been a major focus of recent research. However, less is known about the means by which conditioned fears re-emerge (i.e., spontaneously recover) as time passes or contexts change following extinction. The periaqueductal gray represents the final common pathway mediating defensive reactions to fear and we have reported previously that the dorsolateral PAG (dlPAG) exhibits a small but reliable increase in neural activity (as measured by c-fos protein immunoreactivity) when spontaneous recovery (SR) of a conditioned taste aversion (CTA) is reduced. Here we extend these correlational studies to determine if inducing dlPAG c-fos expression through electrical brain stimulation could cause a reduction in SR of a CTA. Male Sprague-Dawley rats acquired a strong aversion to saccharin (conditioned stimulus; CS) and then underwent CTA extinction through multiple non-reinforced exposures to the CS. Following a 30-day latency period after asymptotic extinction was achieved; rats either received stimulation of the dorsal PAG (dPAG) or stimulation of closely adjacent structures. Sixty minutes following the stimulation, rats were again presented with the saccharin solution as we tested for SR of the CTA. The brain stimulation evoked c-fos expression around the tip of the electrodes. However, stimulation of the dPAG failed to reduce SR of the previously extinguished CTA. In fact, dPAG stimulation caused rats to significantly suppress their saccharin drinking (relative to controls) - indicating an enhanced SR. These data refute a cause-and-effect relationship between enhanced dPAG c-fos expression and a reduction in SR. However, they highlight a role for the dPAG in modulating SR of extinguished CTAs.

Periaqueductal gray c-Fos expression varies relative to the method of conditioned taste aversion extinction employed

A conditioned taste aversion (CTA) is acquired when an animal consumes a novel taste (CS) and then experiences the symptoms of poisoning (US). Following CTA training, animals will avoid the taste that was previously associated with malaise. This defensive reaction to a learned fear can be extinguished by repeated exposure to the CS alone (CS-only; CSO-EXT). However, following a latency period in which the CS is not presented, the CTA will spontaneously recover (SR). Through the use of an explicitly unpaired extinction procedure (EU-EXT) we have shown that we can speed up extinction and attenuate SR of the CTA. Here we compared and contrasted the ability of CSO and EU extinction procedures to affect c-Fos expression in the periaqueductal gray (PAG). Fluid-deprived Sprague-Dawley rats acquired a strong CTA [via 3 pairings of 0.3% oral saccharin (SAC; the CS) and 81mg/kg i.p. lithium chloride (LiCl; the US)] followed by extinction trials consisting of multiple exposures to either, (a) the CS every-other day (CSO-EXT), or (b) CS and US on alternate days (EU-EXT). A different group of rats did not receive multiple CS exposures and served as a "no extinction" (NE) control. Both extinction procedures resulted in ≥90% reacceptance of SAC (achieving asymptotic extinction). Some of the animals were sacrificed for c-Fos immunohistochemical analysis following asymptotic extinction. Other rats entered a 30-day latency period where they drank water only. These remaining animals were then tested for SR with a final exposure to SAC before being sacrificed for c-Fos immunohistochemistry. As reported previously, rats in the CS-only group exhibited a significant SR of the CTA. However, animals in the EU extinction group reached asymptotic extinction more rapidly than did CSO rats and they did not show SR of the CTA. As compared to rats that retained their CTA, both groups of extinguished rats showed suppression in the number of c-Fos-labeled neurons in all 4 longitudinal columns of the PAG. The number of c-Fos-labeled cells in the PAG was generally low but there was a reliable increase in c-Fos expression in dorsolateral PAG (dlPAG) following the SR test in the brains of rats that went through the EU-EXT procedure as compared with those that either went through the more-traditional CSO extinction procedure or experienced no extinction at all. The number of c-Fos-labeled neurons in the dlPAG was significantly correlated with the amount of SAC consumed at the SR test. Surprisingly, the brains of EU-extinguished rats and CSO extinguished rats did not differ in the number of c-Fos-labeled neurons in gustatory neocortex, medial prefrontal cortex, basolateral amygdala, or the central nucleus of the amygdala. Thus, behavioral differences in SR between the EU and CSO extinction animals were not represented by corresponding changes in the neural activity of several brain nuclei classically associated with extinction learning. However a detailed analysis of PAG c-Fos expression provided hints about some of the physiological changes evoked by these 2 extinction paradigms that produce very different behavioral outcomes. The findings are clinically relevant as we seek the development of treatments for deficits in fear extinction (e.g. PTSD, phobias).