Posterior Basolateral Amygdala Research Articles

Synergism between two BLA-to-BNST pathways for appropriate expression of anxiety-like behaviors in male mice

Understanding how distinct functional circuits are coordinated to fine-tune mood and behavior is of fundamental importance. Here, we observe that within the dense projections from basolateral amygdala (BLA) to bed nucleus of stria terminalis (BNST), there are two functionally opposing pathways orchestrated to enable contextually appropriate expression of anxiety-like behaviors in male mice. Specifically, the anterior BLA neurons predominantly innervate the anterodorsal BNST (adBNST), while their posterior counterparts send massive fibers to oval BNST (ovBNST) with moderate to adBNST. Optogenetic activation of the anterior and posterior BLA inputs oppositely regulated the activity of adBNST neurons and anxiety-like behaviors, via disengaging and engaging the inhibitory ovBNST-to-adBNST microcircuit, respectively. Importantly, the two pathways exhibited synchronized but opposite responses to both anxiolytic and anxiogenic stimuli, partially due to their mutual inhibition within BLA and the different inputs they receive. These findings reveal synergistic interactions between two BLA-to-BNST pathways for appropriate anxiety expression with ongoing environmental demands.

Anterior and Posterior Basolateral Amygdala Projections of Cell Type–Specific D1-Expressing Neurons From the Medial Prefrontal Cortex Differentially Control Alcohol-Seeking Behavior

BackgroundAlcohol use disorder is characterized by compulsive alcohol-seeking behavior, which is associated with dysregulation of afferent projections from the medial prefrontal cortex to the basolateral amygdala (BLA). However, the contribution of the cell type–specific mechanism in this neuronal circuit to alcohol-seeking behavior remains unclear. MethodsMice were trained with 2-bottle choice and operant alcohol self-administration procedures. Anterograde and retrograde viral methods traced the connection between dopamine type 1 receptor (D1R) neurons and BLA neurons. Electrophysiology and in vivo optogenetic techniques were used to test the function of neural circuits in alcohol-seeking behavior. ResultsChronic alcohol consumption preferentially changed the activity of posterior BLA (pBLA) neurons but not anterior BLA (aBLA) neurons and overexcited D1R neurons in the medial prefrontal cortex. Interestingly, we found that 2 populations of D1R neurons, anterior and posterior (pD1R) neurons, separately targeted the aBLA and pBLA, respectively, and only a few D1R neurons innervated both aBLA and pBLA neurons. Furthermore, pD1R neurons exhibited more excitability than anterior D1R neurons in alcohol-drinking mice. Moreover, we observed enhanced glutamatergic transmission and an increased NMDA/AMPA receptor ratio in the medial prefrontal cortex inputs from pD1R neurons to the pBLA. Optogenetic long-term depression induction of the pD1R-pBLA circuit reduced alcohol-seeking behavior, while optogenetic long-term depression or long-term potentiation induction of the anterior D1R–aBLA circuit produced no change in alcohol intake. ConclusionsThe pD1R-pBLA circuit mediates chronic alcohol consumption, which may suggest a cell type–specific neuronal mechanism underlying reward-seeking behavior in alcohol use disorder.

High emotional reactivity is associated with activation of a molecularly distinct hippocampal-amygdala circuit modulated by the glucocorticoid receptor

Emotions are characterized not only by their valence but also by whether they are stable or labile. Yet, we do not understand the molecular or circuit mechanisms that control the dynamic nature of emotional responses. We have shown that glucocorticoid receptor overexpression in the forebrain (GRov) leads to a highly reactive mouse with increased anxiety behavior coupled with greater swings in emotional responses. This phenotype is established early in development and persists into adulthood. However, the neural circuitry mediating this lifelong emotional lability remains unknown. In the present study, optogenetic stimulation in ventral dentate gyrus (vDG) of GRov mice led to a greater range and a prolonged duration of anxiety behavior. cFos expression analysis showed that the amplified behavioral response to vDG activation in GRov mice is coupled to increased neuronal activity in specific brain regions. Relative to wild type mice, GRov mice displayed glutamatergic/GABAergic activation imbalance in ventral CA1 (vCA1) and selectively increased glutamatergic activation in the basal posterior amygdaloid complex. Moreover, forebrain GR overexpression led to increased activation of molecularly distinct subpopulations of neurons within the hippocampus and the posterior basolateral amygdala (pBLA) as evident from the increased cFos co-labeling in the calbindin1+ glutamatergic neurons in vCA1 and in the DARPP-32/Ppp1r1b+ glutamatergic neurons in pBLA. We propose that a molecularly distinct hippocampal-amygdala circuit is shaped by stress early in life and tunes the dynamics of emotional responses.

Infralimbic medial prefrontal cortex signalling to calbindin 1 positive neurons in posterior basolateral amygdala suppresses anxiety- and depression-like behaviours

Generalization is a fundamental cognitive ability of organisms to deal with the uncertainty in real-world situations. Excessive fear generalization and impaired reward generalization are closely related to many psychiatric disorders. However, the neural circuit mechanism for reward generalization and its role in anxiety-like behaviours remain elusive. Here, we found a robust activation of calbindin 1-neurons (Calb 1) in the posterior basolateral amygdala (pBLA), simultaneous with reward generalization to an ambiguous cue after reward conditioning in mice. We identify the infralimbic medial prefrontal cortex (IL) to the pBLACalb1 (Calb 1 neurons in the pBLA) pathway as being involved in reward generalization for the ambiguity. Activating IL–pBLA inputs strengthens reward generalization and reduces chronic unpredictable mild stress-induced anxiety- and depression-like behaviours in a manner dependent on pBLACalb1 neuron activation. These findings suggest that the IL–pBLACalb1 circuit could be a target to promote stress resilience via reward generalization and consequently ameliorate anxiety- and depression-like behaviours.

Oxytocin Receptor-Expressing Neurons in the Paraventricular Thalamus Regulate Feeding Motivation through Excitatory Projections to the Nucleus Accumbens Core.

Oxytocin receptors (OTR) have been found in the paraventricular thalamus (PVT) for the regulation of feeding and maternal behaviors. However, the functional projections of OTR-expressing PVT neurons remain largely unknown. Here, we used chemogenetic and optogenetic tools to test the role of OTR-expressing PVT neurons and their projections in the regulation of food intake in both male and female OTR-Cre mice. We found chemogenetic activation of OTR-expressing PVT neurons promoted food seeking under trials with a progressive ratio schedule of reinforcement. Using Feeding Experimentation Devices for real-time meal measurements, we found chemogenetic activation of OTR-expressing PVT neurons increased meal frequency but not cumulative food intake because of a compensatory decrease in meal sizes. In combination with anterograde neural tracing and slice patch-clamp recordings, we found optogenetic stimulation of PVT OTR terminals excited neurons in the posterior basolateral amygdala (pBLA) and nucleus accumbens core (NAcC) as well as local PVT neurons through monosynaptic glutamatergic transmissions. Photostimulation of OTR-expressing PVT-NAcC projections promoted food-seeking, whereas selective activation of PVT-pBLA projections produced little effect on feeding. In contrast to selective activation of OTR terminals, photostimulation of a broader population of glutamatergic PVT terminals exerted direct excitation followed by indirect lateral inhibition on neurons in both NAcC and anterior basolateral amygdala. Together, these results suggest that OTR-expressing PVT neurons are a distinct population of PVT glutamate neurons that regulate feeding motivation through projections to NAcC.SIGNIFICANCE STATEMENT The paraventricular thalamus plays an important role in the regulation of feeding motivation. However, because of the diversity of paraventricular thalamic neurons, the specific neuron types promoting food motivation remain elusive. In this study, we provide evidence that oxytocin receptor-expressing neurons are a specific group of glutamate neurons that primarily project to the nucleus accumbens core and posterior amygdala. We found that activation of these neurons promotes the motivation for food reward and increases meal frequency through projections to the nucleus accumbens core but not the posterior amygdala. As a result, we postulate that oxytocin receptor-expressing neurons in the paraventricular thalamus and their projections to the nucleus accumbens core mainly regulate feeding motivation but not food consumption.

Posterior basolateral amygdala to ventral hippocampal CA1 drives approach behaviour to exert an anxiolytic effect

The basolateral amygdala (BLA) and ventral hippocampal CA1 (vCA1) are cellularly and functionally diverse along their anterior–posterior and superficial-deep axes. Here, we find that anterior BLA (aBLA) and posterior BLA (pBLA) innervate deep-layer calbindin1-negative (Calb1−) and superficial-layer calbindin1-positive neurons (Calb1+) in vCA1, respectively. Photostimulation of pBLA–vCA1 inputs has an anxiolytic effect in mice, promoting approach behaviours during conflict exploratory tasks. By contrast, stimulating aBLA–vCA1 inputs induces anxiety-like behaviour resulting in fewer approaches. During conflict stages of the elevated plus maze task vCA1Calb1+ neurons are preferentially activated at the open-to-closed arm transition, and photostimulation of vCA1Calb1+ neurons at decision-making zones promotes approach with fewer retreats. In the APP/PS1 mouse model of Alzheimer’s disease, which shows anxiety-like behaviour, photostimulating the pBLA–vCA1Calb1+ circuit ameliorates the anxiety in a Calb1-dependent manner. These findings suggest the pBLA–vCA1Calb1+ circuit from heterogeneous BLA–vCA1 connections drives approach behaviour to reduce anxiety-like behaviour.

Context and topography determine the role of basolateral amygdala metabotropic glutamate receptor 5 in appetitive Pavlovian responding.

Preclinical data have shown that the excitatory metabotropic Gαq-coupled glutamate receptor, mGluR5, has a role in substance abuse and relapse. However, little is known about the contribution of mGluR5 to the expression of conditioned responding elicited by appetitive Pavlovian cues. We investigated this question in rats that were trained to associate a discrete, auditory conditioned stimulus (CS) with a fructose-glucose solution (5.5% fructose/4.5% glucose; "sugar"). In subsequent tests for the expression of conditioned responding without sugar delivery, CS-elicited fluid port entries were elevated in a context associated with sugar, relative to an equally familiar, neutral context. Inhibiting mGluR5 via systemic injections of a negative allosteric modulator (MTEP; 5 mg/kg) reduced CS port entries in both the sugar context and neutral context. Targeting MTEP microinjections (3 µg/side; 0.3 µl/min) to the nucleus accumbens (Acb) core had no effect on CS port entries at test, whereas the same manipulation in the basolateral amygdala (BLA) produced effects that were topographically dependent. Specifically, microinjecting MTEP in the posterior BLA had no effect on behavior, whereas inhibiting mGluR5 in the anterior BLA enhanced the contextual discrimination of CS port entries. These data are the first to show a role of mGluR5 in the context-dependent expression of appetitive Pavlovian conditioned responding, with a topographically defined arrangement of mGluR5 in the BLA being particularly important for context-based responding to a discrete, appetitive cue.

Relative contributions and mapping of ventral tegmental area dopamine and GABA neurons by projection target in the rat.

The ventral tegmental area (VTA) is a heterogeneous midbrain structure that contains dopamine (DA), GABA, and glutamate neurons that project to many different brain regions. Here, we combined retrograde tracing with immunocytochemistry against tyrosine hydroxylase (TH) or glutamate decarboxylase (GAD) to systematically compare the proportion of dopaminergic and GABAergic VTA projections to 10 target nuclei: anterior cingulate, prelimbic, and infralimbic cortex; nucleus accumbens core, medial shell, and lateral shell; anterior and posterior basolateral amygdala; ventral pallidum; and periaqueductal gray. Overall, the non-dopaminergic component predominated VTA efferents, accounting for more than 50% of all projecting neurons to each region except the nucleus accumbens core. In addition, GABA neurons contributed no more than 20% to each projection, with the exception of the projection to the ventrolateral periaqueductal gray, where the GABAergic contribution approached 50%. Therefore, there is likely a significant glutamatergic component to many of the VTA's projections. We also found that VTA cell bodies retrogradely labeled from the various target brain regions had distinct distribution patterns within the VTA, including in the locations of DA and GABA neurons. Despite this patterned organization, VTA neurons comprising these different projections were intermingled and never limited to any one subregion. These anatomical results are consistent with the idea that VTA neurons participate in multiple distinct, parallel circuits that differentially contribute to motivation and reward. While attention has largely focused on VTA DA neurons, a better understanding of VTA subpopulations, especially the contribution of non-DA neurons to projections, will be critical for future work.

Reduced resting-state functional connectivity of the basolateral amygdala to the medial prefrontal cortex in preweaning rats exposed to chronic early-life stress.

Early-life stress (ELS) exposure has long-term consequences for both brain structure and function and impacts cognitive and emotional behavior. The basolateral amygdala (BLA) plays an important role in anxiety and fear conditioning through its extensive anatomical and functional connections, in particular to the medial prefrontal cortex (mPFC). However, how ELS affects amygdala function and connectivity in developing rats is unknown. We used the naturalistic limited bedding/nesting (LB) paradigm to induce chronic stress in the pups between postnatal day (PND) 1-10. Male normal bedding (NB, control) or LB offspring underwent structural and resting-state functional MRI (rs-fMRI) on PND18 and in adulthood (PND74-76). Adult male rats were tested for fear conditioning and extinction behavior prior to scanning. Seed-based functional connectivity maps were generated based on four BLA seeds (left, right, anterior and posterior). At both ages, LB induced different effects on anterior and posterior BLA networks, with significant reductions in rs-fMRI connectivity between the anterior BLA and mPFC in LB compared to NB offspring. BLA connectivity was lateralized by preweaning age, with the right hemisphere displaying more connectivity changes than the left. Weak negative volumetric correlations between the BLA and mPFC were also present, mostly in preweaning LB animals. rs-fMRI connectivity and volumetric changes were associated with enhanced fear behaviors in adult LB offspring. Activation of the LB-exposed neonatal amygdala described previously might accelerate the maturation of BLA-mPFC projections and/or modify the activity of reciprocal connections between these structures, leading to a net reduction in rs-fMRI connectivity and increased fear behavior.

Role of amygdala–prefrontal cortex circuitry in regulating the expression of contextual fear memory

The basolateral amygdala (BLA) and medial prefrontal cortex (mPFC) are inter-connected regions involved in fear memory expression. The reciprocal nature of projections between these areas differs along the rostrocaudal extent of BLA. This study investigated the role of functional interactions between BLA and the prelimbic (PL) subregion of mPFC in mediating contextual fear memory. Freezing served as the measure of conditioned fear. Experiments 1–3 examined the effects of left, right or bilateral infusion of bupivacaine into anterior BLA (aBLA), posterior BLA (pBLA) or PL on fear memory expression. Reversible inactivation of left, right or bilateral aBLA impaired fear memory expression. Bilateral inactivation of pBLA or PL also disrupted the expression of fear memory, although left or right inactivation alone had no significant effects in either region. Experiment 4 examined the effects of functionally disconnecting pBLA and PL on contextual fear memory by infusing bupivacaine unilaterally into pBLA and PL in the ipsilateral or contralateral hemisphere. Fear memory expression was impaired by asymmetric inactivation of pBLA and PL; however, a similar effect was also observed with symmetric inactivation of these regions. Bupivacaine infusion did not affect behavior in the open field, likely ruling out non-specific effects of inactivation on innate fear and locomotor activity. These results demonstrate different roles for rostral and caudal BLA in mediating the expression of contextual fear memory. They also raise the possibility that pBLA–PL circuitry is involved in subserving fear memory expression via complex processing mechanisms, although further research is needed to confirm this preliminary finding.

Brain Region-Specific Mechanisms for Acute Morphine-Induced Mitogen-Activated Protein Kinase Modulation and Distinct Patterns of Activation during Analgesic Tolerance and Locomotor Sensitization

Opioid-receptor activation in cell lines results in phosphorylation of p42/44 mitogen-activated protein kinase (MAPK), which contributes to agonist-induced desensitization of adenylate cyclase signaling. In this study, morphine-induced MAPK modulation was examined in the mouse brain using antibodies against phosphorylated MAPK. Thirty minutes after systemic morphine, MAPK modulation was observed in brain areas associated with analgesia and reward. Activation of MAPK was increased in the anterior cingulate (Acc), somato-sensory and association cortices, and locus ceruleus (LC). In contrast, MAPK activation was decreased in the nucleus accumbens and central amygdala (CeA). Double-label confocal microscopy revealed that morphine-induced MAPK modulation occurred predominantly in cells not expressing mu-opioid receptors, with the exception of the LC. Furthermore, the NMDA receptor antagonist 3,3-(2-carboxypiperazine-4-yl)-propyl-1-phosphonate blocked morphine-induced MAPK modulation in several cortical areas including the Acc. We then examined morphine-induced MAPK modulation during expression of either analgesic tolerance or locomotor sensitization, which were differentiated by two repeated morphine regimens. Analgesic tolerance was accompanied by tolerance to morphine-induced MAPK modulation in all of the brain areas examined except the CeA. Locomotor sensitization resulted in sensitization to morphine-induced MAPK activation in the posterior basolateral amygdala. Additionally, a pronounced instatement of morphine-induced MAPK activation was observed in CA3 hippocampal processes. This instatement was observed during expression of tolerance; however, it was not significant during sensitization. In summary, these results provide distinct, region-specific mechanisms for morphine-induced MAPK modulation in the mouse brain and give insight into the brain circuitry involved in acute and adaptive opioid behaviors.

Estrous cycle and food availability affect feeding induced by amygdala 5-HT receptor blockade

We have recently reported that bilateral infusions of the 5-HT receptor antagonist metergoline (MET) into the posterior basolateral amygdala (pBLA) elicit feeding in female rats tested at mid-light cycle. The present study was performed to determine whether (1) testing at two different phases of the estrous cycle, and/or (2) the palatability of the food might modify this effect. Subjects were 18 adult females with bilateral pBLA cannulae. Following familiarization with Froot Loops cereal, a within-subjects design tested all animals for 1- and 2-h food intake under 2 Drug (0.3 nmol MET vs. Vehicle), 2 Estrous Cycle (diestrus vs. estrus) and 2 Food (lab chow vs. Froot Loops) conditions. Rats weighed more at diestrus than at proestrus ( P<.05) or estrus ( P<.005). Multivariate analyses of variance (MANOVAs) revealed a preference for Froot Loops over lab chow ( P<.0001). MET increased feeding regardless of food type ( P<.0001). Rats ate more Froot Loops ( P<.01), but not lab chow, at diestrus vs. estrus. A three-way interaction ( P<.05) showed rats ate more during the first hour in estrus than in diestrus to lab chow but not Froot Loops. These data suggest pBLA MET differentially affects feeding over the estrous cycle depending on the palatability of food available.

Lesions of the posterior basolateral amygdala block feeding induced by systemic 8-OH-DPAT

We have recently reported that bilateral electrolytic lesions of the posterodorsal amygdala (PDA) in female rats which induce protracted overeating and weight gain also attenuate short-term feeding stimulated by intraraphe infusions of the serotonin (5-HT) 1A agonist, (±)-8-hydroxy-2-(di- n-propylamino)tetralin, (8-OH-DPAT). Bilateral lesions of the posterior basolateral amygdala (pBLA) in male rats have also been reported to enhance feeding and weight gain, but much less so than PDA lesions do in female rats. The present study was performed to determine if pBLA lesions in female rats might attenuate 8-OH-DPAT feeding and what, if any, relationship exists between 8-OH-DPAT-induced feeding and lesion-induced weight gain. Lesioned rats showed reliable increases in 24-h food intake and weight gain relative to shams during the days between surgery and acute drug-induced feeding tests. 8-OH-DPAT (0, 60, 120 or 240 μg/kg in saline) increased feeding of shams in a dose-dependent manner over 2 h. Feeding at the most effective dose (120 μg/kg) was reduced to vehicle levels in lesioned rats. The feeding induced by this dose correlated inversely ( r=−.59, P<.01) with the magnitude of weight gained following lesions. Feeding at the highest dose (240 μg/kg) showed a biphasic effect of feeding inhibition over the first vs. second hour that was unaffected by lesions. These findings imply that either fibers of passage and/or cellular elements in both the PDA and pBLA normally inhibit overeating and weight gain via intact serotonergic mechanisms.

5-HT receptor blockade in the posterior amygdala elicits feeding in female rats.

Previous work suggests that feeding following intraventricular (i.v.t.) injections of the serotonin (5-HT)(1/2/7) antagonist metergoline (MET) is not localized to the hypothalamus. Since lesions of the posterior basolateral amygdala (pBLA) block feeding following systemic 5-HT1A agonist 8-hydroxy-2(di-n-propylamino)tetralin, the ability of intra-pBLA MET to elicit feeding was investigated. In two separate experiments, feeding of female rats was measured over 2 h following 0, 3, 10 and 30 nmol and 0, 0.03, 0.3 and 3 nmol MET (mol. wt. 403.5) injected bilaterally into each pBLA. All three doses used in Experiment 1 increased feeding over 2 h. In Experiment 2, feeding over the first hour was enhanced after the two highest doses. Since intra-pBLA MET elicits feeding comparable to that seen using much higher doses administered i.v.t. these data implicate the pBLA as an extra-hypothalamic site mediating the effects of 5-HT in feeding control.

Anterograde tracing of trigeminal afferent pathways from the murine tooth pulp to cortex using herpes simplex virus type 1

Due to its predominantly nociceptive innervation, viral tracing from the tooth pulp provides a potential means for tracing central pain pathways. The neural pathways from the tooth pulp to cortex were determined using in situ hybridization to detect the anterograde transneuronal spread of herpes simplex virus type 1 strain H129 following inoculation into the murine mandibular incisor pulp. Virus first appeared in the brain at day 3 in the dorsomedial region of all three subnuclei of the spinal trigeminal nucleus and the principal sensory nucleus. By days 5-6 virus had spread to the contralateral medial nucleus of the medial geniculate complex, posterior thalamus, and ventroposteromedial thalamus. At days 7-8 virus was detected in laminae IV and Va of the primary somatosensory cortex and lamina IV of the secondary somatosensory cortex in regions previously shown to receive input from the lower jaw. Several mice also showed infection of laminae II/III of the ipsilateral dysgranular insular cortex, along with labeling for virus in the ipsilateral external lateral parabrachial nucleus, posterior thalamus, and posterior basolateral amygdala. Our results are highly consistent with previous tracing and electrophysiological studies utilizing the tooth pulp and with studies implicating the infected structures in nociception. Viral spread appeared to define two separate afferent systems with infection of structures which have been implicated in the sensory-discriminative aspects of pain, such as the ventroposteromedial thalamus and somatosensory cortex, as well as in the dysgranular insular cortex and related subcortical nuclei which may have a role in the affective-motivational aspects of pain.

The basolateral amygdala, dopamine and gastric stress ulcer formation in rats

Bilateral microinjections of dopamine (DA, 0.3, 3.0 or 30.0 μg) or the DAagonist, bromocryptine (3.0 μg) into the basolateral amygdala (BLA) dose-dependently attenuated cold restraint stress (3 h at 4 °C) -induced gastric ulcer formation in rats. On the other hand, intra-BLA injections of the neurotoxin, 6-hydroxydopamine (10 μg) or the DA-antagonist, haloperidol (0.1 or 1.0 μg) aggravated such stress ulcer formation. All these effects were seen only when the injection sites were localized in the posterior (and not the anterior) BLA. Further, pretreatment of rats with haloperidol (0.1 μg) clearly antagonized the gastric cytoprotective effects of DA or bromocryptine (both at 3.0 μg), when both chemicals were injected in the posterior BLA. The results indicate that DA-ergic mechanisms in the posterior BLA are important for the regulation of gastric mucosal integrity during cold restraint stress.