Muscarinic neurotransmission in the anterior basolateral amygdalar nucleus (BLa) mediated by the M1 receptor (M1R) is critical for memory consolidation. Although knowledge of the subcellular localization of M1R in the BLa would contribute to an understanding of cholinergic mechanisms involved in mnemonic function, there have been no ultrastructural studies of this receptor in the BLa. In the present investigation, immunocytochemistry at the electron microscopic level was used to determine which structures in the BLa express M1R. The innervation of these structures by cholinergic axons expressing the vesicular acetylcholine transporter (VAChT) was also studied. All perikarya of pyramidal neurons were labeled, and about 90% of dendritic shafts and 60% of dendritic spines were M1R+. Some dendrites had spines suggesting that they belonged to pyramidal cells, whereas others had morphological features typical of interneurons. M1R immunoreactivity (M1R-ir) was also seen in axon terminals, most of which formed asymmetrical synapses. The main targets of M1R+ terminals forming asymmetrical synapses were dendritic spines, most of which were M1R+. The main targets of M1R+ terminals forming symmetrical synapses were M1R+ perikarya and dendritic shafts. About three-quarters of VAChT+ cholinergic terminals formed synapses; the main postsynaptic targets were M1R+ dendritic shafts and spines. In some cases M1R-ir was seen near the postsynaptic membrane of these processes, but in other cases it was found outside of the active zone of VAChT+ synapses. These findings suggest that M1R mechanisms in the BLa are complex, involving postsynaptic effects as well as regulating release of neurotransmitters from presynaptic terminals.

Modulatory interactions of opioids and norepinephrine (NE) in the anterior subdivision of the basolateral nucleus of the amygdala (BLa) are critical for the consolidation of memories of emotionally arousing experiences. Although there have been several studies of the noradrenergic system in the amygdalar basolateral nuclear complex (BLC), little is known about the chemical neuroanatomy of opioid systems in this region. To address this knowledge gap the present study first examined the distribution of met-enkephalin-like immunoreactivity (ENK-ir) in the BLC at the light microscopic level, and then utilized dual-labeling immunocytochemistry combined with electron microscopy to investigate the extent of convergence of NE and ENK terminals onto common structures in the BLa. Antibodies to ENK and the norepinephrine transporter (NET) were used in these studies. Light microscopic examination revealed that a subpopulation of small nonpyramidal neurons expressed ENK-ir in all nuclei of the BLC. In addition, the somata of some pyramidal cells exhibited light to moderate ENK-ir. ENK+ axon terminals were also observed. Ultrastructural analysis confined to the BLa revealed that most ENK+ axon terminals formed asymmetrical synapses that mainly contacted spines and shafts of thin dendrites. ENK+ terminals forming symmetrical synapses mainly contacted dendritic shafts. Approximately 20% of NET+ terminals contacted a structure that was also contacted by an ENK+ terminal and 6% of NET+ terminals contacted an ENK+ terminal. These findings suggest that ENK and NE terminals in the BLa may interact by targeting common dendrites and by direct interactions between the two types of terminals.

The basolateral nuclear complex of the amygdala (BLC) receives a dense serotonergic innervation that appears to play a critical role in the regulation of mood and anxiety. However, little is known about how serotonergic inputs interface with different neuronal subpopulations in this region. To address this question, dual-labeling immunohistochemical techniques were used at the light and electron microscopic levels to examine inputs from serotonin-immunoreactive (5-HT+) terminals to different neuronal subpopulations in the rat BLC. Pyramidal cells were labeled by using antibodies to calcium/calmodulin-dependent protein kinase II, whereas different interneuronal subpopulations were labeled by using antibodies to a variety of interneuronal markers including parvalbumin (PV), vasoactive intestinal peptide (VIP), calretinin, calbindin, cholecystokinin, and somatostatin. The BLC exhibited a dense innervation by thin 5-HT+ axons. Electron microscopic examination of the anterior basolateral nucleus (BLa) revealed that 5-HT+ axon terminals contained clusters of small synaptic vesicles and a smaller number of larger dense-core vesicles. Serial section reconstruction of 5-HT+ terminals demonstrated that 76% of these terminals formed synaptic junctions. The great majority of these synapses were symmetrical. The main targets of 5-HT+ terminals were spines and distal dendrites of pyramidal cells. However, in light microscopic preparations it was common to observe apparent contacts between 5-HT+ terminals and all subpopulations of BLC interneurons. Electron microscopic analysis of the BLa in sections dual-labeled for 5-HT/PV and 5-HT/VIP revealed that many of these contacts were synapses. These findings suggest that serotonergic axon terminals differentially innervate several neuronal subpopulations in the BLC.

Activation of M2 muscarinic receptors (M2Rs) in the rat anterior basolateral nucleus (BLa) is critical for the consolidation of memories of emotionally arousing events. The present investigation used immunocytochemistry at the electron microscopic level to determine which structures in the BLa express M2Rs. In addition, dual localization of M2R and the vesicular acetylcholine transporter protein (VAChT), a marker for cholinergic axons, was performed to determine whether M2R is an autoreceptor in cholinergic axons innervating the BLa. M2R immunoreactivity (M2R-ir) was absent from the perikarya of pyramidal neurons, with the exception of the Golgi complex, but was dense in the proximal dendrites and axon initial segments emanating from these neurons. Most perikarya of nonpyramidal neurons were also M2R-negative. About 95% of dendritic shafts and 60% of dendritic spines were M2 immunoreactive (M2R(+) ). Some M2R(+) dendrites had spines, suggesting that they belonged to pyramidal cells, whereas others had morphological features typical of nonpyramidal neurons. M2R-ir was also seen in axon terminals, most of which formed asymmetrical synapses. The main targets of M2R(+) terminals forming asymmetrical (putative excitatory) synapses were dendritic spines, most of which were M2R(+) . The main targets of M2R(+) terminals forming symmetrical (putative inhibitory or neuromodulatory) synapses were unlabeled perikarya and M2R(+) dendritic shafts. M2R-ir was also seen in VAChT(+) cholinergic terminals, indicating a possible autoreceptor role. These findings suggest that M2R-mediated mechanisms in the BLa are very complex, involving postsynaptic effects in dendrites as well as regulating release of glutamate, γ-aminobutyric acid, and acetylcholine from presynaptic axon terminals. J. Comp. Neurol. 524:2400-2417, 2016. © 2016 Wiley Periodicals, Inc.

Specific neuronal populations in the basolateral amygdala (ABL) exhibit immunoreactivity for distinct neuropeptides and calcium-binding proteins. In the present study, immunohistochemical techniques were used to analyze neurons in the rat ABL that contain cholecystokinin (CCK). Some pyramidal projection neurons in the anterior subdivision of the basolateral nucleus exhibited low levels of CCK immunoreactivity in rats that received injections of colchicine to interrupt axonal transport; staining was concentrated in the axon initial segments of these cells. High levels of CCK immunoreactivity were observed in two subpopulations of nonpyramidal interneurons in all nuclei of the ABL: (1) type L neurons (characterized by large somata and thick dendrites), and (2) type S neurons (characterized by small somata and thin dendrites). Dual-labeling immunofluorescence studies using confocal laser scanning microscopy revealed that many (30-40%) type L CCK+ interneurons exhibited immunoreactivity for calbindin (CB), but not for parvalbumin (PV), calretinin (CR), or vasoactive intestinal polypeptide (VIP). In contrast, there was extensive colocalization of CR and VIP with CCK in type S neurons, but no significant colocalization with CB or PV. In addition, the majority of CR and VIP interneurons exhibited colocalization of both neurochemicals. Collectively, the results of this and previous studies indicate that there are at least four distinct interneuronal subpopulations in the ABL: (1) PV+ neurons (the great majority of which are CB+); (2) SOM+ neurons (many of which are CB+ and NPY+); (3) large CCK+ neurons (some of which are CB+); and (4) small bipolar/bitufted neurons that exhibit various amounts of colocalization of CCK, VIP, and CR.

Inhibitory circuits in the basolateral nuclear complex of the amygdala (BNC) critical for controlling the acquisition, expression, and extinction of emotional responses are mediated by GABAergic interneurons (INs). Studies in rodents have demonstrated that separate IN subpopulations, identified by their expression of calcium-binding proteins and neuropeptides, play discrete roles in the intrinsic circuitry of the BNC. Far less is known about IN subpopulations in primates. In order to fill in this gap in our understanding of primate INs, the present investigation used dual-labeling immunohistochemistry for IN markers to identify subpopulations expressing cholecystokinin (CCK), calbindin (CB), calretinin (CR), and somatostatin (SOM) in somata and axon terminals in the monkey BNC. In general, colocalization patterns seen in somata and axon terminals were similar. It was found that there was virtually no colocalization of CB and CR, the two calcium-binding proteins investigated. Three subtypes of CCK-immunoreactive (CCK+) INs were identified on the basis of their expression of CR or CB: (1) CCK+/CR+; (2) CCK+/CB+); and (3) CCK+/CR-/CB-. Almost no colocalization of CCK with SOM was observed, but there was extensive colocalization of SOM and CB. CCK+, CR+, and CCK+/CR+ double-labeled axon terminals were seen surrounding pyramidal cell somata in basket-like plexuses, as well as in the neuropil. CB+, SOM+, and CB+/SOM+ terminals did not form baskets, suggesting that these IN subpopulations are mainly dendrite-targeting neurons. In general, the IN subpopulations in the monkey are not dissimilar to those seen in rodents but, unlike rodents, CB+ INs in the monkey are not basket cells.

The cholinergic innervation of the neocortex, hippocampus, and basolateral amygdala is critical for higher cognitive functions, including attention and memory. One action of ACh in the hippocampus is the potentiation of NMDA receptor (NMDAR) currents in pyramidal neurons (PNs) by M1 muscarinic receptors (M1Rs). The increase in these currents enhances long-term potentiation (LTP), an important mechanism for memory formation. Ultrastructural observations in the hippocampus revealed that M1Rs and NMDARs were colocalized in hippocampal PN dendrites, consistent with electrophysiological studies demonstrating M1R-NMDAR interactions. Similar to the hippocampus, activation of M1Rs have been shown to be critical for mnemonic functions in the anterior basolateral nucleus of the amygdala (BLa). In the present study dual-labeling immunoelectron microscopy was used to determine if there was colocalization of M1Rs and NMDARs in neurons of the mouse BLa. We found extensive colocalization of these receptors in dendrites and spines of BLa PNs, and most of the M1Rs were membrane-associated where they could be activated by released acetylcholine. These results suggest that M1Rs in BLa PNs could be targeted by M1R positive allosteric modulators (PAMs), resulting in amelioration of memory impairments in neuropsychiatric disorders, such as Alzheimer's disease, by potentiating NMDAR currents in the amygdala.

The hippocampus and amygdala are key structures of the limbic system whose connections include reciprocal interactions with the basal forebrain (BF). The hippocampus receives both cholinergic and GABAergic afferents from the medial septal area of the BF. Hippocampal projections back to the medial septal area arise from non-pyramidal GABAergic neurons that express somatostatin (SOM), calbindin (CB), and neuropeptide Y (NPY). Recent experiments in our lab have demonstrated that the basolateral amygdala, like the hippocampus, receives both cholinergic and GABAergic afferents from the BF. These projections arise from neurons in the substantia innominata (SI) and ventral pallidum (VP). It remained to be determined, however, whether the amygdala has projections back to the BF that arise from GABAergic non-pyramidal neurons. This question was investigated in the present study by combining Fluorogold (FG) retrograde tract tracing with immunohistochemistry for GABAergic non-pyramidal cell markers, including SOM, CB, NPY, parvalbumin, calretinin, and glutamic acid decarboxylase (GAD). FG injections into the BF produced a diffuse array of retrogradely labeled neurons in many nuclei of the amygdala. The great majority of amygdalar FG+ neurons did not express non-pyramidal cell markers. However, a subpopulation of non-pyramidal SOM+ neurons, termed "long-range non-pyramidal neurons" (LRNP neurons), in the external capsule, basolateral amygdala, and cortical and medial amygdalar nuclei were FG+. About one-third of the SOM+ LRNP neurons were CB+ or NPY+, and one-half were GAD+. It remains to be determined if these inhibitory amygdalar projections to the BF, like those from the hippocampus, are important for regulating synchronous oscillations in the amygdalar-BF network.

The basolateral amygdala receives a very dense cholinergic innervation from the basal forebrain that is important for memory consolidation. Although behavioral studies have shown that both M1 and M2 muscarinic receptors are critical for these mnemonic functions, there have been very few neuroanatomical and electrophysiological investigations of the localization and function of different types of muscarinic receptors in the amygdala. In the present study we investigated the subcellular localization of M2 muscarinic receptors (M2Rs) in the anterior basolateral nucleus (BLa) of the mouse, including the localization of M2Rs in parvalbumin (PV) immunoreactive interneurons, using double-labeling immunoelectron microscopy. Little if any M2R-immunoreactivity (M2R-ir) was observed in neuronal somata, but the neuropil was densely labeled. Ultrastructural analysis using a pre-embedding immunogold-silver technique (IGS) demonstrated M2R-ir in dendritic shafts, spines, and axon terminals forming asymmetrical (excitatory) or symmetrical (mostly inhibitory) synapses. In addition, about one-quarter of PV+ axon terminals and half of PV+ dendrites, localized using immunoperoxidase, were M2R+ when observed in single thin sections. In all M2R+ neuropilar structures, including those that were PV+, about one-quarter to two-thirds of M2R+ immunoparticles were plasma-membrane-associated, depending on the structure. The expression of M2Rs in PV+ and PV-negative terminals forming symmetrical synapses indicates M2R modulation of inhibitory transmission. Electrophysiological studies in mouse and rat brain slices, including paired recordings from interneurons and pyramidal projection neurons, demonstrated M2R-mediated suppression of GABA release. These findings suggest cell-type-specific functions of M2Rs and shed light on organizing principles of cholinergic modulation in the BLa.

The basolateral nucleus of the amygdala receives an extremely dense cholinergic innervation from the basal forebrain that is critical for memory consolidation. Although previous electron microscopic studies determined some of the postsynaptic targets of cholinergic afferents, the majority of postsynaptic structures were dendritic shafts whose neurons of origin were not identified. To make this determination, the present study analyzed the cholinergic innervation of the anterior subdivision of the basolateral amygdalar nucleus (BLa) of the rat using electron microscopic dual-labeling immunocytochemistry. The vesicular acetylcholine transporter (VAChT) was used as a marker for cholinergic terminals; calcium/calmodulin-dependent protein kinase II (CaMK) was used as a marker for pyramidal cells, the principal neurons of the BLa; and parvalbumin (PV) was used as a marker for the predominant interneuronal subpopulation in this nucleus. VAChT(+) terminals were visualized by using diaminobenzidine as a chromogen, whereas CAMK(+) or PV(+) neurons were visualized with Vector very intense purple (VIP) as a chromogen. Quantitative analyses revealed that the great majority of dendritic shafts receiving cholinergic inputs were CAMK(+) , indicating that they were of pyramidal cell origin. In fact, 89% of the postsynaptic targets of cholinergic terminals in the BLa were pyramidal cells, including perikarya (3%), dendritic shafts (47%), and dendritic spines (39%). PV(+) structures, including perikarya and dendrites, constituted 7% of the postsynaptic targets of cholinergic axon terminals. The cholinergic innervation of both pyramidal cells and PV(+) interneurons may constitute an anatomical substrate for the generation of oscillatory activity involved in memory consolidation by the BLa.

Although it is known that acetylcholine acting through M1 muscarinic receptors (M1Rs) is essential for memory consolidation in the anterior basolateral nucleus of the amygdala (BLa), virtually nothing is known about the circuits involved. In the hippocampus M1R activation facilitates long-term potentiation (LTP) by potentiating NMDA glutamate receptor (NMDAR) currents. The majority of NMDAR+ profiles in the BLa are spines. Since about half of dendritic spines of BLa pyramidal neurons (PNs) receiving glutamatergic inputs are M1R-immunoreactive (M1R+) it is possible that the role of M1Rs in BLa mnemonic functions also involves potentiation of NMDAR currents in spines. However, the finding that only about half of BLa spines are M1R+ suggests that this proposed mechanism may only apply to a subset of glutamatergic inputs. As a first step in the identification of differential glutamatergic inputs to M1R+ spines in the BLa, the present electron microscopic study used antibodies to two different vesicular glutamate transporter proteins (VGluTs) to label two different subsets of glutamatergic inputs to M1R+ spines. These inputs are largely complimentary with VGluT1+ inputs arising mainly from cortical structures and the basolateral nucleus, and VGluT2+ inputs arising mainly from the thalamus. It was found that about one-half of the spines that were postsynaptic to VGluT1+ or VGluT2+ terminals were M1R+. In addition, a subset of the VGluT1+ or VGluT2+ axon terminals were M1R+, including those that synapsed with M1R+ spines. These results suggest that acetylcholine can modulate glutamatergic inputs to BLa spines by presynaptic as well as postsynaptic M1R-mediated mechanisms.

Several distinct subpopulations of interneurons (INs) in the amygdalar basolateral nuclear complex (BNC) of the rat can be recognized on the basis of their expression of calcium-binding proteins and neuropeptides, including parvalbumin (PV), somatostatin (SOM), calretinin (CR), and cholecystokinin (CCK). In the rat BNC CCK is expressed in two separate IN subpopulations, termed large (CCKL ) and small (CCKS ). These subpopulations exhibit distinct connections indicative of discrete functional roles in the circuitry of the BNC. Although there have been several studies of PV+, SOM+, and CR+ INs in the primate BNC, there is almost no information regarding CCK+ INs in these species. Therefore, in the present study the distribution and morphology of CCK+ INs and their axon terminals in the BNC of the monkey was investigated. CCK immunoreactivity in the BNC was observed in somata and proximal dendrites of nonpyramidal neurons, as well as in axon terminals. A moderate density of CCK+ INs was found in all nuclei of the BNC. CCK+ INs in the BNC were morphologically heterogeneous, with both small and large varieties observed. All CCK+ somata gave rise to 2-4 dendrites that branched sparingly and were aspiny. CCK+ axon terminals in the BNC were found both in the neuropil and forming pericellular baskets contacting somata of pyramidal cells. In addition, many CCK+ neurons were contacted by multiple CCK+ terminals, indicative of the existence of a CCK interneuronal network. These data indicate that the morphology of CCK+ INs in the monkey is very similar to that of the rat.

Studies in rodents have shown that interactions between cholecystokinin (CCK) and the endogenous cannabinoid system in the basolateral nuclear complex of the amygdala (BNC) modulate anxiety-like behavior and fear learning/expression. One of the main cell types implicated is a CCK-immunoreactive (CCK+) basket cell that innervates the somata of pyramidal projection neurons (PNs) and expresses the type 1 cannabinoid receptor (CB1R) in its axon terminals. Although numerous studies have elucidated the anatomy and physiology of these CCK+/CB1R + interneurons in rodents, it has not been determined if they exist in primates. The present investigation used immunohistochemical techniques in the monkey to answer this question. It was found that the monkey BNC, as in rodents, has a very high density of CB1R + axons, including CB1R + axon terminals that form basket-like plexuses contacting somata of PNs. These axons, as well as axons in the neuropil, exhibit extensive colocalization of CCK and CB1R. These findings suggest that the same synaptic mechanisms involved in CCK-CB1R interactions in rodents may also apply to primates, and that therapies that target the cannabinoid system in the BNC may be useful for treating fear and anxiety in human patients.

There are discrete subpopulations of GABAergic interneurons in the basolateral amygdala (ABL) that contain particular neuropeptides or calcium-binding proteins (calbindin-D28k, parvalbumin (PV), or calretinin). The present study employed a dual-labeling immunofluorescence technique combined with confocal laser scanning microscopy to investigate the neurochemical characteristics of the interneuronal subpopulation containing somatostatin (SOM). The great majority of SOM+ neurons in the ABL exhibited GABA immunoreactivity (66-82% depending on the nucleus). These SOM+ neurons constituted 11-18% of the GABA+ population. There was also extensive colocalization of SOM with calbindin (CB) in all nuclei of the ABL, but no colocalization of SOM with parvalbumin, calretinin, or vasoactive intestinal polypeptide. In the basolateral nucleus more than 90% of SOM+ neurons also exhibited CB immunoreactivity, whereas in the lateral nucleus about two-thirds of SOM+ neurons contained significant levels of CB. These SOM/CB neurons constituted about one quarter of the CB+ population in the basolateral nucleus and about one third of the CB+ population in the lateral nucleus. These results, in conjunction with the findings of previous studies, indicate that there are at least three major subpopulations of GABAergic interneurons in the ABL: (i) SOM+ neurons (most of which also contain CB and/or neuropeptide Y); (ii) PV+ neurons (most of which also contain CB); and (iii) CR+ neurons (most of which also contain vasoactive intestinal polypeptide).

The basolateral amygdala contains several subpopulations of inhibitory interneurons that can be distinguished on the basis of their content of calcium-binding proteins or peptides. Although previous studies have shown that interneuronal subpopulations containing parvalbumin (PV) or vasoactive intestinal peptide (VIP) innervate distinct postsynaptic domains of pyramidal cells as well as other interneurons, very little is known about the synaptic outputs of the interneuronal subpopulation that expresses somatostatin (SOM). The present study utilized dual-labeling immunocytochemical techniques at the light and electron microscopic levels to analyze the innervation of pyramidal cells, PV+ interneurons, and VIP+ interneurons in the anterior basolateral amygdalar nucleus (BLa) by SOM+ axon terminals. Pyramidal cell somata and dendrites were selectively labeled with antibodies to calcium/calmodulin-dependent protein kinase II (CaMK); previous studies have shown that the vast majority of dendritic spines, whether CAMK+ or not, arise from pyramidal cells. Almost all SOM+ axon terminals formed symmetrical synapses. The main postsynaptic targets of SOM+ terminals were small-caliber CaMK+ dendrites and dendritic spines, some of which were CaMK+. These SOM+ synapses with dendrites were often in close proximity to asymmetrical (excitatory) synapses to these same structures formed by unlabeled terminals. Few SOM+ terminals formed synapses with CaMK+ pyramidal cell somata or large-caliber (proximal) dendrites. Likewise, only 15% of SOM+ terminals formed synapses with PV+, VIP+, or SOM+ interneurons. These findings suggest that inhibitory inputs from SOM+ interneurons may interact with excitatory inputs to pyramidal cell distal dendrites in the BLa. These interactions might affect synaptic plasticity related to emotional learning.