Subsynaptic expression of glutamate delta receptor 1 at calcitonin gene-related peptide-containing projections from the parabrachial region in the central amygdala of mice and monkeys

The lateral and capsular divisions of the central amygdala (CeLC) is a key region involved in the processing of emotional-affective dimensions of pain. In the spino-parabrachio-amygdaloid pain pathway, the CeLC receives nociceptive information from calcitonin gene-related peptide (CGRP) neurons in the parabrachial nucleus. Recent evidence indicates that glutamate delta receptor 1 (GluD1) regulates this projection in mice. GluD1 is an atypical ionotropic glutamate receptor that largely functions as a synaptogenic molecule involved in the formation and maintenance of synapses. Despite its strong cellular expression, little is known about the subsynaptic localization of GluD1 and its potential interaction with CGRP terminals in CeLC neurons. To address this issue, we used single and double immuno-electron microscopy techniques in rodents and monkeys. In both species, CGRP-positive (CGRP+) terminals formed symmetric and asymmetric synapses with dendrites, symmetric synapses with soma, and less commonly, asymmetric synapses with spines. Approximately 80% of CGRP+ terminals forming clear symmetric synapses expressed vGluT2 immunoreactivity and none were immunoreactive for GABA, confirming the glutamatergic nature of this projection and suggesting that some PB-CGRP terminals may modulate transmission to the CeLC in a peptidergic manner. GluD1 was expressed in the core of symmetric axo-dendritic and axo-somatic synapses and peri-synaptic to asymmetric synapses. These findings show that the ascending CGRP+PB-CeLC projection mediates its effects through a heterogenous population of terminals that display strong synaptic relationships with GluD1 in both mice and monkeys. Title: Subsynaptic expression of glutamate delta receptor 1 at calcitonin gene-related peptide-containing projections from the parabrachial region in the central amygdala of mice and monkeys Authors: Diane Choi 1,2, Karina Dalal 2, Chris Kozuch 2, Jean-Francois Paré 2, Shashank M. Dravid 4, and Yoland Smith 1,2,3 Affiliations: 1 Graduate Program in Molecular and Systems Pharmacology, Emory University, Atlanta, GA 30329 2 Emory National Primate Research Center, Emory University, Atlanta, GA 30329 3 Department of Neurology, Emory University, Atlanta, GA 30329 4 Department of Psychiatry and Behavioral Sciences, Texas A&M University, College Station, TX 77840 Corresponding Author : Yoland Smith, [email protected]

We would like to thank Susan Jenkins for her technical assistance. This work was supported by T32 grant GM 8602-24, NIH grant R01 NS118731, and the NIH/ORIP Emory Primate Center base grant P51OD011132. Conflict of Interest: The authors declare no conflict of interest.

pain, CGRP, parabrachial, electron microscopy, immunogold, vGluT2 Graphical Text: This study used immuno-electron microscopy approaches to demonstrate that GluD1 is strongly expressed at symmetric axo-somatic and axo-dendritic synapses formed by vGluT2-containing CGRP+ terminals in the mouse and monkey central amygdala.

The lateral and capsular divisions of the central amygdala (CeLC) is a key region involved in the processing of emotional-affective dimensions of pain. In the spino-parabrachio-amygdaloid pain pathway, the CeLC receives nociceptive information from calcitonin gene-related peptide (CGRP) neurons in the parabrachial nucleus. Recent evidence indicates that glutamate delta receptor 1 (GluD1) regulates this projection in mice. GluD1 is an atypical ionotropic glutamate receptor that largely functions as a synaptogenic molecule involved in the formation and maintenance of synapses. Despite its strong cellular expression, little is known about the subsynaptic localization of GluD1 and its potential interaction with CGRP terminals in CeLC neurons. To address this issue, we used single and double immuno-electron microscopy techniques in rodents and monkeys. In both species, CGRP-positive (CGRP+) terminals formed symmetric and asymmetric synapses with dendrites, symmetric synapses with soma, and less commonly, asymmetric synapses with spines. Approximately 80% of CGRP+ terminals forming clear symmetric synapses expressed vGluT2 immunoreactivity and none were immunoreactive for GABA, confirming the glutamatergic nature of this projection and suggesting that some PB-CGRP terminals may modulate transmission to the CeLC in a peptidergic manner. GluD1 was expressed in the core of symmetric axo-dendritic and axo-somatic synapses and peri-synaptic to asymmetric synapses. These findings show that the ascending CGRP+PB-CeLC projection mediates its effects through a heterogenous population of terminals that display strong synaptic relationships with GluD1 in both mice and monkeys. Abbreviations BSA (Bovine Serum Albumin) CeA (central amygdala) CeLC (lateral and capsular divisions of the central amygdala) Cbln (cerebellin) CGRP (calcitonin gene-related peptide) De (dendrite) DAB (diaminobenzidine tetrahydrochloride) EM (electron microscopy) GABA (gamma-aminobutyric acid) GluD1 (glutamate delta receptor 1) LM (light microscopy) m (mitochondria) PB (parabrachial nucleus) PBS (phosphate buffer saline) PKCδ+ (protein kinase C δ+) RT (room temperature) Sp (spine) TRIS (tris(hydroxymethyl)aminomethane) vGluT2 (vesicular glutamate transporter 2) 1 Introduction Approximately 21% of U.S. adults experience chronic pain [56], which can severely impact quality-of-life due to physical and emotional burdens. Pain is a complex disorder that consists of multiple dimensions, including sensorimotor, cognitive, and emotional-affective aspects. The central nucleus of the amygdala (CeA) has been implicated as a key region for the processing of emotional-affective components of pain [66, 70]. In particular, the lateral and capsular divisions of the central amygdala (CeLC), also coined as the “nociceptive amygdala,” serves as an integrative hub for nociceptive and emotional-affective processing of pain [13, 66]. The CeLC receives glutamatergic projections from the external lateral parabrachial nucleus (PB) via the spino-parabrachio-amygdaloid pain pathway [13, 38, 63], which is well-studied for its role in nociceptive pain modulation [3, 48, 66]. The PB is the sole source of calcitonin gene-related peptide (CGRP)-containing inputs to the CeLC [7, 8, 34, 58, 61], a projection that plays a critical role in the transduction of emotional-affective pain signals through the mediation of nociceptive synaptic transmission and plasticity [15-17, 50, 62]. The PB-CGRP+ terminals mainly target protein kinase C δ (PKCδ+) neurons, one of two major cell types in the CeLC [1, 28, 37, 71], considered “pronociceptive” because their activation increases pain-related responses [71]. Thus, it is presumed that the PB-CeLC CGRP+ projection modulates pain through its synaptic signaling with PKCδ+ neurons in the CeLC [38, 71]. However, there remains a significant knowledge gap in our understanding of the structural synaptic signaling occurring at these PB-CeLC synapses. Recent evidence suggests that glutamate delta receptors (GluDs) may also play a role in regulating transmission of nociceptive pain signals to the CeLC. GluDs are unusual members of the ionotropic glutamate receptor superfamily that do not exhibit typical agonist-induced ionotropic activity [33, 40]. Instead, they function as synaptogenic molecules, forming trans-synaptic complexes with the presynaptic proteins neurexin and cerebellin (Cbln) to regulate synapse formation and maintenance [10, 42, 68]. The role of GluD1 in synaptic transmission has been explored in various regions, including the hippocampus and striatum [35, 39, 65]. A recent study has demonstrated that disruption of GluD1 impairs excitatory transmission at PB-CeLC synapses in mice [12]. Furthermore, dysfunction of the trans-synaptic signaling between GluD1 and Cbln1, which is heavily expressed in PB neurons [47, 52], has been implicated in both inflammatory and neuropathic rodent models of pain [12]. GluD1-Cbln1 interactions contribute to the regulation of PB-CeLC synapses and pain-related behaviors [12], our understanding of the underlying substrate of the functional interactions between GluD1 and CGRP+PB-CeLC synapses remains incomplete. In this study, we used double immuno-electron microscopy approaches to map the subsynaptic localization of GluD1 at CGRP+PB-CeLC synapses in mice and monkeys. The extension of this analysis to nonhuman primates will strengthen the foundation for the translation of knowledge gained through this study to humans. 2 Materials & Methods 2.1 Animals For the characterization of CGRP and GluD1 in the CeLC, we used tissue from a total of 8 mice (C57Bl/6J, 1 female, 7 males, 9 - 16 weeks old) and 7 rhesus monkeys (Macaca mulatta, 3 females, 4 males, 3 - 11 years old). Five mice brains were received from Dr. Shashank Dravid’s laboratory. The remaining 3 mice and all monkeys were from the Emory National Primate Research Center breeding colonies. All animals were deeply anesthetized with either isoflurane and an overdose of ketamine (100-150 mg/kg, IP) and dormitor (0.1 mg/kg, IP) for mice or pentobarbital (100 mg/kg, IV) for monkeys. Animals were then transcardially perfused with cold oxygenated Ringer’s solution followed by a fixative containing 4% paraformaldehyde and 0.1% glutaraldehyde in phosphate buffer (0.1M, pH 7.4). Afterwards, brains were collected and post-fixed in paraformaldehyde for 24 hours. Tissue blocks were cut in 50 µm-thick coronal sections with a vibrating microtome. All animal procedures were in line with the National Institutes of Health guidelines for the use of animals in research and approved by the Creighton University and Emory University Institutional Animal Care and Use Committee Policies and Procedures (IACUC). 2.2 Antibodies All commercially available antibodies used in these studies are well characterized and recorded in The Antibody Registry©. The specificity of the GluD1 antibody has been previously reported [22] and validated by the lack of staining in the striatum of GluD1 KO mice [6, 39]. Table 1. List of primary antibodies used in this study | CGRP – 414 004 | Guinea Pig | Synaptic Systems | AB_2737049 | 1:5000 | | GluD1 – AF1390 | Rabbit | Frontiers Inst. Company Ltd | AB_2571757 | 1:3000 | | GABA – A2052 | Rabbit | Sigma-Aldrich | AB_477652 | 1:1000 | | vGluT2 – VGT-3 (rodent) | Rabbit | MAb Technologies | AB_2315568 | 1:5000 | | vGluT2 – VGT2-6 (monkey) | Rabbit | MAb Technologies | AB_2315569 | 1:1000 | 2.3 Light Microscopy (LM) Immunohistochemistry 2.3.1 CGRP Immunoperoxidase localization studies Sections of mice and monkey brain tissue containing the right CeLC were placed in sodium borohydride (1% in phosphate buffered saline, PBS 0.01 M, pH 7.4) for 20 minutes and then rinsed five times in PBS. Sections were then placed in a pre-incubation solution [1% normal goat serum, 1% Bovine Serum Albumin (BSA), 0.3% Triton-X-100, and PBS] for 60 minutes at room temperature (RT). This was followed by incubation in the primary antibody solution [CGRP antibody (Table 1), 1% normal goat serum, 1% BSA, 0.3% Triton-X-100, and PBS] for 24 hours at RT. After primary antibody incubation, sections were rinsed in PBS and then placed in the secondary antibody solution [goat anti-guinea pig biotinylated IgG (1:200 in PBS; Vector Laboratories), 1% normal goat serum, 1% BSA, 0.3% Triton-X-100, and PBS] for 90 minutes at RT. Sections were then rinsed with PBS and placed in an avidin-biotin-peroxidase complex (ABC; Vector Laboratories) solution for 90 minutes at RT. Sections were rinsed in PBS and then with tris(hydroxymethyl)aminomethane (TRIS, 0.05M, pH 7.6) before being incubated in a solution containing 0.01M imidazole, 0.005% hydrogen peroxide, and 0.025% 3,3′-diaminobenzidine tetrahydrochloride (DAB; Sigma) in Tris for 10 minutes at RT. After several rinses in PBS, sections were mounted on gelatin-coated slides, dehydrated, and cover-slipped. Sections were digitally scanned by an Aperio ScanScope CS system (Aperio Technologies, Vista, CA) and analyzed using ImageScope software (Aperio Technologies). 2.3.2 GluD1 Immunoperoxidase localization studies Sections underwent LM immunohistochemistry, as described above, with the primary GluD1-rabbit antibody solution (Table 1) and secondary goat anti-rabbit biotinylated IgG (1:200 in PBS; Vector Laboratories) solution. 2.4 Electron Microscopy (EM) Immunohistochemistry 2.4.1 Pre-embedding CGRP immunoperoxidase labeling After sodium borohydride treatment, sections were placed in a 100% cryoprotectant solution (25% sucrose and 10% glycerol in PB 0.1M, pH 7.4) for 20 minutes at RT and then placed in an -80°C freezer for 20 minutes. Next, sections were reintroduced to 100% cryoprotectant solution for 10 minutes at RT, followed by diluted (70%, 50%, 30%) cryoprotectant solutions for 10 minutes each. Sections then were processed through the same primary and secondary incubation steps as the LM immunohistochemistry method above, with two changes: (1) the omission of Triton-X-100 from all solutions, and (2) the primary incubation was extended to 48 hours at 4°C. After the DAB/peroxidase reaction, sections were rinsed in PB (0.1M, pH 7.4) and underwent EM processing. In brief, the sections were post-fixed in 1% osmium tetroxide solution for 20 minutes, followed by washes in PB. Sections were then dehydrated with increasing concentrations of alcohol before being placed in propylene oxide. For the 70% alcohol solution, 1% uranyl acetate was added to increase contrast at the EM level. Afterwards, sections were embedded in epoxy resin (Durcupan, ACM; Fluka, Buchs, Switzerland) for 12 hours, mounted onto oil-coated slides, and cover-slipped before being placed in a 60°C oven for 48 hours. Blocks of tissue containing the CeLC were taken from the slides, glued on resin blocks, trimmed and serially cut into 60-nm sections using an ultramicrotome (Ultra-cut T2; Leica, Germany Leica). The ultrathin sections were collected onto single slot Pioloform-coated copper grids, stained with lead citrate for 5 minutes, and examined with a JEOL electron microscope (JEOL JEM 1011). 2.4.2 Double pre-embedding immunoperoxidase and immunogold labeling 2.4.2.1 CGRP-GluD1 To characterize the relationship between CGRP+ terminals and GluD1 on the surface of CeLC neuronal elements, sections from the right CeLC were processed for double immuno-electron microscopy localization of CGRP (immunoperoxidase) and GluD1 (pre-embedding immunogold). After sodium borohydride and cryoprotectant treatments, sections were placed in a pre-incubation solution containing 5% dry milk and PBS for 30 minutes at RT. Sections were then rinsed with TBS containing 0.1% gelatin (TBS-gelatin) and incubated in the primary antibody solution [CGRP primary guinea pig antibody, GluD1 primary rabbit antibody (Table 1), 1% dry milk, and TBS-gelatin] for 24 hours at RT. After rinses in TBS-gelatin, sections were incubated in the secondary antibody solution [15 nm gold-conjugated goat anti-rabbit IgG (1:100 in TBS-gelatin; Nanoprobes Inc., goat anti-guinea pig biotinylated IgG (1:200 in TBS-gelatin; Vector Laboratories), 1% dry milk, and TBS-gelatin] for 2 hours at RT. Sections were rinsed in TBS-gelatin and 2% sodium acetate buffer before gold particles were silver intensified with the HQ silver kit (Nanoprobes Inc.) for approximately 16-20 minutes. After silver intensification, the ABC and DAB procedures were the same as described above in the pre-embedding immunoperoxidase method. After the DAB reaction, sections were rinsed with PB (0.1M, pH 7.4) before they underwent EM processing (osmification, dehydration, and embedding) as described in the pre-embedding immunoperoxidase procedure with the following changes: (1) the sections were fixed in 0.5% osmium tetroxide solution for 10 minutes instead of 20, and (2) sections were stained with 1% uranyl acetate solution for 10 minutes instead of 35. 2.4.2.2 CGRP-vGluT2 To assess the extent of the co-expression of vGluT2 and CGRP in axon terminals, sections from the right CeLC were processed for double immuno-electron microscopy localization of CGRP (immunogold) and vGLuT2 (immunoperoxidase) as described above with the following changes: (1) sections were incubated in the primary antibody solution [CGRP guinea pig primary antibody, vGluT2 primary rabbit antibody] (Table 1), and (2) sections were incubated in the secondary antibody solution [15nm gold-conjugated goat anti-guinea pig (1:100 in TBS-gelatin; Nanoprobes Inc.), goat anti-rabbit biotinylated IgG (1:200 in TBS-gelatin; Vector Laboratories)]. Because of differences in the COOH terminus between rodents and primates, different well-characterized vGluT2 primary antibodies were used for rodent and monkey tissues (Table 1). 2.5 Post-embedding Immunogold Ultrathin sections of CeLC tissue, previously immunostained with the CGRP and GluD1 antibodies from the double pre-embedding immunoperoxidase and immunogold protocol mentioned above, were serially cut using an ultramicrotome and collected onto gold grids. Once dry, the grids were placed in the pre-incubation solution [0.01% Triton-X-100 (TBS-T 0.01%; pH 7.6)] for 10 minutes at RT. Grids were then incubated in the primary antibody solution [GABA antibody (Table 1) and TBS-T 0.01%] for 24 hours at RT. After rinses with TBS-T 0.01% and TBS (0.5M, pH 8.2), the grids were incubated in the secondary solution [15 nm gold-conjugated goat anti-rabbit IgG (1:50 in TBS (pH 8.2); British Biocell)] for 90 minutes at RT before being in TBS (pH 8.2) for 20 minutes, washed in distilled water for 5 minutes, stained with 1% uranyl acetate (in distilled water) for 5 minutes, washed in distilled water and stained with lead citrate for 5 minutes. 2.6 Data Analysis All electron micrographs were taken at 40,000X magnification with a CCD camera (Gatan Model 785; Gatan, Inc., Pleasanton, CA). Images were analyzed with the Gatan Digital Micrograph software (Version 3.10.1). Some of the micrographs were adjusted for brightness or contrast using either the Digital Micrograph software or Adobe Photoshop (Version 25.9.1). Micrographs were compiled into figures in Adobe Photoshop (Version 25.9.1). Statistical analyses using chi-squared tests were conducted using R (Version 4.4.1 (2024-06-14 ucrt)). Statistical analyses using unpaired t-tests (i.e., comparisons between species) were conducted using GraphPad Prism Software (Version 10.4.1 (627)). All graphical data are presented as average percentage values ± Standard Error of the Mean (SEM). 2.6.1 Immunoperoxidase-stained tissue Data were collected from a total of 6 blocks of right CeLC tissue (1 block/animal for 3 mice and 3 monkeys each), with 50-100 electron micrographs taken of randomly selected immunoperoxidase-labeled CGRP terminals per animal for a total surface area of 2886 µm 2 in mice and 2190 µm 2 in monkeys. Labeled terminals were categorized as forming axo-dendritic, axo-spinous, or axo-somatic synapses based on ultrastructural features [54]. The length, or cross-sectional diameter, of axon terminals was determined by measuring the longest axis of the terminal parallel to the synaptic junction. The relative percentage of symmetric versus asymmetric synapses formed by CGRP+ terminals was calculated by dividing the number of each type of synapse by the total number of synapses formed by CGRP+ terminals per animal. The mean percentage of each type of synapse was compared between mice and monkeys through analysis of n=200 and n=163 synapses formed by CGRP+ terminals in mice and monkeys, respectively. The mean number of CGRP-labeled terminals in each category was calculated and statistically compared for each animal species using chi-squared analyses. The percentage of symmetric (lack of thick post-synaptic density) versus asymmetric (presence of thick post-synaptic density) synapses [54] formed by CGRP+ terminals was compared between species using unpaired t-tests. 2.6.2 Double pre-embedding immunoperoxidase and immunogold stained tissue To determine the synaptic relationships between CGRP+ terminals and post-synaptic GluD1 localization, ultrathin sections of double-immunostained tissue (CGRP/GluD1) were examined. Data were collected from a total of 6 blocks of right CeLC tissue (1 block/animal for 3 mice and 3 monkeys each). Approximately 60-100 electron micrographs of randomly selected CGRP immunoperoxidase-labeled terminals were taken from the most superficial sections of tissue blocks that contained both reaction products. A minimum of two or more gold particles was required to be considered GluD1-immunoreactive. A total surface area of 3137 µm 2 and 2753 µm 2 of right CeLC tissue was examined from mice and monkeys, respectively. The percentages of CGRP-labeled terminals that expressed post-synaptic GluD1 immunogold labeling was calculated for CGRP-labeled terminals from mice (n=197) and monkeys (n=260). The mean percentage of CGRP-containing terminals with post-synaptic GluD1 labeling was calculated for each animal species. Data were analyzed for significant differences between species using unpaired t-tests. To determine the extent of vGluT2 co-expression in CGRP+ terminals, ultrathin sections of double-immunostained tissue (CGRP/vGluT2) were examined as described above. Data were collected from a total of 6 blocks of right CeLC tissue (1 block/animal for 3 mice and 3 monkeys each). Approximately 60-100 electron micrographs of randomly selected CGRP immunogold-labeled terminals per animal for a total surface area of 3019 µm 2 in mice and 3197 µm 2 in monkeys. A minimum of five or more gold particles was required for a terminal to be considered CGRP-immunoreactive. The mean percentage of CGRP-labeled terminals that co-expressed vGluT2 immunoperoxidase-labeling was compared between mice (n=275) and monkeys (n=235) using unpaired t-tests. 2.6.3 Triple pre-embedding immunogold GluD1 & immunoperoxidase CGRP /Post-embedding immunogold GABA To determine if CGRP+ terminals that expressed post-synaptic GluD1 immunogold labeling were GABA-positive, ultrathin sections of triple-immunostained tissue (CGRP/GluD1/GABA) were examined. From this tissue, electron micrographs of CGRP+ terminals that expressed post-synaptic GluD1 labeling were taken from both mice and monkey right CeLC tissue and categorized as GABA-immunoreactive or not based on the density of 15nm gold particle labeling associated with them. In this material, the specificity of the post-embedding immunogold labeling for putative GABA-containing terminals was determined by calculating the density of gold particles in every terminal that formed clear symmetric or asymmetric CGRP/GluD1-labeled synapses and dividing the number of gold particles in each individual terminal by the cross-sectional area of that terminal (ImageJ, Version 1.41o). A terminal would be categorized as “Symmetric/GABA+” if it met the following criteria: (1) it formed a clear symmetric synapse in the plane of section examined, and (2) the gold particles density was 2.5 times greater than the average gold particle density associated with terminals in the same tissue forming asymmetric synapses (i.e., putative glutamatergic). A terminal would be categorized as “Symmetric/GABA-” if it formed a symmetric synapse, but expressed a gold particle density below the aforementioned cutoff. In this series of experiments, we qualitatively assessed the gold particle density of CGRP-GluD1 synapses and compared them to the gold particle density of terminals forming clear asymmetric and symmetric synapses. To account for variability in the absolute number of gold particles across post-embedding reactions and animals, micrographs were collected from each run of post-embedding reaction from each animal. Due to technical difficulties in retaining immunogold labeling for GluD1 after post-embedding GABA procedure, we were unable to acquire the necessary number of double-labeled synapses to perform statistical analyses. Thus, only a qualitative description of the results from these experiments is provided. 3 Results Consistent with previous studies, CGRP immunoreactivity is strongly expressed in the mouse CeA [58]. Our light microscopic (LM) observations confirmed that there is profuse CGRP neuropil immunostaining found in both the mouse (Figure 1A) and monkey right CeLC (Figure 1B). In keeping with recent findings [12], our LM observations confirm close overlap between GluD1 and CGRP immunoreactivity in the mouse CeLC (Figure 1C). Our findings further demonstrate that this co-registration is also seen in the monkey CeLC (Figure 1D). Given well-documented evidence for pain lateralization to the right CeA [2, 5, 24], all experiments in this study were conducted using tissue from the right hemisphere. 3.1 Ultrastructural features of CGRP+ terminals in the CeLC At the electron microscopic level, CGRP+ terminals varied in size, ranging from ~ 0.25 µm to over 1 µm in cross-sectional diameter, with the presence of mitochondria in some boutons (Figures 2D & E). When somata were targeted by CGRP+ terminals, we observed multiple CGRP+ inputs, whereas other somata were devoid of CGRP innervation (Figure 2G). Most CGRP+ terminals contained small, spherical or oval, electro-lucent vesicles as well as large dense core vesicles (DCVs). The density of these vesicles varied between terminals; some CGRP+ terminals contained many electron-lucent vesicles and few DCVs (Figure 2C), while others contained many DCVs and few electron-lucent vesicles (Figure 3H). In tissue from which CGRP was localized with pre-embedding immunogold, strong gold labeling was found around these DCVs (white arrowheads, Figure 3H). We did not observe notable patterns of DCV distribution within CGRP+ terminals in relation to synaptic active zones. Furthermore, there were no notable differences in the morphology of CGRP+ terminals between mice and monkeys. 3.2 Post-synaptic targets of CGRP+ terminals in the CeLC To determine the post-synaptic targets of CGRP+ terminals in the CeLC, blocks of immunoperoxidase-stained tissue from 3 mice and 3 monkeys were examined in the electron microscope (EM). We observed strong CGRP immunoreactivity in presynaptic terminals that formed axo-dendritic (Figures 2A & E), axo-spinous (Figures 2B & E), or axo-somatic synapses (Figures 2C, F, & G). Analysis of approximately 50 micrographs of randomly selected CGRP-immunoreactive terminals/animal showed that most CGRP+ terminals formed axo-dendritic synapses in both mice (chi-squared test, p <0.001) and monkeys (chi-squared test, p <0.001). In mice, 56.6 ± 3.8% of CGRP-labeled terminals formed axo-dendritic synapses, while in monkeys, 70.1 ± 3.9% of CGRP-labeled terminals formed axo-dendritic synapses (Figure 2H). To a lesser extent, CGRP+ terminals also formed axo-spinous (24.4 ± 2.7% in mice and 13.2 ± 5.9% in monkeys) and axo-somatic (19.0 ± 1.9% in mice and 17.9 ± 2.2% in monkeys) synapses (Figure 2H). It is noteworthy that individual somas were often surrounded by several CGRP+ terminals (Figure 2G), a pattern reminiscent of the pericellular CGRP+ innervation of CeLC neurons recently described by Gandhi et al. in mice [12]. The distribution of post-synaptic targets to CGRP+ terminals was not significantly different between species (Figure 2H; unpaired t-test, axo-dendritic, p =0.068; axo-spinous, p =0.161; axo-somatic, p =0.721). 3.3 CGRP+ terminals form symmetric or asymmetric synapses in the CeLC Synapses formed by CGRP+ terminals in the CeLC were categorized as symmetric or asymmetric based on the absence or presence of a dense post-synaptic density, respectively (126). Based on the analysis of 50-100 CGRP+ terminals/animal forming clear synaptic junctions, 55.0 ± 9.8% and 50.9 ± 8.2% were categorized as forming symmetric synapses (black arrowhead, Figures 2C, D, F, & G), while 45.0 ± 9.8% and 49.1 ± 8.2% formed asymmetric synapses (black arrows, Figures 2A, B, & E) in mice and monkeys, respectively. However, the prevalence of either type of synapse varied significantly between post-synaptic targets (Figure 2I – K). For instance, most axo-spinous synapses formed by CGRP+ terminals were asymmetric (75.1 ± 11.6% and 78.6 ± 14.5% in mice and monkeys, respectively) (Figure 2I), whereas most axo-somatic synapses were symmetric (85.6 ± 8.7% and 78.9 ± 2.0% in mice and monkeys, respectively) (Figures 2K). Axo-dendritic synapses were divided evenly between asymmetric (42.7 ± 8.1% and 51.4 ± 8.2% in mice and monkeys, respectively) and symmetric (57.3 ± 8.1% and 50.4 ± 6.8% in mice and monkeys, respectively) (Figure 2J) synapses. There was no significant difference in the overall percentage and prevalence of symmetric versus asymmetric synapses on specific post-synaptic targets formed by CGRP+ terminals in the CeLC between mice and monkeys (unpaired t-test; axo-dendritic: symmetric p=0.548, asymmetric p=0.491; axo-somatic: symmetric p=0.499, asymmetric p=0.499; axo-spinous: symmetric p=0.864, asymmetric p=0.864). 3.4 CGRP+ terminals largely express vGluT2 in the CeLC Although there is strong literature evidence that the PB-CeLC projection is excitatory and glutamatergic [49, 64], the fact that a large proportion of CGRP+ terminals in the CeLC form symmetric synapses, a type of synaptic junction commonly associated with inhibitory transmission, prompted us to confirm that CGRP+ terminals in the mouse and monkey CeLC express vGluT2, the primary glutamatergic marker found in PB neurons [36]. To determine the co-expression of CGRP and vGluT2 immunoreactivity in presynaptic terminals in the CeLC, blocks of double-immunostained (CGRP-vGluT2) tissue from 3 mice and 3 monkeys were examined. In these experiments, CGRP was labeled with immunogold, while immunoperoxidase was used to label vGluT2. Electron micrographs were only collected from superficial ultrathin sections where both CGRP and vGluT2 labeling were co-expressed. A threshold of five or more gold particles was required for a terminal to be considered CGRP-immunoreactive. As expected, we observed frequent co-expression of vGluT2 and CGRP labeling in both the mouse (Figures 3A – C) and monkey (Figures 3D – H) CeLC. However, there were also CGRP+ terminals that were devoid of vGluT2 labeling (Figures 3B & G). Because strong vGluT2 labeling was observed in other terminals in the close vicinity of the vGluT2-negative/CGRP+ terminals (Figure 3G), the lack of vGluT2 labeling in these CGRP+ terminals is unlikely to be the result of antibody penetration issues. Of all CGRP+ terminals examined (n=274 in mice and n=235 in monkeys), 66.7 ± 4.2% and 69.5 ± 3.6% expressed vGluT2 immunoreactivity in the mouse and monkey CeLC, respectively (Figure 3I). There were no significant differences in vGluT2 expression within CGRP-labeled terminals between the mouse and monkey CeLC (Figure 3I; unpaired t-test, p =0.633). Of all CGRP+ terminals that formed clear synaptic junctions (n=63 in mice and n=80 in monkeys), we determined the relative proportion of CGRP+ terminals forming asymmetric and symmetric synapses. As expected, most CGRP+ terminals that formed asymmetric synapses (n=35 in mice and n=37 in monkeys) displayed vGluT2 immunoreactivity (black arrowheads, Figures 3A, B, D, & E). Similarly, of the CGRP+ terminals involved in symmetric axo-dendritic or axo-somatic synapses (n=28 in mice and n=43 in monkeys), 88.9 ± 5.6% and 79.0 ± 1.0% of CGRP+ terminals involving symmetric axo-dendritic and axo-somatic synapses also displayed vGluT2 immunostaining (Figures 3C & F), while 11.1 ± 5.6 % and 21.0 ± 1.0% % were devoid of vGluT2 labeling (Figure 3G) in mice and monkeys, respectively. 3.5 GluD1 expression at synapses formed by CGRP+ terminals in the CeLC To determine the association between CGRP+ terminals and post-synaptic GluD1 localization, blocks of double-immunostained (CGRP-GluD1) tissue from 3 mice and 3 monkeys were examined. In these experiments, CGRP was labeled with immunoperoxidase and GluD1 with immunogold. EM observations were collected from only the most superficial tissue sections where both CGRP and GluD1 labeling was co-expressed. A threshold of two or more gold particles was required to be considered GluD1-immunoreactive. We observed strong GluD1 synaptic labeling within the main body of symmetric axo-dendritic (Figures 4A & D) and axo-somatic synapses (Figures 4C & F) formed by CGRP+ terminals, while peri-synaptic GluD1 labeling was found at the edges of asymmetric axo-dendritic (Figures 4B & E) and axo-spinous synapses. It should also be noted that we observed multiple GluD1-labeled axo-somatic synapses formed by CGRP+ terminals that converged around the same soma (Figure 4G). Of all synapses formed by CGRP-labeled terminals examined in this material, 42.5 ± 2.4% and 51.6 ± 2.4% expressed GluD1 immunoreactivity in the mouse and monkey CeLC, respectively (Figure 4H). Statistical analyses revealed no significant difference in GluD1 expression at CGRP-labeled terminals between the mouse and monkey CeLC (Figure 4H; unpaired t-test, p=0.056). Tables 2 and 3 show the distribution of GluD1 labeling at symmetric or asymmetric synapses formed by CGRP+ terminals on different post-synaptic targets (dendrites, soma, and spines) in mice and monkeys. Table 2. Distribution of GluD1 labeling at synapses formed by CGRP+ terminals in mice CeLC. Mean percentages (± SEM) of GluD1-labeled symmetric and asymmetric synapses formed by different post-synaptic targets. | Symmetric | Asymmetric | Symmetric | Asymmetric | | | Dendrite (n=56) | 69.1 ± 4.8% | 3.2 ± 1.6% | 4.8 ± 4.8% | 23.0 ± 7.9% | | Soma (n=34) | 95.2 ± 2.4% | 2.6 ± 2.6% | - | 2.2 ± 2.2% | | Spine (n=8) | 33.3 ± 33.3% | - | - | 66.7 ± 33.3% | Table 3. Distribution of GluD1 labeling at synapses formed by CGRP+ terminals in monkey CeLC. Mean percentages (± SEM) of GluD1-labeled symmetric and asymmetric synapses formed by different post-synaptic targets. | Symmetric | Asymmetric | Symmetric | Asymmetric | | | Dendrite (n=63) | 74.3 ± 3.0% | 2.9 ± 2.9% | 4.7 ± 2.5% | 18.2 ± 6.9% | | Soma (n=17) | 100.0 ± 0.0% | - | - | - | | Spine (n=6) | 60.0 ± 30.6% | - | - | 6.7 ± 6.7% | 3.6 Lack of GABA expression in CGRP+ terminals Given evidence that a proportion of CGRP+ terminals forming GluD1+ symmetric synapses were vGluT2-negative, we tested the possibility that some of these CGRP+ terminals may be GABAergic. To do so, we used a triple immunostaining approach in which the post-embedding immunogold procedure was used to label GABA in ultrathin sections of CeLC tissue double-immunostained for CGRP and GluD1, as described above. In this tissue, GABA immunoreactivity was depicted by large numbers of 15 nm gold particles in axon terminals, while GluD1 labeling was indicated by larger-sized post-synaptic gold labeling most commonly attached to a synapse (black arrows, Figure 5), and CGRP immunostaining was indicated by immunoperoxidase. Using the quantitative approach described in the methods section to assess the specificity of the GABA labeling, none of the 45 and 24 CGRP+ terminals that also expressed post-synaptic GluD1 localization, in mice and monkeys, respectively, were considered GABA-positive (Figures 5A, B, D, and E). From this material, we also determined whether any CGRP+ terminals displayed GABA immunoreactivity irrespective of their synaptic expression or not of GluD1. Of the 81 and 29 CGRP+ terminals analyzed, 69.9 ± 5.1% and 83.7 ± 2.3% formed clear symmetric synapses in mice and monkeys, respectively. None of the CGRP+ terminals forming symmetric synapses displayed GABA immunoreactivity (Figures 5C & F). To confirm the specificity of the post-embedding immunogold reaction, we also observed the presence of symmetric, putatively GABAergic, synapses formed by GABA-positive terminals (Figure 5H), as well as the presence of asymmetric, putatively glutamatergic, synapses formed by GABA-negative terminals (Figure 5G). 4 Discussion The CeLC is a key region for the integration of nociceptive and emotional-affective processing of pain [13, 48, 51, 66, 70]. Through the spino-parabrachio-amygdaloid pain pathway, the CeLC receives direct monosynaptic inputs from CGRP-containing PB neurons [38, 49, 58, 64]. Pain-related neuroplasticity, such as enhanced transmission and CeLC neuronal excitability, at PB-CeLC synapses has been well studied in acute inflammatory [16, 50, 55] and neuropathic [17, 23] pain models. However, our knowledge of the anatomical substrate through which nociceptive signals are transmitted through PB-CeLC synapses remains limited. In the present study, we used immuno-EM techniques to characterize the ultrastructural features and the pattern of synaptic connections of CGRP+ terminals in the mouse and monkey right CeLC. We observed dense CGRP immunoreactivity throughout the CeLC neuropil. At the electron microscopic level, CGRP+ terminals formed symmetric and asymmetric synapses with dendritic profiles and asymmetric synapses with spines. In line with previous literature [60], numerous large CGRP+ boutons also formed symmetric synapses with cell bodies. Using double and triple immuno-EM approaches, we showed that most of the CGRP+ terminals forming asymmetric axo-spinous and axo-dendritic synapses expressed vGluT2 and were devoid of GABA immunoreactivity, confirming the glutamatergic nature of this projection [49, 64]. We also found a large proportion of vGluT2+/GABA- CGRP+ terminals forming symmetric synapses, while approximately 10% and 20% of CGRP+ terminals forming symmetric synapses were devoid of both vGluT2 and GABA immunoreactivity in mice and monkeys, respectively. Given recent data showing that GluD1 regulates excitatory transmission at PB-CeLC synapses and contributes to pain-related behaviors [12], we further characterized GluD1 expression at synapses formed by CGRP+ terminals in the CeLC nucleus using double immuno-EM methods. In both mice and monkey CeLC tissue, we observed strong post-synaptic GluD1 expression at synapses formed by CGRP+ terminals. Most GluD1 immunogold labeling was found in the core of symmetric synapses or was peri-synaptically localized at the edges of asymmetric CGRP+ synapses. Overall, these findings extend recent data showing the importance of GluD1 in regulating PB-CeLC CGRP+ synapses and further demonstrate that CGRP+ terminals in the CeLC are heterogenous in their ultrastructural and chemical phenotypes, raising the possibility that the PB-CeLC projection originates from different subsets of PB neurons. Our results also demonstrate that the chemical and ultrastructural features of PB-CeLC synapses are maintained between rodents and monkeys, providing a strong foundation to translate animals’ preclinical findings gathered about this nociceptive pathway to the human brain. 4.1 Transmitter content of CGRP+ terminals forming symmetric synapses Our ultrastructural data revealed that there is a heterogenous population of CGRP+ terminals in the mouse and monkey CeLC (Figure 6). That is, most CGRP+ terminals forming axo-spinous synapses were asymmetric (Figure 6A), while most axo-somatic synapses were symmetric (Figures 6C & C’). Moreover, CGRP+ terminals forming axo-dendritic synapses were evenly distributed between asymmetric (Figure 6B) and symmetric (Figures 6B’ & B”) junctions. These results indicate that a subset of CGRP+ terminals in both the mouse and monkey CeLC form symmetric synapses. Given evidence that the PB input to the CeLC is glutamatergic and excitatory [49, 64], we expected most terminals to form asymmetric synapses, which is most commonly associated with glutamatergic synapses [19, 25], whereas symmetric synapses are typically associated with GABAergic synapses [67]. Approximately 80% of CGRP+ axo-somatic synapses and ~ 50% of axo-dendritic synapses were symmetric. However, our double immuno-EM data confirmed that the CGRP+ terminals forming these symmetric synapses are GABA-negative and that a majority exhibit vGluT2 immunoreactivity. Yet it is noteworthy that approximately 15% of CGRP terminals forming symmetric synapses did not express GABA or vGluT2 immunoreactivity. Other EM studies have also reported the appearance of PB-CGRP terminals forming symmetric synapses onto soma and some dendritic shafts in the rodent CeLC [9, 21, 41, 60] as well as in the bed nucleus of the stria terminalis [32, 60]. Our results extend these findings to the monkey CeLC, showing no significant differences in the pattern of synaptic connection of CGRP+ terminals between mice and monkeys. In light of these observations, two important issues remain unanswered: (1) Is CGRP/glutamate signaling mediated in a similar fashion at symmetric versus asymmetric synapses, and (2) Are the pre- and post-synaptic mediators of synaptic transmission the same at CGRP+ terminals that do or do not co-express vGluT2? It has been previously shown that PB-CGRP neurons co-express vGluT2 and other neuropeptides [53]. Perhaps the vGluT2-positive PB-CGRP terminals mainly release glutamate as a primary transmitter, whereas the CGRP+/vGluT2-negative terminals largely function in a peptidergic manner. If such is the case, why do synapses formed by these glutamatergic CGRP+ terminals display a symmetric membrane specialization? A possible explanation may lie in the composition of the post-synaptic density (PSD) associated with the synapses formed by these CGRP+ terminals. The dense PSD of asymmetric, putative glutamatergic, synapses can be attributed to large numbers of receptors (NMDARs, AMPARs, mGluRs), scaffolding proteins (i.e., PSD-95), signaling proteins (i.e., CaMKII), and cytoskeletal components (i.e., actin, calmodulin) that arrange its framework [25, 26, 59, 72]. On the other hand, the thin PSD of symmetric, putative GABAergic, synapses mainly consist of gephyrin, glycine and GABA A receptors, and cytoskeletal elements (i.e., microtubules, actin) [29, 30, 46], suggesting that the lack of a thick post-synaptic specialization may presumably be due to a more simplified PSD architecture. Thus, the subset of CGRP+ terminals forming symmetric synapses may have limited post-synaptic machinery, possibly modified in composition to mediate peptidergic over glutamatergic transmission. Further studies are needed to fully elucidate the composition of the PSDs and determine if such a structural and biochemical difference has an impact upon the properties of these synapses. 4.2 Could CGRP be the primary transmitter released at some PB-CeLC synapses? An interesting observation from our immunolabeling studies is the prominent CGRP localization surrounding large DCVs in PB terminals. It is known that CGRP is stored in DCVs [43, 45], and EM studies have described CGRP localization around large DCVs in the dorsal horn of the spinal cord in monkeys and humans and in frog nerve muscles [4, 18, 43, 44]. Our immunogold data depict a similar pattern of CGRP immunoreactivity in PB-CeLC terminals in both mice and monkey tissues. While it can be said that most CGRP+ terminals examined in this study contained DCVs, the density of such vesicles varied significantly between terminals, such that some terminals were packed with electron-lucent vesicles among which were interspersed a few DCVs, while others were enriched in DCVs with less electron-lucent vesicles. Although this may suggest that PB terminals are heterogenous in their relative content in CGRP-containing vesicles, vesicle counts from single ultrathin sections must be interpreted with caution as they do not provide a complete view of the terminals. Future 3D-EM reconstruction of individual CGRP+ terminals may help address this issue. However, it is difficult to determine whether the number of DCVs is correlated with increased storage or synaptic release of CGRP. Additional studies that directly examine this issue are needed to further relate CGRP+ vesicular content with CGRP release at synaptic sites. Another critical regulator of CGRP-mediated effects in the CeLC is the localization and relative abundance of CGRP receptors in relation to synapses formed by PB-CeLC CGRP+ terminals. There is evidence that CGRP receptor 1 is strongly expressed in the CeA [34, 69], and that CGRP receptor 1 activation in the CeA regulates PKA- and NMDA-mediated synaptic plasticity and pain-related behaviors [16]. Given recent evidence that CGRP itself acts as a peptidergic modulator of threat behavior in the parabrachio-amygdaloid pathway [27], there may be a subpopulation of PB-CGRP+/vGluT2- terminals that mediate some aspects of pain plasticity and pain-related beahviors selectively through peptidergic transmission. 4.3 Interaction between CGRP and GluD1 in the CeLC Though GluD1 exhibits strong mRNA and protein expression in the CeA [20, 31], there is limited knowledge of its subcellular and subsynaptic localization and relationships with specific afferents in this region. In a recent study, Gandhi et. al have shown that GluD1 localizes post-synaptically to CGRP terminals in the soma of CeLC neurons [12]. Our immuno-EM studies confirm these findings, as we observed strong GluD1-CGRP colocalization in both mice (~40%) and monkey (~50%) right CeLC tissues. In addition, we found that most GluD1 immunogold labeling was localized to the main body, or less frequently, at the edges of synapses, formed by CGRP+ terminals, further strengthening the foundation that GluD1 is involved in synaptic signaling at CGRP-CeLC synapses [12]. Interestingly, the CGRP+/GluD1+ synapses were either asymmetric or symmetric based on the thickening of their post-synaptic densities Although GluD1 has been shown to localize post-synaptically at symmetric synapses associated with inhibitory transmission in other brain regions [6, 11, 14], our triple immuno-EM data showed that none of the CGRP+ terminals that also expressed post-synaptic GluD1 labeling displayed GABA immunoreactivity. Moreover, studies have shown that disruption of GluD1 in the CeLC leads to a reduction in mEPSC frequency and amplitude but no changes in mIPSCs [12]. Thus, while we confirmed that GluD1 is not involved in inhibitory transmission at PB-CeLC CGRP+ terminals, the unexpected localization of GluD1 at symmetric synapses questions whether GluD1 only contributes to excitatory transmission in the CeLC. Furthermore, given that there is reduced GluD1 expression in both inflammatory and neuropathic pain models [12], it would be prudent to determine how these pain states affect GluD1 expression as well as its localization with CGRP at the ultrastructural level. Studies exploring morphometric measurements that assess changes in strength and physiological properties of CGRP-GluD1 in different pain states would further contribute to our understanding of GluD1’s role in pain processing in the CeLC. Moreover, a recent study has also shown that in the dorsal horn of the spinal cord, GluD1 is localized post-synaptically to CGRP+ terminals from sensory neurons [57]. The observed downregulation of spinal cord GluD1 expression in inflammatory pain models [57] demonstrates another potential site of GluD1-Cbln1 interaction for mediating pain states in the spino-parabrachio-amygdaloid pain pathway. 5 Concluding Remarks Synaptic transmission and plasticity at PB-CeLC synapses is well-studied, however our understanding of the anatomical substrate through which nociceptive signals are transmitted through PB-CeLC synapses remains limited. In this study, we characterized the ultrastructural features and synaptic connections of PB-CGRP+ terminals in both the mouse and monkey CeLC. Our electron microscopic analysis demonstrates that CGRP+ terminals form symmetric and asymmetric synapses with dendrites, symmetric synapses with soma, and asymmetric synapses with spines in the CeLC. Double immuno-EM studies confirmed that the majority of CGRP+ terminals are vGluT2-positive and devoid of GABA immunoreactivity. Furthermore, our data demonstrate strong post-synaptic GluD1 localization in the core of symmetric synapses and at the edges of asymmetric synapses formed by CGRP+ terminals, extending recent evidence that GluD1 regulates CGRP+PB-CeLC synapses. Altogether, these findings extend knowledge of the anatomical substrates that mediate transmission of nociceptive signals through the parabrachio-amygdaloid pathway and demonstrate that the synaptic organization of this network is comparable between mice and monkeys, increasing the translational value of these findings to the human brain. 6 References 1. Adke, A.P., et al., Cell-Type Specificity of Neuronal Excitability and Morphology in the Central Amygdala. eNeuro, 2021. 8 (1).2. Allen, H.N., et al., A Parabrachial-to-Amygdala Circuit That Determines Hemispheric Lateralization of Somatosensory Processing. Biol Psychiatry, 2023. 93 (4): p. 370-381.3. Bernard, J.F. and J.M. Besson, The spino(trigemino)pontoamygdaloid pathway: electrophysiological evidence for an involvement in pain processes. J Neurophysiol, 1990. 63 (3): p. 473-90.4. Carlton, S.M., et al., A light and electron microscopic level analysis of calcitonin gene-related peptide (CGRP) in the spinal cord of the primate: An immunohistochemical study. Neurosci Lett, 1987. 82 (2): p. 145-150.5. Carrasquillo, Y. and R.W.t. Gereau, Hemispheric lateralization of a molecular signal for pain modulation in the amygdala. Mol Pain, 2008. 4 : p. 24.6. Choi, D., et al., Ultrastructural Localization of Glutamate Delta Receptor 1 in the Rodent and Primate Lateral Habenula. J Comp Neurol 2025. 533 (1): p. e70019.7. D’Hanis, W., R. Linke, and D.M. Yilmazer-Hanke, Topography of thalamic and parabrachial calcitonin gene-related peptide (CGRP) immunoreactive neurons projecting to subnuclei of the amygdala and extended amygdala. J Comp Neurol, 2007. 505 (3): p. 268-91.8. Dobolyi, A., et al., Calcitonin gene-related peptide-containing pathways in the rat forebrain. J Comp Neurol, 2005. 489 (1): p. 92-119.9. Dong, Y.L., et al., Differential postsynaptic compartments in the laterocapsular division of the central nucleus of amygdala for afferents from the parabrachial nucleus and the basolateral nucleus in the rat. J Comp Neurol, 2010. 518 (23): p. 4771-91.10. Elegheert, J., et al., Structural basis for integration of GluD receptors within synaptic organizer complexes. Science, 2016. 353 (6296): p. 295-9.11. Fossati, M., et al., Trans-Synaptic Signaling through the Glutamate Receptor Delta-1 Mediates Inhibitory Synapse Formation in Cortical Pyramidal Neurons. Neuron, 2019. 104 (6): p. 1081-1094.e7.12. Gandhi, P.J., et al., Dysfunction of Glutamate Delta-1 Receptor-Cerebellin 1 Trans-Synaptic Signaling in the Central Amygdala in Chronic Pain. Cells, 2021. 10 (10): p. 2644.13. Gauriau, C. and J.F. Bernard, Pain pathways and parabrachial circuits in the rat. Exp Physiol, 2002. 87 (2): p. 251-8.14. Gawande, D.Y., et al., Glutamate Delta-1 Receptor Regulates Inhibitory Neurotransmission in the Nucleus Accumbens Core and Anxiety-Like Behaviors. Mol Neurobiol, 2021. 58 (10): p. 4787-4801.15. Han, J.S., et al., Facilitation of synaptic transmission and pain responses by CGRP in the amygdala of normal rats. Mol Pain, 2010. 6 : p. 10.16. Han, J.S., W. Li, and V. Neugebauer, Critical role of calcitonin gene-related peptide 1 receptors in the amygdala in synaptic plasticity and pain behavior. J Neurosci, 2005. 25 (46): p. 10717-28.17. Han, J.S. and V. Neugebauer, Synaptic plasticity in the amygdala in a visceral pain model in rats. Neurosci Lett, 2004. 361 (1-3): p. 254-7.18. Harmann, P.A., et al., Calcitonin gene-related peptide (CGRP) in the human spinal cord: A light and electron microscopic analysis. J Comp Neurol 1988. 269 (3): p. 371-380.19. Hassel, B. and R. Dingledine, Chapter 17 - Glutamate and Glutamate Receptors, in Basic Neurochemistry (Eighth Edition), S.T. Brady, et al., Editors. 2012, Academic Press: New York. p. 342-366.20. Hepp, R., et al., Glutamate receptors of the delta family are widely expressed in the adult brain. Brain Struct Funct, 2015. 220 (5): p. 2797-2815.21. Honkaniemi, J., et al., Simultaneous localization of calcitonin gene-related peptide and neurotensin in rat central amygdaloid nucleus. Neurosci Lett, 1990. 113 (1): p. 1-6.22. Hoover, A.H., et al., Ultrastructural localization of glutamate delta 1 (GluD1) receptor immunoreactivity in the mouse and monkey striatum. J Comp Neurol 2021. 529 (7): p. 1703-1718.23. Ikeda, R., et al., NMDA receptor-independent synaptic plasticity in the central amygdala in the rat model of neuropathic pain. Pain, 2007. 127 (1-2): p. 161-72.24. Ji, G. and V. Neugebauer, Hemispheric lateralization of pain processing by amygdala neurons. J Neurophysiol, 2009. 102 (4): p. 2253-64.25. Kennedy, M.B., The postsynaptic density at glutamatergic synapses. Trends in Neurosciences, 1997. 20 (6): p. 264-268.26. Kennedy, M.B., Signal-Processing Machines at the Postsynaptic Density. Science, 2000. 290 (5492): p. 750-754.27. Kim, D.-I., et al., Presynaptic sensor and silencer of peptidergic transmission reveal neuropeptides as primary transmitters in pontine fear circuit. Cell, 2024. 187 (18): p. 5102-5117.e16.28. Kim, J., et al., Basolateral to Central Amygdala Neural Circuits for Appetitive Behaviors. Neuron, 2017. 93 (6): p. 1464-1479.e5.29. Kirsch, J. and H. Betz, The postsynaptic localization of the glycine receptor-associated protein gephyrin is regulated by the cytoskeleton. J Neurosci, 1995. 15 (6): p. 4148-56.30. Kneussel, M. and H. Betz, Receptors, gephyrin and gephyrin-associated proteins: novel insights into the assembly of inhibitory postsynaptic membrane specializations. J Physiol, 2000. 525 Pt 1 (Pt 1): p. 1-9.31. Konno, K., et al., Enriched expression of GluD1 in higher brain regions and its involvement in parallel fiber-interneuron synapse formation in the cerebellum. J Neurosci, 2014. 34 (22): p. 7412-24.32. Kozicz, T. and A. Arimura, Axon terminals containing CGRP-immunoreactivity form synapses with CRF- and Met-enkephalin-immunopositive neurons in the laterodorsal division of the bed nucleus of the stria terminalis in the rat. Brain Res, 2001. 893 (1): p. 11-20.33. Kristensen, A.S., et al., Pharmacology and Structural Analysis of Ligand Binding to the Orthosteric Site of Glutamate-Like GluD2 Receptors. Mol Pharmacol, 2016. 89 (2): p. 253-62.34. Kruger, L., et al., Calcitonin gene-related peptide (CGRP) in the rat central nervous system: patterns of immunoreactivity and receptor binding sites. Brain Res, 1988. 463 (2): p. 223-44.35. Kusnoor, S.V., et al., Extracerebellar role for Cerebellin1: modulation of dendritic spine density and synapses in striatal medium spiny neurons. J Comp Neurol, 2010. 518 (13): p. 2525-37.36. Lein, E.S., et al., Genome-wide atlas of gene expression in the adult mouse brain. Nature, 2007. 445 (7124): p. 168-176.37. Li, H., et al., Experience-dependent modification of a central amygdala fear circuit. Nat Neurosci, 2013. 16 (3): p. 332-9.38. Li, J.N. and P.L. Sheets, Spared nerve injury differentially alters parabrachial monosynaptic excitatory inputs to molecularly specific neurons in distinct subregions of the central amygdala. Pain, 2020. 161 (1): p. 166-176.39. Liu, J., et al., Striatal glutamate delta-1 receptor regulates behavioral flexibility and thalamostriatal connectivity. Neurobiol Dis, 2020. 137 : p. 104746-104746.40. Lomeli, H., et al., The rat delta-1 and delta-2 subunits extend the excitatory amino acid receptor family. FEBS Letters, 1993. 315 (3): p. 318-322.41. Lu, Y.-C., et al., Neurochemical properties of the synapses between the parabrachial nucleus-derived CGRP-positive axonal terminals and the GABAergic neurons in the lateral capsular division of central nucleus of amygdala. Mol Neurobiol, 2015. 51 (1): p. 105-118.42. Matsuda, K., et al., Cbln1 is a ligand for an orphan glutamate receptor delta2, a bidirectional synapse organizer. Science, 2010. 328 (5976): p. 363-8.43. Matteoli, M., et al., Differential effect of alpha-latrotoxin on exocytosis from small synaptic vesicles and from large dense-core vesicles containing calcitonin gene-related peptide at the frog neuromuscular junction. Proc Natl Acad Sci U S A, 1988. 85 (19): p. 7366-70.44. Merighi, A., Costorage and coexistence of neuropeptides in the mammalian CNS. Prog Neurobiol, 2002. 66 (3): p. 161-190.45. Merighi, A., Costorage of High Molecular Weight Neurotransmitters in Large Dense Core Vesicles of Mammalian Neurons. Front Cell Neurosci, 2018. 12 : p. 272.46. Meyer, G., et al., Identification of a gephyrin binding motif on the glycine receptor beta subunit. Neuron, 1995. 15 (3): p. 563-72.47. Miura, E., et al., Distinct expression of Cbln family mRNAs in developing and adult mouse brains. Eur J Neurosci, 2006. 24 (3): p. 750-760.48. Neugebauer, V., Amygdala pain mechanisms. Handb Exp Pharmacol, 2015. 227 : p. 261-84.49. Neugebauer, V. and W. Li, Processing of nociceptive mechanical and thermal information in central amygdala neurons with knee-joint input. J Neurophysiol, 2002. 87 (1): p. 103-12.50. Neugebauer, V., et al., Synaptic plasticity in the amygdala in a model of arthritic pain: differential roles of metabotropic glutamate receptors 1 and 5. J Neurosci, 2003. 23 (1): p. 52-63.51. Neugebauer, V., et al., The amygdala and persistent pain. Neuroscientist, 2004. 10 (3): p. 221-34.52. Otsuka, S., et al., Roles of Cbln1 in Non-Motor Functions of Mice. J Neurosci, 2016. 36 (46): p. 11801-11816.53. Pauli, J.L., et al., Molecular and anatomical characterization of parabrachial neurons and their axonal projections. eLife, 2022. 11 : p. e81868.54. Peters, A., P.S. Palay. SL, and H.d. Webster, The fine structure of the nervous system: the neurons and supporting cells (3rd ed.) . 3rd ed. 1991: Oxford University Press.55. Ren, W. and V. Neugebauer, Pain-related increase of excitatory transmission and decrease of inhibitory transmission in the central nucleus of the amygdala are mediated by mGluR1. Mol Pain, 2010. 6 : p. 93.56. S. Michaela Rikard, A.E.S., Kristine M. Schmit, Gery P. Guy Jr., Chronic Pain Among Adult - United States, 2019-2021. MMWR Morb Mortal Wkly Rep 2023, 2023. 72 : p. 379-385.57. Sabnis, S.S., et al., Intravenous recombinant cerebellin 1 treatment restores signalling by spinal glutamate delta 1 receptors and mitigates chronic pain. Br J Pharmacol, 2024. 181 (9): p. 1421-1437.58. Schwaber, J.S., et al., Neurons containing calcitonin gene-related peptide in the parabrachial nucleus project to the central nucleus of the amygdala. J Comp Neurol, 1988. 270 (3): p. 416-26, 398-9.59. Sheng, M. and C.C. Hoogenraad, The Postsynaptic Architecture of Excitatory Synapses: A More Quantitative View. Annual Review of Biochemistry, 2007. 76 (Volume 76, 2007): p. 823-847.60. Shimada, S., et al., Light and electron microscopic studies of calcitonin gene-related peptide-like immunoreactive terminals in the central nucleus of the amygdala and the bed nucleus of the stria terminalis of the rat. Exp Brain Res, 1989. 77 (1): p. 217-220.61. Shimada, S., et al., Calcitonin gene-related peptidergic projection from the parabrachial area to the forebrain and diencephalon in the rat: an immunohistochemical analysis. Neuroscience, 1985. 16 (3): p. 607-16.62. Shinohara, K., et al., Essential role of endogenous calcitonin gene-related peptide in pain-associated plasticity in the central amygdala. Eur J Neurosci, 2017. 46 (6): p. 2149-2160.63. Spike, R.C., et al., A quantitative and morphological study of projection neurons in lamina I of the rat lumbar spinal cord. Eur J Neurosci, 2003. 18 (9): p. 2433-48.64. Sugimura, Y.K., et al., Synaptic and network consequences of monosynaptic nociceptive inputs of parabrachial nucleus origin in the central amygdala. J Neurophysiol, 2016. 115 (6): p. 2721-39.65. Tao, W., et al., Postsynaptic δ1 glutamate receptor assembles and maintains hippocampal synapses via Cbln2 and neurexin. Proc Natl Acad Sci U S A, 2018. 115 (23): p. E5373-E5381.66. Thompson, J.M. and V. Neugebauer, Amygdala Plasticity and Pain. Pain Res Manag, 2017. 2017 : p. 8296501.67. Tretter, V. and S.J. Moss, GABA(A) Receptor Dynamics and Constructing GABAergic Synapses. Front Mol Neurosci, 2008. 1 : p. 7.68. Uemura, T., et al., Trans-synaptic interaction of GluRdelta2 and Neurexin through Cbln1 mediates synapse formation in the cerebellum. Cell, 2010. 141 (6): p. 1068-79.69. van Rossum, D., U.K. Hanisch, and R. Quirion, Neuroanatomical localization, pharmacological characterization and functions of CGRP, related peptides and their receptors. Neurosci Biobehav Rev, 1997. 21 (5): p. 649-78.70. Veinante, P., I. Yalcin, and M. Barrot, The amygdala between sensation and affect: a role in pain. J Mol Psychiatry, 2013. 1 (1): p. 9.71. Wilson, T.D., et al., Dual and Opposing Functions of the Central Amygdala in the Modulation of Pain. Cell Rep, 2019. 29 (2): p. 332-346.e5.72. Ziff, E.B., Enlightening the Postsynaptic Density. Neuron, 1997. 19 (6): p. 1163-1174. Figure Legends. Figure 1. CGRP and GluD1 expression in the mouse and monkey right CeA. Light micrographs of CGRP and GluD1 immunoperoxidase staining in the right CeA of a mouse ( A, C ) and monkey ( B, D ). There is close overlap of CGRP and GluD1 expression in the mouse ( A & B ) and monkey ( C & D ) right CeA. Abbreviations : CeA: central amygdala, Th: thalamus, GPi: globus pallidus internus, OT: optic tract. Scale bar in A, C = 500 µm. Scale bar in B, D = 1mm. Figure 2. CGRP expression in the mouse and monkey CeLC. Examples of CGRP-immunoreactive terminals in mice ( A – C ) and monkey ( D – G ) right CeLC. CGRP+ terminals form axo-dendritic ( A & D ), axo-spinous ( B & E ), and axo-somatic ( C & F ) synapses. CGRP+ terminals form both asymmetric (black arrows) and symmetric (black arrowheads) synapses. ( G ) Note that many CGRP+ terminals often converged around single soma of CeLC neurons. ( H ). Graph demonstrates mean percentages (± SEM) of CGRP+ terminals that form axo-spinous, axo-dendritic, and axo-somatic synapses from 3 mice and 3 monkeys each. Most CGRP+ terminals formed axo-dendritic synapses in both mice (chi-squared test, p<0.001) and monkeys (chi-squared test, p<0.001). The distribution of each synapse type at CGRP+ terminals was conserved across species (unpaired t-test, axo-spinous, p=0.161; axo-dendritic, p=0.068; axo-somatic, p=0.721). ( I – K ) Graphs show mean percentages (± SEM) of CGRP+ terminals that form symmetric or asymmetric axo-spinous ( I ), axo-dendritic ( J ), or axo-somatic ( K ) synapses, from 3 mice and 3 monkeys each. ( I ) Most axo-spinous synapses formed by CGRP+ terminals were asymmetric. ( J ) CGRP+ axo-dendritic synapses were evenly distributed between symmetric and asymmetric synapses. ( K ) Most axo-somatic synapses formed by CGRP+ terminals were symmetric. The distribution of CGRP+ terminals forming symmetric and asymmetric synapses for each post-synaptic target was conserved across species (unpaired t-test; axo-spinous, symmetric: p=0.864, asymmetric: p=0.864; axo-dendritic, symmetric: p=0.548, asymmetric: p=0.491; axo-somatic, symmetric: p=0.499, asymmetric: p=0.499). Abbreviations: De: dendrite, Sp: spine, m: mitochondria. Scale bar in A (applies B, D, E) = 0.5 µm. Scale bar in C (applies F) = 0.5 µm. Scale bar in G = 0.5 µm. Figure 3. CGRP+ terminals express vGluT2 in the CeLC. Micrographs of double-immunostained CGRP (immunogold) and vGluT2 (peroxidase) mouse ( A – C ) and monkey ( D – H ) right CeLC tissue. CGRP-labeled terminals displaying vGluT2 immunoreactivity are indicated by a white star. Many of these CGRP-vGluT2 terminals form asymmetric synapses (black arrowheads; A, B, D, & E ). Some double-labeled CGRP-vGluT2 ( C & F ) and single-labeled CGRP+ ( G ) terminals form symmetric synapses (black arrows). ( H ) Strong CGRP immunogold labeling surrounds dense core vesicles (white arrowheads). ( I ) Graph demonstrates mean percentages (± SEM) of CGRP+ terminals that are labeled with vGluT2 from 3 mice and 3 monkeys each. Of all CGRP-labeled terminals, 66.7 ± 4.2% and 69.5 ± 3.6% expressed vGluT2 immunoreactivity in the mouse and monkey CeLC, respectively. There were no significant differences in vGluT2 expression within CGRP-labeled terminals between the mouse and monkey CeLC (unpaired t-test, p =0.560). Abbreviations : De: dendrite, Sp: spine. Scale bar in A (applies B – H) = 0.5 µm. Figure 4. GluD1 is associated with CGRP+ terminals in the CeLC. Micrographs of double-immunostained GluD1 (immunogold) and CGRP (peroxidase) mouse ( A – C ) and monkey ( D – F ) right CeLC tissue. CGRP+ terminals in contact with synaptic ( A, C, D, & F ) and peri-synaptic ( B & E ) GluD1 labeling. Synaptic GluD1 labeling is observed at symmetric synapses (black arrows) in both dendrites ( A & D ) and soma ( C & F ). Peri-synaptic GluD1 labeling is shown at asymmetric synapses (black arrowheads) in dendrites ( B & E ). ( G ) Multiple CGRP/GluD1-positive synapses converging around the same soma in CeLC neurons. ( H ) Graph demonstrates mean percentages (± SEM) of synapses formed by CGRP+ terminals that are labeled with GluD1 from 3 mice and 3 monkeys each. Of all CGRP-labeled terminals, 42.5 ± 2.4% and 51.6 ± 2.4% expressed GluD1 immunoreactivity in the mouse and monkey CeLC, respectively. There was no significance difference in GluD1 expression at CGRP+ terminals between the mouse and monkey CeA (unpaired t-test, p =0.056). Abbreviations : De: dendrite. Scale bar in A (applies B-F) = 0.5 µm. Scale bar in G = 2 µm. Figure 5. Post-embedding GABA in the mouse and monkey CeLC. Examples of CGRP (immunoperoxidase), GluD1 (~25-35nm immunogold), and GABA (15nm immunogold) synaptic co-immunolabeling in ( A – C ) and monkey ( D – F ) right CeLC. ( A, B, D, & E ) Electron micrographs of GABA-negative CGRP+ terminals forming symmetric synapses (black arrows) labeled for GluD1. ( C & F ) Micrographs of GABA-negative CGRP+ terminals forming symmetric synapses that are not labeled for GluD1. ( G ) Example of a GABA-negative terminal forming an asymmetric synapse (black arrowhead) in mouse CeLC. ( H ) Example of a GABA-positive terminal forming a symmetric synapse in monkey CeLC. Abbreviation : De: dendrite. Scale bar in A (applies B – F) = 0.5 µm. Figure 6. Schematic summary detailing the heterogenous population of CGRP+ terminals in the mouse and monkey CeLC. ( A ) Most CGRP+ terminals that formed axo-spinous synapses were asymmetric (defined by a thick post-synaptic density, dark blue). Most of these terminals expressed vGluT2 immunoreactivity. ( B ) Axo-dendritic contacts from CGRP+ terminals were divided evenly between asymmetric ( B ) and symmetric ( B’, B’ ’ ) synapses. Most axo-dendritic symmetric synapses expressed vGluT2 immunoreactivity ( B’ ), though a small proportion only expressed CGRP ( B’ ’ ). ( C ) Most CGRP+ terminals that formed axo-somatic synapses were symmetric. Most of these symmetric synapses expressed vGluT2 immunoreactivity ( C ) and a small proportion only expressed CGRP ( C’ ). Created in https://BioRender.com. Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License.

Authors Metrics & Citations Metrics Article Usage 343views 199downloads Citations Download citation Diane Choi, Karina Dalal, Chris Kozuch, et al. Subsynaptic expression of glutamate delta receptor 1 at calcitonin gene-related peptide-containing projections from the parabrachial region in the central amygdala of mice and monkeys. Authorea. 17 May 2025. DOI: https://doi.org/10.22541/au.174745090.02304514/v1 DOI: https://doi.org/10.22541/au.174745090.02304514/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.