Pancreatic amylin dynamically reconfigures distributed brain networks governing appetite regulation in mice

Amylin, also known as islet amyloid polypeptide, is co-secreted with insulin by pancreatic β-cells in response to nutrient stimuli. Amylin reduces food intake and body weight and also acts as adiposity signal to control energy expenditure (EE). Circulating amylin acts centrally by primarily activating neurons of the area postrema (AP), a circumventricular organ (CVO) located in the hindbrain. Amylin receptors (AMYs) are dimers of the calcitonin-like receptor isoform a (CTRa) and of one member of the receptor activity modifying proteins (RAMPs). The presence of CTRa and the RAMPs has been described in several brain areas, including the AP. Amylin synergistically interacts with the adipokine leptin to reduce food intake and body weight. Although, brain areas where this interaction occurs have not been identified yet, the AP recently emerged as a good candidate. However, evidence of co-expression of all the subunits of the AMYs and leptin receptors (LepRb) in individual AP-neurons is still lacking. Therefore we applied a combination of laser capture microdissection and single-cell qPCR to investigate the co-expression of AMYs components and LepRb in individual AP-neurons. Our results demonstrated that CTRa and one or more RAMPs transcripts are co-expressed in single AP- neurons. Moreover, acute amylin treatment differentially regulates its own receptor: while CTRa is un-affected, RAMP1 and RAMP3 mRNAs are consistently down-regulated. On the contrary, RAMP2 mRNA is up-regulated. Furthermore, 30% of amylin-activated AP-neurons, which co-express all the transcripts necessary to generate a functional AMY, also co-express LepRb mRNA. Interestingly, these results outline the possibility that more than one RAMP can be co-expressed with CTRa, raising the question whether different CTRa/RAMPs might mediate different physiological effects in the feeding behavior. Indeed, not only does RAMP1 generate an AMY1 by coupling with CTR, but it also has the potential to bind to the calcitonin-like receptor and generate a calcitonin-gene-related peptide receptor (CGRP). Since both amylin and CGRP decrease food intake and body weight we speculate that RAMP1 might exert a critical role in the regulation of energy balance. In fact, transgenic mice expressing the human (h) RAMP1 (Nestin/hRAMP1) on a regular diet are characterized by a markedly lean phenotype. This suggests that RAMP1 might exert a protective effect against obesity. To further validate this hypothesis, we challenged the transgenic mice with a high-fat diet (HFD). Our findings revealed that the lean phenotype is maintained on HFD and the reduction in body weight results from increased energy expenditure rather than from a reduction in food intake. Hence, RAMP1 appears to be involved in the regulation of energy balance. Finally, CVOs, such as the AP, have emerged as new neurogenic niches in the adult brain, and amylin has been recently shown to exert a neurotrophic effect. Therefore, we combined transcriptome analysis and immunohistochemical techniques to investigated acute and chronic effects of amylin treatment on the AP of adult rats. Our results showed that acute amylin regulates genes involved in multiple pathways and processes of adult neurogenesis. Moreover, chronic amylin infusion increases the number of newly proliferating cells in the adult AP, and determines the fate-commitment of these adult-born cells into mature neurons rather than glia. In summary, we demonstrated that all the AMY subunits and the LepRb are co-expressed in single AP-neurons. Acute amylin administration differentially regulates the transcripts of the RAMPs and the LepRb in the AP. Additionally, acute amylin affects genes involved in the regulation of multiple pathways and processes of adult neurogenesis. Chronic amylin treatment increases the number of newly proliferating cells in the AP and determines their neuronal fate over glia. Thus, amylin, in addition to its role as a satiating hormone, plays a novel role in the regulation of neurogenic processes in the AP of adult rats.

Amylin brain circuitry

SUMMARYNausea is a discomforting sensation of gut malaise that remains a major clinical challenge. Several visceral poisons induce nausea through the area postrema, a sensory circumventricular organ that detects blood-borne factors. Here, we use genetic approaches based on an area postrema cell atlas to reveal inhibitory neurons that counteract nausea-associated poison responses. The gut hormone glucose insulinotropic peptide (GIP) activates area postrema inhibitory neurons that project locally and elicit inhibitory currents in nausea-promoting excitatory neurons through γ-aminobutyric acid (GABA) receptors. Moreover, GIP blocks behavioral responses to poisons in wild-type mice, with protection eliminated by targeted area postrema neuron ablation. These findings provide insights into the basic organization of nausea-associated brainstem circuits and reveal that area postrema inhibitory neurons are an effective pharmacological target for nausea intervention.

The aim of this study was to elucidate the influence of receptor activity modifying protein 1 (RAMP1) overexpression on the expression and distribution of calcitonin receptor-like receptor (CRLR) in MG-63 cells. Our research also focused on whether RAMP1 overexpression enhanced the promoting effect of exogenous CGRP on osteogenic differentiation in MG-63 cells. We first constructed a eukaryotic expression vector containing human RAMP1 and stably transfected it into MG-63 cells. Real-time PCR and Western blotting were used to determine the expression levels of RAMP1 and CRLR mRNA and protein, respectively. Immunofluorescence analysis was employed to compare the distribution of CRLR in transfected cells. After treatment with CGRP, the extent of osteogenic differentiation was evaluated by simultaneous monitoring of alkaline phosphatase activity, the expression patterns of osteoblastic markers and mineralisation staining. We found that RAMP1 was more highly expressed in the transfected group compared with the control groups (P < 0.01). The CRLR expression was significantly higher than that in the control groups (P < 0.05). In addition, after 7 days of CGRP treatment to induce osteogenic differentiation, the expression of collagen I mRNA was markedly increased in the transfected group (P < 0.05). The transfected group exhibited more granular precipitation in the cytoplasm with alkaline phosphatase staining after 7 and 14 days of differentiation. When stained with Alizarin Red, cells overexpressing RAMP1 were darker and formed many mineralised nodules with clear boundaries and calcium deposition typical of mineralised bone matrix structures at 28 days post-induction of differentiation. The CGRP-induced ALP activity in the RAMP1 overexpression group was significantly higher 3, 6 and 9 days after induction than that in the two control groups (P < 0.05). RAMP1 overexpression promotes CRLR expression, localisation on the cell membrane and enhanced CGRP-mediated differentiation of MG-63 cells. This study contributes to a better understanding of the molecular mechanisms governing CGRP-induced MG-63 differentiation.

The pancreatic hormone amylin plays a central role in regulating energy homeostasis and glycaemic control by stimulating satiation and reducing food reward, making amylin receptor agonists attractive for the treatment of metabolic diseases. Amylin receptors consist of heterodimerized complexes of the calcitonin receptor and receptor-activity modifying proteins subtype 1-3 (RAMP1-3). Neuronal activation in response to amylin dosing has been well characterized, but only in selected regions expressing high levels of RAMPs. The current study identifies global brain-wide changes in response to amylin and by comparing wild type and RAMP1/3 knockout mice reveals the importance of RAMP1/3 in mediating this response. Amylin dosing resulted in neuronal activation as measured by an increase in c-Fos labelled cells in 20 brain regions, altogether making up the circuitry of neuronal appetite regulation (e.g., area postrema (AP), nucleus of the solitary tract (NTS), parabrachial nucleus (PB), and central amygdala (CEA)). c-Fos response was also detected in distinct nuclei across the brain that typically have not been linked with amylin signalling. In RAMP1/3 knockout amylin induced low-level neuronal activation in seven regions, including the AP, NTS and PB, indicating the existence of RAMP1/3-independent mechanisms of amylin response. Under basal conditions RAMP1/3 knockout mice show reduced neuronal activity in the hippocampal formation as well as reduced hippocampal volume, suggesting a role for RAMP1/3 in hippocampal physiology and maintenance. Altogether these data provide a global map of amylin response in the mouse brain and establishes the significance of RAMP1/3 receptors in relaying this response.

Viral depletion of calcitonin receptors in the area postrema: A proof-of-concept study

Evidence from human and animal studies indicates that mechanical loads to breathing are stressful stimuli and evoke compensatory behaviours. Conditioning of stressful stimuli is known to cause changes in basal stress levels and behaviour. Individuals with respiratory obstructive diseases repeatedly experience bouts of airway obstruction, which may act as a form of conditioning, and often have affective disorders, such as anxiety and depression. It is unknown whether the development of affective disorders in these individuals results from the unexpected recurring respiratory perturbations. To investigate this possibility, we developed a model to elicit tracheal occlusion (TO) in conscious rats and exposed them to 10 days of TO conditioning. We hypothesized that healthy, conscious animals exposed to TO conditioning would develop stress and anxiety and would have modulated neural activity in respiratory, stress, discriminative and affective neural regions. Following TO conditioning, rats had increased basal corticosterone levels, greater adrenal weights and elevated anxiety levels compared with animals not receiving TO. Significant increases in cytochrome oxidase staining were found in brainstem respiratory nuclei, periaqueductal grey, dorsal raphe, thalamus and insular cortex. These results suggest that healthy animals develop stress and anxiety responses to respiratory load conditioning via inescapable tracheal occlusions, which may be mediated through state changes in specific brain nuclei.

The LPBN (lateral parabrachial nucleus) plays an important role in feeding control. CGRP (calcitonin gene-related peptide) LPBN neurons activation mediates the anorectic effects of different gut-derived peptides, including amylin. Amylin and its long acting analog sCT (salmon calcitonin) exert their anorectic actions primarily by directly activating neurons located in the area postrema (AP). A large proportion of projections from the AP and the adjacent nucleus of the solitary tractNTS to the LPBN, are noradrenergic (NA), and amylin-activated NAAP neurons are critical in mediating amylin's hypophagic effects. Here, we determine the functional role of NAAP amylin activated neurons to activate CGRP and non-CGRP LPBN neurons. To this end, NA was specifically depleted in the rat LPBN through a stereotaxic microinfusion of 6-OHDA, a neurotoxic agent that destroys NA terminals. While amylin (50μg/kg) and sCT (5μg/kg) reduced eating in sham-lesioned rats, no reduction in feeding occurred in NA-depleted animals. Further, the amylin-induced c-Fos response in the LPBN and c-Fos/CGRP colocalization were reduced in NA-depleted animals compared to controls. We conclude that AP→LPBN NA signaling, through the activation of LPBN CGRP neurons, mediates part of amylin's hypophagic effect.

We have recently reported that an acute intragastric hypertonic saline load increases Fos immunoreactivity in several central nuclei, including the supraoptic nucleus (SON), paraventricular nucleus (PVN), nucleus of the solitary tract (NTS), area postrema (AP), and lateral parabrachial nucleus (LPBN). We have also shown that these responses are mediated by stimulation of peripheral osmoreceptors with splanchnic and vagal afferent projections. However, it is unclear whether the primary projections of peripheral osmoreceptors terminate in the NTS or the AP, both of which project to the SON and PVN. This study tested the hypothesis that efferent projections from the AP were necessary for the Fos responses in the SON, PVN, and LPBN. We examined the effect of AP lesion on the response of central Fos immunoreactivity to intragastric hypertonic saline infusion in conscious rats. Compared with sham-lesioned rats (n = 5), Fos expression in AP-lesioned rats (n = 6) was similar in the SON following the intragastric sodium load. However, in contrast to the sham group, Fos expression was significantly reduced in the PVN of AP-lesioned rats. Fos levels observed in the NTS and LPBN were similar in both groups. These results suggest that the PVN response to intragastric hypertonic saline is dependent on efferent projections from the AP. In contrast, Fos responses to this stimulus in the NTS, SON, and LPBN are independent of the activity of the AP.

The aim of this study was to elucidate the influence of receptor activity modifying protein 1 (RAMP1) overexpression on the expression and distribution of calcitonin receptor-like receptor (CRLR) in MG-63 cells. Our research also focused on whether RAMP1 overexpression enhanced the promoting effect of exogenous CGRP on osteogenic differentiation in MG-63 cells. We first constructed a eukaryotic expression vector containing human RAMP1 and stably transfected it into MG-63 cells. Real-time PCR and Western blotting were used to determine the expression levels of RAMP1 and CRLR mRNA and protein, respectively. Immunofluorescence analysis was employed to compare the distribution of CRLR in transfected cells. After treatment with CGRP, the extent of osteogenic differentiation was evaluated by simultaneous monitoring of alkaline phosphatase activity, the expression patterns of osteoblastic markers and mineralisation staining. We found that RAMP1 was more highly expressed in the transfected group compared with the control groups (P < 0.01). The CRLR expression was significantly higher than that in the control groups (P < 0.05). In addition, after 7 days of CGRP treatment to induce osteogenic differentiation, the expression of collagen I mRNA was markedly increased in the transfected group (P < 0.05). The transfected group exhibited more granular precipitation in the cytoplasm with alkaline phosphatase staining after 7 and 14 days of differentiation. When stained with Alizarin Red, cells overexpressing RAMP1 were darker and formed many mineralised nodules with clear boundaries and calcium deposition typical of mineralised bone matrix structures at 28 days post-induction of differentiation. The CGRP-induced ALP activity in the RAMP1 overexpression group was significantly higher 3, 6 and 9 days after induction than that in the two control groups (P < 0.05). RAMP1 overexpression promotes CRLR expression, localisation on the cell membrane and enhanced CGRP-mediated differentiation of MG-63 cells. This study contributes to a better understanding of the molecular mechanisms governing CGRP-induced MG-63 differentiation.

Body weight lowering effect of glucose-dependent insulinotropic polypeptide and glucagon-like peptide receptor agonists is more efficient in RAMP1/3 KO than in WT mice

Effects of parabrachial stimulation on angiotensin and blood pressure sensitive area postrema neurons

Objectives: The aim of the study was to understand the effects of suckling on the brain of the pups by mapping their brain activation pattern in response to suckling.Methods: The c-fos method was applied to identify activated neurons. Fasted rat pups were returned to their mothers for suckling and sacrificed 2 hours later for Fos immunohistochemistry. Double labeling was also performed to characterize some of the activated neurons. For comparison, another group of fasted pups were given dry food before Fos mapping.Results: After suckling, we found an increase in the number of Fos-immunoreactive neurons in the insular and somatosensory cortices, central amygdaloid nucleus (CAm), paraventricular (PVN) and supraoptic hypothalamic nuclei, lateral parabrachial nucleus (LPB), nucleus of the solitary tract (NTS), and the area postrema. Double labeling experiments demonstrated the activation of calcitonin gene-related peptide-ir (CGRP-ir) neurons in the LPB, corticotropin-releasing hormone-ir (CRH-ir) but not oxytocin-ir neurons in the PVN, and noradrenergic neurons in the NTS. In the CAm, Fos-ir neurons did not contain CRH but were apposed to CGRP-ir fiber terminals. Refeeding with dry food-induced Fos activation in all brain areas activated by suckling. The degree of activation was higher following dry food consumption than suckling in the insular cortex, and lower in the supraoptic nucleus and the NTS. Furthermore, the accumbens, arcuate, and dorsomedial hypothalamic nuclei, and the lateral hypothalamic area, which were not activated by suckling, showed activation by dry food.Discussion: Neurons in a number of brain areas are activated during suckling, and may participate in the signaling of satiety, taste perception, reward, food, and salt balance regulation.

In this study we examined fasted and refed cfos activation in cortical, brainstem, and hypothalamic brain regions associated with appetite regulation. We examined a number of time points during refeeding to gain insight into the temporal pattern of neuronal activation and changes in endocrine parameters associated with fasting and refeeding. In response to refeeding, blood glucose and plasma insulin returned to basal levels within 30 minutes, whereas plasma nonesterified fatty acids and leptin returned to basal levels after 1 and 2 hours, respectively. Within the hypothalamic arcuate nucleus (ARC), fasting increased cfos activation in ∼25% of neuropeptide Y neurons, which was terminated 1 hour after refeeding. Fasting had no effect on cfos activation in pro-opiomelanocortin neurons; however, 1 and 2 hours of refeeding significantly activated ∼20% of ARC pro-opiomelanocortin neurons. Acute refeeding (30, 60, and 120 minutes), but not fasting, increased cfos activation in the nucleus accumbens, the cingulate cortex (but not the insular cortex), the medial and lateral parabrachial nucleus, the nucleus of the solitary tract, the area postrema, the dorsal raphe, and the ventromedial nucleus of the hypothalamus. After 6 hours of refeeding, cfos activity was reduced in the majority of these regions compared with that at earlier time points. Our data indicate that acute refeeding, rather than long-term fasting, activates cortical, brainstem, and hypothalamic neural circuits associated with appetite regulation and reward processing. Although the hypothalamic ARC remains a critical sensory node detecting changes in the metabolic state and feedback during fasting and acute refeeding, our results also reveal the temporal pattern in cfos activation in cortical and brainstem areas implicated in the control of appetite and body weight regulation.

Functional seizures (FS) often disrupt the key regions integral to cognitive processing and emotional regulation (anterior insula, anterior cingulate, and temporoparietal junction). We investigated the potential neurophysiologic mechanism of action (MOA) of neurobehavioral therapy (NBT) using resting-state functional MRI seed-based whole-brain functional connectivity within these regions in adults with FS. We hypothesized that NBT would induce changes in functional connectivity in parallel with improving seizure frequency and behavioral outcomes. Forty patients with traumatic brain injury and FS (TBI+FS) underwent 12 weekly sessions of NBT and provided pre-/post-intervention resting-state functional magnetic resonance imaging (MRI), seizure logs, and behavioral assessments. Fifty-five individuals with TBI without FS (TBI-only) completed the same measures, received standard medical care but not NBT, and functional MRI ~12 weeks apart. For each key region, two-sample t-tests assessed direct group comparison. Repeated measures analysis of covariance assessed how group differences evolved over time and how these changes were modulated by the changes in seizure frequency, diagnosis duration, or behavioral scores (false discovery rate corrected at p < .05). With NBT, seed-based whole-brain functional connectivity was significantly higher between right anterior insula and left supplementary motor area in TBI+FS compared to TBI-only, and between left anterior insula and left postcentral gyrus in seizure-free TBI+FS compared to those who were not seizure-free. Percentage decrease in seizure frequency with NBT was associated with lower functional connectivity between bilateral insula and left superior medial frontal gyrus in patients with FS. Improvements in behavioral measures did not correspond to changes in functional connectivity. The study underscores the relationship between the changes in resting-state functional connectivity of the anterior insula in FS and treatment response to NBT and illustrates the potential neurophysiologic MOA of NBT for the treatment of FS; it suggests an independence of this MOA from the potential effects of NBT on behavioral measures.