Review
6 July 2023
Molecular Signaling Section, Laboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD 20892, USA
Mouse Metabolism Core, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD 20892, USA
Author to whom correspondence should be addressed.
Precise neural regulation is required for maintenance of energy homeostasis. Essential to this are the hypothalamic and brainstem nuclei which are located adjacent and supra-adjacent to the circumventricular organs. They comprise multiple distinct neuronal populations which receive inputs not only from other brain regions, but also from circulating signals such as hormones, nutrients, metabolites and postprandial signals. Hence, they are ideally placed to exert a multi-tier control over metabolism. The neuronal sub-populations present in these key metabolically relevant nuclei regulate various facets of energy balance which includes appetite/satiety control, substrate utilization by peripheral organs and glucose homeostasis. In situations of heightened energy demand or excess, they maintain energy homeostasis by restoring the balance between energy intake and expenditure. While research on the metabolic role of the central nervous system has progressed rapidly, the neural circuitry and molecular mechanisms involved in regulating distinct metabolic functions have only gained traction in the last few decades. The focus of this review is to provide an updated summary of the mechanisms by which the various neuronal subpopulations, mainly located in the hypothalamus and the brainstem, regulate key metabolic functions.
Extensive evidence has unequivocally confirmed the importance of the brain in metabolic disorders and obesity [1]. Research identifying its pathophysiological role has spanned over several decades. From its humble beginnings of employing rodents with hypothalamic lesions, which aided in identifying the role of distinct brain regions in appetite/satiety regulation, to the use of more sophisticated approaches such as chemogenetics and genome-wide association study (GWAS) to identify novel therapeutic targets/pathways in the brain, the central nervous system (CNS) has now been firmly established as a critical component that is dysregulated in the development of obesity [2,3,4]. More recently, research on the metabolic role of the CNS has also paved the way for the identification of drug targets in metabolic disorders such as Type 2 diabetes and obesity. Furthermore, obesity results in long-lasting changes in the cytoarchitecture and synaptic plasticity of the brain, particularly in the hypothalamus [5,6,7]. Hence, a comprehensive understanding of how the CNS fine-tunes metabolic functions could aid in the further development of therapeutics for various metabolic disorders. A key aspect of current metabolic research is focused on understanding the contributions of the hypothalamic and the brainstem circuitry in the regulation of appetite and energy homeostasis. By manipulating neuronal populations located in these regions, research has uncovered several major neural circuits that exert control over appetite and metabolic functions.
The hypothalamus and brainstem are critical components of the homeostatic system that regulates appetite and energy balance. These key brain regions have distinct neuronal populations and nuclei, having both complimentary and contrasting roles, exert tremendous control over several facets of energy balance. This occurs both at the level of energy intake and energy expenditure. This section serves to introduce the distinct hypothalamic and brainstem nuclei, and its primary role in the regulation of energy balance. More detailed mechanisms involving the regulation of glycemia and lipid metabolism will be discussed in the next section.
The hypothalamus is one of the most well-studied brain regions in metabolism. Apart from regulating a broad range of thermoregulatory, reproductive, and cardiovascular functions, it also exerts tremendous influence on several aspects of energy balance. The hypothalamus is composed of multiple nuclei located adjacent to the third ventricle. These nuclei comprise a distinct subpopulation of neurons, capable of altering energy intake and/or energy expenditure via anabolic or catabolic functions. The complex nexus of hypothalamic neuronal interconnections can integrate responses from peripheral signals (hormones, nutrients, and metabolites), to modulate appetite centrally, and to influence lipid and glucose metabolism, peripherally. Additionally, they also have reciprocal projections to and from extrahypothalamic nuclei located in the brainstem, midbrain, and forebrain which can also alter synaptic activity in the hypothalamic metabolic circuits. Consequently, via integration and coordination of responses from the brain and periphery, hypothalamic nuclei are key regulators of energy homeostasis.
The ARC is considered as one of the most important brain regions involved in the regulation of appetite and energy expenditure. Located near the median eminence, a region enriched in fenestrated capillaries, the ARC is accessible to circulating hormones, nutrients and metabolites, thus, serving as an ideal relay center to communicate circulating peripheral signals to the brain. The ARC comprises two distinct neuronal subpopulations that have opposing roles in energy homeostasis, the anabolic neuropeptideY/Agouti-related protein (NPY/AGRP) neurons and the catabolic, pro-opiomelanocortin (POMC) and cocaine-and amphetamine-regulated transcript (CART) or POMC/CART neurons (referred to henceforth as AGRP and POMC neurons). Both these neurons are first order neurons, which have glucose and nutrient sensing capabilities, in addition to receiving input from circulating hormones and satiety signals [8,9,10,11]. These counterregulatory neuronal populations are modulated by energy status. Food deprivation rapidly activates AGRP neurons and inhibits POMC neurons [12,13,14]. AGRP neurons release both NPY, which is an agonist for the Y1-5 receptors, and AGRP, an inverse agonist for melanocortin receptors [15,16]. Ablation of AGRP neurons results in a dramatic reduction in feeding, while acute activation results in a robust increase in food intake, weight gain, and altered autonomic outflow to several organs and tissues [17,18,19,20]. NPY was one of the first orexigenic neuropeptides to be identified, and subsequent functional studies revealed a potent, albeit fleeting, appetite-stimulating effect [21]. More recently, it has been revealed that NPY-mediated effects on feeding are mediated via the Y1 receptor, while its effects on energy expenditure are driven via the Y2 receptor [22]. Although both of these orexigenic neuropeptides, NPY and AGRP, have complimentary roles in triggering a hyperphagic response and reducing energy expenditure, the longer-lasting or sustained effect of these neurons on food intake is dependent on AGRP release, while the more rapid effect on food intake is dependent on NPY secretion [20,21,23]. Additionally, AGRP neurons also release GABA, which plays an integral role in AGRP-mediated effects on appetite and energy balance [21,24]. Furthermore, diet-induced obesity blunts AGRP responsiveness to circulating hormones [25]. In stark contrast to AGRP neurons, POMC neurons have a pronounced catabolic effect due to their ability to release the anorectic neuropeptide, α-melanocyte-stimulating hormone (α-MSH), a major satiety neuropeptide which is an agonist of melanocortin receptors [26]. Ablation of POMC neurons was reported to result in a mild obesity phenotype characterized by both reduced and increased food intake [27,28]. Interestingly, only chronic, but not acute chemogenetic activation of these neurons results in suppression of food intake, suggesting a role for POMC in maintaining long-term energy homeostasis [28]. POMC neurons have been reported to exhibit functional and spatial heterogeneity characterized by differences in both molecular architecture and anatomical projections to distinct brain regions, suggesting a more complex neural network involved in metabolic control [29,30,31]. POMC neuronal activity is also regulated by AGRP neurons. Anatomic and functional evidence indicates that GABA-releasing AGRP neurons are involved in inhibiting POMC neuronal activity and α-MSH release [32,33,34]. Apart from α-MSH, it is also to be noted that POMC neurons also release β-endorphin, which binds to the μ-opioid receptors. Both these POMC-derived neuropeptides have functionally antagonistic roles in the regulation of energy balance [35]. Both hypothalamic AGRP and POMC neurons are known to express the μ-opioid receptors (MOR). In the case of POMC neurons, the MORs function as autoinhibitory receptors that are activated by the release of β-endorphins [36]. Interestingly, while α-MSH is predominantly involved in suppressing appetite, β-endorphin were shown to play a major role in promoting a palatability-driven feeding response [37,38]. Naltrexone, a MOR antagonist which has been shown to suppress feeding on a short-term basis, has been shown to have stimulatory effects on POMC neurons in both rodents and humans [39,40]. More about their therapeutic utility will be covered in a later section.
Both AGRP and POMC neurons also express receptors for insulin (IR) and leptin (LepR). Leptin depolarizes and increases firing frequency of POMC neurons, while hyperpolarizing and inhibiting AGRP/NPY neuronal activity and neuropeptide release [41,42,43,44,45]. Mechanistic studies revealed that deletion of Rho-kinase 1, a protein kinase involved in cytoskeletal reorganization and neuropeptide release, in both AGRP and POMC neuronal populations resulted in leptin resistance and obesity [46,47]. Collectively, these data point to a crucial central mechanism by which leptin can induce a negative energy balance. Studies investigating the role of insulin signaling in both AGRP and POMC neurons on appetite regulation have yielded contradictory results. While some studies reported on little-to-no effect on appetite and body weight change with IR deletion in AGRP neurons, others have described a more nuanced role of AGRP-specific insulin signaling on regulating meal size [48,49]. A context-dependent appetite suppression role is reported for insulin signaling in the AGRP neurons, which is characterized by acute repression of feeding bouts without altering total calorie intake, and the suppression of highly palatable high-fat-diet food over standard chow [49]. In the case of POMC neurons, while the deletion of LepR results in mild obesity, knockout of IR in these neurons had no significant effect on body weight [50,51]. Furthermore, both AGRP and POMC neurons are modulated by postprandial signals, such as ghrelin, incretins, and amylin, to regulate food intake [52,53,54,55,56]. Apart from having integral roles in appetite and satiety regulation, these neuronal populations are also involved in maintaining glucose homeostasis as chemogenetic activation of AGRP and POMC neurons revealed distinct roles of G protein activation on food intake and glycemic control [18,20,28,57]. The mechanisms through which both these neuronal populations regulate glucose homeostasis will be discussed in later sections of this review. Additionally, both AGRP and POMC neurons can also regulate energy balance via the hypothalamic–pituitary–thyroid (HPT) axis. HPT axis is well-known to stimulate energy expenditure. Thyroid hormones play an important role in maintenance of homeothermia, and stimulation of the thyroid axis is known to increase energy expenditure via thermogenesis [58]. ICV administration of NPY has been shown to suppress circulating levels of thyroid hormones [59]. Interestingly, the melanocortin system has also been shown to regulate the HPT axis. Both in vivo and in vitro studies revealed that α-MSH can stimulate the HPT axis by increasing the levels of thyroid stimulating hormone (TSH), while AGRP on the other hand inhibits it [60,61]. For more information on the role of the melanocortin system in regulating the HPT axis, readers can refer to other reviews on this topic [62].
Another key aspect of the ARC neurons, especially POMC, is that they exhibit sexual dimorphism. Higher number of POMC neurons and increased neural activity were observed in female animals when compared to their male counterparts [63]. Disruption of key genes in POMC neurons in female mice resulted in the development of obesity [63,64,65,66]. More recently, POMC-specific alteration of certain highly expressed CNS genes, resulted in changes in glucoregulation and energy balance in female mice only [67,68,69].
ARC is highly susceptible to synaptic plasticity in response to the hormonal milieu. Both AGRP and POMC neurons have been described as exhibiting some level of synaptic rewiring under periods of food deprivation and overfeeding conditions [70]. Particularly, the melanocortin system has been reported to exhibit synaptic remodeling under both extreme metabolic changes, such as starvation and overfeeding, but also under physiological feeding states which results in modest metabolic changes [71,72,73]. Plasticity of the ARC has important implications in obesity, as diet-induced obesity has been demonstrated to suppress hypothalamic remodeling and neurogenesis resulting in reduced neuronal turnover [74]. It was also demonstrated to result in reactive gliosis in the ARC with altered synaptic architecture of the NPY and POMC neurons [75]. High fat diet (HFD)-induced neurogenesis is not restricted to the neuronal populations alone in the ARC. HFD activated neurogenesis in the median eminence however leads to energy storage, while prevention of it results in a reduction in weight gain [76]. Stimulation of neurogenesis in response to HFD is observed in female mice and not in males, suggesting a sexual dimorphic nature of hypothalamic neurogenesis [77].
The PVH serves as an important convergence/termination point for orexigenic and anorexigenic projections arising from the ARC and other hypothalamic regions. Neurons present in this region express two different types of melanocortin receptors subtypes (MC3R and MC4R) that can
