Dong Kong, PhD - US grants

DESCRIPTION (provided by applicant): Agouti-related peptide (AgRP)-expressing neurons in the arcuate nucleus of the hypothalamus are critical regulators of energy balance. AgRP neurons are anabolic: optogenetic or pharmaco-genetic stimulation of AgRP neurons drives intense feeding behavior and promotes obesity; disruption of these neurons in adult mice causes severe anorexia. Given the important roles played by AgRP neurons, there is great interest in understanding the factors that regulate their activity. Most previous studies have been placed on examining their direct regulation by circulating factors, such as leptin, insulin, and ghrelin. Their synaptic regulation by neurotransmitters released from other neurons in the brain, however, has been greatly overlooked. This is unfortunate because defective synaptic transmission on these neurons could also contribute to eating disorders. Furthermore, it is likely that the mechanism-of-action for hormonal regulation of AgRP neurons, for example by ghrelin, is modulation of afferent synaptic transmission. Through the recent work at Dr. Brad Lowell group (Prof of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School), the candidate has found that glutamatergic synaptic transmission plays a key role in AgRP neurons. In particular, he discovered that AgRP neurons but not the adjacent POMC neurons have dendritic spines, 1um3 protrusions where majority of glutamatergic synapses reside and within which glutamate NMDA receptors operate to control synaptic plasticity. In addition, he found that the fasting- induced activation of AgRP neurons and its related feeding behavior are paralleled (and likely caused) by a marked increase in the number of spines (i.e. spinogenesis), and this is dependent on postsynaptic NMDARs. Thus, glutamatergic transmission and its plasticity, as modulated by postsynaptic NMDARs, play critical roles in controlling AgRP neuron activity and their related feeding behaviors. These findings, which are recently published on Neuron, provide the candidate a unique opportunity to interrogate synaptic regulations in the feeding circuits. However, to pursue such studies, some state-of-art technologies (such as electrophysiology combined with 2-photon microscope imaging), which are beyond the scope of Dr. Lowell's lab and not available at BIDMC, are required. Toward these ends, the candidate is now trained by Dr. Bernardo Sabatini (Prof of Neurobiology, HHMI, Dept. of Neurobiology, Harvard Medical School), to use such advanced technologies to study structural and functional properties of spines. In this K01 mentored career development award, under the mentorship of Dr. Sabatini, and co- mentorship of Dr. Lowell, the candidate proposes to obtain acquisition in both scientific knowledge and in technologies (electrophysiology combined with 2-photon microscope imaging) related to synapse studies, and develop other necessary skills toward his career independence (immediate goal). The candidate is now Instructor in Medicine at BIDMC and Harvard Medical School. Once he finishes training with Dr. Sabatini, he will be transitioned to Assistant Professor at BIDMC and establish his own laboratory, become an independent investigator in the area of nutrition, obesity and neuroscience research, and apply multi- disciplinary methodology to understand synaptic plasticity in hypothalamic neurons controlling feeding, energy expenditure, and fuel metabolisms (long-term goal). Therefore, the K01 award will provide the candidate protected time to obtain necessary training before he becomes independent. At the same time, the proposed project in this award will greatly help the candidate to obtain subsequent R01 grant support.

? DESCRIPTION (provided by applicant): Neurons in the brain detect changes in nutritional status and environment, and relay signals to their downstream targets to regulate food intake and energy expenditure, the balance of which is critical to maintain normal body weight and protect from obesity. Given the complexity of the brain, the neurobiological mechanisms underlying these processes are poorly understood. Efficient treatment of obesity is thus still lacking. Although a lot of success has been recently achieved in dissecting the neural circuitry of feeding behaviors, the research to understand the neural basis of energy expenditure is still in its infancy. In a recent study, we focused on a group of hypothalamic neurons labeled by cre activity in Rip-cre transgenic mice, thereafter referred to as RIP neurons, and uncovered an arcuate-based circuit that selectively drives brown adipose tissue (BAT) activity and energy expenditure. Specifically, we disrupted GABAergic neurotransmission from these neurons in a cre-dependent manner and observed that mice lacking synaptic GABA release from RIP neurons have reduced energy expenditure and become obese, and are extremely sensitive to high fat diet-induced obesity due to defective thermogenesis. Leptin's ability to stimulate energy expenditure is also attenuated in these animals. With pharmacogenetic DREADDs, we acutely and selectively activated the subset of RIP neurons in the arcuate nucleus (ARC) and rapidly stimulated BAT-mediated energy expenditure. Moreover, with channelrhodopsin-assisted circuit mapping (CRACM), we characterized that ARC RIP neurons project to the paraventricular nucleus (PVH) and specifically innervate the PVH neurons that project to the nucleus of solitary tract (NTS) in the brain stem. Of great interest, we observed that RIP neurons have no effects in regulating food intake. These findings demonstrate that GABAergic RIP neurons in the ARC selectively drive energy expenditure, contribute to leptin's stimulatory effect on thermogenesis, and protect against diet-induced obesity. Given the importance of these neurons in maintaining body weight and resisting obesity, it is crucial to comprehensively understand their related neural circuitry. In Aim 1, we set out to employ advanced optogenetic and deep brain imaging approaches to investigate the regulations of RIP neurons during thermoregulation and functionally assess their projection to the PVH in stimulating energy expenditure. In Aim 2, we will focus on the output signals of RIP neurons in the PVH and identify their efferent subset of neurons that convey their signals to the BAT. Finally, in Aim 3, we will survey the afferent inputs of RIP neurons within a microcircuit in the arcuate nucleus and scrutinize their functions in regulating energy expenditure. In total, these proposed studies could significantly advance our understanding of the neural basis of energy expenditure and provide novel information to prevent obesity.

AMP-activated protein kinase (AMPK), an evolutionarily conserved serine/threonine kinase stimulated by both decreased cellular energy status and increased calcium, is an important player acting at the interface between metabolism and brain function. In addition to metabolic diseases like obesity and diabetes, abnormal AMPK activities have been implicated in a variety of neurological disorders with dysfunctional neurotransmission. The neurobiological mechanisms of AMPK responsible for these effects, however, are largely unknown. Recent studies have suggested that agouti-related peptide (AgRP)-expressing neurons in the hypothalamus, a master controller of feeding and energy balance, receive intense glutamatergic input and their excitatory synaptic plasticity plays an essential role in regulating AgRP neuron firing and related feeding. Importantly, our prior findings demonstrate that fasting significantly induces dendritic spinogenesis, glutamatergic synaptogenesis, and firing in AgRP neurons, and this fasting-induced plasticity requires postsynaptic NMDA receptors on AgRP neurons and contributes essentially to their fasting-induced activation. The neurobiological mechanism that underlies fasting-induced plasticity in AgRP neurons, however, is left unknown. In this context, AMPK in the hypothalamus is activated by fasting and manipulation of AMPK activity in this region affects feeding. In addition, when stimulated pharmacologically in brain slices, AMPK increases glutamatergic input to AgRP neurons. These findings suggest that AMPK likely trigger this fasting-induced plasticity. However, given the wide expression of AMPK in the brain and its multi-faceted roles in cellular biology, whether AMPK in AgRP neurons mediates fasting-induced feeding is still in debate. How fasting modulates AMPK dynamics is also unclear. By employing a battery of neuron-specific approaches, including neuron-specific transgenic and knockout mouse lines, cre-dependent AAV viral vectors, 2-photon laser scanning microscopy (2PLSM) combined with whole cell patch-clamp electrophysiology, and particularly 2PLSM-based fluorescence lifetime imaging (FLIM), this proposal aims to provide a unique, multi-faceted study to understand AMPK signaling and its physiology in the neurotransmission of AgRP neurons. Based on our compelling preliminary findings, we hypothesize that a postsynaptic pathway engaged by AMPK in AgRP neurons drives fasting induced excitatory synaptic plasticity and the plasticity brought about by this pathway accounts for the effects of AMPK on energy balance (Aim 1). We further hypothesize that AMPK functions as a critical integrator of diverse inputs (such as fasting, ghrelin, and leptin) of AgRP neurons and mediates both synaptic and cellular changes (Aim 2). Our novel findings on synaptic plasticity and AMPK will provide innovative knowledge in the feeding circuits. Given the wide distribution of AMPK and its substrates, the uncovered pathway engaged by AMPK in AgRP neurons will likely operate both within and beyond the hypothalamus, and have important implications for many processes where synaptic plasticity plays a key regulatory role.

There is a strong comorbidity of narcolepsy and diabetes/obesity; however, the causal underlying mechanism is unclear. We recently performed studies using astrocyte-specific connexin 43 (Cx43) knockout mice (Cx43 KO) and found that they display both a narcolepsy-like phenotype and metabolic dysregulation. These linked phenotypes, arising from a single genetic manipulation, raise the potential that we have identified a cellular and molecular underpinning of this clinical linkage. We will use the collective strengths of Drs. Haydon and Kong, who are highly experienced in studying astrocytes and the control of sleep/wake cycles (Haydon) and the study of the hypothalamic neural circuits and metabolic control (Kong). Together, we will test the hypothesis that astrocytic connexins are essential for the supply of lactate as an energy substrate to orexinergic neurons, and in doing so, modulate orexinergic control of wakefulness and metabolic control. Pierre Magistretti and colleagues proposed an attractive hypothesis concerning metabolic coupling between astrocytes and neurons in which the astrocyte metabolizes glucose to lactate, then shuttle this energy source to neurons where lactate is converted to pyruvate, which is used in oxidative phosphorylation. This Astrocyte- Neuron Lactate Shuttle (ANLS) is attractive because: i) astrocytes contact the vasculature and express GLUT1, a glucose transporter, they can take up glucose. ii), astrocytes are considered to be biased towards glycolysis, and iii) neurons express monocarboxylate transporters (MCT) that are required for the uptake of lactate. We will extend our initial observations to test the hypothesis that astrocyte-derived lactate is required by orexinergic neurons to promote their electrical activity and that experimental manipulation of orexinergic neuronal activity is both necessary and sufficient to cause narcolepsy and metabolic disorders. Aim I: We will test the hypothesis that the deletion of Cx30 and Cx43 impairs the Astrocyte-Neuron Lactate Shuttle AND promotes narcolepsy and systemic metabolic dysfunction. Aim II: We will test the hypothesis that astrocyte-derived lactate modulates orexinergic neuron activity. Aim III: We will test the hypothesis that the activation of orexinergic neurons is sufficient to rescue normal phenotypes in connexin KO mice.

PROJECT SUMMARY Intellectual disability (ID) is the most frequent cause of developmental disabilities and affects 1-3% of the population. Most cases of ID, unfortunately, are still lacking effective prevention and intervention options. Among the heterogeneous causes of ID, metabolic dysfunctions in the brain are playing an essential role, with a particular emphasis on lipids, which form a fundamental basis for both neurobiology and brain function. Given the extreme complexity of lipid metabolism and the central nervous system, how lipids and which species of them regulate neuronal functions and contribute to the development of ID, however, are still poorly understood. In this context, Acsl4, a gene encoding an isozyme of long chain acyl-CoA synthetase family, was found mutated in X-linked ID or Alport syndrome, thus providing a valuable interface to understand both mental health and lipid metabolism. Due to the lack of effective approaches to perturb it in neurons, ACSL4?s precise functions in the brain are still undetermined. Whether dysregulations of ACSL4 contribute to the development of ID is also unknown. Recently, based on multiple lines of evidence and our preliminary observation, we hypothesize that ACSL4 in hippocampal neurons is essential for neuron development and cognition. We further propose that lacking ACSL4 in the mouse hippocampus affects learning and memory (Aim-1) by dampening synapse formation and synaptic plasticity (Aim-2). Importantly, we postulate that ACSL4 achieves such regulatory functions in the brain by altering lipid metabolism (Aim-3). Through a recent effort to investigate the role of ACSL4 in adipose tissue in vivo, we have established a floxed-Acsl4 conditional mouse model, which allows tissue-specific deletion of the gene in a cre-dependent manner. We then plan to leverage this mouse line to establish a neuron-specific ACSL4 knockout mouse model to assess the above hypotheses. The proposed studies in the current application will discover a novel neurobiological mechanism by which lipid homeostasis orchestrated by ACSL4 regulates synaptic and neural functions in the brain and maintains mental health. A battery of rigorous, comprehensive approaches at molecular, cellular and system levels will be employed. Fulfillment of the research will likely reveal new biology and novel biomarkers relevant to the etiology and treatment of mental disorders and will therefore be of great values in both basic and translational research on mental health.