The hypothalamic arcuate nucleus and the

Full text

8
The hypothalamic arcuate nucleus and the
control of peripheral substrates
Aurélie Joly-Amado, PhD, Post Doctoral Fellow
1
,
Céline Cansell, PhD, PhD Student
1
,
Raphaël G.P. Denis, PhD, Research Engineer
1
,
Anne-Sophie Delbes, MD, Engineer,
Julien Castel, BD, Technician, Sarah Martinez, MD, Engineer,
Serge Luquet, PhD, Research Scientist/Principal Investigator
*
Univ Paris Diderot, Sorbonne Paris Cité, Unité de Biologie Fonctionnelle et Adaptative (BFA) UMR 8251 CNRS,
F-75205 Paris, France
Keywords:
obesity
diabetes
nutrient partitioning
substrate utilization
hypothalamus
agouti-related peptide
neuropeptide Y
autonomic nervous system
The arcuate nucleus (ARC) of the hypothalamus is particularly
regarded as a critical platform that integrates circulating signals of
hunger and satiety reflecting energy stores and nutrient availability.
Among ARC neurons, pro-opiomelanocortin (POMC) and agouti-
related protein and neuropeptide Y (NPY/AgRP neurons) are
considered as two opposing branches of the melanocortin signaling
pathway. Integration of circulating signals of hunger and satiety
results in the release of the melanocortin receptor ligand
a
-mela-
nocyte-stimulating hormone (
a
MSH) by the POMC neurons system
and decreases feeding and increases energy expenditure. The
orexigenic/anabolic action of NPY/AgRP neurons is believed to rely
essentially on their inhibitory input onto POMC neurons and second-
orders targets. Recent updates in the field have casted a new light on
the role of the ARC neurons in the coordinated regulation of pe-
ripheral organs involved in the control of nutrient storage, trans-
formation and substrate utilization independent of food intake.
Ó2014 Elsevier Ltd. All rights reserved.
*Corresponding author. Unité “Biologie Fonctionnelle & Adaptative”(BFA), Université Paris Diderot-Paris 7, CNRS UMR 8251,
4 rue Marie-Andrée Lagroua Weill-Hallé, Bâtiment Buffon, Case courrier 7126, 75205 Paris Cedex 13, France. Tel.: þ33 1 57 27 77
93; Fax: þ33 1 57 27 77 96.
E-mail address: serge.luquet@univ-paris-diderot.fr (S. Luquet).
1
These authors contributed equally to this work.
Contents lists available at ScienceDirect
Best Practice & Research Clinical
Endocrinology & Metabolism
journal homepage: www.elsevier.com/locate/beem
http://dx.doi.org/10.1016/j.beem.2014.03.003
1521-690X/Ó2014 Elsevier Ltd. All rights reserved.
Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737

Introduction
Obesity and correlated diseases such as hypertension, atherosclerosis, dyslipidemia, coronary dis-
eases and diabetes mellitus are now clearly identified as a worldwide pandemic in both developing and
developed countries [1]. Obesity per se, as well as the constellation of associated pathophysiology
defined as the metabolic syndrome is inflicting an escalating public health burden that, according to
the World Health Organization (WHO) (http://www.who.int/mediacentre/factsheets/fs311/en/) origi-
nates mostly from a drastic change in lifestyle involving an increased consumption of energy-rich diet
and reduced energy expenditure.
Whereas some genetic loci were clearly identified, and extensively studied as monogenic causes for
obesity, it is widely accepted that the metabolic syndrome is in essence a multifactorial diseases that
encloses a complex network of molecular, cellular and physiologic alterations [2,3]. Although some
debate exists on the exact number, a study from the American Center for Disease Control andPrevention
recently made a case for obesity as the next top American killer before tobacco [4]. The urge to under-
standing the fundamental determinants of obesity and obesity-related disease has therefore generated a
frantic race in the scientific community to unravel the mechanism involved and build a potential
therapeutic strategy. Appropriate energy homeostasis results from the exquisite balance between en-
ergy intake and energy expenditure. In that regard, several determinants of feeding behavior are
extensively studied encompassing the homeostatic regulation of nutrient intake, generally attributed to
a hypothalamic-brainstem circuitry, but also the hedonic and motivational aspect of feeding relying, at
least in part, on dopamine release in the mesocortico-limbic system. Different aspects of energy
expenditure that include adaptive thermogenesis, physical activity and basal metabolic rate have also
been dissected out as an essential part of obesity etiology. However, a fundamental component of the
energy balance involves the ability of the brain tocoordinate the activity of peripheral tissue to insure the
fate of a nutrient once ingested. It is becoming evident that, aside of excessive energy consumption,
obesity-related metabolic complications involve the inappropriate conversion, storage and utilization of
nutrients: an integrated process referred as to “nutrient partitioning”.
Feeding inputs, circulating signals reflecting energy stores as well as cognitive and circadian control
are among the many parameters that will be integrated at the level of the central nervous (CNS) system
which will in turn orchestrate peripheral organ activity through the modulation of the autonomic
nervous system (ANS). Although genetic and pharmacological interventions have allowed deciphering
key molecular pathways underlying the regulation of energy expenditure and food intake, these
studies also provided evidence that obesity-related substrate utilization could be manipulated inde-
pendently of food intake and body weight.
Bariatric surgery offers a vivid example in which type 2 diabetes can be corrected in a few days after
surgery showing a time-lapse that does not correlate with body weight loss. Although the mechanisms
are still an active matter of research, it illustrates how obesity and its most common corollary disease
can be separated [5–8]. It is tempting to speculate that a rapid reshaping of the reciprocal nervous
dialogue between the brain and the periphery could be instrumental in the restoration of proper
nutrient partitioning. In that view, the links between obesity and obesity-related diseases such as
diabetes and dyslipidemia, could originate from a primary dysfunction in the ability of the brain to
orchestrate the activity of peripheral tissues [9,10].
The arcuate nucleus (ARC) of the hypothalamus contains at least two crucial populations of neurons
that continuously monitor signals reflecting energy status and promote the appropriate behavioral and
metabolic responses to changes in energy demand. Neurons making pro-opiomelanocortin (POMC)
decrease food intake and increase energy expenditure through activation of G protein-coupled mel-
anocortin receptors (MCR) via the release of
a
-melanocyte-stimulating hormone (
a
MSH). Until
recently, the prevailing idea was that the neighboring neurons expressing the orexigenic neuropep-
tides, agouti-related protein (AgRP) and neuropeptide Y (NPY) (NPY/AgRP neurons) increased feeding
and decrease energy expenditure primarily by opposing the anorexigenic/catabolic actions of the
POMC through both the competitive inhibition of melanocortin tone at the postsynaptic level and via
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737726

Fig. 1. The arcuate nucleusand the control of nutrient partitioning Sagittal section(upper panel)of a mouse brain showinginterconnected
nuclei engaged in the control of energy balance. Arcuate neurons project to second order targets including the paraventricular nucleus
(PVN), thelateral hypothalamus(HyLat), the parabrachialnucleus (PBN), the dopaminergicneurons in the ventral tegmental area (VTA) of
the midbrain and the nucleus tractussolitarii (NTS)in the brainstem. Hypothalamic nucleialso receiveinput from the dorsal vagalcomplex
(DMX).The PVN integratesinputsfrom ARC neuronsand from the suprachiasmaticnucleus (SCh).The arcuatenucleus of the hypothalamus
(ARC)contains at least twopopulations of neurons that control energybalance the orexigenic/anabolic neurons producing neuropeptideY
andAgouti-relatedprotein(NPY/AgRP)and the anorexic/catabolicneuronsproducingpro-opiomelanocortin(POMC) andthe naturalligand
for the melanocortin receptor. These first order neurons are located exquisitely close to the median eminence (ME), one of the brain’s
circumventricular organs that lies at the bottom of the third ventricle (3
rd
V). The ME represents a microenvironment composed of fenes-
tratedcapillaries(red) andglia cells highlystructuredby tight junctions(green).Together, thiscreates a privilegedregion of the blood–brain
barrier (BBB) in which macromolecules and energy-related (leptin, insulin, ghrelin) peptides access target neurons inthe ARC through a
regulated passage involving highly specialized hypothalamic glial cells: the tanycytes. ARC neurons are of diverse nature and display
segregated projection to hypothalamic and extra hypothalamic nuclei. Integration of SCh “Clock”input and ARC-relayed nutrient input
results in the ANS-mediated coordination of peripheral organ activity. This process can be independent from the regulation feeding,
neuroendocrine release or metabolic rate and results in the concertedorchestration of nutrient transformation, storage and utilization.
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737 727

directed inhibition of POMC firing rate (Fig. 1). This review is an attempt to cover some of the new
functions, mechanisms and neurocircuitry by which ARC neurons can specifically control nutrient
partitioning trough feeding independent manners.
The two ages of the MONA LISA hypothesis
In the CNS, the hypothalamus has rapidly been recognized as a primary integrator of circulating
signal of hunger and satiety and extensively studied for its intimate implication in the control of ANS
output, behavior and endocrine release. Several essential physiological functions such as salt & water
intake, reproduction, wake & sleep, body temperature, circadian rhythmicity and metabolic rate rely on
the coordinated output of a highly differentiated network of hypothalamic nuclei.
In the early 1940s, experiments using electrical stimulation and lesioning allowed the identification
of functional different nuclei in the mediobasal hypothalamus (MBH) that had specific actions on energy
homeostasis [11,12]. The ventromedial hypothalamus (VMH) was first considered as a satiety area
because destruction of the VMH resulted in hyperphagia and obesity, whereas electric stimulation of
VMH led to decreased food intake and body weight. Conversely, the destruction of the lateral hypo-
thalamus (LH) led to anorexia, while stimulation of LH caused voracious feeding and obesity. The overall
conceptual framework that emerged from these observations, i.e., one in which a “satiety center”kept
the “feeding center”in check, was largely abandoned because of the realization that the LH lesions
disrupted catecholaminergic nerve tracts passing through the hypothalamus that were essential for
normal feeding and movement, and that the VMH lesions had a major impact on the autonomic output
[13]. Nevertheless, this experimental paradigm already provided an illustration of the fact that obesity
could not be solely attributable to hyperphagia, but rather also involved increased insulinemia [14].In
the same line, mono-sodium-glutamate (MSG) treatment at an early stage after birth results in an
extensive damage of the ARC and leadsto late onset obesity despite normal feeding [15–17]. A common
feature displayed by these obese models was a global decrease in sympathetic tone associated with
endocrine alterations, amongst others in corticosterone. These observations prompted the statement by
Bray that Most Obesities kNown Are Low InSympathetic Activity: the “MONA LISA”hypothesis that
already pointed defective nutrient partitioning as a preponderant mechanism driving obesity [18].
The identification of the obese gene encoding a 16 kDa protein called leptin (Lep
ob
;[19]) has pro-
moted a radical change in the general conceptual framework describing the central regulation of en-
ergy balance. The feeding & satiety center hypothesis has been replaced by a notion of intermingled
neuronal networks in which highly specialized neurons [20] are able to encode primary signals
reflecting blood-borne information about energy stores into synaptic transmission. Leptin levels rise
and fall in direct proportion to adipose tissue mass and are relatively insensitive to daily changes in
food intake. Food deprivation causes leptin levels to drop as energy stores are utilized, and this decline
promotes endocrine and behavioral alterations that result in increased appetite and decreased energy
expenditure. Leptin’s targets are found not only in the CNS, but also in peripheral tissues [21,22]. Mice
lacking leptin (ob/ob) become morbidly obese as a consequence of metabolic disturbances and hy-
perphagia; they are also cold intolerant, diabetic and infertile [13,23]. The discovery of leptin and later
on the leptin receptor, a single-pass transmembrane protein of the gp130 cytokine receptor family [24],
which when mutated is responsible for the db/db phenotype Lepr
db
[25], fueled an active research field
in which a leptin-based treatment was hoped to fulfill the promises of a pharmaceutical approach to
surfeit the obesity epidemic. Although, leptin deficiency provided a fantastic model to study the
metabolic syndrome it accounted as for other monogenic obesity for only rare cases of obesity [26].
However, the identification of the leptin signaling pathway opened up a new era in the understanding
of the central control of energy homeostasis and allowed the dissection of several fundamental neu-
rocircuitries and molecular mechanisms underpinning the hypothalamic control of feeding, energy
expenditure and neuroendocrine control [27]. In the new leptin era, the MONA LISA hypothesis still
stands [20]. Leptin, as well as other factors are impinging directly onto ARC neurons to modulate their
activity, in turn these “first order”neurons project to several structures that can independently
modulate ANS output and separately affect different peripheral organs [9,28,29]. This mechanism
provides a circuit-based blueprint to control peripheral substrate utilization and prompted the pro-
vocative hypothesis that sees several features of the metabolic syndrome including dyslipidemia,
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737728

insulin resistance, high blood pressure, abdominal fat, as a primary defect of brain ANS control [9].
Instead of a global decrease in catecholamine release, the MONA LISA hypothesis could be replaced for
a more selective one, i.e., obesity is the reflection of tissue-specific changes in ANS output which will in
turn induce a global change in energy fluxes.
The arcuate nucleus of the hypothalamus: exquisite location for exquisite regulation
ARC of the hypothalamus is located at the bottom of the third ventricle (3
rd
V) in close vicinity to one
of the circumventricular organs (CVO): the median eminence (ME), a structure of the blood–brain
barrier (BBB) that has evolved as to allow selective exchange between blood-borne peptide and ce-
rebrospinal fluid and ARC neurons (Fig. 1). Proper regulation of body weight relies on the ability of
peripheral signals to reach and modulate the activity of effectors neurons. These neurons will primarily
integrate circulating signals and encode energy-related signals into synaptic transmission that will in
turn impinge onto secondary target neurons or “second-order neurons”.
The unique specification of the ME was vividly illustrated by the identification of the role of
tanycytes as gatekeepers of the BBB [30]. Tanycytes are highly specialized hypothalamic glial cells that
extend from the ependymal surface of the 3
rd
V to a plexus of permeable fenestrated capillaries [31]
that can accommodate a rapid transport of energy-relevant peptides such as ghrelin [32]. Tanycytes
are layered around tight junctions and represent the first rampart between the blood and the CSF [33].
Changes in nutrient availability were shown to directly affect ME permeability through a glucose
sensing- vascular endothelial growth factor A (VEGF-A) releasing mechanism from the tanycytes to
endothelial cells [31]. Tanycytes were also shown to be the first-and critical-step into leptin’s entry into
the hypothalamus [34] (Fig. 1). This work comes in addition to previous observations [35] and em-
phasizes the role of the ME as an integral component of energy balance regulation. Highly specialized
neuronal subsets that define the two branches of the melanocortin system are located in close vicinity
to the ME. This anatomical feature allows the ARC neurons to rapidly engage in electrophysiological
changes in response to the entry of circulating hunger and satiety hormones, hence their appellation of
“first order”neurons (Fig. 1).
Among the most well-characterized “first order neurons”are the neurons that make the neuro-
peptide Y (NPY) and the Agouti-related protein (AgRP) and the neurons that produce the pro-
opiomelanocortin (POMC) and cocaine-and amphetamine-related transcript (CART). NPY, initially
discovered by Tatemoto et al. [36], was later on found to be a powerful stimulator of feeding [37]. AgRP
was discovered as an inverse agonist for the melanocortin receptors [38–40]. Co-localization of the two
peptide was described soon after [41], and NPY and AgRP were found to be present in the same pro-
cesses arising from hunger-associated neurons. In addition, these neurons release gamma amino-
butyric acid (GABA) establishing further the inhibitory action of NPY/AgRP neurons [42].
The neighboring POMC neurons are intermingled with NPY/AgRP neurons. POMC neurons are of
mixed excitatory and inhibitory nature and release the neurotransmitters glutamate and GABA [43] as
well as posttranslational products of the POMC peptide, including
a
,
b
,
g
-melanocyte-stimulating
hormone (MSH) and the adrenocorticotropic hormone (ACTH). POMC and NPY/AgRP neurons have
reciprocal antagonistic actions; GABAergic outputs from the NPY/AgRP neurons synaps onto POMC
neurons [44–48], moreover the release of NPY by NPY/AgRP neurons leads to activation of the Gi-
coupled NPY-Y1 receptor located onto POMC. Both neuronal population share common “second-
order”targets in the CNS that are found not only in the paraventricular (PVN), ventromedial (VMH),
dorsomedial (DMH) and lateral (LH) nuclei of the hypothalamus but also in extra hypothalamic targets
such as the nucleus of the tractus solitarii (NTS), the parabrachial nucleus (PBN) [16] or the intermedio-
lateral cell column (IML) [49]. In these structures, the release of
a
-MSH by POMC neurons initiates the
anorectic/catabolic melanocortin signaling cascade through the binding of
a
-MSH to the Gs-coupled
melanocortin receptor (MCR). The MCRs (MCR1 to MCR5) are G protein-coupled receptors distrib-
uted throughout the body, with MC3R and MC4R having theirexpression restricted to the CNS [50,51].
Conversely, during energy deprivation electrophysiological NPY/AgRP neurons will increase their firing
leading to enhanced release of AgRP, which will oppose
a
-MSH binding in postsynaptic targets [52]
(Fig. 1).
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737 729

The PVN represents a very important second order structure downstream of ARC neurons outputs.
The PVN contains several key neuroendocrine neurons including oxytocin (OT), vasopressin,
corticotropin-releasing hormone (CRH) and thyrotropin-releasing hormone-synthesizing neurons
(TRH). In addition to this endocrine function of many PVN neurons, several pre-autonomic neurons
that project to the dorsal motor nucleus of the vagus nerves (DMX) and as well as to the IML allow the
PVN to directly contribute to both parasympathetic and sympathetic output as well. In turn the PVN
receives input from the NTS and the trigeminal pars caudalis. The PVN receives also inputs from the
PBN, DMN and LH. Finally, intra-PVN neuronal encoding is under the control of circadian clock signals
emanating from the suprachiasmatic nucleus (SCN) [53].
MC4R is expressed onto CRH, TRH and OTcontaining neurons [50]. Intra-PVN competition between
AgRP and
a
-MSH tone will directly impact onto energy conservation. During times of plenty
a
-MSH
binding to MCR will positively regulate both the hypothalamic–pituitary–thyroid (HPT) and hypo-
thalamic–pituitary–adrenal (HPA) axis, while the fasting-mediated increase of AgRP release was shown
to be instrumental in the adaptive response of the HPT- and HPA-axis during negative energy balance
[54,55]. NPY/AgRP and POMC neurons are hence considered as the two opposed branches of the
“melanocortin system”,defining a paradigmatic antagonistic regulation of nutrient intake and energy
expenditure. Mutations in the melanocortin signaling pathway including MC3R or MC4R null mutants
[56] or any enzyme involved in the processing of melanocortin peptide, together with ectopic
expression of the MCR antagonist agouti Agouti yellow (A
y
)[57], result in increased feeding and
decreased energy expenditure and invariably morbid obesity in both humans and experimental ani-
mals [58,59].
ARC neurons also project to PVN-pre-autonomic neurons and a direct projectionwas described for a
subset of POMC neurons onto MC4R positive neurons in the IML (Reference). Altogether these ob-
servations support the implication of the MCR signaling pathway in the autonomic regulation and
predict that both direct ARC-IML and ARC-PVN-IML network might account for autonomic and
metabolic effects of melanocortin signaling molecules.
Recently, a reciprocal control from PVN neurons onto ARC NPY/AgRP neurons was described as a
novel regulatory loop involved in feeding behavior neurons [60]. This complex interplay of MCR-
bearing neuronal subpopulations defines the so-called melanocortin signaling pathway and is a key
neurocircuit for the regulation of the energy balance. Hitherto the ARC neurons have been considered
as a primary integrative structure for circulating signals entering the ME. NPY/AgRP neurons are
classically envisioned as a natural opponent of POMC neurons activity, mostly through their antago-
nistic action onto the melanocortin signaling pathway [61,62]. It is only recently that the integrated role
has expended beyond the strict regulation of food intake, to melanocortin dependent and independent
regulation of peripheral nutrient partitioning.
Arcuate control of peripheral nutrient partitioning
ANS-modulation of peripheral organ activity via efferent nerves is a crucial component of an in-
tegrated adaptive response that is initiated at the level of the brain as a result of the integration of
hormonal and nervous afferent inputs. Appropriate regulation of ANS output onto metabolically active
tissues is required to finely orchestrate inter-organ communication during post-absorptive states.
Virtually every tissue including pancreas, liver, brown adipose tissue, white depot, as well as striated,
cardiac and smooth muscle, intestinal tract and bone tissue receives ANS inputs [9,63]. The increased
sympathetic tone during fasting represents a key mechanism for increased adipose tissue lipolysis and
liver glucose production, as well as decreased beta-cell insulin release. At the opposite increased
parasympathetic activity will promote energy storage and conversion [9,64]. Viral-based tracing
studies have shown that pre-autonomic hypothalamic neurons have a distinct organization according
to their efferent organs [64]. This organization supports the concept that a discrete subset of neurons
could participate in and control the selective ANS outflow to one specific tissue. Both POMC and NPY/
AgRP neurons provide dense synaptic inputs to pre-autonomic structures such as the PVN and a direct
action onto peripheral tissue activity, independent from acute regulation of feeding are now largely
documented. Central manipulation of the melanocortin signaling pathway was shown to affect pe-
ripheral cholesterol and lipid metabolism independent from food intake [65–67]. In these experiments
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737730

central blockade of MCR signaling, in otherwise pair-fed conditions, led to increased lipid synthesis and
storage in the white adipose (WAT) and increased insulin-stimulated glucose uptake in the WAT, but
resulted in decreased glucose metabolism in muscle and BAT [65]. This effect was mediated by the
sympathetic nervous system and associated with increased triglyceride synthesis in and export from
the liver [65]. Conversely central stimulation of MCRs was shown to trigger lipid mobilization in WAT
[65,66,68]. In the same line, central manipulation of MCRs selectively increased high-density-
circulating lipoprotein cholesterol (HDL) through reduced liver uptake [67]. These results were
observed independently form food intake and provide a vivid example of selective substrate changes
induced peripherally (lipid vs carbohydrate) through central manipulation. Fasting represents a
physiological situation in which NPY/AgRP neurons have the highest firing rate and exert a profound
inhibitory tone onto POMC neuron activity, antagonize MCR tone through AgRP release and promote
postsynaptic inhibition through NPY release. A recent study, perfectly in line with the results cited
above showed that during fasting, AgRP release is required to promote the fasting-induced decrease in
hepatic SNS activity and increased liver TG synthesis while central knock-down of AgRP in wild type
animal prevented high fat induced liver steatosis [69]. Central injection of NPY was shown to promote
increased liver very-low-density lipoprotein secretion [70], reduced insulin inhibitory action onto
VLDL secretion [71] and to induce hepatic insulin resistance via sympathetic innervations [71,72]. More
recently, the release of NPY by NPY/AgRP neurons was shown to control tyrosine hydroxylase (TH)
expression in the PVN and other brainstem regions via NPY-Y1 receptor activation. The overall
consequence was to increase BAT thermogenesis [73].
Taken together these studies converge toward the notion that MCR blockage and increased NPY/
AgRP activity will promote a concerted change in liver lipid synthesis and export, via the ANS, asso-
ciated with increased WAT lipid storage and BAT lipid catabolism. These observations also illustrate
how central ANS manipulation can simultaneously enhance substrate synthesis in one tissue (liver) to
promote its catabolism to another tissue (BAT) via the exquisite coordination of effector organs. From a
strict thermodynamic stand point, these changes can be independent from actual nutrient intake and
allow the re-direction of energy fluxes from one tissue to another (Fig. 1).
Neuronal and physiologic timing for ARC neuron activity
When considering the ARC based neurocircuitry, one has to conciliate the proper timingduring which
ARC neuron will fire, the differential delay and action of fast-acting neurotransmitters vs slow-acting
neuropeptides together with the differential activity of ARC neurons during different physiological
conditions.In that view, the criticalrole for GABA releasefrom NPY/AgRP neurons wasprovided by several
independent studies. The strict necessity for AgRP neurons in the maintenance of feeding came from
several studiesusing selective ablation of NPY/AgRP neurons.Acute depletion of NPY/AgRP neurons leads
to profound anorexia [74–76], which was not the consequence of an increased melanocortin tone but
rather involved the sudden loss of GABAergic inputs from NPY/AgRP neurons onto the PBN [77]. GABA-A
receptor agonist delivery specifically into the PBN could prevent anorexia following NPY/AgRP neurons
ablation. This study demonstrated that GABA made by NPY/AgRP neurons was critically required to
maintain feeding in a melanocortin independent manner [78,79]. The same group recently identified
calcitonin gene-related peptide-expressing neurons in the outer external lateral subdivision of the PBN
and their projection to the central nucleus of the amygdala as the neuronal substrate for suppressing
appetite [80].In vivo manipulation of NPY/AgRP neurons through forcedexpression of designer receptors
exclusively activated by designerdrugs (DREADD) orphoto-actiavable channel rhodopsin (ChR2) [81,82]
demonstrated the necessity and sufficiency of NPY/AgRP neurons to promote the full feeding sequence
[83], but also confirmed that acute feeding evoked by NPY/AgRP neurons activation relied on the com-
bination of NPY and GABA mediated inhibition of PVN oxytocin (OT) neurons independently from MCR
inhibition [84]. Interestingly, activation of ARC fibers synapsing in the PVN led to a prolonged asyn-
chronous GABA release at the NPY/AgRP synapse that sustained inhibition for hundreds of milliseconds
after action potential-mediated Inhibitory Postsynaptic Currents (IPSCs) occurred. This timescale is
usually associated with neuromodulation, but was here fully recapitulated by biphasic GABA release.
Hence, NPY/AgRP neurons that synapse onto the PVN can modulate postsynaptic activity two orders of
magnitude longer after action potential occurred [84]. One can therefore envision a delay in fast- and
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737 731

slow-acting neurotransmitters, initiated bychanges in AgRP neuronal activity, but occurring at different
timescales and in different postsynaptic structures. A recent study from Krashes et al. definitively
established that AgRP release by NPY/AgRP neurons is sufficient to induce feeding during a prolonged
period of time, while GABA and NPY co release is critical to induce rapid feedingevents [85].
In addition to the intrinsic nutrient-entrained activity of NPY/AgRP and POMC neurons, the PVN also
integrates circadian-related signals relayed from the SCN [53,64]. SCN input and energy-related input
conveyed to ARC neurons could therefore be instrumental in the control of carbohydrate vs lipid-
substrate production and utilization [86]. For instance SCN input to the PVN was shown to critically
operate SNS-mediated liver glucose production [87] and the simple manipulation of feeding schedule,
with no change in calories consumed, was shown to be sufficient in preventing liver steatosis, obesity
and diabetes associated with high fat feeding [88]. In the latter experiment, animals subjected to
restricted feeding schedules displayed a drastic change in the respiratory quotient (RQ) indicative of
the nature of the substrate being used by an organism (RQ ¼1 for glucose utilization and RQ ¼0.7 for
lipid utilization). The drastic RQ changes indicate a rapid change in peripheral substrate selectivityand
utilization, while the smaller changes in RQ observed in animals fed ad libitum indicate a limited –
essentially gluco-lipidic –metabolic plasticity [88]. Hence, manipulating the timing of nutrient input is
sufficient to change the inter-organ dialog and nutrient partitioning. This manipulation illustrates the
concept that nutrient excess per se, does not necessarily translates into metabolic disease if appropriate
manipulation of nutrient partitioning is applied to optimize nutrient fate and utilization. Finally, the
intrinsic timing during which the ARC neurocircuitry is active, i.e., besides external clues of acute
energy deprivation or nutrient oversupply, is still a matter of debate. In that regards, NPY/AgRP neu-
rons were shown to exhibit pacemaker activity [89] suggesting that their activity might extend beyond
the period of energy deprivation, that is also during the post-prandial period. In that period, the co-
ordinated input from NPY/AgRP neurons to pre-autonomic structures could be instrumental in the
determination of peripheral carbohydrate handling independently from nutrient intake, for instance
through long-lasting GABAergic inhibitory post-synaptic currents (IPSCs)’s at the synaptic button.
One can therefore envision a role for both POMC and NPY/AgRP neurons in a segregated action onto
efferent organ activity, beside their antagonistic regulation of feeding, through the control of ANS
output. Both neuronal networks could then operate on different timescales by the combinatorial use of
slow and fast-acting neurotransmitters and peptides, in a melanocortin dependent and independent
manner to independently affect peripheral nutrient partitioning (carbohydrate vs fat) in a coordinated,
but not necessarily opposite manner.
NPY/AgRP neurons: a central switch for peripheral lipid vs carbohydrate utilization
Using an animal model allowing for the selective neonatal ablation of NPY/AgRP neurons [76,90]
our group substantiated this hypothesis by showing that mice lacking NPY/AgRP neurons from the
perinatal period onwards displaya drastic change in ANS output to peripheral tissues characterized by
a decreased norepinephrine turn-over rate (an indirect readout of sympathetic outflow) onto the
pancreas, liver and white-glycolytic muscle, and an increased norepinephrine turn-over rate in
oxidative fat-burning muscle. When fed a regular chow (carbohydrate rich) diet, animal lacking NPY/
AgRP neurons displayed increased feeding efficiency together with hyperinsulinemia and late onset
obesity. Obesity was not the result of increased caloric intake, but rather involved a shift in substrate
utilization due to simultaneous metabolic changes in a number of peripheral tissues. The RQ revealed a
change in substrate utilization towards lipid oxidation that correlates with enhanced conversion of
carbohydrate in the liver associated with increased TG synthesis and export [91]. These change in RQ
clearly followed a daily pattern with a lowest value at the entry of the dark period, a time at which NPY/
AgRP neurons should be activated by nutrient deprivation. The shift towards lipid-substrate preference
was evidenced by both increased peripheral lipid synthesis and export by the liver, and enhanced lipid-
substrate preference at the level of soleus muscle mitochondrial respiration.
Mice lacking NPY/AgRP neurons were then shifted on a high fat diet in order to challenge the hy-
pothesis that these changes in ANS output would indeed change the global adaptation to lipid fuel.
NPY/AgRP-ablated mice normalized their feed efficiency and had both a paradoxical improvement of
glucose tolerance and a reduction in body weight gain compared to control mice [92,93].
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737732

Furthermore, RQ profile and white adipose tissue (WAT) gain in mice lacking NPY/AgRP neurons fed
a regular chow diet could be selectively normalized through GABA-A receptor agonist treatment. One
can hypothesize that ANS output by NPY/AgRP neurons relies more prominently on the slower, tonic
GABAergic inputs that are long-lasting currents compatible with the partitioning of nutrients that
occurs after the meal, during post-ingestive processes. Thus, beyond their well described implication in
the acute regulation of food intake, NPY/AgRP neurons might also directly regulate the fate of nutrients
once ingested through the orchestration of a coordinated dialog between organs including post-
prandial insulin release from pancreas, nutrient conversion and storage in the liver and adipose tis-
sue, and glucose vs lipid utilization in muscle [65,66,94].
Remarkably, despite marked obesity mice lacking NPY/AgRP neurons retained the ability to expand
their adipose mass and displayed improved glucose tolerance and insulin sensitivity upon high fat
feeding. This counter-intuitive observation can be solved by considering the role of NPY/AgRP neurons,
beyond feeding, as a central switch operating the metabolic balance between carbohydrate and lipid
utilization. In that view, the lack of NPY/AgRP neurons might have translated into better adaptation to
high fat diet and the optimization of carbohydrate vs lipid oxidation [93]. This result is in line with a
recent study showing that Sirt1 invalidation selectively in NPY/AgRP neurons results in a shift in the
overall metabolic profile, the impairment of metabolic adaptation to fasting and a change in ghrelin-
induced excitability [95].
Energy-relevant neurons of the ARC are not limited to the POMC and NPY/AgRP population and
more recently other players have been revealed. GABAergic RIP-Cre neurons are a newly described
population whose activity directly controls energy expenditure but not feeding [96]. In addition, AgRP,
NPY and
a
-MSH containing fibers are distributed widely in the brain [16,46] and the existence of
segregated populations within NPY/AgRP neurons projecting to specific second order targets has been
suggested. Using cell-type-specific neuron manipulation and projection-specific anatomical analysis
Betley et al. found that NPY/AgRP neurons indeed display segregated axonal projections that target
different brain regions and originate from distinct ARC subpopulations among which a subset could
control feeding [97]. It is formally possible that the control of nutrient partitioning independent of food
intake also originates from a segregated, non-feeding related, neuronal network based in the ARC.
ARC control of nutrient partitioning: the link between obesity and obesity-related disorders?
The limit between a healthy obesity and an obesity associated with corollary diseases could be the
result of a defective central switch and aberrant ANS output and nutrient partitioning as stated by the
MONA LISA hypothesis [18,20] and Buijs and Kreier [9]. Proper balancing of ANS tone into peripheral
tissues could be a primary set-point defined during postnatal development to establish the balance
between carbohydrate vs lipid adaptation of an organism for the rest of adulthood. Theweaning period
is a critical period of development in which a switch between essentially lipid nutrients to
carbohydrate-based feeding occurs in a short period of time. One could envision that NPY/AgRP
neurons are critical for this behavioral and metabolic switch from lipid to carbohydrate metabolism.
Indeed NPY/AgRP and POMC network development coincides with weaning in rodents (i.e., the switch
from a lipid-rich milk diet to a carbohydrate-rich solid diet) [98] and the metabolic imprinting of ARC
projections have been shown to rely on the trophic action of leptin [99–101]. Mice lacking NPY/AgRP
neurons could have remained in a “default”metabolic mode in which lipid substrates are still
preferred. In that regards we found that the body weight curves of naïve and NPY/AgRP-ablated mice
were similar when fed a HFD since weaning [92]. Similarly, in adulthood an attempt to modify or
restore ARC mediated coordination of peripheral substrate utilization could be a promising avenue to
treat diabetes, dyslipidemia or cardiovascular diseases independently from actual body weight loss or
feeding manipulation.
Summary
The arcuate nucleus of the hypothalamus contains several neuronal populations that are exquisitely
positioned, close to a blood–brain barrier entry point, to integrate circulating signals of hunger and
satiety. Among ARC neurons, POMC and NPY/AgRP neurons are considered as two opposed branches of
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737 733

the melanocortin signaling pathway. NPY/AgRP neurons are segregated populations that have different
and overlapping projections dedicated to food-related and non-food-related functions [102]. Manip-
ulation of individual neurotransmitters produced by these neurons, as well as in vivo activity have
highlighted their implication in ANS-mediated control of peripheral organ activity. Activity of ARC
neurons can now be considered to extend beyond the strict regulation of feeding to the control of
nutrient storage, utilization and transformation. This fundamental process is a cornerstone to a new
conceptual framework that sees obesity and obesity associated disorders as parallel diseases, inherited
through the simultaneous alteration of discrete hypothalamic neurocircuitry.
Declaration of conflict of interest
The authors declare that the research was conducted in the absence of anycommercial or financial
relationship that could be construed as potential conflict of interest.
Acknowledgments
This work was supported by young investigator ATIP grant from the CNRS, an equipment grant from
the Région Île-de-France, an equipment grant from the University Paris Diderot-Paris 7, a research
fellowship from the Société Francophone du Diabète-Lillyand the grant by the “Agence Nationale de la
Recherche”ANR-09-BLAN-0267-02 and ANR 11 BSV1 021 01. A.J-A. received a National Merit Schol-
arship from the French Department of National Education and Research, a research grant from the
Société Francophone du Diabète and a research grant from the SFNEP-ANTADIR. R.D. received a
research fellowship from the Région Île-de-France. C.C. received a PhD fellowship from the Centre
National la Recherche Scientifique (CNRS) and a research grant from the Société Francophone du Di-
abète-Roche.
References
[1] Molavi B, Rasouli N, Kern PA. The prevention and treatment of metabolic syndrome and high-risk obesity. Curr Opin
Cardiol 2006;21:479–85.
[2] Kahn BB, Flier JS. Obesity and insulin resistance. J Clin Invest 2000;106:473–81.
[3] Kopelman PG, Hitman GA. Diabetes. Exploding type II. Lancet 1998;352(Suppl 4):SIV5.
[4] Mokdad AH, Marks JS, Stroup DF, et al. Actual causes of death in the United States, 2000. JAMA: J Am Med Assoc 2004;
291:1238–45.
[5] Troy S, Soty M, Ribeiro L, et al. Intestinal gluconeogenesis is a key factor for early metabolic changes after gastric bypass
but not after gastric lap-band in mice. Cell Metab 2008;8:201–11.
Practice points
Obesity and obesity-related disorders can be separated through central manipulation of ARC
neurocircuitry.
The establishment and activity of a hypothalamic network can define the equilibrium be-
tween carbohydrate and lipid preference.
Research agenda
Pursue the characterization of neural substrate for central control of nutrient partitioning.
Define the limit of hypothalamic network plasticity in adulthood.
Explore the dichotomy between obesity per se, and obesity-related disorders.
Explore pharmacotherapy that would take into account the internal clock and the proper
timing for neuronal activation.
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737734

[6] Cummings DE, Overduin J, Foster-Schubert KE. Gastric bypass for obesity: mechanisms of weight loss and diabetes
resolution. J Clin Endocrinol Metab 2004;89:2608–15 .
[7] Hickey MS, Pories WJ, MacDonald Jr KG, et al. A new paradigm for type 2 diabetes mellitus: could it be a disease of the
foregut? Ann Surg 1998;227:637–43. discussion 43–4.
[8] Perez-Tilve D, D’Alessio DA, Tschop MH. A sweet spot for the bariatric surgeon. Cell metab 2008;8:177–9.
[9] Buijs RM, Kreier F. The metabolic syndrome: a brain disease? J Neuroendocrinol 2006;18:715–6.
[10] Girardet C, Butler AA. Neural melanocortin receptors in obesity and related metabolic disorders. Biochim Biophys Acta
2014;1842:482–94.
[11] Anand BK, Brobeck JR. Hypothalamic control of food intake in rats and cats. Yale J Biol Med 1951;24:123–40.
[12] Heatherington AW, Ranson SW. Hypothalamic lesions and adiposity in the rat. Anat Rec 1940;78:149–72.
[13] Elmquist JK, Elias CF, Saper CB. From lesions to leptin: hypothalamic control of food intake and body weight. Neuron
1999;22:221–32.
[14] Hansen FM, Nilsson P, Hustvedt BE, et al. Significance of hyperinsulinemia in ventromedial hypothalamus-lesioned rats.
Am J Physiol 1983;244:E203–8.
[15] Olney JW. Brain lesions, obesity, and other disturbances in mice treated with monosodium glutamate. Science 1969;164:
719–21.
[16] Broberger C, Johansen J, Johansson C, et al. The neuropeptide Y/agouti gene-related protein (AGRP) brain circuitry in
normal, anorectic, and monosodium glutamate-treated mice. Proc Natl Acad Sci U S A 1998;95:15043–8.
[17] Abe M, Saito M, Shimazu T. Neuropeptide Y in the specific hypothalamic nuclei of rats treated neonatally with mono-
sodium glutamate. Brain Res Bull 1990;24:289–91.
*[18] Bray GA. Obesity, a disorder of nutrient partitioning: the MONA LISA hypothesis. J Nutr 1991;121:1146–62.
[19] Zhang Y. Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425–32.
[20] Bray GA, York DA. The MONA LISA hypothesis in the time of leptin. Recent Prog Hormone Res 1998;53:95–117. discussion
8.
[21] Sierra-Honigmann MR, Nath AK, Murakami C, et al. Biological action of leptin as an angiogenic factor. Science 1998;281:
1683–6.
[22] Unger RH, Zhou YT, Orci L. Regulation of fatty acid homeostasis in cells: novel role of leptin. Proc Natl Acad Sci U S A 1999;
96:2327–32.
[23] Friedman JM, Halaas JL. Leptin and the regulation of body weight in mammals. Nature 1998;395:763–70.
[24] Tartaglia LA. The leptin receptor. J Biol Chem 1997;272:6093–6.
[25] Chen H, Charlat O, Tartaglia LA, et al. Evidence that the diabetes gene encodes the leptin receptor: identification of a
mutation in the leptin receptor gene in db/db mice. Cell 1996;84:491–5.
[26] Banks WA. Is obesity a disease of the blood–brain barrier? Physiological, pathological, and evolutionary considerations.
Curr Pharm Des 2003;9:801–9.
[27] Morton GJ, Cummings DE, Baskin DG, et al. Central nervous system control of food intake and body weight. Nature 2006;
443:289–95.
[28] Kreier F, Buijs RM. Evidence for parasympathetic innervation of white adipose tissue, clearing up some vagaries. Am J
Physiol Regul Integr Comp Physiol 2007;293:R548–9. author reply R50–2, discussion R53–4.
[29] Kreier F, Kap YS, Mettenleiter TC, et al. Tracing from fat tissue, liver, and pancreas: a neuroanatomical framework for the
role of the brain in type 2 diabetes. Endocrinology 2006;147:1140–7.
[30] Gao Y, Tschop MH, Luquet S. Hypothalamic tanycytes: gatekeepers to metabolic control. Cell Metab 2014;19:173–5.
*[31] Langlet F, Levin BE, Luquet S, et al. Tanycytic VEGF-A boosts blood–hypothalamus barrier plasticity and access of
metabolic signals to the arcuate nucleus in response to fasting. Cell Metab 2013;17:607–17.
*[32] Schaeffer M, Langlet F, Lafont C, et al. Rapid sensing of circulating ghrelin by hypothalamic appetite-modifying neurons.
Proc Natl Acad Sci U S A 2013;110:1512–7.
[33] Mullier A, Bouret SG, Prevot V, et al. Differential distribution of tight junction proteins suggests a role for tanycytes in
blood–hypothalamus barrier regulation in the adult mouse brain. J Comp Neurol 2010 Apr 1;518(7):943–62.
*[34] Balland E, Dam J, Langlet F, et al. Hypothalamic tanycytes are an ERK-gated conduit for leptin into the brain. Cell metab
2014;19:293–301.
[35] Banks WA. Role of the blood–brain barrier in the evolution of feeding and cognition. Ann New York Acad Sci 2012;1264:
13–9.
[36] Tatemoto K, Carlquist M, Mutt V. Neuropeptide Y –a novel brain peptide with structural similarities to peptide YY and
pancreatic polypeptide. Nature 1982;296:659–60.
[37] Clark JT, Kalra PS, Crowley WR, et al. Neuropeptide Y and human pancreatic polypeptide stimulate feeding behavior in
rats. Endocrinology 1984;115:427–9.
[38] Michaud EJ, Bultman SJ, Klebig ML, et al. A molecular model for the genetic and phenotypic characteristics of the mouse
lethal yellow (Ay) mutation. Proc Natl Acad Sci U S A 1994;91:2562–6.
[39] Ollmann MM, Wilson BD, Yang YK, et al. Antagonism of central melanocortin receptors in vitro and in vivo by agouti-
related protein. Science 1997;278:135–8.
[40] Shutter JR, Graham M, Kinsey AC, et al. Hypothalamic expression of ART, a novel gene related to agouti, is up-regulated in
obese and diabetic mutant mice. Genes Dev 1997;11:593–602.
[41] Hahn TM, Breininger JF, Baskin DG, et al. Coexpression of Agrp and NPY in fasting-activated hypothalamic neurons. Nat
Neurosci 1998;1:271–2.
[42] Horvath TL, Bechmann I, Naftolin F, et al. Heterogeneity in the neuropeptide Y-containing neurons of the rat arcuate
nucleus: GABAergic and non-GABAergic subpopulations. Brain Res 1997;756:283–6.
[43] Hentges ST, Otero-Corchon V, Pennock RL, et al. Proopiomelanocortin expression in both GABA and glutamate neurons.
J Neurosci 2009;29:13684–90.
[44] Horvath TL, Naftolin F, Kalra SP, et al. Neuropeptide-Y innervation of beta-endorphin-containing cells in the rat mediobasal
hypothalamus: a light and electron microscopic double immunostaining analysis. Endocrinology 1992;131:2461–7.
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737 735

[45] Cowley MA, Smart JL, Rubinstein M, et al. Leptin activates anorexigenic POMC neurons through a neural network in the
arcuate nucleus. Nature 2001;411:480–4.
[46] Broberger C, Hokfelt T. Hypothalamic and vagal neuropeptide circuitries regulating food intake. Physiol Behav 2001;74:
669–82.
[47] Williams G, Bing C, Cai XJ, et al. The hypothalamus and the control of energy homeostasis: different circuits, different
purposes. Physiol Behav 2001;74:683–701.
[48] Pinto S, Roseberry AG, Liu H, et al. Rapid rewiring of arcuate nucleus feeding circuits by leptin. Science 2004;304:110–5.
[49] Zhou L, Sutton GM, Rochford JJ, et al. Serotonin 2C receptor agonistsimprove type 2 diabetes via melanocortin-4 receptor
signaling pathways. Cell Metab 2007;6:398–405.
[50] Liu H, Kishi T, Roseberry AG, et al. Transgenic mice expressing green fluorescent protein under the control of the
melanocortin-4 receptor promoter. J Neurosci 2003;23:7143–54.
[51] Mountjoy KG, Wong J. Obesity, diabetes and functions for proopiomelanocortin-derived peptides. Mol Cell Endocrinol
1997;128:171–7.
[52] Cone RD. Anatomy and regulation of the central melanocortin system. Nat Neurosci 2005;8:571–8.
[53] Buijs RM, Hermes MH, Kalsbeek A. The suprachiasmatic nucleus-paraventricular nucleus interactions: a bridge to the
neuroendocrine and autonomic nervous system. Prog Brain Res 1998;119:365–82.
[54] Fekete C, Legradi G, Mihaly E, et al. alpha-Melanocyte-stimulating hormone is contained in nerve terminals innervating
thyrotropin-releasing hormone-synthesizing neurons in the hypothalamic paraventricular nucleus and prevents fasting-
induced suppression of prothyrotropin-releasing hormone gene expression. J Neurosci 2000;20:1550–8.
[55] Lechan RM, Fekete C. The TRH neuron: a hypothalamic integrator of energy metabolism. Prog Brain Res 2006;153:209–35.
[56] Huszar D, Lynch CA, Fairchild-Huntress V, et al. Targeted disruption of the melanocortin-4 receptor results in obesity in
mice. Cell 1997;88:131–41.
[57] Miltenberger RJ, Mynatt RL, Wilkinson JE, et al. The role of the agouti gene in the yellow obese syndrome. J Nutr 1997;
127:1902S–7S.
[58] Hinney A, Schmidt A, Nottebom K, et al. Several mutations in the melanocortin-4 receptor gene including a nonsense and
a frameshift mutation associated with dominantly inherited obesity in humans. J Clin Endocrinol Metab 1999;84:1483–6.
[59] Krude H, Schnabel D, Luck W, et al. Implications of the phenotype of POMC deficiency for the role of POMC-derived
peptides in skin physiology. Ann N Y Acad Sci 1999;885:419–21.
[60] Krashes MJ, Shah BP, Madara JC, et al. An excitatory paraventricular nucleus to AgRP neuron circuit that drives hunger.
Nature 2014;507:238–42.
[61] Beck B. KO’s and organisation of peptidergic feeding behavior mechanisms. Neurosci Biobehav Rev 2001;25:143–58.
[62] Kalra SP, Dube MG, Sahu A, et al. Neuropeptide Y secretion increases in the paraventricular nucleus in association with
increased appetite for food. Proc Natl Acad Sci U S A 1991;88:10931–5.
[63] Kalsbeek A, Bruinstroop E, Yi CX, et al. Hypothalamic control of energy metabolism via the autonomic nervous system.
Ann N Y Acad Sci 2010;1212:114–29.
*[64] Buijs RM, la Fleur SE, Wortel J, et al. The suprachiasmatic nucleus balances sympathetic and parasympathetic output to
peripheral organs through separate preautonomic neurons. J Comp Neurol 2003;464:36–48.
*[65] Nogueiras R, Wiedmer P, Perez-Tilve D, et al. The central melanocortin system directly controls peripheral lipid meta-
bolism. J Clin Invest 2007;117:3475–88.
*[66] Nogueiras R, Lopez M, Dieguez C. Regulation of lipid metabolism by energy availability: a role for the central nervous
system. Obes Rev 2010;11:185–201.
[67] Perez-Tilve D, Hofmann SM, Basford J, et al. Melanocortin signaling in the CNS directly regulates circulating cholesterol.
Nat Neurosci 2010;13:877–82.
[68] Brito MN, Brito NA, Baro DJ, et al. Differential activation of the sympathetic innervation of adipose tissues by melano-
cortin receptor stimulation. Endocrinology 2007;148:5339–47.
[69] Warne JP, Varonin JM, Nielsen SS, et al. Coordinated regulation of hepatic energy stores by leptin and hypothalamic
agouti-related protein. J Neurosci Off J Soc Neurosci 2013;33:11972–85.
[70] Stafford JM, Yu F, Printz R, et al. Central nervous system neuropeptide Y signaling modulates VLDL triglyceride secretion.
Diabetes 2008;57:1482–90.
[71] van den Hoek AM, Voshol PJ, Karnekamp BN, et al. Intracerebroventricular neuropeptide Y infusion precludes inhibition
of glucose and VLDL production by insulin. Diabetes 2004;53:2529–34.
[72] van den Hoek AM, van Heijningen C, Schroder-van der Elst JP, et al. Intracerebroventricular administration of neuro-
peptide Y induces hepatic insulin resistance via sympathetic innervation. Diabetes 2008;57:2304–10.
[73] Shi YC, Lau J, Lin Z, et al. Arcuate NPY controls sympathetic output and BAT function via a relay of tyrosine hydroxylase
neurons in the PVN. Cell Metab 2013;17:236–48.
[74] Gropp E, Shanabrough M, Borok E, et al. Agouti-related peptide-expressing neurons are mandatory for feeding. Nat
Neurosci 2005;8:1289–91.
[75] Bewick GA, Gardiner JV, Dhillo WS, et al. Post-embryonic ablation of AgRP neurons in mice leads to a lean, hypophagic
phenotype. FASEB J 2005;19:1680–2.
[76] Luquet S, Perez FA, Hnasko TS, et al. NPY/AgRP neurons are essential for feeding in adult mice but can be ablated in
neonates. Science 2005;310:683–5.
[77] Wu Q, Howell MP, Cowley MA, et al. Starvation after AgRP neuron ablation is independent of melanocortin signaling. Proc
Natl Acad Sci U S A 2008;105:2687–92.
[78] Wu Q, Boyle MP, Palmiter RD. Loss of GABAergic signaling by AgRP neurons to the parabrachial nucleus leads to star-
vation. Cell 2009;137:1225–34.
[79] Wu Q, Palmiter RD. GABAergic signaling by AgRP neurons prevents anorexia via a melanocortin-independent mecha-
nism. Eur J Pharmacol; 2011.
*[80] Carter ME, Soden ME, Zweifel LS, et al. Genetic identification of a neural circuit that suppresses appetite. Nature 2013;
503:111–4.
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737736

[81] Hegemann P, Moglich A. Channel rhodopsin engineering and exploration of newoptogenetic tools. Nat Methods 2011;8:
39–42.
[82] Krashes MJ, Koda S, Ye C, et al. Rapid, reversible activation of AgRP neurons drives feeding behavior in mice. J Clin Invest;
2011.
[83] Aponte Y, Atasoy D, Sternson SM. AGRP neurons are sufficient to orchestrate feeding behavior rapidly and without
training. Nat Neurosci 2011;14:351–5.
[84] Atasoy D, Betley JN, Su HH, et al. Deconstruction of a neural circuit for hunger. Nature 2012;488:172–7.
[85] Krashes MJ, Shah BP, Koda S, et al. Rapid versus delayed stimulation of feeding by the endogenously released AgRP
neuron mediators GABA, NPY, and AgRP. Cell Metab 2013;18:588–95.
[86] Peek CB, Ramsey KM, Marcheva B, et al. Nutrient sensing and the circadian clock. Trends Endocrinol Metab TEM 2012;23:
312–8.
*[87] Kalsbeek A, La Fleur S, Van Heijningen C, et al. Suprachiasmatic GABAergic inputs to the paraventricular nucleus control
plasma glucose concentrations in the rat via sympathetic innervation of the liver. J Neurosci 2004;24:7604–13.
[88] Hatori M, Vollmers C, Zarrinpar A, et al. Time-restricted feeding without reducing caloric intake prevents metabolic
diseases in mice fed a high-fat diet. Cell Metab 2012;15:848–60.
[89] van den Top M, Lee K, Whyment AD, et al. Orexigen-sensitive NPY/AgRP pacemaker neurons in the hypothalamic arcuate
nucleus. Nat Neurosci 2004;7:493–4.
[90] Luquet S, Phillips CT, Palmiter RD. NPY/AgRP neurons are not essential for feeding responses to glucoprivation. Peptides
2007;28:214–25.
[91] Cansell C, Denis RG, Joly-Amado A, et al. Arcuate AgRP neurons and the regulation of energy balance. Front Endocrinol
(Lausanne) 2012;3:169.
[92] Joly-Amado A, Denis RG, Castel J, et al. Hypothalamic AgRP-neurons control peripheral substrate utilization and nutrient
partitioning. EMBO J; 2012.
[93] Varela L, Horvath TL. AgRP neurons: a switch between peripheral carbohydrate and lipid utilization. EMBO J 2012;31:
4252–4.
[94] Yamada T, Oka Y, Katagiri H. Inter-organ metabolic communication involved in energy homeostasis: potential thera-
peutic targets for obesity and metabolic syndrome. Pharmacol Ther 2008;117:188–98.
[95] Dietrich MO, Antunes C, Geliang G, et al. Agrp neurons mediate Sirt1’s action on the melanocortin system and energy
balance: roles for Sirt1 in neuronal firing and synaptic plasticity. J Neurosci 2010;30:11815–25.
[96] Kong D, Tong Q, Ye C, et al. GABAergic RIP-Cre neurons in the arcuate nucleus selectively regulate energy expenditure.
Cell 2012;151:645–57.
*[97] Betley JN, Cao ZF, Ritola KD, et al. Parallel, redundant circuit organization for homeostatic control of feeding behavior. Cell
2013;155:1337–50.
[98] Nilsson I, Johansen JE, Schalling M, et al. Maturation of the hypothalamic arcuate agouti-related protein system during
postnatal development in the mouse. Brain Res Dev Brain Res 2005;155:147–54.
[99] Bouret SG, Draper SJ, Simerly RB. Trophic action of leptin on hypothalamic neurons that regulate feeding. Science 2004;
304:108–10.
[100] Bouret SG, Draper SJ, Simerly RB. Formation of projection pathways from the arcuate nucleus of the hypothalamus to
hypothalamic regions implicated in the neural control of feeding behavior in mice. J Neurosci 2004;24:2797–805.
[101] Bouret SG, Simerly RB. Minireview: leptin and development of hypothalamic feeding circuits. Endocrinology 2004;145:
2621–6.
[102] Dietrich MO, Bober J, Ferreira JG, et al. AgRP neurons regulate development of dopamine neuronal plasticity and
nonfood-associated behaviors. Nat Neurosci 2012;15:1108–10 .
A. Joly-Amado et al. / Best Practice & Research Clinical Endocrinology & Metabolism 28 (2014) 725–737 737