Received: 5 December 2020 | Revised: 6 February 2021 | Accepted: 12 February 2021

DOI: 10.1111/jne.12960

EDITOR INVITED REVIEW

Role of serotonin in body weight, insulin secretion and

glycaemic control

Teodora Georgescu1 | David Lyons2 | Lora K. Heisler3

1
Department of Anatomy, Centre for
Neuroendocrinology, School of Biomedical Abstract
Sciences, University of Otago, Dunedin, Obesity and type 2 diabetes are key healthcare challenges of the 21st century.
New Zealand
2 Subsequent to its discovery in 1948, serotonin (5-­hydroxytryptamine; 5-­HT) has
School of Physiology, Pharmacology &
Neuroscience, University of Bristol, Bristol, emerged as a principal modulator of energy homeostasis and body weight, prompt-
UK
ing it to be a target of weight loss medications (eg, fenfluramine, D-­fenfluramine,
3
Rowett Institute, University of Aberdeen,
Aberdeen, UK
fenfluramine-­phentermine and sibutramine). The potential risk of off-­target effects
led to these medications being withdrawn from clinical use and spurred drug discov-
Correspondence
Lora K. Heisler, Rowett Institute, University
ery into 5-­HT receptor selective ligands. The serotonin 2C receptor (5-­HT2CR) is the
of Aberdeen, Aberdeen, UK. primary receptor through which 5-­HT impacts feeding and body weight and 5-­HT2CR
Email: [email protected]
agonist lorcaserin was released for obesity treatment in 2012. Obese patients with
Funding information type 2 diabetes prescribed medications that produce weight loss commonly observe
Biotechnology and Biological Sciences
Research Council, Grant/Award Number:
improvements in type 2 diabetes. However, recent research has provided compel-
BB/R01857X/1 and BBN017838/1; ling evidence that 5-­HT2CR agonists produce effects on blood glucose and insulin
Diabetes UK, Grant/Award Number:
18/0005884
sensitivity independent of weight loss. As such, neuroactive 5-­HT2CR agonists are a
potential new category of type 2 diabetes medications. 5-­HT is also expressed within
pancreatic β cells, is co-­released with insulin and may have a role in modulating insulin
secretion. This review highlights the latest advances in the function of 5-­HT in body
weight, insulin release and glycaemic control.

KEYWORDS

5-­HT, body weight, glycaemic control, insulin, obesity, serotonin

1 | I NTRO D U C TI O N increase in average body weight is evident in children. From 1975 to

2016, the proportion of obese children and adolescents increased
The serious health consequences of diabetes are illustrated by the from less than 1% to 5.6% in boys and 7.8% in girls.3
statistic that diabetes causes 1 death every 6 seconds and consti- A typical strategy to treat obesity and type 2 diabetes is life-
tutes approximately 15% of human mortality worldwide.1 Obesity style modification, including reducing caloric intake and increasing
is the strongest risk factor for the development of type 2 diabetes, physical exercise. However, this approach is commonly associated
2
accounting for approximately 80% of the risk. Obesity is defined with modest and often transient weight loss, rarely exceeding 10%
as a body mass index (BMI) greater than 30 kg m2 and arises when of body weight.4 Pharmacotherapies have been developed to use in
energy intake exceeds energy expenditure. The number of obese conjunction with lifestyle modification to facilitate successful weight
and overweight adults has almost tripled worldwide since 1975, with loss. In more severe cases of obesity, bariatric surgery may be recom-
3
more than 2.1 billion adults now meeting criteria. A more striking mended. Recent advances in the molecular mechanisms regulating

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2021 The Authors. Journal of Neuroendocrinology published by John Wiley & Sons Ltd on behalf of British Society for Neuroendocrinology

Journal of Neuroendocrinology. 2021;33:e12960. wileyonlinelibrary.com/journal/jne | 1 of 12

https://doi.org/10.1111/jne.12960
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 of 12 | GEORGESCU et al.

satiety and energy homeostasis have provided new targets for drug 5-­hydroxytryptophan. Tryptophan hydroxylase catalyses this reac-
discovery for obesity treatment. The brain has emerged as the mas- tion and is the rate-­limiting enzyme for the pathway.5 There are two
ter orchestrator of food intake and energy balance, integrating func- isoforms of this enzyme: tryptophan hydroxylase 1 (Tph1) expressed
tionally relevant signals within pathways and regulatory nodes to in peripheral tissues and Tph2 expressed in the brain.6 In the sec-
exert coordinated control of whole body metabolism. Accumulating ond and final step, aromatic l-­amino acid decarboxylase converts
evidence over the past decade has also revealed a role for the brain 5-­hydroxytryptophan into 5-­HT.
in glycaemic control. 5-­HT has a broad range of physiological and behavioural func-
The focus of this review is serotonin (5-­
hydroxytryptamine; tions and is present in most species. In mammals, 5-­HT is synthe-
5-­HT), which is a key modulator of food intake and energy me- sised in both the periphery and the central nervous system (CNS). In
tabolism and a target of weight loss medications (eg, fenfluramine, the periphery, 5-­HT is primarily produced by enterochromaffin cells
D-­fenfluramine, fenfluramine-­phentermine, sibutramine and lorca- of the gastrointestinal tract where it regulates intestinal motility.7
serin). Obese patients with type 2 diabetes prescribed medications 5-­HT has also been reported in other organs and tissues and has
that produce weight loss commonly observe improvements in type been linked to the function of brain, liver, bone, mammary glands
2 diabetes. Recent evidence indicates that 5-­HT weight loss med- and pancreatic β cells.8-­10
ications also directly improve type 2 diabetes without weight loss
by reducing hepatic glucose production and increasing insulin sen-
sitivity. 5-­HT is also expressed within the pancreatic β cells, being 3 | 5 - ­H T R E LE A S E A N D M E TA B O LI S M
co-­released with insulin, and may have a role in modulating insulin
secretion. These topics are discussed below. Once synthesised, 5-­
HT is stored in vesicles and released
through regulated exocytosis allowing it to bind to 5-­HT recep-
tors (Figure 1). The 5-­HT transporter (5-­HTT or SERT) is respon-
2 | 5 - ­H T S Y NTH E S I S sible for the reuptake of 5-­HT. Metabolism of 5-­HT is carried out
in two steps, with the first being oxidative deamination by mono-
The synthesis of the monoamine transmitter 5-­HT is carried out in amine oxidase A, yielding 5-­hydroxyindole-­3-­acetaldehyde. In the
a two-­step process that requires the essential dietary amino acid second step, aldehyde dehydrogenase catalyses the oxidation of
tryptophan. The first step involves the conversion of tryptophan to 5-­hydroxyindole-­3-­acetaldehyde to 5-­hydroxyindoleacetic acid.

Tryptophan
Tryptophan
Pre-synaptic neurone hydroxylase
5-hydroxy-trytophan
(5-HTP)
Aromatic amino acid
decarboxylase
5-hydroxytryptamine
(5-HT)

5-HT1R
autoreceptors
5-HTT
F I G U R E 1 Serotonin
5-HT1R
5-HT Post-synaptic (5-­hydroxytryptamine; 5-­HT) synthesis
5-HT5R 5-HT2R neurone and neuronal signalling. 5-­HT is
synthesised in the presynaptic neurone
from the amino acid tryptophan.
Gi 5-HT4R Following activation of this neurone, 5-­HT
Gs 5-HT6R Gq/11 is packed into vesicles and released into
Gi/Go 5-HT7R the synaptic cleft, where it signals through
Na+ its receptors (5-­HT1R to 5-­HT 7R). The
PLC signalling is terminated by re-­uptake of
AC
5-­HT from the cleft by 5-­HT transporter
5-HT3R cAMP IP3 DAG (5-­HTT). AC, adenylate cyclase;
DAG, 1,2-­diacylglycerol; IP3, inositol
cellular response 1,4,5-­trisphosphate; PLC, phospholipase C
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
GEORGESCU et al. | 3 of 12

4 | 5 - ­H T R EC E P TO R S not exclusively, expressed within the CNS, whereas 5-­HT2BRs are

primarily expressed in the periphery.13
Beginning in the 1950s, numerous 5-­HT receptors have been dis- Receptors within the 5-­HT1 family are Gi/o-­coupled and may also
covered in mammals.11 Following their discovery, some receptors be positioned as autoreceptors and heteroreceptors. 5-­HT5Rs are
have been renamed. 5-­HT receptors are now grouped into seven also Gi/o-­coupled and function to hyperpolarise cells. The 5-­HT4R,
families (5-­HT1R to 5-­HT 7R) based on sequence homology and in- 5-­HT6R and 5-­HT 7R families are Gs-­coupled and their activation
tracellular effectors (Figure 2). The change in receptor nomenclature results in cell depolarisation. 5-­HT2Rs couple to Gq/11. In addition
highlights the importance of the awareness of earlier names of a sin- to these canonical signalling pathways, recent evidence suggests
gle receptor when reading and interpreting earlier reports. 5-­HT3Rs that metabotropic 5-­HT receptors also signal through a variety of
are ligand gated cation channels. All the other 5-­HT receptors com- non-­canonical pathways.11 This is worth considering when predict-
prise a G-­protein coupled with an extracellular N-­terminus, seven ing a cellular response following 5-­HT receptor activation. 5-­HT re-
transmembrane domains connected by three extracellular and three ceptors are similarly distributed in vertebrates, with the following
intracellular loops, and an intracellular C-­terminus.11,12 All 5-­HT re- exceptions: rodents do not have 5-­ht1eRs, humans do not express
ceptors are expressed postsynaptically. All 5-­HT receptors are ex- full-­length 5-­ht5bRs, and humans and rodents exhibit a different pat-
pressed in both the brain and periphery to varying degrees, with the tern of 5-­HT3R and 5-­HT6R distribution within the forebrain.11 5-­
following exceptions: 5-­HT2CRs and 5-­HT6Rs are predominantly, if ht1eR and 5-­ht5bR are denoted with lower case appellation because a

5-HT RECEPTOR NOMENCLATURE

1950 D M

1979 5-HT1

1986 5-HT1 5-HT2 5-HT3

5-HT1A 5-HT1B 5-HT1C 5-HT1D 5-HT1E 5-HT2F 5-HT4

1988 HTR1A HTR1Dβ HTR2C HTR1Dα HTR1E HTR1F HTR2A HTR2B HTR3A HTR3B HTR3C,D,E HTR4 HTR5A HTR5B HTR6 HTR7

2009 5-HT1A 5-HT1B 5-HT1D 5-ht1e 5-HT1F 5-HT2A 5-HT2B 5-HT2C 5-HT3 5-HT4 5-HT5A 5-HT5b 5-HT6 5-HT7

F I G U R E 2 Serotonin (5-­hydroxytryptamine; 5-­HT) receptor discovery and nomenclature. 5-­HT receptors were first discovered in the
1950s and designated D and M. Subsequently, additional receptors have been identified and the receptors have been grouped into seven
families based on sequence homology and effector pathways. Currently, 14 5-­HT receptors have been identified in mammals. Adapted from
Sharp & Barnes11 and Marin et al103
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 of 12 | GEORGESCU et al.

functional response in native cells has not yet been demonstrated.13 In addition to receptor-­
mediated transductional processes,
Isoforms of 5-­HT2CRs, 5-­HT3Rs, 5-­HT4Rs and 5-­HT 7Rs have been receptor-­
independent 5-­
HT signalling may also influence insulin
identified, although the specific function of these isoforms remains secretion. For example, non-­canonical transductional mechanisms
to be clearly established. 5-­HT2CRs are the only 5-­HTR that undergo promote insulin exocytosis from β cells via serotonylation of Rab
RNA editing. proteins. 21 As demonstrated in other tissues, serotonylation relies
on transamination by transglutaminases of 5-­HT to small GTPases
resulting in these being constitutively active. 28 Further supporting
5 | β C E LL 5 - ­H T this receptor-­independent effect of 5-­HT, transglutaminase 2 null
mice are glucose-­intolerant and show decreased levels of insulin
5.1 | 5-­HT and receptor expression following glucose stimulation. 29 Walther et al28 propose a dynamic
model in which progressive insulin and 5-­HT co-­secretion promotes
Evidence of 5-­HT in the pancreatic islets of Langerhans dates to high extracellular 5-­HT concentrations that, via an autocrine feed-
studies carried out more than 50 years ago.14 β cells co-­release 5-­HT back loop, attenuates additional insulin (and 5-­HT) release through
15,16
alongside insulin and ATP. Expression of tryptophan hydroxy- inhibitory 5-­HT1ARs located on β cells. Extracellular 5-­HT is subse-
lase (TPH1) in β cells indicates the capacity for de novo synthesis quently cleared via 5-­HTT. They propose that as extracellular 5-­HT
of 5-­HT in the islets of Langerhans.17-­19 Mice lacking TPH1 only in levels fall and intracellular 5-­HT levels build, insulin (and 5-­HT) se-
β cells are glucose intolerant and secrete less insulin when fed a cretion is once more enhanced via serotonylation (Figure 3).
20
high-­fat diet for 6 weeks. This link is specific to β cells because Although 5-­HT was identified in the islet more than half a cen-
TPH1 knockout in gut enterochromaffin cells does not affect insu- tury ago, the physiological function of 5-­HT co-­release with insulin
lin secretion. 21,22 Human and rodent β cells also express 5-­HTT, the is still largely unknown. Seemingly contradictory reports may be
transporter responsible for 5-­HT reuptake.17,19 Research suggests reconciled by differing functions of distinct 5-­HT receptors in dif-
that 5-­HT acts as an autocrine signal modulating β cell function and ferent cell types and receptor-­independent 5-­HT action. Additional
proliferation20,23-­25 (Figure 3). research is required to clarify the function of 5-­HT in the islet and in
The presence of 5-­HT1ARs, 5-­HT1DRs, 5-­HT1FRs, 5-­HT2BRs, 5-­ the modulation of insulin secretion.
HT3Rs and 5-­HT5ARs within human islets has been revealed via a
revealed via gene and/or protein expression techniques.19,25,26
Different receptors appear to be primarily expressed within specific 6 | B R A I N 5 - ­H T
cell types, with 5-­HT1ARs, 5-­HT1DRs, 5-­HT2BRs and 5-­HT3Rs in β
cells, 5-­HT1FRs in α cells, and 5-­HT5ARs in δ cells.19,25,26 6.1 | 5-­HT and receptor expression

Although comparatively little is known about the function of 5-­HT

5.2 | Insulin secretion and β cell proliferation in pancreatic insulin secretion and β cell proliferation, more intense
research efforts have focussed on the role of 5-­HT in the brain. CNS
5-­HT2BR agonists promote insulin secretion, an effect partly regu- 5-­HT influences a wide variety of processes, such as energy balance,
lated by a rise in intracellular [Ca2+] and enhanced mitochondrial anxiety, mood, migraine, sleep, locomotion, circadian rhythms and
activity. 25 Expression of 5-­HT2BRs in mouse islets increases during aggression, amongst others. This review is focussed on the role of
pregnancy and it serves to enhance β cell proliferation during this CNS 5-­HT in the modulation of feeding behaviour, body weight and
23
period of high insulin demand. Pharmacological inhibition of the glycaemic control.
5-­HT2BR impaired glucose tolerance and reduced the pregnancy-­ 5-­
HT synthesis requires the essential amino acid tryptophan
associated increase in β cell mass, highlighting this receptor as obtained from food. Tryptophan is carried by an active transport
a favourable target for the treatment of gestational diabetes. 23 mechanism into the brain where it is converted into 5-­HT. Because
Regulation of insulin secretion and β cell function has also been tryptophan competes with the other large neutral amino acids (LNAA)
described during lactation, and this phenomenon is dependent for transport into the brain, the ratio of tryptophan to other LNAA is
on 5-­HT signalling at the 5-­HT2BR. 27 However, mice with a β cell-­ an important determinant for 5-­HT synthesis.30 Peripherally derived
specific knockout of 5-­HT2BRs are not glucose intolerant after a 5-­HT does not cross the blood-­brain barrier. However, recent stud-
6-­
week period of high-­
fat diet, whereas 5-­
HT3R knockout mice ies reveal the presence of 5-­HTT in the brain endothelium and have
display diminishes glucose-­stimulated insulin secretion. 20 Glucose-­ linked this to transport of augmented 5-­HT from the brain into the
dependent 5-­HT release from human β cells also acts on α cells to in- blood, but not vice-­versa.31-­33 Brain 5-­HT is produced in a collection
hibit glucagon secretion via action at the 5-­HT1FR.19 Administration of midline brainstem nuclei called raphe, meaning a ‘seam connect-
of a 5-­HT1FR agonist lowered glucagon secretion and improved gly- ing two halves’. The raphe nuclei are comprised of the raphe pallidus
caemic levels in diabetic mice.19 These findings suggest that pan- (RPa; B1), raphe obscurus (ROb; B2), raphe magnus (RMg; B3), raphe
creatic 5-­HT plays a significant role in insulin secretion via action at pontis (RPn; B4), median raphe (RMn; B5, B8), dorsal raphe (DRN;
different 5-­HT receptors expressed on distinct cell types. 20,23 B6, B7) and the supralemniscal nucleus (SLN; B9). Unlike the other
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
GEORGESCU et al. | 5 of 12

K+ Ca2+
5-HT1D 5-HT2B
5-HT3
∆ψ

VDCC
KATP Channel
[Ca2+]
TGase
inactive
ATP

TGase
active

Glc Rab
[Gluc]i Serotonylation
GDP
of Rab proteins
Glut2

5-HT
VMAT2 Effectors

Rab
[5-HT]i GTP
GTP
GDP
cAMP
SERT 5-HT1A

Insulin
exocytosis
Autocrine
[5-HT]e Paracrine

[5-HT]e >> [5-HT]i Insulin release

[5-HT]i >> [5-HT]e Insulin release

F I G U R E 3 Schematic of model of exocytosis of β granules from β cells in response to serotonin (5-­hydroxytryptamine; 5-­HT). Glucose
(Glu) is transported into the cell, oxidised by the mitochondria to produce ATP, which then causes the closure of the K ATP channel and leads
to depolarisation of the membrane. This opens the voltage-­dependent Ca2+ channel and the influx of Ca2+ ions activate transglutaminase
(TGases) and serotonylate a number of proteins. Amongst these, Rab3a and Rab27a are essential for insulin release. Thereby their activation
causes co-­secretion of 5-­HT and insulin. By signalling at the 5-­HT1AR, high extracellular 5-­HT ([5-­HT]e) reduces further secretion of
insulin. However, this effect is dampened as 5-­HT reuptake occurs via the 5-­HT transporter (5-­HTT). When intracellular 5-­HT ([5-­HT]i)
is substantially higher than extracellular 5-­HT ([5-­HT]e), this triggers another event of insulin secretion by serotonylation. This dynamic
regulation of 5-­HT levels may explain the established oscillation of insulin exocytosis from glucose stimulated β cells. Adapted from
Paulmann et al. 21GLUT2, glucose transporter 2; 5-­HTT, serotonin transporter; VDCC, voltage-­dependent Ca2+ channel; VMAT, vesicular
monoamine transporter

raphe nuclei, the SLN is not expressed on the midline. These nuclei terminals to some extent. The cluster of caudal raphe nuclei (B1-­B 4)
were previously designated B1-­B9 and project to distinct parts of are responsible for descending 5-­HT projections that innervate re-
the brain such that almost all brain regions are innervated by 5-­HT gions including the cerebellum, midbrain, pons, medulla and spinal
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 of 12 | GEORGESCU et al.

cord, whereas the more rostral raphe nuclei (B5-­B9) give rise to as- 5-­HT reuptake (selective serotonin reuptake inhibitor; SSRI) and
cending projections that target areas such as the cortex, hippocam- stimulated 5-­HT release.
pus, thalamus, hypothalamus, striatum and amygdala34-­37 (Figure 4).
The DRN is the largest raphe nuclei and distinct DRN subdivisions
are relevant to energy homeostasis.34 The arcuate nucleus of the 7 | C LI N I C A L A N D PR EC LI N I C A L 5 - ­H T
hypothalamus (ARC) melanocortin pro-­
opiomelanocortin (POMC) M E D I C ATI O N S
and agouti-­related peptide (AgRP) neuronal populations are primary
regulators of energy homeostasis and are innervated by 5-­HT fi- 7.1 | Medications augmenting 5-­HT bioavailability
bres.36,38-­41 The melanocortin system is required for 5-­HT weight
loss medications to produce their therapeutic effect.42-­47 7.1.1 | Fenfluramine, d-­fenfluramine and
fenfluramine-­phentermine (fen-­phen)

6.2 | Food intake and body weight Medications specifically targeting 5-­HT for obesity treatment were
fenfluramine (Pondimin) and the d-­
enantiomer of fenfluramine (d-­
Early research with 5-­HT revealed a strong inverse relationship be- fenfluramine, dexfenfluramine, Redux), which stimulate the release of
tween 5-­HT and food intake. Although brain 5-­HT accounts for a 5-­HT and block its reuptake. Fenfluramine and d-­fenfluramine alone
relatively small percentage of 5-­HT (approximately 3%), research in and in combination with phentermine (fen-­phen) were used to treat
the 1970s revealed that brain 5-­HT is necessary for normal appetite human obesity until the late 1990s when they were withdrawn from
and body weight regulation. Specifically, in back-­to-­back publica- clinical use as a result of valvular side effects in a small group of obese
tions in Science in 1976, both Breisch et al48 and Saller and Stricker49 patients.50 Earlier this year, the US Food and Drug Administration
reported that selectively decreasing brain 5-­HT synthesis promoted (FDA) and European Medicines Agency granted or recommended
weight gain and the development of obesity. Early pharmacological granting the marketing of fenfluramine (Fintepla) for the treatment of
strategies to elevate brain 5-­HT to treat obesity selectively blocked seizures associated the severe form of epilepsy Dravet syndrome.

V1
Cortex

ACC

FC Hippocampus
Cerebellum
3V

Septum PAG Aq

DRN
Thalamus B6,B7
BNST 4V

RPn
NAcc PVH RMn B4 NTS
VTA B8,B5

Hypothalamus SLN ROb

3V B9 B2 SC
ARC
RMg
B3 RPa
B1

F I G U R E 4 Schematic of raphe nuclei localisation and selected projections. 5-­HT neurones within the raphe nuclei cluster along the
midline and broadly innervate the CNS. Displayed is a schematic of a sagittal midline brain view illustrating the relative localisation of the
raphe nuclei and some selected projections. 3V, third ventricle; 4V, fourth ventricle; ACC, anterior cingulate cortex; Aq, aqueduct; ARC,
arcuate nucleus of the hypothalamus; BNST, bed nucleus of the stria terminalis; RPa, raphe pallidus nucleus, B1; ROb, raphe obscurus
nucleus, B2; RMg, raphe 2magnus nucleus, B3; RPn, raphe pontis nucleus, B4; RMn, median raphe nucleus, B5, B8; DRN, dorsal raphe
nucleus, B6, B7; SLN, supralemniscal nucleus, B9; FC, frontal cortex; NAcc, nucleus accumbens; NTS, nucleus of the solitary tract; PAG,
periaqueductal grey; PVH, paraventricular hypothalamic nucleus; SC, spinal cord; V1, primary visual cortex; and VTA, ventral tegmental area
(adapted from Lam & Heisler 2007)36
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
GEORGESCU et al. | 7 of 12

The primary therapeutic mechanism of fenfluramine and d-­ associated with a lowering of energy expenditure. Although one
fenfluramine in treating obesity is achieved via decreasing food in- study reported that sibutramine partially attenuated the decline in
take.51 However, some evidence suggests that d-­fenfluramine may energy expenditure with weight loss in obese people (P = 0.09), 62
also increase resting and postprandial energy expenditure in obese other studies report no effect of sibutramine on energy expen-
patients.52 Fenfluramine was also found to have effects on glucose diture in people. 63,64 Alongside its anorectic effects, sibutramine
tolerance. For example, following a 7-­day treatment with fenflu- action also improves glycaemic control in obese people with
ramine, blood glucose of obese patients with type 2 diabetes was type 2 diabetes. 58 A randomised prospective placebo-­controlled
53
significantly lowered. The glycaemic effect may be at least in part double-­b linded study in patients with type 2 diabetes prescribed
independent of the anorexigenic effects of fenfluramine.53-­55 metformin indicated that patients treated with sibutramine
Following the withdrawal of fenfluramine, d-­fenfluramine and showed significant weight loss and improved glycaemic control,
fen-­p hen for weight loss, research effort turned to establishing and demonstrated fasting blood glucose in parallel with weight
the receptors underpinning both the off-­target and therapeutic loss over the 12-­m onth period. 65 Sibutramine treated patients also
effects. The receptor implicated in the adverse effect is peripheral had reduced levels of A1c (HbA1C), a glycated form of haemoglo-
5-­HT2B Rs. Compelling evidence indicating that brain 5-­HT2CRs are bin that is used as a clinical biomarker for diabetes. 65 Similar to
required for the therapeutic effect was provided by transgenic d -­f enfluramine, 5-­HT2CRs and POMC are implicated in the obesity
mouse studies. Specifically, 5-­HT2CR knockout mice do not re- therapeutic effect of sibutramine.43,66,67
42,56
spond to the anorectic effect of d -­f enfluramine. Given that Medications broadly increasing 5-­
HT bioavailability are ef-
the 5-­HT2CRs are expressed in a variety of brain regions where fective for treating human obesity and commonly improve insu-
they produce an array of behavioural effects, additional studies lin sensitivity and glycaemic control in obese patients with type
aimed to define the specific subset of 5-­HT2CRs and the circuit 2 diabetes by promoting weight loss. However, the risk of off-­
through which d-­fenfluramine reduces feeding. 5-­HT2CRs are co-­ target effects was deemed to outweigh the therapeutic benefit
expressed with ARC POMC neurones and d-­fenfluramine and 5-­ of fenfluramine, d -­
fenfluramine, fen-­p hen and sibutramine and
HT2CR agonists increase POMC neurone activity.47 In mice null they were subsequently withdrawn from human treatment. Other
for 5-­HT2CRs, re-­expression of 5-­HT2CRs only in POMC neurones SSRIs are still commonly prescribed for conditions such as anxi-
restored the anorectic effect of d-­fenfluramine.42 Illustrating the ety and depression (eg, fluoxetine/Prozac). Obesity increases the
necessity of POMC peptides the effect of d-­fenfluramine on food risk of heart disease and stroke. This suggests that a subset of
intake, pharmacological (SHU9119) or genetic (Mc4r null mice) obese people with increased risk of heart disease and stroke may
blockade of one of the receptor targets for POMC peptides, the be more susceptible cardiovascular complications associated with
melanocortin4 receptor (MC4R), attenuated the effects of d -­ 5-­HT medications.
42,47
fenfluramine. The circuit was further clarified through trans- Genetic and pharmacological research indicate that the ther-
genic MC4R re-­expression, which revealed that MC4Rs within the apeutic effect of fenfluramine, d-­
fenfluramine and sibutramine
paraventricular nucleus of the hypothalamus (PVH) are sufficient is primarily achieved via 5-­HT action at brain 5-­HT2CRs and this
for d-­fenfluramine food intake suppression.42 In summary, a key spurred drug discovery efforts toward the development of selective
mechanism through which d -­
fenfluramine decreases feeding is 5-­HT2CR agonists. Of the 5-­HT receptors, only 5-­HT2CR null mice ex-
via activation of the subset of 5-­HT2CRs co-­expressed with ARC hibit hyperphagia and obesity and develop characteristics of type 2
POMC, which stimulates the release of POMC peptides to act at diabetes.68,69 5-­HTR knockout studies also highlight the importance
MC4Rs within the PVH. of the 5-­HT2BRs in cardiac structure and function.70 The only 5-­HT
receptor that is fatal if knocked out in mice is the 5-­HT2BR.71 Thus,
a challenge during this drug discovery period was to develop a com-
7.1.2 | Sibutramine pound with high affinity for the 5-­HT2CRs but with low affinity for
the 5-­HT2BRs.
Sibutramine (Meridia) is a 5-­HT and noradrenaline reuptake inhibi-
tor that was prescribed for obesity treatment between 1997 and
2010 and typically produced 5%-­7% weight loss. 57,58 Patient stud- 7.2 | 5-­HT2CR agonists
ies indicate that the anorectic effects of sibutramine are main-
tained during 10 months of treatment. 59 However, sibutramine 7.2.1 | Preclinical 5-­HT2CR agonists
was withdrawn from clinical use in 2010 as a result of reports as-
sociating it with increased risk of myocardial infarction and stroke Preclinical meta-­chlorophenylpiperazine (mCPP) provided some of
in obese patients. 60 the first insights into the potential of a 5-­HT2CR agonist for obesity
Similar to fenfluramine, sibutramine decreases feeding and treatment. However, this compound was unlikely to be suitable for
enhances satiety. 59 In rats, it has also been reported to exert ef- widespread human use because it has affinity for multiple recep-
fects on energy expenditure 61; however, in humans, the effect tors in addition to the 5-­HT2CR (pEC50 7.00), including the 5-­HT2BR
on energy expenditure is not apparent. Weight loss is typically (pEC50 7.2). Similar to former 5-­HT weight loss medications, mCPP
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 of 12 | GEORGESCU et al.

reduces food intake and body weight in rodents 72

and humans.73 7.2.2 | Weight loss medication lorcaserin
This effect appears to be primarily achieved via the 5-­HT2CRs given
that mCPP does not reduce feeding in mice lacking 5-­HT2CRs.68 The USA FDA approval of the selective 5-­HT2CR agonist lorcaserin
The first clear indication of the potential of 5-­HT medications in 2012 represented the first weight loss medication approved by
in the direct improvement of glycaemic control was observed with the FDA in over a decade. Lorcaserin is a full agonist at the 5-­HT2CR,
mCPP. In a dietary-­induced obese mouse model of type 2 diabe- where it exhibits 104-­fold selectivity over the 5-­HT2BR.78 Lorcaserin
tes, a single treatment with mCPP rapidly improved glucose and improves obesity by reducing food intake.44,45,78,79 It does not alter
74
insulin tolerance over the next 2 hours. Significant improve- energy expenditure.80 In addition to homeostatic feeding, lorcaserin
ments in glycaemia were achieved at doses of mCPP that were also decreases operant responding for food reward, impulsivity and
too low to reduce feeding. Specifically, improvements in glycae- binge eating.81-­84 In clinical trials, lorcaserin reduced body weight by
mia were achieved at 1 mg/kg, whereas more than double of this approximately 5% in obese patients.85-­87 Lorcaserin's weight loss is
dose (2.5 mg/kg) was required to impact feeding behaviour. A sin- characterised by a reduction of fat mass but not lean mass.78 In an-
gle pre-­t reatment of mCPP prior to a bolus of insulin also signifi- other study, lorcaserin treatment led a reduction in central adiposity
74
cantly increased insulin sensitivity in muscle and liver. Similarly, compared to placebo.88 Trunk fat mass has been associated with a
prolonged treatment with mCPP improved glucose and insulin higher risk of obesity-­associated comorbidities such as cardiovascu-
tolerance, reduced hyperinsulinaemia and decreased enzymes lar diseases, further highlighting the potential benefits of lorcaserin
associated with hepatic glucose production.74 These beneficial treatment. These cardiometabolic health improvements associated
effects were achieved at a dose of mCPP that was too low to im- with lorcaserin treatment have been shown to be dependent on a
pact feeding or body weight.74 These findings illustrated the first reduction in total atherogenic lipoproteins.89 A large scale, 5-­year
dissociation between the effects of a 5-­HT drug on appetite/body cardiovascular analysis was performed in the CAMELLIA-­TIMI 61
weight and glycaemic control/insulin sensitivity. trial described below.
Research investigating the mechanism of action of mCPP and Lorcaserin produces an improvement in glycaemic control in
other preclinical 5-­HT2CR agonists (eg BVT.X and WAY161,503) on mouse models of type 2 diabetes independent of weight loss.45
appetite and glycaemic control/insulin sensitivity also point to the Indeed, the doses of drug required to improve glycaemia (4 mg/
melanocortin system. WAY161,503 increases the activity of ARC kg) were approximately half that required to decrease food intake
43
POMC neurones. Similarly, mCPP increases the activity of ARC (7.5 mg/kg).45 These benefits of lorcaserin were achieved via reduc-
POMC cells and this is achieved via a postsynaptic mechanism reli- ing hepatic glucose production and increasing insulin sensitivity.45
47,75
ant on the presence of TRPC channels. Illustrating the necessity Lorcaserin does not appear to influence insulin secretion in mice.45
of the subset of 5-­HT2CRs specifically in POMC neurones, mCPP Demonstrating the translational potential of lorcaserin for the treat-
does not impact feeding in mice selectively lacking 5-­HT2CRs only ment of type 2 diabetes, clinical trials in obese patients with type 2
in POMC neurones.76 D'Agostino et al44 revealed that POMC pep- diabetes revealed that lorcaserin improved HbA1C levels and fast-
tides are required for the complete anorectic effect of WAY161,503 ing glucose levels86 irrespective of the weight loss.90 Data obtained
using both a transgenic knockout and a brain region selective Pomc from the approximately 12,000 patient CAMELLIA-­TIMI 61 trial in-
CRISPR/Cas9 knockdown approach in mice. Similarly, BVT.X is inef- dicated that lorcaserin significantly reduced the risk of developing
fective at reducing food intake in MC4R knockout mice.77 diabetes by 23% in participants without prediabetes or diabetes and
As noted above, whole brain 5-­HT2CR knockout mice exhibit by 19% in participants with prediabetes.91 In participants with dia-
obesity and parameters associated with type 2 diabetes (eg, insu- betes, lorcaserin significantly lowered HbA1C and the risk of diabetic
60
lin resistance and impaired glucose tolerance). On a high-­fat diet, microvascular complications.91
60
5-­HT2CR knockout mice also develop hyperglycaemia. Selectively Analysis of the mechanism of action revealed that lorcaserin re-
knocking out 5-­HT2CRs only in POMC neurones does not promote quires POMC to produce its effects on food intake.44 Similalrly, the
obesity, but does cause hyperinsulinaemia, hyperglycaemia, hy- glucoregulatory effects of lorcaserin are absent in brain Pomc null
76
perglucagonaemia and insulin resistance. These findings provide mice, but the specific re-­expression of Pomc only in 5-­HT2CRs neu-
further evidence that 5-­HT2CRs play a key role in glycaemic con- rones is sufficient to restore the glycaemic actions of lorcaserin.45
trol independently of their established effect on appetite and body The effects of lorcaserin on food intake, glycaemia and insulin sen-
weight. Pharmacologically or genetically blocking one of the down- sitivity require downstream MC4Rs.45 The effects on glycaemia and
stream receptor targets of POMC peptides, the MC4R, prevents the insulin sensitivity, but not food intake, require the specific subset of
effects of mCPP on glycaemic control and insulin sensitivity.74 MC4Rs expressed in preautonomic cholinergic neurones in the dor-
These preclinical findings highlighted the potential for a 5-­HT2CR sal vagal complex and spinal cord.45 These data illustrate the disso-
agonist for the treatment of both obesity and type 2 diabetes and ciation of the circuitry through which lorcaserin modulates appetite
indicate that both effects are achieved via the melanocortin system. and blood glucose/insulin sensitivity.
Follow-­up research with 5-­HT2CR agonist lorcaserin, as described With regards to safety, the FDA requested a long-­term trial eval-
below, reveals the different branches of melanocortin circuitry pro- uating the potential cardiovascular effects of lorcaserin as part of a
ducing these effects. postmarketing requirement. The CAMELLIA-­TIMI 61 trial comprised
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
GEORGESCU et al. | 9 of 12

a 5-­year, randomised, double-­blind, placebo controlled, multicentre may impact the efficacy of 5-­HT2CR medications in patients. This
trial using 473 sites in eight countries. Involving approximately possibility remains to be investigated.
12,000 overweight or obese patients with atherosclerotic cardiovas-
cular disease or multiple cardiovascular risk factors, CAMELLIA-­TIMI
61 comprises the largest cardiovascular outcome study for a weight 8 | CO N C LU S I O N S
loss medication to date. No significant differences between treat-
ment groups were seen for any major adverse events.92,93 Rather, Obesity and type 2 diabetes are currently a growing global health
lorcaserin promoted significant weight loss and provided further concern. The brain, acting as the principal orchestrator of feeding
health benefits, decreasing the risk of the development of diabetes behaviours, hosts a complex array of networks regulating energy
and improving diabetes and diabetic microvascular complications in homeostasis. Key amongst these are 5-­HT circuits and, correspond-
participants with type 2 diabetes.92,93 In a decision that is difficult ingly, various therapeutics target components of this system, par-
to interpret, in 2020, the FDA concluded that the potential risks of ticularly the 5-­
HT2CRs. Weight loss commonly improves type 2
lorcaserin outweigh its benefits. Despite the finding that lorcaserin diabetes and, as such, 5-­HT obesity drugs would be expected to im-
did not increase the incidence of any serious adverse events, the prove glycaemic control and insulin sensitivity in patients with type
FDA focussed on a small numerical difference in the incidence of 2 diabetes. However, recent research has provided compelling evi-
malignancies during the CAMELLIA-­TIMI 61 trial: 7.7% of lorcaserin dence that 5-­HT2CR agonists produce effects on blood glucose and
treated patients compared to 7.1% of placebo treated patients were insulin sensitivity independently of weight loss. As such, this opens
diagnosed with some form of cancer over the 5-­year period. Given up a potentially new category of type 2 diabetes medications, com-
that obesity is established to increase the risk of cancer (by approxi- prising a class of drugs that target the brain.
mately 7.1% over 5 years as suggested by the placebo-­treated group) Early 5-­HT weight loss medications (fenfluramine, d-­fenfluramine
and that there was no statistical difference between the lorcaserin and sibutramine) alter whole body 5-­HT bioavailability and therefore
and placebo groups with respect to the incidence of malignancies, influence multiple 5-­HT receptors, including those not directly in-
the FDA’s interpretation of the similar rates of malignancies over volved in the regulation of ingestive behaviour and metabolism. Side
the 5 year monitoring period is difficult to understand. The man- effects led to their withdrawal and the emergence of the selective
ufacturer of lorcaserin, Eisai, voluntarily withdraw lorcaserin from 5-­HT2CR agonist lorcaserin. Lorcaserin directly targets the principal
the USA market for obesity treatment in February 2020. However, 5-­HT receptor influencing body weight. The therapeutic efficacy
lorcaserin is still marketed in other countries for obesity and, in of lorcaserin is associated with an improved side effect profile at-
consultation with the FDA, lorcaserin is entering into a phase 3 trial tributed to lack of agonism at the 5-­HT2BR.85 The clinical utility of
(MOMENTUM 1) for Dravet syndrome, a severe form of epilepsy medications such as lorcaserin has nonetheless highlighted the need
that first presents in infancy. Mice lacking 5-­HT2CRs are prone to to further understand the mechanisms through which their benefi-
seizures that can be fatal,68 providing a rationale that a 5-­HT2CR ag- cial effect is achieved. Research indicates that lorcaserin requires
onist could be effective for treating seizures associated with Dravet melanocortin pathways to reduce food intake and improve glycae-
syndrome. mic control/insulin sensitivity. Principal MC4Rs controlling feeding
In summary, 5-­HT2CR agonists are effective at improving obesity reside within the paraventricular nucleus of the hypothalamus,97,98
and show therapeutic benefit with respect to glycaemic control and whereas MC4Rs controlling the effects of lorcaserin on glycaemia
insulin sensitivity in obese individuals with prediabetes or type 2 di- and insulin sensitivity reside in the brainstem and spinal cord.45
abetes. Lorcaserin is also effective at reducing the onset of new type Current weight loss medications and those in clinical trials com-
2 diabetes. These effects on diabetes parameters are not dependent bine different ligands to produce a greater therapeutic effect (eg,
upon weight loss, indicating a parallel mechanism of action. Current contrave: bupropion and naltrexone; quismia: phentermine and
frontline diabetes medications target peripheral tissues to produce topiramate; and various emerging gut peptide combinations). Thus,
a therapeutic benefit. Thus, these findings with 5-­HT2CR agonists another future strategy to improve the therapeutic profile of lor-
represent a first indication that the brain may be harnessed to im- HT1BR agonists CP-­
caserin is combination therapy. 5-­ 94253 and
prove type 2 diabetes in humans. Lorcaserin reduces body weight RU-­24969 promote food intake reductions and this effect is blocked
by promoting satiety and also decreases motivation for food reward by pre-­treatment with the 5-­HT1BR antagonists.40,99 Combined ad-
and impulsivity.81,82,94 It is therefore perhaps surprising that the ef- ministration of sub-­anorectic doses of 5-­HT1BR and 5-­HT2CR ago-
fects on body weight are not larger. 5-­HT2CRs are subjected to RNA nists results in a synergistic reduction in food intake in mice100 and
editing events in which adenosine deaminase converts adenosine rats.101 5-­HT1BRs are primarily expressed as heteroreceptors on
residues to inosines. These post-­transcriptional modifications can non-­5-­HT terminals, where they prevent the release of other neu-
impact the efficacy of receptor to G-­protein coupling, suggesting rotransmitters. Evidence suggests that a key mechanism underpin-
an additional mechanism by which 5-­HT signal transduction is reg- ning the anorectic effects of 5-­HT1BRs is via hyperpolarisation of
ulated.95 In a genetically engineered mouse line, increased 5-­HT2CR AgRP neurones and the resultant reduction in the inhibitory tone
RNA editing attenuated the anorectic effect of 5-­HT2CR agonist onto ARC POMC neurones.40,100 The combination of 5-­HT1BR and
WAY161,503.96 It is therefore possible that 5-­HT2CR RNA editing 5-­HT2CR agonists thereby increases the absolute number of POMC
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 of 12 | GEORGESCU et al.

neurones activated compared to each agonist administered alone 5. Grahame-­Smith DG. The biosynthesis of 5-­hydroxytryptamine in
brain. Biochem J. 1967;105:351-­360.
and this results in a correspondingly greater reduction in food in-
6. Walther DJ, Bader M. A unique central tryptophan hydroxylase
take.102 Whether this combination also increases the effects of lor- isoform. Biochem Pharmacol. 2003;66:1673-­1680.
caserin on glycaemic control and insulin sensitivity has not yet been 7. Keszthelyi D, Troost FJ, Masclee AAM. Understanding the role of
reported. These findings illustrate the potential of a combination of tryptophan and serotonin metabolism in gastrointestinal function.
Neurogastroenterol Motil. 2009;21:1239-­1249.
a 5-­HT2CR agonist with other anorectic ligands to produce an im-
8. Matsuda M, Imaoka T, Vomachka AJ, et al. Serotonin regulates
proved therapeutic profile. mammary gland development via an autocrine-­paracrine loop. Dev
Compelling genetic, pharmacological and anatomical evidence Cell. 2004;6:193-­203.
generated over the last half century points to the 5-­HT system as 9. Lesurtel M, Graf R, Aleil B, et al. Platelet-­derived serotonin medi-
having a key role in the regulation of energy homeostasis and in ates liver regeneration. Science. 2006;312:104-­107.
10. Namkung J, Kim H, Park S. Peripheral serotonin: a new player in
insulin secretion from β cells. Brain 5-­HT via action at 5-­HT2CRs
systemic energy homeostasis. Mol Cells. 2015;38:1023-­1028.
impacts feeding, body weight, glycaemic control and insulin sensi- 11. Sharp T, Barnes NM. Central 5-­HT receptors and their function;
tivity. Pancreatic 5-­HT is co-­secreted from β cells where it acts at present and future. Neuropharmacology. 2020;177:108155.
5-­HT1ARs, 5-­HT1FRs and 5-­HT2BRs to influence insulin secretion. 12. Kroeze WK, Kristiansen K, Roth BL. Molecular biology of sero-
tonin receptors structure and function at the molecular level. Curr
The mechanisms through which 5-­HT modulates insulin sensitivity
Top Med Chem. 2002;2:507-­528.
and insulin release are still not well understood and provide an excit- 13. Barnes NM, Sharp T. A review of central 5-­HT receptors and their
ing opportunity for future research. function. Neuropharmacology. 1999;38:1083-­1152.
14. Falck B, Hellman B. Evidence for the presence of biogenic amines
in pancreatic islets. Experientia. 1963;19:139-­140.
AC K N OW L E D G E M E N T S
15. Gylfe E. Association between 5-­hydroxytryptamine release and
We acknowledge funding from Diabetes UK (18/0005884) and the insulin secretion. J Endocrinol. 1978;78:239-­248.
BBSRC (BB/R01857X/1 and BBN017838/1). 16. Braun M, Wendt A, Karanauskaite J, et al. Corelease and differ-
ential exit via the fusion pore of GABA, serotonin, and ATP from
LDCV in rat pancreatic β cells. J Gen Physiol. 2007;129:221-­231.
C O N FL I C T O F I N T E R E S T S
17. Ohta Y, Kosaka Y, Kishimoto N, et al. Convergence of the insu-
The authors declare that they have no conflicts of interest. lin and serotonin programs in the pancreatic β-­cell. Diabetes.
2011;60:3208-­3216.
AU T H O R C O N T R I B U T I O N S 18. Bennet H, Balhuizen A, Medina A, et al. Altered serotonin (5-­HT)
1D and 2A receptor expression may contribute to defective in-
Teodora Georgescu: Writing –­ original draft; Writing –­ review & ed-
sulin and glucagon secretion in human type 2 diabetes. Peptides.
iting. David Lyons: Supervision; Writing –­ original draft; Writing –­ re- 2015;71:113-­120.
view & editing. Lora K. Heisler: Supervision; Writing –­ original draft; 19. Almaça J, Molina J, Menegaz D, et al. Human beta cells produce
Writing –­ review & editing. and release serotonin to inhibit glucagon secretion from alpha
cells. Cell Rep. 2016;17:3281-­3291.
20. Kim K, Oh CM, Ohara-­Imaizumi M, et al. Functional role of sero-
PEER REVIEW
tonin in insulin secretion in a diet-­induced insulin-­resistant state.
The peer review history for this article is available at https://publo​ Endocrinology. 2015;156:444-­452.
ns.com/publo​n/10.1111/jne.12960. 21. Paulmann N, Grohmann M, Voigt JP, et al. Intracellular serotonin
modulates insulin secretion from pancreatic β-­cells by protein se-
rotonylation. PLoS Biol. 2009;7:1-­10.
ORCID
22. Sumara G, Sumara O, Kim JK, Karsenty G. Gut-­derived serotonin
Teodora Georgescu https://orcid.org/0000-0002-4586-8319 is a multifunctional determinant to fasting adaptation. Cell Metab.
Lora K. Heisler https://orcid.org/0000-0002-7731-1419 2012;16:588-­600.
23. Kim H, Toyofuku Y, Lynn FC, et al. Serotonin regulates pancreatic
beta cell mass during pregnancy. Nat Med. 2010;16:804-­8 08.
REFERENCES
24. Ohara-­Imaizumi M, Kim H, Yoshida M, et al. Serotonin regulates
1. Hex N, Bartlett C, Wright D, Taylor M, Varley D. Estimating the glucose-­stimulated insulin secretion from pancreatic β cells during
current and future costs of Type1 and Type2 diabetes in the UK, pregnancy. Proc Natl Acad Sci U S A. 2013;110:19420-­19425.
including direct health costs and indirect societal and productivity 25. Bennet H, Mollet IG, Balhuizen A, et al. Serotonin (5-­ HT) re-
costs. Diabet Med. 2012;29:855-­862. ceptor 2b activation augments glucose-­ stimulated insulin se-
2. Jones PM, Persaud SJ. Classification and Diagnosis of Diabetes. In: cretion in human and mouse islets of Langerhans. Diabetologia.
Holt RIG, Cockram CS, Flyvbjerg A, Goldstein BJ, eds. Textbook of 2016;59:744-­754.
Diabetes, 5th ed. Oxford, UK: Wiley-­Blackwell; 2010. 26. Blodgett DM, Nowosielska A, Afik S, et al. Novel observations
3. Bentham J, Di Cesare M, Bilano V, et al. Worldwide trends in body-­ from next-­generation RNA sequencing of highly purified human
mass index, underweight, overweight, and obesity from 1975 to adult and fetal islet cell subsets. Diabetes. 2015;64:3172-­3181.
2016: a pooled analysis of 2416 population-­based measurement 27. Moon JH, Kim H, Kim H, et al. Lactation improves pancreatic β cell
studies in 128·9 million children, adolescents, and adults. Lancet. mass and function through serotonin production. Sci Transl Med.
2017;390:2627-­2642. 2020;12:eaay0455.
4. Carvajal R, Wadden TA, Tsai AG, Peck K, Moran CH. Managing 28. Walther DJ, Peter JU, Winter S, et al. Serotonylation of Small
obesity in primary care practice: a narrative review. Ann N Y Acad GTPases Is a signal transduction pathway that triggers platelet α-­
Sci. 2013;1281:191-­206. granule release. Cell. 2003;115:851-­862.
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
GEORGESCU et al. | 11 of 12

29. Bernassola F, Federici M, Corazzari M, et al. Role of transglutami- 49. Saller CF, Stricker EM. Hyperphagia and increased growth in
nase 2 in glucose tolerance: knockout mice studies and a putative rats after intraventricular injection of 5,7-­dihydroxytryptamine.
mutation in a MODY patient. FASEB J. 2002;16:1371-­1378. Science. 1976;191:385-­387.
30. Fernstrom JD, Wurtman RJ. Brain serotonin content: phys- 50. Connolly HM, Crary JL, McGoon MD, et al. Valvular heart dis-
iological dependence on plasma tryptophan levels. Science. ease associated with fenfluramine phentermine. N Engl J Med.
1971;173:149-­152. 1997;337:581-­588.
31. Wakayama K, Ohtsuki S, Takanaga H, Hosoya KI, Terasaki T. 51. Fisler JS, Underberger SJ, York DA, Bray GA. d-­Fenfluramine in a
Localization of norepinephrine and serotonin transporter rat model of dietary fat-­induced obesity. Pharmacol Biochem Behav.
in mouse brain capillary endothelial cells. Neurosci Res. 1993;45:487-­493.
2002;44:173-­180. 52. Scalfi L, D'Arrigo E, Carandente V, Coltorti A, Contaldo F. The
32. Brust P, Friedrich A, Krizbai IA, Bergmann R. Functional expression acute effect of dexfenfluramine on resting metabolic rate
of the serotonin transporter in immortalized rat brain microvessel and postprandial thermogenesis in obese subjects: a double-­
endothelial cells. J Neurochem. 2000;74:1241-­1248. blind placebo-­controlled study. Int J Obes Relat Metab Disord.
33. Nakatani Y, Sato-­Suzuki I, Tsujino N, et al. Augmented brain 5-­HT 1993;17:91-­96.
crosses the blood-­brain barrier through the 5-­HT transporter in 53. Larsen S, Vejtorp L, Hornnes P, Bechgaard H, Sestoft L, Lyngsoe
rat. Eur J Neurosci. 2008;27:2466-­2472. J. Metabolic effects of fenfluramine in obese diabetics. Br J Clin
34. Hale MW, Lowry CA. Functional topography of midbrain and pon- Pharmacol. 1977;4:529-­533.
tine serotonergic systems: Implications for synaptic regulation of 54. Turtle JR, Burgess JA. Hypoglycemic action of fenfluramine in dia-
serotonergic circuits. Psychopharmacology. 2011;213:243-­264. betes mellitus. Diabetes. 1973;22:858-­867.
35. Dahlstroem A, Fuxe K. Evidence for the existence of monoamine-­ 55. Verdy M, Charbonneau L, Verdy I, Belanger R, Bolte E, Chiasson JL.
containing neurons in the central nervous system. I. Demonstration Fenfluramine in the treatment of non-­insulin-­dependent diabetics:
of monoamines in the cell bodies of brain neurons. Acta Physiol hypoglycemic versus anorectic effect. Int J Obes. 1983;7:289-­297.
Scand Suppl. 1964;232:1-­55. 56. Vickers SP, Clifton PG, Dourish CT, Tecott LH. Reduced satiating
36. Lam DD, Heisler LK. Serotonin and energy balance: molecular effect of d-­fenfluramine in serotonin 5-­HT(2C) receptor mutant
mechanisms and implications for type 2 diabetes. Expert Rev Mol mice. Psychopharmacology. 1999;143:309-­314.
Med. 2007;9:1-­24. 57. Bray GA, Ryan DH, Heidingsfeldervi S, et al. A double-­ blind
37. Vertes RP, Crane AM. Distribution, quantification, and morpho- randomized placebo –­Controlled Trial of Sibutramine. Obes Res
logical characteristics of serotonin-­immunoreactive cells of the 1995;4:263-­270.
supralemniscal nucleus (B9) and pontomesencephalic reticular 58. Arterburn DE, Crane PK, Veenstra DL. The efficacy and safety of
formation in the rat. J Comp Neurol. 1997;378:411-­424. sibutramine for weight loss: a systematic review. Arch Intern Med.
38. Kiss J, Léránth C, Halász B. Serotoninergic endings on VIP-­neurons 2004;164:994-­1003.
in the suprachiasmatic nucleus and on ACTH-­neurons in the ar- 59. Barkeling B, Elfhag K, Rooth P, Rössner S. Short-­term effects of
cuate nucleus of the rat hypothalamus. A combination of high sibutramine (Reductil) on appetite and eating behaviour and the
resolution autoradiography and electron microscopic immunocy- long-­term therapeutic outcome. Int J Obes Relat Metab Disord.
tochemistry. Neurosci Lett. 1984;44:119-­124. 2003;27:693-­700.
39. Willoughby JO, Blessing WW. Origin of serotonin innervation 60. James WPT, Caterson ID, Coutinho W, et al. Effect of sibutramine
of the arcuate and ventromedial hypothalamic region. Brain Res. on cardiovascular outcomes in overweight and obese subjects. N
1987;418:170-­173. Engl J Med. 2010;363:905-­917.
40. Heisler LK, Jobst EE, Sutton GM, et al. Serotonin reciprocally reg- 61. Connoley IP, Liu YL, Frost I, Reckless IP, Heal DJ, Stock MJ.
ulates melanocortin neurons to modulate food intake. Neuron. Thermogenic effects of sibutramine and its metabolites. Br J
2006;51:239-­249. Pharmacol. 1999;126:1487-­1495.
41. Zhou L, Williams T, Lachey JL, Kishi T, Cowley MA, Heisler LK. 62. Walsh KM, Leen E, Lean MEJ. The effect of sibutramine on rest-
Serotonergic pathways converge upon central melanocortin sys- ing energy expenditure and adrenaline-­induced thermogenesis in
tems to regulate energy balance. Peptides. 2005;26:1728-­1732. obese females. Int J Obes. 1999;23:1009-­1015.
42. Xu Y, Jones JE, Lauzon DA, et al. A Serotonin and melano- 63. Van Mil EGAH, Westerterp KR, Kester ADM, Delemarre-­van De
cortin circuit mediates D-­ fenfluramine anorexia. J Neurosci. Waal HA, Gerver WJM, Saris WHM. The effect of sibutramine on
2010;30:14630-­14634. energy expenditure and body composition in obese adolescents. J
43. Burke LK, Doslikova B, D'Agostino G, et al. 5-­HT obesity medi- Clin Endocrinol Metab. 2007;92:1409-­1414.
cation efficacy via POMC activation is maintained during aging. 64. Starling RD, Liu X, Sullivan DH. Influence of sibutramine on
Endocrinology. 2014;155:3732-­3738. energy expenditure in African American women. Obes Res.
44. D’Agostino G, Lyons D, Cristiano C, et al. Nucleus of the solitary 2001;9:251-­256.
tract serotonin 5-­ HT2C receptors modulate food intake. Cell 65. McNulty SJ, Ur E, Williams G. A randomized trial of sibutramine
Metab. 2018;28:619-­630.e5. in the management of obese type 2 diabetic patients treated with
45. Burke LK, Ogunnowo-­Bada E, Georgescu T. et al. Lorcaserin im- metformin. Diabetes Care. 2003;26:125-­131.
proves glycemic control via a melanocortin neurocircuit. Mol 66. Jackson HC, Bearham MC, Hutchins LJ, Mazurkiewicz SE, Needham
Metab. 2017;6:1092-­1102. AM, Heal DJ. Investigation of the mechanisms underlying the hy-
46. Xu Y, Jones JE, Kohno D, et al. 5-­ HT2CRs expressed by pro-­ pophagic effects of the 5-­HT and noradrenaline reuptake inhibitor,
opiomelanocortin neurons regulate energy homeostasis. Neuron. sibutramine, in the rat. Br J Pharmacol. 1997;121:1613-­1618.
2008;60:582-­589. 67. Higgs S, Cooper AJ, Barnes NM. Reversal of sibutramine-­
47. Heisler LK, Cowley MA, Tecott LH, et al. Activation of central me- induced anorexia with a selective 5-­HT2C receptor antagonist.
lanocortin pathways by fenfluramine. Science. 2002;297:609-­611. Psychopharmacology. 2011;214:941-­947.
48. Breisch ST, Zemlan FP, Hoebel BG, Url S. Hyperphagia and 68. Tecott LH, Sun LM, Akana SF, et al. Eating disorder and epilepsy in
obesity following serotonin depletion by intraventricular p-­ mice lacking 5-­HT2c serotonin receptors. Nature. 1995;374:542-­546.
Chlorophenylalanine Hyperphagia and obesity following sero- 69. Nonogaki K, Strack AM, Dallman MF, Tecott LH. Leptin-­independent
tonin depletion by Intraventricularp-­Chlorophenylalanine. Science. hyperphagia and type 2 diabetes in mice with a mutated serotonin
1976;192:382-­385. 5-­HT2C receptor gene. Nat Med. 1998;4:1152-­1156.
 13652826, 2021, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jne.12960 by Libya Hinari NPL, Wiley Online Library on [23/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 of 12 | GEORGESCU et al.

70. Nebigil CG, Etienne N, Messaddeq N, Maroteaux L. Serotonin is a factors in obese adults: A six-­ month, randomized, placebo-­
novel survival factor of cardiomyocytes: mitochondria as a target controlled, double-­ blind clinical trial. Diabetes, Obes Metab.
of 5-­HT2B receptor signaling. FASEB J. 2003;17:1373-­1375. 2019;21:1487-­1492.
71. Nebigil CG, Choi DS, Dierich A, et al. Serotonin 2B recep- 90. Pi-­Sunyer X, Shanahan W, Fain R, Ma T, Garvey WT. Impact of lor-
tor is required for heart development. Proc Natl Acad Sci U S A. caserin on glycemic control in overweight and obese patients with
2000;97:9508-­9513. type 2 diabetes: analysis of week 52 responders and nonrespond-
72. Vickers SP, Easton N, Webster LJ, et al. Oral administration ers. Postgrad Med. 2016;128:591-­597.
of the 5-­ HT2Creceptor agonist, mCPP, reduces body weight 91. Bohula EA, Wiviott SD, McGuire DK, et al. Cardiovascular safety
gain in rats over 28 days as a result of maintained hypophagia. of lorcaserin in overweight or obese patients. N Engl J Med.
Psychopharmacology. 2003;167:274-­280. 2018;379:1107-­1117.
73. Sargent PA, Sharpley AL, Williams C, Goodall EM, Cowen PJ. 5-­ 92. Bohula EA, Wiviott SD, McGuire DK, et al. Cardiovascular safety
HT(2C) receptor activation decreases appetite and body weight in of lorcaserin in overweight or obese patients. N Engl J Med.
obese subjects. Psychopharmacology. 1997;133:309-­312. 2018;379:1107-­1117.
74. Zhou L, Sutton GM, Rochford JJ, et al. Serotonin 2C receptor ago- 93. Bohula EA, Scirica BM, Fanola C, et al. Design and rationale for
nists improve type 2 diabetes via melanocortin-­4 receptor signal- the cardiovascular and metabolic effects of lorcaserin in over-
ing pathways. Cell Metab. 2007;6:398-­4 05. weight and obese patients-­thrombolysis in myocardial infarction
75. Sohn J-­W, Xu Y, Jones JE, Wickman K, Williams KW, Elmquist 61 (CAMELLIA-­TIMI 61) trial. Am Heart J. 2018;202:39-­48.
JK. Serotonin 2C receptor activates a distinct population of ar- 94. Higgins GA, Silenieks LB, Altherr EB, MacMillan C, Fletcher PJ,
cuate pro-­opiomelanocortin neurons via TRPC channels. Neuron. Pratt WE. Lorcaserin and CP-­8 09101 reduce motor impulsivity and
2011;71:488-­497. reinstatement of food seeking behavior in male rats: Implications
76. Berglund ED, Liu C, Sohn JW, et al. Serotonin 2C receptors in pro-­ for understanding the anti-­obesity property of 5-­HT2C receptor
opiomelanocortin neurons regulate energy and glucose homeo- agonists. Psychopharmacology. 2016;233:2841-­2856.
stasis. J Clin Invest. 2013;123:5061-­5070. 95. Burns CM, Chu H, Rueter SM, et al. Regulation of sero-
77. Lam DD, Przydzial MJ, Ridley SH, et al. Serotonin 5-­HT2C receptor tonin-­2C receptor G-­ protein coupling by RNA editing. Nature.
agonist promotes hypophagia via downstream activation of mela- 1997;387:303-­3 08.
nocortin 4 receptors. Endocrinology. 2008;149:1323-­1328. 96. Garfield AS, Davies JR, Burke LK, et al. Increased alternate splicing
78. Thomsen WJ, Grottick AJ, Menzaghi F, et al. Lorcaserin, a novel of Htr2c in a mouse model for Prader-­Willi syndrome leads disrup-
selective human 5-­hydroxytryptamine 2C agonist : in Vitro and tion of 5HT2C receptor mediated appetite. Mol Brain. 2016;9:95.
in Vivo pharmacological characterization. J Pharmacol Exp Ther. 97. Balthasar N, Dalgaard LT, Lee CE, et al. Divergence of melanocor-
2008;325:577-­587. tin pathways in the control of food intake and energy expenditure.
79. Higgs S, Cooper AJ, Barnes NM. The 5-­HT2C receptor agonist, Cell. 2005;123:493-­505.
lorcaserin, and the 5-­HT6 receptor antagonist, SB-­742457, pro- 98. Krashes MJ, Lowell BB, Garfield AS. Melanocortin-­4 receptor–­
mote satiety; A microstructural analysis of feeding behaviour. regulated energy homeostasis. Nat Neurosci. 2016;19:206-­219.
Psychopharmacology. 2016;233:417-­424. 99. Kennett GA, Curzon G. Evidence that hypophagia induced by
80. Martin CK, Redman LM, Zhang J, et al. Lorcaserin, a 5-­HT2C re- mCPP and TFMPP requires 5-­HT1C and 5-­HT1B receptors; hy-
ceptor agonist, reduces body weight by decreasing energy intake pophagia induced by RU 24969 only requires 5-­HT1B receptors.
without influencing energy expenditure. J Clin Endocrinol Metab. Psychopharmacology. 1988;96:93-­100.
2011;96:837-­8 45. 100. Doslikova B, Garfield AS, Shaw J, et al. 5-­ HT2C receptor ag-
81. Valencia-­Torres L, Olarte-­Sánchez CM, Lyons DJ et al. Activation onist anorectic efficacy potentiated by 5-­ HT1B receptor
of ventral tegmental area 5-­HT2C receptors reduces incentive mo- agonist coapplication: an effect mediated via increased pro-
tivation. Neuropsychopharmacology. 2016;42:1511-­1521. portion of pro-­opiomelanocortin neurons activated. J Neurosci.
82. Higgins GA, Zeeb FD, Fletcher PJ. Role of impulsivity and re- 2013;33:9800-­9804.
ward in the anti-­obesity actions of 5-­HT2C receptor agonists. J 101. Schreiber R, De Vry J. Role of 5-­HT2C receptors in the hypophagic
Psychopharmacol. 2017;31:1403-­1418. effect of m-­CPP, ORG 37684 and CP-­94,253 in the rat. Prog Neuro-­
83. Higgins GA, Silenieks LB, Roßmann A, et al. The 5-­HT 2C receptor Psychopharmacology Biol Psychiatry. 2002;26:441-­4 49.
agonist lorcaserin reduces nicotine self-­administration, discrimi- 102. Doslikova B, Garfield AS, Shaw J, et al. 5-­ HT2C receptor ag-
nation, and reinstatement: Relationship to feeding behavior and onist anorectic efficacy potentiated by 5-­ HT1B receptor
impulse control. Neuropsychopharmacology. 2012;37:1177-­1191. agonist coapplication: An effect mediated via increased pro-
84. Xu P, He Y, Cao X, et al. Activation of serotonin 2C receptors in do- portion of pro-­opiomelanocortin neurons activated. J Neurosci.
pamine neurons inhibits binge-­like eating in mice. Biol Psychiatry. 2013;33:9800-­9804.
2017;81:737-­747. 103. Marin P, Bécamel C, Chaumont-­Dubel S, Vandermoere F, Bockaert
85. Taylor JR, Dietrich E, Powell J. Lorcaserin for weight management. J, Claeysen S. Classification and signaling characteristics of 5-­HT
Diabetes, Metab Syndr Obes Targets Ther. 2013;6:209-­216. receptors: toward the concept of 5-­ HT receptosomes. Handb
86. O'Neil PM, Smith SR, Weissman NJ, et al. Randomized placebo-­ Behav Neurosci. 2020;31:91-­120.
controlled clinical trial of lorcaserin for weight loss in type 2 diabe-
tes mellitus: the BLOOM-­DM study. Obesity. 2012;20:1426-­1436.
87. Smith SR, Weissman NJ, Anderson CM, et al. Multicenter, placebo-­
How to cite this article: Georgescu T, Lyons D, Heisler LK.
controlled trial of lorcaserin for weight management. N Engl J Med.
Role of serotonin in body weight, insulin secretion and
2010;363:245-­256.
88. Apovian C, Palmer K, Fain R, Perdomo C, Rubino D. Effects of lor- glycaemic control. J Neuroendocrinol. 2021;33:e12960.
caserin on fat and lean mass loss in obese and overweight patients https://doi.org/10.1111/jne.12960
without and with type 2 diabetes mellitus: the BLOSSOM and
BLOOM-­DM studies. Diabetes, Obes Metab. 2016;18:945-­948.
89. Tuccinardi D, Farr OM, Upadhyay J, et al. Lorcaserin treat-
ment decreases body weight and reduces cardiometabolic risk