Rights / License: Research Collection In Copyright - Non ......Dr. Matthew Hayes, co-examiner Dr....

Research Collection

Doctoral Thesis

Role of the Vagus Nerve in the Regulation of Energy Homeostasisand Glycemia by Intestinal Glucagon-Like Peptide-1

Author(s): Krieger, Jean-Philippe

Publication Date: 2016

Permanent Link: https://doi.org/10.3929/ethz-a-010654308

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

https://doi.org/10.3929/ethz-a-010654308

http://rightsstatements.org/page/InC-NC/1.0/

https://www.research-collection.ethz.ch

https://www.research-collection.ethz.ch/terms-of-use

DISS. ETH NO. 23358

Role of the Vagus Nerve in the Regulation of Energy Homeostasis and Glycemia by Intestinal Glucagon-Lik e

Peptide-1

A thesis submitted to attain the degree of

DOCTOR OF SCIENCES of ETH ZURICH

(Dr. sc. ETH Zurich)

presented by

JEAN-PHILIPPE KRIEGER

Ingénieur diplômé de l’Institut des Sciences et Industries du Vivant et de l’Environnement AgroParisTech (Paris, France)

born on 29.03.1987

citizen of France

Prof. Dr. Wolfgang Langhans, examiner

Prof. Rémy Burcelin, co-examiner

Dr. Matthew Hayes, co-examiner

Dr. Shin Jae Lee, co-examiner

Prof. Dr. Thomas Lutz, co-examiner

2016

« People ask me, 'What is the use of climbing Mount Everest?' and my

answer must at once be, 'It is of no use.' There is not the slightest prospect of any

gain whatsoever. Oh, we may learn a little about the behaviour of the human body at

high altitudes, and possibly medical men may turn our observation to some account

for the purposes of aviation. […]

If you cannot understand that there is something in man which

responds to the challenge of this mountain and goes out to meet it, that the

struggle is the struggle of life itself upward and forever upward, then you won't

see why we go. What we get from this adventure is just sheer joy. And joy is,

after all, the end of life. »

George Mallory

Climbing Everest

Acknowledgments

1

ACKNOWLEDGMENTS

WARNING: The acknowledgment section usually is the most carefully read section of an

entire thesis, so I will try my best not to forget anyone. If you don’t find yourself in the next

lines, let me know and I will offer profuse apologies, blame it on my temporary exhaustion

and get you a new copy of the thesis with your name on it.

It has been a busy, rich and fruitful journey. Now is the time to do what I could so rarely do

during my busy doctoral life: take the time to say thank you and acknowledge those who

contributed to this work, either by their scientific inputs, experimental help, administrative

support, adventurous spirit, craziness, food supply, friendship and love.

First and foremost, I would like to express my immense gratitude to my thesis supervisor Pr.

Langhans. Thank you for giving me the opportunity to do my doctoral thesis in your

laboratory and providing me with the perfect environment to do research. I really appreciated

your constant encouragements in pursuing my experiments, and your positive feedbacks on

my presentations, posters and manuscripts (even when they were probably not that good!).

Thank you for the time spent reading my manuscripts, including this dissertation, and your

willingness to send me to numerous meetings over the years. I am sure we will find a way to

prolong our fruitful association in the future.

Moreover, I am also extremely grateful to Dr. Shin Lee. Thank you for your guidance and

critical inputs along the way. We almost started in this laboratory at the same time and I can

say that our work together and our constant interactions really shaped my scientific thinking.

Especially, thank you for opening my mind to new approaches and showing me that we can

always go one step further. The work that follows is also yours and it’s needless to say I

could not have done it without you.

I would like to thank Pr. Thomas Lutz, Pr. Rémy Burcelin and Dr. Matthew Hayes for

following my work over the last years, taking the time to read my thesis, and being part of my

thesis committee.

A big thank you also goes to Dr. Guillaume de Lartigue. I am still amazed by how much we

have achieved during your 3 months in Zürich. With you, my experimental plans made a

giant leap forward and I learned a lot. I also would like to thank our other “Schwerzenbach

special guests” (Pr. Woods, Levin, Watts, Spector, …) for the scientific discussions and

inspirational presence.

I am also extremely grateful to Myrtha Arnold for showing me some of her surgical secrets.

Especially, thank you for developing the technique of bilateral nodose ganglion injection and

taking the time to teach me how to perform it. Thank you for the long hours you spent in the

Acknowledgments

2

surgery room for my project. Your hard work and constant willingness to improve will always

be remembered.

I also would like to thank the members of my laboratory, with whom we shared thoughts, joy,

despair, coffee and chocolate. Very special thanks to Shahana (for her patience when I tried

to speak Italian), Deepti (for her patience when I tried to eat spicy food), Sharon (for piloting

our “Lymph for life” venture), Melanie (for showing me how to do bricolage in the behavioral

rooms), as well as Nino (for all the gifts I received), Nadja (welcome to the team), Fohr (for

being the only person who likes my music in the lab), Flavia (for almost working for me once)

and Abdelhak (for all the job advice). I would like to give special thanks to Marie Labouesse

for convincing me to join ETH on a snowy day of February 2012. Also, thanks for inspiring

me with all your brilliant research projects and for teaching me that a second day starts in a

lab after 8 pm. Moreover, a very big thank you goes to Klaus: your willingness to give

everything to science so early in your career was impressive. Thanks for entertaining us in

the office, for making us work late at night in exchange of caipirinhas and for giving me the

“tiger”. You will be an amazing scientist. Also, thanks to Pius for the unforgettable stories and

for teaching me that deadlines are made to be broken… I mean … really broken. Thanks to

Rahel, Tim, Katharina and Marie for choosing to do their thesis in Schwerzenbach and

sharing their sense of humor and some good times with us. Naturally, an immense thank you

goes to Rosi. Thanks for explaining me thousand times how GLP-1 is secreted. Thanks for

contaminating me a bit more with the love of mountains and crazy outdoor adventures. Most

importantly, thank you for sharing your endless optimism throughout the years and for all the

things you’ve done outside the lab.

I am also extremely grateful to the former members of our lab, especially Elnaz, Guggi, Ulli

and Sandra for their inspirational work and the fun moments spent together. Thanks to Urs

for making science fun and a little less serious, and to the members of his lab for providing

us with lots of ideas and new behavioral paradigms.

Also I would like to thank the Wolfrums for being such great and fun neighbors. Thanks for

teaching me how to produce viruses and for your willingness to share the molecular tools

and equipment you have. A great thank you goes to Matthias and Elke for showing me how

all this stuff works. In addition, a big thank you goes to the Lutz’s lab for providing me with a

fully functional Phenomaster system (thank you Christina, I realize how difficult that is now

that we have one!) and for the kind comments on my projects during our joined lab meetings.

Thanks to the FACS facility in Irchel for making cell sorting so easy for us.

Some other special people made my life in Schwerzenbach much easier: I would particularly

like to thank Ruth Hauser for her immense administrative work, Marcel Morf for making my

Acknowledgments

3

stay completely bug-free and our technical service crew for their ability to instantly repair

everything I broke. Also, an immense thank you goes to Silke and the whole animal

caretaker crew: you guys simply keep this place running! I am really looking forward to

sharing with you my results in one of your meetings, because you greatly contributed to this

thesis.

On a personal note, I would like to express a deep thank you to four of my early science

mentors. Thank you to Christophe Laville, Olivier Loudet, Damian Holsinger and Jeff

Ghiglione. I had a wonderful experience in your class / lab and, hence, you are all a bit

responsible for where I am today.

Also, thanks to Camille, Océane and Axel for being my friends, but also a scientific source of

inspiration. Thanks to all the friends who remained my friends although I barely gave news in

the last four years. Thanks to Clémentine, Denis, Virgile, Daphney, PE, Bérengère, Solenne,

PM, Delphine …

I also would like to express my warm feelings to former colleagues and friends at Decathlon.

Working with you was amazing. Your support when I decided to leave industry and start a

PhD will always be remembered. Thanks to Maud, Sophie, Camille, Philou, Damien, Elodie,

Nico, Jérémy, …

Biking, running and swimming kept me sane during these years and I would like to thank all

those who shared training, races and adventures with me. Thanks to Camille (again), Rosi

(again), huge thanks to Antoine, the Trigether Club and the Mont Blanc Crew ‘12, ‘14 and

‘15.

Finally, I would like to gratefully thank my family. Thank you for always giving me the

opportunity to study in the best conditions. Thank you for your trust and encouragement in

pursuing the path that I chose. Thank you to my brother Olivier for really saving the world

while I was just “playing with pipettes in Switzerland”.

Table of contents

4

TABLE OF CONTENTS

ACKNOWLEDGMENTS

LIST OF ABBREVIATIONS

SUMMARY

RESUME

1

8

11

13

CHAPTER 1: GENERAL INTRODUCTION 17

1 The worldwide obesity epidemic: trends, causes and possible solutions 18

The epidemiology of obesity: a short overview of current trends 18 1.1

A complex network of causes leads to obesity 20 1.2

Current approaches used to tackle obesity 22 1.3

Understanding the control of food intake to tackle obesity. 24 1.4

2 The vagus nerve, a bidirectional information highwa y linking brain and periphery

for the regulation of body homeostasis 27

Anatomy of the vagus nerve 27 2.1

Overview of vagus nerve homeostatic functions 30 2.2

Role of the vagus nerve in the control of eating and post-absorptive mechanisms 30 2.3

Role of the vagus nerve in the control of glucose homeostasis 34 2.4

Role of the vagus nerve in the control of energy expenditure 35 2.5

3 The glucagon-like peptide-1, a multifaceted peptide involved in energy

homeostasis 36

Secretion of intestinal GLP-1 36 3.1

The biological functions of GLP-1 in glucose and energy homeostasis 38 3.2

The central GLP-1 system: effects and interaction with the periphery 41 3.3

The GLP-1R agonists for anti-diabetes and anti-obesity therapies 44 3.4

4 Aims of the thesis 45

Table of contents

5

CHAPTER 2: VAGAL MEDIATION OF GLP-1'S EFECTS ON FOO D INTAKE AND

GLYCEMIA 59

1 Introduction 61

2 Vagal GLP-1R activation in the control of food inta ke 62

Necessity of vagal afferents 62 2.1

Roles of vagal subbranches 63 2.2

Mediation of acute vs chronic effects of GLP-1R agonists by the vagus nerve 65 2.3

Summary 66 2.4

3 Vagal GLP-1R activation and the neuroincretin effec t 67

The vagus nerve can mediate GLP-1 effects on post-prandial glycemia 67 3.1

Roles of vagal subbranches 67 3.2

Relevance of the vagal neuroincretin effect of GLP-1 70 3.3

Summary 71 3.4

4 Putative mechanisms of vagal GLP-1R effects on food intake and glycemia 72

Central pathways activated by vagal GLP-1R activation 72 4.1

Neuropeptides mediating vagal GLP-1R effects in VAN 75 4.2

5 Methodological obstacles to overcome: suggestions f or future vagal

manipulations 76

Lesion-based approaches and genetic manipulations of the vagus nerve 76 5.1

Proposed approaches for future studies of vagal signaling in the control of eating and 5.2

glycemia 77

6 Concluding comments 78

CHAPTER 3: KNOCKDOWN OF GLP-1 RECEPTORS IN VAGAL AF FERENTS

AFFECTS NORMAL FOOD INTAKE AND GLYCEMIA 91

CHAPTER 4: THE COCAINE- AND AMPHETAMINE-REGULATED

TRANSCRIPT MEDIATES THE SATIATING EFFECTS OF GLP-1 IN VAGAL

AFFERENT NEURONS 115

Table of contents

6

CHAPTER 5: GLP-1 RECEPTORS IN VAGAL AFFERENT NEURON S

MODULATE ENERGY EXPENDITURE VIA BROWN ADIPOSE TISSU E

THERMOGENESIS 137

CHAPTER 6: GENERAL DISCUSSION 1 65

1 Overview of the main findings 166

2 Established and potential non-vagal targets of inte stinal GLP-1 in the control

energy balance and glycemia 168

Potential satiating effects of circulating GLP-1 via central GLP-1R 168 2.1

Direct insulinotropic effects of GLP-1 in pancreatic β-cells 169 2.2

Potential other GLP-1 targets in the control of energy balance and glycemia 170 2.3

3 A role for vagal afferent neurons in the long-term control of energy balance:

potential implication in the development of obesity 171

The chronic effect of VAN GLP-1R kd in the overall energy balance is revealed by 3.1

HFD challenge 172

Evidence for a modulation of vagal afferent function in obesity 175 3.2

4 Putative evolution of the endogenous GLP-1 system d uring the development of

obesity and therapeutic implications 177

Our model tested the role of VAN GLP-1R in response to an obesogenic diet, not in 4.1

an established obese phenotype 177

Intestinal GLP-1 secretion is amplified during the development of HFD-induced 4.2

obesity 178

Putative effects of high intestinal GLP-1 levels on glucose and lipid metabolism 179 4.3

5 GLP-1 and the vagus nerve, the future of obesity tr eatment? 181

New pharmacological ways to access GLP-1 receptors on vagal afferents 181 5.1

The promises of neuromodulation approaches 183 5.2

The use of recent neuroscience tools to further understand the role of vagal afferents 5.3

in the control of energy balance 186

Table of contents

7

APPENDICES (not present in the online version) 197

CURRICULUM VITAE

LIST OF PUBLICATIONS, PRESENTATIONS AND AWARDS

198

200

List of abbreviations

8

LIST OF ABBREVIATIONS

NB: specific abbreviations used for figures and tables can be found in respective figure

legends

ACC AgRP AMPK ANOVA AP Arc ATP AUC B3AR BAT BBB cAMP CART CCK CCKaR CeA CHB CHBX CNO CNS CREB CVO DIO DMEM DMH DMV DPP-IV DREADD DRG ENS Epac2 EPODE Ex-4 Ex-9 FAS FDA Fig GABA GFP GI GIP

acetyl-coA carboxylase agouti-related peptide adenosine monophosphate-activated protein kinase analysis of variance area postrema arcuate nucleus of the hypothalamus adenosine triphosphate area under the curve β3-adrenergic receptor brown adipose tissue blood-brain barrier cyclic adenosine monophosphate cocaine- and amphetamine-regulated transcript cholecystokinin cholecystokinin a receptor central nucleus of amygdala common hepatic branch common hepatic branch vagotomy clozapine-N-oxide central nervous system cAMP response element binding protein circumventricular organs diet-induced obese Dulbecco’s modified Eagle Medium dorsomedial hypothalamus dorsal motor nucleus of the vagus nerve dipeptidyl peptidase-IV designer receptor exclusively activated by designer drugs dorsal root ganglion enteric nervous system exchange protein directly activiated by cAMP 2 ensemble, prévenons l’obésité des enfants exendin-4 exendin-(9-39) fatty acid synthase Federal Drug Adminstration figure γ-aminobutyric acid green fluorescent protein gastrointestinal gastric inhibitory peptide


9

GLP-1 GLP-1R GPCR GWAS H&E HBSS HFD HPV ICV IGLE IMA IP IPGTT IPIST IR IV kd LHA LV LXRα/β MANOVA MAPK MCH NCD NG NHANES NMDA NPY NR NTS OGTT PBN PBS PC1/3 PC2 Pdx1 PEG PFA Phox2b PI3K PKA PKC PLC PMCH POMC PPARγ PPG PRV

glucagon-like peptide-1 glucagon-like peptide-1 receptor G-protein coupled receptor genome wide association studies haematoxylin and eosin stain Hank’s balanced salt solution high-fat diet hepatic portal vein intracerebroventricular intraganglionic laminar endings intramuscular arrays intraperitoneal intraperitoneal glucose totlerance test intraperitoneal insulin sensitivity infra-red intravenous knockdown lateral hypothalamus lentivirus liver X receptor α/β multiple analysis of variance mitogen-activated protein kinase melanin-concentrating hormone noncommunicable disease nodose ganglion National Health and Nutrition Examination Survey N-methyl-D-aspartate neuropeptide-y nuclear receptor nucleus of the tractus solitarii oral glucose totlerance test parabrachial nucleus of the hypothalamus phosphate buffer saline prohormone convertase 1/3 prohormone convertase 2 pancreatic and duodenal homeobox 1 polyethylene glycol paraformaldehyde paired-like homeobox 2b phosphoinositide 3-kinase protein kinase A protein kinase C phospholipase C pro-melanin concentrating hormone proopiomelanocortin peroxisome proliferator-activated receptor γ preproglucagon pseudorabies virus


10

PVN PYY qPCR RER RNA rRPa RT SC ScWAT SDA SEM shRNA subP UCP1 VAN VEN VGX VMH VN VNB VNS VO2

vWAT WAT YFP

paraventricular nucleus of the hypothalamus peptide YY quantitative polymerase chain reaction respiratory exchange ratio ribonucleic acid rostral raphe pallidus reverse transcription subcuatenous subcutaneous white adipose tissue subdiaphragmatic vagal deafferentation standard error of the mean short hairpin RNA substance P uncoupling protein-1 vagal afferent neurons vagal efferent neurons total subdiaphragmatic vagotomy ventromedial nucleus of the hypothalamus vagus nerve vagus nerve blockade vagus nerve stimulation rate of oxygen consumption visceral white adipose tissue white adipose tissue yellow fluorescent protein

Abstract

11

ABSTRACT

In a world where 600 million adults are obese, understanding this condition and the

associated comorbidities has become urgent. Because obesity is due to an imbalance

between energy intake and energy expenditure, our field of research aims at providing the

fundamental knowledge of how the energy balance is normally regulated and how this

regulation may fail in pathological states. In this context, the role of the so-called “gut-brain

axis” in the control of food intake has received increasing attention. Indeed, the mechanical,

metabolic or hormonal signals generated in the periphery are centrally integrated and

translated into an appropriate eating behavior. In this gut-brain communication, the vagus

nerve (VN) plays an essential role. It consists of a bidirectional neuronal route, conveying

bottom-up information from the gastrointestinal tract about food quantity and composition,

and exerting a top-down control on gastrointestinal motility and secretion.

One of the intestinal peptides interacting with the VN is the glucagon-like peptide-1 (GLP-1).

Originally identified as an incretin hormone, GLP-1 has been found to inhibit eating and

gastric emptying. Based on these findings, GLP-1 receptor (GLP-1R) agonists have been

developed and are promising pharmacological tools in the treatment of obesity and diabetes.

In the present thesis, we dissect the molecular and functional interaction between

endogenous intestinal GLP-1 and the VN in the control of food intake, energy expenditure

and glycemia.

Initially, we reviewed the available literature (CHAPTER 2) testing the hypothesis that vagal

afferent neurons (VAN) mediate the effects of intestinal GLP-1 on food intake and glycemia.

This literature review revealed several important gaps and remaining questions to be

addressed.

First, the previously-available evidence supporting a role for the vagus nerve in mediating the

effects of intestinal GLP-1 on food intake and glycemia were based on methods that lesioned

the VN. The results of lesion studies are difficult to interpret because the lesions affect both

VAN and vagal efferents neurons (VEN) and may mask potential antagonist effects of

different vagal fibers. Hence, the exact role of VAN GLP-1R in mediating the effects of

endogenous intestinal GLP-1 was still unclear and required a new experimental approach.

We set up a bilateral nodose ganglion (NG) injection technique to deliver a lentiviral vector

and knock down GLP-1R in VAN via RNA interference. Using this approach, we showed that

GLP-1R in VAN are necessary for the normal control of meal size, gastric emptying and post-

meal glycemia (CHAPTER 3).

Abstract

12

Second, we further investigated potential downstream mediators of the GLP-1 signal within

VAN and their relevance for the control of food intake and gastric emptying. Multiple

neuropeptides have been identified in the NG and some of them, including the Cocaine- and

Amphetamine-Regulated Transcript (CART) had already been implicated in the eating-

inhibition induced by cholecystokinin (CCK). We screened several neuropeptides in the NG

for their transcriptional response to GLP-1 in a primary VAN culture model and found that

only CART expression was modulated. We further showed that GLP-1R activation induces

VAN CART expression in vivo. Moreover, using two models of CART inhibition, we

demonstrated that a full CART expression in VAN and CART release in the NTS are

necessary for the full eating and gastric emptying inhibitory effects induced by intraperitoneal

GLP-1 (CHAPTER 4). These results implicate for the first time CART as a neuropeptide

transmitting the signal of GLP-1 to the brainstem. The function and neuronal targets of CART

within the NTS remain to be further investigated.

Additionally, the idea that endogenous peripheral GLP-1 modulates energy expenditure was

largely unexplored. We set up a protocol to investigate the effects of peripheral GLP-1R

activation on energy expenditure in wild-type rats while controlling for food intake and gastric

emptying, two potential confounding factors. We found that peripheral GLP-1R activation

induced a transient decrease of whole-body energy expenditure, accompanied by a

decrease in interscapular brown adipose tissue (iBAT) temperature. Moreover, upon high-fat

diet exposure, reduced GLP-1R expression in VAN increased energy expenditure and BAT

thermogenesis, and alleviated the development of HFD-induced obesity and glucose

intolerance. Together with a retrograde tracing study from the iBAT and central gene

expression analysis, this led us to propose a new vagally-mediated pathway by which

endogenous peripheral GLP-1 tunes down iBAT activity after a meal (CHAPTER 5).

Considered together, these findings provide novel mechanistic and functional data about the

gut-to-brain communication in the control of energy balance. Beyond the classic view that

VAN are involved in the acute mediation of satiation signals, our findings indicate that vagal

GLP-1R also play a role in the acute control of glycemia and the chronic control of energy

expenditure. Our results also raised new questions about the exact role of vagally-expressed

neuropeptides in the control of energy balance and their contribution to the development of

obesity. Moreover, based on our findings, we developed the novel idea of a specific time

window at the onset of diet-induced obesity, in which HFD-induced secretion of endogenous

intestinal GLP-1 may amplify the development of an obese phenotype (CHAPTER 6).

Eventually, our findings and the questions that they raised may contribute to the fundamental

knowledge needed for the further development of GLP-1-based or vagal-targeting therapies.

Résumé

13

RESUME

Dans le monde, plus de 600 millions d’individus sont obèses; la compréhension des

mécanismes de cette pathologie, ainsi que celle des maladies qui lui sont associées, est

devenue urgente. Le développement de l’obésité est dû à un déséquilibre quantitatif et

chronique entre l’énergie ingérée et l’énergie dépensée. Notre domaine de recherche

s’efforce donc d’éclaircir les mécanismes de contrôle de la balance énergétique, et de

comprendre comment un dérèglement de ces mécanismes aboutit à l’obésité. Dans ce

contexte, le rôle que joue l’« axe intestin-cerveau » dans le contrôle de la prise alimentaire

est devenu un thème de recherche incontournable. En effet, les signaux mécaniques,

métaboliques ou hormonaux générés en périphérie sont intégrés au niveau du cerveau et

aboutissent à la formation d’un comportement alimentaire approprié. Dans cette

communication entre l’intestin et le cerveau, le nerf vague joue un rôle essentiel. Le nerf

vague est une voie neuronale bidirectionnelle qui, d’une part, transmet de manière

ascendante des informations du tube digestif concernant la quantité et la qualité des

aliments ingérés et, d’autre part, exerce un contrôle descendant sur les propriétés

contractiles et sécrétrices de l’estomac et l’intestin.

Le « glucagon-like peptide-1 » (GLP-1) est un des signaux peptidiques provenant de

l’intestin et interagissant avec le nerf vague. GLP-1 fut identifié en premier lieu comme une

hormone amplifiant la sécrétion d’insuline induite par l’élévation du glucose plasmatique.

Plus tard, de nombreux autres effets métaboliques de cette hormone ont été découverts,

notamment son rôle dans la réduction de l’appétit et la régulation de la vidange gastrique. En

se basant sur ces effets endogènes, des agonistes du récepteur au GLP-1 ont été

développés et sont désormais à la base de prometteurs traitements contre l’obésité et le

diabète. Dans cette thèse, nous analysons en détail les interactions moléculaires et

fonctionnelles entre le nerf vague et le GLP-1 intestinal dans le contrôle de l’homéostat

énergétique et glucidique.

Nous débutons par une revue de la littérature scientifique (CHAPITRE 2) testant l’hypothèse

selon laquelle les afférences vagales sont des intermédiaires indispensables aux effets du

GLP-1 intestinal sur la prise alimentaire et la glycémie. Cette revue de la littérature a révélé

d’importants manques dans notre compréhension des interactions GLP-1/nerf vague et

souligne les questions en suspens auxquels nous nous sommes efforcés de répondre par

une série d’approches expérimentales.

Tout d’abord, les études disponibles concernant ce sujet sont basées principalement sur des

lésions du nerf vague. Les études basées sur des lésions sont difficiles à analyser de

manière précise car les afférences comme les efférences vagales sont coupées et cela

Résumé

14

masque la potentielle complexité des effets des différentes fibres vagales. Par conséquent,

le rôle exact des récepteurs au GLP-1 présents sur les afférences vagales restait à éclaircir

par le biais d’une nouvelle approche expérimentale. Pour cela, nous avons développé une

technique d’injection bilatérale dans le ganglion pléxiforme afin d’y délivrer un vecteur viral et

de réduire l’expression des récepteurs au GLP-1 par interférence ARN. Grâce à cette

approche, nous avons pu montrer que les récepteurs au GLP-1 présents sur les afférences

vagales sont en effet nécessaires au contrôle normal de la satiété, de la vidange gastrique,

et de la glycémie post-prandiale (CHAPITRE 3).

Dans un second temps, nous souhaitions investiguer l’implication des différents

neuropeptides exprimés par les fibres vagales dans le contrôle de la prise alimentaire et de

la vidange gastrique par GLP-1. En effet, de nombreux neuropeptides ont été détectés dans

le ganglion pléxiforme et certains d’entre eux, tels que le “Cocaine- and Amphetamine-

Regulated Transcript” (CART), ont déjà été impliqués dans l’effet satiétogène de la

cholécystokinine. En revanche, l’implication de tels neuropeptides dans le signal vagal de

GLP-1 était une thématique inexplorée. Nous avons étudié la réponse transcriptionnelle de

plusieurs neuropeptides à l’application de GLP-1 dans une culture primaire d’afférences

vagales et avons détecté que seule l’expression de CART était modulée. Nous avons par la

suite vérifié que l’activation des récepteurs au GLP-1 induisait également l’expression de

CART in vivo. De plus, en utilisant deux modèles distincts d’inhibition de CART, nous avons

démontré que l’expression de CART par les afférences vagales ainsi que la sécrétion de

CART dans le noyau du tractus solitaire (NTS) sont nécessaires aux effets sur la satiété et la

vidange gastrique d’une injection intra-péritonéale de GLP-1 (CHAPITRE 4). Ces résultats

impliquent pour la première fois CART en tant que transmetteur du signal vagal de GLP-1

vers le tronc cérébral. Les cibles neuronales ainsi que la fonction exacte de CART dans le

NTS restent à investiguer.

De plus, l’idée que GLP-1 puisse moduler la dépense énergétique restait également peu

étudiée. Nous avons initialement développé un protocole d’étude des effets de l’activation

périphérique des récepteurs au GLP-1 sur la dépense énergétique de rats (phénotype

sauvage) tout en contrôlant la prise alimentaire et la vidange gastrique, deux potentiels

facteurs de confusion. Nous avons constaté que l’activation des récepteurs au GLP-1

périphérique conduisait à une diminution de la dépense énergétique, accompagnée d’une

diminution de la température du tissu adipeux brun interscapulaire (iBAT). Nous avons par la

suite montré qu’un knockdown du récepteur au GLP-1 dans les afférences vagales, pendant

l’exposition à un régime riche en gras, engendrait une augmentation de la dépense

énergétique et de l’activité du iBAT. Ces changements étaient suffisants pour freiner le

développement d’un phénotype obèse et l’apparition de l’intolérance au glucose. En

Résumé

15

complément, un marquage rétrograde des neurones provenant du iBAT et l’analyse de

l’expression centrale de gènes impliqués dans le métabolisme énergétique nous permettent,

au final, de proposer un nouvelle route neuronale par laquelle le GLP-1 intestinal réduit

l’activité du iBAT après un repas (CHAPITRE 5).

Pris dans leur ensemble, nos résultats apportent de nouvelles connaissances sur les aspects

mécanistiques et fonctionnels de la communication intestin-cerveau. Au-delà de la vue

classique selon laquelle les afférence vagales sont impliqués dans les signaux de satiation à

court terme, nous montrons également qu’elles sont impliquées dans la régulation de la

glycémie et dans le contrôle chronique de la dépense énergétique. Nos résultats posent

également la question du rôle des neuropeptides dans les afférences vagales et de leur

contribution au développement de l’obésité. De plus, nos résultats nous ont permis de

formuler l’hypothèse selon laquelle le développement de l’obésité (par opposition à une

obésité établie) consiste en une fenêtre spécifique durant laquelle une importante sécrétion

de GLP-1 par l’intestin peut participer à l’aggravation du phénotype obèse (CHAPITRE 6).

Finalement, nos résultats et les questions nouvelles qu’ont soulevés ces expériences

permettront d’apporter des connaissances fondamentales au développement de thérapies

ciblant GLP-1 et le nerf vague.

Chapter 1: Obesity, the vagus nerve and the glucagon-like peptide-1

17

CHAPTER 1: GENERAL INTRODUCTION

OBESITY AND THE CONTROL OF ENERGY BALANCE BY THE

VAGUS NERVE AND THE GLUCAGON-LIKE PEPTIDE-1


18

1 The worldwide obesity epidemic: trends, causes an d possible

solutions

Research articles and talks on energy homeostasis are often introduced with the statement

that obesity is on the rise. The use of this introductory statement implies that our field of

research is dedicated to provide the fundamental knowledge that, in fine, will help to alleviate

obesity and its consequences. Hence, to introduce my work, I wanted to relay the current

views on trends, causes, and consequences of the obesity epidemic. Furthermore, I will

discuss how the approach of our laboratory, focusing on the control of food intake, can help

understand and tackle the obesity epidemic.

The epidemiology of obesity: a short overview of c urrent trends 1.1

Obesity itself is defined by an increase in the amount of body fat, approximated at the

population level by an elevated body mass index (BMI). In 2014, more than 1.9 billion adults

were overweight (BMI > 25) and 600 million were obese (BMI > 30), representing 39% and

13% of the adult world population, respectively (1). Moreover, these figures have doubled

since 1980 and this strong upward trend seems even more alarming than the worldwide

prevalence itself. The rise of obesity is a major public health concern because obesity is

strongly associated with an impressive array of comorbidities (including type 2 diabetes

mellitus, hypertension, coronary heart disease, dyslipidemia, … (2)) and, as a result, with a

reduction in life expectancy (3). Forecast models predict that the life-shortening effects of

obesity-associated diseases may reverse the increase in life expectancy induced by medical

advancements in the United States (US) during the 21st century (4).

The American continent, as well as Europe and the Eastern Mediterranean area, show

worldwide the highest prevalence of obesity (Fig. 1). It would however be misleading to

consider obesity a high-income country problem: indeed, in low- and middle-income

countries (as classified by the World Bank), where the prevalence of overweight and obesity


19

is still relatively low, obesity is rising dramatically, especially in large urban areas and among

children. Moreover, in all regions (as defined by the World Health Organization), the obesity

rates are higher among women than men (1).

Figure 1: Prevalence of obesity by sex and regions as defined by the World Health Organization (1) Most of the key facts of the prevalence of overweight and obesity are based on the

classification of individuals in discrete BMI ranges (i.e., overweight: 25 to 30, mildly obese:

30 to 35, etc). This does not present a full picture of the overall distribution of individuals BMI

within the population and how this distribution evolved in recent years. Indeed, a slight shift

of the population weight could have dramatically increased the number of individuals falling

into the superior BMI category. This data presentation in BMI groups has been debated and,

in some cases, accused to artificially magnify the problem (5). Figure 2 presents a less often

seen, more complete picture of the BMI shifts at the population level. It compares the BMI

distribution of adults and adolescents in the United Stated between the 1976-1980 and a

1999-2004 National Health and Nutrition Examination Surveys (NHANES). This presentation

shows that, both for adults and adolescents, the distribution has shifted to the right, with a

greater shift for the upper BMI percentiles (6).

Together, these data indicate that obesity is indeed on the rise and that a better

understanding of the causes of such a profound trend is needed.


20

Figure 2: Change in the distribution of BMI between the 1976-80 and the 1999-2004 NHANES in adults (20-74 years old; left) and children and ado lescents (6-19 years old; right) in the US (6) It is, however, important to emphasize that these data are of epidemiological nature and are

thus, limited in two ways. First, they may not translate at the individual level. This is

exemplified by the case of “healthy” obese individuals, i.e., individuals that are obese but do

not show the epidemiologically-associated symptoms of one or several comorbidities. When

these healthy obese subjects are followed long enough, comorbidities may eventually

develop (7), thus reconciliating epidemiological and individual data. Second, epidemiological

data do not show causality, and understanding the true causes of obesity requires other

approaches. In the next paragraph, we will try to apprehend the complexity of the causes

leading to obesity.

A complex network of causes leads to obesity 1.2

Looking at obesity from a medical point of view may give the impression that obesity is a

rather monocausal disease that can be traced to a simple disruption of the energy balance. It

is, however, a more complex network of causes that drives the obesity epidemic.

1.2.1 One cause, several causes, layers of influenc es

Initially reserved for infectious diseases, the word “epidemic” is now widely used to also refer

to non-communicable diseases (NCD) that are spreading rapidly on a worldwide scale.

Although the same word is used to describe both infectious diseases and NCD, causes of

NCD are thought to reside in a more complex network of causes.

ADULTS CHILDREN


21

Figure 3: Conceptual hierarchy of determinants in t he etiology of a NCD (8)

Egger (8) proposes to conceptually divide the causes of a NCD into several “layers of

influence” (see Fig. 3): risk factors are the most immediate causes of the disease. They are,

in turn, influenced by proximal (i.e., nutrition, sedentary time), medial (i.e., food and exercise

environments) or distal (i.e., economics, social context) determinants.

1.2.2 Illustrating the multifactorial origin of obe sity

As an exhaustive list of all proposed causes for the rise in obesity would be beyond the

scope of this introduction, we chose to illustrate these “layers of influence” with a few

examples, starting with the more distal determinants.

Economic aspects may be one of the most tangible distal determinants of the rise in obesity.

As mentioned in a very complete review by J. Cawley (9), the first simple economic

determinant of obesity could be a differential evolution of the cost of energy-dense vs.

nutritious foods: a striking example is the dramatic decrease in the prices of sugar and

sweetened beverages between 1990 and 2007 in the US (2L bottle of Coca Cola -34.9%).

Meanwhile, the prices of fresh fruits and vegetables steadily rose (10). Moreover, the

NHANES 2005-2008 showed a negative correlation between weight and income, especially

among women (11), supporting the idea that the increasing cost of food, together with low

income, may impact food choices and caloric intake. Whether these changes in prices were

large enough to contribute to the rise of obesity in the US is however debated (9).

Typical medial or proximal determinants of obesity are often grouped under the term of

“lifestyle”. The emergence of modern western lifestyles, including a shift in nutrition and

exercise habits, seems indeed to temporally correlate with the emergence of obesity as an


22

issue at the population level (12). To illustrate the diversity of these so-called lifestyle factors,

we can use Egger and Dixon’s attempt to summarize them under the acronym “NASTIE

ODOURS”: Nutrition, (In)Activity, Stress, Technology-induced pathologies, Inadequate sleep,

Environment, Occupation, Drugs/smoking/alcohol, Over/underexposure, Relationships,

Social factors (8).

Moreover, genetic factors play a role in the variation of body fatness. Early evidence came

from family, twin or adoption studies, in which the contributions of the environmental and

genetic factors could be differentiated and quantitatively estimated (see (13) for review).

Based on these studies, 40-70% of the BMI variability between individuals has been

attributed to genetics. With the technical advances of genome sequencing, genetic

polymorphism of large populations can be correlated with BMI in genome-wide association

studies (GWAS). This new approach led to the identification of around 100 loci linked to an

obese phenotype, explaining around 21% of BMI (14). Consequently, GWAS further

emphasize the strong genetic component in BMI variations. The discrepancy between the

contribution of genetic factors seen in GWAS (21%) and in family studies (40 - 70%) may,

however, indicate that a large component of the genetic influence on BMI is due to non-

common variants, that are thus not found in large population screening.

Together, these determinants of obesity illustrate the multifactorial and complex nature of the

pathology of obesity.

Current approaches used to tackle obesity 1.3

In this section, we will briefly review the panel of existing approaches currently being

available. This will underline why understanding the control of food intake (i.e., the topic of

our laboratory) is relevant for obesity.

1.3.1 “Lifestyle” interventions

Many clinical interventions have been designed to promote healthy nutrition and exercise

and induce weight loss in overweight or obese individuals. A meta-analysis of several of


23

these interventions, however, indicates only a modest weight loss (< 5 kg) after 2 to 3 years

(15). The reasons for this limited long-term weight loss are unclear but do not seem to be

linked with a low long-term compliance to the new dietary and exercise habits (15).

Interestingly, despite the relatively small weight loss, improvements in cardiovascular

outcomes and glucose metabolism persist for years after the initial intervention (15,16),

suggesting a weight-independent benefit of the lifestyle interventions.

In addition, broad multi-approach programs have been designed to prevent the development

of overweight and obesity. The EPODE program (French for « Together Let’s prevent

childhood obesity ») for example, uses an approach based on social marketing and help of

local communities to promote exercise and healthy nutrition in children (17), thus attempting

to soften the effects of economic and lifestyle determinants. The diversity of the actions

undertaken, the numbers of actors involved and their long-term engagement are key factors

for the success of such programs. The main drawbacks of these programs are their cost and

the relative difficulty to monitor their impact (9).

1.3.2 Weight-reducing drugs

Where prevention and lifestyle interventions have not sufficed, medical solutions have

emerged. A first category of medical help can be found in weight-reducing drugs. Some of

these drugs primarily reduce the absorption of calories, such as Orlistat, the most common

anti-obesity drug. Orlistat inhibits the pancreatic lipase and, thus, the absorption of food-

derived triglycerides in the intestine. Another class of drugs is designed to mimic or target

appetite-controlling pathways, such as rimonabant (a cannabinoid receptor 1 antagonist),

pramlintide (a synthetic amylin analog), or the glucagon-like peptide-1 receptor agonists

(exenatide, liraglutide, …). A historical perspective indicates that the toxicity of appetite-

regulated drugs needs to be carefully reviewed, because several of them were withdrawn

due to life-threatening side effects, including alterations of the cardiovascular functions (such

as sibutramin or the fenfluramine-phentermine combination).


24

It is however interesting to note that pharmacological interventions with the above-mentioned

drugs reduce body weight by 5-10 kg after 2 years in overweight and obese individuals (15),

indicating an improved weight loss efficacy compared to a diet-only therapy.

1.3.3 Bariatric surgery

Bariatric surgery, which comprises several techniques reorganizing the anatomy of the

gastrointestinal (GI) tract, is regarded as the most effective weight loss strategy for obese

individuals (18,19). Indeed, techniques such as the Roux-en-Y gastric bypass surgery, which

reduces the stomach to a pouch and bypasses the duodenum and a part of the jejunum,

results in a rapid weight loss and in the resolution of diabetes, and shows a greater

maintenance of weight loss than any other method to date (18). Although irreversible and

linked with the inherent risks of an invasive surgery, the number of such interventions has

peaked in the early 2000’s and may have stabilized today between 100,000 and 200,000

bypass surgeries per year in the US (20,21). Surprisingly, the mechanisms of the

effectiveness of bypass surgeries are relatively unclear, and research projects identifying

these mechanisms may lead to new efficient therapies.

Understanding the control of food intake to tackle obesity. 1.4

The first law of thermodynamics states that the energy consumed must be either expended,

excreted or stored. This indicates that obesity, i.e., the storage induced by excess energy

intake, can theoretically be reduced by a change in the energy balance and, in particular, the

energy intake. As mentioned earlier, drugs, surgeries or lifestyle interventions aim to treat

obesity via, at least partly, a reduction in the energy consumed by individuals. To provide the

fundamental knowledge needed for this medical approach, our laboratory has focused on the

physiological mechanisms that control food intake, as well as on the potential factors that

may override a normal energy balance during obesity.

The science behind eating behavior was first dominated by the view that the utilization of

glucose was a primary factor in the initiation and termination of meals (22) and that specific

hypothalamic brain areas were responsible for one particular aspect of eating behavior (23).


25

In the next 40 years, a much more complex picture has emerged: according to this

contemporary view, a complex central neuronal network responds to a spectrum of signals

emanating from the GI tract (hormonal, neuronal, metabolic) and adipose tissue. The central

command also integrates signals from the external environment or learned eating habits

such as food cues and, in fine, translates them into an appropriate behavioral response.

A first set of signals that control food intake are those that are generated during meals and

influence satiation (i.e., meal termination) or satiety (i.e., meal intervals). Typically, these

signals are of mechanical nature (distention of the stomach) or are peptides secreted by the

gut (e.g., cholecystokinin, CCK (24)). Moreover, hormones secreted in proportion to body fat,

such as insulin and leptin, act as adiposity signals. They gain access to the hypothalamic

and hindbrain neurons influencing food intake and energy homeostasis, thus exerting a

longer-term control on these parameters via the modulation of autonomic and

neuroendocrine pathways (25). It is generally assumed that an inability of the brain to

integrate these adiposity signals contributes to the development of an obese phenotype (26).

Among the central circuit integrating satiety and adiposity signals, neurons of the arcuate

nucleus of the hypothalamus (Arc) were broadly studied for their role in the neural control of

energy homeostasis. Briefly, one population of neurons expresses the proopiomelanocortin

(POMC) gene and stimulates a melanocortin signaling pathway to exert a catabolic action. A

second set of neurons express the orexigenic neuropeptides agouti-related peptide (AgRP)

and neuropeptide-Y (NPY) to induce anabolic processes, including the stimulation of

appetite. There is evidence that the action of adiposity signals relies on the activation of the

POMC neurons while suppressing the activity of the NPY/AgRP neurons (see (25) for

review).

It would, however, be too simplistic to assume that this neuronal circuit in the Arc integrates

all peripheral and environmental signals. A whole field of research still investigates the roles

of a broad array of brain areas in the control of energy intake and expenditure, and it would


26

be beyond the scope of this introduction to detail all of them (see (27) for a more complete

overview of brain circuits).

In brief, the mechanisms that control food intake have progressively been unraveled in the

last 60 years (see Fig. 4). Most notably, the role of the peripheral signals has been better

characterized, with more than 30 gut peptides identified (28). Nevertheless, how exactly

peripheral signals get into the brain, are integrated and translated into eating behavior, is still

largely unclear. In particular, the deregulation of this well-integrated system that may lead to

obesity remains to be fully discovered.

Figure 4: A model summarizing the control of energy intake by the integration of peripheral signals by hypothalamic and hindbrain circuits (26) Abbreviations: ARC, arcuate nucleus of the hypothalamus; CCK, cholecystokinin; GI, gastrointestinal; LHA, lateral hypothalamus; NPY, neuropeptide Y; NTS, nucleus of the tractus solitarii; PFA, perifornical area; POMC, proopiomelanocortin; PVN, paraventricular nucleus of the hypothalamus

In summary, we highlighted in this section the current rise in worldwide obesity and the

complexity of causes that drive this trend. Because obesity is caused by an imbalance

between energy intake and energy expenditure, our lab’s research aims at providing the


27

fundamental knowledge of how eating is normally controlled and how this system can get

deregulated.

In the next sections, I will introduce the two systems that will be further studied in the

subsequent chapters. Consecutively, I will describe how the vagus nerve (VN) and the gut

hormone glucagon-like peptide-1 (GLP-1) are involved in the control of eating, the regulation

of glycemia and the control of energy expenditure.

2 The vagus nerve, a bidirectional information high way linking

brain and periphery for the regulation of body home ostasis

The gut-brain axis is a conceptual entity that names all information that is conveyed from the

GI tract to the brain. In the context of eating, the signals conveyed by the gut-brain axis

inform about food amount, composition and energy content. The VN has been recognized as

a major neuronal component of the gut-brain axis in the control of food intake (29). Before

explaining its role in more details, I will briefly describe the anatomy of the VN as well as

some of its other classic functions unrelated to eating.

Anatomy of the vagus nerve 2.1

The VN (cranial nerve X) innervates all organs in the thoracic and abdominal cavities. Its

complicated journey between the brain and the abdomen earned the VN the name of

« wanderer ». Vagal efferent neurons (VEN) primarly arise from the nucleus ambiguus and

the dorsal motor nucleus of the VN (DMV), whereas vagal afferent neurons (VAN) mainly

project to the nucleus tractus solitarii (NTS). Numerous rootlets exit the medulla and join

before exiting the jugular foramen. VN fibers then pass through two ganglia, where the

afferent cell bodies are located (jugular/superior and nodose/inferior ganglion), and separate

in several branches (auricular, meningeal, sympathetic, pharyngeal and laryngeal branches).

The remaining VN fibers travel along the carotid artery and form the laryngeal nerve or send

branches to the esophagus, heart and lungs. Finally, after passing the diaphragm, VN fibers


28

innervate all abdominal organs and related ganglia or plexuses (30). The ventral (left) and

dorsal (right) trunks separate into several branches: the gastric branches, the celiac

branches and the common hepatic branch. Gastric branches innervate the stomach and

proximal intestine whereas celiac branches innervate the small intestine, cecum and colon

(31). The common hepatic branch only originates from the ventral trunk. It innervates not

only the liver and the hepatic portal vein (HPV), but also the stomach, pancreas and proximal

intestine (31) (see Fig. 5).

As mentioned above, the VN contains VAN and VEN. VAN are pseudounipolar neurons

whose cell bodies reside in the nodose ganglia (NG). They are involved in the sensing of

numerous peripheral stimuli (mechanical, chemical, nociceptive, …) and relay the

corresponding signals to the dorso-vagal complex in the hindbrain. Preganglionic VEN have

their cell bodies in the DMV and project to postganglionic neurons located in close proximity

to the target organs. VEN mostly convey motor signals to the periphery and constitute a large

part of the parasympathetic nervous system. In the VN, the number of afferents greatly

outweighs the number of efferents, with 60 to 80% of vagal neurons being afferents coming

from visceral organs (30). The absolute number of vagal neurons has been estimated to be

around 16,000 to 18,000 and to be conserved across species (32).

The VN contains A- (large, myelinated), B- (intermediate, lightly myelinated) and C-fibers

(small, unmyelinated) but most of the fibers (60-80%) are afferent C-fibers from visceral

organs (30). VEN use acetylcholine as a major neurotransmitter, whereas VAN are mostly

glutamatergic. A wide array of neurotransmitters and neuropeptides, however, are expressed

in the NG, but their roles are still unclear (32).


29

Figure 5: Schematic representation of the anatomy o f the vagus nerve (adapted from (31)). Cervical and thoracic branches have been simplified by the author for clarity but are all bilateral. Subdiaphragmatic branches have been highlighted over the original figure: hepatic branch (yellow), gastric branch (green) and celiac branch (green). Abbreviations for periphery: abd, abdominal; ac, celiac artery; access, accessory; agd, right gastric artery; ags, left gastric artery; ahc, common hepatic artery; al, splenic artery; ams, superior mesenteric artery; auric, auricular; br(s), branch(es); cel, celiac; commun, communicating; dep, depressor; gangl, ganglion; hep, hepatic; laryng, laryngeal; la, larynx; mening, meningeal; mes, mesenteric; n, nerve; ph, pharynx; reccur, reccurent; sup, superior; tr, trachea; v,vein. Abbreviations for the brain: amb, nucleus ambiguus; dmnX, dorsal motor nucleus of the vagus nerve; NTS, nucleus of the tractus solitarii; AP, area postrema.


30

Overview of vagus nerve homeostatic functions 2.2

The VN functions as an « unconscious inner brain », integrating peripheral signals via VAN

and providing metabolic feedback regulation via VEN (30). With its broad neuronal network,

the VN plays a role in the regulation of numerous organ functions, such as breathing, heart

rate, blood pressure and vascular stiffness (30). It would be beyond the scope of this

introduction to detail the role of the VN in regulating all these functions but I can, as an

example, briefly mention the predominant role of the VN in mediating two well-described

reflexes. First, the cough reflex is largely mediated by VAN inputs from the larynx and

trachea via activation of so-called cough receptors on C and Aδ fibers. The appropriate

response to this stimulation originates in the NTS: the nucleus ambiguous is activated and

efferent fibers controlling respiratory and pharyngolaryngeal muscles generate the cough

(33). Moreover, a second well-described vagal loop is the Hering-Breur reflex, also known as

respiratory sinus arrhythmia: sensing of bronchial stretch by VAN is relayed in the NTS to

activate cardiac vagal motor neurons. This results in the heart rate variation due to

respiration (33).

In the next paragraphs, I will describe the role of the VN in the following physiological

functions: the control of eating and energy expenditure, as well as the regulation of glucose

metabolism.

Role of the vagus nerve in the control of eating a nd post-absorptive 2.3

mechanisms

As mentioned earlier, the VN densely innervates the GI tract. VAN sense post-absorptive

mechanical and chemical signals relevant for the control of food intake, and VEN are

involved in the control of digestion and absorption of nutrients. Both VAN and VEN play a

role in the control of eating at several levels. In the stomach, they mostly relay mechanical

stimuli and control gastric secretion; in the intestine, VAN are mostly involved in the

chemosensing of gut peptides.


31

2.3.1 Vagal sensing and control of gastric motility and secretion

Electrophysiological and physiological evidence suggests that VAN convey information about

the volume of food ingested via gastric distension. Indeed, gastric loads were shown to

activate VAN (34) and induce neuronal activation in the NTS and DMV (35). Moreover,

infused gastric loads of identical volumes but of different nutritional content were associated

with similar firing rates of gastric VAN (36) and similar eating responses (37). Most

mechanosensitive VAN terminals innervating the stomach are intraganglionic laminar

endings (IGLE) located in the connective “capsule” of the myenteric plexus (38), and

intramuscular arrays (IMA) located in the stomach longitudinal and circular muscle layers

(39). Although structurally different, both terminals respond to the mechanical cues

generated by stretch and contraction of the stomach muscle layers (40,41). How they

functionally differ is, however, still unclear (42). A less abundant subgroup of gastric VAN

terminates in the gastric mucosa and is sensitive to touch (41,43), i.e., to the direct contact

with a gastric load, rather than gastric distension. Together, these structural and functional

observations led to the view that gastric VAN primarily integrate the mechanical signals

induced by a gastric load.

Several observations suggested that VE are involved in the coordination of secretion and

motility of the GI tract. First, vagotomy acutely impairs the reservoir function of the stomach

(44) and differentially modulate the gastric emptying of solids and fluids (de Lartigue,

unpublished). Moreover, numerous GI pathologies in which gastric motility or secretion is

impaired (gastroparesis, esophageal reflux, …) are associated with dysfunctional vagal

reflexes (45). Two distinct pathways in the parasympathetic postganglionic neurons mediate

the effects of gastric VEN on gastric motility. The first is an excitatory cholinergic pathway

that results in the activation of muscarinic cholinergic receptors in the gastric smooth muscle

and eventually increases gastric motility and secretion. The second pathway is a non-

cholinergic, non-adrenergic pathway that inhibits gastric function via the release of nitric


32

oxide and or vasoactive intestinal neuropeptide (46). Together, these two excitatory and

inhibitory pathways provide the possibility of a precise descending control of gut motility.

Preganglionic DMV neurons projecting to the stomach show spontaneous tonic activity in a

pacemaker-like fashion (47) but are under constant modulation by humoral or descending

CNS inputs and via VAN inputs (via a relay in the NTS) (see (48) and (45) for review).

Neuronal projections from the NTS to the DMV (mostly glutamatergic, GABAergic and

catecholaminergic inputs) directly link VAN to VEN, providing the neuronal substrate for so-

called « vago-vagal reflexes ». Several vago-vagal reflexes have been studied and are

relevant for the maintenance of the regional coordination of food transit. The gastric

accommodation reflex, for example, causes a reduction in pressure of the stomach corpus

when the antrum is stretched (49). This allows for filling of the stomach without an increase in

gastric pressure and, thus, accommodates various meal sizes. Moreover, it induces a

retropropulsion of the chyme that greatly participates in the mechanical disruption of food

particles during digestion. This reflex is initiated by gastric vagal mechanoreceptors from the

antrum that cause, after a relay in the NTS, the inhibition of DMV neurons maintaining gastric

tone in the corpus (50). Vago-vagal reflexes also coordinate gastric secretion during the

gastric phase of the digestion: VEN cholinergic inputs to the stomach stimulate H+ release

by parietal cells. This acid release is further amplified by the vagally-induced release of

histamine from ECL cells. Moreover, among other contributors, VEN induce the release of

gastrin from the G-cells and attenuate the inhibition of gastrin release by somatostatin.

Gastrin, in turn, stimulates H+ secretion from parietal cells and the release of peptidases by

chief cells. Together, these vagally-mediated mechanisms coordinate the secretion of acids

and enzymes needed for the chemical processing of the ingested food.

In summary, gastric VEN activation via vago-vagal reflexes results in:

a/ the regional contraction and relaxation of the stomach, which facilitates the mechanical

disruption of the food and allows its transit to the duodenum for further nutrient absorption


33

b/ the control of acid secretion and the release of digestive enzymes resulting in the chemical

breakdown of food particles.

2.3.2 Vagal chemosensing in the intestine

Several early observations support the view that intestinal VAN are predominantly involved in

chemosensing rather than mechanosensing. First, although the several types of

mechanoreceptors described in the stomach (IMA, IGLE and mucosal afferents) are also

present in the small intestine, their abundance is greatly reduced (51). Moreover, nutrient

infusions into the proximal small intestine elicited a nutrient-specific pattern of activation in

intestinal VAN (52).

As demonstrated by tracing studies, VAN are mainly present within the crypts and villi of the

lamina propria (Fig. 6) (53). They are therefore in close proximity to the intestinal epithelial

cells, but not directly exposed to luminal contents. It is therefore unlikely that their principal

role is to sense luminal nutrients (51). They are, however, ideally positioned to sense the gut

peptides released by enteroendocrine cells (54). Moreover, intestinal VAN express receptors

for a wide spectrum of gut peptides (55). Due to the proximity between enteroendocrine cells

and VAN terminals, the classical view is that VAN sense gut peptides in a paracrine-like

fashion. Moreover, it was recently reported that some enteroendocrine cells show long axon-

like processes called « neuropods » (Fig. 6) (56). This opened the possibility that

enteroendocrine cells form synapses with sensory fibers. The in vivo functionality of these

neuropods and how exactly they connect to VAN remains to be further explored.

Gut-derived CCK is the best-characterized example of mediation of a gut peptide’s effects on

food intake and gastric motility by VAN. Indeed, VAN express the CCKa receptor (CCKaR),

CCK stimulates VAN discharge and VAN lesions block the effects of CCK on food intake and

gastric motility. Moreover, CCK regulates the expression of certain receptors and

neuropeptides in VAN to encode the nutritional status and modulate the capacity of food

stimulation (see (58) for review). There is also substantial evidence that VAN mediate the


34

effects of other anorectic peptides, including the peptide YY (59) and the glucagon-like

peptide-1 (see below and chapter 2).

Figure 6: Innervation of the gut mucosa by vagal af ferent neurons and possible synaptic connections with neuropods of enteroendocrine cells (Left) DiI tracing from the nodose ganglia show mucosal endings close to the epithelium (e) in a villus of the rat proximal duodenum (53) (Right) Schematic model of an enteroendocrine cell’s neuropod, in close apposition with an enteric glial cell (57), modified.

To summarize, intestinal VAN mostly encode nutrient composition via the sensing of gut

peptides and are thus involved in the control of FI. While gastric VEN are clearly involved in

the control of gastric motility and secretion, the role of intestinal VEN is unclear. Indeed, the

enteric nervous system (ENS) is the main regulator of intestinal motility but a functional

interaction between VEN and ENS is probable. To which extent VEN can control intestinal

motility is, however, unknown.

Role of the vagus nerve in the control of glucose homeostasis 2.4

This section will discuss the current literature describing the role of the VN in glucoregulatory

processes in the pancreas and the liver.

First, VAN terminating in the hepatoportal bed are implicated in the control of glucose

homeostasis. Indeed, a glucosensory site localized in the HPV, upstream of the hepatic hilus

(60) is connected to VAN (61). These afferents fire upon glucose stimulation in a

concentration-dependent manner (62) and project to the NTS (63), which makes neural


35

connections to hypothalamic glucoregulatory centers (64). The exact role of VAN in this

process has, however, not been fully established, mainly due to the use of unspecific

lesioning techniques (common hepatic branch vagotomy). In the next chapter, we will

discuss the possibility that VAN can mediate an effect on glucose homeostasis initiated by

the activation of GLP-1 receptors. The activation of the VAN of the hepatoportal bed has

been shown to increase whole-body glucose utilization by both insulin-dependent and –

independent mechanisms (65). Briefly, recent findings support the idea of a gut-to-brain-to-

periphery axis controlling glucose homeostasis via a VEN feedback on hepatic glucose

production (hepato-incretin axis) (66) and pancreatic islet insulin release (neuroincretin

effect) (67).

Role of the vagus nerve in the control of energy e xpenditure 2.5

Indications for a role of the VN in the control of energy expenditure come in part from animal

and human studies involving electrical VN stimulation (VNS). Indeed, stimulation of the

cervical VAN decreased brown adipose tissue (BAT) temperature and indicators of energy

expenditure in anesthetized rats (68,69). Conversely, auricular VNS in diet-induced obese

rats induced weight loss without changes in daily intake, indicative for an increased energy

expenditure (70). In humans, a temporary interruption of VNS in epileptic patients led to a

reduction in energy expenditure, correlating with a reduced BAT activity (71). On the other

hand, another study involving chronic subdiaphragmatic VNS in pigs showed no changes in

energy expenditure, although the measurement was technically limited (72). These

apparently contradictory findings may be explained by the use of slightly different models of

VNS. Moreover, several molecular manipulations of VAN in mice, have led to changes in

energy expenditure. Indeed, downregulation of the nuclear receptors peroxisome proliferator-

activated receptor γ (PPARγ) and liver X receptor α and β (LXRα/β) in paired-like homeobox

2b (Phox2b)-expressing neurons amplified the high-fat diet (HFD)-induced increase in

energy expenditure (73,74). Although there are concerns with respect to the use of Phox2b

as a marker of VAN (75), these results support the idea that disrupting vagal sensing can


36

impact whole-body energy expenditure. A recurrent finding, though, is the correlation

between changes in whole-body energy expenditure and interscapular BAT activity (69-71).

How VAN inputs to the NTS affect the sympathetic outputs to BAT is largely unknown (68)

and remains to be further investigated.

3 The glucagon-like peptide-1, a multifaceted pepti de involved in

energy homeostasis

Secretion of intestinal GLP-1 3.1

Intestinal GLP-1 is secreted by enteroendocrine L-cells. Such cells are found all along the

intestine; within the small intestine, the highest density of L-cells is found in the ileum (76,77).

The peptide GLP-1 is encoded by the preproglucagon gene and further post-translationally

processed by the prohormone convertase PC1/3 (78). GLP-1 is found in vivo in several

forms but only the GLP-1(7-36) and GLP-1(7-37) are believed to be active: both forms were

shown to have similar biological effects and kinetics in humans, although the GLP-1(7-36)

form is predominant (79). L-cells release GLP-1 upon nutrient stimulation: the current view is

that several G-protein coupled receptors (GPCRs) expressed by the L-cells sense the

digestion products of carbohydrates, lipids and proteins (80). The intracellular signals

generated by the binding of these compounds to the receptors interact with glucose

utilization and cation transporters to induce a calcium efflux from the endoplasmic reticulum,

with the resulting increase in intracellular calcium triggering GLP-1 release (80) (Fig. 7).

Moreover, recent findings indicate the possibility that certain fatty acids or bacterial

compounds induce the release of GLP-1 via a metabolic effect on L-cells, involving

mitochondrial uncoupling and changes in ATP production (81,82). After its release, GLP-1 is

rapidly inactivated by the enzyme dipeptidyl peptidase-IV (DPP-IV) via N-terminal

degradation at the alanine position 2 (83). Plasmatic and liver concentrations of DPP-IV are

high, thus limiting the half-life of GLP-1 in the systemic circulation (84).


37

Figure 7: Summary of the known nutrient and not-nut rient related pathways facilitating the release of incretin hormones, including GLP-1, by e nteroendocrine cells (80) Abbreviations: CaSR, calcium-sensing receptor; CnR1, cannabinoid receptor 1; Epac, exchange protein activated by cAMP; FFAR, free fatty acid receptor; GLUT, glucose transporter; GPR, G-protein coupled receptor; PKA, phosphate kinase A; SGLT, sodium glucose transporter; SSTR, somatostatin receptor

The GLP-1 receptor (GLP-1R) is a 7 transmembrane-spanning heterotrimeric GPCR class B

(85). It is expressed in many tissues, including pancreatic α-, β-, and δ-cells but also in heart,

muscles, white adipose tissue, stomach, skin and numerous brain regions (86). The GLP-1R

has been shown to couple to multiple intracellular downstream pathways (adenylate cyclase,

phospholipase C (PLC), phosphokinase A or C (PKA, PKC), phosphoinositide 3-kinase

(PI3K), mitogen-activated protein kinase (MAPK), …) depending on the cell type studied

(87,88) and can thus be either a Gs-, Gi-, Gq- or Go-coupled receptor (89). Desensitization

has been observed in vitro in β-cell models (90,91), but, in vivo, subchronic GLP-1R agonist

treatment was not associated with a reduction of the GLP-1R-dependent insulin response

(92). Nevertheless, internalization of the ligand/receptor complex and trafficking between

membrane and intracytoplasmic locations were observed in vivo in the brain (93) and in the

NG (94). These observations led to the idea that, in certain tissues, GLP-1R sensitivity can

be temporarily modulated by previous GLP-1R activation and the nutritional status.


38

The biological functions of GLP-1 in glucose and e nergy homeostasis 3.2

3.2.1 Peripheral GLP-1’s effects on glucose homeost asis

GLP-1 was initially identified as an incretin hormone both in rats and humans (95,96):

together with the gastric inhibitory peptide (GIP), GLP-1 enhances the glucose-induced

insulin secretion and, thus, limits postprandial glucose excursion. The insulinotropic effect of

GLP-1 is at least partly mediated by the direct binding of GLP-1 to its receptor on the

pancreatic β-cells (97). The β-cell GLP-1R activates the adenylate cyclase, via a stimulatory

G-protein, and increases intracellular cAMP levels. A broad array of intracellular pathways

are then activated by PKA and the exchange protein activated by cAMP-2 (Epac2) that lead

to β-cell membrane depolarization, increases in intracellular ATP and Ca2+ and, eventually,

to the release of insulin storage vesicles (see (98) for review). Importantly, the insulinotropic

action of GLP-1 is strictly glucose-dependent, as shown by the absence of insulin secretion

in islets perfused with low glucose levels (99). Beyond the enhancement of glucose-induced

insulin secretion, GLP-1R activation on the β-cells also promotes insulin synthesis at the

transcriptional (translocation of the cAMP responsive binding element protein (CREB)

promoter to the nucleus) and post-transcriptional level (enhancement of mRNA stability)

(100). Moreover, GLP-1R activation increases β-cell mass in vitro and in vivo, via increased

β-cell proliferation and reduced β-cell apoptosis (101,102). This protective mechanism may

involve an intra-islet secretion of GLP-1 by α-cells, shifting their prohormone convertase

expression profile from PC2 (leading to glucagon synthesis) to PC1/3 (67). In the pancreatic

islets, GLP-1 also inhibits glucagon secretion by α-cells and stimulates somatostatin

secretion by δ-cells, likely via a direct effect on their GLP-1R (103). In vitro evidence also

suggests that GLP-1 controls glucose homeostasis via direct insulin-independent effects on

muscles, adipose tissue and liver (see (86) for review). In vivo, it is difficult to highlight

potential direct GLP-1 effects on glucose homeostasis because insulin, glucagon and

somatostatin secretions are affected by the pharmacological administration of GLP-1.

Experiments in humans undergoing an euglycemic hyperinsulinemic clamp together with a


39

normalization of somatostatin and glucagon levels have not shown insulin-independent

effects of GLP-1 on glucose infusion rates (104). Further studies suggested little to no acute

or chronic insulin-independent effects of GLP-1 on glucose turnover at the whole-body level

(97,105).

3.2.2 Peripheral effects of GLP-1 on eating

The effects of endogenous GLP-1 and exogenous GLP-1R agonists on eating will be further

discussed along the thesis. Briefly, I would like to provide here evidence that endogenous

GLP-1 inhibits eating. Peripheral administration of GLP-1 or GLP-1R agonists reduced food

intake in a variety of species, including rodents, primates and humans (106-108). The classic

view is that endogenous intestinal GLP-1 is mainly implicated in meal termination: indeed,

GLP-1 secretion is rapidly induced by luminal nutrient stimulation and GLP-1 HPV

concentrations rise already during a meal, which is consistent with an intra-meal effect.

Moreover, acute intraperitoneal (IP) injections of exendin-4 (Ex-4, a GLP-1R agonist) or

GLP-1, shortly before or during a meal, reduced food intake via an effect on meal size (i.e.,

by inducing satiation) (109,110). The mechanisms by which endogenous intestinal GLP-1

reduces meal size have not been fully elucidated (see (111) for review). An endocrine effect

on brain regions expressing GLP-1R and involved in the control of eating has received some

experimental support (112). The very limited GLP-1 half-life in the plasma, however, is more

consistent with a paracrine effect on VAN. In the next chapter, the potential vagal mediation

of GLP-1’s effects on food intake and glycemia will be further discussed. Intestinal GLP-1 is

also thought to exert a potent control on gastric acid secretion and motility via the VN (113).

Slowing nutrient transit in the stomach, in turn, affects glucose homeostasis (slower nutrient

absorption). Moreover, it likely increases the duration of nutrient interactions with vagal

mechanoreceptors and may therefore participate in the eating inhibition induced by GLP-1.


40

Off note, diverse other organs have been shown to express the GLP-1R (heart, bones,

kidney, testis, …) and a broad spectrum of functions have been reported. It is beyond the

scope of this introduction to detail them, but the most studied non-classic functions involve

cardioprotection (see (114) for review) and neuroprotection (see (115) for review) (Fig. 8).

Figure 8: overview of the known effects of GLP-1 on eating, glucose homeostasis, cardiac and brain functions (86)

3.2.3 Evidence for an effect of GLP-1 on energy exp enditure

There is evidence indicating that direct activation of central GLP-1R in the ventromedial

hypothalamus (VMH) increases energy expenditure via the activation of BAT thermogenesis

(116). Whether this mechanism is relevant for the control of energy expenditure by

endogenous GLP-1 has not been established. In addition, the effect of peripheral GLP-1R

activation on energy expenditure has scarcely been studied. Conflicting reports regarding the

effects of GLP-1 on energy expenditure can be found in the literature (Table 1). Briefly,

elevated circulating GLP-1 levels have been associated with increased energy expenditure in

mice (117) and humans (118). This has been supported by the fact that central as well as


41

peripheral administration of GLP-1 or GLP-1R agonists acutely increased VO2 or heat

production in mice (116), rats (119-121) and humans (121). Inversely, others reported no

change (122) or an acute decrease in VO2, heat production or diet-induced thermogenesis in

mice (123), hamsters (124) and humans (125,126) (Table 1). These discrepancies may be

explained by the different doses, agonists (GLP-1, Ex-4, liraglutide) or models (mice, rats,

hamsters, humans) used in these studies. More importantly, the different routes of

administration used (IP, intravenous (IV) or subcutaneous (SC)) may provide access to

distinct GLP-1R populations and, hence, contribute to the conflicting reports.

The central GLP-1 system: effects and interaction with the periphery 3.3

There is early evidence for the presence of glucagon-like peptides in the mammalian brain

(127,128). More recently, the cell bodies of PPG-expressing neurons have been localized in

the NTS, the reticular formation and the olfactory bulb (129-131). PPG-expressing neurons

project to most of the known GLP-1R-expressing brain areas (54,132), with an important

density of projections to the paraventricular and dorsomedial nuclei of the hypothalamus

(PVN and DMH) (133). Data from a mouse model expressing the yellow fluorescent protein

(YFP) in PPG-expressing cells suggested that GLP-1 neurons in the olfactory bulb are

mainly local interneurons, and that most of the projections seen in the brain originate from

brainstem PPG neurons (132,134). PPG neuronal projections are found in many GLP-1R-

expressing brain areas, suggesting a functional role for the central GLP-1 system. Several

studies (reviewed in (135)) used the central administration of GLP-1 or GLP-1R agonists to

test the potential functions of central GLP-1R populations: the range of response observed

was strikingly similar to the effects of peripheral GLP-1, i.e., notably a suppression of food

intake and an improved glucose tolerance. Pharmacological activation of brain GLP-1R,

however, is not a direct assessment of the physiological role of the central GLP-1 system.

More recently, a PPG knockdown in the NTS of rats via RNA interference indicated a

physiological role for NTS PPG neurons to regulate HFD-induced weight gain and glucose

intolerance (136). Moreover, an activation of Phox2b neurons in the NTS (among which are


42

Model Drug Route Dose Effects described ReferenceMice Ex4 ICV 1 µg Reduced VO2 / Reduced RER Baggio 2004 (123)

Mice Liraglutide ICV 3 µg Increased EE Beiroa 2004 (116)

Rat GLP-1 ICV 1 µg Increased VO2 (for 2h, in lean but not in obese rats) Hwa 1998 (119)

Rats GLP-1 ICV 3 nmol Increased VO2 (still occurs in vagotomized rats) Osaka 2005 (120)

Humans GLP-1 IV (post meal) 50 pmol/kg/h Decreased EE and DIT Flint 2000 (125)

Humans GLP-1 IV (post meal) 50 pmol/kg/h Decreased EE and DIT Flint 2001(126)

Humans GLP-1 IV (180' infusions) Increased VO2, VCO2, EE Shalev 1997 (121)

Rats GLP-1 IV 0.2 to 5 nmol Increased VO2 and increased body temperature Osaka 2005 (120)

Mice Ex4 IP 1.5 µg Reduced VO2 / Reduced RER (for about 2h) Baggio 2004 (123)

Rats Ex4 / Ex9 SC 5 / 50 µg/kg No changes in EE Abegg 2013 (122)

Mice Increased VO2 (throughout the day) Conarello 2002 (117)

Humans Correlative study Fasting levels of GLP-1 are positively correlated with basal EE, independent of body composition

Pannacciulli 2006 (118)

DPP IV -/-

Table 1: Summary of available evidence that central or peripheral GLP-1R modulate whole-body energy ex penditure and its surrogate markers (Krieger 2015, unpublished) Abbreviations: DIT, diet-induced thermogenesis; DPP IV -/-, dipeptidyl-peptidase IV knockout mice; EE, energy expenditure; Ex-4, exendin-4; Ex-9, exendin-9; ICV, intracerebroventricular; IP, intraperitoneal; IV, intravenous; RER, respiratory exchange ratio; SC; subcutaneous.

Reduced EE during fasting


43

the PPG neurons) using “Designer Receptors Exclusively Activated by Designer Drugs”

(DREADD) supported a role for PPG neurons in reducing the intake of palatable foods (137).

Further experiments involving site- and time-specific manipulations of central GLP-1R are

however needed to critically examine the physiological relevance of endogenous central

GLP-1 and the roles of distinct GLP-1R populations.

As mentioned earlier, the functions of central and peripheral GLP-1 seem redundant. Little is

known, though, about the interplay between intestinal and central GLP-1 (138) and whether

the peripheral release of GLP-1 in a postprandial situation induces the release of central

GLP-1 is still unclear. The possibility that peripheral GLP-1 directly activates PPG neurons,

via GLP-1R located on dendrites in the area postrema (AP) for example, is very unlikely.

Indeed, PPG neurons do not express the GLP-1R and are not electrically activated by GLP-1

(139). Alternatively, GLP-1-sensitive neurons in the AP may serve as a relay for the

activation of PPG neurons: GLP-1R expressing neurons in the AP, however, are mostly

catecholaminergic neurons and recent results indicate that NTS PPG neurons do not receive

catecholaminergic inputs (140). Peripheral GLP-1 may activate PPG neurons via VAN,

because PPG neurons receive direct monosynaptic glutamatergic inputs from VAN (139).

Unpublished observations (from Lutz and Vrang), however, did not show co-localization

between Ex-4-induced c-Fos and GLP-1 immunoreactivity in the NTS of rats. In brief,

although it is tempting to think that peripheral GLP-1 can stimulate central GLP-1 secretion,

there is limited functional and mechanistic evidence to support this idea. The peripheral and

central GLP-1 systems may be two separate entities with converging functions. Other

peripheral signals activate PPG neurons: indeed, the electrical activity of these neurons is

modulated by other peripheral signals such as CCK, leptin (139) and gastric distension (141).

The current understanding of how central GLP-1 neurons interact with the periphery and

exert their functions was summarized by (142) (see Fig. 9).


44

Figure 9: Current understanding of PPG neurons acti vation by peripheral signals, projections and functions (142). Abbreviations: AMG, amygdala; ARC, arcuate nucleus of the hypothalamus; BNST, bed nucleus of the stria terminalis; BP, blood pressure; CSF, cerebrospinal fluid; DMH, dorsomedial nucleus of the hypothalamus; DMNX, dorsal motor nucleus of the vagus nerve; HR, heart rate; NAc, nucleus accumbens; PPG, preproglucagon; PVH, paraventricular nucleus of the hypothalamus; VLM, ventrolateral medulla; VTA, ventral tegmental area.

The GLP-1R agonists for anti-diabetes and anti-obe sity therapies 3.4

While the studies presented in this thesis mainly investigate the mechanisms by which

endogenous GLP-1 exerts its functions, it is important to note that GLP-1R agonists are

currently used as anti-diabetes and anti-obesity treatments, and that their action does not

necessarily mimic the mechanisms of action of endogenous GLP-1. The differences between

endogenous GLP-1 and the pharmacological administration of GLP-1R agonists are

highlighted.

Several pharmaceutical companies have tested numerous GLP-1R agonists for their ability

to improve glucose tolerance in type 2 diabetic patients. These GLP-1R agonists are either

short-acting (e.g., exenatide, also called Ex-4, or lixisenatide) or long-acting compounds

(e.g., liraglutide, albiglutide, dulaglutide, …) (143). It is generally conceptualized that short-

acting compounds primarily have a positive effect on glycemia via a strong inhibition on

gastric emptying combined with an insulinotropic and glucagonostatic action. Long-acting


45

compounds, however, seem to exert a more profound effect on glucose homeostasis.

Indeed, chronic administration of GLP-1R agonists improved glucose tolerance in obese

individuals, via mechanisms that involve an improved glucose-induced insulin secretion and

reduced glucagon secretion, but also an improved capacity of resistant β-cells to secrete

insulin and the improvement of β-cell mass (144). Reports that chronic administration of

GLP-1R agonists induced weight loss in obese patients led to the recent FDA approval of

liraglutide as a weight loss drug (145). The mechanisms, however, are less clear. In a recent

study, Secher and colleagues showed that liraglutide has direct access to the brain and

induces weight loss that is mediated via GLP-1R in the Arc (93).

This study, among others, highlights the idea that the mechanisms by which GLP-1R

agonists inhibit eating and regulate glucose homeostasis, may be distinct from the

mechanisms by which endogenous peripheral GLP-1 fulfills similar functions. In other words,

pharmacological administration of GLP-1R agonists may bypass the normal route by which

intestinal GLP-1 signals the brain and may therefore rather mimic several aspects of the

central GLP-1 system.

In summary, we can conclude that endogenous peripheral and central GLP-1 exert several

physiological functions including the control of food intake, the regulation of glycemia and

energy expenditure. The physiological mechanism by which endogenous peripheral GLP-1

signals the brain is unclear and may not be recapitulated by the use of GLP-1R agonists.

Understanding this physiological mechanism may provide a new approach to target the GLP-

1 system.

4 Aims of the thesis

Early in my thesis, I was intrigued and interested in the concept of a gut-brain axis controlling

the energy balance. I rapidly focused on the possibility that the VN, specifically VAN, can

mediate the effects of peripheral endogenous GLP-1 on food intake, glycemia and energy

expenditure. Therefore, I structured my thesis around the following aims.


46

The first aim was to evaluate the available evidence that the VN mediates the effects of

peripheral endogenous GLP-1 on food intake and glycemia. Indeed, a broad array of studies

points towards a role of VAN in mediating the endogenous peripheral GLP-1 signal to the

brain. Nonetheless, these studies only provide indirect evidence for a role of VAN GLP-1R

signaling in the normal control of eating and regulation of glycemia. The article found in

CHAPTER 2 reviews these studies.

This literature review showed that the experimental methods used to date did not specifically

test the physiological role of VAN GLP-1R in energy balance. Therefore, the second aim of

my thesis was to develop a new method to specifically test whether VAN relay a GLP-1

signal to the brain relevant for the control of food intake (including gastric emptying) and

glycemia. Thus, in CHAPTER 3 , I present the bilateral delivery of a lentiviral vector into the

NG as a way to specifically knock down GLP-1R in rat VAN. Building on this new method, we

critically examined the physiological role of the endogenous gut-derived GLP-1 acting on

vagal GLP-1R to control food intake and participate in gastric emptying and glucose

homeostasis.

As mentioned earlier, the presence of numerous neuropeptides has been reported in the NG,

and some of these peptides have been implicated in the control of food intake. Whether a

vagally-mediated GLP-1 signal involves such neuropeptides is, however, unknown. Based on

our previous results, we hypothesized that the effects of GLP-1R activation in the VAN may

rely on the expression and the release of one of these neuropeptides. CHAPTER 4 presents

our current results showing that CART is a necessary mediator of the control of food intake

and gastric emptying by GLP-1.

In the initial literature review presented in Chapter 2, I did not comment on a role of the VN in

the effects of GLP-1 on energy expenditure because the role of peripheral GLP-1 on energy

expenditure itself was understudied and remained unclear. Hence, the fourth aim of the study

was to investigate whether endogenous peripheral GLP-1 can modulate energy expenditure

and if so, whether this is vagally-mediated. Moreover, based on preliminary reports indicating


47

a modulation of BAT activity by VNS or central GLP-1 activation, we also investigated the

role of the BAT in a potential modulation of energy expenditure by peripheral GLP-1. Overall,

in CHAPTER 5, we present evidence that, with high-fat diet exposure, the activation of VAN

GLP-1R reduces energy expenditure and BAT activity and that a specific attenuation of this

gut-BAT crosstalk can alleviate the course of HFD-induced obesity.

In summary, this thesis aims at providing a clearer view on the vagal mediation of GLP-1’s

effects in the control of energy balance, both in a physiological and obesogenic context.

The present thesis comprises research articles originally written for peer-reviewed journal

publications. The articles are presented in the form in which they have been published

(CHAPTERS 2 and 3 ) or in a form that reflects our current understanding of ongoing studies

(CHAPTERS 4 and 5 ). A more detailed description of the research aims can be found in

each chapter, as well as a discussion of the findings. Eventually, considering these articles

together, new scientific questions emerged, that we will discuss in CHAPTER 6.


48

REFERENCES

1. World Health Organization - Obesity and overweight fact sheet n°311. [Internet] 2015. [cited Nov 2015] Available from: http://www.who.int/mediacentre/factsheets/fs311/en/

2. Pi-Sunyer FX. Comorbidities of overweight and obesity: current evidence and research issues. Med Sci Sports Exerc. 1999; 31(11):S602.

3. Walls HL, Backholer K, Proietto J, McNeil JJ. Obesity and Trends in Life Expectancy. J Obes. 2012; (1-2):1–4.

4. Olshansky SJ, Passaro DJ, Hershow RC, Layden J, Carnes BA, Brody J, Hayflick L, Butler R, Allison D, Ludwig D. A potential decline in life expectancy in the United States in the 21st century. N Engl J Med. 2005; 352(11):1138–45.

5. Campos P. The epidemiology of overweight and obesity: public health crisis or moral panic? Int J Epidemiol. 2005; 35(1):55–60.

6. Ogden CL, Yanovski SZ, Carroll MD, Flegal KM. The Epidemiology of Obesity. Gastroenterology. 2007; 132(6):2087–102.

7. Bell JA, Hamer M, Sabia S, Singh-Manoux A, Batty GD, Kivimaki M. The Natural Course of Healthy Obesity Over 20 Years. J Am Coll Cardiol. 2015; 65(1):101–2.

8. Egger G, Dixon J. Beyond Obesity and Lifestyle: A Review of 21st Century Chronic Disease Determinants. BioMed Res Int. 2014; 731685:1-12.

9. Cawley J. An economy of scales: A selective review of obesity's economic causes, consequences, and solutions. J Health Econ. 2015; 43:244-268.

10. Christian T, Rashad I. Trends in U.S. food prices, 1950-2007. Econ Hum Biol. 2009;7(1):113–20.

11. McLaren L. Socioeconomic Status and Obesity. Epidemiol Rev. 2007; 29(1):29–48.

12. Finucane MM, Stevens GA, Cowan MJ, Danaei G, Lin JK, Paciorek CJ, Singh MG, Gutierrez HR, Lu Y, Bahalim AN. National, regional, and global trends in body-mass index since 1980: systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9·1 million participants. Lancet. 2011; 377(9765):557–67.

13. Maes HHM, Neale MC, Eaves LJ. Genetic and Environmental Factors in Relative Body Weight and Human Adiposity. Behav Genet. 1997; 27(4):325–51.

14. Locke AE, Kahali B, Berndt SI, Justice AE, Pers TH, Day FR, and 342 more authors. Genetic studies of body mass index yield new insights for obesity biology. Nature. 2015; 518(7538):197–206.

15. Douketis JD, Macie C, Thabane L, Williamson DF. Systematic review of long-term weight loss studies in obese adults: clinical significance and applicability to clinical practice. Int J Obes. 2005; 29(10):1153–67.

16. Li G, Zhang P, Wang J, Gregg EW, Yang W, Gong Q, Li H, Jiang Y, An Y, Shuai Y, Zhang B, Zhang J, Thompson TJ, Gerzoff RB, Roglic G, Hu Y, Bennett PH. The


49

long-term effect of lifestyle interventions to prevent diabetes in the China Da Qing Diabetes Prevention Study: a 20-year follow-up study. Lancet. 2000; 371(9626):1783–9.

17. Borys J-M, Le Bodo Y, Jebb SA, Seidell JC, Summerbell C, Richard D, De Henauw S, Moreno LA, Romon M, Visscher TL, Raffin S, Swinburn B; EEN Study Group. EPODE approach for childhood obesity prevention: methods, progress and international development. Obes Rev. 2012; 13(4):299–315.

18. Colquitt JL, Pickett K, Loveman E, Frampton GK. Surgery for weight loss in adults. Cochrane Rev. 2014; 1-214

19. Gloy VL, Briel M, Bhatt DL, Kashyap SR, Schauer PR, Mingrone G, Bucher HC, Nordmann AJ. Bariatric surgery versus non-surgical treatment for obesity: a systematic review and meta-analysis of randomised controlled trials. BMJ. 2013; 347:f5934.

20. Livingston EH. The incidence of bariatric surgery has plateaued in the U.S. Am J Surg. 2010; 200(3):378–85.

21. Nguyen NT, Masoomi H, Magno CP, Nguyen X-MT, Laugenour K, Lane J. Trends in use of bariatric surgery, 2003-2008. J Am Coll Surg. 2011; 213(2):261–6.

22. Mayer J. Glucostatic mechanism of regulation of food intake. N Engl J of M. 1953; 1–4.

23. Woods SC. Metabolic signals and food intake. Forty years of progress. Appetite. 2013; 71:440–4.

24. Gibbs J, Young RC, Smith GP. Cholecystokinin elicits satiety in rats with open gastric fistulas. Nature. 1973; 245(5424):323–5.

25. Woods SC, D'Alessio DA. Central control of body weight and appetite. J Clin Endocrinol Metab. 2008; 93(11 Suppl 1):S37–50.

26. Woods SC. Signals that influence food intake and body weight. Physiol Behav. 2005; 86(5):709–16.

27. Begg DP, Woods SC. The endocrinology of food intake. Nat Rev Endocrinol. 2013; 9(10):584–97.

28. Rehfeld JF. Gastrointestinal hormones and their targets. Adv Exp Med Biol. 2014; 817:157–75.

29. Berthoud H-R. The vagus nerve, food intake and obesity. Regul Pept. 2008; 149(1-3):15–25.

30. Yuan H, Silberstein SD. Vagus Nerve and Vagus Nerve Stimulation, a Comprehensive Review: Part I. Headache. 2015; 56(1):71-8.

31. Berthoud HR, Neuhuber WL. Functional and chemical anatomy of the afferent vagal system. Auto Nuerosci Basic. 2000; 85(1-3):1–17.

32. de Lartigue G. Putative role of neuropeptides in vagal afferent signaling. Physiol Behav. 2014; 1–15.


50

33. Yuan H, Silberstein SD. Vagus Nerve and Vagus Nerve Stimulation, a Comprehensive Review: Part III. Headache. 2015; EPub head.12650

34. Schwartz GJ, McHugh PR, Moran TH. Integration of vagal afferent responses to gastric loads and cholecystokinin in rats. Am J Physiol. 1991; 261(1 Pt 2):R64–9.

35. Willing AE, Berthoud HR. Gastric distension-induced c-fos expression in catecholaminergic neurons of rat dorsal vagal complex. Am J Physiol. 1997; 272(1 Pt 2):R59–67.

36. Mathis C, Moran TH, Schwartz GJ. Load-sensitive rat gastric vagal afferents encode volume but not gastric nutrients. Am J Physiol. 1998; 274(2 Pt 2):R280–6.

37. Phillips RJ, Powley TL. Gastric volume rather than nutrient content inhibits food intake. Am J Physiol. 1996; 271(3 Pt 2):R766–9.

38. Neuhuber WL, Kressel M, Stark A, Berthoud HR. Vagal efferent and afferent innervation of the rat esophagus as demonstrated by anterograde DiI and DiA tracing: focus on myenteric ganglia. J Auton Nerv Syst. 1998; 70(1-2):92–102.

39. Berthoud HR, Powley TL. Vagal afferent innervation of the rat fundic stomach: morphological characterization of the gastric tension receptor. J Comp Neurol. 1992; 319(2):261–76.

40. Zagorodnyuk VP, Chen BN, Brookes SJ. Intraganglionic laminar endings are mechano-transduction sites of vagal tension receptors in the guinea-pig stomach. J Physiol. 2001; 534(Pt 1):255–68.

41. Page AJ, Martin CM, Blackshaw LA. Vagal mechanoreceptors and chemoreceptors in mouse stomach and esophagus. J Neurophysiol. 2002; 87(4):2095–103.

42. Phillips RJ, Powley TL. Tension and stretch receptors in gastrointestinal smooth muscle: re-evaluating vagal mechanoreceptor electrophysiology. Brain Res Brain Res Rev. 2000; 34(1-2):1–26.

43. Berthoud HR, Lynn PA, Blackshaw LA. Vagal and spinal mechanosensors in the rat stomach and colon have multiple receptive fields. Am J Physiol Regul Integr Comp Physiol. 2001; 280(5):R1371–81.

44. Li Y, Owyang C. Musings on the wanderer: what's new in our understanding of vago-vagal reflexes? V. Remodeling of vagus and enteric neural circuitry after vagal injury. Am J Physiol - Gastr L. 2003; 285(3):G461–9.

45. Browning KN, Travagli RA. Plasticity of vagal brainstem circuits in the control of gastrointestinal function. Auton Neurosci - Basic. 2011; 161(1-2):6–13.

46. Travagli RA, Hermann GE, Browning KN, Rogers RC. Brainstem circuits regulating gastric function. Annu Rev Physiol. 2006; 68:279–305.

47. Fogel R, Zhang X, Renehan WE. Relationships between the morphology and function of gastric and intestinal distention-sensitive neurons in the dorsal motor nucleus of the vagus. J Comp Neurol. 1996; 364(1):78–91.

48. Rogers RC, McTigue DM, Hermann GE. Vagovagal reflex control of digestion: afferent modulation by neural and “endoneurocrine” factors. Am J Physiol. 1995; 268(1 Pt 1):G1–10.


51

49. Abrahamsson H. Vagal Relaxation of the Stomach Induced from the Gastric Antrum. Acta Physiol Scand. 1973; 89(3):406–14.

50. McCann MJ, Rogers RC. Impact of antral mechanoreceptor activation on the vago-vagal reflex in the rat: functional zonation of responses. J Physiol. 1992;453:401–11.

51. Berthoud HR. Vagal and hormonal gut-brain communication: from satiation to satisfaction. Neurogastroenterol Motil. 2008; 20 Suppl 1:64–72.

52. Schwartz GJ, Moran TH. Duodenal nutrient exposure elicits nutrient-specific gut motility and vagal afferent signals in rat. Am J Physiol. 1998; 274(5 Pt 2):R1236–42.

53. Berthoud HR, Kressel M, Raybould HE, Neuhuber WL. Vagal sensors in the rat duodenal mucosa: distribution and structure as revealed by in vivo DiI-tracing. Anat Embryol. 1995; 191(3):203–12.

54. Richards P, Parker HE, Adriaenssens AE, Hodgson JM, Cork SC, Trapp S, Gribble FM, Reimann F. Identification and characterisation of glucagon-like peptide-1 receptor expressing cells using a new transgenic mouse model. Diabetes. 2014; 63(4):1224-33.

55. Raybould HE. Gut chemosensing: interactions between gut endocrine cells and visceral afferents. Auton Neurosci - Basic. 2010; 153(1-2):41–6.

56. Bohórquez DV, Liddle RA. Axon-like basal processes in enteroendocrine cells: characteristics and potential targets. Clin Transl Sci. 2011; 4(5):387–91.

57. Bohórquez DV, Samsa LA, Roholt A, Medicetty S, Chandra R, Liddle RA. An enteroendocrine cell-enteric glia connection revealed by 3D electron microscopy. PLoS ONE. 2014; 9(2):e89881.

58. Dockray GJ. Cholecystokinin and gut-brain signalling. Regul Pept. 2009; 155(1-3):6–10.

59. Koda S, Date Y, Murakami N, Shimbara T, Hanada T, Toshinai K, Niijima A, Furuya M, Inomata N, Osuye K, Nakazato M. The role of the vagal nerve in peripheral PYY3-36-induced feeding reduction in rats. Endocrinology. 2005; 146(5):2369–75.

60. Hevener AL, Bergman RN, Donovan CM. Novel glucosensor for hypoglycemic detection localized to the portal vein. Diabetes. 1997; 46(9):1521–5.

61. Martin JR, Novin D, Vanderweele DA. Loss of glucagon suppression of feeding after vagotomy in rats. Am J Physiol. 1978; 234(3):E314–8.

62. Niijima A. The effect of D-glucose on the firing rate of glucose-sensitive vagal afferents in the liver in comparison with the effect of 2-deoxy-D-glucose. J Auton Nerv Syst. 1984 May;10(3-4):255–60.

63. Adachi A, Shimizu N, Oomura Y, Kobáshi M. Convergence of hepatoportal glucose-sensitive afferent signals to glucose-sensitive units within the nucleus of the solitary tract. Neurosci Lett. 1984; 46(2):215–8.

64. Shimizu N, Oomura Y, Novin D, Grijalva CV, Cooper PH. Functional correlations between lateral hypothalamic glucose-sensitive neurons and hepatic portal glucose-sensitive units in rat. Brain Res. 1983; 265(1):49–54.


52

65. Burcelin R, Dolci W, Thorens B. Portal glucose infusion in the mouse induces hypoglycemia: evidence that the hepatoportal glucose sensor stimulates glucose utilization. Diabetes. 2000; 49(10):1635–42.

66. Ahrén B. Hepato-incretin function of GLP-1: novel concept and target in type 1 diabetes. Diabetes. 2015; 64(3):715–7.

67. Donath MY, Burcelin R. GLP-1 effects on islets: hormonal, neuronal, or paracrine? Diabetes Care. 2013 Aug;36 Suppl 2:S145–8.

68. Morrison SF, Madden CJ, Tupone D. Central Neural Regulation of Brown Adipose Tissue Thermogenesis and Energy Expenditure. Cell Metabolism. 19(5):741–56.

69. Morrison SF, Tupone D, Madden CJ. Inhibition of brown adipose tissue (BAT) and shivering thermogenesis by vagal nerve stimulation (VNS). Auton Neurosci-Basic. 2015; 192:2–3.

70. Han L, Jian-Bin Z, Chen X, Qing-Qing T, Wei-Xing S, Jing-Zhu Z, Jian-De C, Yin-Ping W. Effects and mechanisms of auricular vagus nerve stimulation on high-fat-diet-induced obese rats. Nutrition. 2015; 31(11-12):1416–22.

71. Vijgen GHEJ, Bouvy ND, Leenen L, Rijkers K, Cornips E, Majoie M, Brans B, van Marken Lichtenbelt WD. Vagus nerve stimulation increases energy expenditure: relation to brown adipose tissue activity. PLoS ONE. 2013; 8(10):e77221.

72. Val-Laillet D, Biraben A, Randuineau G, Malbert CH. Chronic vagus nerve stimulation decreased weight gain, food consumption and sweet craving in adult obese minipigs. Appetite. 2010; 55(2):245–52.

73. Liu C, Bookout AL, Lee S, Sun K, Jia L, Lee C, Udit S, Deng Y, Scherer PE, Mangelsdorf DJ, Gautron L, Elmquist JK. PPARg in Vagal Neurons Regulates High-Fat Diet Induced Thermogenesis. Cell Metab. 2014; 19:722–730.

74. Mansuy-Aubert V, Gautron L, Lee S, Bookout AL, Kusminski C, Sun K, Zhang Y, Scherer PE, Mengelsdorf DJ, Elmquist JK. Loss of the liver X receptor LXRα/β in peripheral sensory neurons modifies energy expenditure. Elife. 2015;4.

75. Scott MM, Williams KW, Rossi J, Lee CE, Elmquist JK. Leptin receptor expression in hindbrain Glp-1 neurons regulates food intake and energy balance in mice. J Clin Invest. 2011; 121(6):2413–21.

76. Mortensen K, Christensen LL, Holst JJ, Orskov C. GLP-1 and GIP are colocalized in a subset of endocrine cells in the small intestine. Regul Pept. 2003; 114(2-3):189–96.

77. Theodorakis MJ, Carlson O, Michopoulos S, Doyle ME, Juhaszova M, Petraki K, Egan JM. Human duodenal enteroendocrine cells: source of both incretin peptides, GLP-1 and GIP. Am J Physiol Endocrinol Metab. 2006; 290(3):E550–9.

78. Rouillé Y, Martin S, Steiner DF. Differential processing of proglucagon by the subtilisin-like prohormone convertases PC2 and PC3 to generate either glucagon or glucagon-like peptide. J Biol Chem. 1995; 270(44):26488–96.

79. Orskov C, Wettergren A, Holst JJ. Biological effects and metabolic rates of glucagonlike peptide-1 7-36 amide and glucagonlike peptide-1 7-37 in healthy


53

subjects are indistinguishable. Diabetes. 1993; 42(5):658–61.

80. Ezcurra M, Reimann F, Gribble FM, Emery E. Molecular mechanisms of incretin hormone secretion. Curr Opin Pharmacol. 2013; 13(6):922–7.

81. Chimerel C, Emery E, Summers DK, Keyser U, Gribble FM, Reimann F. Bacterial metabolite indole modulates incretin secretion from intestinal enteroendocrine L cells. Cell Rep. 2014; 9(4):1202–8.

82. Clara R, Langhans W, Mansouri A. Oleic acid stimulates glucagon-like peptide-1 release from enteroendocrine cells by modulating cell respiration and glycolysis. Metab Clin Exp. 2015; 65(3):8–17.

83. Deacon CF, Johnsen AH, Holst JJ. Degradation of glucagon-like peptide-1 by human plasma in vitro yields an N-terminally truncated peptide that is a major endogenous metabolite in vivo. J Clin Endocrinol Metab. 1995; 80(3):952–7.

84. Mentlein R, Gallwitz B, Schmidt WE. Dipeptidyl-peptidase IV hydrolyses gastric inhibitory polypeptide, glucagon-like peptide-1(7-36)amide, peptide histidine methionine and is responsible for their degradation in human serum. Eur J Biochem. 1993; 214(3):829–35.

85. Mayo KE, Miller LJ, Bataille D, Dalle S, Göke B, Thorens B, Drucker DJ International Union of Pharmacology. XXXV. The glucagon receptor family. Pharmacol Rev. 2003; 55(1):167–94.

86. Baggio LL, Drucker DJ. Biology of Incretins: GLP-1 and GIP. Gastroenterology. 2007; 132(6):2131–57.

87. Montrose-Rafizadeh C, Avdonin P, Garant MJ, Rodgers BD, Kole S, Yang H, Levine MA, Schwindinger W, Bernier M. Pancreatic glucagon-like peptide-1 receptor couples to multiple G proteins and activates mitogen-activated protein kinase pathways in Chinese hamster ovary cells. Endocrinology. 1999; 140(3):1132–40.

88. Wheeler MB, Lu M, Dillon JS, Leng XH, Chen C, Boyd AE. Functional expression of the rat glucagon-like peptide-I receptor, evidence for coupling to both adenylyl cyclase and phospholipase-C. Endocrinology. 1993; 133(1):57–62.

89. Hällbrink M, Holmqvist T, Olsson M, Ostenson CG, Efendic S, Langel U. Different domains in the third intracellular loop of the GLP-1 receptor are responsible for Galpha(s) and Galpha(i)/Galpha(o) activation. Biochim Biophys Acta. 2001; 1546(1):79–86.

90. Fehmann HC, Habener JF. Homologous desensitization of the insulinotropic glucagon-like peptide-I (7-37) receptor on insulinoma (HIT-T15) cells. Endocrinology. 1991; 128(6):2880–8.

91. Gromada J, Dissing S, Rorsman P. Desensitization of glucagon-like peptide 1 receptors in insulin-secreting beta TC3 cells: role of PKA-independent mechanisms. Br J Pharmacol. 1996; 118(3):769–75.

92. Baggio LL, Kim J-G, Drucker DJ. Chronic exposure to GLP-1R agonists promotes homologous GLP-1 receptor desensitization in vitro but does not attenuate GLP-1R-dependent glucose homeostasis in vivo. Diabetes. 2004; 53 Suppl 3:S205–14.


54

93. Secher A, Jelsing J, Baquero AF, Hecksher-Sørensen J, Cowley MA, Dalbøge LS, Hansen G, Grove KL, Pyke C, Raun K, Schäffer L, Tangs-Christensen M, Verma S, Witgen BM, Vrang N, Knudsen LB. The arcuate nucleus mediates GLP-1 receptor agonist liraglutide-dependent weight loss. J Clin Invest. 2014; 124:4473-4488.

94. Ronveaux CC, de Lartigue G, Raybould HE. Ability of GLP-1 to decrease food intake is dependent on nutritional status. Physiol Behavi. 2014; 135:222–9.

95. Mojsov S, Weir GC, Habener JF. Insulinotropin: glucagon-like peptide I (7-37) co-encoded in the glucagon gene is a potent stimulator of insulin release in the perfused rat pancreas. J Clin Invest. 1987; 79(2):616–9.

96. Kreymann B, Ghatei MA, Willams G, Bloom SR. Glucagon-like peptide-1 7-36 : a physiological incretin in man. Lancet. 1987; 2(8571):1300-4.

97. Lamont BJ, Li Y, Kwan E, Brown TJ, Gaisano H, Drucker DJ. Pancreatic GLP-1 receptor activation is sufficient for incretin control of glucose metabolism in mice. J Clin Invest. 2012; 122(1):388–402.

98. Holst JJ. The Physiology of Glucagon-like Peptide 1. Physiol Rev. 2007; 87(4):1409–39.

99. de Heer J, Holst JJ. Sulfonylurea compounds uncouple the glucose dependence of the insulinotropic effect of glucagon-like peptide 1. Diabetes. 2007; 56(2):438–43.

100. Drucker DJ, Philippe J, Mojsov S, Chick WL, Habener JF. Glucagon-like peptide I stimulates insulin gene expression and increases cyclic AMP levels in a rat islet cell line. Proc Natl Acad Sci USA. 1987; 84(10):3434–8.

101. Li Y, Hansotia T, Yusta B, Ris F, Halban PA, Drucker DJ. Glucagon-like peptide-1 receptor signaling modulates beta cell apoptosis. J Biol Chem. 2003; 278(1):471–8.

102. Yusta B, Baggio LL, Estall JL, Koehler JA, Holland DP, Li H, Pipeleers D, Ling Z, Drucker DJ. GLP-1 receptor activation improves beta cell function and survival following induction of endoplasmic reticulum stress. Cell Metab. 2006; 4(5):391–406.

103. Fehmann HC, Habener JF. Functional receptors for the insulinotropic hormone glucagon-like peptide-I(7-37) on a somatostatin secreting cell line. FEBS Lett. 1991; 279(2):335–40.

104. Orskov L, Holst JJ, Møller J, Orskov C, Møller N, Alberti KG, Schmitz O. GLP-1 does not not acutely affect insulin sensitivity in healthy man. Diabetologia. 1996; 39(10):1227–32.

105. Vella A, Shah P, Basu R, Basu A, Holst JJ, Rizza RA. Effect of glucagon-like peptide 1(7-36) amide on glucose effectiveness and insulin action in people with type 2 diabetes. Diabetes. 2000; 49(4):611–7.

106. Larsen PJ, Fledelius C, Knudsen LB, Tang-Christensen M. Systemic administration of the long-acting GLP-1 derivative NN2211 induces lasting and reversible weight loss in both normal and obese rats. Diabetes. 2001; 50(11):2530–9.

107. Scott KA, Moran TH. The GLP-1 agonist exendin-4 reduces food intake in nonhuman primates through changes in meal size. Am J Physiol Regul Integr Comp Physiol 2007; 293(3):R983–7.


55

108. Gutzwiller J-P, Göke B, Drewe J, Hildebrand P, Ketterer S, Handschin D, Winterhalder R, Conen D, Beglinger C.. Glucagon-like peptide-1: a potent regulator of food intake in humans. Gut. 1999; 44(1):81–6.

109. Williams DL, Baskin DG, Schwartz MW. Evidence that Intestinal Glucagon-Like Peptide-1 Plays a Physiological Role in Satiety. Endocrinology. 2009; 150(4):1680–7.

110. Ruttimann EB, Arnold M, Hillebrand JJ, Geary N, Langhans W. Intrameal Hepatic Portal and Intraperitoneal Infusions of Glucagon-Like Peptide-1 Reduce Spontaneous Meal Size in the Rat via Different Mechanisms. Endocrinology. 2008; 150(3):1174–81.

111. Punjabi M, Arnold M, Geary N, Langhans W, Pacheco-López G. Peripheral Glucagon-like Peptide-1 (GLP-1) and Satiation. Physiol Behav. 2011;105(1):71–6.

112. Punjabi M, Arnold M, Rüttimann E, Graber M, Geary N, Pacheco-López G, Langhans W. Circulating glucagon-like peptide-1 (GLP-1) inhibits eating in male rats by acting in the hindbrain and without inducing avoidance. Endocrinology. 2014; 155(5):1690–9.

113. Imeryüz NI, Yegen BC, Bozkhurt A, Coskun T, Villanueva-Penacarrillo ML, Ulusoy NB. Glucagon-like peptide-1 inhibits gastric emptying via vagalafferent-mediated central mechanisms. Am J Physiol Gastrointest Liver Physiol. 1997; 273: G920–7.

114. Ravassa S, Zudaire A, Diez J. GLP-1 and cardioprotection: from bench to bedside. Cardiovasc Res. 2012; 94(2):316–23.

115. Hölscher C. Potential role of glucagon-like peptide-1 (GLP-1) in neuroprotection. CNS Drugs. 2012; 26(10):871–82.

116. Beiroa D, Imbernon M, Gallego R, Senra A, Herranz D, Villaroya F, Serrano M, Ferno J, Salvador J, Escalada J, Dieguez C, Lopez M, Frühbeck G, Nogueiras R. GLP-1 Agonism Stimulates Brown Adipose Tissue Thermogenesis and Browning Through Hypothalamic AMPK. Diabetes. 2014; 63(30):3346-58.

117. Conarello SL, Li Z, Ronan J, Roy RS, Zhu L, Jiang G, Liu F, Woods J, Zycband, Moller D, Thornberry, Zhang BB. Mice lacking dipeptidyl peptidase IV are protected against obesity and insulin resistance. Proc Natl Acad Sci USA. 2003; 100(11):6825–30.

118. Pannacciulli N, Bunt JC, Koska J, Bogardus C, Krakoff J. Higher fasting plasma concentrations of glucagon-like peptide 1 are associated with higher resting energy expenditure and fat oxidation rates in humans. Am J Clin Nutr. 2006; 84(3):556–60.

119. Hwa JJ, Ghibaudi L, Williams P, Witten MB, Tedesco R, Strader CD. Differential effects of intracerebroventricular glucagon-like peptide-1 on feeding and energy expenditure regulation. Peptides. 1998; 19(5):869–75.

120. Osaka T, Endo M, Yamakawa M, Inoue S. Energy expenditure by intravenous administration of glucagon-like peptide-1 mediated by the lower brainstem and sympathoadrenal system. Peptides. 2005; 26(9):1623–31.

121. Shalev A, Holst JJ, Keller U. Effects of glucagon-like peptide 1 (7-36 amide) on whole-body protein metabolism in healthy man. Eur J Clin Invest. 1997; 27(1):10–6.


56

122. Abegg K, Schiesser M, Lutz TA, Bueter M. Acute peripheral GLP-1 receptor agonism or antagonism does not alter energy expenditure in rats after Roux-en-Y gastric bypass. Physiol Behav. 2013; 121:70–8.

123. Baggio LL, Huang Q, Brown TJ, Drucker DJ. Oxyntomodulin and glucagon-like peptide-1 differentially regulate murine food intake and energy expenditure. Gastroenterology. 2004; 127(2):546–58.

124. Taher J, Baker CL, Cuizon C, Masoudpour H, Zhang R, Farr S, Naples M, Bourdon C, Pausova Z, Adeli K. GLP-1 receptor agonism ameliorates hepatic VLDL overproduction and de novo lipogenesis in insulin resistance. Mol Metab. 2014; 3(9):823–33.

125. Flint A, Raben A, Rehfeld JF, Holst JJ, Astrup A. The effect of glucagon-like peptide-1 on energy expenditure and substrate metabolism in humans. Int J Obes. 2000; 24(3):288–98.

126. Flint A, Raben A, Ersbøll AK, Holst JJ, Astrup A. The effect of physiological levels of glucagon-like peptide-1 on appetite, gastric emptying, energy and substrate metabolism in obesity. Int J Obes. 2001; 25(6):781–92.

127. Jin SL, Han VK, Simmons JG, Towle AC, Lauder JM, Lund PK. Distribution of glucagonlike peptide I (GLP-I), glucagon, and glicentin in the rat brain: an immunocytochemical study. J Comp Neurol. 1988; 271(4):519–32.

128. Conlon JM, Samson WK, Dobbs RE, Orci L, Unger RH. Glucagon-like polypeptides in canine brain. Diabetes. 1979; 28(7):700–2.

129. Vrang N, Larsen PJ. Preproglucagon derived peptides GLP-1, GLP-2 and oxyntomodulin in the CNS: Role of peripherally secreted and centrally produced peptides. Prog Neurobiol. 2010; 92(3):442–62.

130. Vrang N, Grove K. The brainstem preproglucagon system in a non-human primate (Macaca mulatta). Brain Res. 2011; 1397:28–37.

131. Merchenthaler I, Lane M, Shughrue P. Distribution of pre-pro-glucagon and glucagon-like peptide-1 receptor messenger RNAs in the rat central nervous system. J Comp Neurol. 1999; 403(2):261–80.

132. Llewellyn-Smith IJ, Reimann F, Gribble FM, Trapp S. Preproglucagon neurons project widely to autonomic control areas in the mouse brain. Neuroscience. 2011; 180(C):111–21.

133. Larsen PJ, Tang-Christensen M, Holst JJ, Orskov C. Distribution of glucagon-like peptide-1 and other preproglucagon-derived peptides in the rat hypothalamus and brainstem. Neuroscience. 1997; 77(1):257–70.

134. Llewellyn-Smith IJ, Gnanamanickam GJE, Reimann F, Gribble FM, Trapp S. Preproglucagon (PPG) neurons innervate neurochemicallyidentified autonomic neurons in the mouse brainstem. Neuroscience. 2013; 229(C):130–43.

135. Trapp S, Richards JE. ScienceDirectThe gut hormone glucagon-like peptide-1 produced in brain: is this physiologically relevant? Curr Opinion Pharmacol. 2013; 13(6):964–9.


57

136. Barrera JG, Jones KR, Herman JP, D'Alessio DA, Woods SC, Seeley RJ. Hyperphagia and Increased Fat Accumulation in Two Models of Chronic CNS Glucagon-Like Peptide-1 Loss of Function. J Neurosci. 2011; 31(10):3904–13.

137. Wang X-F, Liu J-J, Xia J, Liu J, Mirabella V, Pang ZP. Endogenous Glucagon-like Peptide-1 Suppresses High-Fat Food Intake by Reducing Synaptic Drive onto Mesolimbic Dopamine Neurons. Cell Rep. 2015; 12(5):726–33.

138. Trapp S, Hisadome K. Glucagon-like peptide 1 and the brain: Central actions–central sources? Auton Neurosci - Basic. 2011; 161(1-2):14–9.

139. Hisadome K, Reimann F, Gribble FM, Trapp S. Leptin Directly Depolarizes Preproglucagon Neurons in the Nucleus Tractus Solitarius. Diabetes. 2010; 59(8):1890-8.

140. Hisadome K, Reimann F, Gribble FM, Trapp S. CCK stimulation of GLP-1 neurons involves alpha1-adrenoceptor-mediated increase in glutamatergic synaptic inputs. Diabetes. 2011; 60(11):2701–9.

141. Vrang N, Phifer CB, Corkern MM, Berthoud H-R. Gastric distension induces c-Fos in medullary GLP-1/2-containing neurons. Am J Physiol Regul Integr Comp Physiol. 2003; 285:R470-8.

142. Trapp S, Cork SC. PPG neurons of the lower brainstem and their role in brain GLP-1 receptor activation. Am J Physiol Regul Integr Comp Physiol. 2015; 309:R795-804

143. Meier JJ. GLP‐1 receptor agonists for individualized treatment of type 2 diabetes mellitus. Nat Rev Endocrinol. 2012; 8(12):728–42.

144. Holz GG, Kühtreiber WM, Habener JF. Pancreatic beta-cells are rendered glucose-competent by the insulinotropic hormone glucagon-like peptide-1(7-37). Nature. 1993; 361(6410):362–5.

145. FDA US. Press Announcements - FDA approves weight-management drug Saxenda [Internet]. 2014 [cited 2015 Jun 3]. Available from: http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm427913.htm

Chapter 2: Vagal mediation of GLP-1’s effects on food intake and glycemia

59

CHAPTER 2:

VAGAL MEDIATION OF GLP-1’S EFFECTS ON FOOD INTAKE A ND

GLYCEMIA

AUTHORS

Krieger JP, Langhans W, Lee SJ

Physiology and Behavior Laboratory, ETH Zürich, Switzerland

STATUS

Published in Physiology and Behavior (2015; 152(Pt B):372-80)


60

ABSTRACT

Nutrient stimulation of the enteroendocrine L-cells induces the release of the glucagon-like

peptide-1 (GLP-1), an incretin and satiating peptide. Due to its short half-life, meal-induced

GLP-1’s effects on food intake and glycemia are likely to be mediated in part by a paracrine

signaling mechanism near the site of release. Early and recent findings from vagus nerve

(VN) lesion studies scrutinized in this review strongly support an important role of the VN in

mediating GLP-1’s effects. Peripheral GLP-1 or GLP-1R agonist treatment failed to elicit the

full satiating effects and maintain glucose homeostasis in various lesion models. The

potential mechanisms underlying the vagal GLP-1R mediated satiation and glycemic control

presumably involve the activation of caudal brainstem neurons via glutamatergic signaling,

which activate a vagal reflex loop or/and relay the information to higher brain centers. Recent

studies also presented here, however, diminish the relevance of the VN for the

pharmacological intervention of obesity and diabetes with chronic GLP-1R agonist

treatments, suggesting that endogenous intestinal GLP-1 and GLP-1R agonists may activate

different GLP-1R populations. Finally, lesion-based approaches are limited and new

technical approaches are discussed to improve the understanding of vagal GLP-1R functions

in maintaining normal energy balance and its relevance in pharmacological interventions.


61

1 Introduction

Eating stimulates the release of glucagon-like peptide-1 (GLP-1) from intestinal L-cells (see

(1-4) for review). GLP-1 is mainly known for its incretin effect, i.e., for its ability to enhance

glucose-induced insulin secretion (5-7), and for its satiating effect, which is predominantly

expressed as a reduction in meal size (see (8) for review). Once in the blood stream, GLP-1

is subject to rapid degradation by the enzyme dipeptidyl-peptidase-4 (DPP-IV) (9,10), which

is also present in the liver (11). This limits the amount of active GLP-1 reaching the systemic

circulation and its potential access to the brain and to remote areas in the periphery. It

therefore stands to reason that a paracrine mechanism, i.e., the activation of GLP-1

receptors (GLP-1R) located in the vicinity of L-cells, can mediate some of GLP-1’s effects on

eating and glycemia. The vagus nerve has initially been proposed to relay the GLP-1 initiated

signals to the brain: vagal afferent neurons (VAN) terminate in the lamina propria of the

intestinal mucosa (12), express GLP-1R (2,13) and are activated in response to GLP-1

(2,14,15), giving them both an ideal position and the cellular machinery to sense L-cell-

released GLP-1, and to relay the resulting signals to the brain. Moreover, central terminals of

VAN are mainly found in the nucleus tractus solitarii (NTS), an area that is known to integrate

many energy-related signals from the periphery and descending neural signals from the

forebrain (see (8,16) for review). Numerous results reviewed below further implicate the VN

in the mediation of endogenous GLP-1’s effects on eating and glycemia. Paradoxically,

recent data investigating the mode of actions of long-lasting GLP-1R agonists such as

liraglutide in appetite reduction and glycemic control question the role of VAN in mediating

endogenous GLP-1 effects on the maintenance of energy balance. This review aims at

dissecting these findings in order to distinguish between the physiological and

pharmacological relevance of vagal GLP-1R activation. Moreover, the mechanisms by which

vagal GLP-1R activation decreases food intake and controls glycemia, including VAN GLP-

1R signaling and central relays, are unclear. We will discuss recent findings and present a


62

brief overview of potential pathways involved in vagal GLP-1R effects on food intake and

glycemia. Finally, we emphasize that previous studies have often used unspecific lesion

techniques to demonstrate the necessity of the VN in mediating GLP-1’s effects on food

intake and glycemia, and we therefore advocate for further careful investigation of the

relationship between GLP-1 and the VN.

2 Vagal GLP-1R activation in the control of food in take

Necessity of vagal afferents 2.1

Early studies reported that peripheral administration of GLP-1 promoted satiation and

reduced energy intake in humans and rats (17-20), raising the possibility that GLP-1 therapy

could be effective in treating obesity. Therefore, many laboratories attempted to unravel the

mechanisms of peripheral GLP-1 mediated satiation (21-23). The eating-inhibitory effect of

intraperitoneally (IP) injected GLP-1 was blocked by IP, but not by intracerebroventricular

(ICV) administration of the GLP-1R antagonist, exendin-(9-39) (Ex-9) (24), suggesting that

the effect of peripheral GLP-1 on food intake is mediated by GLP-1R in the periphery. Given

the short half-life and the generally low concentration of GLP-1 in the systemic circulation,

the hypothesis that a neural pathway mediates peripheral GLP-1’s effects on food intake has

been explored. A number of studies (summarized in Table 1) used a combination of vagal

manipulation with activation or blockade of peripheral GLP-1R. The inhibitory effect on food

intake of low doses of IP exendin-4 (Ex-4, a GLP-1R agonist) was lost in mice pretreated

with capsaïcin (25), suggesting the involvement of sensory fibers. Moreover, bilateral total

subdiaphragmatic vagotomy (VGX) attenuated the reduction of food intake induced by IP

injection of GLP-1 in rats (26), suggesting that the eating-inhibitory effect of peripheral GLP-1

requires an intact VN. More recent data in patients with VGX and pyloroplasty extended this

conclusion to humans, demonstrating that an intact VN is needed for the inhibition of food

intake by intravenous (IV) GLP-1 (27). The VGX procedure, however, lesions VAN as well as

vagal efferent neurons (VEN), and the absence of efferent projections from the brainstem


63

can significantly alter gut motility, exocrine and endocrine secretion and, subsequently,

eating behavior (28). Subdiaphragmatic vagal deafferentation (SDA), the most complete and

selective vagal deafferentation method available to date (29-31), leaves about 50% of the

VEN intact and, hence, largely avoids the strong side effects of VGX. Studies using SDA

allow for a markedly improved interpretation of the lesion-induced attenuation of GLP-1’s

effects on food intake (32-35). Altogether, they indicate the necessity of VAN in mediating

peripheral GLP-1 satiation, consistent with the idea that intestinal GLP-1 can induce meal

termination via activation of vagal GLP-1R.

Roles of vagal subbranches 2.2

GLP-1R are expressed on vagal afferent fibers innervating the hepatoportal bed (i.e., on the

common hepatic branch - CHB) as well as on fibers innervating the gastrointestinal (GI) tract

(i.e., the CHB as well as the gastric and celiac branches) (12). Therefore, the eating-

inhibitory effects of intestinal GLP-1 could either be mediated by GLP-1R signaling through

the CHB of the VN (endocrine action) or by the vagal afferent branches innervating the

intestine (paracrine action). One way to address this question is to administer GLP-1 via

different routes. Indeed, IP injection of GLP-1 appears to provide an easy access to the

lamina propria of the intestinal mucosa, where VAN innervating the GI tract terminate. An

infusion of GLP-1 into the hepatic portal vein (HPV), however, should mainly activate GLP-

1R present in the hepatoportal bed, i.e. on the CHB. To differentiate these possible

mechanisms of eating inhibition by peripheral GLP-1 infusions, a previous study from our

laboratory used a combination of remotely controlled intrameal GLP-1 infusions (IP or HPV)

and SDA, and showed that intact vagal afferents are required for the inhibition of eating after

IP, but not after HPV, GLP-1 infusions (35), as previously suggested by other studies

(14,36). Whether the CHB of the VN is required for the reduction of food intake after IP GLP-

1 was directly addressed by Hayes and colleagues. In this study, rats showed a blunted

eating-inhibitory response to IP GLP-1 after SDA, but not after a selective ablation of the

CHB (CHBX), indicating that IP injected GLP-1 does not control food intake


64

Table 1: Summary of animal studies investigating th e role of the vagus nerve in mediating GLP-1's effe cts on food intake by combining vagal manipulation with peripheral GLP-1R activation or b lockade. Abbreviations: ICV: intracerebroventricular, IP: intraperitoneum, JV: jugular vein, HPV: hepatic portal vein, SC: subcutaneous, VC: vena cava; CHBX: common hepatic branch vagotomy, SDA: subdiaphragmatic vagal deafferentation, VGX: vagotomy

GLP-1R activation or blockade Vagal manipulation Species Finding Reference

GLP-1 SC (6 pmol/kg) – ICV (75 fmol/rat)

Capsaïcin application to the vagus nerve

Rat Vagal afferents are necessary for the inhibition of gastric emptying induced by SC GLP-1

Imeryüz et al., 1997 (21)

Ex-4 IP (0.06 µg/mouse)

Capsaïcin treatment by IP injections

Mouse The full effect of IP Ex-4 on food intake requires intact vagal afferents

Talsania et al., 2005 (25)

GLP-1 IP (100 nmol/kg)

VGX Rat The vagus nerve is required for the reduction of food intake induced by IP GLP-1

Abbott et al., 2005 (26)

GLP-1 (intrameal) IP (10 nmol/kg) - HPV (0.25 to 10 nmol/kg) - VC (1 nmol/kg)

SDA Rat Circulating GLP-1 reduces meal size by nonvagal mechanisms but IP GLP-1 requires intact vagal afferents

Rüttimann et al., 2009 (35)

GLP-1 JV (1.25 to 10 µg/rat)

VGX Capsaïcin treatment by IP injections

Rat Circulating GLP-1 reduces food intake by nonvagal mechanisms

Zhang and Ritter, 2011 (36)

DPP-IV inhibitor diprotinin A HPV (4 µmol/kg)

CHBX Rat The inhibition of 1h and 12h food intake by HPV administration of DPP-IV inhibitor is abolished in CHBX rats

Fujiwara et al., 2011 (37)

GLP-1 IP (0.5 mg/kg)

CHBX vs SDA Rat Vagal afferents from gastrointestinal origins, but not the CHB, are necessary for the eating inhibition induced by IP GLP-1

Hayes et al., 2011 (33)

Liraglutide / IP (10 to 50 µg/kg) Ex-4 / IP (1 and 3 µg/kg)

SDA Rat Ex-4 and liraglutide acutely suppress food intake by activation of both CNS and vagal GLP-1R

Kanoski et al., 2011 (34)

Ex-4 IP (0.1 and 1.0 µg/kg)

SDA Rat Vagal afferents mediate the early satiation induced by peripheral administration of low dose Ex-4 but not later eating-inhibitory actions

Labouesse et al., 2012 (32)

Liraglutide / IP (400 µg/kg) for acute studies and SC (300 to 1000 µg/kg) for chronic treatment

Visceral nerve specific genetic deletion of Glp1r

Mouse Visceral nerve GLP-1R do not mediate the reduction in food intake induced by liraglutide

Sisley et al., 2014 (39)


65

via the activation of vagal GLP-1R in the hepatoportal bed (33). Consistent with this

interpretation, HPV and vena cava infusions of GLP-1 reduced meal size similarly (34).

These findings collectively suggest that an acute IP injection of GLP-1 reduces food intake

via an activation of GLP-1R located on VAN of gastrointestinal but not hepatic origin. As

GLP-1 released during meals is likely to activate GLP-1R present on VAN terminals in the

lamina propria, it stands to reason that GLP-1 released during meals reduces food intake via

a similar mechanism. These findings also suggest, however, that high systemic blood levels

of GLP-1 or of its DPP-IV-resistant agonists reduce food intake via a different mechanism

that may involve central GLP-1R activation.

Recent studies challenged the prevailing view that vagal afferents innervating the GI tract are

the main neural pathway mediating peripheral GLP-1 effects on food intake. Injection of a

DPP-IV inhibitor in the HPV that raises HPV (but not systemic) GLP-1 levels within the

physiological range have been shown to reduce 1 h and 12 h food intake (37). Moreover, this

effect required an intact CHB. These findings can only be reconciled by the idea that

pharmacological inhibition of DPP-IV might raise HPV GLP-1 in a more pronounced and

sustained way in contrast to IP or HPV injection of GLP-1; the duration of GLP-1 elevation in

the HPV may be a critical factor for the activation of the CHB.

Mediation of acute vs chronic effects of GLP-1R ag onists by the vagus 2.3

nerve

The use of GLP-1R agonists to treat obesity is looming above the horizon. It is therefore

important to ask whether vagal afferents are also required for the eating-inhibitory and anti-

obesity effects of long-lasting GLP-1R agonists. Two studies examined the acute effects of

IP infusions of variable doses of Ex-4 or liraglutide on food intake (1 to 24 h) in SDA rats

(32,34). Together, these studies demonstrated that the lesion of vagal afferents either

decreased the sensitivity (i.e. the threshold dose needed to decrease food intake) or blunted

the short-term effect of GLP-1R agonists on food intake. This suggests that vagal afferents

are needed for the full expression of the acute effects of GLP-1R agonists on food intake.


66

GLP-1R agonists are, however, administered chronically at pharmacological doses and the

same conclusion may not hold for the chronic effects of GLP-1R agonist on food intake and

body weight. In fact, food intake and body weight were decreased to the same extent in

Sham and SDA rats after a 14-day treatment with subcutaneous (SC) liraglutide (38).

Moreover, a genetic deletion of the glp-1r gene in paired-like homeobox 2b (Phox2b)-

expressing visceral nerves did not affect the effects of liraglutide on short-term (4 h and 24 h)

chow intake or 14-day cumulative food intake and body weight change (39). These data

suggest that vagal afferents do not play a major role in mediating the chronic changes in food

intake and body weight after pharmacological administration of GLP-1R agonists.

Summary 2.4

Together, these results advocate for prudence when it comes to make statements about the

role of vagal GLP-1R activation in controlling food intake.

a) There is strong evidence indicating that endogenous GLP-1 released during meals

reduces meal size by activating GLP-1R located on intestinal vagal afferents.

b) This mechanism may be mimicked by IP injected GLP-1 and also be the basis for the

acute effects of GLP-1R agonists.

c) High levels of circulating GLP-1, which can be achieved by intravenous infusions of GLP-1

or pharmacological inhibition of DPP-IV activity, reduce food intake by recruiting other

mechanisms distinct from vagal GLP-1R activation.

c) Chronic pharmacological administration of liraglutide may exert its eating-inhibitory and

body weight lowering effects via non-vagal mediated mechanisms. It is possible to speculate

from the available data that the eating-inhibitory effects of central GLP-1R activation

dominate that of the vagal activation in the context of pharmacological intervention. Indeed,

central GLP-1R activation elicits multiple behavioral effects beyond the homeostatic control

of food intake, including reward suppression and activation of stress responses (40-45).


67

Further research is needed to understand which central GLP-1R population mediates the

effects of GLP-1R agonists on food intake.

3 Vagal GLP-1R activation and the neuroincretin eff ect

The vagus nerve can mediate GLP-1 effects on postprandial glycemia 3.1

The interaction between autonomic nerves and incretin hormones to control islet function has

been conceptualized as the “neuroincretin effect” (14,46,47). Although the neuroincretin

effect of GLP-1 has been reviewed elsewhere (47-49), we aimed at reviewing both early and

recently published data involving a manipulation of the VN, together with GLP-1R activation

or blockade (Table 2), through the prism of the physiology/pharmacology dichotomy.

Early findings showing an increase in the discharge rate of hepatic VAN and pancreatic VEN

upon GLP-1 stimulation provided evidence for a vagal mediation of the neuroincretin effect of

GLP-1 (14). In a study published by Imeryüz et al., a SC injection of GLP-1 did not change

blood glucose levels in rats treated with perivagal application of capsaïcin or Sham rats (21).

In the same study, capsaïcin-treated rats, however, showed an early hyperglycemia after a

glucose test meal compared to Sham animals, consistent with the idea that activation of VAN

by prandially released GLP-1 could play a role in the control of post-meal blood glucose.

Human data showed that post-meal insulin secretion was unaffected by IV GLP-1 in VGX

patients with pyloroplasty, but was suppressed in healthy controls (27). In addition, the same

group demonstrated that gastrointestinal-mediated glucose disposal was reduced in VGX

patients with pyloroplasty (50), suggesting that a vagal mediation of GLP-1’s effects on

glycemia is potentially relevant to control post-meal blood glucose in humans.

Roles of vagal subbranches 3.2

More specifically, because Nakabayashi and Nishizawa (14) addressed the hepatic vagal

afferents as a potential route for the mediation of GLP-1’s neuroincretin effect, later reports

focused on whether this «portal sensor» is indeed involved. A series of studies tested the


68

role of GLP-1R localized on VAN terminals of the CHB by performing HPV injections of GLP-

1 (51,52), DPP-IV inhibitor (37) or a GLP-1R antagonist (52,53). Together, these studies

demonstrated that a local elevation of HPV GLP-1 could reduce blood glucose after an

OGTT or after IV glucose infusion and that this glucose lowering effect was lost after lesion

of the CHB. These studies assumed that GLP-1 concentrations in the HPV after a meal

activate GLP-1R on the sensory terminals in the hepatoportal bed (the CHB) and generate

neural signals to the dorsal vagal complex to regulate blood glucose levels. Prandially

released GLP-1, however, could also activate GLP-1R located on intestinal VAN terminals.

To compare the role of paracrine (intestinal) vs endocrine (hepatic) GLP-1R signaling in

glucose regulation, Hayes and colleagues compared the glucose responses to an OGTT

after IP Ex-9 pretreatment in Sham, CHBX or SDA rats (33). While IP Ex-9 treatment

increased blood glucose levels similarly in CHBX and Sham rats, it failed to elicit a further

increase in glucose response in SDA rats. This suggests that the CHB is not required to

mediate the glucose lowering effects of GLP-1.

It may be possible to reconcile these apparently inconsistent findings by conceptualizing the

GLP-1 sensing of intestinal and hepatic afferents as a collaborative integrated system. After

a meal, high local concentrations of GLP-1 in the vicinity of intestinal vagal afferent terminals

initiate the neuroincretin effect via a vagal pancreatic reflex pathway. Then, if the meal leads

to high portal GLP-1 concentrations, hepatic afferents may play an additive role to further

maintain the glucose-lowering effect of vagal GLP-1R activation. As we speculated above for

the control of food intake, it is plausible that hepatic afferents require a sustained activation

to contribute to the regulation of prandial glucose. In summary, whereas the exact role of

different vagal branches is still unclear, it is reasonable to assume that the VN contributes to

GLP-1’s neuroincretin effect.


69

GLP-1R activation or blockade Vagal manipulation Species Finding Reference Meal-induced GLP-1 Capsaïcin

application to the vagus nerve

Rat Vagal afferent denervation by capsaïcin causes early hyperglycemia after a test meal

Imeryüz et al., 1997 (21)

GLP-1 HPV or JV (10 pmol/kg)

Ganglion blocker chlorisondamine

Rat Insulin response to a glucose and GLP-1 HPV infusion is mediated by a neural reflex

Balkan et al., 2000 (54)

GLP-1 Tail vein (0.1 or 10 pmol/kg)

Capsaïcin treatment by IP injections

Mouse GLP-1-induced insulin secretion requires intact sensory nerves in mice when GLP-1 is administered IV at low doses

Ahren, 2004 (55)

DPP-IV inhibitor diprotinin A HPV (4 µmol/kg)

VGX Rat HPV infusion of a DPP-IV inhibitor decreases blood glucose and increases plasma insulin : no effects in VGX rats

Fujiwara et al., 2011 (37)

Ex-9 IP (0.5 mg/kg)

CHBX vs SDA Rat Vagal afferents from gastrointestinal origings, but not the CHB, are necessary for the control of glycemia by vagal GLP-1R

Hayes et al., 2011 (33)

GLP-1 HPV (1 or 3 pmol/kg/min)

CHBX vs SDA Rat HPV infusion of a low dose of GLP-1 stimulates insulin secretion through the CHB

Nishizawa et al., 2013 (51)

Liraglutide / IP (400 µg/kg) for acute studies and SC (300 to 1000 µg/kg) for chronic treatment

Visceral nerve specific genetic deletion of Glp1r

Mouse Visceral nerve GLP-1R do not mediate the glucose-lowering effects of liraglutide

Sisley et al., 2014 (39)

Table 2: Summary of animal studies investigating th e role of the vagus nerve in mediating GLP-1's effe cts on glycemia by combining vagal manipulation with peripheral GLP-1R activation or b lockade. Abbreviations: IP: intraperitoneal, IV: intravenous, JV: jugular vein, HPV: hepatic portal vein, SC: subcutaneous; CHBX: common hepatic branch vagotomy, SDA: subdiaphragmatic vagal deafferentation, VGX: vagotomy


70

Relevance of the vagal neuroincretin effect of GLP -1 3.3

The incretin effect of circulating GLP-1 is mediated in large part by the direct activation of

GLP-1R on pancreatic β-cells (5-7). An additional vagal component of GLP-1’s effects on

glycemia is generally conceptualized as a “gut-to-brain-to-pancreas” pathway that results in

an increase in insulin release in addition to the conventional “incretin” effects (47,48). A few

studies addressed this topic by measuring insulin after a glucose challenge in the context of

vagal manipulation. The glucose-stimulated insulin secretion induced by intravenous infusion

of GLP-1 was attenuated by the ganglionic blocker chlorisondamine (54), by capsaïcin

pretreatment (55) or by CHBX (51). Similarly, the effect of an HPV administration of a DPP-

IV inhibitor on glucose-stimulated insulin secretion was attenuated by VGX (37). Moreover,

IP infusion of atropin, a blocker of the muscarinic receptors, reduced the insulin response

after intravenous co-infusion of glucose and GLP-1 (56). This further suggests that the

activation of vagal afferent GLP-1R signaling increases insulin secretion through a

descending signal mediated by pancreatic vagal efferents.

Interestingly, quantifications of the neuroincretin effect, defined as the supplementary insulin

secretion induced by the activation of vagal GLP-1R, ranged from 0 to 60% (51,55). It seems

to be clear from these studies that the neuroincretin effect is relevant only when small

(physiological) doses of GLP-1 are used and that insulin secretion does not increase in a

dose dependent manner. Mice with a genetic deletion of the glp1r in visceral nerves,

however, did not show any significant impairment in glucose tolerance or insulin secretion

(39). Moreover, pancreatic re-expression of the human GLP-1R in glp-1r -/- mice (Pdx1-

hGLP1R:glp1r-/-) was sufficient to normalize glucose tolerance after an OGTT or an IPGTT

(57), suggesting that VAN GLP-1R are not crucial for glycemic control. It is, however,

possible that the relatively large bolus of glucose used in these studies raised circulating

GLP-1 values much higher than the post-meal circulating GLP-1 levels, consequently

masking the vagally-mediated component of the incretin effect.


71

Moreover, some studies have reported changes in blood glucose after activation or blockade

of portal GLP-1R without significant changes in insulin levels compared to vehicle injections

(52,53,58). These findings are consistent with the idea that GLP-1 can affect glycemia in an

insulin-independent manner (59-61), and fit studies showing that GLP-1R activation can

produce a beneficial effect on glucose regulation in Type 1 diabetes patients, despite their

inherent lack of insulin production (62-67). More recent studies have provided experimental

support for this idea: GLP-1R blockade by Ex-9 or loss of GLP-1 action in glp-1r KO mice

increased glucose levels in streptozotocin animals with no glucagon signaling (68,69).

Evidence presented above and other findings also support the idea that GLP-1 in the HPV

stimulates VAN there and enhances glucose disposal (14,70). Nevertheless, whether GLP-1

acts directly on hepatocytes to regulate hepatic glucose production or whether its effect is

executed through autonomic nerves is currently unknown. The idea that a “gut-brain-liver”

axis controls glucose homeostasis by both insulin-dependent and independent mechanisms

is getting increasing experimental support, but an unequivocal demonstration of this pathway

requires further research (see (71-73) for review).

Finally, sustained improvements in glucose tolerance were seen in the mouse model of glp1r

KO in visceral nerves after chronic liraglutide administration, indicating no requirements of

vagal GLP-1R in blood glucose regulation by GLP-1R agonists (39). This finding further

suggests that the mechanism by which GLP-1R agonists exert their chronic effects on

glucose tolerance does not require intact GLP-1R signaling in the visceral nerves.

Summary 3.4

Altogether, the available evidence suggests that:

a) peripheral GLP-1 can activate a reflex loop that involves subdiaphragmatic VAN and

pancreatic VEN to further enhance the direct effect of GLP-1 on pancreatic beta-cells and to

increase glucose-stimulated insulin secretion. The exact roles of vagal subbranches, as well

as their physiological relevance, remains to be further investigated.


72

b) more specific approaches manipulating GLP-1R populations in a tissue specific manner

indicate no involvement of vagal receptors in the control of blood glucose after glucose

challenges (OGTT or IPGTT). These studies, however, did not investigate these parameters

in the context of a physiological mixed-nutrient meal.

c) recent data raise the possibility that GLP-1 can decrease hepatic glucose production

through insulin-dependent mechanisms that may involve VAN.

d) chronic liraglutide treatment improves glycemia in the absence of vagal GLP-1R signaling,

but further research is needed to confirm that the VN is not involved in the beneficial effects

of chronic GLP-1R agonist treatment.

4 Putative mechanisms of vagal GLP-1R effects on fo od intake

and glycemia

Whereas the studies reviewed above provide strong evidence that GLP-1 can reduce food

intake and glycemia through VAN GLP-1R activation, the mechanisms by which it does so

remain poorly understood. Indeed, there are two important points to be considered: 1) the

brain circuits that process the satiation signals generated by vagal GLP-1R activation remain

unclear; 2) the neurotransmitters/neuropeptides involved in mediating the effects of VAN

GLP-1R activation are still unknown. In this section, we will review the current knowledge

related to this topic and propose putative mechanisms that could link VAN GLP-1R activation

with the behavioral outputs governed by the central processing of peripheral signals.

Central pathways activated by vagal GLP-1R activat ion 4.1

VAN terminals project to the caudal brainstem where they make synaptic connections to

second order neurons in the NTS (12). Glutamate is considered to be the major

neurotransmitter of VAN, and numerous findings have implicated the release of glutamate in

the NTS in the control of eating (see (74) for review). Therefore, it is reasonable to speculate

that VAN GLP-1R activation results in the activation of NTS neurons via glutamatergic


73

synaptic transmission, initiating the processing of the eating-inhibitory signal in various brain

areas including hypothalamic and forebrain nuclei. Indeed, VGX blunted the c-fos mRNA

expression induced by IP Ex-4 injection in the medial part of the NTS (75). The

neurochemical phenotype of the NTS neurons receiving vagal GLP-1R mediated signals is

currently unknown. One potential candidate are catecholamine (CA) neurons in the caudal

brainstem that are linked to the control of autonomic processes, food intake and stress (76).

CA neurons in the NTS receive direct monosynaptic inputs from visceral afferent neurons

and are activated by IV administration of GLP-1R agonists (77,78). Therefore, based on their

relevance and anatomical location, it is likely that CA neurons are key players in relaying the

VAN GLP-1R signal to higher brain centers.

Peripheral GLP-1R agonist treatment induces neuronal activation in multiple other brains

sites in the ascending visceral afferent pathway, including the parabrachial nucleus (PBN),

the paraventricular nucleus of the hypothalamus (PVN), and the central nucleus of amygdala

(CeA) (32,75). It is important to consider that the long lasting GLP-1R agonists can bind to

central GLP-1R in leaky brain areas such as the arcuate nucleus of the hypothalamus (Arc)

and in circumventricular organs such as the area postrema (AP) (38). They can therefore

indirectly activate other brain areas such as the ones mentioned above. Nonetheless, the c-

Fos expression pattern after HPV administration of a DPP-IV inhibitor (that raises HPV, but

not systemic GLP-1 levels) is to a great extent similar to the expression pattern induced by IP

Ex-4 (37). In the same study, VGX blocked the DPP-IV inhibitor-induced c-Fos expression in

all brain areas including the NTS, PVN, PBN, and CeA, suggesting that vagal GLP-1R

stimulation is required for the activation of brainstem and forebrain neurons linked to

ascending visceral pathways.

Abbott et al. provided evidence for a role of the vagal-brainstem-hypothalamic pathway in the

effects of GLP-1 on food intake by interrupting the connection between caudal brainstem and

hypothalamus using midbrain knife cuts (26). The PVN is one of the major hypothalamic sites

receiving afferent projections from the NTS involved in the control of food intake and energy


74

expenditure (79-81). The c-Fos activation in the PVN after IP Ex-4 was attenuated in VGX

(75) as well as in SDA rats (32). This decrease in PVN neuronal activation was associated

with an attenuated acute eating-inhibitory response to IP Ex-4 (32). This finding is consistent

with the idea that the PVN modulates food intake by integrating neural signals generated in

the brainstem as well as the hypothalamus in response to nutritional and hormonal factors

(82,83). Moreover, lesions (84,85) or a functional disruption of the PVN (86,87) both lead to

severe hyperphagia and obesity. Altogether, the available evidence strongly suggests that

the PVN is a major hypothalamic site mediating the eating-inhibitory effects of peripheral

GLP-1R activation.

To challenge this idea, Hayes and colleagues interrupted the forebrain-caudal brainstem

communication by generating supracollicular decerebrated rats and tested the effects of

peripheral GLP-1R stimulation on eating behavior in these animals. Ex-4 decreased food

intake and gastric emptying similarly between control and chronic decerebrated rats (88),

suggesting that the caudal brainstem is sufficient to mediate these effects of peripheral GLP-

1 and that the hypothalamic/forebrain processing is not necessary. In support of this

interpretation, the existence of a monosynaptic vago-vagal input output circuit in the NTS and

dorsal motor nucleus of the vagus (DMV) provides additional evidence for the sufficiency of

the caudal brainstem for the effect of control of food intake under peripheral GLP-1R

stimulation on eating (89). The contradictory findings of the midbrain knife cut and

decerebrated rat models are difficult to reconcile, but the choice of drug, GLP-1 (100

nmol/kg) vs. Ex-4 (1.2 - 3 µg/kg) may play a role. It is possible that the dose of Ex-4 used in

this study induced visceral malaise and a taste aversion that contributed to the eating

suppression (90). Furthermore, given the high lipid solubility of Ex-4, it is possible that it

exerted independent effects on forebrain circuits to inhibit eating without the brainstem

connections. Future studies are needed to further address the necessity of the brainstem-

forebrain connection in processing the vagal GLP-1R mediated inhibition of eating.


75

Whether activation of VAN GLP-1R initiates effects of food intake and glycemia through a

common neuronal pathway remains an open question. A recent study using visceral and

neuronal glp-1r KO mouse models disregarded the importance of GLP-1R in VAN and in the

CNS in the effects of liraglutide on glucose homeostasis (39). Although the role of VAN GLP-

1 signaling in glucose homeostasis is still controversial (57), it is generally thought that meal-

induced GLP-1 secretion leads to the activation of vago-vagal reflexes that control insulin

secretion indirectly (see above and (91)). Moreover, as mentioned earlier, the possibility that

the VN can also mediate an insulin-independent effect of GLP-1 on endogenous glucose

production is getting increasing attention (71-73). The exact mechanism of the neural effects

of GLP-1 on glycemia, including its central processing, still needs to be explored.

Neuropeptides mediating vagal GLP-1R effects in VA N 4.2

Due to limited viscerotopic organization of NTS neurons receiving vagal inputs (92), it is

unlikely that glutamatergic transmission alone would permit an unambiguous transmission of

vagal-branch specific information of gut-derived factors to the brainstem. Recently, vagally-

expressed neuropeptides have been found to play important roles in meal-induced satiation

and diet–induced obesity (74). The nodose ganglia (NG) express notably the cocaine- and

amphetamine-regulated transcript (CART) (93-95) and melanin-concentrating hormone

(MCH) (96-98). Both neuropeptides are also abundantly expressed in other brain areas

involved in the control of food intake. There is evidence that the expression of CART and

MCH in the NG is differentially regulated by the nutritional status and cholecystokinin (CCK).

Indeed, CART co-localizes with CCK1a receptors (CCK1aR) in VAN (94) and its expression

is induced in, and the peptide is released from, cultured primary VAN in response to CCK

(93). Moreover, leptin and ghrelin potentiate or inhibit this effect of CCK on CART expression

(93,99), raising the possibility that the VN is an important hub for the integration of gastric,

intestinal and adiposity signals. The CART regulation by CCK has enhanced our

understanding of the cellular mechanisms in VAN, which can be applied to the studies of

other vagally-mediated satiating signals. Currently, no information is available regarding a


76

neuropeptide expression under the control of GLP-1R signaling. It has been shown,

however, that the GLP-1R localization in the plasma membrane is nutritionally regulated and

increased in the NG by eating (100). Furthermore, GLP-1R expression is down-regulated in

the NG of rats prone to diet-induced obesity (DIO) fed with a high-fat diet (101), with a

concomitant decrease in circulating GLP-1 levels (24,101,102). This suggests an impaired

GLP-1R signaling that may contribute to the hyperphagia and obesity phenotypes of DIO rats

(24). Based on these findings, it is imperative to identify the neuropeptide(s) mediating the

vagal GLP-1R signaling and investigate the molecular events associated with the impaired

GLP-1R functions in obesity.

5 Methodological obstacles to overcome: suggestions for future

vagal manipulations

As mentioned briefly in the introduction, the research into the vagal mediation of GLP-1

effects has been hampered by a few technical caveats. In this last part, we aim at

highlighting the limitations of the techniques described above in order to suggest alternative

approaches for future vagal manipulations.

Lesion-based approaches and genetic manipulations of the vagus nerve 5.1

Until recently, manipulations of the VN could only be achieved by a lesion-based approach,

such as capsaïcin treatment or different variants of vagotomy. Although surgical techniques

have been modified to increase the specificity to VAN or VEN (29) or to a specific branch of

the VN, lesion approaches result in a non-specific impairment of the vagal neurons,

compromising a broad range of signaling. An example can be seen in SDA rats: food intake

and gastric emptying rate is already impaired in SDA compared to Sham rats (unpublished

data - 50% of the efferents are also sectioned). Consequently, the question remains whether

the SDA model is ideal for assessing the essential role of the vagus nerve in mediating GLP-


77

1 effects on food intake because food intake may already be limited by unspecific

impairments of vagal function.

Moreover, a lesioning approach also impairs other pathways involved in the control of food

intake, such as CCK satiation. To overcome this lack of specificity, many studies combined a

vagal manipulation with an exogenous infusion of GLP-1 or one of its agonists/antagonists.

This approach, however, presents the problems of choosing the right dose and route of

administration that can simulate the nutrient-induced release of GLP-1. Whereas IP

administration targets the activation of gastro-intestinal GLP-1R populations, more targeted

IV infusions - such as HPV – may help dissecting out the role of CHB, but do not seem to be

the ideal model to study the intestinal release of GLP-1 (35,103). Moreover, as mentioned

above, the use of pharmacological doses of GLP-1 or GLP-1R agonists rather than

physiological doses of GLP-1 increase the likelihood of direct activation of central GLP-1R

and may mask vagal-specific effects.

Recently, a study has overcome the need of vagal lesioning by performing a genetic deletion

of the glp1r gene based on the cre-lox recombination system in mice (39). Vagal-specific

genetic deletions published so far were based on the expression of a cre-recombinase driven

by a Phox2b (39,104) or Nav1.8 (105) promotor. Expression of these two genes is, however,

not restricted to the NG (106,107). Hence, the use of their promoters can drive non vagal cre

expression and potential non specific glp1r deletion, which could complicate interpretation.

Proposed approaches for future studies of vagal si gnaling in the control 5.2

of eating and glycemia

Based on these limitations, we propose a lesion-free technique that could specifically target

GLP-1R on VAN. First, as just mentioned, the main limitation of the cre-lox approach in this

context is whether or not it drives cre recombinase expression in a VAN-specific manner.

Consequently, it is reasonable to suggest that gene expression screening of VAN is needed

to identify candidate genes specific to these neurons. Another approach could involve the

delivery of viral vectors to unlock a new spectrum of vagal-specific genetic manipulations.


78

The NG contains the cell bodies of VAN and hence is a well-suited delivery site (108).

Whether viral vector delivery could potentially be efficient at manipulating gene expression in

VAN, and which type of viral vectors are suitable for these neurons, has not being

investigated yet. To be more specific, this approach can be used for the delivery of short

hairpin RNA (shRNA)-based vectors and opens the door for optogenetics or DREADD

(Designer Receptors Exclusively Activated by Designer Drugs) manipulations of the VN.

Furthermore, the field of vagus nerve stimulation (VNS) has recently received attention in

relation to the control of eating and the regulation of body weight (see (109) for review).

Approaches based on this technique could allow modulating vagal activity, which could be of

interest not only for therapeutic, but also for experimental purposes. None of these

approaches, however, distinguishes between the different vagal branches, i.e., receptor

populations. Consequently, further research is needed to overcome this methodological

obstacle in investigating vagal branch-specific functions. Altogether, considering these

perspectives will further enhance our understanding of the role of the VN in the control of

food intake and glucose homeostasis.

6 Concluding comments

In summary, evidence presented in this review strongly suggests that the contribution of the

VN to mediating GLP-1 effects on food intake and glycemia is context specific. Indeed,

although not being fully demonstrated, it is commonly assumed that the VN is a key player in

mediating the effects of intestinal GLP-1 on meal size and post-meal control of blood

glucose. The importance of this vagal pathway is, however, diminished when levels of

circulating GLP-1 are markedly elevated, such as after an IV GLP-1 infusion, or after DPP-IV

inhibitor treatment (see summary figure). Moreover, some reports indicate that the acute

effects of long-lasting GLP-1R agonists on food intake are at least in part vagally mediated,

whereas the weight loss and improved glucose tolerance after chronic treatments do not

seem to require vagal afferents. The mechanisms involved in the beneficial effects of chronic


79

agonist treatments are still investigated, but evidence indicates a crucial involvement of

central nervous system GLP-1R populations, especially in the hypothalamus, in appetite

reduction (38,39). Despite the recent reports questioning a role of the VN in mediating GLP-

1R agonist actions, further research is needed to address whether VAN GLP-1R activation

leads to appetite reduction and weight loss without the side effects normally associated with

GLP-1R agonist treatment in humans (110).

In addition, this review highlighted the fact that the central mechanisms mediating the effects

of vagal GLP-1R activation on food intake and glucose homeostasis, as well as the GLP-1R

VAN signaling, are not yet understood. We advocate further research to provide a complete

understanding of the neural mechanisms that relay intestinal GLP-1 signal to the brain and

induce satiation and glucoregulatory responses. Ideally, future studies should avoid lesion-

based approaches for the benefit of specific molecular manipulation of vagal GLP-1R

signaling.

This work was funded by ETH Zürich grant 4712-2 and SNSF MHV grant PMPDP3_151360


80

Summary figure : Meal-induced release of intestinal GLP-1 can access GLP-1R on vagal afferents of gastrointestinal (a) and hepatic origins (b). Vagal information is processed in the brainstem to activate a vago-vagal loop and/or higher brain centers (c). Vagal efferent projections to the stomach and pancreas control gastric emptying and participate in glucose-induced insulin production (d), supplementary to the direct incretin GLP-1 effect on pancreatic cells (e). New lesion-free approaches are needed to provide an unequivocal demonstration of this pathway. GLP-1R agonists have been shown to require vagal afferents to exert their acute effects on food intake when administered IP (a). Their chronic pharmacological effects on body weight and glucose tolerance, however, seem to be primarily mediated by a direct activation of central GLP-1R (f). Which central GLP-1R population mediates these effects requires further investigation


81

REFERENCES

1. Diakogiannaki, E., Gribble, F. M. & Reimann, F. Nutrient detection by incretin hormone secreting cells. Physiol Behav. 2012; 106:387–393.

2. Bucinskaite, V., Tolessa T, Pedersen J, Rydqvist B, Zerihun L, Holst JJ, Hellström PM. Receptor-mediated activation of gastric vagal afferents by glucagon-like peptide-1 in the rat. Neurogastroenterol Motil. 2009; 21:978–e78.

3. Ezcurra, M., Reimann, F., Gribble, F. M. & Emery, E. Molecular mechanisms of incretin hormone secretion. Curr Opin Pharmacol; 2013; 13:922–927.

4. Lim, G. E. & Brubaker, P. L. Glucagon-like peptide 1 secretion by the L-cell - The view from within. Diabetes. 2006; 55:S70–S77

5. Kreymann, B., Ghatei, M. A., Willams, G. & Bloom, S. R. Glucagon-like peptide-1 7-36: a physiological incretin in man. Lancet. 1987; 2(8571):1300-4.

6. Drucker, D. J., Philippe, J., Mojsov, S., Chick, W. L. & Habener, J. F. Glucagon-like peptide I stimulates insulin gene expression and increases cyclic AMP levels in a rat islet cell line. Proc Natl Acad Sci USA. 1987: 84,3434–3438.

7. Mojsov, S., Weir, G. C. & Habener, J. F. Insulinotropin: glucagon-like peptide I (7-37) co-encoded in the glucagon gene is a potent stimulator of insulin release in the perfused rat pancreas. J Clin Invest. 1987; 79:616–619.

8. Punjabi, M., Arnold, M., Geary, N., Langhans, W. & Pacheco-López, G. Peripheral Glucagon-like Peptide-1 (GLP-1) and Satiation. Physiol Behav; 2011:105(1):71-6.

9. Mentlein, R., Gallwitz, B. & Schmidt, W. E. Dipeptidyl-peptidase IV hydrolyses gastric inhibitory polypeptide, glucagon-like peptide-1(7-36)amide, peptide histidine methionine and is responsible for their degradation in human serum. Eur J Biochem; 1993; 214:829–835.

10. Deacon, C. F., Johnsen, A. H. & Holst, J. J. Degradation of glucagon-like peptide-1 by human plasma in vitro yields an N-terminally truncated peptide that is a major endogenous metabolite in vivo. J Clin Endocrinol Metab. 1995; 80:952–957.

11. Heymann, E. & Mentlein, R. Liver dipeptidyl aminopeptidase IV hydrolyzes substance P. FEBS Lett. 1978; 91:360–364.

12. Berthoud, H. R. & Neuhuber, W. L. Functional and chemical anatomy of the afferent vagal system. Auton Neurosci - Basic. 2000; 85:1–17.

13. Nakagawa, A. Satake H, Nakabayashi H, Nishizawa M, Furuya K, Nakano S, Kigoshi T, Nakayama K, Uchida K. Receptor gene expression of glucagon-like peptide-1, but not glucose-dependent insulinotropic polypeptide, in rat nodose ganglion cells. Auton Neurosci - Basic. 2004; 110:36-43.

14. Nakabayashi, H., Nishizawa, M., Nakagawa, A., Takeda, R. & Niijima, A. Vagal hepatopancreatic reflex effect evoked by intraportal appearance of tGLP-1. Am. J. Physiol. 1996; 271:E808–13.

15. Kakei, M., Yada, T., Nakagawa, A. & Nakabayashi, H. Glucagon-like peptide-1 evokes action potentials and increases cytosolic Ca. Auton Neurosci - Basic. 2002; 102:39-44.


82

16. Grill, H. J. & Hayes, M. R. Hindbrain Neurons as an Essential Hub in the Neuroanatomically Distributed Control of Energy Balance. Cell Metab. 2012; 16:296–309.

17. Flint, A., Raben, A., Astrup, A. & Holst, J. J. Glucagon-like peptide 1 promotes satiety and suppresses energy intake in humans. J Clin Invest. 1998; 101:515–520.

18. Gutzwiller, J.-P. Göke B, Drewe J, Hildebrand P, Ketterer S, Handschin D, Winterhalder R, Conen D, Beglinger C. Glucagon-like peptide-1: a potent regulator of food intake in humans. Gut. 1999; 44:81–86.

19. Flint, A., Raben, A., Ersbøll, A. K., Holst, J. J. & Astrup, A. The effect of physiological levels of glucagon-like peptide-1 on appetite, gastric emptying, energy and substrate metabolism in obesity. Int J Obes. 2001; 25:781–792.

20. Chelikani, P. K., Haver, A. C. & Reidelberger, R. D. Intravenous infusion of glucagon-like peptide-1 potently inhibits food intake, sham feeding, and gastric emptying in rats. Am J Physiol Regul Integr Comp Physiol. 2005; 288:R1695–R1706.

21. IImeryüz NI, Yegen BC, Bozkhurt A, Coskun T, Villanueva-Penacarrillo ML, Ulusoy NB. Glucagon-like peptide-1 inhibits gastric emptying via vagal afferent-mediated central mechanisms. Am J Physiol Gastrointest Liver Physiol 1997; 273:G920-927.

22. Larsen, P. J., Fledelius, C., Knudsen, L. B. & Tang-Christensen, M. Systemic administration of the long-acting GLP-1 derivative NN2211 induces lasting and reversible weight loss in both normal and obese rats. Diabetes. 2011; 50:2530–2539.

23. Rodriquez de Fonseca F, Navarro M, Alvarez E, Roncero i, Chowen JA, Maestre O, Gómez R, Muñoz RM, Eng J, and Blázquez E. Peripheral versus central effects of glucagon-like peptide-1 receptor agonists on satiety and body weight loss in Zucker obese rats. Metab Clin Exp. 2000; 49:709–717.

24. Williams, D. L., Baskin, D. G. & Schwartz, M. W. Evidence that Intestinal Glucagon-Like Peptide-1 Plays a Physiological Role in Satiety. Endocrinology. 2009; 150:1680–1687.

25. Talsania, T., Anini, Y., Siu, S., Drucker, D. J. & Brubaker, P. L. Peripheral exendin-4 and peptide YY(3-36) synergistically reduce food intake through different mechanisms in mice. Endocrinology. 2005; 146:3748–3756.

26. Abbott, C. R. Monteiro M, Small CJ, Sajedi A, Smith KL, Parkinson JR, Ghatei MA, Bloom SR The inhibitory effects of peripheral administration of peptide YY3–36 and glucagon-like peptide-1 on food intake are attenuated by ablation of the vagal–brainstem–hypothalamic pathway. Brain Res. 2005;1044:127–131.

27. Plamboeck, A. Veedfald S, Deacon CF, Hartmann B, Wettergren A, Svendsen LB, Meisner S, Hovendal C, Vilsbøll T, Knop FK, Holst JJ. The effect of exogenous GLP-1 on food intake is lost in male truncally vagotomized subjects with pyloroplasty. Am J Physiol Gastrointest Liver Physiol. 2013; 304:G1117–27.

28. Kraly, F. S., Jerome, C. & Smith, G. P. Specific postoperative syndromes after total and selective vagotomies in the rat. Appetite. 1986; 7:1–17.

29. Norgren, R. & Smith, G. P. A method for selective section of vagal afferent or efferent axons in the rat. Am J Physiol. 1994;1136–1141.


83

30. Walls EK, Wang FB, Holst MC, Philipps RJ, Voreis JS, Perkins A, Pollard LE, Powley TL. Selective vagal rhizotomies: a new dorsal surgical approach used for intestinal deafferentations. Am J Physiol. 1995; 269:R1279–88.

31. Schwartz, G. J., Plata-Salamán, C. R. & Langhans, W. Subdiaphragmatic vagal deafferentation fails to block feeding-suppressive effects of LPS and IL-1 beta in rats. Am J Physiol. 1997; 273:R1193–8.

32. Labouesse MA, Stadlbauer U, Weber E, Arnold M, Langhans W, Pacheco-Lopez. Vagal afferents mediate early satiation and prevent flavour avoidance learning in response to intraperitoneally infused exendin-4. J Neuroendocrinol. 2012; 24(12):1505-16.

33. Hayes MR, Kanoski SE, De Jonghe BC, Leichner TM, Alhadeff AL, Fortin SM, Arnold M, Langhans W, Grill HJ. The common hepatic branch of the vagus is not required to mediate the glycemic and food intake suppressive effects of glucagon-like-peptide-1. Am J Physiol Regul Integr Comp Physiol 2011; 301:R1479–1485

34. Kanoski, S. E., Fortin, S. M., Arnold, M., Grill, H. J. & Hayes, M. R. Peripheral and Central GLP-1 Receptor Populations Mediate the Anorectic Effects of Peripherally Administered GLP-1 Receptor Agonists, Liraglutide and Exendin-4. Endocrinology. 2011; 152:3103–3112.

35. Ruttimann, E. B., Arnold, M., Hillebrand, J. J., Geary, N. & Langhans, W. Intrameal Hepatic Portal and Intraperitoneal Infusions of Glucagon-Like Peptide-1 Reduce Spontaneous Meal Size in the Rat via Different Mechanisms. Endocrinology. 2008; 150:1174–1181.

36. Zhang, J. & Ritter, R. C. Circulating GLP-1 and CCK-8 reduce food intake by capsaicin-insensitive, nonvagal mechanisms. Am J Physiol. 2012; 302(2):R246-73.

37. Fujiwara K, Gotoh K, Chiba S, Masaki T, Katsuragi I, Kakuma T, Yoshimatsu H. Intraportal administration of DPP-IV inhibitor regulates insulin secretion and food intake mediated by the hepatic vagal afferent nerve in rats. J Neurochem 2012; 121:66–76.


39. Sisley S, Gutierrez-Aguilar R, Scott M, D'Alessio DA, Sandoval DA, Seeley RJ. Neuronal GLP1R mediates liraglutide’s anorectic but not glucose-lowering effect. J Clin Invest 2014; 124:2456-2463

40. Dossat, A. M., Lilly, N., Kay, K. & Williams, D. L. Glucagon-Like Peptide 1 Receptors in Nucleus Accumbens Affect Food Intake. J Neurosci. 2011; 31:14453–14457.

41. Dossat, A. M., Diaz R, Gallo L, Panagos A, Kay K, Williams DL. Nucleus accumbens GLP-1 receptors influence meal size and palatability. Am J Physiol Endocrinol Metab. 2013; 304:E1314–20.

42. Alhadeff, A. L., Rupprecht, L. E. & Hayes, M. R. GLP-1 Neurons in the Nucleus of the Solitary Tract Project Directly to the Ventral Tegmental Area and Nucleus Accumbens to Control for Food Intake. Endocrinology. 2012; 153:647–658.


84

43. Dickson, S. L., Shirazi RH, Hansson C, Bergquist F, Nissbrandt H Skibicka KP. The Glucagon-Like Peptide 1 (GLP-1) Analogue, Exendin-4, Decreases the Rewarding Value of Food: A New Role for Mesolimbic GLP-1 Receptors. J Neurosci. 2012; 32:4812–4820.

44. Mietlicki-Baase, E. G., Ortinski PI, Rupprecht LE, Olivos DR, Alhadeff AL, Pierce CR, Hayes MR. The food intake-suppressive effects of glucagon-like peptide-1 receptor signaling in the ventral tegmental area are mediated by AMPA/kainate receptors. Am J Physiol Endocrinol Metab. 2013; 305:E1367–74.

45. Skibicka, K. P. & Dickson, S. L. Enteroendocrine hormones-central effects on behavior. Curr Opin Pharmacol. 2013; 13:977–982.

46. Nishizawa, M., Nakabayashi, H., Uchida, K., Nakagawa, A. & Niijima, A. The hepatic vagal nerve is receptive to incretin hormone glucagon-like peptide-1, but not to glucose-dependent insulinotropic polypeptide, in the portal vein. J Auton Nerv Syst. 1996; 61:149–154.

47. Ahrén, B. The neuro-incretin concept. Regul Pept. 2014; 194-195:3-5.

48. Donath, M. Y. & Burcelin, R. GLP-1 effects on islets: hormonal, neuronal, or paracrine? Diabetes Care. 2013; 36 Suppl 2:S145–8.

49. Hayes, M. R., Mietlicki-Baase, E. G., Kanoski, S. E. & De Jonghe, B. C. Incretins and amylin: neuroendocrine communication between the gut, pancreas, and brain in control of food intake and blood glucose. Annu Rev Nutr. 2014; 34:237–260.

50. Plamboeck, A, Veedfald S, Deacon CF, Hartmann B, Wettergren A, Svendsen L, Meisner S, Hovendal C, Knop FK, Vilsboll T, Holst JJ. Characterisation of oral and i.v. glucose handling in truncally vagotomised subjects with pyloroplasty. Eur J Endocrinol. 2013; 169:187–201.

51. Nishizawa M, Nakabayashi H, Uehara K, Nakagawa A, Uchida K, Koya D. Intraportal GLP-1 stimulates insulin secretion predominantly through the hepatoportal-pancreatic vagal reflex pathways. Am J Physiol Endocrinol Metab 2013; 305:E376–387.

52. Burcelin, R., Da Costa, A., Drucker, D. & Thorens, B. Glucose competence of the hepatoportal vein sensor requires the presence of an activated glucagon-like peptide-1 receptor. Diabetes. 2001; 50:1720–1728.

53. Vahl, T. P., Tauchi M, Durler TS, Elfers EE, Ellis KS, Woods SC, Seeley RJ, Herman JP, D'Alessio DA. Glucagon-like peptide-1 (GLP-1) receptors expressed on nerve terminals in the portal vein mediate the effects of endogenous GLP-1 on glucose tolerance in rats. Endocrinology. 2007; 148:4965–4973.

54. Balkan, B. & Li, X. Portal GLP-1 administration in rats augments the insulin response to glucose via neuronal mechanisms. Am J Physiol Regul Integr Comp Physiol. 2000; 279:R1449–R1454.

55. Ahrén, B. Sensory nerves contribute to insulin secretion by glucagon-like peptide-1 in mice. Am J Physiol Regul Integr Comp Physiol. 2004; 286:R269–72.

56. Ahlkvist, L. & Ahrén, B. Inhibiting the cholinergic effector system of the vagus nerve reduces the insulinotropic effects of both GLP-1 and GIP. Diabetologia. 2014; 57:1–564.


85

57. Lamont, B. J., Li Y, Kwan E, Brown TJ, Gaisano, Drucker DJ. Pancreatic GLP-1 receptor activation is sufficient for incretin control of glucose metabolism in mice. J Clin Invest. 2012; 122:388–402.

58. Ionut, V., Hucking, K., Liberty, I. F. & Bergman, R. N. Synergistic effect of portal glucose and glucagon-like peptide-1 to lower systemic glucose and stimulate counter-regulatory hormones. Diabetologia. 2005; 48:967–975.

59. Dardevet, D, Moore MC, DiCostanzo CA, Lautz M, Cherrington AD. Insulin secretion-independent effects of GLP-1 on canine liver glucose metabolism do not involve portal vein GLP-1 receptors. Am J Physiol Gastrointest Liver Physiol. 2015; 289:G806–14.

60. Seghieri, M., Rebelos E, Gastaldelli A, Astiarraga BD, Casolaro A, Barsotti, Pocai A, Nauck M, Muscelli E, Ferrannini E. Direct effect of GLP-1 infusion on endogenous glucose production in humans. Diabetologia. 2013; 56:156–161.

61. Ohlsson, L, Alsalim W, Carr RD, Tura A, Pacini G, Mari A, Ahrén B. Glucose-lowering effect of the DPP-4 inhibitor sitagliptin after glucose and non-glucose macronutrient ingestion in non-diabetic subjects. Diabetes Obes Metab. 2013; 15:531–537.

62. Behme MT, Dupré J, McDonald T. Glucagon-like peptide 1 improved glycemic control in type 1 diabetes. BMC Endocr Disord. 2003; 3:3-12.

63. Dupre, J, Behme T, Hramiak IM, McFarlane P, Williamson MP, Zabel P, McDonald TJ. Glucagon-Like Peptide I Reduces Postprandial Glycemic Excursions in IDDM. Diabetes. 1995; 44:626-630.

64. Creutzfeldt, WO, Kleine N, Willms B, Orskov C, Holst JJ, Nauck MA. Glucagonostatic actions and reduction of fasting hyperglycemia by exogenous glucagon-like peptide-1 (7-36) amide in type I diabetic patients. Diabetes Care. 1996; 19(6):580-6.

65. Kielgast, U., Krarup, T., Holst, J. J. & Madsbad, S. Four weeks of treatment with liraglutide reduces insulin dose without loss of glycemic control in type 1 diabetic patients with and without residual beta-cell function. Diabetes Care. 2011; 34:1463–1468.

66. Ghazi, T., Rink, L., Sherr, J. L. & Herold, K. C. Acute metabolic effects of exenatide in patients with type 1 diabetes with and without residual insulin to oral and intravenous glucose challenges. Diabetes Care. 2014; 37:210–216.

67. Farngren, J., Persson, M., Schweizer, A., Foley, J. E. & Ahrén, B. Vildagliptin reduces glucagon during hyperglycemia and sustains glucagon counterregulation during hypoglycemia in type 1 diabetes. J Clin Endocrinol Metab. 2012; 97:3799–3806.

68. Jun, L. S., Millican RL, Haawkins ED, Konkol DL, Showalter AD, Christe ME, Michael MD, Sloop KW. Absence of Glucagon and Insulin Action Reveals a Role for the GLP-1 Receptor in Endogenous Glucose Production. Diabetes. 2015; 64:819–827.

69. Omar, B. A, Andersen B, Hald J, Raun K, Nishimura E, Ahrén B. Fibroblast growth factor 21 (FGF21) and glucagon-like peptide 1 contribute to diabetes resistance in glucagon receptor-deficient mice. Diabetes. 2014; 63:101–110.

70. Nishizawa, M. Moore MC, Shiota M, Gustavson SM, Snead WL, Neal DW, Cherrington AD. Effect of intraportal glucagon-like peptide-1 on glucose metabolism in conscious dogs. Am J Physiol Endocrinol Metab. 2003; 284:E1027–36.


86

71. Schwartz MW, Seeley RJ, Tschöp MH, Woods SC, Morton GJ, Myers MG, D'Alessio DA. Cooperation between brain and islet in glucose homeostasis and diabetes. Nature 2013; 503:59–66.

72. Jarrad M Scarlett, M. W. S. Gut-brain mechanisms controlling glucose homeostasis. F1000Prime Rep. 2015; 7:12.

73. Ahrén, B. Hepato-incretin function of GLP-1: novel concept and target in type 1 diabetes. Diabetes 64, 715–717 (2015).

74. de Lartigue, G. Putative role of neuropeptides in vagal afferent signaling. Physiol Behav; 2014. j.physbeh.2014.03.011

75. Baraboi, E.-D., St-Pierre, D. H., Shooner, J., Timofeeva, E. & Richard, D. Brain activation following peripheral administration of the GLP-1 receptor agonist exendin-4. Am J Physiol Regul Integr Comp Physiol. 2011; 301:R1011–24.

76. Yamamoto, H. Kishi T, Lee CE, Choi BJ, Fang H, Hollenberg AN, Drucker DJ, Elmquist JK. Glucagon-like peptide-1-responsive catecholamine neurons in the area postrema link peripheral glucagon-like peptide-1 with central autonomic control sites. J Neurosci. 2003; 2939–2946.

77. Sumal, K. K., Blessing, W. W., Joh, T. H., Reis, D. J. & Pickel, V. M. Synaptic interaction of vagal afferents and catecholaminergic neurons in the rat nucleus tractus solitarius. Brain Res. 1983; 277:31–40.

78. Appleyard, S. M., Marks D, Kobayashi K, Okano H, Low MJ, Andresen MC. Visceral afferents directly activate catecholamine neurons in the solitary tract nucleus. J Neurosci. 2007; 27:13292–13302.

79. Guevara-Aguilar, R., Jimenez-Montufar, L. L., Garcia-Diaz, D. E., Wayner, M. J. & Armstrong, D. L. Olfactory and visceral projections to the paraventricular nucleus. Brain Res. 1988; 20:799–801.

80. Rinaman, L. Ascending projections from the caudal visceral nucleus of the solitary tract to brain regions involved in food intake and energy expenditure. Brain Res. 2010; 1350:18–34.

81. Rinaman, L. Visceral sensory inputs to the endocrine hypothalamus. Front Neuroendocrinol. 2007; 28:50–60.

82. Cowley, M. A, Pronchuk N, Fan W, Dinulescu DM, Colmers WF, Cone RD. Integration of NPY, AGRP, and melanocortin signals in the hypothalamic paraventricular nucleus: evidence of a cellular basis for the adipostat. Neuron. 1999; 24:155–163.

83. McKinley, M. J., Johnson AK. The Physiological Regulation of Thirst and Fluid Intake. News Physiol Sci. 2004; 19:1–6.

84. Gold, R. M. Hypothalamic obesity: the myth of the ventromedial nucleus. Science. 1973; 182:488–490.

85. Sims, J. S. & Lorden, J. F. Effect of paraventricular nucleus lesions on body weight, food intake and insulin levels. Behav Brain Res; 1986; 22:265–281.

86. Michaud, J. L., Boucher F, Melnyk A, Gauthier F, Goshu E, Lévy E, Mitchell GA, Himms-Hagen J, Fan CM. Sim1 haploinsufficiency causes hyperphagia, obesity and


87

reduction of the paraventricular nucleus of the hypothalamus. Hum Mol Genet. 2001; 10:1465–1473.

87. Holder, J. L., Butte, N. F. & Zinn, A. R. Profound obesity associated with a balanced translocation that disrupts the SIM1 gene. Hum Mol Genet. 2000; 9(1):101-8.

88. Hayes, M. R., Skibicka, K. P. & Grill, H. J. Caudal brainstem processing is sufficient for behavioral, sympathetic, and parasympathetic responses driven by peripheral and hindbrain glucagon-like-peptide-1 receptor stimulation. Endocrinology. 2008; 149: 4059–4068.

89. Rinaman, L., Card, J. P., Schwaber, J. S. & Miselis, R. R. Ultrastructural demonstration of a gastric monosynaptic vagal circuit in the nucleus of the solitary tract in rat. J Neurosci. 1989; 9:1985–1996.

90. Tang-Christensen, M, Larsen PJ, GÖke R, Fink-jensen A, Jessop DS, Moller M, Sheikh SP. Central administration of GLP-1-(7-36)amide inhibits food and water intake in rats. Am Physiol. 1996; 271(4 Pt 2):R848-56.

91. Nadkarni, P., Chepurny, O. G. & Holz, G. G. Regulation of Glucose Homeostasis by GLP-1. Prog Mol Biol Transl Sci. 2014; 121. 2014; 121:23-65.

92. Paton, J. F., Li, Y. W., Deuchars, J. & Kasparov, S. Properties of solitary tract neurons receiving inputs from the sub-diaphragmatic vagus nerve. Neurosci. 2000; 95:141–153.

93. de Lartigue, G., Dimaline, R., Varro, A. & Dockray, G. J. Cocaine- and amphetamine-regulated transcript: stimulation of expression in rat vagal afferent neurons by cholecystokinin and suppression by ghrelin. J Neurosci. 2007; 27:2876–2882.

94. Broberger, C., Holmberg, K., Kuhar, M. J. & Hökfelt, T. Cocaine- and amphetamine-regulated transcript in the rat vagus nerve: A putative mediator of cholecystokinin-induced satiety. Proc Natl Acad Sci USA. 1999; 96:13506–13511.

95. Zheng, H., Patterson, L. M. & Berthoud, H.-R. CART in the dorsal vagal complex: sources of immunoreactivity and effects on Fos expression and food intake. Brain Research 957, 298–310 (2002).

96. Skofitsch, G., Jacobowitz, D. M. & Zamir, N. Immunohistochemical localization of a melanin concentrating hormone-like peptide in the rat brain. Brain Res Bull. 1985; 15:635–649.

97. Zamir, N., Skofitsch, G. & Jacobowitz, D. M. Distribution of immunoreactive melanin-concentrating hormone in the central nervous system of the rat. Brain Res. 1986; 373, 240–245.

98. Zamir, N., Skofitsch, G., Bannon, M. J. & Jacobowitz, D. M. Melanin-concentrating hormone: unique peptide neuronal system in the rat brain and pituitary gland. Proc Natl Acad Sci USA. 1986; 83(5):1528-31.

99. de Lartigue, G, Lur G, Dimaline R, Varro A, Raybouald H, Dockray GJ. EGR1 Is a Target for Cooperative Interactions between Cholecystokinin and Leptin, and Inhibition by Ghrelin, in Vagal Afferent Neurons. Endocrinology. 2010; 151:3589–3599.

100. Ronveaux, C. C., de Lartigue, G. & Raybould, H. E. Ability of GLP-1 to decrease food


88

intake is dependent on nutritional status. Physiol Behav. 2014; 135:222–229.

101. Duca, F. A., Sakar, Y. & Covasa, M. Combination of obesity and high-fat feeding diminishes sensitivity to GLP-1R agonist, Exendin-4. Diabetes. 2013; 62(7):2410:5.

102. Duca, F. A., Swartz, T. D., Sakar, Y. & Covasa, M. Decreased intestinal nutrient response in diet-induced obese rats: role of gut peptides and nutrient receptors. Int J Obes. 2012; 37(3):375-81.

103. Punjabi M, Arnold M, Rüttimann E, Graber M, Geary N, Pacheco-López G, Langhans W. Circulating glucagon-like peptide-1 (GLP-1) inhibits eating in male rats by acting in the hindbrain and without inducing avoidance. Endocrinology. 2014; 155(5):1690–9. .


105. de Lartigue, G., Ronveaux, C. C. & Raybould, H. E. Deletion of leptin signaling in vagal afferent neurons results in hyperphagia and obesity. Mol Metab. 2014; 3:595–607.

106. Scott, M. M., Williams, K. W., Rossi, J., Lee, C. E. & Elmquist, J. K. Leptin receptor expression in hindbrain Glp-1 neurons regulates food intake and energy balance in mice. J Clin Invest. 2011; 121:2413–2421.

107. Stirling, L. C., Forlani G, Baker MD, Wood JN, Matthews EA, Dickenson AH, Nassar MA. Nociceptor-specific gene deletion using heterozygous NaV1.8-Cre recombinase mice. Pain. 2005; 113:27–36.

108. Calik, M. W., Radulovacki, M. & Carley, D. W. A Method of Nodose Ganglia Injection in Sprague-Dawley Rat. J Vis Exp. 2014; (93):e52233.

109. McClelland, J., Bozhilova, N., Campbell, I. & Schmidt, U. A systematic review of the effects of neuromodulation on eating and body weight: evidence from human and animal studies. Eur Eat Disord Rev. 2013; 21:436–455.

110. Meier, J. J. GLP-1 receptor agonists for individualized treatment of type 2 diabetes mellitus. Nat Rev Endocrinol. 2012; 8:728–742.

89

Chapter 3: GLP-1R knockdown in vagal afferent neurons, food intake and glycemia

91

CHAPTER 3:

KNOCKDOWN OF GLP-1 RECEPTORS IN VAGAL AFFERENTS

AFFECTS NORMAL FOOD INTAKE AND GLYCEMIA

AUTHORS

Krieger JP, Arnold M, Pettersen KG, Lossel P, Langhans W, Lee SJ


STATUS

Published in Diabetes (2016; 65(1):34-43)


92

ABSTRACT

Nutrient stimulation of enteroendocrine L-cells induces the release of the incretin and

satiating peptide glucagon-like peptide-1 (GLP-1). The vagus nerve innervates visceral

organs and may contribute to the mediation of gut-derived GLP-1’s effects on food intake

and glycemic control. To test the hypothesis that vagal afferent neuron (VAN) GLP-1

receptors (GLP-1R) are necessary for the effect of endogenous GLP-1 on eating and energy

balance, we established a novel bilateral nodose ganglia (NG) injection technique to deliver a

lentiviral vector and to knock down VAN GLP-1R in male Sprague Dawley rats. We found

that a full expression of VAN GLP-1R is not necessary for the maintenance of long-term

energy balance in normal eating conditions. VAN GLP-1R knockdown (kd) did, however,

increase meal size and accelerated gastric emptying. Moreover, post-meal glycemia was

elevated and insulin release was blunted in GLP-1R kd rats, suggesting that VAN GLP-1R

are physiological contributors to the neuroincretin effect after a meal. Collectively, our results

highlight a crucial role for the VAN in mediating the effects of endogenous GLP-1 on food

intake and glycemia and may promote the further development of GLP-1-based therapies.


93

INTRODUCTION

Glucagon-like peptide-1 (GLP-1) is an incretin and satiating hormone that has provided new

tools for the pharmacotherapy of obesity and diabetes (1,2). Yet, despite the clinical

effectiveness of GLP-1-based drugs in ameliorating the symptoms of type 2 diabetes, the

role of endogenous GLP-1 in the control of energy intake and glucose homeostasis is not

fully understood. Vagal afferent neurons (VAN) express GLP-1R (3,4) and terminate in the

lamina propria of the intestinal mucosa as well as in the wall of the hepatic portal vein (HPV)

(5). VAN may therefore relay the gut GLP-1-derived signals to the brain and, hence, mediate

satiating and glucoregulatory responses. Previous studies using lesioning approaches have

implicated the vagus nerve in the effects of peripherally administered GLP-1 on food intake

and glycemia (see (6) and (7) for review). In more recent studies, sudiaphragmatic vagal

deafferentation (SDA) in rats clearly attenuated the acute eating-inhibitory effect of

intraperitoneally (IP) infused GLP-1 (8) and exendin-4 (Ex-4, a GLP-1R agonist) (9).

Moreover, unlike Sham-operated rats, SDA rats failed to show a GLP-1R-mediated incretin

response (10). Based on these findings, it is reasonable to hypothesize that endogenous gut-

derived GLP-1 could activate GLP-1R on VAN in a paracrine-like fashion to reduce food

intake, limit gastric emptying and trigger a neural component of the incretin effect. Disruption

of this endogenous GLP-1 signaling mechanism in the VAN due to genetic or environmental

factors may contribute to the pathophysiology of obesity and diabetes. Hence, we examined

the physiological role of VAN GLP-1R in the control of food intake and regulation of glucose

homeostasis by generating a specific knockdown (kd) of VAN GLP-1R expression in rats.

Our approach is based on the delivery of a shRNA construct targeting the GLP-1R mRNA

transcript by injecting a lentiviral vector bilaterally into the nodose ganglia (NG) of rats. Using

RNA interference to manipulate gene expression in a tissue specific manner, we report that

VAN GLP-1R 1) are required for the physiological control of meal size and gastric emptying,

but not for the regulation of long-term energy intake and body weight; 2) are necessary for

the full effects of acute IP GLP-1 and Ex-4 administration on eating and gastric emptying; 3)


94

mediate a neural component of GLP-1’s incretin effect that is physiologically relevant for the

post-prandial control of blood glucose.

Collectively, our findings establish the VAN as a major mediator of endogenous GLP-1’s

short-term effects on eating, gastric emptying and glycemia. They also indicate, however,

that the reduction of GLP-1R expression in VAN is not sufficient to promote obesity under

normal eating conditions.

METHODS

Animals and housing

Male Sprague Dawley rats (Charles River) were individually housed (21±1°C, 55±5% HR)

with a 12/12 h dark/light cycle. Unless otherwise noted, animals had ad libitum access to

water and standard chow (Kliba 3433, energy density: 3.13 kcal/g). All experimental

procedures were approved by the Zürich Cantonal Veterinary Office.

Lentivirus-mediated short hairpin RNA interference

pLKO.1-puro vectors expressing turboGFP and the U6 promoter-driven shRNA sequence

targeting the rat GLP-1R mRNA or a non-target shRNA sequence were obtained from

Sigma-Aldrich. Efficiency of the GLP-1R-targeting shRNA construct was verified in vitro in

INS-1E cells (Prof. Maechler and Prof. Wollheim, Geneva University). GLP1-R-targeting or

control lentiviral particles were produced in HEK 293T cells using the pMD2.G and psPAX2

plasmids (Prof. Trono, Addgene #12259 and 12260) and concentrated to 1010 particles/mL

using 8% PEG6000 (Millipore) and resuspended in PBS.

Surgery

Rats (290-340 g on surgery day) were anesthetized by an IP injection of ketamine (88 mg/kg,

Ketalar, Kanthonsapotheke Zürich) and xylazin (5 mg/kg, Rompun 2%, Kantonsapotheke

Zürich) and nodose ganglia (NG) were exposed. A glass capillary (50 µm tip) was used to


95

administer 1.5 µL of viral solution into each NG with a Picospritzer III injector (Parker

Hannifin). To ensure expression of the viral constructs, animals were allowed to recover for

20 days. IP and HPV catheters were implanted as previously described (see (11) for

complete description). Intracerebroventricular (ICV) cannulas were implanted in the 4th

ventricle (stereotaxic coordinates: 2.5 mm posterior to lambda, 0 mm lateral to midline, 5 mm

below skull surface) and placement was verified functionally with infusion of 5-thioglucose

(Sigma-Aldrich) using a 2.5 mm injector (210 µg/rat) and anatomically post-mortem.

Tissue collection

Animals received an IP injection of pentobarbital (100 mg/kg, Kantonsapotheke Zürich) and

NG, brain and pancreas were immediately collected. For gene and protein analysis tissues

were frozen in liquid nitrogen and stored at -80°C. For GFP visualization, NG were fixed for 2

h in 4% paraformaldehyde, 25% sucrose solution in PBS and cut at 10 µm in a cryostat and

mounted on glass slides.

Gene expression and protein analysis

The nucleus tractus solitarii (NTS), the hypothalamic dorsomedial, paraventricular, and

arcuate nuclei (DMH, PVH and Arc, respectively) were mircropunched using anatomical

landmarks, and NG from the same animal were pooled before RNA and proteins were

extracted using Trizol (Life Technologies). RT-qPCR was performed using SybR Green on a

OneStep Plus instrument (Applied Biosystems) and results were analyzed using the 2ddCt

method. A western blot was performed to detect the GLP-1R protein (Rabbit antibody 39072,

1:400, Abcam) using β-actin as reference (Mouse antibody AC-74, 1:3000, Sigma-Aldrich).

Drugs

GLP-1(7-36)amide (GLP-1, Bachem H-6795), exendin-4 (Ex-4, Bachem H-8730) and

cholecystokinin octapeptide (CCK, Bachem H-2080) were resuspended in sterile PBS and

administered at doses of 33 µg/kg (GLP-1), 0.3 µg/kg (Ex-4) and 4 µg/kg (CCK) via IP

catheters. Ex-4 was administered into the 4th ventricle at a dose of 0.3 µg/rat. Rats were


96

habituated to IP or ICV injections with vehicle solutions on three occasions before

experiments.

Food intake measurement and meal pattern analysis

Food was available through a niche and placed on scales (XS4001S, Mettler-Toledo) for

continuous measurement (see (12) for description). Meal patterns were analyzed with

custom software (LabX meal analyzer 1.4, Mettler-Toledo). Data are presented as average of

3 days. For food intake experiments after IP GLP-1, rats were fasted overnight and received

a 3 g pre-meal 1 h before dark onset to allow GLP-1R trafficking to the VAN membrane (4).

For IP CCK, Ex-4 or ICV Ex-4, rats were fasted for 4 h before dark onset. In all cases, rats

received IP or ICV injections right before dark onset and were brought immediately to their

home cages.

Gastric emptying assay

One week prior to the experiment, rats were habituated to test meals and restricted feeding

schedule (Test meal at dark onset, ad libitum food access from 3 to 8 h after dark onset, food

deprivation otherwise). On experimental days, rats received a 4 g chow meal containing 1%

(w/w) paracetamol (4-acetamido-phenol, Sigma-Aldrich) and 0.25% (w/w) saccharin (Sigma-

Aldrich). IP or ICV injections were given 5 min prior to the test meal. Baseline tail vein blood

was taken 30 min prior to test meal onset and post-meal blood was collected according to

the scheduled timepoints. Paracetamol concentrations were measured with a commercial kit

(Cambridge Life Sciences K8002) .

Indirect calorimetry

Measurements were conducted in an open circuit calorimetry Phenomaster system (TSE)

after 5 days of habituation. Data are presented as 1 h time bins averaged over 3 days.

Plasma analysis after test meal

Blood was sampled from HPV catheters in unrestrained animals 30 min prior to (baseline)

and according to the scheduled timepoints after the beginning of a 5 g chow test meal.


97

Glucose was measured twice using AccuCheck (Roche) and 150 µL of blood was

immediately mixed with EDTA (Titriplex, Merck), Aprotinin (Sigma-Aldrich) and DPP-IV

inhibitor (Millipore) before centrifugation and storage of the plasma at -80°C. Total active

GLP-1, insulin and glucagon were measured simultaneously using an immunoassay

(MesoScale Discovery multi-spot K15171C).

Oral glucose tolerance test

16 h-food deprived rats adapted to gavage received an oral bolus of 40% glucose solution (2

g/kg). Blood samples for glucose and insulin were taken from tail vein at baseline and 15, 30,

60, 90, and 120 min after the oral glucose bolus. Insulin was measured using an

immunoassay (MesoScale Discovery single-spot for mouse/rat K152BZC).

Statistical analysis

Data normality was verified using the Shapiro-Wilk (when n≥7) and the Kolmogorov-Smirnov

(when n≤6) tests and homoscedasticity was checked by visualizing the distribution of

residuals. Non-parametric tests were used otherwise. When data distribution was compatible

with normality, outliers were detected using the Grubb’s test. Differences were analyzed by a

Student t-test for unpaired normally distributed values of equal variance (Fig. 1 C, E, F, G;

Fig. 2 D; Fig 3 A-I, K; Fig. 5 A, B), or a Mann-Whitney test for unpaired comparison of non

normally distributed data (Fig. 1 D; Fig. 3 L) using GraphPad Prism (version 6.05 for

Windows). Where the dependent variable was affected by two factors – one within-subject

factor (time or injection) and one between-subject factor (surgery group), the data were

analyzed with a mixed ANOVA (Fig. 2 A-C; Fig. 3 J; Fig. 4; Fig. 5 C-H) using SAS (version

9.3). When the main effect- or interaction terms were significant, post-hoc analyses using the

Bonferroni correction were performed. Data are presented as mean ± SEM. P-values < 0.05

were considered significant. All graphs were generated using GraphPad Prism (version

6.05).


98

RESULTS

Histological confirmation of viral infection and qu antification of in vivo GLP-1R kd

Three weeks after bilateral NG injection of a lentivirus containing a GLP-1R-targeting shRNA

construct (Fig. 1A), infection of VAN was confirmed by visualizing GFP expression in NG

sections (Fig. 1B). GLP-1R mRNA expression in NG was reduced by 52.5% in GLP-1R kd

rats compared to control rats injected with LV-shCTL containing non-specific target shRNA

(Fig. 1C). In addition, reduction of GLP-1R protein was confirmed using Western Blot from

NG protein extracts of control and GLP-1R kd rats (Fig. 1D). GLP-1R expression was

unchanged in the pancreas, where GLP-1R activation improves insulin secretion, as well as

in the key GLP-1R expressing regions in the brain involved in the control of food intake and

glucose homeostasis (Fig. 1E), indicating tissue specificity of the LV-mediated gene kd

approach. Moreover, VAN genes involved in the control of food intake, such as the

cholecystokinin A receptor (CCKaR), leptin receptor (LepR) and peroxisome proliferator

activated receptor γ (PPARγ), were similarly expressed in the NG of control and GLP-1R kd

rats (Fig. 1F), suggesting the GLP-1R shRNA construct employed was specific. Finally, we

measured food intake 30 min following an IP injection of CCK and found a 25-40% reduction

(9,13) in both groups, demonstrating the preservation of VAN functional integrity (Fig. 1G).

Together, these results indicate that LV-mediated delivery of a shRNA construct by bilateral

NG injection is tissue-specific, target-specific and does not impair vagal afferent function.

Endogenous GLP-1R signaling in the VAN is not requi red for normal long-term energy

balance

Body weights of control and GLP-1R kd rats remained similar during the entire course of the

experiments when fed ad libitum (Fig. 2A). Moreover, energy balance remained undisturbed

as documented by similar daily chow intake (Fig. 2B) and daily energy expenditure (Fig. 2C,

D).


99

Figure 1: NG injection of a GLP-1R-targeting lentiv irus led to a specific reduction in GLP-1R expression in the VAN (A) Schematic representation of lentiviral injection site in a rat nodose ganglion. (B) Visualization of GFP expression in the nodose ganglion of a LV-shGLP-1R injected rat, representative picture. Scale bar, 50 µm. (C) Relative expression of GLP-1R mRNA in the nodose ganglia of control and GLP-1R kd rats (n = 10/8; Student t-test, P < 0.0001). (D) Relative expression of the GLP-1R protein levels in the nodose ganglia of control and GLP-1R kd rats (n = 5/5; Mann-Whitney test, P < 0.01) as measured by the relative intensity of the GLP-1R detection band normalized by the intensity of the β-actin band, with representative examples. The dotted lane indicates where non-contiguous bands were grouped. (E) Relative expression of GLP-1R mRNA in the pancreas (n = 6/7; Student t-test, ns), NTS (n = 7/7; Student t-test, ns), Arc (n = 7/7; Student t-test, ns), PVH (n = 7/6; Student t-test, ns) and DMH (n = 7/7; Student t-test, ns) of control and GLP-1R kd rats. (F) Relative expression of LepR (n = 6/7; Student t-test, ns), CCKaR (n = 6/7; Student t-test, ns) and PPARg (n = 6/7; Student t-test, ns) mRNA in the nodose ganglia of control and GLP-1R kd rats. (G) Percentage of decrease in 1 h food intake after IP injection CCK (4 µg/kg) relative to vehicle injection (n = 6/6; Student t-test, ns) of control and GLP-1R kd rats.* indicates a significant difference between the control and GLP-1R kd groups (P < 0.05) Endogenous GLP-1R signaling in the VAN controls mea l size and gastric emptying

Measurements of undisturbed meal patterns showed that GLP-1R kd induced increases in

meal size (Fig. 3A) and meal duration (Fig. 3D). Although these changes were significant

over 24 h, the increase in meal size and duration was only evident during the dark phase

with no significant differences in the light phase (Fig. 3B, 3C, 3E and 3F). Consistent with the

long-term daily food intake measurements (Fig. 2A), 24 h food intake was not affected by the

kd during the period of meal pattern measurements (data not shown): this was mainly due to

a compensatory decrease in the number of meals in the GLP-1R kd rats (Fig. 3G), which

was also evident only during the dark phase (Fig. 3H, 3I). Moreover, the rate of gastric


100

emptying after a meal, as measured by the appearance of paracetamol in the plasma after a

test meal, was enhanced in GLP-1R kd rats compared to controls (Fig. 3J). Finally, in line

with their ad libitum meal pattern, GLP-1R kd rats showed an increase in food intake during

the 1 h refeeding period after a 16 h fast (Fig. 3K), associated with an amplified peak of

energy expenditure during the 1 h refeeding period (Fig. 3L).

Figure 2: VAN GLP-1R are not required for normal bo dy weight, daily food intake and energy expenditure (A) Body weight (n = 11/11; ANOVA, group F(1,20) = 0.26, ns) and (B) daily food intake (n = 11/11; ANOVA, group F(1,20) = 1.10, ns) of control and GLP-1R kd rats maintained on chow after surgical injection (C) 24 h time course (n = 8/7; ANOVA, group F(1,13) = 0.08, ns) and (D) cumulative energy expenditure (n = 8/7; Student t-test, ns) in control and GLP-1R kd rats fed ad libitum with chow

Endogenous GLP-1 signaling in the VAN is required f or the effects of IP GLP-1 and Ex-

4 but not ICV Ex-4 on food intake and gastric empty ing

According to previous studies (see (6,7) for review), GLP-1 and low-dose Ex-4 require intact

VAN to exert their full inhibitory effects on short-term food intake. Hence, we further tested

whether the satiating effects of exogenous GLP-1 and Ex-4 were attenuated in GLP-1R kd

rats. Indeed, IP injections of GLP-1 – at a dose that elevates intestinal lymph GLP-1 similar

as a meal (14) – or low dose Ex-4 failed to significantly reduce 1 h food intake in GLP-1R kd

rats (Fig. 4A, B). In a similar design, we tested whether GLP-1R in the VAN mediate


101

Figure 3: VAN GLP-1R controlled the size and the ga stric emptying of a meal (A) Average meal size over 24 h, (B) 12 h dark phase and (C) 12 h light phase (n = 8/7; Student t-tests, respectively P < 0.05, P < 0.05, ns) in ad libitum-fed control and GLP-1R kd rats. (D) Average meal duration over 24 h, (E) 12 h dark phase and (F) 12 h light phase (n = 8/7; Student t-tests, respectively P < 0.05, P = 0.07, ns) in ad libitum-fed control and GLP-1R kd rats. (G) Number of meals in 24 h, (H) 12 h dark phase, (I) 12 h light phase (n = 8/7; Student t-tests, respectively P < 0.05, P < 0.01, ns) in ad libitum-fed control and GLP-1R kd rats. (J) Plasma paracetamol concentrations of control and GLP-1R kd rats after a 4 g powdered chow test meal containing 1% paracetamol (w/w) (n = 9/9; ANOVA, group F(1,16) = 17.98, P < 0.0001; group x time F(6,96) = 4.71, P < 0.0001). (K) Food intake (n = 10/10; Student t-test, P < 0.05) and (L) peak of energy expenditure in the first hour of ad libitum refeeding with chow after a 16 h fast (n = 8/7; Mann-Whitney test, P < 0.05). * indicates a significant difference between the control and GLP-1R kd groups (P < 0.05)


102

the GLP-1- or Ex-4-induced inhibition of gastric emptying using the paracetamol test. IP

GLP-1 and Ex-4 failed to inhibit gastric emptying in the GLP-1R kd group as shown by early

appearance of paracetamol in the plasma 20 minutes after a test meal (Fig. 4D, E). To test

whether these effects of GLP-1R kd are mediated by receptors expressed on peripheral or

central terminals of vagal afferents, we performed 4th ventricular (4th ICV) injection of Ex-4.

In contrast to the attenuated effects of IP Ex-4, 4th ICV injection of Ex-4 showed the full

expression of the inhibitory effects on eating and gastric emptying in kd rats (Fig. 4C, F),

consistent with the idea that peripheral but not central GLP-1R expressed on VAN terminals

mediate the effects of IP infused GLP-1 on food intake and gastric emptying. Together, these

results indicate that activation of GLP-1R on the VAN in the gut mediates the inhibitions of

eating and gastric emptying induced by IP GLP-1 or low-dose Ex-4.

Endogenous GLP-1 signaling in the VAN is required f or normal glycemia after a

mixed-nutrient meal but not after oral glucose chal lenge

GLP-1R on VAN have been implicated in the neuroincretin effects of endogenous GLP-1. We

measured HPV blood glucose in the fasted and fed conditions in control and GLP-1R kd rats.

After an overnight fast (16 h), HPV blood glucose was not different between GLP-1R kd and

control rats (Fig. 5A), whereas kd rats showed a higher blood glucose than controls in the fed

state (2 h of fasting after ad libitum food access 5 h into the dark phase- Fig. 5B). This

indicated that GLP-1R in the VAN are necessary for the full incretin effect of meal-induced

GLP-1. To test this hypothesis, we measured HPV blood glucose (Fig. 5C), plasma insulin

(Fig. 5D), GLP-1 (Fig. 5E), and glucagon (Fig. 5F) after a 16 h fast followed by a 5 g chow

test meal. HPV blood sampling was chosen to allow for the concomitant measurement of the

meal-induced increase in HPV GLP-1 levels (which is subject to a rapid degradation by DPP-

IV in the systemic circulation and in the liver). The test meal elevated HPV blood glucose in

both groups but resulted in a higher post-meal blood glucose level in the GLP-1R kd rats.

Interestingly, post-meal HPV GLP-1 and glucagon were similar in both groups, but insulin

appearance in the HPV was blunted in the GLP-1R kd group. An oral GTT, however, did not


103

reveal differences in plasma glucose and insulin levels between control and kd rats (Fig. 5G,

H).

Figure 4: VAN GLP-1R kd attenuated the effects of I P GLP-1 and Ex-4, but not ICV Ex-4, on food intake and gastric emptying (A) 1 h food intake after IP GLP-1 (33 µg/kg; n = 7/8), (B) IP Ex-4 (0.3 µg/kg; n = 12/11) and (C) ICV Ex-4 (0.3 µg; n = 6-8). (D) Plasma paracetamol concentrations of control and GLP-1R kd rats 20 minutes after allowing access to a 4 g powdered chow test meal containing 1% paracetamol (w/w) and injected with IP GLP-1 (33 µg/kg; n = 7/7), (E) IP Ex-4 (0.3 µg/kg; n = 7/7) and (F) ICV Ex-4 (0.3 µg; n = 6-8). For all results, ANOVA was followed by post-hoc comparisons. Different letters indicate a significant difference between two groups after post-hoc Bonferroni-corrected comparisons (P < 0.05)


104

Figure 5: VAN GLP-1R kd disturbed post-meal glycemi a and insulinemia but did not impair tolerance of an oral glucose bolus (A) HPV glucose after an overnight fast (n = 8/8; Student t-test, ns) or (B) 2 h after food deprivation of ad libitum fed animals (n = 6/6; Student t-test, P<0.05). (C) HPV blood glucose (n = 8/8; ANOVA; group F(1,14) = 28.4, P < 0.0001; group x time F(5,70) = 3.440, P < 0.01), (D) plasma insulin (n = 8/8; ANOVA; group F(1,14) = 0.012, ns; group x time F(5,70) = 2.90, P < 0.05), (E) GLP-1 (n = 8/8; ANOVA; group F(1,14) = 0.73, ns; group x time F(5,70) = 0.68, ns) and (F) glucagon (n = 8/8; ANOVA; group F(1,14) = 0.13, ns; group x time F(5,70) = 0.29, ns) after a chow test meal. (G) Tail vein blood glucose (n = 7/7; ANOVA; group F(1,12) = 0.30, ns; group x time F(6,72) = 1.29, ns) and (H) plasma insulin (n = 7/7; ANOVA; group F(1,12) = 0.16, ns; group x time F(4,48) =1.12, ns) after an OGTT (2 g/kg).* indicates a significant difference between the control and GLP-1R kd groups (P < 0.05)

DISCUSSION

It has long been hypothesized that VAN could control eating behavior by serving as a key

mediator of nutritional cues from the intestine to the brain (6,7). Vagal lesioning methods

(including SDA, the most specific method for the disconnection of subdiaphragmatic vagal

afferents (15)) provided initial evidence for the role of the VAN in mediating peripheral

exogenous GLP-1 effects on food intake and glycemia (9,16). These methods, however,

resulted in a complete impairment of VAN signaling and function and they did not specifically

test the role of VAN in the effects of endogenous GLP-1. Therefore, the role of VAN GLP-1R

signaling in mediating the effects of endogenous GLP-1 on energy homeostasis has been

difficult to elucidate. Lately, vagal-specific genetic deletions of receptors involved in nutrient

sensing have been attempted in mouse models using paired-like homeobox 2b (Phox2b) or

Nav1.8 genes, whose promoters drive cre-recombinase expression (17-19). The cre


105

expression in these mouse models is, however, not limited to the VAN and extends to the

spinal afferents and brainstem (20,21). Moreover, for a tightly controlled system such as

eating behavior, gene deletion approaches are suspected to yield compensatory

mechanisms during development to maintain overall energy balance (18,22,23). To

overcome these obstacles, we used the bilateral delivery of a shRNA expressing-viral vector

into the NG to accomplish an inducible molecular manipulation of VAN function in adult rats.

To our knowledge, NG injection has so far been limited to nonsurvival administration of

compounds for electrophysiological recordings (24) and, more recently, to the unilateral

delivery of viral tracers and optogenetics-related tools (25). Here we demonstrate that

bilateral NG injection of a viral-mediated shRNA yields a specific and long-lasting reduction

of GLP-1R expression in the VAN.

When fed ad libitum, body weights of control and GLP-1R kd rats were similar over the entire

course of the experiment. Also, daily food intake was unchanged which, together with the

unchanged energy expenditure, demonstrates that a reduction of GLP-1R expression in VAN

does not chronically disturb energy balance. Overnight fasting and refeeding, however, led to

a much larger food intake and peak of energy expenditure in the GLP-1R kd rats,

presumably due to an increased meal-induced thermogenesis. A major caveat of the viral

mediated GLP-1R kd approach is that the reduction of GLP-1R expression in the VAN is

partial. Therefore, it cannot be excluded that the remaining expression of GLP-1R in the VAN

accounted for the absence of chronic changes in energy homeostasis. Nevertheless, this

negative phenotype in the body weight and daily food intake in our kd model is in accordance

with the GLP-1R kd∆Phox2b mouse model (17).

Previous data using surgical lesions concluded that vagal afferents are needed for the full

expression of IP-injected GLP-1 or Ex-4 effects on food intake (8,9,16,26). Considering the

fact that GLP-1 is cleared from plasma within minutes (27,28), these results suggested that

endogenous GLP-1 released from intestinal L-cells activates GLP-1R located on intestinal

VAN in a paracrine-like fashion. Here, we present strong evidence that GLP-1R in the VAN in


106

fact mediate the satiating and gastric emptying inhibiting effects of endogenous GLP-1, as

well as IP-infused GLP-1 and Ex-4.

Our data specifically confirm the importance of VAN GLP-1R signaling in the short-term

control of eating by endogenous GLP-1. GLP-1R kd specifically delayed nocturnal meal

termination (satiation) without affecting post-meal satiety, consistent with a paracrine effect of

endogenous GLP-1 on VAN GLP-1R. It stands to reason that the effects are mostly

nocturnal, when rats consume most of their calories, because intestinal GLP-1 is being

secreted via luminal nutrient stimulation. Moreover, recent data demonstrated a circadian

rhythm for GLP-1 (29), with a maximal GLP-1 release upon glucose stimulation before dark

onset. The effects of the GLP-1R kd may therefore be magnified during the early dark phase

when circulating GLP-1 levels are high. Finally, GLP-1R in the VAN are internalized during

fasting and translocated to the membrane in the fed state (4). The absence of an effect on

meal size and duration during the light phase may therefore be due to the internalization of

GLP-1R in the VAN when food intake is low and intermeal intervals are prolonged.

In addition, we demonstrated that VAN GLP-1R expression is necessary for the normal

gastric emptying of a meal, as hypothesized from earlier studies (27,30,31), consistent with a

paracrine effect of endogenous GLP-1 on VAN GLP-1R. Whether the GLP-1-induced

reduction in gastric emptying is a major mechanism by which endogenous GLP-1 controls

meal size is unclear. Further experiments should test if the eating-inhibitory effect of GLP-1

on meal size is still present in animals where gastric emptying is eliminated by gastric fistula.

Surprisingly, despite the consistent and lasting increase in meal size, daily food intake was

not altered by VAN GLP-1R kd, which was the result of a compensatory decrease in meal

number. Given the fact that IP administration of GLP-1 induces short-term satiation by a

reduction in meal size (8,32), it appears that the increase in meal size is the primary effect of

VAN GLP-1R kd, reflecting a specific satiating effect of endogenous GLP-1 via a paracrine

action. The decrease in meal number is likely a secondary, compensatory mechanism. It is

also possible to speculate that bigger meals in VAN GLP-1R kd rats trigger the release of


107

other eating-inhibitory hormones (e.g., CCK, PYY) whose signaling remains intact, and delay

the rise in ghrelin. Moreover, VAN GLP-1R KD may lead to compensatory changes in

neuronal activity of the dorsal vagal complex (unpublished observations), that, in turn, may

prolong the intermeal interval and hence decrease meal number. Together, this strengthens

the classical view that VAN GLP-1R mainly mediate short-term satiation and may be one of

several redundant eating control mechanisms comprising the gut-brain axis (7,9,16,33).

Activation of pancreatic GLP-1R by gut-derived endogenous GLP-1 is considered to be the

classical mechanism of GLP-1’s incretin effect. Several findings in rodents and humans,

however, suggest an additional involvement of a neural pathway in the GLP-1R-dependent

release of insulin (34-38). Our findings demonstrate that vagal GLP-1R signaling is

necessary for the normal control of glycemia after eating. Interestingly, the elevated post-

meal glycemia in GLP-1R kd rats was concomitant with lower levels of HPV insulin 15

minutes after the beginning of the meal without changes in GLP-1 or glucagon levels. These

results indicate that endogenous meal-induced GLP-1 acts on VAN GLP-1R to control post-

meal glycemia via a neural component of the incretin effect. Moreover, recent data showed

that IP infusion of atropin, a blocker of muscarinic receptors, reduces the insulin response

after intravenous co-infusion of glucose and GLP-1 (39). Together with our findings, this

supports the idea that ascending VAN and descending pancreatic efferents form a «gut-

brain-pancreas» axis mediating some of the effects of intestinal GLP-1 on insulin secretion.

Intriguingly, after an OGTT, no significant differences in blood glucose or plasma insulin were

seen between the two groups. Several differences between the use of a test meal or a

glucose bolus could explain this discrepancy. First, OGTT is thought to induce the release of

GLP-1 with a different amplitude and/or time-course than a solid meal ingested over several

minutes (40). As vagal lesion studies indicated that high levels of circulating GLP-1 can exert

a VAN-independent effect (8,41), it is plausible that higher GLP-1 levels after OGTT mask

the role of VAN GLP-1R. Second, glucose measurements from HPV blood sampling may not

reflect the incretin effect of GLP-1R signaling in the liver. The contribution of the liver to both


108

insulin-dependent and insulin-independent mechanisms is indeed receiving increasing

support (42-44) and it is consequently possible that HPV blood glucose and insulin levels

differ from systemic blood values. Finally, the stress caused by the OGTT procedure (gavage

and tail vein sampling) may have masked the difference in glucose levels between GLP-1R

kd and controls. Corticosterone has a powerful effect on glucose levels because it inhibits

insulin secretion and increases hepatic gluconeogenesis. Therefore, a stress-free HPV

sampling (voluntary meal followed by unrestrained blood sampling through HPV catheter)

may be a more accurate way to differentiate VAN GLP-1R kd effects on glucose/hormonal

changes than tail vein sampling.

Together, our findings demonstrate a crucial role for a vagal pathway in the maintenance of

normal eating behavior and post-prandial glycemia by endogenous GLP-1. Recent studies,

however, have shown that GLP-1R agonists such as liraglutide, exert their body weight and

glucose-lowering effects independent of vagal afferents (45) and do not require VAN GLP-1R

(17). Most likely, GLP-1R agonists do not access VAN GLP-1R when administered

subcutaneously. Instead, GLP-1R agonist effects on body weight may be mediated by the

activation of central GLP-1R (45). Based on our data, it is, however, possible to consider the

vagus nerve as a target organ to modulate satiation and glycemia.


109

AUTHORS CONTRIBUTIONS

Krieger J.-P. participated in the study conception and design, conducted the experiments,

analyzed the data and wrote the manuscript.

Arnold M. set up and performed the nodose ganglion injections.

Grossi Pettersen K. participated in the experiments and data interpretation.

Langhans W. participated in the study conception and design, participated in data

interpretation and edited/reviewed the manuscript.

Lee S.J. participated in the study conception and design, participated in the experiments/data

intepretation and edited/reviewed the manuscript.

ACKNOWLEDGEMENTS

The authors thank Dr M. Hayes (University of Pennsylvania), Dr R. Burcelin (INSERM

Toulouse) and Dr T. Lutz (University of Zürich) for their scientific advice while preparing this

study. R. Clara, S. Fedele, N. Jejelava, S. Kaufman, R. Kästli, M. Klarer and M. Labouesse

(ETH Zürich) are acknowledged for their precious help during animal experiments and data

analysis. We thank Dr C. Boyle (University of Zürich) for support and advice related to

indirect calorimetry measurements.


110

REFERENCES

1. Vilsbøll T, Christensen M, Junker AE, Knop FK, Gluud LL. Effects of glucagon-like peptide-1 receptor agonists on weight loss: systematic review and meta-analyses of randomised controlled trials. BMJ. 2012; 344:d7771.

2. FDA approves weight-management drug Saxenda [article online], 2014. Available from: http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm427913.html. Accessed 10 June 2015.

3. Nakagawa A, Satake H, Nakabayashi H, Nishizawa M, Furuya K, Nakano S, Kigoshi T, Nakayama K, Uchida K. Receptor gene expression of glucagon-like peptide-1, but not glucose-dependent insulinotropic polypeptide, in rat nodose ganglion cells. Auton Neurosci. 2004; 110:36-43.

4. Ronveaux CC, de Lartigue G, Raybould HE. Ability of GLP-1 to decrease food intake is dependent on nutritional status. Physiol Behav. 2014; 135:222–229.

5. Berthoud HR, Neuhuber WL. Functional and chemical anatomy of the afferent vagal system. Auton Neurosci. 2000; 85:1–17.

6. Ronveaux CC, Tomé D, Raybould HE. Glucagon-like peptide 1 interacts with ghrelin and leptin to regulate glucose metabolism and food intake through vagal afferent neuron signaling. J Nutr. 2015; 145:672–680.

7. Krieger JP, Langhans W, Lee SJ. Vagal mediation of GLP-1's effects on food intake and glycemia. Physiol Behav. 2015; 152(Pt B):372-80.

8. Ruttimann EB, Arnold M, Hillebrand JJ, Geary N, Langhans W. Intrameal hepatic portal and intraperitoneal infusions of glucagon-like peptide-1 deduce spontaneous meal size in the rat via different mechanisms. Endocrinology. 2009; 150:1174–1181.

9. Labouesse MA, Stadlbauer U, Weber E, Arnold M, Langhans W, Pacheco-López G. Vagal afferents mediate early satiation and prevent flavour avoidance learning in response to intraperitoneally infused exendin-4. J Neuroendocrinol. 2012; 24:1505-1516.

10. Hayes MR, Kanoski SE, De Jonghe BC, Leichner TM, Alhadeff AL, Fortin SM, Arnold M, Langhans W, Grill HJ. The common hepatic branch of the vagus is not required to mediate the glycemic and food intake suppressive effects of glucagon-like-peptide-1. Am J Physiol Regul Integr Comp Physiol. 2011; 301:R1479–1485.

11. Surina-Baumgartner DM, Langhans W, Geary N. Hepatic portal insulin antibody infusion increases, but insulin does not alter, spontaneous meal size in rats. Am J Physiol Regul Integr Comp Physiol. 1995; 269:R978–982.

12. Karimian Azari E, Leitner C, Jaggi T, Langhans W, Mansouri A. Possible role of intestinal fatty acid oxidation in the eating-inhibitory effect of the PPAR-α agonist Wy-14643 in high-fat diet fed rats. PLoS One. 2013; 8:e74869.

13. Klarer M, Arnold M, Günther L, Winter C, Langhans W, Meyer U. Gut vagal afferents differentially modulate innate anxiety and learned fear. J Neurosci. 2014; 34:7067–7076.

14. Arnold M, Thurnherr A, Dai Y, Graber M, Pacheco-López G, Langhans W.


111

Intraperitoneal (IP) glucagon-like peptide-1 (GLP-1) injections and meals in rats increase intestinal lymphatic GLP-1 similarly. (Abstract) Appetite. 2012; 59:e1–e61.

15. Norgren R, Smith GP. A method for selective section of vagal afferent or efferent axons in the rat. Am J Physiol Regul Integr Comp Physiol. 1994; 267:R1136–1141.

16. Kanoski SE, Fortin SM, Arnold M, Grill HJ, Hayes MR. Peripheral and central GLP-1 receptor populations mediate the anorectic effects of peripherally administered GLP-1 receptor agonists, liraglutide and exendin-4. Endocrinology. 2011; 152:3103–3112.

17. Sisley S, Gutierrez-Aguilar R, Scott M, D'Alessio DA, Sandoval DA, Seeley RJ. Neuronal GLP1R mediates liraglutide’s anorectic but not glucose-lowering effect. J Clin Invest. 2014; 124:2456-2463.

18. de Lartigue G, Ronveaux CC, Raybould HE. Deletion of leptin signaling in vagal afferent neurons results in hyperphagia and obesity. Mol Metab. 2014; 3:595–607.

19. Liu C, Bookout AL, Lee S, Sun K, Jia L, Lee C, Udit S, Deng Y, Scherer PE, Mangelsdorf DJ, Gautron L, Elmquist JK. PPARγ in vagal neurons regulates high-fat diet induced thermogenesis. Cell Metab. 2014; 19:722–30.

20. Scott MM, Williams KW, Rossi J, Lee CE, Elmquist JK. Leptin receptor expression in hindbrain Glp-1 neurons regulates food intake and energy balance in mice. J Clin Invest. 2011; 121:2413–2421.

21. Stirling LC, Forlani G, Baker MD, Wood JN, Matthews EA, Dickenson AH, Nassar MA. Nociceptor-specific gene deletion using heterozygous Nav1.8-Cre recombinase mice. Pain. 2005; 113:27–36.

22. Erickson JC, Clegg KE, Palmiter RD. Sensitivity to leptin and susceptibility to seizures of mice lacking neuropeptide Y. Nature. 1996; 381:415–421.

23. Qian S, Chen H, Weingarth D, Trumbauer ME, Novi DE, Guan X, Yu H, Shen Z, Feng Y, Frazier E, Chen A, Camacho RE, Shearman LP, Gopal-Truter S, MacNeil DJ, Van der Ploeg LHT, Marsh DJ. Neither agouti-related protein nor neuropeptide Y is critically required for the regulation of energy homeostasis in mice. Mol Cell Biol. 2002; 22:5027–5035.

24. Calik MW, Radulovacki M, Carley DW. A method of nodose ganglia injection in Sprague-Dawley rat. J Vis Exp. 2014; 93:e52233.

25. Chang RB, Strochlic DE, Williams EK, Umans BD, Liberles SD. Vagal sensory neuron subtypes that differentially control breathing. Cell. 2015; 161:622–633.

26. Abbott CR, Monteiro M, Small CJ, Sajedi A, Smith KL, Parkinson JRC, Ghatei MA, Bloom SR. The inhibitory effects of peripheral administration of peptide YY3–36 and glucagon-like peptide-1 on food intake are attenuated by ablation of the vagal–brainstem–hypothalamic pathway. Brain Res. 2005; 1044:127–131.

27. Schjoldager BT, Mortensen PE, Christiansen J, Orskov C, Holst JJ. GLP-1 (glucagon-like peptide 1) and truncated GLP-1, fragments of human proglucagon, inhibit gastric acid secretion in humans. Dig Dis Sci. 1989; 34:703–708.

28. Deacon CF, Pridal L, Klarskov L, Olesen M, Holst JJ. Glucagon-like peptide 1 undergoes differential tissue-specific metabolism in the anesthetized pig. Am J


112

Physiol. 1996; 271:E458–464.

29. Gil-Lozano M, Mingomataj EL, Wu WK, Ridout SA, Brubaker PL. Circadian secretion of the intestinal hormone GLP-1 by the rodent L cell. Diabetes. 2014; 63:3674-3685.

30. Anvari M, Paterson CA, Daniel EE, McDonald TJ. Effects of GLP-1 on gastric emptying, antropyloric motility, and transpyloric flow in response to a nonnutrient liquid. Dig Dis Sci. 1998; 43:1133–1140.

31. Imeryüz NI, Yegen BC, Bozkhurt A, Coskun T, Villanueva-Penacarrillo ML, Ulusoy NB. Glucagon-like peptide-1 inhibits gastric emptying via vagal afferent-mediated central mechanisms. Am J Physiol Gastrointest Liver Physiol. 1997; 273:G920-927.

32. Williams DL, Baskin DG, Schwartz MW. Evidence that intestinal glucagon-like peptide-1 plays a physiological role in satiety. Endocrinology. 2009; 150:1680–1687.

33. Lenard NR, Berthoud HR. Central and peripheral regulation of food intake and physical activity: pathways and genes. Obesity. 2008; 16:S11–22.

34. Balkan B, Li X. Portal GLP-1 administration in rats augments the insulin response to glucose via neuronal mechanisms. Am J Physiol Regul Integr Comp Physiol. 2000; 279:R1449–1154.

35. Ahrén B. Sensory nerves contribute to insulin secretion by glucagon-like peptide-1 in mice. Am J Physiol Regul Integr Comp Physiol. 2004; 286:R269–272.

36. Nishizawa M, Nakabayashi H, Uehara K, Nakagawa A, Uchida K, Koya D. Intraportal GLP-1 stimulates insulin secretion predominantly through the hepatoportal-pancreatic vagal reflex pathways. Am J Physiol Endocrinol Metab. 2013; 305:E376–387.

37. Fujiwara K, Gotoh K, Chiba S, Masaki T, Katsuragi I, Kakuma T, Yoshimatsu H. Intraportal administration of DPP-IV inhibitor regulates insulin secretion and food intake mediated by the hepatic vagal afferent nerve in rats. J Neurochem. 2012; 121:66–76.

38. Vahl TP, Tauchi M, Durler TS, Elfers EE, Fernandes TM, Bitner RD, Ellis KS, Woods SC, Seeley RJ, Herman JP, D'Alessio DA. Glucagon-like peptide-1 (GLP-1) receptors expressed on nerve terminals in the portal vein mediate the effects of endogenous GLP-1 on glucose tolerance in rats. Endocrinology. 2007; 148:4965–4973.

39. Ahlkvist L, Ahrén B. Inhibiting the cholinergic effector system of the vagus nerve reduces the insulinotropic effects of both GLP-1 and GIP (Abstract). Diabetologia. 2014; 57:1–564.

40. Calanna S, Christensen M, Holst JJ, Laferrère B, Gluud LL, Vilsbøll T, Knop FK. Secretion of glucagon-like peptide-1 in patients with type 2 diabetes mellitus: systematic review and meta-analyses of clinical studies. Diabetologia. 2013; 56:965–972.

41. Zhang J, Ritter RC. Circulating GLP-1 and CCK-8 reduce food intake by capsaicin-insensitive, nonvagal mechanisms. Am J Physiol Regul Integr Comp Physiol. 2012; 302:R264-273.

42. Schwartz MW, Seeley RJ, Tschöp MH, Woods SC, Morton GJ, Myers MG, D'Alessio DA. Cooperation between brain and islet in glucose homeostasis and diabetes.


113

Nature. 2013; 503:59–66.

43. Scarlett JM, Schwartz M. Gut-brain mechanisms controlling glucose homeostasis. F1000Prime Rep. 2015; 7:12.

44. Ahrén B. Hepato-incretin function of GLP-1: novel concept and target in type 1 diabetes. Diabetes. 2015; 64:715–717.


Chapter 4: CART mediates the effects of GLP-1 in vagal afferent neurons

115

CHAPTER 4:

THE COCAINE- AND AMPHETAMINE-REGULATED TRANSCRIPT

MEDIATES THE SATIATING EFFECTS OF GLP-1 IN VAGAL

AFFERENT NEURONS

AUTHORS

Krieger JP1, Arnold M1, Langhans W1, de Lartigue G2, Lee SJ1

1 Physiology and Behavior Laboratory, ETH Zürich, Switzerland

2 John B. Pierce Laboratory, New Haven, USA

STATUS

In preparation


116

ABSTRACT

GLP-1 is an incretin and satiation hormone produced in the intestine in response to eating.

Using a rat model with a specific knockdown (kd) of GLP-1 receptors (GLP-1R) in vagal

afferents neurons (VAN), we previously demonstrated that VAN GLP-1R are required for

endogenous GLP-1-induced satiation. The molecular mechanism by which VAN GLP-1R

activation decreases food intake remains unknown. We screened the expression of

neuropeptides involved in the control of food intake in a primary culture of VAN and found

that the Cocaine- and Amphetamine-Regulated Transcript (CART) mRNA and protein levels

were upregulated in response to GLP-1 stimulation in a GLP-1R dependent manner. CART

was co-localized in GLP-1R expressing cells in the rat nodose ganglia (NG), and

intraperitoneal (IP) GLP-1 (10 nmol/kg) was sufficient to induce CART mRNA and protein

expression. We therefore hypothesized that CART mediates the eating-inhibitory effects of

vagal GLP-1R activation and tested this using two rat models of CART inhibition. First, we

reduced VAN CART expression using a lentiviral-mediated CART kd. Second, we blocked

CART signaling in the nucleus tractus solitarii (NTS) by administering a CART antibody into

the NTS. In both models, CART inhibition abolished IP GLP-1-induced (10 nmol/kg) satiation

and attenuated GLP-1-induced gastric emptying. We conclude that GLP-1 induces CART

transcription and synthesis in VAN and that CART released from VAN into the NTS mediates

the physiological effects of GLP-1 on satiation and gastric emptying.


117

INTRODUCTION

Vagal afferent neurons (VAN) are key components of the gut-brain axis, conveying

mechanical and chemical signals from the gastrointestinal tract to the brain. They participate

in the normal control of food intake and body weight (see (1) for review) and can respond to

nutritional challenges (2). Recently, a series of studies using viral or transgenic approaches

began to unravel the molecular mechanisms by which VAN can modulate energy balance (3-

7). Specifically, we demonstrated that intestinal glucagon-like peptide-1 (GLP-1) acts in a

paracrine fashion on VAN GLP-1 receptors (GLP-1R) to induce satiation and to slow gastric

emptying (7). The vagal and central mechanisms by which VAN GLP-1R activation

decreases food intake remain, however, unclear. Substantial evidence indicates that

glutamate is the main neurotransmitter involved in the transmission of sensory information

from VAN to the nucleus tractus solitarii (NTS, see (8) for review). Yet VAN also express a

broad array of peptide-based neurotransmitters, such as Cocaine- and Amphetamine-

Regulated Transcript (CART, (9)), Melanin-Concentrating Hormone (MCH, (10)), substance

P (subP, (11,12)) and neurokinin A (11,13). Their roles in the vagal control of food intake is

however largely unknown (8). Among these neuropeptides, CART has received attention due

to its broad role in the control of food intake. Indeed, since its original discovery (14), CART

has been detected in many brain areas involved in the control of eating (see (15) for review),

and intracerebroventricular (ICV) injections of the CART peptide or the CART antibody have

highlighted an anorexigenic property of CART in the CNS (16-18). CART is also expressed in

nearly half of the neurons of the medial and caudal nodose ganglion (NG) (9,19,20) and

vagal CART positive fibers terminate in the commissural and medial NTS (16,19). Several

studies demonstrated that CART is a mediator of the eating-inhibitory effects of

cholecystokinin (CCK) in VAN (9,20-22). Indeed, CART is colocalized with CCKa receptors in

the NG (9), and IP administration of CCK increased CART protein expression in the NG (20).

Moreover, in cultured VAN, administration of CCK induced CART expression and release


118

and this effect was modulated by leptin and ghrelin (21,23). Finally, silencing CART in vivo in

VAN blocked the effects of IP-infused CCK and leptin on short-term food intake (22).

In this study, we hypothesized that vagally expressed neuropeptides play an essential role in

the eating inhibition induced by VAN GLP-1R activation. We initially screened a set of

neuropeptides for their transcriptional response to GLP-1 stimulation in primary rat VAN

cultures and found that CART was upregulated by GLP-1 via the GLP-1R. Consequently, we

further tested the idea that CART in VAN is a necessary mediator of GLP-1’s effects on food

intake and gastric emptying and present here in vitro and in vivo evidence that support this

hypothesis.

METHODS

Primary culture of vagal afferent neurons

Primary cultures of rat VAN were carried out as previously described (20). Immediately after

collection, NG were digested for 2 h with collagenase type Ia (Roche) (2 mg/mL) in 37°C

HBSS (Sigma-Aldrich), dissociated with coated glass pipettes (Sigmacoat, Sigma-Aldrich),

and plated in poly-L-lysin-coated chamber slides (Millicell EZ Slide, Millipore). Cultures were

grown for 72 h in 4.5 g/L DMEM with 10% bovine serum. For experiments, VAN cultures

were transferred to a serum-free medium (20% glucose in cell-culture grade water) for 2 h

and incubated for 1 h with GLP-1 (10 nM final concentration, Bachem) or PBS. The GLP-1R

antagonist exendin-(9-39) (Ex-9, 100 nM final concentration, Bachem) was added 30 min

prior to GLP-1 or PBS. VAN mRNA was extracted using Trizol according to the

manufacturer’s protocol (Life Technologies) and levels of the calcitonin (calc), neuropeptide y

(npy), pro-melanin-concentrating hormone (pmch), substance P (subP) and CART (cart)

genes were measured using RT-qPCR. For immunocytochemistry, VAN cultures were fixed

for 30 min with 4% PFA and stained for CART (Rabbit anti-CART(55-102), H-003-62,

Phoenix Pharmaceuticals, 1:1K; Cy-3-conjugated donkey anti-rabbit, 711-165-152, Jackson

ImmunoResearch, 1:400).


119

Immunohistochemistry in the nodose ganglia

NG were collected and immediately fixed for 2 h in 4% PFA, and immersed in a 25% sucrose

PBS solution overnight. NG were sectioned at 10-14 µm in a cryostat (CM 1950, Leica) and

placed on polysine-coated microscope slides (Menzel, ThermoScientific). The co-

immunostaining for CART and GLP-1R was performed in a three-step procedure. First, the

GLP-1R staining was carried out (Rabbit anti-GLP-1R, ab39072, Abcam, 1:1K; Alexa 488-

conjugated donkey anti-rabbit, A21206, Invitrogen, 1:400). Second, the GLP-1R detection

complex was protected by successive incubation with rabbit serum followed by F(Ab)

fragment (Donkey anti-rabbit F(ab) fragment, 711-007-003, Jackson ImmunoResearch, 40

µg/mL). Third, the CART staining was carried out as described above. To test the induction

of CART by vagal GLP-1R activation, NG were collected from fasted rats (24 h fasting) and

refed rats (24 h fasting + 4 h ad libitum refeeding) or after an IP GLP-1 injection (24 h fasting

+ 1 h after injection of IP GLP-1 10 nmol/kg, Bachem). NG were fixed and stained for CART

as described above.

CART knockdown in the nodose ganglion

A shRNA sequence targeting the rat CART mRNA or a non-target shRNA sequence were

driven by a U6 promoter in a pLKO.1-puro vectors expressing turboGFP (Sigma-Aldrich).

The efficiency of the CART-targeting shRNA construct was verified in vitro in GH3 cells (CCL

82.1, ATCC) after sorting turboGFP-expressing cells (FACS Aria III FCF, Flow Cytometry

Facility, University of Zurich). CART-targeting (LV-shCART) or control (LV-shCTL) lentiviral

particles were produced in HEK 293T cells using the pMD2.G and psPAX2 plasmids (gifts

from Dider Trono, Ecole Polytechnique Fédérale de Lausanne; cat. no. 12259 and 12260)

and concentrated to 1010 particles/mL using 8% PEG6000 (Millipore) in 0.3M NaCl and

resuspended in PBS. Rats (290-340 g on surgery day) were anesthetized by an IP injection

of ketamine (88 mg/kg, Ketalar, Zurich Cantonal Pharmacy) and xylazin (5 mg/kg, Rompun

2%, Zurich Cantonal Pharmacy) and NG were exposed. A custom-made glass capillary (50

µm tip) was used to administer 1.5 µL of viral solution (LV-shCTL or LV-shCART) into each


120

NG with a Picospritzer III injector (Parker Hannifin). To ensure expression of the viral

constructs, animals were allowed to recover for 20 days after surgery. Rats were handled

and habituated to IP injections of NaCl prior to the experiment. Animals had ad libitum

access to water and chow food (3433, Kliba-Nafag) unless otherwise noted.

NTS cannulation

Rats (290-340 g on surgery day) were anesthetized by an IP injection of ketamine and

ketamine as described above and placed in a stereotaxic apparatus. Cannulas (PlasticOne)

were implanted in the NTS (stereotaxic coordinates: 1.0 mm posterior to occipital crest, 0.7

mm lateral to midline, 5.9 mm below skull surface). Animals were allowed to recover for 10

days and habituated to the cannula manipulation and to the NTS injection procedure before

experiments. Animals had ad libitum access to water and chow (3433, Kliba-Nafag) unless

otherwise noted. Cannula placement was verified by dye injection (500 nL) and post-mortem

anatomical visualization of the solution spread.

Effect of IP GLP-1 on food intake in the CART inhib ition models

The effect of IP GLP-1 on short-term food intake was assessed in a similar manner in two

experiments, in CART knockdown (kd) rats and in wild-type rats receiving a CART antibody

injection. Animals were habituated to acrylic cages in which powdered chow (3436, Kliba-

Nafag) was available through a niche in feeding cups placed on scales (XS4001S, Mettler-

Toledo) for continuous measurement of meal patterns (see (24) for detailed description).

After 16 h of food deprivation, animals were allowed access to a pre-meal composed of 3 g

powdered chow and 3 mL water 30 min before dark onset. A pre-meal allows GLP-1R

trafficking to the VAN membrane (25). Only animals that finished the pre-meal within 10 min

were included in the experiments. In the CART kd experiment, rats were IP injected with

GLP-1 (10 nmol/kg) or a vehicle solution 5 min before dark onset. In the CART antibody

experiment, rats were first injected with a CART antibody solution (1:500 in PBS, 500 nL over

30 sec) or a vehicle solution using an electronically controlled pump (UltraMicroPump UMP3

with SYS-Micro4 controller, World Precision Instruments) 15 min prior to dark onset. Rats


121

subsequently received an IP injection of GLP-1 (10 nmol/kg) or vehicle solution 5 min before

dark onset. Food was returned at dark onset and food intake was recorded.

Effect of IP GLP-1 on gastric emptying in the CART inhibition models

One week prior to the experiment, rats were habituated to receiving test meals, to the

restricted feeding schedule (test meal at dark onset, ad libitum food access from 3 to 8 h

after dark onset, food deprivation otherwise) and to the restraint for tail vein blood sampling.

On experimental days, all rats received a 4 g chow meal containing 1% (w/w) paracetamol

(4-acetamido-phenol, Sigma-Aldrich) and 0.25% (w/w) saccharin (Sigma-Aldrich). Injections

of IP GLP-1 and/or NTS CART antibody solution were done in identical conditions as during

food intake measurements (described above). In both CART inhibition models, drug

administration was randomized in a crossover design. Baseline tail vein blood was taken 30

min prior to test meal onset and post-meal blood was collected according to the scheduled

timepoints. Blood was immediately placed on ice, centrifuged (10 min at 8370 g) and stored

at -80°C for further processing. Paracetamol concentrations were measured with a

commercial kit (Cambridge Life Sciences K8002) adapted for small plasma volumes and

reading in a 96-well plate.

Figures and statistical analysis

Data normality was verified using the Shapiro-Wilk (when n ≥ 7) and the Kolmogorov-

Smirnov (when 5 ≤ n ≤ 6) tests and homoscedasticity was checked by visualizing the

distribution of residuals. Non-parametric tests were used otherwise. Tests used and P-values

are given in the figure legends. Data are presented as means ± SEM. P-values < 0.05 were

considered significant. All graphs were generated using GraphPad Prism (version 6.05 for

Windows).


122

RESULTS

GLP-1 increased CART expression in primary vagal af ferent neurons

In a primary culture of VAN, we first tested whether GLP-1 modulates the expression profile

of known neuropeptides involved in eating regulation. Among the peptides screened, we

found that only CART was increased by 2 h incubation with GLP-1 (10 nM), whereas others

remained unchanged (Fig. 1A). Using immunocytochemistry, we measured an increase in

the number of CART positive cells after GLP-1 incubation (Fig. 1C, D). We further tested

whether this effect was dependent on the presence of GLP-1R. Pre-incubation of the primary

VAN culture with the GLP-1R antagonist Ex-9 (100 nM) prevented the GLP-1-induced

increase in CART gene expression (Fig. 1B) and number of CART-expressing neurons (Fig.

1C, D). Together, these results show that GLP-1 can induce CART mRNA and protein

expression in a subpopulation of VAN via the GLP-1R.

Figure 1: GLP-1 (10 nM) induced CART expression in primary VAN via the GLP-1R (A) Relative mRNA expression of the calcitonin (calc), neuropeptide y (npy), pro-melanin-concentrating hormone (pmch), substance P (subP) and CART (cart) genes in primary VA incubated with 10 nM GLP-1 or the vehicle solution for 1 h (n = 5; Student t-test, P < 0.05 for cart). (B) Relative mRNA expression of the cart gene in primary VAN incubated with vehicle, 100 nM of Ex-9, 10 nM


123

GLP-1 or co-incubated with GLP-1 and Ex-9 (n = 5; Two-way ANOVA GLP-1 x antagonist; GLP-1 effect P = 0.1, antagonist effect P = 0.1, interaction P < 0.05; * indicates significant difference from control in Bonferroni post hoc comparisons). (C) Representative pictures of cultured VA incubated with vehicle (top left), 100 nM of Ex-9 (top right), 10 nM GLP-1 (bottom left) or co-incubated with GLP-1 and Ex-9 (bottom right). (D) Percentage of CART positive cells among all cells in a primary culture of VA after incubation with vehicle, 100 nM of Ex-9, 10 nM GLP-1 or co-incubated with GLP-1 and Ex-9 (n = 5; Two-way ANOVA GLP-1 x antagonist; GLP-1 effect P < 0.05, antagonist effect P < 0.05, interaction P < 0.05; * indicates significant difference from control in Bonferroni post-hoc comparisons). CART is expressed in GLP-1R-expressing vagal affere nt neurons and is induced by IP

administration of GLP-1

The data above suggest that CART expression is induced in the VAN expressing GLP-1R.

Colocalization of CART and GLP-1R in VA is a pre-requisite for CART to mediate the effects

of vagal GLP-1R activation. Thus, we next visualized CART protein expression in the GLP-

1R expressing cells of the rat NG using double-immunostaining. CART was visible in a

perinuclear localization in nearly half of the VA and GLP-1R were mainly expressed at the

cell membrane (fed state, Fig. 2A, B, F and G). Due to the weak labeling of the GLP-1R

antibody, we chose to determine whether the subpopulation of neurons clearly labeled as

GLP-1R positive were also labeled as CART positive (CART+), and saw a strong overlap

between the expression of the two proteins (Fig. 2A-D). CART expression was, however, not

limited to GLP-1R expressing neurons. As previously demonstrated, CART expression

depended on the feeding status, i.e., CART staining was not visible in the NG of fasted rats,

but appeared when animals were refed ad libitum for 4 h (Fig. 2E and G). An IP injection of

GLP-1 (10 nmol/kg) was sufficient to induce CART mRNA and protein expression in the NG

(Fig. 2F, G). Together, these data raise the possibility that endogenous intestinal GLP-1

induces CART expression in the NG via activation of VAN GLP-1R. We consequently further

tested the functional role of CART in the physiological effects of endogenous GLP-1 on food

intake and gastric emptying.


124

Figure 2: CART and GLP-1R colocalized, and IP GLP-1 is sufficient to induce CART expression in the nodose ganglion of fasted rats (A) GLP-1R staining, (B) CART staining and (C) merged picture of a rat nodose ganglion (ad libitum fed). (D) Percentage of CART-positive cells among GLP-1R-positive cells in a rat nodose ganglion (n = 4). (E) CART staining in the nodose ganglion of a rat fasted for 24 h and injected IP with a vehicle solution, (F) injected IP with GLP-1 (10 nmol/kg) or (G) refed ad libitum for 4 h. (H) Relative mRNA expression of the cart gene in the nodose ganglia (n = 3/5/2; Kriskal-Wallis test, P < 0.05; * indicates significant difference with the fasting condition after Dunn’s comparisons).

Full CART expression in vagal afferents is necessar y for the IP GLP-1-induced

reduction in food intake and gastric emptying

To test whether CART is a necessary player in the vagal mediation of GLP-1’s effects, we

specifically knocked down the CART mRNA transcript in VAN. For this, we used our

previously described lentivirus-based shRNA delivery via bilateral NG injection. Using this

approach, CART expression was reduced by 35 % (Fig. 3A) in the NG of rats injected with

LV-shCART compared to control injected rats. In this CART kd model, the eating-inhibition

induced by the IP injection of GLP-1 (10 nmol/kg) was significantly attenuated (Fig. 3B).

Moreover, we previously described that VAN GLP-1R activation is associated with a

reduction in gastric emptying that parallels the eating inhibition. In VAN CART kd rats, meal-

induced gastric emptying was accelerated, as shown by a faster appearance of paracetamol

in the plasma. The gastric emptying inhibition induced by IP GLP-1, however, was abolished

by the CART kd (Fig. 3C).


125

Figure 3: CART kd in the NG attenuated the inhibiti on of eating and gastric emptying induced by IP GLP-1

(A) Relative mRNA expression of the cart gene in the nodose ganglia of rats injected with an anti-CART shRNA-carrying lentivirus, or a control lentivirus (n = 8/7; Student t-test, P < 0.05). (B) 30 min food intake (n = 8/8/7/7; Two-way mixed ANOVA knockdown x drug ; Knockdown effect NS, drug effect P<0.05, interaction P=0.01; different letters indicate a significant difference by Bonferroni-adjusted post-hoc comparisons) and (C) plasma paracetamol concentrations after IP injection of GLP-1 (10 nmol/kg) or vehicle in control and CART KD rats (n = 8/8/7/7 ; Mixed MANOVA knockdown x drug x time; knockdown effect P<0.01, drug effect NS, time p<0.01, interaction knockdown * drug NS)

Together, these results demonstrate that the food intake reduction and the gastric emptying

inhibition following IP GLP-1 injection require unaltered VAN CART expression, suggesting

that CART is a downstream mediator of VAN GLP-1R signaling.

NTS CART antibody attenuated IP GLP-1-induced reduc tion in food intake and gastric

emptying

A possible role for CART in the eating-inhibitory effect of GLP-1 may reside in the release of

CART vesicles in the NTS. Thus, to test whether the presence of CART in the NTS is

necessary for GLP-1 to inhibit eating and gastric emptying, we infused a CART antibody into

the NTS prior to GLP-1 IP administration. The pre-injection of a CART antibody into the NTS

blocked the effects of IP GLP-1 on food intake (Fig. 4A) and gastric emptying (Fig. 4B).

These results are consistent with the idea that the release of CART in the NTS from VAN is

necessary for the eating-inhibition induced by GLP-1.


126

Figure 4: CART antibody in the NTS blocked the inhi bition of eating and gastric emptying by IP GLP-1

(A) 30 min food intake (n = 10 ; Two-way ANOVA antibody x ip drug; antibody effect P < 0.05, IP drug effect P = 0.06, interaction P < 0 .05 ; different letters indicate a significant difference among groups by Bonferroni-adjusted post-hoc comparisons) and (B) plasma paracetamol concentrations after IP injection of GLP-1 (10 nmol/kg), NTS pre-injection with the CART Ab and respective control vehicle injections (n = 5/6/9/8 ; MANOVA IP drug x NTS Ab x time; IP drug effect P < 0.05, NTS Ab effect P = 0.12, time effect P < 0.01, interaction IP drug * NTS Ab P = 0.1).

DISCUSSION

We here report that GLP-1 induced CART mRNA and protein expression via the GLP-1R in

primary VAN. Moreover, in vivo, CART co-localized with GLP-1R, and IP GLP-1 was

sufficient to induce CART expression in the NG of fasted rats. We then further tested

whether CART mediates the eating-inhibitory effect of VAN GLP-1R activation. First, we

knocked down CART expression in VAN using a lentiviral-mediated RNA interference

approach. Second, we blocked CART signaling in the NTS by administering a CART

antibody into the NTS. In both models, CART inhibition abolished IP GLP-1-induced satiation

and gastric emptying. We conclude that GLP-1 induces CART synthesis in VAN and that

CART released from VAN into the NTS is involved in the mediation of the physiological

effects of GLP-1 on satiation and gastric emptying. Together, our data indicate that VAN

CART is a necessary downstream mediator of endogenous GLP-1’s effects.

Based on our results, it is possible to speculate that VAN GLP-1R activation is involved in a/

an increase in CART transcription, and protein expression, and/or b/ the transport of the

CART in vesicles to VAN terminals for their release into the NTS. This sequence of events is


127

further suggested by the study of CCK-induced CART release in VAN. CCK induces CART

expression as early as after 10 min incubation in primary VAN cultures (23) but the highest

expression is reached after 2 h, indicating that CCK action on CART gene transcription does

not fully explain the short term (<30min) satiating effect of CCK. Moreover, the transport of

CART from the soma to the VAN terminals is likely to be a rate-limiting step between CCK-

induced CART expression and the eating inhibition. Indeed, CART vesicles have been

reported in neurite terminals only after 85 min of incubation with CCK (in vitro) (20), and

CART positive fibers appear in the NTS within 2 h of refeeding (26). Based on all these facts,

it is likely that CCK promotes CART release by increasing VAN firing in addition to the

transcriptional regulation. Similarly, considering the short-term GLP-1 effect on food intake, it

seems more plausible that VAN GLP-1R activation simultaneously induces gene

transcription and the release of pre-existing CART vesicles into the NTS. The activation of

CART transcription may maintain and increase CART abundance for subsequent eating

bouts, rather than exert an intra-meal effect. Thus, it stands to reason that a permanent

downregulation of VAN CART (that reduces the amount of CART produced and, hence,

released), and an acute neutralization of CART in the NTS via an antibody, have similar

effects on GLP-1-induced inhibition of eating and gastric emptying. It is, however, surprising

that a relatively mild downregulation of VAN CART expression leads to a substantial

reduction in vagal GLP-1R-induced eating inhibition. This may indicate that reduced CART

transcription leads to a rapid decrease in the amount of CART peptides available for release

into the NTS, and therefore blocks the primary mechanism underlying GLP-1 induced eating

inhibiton.

Several studies indicate that CART can inhibit eating and gastric emptying via hindbrain

mechanisms. Indeed, 4th ventricular injection of the CART peptide reduced food intake

(19,27-30) and gastric emptying (29). While injection of CART in the lateral or 3rd ventricles

produced similar effects to 4th ventricular injections (27,31), cerebral aqueduct occlusion

attenuated the effects of 3rd ICV injection of CART on eating inhibition, further suggesting


128

that the inhibitory effect of CART on food intake originates within the hindbrain. In addition,

4th ventricular injection of the CART peptide induced c-Fos in the NTS but the phenotype and

projections of these CART-sensitive neurons are unknown (19).

At the neuronal and synaptic levels, the exact role and the mechanisms of CART action in

the NTS remain speculative. Zheng and colleagues (19) hypothesized that CART could be

co-released from VAN with glutamate in the NTS, thus providing a modulation of the

glutamatergic satiation signal. Rather than having an immediate effect, CART could induce

longer lasting pre- or post-synaptic changes in NTS neurons and, consequently, a longer-

term influence on the control of food intake. Indeed, structural differences between classical

neurotransmitters and neuropeptides suggest that neuropeptides can modulate synaptic

transmission rather than being a fast-acting signal: glutamate is present in small synaptic

vesicles that are released and recycled at the synapse while neuropeptides like CART are

synthesized in the NG and transported in large vesicles to the axon terminals (32); the

release of such large vesicles has been shown to occur outside of the synaptic terminals

(33), indicating a modulatory role more than the ability to transmit a signal from neuron-to-

neuron. CART has also been speculated to provide the early depolarizing event required for

NMDA receptor activation in NTS neurons, thus increasing neuron specificity and reducing

potential negative outcomes of glutamate spillover (8). Finally, meal-to-meal variations in the

release of CART in the NTS based on feeding status could provide a certain plasticity to the

vagal control of food intake that cannot be obtained with glutamatergic transmission alone

(8).

Interestingly, several reports suggest that activation of central GLP-1R is necessary for the

hypophagia induced by ICV CART. Indeed, pre-treatment (lateral ventricle) or co-infusions

(4th ventricule) of the GLP-1R antagonist attenuated the effects of ICV CART on food intake

(30,34). This indicates that the effects of 4th ICV CART require the central release of GLP-1.

We can thus speculate that one of the neuronal targets of vagal-derived CART may be the

GLP-1 producing neurons in the NTS. This idea requires further testing but it is tempting to


129

think that the release of CART in the NTS by VAN relays the signal of intestinal GLP-1

secretion to the central GLP-1-producing neurons. The lack of an identified CART receptor

(35,36), though, is a major obstacle to the identification of the central targets of CART.

Figure 4: Proposed mechanisms by which CART mediate s the satiating effects of endogenous intestinal GLP-1 via vagal afferent neurons. The activation of GLP-1R in the gut mucosa induces CART transcription (via an intracellular pathway to be identified) in VAN and activates the release of CART vesicles in the NTS. The role and neuronal targets of CART in the NTS are still unknown but it is probable that CART provides specificity and/or plasticity to the neuron-to-neuron glutamatergic transmission.

Despite the uncertain role of CART in the NTS, our data reiterate CART as an integrator

of nutrient- and energy-related signals in VAN. Indeed, as mentioned earlier, CART has been

shown to mediate the effects of CCK in VAN, and these effects are modulated by both leptin

and ghrelin. Showing that CART is also a major downstream mediator of vagal GLP-1’s

effects further enhances the potential importance of CART in the neural control of food


130

intake. Moreover, it naturally leads to the question whether CART integrates these different

signals at the cellular level. Studies investigating the effects of various combinations of GLP-

1, CCK, leptin, ghrelin or other vagally-sensed metabolic signals on CART expression and

release are needed to understand how VAN CART may compute these diverse inputs to

generate behavioral outcomes. Given its role as a mediating hub for several nutrient- and

energy-related signals, VAN CART has also been hypothesized to contribute to the

pathophysiology of obesity. Indeed, the eating-induced increase in CART levels is blocked in

diet-induced obese rats (preliminary observations reported in (2)). Moreover, unpublished

observations of our lab (Krieger, Lee, de Lartigue, 2015) indicate that the kd of CART in VAN

is sufficient to induce hyperphagia and weight gain. Further studies are required to confirm a

causal role for declining VAN CART levels in diet-induced obesity.

In sum, our data are the first demonstration to date that CART mediates the effects of VAN

GLP-1R activation on food intake and gastric emptying. Our study is therefore a further step

towards considering VAN CART as a major integrator of nutrient-derived signals along the

gut-brain axis. The possibility that VAN CART plays a role in the pathophysiology of obesity

makes it a potential therapeutic target.


131

AUTHORS CONTRIBUTIONS

Krieger J.-P. participated in the study conception and design, conducted the experiments,

analyzed the data and wrote the manuscript.

Arnold M. set up and performed the nodose ganglion injections.

Langhans W. participated in the study conception and design, participated in data

interpretation and edited/reviewed the manuscript.

De Lartigue G. participated in the study conception and design, participated in the

experiments/data intepretation and edited/reviewed the manuscript.

Lee S.J. participated in the study conception and design, participated in the experiments/data

intepretation and edited/reviewed the manuscript.

ACKNOWLEDGEMENTS

Rosmarie Clara, Shahana Fedele, Nino Jejelava, Marie Rouault for help during animal

experiments


132

REFERENCES


2. de Lartigue G, la Serre de CB, Raybould HE. Vagal afferent neurons in high fat diet-induced obesity; intestinal microflora, gut inflammation and cholecystokinin. Physiol Behav. 2011; 105(1):100–5.

3. Sisley S, Gutierrez-Aguilar R, Scott M, D'Alessio DA, Sandoval DA, Seeley RJ. Neuronal GLP1R mediates liraglutide’s anorectic but not glucose-lowering effect. J Clin Invest. 2014; 124(6):2456-63.

4. de Lartigue G, Ronveaux CC, Raybould HE. Deletion of leptin signaling in vagal afferent neurons results in hyperphagia and obesity. Mol Metab. 2014; 3(6):595–607.

5. Liu C, Bookout AL, Lee S, Sun K, Jia L, Lee C, Udit S, Deng Y ,Scherer PE, Mangelsdorf DJ, Gautron L, Elmquist JK. PPARg in Vagal Neurons Regulates High-Fat Diet Induced Thermogenesis. Cell Metab. 2014; 19(4):722–30.


7. Krieger JP, Arnold M, Pettersen KG, Lossel P, Langhans W, Lee SJ. Knockdown of GLP-1 receptors in vagal afferents affects normal food intake and glycemia. Diabetes. 2016; 65(1):34-43.

8. de Lartigue G. Putative role of neuropeptides in vagal afferent signaling. Physiol Behav. 2014; 136:155-69.

9. Broberger C, Holmberg K, Kuhar MJ, Hökfelt T. Cocaine- and amphetamine-regulated transcript in the rat vagus nerve: A putative mediator of cholecystokinin-induced satiety. Proc Natl Acad Sci USA. 1999 Nov 9;96(23):13506–11.

10. Burdyga G, Varro A, Dimaline R, Thompson DG, Dockray GJ. Feeding-dependent depression of melanin-concentrating hormone and melanin-concentrating hormone receptor-1 expression in vagal afferent neurones. Neurosci. 2006; 137(4):1405–15.

11. Helke CJ, Hill KM. Immunohistochemical study of neuropeptides in vagal and glossopharyngeal afferent neurons in the rat. Neurosci. 1988; 26(2):539–51.

12. Helke CJ, Niederer AJ. Studies on the coexistence of substance P with other putative transmitters in the nodose and petrosal ganglia. Synapse. 1990; 5(2):144–51.

13. Czyzyk-Krzeska MF, Bayliss DA, Seroogy KB, Millhorn DE. Gene expression for peptides in neurons of the petrosal and nodose ganglia in rat. Exp Brain Res. 1991; 83(2):411–8.

14. Douglass J, McKinzie AA, Couceyro P. PCR differential display identifies a rat brain mRNA that is transcriptionally regulated by cocaine and amphetamine. J Neurosci. 1995; 15(3 Pt 2):2471–81.

15. Subhedar NK, Nakhate KT, Upadhya MA, Kokare DM. CART in the brain of vertebrates: Circuits, functions and evolution. Peptides. 2014; 54:108–30.


133

16. Koylu EO, Couceyro P, Lambert PD, Kuhar MJ. Cocaine- and amphetamine-regulated transcript peptide immunohistochemical localization in the rat brain. J Comp Neurol. 391:115–32.

17. Kristensen P, Judge ME, Thim L, Ribel U, Christjansen KN, Wulff BS, Clausen JT, Jensen PB, Madsen OD, Vrang N, Larsen PJ, Hastrup S. Hypothalamic CART is a new anorectic peptide regulated by leptin. Nature. 1998; 393(6680):72–6.

18. Lambert PD, Couceyro PR, McGirr KM, Dall Vechia SE, Smith Y, Kuhar MJ. CART peptides in the central control of feeding and interactions with neuropeptide Y. Synapse. 1998; 29(4):293–8.

19. Zheng H, Patterson LM, Berthoud H-R. CART in the dorsal vagal complex: sources of immunoreactivity and effects on Fos expression and food intake. Brain Res. 2002; 957(2):298–310.

20. de Lartigue G, Dimaline R, Varro A, Dockray GJ. Cocaine- and amphetamine-regulated transcript: stimulation of expression in rat vagal afferent neurons by cholecystokinin and suppression by ghrelin. J Neurosci. 2007; 27(11):2876–82.

21. de Lartigue G, Dimaline R, Varro A, Raybould H, la Serre de CB, Dockray GJ. Cocaine- and amphetamine-regulated transcript mediates the actions of cholecystokinin on rat vagal afferent neurons. Gastroenterology. 2010; 138(4):1479–90.

22. Heldsinger A, Lu Y, Zhou SY, Wu X, Grabauskas G, Song I, Owyang C. Cocaine- and amphetamine-regulated transcript is the neurotransmitter regulating the action of cholecystokinin and leptin on short-term satiety in rats. Am J Physiol Gastrointest Liver Physiol. 2012; 303(9):G1042–51.

23. de Lartigue G, Lur G, Dimaline R, Varro A, Raybould H, Dockray GJ. EGR1 Is a Target for Cooperative Interactions between Cholecystokinin and Leptin, and Inhibition by Ghrelin, in Vagal Afferent Neurons. Endocrinology. 2010; 151(8):3589–99.

24. Karimian Azari E, Leitner C, Jaggi T, Langhans W, Mansouri A. Possible role of intestinal fatty acid oxidation in the eating-inhibitory effect of the PPAR-α agonist Wy-14643 in high-fat diet fed rats. PLoS ONE. 2013; 8(9):e74869.

25. Ronveaux CC, de Lartigue G, Raybould HE. Ability of GLP-1 to decrease food intake is dependent on nutritional status. Physiol Behav. 2014; 135:222–9.

26. de Lartigue G. Cocaine- and Amphetamine-Regulated Transcript (CART) Derived From Vagal Afferent Neurons Is Required for Satiation. Neurogastroenterol Motil. 2013; 25:1–12.

27. Aja S, Sahandy S, Ladenheim EE, Schwartz GJ, Moran TH. Intracerebroventricular CART peptide reduces food intake and alters motor behavior at a hindbrain site. Am J Physiol Regul Integr Comp Physiol. 2001; 281(6):R1862–7.

28. Aja S, Robinson BM, Mills KJ, Ladenheim EE, Moran TH. Fourth ventricular CART reduces food and water intake and produces a conditioned taste aversion in rats. Behav Neurosci. 2002; 116(5):918–21.

29. Smedh U, Moran TH. Peptides that regulate food intake: separable mechanisms for dorsal hindbrain CART peptide to inhibit gastric emptying and food intake. Am J


134

Physiol Regul Integr Comp Physiol. 2003; 284(6):R1418–26.

30. Skibicka KP, Alhadeff AL, Grill HJ. Hindbrain cocaine- and amphetamine-regulated transcript induces hypothermia mediated by GLP-1 receptors. Journal of Neuroscience. 2009; 29(21):6973–81.

31. Aja S, Schwartz GJ, Kuhar MJ, Moran TH. Intracerebroventricular CART peptide reduces rat ingestive behavior and alters licking microstructure. Am J Physiol Regul Integr Comp Physiol. 2001; 280(6):R1613–9.

32. Zhu PC, Thureson-Klein A, Klein RL. Exocytosis from large dense cored vesicles outside the active synaptic zones of terminals within the trigeminal subnucleus caudalis: a possible mechanism for neuropeptide release. Neurosci. 1986; 19(1):43–54.

33. Verhage M, McMahon HT, Ghijsen WE, Boomsma F, Scholten G, Wiegant VM, Nicholls DG. Differential release of amino acids, neuropeptides, and catecholamines from isolated nerve terminals. Neuron. 1991; 6(4):517–24.

34. Aja S, Ewing C, Lin J, Hyun J, Moran TH. Blockade of central GLP-1 receptors prevents CART-induced hypophagia and brain c-Fos expression. Peptides. 2006; 27(1):157–64.

35. Vicentic A, Lakatos A, Jones D. The CART receptors: background and recent advances. Peptides. 2006; 27(8):1934–7.

36. Lin Y, Hall RA, Kuhar MJ. CART peptide stimulation of G protein-mediated signaling in differentiated PC12 Cells: Identification of PACAP 6–38 as a CART receptor antagonist. Neuropeptides. 2011; 45(5):351–8.

135

Chapter 5: GLP-1R in vagal afferents, energy expenditure and brown adipose tissue

137

CHAPTER 5:

GLP-1 RECEPTORS IN VAGAL AFFERENT NEURONS MODULATE

ENERGY EXPENDITURE VIA BROWN ADIPOSE TISSUE

THERMOGENESIS

AUTHORS

Krieger JP, Arnold M, Pettersen KG, Lossel P, Langhans W, Lee SJ


STATUS

In preparation


138

ABSTRACT

The glucagon-like peptide-1 (GLP-1) is a gut peptide involved in the control of energy intake

and glucose metabolism. Whether intestinal GLP-1 also plays a role in the regulation of

energy expenditure, is, however, unknown. Recent studies have shown that central GLP-1

receptor (GLP-1R) activation acutely increases energy expenditure via a stimulation of brown

adipose tissue (BAT) thermogenesis, raising the possibility that peripheral GLP-1R activation

may also be involved in the regulation of BAT thermogenesis. To test this hypothesis, we first

investigated the effects of peripheral GLP-1R activation on energy expenditure in male

Sprague-Dawley rats while controlling for potential confounding factors, i.e., food intake and

gastric emptying. In contrast to central GLP-1R activation, peripheral GLP-1R activation by

the GLP-1R agonist exendin-4 induced a transient decrease of whole-body energy

expenditure, accompanied by a decrease in interscapular BAT (iBAT) temperature. The

absence of GLP-1R in iBAT prompted us to search for a new pathway by which endogenous

peripheral GLP-1 can modulate BAT activity. Using a retrograde viral tracing technique, we

here present evidence that iBAT adipocytes are synaptically connected to vagal afferent

neurons (VAN). To test the physiological relevance of this pathway, we knocked down GLP-

1R in VAN. When exposed to a high-fat diet (HFD), GLP-1R kd rats showed an increase in

energy expenditure and BAT activity during the dark phase compared to control rats, which

alleviated HFD-induced obesity and glucose intolerance. Together, our data indicate that

endogenous GLP-1 plays a role in energy expenditure and that a functional gut-BAT

crosstalk via VAN is important for the pathophysiology of obesity.


139

INTRODUCTION

The development of obesity is indisputably influenced by fat- and carbohydrate-rich

« western » diets, but the nutrient sensing mechanisms leading to the dysregulation of

energy balance when exposed to such diets remains to be fully elucidated. The vagus nerve

is a key neuronal element in the gut-brain communication: therefore, its role in nutrient

sensing and, more broadly, in the regulation of energy metabolism has recently received

increasing attention. In recent years, newly developed strategies allowed specific genetic

manipulation targeting vagal afferent neurons (VAN) and provided important insights into the

molecular mechanisms controlling eating and energy expenditure (1-5). Recently, using the

bilateral injection of a viral vector in the nodose ganglion (NG), we demonstrated that VAN

GLP-1R expression is necessary for the effects of endogenous intestinal glucagon-like

peptide-1 (GLP-1) on eating and glycemia (5). GLP-1R knockdown (kd) in VAN increased

meal size and slowed down gastric emptying without affecting 24 h food intake or body

weight, and raised post-meal glucose levels by blunting insulin secretion.

The role of endogenous intestinal GLP-1 in the control of energy intake and glucose

metabolism is well characterized, whereas the physiological effects of intestinal GLP-1 on

energy expenditure and its components have not yet been systematically investigated.

Central GLP-1R activation acutely increases whole-body energy expenditure (6-8) and

stimulates brown adipose tissue (BAT) thermogenesis (7) in rodents, but whether

endogenous intestinal GLP-1 can access central GLP-1R is unclear. Studies using peripheral

administration of GLP-1 or GLP-1R agonists should better address the role of intestinal GLP-

1 but led to conflicting reports on energy expenditure and its surrogate measures in mice,

rats and humans (8-13). In addition, there is evidence indicating a role of the vagus nerve in

controlling energy expenditure and BAT thermogenesis. Indeed, diverse techniques of vagus

nerve stimulation (VNS) were linked with a modulation of whole-body energy expenditure

and BAT activity in rodents and humans (14-17). Moreover, alterations of lipid sensing

pathways in VAN (via downregulation of the nuclear receptors peroxisome proliferator-


140

activated receptor γ, PPARγ, and liver X receptors α and β, LXRα/β) alleviated high-fat diet

(HFD)-induced obesity, mainly by increasing BAT thermogenesis (2,3). The exact role of

VAN in the regulation of whole-body energy expenditure and BAT thermogenesis, however,

is still elusive.

Considering the pivotal role of VAN in mediating the GLP-1 signal to the brain and the

common feature of VAN and GLP-1 to modulate BAT activity, we hypothesized that a/

peripheral GLP-1R activation can modulate whole-body energy expenditure, b/ this

modulation is dependent on VAN GLP-1R and c/ the effect of peripheral GLP-1 on energy

expenditure is at least partly due to changes in BAT activity. In order to clarify the effect of

endogenous intestinal GLP-1 on energy expenditure, we activated peripheral GLP-1R using

an intraperitoneal (IP) injection of exendin-4 (Ex-4) in rats and measured whole-body energy

expenditure and BAT thermogenesis. We further investigated the potential role of VAN for

the modulation of BAT activity using retrograde tracing studies from the interscapular BAT

(iBAT) pads. Finally, we tested the physiological relevance of this vagal pathway in rats with

a reduced GLP-1R expression in VAN. Overall, our data propose a new role for endogenous

GLP-1 to tune down energy expenditure after a meal via a previously undescribed gut-brain-

BAT pathway. Furthermore, increased intestinal GLP-1 secretion during HFD feeding may

reduce the dissipation of nutrients as heat, thus contributing to body weight gain and glucose

intolerance.

METHODS

Animals and housing

Male Sprague-Dawley rats (Charles River) were individually housed in a temperature and

humidity-controlled environment (21±1°C, 55±5% HR) with a 12/12 h dark/light cycle.

Standard laboratory chow (3433, Kliba Nafag, energy density: 3.13 kcal/g) was provided ad


141

libitum unless otherwise noted. All procedures were approved by the Zurich Cantonal

Veterinary Office.

Acute effects of IP Ex-4 injection on energy expend iture, BAT temperature and

locomotor activity

Rats were individually housed and habituated for 6 days to the new environment and to IP

injections of NaCl prior to the experiment. The routine daily health check was performed 2 h

before the dark phase onset. On experimental days, rats (350-450 g) were allowed food

access ad libitum until food was withdrawn 2 h into the dark phase. Two h later, rats received

an IP injection of NaCl or Ex-4 (0.3 µg/kg) and were immediately brought back to their cages.

This protocol was repeated four times to a/ measure whole-body energy expenditure and

RER, b/ measure iBAT temperature with infrared imaging, c/ measure locomotor activity and

d/ collect tissue for subsequent molecular analysis. Injections were randomized and a

minimum of 4 days separated each experimental day.

Indirect calorimetry

Rats were individually housed in an open circuit indirect calorimetry Phenomaster system

(TSE). Individual body weights were used to normalize energy expenditure. Data are

presented as 1 h time bins.

BAT Infrared imaging

To increase the sensitivity of the BAT temperature measurements, the interscapular area

was shaved. Infra-red (IR) pictures were taken with an E60 camera (FLIR), mounted

vertically over the area of interest (distance 30 cm). Rats were lightly restrained in a

stretched position under the camera and 3 snapshots/timepoint were taken for further

analysis. Baseline IR pictures were taken 2 h after dark onset and a second series of IR

pictures was taken 2 h after the IP injection. Final analysis of the IR pictures was done by

defining a standard rectangular area (60 x 20 px) over the interscapular region and averaging

the mean area temperatures of two pictures (FLIR Tool software for PC).


142

Locomotor activity

Ninety min after IP injection of Ex-4 (to overlap the phase during which EE was at the

lowest), rats locomotor activity was monitored for 45 min during an open field test. The open

field test was conducted in two identical arenas (80x80x50 cm), filmed by a digital camera

and analyzed with a computer running the EthoVision tracking system (see (18) for a full

description).

Tissue collection

Subcutaneous white adipose tissue (ScWAT) and brown adipose tissue (BAT) were

immediately collected after decapitation of the animals 2 h after IP Ex-4 injection. Tissues

were immediately frozen in liquid nitrogen and stored at -80°C until further processing.

Retrograde labeling from the iBAT

Rats (520-570 g on surgery day) were anesthetized with ketamin (Ketalar, 88 mg/kg IP,

Zurich Cantonal Pharmacy) and xylazin (Rompun 2%, 5 mg/kg IP, Zurich Cantonal

Pharmacy). An incision of 4 cm was made above the interscapular area to allow visualization

of and access to the brown fat pads. One µL of a pseudorabies 263 virus (PRV-263, 1.4x109

pfu/mL, provided by Lynn Enquist, Princeton University, virus center grant p40RR018604)

was injected into each iBAT pad. After injection, the incision was immediately stitched and

rats received appropriate post-surgical care (20 mg/kg sulfadoxin, 5 m/kg trimethoprim,

subcutaneously, SC and 5 mg/kg carprofenum, SC, Zurich Cantonal Pharmacy). Four days

after PRV-263 injection, rats were euthanized with CO2. Nodose ganglia were immediately

collected and fixed in 4% PFA. Fixed nodose ganglia were cut at 10 µm in a cryostat (CM

1950, Leica) and fluorescence was visualized with an Axioscope.A1 microspoce (Zeiss)

equipped with an AxioCam MRC camera (Zeiss).


143

GLP-1R knockdown in vagal afferent neurons

Lentiviral-mediated knockdown

The production of lentiviral particles targeting the GLP-1R mRNA via a shRNA construct was

carried out as previously described (5). Briefly, a U6 promoter-driven shRNA sequence

targeting the rat GLP-1R mRNA was expressed in a pLKO.1-puro vector (Sigma-Aldrich). A

non-target shRNA sequence (Sigma-Aldrich) expressed in a similar vector was used as a

control. Each of these plasmids has been co-transfected with the pMD2.G and psPAX2

plasmids (Didier Trono, Addgene #12259 and 12260) in HEK 293T cells, and lentiviral

particles were collected, concentrated to 1010 particles/mL using 8% PEG6000 (Millipore) in

0.3M NaCl and resuspended in PBS. Rats (270-330 g on surgery day) were anesthetized

with ketamin (Ketalar, 88 mg/kg IP, Zurich Cantonal Pharmacy) and xylazin (Rompun 2%, 5

mg/kg IP, Zurich Cantonal Pharmacy). An incision in the anterior neck was made to provide

successive access to the left and right cervical branches of the vagus nerve. 1.5 µL of the

concentrated viral solution were injected into each nodose ganglion (NG) with a Picospritzer

III injector (Parker Hannifin). The incision was sutured and post-surgery care was given

according to the above-mentioned protocol. Before and during 20 days after surgery, all

animals had access to standard chow. After the 20-day period left for the knockdown to fully

establish, animals had ad libitum access to a HFD (D12492, Sniff, 60% energy from fat,

energy density: 5.16 kcal/g).

Measurement of food intake and body composition

Daily food intake and meal patterns were measured by an automated system (see (19) for

description) unless animals were placed in the indirect calorimetry system (see below).

Analysis of body composition was conducted 10 weeks after introduction of the HFD using a

computerized tomography system (La Theta LCT-100, Aloka) with a previously validated

method (20). Briefly, rats were anesthesized with isoflurane and placed supine in a cylindric

holder (120 mm inner diameter). An initial whole-body sagittal image was obtained to ensure

proper placement of the animal. Scans were done from vertebrae L1 to L6 with a 2 mm pitch


144

size. Volumes of adipose tisse (fat mass), bone, air, and the remainder (lean mass) are

detected by the Aloka software based on their X-ray absorption. Fat mass and lean mass

were computed using the density factors of 0.92 g/cm3 and 1.10 g/cm3, respectively. Fat ratio

is defined as fat mass / (lean + fat mass).

Continuous measurements of energy expenditure, locomotor activity, body (IP) and

iBAT temperatures

A telemetry sensor (TA-F40, DSI) was surgically implanted IP to monitor activity in freely-

moving animals. After recovery, rats were placed in an indirect calorimetry system

(Phenomaster, TSE) and habituated for 5 days to the environment prior to the data

collection. Energy expenditure and activity were monitored in parallel in the indirect

calorimetry chambers, and averaged over 3 days. Daily health check of the rats was

performed 2 hours before the dark phase onset. Individual body weights were used to

normalize energy expenditure. To monitor BAT temperature, a dual telemetry sensor (F40-

TT, DSI) was surgically implanted (SC) in a separate rat cohort 10 weeks after introduction of

a HFD. The two flexible temperature probes were placed in close apposition with the

interscapular brown fat pads, or in the peritoneal cavity, respectively, and secured with

sutures. After 10 days of recovery, BAT and IP temperatures were continuously measured in

ad libitum HFD-fed animals (weeks 11 to 12, Dataquest ART v3.1, DSI).

Oral glucose tolerance test (OGTT)

An OGTT was performed 11 weeks after the introduction of the HFD. Fourteen h-food

deprived rats adapted to gavage and restraining received an oral bolus of 40% glucose

solution (2 g/kg). Tail vein blood samples for glucose and insulin were taken at baseline and

15, 30, 60, 90, and 120 min after the oral glucose bolus. Blood glucose was measured twice

using Accucheck (Roche). Insulin was measured from plasma using an immunoassay

(MesoScale Discovery single-spot for mouse/rat K152BZC).


145

Intraperitoneal insulin sensitivity test (IPIST)

An IPIST was performed 12 weeks after the introduction of the HFD. Fourteen h-food

deprived rats adapted to injections and restraining received an intraperitoneal injection of 1

IU/kg of insulin (Actrapid HM, Novo Nordisk). Tail vein blood glucose was measured twice

per timepoint using Accucheck (Roche) at baseline and 15, 30, 45, 60, 90, 120 and 180 min

after injection.

HFD test meal-induced GLP-1 levels

10 weeks after introduction of the HFD, rats were habituated to test meals and restricted

feeding schedule for a week (Test meal at dark phase onset, ad libitum food access from 3 to

8 h after dark phase onset, food deprivation otherwise). On experimental days, rats received

a 3 g HFD test meal at dark phase onset. Blood was sampled from hepatic portal vein (HPV)

catheters in unrestrained animals 30 min prior to (baseline) and according to the scheduled

timepoints. At each time point, 150 µL of blood was immediately mixed with EDTA (Titriplex,

Merck), Aprotinin (Sigma-Aldrich) and DPP-IV inhibitor (Millipore) before centrifugation and

storage of the plasma at -80°C. Total active GLP-1 was measured using an immunoassay

(MesoScale Discovery multi-spot K15171C).

Tissue collection of GLP-1R kd rats

Animals received an IP injection of pentobarbital (100 mg/kg, Zurich Cantonal Pharmacy)

and NG (ganglia from the same animal were pooled), brain, subcutaneous white adipose

tissue (ScWAT) and iBAT were immediately collected. For gene and protein expression

analysis, tissues were immediately frozen in liquid nitrogen and stored at -80°C until further

processing. The nucleus tractus solitarii (NTS), arcuate nucleus (Arc) and paraventricular

nucleus of the hypothalamus (PVH) were collected after brain cryosection and

micropunching. For stainings with haematoxylin and eosin (H&E), adipose tissues were fixed

in 4% PFA (Sigma-Aldrich) and processed in a STP 120 (Microm) according to the standard

protocol. Paraffin-embedded samples were cut at 5 µm on a Hyrax M55 (Zeiss),


146

deparaffinized and stained on a Varistain 24-4 (Shandon). Pictures were obtained with an

Axioscope A.1 microscope (Zeiss).

Gene expression and protein analysis

RNA and proteins were extracted using Trizol (Life Technologies). RT-qPCR analysis was

performed using SybR Green on a OneStep Plus instrument (Applied Biosystems) and

results were analyzed using the 2ddCt method. Western blots were performed to detect the

GLP-1R protein (Rabbit antibody 39072, 1:400, Abcam) or the Uncoupling protein-1 (UCP1;

Rabbit antibody PA1-24894 1:2000, ThermoFisher) using β-actin as reference (Mouse

antibody AC-74, 1:3000, Sigma-Aldrich).

Statistical analysis

Data normality was verified using the Shapiro-Wilk (when n ≥ 7) and the Kolmogorov-

Smirnov (when 5 ≤ n ≤ 6) tests and homoscedasticity was checked by visualizing the

distribution of residuals. Non-parametric tests were used otherwise. Tests used and P-values

are given in the figure legends. Data are presented as means ± SEM. P-values < 0.05 were

considered significant. All graphs were generated using GraphPad Prism (version 6.05 for

Windows).

RESULTS

Peripheral GLP-1R activation induced a transient de crease of energy expenditure

To test the effects of peripheral GLP-1R activation on whole-body energy expenditure, we

performed an IP injection of the GLP-1R agonist Ex-4 at a dose that previously inhibited food

intake via the activation of VAN GLP-1R (0.3 µg/kg) (5). Moreover, to isolate the effects of

GLP-1R activation on EE per se, we standardized food intake and gastric emptying using a

restricted feeding schedule (no access to food 2h before and 4h after the IP Ex-4 injection-

Fig. 1A). Our data demonstrate that IP injection of Ex-4 transiently reduced whole-body


147

Figure 1: IP injection of the GLP-1R agonist Ex-4 ( 0.3 µg/kg) transiently reduced energy expenditure, RER and iBAT thermogenesis (A) Experimental protocol used to study the effects of IP Ex-4 on energy homeostasis independently of the effects on food intake and gastric emptying (B) Hourly heat production following IP Veh or Ex-4 injection (n = 7/9; Two-way ANOVA group x time; group effect NS, time effect P < 0.0001, interaction P = 0.1) (C) Cumulative energy expenditure during the 4 h following IP Veh or Ex-4 injection (n = 7/9; Student t-test; P < 0.05) (D) Hourly evolution of the respiratory exchange ratio following IP Veh or Ex-4 injection (n = 7/9; Two-way ANOVA group x time; group effect NS, time effect P < 0.0001, interaction P < 0.005) (E) Total distance moved during a 45-min open field test (n = 8/7; Student t-test; NS) (F) Average temperature of the interscapular area before and 2 h after IP Veh or Ex-4 injection, with representative pictures. White rectangles indicate the area considered to evaluate interscapular skin temperatures (n = 8; Two-way ANOVA group x time; group effect NS, time effect NS, interaction P < 0.01; * indicates a significant difference with Veh-injected rats 2 h post-injection - Bonferronic post-hoc test) (G) Relative mRNA expression of thermogenic markers in the iBAT; (n = 8/7; Student t-tests; P < 0.01 for B3AR; P < 0.1 for PGC1α and PPARγ; P < 0.2 for UCP1, Cidea, FGF21, HSL and MGL; P > 0.3 for ATGL and DIO2 (H) Relative UCP1 protein levels in the iBAT 2 h after IP Veh or Ex-4 injection, with representative bands (n = 7; Student t-test; NS) Abbreviations: OF, open field; UCP1, Uncoupling protein-1; B3AR, β-3 adrenergic receptor; PGC1α, peroxisome proliferator-activated receptor γ coactivator α-1; PPARγ, peroxisome proliferator-activated receptor γ; DIO2, type 2 iodothyronine deiodinase; FGF21, fibroblast growth factor 21; ATGL, adipose triglyceride lipase; HSL, hormone-sensitive lipase; MGL, monoacylglycerol lipase


148

energy expenditure (Fig. 1B, C) and respiratory exchange ratio (RER) (Fig. 1D) in a

timeframe of 4 h after injection. Because Ex-4 is known to induce visceral malaise (21), it

may impact energy expenditure via a reduction in locomotor activity. To test this, we

measured locomotor activity in a 45 min open field test after IP injection of Veh or Ex-4 and

showed that Ex-4 at the dose effective for reducing energy expenditure did not reduce

locomotor activity (Fig. 1E and Supp. Fig. 1). Together, our data indicate that peripheral

GLP-1R activation reduces whole-body energy expenditure independent of a decrease in

food intake and without reducing locomotor activity.

Peripheral GLP-1R activation induced a transient de crease in iBAT temperature

We further tested the idea that the reduction in energy expenditure induced by peripheral

GLP-1R activation was at least partially due to the changes in BAT thermogenesis. Using a

similar paradigm as above, we found that interscapular skin temperature was decreased 2 h

after IP Ex-4 injection compared to vehicle injection (Fig. 1F), indicative for a reduced iBAT

activity. Furthermore, at the same time point, the expression of thermogenic markers tended

to be lower in the iBAT of Ex-4 injected rats compared to controls (Fig. 1G). We, however,

failed to detect the decrease in the protein levels of UCP1 in iBAT of Ex-4 injected rats (Fig.

1H).

Potential control of iBAT activity by VAN

iBAT showed no detectable mRNA or protein expression of the GLP-1R, as measured by

RT-PCR and Western blot of iBAT and NG extracts (Fig. 2A, B). This eliminates the

possibility of a direct action of GLP-1 on brown adipocytes. To test if VAN are involved in the

control of BAT activity, we first tested whether iBAT and NG are synaptically linked using a

neuroanatomical tracing technique. For this, we injected the retrograde tracer PRV-263

expressing tdTomato bilaterally in the iBAT pads. 4 days after PRV injection, a subset of

neurons in both NG were labeled with tdTomato (Fig. 2C). These data indicate a neuronal

network linking the gut and the iBAT. Together with our previous results, it suggests that

endogenous GLP-1 can modulate energy expenditure and BAT thermogenesis via VAN.


149

Figure 2: Absence of GLP-1R in the iBAT of the rat and synaptic connections between the iBAT and VA (A) Products of a RT-PCR ran with GLP-1R primers only (lane 1 - “Neg”) or together with iBAT cDNAs (lanes 2 to 5 - “iBAT”) or with NG cDNAs (lanes 6 to 8 - “NG”); expected GLP-1R amplicon size: 79bp (B) Relative expression of the GLP-1R protein in the iBAT and in the NG, with representative bands (n = 4/5; Mann-Whitney test; P < 0.05) (C) Representative fluorescent picture of a NG 4 days after injection of a PRV-263 virus into the iBAT. Arrows indicate labeled cells, top right frame shows labeling of fibers within the NG GLP-1R kd in VAN reduced HFD-induced body weight ga in and impairments of

glucose homeostasis

To test the relevance of GLP-1R activation in VAN for the control of BAT activity, we used

our established model of Sprague-Dawley rats with a specific kd of GLP-1R in VAN (5).

Injection of lentiviral particles containing a GLP-1R targeting shRNA construct led to a

significant reduction of GLP-1R mRNA (Fig. 3A) and protein (Fig. 3B) expression in the NG

compared to control injected rats. In this model, body weight is equivalent to the one of

control rats (5). When fed a 60% HFD, however, the body weight gain of GLP-1R kd rats

diverged from controls, leading to a significantly lower body weight gain after week 9 in GLP-

1R kd rats (Fig. 3C). As indicated by a trend towards a lower fat mass without any change in

lean mass, the reduction in HFD-induced weight gain in GLP-1R kd rats may predominantly

be explained by less fat accumulation (Fig. 3E). Surprisingly, although GLP-1R kd rats


150

Figure 3: GLP-1R kd in VAN alleviated HFD-induced b ody weight gain and glucose intolerance (A) Relative GLP-1R mRNA expression in control and GLP-1R kd rats (n = 10/9; Student t-test; P < 0.001) (B) Relative GLP-1R protein expression in control and GLP-1R kd rats, with representative bands (n = 7/6; Student t-test; P < 0.05) (C) Body weight gain of control and GLP-1R kd rats after introduction of the HFD (n = 13/14; Two-way ANOVA, group x time; group effect P < 0.05, time effect P < 0.0001, interaction effect P < 0.01; * indicates a significant difference between control and GLP-1R kd rats - post-hoc Bonferroni comparisons) (D) Daily intake of HFD by control and GLP-1R kd rats (n = 13/14; Two-way ANOVA, group x time; group effect NS, time effect P < 0.0001, interaction effect NS) (E) Lean and fat mass of control and GLP-1R kd rats (n = 13/13; Student t-tests; NS) (F) Blood glucose during an OGTT (2 g/kg) in control and GLP-1R kd rats (n = 13/14; Two-way ANOVA group x time; group effect NS, time effect P < 0.0001, interaction effect P < 0.05; * indicates a significant difference between control and GLP-1R kd rats - post-hoc Bonferroni comparisons) (G) Area under curve of blood glucose during OGTT (n = 13/14; Student t-test; P < 0.01) in control and GLP-1R kd rats (H) Plasma insulin durin OGTT (n = 12/14; Two-way ANOVA group x time; group effect P < 0.05, time effect P < 0.0001, interaction effect NS; * indicates a main group effect) in control and GLP-1R kd rats (I) Blood glucose during an IPIST in control and GLP-1R kd rats (n = 6/7; Two-way ANOVA group x time; group effect NS, time effect P < 0.0001, interaction effect P < 0.05; * indicates a significant difference between control and GLP-1R kd rats - post-hoc Bonferroni comparisons)


151

showed a larger meal size than controls before being switched to HFD, HFD exposure did

not lead to alterations in either daily food intake or meal patterns (Fig. 3D, Supp. Fig. 3). We

further tested whether the lower body weight gain in GLP-1R kd rats was associated with

improved glucose intolerance and insulin sensitivity. GLP-1R kd rats showed a reduced

glucose excursion compared to controls from 60 to 120 min after an oral glucose bolus (Fig.

3F), leading to an overall reduced glucose AUC (Fig. 3G). Consistent with glucose levels,

plasma insulin levels were also lower in GLP-1R kd rats than in controls after the oral

glucose bolus (Fig. 3H). An IP insulin sensitivity test (IPIST) also showed a more robust

reduction in blood glucose 15 min after IP injection of insulin in GLP-1R kd rats than in

controls (Fig. 3I). These results indicate that GLP-1R kd in VAN confered a relative

protection from HFD-induced weight gain, glucose intolerance and insulin resistance.

GLP-1R kd in VAN increased dark phase energy expend iture and iBAT temperature

during HFD exposure

When chronically challenged with a 60% HFD, GLP-1R kd rats showed an increased heat

production in the early dark phase compared to their controls (Fig. 4A, B). This increase in

energy expenditure was not due to changes in locomotor activity (Supp. Fig. 2). Consistent

with the absence of changes in food intake, RER was similar between the two groups (Fig.

4C). In another cohort, we implanted a dual temperature probe and measured in parallel IP

and BAT temperatures in control and GLP-1R kd rats. iBAT temperature was constantly

elevated during the dark phase in GLP-1R kd rats compared to controls, whereas it remained

similar to controls during the light phase (Fig. 4E). IP temperature remained similar

throughout the day between the two groups (Fig. 4D).

GLP-1R kd in VAN increased thermogenic capacity in both BAT and ScWAT during

HFD exposure

iBAT H&E staining showed a smaller average cell size in the GLP-1R kd group, indicative for

increased BAT thermogenesis in this group (Fig. 4G). In addition, the molecular analysis of

the iBAT after 12 weeks on HFD revealed an increased UCP1 mRNA and protein expression


152

Figure 4: GLP-1R kd in VAN increased dark phase ene rgy expenditure and BAT and ScWAT thermogenic capacity during HFD exposure (A) Hourly evolution of heat production over 24 h in control and GLP-1R kd rats (n = 7/8; Two-way ANOVA; group x time; group effect P = 0.1, time effect P<0.0001, interaction effect NS) (B) Cumulative energy expenditure split in day quarters in control and GLP-1R kd rats (n = 7/8; Sidak-Bonferroni corrected t-test; P < 0.05 for 0 - 6 h, NS for 6 - 12 h and 12 - 18 h, P = 0.1 for 18 – 24 h) (C) Hourly evolution of respiratory exchange ratio (RER) over 24 h in control and GLP-1R kd rats (n = 7/8; Two-way ANOVA; group x time; group effect NS, time effect P<0.0001, interaction effect NS) (D) Real-time 24 h monitoring of IP temperatures (n = 6/6; Two-way ANOVA group x time; group effect NS, time effect NS, interaction effect NS) (E) Real-time 24 h monitoring of iBAT temperatures (n = 6/6; Two-way ANOVA group x time; group effect P = 0.06; time P < 0.0001; interaction P < 0.001) (F) Relative mRNA expression of thermogenic markers in the iBAT (n = 8/8; Student t-tests; P<0.05 for UCP1 and


153

PPARΥ; NS for other genes) (G) Representative pictures of H&E staining of iBAT of control and GLP-1R kd rats (H) Relative mRNA expression in the ScWAT (n = 6/7; P < 0.05 for B3AdrR and PPARΥ; P = 0.1 for FAS and ACC and NS for other genes (I) Relative expression of UCP1 protein levels in the ScWAT, with representative bands (n = 6/7; Student t-test; P < 0.05) Abbreviations: UCP1, Uncoupling protein-1; B3AdrR, β3-adrenergic receptor; PGC1α, peroxisome proliferator-activated receptor γ coactivator α-1; PPARγ, peroxisome proliferator-activated receptor γ; DIO2, type 2 iodothyronine deiodinase; FGF21, fibroblast growth factor 21; ATGL, adipose triglyceride lipase; HSL, hormone-sensitive lipase; MGL, monoacylglycerol lipase; FAS, fatty acid synthase; ACC, acetyl-coA carboxylase

and a trend towards an increase in β3-adrenergic receptor (B3AR) expression in the GLP-1R

kd rats compared to controls (Fig. 4F). In addition, RT-qPCR indicated an increase in

thermogenic markers (including UCP-1, PPARγ, and B3AR) and a trend towards a reduction

in the expression of lipogenic enzymes (FAS, ACC) in the scWAT of GLP-1R kd rats

compared to controls (Fig. 4H). A robust increase in UCP1 was also detected at the protein

level (Fig. 4I). Together, these results indicate an increased BAT thermogenic phenotype and

browning of the ScWAT via increased sympathetic inputs as a result of VAN GLP-1R kd.

GLP-1R kd in VAN led to a maintenance of ppg expression in the NTS during HFD

exposure

We further investigated whether central regulators of energy balance and glucose

homeostasis were modulated by the GLP-1R kd in VAN. In the arcuate nucleus of the

hypothalamus (Arc), there was no difference in the expression of feeding related

neuropeptides, consistent with the similar food intake between two groups (Fig. 5B). Similary,

in the paraventricular nucleus of the hypothalamus (PVN), there was no change in genes

involved in the neuroendocrine control of energy metabolism (Supp. Fig. 4A).

Preproglucagon (PPG) is a precursor for the GLP-1 and is produced in the NTS in the

hindbrain. Interestingly, we found that HFD feeding decreased ppg mRNA expression in the

NTS compared to normal chow feeding in control rats (Supp. Fig. 4B). GLP-1R kd prevented

this HFD-induced decrease in PPG mRNA expression (Fig. 5A and Supp. Fig. 4B) whereas it

did not affect the central GLP-1R expression (Fig. 5A, B and Supp. Fig. 4B). Because central

GLP-1R activation modulates BAT functions, the protective effect of VAN GLP-1R kd on NTS


154

ppg expression suggests a potential central mechanism linking peripheral GLP-1R signaling

to the adipose tissue function.

Figure 5: GLP-1R kd in VAN increased NTS ppg expression during HFD exposure (A) Relative mRNA expression in the NTS (n = 6/7; Student t-test; P < 0.05 for ppg, NS for other genes) (B) Relative mRNA expression in the Arc (n = 7/8; Student t-test; NS) Abbreviations: AgRP, Agouti-Related Peptide; Arc, arcuate nucleus of the hypothalamus; DBH, dopamine β-hydroxylase; GLP-1R, glucagon-like peptide-1 receptor; NTS, nucleus tractus solitarii; POMC, proopiomelanocortin; PPG, preproglucagon; NPY, neuropeptide-Y

DISCUSSION

Recent studies have highlighted the newly emerging role of GLP-1 in the control of energy

expenditure and, in particular, the stimulatory effects of central GLP-1R activation on BAT

thermogenesis. This study is, to our knowledge, the first to unequivocally demonstrate the

differential role of peripheral GLP-1R activation on EE and BAT thermogenesis.

First, our study shows that peripheral GLP-1R agonist treatment decreases whole-body

energy expenditure and BAT thermogenesis. There were conflicting results in the literature

regarding the effects of peripheral GLP-1 or GLP-1R agonist treatment on energy

expenditure (8,9,11-13). These discrepancies may be explained by the use of different routes

of administration (IP, IV, SC), doses (see chapter 1, table 1), agonists (GLP-1, Ex-4,

liraglutide) or species (mice, rats, hamsters, humans). Furthermore, it is important to

consider that the reduction in food intake and gastric emptying induced by GLP-1

administration can lead to adaptive thermogenesis. Therefore, it was crucial to dissect the

direct effect of GLP-1R activation on energy expenditure by eliminating these parameters


155

with a restricted feeding schedule. Our data clearly demonstrate that peripheral GLP-1R

activation by a low dose of IP Ex-4 reduces RER, energy expenditure and BAT temperature

in the absence of eating, suggesting a direct effect. It remains unclear how IP Ex-4 reduces

iBAT temperature, given the relatively mild modulation of thermogenic markers and the

absence of apparent change in UCP1 protein levels. Because we measured gene and

protein expression 2 h after Ex-4 injection, when energy expenditure started to increase

again, it remains therefore possible that the BAT thermogenic activity is already normalized

to the baseline level.

Since endogenous GLP-1 requires normal GLP-1R expression in VAN for the control of food

intake and glycemia, we next hypothesized that this gut-brain pathway may also be involved

in the regulation of BAT thermogenesis. Injections of retrogradely transported viral tracers

have been widely used to find the CNS origins of the sympathetic regulation of the BAT

((22,23) among others). Using a fluorescent-labeled transynaptic PRV, we visualized VAN in

the NG retrogradely labeled from the iBAT, suggesting a vagally-mediated control of BAT

thermogenesis rather than a direct activation of GLP-1R in the BAT. We acknowledge that

our tracing data identified synaptically connected neurons between VAN and iBAT but did not

provide indisputable evidence that this pathway is relevant for the effects of peripheral GLP-

1R activation on energy expenditure and BAT activity. Further studies are on the way to

investigate whether these PRV-infected VAN express GLP-1R and whether the acute effects

of IP Ex-4 on energy expenditure and BAT temperature are attenuated in GLP-1R kd rats.

Nonetheless, we could speculate on the potential central pathways involved in this gut-to-

BAT pathway. The premotor neurons in the rostral raphe pallidus (rRPa) were shown to be

the key mediators of the sympathetic control of BAT thermogesis (24,25), providing an

excitatory drive to BAT sympathetic preganglionic neurons in the spinal intermediolateral

nucleus. rRPa premotor neurons activity is modulated by hypothalamic and brainstem

projections (25-27). Indeed, decerebrate rats showed higher BAT temperature than intact

rats after injection of a melanocortin receptor agonist into the rRPa (28), indicating that the


156

caudal brainstem circuit is sufficient for tuning BAT thermogenesis. Moreover, disinhibition of

NTS neurons via a GABAa receptor antagonist blocked the increase in BAT sympathetic

nerve activity and BAT temperature induced by various stimuli (including cold, and direct

activation of rRPa neurons) in an anesthetized rat model (26). Therefore, it is reasonable to

speculate that rRPa neurons integrate vagally-mediated signals via inhibitory NTS

projections to modulate BAT sympathetic outflow.

Our study further demonstrated the physiological relevance of a gut-BAT crosstalk involving

gut-derived GLP-1 in the pathophysiology of obesity. Indeed, permanent downregulation of

GLP-1R kd in VAN increased BAT thermogenesis and energy expenditure, which was

associated with reduced HFD-induced body weight gain and markedly improved glucose

homeostasis. Several recent studies have suggested the involvement of the vagus nerve in

the regulation of HFD thermogenesis. VA express genes that are involved in lipid sensing

mechanism and are thus relevant for the sympathetic regulation of adipose tissue and lipid

metabolism. Indeed, the loss of nuclear receptor expression involved in lipid sensing,

(PPARγ (2) and LXRα/β (3)) in VAN of mice increased mRNA levels of BAT thermogenic

markers such as UCP1 and β3-adrenergic receptors (2), which were associated with an

increase in energy expenditure, and with attenuated lipid accumulation during HFD exposure

(3). Moreover, hepatic vagotomy restored the reduced BAT thermogenic capacity and rRPa

c-Fos expression induced by a liver glucokinase overexpression in HFD-fed mice (29),

further strengthening the idea that VAN could be the crucial link relaying the effects of

peripheral signals to the CNS to modulate BAT activity.

The fact that, in the GLP-1R kd rat model, the disruption of a gut-BAT crosstalk translates

into whole-body energy expenditure and body weight effects during exposure of HFD, but not

chow, raises an important question. Recent data have shown that 3 to 13 wk HFD-feeding

increases the GLP-1 response to a mixed-meal compared to low fat diet-fed rats (30). In our

study, chronic HFD-fed rats had 3-fold higher meal-induced GLP-1 levels in the hepatic

portal vein than the chow-fed rats after an isocaloric test meal (Supp. Fig. 3E). Hence, it is a


157

plausible speculation that an increased endogenous gut GLP-1 secretion in response to

HFD, and the vagal GLP-1R activation that follows, may be the gut-brain mechanism that

signals the high amount of fats in the diet and promote the central mechanisms favoring the

storage of the excess fat. Therefore, blocking this signaling pathway by GLP-1R kd in VAN

prevented the fat accumulation via an increase in BAT thermogenesis and browing of the

ScWAT. As mentioned earlier, HFD exposure modulates the expression pattern of the

nuclear receptors involved in lipid sensing in VAN (2). Because these nuclear receptors have

been shown to regulate the thermogenic response to HFD exposure, we can also speculate

that GLP-1 signal within the NG interacts with these nuclear receptors. Further investigation

of the role of the lipid sensing in VAN and its interaction with GLP-1 signaling in regards to

the control of energy expenditure is clearly warranted. Intriguingly, our data show a decrease

in NTS ppg expression after chronic HFD exposure in control, but not in GLP-1R kd rats.

Acute central injection of GLP-1 or GLP-1R agonists has been shown to transiently increase

energy expenditure in rodents (6-8), via a mechanism involving hypothalamic GLP-1R,

adenosine monophosphate-activated protein kinase (AMPK) (7) and increased sympathetic

outflow to the BAT (31). This suggests that increased central GLP-1 production in GLP-1R kd

rats during HFD exposure may be one of the mechanisms contributing to the increased BAT

activity in this model. NTS ppg-expressing neurons respond to food-related cues, such as

CCK, gastric distension, and leptin, and ppg kd results in hyperphagia and fat accumulation,

suggesting the involvement of central GLP-1 system in the regulation of energy balance (32).

Although GLP-1 does not directly activate ppg neurons (33), chronic peripheral treatment

with the GLP-1R agonist liraglutide decreased central ppg gene expression compared to ad

libitum fed and weight-matched control rats (34). The results from our GLP-1R kd study

indicate that GLP-1R in VAN are necessary for the modulation of central ppg expression

induced by HFD exposure. Together, this raises the possibility that the amplified gut

secretion of GLP-1 during HFD exposure leads to a downregulation of the central GLP-1

system via VAN, which contributes to reduced BAT thermogenesis and fat accumulation.


158

In the light of our data, antagonizing elevated GLP-1 levels early in the development of

obesity could be a promising strategy to alleviate the course of body weight gain and insulin

resistance. This may seem contradictory to the growing body of evidence showing that GLP-

1R agonists treatments are effective at treating obesity via a reduction in eating and a

possible effect on energy expenditure. Several studies, however, have shown that the effects

of the GLP-1R agonist liraglutide on body weight and food intake are mediated by central

GLP-1R (4,34), i.e. receptors physiologically activated by the central GLP-1 system. The

effects of GLP-1R agonists on eating and body weight may predominantly compensate for

the alteration of the central GLP-1 system induced by the gut-brain mechanisms speculated

here.

The existence of a pathway that tunes down BAT thermogenesis after nutrient-induced GLP-

1 secretion implies that, at the whole organism level, heat dissipation of nutrients/fat is being

limited, probably to favor the uptake of upcoming nutrients by peripheral organs. In other

words, GLP-1 may elicit a counter-regulatory mechanism against meal-induced

thermogenesis (the supplementary energy cost above basal metabolic rate for the use and

storage of nutrient) by decreasing heat generation in the BAT. This idea supports recent

findings indicating that diet-induced thermogenesis is limited in the BAT (see (35) for review).

This “thrift” mechanism may better distribute the energy source to maintain the overall energy

balance between the meals, which could be an evolutionary advantage in an environment

where food is scarce. During exposure to a high-caloric diet rich in fat, this mechanism may

however serve to promote the positive energy balance which contributes to the development

of obesity.

In this study, we demonstrated a new role for endogenous intestinal GLP-1 in the control of

energy expenditure via BAT thermogenesis. The clinical relevance of this mechanism in the

development of obesity requires further investigation but our data raised the possibility that

interfering with the GLP-1 gut-BAT crosstalk can alleviate the development of diet-induced

obesity.


159

REFERENCES

1. de Lartigue G, Ronveaux CC, Raybould HE. Deletion of leptin signaling in vagal afferent neurons results in hyperphagia and obesity. Mol Metab. 2014;3(6):595–607.



4. Sisley S, Gutierrez-Aguilar R, Scott M, D'Alessio DA, Sandoval DA, Seeley RJ. Neuronal GLP1R mediates liraglutide’s anorectic but not glucose-lowering effect. J Clin Invest. 2014; 124(6):2456-63.

5. Krieger JP, Arnold M, Pettersen KG, Lossel P, Langhans W, Lee SJ. Knockdown of GLP-1 receptors in vagal afferents affects normal food intake and glycemia. Diabetes. 2016; 65(1):34-43.

6. Hwa JJ, Ghibaudi L, Williams P, Witten MB, Tedesco R, Strader CD. Differential effects of intracerebroventricular glucagon-like peptide-1 on feeding and energy expenditure regulation. Peptides. 1998; 19(5):869–75.

7. Beiroa D, Imbernon M, Gallego R, Senra A, Herranz D, Villaroya F, Serrano M, Fernø J, Salvador J, Escalada J, Dieguez C, Lopez M, Frühbeck G, Nogueiras R. GLP-1 Agonism Stimulates Brown Adipose Tissue Thermogenesis and Browning Through Hypothalamic AMPK. Diabetes. 2014; 63(10):3346-58.

8. Osaka T, Endo M, Yamakawa M, Inoue S. Energy expenditure by intravenous administration of glucagon-like peptide-1 mediated by the lower brainstem and sympathoadrenal system. Peptides. 2005; 26(9):1623–31.

9. Shalev A, Holst JJ, Keller U. Effects of glucagon-like peptide 1 (7-36 amide) on whole-body protein metabolism in healthy man. Eur J Clin Invest. 1997; 27(1):10–6.

10. Conarello SL, Li Z, Ronan J, Roy RS, Zhu L, Jiang G, Liu F, Woods J, Zycband E, Moller DE, Thornberry NA, Zhang BB. Mice lacking dipeptidyl peptidase IV are protected against obesity and insulin resistance. Proc Natl Acad Sci USA. 2003; 100(11):6825–30.

11. Baggio LL, Huang Q, Brown TJ, Drucker DJ. Oxyntomodulin and glucagon-like peptide-1 differentially regulate murine food intake and energy expenditure. Gastroenterology. 2004; 127(2):546–58.

12. Flint A, Raben A, Rehfeld JF, Holst JJ, Astrup A. The effect of glucagon-like peptide-1 on energy expenditure and substrate metabolism in humans. International Journal of Obesity. 2000; 24(3):288–98.

13. Flint A, Raben A, Ersbøll AK, Holst JJ, Astrup A. The effect of physiological levels of glucagon-like peptide-1 on appetite, gastric emptying, energy and substrate metabolism in obesity. International Journal of Obesity. 2001; 25(6):781–92.

14. Morrison SF, Madden CJ, Tupone D. Central Neural Regulation of Brown Adipose


160

Tissue Thermogenesis and Energy Expenditure. Cell Metab. 2014; 19(5):741–56.

15. Morrison SF, Tupone D, Madden CJ. Inhibition of brown adipose tissue (BAT) and shivering thermogenesis by vagal nerve stimulation (VNS). Auton Neurosci - Basic. 2015; 192:2–3.

16. Han L, Jian-Bin Z, Chen X, Qing-Qing T, Wei-Xing S, Jing-Zhu Z, Jian-De C, Yin-Ping W. Effects and mechanisms of auricular vagus nerve stimulation on high-fat-diet-induced obese rats. Nutrition. 2015; 31(11-12):1416–22.


18. Klarer M, Arnold M, Gunther L, Winter C, Langhans W, Meyer U. Gut Vagal Afferents Differentially Modulate Innate Anxiety and Learned Fear. J Neurosci. 2014; 34(21):7067–76.

19. Karimian Azari E, Leitner C, Jaggi T, Langhans W, Mansouri A. Possible role of intestinal fatty acid oxidation in the eating-inhibitory effect of the PPAR-α agonist Wy-14643 in high-fat diet fed rats. PLoS ONE. 2013;8(9):e74869.

20. Hillebrand JJG, Langhans W, Geary N. Validation of Computed Tomographic Estimates of intra-abdominal and Subcutaneousadipose Tissue in Rats and Mice. Obesity. 2009; 18(4):848–53.

21. Kanoski SE, Rupprecht LE, Fortin SM, De Jonghe BC. The role of nausea in food intake and body weight suppression by peripheral GLP-1 receptor agonists, exendin-4 and liraglutide. Neuropharmacology. 2012; 62(5-6): 1916-27.

22. Bamshad M, Song CK, Bartness TJ. CNS origins of the sympathetic nervous system outflow to brown adipose tissue. Am J Physiol. 1999; 276(6 Pt 2):R1569–78.

23. Oldfield BJ, Giles ME, Watson A, Anderson C, Colvill LM, McKinley MJ. The neurochemical characterisation of hypothalamic pathways projecting polysynaptically to brown adipose tissue in the rat. Neurosci. 2002; 110(3):515–26.

24. Cano G, Passerin AM, Schiltz JC, Card JP, Morrison SF, Sved AF. Anatomical substrates for the central control of sympathetic outflow to interscapular adipose tissue during cold exposure. J Comp Neurol. 2003; 460(3):303–26.

25. Morrison SF, Cao W-H, Madden CJ. Dorsomedial hypothalamic and brainstem pathways controlling thermogenesis in brown adipose tissue. J Therm Biol. 2004; 29(7–8):333–7.

26. Cao W-H, Madden CJ, Morrison SF. Inhibition of brown adipose tissue thermogenesis by neurons in the ventrolateral medulla and in the nucleus tractus solitarius. Am J Physiol Regul Integr Comp Physiol. 2010; 299(1):R277–90.

27. Morrison SF. 2010 Carl Ludwig Distinguished Lectureship of the APS Neural Control and Autonomic Regulation Section: Central neural pathways for thermoregulatory cold defense. J Appl Physiol. 2011; 110(5):1137–49.

28. Skibicka KP, Grill HJ. Energetic responses are triggered by caudal brainstem melanocortin receptor stimulation and mediated by local sympathetic effector circuits.


161

Endocrinology. 2008; 149(7):3605–16.

29. Tsukita S, Yamada T, Uno K, Takahashi K, Kaneko K, Ishigaki Y, Imai J, Hasegawa Y, Sawada S, Ishihara H, Oka Y, Katagiri H. Hepatic glucokinase modulates obesity predisposition by regulating BAT thermogenesis via neural signals. Cell Metab. 2012; 16(6):825–32.

30. Wang F, Yoder SM, Yang Q, Kohan AB, Kindel TL, Wang J, Tso P. Chronic High-Fat Feeding Increases Mixed Meal-Induced Incretin Secretion in Sprague-Dawley Rats. Am J Physiol Gastrointest Liver Physiol. 2015; 309(10):G807-15.

31. Kooijman S, Wang Y, Parlevliet ET, Boon MR, Edelschaap D, Snaterse G, Pijl H; Romijn JA; Rensen PC. Central GLP-1 receptor signalling accelerates plasma clearance of triacylglycerol and glucose by activating brown adipose tissue in mice. Diabetologia. 2015; 58(11):2637-46.




35. Kozak LP. Brown fat and the myth of diet-induced thermogenesis. Cell Metab. 2010; 11(4):263–7.


162

SUPPLEMENTS

Supplementary figure 1: Locomotor activity during a 45-min open field test in rats injected with IP Veh or Ex-4 90 min before the start of the test (n = 8/7; Two-way ANOVA group x time; group effect NS, time effect P<0.0001, interaction NS)

Supplementary figure 2: 24 h locomotor activity in freely-moving ad libitum fed control and GLP-1R kd rats (n = 7/8; Two-way ANOVA group x time; group effect NS, time effect P < 0.0001, interaction NS)


163

Supplementary figure 3: (A) Changes in 24 h energy intake, (B) average daily meal size, (C) average daily meal duration and (D) average daily meal numbers before and after introduction of HFD (n = 7/8; Sidak-Bonferroni corrected t-tests; * indicates a difference between control and GLP-1R kd rats) (E) HPV plasma levels of GLP-1 in rats maintained on chow after a chow meal or in rats maintained on HFD after an isocaloric HFD meal (n = 6/7/8/8; MANOVA group x meal x time; group effect NS; meal effect P < 0.0001; time effect P < 0.0001; group*diet NS); data of the chow group have been published elsewhere (5)

Supplementary figure 4: (A) Relative mRNA expression the paraventricular nucleus of the hypothalamus of HFD-fed control and GLP-1R kd rats (n = 8/8 ; Student t-test ; NS) (B) Relative mRNA expression of ppg in the NTS of age-matched chow- or HFD-fed control and GLP-1R kd rats (n = 7/7/5/7; Two-way ANOVA group x diet ; group effect NS, diet effect P < 0.05 ; interaction effect P < 0.05; * indicates a significant difference with chow-fed control rats with Bonferroni post-hoc test ; HFD levels of ppg expression presented here are the same data presented in fig. 5G, to allow comparison with the age-matched chow fed control and GLP-1R kd rats) Abbreviations: AVP, arginine vasopressin; CRH, corticotropin releasing hormone; GLP-1R, Glucagon-like peptide-1; PPG, preproglucagon; TRH, thyrotropin releasing hormone

Chapter 6: General discussion

165

CHAPTER 6:

GENERAL DISCUSSION

Appendix 2: List of publications, presentations and awards

166

1 Overview of the main findings

In the manuscripts compiled in this thesis, we contributed to a better understanding of how

the gut communicates with the brain to control energy intake and energy expenditure. We

specifically investigated the role of the vagus nerve (VN), the main neuronal route of

communication between the gut and the brain, in mediating the effects of the glucagon-like

peptide-1 (GLP-1), a powerful incretin and satiating peptide.

An initial review of the available literature (CHAPTER 2) revealed several points that guided

our subsequent experimental approach:

1. Lesions of the VN suggested a role for vagal afferent neurons (VAN) in mediating GLP-1’s

effects on food intake and glycemia. The exact role of VAN GLP-1 receptors (GLP-1R) in

mediating the effects of endogenous intestinal GLP-1 was unknown and required a new

experimental approach. We set up a bilateral nodose ganglion (NG) injection technique to

deliver a lentiviral vector and knock down GLP-1R in VAN via RNA interference. Using this

approach, we showed that GLP-1R in VAN are necessary for the normal control of meal size,

gastric emptying and post-meal glycemia (CHAPTER 3).

2. Several neuropeptides have been identified in the NG and some of them, i.e., the melanin-

concentrating hormone (MCH) and the Cocaine- and Amphetamine-Regulated Transcript

(CART) were already implicated in the eating-inhibition induced by cholecystokinin (CCK).

Whether neuropeptides were involved in the vagally-mediated effects of GLP-1 was,

however, unknown. We screened neuropeptides in the NG for their transcriptional response

to GLP-1 in a primary VAN culture model and found that only CART was upregulated. We

further showed that GLP-1R activation induces VAN CART expression in vivo. Moreover,

using two models of loss of CART function, we demonstrated that a full CART expression in

VAN and CART signaling in the NTS are necessary for the full eating and gastric emptying

effects induced by intraperitoneal (IP) GLP-1. These results implicate for the first time CART

as a neuropeptide involved in the brainstem processing of the GLP-1 signal (CHAPTER 4).


167

The neuronal targets and the functional consequences of CART release in the NTS by VAN

remain to be further investigated.

3. The initial review of the literature did not comment on energy expenditure because the role

of the VN in the control of energy expenditure has been unexplored in the field until recently.

Moreover, whether peripheral GLP-1 can modulate energy expenditure also remained to be

determined, notably because the effects of GLP-1 on eating and gastric emptying were

confounding factors. We initially set up a protocol to investigate the effects of peripheral

GLP-1R activation on energy expenditure in wild-type rats while controlling for food intake

and gastric emptying. We found that peripheral GLP-1R activation induced a transient

decrease of whole-body energy expenditure, accompanied by a decrease in interscapular

brown adipose tissue (iBAT) temperature. We further tested whether a permanent

downregulation of GLP-1R in VAN was associated with changes in energy expenditure.

Indeed, during HFD feeding, when endogenous peripheral GLP-1 secretion is stimulated,

VAN GLP-1R KD prevented the HFD-induced reduction in energy expenditure presumably

by increasing BAT activity during the dark phase. This was sufficient to alleviate the HFD-

induced body weight gain and glucose intolerance. Together with a retrograde tracing from

the iBAT and central gene expression analysis, we proposed a new vagally-mediated

pathway by which endogenous peripheral GLP-1 tunes down iBAT activity after a meal

(CHAPTER 5).

Together, these results enhanced our understanding of the gut-to-brain GLP-1 signal. Many

questions, derived directly or indirectly from our results, have emerged and remain to be

answered. Some of the specific questions are already discussed separately in each chapter.

Broader questions, however, will be discussed in the following sections.


168

2 Established and potential non-vagal targets of in testinal GLP-1

in the control energy balance and glycemia

The results presented in this thesis clearly showed a role for VA in meditating the effects of

endogenous intestinal GLP-1 on both food intake and energy expenditure, which are the key

factors in maintaining energy balance. This, however, does not mean that the VN is the only

or most important route for the GLP-1 signal. Indeed, GLP-1 acts in a paracrine fashion but

also enters the systemic blood and lymph circulation (1). Thus, despite its degradation by

DPP-IV in the liver and the systemic circulation, it can have direct effects via the activation of

non-vagal peripheral GLP-1R. I will briefly discuss other non-vagal potential mechanisms of

GLP-1 effects on parameters implicated in the energy homeostasis.

Potential satiating effects of circulating GLP-1 v ia central GLP-1R 2.1

It is possible that circulating GLP-1 can bind to central GLP-1R. Indeed, GLP-1 may have

access to brain areas outside the blood-brain barrier (BBB) and may also diffuse through the

BBB to get access to other brain areas (2). As mentioned in CHAPTER 2, several

experiments aimed at mimicking the release of GLP-1 in the circulation (via injection into the

hepatic portal vein - HPV) supported the idea that the effect of GLP-1 on food intake may

also rely on a direct access to central GLP-1R (3), especially in the area postrema (AP) (4).

In support of this, injection of 125I-labeled GLP-1 in the aorta of anesthetized mice was found

in the subfornical organ and the AP (5). Pharmacological doses of the fluorescent-labeled

GLP-1R agonist liraglutide were shown to access both circumventricular and hypothalamic

areas (6). It is then possible that elevations of the GLP-1 levels in the HPV following a meal

can exert an eating inhibition by binding directly to GLP-1R in the brain, independently of

VAN. The direct activation of central GLP-1R, however, can also be an artifact induced by

intravenous (IV) injections of GLP-1 or an exclusive effect of pharmacological doses of GLP-

1R agonists and do not necessarily reflect the physiological mechanism by which

endogenous intestinal GLP-1 exerts its satiating effect.


169

Several approaches can be suggested to address the question whether endogenous

intestinal GLP-1 activates central GLP-1R in physiological conditions. First, based on

quantitative data about meal-induced GLP-1 secretion, degradation by DPP-IV, and

penetration through the BBB, a compartmental model can be established to estimate the

concentration of endogenous GLP-1 that may reach central GLP-1R (a classic approach in

nutrition or toxicity studies to estimate the concentration of a substance in an organ (7)).

More physiologically, GLP-1 can also be fluorescently labeled with a small fluorochrome

(such as “VivoTag”) and injected via catheters in the HPV or in the superior mesenteric

artery, at a dose and a rate that mimics the meal-induced release of GLP-1, before being

visualized in the brain. This approach, however, may present several caveats. First, similarly

to the study by Orskov and colleagues (5), it involves an IV injection, which might not

replicate the spread of endogenous GLP-1 in other compartments, such as the mesenteric

lymph, and raises the question of the adequate dose to simulate meal-induced GLP-1 levels.

Moreover, the attachment of a tag to GLP-1 may modify its kinetics, including its penetration

through the BBB. Alternatively, in mice with a CNS-specific deletion of ppg expression (that

abolishes central GLP-1 production), detecting GLP-1 (or GLP-1/GLP-1R interaction) in the

brain would give the indication that endogenous gut GLP-1 can access to central GLP-1R.

Eventually, an intestinal-specific modification of the ppg gene can lead to the secretion by the

gut of a GLP-1 molecule that differs from the central GLP-1 production (via the

insertion/deletion of a few amino acids without affecting GLP-1 kinetics) and the subsequent

identification of this modified GLP-1 molecule in the brain by proteomics.

Direct insulinotropic effects of GLP-1 in pancreat ic β-cells 2.2

For its glycemic effects, GLP-1 is thought to be in high enough concentration to exert its

insulinotropic effect on β-cells via a direct activation of the GLP-1R. The re-expression of the

human GLP-1R in the pancreatic islets and ductal cells of Glp-1r -/- mice was sufficient to

normalize the glucose and insulin responses to glucose tolerance tests (both oral and IP) (8).

This result strongly indicates that GLP-1R on the pancreatic β-cells largely mediate the


170

incretin effect of endogenous GLP-1. Our results showed a role for VA GLP-1R in amplifying

the early insulin response to a meal. We propose to conceptualize the insulinotropic effect of

endogenous intestinal GLP-1 as a collaboration between an early neuroincretin pathway

involving a gut-brain-pancreas vago-vagal reflex and an endocrine effect of GLP-1 on

pancreatic β-cells. Indeed, because the direct incretin effect of GLP-1 on β-cells is mild at

euglycemic levels (9), we can speculate that a neuronal activation of islet cells participates to

an early insulin production, before glycemia starts to rise significantly and before the direct

incretin effect of GLP-1 on β-cells intensifies. Consequently, this early insulin release can be

a sensitization signal for peripheral organs to activate the insulin-sensitive intracellular

pathways and thus, efficiently anticipate the elevation in glycemia. Further elevation of blood

glucose levels, then, may mainly recruit the direct incretin effect, as suggested by the above-

mentioned experiments (8).

Potential other GLP-1 targets in the control of en ergy balance and 2.3

glycemia

Several other GLP-1R populations may play a direct or indirect role in the control of energy

balance. GLP-1R expression has been reported in the dorsal root ganglia (DRG) (10): this

opens the possibility of another route for the GLP-1 signal to the brain involving spinal

afferents. Available studies, however, indicate a functional relevance of the DRG GLP-1R in

neuroprotection and pain sensitivity (11,12) rather than in the control of energy balance.

GLP-1R have also been detected in the enteric nervous system (ENS) of mice (10) and

primates (13). The clear interaction between GLP-1 and the ENS has not been shown. Some

reports, however, indicate that GLP-1 could play a local role on the regulation of gastric

emptying by controlling smooth muscle activation via a nitric oxide-dependent pathway (14).

Additionally, GLP-1R are expressed in both subcutaenous and visceral white adipose tissue

(ScWAT and vWAT) (15). GLP-1 has been shown to increase lipolytic processes in 3T3-L1

cells (15). Moreover, GLP-1R activation by GLP-1 or liraglutide increased the differentiation

and reduced the apoptosis of pre-adipocytes both in vitro and in vivo (16), thus favoring


171

adipose tissue hyperplasia. Interestingly, morbidly obese women showing hyperplasia in

ScWAT and vWAT showed a better glucose, insulin and lipid profiles compared to individuals

with adipose hypertrophy (17). Thus, it is possible to speculate that endogenous gut GLP-1

access to GLP-1R on white adipocytes to induce lipolysis and induce the differentiation of

new adipocytes, thus favoring adipose tissue hyperplasia. This idea, however, requires

further research.

In brief, based on our results and the available literature, we can say that VAN are the key

mediator of endogenous peripheral GLP-1’s effects on eating, gastric emptying, post-meal

glycemia and energy expenditure. On the other hand, circulating GLP-1 also participates in

the insulinotropic effects of GLP-1 in the pancreas and may access central GLP-1R to further

control food intake. Our data from the HFD-induced obese rats suggest that this apparent

role of VAN GLP-1R cannot be generalized and depends on the diet and the obesity status.

3 A role for vagal afferent neurons in the long-ter m control of

energy balance: potential implication in the develo pment of

obesity

As pointed out in CHAPTER 2, lesions of VAN resulted in acute effects on meal size when

animals were exposed to standard chow. Compensations in meal frequency, however,

prevented chronic changes in daily food intake and body weight, leading to the view that

VAN are only involved in the short-term control of food intake (18,19). As mentioned

previously, lesions studies are very limited in their interpretation. Based on our and other

recent findings, we ask the following questions: a/ How does the exposure to a HFD

modulate the apparent function of VAN GLP-1R? b/ Aside from GLP-1R, is there evidence

that VAN function is modulated during HFD exposure or obesity?


172

The chronic effect of VAN GLP-1R kd in the overall energy balance is 3.1

revealed by HFD challenge

In CHAPTER 3, we found that GLP-1R in VAN are necessary for normal satiation. The daily

energy intake, however, was similar between control and VAN GLP-1R kd rats, due to

compensations in meal number. Overall, the energy balance and, thus, body weight, was not

affected by the kd when rats were fed ad libitum with standard chow diet. We therefore

concluded that VAN GLP-1R had a role in the meal-to-meal control of food intake rather than

in the long-term control of the energy balance. In CHAPTER 5, however, we report that the

introduction of a high-fat diet (HFD) to VAN GLP-1R kd rats resulted in unexpected and

interesting phenotypes. First, the previously described changes in meal pattern (increased

meal size and duration) are no longer present after the introduction of a HFD. More

noticeable is a clear disruption in the energy balance: the energy expenditure was increased

in the dark phase in VAN GLP-1R kd rats with no changes in food intake, leading to a smaller

body weight gain than control rats after 8-9 weeks of HFD exposure. Together, these results

suggest that exposure to HFD changed the relevance of VAN GLP-1R effects in the chronic

regulation of energy balance. Potential mechanistic explanations of this phenomenon are in

part discussed in CHAPTER 5 and will be further discussed below.

First, the introduction of a 60% HFD may mask the control of meal size by VAN GLP-1R.

Indeed, there is a large body of evidence showing that the hedonic processes triggered by

the consumption of palatable food can result in caloric intake largely exceeding energy

requirements (see (20) for review). It is then plausible that the reward-related consumption of

palatable food rapidly overrides the homeostatic effects on meal size induced by VAN GLP-

1R activation, thus eliminating the differences in meal size between GLP-1R kd and control

rats. In addition, unpublished data of our laboratory (Rouault and Krieger, 2015) suggest that

VAN GLP-1R kd affect the consumption of palatable diet but this effect is short-lived. While

the mechanisms explaining this observation remain unknown, it would be interesting to find

the molecular switch counteracting the normal satiating process during exposure to a HFD.


173

Moreover, the introduction of HFD in a rodent model causes profound changes in the

intestinal secretion of GLP-1. Indeed, we found a 3-fold increase in HPV postprandial levels

of GLP-1 in rats exposed to HFD receiving a HFD meal compared to rats fed with chow

receiving an isocaloric chow test meal. We can speculate that the high GLP-1 secretion

during HFD exposure leads to a more pronounced activation of the VAN GLP-1R, thus

amplifying the effects of the kd on energy expenditure and BAT thermogenesis. Moreover, in

chow-fed rats, the exogenous administration of Ex-4 was sufficient to decrease energy

expenditure and BAT thermogenesis, likely via a vagal mechanism. In summary, these

findings support the idea that VAN GLP-1R mediate an effect of endogenous secretion of

GLP-1 on energy expenditure and BAT activity that is revealed by a HFD challenge.

The modulatory effects of HFD on the central GLP-1 system may provide another interesting

possibility regarding the chronic effects of VAN GLP-1R on energy balance. Indeed, the

decrease in NTS ppg expression induced by HFD exposure was greatly attenuated in VAN

GLP-1R kd rats. Because the central GLP-1 system has been shown to play a role in the

HFD-induced hyperphagia and fat accumulation (21), the decreased NTS ppg expression

would provide another possible explanation by which the VAN GLP-1R kd can exert a

chronic influence on the energy balance during HFD exposure. HFD-induced downregulation

of NTS ppg expression and its inverse relationship with body weight is, however, inconsistent

in the literature. Indeed, NTS ppg expression was strongly upregulated in obese Zucker rats,

independent of the diet (22), and was correlated with fat mass in HFD-exposed Long-Evans

rats (21). Moreover, a 2-week exposure to a HFD was sufficient to increase NTS ppg

expression in mice (23). Whereas NTS PPG neurons express the leptin receptor and ppg

expression is regulated by leptin in mice, there is no evidence for leptin receptor expression

and leptin regulation in rat PPG neurons, highlighting species differences in the central GLP-

1 system between rats and mice (24).

In addition, because intestinal GLP-1 secretion is elevated in our model of HFD-fed, the

HFD-induced downregulation of NTS ppg expression suggests a/ an inverse relationship


174

between peripheral and central GLP-1 production and b/ that VAN GLP-1R are necessary

mediators of the balance between peripheral and central GLP-1. Little is known about the

interplay between the central and peripheral GLP-1. As explained in our general introduction

(CHAPTER 1), the regulation of NTS PPG neurons by endogenous peripheral GLP-1 is

unlikely to happen directly (because NTS PPG neurons do not express the GLP-1R (25)) or

via a relay in the AP (because GLP-1R expressing AP neurons are catecholaminergic and

NTS PPG neurons do not receive catecholaminergic inputs (26)). Alternatively, PPG neurons

may be acutely modulated by VAN GLP-1R activation, because they receive direct

monosynaptic glutamatergic inputs from VAN (25). An acute IP Ex-4 injection, however, did

not induce c-Fos protein expression in the rat NTS GLP-1-expressing neurons (unpublished

observations, Lutz and Vrang). There is, however, indication that pharmacological doses of

IP Ex-4 (unpublished observations, Lee SJ, 2015) or the chronic administration of liraglutide

reduces the expression of ppg in the NTS (6). These observations are consistent with the

idea that a sustained activation of GLP-1R, including peripheral GLP-1R, may lead to a

reduction in NTS ppg expression, although peripheral GLP-1 does not acutely activate NTS

PPG neurons. A crosstalk between the high intestinal secretion of GLP-1 and the NTS ppg

expression in the context of a HFD, as well as the potential mediators of this crosstalk,

deserve further investigation. Indeed, interfering with this crosstalk during obesity may

maintain NTS ppg expression and reduce hyperphagia and fat accumulation (21).

In addition, it is plausible that the apparent function of VAN changes in an obesogenic

context following the modulation of lipid sensing pathways within VAN. Indeed, VAN express

nuclear receptors (NR) involved in fat metabolism, and HFD exposure modulates the

expression pattern of these NR in mice (27). The loss of these NR have been linked to an

increase in the thermogenic response to HFD exposure (27,28). In other words, the exposure

to HFD reduces the thermogenic power of VAN via NR signaling. How NR in the VAN

detects the presence of fat in the gut and how this translates into metabolic effects is

currently unknown. Based on our findings, it is interesting to speculate that VAN GLP-1R


175

activation may interact with this vagal lipid sensing and thus exerts its effect on energy

expenditure. This would explain why only HFD-exposed, but not chow-fed, VAN GLP-1R kd

rats showed changes in energy expenditure compared to their controls.

In conclusion, the role of VAN GLP-1R in the regulation of energy homeostasis is probably

context-specific. In brief, we can say that, in a physiological context, VAN GLP-1R activation

is short-lived and influences the acute post-meal control of eating and glycemia. In an

obesogenic context, however, the introduction of palatable food may modulate the apparent

role of VAN GLP-1R activation in the regulation of energy balance via the high peripheral

GLP-1 secretion, the decreased central ppg expression and the interaction with lipid sensing

mechanisms within VAN.

Evidence for a modulation of vagal afferent functi on in obesity 3.2

Beyond the GLP-1R signaling, the exposure to an obesogenic diet or obesity can modulate

vagal afferent function. There is evidence that the satiating effects of gut nutrients and gut

peptides are reduced in diet-induced obesity (DIO) (29-31). Moreover, maintenance on a

HFD attenuated the NTS response to the intestinal infusion of oleate (32) and to the injection

of CCK (33), as shown by a reduction in the number of neurons stained for the activity

marker c-Fos. This suggests that the VAN molecular pathways governing satiation are

altered during the development of diet-induced obesity. Indeed, maintenance on a HFD

decreases mechanosensory properties of gastric VAN (34,35), which may be related to the

modulatory effect of ghrelin on gastric VAN sensitivity to distension and contact. Moreover,

there is overwhelming evidence that the sensitivity of VAN to CCK, as well as other gut

peptides (e.g., bombesin) and serotonin, is reduced in diet-induced obese (DIO) rats and

mice (32,36-38). The reduced sensitivity of VA to mechanical signals and intestinal satiating

compounds coincides with a reduced excitability of the neuronal cell membrane (35), which

may lead to an overall decrease in action potential frequency. Additionally, the development

of leptin resistance in VAN has been proposed as a potential mechanism for the decreased

sensitivity of VAN to satiating signals. Indeed, VAN develop leptin resistance early in the


176

onset of obesity, which coincides with the onset of hyperphagia and the reduction of VAN

sensitivity to CCK (39). Moreover, conditional knockout of the leptin receptor in Nav1.8-

expressing neurons led to an uncompensated increase in meal size, higher body weight and

adiposity compared to control mice (40). This showed that the development of leptin

resistance in VAN can induce hyperphagia and weight gain. This is of particular interest

knowing the characterized interaction between the anorectic effects of leptin and GLP-1.

Indeed, IP leptin in rats strongly enhanced the acute eating-inhibition and weight loss

induced by IP GLP-1 or Ex-4 in rats (41,42). Moreover, reduced circulating leptin levels

induced by fasting (41) or leptin resistance induced by the maintenance on a HFD (31) were

both shown to reduce the anorectic effect of IP GLP-1. Hence, knowing the pivotal role of

VAN in mediating the GLP-1R-induced satiation, it is possible to speculate that the HFD-

induced leptin resistance in VAN drives a reduction of the anorectic effect of endogenous gut

GLP-1, thus favoring HFD-induced hyperphagia. Moreover, because leptin potentiates the

effects of CCK on CART abundance in VAN (43), it is likely that an interaction between VAN

leptin and GLP-1R signaling also happens via the regulation of CART expression.

In addition, the normal plasticity of VAN is lost after chronic exposure to a HFD. Indeed, in

physiological situations, the feeding status modulates the sensitivity of VAN to peripheral

signals: this is achieved by the modulation of the expression of receptors and neuropeptides

involved in satiation (see (44) for review). During chronic HFD exposure, the postprandial

rise of expression of Y2 receptors and CART, as well as the downregulation of MCH, its

receptor, and the cannabinoid 1 receptor, are blunted (39). Moreover, unpublished results

from our lab in collaboration with G. de Lartigue (2015) indicate that knocking down one of

the anorexigenic peptides, CART, in VAN, is sufficient to induce hyperphagia and an

increase in body weight gain in chow-fed rats. This further strengthens the idea that VAN

plasticity is necessary for the mediation of appropriate satiation signals to the brain and, thus,

for the long-term control of energy intake.


177

In summary, these studies confirm that VAN are indeed the relays of short-term satiation

signals, but they also indicate that a diet-induced alteration of VAN function can promote the

overconsumption of food and the development of an obese phenotype. Together with our

findings, this further suggests that VAN play a role in the regulation of energy balance that

depends on the diet and the adiposity status.

4 Putative evolution of the endogenous GLP-1 system during the

development of obesity and therapeutic implications

Our results (CHAPTER 3) highlighted the protective effects against HFD-induced metabolic

impairments conferred by the VAN GLP-1R kd. At first, these results seemed counter-

intuitive for us. Indeed, based on the known role of GLP-1 and our previous results

(CHAPTERS 3 & 4 ), interfering with normal satiation during HFD exposure should

exacerbate food intake and lead to an increase in body weight gain. Moreover, the

pharmacological administration of GLP-1R agonists, although not targeted only at VAN GLP-

1R, leads to a reduction in body weight and an improvement in glucose homeostasis in

obese patients. Several reports, however, were discordant with this rather simplistic view:

indeed, acute systemic blockade of the GLP-1R with Ex-9 decreased food intake in rats

exposed to a HFD (31). Moreover, glucose tolerance was improved in mice fed a HFD

injected for 4 weeks with Ex-9 (45), suggesting that decreased GLP-1R signaling could be

beneficial in some conditions. We will try to reconcile these findings by exploring the idea that

a high endogenous intestinal GLP-1 response to HFD exposure can be detrimental during

the development of obesity.

Our model tested the role of VAN GLP-1R in respons e to an obesogenic 4.1

diet, not in an established obese phenotype

First, in the experimental model used in CHAPTER 5, the VAN GLP-1R kd was established

prior to HFD exposure. This brings an important point to consider: we interfered with the gut-


178

brain communication during the dynamic phase in which the whole-body metabolism is

modulated by the introduction of a HFD rather than in a context of established obesity with

several established metabolic impairments. Thus, our model helps us to investigate the role

of peripheral endogenous GLP-1 acting on VAN in the development of HFD-induced obesity,

but does not answer the question whether VAN GLP-1R can alleviate or worsen an already-

established obese phenotype. Knocking down the GLP-1R in VAN of DIO rats would be an

interesting experiment to further investigate this point.

Intestinal GLP-1 secretion is amplified during the development of HFD-4.2

induced obesity

It is often reported that fasting and postprandial plasma GLP-1 levels are reduced in obese

animals (29,46) and humans (47,48) compared to their lean controls. We, however, have

found much higher meal-induced GLP-1 levels in HFD-maintained rats compared to chow-

maintained rats receiving an isocaloric amount of their respective diet as a test meal. This

apparent discrepancy may be a key to understand the role of endogenous GLP-1 in the

pathophysiology of obesity. First, we measured the meal-induced GLP-1 release after 10

weeks of exposure to a HFD. Others have found a similar upregulation of GLP-1 secretion in

rodents fed an obesogenic diet for up to 13 weeks (49,50) or in humans briefly exposed to a

high-fat diet (51). Together, these data suggest that HFD exposure results in an elevation of

endogenous GLP-1 secretion, due to the increase in fat and energy intake for an initial period

of at least 13 weeks in rat models (49,50). Fasting and meal-induced GLP-1 levels appear to

diminish in obese animals and humans and may be due to the development of leptin

resistance in L-cells (46).

Moreover, technical considerations may also have allowed us to highlight a clear difference

in meal-induced GLP-1 levels between chow-fed and HFD-fed animals. Indeed, in contrary to

other reports (46,49), we consistently sampled HPV blood rather than tail vein blood as a

more dynamic readout of GLP-1 secretion, and repeatedly did so in freely-moving animals. In

addition, we gave HFD-maintained rats a HFD test meal, i.e. a meal that obviously differs in


179

nutrient composition to the chow meal given to chow-maintained rats (although isocaloric)

rather than a similar test meal. We argue that this approach is more informative about the

endogenous GLP-1 levels at which rats are exposed during HFD feeding.

To summarize, based on our data as well as the literature, it is undisputable that GLP-1

secretion is affected by the introduction of a high fat content in the diet and must have an

important function in the development of obesity. The time course of the rise and fall of GLP-

1 during the development of obesity, however, needs to be further systematically

investigated in the future.

Putative effects of high intestinal GLP-1 levels o n glucose and lipid 4.3

metabolism

In our model, the repeated elevation of postprandial GLP-1 levels induced by hyperphagia

and by the nutrient composition of HFD meals is associated with a/ the tuning down of BAT

activity and whole-body energy expenditure and b/ the downregulation of the central GLP-1

system, which may contribute to hyperphagia and fat accumulation (52). In addition, we

cannot rule out that an increased GLP-1 secretion during HFD exposure exacerbates the

insulinotropic action of GLP-1 (including its neuroincretin component) and, therefore, may

have a direct effect on the morphology and gene expression pattern of the adipose tissues.

Indeed, the BATIRKO mouse model (BAT-specific deletion of the insulin receptor) showed a

reduced BAT cell size, likely due to reduced lipid accretion, and an elevated level of UCP1

compared to controls (53). Together, these results suggest that the initial elevation of

intestinal GLP-1 secretion upon HFD exposure can have detrimental consequences on

components of energy balance and glucose homeostasis by generating hyperinsulinemia

and altering insulin sensitivity.

To test this idea, a chronic, systemic, blockade of peripheral GLP-1R using GLP-1R

antagonist can be performed during the onset of diet-induced obesity. To our knowledge, this

experiment has not been specifically performed for the purpose of testing this hypothesis.

We can, however, learn from other experimental paradigms. Indeed, the Glp-1r KO mice is


180

protected from HFD-induced insulin resistance and, to some extent, body weight gain (in

females) (54). A similar phenotype has been observed in the double-incretin receptor KO

(DIRKO) mice (55). Moreover, chronic infusions with exendin-9 (Ex-9) in wild-type mice

exposed simultaneously to a HFD for one month block the HFD-increase in fasting glycemia

and improve oral glucose tolerance (45). In the same experiment, body weights were

unchanged in Ex-9-treated mice compared to their non-treated controls, but energy intake

was increased, suggesting an effect of the peripheral GLP-1R activation in the control of

energy expenditure. These results would support the hypothesis of a time window during

which HFD-induced elevated GLP-1 levels foster a modulation of energy expenditure and

glucose homeostasis (see Fig.1).

This idea remains, however, largely speculative and lacks further scientific support.

Moreover, the pharmacological use of a GLP-1R antagonist as a way to temper or slow the

development of an obese phenotype seems unrealistic. Indeed, obese animal models can be

experimentally generated by a diet manipulation, which allows testing the effects of a

pharmacological intervention in a controlled manner. In humans, identifying a period in which

high GLP-1 levels should be neutralized seems a difficult therapeutic strategy. In this regard,

the pharmacological use of GLP-1R agonists when obesity and diabetes are established is

an effective strategy to enhance β-cell insulin production and reduce appetite via central

GLP-1R (6). Nevertheless, measurements of endogenous GLP-1 levels could informative in

the diagnosis of pre-diabetics states. Hence, it is important to understand how GLP-1 can be

a contributing factor to fat accumulation in response to an obesogenic diet.

To summarize, our results presented in CHAPTER 5, which may appear counter-intuitive to

the conventional GLP-1 functions, shed new light on the contribution of intestinal

endogenous GLP-1 secretion to the development of an obese phenotype in a DIO model.

These results may suggest a benefit of blocking peripheral GLP-1R during the development

of obesity to slow weight gain and impairment of glucose homeostasis, but are by no means


181

contradictory to the beneficial effects of pharmacological GLP-1R agonists in established

obesity.

Figure 1: Proposed evolution of the secretion of en dogenous GLP-1 (peripheral and central) during the development of diet-induced obesity and summary of the phenotypes of VAN GLP-1R kd rats

5 GLP-1 and the vagus nerve, the future of obesity treatment?

The manuscripts in this thesis are part of a broader literature suggesting a role for the VN in

the control of energy balance. Hence, the VN can be seen as a therapeutic target against

obesity and its comorbidities. In this final section, we will discuss potential therapeutic

strategies that derive from our findings as well as current developing approaches targeting

the VN.

New pharmacological ways to access GLP-1 receptors on vagal afferents 5.1

First, as mentioned earlier, the currently used GLP-1R agonists are thought to reduce body

weight via a direct access to central GLP-1R. In other words, GLP-1R agonists do not require

the vagal route to produce the beneficial effects on body weight and appetite reduction (6).


182

Based on our results showing a role for GLP-1R in VAN to mediate satiation, recruiting the

vagal route may have an additional therapeutic advantage. Two possibilities (not exclusive)

can explain that GLP-1R agonists do not require an intact vagal route: a/ long-acting GLP-1R

gain access to peripheral and central GLP-1 but the effects of central GLP-1R activation is

largely predominant or b/ agonists do not gain access to GLP-1R on VAN because they are

injected subcutaneously for therapeutic usage. This second possibility leads to the idea of

delivering GLP-1 or GLP-1R agonists by an oral route, which would provide them access to

VAN in the intestinal mucosa. Orally administered GLP-1 is absorbed by the gut mucosa and

significantly elevates circulating GLP-1 levels in humans (56). In addition, oral GLP-1 affects

glucose homeostasis after an OGTT and the total meal energy intake in humans (57,58).

Moreover, the oral GLP-1R agonist Semaglutide (Novo Nordisk) has recently moved to a

phase 3a clinical trial (59), based on previous “encouraging results”. For oral delivery, GLP-1

and GLP-1R agonists must withstand the gastrointestinal mileu and were stabilized with

specific proprietary agent (SNAC or Eligen Carrier Technology ®). Nevertheless, the oral

dose needed to achieve comparable effect to the once-daily injectable semaglutide is likely

to be high (in the phase 3 trial, the injected does to be tested fall between 0.5 to 1 mg,

whereas oral doses range between 3 and 14 mg). The design of oral GLP-1R agonist was

primarily motivated by the desire to find an alternative to injectable drugs. It will however be

interesting to observe whether an oral delivery of GLP-1R agonists, that may recruit the

vagal route, have similar effects on body weight and glucose homeostasis to the injectable

versions or whether it can, indeed, enhance the effect.

A way to (partly) avoid enzymatic degradation is to have GLP-1 or an agonist secreted by a

host directly in the intestine. Indeed, bacteria and yeast have been engineered to secrete

GLP-1 and administered by gavage to rodent models of diabetes (60,61). In both cases, the

authors reported acute improvements of glucose homeostasis. Hence, using a micro-

organism as a vector to deliver GLP-1 in close proximity to their site of endogenous intestinal

release may represent another viable therapeutic strategy. Intriguingly, the same strategy


183

applied to a commensal bacterial strain engineered to secrete the non-insulinotropic form of

GLP-1, GLP-1(1-37), has been shown to reprogram epithelial cells in the intestine to produce

insulin (62). This unexpected source of insulin has been shown to reduce hyperglycemia in a

rat model of type 1 diabetes. In both cases, however, the therapeutic potential of these

approaches in humans might be legally limited by the fact that they involved the ingestion of

genetically-modified micro-organisms. Alternatively, the use of certain prebiotics, including

non digestible fermentable carbohydrate, has been shown to reduce appetite and improve

glucose tolerance in humans and correlate with increased post-prandial GLP-1 levels in

humans (63,64). The mechanisms are partially understood and may rely on the production of

short-chain fatty acids by colonic microbes and changes in the colonic microbial ecosystem.

In summary, delivering GLP-1 or an agonist orally or via an intestinal host may provide them

access to GLP-1R on VA and have some therapeutic benefits. Whether the effects of these

new therapeutic approaches are indeed mediated by VAN, and whether they will bring a

distinct therapeutic benefit to the currently-used GLP-1R injectable agonists remains to be

explored.

The promises of neuromodulation approaches 5.2

Alternatively, therapeutic approaches that target not only the GLP-1R signaling but the VN

itself may be successful to alleviate the development of obesity. Initially, results of vagotomy

or subdiaphragmatic deafferentation experiments suggested that blocking the VN signal

could acutely impact satiation, but could not affect body weight over the long-term, due to an

increase in meal number that compensated for the decrease in meal size. As we pointed out

in CHAPTER 2, results of lesion studies are difficult to analyze because most of the lesions

affect both VAN and VEN and may mask potential antagonist effects of different vagal fibers.

Despite these results, there is recently growing interest in the therapeutic potential of

another, rather unspecific, VN-targeting approach, i.e., VN neuromodulation.

Neuromodulation techniques can be divided into two apparently antagonists approaches.

First, VN stimulation (VNS) involves the implantation of electrodes on the VN to generate


184

electrical pulses and, thus, initiate action potentials. Second, VN blockade (VNB) uses

implantable electrodes in order to locally disturb the propagation of action potentials in

subdiaphragmatic VN branches. Both techniques present several advantages: first, they

provide the possibility to switch the device on or off based on the patient’s needs and to

individualize the characteristics of the currents applied. Moreover, the implantation of the

device is a relatively minor surgery compared to gastric bypass techniques. Finally, VNS is

already FDA-approved for the treatment of refractory epilepsy. So, are these VN-targeting

techniques the future of anti-obesity strategies, based on the available results in animals and

humans?

Initially, diverse VNS techniques produced a reduction in body weight and food intake in

dogs (65), pigs (66), rabbits (67,68) and rats (69,70) compared to their unstimulated controls.

More interestingly, similar techniques were shown to attenuate the weight gain induced by an

obesogenic diet in rats (71-73) and pigs (74). Retrospective studies in patients treated with

VNS for epilepsia or depression showed mixed results: two studies indicated that a large

proportion of the followed cohort showed weight loss or stable weight (75,76), whereas two

other studies found no effects of VNS on body weight (77,78). Interestingly, studies in

depressed patients with VNS indicated a change in the emotional response to sweets and

reduction in food cravings (79,80). These behavioral changes may correlate with and/or

explain the VNS-induced reduction in body weight measured in some studies. To summarize,

there is pre-clinical evidence suggesting that VNS can be an interesting therapeutic

approach to reduce obesity. The results of large clinical studies targeting obese patients are

required to further substantiate this claim.

Interestingly, VNB showed positive results to reduce eating and body weight in advanced

clinical trials. Indeed, in a first set of clinical study, VNB led to a sustained reduction in body

weight in obese patients that had not achieved satisfactory weight loss with behavioral

intervention or weight-loss drugs (81,82). Larger clinical trials confirmed these findings (83-

86).


185

In summary, VN neuromodulation approaches, and especially VNB, have shown clinical

potential to reduce body weight in obese patients. Mainly, a reduction in caloric intake has

been reported and seems to account for the loss of body weight. An effect of VNS on energy

expenditure may also contribute to the shift in energy balance (87). It is, however, difficult to

conceptualize that VNB can lead to a reduction in body weight via a decrease in food intake.

Indeed, our results and others have shown that VAN carry satiation signals and blocking

them via VNB should result in an increase in meal size and hyperphagia. There is evidence,

however, that some compounds that stimulate food intake, such as the fatty acid oxidation

inhibitor mercaptoacetate, act via an activation of VAN (88,89). Moreover, VNB also blocks

the electrical activity of VEN, which probably induces a reduction in gastric motility and

secretion, thus reducing nutrient absorption. Eventually, the effects of VNB on food intake

and body weights probably are the results of blocking both feeding stimulatory afferent

signals as well as inhibiting the efferent control of GI secretion and motility.

Moreover, the reason why VNB, an intermittent blockade of vagal signals, showed a greater

therapeutic potential than a permanent disruption of vagal signaling using subdiaphragmatic

vagotomy or SDA, is unclear. The chronic and constant application of an electrical current to

the VN may increase the risk of VN damage, as shown in dogs (90) and pigs (74), although

the risk appears to be limited. Intermittent strategies, however, may be a superior option to

prevent the emergence of stimulation-induced neuropathy and the compensatory

mechanisms associated with the absence of normal autonomic and sensory nerve functions.

Further research should clarify the mechanisms by which VN neuromodulation techniques

induce weight loss.

In addition, these neuromodulation techniques brought us to further speculate the cellular

and molecular mechanisms underlying the beneficial effects on body weight and appetite

control. Our study in CHAPTER 4, among others (see (91) for review), highlighted the role of

CART as a vagal neuropeptide required for the effects of GLP-1 on eating and gastric

emptying. How neuromodulation techniques affect the expression and release of vagal


186

neuropeptides is unkown. Moreover, in the case of diet-induced obesity, a reduced leptin

signaling is thought to be responsible for a phenotypic switch of VAN to an orexigenic state

(40,44), with low CART and high MCH expression. If neuromodulation in obese patients

restores a normal VAN function, i.e. normalizes neuropeptide expression in VAN, has not

been investigated. Probably, neuromodulation techniques that primarily trigger or block

action potentials in VN neurons may not fully recapitulate the satiation-related cellular and

molecular events induced by endogenous vagal chemosensing. A deeper understanding of

the role of the neuropeptides in the vagal control of energy balance may lead to the

development of new therapeutic approaches or complement existing strategies.

The use of recent neuroscience tools to further un derstand the role of 5.3

vagal afferents in the control of energy balance

The research on the VN functions can greatly benefit from newly developed neuroscience

tools to manipulate neuronal activity in a cell-specific manner. Indeed, a controlled activation

or inhibition of VAN is possible in animal models using optogenetics or DREADD (Designer

Receptors Exclusively Activated by Designer Drugs). To our knowledge, optogenetics has

recently been used in VAN to test the role of distinct neuronal populations in the control of

breathing (with unilateral light activation of VAN in the NG of anesthetized rat) (92).

Moreover, the use of the DREADD technique does not require a sophisticated optogenetics

setup and may also be useful to investigate the effects of activating or inhibiting a specific

VAN population. To my knowledge, so far no studies have used these established

neuroscience tools to study VAN function in the context of eating behavior. The development

of the bilateral NG injection technique during my thesis yields the possibility to inject viral

vectors bilaterally and thus apply these new techniques to the study of VAN functions.

Moreover, conditional (i.e., cre-dependent) expression of DREADD-activated receptors can

be used to specifically test the effect of a subgroup of VAN, such as those expressing a

particular marker or the VAN coming from a particular vagal branch. More recently, these

techniques were further utilized to specifically express light-activated receptors such as


187

channelrhodopsin or DREADD in neurons previously activated by a drug or an environmental

stimulus (using an immediate early gene-driven tagging by tetracycline and conditional

expression of the appropriate receptor (93,94)). This allows for the later reactivation of these

functionally-defined neurons by light (optogenetics) or CNO (DREADD). Such tools may be

helpful to provide the signature pattern of VAN activation during a meal. By manipulating this

signature pattern, we may reproduce or block nutrient-induced satiation and further study the

contribution of VAN to eating behavior. In summary, many innovative and promising clinical

approaches are targeting the GLP-1 system and/or the VN and are on the verge of becoming

real therapeutic alternatives to current treatments. Moreover, constant improvements in

neuronal mapping and in vivo control of neuronal activity allow to further investigate the

potential of VAN in regulating the energy balance.

Figure 2: Overview of current innovative therapeuti c and research strategies involving GLP-1 receptors on vagal afferents or neuromodulation of the vagus nerve

Clinically-tested strategies involve neuromodulation of the vagus nerve using stimulation or blockade and oral administration of GLP-1R agonists. Emerging approaches are based on the oral administration of prebiotics to increase the short-chain fatty acid-induced secretion of GLP-1 or the genetic engineering of commensal bacteria to secrete GLP-1. Current research may benefit from the use of cell-specific manipulation of VAN activity via DREADD or optogenetics. DREADD, Designer Receptors Exclusively Activated by Designer Drugs; GLP-1R : glucagon-like peptide-1 receptor; VA: vagal afferents; VN: vagus nerve


188

In conclusion, I can emphasize that the work presented in this dissertation is of particular

interest in the context of the development of new strategies against diabetes and the

associated metabolic diseases. Indeed, the development of treatments targeting the gut-

brain communication itself (VNS, VNB, …) requires a deeper understanding of the gut-brain

signals both at the functional and molecular levels. By dissecting the role of the VN in

mediating the effects of gut endogenous GLP-1 on eating, glycemia and energy expenditure,

I believe that this thesis work extends the current knowledge about the role of the gut-brain

axis in the regulation of the energy balance. Many important questions, however, arose from

this work and remained unanswered. The experimental tools developed during my thesis, as

well as the new approaches suggested in this dissertation, will allow to further improve our

understanding of the role of VAN in regulating the energy balance.


189

REFERENCES

1. D’Alessio D, Lu W, Sun W, Zheng S, Yang Q, Seeley R, Woods SC, Tso P. Fasting and postprandial concentrations of GLP-1 in intestinal lymph and portal plasma: evidence for selective release of GLP-1 in the lymph system. Am J Physiol Regul Integr Comp Physiol. 2007; 293(6):R2163–9.

2. Kastin AJ, Akerstrom V, Pan W. Interactions of glucagon-like peptide-1 (GLP-1) with the blood-brain barrier. J Mol Neurosci. 2002; 18(1-2):7–14.

3. Ruttimann EB, Arnold M, Hillebrand JJ, Geary N, Langhans W. Intrameal Hepatic Portal and Intraperitoneal Infusions of Glucagon-Like Peptide-1 Reduce Spontaneous Meal Size in the Rat via Different Mechanisms. Endocrinology. 2008; 150(3):1174–81.

4. Punjabi M, Arnold M, Rüttimann E, Graber M, Geary N, Pacheco-López G, Langhans W. Circulating glucagon-like peptide-1 (GLP-1) inhibits eating in male rats by acting in the hindbrain and without inducing avoidance. Endocrinology. 2014; 155(5):1690–9.

5. Orskov C, Poulsen SS, Moller M, Holst JJ. Glucagon-like peptide I receptors in the subfornical organ and the area postrema are accessible to circulating glucagon-like peptide 1. Diabetes. 1996; 45(6):832–5.


7. Cifelli CJ, Green JB, Green MH. Use of model-based compartmental analysis to study vitamin A kinetics and metabolism. Vitam Horm. 2007; 75:161–95.

8. Lamont BJ, Li Y, Kwan E, Brown TJ, Gaisano H, Drucker DJ. Pancreatic GLP-1 receptor activation is sufficient for incretin control of glucose metabolism in mice. J Clin Invest. 2012; 122(1):388–402.

9. Holst JJ. The Physiology of Glucagon-like Peptide 1. Physiological Reviews. 2007; 87(4):1409–39.

10. Richards P, Parker HE, Adriaenssens AE, Hodgson JM, Cork SC, Trapp S, Gribble FM, Reimann F. Identification and characterisation of glucagon-like peptide-1 receptor expressing cells using a new transgenic mouse model. Diabetes. 2014; 63(4):1224-33.

11. Jolivalt CG, Fineman M, Deacon CF, Carr RD, Calcutt NA. GLP-1 signals via ERK in peripheral nerve and prevents nerve dysfunction in diabetic mice. Diabetes, Obes Metab. 2011 Nov;13(11):990–1000.

12. Gong N, Xiao Q, Zhu B, Zhang CY, Wang YC, Fan H, Ma AN, Wang YX. Activation of Spinal Glucagon-Like Peptide-1 Receptors Specifically Suppresses Pain Hypersensitivity. J Neurosci. 2014; 34(15):5322–34.

13. Pyke C, Heller RS, Kirk RK, Orskov C, Reedtz-Runge S, Kaastrup P, Hvelplund A, Bardram L. GLP-1 receptor localization in monkey and human tissue: novel distribution revealed with extensively validated monoclonal antibody. Endocrinology. 2014; 155(4):1280–90.

14. Tolessa T, Gutniak M, Holst JJ, Efendic S, Hellström PM. Inhibitory effect of glucagon-


190

like peptide-1 on small bowel motility. Fasting but not fed motility inhibited via nitric oxide independently of insulin and somatostatin. J Clin Invest. 1998; 102(4):764–74.

15. Vendrell J, Bekay El R, Peral B, Garcia-Fuentes E, Megia A, Macias-Gonzalez M, Fernandez Real J, Jmenez Gomez Y, Escoté X, Pachon G, Simo R, Selva DM, Malagon MM, Tinahones FJ. Study of the potential association of adipose tissue GLP-1 receptor with obesity and insulin resistance. Endocrinology. 2011; 152(11):4072–9.

16. Challa TD, Beaton N, Arnold M, Rudofsky G, Langhans W, Wolfrum C. Regulation of Adipocyte Formation by GLP-1/GLP-1R Signaling. J Physiol. 2012 Feb; 287(9):6421–30.

17. Hoffstedt J, Arner E, Wahrenberg H, Andersson DP, Qvisth V, Löfgren P, Ryden M, Thöorne A, Wiren M, Palmer M, Thorell A, Toft E, Arner P. Regional impact of adipose tissue morphology on the metabolic profile in morbid obesity. Diabetologia. 2010; 53(12):2496–503.

18. Konturek SJ, Konturek JW, Pawlik T, Brzozowski T. Brain-gut axis and its role in the control of food intake. J Physiol Pharmacol. 2004; 55(1 Pt 2):137–54.


20. Kenny PJ. Reward mechanisms in obesity: new insights and future directions. Neuron. 2011; 69(4):664–79.


22. Vrang N, Larsen PJ, Jensen PB, Lykkegaard K, Artmann A, Larsen LK, Tang-Christensen M. Upregulation of the brainstem preproglucagon system in the obese Zucker rat. Brain Res. 2008; 1187:116–24.

23. Knauf C, Cani PD, Ait-Belgnaoui A, Benani A, Dray C, Cabou C, Colom A, Uldry M, Rastrelli S, Sabatier E, Godet N, Waget A, Pénicaud L, Valet P, Burcelin R. Brain Glucagon-Like Peptide 1 Signaling Controls the Onset of High-Fat Diet-Induced Insulin Resistance and Reduces Energy Expenditure. Endocrinology. 2008; 149(10):4768–77.

24. Lachey JL, D'Alessio DA, Rinaman L, Elmquist JK, Drucker DJ, Seeley RJ. The role of central glucagon-like peptide-1 in mediating the effects of visceral illness: differential effects in rats and mice. Endocrinology. 2005; 146(1):458–62.


26. Hisadome K, Reimann F, Gribble FM, Trapp S. CCK stimulation of GLP-1 neurons involves alpha1-adrenoceptor-mediated increase in glutamatergic synaptic inputs. Diabetes. 2011; 60(11):2701–9.



191


29. Duca FA, Swartz TD, Sakar Y, Covasa M. Decreased intestinal nutrient response in diet-induced obese rats: role of gut peptides and nutrient receptors. Int J Obes. 2013; 37(3):375-81.

30. Duca FA, Sakar Y, Covasa M. Combination of obesity and high-fat feeding diminishes sensitivity to GLP-1R agonist, Exendin-4. Diabetes. 2013; 62(7):2410-5.

31. Williams DL, Hyvarinen N, Lilly N, Kay K, Dossat A, Parise E, Torregossa AM. Maintenance on a high-fat diet impairs the anorexic response to glucagon-like-peptide-1 receptor activation. Physiol Behav. 2011; 103(5):557–64.

32. Covasa M, Grahn J, Ritter RC. Reduced hindbrain and enteric neuronal response to intestinal oleate in rats maintained on high-fat diet. Auton Neurosci - Basiv. 2000; 84(1-2):8–18.

33. Covasa M, Grahn J, Ritter RC. High fat maintenance diet attenuates hindbrain neuronal response to CCK. Regul Pept. 2000; 86(1–3):83–8.

34. Kentish S, Li H, Philp LK, O’Donnell TA, Isaacs NJ, Young RL, Wittert GA, BlackshawLA, Page AJ. Diet induced adaptation of vagal afferent function. J Physiol. 2012; 590(Pt 1):290-21.

35. Daly DM, Park SJ, Valinsky WC, Beyak MJ. Impaired intestinal afferent nerve satiety signalling and vagal afferent excitability in diet induced obesity in the mouse. J Physiol. 2011; 589(Pt 11):2857–70.

36. Duca FA, Zhong L, Covasa M. Reduced CCK signaling in obese-prone rats fed a high fat diet. Horm Behav. 2013; 64(5):812–7.

37. Savastano DM, Covasa M. Adaptation to a high-fat diet leads to hyperphagia and diminished sensitivity to cholecystokinin in rats. J Nutrition. 2005; 135(8):1953–9.

38. Swartz TD, Savastano DM, Covasa M. Reduced sensitivity to cholecystokinin in male rats fed a high-fat diet is reversible. J Nutrition. 2010; 140(9):1698–703.

39. de Lartigue G, Barbier de la Serre C, Espero E, Lee J, Raybould HE. Leptin Resistance in Vagal Afferent Neurons Inhibits Cholecystokinin Signaling and Satiation in Diet Induced Obese Rats. PLoS ONE. 2012; 7(3):e32967.

40. de Lartigue G, Ronveaux CC, Raybould HE. Deletion of leptin signaling in vagal afferent neurons results in hyperphagia and obesity. Mol Metab. 2014; 3(6):595–607.

41. Williams DL, Baskin DG, Schwartz MW. Leptin regulation of the anorexic response to glucagon-like peptide-1 receptor stimulation. Diabetes. 2006; 55(12):3387–93.

42. Bojanowska E, Nowak A. Interactions between leptin and exendin-4, a glucagon-like peptide-1 agonist, in the regulation of food intake in the rat. J Physiol Pharmacol. 2007; 58(2):349–60.

43. de Lartigue G, Lur G, Dimaline R, Varro A, Raybould H, Dockray GJ. EGR1 Is a Target for Cooperative Interactions between Cholecystokinin and Leptin, and Inhibition by Ghrelin, in Vagal Afferent Neurons. Endocrinology. 2010; 151(8):3589–99.


192

44. de Lartigue G, la Serre de CB, Raybould HE. Vagal afferent neurons in high fat diet-induced obesity; intestinal microflora, gut inflammation and cholecystokinin. Physiol Behav. 2011; 105(1):100–5.

45. Knauf C, Cani PD, Ait-Belgnaoui A, Benani A, Dray C, Cabou C, Colom A, Uldry M, Rastrelli S, Sabatier E, Godet N, Waget A, Pénicaud L, Valet P, Burcelin R. Brain Glucagon-Like Peptide 1 Signaling Controls the Onset of High-Fat Diet-Induced Insulin Resistance and Reduces Energy Expenditure. Endocrinology. 2008; 149(10):4768–77.

46. Anini Y, Brubaker PL. Role of Leptin in the Regulation of Glucagon-Like Peptide-1 Secretion. Diabetes. 2003; 52(2):252–9.

47. Ranganath LR, Beety JM, Morgan LM, Wright JW, Howland R, Marks V. Attenuated GLP-1 secretion in obesity: cause or consequence? Gut. 1996; 38(6):916–9.

48. Verdich C, Toubro S, Buemann B, M L, Lysgard Madsen J, Holst JJ, Astrup A. The role of postprandial releases of insulin and incretin hormones in meal-induced satiety - effect of obesity and weight reduction. Int J Obes Relat Metab Disorg. 2001; 225(8):1206–14.

49. Nakajima S, Hira T, Hara H. Postprandial glucagon-like peptide-1 secretion is increased during the progression of glucose intolerance and obesity in high-fat/high-sucrose diet-fed rats. Br J Nutr. 2015; 113(9):1477–88.

50. Wang F, Yoder SM, Yang Q, Kohan AB, Kindel TL, Wang J, Tso P. Chronic High-Fat Feeding Increases Mixed Meal-Induced Incretin Secretion in Sprague-Dawley Rats. Am J Physiol Gastrointest Liver Physiol. 2015; 309(10):G807-15..

51. Numao S, Kawano H, Endo N, Yamada Y, Konishi M, Takahashi M, Sakamoto S. Short-term low carbohydrate/high-fat diet intake increases postprandial plasma glucose and glucagon-like peptide-1 levels during an oral glucose tolerance test in healthy men. Eur J Clin Nutr. 2012; 66(8):926–31.

52. Barrera JG, D'Alessio DA, Drucker DJ, Woods SC, Seeley RJ. Differences in the central anorectic effects of glucagon-like peptide-1 and exendin-4 in rats. Diabetes. 2009; 58(12):2820–7.

53. Guerra C, Navarro P, Valverde AM, Arribas M, Brüning J, Kozak LP, Kahn CR, Benito M. Brown adipose tissue–specific insulin receptor knockout shows diabetic phenotype without insulin resistance. J Clin Invest. 2001;108(8):1205–13.

54. Ayala JE, Bracy DP, James FD, Burmeister MA, Wasserman DH, Drucker DJ. Glucagon-Like Peptide-1 Receptor Knockout Mice Are Protected from High-Fat Diet-Induced Insulin Resistance. Endocrinology. 2010; 151(10):4678–87.

55. Ayala JE, Bracy DP, Hansotia T, Flock G, Seino Y, Wasserman DH, Drucker DJ. Insulin action in the double incretin receptor knockout mouse. Diabetes. 2008; 57(2):288–97.

56. Beglinger C, Poller B, Arbit E, Ganzoni C, Gass S, Gomez-Orellana I, Drewe J. Pharmacokinetics and pharmacodynamic effects of oral GLP-1 and PYY3-36: a proof-of-concept study in healthy subjects. Clin Pharmacol Ther. 2008; 84(4):468–74.

57. Steinert RE, Poller B, Castelli MC, Friedman K, Huber AR, Drewe J, Beglinger C. Orally administered glucagon-like peptide-1 affects glucose homeostasis following an


193

oral glucose tolerance test in healthy male subjects. Clin Pharmacol Ther. 2009; 86(6):644–50.

58. Steinert RE, Poller B, Castelli MC, Drewe J, Beglinger C. Oral administration of glucagon-like peptide 1 or peptide YY 3-36 affects food intake in healthy male subjects. Am J Clin Nutr. 2010; 92(4):810–7.

59. Novo Nordisk to initiate phase 3a development of oral semaglutide, a once-daily oral GLP-1 analogue [article online]. Available from : https://www.novonordisk.com/bin/getPDF.1947638.pdf. 2015, accessed on Dec 2015.

60. Agarwal P, Khatri P, Billack B, Low W-K, Shao J. Oral delivery of glucagon like peptide-1 by a recombinant Lactococcus lactis. Pharm Res. 2014; 31(12):3404–14.

61. Ri W, Chao M, Xiaodan L, Huikun D, Yanli J, Yu W, Pingzhe J, Haisong W, Peipei T, Miao L, Gangang N, Baicheng M, Minggang L. Construction of yeast strains expressing long-acting glucagon-like peptide-1 (GLP-1) and their therapeutic effects on type 2 diabetes mellitus mouse model. Yi Chuan. 2015; 37(2):183–91.

62. Duan FF, Liu JH, March JC. Engineered commensal bacteria reprogram intestinal cells into glucose-responsive insulin-secreting cells for the treatment of diabetes. Diabetes. 2015; 64(5):1794–803.

63. Cani PD, Lecourt E, Dewulf EM, Sohet FM, Pachikian BD, Naslain D, De Backer F, Neyrinck AM, Delzenne NM. Gut microbiota fermentation of prebiotics increases satietogenic and incretin gut peptide production with consequences for appetite sensation and glucose response after a meal. Am J Clin Nutr. 2009; 90(5):1236–43.

64. McCarty MF, DiNicolantonio JJ. Acarbose, lente carbohydrate, and prebiotics promote metabolic health and longevity by stimulating intestinal production of GLP-1. Open Heart. 2015; 2(1):e000205.

65. Roslin M, Kurian M. The Use of Electrical Stimulation of the Vagus Nerve to Treat Morbid Obesity. Epilepsy Behav. 2001; 2(3, Supplement):S11–6.

66. Matyja A, Thor PJ, Sobocki J, Laskiewicz J, Kekus J, Tuz R, Koczanowski J, Zaraska W. Effects of vagal pacing on food intake and body mass in pigs. Folia Med Cracov. 2004; 45(3-4):55–62.

67. Sobocki J, Thor PJ, Uson J, Diaz-Guemes I, Lipinski M, Calles C, Pascual S. Microchip vagal pacing reduces food intake and body mass. Hepatogastroenterol. 2001; 48(42):1783–7.

68. Sobocki J, Thor P, Krolczyk G, Uson J, Diaz-Guemes I, Lipinski M. The cybergut. An experimental study on permanent microchip neuromodulation for control of gut function. Acta Chir Belg. 2002; 102(2):68–70.

69. Krolczyk G, Zurowski D, Sobocki J, Słowiaczek MP, Laskiewicz J, Matyja A, Zaraska K, Zaraska W, Thor PJ. Effects of continuous microchip (MC) vagal neuromodulation on gastrointestinal function in rats. J Physiol Pharmacol. 2001; 52(4 Pt 1):705–15.

70. Sobocki J, Fourtanier G, Estany J, Otal P. Does vagal nerve stimulation affect body composition and metabolism? Experimental study of a new potential technique in bariatric surgery. Surgery. 2006; 139(2):209–16.


194

71. Bugajski AJ, Gil K, Ziomber A, Zurowski D, Zaraska W, Thor PJ. Effect of long-term vagal stimulation on food intake and body weight during diet induced obesity in rats. J Physiol Pharmacol. 2007; 58 Suppl 1:5–12.

72. Gil K, Bugajski A, Kurnik M, Thor P. Chronic vagus nerve stimulation reduces body fat, blood cholesterol and triglyceride levels in rats fed a high-fat diet. Folia Med Cracov. 2012; 52(3-4):79–96.

73. Gil K, Bugajski A, Thor P. Electrical vagus nerve stimulation decreases food consumption and weight gain in rats fed a high-fat diet. J Physiol Pharmacol. 2011; 62(6):637–46.

74. Val-Laillet D, Biraben A, Randuineau G, Malbert CH. Chronic vagus nerve stimulation decreased weight gain, food consumption and sweet craving in adult obese minipigs. Appetite. 2010 Sep 30;55(2):245–52.

75. Burneo JG, Faught E, Knowlton R, Morawetz R, Kuzniecky R. Weight loss associated with vagus nerve stimulation. Neurology. 2002; 59(3):463–4.

76. Pardo JV, Sheikh SA, Kuskowski MA, Surerus-Johnson C, Hagen MC, Lee JT, Rittberg BR, Adson DE. Weight loss during chronic, cervical vagus nerve stimulation in depressed patients with obesity: an observation. Int J Obes. 2007; 31(11):1756–9.

77. Koren MS, Holmes MD. Vagus nerve stimulation does not lead to significant changes in body weight in patients with epilepsy. Epilepsy Behav. 2006; 8(1):246–9.

78. Rush AJ, Sackeim HA, Marangell LB, George MS, Brannan SK, Davis SM, Lavori P, Howland R, Kling MA, Rittberg B, Carpenter L, Ninan P, Moreno F, Schwartz T, Conway C, Burke M, Barrry JJ. Effects of 12 months of vagus nerve stimulation in treatment-resistant depression: a naturalistic study. Biol Psychiatry. 2005; 58(5):355–63.

79. Bodenlos JS, Kose S, Borckardt JJ, Nahas Z, Shaw D, O'Neil PM, George MS. Vagus nerve stimulation acutely alters food craving in adults with depression. Appetite. 2007; 48(2):145–53.

80. Bodenlos JS, Kose S, Borckardt JJ, Nahas Z, Shaw, O'Neil PM, Pagoto SL, George MS. Vagus nerve stimulation and emotional responses to food among depressed patients. J Diabetes Sci Technol. 2007; 1(5):771–9.

81. Camilleri M, Toouli J, Herrera MF, Kulseng B, Kow L, Pantoja JP, Marvik R, Johnsen G, Billington CJ, Moody FG, Knudson MB, Tweden KS, Vollmer M, Wilson RR, Anvari M. Intra-abdominal vagal blocking (VBLOC therapy): clinical results with a new implantable medical device. Surgery. 2008; 143(6):723–31.

82. Camilleri M, Toouli J, Herrera MF, Kow L, Pantoja JP, Billington CJ, Tweden KS, Wilson RR, Moody FG. Selection of electrical algorithms to treat obesity with intermittent vagal block using an implantable medical device. Surg Obes Relat Dis. 2009; 5(2):224–9.

83. Sarr MG, Billington CJ, Brancatisano R, Brancatisano A, Toouli J, Kow L, Nguyen NT, Blackstone R, Maher JW, Shikora S, Reeds DN, Eagon JC, Wolfe BM, O'Rourke RW, Fujioka K, Takata M, Swain JM, Morton JM, Ikramuddin S, Schweitzer M, Chand B, Rosenthal R; EMPOWER Study Group. The EMPOWER study: randomized, prospective, double-blind, multicenter trial of vagal blockade to induce weight loss in


195

morbid obesity. Obes Surg. 2012 Nov;22(11):1771–82.

84. Ikramuddin S, Blackstone RP, Brancatisano A, Toouli J, Shah SN, Wolfe BM, Fujioka K, Maher JW, Swain J, Que FG, Morton JM, Leslie DB, Brancatisano R, Kow L, O'Rourke RW, Deveney C, Takata M, Miller CJ, Knudson MB, Tweden KS, Shikora SA, Sarr MG, Billington CJ. Effect of reversible intermittent intra-abdominal vagal nerve blockade on morbid obesity: the ReCharge randomized clinical trial. JAMA. 2014; 312(9):915–22.

85. Shikora SA, Toouli J, Herrera MF, Kulseng B, Brancatisano R, Kow L, Pantoja JP, Johnsen G, Brancatisano A, Tweden KS, Knudson MB, Billingto CJ. Intermittent Vagal Nerve Block for Improvements in Obesity, Cardiovascular Risk Factors, and Glycemic Control in Patients with Type 2 Diabetes Mellitus: 2-Year Results of the VBLOC DM2 Study. Obes Surg. 2015; Epub 10.1007/s11695-015-1914-1.

86. Shikora SA, Wolfe BM, Apovian CM, Anvari M, Sarwer DB, Gibbons RD, Ikramuddin S, Miller CJ, Knudson MB, Tweden KS, Sarr MG, Billington CJ. Sustained Weight Loss with Vagal Nerve Blockade but Not with Sham: 18-Month Results of the ReCharge Trial. J Obes. 2015; 365604:1-8.


88. Lutz TA, Diener M, Scharrer E. Intraportal mercaptoacetate infusion increases afferent activity in the common hepatic vagus branch of the rat. Am J Physiol Regul Integr Comp Physiol. 1997; 273(1):R442–5.

89. Brandt K, Arnold M, Geary N, Langhans W, Leonhardt M. Vagal afferents mediate the feeding response to mercaptoacetate but not to the beta-3 adrenergic receptor agonist CL 316,243. Neurosci Lett. 2007; 411(2):104–7.

90. Cohen ML, Georgievskaya Z. Histopathology of the stimulated Vagus nerve: primum non nocere. Heart Fail Rev. 2011; 16(2):163–9.

91. de Lartigue G. Putative role of neuropeptides in vagal afferent signaling. Physiol Behav. 2014; 136:155-69.

92. Chang RB, Strochlic DE, Williams EK, Umans BD, Liberles SD. Vagal Sensory Neuron Subtypes that Differentially Control Breathing. Cell. 2015; 161(3):622–33.

93. Ramirez S, Liu X, Lin P-A, Suh J, Pignatelli M, Redondo RL, Ryan TJ, Tonegawa S. Creating a false memory in the hippocampus. Science. 2013; 341(6144):387–91.

94. Zhang Z, Ferretti V, Güntan İ, Moro A, Steinberg EA, Ye Z, Zecharia AY, Yu X, Vyssotski AL, Brickley SG, Yustos R, Pillidge ZE, Harding EC, Wisden W, Franks NP. Neuronal ensembles sufficient for recovery sleep and the sedative actions of alpha-2 adrenergic agonists. Nat Neurosci. 2015; 18(4):553–61.

Date post:	04-Aug-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Rights / License: Research Collection In Copyright - Non ......Dr. Matthew Hayes, co-examiner Dr....

Documents