RA

PT

*D

Bw(gWc(abt(rqaesewmAbwewsdatcbwoona

IaTlam

GASTROENTEROLOGY 2008;135:1267–1276

ole of Ghrelin in the Relationship Between Hyperphagia andccelerated Gastric Emptying in Diabetic Mice

IETER–JAN VERHULST,* BETTY DE SMET,* INGE SAELS,* THEO THIJS,* LUC VER DONCK,‡ DIEDER MOECHARS,‡

HEO L. PEETERS,* and INGE DEPOORTERE*

Centre for Gastroenterological Research, Catholic University of Leuven, Leuven, Belgium; and ‡Johnson & Johnson Division of Pharmaceutical Research and

evelopment, Janssen Pharmaceutica, Beerse, Belgium

sipbwlrtiiitlwppistuhktstcuGb

tanaic

rmm

BA

SIC–

ALI

MEN

TARY

TRA

CT

ackground & Aims: Ghrelin is an orexigenic peptideith gastroprokinetic effects. Mice with streptozotocin

STZ)-induced diabetes exhibit hyperphagia, alteredastric emptying, and increased plasma ghrelin levels.

e investigated the causative role of ghrelin herein byomparing changes in ghrelin receptor knockoutgrowth hormone secretagogue receptor [GHS-R]�/�)nd wild-type (GHS-R�/�) mice with STZ-induced dia-etes. Methods: Gastric emptying was measured with

he [13C]octanoic acid breath test. The messenger RNAmRNA) expression of neuropeptide Y (NPY), agouti-elated peptide (AgRP), and proopiomelanocortin wasuantified by real-time reverse-transcription polymer-se chain reaction. Neural contractions were elicited bylectrical field stimulation in fundic smooth muscletrips. Results: Diabetes increased plasma ghrelin lev-ls to a similar extent in both genotypes. Hyperphagiaas more pronounced in GHS-R�/� than in GHS-R�/�

ice between days 12 and 21. Increases in NPY andgRP mRNA expression were less pronounced in dia-etic GHS-R�/� than in GHS-R�/� mice from day 15 on,hereas decreases in proopiomelanocortin mRNA lev-

ls were similar in both genotypes. Gastric emptyingas accelerated to a similar extent in both genotypes,

tarting on day 16. In fundic smooth muscle strips ofiabetic GHS-R�/� and GHS-R�/� mice, neuronal relax-tions were reduced, whereas contractions were increased;his increase was related to an increased affinity of mus-arinic and tachykinergic receptors. Conclusions: Dia-etic hyperphagia is regulated by central mechanisms inhich the ghrelin-signaling pathway affects the expressionf NPY and AgRP in the hypothalamus. The accelerationf gastric emptying, which is not affected by ghrelin sig-aling, is not the cause of diabetic hyperphagia and prob-bly involves local contractility changes in the fundus.

n 1996, the growth hormone secretagogue receptor(GHS-R) was cloned and identified as the receptor forfamily of synthetic growth hormone secretagogues.1

he endogenous ligand of this receptor, ghrelin, an acy-ated 28-amino acid peptide, was identified 3 years later2

nd, as expected, was shown to stimulate growth hor-

one release. Because ghrelin is mainly produced by the

tomach, an organ well positioned to detect recentlyngested food, it was soon demonstrated that ghrelinlays an important role in the regulation of the energyalance. Peripheral administration of ghrelin induceseight gain by decreasing fat utilization3 and by stimu-

ating food intake in rodents4 as well as in humans.5 Aole for ghrelin in meal initiation was also confirmed byhe fluctuations in plasma ghrelin levels, which increasemmediately before a meal and rapidly fall after foodntake, following a 24-hour pattern reciprocal to that ofnsulin.6 This raises the question whether insulin nega-ively regulates ghrelin. Indeed, increased plasma ghrelinevels were observed during insulin deficiency in rodentsith streptozotocin (STZ)-induced diabetes.7,8 Also inatients with type 1 diabetes, absolute insulin deficiencyrevented prandial plasma ghrelin suppression until the

nsulin deficiency was corrected with an intravenous in-ulin bolus.9 In both conditions, the hyperghrelinemia ishought to play a role in the hyperphagia associated withncontrolled type 1 diabetes. Important evidence for thisypothesis was provided in a study with diabetic ghrelinnockout mice.8 In the absence of its endogenous ligand,he ghrelin receptor still maintains an important level ofignaling, and it is unclear to what extent this contributeso the hyperphagia.10 The physiologic importance of thisonstitutive activity in humans was demonstrated in 2nrelated families harboring a missense mutation in theHS-R, which leads to a syndrome characterized not onlyy short stature but also by obesity.11

Ghrelin stimulates food intake through activation ofhe ghrelin receptor present on neuropeptide Y (NPY)/gouti-related peptide (AgRP) neurons in the arcuateucleus either by crossing the blood/brain barrier or viactivation of vagal nerve activity.5,12–14 Ghrelin also hasmportant effects on gastrointestinal motility that mayontribute to appetite signaling. It induces strong “hun-

Abbreviations used in this paper: ACh, acetylcholine; AgRP, agouti-elated peptide; EFS, electrical field stimulation; GHS-R, growth hor-one secretagogue receptor; NPY, neuropeptide Y; POMC, proopio-elanocortin; STZ, streptozotocin.

© 2008 by the AGA Institute0016-5085/08/$34.00

doi:10.1053/j.gastro.2008.06.044

gstuc

wsitdwotaigtdawtfm

tow

RvddCv

mcaipstbmflk(iGlB

ASIC

–A

LIMEN

TARY

TRA

CT

1268 VERHULST ET AL GASTROENTEROLOGY Vol. 135, No. 4

er” contractions in the fasted state originating in thetomach and migrating distally15,16 and accelerates gas-ric emptying17–20 in rodents and man. Until now, it isnclear to what extent the prokinetic effects of ghrelinontribute to its effects on food intake.

The aim of our study was therefore multiple. First,e determined the functional relevance of the ghrelin-

ignaling pathway in diabetic hyperphagia by eliminat-ng the constitutive activity of the ghrelin receptor. Tohis end, we compared food intake in STZ-inducediabetic ghrelin receptor knockout (GHS-R�/�) andild-type (GHS-R�/�) mice. Second, the contributionf a central pathway in the ghrelin-dependent regula-ion of hyperphagia was determined by comparinglterations in the expression of orexigenic and anorex-genic neuropeptides in the hypothalamus of bothenotypes. Third, we investigated whether altered gas-ric emptying, known to occur in rats with STZ-in-uced diabetes,21,22 may contribute to hyperphagia,nd we determined the role of ghrelin herein. Finally,e studied the effect of diabetes on local alterations in

he in vitro contractility of smooth muscle strips of theundus to elucidate whether peripheral mechanisms

ediate the changes in gastric emptying.This multiparametric analysis should help us reveal

he role of the ghrelin-signaling pathway in hyperphagiaf mice with uncontrolled diabetes. It should also clarifyhether central or peripheral pathways are involved.

Materials and MethodsAnimalsMale (40 –50 weeks of age) GHS-R�/� and GHS-

�/� mice were housed in a temperature-controlled en-ironment (20°C–22°C) under a 14-hour:10-hour light-ark cycle and had ad libitum access to food andrinking water. This research was approved by the Ethicalommittee for Animal Experiments of the Catholic Uni-ersity of Leuven.

Generation of GHS-R�/� MiceGHS-R�/� mice were developed by Janssen Phar-

aceutica (Beerse, Belgium) in collaboration with Lexi-on Genetics, Inc (The Woodlands, TX). With a polymer-se chain reaction (PCR) probe, genomic clones weresolated by screening of the 129SvEvBrd-derived lambdaKOS genomic library.23 A 9.5-kilobase genomic clonepanning exon 1 and exon 2 was used to generate theargeting vector via yeast-mediated homologous recom-ination. In this vector, a 1325-base pair genomic frag-ent, spanning exon 1 and exon 2, was replaced by a

oxed version of exon 1 and exon 2 including a 1.7-ilobase PGK-neo selection cassette flanked by 2 Frt sitesFigure 1). The NotI-linearized vector was electroporatednto 129 Sv/Evbrd (LEX1) embryonic stem (ES) cells, and

418-fialuridine (FIAU)-resistant ES cell clones were iso-ated and analyzed for homologous recombination by

Figure 1. Targeted disruption of theGHS-R1 gene. (A) Structure of the tar-geting vector, wild-type locus, tar-geted locus, and Cre-excised locus.Boxes representing the exons in openand shaded boxes are the noncodingand coding regions, respectively. Ar-rows indicate the position of the PCRprimers used for genotyping the wild-type and targeted allele. (B) Expres-sion of the GHS-R1 transcript in brainand pituitary of both genotypes. Re-sults are means � SEM (n � 6) relativeexpression levels after normalization

to �-actin.

SjigaCflwGcwftsqr

wtsfSmdueaGa

anbdARsraaRra

m(a(agtfia1

bcTwMIsdecev

pbm1

I

Rf

T

A

B

C

BA

SIC–

ALI

MEN

TARY

TRA

CT

October 2008 ROLE OF GHRELIN IN DIABETIC MICE 1269

outhern blot analysis. Targeted ES cell clones were in-ected into C57BL/6 (albino) blastocysts, and the result-ng chimeras were mated to C57BL/6 (albino) females toenerate animals heterozygote for the floxed GHS-R1llele. These were subsequently crossed with Protamine-re mice.24 Male descendants heterozygote for both theoxed GHS-R1 allele and the Protamine Cre transgeneere crossed to C57Bl/6 females to obtain heterozygoteHS-R1 knockout animals. These were subsequently

rossed to generate the genotypes used in the study. PCRas used to screen genotypes by using DNA isolated

rom mouse tail biopsy samples. Real-time reverse-ranscription (RT) PCR analysis was used to show expres-ion or absence of the GHS-R1 transcript. Primers se-uences are summarized in Table 1, sections A and B,espectively.

Experimental DesignGHS-R�/� (n � 26) and GHS-R�/� mice (n � 22)

ere fasted overnight (14 hours) and injected intraperi-oneally with 160 mg/kg STZ dissolved in 5 mmol/Lodium citrate buffer (pH 4.0). Body weight, 24-hourood intake, and tail-blood glucose levels (Accu-Chekensor Comfort; Roche Diagnostics, Mannheim, Ger-any) of 14 GHS-R�/� and 11 GHS-R�/� were monitored

aily at 10 AM from 3 days before the injection of STZntil 27 days after. Mice were subjected to a 13C gastricmptying breath test before induction of STZ diabetesnd 6, 10, 16, and 22 days thereafter. At day 27, mice (6HS-R�/�, 6 GHS-R�/�) were killed at 10 AM. Hypothal-

mi were snap frozen in liquid nitrogen for real-time PCR

able 1. Forward and Reverse Primer Sequences

Gene Primer sequences

. Wild-type allele 5=-TgggggTgCgAACATTAgC-3= (forward)5=-CTgAAggCATCTTTCACTACg-3=(reverse)

Knockout allele 5=-ACATATTCTATgTgAggCACC-3=(forward)5=-CTgAAggCATCTTTCACTACg-3=(reverse)

. GHS-R 5=-CCgCCTCTggCAgTATCg-3= (forward)5=-gCTgACAAACTggAAgAgTTTgC-3=(reverse)5=-CCCTggAACTTCggCgACCTgC-3= [5=]FAM

[3=]TAMRA (probe)�-actin 5=-CATCTTggCCTCACTgTCCAC-3=(forward)

5=-gggCCggACTCATCgTACT-3=(reverse)5=-TgCTTgCTgATCCACATCTgCTggA-3= [5=]FAM

[3=]TAMRA (probe). AgRP 5=-gCggAggTgCTAgATCCA-3= (forward)

5=-AggACTCgTgCAgCCTTA-3= (reverse)NPY 5=-CCgCTCTgCgACACTACAT-3= (forward)

5=-TgTCTCAgggCTggATCTCT-3= (reverse)POMC 5=-ACCTCACCACggAgAgCA-3= (forward)

5=-gCgAgAggTCgAgTTTgC-3= (reverse)GHS-R 5=-TCAgggACCAgAACCACAAA-3= (forward)

5=-CCAgCAgAggATgAAAgCAA-3= (reverse)Ghrelin 5=-CCAgAggACAgAggACAAgC-3= (forward)

5=-ACATCgAAgggAgCATTgAA-3= (reverse)GAPDH 5=-CCCCAATgTgTCCgTCgTg-3= (forward)

a5=-gCCTgCTTCACCACCTTCT-3= (reverse)

nalysis, fat pads (epididymal, inguinal, and retroperito-eal fat) were collected and weighed, blood was collectedy cardiac puncture for ghrelin measurements, and fun-us was dissected for in vitro contractility measurements.nother group of diabetic GHS-R�/� (n � 12) and GHS-�/� (n � 11) mice was not used in the gastric emptying

tudy but was killed for real-time PCR analysis and gh-elin measurements at day 7 (5 GHS-R�/�, 5 GHS-R�/�)nd day 15 (7 GHS-R�/�, 6 GHS-R�/�). A group ofge-matched nondiabetic ad libitum fed mice (7 GHS-�/�, 5 GHS-R�/�) was used as a control group for

eal-time PCR analysis, plasma ghrelin measurements,nd contractility studies.

Radioimmunoassay for Plasma Ghrelin LevelsAcidified plasma samples from ad libitum fed

ice (10 AM) were extracted on a Sep-Pak C18 cartridgeWaters Corporation, Milford, MA). The radioimmuno-ssay was performed with [125I] rabbit ghrelin as tracerBachem, Torrance, CA) and with a rabbit antibody raisedgainst human ghrelin[14 –28] (Eurogentec, Seraing, Bel-ium) (final dilution 1:8000), which recognizes both oc-anoylated and desoctanoylated ghrelin. Intraassay coef-cient of variation of the radioimmunoassay was 6.4%nd 7.7%, respectively. The minimal detectable dose was5.6 pg/mL.

Quantitative Real-Time PCRTotal RNA was extracted from the hypothalamus

y means of the TRIzol reagent and reverse transcribed toomplementary DNA (cDNA) with Superscript II Reverseranscriptase. The quantitative real-time PCR reactionas run on a Lightcycler 480 system (Roche Diagnostics,annheim, Germany) with LightCycler 480 SYBR GreenMaster mix. Primer sequences are shown in Table 1,

ection C. An interrun calibrator was used, and a stan-ard curve was created for each gene to obtain PCRfficiencies. Relative expression levels of all samples werealculated with the LightCycler 480 software and werexpressed relative to GAPDH and corrected for interrunariability.

Breath Test for Gastric EmptyingGastric emptying was measured during the light

hase (10 AM) with a noninvasive [13C]octanoic acidreath test in fasted (19 hours) mice according to theethod of Kitazawa et al,19 except that the amount of

3CO2 in the exhaled air was determined by an Infraredsotope Analyser (IRIS, Wagner, Germany).

In Vitro Contractility Studies With SmoothMuscle Strips From the FundusThe fundus from control (5 GHS-R�/�, 3 GHS-

�/�) and diabetic (4 GHS-R�/�, 4 GHS-R�/�) mice wasreed from mucosa, and strips were cut and suspended

long their circular axis in a tissue bath filled with Krebs

sr1mmmP

usi(Na(warc

(imswAv(c

tEiPtswGi(

iimcva

dao

Rtrea

gi.ac

FdGm.

BA

SIC–

ALIM

ENTA

RY

TRA

CT

1270 VERHULST ET AL GASTROENTEROLOGY Vol. 135, No. 4

olution as described by De Smet et al.25 The strips’esponse to increasing concentrations of ACh (10�4 to0�9 mol/L) and substance P (10�5 to 10�9 mol/L) waseasured isometrically. The negative logarithm of theedian effective concentration (pEC50) value and theaximal contraction were calculated with the GraphPad

rism 4.00 software (San Diego, CA).Neural responses were elicited by electrical field stim-

lation (EFS) of strips as described previously.25 Re-ponses were characterized pharmacologically by repeat-ng the frequency spectrum in the presence of L-NAME3 � 10�4 mol/L) in combination with atropine, the

K1-antagonist SR140333 (5 � 10�7 mol/L), the NK2-ntagonist SR48968 (5 � 10�7 mol/L), and tetrodotoxin3 � 10�4 mol/L), respectively. Strips were pretreatedith each antagonist for 25 minutes before EFS waspplied. The on-relaxations were expressed relative to theesponse induced by 10�5 mol/L nitroglycerin. The off-ontractions were expressed in grams/square millimeters.

Statistical AnalysisResults are presented as means � SEM. Data

body weight, glucose levels, food intake, gastric empty-ng, and in vitro contractility) obtained from repeated

easurements performed at different time points in theame mice were analyzed using the Proc Mixed procedureith slice option from SAS (SAS Institute Inc, Cary, NC).ll other data were analyzed with a factorial analysis ofariance followed by Newmann–Keuls as a post hoc testStatistica 6.0; StatSoft, Tulsa, OK). Significance was ac-epted at the 5% level.

ResultsGeneration of the GHS-R Knockout MiceThe strategy applied resulted in the deletion of

he first 2 exons that encode GHS-R1 (ENSEMBL:NSMUSG00000051136) (Figure 1A). Correct target-

ng in ES cells was confirmed by Southern analysis, andCR analysis demonstrated the ablation of the wild-ype GHS-R1 allele (results not shown). Loss of expres-ion of the GHS-R1 transcript in the knockout miceas confirmed by quantitative real-time PCR. TheHS-R1 transcript was absent in the brain and pitu-

tary derived from the homozygote GHS-R�/� miceFigure 1B).

Effect of STZ Diabetes on Blood GlucoseLevels, Food Intake, and Body WeightBoth genotypes displayed a significant (P � .001)

ncrease in blood glucose levels within 24 hours after thenjection of STZ (Figure 2A). Blood glucose levels were

aximal from day 3 onward and did not differ signifi-antly between both genotypes during the entire obser-ation period. In both genotypes, body weight loss was

pparent from day 2 after the induction of diabetes and s

id not differ between GHS-R�/� mice (�24.7% � 1.8%)nd GHS-R�/� mice (�25.6% � 3.8%) during the coursef the experiment (Figure 2B).

Total fat pad mass was similar in GHS-R�/� and GHS-�/� mice at the start of the experiment and amounted

o 4.4% � 0.7% and to 4.3% � 0.6% of body weight,espectively. At day 27 after the induction of diabetes, nopididymal, inguinal, or retroperitoneal fat was detect-ble in either genotype (data not shown).

Daily 24-hour food intake did not differ between bothenotypes before the start of the experiment. After thenduction of diabetes, food intake was significantly (P �05) increased at day 5 in the wild-type mice. Mutant micelso became hyperphagic, but this did not reach statisti-al significance until day 9 (P � .01). This hyperphagic

igure 2. Changes in blood glucose levels (A), body weight (B), andaily (24 hours) food intake (C) in STZ-induced diabetic GHS-R�/� andHS-R�/� mice as a function of time. Results are represented as theeans � SEM (n � 11–14 animals/group). *P � .05; **P � .01; ***P �

001 (GHS-R�/� vs GHS-R�/� mice).

tate remained during the further course of the experi-

mt1

lttmg7dCc

tsmAw.g

Re

tmsmd3NeAdos

wmda

FdaorPGfir3.R

BA

SIC–

ALI

MEN

TARY

TRA

CT

October 2008 ROLE OF GHRELIN IN DIABETIC MICE 1271

ent, but hyperphagia was significantly less (P � .01) inhe GHS-R�/� than in the GHS-R�/� mice between days2 and 21 (Figure 2C).

Effect of STZ Diabetes on Plasma GhrelinLevels, Ghrelin Messenger RNA, and GHS-RMessenger RNA Expression in theHypothalamusBefore the induction of diabetes, plasma ghrelin

evels did not differ between both genotypes, neither inhe fed (GHS-R�/�: 528, GHS-R�/�: 435 pg/mL) nor inhe fasted state (GHS-R�/�: 1221, GHS-R�/�: 1193 pg/

L). In diabetic GHS-R�/� and GHS-R�/� mice, plasmahrelin levels were significantly (P � .05) increased at dayafter the induction of diabetes and remained elevated

uring the further course of the experiment (Figure 3A).hanges in plasma ghrelin levels did not differ signifi-

antly between both genotypes.In the hypothalamus of diabetic GHS-R�/� mice, the

ime course of the changes in ghrelin and GHS-R mes-enger RNA (mRNA) expression was bell-shaped with

aximal increases (P � .01) at day 15 (Figure 3B and C).t day 27, ghrelin mRNA expression was normalized,hereas GHS-R mRNA expression was decreased (P �

01) compared with day 0. No difference in hypothalamichrelin mRNA expression between GHS-R�/� and GHS-

igure 3. Time-dependent effects ofiabetes on plasma ghrelin levels (A)nd hypothalamic mRNA expressionf ghrelin (B); GHS-R (C); and the neu-opeptides AgRP (D), NPY (E), andOMC (F) in ad libitum fed (10 AM)HS-R�/� and GHS-R�/� mice be-

ore and 7, 15, and 27 days after thenduction of diabetes. Results are rep-esented as the means � SEM (n �–7 animals/group). *P � .05; **P �01; ***P � .001 (GHS-R�/� vs GHS-

�/� mice).

�/� mice was observed. As expected, GHS-R mRNAxpression was absent in the mutant mice.

Effect of STZ Diabetes on NPY, AgRP, andProopiomelanocortin mRNA Expression inthe HypothalamusIn the wild-type mice, induction of diabetes led

o a gradual and significant increase in NPY and AgRPRNA expression from day 15 onward (P � .01) and a

ignificant decrease in proopiomelanocortin (POMC)RNA expression, which was already maximal from

ay 7 onward (P � .001) (Figure 3D–F). At day 27, a5-fold increase in AgRP mRNA, a 9-fold increase inPY mRNA, and a 7-fold decrease in POMC mRNA

xpression were observed. The increase in NPY andgRP mRNA expression started to diverge betweeniabetic GHS-R�/� and GHS-R�/� mice from day 15nward. No difference in POMC expression was ob-erved between both genotypes.

Effect of STZ Diabetes on Gastric EmptyingNo difference in gastric half excretion time (Thalf)

as observed between wild-type (Thalf: 109 � 5 minutes) andutant mice (Thalf: 110 � 7 minutes) before the onset of

iabetes. Gastric emptying of the solid meal was markedlyccelerated in both genotypes at day 16 (Thalf: GHS-R�/�:

5Paotim

tssoo25cG

rstC(w(th

f(ed

sfptlqaraSa

tdstar(.0f(m

Foar(a*malwcg

BA

SIC–

ALIM

ENTA

RY

TRA

CT

1272 VERHULST ET AL GASTROENTEROLOGY Vol. 135, No. 4

6 � 10 minutes, P � .01, and GHS-R�/�: 55 � 9 minutes,� .001) and day 22 (GHS-R�/�: 56 � 18 minutes, P � .01,

nd GHS-R�/�: 48 � 5 minutes, P � .001) after inductionf diabetes (Figure 4A). Similar effects were observed on lagime (Figure 4B). Gastric emptying correlated with foodntake in diabetic wild-type (r � 0.48, P � .01) and mutant

ice (r � 0.65, P � .01).

Effect of STZ Diabetes on the In VitroContractile Responses in the Mouse FundusNeuronal responses under nondiabetic condi-

ions. The responses of fundic smooth muscle strips,ubjected to EFS, were frequency dependent and con-isted mainly of on-relaxations. Off-contractions werebserved at higher frequencies (8 –32 Hz). EFS-inducedn-relaxations were more pronounced (P � .05) betweenand 16 Hz in GHS-R�/� than in GHS-R�/� mice (Figure

A). Within the same frequency range, EFS-induced off-ontractions were significantly (P � .05) increased inHS-R�/� compared with GHS-R�/� mice (Figure 5B).

Effect of STZ diabetes on neuronal responses. Aepresentative tracing of the neuronal response of fundictrips from GHS-R�/� and GHS-R�/� mice 27 days afterhe induction of diabetes is shown in Figure 6A and B.hanges in tension are summarized in the same Figure

Figure 6C–F). In both genotypes, inhibitory responsesere reduced at low frequencies and abolished at high

16 –32 Hz) frequencies (Figure 6C and D). In addition,he poststimulus off-contractions were strongly en-anced after the induction of diabetes at all investigated

igure 5. Electrical field stimulationf strips from nondiabetic GHS-R�/�

nd GHS-R�/� mice resulted in on-elaxations (A) and off-contractionsB). Results are means � SEM (n � 3nimals/group; n � 3 strips/animal).P � .05 (GHS-R�/� vs GHS-R�/�

ice). On-relaxations are expresseds a percentage of the maximal re-

axation obtained after stimulationith 10�5 mol/L nitroglycerin; off-ontractions are expressed in

rams/square millimeters.

requencies in wild-type (P � .001) as well as in mutantP � .05) mice (Figure 6E and F). No significant differ-nces in the contractile responses were observed betweeniabetic GHS-R�/� and GHS-R�/� mice.

Effect of STZ diabetes on the contractile re-ponse to ACh and substance P. Neural responses ofundic strips from nondiabetic mice were characterizedharmacologically (Figure 7A and B). Pretreatment withhe NO-synthase inhibitor L-NAME reversed the on-re-axations into strong contractions over the entire fre-uency spectrum. These contractions were abolished byddition of atropine at low frequencies. The atropine-esistant contractions were further reduced by successiveddition of the NK1 and the NK2 receptor antagonists,R140333 and SR48968, respectively. All responses werebolished in the presence of tetrodotoxin.

Because ACh and substance P are the 2 main neuro-ransmitters mediating excitatory responses in the fun-us, the effect of diabetes on the susceptibility of themooth muscle strips to ACh and substance P was inves-igated by establishing dose-response curves to bothgents (Figure 7C and D). In both genotypes, the dose-esponse curve to ACh was shifted to the left at day 27GHS-R�/�: pEC50 from 6.13 � 0.05 to 6.38 � 0.09, P �01, and GHS-R�/�: pEC50 from 6.09 � 0.03 to 6.42 �.05, P � .01). A significant increase of the pEC50 valueor substance P was also observed in diabetic wild-typefrom 6.37 � 0.10 to 7.24 � 0.20, P � .001) and mutant

ice (from 6.27 � 0.11 to 7.31 � 0.23, P � .001). Neither

Figure 4. Time-dependent changesin gastric half-excretion time (Thalf) (A)and lag time (Tlag) (B) as determined bythe [13C]octanoic acid breath test indiabetic GHS-R�/� and GHS-R�/�

mice. Results are means � SEM (n �11–14 animals/group). †P � .01(GHS-R�/�, compared with day 0);‡P � .05; ‡‡P � .01; ‡‡‡P � .001(GHS-R�/�, compared with day 0).

tc

sphlcdntc

Sbrtfifwwtap

dasapmnwfaaon

gGwePmdmwt

Fbdmid(motaa(a.

BA

SIC–

ALI

MEN

TARY

TRA

CT

October 2008 ROLE OF GHRELIN IN DIABETIC MICE 1273

he genotype nor the treatment affected the maximalontraction to ACh or substance P.

DiscussionIn the present study, we showed that the ghrelin-

ignaling pathway plays an important role in the hy-erphagia associated with STZ-induced diabetes. Thisyperphagia is mainly driven by central pathways stimu-

ated by the high circulating plasma ghrelin levels. Inontrast, the acceleration of gastric emptying during theiabetic process occurs independently of ghrelin and isot the cause of the hyperphagia. Impaired accommoda-ion of the stomach because of local ghrelin-independenthanges in contractility may accelerate gastric emptying.

In the pathogenesis of hyperphagia associated withTZ-induced diabetes, a model of insulin-deficient dia-etes, reduced signaling by insulin and leptin plays a keyole.26 –28 Recent studies suggest that STZ-induced diabe-es in rats is characterized by hyperghrelinemia,7,29 anding that was confirmed in the present study. Evidence

or a causative role of endogenous ghrelin in hyperphagiaas further provided in diabetic ghrelin knockout mice,hich showed an attenuation of hyperphagia.8 Likewise,

reatment of diabetic rodents with a ghrelin receptorntagonist was able to partially inhibit diabetic hy-

igure 6. Effect of STZ-induced dia-etes on the neural responses of fun-ic strips from wild-type and mutantice. Representative tracings of EFS-

nduced responses in nondiabetic andiabetic GHS-R�/� (A) and GHS-R�/�

B) mice are shown. The graphs sum-arize the change in tension of then-relaxations (C and D) and off-con-ractions (E and F) in GHS-R�/� (left)nd GHS-R�/� (right) mice. Resultsre represented as the means � SEM

n � 3–5 animals/group; n � 3 strips/nimal). *P � .05; **P � . 01; ***P �001 (nondiabetic vs diabetic mice).

erphagia.7,8 R

In diabetic ghrelin knockout mice the ligand-indepen-ent signaling activity of the ghrelin receptor may still playn important role in hyperphagia. In the present study, wehowed that ablation of the ghrelin receptor did not causefurther reduction of hyperphagia but rather prolonged theeriod of reduced hyperphagia. In the ghrelin knockoutice, the initial attenuation of hyperphagia had already

ormalized at day 10 after the induction of diabetes,hereas, in the ghrelin receptor knockout mice, the reduced

ood intake was apparent at a later stage between days 12nd 21. Thus, it appears that an increased constitutivectivity of the ghrelin receptor may compensate for the lossf ghrelin in the ghrelin knockout mice and lead to an earlyormalization of hyperphagia in these mice.The role of hypothalamic neuropeptides in the diver-

ence of hyperphagic responses between GHS-R�/� andHS-R�/� mice was further investigated. In accordanceith earlier studies,30 –32 we observed an increased mRNA

xpression of NPY and AgRP and a reduced expression ofOMC after the induction of STZ diabetes in wild-typeice. However, in diabetic GHS-R�/� mice, the time-

ependent increase in the expression of NPY and AgRPRNA was less pronounced. These changes coincidedith the divergence in food intake between both geno-

ypes. Because the AgRP/NPY neurons express the GHS-

33 and are the primary targets of ghrelin’s orexigenic

aatdemdmgAn

iipcrrpiiirrtfh

dct

gimb1aodtiadb

tsmOwh

t

BA

SIC–

ALIM

ENTA

RY

TRA

CT

1274 VERHULST ET AL GASTROENTEROLOGY Vol. 135, No. 4

ctions in the hypothalamus,12,13 this observation impliesrole for the NPY/AgRP neuron as a downstream effec-

or of ghrelin in diabetic hyperphagia. The time-depen-ent changes in hypothalamic ghrelin and GHS-R mRNAxpression were bell-shaped. The decrease of GHS-RRNA expression at the end of the observation period in

iabetic wild-type mice may point to a negative feedbackechanism initiated by the sustained increase in plasma

hrelin levels and by the higher mRNA expression ofgRP and NPY to prevent overeating. Nevertheless, it isot sufficient to prevent hyperphagia.During the ingestion of food, the brain also receives

mportant information from mechanoreceptors, quantitat-ng stretch and chemoreceptors, activated by nutrientsresent in the gastrointestinal tract. In addition, the well-ontrolled process of gastric emptying also plays a crucialole in the regulation of satiety. Rapid emptying wouldeduce the negative feedback satiety signal produced by theresence of nutrients inside the stomach and, thus, precip-

tate a feeling of hunger. However, a recent study showed anncrease of postprandial symptoms and decreased caloricntake following acceleration of gastric emptying.34 Theapid downloading of nutrients in the duodenum mayelease intestinal peptides such as CCK, PYY, and GLP-1hat may enhance satiety signaling. In the present study, weound a good correlation between gastric emptying and

yperphagia in both genotypes. Nevertheless, our data in- d

icate that the accelerated gastric emptying was not theause of hyperphagia because it occurred in a later phase ofhe diabetic process.

In human, diabetes is often accompanied by delayedastric emptying. A dysfunction of the vagal nerve result-ng from autonomic neuropathy is believed to be the

ajor cause.35 However, accelerated gastric emptying cane seen in subgroups of patients in the early stage of type36 or type 2 diabetes.37,38 Likewise, there is evidence forn accelerated gastric emptying in several rodent modelsf insulin-dependent21,39,40 and noninsulin-dependentiabetes mellitus,41,42 but only few studies investigatedhe mechanisms involved. Our study showed that thencreased plasma ghrelin levels do not contribute to theccelerated gastric emptying in diabetic mice because noifference in gastric emptying rate could be establishedetween GHS-R�/� and GHS-R�/� mice.

It is also unlikely that acute hyperglycemia contributedo accelerated gastric emptying because earlier studieshowed that high glucose induces a reduction of gastric

otility43,44 and markedly delays gastric emptying.45

ther studies in STZ-induced diabetic rats46 and patientsith type 1 diabetes47 also report a slowing effect of acuteyperglycemia on gastric emptying.Gastric emptying results from the coordinated activity of

he proximal stomach (fundus), antrum, pylorus, and duo-

Figure 7. Pharmacologic character-ization of neural responses of nondia-betic wild-type mice (A and B). (A)Representative tracings obtained inthe presence of L-NAME (3 � 10�4

mol/L), and after cumulative additionof atropine (5 � 10�6 mol/L),SR140333 (5 � 10�7 mol/L),SR48968 (5 � 10�7 mol/L), and tetro-dotoxin (TTX) (3 � 10�4 mol/L), re-spectively. (B) Tension of the off-con-tractions obtained with the variouspharmacologic treatments. Resultsare mean � SEM (n � 3 animals/group; n� 3 strips/animal). *P � .05;** P � .001. Effect of diabetes on thecontractile response to ACh and sub-stance P in fundic strips (C and D).Dose-response curves to ACh (C) andsubstance P (D) in strips from nondia-betic (open symbols) and diabetic(solid symbols) GHS-R�/� and GHS-R�/� mice. Results are represented asthe means � SEM (n � 3–5 animals/group; n � 3 strips/animal).

enum. Loss of fundic relaxation is expected to accelerate

gagctilotncsrtrerir

rtwptsliittt

tacidcbp

rgsgils

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

BA

SIC–

ALI

MEN

TARY

TRA

CT

October 2008 ROLE OF GHRELIN IN DIABETIC MICE 1275

astric emptying, whereas loss of pyloric or duodenal relax-tion may delay gastric emptying. We focused on the re-ional alterations in fundic smooth muscle contractility inontrol and diabetic mice of both genotypes. In both geno-ypes, induction of diabetes abolished or reduced nitrergicnhibitory responses in fundic strips. It has been well estab-ished that diabetes is associated with low NO levels becausef a decreased neuronal NO synthase (nNOS) expression inhe gastric myenteric plexus.48–50 The molecular mecha-isms regulating changes in expression of nNOS are un-lear. Insulin-induced reversal of nNOS loss in diabetic miceuggests that insulin or insulin-like growth factors mayegulate the nNOS expression in these systems.50 Alterna-ively, continuous high glucose levels may directly or indi-ectly influence nNOS expression, especially because thenteric nervous system contains glucoseresponsive neu-ons.51 Our results therefore suggest that gastric emptyings accelerated because of a major loss of nitrergic inhibitoryesponses in the fundus.

In addition, in both genotypes, the excitatory reboundesponses were markedly enhanced compared with con-rols. The dose-response curves to ACh and substance Pere shifted to the left in fundic smooth muscle strips,ointing to an increased affinity of the muscarinic andachykinergic receptors for their substrates. An increasedensitivity to ACh was also reported by Kamata et al52 inongitudinal smooth muscle strips of the fundus of STZ-nduced diabetic rats. Because the changes in contractil-ty were similar between GHS-R�/� and GHS-R�/� mice,hese data confirm our finding that the accelerated gas-ric emptying in diabetic mice occurs independently fromhe ghrelin-signaling pathway.

Reduced relaxations and increased rebound contrac-ions in the diabetic mice may lead to impaired fundicccommodation and accelerated gastric emptying. Be-ause impaired gastric accommodation normally resultsn increased satiety,53 it cannot play an important role iniabetic hyperphagia. This reinforces our hypothesis thathanges in the central mechanisms associated with dia-etic hyperphagia overrule the symptoms induced byeripheral changes in gastric contractility.In conclusion, diabetic hyperphagia is predominantly

egulated by central mechanisms in which peripheralhrelin plays a predominant role by affecting the expres-ion of NPY and AgRP in the hypothalamus. Acceleratedastric emptying, which occurs independently of ghrelin,s not the cause of increased food intake and involvesocal changes in the enteric nervous system and inmooth muscle contractility.

References

1. Howard AD, Feighner SD, Cully DF, et al. A receptor in pituitaryand hypothalamus that functions in growth hormone release.Science 1996;273:974–977.

2. Kojima M, Hosoda H, Date Y, et al. Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature

1999;402:656–660.

3. Tschop M, Smiley DL, Heiman ML. Ghrelin induces adiposity inrodents. Nature 2000;407:908–913.

4. Nakazato M, Murakami N, Date Y, et al. A role for ghrelin in thecentral regulation of feeding. Nature 2001;409:194–198.

5. Wren AM, Seal LJ, Cohen MA, et al. Ghrelin enhances appetiteand increases food intake in humans. J Clin Endocrinol Metab2001;86:5992.

6. Cummings DE, Purnell JQ, Frayo RS, et al. A preprandial rise inplasma ghrelin levels suggests a role in meal initiation in hu-mans. Diabetes 2001;50:1714–1719.

7. Ishii S, Kamegai J, Tamura H, et al. Role of ghrelin in streptozo-tocin-induced diabetic hyperphagia. Endocrinology 2002;143:4934–4937.

8. Dong J, Peeters TL, De Smet B, et al. Role of endogenous ghrelinin the hyperphagia of mice with streptozotocin-induced diabetes.Endocrinology 2006;147:2634–2642.

9. Murdolo G, Lucidi P, Di Loreto C, et al. Insulin is required forprandial ghrelin suppression in humans. Diabetes 2003;52:2923–2927.

0. Holst B, Cygankiewicz A, Jensen TH, et al. High constitutivesignaling of the ghrelin receptor—identification of a potent in-verse agonist. Mol Endocrinol 2003;17:2201–2210.

1. Pantel J, Legendre M, Cabrol S, et al. Loss of constitutive activityof the growth hormone secretagogue receptor in familial shortstature. J Clin Invest 2006;116:760–768.

2. Seoane LM, Lopez M, Tovar S, et al. Agouti-related peptide,neuropeptide Y, and somatostatin-producing neurons are targetsfor ghrelin actions in the rat hypothalamus. Endocrinology 2003;144:544–551.

3. Chen HY, Trumbauer ME, Chen AS, et al. Orexigenic action ofperipheral ghrelin is mediated by neuropeptide Y and agouti-related protein. Endocrinology 2004;145:2607–2612.

4. Date Y, Murakami N, Toshinai K, et al. The role of the gastricafferent vagal nerve in ghrelin-induced feeding and growthhormone secretion in rats. Gastroenterology 2002;123:1120–1128.

5. Fujino K, Inui A, Asakawa A, et al. Ghrelin induces fasted motoractivity of the gastrointestinal tract in conscious fed rats.J Physiol 2003;550:227–240.

6. Tack J, Depoortere I, Bisschops R, et al. Influence of ghrelin oninterdigestive gastrointestinal motility in humans. Gut 2006;55:327–333.

7. Masuda Y, Tanaka T, Inomata N, et al. Ghrelin stimulates gastricacid secretion and motility in rats. Biochem Biophys Res Com-mun 2000;276:905–908.

8. Tack J, Depoortere I, Bisschops R, et al. Influence of ghrelin ongastric emptying and meal-related symptoms in idiopathic gas-troparesis. Aliment Pharmacol Ther 2005;22:847–853.

9. Kitazawa T, De Smet B, Verbeke K, et al. Gastric motor effects ofpeptide and non-peptide ghrelin agonists in mice in vivo and invitro. Gut 2005;54:1078–1084.

0. Depoortere I, De Winter B, Thijs T, et al. Comparison of thegastroprokinetic effects of ghrelin, GHRP-6 and motilin in rats invivo and in vitro. Eur J Pharmacol 2005;515:160–168.

1. Granneman JG, Stricker EM. Food intake and gastric emptying inrats with streptozotocin-induced diabetes. Am J Physiol 1984;247:R1054–R1061.

2. Miyamoto Y, Yoneda M, Morikawa A, et al. Gastric neuropeptidesand gastric motor abnormality in streptozotocin-induced diabeticrats: observation for four weeks after streptozotocin. Dig Dis Sci2001;46:1596–1603.

3. Wattler S, Kelly M, Nehls M. Construction of gene targetingvectors from lambda KOS genomic libraries. Biotechniques1999;26:1150–1160.

4. O’Gorman S, Dagenais NA, Qian M, et al. Protamine-Cre recom-

binase transgenes efficiently recombine target sequences in the

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

4

4

4

4

4

4

4

5

5

5

5

GL1

RM

Ja

BA

SIC–

ALIM

ENTA

RY

TRA

CT

1276 VERHULST ET AL GASTROENTEROLOGY Vol. 135, No. 4

male germ line of mice, but not in embryonic stem cells. Proc NatlAcad Sci U S A 1997;94:14602–14607.

5. De Smet B, Thijs T, Peeters TL, et al. Effect of peripheral obesta-tin on gastric emptying and intestinal contractility in rodents.Neurogastroenterol Motil 2007;19:211–217.

6. Booth DA. Some characteristics of feeding during streptozotocin-induced diabetes in the rat. J Comp Physiol Psychol 1972;80:238–249.

7. Sipols AJ, Baskin DG, Schwartz MW. Effect of intracerebroven-tricular insulin infusion on diabetic hyperphagia and hypotha-lamic neuropeptide gene expression. Diabetes 1995;44:147–151.

8. Sindelar DK, Havel PJ, Seeley RJ, et al. Low plasma leptin levelscontribute to diabetic hyperphagia in rats. Diabetes 1999;48:1275–1280.

9. Masaoka T, Suzuki H, Hosoda H, et al. Enhanced plasma ghrelinlevels in rats with streptozotocin-induced diabetes. FEBS Lett2003;541:64–68.

0. Sindelar DK, Mystkowski P, Marsh DJ, et al. Attenuation ofdiabetic hyperphagia in neuropeptide Y-deficient mice. Diabetes2002;51:778–783.

1. Havel PJ, Hahn TM, Sindelar DK, et al. Effects of streptozotocin-induced diabetes and insulin treatment on the hypothalamicmelanocortin system and muscle uncoupling protein 3 expres-sion in rats. Diabetes 2000;49:244–252.

2. Qu SY, Yang YK, Li JY, et al. Agouti-related protein is a mediatorof diabetic hyperphagia. Regul Pept 2001;98:69–75.

3. Willesen MG, Kristensen P, Romer J. Co-localization of growthhormone secretagogue receptor and NPY mRNA in the arcuatenucleus of the rat. Neuroendocrinology 1999;70:306–316.

4. Alsina ST, Bruix SA, Sunyé P, et al. Pharmacological accelerationof gastric emptying decreases caloric intake and increases post-prandial symptoms in obese subjects. Gastroenterology 2008;134:A-315.

5. Feldman M, Corbett DB, Ramsey EJ, et al. Abnormal gastricfunction in longstanding, insulin-dependent diabetic patients.Gastroenterology 1979;77:12–17.

6. Nowak TV, Johnson CP, Kalbfleisch JH, et al. Highly variablegastric emptying in patients with insulin dependent diabetesmellitus. Gut 1995;37:23–29.

7. Phillips WT, Schwartz JG, McMahan CA. Rapid gastric emptying inpatients with early non-insulin-dependent diabetes mellitus.N Engl J Med 1991;324:130–131.

8. Bertin E, Schneider N, Abdelli N, et al. Gastric emptying is accel-erated in obese type 2 diabetic patients without autonomic neu-ropathy. Diabetes Metab 2001;27:357–364.

9. Nowak TV, Roza AM, Weisbruch JP, et al. Accelerated gastricemptying in diabetic rodents: effect of insulin treatment andpancreas transplantation. J Lab Clin Med 1994;123:110–116.

0. Young AA, Gedulin B, Vine W, et al. Gastric emptying is acceler-ated in diabetic BB rats and is slowed by subcutaneous injec-

tions of amylin. Diabetologia 1995;38:642–648.

1. Itoh H, Yoneda M, Tamori K, et al. Rapid gastric emptying andpathological changes of vagus nerve in the spontaneouslydiabetic Chinese hamster. Diabetes Res Clin Pract 1995;28:89–95.

2. Green GM, Guan D, Schwartz JG, et al. Accelerated gastric emp-tying of glucose in Zucker type 2 diabetic rats: role in postpran-dial hyperglycaemia. Diabetologia 1997;40:136–142.

3. Sakaguchi T, Shimojo E. Inhibition of gastric motility induced byhepatic portal injections of D-glucose and its anomers. J Physiol1984;351:573–581.

4. Raybould HE, Zittel TT. Inhibition of gastric motility induced byintestinal glucose in awake rats: role of Na(�)-glucose co-trans-porter. Neurogastroenterol Motil 1995;7:9–14.

5. Ishiguchi T, Tada H, Nakagawa K, et al. Hyperglycemia impairsantro-pyloric coordination and delays gastric emptying in con-scious rats. Auton Neurosci 2002;95:112–120.

6. Chang FY, Lee SD, Yeh GH, et al. Hyperglycaemia is responsiblefor the inhibited gastrointestinal transit in the early diabetic rat.Acta Physiol Scand 1995;155:457–462.

7. Fraser RJ, Horowitz M, Maddox AF, et al. Hyperglycaemia slowsgastric emptying in type 1 (insulin-dependent) diabetes mellitus.Diabetologia 1990;33:675–680.

8. Takahashi T, Nakamura K, Itoh H, et al. Impaired expression ofnitric oxide synthase in the gastric myenteric plexus of sponta-neously diabetic rats. Gastroenterology 1997;113:1535–1544.

9. Wrzos HF, Cruz A, Polavarapu R, et al. Nitric oxide synthase (NOS)expression in the myenteric plexus of streptozotocin-diabeticrats. Dig Dis Sci 1997;42:2106–2110.

0. Watkins CC, Sawa A, Jaffrey S, et al. Insulin restores neuronalnitric oxide synthase expression and function that is lost indiabetic gastropathy. J Clin Invest 2000;106:373–384.

1. Liu M, Seino S, Kirchgessner AL, et al. Identification and charac-terization of glucoresponsive neurons in the enteric nervous sys-tem. J Neurosci 1999;19:10305–10317.

2. Kamata K, Sakamoto A, Kasuya Y. Changes in sensitivity of therat stomach fundus to various drugs in streptozotocin-induceddiabetic rats. Jpn J Pharmacol 1988;47:99–102.

3. Tack J, Piessevaux H, Coulie B, et al. Role of impaired gastricaccommodation to a meal in functional dyspepsia. Gastroenter-ology 1998;115:1346–1352.

Received November 20, 2007. Accepted June 19, 2008.Address requests for reprints to: Inge Depoortere, PhD, Centre for

astroenterological Research, Gasthuisberg O&N1, bus 701, 3000euven, Belgium. e-mail: inge.depoortere@med.kuleuven.be; fax: (32)6-345-939.Supported by grants from the Flemish Foundation for Scientific

esearch (contract FWO G.0144.04 and 1.5.125.05) and the Belgianinistry of Science (contract GOA 03/11).Conflicts of interest: D. Moechars and L. Ver Donck are employees of

ohnson & Johnson Pharmaceutical Research and Development. Otheruthors have no conflicts of interest.

P.J.V. and B.D.S. equally contributed to this paper.