Noradrenergic inhibition of canine gallbladder contraction and murine pancreatic secretion during stress by corticotropin releasing factor

Noradrenergic inhibition of canine gallbladder

contraction and murine pancreatic secretion

during stress by corticotropin-releasing factor.

H J Lenz, … , B Messmer, F G Zimmerman

J Clin Invest. 1992;89(2):437-443. https://doi.org/10.1172/JCI115603.

Gastrointestinal secretory and motor responses are profoundly altered during stress; but the effects of stress and its mediator(s) on the two major gut functions, exocrine pancreatic secretion and gallbladder motility, are unknown. We therefore developed two animal models that allowed us to examine the effects of acoustic stress on canine gallbladder contraction and restraint stress on rat exocrine pancreatic secretion. Acoustic stress

inhibited cholecystokinin-8 (CCK)- and meal-induced gallbladder contraction, and restraint stress inhibited basal and CCK/secretin-stimulated pancreatic secretion. These inhibitory responses were mimicked by cerebral injection of corticotropin-releasing factor (CRF) and abolished by the CRF antagonist, alpha-helical CRF-(9-41). The effects of stress and

exogenous CRF were simulated by intravenous infusion of norepinephrine but prevented by ganglionic, noradrenergic, and alpha-adrenergic but not beta-adrenergic receptor blockade. Vagotomy, adrenalectomy, and--in rats--hypophysectomy did not alter the effects produced by stress and CRF. These results indicate that endogenous CRF released in response to different stressors in distinct species inhibits canine gallbladder contraction and murine exocrine pancreatic secretion via activation of sympathetic efferents. Release of

norepinephrine appears to be the final common pathway producing inhibition of biliary and pancreatic digestive function during stress mediated by cerebral CRF.

Research Article

Find the latest version:

Noradrenergic

Inhibition of

Canine Gallbladder Contraction and

Murine

Pancreatic Secretion during

Stress by

Corticotropin-releasing

Factor

H.JurgenLenz,** BernhardMessmer,*andFrank G. Zimmerman*

*Division ofGastroenterology, DepartmentofMedicine, UniversityofCaliforniaatSan Diego,School ofMedicine,LaJolla, California 92093-0671;and tNeurogastroenterology Laboratory, DepartmentofMedicine, University ofHamburg,

2000 Hamburg 20, Federal Republic of Germany

Abstract

Gastrointestinalsecretory andmotorresponsesareprofoundly altered during stress; but the effects ofstress and its media-tor(s)onthetwomajorgut functions, exocrine pancreatic

secre-tion andgallbladder motility,areunknown. We therefore devel-oped two animal models that allowedus toexamine the effects of acoustic stress on canine gallbladder contraction and

re-straint stress on rat exocrine pancreatic secretion. Acoustic stress inhibited cholecystokinin-8

(CCK)-

and meal-induced gallbladdercontraction,andrestraintstressinhibited basal and CCK/secretin-stimulated pancreatic secretion. These inhibi-tory responses were mimicked by cerebralinjectionof cortico-tropin-releasingfactor(CRF)and abolishedby the CRF antago-nist,a-helicalCRF-(941).Theeffects of stress and exogenous CRF were simulated by intravenous infusion ofnorepinephrine butprevented by ganglionic, noradrenergic, and a-adrenergic butnot

_{fl-adrenergic}

receptor blockade.Vagotomy, adrenalec-tomy, and-in rats-hypophysectomy didnotalter the effects produced by stress and CRF.Theseresultsindicate that endoge-nous CRF released in response to different stressors in distinct speciesinhibitscaninegallbladder contractionandmurine

exo-crinepancreatic secretion via activation ofsympathetic effer-ents. Release ofnorepinephrineappearstobe the finalcommon

pathway producinginhibition ofbiliaryandpancreatic diges-tivefunction during stress mediated by cerebral CRF.(J. Clin. Invest. 1992. 89:437-443.) Key words: gallbladder contraction *pancreaticsecretion *stress*corticotropin-releasing

factor.

neuropeptide

Introduction

Afteringestionof a meal under normalconditions,thevarious digestiveprocesses suchassecretion,gutmotility,and absorp-tionareratherindependentfrom central nervous system con-trol. Theyarecoordinatedby theentericnervoussystem, gut hormones, and paracrinefactors(1). Duringstress, however, brain regions(cerebral cortex, limbicsystem, and hypothala-mus) influence the informational outflow from brain stem

centers tothe gut and therebyaffectgastrointestinalfunctions (2). Stress-related postprandial syndromes like abdominal crampinganddiarrhea may be secondary to abnormalgastric

Addressreprint requests to H.JurgenLenz, University of California, SanDiego, School of Medicine, La Jolla, CA 92093-0671.

Receivedforpublication 15 January 1991 and in revised form 4 October1991.

motor function and enhanced colonic motility (3-5). How-ever, theeffects ofmentalorphysical stressonthetwomajor digestiveprocesses,gallbladder contractionandexocrine pan-creatic secretion,are, to the bestofourknowledge, completely unknown.This paucityof knowledge appearstobesecondary tothe lackof suitable experimentalmodels.Itis of interest that both the gallbladder and theexocrinepancreasaredensely in-nervated byadrenergic efferents buttheirphysiological func-tions are not well defined (6, 7). Inthe endocrine pancreas, noradrenergic inputinhibits insulin and stimulates glucagon secretion, andneuroglucopenia produces selective activation ofthesympathetic efferentstothe pancreas(8-1 1).

The41-residue hypothalamic peptide, corticotropin-releas-ingfactor (CRF; reference 12), appears to be the major endoge-nous central nervous systemtransmitterthat produces charac-teristic gastrointestinal stress responses: inhibition ofgastric acid secretion and gastrointestinal transit (3, 4, 13-17) and stimulationofgastroduodenalbicarbonate secretions and

co-lonic motility (3, 4, 16, 18, 19). The central nervous system effects ofCRF and stressaremediatedprimarilyby increased sympathetictoneand, in part, byparasympathetic withdrawal

(13-20)

or, in one instance, by activation of a functional pitu-itary-gut axis(18).However, the effects of CRF on biliary and pancreatic digestiveprocesses are unknown. Wetherefore de-veloped two animal models that allowed us to examine the effects ofstressandexogenous CRF oncaninegallbladder con-tractionandratexocrinepancreatic secretions.

Theresults of these studies indicate that CRF, released en-dogenouslyin response toacousticandrestraintstress,inhibits gallbladder contraction andpancreaticsecretion,respectively. Release of norepinephrine acting on a-adrenergic receptors ap-pearsto bethe finalcommon pathway inhibiting biliary and pancreatic digestive function duringstress mediated by endoge-nousCRF.

Methods

Animal preparations. Male Sprague-Dawley rats (250-300 g) and male beagle dogs (9-10kg) were obtained from Wiga(Sulzfeld,FRG). Ani-malswere fed a standard diet and housed in single quarters under temperature-, humidity-, and illumination-controlled conditions. Afterabdominalsurgery, animals were fedaliquid diet (Altromin Son-derdiat, Hamburg, FRG) forup to 7d. Dogs were fitted with a third cerebral ventricular cannula and a gastric fistula as previously de-scribed(15,21). Some dogs subsequently underwent truncalvagotomy andpyloroplastyorbilateral adrenalectomy(15,21,22). Adrenalecto-mized animals received prednisone (5 mg/d with meals; Apotheke, Universitatskrankenhaus-Eppendorf, Hamburg, FRG) and had free

accessto0.075MNaCl.Completeness of vagotomywasverified bya

1.Abbreviations used in this paper: CCK, cholecystokinin 8; CRF, corticotropin-releasing factor.

J.Clin. Invest.

©TheAmericanSociety for Clinical Investigation, Inc. 0021-9738/92/02/0437/07 $2.00

negative 2-deoxy-D-glucose test (15). Animals were allowed to recover for 23 wk before the experiment. To obtain venous access, an 18 Frenchplasticcatheter wasinsertedinto afrontlegvein on the day of theexperiment.

4d before theexperiment, rats were fitted with cerebral right lateral ventricular cannulae, a gastric fistula, and a right jugular venous cath-eter(16-18).2dbeforetheexperiment, under general anesthesia (xyla-zineat6 mg/kg; Rompun, Bayer AG, Leverkusen, FRG and ketamine hydrochloride at 50mg/kg;Ketanest, Parke-Davis, Morris Plains, NJ), they were fitted with apancreaticcatheter. Theproximal bileduct was ligated and connected with theduodenum by polyethylene tubing (0.36 mm i.d., 0.96 mm o.d.). The distal commonbile-pancreaticduct was cannulatedwitha secondpolyethylene catheter that was routed subcutaneously to exit in theinterscapularregion of the animal's neck. For 2 dbefore theexperiment,pancreatic juice drained into a small plasticsack thatwasemptiedevery 8 h. Some animals were subjected tobilateraltruncal vagotomy or adrenalectomy(17),and hypophysec-tomized animalswerecommercially obtained (Wiga).Completeness of vagotomy wasverifiedby anegative 2-deoxy-D-glucose test (17) and completeness ofadrenalectomyand hypophysectomy bymacroscopic inspection. Adrenalectomized animals andhypophysectomized ani-mals hadfreeaccess to0.15MNaCl.

Experimental design. Aftera 24-hfast,dogs werestudied in modi-fied Pavlov-type slings.Tofacilitateaccurate measurementof gallblad-derdiameters,theabdomen was shaved on the daybeforethe experi-ment. RatCRF, the CRF receptorantagonist, a-helical CRF-(9-41) (23),orsterile water as the control(vehicleforpeptides)wasinjected cerebroventricularlyin 10-ul volumes andintravenouslyin_200-,41 vol-umes over 1 min asdescribed (15, 21).Peptidedoses weregivenon separatedaysand in random order.Peptideswereobtained fromDr. J. E.Rivier (SalkInstitute,LaJolla,CA).In mostexperiments, gallblad-dercontractionwasinduced by infusion ofcholecystokinin-8(CCK; 20 pmol/kg per hi.v.). In someexperiments gallbladder contraction wasinducedbyintragastric administration ofablenderizedmeal con-sisting ofcookedgroundbeef, bread,and butter(240ml,14 gfat,32 g protein, 12 gcarbohydrate).Tomimicapostprandialpancreatic secre-tory response,exocrinepancreatic secretionwasstimulated by simulta-neousinfusion of CCKandsecretin(10 pmol/kgper heach).CRF (or control)wasadministeredatthetimethe CCKinfusionwasbegunor at thetimethemeal wasgiven.The CRFantagonistwasadministered15 min before CRF. In stressexperiments, dogswerefitted with head-phones andexposed for1 h to"pop"music(70-80 dB)thatwas prere-corded atdifferent speeds (17, 33,and 45 rpm in randomorder,1min each). Incontrol experiments, dogswereequippedwith headphones but notexposedto themusic.

Toexamine theneurohumoralpathwaysmediatingtheeffectsof CRF and stress,pharmacologicalapproacheswereutilized: ganglionic blockadewithchlorisondamine (1mg/kgi.v.;Ecolid,CIBA-Geigy, Ba-sel,Switzerland);noradrenergicblockadewithbretylium (10 mg/kgper h, Bretylol; American Critical Care, American Hospital Supply, McGawPark, IL);

03-adrenergic

blockade withpropranolol (2mg/kg perh,Dociton;Rhein-Pharma,ArzneimittelwerkGmbH,Heidelberg,

FRG); opiatereceptorblockade withnaloxone (1 mg/kgi.v.;Sigma ChemieGmbH,Deisenhofen, FRG);a-adrenergicreceptor blockade with phentolamine (0.1-10 gmol/kg per h; Regitin, CIBA-Geigy GmbH, Wehr, FRG).These agentsweregiven15minbefore gallblad-der contraction was induced.Norepinephrine(100nmol/kgperh,

Ar-terenol;HoechstAG,Frankfurt,FRG)wasadministeredatthesame

timegallbladdercontractionwasinduced.Completenessofautonomic nervous system, noradrenergic, andopiateblockades has been vali-datedpreviously (15-18, 21, 22).Inadditional experiments, adreno-corticotropichormone (ACTH)or_fl-endorphin(100pmol/kgperh) was intravenously infused for 1 h.

Ratswere fasted for 24 h but had free access to water until the beginningoftheexperiment.Animals were placed in 15-liter bucketsto

which they were accustomed. The gastric cannula was opened, the gas-triccontentinspected,and thepancreaticcatheter connectedto a

0.5-mlsyringethatwasattachedtothe rim of the bucketwithadhesive

tape. The animalwasnotstudied if residualgastriccontent wasfound. Pancreatic juice was aspirated manually in freely moving rats and quantitatedin 15-minintervals for 1 h. In stressexperiments,physical restraint, apreviouslyvalidated stressor (5, 18), was used.Briefly,the animal's trunk was secured to a woodenplatefor 1 h with three metal clamps. This restraint at room temperature results in characteristic neuroendocrine and autonomic stress responses(5, 18).Control (sterile water)orpeptideswereinjected cerebroventricularlyin5-41 volumes andintravenouslyin50-,glvolumes as described(5, 17, 18).CRF or control wasinjectedatthezerotimepoint,whereasa-helical CRF-(9-41)wasinjectedatthe -1 5-mintimepoint. Peptidedoses were admin-istered in random order. At the end of theexperiment, animalswere killedbycervicaldislocation;aclean empty stomach and correct cere-bral cannulaplacementwerealso verified(17).Todelineate the neuro-humoralpathwaysmediatingtheeffectsof stress andofCRF on exo-crine pancreatic secretion in rats, the same pharmacological ap-proacheswereusedasdescribed inthe dog experiments.Dosages and timecourseofadministrationwerelikewisethe same.

Biologicalmeasurements.Gallbladder sizewasmeasured by real-timeultrasonography usinga5-MHz transducer(Toshibamodel SSA-77A)in theupright position.The transducer wasplacedin the right upper abdominalquadrant,and the greatestlengthofthe gallbladder

wasdeterminedon anoscilloscopescreenwith standardized calipers. Rotatingthe transducer by 900, wedetermined the greatest width (transversediameter)and the greatestdepth(anterior-posterior diame-ter). Gallbladder volume was calculated using the sum-of-cylinder methodasdescribed (24). Thismethodhas been shownpreviouslynot

tooverestimategallbladdervolume(24).Toreduce thevariabilityof gallbladder volumes, three measurements were taken ateach time

point (timepoint±2 min)andaveraged. Pancreaticvolumeswere esti-matedbyweighingthe collected 1 5-minsamples, assumingadensity of 1.0. Bicarbonate wasdetermined on anautomatedtitration system (Radiometer, Copenhagen, Denmark) using apreviously validated back-titration method for low bicarbonate concentrations(18).Protein

wasmeasuredasdescribed (25).

Plasma measurements. 2 mlofvenousblood wasremoved, col-lected inchilled tubescontaining100,l0.5M EDTA(EGTA for cate-cholamines),andcentrifugedat3,000gat4VC for10min;plasmawas storedat-20'C.Plasma concentrations ofcortisol, ACTH,and (3-en-dorphinweredetermined induplicate byradioimmunoassay (DRG-InstrumentsGmbH,Marburg, FRG) (18).Plasmaconcentrations of epinephrineand norepinephrineweremeasured by asingle isotope

radioenzymatic method(5, 17, 21).

S-[methyl-3]Adenosyl-L-methio-ninewaspurchasedfrom NewEnglandNuclear(Boston,MA).Plasma glucose concentrations were determined by the glucose-oxidase method(Beckman Instruments,Inc.,PaloAlto,CA)(5,21).

Statisticalanalysis.The dataweresubjectedtoanalysisof variance and differences between treatmentgroups weredetermined by the Newman-Keuls multiple-rangetest(21). Results obtained from five animalswereexpressedasmeans±SEM,andtheywereconsidered sig-nificant if P<0.05.

Results

Effects ofCRFandstress.Acousticstressand CRF(0.5nmol/ kg given cerebroventricularly) significantly increased plasma

concentrations ofnorepinephrine, epinephrine, glucose,

corti-sol,ACTH,and

_,l-endorphin

(Table I). Basalgallbladder

vol-umesrangedbetween 12.5 and 13.5 cm3(Fig. 1A),andbasal

pancreaticvolume secretionrangedbetween 465 and515 ,ul/h (Fig. 1B).Indogs,CCKalonereduced gallbladdervolume to 3.5 cm3at45 min. Increasingdoses of cerebral CRF signifi-cantlyreversed CCK-induced gallbladdercontraction; atthe

highestdose ofCRF(0.5 nmol/kg),gallbladder volumeswere not significantly different compared with resting gallbladder

Table I.NeuroendocrineEffectsofAcousticStress and CRF

Control Stress CRF

0 min 20 min 0min 20min 0min 20 min

Norepinephrine (nM) 1.17±0.21 1.37±0.24 1.00±0.21 2.74±0.44 1.10±0.18 2.98±0.53

Epinephrine(nM) 0.52±0.11 0.59±0.17 0.61±0.10 1.12±0.24 0.48±0.08 1.06±0.13

Cortisol_(AM) 0.51±0.11 0.60±0.21 0.53±0.08 2.22±0.54 0.49±0.24 2.08±0.37

ACTH (pM) 20±3 23±4 24±4 230±39 26±4 271±41

$-Endorphin(pM) 28±3 36±5 25±3 210±27 31±4 202±25

Glucose(mM) 4.8±0.2 5.0±0.3 4.9±0.3 7.4±0.6 4.8±0.4 7.2±0.4

Acousticstress was begun at 0 min and CRF (0.5 nmol/kg) or control(sterilewater) were givencerebroventricularlyat0 min. At 20min,there was asignificant(P < 0.01) increase in plasma levelsofall parameters measured in stressed and CRF-treatedanimals.

CRFsignificantly inhibited pancreaticvolume secretionfor1h (Fig. 1 B). In addition, cerebral CRF (1 nmol/kg) inhibited total pancreatic protein secretion from 17.4±1.9 to 8.1±0.9

mg/h (P < 0.01) and total pancreatic bicarbonate secretion from 29.5±3.1 to 10.7±0.9 Lmol/h (P < 0.01). In contrast,

intravenousadministration of CRF (0.5 nmol/kg in dogs and 1.0 nmol/kg in rats) didnot significantly alter CCK-induced gallbladder contractionorexocrinepancreaticsecretion (data

notshown).

Acousticstressindogssignificantly reversed CCK-induced gallbladder contraction; theeffect ofstresswascomparableto

15-

10-5-1

A

*stress

A0.5

_)CRIF

.0.05)nmoVkg

O control

0 30

Time (min)

151

_c

10

-

5-CRF (0.1 nmol/kg

/Z

600

-400

-200

-60L.

0

TCl)

0

5).)

L-c

600

-400

-200

-theeffectof cerebral CRF(0.1 nmol/kg; Fig. 1 C). Restraint

stressinratsand cerebral CRF(1 nmol/kg) inhibited

pancre-aticvolume secretion similarly (P<0.01; Fig. 1 D). Restraint

stressalsoinhibited totalproteinoutputby 52% and total bi-carbonateoutputby 63% (P<0.01). Cerebraladministration ofa-helical CRF-(9-4 1) 15 minbeforestress exposure or

cere-braladministrationofCRFcompletely reversedtheir effectson

gallbladder volume(Fig. 1C) andexocrine pancreatic secretion (Fig. 1 D). Theseresponses weremeasured45 minafter CRF (or control)administrationorstressexposure. InCRF-treated animalsand in stressed animals, thehighestdoseofa-helical

Figure 1. Effects of stress and cere-bralCRF on caninegallbladder contractionand on ratpancreatic B secretion. (A)All groupsreceived

CCK(20 pmol/kgperh) from time zero to 60min.Acousticstress(0-60

min)andcerebral CRF(at0.1 and

KT*

0.5nmol/kg)givenattimezero

sig-nificantly (P<0.01)reversed CCK-induced gallbladder contraction for

1 h. *P<0.05comparedwith

con-trol._(B)Restraintstress_(0-60

_min)

*stress _{and cerebral CRF (at 1 nmol/kg)}

*0

}5

CRF given attimezerosignificantly(P *₀₁

nmo/g

<0.01)inhibitedpancreatic secre-o csecre-ontrsecre-ol tion for 1 h. *P < 0.05 and **P

<0.01

comparedwith control.(C)

0 30 60 _{All groups received CCK (20}

_pmol/

Time(min) kg per h), and gallbladder volumes were measured at 45 min. The CRF

D

antagonist,a-helicalCRF-(9-41) [CRF-(9-41)], given 15 min before C RF

_n**

stressexposure orcerebral CRF, (1 nmol/kg) > dose-dependently reversed

stress-\,1>and CRF-induced gallbladder*

relax-/t*

ation. Conabscissa denotes animals treatedwith CCK alone. *P<0.05

stress and **P <0.01 compared with zero

dose(sterilewater).(D) Cerebral CRF-(9-41)given 15 min before stressexposure or cerebral CRF dose-dependently reversed

stress-and CRF-induced inhibition of

pan-O

O 1

1 0

1

O

E3

creaticsecretion.Bonabscissa

de-0 0.1 1.0 10 B

notesbasalsecretion. P<0.05 and CRF-(9-41)(nmol/kg) **P<0.01compared with zero dose. E

E

a)

D

.0

0 0.1 1.0 10 C

CRF-(9-4 1) (10 nmol/kg) produced responses that were similar tothose observed with CCK alone (Fig. 1 C) and that were similarto basalpancreaticsecretion (Fig. 1 D).

Stimulated gallbladder contraction and pancreatic secre-tion. Toexaminetheeffects ofstress and CRF on endogenously induced gallbladder contraction, a balanced, blenderized mixed meal was administered intragastrically. Both acoustic stress and cerebral CRF (0.5 nmol/kg) reversed meal-induced gallbladder contraction (Fig.2A). At the30-,45-, and 60-min intervals, gallbladder volumes in stressed and in CRF-treated animalswere nearly identicalto basal volumes. To simulate postprandial exocrine pancreaticsecretion, a combined infu-sion of CCK and secretin was administered intravenously. Both restraint stress and CRF abolished the protein and bicar-bonateresponses(Fig.2,Band C)aswellasvolumesecretion (not shown). In control experiments, the stimulated responses were protein, 34.3±0.4 mg/h; bicarbonate, 40.1±0.5 umol/h; and

volume,

985±57 ul/h. Instressedanimals, therespective responses were 20.4±1.5 mg/h, 29.4±2.4Mmol/h,and 595±42

il/h

(P

< 0.01

compared

with the

_{appropriate control).}

In CRF-treated animals,therespective responses were 20.5±0.9 mg/h,

28.5±2.9,umol/h,

and

575±44,ul/h

(P<0.01 compared with the appropriate control).

151

A

C

E

E

10--o

5-cu

40

-C)

E

30-o

6

iw

0-0

50-E

3-cn

° 30

-I s

k - --I

* stress

A CRF(0.5nmoVkg)

O control

B ---o

.

Ht--I

C

-

~~~-.1 1

.,

0 30

Time(min)

Figure 2. Effects ofstressandcerebral CRFonmeal-stimulated gall-bladder contraction(A)andCCK/secretin-stimulatedpancreatic

se-cretions (B andC).Themixed meal(240 ml)wasadministered in-tragastricallyattimezero.CCK and secretin(10pmol/kgper heach) weregivenintravenously (0-60 min). Bothstressand CRF signifi-cantly(P<0.01)reversed the effects of the mixed meal(A,30-60 min) and the effects ofCCK/secretin (BandC, 15-60min).

Neurohumoralpathways. Fig. 3 depicts the effects ofgangli-onicblockadewithchlorisondamine, noradrenergic blockade with bretylium, and truncal vagotomy on gallbladder volume (Fig. 3 A) and pancreatic secretion (Fig. 3 B) in the resting state, during stress, and aftercerebral administration of CRF (0.1 nmol/kgindogs and 1.0nmol/kg in rats). Neither stress nor cerebral administration of CRFaltered resting (fasting) gall-bladdervolume(Fig.3 A), butchlorisondamine(P > 0.05) and vagotomy (P < 0.05) diminished basal pancreatic secretion, whereasbretyliumdid not (Fig. 3 B). In dogs, both chlorisonda-mine and bretylium, but not vagotomy, prevented the reversal of CCK-induced gallbladder contraction produced by CRF and stress that was observed in animals treated with NaCl (Fig. 3 A).Similarly,in rats,chlorisondamineand bretylium, but not vagotomy, abolished the inhibitory actions of CRF and stress onexocrinepancreatic secretion (Fig. 3 B).Hypophysectomy in rats,adrenalectomy, propranolol, and naloxone in rats and dogs, as well as exogenous infusionofACTH or

_f3-endorphin

(100pmol/kg per h each) administered to rats and dogs, did not significantlyalter theeffects ofcerebral CRF and stress on gall-bladder volume andpancreatic secretionduring the basal state (TablesIIandIII).

Noradrenergic pathways. To determine whether norepi-nephrineis the common peripheral neurotransmitter mediat-ingtheeffects ofstressand CRF on gallbladder motility and exocrinepancreatic secretion, this catecholaminewas given ex-ogenouslyand thebiological responses were measured. Table IV shows that norepinephrine (100 nmol/kg per h) produced plasma levels that were similar to those previously observed in response to CRF andrestraintstress in rats (4) and to those observed in response to CRF in dogs (21) and in response to acousticstress(TableI).Thisdoseofnorepinephrinealso pro-duced areversal of CCK-induced gallbladdercontraction(Fig. 4A)andinhibition of exocrine pancreaticsecretion (Fig. 4 B). These responses were similartothose produced by stress and cerebral CRF

(0.1

nmol/kg in dogs and 1.0 nmol/kg in rats). Increasingdoses of thea-adrenergicreceptor antagonist phen-tolamine significantly reversed the effects ofcerebral CRF, stress, and exogenous norepinephrine ongallbladder volume andpancreatic secretion (Fig.4). At thehighestdoseof phen-tolamine(10

,mol/kg

perh), gallbladdervolumeand pancre-aticsecretion in all treatmentgroups weresimilarcompared withgallbladdervolume indogs treatedwith CCKalone and in ratsat basal, respectively. Thedifference between pancreatic secretion in stressed animals treated with

phentolamine (10

,gmol/kg

per h) and basal secretion was not significant (Fig.

4B).

Discussion

Gallbladder contractionandexocrinepancreaticsecretion sub-serveimportant digestivefunctionsfacilitatingthe coordinated postprandial digestionandsubsequentabsorption offat, pro-tein,andcarbohydrates

(1).

Both thegallbladderandthe

exo-crine pancreas enjoy a rich supply of

noradrenergic

nerve fibers, but theirphysiological (or pathophysiological)

impor-tancefor gut functionhas notbeendefined(6, 7).

Using

two new animal models, wehavefound that,

during

stress, activa-tionofnoradrenergic outflowtothe gut mediatedbycerebral

CRFproduces inhibition ofcaninegallbladder contractionand

Figure3.Neuronal pathways mediatingstress-and

CRF-inducedgallbladder relaxation and inhibition

ofexocrinepancreatic secretion. (A) Gallbladder

vol-umewasmeasured 45 min after administration of

CCK,cerebralCRF (0.1 nmol/kg), and acousticstress.

(B)Pancreatic secretionwasmeasured inresponseto

cerebralCRF (1 nmol/kg) and restraintstress. Gan-glionic blockade with chlorisondamine(1 mg/kg), truncalvagotomy,andnoradrenergic blockade with bretylium (10 mg/kgperh)wereperformed.**P

<0.01 compared with CCK + NaCl (A) or with basal(B).

Cerebral administration of CRFdose-dependently

inhib-ited CCK-induced gallbladder contractioninawakedogsand basalpancreaticsecretions(volume, protein,andbicarbonate)

in awake, freely movingrats. CRF alsoinhibited gallbladder contractionproduced byaphysiological stimulus,abalanced mixedmeal, and CCK/secretin-stimulated exocrine pancreatic

secretions. Incontrast,ahigh doseofCRF administered intra-venously didnotalterbiliary and pancreatic functions, suggest-ing that CRF exerts its biological actions within the central nervoussystem atayet-undefined site.

We havepreviously shown that restraintstresselicits char-acteristic neuroendocrinestressresponsesinrats(4). We have nowdemonstrated thatacoustic stressincreasesplasma

con-centrations ofnorepinephrine, epinephrine, glucose, cortisol, ACTH, and

f3-endorphin.

These responses arequantitatively

similar to those observed after exogenous administration of

CRF(4, 21).Asimilartypeof acousticstresshas been

demon-stratedtoinhibitcaninegastricmotorfunction (26, 27).Ina mannercomparabletothehigher doses of exogenously admin-istered CRF, acousticstressinhibits CCK- and meal-induced

caninegallbladder contraction and restraintstressinhibits

ba-salexocrinepancreatic secretioninrats.Theresponseselicited bybothstressandexogenousCRFwereinhibitedbycerebral

administrationof the CRF receptorantagonist,a-helical

CRF-(9-41)(23), inadose-related fashion. Theseresults indicatethat CRF is released in response to different stressors in distinct

species, exertingitscentralnervoussystem effectsongut

func-tion via specific CRF receptors. Ofinterest, the dose of the

antagonist requiredtoabolish theeffects ofCRFwas 10times

less inratsthanindogs. Thismaybesecondarytodifferences in peptide distribution ortissue penetration ortodecreased affinity of the canine CRFreceptorfora-helicalCRF(9-41).

Previously validated surgical and pharmacological

ap-proacheswereutilizedtodelineate the neurohumoral efferents mediating the effects ofstressand CRF. Ganglionic and norad-renergic blockade and truncalvagotomydidnotsignificantly altergallbladder volume, indicatingthatautonomicefferents

donotexertatoniccontrolovergallbladdervolumeatrest(6).

Incontrast,ganglionic blockade with chlorisondamine slightly (but insignificantly) and vagotomy significantly diminished

resting pancreatic secretion, confirming that vagal efferents

controlrestingexocrinepancreatic secretion (7). However,the effectsof acousticstressindogsand the effectsofrestraintstress inratsongallbladder volume andonpancreatic secretion, re-spectively,werecompletelyabolishedbyganglionicand norad-renergic blockade.Incontrast, vagotomy,adrenalectomy, and

Table II. RoleofNeuroendocrine PathwaysMediatingStress- and CRF-induced Gallbladder Relaxation

Control Stress CRF

0 min 30 min 0 min 30min 0 min 30 min

cm3

Adrenalectomy 12.8±1.9 5.0±0.9 12.1±0.9 11.4±0.8 10.4±0.9 9.9±1.6

Propranolol 11.1±1.4 5.1±1.1 10.4±1.3 10.4±1.8 11.4±1.4 9.7±0.9

Naloxone 12.9±1.4 6.2±1.4 11.3±2.0 10.5±1.8 10.7±1.9 9.6±1.1

fl-Endorphin 13.1±1.5 5.9±1.3 12.3±0.8 10.4±0.8 13.0±1.8 9.9±1.4

ACTH 10.7±2.1 4.4±0.8 13.2±1.7 11.2±0.9 11.3±1.1 9.4±0.8

Ineachtreatmentgroup, acoustic stress and cerebral CRF (0.5 nmol/kg) reversed CCK-induced gallbladder contraction compared with control experiments. Agents were administeredasdescribed in Methods.

15i

A

** **

10-

5-**

Ofasting * CCK+NaCI

0CCK+CRF

gqCCK+stress

0)

E

0

0

0

0

Table III. Role ofNeuroendocrine Pathways Mediating Stress- and CRF-induced Inhibition ofPancreatic Secretion

Control Stress CRF

0min 30 min 0min 30min 0 min 30min

,I/h

Hypophysectomy 440±49 420±39 470±41 220±27 460±36 23 1±40

Adrenalectomy 490±56 459±46 474±49 279±43 474±39 230±47

Propranolol 510±67 480±57 539±86 241±68 510±47 276±56

Naloxone 495±45 439±44 522±37 260±34 513±56 269±37

f3-Endorphin 420±39 403±42 408±44 243±31 428±47 210±33

ACTH 481±56 473±71 505±49 257±56 496±39 260±22

Ineach treatment group, restraintstressandcerebral CRF(1 nmol/kg) significantly(P<0.01) inhibited pancreaticvolumesecretion. Agents wereadministeredasdescribedin Methods.

hypophysectomy(in rats) didnotalter theeffects ofstressand CRFonbiliaryandpancreaticfunction. These results indicate thatbothstress and CRFexerttheir effectsonthegallbladder and theexocrinepancreasviasympathetic noradrenergic effer-ents. This conclusion is further strengthened by the lack of effect of adrenalectomy,

03-adrenergic

blockade, naloxone, or

exogenous administrationof ACTH and

f3-endorphin.

These results concur with previous observationsthat indicate that CRFexertsitseffectsongastrointestinal secretoryandmotor function by modulating autonomic nervous system activity (13-17,21,22).Incontrast,stress-induced stimulationof

mu-rine duodenal bicarbonate secretion ismediated by endoge-nous CRF andsubsequentreleaseof

f-endorphin

(18),butthis pathway appearstobeanexception.

Cuttingthesplanchnicnervesincreasesvagallystimulated

pancreatic

secretion

(28),

whereas splanchnic nerve stimula-tion inhibitspancreatic secretion (29). Catecholamines have been

implicated

tomediatethese responses either

directly

or indirectly by decreasing blood flow to the pancreas

(30).

Whereas

,3-adrenergic

agonists have been reported either to

have no effector tocause inhibition of stimulated

secretion,

a-adrenergic agonists clearlycauseinhibition in vivo

(31

and referencestherein). Similarly, stimulation of

splanchnic

nerves

inhibitsthecontractile effects of CCKonfeline

gallbladder (6),

and norepinephrine relaxes

guinea

pig

and cat

gallbladder

in vivoand invitro(6andreferences

therein;

also

30, 32-34).

In additiontothese

pharmacological observations,

ourresults in-dicatethatendogenous stimulation of

sympathetic

noradrener-15I

A

co

E

E10-Xc 5-0

D

CCK+NE

CCK+stress

C

OCK

CCK+CRF

gicoutflowby twodifferent stressors inhibits murine pancre-aticandcanine biliarymotorfunction.

It wasof interest that endogenouslyreleased CRF appeared to exertits actionson twodifferent peripheralorganfunctions in distinct species by a common noradrenergic pathway. To further investigate the importance of noradrenergic control overbiliaryandpancreatic function duringstress, norepineph-rinewasadministeredat adosethat produced plasma concen-trations similar to those observed in response to CRF and stress. Ofnote, intravenousadministration ofnorepinephrine produced reversal of CCK-induced gallbladder contraction and inhibition ofpancreatic secretion (29, 32, 34) that were similartothose observed in response to stress orCRF. Further-more, theeffects ofall threestimuli-exogenous CRF, stress, andnorepinephrine-weredose-dependentlyabolishedby pe-ripheraladministration ofthea-adrenergicreceptorantagonist phentolamine but notby propranolol. These results indicate thata-adrenergic receptorsarelikelytobe the final target for norepinephrinereleased inresponse to stress and exogenously administeredCRF. Thesefindingsareinaccordwithprevious observationsthat suggest that CRF and stress exerttheircentral nervoussystem actionsonthe uppergastrointestinaltractvia sympathetic

noradrenergic

efferents (5, 15-17, 21, 22, 35).

Our results differ from previous studies in anesthetized cats,which indicatedthat therelaxing effect of

norepinephrine

onthegallbladderwasmediated by,B-adrenergicreceptors(33).

Inthisinvestigation,removalof the

primary

_{f3-adrenergic}

ago-nistepinephrinebyadrenalectomyand

fl-adrenergic

blockade

6001 B

c

.O

400-0

D 0

*.a

cn₀

200-0

CL

I I I

0 0.1 1 10

Phentolamine

_(jgmoL/k

NE Figure4. Reversal ofCRF-,stress-,and norepineph-.4

-^ rine

(NE)-induced

gallbladder relaxation and inhibi-** tion ofpancreatic secretion by the a-adrenergic re-** CRF ceptorantagonistphentolamine. (A) Gallbladder

vol-ume wasmeasuredat45 min after administration of CCK,cerebral CRF(0.1nmol/kg),intravenous NE

stress (100 nmol/kgperh), and acoustic stress.(B) Pancre-atic secretionwasmeasuredin responsetocerebral CRF(1 nmol/kg), intravenousNE(100 nmol/kg per h), and restraintstress. Inboth species, phentolamine ' ' (at 1 and 10,umol/kgperh)significantly (**P<0.01) 0 0.1 1 10 reversed the biological actions produced by stress,

TableIV.Plasma Concentrations ofNorepinephrine

Norepinephrine

0 min 20 min 40 min 60 min

nM

Rats 0.78±0.15 1.90±0.31 2.47±0.48 2.22±0.57

Dogs

_1.10±0.27

_2.89±0.41

_3.22±0.65

_3.02±0.52

Norepinephrine (100 nmol/kg per h) wasinfused intravenously from 0 to 60 min.Plasma levels weresignificantly(P<0.01)increased (20-60 min) butnotsignificantlydifferent from those after acoustic stressand CRF indogs (Table I)orafter restraintstressand CRF in rats(reference4).

did not alter CRF- and stress-inducedinhibition ofgallbladder contraction in awake, nonanesthetized dogs. Furthermore, propranololdidnotinhibit therelaxingeffects of norepineph-rine on CCK-induced gallbladder contraction (unpublished observations). Finally, inthis study, a "physiologic" dose of

norepinephrine was used that mimicked plasma concentra-tions that were observed in responseto stress. Species differ-ences andstudydesign (bolusvs.continuousinfusion of ago-nist and antagoago-nist, as well as the state ofanesthesia)may ac-count for these diverse observations. Thespecificsite of action fornorepinephrineto inducea-receptor-mediatedcanine gall-bladdercontraction remainstobe definedinvitro.

We conclude that stress, exogenousCRF, and norepineph-rine inhibit canine gallbladder contraction and murine

exo-crine pancreatic secretion. Different stressors (acoustic and

restraint stress) in distinct species release CRF that acts on

specificCRF receptors within the brain. Release of norepineph-rine appears to be the final commonpathwayproducing inhibi-tionofbiliaryandpancreaticfunctionduringstressmediated by endogenous CRF.

Acknowledgments

Theauthors thank K. Feutlinske and E. Friemel for technical

assis-tanceand J. E.Rivier forgeneroussupply ofpeptides.

This work wassupported by the Deutsche Forschungsgemeins-chaft, Hamburgische Wissenschaftliche Stiftung,and theNational In-stitutes ofHealth(NIH). H. J. Lenzisrecipient ofanNIHClinical InvestigatorAward.

References

1.Johnson,L.R.1987.PhysiologyoftheGastrointestinalTract. RavenPress,

NewYork. 1780 pp.

2.Holzl,R., and W. E. Whitehead. 1983.Psychophysiology

oftheGastrointes-tinal Tract. Plenum Press, New York. 360 pp.

3.Williams,C. L., J. M. Peterson, R. G. Villar, and T. F. Burks. 1987.

Corti-cotropin-releasing factor directly mediatescolonic responses to stress. Am.J.

Physiol. 253:G582-G586.

4.Lenz, H. J.,A.Raedler, H. Greten, W. W. Vale, and J. E. Rivier. 1988.

Stress-induced gastrointestinal secretoryand motor responsesin rats are

me-diated by endogenous corticotropin-releasingfactor.Gastroenterology.95:15

10-1517.

5. Lenz, H. J. 1989. Neurohumoral pathways mediating stress-induced

changesinratgastrointestinaltransit.Gastroenterology.97:216-218. 6. Ryan, J. P. 1987. Motility of the gallbladder and biliary tree. In Physiology of the Gastrointestinal Tract.L. R.Johnson, editor. Raven Press, New York. 695-722.

7.Singer,M.V.1986. Neurohumoral controlofpancreatic enzyme secretion in animals. In The Exocrine Pancreas: Biology, Pathobiology and Diseases.

V.L.W.Go,J.D.Gardner,F. P.Brooks,E.Lebenthal,E.P.DiMagno,and G.A.Scheele,editors. RavenPress,NewYork. 315-332.

8.Porte, D.,Jr. 1973. Neuralregulationof insulin secretion in thedog.J. Clin.

Invest.52:210-214.

9.Ahren, B.,R.C.Veith,andG.J.Taborsky,Jr. 1987.Sympatheticnerve

stimulationvs.pancreatic norepinephrineinfusioninthedog.1.Effectsonbasal

insulin and glucagon release. Endocrinology. 121:323-331.

10.Marliss,E.B.,L.Girardier,J.Seydoux,C.B.Wollheim,Y.Kanazawa,L.

Orci,A.E.Renold,andD.Porte,Jr.1973.Glucagon secretion induced by pancre-aticnervestimulation in thedog.J. Clin. Invest. 52:1246-1249.

11. Havel, P.J.,R.C. Veith, B. E. Dunning,and J. Taborsky,Jr. 1990. Pancreaticnoradrenergic nerves areactivatedbyneuroglucopeniabutnotby

hypotensionorhypoxiain thedog. Evidence forstress-specificandregionally

selective activation ofthesympatheticnervoussystem. J. Clin. Invest. 82:1538-1545.

12.Rivier,J.,J.Spiess,and W. Vale.1983. Characterization ofrat

hypotha-lamiccorticotropin-releasingfactor.Proc.Natl. Acad.Sci. USA.80:4851-4855.

13.Tache, Y.,Y.Goto,W.Gunion,W.Vale,J.Rivier,and M. Brown. 1983.

Inhibitionofgastric acid secretioninratsbyintracerebralinjectionof

corticotro-pin-releasing factor.Science(Wash. DC).222:935-937.

14.Tache, Y.,M. Maeda-Hagiwara,andC. M.Turkelson. 1987. Central

nervoussystem actionofcorticotropin-releasingfactortoinhibitgastricemptying

inrats.Am.J.Physiol.253:G241-G245.

15.Lenz,H.J.,S. E.Hester,and M. R. Brown.1985.Corticotropin-releasing

factor: mechanismstoinhibitgastricacid secretioninconsciousdogs.J. Clin. Invest.75:889-895.

16.Lenz,H.J.,M.Burlage,A.Raedler,and H.Greten.1988. Centralnervous

systemeffects ofcorticotropin-releasingfactorongastrointestinaltransit in the

rat.Gastroenterology.94:598-602.

17.Druge, G.,A.Raedler,H.Greten,and H. J. Lenz. 1989.Pathways

mediat-ingCRF-inducedinhibition ofgastricacidsecretioninrats. Am. J.Physiol.

256:G214-G219.

18.Lenz,H.J. 1989.Regulationofduodenalbicarbonatesecretionduring

stressbycorticotropin-releasingfactorandf3-endorphin.Proc.Natl.Acad.Sci. USA. 86:1417-1420.

19.Gunion,M.W.,G. L.Kauffman,Jr.,andY.Tache. 1990.

Intrahypothala-miccorticotropin-releasingfactorelevatesgastricbicarbonate andinhibitsstress

ulcers inrats. Am.J.Physiol258:GI52-G157.

20. Brown,M.R.,L.A.Fisher,J.Spiess,C.Rivier,J.Rivier,and W. Vale.

1982.Corticotropin-releasingfactor:actiononthesympatheticnervoussystem andmetabolism.Endocrinology. 11:928-931.

21.Lenz,H.J.,A.Raedler,H.Greten,and M. R. Brown.1987.

Corticotro-pin-releasingfactor initiatesbiologicalactionswithinthebrainthatareobserved in responseto stress.Am.J.Physiol.252:R34-R39.

22.Lenz,H. J. 1989.Neuroendocrineregulationofcaninegastricsecretion, emptyingandbloodflow.J.Neuroendocrinol.1:163-167.

23.Rivier,J.,C.Rivier,and W. Vale.1984.Synthetic competitiveantagonists

ofcorticotropin-releasingfactor: effectonACTHsecretionin therat.Science

(Wash. DC).224:889-891.

24.Everson,G.T.,D. Z.Braverman,M. L.Johnson,and F.Kern,Jr.1980.A

critical evaluationof real-timeultrasonographyforthestudyofgallbladder

vol-umeandcontraction.Gastroenterology.79:40-46.

25.Lowry, 0.H.,N. J.Rosebrough,A.L.Farr,and R. J. Randall. 1951.

Proteinmeasurementwith the Folinphenolreagent. J.Biol. Chem. 193:265-275. 26.Gue, M.,and L. Bueno.1986.Diazepamandmuscimol blockadeofthe

gastrointestinalmotordisturbances inducedbyacousticstressindogs.Eur. J.

Pharmacol. 131:123-127.

27.Gue,M.,J.Fioramonti,J.Frexinos,M.Alvinerie,and L. Bueno. 1987.

Influence of acousticstressbynoiseongastrointestinal motilityindogs. Dig.Dis. Sci.32:1411-1417.

28.Holst,J.J.,S.Khnuhtsen,S. L.Jensen,0. V.Nielsen,and S. S. Poulsen. 1984. ParacrinesomatostatinandGRP-nerves in the controlofantralgastrin

release.Dig.Dis.Sci.29(Suppl.):37S.

29.Barlow,T.E.,J. R.Greenwell,A. A.Harper,andT.Scratcherd. 1971. The effect of adrenaline andnoradrenalineonbloodflow,electrical conductance and external secretionofthe pancreas. J.Physiol. (Lond.).217:665-678.

30.Demol, P.,andH.Sarles. 1980. Actionofcatecholaminesonthe exocrine

pancreaticsecretionof consciousrats.Arch.Int.Pharmacodyn.Ther. 243:149-163.

31.Holst,J.J. 1986. Neuralregulationofpancreaticexocrinefunction.In TheExocrinePancreas:Biology, Pathobiology,andDiseases.V. L. W.Go,J. D.

Gardner,F. P.Brooks,E.Lebenthal,E. P.DiMagno,andG.A.Scheele,editors.

RavenPress,NewYork.287-300.

32.Palin, B.,and S.Skoglund.1964. Neural and humoral control of

gallblad-deremptyingmechanism in thecat.ActaPhysiol.Scand.60:358-362. 33.Persson,C. G.A.1973. DualeffectsonthesphincterofOddi and

gallblad-derinducedbystimulationof theright splanchnicnerves.ActaPhysiol.Scand.

87:334-343.

34.Bertaccini, G.,G.DeCaro,R.Endean,V.Erspamer,and M.Impicciatore.

1986.Theactionof caeruleinonthe smoothmuscleofthegastrointestinaltract

and the gallbladder. Br. J.Pharmacol. 34:291-3 10.

35.Lenz, H.J.,andG. Druge. 1990. Neurohumoralpathwaysmediating

stress-induced inhibition ofgastric acid secretion in rats. Gastroenterology.