doi:10.1152/ajpregu.00781.2009 298:1240-1248, 2010.Am J Physiol Regulatory Integrative Comp Physiol

Henrik Seth, Albin Gräns and Michael Axelsson

You might find this additional information useful...

46 articles, 18 of which you can access free at: This article cites http://ajpregu.physiology.org/cgi/content/full/298/5/R1240#BIBL

on the following topics: http://highwire.stanford.edu/lists/artbytopic.dtlcan be found at Medline items on this article's topics

Veterinary Science .. Trout Physiology .. Salmoniformes Physiology .. Blood Circulation Medicine .. Hyperemia Oncology .. Cholecystokinin/pancreozymin Medicine .. Cholecystokinin

including high-resolution figures, can be found at: Updated information and services http://ajpregu.physiology.org/cgi/content/full/298/5/R1240

can be found at: and Comparative PhysiologyAmerican Journal of Physiology - Regulatory, Integrativeabout Additional material and information

http://www.the-aps.org/publications/ajpregu

This information is current as of April 29, 2010 .  

http://www.the-aps.org/.ISSN: 0363-6119, ESSN: 1522-1490. Visit our website at Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2005 by the American Physiological Society. ranging from molecules to humans, including clinical investigations. It is published 12 times a year (monthly) by the Americanilluminate normal or abnormal regulation and integration of physiological mechanisms at all levels of biological organization,

publishes original investigations thatThe American Journal of Physiology - Regulatory, Integrative and Comparative Physiology

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

Cholecystokinin as a regulator of cardiac function and postprandialgastrointestinal blood flow in rainbow trout (Oncorhynchus mykiss)

Henrik Seth, Albin Gräns, and Michael AxelssonDepartment of Zoology, University of Gothenburg, Gothenburg, Sweden

Submitted 24 November 2009; accepted in final form 16 February 2010

Seth H, Gräns A, Axelsson M. Cholecystokinin as a regulator ofcardiac function and postprandial gastrointestinal blood flow in rain-bow trout (Oncorhynchus mykiss). Am J Physiol Regul Integr CompPhysiol 298: R1240–R1248, 2010. First published February 17, 2010;doi:10.1152/ajpregu.00781.2009.—We have studied the potential roleof CCK as a regulator/modulator of the postprandial increase ingastrointestinal blood flow. Rainbow trout (Oncorhynchus mykiss)were instrumented with pulsed Doppler flow probes to measure theeffects of CCK on cardiac output and gastrointestinal blood flow.Furthermore, vascular preparations were used to study the directeffects of CCK on the vessels. In addition, we used in situ perfusedhearts to further study the effects of CCK on the cardiovascularsystem. When the sulfated form of CCK-8 was injected at a physio-logical concentration (0.19 pmol/kg) in vivo, there was a significantincrease in the gastrointestinal blood flow (18 � 4%). This increase ingastrointestinal blood flow was followed by a subsequent increase incardiac output (30 � 6%). When the dose was increased to 0.76pmol/kg, there was only a 14 � 6% increase in gastrointestinal bloodflow; possibly due to a dose-dependent increase in the gill vascularresistance as previously reported or a direct effect on the heart.Nevertheless, CCK did not affect the isolated vessel preparations, andthus, it seems unlikely that CCK has a direct effect on the bloodvessels of the second or third order. CCK did, however, have profoundeffects on the dynamics of the heart, and without a change in cardiacoutput, there was a significant increase in the amplitude (59 � 4%)and rate (dQ/dt: 55 � 4%; -dQ/dt: 208 � 49%) of the phasic flowprofile. If and how this might be coupled to a postprandial gastroin-testinal hyperemia remains to be determined. We conclude that CCKhas the potential as a regulator of the postprandial gastrointestinalblood flow in fish and most likely has its effect by inducing agastrointestinal hyperemia. The mechanism by which CCK acts is atpresent unknown.

hormonal regulation; CCK-8s; circulating plasma levels; hyperemia;teleost

WHEN FOOD IS INGESTED, SEVERAL cardiovascular changes occur toenable the animal to efficiently digest, absorb, and redistributethe nutrients. The nature of these rearrangements, as well as theexact regulation governing these cardiovascular changes hasonly received limited attention in fish.

We have previously shown, in two separated teleost species,the rainbow trout (Oncorhynchus mykiss) (45, 46) and theshort-horn sculpin (Myoxocephalus scorpius) (44), that thepostprandial cardiovascular changes are influenced by mechan-ical and chemical stimuli, as well as the nutrient composition ina manner that is similar to what has been reported previouslyin mammals (reviewed by Refs 34 and 36). As a meal entersthe stomach, there is a �-adrenergic increase in the systemicvascular resistance. Subsequently, as food enters and is hydro-

lyzed in the proximal intestine, the free nutrient componentsinduce a local decrease in the gastrointestinal vascular resis-tance, and blood is thus shunted from the systemic to thegastrointestinal circulation. The increase in gastrointestinalblood flow (Qcma), resulting from this hyperemia is, in mostfish species studied, enhanced by an increase in cardiac output(CO) (5, 6).

The fact that the postprandial hyperemia is influenced by thenutrient composition of the diet is also in concordance withwhat has been seen in mammals (16, 24). The reason for thisdifference and the exact regulation governing the nutrient-induced hyperemia has not been determined. Several mecha-nisms have been proposed in mammals, such as direct effectsof the absorbed nutrients, metabolic and nonmetabolic vasoac-tive factors, as well as endocrine and neural involvement.These mechanisms have been extensively reviewed (25, 36),and the local decrease in tissue oxygen content, or alternativelyan increase in osmolarity, that occurs as a consequence ofnutrient uptake and assimilation, is most likely the most im-portant initial trigger (8). Also, other factors could contributeand modulate the response, since an increase in gastrointestinalblood flow is not necessarily linked to an increase in oxygenconsumption (specific dynamic action) and vice versa. It isnow believed that most of the specific dynamic action dependon factors outside the gastrointestinal tract (3), such as proteinassimilation (10, 11).

In mammals, several hormonal components have also beenproposed to be involved in this regulation. These includepeptides like CCK and vasoactive intestinal polypeptide (23).CCK is also a possible modulator of the postprandial cardio-vascular response to different diets in rainbow trout, since ithas been shown that the circulating plasma levels of CCKchange depending on the protein and lipid content of theingested meal (32).

CCK is a diverse peptide found in several different isoforms,as well as in numerous tissues in both mammals (13) andteleosts (28, 40). In rainbow trout, three different forms havebeen discovered, differing in the amino acid in position 6counting from the c-terminal (Fig. 1). Different posttranscrip-tional splice variants result in peptides of differing length (7, 8,or 21 amino acids), all sharing the same conserved region and asulfated tyrosine in position 7 from the c-terminal. The physio-logical significance of the difference in length and amino acidcomposition in position 6 is unknown; the different forms are,nevertheless, expressed differently in different parts of the trout(29). CCK is involved in the secretion of pancreatic juices indogs (38), stimulating gallbladder contractions (1, 2), aswell as regulating gastric emptying in rainbow trout (39)and controlling functions of the central nervous system,such as satiety in rats (12).

Address for reprint requests and other correspondence: H. Seth, Dept. ofZoology, Univ. of Gothenburg, Box 463, S-405 30 Gothenburg, Sweden(e-mail: henrik.seth@zool.gu.se).

Am J Physiol Regul Integr Comp Physiol 298: R1240–R1248, 2010.First published February 17, 2010; doi:10.1152/ajpregu.00781.2009.

0363-6119/10 $8.00 Copyright © 2010 the American Physiological Society http://www.ajpregu.orgR1240

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

CCK has previously been shown to affect the gastrointesti-nal blood flow in mammalian species (15, 27, 42). This effectcould be mediated through a neurogenic release of NO withCCK acting peripherally (43), although some suggest thiseffect is indirect via an increase in, for example, the pancreaticsecretion and thus a local vasodilation in this tissue (38), andstill others suggest that it is only vasoactive in high doses (41),doses that nevertheless could occur at discrete locations withinthe intestinal wall.

Overall, there seems to be many contradictory results con-cerning the importance of CCK in regulating or modulating thepostprandial hyperemia in mammals. In fish, even less isknown about how CCK might affect the gastrointestinal vas-culature, as well as the cardiovascular system, in general. We,therefore, aimed to investigate whether CCK participates in thepostprandial regulation of the gastrointestinal blood flow inrainbow trout. This was done in vivo with intra-arterial injec-tions of the sulfated form of CCK-8. The effect of CCK onisolated vascular preparations, as well as in situ perfused heartpreparations, was also assessed.

MATERIALS AND METHODS

Experimental Animals

Rainbow trout (O. mykiss), ranging in size between 380 and 750 g(mean: 487 � 21 g, n � 27), were purchased from a local hatchery.The fish were held in 2 m3 fiberglass tanks supplied with aeratedfreshwater (10°C) from the departmental recirculating water systemand fed dry trout pellets at regular intervals. The photoperiod wasadjusted to 12:12-h light-dark conditions. Upon arrival, fish were leftfor at least 1 wk prior to any experimental procedures. Ethical permit13/2007 from the Animal Ethics Committee of Gothenburg, coveredall experiments reported here.

In Vivo Surgical Procedures

Fish were fasted for �1 wk prior to surgery. Individual fish wereanesthetized in water containing MS-222 (150 mg/l) buffered withsodium bicarbonate (300 mg/l) and transferred to an operating tablecovered with soft water-soaked rubber foam. The gills were continu-ously irrigated with aerated freshwater (10°C) containing MS-222 (75mg/l) buffered with sodium bicarbonate (150 mg/l). To allow for theinjection of CCK into the bloodstream, the dorsal aorta was cannu-lated through the roof of the buccal cavity with a tapered polyethylene(PE)-50 (polyethylene) catheter, filled with heparinized (100 IU/ml)saline (0.9%), using a guide wire, according to the technique de-scribed by Axelsson and Fritsche (4). The catheter was tunneledthrough the snout and secured to the skin with sutures (3/0). CO was

measured with a 1.8-mm Doppler flow probe placed around theventral aorta. To place the cuff, the fish was positioned on its left side,and a small incision was made at the base of the fourth gill archdorsal to the ventral aorta. Connective tissue was carefully re-moved until the ventral aorta was visible, and there was enoughroom for the cuff. The cuff lead was secured with one suture closeto the cuff and one to the back of the fish. To measure gastrointestinalblood flow, a Doppler flow probe with a diameter of 1.1–1.2 mm wasplaced around the coeliacomesenteric artery (CMA), roughly 10 mmdistal to the bifurcation of the coeliacomesenteric artery, as describedby Seth at al (46). To access the coeliacomesenteric artery, the fishwas placed on its left dorsolateral side, and a 25-mm incision wasmade ventrodorsally from the base of the pectoral fin. The vessel wasdissected free using blunt dissection, taking care not to damage any ofthe surrounding nerves. Once the cuff was positioned, the lead wassecured with sutures to the skin, and the incision was closed withuninterrupted sutures.

In Situ Perfused Heart Preparations

As described previously, fish were anesthetized in water containingMS-222 (150 mg/l) buffered with sodium bicarbonate (300 mg/l) andtransferred to an operating table covered with soft water-soakedrubber foam. The technique used for the perfused heart preparationwas a modified version of the technique described by Franklin andAxelsson (22). On the operating table, the gills were continuouslyirrigated with aerated freshwater (10°C), and 2 ml of heparin (5000iIU/ml) was injected into the caudal artery. The hepatic veins werethen visualized through a ventral midline incision, and all but one wastied of using silk suture. A custom-made stainless-steel cannula(inflow) was introduced via the open hepatic vein, the cannula wassecured using a silk suture (3/0) and then checked for leakage. Theinflow cannula was connected to a Marriot bottle filled with heparanized(1%) Ringer solution (NaCl, 140.0 mM; NaHCO3, 15.0 mM; KCl, 2.5mM; CaCl2 � 2 H2O, 1.5 mM; KH2PO4, 1.0 mM; MgSO4 � 7 H2O, 0.8mM; HEPES, 5.0 mM; glucose, 10 mM; and adrenaline, 5 nM), with pHset to 7.8. The unbranched ventral aorta was then visualized by removingthe gills and another stainless-steel cannula (outflow) was then placedinto the ventral aorta just before the gills (secured with silk suture asabove). The fish was then removed from the operating table, and theentire fish was immersed into a cooled (10°C) bath containing saline(0.9%). The pericardium, as well as the surrounding nerves, was leftintact for the entire experimental protocol.

The inflow cannula was then disconnected from the Marriot bottleand attached to a constant pressure device, connected to a cooled(10°C) reservoir containing 1,000 ml of well aerated (0.3% CO2)Ringer solution (�0.5 nM adrenaline). The outflow cannula wasconnected to a transit-time flow probe (2N688 Flow through probe;Transonic, Ithaca, NY). The inflow and outflow pressures wereadjusted by means of a hoist and to measure the inflow and outflowpressures, the cannulas were connected to pressure transducers (MLD-type disposable pressure transducers).

Vascular Preparations

Fish were euthanized with a blow to the head, and the CMA wasdissected free and put on ice. Small rings with a width of �1 mm werecut from second- to third-order branches of the CMA and mounted onforce transducers (SENSELab 10-700-0003; Somedic Sales AB,Hörby, Sweden) for the measurement of vascular wall isometrictension. Preparations were submersed in cooled (10°C) and well-aerated (0.3% CO2) Ringer solution (NaCl, 140.0 mM; NaHCO3, 15.0mM; KCl, 2.5 mM; CaCl2 � 2 H2O, 1.5 mM; KH2PO4, 1.0 mM;MgSO4 � 7 H2O, 0.8 mM; HEPES, 5.0 mM), kept under a normal-ized tension (�1.0–1.5 g), as described earlier (47), and allowed torecover for 1 h to normalize the preparations. Preparations werechecked for viability using potassium-rich (�50 mM) Ringer solution.

Fig. 1. The sequence alignment of three different trout CCKs and a gastrin,compared with the mammalian CCK used in this study (in bold). Dashed boxindicate conserved region at the amidated c-terminal. Asterisks denote asulfated amino acid (tyrosine). Note the difference between the CCKs and thegastrin. -i- indicates that the peptide exists in several forms of varying length.

R1241INFLUENCE OF CCK ON THE CARDIOVASCULAR SYSTEM

AJP-Regul Integr Comp Physiol • VOL 298 • MAY 2010 • www.ajpregu.org

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

Experimental Protocols

In vivo recordings. Animals were allowed to recover for at least 24h prior to the experiment. To avoid too much twisting of the wires, thefish was not connected to either the Doppler amplifier or the pressuretransducer during the initial 6–12 h. Thereafter, the fish was con-nected to the devices to allow for an undisturbed recovery and therecording of baseline parameters. After this initial recovery period,routine baseline cardiovascular variables were recorded for 1 h. Oncea stable baseline was attained, the sulfated form of CCK-8 (see below)was injected through the dorsal aortic catheter. One group (n � 6)received an injection of 0.19 pmol/kg followed by a recovery periodof 4 h before another injection of 0.76 pmol/kg. The control group(n � 6), received two saline injections of corresponding volume(0.3–0.4 ml) separated by a recovery period of 4 h. Cardiovascularparameters were recorded at 2–5 min postinjection. The CCK-8sconcentrations were chosen as to give an estimated final plasmaconcentration of 50 pM [physiological level according to Jönssonet al. (32)] and 200 pM respectively, although the absolute concen-tration would depend on several factors, such as tissue uptake,degradation, and clearance.

In situ perfused heart preparations. The inflow and outflow pres-sures were adjusted to within physiological levels (inflow: 0 � 0.20kPa; outflow: 4.5 � 0.50 kPa), and the heart preparations (n � 8) werethen allowed to stabilize for a period. To confirm the viability of thepreparation, a standard power test was performed on each preparationby a stepwise increase in the outflow pressure (data not shown). Theoutflow pressure was returned, and the preparation was allowed tostabilize again. After a 10-min baseline recording, CCK was added tothe Ringer solution to give a final concentration ranging from 0.1 nMto 0.1 �M.

Vascular preparations. The vascular preparations (n � 8) wereprecontracted with 0.3 �M of adrenaline and allowed to reach a stablebaseline before the protocol commenced (�30 min). CCK was thenadded to the bath to give a final concentration of 0.1 nM. The dosagewas then stepwise increased to 0.1 �M. Finally, the ability of thepreparation to relax was examined using atropine 0.1 mM, which,except for being a muscarinic receptor antagonist, is a very potentvasorelaxant, at least in adrenergically stimulated vessels (33, 35).

Drugs

CK-8 in its sulfated form was obtained from Ansynth Service BY(Roosendaal, The Netherlands). CCK exists in several subforms, outof which CCK-7, CCK-8, and CCK-21 have been found in rainbowtrout. It is related to gastrin, both functionally and structurally (Fig. 1).The biological active site is situated near the C-terminal just by aconserved sulfated tyrosine residue. We used a mammalian CCK 8s,which differ from the trout forms by having methionine in the sixthposition from the fully amidated C-terminal, compared with leucine(CCK-l), asparagine (CCK-n), and threonine (CCK-t) in the troutforms (29). However, it shares the conserved region with all of thetrout forms and has been characterized as reviewed by Johnsen (30).CCK was diluted in saline (0.9%).

Atropine (Atropine sulfate salt, �97%, A-0257) and adrenaline[(-)-epinephrine (�)-bitartrate salt, E4375] were obtained from Sig-ma-Aldrich (St. Louis, MO) and diluted in Ringer solution.

Data Acquisition

The Doppler flow probes were connected to a directional pulsedDoppler flowmeter (model 545C-4; The University of Iowa, Iowa City,IA). For the in situ perfused heart preparations, transit-time flow probes(2N688 flow through probe; Transonic) were connected to a small animalblood flow meter (T206; Transonic). Pressure transducers (MLD-type)were connected to a 4-channel Amplifier (SENSELab, 4CAMP, So-medic, Hörby, Sweden). For the strip preparations, eight force transduc-ers (SENSELab 10-700-0003; Somedic) were connected to two 4-chan-

nel amplifiers (SENSELab, 4CHAMP 700; Somedic). In all cases, aPowerLab system connected to a PC running Chart 6 (ADInstruments,Castle Hill, Australia) was used for the analog/digital conversion and dataacquisition.

Data Analysis and Statistics

Heart rate (HR) was obtained from the phasic pressure traces, andcardiac stroke volume (SV) was calculated as SV � CO/HR. In vivoblood flows (CO and Qcma) are presented as relative changes with thebaseline set to 100%. The routine values were averaged and comparedwith the values after the CCK injection by a two-sample t-test,confirming equal variance. The wall tension of the vascular prepara-tions was analyzed for treatment effects using a repeated-measuresANOVA followed by a Tukey’s post hoc test.

Systolic flow slope (�dQ/dt), diastolic flow slope (-dQ/dt), ampli-tude, as well as time to peak (X) and time to return (Y), were obtainedfrom each flow pulse (Fig. 2) and averaged over a arbitrarily chosenperiod of time for each treatment. The in situ perfused heart prepa-rations were used to analyze the effects on CO, HR, and SV, as wellas the above described parameters, using a repeated-measuresANOVA followed by a Dunnet’s post hoc test. All statistical com-parisons were made on raw untransformed data and to correct formultiple comparisons, the Holm-Bonferroni algorithm was used. Allreported values are expressed as means � SE. * and ** denote asignificant difference from routine as well as control values (P 0.05and P 0.01, respectively). † denotes a significant (P 0.05) changecompared with previous treatment.

RESULTS

In Vivo Recordings

The injection of the gastrointestinal hormone CCK-8s (0.19pmol/kg) produced a significant (P 0.05) increase (20%;18 � 4% compared with routine) in gastrointestinal blood flow(Qcma) compared with a control injection of saline (Fig. 3A).This increase is most likely due to a decrease in the resistanceof the gastrointestinal vasculature and a significant increase inCO (21% compared with control; 30 � 6% compared withroutine; Fig. 3B), mediated through an increase in SV (16%

Fig. 2. Different parameters that were used to characterize the phasic cardiacoutput flow profile of in situ perfused heart preparations (See Data analysisand Statistics for details).

R1242 INFLUENCE OF CCK ON THE CARDIOVASCULAR SYSTEM

AJP-Regul Integr Comp Physiol • VOL 298 • MAY 2010 • www.ajpregu.org

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

compared with control; 32 � 11% compared with routine; Fig.3C). The increase in stroke volume is, however, not significantcompared with the control injection. Saline injections did notproduce any significant changes in cardiac output or strokevolume; the small, yet not significant, increase in cardiacoutput and stroke volume when injecting saline could mostlikely be attributed to a small increase in the circulating bloodvolume (5–6%), as well as a subsequent decrease in viscositydue to a dilution of the blood (Fig. 3, A–D).

When increasing the amount of injected CCK to 0.76 pmol/kg, the response is decreased compared with the lower dose.The increase in Qcma is only 15% compared with the controlinjection (14 � 6% compared with routine). This is most likelydue to a now insignificant increase in CO (5% compared withcontrol; 12 � 2% compared with routine) and SV (6%; 20 �3% compared with routine), thereby removing the CO-medi-ated part of the increase in Qcma. The reason for this could bea dose-dependent CCK-induced increase in the gill resistances(below).

No significant change in HR was seen with any of thetreatments (Fig. 3D).

In Situ Perfused Heart

CCK had no significant effect on cardiac output in the in situperfused heart preparations (Fig. 4A). Stroke volume did notchange, but heart rate increased significantly (11 � 2%) at a

CCK concentration of 10 nM and remained at the same levelwhen increasing the concentration to 100 nM (Fig. 4, B and C).Despite only a minor change in heart rate, there was a consid-erable change in the shape of the phasic flow profile, as can beseen in Fig. 5. The maximum amplitude of each pulse in-creased by 35 � 12% at a CCK concentration of 10 nMcompared with the baseline. The amplitude increased evenfurther (50 � 17%) at 100 nM (Fig. 6). This increase can beexplained by a change in systolic/diastolic slope (�dQ/dt; Fig. 7,A and B) or the time to peak/baseline (X and Y, respectively;Fig. 7, C and D). The systolic slope increased with 39 � 12%and 55 � 15% at 10 nM and 100 nM, respectively. There wasan even more profound change in the relaxation slope, whichincreased with 96 � 26% and 208 � 48%, respectively. Thechanges in the above variables resulted in a decreased time topeak (11 � 6%) and time to return (40 � 6%) at a CCK dosageof 100 nM.

The relative ventricular mass of the perfused hearts was0.095 � 0.004%.

Vascular Ring Preparations

There was no change in the wall tension of the precontracted(0.3 �M adrenaline) vascular ring preparations with either alow (0.1 nM) or a high (100 nM) CCK concentration. Addingatropine (0.1 mM) relaxed the vessels and decreased vesselwall tension by 12.6 � 2% (Fig. 8).

Fig. 3. Means � SE of gastrointestinal blood flow (Qcma) (A) at the coeliacomesenteric artery (CMA), cardiac output (CO) (B), stroke volume (SV) (C) and heartrate (HR) (D) measured in vivo in rainbow trout (Oncorhynchus mykiss). 0.19 pmol/kg and 0.76 pmol/kg sulfated CCK-8 were injected intra-arterially (blackbars) at an interval of 4 h in animals weighing 490 � 58 g (n � 6). Control animals (n � 6, 625 � 48 g) received an intra-arterial injection of saline only (graybars). Dashed line (black) indicates routine pretreatment values for treated animals, whereas dashed line (gray) indicates routine pretreatment values for thecontrol group. A two-sample t-test, assuming equal variance, was used to test for a significant difference between treatments. To correct for multiple comparisonsa Holm-Bonferroni test was used. *Significant difference from both routine and control values for CCK-treated group (P 0.05).

R1243INFLUENCE OF CCK ON THE CARDIOVASCULAR SYSTEM

AJP-Regul Integr Comp Physiol • VOL 298 • MAY 2010 • www.ajpregu.org

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

DISCUSSION

Different nutrients are not equally important in inducing apostprandial hyperemia, and several theories have been pro-posed as to what cause this locally defined hyperemia inmammals (14, 24). One possible theory is that different gas-trointestinal hormones might contribute. CCK is one of thesepotentially important peptides that possibly modulate the gas-trointestinal blood flow and thus could be involved in thepostprandial hyperemia, as well as other cardiovascularchanges (43).

For fish, little is known about where and how the postpran-dial hyperemia is regulated and the importance of endocrinesubstances such as CCK. The observed, CCK-stimulated in-crease in gastrointestinal blood flow, is in concordance withmammalian studies, in which CCK have either direct (15, 27)or indirect neurogenic effects [(7, 43); reviewed by Ruiz-Gayoet al. (42)] on the gastrointestinal vessels. The effect of CCKcan also be secondary to increases in, for example, pancreaticsecretions and thus pancreatic blood flow, leading to overallincrease in gastrointestinal blood flow (38).

Furthermore, results from the isolated vascular preparationsindicate that there are, at least in rainbow trout, no directeffects of physiological levels of CCK on isolated vessels fromsecond- to third-order branches of the coeliacomesenteric ar-tery. Nevertheless, disparity exists, and many studies indicatethat the level needed to induce a gastrointestinal hyperemia ismuch higher than the circulating plasma levels recorded in thepostprandial state (23, 41). It is, thus possible that the hormonework in a more paracrine mode, assuming that higher CCKlevels can be produced locally in the intestine. Our results,however, challenge this, given that there was little effect ofhigh CCK concentrations. It could also be that the directvascular effect of CCK is limited to the smaller-diametersubmucosal/mucosal vessels that are not possible to use in thepresent experimental setup. This is less likely consideringthat, at least in mammals, for the most part, the control ofthe gastrointestinal vascular resistance lies within the majorvessels before the beginning of the regional intestinal vas-culature (9).

The increase in gastrointestinal blood flow initiated bycirculating CCK seen in this study is considerably lower thanpostfeeding values presented for rainbow trout in previousstudies, showing an increase of 100% and above (17, 26). It is,however, unlikely that CCK would account for the entireincrease in gastrointestinal blood flow seen after feeding, and itis probably more important in adapting the cardiovascularresponse to the nutrient composition of the diet. The dynamicsof the CCK release and the subsequent circulating levels arealso unknown, and it could very well be that there is acontinuous release of CCK during postprandial conditions.Therefore, the levels achieved in this study could be transient,ultimately depending on the rate of turnover or breakdown.Furthermore, by injecting the dose into the systemic circula-tion, it could also be that we see a generalized vasodilation notlimited to the gastrointestinal tract and, consequently, therewould be no redistribution of blood from the systemic to thegastrointestinal circulation. During normal postfeeding condi-tions, it may well be that most of the CCK is confined to theintestine and thus only limited amounts reach the systemiccirculation.

The observed increase in gastrointestinal blood flow in thepresent study is most likely due to a decrease in the vascularresistance of the gastrointestinal tract. However, we did notsimultaneously record arterial and venous pressure; thus, anycalculation of vascular resistances is not possible. The reasonsfor this are numerous. First of all, an additional venous catheterwould have imposed further stress on the animal and thus couldhave hidden the observed cardiovascular changes. Second, itwould have proven hard to get accurate measurements of dorsalaortic pressure so soon after the intra-arterial injection of CCK,and artifacts from the injection would have been a major concern.

Fig. 4. Means � SE of CO (A), HR (B), and SV (C) measured in the in situperfused hearts (n � 8; 422 � 14 g with a mean relative ventricular mass of0.095 � 0.004%) of rainbow trout (O. mykiss). CCK was added to theperfusate at increments of 10 from 0.1 nM to 100 nM, and samples were testedfor a significant difference from routine using a repeated-measures ANOVAfollowed by a Dunnet’s post hoc test. Dashed line indicates routine pretreat-ment values. *Significant difference from routine values (P 0.05).

R1244 INFLUENCE OF CCK ON THE CARDIOVASCULAR SYSTEM

AJP-Regul Integr Comp Physiol • VOL 298 • MAY 2010 • www.ajpregu.org

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

To maintain the increase in gastrointestinal blood flow, there wasa subsequent increase in cardiac output. The increase was of sucha magnitude that it most likely could account for the entireincrease in gastrointestinal blood flow, without an additionalredistribution of blood. The increase in cardiac output was medi-ated via an increase in stroke volume since heart rate did notchange. However, the in vivo increase in cardiac output was notseen in the isolated perfused heart preparations, and thus, it isprobably not a direct effect of CCK on the heart itself. The factthat neither the isolated vessels nor the heart preparations showedany change strengthens the notion that CCK could work througha neurogenic mechanism. This, however, remains to be estab-lished. The increase in cardiac output could also have been apassive effect, resulting from the decrease in vascular resistance,although this is unlikely, considering the intrinsic property of theheart to maintain the stroke volume over a wide range of aorticpressures [homeometric regulation; (18)].

Even though cardiac output did not change, interestingchanges were seen upon CCK exposure in the perfused heart.The most obvious effect was the increase in the amplitude ofcardiac flow profile. Amplitude increased as a consequence of

an increase in systolic and diastolic rate (dQ/dt) with only aminor increase in heart rate. This increase of the systolic anddiastolic rate of the phasic flow profile could be caused by achange in the wind-kessel effect of the bulbus arteriosus (19).The bulbus probably enables the very large stroke volumesseen in many fishes (37), as well as having the capability toadjust to changes in blood pressure (31). The bulbus receivessympathetic innervation, and Farrell et al. (21) indicated that abulbus-mediated regulation of the pulsatility and hence, the bloodflow patterns through the gills could be important in governing itssynchrony with the ventilatory movements in fish. A stiffening ofthe bulbus would explain the change in flow profile seen, andCCK could possibly act through a NO-mediated mechanism,given the wide repertoire of functions of NO in the heart (49). Thedosage at which CCK has its effect is high compared withcirculating levels, but it could, nevertheless, be attained at definedlocations, such as at synapses or in the vicinity of neuroendocrinecells.

Nonetheless, when increasing the injected dose of CCK,there was still a significant increase in gastrointestinal bloodflow, but the response was reduced compared with the lowerphysiological dose. There are several possible explanationsfor this. It could be that the maximal dose needed hadalready been reached. This is, however, unlikely consideringthat CCK in local areas of the gastrointestinal tract probablyreach much higher levels as has been speculated in severalmammalian studies (41). Another much more plausibleexplanation is that there is a dose-dependent increase in thevascular resistance of the gill circulation, as previouslyreported by Sundin and Nilsson (48). An increase in theresistance of the gill vessels would increase the workload ofthe heart and thus lead to a decrease in stroke volume (whenoutside of the homeometric regulatory scope) through aStarling mechanism (20), causing the observed decrease incardiac output. This is also consistent with in vivo record-ings of CO and ventral aortic pressure (Pva) in cod (48),where CO decreases with an increase in Pva during increas-ing levels of CCK exposure. Ultimately, this means thatwhen cardiac output is decreased, the increase in gastroin-testinal blood flow cannot be maintained and gastrointesti-nal blood flow decreases, as seen here. Whether or not thisholds true remains to be determined.

Fig. 5. Representative trace illustrating the change in the cardiac output flow profile associated with an application of CCK at levels of 10 nM and above.

Fig. 6. Means � SE in the amplitude of the cardiac output flow profile (Fig. 2)measured in the in situ perfused hearts (n � 8; 422 � 14 g with a mean relativeventricular mass of 0.095 � 0.004%) of rainbow trout (O. mykiss). Sampleswere tested for a significant difference from routine using a repeated-measureANOVA followed by a Dunnet’s post hoc test. Dashed line indicates routinepretreatment values. **Significant difference from routine values (P 0.01).

R1245INFLUENCE OF CCK ON THE CARDIOVASCULAR SYSTEM

AJP-Regul Integr Comp Physiol • VOL 298 • MAY 2010 • www.ajpregu.org

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

Conclusions and Future Perspectives

In conclusion, CCK seems to affect the gastrointestinalblood flow and the dynamics of the phasic flow profile in theventral aorta. CCK could be important in regulating the car-diovascular postprandial response to different diets, but CCKdoes not have a direct effect on the vasculature but insteadprobably acts on peripheral nerves relaying the signal to its

target. Furthermore, one rather peculiar effect on the heart wasdiscovered. This CCK-induced change in flow pattern could beimportant in modulating the flow through the gills duringchanging conditions, such as, for example, after feeding orduring stress, such as hypoxia. To undoubtedly establish therole of CCK in the postprandial regulation of gastrointestinalblood flow, more studies are, however, needed. First and most

Fig. 7. Means � SE of systolic slope (�dQ/dt) (A), diastolic slope (-dQ/dT) (B), X (time to peak) (C) and Y (time to routine) (D) of the cardiac output flow profile(Fig. 2) measured in the in situ perfused hearts (n � 8; 422 � 14 g with a mean relative ventricular mass of 0.095 � 0.004%) of rainbow trout (O. mykiss).Samples were tested for a significant difference from routine values using a repeated-measure ANOVA followed by a Dunnet’s post hoc test. Dashed lineindicates routine pretreatment values. * and ** denotes a Significant difference from routine values (P 0.05) and (P 0.01), respectively.

Fig. 8. Representative recording of a vascularpreparation from the coeliacomensenteric ar-tery and its branches, precontracted with 0.3�M adrenaline and allowed to normalize be-fore application of CCK in stepwise incre-ments from 0.1 nM to 100 nM. To verify thecapacity of the preparation to relax, atropine(0.1 mM) was added at the end of the exper-iment. The small graph in the upper right cor-ner shows the means � SE for the change incontractile force of n � 32 vascular prepara-tions obtained from n � 8 (465 � 14 g) rain-bow trout (O. mykiss). Samples were tested fora significant difference from routine and pre-contracted values using a repeated-measuresANOVA followed by a Tukey’s post hoc test.Dashed line indicates routine pretreatment val-ues and † denotes a significant difference fromthe precontracted values (P 0.05) for theatropine treatment.

R1246 INFLUENCE OF CCK ON THE CARDIOVASCULAR SYSTEM

AJP-Regul Integr Comp Physiol • VOL 298 • MAY 2010 • www.ajpregu.org

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

importantly, one would have to resolve the dynamics of theCCK in the plasma. Given the huge lack of knowledge con-cerning such aspects in fish, this holds true for most otherpeptides or pharmaceuticals as well. This could be achieved athigh resolution using a microdialysis approach, enabling notonly the exact determination of postprandial levels of CCK, butalso the dynamics of the injected molecule. This would makea setup where CCK (and other peptides) can be infused, so asto give physiological plasma levels on a long-term basispossible. Furthermore, the modulatory capacity of CCK on theheart or the bulbus arteriosus warrants future investigations.

ACKNOWLEDGMENTS

The authors wish to thank Susanne Salmela for her helpduring the practical part of this manuscript and scientists fromthe Fish Endocrinology Laboratory at the University of Goth-enburg for helpful advice regarding the CCK 8s-peptide.

GRANTS

Funding was received from Swedish Research Council (M. Axelsson), andWilhelm and Martina Lundgren’s foundation (H. Seth).

DISCLOSURES

No conflicts of interest are declared by the authors.

REFERENCES

1. Aldman G, Grove D, Holmgren S. Duodenal acidification and intra-arterial injection of CCK8 increase gallbladder motility in the rainbowtrout, Oncorhynchus mykiss. Gen Comp Endocrinol 86: 20–25, 1992.

2. Aldman G, Holmgren S. Intraduodenal fat and amino acids activategallbladder motility in the rainbow trout, Oncorhynchus mykiss. GenComp Endocrinol 100: 27–32, 1995.

3. Altimiras J, Claireaux G, Sandblom E, Farrell AP, McKenzie DJ,Axelsson M. Gastrointestinal blood flow and postprandial metabolism inswimming sea bass Dicentrarchus labrax. Physiol Biochem Zool 81:663–672, 2008.

4. Axelsson M, Fritsche R. Cannulation techniques. In: Biochemistry andMolecular Biology of Fishes, edited by Hochachka PW and MommensenTP. New York: Elsevier Science, 1994, p. 17–38.

5. Axelsson M, Fritsche R. Effects of exercise, hypoxia and feeding on thegastrointestinal blood flow in the Atlantic cod Gadus morhua. J Exp Biol158: 181–198, 1991.

6. Axelsson M, Thorarensen H, Nilsson S, Farrell AP. Gastrointestinalblood flow in the red Irish lord, Hemilepidotus hemilepidotus: long-termeffects of feeding and adrenergic control. J Comp Physiol [B] 170:145–152, 2000.

7. Biber B, Fara J, Lundgren O. A pharmacological study of intestinalvasodilator mechanisms in the cat. Acta Physiol Scand 90: 673–683, 1974.

8. Bohlen HG. Integration of intestinal structure, function, and microvascu-lar regulation. Microcirculation 5: 27–37, 1998.

9. Bohlen HG, Henrich H, Gore RW, Johnson PC. Intestinal muscle andmucosal blood flow during direct sympathetic stimulation. Am J PhysiolHeart Circ Physiol 235: H40–H45, 1978.

10. Brown CR, Cameron JN. The induction of specific dynamic action inchannel catfish by infusion of essential amino acids. Physiol Zool 64:276–297, 1991.

11. Brown CR, Cameron JN. The relationship between specific dynamicaction (SDA) and protein synthesis rates in the channel catfish. PhysiolZool 64: 298–309, 1991.

12. Burton-Freeman B, Gietzen DW, Schneeman BO. Cholecystokinin andserotonin receptors in the regulation of fat-induced satiety in rats. Am JPhysiol Regul Integr Comp Physiol 276: R429–R434, 1999.

13. Chandra R, Liddle RA. Cholecystokinin. Curr Opin Endocrinol Diabe-tes Obes 14: 63–67, 2007.

14. Chou CC, Coatney RW. Nutrient-induced changes in intestinal bloodflow in the dog. Br Vet J 150: 423–437, 1994.

15. Chou CC, Hsieh CP, Dabney JM. Comparison of vascular effects ofgastrointestinal hormones on various organs. Am J Physiol Heart CircPhysiol 232: H103–H109, 1977.

16. Chou CC, Kvietys P, Post J, Sit SP. Constituents of chyme responsiblefor postprandial intestinal hyperemia. Am J Physiol Heart Circ Physiol235: H677–H682, 1978.

17. Eliason EJ, Higgs DA, Farrell AP. Postprandial gastrointestinal bloodflow, oxygen consumption and heart rate in rainbow trout (Oncorhynchusmykiss). Comp Biochem Physiol 149: 380–388, 2008.

18. Farrell AP. A review of cardiac performance in the teleost heart: Intrinsicand humoral regulation. Can J Zool 62: 523–536, 1984.

19. Farrell AP. The Wind-kessel effect of the bulbus arteriousus in trout. JExp Zool 209: 169–173, 1979.

20. Farrell AP, Jones DR. The heart. In: Fish Physiology, The Cardiovas-cular System. San Diego, CA: Academic, 1992, p. 1–88.

21. Farrell AP, Sobin SS, Randall DJ, Crosby S. Intralamellar blood flowpatterns in fish gills. Am J Physiol Regul Integr Comp Physiol 239:R428–R436, 1980.

22. Franklin C, Axelsson M. The intrinsic properties of an in situ perfusedcrocodile heart. J Exp Biol 186: 269–288, 1994.

23. Gallavan RH Jr, Chen MH, Joffe SN, Jacobson ED. Vasoactiveintestinal polypeptide, cholecystokinin, glucagon, and bile-oleate-inducedjejunal hyperemia. Am J Physiol Gastrointest Liver Physiol 248: G208–G215, 1985.

24. Gallavan RH Jr, Chou CC. Possible mechanisms for the initiation andmaintenance of postprandial intestinal hyperemia. Am J Physiol Gastroi-ntest Liver Physiol 249: G301–G308, 1985.

25. Granger DN, Kvietys PR. The splanchnic circulation: intrinsic regula-tion. Annu Rev Physiol 43: 409–418, 1981.

26. Grans A, Albertsson F, Axelsson M, Olsson C. Postprandial changes inenteric electrical activity and gut blood flow in rainbow trout (Oncorhyn-chus mykiss) acclimated to different temperatures. J Exp Biol 212: 2550–2557, 2009.

27. Guth PH, Smith E. The effect of gastrointestinal hormones on the gastricmicrocirculation. Gastroenterology 71: 435–438, 1976.

28. Holmgren S. Neuropeptide functions in the fish gut. Peptides 6 Suppl 3:363–368, 1985.

29. Jensen H, Rourke IJ, Moller M, Jonson L, Johnsen AH. Identificationand distribution of CCK-related peptides and mRNAs in the rainbow trout,Oncorhynchus mykiss. Biochim Biophys Acta 1517: 190–201, 2001.

30. Johnsen AH. Phylogeny of the cholecystokinin/gastrin family. FrontNeuroendocrinol 19: 73–99, 1998.

31. Jones DR, Perbhoo K, Braun MH. Necrophysiological determination ofblood pressure in fishes. Naturwissenschaften 92: 582–585, 2005.

32. Jönsson E, Forsman A, Einarsdottir IE, Egner B, Ruohonen K,Bjornsson BT. Circulating levels of cholecystokinin and gastrin-releasingpeptide in rainbow trout fed different diets. Gen Comp Endocrinol 148:187–194, 2006.

33. Kwan CY, Zhang WB, Kwan TK, Sakai Y. In vitro relaxation ofvascular smooth muscle by atropine: involvement of K� channels andendothelium. Naunyn-Schmiedeberg’s Arch Pharmacol 368: 1–9, 2003.

34. Kvietys PR, Granger DN. Regulation of colonic blood flow. Fed Proc41: 2106–2110, 1982.

35. Liu SQ, Zang WJ, Li ZL, Yu XJ, Li BP. Effect of atropine ondenervated rabbit ear blood vessels. J Cardiovasc Pharmacol 43: 99–105,2004.

36. Matheson PJ, Wilson MA, Garrison RN. Regulation of intestinal bloodflow. J Surg Res 93: 182–196, 2000.

37. Mendonca PC, Genge AG, Deitch EJ, Gamperl AK. Mechanismsresponsible for the enhanced pumping capacity of the in situ winterflounder heart (Pseudopleuronectes americanus). Am J Physiol RegulIntegr Comp Physiol 293: R2112–R2119, 2007.

38. Nakajima M, Naruse S, Kitagawa M, Ishiguro H, Jin C, Ito O,Hayakawa T. Role of cholecystokinin in the intestinal phase of pancreaticcirculation in dogs. Am J Physiol Gastrointest Liver Physiol 280: G614–G620, 2001.

39. Olsson C, Aldman G, Larsson A, Holmgren S. Cholecystokinin affectsgastric emptying and stomach motility in the rainbow trout Oncorhynchusmykiss. J Exp Biol 202: 161–170, 1999.

40. Olsson C, Holmgren S. The control of gut motility. Comp BiochemPhysiol 128: 481–503, 2001.

41. Premen AJ, Kvietys PR, Granger DN. Postprandial regulation of intes-tinal blood flow: role of gastrointestinal hormones. Am J Physiol Gastroi-ntest Liver Physiol 249: G250–G255, 1985.

R1247INFLUENCE OF CCK ON THE CARDIOVASCULAR SYSTEM

AJP-Regul Integr Comp Physiol • VOL 298 • MAY 2010 • www.ajpregu.org

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from

42. Ruiz-Gayo M, Gonzalez MC, Fernandez-Alfonso S. Vasodilatory ef-fects of cholecystokinin: new role for an old peptide? Regul Pept 137:179–184, 2006.

43. Sanchez-Fernandez C, Gonzalez MC, Beart PM, Mercer LD, Ruiz-GayoM, Fernandez-Alfonso MS. A novel role for cholecystokinin: regulation ofmesenteric vascular resistance. Regul Pept 121: 145–153, 2004.

44. Seth H, Axelsson M. Effects of gastric distension and feeding on cardio-vascular variables in the shorthorn sculpin (Myoxocephalus scorpius). AmJ Physiol Regul Integr Comp Physiol 296: R171–R177, 2009.

45. Seth H, Sandblom E, Axelsson M. Nutrient-induced gastrointestinalhyperemia and specific dynamic action in rainbow trout (Oncorhynchusmykiss)—importance of proteins and lipids. Am J Physiol Regul IntegrComp Physiol 296: R345–R352, 2009.

46. Seth H, Sandblom E, Holmgren S, Axelsson M. Effects of gastricdistension on the cardiovascular system in rainbow trout (Oncorhynchusmykiss). Am J Physiol Regul Integr Comp Physiol 294: R1648–R1656,2008.

47. Shahbazi F, Conlon JM, Holmgren S, Jensen J. Effects of codbradykinin and it’s analogs on vascular and intestinal smooth mus-cle of the Atlantic cod, Gadus morhua. Peptides 22: 1023–1029,2001.

48. Sundin L, Nilsson S. Arterio-venous branchial blood flow in the AtlanticCod Gadus morhua. J Exp Biol 165: 73–84, 1992.

49. Tota B, Amelio D, Pellegrino D, Ip YK, Cerra MC. NO modulation ofmyocardial performance in fish hearts. Comp Biochem Physiol 142:164–177, 2005.

R1248 INFLUENCE OF CCK ON THE CARDIOVASCULAR SYSTEM

AJP-Regul Integr Comp Physiol • VOL 298 • MAY 2010 • www.ajpregu.org

on April 29, 2010

ajpregu.physiology.orgD

ownloaded from