Comparative Biochemistry and Physiology Part A 134(2003) 471–480

1095-6433/03/$ - see front matter� 2002 Elsevier Science Inc. All rights reserved.PII: S1095-6433Ž02.00324-0

Control of cardiovascular function in the icefishChionodracohamatus: involvement of serotonin and nitric oxide

D. Pellegrino , R. Acierno , B. Tota *a b a,c,

Department of Cell Biology, University of Calabria, Arcavacata di Rende 87030 (CS), Italya

Department of Biology, University of Lecce, Prov.le Monteroni, 73100 Lecce, Italyb

Zoological Station ‘A. Dohrn’, Villa Comunale, I-80121 Naples, Italyc

Received 17 July 2002; received in revised form 4 November 2002; accepted 5 November 2002

Abstract

The involvement of nitric oxide(NO) in the branchial circulation and cardiac performance of the Antarctichemoglobinless icefishChionodraco hamatus was investigated using isolated and perfused head and working heartpreparations. In the branchial vasculature under basal(i.e. unstimulated conditions), the nitric oxide synthase(NOS)inhibitor L-NIO (L-N -(1-iminoethyl) ornithine, 10 and 10 M), caused a marked vasoconstriction(20%), indicating5 y5 y4

a basal nitrergic vasodilator tone, while the dose–response curve of the NO donor SIN-1(3-morpholinosydnonimine)showed a dose-dependent vasodilator effect. Acetylcholine induced a dose-dependent branchial vasoconstriction mediatedby muscarinic receptors, since the effects were abolished by pre-treatment with atropine(10 M). Serotonin(5-HT)y4

induced a dose-dependent branchial methysergide-sensitive vasoconstriction which was abolished by pre-treatment withL-NIO, indicating a NO-dependent mechanism. On the isolated heart, the NOS inhibitorL-NMMA (N -monomethyl-L-G

arginine) 10 M produced a small, but significant decrease of heart rate and, as a consequence, a decrease of powery4

output, while the NO donor sodium nitroprusside(SNP) 10 M elicited increases of stroke volume, stroke work andy4

power output, suggesting an exogenous NO-dependent positive inotropism. Exposure of the bulbus arteriosus toL-NMMA was without any appreciable effect. In contrast, SNP produced a notable relaxation of the bulbus wall with amarked increase of its stiffness, as indicated by the increase of systolic and diastolic dPydt max (23 and 104%,respectively).� 2002 Elsevier Science Inc. All rights reserved.

Keywords: Antarctic fish; Acetylcholine; Serotonin; Gill hemodynamics; Heart; Nitric oxide; Notothenioidei

1. Introduction

The Antarctic perciform suborder Notothenioid-ei is the dominant fish group, in terms of bothspecies numbers and abundance, inhabiting thecoldest and most thermally stable waters in theSouthern Ocean where they have evolved in iso-lation in the last 25–40 My. They include theChannichthyidae family which are unique among

*Corresponding author. Tel.yfax: q39-984-492905.E-mail address: tota@unical.it(B. Tota).

all vertebrates, as adults, because they lack hemo-globin (Hb) (Ruud, 1954) and, in some specieslike Chaenocephalus aceratus, also cardiac myo-globin (Moylan and Sidell, 2000). The hemoglo-binless condition, associated with a colorless bloodwith few nonfunctional erythrocytelike cells, givesthese fishes a peculiar pale color for which theyare named icefish. Because icefish lack Hb, oxygenis carried exclusively in physical solution andtherefore the oxygen-carrying capacity of the ice-fish blood, despite the gain in gas solubility at lowtemperature, is only one-tenth of that of red-

472 D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

blooded teleosts(Ruud, 1954). However, the rateof oxygen consumption is only modestly lowerthan that of other teleosts(Hemmingsen et al.,1972; Hemmingsen, 1991), due to adaptationalrearrangements of the cardio-circulatory systemthat make the icefish unusual among other teleosts(Tota et al., 1991a,b; Hemmingsen, 1991; Tota etal., 1998). These cardio-circulatory traits include:increase in blood volume, from 2 to 4 times largerthan that in many teleosts(C. aceratus: Hemming-sen and Douglas, 1970;Chionodraco hamatus:Acierno et al., 1995); large cardiac output(CO)attained with large stroke volume(SV) (6–15times greater than in other teleosts) at low heartrate (HR) and at relatively low ventral aortapressure(Hemmingsen et al., 1972), typical of avery specialized volume pump heart ventricle(Totaand Gattuso, 1996); impressive cardiac enlarge-ment resulting in a relative heart weight similar tothat of small mammals(Johnston et al., 1983);enlarged bore of both the capillaries and mainvessels(arteries and veins), consistent with a highvolume-high flowylow pressure vascular system(Egginton et al., 2002, and references therein). Inline with the uncommonly low pressure drops fromthe ventral aorta to the dorsal aorta(Hemmingsen,1991), the gill vasculature also shows larger diam-eters and thinner walls of the main branchialvessels as well as exceptionally large lacunae andmarginal channels of the lamellae(Davison et al.,1997). The bulbus arteriosus is much larger thanthe general size of teleostean bulbus and its func-tional morphology appears well suited to accom-modate the very high SV of the icefish(Icardo etal., 1999). To provide a quite constant branchialflow, despite the very high HR, a remarkablestretching capability and compliance of the bulbuswall is required.

Due to the impaired total oxygen content of theblood, the neuro-humoral control of cardiac func-tion and branchial perfusion must play a crucialrole in icefish respiratory homeostasis, especiallyin the case of burst swimming activity. However,how this is achieved in the icefish is unknown.The few studies on the cardio-vascular control ofAntarctic fish have been conducted on red-bloodedspecies(Davison et al., 1997; Sundin et al., 1998;Egginton et al., 2001, and references therein). Intwo red-blooded species,Pagothenia borchgrev-inki andTrematomus bernacchii, the latter being abenthic ecotype very similar toC. hamatus, thevery low resting HR is due to an unusually high

inhibitory vagal tone of 55 and 80%, respectively(Axelsson et al., 1992) and their branchial vascu-lature also appears under cholinergic(i.e. vasocon-striction) tone, although alpha- (i.e.vasoconstriction) and beta-(i.e. vasodilation) adre-nergic influences have been detected(Axelsson etal., 1994; Foster et al., 1998). Both neuro-humoralinfluences appeared less potent than the serotoni-nergic-induced branchial vasoconstriction(Fosteret al., 1998; Sundin et al., 1998), which is 1.000-fold higher that found in trout(Sundin et al.,1995). The importance of the serotoninergicvasomotion, confirmed in few other bottom-dwell-ing and sluggish red-blooded species(Egginton etal., 2001), has been corroborated by the detectionof 5-HT storage sites in the gills ofP. borchgrev-inki (Sundin et al., 1998). Although more speciesneed to be analysed before making any generali-zation, these studies stress at the same time therole of the nonadrenergic, noncholinergic controlof the branchial circulation in the Antarctic teleostsand the importance of uncovering the mechanismsunderlying their low peripheral vascular resistance.This points obviously to nitric oxide(NO), whoserole as the major modulator of both vascularresistance and cardiac function has been firmlyestablished in mammals(Moncada and Higgs,1995). Surprisingly, the role of NO has been poorlyanalysed in fish. The few studies made in teleostshave led to contradictory results, depending eitheron different organ–tissue preparations used orpossible species-specificities(Olson and Villa,1991; McGeer and Eddy, 1996; Mustafa et al.,1997; Imbrogno et al., 2001; Pellegrino et al.,2002, and references therein). For example, someauthors presented evidence against the presence ofa NO system in the teleost systemic circulation(Olson and Villa, 1991), whereas others suggestedthat NO-releasing compounds cause vasodilatationin rainbow trout in vivo by activation of solubleguanylate cyclase(McGeer and Eddy, 1996). Dueto their intriguing diversification and the unique-ness of their physiological adaptations, icefishoffer an exceptional opportunity to explore aspectsof unity and diversity in the neuro-humoral controlof the teleost cardiovascular system. Since noinformation is available on the functions of NO inicefish, we have investigated inC. hamatus theinfluence of NO on the branchial vasculature andheart performance under basal(i.e. unstimulated)conditions, using both exogenous NO-donors andinhibitors. In addition, we have characterized for

473D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

the first time in an icefish the branchial vasocon-strictory effects of acetylcholine(ACh) and 5-HT,demonstrating that the serotoninergic influenceinvolves a NO-mediated mechanism.

2. Materials and methods

2.1. Animals

The study was conducted on 30 specimens ofthe icefish C. hamatus of both sexes(weighing438.45"43.7 g; mean value"S.E.M.). Fish werecaught by nets in the Terranova Bay, Ross Sea,Antarctica, during February 1999. Identificationwas based on descriptions of Fisher and Hureau(1985). After capture, the animals were kept notfed for at least 10 days in aerated, running seawater at temperatures between 0 and 28C. Onlyfish that appeared to be healthy were used in theexperiments. All the experiments were done in theItalian Antarctic Base laboratories, Terranova Bay(748429S, 1648069E), during the XIV Italian Ant-arctic Expedition(January–February 1999).

2.2. Saline

The perfusate was a modified version of theCortland saline(Wolf, 1963), with an increasedNaCl content to bring the ionic concentration upto the level found in channichthyd blood(Holeton,1970). Its composition(in millimoles per liter)was: NaCl 252.4. KCl 5.0, MgSOØ7H O 2.0,4 2

Glucose 5.56, CaCl 2.3, NaH POØH O 0.2,2 2 4 2

Na HPOØ2H O 2.3. The saline was aerated and2 4 2

the pH was adjusted to 7.8 at 18C, (i.e. thephysiological pH in icefish) by using NaHCO3(Tota et al., 1991a). The effects of the differentsubstances were assessed by adding them to theperfusion saline. Chemicals were obtained fromSigma-Aldrich.

2.3. Isolated preparations

2.3.1. Isolated perfused headThe animals(ns24) were heparinized with a

caudal veinyartery injection of 0.7 mlykg of sodi-um heparine in saline(50 UIyml), anaesthetizedwith ethyl p-amino-benzoate(benzocaine, 0.2 gyl), killed by spinectomy and pithed. The animalwas opened ventrally behind the pectoral fins; theventral aorta was cannulated and the gills werewashed immediately with heparinized saline(50

UIyml). The head was isolated by cutting behindthe opercular openings and connected with theperfusion apparatus(modified from that describedby Perry et al., 1982) and the perfusion wasimmediately started from a saline reservoir. Prep-aration time was less than 5 min. The isolatedhead was perfused at constant pulsatile flow usinga perystaltic pump and a compliant system at thephysiological value of 100 mlyminykg. All theexperiments were performed in a cold thermostat-ted cabinet (LKB 2021 Maxicoldlab) whichallowed the perfusion system to be maintained at0 8C.

2.3.2. Isolated perfused working heartThe isolated working heart preparation was

achieved according to the protocol described byTota et al. (1991a), using six specimens ofC.hamatus. The animals were anaesthetized in ben-zocaine(0.2 gyl) for 15 min. The ventral aortaand the sinus venosus were cannulated by polyeth-ylene cannulae and the heart was connected witha perfusion apparatus. Perfusion was immediatelystarted, the heart received saline from an inputreservoir and pumped against an afterload pressuregiven by the height of an output reservoir. Isolationtime was 5 min. All the experiments were per-formed in a cold thermostatted cabinet(LKB 2021Maxicoldlab), which allowed the heart and perfu-sion system to be maintained at near zero temper-ature. Each heart generated its own rhythm.Cardiac parameters were simultaneously measuredduring experiments. The fish were weighed beforeeach experiment and the blotted wet weights ofwhole hearts(atrium, ventricle and bulbus) andventricles were determined at the end.

2.4. Experimental protocols

2.4.1. Isolated headExperiments started 20 min after stabilization of

the ventral aortic pressure, which usually occurred20–30 min from the beginning of perfusion. Unsta-ble preparations, usually showing progressiveincreases in resistance, were discarded.

After baseline stabilization, each preparationwas subsequently exposed to ACh, 5-HT or SIN-1, in increasing concentrations. In the case of pre-treatment with the nitric oxide synthase(NOS)inhibitor, L-NIO, atropine or pirenzepine, the prep-arations were exposed to these agents for 20 minbefore the onset of any further test.

474 D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

Fig. 1. Effect of ACh perfusion on input pressure(Pi) in the isolated icefish head preparation; the inlet panel shows the effect of ACh10 M before and after atropine(10 M) treatment. Results are expressed as percent changes(D%) from control values; data arey5 y4

means"S.E. of 3(AtropineqACh) and 4(ACh alone) experiments(*P-0.05).

2.4.2. Isolated working heartHearts that did not stabilize within 15 min were

discarded. After stabilizing cardiac parameters ata physiological flow value of 100 mlyminykg, theheart was subsequently perfused with the NOSinhibitor L-NIO and the NO donor sodium nitro-prusside(SNP), both at a concentration 10 My4

in saline.

2.5. Measurements and calculations

Ventral aortic perfusion pressure was measuredvia a MP20D pressure transducer(Micron instru-ments, Simi Valley, CA) connected through a T-tube placed immediately before the input cannula,and its output displayed on a pen recorder(Uni-record 7050, Ugo Basile, Comerio, Italy). Theisolated heart preload and afterload were measuredwith a Bentley Trantec 800 transducer connectedto a Gemini 7070 2-channels recorder(Ugo Basi-le) and were referred to the saline level in theperfusion chamber. HR was calculated from thepressure waves recording. The CO was measuredby using a Gould-Statham SP2202 flowmeter andsimultaneously recorded with the bulbus pressureon the Gemini recorder. Pressure measurementswere expressed as mean"S.E. of the absolutevalues in kPa or as percent changes from individualexperiments. dPydt was calculated from therecorded pressure curves. Statistical significanceof differences was evaluated on absolute valuesusing paired Student’st-test.

3. Results

3.1. Isolated head preparation

Under basal conditions the preparation showedthe high flowylow pressure circulatory patterntypical of the icefish, as detailed below. Thefunctional characteristics of the isolated head prep-aration were assessed testing ACh and 5-HT dose–response curves. Fig. 1 shows the characteristicACh vasoconstrictory effect on teleost gill vascu-lature (Pellegrino et al., 2002, and referencestherein). The cholinergic response is abolished byatropine, indicating a mechanism dependent on theactivation of muscarinic receptors. Infusion of theicefish head preparation with saline containingvarious concentrations of 5-HT elicited a dose-dependent vasoconstriction(Fig. 2). The 5-HTvasoconstrictory effect was significant at higherconcentrations(10 and 10 M) and was inhib-y5 y4

ited by the 5-HT receptor antagonist methysergide10 M (Fig. 2).y5

The NOS inhibitorL-NIO (10 and 10 M)y5 y4

alone exerted a vasoconstrictory effect(20%) (Fig.3), while the NO donor SIN-1(from 10 toy12

10 M) induced a remarkable decrease of vas-y5

cular resistance in the gills(Fig. 4). These dataare consistent with a vasodilatory nitrergic tone inthe branchial vasculature of the icefish. The 5-HTvasoconstrictory response was completely abol-ished byL-NIO (10 M), demonstrating a NO-y5

dependent mechanism(Fig. 5).

475D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

Fig. 3. Effect ofL-NIO perfusion on input pressure(Pi) in theisolated icefish head preparation. Results are expressed as per-cent changes(D%) from control values; data are means"S.E.of 3 experiments(*P-0.05).

Fig. 2. Effect of 5-HT perfusion on input pressure(Pi) in the isolated icefish head preparation before and after methysergide(10y5

M) treatment. Results are expressed as percent changes(D%) from control values; data are means"S.E. of 3(5-HTqmethysergide)and 4(5-HT alone) experiments(*P-0.05).

3.2. Isolated perfused heart preparation

The performance of the isolated working heartof the icefish showed remarkably high CO at lowHR, confirming previous results in vitro onC.hamatus by Tota et al.(1991a).

Exposure of the cardiac preparation to the NOSinhibitor L-NMMA (10 M) induced a negativey4

chronotropism, reducing HR by 12.6%(Fig. 6).Since there were no effects on SV and SW, thedecrease in PO can be attributed to the decreasein HR. However, in two heart preparations exposedto another inhibitor,L-NIO, a mild negative ino-tropism(i.e. SVsy7.3%, SWsy9.9%P symax

7.8%) was detected. More experiments are neededto validate this preliminary result. The NO donorSNP (10 M) caused increases of SV, SW andy5

PO (Fig. 6), which may be consistent with apositive inotropic influence of exogenous NO. Ofnote are the effects elicited by the drugs on thebulbus arteriosus. WhileL-NMMA 10 M didy4

not show any appreciable effect on bulbar pressure,the NO donor SNP 10 M produced increases ofy4

dPydt max from 8.5"1.9 to 12.3"3.1 cm H O2

and dPydt min from y8.7"1.9 to y12.3"4.2cm H O. The modification of the bulbar pressure2

curve under the actions of the two substances,reported in Fig. 7, reveals increases of both bulbusstiffness, indicated by the steeper pressure curve,and P due to the positive inotropic effect onmax

the ventricle. SNP exposure caused the completerelaxation of the bulbar wall, evident at visual

inspection and documented by the bulbar pressurecurve showing the impairment of the windkesselfunction.

4. Discussion

This study was aimed to obtain in the icefishpreliminary information on the involvement of NOin the control of cardiac function and gill circula-tion, together with an assessment of the sensitivityof the branchial vasculature to exogenous AChand 5-HT. We used the in vitro isolated andperfused working heart preparation ofC. hamatus,set up and standardized by us(Tota et al., 1991a),which perfused at constant pressure generatedphysiologically comparable values of output pres-sure, CO, ventricle work, and power(Hemmingsen

476 D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

Fig. 4. Effect of SIN-1 perfusion on input pressure(Pi) in the isolated icefish head preparation. Results are expressed as percent changes(D%) from control values; data are means"S.E. of 4 experiments(*P-0.05).

Fig. 5. Effect ofL-NIO pre-treatment on the action of 5-HT in the isolated icefish head preparation. Results are expressed as percentchanges(D%) from control values; data are means"S.E. of 3 experiments(*P-0.05).

et al., 1972), typical of the volume pump heart ofthe icefish(Tota et al., 1991a; Tota and Gattuso,1996). The reactivity of the branchial circulationwas analysed using an isolated head preparationperfused at constant physiological flow. A limita-tion of the preparation was the lack of physiolog-ical afterload on the efferent branchial circulation,so that changes in input pressure reflected princi-pally changes in vascular resistance between theventral aorta and the efferent vasculature. On theother hand, due to its extensive endothelial surface,which is the major putative source of NO, thispreparation may be a useful bioassay system foran overall assessment of NO-mediated vascularresponses.

There is no information on the regulation of thevascular resistance of the icefish branchial vascu-lature. Changes at this level, particularly in thearterio-arterial pathway, may have great influenceon the post-branchial pressure, which is the drivingpressure for perfusing the whole systemic vascu-lature. Moreover, local modulatory changes of thebranchial vascular bed may affect the perfusionpressure of the respiratory part of the gills, therebyinfluencing the oxygenation of the blood. A norad-renaline-mediated branchial vasodilation was notedin the icefish C. aceratus, but what type ofadrenoceptor is involved is unknown(Rankin,1989). Axelsson et al.(1994) found that in theAntarctic cryopelagic red-bloodedP. borchgrevinki

477D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

Fig. 6. Effects ofL-NMMA (10 M) and SNP(10 M) on cardiac parameters in isolated and perfused icefish heart(CO, cardiacy4 y4

output; HR, heart rate; SV, stroke volume; SW, stroke work; PO, power output). Results are expressed as percent changes(D%) fromcontrol values; data are means"S.E. of 6 experiments(*P-0.05).

Fig. 7. Pressure waveforms measured in the bulbus arteriosus of the icefish during control(a) and exposure to SNP(b).

there was a 30% increase of branchial resistanceduring exercise, which was blocked by the mus-carinic receptor antagonist atropine. In the samestudy these authors found evidence of ana-adrenergic vasoconstrictor tonus dominant over ab-adrenoceptor-mediated vasodilation. The presentstudy demonstrated the presence of a remarkableACh-mediated branchial vasoconstriction in theicefish, thus confirming the general vasoconstric-tory action of the cholinergic agonists so fardescribed in virtually all teleosts(see for refer-ences Olson and Villa, 1991; Pellegrino et al.,2002). The cholinergic vasomotion was mediatedby muscarinic receptors, since it was abrogated byatropine. Interestingly, in the same preparation, 5-

HT exerted a dose-dependent vasoconstriction, butthis agonist appeared much less potent than ACh.The 5-HT effect was completely suppressed bymethysergide(a 5-HT y5-HT receptor antago-1 2

nist). This confirms the notion that in fish 5-HTis a vasoconstrictor of the branchial vasculature,as also documented in the AntarcticT. bernacchiiand P. borchgrevinki, following in vivo intravas-cular injections of 5-HT(Axelsson et al., 1994).Sundin et al. (1998) provided evidence inP.borchgrevinki of a 5-HT receptor-mediated vaso-2

constriction in the branchial vasculature, but themechanisms involved in this regulation are stillunknown. It has been speculated that the highsensitivity to 5-HT, detected so far in few Antarctic

478 D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

red-blooded notothenioids, may be due to eitherspecies differences or to cold adaptation. Indeed,cold adaptation has been suggested as an expla-nation for the dominance ofa-adrenoceptor overb-adrenoceptor responses inP. borchgrevinki(Axelsson et al., 1994), mimicking the acclimatoryresponse of winter fish from temperate waters(Randall and Perry, 1992). In mammals, coolingenhances both 5-HT -receptor anda -adrenocep-2 2

tor-mediated vasoconstriction(Vanhoutte and Fla-vahan, 1986), presumably by an increase inreceptor affinity. If a direct effect of temperatureon receptor affinity applies also to 5-HT receptors2

in notothenioid fish and if there is an equivalentreceptor density among species, then sub-Antarcticand Antarctic species should display different sen-sitivities to serotonin depending on the environ-mental water temperatures. Our data show amarked predominance of the cholinergic vs. sero-toninergic vasoconstriction in the icefish gills, thusprompting analysis of a larger number of nototh-enioid species before accepting any generalizationon neuro-humoral pattern predominance of vascu-lar control in Antarctic teleosts.

Of note, the branchial vasoconstriction causedby serotonin inC. hamatus was NO-dependent.Since it has been shown thatL-NAME is also amuscarinic receptor antagonist(Buxton et al.,1993) and we found that the icefish gills are undera powerful cholinergic tone, we have used themore specific NOS inhibitorL-NIO (10 M) toy4

abolish the release of endogenous NO withoutinterfering with the cholinergic tone. Pre-treatmentwith this inhibitor abolished the serotoninergicvasoconstriction, demonstrating in the icefish gillsat the same time the presence of the nitrergicmodulation and its importance in 5-HT signalling.In fish the mechanism involved in 5-HT effects isstill unknown. In contrast, evidence in mammalshas shown the ability of 5-HT to release NO fromthe endothelium as well as the involvement of NOin the 5-HT-postreceptor signal transduction path-way. For example, the coronary vasodilationinduced by 5-HT results from stimulation of NOrelease from vascular endothelial cells(Ellwoodand Curtis, 1996) and 5-HT-induced NO produc-tion is mediated by both 5-HT B and 5-HT B1 2

receptor activation in human coronary artery endo-thelial cells(Ishida et al. 1998).

Along with this finding, we have also shownthat endogenous NO exerts a basal vasodilatortone in the unstimulated branchial circulation of

the icefish, since NOS inhibition byL-NIO induceda consistent vasoconstriction(20%), while theexogenous NO donor SIN-1 elicited a dose-dependent vasodilation. Interestingly, this NO-dependent vasodilation in the icefish, which istypical of mammals(see for references Moncadaand Higgs, 1995), contrasts the NO-mediated bran-chial vasoconstriction that we have detected in thefresh water European eel(Anguilla anguilla) (Pel-legrino et al., 2002). Whether this oppositeresponse of the gill vasculature to NO reflectsdiversities either in phylogeny or in adaptation orin variety of structural vascular sites remainsunknown.

The present investigation also shows that theicefish heart is sensitive to endogenous and exog-enous NO. In the working heart under basalconditions perfusion with the NOS inhibitorL-NMMA affected HR, eliciting a negative chrono-tropic response. In absence of any detectable effecton SV, the decrease of SW can be attributable tothe slight reduction of pressure, while the decreasein ventricular power output is influenced by thedecrease in HR. In contrast with the response ofthe branchial vasculature,L-NMMA did not showany appreciable difference on bulbus basal con-tractility, suggesting absence of a significantnitrergic tone on the smooth muscle activity ofthis chamber. In spite of that, the exposure to theexogenous NO donor SNP produced definedeffects both on bulbus and heart functionality. Asin the branchial vasculature, SNP exerted a dilationof the bulbus wall. This was a sudden and longlasting action, resulting in chamber enlargementwith loss of the windkessel function. This effectwas reflected by a modification of the pressurecurve, indicating an increase of the stiffness(dPydV) of the bulbus wall, with dP increasing signif-icantly and the chamber volume constantly beingenlarged, a phenomenon likely due to a state of‘atonic paralysis’ of the muscle. It is notable thata thorough study of the structure of the bulbus inC. hamatus has provided clear evidence that,despite the absence of nerve terminals, the smoothmuscle cells, located in the middle layer of thebulbus, appear to contribute significantly to themaintenance of the physical properties of the wall(Icardo et al., 1999). Moreover, the presence ofatrial natriuretic peptide receptors in the icefishbulbus is consistent with an autocrine–paracrinemodulation of the chamber tonus(Cerra et al.,1997).

479D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

On the heart performance, SNP exerted a posi-tive inotropic action, since it elicited increases ofSV, SW and PO. Although the present work cannotclarify if, and to which extent, these cardiacresponses are of physiological significance, itclearly indicates that the icefish heart is a targetof NO. Moreover, it appears that the icefish heart,as noted above with the gill vessels, responds toNO in a way opposite to that shown by someteleost and amphibian heart preparations. In fact,in the isolated and perfused working hearts of theeel A. anguilla (Imbrogno et al., 2001), the salmoSalmo salar (Gattuso et al., 2002), and the frogRana esculenta (Sys et al., 1997), endogenous andexogenous NO elicited negative inotropic effects.The same negative inotropic effects were alsodetected in several mammalian hearts(see Balli-gand, 2000, for references). The mechanismsunderlying these ‘anomalous’ nitrergic cardiovas-cular responses in the icefish are a challenge forfuture work.

Acknowledgments

This work was supported by the Italian NationalResearch Program in Antarctica(PNRA).

References

Acierno, R., MacDonald, J.A., Agnisola, C., Tota, B., 1995.Blood volume in the hemoglobinless antarctic teleostChion-odraco hamatus (Lonnberg). J. Exp. Zool. 272, 407–409.¨

Axelsson, M., Davison, B., Forster, M., Nilsson, S., 1994.Blood pressure control in the Antarctic fishPagotheniaborchgrevinki. J. Exp. Biol. 190, 265–279.

Axelsson, M., Davison, W., Forster, M.E., Farrel, A.P., 1992.Cardiovascular responses of the red-blooded antarctic fishes,Pagoteniia bernacchii and Pagothenia borchgrevinki. J.Exp. Biol. 167, 179–201.

Balligand, J.L., 2000. Regulation of cardiac function by nitricoxide. In: Mayer, B. (Ed.), Nitric Oxide. Handbook ofExperimental Pharmacology, vol. 143. Verlag, Berlin, Hei-delberg, New York, pp. 206–234.

Buxton, I.O., Cheek, D.J., Eckman, D., Westfall, D.P., Sanders,D.P., Keef, K.D., 1993.N -L-arginine methyl ester andG

other alkyl esters of arginine are muscarinic receptor antag-onists. Circ. Res. 72, 387–395.

Cerra, M.C., Canonaco, M., Acierno, R., Tota, B., 1997.Different binding activities of A- and B-type natriuretichormones in the heart of two Antarctic teleosts, the red-blooded Trematomus bernacchii and the hemoglobinlessChionodraco hamatus. Comp. Biochem. Physiol. 118A,993–999.

Davison, W., Axelsson, M., Nilsson, S., Forster, M.E., 1997.Cardiovascular function in Antarctic notothenioid fishes.Comp. Biochem. Physiol. 118A, 1001–1008.

Egginton, S., Skilbeck, C., Hoofd, L., Calvo, J., Johnston,I.A., 2002. Peripheral oxygen transport in skeletal muscleof Antarctic and sub Antarctic notothenioid fish. J. Exp.Biol. 205, 769–779.

Egginton, S., Forster, M.E., Davison, W., 2001. Control ofvascular tone in notothenioid fishes is determined by phy-logeny, not environmental temperature. Am. J. Physiol. 280,R1197–R1205.

Ellwood, A.J., Curtis, M.J., 1996. Mechanism of 5-hydroxy-tryptamine-induced coronary vasodilatation assessed bydirect detection of nitric oxide production in guinea-pigisolated heart. Br. J. Pharmacol. 119, 721–729.

Foster, M.E., Davison, W., Axelsson, M., Sundin, L., Franklin,C.E., Gieseg, S., 1998. Catecholamine release in heat-stressed Antarctic fish causes proton extrusion by red cells.J. Comp. Physiol. B 168, 345–352.

Fisher, W., Hureau, J.C., 1985. FAO species identificationsheets for fisheries purposes. Southern Ocean(Fishing areas45, 58 and 88). Commission for the Conservation ofAntarctic Marine Living Resources, vol. 2. FAO, Rome, pp.233–270.

Gattuso, A., Mazza, R., Imbrogno, S., Sverdrup, A., Tota, B.,Nylund, A., 2002. Cardiac performance inSalmo salar withInfectious Salmon Anaemia(ISA): putative role of nitricoxide. Dis. Aquat. Org. 52, 11–20.

Hemmingsen, E.A., 1991. Respiratory and cardiovascularadaptations in hemoglobin-free fish: resolved and unresolvedproblems. In: Di Prisco, G., Maresca, B., Tota, B.(Eds.),Biology of Antarctic Fish. Springer, Berlin, pp. 191–203.

Hemmingsen, E.A., Douglas, E.L., Johansen, K., Millard,R.W., 1972. Aortic blood flow and cardiac output in thehemoglobin-free fish Chaenocephalus aceratus. Comp.Biochem. Physiol. 43A, 1045–1051.

Hemmingsen, E.A., Douglas, E.L., 1970. Respiratory charac-teristics of the hemoglobin-free fishChaenocephalus acer-atus. Comp. Biochem. Physiol. 33, 733–744.

Holeton, G.F., 1970. Oxygen uptake and circulation by ahaemoglobinless fish(Chaenocephalus aceratus Lonnberg)compared with three red-blooded Antarctic fish. Comp.Biochem. Physiol. 34, 457–471.

Icardo, J.M., Colvee, E., Cerra, M.C., Tota, B., 1999. Bulbusarteriosus of the Antarctic teleost. I. The white-bloodedChionodraco hamatus. Anat. Rec. 254, 396–407.

Imbrogno, S., De Iuri, L., Mazza, R., Tota, B., 2001. Nitricoxide modulates cardiac performance in the heart ofAnguil-la anguilla. J. Exp. Biol. 204, 1719–1727.

Ishida, T., Kawashima, S., Hirata, K., Yokoyama, M., 1998.Nitric oxide is produced via 5-HT1B and 5-HT2B receptoractivation in human coronary artery endothelial cells. KobeJ. Med. Sci. 44, 51–63.

Johnston, I.A., Fitch, N., Zummo, G., Wood, R.E., Harrison,P., Tota, B., 1983. Morphometric and ultrastructural featuresof the ventricular myocardium of the haemoglobin-lessicefish Chaenocephalus aceratus. Comp. Biochem. Physiol.76A, 475–480.

McGeer, J.C., Eddy, F.B., 1996. Effects of sodium nitroprussideon blood circulation and acid–base and ionic balance inrainbow trout: indications for nitric oxide induced vasodi-lation. Can. J. Zool. 74, 1211–1219.

Moncada, S., Higgs, E.A., 1995. Molecular mechanisms andtherapeutic strategies related to nitric oxide. FASEB J. 9,1319–1330.

480 D. Pellegrino et al. / Comparative Biochemistry and Physiology Part A 134 (2003) 471–480

Moylan, T.J., Sidell, B.D., 2000. Concentrations of myoglobinand myoglobin mRNA in heart ventricles of antarctic fishes.J. Exp. Biol. 203, 1277–1286.

Mustafa, T., Agnisola, C., Hansen, J.K., 1997. Evidence forNO-dependent vasodilation in the trout(Oncorhynchusmykiss) coronary system. J. Comp. Physiol. B 167, 98–104.

Olson, K.R., Villa, J., 1991. Evidence against nonprostanoidendothelium-derived relaxing factor(s) in trout vessels. Am.J. Physiol. 260, R925–R933.

Pellegrino, D., Sprovieri, E., Mazza, R., Randall, D.J., Tota,B., 2002. Nitric oxide-cGMP-mediated vasoconstriction andeffects of acetylcholine in the branchial circulation of theeel. Comp. Biochem. Physiol. 132(A), 447–457.

Perry, S.F., Davie, P.S., Daxboeck, C., Ellis, A.G., Randall,D.J., 1982. A comparison of CO excretion in a spontane-2

ously ventilating blood-perfused trout preparation andsaline-perfused gill preparations: contribution of the bran-chial epithelium and red blood cell. J. Exp. Biol. 101,47–60.

Randall, D.J., Perry, S.F., 1992. Catecholamines. In: Hoar,W.S., Randall, D.J., Farrell, A.P.(Eds.), Fish Physiology,vol. XIIB. Academic Press, San Diego, CA, pp. 255–300.

Rankin, J.C., 1989. Blood circulation and gill water fluxes inthe icefish, Chaenocephalus aceratus Lonnberg. Antarct.¨Spec. Top. 87–91.

Ruud, J.T., 1954. Vertebrates without erythrocytes and bloodpigment. Nature 173, 848–850.

Sys, S.U., Pellegrino, D., Mazza, R., Gattuso, A., Andries,L.J., Tota, B., 1997. Endocardial endothelium in the avas-

cular heart of the frog: morphology and role of nitric oxide.J. Exp. Biol. 200, 3109–3118.

Sundin, L., Davison, W., Forster, M., Axelsson, M., 1998. Arole of 5-HT receptors in the gill vasculature of the2

Antarctic fishPagothenia borchgrevinki. J. Exp. Biol. 201,2129–2138.

Sundin, L., Nilsson, G.E., Block, M., Lofman, C.O., 1995.¨Control of gill filament blood flow by serotonin in therainbow troutOncorhynchus mykiss. Am. J. Physiol. 268,R1224–R1229.

Tota, B., Cerra, M.C., Mazza, R., Pellegrino, D., Icardo, J.,1998. The heart of antarctic icefish as paradigm of coldadaptation. J. Therm. Biol. 22, 409–417.

Tota, B., Gattuso, A., 1996. Heart ventricle pumps in teleostsand elasmobranchs: a morphodynamic approach. J. Exp.Zool. 275, 162–171.

Tota, B., Acierno, R., Agnisola, C., 1991a. Mechanical per-formance of the isolated and perfused heart of the haemo-globinless antarctic icefish Chionodraco hamatus(Lonnberg): effects of loading conditions and temperature.¨Philos.Trans. R. Soc. Lond.wBiol.x 332, 191–198.

Tota, B., Agnisola, C., Schioppa, M., Acierno, R., Harrison,P., Zummo, G., 1991b. Structural and mechanical character-istics of the heart of the icefishChionodraco hamatus,Lonnberg. In: Di Prisco, G., Maresca, B., Tota, B.(Eds.),¨Biology of Antarctic Fish. Springer, Berlin, pp. 205–219.

Vanhoutte, P.M., Flavahan, N.A., 1986. Effects of temperatureon -adrenoceptors in limb veins: role of receptor reserve.Fedn. Proc. Fedn. Am. Socs. Exp. Biol. 45, 2347–2354.

Wolf, K., 1963. Physiological salines for freshwater teleosts.Prog. Fish Cult. 25, 135–140.