The Journal of Experimental Biology 203, 1225Ð1239 (2000) 1225 Printed in Great Britain © The Company of Biologists Limited 2000 JEB2501

BRANCHIAL RECEPTORS AND CARDIORESPIRATORY REFLEXES IN A NEOTROPICAL FISH, THE TAMBAQUI (COLOSSOMA MACROPOMUM)

LENA SUNDIN1, STEPHEN G. REID1, F. TADEU RANTIN2 AND WILLIAM K. MILSOM1,* 1Department of Zoology, University of British Columbia, 6270 University Boulevard, Vancouver, British Columbia, Canada V6T 1Z4 and 2Department of Physiological Sciences, Federal University of São Carlos, 13565-905 São Carlos, SP, Brazil *Author for correspondence (e-mail: [email protected])

Accepted 26 January; published on WWW 9 March 2000

Summary This study examined the location and physiological triggering the elevation in systemic vascular resistance, roles of branchial chemoreceptors involved in the breathing amplitude, swelling of the inferior lip and that cardiorespiratory responses to hypoxia and hypercarbia in induce aquatic surface respiration during hypoxia are a neotropical fish that exhibits aquatic surface respiration, extrabranchial, although branchial receptors also the tambaqui (Colossoma macropomum). Fish were contribute to the latter two responses. Hypercarbia also exposed to abrupt progressive environmental hypoxia produced bradycardia and increases in breathing (18.6Ð1.3 kPa water PO∑) and hypercarbia (water frequency, as well as hypertension, and, while the data equilibrated with 5 % CO2 in air, which lowered the water suggest that there may be receptors uniquely sensitive to pH from 7.0 to 5.0). They were also subjected to injections changes in CO2/pH involved in cardiorespiratory control, of NaCN into the ventral aorta (to stimulate receptors this is based on quantitative rather than qualitative monitoring the blood) and buccal cavity (to stimulate differences in receptor responses. These data reveal receptors monitoring the respiratory water). All tests were yet another novel combination for the distribution of performed before and after selective denervation of cardiorespiratory chemoreceptors in fish from which branchial branches of cranial nerves IX and X to the gill teleologically satisfying trends have yet to emerge. arches. The data suggest that the O2 receptors eliciting reflex bradycardia and increases in breathing frequency Key words: fish, tambaqui, Colossoma macropomum, are situated on all gill arches and sense changes in both the chemoreceptor, hypoxia, hypercarbia, oxygen, carbon dioxide, gill, blood and respiratory water and that the O2 receptors blood pressure, heat rate, ventilation.

Introduction While much is known about the reflex cardiorespiratory primarily to arterial hypoxaemia, while less tolerant fish responses of teleost fishes to hypoxia and their ventilatory respond more immediately to aquatic hypoxia. responses to hypercarbia, there are few reports concerning The distribution of these receptors among the different gill cardiovascular responses to hypercarbia and surprisingly few arches also does not seem to be uniform amongst species. For studies, on only a handful of species, designed to determine the example, in the Atlantic cod (Gadus morhua) (Fritsche and locations and innervation of the receptors eliciting these reflex Nilsson, 1989), rainbow trout (Oncorhynchus mykiss) (Smith responses. and Jones, 1978; Daxboeck and Holeton, 1978), coho salmon Most teleosts respond to hypoxia with a substantial decrease (O. kisutch) (Smith and Davie, 1984) and traira (Sundin et al., in heart rate. In teleosts from temperate waters, this 1999), these receptors are located on the first gill arch, but in bradycardia appears to be triggered by activation of externally the catfish (Ictalurus punctatus) they appear to be located on oriented branchial receptors (Randall and Smith, 1967; all of the first three gill arches (Burleson and Smatresk, 1990a). Saunders and Sutterlin, 1971; Smith and Jones, 1978; Smatresk Furthermore, in an elasmobranch, the dogfish (Scyliorhinus et al., 1986; Burleson and Smatresk, 1990b; McKenzie et al., canicula), they are not confined to the gills but are also located 1991; Burleson and Milsom, 1993), while in a hypoxia-tolerant throughout the orobranchial cavity, where they are innervated neotropical fish, the traira (Hoplias malabaricus), the hypoxic by cranial nerves V (trigeminal) and VII (facial) (Butler and bradycardia is primarily elicited reflexively by activation of Taylor, 1971; Butler et al., 1977). The reasons for these internally oriented branchial receptors (Sundin et al., 1999). differences are also unclear. This may suggest that hypoxia-tolerant species respond The ventilatory response to hypoxia in teleosts consists of 1226 L. SUNDIN AND OTHERS an increase in both breathing frequency and amplitude. In the well-oxygenated surface layer (see Rantin and Kalinin, several studies on various species, denervation of cranial 1996). To facilitate this, the inferior lip swells to form a funnel nerves IX and X to the gills and the pseudobranch have failed that can direct the surface water into the mouth and over the to eliminate ventilatory responses to hypoxia (Saunders and gills. The lower lip is not involved in gas exchange but serves Sutterlin, 1971; Sundin et al., 1999). However, it has been purely as a mechanical structure enhancing skimming of the reported that complete denervation of the branchial branches surface water (see Val and Almeida-Val, 1995). While the of cranial nerves IX and X eliminates all the ventilatory mechanisms that induce aquatic surface breathing and lip response in the catfish (Burleson and Smatresk, 1990a) and the swelling are unknown, this mechanism provides these fish with gar (Lepisosteus osseus) (Smatresk, 1989). Thus, the results an alternative strategy to ventilation in hypoxic waters. Thus, from fish species studied to date suggest that ventilatory O2 the goals of this study were to document respiratory, chemoreceptors do not share common locations and cardiovascular, behavioural and morphological responses to distribution among species, just as the O2-sensitive hypoxia and hypercarbia in Colossoma macropomum and to chemoreceptors eliciting cardiac reflexes do not. identify the location (internal and external) and the distribution During exposure to environmental hypercarbia, arterial (branchial and extrabranchial) of O2-sensitive and CO2/pH- hypercapnia or intra-arterial injections of acid, fish increase sensitive chemoreceptors involved in these responses. As such, their ventilation rate and/or amplitude (Randall and Jones, this study was designed to contribute to our knowledge of 1973; Janssen and Randall, 1975; Randall et al., 1976; Thomas cardiorespiratory control in fish in general and, by examining and Le Ruz, 1982; Smith and Jones, 1982; Thomas et al., 1983; differences in receptor distribution promoting reflex responses Reid et al., 1999). This increase in breathing has been to hypoxia versus hypercarbia, to help resolve questions attributed to both indirect and direct effects of the changes in concerning the presence or absence of receptors uniquely pH/CO2 (Smith and Jones, 1982; Randall and Taylor, 1991; sensitive to changes in CO2/pH involved in respiratory control. Perry et al., 1992), but much of the evidence suggests that hypercarbia and/or hypercapnia can cause ventilatory increases directly, independent of changes in arterial oxygen content or Materials and methods plasma catecholamine levels (Butler and Taylor, 1971; Heisler Experimental animals et al., 1988; Perry and Kinkead, 1989; Wood et al., 1990; For these experiments, juvenile tambaqui Colossoma Burleson et al., 1992). There have been few studies designed macropomum (Cuvier, 1818) (659±32 g; mean ± S.E.M., N=28) to determine the locations and distribution of the receptors were obtained from CEPTA (Tropical Fish Research Centre)/ involved in the hypercarbic ventilatory response. One such IBAMA in Pirassununga SP, Brazil, and transported to the study in a neotropical fish, the traira (Hoplias malabaricus), Federal University of São Carlos. These fish are third- or showed that the hypercarbia-induced increases in both fourth-generation descendants of native tambaqui taken from breathing frequency and amplitude arose from receptors with the Amazon in 1993. Animals were maintained outdoors in a similar distribution to those that elicited the hypoxic fibreglass aquaria supplied with aerated and dechlorinated City ventilatory responses (Reid et al., 1999). The traira also of São Carlos tapwater. Temperature was maintained at 25 ¡C, exhibited a bradycardia and mild hypotension during exposure and the animals were exposed to a natural photoperiod. Fish to environmental hypercarbia (Reid et al., 1999) and, in this were fed ad libitum every second day, and experiments were case, the distribution of receptors involved in producing the performed between September and October. hypoxic and the hypercarbic bradycardia were different; those producing the hypoxic bradycardia were located exclusively in Animal preparation the first gill arch, while those producing the hypercarbic Animals were anaesthetized in an aqueous solution of bradycardia were found in other gill arches as well. These data benzocaine (100 mg l−1) predissolved in 2 ml of 70 % ethanol. suggested that receptors uniquely sensitive to changes in During surgery, the gills were ventilated with a second solution −1 CO2/pH may exist in fish and that their distribution may be of benzocaine (50 mg l ) gassed with air. Impedance distinct from that of receptors involved in the hypoxic electrodes were sutured to each operculum to monitor the ventilatory response. breath-by-breath displacement of the operculum and measure These previous results suggest that studies of the distribution ventilation rate (fV) and an index of ventilation amplitude of cardiorespiratory chemoreceptor populations for a variety of (VAMP). Using a dremel tool, a hole was drilled through the species, adapted to different habitats, are still needed before snout between the nostrils, and a flared cannula (PE 160) was testable hypothesis concerning the evolution, phylogeny and fed from inside the mouth out through the hole and was secured adaptive significance of these receptors and their distribution with a cuff. This allowed administration of NaCN and HCl can be formulated. The present study focuses on the solutions into the buccal cavity to stimulate putative O2 and cardiorespiratory responses to hypoxia and hypercarbia of CO2/pH chemoreceptors (external) on the gills monitoring the another hypoxia-tolerant neotropical fish, the tambaqui respiratory water. A second cannula (PE 50) was inserted into (Colossoma macropomum). The tambaqui, however, is an the afferent branchial artery of the third gill arch and advanced aquatic surface breather that, under conditions of towards the ventral aorta. This cannula was used to measure environmental hypoxia, will come to the surface and siphon ventral aortic blood pressure (PVA) and heart rate (fH) as well Branchial chemoreceptors in tambaqui 1227 as to inject solutions of saline, NaCN and HCl to stimulate following order: (1) internal saline, (2) internal NaCN (0.5 ml −1 −1 putative O2 and CO2/pH chemoreceptors (internal) that of 2mgml NaCN in saline), (3) internal HCl (0.125 mmol l monitor the blood. in 0.3 ml of saline), (4) external water, (5) external NaCN The operculum was reflected forward, and a small incision (1 ml of 2 mg ml−1 NaCN in water) and (6) external HCl (approximately 1 cm) was made in the epithelium at the dorsal (0.125 mmol l−1 in 0.5 ml of water). In each case, the cannula end of the first and second gill arches where they meet the roof was flushed with 0.5 ml of saline (for internal injections) or of the opercular cavity. This permitted access to cranial nerve 1.0 ml of water (for external injections) to ensure complete IX (glossopharyngeal) and the pretrematic branch of cranial drug delivery. After each injection, cardiorespiratory variables nerve X (vagus) innervating the first gill arch (G1 group, were recorded for 3 min. If pre-injection levels of fV, VAMP, N=10). Tambaqui do not have a pseudobranch. For complete PVA and fH were not restored within that period, subsequent branchial denervation, the incision was enlarged injections were delayed until these variables returned to approximately 1 cm in the caudal direction. The branchial previous levels or stabilized at new levels. nerves to all gill arches were carefully dissected free of Next, the animals were subjected to abrupt, progressive connective tissue and cut with fine iris scissors (G4 group, environmental hypoxia by shutting off the airflow and gassing N=7). In all cases, the cardiac and visceral branches of the the tank with nitrogen. The PwO∑ was lowered from an air- vagus (X) were left intact. In the control group (N=11), the saturated level of 18.6 kPa (140 mmHg; 25 ¡C) to 1.3 kPa nerves were left intact, but in three of these animals the nerves (10 mmHg) over approximately 10 min. At this point, the were exposed (sham operation) but not sectioned. There was nitrogen flow was halted, airflow was restored and the water no sham effect. The healing process in this species was rapid, PO∑ gradually returned to normoxic levels. Finally, the animals and the incision was covered with ‘scar tissue’ within 24 h. All were subjected to abrupt, progressive environmental denervations were confirmed post mortem by autopsy. hypercarbia by gassing the tank with 5 % CO2. Initial After surgery, the animals were manually ventilated with hypercarbic experiments demonstrated that equilibrating the aerated water, and as soon as they showed signs of coming out water with 0.1 %, 0.25 %, 0.5 % or 0.75 % CO2 had no effect of the anaesthesia they were placed into individual cylindrical on ventilation or heart rate/blood pressure, whilst 1.25 % and tubes housed within larger experimental tanks (approximately 2.5 % CO2 elicited very modest changes. Consequently, the 80 l). Mesh covered the ends of the cylindrical tube. This animals were exposed to 5.0 % CO2 such that the water pH fell facilitated rapid equilibration of the water within the tube with from approximately 7.0 to 5.0 over 10 min. This reliably the water in the holding tank. A large slit on top of the tubes stimulated ventilation in all control animals. The animals were permitted the impedance leads/cannulae to exit the tank. The then returned to normocapnic conditions, and cardiorespiratory tank was covered to maintain a dark and quiet environment for variables were monitored until the water pH returned to at least the fish. Animals were allowed to recover for a minimum of 6.5. These changes in pH and PO∑ (both magnitude and rate of 24 h prior to experimentation. change) are well within physiological ranges for these fish. For example, measurements made at different times during the Experimental protocols annual flood cycle in the Amazon and Orinoco rivers, in which Following the 24 h recovery period, the opercular impedance these fish are typically found, show pH ranging from 3.8 to 8.0 −1 leads were connected to an impedance converter to measure fV with dissolved O2 concentration in the range 5.1Ð0 mg l −1 (breaths min ) and VAMP (arbitrary units). The afferent through the water column. During the dry periods, levels of branchial artery cannula was connected to a pressure CO2 are lower, and pH and O2 levels are higher, in these waters −1 transducer to measure PVA (kPa) and fH (beats min ). The (Val and Almeida-Val, 1995). Since these experiments were partial pressure of oxygen in the water (PwO∑) was monitored designed to study the effects of progressive branchial continuously with an oxygen electrode (FAC 001 O2 electrode denervation on cardiorespiratory responses, rapid progressive and FAC 204A oxygen analyser) supplied, via siphon, with a changes in pH and PO∑ were used to produce strong responses steady flow of water from the experimental chamber. The whose modulation by denervation would be most apparent. electrode was calibrated with solutions of sodium bisulphate Removal of a modest response by denervation would be much in borax (PO∑=0 kPa) and air-equilibrated water (PO∑=18.6 kPa; less convincing. 25 ¡C). Water pH was continuously measured with a pH Because of a persistent decrease in fH during hypoxia and electrode calibrated with standard solutions. Prior to initiating hypercarbia in the G4 fish (see Results), these animals were the experimental protocol, the fish were left undisturbed for subsequently treated with atropine (1 mg kg−1, Sigma), and the approximately 30 min to allow fV, VAMP, fH and PVA to experiments were repeated to test for direct effects of hypoxia, stabilize at steady levels. hypercarbia and NaCN on the heart. This procedure should First, the animals were subjected to a series of internal (into block receptor-mediated reflex responses at the heart and the branchial artery cannula) and external (into the snout reveal whether the bradycardia demonstrated by the G4 fish cannula) injections of NaCN and HCl to stimulate putative O2 was due to a direct effect of each treatment on the heart. and CO2/pH chemoreceptors. Injections of the vehicle alone (0.9 % NaCl for internal and water for external injections) Buccal and opercular pressure versus respiratory impedance served as controls. The injections were administered in the Our primary measure of breathing was the change in 1228 L. SUNDIN AND OTHERS

Fig. 1. (A) Frontal view of a tambaqui with the mouth agape and the mouth flap open during buccal expansion, and (B) with the mouth flap closed during the buccal compression phase of the ventilatory cycle. (C) Lateral view of the head of a tambaqui approaching the water surface illustrating the initial swelling of the lower lip following hypoxic exposure, and (D) a dorsal view of the same fish illustrating the forward expansion of a lip near full development. electrical impedance measured across the orobranchial cavity. hypoxia, for the final 30 s of the hypoxic exposure (water To confirm the accuracy of our impedance measurements, we PO∑=1.3 kPa) and after 30 min of recovery as well as alternately measured buccal and opercular pressures in three immediately prior to the initiation of hypercarbia, for the final fish, using appropriate cannulae, and simultaneously measured minute of the hypercarbic exposure (water pH=5.0) and after respiratory impedance. To stimulate breathing further under 30 min of recovery from hypercarbia. During the injection these conditions, we injected NaCN (1 ml of 500 µgml−1 experiments, data were analyzed for a 30 s control period NaCN) into the mouth through the buccal cannula. immediately prior to an injection of water, saline, NaCN or HCl and at 10 s intervals for the first minute post-injection. During Lip swelling and aquatic surface respiration the second and third minute post-injection, data were analyzed Intact and G4 fish (the same fish that had been used in for a 30 s period each minute. Maximum responses are reported. the protocols described above) were further exposed to fH, PVA and fV are reported as absolute values. Since VAMP ∑=2.0 kPa) for approximately was measured in arbitrary units, V and total ventilation environmental hypoxia (PwO . AMP 3.5 h (time taken to reduce PO∑ was approximately 0.5 h). (VTOT=VAMP×fV) are reported as a percentage change from the Before these experiments, however, all instrumentation except control value. In the second series of experiments, either the the PVA cannula was removed under anaesthesia, and the animals performed aquatic surface respiration or they did not. No animals were transferred to a large glass aquarium where they quantification, in terms of frequency or time spent at the surface, were allowed to recover for 1Ð2 days. Three sides of the was made. The extent of lip formation was estimated visually. aquarium were covered to minimize disturbance, but one side was left open to allow visual observation of the fish. Hypoxia Statistical analyses was induced by gassing the water with nitrogen, and a fan The data are reported as the mean ±1 standard error of the directed a constant stream of air across the surface of the water mean (S.E.M.). Differences in resting values for each variable to prevent this gas from accumulating above the surface. The between the intact animals and the denervated groups, before effects of denervation on the degree of inferior lip swelling and and after treatment with atropine, were tested by unpaired and the incidence of aquatic surface respiration were monitored paired Student’s t-tests, where appropriate. Data were visually. Tambaqui normally develop inferior lips fully within compared using one-way repeated-measures analysis of approximately 2Ð3 h of exposure to hypoxia (Braum and Junk, variance (ANOVA) to test for the significance of changes in 1982; Val and Almeida-Val, 1995). response to each stimulus. If significant differences (Pр0.05) were found, a Dunnett’s multiple-comparison test was used as Data analysis a post-hoc test. To evaluate the effect of selective denervations Cardiovascular and respiratory variables were analyzed for a on the responses to the different treatments, a two-way 1 min control period immediately prior to the initiation of repeated-measures ANOVA was used. Branchial chemoreceptors in tambaqui 1229

A Buccal opercular

Pressure

Fig. 2. Recordings of the respiratory impedance and Impedance the pressure in the buccal or opercular cavity in a Ratio tambaqui under normoxic, normocarbic conditions. 1:1 1:1.5 1:1.3 1:1.6 In A, note that, when the input to the pressure transducer was switched from the buccal cannula to B the opercular cannula, the pressure and impedance NaCN fluctuations were no longer correlated. The Opercular opercular rhythm was slower than the buccal and pressure impedance rhythms. In B, when NaCN was applied to stimulate respiration (at the arrow), the frequency and amplitude of buccal movements increased (as Impedance indicated by the changes in impedance), while opercular movements remained unchanged. Ratio 1:1.2 1:1.6

Results tend to be asynchronous (coupling ratios from 1:1 to 1:1.6, Ventilatory mechanics: buccal and opercular pressure versus equivalent to approximately 2:3). When breathing was respiratory impedance stimulated by the injection of NaCN into the respiratory water Tambaqui skimmed water at the surface of open tanks under flow, the frequency and amplitude of the impedance hypoxic conditions. At this time, the inferior lip became fluctuations (as well as the buccal pressure fluctuations, which swollen and was used as a funnel to direct the surface film of are not shown) increased, while the rate and magnitude of the water into the mouth. The mouth remained agape the entire time opercular pressure fluctuations were unchanged (Fig. 2B). the fish was breathing the surface water. On close examination, Similar results were obtained in all three animals studied. we observed that there was no water reflux through the mouth Aquatic surface breathing and lip swelling during buccal contraction despite the fact the mouth was agape. This was because these fish possess loose epithelial flaps During hypoxia, tambaqui with access to the water surface attached to the upper and lower jaw that can act as a ‘pocket initiated aquatic surface respiration and, to aid skimming of the valve’ to prevent reflux when the pressure in the buccal cavity surface water, their lower lips swelled to form a funnel exceeds ambient pressure. Fig. 1A,B shows the opening to the (Fig. 1C,D). From Table 1 it is clear that, even when the nerves mouth from the front with the mouth open wide and the mouth to all gill arches were sectioned, the tambaqui still performed flap open (Fig. 1A) and closed (Fig. 1B). These flaps are drawn open, presumably by negative pressure, during buccal Table 1. Effect of total gill denervation on the incidence of expansion and are forced to close during buccal contraction. aquatic surface respiration and the degree of lower lip Fig. 1C depicts a fish skimming water at the surface of the tank, swelling in tambaqui while Fig. 1D shows the swollen lower lip from a dorsal view. Denervated Our primary measure of breathing was the change in Control (G4) electrical impedance measured across the orobranchial cavity. In trying to determine the source of apparent recording Lip Lip artefacts arising from injections into the mouth through the Individual swelling ASR Individual swelling ASR buccal cannula, we discovered that the buccal and opercular C1 +++ * D1 + * rhythms could vary independently. Thus, to confirm the C2 ++ * D2 + * accuracy of our impedance measurements, we alternately C3 +++ * D3 ++ * measured buccal and opercular pressures in three fish, using C4 +++ * D4 ++ * C5 + * D5 ++ * appropriate cannulae, and simultaneously measured C6 +++ * D6 + * respiratory impedance to examine the effects of the C7 +++ * D7 + * buccal/opercular asynchrony on our impedance measurements. C8 +++ * Fig. 2 illustrates simultaneous pressure and impedance C9 +++ * recordings obtained in one fish. Note that the buccal pressure fluctuations (Fig. 2A, initial portion of upper trace) are ASR, aquatic surface respiration. synchronous with the changes in respiratory impedance (a +++, Fully developed; ++, half developed; +, poorly developed; coupling ratio of 1:1), but that opercular pressure fluctuations *fish performs ASR. 1230 L. SUNDIN AND OTHERS A Nitrogen on Off

5.5

4.0 (kPa)

Blood pressure 2.5

1 min Ventilation

B 2

O 18.7 10.7 5.3 2.7 2.7 5.3 10.7 18.7 P (kPa) 5.5

4.0 (kPa)

Blood pressure 2.5 Ventilation

10 s Fig. 3. Representative recordings showing cardiovascular and respiratory changes in a sham-operated tambaqui during progressive hypoxia and recovery. (A) Recordings in a compressed form to show the entire response. (B) Traces made at specific water PO∑ levels to display details of the response. Note the fast recovery of heart rate when the nitrogen was turned off and the difference in heart rate at the same PO∑ (2.7 kPa) between hypoxia and recovery. aquatic surface respiration and began to develop swelling of figure that, when the water PO∑ falls (past approximately 16 kPa the lower lip in response to hypoxia. Subjectively, the degree in this figure), fH begins to decrease whilst both fV and VAMP of swelling of the lower lip, however, was attenuated. increase. Upon the restoration of airflow, fH increased rapidly and rose to substantially greater levels than observed prior to Effects of progressive denervation on resting levels of the initiation of hypoxia. Ventilation (fV and VAMP) was slower cardiorespiratory variables to respond to the restoration of O2 levels in the water. There was a trend for resting fH and fV to increase with Fig. 4A shows original recordings depicting the changes in progressive levels of denervation of the branchial arches (G1 ventilation (fV and VAMP) and in fH and PVA during the abrupt and G4); however, progressive denervation had no statistical transition from normocarbia (air) to severe hypercarbia (water effect on resting PVA, fH or fV (Table 2). pH 5.0) and back to normocarbia in one control fish. It is evident from the extracted data (Fig. 4B) that, when the water Responses to abrupt hypoxia, hypercarbia and NaCN and was gassed with 5.0 % CO2, causing the water pH to fall from HCl injections in intact animals 7.0 to 5.0, fH clearly decreased whilst PVA, fV and VAMP In Fig. 3A,B, original recordings of fV and VAMP and of fH increased. These changes (bradycardia and hyperventilation) and PVA in a sham-operated fish (intact gill innervation) during were typical responses of tambaqui to environmental abrupt, progressive hypoxia are shown. It is evident from this hypercarbia prior to any gill denervation. Branchial chemoreceptors in tambaqui 1231 A 5% CO2 on Off

5.5

4.0 (kPa)

Blood pressure 2.5

Ventilation 1 min

B 7.0 6.0 5.5 5.0 5.0 5.5 6.0 6.5 pH 5.5

4.0 (kPa)

Blood pressure 2.5 Ventilation 10 s Fig. 4. Representative recordings from an intact fish showing the cardiovascular and respiratory changes during acute progressive hypercarbia and recovery. (A) Recordings in a compressed form to show the entire response. (B) Recordings taken at specific water pH levels to display details of the response.

Table 2. Resting values of cardiorespiratory variables before without effect (data not shown). It is possible that the dose of and after selective denervation and atropine treatment acid used was not sufficient to produce significant pH changes Control, G1, G4, G4 + atropine, at the receptor sites. Alternatively, it is possible that the Variable N=11 N=10 N=7 N=6 receptors mediating the hypercarbic responses are primarily + − stimulated by CO2 rather than extracellular H per se. Fish did fH (beats min 1) 49.1±6.0 49.6±4.9 58.6±3.1 64.4±2.0 not tolerate stronger doses of acid well, however, preventing PVA (kPa) 3.7±0.2 3.7±0.3 3.8±0.4 3.8±0.5 − us from pursuing this further. The lack of response to the acid fV (breaths min 1) 32.5±2.5 38.9±2.4 37.1±2.5 40.7±3.8 injections also makes it impossible to determine whether the Values are mean ± S.E.M. at 25 ¡C. receptors mediating the hypercarbic response were internally Nerve transection did not alter any variable significantly. or externally oriented. fH, heart rate; PVA, ventral aortic pressure; fV, ventilation rate; G1, nerves IX and the pretrematic branch of nerve X innervating the first Cardiovascular reflexes gill arch sectioned; G4, complete branchial denervation (all gill Direct stimulation of oxygen-sensitive receptors (hypoxia, arches). internal and external injections of NaCN) produced a rapid decrease in fH that remained after the first gill arch had been Both internal and external NaCN injections also produced denervated (Fig. 5). After all the gill arches had been reflex bradycardia and increased ventilation (Figs 5, 7, 8, 9). denervated (G4 fish), hypoxia and internally (but not Acid injections, both internal and external, were notably externally) injected NaCN still produced a significant, small, 1232 L. SUNDIN AND OTHERS A B Hypoxia Hypercarbia 80 )

-1 60 * * *

40 *

Fig. 5. The effects of (A) abrupt * hypoxia, (B) abrupt hypercarbia, Heart rate (beats min 20 (C) 1Ð2 mg ml−1 NaCN injected internally into the ventral aorta and (D) 1.0 ml of 1Ð2 mg ml−1 NaCN 0 added externally via a snout C G1 G4 G4+A C G1 G4 G4+A cannula into the respiratory water on heart rate in intact fish (C), in fish with nerve IX and the C D pretrematic branch of nerve X to Internal NaCN External NaCN the first gill arch sectioned (G1), in 80 fish with nerve IX and branchial branches of nerve X to all four gill arches sectioned (G4) and in the G4 ) 60 fish after pretreatment with atropine -1 * * (G4+A). The open columns represent the starting condition, the * * grey columns represent the final 40 response and the black columns represent the recovery values. The * * data are shown as the mean + S.E.M.

Values marked with an asterisk are Heart rate (beats min 20 significantly different (Pр0.05) from starting values in that condition. See Table 2 for values of 0 N. C G1 G4 G4+A C G1 G4 G4+A

− but more slowly developing, bradycardia. By pre-treating the Thus, as fH markedly fell from 47 to 20 beats min 1 in control fish with atropine, we tested whether this response was a and G1 fish (Fig. 5), PVA also fell (Fig. 6). Just like the cholinergic reflex response or a nonspecific direct effect on the bradycardia, this pressure decrease was abolished in G4 heart. Despite atropine treatment, this small, slowly developing animals. bradycardia still remained (Fig. 5). The effects of hypercarbia on PVA are also illustrated in Hypercarbia (exposure to 5.0 % CO2) with the gill Fig. 6. In the control group, despite the bradycardia, there was innervation intact also caused heart rate to decrease. This a progressive increase in blood pressure. While this response response was abolished in the G1 denervated group; there was appeared unchanged following denervation of the first gill arch a trend for heart rate to decrease during hypercarbia in this (G1), this increase was no longer statistically significant. Blood group, but this decrease was not statistically significant. This pressure again increased significantly in the G4 fish during was not affected by total gill denervation (G4) nor by pre- hypercarbia (pH 5.0), despite the tendency for heart rate to treatment with atropine (1 mg kg−1) to block cholinergic reflex decrease. Identical changes occurred in this group following responses on the heart (Fig. 5). pre-treatment with atropine, although these changes were again Despite the reflex hypoxic bradycardia, there were no not significant because of high interindividual variability. significant changes in PVA in any of the fish groups in response to hypoxia (Fig. 6). When NaCN was injected into the blood, Respiratory reflexes all groups of animals maintained or slightly increased Hypoxia and internal NaCN produced a rapid increase in fV (although not significantly) PVA (Fig. 6). Changes in PVA in both the control and the G1 fish (Fig. 7) that was abolished following external NaCN injection reflected the changes in fH. by complete branchial denervation. External NaCN also Branchial chemoreceptors in tambaqui 1233

A B Hypoxia Hypercarbia

5 * *

4

3

2 Blood pressure (kPa) 1

0 C G1 G4 G4+A C G1 G4 G4+A

C D Internal NaCN External NaCN

5

Fig. 6. The effects on blood pressure (kPa) of abrupt hypoxia (A), abrupt 4 hypercarbia (B) and injections of * NaCN either internally into the 3 * ventral aorta (C) or externally into the respiratory water (D). The data are shown as the mean + S.E.M. See 2 Fig. 5 for abbreviations. Values marked with an asterisk are Blood pressure (kPa) significantly different (Pр0.05) 1 from starting values in that condition. See Table 2 for values of 0 N. C G1 G4 G4+A C G1 G4 G4+A produced a rapid increase in fV in the control and G1 fish that amplitude, while others (4/11 in the control group and 4/10 in was not eliminated by complete branchial denervation; the G1 group) showed no change or a decrease in amplitude. . external NaCN still increased fV, albeit more slowly. In the Because of the amplitude responses, VTOT also increased in control and G1 fish, ventilation rate also increased during all the groups during hypoxia, with the increase in the control exposure to hypercarbia. Ventilation rate in the G4 fish did not fish being substantially greater than in the G1 and the G4 fish increase significantly during hypercarbia. (Fig. 9). Internal and external NaCN injections caused rapid . in in both the control and G1 groups, but only Hypoxia and internal NaCN significantly elevated VAMP increases in VTOT . all experimental groups (control, G1 and G4), and there were external NaCN treatment produced an increase in VTOT in the no differences in responses to the treatments detected between G4 group. During hypercarbia, the control group exhibited a . the groups (Fig. 8). External NaCN injection also increased doubling of VTOT, which was attenuated in the G1 group and VAMP in the intact animals and to a lesser extent in the G1 and absent in the G4 group. G4 fish. Ventilation amplitude did not increase significantly in any group during hypercarbic exposure (Fig. 8). There was a Discussion tendency for ventilation amplitude to increase in some Ventilatory mechanics: buccal and opercular pressure versus individuals in the control and G1 groups, but there was respiratory impedance tremendous variability in this response. Some fish (7/11 in the While observing tambaqui breathing during hypoxia in an control group and 6/10 in the G1 group), such as the one open tank, we noticed that these fish possessed a flap-like valve depicted in Fig. 4, showed large increases in breathing in the mouth (Fig. 1A,B). This flap appears to be composed of 1234 L. SUNDIN AND OTHERS A B Hypoxia Hypercarbia

) 60 * * -1 * 50 * 40

30

20

10 Ventilation rate (breaths min 0 C G1 G4 C G1 G4

C D IInternal NaCN External NaCN

) 60 -1 * * * * 50 * 40 Fig. 7. The effects on ventilation rate of abrupt hypoxia (A), abrupt hypercarbia (B) and injections of NaCN 30 either internally into the ventral aorta (C) or externally into the respiratory water (D). The data are shown as 20 the mean + S.E.M. See Fig. 5 for abbreviations. Values marked with an asterisk are significantly different 10 р Ventilation rate (breaths min (P 0.05) from starting values in that condition. See 0 Table 2 for values of N. C G1 G4 C G1 G4 thin epithelial sheets, which extend from the margins of the separate central rhythm generators for buccal and opercular upper and lower jaw and act like a pocket valve. These flaps rhythms (which are often entrained) and that only the buccal collapse against the roof and floor of the mouth during the rhythm generator responds to aquatic respiratory stimuli negative pressure expansion phase of the buccal cycle (unfortunately no internal injections of NaCN were given in (Fig. 1A), but fill with water and close, sealing the entrance to this experimental series). This intriguing suggestion requires the mouth, during the positive-pressure compression phase further research. This is quite different from the dissociation (Fig. 1B). As such, they prevent reflux of water back through seen between buccal and opercular movements in air- the open mouth, allowing the fish to ventilate the gills breathing fish, where opercular movements stop while buccal efficiently while still maintaining the mouth gape. It also movements continue during air breaths. In the tambaqui, the undoubtedly serves to prevent disturbance of the surface layer. opercular rhythm does not simply stop but continues at a Since the top 1Ð5 mm of the water column can be the sole distinctly different rate from that of the respiratory impedance source of oxygen during periods of extreme hypoxia in natural (buccal rhythm). habitats, this valve mechanism may serve to prevent disturbances of the surface layer that would make aquatic Responses to hypoxia: lip swelling and aquatic surface surface respiration pointless. breathing We also inadvertently discovered that the buccal rhythm did To ascertain whether the aquatic surface respiration and lip not always coincide with the rate of opercular expansion and swelling were induced by receptors located in the gills, free- compression. In examining this further, we discovered that swimming tambaqui were exposed to hypoxia prior to and buccal pressure cycles were synchronous with the changes in after complete branchial denervation. Our results clearly respiratory impedance (a coupling ratio of 1:1) and that both show that fish still displayed aquatic surface respiration responded well to respiratory stimuli. Opercular pressure following complete branchial denervation which, fluctuations, however, could, from time to time, be subjectively, was indistinguishable from the behaviour asynchronous with the changes in respiratory impedance exhibited by intact fish. Denervated fish did show a clear (coupling ratios 1:1Ð2:3) and did not respond to respiratory reduction in the degree of swelling of the lower lip compared stimuli. This suggests the possibility that there may be with intact animals, however, indicating that lip swelling Branchial chemoreceptors in tambaqui 1235

A B Hypoxia Hypercarbia 120 * * 100 * 80 60 40 20 0

Ventilation amplitude (% change) -20 C G1 G4 C G1 G4

C D Internal NaCN External NaCN 120

100 * Fig. 8. The effects on ventilation amplitude (percentage change from starting values) of abrupt 80 hypoxia (A), abrupt hypercarbia (B) and injections of NaCN either internally into the ventral aorta (C) or 60 * * * * externally into the respiratory water (D). The data are * 40 shown as the mean + S.E.M. The grey columns depict the final response and the black columns represent the 20 recovery values. Columns marked with an asterisk are

р Ventilation amplitude (% change) significantly different (P 0.05) from starting values 0 (100 %) in that condition. See Table 2 for values of N. C G1 G4 C G1 G4 resulted from the stimulation of both branchial and and Nilsson, 1989; rainbow trout, Smith and Jones, 1978; extrabranchial receptors. Daxboeck and Holeton, 1978; Smith and Davie, 1984; traira, Sundin et al., 1999), the component of this response that Cardiovascular responses: heart rate appeared to be reflex in tambaqui was not abolished by Acute and rapidly induced hypoxia or hypercarbia produced denervation of the IXth and Xth cranial nerves to only the first a marked bradycardia in the present study. During hypoxic gill arch. Other examples where O2-sensitive receptors exposure, progressive denervation reduced the magnitude of eliciting cardiac reflexes are situated outside the first gill arch the bradycardia, but did not eliminate it completely. When all are the channel catfish, in which these receptors are found on branchial nerves were sectioned in the G4 group of tambaqui, the first three gill arches (Burleson and Smatresk, 1990a), and the bradycardia that was still present was not eliminated by an elasmobranch, the dogfish shark (Scyliorhinus canicula), in pre-treatment with atropine, indicating that the sustained which they are found on all the gill arches as well as within bradycardia was not a receptor-mediated vagal reflex response. the orobranchial cavity (Butler et al., 1977). It is not yet clear Branchial motor nerves control the positioning of the gill what these differences in distribution reflect since, at present, curtain in the respiratory water flow and the distribution of they cannot be correlated with degree of hypoxia tolerance, blood flow through the gill filaments (Nilsson, 1984), and a habitat preference or phylogenetic position. large reduction in arterial PO∑ following complete bilateral Injections of NaCN into the blood or the respiratory water denervation of cranial nerves IX and X has been reported in produced a rapid bradycardia in both the control and G1 fish, the sea raven (Hemitripterus americanus) (Saunders and suggesting that both internally and externally oriented O2 Sutterlin, 1971). Therefore, it is likely that the hypoxaemia chemoreceptors are involved in eliciting the reflex decrease in developed in the G4 fish during hypoxic exposure could have fH in tambaqui. Similar results have been obtained in the been particularly severe and that this could directly have bimodally breathing gar (Lepisosteus osseus) (Smatresk et al., affected the heart. 1986). In contrast, these receptors appear to be exclusively A hypoxic bradycardia is a typical response in fish, but in externally oriented in most water-breathing teleosts (for contrast to most teleosts, in which this reflex arises from review, see Burleson et al., 1992) but internally oriented in receptors located on the first gill arch (Atlantic cod, Fritsche traira (Sundin et al., 1999). These intra-species differences in 1236 L. SUNDIN AND OTHERS

A B Hypoxia Hypercarbia 280 * 240 200 160 * 120 * * 80 * 40 Total ventilation (% change) 0 C G1 G4 C G1 G4

C D Internal NaCN External NaCN 280 * 240 Fig. 9. The effects on total ventilation (percentage 200 change from the starting values) of abrupt hypoxia * (A), abrupt hypercarbia (B) and injections of NaCN 160 either internally into the ventral aorta (C) or 120 * externally into the respiratory water (D). The data are * * shown as the mean + S.E.M. The grey columns depict 80 the final response and the black columns represent the recovery values. Columns marked with an asterisk 40 Total ventilation (% change) are significantly different (Pр0.05) from starting 0 values in that condition. See Table 2 for values of N. C G1 G4 C G1 G4 interbranchial distribution are even harder to interpret than bradycardia being initiated by receptors on only the first their intrabranchial distribution and do not support the arch. suggestion that hypoxia-tolerant species are more sensitive to arterial hypoxaemia while less tolerant species are more Cardiovascular responses: blood pressure sensitive to aquatic hypoxia. Although hypoxia in teleost fish commonly produces In the case of hypercarbia, the fall in heart rate in the G1 hypertension, as a result of hypoxic bradycardia, some fish and G4 fish was not significant, suggesting that the reflex species exhibit a constant blood pressure or even hypotension was elicited by branchial receptors confined to the first gill (see Fritsche and Nilsson, 1993; Sundin et al., 1999). In arch. Thus, the distribution of receptors producing the tambaqui following internal and external NaCN treatment and hypercarbic bradycardia in tambaqui appears to be different hypoxia, the only significant change in blood pressure was a from that of the receptors mediating the hypoxic bradycardia decrease of approximately 1 kPa in the control and the G1 fish and strengthens arguments that changes in CO2/pH may caused by external injection of NaCN. Judging from these act as independent cardiac stimuli in fish. A similar results, it appears that tambaqui have good barostatic control conclusion was drawn in the only other study of which we and can maintain blood pressure within a narrow range despite know that documents the distribution of chemoreceptors large changes in fH. Reflex corrections to maintain blood mediating hypercarbic bradycardia in a teleost (Reid et al., pressure could include either increased cardiac stroke volume 1999). The receptors involved in eliciting the hypercarbic or systemic resistance or both. bradycardia in traira, however, were distributed differently The overall effects of hypercarbia on blood pressure were from those in tambaqui. While the receptors mediating the quite small. In all groups of fish, however, there was a mild hypercarbic bradycardia in tambaqui are located on the hypertension (a 15Ð20 % increase in blood pressure). This was first arch, those mediating the hypoxic bradycardia are significant in the control and total-gill-denervated groups located possibly on all gill arches; in traira, the situation is despite a significant bradycardia in the control fish. This could reversed, with the hypercarbic bradycardia being mediated only arise from a concomitant increase in stroke volume or by receptors possibly on all gill arches and the hypoxic systemic vascular resistance and, since this still occurred in Branchial chemoreceptors in tambaqui 1237 animals after complete gill denervation, it must have been which ventilatory responses were unaffected by complete gill mediated by receptors outside the gills or by direct effects of denervation (Hughes and Shelton, 1962; Saunders and the changes in pH/PCO∑ on the vasculature. Extrabranchial Sutterlin, 1971; Sundin et al., 1999). In contrast, in the channel receptors appear to trigger increases in systemic vascular catfish and the longnose gar, complete branchial denervation resistance during hypoxia in other teleosts (see Sundin et al., abolished all the ventilatory responses (Smatresk, 1989; 1999), and these or similar receptors may be involved here. Burleson and Smatresk, 1990a). As with the hypoxic stimulus, complete gill arch denervation Ventilatory responses: ventilation rate did not attenuate the increase in VAMP elicited by internal or Hypoxia is a powerful respiratory stimulant in all fish (for external NaCN injections. This suggests that there is a references, see Shelton et al., 1986; Fritsche and Nilsson, contribution to the increase in VAMP from extrabranchial 1993) and, as has been reported previously, tambaqui increase receptors that are both externally and internally oriented. − resting fV from approximately 35 to 55Ð66 breaths min 1 in Several potential locations have been suggested as the site for response to acute exposure to low oxygen tensions (Rantin and extrabranchial oxygen receptors. They may exist in the Kalinin, 1996). The denervation experiments in the present orobuccal cavity, innervated by cranial nerves V and VII study revealed that the receptors involved in producing this (Hughes and Shelton, 1962; Butler et al., 1977), or be located increase are solely branchial and are probably distributed on within the central nervous system (Satchell, 1961; Saunders all the gill arches. Both internal and external injections of and Sutterlin, 1971). Since all attempts to evoke a ventilatory NaCN were able to mimic hypoxia and rapidly increased fV response by central stimulation of the brain in vitro or in vivo by a comparable amount. On the basis of these results, we have failed (Rovainen, 1977; Kawasaki, 1980; Hedrick et al., conclude that the branchial O2-sensitive chemoreceptors 1991), support for the existence of central chemoreceptors in mediating the increase in fV monitor both the blood and the fish is weak. They may also be located in the heart or in the respiratory water. There is a striking similarity between the ventral aorta innervated by the cardiac and visceral branches distribution and location of the receptors triggering the of the vagus nerve since these branches were not sectioned in hypoxic bradycardia and the increase in fV in this species (see the present study (Smatresk et al., 1986). above), raising the possibility that the same receptor Studies on other fish species have typically found increases population mediates both reflex responses. This is not the case in ventilation amplitude in response to hypercarbia (Heisler et for most other species since, in most, the receptors responsible al., 1988; Wood et al., 1990; Kinkead and Perry, 1991; for producing the fall in heart rate are found only on the first Gilmour and Perry, 1994; Reid et al., 1999). In tambaqui, gill arch (see above). hypercarbia produced a variety of changes in ventilation Ventilation rate also increased significantly during amplitude in different individuals. The net result was that there hypercarbia in intact fish. This is consistent with other studies were no significant changes in mean ventilation amplitude in on a wide variety of fish (dogfish Scyliorhinus stellaris, Heisler response to hypercarbia in any group. In the control and G1 et al., 1988; skate Raja oscellata, Wood et al., 1990; rainbow fish, seven out of 11 and six out of 10 animals, respectively, trout, Kinkead and Perry, 1991; Gilmour and Perry, 1994; increased ventilation amplitude in each group. In several traira, Reid et al., 1999). The levels of CO2 required to produce animals (such as the one depicted in Fig. 4), the increases were these responses in tambaqui, as well as in traira, are substantial. After complete gill denervation, only one out of substantially greater than in more temperate fishes (4.0–5.3 kPa seven animals showed any notable increase in ventilation versus 0.7 kPa), and this may reflect the extremely acidic amplitude. Given this variability, it is difficult to draw firm waters (pH 3.5–5.0) in which these fish often live. The increase conclusions, but the data suggest that increases in ventilation in ventilation rate on exposure to hypercarbia was still present amplitude do occur in some animals and that, for the most part, in fish following denervation of the first gill arch, but not this appears to be a response to stimulation of branchial following total denervation of all gill arches. The receptors receptors. mediating the hypercarbic tachypnoea in the only other species studied so far, the traira (Reid et al., 1999), as well as the Concluding remarks receptors mediating the hypoxic tachypnoea in both the traira The results of the present study do not reveal a simple and the tambaqui (Sundin et al., 1999), are all confined to the picture of cardiorespiratory chemoreceptor control in gills. Thus, it is possible that the same receptors are involved tambaqui. The responses of this species to hypoxia and in mediating the increase in ventilation rate during both hypercarbia appear to involve several putative receptor hypercarbia and hypoxia in both species. populations, in different locations, that have different central projections producing different reflex motor outputs. The basis Ventilatory responses: ventilation amplitude for the intraspecies variability in the location and distribution Even complete branchial denervation failed to attenuate of branchial and extrabranchial chemoreceptors in fish that significantly the increase in VAMP induced by hypoxia, clearly have been studied remains elusive. More observational data of showing that activation of extrabranchial receptors alone is this sort will be needed, for a variety of species adapted to sufficient to produce this increase. These results agree with different habitats, before testable hypothesis concerning the earlier studies in tench (Tinca tinca), sea raven and traira in evolution, phylogeny and adaptive significance of differences 1238 L. SUNDIN AND OTHERS in the distribution of cardiorespiratory chemoreceptors can be Daxboeck, C. and Holeton, G. F. (1978). Oxygen receptors in the formulated. Finally, while there is some evidence to suggest rainbow trout, Salmo gairdneri. Can. J. Zool. 56, 1254Ð1259. that the receptors involved in producing the hypercarbic Fritsche, R. and Nilsson, S. (1989). Cardiovascular responses to bradycardia and increase in ventilation amplitude, in those hypoxia in the Atlantic cod, Gadus morhua. Exp. Biol. 48, animals that showed an increase in ventilation amplitude, were 153Ð160. Fritsche, R. and Nilsson, S. (1993). Cardiovascular and ventilatory different from the O2-sensitive receptors that elicited similar control during hypoxia. In Fish Ecophysiology (ed. J. C. Rankin changes in response to hypoxia, the distributions of receptors and F. B. Jensen), pp. 180Ð206. London, New York: Chapman & involved in producing increases in systemic vascular resistance Hall. and breathing frequency during both hypercarbia and hypoxia Gilmour, K. M. and Perry, S. F. (1994). The effects of hypoxia, were similar. Viewed subjectively, this evidence is not hyperoxia or hypercapnia on acidÐbase disequilibrium in the overwhelming because the differences appear to be more arterial blood of rainbow trout. J. Exp. Biol. 192, 269Ð284. quantitative than qualitative. Furthermore, even if there are Hedrick, M. S., Burleson, M. L., Jones, D. R. and Milsom, W. K. differences in distribution, the data do not necessarily imply (1991). An examination of central chemosensitivity in an air- that the hypercarbic responses arise from receptors that are breathing fish (Amia calva). J. Exp. Biol. 155, 165Ð174. insensitive to hypoxia, but that only some are also sensitive to Heisler, N., Toews, D. P. and Holeton, G. F. (1988). Regulation of hypercarbia and have a different distribution from those that ventilation and acidÐbase status in the elasmobranch Scyliorhinus are not. True resolution of this question will require single- stellaris during hyperoxia-induced hypercarbia. Respir. Physiol. 71, 227Ð246. fibre recordings from putative branchial receptors. Hughes, G. M. and Shelton, G. (1962). Respiratory mechanisms and their nervous control in fish. Adv. Comp. Physiol. Biochem. 1, This study was supported by NSERC of Canada scientific 275Ð364. and operating grants to W.K.M. and FAPESP (Fundação de Janssen, R. G. and Randall, D. J. (1975). The effects of changes in Amparo à Pesquisa do Estado de São Paulo), and CNPq (the pH and PCO∑ in blood and water on breathing in rainbow trout, Brazilian National Research Council for Development of Salmo gairdneri. Respir. Physiol. 25, 235Ð245. Sciences and Technology) grants to T.R. L.S. was the Kawasaki, R. (1980). Maintenance of respiratory rhythm-generation recipient of Swedish Foundation for International Cooperation by vascular perfusion with physiological saline in the isolated head in Research and Higher Education (STINT) and Isaak Walton of the carp. Jap. J. Physiol. 30, 575Ð589. Killam Foundation postdoctoral fellowships. Travel grants for Kinkead, R. and Perry, S. F. (1991). The effects of catecholamines L.S. were provided by the Wenner-Gren Center Foundation on ventilation in rainbow trout during hypoxia or hypercapnia. Respir. Physiol. 84, 77Ð92. and the Royal Society of Arts and Sciences in Göteborg. McKenzie, D. J., Burleson, M. L. and Randall, D. J. (1991). The S.G.R. was the recipient of an NSERC of Canada postdoctoral effects of branchial denervation and pseudobranch ablation on fellowship. cardio-ventilatory control in an air-breathing fish. J. Exp. Biol. 161, 347Ð365. Milsom, W. K. and Brill, R. W. (1986). Oxygen sensitive afferent References information arising from the first gill arch of yellowfin tuna. Respir. Braum, E. and Junk, W. (1982). Morphological adaptation of two Physiol. 66, 193Ð203. Amazonian characoids (Pisces) for surviving in oxygen deficient Nilsson, S. (1984). Innervation and pharmacology of the gills. In Fish waters. Int. Rev. Ges. Hydrobiol. 67, 869Ð886. Physiology, vol. XA (ed. W. S. Hoar and D. J. Randall), pp. Burleson, M. L. and Milsom, W. K. (1993). Sensory receptors 185Ð227. Orlando: Academic Press. in the first gill arch of rainbow trout. Respir. Physiol. 93, Perry, S. F. and Kinkead, R. (1989). The role of catecholamines in 97Ð110. regulating arterial oxygen content during acute hypercarbic Burleson, M. L. and Smatresk, N. J. (1990a). Effects of sectioning acidosis in rainbow trout (Salmo gairdneri). Respir. Physiol. 77, cranial nerves IX and X on cardiovascular and ventilatory reflex 365Ð378. responses to hypoxia and NaCN in channel catfish. J. Exp. Biol. Perry, S. F., Kinkead, R. and Fritsche, R. (1992). Are circulating 154, 407Ð420. catecholamines involved in the control of breathing by fishes? Rev. Burleson, M. L. and Smatresk, N. J. (1990b). Evidence for two Fish Biol. Fish. 2, 65Ð83. oxygen-sensitive chemoreceptor loci in channel catfish, Ictalurus Randall, D. J., Heisler, N. and Drees, F. (1976). Ventilatory punctatus. Physiol. Zool. 63, 208- 221. responses to hypercapnia in the larger spotted dogfish Scyliorhinus Burleson, M. L., Smatresk, N. J. and Milsom, W. K. (1992). stellaris. Am. J. Physiol. 230, 590Ð594. Afferent inputs associated with cardioventilatory control in fish. In Randall, D. J. and Jones, D. R. (1973). The effects of deafferentation Fish Physiology, vol. XIIB (ed. W. S. Hoar, D. J. Randall and A. of the pseudobranch on the respiratory response to hypoxia and P. Farrell), pp. 389Ð426. New York: Academic Press. hyperoxia in the trout (Salmo gairdneri). Respir. Physiol. 17, Butler, P. J. and Taylor, E. W. (1971). Response of the dogfish 291Ð302. (Scyliorhinus canicula L.) to slowly induced and rapidly induced Randall, D. J. and Smith, J. C. (1967). The regulation of cardiac hypoxia. Comp. Biochem. Physiol. 39A, 307Ð323. activity in fish in a hypoxic environment. J. Exp. Biol. 40, 104Ð113. Butler, P. J., Taylor, E. W. and Short, S. (1977). The effect of Randall, D. J. and Taylor, E. W. (1991). Evidence of a role for sectioning cranial nerves V, VII, IX and X on the cardiac response catecholamines in the control of breathing in fish. Rev. Fish Biol. of the dogfish Scyliorhinus canicula to environmental hypoxia. J. Fish. 1, 139Ð157. Exp. Biol. 69, 233Ð245. Rantin, F. T. and Kalinin, A. L. (1996). Cardiorespiratory function Branchial chemoreceptors in tambaqui 1239

and aquatic surface respiration in Colossoma macropomum evidence for two chemoreceptive loci. Am. J. Physiol. 251, exposed to graded and acute hypoxia. In Physiology and R116ÐR125. Biochemistry of the Fishes of the Amazon (ed. A. L. Val, V. M. Smith, F. M. and Davie, P. S. (1984). Effects of sectioning cranial F. de Almeida-Val and D. J. Randall), pp. 169Ð180. Manaus: nerves IX and X on the cardiac response to hypoxia in the coho INPA. salmon, Oncorhynchus kisutch. Can. J. Zool. 62, 766Ð768. Reid, S. G., Sundin, L., Kalinin, A. L., Rantin, F. T. and Milsom, Smith, F. M. and Jones, D. R. (1978). Localization of receptors W. K. (1999). Regulation of cardiorespiratory reflexes in the traira causing hypoxic bradycardia in trout (Salmo gairdneri). Can. J. (Hoplias malabaricus): The role of pH/CO2 receptors. Respir. Zool. 56, 1260Ð1265. Physiol. (in press). Smith, F. M. and Jones, D. R. (1982). The effect of changes in blood Rovainen, C. M. (1977). Neural control of ventilation in the lamprey. oxygen carrying capacity on ventilation volume in the rainbow Fedn. Proc. 36, 2386Ð2389. trout (Salmo gairdneri). J. Exp. Biol. 97, 325Ð334. Satchell, G. H. (1961). The response of the dogfish to anoxia. J. Exp. Sundin, L., Reid, S. G., Kalinin, A. L., Rantin, F. T. and Milsom, Biol. 38, 531Ð543. W. K. (1999). Cardiovascular and respiratory reflexes in the Saunders, R. L. and Sutterlin, A. M. (1971). Cardiac and respiratory tropical fish (Hoplias malabaricus). I. Role of gill O2 responses to hypoxia in the sea raven, Hemitripterus americanus chemoreceptors. Respir. Physiol. (in press). and an investigation of possible control mechanisms. J. Fish. Res. Thomas, S., Fievet, B., Barthelemy, L. and Peyraud, C. (1983). Bd Can. 28, 491Ð503. Comparison of the effects of exogenous and endogenous Shelton, G., Jones, D. R. and Milsom, W. K. (1986). Control of hypercapnia on ventilation and oxygen uptake in the rainbow trout breathing in ectothermic vertebrates. In Handbook of Physiology, (Salmo gairdneri R.). J. Comp. Physiol. B 151, 185Ð190. section 3, The Respiratory System, vol II, Control of Breathing, part Thomas, S. and Le Ruz, H. (1982). A continuous study of rapid 2 (ed. S. R. Geiger, A. P. Fishman, N. S. Cherniack and J. G. changes in blood acidÐbase status of trout during variations of Widdicombe), pp. 857Ð909. Baltimore: Waverly Press. water PCO∑. J. Comp. Physiol. B 148, 123Ð130. Smatresk, N. J. (1989). Chemoreflex control of respiration in an Val, A. L. and Almeida-Val, V. M. F. (1995). Fishes of the Amazon air-breathing fish. In Chemoreceptors and Chemoreflexes in and their Environment. Physiological and Biochemical Features. Breathing Ð Cellular and Molecular Aspects (ed. S. Lahiri, R. E. Heidelberg: Springer Verlag. Forster II, R. O. Davies and A. I. Pack), pp. 52Ð29. London: Wood, C. M., Turner, J. D., Munger, R. S. and Graham, M. S. Oxford University Press. (1990). Control of ventilation in the hypercarbic skate Raja Smatresk, N. J., Burleson, M. L. and Azizi, S. Q. (1986). oscellata. II. Cerebrospinal fluid and intracellular pH in the brain Chemoreflexive responses to hypoxia and NaCN in longnose gar: and other tissues. Respir. Physiol. 80, 279Ð298.