Ventilatory Responses to Carboxyhaemoglobinaemia and Hypoxic Hypoxia in Bufo Paracnemis

Most early studies on pulmonary ventilation in amphibians focused on the mechanics of pulmonary ventilation: the mechanism behind the positive pressure inflation of the lungs (DeJongh and Gans, 1969; Jones, 1982). In addition, the periodic pattern of ventilation was described as breath-holding alternating with bursts of pulmonary ventilation, which in some anuran amphibians are initiated by stepwise pulmonary deflation followed by lung reinflation (Boutilier, 1984; Shelton and Boutilier, 1982).

Recently, studies have evaluated the chemical drive to breathe and the receptors involved. The ventilatory responses to hypoxia and change of temperature were assessed in Bufo paracnemis (Kruhøffer et al. 1987), and other studies characterized the arterial O₂ receptors (Ishii et al. 1985; Van Vliet and West, 1992). In addition, central acid–base receptors have been documented in Bufo marinus (Smatresk and Smits, 1991) and Bufo paracnemis (Branco et al. 1992, 1993). The central chemosensitivity of Bufo marinus depends on alterations of pH and/or CO₂ levels (Smatresk and Smits, 1991). The O₂ stimulus to breathe has also been evaluated in Bufo paracnemis (Wang et al. 1994), with the conclusion that pulmonary ventilation is modulated by changes of arterial and not by of the blood. The experimental approach used in that study (bleeding with Ringer substitution) can, however, be criticized since cardiovascular alterations occurred. To avoid this problem, the present study applies reductions of by formation of carboxyhaemoglobin instead of bleeding. The animals were exposed to CO hypoxia and/or to reductions of inspired ⁠. The effects on blood variables were measured, and the ventilatory responses recorded together with changes in arterial blood pressure and heart rate.

Toads, Bufo paracnemis (Lutz) (mass 160–510 g, N=27), were collected at the Campus of the University of São Paulo in Ribeirão Preto from November to January (summer season). The toads were maintained in containers with free access to water and a basking area (temperature 25–27 °C). Food was withheld during the 4 days preceding surgery.

Arterial cannulation was performed as described in detail by Boutilier et al. (1979). A PE 50 catheter was occlusively inserted into the right femoral artery, after which skin incisions were closed and the catheter secured. This procedure was performed within 15 min, during which ether anaesthesia was applied. For initial anaesthesia, the toad was placed in a closed box saturated with ether vapour. During surgery, ether was evaporated from cotton pads placed under the belly. The first indications of recovery were observed within a few minutes after removal of ether, and complete recovery was apparent after a few hours. Experiments were initiated 24 h after surgery.

Pulmonary ventilation was measured directly using a pneumotachographic method described by Glass et al. (1978). Briefly, a light-weight face mask was constructed for each toad. Initially, an alginate resin was used to obtain a negative impression of the head, and then a positive plaster cast was made for moulding the masks (Bioplast, Scheu-Dental, Iserlohn, Germany). The masks were placed on the toads and provided an airtight fit between the nostrils and a Fleisch tube. A direct relationship exists between laminar air flow and pressure difference across this tube (Fleisch, 1925; Grenvik et al. 1966). The latter was monitored by a highly sensitive differential air pressure transducer (Statham, 12123) connected to a Narco four-channel recorder (Narco Bio-systems, Inc., Houston, Texas, USA; type 7179; model 4-A). Calibration was performed according to recommended routines by passing known volumes of gas through the pneumotachograph (for details, see Hobbes, 1967).

The arterial cannula was connected to a Statham pressure transducer (P23Dd) kept at the same level as the animal’s heart. A mercury column was used to calibrate the signals of this transducer before and after the experiment.

Immediately after withdrawal, arterial blood samples were analyzed for and pH using a FAC 204A O₂ analyzer (FAC Instruments, São Carlos, Brazil) and a Micronal B374 pH meter (São Paulo, Brazil). Electrodes were kept at the experimental temperature. Oxygen electrodes were calibrated with pure N₂ and humidified atmospheric air. The pH electrode was adjusted using high-precision buffers: S1500 and S1510; Radiometer, Copenhagen, Denmark (see Siggaard-Andersen, 1976, for exact composition and calibration values at different temperatures). Arterial O₂ content, ⁠, was measured according to the micromethod of Tucker (1967). Blood was estimated using the Astrup technique (Astrup, 1956). This involves equilibration of blood samples with gas mixtures of known ⁠, after which the relationship is plotted for each animal. The of the blood samples can then be interpolated from this relationship. Standard calibration gases (AGA, Brazil) with fractional CO₂ concentrations of 0.01, 0.03 and 0.06 were used for equilibration.

The fully recovered toad was equipped with a face mask and left undisturbed for 2 or 3 h in a chamber (volume 2 l) supplied with a continuous flux of humidified air (5 l min⁻¹). After this control period, hypoxic gas mixtures (fractional concentrations, FO₂, of 0.10 and 0.05) were prepared by mixing pure nitrogen with air. Relative flow rates (air and N₂) were controlled by means of precision needle valves. After humidification, the resulting mixture entered the chamber. Its concentration was continuously monitored using a Beckman OM-11 oxygen analyzer. All exposures to test gases were applied for 45 min, at the end of which ventilation was recorded and arterial blood withdrawn. Subsequently, known quantities of pure CO were mixed into the animal chamber to provide initial fractional concentrations of 0.001 or 0.01. To achieve this, 2 or 20 ml of air, respectively, was withdrawn from the 2 l animal chamber and replaced with pure CO. The resulting CO exposure was maintained for 30 min. After this treatment, the chamber was flushed with fresh air for 45 min. Then the series of hypoxic exposures was repeated along with measurements of ventilation and analysis of blood samples. In some experiments, inspiration of FCO=0.01 was accompanied by bicarbonate infusion to keep pH values constant. This was accomplished by infusion of NaHCO₃ at 1 mmol kg⁻¹ body mass into the dorsal lymph sac. This quantity was previously determined using stepwise NaHCO₃ infusion, accompanied by blood sampling. For more details concerning compensation of pH, see Siggaard-Andersen (1976).

Pulmonary ventilation was distinguished from buccal movements by a biphasic flow profile during expiration (Jones, 1982) and ventilation was calculated excluding buccal movements. Ventilation, blood pressure and heart rate were calculated on the basis of 10 min recording periods. Tidal volume was obtained from the integrated area of the expired flow signal. Effects of hypoxic hypoxia and CO treatment were evaluated using analysis of variance (ANOVA) and the difference between means was assessed by Tuckey’s test. A P value of less than 0.05 was considered significant.

Table 1 shows the data for the effects of hypoxic hypoxia and/or CO hypoxia on of the arterial blood. Hypoxic hypoxia down to FO₂=0.05 reduced of the blood by about 50 %. CO hypoxia decreased by the same amount when FCO=0.01 was inspired. By combining the treatments of hypoxic hypoxia (inspired FO₂=0.05) and CO hypoxia (FCO=0.01), declined to 30 % of the normoxic control value.

Table 2 shows the effects of treatments on blood pressure, heart rate and blood gases. None of the experimental conditions had any significant effect on mean arterial pressure. Heart rate rose during hypoxic hypoxia, but was not affected by CO hypoxia. Hypoxic hypoxia (FO₂=0.05) reduced to 30 % of the normoxic control value. Carboxyhaemoglobinaemia did not affect ⁠. Arterial pH increased during hypoxic hypoxia as a result of hyperventilation (Fig. 1). Treatment with FCO=0.01 caused a reduction of pHa. In some experiments, this reduction was compensated by bicarbonate infusion. Arterial was not influenced by inspiration of FCO=0.01.

Fig. 1 shows the effects of hypoxic and CO hypoxia on pulmonary ventilation. Hypoxic hypoxia was accompanied by marked increases of ventilation. Inspiration of CO at low levels (FCO=0.001) did not affect ventilation which, however, increased slightly at the higher level (FCO=0.01). This increase did not occur when arterial pH was controlled through bicarbonate infusion.

Studies on the carotid bodies of mammals indicate that O₂ partial pressure is the specific stimulus modulator for the O₂ receptor (Mills and Edwards, 1968; Garland et al. 1994). The transducer mechanism may involve cytochrome (Larihi, 1994). Participation of O₂-sensitive K⁺ channels is also suggested (Lópes-Barneo et al. 1993; Lópes-Lópes et al. 1989). A complete and consistent picture is, however, not yet available in spite of considerable recent progress (see Lahiri, 1994). The final stimulus modulator may be ⁠, but receptor function depends on blood flow and oxygen delivery to the receptor site. If blood flow is relatively low, then the output from the receptor becomes dependent on ⁠, which has been documented for the aortic O₂ receptors of the cat (Lahiri et al. 1981). Conversely, carotid O₂ receptors are heavily perfused and their output strongly depends on O₂ partial pressure; they are largely unaffected by carboxyhaemoglobinaemia (Lahiri et al. 1981). This implies that the final O₂ modulator is masked unless the receptor site is heavily perfused and, thus, perfusion characteristics determine the apparent arterial O₂ stimulus (⁠ or [O₂]).

In vertebrates, the apparent O₂ stimulus to breathe has been most widely studied in mammals and, according to the traditional view, their O₂ drive to breathe depends on arterial O₂ partial pressure rather than on (cf. Comroe, 1974). In a recent study, Garland et al. (1994) reported a considerable ventilatory response to carboxyhaemoglobinaemia in squirrels and rats. The ventilatory responses to hypoxic hypoxia seem to exceed those to CO hypoxia in both species. Increases of ventilation in response to CO hypoxia have also been reported for cats and goats (Santiago and Edelman, 1976; Gautier and Bonora, 1983; Gautier et al. 1990). It is not yet possible to relate these responses to the relative contributions from carotid and aortic receptor groups.

Information is scarce for other vertebrate classes. Positive evidence for a dependence of ventilation on has been obtained for birds (Tschorn and Fedde, 1974) and teleost fish (Smith and Jones, 1982). Corresponding data are few for reptiles and amphibians. Studying the toad Bufo paracnemis, Wang et al. (1994) reduced by replacing blood with Ringer’s solution. Ventilation is unaffected by this treatment, whereas decreases of arterial stimulate ventilation. This anaemia is, however, accompanied by tachycardia and probably also by viscosity changes, and this complicates interpretation of the results. An increased heart rate may partially restore O₂ delivery to receptor sites (Lahiri et al. 1981). In the present study, we reduced by applying CO hypoxia and thus avoided cardiovascular side effects. Moreover, we controlled acid–base status through bicarbonate infusion. Because of these precautions, the reduction of was not accompanied by significant stimulation of ventilation. It therefore seems justifiable to conclude that the O₂ drive in Bufo paracnemis is, indeed, linked to changes of O₂ partial pressure and is largely independent of ⁠. The possibility that very severe reductions of may, eventually, cause ventilatory responses cannot be excluded. Furthermore, it should be noted that recent work suggests the possibility that CO is an atypical neurotransmitter (Snyder, 1992). Such a role for CO may complicate the interpretation of data obtained by application of CO hypoxia (Garland et al. 1994). Carbon monoxide has been used as a probe to study in vitro the sensor mechanism of mammalian O₂ receptors (Lahiri, 1994). The receptors were excited when very high levels of CO were applied (6.67 kPa=500 mmHg). These findings cannot be applied directly to the present study because of differences in experimental protocols and levels of CO at the receptor site.

An increase of heart rate during hypoxic hypoxia was reported for Bufo marinus (Boutilier and Toews, 1977). The increase was, however, marginal, and severe hypoxia caused a pronounced bradycardia relative to normoxic control values. Recently, Wang et al. (1994) observed a progressive increase of heart rate with reduction of inspired ⁠, which is consistent with the present results. The present blood gas data agree well with previous studies on Bufo paracnemis (Kruhøffer et al. 1987; Branco et al. 1993; Wang et al. 1994) and Bufo marinus (Boutilier et al. 1987; Wood and Malvin, 1991). The observation that was not altered by CO hypoxia indicates that the reduction of pH during treatment with FCO=0.01 was due to a metabolic acidosis. Our ventilation data are also in close agreement with earlier measurements on Bufo paracnemis at 25 °C (Kruhøffer et al. 1987; Branco et al. 1992; Wang et al. 1994).

A number of studies report on peripheral O₂ receptors in anuran amphibians. Ishii et al. (1966, 1985) located such receptors in the carotid labyrinth as well as in the aortic arch. Recently, Van Vliet and West (1992) managed to record the afferent neural activity from these regions, applying reductions of as well as alterations of ⁠. They concluded that the output is related to changes of ⁠, which is consistent with the data on the ventilatory drive (Wang et al. 1994; present study). This suggests that the O₂ receptors that dominate the hypoxic response are not situated in O₂-demanding tissue but instead in a well-perfused region.

Anuran amphibians seem to share some essential features of ventilatory control with mammals: in both vertebrate classes, the O₂ drive to breathe depends on peripheral receptors. Moreover, ventilatory acid–base regulation is based on central chemoreceptors, backed up by a peripheral component (Smatresk and Smits, 1991; Branco et al. 1992). These common characteristics seem unexpected and are intriguing, considering the large evolutionary distance between the groups, the very different mechanism of pulmonary ventilation employed by the two groups and the high dependence on cutaneous respiration in amphibians.

This study was supported by an equipment grant (94/1239-4) from FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo). CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) provided research support to the authors (521995/94-9 and 520769/93-7). We also thank Dr Geraldo M. Campos for advice on statistics and FAEPA (Fundação de Apoio ao Ensino, Pesquisa e Assistência, do Hospital das Clínicas, da Faculdade de Medicina de Ribeirão Preto/USP) for financial support.