ABSTRACT
It is well established that arterial pH decreases with increased temperature in amphibians and reptiles through an elevation of arterial PCO2, but the underlying regulation remains controversial. The alphastat hypothesis ascribes the pH fall to a ventilatory regulation of protein ionisation, but the pH reduction with temperature is lower than predicted by the pKa change of the imidazole group on histidine. We hypothesised that arterial pH decreases at high, but not at low, temperatures when toads (Rhinella marina) and snakes (Python molurus) are exposed to hyperoxia. In toads, hyperoxia caused similar elevations of arterial PCO2 at 20 and 30°C, indicative of a temperature-independent oxygen-mediated drive to breathing, whereas PCO2 was unaffected by hyperoxia in snakes at 25 and 35°C. These findings do not support our hypothesis of an increased oxygen-mediated drive to breathing as body temperature increases.
INTRODUCTION
The acid–base status of arterial blood is influenced by body temperature in ectothermic vertebrates, and arterial pH generally decreases as body temperature increases (e.g. Reeves, 1972; Glass et al., 1985; Ultsch and Jackson, 1996; Burton, 2002; Wang and Jackson, 2016). In aquatic species, the reduction in arterial pH is associated with an alteration in plasma bicarbonate concentration ([HCO3−]) (Heisler, 1986; Ultsch and Jackson, 1996). In contrast, air-breathing vertebrates alter arterial pH by elevating the partial pressure of CO2 (PCO2) in the arterial blood by a reduction in pulmonary ventilation relative to the rate of CO2 production (Jackson et al., 1974; Cameron and Kormanik, 1982; Glass et al., 1983, 1985; Jackson, 1986; Boutilier et al., 1987; Amin-Naves et al., 2004). Although the influence of temperature on acid–base status has been documented in many different species, the underlying regulation remains controversial and unresolved, and there is even little agreement on which parameters are actually being regulated (Wang and Jackson, 2016; Milsom et al., 2022). In addition, some species such as monitor lizards do not appear to conform to this general pattern (e.g. Wood et al., 1981; Zena et al., 2016a).
As the most influential (and contentious) model, the alphastat theory proposes that animals regulate the ionisation of the imidazole moieties of histidine in proteins (Reeves, 1972). This proposal is appealing because such regulation would serve to preserve enzyme activities and protein conformations as well as stabilise ion and water distribution across cell membranes (Reeves, 1972; Burton, 2002). The alphastat hypothesis predicts that air-breathing vertebrates regulate arterial pH through ventilatory control of arterial PCO2 without alterations in total CO2 concentration (Wang and Jackson, 2016). However, the reduction in arterial pH with temperature (ΔpH/ΔT) is lower than predicted from the change in pKa of the imidazole group on histidine (Wood et al., 1981; Heisler, 1986; Ultsch and Jackson, 1996). This remains a strong argument against the alphastat hypothesis.
Air-breathing vertebrates ventilate their lungs to obtain O2 and to eliminate the CO2 that is produced by respiration, and lung ventilation is regulated by central chemoreceptors that are sensitive to CO2/pH (Branco and Wood, 1993; Branco et al., 1993; Santin et al., 2013; Zena et al., 2016b; Milsom et al., 2022). The peripheral chemoreceptors are typically stimulated when arterial partial pressure of O2 (PO2) is below that required for complete saturation of haemoglobin (Milsom and Wang, 2017; Milsom et al., 2022). The ventilatory responses to hypercapnia (high CO2) are very robust in air-breathing vertebrates, while the oxygen shortage typically needs to be rather severe before a ventilatory response is elicited (Dejours, 1988). Therefore, it is generally assumed that tetrapods regulate ventilation to maintain CO2 levels and acid–base balance (Milsom et al., 2022). Nevertheless, the ventilatory response to hypoxia is more vigorous when the temperature increases (Glass et al., 1985; Kruhøffer et al., 1987). Also, the ventilatory response to hypercapnia is much more pronounced as body temperature increases in both amphibians and reptiles (e.g. Branco et al., 1993; Branco and Wood, 1993; Bicego-Nahas and Branco, 1999; Klein et al., 2002; Milsom et al., 2022). Thus, given that temperature affects the ventilatory response to oxygen and CO2, we wondered whether there is a shift in the balance toward oxygen regulation as temperature increases, such that an increased oxygen-mediated drive to lung ventilation may explain why arterial pH does not decrease as much as predicted by the alphastat hypothesis.
We investigated this possibility by exposing two species of ectothermic air-breathing vertebrates to increased temperatures with and without hyperoxia. We predicted that animals would exhibit higher arterial PCO2 when exposed to hyperoxia, and this effect would be particularly increased at higher temperatures. Comparably, the pH reduction should be higher at higher temperatures if the animals are exposed to hyperoxia. Therefore, if there is no oxygen-related breathing drive, an air-breathing ectothermic animal will probably exhibit higher ΔpH/ΔT when exposed to hyperoxia.
MATERIALS AND METHODS
Experimental animals
We studied eight Rhinella marina (Linnaeus 1758) (312±37 g) and seven Python molurus (Linnaeus 1758) (265±18 g) of either sex. The cane toads were purchased from Exotic Tropicals Herpetoculture (Barbados, West Indies) and maintained in containers with free access to water and substrate for burrowing. The snakes were obtained from an animal supplier and maintained in individual compartments (72×40×20 cm) with free access to water. The toads were fed insects on a daily basis, and the snakes were fed dead mice on a weekly basis. Feeding was interrupted 48 h before instrumentation for the toads and at least a week before instrumentation for the snakes. These fasting periods were chosen to reduce the influence of digestion on metabolism and blood gases (Overgaard et al., 1999; Andersen and Wang, 2003). Both species were kept in a temperature-controlled system on a 12 h:12 h light:dark photoperiod. All experiments were performed in accordance with Danish Regulations for animal experimentation.
Animal instrumentation
The cane toads were anaesthetised by immersion in benzocaine (1 g l−1) until they lost the corneal reflex and the retraction reflex of the pelvic extremities. Pythons were anaesthetised in a sealed container with gauze soaked in isoflurane (IsoFlo Vet 100%; Orion Pharma Animal Health AS). When righting reflexes subsided, the trachea was intubated for mechanical ventilation with a recycling oxygen ventilator (Anesthesia Workstation, Hallowell EMC) and a vaporiser with 2% isoflurane (total volume of 200 ml of air kg−1 min−1 and 4 breaths min−1) (Fluotec Mark 3 vaporiser, Simonsen & Well A/S). Snakes and cane toads were instrumented with catheters (PE50) by occlusive cannulation of the vertebral artery and the femoral artery, respectively, allowing arterial blood samples to be collected without disturbance. Upon recovery from anaesthesia, each animal was placed in a temperature-controlled chamber (cane toads: 20 and 30°C; snakes: 25 and 35°C), and kept in these conditions for 12–24 h until blood samples were collected. The temperature exposure was randomised. After experiments, the animals were euthanised with an overdose of pentobarbital (100 mg kg−1 by intravascular infusion through the catheter).
Experimental protocol
Blood samples (0.3 ml) were collected into heparinised syringes and analysed immediately after sampling using a GEM Premier 3500 blood gas analyser (Instrumentation Laboratory, Bedford, MA, USA) to obtain arterial pH (pHa), arterial PCO2 (PaCO2) and arterial PO2(PaO2), as well as plasma Na+ and K+ concentration. We calculated plasma [HCO3−] by rearranging the Henderson–Hasselbalch equation. To characterise the oxygen drive for ventilation and its correlation to acid–base status, we exposed animals to normoxia and hyperoxia at both temperatures (100% O2 for 2 h). This high level of hyperoxia was chosen to ensure that PaO2 would be elevated despite the presence of a cardiac right-to-left shunt (Wood, 1984).
Statistics
The effect of temperature on the measured and derived parameters in normoxia and hyperoxia was tested using two-way repeated-measures ANOVA and Newman–Keuls post hoc test using GraphPad Prism 7. All data are presented as means±s.e.m. Differences were considered statistically significant when P<0.05.
RESULTS AND DISCUSSION
pHa decreased with increased temperature in both R. marina and P. molurus (Fig. 1). In both species, the pHa reduction was achieved primarily by a rise in PCO2 (Fig. 1), while the calculated plasma [HCO3−] did not change (Table 1). This pattern is consistent with findings in other air-breathing ectotherms where the reduction in pH with temperature is due to an elevation of PaCO2 at stable plasma [HCO3−] (Reeves, 1977; Boutilier et al., 1987; Amin-Naves et al., 2004). However, in some previous studies on R. marina (formerly known as Bufo marinus), [HCO3−] did change with temperature, but contributed little to the fall in pHa compared with the rise in PaCO2 (Boutilier et al., 1987; Stinner et al., 1994). With the exception of a small rise in plasma Na+ concentration in toads exposed to hyperoxia at 30°C, plasma ion levels (Na+, K+ and Ca2+) were not affected by either temperature changes or exposure to hyperoxia (Table 1). This is consistent with the alphastat hypothesis predicting that the stable protein ionisation avoids ion and water shifts, but cannot necessarily be interpreted as evidence in favour of the alphastat hypothesis. The ion levels resembled those found in previous studies on these species (e.g. Overgaard et al., 1999; Andersen and Wang, 2003).
In both species and at both temperatures, hyperoxia achieved the desired elevation of PaO2 without a change in haemoglobin concentration in the blood (Fig. 1, Table 1), but hyperoxia did not affect the hypothesised reduction in pHa with increased temperature. Thus, in the toads, ΔpH/ΔT was 0.014 and 0.013 units °C−1 in normoxia and hyperoxia, respectively, whereas ΔpH/ΔT decreased from 0.016 units °C−1 in the normoxic snakes to 0.008 units °C−1 in the hyperoxic snakes. The unaffected pH reduction with temperature in hyperoxic toads, and the smaller ΔpH/ΔT in the hyperoxic snakes are actually contrary to our hypothesis. Accordingly, our study reveals no indication of increased oxygen-mediated drive to breathing as body temperature increases in toads and pythons.
In the normoxic pythons, the rise in PaO2 with increased temperature is probably explained by the temperature-induced lowering of blood oxygen affinity that in the presence of cardiac right-to-left shunts elevates PaO2 (Wood, 1984). However, the small rise in arterial saturation (SO2) also contributed to this (Table 1), and the high SO2 values are consistent with the low capacity for cardiac shunting in pythons (Jensen et al., 2010). In the normoxic toads, we did not see a similar rise in PaO2 (Fig. 1), and this differs from previous studies on toads where PaO2 increased with temperature (Boutilier et al., 1987; Kruhøffer et al., 1987; Wang et al., 1998). As in the pythons, there was a high arterial SO2 at both temperatures, and it seems that toads in our study had a low right-to-left shunt in both normoxia and hyperoxia.
In the snakes, hyperoxia did not affect PaCO2, showing that lung ventilation relative to metabolism (i.e. the air convection requirement that determines PaCO2) was not affected. We, therefore, found no evidence for an oxygen-mediated drive to breathing at either temperature in the pythons. Previous studies on turtles, lizards, snakes and alligators show that hyperoxia depresses ventilation (Randall et al., 1944; Glass et al., 1978). Frankel et al. (1969) also observed that prolonged oxygen breathing depressed respiration and increased PaCO2 in Pseudemys scripta at 28°C. In lightly anaesthetised tortoises (Testudo horsfieldi) breathing room air, a single inhalation of pure oxygen led to transient reductions in ventilation at 35°C, but not at 25 and 30°C, indicating that the oxygen-mediated drive to breathing increased with temperature (Benchetrit et al., 1977). The differences between species may reflect genuine differences among the various clades of reptiles. However, it is also possible that the exposure regime matters, as single inhalations of hyperoxic gases may have more significant, albeit short-lasting, effects than the more prolonged exposures used in our present study (Benchetrit et al., 1977).
In amphibians, including toads, the regulation of arterial blood gases in response to altered body temperature is further complicated by the large degree of cutaneous gas exchange (Jackson, 1978). As metabolism increases with body temperature, the lungs become more important for both O2 uptake and CO2 excretion, and the rise in the respiratory exchange ratio over the lungs contributes to the reduction in pHa with increased temperature (Wang et al., 1998). Nevertheless, cutaneous CO2 excretion is a passive process determined by a rather temperature-insensitive CO2 conductance, and it is lung ventilation, controlled by the peripheral and central chemoreceptors, that regulates PaCO2. This means that reductions in ventilation will cause PaCO2 to increase and this effect will be particularly pronounced at a high temperature.
Hyperoxia caused significant elevations of PaCO2 at both temperatures in the cane toads (Fig. 1) and provided evidence for an oxygen-mediated drive to breathing in normoxia. Consistent with this finding, Zena et al. (2016a) reported decreased ventilation in a closely related species of toad (Rhinella schneideri) when exposed to 30% O2 at 25°C. The similar effects at 20 and 30°C in our study are not intuitive as the ventilatory responses to hypoxia are more pronounced at high temperatures and the rise in ventilation occurs at higher levels of inspired PaO2 (e.g. Kruhøffer et al., 1987). However, it is clear that the principal drive to lung ventilation is mediated by CO2, and the central chemoreception for CO2 seems to dominate over a wide range of temperatures in toads (Branco et al., 1993). In contrast, fish respond much less to hypercapnia and ventilation is clearly depressed by hyperoxia and hence associated with sizeable elevations of PaCO2 (e.g. Berschick et al., 1987; McArley et al., 2018).
In conclusion, our study reveals no indication of an increased oxygen-mediated drive to breathing as body temperature rises in toads and pythons. However, it confirms that CO2 is the most important determinant for ventilation, irrespective of temperature.
Acknowledgements
We would like to thank the Zoophysiology Department for hosting this project in their facilities. Special thanks to Rasmus Buchanan for technical assistance and Heidi Meldgaard Jensen for animal care. We are also grateful to Katja Bundgaard and Nadine Schmidbauer for their help with the cannulations.
Footnotes
Author contributions
Conceptualization: S.A.C., C.A.C.L., T.W.; Validation: T.W.; Formal analysis: S.A.C.; Investigation: S.A.C.; Resources: T.W.; Data curation: S.A.C.; Writing - original draft: S.A.C., C.A.C.L., T.W.; Writing - review & editing: S.A.C., C.A.C.L., T.W.; Visualization: S.A.C., T.W.; Supervision: C.A.C.L., T.W.; Project administration: C.A.C.L., T.W.; Funding acquisition: S.A.C., C.A.C.L., T.W.
Data availability
The data are available from the UFSCar repository: https://repositorio.ufscar.br/handle/ufscar/18835
Funding
This study was funded by Danmarks Frie Forskningsfond (Independent Research Fund of Denmark) and S.A.C. received support from The Company of Biologists (travelling fellowship JEBTFF 2208797) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq 142386/2020-0).
References
Competing interests
The authors declare no competing or financial interests.