The aim of this study was to determine the effects of temperature upon pH, protein charge and acid–base-relevant ion exchange in air-breathing ectotherms. Plasma and skeletal muscles in cane toads (Bufo marinus) and bullfrogs (Rana catesbeiana) were examined at 30, 20 and 10 °C. In addition, skeletal muscle ion concentrations were examined in black racer snakes (Coluber constrictor) at 30 and 10 °C. Cooling the amphibians produced a reduction in most of the plasma ion concentrations (Na+, K+, Ca2+, Cl, SO42−) and in protein concentration because of increased hydration. Between 30 and 10 °C, total plasma osmolality fell by 14 % in the toads and by 5 % in the frogs. Plasma protein charge, calculated using the principle of electroneutrality, was unaffected by temperature, except possibly for the toads at 10 °C. The in vivo skeletal muscle ΔpHi/ΔT ratio, where pHi is intracellular pH and T is temperature, between 30 and 20 °C averaged −0.014 °C−1 in the toads and −0.019 °C−1 in the frogs. Between 20 and 10 °C, there was no change in pHi in the toads and a −0.005 °C−1 change in the frogs. The in vitro skeletal muscle ΔpHi/ΔT averaged −0.011 °C−1 in both toads and frogs. In all three species, skeletal muscle inulin space declined with cooling. Intracellular ion concentrations were calculated by subtracting extracellular fluid ion concentrations from whole-muscle ion concentrations. In general, temperature had a large effect upon intracellular ion concentrations (Na+, K+, Cl) and intracellular CO2 levels. The relevance of the changes in intracellular ion concentration to skeletal muscle acid–base status and protein charge and the possible mechanisms producing the adjustments in intracellular ion concentration are discussed. It is concluded that ion-exchange mechanisms make an important contribution to adjusting pH with changes in temperature.

Studies of acid–base status as a function of temperature in ectotherms have been of interest since the seminal work by Stadie et al. (1925) and Austin et al. (1927). These early studies established that, in sealed blood samples, the principal factor determining the change in plasma pH (pHpl) with changing temperature (ΔpHplT) is the change in ionization constant of the plasma proteins (ΔpKproT). In addition, from the buffer equation of Van Slyke, it was recognized that as long as an animal maintained a constant pHpl−pKpro with changes in body temperature, then protein charge was conserved. Austin et al. (1927) sampled blood from alligators at 9 and 35 °C and concluded that the animals regulated pH in vivo to maintain a constant plasma protein charge. Subsequent studies extended this early work to other reptiles, amphibians, fish and invertebrates (for reviews, see Heisler, 1986b; Reeves, 1977). Reeves (1972) focused attention upon the imidazole of histidine as the principal ionizing group of proteins.

The same model of constant protein charge and variable Ph was first applied to the intracellular compartment by Malan et al. (1976). They concluded that intracellular pH (pHi) was regulated to preserve a constant pHi−pKpro over a wide temperature range just as in the extracellular fluid. By preserving protein charge over a wide temperature range, ectotherms supposedly prevent changes in enzyme activity as well as shifts in ions and water. However, as more studies of individual species became available, it was clear that there are considerable differences in the magnitude of the ΔpH/ΔT relationship among species, between tissues within an individual animal and over the temperature range of an animal (Whiteley et al., 1995; Lutz et al., 1989; Rocha and Branco, 1998; Butler and Day, 1993; Lehoux and Guderley, 1997; Boutilier et al., 1987; Stinner and Wardle, 1988). While some recent studies conclude that protein (i.e. imidazole) charge is independent of temperature (Van Dijk et al., 1997; Hitzig et al., 1994; Rodeau, 1984; Graber et al., 1992; Gaillard and Malan, 1985; Johnson et al., 1993; Douse and Mitchell, 1991;

Ultsch and Jackson, 1996), no unifying theory of acid–base regulation in animals has yet been formulated.

Another concept, in addition to regulating pH to maintain a constant protein charge, emerging from the studies of temperature and acid–base status in ectotherms is that water-breathers use primarily ion-exchange mechanisms to adjust pH whereas air-breathers use ventilation to adjust pH (Reeves, 1977; Cameron, 1984). The low solubility of oxygen in water relative to air imposes constraints on reductions in ventilation in water-breathers. In air-breathers, the decrease in pH that occurs with an increase in temperature is associated with substantial reductions in minute ventilation relative to metabolic rate . This allows for the rise in which is caused by passive temperature-induced decreases in the solubility coefficient and pK of the CO2/HCO3 buffer system, with virtually no change in CO2 content. In water-breathers such as fish, however, is relatively independent of temperature so that, with transepithelial and transmembrane ion exchange, is nearly constant and [HCO3] varies inversely with temperature. The putative advantage of using to adjust pH in air-breathers is that it can occur almost instantaneously with the change in body temperature, whereas ion-exchange mechanisms take many hours to reach a new steady-state pH. Hence, it would appear that air-breathers are better able to avoid disruption of pH-dependent biochemical homeostasis, including enzyme activity, with changes in temperature.

However, an exception to this view of air-breathers rapidly adjusting acid–base status with changes in body temperature is the black racer snake Coluber constrictor (Stinner et al., 1996; Stinner and Wardle, 1988). When the snakes were cooled from 30 to 10 °C and from 30 to 5 °C, a steady-state plasma pH was not reached for approximately 35 h and 60 h respectively. There were also gradual increases in [CO2]pl. In a more recent study, Stinner et al. (1998) concluded that these slow acid–base changes resulted from a gradual decline in with concomitant CO2 retention. CO2 retention also occurred in skeletal muscle, and the ΔpHi/ΔT ratio of skeletal muscle was consequently much lower than the estimated ΔpKproT ratio. Stinner et al. (1998) found no evidence for strong ion shifts, and hence ion-exchange mechanisms, adjusting pHpl with the change in temperature in the snakes. The authors also concluded that acid–base-relevant ion exchange did not occur in the muscle cells, but this was based upon measurements of strong ion levels in whole muscle rather than in intracellular fluid. The possibility of temperature affecting skeletal muscle extracellular fluid space, as observed in the cane toad Bufo marinus (Stinner et al., 1994), weakened this conclusion.

In their study of B. marinus and the African clawed frog Xenopis laevis, Boutilier et al. (1987) reported large differences in the ΔpHi/ΔT relationship among skeletal muscles and concluded that ion-exchange mechanisms must be operating. However, they did not examine muscle cell strong ion concentrations. It is possible that their findings resulted from differences in passive ΔpKproT resulting from different protein compositions among skeletal muscles (Cameron,

1989). The purpose of the present study was to determine acid–base-relevant intracellular ion concentrations and protein charge as a function of temperature in C. constrictor, B. marinus and the bullfrog R. catesbeiana. We used the principle of electroneutrality (the sum of the cations in mequivalents equals the sum of the anions in mequivalents) to estimate plasma protein charge and compared this with the in vivo and in vitro ΔpHplT relationships. We also examined skeletal muscle intracellular ion concentrations and their relationship to the in vivo and in vitro ΔpHi/ΔT ratios. The two amphibian species played prominent roles in the development of the commonly held view that air-breathing ectotherms maintain a constant protein charge over a wide temperature range by p.assiv.e shifts in the buffer equilibria and active adjustment of (Reeves, 1972, 1977).

Animals

Cane toads (Bufo marinus) (mean mass 177.1±61.6 g, N=181), bullfrogs (Rana catesbeiana) (mean mass 307.4±97.9 g, N=136) and black racer snakes (Coluber constrictor) (mean mass 212.9±56.7 g, N=20) were purchased from commercial suppliers. They were housed in the animal facility at The University of Akron, where room temperature was maintained at 25–30 °C. The cages for the cane toads and black racer snakes contained light bulbs at one end that provided basking sites at 35–40 °C. The lights were set to a 12 h:12 h L:D cycle. Approximately once weekly, the animals were fed on small rats. The animals were starved for 1 week prior to experimentation to avoid acid–base and electrolyte changes associated with the digestion and absorption of food (Coulson and Hernandez, 1964).

Surgery, experimental apparatus and blood analyses

The sciatic artery of cane toads and bullfrogs was occlusively cannulated (PE 50). The toads were anesthetized by exposure in a sealed chamber to cotton soaked in isoflurane. The bullfrogs were anesthetized in a 1 % solution of MS-222 buffered to a pH 7.3–7.4 by addition of 25 ml l−1 of 0.5 mol l−1 Na2HPO4. After a post-operative recovery period of at least 24 h, the toads and frogs were placed inside individual chambers within a darkened Precision (model 815) temperature-controlled cabinet (±0.5 °C). The individual animal chambers measured 32 cm×22 cm×17 cm and were made of an opaque plastic. Each chamber contained 1–1.5 l of water and was tipped at an angle to provide a dry area.

Pumps inside the cabinets continuously circulated water-saturated cabinet air through the animal chambers. At 20 and 30 °C, the chamber water was siphoned off and replaced with fresh temperature-equilibrated tapwater at least once every 48 h. For blood sampling purposes, the trailing end of the cannula was passed through a hole in the chamber lid and tied to a stick which prevented the end of the cannula from being pulled inside the chamber. The cannulas were flushed once every 2–3 days with heparinized (20 i.u. ml−1) amphibian Ringer’s solution (2.15 mmol l−1 KCl, 2.0 mmol l−1 CaCl2.2H2O, 68.1 mmol l−1 NaCl, 3.89 mmol l−1 Na2HPO4, 15.0 mmol l−1 NaHCO3, 1.2 mmol l−1 MgSO4.7H2O).

Prior to opening the temperature cabinet door for blood sampling, the laboratory lights were turned off to minimize disturbance to the animals. For pH measurements, blood was sampled directly into heparinized microcapillary hematocrit tubes, which were then quickly immersed in iced water. pH was measured using temperature-equilibrated Radiometer BMS-3 electrodes, as described previously (Stinner et al., 1998). The blood samples were centrifuged at 15 000 g for 5 min in a microcapillary centrifuge, and each plasma sample was then measured at 10 and 30 °C for determination of the in vitro ΔpHplT relationship. For ion and water analyses, 0.5 ml of blood was collected and immediately centrifuged at 2000 g for 5 min in an Eppendorf microcentrifuge. The plasma was sealed in a microcentrifuge tube and frozen until analysis. [Na+], [K+], [Cl], [Ca2+], [Mg2+], [inorganic phosphate], [protein], [SO42−] and osmolality were determined as described previously (Stinner et al., 1998). Water content was determined by drying weighed plasma samples (100–200 μl) overnight at 60 °C. Animals were sampled for blood at one temperature only. Animals were kept for at least 48 h at 30 °C, 72 h at 20 °C and 96 h at 10 °C. None was kept for more than 1 week in the experimental apparatus before blood sampling.

Skeletal muscle

The experimental arrangement and temperature exposure times were the same as described above. Animals were quickly decapitated, and [CO2] was then determined on muscle samples weighing 0.18–0.22 g, after overnight soaking in 0.1 mol l−1 NaOH using a Capni-con 3a CO2 analyzer, as described previously (Stinner et al., 1994, 1998). Each muscle was sampled twice, and each sample was measured in duplicate for [CO2]. To minimize the time between decapitation and sealing the samples in vials, no more than five muscles were sampled from each animal. Muscle [CO2] values for B. marinus at 10 °C were taken from an earlier study (Stinner et al., 1994). For [Na+], [K+] and [Cl] determinations, additional muscle samples weighing approximately 0.25 g were harvested and soaked in 4 ml of 0.1 mol l−1 HNO3 for 24 h with gentle shaking (Boutilier et al., 1986). Each muscle was sampled twice. [Na+] and [K+] were measured in duplicate using flame photometry (Coleman model 51; Bacharach, Inc., Pittsburgh, PA, USA). [Cl] was measured (4–6 replicates) using a Digital chloridometer (Buchler Instruments, Lenexa, KS, USA). For water contents, duplicate muscle samples weighing 0.2–0.4 g were dried overnight to a constant mass at 60 °C. In those cases where previously cannulated animals were used, care was taken to sample muscles from the uncannulated leg.

Skeletal muscle pHi was determined using the homogenate technique of Pörtner et al. (1990). Details of this procedure in our laboratory have been described previously by Stinner et al. (1998). Prior to muscle sampling, the toads were anesthetized by exposure to cotton soaked in isoflurane after siphoning off the chamber water and turning the air pump off. The bullfrogs were anesthetized by siphoning off the chamber water and replacing it with 1 l of temperature-equilibrated 1 % MS-222 solution buffered to pH 7.3–7.4. Anesthesia took 15–20 min. No more than five muscles were sampled from each animal. Each muscle was sampled twice, and each sample was used to prepare two tubes of muscle powder plus inhibitor solution. For animals held at 10 °C, the supernatant pH of one tube was measured at 10 °C, and the supernatant pH of the other tube was measured at 30 °C. The difference in pH was used to determine the passive component of the temperature-dependent adjustment of pHi (i.e. ΔpKproT). For animals held at 20 and 30 °C, both samples taken from the same ground muscle were measured at the animal’s body temperature and the values were averaged.

Extracellular fluid volume was determined using [3H]inulin.

The sciatic artery of the toads and frogs was occlusively cannulated. A posterior branch of the dorsal aorta of the black racer snakes was occlusively cannulated as described previously (Stinner et al., 1996). After a recovery period of at least 24 h, followed by exposure to 10, 20 or 30 °C (see above), 148 kBq 100 g−1 of [3H]inulin followed by approximately 0.5 ml of amphibian Ringer’s solution or reptilian Ringer’s solution (3.8 mmol l−1 KCl, 2.9 mmol l−1 CaCl2.2H2O, 128.0 mmol l−1 NaCl, 1.55 mmol l−1 Na2HPO4, 24.0 mmol l−1 NaHCO3, 1.2 mmol l−1 MgSO4.7H2O) were injected through the cannula. After 12–15 h, the toads and snakes were anesthetized with cotton soaked in isoflurane and the bullfrogs were anesthetized with MS-222 (see above). An incision was made through the ventral body wall, and 1.0 ml of blood was taken by heart puncture for six plasma sample (approximately 0.1 g) isotope analyses by liquid scintillation counting (Beckman model LS-6500, Beckman Instruments, Fullerton, CA, USA). Immediately after blood collection, five samples (approximately 0.2 g) from each skeletal muscle were placed inside scintillation vials and dissolved in 2 ml of Solvable at 60 °C, with occasional swirling, for approximately 4 h (Packard Instrument Co., Meriden, CT, USA). The dissolved muscle samples were allowed to cool to room temperature, and 10 ml of Ultima Gold (Packard Instrument Co.) was added for determination of radioactivity in the liquid scintillation counter. Inulin space and plasma concentrations of ions and water were used to correct whole-muscle ion and water concentrations to intracellular values (Stinner et al., 1994).

Results are presented as means ± S.D. (N).

The body temperatures of the cane toads and bullfrogs, measured with a Schulteis quick-reading thermometer inserted into the cloaca, were 10.1±0.8 °C (N=44), 20.0±0.8 °C (N=62) and 30.2±1.0 °C (N=57). The body temperature of the black racer snakes was assumed to be the same as the cabinet temperature.

Plasma

In vitro ΔpHplT measured in 10 B. marinus was −0.014±0.002 °C−1. The in vitro ΔpHplT coefficient varied with plasma protein concentration according to the linear regression ΔpHplT=−7.95×10−3−2.43×10−4[protein], where protein concentration is in g l−1 (P<0.01). Cooling caused dilution of the plasma in the cane toads (Table 1). Although not significantly different, water content was lowest at 30 °C, and plasma osmolality fell significantly by about 7 % for each 10 °C drop in temperature. This dilution is evident for most of the individual ions (Na+, K+, Ca2+, Cl, SO42−) and for protein. Concentrations of three ions did not decrease with a reduction in temperature, Mg2+, HCO3 and phosphate. Plasma [HCO3] was estimated by subtracting dissolved CO2 (i.e. α ) from total [CO2]. Total [CO2] and were taken from Stinner et al. (1994) and Howell et al. (1970), respectively, and α was calculated using the equation of Boutilier (Heisler, 1986a). Protein charge was calculated from the principle of electroneutrality as (the sum of the concentrations of Na+, K+, 2×Ca2+ and 2×Mg2+) minus (the sum of the concentrations of Cl, lactate, 2×SO42−, 2×HPO42− and H2PO4). [Lactate] was taken from Stinner et al. (1994). [HPO42−] and [H2PO4] were estimated from the Henderson–Hasselbalch equation, using pH values from Howell et al. (1970) and a pK of 6.791 at 30 °C (Siggaard-anderson, 1974). The van’t Hoff equation, pK−6.791=(ΔH°/4.576)[(1/T)−(1/310)], was used to correct pK to 30, 20 and 10 °C, where ΔH° (the enthalpy change) is 982 cal mol−1 (Edsall and Wyman, 1958). Plasma protein charge was the same (−0.31 mequiv g−1 protein) at 30 and 20 °C, but was higher (−0.37 mequiv g−1 protein) at 10 °C. This increase in protein charge was the result of a 7 % decrease in protein concentration between 20 and 10 °C, while the nonprotein anion gap (defined as the sum of the cations in mequivalents minus the sum of the nonprotein anions in mequivalents) remained unchanged. Simple dilution of the plasma should have decreased the nonprotein anion gap. However, the decline in the nonprotein anion concentration was relatively large (10.7 %), exceeding the decrease in cation concentration, suggesting a net active removal of anions (e.g. Cl) from the extracellular fluid.

Table 1.

Systemic arterial plasma electrolyte levels and water content in the cane toad Bufo marinus

Systemic arterial plasma electrolyte levels and water content in the cane toad Bufo marinus
Systemic arterial plasma electrolyte levels and water content in the cane toad Bufo marinus

In the bullfrogs, there was little or no difference in plasma ion concentrations and osmolality at 30 and 20 °C (Table 2). However, the values were approximately 5 % lower at 10 °C, indicative of some dilution by water, although [Mg2+] and [phosphate] were significantly higher at 10 °C. [HCO3] was calculated, as described above, using total [CO2] and taken from Mackenzie and Jackson (1978). Plasma protein charge, calculated as described above using a lactate concentration of 0.8 mmol l−1 (Lindinger and McDonald, 1986), was unaffected by temperature.

Table 2.

Systemic arterial plasma electrolyte levels in the bullfrog Rana catesbeiana

Systemic arterial plasma electrolyte levels in the bullfrog Rana catesbeiana
Systemic arterial plasma electrolyte levels in the bullfrog Rana catesbeiana

Skeletal muscle

Skeletal muscle cell pH (pHi) in the cane toads and bullfrogs followed a similar pattern with changes in temperature. Between 30 and 20 °C, there was a relatively large increase in pHi (−0.014 °C−1 in the cane toad and −0.019 °C−1 in the bullfrog). Between 20 and 10 °C, the pH values did not change significantly in the cane toad and only increased by approximately 0.005 °C−1 in the bullfrog. There were no significant differences in the in vitro ΔpHi/ΔT ratios among the muscles, which averaged −0.011 °C−1 between 30 and 10 °C (Tables 3, 4).

Table 3.

Skeletal muscle pH and in vitro ΔpH/ΔT in the cane toad Bufo marinus

Skeletal muscle pH and in vitro ΔpH/ΔT in the cane toad Bufo marinus
Skeletal muscle pH and in vitro ΔpH/ΔT in the cane toad Bufo marinus
Table 4.

Skeletal muscle pH and in vitro ΔpH/ΔT in the bullfrog Rana catesbeiana

Skeletal muscle pH and in vitro ΔpH/ΔT in the bullfrog Rana catesbeiana
Skeletal muscle pH and in vitro ΔpH/ΔT in the bullfrog Rana catesbeiana

In the toads and frogs, inulin space was largest in the submaxillary, followed by the obliques and rectus abdominis, then the sartorius, and finally the triceps femoris, gracilis major, semimembranosus and the gastrocnemius muscle, for which values were all similar (Tables 5, 6). In general, inulin space was larger in the cane toads than in the bullfrogs, especially in the hindlimb muscles where the inulin space was approximately twice as great in the toads. For all three species, there was a general trend for inulin space to decrease with decreasing temperature (Tables 57).

Table 5.

Skeletal muscle inulin space in the cane toad Bufo marinus

Skeletal muscle inulin space in the cane toad Bufo marinus
Skeletal muscle inulin space in the cane toad Bufo marinus
Table 6.

Skeletal muscle inulin space in the bullfrog Rana catesbeiana

Skeletal muscle inulin space in the bullfrog Rana catesbeiana
Skeletal muscle inulin space in the bullfrog Rana catesbeiana
Table 7.

Skeletal muscle inulin space in the black racer snake Coluber constrictor

Skeletal muscle inulin space in the black racer snake Coluber constrictor
Skeletal muscle inulin space in the black racer snake Coluber constrictor

For both the toads and frogs, intracellular [Na+] and [Cl] were highest in the submaxillary, and lowest in the triceps femoris, gracilis major, semimembranosus and sartorius muscles (Tables 8, 9). Intracellular [K+] followed the reverse order. Despite these differences in intracellular ion concentrations, all the muscles within each species followed the same general pattern with respect to temperature. Cooling the cane toads from 30 to 20 °C resulted in increases of [Na+]i, [Cl]i and [CO2]i that averaged 12.9, 8.2 and 4.2 mmol l−1 respectively. [K+]i decreased by 11.1 mmol l−1. Cooling the toads from 20 to 10 °C decreased [Na+]i, [Cl]i, [CO2]i and [K+]i by an average of 8.3, 1.0, 1.7 and 3.5 mmol l−1 respectively. Cooling the toads increased the water content of the cells. In the bullfrogs, cooling from 30 to 20 °C produced an increase in [Na+]i, [CO2]i and [K+]i of 5.1 1.0 and 5.3 mmol l−1 respectively. [Cl]i was not affected. In response to cooling from 20 to 10 °C, [Na+]i, [CO2]i and [K+]i decreased by 6.2, 1.0 and 10.7 mmol l−1 respectively. [Cl]i was little affected. In the black racer snakes, cooling from 30 to 10 °C produced large reductions in [Na+]i (10.6 mmol l−1) and [Cl]i (10.7 mmol l−1), a 2.8 mmol l−1 increase in [CO2]i and no change in [K+]i (Table 10).

Table 8.

Skeletal muscle intracellular electrolyte concentrations, total carbon dioxide concentration and water content in the cane toad Bufo marinus

Skeletal muscle intracellular electrolyte concentrations, total carbon dioxide concentration and water content in the cane toad Bufo marinus
Skeletal muscle intracellular electrolyte concentrations, total carbon dioxide concentration and water content in the cane toad Bufo marinus
Table 9.

Skeletal muscle intracellular electrolyte concentrations and total carbon dioxide concentration in the bullfrog Rana catesbeiana

Skeletal muscle intracellular electrolyte concentrations and total carbon dioxide concentration in the bullfrog Rana catesbeiana
Skeletal muscle intracellular electrolyte concentrations and total carbon dioxide concentration in the bullfrog Rana catesbeiana
Table 10.

Skeletal muscle intracellular electrolyte concentrations and total carbon dioxide concentration in the black racer snake Coluber constrictor

Skeletal muscle intracellular electrolyte concentrations and total carbon dioxide concentration in the black racer snake Coluber constrictor
Skeletal muscle intracellular electrolyte concentrations and total carbon dioxide concentration in the black racer snake Coluber constrictor

The results of this study document for the first time a strong thermal dependence of acid–base-relevant ion concentration in air-breathing ectotherms. In the cane toad, there is a large reduction in plasma electrolyte concentration with cooling, which largely reflects changes in hydration. It is known that some frogs take up water when transferred from room temperature to 2–5 °C, resulting in a weight gain of approximately 5 % and dilution of electrolytes in the plasma and gastrocnemius muscle (Weathers, 1975; Miller et al., 1968). This water retention is thought to be associated with a reduction in kidney function and a change in the set point for osmoregulation. In the present study, plasma protein concentration in B. marinus fell by 10.1 % between 30 and 20 °C and by 7.4 % between 20 and 10 °C, which is in excellent agreement with the reductions in hematocrit observed previously (Stinner et al., 1994). Hydration did not, however, lower the [HCO3]pl, which presumably would have produced an acidosis. Simple dilution of [HCO3]pl between 30 and 10 °C would have reduced [HCO3]pl by approximately 4 mmol l−1. The generation of HCO3 by the toad was probably associated with an equivalent transfer of Na+ into the extracellular fluid since dilution of Na+ between 30 and 10 °C should have reduced its concentration by approximately 4 mmol l−1 below that actually measured. In the bullfrog, water retention in the plasma was not evident between 30 and 20 °C, but between 20 and 10 °C there was a modest 5 % reduction in osmolality. With the possible exception of K+, the thermal dependence of intracellular ion concentrations cannot be explained by changes in water content in the toads and frogs (see below).

In B. marinus, the in vivo ΔpHplT ratio is the same as we measured for the in vitro ΔpH/ΔT ratio (−0.014 °C−1; Howell et al., 1970; Boutilier et al., 1987). Our results show that the in vivo ΔpH/ΔT relationship was produced not only by the passive ΔpK/ΔT of the plasma proteins but also by ion exchange (e.g. NaHCO3) offsetting a potential dilution acidosis. Since the in vivo ΔpHpl/ΔT is close to that for the ΔpKproT, this suggests that plasma protein charge is independent of temperature (Stadie et al., 1925; Reeves, 1972). Estimated plasma protein charge in the toad is the same at 30 and 20 °C, but tends to be higher at 10 °C (Table 1). The increase between 20 and 10 °C may have resulted from a net removal of anions (e.g. Cl) from the plasma, or could simply reflect the cumulative effect of measurement errors associated with the electrolyte determinations and variation in electrolyte concentration among individual animals. Plasma protein charge is independent of temperature in R. catesbeiana (Table 2). Presumably, the in vitro ΔpHplT coefficient in the bullfrog is close to that in the cane toad (−0.014 °C−1). However, the in vivo ΔpHplT relationship in R. catesbeiana is uncertain. The studies by Howell et al. (1970), Mackenzie and Jackson (1978) and Rocha and Branco (1998) are in reasonable agreement at 30 and 20 °C, but not at 10 °C. Between 20 and 10 °C, Howell et al. (1970) reported an in vivo ΔpHplT ratio of −0.022 °C−1, while Mackenzie and Jackson (1978) and Rocha and Branco (1998) report a value of −0.006 °C−1. Unfortunately, the frequently cited ΔpHplT values reported by Reeves (1972) and Malan et al. (1976) were from blood samples taken by heart puncture on pithed frogs. Thus, these values may be seriously in error because of struggling by the animals and sampling varying proportions of venous and arterial blood from the heart. Between 30 and 10 °C, plasma protein charge decreased by 21 % in C. constrictor (Stinner et al., 1998).

In B. marinus and R. catesbeiana, most of the in vivo skeletal muscle ΔpHi occurred between 30 and 20 °C and there was little change in pHi between 20 and 10 °C (Tables 3, 4). Hence, the in vivo ΔpHi/ΔT did not result simply from the passive physicochemical response of the intracellular buffers, as occurs in the muscle homogenate. Active processes, e.g. adjustment in and/or ion exchange, must have been involved. Boutilier et al. (1987) also concluded that temperature-induced acid–base-relevant ion exchange occurs in the skeletal muscle of B. marinus because they also found nonlinear ΔpHi/ΔT relationships. The lack of agreement between the in vivo and in vitro results strongly suggests that muscle protein charge is affected by temperature. For example, the low ΔpHi/ΔT between 20 and 10 °C in both species suggests acidification of the intracellular proteins.

The pHi values listed in Table 3 for cane toads at 20 °C are close to those reported by Pörtner et al. (1990) and Boutilier et al. (1987), but our pHi values at 30 and 10 °C are at least 0.2 and 0.1 pH units lower, respectively, than those of Boutilier et al. (1987). In addition, Malan et al. (1976) reported much lower pHi values in R. catesbeiana, e.g. only 6.8 at 30 °C compared with our value of 7.085 (Table 4), and a linear ΔpHi/ΔT relationship between 30 and 10 °C. We cannot explain the differences in reported pHi values. No systematic study, other than at 20 °C, has compared the DMO (5,5-dimethyl-2,4-oxazolidinedione) and homogenate techniques. Pörtner et al. (1990) discuss possible reasons for expecting different pHi values between the DMO and tissue homogenate methods. As discussed below, the ΔpHi/ΔT relationships in Tables 3 and 4 are consistent with adjustments in muscle ion concentrations reported in Tables 8 and 9.

The results of the present study do not support the commonly held view that air-breathing ectotherms use only ventilatory regulation and passive physicochemical processes to adjust pH with changes in temperature. Tables 8 and 9 document large changes in [Na+]i, [K+] i, [Cl]i and [CO2]i in the skeletal muscles of toads and frogs. Between 30 and 20 °C, [Na+]i and [CO2]i increase, and between 20 and 10 °C [Na+]I and [CO2]i decrease. While adjustments in (ventilatory regulation) with changes in temperature undoubtedly affect the ΔpHi/ΔT coefficients, it is evident that there are important changes in [NaHCO3] that also contribute to the adjustment of pHi. For example, the decrease in [NaHCO3] between 20 and 10 °C helps to explain why the in vivo ΔpHi/ΔT ratio is much lower than the in vitro ΔpHi/ΔT ratio.

The large changes in intracellular ion concentrations observed in toads cooled from 30 to 20 °C are consistent with a general inhibition of ion pumps within the sarcolemma or possibly a general increase in permeability to ions across the sarcolemma. [Na+]i, [Cl]i and [CO2]i increased and [K+]i decreased, suggesting movement of these ions down their respective concentration gradients between the extracellular fluid and the sarcoplasm. The large inward movement of Na+ was probably accompanied by some outward transfer of H+. Na+/H+ exchangers are known to exist in the plasma membrane of frog skeletal muscle cells (Abercrombie et al., 1983). No evidence has been found for K+/H+ exchangers in amphibian skeletal muscle (Amorena et al., 1990). Between 20 and 10 °C, [K+]i decreased by approximately 4 %, possibly reflecting dilution by water. [Na+]i, [Cl]i and [CO2]i also declined, which cannot, of course, be explained by a general inhibition of ion pumps or increase in membrane permeability. To maintain electroneutrality, the large loss of Na+ was probably accompanied by H+ entering the cells and titrating protein, phosphate and bicarbonate buffers. In R. catesbeiana, between 30 and 20 °C, there was an increase in [Na+]i and [K+]i with little change in [Cl]i. To maintain electroneutrality, there would have been a reciprocal shift of H+ out of the cell, raising pHi, intracellular net negative protein charge and [HCO3]i. Between 20 and 10 °C, the opposite occurred, [Na+]i and [K+]i decreased with little change in [Cl]i, so [H+]i would have increased. In the snakes, acidification of skeletal muscle was accompanied by decreases in [Na]i and, to a lesser extent, [Cl]i (Table 10). Because cells contain many kinds of ions, including organic compounds, it is virtually impossible to calculate intracellular protein charge from the principle of electroneutrality as was done in this study for plasma proteins. Despite this, the large intracellular ion concentration changes and the ΔpHi/ΔT data relative to the in vitro ΔpH/ΔT coefficients strongly suggest that skeletal muscle protein charge is not constant between 30 and 10 °C.

The bullfrog whole skeletal muscle [CO2] values obtained in the present study are approximately 4 mmol l−1 lower than previously published by Stinner et al. (1998) and Reeves (1972), despite using similar techniques. These large differences in [CO2] may reflect seasonal variation. Rocha and Branco (1998) reported that arterial is twice as high in winter-acclimatized as in summer- and spring-acclimatized R. catesbeiana, but pH was not affected, indicating increased [HCO3]pl levels during winter. The data of Reeves (1972) were collected in March–June, the data of Stinner et al. (1998) were collected in October–April, while the muscle CO2 results in the present report were obtained during July and August.

In summary, the results of this study show that, in cane toads and bullfrogs, plasma protein charge is independent of temperature between 30 and 10 °C, with the possible exception of cane toads at 10 °C. Increased hydration with cooling in the toads did not lower their [HCO3]pl because of a compensatory increase in [NaHCO3]pl. In contrast to plasma, skeletal muscle protein charge in the cane toads and bullfrogs is temperature-dependent. Large adjustments in intracellular ion concentrations took place when the temperature was changed. While differences in hydration, membrane permeability to ions and Q10 effects upon ion pump activity may all be involved, NaHCO3 transfer across the sarcolemma played a significant role in the adjustment of pHi. Large changes in skeletal muscle intracellular ion concentrations also occurred in the black racer snakes, possibly in response to temperature-induced changes in and carbonic acid levels. We conclude that air-breathing ectotherms utilize ion-exchange mechanisms, in addition to passive physicochemical adjustments of the buffers and active ventilatory regulation, to adjust their pH in response to changes in body temperature and that skeletal muscle protein charge is not independent of temperature.

Abercrombie
,
R. F.
,
Putnam
,
R. W.
and
Roos
,
A.
(
1983
).
The intracellular pH of frog skeletal muscle: Its regulation in isotonic solutions
.
J. Physiol., Lond.
345
,
175
187
.
Amorena
,
C. E.
,
Wilding
,
T. J.
,
Manchester
,
J. K.
and
Roos
,
A.
(
1990
).
Changes in intracellular pH caused by high K in normal and acidified frog muscle
.
J. Gen. Physiol.
96
,
959
972
.
Austin
,
J. H.
,
Sunderman
,
F. W.
and
Camack
,
J. G.
(
1927
).
Studies in serum electrolytes. II. The electrolyte composition and the pH of serum of a poikilothermous animal at different temperatures
.
J. Biol. Chem.
72
,
677
685
.
Boutilier
,
R. G.
,
Emilio
,
M. G.
and
Shelton
,
G.
(
1986
).
The effects of mechanical work on electrolyte and water distribution in amphibian skeletal muscle
.
J. Exp. Biol.
120
,
333
350
.
Boutilier
,
R. G.
,
Glass
,
M. L.
and
Heisler
,
N.
(
1987
).
Blood gases and extracellular/intracellular acid–base status as a function of temperature in the anuran amphibians Xenopus laevis and Bufo marinus
.
J. Exp. Biol.
130
,
13
25
.
Butler
,
P. J.
and
Day
,
N.
(
1993
).
The relationship between intracellular pH and seasonal temperature in the brown trout Salmo trutta
.
J. Exp. Biol.
177
,
293
297
.
Cameron
,
J. N.
(
1984
).
The acid–base status of fish at different temperatures
.
Am. J. Physiol.
246
,
R452
R459
.
Cameron
,
J. N.
(
1989
).
Acid–base homeostasis: past and present perspectives
.
Physiol. Zool.
62
,
845
865
.
Coulson
,
R. A.
and
Hernandez
,
T.
(
1964
).
Biochemistry of the Alligator: A Study of Metabolism in Slow Motion. Louisiana State University Press
.
Douse
,
M. A.
and
Mitchell
,
G. S.
(
1991
).
Time course of temperature effects on arterial acid–base status in Alligator mississippiensis
.
Respir. Physiol.
83
,
87
102
.
Edsall
,
J. T.
and
Wyman
,
J.
(
1958
).
Biophysical Chemistry, vol. 1
.
New York
:
Academic Press
.
Gaillard
,
S.
and
Malan
,
A.
(
1985
).
Intracellular pH–temperature relationships in a water breather, the crayfish
.
Molec. Physiol.
7
,
1
16
.
Graber
,
M.
,
Barry
,
C.
,
Dipaola
,
J.
and
Hasagawa
,
A.
(
1992
).
Intracellular pH in OK cells. II. Effects of temperature on cell pH
.
Am. J. Physiol.
262
,
F723
F730
.
Heisler
,
N.
(
1986a
).
Buffering and transmembrane ion transfer processes
. In
Acid–Base Regulation in Animals
(ed.
N.
Heisler
), pp.
3
47
.
Amsterdam
:
Elsevier Biomedical Press
.
Heisler
,
N.
(
1986b
).
Comparative aspects of acid–base regulation
. In
Acid–Base Regulation in Animals
(ed.
N.
Heisler
), pp.
397
450
.
Amsterdam
:
Elsevier Biomedical Press
.
Hitzig
,
B. M.
,
Perng
,
W.
,
Burt
,
T.
,
Okunieff
,
P.
and
Johnson
,
D. C.
(
1994
).
1H-NMR measurement of fractional dissociation of imidazole in intact animals
.
Am. J. Physiol.
266
,
R1008
R1015
.
Howell
,
B. J.
,
Baumgardner
,
F. W.
,
Bondi
,
K.
and
Rahn
,
H.
(
1970
).
Acid–base balance in cold blooded vertebrates as a function of body temperature
.
Am. J. Physiol.
218
,
600
606
.
Johnson
,
D. C.
,
Burt
,
C. T.
,
Perng
,
W.
and
Hitzig
,
B. M.
(
1993
).
Effects of temperature on muscle pHi and phosphate metabolites in newts and lungless salamanders
.
Am. J. Physiol.
34
,
R1162
R1167
.
Lehoux
,
E. A.
and
Guderley
,
H. E.
(
1997
).
Thermally induced changes in intracellular pH and modulators of phosphofructokinase in trout white muscle
.
J. Exp. Biol.
200
,
931
939
.
Lindinger
,
M. I.
and
McDonald
,
D. G.
(
1986
).
Cutaneous and renal responses to intravascular infusions of HCl and NH4Cl in the bullfrog (Rana catesbeiana)
.
Comp. Biochem. Physiol.
84A
,
113
122
.
Lutz
,
P. L.
,
Bergey
,
A.
and
Bergey
,
M.
(
1989
).
Effects of temperature on gas exchange and acid–base balance in the sea turtle Caretta caretta at rest and during routine activity
.
J. Exp. Biol.
144
,
155
169
.
Mackenzie
,
J. A.
and
Jackson
,
D. C.
(
1978
).
The effect of temperature on cutaneous CO2 loss and conductance in the bullfrog
.
Respir. Physiol.
32
,
313
323
.
Malan
,
A.
,
Wilson
,
T. L.
and
Reeves
,
R. B.
(
1976
).
Intracellular pH in cold-blooded vertebrates as a function of body temperature
.
Respir. Physiol.
28
,
29
47
.
Miller
,
D. A.
,
Standish
,
M. L.
and
Thurman
,
A. E.
(
1968
).
Effects of temperature on water and electrolyte balance in the frog
.
Physiol. Zool.
41
,
500
506
.
Pörtner
,
H. O.
,
Boutilier
,
R. G.
,
Tang
,
Y.
and
Toews
,
D. P.
(
1990
).
Determination of intracellular pH and after metabolic inhibition by fluoride and nitrilotriacetic acid
.
Respir. Physiol.
81
,
255
274
.
Reeves
,
R. B.
(
1972
).
An imidazole alphastat hypothesis for vertebrate acid–base regulation: tissue carbon dioxide content and body temperature in bullfrogs
.
Respir. Physiol
.
14
,
219
236
.
Reeves
,
R. B.
(
1977
).
The interaction of body temperature and acid–base balance in ectothermic vertebrates
.
Annu. Rev. Physiol
.
39
,
559
586
.
Rocha
,
P. L.
and
Branco
,
L. G. S.
(
1998
).
Seasonal changes in the cardiovascular, respiratory and metabolic responses to temperature and hypoxia in the bullfrog Rana catesbeiana
.
J. Exp. Biol.
201
,
761
768
.
Rodeau
,
J.-L.
(
1984
).
Effects of temperature on intracellular pH in crayfish neurons and muscle fibers
.
Am. J. Physiol.
246
,
C45
C49
.
Siggaard-Anderson
,
O.
(
1974
).
The Acid–Base Status of the Blood, fourth edition
.
Copenhagen
:
Munksgaard
.
Stadie
,
W. C.
,
Austin
,
J. H.
and
Robinson
,
H. W.
(
1925
).
The effect of temperature on the acid–base protein equilibrium and its influence on the CO2 absorption curve of whole blood, true and separated serum
.
J. Biol. Chem.
66
,
901
920
.
Stinner
,
J. N.
,
Grguric
,
M. R.
and
Beaty
,
S. L.
(
1996
).
Ventilatory and blood acid–base adjustments to a decrease in body temperature from 30 to 10 °C in black racer snakes Coluber constrictor
.
J. Exp. Biol.
199
,
815
823
.
Stinner
,
J. N.
,
Hartzler
,
L. K.
,
Grguric
,
M. R.
and
Newlon
,
D. L.
(
1998
).
A protein titration hypothesis for the temperature-dependence of tissue CO2 content in reptiles and amphibians
.
J. Exp. Biol.
201
,
415
424
.
Stinner
,
J. N.
,
Newlon
,
D. L.
and
Heisler
,
N.
(
1994
).
Extracellular and intracellular CO2 concentration as a function of body temperature in the toad Bufo marinus
.
J. Exp. Biol
.
195
,
345
360
.
Stinner
,
J. N.
and
Wardle
,
R. L.
(
1988
).
Effect of temperature upon carbon dioxide stores in the snake Coluber constrictor and the turtle Chrysemys scripta
.
J. Exp. Biol.
137
,
529
548
.
Ultsch
,
G. R.
and
Jackson
,
D. C.
(
1996
).
pH and temperature in ectothermic vertebrates
.
Bull. Alabama Mus. Nat. Hist.
18
,
1
41
.
Van Dijk
,
P. L. M.
,
Hardewig
,
I.
and
Pörtner
,
H. O.
(
1997
).
Temperature-dependent shift of pHi in fish white muscle: contributions of passive and active processes
.
Am. J. Physiol.
272
,
R84
R89
.
Weathers
,
W. W.
(
1975
).
Circulatory responses of Rana catesbeiana to temperature, season and previous thermal history
.
Comp. Biochem. Physiol.
51A
,
43
52
.
Whiteley
,
N. M.
,
Taylor
,
J. K.
and
Taylor
,
E. W.
(
1995
).
Extracellular and intracellular acid–base status in the freshwater crayfish Austropotamobius pallipes between 1 and 12 °C
.
J. Exp. Biol.
198
,
567
576
.