Apneic Oxygen Uptake in the Torpid Bat, Eptesicus Fuscus

Many hibernating mammals alternate periods of breathing with apneas of variable duration. During studies of acid-base state and ventilation in the torpid bat, Eptesicus fuscus, it became apparent that significant O₂ uptake was occurring during these apneas. A previous study demonstrated cutaneous exchange of CO₂ in euthermic Eptesicus fuscus (Herreid et al. 1968), but did not reveal significant cutaneous exchange of O₂. Theoretical calculations support the notion that apnea is prolonged in torpid hedgehogs by passive diffusion of O₂ into the lungs, a process termed apneic oxygenation (Clausen and Ersland, 1968; Malan, 1982). This study presents direct evidence for O₂ uptake during apnea in Eptesicus fuscus and how it is influenced by changing body temperature (T_b).

Big brown bats, Eptesicus fuscus, were captured locally in accordance with a scientific collector’s permit issued by the Rhode Island Department of Environmental Management. Bats were kept in a specially prepared colony room maintained at 25 °C. Outgoing ventilation from this colony room flowed through anti-viral filters. The bats were housed in stainless-steel cages in groups according to cage size and fed live mealworms raised on a diet of flour, chicken feed, potato slices and monkey chow. They had free access to vitamin-supplemented water (Poly-Vi-Sol or equivalent) at all times. Individual bats were identified by numbered bands on their forearms. Body mass upon capture ranged from 14 to 16 g, but increased to over 20 g in some individuals during captivity.

Bats were placed in the experimental chamber at ambient temperatures selected to reach target T_b values of 5, 10, 20 or 30 °C while simultaneously monitoring Tb, ventilation, instantaneous O₂ uptake and the electrocardiogram (ECG). At least 2 h of steady-state physiological conditions were observed before acquiring data.

A specially made chamber permitted simultaneous monitoring of Tb, ventilation, instantaneous O2 uptake and ECG without disturbing the bat. Details of the chamber and analytical procedures are described elsewhere (Szewczak and Jackson, 1992a,b). In brief, the chamber exploits this bat’s penchant for roosting in crevices. Although the chamber’s interior is only slightly larger than the bat’s, they voluntarily crawled into it without a complete arousal from torpor after only a few training sessions. The small dead space within the chamber facilitated the detection of respiratory movements, while providing a short time constant for monitoring gas exchange with a sensitivity sufficient to reveal O₂ uptake from single breaths. To ensure that the system reliably measured apneic O₂ uptake, it was tested using a faux bat in the chamber. This procedure yielded zero O₂ uptake following the identical calibration protocol used with the experimental bats. A computerized data-acquisition system collected oxygen uptake and respiratory movement data using programs that we had developed (IBM XT with Data Translation DT2801 A/D board, ASYST data acquisition and data management software: Keithly-ASYST Software Technologies, Inc., version 3.0).

During experiments, gas flow through the chamber was from tanks of pressurized air. This eliminated any potential errors from variations in gas composition. A bypass circuit, however, enabled room air to ventilate the chamber during Tb equilibration and analyzer calibration of the experimental gas flow (Szewczak and Jackson, 1992b). Total O₂ uptake was calculated by integrating the instantaneous rate of O₂ uptake record with respect to time. As far as possible, total O₂ uptake was calculated in phase with ventilatory cycles (the onset of a ventilatory bout to the end of an apneic interval). Steady-state sections from apneic periods of the instantaneous O₂ uptake recording were separately integrated with respect to time to determine rates of apneic O₂ uptake. The fraction of time that was apneic was calculated from the ventilation recordings and then applied to the rates of apneic O₂ uptake to determine the total apneic O₂ uptake. We assumed somatic metabolism to be independent of ventilatory state because of the uniform heart rate from ventilation to apnea.

Below T_b=30°C, a typical ventilatory cycle consisted of a 1-to 9-min bout of rather evenly spaced breaths, followed by an apneic interval. Occasional sporadic breaths punctuated the apneic interval, most commonly at T_b=20°C. (Apneic O₂ uptake was calculated only from apneic intervals free from these sporadic breaths.) Apneas averaged longer at T_b=10°C (56.7415.3 min) than at T_b=20°C (6.1±0.4min; maximum: 13.7 min) or T_b=5 °C (6.5±0.8 min; maximum: 40.9 min). Nevertheless, the fraction of time apneic was similar for these temperatures (Table 1). The longest recorded apnea was 147 min at T_b=10 °C. This bat was apneic for 95 % of a complete ventilatory cycle with an apneic O₂ uptake rate of 24.9 panol O₂ h⁻¹.

Heart rate was essentially constant from ventilation to apnea, particularly if compared to heart rates of typical diving mammals, which may reduce heart rate by 80 % during apnea (Elsner, 1965). Cardiogenic pulses synchronous with ECG signals were visible in the plethysmograph pressure records (Fig. 1). These pulses were most pronounced following a ventilatory bout, then steadily decreased during apnea, often becoming lost in the noise. Heart rates were 8.6±0.6, 13.5±0.4 and 35.443.8 beats min⁻¹ at T_b=5, 10 and 20 °C, respectively.

Instantaneous O₂ uptake records (Fig. 2) consisted of spikes due to ventilatory bouts, separated by periods during apnea with O₂ uptake remaining above zero. Simultaneous plethysmography confirmed the apneas indicated in these records. Apneic O₂ uptake was often lower immediately following a ventilatory bout, then increased to a steady level. Total its well with allometric data from other hibernators (Geiser, 1988) only if the O₂ uptake calculation includes the apneic contribution between ventilatory spikes. Apneic intervals were too brief to measure apneic O₂ uptake confidently at T_b=30 °C.

The rate of apneic O₂ uptake increased as Tb was raised from 5 to 20 °C, but did not maintain a constant fraction of total O₂ uptake (Fig. 3). The maximum fraction of total O2 uptake from apnea was at T_b=10 °C.

Oxygen uptake during apnea may occur either by cutaneous exchange of gas or via the airways into the lung. Several lines of evidence indicate that airway exchange is the more likely process in Eptesicus fuscus. A study of cutaneous gas exchange in euthermic Eptesicus fuscus measured gas exchange from the head and body separately (Herreid et al. 1968). The cutaneous output of CO₂ ranged from 18.8 to 104μmol CO₂ h⁻¹ for ambient temperatures of 18 and 37.5 °C, respectively, with cutaneous O₂ uptake reported to be insignificant. In that study, the wings of the bat were held open, increasing the available cutaneous surface area by approximately five times over the folded wing posture of the torpid bats in the present study. Because the cutaneous conductance of O₂ is 20 times less than that of CO₂ (Piiper et al. 1976), the potential for cutaneous O₂ uptake of the bats in the present study would be two orders of magnitude less than the cutaneous CO₂ exchange of the bats in the prior study. Therefore, the expected cutaneous O₂ uptake would be at most 0.2 μmol O₂ h⁻¹ for bats at T_b=20 °C and below, considerably less than the 8.9–33.9 μmolO₂⁻¹ range of apneic O₂ uptake measured in the present study. It is thus unlikely that cutaneous O₂ uptake could account for more than a minor portion of the apneic O₂ uptake measured in this study.

For effective gas exchange to occur by diffusion down the airway, the glottis must remain open during apnea. The cardiogenic pulsations recorded on the plethysmographic pressure trace (Fig. 1) strongly support an open glottis during apnea (Malan, 1973, 1982). There is no abrupt change in the quality of the pulsations in the transition from breathing to apnea that would indicate glottal closing. Based upon an analysis of breathing pattern and oxygen uptake, it was similarly concluded that the glottis remained open during apnea in torpid pipistrelle bats (Hays et al. 1991). Nevertheless, an investigation of non-ventilatory O₂ uptake during apnea in the little brown bat, Myotis lucifugas, concluded that its glottis was closed during apnea (Thomas et al. 1990). This conclusion was based upon measured rates of non-ventilatory O₂ uptake derived from numerical transformation of the instantaneous O₂ uptake data. The transformation procedure may possibly have introduced uncertainty, particularly at the low levels of O₂ uptake of 6.5 g Myotis lucifugus at T_b=5 °C. Nonetheless, it is possible that there may be interspecies differences and that the glottis of Myotis lucifugus does remain closed during apnea, but the direct measurements of apneic O₂ uptake and observation of cardiogenic pulses of the present study suggest otherwise. To explore this issue further, the following sections quantitatively examine O₂ management during apnea in Eptesicus fuscus.

During apnea, the bat’s heart rate remains essentially constant. This continued circulation enables the tissues to withdraw O₂ stored in the blood, which reduces its and enables it to remove O₂ stored in the lungs. The extent of these O₂ stores will be calculated and then used to determine the duration of apnea that they can support. Comparing this result with the observed durations of apnea should suggest whether it is necessary to invoke a mechanism for replenishing O₂ during apnea. The following calculations assume a fully O₂-loaded bat at the onset of a 1 h apnea at T_b=10 °C. The bat is presumed to be consuming O₂ at the experimentally measured metabolic rate of 35.2μmol O₂h⁻¹. All gas volumes are corrected for 10 °C; the results from these calculations are compiled in Table 2.

The blood volume of bats is greater than that of other mammals; it has been determined to be about 13 ml 100 g⁻¹ body mass (M) in the bat Myotis lucifugus (Kallen, 1960). Because blood volume scales proportional to M^1.02 (Stahl, 1967), it may be estimated to be 1.9 ml for a 15 g Eptesicus fuscus. Assuming 25 % of this to be arterial blood with an O₂-carrying capacity of 6.83 mmol l⁻¹ (Malan, 1982), arterial blood should hold 3.2 μmol O₂. Assuming a decrease in arterial saturation to 50 % provides 4.9 μmol O₂ in venous blood for a total blood store of 8.1 μmol O₂ (an estimate, since the respiratory properties of this bat’s blood have not been determined). However, bats are known to reduce haematocrit during torpor by splanchic sequestering (Kallen, 1960; Martin and Stehn, 1977), so the total blood store of O₂ is probably unavailable. Thus, for this calculation, the estimated blood store will be reduced by one-third, yielding 5.4μmol O₂

Because of its high oxygen affinity, oxygen bound to myoglobin is not available to reenter the circulation (Dejours, 1981), and thus is only useful to the tissue in which it resides, which is primarily muscle. Since apneic bats are inactive, myoglobin O₂ stores are considered to be inconsequential.

Initial blood and lung O₂ stores can account for only 21% of the metabolic O₂ requirement during a 1 h apnea and are thus incapable of supporting apneas much longer than 12 min. If the presumption of splanchic sequestering were removed, and all blood was available to the bat, then initial O₂ stores would be capable of supporting an apnea of 17 min, which is still less than the 56.7 min mean duration of apnea at T_b=10°C. There must therefore be a process for replenishing O₂ during apnea, which is consistent with the apneic O₂ uptake measured by the present study.

Oxygen can enter the blood either through the skin or via the respiratory airways and membranes. The theoretical calculation of airway diffusion potential follows from Malan (1982). The conditions stated above also apply to the following calculations.

Based upon the study by Herreid et al. (1968) the cutaneous O₂ uptake would be at most 0.2 μmolO₂h⁻¹, which is probably a conservative overestimate for T_b=10°C. This is equivalent to only 0.6% of the bat’s total O₂ requirement, and capable of extending apnea by perhaps 1 min following exhaustion of initial O₂ stores.

As a consequence of the respiratory quotient and the high CO₂ capacitance of blood and tissues (Dejours, 1981), CO₂ incompletely replaces the volume of O₂ absorbed from the lungs. Instead of contracting, the mechanical forces acting on the lung maintain it at a constant volume, and thus it draws an influx of ambient air (Malan, 1982). The diffusive efflux of CO₂ is limited by flowing counter to this influx and by its smaller concentration gradient and lower diffusivity compared to O₂ (0.144cm²s⁻¹ at 10°C) (Weast, 1979).

The mean ventilatory to apneic was 3.2 kPa (Szewczak and Jackson, 1992zz). From this value, the net diffusive efflux of CO2 is estimated to be 3.98p,mol CO₂ h⁻ 1(0.092 ml h⁻¹). For an assumed respiratory quotient of 0.78, the total would be 27.5 μmol h⁻¹ (0.635 ml h⁻¹). This leaves 23.5μmol h⁻¹ (0.543 ml ⁻¹) of CO₂ to accumulate in the blood and tissues, with some cutaneous release, but it is unlikely that all of it fills the lungs. Metabolic absorption of the diffusive influx of 0.385 ml O₂ h⁻¹, along with the diffusive efflux of 0.092 ml CO₂h⁻¹, combine to yield a continuous convective influx of 0.477 ml h⁻¹ of air, providing an additional 4.3μmolChh⁻¹. Absorption of the initial lung store of 0.044 ml O₂ provides a one-time bulk convection of 0.044 ml of air, furnishing 0.4 μmol O_2. Thus, the total bulk influx of O₂ during the first hour of apnea would be 4.7μmol. This process would contribute 13.3% of the bat’s total O₂ requirement.

The processes of diffusion and bulk convection combine to provide 21.3μmol O₂ during this theoretical first hour of apnea. This provides 60.5% of the bat’s O₂ requirement. Because it is based upon diffusion, the rate of this process should be dependent upon concentration gradient. This notion is consistent with the time course of instantaneous O₂ uptake recordings (Fig. 2). Following a ventilatory bout, O₂ stores are at capacity. This minimizes the concentration gradient, and hence O₂ influx. Consuming the initial O₂ stores improves the gradient and enhances the rate of O₂ uptake. This continues until an equilibrium is achieved between O₂ consumption and diffusion capacity, which then stabilizes the rate. The calculated equilibrium apneic O₂ uptake rate of 20.9μmolO₂h⁻¹ (diffusive plus convective flux, following exhaustion of initial stores) also compares favorably with the measured rate of 18.9±2.9μmolO₂h⁻¹ for bats at T_b=10°C. We thus conclude that passive O₂ influx down the airway is the most likely mechanism by which these bats acquire O₂ during apnea.

The idealized airway cylinder used in the above calculation did not consider potential resistances such as those from oral or nasal airways. Thus, the actual O₂ influx might be expected to be less than the calculated value. However, some experimentally measured rates of apneic O₂ uptake were actually higher (e.g. 24.9 μmol O₂h⁻¹ measured vs 20.9μmol O₂h⁻¹ calculated for T_b=10°C). Because the above calculation assumed passive diffusion in still air, the difference may result from mechanical agitation of gases within the airways. It has been previously suggested that the mechanical action from the heart, i.e. cardiogenic mixing, may facilitate diffusion potential (West and Hugh-Jones, 1961; Fukuchi et al. 1976). Indeed, the cardiogenic pulses recorded from Eptesicus fuscus are suggestively prominent relative to the ventilation movements (Fig. 1).

Cardiogenic mixing may also influence the observed temperature-dependency of apneic O₂ uptake. The Q₁₀ for diffusion of O2 in air is about 1.2 (Dejours, 1981), whereas the Q₁₀ of the rate of apneic O₂ uptake from T_b=10 to 20 °C was 2.4, and from T_b=5 to 10 °C was 4.5. The departure of these values from a Q₁₀ of 1.2 indicates a dependency upon some other factor in addition to temperature. (Any temperature effects on haemoglobin affinity for O₂ would favorably influence the O₂ concentration gradient and reduce the Q₁₀, rather than increase it.) The effect of cardiogenic mixing may be accentuated at low heart rates, in which it first breaks up the stratification of gases that inhibits diffusion (Fukuchi et al. 1976). At T_b=5 °C, the average interval between heartbeats is 7 s, a plausible interval for stratification to occur in small airways. Thus, the large Q₁₀ of 4.5 from T_b=5 to 10°C could be attributed to the differential effects of cardiogenic mixing at these temperatures.

The observation of cardiogenic pulsations and a theoretical calculation of O₂ diffusion support the conclusion that apneic O₂ uptake occurs down an open airway. This process is apparently enhanced by the mechanical action of the heart, which could properly be considered to be an important respiratory organ for this animal.

We thank Dr Andre Malan for his helpful comments and encouragement of this investigation. We gratefully acknowledge the assistance of Dr Odile Mathieu-Costello who provided airway morphometric data. This work was supported by Grant DCB8802045 from the National Science Foundation (D.C.J.).