Ventilation patterns in red kangaroos (Macropus rufusDesmarest): juveniles work harder than adults at thermal extremes, but extract more oxygen per breath at thermoneutrality

Red kangaroos (Macropus rufus Desmarest) are one of the largest extant marsupials and inhabit much of arid and semi-arid inland Australia. Their unpredictable environment is typified by stochastic rainfall and extremes of ambient temperature, ranging from below freezing at night to over 50°C diurnally (Dawson,1995). The survival of juvenile red kangaroos in this environment is problematic and significant population recruitment occurs mainly after periods of high or extended rainfall. Such events provide juvenile kangaroos with access to the high-quality forage needed for their long-term growth and survival (Munn et al., 2006; Munn and Dawson, 2006). However, in addition to their requirement for high-quality forage, which occurs over a relatively long physiological time-scale (i.e. weeks to months),juvenile red kangaroos must be able to cope with the more immediate challenges of their environment. These challenges include high radiant heat loads, daily thermal extremes and scarce free water(Dawson, 1995).

The red kangaroos' reproductive strategy affords their offspring some protection from environmental extremes in the form of maternal care while the young develop inside the pouch, but only during the early stages of development. By 190 days old (d) and weighing around 2 kg, the young is fully furred and will venture out of the pouch for short periods. At this`in–out' stage, the young kangaroo is still unable to maintain deep body temperature (T_b) for long periods, regularly returning to the pouch for warmth and safety (Frith and Sharman, 1964; Dawson,1995). By 230–250 d, thermoregulation is well developed(Munn and Dawson, 2001) and the young kangaroo permanently leaves the pouch, becoming a young-at-foot(YAF; mass 4–5 kg). YAF red kangaroos forage in association with their mothers but also continue to take milk by putting their head into the pouch to access the same teat used during pouch-life. It is not for another 100 days or so that the YAF is fully weaned, at around 360 d and weighing 10–11 kg(Frith and Sharman, 1964; Dawson, 1995). Once young kangaroos have permanently left the mother's pouch they face the same environmental challenges as adults, but their smaller body size and high-energy requirements for growth (Munn and Dawson, 2001; Munn and Dawson, 2003) potentially impact on their ability to meet such thermal challenges.

Munn and Dawson showed that the energy and water requirements of YAF red kangaroos differ from those of adults(Munn and Dawson, 2001; Munn and Dawson, 2003). For example, at thermoneutral ambient temperatures (T_a) around 25°C, YAF red kangaroos had absolute rates of oxygen consumption (ml min^–1) that were 58% that of mature, non-lactating females,despite being only 30% of maternal body mass. To maintain heat balance at cold T_as, at –5°C, the absolute oxygen consumption(ml min^–1) of YAFs increased to 68% that of mature females(Munn and Dawson, 2001). In addition, YAF also had high water requirements at high T_a(∼45°C), where juvenile kangaroos evaporated 2.5-times more water per unit body mass than mature females. These high evaporative water losses were achieved largely through elevated rates of respiratory evaporative cooling(i.e. panting) (Munn and Dawson,2001). There is clearly a requirement for a larger relative capacity of the ventilatory system of the juvenile kangaroo compared to the adult. Therefore, we here explore the ventilatory physiology of juvenile red kangaroos by asking whether their ventilatory characteristics are similar to those of adults.

To evaluate the capabilities of the juvenile kangaroo ventilatory system,relative to adults, we used an allometric approach(Schmidt-Nielsen, 1984). That is, we compared the juvenile and adult red kangaroos' ventilation/respiratory patterns (including oxygen consumption) using the relevant body-mass exponents predicted from quarter-power scaling (West et al., 2000). The body mass (m) exponents that we used for oxygen consumption (ml min^–1; m^0.75),respiration rate (breaths min^–1; m^–0.25), tidal volume (ml; m^1.0), ventilation rate (ml min^–1; m^0.75) and oxygen extraction (%; m^0.0)are supported empirically from studies on both eutherians(Stahl, 1967) and marsupials(Dawson and Needham, 1981; Frappell and Baudinette, 1995; Withers et al., 2006). Because there is some controversy about exponent values for various parameters(Dodds et al., 2001; Kozlowski and Konarzewski,2004; West and Brown,2005; White and Seymor,2005), we also present results using an alternative exponent of 0.67 for oxygen consumption and ventilation. We therefore compared allometrically adjusted data for the ventilation/respiration characteristics of juvenile and adult kangaroos at thermoneutral (25°C) and extreme T_as of –5°C and 45°C.

Seven juvenile red kangaroos (five females and two males) were hand-reared. Four weeks prior to experimentation the animals were transferred to our laboratory at the University of New South Wales. During this time the kangaroos were familiarised with the experimental procedure and implanted intraperitoneally with temperature-sensitive radio transmitters(±0.1°C) (Sirtrack, Havelock North, Hawkes Bay, New Zealand), which had been calibrated (±0.1°C) against a certified mercury in glass thermometer (National Testing Authorities, Sydney, NSW, Australia). The kangaroos were maintained in pens (1.65×0.9×1.2 m) under a 12 h:12 h light:dark regime. Feed [Gordons rabbit pellets (Gordons Specialty Stock Feeds, Yanderra, NSW, Australia), Kangaroo cubes (Doust and Rabbidge, Forbes,NSW, Australia) and Lucerne/Bran mix (Kensington Stock Feeds, Sydney, NSW,Australia)] and water were available ad libitum. The diet of YAF kangaroos was supplemented with a ration of low lactose milk (Digestelact,Sharpe Laboratories, Sydney, Australia), the quantity of which was reduced over time until it was eliminated at normal weaning age. Juvenile red kangaroos on this diet regimen show growth and development patterns that are comparable to those raised on natural diets(Frith and Sharman, 1964; Sharman et al., 1964; Munn and Dawson, 2003). Food was withheld for 24 h prior to experimentation. During the major part of this study the juvenile red kangaroos were at an average age of 313±5 d,with a mass of 8.6±0.3 kg (mean ± s.e.m.).

For metabolic and ventilatory measurements, kangaroos were weighed to the nearest 0.1 kg (Salter Scales, Sydney, Australia) and placed in an open-circuit metabolism chamber (69.5×45×58 cm) within a temperature-controlled room. T_a was regulated to±0.5°C of a set point and each animal was tested at T_as of –5°, 25° and 45°C. T_a was measured (±0.1°C) using a thermocouple placed in the excurrent port of the chamber. The metabolism chamber had a mesh floor above a bath of vegetable oil to trap excreta. The walls of the metabolism chamber were painted flat black to reduce radiation reflection(Porter, 1969; Maloney and Dawson, 1994). Experiments were performed between 08.00 and 16.00 h, corresponding to the resting phase of adult red kangaroo circadian rhythm(Watson and Dawson, 1993). Animals were monitored throughout each experiment via a low-light CCD camera (Oatley Electronics, Sydney, Australia) mounted inside the chamber.

After an animal was sealed into the chamber, at least 3 h was allowed for equilibration at any given temperature; data collection then commenced when the animal's body temperature had stabilised. Flow rate (FR) of dry air into the metabolism chamber was measured upstream using a Hastings Mass flowmeter(model HFM-201; John Morris Scientific, Sydney, Australia). Flow rate was adjusted to prevent water vapour pressure inside the chamber from exceeding 15 mmHg (1 mmHg=133.3 Pa). A sub-sample of air (125 ml min^–1)was drawn from the excurrent port using a flow controller (Ametek Applied Electrochemistry R2 flow controller; Pittsburgh, PA, USA) and passed through a capacitance type relative humidity (%RH) sensor (±0.1%)(CHK-Engineering, Sydney, Australia), which was calibrated regularly using saturated solutions of lithium chloride, sodium chloride and magnesium chloride (Winston and Bates,1960). After leaving the humidity sensor, the excurrent air was dried with drierite, scrubbed of CO₂ with ascarite and re-dried before passing through an oxygen analyser (Ametek Applied Electrochemistry S3A-III).

Sensor outputs (FR, %RH, %O₂) were logged on a personal computer at 5 s intervals using Warthog Labhelper software (Warthog Systems, University of California, Riverside, CA, USA) running a 12-bit analog/digital (A/D)converter (National Instruments Lab-NB card, North Ryde, NSW, Australia). This system averaged approximately 120 readings for each recorded value and gave a maximum resolution of 0.006% for %O₂. The whole system was calibrated regularly using the iron-burn method(Young et al., 1984). Oxygen consumption (V̇_O2)(ml min^–1) was calculated as described in detail elsewhere(Munn and Dawson, 2001). At least 20 min of continuous minimal V̇_O2 during an exposure was used to determine the mean metabolic rate at each T_a.

The metabolism chamber acted as a whole-body plethysmograph, which allowed measurement of respiratory frequency (f_R; breaths min^–1) and tidal volume (V_T; ml)(Malan, 1973; Maloney and Dawson, 1994). Described in detail by Maloney and Dawson(Maloney and Dawson, 1994),plethysmography measures changes in the chamber air pressure caused by the humidification and warming/cooling of chamber air during inspiration. Chamber pressure was measured with a pressure transducer (PT-100; Sable Systems International, Las Vegas, NV, USA) and logged on a personal computer via an analog/digital converter. Data were logged at 0.05 s intervals at –5 and 25°C and at 0.03 s intervals at 45°C. The system was calibrated by injecting 115 ml of air into the chamber at the end of each experiment. Calibrations were repeated until the injection deflections were stable and matched the deflection kinetics exhibited by the respiring animal(Maloney and Dawson,1994).

For each animal, V_T was established as the mean of five sets of respiratory traces taken throughout the 20-min period for minimal V̇_O2 at each T_a. V_T and respiratory minute volume(ventilation rate, V̇_I=V_T×f_R)was measured at body temperature and pressure, saturated, but converted to standard temperature and pressure, dry, for analysis.

Data for juveniles were compared directly to that for adult red kangaroos obtained previously (Dawson et al.,2000a) (N=7). Mean mass of the adults was 23.5±1.1 kg.

The effect of temperature on ventilation in the YAF red kangaroos was assessed by comparing responses at T_as of –5°C,25°C and 45°C using one-way repeated-measures ANOVA. Assumptions for ANOVA were tested by Levene's test for variances and Kolmogorov–Smirnov test for normality. To account for heterogeneity of variances in YAF, f_R, V̇_Iand E_O2, these data sets were log₁₀transformed (Zar, 1999). Allometrically adjusted data for the YAF and adult red kangaroos were compared at –5°, 25° and 45°C using a two-way repeated measures ANOVA. Assumptions for ANOVA were tested by Levene's test for variances and Kolmogorov–Smirnov test for normality. To account for heterogeneity of variances in the YAF versus adult f_R, V̇_I (for data using body-mass exponents of 0.67 or 0.75) and E_O2, these data sets were log₁₀ transformed(Zar, 1999). When ANOVA yielded significant differences, a Student–Newman–Keuls (SNK) test was performed to compare individual means. All statistical analyses were performed using Statistica 4.1 for Macintosh (Statsoft, Tulsa) and results are presented as means ± s.e.m.

Under cold (–5°C) and thermoneutral (25°C) T_as, the YAF had similar, stable T_bs ranging from 36.9°C to 37.4°C. At T_a–5°C, the YAF kangaroos'V̇_O2 was 2.6 times higher than it was under thermoneutral conditions at T_a25°C (P<0.05, N=7; Table 1). The YAF accommodated their higher V̇_O2at T_a –5°C by taking significantly deeper breaths (i.e. higher V_T) and more of them (i.e. higher f_R), leading to a 2.1 times higher V̇_I than at T_a 25°C (P<0.05, N=7; Table 1). The high V̇_O2 was also accommodated by a slight, though non-significant, rise in E_O2 of around 4.3% units (P=0.2; Table 1).

At 45°C, the YAF kangaroos were unable to maintain their T_b at the levels exhibited under thermoneutrality. Although stable, the YAF T_b at T_a45°C was significantly higher than that at T_a25°C, being 38.3±0.2°C (P<0.05, N=7; Table 1). At T_a 45°C, the YAF panted heavily, with respiratory rates 30 times higher than at T_a 25°C(Table 1). This was accompanied by significant reductions in V_T. There was also a significant reduction in E_O2 to around 1.5% at T_a 45°C, compared with 21% at T_a25°C (P<0.05; Table 1). Notably, V̇_Iin the YAF at T_a 45°C was 20 times that seen at thermoneutral T_as of 25°C (P<0.05; Table 1).

Using the exponents predicted from quarter-power scaling(West et al., 2000), we compared the YAF's ventilatory characteristics directly with those obtained from adult red kangaroos (Dawson et al.,2000a). The YAF red kangaroos showed respiratory responses to T_as of –5°, 25° and 45°C that were similar to those of mature females (Fig. 1) but at all T_as the YAF V̇_O2 (ml min^–1 kg^–0.75) was around 1.6 times that of the mature females (P<0.05; Fig. 1A). At a thermoneutral T_a of 25°C, we found no significant difference between the YAF and adult red kangaroos' allometrically adjusted f_R (breaths min^–1 kg^0.25; Fig. 1B), V_T (ml kg^–1; Fig. 1C) or V̇_I (ml min^–1 kg^–0.75; Fig. 1D), although differences were apparent at thermal extremes. The higher oxygen use by juveniles at 25°C compared to adults was accommodated by higher E_O2 (%), being 21.4±1.8% compared to 16.6±1.9% in the adults (P<0.05; Fig. 1E).

At cold T_as of –5°C, both the YAF and mature female red kangaroos significantly increased V̇_O2(Fig. 1A) and V̇_I(Fig. 1D) by around 1.5 and 2.2 times, respectively, from thermoneutral levels. The YAF V̇_I at T_a–5°C was significantly higher than that of the mature females by a factor of 1.5 (Fig. 1D). This was associated with the YAF's V_T being 35% greater at T_a –5°C than it was at T_a25°C (P<0.01), and it was also 40% higher than that of the mature females (P<0.05; Fig. 1C). Furthermore, the mature females did not significantly increase V_T at T_a –5°C compared with that at T_a 25°C(Fig. 1; P>0.05). Notably, the YAF and adult red kangaroos showed distinctly higher f_Rs at T_a –5°C compared with those seen at T_a 25°C, by between 7.4 and 4.5 breaths min^–1 kg^0.25, respectively, and these differences showed a strong tendency for significance (P=0.06) in each case (see Fig. 1B).

At high T_as of around 45°C, well above the T_b of juvenile and adult red kangaroos, the V̇_I of YAF was over 20 times higher than at thermoneutral conditions and was over 1.9 times that of the mature females on an allometrically adjusted basis(Fig. 1D). There was no significant difference between the V_T of the YAF and the mature female kangaroos at 45°C (Fig. 1C). Therefore, the higher rate of V̇_I exhibited by the YAF at T_a 45°C compared with mature females(Fig. 1D) was due solely to the YAF's higher f_R (Fig. 1B), which exceeded 320 breaths min^–1 on average(Table 1), compared with 150 breaths min^–1 for the mature females.

Munn and Dawson (Munn and Dawson,2001) found that the resting metabolic rate (ml O₂min^–1 kg^–0.75) of YAF red kangaroos was around 1.5 times that of fully mature, non-reproductive females, indicating that juvenile red kangaroos have intrinsically higher rates of energy turnover. We have re-iterated that finding here, placing our study of their ventilation patterns within the appropriate context. Juvenile mammals generally have higher basal metabolic rates than adults, due mainly to the extra energy turnover associated with growth(Robbins, 1993; Munn and Dawson, 2003). Here,we have shown that the YAF's higher V̇_O2 was not accommodated simply through higher f_R or even a higher overall V̇_I but was accommodated by a higher level of oxygen being extracted from each breath(i.e. a higher E_O2; Fig. 1E). It is generally assumed that E_O2 is a fairly constant fraction of V̇_I across adult animals from different mammal groups, but we have shown that E_O2 can vary with animal age. This is reminiscent of the conclusion of Chappell et al.(Chappell et al., 2003) that E_O2 decreased with age in deer mice. Chappell's results (Chappell et al., 2003)were most likely due to senescence, but the higher E_O2s reported here for the very young animal deserves further investigation. Differences in E_O2 can arise from either changes in ventilation/perfusion matching or differences in the oxygen content of mixed venous blood because arterial blood is nearly 100% saturated under non-pathological conditions (Bartels,1971; Withers,1992). Thus, an increased E_O2reflects either improved ventilation/perfusion matching or increased O₂ unloading from haemoglobin in tissues. Whichever the reason, it points to differences in the handling of oxygen with age in these large macropods.

Our red kangaroos showed typical mammalian responses to the higher O₂ demands imposed by cold T_as(Fig. 1A). Both the YAF and mature females increased V̇_O2 by around 2-fold at T_a –5°C compared with that at T_a 25°C. In both age groups, the higher V̇_O2 at T_a –5°C was met largely by increases in V̇_I. In the adult animals, a significant increase in E_O2 also contributed(Fig. 1E), which is a typical mammalian response to cold exposure(Mortola and Frappell, 2000). That the YAF's E_O2 did not increase at T_a –5°C relative to that seen at T_a 25°C, as was the case for the adults, suggests that they may be operating near their maximum level of extraction even under standard conditions. Therefore, unlike adult kangaroos, the YAF would appear to have little reserve capacity in their E_O2potential to ameliorate the extreme cold(Fig. 1E), and their higher V̇_O2 at T_a –5°C was met solely through increased V̇_I(Table 1). Notably, we were unable to push our juveniles (or adults) to the limits of their abilities, as indicated by the maintenance of constant T_bs at all T_as; further research is required to determine their maximum levels of E_O2 at even lower T_as or in response to exercise. Nonetheless, the higher V̇_I of the YAF at a T_a of –5°C, compared with that at T_a 25°C, was accomplished by an increase in V_T, while V_T of the adults remained unchanged between thermoneutral and cold T_as. Therefore,the juvenile kangaroos' ventilation system needed to work harder than that of adults at T_a –5°C in order to maintain T_b. Arid-zone kangaroos routinely face such cold conditions, particularly in winter when T_a regularly falls below freezing at night(http://www.bom.gov.au/climate/averages/– date viewed 01/12/2006).

In addition to managing homeostasis during cold conditions, free-ranging red kangaroos often encounter particularly hot conditions. During summer months, for example, the average incident radiation across much of the red kangaroo's range exceeds 1000 W m^–2(Dawson, 1995), and they experience average daily T_as of 30 to >45°C(http://www.bom.gov.au/climate/averages/– date viewed 01/12/2006). Munn and Dawson(Munn and Dawson, 2001) and Dawson et al. (Dawson et al.,2000a; Dawson et al.,2000b) found that both YAF and mature female red kangaroos,respectively, showed typical mammalian responses to high T_as of ∼45°C. These principally included increases in evaporative cooling through both cutaneous (i.e. licking) and respiratory(i.e. panting) routes. Notably, panting was the major source of evaporative cooling under hot conditions for the adult red kangaroos, accounting for∼60% of total evaporative heat loss(Dawson et al., 2000b). Mature kangaroos do not sweat in response to thermal heat loads (although they do in response to exercise) (Dawson et al.,1974), and so the remaining portion of their evaporative heat loss at T_a 45°C can be attributed to insensible cutaneous evaporation and licking of the body surface, especially the forearms(Needham et al., 1974; Dawson et al., 2000b). Conversely, the YAF red kangaroos relied equally on cutaneous and respiratory evaporative cooling to maintain T_b at T_a 45°C (Munn and Dawson, 2001). We do not know if the YAF were sweating. Sweating in adult kangaroos occurs only during exercise, when T_bincreases markedly. But if there is a threshold T_b above which sweating is initiated in resting kangaroos, then perhaps the YAF reached that T_b while the adults did not. Juvenile T_b reached 38.3°C while adults reached only 36.8°C at T_as around 45°C (see Munn and Dawson, 2001). By weaning age, when they had increased to a body mass of around 11–12 kg,the juveniles became more reliant on respiratory evaporative cooling at high T_as (Munn and Dawson,2001), suggesting that the smaller YAF kangaroos' ventilatory system alone was not sufficient to support the higher rates evaporation required to maintain T_b, thereby necessitating the higher levels of cutaneous evaporation observed. The data presented here support this conclusion, and at a T_a of 45°C, well in excess of T_b, there were indications that the YAFs' ventilatory system was working harder than that of adults. In particular, at T_a 45°C, the YAF red kangaroos showed one of the highest f_Rs that we are aware of for any mammal, once body mass was taken into account. On an allometric basis, the fastest previously recorded f_Rs were 189 breaths min^–1kg^0.25 for an exercising 2 g Etruscan shrew (Suncus etruscus) (Jürgens et al.,1996) and 253 breaths min^–1 kg^0.25 for the masked shrew (Sorex cinereus)(Morrison et al., 1959) (but see Jürgens et al.,1996). These values are less than half our measured f_R of 560 breaths min^–1 kg^0.25(i.e. 323 breaths min^–1; Table 1) for the heat-stressed YAF red kangaroos at an average body mass of 9 kg.

In this study we have focussed on scaling exponents empirically derived for adult mammals (Stahl, 1967; Dawson and Needham, 1981; Frappell and Baudinette, 1995; Withers et al., 2006) and that are consistent with quarter-power scaling(West et al., 2000). However,we recognise that there is considerable controversy about the appropriate scaling exponent for metabolic rate in mammals (e.g. Dodds et al., 2001; Kozlowski and Konarzewski,2004; West and Brown,2005; White and Seymor,2005). For example, a body-mass exponent of 0.67 has been suggested as more appropriate for both intra- and inter-specific comparisons(Heusner, 1991; White and Seymour,2003). Choosing an appropriate exponent is further complicated when juvenile animals are considered(Wieser, 1984), but our objective was to examine whether juveniles were small adults in terms of ventilatory physiology (Munn and Dawson,2001). We therefore chose an exponent of 0.75 for comparisons of V̇_O2 and V̇_I, which is consistent with adult data (Stahl, 1967; Dawson and Needham, 1981; Frappell and Baudinette, 1995; Withers et al., 2006). Our results, however, were consistent both in pattern and implication when we compared V̇_O2s and V̇_Is of the YAF and adult kangaroos using a body-mass exponent of 0.67 rather than 0.75(Table 2). Specifically, using a scaling exponent for body mass of 0.67, we found that the YAFs' resting V̇_O2 (at T_a 25°C) was still 1.4 times that of adult animals(P<0.01; Table 2). We are unaware of alternative scaling exponents for V_T, f_R or E_O2, but regardless of which body-mass exponent is used for V̇_O2 and V̇_I, the theoretical and measured body-mass exponent for E_O2 should remain constant at zero. In other words, regardless of the choice of exponent,our conclusion that the juvenile kangaroos meet higher resting oxygen consumptions by extracting more oxygen per breath is supported. Moreover, even with the more conservative body-mass exponent of 0.67, the YAF kangaroos had significantly higher V̇_O2s and V̇_Is compared with those of adults at T_a –5°C and 45°C(Table 2).

There are two primary conclusions that can be drawn from this study. Firstly, in terms of their f_R, V_T and V̇_I, the ventilatory system of juvenile red kangaroos at thermoneutral T_as was not significantly different from that expected for an adult marsupial of similar body mass. However, the YAF did have a higher E_O2 (%) at T_a 25°C,compared with adults, which explains how they meet their comparatively higher O₂ demands at this T_a. How this may be involved with active growth, which is usually cited as the main reason for juveniles having higher V̇_O2s, relative to adults (Robbins, 1993), is unknown. The second conclusion from this study is that, in the face of thermal extremes, the juvenile kangaroo ventilatory system must work considerably harder than that of adult animals to maintain heat balance. Overall, juvenile kangaroos appear more sensitive to extreme conditions, not only with respect to long-term stresses, such as food limitation(Munn and Dawson, 2006), but also to short-term extremes, such as severe cold or heat (present study)(Munn and Dawson, 2001). Short-term stress has been implicated in the mortalities of other large herbivores, such as caribou (Rangifer tarandus), where severe winter storms, rather than starvation, appeared to be the immediate cause of mortality in animals that were also exposed to long-term malnutrition(Dau, 2005). It is therefore important that these facets of juvenile mammal physiology be taken into account when considering the possible impacts of a changing climate. Currently, the bulk of models that evaluate species' extinction risks in relation to climate change often neglect the potential impacts on juvenile survival (e.g. Thomas et al.,2004; Humphries et al.,2004). Variation in juvenile survival, however, is particularly important for large mammalian herbivores from a range of climate and habitat types (Gaillard et al.,1998).

An Australian Research Council Postgraduate Award supported A.J.M. This research was in part supported by a grant from the Australian Research Council to T.J.D. Kangaroos were held under a licence from the New South Wales National Parks and Wildlife Service, and the University of New South Wales Animal Care and Ethics Committee gave approval for this project. Prof. Mark Chappell kindly provided Warthog Lab helper/analysist software for respirometry and plethysmography. Special thanks must go to the animal carers who participated in this project, especially Anne Williams.