Developmental allometry of pulmonary structure and function in the altricial Australian pelican Pelecanus conspicillatus
1 Environmental Biology, University of Adelaide, Adelaide, SA 5005,
Australia
2 Anatomy and Histology, Flinders University of South Australia, Adelaide,
SA 5001, Australia
3 Cardiac Physiology, National Cardiovascular Center, Suita, Osaka, Japan
565-8565
* Author for correspondence (e-mail: roger.seymour{at}adelaide.edu.au)
Accepted 4 May 2004
![]() |
Summary |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: bird, lung, juvenile, development, respiration, morphometry, diffusing capacity, symmorphosis
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The concept of symmorphosis is that the maximum functional capacity of
parts of a system should be matched; there are no over- or under-designed
links in the chains. The hypothesis was named and chiefly developed by Taylor
and Weibel (1981) in their
consideration of the oxygen cascade in exercising mammals. Their approach was
to test the hypothesis by measuring structure and function of each level in
mammals and compare them interspecifically with allometric techniques. They
demonstrated that the oxygen transport capacities are matched to a large
extent at each level from the lung to the mitochondrion, except for the
morphometric pulmonary diffusing capacity, which seemed to be somewhat
over-endowed in larger species (Taylor et
al., 1989
). However, physiological pulmonary diffusing capacity
was subsequently shown to be well matched to the maximum capacity of the
cardiovascular system (Hsia,
1998
). Interspecific analyses have also been carried out on
pulmonary morphometry in birds (Maina,
1993
,
1998
;
Maina et al., 1989
), but never
explicitly compared to maximum metabolic rates of the same species.
The symmorphosis paradigm is an ideal that may or may not be realised by
natural selection. All studies addressing the question have been directed at
adult animals. Whether it is evident in developmental stages is not known.
Therefore we have begun to evaluate the changes in transport capacities of the
gas exchange apparatus and the cardiovascular system that occur during
development in birds. We are unaware of any studies that measure the
development of pulmonary diffusing capacity in relation to maximum oxygen
uptake rate in paranatal and juvenile birds. The ontogeny of pulmonary
structure has been measured in domestic turkeys Meleagris gallopavo,
including morphometrics of compartment volumes and surface areas
(Timmwood et al., 1987), but
the bloodgas tissue thickness was not provided, so diffusing capacity
cannot be calculated, and there are no data on the metabolic scope of
developing turkeys. As part of a larger study on the development of the
cardiovascular and respiratory systems in birds of different hatchling
maturity, we present data on pulmonary diffusing capacity in relation to
maximum aerobic metabolism in the altricial Australian pelican. The Order
Pelicaniformes produces the largest altricial hatchlings that grow to become
some of the largest extant birds capable of flight. Information on the
energetics and respiration of pelican embryos has been published
(Pearson et al., 2002
), and a
fuller morphometric analysis of embryonic and post-hatching lung development
is to be published elsewhere. Here we focus on the question of symmorphosis,
and use the same allometric approach as Taylor et al.
(1989
).
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Respiration
Metabolic rate was determined by open-flow respirometry, following standard
techniques (Withers, 1977).
Air flow rates through the systems were measured with mass-flow meters or
controllers (Sierra Instruments Mass-Trak, CA, USA) and they varied from 350
ml min1 in hatchlings up to 18 l min1 in
the largest birds. The air leaving the system was scrubbed with Drierite (W.
A. Hammond, Xenia, Ohio, USA), soda lime and Drierite columns, and delivered
by a sub-sampling circuit either to a fuel-cell oxygen analyser (David Bishop
Instruments 280/0427 Combo, UK), thermostatted at 36°C, or to a
solid-state zirconium oxygen analyser (Ametek S-3A/I, Pittsburgh,
Pennsylvania, USA). Both analysers were calibrated with pure nitrogen and dry,
CO2-free air. The outputs from these instruments were recorded with
A/D converters and recording software (Sable Systems Universal Interface and
DATACAN v5.2, Las Vegas, Nevada, USA).
For resting metabolism, birds were placed in darkened, plastic containers of appropriate size, inside a constant temperature cabinet or room at 36°C for hatchlings or 27°C for older birds. Metabolic rate was measured until it dropped to stable values, which occurred after 2030 min. The resulting minimum values are not basal metabolic rates, because the birds were not fasting, they were growing, and they were measured during the day.
Two approaches were used in attempts to elicit maximum aerobic metabolism, cold exposure and exercise. Neither treatment increased the metabolic rates of hatchlings significantly, because exposure to cold temperatures caused an immediate drop in body temperature and metabolic rate, and the birds were incapable of any kind of energetic movement. Birds from 11 to 21 days old (crèche phase) were also incapable of exercising, but they exhibited strong shivering and metabolic rate increase when exposed to low temperature. Respiration was therefore increased to the highest obtainable value by abruptly lowering ambient temperature to about 5°C in the respirometry chamber and measuring oxygen consumption until it peaked and body temperature (measured via a thermocouple about 2 cm in the cloaca) began to fall. This procedure required about 1530 min.
Larger birds from the field produced visible shivering and large metabolic
responses to cold, but did not produce maximum metabolic rate on cold
exposure, including trials with 20% oxygen in helium at about 0°C, which
greatly increases thermal conductance
(Hinds et al., 1993). However,
they were capable of running on a treadmill and swimming in a flume, and the
resulting values were 1.23.9 times higher than the maxima derived from
cold exposure. Clear plastic, flow-through masks, taped loosely around the
neck, were used during exercise. Exercise was induced by running on a
treadmill at speeds up to 1.4 m s1 and by swimming in a
flume up to 1 m s1, all at room temperature. The birds were
encouraged to exercise by holding or gently pulling the feathers on the tail.
Exercise periods were variable, from 210 min. Runs in which the
respirometry values were reasonably stable for at least 2 min during
continuous activity were considered successful. Values were averaged during
stable periods lasting 26 min. The values for running and swimming were
similar, and without consistent directional bias. In all cases, reported
values are averages from the two exercise methods.
Pulmonary morphometry
Lung morphometry was carried out as part of a larger study on the
developmental anatomy of embryonic and juvenile pelicans. The birds were
killed by rapid asphyxiation with pure carbon dioxide flowing through their
chamber or head mask. This technique also facilitated the infusion of fixative
into the airways, because the gas was absorbed into the tissues. Lungs were
fixed with 2.5% glutaraldehyde in a 0.01 mol l1 phosphate
buffer (pH 7.4, 350 mOsm). The birds were placed in a supine position and the
fixative was introduced via a cannula in the trachea with the
reservoir located 20 cm above the sternum. The trachea was ligated below the
cannula when the flow of fixative stopped. The lungs were removed from the
thorax after 2 h. The fixed lungs were sliced in the plane of the costal
grooves from apex to base into 1020 slices, depending on lung size.
Lung volume (VL), including air and tissue, was determined
using the Cavalieri method of point counting on slices from the left and the
right lungs (Howard and Reed,
1998). Total lung volume was obtained by summing the volumes of
the two lungs.
The right lung was used for transmission electron microscopy. Ten pieces of
tissue, 2 mm in diameter, were sampled using the independent area-weighted
periodic sampling technique described by Cruz-Orive and Weibel
(1981). Lung pieces were
stained with osmium, en bloc stained with uranyl acetate, dehydrated
in an ascending series of alcohol, cleared in propylene oxide and embedded in
Durcupan embedding resin (Fluka Chemie AG; Sigma Chemicals, St Louis, MO,
USA). Gold to silver sections with sides of about 1.5 mm were cut with a
diamond knife with the ultramicrotome set to cut at 100 nm. Sections were
collected on copper thin bar 200 square mesh grids and stained with lead
citrate prior to viewing in a Joel JEM 1200-EX (Tokyo, Japan) transmission
electron microscope at an acceleration voltage of 80 kV.
Ten sections from each bird lung were viewed at 2000x magnification
and sampled using the systematic area-weighted quadrats subsampling technique
(Muller et al., 1981). Six
fields were sampled from each section, giving 60 fields per lung. Images of
the sampled fields were imported into CorelDraw 9 (Corel Corporation), a Merz
grid was superimposed on the image, and point and intersection counts were
carried out.
The pertinent data for the present analysis are the total surface area and
the harmonic mean thickness of the bloodgas tissue barrier, where the
blood capillaries oppose the air capillaries. The gas exchange surface density
of the blood capillaries (SVc) was determined by counting
intersections that ran from air to blood and calculated using
SVc=2I/L, where I is the number
of intersections counted and L is the total length of the test
system. The surface area of the bloodgas barrier,
St, was calculated from the reference volume of the
exchange tissue (Maina et al.,
1989).
The harmonic mean thickness of the bloodgas tissue barrier
(ht,) was calculated as
2/3[n/
(1/L)]M1, where
n is the number of measured intercepts,
(1/L) is the
sum of the reciprocal of the intercept lengths and M is the final
magnification (Maina et al.,
1989
). The ratio St/
ht,
multiplied by Krogh's coefficient of oxygen diffusion, provides the anatomical
(tissue) diffusing capacity of the bloodgas barrier. The value for
Krogh's coefficient was assumed to be 4.1x1010
cm2 s1 mbar1, to be comparable
to the study of Maina et al.
(1989
).
Statistics
Statistics include linear regressions (model 1, least squares) on
log10-transformed data, excluding the single adult bird. The units
of mass are grams. Regression statistics include coefficients of determination
(r2) and 95% confidence intervals of the slope (CI),
calculated in Excel with StatistiXL add-in software
(statistixl.com),
as were tests for relationships among residuals and for outlier points.
Differences in regression slopes and elevations were tested with analysis of
covariance (ANCOVA) according to Zar
(1998). Where slopes were
significantly different, the regions of significantly different elevations
were identified with the JohnsonNeyman technique
(White, 2003
). Means of other
values are given with 95% confidence intervals.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Resting and maximal metabolic rates diverged in birds aged 11 days or older. The increases in resting metabolic rate were not linear on a double log plot, because 11- and 21-day-old birds had relatively high values. Maximal rate, however, was linear over two orders of magnitude body mass, with a slope of 1.28±0.07 CI (Fig. 1). At the maximum body mass of 8834 g, the regression equations indicate a resting metabolic rate of 144 ml min1 and a maximum metabolic rate of 436 ml min1. This indicates a total metabolic scope of 292 ml min1 or a factorial scope of 3.0. There was no apparent difference in maximal rate of the adult and the juveniles of similar mass.
Pulmonary morphometry
Lung volume data were available for 13 post-hatch pelicans
(Fig. 2). There was a strong,
linear relationship between log-transformed lung volume and body mass, and the
slope of the relationship (1.05±0.06 CI) was not significantly greater
than 1.0. The value from the adult bird was not exceptional.
|
The surface area of the bloodgas barrier increased allometrically
with body size with a slope of 1.25±0.15 CI
(Fig. 3). There appeared to be
further increase in surface area in the single adult, and the difference was
significant (Outlier test: Studentised residual 2.66 > critical value 2.23
at =0.05).
|
Harmonic mean thickness of the bloodgas tissue barrier did not
change significantly in the juvenile pelicans during development, the slope
being 0.02±0.08 CI (Fig.
4). However, the value from the adult was significantly lower than
the other juveniles of similar body mass (Outlier test: Studentised residual
4.58 > critical value 2.23 at =0.05).
|
The anatomical diffusing capacity of the bloodgas tissue barrier was
significantly allometric, with a slope of 1.23±0.20 CI. Because of the
dramatic decrease in barrier thickness, diffusing capacity in the adult was
significantly higher than values for the other juveniles of similar mass
(Outlier test: Studentised residual 3.78 > critical value 2.23 at
=0.05).
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
The maximum metabolic rates we recorded during exercise in the 8.8 kg
juvenile (402 ml min1) and the 7.2 kg adult (357 ml
min1) were apparently sufficient to power flight. The cost
of flight for such a large bird has not been measured, but it is estimated
that the largest flying bird would weigh about 12 kg when the power required
for flight is equal to the available power
(Norberg, 1996). This point is
115 W, or 345 ml O2 min1, assuming 20 J
ml1. Although the cardio-pulmonary system may have the
capacity to deliver enough oxygen for powered flight, it is unlikely that the
oldest juveniles could sustain flight. Indeed, older juveniles (partial
primary feather development) that were observed at the colony were only
capable of short gliding flight, and then only after long run-ups on open
ground. We judge that the juveniles we sampled had never flown sustainably. It
appears that the capacity of the oxygen delivery system is developed in
advance of their need to use it.
Diffusing capacity of the bloodgas barrier increases throughout most of post-hatch development by an increase of gas-exchange surface area (Fig. 3), rather than decreased blood-gas barrier thickness (Fig. 4). However, thickness decreases significantly, and surface area further increases, in the adult. This bird is the only one in this study with a barrier thickness in the range for other adult birds (Fig. 4). Although only one adult was available to us, it was a statistically significant outlier that leads to the conclusion that the gas-exchange barrier decreases in thickness and increases in area during maturation of sedentary juveniles to volant adults, although body mass and lung volume do not change.
It is instructive to compare the allometries of growing pelicans and adults
of other species to distinguish the effects of development and inherent
scaling relationships of the products of development.
Fig. 1 compares maximum
metabolic rate in pelicans with 33 species of flying birds
(Norberg, 1996) and nine
species exposed to cold (Hinds et al.,
1993
). Analysis of covariance reveals significant differences in
slope between growing pelicans and both studies of adult birds
(P<0.0001). While the slope for pelicans is greater than unity, it
is significantly less in adult species, indicating that the mass-specific
maximum rate decreases in larger species. The JohnsonNeyman technique
shows that the data for developing pelicans were significantly lower than for
cold-exposed adult birds at body masses less than 2.2 kg, that is, the
pelicans up to and including 21 days old. The technique also showed that
values from larger pelicans, including the adult, lie between extrapolated
regressions for cold-exposed and flying birds, but are significantly lower
than flying birds.
Fig. 5 shows a similar
comparison of diffusing capacity of the bloodgas barrier between
pelicans and 26 species of adult birds from a morphometric study by Maina et
al. (1989). (There are slight
differences between our regression equations for their data, possibly because
of rounding errors in their tabled values.) Once again, ANCOVA reveals
significantly different slopes (P=0.0003) and the
JohnsonNeyman technique identifies the pelican data as significantly
lower at all body masses.
The striking similarity of allometric slopes of maximum metabolic rate (b=1.28) and diffusing capacity of the bloodgas barrier (b=1.23) is indicative of a parallel development of these variables during post-hatching growth of pelicans (cf. Figs 1 and 5). These results suggest either that pulmonary diffusion limits maximum metabolic rate or that the oxygen delivery system is to some extent symmorphic. If the cardiovascular system or processes at the cellular level limited oxygen uptake, then one would expect lower slopes for metabolic rate than for diffusing capacity. A similar congruence is apparent in adult birds in which the slope for metabolic rate during flapping flight (b=0.80) is almost identical to that of pulmonary tissue diffusion capacity (b=0.79) (cf. Figs 1 and 5). The allometry of maximum metabolic rate in birds is not known.
Our conclusion that bloodgas barrier diffusing capacity is linked to
maximum oxygen uptake should be viewed with some caution, however, for three
reasons. First, although similarity in allometric slopes is consistent with
symmorphosis, allometry cannot prove it, because it does not explicitly
measure capacity matching in the oxygen cascade, and other explanations are
possible (Dudley and Gans,
1991). Secondly, the 95% confidence interval of the slope of
diffusing capacity is broad (±0.20) in this study. Third, there is no
relationship between the residuals of diffusing capacity and maximum metabolic
rate in pelicans (r2=0.009). If pulmonary diffusion
limited oxygen uptake, one would expect birds with higher diffusing capacity
to have higher maximum metabolic rate. Until there are other studies on the
development of the oxygen delivery system in birds, we can only tentatively
conclude that the data from Australian pelicans are consistent with the
concept of symmorphosis during post-hatching development in birds.
![]() |
Critique |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
While the morphometric techniques are fairly standard and comparable,
measurements of resting and maximal metabolic rates are fraught with
difficulties. How can we be certain that our measurements accurately
represented the total range of aerobic metabolic rate? We did not measure
blood lactate levels to determine whether the birds had exceeded their aerobic
limit. However, we have other evidence that they were reaching maximum values.
First, they exhibited signs of fatigue during exercise, such as slower speeds,
poorer coordination, increasing reluctance to continue, and rapid breathing.
These occurred after at least 2 min, which would have given time for them to
achieve a steady state and the respirometry system to stabilize. Second, there
was good congruence between data during running and swimming. Third, the
values from the largest pelicans exceed the extrapolation from adult birds
exposed to cold and converge with the extrapolation from flying species
(Fig. 1). Bynecessity, this
study involved inducing maximum metabolic rate by cold exposure and exercise.
It is well known that in adult birds, exercise results in metabolic rates
about three times higher than does cold
(Hinds et al., 1993). However,
this difference may not occur at all developmental stages. Hatchling pelicans
were not capable of sustained exercise or thermoregulation. Shivering in young
juveniles produced higher metabolic rates than we could induce by exercise,
but exercise became more effective in the oldest pelicans. In the five
juveniles for which we have data on both exercise-induced and cold-induced
metabolism, the ratio (exercise/cold) increased from 0.9 in a 4.2 kg juvenile
to 3.9 in the 7.2 kg adult. Therefore our birds were reaching the threefold
difference expected for adults.
The factorial metabolic scope increased from 1 at hatching to 3.0 in the
largest juveniles (Fig. 1).
This value is considerably lower than metabolic scopes in adult birds that
range from 4.3 to 6.5 in response to cold
(Hinds et al., 1993) and up to
14 during flight (Norberg,
1996
). Part of the explanation may be that the oxygen delivery
system is simply not as developed in the juveniles as would be expected in the
adult, or that resting metabolic rate is not basal. Evidence for the latter is
that the expected minimal metabolic rate for birds equals
0.149M0.68 (Frappell et
al., 2001
), which is 72 ml min1 for a 8834 g
bird. Our resting value of 144 ml min1 is 2 times higher,
and is doubtless influenced by the heat increment of digestion (specific
dynamic action) of the birds' high protein diets and their rapid growth rate.
Because resting metabolism was not the focus of the study, failure to achieve
conditions for measuring basal metabolic rate does not compromise the
conclusions.
We are reasonably confident that no treatment could have elicited higher
metabolic rates in any of the juveniles, but we cannot dismiss the possibility
that running and swimming were insufficient to elicit maximum values in the
adult bird that was capable of flight. Its flight muscles were not exercised.
Given that this bird had a tissue diffusing capacity significantly higher than
the juveniles, but did not show a higher metabolic rate, it is quite likely
that metabolic rate could be higher during flight than during running or
swimming. Furthermore, aerobic metabolic rate during sustained flight is not
usually maximal (Norberg,
1996). To understand the development of the gas transport systems
in pelicans more fully, it would be desirable to examine flight respiration
and pulmonary morphometric development between juveniles and similarly sized
adults, but this would have to be done in relation to age rather than body
mass.
![]() |
Acknowledgments |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Crapo, J. D. and Crapo, R. O. (1983). Comparison of total lung diffusion capacity and the membrane component of diffusion capacity as determined by physiologic and morphometric techniques. Resp. Physiol. 51,183 -194.[CrossRef][Medline]
Cruz-Orive, L. M. and Weibel, E. W. (1981). Sampling designs for stereology. J. Microsc. 122,235 -257.[Medline]
Dudley, R. and Gans, C. (1991). A critique of symmorphosis and optimality models in physiology. Physiol. Zool. 64,627 -637.
Frappell, P. B., Hinds, D. S. and Boggs, D. F. (2001). Scaling of respiratory variables and the breathing pattern in birds: an allometric and phylogenetic approach. Physiol. Biochem. Zool. 71,75 -89.[CrossRef]
Hinds, D. S., Baudinette, R. V., Macmillen, R. E. and Halpern,
E. A. (1993). Maximum metabolism and the aerobic factorial
scope of endotherms. J. Exp. Biol.
182, 41-56.
Howard, C. V. and Reed, M. G. (1998). Unbiased Stereology. Three-Dimensional Measurement in Microscopy. Oxford: BIOS Scientific Publishers.
Hsia, C. C. W. (1998). Limits of adaptation in pulmonary gas exchange. In Principles of Animal Design (ed. E. R. Weibel, C. R. Taylor and L. Bolis), pp.168 -176. Cambridge: Cambridge University Press.
Maina, J. N. (1993). Morphometries of the avian lung: the structuralfunctional correlations in the design of the lungs of birds. Comp. Biochem. Physiol. 105A,397 -410.
Maina, J. N. (1998). The lungs of flying vertebrates birds and bats: is their structure optimized for this elite mode of locomotion? In Principles of Animal Design (ed. E. R. Weibel, C. R. Taylor and L. Bolis), pp.177 -185. Cambridge: Cambridge University Press.
Maina, J. N., King, A. S. and Settle, G. (1989). An allometric study of pulmonary morphometric parameters in birds, with mammalian comparisons. Phil. Trans. R. Soc. Lond. B 326,1 -57.[Medline]
Muller, A. E., Cruz-Orive, L. M., Geh, P. and Weibel, E. R. (1981). Comparison of two subsampling methods for electron microscopic morphometry. J. Microsc. 123, 35-39.[Medline]
Norberg, U. M. (1996). Energetics of flight. In Avian Energetics and Nutritional Ecology (ed. C. Carey), pp. 199-249. New York: Chapman & Hall.
Pearson, J. T., Seymour, R. S., Baudinette, R. V. and Runciman,
S. (2002). Respiration and energetics of embryonic
development in a large altricial bird, the Australian pelican (Pelecanus
conspicillatus). J. Exp. Biol.
205,2925
-2933.
Powell, F. L. and Scheid, P. (1997). Lung structure and function: Birds. In Comparative Pulmonary Physiology: Current Concepts (ed. S. C. Wood), pp.237 -255. New York: Marcel Dekker.
Spicer, J. I. and Burggren, W. W. (2003). Development of physiological regulatory systems: altering the timing of crucial events. Zoology 106, 91-99.
Taylor, C. R. and Weibel, E. R. (1981). Design of the mammalian respiratory system. I. Problem and strategy. Resp. Physiol. 44,1 -10.[CrossRef][Medline]
Taylor, C. R., Weibel, E. R., Karas, R. H. and Hoppeler, H. (1989). Matching structures and functions in the respiratory system. In Comparative Pulmonary Physiology. Current Concepts, vol. 39 (ed. S. C. Wood), pp.27 -65. New York, Basel: Marcel Dekker, Inc.
Timmwood, K. I., Hyde, D. M. and Plopper, C. G. (1987). Lung growth of the turkey, Meleagris gallopavo: I. morphologic and morphometric description. Am. J. Anat. 178,144 -157.[Medline]
Weibel, E. R. (1984). The Pathway for Oxygen Structure and Function in the Mammalian Respiratory System. Cambridge, Massachusetts: Harvard University Press.
White, C. R. (2003). Allometric analysis beyond heterogeneous regression slopes: use of the Johnson-Neyman technique in comparative biology. Physiol. Biochem. Zool. 76,135 -140.[CrossRef][Medline]
Withers, P. C. (1977). Measurement of
O2,
CO2 and
evaporative water loss with a flow-through mask. J. Appl. Physiol.:
Resp., Environ. Ex. Physiol. 42,120
-123.
Zar, J. H. (1998). Biostatistical Analysis. Englewood Cliffs, New Jersey: Prentice Hall.