Maturation of cardiovascular control mechanisms in the embryonic emu(Dromiceius novaehollandiae)

Our understanding of cardiovascular regulation during avian development is primarily derived from embryonic chickens. However, the degree to which the developmental patterns in embryonic chickens may apply to other bird species is unknown. Given the diversity within this group of vertebrates, differences in physiological development among species might be anticipated. This point is supported by the differences in autonomic tone determined in two different strains of adult chickens (Kuo et al.,2001). However, Pearson and Tazawa(1999) recently compared heart rate changes during avian ontogeny and suggested that many altricial species may share similar patterns of cardiovascular maturation during early development. They documented the relationship of changes in heart rate with embryonic mass and incubation length of several altricial bird species. Given this relationship, the activation of the underlying mechanisms that regulate heart rate (as well as other cardiovascular parameters) may also occur at similar times during maturation among avian species.

Studies have established that chickens exhibit several components of the mature cardiovascular regulatory system early in ontogeny(Ignarro and Shideman, 1968; Pappano, 1977). These include responsive adrenergic and cholinergic receptor populations in the developing cardiovascular system as well as anabolic and catabolic enzymes for catecholamines and acetylcholine (Berry,1950; Zacks, 1954; Ignarro and Shideman, 1968). Despite the presence of regulatory components from both arms of the autonomic nervous system, at day 12 [60% of the incubation period (60%I)] embryonic white leghorn chickens (Gallus gallus) possess only a tonic adrenergic stimulation of the cardiovascular system(Crossley and Altimiras, 2000). Several factors may dictate the timing of onset and subsequent effectiveness of cardiovascular regulatory mechanisms during chicken development. However,due to the lack of data from any avian species other than chickens,comparative hypotheses focusing on cardiovascular control during development cannot be evaluated.

Ratite birds are a useful group for addressing evolutionary hypotheses focusing on developmental cardiovascular physiology in birds. Ratites represent a distinct lineage of avian vertebrates, separated from other groups. Thus, the study of ratites could further the understanding of cardiovascular control patterns that are conserved during development in avian lineages (Burggren and Crossley,2002).

The goal of this study was to characterize the development of cardiovascular control systems in the emu. We hypothesized that, given the taxonomic differences between emus and chickens, emus would possess different cardiovascular regulation mechanisms. Cardiovascular responses (change in heart rate and arterial pressure) to graded levels of hypoxia were used as indicators of the embryonic emus' ability to respond to the environment as well as to determine the timing of onset of a functional chemoreflex. Furthermore, cholinergic and adrenergic receptor antagonists were applied during the interval from 70%I to 90%I to assess the function of these receptors during this 30% developmental window of emu incubation.

Freshly laid emu (Dromiceius novaehollandiae) eggs (N=20)were collected from adult holding pens at the Cross Timbers Emu Ranch (Flower Mound, TX, USA) and refrigerated at 3°C for a maximum of one week to temporarily arrest development pending collection of a full cohort of eggs. All eggs were then transported to the University of North Texas and placed in a force ventilation incubator (G.Q.F. Model 1536E) at 36°C with a relative humidity of 30%. Eggs were removed from incubation for study on days 30, 35,40 and 45, which corresponded to 60%I, 70%I, 80%I and 90%I of a 50-day incubation period.

Eggs were placed in a water-jacketed temperature control dish for surgery. A 3 cm² window was cut in the shell to expose the vessels of the underlying chorioallantoic membrane (CAM). Both a CAM artery and vein were occlusively catheterized using heat-pulled polyethylene tubing (PE-90) filled with heparinized saline (0.9%) (Crossley and Altimiras, 2000).

Following catheterization, eggs were placed in a 1200 ml temperature-controlled chamber fitted with a lid containing multiple ports for externalization of catheters and inflow tubes for changing gas mixtures within the chamber. The arterial catheter was then passed through a port and connected to a pressure transducer (WPI) attached to a bridge amplifier (CB Sciences 400). Humidified 36°C air was continuously circulated into each experimental chamber during the period of study at a rate of 1.5 litres min^-1. This rate of gas flow was also used during the acute hypoxic response experiments. A data acquisitions system (Powerlab; AD Instruments,Colorado Springs, CO, USA) was used to collect the output signal at a sampling frequency of 100 Hz.

Eggs were equilibrated in their experimental chamber for 1 h to allow cardiovascular values to stabilize. Prior to experimental manipulations,baseline data were collected from 20 instrumented embryos for 30 min. Arterial pressure was recorded directly and heart rate determined from the blood pressure trace using a software tachograph. Two experimental series were conducted on each instrumented embryo.

20 embryos were used during this series. Each embryo was exposed to 15%O₂ for 5 min while changes in heart rate and arterial pressure were recorded. After a 30 min recovery period (a length of time that was sufficient for cardiovascular variables to return to control levels), the hypoxic procedure was then repeated with 10% O₂. This hypoxic protocol was completed in all of the embryos at each of the incubation intervals tested.

Following a 1.5 h recovery period, five embryos from three stages (70%I,80%I and 90%I; total N=15) from the hypoxic series study were used to assess the autonomic tone on the emu cardiovascular system. Three autonomic receptor antagonists were serially injected to determine the endogenous autonomic receptor (cholinergic and adrenergic) tone on heart rate and arterial pressure.

Following the hypoxic recovery period, a single dose (1 mg kg^-1)of the muscarinic antagonist atropine was injected to determine the resting cholinergic receptor stimulation (tone) on heart rate and arterial pressure. Once a new baseline for heart rate and pressure was established, a second drug, the β-adrenergic antagonist propranolol (3 mg kg^-1), was injected into each embryo. Heart rate and blood pressure changes were recorded and allowed to stabilize at a new resting level prior to the injection of theα-adrenergic antagonist phentolamine (3 mg kg^-1).

After the injection of phentolamine, blood samples were taken to determine catecholamine concentration in a subset of embryos. Blood samples(approximately 100 μl) were collected by allowing blood to flow freely from the arterial catheter into a 500 μl Eppendorf collection tube. Catecholamine blood samples were mixed with 5 μl of an EGTA/glutathione solution (0.2 mol l^-1/0.2 mol l^-1) to prevent catecholamine oxidation and immediately centrifuged at 13 700 g for 5 min. High-performance liquid chromatography (HPLC)analysis of plasma catecholamines was carried out as previously described by Fritsche and Nilsson (1990). Samples were maintained at -70°C until analysis was carried out (within one month). These embryos were then euthanized with an intravenous injection of sodium pentobarbital (50 mg kg^-1). All embryos were then weighed to the nearest 1.0 g todetermine wet mass of the animals(Table 1).

Non-parametric tests were conducted on the data due to small sample sizes and the comparison of percentage changes. A Wilcoxon non-parametric test was used to assess statistical significance of the responses to treatments(hypoxia and drug administration; atropine, propranolol and phentolamine) at each stage of incubation studied in the emu. A Mann–Whitney Utest on the percentage change in heart rate (fh) and mean arterial pressure (P̄) following each treatment was used to determine differences in responses between developmental intervals. Since repeated tests were carried out, thereby using the same data more than once, the fiduciary limit (P<0.05) was Bonferroni-corrected according to the number of times (2-3 times) each data set was used (thereby comparing 60%I to 70%I, 70%I to 80%I, 70%I to 90%I and 80%I to 90%I). This procedure was conducted to reduce the possibility of incorrectly finding a difference between incubation groups, i.e. a type 1 error. A Mann–Whitney U test was also conducted on the catecholamine levels using the same correction described above. The limited sample size prevented comparisons of the data acquired for the 60%I embryos. All data are shown as means ± 1 s.e.m. A fiduciary level of P<0.05 was used. Statistical analysis was conducted using Statistica (Statsoft, version 5.0).

Resting fh in embryonic emus decreased significantly from 171±5 beats min^-1 at 60%I to 154±6 beats min^-1 at 80%I and then remained constant during the rest of development (Fig. 1A). P̄a increased significantly from 1.2 kPa at 60%I to approximately 2.0 kPa at 70%I, showing no further significant increase during development (Fig. 1B).

Hypoxic exposure (both 10% and 15% O₂) significantly decreased fh at 60%I but had no effect from 70%I to 80%I. At 90%I,10% O₂ caused a significant increase in fh(Fig. 2A,C). The fall in fh at 60%I was accompanied by a reduction in P̄a during exposure to 15% and 10%O₂ (Fig. 2B,D). A significant decrease in P̄a was also recorded during 10% O₂ in 70%I embryos and during 15% O₂in 80%I embryos (Fig. 2B,D). However, hypoxia (10% and 15%) had no effect on P̄a in embryonic emus that had reached 90%I (Fig. 2B,D).

Fig. 3 represents the response of fh and P̄a of an embryonic emu at 90%I to injections of cholinergic and adrenergic receptor antagonists. Following cholinergic blockade, fh rose while P̄a was unchanged. β-adrenergic receptor blockade then caused a decrease in fh and an increase in P̄a. α-adrenergic receptor blockade caused both fh and P̄ to fall(Fig 3A,B).

At all incubation ages tested, fh significantly increased following cholinergic blockade: 70%I (increase of 12±3 beats min^-1), 80%I (increase of 20±3 beats min^-1) and with the greatest increase at 90%I (increase of 50±3 beats min^-1) (Fig. 4A). Cholinergic blockade had no effect on P̄a during this period(Fig. 4B).

β-adrenergic blockade with propranolol, following cholinergic blockade, caused changes in fh and P̄a at each measured age of emu development (Fig. 5A,B). Propranolol elicited a bradycardia that was significantly weaker at 70%I (-33 beats min^-1) compared with at 80%I (-43 beats min^-1) and 90%I (-46 beats min^-1) (Fig. 5A). Propranolol had the opposite effect on arterial pressure,which increased at 70%I (0.15 kPa), 80%I (0.3 kPa) and 90%I (0.9 kPa)(Fig. 5B). But although the absolute change in pressure caused by propranolol appeared greatest in the 90%I age group, the change in pressure relative to the resting arterial pressure for each age group was constant.

Injection of the non-specific α-adrenergic blocker phentolamine,after muscarinic and β-adrenergic blockade, produced a significant decrease in fh and P̄a(Fig. 3A,B). Phentolamine decreased fh by a similar amount in embryos at 70%I (-19 beats min^-1), 80%I (-28 beats min^-1) and 90%I (-25 beats min^-1) (Fig. 5C). Phentolamine injection also caused a significant decrease in P̄a at 70%I (-0.35 kPa), 80%I (-0.60 kPa) and 90%I (-1.50 kPa) (Fig. 5D). Again, while the absolute change in pressure caused by phentolamine appeared greatest in the 90%I age group, the change in pressure relative to the resting arterial pressure for each age group was constant.

Plasma concentrations of norepinephrine increased progressively with embryonic development to a maximum of 80.0±21.0 ng ml^-1 at 90%I (Table 2). The maximum epinephrine level was 53±24 ng ml^-1 at 90%I, but the levels did not differ statistically during incubation(Table 2).

This study was conducted to understand the development of cardiovascular regulation in embryonic emus and provide additional information about developmental cardiovascular physiology in birds. Basal cardiovascular function comes under tonic adrenergic and cholinergic control between 60%I and 70%I. The cardiovascular responses to hypoxia during embryonic development suggest that reflexive control systems may be operational in emus prior to hatching. When compared with the known cardiovascular regulation systems in embryonic white leghorn chickens, emu embryos possess some similar cardiovascular control mechanisms but there are important differences between these species during development.

The heart rate and arterial pressure responses to hypoxia and reaction to autonomic antagonists were conducted in all embryos tested at 70%I, 80%I and 90%I. A recovery period between the hypoxic exposures and autonomic blockade allowed blood chemistry [lactate, pH, oxygen partial pressure(P_O) and carbon dioxide partial pressure(P_CO)] changes associated with hypoxia to return to control values. The recovery period (1.5 h) was based on blood chemistry recovery profiles following 5 min hypoxic exposures (15% and 10%O₂) in embryonic white leghorn chickens. Following each hypoxic exposure in embryonic white leghorn chickens, blood parameters returned to near control values within 1 h (D. A. Crossley, unpublished data). Therefore,the time period allowed prior to the initiation of the autonomic blockade series (1.5 h) in embryonic emus was probably sufficient to allow blood chemistry to return to control levels.

The validity of utilizing a 5 min exposure to assess the cardiovascular response of embryonic emus also must be addressed from two views. The first is that the 5 min period would not allow the P_O to reach the desired level within the experimental chamber. The second is that the exposure period was sufficient to elicit a transient cardiovascular response only. While both of these issues are acknowledged, the flow rate of air into the chambers was sufficient to turn over the volume of gas to the new P_O level in approximately 30 s. This was independently verified prior to each study (D. A. Crossley et al., unpublished data). Therefore, the chamber that the embryo was in reached the desired P_O within the first minute. A 5 min exposure could also be questioned as insufficient in terms of allowing an equilibrium cardiovascular state to be reached by the embryo. The goal of these experiments was to determine the acute cardiovascular response to hypoxia in embryonic emus in an effort to study the cardiovascular chemoreflex and compare it with the response of embryonic white leghorn chickens. The changes in the acute hypoxic response (Fig. 2) and differences from that of the white leghorn chicken embryos indicates that a 5 min period was sufficient to assess the developmental and species response differences to acute hypoxia.

Each of the autonomic blockade agents in the study was given to each embryo in series (atropine, propranolol and then phentolamine). This method of study has been used successfully in embryonic chickens(Girard, 1973; Tazawa et al., 1992; Crossley and Altimiras, 2000),embryonic alligators (D. A. Crossley et al., unpublished data) and larval bullfrogs (Burggren and Doyle,1986) to determine the presence and strength of cardiovascular autonomic tone during ontogeny. Therefore, this method was chosen to allow comparison with existing literature for other species. However, there is a potential for the cardiovascular responses to differ based on the order of drug injection, and this should be considered in all studies.

Drug injection volumes might have induced a hypervolemia in the developing emu. Blood volumes in embryonic emus (Table 1) were estimated using the known relationship between embryonic mass and blood volume in developing chickens(Crossley and Altimiras, 2000). Each drug injection and flush volume totaled maximally less than 6% of the estimated blood volume in the earliest embryos (60%) studied. Estimated blood volume of the developing emu was calculated as 7% of the embryonic wet mass(Table 1) based on previously determined values for embryonic emus (D. A. Crossley, unpublished data). Therefore, the maximal volume of fluid added to the embryo during the length of the study was less than 17% of the estimated blood volume. Using 7% of embryonic mass as an estimation of blood volume may underestimate the actual volume by as much as 25% (based on chickens; Romanoff, 1967). Therefore,the volume of each injection probably did not alter cardiovascular function(Crossley and Altimiras,2000).

Over the interval of emu development tested (60%I to 90%I), resting P̄a increased progressively and was significant only between 60%I and 70%I(Fig. 1B). Routine fh gradually decreased over this time frame, possibly due to the onset and progressive expression of parasympathetic cardiac regulation(Fig. 1A). This speculation is based on the observation that atropine produced progressively larger increases in fh after 60%I (Fig. 4A). If this explanation is correct, ongoing maturation of the cholinergic inhibitory system seems likely to continue after hatching because fh was elevated (3.5 times higher) compared with adult emus (Grubb et al., 1983). While this finding could be attributed to mass changes alone, maturation of cholinergic receptor-mediated regulation may also contribute.

60%I embryonic emus responded to acute hypoxia (15% or 10% O₂)with a bradycardia, while 90%I embryos exhibited a tachycardia during 10%O₂ exposure (Fig. 2). This reversal suggests that at 90%I embryonic emus may possess central or reflexive regulation of fh. The sensory components that detect changes in Pa_O may be functional and the effector systems altering fh are active. If the components of the fh chemoreflex are intact at 90%I, as the data indicate, then the hypoxic tachycardia was probably induced by a reduction of cholinergic depression and/or an increase in adrenergic stimulation of fh. The latter has been demonstrated in other avian species and could be largely responsible for the response documented in this study (Giussani et al.,1994; Dragon et al.,1996; Mulder et al.,2000; Crossley et al.,2003).

The change in the P̄a response to hypoxia over development suggests that a hypoxic reflexive becomes operational during emu incubation. This suggestion is based on the pressure response change from hypotension during 10% O₂ (60%I and 70%I) to no change in pressure in embryos at 80%I and 90%I(Fig. 2B,D). An alternative explanation for no change in P̄a in the 80%I and 90%I groups may be a reduced sensitivity to hypoxia. However, since 10% O₂ was sufficient to induce changes in fhat 90%I, it is likely that the embryos were sensitive to hypoxia at this point(Fig. 2B).

In addition to a cholinergic tone, embryonic emus also had an importantβ-adrenergic receptor-mediated tone on fh(Fig. 5A). Thisβ-adrenergic tone rose between 70%I and 80%I(Fig. 5A). In chicken embryos,there is an increase in β-adrenergic tone on fh that has been partially attributed to an increased stimulation by elevated levels of circulating catecholamines (Dragon et al., 1996; Crossley and Altimiras, 2000; Mulder et al., 2000). Circulating levels of norepinephrine also increased from 70%I to 80%I (Table 2) and may account for the increase in β-adrenergic receptor-mediated tone on fh during this period.

Unlike the fh responses to β-blockade during emu development, the reaction to α-blockade with phentolamine was constant during the period of incubation studied (70%I to 90%I; Fig. 5C). The change in fh after α-blockade may be derived from a secondary response to the accompanied vasodilation, which would increase filling time and reduce fh. However, given the excitatory action of cardiac α-receptors in adult, as well as neonatal and fetal, mammals(Cohen, 1986), directα-receptor excitatory tone on fh may be present in embryonic emus.

Embryonic emus possessed a constant β-adrenergic tone that decreased P̄a during incubation when corrected for the developmental changes in resting arterial pressure(Fig. 5B). These embryos also possessed a constant α-adrenergic tone that increased P̄a(Fig. 5D). In combination,these findings indicate that the embryonic emu relies on both α- andβ-adrenergic receptor types to maintain P̄a during the window of development studied.

The development of cardiovascular regulation in birds has been studied primarily in the domestic chicken. The present study reveals some distinct differences between emus and chickens.

In emus, fh initially falls (between 60%I and 70%I)then remains relatively constant (Fig. 1A), while embryonic chickens maintain fh over the same period of incubation (Girard,1973; Tazawa,1981). This difference may be attributed to the presence of a cholinergic tone on fh in emus(Fig. 4A) and its absence in chickens of the white leghorn strain(Crossley and Altimiras, 2000). Furthermore, the hypoxic tachycardia of late embryonic emus (90%I; Fig. 2C) differs from the bradycardia in white leghorn embryos. Again, this may also be attributed to the differences in cholinergic tone between the species.

Differences in the mechanisms of regulation were also evident between embryonic emus and chickens of the white leghorn strain. The most prominent difference between these avian embryos was the presence of cholinergic tone on fh in embryonic emus (present study) and the absence in white leghorn chickens (Crossley and Altimiras, 2000). Given our understanding of these two models of avian development, the reason for the appearance of cholinergic fh tone in emus and its absence in white leghorn chickens is unclear. This comparison illustrates that the cardiovascular regulatory mechanisms delineated in white leghorn chickens during development may be different in other bird species.

During embryonic development, emu cardiovascular regulation mechanisms are more advanced that those of embryonic white leghorn chickens over a similar window of development. At 70%I, a clear cholinergic receptor-mediated tone(regulating fh) and adrenergic receptor-mediated tone(regulating fh and P̄a) are functional in the emu. As the emu matures, the embryo relies on an increasing cholinergic tone on fh and a β-adrenergic control of fh. Furthermore, embryonic emus exhibit temporal changes in the cardiovascular responses to 10% O₂. Therefore, maturation of cardiovascular regulatory mechanisms should be expected to differ between avian species during development.

We are grateful for the help and egg donations provided by the ownership of the Cross Timbers Emu Ranch in Flower Mound, TX and for Dr Michael Hedrick's MS suggestions as well as egg donations. Funding for the project was provided by an American Heart Association Training Fellowship To D.A.C. and a National Science Foundation grant (1BN-9616138) to W.W.B.