Respiratory Gases and Acid–Base Balance in Shell-Less Avian Embryos

The respiration of the avian embryo is of considerable interest for both theoretical and practical reasons. It is, as Wangensteen & Rahn (1970/1971) first pointed out, a model system for studying the diffusion component of vertebrate gas exchange. Thus, the movement of gases between the atmosphere and the animal ‘s vascular system is, in the avian embryo, dictated by the conductance of the eggshell. This has a fixed capacity which appears to be dictated by the need to limit water vapour losses to about 15 % of the original egg weight, a requirement that clearly constrains respiratory gas exchange. The continually increasing metabolic demand by the developing embryo for oxygen is, therefore, met by an increased blood flow through the chorioallantoic respiratory system, increased haematocrit and haemoglobin concentrations and by changes in the oxygen affinity of the haemoglobin. Despite these adaptations, the oxygen pressure in the embryo ‘s blood falls throughout the latter part of incubation and is only reversed when pulmonary respiration is initiated shortly before hatching (Freeman & Misson, 1970; Tazawa et al. 1983). It is not surprising, therefore, that the earliest measurements of blood gases in the embryo showed that there was also a progressive respiratory acidosis as carbon dioxide accumulated during development. What was, perhaps, less expected was the discovery that the embryo could compensate for this acidosis by a metabolic alkalosis (Dawes & Simkiss, 1969).

This effect is seen as an increase in blood bicarbonate levels. It is not clear whether the additional base is derived from the secretion of acid into the renal tubules of the embryo (Dawes & Simkiss, 1971), or by the resorption of the eggshell by the chorioallantoic membrane (Crooks & Simkiss, 1974).

The physiology of avian embryos has recently attracted additional interest because of the possibility of genetically manipulating various domesticated birds (Freeman & Messer, 1985). The avian system of reproduction makes it particularly difficult to handle the embryo unless the zygote can be made accessible to the experimenter. For this reason there has been renewed interest in the shell-less in vitro techniques developed by Dunn and his coworkers (Dunn & Boone, 1976; Dunn et al. 1981) with their subsequent modification to produce a successful ‘surrogate egg ‘system for culturing embryos (Rowlett & Simkiss, 1987).

Shell-less techniques for culturing embryos of the domestic fowl have been successful in producing offspring with a chronological age of 21 days. They have never succeeded, however, in producing viable hatchlings, which clearly suffer from both calcium and protein deficiencies. Another possible reason for their non viability is the disturbance to their respiratory gas regulation which the absence of a shell induces. This could be due to several causes. Clearly, shell-less cultures have lost the resistance to gas diffusion imposed by the eggshell, but they also have a reduced respiratory area largely dictated by the area of chorioallantoic membrane exposed in the culture vessel. It will also be apparent that if the metabolic alkalosis found in the normal chick embryo is maintained by resorption of base from the eggshell, then this should also be disrupted in shell-less cultures. For these reasons a study was undertaken of the respiratory gases in cultured embryos.

Fertile eggs of a Rhode-Island Red strain of the domestic fowl were used throughout this work. For normal purposes they were incubated in a commercial forced-air incubator set at 38 °C and 60% relative humidity. The shell-less culture system that was used has been described in detail by Rowlett & Simkiss (1987). Eggs were incubated for 72 h and then opened into a ‘cling-film’ sheet of plastic suspended in a cylindrical plastic holder, and covered with a Petri dish (Fig. 1). These shell-less cultures were incubated in a clean-air cabinet (Bassaire, South ampton) in a constant-temperature room set at 38 °C with 80 % relative humidity.

Blood samples (100 μl) were taken from the allantoic artery (deoxygenated) and allantoic vein (oxygenated) of embryos with a heparinized needle and syringe. The blood-sampling equipment was kept on ice before and after sampling. With control embryos, the egg was punctured in the air space and suspended with the narrow pole uppermost. A 2 cm² hole was made in the shell and the shell membrane dampened with saline so that the underlying allantoic blood vessels could be located and blood samples obtained with the minimum of disturbance.

Obtaining samples from shell-less cultures simply involved penetrating the chorioallantois and taking blood from the conspicuous allantoic vessels.

All blood samples were analysed within 10 min by using an automatic blood gas analyser with pH, and electrodes (System 1302, Instrumentation Laboratory, Italy) which were calibrated with precision buffers and analysed gas mixtures.

All the embryos were killed and weighed immediately after taking blood samples.

Results of analyses for pH, and for oxygenated and deoxygenated blood samples are shown for control embryos and for shell-less cultures (Fig. 2). 46 control and 37 shell-less embryos were studied at various stages of development, but it was not always possible to obtain both arterial and venous blood samples from each specimen. The numbers of samples are shown in Fig. 2, with means and standard error values for each stage. Student’s t-test was used to test for statistical significance between the control and shell-less values for each parameter on each day of incubation. Apart from those on day 12, all values were significantly different for both oxygenated (P<0 · 05) and deoxygenated (P< 0 · 001) blood. The values for oxygenated blood were also significantly different (P<0 · 05), but there were no significant differences for in deoxygenated blood or for blood pH.

An estimate of the non-respiratory component of the blood bicarbonate, i.e. base excess, was determined from nomograms published for human blood (Siggaard-Andersen, 1965). Haemoglobin values of 7, 8, 9 and 10g 100cm⁻³ for 12-, 14-, 16-and 18-day embryos were used for these estimates, based on the results of Tazawa et al. (1971). The results are shown in Fig. 3 for both normal and shell-less embryos.

During the period from 10 to 18 days of incubation, the blood of the normal avian embryo shows a decrease in from about 10·6 to 7·6 kPa, an increase in from 1·6 to 4·3 kPa, a rise in blood bicarbonate from 12 to 22 mmol dm⁻³ and an increase in base excess from −8 to 1 mmol dm⁻³ (Tazawa et al. 1971). The results shown in Table 1 and Figs 2 and 3 for the control embryos agree with these data, and may be interpreted as the product of a resistance to gas diffusion across the avian eggshell and increasing metabolic activity from the growing embryo. Accordingly, there is increased respiratory acidosis, which is compensated by metabolic alkalosis (Dawes & Simkiss, 1969).

This interpretation of the respiratory adaptations of the avian embryo has been confirmed elsewhere (Tazawa & Piiper, 1984). It has also been tested experimentally by effectively reducing the area of the eggshell available for respiratory exchange by coating roughly one-quarter of the surface with epoxy resin. Under these conditions blood falls and rises, inducing an increase in blood base excess which tends to regulate the blood pH at about 7·4 (Tazawa et al. 1971).

The area of the chorioallantoic membrane available for respiratory exchange in shell-less culture systems is constrained by the morphology of the system (Fig. 1). The average surface area unobstructed by plastic film was only 38 cm², roughly half of what is normally available (Dunn & Boone, 1976). It is possible that further gas exchange occurs across the plastic film, but this has an oxygen permeability of 3·1×10⁻⁴cmmin⁻¹ Pa⁻¹, i.e. roughly half that of the normal eggshell (Romijn, 1950; Kutchai & Steen, 1971) and the chorioallantois is separated from this by varying thicknesses of albumen. As a consequence, it seemed likely that shell-less cultures would be subjected to hypoxia and hypercapnia in a similar way to embryos in coated eggshells. The results shown in Fig. 2 demonstrate that this is not the case. The blood oxygen levels in the allantoic vein, which drains the respiratory organ, decrease whereas the levels in the allantoic artery do not. The cultured embryos are, therefore, subjected to a considerable decrease in the arteriovenous oxygen gradient (Fig. 2.). However, instead of being hypercapnic, they show reduced levels of blood carbon dioxide. The decreased carbon dioxide levels could be due to the embryos in shell-less culture being retarded in size (Table 2). It is, however, possible to compensate for this, as Tazawa et al. (1971) and Tazawa & Mochizuki (1977) have produced equations relating blood gas levels to chick embryo mass. The observed and calculated blood gas levels (Table 2) indicate that the abnormal levels in shell-less embryos are not simply due to their small size. In fact, the low oxygen and carbon dioxide levels of the blood of shell less cultured embryos are more typical of amphibious vertebrates in which respiratory exchange is strongly influenced by the solubility of the respiratory gases in water. Thus, in the Amphibia, aerial respiration accounts for much of the oxygen uptake whereas diffusion across the skin into the water is responsible for the removal of much of the carbon dioxide. The efficiency of these systems decreases with an increase in temperature (Rahn & Howell, 1976), but it appears likely that the reduced aerial part of the chorioallantois is the main region for oxygen uptake in the cultured embryo and the submerged lower part of the chorioallantois may contribute to the removal of carbon dioxide.

This work was supported by the Agricultural and Food Research Council (AG45/233/317).