The surfactant system, a complex mixture of lipids and proteins, controls surface tension in the lung and is crucial for the first breath at birth, and thereafter. Heterokairy is defined as plasticity of a developmental process within an individual. Here, we provide experimental evidence for the concept of heterokairy, as hypoxia induces a change in the onset and rate of development of surfactant, probably via endogenous glucocorticoids, to produce individuals capable of surviving early hatching. Chicken eggs were incubated under normoxic (21% O2) conditions throughout or under hypoxic (17% O2) conditions from day 10 of incubation. Embryos were sampled at days 16, 18, and 20 and also 24 h after hatching. In a second experiment, dexamethasone (Dex), tri-iodothyronine (T3), or a combination (Dex + T3) was administered 24 and 48 h before each time point. Both hypoxia and Dex accelerated maturation of the surfactant lipids by increasing total phospholipid (PL), disaturated phospholipid (DSP), and cholesterol (Chol) in lavage at days 16 and 18. Maturation of surfactant lipid composition was accelerated, with day 16 %DSP/PL, Chol/DSP, and Chol/PL resembling the ratios of day 20 control animals. The effect of Dex + T3 was similar to that of Dex alone. Hypoxia increased plasma corticosterone levels at day 16, while plasma T3 levels were not affected. Hence, exposure to hypoxia during critical developmental windows accelerates surfactant maturation, probably by increasing corticosterone production. This internal modulation of the developmental response to an external stimulus is a demonstration of physiological heterokairy.
- thyroid hormones
heterokairy, a recently coined term (23), is defined as changes in the timing and/or onset of a particular physiological system during development within a given species. Such alterations in timing can be initiated by disturbances to the developmental process, be they genetic, environmental, or phylogenetic, resulting in variability within species. While many studies demonstrate heterokairy, there are very few that demonstrate a mechanism whereby exogenous (environmental) stimuli influence the endogenous developmental controllers leading to developmental plasticity. Here, we provide experimental evidence for the concept of heterokairy by demonstrating that hypoxia can induce differences in developmental rates, probably by altering the release and activity of endogenous neurohormonal controllers, to produce individuals with (in this case) enhanced lung development and that hatch earlier. Concurrently, we provide evidence for the role of hypoxia in the control of pulmonary surfactant development in vertebrates.
The pulmonary surfactant system is crucial for the hatchling to take its first breath. Surfactant is a complex mixture of lipids and proteins, which acts to regulate the surface tension at the air-liquid interface of the lung. By disrupting the attractive forces between water molecules at low lung volumes, surfactant can reduce surface tension (20) and consequently reduce the work of breathing, prevent fluid transudation into the lung (20), and prevent the adherence of opposing respiratory surfaces (5). Surfactant is composed of phospholipids (PL), including disaturated phospholipids (DSP), neutral lipids [predominantly cholesterol (Chol)], and surfactant-associated proteins. Surfactant is synthesized within alveolar type II cells and stored in secretory vesicles termed lamellar bodies. On receiving an appropriate signal, the lamellar bodies are exocytosed into the liquid lining of the lung where the lipids assemble to create the surfactant film at the air-liquid interface (29).
In general, the maturation of the surfactant system is indicated by an increase in the number of lamellar bodies and by increases in the amount and saturation (%DSP/PL) of total surfactant PL over the last 30% of incubation. The overall pattern of surfactant development in the chicken is similar to that of mammals and other vertebrates (12, 14, 25). Specifically, type II cells develop within the atrial walls of the parabronchi between the 14th and 15th day of incubation (36). The first lamellar bodies within the type II cells appear on day 16 of incubation and then rapidly differentiate and increase in number to a maximum at day 18 (10, 16). Similarly, the amount of total PL and DSP increases in the final stages of incubation, with the most rapid increase after 18 days of incubation (14). Additionally, the DSP/total PL ratio is maximal immediately after pipping, which normally occurs on day 19–20 of incubation (14).
Although the pattern of maturation is similar between species, the onset and completion of the development of the vertebrate surfactant system differ dramatically between species (12). These differences do not appear to be determined by phylogeny, as the most closely related species often have the most different pattern. Instead it appears that the developmental process is determined by birthing strategy, which correlates with the development of relative hypoxia as an embryo develops. For example, green sea turtles and the viviparous lizard Tiliqua rugosa (and also placental mammals) complete their surfactant development much earlier (75–80%) than most oviparous species (90–95%) that lay their eggs in a manner that exposes them to normal air (12). Therefore, one of the factors that drive the development of the vertebrate surfactant system may be hypoxia. We have shown that under mildly hypoxic (17% O2) conditions, development of chicken embryos is accelerated, with hatching being brought forward by 24 h (24).
Glucocorticoids and thyroid hormones also have a role in avian surfactant development, as they do for mammals. The synthetic glucocorticoid dexamethasone (Dex) applied in vitro enhances PL synthesis and secretion from cultured embryonic chicken type II cells (25). Moreover, in ovo administration of glucocorticoids and thyroid hormones to chicken eggs and embryos can accelerate overall lung development, with increased synthesis and secretion of total lung PL (6, 11, 31, 32).
The specific and interactive effects of hypoxia, glucocorticoids, and thyroid hormones on the pulmonary surfactant system are not understood. Given that hypoxia enhances plasma glucocorticoid levels in mammalian embryos (7–9), coincident with surfactant protein mRNA expression (8), and peak plasma thyroid hormone and glucocorticoid concentrations occur simultaneously with increasing hypoxia late in embryonic development in chickens (6, 15, 27, 28), it is likely that these factors interact during late development to regulate surfactant maturation in birds. This study tests the hypotheses that exposure to hypoxia, and in ovo administration of glucocorticoids and thyroid hormones, will accelerate surfactant development in chickens. Furthermore, we hypothesize that the mechanism by which hypoxia stimulates surfactant maturation is via an increase in either or both of these hormones. By experimentally manipulating oxygen tension and the endogenous controllers, we determined whether these factors might result in physiological heterokairy of surfactant development.
MATERIALS AND METHODS
Fertilized chicken eggs were obtained from commercial suppliers (Globe Derby Poultry, Adelaide, Australia or HiChick Breeding, Kapunda, Australia). All eggs were incubated at 39°C in a Bellsouth 100 electronic incubator equipped with a Bellsouth 100AT automatic turner (Bellsouth, Narre Warren, Victoria, Australia). Chicken embryos involved in hypoxia experiments were sampled after 16, 18 (prepip), and 20 days (postpip) of incubation and during the first 24 h after hatching (day 21). Pipping is defined as the process whereby the embryos break the shell membrane and begin aerial respiration within the air cell. Eggs were monitored for the presence of pipping from day 19 onward by both visual and aural observations. Embryos undergoing hormone pretreatment were sampled on days 16, 18, and 20 of incubation. These time points represent important developmental stages of the respiratory epithelium in birds (10, 16) and also match days for which data are available for the development of the pulmonary surfactant system (12, 14).
Animal experiments were performed with ethics approval from the University of Adelaide Animal Ethics Committee.
Control animals were incubated under normoxic conditions (21% O2, 0.03% CO2) until sampled. Experimental animals were incubated under normoxic conditions until day 10 of incubation when eggs were transferred to hypoxic conditions (17.0 ± 0.4% atmospheric O2, 0.03% CO2). Hypoxic conditions were maintained by connecting a cylinder filled with 15% O2 in nitrogen to the Bellsouth incubator via a flowmeter (Sierra Instruments, Monterey, CA) set to a flow rate of 180 ml/min. The porosity of the polystyrene incubator, together with the gas flow rate and the low cylinder O2 concentration, ensured adequate mixing of the gas and maintenance of 17% O2. The O2 and CO2 content of the incubator was continuously monitored from the opposite end of the gas inlet, using an infrared CO2 and Zirconia cell O2 analyzer.
Chicken embryos were pretreated with an initial hormone dose 48 h before each sampling day, and we administered a second dose 24 h before sampling. Embryos were pretreated with water-soluble Dex (2 × 50 μg, Sigma Chemicals, Sydney, Australia), tri-iodothyronine (T3, 2 × 3 μg, Sigma), or a combination of both (Dex + T3). Hormone doses were based on previous hormone pretreatment studies on chicken (6, 32), turtle (19), and crocodile eggs (22, 26). A small hole was drilled into the air cell of the egg using a dental drill, and 100 μl of isotonic saline (0.15 M NaCl) containing the agonists was injected. Dex was dissolved directly into saline whereas T3 was dissolved in a small volume of 0.1 M NaOH before diluting in saline. Control animals were injected with 100 μl of isotonic saline only. Drilling and injection procedures were performed under sterile conditions in a laminar flow hood (model VWS-120, Clyde Industries). After injection, the hole in the top of the shell was sealed with a drop of candle wax, and eggs were returned to the incubator and left in an upright position for 2 h to ensure optimal uptake of the hormone (32). After the 2-h period, normal egg rotation was resumed.
Blood for plasma hormone analysis was obtained from embryos at days 16 and 18 from the vessels in the chorioallantoic membrane at the air cell of the egg using a heparinized syringe. For day 20 embryos and hatchlings, blood was collected directly from the heart. Blood samples were centrifuged (6,000 rpm, 3 min), and the supernatant (plasma) was snap-frozen and stored at −80°C. Levels of plasma free T3 were determined by a standard proprietary autoimmune assay system performed at the Institute of Medical and Veterinary Science (Adelaide, Australia) on an Abbott Architect Instrument (Abbott Diagnostics, Abbott Laboratories, Abbott Park, IL). Levels of plasma corticosterone were determined by ReproMed (Adelaide, Australia) using a rat corticosterone kit (DSL-80100) from Diagnostic Systems Laboratories. This RIA was specially modified for chicken serum, by reducing all volumes by 50% due to small amounts of sample and diluting kit standards with charcoal-stripped chicken serum. This modified standard curve (5.8–2,890 nmol/l) exhibited good parallelism with the kit standard curve. Interassay precision (n = 3) was 10.6 to 11.9%, and intra-assay precision (n = 10) was 6.5 to 6.9%.
At days 16 and 18 of incubation, embryos were killed by briefly dipping eggs in liquid nitrogen. Animals that had commenced pulmonary ventilation (day 20 embryos and hatchlings) were killed by inhalation of 100% carbon dioxide. Embryos were tracheal cannulated, and the lungs were lavaged with 3 vol of chilled 0.15 M saline (0.04–0.07 ml/g body mass) instilled and withdrawn three times per volume. Lavage volume was dependent on the strength of the respiratory tissue (to avoid rupturing the lungs and air sacs). Generally, 0.04 ml/g body mass was used on day 16 embryos and 0.07 ml/g body mass was used on pipped and hatched embryos, while the volume for day 18 embryos was intermediate. Any fetal lung fluid present was incorporated into the lavage. Lavage from individuals was centrifuged (150 g, 5 min, 4°C) to remove macrophages and other cellular debris. The supernatant was snap-frozen in liquid nitrogen and lyophilized. Lipids were extracted from lavage with a chloroform-methanol mixture (1:2 vol/vol) (4). Total PL content was calculated by measuring total inorganic phosphorus in the lavage (3). The DSP fraction was separated from the neutral lipid fraction on aluminum oxide using adsorption column chromatography (17). The DSP content of lavage was quantified by measuring inorganic phosphorus (3). Chol levels in lavage were determined on the neutral lipid fraction collected from the aluminum oxide columns using gas chromatography (33). Briefly, an internal standard (25 μg of 5β-cholestan-3α-ol) was added to the sample. Sterols were derivatized with Sylon BTZ (Supelco) for 35 min at 80°C, and silyl derivatives were extracted into hexane. Samples were measured on an Agilent 6890 gas chromatograph equipped with an Agilent 7683 autosampler, split/splitless injector, flame ionization detector coupled to a Chemstation computerized chromatography data system. One to two microliters of sample in hexane was injected (splitless mode) onto a BPX5 capillary column (5% phenylpolisiloxane, 25 m × 0.22 mm, 0.25-μm film thickness; SGE Australia). The oven temperature was increased from 230°C (for 1.5 min) at 10°C/min to 265°C (for 9 min). Hydrogen was the carrier gas at a linear velocity of 56 cm/s. Chol concentration was calculated from the standard curve using the ratio of the sterol peak area to the peak area of the internal standard.
All data are presented as means ± SE. Patterns of embryonic development of surfactant PLs after hypoxic and normoxic incubation are illustrated with a best-fit regression line. Correlation coefficients were calculated using the least-squares method (35). Differences between hypoxic and normoxic embryos at the same stages were determined using a Student's t-test, with significance at P < 0.05. Differences between hormone-pretreated embryos were determined using an ANOVA followed by either a Dunnett's multiple comparisons test or a Student-Newman-Keuls multiple comparisons test as appropriate, to compare each experimental group to the control for that stage. Differences in plasma T3 and corticosterone between experimental groups and the controls were determined using a Student's t-test with Bonferroni correction.
Under hypoxic conditions, approximately one-third (29.5%) of the embryos pipped/hatched 24 h earlier than normoxic embryos (day 20–21 of incubation as opposed to day 21–22). There were no significant differences in body mass between control and hypoxic animals of the same incubation age [day 16: hypoxia 15.33 ± 0.78 g (n = 11), normoxia 16.25 ± 1.08 g (n = 7); day 18: hypoxia 22.00 ± 1.29 g (n = 11), normoxia 21.76 ± 1.80 g (n = 7); pip: hypoxia 40.68 ± 2.22 g (n = 7), normoxia 40.05 ± 2.10 g (n = 5); hatch: hypoxia 41.85 ± 1.14 g (n = 9), normoxia 40.79 ± 2.23 g (n = 5); P > 0.05]. Animals pretreated with hormones and control animals (saline treated) pipped on the same day of incubation (day 19–20). Hormone pretreatment did not affect body mass at day 16 of incubation [day 16: saline 15.25 ± 1.05 g (n = 8), Dex 11.89 ± 0.90 g (n = 7), T3 16.47 ± 1.76 g (n = 5), Dex + T3, 13.31 ± 1.28 g (n = 6); P > 0.05] or at pipping [day 20: saline 40.70 ± 2.29 g (n = 11), Dex 39.54 ± 2.49 g (n = 6), T3 42.93 ± 2.05 g (n = 5), Dex + T3, 45.90 ± 1.33 g (n = 5); P > 0.05]. However, at day 18, animals treated with Dex + T3 were significantly heavier than controls [day 18: saline 19.38 ± 1.35 g (n = 7), Dex 20.88 ± 1.44 g (n = 9), T3 20.08 ± 1.95 g (n = 7), Dex + T3, 26.74 ± 1.88 g (n = 6); P < 0.05 for Dex + T3 vs. saline].
Effect of hypoxia and hormone pretreatments on plasma T3 levels.
In control animals, plasma T3 was low for days 16 and 18 of incubation and then increased dramatically (P < 0.001) at day 20 (pip) before decreasing (P < 0.01) at day 21 (hatch) (Fig. 1). The control group includes both embryos incubated under normoxic conditions and animals injected with saline. There was no difference in plasma T3 levels between embryos incubated under hypoxic conditions and controls at any time point (Fig. 1). Embryos treated with T3 demonstrated significantly higher levels of free T3 in the plasma on days 16, 18 (P < 0.01), and 20 (P < 0.05) of incubation. Treatment with Dex increased plasma T3 levels at days 16 and 18 (P < 0.01) but not at day 20. As the number of samples in this latter experimental group is low, the results need to be interpreted with caution.
Effect of hypoxia and hormone pretreatments on plasma corticosterone levels.
In control animals, plasma corticosterone levels increased from day 16 to day 20 of incubation (P < 0.05), with no significant differences between days 16 and 18 and days 18 and 20 (Fig. 2). Embryos treated with Dex demonstrated significantly higher levels of plasma corticosterone on day 16 of incubation compared with controls (P < 0.05) but not at days 18 and 20 of incubation (Fig. 2). Embryos incubated under hypoxic conditions had significantly higher plasma corticosterone levels than controls at days 16 and 20 of incubation (Fig. 2). There was no difference in plasma corticosterone in day 18 hypoxic embryos compared with controls (Fig. 2).
Effect of hypoxia on surfactant lipid development and composition.
In this and the following section, we have provided both the total amounts of individual components (Tables 1 and 2) as well as the relative composition of different components (i.e., ratios in Figs. 3 and 4). Hence, the data in Figs. 3 and 4 are derived from Tables 1 and 2. These data provide an indication of the effects of hypoxia and hormone treatment on both surfactant secretion (lipid amounts) and synthesis (ratios).
Direct comparisons between the lavage of hypoxic embryos and controls reveal significant differences in the amounts of total PL, DSP, and Chol in the earlier stages of embryonic development. At days 16 and 18, hypoxia caused an increase in the amounts [μg/mg dry lung wt (DLW)] of total PL and DSP in lavage above that of control animals. However, at pipping and hatching there was no difference in amounts (μg/mg DLW) of total PL and DSP between hypoxic and normoxic animals. The amount of total Chol (μg/mg DLW) was significantly higher in the lavage of hypoxic animals at day 16 compared with control animals. However, at day 18, pipping, and hatching, no differences were observed in the amount of lavagable Chol between hypoxic and normoxic animals (Table 1).
When comparing the effect of hypoxia on the ratios of surfactant lipids at each of the time points, we found that the saturation of the PLs (%DSP/PL) was increased by hypoxia at day 16 (P < 0.05) (relative to controls) but not at day 18, pipping, or hatching (Fig. 3A). Hypoxia decreased the amount of Chol relative to total PL (Chol/PL) in lavage of day 18 embryos relative to controls (P < 0.05) (Fig. 3B). No change in Chol/PL was observed at day 16, pipping, or hatching between hypoxic and normoxic animals. The amount of Chol relative to the amount of DSP (Chol/DSP) in lavage was dramatically reduced in hypoxic animals compared with controls at days 16 and 18 (P < 0.05) (Fig. 3C). No changes were observed in the Chol/DSP ratio in lavage of hypoxic animals that had pipped or hatched compared with controls.
When comparing the pattern of change in composition over time between normoxic and hypoxic embryos, we found that the extent of saturation of the surfactant PL (%DSP/PL) in normoxic embryos increased dramatically from day 16 to pipping (P < 0.001), with no significant changes between pipping and hatching (Fig. 3A). There were also no significant changes in %DSP/PL between pipping and hatching in hypoxic embryos (Fig. 3A). %DSP/PL in the lavage of hypoxic embryos increased significantly from day 16 to pip in an exponential fashion (P < 0.001), but this increase was less dramatic than observed in normoxic embryos (Fig. 3A). The ratio of amounts of Chol to total PL (Chol/PL) decreased linearly in the lavage of normoxic embryos from day 16 to pip (P < 0.01) (Fig. 3B). This linear relationship was altered by hypoxia, such that the Chol/PL ratio decreased significantly from day 16 to pip in a logarithmic fashion (P < 0.01) (Fig. 3B). As observed with normoxia, however, there was no change in the amount of Chol relative to total lavage PL between pipping and hatching in hypoxic embryos. Chol/DSP in lavage decreased exponentially in normoxic embryos from day 16 to pipping (P < 0.001) (Fig. 3C), with no significant changes between pipping and hatching. However, in hypoxic embryos, Chol/DSP decreased logarithmically in a less dramatic fashion from day 16 to pip (P < 0.001), with no change between pipping and hatching (Fig. 3C).
Effect of hormone pretreatment on surfactant lipids.
At day 16 of incubation, treatment with Dex resulted in increases in the amounts (μg/mg DLW) of total PL, DSP, and Chol in lavage above that of control animals (Table 2). Treatment with T3 and Dex + T3 increased only DSP but not PL or Chol. At day 18, compared with the saline control, pretreatment with Dex and Dex + T3 dramatically increased the amounts (μg/mg DLW) of total PL, DSP, and Chol, but not treatment with T3 alone. At pipping (day 20), hormone pretreatment with Dex, T3, and Dex + T3 had no effect on the amounts of total PL, DSP, or Chol (Table 2).
The saturation of the PLs (%DSP/PL) in the lavage of embryonic chickens was dramatically increased at day 16 by pretreatment with Dex, T3, and Dex + T3 (P < 0.01) (Fig. 4A). Chol/PL in lavage was not affected by hormone pretreatment at any stage (Fig. 4B). However, Chol/DSP was greatly decreased by treatment with Dex, T3, and Dex + T3 at day 16 (P < 0.01) (Fig. 4C). At day 18 and pipping, the extent of saturation of the surfactant PLs as well as the amount of Chol relative to DSP was not affected by treatment with Dex, T3, and Dex + T3 (Fig. 4, A and C).
Mild hypoxia (17% O2) had previously induced early hatching in chickens (24) but did not produce animals with apparently different physiological systems or anatomic structures. Similarly, in the present study, hatching was generally brought forward by 24 h. Despite alterations in timing, body mass was not different in hypoxic animals compared with controls at any stage. Incubation under severe conditions of hypoxia (10% O2) has previously resulted in decreased body mass in embryonic chickens (18, 34), presumably because size is sacrificed to allow for a more complete development of crucial organs and systems.
In the present study, hormone doses were based on other studies involving in ovo pretreatment of chickens (6, 32), turtles (19), and crocodiles (22, 26). Neither agonist accelerated pipping when administered on days 18 and 19. However, at day 18, Dex + T3-treated embryos were significantly larger in body mass, possibly due to the combination of hormones stimulating both the overall metabolic rate of the embryos (T3) as well as specific synthetic processes (Dex) that accelerate maturation. The combination of these processes may therefore have allowed the embryo to grow larger in the time available.
Effect of hypoxia and hormone pretreatments on plasma T3 levels.
Our injection method successfully administered T3 into the bloodstream of experimental animals, as pretreatment with T3 significantly elevated plasma free T3 levels above those of control animals (P < 0.01) (Fig. 1). The low levels of T3 in the plasma of 16- and 18-day-old control embryos followed by the dramatic increase in plasma T3 at day 20 corresponds to the normally observed changes in developing chickens during the last days of incubation (6, 28). Hypoxia did not elevate plasma T3 levels (Fig. 1). Hence, the hypoxia-induced effects on surfactant PLs were not likely to be mediated by T3. Although the number of samples in the Dex pretreatment group is low, our findings that Dex stimulates secretion of T3 at days 16 and 18 (Fig. 1) are supported by previous work (6). Furthermore, as with the present study, Darras et al. (6) observed that once the animals matured, the effect of glucocorticoids on plasma T3 was abolished. Glucocorticoids may cause an elevation of plasma T3 levels by inhibiting the enzymes that break down T3 (thyroid hormone-deiodinating enzymes). This inhibition is apparent for embryos during earlier stages of development but is lost once the embryo reaches a more mature state (pipping and posthatch) (6).
Effect of hypoxia and hormone pretreatments on plasma corticosterone levels.
Plasma corticosterone levels in control animals significantly increased from day 16 of incubation to day 20 (Fig. 2; P < 0.05), which is consistent with previous studies that observed peaks of plasma and adrenal corticosterone at around the time of hatching (27). Pretreatment with Dex resulted in elevated levels of plasma corticosterone at day 16 of incubation compared with controls (Fig. 2; P < 0.05). This indicates that the administration of Dex was successful in entering the bloodstream and that the corticosterone assay detected excess plasma Dex. Success is further indicated by the effect of Dex pretreatment on the surfactant lipids (Table 2 and Fig. 4).
Elevated levels of corticosterone were not observed in Dex-treated day 18 and pipped embryos (Fig. 2). This may reflect the maturation of normal self-regulatory mechanisms (i.e., development of the negative-feedback loop in the hypothalamo-pituitary-adrenal axis) and thus suppression of endogenous glucocorticoids.
Hypoxic incubation of chicken embryos resulted in elevated plasma corticosterone at day 16 of incubation (Fig. 2; P < 0.05). Early elevated levels of plasma corticosterone are likely to cause the changes observed in the surfactant lipids on days 16 and 18 after hypoxic incubation. While plasma corticosterone levels were not elevated at day 18, elevated corticosterone at day 16 may have been enough to trigger the changes observed at day 18, just as Dex administration on days 16 and 17 resulted in lipid changes at day 18. Tanabe and coworkers (27) observed that peak plasma corticosterone concentrations occur at hatching. As 29.5% of hypoxic animals pipped earlier than the normoxic control animals, the elevated levels of plasma corticosterone at day 20 in hypoxic animals probably reflect this early hatching.
Effect of hypoxia on surfactant lipid development and composition.
This is the first study to determine whether hypoxia affects surfactant development and composition in birds. In the sea turtle, mild hypoxia (17% O2) did not affect incubation time or the developmental pattern of the surfactant PLs (13). However, green sea turtle eggs are naturally buried under wet sand with a high bacterial and fungal load and are therefore routinely exposed to low oxygen concentrations during development (1, 13).
The pattern of development of the surfactant PLs in chicken embryos incubated under normoxic conditions (Fig. 3) matched those reported previously (12, 14). Hypoxic incubation from day 10 resulted in an overall acceleration of the maturation of the surfactant system but did not alter the end point. Hypoxia induced both the early release of PL, DSP, and Chol (Table 1), as well as an enhanced maturation of the surfactant ratios (%DSP/PL, Chol/PL, and Chol/DSP; Fig. 3), indicating upregulation of the synthetic and secretory pathways of surfactant lipids from day 16. This premature maturation correlated with the early pipping and hatching. This study therefore provides evidence that exposure to hypoxia late in incubation leads to physiological heterokairy of the surfactant system by altering both the rate and onset of development of the surfactant lipids. The mechanism by which hypoxia could act to trigger surfactant maturation is most likely hormonal. Levels of thyroid hormones and glucocorticoids peak at approximately the same time as the maturation of the surfactant system (6, 21, 27, 28) just before pipping, which further correlates with the occurrence of relative hypoxia (30). In this study, levels of corticosterone increased in response to hypoxia early in embryonic development, whereas levels of T3 did not increase at any time. Moreover, exogenous T3 did not alter the maturation of the surfactant system to the same extent as Dex. It is likely, therefore, that the maturation process caused by hypoxia is mediated by glucocorticoids. Catecholamines may also be potential contributors to surfactant development. However, as epinephrine promotes surfactant secretion, but not the synthetic process, and hence the developmental trajectory, the actions of this neurotransmitter were not investigated in this study.
Effect of hormone pretreatment on surfactant lipids.
This is the first study to demonstrate that in ovo administration of hormones can influence the amount and composition of alveolar surfactant in birds. In saltwater crocodiles, administration of Dex or T3 to eggs and embryos increased the total PL content in lavage, but the effect was dependent on gestational age (26). Similarly, in chickens, hormone administration had variable responses at different stages of incubation. Total lipid content of lavage was increased by Dex and Dex + T3 treatment predominantly at the early time points (days 16 and 18) and then appeared to lose efficacy after pipping (Table 2). This is consistent with in vitro studies (25) in which Dex treatment of type II cells isolated from the lung of day 16 and day 18 chicken embryos resulted in increased surfactant secretion, while treatment of cells from day 20 embryos and hatchlings did not. The early maturation effect of Dex is likely to be mediated via receptors on lung fibroblasts (25). This action triggers the synthesis of specific factor(s), which acts on type II cells to promote surfactant synthesis and/or secretion. The level of action depends on the relative timing of hormone treatment and the sensitivity of the type II cells to the fibroblast factor at the time of sampling (25). The loss of efficacy of Dex after pipping (Table 2 and Fig. 4) could be due to a factor similar to that secreted by mature rat type II cells that inhibits fibroblast function, limiting the glucocorticoid response (2).
Administration of T3 alone did not affect total PL or Chol at any time point (Table 2). Therefore, it appears that any effects of pretreatment with Dex + T3 were mediated by Dex alone. Consequently, it does not appear that in the chicken Dex and T3 have a synergistic effect, and T3 does not appear to be as potent in chicken lungs as it is in crocodiles (26). However, administration of T3 did increase total amounts of DSP in lavage of day 16 embryos (Table 2) and thus also increased %DSP/PL (Fig. 4A) and decreased Chol/DSP (Fig. 4C) compared with controls at day 16. This indicates that thyroid hormones may have a role in increasing the saturation of the PLs early in avian surfactant development, presumably via enhanced surfactant synthesis, as opposed to secretion.
Glucocorticoids appear to act only at certain critical windows in time in the embryonic development of the chicken and bring about developmental plasticity in the process of surfactant maturation. The percent saturation of surfactant PLs was affected by hormone pretreatment only before day 16 (Fig. 4A), indicating that this time is important for triggering surfactant lipid synthesis. Similarly, total amounts of surfactant PLs were increased in lavage by hormone pretreatment before both day 16 and day 18 but not at day 20 (Table 2), indicating that day 18 is the crucial time for the hormones to trigger surfactant secretion into the air space.
The pretreatment of chicken embryos with hormones, administered at specific developmental stages, results in physiological heterokairy of the pulmonary surfactant system. The day 16 values for %DSP/PL and Chol/DSP after hormone pretreatment resembled values found in control animals at pipping (Fig. 4, A and C). These effects were predominantly mediated by Dex, indicating that glucocorticoids are crucial to the development and maturation of the surfactant system.
In conclusion, this study provides evidence for the role of and interactions between environmental and hormonal factors during the development of the avian surfactant system. Both exposure to hypoxia and pretreatment of embryos with a synthetic glucocorticoid during critical developmental windows accelerate the maturation of the surfactant system, thus demonstrating physiological heterokairy. As hypoxia did not elevate plasma T3 levels, thyroid hormones are unlikely to represent the mechanism by which hypoxia induces surfactant alterations. Instead, hypoxia-induced alterations are most likely attributed to increased glucocorticoid levels at critical windows early in surfactant development, as plasma corticosterone is increased by hypoxia at this time. Furthermore, alterations in surfactant development in response to hypoxia match closely those associated with Dex administration. Corticosterone, in turn, could initiate the development and maturation of the surfactant system by first triggering surfactant synthesis and then enhancing surfactant secretion in the latter stages of incubation. This evidence suggests that the avian pulmonary surfactant system exhibits a high level of plasticity within the early stages of surfactant maturation. Environmental and endogenous physiological factors can interact during these critical windows and exploit this developmental plasticity, resulting in earlier development of the surfactant system and earlier commencement of air breathing.
We thank M. Mano and C. White for assistance with technical aspects of this project.
This research was supported by Australian Research Council (ARC) grants to S. Orgeig and C. B. Daniels and an ARC Research Fellowship to S. Orgeig.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society