Acclimation to hypothermic incubation in developing chicken embryos(Gallus domesticus): II. Hematology and blood O₂transport

The avian embryo undergoes a thermoregulatory transition from being a poikilotherm in its egg to being a homeotherm at or shortly after hatching. Thermoregulation in hatchling chicks relies on non-shivering endogenous heat production to control body temperature (T_b)(Whittow and Tazawa, 1991; Roberts, 1996; Tazawa et al., 2001), but the physiological adjustments and the timing of their onset as they occur during the transition to homeothermy have not been well described. In a companion study (Black and Burggren,2004), we showed that late-stage chicken embryos (HH 41–44; Hamburger and Hamilton, 1951)maintained their rate of O₂ consumption (and presumably the attendant heat production) during acute reductions in body temperature,suggesting the onset of at least some of the physiological control mechanisms necessary for thermoregulation. Black and Burggren(2004) also investigated the influence of chronic hypothermic incubation in chicken embryos. Not surprisingly, fundamental developmental parameters such as growth and the time of hatching events (internal and external pipping) were slowed by chronic hypothermic incubation (35°C). These changes, however, were all relative– i.e. when expressed as a percentage of total incubation period,embryonic growth and hatching events occurred at predictable points in development in hypothermic chicks. These data suggested that temperature alone did not induce heterokairic changes (heterokairy is the change in the relative timing of onset of physiological processes and their control; Spicer and Burggren, 2003). However, late-stage embryos incubated under hypothermic conditions were not as proficient at maintaining metabolism during a gradual decline in ambient temperature (T_a) as were populations incubated at the normal 38°C, suggesting that chronically low incubation temperature interfered with the normal onset of maturity of their thermoregulatory system,and in particular the ability to elevate oxygen consumption and heat production.

Maturity of both the respiratory and cardiovascular systems is crucial to ensure the efficient delivery of O₂ to support metabolic heat production and regulate body temperature. In this study we examine the effects of hypothermic incubation on the metabolic and thermoregulatory physiology of chicken embryos. We examine hematology, blood O₂ transport characteristics, and in vivo arterial and venous blood P_O2 and pH of late-stage chicken embryos incubated in either normothermal conditions (38°C) or hypothermal conditions (35°C). Our goal was to determine if differences in blood parameters might underlie our finding that lower incubation temperature negatively affects the ability of embryos to respond metabolically to decreases in ambient temperature. Specifically, we hypothesized that chronic incubation at low ambient temperature (35°C) would modify both blood O₂-carrying capacity and hemoglobin–O₂ binding affinity in the late-stage chicken embryo, and that these changes might result in a modified thermoregulatory phenotype.

Fertilized White Leghorn eggs (Gallus domesticus L.) were obtained from Texas A&M University (College Station, Texas, USA) and shipped to the University of North Texas (Denton, TX, USA) where they were incubated in commercial incubators. All experimental procedures were approved by The University of North Texas' Institutional Animal Care and Use Committee.

Samples were incubated at 38.0°C (control), or 35.0°C (hypothermic)and a relative humidity of 60%. To determine the gross effects of incubation temperature on development, nine or more embryos incubated at each temperature were staged for developmental maturity on days (D) 13–14, 15–16,17–18 and 19–20 (Hamburger and Hamilton, 1951). Hypothermic embryos have a slower rate of development than embryos incubated in control conditions, so staging was completed to determine the length of incubation required for the 35°C embryos to reach developmental stages equivalent to those of the 38°C embryos.

All subsequent experiments were conducted on embryos at the following stages: HH 41–42, reached on days 15–16 for 38°C and days 17–18 for 35°C; HH 43–44, reached on days 17–18 for 38°C and days 19–20 for 35°C.

Each egg was candled to locate a large chorioallantoic membrane (CAM)vessel. The overlying shell and membranes were removed without disturbing the underlying vasculature. The tip of a 30 gauge needle attached to a 1 ml syringe was inserted into a CAM artery and/or vein against the direction of blood flow. Dead space in the needle (approximately 10 μl) was filled with heparinized saline (53 mg 100 ml^–1). A minimum of 200 μl of blood was drawn for determination of P₅₀, hemoglobin concentraction and hematocrit, while 100 μl was drawn for measurement of arterial and venous P_O2 and pH or from separate embryos for ATP and 2,3-BPG determinations. Caution was taken to avoid contamination by the accidental uptake of fluid.

A 50 μl blood sub-sample was injected into an OSM2 Hemoximeter(Radiometer, Copenhagen) for determination of the hemoglobin content of the blood (g%). Hematocrit was measured by drawing 10 μl of blood into a heparinized capillary tube, sealing one end, and centrifuging for 5 min in a micro-hematocrit centrifuge. The hematocrit was determined as the volume of packed red blood cells to the total volume of blood.

A 50 μl sub-sample of freshly drawn arterial blood was added to a cuvette containing 4 ml of Hemox solution, 20 μl of Additive A, and 10μl of Anti-Foaming agent, all solutions from TSC Scientific (New Hope, PA,USA). The sample was placed in the measurement cuvette of a HEMOX™Analyzer (TSC Scientific), which continuously records blood O₂saturation while changing blood P_O2, plotting the results as a continuous rather than punctuated O₂ equilibirum curve. Blood placed in the measurement cuvette was allowed to thermally equilibrate at either 35°C or 38°C, depending on the experimental protocol. Measurements of percentage saturation and P_O2 were simultaneously recorded on a computer using Chart software (ADInstruments, Colorado Springs, CO, USA). Humidified air was gently bubbled through the sample until 100% O₂ saturation was obtained (approximately 7 min). The sample was then bubbled with humidified nitrogen for 7 min, until O₂ saturation of the blood reached zero. The sample was then reoxygenated through equilibration with air for an additional 7 min. Recorded saturations and P_O2 values were plotted to obtain a sigmoidal O₂ equilibrium curve. Two P₅₀ measurements were determined on each sample, with the replicates averaged to generate the P₅₀ for that sample.

The percentage saturation (S) of the blood corresponding to P_O2 values of 1.6 kPa, 2.5 kPa, 4.1 kPa, 6.6 kPa, 10.5 kPa and 16.6 kPa was determined for each O₂ equilibrium curve to characterize the shape of the O₂-equilibrium curve for creation of a Hill plot. Values were converted as follows: x=logP_O2, y=log[S/(100–S)], where S is saturation. A linear Hill plot with logP_O2 as the independent variable and log[S/(100–S)] as the dependant variable was fitted with a regression line (Sigmaplot 2001). The`n' coefficient, the slope of the regression line, summarizes the shape of the curve.

A 100 μl sample of CAM arterial or venous blood was injected into thermostatted P_O2 and pH electrodes(Microelectrodes Inc., Bedford, NH, USA) connected to a blood gas analyzer(Blood Gas Meter BMG 200, Cameron Instrument Company, Port Aransas, TX, USA)for simultaneous measurements of P_O2 and pH.

The O₂-binding properties of embryonic chick hemoglobin are highly influenced by pH (Bohr shift). To allow direct comparisons of blood properties between incubation temperatures without the confounding effects of differences in acid–base status, most previously published O₂equilibrium curves have been generated at 7.47, a pH characteristic of late-stage embryos incubated at 38°C(Tazawa, 1980). We elected to measure the P₅₀ of each blood sample at two pH values,7.47, corresponding to previous measurements of arterial pH for embryos incubated at 38°C, and at 7.57, the arterial pH for embryos incubated at 35°C (Tazawa, 1973, 1980). Bohr shifts were calculated for each developmental stage and incubation temperature. Then,using the calculated Bohr shifts, a series of stage- and treatment-specific O₂-equilibrium curves determined in vitro at pH values of 7.27–7.87 (in 0.1 pH unit increments), were constructed. Each in vivo blood P_O2 value collected from a sample was then plotted on the O₂-equilibrium curve most closely corresponding to that sample's pH. The in vivoO₂-equilibrium curves used for placement were then removed to reveal the in vivo O₂-equilibrium curves at the mean arterial and venous pH values.

A 100 μl sample of blood was added to an equal volume of trichloroacetic acid (TCA) for ATP measurement and to 300 μl of TCA for 2,3-bisphosphoglycerate (2,3-BPG) measurement. Samples designated for ATP measurement were stored on ice for a maximum of 2 h before measurement. Samples for 2,3-BPG measurement were stored at –70°C for 4 weeks before measurement, conditions that maintain stable amounts of organic compounds. Concentrations of organic phosphates were determined with an end-point spectrophotometry technique at a wavelength of 345 nm(Sigma–Aldrich measurement kit; St Louis, MO, USA).

All data was tested for normality of distributions (Shapiro–Wilks normality test) and equality of variances before specific statistical analyses were performed. Data from within and between stages and at the two incubation temperatures were tested for statistical significance with an ANOVA. Significance between groups was determined with a Student–Newman–Keuls (SNK) multiple range post hoc test. All statistical tests were conducted using SAS software and decisions were made with a 0.05 level of significance. Values are means ± 1 s.e.m.

Incubation temperature had no significant effect (P>0.05) on hematocrit or [Hb] in HH 41–42 embryos(Fig. 1). Further development to HH 43–44 in 38°C embryos resulted in a significant increase in both hematocrit (F=6.17, P=0.001) and [Hb] (F=6.28, P=0.001). However, the stage-dependent increases in 38°C incubated embryos did not develop in 35°C embryos, where no significant increase occurred in either hematocrit or [Hb].

At HH 41–42, blood of both 35°C and 38°C embryos had relatively high P₅₀ value, low O₂ affinity, and was insensitive to changes in measurement temperature(Fig. 2). By HH 43–44,the blood–O₂ affinity of the 38°C embryos at 38°C had not changed significantly, but was now temperature sensitive, with P₅₀ dropping from 6.0 kPa±0.4 at 38°C to 4.4 kPa±0.4 at 35°C (Fig. 2B, F=4.55, P<0.0001). In considerable contrast to the 38°C embryos, the blood of the 35°C embryos remained temperature insensitive at HH 43–44, showing no change in O₂affinity measured at 38°C (Fig. 2B, SNK, P>0.1).

Temperature- and stage-induced O₂ affinity changes in 35°C and 38°C embryos were not accompanied by any significant change in the shape of the O₂ equilibrium curve, summarized by the Hill coefficients (n). Values of n ranged from 2.42±0.056 to 2.64±0.059 for the HH 41–42 embryos(Fig. 3A inset). Similarly, at HH 43–44 there were no significant changes in the shape of the curves for either the 35°C or 38°C embryos, with Hill coefficients ranging from 2.10±0.19 to 3.10±0.11(Fig. 3B inset).

Blood from the CAM artery, carrying relatively deoxygenated blood from the tissue of the embryos to the CAM gas exchanger underlying the shell, had low P_O2 values, in the order of 3.7–4.2 kPa(Table 1). 35°C embryos at HH 41–42 had a significantly lower arterial P_O2 than the 38°C embryos (SNK, P<0.05), but between HH 41–42 and 43–44 the arterial P_O2 of the 35°C embryos remained unchanged,while that of the 38°C embryos dropped significantly (SNK, P<0.001).

Venous CAM blood (draining the CAM gas exchanger) of HH 43–44 embryos had similar P_O2 values regardless of incubation temperature (between 8.1±0.3 kPa and 9.1±0.4 kPa for 35°C and 38°C, respectively). Only the 38°C embryos experienced a significant drop in venous P_O2 as development continued (SNK, P=0.024). Within each incubation temperature and stage the venous P_O2 was always significantly higher than that of the arterial blood (SNK, P<0.001).

There were no significant differences in the pH of arterial or venous blood between incubation temperatures or stages, and the pH of the arterial blood was not significantly different from the venous blood for any of the treatment groups (Table 1).

When plotted on O₂-equilibrium curves representing in vivo arterial and venous pH conditions, the in vivophysiological range of P_O2 and saturation experienced by the embryo can be determined for each group(Fig. 4). HH 41–42 embryos maintained a 60–80% O₂ saturation in `arterialized'blood in both incubation temperature groups. With further development to HH 43–44, oxygen saturation climbed slightly to 65–90% in both groups. There was more variation in CAM arterial blood saturations. Minimum saturation values of about 20% were evident in 35°C embryos at HH 43–44 compared with less than 10% in 38°C embryos at HH 43–44.

Blood [ATP] ranged from approximately 80–120 μmol dl^–1 (Fig. 5A), while blood [2-3-BPG] ranged from approximately 0.1–0.6μmol dl^–1 (Fig. 5B). The 35°C embryos had significantly lower concentrations of ATP at HH 41–42 (F=4.20, P=0.012), but by HH 43–44 there was no significant difference in [ATP] between incubation temperatures. Concentrations of 2,3-BPG were not significantly different between incubation temperatures at HH 41–42. By HH 43–44 the 38°C embryos experienced a significant decline in blood 2,3-BPG, giving the 35°C embryos a significantly higher [2,3-BPG] at the latest stages(F=6.99, P<0.001).

In general, 35°C embryos maintained constant [ATP] and [2,3-BPG]between HH 41–42 and 43–44, while the concentrations of both organic phosphates tended to decrease between these same stages in the 38°C embryos (Fig. 5).

Hematocrit and hemoglobin content for 38°C late-stage chicken embryos agree with values reported in the literature(Tazawa et al., 1971; Tazawa, 1972; Baumann et al., 1983; Dzialowski et al., 2002). The 38°C embryos showed a significant increase in both hematological parameters between HH 41–42 and 43–44, again in line with previous reports (Tazawa et al.,1971).

Hypoxic incubation can increase both hematocrit and [Hb] of embryonic chicken blood (Dusseau and Hutchins,1988; Dzialowski et al.,2002), suggesting that increasing the total O₂-carrying capacity of the blood may be an important acclimation response ensuring adequate O₂ delivery to tissues prior to hatching. It is interesting to note that, in contrast to the 38°C embryos, there was no similar significant increase in hematocrit and [Hb] values in 35°C embryos with development from HH 41–42 to HH 43–44(Fig. 1). Perhaps hypothermic incubation precludes important acclimation responses that operate at 38°C.

The lower O₂-carrying capacity of the 35°C embryos corresponds to the significant increase in Hb–O₂ affinity between HH 41–42 and 43–44. The concurrence of these events suggests that the 35°C embryos may be maximizing the loading of O₂ at the respiratory surface to compensate for restriction of total O₂-carrying capacity. Blood gas data supports this notion,with the 35°C HH 43–44 embryos creating CAM venous (`arterialized')O₂ saturations of close to 90%(Fig. 4). A left-shifted curve should allow the embryo to saturate blood more completely at P_O2 values in vivo similar to those experienced by the 38°C embryos. HH 41–42 embryos incubated at 35°C and all 38°C embryos maintained CAM venous saturations of less than 80%, corresponding to the findings of other studies(Misson and Freeman, 1972; Tazawa and Mochizuki, 1978; Tazawa, 1980).

Hemoglobin–O₂ binding affinity represents a compromise between the conflicting needs of efficient loading of O₂ at the respiratory surface and adequate delivery of O₂ to the metabolically active tissues (Reeves,1984). Balancing these requirements becomes more complicated in the late stages of incubation in the chicken embryo. An egg incubating in normal temperature and O₂ conditions becomes increasingly hypoxic in ovo as the restrictions of diffusive gas exchange across the egg shell fail to meet the growing O₂ demands of the rapidly developing embryos (Wagensteen et al.,1970; Rahn et al.,1974; Ar et al.,1980; Tazawa,1980; Reeves,1984). As a consequence, air cell P_O2 declines from 18.6 kPa in early incubation to approximately 13.3 kPa just prior to pipping(Reeves, 1984; Burggren et al., 2000). Venous P_O2 follows that same pattern, decreasing from 10.6 kPa at D12 of incubation to 8.1 kPa by D17(Tazawa et al., 1971). Following an inverted pattern, blood P_CO2increases during development, resulting in a progressive decrease in blood pH(Tazawa, 1973).

Maintaining adequate O₂ loading in the face of these developmental respiratory transitions is assisted by Hb–O₂affinity transitions in the chicken embryo. Development is characterized by a progressive left-shift in the O₂ equilibrium curve, from a P₅₀ of 7.0 kPa at D8 to P₅₀ values at D18 down to 4.2–5.8 kPa (Tazawa,1980) or even as low as 4.1 kPa(Reeves, 1984). In the present study, chicken embryos incubated normothermically at 38°C generally followed the expected pattern, with a decrease in P₅₀ from 6.8 kPa at HH 41–42 down to 6.0 kPa at HH 43–44 (days 17 and 18 of incubation), with P₅₀ values at these later stages slightly higher than values reported in the literature. Embryos incubated at 35°C followed the expected pattern more closely, reaching a P₅₀ of about 4.0 kPa by HH 43–44.

Why hemoglobin(s) of HH 41–42 embryos remain insensitive to a 3°C change in measurement temperature is unclear. The sensitivity of whole blood to changes in temperature depends on its hemoglobin types(Baumann and Meuer, 1992). By D8 of incubation the blood of the chicken embryo contains three types of hemoglobins: a hatching hemoglobin (HbH), which plays only a minor role in O₂ transport, and two adult hemoglobins (HbA and HbD). In adult chickens the ratio of HbA to HbD is approximately 3:1, but HbD plays a larger role in the chicken embryo, existing at a ratio of 1:1.05 at D9 of incubation(Baumann and Meuer, 1992). The ratio of HbA to HbD continues to change in the late stages of incubation, and although the influence of each hemoglobin on the O₂ binding affinity of the blood is not yet fully understood, it is possible that the ratio of HbA to HbD may have an important impact on the temperature sensitivity and other characteristics of embryonic hemoglobin. Incubation temperature may impact the rate at which the ratio of HbA to HbD increases and in this fashion delay the onset of blood temperature sensitivity.

In addition to being altered by changes in embryo acid–base status,temperature and Hb type, O₂ affinity of embryonic chicken blood is,of course, also affected by organic phosphate concentrations(Baumann and Meuer, 1992). Up until approximately D12 of incubation in the chicken, ATP is the primary organic modifier of hemoglobin–O₂ affinity(Misson and Freeman, 1972; Bartlett and Borgese, 1976; Baumann and Meuer, 1992; Hochachka and Somero, 2002). The increase in hemoglobin–O₂ affinity between D8 and D18 of incubation corresponds to the period where metabolic rate reaches its maximum pre-pipping plateau, and hypoxia begins to develop within the egg. At this point the aerobic production of ATP is more difficult to achieve and there is a fall in blood [ATP] (Bartlett and Borgese, 1976; Nikinmaa,1990). As [ATP] drops, anaerobic production of 2,3-bisphosphoglycerate (2,3-BPG) increases and this organic phosphate then acts as the primary allosteric modifier of hemoglobin until after hatching,when inositol polyphosphate (IPP) becomes the adult allosteric modifier(Bartlett and Borgese, 1976; Nikinmaa, 1990).

Contrary to our expectations, the significant increase in hemoglobin–O₂ affinity between HH 41–42 and 43–44 in the 35°C embryos is not associated with any change in [ATP] or[2,3-BPG]. This suggests that in hypothermic embryos organic phosphates played little role in modifying hemoglobin–O₂ affinity, and that incubation temperature and the nature of the hemoglobins present were probably the primary reasons for the significant drop in P₅₀observed by HH 43–44. The effect of temperature on organic phosphate concentration in the chicken embryo has never been examined but, in general,[ATP] and [2,3-BPG] from these experiments are similar to those of numerous other studies on chicken embryos (Isaacks and Harkness, 1975; Isaacks et al., 1976; Bartlett and Borgese, 1976; Baumann and Meuer, 1992) and other precocial embryos including turkeys,pheasants, guinea fowl and ducks (Isaacks et al., 1976; Bartlett and Borgese, 1976).

Although incubation temperature had no apparent affect on organic phosphate concentrations in late-stage chicken embryos, hypoxic incubation results in an earlier switch from ATP to 2,3-BPG. Exposure to hyperoxia allows chicken embryos to maintain high levels of ATP throughout incubation and delay the developmental switch to 2,3-BPG (Ingermann et al., 1983; Baumann et al.,1986). The arterial and venous pH and P_O2 values collected for embryos incubated at 35°C and 38°C were similar at each stage of development(Table 1), indicating that a reduction in incubation temperature does not increase the level of hypoxic stress beyond what is normally experienced by the late-stage embryo. From previous research, it is clear that O₂ regulates both the concentrations of organic phosphates and the timing of the switch from ATP to 2,3-BPG. It is not surprising then, that incubation temperature alone did not result in obvious changes in the patterns of organic phosphate concentration in the late-stage chicken embryo.

The adult chicken can increase pulmonary ventilation and cardiac output to elevate O₂ transport rates in support of increased V̇_O2. In contrast,V̇_O2 in the chicken embryo is limited by diffusive gas exchange across the shell, especially in later developmental stages when O₂ demands are increasing(Tazawa et al., 1992). Essential changes occur in both [Hb] and blood Hb–O₂ affinity as the embryo reaches these final, most aerobically demanding stages. Interestingly, embryos incubated at 35°C and 38°C maintained surprisingly similar rates of O₂ consumption at equivalent stages of development – less than might be predicted with from typical temperature effects on animal metabolism(Black and Burggren, 2004). Yet, the mechanisms for achieving these similar O₂ demands appear quite different. Initially, HH 41–42 embryos present themselves very similarly at both incubation temperatures, with near identical basal metabolic rates, hematocrits, total hemoglobin contents and temperature-insensitive hemoglobins with low O₂ affinities. However, by HH 43–44 important differences emerge between 35°C and 38°C embryos. The 38°C embryos have developed significantly higher hematocrit and [Hb], and so presumably have a greater potential O₂-carrying capacity than the 35°C embryos. The 35°C embryos, on the other hand, have hemoglobins with higher O₂-binding affinities and are able to more completely saturate the blood at the respiratory gas exchanger surface. The similar metabolic rates of these two different incubation groups indicate that either set of compensatory changes can effectively support the O₂demands of embryos.

Although basal metabolic rates of embryos incubated at 35°C and 38°C are not significantly different(Black and Burggren, 2004),35°C embryos were less effective in responding metabolically to acute decreases in ambient temperature in late incubation. We speculate that at HH 43–44 the more temperature-sensitive hemoglobins of the 38°C embryos might provide an advantage over the temperature-insensitive hemoglobins of 35°C embryos at the same developmental stage. When exposed to decreases in ambient temperature, the hemoglobins of the 38°C embryos at HH 43–44 could have a higher O₂ affinity at the cooler respiratory surface and a lower O₂ affinity at the site of the warmer tissues,maximizing O₂ loading at the CAM. The presence of descending temperature gradients from the core of the embryo to the surface of the shell in the developing chicken egg is well established, making the temperature-sensitive nature of the hemoglobin molecule in the 38°C embryos an efficient strategy for O₂ transportation(Tazawa et al., 1988; Turner, 1990). The 35°C embryos at HH 43–44 maintain a high O₂-affinity regardless of temperature, which apparently still enables them to achieve efficient loading of O₂ but might inhibit efficient delivery of O₂ to the embryo. As one might expect, the 35°C embryos experienced a significant decline in V̇_O2 with a much smaller acute ambient temperature decrease than the 38°C embryos,suggesting a less developed or otherwise compromised capacity for thermoregulation through elevated aerobic metabolism.

Beyond HH 41–42, 35°C embryos experienced retarded hematological development and a delayed metabolic response to acute temperature changes compared to embryos incubated in control conditions of 38°C. These data providing the first supporting evidence in an endothermic species for`heterokairy' – a within-individual change in the timing of the onset of regulatory mechanisms during development(Spicer and Burggren, 2003). Exposure to decreased ambient temperature represents an ecologically relevant threat to the chicken hatchlings (Whittow and Tazawa, 1991), especially the 35°C embryos with their retarded thermoregulatory abilities. A series of future experiments will examine how the hematological and metabolic differences revealed in these studies affect the ability of the hatchlings to cope with such thermal stress. Of course, physiological temperature-dependent differences induced by incubation temperature differences are not likely to arise in a naturally incubated chicken embryo receiving conductive heat from the incubating hen. However, even though the experimental induction of such changes is not directly relevant to the ecology of natural incubation, the present study has shown that using incubation temperature as a developmental variable is clearly a very useful tool for examining the capability of developing physiological and hematological systems to respond to environmental challenges in ovo.

We thank Ed Dzialowski and Sheva Khorrami for helpful editorial comments. This work was supported by NSF Grant no. IBN-0128043 to W.W.B.