Metabolic response to cooling temperatures in chicken embryos and hatchlings after cold incubation

Comparative Biochemistry and Physiology, Part A 145 (2006) 441 – 448 www.elsevier.com/locate/cbpa

Metabolic response to cooling temperatures in chicken embryos and hatchlings after cold incubation Jacopo P. Mortola ⁎ Department of Physiology, McGill University, 3655 Promenade Sir William Osler, Montreal, Quebec, Canada H3G 1Y6 Received 1 April 2006; received in revised form 28 July 2006; accepted 31 July 2006 Available online 4 August 2006

Abstract We asked to what extent cold exposure during embryonic growth, and the accompanying hypometabolism, may interfere with the normal development of thermogenesis. White Leghorn chicken eggs were incubated in control conditions (38 °C) or at 36 or 35 °C. Embryos incubated at ̇ 2) and body weight (W) throughout a lower temperature (34 °C) failed to hatch. The cold-incubated embryos had lower oxygen consumption (VO ̇ 2 relationship of the cold-incubated embryos was as in incubation, and hatching was delayed by about, respectively, 1 and 2 days. The W–VO ̇ . At embryonic controls, indicating that cold-induced hypometabolism was at the expense of the growth, not the maintenance, component of VO 2 day E11, the metabolic response to changes in ambient temperature (T) over the 30–39 °C range was typically poikilothermic, with Q10 = 1.8–1.9, and similar among all sets of embryos. Toward the end of incubation (E20), the thermogenic responses of the cold-incubated embryos were significantly lower than in controls. This difference occurred also in the few-hour old hatchlings (H1), even though, at this time, W was similar among groups. Exposure to cold during only the last 3 days of incubation (from E18 to H1), i.e. during the developmental onset of the endothermic mechanisms, did not lower the thermogenic capacity of the hatchlings. In conclusion, sustained cold-induced hypometabolism throughout incubation blunted the rate of embryonic growth and the development of thermogenesis. This latter phenomenon could be an example of epigenetic regulation, i.e. of environmental factors exerting a long-term effect on gene expression. © 2006 Elsevier Inc. All rights reserved. Keywords: Embryonic development; Epigenetic adaptation; Hatching; Hypometabolism; Hypothermia; Thermoregulation; Development

1. Introduction In adult mammals and birds, cold exposure stimulates metabolic rate, as part of the endothermic defence of body temperature. A sustained exposure to cold improves the ability to withstand further cold challenges, through an enhancement of the thermogenic capacity (Arieli et al., 1979; Gordon, 1990). Also in neonatal animals cold exposure during the first postnatal weeks stimulates brown fat, non-shivering thermogenesis, and the biomolecular machinery involved in heat production (Skála and Hahn, 1974; Bertin and Portet, 1981; Cannon and Nedergaard, 1983; Bertin et al., 1993; Sant'Anna and Mortola, 2003). Contrary to the postnatal situation, little attention has been given to the implications that a prenatal cold-induced hypometabolic condition may have on the establishment of the ⁎ Tel.: +1 514 3984335; fax: +1 514 3987452. E-mail address: [email protected]. 1095-6433/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpa.2006.07.020

thermogenic capacity. For a large fraction of embryonic development birds and mammals have minimal endothermic capabilities, and a drop in temperature lowers their metabolic rate (Whittow and Tazawa, 1991; Power et al., 2004). In rats, sustained cold during pregnancy reduced fetal growth, resulting in offsprings of smaller size and with lower oxygen consumption ̇ 2) (Saetta et al., 1988). In mammals, however, the maternal (VO protection of the fetal temperature and the fact that maternal thermal stress has an impact on fetal development (FernandezCano, 1958; Hensleigh and Johnson, 1971; Tazumi et al., 2005) complicate the interpretation of the effects of cold-induced hypometabolism on fetal outcome. Experiments in avian preparations circumvent the issue of the maternal interference in the embryo's response to cold. In a recent study (Black and Burggren, 2004a,b), chicken eggs were incubated at 38 °C or at 35 °C; toward the end of incubation, those incubated in the cold were less ̇ 2 during cooling than the embryos capable of maintaining VO incubated at 38 °C. This difference, although small, remained

442

J.P. Mortola / Comparative Biochemistry and Physiology, Part A 145 (2006) 441–448

significant after taking into account the differences in developmental staging between the two groups. On the other hand, studies performed on the hatchlings of turkeys, chickens and Muscovy ducks, incubated at low temperatures during the last few days of incubation, indicated an increase in heat production compared to controls (Minne and Decuypere, 1984; Tzschentke and Nichelmann, 1999; Tzschentke et al., 2001). One interpretation of these discordant results is that cold incubation may have qualitatively different effects on heat production between the late embryo and the hatchling. Perhaps, this difference could be due to the fact that ̇ 2 during the terminal hatching, by lifting the O2-limitation on VO phases of embryonic development, causes a drastic and rapid change in heat production (Tazawa et al., 1988; Whittow and Tazawa, 1991; Nichelmann and Tzschentke, 2003; Black and Burggren, 2004a,b; Mortola and Labbè, 2005). Alternatively, cold exposures could have different effects depending on whether they occur throughout the whole incubation or only toward the end of it, this latter being the time when the thermogenic capacity begins to form. Hence, the primary goal of the current study was to assess the effects of cold exposure throughout incubation on the thermogenic response of both the late embryos and the hatchlings, to evaluate the possibility of a qualitative difference between these two developmental stages. The effects of cold exposure limited to the last few days of incubation have also been considered. A secondary aim was to examine the body weight–oxygen coṅ 2) relationship during embryonic development sumption (W–VO under various degrees of cold exposure. In fact, the drop in body weight (W) in embryos reared at 35 °C was proportional to their ̇ 2 (Black and Burggren, 2004a,b). Should the W– decrease in VO ̇ VO2 relationship of the cold-incubated embryos be similar to controls, it would imply that, of the total embryonic energy budget, the cold-induced hypometabolism has curtailed solely the component related to body growth. 2. Methods Freshly laid fertilized eggs of White Leghorn chickens (Gallus gallus) were obtained from a local supplier. Incubation started around midday (day 0). The eggs were weighed and placed in incubators (Hova-Bator, Savannah, GA, USA) pre-set either at the temperature (T) of 38 °C (Control group), or at lower T, i.e. 36, 35 or 34 °C. The incubators provided a 45° egg rotation four times a day. In all incubators, the relative humidity was 60%. A T-data logger and a hygrometer were placed inside the incubators; the former collected the T value every 10-min interval, while humidity was read daily. A first set of eggs was used to measure body growth and metabolic rate from embryonic day 8 to day 20, at 3-day intervals, and to consider the degree of hatchability under the low T conditions. The metabolic responses to changes in T were measured in embryos at the same chronological age (E11, E20), and in the hatchlings on the day of hatching (H1). 2.1. Gaseous metabolism Metabolic rate was measured by indirect calorimetry (oxẏ 2, and carbon dioxide production, V̇CO2), gen consumption, VO

with an open-flow methodology (Frappell et al., 1992) adapted to the chicken embryo (Menna and Mortola, 2002, 2003; Mortola and Labbè, 2005). Measurements were performed on embryos (grouped in sets of two) and on hatchlings (singly) in a respirometer, which consisted of a 300-mL plastic container maintained at the desired T by a water bath. A steady gas flow of 100 mL/min was continuously delivered through the respirometer, and controlled by a precision flowmeter. The inflow and outflow O2 and CO2 concentrations were monitored by gas analysers (OM-11, Beckman and CD-3A, Applied Electrochemistry) arranged in series, after the gas passed through a drying column. The output of the analysers was displayed on a computer monitor during on-line acquisition. ̇ 2 and V̇CO2 were computed from the flow rate and the inflow– VO outflow concentration difference over a period of 5 min at the end of each condition. The values, calculated at standard temperature, pressure, and dry conditions are presented in mL/ min (1 mL O2 STPD = 0.0446 mmol O2). 2.2. Protocols Eggs were weighed at the beginning of the incubation and on the day of the measurements. The average daily water loss (delta H2O) was computed from the difference between the initial and final egg weight, divided by the number of days of incubation. The difference in partial pressure of water P(H2O) between the egg and the incubator was calculated from the corresponding T and relative humidity. From these data, the water conductance of each egg G(H2O) was equal to delta H2O/delta P(H2O), mL day− 1 Torr− 1. At the end of the experiment, the embryo was killed by exposure to CO2 and cold, the egg was opened, the embryo examined to exclude gross anatomical abnormalities, and body weight (W) and head weight were measured on a digital scale. For each group of embryos, the metabolic response to changes in T was studied on day 11 (E11) and day 20 (E20) embryos, and in hatchlings within the first 24 h after hatching (H1). Embryos were studied in sets of two, while hatchlings were studied individually. Each group, at any given age, had N = 10, i.e., 10 sets of two embryos each, or 10 hatchlings. The metabolic response to T consisted of measurements performed at T = 39–36–33–30 °C, either in ascending or descending sequence, alternating the order among experiments. Each T was maintained for 1 h, and data refer to the last 5 min of each exposure. An additional group of eggs was incubated in control conditions (38 °C) until E18, when it was switched to 35 °C and left at that T until hatching. Then, the metabolic response to changes in T was studied in the hatchlings with the same protocol described above. 2.3. Normalisation and statistics ̇ 2 were normalised by embryo's W. The data of V̇O2 and VCO All group data are presented as means ± 1 S.E.M. Linear regression through the data points was employed to compute the Q10 values of E11 and E20, according to the van't Hoff

J.P. Mortola / Comparative Biochemistry and Physiology, Part A 145 (2006) 441–448 Table 1 Incubation conditions and egg water conductances

daily water loss and water conductance G(H2O) did not differ significantly among groups (Table 1).

Group

38 °C

Number of embryos Incubation temperature, °C Incubation relative humidity, % Egg mass at start, g Water loss, mL/day

119 133 97 37 37.6 (0.1)a 35.8 (0.1)b 35.0 (0.1)c 33.7 (0.2)d 65 (1)a 61 (2)a,b 59 (1)a,b 52 (8)b

G(H2O), mL day− 1 mm Hg− 1

36 °C

a

59.8 (0.5) 0.29 (0.02)a 16.7 (0.9)a

443

35 °C

a

59.5 (0.3) 0.26 (0.01)a 15.7 (0.4)a

34 °C

a

59.9 (0.3) 0.28 (0.01)a 16.3 (0.5)a

a

59.0 (0.6) 0.26 (0.01)a 16.7 (0.3)a

Values are means (±1 S.E.M.). N, total number of eggs. Water loss measured as the egg weight difference between day 0 and day 20. G(H2O), water conductance, computed over the 20-day incubation. Groups with the same letters are not statistically different (ANOVA, P b 0.05).

equation Q10 = (V̇O2'/V̇O2")10/(T′−T ʺ), where V̇O2' and V̇O2" are the values of V̇O2 at 39 °C (T′) and 30 °C (T ʺ). Statistical comparisons of the body growth and metabolic curves, at each embryonic age, and of the Q10 values, were done by ANOVA, with post-hoc Bonferroni's limitations for the comparisons among groups. For the statistical analysis of the T-metabolism curves, two-way repeated-measures ANOVA (T being the onefactor repetition) was adopted, followed by post-hoc Bonferroni's limitations for the comparisons of interest. These were, at any given T, the comparisons between controls and cold groups, and, within each group, the six possible comparisons among the four T. In all cases, a difference was considered statistically significant at P b 0.05. 3. Results

3.2. Embryo's growth Fig. 1 presents the body weight (W) of embryos, collected every 3 days from day 5 to hatching. At all ages, the cold groups had lower W, and the blunting in body growth was significantly more pronounced the lower the incubation T. At day 20, 48% of control embryos were in the external pipping phase, whereas this was the case of 8% of the embryos incubated at 36 °C, and of 0% for the other cold groups. In fact, only a few of the embryos incubated at 34 °C reached day 20, and none of them hatched. No further measurements were performed on this latter group of embryos, and their data are included in Figs. 1–3 solely for comparison purposes. The time of hatching in controls occurred at 20.7 ± 0.1 days of incubation. In the cold-incubated embryos, it occurred at 21.8 ± 0.1 days and 22.6 ± 0.1 days, for the 36 °C and 35 °C groups, respectively. The incubation times of the cold groups differed significantly from controls and from each other. On hatching day, the W of the hatchlings of the 36 °C group was 41.9 ± 0.6 g, not significantly different from controls (41.9 ± 0.4 g), while that of the 35 °C group was 38.9 ± 0.5 g, just slightly below that of the other two groups (P b 0.05). With the progression of incubation, the head of the embryo became a smaller percentage of the total embryo's W, from about 60% at day 8 to about 20% at day 20. The head weight–W relationship during development was described by a curve quite similar for all three groups (Fig. 2), suggesting that the incubation T did not modify the relative proportion of body growth.

3.1. Egg size and conductance 3.3. Gaseous metabolism during incubation The eggs of the various groups had statistically similar weight at the onset of incubation (Table 1). Also the average

Fig. 1. Growth curves during incubation for the four groups of embryos. Symbols are group means, bars represent 1 S.E.M. When not indicated, S.E.M. were within the symbol size. At any age, the cold groups had lower body weight than controls. Hatching time was significantly different among the groups. Body weight at hatching was lower in the 35 °C group than in the 36 °C or 38 °C groups. The 34 °C group (dashed line) failed to hatch.

Throughout incubation, embryos of the cold groups had lower ̇ 2) and carbon dioxide production (VCO ̇ 2) oxygen consumption (VO

Fig. 2. Head weight–body weight for the embryos incubated at 38, 36 and 35 °C. Values are group means, bars represent 1 S.E.M. Oblique dashed line indicate constant head weight–body weight ratios. The data of the 34 °C group are also included (dashed line), although these embryos failed to hatch.

444

J.P. Mortola / Comparative Biochemistry and Physiology, Part A 145 (2006) 441–448

̇ 2 of than control embryos (Fig. 3, left panels). On average, the VO the 36 °C group was 80%, and that of the 35 °C group was 59%, of control. Relative to W, the progression of oxygen consumption ̇ 2) and carbon dioxide production (VCO ̇ 2) throughout incuba(VO tion followed a curvilinear relationship, concave toward the Waxis (Fig. 3, right panels), because the W-specific gaseous metabolism dropped with the progression of incubation from about 25 mL kg− 1 min− 1 at day 8 to about 10 mL kg− 1 min− 1 at day 20 (oblique dashed lines). The metabolic rate of the cold embryos was appropriate for their W; in fact, the W-metabolism curve of the 36 °C- and 35 °C-cold groups overlapped the control relationship for the most part of incubation (Fig. 3, right panels), with the sole exception of the last data points, at E20. 3.4. Metabolic response to changes in temperature: E11 At E11, the W of the embryos used for these measurements (10 sets per group, of two embryos each) averaged 4.2 ± 0.1 g, 2.8 ± 0.1, and 2.5 ± 0.1 in the Control, 36 °C, and 35 °C groups, ̇ 2 values of the cold groups respectively. At any given T, the VO were lower than in controls (Fig. 4, top panel). This was due to the differences in embryo's W; in fact, after W-normalisation ̇ 2 relationships the curves overlapped. After expressing the T–VO in percent of the values at 39 °C (Fig. 4, bottom panel), it was easier to compare the shape of the individual curves; they did not differ significantly from each other. Also the Q10 values of

the three groups did not differ significantly (Controls, 1.9 ± 0.1; 36 °C group, 1.8 ± 0.1; 35 °C group, 1.7 ± 0.1). 3.5. Metabolic response to changes in temperature: E20 At E20, the W of the embryos used for these measurements (10 sets per group, of two embryos each) averaged 40.5 ± 1.6 g, 31.7 ± 1.1, and 24.6 (± 0.6) in the Control, 36 °C, and 35 °C groups, respectively. At any given T, the V̇O2 values of the cold groups were lower than in controls (Fig. 5, top left panel). After ̇ 2 relationships in percent of the values at expressing the T–VO 39 °C (Fig. 5, bottom left panel), the curves of the cold groups were steeper than in controls. In fact, at 30 °C, embryos of the 36 °C and the 35 °C groups dropped their V̇O2 to, respectively, 53 ± 2% and 50 ± 2% of the values they had at 39 °C; these percentages were significantly lower than in Controls (67 ± 3%). The average Q10 value of Controls was 1.56 (±0.07), that of the 36 °C group was 1.92 (± 0.07) (P b 0.015 from Controls) and that of the 35 °C group was 2.06 (± 0.09) (P b 0.001 from Controls). 3.6. Metabolic response to changes in temperature: H1 The W of the hatchlings used for these measurements (N = 10 per group) did not differ significantly among groups (39.3 ± 0.9 g, 39.2 ± 0.8 and 40.8 ± 0.7 in the Control, 36 °C, and

Fig. 3. Gaseous metabolism as function of incubation age (left panels) or embryos body weight (right panels) in the three groups of embryos. Values are group means, bars represent 1 S.E.M. When not indicated, S.E.M. were within the symbol size. Oblique dashed lines indicate constant metabolic rate–body weight ratios (mL/kg/ min). At any age, the cold groups had lower metabolic rate than controls, but appropriate for weight. The data of the 34 °C group are also included (dashed lines), although these embryos failed to hatch.

J.P. Mortola / Comparative Biochemistry and Physiology, Part A 145 (2006) 441–448

445

Upon exposure to lower T, their thermogenic response was similar to (at 36 °C), or greater than (at 33 °C and 30 °C), the response of control hatchlings (Fig. 6). 4. Discussion 4.1. Growth and hatching

Fig. 4. Ambient temperature–oxygen consumption relationships in embryos at day 11. Symbols are the average of 10 sets of two embryos each, except for the 35 °C group, where symbols are averages of 6 sets of two embryos each. Bars represent 1 S.E.M. Within each panel, different letters indicate significant ̇ (two-way ANOVA, with post-hoc Bonferroni's limitations, differences in VO 2 P b 0.05). For the 35 °C group the statistical results (not indicated) were as those of the 36 °C group (in italics).

35 °C groups, respectively), while the incubation age was greater the smaller the incubation T (20.5 ± 2 days, 22.1 ± 0.3 and 23.2 ± 0.3 in the Control, 36 °C, and 35 °C groups, respectively). In all hatchlings, cooling over the 36–33 °C range provoked a clear thermogenic response, which subsided with further cooling to 30 °C (Fig. 5, right top panel). After expressing the curves in percent of the 39 °C values (Fig. 5, right bottom panel), it became clear that the thermogenic responses of the cold groups were not as marked as in controls, with significantly lower thermogenic responses at 33 °C and 36 °C. At 30 °C, the Q10 effect lowered V̇O2 in all groups, and differences were no longer statistically significant. In any age group, the changes in V̇CO2 with T, and the differences among groups, were qualitatively similar to those ̇ 2. described for VO 3.7. Thermogenesis in H1 after cold exposure only in late incubation Embryos incubated in control conditions until E18 and at 35 °C thereafter hatched approximately at the same time (20.8 ± 0.1 days) of controls. Also their V̇O2 at 39 °C did not significantly differ from controls (17.4 ± 0.6 mL kg− 1 min− 1).

During embryonic development, incubation at 36 °C and ̇ 2 and W 35 °C lowered the developmental trajectories of VO proportionately, so that the relationship between the two (Fig. 3) was unaffected. Only toward the end of incubation (E20) the ̇ 2 data points of the cold groups were slightly below the W–VO control curve. Most likely, this small deviation reflected the fact that, toward the end of incubation, the heat produced by the embryo raises body temperature above ambient (Tazawa and Rahn, 1987), which is equivalent to an increase in the incubation temperature of the cold groups above the desired value. Therefore, at this age, the 35 °C and 36 °C temperatures used for the metabolic measurements must have been slightly lower than experienced in the incubator, causing some drop in V̇O2. The fact that, for the most part of incubation, the W–V̇O2 data points of the cold-incubated embryos fell on the control curve is ̇ 2 imposed by the lower T intriguing. It means that the drop in VO (Q10 effect) occurred solely at the expense of the rate of body growth, without compromising the metabolic contribution of the individual organs, or, according to a terminology previously adopted (Vleck and Vleck, 1987; Hoyt, 1987), without compromising the “maintenance component” of embryonic metabolism. In this context the finding of a constant W–head weight curve among groups is pertinent (Fig. 2), because the head represents a large percentage of embryo's W (between 60% at E7 and 30% at E15) and of its total metabolism. Also the chorioallantoic membrane, in cold conditions, has been shown to grow less, but keeping its functional adequacy to the embryo's gas-exchange needs (Tazawa, 1973). Differently, the ̇ 2 curve of the 34 °C embryos fell below the fact that the W–VO control curve implies that this degree of cold not only compromised the rate of growth but also lowered the maintenance component of tissue metabolism. Eventually, incubation at 34 °C was not compatible with the embryo's survival, and did not result in successful hatching. During the ectothermic phase, Q10 was about 2 (Fig. 4), as reported previously (Tazawa et al., 1989; Whittow and Tazawa, 1991; Mortola and Labbè, 2005). With this Q10, lowering T to ̇ 2 to 87% and 81% of the value at 36 °C or to 35 °C decreases VO 38 °C. Over the first 15-day period, the control embryos grew to 15 g (Fig. 1); the 36 °C and 35 °C embryos reached this W in 16.1 and 17.2 days, i.e. with a delay of, respectively, 7.3% and 14.6%. Therefore, a drop in V̇O2 of 13% corresponded to a growth retardation of 7%, and a drop of 19% to a growth retardation of 15%. These values do not agree with hatching times, which, in the 36 °C and 35 °C embryos, were delayed by only 5% and 9%. The difference between the growth delay throughout the first two-thirds of incubation and the delay in hatching time must be due to the fact that in the last few days,

446

J.P. Mortola / Comparative Biochemistry and Physiology, Part A 145 (2006) 441–448

Fig. 5. Ambient temperature–oxygen consumption relationships in 20-day-old embryos (left panels) and in hatchlings (right panels). Symbols are the average of 10 ̇ (two-way ANOVA, sets of two embryos each, or of 10 hatchlings. Bars represent 1 S.E.M. Within each panel, different letters indicate significant differences in VO 2 with post-hoc Bonferroni's limitations, P b 0.05). The statistical results of the 35 °C group (not indicated) were very similar to those of the 36 °C group (in italics).

with the onset of thermogenic mechanisms (Tazawa et al., 1988; Black and Burggren, 2004a,b; Mortola and Labbè, 2005) cold does not lower V̇O2 as much as it does during the previous days of incubation. In summary, for most of the cold incubation, the blunting effects on the normal growth are predictable (about 7– ̇ 2), and, as such, should be similar for 8% for a 10% drop in VO all species. Differently, toward the end of incubation, the growth delay depends on the development of the thermogenic capacity, which varies greatly among species (Tazawa et al., 1988; Nichelmann and Tzschentke, 2003). Although there are no comparative data, cold-induced hypometabolism should delay hatching in altricial species more than in chickens or other precocial birds, because in altricial species thermogenesis begins only after hatching.

cating an increase in the hatchling's thermogenic capacity. Differently, Black and Burggren (2004a,b) exposed chicken embryos to cold (35 °C) for the whole incubation, and tested the embryos toward the end of incubation; the results indicated a lower thermogenic capacity. In fact, upon gradual cooling, the

4.2. Thermogenic response to cold Since the early observation by Romanoff (1972), many studies have indicated that cold blunts the embryo's growth and metabolic rate, and delays hatching. Only a few, though, have examined the implications of the cold-induced hypometabolism on various aspects of heat control, either on hatchlings (Minne and Decuypere, 1984; Givisiez et al., 2003; Nichelmann, 2004; Tzschentke and Nichelmann, 1999; Tzschentke et al., 2001), or on late embryos (Black and Burggren, 2004a,b). All the studies on hatchlings consisted in cold exposures during the last few days of incubation, and the results had been uniform in indi-

Fig. 6. Ambient temperature–oxygen consumption relationships in a few-hourold hatchlings maintained at 38 °C throughout incubation (controls, open circles), or at 38 °C until E18 and at 35 °C thereafter (filled triangles). Values are expressed normalised by body weight (mL/kg/min). Symbols are the averages of, respectively, 10 and 12 animals. Bars represent 1 S.E.M. Symbols with the ̇ not different statistically (two-way ANOVA, same letters indicate values of VO 2 with post-hoc Bonferroni's limitations, P b 0.05).

J.P. Mortola / Comparative Biochemistry and Physiology, Part A 145 (2006) 441–448

35 °C-incubated embryos dropped V̇O2 slightly more, and at higher temperatures, than the 38 °C-incubated embryos. The current experiments in chicken embryos exposed throughout the whole incubation indicated that thermogenesis was blunted in both the embryos and the hatchlings; therefore, the results do not support the possibility that, following embryonic cold, the characteristics of the thermogenic capacity have qualitative differences before and after hatching. Rather, it seems that the timing of the embryonic cold exposure plays an important role. In fact, cold during the last days of incubation, i.e. when the thermogenic abilities have begun to develop, enhanced the thermogenic capacity of the hatchlings, not dissimilarly from what occurs also in mammals with cold exposures during postnatal development (Skála and Hahn, 1974; Bertin and Portet, 1981; Cannon and Nedergaard, 1983; Bertin et al., 1993; Sant'Anna and Mortola, 2003). At E11, the differences in absolute values of V̇O2 between cold-incubated and control embryos disappeared upon normal̇ 2 curves and the Q10 isation by W; also the shape of the T–VO values were the same in cold and control groups. At E20, the cold groups had significantly steeper T–V̇O2 curves, and higher Q10 values, than controls. The results at both ages are the expression of the developmental delay caused by the hypometabolism. However, Black and Burggren (2004a,b) found some ̇ 2 differences between cold- and warm-incubated embryos VO even after matching the developmental stages of the embryos; hence, they concluded that cold incubation not only blunted development, but also caused a significant delay in the relative timing of the onset of thermogenesis. The present results in the hatchlings, with a lower thermogenic response in the coldincubated group, give further support to Black and Burggren's view. In fact, hatching is a developmental stage not linked to chronological time, as shown by the fact that all chicks hatched spontaneously at different ages, after reaching a similar W. Hence, cold incubation mismatched the developmental trajectories of body growth and thermogenesis, forcing the latter to lag behind the former. This conclusion, therefore, is the opposite of what reached from experiments performed during postnatal development (e.g., Arieli et al., 1979; Gordon, 1990; Sant'Anna and Mortola, 2003) or, as mentioned previously, from experiments of cold exposure during the late phases of embryonic development. Presumably, this difference relates to the different metabolic responses to cold. During most of embryonic development cold causes a sustained hypometabolic state, while, at the end of incubation, with the beginning of the thermogenic control, and postnatally, cold causes a stimulation of V̇O2 and a hypermetabolic condition. In mammals, the small size could be, by itself, a factor reducing thermogenesis, because the large body surface, relative to body mass, favours heat loss and because of the reduced expression of the brown fat uncoupling protein (Mostyn et al., 2005). However, this cannot be a factor for the lower thermogenesis of the cold-incubated hatchlings. In fact, only the 35 °C group had smaller W, and the differences in thermogenesis were equally apparent in the 35 °C- and 36 °C groups. If cold-incubation had compromised the normal growth of the chorioallantoic membrane, the ensuing gas-diffusion impair-

447

ment and hypoxemia could have modified the thermogenic abilities of the hatchlings, as it occurs in newborn mammals (Mortola and Naso, 1998; Frappell et al., 1998). However, there is no reason to suspect that the cold-incubation altered the gas diffusion properties of the membranes. In fact, in previous work, the blood gas properties of cold-incubated embryos were as in controls (Tazawa, 1973), and, in the current experiments, the gas conductance of cold-incubated eggs was normal (Table 1). On the other hand, hematocrit and hemoglobin concentration were reduced in cold-incubated chicken embryos (Black and Burggren, 2004a,b); this decrease in the blood O2-carrying capacity, if persisted in the hatchlings, could have contributed to their reduced thermogenic capacity. 4.3. Conclusions Incubation in cold of a degree compatible with successful hatching lowered the embryo's trajectory of body growth in a way commensurate to the hypometabolism, indicating that the primary component curtailed by cold is the cost of growth, not the cost of tissue maintenance. Thermogenesis was blunted not only in the late embryos but also in the hatchlings, as long as the cold-induced hypometabolism occurred throughout the whole incubation. Heterokairy, a term proposed recently to indicate disturbances in the relative timing of the onset of different physiological processes (Spicer and Burggren, 2003), could be one aspect of epigenetic adaptation. This latter expression is used to indicate the possible role of environmental factors in influencing gene expression, as a mechanism to explain the long-term effects of early perturbations on the development of physiological mechanisms, including thermoregulation (Nichelmann, 2002, 2004). The depression of thermogenesis here observed in the hatchlings after cold incubation could be an indication of both heterokairy and epigenetic adaptation. It would be of interest to know to what extent such depression persists during postnatal growth, and whether or not it could occur with other interventions causing embryonic hypometabolism, such as sustained caloric restriction or hypoxia. Acknowledgements Funds for this research were from a Canadian Institute of Health Research grant. References Arieli, A., Meltzer, A., Berman, A., 1979. Seasonal acclimatization in the hen. Poult. Sci. 20, 505–514. Bertin, R., Portet, R., 1981. Effect of ambient temperature on lipid metabolism in brown fat during the perinatal period. Comp. Biochem. Physiol. B 70, 193–197. Bertin, R., De Marco, F., Mouroux, I., Portet, R., 1993. Postnatal development of nonshivering thermogenesis in rats: effects of rearing temperature. J. Dev. Physiol. 19, 9–15. Black, J.L., Burggren, W.W., 2004a. Acclimation to hypothermic incubation in developing chicken embryos (Gallus domesticus). I. Developmental effects and chronic and acute metabolic adjustments. J. Exp. Biol. 207, 1543–1552. Black, J.L., Burggren, W.W., 2004b. Acclimation to hypothermic incubation in developing chicken embryos (Gallus domesticus). II. Hematology and blood O2 transport. J. Exp. Biol. 207, 1553–1561.

448

J.P. Mortola / Comparative Biochemistry and Physiology, Part A 145 (2006) 441–448

Cannon, B., Nedergaard, J., 1983. Biochemical aspects of acclimation to cold. J. Therm. Biol. 8, 85–90. Fernandez-Cano, L., 1958. Effect of increase or decrease of body temperature and hypoxia on pregnancy in the rat. Fertil. Steril. 9, 455–459. Frappell, P., Lanthier, C., Baudinette, R.V., Mortola, J.P., 1992. Metabolism and ventilation in acute hypoxia: a comparative analysis in small mammalian species. Am. J. Physiol. 262, R1040–R1046. Frappell, P.B., León-Velarde, F., Aguero, L., Mortola, J.P., 1998. Response to cooling temperature in infants born at an altitude of 4330 meters. Am. J. Respir. Crit. Care Med. 158, 1751–1756. Givisiez, P.E.N., Furlan, R.L., Malheiros, E.B., Macari, M., 2003. Incubation and rearing temperature effects on Hsp70 levels and heat stress response in broilers. Can. J. Anim. Sci. 83, 213–220. Gordon, C.J., 1990. Thermal biology of the laboratory rat. Physiol. Behav. 47, 963–991. Hensleigh, P.A., Johnson, D.C., 1971. Heat stress effects during pregnancy. I. Retardation of fetal rat growth. Fertil. Steril. 22, 522–527. Hoyt, D.F., 1987. A new model of avian embryonic metabolism. J. Exp. Zool., Suppl. 1, 127–138. Menna, T.M., Mortola, J.P., 2002. Metabolic control of pulmonary ventilation in the developing chick embryo. Respir. Physiol. Neurobiol. 130, 43–55. Menna, T.M., Mortola, J.P., 2003. Ventilatory chemosensitivity in the chick embryo. Respir. Physiol. Neurobiol. 137, 69–79. Minne, B., Decuypere, E., 1984. Effects of late prenatal temperatures on some thermoregulatory aspects in young chickens. Arch. Exp. Vet.Med. 38, 374–383. Mortola, J.P., Labbè, K., 2005. Oxygen consumption of the chicken embryo: interaction between temperature and oxygenation. Respir. Physiol. Neurobiol. 146, 97–106. Mortola, J.P., Naso, L., 1998. Thermogenesis in newborn rats after prenatal or postnatal hypoxia. J. Appl. Physiol. 85, 84–90. Mostyn, A., Litten, J.C., Perkins, K.S., Euden, P.J., Corson, A.M., Symonds, M.E., Clarke, L., 2005. Influence of size at birth on the endocrine profiles and expression of uncoupling proteins in subcutaneous adipose tissue, lung, and muscle of neonatal pigs. Am. J. Physiol. 288, R1536–R1542. Nichelmann, M., 2002. Perinatal development of control systems in birds. Comp. Biochem. Physiol., A 131, 697–699. Nichelmann, M., 2004. Perinatal epigenetic temperature adaptation in avian species: comparison of turkey and Muscovy duck. J. Therm. Biol. 29, 613–619. Nichelmann, M., Tzschentke, B., 2003. Efficiency of thermoregulatory control elements in precocial poultry embryos. Avian Poult. Biol. Rev. 14, 1–19.

Power, G.G., Blood, A.B., Hunter, C.J., 2004. Perinatal thermal physiology, In: Polin, R.A., Fox, W.W., Abman, S.H. (Eds.), 3rd Edition. Fetal and Neonatal Physiology, vol. 55. Saunders, Philadelphia, pp. 541–548. ch. 55. Romanoff, A.L., 1972. Pathogenesis of the avian embryo. An analysis of causes of malformations and prenatal death, vol. 4. Wiley-Interscience, New York, pp. 57–106. ch. 4. Saetta, M., Noworaj, A., Mortola, J.P., 1988. Cold exposure of the pregnant rat and neonatal respiration. Exp. Biol. 47, 177–181. Sant'Anna, G., Mortola, J.P., 2003. Thermal and respiratory control in young rats exposed to cold during postnatal development. Comp. Biochem. Physiol. A 134, 449–459. Skála, J., Hahn, P., 1974. Changes in interscapular brown adipose tissue of the rat during perinatal and early postnatal development and after cold acclimation. VI. Effect of hormones and ambient temperature. Int. J. Biochem. 5, 95–106. Spicer, J.I., Burggren, W.W., 2003. Development of physiological regulatory systems: altering the timing of crucial events. Zool. 106, 91–99. Tazawa, H., 1973. Hypothermal effect on the gas exchange in chicken embryo. Respir. Physiol. 17, 21–31. Tazawa, H., Rahn, H., 1987. Temperature and metabolism of chick embryos and hatchlings after prolonged cooling. J. Exp. Zool., Suppl. 1, 105–109. Tazawa, H., Wakayama, H., Turner, J.S., Paganelli, C.V., 1988. Metabolic compensation for gradual cooling in developing chick embryos. Comp. Biochem. Physiol. A 89, 125–129. Tazawa, H., Okuda, A., Nakazawa, S., Whittow, G.C., 1989. Metabolic responses of chicken embryos to graded, prolonged alterations in ambient temperature. Comp. Biohem. Physiol., A 92, 613–617. Tazumi, T., Hori, E., Uwano, T., Umeno, K., Tanebe, K., Tabuchi, E., Ono, T., Nishijo, H., 2005. Effects of prenatal maternal stress by repeated cold environment on behavioral and emotional development in the rat offspring. Behav. Brain Res. 162, 153–160. Tzschentke, B., Nichelmann, M., 1999. Development of avian thermoregulatory system during the early postnatal period: development of thermoregulatory set-point. Ornis Fenn. 76, 189–198. Tzschentke, B., Basta, D., Nichelmann, M., 2001. Epigenetic temperature adaptation in birds: peculiarities and similarities in comparison to acclimation. News Biomed. Sci. 1, 26–31. Vleck, C.M., Vleck, D., 1987. Metabolism and energetics of avian embryos. J. Exp. Zool., Suppl. 1, 111–125. Whittow, G.C., Tazawa, H., 1991. The early development of thermoregulation in birds. Physiol. Zool. 57, 118–127.