- Email: [email protected]
Sensitivity of organ growth to chronically low oxygen levels during incubation in Red Junglefowl and domesticated chicken breeds
Sensitivity of organ growth to chronically low oxygen levels during incubation in Red Junglefowl and domesticated chicken breeds I. Lindgren and J. Altimiras1 Avian Behavioural Genomics and Physiology, Department of Physics, Chemistry and Biology, Linköping University, SE-58183 Linköping, Sweden affected by hypoxia in any breed, but a substantial stunting effect was observed for the liver and heart at late embryonic ages, with marked differences between breeds. In Red Junglefowl, only the heart was stunted. In White Leghorns, only the liver was stunted, and in broilers, both organs were stunted. These results can be explained in terms of the selection pressure on longterm production traits (reproductive effort) in White Leghorns, requiring a more efficient lipid metabolism, compared with the selection pressure on shorter-term production traits (growth) in broilers, requiring overall metabolic turnover and convective nutrient delivery to all tissues. At the same time, a remarkable sparing of the heart was observed in broilers and Red Junglefowl between E11 and E15, which suggests that cardiac growth can be manipulated during embryonic development. This result could be relevant for manipulating the phenotype of the heart for management purposes at a developmental stage when the bird is most versatile and phenotypically malleable.
Key words: broiler, domestication, organ growth, hypoxia, cardiovascular development 2011 Poultry Science 90:126–135 doi:10.3382/ps.2010-00996
INTRODUCTION
Numerous studies have evaluated postnatal phenotypic differences between strains either for selection purposes (from 2 wk onward) or to study the effect of different poultry management procedures. In relation to growth, for example, the growth rates during the first 2 wk posthatching are of critical importance for future growth (Ricklefs, 1985), but early studies disregarded the relevance of embryonic development because differences in embryonic growth between growth-selected strains could not be detected (Byerly, 1932), probably because of small differences in growth. Later on, however, a clear relationship between embryonic and postembryonic growth was demonstrated (Al-Murrani, 1978), and it was evident that the embryonic growth of modern broilers was faster than that of modern layers (Ohta et al., 2004). Other physiological differences, such as hormone levels and use of metabolic substrates during embryonic stages, between breeds have also been characterized (Christensen et al., 1995; Everaert et al., 2008).
Chicken selection experiments were initiated by the Romans more than 2,000 yr ago (Appleby et al. 2004). Selection programs were taken to new heights in the 20th century, with the aim of improving production traits such as muscle growth or egg productivity, which has given rise to the modern commercial broiler and layer strains. Growth rate, pectoral muscle mass, egg number, and egg size are some of the selected traits in genetic selection programs for poultry strains. In most cases, the genetic architecture of these traits is complex and it is driven by gene networks with multiple epistatic effects (Carlborg et al., 2003; Wahlberg et al., 2009; Rubin et al., 2010; Wright et al., 2010). ©2011 Poultry Science Association Inc. Received July 7, 2010. Accepted September 13, 2010. 1 Corresponding author: [email protected]
126
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
ABSTRACT Genetic selection programs have imposed large phenotypic changes in domesticated chicken breeds that are also apparent during embryonic development. Broilers, for example, have a faster growth rate before hatching in comparison with White Leghorns, indicating that the allocation of resources toward different functions already begins before hatching. Therefore, we hypothesized that embryonic organ growth would follow different developmental trajectories and would be differentially affected by an oxygen shortage during incubation. Heart, brain, and liver growth were studied in broiler, White Leghorn, and Red Junglefowl embryos at embryonic (E) ages E11, E13, E15, E18, and E20, and the results were fitted to growth allometric equations to determine the degree of organ stunting or sparing caused by low oxygen during incubation. Hypoxia caused a 3-fold larger mortality in Red Junglefowl than in the domesticated breeds, with a similar impairment of embryonic growth of 18%, coupled with a reduction in yolk utilization of 56%. Relative brain size was not
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
MATERIALS AND METHODS Incubation Conditions Fertilized eggs from 2 domestic chicken breeds and from Red Junglefowl were stored at 15 to 18°C for a maximum of 2 wk before incubation began. Eggs from the domestic breeds were obtained from local hatcheries, fast-growing Ross 308 broilers from SweHatch AB (Väderstad, Sweden) and Lohman White Leghorns from Gimranäs AB (Herrljunga, Sweden). Red Junglefowl eggs were obtained from an experimental captive population kept at the Götala Research Station (Swedish Agricultural University SLU, Skara, Sweden) since 1993 (Schütz et al., 2001). Incubation took place in commercial poultry incubators (model 25 HS, Masalles Comercial, Barcelona, Spain) at 37.8°C and 45% RH. Incubation was begun at 1500 h and was preceded by a 2-h period at room temperature. Eggs were turned automatically once every hour.
Experimental Protocol Two groups were set for all breeds, a control incubated under normoxic conditions (20.95% oxygen) and an experimental group incubated under low-oxygen
conditions (15% oxygen), which is referred to as the hypoxic treatment. Hypoxic incubation took place from the moment the eggs were placed into the incubator until hatching. Hypoxia was achieved by regulating the flow of nitrogen (nominal rate, 9 L/min) into the incubator using a flow meter (B-125-50, Porter Instrument Company, Hatfield, PA). Oxygen levels were continuously monitored using a galvanic oxygen probe placed inside the incubator (DD103 oxygen sensor, Pico Technology, St. Neots, Cambridgeshire, UK) and logged into a personal computer using proprietary software (PicoLog, Pico Technology). To account for changes in barometric pressure and to keep the oxygen levels as constant as possible, the prevailing barometric pressure was checked 3 times per week and the oxygen values were adjusted accordingly. All eggs from each breed were weighed at the beginning of incubation and systematically distributed in the 2 experimental groups to avoid initial differences in egg mass which could, in itself, have a confounding effect on the embryonic mass at the time of sampling between control and hypoxic groups (Metcalfe et al., 1981). Egg mass varies largely between breeds, but it is also dependent on the age of the hen. At the same hen age, Red Junglefowl eggs are smaller than White Leghorn eggs, which, in turn, are smaller than broiler eggs. This makes egg mass a confounding factor when trying to understand breed differences because the ratio between the area for oxygen diffusion and egg volume is dependent on egg size, and the mass at hatching is on average 68% of the egg mass (Wilson, 1991). To account for this potential confounding effect, 2 groups of broiler eggs were included in the experimental design. The first group came from a hen flock matched in age with the White Leghorn flock (45 wk; hereafter referred to as the “large broiler” group). The second group was chosen to match the egg size of the White Leghorn eggs so the broiler hens were younger (30 wk; hereafter referred to as the “small broiler” group). The initial characteristics of the eggs from the different experimental groups are presented in Table 1.
Handling of Embryos and Dissection All procedures were approved by the local Ethical Committee (Linköpings djurförsöksetiska nämnd diary number 45-03). Candling was carried out on d 8 of incubation and nonviable eggs were sorted out. Together with nonviable eggs found on each sampling day and in the final screening of the surplus embryos, mortality was estimated. All failed eggs were dissected out to assess the cause of failure and to estimate the approximate time of embryonic demise so that it was possible to subdivide failed embryos into early mortality (up to d 3, including nonfertile eggs) and late mortality (demise after d 3 of incubation). Eggs were sampled in the morning of incubation on d 11, 13, 15, 18, and 20 after euthanasia with sodium pentobarbital (injection of 0.25 mL of a 60 mg/mL solution into the egg).
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
The rate of postnatal somatic growth is also dependent on the growth of different organs involved in, for example, digestion (Lilja, 1983), as shown by the unconscious selection of larger intestinal masses in broilers (Jackson and Diamond, 1996; Konarzewski et al., 2000). The same is true during embryonic development, as seen in the larger digestive tracts in fast-growing quail (Lilja and Olsson, 1987). We therefore hypothesized that organ growth in different breeds would follow different developmental trajectories according to the specific requirements for the postembryonic allocation of resources. To evaluate this hypothesis, we analyzed embryonic heart, brain, and liver growth in 2 domesticated breeds and its wild ancestor, the Red Junglefowl, and fitted the data to organ growth equations to derive temporal physiological landmarks of organ growth during development. Because oxygen is the main limiting factor for the growth of the chicken embryo and oxygen diffusion into the egg is dependent on the partial pressure gradient between the blood and the environment (Metcalfe et al., 1981), we used chronic isobaric hypoxia as an experimental manipulation to challenge the embryo to prioritize the allocation of resources (i.e., oxygen) to the organs that would require it most. The results obtained will improve our understanding of how selection for productivity traits affects the allocation of resources between organs and how to manipulate this allocation by challenging the chicken at an early age when the organism can react to environmental manipulation.
127
128
Lindgren and Altimiras
The killed embryo was externalized from the eggshell by cutting open the amnion and leaving the chorioallantoic membrane attached to the eggshell. The yolk sac was cut where it entered the abdomen, and the yolk and embryo were weighed separately (Sartorius BP 221S, Sartorius AG, Göttingen, Germany). In embryos on embryonic day (E) 20, the partially internalized yolk was removed to obtain the yolk-free embryonic mass. The liver, heart, and brain were removed and weighed.
Modeling of Organ Growth and Extraction of Model Parameters
Bird type Small broilers Large broilers White Leghorns Red Junglefowl
Egg mass (g) 58.3 66.3 58.2 39.1
± ± ± ±
0.4 1.4 0.6 0.1
Yolk (%) 29.9 32.2 28.4 32.5
± ± ± ±
0.3 0.4 0.3 0.2
1The initial amount of yolk was obtained from a subset of eggs never set in incubation (n = 20).
module of Statistica (release 8, StatSoft Inc., Tulsa, OK) using the Levenberg-Marquardt algorithm and a least squares loss function. The model was first validated using a larger independent data set obtained using eggs from the same broiler breed under the same experimental conditions as in this study (n = 100; Lindgren, 2004) and was then applied to the specific data subsets (n = 50 each). Because the estimated model parameters (A, B, and C) are purely mathematical entities without biological significance, we focused our interest on some specific landmarks dictated by the interaction between the control and the hypoxic organ growth curves. The following landmarks were obtained by interpolation from the allometry equations (see Figure 1): 1) Mb at transition was the body mass at which the control and hypoxic curves intersected; 2) Mb at maximum sparing was the body mass at which the hypoxic curve was most different from the normoxic curve; 3) percentage of sparing (%sparing) was the intensity of the sparing effect at Mb at the maximum sparing; and 4) percentage of stunting (%stunting) was the intensity of the stunting effect close to hatching. Such landmarks provide relevant information in terms of the timing and magnitude of the sparing-stunting effects, and they are amenable to a discussion of the data in biological terms.
Figure 1. Graphical representation of the potential organ growth models in control conditions (continuous line) and hypoxic conditions (discontinuous line). Although a fourth model is theoretically possible, namely, one in which sparing occurs throughout development without stunting, it was deemed unlikely and is not represented here. The interaction between hypoxic incubation and control incubation was analyzed further by using the landmarks indicated on the curves: Mb at transition is the body mass at which the control and hypoxic curves intersect, Mb at maximum (max) sparing is the body mass at which the hypoxic curve is most different from the normoxic curve, percentage of sparing (%sparing) is the intensity of the sparing effect at Mb at maximum sparing, and percentage of stunting (%stunting) is the intensity of the stunting effect close to hatching.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
Three alternative models, shown in Figure 1, were considered to evaluate the effect of hypoxia on organ growth. Model A assumed that hypoxia had no effect on relative organ growth despite the expected effect on embryonic growth, whereas model B assumed that organ stunting would appear at a given embryonic mass. Finally, model C assumed that stunting could be preceded by a period of organ sparing. A fourth model, in which sparing occurred throughout incubation, is theoretically possible, but it was not considered because such a scenario was never apparent from the results. The mathematical formulation of the model for organ growth was based on the classical allometric equation: organ mass = AMBb . An extension to the classical equation was later formulated (Luecke et al., 1995): (B+C)´(logMb ) organ mass = AMb . This equation is referred to hereafter as the extended equation. In this equation, organ mass (in milligrams) is dependent on body mass (Mb in grams) and 3 independent parameters (A, B, and C) instead of only 2 (A and B), as in the classical equation. Parameter estimation and data fitting to the model above were carried out in the Nonlinear Estimation
Table 1. Initial egg mass (g) of eggs from the different breeds used in the study (mean ± SE)1
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
129
Statistical Analysis Data were analyzed for statistical significance by using an independent samples t-test corrected for multiple comparisons (Curran-Everett, 2000). When dealing with data expressed in percentages, the alternative MannWhitney nonparametric test was applied, followed by the same multiple-comparison correction. All tests were carried out in SPSS (version 15, SPSS Inc., Chicago, IL), with P < 0.05 being considered a significant difference. All values except in Table 1 are expressed as mean (SD) (Curran-Everett and Benos, 2004).
RESULTS Breed Characteristics and Developmental Mortality
Yolk-Free Embryonic Mass and Yolk and Organ Masses The relative changes in masses in the hypoxic group in relation to the control group are shown in Figure 2. As anticipated, hypoxia had a significant effect on embryonic growth at E13 and later ages in all breeds. In relative terms, hypoxia impaired somatic growth equally in all breeds, 11% on average at E13 and by Table 2. Mortality (%) of the different breeds under the 2 experimental treatments: control (normoxia, i.e., 20.95% oxygen) and hypoxia (15% oxygen)1 Control Bird type Small broilers Large broilers White Leghorns Red Junglefowl
Hypoxia
<3 d
>3 d
<3 d
>3 d
5.0 6.7 2.9 3.9
1.7 4.0 2.9 3.9
14.4 11.1 4.3 1.4
10.0 10.0 8.6 30.1
1Mortality results were split between the mortality during early development (before d 3 or nonfertilized) and the rest of development (between d 4 and sampling).
Figure 2. Relative change (%) in yolk-free embryonic mass (A) and yolk mass (B) for hypoxic embryos at different embryonic (E) developmental ages (E11, E13, E15, E18, and E20) and for the different breeds [broilers from small and large eggs (sb and lb, respectively), White Leghorn (wl), and Red Junglefowl (jf)]. The dotted line at 100% corresponds to the reference control values. Shaded bars indicate those experimental groups in which a significant effect of hypoxic incubation was found (P ≤ 0.05). Alternatively, open bars indicate no difference from controls. A box around the embryonic age on the x-axis indicates a significant difference between breeds (Kruskal-Wallis test). Asterisks indicate those breeds that differ significantly from the sb breed. Data are presented as mean (SD) (n = 10).
18% on average at later ages. In agreement with these data, hypoxia also impaired yolk utilization by 56% on average for all breeds at E20. At earlier ages, the pattern of yolk absorption was significantly impaired only in some breeds. Hypoxia also affected organ growth significantly, but the patterns differed for each organ and age, as shown in Figure 3. At E11, only brain mass was significantly reduced, by 12% on average, but neither liver nor heart mass was reduced. Interestingly, liver and heart masses were significantly increased in the large broilers. Because embryonic mass was not affected by hypoxia at E11, the implication is that relative brain mass decreased on hypoxic exposure. In late development, somatic growth was impaired to the same extent as brain growth, so no changes in relative brain size were apparent, and the effect was rather consistent between breeds. The liver and heart displayed different profiles, depending on the breed and developmental age. The general perspective shown in Figure 3 is that relative liver and heart masses increased in hypoxic embryos at E11 and E13 and decreased thereafter. The decrease was faster for liver mass, which showed an average decline in mass of 22% at E18 and E20. Heart mass, instead,
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
Eggs from the LB group had an average size of 66.3 g and a yolk content of 32.2%. Because the White Leghorn eggs from age-matched hens (45 wk) were 12% smaller, a second broiler group matching in egg size and yolk content, but not in hen age (30 wk instead), was included in the study (the small broiler group). The last group was obtained from Red Junglefowl, and the eggs were considerably smaller but had a larger yolk content. The specific masses and yolk contents for each strain are shown in Table 1. Developmental mortalities in the control birds ranged from 1.7 to 6.7%, a smaller percentage than in the hypoxic experimental groups, as shown in Table 2. The most significant result in this regard was the greater mortality in Red Junglefowl (30.1%), which was at least 3-fold larger than in broilers (10%) and White Leghorns (8.6%).
130
Lindgren and Altimiras
changed less even at the late stages for all breeds. The main exception to this was the marked declines in liver (41% change) and heart masses (31% change) in large broilers at E20, which were significantly different from all other breeds and markedly different from the 23% reduction in somatic growth.
Allometric Modeling of Organ Growth
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
Because organ growth is a nonlinear process depending on embryonic mass, genetic factors, and environmental influences, it is difficult to provide a comprehensive picture of the process from discrete sampling studies. Therefore, we fitted our results to obtain organ growth curves to draw more general conclusions regarding how organ growth was differentially affected by hypoxia in the different breeds. The first step required the validation of a suitable organ growth equation. For that purpose, we used a large data set of embryo and organ masses collected from large broiler eggs under the same conditions for an earlier study (Lindgren, 2004). The data were fitted to the extended allometric equation (Luecke et al., 1995). The results from the best fits to the extended equation are shown in Figure 4, together with the original data points. In the validation process, it was noticed that the classical equation sufficed to explain most of the variation in the data for the control data sets, and the extended equation improved the regression fit only marginally. The r2 averaged 0.962 for the classical equation and 0.965 for the extended equation between all breeds and organs. However, when applied to the hypoxic data sets, r2 of the extended equation was improved over r2 in the classical equation (0.964 for the extended equation vs. 0.957 for the classical equation). Important also was that the classical equation provided a skewed distribution of the residuals, so the fit was poorer at the higher end of the curve (i.e., larger organ masses and larger embryos), whereas the residuals for the extended equation were more evenly distributed. These results indicate that the classical equation could not account for the larger effects of hypoxia on organ mass in late development. These larger effects are fully compatible with our physiological understanding of organ growth under limiting conditions.
though in White Leghorns, a certain degree of stunting could make it closer to model B. Quantitatively, the intensity of the sparing and stunting effects was rather variable, as shown in Figure 5. The brain showed the least sparing and stunting for all breeds, whereas the liver and heart were spared early and stunted late by roughly 20% in large broilers. In Red Junglefowl, the heart, but not the liver, showed the sparing-stunting pattern. White Leghorns are the breed in which hypoxia had the smallest effect on the heart and brain, but the liver was spared by 10.3% and stunted by 18%.
Organ Growth Patterns and the Effect of Hypoxia Between Chicken Breeds Because of the facts presented above, only the extended equation was used in the subsequent analysis of organ growth between breeds. This modeling approach revealed many interesting facts. First was that from the 3 original growth models proposed in Figure 1, liver growth under hypoxic conditions was better represented by either model A or C, whereas the heart consistently showed early sparing and late stunting (model C). Finally, the brain was better represented by model A, al-
Figure 3. Relative change (%) in liver mass (A), heart mass (B), and brain mass (C) for hypoxic embryos at different embryonic (E) developmental ages (E11, E13, E15, E18, and E20) and for the different breeds [broilers from small and large eggs (sb and lb, respectively), White Leghorn (wl), and Red Junglefowl (jf)]. The dotted line at 100% corresponds to the reference control values. Shaded bars indicate those experimental groups in which a significant effect of hypoxic incubation was found (P ≤ 0.05). Alternatively, open bars indicate no difference from controls. A box around the embryonic age on the x-axis indicates a significant difference between breeds (Kruskal-Wallis test). Asterisks indicate those breeds that differ significantly from the sb breed. Data are presented as mean (SD) (n = 10).
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
The sparing effect was usually found between 14 and 16 d, and it was very consistent for the heart, for which all breeds showed it consistently at d 14.6 on average (Table 3). It was more variable in the liver (14.9 d), and the Red Junglefowl did not show it at all. The sparing effect on the brain was very small. The point of transition between sparing effects was found approximately 3 d later (17.5 d).
DISCUSSION Selection for Production Traits Decreases Susceptibility to Hypoxia and Promotes Efficiency in Embryonic Growth As expected, hypoxia caused a marked growth reduction that was similar for all breeds (Figure 2A), results
131
Brain and Liver Growth Are Differentially Affected by Hypoxia and Domestication
Figure 4. Validation of the extended allometric equation in large broiler eggs for liver (A), heart (B), and brain (C) masses as a function of yolk-free embryonic mass under control (open symbols) and hypoxic (solid symbols) conditions. Best fit to the modified allometric equation is shown for both data sets (continuous line, control; discontinuous line, hypoxia).
Taking advantage of the modeling approach, we successfully accounted for most of the variation in organ mass (96% on average) and inferred relevant biological parameters associated with organ growth in all breeds (Table 3). For this inference, we reevaluated the mathematical best-curve fits to estimate the degree of asymmetric fetal growth restriction (Lang et al., 2000) and the existence of sparing or stunting in different organs. Organ sparing is defined as a situation in which an organ maintains or increases its relative mass in relation to the embryonic mass even when facing adverse conditions, whereas organ stunting is a situation in which the relative organ mass decreases (Ounsted, 1988). Both concepts are routinely used in clinical fetal and neonatal medicine, even if the appropriateness of the term has been argued against (Hull et al., 1978). Sparing and stunting have also been used to characterize organ
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
comparable with many other studies in domesticated breeds [see Chan and Burggren (2005) for a partial review of previous results]. Growth retardation was initiated at about 13 d of incubation and continued until the time of hatching. Interestingly, the Red Junglefowl was more susceptible to hypoxia than were the domesticated breeds, as shown by the larger number of mortalities in Table 2, and this was paralleled by the decreased yolk utilization seen as early as E11 (Figure 2B), a time when embryonic mass was not yet impaired in any breed (Figure 2A). Because the yolk lipids are the primary source of energy for the chicken embryo (Vleck and Vleck, 1987) and fewer lipids can be catabolized in hypoxic conditions, it is to be expected that more yolk remains in hypoxic embryos, but it is remarkable that yolk accumulation was proportionally larger in Red Junglefowl embryos. The main drive for modern chicken selection has been the improvement of production traits, such as muscle growth or egg laying. Therefore, it is likely that enhanced metabolic performance has been selected for, and this can be translated into a more efficient investment toward muscle growth (broilers) or egg production (layers). In this light, we suggest that the reduced yolk utilization by Red Junglefowl already during development reveals a lower performance in handling metabolic substrates during hypoxia. At the same time, the growth retardation caused by hypoxia was similar between all breeds, indicating that the oxygen shortage imposed the same degree of burden on growth. Despite the differences during embryonic development in protein turnover (Muramatsu et al., 1990), yolk depletion, and basal metabolism (Sato et al., 2006), it is likely that the shortage of oxygen could not be compensated for by the heritable differences in metabolic substrate handling shown by broiler and layer breeds (Sato et al., 2006).
132
Lindgren and Altimiras
Table 3. Yolk-free embryonic mass (Mb) and estimated embryonic age at the time of the maximal sparing effect and at the transition time between sparing and stunting according to the allometric model shown as model C1
Organ
Chicken breed 2
Liver Heart Brain
sb lb wl jf sb lb wl jf sb lb wl jf
Mb at maximum sparing (g)
Age at maximum sparing (d)
Sparing effect (%)
Mb at transition (g)
Age at transition (d)
Stunting effect (%)
13.7 11.3 12.5 No cross 11.0 12.5 9.9 9.2 17.3 15.3 No cross 17.1
15.1 14.2 15.5 14.2 14.6 14.7 14.8 16.0 15.4 17.7
15.7 17.2 10.3 0.0 12.7 22.3 5.3 24.6 0.9 1.9 0.0 0.1
30.9 20.3 19.4 No cross 21.6 23.2 20.2 21.0 25.1 23.8 No cross 23.3
18.7 16.6 17.3 17.0 17.1 17.5 18.8 17.7 17.3 19.4
0.0 20.1 18.0 0.0 9.8 20.3 4.6 14.1 1.0 3.3 7.0 0.1
1Age was estimated by back interpolation of embryonic mass to the age growth curve for each breed. “No cross” entries indicate that the organ growth curves for control and hypoxic birds did not intersect so that model C was not applicable. 2sb = small broiler; lb = large broiler; wl = White Leghorn; jf = Red Junglefowl.
en brain (Asson-Batres et al., 1989). This fixed allometric growth during variable oxygenation conditions (even hyperoxia; Stock et al., 1983) indicates that brain growth is genetically linked to embryonic growth and shows little plasticity to environmental alterations. The absence of brain sparing (as defined by Ounsted, 1988), does not indicate that blood flow in the brain is not prioritized. It is almost certain that brain blood flow was maintained at the expense of blood flow to other organs, such as the liver, as shown in acute hypoxia (Peebles et al., 2003), but brain sparing is a generic concept that cannot discriminate among functional changes that occur in different brain areas, such as the cognitive impairments triggered by acute hypoxia at specific times of development (Rodricks et al., 2008). Liver and heart growth during hypoxia deviated from the typical allometric growth in normoxia and followed the pattern of model C (Figure 1) in most breeds. The
Figure 5. Schematic summary of the magnitude and timing of sparing and stunting in the different breeds. Three time points are defined for each breed and organ. Time point 1: maximum sparing effect. The intensity of the effect is shown as a circle; the larger the circle, the larger the intensity, as shown in the calibration bar. Time point 2: the intersection of the control and hypoxic growth curves, shown as a vertical line. Time point 3: the intensity of the stunting effect at 20 d of incubation. An X in the scheme indicates the absence of sparing or stunting because the data did not fit model C. sb = broilers from small eggs; lb = broilers from large eggs; wl = White Leghorns; jf = Red Junglefowl.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
growth in chickens raised at high altitudes or in altered environments (McCutcheon et al., 1982). Crucial to the discussion on organ sparing and stunting is the prioritization of the blood and oxygen supply to some organs to the detriment of others, and in this light, recent studies use the terms “brain sparing” and “heart sparing” to describe the situation in which the brain or heart blood flow is prioritized (Baschat et al., 1997; Haugen et al., 2005). Embryonic brain sparing, for instance, is linked to liver stunting because it is due to an increase in ductal venous flow, which bypasses the liver (Haugen et al., 2005). Our results demonstrate that hypoxia modifies allometric growth of the heart and liver, but not the brain. The brain grew in line with the embryo during the last week of development and showed no signs of sparing or stunting, which agrees with earlier evidence that hypoxia does not affect protein content, DNA content, or cytochrome oxidase activity in the hypoxic chick-
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
Effects of Hypoxia on Growth of the Heart Cardiac sparing is a known effect of chronic developmental hypoxia (McCutcheon et al., 1982; Dzialowski et al., 2002; Lindgren and Altimiras, 2009), but its onset and duration have been the subject of debate (Chan and Burggren, 2005). The basic tenet is that the reduction in convective oxygen flow caused by hypoxia is partly compensated for by changes in the oxygencarrying capacity (Baumann et al., 1983; Dzialowski et al., 2002) and cardiac output. A relatively enlarged
heart would increase stroke volume, and this in turn would increase cardiac output. Our results indicate that the heart is relatively larger in the second week and in the beginning of the third week of development (cardiac sparing up to E17; Figure 5), but it becomes stunted by 15% before hatching. Many studies have shown cardiac stunting before hatching (McCutcheon et al., 1982; Asson-Batres et al., 1989; Tintu et al., 2009), but others have not (Villamor et al., 2004; Lindgren and Altimiras, 2009). We have no explanation for the discrepancies, but considering that chronic exposure to 15% oxygen is close to the lethal oxygen threshold [12% oxygen is lethal at E5 (Miller et al., 2002, in White Leghorns, and our own unpublished results in broilers)], it could be due in part to small variations in oxygen concentration unaccounted for during incubation. Such variations (in particular in late embryos at the time when growth rate is greatest) would affect oxygen delivery and growth because partial pressure of oxygen in the blood in late development is low and lies in the steep part of the oxygen dissociation curve (Tazawa et al., 1988). The potential for cardiac sparing has also been shown in hypoxic conditions [40 and 60% oxygen (McCutcheon et al., 1982; Stock et al., 1983)], which indicates that cardiac development can accommodate to higher growth rates if oxygen limitations are removed. In human fetuses, heart sparing is associated with increased coronary flow (Baschat et al., 1997), and this may be the case in chickens because coronary flow is spared during acute hypoxia (Mulder et al., 1998). The heart is the only organ studied in which domesticated breeds differ markedly from one another. The same was observed when comparing randombred chickens and broilers (Christensen et al., 1995). The most parsimonious explanation is that selection for reproductive output capitalized on traits for metabolic efficiency and long-term productivity, relieving the selection pressure for short-term productivity.
Conclusions and Implications for Poultry Management Our results demonstrate that chicken genetic selection has resulted in important allometric alterations in organ growth that can already be observed during embryonic development. We interpret our findings in light of the selection history for each breed. In both broilers and layers (but not Red Junglefowl, their wild ancestor), the liver plays an important role as the main engine for protein and lipid metabolism, and liver stunting occurs when oxygen is restricted. Heart stunting is not apparent in layers but it is remarkable in Red Junglefowl and broilers, which we interpret as having a relaxed selection pressure for protein synthesis and accretion in layers to favor lipid metabolism and ultimately, reproductive output.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
turning points for maximal organ sparing were around E14 and the transition to stunting at E16 on average, which corresponded to a relative embryonic growth of 30 and 47% of the final hatching mass [estimated from previously published growth curves (Romanoff, 1960)]. These time estimates obtained from the models fit well with our physiological understanding of oxygen limitation in eggs, which occurs at about E16 to E17, as indicated by the leveling off of the metabolic rate (Rahn et al., 1974; Hoyt and Rahn, 1980), even if the embryo still needs to grow 53% of the final embryonic mass in the last 4 d of development. We interpret the good correlation between the model estimates and the dynamics of oxygen consumption during development as an indication that the organ growth landmarks identified with the modeling approach revealed a physiologically relevant time from which to evaluate the effect of hypoxia on the different breeds. The liver showed the most differences from the effects of hypoxia between breeds, that is, substantial sparing and stunting in domesticated breeds, but there was no effect on Red Junglefowl. Given the fundamental role of the liver in all metabolic processes in the organism, we suggest that selection for productivity traits has uncoupled liver growth from embryonic growth. Until E17, the embryo tries to spare liver function at the expense of other tissues, but it later succumbs to the oxygen shortage when the limitations become too severe. This would not occur in Red Junglefowl because there is no selection process favoring specific productivity traits in which the liver plays an active role. The key role of the liver in the enhanced productivity of broilers and layers is well documented. On the one hand, broilers have a high rate of protein accretion [high synthesis and low degradation (Muramatsu et al., 1990)], and it has been shown that fast-growing chickens have relatively larger livers (Konarzewski et al., 2000). On the other hand, layers grow slower but have higher rates of fat deposition, as indicated by the high levels of circulating very low density lipoproteins (Whitehead and Griffin, 1984; Liu et al., 1995), and these traits are pivotal to their high egg productivity. Thus, it is likely that the significant effects on liver embryonic growth observed in domesticated strains indicate that, already at an early age, the embryo is dependent on the capability of the liver to handle metabolic substrates.
133
134
Lindgren and Altimiras
Finally, the remarkable sparing effect on the heart from E11 to E15 (Figure 5) could be relevant as a management tool for future procedures involving alterations to the incubation environment because it suggests that the heart mass can be manipulated by limiting the oxygen supply. Our results pinpoint the existence of a critical development window for heart mass that spans E11 to E15, and we are committed to understanding the genetic and physiological implications at a deeper cellular and molecular level.
ACKNOWLEDGMENTS
REFERENCES Al-Murrani, W. K. 1978. Maternal effects on embryonic and postembryonic growth in poultry. Br. Poult. Sci. 19:277–281. Appleby, M. C., J. A. Mench, and B. O. Hughes. 2004. Poultry Behaviour and Welfare. CABI Publishing, London, UK. Asson-Batres, M. A., M. K. Stock, J. F. Hare, and J. Metcalfe. 1989. O2 effect on composition of chick embryonic heart and brain. Respir. Physiol. 77:101–109. Baschat, A. A., U. Gembruch, I. Reiss, L. Gortner, and K. Dietrich. 1997. Demonstration of fetal coronary blood flow by Doppler ultrasound in relation to arterial and venous flow velocity waveforms and perinatal outcome—The “heart-sparing effect.” Ultrasound Obstet. Gynecol. 9:162–172. Baumann, R., S. Padeken, E.-A. Haller, and T. Brilmayer. 1983. Effects of hypoxia on oxygen affinity, hemoglobin pattern, and blood volume of early chicken embryos. Am. J. Physiol. 244:R733–R741. Byerly, T. C. 1932. Growth of the chick embryo in relation to its food supply. J. Exp. Biol. 9:15–44. Carlborg, Ö., S. Kerje, S. Schütz, L. Jacobsson, P. Jensen, and L. Andersson. 2003. A global search reveals epistatic interaction between QTL for early growth in the chicken. Genome Res. 13:413–421. Chan, T., and W. Burggren. 2005. Hypoxic incubation creates differential morphological effects during specific developmental critical windows in the embryo of the chicken (Gallus gallus). Respir. Physiol. Neurobiol. 145:251–263. Christensen, V. L., G. B. Havenstein, and G. S. Davis. 1995. Egg characteristics, carbohydrate metabolism, and thyroid hormones in late chick embryos from different genetic lines. Poult. Sci. 74:551–562. Curran-Everett, D. 2000. Multiple comparisons: Philosophies and illustrations. Am. J. Physiol. Regul. Integr. Comp. Physiol. 279:R1–R8. Curran-Everett, D., and D. J. Benos. 2004. Guidelines for reporting statistics in journals published by the Americal Physiological Society. Am. J. Physiol. Regul. Integr. Comp. Physiol. 287:R247– R249. Dzialowski, E. M., D. Von Plettenberg, N. A. Elmonoufy, and W. W. Burggren. 2002. Chronic hypoxia alters the physiological and
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
This study was made possible by the generous provision of eggs by Pierre Molin from SweHatch AB in Väderstad, Sweden. We also acknowledge the assistance of Yi Lin Chou (currently at Taipei Medical School, Taipei, Taiwan) in data collection from Red Junglefowl and White Leghorn breeds. The study was financially supported by a project grant from FORMAS, the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (Stockholm, Sweden) and an infrastructure grant from the Wallenberg Foundation (Stockholm, Sweden).
morphological trajectories of developing chicken embryos. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 131:713–724. Everaert, N., H. Willemsen, L. De Smit, A. Witters, J. De Baerdemaeker, E. Decuypere, and V. Bruggeman. 2008. Comparison of a modern broiler and layer strain during embryonic development and the hatching process. Br. Poult. Sci. 49:574–582. Haugen, G., M. Hanson, T. Kiserud, S. Crozier, H. Inskip, and K. M. Godfrey. 2005. Fetal liver-sparing cardiovascular adaptations linked to mother’s slimness and diet. Circ. Res. 96:12–14. Hoyt, D. F., and H. Rahn. 1980. Respiration of avian embryos—A comparative analysis. Respir. Physiol. 39:255–264. Hull, D., J. Dobbing, and R. W. Miller. 1978. Definition, epidemiology and identification. Pages 69–84 in Abnormal Fetal Growth: Biological Bases and Consequences. F. Naftolin ed. Dahlem Konferenzen, Berlin, Germany. Jackson, S., and J. Diamond. 1996. Metabolic and digestive responses to artificial selection in chickens. Evolution 50:1638–1650. Konarzewski, M., A. Gavin, R. McDevitt, and I. R. Wallis. 2000. Metabolic and organ mass responses to selection for high growth rates in the domestic chicken (Gallus domesticus). Physiol. Biochem. Zool. 73:237–248. Lang, U., R. S. Baker, J. Khoury, and K. E. Clark. 2000. Effects of chronic reduction in uterine blood flow on fetal and placental growth in the sheep. Am. J. Physiol. Regul. Integr. Comp. Physiol. 279:R53–R59. Lilja, C. 1983. A comparative study of postnatal growth and organ development in some species of birds. Growth 47:317–339. Lilja, C., and U. Olsson. 1987. Changes in embryonic development associated with long-term selection for high growth rate in Japanese quail. Growth 51:301–308. Lindgren, I. 2004. Effects of chronic hypoxia on embryonic and organ growth during the last week of development in broiler chickens. LiU-IFM-Biol-Ex-1128. Linköpings Universitet, Linköping, Sweden. Lindgren, I., and J. Altimiras. 2009. Chronic prenatal hypoxia sensitizes β-adrenoceptors in the fetal heart but causes postnatal desensitization. Am. J. Physiol. Regul. Integr. Comp. Physiol. 297:H258–H264. Liu, Z., B. W. McBride, A. Lirette, and J. R. Chambers. 1995. Characterization of embryonic oxygen consumption of two broiler chicken lines differing in body fat content. Can. J. Anim. Sci. 75:115–119. Luecke, R. H., W. D. Wosilait, and J. F. Young. 1995. Mathematical representation of organ growth in the human embryo/fetus. Int. J. Biomed. Comput. 39:337–347. McCutcheon, I. E., J. Metcalfe, A. B. Metzenberg, and T. Ettinger. 1982. Organ growth in hyperoxic and hypoxic chick embryos. Respir. Physiol. 50:153–163. Metcalfe, J., I. E. McCutcheon, D. L. Francisco, A. B. Metzenberg, and J. E. Welch. 1981. Oxygen availability and growth of the chick embryo. Respir. Physiol. 46:81–88. Miller, S. L., L. R. Green, D. M. Peebles, M. A. Hanson, and C. E. Blanco. 2002. Effects of chronic hypoxia and protein malnutrition on growth in the developing chick. Am. J. Obstet. Gynecol. 186:261–267. Mulder, A. L. M., J. C. Van Golde, F. W. Prinzen, and C. E. Blanco. 1998. Cardiac output distribution in response to hypoxia in the chick embryo in the second half of the incubation time. J. Physiol. 508:281–287. Muramatsu, T., K. Hiramoto, and J. Okumura. 1990. Strain differences in whole-body protein turnover in the chicken embryo. Br. Poult. Sci. 31:91–99. Ohta, Y., T. Yoshida, and N. Tsushima. 2004. Comparison between broilers and layers for growth and protein use by embryos. Poult. Sci. 83:783–787. Ounsted, M. 1988. Small-for-dates babies: Heterogeneity and the concept of “brain sparing.” Paediatr. Perinat. Epidemiol. 2:365– 370. Peebles, D. M., J. C. Dixon, J. S. Thornton, E. B. Cady, A. Priest, S. L. Miller, C. E. Blanco, T. A. L. M. Mulder, R. J. Ordidge, and C. H. Rodeck. 2003. Magnetic resonance proton spectroscopy and diffusion weighted imaging of chick embryo brain in ovo. Brain Res. Dev. Brain Res. 141:101–107.
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
Tintu, A., E. Rouwet, S. Verlohren, J. Brinkmann, S. Ahmad, F. Crispi, M. van Bilsen, P. Carmeliet, A. C. Staff, M. Tjwa, I. Cetin, E. Gratacós, E. Hernandez-Andrade, L. Hofstra, M. Jacobs, W. H. Lamers, I. Morano, E. Safak, A. Ahmed, and F. le Noble. 2009. Hypoxia induces dilated cardiomyopathy in the chick embryo: Mechanism, intervention, and long-term consequences. PLoS ONE 4:e5155. Villamor, E., C. G. A. Kessels, K. Ruijtenbeek, R. J. Van Suylen, J. Belik, J. G. R. De Mey, and C. E. Blanco. 2004. Chronic in ovo hypoxia decreases pulmonary arterial contractile reactivity and induces biventricular cardiac enlargement in the chicken embryo. Am. J. Physiol. Regul. Integr. Comp. Physiol. 287:R642–R651. Vleck, C. M., and D. Vleck. 1987. Metabolism and energetics of avian embryos. J. Exp. Zool. Suppl. 1:111–125. Wahlberg, P., Ö. Carlborg, M. Foglio, X. Tordoir, A.-C. Syvänen, M. Lathrop, I. G. Gut, P. B. Siegel, and L. Andersson. 2009. Genetic analysis of an F2 intercross between two chicken lines divergently selected for body-weight. BMC Genomics 10:248. Whitehead, C. C., and H. D. Griffin. 1984. Development of divergent lines of lean and fat broilers using plasma very low density lipoprotein concentration as selection criterion: The first three generations. Br. Poult. Sci. 25:573–582. Wilson, H. R. 1991. Effects of size on hatchability, chick size and posthatching growth. Pages 279–283 in Avian Incubation. S. G. Tullett, ed. Butterworth-Heinemann, London, UK. Wright, D., C.-J. Rubin, Á. Martínez Barrio, K. E. Schütz, S. Kerje, H. Brändström, A. Kindmark, P. Jensen, and L. Andersson. 2010. Pleiotropic and epistatic QTL affecting behaviour and morphology in a large F2 intercross in the chicken. Mol. Ecol. 19:5079–5320. doi:10.1111/j.1365-294X.2010.04882.x
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
Rahn, H., C. V. Paganelli, and A. Ar. 1974. The avian egg: Air-cell gas tension, metabolism and incubation time. Respir. Physiol. 22:297–309. Ricklefs, R. E. 1985. Modification of growth and development of muscles of poultry. Poult. Sci. 64:1563–1576. Rodricks, C. L., M. E. Gibbs, G. Jenkin, and S. L. Miller. 2008. The effect of hypoxia at different embryonic ages on impairment of memory ability in chicks. Int. J. Dev. Neurosci. 26:113–118. Romanoff, A. L. 1960. The Avian Embryo: Structural and Functional Development. MacMillan Company, New York, NY. Rubin, C.-J., M. C. Zody, J. Eriksson, J. R. D. Meadows, E. Sherwood, M. T. Webster, L. Jiang, M. Ingman, T. Sharpe, S. Ka, F. Hallböök, F. Besnier, Ö. Carlborg, B. Bed’holm, M. Tixier-Boichard, P. Jensen, P. B. Siegel, K. Lindblad-Toh, and L. Andersson. 2010. Whole-genome resequencing reveals loci under selection during chicken domestication. Nature 464:587–591. Sato, M., T. Tachibana, and M. Furuse. 2006. Heat production and lipid metabolism in broiler and layer chickens during embryonic development. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 143:382–388. Schütz, K. E., B. Forkman, and P. Jensen. 2001. Domestication effects on foraging strategy, social behaviour and different fear responses: A comparison between the red junglefowl (Gallus gallus) and a modern layer strain. Appl. Anim. Behav. Sci. 74:1–14. Stock, M. K., D. L. Francisco, and J. Metcalfe. 1983. Organ growth in chick embryos incubated in 40% or 70% oxygen. Respir. Physiol. 52:1–11. Tazawa, H., S. Nakazawa, A. Okuda, and G. C. Whittow. 1988. Short-term effects of altered shell conductance on oxygen uptake and hematological variables of late chicken embryos. Respir. Physiol. 74:199–209.
135
Key words: broiler, domestication, organ growth, hypoxia, cardiovascular development 2011 Poultry Science 90:126–135 doi:10.3382/ps.2010-00996
INTRODUCTION
Numerous studies have evaluated postnatal phenotypic differences between strains either for selection purposes (from 2 wk onward) or to study the effect of different poultry management procedures. In relation to growth, for example, the growth rates during the first 2 wk posthatching are of critical importance for future growth (Ricklefs, 1985), but early studies disregarded the relevance of embryonic development because differences in embryonic growth between growth-selected strains could not be detected (Byerly, 1932), probably because of small differences in growth. Later on, however, a clear relationship between embryonic and postembryonic growth was demonstrated (Al-Murrani, 1978), and it was evident that the embryonic growth of modern broilers was faster than that of modern layers (Ohta et al., 2004). Other physiological differences, such as hormone levels and use of metabolic substrates during embryonic stages, between breeds have also been characterized (Christensen et al., 1995; Everaert et al., 2008).
Chicken selection experiments were initiated by the Romans more than 2,000 yr ago (Appleby et al. 2004). Selection programs were taken to new heights in the 20th century, with the aim of improving production traits such as muscle growth or egg productivity, which has given rise to the modern commercial broiler and layer strains. Growth rate, pectoral muscle mass, egg number, and egg size are some of the selected traits in genetic selection programs for poultry strains. In most cases, the genetic architecture of these traits is complex and it is driven by gene networks with multiple epistatic effects (Carlborg et al., 2003; Wahlberg et al., 2009; Rubin et al., 2010; Wright et al., 2010). ©2011 Poultry Science Association Inc. Received July 7, 2010. Accepted September 13, 2010. 1 Corresponding author: [email protected]
126
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
ABSTRACT Genetic selection programs have imposed large phenotypic changes in domesticated chicken breeds that are also apparent during embryonic development. Broilers, for example, have a faster growth rate before hatching in comparison with White Leghorns, indicating that the allocation of resources toward different functions already begins before hatching. Therefore, we hypothesized that embryonic organ growth would follow different developmental trajectories and would be differentially affected by an oxygen shortage during incubation. Heart, brain, and liver growth were studied in broiler, White Leghorn, and Red Junglefowl embryos at embryonic (E) ages E11, E13, E15, E18, and E20, and the results were fitted to growth allometric equations to determine the degree of organ stunting or sparing caused by low oxygen during incubation. Hypoxia caused a 3-fold larger mortality in Red Junglefowl than in the domesticated breeds, with a similar impairment of embryonic growth of 18%, coupled with a reduction in yolk utilization of 56%. Relative brain size was not
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
MATERIALS AND METHODS Incubation Conditions Fertilized eggs from 2 domestic chicken breeds and from Red Junglefowl were stored at 15 to 18°C for a maximum of 2 wk before incubation began. Eggs from the domestic breeds were obtained from local hatcheries, fast-growing Ross 308 broilers from SweHatch AB (Väderstad, Sweden) and Lohman White Leghorns from Gimranäs AB (Herrljunga, Sweden). Red Junglefowl eggs were obtained from an experimental captive population kept at the Götala Research Station (Swedish Agricultural University SLU, Skara, Sweden) since 1993 (Schütz et al., 2001). Incubation took place in commercial poultry incubators (model 25 HS, Masalles Comercial, Barcelona, Spain) at 37.8°C and 45% RH. Incubation was begun at 1500 h and was preceded by a 2-h period at room temperature. Eggs were turned automatically once every hour.
Experimental Protocol Two groups were set for all breeds, a control incubated under normoxic conditions (20.95% oxygen) and an experimental group incubated under low-oxygen
conditions (15% oxygen), which is referred to as the hypoxic treatment. Hypoxic incubation took place from the moment the eggs were placed into the incubator until hatching. Hypoxia was achieved by regulating the flow of nitrogen (nominal rate, 9 L/min) into the incubator using a flow meter (B-125-50, Porter Instrument Company, Hatfield, PA). Oxygen levels were continuously monitored using a galvanic oxygen probe placed inside the incubator (DD103 oxygen sensor, Pico Technology, St. Neots, Cambridgeshire, UK) and logged into a personal computer using proprietary software (PicoLog, Pico Technology). To account for changes in barometric pressure and to keep the oxygen levels as constant as possible, the prevailing barometric pressure was checked 3 times per week and the oxygen values were adjusted accordingly. All eggs from each breed were weighed at the beginning of incubation and systematically distributed in the 2 experimental groups to avoid initial differences in egg mass which could, in itself, have a confounding effect on the embryonic mass at the time of sampling between control and hypoxic groups (Metcalfe et al., 1981). Egg mass varies largely between breeds, but it is also dependent on the age of the hen. At the same hen age, Red Junglefowl eggs are smaller than White Leghorn eggs, which, in turn, are smaller than broiler eggs. This makes egg mass a confounding factor when trying to understand breed differences because the ratio between the area for oxygen diffusion and egg volume is dependent on egg size, and the mass at hatching is on average 68% of the egg mass (Wilson, 1991). To account for this potential confounding effect, 2 groups of broiler eggs were included in the experimental design. The first group came from a hen flock matched in age with the White Leghorn flock (45 wk; hereafter referred to as the “large broiler” group). The second group was chosen to match the egg size of the White Leghorn eggs so the broiler hens were younger (30 wk; hereafter referred to as the “small broiler” group). The initial characteristics of the eggs from the different experimental groups are presented in Table 1.
Handling of Embryos and Dissection All procedures were approved by the local Ethical Committee (Linköpings djurförsöksetiska nämnd diary number 45-03). Candling was carried out on d 8 of incubation and nonviable eggs were sorted out. Together with nonviable eggs found on each sampling day and in the final screening of the surplus embryos, mortality was estimated. All failed eggs were dissected out to assess the cause of failure and to estimate the approximate time of embryonic demise so that it was possible to subdivide failed embryos into early mortality (up to d 3, including nonfertile eggs) and late mortality (demise after d 3 of incubation). Eggs were sampled in the morning of incubation on d 11, 13, 15, 18, and 20 after euthanasia with sodium pentobarbital (injection of 0.25 mL of a 60 mg/mL solution into the egg).
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
The rate of postnatal somatic growth is also dependent on the growth of different organs involved in, for example, digestion (Lilja, 1983), as shown by the unconscious selection of larger intestinal masses in broilers (Jackson and Diamond, 1996; Konarzewski et al., 2000). The same is true during embryonic development, as seen in the larger digestive tracts in fast-growing quail (Lilja and Olsson, 1987). We therefore hypothesized that organ growth in different breeds would follow different developmental trajectories according to the specific requirements for the postembryonic allocation of resources. To evaluate this hypothesis, we analyzed embryonic heart, brain, and liver growth in 2 domesticated breeds and its wild ancestor, the Red Junglefowl, and fitted the data to organ growth equations to derive temporal physiological landmarks of organ growth during development. Because oxygen is the main limiting factor for the growth of the chicken embryo and oxygen diffusion into the egg is dependent on the partial pressure gradient between the blood and the environment (Metcalfe et al., 1981), we used chronic isobaric hypoxia as an experimental manipulation to challenge the embryo to prioritize the allocation of resources (i.e., oxygen) to the organs that would require it most. The results obtained will improve our understanding of how selection for productivity traits affects the allocation of resources between organs and how to manipulate this allocation by challenging the chicken at an early age when the organism can react to environmental manipulation.
127
128
Lindgren and Altimiras
The killed embryo was externalized from the eggshell by cutting open the amnion and leaving the chorioallantoic membrane attached to the eggshell. The yolk sac was cut where it entered the abdomen, and the yolk and embryo were weighed separately (Sartorius BP 221S, Sartorius AG, Göttingen, Germany). In embryos on embryonic day (E) 20, the partially internalized yolk was removed to obtain the yolk-free embryonic mass. The liver, heart, and brain were removed and weighed.
Modeling of Organ Growth and Extraction of Model Parameters
Bird type Small broilers Large broilers White Leghorns Red Junglefowl
Egg mass (g) 58.3 66.3 58.2 39.1
± ± ± ±
0.4 1.4 0.6 0.1
Yolk (%) 29.9 32.2 28.4 32.5
± ± ± ±
0.3 0.4 0.3 0.2
1The initial amount of yolk was obtained from a subset of eggs never set in incubation (n = 20).
module of Statistica (release 8, StatSoft Inc., Tulsa, OK) using the Levenberg-Marquardt algorithm and a least squares loss function. The model was first validated using a larger independent data set obtained using eggs from the same broiler breed under the same experimental conditions as in this study (n = 100; Lindgren, 2004) and was then applied to the specific data subsets (n = 50 each). Because the estimated model parameters (A, B, and C) are purely mathematical entities without biological significance, we focused our interest on some specific landmarks dictated by the interaction between the control and the hypoxic organ growth curves. The following landmarks were obtained by interpolation from the allometry equations (see Figure 1): 1) Mb at transition was the body mass at which the control and hypoxic curves intersected; 2) Mb at maximum sparing was the body mass at which the hypoxic curve was most different from the normoxic curve; 3) percentage of sparing (%sparing) was the intensity of the sparing effect at Mb at the maximum sparing; and 4) percentage of stunting (%stunting) was the intensity of the stunting effect close to hatching. Such landmarks provide relevant information in terms of the timing and magnitude of the sparing-stunting effects, and they are amenable to a discussion of the data in biological terms.
Figure 1. Graphical representation of the potential organ growth models in control conditions (continuous line) and hypoxic conditions (discontinuous line). Although a fourth model is theoretically possible, namely, one in which sparing occurs throughout development without stunting, it was deemed unlikely and is not represented here. The interaction between hypoxic incubation and control incubation was analyzed further by using the landmarks indicated on the curves: Mb at transition is the body mass at which the control and hypoxic curves intersect, Mb at maximum (max) sparing is the body mass at which the hypoxic curve is most different from the normoxic curve, percentage of sparing (%sparing) is the intensity of the sparing effect at Mb at maximum sparing, and percentage of stunting (%stunting) is the intensity of the stunting effect close to hatching.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
Three alternative models, shown in Figure 1, were considered to evaluate the effect of hypoxia on organ growth. Model A assumed that hypoxia had no effect on relative organ growth despite the expected effect on embryonic growth, whereas model B assumed that organ stunting would appear at a given embryonic mass. Finally, model C assumed that stunting could be preceded by a period of organ sparing. A fourth model, in which sparing occurred throughout incubation, is theoretically possible, but it was not considered because such a scenario was never apparent from the results. The mathematical formulation of the model for organ growth was based on the classical allometric equation: organ mass = AMBb . An extension to the classical equation was later formulated (Luecke et al., 1995): (B+C)´(logMb ) organ mass = AMb . This equation is referred to hereafter as the extended equation. In this equation, organ mass (in milligrams) is dependent on body mass (Mb in grams) and 3 independent parameters (A, B, and C) instead of only 2 (A and B), as in the classical equation. Parameter estimation and data fitting to the model above were carried out in the Nonlinear Estimation
Table 1. Initial egg mass (g) of eggs from the different breeds used in the study (mean ± SE)1
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
129
Statistical Analysis Data were analyzed for statistical significance by using an independent samples t-test corrected for multiple comparisons (Curran-Everett, 2000). When dealing with data expressed in percentages, the alternative MannWhitney nonparametric test was applied, followed by the same multiple-comparison correction. All tests were carried out in SPSS (version 15, SPSS Inc., Chicago, IL), with P < 0.05 being considered a significant difference. All values except in Table 1 are expressed as mean (SD) (Curran-Everett and Benos, 2004).
RESULTS Breed Characteristics and Developmental Mortality
Yolk-Free Embryonic Mass and Yolk and Organ Masses The relative changes in masses in the hypoxic group in relation to the control group are shown in Figure 2. As anticipated, hypoxia had a significant effect on embryonic growth at E13 and later ages in all breeds. In relative terms, hypoxia impaired somatic growth equally in all breeds, 11% on average at E13 and by Table 2. Mortality (%) of the different breeds under the 2 experimental treatments: control (normoxia, i.e., 20.95% oxygen) and hypoxia (15% oxygen)1 Control Bird type Small broilers Large broilers White Leghorns Red Junglefowl
Hypoxia
<3 d
>3 d
<3 d
>3 d
5.0 6.7 2.9 3.9
1.7 4.0 2.9 3.9
14.4 11.1 4.3 1.4
10.0 10.0 8.6 30.1
1Mortality results were split between the mortality during early development (before d 3 or nonfertilized) and the rest of development (between d 4 and sampling).
Figure 2. Relative change (%) in yolk-free embryonic mass (A) and yolk mass (B) for hypoxic embryos at different embryonic (E) developmental ages (E11, E13, E15, E18, and E20) and for the different breeds [broilers from small and large eggs (sb and lb, respectively), White Leghorn (wl), and Red Junglefowl (jf)]. The dotted line at 100% corresponds to the reference control values. Shaded bars indicate those experimental groups in which a significant effect of hypoxic incubation was found (P ≤ 0.05). Alternatively, open bars indicate no difference from controls. A box around the embryonic age on the x-axis indicates a significant difference between breeds (Kruskal-Wallis test). Asterisks indicate those breeds that differ significantly from the sb breed. Data are presented as mean (SD) (n = 10).
18% on average at later ages. In agreement with these data, hypoxia also impaired yolk utilization by 56% on average for all breeds at E20. At earlier ages, the pattern of yolk absorption was significantly impaired only in some breeds. Hypoxia also affected organ growth significantly, but the patterns differed for each organ and age, as shown in Figure 3. At E11, only brain mass was significantly reduced, by 12% on average, but neither liver nor heart mass was reduced. Interestingly, liver and heart masses were significantly increased in the large broilers. Because embryonic mass was not affected by hypoxia at E11, the implication is that relative brain mass decreased on hypoxic exposure. In late development, somatic growth was impaired to the same extent as brain growth, so no changes in relative brain size were apparent, and the effect was rather consistent between breeds. The liver and heart displayed different profiles, depending on the breed and developmental age. The general perspective shown in Figure 3 is that relative liver and heart masses increased in hypoxic embryos at E11 and E13 and decreased thereafter. The decrease was faster for liver mass, which showed an average decline in mass of 22% at E18 and E20. Heart mass, instead,
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
Eggs from the LB group had an average size of 66.3 g and a yolk content of 32.2%. Because the White Leghorn eggs from age-matched hens (45 wk) were 12% smaller, a second broiler group matching in egg size and yolk content, but not in hen age (30 wk instead), was included in the study (the small broiler group). The last group was obtained from Red Junglefowl, and the eggs were considerably smaller but had a larger yolk content. The specific masses and yolk contents for each strain are shown in Table 1. Developmental mortalities in the control birds ranged from 1.7 to 6.7%, a smaller percentage than in the hypoxic experimental groups, as shown in Table 2. The most significant result in this regard was the greater mortality in Red Junglefowl (30.1%), which was at least 3-fold larger than in broilers (10%) and White Leghorns (8.6%).
130
Lindgren and Altimiras
changed less even at the late stages for all breeds. The main exception to this was the marked declines in liver (41% change) and heart masses (31% change) in large broilers at E20, which were significantly different from all other breeds and markedly different from the 23% reduction in somatic growth.
Allometric Modeling of Organ Growth
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
Because organ growth is a nonlinear process depending on embryonic mass, genetic factors, and environmental influences, it is difficult to provide a comprehensive picture of the process from discrete sampling studies. Therefore, we fitted our results to obtain organ growth curves to draw more general conclusions regarding how organ growth was differentially affected by hypoxia in the different breeds. The first step required the validation of a suitable organ growth equation. For that purpose, we used a large data set of embryo and organ masses collected from large broiler eggs under the same conditions for an earlier study (Lindgren, 2004). The data were fitted to the extended allometric equation (Luecke et al., 1995). The results from the best fits to the extended equation are shown in Figure 4, together with the original data points. In the validation process, it was noticed that the classical equation sufficed to explain most of the variation in the data for the control data sets, and the extended equation improved the regression fit only marginally. The r2 averaged 0.962 for the classical equation and 0.965 for the extended equation between all breeds and organs. However, when applied to the hypoxic data sets, r2 of the extended equation was improved over r2 in the classical equation (0.964 for the extended equation vs. 0.957 for the classical equation). Important also was that the classical equation provided a skewed distribution of the residuals, so the fit was poorer at the higher end of the curve (i.e., larger organ masses and larger embryos), whereas the residuals for the extended equation were more evenly distributed. These results indicate that the classical equation could not account for the larger effects of hypoxia on organ mass in late development. These larger effects are fully compatible with our physiological understanding of organ growth under limiting conditions.
though in White Leghorns, a certain degree of stunting could make it closer to model B. Quantitatively, the intensity of the sparing and stunting effects was rather variable, as shown in Figure 5. The brain showed the least sparing and stunting for all breeds, whereas the liver and heart were spared early and stunted late by roughly 20% in large broilers. In Red Junglefowl, the heart, but not the liver, showed the sparing-stunting pattern. White Leghorns are the breed in which hypoxia had the smallest effect on the heart and brain, but the liver was spared by 10.3% and stunted by 18%.
Organ Growth Patterns and the Effect of Hypoxia Between Chicken Breeds Because of the facts presented above, only the extended equation was used in the subsequent analysis of organ growth between breeds. This modeling approach revealed many interesting facts. First was that from the 3 original growth models proposed in Figure 1, liver growth under hypoxic conditions was better represented by either model A or C, whereas the heart consistently showed early sparing and late stunting (model C). Finally, the brain was better represented by model A, al-
Figure 3. Relative change (%) in liver mass (A), heart mass (B), and brain mass (C) for hypoxic embryos at different embryonic (E) developmental ages (E11, E13, E15, E18, and E20) and for the different breeds [broilers from small and large eggs (sb and lb, respectively), White Leghorn (wl), and Red Junglefowl (jf)]. The dotted line at 100% corresponds to the reference control values. Shaded bars indicate those experimental groups in which a significant effect of hypoxic incubation was found (P ≤ 0.05). Alternatively, open bars indicate no difference from controls. A box around the embryonic age on the x-axis indicates a significant difference between breeds (Kruskal-Wallis test). Asterisks indicate those breeds that differ significantly from the sb breed. Data are presented as mean (SD) (n = 10).
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
The sparing effect was usually found between 14 and 16 d, and it was very consistent for the heart, for which all breeds showed it consistently at d 14.6 on average (Table 3). It was more variable in the liver (14.9 d), and the Red Junglefowl did not show it at all. The sparing effect on the brain was very small. The point of transition between sparing effects was found approximately 3 d later (17.5 d).
DISCUSSION Selection for Production Traits Decreases Susceptibility to Hypoxia and Promotes Efficiency in Embryonic Growth As expected, hypoxia caused a marked growth reduction that was similar for all breeds (Figure 2A), results
131
Brain and Liver Growth Are Differentially Affected by Hypoxia and Domestication
Figure 4. Validation of the extended allometric equation in large broiler eggs for liver (A), heart (B), and brain (C) masses as a function of yolk-free embryonic mass under control (open symbols) and hypoxic (solid symbols) conditions. Best fit to the modified allometric equation is shown for both data sets (continuous line, control; discontinuous line, hypoxia).
Taking advantage of the modeling approach, we successfully accounted for most of the variation in organ mass (96% on average) and inferred relevant biological parameters associated with organ growth in all breeds (Table 3). For this inference, we reevaluated the mathematical best-curve fits to estimate the degree of asymmetric fetal growth restriction (Lang et al., 2000) and the existence of sparing or stunting in different organs. Organ sparing is defined as a situation in which an organ maintains or increases its relative mass in relation to the embryonic mass even when facing adverse conditions, whereas organ stunting is a situation in which the relative organ mass decreases (Ounsted, 1988). Both concepts are routinely used in clinical fetal and neonatal medicine, even if the appropriateness of the term has been argued against (Hull et al., 1978). Sparing and stunting have also been used to characterize organ
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
comparable with many other studies in domesticated breeds [see Chan and Burggren (2005) for a partial review of previous results]. Growth retardation was initiated at about 13 d of incubation and continued until the time of hatching. Interestingly, the Red Junglefowl was more susceptible to hypoxia than were the domesticated breeds, as shown by the larger number of mortalities in Table 2, and this was paralleled by the decreased yolk utilization seen as early as E11 (Figure 2B), a time when embryonic mass was not yet impaired in any breed (Figure 2A). Because the yolk lipids are the primary source of energy for the chicken embryo (Vleck and Vleck, 1987) and fewer lipids can be catabolized in hypoxic conditions, it is to be expected that more yolk remains in hypoxic embryos, but it is remarkable that yolk accumulation was proportionally larger in Red Junglefowl embryos. The main drive for modern chicken selection has been the improvement of production traits, such as muscle growth or egg laying. Therefore, it is likely that enhanced metabolic performance has been selected for, and this can be translated into a more efficient investment toward muscle growth (broilers) or egg production (layers). In this light, we suggest that the reduced yolk utilization by Red Junglefowl already during development reveals a lower performance in handling metabolic substrates during hypoxia. At the same time, the growth retardation caused by hypoxia was similar between all breeds, indicating that the oxygen shortage imposed the same degree of burden on growth. Despite the differences during embryonic development in protein turnover (Muramatsu et al., 1990), yolk depletion, and basal metabolism (Sato et al., 2006), it is likely that the shortage of oxygen could not be compensated for by the heritable differences in metabolic substrate handling shown by broiler and layer breeds (Sato et al., 2006).
132
Lindgren and Altimiras
Table 3. Yolk-free embryonic mass (Mb) and estimated embryonic age at the time of the maximal sparing effect and at the transition time between sparing and stunting according to the allometric model shown as model C1
Organ
Chicken breed 2
Liver Heart Brain
sb lb wl jf sb lb wl jf sb lb wl jf
Mb at maximum sparing (g)
Age at maximum sparing (d)
Sparing effect (%)
Mb at transition (g)
Age at transition (d)
Stunting effect (%)
13.7 11.3 12.5 No cross 11.0 12.5 9.9 9.2 17.3 15.3 No cross 17.1
15.1 14.2 15.5 14.2 14.6 14.7 14.8 16.0 15.4 17.7
15.7 17.2 10.3 0.0 12.7 22.3 5.3 24.6 0.9 1.9 0.0 0.1
30.9 20.3 19.4 No cross 21.6 23.2 20.2 21.0 25.1 23.8 No cross 23.3
18.7 16.6 17.3 17.0 17.1 17.5 18.8 17.7 17.3 19.4
0.0 20.1 18.0 0.0 9.8 20.3 4.6 14.1 1.0 3.3 7.0 0.1
1Age was estimated by back interpolation of embryonic mass to the age growth curve for each breed. “No cross” entries indicate that the organ growth curves for control and hypoxic birds did not intersect so that model C was not applicable. 2sb = small broiler; lb = large broiler; wl = White Leghorn; jf = Red Junglefowl.
en brain (Asson-Batres et al., 1989). This fixed allometric growth during variable oxygenation conditions (even hyperoxia; Stock et al., 1983) indicates that brain growth is genetically linked to embryonic growth and shows little plasticity to environmental alterations. The absence of brain sparing (as defined by Ounsted, 1988), does not indicate that blood flow in the brain is not prioritized. It is almost certain that brain blood flow was maintained at the expense of blood flow to other organs, such as the liver, as shown in acute hypoxia (Peebles et al., 2003), but brain sparing is a generic concept that cannot discriminate among functional changes that occur in different brain areas, such as the cognitive impairments triggered by acute hypoxia at specific times of development (Rodricks et al., 2008). Liver and heart growth during hypoxia deviated from the typical allometric growth in normoxia and followed the pattern of model C (Figure 1) in most breeds. The
Figure 5. Schematic summary of the magnitude and timing of sparing and stunting in the different breeds. Three time points are defined for each breed and organ. Time point 1: maximum sparing effect. The intensity of the effect is shown as a circle; the larger the circle, the larger the intensity, as shown in the calibration bar. Time point 2: the intersection of the control and hypoxic growth curves, shown as a vertical line. Time point 3: the intensity of the stunting effect at 20 d of incubation. An X in the scheme indicates the absence of sparing or stunting because the data did not fit model C. sb = broilers from small eggs; lb = broilers from large eggs; wl = White Leghorns; jf = Red Junglefowl.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
growth in chickens raised at high altitudes or in altered environments (McCutcheon et al., 1982). Crucial to the discussion on organ sparing and stunting is the prioritization of the blood and oxygen supply to some organs to the detriment of others, and in this light, recent studies use the terms “brain sparing” and “heart sparing” to describe the situation in which the brain or heart blood flow is prioritized (Baschat et al., 1997; Haugen et al., 2005). Embryonic brain sparing, for instance, is linked to liver stunting because it is due to an increase in ductal venous flow, which bypasses the liver (Haugen et al., 2005). Our results demonstrate that hypoxia modifies allometric growth of the heart and liver, but not the brain. The brain grew in line with the embryo during the last week of development and showed no signs of sparing or stunting, which agrees with earlier evidence that hypoxia does not affect protein content, DNA content, or cytochrome oxidase activity in the hypoxic chick-
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
Effects of Hypoxia on Growth of the Heart Cardiac sparing is a known effect of chronic developmental hypoxia (McCutcheon et al., 1982; Dzialowski et al., 2002; Lindgren and Altimiras, 2009), but its onset and duration have been the subject of debate (Chan and Burggren, 2005). The basic tenet is that the reduction in convective oxygen flow caused by hypoxia is partly compensated for by changes in the oxygencarrying capacity (Baumann et al., 1983; Dzialowski et al., 2002) and cardiac output. A relatively enlarged
heart would increase stroke volume, and this in turn would increase cardiac output. Our results indicate that the heart is relatively larger in the second week and in the beginning of the third week of development (cardiac sparing up to E17; Figure 5), but it becomes stunted by 15% before hatching. Many studies have shown cardiac stunting before hatching (McCutcheon et al., 1982; Asson-Batres et al., 1989; Tintu et al., 2009), but others have not (Villamor et al., 2004; Lindgren and Altimiras, 2009). We have no explanation for the discrepancies, but considering that chronic exposure to 15% oxygen is close to the lethal oxygen threshold [12% oxygen is lethal at E5 (Miller et al., 2002, in White Leghorns, and our own unpublished results in broilers)], it could be due in part to small variations in oxygen concentration unaccounted for during incubation. Such variations (in particular in late embryos at the time when growth rate is greatest) would affect oxygen delivery and growth because partial pressure of oxygen in the blood in late development is low and lies in the steep part of the oxygen dissociation curve (Tazawa et al., 1988). The potential for cardiac sparing has also been shown in hypoxic conditions [40 and 60% oxygen (McCutcheon et al., 1982; Stock et al., 1983)], which indicates that cardiac development can accommodate to higher growth rates if oxygen limitations are removed. In human fetuses, heart sparing is associated with increased coronary flow (Baschat et al., 1997), and this may be the case in chickens because coronary flow is spared during acute hypoxia (Mulder et al., 1998). The heart is the only organ studied in which domesticated breeds differ markedly from one another. The same was observed when comparing randombred chickens and broilers (Christensen et al., 1995). The most parsimonious explanation is that selection for reproductive output capitalized on traits for metabolic efficiency and long-term productivity, relieving the selection pressure for short-term productivity.
Conclusions and Implications for Poultry Management Our results demonstrate that chicken genetic selection has resulted in important allometric alterations in organ growth that can already be observed during embryonic development. We interpret our findings in light of the selection history for each breed. In both broilers and layers (but not Red Junglefowl, their wild ancestor), the liver plays an important role as the main engine for protein and lipid metabolism, and liver stunting occurs when oxygen is restricted. Heart stunting is not apparent in layers but it is remarkable in Red Junglefowl and broilers, which we interpret as having a relaxed selection pressure for protein synthesis and accretion in layers to favor lipid metabolism and ultimately, reproductive output.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
turning points for maximal organ sparing were around E14 and the transition to stunting at E16 on average, which corresponded to a relative embryonic growth of 30 and 47% of the final hatching mass [estimated from previously published growth curves (Romanoff, 1960)]. These time estimates obtained from the models fit well with our physiological understanding of oxygen limitation in eggs, which occurs at about E16 to E17, as indicated by the leveling off of the metabolic rate (Rahn et al., 1974; Hoyt and Rahn, 1980), even if the embryo still needs to grow 53% of the final embryonic mass in the last 4 d of development. We interpret the good correlation between the model estimates and the dynamics of oxygen consumption during development as an indication that the organ growth landmarks identified with the modeling approach revealed a physiologically relevant time from which to evaluate the effect of hypoxia on the different breeds. The liver showed the most differences from the effects of hypoxia between breeds, that is, substantial sparing and stunting in domesticated breeds, but there was no effect on Red Junglefowl. Given the fundamental role of the liver in all metabolic processes in the organism, we suggest that selection for productivity traits has uncoupled liver growth from embryonic growth. Until E17, the embryo tries to spare liver function at the expense of other tissues, but it later succumbs to the oxygen shortage when the limitations become too severe. This would not occur in Red Junglefowl because there is no selection process favoring specific productivity traits in which the liver plays an active role. The key role of the liver in the enhanced productivity of broilers and layers is well documented. On the one hand, broilers have a high rate of protein accretion [high synthesis and low degradation (Muramatsu et al., 1990)], and it has been shown that fast-growing chickens have relatively larger livers (Konarzewski et al., 2000). On the other hand, layers grow slower but have higher rates of fat deposition, as indicated by the high levels of circulating very low density lipoproteins (Whitehead and Griffin, 1984; Liu et al., 1995), and these traits are pivotal to their high egg productivity. Thus, it is likely that the significant effects on liver embryonic growth observed in domesticated strains indicate that, already at an early age, the embryo is dependent on the capability of the liver to handle metabolic substrates.
133
134
Lindgren and Altimiras
Finally, the remarkable sparing effect on the heart from E11 to E15 (Figure 5) could be relevant as a management tool for future procedures involving alterations to the incubation environment because it suggests that the heart mass can be manipulated by limiting the oxygen supply. Our results pinpoint the existence of a critical development window for heart mass that spans E11 to E15, and we are committed to understanding the genetic and physiological implications at a deeper cellular and molecular level.
ACKNOWLEDGMENTS
REFERENCES Al-Murrani, W. K. 1978. Maternal effects on embryonic and postembryonic growth in poultry. Br. Poult. Sci. 19:277–281. Appleby, M. C., J. A. Mench, and B. O. Hughes. 2004. Poultry Behaviour and Welfare. CABI Publishing, London, UK. Asson-Batres, M. A., M. K. Stock, J. F. Hare, and J. Metcalfe. 1989. O2 effect on composition of chick embryonic heart and brain. Respir. Physiol. 77:101–109. Baschat, A. A., U. Gembruch, I. Reiss, L. Gortner, and K. Dietrich. 1997. Demonstration of fetal coronary blood flow by Doppler ultrasound in relation to arterial and venous flow velocity waveforms and perinatal outcome—The “heart-sparing effect.” Ultrasound Obstet. Gynecol. 9:162–172. Baumann, R., S. Padeken, E.-A. Haller, and T. Brilmayer. 1983. Effects of hypoxia on oxygen affinity, hemoglobin pattern, and blood volume of early chicken embryos. Am. J. Physiol. 244:R733–R741. Byerly, T. C. 1932. Growth of the chick embryo in relation to its food supply. J. Exp. Biol. 9:15–44. Carlborg, Ö., S. Kerje, S. Schütz, L. Jacobsson, P. Jensen, and L. Andersson. 2003. A global search reveals epistatic interaction between QTL for early growth in the chicken. Genome Res. 13:413–421. Chan, T., and W. Burggren. 2005. Hypoxic incubation creates differential morphological effects during specific developmental critical windows in the embryo of the chicken (Gallus gallus). Respir. Physiol. Neurobiol. 145:251–263. Christensen, V. L., G. B. Havenstein, and G. S. Davis. 1995. Egg characteristics, carbohydrate metabolism, and thyroid hormones in late chick embryos from different genetic lines. Poult. Sci. 74:551–562. Curran-Everett, D. 2000. Multiple comparisons: Philosophies and illustrations. Am. J. Physiol. Regul. Integr. Comp. Physiol. 279:R1–R8. Curran-Everett, D., and D. J. Benos. 2004. Guidelines for reporting statistics in journals published by the Americal Physiological Society. Am. J. Physiol. Regul. Integr. Comp. Physiol. 287:R247– R249. Dzialowski, E. M., D. Von Plettenberg, N. A. Elmonoufy, and W. W. Burggren. 2002. Chronic hypoxia alters the physiological and
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
This study was made possible by the generous provision of eggs by Pierre Molin from SweHatch AB in Väderstad, Sweden. We also acknowledge the assistance of Yi Lin Chou (currently at Taipei Medical School, Taipei, Taiwan) in data collection from Red Junglefowl and White Leghorn breeds. The study was financially supported by a project grant from FORMAS, the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (Stockholm, Sweden) and an infrastructure grant from the Wallenberg Foundation (Stockholm, Sweden).
morphological trajectories of developing chicken embryos. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 131:713–724. Everaert, N., H. Willemsen, L. De Smit, A. Witters, J. De Baerdemaeker, E. Decuypere, and V. Bruggeman. 2008. Comparison of a modern broiler and layer strain during embryonic development and the hatching process. Br. Poult. Sci. 49:574–582. Haugen, G., M. Hanson, T. Kiserud, S. Crozier, H. Inskip, and K. M. Godfrey. 2005. Fetal liver-sparing cardiovascular adaptations linked to mother’s slimness and diet. Circ. Res. 96:12–14. Hoyt, D. F., and H. Rahn. 1980. Respiration of avian embryos—A comparative analysis. Respir. Physiol. 39:255–264. Hull, D., J. Dobbing, and R. W. Miller. 1978. Definition, epidemiology and identification. Pages 69–84 in Abnormal Fetal Growth: Biological Bases and Consequences. F. Naftolin ed. Dahlem Konferenzen, Berlin, Germany. Jackson, S., and J. Diamond. 1996. Metabolic and digestive responses to artificial selection in chickens. Evolution 50:1638–1650. Konarzewski, M., A. Gavin, R. McDevitt, and I. R. Wallis. 2000. Metabolic and organ mass responses to selection for high growth rates in the domestic chicken (Gallus domesticus). Physiol. Biochem. Zool. 73:237–248. Lang, U., R. S. Baker, J. Khoury, and K. E. Clark. 2000. Effects of chronic reduction in uterine blood flow on fetal and placental growth in the sheep. Am. J. Physiol. Regul. Integr. Comp. Physiol. 279:R53–R59. Lilja, C. 1983. A comparative study of postnatal growth and organ development in some species of birds. Growth 47:317–339. Lilja, C., and U. Olsson. 1987. Changes in embryonic development associated with long-term selection for high growth rate in Japanese quail. Growth 51:301–308. Lindgren, I. 2004. Effects of chronic hypoxia on embryonic and organ growth during the last week of development in broiler chickens. LiU-IFM-Biol-Ex-1128. Linköpings Universitet, Linköping, Sweden. Lindgren, I., and J. Altimiras. 2009. Chronic prenatal hypoxia sensitizes β-adrenoceptors in the fetal heart but causes postnatal desensitization. Am. J. Physiol. Regul. Integr. Comp. Physiol. 297:H258–H264. Liu, Z., B. W. McBride, A. Lirette, and J. R. Chambers. 1995. Characterization of embryonic oxygen consumption of two broiler chicken lines differing in body fat content. Can. J. Anim. Sci. 75:115–119. Luecke, R. H., W. D. Wosilait, and J. F. Young. 1995. Mathematical representation of organ growth in the human embryo/fetus. Int. J. Biomed. Comput. 39:337–347. McCutcheon, I. E., J. Metcalfe, A. B. Metzenberg, and T. Ettinger. 1982. Organ growth in hyperoxic and hypoxic chick embryos. Respir. Physiol. 50:153–163. Metcalfe, J., I. E. McCutcheon, D. L. Francisco, A. B. Metzenberg, and J. E. Welch. 1981. Oxygen availability and growth of the chick embryo. Respir. Physiol. 46:81–88. Miller, S. L., L. R. Green, D. M. Peebles, M. A. Hanson, and C. E. Blanco. 2002. Effects of chronic hypoxia and protein malnutrition on growth in the developing chick. Am. J. Obstet. Gynecol. 186:261–267. Mulder, A. L. M., J. C. Van Golde, F. W. Prinzen, and C. E. Blanco. 1998. Cardiac output distribution in response to hypoxia in the chick embryo in the second half of the incubation time. J. Physiol. 508:281–287. Muramatsu, T., K. Hiramoto, and J. Okumura. 1990. Strain differences in whole-body protein turnover in the chicken embryo. Br. Poult. Sci. 31:91–99. Ohta, Y., T. Yoshida, and N. Tsushima. 2004. Comparison between broilers and layers for growth and protein use by embryos. Poult. Sci. 83:783–787. Ounsted, M. 1988. Small-for-dates babies: Heterogeneity and the concept of “brain sparing.” Paediatr. Perinat. Epidemiol. 2:365– 370. Peebles, D. M., J. C. Dixon, J. S. Thornton, E. B. Cady, A. Priest, S. L. Miller, C. E. Blanco, T. A. L. M. Mulder, R. J. Ordidge, and C. H. Rodeck. 2003. Magnetic resonance proton spectroscopy and diffusion weighted imaging of chick embryo brain in ovo. Brain Res. Dev. Brain Res. 141:101–107.
EMBRYONIC ORGAN GROWTH IN DOMESTICATED CHICKEN BREEDS
Tintu, A., E. Rouwet, S. Verlohren, J. Brinkmann, S. Ahmad, F. Crispi, M. van Bilsen, P. Carmeliet, A. C. Staff, M. Tjwa, I. Cetin, E. Gratacós, E. Hernandez-Andrade, L. Hofstra, M. Jacobs, W. H. Lamers, I. Morano, E. Safak, A. Ahmed, and F. le Noble. 2009. Hypoxia induces dilated cardiomyopathy in the chick embryo: Mechanism, intervention, and long-term consequences. PLoS ONE 4:e5155. Villamor, E., C. G. A. Kessels, K. Ruijtenbeek, R. J. Van Suylen, J. Belik, J. G. R. De Mey, and C. E. Blanco. 2004. Chronic in ovo hypoxia decreases pulmonary arterial contractile reactivity and induces biventricular cardiac enlargement in the chicken embryo. Am. J. Physiol. Regul. Integr. Comp. Physiol. 287:R642–R651. Vleck, C. M., and D. Vleck. 1987. Metabolism and energetics of avian embryos. J. Exp. Zool. Suppl. 1:111–125. Wahlberg, P., Ö. Carlborg, M. Foglio, X. Tordoir, A.-C. Syvänen, M. Lathrop, I. G. Gut, P. B. Siegel, and L. Andersson. 2009. Genetic analysis of an F2 intercross between two chicken lines divergently selected for body-weight. BMC Genomics 10:248. Whitehead, C. C., and H. D. Griffin. 1984. Development of divergent lines of lean and fat broilers using plasma very low density lipoprotein concentration as selection criterion: The first three generations. Br. Poult. Sci. 25:573–582. Wilson, H. R. 1991. Effects of size on hatchability, chick size and posthatching growth. Pages 279–283 in Avian Incubation. S. G. Tullett, ed. Butterworth-Heinemann, London, UK. Wright, D., C.-J. Rubin, Á. Martínez Barrio, K. E. Schütz, S. Kerje, H. Brändström, A. Kindmark, P. Jensen, and L. Andersson. 2010. Pleiotropic and epistatic QTL affecting behaviour and morphology in a large F2 intercross in the chicken. Mol. Ecol. 19:5079–5320. doi:10.1111/j.1365-294X.2010.04882.x
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2015
Rahn, H., C. V. Paganelli, and A. Ar. 1974. The avian egg: Air-cell gas tension, metabolism and incubation time. Respir. Physiol. 22:297–309. Ricklefs, R. E. 1985. Modification of growth and development of muscles of poultry. Poult. Sci. 64:1563–1576. Rodricks, C. L., M. E. Gibbs, G. Jenkin, and S. L. Miller. 2008. The effect of hypoxia at different embryonic ages on impairment of memory ability in chicks. Int. J. Dev. Neurosci. 26:113–118. Romanoff, A. L. 1960. The Avian Embryo: Structural and Functional Development. MacMillan Company, New York, NY. Rubin, C.-J., M. C. Zody, J. Eriksson, J. R. D. Meadows, E. Sherwood, M. T. Webster, L. Jiang, M. Ingman, T. Sharpe, S. Ka, F. Hallböök, F. Besnier, Ö. Carlborg, B. Bed’holm, M. Tixier-Boichard, P. Jensen, P. B. Siegel, K. Lindblad-Toh, and L. Andersson. 2010. Whole-genome resequencing reveals loci under selection during chicken domestication. Nature 464:587–591. Sato, M., T. Tachibana, and M. Furuse. 2006. Heat production and lipid metabolism in broiler and layer chickens during embryonic development. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 143:382–388. Schütz, K. E., B. Forkman, and P. Jensen. 2001. Domestication effects on foraging strategy, social behaviour and different fear responses: A comparison between the red junglefowl (Gallus gallus) and a modern layer strain. Appl. Anim. Behav. Sci. 74:1–14. Stock, M. K., D. L. Francisco, and J. Metcalfe. 1983. Organ growth in chick embryos incubated in 40% or 70% oxygen. Respir. Physiol. 52:1–11. Tazawa, H., S. Nakazawa, A. Okuda, and G. C. Whittow. 1988. Short-term effects of altered shell conductance on oxygen uptake and hematological variables of late chicken embryos. Respir. Physiol. 74:199–209.
135