1st Page

Journal of Animal Science - Animal Nutrition

Prenatal nutritional manipulation by in ovo enrichment influences bone structure, composition, and mechanical properties


This article in JAS

  1. Vol. 91 No. 6, p. 2784-2793
    Received: July 11, 2012
    Accepted: Mar 07, 2013
    Published: November 25, 2014

    1 Corresponding author(s):

  1. R. Yair*†,
  2. R. Shahar and
  3. Z. Uni 1
  1. Department of Animal Science
    The Koret School of Veterinary Medicine, The Robert H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University, Rehovot 76100, Israel


The objective of this study was to examine the effect of embryonic nutritional enrichment on the development and properties of broiler leg bones (tibia and femur) from the prenatal period until maturity. To accomplish the objective, 300 eggs were divided into 2 groups: a noninjected group (control) and a group injected in ovo with a solution containing minerals, vitamins, and carbohydrates (enriched). Tibia and femur from both legs were harvested from chicks on embryonic days 19 (E19) and 21 (E21) and d 3, 7, 14, 28, and 54 posthatch (n = 8). The bones were mechanically tested (stiffness, maximal load, and work to fracture) and scanned in a micro-computed tomography (μCT) scanner to examine the structural properties of the cortical [cortical area, medullary area, cortical thickness, and maximal moment of inertia (Imax)] and trabecular (bone volume percent, trabecular thickness, and trabecular number) areas. To examine bone mineralization, bone mineral density (BMD) of the cortical area was obtained from the μCT scans, and bones were analyzed for the ash and mineral content. The results showed improved mechanical properties of the enriched group between E19 and d 3 and on d 14 (P < 0.05). Differences in cortical morphology were noted between E19 and d 14 as the enriched group had greater medullary area on E19 (femur), reduced medullary area on E21 (both bones), greater femoral cortical area on d 3, and greater Imax of both bones on d 14 (P < 0.05). The major differences in bone trabecular architecture were that the enriched group had greater bone volume percent and trabecular thickness in the tibia on d 7 and the femur on d 28 (P < 0.05). The pattern of mineralization between E19 and d 54 showed improved mineralization in the enriched group on E19 whereas on d 3 and 7, the control group showed a mineralization advantage, and on d 28 and 54, the enriched group showed again greater mineralization (P < 0.05). In summary, this study demonstrated that in ovo enrichment affects multiple bone properties pre- and postnatally and showed that avian embryos are a good model for studying the effect of embryonic nutrition on natal and postnatal development. Most importantly, the enrichment led to improved mechanical properties until d 14 (roughly third of the lifespan of the bird), a big advantage for the young broiler. Additionally, the improved mineralization and trabecular architecture on d 28 and 54 indicate a potential long-term effect of altering embryonic nutrition.


Current broiler strains suffer from numerous bone problems, especially in their leg bones (Angel, 2007; Dibner et al., 2007). The extremely high growth potential and metabolic rate of current broiler strains (Havenstein et al., 2003; Tona et al., 2004) leads to increasing incidence of skeletal problems (Julian, 1998; Angel, 2007; Shim et al., 2012). Additionally, it is hypothesized that because of the increased metabolic rate of the embryos of today, the embryonic nutrient reserves are insufficient and might be depleted in the prenatal period. Because embryonic nutrition has a pronounced effect on progeny performance (Barker et al., 1989; Petry and Hales, 2000; McMillen and Robinson, 2005), such nutritional insufficiencies may induce adaptive responses with long-lasting adverse consequences, which were previously termed “programming” (Lucas, 1991).

During the prenatal period, the yolk (the major mineral source for the embryo) contains limited reserves of P, Zn, Cu, and Mn, and therefore the embryo consumes little of those minerals during that period (Yair and Uni, 2011). Because those minerals are important for broiler bone development (Kidd, 2003; Angel, 2007; Dibner et al., 2007), we hypothesized that the prenatal mineral limitation will probably hinder bone development. On the other hand, embryonic enrichment with those minerals along with vitamins and carbohydrates by the in ovo feeding method was able to increase yolk mineral content and consumption during the prenatal period (Yair and Uni, 2011). It was then hypothesized that similar embryonic enrichment will have a positive effect on broiler bone development in the embryonic period and possibly in later developmental stages. Accordingly, in this study, the effect of a mineral, vitamin, and carbohydrate enrichment by in ovo feeding on the development and the structural and mechanical properties of broiler leg bones (tibia and femur) from the prenatal period until maturity was examined.


The experiment was approved by the Ethics Committee for Animal Experimentation, Faculty of Agricultural, Food and Environmental Sciences, the Hebrew University of Jerusalem.

General Procedure

Three hundred hatching eggs from Cobb 500 hens (38 wk of age), which were fed according to the protocol (Cobb, 2008), were obtained from a commercial hatchery (Brown, Hod Hasharon, Israel). The eggs were divided into 2 groups of similar weight distribution, and the enriched group and the nonenriched group (control) were incubated in a commercial incubator (Petersime, Zulte, Belgium) located in the Faculty of Agricultural, Food and Environmental Sciences (Rehovot, Israel) according to routine incubation procedures (37.8°C and 56% relative humidity). Inside the incubator, the eggs of each group were placed in a 150-egg tray. The trays of both groups were placed adjacent to each other and their position was switched every 24 h to minimize the variability in incubation condition between the groups.


On embryonic d 17 the amniotic fluids of the fertile eggs of the enriched group were injected by the in ovo feeding methodology (Uni and Ferket, 2004) with 0.6 mL solution. The formulation of the solution was designed based on a preliminary experiment, in which the levels and consumption of minerals that are important for bone development during the perinatal period were elevated (Yair and Uni, 2011). The solution included important nutrients for bone development, such as organic Zn, Cu, and Mn (Bioplex Zn, Cu, and Mn; Alltech Inc., Nicholasville, KY), organic Ca [Ca-β-hydroxy-β-methylbutyrate (HMB); Sigma-Aldrich, St. Louis, MO), and vitamin D3 (Phibro Animal Health Corp., Teaneck, NJ). The complete formulation of the enrichment solution is presented in Table 1. The control group was not enriched with any solution but was subjected to the same handling procedures as the in ovo feeding group. This was based on several papers that used noninjected groups as control because previous experiments have shown that water or saline injection did not affect multiple embryo and chick variables (Tako et al., 2004; Uni et al., 2005; Foye et al., 2007; Kornasio et al., 2011; Zhai et al., 2011). Similarly, in a previous preliminary experiment, both nonenriched and saline-enriched control groups were used and found no differences between them in terms of tibial weight, length, and mechanical properties until d 7.

View Full Table | Close Full ViewTable 1.

Formulation of the enrichment solution

Mineral Chemical form Concentration Amount/embryo
Fe Bioplex Fe1 1.6 mg/mL 0.96 mg
Zn Bioplex Zn1 1 mg/mL 0.6 mg
Mn Bioplex Mn1 0.6 mg/mL 0.36 mg
Ca Organic Ca-HMB2 0.6 mg/mL 0.36 mg
Cu Bioplex Cu1 0.03 mg/mL 0.018 mg
P KH2Po4 1.4 mg/mL 0.845 mg
Maltodextrin1 4 mg/mL 2.4 mg
Vitamin A3 1,333 IU/mL 800 IU
Vitamin D3 303 IU/mL 182 IU
Vitamin E3 1.33 IU/mL 0.8 IU
1Bioplex (Alltech inc., Nicholasville, KY); chelated minerals are linked to a mixture of AA and di- and tripeptides.
2HMB = β-hydroxy-β-methylbutyrate. Sigma-Aldrich (St. Louis, MO).
3Phibro Animal Health Corp. (Teaneck, NJ).

Bone Samples

On embryonic d 19 and 21 (E19 and E21), 8 eggs from each group were randomly selected, and the tibia and femur were removed from both legs of each of the embryos. The bones were cleaned of all soft tissues, externally measured (weight and length), wrapped in saline-soaked gauze, and stored at –20°C. After hatching (hatchability was 86.4% for the enriched group and 90.9% for the control group) of the remaining eggs, each hatchling was identified by a neck tag number and moved to a pen, and the floor was covered with soft pine-wood shavings. All chicks were housed in 1 room under commercial conditions and were given ad libitum access to water. The chicks were first fed 36 h after hatch and were given ad libitum access to commercial diet (Table 2) thereafter. On posthatch d 3, 7, 14, 28, and 54, 8 randomly selected chicks from each group were sacrificed by cervical dislocation, and the tibia and femur from both legs were removed. The bones were cleaned of all soft tissues, externally measured (weight and length), wrapped in saline-soaked gauze, and stored at –20°C.

View Full Table | Close Full ViewTable 2.

Composition of starter (36 h posthatch until d 10), grower (d 11 to 21), and finisher (d 22 to 54) diets fed to both groups in the experiment

Item Starter Grower Finisher
Ingredient, %
    Wheat 39.95 45.06 50.03
    Corn 17.80 18.40 10.80
    Dried distillers grains 4.00 9.40
    Soybean meal 27.18 19.44 14.01
    Sunflower meal 4.00 6.00 7.00
    Wheat middling 4.00
    Vegetable fat 2.64 3.02 5.06
    Limestone 1.43 1.25 1.22
    Dicalcium phosphate 1.50 1.30 1.00
    NaCl 0.22 0.20 0.18
    Vitamin–trace mineral premix1 0.90 0.80 0.75
    Hy·D2 0.05 0.05
    L-Lysine sulfate 0.11 0.23 0.32
    Methionine hydroxyl analog 0.11 0.12 0.12
    L-Thr 0.04 0.07
Calculated energy and nutrient composition, as-fed
    ME, kcal/kg 2,955 3,050 3,170
    CP, % 21.0 19.0 18.0
    Ca, % 1.16 1.00 0.90
    Total P, % 0.70 0.64 0.60
    Available P, % 0.50 0.45 0.40
    Fat, % 4.5 5.0 7.2
    Fiber, % 3.3 3.6 3.9
    Ash, % 6.19 5.57 5.24
1Provided per kilogram of grower diet (The starter diet contained 12.5% higher mineral and vitamin amounts and the finisher diet contained 6.25% lower mineral and vitamin amounts): vitamin A, 12,000 IU; vitamin D3, 5,000 IU; vitamin E, 70 IU; vitamin K, 3 mg; vitamin B1, 3 mg; vitamin B2, 7.5 mg; nicotinic acid, 50 mg; pantothenic acid, 14 mg; vitamin B6, 4 mg; folic acid, 1.5 mg; vitamin B12, 0.015 mg; biotin, 0.25 mg; Fe, 30 mg as ferrous sulfate; Zn, 100 mg as zinc oxide; Mn, 100 mg a manganous oxide; Cu, 20 mg as copper sulfate; I, 1.2 mg as potassium iodide; Co, 0.2 mg as cobalt oxide; and Se, 0.3 mg as sodium selenite.
2Hy·D = 25-hydroxycholecalciferol (DSM, Basel, Switzerland).

Mechanical Testing

On the day of test, the right tibia and femur of each embryo and chick were slowly thawed to room temperature. Biomechanical test of tibia and femur collected between E19 to d 7 was performed using a custom-built micromechanical testing device (Fig. S1) as previously described (Yair et al., 2012). Biomechanical test of tibia and femur collected between d 14 and 54 was performed using a commercial materials testing machine (Model 3345; Instron, Norwood, MA) as previously described by Shipov et al. (2010).

After measuring both sizes of bones and instruments, the distance between the supports was adjusted to be the maximal distance possible in the cylindrical part of the diaphysis (Table S1). Loading was conducted at a constant rate (2 or 2.6 mm/min; Table S1) up to fracture point, as identified by a sudden decrease in load, or to a maximum load of 700 N in the larger bones. Force and displacement data were collected at 50 Hz using custom-written software (NI Labview, Austin, TX) for the smaller bones and at 20 Hz by a software (BlueHill; Instron) for the larger bones. The resulting load-displacement curves were used to calculate whole bone stiffness (slope of the linear portion of the load-displacement curve), ultimate load, and work to fracture (WTF; Lanyon et al., 1982).

Structural Analysis

The left tibia and femur of each embryo and chick were scanned using a high-resolution micro-computed tomography (μCT) scanner (SkyScan 1174; Bruker microCT, Kontich, Belgium) as previously described (Yair et al., 2012). Specific scanning variables for each sampling day and each bone can be found in Table S1. For each set of scanning variables, 2 phantoms of known mineral density (0.25 and 0.75 g/cm3) supplied by the manufacturer of the scanner (Bruker microCT) were also scanned using these variables. This allowed calibration of the attenuation levels directly to bone mineral density (BMD) values.

Cortical bone analysis was performed at mid diaphysis of the examined bone by selecting a 100-slice region of interest. The region length differed depending on bone size (Table S1). These cortical properties were measured: BMD, mean cortical thickness, medullary area, cortical cross-sectional area, and maximal moment of inertia (Imax). The Imax was calculated using ImageJ software (National Institutes of Health, Bethesda, MD) with the BoneJ plug-in (Doube et al., 2010). Bone mineral density of the cortical bone was determined starting d 3 because until that developmental stage, the cortical bone is very porous, therefore yielding unreliable BMD values. This limitation is similar to the limitation of determining tissue mineral density of cancellous bone by μCT scanning and is due primarily to partial volume effects (Bouxsein et al., 2010).

Cancellous bone (spongy bone) analysis was performed at the distal epiphyseal area of each examined bone over a 100-slice region of interest; the region length differed depending on bone size (Table S1). The first slice of the selected region of interest was set at the most proximal area where the cross-section was filled with trabeculae and the region of interest consisted of the next 100 slices in the distal direction. These trabecular properties were measured: bone volume (%), mean trabecular thickness (μm), and mean trabecular number (1/μm). Cancellous bone analysis was only performed from d 7 because of the scarcity of cancellous bone struts at the embryonic stage and posthatch, which can lead to unreliable results.

Ash Content and Mineral Analysis

After the left tibia and femur bones were mechanically tested, a small section from the mid diaphysis of each bone (about 10 to 20% of the length of the bone) was removed and taken for bone ash quantification using a previously described method (Yair et al., 2012). Between E19 and d 7, a sample (100 to 150 mg) from each tibia ash was collected (after the bone ash quantification procedure) to examine the bone mineral content. Each sample was digested with a mixture of 2 mL 30% H2O2 and 4 mL 70% HNO3 in a 50-mL plastic tube for 6 h in a 95°C bath. The digested samples were then analyzed for their mineral content using an inductively coupled plasma atomic emission spectroscopy instrument (Spectro Arcos, Kleve, Germany). The results of the Ca, Mn, P, and Zn content were calculated as percent of the dry bone weight.

Statistical Analysis

Values are presented as means ± SE. Data were analyzed by 1-way ANOVA (JMP 7 software; SAS Inst. Inc., Cary, NC), and separation between groups at each sampling time was performed using Student’s t test. Differences were considered statistically significant at P < 0.05.


Length, Weight, and Relative Weight

Tibia and femur length, weight, and relative weight are presented in Table 3. The major difference between the groups was observed on d 14. The tibia and femur of the enriched group were longer, and their weight and relative weight were greater in comparison with the control (P < 0.01) whereas no consistent differences were detected until this age or later.

View Full Table | Close Full ViewTable 3.

Tibia and femur length, weight, and relative weight of the control and enriched groups in the prenatal and growing periods

Length, cm
Weight, g
Relative weight, mg/g BW
Sampling day1 Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE
E19 Tibia 2.94 0.03 2.94 0.03 0.23 0.10 0.25 0.10 6.97 0.28 7.30 0.28
Femur 2.08 0.04 2.15 0.03 0.17 0.01 0.18 0.01 4.82 0.17 5.10 0.16
E21 Tibia 3.17 0.03 3.17 0.03 0.29 0.01 0.30 0.01 6.91 0.17 7.07 0.17
Femur 2.38 0.03 2.31 0.03 0.20 0.01 0.21 0.01 4.90 0.09 4.83 0.10
d 3 Tibia 3.54 0.02 3.54 0.02 0.50 0.01 0.48 0.01 7.65 0.21 7.08 0.21
Femur 2.58 0.03 2.54 0.03 0.32 0.01 0.30 0.01 4.87*2 0.10 4.47 0.11
d 7 Tibia 4.30 0.04 4.44* 0.04 1.02 0.03 1.04 0.03 7.08 0.19 7.17 0.19
Femur 3.23 0.05 3.21 0.05 0.71 0.02 0.73 0.02 4.88 0.14 5.00 0.14
d 14 Tibia 5.55 0.07 5.99* 0.06 2.73 0.11 3.29* 0.09 7.03 0.20 8.02* 0.18
Femur 4.07 0.07 4.25 0.06 2.05 0.08 2.35* 0.07 5.17 0.10 5.74* 0.10
d 28 Tibia 8.44 0.10 8.47 0.11 9.37 0.54 9.37 0.57 7.22 0.20 7.47 0.21
Femur 5.83 0.07 5.84 0.08 6.66 0.40 6.58 0.43 5.14 0.16 5.24 0.17
d 54 Tibia 11.88 0.14 12.00 0.18 24.36 1.40 23.52 1.76 6.87 0.17 6.63 0.21
Femur 7.99 0.12 7.94 0.12 16.81 0.89 14.64 0.89 4.80 0.19 4.45 0.19
1E19 and E21 = embryonic days.
2Means with asterisk mark a significant difference between the groups (P < 0.05); n = 8 per group on each sampling day.

Mechanical Properties

The mechanical properties of both bones are presented in Table 4. Generally, the tibia and femur of the enriched group showed superior mechanical properties (stiffness, maximal load, and WTF) to those of the control between E19 and d 3 (P < 0.05). By d 7, both groups did not differ in their mechanical properties; however, on d 14, the tibia again showed increased stiffness and maximal load (P < 0.05).

View Full Table | Close Full ViewTable 4.

Bone mechanical properties in the prenatal and growing periods. stiffness, maximal load and work to fracture (WTF) of the tibia and femur1,2

Stiffness, N/mm
Maximal load, N
WTF, N x mm
Sampling day Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE
E19 Tibia 32.26 2.15 33.19 2.15 12.11 0.81 12.94 0.81 21.68 1.81 21.44 1.57
Femur 44.42 2.57 52.20* 2.38 14.93 1.13 18.78* 1.04 12.67 1.67 19.40* 1.55
E21 Tibia 44.89 2.80 58.55* 2.80 14.28 1.16 18.66* 1.16 37.11 2.76 45.99* 2.76
Femur 35.79 2.10 44.16* 2.24 13.62 0.80 17.48* 0.86 24.98 2.55 23.99 2.73
d 3 Tibia 58.37 2.12 69.26* 2.12 22.87 0.58 22.09 0.58 54.77 1.85 53.05 1.85
Femur 80.44 2.85 89.10* 2.85 24.00 1.15 23.11 1.15 47.04 3.65 43.64 3.65
d 7 Tibia 156.5 25.7 146.9 29.5 73.7 12.6 70.6 9.9 103.5 7.2 99.7 6.7
Femur 126.4 10.3 124.7 10.3 52.1 4.5 57.5 4.5 65.6 3.1 56.5 3.1
d 14 Tibia 246.7 14.1 294.3* 14.1 139.3 7.7 185.8* 7.7 154.1 15.4 149.7 15.4
Femur 219.0 12.8 229.7 12.0 160.3 8.7 177.2 8.1 173.3 21.7 187.1 20.3
d 28 Tibia 361.2 38.7 380.6 44.7 270.1 19.3 237.8 22.3 445.8 103.0 463.5 119.0
Femur 266.1 25.1 311.4 29.0 197.8 17.9 210.1 20.7 546.3 61.7 679.5 71.3
d 54 Tibia 309.2 28.5 344.3 28.5 455.1 46.0 429.8 46.0 710.8 95.9 642.3 95.9
Femur 430.5 29.9 409.8 34.5 312.5 20.7 285.8 24.0 527.1 71.5 485.2 82.6
1E19 and E21 = embryonic days.
2Means with asterisk mark a significant difference between the groups (P < 0.05); n = 8 per group on each sampling day.

Cortical Bone Structural Properties

Cortical bone structural properties are presented in Table 5. The differences in cortical bone structural properties between the 2 groups occurred only during the perinatal period (E19 to d 3) and on d 14. On E19, the enriched group femur had larger (P < 0.05) medullary area whereas on E21, it was the control group that had larger tibial and femoral medullary area (P < 0.05). The enriched group exhibited larger femoral cortical area at d 3 and greater femoral and tibial Imax at d 14 than the control (P < 0.05).

View Full Table | Close Full ViewTable 5.

Cortical bone structural properties in the prenatal and growing periods. Cortical area, medullary area, crossectional thickness, and maximal moment of inertia (Imax) of the tibia and femur

Cortical area, mm2
Medullary area, mm2
Crossectional thickness, mm
Imax, mm4
Sampling day1 Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE
E19 Tibia 1.95 0.07 1.82 0.08 0.57 0.06 0.57 0.06 0.25 0.01 0.28 0.02 0.49 0.04 0.46 0.04
Femur 1.89 0.10 1.81 0.10 0.66 0.05 0.87*2 0.06 0.25 0.01 0.25 0.01 0.46 0.04 0.53 0.04
E21 Tibia 1.45 0.10 1.51 0.11 1.24* 0.08 0.98 0.08 0.27 0.02 0.31 0.02 0.49 0.04 0.44 0.05
Femur 1.80 0.10 1.98 0.09 1.11* 0.07 0.88 0.06 0.32 0.01 0.34 0.01 0.53 0.04 0.57 0.04
d 3 Tibia 1.58 0.04 1.60 0.04 1.73 0.10 1.80 0.10 0.26 0.01 0.26 0.01 0.70 0.04 0.73 0.04
Femur 1.67 0.04 1.81* 0.04 1.63 0.27 1.50 0.27 0.28 0.01 0.29 0.01 0.67 0.04 0.72 0.04
d 7 Tibia 3.30 0.15 3.58 0.14 2.26 0.08 2.22 0.08 0.43 0.02 0.44 0.02 2.38 0.18 2.44 0.18
Femur 4.55 0.14 4.60 0.14 2.24 0.08 2.16 0.08 0.53 0.01 0.53 0.01 3.97 0.18 3.89 0.18
d 14 Tibia 11.13 0.63 12.47 0.59 2.98 0.15 3.37 0.14 0.94 0.03 1.02 0.03 16.3 1.66 20.45* 1.55
Femur 13.31 0.59 14.93 0.55 4.32 0.33 4.80 0.31 0.88 0.01 0.90 0.01 23.3 2.05 29.97* 1.92
d 28 Tibia 22.32 1.27 23.15 1.36 11.11 1.26 11.67 1.35 1.21 0.07 1.21 0.07 93.7 11.2 102.8 12.0
Femur 27.87 0.01 28.97 1.16 19.72 1.48 19.02 1.71 1.07 0.06 0.98 0.07 157.9 13.7 154.1 14.6
d 54 Tibia 37.99 3.29 38.87 3.80 38.32 2.88 33.68 3.32 1.00 0.09 1.20 0.10 433.4 61.6 367.0 72.8
Femur 39.02 2.97 36.76 2.97 56.57 4.97 60.64 4.97 0.71 0.06 0.67 0.06 485.3 60.9 471.2 60.9
1E19 and E21 = embryonic days.
2Means with asterisk mark a significant difference between the groups (P < 0.05); n = 8 per group on each sampling day.

Trabecular Bone Structural Properties

The trabecular bone structural properties are presented in Fig. 1. On d 7, the tibial bone volume (%) and trabecular thickness were greater (P < 0.05) in the enriched group in comparison with the control. On d 14, only femoral trabecular thickness was greater in the enriched group than in the control group (P < 0.05). In addition, femoral trabecular number was greater in the control group on d 7 and 14 (P < 0.05). On d 28, both the tibia and femur exhibited a similar pattern of increased (P < 0.05) bone volume and trabecular number in the enriched group.

Figure 1.
Figure 1.

Trabecular bone structural properties between d 7 and 54. Bone volume percent of the tibia (A) and femur (B), trabecular thickness of the tibia (C) and femur (D), and trabecular number of the tibia (E) and femur (F). Within a day, the difference between 2 means is marked with an asterisk (P < 0.05).


Bone Mineralization

Tibia and femur ash content and BMD are presented in Fig. 2. On E19, the tibial ash content of the enriched group embryos was greater than the control (P < 0.001). However, on d 3 and 7 posthatch, the control group showed greater BMD and ash content than the enriched group in both tibia and femur (P < 0.05). On d 28 and 54, the tibia of the enriched group again seems to be more mineralized than the control as can be observed by the elevated BMD on d 28 and ash content on d 54 (P < 0.05).

Figure 2.
Figure 2.

Mineralization of the tibia and femur in the prenatal and growing periods. Ash content of the tibia (A) and femur (B) and bone mineral density (BMD) of the tibia (C) and femur (D). Within a day, the difference between means is marked with an asterisk (P < 0.05).


The mineral content of the tibia between E19 and d 7 is presented in Table 6. The Ca and P content of the tibia showed a similar trend. On E19, the enriched group had about 50% greater Ca and P content than the control (P < 0.01). On E21 and d 3, no differences between the groups were observed whereas on d 7, the control group had 17 to 20% greater Ca and P content than the enriched group (P < 0.05). The Mn content of the control group was greater (P < 0.001) throughout the examination period (E19 to d 7) whereas no differences (P > 0.08) between the groups were observed in the Fe content on any of the sampling days.

View Full Table | Close Full ViewTable 6.

Analysis of some of the different minerals in the tibia expressed in percent of the dry weight of the bone

Ca, %
Mn, %
P, %
Zn, %
Sampling day1 Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE
E19 17.46 1.23 26.83*2 2.39 0.002 0.001 0.008* 0.001 9.32 0.60 14.00* 1.23 0.01 0.002 0.02 0.004
E21 42.18 0.97 43.48 1.77 0.002 0.000 0.012* 0.001 22.14 0.53 22.66 0.92 0.06 0.004 0.06 0.003
d 3 41.64 2.65 35.09 3.83 0.002 0.000 0.006* 0.001 21.04 1.41 17.77 1.79 0.05 0.005 0.05 0.006
d 7 47.19* 2.55 39.26 1.61 0.002 0.000 0.003* 0.000 24.24* 1.32 20.57 0.87 0.05 0.003 0.06 0.010
s1E19 and E21 = embryonic days.
2Means with asterisk differ from the control (P < 0.05); n = 8 per group on each sampling day.


This study demonstrated that changing embryonic nutrition by in ovo feeding methodology affects the structural and mechanical properties of the tibia and femur and changes the dynamics of skeletal development. The effect of prenatal nutrient supplementation was detectable during the perinatal period, up to d 14. The changes between the groups in bone mineralization and trabecular structure on d 28 and d 54 indicated the potential for a long-term effect although no clear lasting (“programming”) effect can be observed in late broiler growth.

Inadequate fetal and early postnatal nutrition were previously shown in humans to be responsible for increased risk for cardiovascular disease, diabetes, and metabolic syndrome in adult life (Barker et al., 1989; Langley and Jackson, 1994; Wildman et al., 1995; McMillen and Robinson, 2005). The development of the skeletal system and its subsequent properties are also affected by embryonic nutrition. Neonatal essential fatty acid deficiency reduced trabecular volumetric BMD in the femur but increased femoral cortical area, thickness, and BMD in adult rats (Korotkova et al., 2005). In mice, prenatal nutritional restriction by bilateral uterine vessel ligation caused reduced femoral cortical thickness as well as periosteal and endosteal circumference 6 mo postnatally even though the prenatally restricted mice were allowed normal lactation (Romano et al., 2009).

Previous work by our group described the changes in the mechanical, structural, and compositional properties of the tibia and femur between embryonic d 14 (E14) and d 7. The results showed that during the perinatal period, bone development “slows down,” as reflected by the fact that most mechanical and geometric properties remain unchanged (Yair et al., 2012). This slow-down trend is different than the trend that was reported for other important tissues such as the digestive tract that can develop rapidly during the last days of incubation and the first days postnatally (Uni et al., 2003). One of the hypotheses that might explain the “slow-down” phenomenon in the pre- to posthatch period in chickens is a nutrient depletion for the developing embryo in the prenatal period; possibly, the shortage of P, Zn, Cu, and Mn in the yolk (the major mineral storage) during the prenatal period (Yair and Uni, 2011) can lead to impaired bone development. There are few examples of the effect of mineral shortage on bone in chickens; Cu deficiency leads to decreased collagen crosslinking and bone mineralization (Opsahl et al., 1982) and Zn deficiency decreases bone collagen turnover and is accompanied by leg deformities (Starcher et al., 1980). Increasing egg mineral content might reduce the incidence of those mineral shortages. However, multiple studies have shown that increasing the concentration of most minerals in the diet of the hen has little or no effect on their concentrations in the egg (Naber, 1979; Angel, 2007). It is, therefore, important to investigate other methods to elevate egg mineral content, such as in ovo feeding.

The in ovo feeding method (Uni and Ferket, 2004) is a suitable tool for manipulating embryonic nutrition because it enables the administration of exogenous nutrients into the amniotic fluid of the late-term avian embryo, thus enriching its nutrient reserves. Previous research with the in ovo feeding method showed that it enhances muscle growth and small intestine development in the pre- and postnatal periods (Tako et al., 2004; Uni and Ferket, 2004; Foye et al., 2007; Cheled-Shoval et al., 2011; Kornasio et al., 2011). The in ovo solution used in this research was previously shown to increase yolk mineral content and consumption during the prenatal period (Yair and Uni, 2011); therefore, such enrichment is hypothesized to have a considerable effect on broilers skeletal development and properties.

In this study, in ovo enrichment resulted in numerous changes in the structure of long bones (tibia and femur) in both the cortical and medullary zones during the pre- to posthatch period (E19 to d 3). Generally, changes in bone structure occur through the mechanisms of bone modeling (Frost, 1990a; Pearson and Lieberman, 2004) and remodeling (Frost, 1990b; Pearson and Lieberman, 2004), which involve the synchronized activity of the various cells of the bone tissue, osteocytes, osteoblasts, and osteoclasts (Lanyon, 1973; Uhthoff and Jaworski, 1978; Reich et al., 2008). Osteocytes, residing within the bone tissue, serve as strain transducers (Marotti, 1996; Pearson and Lieberman, 2004; Bonewald, 2011), osteoblasts are bone forming cells, and osteoclasts are bone resorbing cells.

Quantifying the changes in bone structure, morphology, and composition allows estimation of the orchestrated activity of osteoblasts and osteoclasts in time. Osteoclast activity occurs mainly on the endosteal surface by endosteal resorption, which increases the medullary area. Therefore, endosteal resorption rate between 2 time points can be estimated using the following formula: EBR = MA2 – MA1, in which EBR is the endosteal resorption rate, MA1 is the medullary area on the previous timepoint, and MA2 is the medullary area on the current timepoint. The EBR is a good proxy for osteoclastic activity. On the other hand osteoblast activity occurs primarily on the periosteum by periosteal bone formation (PBF). The changes in cortical area are created by formation in the periosteum and resorption in the endosteal surface (EBR). Hence, an estimate of PBF can be obtained by using the cortical area values and the corresponding EBR value using the following formula: PBF = (CA2 – CA1) + EBR, in which CA1 is the cortical area on the previous timepoint and CA2 is the cortical area on the current timepoint. For example, in the current study we showed that between E19 and E21, femoral EBR of the enriched group was minor (0.01 mm2, which is 0.55% of the femoral cortical area on E19) whereas in the control group, a much greater resorption rate occurred (0.45 mm2, which is 24.34% of the cortical area on E19). During the same period, the PBF was 0.18 mm2 (which is 9.94% of the cortical area on E19) in the enriched group and 0.36 mm2 in the control group (19.04% of the cortical area on E19). This example shows that it is reasonable to assume that in ovo enrichment reduced osteoclastic and osteoblastic activities perinatally and demonstrates the changes in bone development dynamics in response to the in ovo enrichment.

The differences in BMD (and ash content) between E19 and d 54 may result from the differences between the 2 groups in bone formation and resorption rates occurring in the endosteal vs. periosteal areas. For example, between E21 and d 3, the control group showed increased PBF and reduced EBR (compared with the enriched group). Because the endosteal area is older and, therefore, more highly mineralized and the periosteal bone is new and more poorly mineralized, the net effect of the modeling process between E21 and d 3 is a decrease in the enriched group mean BMD. However, it should be noted that not all changes in BMD can be similarly explained. For instance, the increase in BMD of the tibial cortex in the enriched group on d 28 occurs despite the fact that the resorption and formation rates of both groups are similar. This increase must therefore be due to other mechanisms that control bone mineralization, which are complex and involve coordination and feedback loops between numerous control levels. Therefore, BMD changes may be due to systemic hormones [such as fibroblast growth factor 23, 1,25(OH)2D3, and parathyroid hormone], bone cells (osteoblasts, osteoclasts, and osteocytes), and bone cell products (such as inorganic pyrophosphate), which act in a paracrine/autocrine manner (Quarles, 2008; Sapir-Koren and Livshits, 2011). Additionally, the differences in BMD and ash content, as a result of the enrichment, can be related to the greater availability of minerals to the embryo (Yair and Uni, 2011), which led to greater deposition of minerals into the bone. The duration of this effect depended on the relative concentration of each mineral in the enrichment solution. In the solution, Mn was presented in great amounts relative to its amount in the yolk, which previously resulted in a 25- to 80-fold increase in the Mn content (Yair and Uni, 2011) and led to a greater Mn content of the tibia at least until d 7. On the other hand, P and Ca were presented in the solution in decreased relative amounts, which resulted in a temporal elevation in their tibial concentration and in bone ash content (at E19). However, this effect rapidly disappeared, and by E21, no difference in Ca, P, or ash content was observed.

In ovo enrichment showed a positive effect on the trabecular architecture until d 28, as expressed by increased bone volume percent of the tibia on d 7 and in the femur on d 28, and increased trabecular thickness in the tibia on d 7 and in the femur on d 14. These improved trabecular properties would further improve the mechanical properties of the whole bone (McCalden et al., 1997; Ito et al., 2002; Mittra et al., 2005; Barak et al., 2010).

Although bone has many important functions, such as homeostasis and endocrine control, its main function is mechanical (Frost, 2000); therefore, the marked effect of in ovo enrichment on the mechanical properties of the tibia and femur until the age of 2 wk posthatch is of great importance. Although there was no mechanical difference between the groups on d 28 and 54, greater tibial BMD and ash content (which usually correlate with increasing mechanical properties) was observed on these days in the enriched group. This finding indicates a mechanical effect of in ovo enrichment on the mature bone as well.

Because the mechanical function of bones is determined by both their structure and composition (Weiner and Wagner, 1998; Sharir et al., 2008), the mechanical differences between the enriched group and the control should reflect the combined changes in the structural and compositional properties. In this study, the increased stiffness of the tibia of the enriched group on d 14 is likely to be the result of their greater Imax.

At hatch, the enriched group showed better mechanical properties, but as time progressed, this advantage became less apparent and by d 7 it disappeared, possibly because of the greater mineral content in the control group on d 3 and 7, which can be observed by their greater ash percent or P and Ca contents. This increase in mineral content might be a compensatory response to the inferior mechanical properties of the hatchlings of the control group (Boersma and Wit, 1997; Gibson et al., 2000; Metcalfe and Monaghan, 2001; Cooke, 2010). Although compensatory responses could have accelerated mineralization rate in the control group between hatch and d 7, mechanical changes between the groups were observed only a week later as the enriched group showed superior mechanical properties on d 14. Compensatory growth is associated with a variety of negative effects in adult life (Forsen et al., 1999; Victora and Barros, 2001; Hales and Ozanne, 2003; Cettour-Rose et al., 2005), which could explain the inferior mechanical and structural properties of the control group on d 14.

Although in ovo enrichment affected bone development dynamics in general, it had a slightly different effect on tibial properties than on femoral properties. In a previous work, the tibia and femur were shown to have different trends of changes in their mechanical, structural, and compositional properties between E14 and d 7 (Yair et al., 2012). It might be suggested that the differences in the effect of in ovo feeding on each bone are related to the specific development dynamic of each bone.

Using the chicken as a model animal for prenatal nutrition has inherent advantages. The entire embryonic nutrition is finite and enclosed within the egg. Furthermore, it is easily accessible and can be maneuvered, and the nutritional and functional effects in the egg can be measured. An additional advantage to the use of meat-producing strains of chicken (broiler) is their extremely high growth rate and fast maturity (Havenstein et al., 1994, 2003), thus amplifying the influence of any nutritional limitation on the development of critical organs and growth rates (Gous, 2010; Uni et al., 2012). Therefore the current study is a good model for the potential long-term effects of embryonic nutrition on the adult skeleton.

In conclusion, this study described the effect of in ovo enrichment with minerals, vitamins, and carbohydrates on the structural, mechanical, and compositional properties of long bones from the embryonic period until maturity. Generally, there was an early positive effect on the bones of hatchlings, which slowly faded by d 7. On d 14, again the enriched group bones showed improved properties. No clear long-term programming effect was observed; however, the increased mineralization levels and improved trabecular bone morphology during the last days of growth (d 28 to 54) indicate a possibility of long-term effects of in ovo enrichment on the dynamics of bone development. This work demonstrates the potential influence of prenatal nutrition on the characteristics of organisms at least until 2 wk posthatch, which account for 33% of the lifespan of broiler chickens.




Be the first to comment.

Please log in to post a comment.
*Society members, certified professionals, and authors are permitted to comment.