Fatty acid composition in the yolk and yolk sac, embryo development, IGF-I and VEGF-A gene expressions and hatching results in eggs obtained from local and commercial breeders

TL;DR

This study compares local and commercial chicken breeds to understand differences in egg development, gene expression, and hatching outcomes.

Contribution

The study identifies genotype-specific differences in fatty acid composition, gene expression (IGF-I and VEGF-A), and hatching performance in local and commercial chicken breeds.

Findings

Local breeds like Denizli showed distinct gene expression patterns compared to commercial hybrids.

Broiler embryos had higher embryo weights, while local breeds retained more residual yolk sac weight.

Hybrid breeds exhibited higher fertility and hatchability compared to local breeds.

Abstract

This study investigated fatty acid composition in yolk and yolk sac, embryo development, IGF-I and VEGF-A gene expressions, and hatching results in eggs from local (Denizli and Gerze) and commercial (broiler (ROSS-308 and layer (Hy-Line W-80) breeders. A total of 1032 eggs per genotype were incubated. Eggs were weighed, numbered, and sampled at the beginning and on days 10, 13, 16, and 19 for yolk, embryo, fatty acid, and gene expression analyses. At hatch, all chicks were weighed, and eight per genotype were sampled for fatty acid composition in the residual yolk sac and gene expression in breast muscle. Significant genotypic differences were found in egg and chick weights, with layer hybrids having the greatest (59.68 and 41.13 g, respectively) and Gerze the lowest values (52.22 and 36.40 g, respectively). Relative embryo weights increased during incubation, with broilers showing the greatest values and local breeds having higher residual yolk sac weights. While chick weights differed, relative chick weights did not vary significantly. Fertility, embryonic mortality, and hatchability were affected by genotype; hybrids had higher fertility and hatchability, whereas Gerze showed the greatest embryonic mortality and lowest hatchability (11.23 and 67.76%, respectively). Fatty acid profiles showed palmitic (approximately 24–25%) and stearic acids (approximately 9–10%) as dominant saturated fatty acids, oleic acid (approximately 40–45%) as the main MUFA, and linoleic acid (approximately 15–20%) as the major PUFA. In terms of the IGF-1 gene, embryos of the Denizli breed were downregulated by 0.38 fold compared to laying hybrids while in chicks this change was found to be an 8.66 fold upregulation. It was determined that these values were downregulated by 0.08 fold in embryos and 0. 71 fold in chicks in terms of the VGEFA gene. PPI analysis of 32 IGF1/VEGFA interactors revealed a core module centered on FGFs, PGF, neuropilins, and VEGF receptors. Enrichment pointed to system development and angiogenesis, with MAPK, receptor tyrosinekinase, IGF1RIRS, and calcium signaling as key pathways. Genotypespecific differences in this signaling hub underscore the value of local breeds like Denizli for optimizing embryonic vascular and muscle development in sustainable poultry production.

Figures

Gene expression analyses (A and B: Layer hybrid was control; C and D: broiler hybrid was control). *: *p* < 0.05, : *p* < 0.01, *: *p* < 0.001

PPI analysis (A: global network proteins interacting with IGF‑1 and VEGF-A; B: core module identified by MCODE (score 21.273))

Functional enrichment analysis results. (A: GO enrichment across BP, MF and CC; B: pathway enrichment based on KEGG and Reactome; FDR: false discovery rate)

Full text

Introduction

Poultry farming is currently one of the fastest growing sectors within animal production. With the increase in population and urbanization, the demand for meat and eggs continues to rise, leading to a significant expansion in production (Mottet and Tempio 2017). The efficiency of poultry production depends on genotype, breeder age as well as incubation and rearing conditions (Onbaşılar et al. 2018a; Erisir et al. 2009; Nabati et al. 2025; Avcılar et al. 2018, Onbaşılar et al. 2020, Onbaşılar et al. 2025b; Tekin Demir et al. 2025). Today, commercial production is largely carried out using broiler and layer hybrid lines. However, local chicken genotypes are preserved by countries as valuable genetic resources and are considered strategically important for sustainable animal production. In Turkey, two native chicken breeds are preserved as genetic resources: Denizli and Gerze.

The embryonic period is the most critical stage in determining a chick’s growth performance, viability, and resistance to diseases (Onbaşılar et al. 2025a; Varol Avcılar et al 2024). This stage is particularly crucial in fast-growing broilers, as it forms a substantial part of their life span (Givisiez et al. 2020). During incubation, the embryo fulfills all its nutritional requirements from the albumen, yolk, and shell. Approximately 90% of the energy requirement is met by lipids from the yolk sac, while the remaining 10% is supplied by proteins and carbohydrates. From the beginning to the end of embryonic development, linoleic acid (C18:2n-6) and alpha-linolenic acid (C18:3n-3) are present alongside other polyunsaturated fatty acids to meet the tissue requirements. High levels of arachidonic acid (C20:4n-6) and docosahexaenoic acid (C22:6n-3) are found in egg yolk phospholipids (Deniz et al. 2025).

Skeletal muscle myogenesis is a process that begins in the early stages of embryonic development and continues throughout life. Muscle cells originate from stem cells known as myoblasts, which first proliferate, then differentiate into myotubes, and eventually fuse to form multinucleated muscle fibers (Chal and Pourquié 2017; Sobolewska et al. 2011). During embryonic myogenesis, transcription factors such as myogenic determination factor, myogenic factor 4, and myogenic factor 5 play significant roles. Myogenin gene is involved in myoblast differentiation and regulates the growth of adult myofibrils (Te Pas and Soumillion 2001). Insulin-like growth factor I (IGF-I) plays a crucial role in stimulating myoblast proliferation and muscle formation during embryogenesis (Yu et al. 2015), whereas vascular endothelial growth factor A (VEGF-A) is a key regulator of angiogenesis and is responsible for the vascular development of muscle tissue (Ferrara and Gerber 2002).

Despite the increasing interest in the genetic and physiological diversity of poultry breeds, there remains a significant gap in comparative data on embryonic development, hatchability parameters, fatty acid profiles of residual yolk sac, and growth-related gene expression among local breed and commercial chicken genotypes. In particular, limited information is available regarding the molecular and biochemical characteristics of local Turkish breeds such as Denizli and Gerze. In this context, the present study aims to comprehensively evaluate embryonic development, hatchability traits, fatty acid composition of residual yolk sac, and the expression levels of IGF-I and VEGF-A genes in embryos and newly hatched chicks obtained from Denizli and Gerze local breeds, as well as from commercial broiler and layer hybrids.

Materials and methods

Hatching eggs and incubation

Hatching eggs from Denizli and Gerze local breeds, as well as from commercial broiler (ROSS-308) and layer breeders (Hy-Line W-80), were used in this study. Denizli and Gerze breeds, as well as laying hybrids, produce white-shelled eggs, whereas broiler hybrids produce brown-shelled eggs. Since Denizli and Gerze are generally regarded as laying-type breeds, the laying hybrid was selected for its production of white-shelled eggs. For the broiler hybrid, ROSS-308, which is the most commonly used hybrid in the region, was chosen. A total of 258 hatching eggs from each genotype were included. All eggs were collected from breeder flocks of similar age (30–32 weeks). Eggs were collected on the same day from each flock and no pre-incubation storage was applied. All eggs were numbered and weighed. Before incubation, eight eggs from each genotype were opened at the blunt end, and the yolks were separated, weighed, and sampled for fatty acid analysis. The remaining eggs were randomly distributed across the trays of each level of the setter (Çimuka, Turkey), with equal numbers per genotype to eliminate positional effects. The incubation conditions were maintained at 37.6 °C and 60% humidity in the setter to end of day 18 and 36.5 °C and 70% humidity in the hatcher to end of day 21. On days 10, 13, 16, and 19 of incubation, eight fertile eggs per genotype were sampled. Eggs were weighed and opened; embryos were euthanized via cervical dislocation. Residual yolk sac and embryo weights were measured and calculated as a percentage of the total egg weight. Residual yolk sac samples were collected for fatty acid analysis, and breast muscle tissue was sampled on day 19 for IGF-I and VEGF-A gene expression analysis. On the day of hatching, all chicks were weighed. Eight chicks per genotype were euthanized by cervical dislocation, and residual yolk sacs were weighed and expressed relative to egg weight. Residual yolk sac was collected for fatty acid profiling, and breast muscle was also sampled for IGF-I and VEGF-A gene expression analysis. Fertility rate, hatchability of fertile and set eggs, and relative chick weights were calculated after incubation (Coulibaly et al. 2024; Onbaşılar et al. 2014).

Fatty acid analyses

Yolk fat was extracted according to the modified Soxhlet method (AOAC 2000). For this purpose, approximately 5 g of egg yolks were taken into 50 ml falcon tubes and 20 ml diethyl ether was added and mixed in an orbital mixer 2 h. Then all samples were centrifuged and the upper phase was taken into balloon jugs without mixing with the yolk. The residual ether in these samples was removed in a rotary evaporator was extracted from the samples exposed to atmospheric nitrogen for one hour. Then, approximately 50 mg of extracted fat was taken into reaction tubes and saponified with 2 ml of 0.5 N methanolic NaOH for 2 min/90 °C. Then 5 ml of 27% methanolic BorTri Fluoride was added and heated for another 5 min at 90 °C. After cooling the reaction tubes, 2 ml of n-Heptane was added and heated again for 1 min. Then 3 ml of saturated NaCl was added and the tubes were inverted and centrifuged to separate the organic phase. Fatty acid methyl esters were transferred from the top layer to the vial in n-Heptane phase and stored at − 20 °C until GC- MS analysis (Yakan et al. 2016). Shimadzu GCMS-QP2020NX model GC-MS equipped with Restek Capillary Column (100 m length, 0.25 mm i.d. × 0.20 mm film) was used for fatty acid separation. The injector temperature was set to 240 °C, the MS detector interface temperature was set to 270 °C and the ion source was set to 200 °C. The split ratio was 1:100 and the total injection volume was 1 µl. The oven temperature was initially programmed to 100 °C for 4 min and ramped to 240 °C at a ramp rate of 5 °C/min. The peaks obtained as a result of the chromatogram were identified in the MS library and confirmed with FAME Mix 37 as an internal control to show the detected fatty acids. Helium was used as carrier gas.

Gene expression analyses

Total RNA was extracted from embryo and chick pectoral muscle using a modified Trizol protocol (Yakan et al. 2024). RNA purity and concentration were determined spectrophotometrically (Epoch 2, Agilent), and integrity was confirmed by visualization of 18S and 28S rRNA bands on a 1% agarose gel. Samples failing to meet purity, concentration, or integrity criteria were re‑extracted.

DNase (Cat. No: EN0521, DNase I, Thermo Fisher Scientific, USA) treatment was performed to prevent possible DNA contamination in the samples. For this purpose, 1 µl DNase I and 1 µl 10x Reaction Buffer with MgCl_2_ were added to the RNA samples fixed at a concentration of 1000 ng and incubated at 37 °C for 30 min. DNase enzyme inhibition as a result of DNase treatment was performed by adding 1 µl 50 mM EDTA and incubating at 65 °C for 10 min. Synthesis of cDNA from the samples was performed using a commercial kit (Cat. No: 4,368,814, High-Capacity cDNA Reverse Transcription Kit, Thermo Fisher Scientific, USA). After incubation at 25 °C for 10 min, 37 °C for 120 min and 85 °C for 5 min, cDNA synthesis was performed using a Thermal Cycler (SimpliAmp™, Thermo Fisher Scientific, USA). After cDNA synthesis, the samples were filled to 200 µl with nuclease-free water and stored at − 20 °C until analysis (Yakan et al. 2023).

Amplifications of IGF-1 and VEGF-A target genes and ACTB housekeeping gene were performed using RT-qPCR (Rotorgene Q, Qiagen, USA). For this purpose, a commercial kit containing SYBR Green (Cat. No: 4,367,659, Power SYBR™ Green PCR Master Mix, Thermo Fisher Scientific, USA) was used. The RT-qPCR reaction was arranged as 40 cycles of denaturation at 95 °C for 10 min, followed by 15 seconds at 95 °C, 60 seconds at 60 °C, and 30 seconds at 72 °C. All samples were run in duplicate. The primer sequences for ACTB were adapted from Ruiz-Castañeda et al. (2016), for IGF-1 from Wang et al. (2021), and for VEGF-A from Reheman et al. (2024).

Bioinformatic analysis for protein- protein interaction

Protein–protein interaction networks for IGF1 and VEGF-A were constructed using the STRING database (version 12.0, https://string-db.org/) (Szklarczyk et al. 2025). Candidate genes were first evaluated across Text mining, Experiments, Databases, Coexpression, Neighborhood, Gene Fusion, and Cooccurrence channels, and only interactions with a combined score ≥0.40 were retained (Özkan et al. 2025a). To reduce network complexity, the first shell was restricted to a maximum of 10 interactors per query protein, and the second shell to no more than 20 interactors (Özkan et al. 2025b). Unconnected proteins were excluded, and the resulting PPI network was imported into Cytoscape software (version 3.10.3) for visualization. Densely connected protein modules were identified using the Molecular Complex Detection (MCODE) plugin (version 2.0.3), with default parameters applied (degree cutoff ≥2, node score cutoff ≥0.2, k-core ≥2, maximum depth = 100) (Setiani et al. 2023). The cluster with the greatest MCODE score was selected and visualized.

Functional enrichment analysis of the thirty-two proteins interacting with IGF-1 and VEGF-A was conducted using the STRING database (version 12.0). Gene Ontology (GO) was used to evaluate biological process (BP), molecular function (MF) and cellular component (CC) categories, while pathway annotation was based on the KEGG and Reactome databases (Szklarczyk et al. 2023).

Statistical analyses

The normality of all variables was confirmed using the Shapiro-Wilk test. Descriptive statistics for each variable were calculated and presented as “Mean ± Standard Error of Mean (SEM)”. One-way ANOVA followed by Tukey’s test was used to evaluate the effects of genotype on egg and chick weights according to the model:

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Yij} = \mu + \mathrm{Gi} + \mathrm{eij}$$\end{document}

Where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Yij}$$\end{document} is the single observation, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document} is the general mean, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Gi}$$\end{document} is the effect of genotype and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{eij}$$\end{document} is the random error. Two-way ANOVA was applied to assess genotype and day effects on embryo and yolk sac weights and fatty acid composition according to the model:

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Yijk} = \mu + \mathrm{Gi} + \mathrm{Dj} + \left( \text{G x D} \right)\mathrm{ij} + \mathrm{eijk}$$\end{document}

Where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Yijk}$$\end{document} is the single observation, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document} is the general mean, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Gi}$$\end{document} is the effect of genotype, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{Dj}$$\end{document} is the effect of day, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\text{G x D}} \right)\mathrm{ij}$$\end{document} is the interaction term and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{eijk}$$\end{document} is the random error. For fatty acid comparisons, we conducted ANOVA followed by Benforroni correction to adjust p-values for multiple comparisons. Chi-square tests were used to analyze differences in fertility, total embryonic mortality and hatchability among genotypes. For gene expression analysis, expression levels of target genes were normalized against the housekeeping gene ACTB and analyzed using the 2^−ΔΔCt method (Livak and Schmittgen 2001), which incorporates Bonferroni correction. When determining the minimum sample size, the type I error probability (α) was set to 0.05 for a power of (1 – β) = 0.80 and an effect size (f) of 0.50. Accordingly, it was calculated that a minimum of 8 samples for each group would be sufficient for the current study. The power analysis was performed using the PASS 11 and G*Power Version 3.1 statistical software programs. Results were expressed as fold changes, and p-values less than 0.05 were considered statistically significant.

Statistical significance for protein- protein interaction (PPI) was assessed by applying the false discovery rate correction method of Benjamini and Hochberg, and terms or pathways with adjusted p values below 0.01 were considered significant. The ten most relevant GO terms and pathways consistent with the study hypothesis were plotted in R Studio (version 2024.09.1) using the ggplot2 package (version 3.5.1).

Results

Table 1 presents the egg weights, chick weights, and relative chick weights for different genotypes collected during the same laying period. Significant differences were observed among genotypes in terms of egg weight (p < 0.001), with layer hybrids exhibiting the greatest values and Gerze breeds the lowest.Table 1. Egg weight, chick weight, and relative chick weight in different genotypes (n = 258 each genotype)GenotypesEgg weight (g)Chick weight (g)*Relative chick weight (%)*Denizli breed53.20 ± 0.30^c^38.28 ± 0.29^c^72.04 ± 0.58Gerze breed52.22 ± 0.28^d^36.40 ± 0.32^d^69.90 ± 0.78Broiler hybrid56.52 ± 0.24^b^40.12 ± 0.24^b^70.58 ± 0.70Layer hybrid59.68 ± 0.23^a^41.13 ± 0.24^a^69.20 ± 0.30P<0.001<0.0010.058^a, b, c, d:^ Differences among genotypes means different lettersin the same column are statistically significant (P<0.05). *:Samples collected throughout the trial were not included in the calculation ofchick weight and relative chick weight means

In the Denizli and Gerze breeds, relative embryo weights increased from 4.68% and 4.64% on day 10 to 49.37% and 46.64% on day 19, respectively (p < 0.001, Table 2). Similarly, broiler and layer hybrids showed increases from 5.29% and 3.80% on day 10 to 54.58% and 49.16% on day 19, respectively (p < 0.001). Genotype had a statistically significant impact on both relative embryo weight and residual yolk sac weight (p < 0.001). Among the genotypes, broiler hybrids had the greatest relative embryo weight, while Denizli and Gerze breeds showed the greatest relative yolk sac weight. Significant genotype × incubation day interactions were detected for both traits (p < 0.01 for embryo weight; p < 0.05 for yolk sac weight).Table 2. Relative embryo weight and relative residual yolk sac weight by genotype on different incubation dayGenotypeIncubationRelative embryoRelative residual yolk sacday (n = 8)weight (%)weight (%)Denizli breed104.6848.411312.5837.681629.8436.431949.3728.28Gerze breed104.6447.551315.2136.721629.0136.031946.6428.64Broiler hybrid105.29441315.6534.371635.9130.721954.5822.74Layer hybrid103.850.621311.9534.251628.5429.561949.1623.86Denizli breed24.12^y^37.70^x^Gerze breed23.88^y^37.23^x^Broiler hybrid27.86^x^32.96^y^Layer hybrid23.36^y^34.57^y^Day 104.60^d^47.64^a^Day 1313.85^c^35.75^b^Day1630.83^b^33.19^c^Day 1949.94^a^25.88^d^SEM0.2560.32PGenotype<0.001<0.001Incubation day<0.001<0.001Genotype × Incubation day0.0020.013^x, y^: Differences among genotypes means within the same column are statistically significant (P<0.05).^a, b, c, d^: Differences among incubationdays means within the same column are statistically significant (P<0.05)

Chick weights differed significantly among genotypes (p < 0.001), whereas relative chick weights showed no significant variation. Fertility (p < 0.05), embryonic mortality (p < 0.05), hatchability of fertile eggs (p < 0.05), and hatchability of set eggs (p < 0.001) were all significantly influenced by genotype (Table 3).Table 3. Hatching results by genotypeGenotypeFertilityrate (%)Total embryonicmortality (%)Hatchability of fertile eggs (%)*Hatchability ofset eggs (%)*Denizli breed (n = 218)87.00^b^4.64^c^94.16^a^79.24^b^Gerze breed (n = 218)87.89^b^11.23^a^84.93^b^67.76^c^Broiler hybrid (n = 218)94.17^a^4.76^c^94.12^a^87.43^a^Layer hybrid (n = 218)91.93^ab^8.78^b^89.09^ab^80.33^b^χ^2^8.709.1210.8121.70P0.0340.0280.013<0.001^a, b, c, d^ Differences among genotypes means different letters in the same column are statistically significant (p < 0.05). *:Samples collected throughout the trial were not included in the calculation

Fertility rates were higher in hybrid genotypes compared to the local breeds, while Gerze showed the greatest embryonic mortality. The greatest hatchability of fertile and set eggs were observed in broiler hybrids, with Gerze exhibiting the lowest efficiency (p < 0.05). Tables 4, 5, 6 present the mean values for fatty acid profiles in the yolk. In yolk samples, palmitic acid (C16:0) and stearic acid (C18:0) were the most abundant saturated fatty acids (SFA). Genotype had a significant effect on total SFA content (p < 0.01), and temporal changes in SFA during incubation varied by genotype (p < 0.01). Except for C17:0, C20:0, and C22:0, genotype significantly influenced all SFAs (p < 0.001). Among monounsaturated fatty acids (MUFA), C16:1 and C18:1n-9 were most prevalent, with total MUFA levels differing significantly by genotype (p < 0.001). The same trend was observed over the incubation period (p < 0.01). Polyunsaturated fatty acid (PUFA) content was dominated by C18:2n-6, with total PUFA levels significantly affected by genotype (p < 0.01). With the exception of C18:3n-6 and C20:5n-3, genotype significantly influenced all PUFAs (p < 0.001).Table 4. Saturated fatty acid (SFA) composition (%) in the egg yolk at different days of incubation in various genotypesGenotypeIncubation day (n = 8)C14:0C15:0C16:0C17:0C18:0C20:0C21:0C22:0C23:0ΣSFADenizlibreedPre-incubation0.430.0725.450.2611.170.050.220.100.2938.03Day 100.420.0525.450.239.980.040.150.090.2436.58Day 130.470.0524.310.258.930.050.120.100.2134.35Day 160.450.0523.810.269.070.110.130.080.1834.11Day 190.470.0624.920.229.410.110.180.090.2635.62Hatch0.430.0525.160.239.060.050.150.070.2435.32GerzebreedPre-incubation0.420.0625.420.258.390.050.240.090.3435.25Day 100.350.0725.010.2610.490.070.210.090.3036.79Day 130.400.0624.500.259.180.060.170.080.2134.87Day 160.420.0824.570.268.780.150.180.080.2534.65Day 190.400.0724.650.2610.100.050.290.080.3136.05Hatch0.410.0624.250.239.080.060.210.080.2834.59Broiler hybridPre-incubation0.470.0925.230.278.610.050.290.050.3235.35Day 100.440.0825.130.248.230.040.240.040.2534.66Day 130.430.0826.200.248.160.050.240.050.2235.57Day 160.470.0726.160.238.380.150.220.060.2435.92Day 190.430.0725.500.228.550.070.250.080.2235.32Hatch0.470.0724.950.239.500.050.230.050.2536.48Layer hybridPre-incubation0.300.0624.190.2511.050.050.310.190.1436.46Day 100.310.0523.460.2310.510.050.250.050.1235.00Day 130.340.0523.410.249.570.040.220.060.1133.98Day 160.340.0523.520.249.150.100.240.060.1233.66Day 190.320.0523.330.259.760.060.290.050.1034.17Hatch0.340.0523.350.249.350.010.290.060.1033.72Denizli breed0.44^x^0.05^z^24.85^y^0.249.61^x^0.070.16^w^0.090.24^y^35.67^x^Gerze breed0.40^y^0.07^y^24.73^y^0.259.34^x^0.070.21^z^0.080.28^x^35.37^x^Broiler hybrid0.45^x^0.07^x^25.53^x^0.248.57^y^0.070.24^y^0.050.25^y^35.55^x^Layer hybrid0.32^z^0.05^z^23.54^z^0.249.90^x^0.060.27^x^0.080.12^z^34.50^y^Pre-incubation0.400.07^a^25.070.269.81^a^0.05^b^0.26^a^0.110.27^a^36.27Day 100.380.06^ab^24.760.249.80^a^0.05^b^0.21^bcd^0.070.23^b^35.76Day 130.410.06^b^24.610.258.96^c^0.05^b^0.19^cd^0.070.19^c^34.69Day 160.420.06^ab^24.520.248.84^c^0.13^a^0.19^d^0.070.20^ab^34.58Day 190.410.06^b^24.600.249.46^ab^0.07^b^0.24^ab^0.080.22^b^35.29Hatch0.410.06^b^24.430.239.25^ab^0.07^b^0.22^bc^0.070.22^b^35.03SEM0.0040.0010.0670.0030.1020.0040.0040.0100.0050.121PGenotype<0.001<0.001<0.0010.350<0.0010.910<0.0010.551<0.0010.004Incubation day0.1840.0080.0890.1470.020<0.001<0.0010.611<0.001<0.001Genotype ×Incubation day0.6030.4150.0020.6960.0170.5830.9830.9410.6270.001^x, y, z^ Differences among genotypes means within the same column are statistically significant (p < 0.05). ^a,b,c,d:^ Differences among incubation days means within the same column are statistically significant (p < 0.05)Table 5. Monounsaturated fatty acid (MUFA) composition (%) in the egg yolk at different ıncubation days in various chicken genotypesGenotypeIncubation day (n = 8)C14:1C16:1C17:1C18:1n-9C20:1ΣMUFADenizlibreedPre-incubation0.073.890.1340.920.3045.32Day 100.083.870.1142.480.2446.77Day 130.084.440.1447.160.3151.63Day 160.084.320.1346.780.3151.57Day 190.083.980.1044.800.2949.24Hatch0.063.350.1145.020.3348.85GerzebreedPre-incubation0.094.260.1544.060.3648.86Day 100.063.690.1341.990.2346.10Day 130.074.020.1444.530.3149.06Day 160.084.320.1543.640.2848.47Day 190.073.630.1141.860.2545.88Hatch0.083.710.1043.950.3348.12Broiler hybridPre-incubation0.114.810.1838.370.3143.75Day 100.104.710.1639.910.3145.19Day 130.104.800.1638.460.3043.82Day 160.114.960.1438.450.2843.86Day 190.104.830.1439.610.3344.99Hatch0.084.480.1639.050.3444.07Layer hybridPre-incubation0.052.710.1736.870.3040.09Day 100.032.910.1337.810.2641.29Day 130.043.150.1440.240.2643.82Day 160.043.480.1540.630.2844.57Day 190.042.970.1338.230.3041.61Hatch0.042.910.1140.070.3043.37Denizli breed0.08^y^3.97^y^0.12^z^44.53^x^0.3048.89^x^Gerze breed0.08^y^3.94^y^0.13^yz^43.34^y^0.2947.75^y^Broiler hybrid0.10^x^4.77^x^0.16^x^38.97^z^0.3144.28^z^Layer hybrid0.04^z^3.02^z^0.14^y^38.97^z^0.2842.46^z^Pre-incubation0.083.92^bc^0.16^a^40.06^b^0.32^ab^44.50^c^Day 100.073.79^bc^0.13^bcd^40.54^b^0.26^c^44.84^c^Day 130.084.10^ab^0.14^ab^42.60^a^0.29^abc^47.08^a^Day 160.084.27^a^0.14^abc^42.37^a^0.29^bc^47.12^a^Day 190.073.86^bc^0.12^d^41.13^ab^0.29^abc^45.43^bc^Hatch0.073.61^c^0.12^cd^42.02^a^0.32^a^46.10^ab^SEM0.0020.0470.0030.1910.0050.195PGenotype<0.001<0.001<0.001<0.0010.225<0.001Incubation day0.1730.002<0.001<0.0010.002<0.001GenotypeX Incubation day0.7810.7070.7390.0070.3250.002^x, y:^ Differences among genotypes means within the same column are statistically significant (p < 0.05). ^a,b,c,d:^ Differences among incubation days means within the same column are statistically significant (P 0.05)Table 6. Polyunsaturated (PUFA) and total unsaturated fatty acid (UFA) composition (%) in the egg yolk at different incubation days in various chicken genotypesGenotypeIncubation day (n = 8)C18:2n-6C18:3n-6C18:3n-3C20:3n-6C20:4n-6C20:5n-3C22:6n-3ΣPUFAΣUFADenizlibreedPre-incubation13.330.140.200.302.410.070.4116.6561.97Day 1013.520.140.180.202.290.050.3416.6663.42Day 1312.510.140.220.131.090.060.1214.0265.65Day 1612.750.190.200.151.010.300.1514.3265.89Day 1913.140.160.190.181.38-0.1615.1464.38Hatch13.920.130.220.161.320.150.1515.8364.68GerzebreedPre-incubation14.310.160.230.271.510.080.3315.8964.75Day 1013.620.170.170.272.430.230.3217.1263.22Day 1315.100.160.250.171.210.070.1316.0865.14Day 1614.180.200.250.221.480.120.1216.8865.35Day 1915.220.180.220.242.050.080.1518.0763.96Hatch14.160.150.200.222.37-0.2617.2965.41Broiler hybridPre-incubation17.380.150.800.291.550.190.5420.8964.65Day 1017.360.150.780.211.160.140.3720.1665.34Day 1318.040.140.840.191.020.140.2820.6164.43Day 1617.580.190.810.191.240.130.2820.2264.08Day 1917.260.160.780.201.050.120.1719.6964.68Hatch16.990.170.720.211.090.110.1819.4563.52Layer hybridPre-incubation19.610.350.950.251.570.080.7523.4663.54Day 1019.760.150.950.201.720.080.7823.7165.00Day 1319.240.150.850.161.170.070.3122.2066.02Day 1619.080.160.960.170.950.070.2421.7766.34Day 1921.410.171.200.201.210.080.3224.2265.84Hatch20.710.150.960.190.890.130.2022.9266.28Denizli breed13.20^w^0.15^z^0.20^z^0.19^y^1.58^x^0.120.22^yz^15.44^w^64.33^y^Gerze breed14.43^z^0.17^z^0.22^z^0.23^x^1.84^x^0.110.22^z^16.89^z^64.64^y^Broiler hybrid17.44^y^0.16^y^0.79^y^0.22^x^1.18^y^0.140.30^y^20.17^y^64.45^y^Layer hybrid19.97^x^0.19^x^0.98^x^0.19^y^1.25^y^0.090.43^x^23.05^x^65.50^x^Pre-incubation16.160.200.540.28^a^1.76^ab^0.110.51^a^19.2263.73^c^Day 1016.070.150.520.22^b^1.90^a^0.120.45^a^19.4164.24^bc^Day 1316.220.150.540.16^d^1.12^c^0.090.21^b^18.2365.31^a^Day 1615.900.180.550.18^cd^1.17^c^0.150.20^b^18.3065.42^a^Day 1916.760.170.600.21^bc^1.42^bc^0.100.20^b^19.2864.71^ab^Hatch16.450.150.520.19^bc^1.41^bc^0.130.20^b^18.8764.97^ab^SEM 0.1740.0070.0140.0040.0560.0110.0130.1760.121PGenotype<0.0010.195<0.001<0.001<0.0010.128<0.001<0.0010.004Incubation day 0.7690.1710.628<0.001<0.0010.464<0.0010.213<0.001Genotype ×Incubation day0.9060.0660.4700.2940.0430.0750.0080.5280.001^x, y, z^ Differences among genotypes means within the same column are statistically significant (p < 0.05). ^a,b,c,d:^ Differences among incubation days means within the same column are statistically significant (p < 0.05)

For IGF-1 gene expression, Denizli breed and Broiler hybrid chicks showed significantly higher expression in the breast muscle compared to layer hybrids (p < 0.001 and p < 0.05, respectively) (Fig. 1).In contrast, IGF-I was downregulated in the embryonic muscle of Denizli compared to both layer and broiler hybrids (p < 0.05). Regarding VEGF-A expression, broiler embryos showed upregulation compared to Gerze (p < 0.01), while chick muscle tissue showed downregulation (p < 0.05). VEGF-A was significantly upregulated in Gerze embryos and chicks relative to Denizli (p < 0.05). Furthermore, both local breeds exhibited higher VEGF-A expression in embryonic tissue than layer hybrids (p < 0.05).Fig. 1. Gene expression analyses (A and B: Layer hybrid was control; C and D: broiler hybrid was control). *: p < 0.05, **: p < 0.01, ***: p < 0.001

The PPI network comprised 32 IGF1/VEGF-A interacting proteins connected by 342 edges (Fig. 2A). MCODE analysis of this network identified a densely interconnected module of 23 proteins linked by 234 edges, yielding the greatest cluster score (21.273) (Fig. 2B). Notably, this subnetwork is dominated by diverse growth factors (e.g. fibroblast growth factors, proepidermal growth factor, placental growth factor) alongside key angiogenic regulators, including neuropilins (NRP1, NRP2) and VEGF receptors FLT1 and FLT4. These findings imply that intricate interplay among these molecules may coordinate IGF1/VEGF-A driven vascular remodeling.Fig. 2PPI analysis (A: global network proteins interacting with IGF‑1 and VEGF-A; B: core module identified by MCODE (score 21.273))

GO enrichment of the 32protein network highlighted system development, cellular response to growth factor stimulus, animal organ development and blood vessel development as the most significant biological processes, with growth factor activity, growth factor binding and fibroblast growth factor receptor binding predominating among molecular functions and receptor complex enriched as the cellular component (Fig. 1A). Moreover, pathway analysis (Figure 1B) identified MAPK signaling in KEGG, broad signal transduction and receptor tyrosine kinase signaling in Reactome, FGFR2 ligand binding and activation, the PI3K cascade via FGFR1, VEGF-A binding to VEGFR leading to receptor dimerization, negative regulation of FGFR1 signaling, hemostasis, IRSrelated events triggered by IGF1R and the calcium signaling pathway as the top pathways.Fig. 3. Functional enrichment analysis results. (A: GO enrichment across BP, MF and CC; B: pathway enrichment based on KEGG and Reactome; FDR: false discovery rate)

Discussion

Animal biodiversity management has become an important issue because of changes in large-scale production systems. Local breeds appear to have the advantage of being well adapted to the local conditions in rural conditions (Onbaşılar et al. 2017, 2018a). In this study, superior egg weights observed in commercial layer hybrids are consistent with breeding programs that emphasize traits such as high egg yield combined with optimal egg weight, and good shell quality (Thiruvenkadan et al. 2010). In contrast, the lower egg weights in the local Denizli and Gerze breeds may be attributed to the absence of systematic selection for egg production traits. Notably, eggs from the Denizli breed were consistently heavier than those from the Gerze breed, supporting previous reports indicating superior performance of the Denizli breed (Özdoğan and Gürcan 2006).

Embryogenesis is a highly energy-demanding process, with roughly 80% of the yolk’s total energy allocated to tissue growth. After accounting for approximately 5% of energy loss through metabolic waste, about 15% of the initial yolk energy remains to sustain other crucial physiological processes throughout development (Buzała et al. 2015). Egg yolk comprises a complex mixture of biologically active substances, including lipids, carbohydrates, hormones, and antibodies, all of which are vital for supporting embryonic development (Buzała et al. 2015). During early incubation, yolk nutrients serve as the primary energy source, directly affecting the pace of embryonic growth (Ho et al. 2011). In this study, embryonic development varied significantly among genotypes, with broiler hybrids exhibiting the greatest relative embryo weights. This finding is consistent with Everaert et al. (2011), who reported that layer embryos from older flocks develop more slowly than broiler embryos, resulting in lower embryo weights up to day 16. Likewise, Ohta et al. (2004) observed similar differences in embryonic growth between broiler and layer hybrids. These distinctions likely arise from intensive genetic selection aimed at enhancing muscle development and growth efficiency in commercial broiler hybrids. Moreover, selection for rapid post-hatch growth appears to influence nutrient mobilization during embryogenesis. Conversely, local breeds displayed reduced yolk utilization, evidenced by higher relative yolk sac weights at hatch. This slower yolk absorption may reflect a more conservative developmental strategy, potentially representing an evolutionary adaptation to variable or less-controlled environmental conditions. Onbaşılar et al. (2018b) reported that embryos of white layer genotypes utilized the yolk sac more quickly during incubation than the embryos of brown layer genotypes.

Chick weights were higher in commercial hybrids and lower in local breeds. However, when adjusted for initial egg weight, relative chick weight did not differ significantly among genotypes, suggesting that smaller local eggs are nonetheless effective in supporting embryonic growth. This could reflect improved shell quality or reduced moisture loss during incubation, which may confer certain physiological advantages. Such a pattern highlights the need for further investigation into egg shell microstructure and water conductance in local breeds.

In terms of hatchability traits, commercial hybrids outperformed local breeds in fertility and hatching efficiency. Broiler hybrids in particular achieved the greatest hatchability rates and lowest embryonic mortality. The Gerze breed, on the other hand, exhibited reduced fertility and higher embryonic loss, suggesting genetic or physiological limitations under standardized incubation conditions. The Denizli breed demonstrated intermediate outcomes, with hatchability metrics that approached those of commercial lines. This finding supports the potential of Denizli breed as a more viable local resource for future breeding or conservation programs. Variability in hatch performance among genotypes may also be influenced by pre-incubation maternal factors, such as egg quality and shell conductance, as well as genetic differences in embryonic metabolism. The relatively poorer hatch outcomes in local breeds may therefore reflect inherent genetic differences. These findings underscore the importance of optimizing both genetic and environmental conditions to enhance hatchability and early chick viability in diverse poultry genotypes.

Significant differences were found between the genotypes in terms of SFAs in the yolk sac. C16:0, a which major SFA, was detected at intermediate levels in broiler and layer hybrids compared to Denizli and Gerze samples. The effect of genotype on fatty acid composition in egg yolk is consistent with the literature (Johansson 2010). One reason for this consistency is that the eggs in this study were obtained from breeders fed with different rations based on their genotypes. Previous studies have reported that nutrition has an effect on egg fatty acid composition (Cherian et al. 2001; Poureslami et al. 2012). Many studies have shown that the fatty acid profile of chicken embryos varies with the age of the breeders and the feed consumed, genotype and geographical region (Kostaman et al. 2021; Milinsk et al. 2003). However, in this study, C18:2n6 and C18:3n3 values were lower in local genotypes (Denizli and Gerze) compared to hybrid genotypes, while C20:3n6 value was high. This is consistent with the report of Kostaman et al. (2021) for Cemani, a local breed, and White Leghorn, a layer hybrid. This is an important finding, showing the difference between local breeds and hybrid genotypes. The fact that fatty acids such as C18:2n6 and C18:3n3 were higher in hybrid genotypes than in native genotypes can be explained by their presence in the cell wall structure and their suggested importance for the development of the nervous system.

Individual differences in SFA levels also affected SFA, with values generally consistent with previous findings (Burnham et al. 2001; Poureslami et al. 2012; Yalçın et al. 2008). On the other hand, incubation period had no significant effect on SFA. Individual fatty acids can be transformed during embryonic development through various mechanisms, such as elongation and desaturation. Kucharska-Gaca et al. (2023) reported changing ratios of fatty acids in geese at different periods of incubation. Similar to SFA, genotype and incubation period had a significant effect on MUFA. In our study, the MUFA values in Denizli and Gerze breeds were higher than those of broiler and layer hybrids, which can be considered advantageous. Both the individual and total index SFA and MFA values were significantly different regarding the genotype and incubation period interaction. More specifically, genotype, but not incubation period, had a significant effect on PUFA levels. Previous studies have reported that temperature and humidity have significant effects on PUFA levels during incubation (Burnham et al. 2001; Yalçın et al. 2008), but no direct effect of incubation period has been reported. In the present study, hybrid genotypes had higher PUFA levels than local genotypes, which can be considered an advantage of hybrid genotypes because PUFA is thought to have an effect on both nervous system and organ development (Yalçın et al. 2008).

Several genes are involved in muscle development during embryonic myogenesis, with insulin-like growth factor I (IGF-I) playing a crucial role in stimulating myoblast proliferation and muscle formation during early embryogenesis (Yu et al. 2015). Al-Musawi et al. (2011) reported that, on day 15 of embryonic development, myoblast proliferation was higher in broiler embryos than in layers, and was followed by an increase in IGF-I levels on day 16. In our study, the IGF-1 gene expression was similar between broiler and the layer genotypes. This finding is similar to that reported by Zhu et al. (2021) for muscle IGF1 gene expression in 17-day-old broiler and layer hybrids. On the other hand, on the day of hatching, chick muscle IGF-1 gene expression values were significantly upregulated in broiler hybrids compared to layer hybrids. These findings confirm our prediction that myoblast proliferation is higher in broiler hybrids than in layers (Al-Musawi et al. 2011). IGF-1 gene expression was lower in embryo muscle tissue in the Denizli breed than in both laying and broiler hybrids, thereby supporting the findings of El-Attrouny et al. (2020). This shows that IGF-1 gene expression in non-selected local breeds is lower in terms of muscle proliferation than in hybrid genotypes. Kita et al. (2005) reported that IGF-1 is expressed at different levels in the brain, eye, and liver at different times of embryo development, while Bhattacharya et al. (2015) showed that IGF-1 expression changes at different periods of embryonic development.

IGF-1 gene expression was significantly downregulated in Denizli breed embryos compared to laying hybrids, but upregulated in hatching day chicks. In chicken embryos, IGF1 expression can vary at different developmental stages and there may be differences even during embryonic and hatching stages (Bhattacharya et al. 2015; Kanački et al. 2012). Taking the laying and broiler hybrid as controls, IGF-1 gene expression in both embryo and chick muscle tissue was not significantly different from the Gerze breed. However, it is noteworthy that gene expression was lower than in broiler hybrids. It was expected that IGF-1 expression would be higher in broiler hybrids than Gerze birds, which is a native non-reproducing breed. Taking the Denizli breed as the control, IGF-1 gene expression was upregulated in Gerze breed embryos on day 19, although the difference was insignificant. In newly hatched chicks, the Gerze breed was significantly down- regulated. This suggests that the Gerze breed is similar to Denizli breed in terms of embryonic muscle development; however, the Denizli breed may show better values in the chick stage.

The development of muscle vascular network is necessary to provide nutrients and oxygen to the cells. VEGF-A is an angiogenesis regulator responsible for the development of the muscle vascular system (Ferrara and Gerber 2002). In the present study, muscle VEGF-A gene expression did not differ between laying and broiler hybrids in both day 18 embryos and newly hatched chicks. Taking the laying hybrids as the control, VEGF-A gene expression was lower in embryo muscle tissue at day 19 in the native genotypes. The fact that a vital gene such as VEGF-A differs between genotypes suggests that genetic structure is very important in terms of both productivity and other traits (Oosterbaan et al. 2012). It is noteworthy that muscle tissue VEGF-A gene expression during the transition from the embryonic to chick stage increased more in native breeds than laying hybrids. This suggests that muscle tissue angiogenesis occurs later in the former than the latter birds. These findings also support the basic finding that embryos of native breeds generally develop more slowly (Akçapınar and Özbeyaz 2021).

Taking broiler hybrids as the control, VEGF-A gene expression in muscle tissue in Denizli breed birds did not differ in day 18 embryos and newly hatched chicks. This can be considered advantageous for Denizli breeds, an indigenous breed without a formal breeding program, unlike the Gerze breed. Using broiler hybrids as the control, VEGF-A gene expression in muscle tissue was significantly down-regulated in day Gerze embryos but upregulated in newly hatched chicks. In Gerze breeds, VEGF-A gene expression was higher in day-old chicks than on embryonic day 18. This suggests that important angiogenesis activity may have been triggered before hatching. Yalçın et al. (2022) reported that muscle tissue VEGF-A gene expression levels differed between two broiler hybrids in both day 19 embryos and newly hatched chicks. This suggests that VEGF-A gene expression may vary not only by embryonic developmental stage but also by genotype.

PPI network analysis revealed that the IGF-1/VEGF-A interactome extends beyond classical angiogenic regulators (FLT1, FLT4), encompassing placental growth factor (PGF) and multiple fibroblast growth factors (FGF1, FGF3, FGF5, FGF7, FGF22). PGF is a critical angiokine co-expressed with VEGF during early embryonic vasculogenesis and has been shown to stimulate endothelial cell proliferation and migration at levels comparable to or exceeding those induced by bFGF and VEGF (Pei et al. 2022). The downregulation of IGF-1observed in Denizli and Gerze embryos may attenuate PGF-mediated pro-angiogenic signaling, potentially contributing to their reduced hatchling weights. Conversely, the post-hatch upregulation of IGF-1 in Denizli chicks may reflect the termination of PGF’s extraembryonic role. MCODE clustering further identified FGF1, FGF3, FGF5, FGF7, and FGF22 as central network hubs. In chicken embryos, FGFs are recognized as essential factors promoting myoblast proliferation and delaying differentiation to ensure sufficient progenitor pools for muscle development (Truong et al. 2022). The high connectivity of these FGFs within the IGF-1/VEGF-A network underscores the pivotal role of FGF-dependent proliferative stages in embryonic myogenesis.

Pathway enrichment analysis showed that the identified growth factors and their receptors primarily signal through MAPK and tyrosine kinase pathways, which coordinate cellular functions through ligand–receptor binding and phosphorylation mechanisms. The network structure suggests that when VEGF-A is downregulated, other growth factors may compensate for its function. In summary, the findings highlight a robust signaling network involving IGF-1, VEGF-A, PGF, and FGF pathways. To further elucidate the biological significance of this network, detailed investigations focusing on vascular and muscular development in embryonic chicken models are warranted. These studies may facilitate the identification of molecular targets aimed at enhancing both embryonic development and post-hatch performance in native and commercial chicken breeds.

Conclusion

Considering the slower developmental and gene expression patterns in local genotypes, future studies should prioritize the optimization of genotype-specific incubation protocols and the design of targeted nutritional strategies to enhance embryonic growth and post-hatch viability. Furthermore, the integration of local genotypes into sustainable breeding programs may support the conservation of genetic diversity while enabling selective enhancement of favorable traits such as muscle vascularization and yolk utilization efficiency. Long-term studies involving larger populations, multi-generational designs, and the molecular analysis of additional regulatory genes involved in metabolism, immunity, and stress responses will be essential to gain a comprehensive understanding of the physiological trade-offs between intensive performance selection and adaptive genetic potential in poultry. In addition, exploring the influence of fatty acid composition in egg yolks on key biological processes in chicks including inflammation, gene regulation, cellular function, and metabolic activity may enhance the applicability of the present findings and provide a foundation for more effective breeding and management strategies.

Limitation of the study

The most significant limitation of this study is that maternal feed could not be standardized according to genotypes.