Genetic-economic aspects of feed efficiency in laying hens

Genome-wide association studies for feed intake and efficiency in two laying periods of chickens

Background

Feed contributes to over 60 % of the total production costs in the poultry industry. Increasing feed costs prompt geneticists to include feed intake and efficiency as selection goals in breeding programs. In the present study, we used an F₂ chicken population in a genome-wide association study (GWAS) to detect potential genetic variants and candidate genes associated with daily feed intake (FI) and feed efficiency, including residual feed intake (RFI) and feed conversion ratio (FCR).

Methods

A total of 1534 F₂ hens from a White Leghorn and Dongxiang reciprocal cross were phenotyped for feed intake and efficiency between 37 and 40 weeks (FI1, RFI1, and FCR1) and between 57 and 60 weeks (FI2, RFI2, and FCR2), and genotyped using the chicken 600 K single nucleotide polymorphism (SNP) genotyping array. Univariate, bivariate, and conditional genome-wide association studies (GWAS) were performed with GEMMA, a genome-wide efficient mixed model association algorithm. The statistical significance threshold for association was inferred by the simpleM method.

Results

We identified eight genomic regions that each contained at least one genetic variant that showed a significant association with FI. Genomic regions on Gallus gallus (GGA) chromosome 4 coincided with known quantitative trait loci (QTL) that affect feed intake of layers. Of particular interest, eight SNPs on GGA1 in the region between 169.23 and 171.55 Mb were consistently associated with FI in both univariate and bivariate GWAS, which explained 3.72 and 2.57 % of the phenotypic variance of FI1 and FI2, respectively. The CAB39L gene can be considered as a promising candidate for FI1. For RFI, a haplotype block on GGA27 harbored a significant SNP associated with RFI2. The major allele of rs315135692 was favorable for a lower RFI, with a phenotypic difference of 3.35 g/day between opposite homozygous genotypes. Strong signals on GGA1 were detected in the bivariate GWAS for FCR.

Conclusions

The results demonstrated the polygenic nature of feed intake. GWAS identified novel variants and confirmed a QTL that was previously reported for feed intake in chickens. Genetic variants associated with feed efficiency may be used in genomic breeding programs to select more efficient layers.

Electronic supplementary material

The online version of this article (doi:10.1186/s12711-015-0161-1) contains supplementary material, which is available to authorized users.

Genetic parameters of feed efficiency traits in laying period of chickens

Laying records on 1,534 F₂ hens, derived from a reciprocal cross between White Leghorns and Dongxiang blue-shelled chickens, were used to estimate genetic parameters for residual feed intake (RFI), feed conversion ratio (FCR), daily feed intake (FI), metabolic BW (MBW), BW gain (BWG), and daily egg mass (EM) at 37 to 40 (T1) and 57 to 60 wk age (T2), respectively. Genetic analysis was subsequently conducted with the AI-REML method using an animal model. Estimates for heritability of RFI, FCR, and FI were 0.21, 0.19, and 0.20 in T1, and 0.29, 0.13, and 0.26 in T2, respectively. In T1 and T2, RFI showed high and positive genetic correlations with FCR (0.51, 0.43) and FI (0.72, 0.84), whereas the genetic correlation between FI and FCR was very low (−0.09, 0.11). Genetically, negative correlations were found between RFI and its component traits (−0.01 to −0.47). In addition, high genetic correlations, from 0.76 to 0.94, were observed between T1 and T2 for RFI, FCR, and FI, suggesting that feed efficiency traits in the 2 stages had a similar genetic background. The results indicate that selection for low RFI could reduce FI without significant changes in EM, while selection on FCR will increase EM. The present study lays the foundation for genetic improvement of feed efficiency during the laying period of chickens.

Pedigree and genomic analyses of feed consumption and residual feed intake in laying hens

Efficiency of production is increasingly important with the current escalation of feed costs and demands to minimize the environmental footprint. The objectives of this study were 1) to estimate heritabilities for daily feed consumption and residual feed intake and their genetic correlations with production and egg-quality traits; 2) to evaluate accuracies of estimated breeding values from pedigree- and marker-based prediction models; and 3) to localize genomic regions associated with feed efficiency in a brown egg layer line. Individual feed intake data collected over 2-wk trial periods were available for approximately 6,000 birds from 8 generations. Genetic parameters were estimated with a multitrait animal model; methods BayesB and BayesCπ were used to estimate marker effects and find genomic regions associated with feed efficiency. Using pedigree information, feed efficiency was found to be moderately heritable (h(2) = 0.46 for daily feed consumption and 0.47 for residual feed intake). Hens that consumed more feed and had greater residual feed intake (lower efficiency) had a genetic tendency to lay slightly more eggs with greater yolk weights and albumen heights. Regions on chromosomes 1, 2, 4, 7, 13, and Z were found to be associated with feed intake and efficiency. The accuracy from genomic prediction was higher and more persistent (better maintained across generations) than that from pedigree-based prediction. These results indicate that genomic selection can be used to improve feed efficiency in layers.

Individual production differences do not explain cannibalistic behaviour in laying hens

1. Sixty-three laying hens were collected from 5 commercial farms experiencing outbreaks of cannibalism. It is known that cannibalistic behaviour varies strongly between hybrids and between individuals of the same hybrid. The aim of this study was to test specific hypotheses about the causation of cannibalism by comparing matched individuals varying in their cannibalistic behaviour. 2. To investigate feed efficiency, egg production and feed consumption, data were collected over 30 d. Nutrient deficiencies have been associated with outbreaks of cannibalism and we hypothesised that cannibals are more feed efficient than their flock mates and, therefore, more susceptible to transient nutrient deficiencies or imbalances that trigger cannibalistic behaviour. 3. Video recordings were made of victims of cloacal cannibalism and their matched controls during oviposition. Victims were expected to have longer oviposition durations and a higher incidence of short-term cloacal prolapses than controls, because the starting point for cannibalism is often suggested to be pecking at the red mucosa of the victim during, or just after, oviposition. 4. Oviduct swab samples were taken from surviving victims of cloacal cannibalism and their matched controls 4 weeks after the cannibalistic attack. If peck wounds provide a route for subsequent bacterial invasion of the reproductive tract, victims would be expected to have greater bacterial colonisation of the oviduct. 5. Mortality within each matched trio was recorded over a 23-month period after the cannibalistic attack. Victims that survived a cannibalistic attack were expected to have reduced longevity compared to controls and cannibals. This could occur if they developed secondary infections as a result of the cannibalistic attack, or if they had poorer health even before the attack. 6. No significant differences were found between cannibals, victims and controls in feed efficiency or any of the production variables measured. Victims of cloacal cannibalism did not expose their cloacal mucosa more, or for longer, than matched control hens during oviposition, and had no more bacteria in their oviduct. Victims tended to die sooner than their matched cannibals or cannibals during a 23-month period after a cannibalistic attack.

Characterization of energetic efficiency in adult broiler breeder hens

This trial characterized residual feed intake (RFI) and residual maintenance requirement (RME(m)) as measures of energetic efficiency in broiler breeder hens. The RFI was defined as the difference between observed and expected ME intake and RME(m) as the difference between observed and expected maintenance requirements. A total of 600 Ross 708 1-d-old pullets were placed in floor pens. At 16 wk, 144 hens were caged and randomly assigned to 1 of 2 feed allocation treatments (72 birds each). The control treatment had feed allocated on a group basis (GRP) following the standard BW target. A second treatment had feed allocated on an individual-bird basis (IND) and followed the same BW target as GRP. Sexual maturity age, egg and chick production, and several feed conversion ratios were correlated to standardized efficiency indices of RFI (SRFI) and RME(m) (SRME(m)) in each treatment. Greater SRFI and SRME(m) values described a greater energetic efficiency. Residual feed intake was more variable in IND than GRP hens (P < 0.001). The variability of RME(m) did not differ between treatments (P = 0.14). The SRFI was positively correlated to egg production in the GRP hens (r = 0.31), but negatively correlated in IND hens (r = -0.40) and was correlated to feed conversion per chick only in the GRP-based feed allocation (r = -0.44). The SRME(m) correlated strongly to egg production (r = 0.64), chick production (r = 0.64), and feed conversion per chick (r = -0.59) in both feed allocation treatments. Feed intake confounded the RFI calculation, which limits the value of RFI as a selection criterion in meat-producing animals. The independence of RME(m) from feed intake is desirable for energetic efficiency assessment in selection programs because consistent values can be obtained across different management schemes. Hens with lower maintenance requirements (greater RME(m) efficiency) partitioned more energy toward reproduction than did high-maintenance hens. The RME(m) methodology provided an unbiased estimate of energetic efficiency by adjusting the maintenance requirement for the effect of dietary thermogenesis.

Effects of maternal energetic efficiency on egg traits, chick traits, broiler growth, yield, and meat quality

This study assessed egg traits, chick traits, growth, yield, and meat quality characteristics of the offspring from broiler breeders classified by 2 measurements of energetic efficiency: residual feed intake (RFI), defined as the difference between observed and expected ME intake, and residual maintenance requirement (RME(m)), defined as the residual of the relationship between hen maintenance requirement and feed intake. A group of 72 pullets were placed in laying cages from 16 to 60 wk of age. Individual hen-based feed allocation was provided following a standard BW target. At 41wk, eggs from 8 d of production were collected and pedigree hatched. Chicks were assigned to 1 of 3 maternal RFI (RFI(mat)) categories: low, average, and high. A total of 366 chicks were placed in 36 floor pens, 6 per sex x RFI(mat) interaction, and raised to 38 d. At the end of the breeder experiment (60 wk), broilers were retrospectively assigned to a low or high maternal RME(m) (RME(mmat)) category. Low RFI(mat) broilers had greater 38-d BW than average and high RFI(mat) broilers. That was achieved through a greater BW gain and feed intake of low RFI(mat) broilers from 21 to 28 d. It was found that RFI(mat) had no effect on feed conversion, yield, or meat quality characteristics. Low RME(m) hens produced heavier eggs (62.3 g) and chicks (42.5 g) than high RME(m) hens (60.0 g; 41.0 g), but RME(mmat) did not affect broiler 38-d BW. High RME(mmat) broilers had greater breast yield (29.5%) and lower breast shear force (4.7 kg of force/g) than low RME(mmat) broilers (28.5%; 5.6 kg of force/g). The low RFI(mat) x high RME(mmat) broilers had the greatest growth to 38 d. It was found that RFI(mat) was inversely related to broiler growth, particularly when RME(mmat) was high. Although low maintenance requirements may be desirable for egg and chick production, hens with a high maintenance requirement produced broilers with greater breast yield and tenderness. Minimizing maintenance requirements may not be compatible with maximizing broiler performance and meat yield.