Genotyping by low-coverage whole-genome sequencing in intercross pedigrees from outbred founders: a cost-efficient approach.

BACKGROUND: Experimental intercrosses between outbred founder populations are powerful resources for mapping loci that contribute to complex traits i.e. quantitative trait loci (QTL). Here, we present an approach and its accompanying software for high-resolution reconstruction of founder mosaic genotypes in the intercross offspring from such populations using whole-genome high-coverage sequence data on founder individuals (~ 30×) and very low-coverage sequence data on intercross individuals (< 0.5×). Sets of founder-line informative markers were selected for each full-sib family and used to infer the founder mosaic genotypes of the intercross individuals. The application of this approach and the quality of the estimated genome-wide genotypes are illustrated in a large F2 pedigree between two divergently selected lines of chickens. RESULTS: We describe how we obtained whole-genome genotype data for hundreds of individuals in a cost- and time-efficient manner by using a Tn5-based library preparation protocol and an imputation algorithm that was optimized for this application. In total, 7.6 million markers segregated in this pedigree and, within each full-sib family, between 10.0 and 13.7% of these were fully informative, i.e. fixed for alternative alleles in the founders from the divergent lines, and were used for reconstruction of the offspring mosaic genotypes. The genotypes that were estimated based on the low-coverage sequence data were highly consistent (> 95% agreement) with those obtained using individual single nucleotide polymorphism (SNP) genotyping. The estimated resolution of the inferred recombination breakpoints was relatively high, with 50% of them being defined on regions shorter than 10 kb. CONCLUSIONS: A method and software for inferring founder mosaic genotypes in intercross offspring from low-coverage whole-genome sequencing in pedigrees from heterozygous founders are described. They provide high-quality, high-resolution genotypes in a time- and cost-efficient manner. The software is freely available at https://github.com/CarlborgGenomics/Stripes .

Artificial Selection Response due to Polygenic Adaptation from a Multilocus, Multiallelic Genetic Architecture.

The ability of a population to adapt to changes in their living conditions, whether in nature or captivity, often depends on polymorphisms in multiple genes across the genome. In-depth studies of such polygenic adaptations are difficult in natural populations, but can be approached using the resources provided by artificial selection experiments. Here, we dissect the genetic mechanisms involved in long-term selection responses of the Virginia chicken lines, populations that after 40 generations of divergent selection for 56-day body weight display a 9-fold difference in the selected trait. In the F15 generation of an intercross between the divergent lines, 20 loci explained >60% of the additive genetic variance for the selected trait. We focused particularly on fine-mapping seven major QTL that replicated in this population and found that only two fine-mapped to single, bi-allelic loci; the other five contained linked loci, multiple alleles or were epistatic. This detailed dissection of the polygenic adaptations in the Virginia lines provides a deeper understanding of the range of different genome-wide mechanisms that have been involved in these long-term selection responses. The results illustrate that the genetic architecture of a highly polygenic trait can involve a broad range of genetic mechanisms, and that this can be the case even in a small population bred from founders with limited genetic diversity.

Genome-wide standing variation facilitates long-term response to bidirectional selection for antibody response in chickens.

BACKGROUND: Long-term selection experiments provide a powerful approach to gain empirical insights into adaptation, allowing researchers to uncover the targets of selection and infer their contributions to the mode and tempo of adaptation. Here we implement a pooled genome re-sequencing approach to investigate the consequences of 39 generations of bidirectional selection in White Leghorn chickens on a humoral immune trait: antibody response to sheep red blood cells. RESULTS: We observed wide genome involvement in response to this selection regime. Many genomic regions were highly differentiated resulting from this experimental selection regime, an involvement of up to 20% of the chicken genome (208.8 Mb). While genetic drift has certainly contributed to this, we implement gene ontology, association analysis and population simulations to increase our confidence in candidate selective sweeps. Three strong candidate genes, MHC, SEMA5A and TGFBR2, are also presented. CONCLUSIONS: The extensive genomic changes highlight the polygenic genetic architecture of antibody response in these chicken populations, which are derived from a common founder population, demonstrating the extent of standing immunogenetic variation available at the onset of selection.

Delayed access of low body weight-selected chicks to food at hatch is associated with up-regulated pancreatic glucagon and glucose transporter gene expression.

Chickens selected for low (LWS) and high (HWS) juvenile body weight (BW) for 55 generations differ in BW by 10-fold at selection age. High (HWR) and low (LWR) body weight-relaxed lines have been random-bred since the 46th generation. Our objective was to evaluate the developmental and nutritional regulation of pancreatic mRNA abundance of pancreatic and duodenal homeobox 1 (PDX1), preproinsulin (PPI), preproglucagon (PPG), and glucose transporter 2 (GLUT2). At day of hatch (DOH) and days 1, 3, 7, and 15 (D1, 3, 7 and 15, respectively), pancreas was collected and real time PCR was performed in Experiment 1. In Experiment 2, HWS and LWS were fed or delayed access to food for 72 h post-hatch, and pancreas collected at D15. There was an interaction of line and age for GLUT2 (P=0.001), PPI (P<0.0001), PPG (P=0.034), and PDX1 (P<0.0001). Expression was greater in chicks from LWR and LWS than HWR and HWS. There was an interaction of line and nutrition on PPG (P<0.0001) and GLUT2 (P=0.001) mRNA, where expression was similar among chicks that were fed but greater in LWS than HWS when chicks were delayed access to food. Thus, the first two weeks is important for maturation of pancreatic endocrine function. Long-term selection for BW is associated with differences in pancreas development, and delaying access to food at hatch may have persisting effects on glucose regulatory function.

Housing systems interacting with sex and genetic line affect broiler growth and carcass traits.

Housing systems used in the production of poultry meat vary worldwide dependent on climate, land availability, and other resources essential for production. Reported here are comparisons between pen and cage rearing (the housing system, denoted HS: ), line crosses LC: ), two native Chinese lines (EM males were mated to Y1 and Y2 and their offspring denoted as EMY1 and EMY2), and sex in determining broiler traits. At hatch, 320 males and 320 females from each LC (giving a total of 1,280 chicks) were randomly assigned within each subgroup to 16 battery pens. There were 4 replicates for each combination of LC by sex. On d 28, half of the chicks were transferred to indoor floor pens, and the others were raised in single cages from d 29 to 91. Weekly body weights, livability, and feed conversion ratios ( FCR: ) were obtained to d 91, the age at which the broilers were slaughtered for carcass measurements. The caged males and females were heavier (P < 0.05) than their penned counterparts (2,292 vs 2,219 g). Except for females from line EMY1 (94.9%), the livability for each unit from 1 to 28 d, and 29 to 91 d was greater than 95%. Penned EMY2 broilers had the highest FCR (3.02), whereas penned EMY1 broilers had the lowest FCR (2.96) among the housing systems by LC combinations (P < 0.05). Caged chickens had thicker subcutaneous fat (7.24 mm), a higher percentages of abdominal fat (5.01%) and liver mass (3.13%) , but lower eviscerated carcass (60.63%) and breast muscle weights (pectoralis major and minor, 17.10%). Males were heavier and had higher percentages of leg muscle (boneless drum plus thigh, 24.22%) and heart muscle (1.08%) than the females (P < 0.05). However, the females had thicker subcutaneous fat (7.19 mm) and higher percentages of carcass weight (87.28%), breast muscle (18.11%), abdominal fat (6.54%), and liver mass (3.15%) than males. Penned females had the highest percentage of breast muscle (18.94%), and caged females had the highest percentage of liver mass (3.72%). Females of EMY1 had the highest percentage of breast muscle (18.40%). Generally, the housing system employed and the sex of the broilers greatly affect the carcass traits.

Growth of White Leghorn Chicken Immune Organs after Long-Term Divergent Selection for High or Low Antibody Response to Sheep Red Blood Cells.

Long-term divergent selection from a common founder population for a single trait-antibody response to sheep erythrocytes 5 days post-injection-has resulted in two distinct lines of White Leghorn chickens with a well-documented difference in antibody titers: high (HAS)- and low (LAS)-antibody selected lines. Subpopulations-high (HAR)- and low (LAR)-antibody relaxed-were developed from generation 24 of the selected lines to relax selection. The objective of the current experiment was to determine if this long-term selection and relaxation of selection impacted the growth of two organs important to chicken immunity: the spleen and the bursa of Fabricius. Spleens and bursae were obtained from ten chickens per line at nine timepoints (E18, D0, D6, D13, D20, D35, D49, D63, and D91) throughout their rapid growth phase and presented as a percent of body weight. Significance was set at p ≤ 0.05. For the spleen, all lines consistently increased in size relative to body weight to D49, followed by a consistent decline. All lines had a similar growth pattern, but HAS spleens grew faster than LAS spleens. For the bursa, LAS was smaller than the other three lines as an embryo and also smaller than HAS through D63. In the selected lines, bursa weight peaked at D35, whereas the relaxed lines peaked at D49. By D91, there was no difference between lines. Artificial and natural selection, represented by the long-term selected and relaxed antibody lines, resulted in differences in the growth patterns and relative weights of the spleen and bursa of Fabricius.

Selection for high and low antibody responses to sheep red blood cells influences cytokine and chemokine expression in chicken peripheral blood leukocytes and splenic tissue.

White Leghorn chickens from a common founder population have been divergently selected for high (HAS) or low (LAS) antibody responses to sheep red blood cells (SRBC) for 49 generations resulting in 2 diverse lines for this trait. Much has been studied in these two lines; however, the impact of these selection pressures on cytokine and chemokine expression is not fully understood. The purpose of this study is to determine if selection for antibody response to SRBC impacts cytokine and chemokine expression in peripheral blood leukocytes (PBL) and spleen from HAS and LAS chickens. Total RNA was isolated from PBL and spleen after which mRNA expression of cytokines (IL4, IL6, IL10, TGF-ß4) and chemokines (CXCL8, CCL4) were determined by quantitative real-time RT-PCR (qRT-PCR). The data were analyzed using Student's t test comparing HAS and LAS (P < 0.05) and are reported as corrected 40-CT. PBL and spleen samples were analyzed separately. With respect to PBL, expression of IL6 was higher (P < 0.05) in PBL isolated from LAS chickens compared to those from the HAS line whereas there were no differences (P > 0.05) in IL4, IL10, CXCL8, CCL4, or TGF-ß4. The cytokine and chemokine mRNA expression profiles were different in the spleen between the two lines. IL4 and CXCL8 expression were higher (P < 0.05) in spleen samples from HAS chickens than LAS. The expression of IL6, IL10, CCL4, or TGF-ß4 in the spleens did not differ (P > 0.05) between the lines. The data indicate that selection for specific antibody responses to SRBC impacts the cytokine and chemokine expression profile in PBL and spleens but in different ways in HAS and LAS. These studies provide insight into the influence that selection pressures for antibody responses have on different immune response components, specifically cytokines and chemokines typically involved in the innate response.

Early intestinal development of chickens divergently selected for high or low 8-wk body weight and a commercial broiler.

The early posthatch period is crucial to intestinal development, shaping long-term growth, metabolism, and health of the chick. The objective of this study was to determine the effect of genetic selection on morphological characteristics and gene expression during early intestinal development. Populations of White Plymouth Rocks have been selected for high weight (HWS) and low weight (LWS) for over 63 generations, and some LWS display symptoms of anorexia. Intestinal structure and function of these populations were compared to a commercial broiler Cobb 500 (Cobb) during the perihatch period. Egg weights, yolk-free embryo BW, yolk weights, and jejunal samples from HWS, LWS, and Cobb were collected on embryonic day (e) 17, e19, day of hatch, day (d) 3, d5, and d7 posthatch for histology and gene expression analysis. The RNAscope in-situ hybridization method was used to localize expression of the stem cell marker, olfactomedin 4 (Olfm4). Villus height (VH), crypt depth (CD), and VH/CD were measured from Olfm4 stained images using ImageJ. mRNA abundance for Olfm4, stem cell marker Lgr5, peptide transporter PepT1, goblet cell marker Muc2, marker of proliferation Ki67, and antimicrobial peptide LEAP2 were examined. Two-factor ANOVA was performed for measurements and Turkey's HSD was used for mean separation when appropriate. Cobb were heaviest and LWS the lightest (P < 0.01). at each timepoint. VH increased in Cobb and CD increased in HWS compared to LWS (P < 0.01). PepT1 mRNA was upregulated in LWS (P < 0.01), and Muc2 mRNA was decreased in both HWS and LWS compared to Cobb (P < 0.01). Selection for high or low 8-wk body weight has caused differences in intestinal gene expression and morphology when compared to a commercial broiler.

Tissue cytokines in chickens from lines selected for high or low humoral antibody responses, given supplemental Limosilactobacillus reuteri and challenged with Histomonas meleagridis.

Histomonas meleagridis, a protozoan parasite, induces blackhead disease (histomoniasis) in poultry. During hatching, chicks from lines divergently selected for high (HAS) and low (LAS) antibody responses to sheep red blood cells were divided into two groups, each of HAS and LAS, and placed in pens with wood shavings as litter. Feed and water were allowed ad libitum. Half of the chicks from each line had Limosilactobacillus reuteri (L. reuteri) inoculated to their drinking water. On day 18, all chicks were given a transcloacal inoculation of 100,000 H. meleagridis cells. Then, 10 days later, they were euthanized, followed by collection of tissues from the brain, cecal tonsil, ceca, liver, thymus, and spleen for qPCR analyses of cytokines involved in immunological development. Changes in cytokine expressions were most numerous in the cecal tonsil, ceca, and liver. In the absence of a functional medication for control of histomoniasis, L. reuteri and/or its secretory product, reuterin, might serve, in some genetic populations, as a means to reduce the impact of histomoniasis in chickens. The data demonstrate that L. reuteri treatment had tissue specificity between the two genetic lines, in which the effects were targeted primarily toward the cecal tonsil, ceca, and liver, which are the primary tissue targets of the parasite (H. meleagridis), as well as the thymus and spleen. However, interactions among main effects reflect that responses to inflammatory markers observed in tissues for one genetic line may not be observed in another.

Association of MHCY genotypes in lines of chickens divergently selected for high or low antibody response to sheep red blood cells.

The chicken MHCY region contains members of several gene families including a family of highly polymorphic MHC class I genes that are structurally distinct from their classical class I gene counterparts. Genetic variability at MHCY could impart variability in immune responses, but robust tests for whether or not this occurs have been lacking. Here we defined the MHCY genotypes present in 2 sets of chicken lines selected for high or low antibody response, the Virginia Tech (VT) HAS and LAS, and the Wageningen University (WU) HA and LA lines. Both sets were developed under long-term bidirectional selection for differences in antibody responses following immunization with the experimental antigen sheep red blood cells. Lines in which selection was relaxed (VT HAR and LAR) or lacking (WU C) provided controls. We looked for evidence of association between MHCY genotypes and antibody titers. Chickens were typed for MHCY using a recently developed method based on a multilocus short tandem repeat sequence found across MHCY haplotypes. Five MHCY haplotypes were found segregating in the VT HAS and LAS lines. One haplotype was present only in HAS chickens, and another was present only in LAS chickens with distribution of the remaining 3 haplotypes differing significantly between the lines. In the WU HA and LA lines, there was a similar MHCY asymmetry. The control populations lacked similar asymmetries. These observations support the likelihood of MHCY genetics affecting heritable antibody responses and provide a basis for further investigations into the role of MHCY region genes in guiding immune responses in chickens.