Brief Summary of Minutes of Annual Meeting

NC1170 Business Meeting

January 13, 2019 San Diego NC1170/NRSP8

Attendees: (AR) Douglas Rhoads, (DE) Carl Schmidt, (AR) Adnan Alrubay, (IA) Susan J. Lamont, (WUHS) Yvonne Drechler, (MS) Bindu Nanduri, (AZ) Fiona McCarthy, (DE) Behnam Abasht, (AR) Wayne Kuenzel, (CA) Huaijun Zhou, (VA) Eric Wong, (CSU Fresno) Katy Tarrant, (WI) Roger Sunde, (MN) Kent Reed, (MI) Gale Strasburg.

Agenda approved unanimously

Sue Lamont USDA Administrative Advisor

• New project was approved and implemented for a five-year term Oct 1, 2018 – Sept 30, 2023
• Gale was thanked for his leadership on assembling and submitting the proposal.
• Publication lists annual reports need to report categories, but only report for this project, what is relevant to the project.
• All publications should acknowledge USDA support from the Hatch program; wording: USDA/NIFA funding through Hatch program (program #)

Comments from NIFA Representative: Dr. Matukumalli was not present due to government shut-down

Comments from NRSP8 Representative Huaijun Zhou

• NRSP8 renewed but significant reduction in NRSP8 because has been successful. Need to think about new directions. NRSP8 being sunset?
• Coordination funding -faculty need to be a member of NRSP8 to get travel support etc. for students.
• Support essential small projects especially junior faculty
• New farm bill support $40,000,000 for high throughput phenotyping and genotyping of plants and animals. Still need to get appropriation (money) to actually fund this effort.  Need to think about priorities of species that might be funded for this effort.
• ADOL genetic lines: USDA indicates ADOL will move to Georgia by 2022 but facility not sufficient until 2022, but facility not sufficient to all of the lines. Michigan ADOL location will be shut down.Are there any stations that could adopt one or more lines as there may not be spaces?

Discussion of workshop external speakers

Purpose of Poultry workshop

• Lamont: multistate meeting at PAG. Need to do annual written reporting,
• No requirement for every member to speak every year.
• Invite relevant outside scientists Wes Warren, Wageningen(?).
• Who do we invite to speak?
• Comments fromcommunity:
  • Wayne Kuenzel - liked concept of minisymposium 2,3 speakers per year in such a relevant target (microbiome, gene knockout).
  • Support for Wayne’s idea: Eric Wong
  • Susan Lamont: value of meeting is learning about people outside of group and what they are doing. Should we shorten technical talks? 
  • Vet school poultry also supports idea of bringing in outside talks. There was a general consensus of support for bringing outside speakers.
  • Perhaps 15 minutes as a limit for the members?

Suggestion shorten the members talk to 15 minutes.

• Doug Rhoads: travel funds for outside speakers an issue
• What is the next minisymposium?

Chicken genome improvement:  Fiona -suggestions for improving chicken genome

Wes-sequencing the Trio of parents/offspring

May help with MHC, minichromosome.

There was support for the idea. 

Red Jungle fowl x parent with very small variability

Perhaps back to UCD red jungle fowl line x something as different as possible.

Fiona will contact Wes with discussion of email 

Doug Rhoads moved have meeting on weekend prior to 2020 PAG meted

Unanimously supported.

 Meeting adjourned at 11:55 January 13, 2019

Accomplishments by Objective

This document is a compilation of annual reports submitted by members of NC1170. Below is a summary of achievements, by objective, reported by the NC1170 membership for the calendar year 2018. One hundred and seven peer reviewed publications were reported for this period. This number improved over the previous report year; furthermore the number of the publications with collaborations among NC1170 member institutions is notable. Over $25 Million in funding was reported; this funding is enabling ever greater advances into avian genomics – from investigations of fundamental biology to development of novel tools and databases. The funding also supported training of the next generation of talent – eight Ph.D dissertations or M.S. theses were completed in this project year. The current membership of NC1170 stands at 34 members spread across 23 institutions.

Objective 1: Create and share data and technology to enhance the development and application of genomics and systems biology in poultry.

AZ

AgBase: supporting functional modeling in agricultural organisms.

AgBase (http://www.agbase.msstate.edu) provides resources to facilitate modeling of functional genomics data and structural and functional annotation of agriculturally important animal, plant, microbe and parasite genomes. We provide 2,069,320 Gene Ontology (GO) annotations for 394,599           gene products from more than 50 agriculturally important species and their pathogens. During 2018 AgBase was moved to a new server system which will provide enhanced performance. We also partnered with Phoenix Bioinformatics, a non-profit organization, to implement a subscription-based funding model for AgBase tools and large-scale analyses, and this will go into effect during 2019.

 Standardized Gene Nomenclature.

The Chicken Gene Nomenclature Consortium (CGNC; http://birdgenenames.org/cgnc) is the sole source globally for chicken gene nomenclature. During 2018 we mapped and updated gene names for the new Galgal5 release. This resulted in 817 obsolete records, 2,693 records with duplicated gene symbols and 1,111 records for manual review.

Chickspress – developing a tissue specific compendium of gene expression for chicken gene products.

The Chickspress resource (http://geneatlas.arl.arizona.edu) provides a detailed “atlas” of chicken gene expression, collating experimental information from Red Jungle Fowl and chicken gene expression studies. During 2018 we have completed remapping all expression data to the new Galgal5 annotations and the publication describing this resources have recently been provisionally accepted (pending minor revisions).

 CA

Identification of Regulatory Elements in Livestock Species

The recent international FAANG (Functional Annotation of ANimal Genomes) initiative has stimulated efforts to functionally annotated important livestock species, which will ultimately be leveraged to improve production efficiency, animal welfare, and food safety. As one of the FAANG pilot projects coordinated by UC Davis, we present the current progress in generating and analyzing data from chicken, cattle, and pig. Samples were collected from adipose, cerebellum, cortex, hypothalamus, liver, lung, muscle, and spleen in two male biological replicates from each species, allowing the identification of both universal and tissue-specific functional elements. We aims to perform RNA-seq, DNase-seq/ATAC-seq, and five ChIP-seq assays on eight tissues in three farm animal species. Currently, we have completed data collection for eight tissues in cattle, three tissues in pig, and five tissues in chicken. Three ChromHMM models, one for each species, were trained to predict genome-wide chromatin states specific to each tissue. To facilitate comparison of models between species, all models were created with 14 states. Among the three species, 11 states were consistent between models while the remaining three states in each model were either unique to that species or appear in the model for only two species. The states common to all species were four promoter-associated states, four enhancer states, a repressive state, an insulator state, and a low-signal state with no strong prevalence of any ChIP-seq marks. The predicted states of all transcription start sites (TSS) were compared in liver, lung, and spleen across three species. The number of TSS with active promoter states was generally higher in pig with 49%, 45%, and 46% in liver, lung and spleen, respectively, compared to 38%, 34%, and 37% in chicken and 39%, 40%, and 38% in cattle. Less than 10% of TSS were predicted in repressed and bivalent states in all species, with the majority of the remaining TSS having a low-signal state. A small portion of TSS were predicted in an enhancer state, which is likely due to missing H3K4me3 signal. A more robust comparison between tissues and species could provide novel insights of evolutionary divergence of regulatory elements when all data collection is completed.

 COH

MHC-Y region genomic sequence data and annotation (Goto, Warden, Zhang, Wu, Kang, McPherson, Delany, Stadtmueller, Bjorkman, Shiina, Hosomichi, Inoko, & Miller). MHC-Y does not fit readily within expectations for a region containing MHC genes. It is a second region of polymorphic MHC genes in chickens. The genes have distinctive features. No homologous region has been identified in mammals. MHC-Y is located on the same chromosome as the classical chicken MHC (MHC-B) and CD1 region, but haplotypes at MHC-Y are inherited independently as if located on another chromosome. MHC-Y is enigmatic with its function remaining mostly undefined.

In the past year we completed annotation of the total MHC-Y region sequence determined (830,931 bp). As we reported last year, 649 kbp is the MHC-Y region and contains 115 genes. The remaining 137 kbp contains four rRNA genes and intervening sequences representing the closely adjacent NOR. Soon we will be submitting a manuscript describing MHC-Y in the Red Jungle Fowl (RJF) reference genome. All that remains is completion of an analysis of the distribution of polymorphic residues in the MHC-Y encoded MHC class I-like molecules.

Data on MHC-Y class I polymorphism (Miller, Goto, Zhang, Stadtmueller, Bjorkman). Genetic/structural polymorphism is a major feature distinguishing MHC class I molecules. The binding and presentation of peptide antigens by classical MHC class I molecules is well understood. Polymorphic residues in and around the antigen binding groove, also called “the MHC-fold”, of the highly polymorphic classical MHC class I molecules determine which peptides are held within the groove of different isoforms. Other molecules also bear the MHC-fold but nearly all of these are monomorphic and are described as non-classical. The highly polymorphic class I-like molecules encoded by MHC-Y have properties that are both classical and non-classical, making them unusual and difficult to classify. They display considerable polymorphism (Afanassieff et al., 2001; Hunt et al., 2006; Thoraval et al., 2003), but do not bind peptides (Hee et al., 2010). We are now completing structural analyses defining position and side-chain orientation of the polymorphic residues found in the MHC-Y class I molecules encoded in the RJF haplotype. The distribution of the polymorphic residues shows interesting patterns.

IA

Transcriptome data made public. Several data sets from RNAseq experiments on chickens were deposited in public databases upon submission for journal publication of the manuscripts describing those studies.

Copy number variations. Copy number variations (CNV) are an important source of genetic variation that has gained increasing attention over the last couple of years. In this study, we performed CNV detection and functional analysis for 18,719 individuals from four pure lines and one commercial cross of layer chickens. Samples were genotyped on four single nucleotide polymorphism (SNP) genotyping platforms, i.e. the Illumina 42K, Affymetrix 600K, and two different customized Affymetrix 50K chips. CNV recovered from the Affymetrix chips were identified by using the Axiom® CNV Summary Tools and PennCNV software and those from the Illumina chip were identified by using the cnvPartition in the Genome Studio software. The mean number of CNV per individual varied from 0.50 to 4.87 according to line or cross and size of the SNP genotyping set. The length of the detected CNV across all datasets ranged from 1.2 kb to 3.2 Mb. The number of duplications exceeded the number of deletions for most lines. Between the lines, there were considerable differences in the number of detected CNV and their distribution. Most of the detected CNV had a low frequency, but 19 CNV were identified with a frequency higher than 5% in birds that were genotyped on the 600K panel, with the most common CNV being detected in 734 birds from three lines. Commonly used SNP genotyping platforms can be used to detect segregating CNV in chicken layer lines. The sample sizes for this study enabled a detailed characterization of the CNV landscape within commercially relevant lines. The size of the SNP panel used affected detection efficiency, with more CNV detected per individual on the higher density 600K panel. In spite of the high level of inter-individual diversity and a large number of CNV observed within individuals, we were able to detect 19 frequent CNV, of which, 57.9% overlapped with annotated genes and 89% overlapped with known quantitative trait loci.

MN

Additional RNAseq datasets compiled for the turkey have been accessioned in NCBI’s Gene Expression Omnibus (GEO) repository.

MS

MSU continues to generate data and develop bioinformatics resources to facilitate the analysis of functional genomics data in agricultural species.

MSU contributed to Host-Pathogen Interaction Database (HPIDB): HPIDB provides predicted and curated host-pathogen protein-protein interaction data to support animal health/disease studies. During 2018 HPIDB provides predicted and curated host-pathogen protein-protein interaction data to support animal health/disease studies. During 2018 HPIDB continued to develop a set of predicted interologs (a total of 130,966 http://hpidb.igbb.msstate.edu/hpi30_interologs.html). Working with University of Florida, We identified active kinases in spleen and liver tissues in chicken based on their reactivity with the ATP and ADP desthiobiotin acyl phosphate probes combined with mass spectrometry. We identified 188 chicken kinases and their ATP-binding regions to create a tissue-specific atlas of active kinase expression in chicken. We also determined the possible functions of these kinases by utilizing bioinformatics approach by comparing functional pathways and disease involvement of human, murine and rat orthologs of these kinases. We also performed chemical proteomic profiling of active deubiquitinases (DUBs) by utilizing active-site directed ubiquitin (Ub)-vinyl sulfone (VS)-HA probe labeling combined with western blot-based or mass spectrometric identification of DUBs. By using these techniques we identified 29 DUBs in cecum, liver and spleen tissues. We have identified that DUBs such as USP5, USP4, UCH-6, UNP, USP7, UCH- L5, USP9x, USP10, USP19, USP47, OTUD6B, and USP8 are the top 12 DUBs identified in liver while in spleen there were USP5, USP4, UCH-6, UNP, UCH-L5, UCH-L1, USP9x, USP10, USP19, USP16, OTUD6B, and USP8. Cecum overall contained less active DUBs, and the most active DUBs were USP5, USP4, UCH-6, UNP, USP7, UCH-L5, UCH-L1, USP47, USP8, USP10, USP19 and USP9x.

TX

Structural variants are an important source of phenotypic variation in domestic species, including chicken, standardized databases for comparison across phenotypes are not available at present. Athrey Lab at continued development and testing of new structural variant cataloging tools, Girar.  After initial development in 2017, additional datasets downloaded from NCBI were tested using the approach. Over 400 unique structural variant variants have been identified and are being verified by standard laboratory techniques. One main approach to validate structural variants has been a partnership with Bionano, who are currently running samples on the Saphyr platform to validate structural variants identified and validated using the in-silico methods.

Also towards Objective 1, Athrey lab graduate student Travis Williams continued ongoing work on in-silico approach to validate currently predicted miRNAs in the chicken genome. The results of this work were presented at the Poultry Science Association annual meeting. This work also produced a short pipeline tool that will deposited on Github.

WI

R Sunde

RNAseq datasets for the turkey transcriptome have been collected in collaboration with K. Reed (U Minnesota); better annotation of selenoprotein transcripts for NCBI is being developed.

1. Rosa

Quantitative Genetic Analysis of Reproductive Traits on a Male Line of Turkey

Turkey body weight and growth rate have been successfully improved with selective breeding. However, reproductive traits such as egg production, fertility, hatchability and egg weight have not been actively selected for, especially on male lines. Improving performance for these traits is crucial to further increase production and decreasing cost. In this context, understanding the genetic background of reproductive traits is important so that efficient breeding programs can be developed to meet the demand for turkey meat. The aim of this study was to estimate genetic parameters and perform a genome-wide association analyses for reproductive traits in a male turkey line. Data on 11,467 females was available for total egg production (TEP), fertility (FERT), hatch of fertile eggs (HOF), egg weight (EW), and body weight at 18 weeks of age (BW18). A total of 1,911 hens were genotyped using a 65K single nucleotide polymorphism (SNP) array, from which 49,289 SNP were available after quality control. Variance components, heritabilities, and genetic correlations were estimated using a multi-trait single- step genomic BLUP model (ssGBLUP) via Restricted Maximum Likelihood (REML). In addition, a genome-

wide association study (GWAS) was implemented for TEP. A relatively high heritability was obtained for EW (h2 = 0.52), a moderate value for TEP (h2 = 0.35), and much lower values for FERT and HOF (h2 = 0.10 and h2 = 0.13, respectively). EW was positively correlated with BW18 (0.20 ± 0.03), whereas it was negatively correlated with FERT (-0.04 ± 0.05), HOF (-0.51 ± 0.04) and TEP (-0.12 ± 0.04). FERT was found positively correlated with HOF (0.40 ± 0.08) and TEP (0.28 ± 0.06). HOF was negatively correlated with BW18 (-0.20 ± 0.05) and positively correlated with TEP (0.23 ± 0.06). TEP was negatively correlated with BW18 (-0.17 ± 0.05). Such high negative (and genetically unfavorable) correlation between EW and HOF, as well as the negative correlation between TEP and BW18 should be carefully considered or the development of a multiple trait breeding strategy. The GWAS detected some interesting genomic regions, including an SNP located in GREB1L on chromosome 3 (q-value < 0.05, p-value = 1.10543e-06), which might play a role on TEP. The region located 0.5 Mb upstream and 0.5 Mb downstream of that SNP (1 Mb window including significant SNP) harbors candidate genes related to egg formation (i.e. maturation of follicle/oocyte, ovulation or reproductive track development). The current study aimed to shed some light on the genetic background of reproductive traits in Turkeys. It combined pedigree and genomic information for the estimation of genetic parameters, and presented the first GWAS analysis for total egg production on a male turkey line.

Including Phenotypic Causal Networks in Genome-Wide Association Studies Using Mixed Effects Structural Equation Models

Network based statistical models accounting for putative causal relationships among multiple phenotypes can be used to infer single-nucleotide polymorphism (SNP) effect which transmitting through a given causal path in genome-wide association studies (GWAS). In GWAS with multiple phenotypes, reconstructing underlying causal structures among traits and SNPs using a single statistical framework is essential for understanding the entirety of genotype-phenotype maps. A structural equation model (SEM) can be used for such purposes. We applied SEM to GWAS (SEM-GWAS) in chickens, taking into account putative causal relationships among breast meat (BM), body weight (BW), hen-house production (HHP), and SNPs. We assessed the performance of SEM-GWAS by comparing the model results with those obtained from traditional multi-trait association analyses (MTM-GWAS). Three different putative causal path diagrams were inferred from highest posterior density (HPD) intervals of 0.75, 0.85, and 0.95 using the inductive causation algorithm. A positive path coefficient was estimated for BM→ BW, and negative values were obtained for BM→ HHP and BW→ HHP in all implemented scenarios. Further, the application of SEM-GWAS enabled the decomposition of SNP effects into direct, indirect, and total effects, identifying whether a SNP effect is acting directly or indirectly on a given trait. In contrast, MTM-GWAS only captured overall genetic effects on traits, which is equivalent to combining the direct and indirect SNP effects from SEM-GWAS. Although MTM-GWAS and SEM-GWAS use the similar probabilistic models, we provide evidence that SEM-GWAS captures complex relationships in terms of causal meaning and mediation and delivers a more comprehensive understanding of SNP effects compared to MTM-GWAS. Our results showed that SEM-GWAS provides important insight regarding the mechanism by which identified SNPs control traits by partitioning them into direct, indirect, and total SNP effects.

Predictive ability of genome-assisted statistical models under various forms of gene action

Recent work has suggested that the performance of prediction models for complex traits may depend on the architecture of the target traits. Here we compared several prediction models with respect to their ability of predicting phenotypes under various statistical architectures of gene action: (1) purely additive, (2) additive and dominance, (3) additive, dominance, and two-locus epistasis, and (4) purely epistatic settings. Simulation and a real chicken dataset were used. Fourteen prediction models were compared: BayesA, BayesB, BayesC, Bayesian LASSO, Bayesian ridge regression, elastic net, genomic best linear unbiased prediction, a Gaussian process, LASSO, random forests, reproducing kernel Hilbert spaces regression, ridge regression (best linear unbiased prediction), relevance vector machines, and support vector machines. When the trait was under additive gene action, the parametric prediction models outperformed non-parametric ones. Conversely, when the trait was under epistatic gene action, the non-parametric prediction models provided more accurate predictions. Thus, prediction models must be selected according to the most probably underlying architecture of traits. In the chicken dataset examined, most models had similar prediction performance. Our results corroborate the view that there is no universally best prediction models, and that the development of robust prediction models is an important research objective.

Objective 2: Facilitate the creation and sharing of poultry research populations and the collection and analysis of relevant new phenotypes including those produced by gene transfer

ADOL

ADOL chicken populations: A major strength of ADOL is the large number of chicken lines that are characterized for a number of traits, especially those associated with viral diseases, and maintained under specific pathogen free (SPF) conditions. Besides providing unique genetic resources to ADOL, ~1,500 embryos or chicks are supplied yearly to academic institutions or companies in the United States. The lines and maintenance are briefly summarized below.

ADOL maintains 35 chicken lines with special genetic characteristics for tumor or viral susceptibility that also differ remarkably for immunological and physiological traits. All but 3 (C, N and P) were developed at the ADOL over the last 67 years. These include 4 of the world’s most highly inbred lines (63, 71, 72,and 15I5,), all of which are well defined for avian leukosis virus (ALV) receptor genes, endogenous virus loci (EV), and resistance to MD. Two of the lines are outbred, 2 of which are highly utilized worldwide for ALV analyses (0 and 15B1). Four congenic lines exist for analysis of EV genes; 3 (0.44-TVBS1- EV21, 0.44-TVBS3-EV21, and RFS) were developed from line 0 and 1 (100B) from line 72. Eight congenic lines exist for analysis of the influence of the MHC (B haplotype) on resistance to tumor diseases, immune responses or vaccinal immunity; 7 (15.6-2, 15.7-2, 15.15I-5, 15.C-12, 15.P-13, 15.P-19, and 15.N-21) were developed from line 15I5, and 1 (15.N-21) from line 0. Lines 63 and 72 differ markedly for MD resistance and immune function traits, as well as ALV and EV genes, but have the same B haplotype. Nineteen recombinant congenic strains (RCS) are under development to identify non-MHC genes that influence traits differing between lines 63 and 72. ADOL also developed one transgenic chicken line (0.ALV6) that is very beneficial for analysis of ALV.

ADOL lines are routinely tested by blood-typing using 40 antisera either to ensure purity or to maintain heterozygosity (EV21, 100B, and O.P-13) during annual line reproduction. The breeders are unique in that they are maintained in a quarantined state and, on the basis of frequent serologic tests for 11 pathogens, are considered free of infection from common poultry pathogens.

Estimates for the completion of the Athens, GA facilities is fall 2022. Unfortunately, despite promises that ADOL and all of its resources would remain intact, it is apparent that the buildings to house the ADOL lines are not sufficient. Thus, the number of lines kept will have to be reduced. Still worrisome is that no plans have been established to move the lines, which will require at least 2 years of having lines at both ADOL and Athens.

COH

Evaluating MHC-Y haplotypes segregating in experimental inbred and highly-selected lines (Zhang, Goto, Lamont, Siegel, Honaker, and Miller). PCR-based genotyping (see below) has allowed us to readily define MHC-Y haplotypes in different populations. We are investigating the relationship between MHC-Y haplotype and immune responses in a variety of experimental populations. In the Fayoumi and white leghorn lines maintained at Iowa State University, we found only one MHC-Y haplotype present in each line, consistent with the highly inbred nature of these two lines. In a retrospective analysis of RNASeq data from an earlier challenge trial of these lines with Newcastle disease virus (NDV), we found major differences of MHC-Y class I gene expression in spleen with greater expression in the Fayoumi line. Another recently reported study shows enhanced expression of MHC-Y after NDV challenge in the respiratory tract of the Fayoumi birds (Deist et al., 2018).

In the closed Virginia Tech HAS and LAS lines, we have defined five haplotypes. The HAS and LAS lines, maintained by matings among individuals at each generation, exhibiting high antibody responses in HAS and low antibody responses in LAS (mate-selection is restricted by no sib matings and no sire or dam families over represented). Selection is still ongoing (Lillie et al., 2017). We have MHC-Y genotyped the 44th and 45th generations (nearly 200 birds in each generation). We first defined patterns and then used these guided by pedigree information for the two generations to define haplotypes. Among the five haplotypes, one is found exclusively within the HAS line. Three others are very common in the LAS line. MHC- Y could be guiding immune responses such that MHC-Y haplotypes have become asymmetrically distributed during the multiple rounds of selection for antibody response.

From our experience with typing, we have learned that multiple MHC-Y haplotypes are typically segregating in non-selected lines. We are working on ways to assign MHC-Y haplotypes in these lines in the absence of pedigree data. Typing is ongoing of birds in Campylobacter colonization trials (Roslin and Alberta) and a study of infection-related lameness (Arkansas).

IA

Iowa State University chicken resource populations maintained, but reduced in number. For most of the year, Iowa State University maintained 13 unique chicken research lines [including highly inbred, MHC-congenic, closed populations; and advanced intercross lines (AIL)] to serve as resources for identifying genes, genetic elements and genomic regions of economic importance; as well as defining unique aspects of chicken genomic architecture. All adult breeders were housed in individual cages and matings done by artificial insemination to ensure pedigree accuracy. All MHC-defined lines were blood-typed to verify MHC serologic haplotype. Two AILs (now at generation F28) were maintained to facilitate fine-mapping of QTL with the goal of identifying genomic regions and candidate genes controlling important phenotypes. Because of reduction in space allotted to the genetics program in the new poultry farm (construction to begin in 2019), the number of genetic lines was reduced from 13 to eight lines. The eight remaining lines are: one AIL (broiler X Fayoumi), two inbred Fayoumi lines, two inbred Leghorn lines, one inbred Spanish line, one antique inbred line (inbreeding started in 1925) and one closed population of broilers from 1980s industry genetics.

Utilization and sharing of research populations. The ISU genetic lines formed a discovery platform for research on the genomics of heat resistance in a USDA-AFRI-NIFA project (PD: C Schmidt, U Del) and a USAID project on genomics of resistance to Newcastle disease virus and heat (PD: H Zhou, UC-Davis) because of defined, distinct responses among lines. Genetic material (chicks, fertile eggs, blood, tissues, DNA or RNA) was shared with many cooperating investigators to expand studies on the chicken genome. Active collaborations utilizing ISU chicken genetic lines or biological materials include H Zhou, UC-Davis (NDV and heat-stress response); C. Schmidt, U Delaware (heat stress); R Coulombe, Utah State (aflatoxin sensitivity); B Abasht, U Delaware (allele-specific expression); E Wong, Virginia Tech (Eimeria response) and V Kapur, Penn State (NDV-embryo assays).

TX

Athrey lab is maintaining a Red Junglefowl population of the Richardson strain. The current population size stands at 42, of which 14 are males. We completed an undergraduate project that used microsatellite markers to assess population genetic status, and pedigree information in this captive population, which resulted in a publication. The well documented research population has become key to a new project we started with Prof. Leif Andersson, to look at the developmental biology of morphological traits in chicken. We will be generating F2 crosses between RJF and heritage breeds.

 Also, as the captive RJF are generating more eggs than we can hatch out and place in the colony (due to space limitations), we are able to collect and ship fertile RJF eggs.

Objective 3. Elucidate genetic mechanisms that underlie economic traits and develop new methods to apply that knowledge to poultry breeding practices.

ADOL

Characterizing the genomic landscape of Marek’s disease virus-induced lymphomas

Meq, a bZIP transcription factor and the viral oncogene for pathogenic strains of Marek’s disease virus (MDV), is required to induce CD4 T cell lymphomas that characterize Marek’s disease (MD) in chickens. However, Meq is not sufficient for neoplastic transformation as not all birds infected with pathogenic strains of MDV developed MD. We hypothesize that additional drivers or somatic mutations in the chicken genome are required for MDV-induced transformation. Using and integrating DNA and RNA genomic screens of MD tumors from genetically-defined experimental layers, our analyses reveal 0.3 somatic mutations per megabase consisting primarily of somatic single nucleotide variants (SNVs) and small insertions and deletions (Indels). Somatic deletions, insertions, and point mutations were enriched in IKZF1 (Ikaros), the first driver gene of MD lymphomas. Ikaros, a Zn-finger transcription factor and the master regulator of lymphocyte development, is a known tumor suppressor in human and murine acute leukemias and lymphomas. In our surveyed MD tumors, 41% of the samples had somatic mutations in key N-terminal Zn-finger binding domains, strongly suggesting perturbed Ikaros function in its ability to bind DNA and regulate transcription. Interesting, somatic mutations in IKZF1 were preferentially found in tumors of gonadal tissues as well as their metastatic clones. IKZF1 mutant MD tumors revealed gene expression profiles indicative of Ikaros perturbation. In addition to IKZF1, other putative somatic mutations reside in FLT3, SOX1, VWF, ROBO1, and ROBO2 and warrant future evaluation. Our results suggest MDV-induced tumors are driven by both Meq expression and Ikaros somatic mutations, which in combination lead to unregulated proliferation, increased cell adhesion, increased migration, and dedifferentiation

Contribution of the T cell receptor (TCR) repertoire to Marek’s disease resistance

Marek’s disease (MD) is a commonly diagnosed herpesviral-induced T cell lymphoproliferative disease of chickens, which has increased in virulence over time and prompted the search for continued improvements in control, both through improved vaccines and increased flock genetic resistance. Model pairs of genetically MD-resistant and susceptible chickens that are either B2 MHC-matched or B21 and B19 MHC-congenic has allowed the study of both non-MHC-linked and MHC-linked genetic resistance; here, we have applied these models to characterizing the T cell receptor (TCR) repertoires in MDV infection. Chickens resistant to MD showed higher usage of Vbeta-1 TCRs than susceptible chickens, in both the CD8 and CD4 subsets in the MHC-matched model, and in CD8 subset only in the MHC-congenic model; and Vbeta-1+ CD8 cells expanded during MDV infection. The TCR locus was found to be divergent between MD-resistant and susceptible chickens in the MHC-matched model, with MD-resistant chickens expressing a greater number of Vbeta-1 TCRs and an increased representation of Vbeta-1 CDR1 loops with an aromatic residue at position 45. TCR Vbeta-1 CDR1 usage in resistant x susceptible F1 birds indicated that the most commonly used CDR1 variant was present only in the susceptible line, suggesting that selection for resistance in the MHC-matched model has optimized the TCR repertoire away from dominant recognition of one of the B2 MHC molecules. Finally, TCR downregulation during MDV infection in the MHC-matched model was observed most strongly in the MD-susceptible line, and TCR downregulation due to viral reactivation in a tumor cell line could be demonstrated to be virus-specific and not due to apoptosis induction.

Neurovirulence of a Meq-deleted Marek’s disease virus in early in ovo challenge

Marek’s disease virus is an important pathogen of chickens which causes neuropathic as well as lymphoproliferative disease. Virulent MDV strains include an oncogene, Meq, while avirulent MDV and closely related viruses are nononcogenic and apparently non-neuropathic, thus allowing there use as vaccine strains. Here, we describe a fatal neuropathy of chicks induced by early in ovo injection (prior to 15 days of embryogenesis) of non-oncogenic Meq-deleted MDV, which induces severe bursa and thymic atrophy as well as mild lymphocytic peripheral nerve lesions. We suggest that the previously identified correlation between neuropathogenicity and a single MDV gene (pp14) presents a strong case for an immune-mediated component to MDV neuropathy, and that other vaccine strains may be capable of inducing neuropathic disease in immune- dysregulated birds.

Identifying the molecular basis for MD vaccine synergy

Marek’s disease (MD) vaccines utilize protective synergism, a phenomenon where the protective efficacy of two vaccines in combination is greater than either vaccine when administered alone. A key example is the bivalent MD vaccine of serotype 2 (SB-1) and serotype 3 HVT (FC-126). Despite this widespread usage, the biological mechanism of the synergistic effect has never been elucidated.

We previously found that SB-1 replicated to the highest levels in spleen, bursa, and thymus, respectively, while HVT showed replication only in bursa at 1 dpi and could not be detected at any other time point or tissue type. Looking at later time points, we find that only SB-1 + HVT (bivalent MD vaccine) can control MDV replication compared to the other two MD vaccines administered alone. Currently, we are screening cytokines to determine if unique or enhanced expression of one or more might account for protective synergism.

Hypothesizing that CD8 T cells may play a role in the protective ability of MD vaccines, we injected chicks with anti-CD8 antibody, which reduced CD8 T cell populations by 50-80%. Regardless of the MD vaccine administered, disease incidence was significantly higher in CD8-depleted birds, indicating that CD8 is responsible for at least part of the protection afforded by MD vaccines. The question is whether CD8 T cells are essential for anti-viral or anti-tumor response.

Comparative transcriptome analysis of spontaneous avian leukosis virus (ALV)-like bursal lymphoma and normal bursal tissues to explore genomic basis associated with the tumor- incidence in chicken.

Sporadic ALV-like bursal lymphoma, also known as spontaneous lymphoid leukosis (LL)-like tumors, were identified in primary breeders’ elite, grandparent and parent hens of a commercial broiler breeder in the absence of exogenous ALV infection and a few experimental lines of White Leghorn. The role of ALV-E field isolates in development of spontaneous LL-like tumors and the potential joint impact in conjunction of a Marek’s disease (MD) vaccine (SB-1) on enhancement of the spontaneous LL-like tumor formation were characterized in chickens of an experimental line of White Leghorn. The spontaneous LL-like tumor incidences of 8%, 17%, 14%, and 42% were observed in the negative control, SB-1 inoculated, ALV-E inoculated, and SB-1 plus ALV-E inoculated groups, respectively, under control conditions. The spontaneous tumor and normal bursa tissues were sampled for total RNAs followed by deep RNA sequencing. After bioinformatical analyses, a total of 923 genes was identified with significantly differentiated expression, which are reportedly involved in over 100 gene ontology terms and pathways, including KEGG pathways, NOD-like receptor signaling pathways, and Toll-like receptor signaling pathways.

AR

Identification of genes affecting ascites susceptibility

AR used whole genome resequencing to identify 31 chromosomal regions as candidate QTLs for affecting ascites.  Three of these regions have been further validated, representing the first validated QTLs for ascites. We are now working on genotyping for all 31 regions in multiple lines including commercial products.  We are preparing breeders for marker assisted selection using two regions to assess utility for selection for resistance to ascites.  AR evaluated the developmental, gender and ontological aspects of mitochondrial biogenesis in broilers, primarily focused on skeletal muscle and the cardio-pulmonary system.  Significant differences were found for gender, tissue, age, and ascites susceptibility.  Muscle mitochondrial content was positively correlated with ascites phenotype.  This may be an easily evaluated trait for selection against ascites. 

Bacterial chondronecrosis with osteomyelitis and lameness in broilers

AR performed one trial for reducing bacterial chondronecrosis with osteomyelitis (BCO) lameness with a commercial organic supplement.  We determined the specific treatment significantly reduced BCO lameness.  The effects of the supplement on specific immune functions were determined.  AR has been evaluating bacterial isolates from BCO lesions to understand the virulence determinants specific to colonization and pathogenesis in broilers. Virulence has been assessed in direct challenge to live broilers as well as phagocytosis assays using chicken macrophage in culture.  Macrophage response may be a simple test to select for resistance to bacterial colonization. We have also been evaluating embryo lethality assays to assess pathogenicity.  We have completed two rounds of directed genome evolution to identify particular pathogenicity determinants in BCO isolates.

Broiler breast muscle and myopathies 

AR performed analyses for differential abundance of microRNA (miRNA) in bigger breast muscle of modern pedigree male broilers and unselected counterparts. A total of 994 miRNA were identified in breast muscle tissues from small RNA sequencing method. Eight higher abundant miRNAs (miR-2131-5p, miR-221-5p, miR-126-3p, miR-146b-5p, miR-10a-5p, let-7b, miR-125b-5p, and miR-146c-5p) and a lower abundant miRNA (miR-206) were identified in modern broiler breast muscles compared to unselected counterparts. Results were integrated with differentially expressed (DE) mRNA in the same tissues and 118 down-regulated mRNAs may be targeted by the up-regulated miRNAs, while 35 up-regulated mRNAs appear to be due to a down-regulated miRNA (i.e., miR-206). Functional network analyses of target genes of DE miRNAs showed their involvement in calcium signaling, axonal guidance signaling, and NRF2-mediated oxidative stress response pathways suggesting their involvement in breast muscle growth in chickens.

Genes and proteins involved in the stress response of broilers

AR- in collaboration with the University of Missouri has completed studies investigating the function of a newly discovered structure in the broiler brain that is involved in stress. The structure, the nucleus of the hippocampal commissure (NHpC), contains a significant population of corticotropin-releasing hormone (CRH) neurons. Using food deprivation (FD) as a gradual stressor that was imposed on male broiler chicks beginning at two weeks of age, chicks were sampled every hour for the first  4 h and at 8 h.  Brain, anterior pituitary gland (PIT) and a blood sample was taken for each time period. Within the brain two structures were dissected, the NHpC as well as the paraventricular nucleus (PVN), the latter a main hypothalamic nucleus involved in the stress response.  Gene expression for CRH neurohormone and its major receptors of CRH, CRHR1 and CRHr2, were determined for each sampling period and compared between the NHpC and PVN. Additionally, gene expression of corticotropes in the PIT that expressed the pre-prohormone proopiomelanocortin (POMC) was determined.  Plasma corticosterone (CORT) was determined by radioimmunoassay.  The NHpC showed the first peak of CRH mRNA within the first 2h of FD.  In contrast, CRH neurons in the PVN showed peak gene expression for CRH at 8h following FD.  CRHR1, the major receptor of CRH neurons within the NhpC showed a negative relationship with CRH mRNA.  In contrast, CRHR1, in the PVN, showed a positive relationship. In other words, as CRH mRNA increased, CRHR1 also increased in the PVN throughout the 8 h of sampling.  Both CRH neurons in the NHpC and PVN were responsible for the significant increase in pituitary POMC mRNA as well as the first detectable plasma increase in the stress hormone CORT.  To date, data support the hypothesis that the NHpC appears to be part of the neuroendocrine system working with the hypothalamic PVN that activates the anterior pituitary and ultimately the adrenal gland to produce the stress hormone CORT during the stress response. 

CA

Improving food security in Africa by enhancing resistance to Newcastle disease virus and heat stress in chickens
Within a USAID funded Feed the Future Innovation Lab for Genomics to Improve Poultry project (H. Zhou, PI) through a partnership of the University of California at Davis (H. Zhou, D. Bunn, R. Gallardo, T. Kelly), Iowa State University (S.J. Lamont, J. Dekkers), Sokoine University of Agriculture (SUA) -Tanzania, the University of Ghana (UOG), and the University of Delaware (C. Schmidt). The five-year research program applied advanced genetics and genomics approaches to sustainably enhance innate resistance to Newcastle disease virus (NDV) and heat stress in chickens to improve poultry production in Africa. We are investigating two stressors (biotic: NDV and abiotic: heat stress). Birds of two genetically distinct and highly inbred lines (Fayoumi and Leghorn), and Hy-Line Brown were either exposed to NDV only (Iowa State) or NDV and heat stress (UCD). Measures of body temperature, blood gas parameters, NDV titers from tears, and antibody response in serum were taken on the live birds, and tissues were collected for transcriptome analysis. Three ecotypes each in Ghana and Tanzania were exposed to La sota NDV and natural exposure to velogenic NDV. DNA isolated from Hy-Line Brown and African ecotypes were genotyped using chicken 600K SNP for GW AS.

At UCD, three manuscripts are prepared on transcriptome analysis in three tissues: Harderian gland, lung and trachea. Harderian gland transcriptome analysis revealed that a limited early response in both Fayoumi and Leghorn lines occurred at 2 dpi. Leghorns eventually had a substantially stronger response at 6 dpi, while Fayoumi had a robust response at the later stages of infection under heat stress. In addition, very few differentially expressed genes overlapped between the lines at each time point suggesting a distinct, line-specific host response to NDV. Especially, GP6 signaling pathway plays significant role in the disease resistance in chickens.

Harderian gland manuscript is published. The other two manuscripts are expected to be submitted in the next few months.

Completed genome-wide association analysis for challenge experiments on Hy-Line Brown: All three replicate trials with over 1,100 challenged birds were completed at UCD and ISU. NDV titers were measured in tears collected at 2 and 6 days post infection. In addition, NDV serum antibody response was measured at 10 days post-infection. The major findings were that the 600K SNP panel and GWAS identified few significant regions associated with the measured phenotypes and these were generally of low genetic effect, emphasizing the highly polygenic nature of the NDV response traits. Additional regions of marginal significance were supported by independent studies that demonstrated differential expression in resistant- susceptible line contrasts of genes that were in the QTL regions. One manuscript reporting these results has been submitted for journal review and another one (ISU) has been published.

For African ecotype NDV challenges, replicate trials involving a total of 2,653 chicks (UOG) and 1789 chicks (SUA) were completed in the challenge facilities. For each replicate, blood was collected for DNA isolation. At four weeks of age, the chicks were challenged with NDV. Tear samples were collected at 2 and 6 dpi for NDV titers. Serum samples were also collected pre-challenge and at 10 dpi for NDV antibody titers. Body weight data was also collected at hatch, weekly until challenge, and at 6 and 10 dpi.

To utilize replicates of previously challenged birds by La Sota NDV strain with available 600K SNP genotype data, the NDV resistance of indigenous African chickens in Tanzania and Ghana undergo further evaluation under field conditions. A RT-qPCR assay that specifically detects mesogenic and velogenic NDV strains was established and validated at SUA. The assay is being used to confirm natural exposure to velogenic NDV. Following natural NDV exposure, data on survival times, body weight, antibody response, and pathological lesion scores were collected. Data analyses are underway.

Serum isolated from the blood samples of 800 birds (~800 birds) have been analyzed to determine antibody levels at pre-exposure and at two time points during the natural exposure trial. ND viral RNA isolation from tear samples and determination of virus titers by RT-qPCR is ongoing. We have currently completed 1436 birds at 2 days post-infection (dpi) and 1301 birds at 6 dpi at UoG and a total of 2,954 birds at SUA.

Using SNP genotypes, birds from each ecotype were reassigned to the emergent population structure, which was then used for estimation of genetic parameters and GWAS. Estimates of heritabilities were moderate to high (0.20 to 0.55) for all traits measured, including pre- and post-infection growth rate, antibody response, and viral levels. Viral levels were, however, not yet available for all birds. Estimates of genetic correlations between antibody level and viral levels were moderately positive but with large standard errors because of the yet small data size for viral levels. Genome-wide Association Studies were completed on the available data using both single-SNP analyses and multi-SNP Bayesian variable selection analyses. Multiple genomic regions with suggestive significance on traits were identified but no QTL with major effects were identified. Overall, response to infection was found to be highly polygenic, with many QTL with small effects but that add up to a substantial genetic component that can be capitalized on using phenotypic or genomic selection with SNPs across the genome. These analyses will be finalized and the identified regions will be explored for underlying genes once the full data is available.

COH

PCR-based method for MHC-Y genotyping (Miller, Goto, Zhang, Fulton, Psifidi, Stevens). We are working to further improve MHC-Y typing methods so that they can be even more easily used in typing large populations of birds. Initially, MHC-Y haplotypes were defined through the patterns revealed in Southern hybridizations. Southern hybridizations are especially time-consuming and not well-suited for typing large numbers of birds. Currently, we are using a PCR-fragment typing method based on microsatellite sequences that are present upstream of the MHC-Y class I genes. These microsatellite sequences vary in repeat number between loci and among haplotypes. Fragments produced from PCR amplification across this microsatellite sequences make it possible to distinguish MHC-Y genotypes on the basis of patterns revealed in ethidium bromide-stained agarose gels. Over the past year, we have improved this typing method by adopting a hot-start polymerase suitable for GC-rich DNA, improving resolution through optimizing gel conditions, recording images with a higher-resolution camera, and routinely typing with two similar but not identical primer pairs. We are working on additional methods that may further reduce the challenge of MHC-Y typing in large populations of chickens.

CU

The liver is particularly important in birds because yolk is produced in the liver. Our preliminary data indicate that full fed broiler breeders develop a fatty liver with an associated decreased egg production. Generally, when an animal is in good health, the somatotropic axis is coupled and there is an abundant population of growth hormone receptor (GHR) in the liver as well as high circulating growth hormone (GH) and insulin-like growth factor (IGF1). We hypothesize that excessive feed intake uncouples the GH/IGF1 axis and disrupts the usual relation between feed intake and ovarian function. Follicle development in the laying hen is a highly efficient and regulated process. Maintenance of a well ordered follicular hierarchy is essential for optimum follicle selection and subsequent egg production in hens. We have previously found that IGF1 is significantly elevated in full fed (FF) broiler breeder hens as compared to restricted fed (RF) broiler breeder hens. Associated with this is excessive and disorganized development of follicles. The relationship between the metabolic and reproductive axis in broiler breeder hens was examined by investigating the in vitro response of granulosa cells to treatment with IGF1. IGF1 does not affect granulosa mRNA expression of FSHR and AMH, factors associated with follicle selection. FSH decreases AMH mRNA expression in cultured granulosa cells from 3-5 mm follicles. IGF1R mRNA is present in the oocyte, a proposed site of IGF1 action. IGF1R mRNA is higher in ooplasm of 3 mm follicles as compared to granulosa cells of 3 mm follicles and to ooplasm and granulosa cells of 5 mm follicles. IGF1 treatment of whole follicles does not alter E2 secretion, although these follicles maintain responsiveness to LH. Preliminary results show no significant effect of IGF1 on mRNA expression of BMP15 or FSHR. In conclusion, localization of IGF1R in the oocyte suggests that IGF1 may have important effects in the oocyte to promote follicle development.

One Ph.D student has initiated her program while working on this project. An undergraduate student was supported by this project during the academic year.

DE

Transcriptomic Analysis of Pectoral Muscles in Broiler Chickens during the Early Phase of Wooden Breast Disorder.

Wooden Breast (WB) is a muscle/meat quality disorder in modern broiler chickens that is clinically distinguished by abnormally firm consistency of the pectoral muscles. To characterize the transcriptome associated with the early pathogenesis of WB in commercial broiler chickens, a time-series study was conducted on the Pectoralis (P.) major muscles between affected and unaffected chickens from a purebred broiler line. To accomplish this objective, chickens were raised for up to 7 weeks of age with muscle biopsy samples from the cranial or the caudal aspect of the P. major muscle belly harvested at week 2, 3 and 4 time points. Cranial P. major muscle specimens from the 3rd week of age are the focus of the present study as this time point precedes the onset of clinically detectable gross lesions at approximately 4 weeks of age. Biopsy samples, including 4 unaffected (U) and 11 affected (A) samples, were processed for RNA-sequencing producing 618 differentially expressed (DE) genes at fold-change (A/U or U/A) >1.3 and False Discovery Ratio (FDR)

Blood Analysis and Proportional Muscle and Organ Weights in Broilers with Wooden Breast.

Wooden Breast is a myopathy of fast growing, commercial broilers causing myofiber necrosis, vasculitis (phlebitis), myoregeneration, and fibrosis with extensive fibrillar collagen deposition in the superficial part of the pectoralis major, presenting clinically as palpably firm breast muscle. Rapid growth, high feed efficiency, and large breast muscle yield are predisposing factors, although the etiology of the disease is still poorly understood. A group of 103 7-week-old Cobb 500 broilers were used to determine the effect of Wooden Breast on 13 blood parameters and the relative weights of the pectoralis major muscle, pectoralis minor muscle, external oblique muscle, wing, heart, lungs, liver, and spleen. Blood analysis performed with the i- STAT handheld blood analyzer on samples taken from the brachial vein of live birds revealed significant differences in blood gases between affected and unaffected chickens, with affected chickens exhibiting higher partial pressure of CO2, total CO2, bicarbonate, and base excess, and lower partial pressure of O2, oxygen saturation, and pH. Affected chickens also possessed a significantly larger pectoralis major muscle and wing relative to body weight. Hypercapnia and hypoxemia in affected chickens suggest greater metabolic demand and insufficient gas exchange, potentially caused by disturbances in circulation, cardiac output, or respiration. Disproportionately slower growth in the external oblique muscle may indicate inadequate development of respiratory muscles relative to the larger breast muscle size. These results support tissue hypoxia and the buildup of metabolic wastes as major contributors to Wooden Breast development and give a more systemic view of Wooden Breast pathology.

The metabolic basis of susceptibility to Wooden Breast Disease in chickens with high feed efficiency.

This study was conducted to characterize metabolic differences between high feed efficiency (HFE) and low feed efficiency (LFE) chickens to investigate why feed efficient chickens are more susceptible to muscle abnormalities such as Wooden Breast Disease. Gene expression profiles were generated by RNA-sequencing of pectoralis major muscle samples from 10 HFE and 13 LFE broiler chickens selected from a modern broiler population. Metabolism-associated differentially expressed genes were identified and interpreted by Ingenuity Pathway Analysis and literature mining. Our RNA-seq data indicates decreased glycolytic capacity, increased fatty acid uptake, mitochondrial oxidation of fatty acids and several other metabolic alterations in the pectoralis major muscle of HFE chickens. We also quantified glycogen content of the pectoralis major muscle and found that the HFE chickens had a significantly (P ≤ 0.05) lower glycogen content. Collectively, this study indicates extensive metabolic differences in the pectoralis major muscle between HFE and LFE chickens and helps identify metabolic features of susceptibility to muscle disorders in modern broiler chickens.

 GA

Molecular and cellular mechanisms that underlie genes and antioxidant enzyme activities during heat stress

Heat stress causes critical molecular dysfunction and cellular changes that affect productivity and potentially compromises bird's welfare. Global temperatures have increased in the past few decades, and climate change will lead to frequent heat waves and longer hot seasons. Molecular mechanisms that underlie nutrient partitioning and metabolism of poultry under heat stress would allow for strategies to mitigate the effects of heat stress. We investigated the immediate and long term transcriptomics changes in chickens under heat stress. Forty- eight Cobb500 male birds were divided into two groups and raised under either constant 25oC or 35oC from 14-26 days of age in individual cages and fed ad libitum on a diet containing 21% CP and 3100kcal ME/kg. Five birds per treatment at 1 and 12 days after heat treatment were euthanized and the liver was sampled for gene expression analysis. We evaluated effect of heat stress (HS) on the expression of select genes in the oxidation/antioxidation pathway in the liver of chickens and further assess changes in antioxidant enzyme activity and biomarkers for oxidative stress in the liver and the Pectoralis (P.) major muscle.

mRNA expression of Nrf2, oxidants NADPH(NOX): [NOX1, NOX2, NOX3, NOX4, NOX5 and DUOX2], and antioxidants [SOD1, CAT, GR, GPx1, NQO1] in the liver were analyzed at 1 and 12 days post-HS. We show that, HS changes the mRNA expression of oxidants thereby increasing cellular reactive oxygen species (ROS). Additionally, persistent HS up- regulates SOD which converts superoxides to hydrogen peroxide. We further demonstrated the dynamic relationship between catalase, GSH peroxidase (GPx) and NADPH under both acute and chronic heat stress. The pentose phosphate pathway could be important under HS since it generates NADPH which serves as a cofactor for GPx. Also, methionine, a precursor of cysteine has been shown to have reducing properties and thereby makes for an alternative fuel for redox processes. Genes in the ROS and antioxidant generation pathways may provide insight into nutritional intervention strategies, especially the use of methionine and/or cysteine when birds are suffering from heat stress. Heat stress was also associated with increased lipid and protein oxidations in the P. major. Molecular and cellular changes in the oxidation/antioxidation pathway may provide insight into interventional strategies.

Effect of Heat stress on Eimeria replication in broiler chickens

Eimeria infection is one of the most important diseases affecting poultry production, and is characterized by bloody or watery diarrhea, weight loss, poor feed conversion and moderate to high mortality. Heat stress (HS) is among the major environmental stressors in poultry, predisposing broiler chickens to immunosuppression and rendering them susceptible to diseases. There are some suggestions that HS reduces Eimeria oocyst output in chickens, however, the relationship between HS and coccidiosis is not elucidated. Our objective was to investigate the effect of HS on the development of E. tenella. Fifty-four 21 day old Cobb500 broiler chickens were infected via gavage with 15x104 E. tenella sporulated oocysts suspended in water and raised in either a thermoneutral (control: 25oC) or a heat-stressed (treatment: 35oC) environment. At 6-days post-infection (dpi), 9 birds in each group were euthanized humanely, and the caecal lesion scores, merozoite and oocyst counts were evaluated. The rectal temperatures were also taken. The HS group had significantly higher cloacal temperature (43.03±0.45oC versus 40.72±0.40; P<0.001) as compared to the control group. At 6 dpi, merozoites, caeca lesion scores and oocyst counts were evaluated in both groups. The HS chickens had lower caeca lesion scores (0.33±0.16 versus 1.89±0.45; P=0.014), merozoite (26.67±24.26 versus 823.21±262.31; P=0.0002) and oocyst counts (80.40±24.36 versus 1802±266.34; P=0.008) as compared to control chickens. Overall our results indicate an interruption of the cycle of E. tenella in chickens housed under heat stress conditions.

High density marker panels, SNPs prioritizing and accuracy of genomic selection

The availability of high-density (HD) marker panels, genome wide variants and sequence data creates an unprecedented opportunity to dissect the genetic basis of complex traits, enhance genomic selection (GS) and identify causal variants of traits of economic importance. The disproportional increase in the number of parameters in the genetic association model compared to the number of phenotypes has led to further deterioration in statistical power and an increase in co-linearity and false positive rates. High density panels do not significantly improve GS accuracy and, in some instances reduce accuracy. This is true for both regression and variance component approaches. To remedy this situation, some form of SNP filtering or external information is needed. Current methods for prioritizing SNP markers (i.e. BayesB, BayesC, etc) are sensitive to the increased co-linearity in HD panels which could limit their performance. Knowledge of genetic diversity based on evolutionary forces is beneficial for tracking loci influenced by selection. The fixation index (FST), as a measure of allele frequency variation among sub- populations, provides a tool to reveal genomic regions under selection pressure. The utility of FST, a measure of allele frequency variation among populations, as an external source of information in GS was evaluated.

A simulation was carried out for a trait with heritability of 0.4. Data was divided into three subpopulations based on phenotype distribution (bottom 5%, middle 90%, top 5%). Marker data were simulated to mimic a 770 K and 1.5 million SNP marker panel. A ten- chromosome genome with 200 K and 400 K SNPs was simulated. Several scenarios with varying distributions for the quantitative trait loci (QTL) effects were simulated. Using all 200 K markers and no filtering, the accuracy of genomic prediction was 0.77. When marker effects were simulated from a gamma distribution, SNPs pre-selected based on the 99.5, 99.0 and 97.5% quantile of the FST score distribution resulted in an accuracy of 0.725, 0.797, and 0.853, respectively. Similar results were observed under other simulation scenarios.

The accuracy obtained using all SNPs can be easily achieved using only 0.5 to 1% of all markers. These results indicate that SNP filtering using already available external information could increase the accuracy of GS. This is especially important as next-generation sequencing technology becomes more affordable and accessible to poultry applications.

IA

Liver and breast muscle transcriptome of laying hens altered by exposure to high ambient temperature.

To maintain productivity, chickens undergo changes in gene expression to maintain metabolic homeostasis during heat exposure. Although previously described for broiler chickens, little is known of the effects of heat on the transcriptome in mature laying hens. We profiled the transcriptome of breast muscle and liver, two major metabolic tissues in poultry, during a 4-week cyclic heating study performed on layers in egg production. Both treatment versus control contrasts and time-based contrasts were analyzed to determine di