Genetic analysis of the blood transcriptome of young healthy pigs to improve disease resilience

BACKGROUND

Disease resilience is the ability of an animal to maintain productive performance under disease conditions and is an important selection target. In pig breeding programs, disease resilience must be evaluated on selection candidates without exposing them to disease. To identify potential genetic indicators for disease resilience that can be measured on selection candidates, we focused on the blood transcriptome of 1594 young healthy pigs with subsequent records on disease resilience. Transcriptome data were obtained by 3'mRNA sequencing and genotype data were from a 650 K genotyping array.

RESULTS

Heritabilities of the expression of 16,545 genes were estimated, of which 5665 genes showed significant estimates of heritability (p < 0.05), ranging from 0.05 to 0.90, with or without accounting for white blood cell composition. Genes with heritable expression levels were spread across chromosomes, but were enriched in the swine leukocyte antigen region (average estimate > 0.2). The correlation of heritability estimates with the corresponding estimates obtained for genes expressed in human blood was weak but a sizable number of genes with heritable expression levels overlapped. Genes with heritable expression levels were significantly enriched for biological processes such as cell activation, immune system process, stress response, and leukocyte activation, and were involved in various disease annotations such as RNA virus infection, including SARS-Cov2, as well as liver disease, and inflammation. To estimate genetic correlations with disease resilience, 3205 genotyped pigs, including the 1594 pigs with transcriptome data, were evaluated for disease resilience following their exposure to a natural polymicrobial disease challenge. Significant genetic correlations (p < 0.05) were observed with all resilience phenotypes, although few exceeded expected false discovery rates. Enrichment analysis of genes ranked by estimates of genetic correlations with resilience phenotypes revealed significance for biological processes such as regulation of cytokines, including interleukins and interferons, and chaperone mediated protein folding.

CONCLUSIONS

These results suggest that expression levels in the blood of young healthy pigs for genes in biological pathways related to immunity and endoplasmic reticulum stress have potential to be used as genetic indicator traits to select for disease resilience.

Exploration of plasma metabolite levels in healthy nursery pigs in response to environmental enrichment and disease resilience

The purpose of this study was to explore plasma metabolite levels in young healthy pigs and their potential association with disease resilience and estimate genetic and phenotypic correlation with the change in lymphocyte concentration following disease challenge. Plasma samples were collected from 968 healthy nursery pigs over 15 batches at an average of 28 ± 3.23 d of age. Forty-four metabolites were identified and quantified by nuclear magnetic resonance. Pigs were then introduced into a natural disease challenge barn, and were classified into four groups based on the growth rate of each animal in the grow-to-finish phase (GFGR) and treatment rate (TR): resilient (RES), average (MID), susceptible (SUS), and dead (pigs that died before harvest). Blood samples were collected from all pigs before and 2 wk after disease challenge and complete blood count was determined. Environmental enrichment (inedible point source objects) was provided for half of the pigs in seven batches (N = 205) to evaluate its impact on resilience and metabolite concentrations. Concentration of all metabolites was affected by batch, while entry age affected the concentration of 16 metabolites. The concentration of creatinine was significantly lower for pigs classified as "dead" and "susceptible" when compared to "average" (P < 0.05). Pigs that received enrichment had significantly lower concentrations of six metabolites compared with pigs that did not receive enrichment (P ≤ 0.05). Both, group classification and enrichment affected metabolites that are involved in the same pathways of valine, leucine, and isoleucine biosynthesis and degradation. Resilient pigs had higher increase in lymphocyte concentration after disease challenge. The concentration of plasma l-α-aminobutyric acid was significantly negatively genetically correlated with the change in lymphocyte concentration following challenge. In conclusion, creatinine concentration in healthy nursery pigs was lower in pigs classified as susceptible or dead after disease challenge, whilst l-α-aminobutyric may be a genetic biomarker of lymphocyte response after pathogen exposure, and both deserve further investigation. Batch, entry age, and environmental enrichment were important factors affecting the concentration of metabolites and should be taken into consideration in future studies.

Plasma protein levels of young healthy pigs as indicators of disease resilience

Selection for disease resilience, which refers to the ability of an animal to maintain performance when exposed to disease, can reduce the impact of infectious diseases. However, direct selection for disease resilience is challenging because nucleus herds must maintain a high health status. A possible solution is indirect selection of indicators of disease resilience. To search for such indicators, we conducted phenotypic and genetic quantitative analyses of the abundances of 377 proteins in plasma samples from 912 young and visually healthy pigs and their relationships with performance and subsequent disease resilience after natural exposure to a polymicrobial disease challenge. Abundances of 100 proteins were significantly heritable (false discovery rate (FDR) <0.10). The abundance of some proteins was or tended to be genetically correlated (rg) with disease resilience, including complement system proteins (rg = -0.24, FDR = 0.001) and IgG heavy chain proteins (rg = -0.68, FDR = 0.22). Gene set enrichment analyses (FDR < 0.2) based on phenotypic and genetic associations of protein abundances with subsequent disease resilience revealed many pathways related to the immune system that were unfavorably associated with subsequent disease resilience, especially the innate immune system. It was not possible to determine whether the observed levels of these proteins reflected baseline levels in these young and visually healthy pigs or were the result of a response to environmental disturbances that the pigs were exposed to before sample collection. Nevertheless, results show that, under these conditions, the abundance of proteins in some immune-related pathways can be used as phenotypic and genetic predictors of disease resilience and have the potential for use in pig breeding and management.

PSXIII-B-14 Investigation of the Blood Transcriptome of Young Healthy Pigs to Identify Genetic Indicators for Disease Resilience

The objective of this study was to investigate the genetic control of the blood transcriptome of young healthy pigs (~27 days of age) to identify genetic indicators for disease resilience. We estimated the heritability of the expression of 16,545 genes and their genetic correlations with 26 measures of performance and resilience before and after exposure to a natural polymicrobial disease challenge. Weaned barrows (n=3,205, Yorkshire*Landrace, 50 batches) from healthy multiplier farms were evaluated for disease resilience in an experimental facility consisting of a high-health quarantine nursery and a challenge nursery and finisher. All pigs were genotyped with a 650k array. Blood samples collected on 1,591 pigs in the quarantine nursery were used for 3’mRNA sequencing and analysis of white blood cell (WBC) counts. Heritability of gene expression was estimated using mixed linear models with (WI) or without (WO) accounting for WBC. The number of genes with significantly heritable expression levels (p&lt; 0.05) was similar for the WI (4,994) and WO models (5,515). Genes with heritable expression levels were significantly enriched for biological processes such as cell activation, immune system process, stress response, and leukocyte activation (q&lt; 1.0×10-7). One genomic region with heritable expression levels, based on average heritability estimates of genes in windows of 0.5Mb, overlapped with the major histocompatibility complex. Significant genetic correlations (p&lt; 0.05) were observed with all recorded phenotypes but not beyond expected false discovery rates (FDR). However, enrichment analysis of genes ranked by estimates of genetic correlations with recorded phenotypes revealed 7 significant GO biological processes (FDR&lt; 0.05), of which 5 were related to innate and/or adaptive immunity. These results suggest that expression levels in blood of young healthy pigs for genes in specific biological pathways have potential as indicator traits to select for disease resilience. Funding from USDA-NIFA #2017-67007-26144, Genome Canada, Genome Alberta, and PigGen Canada.

Genome-wide association study of disease resilience traits from a natural polymicrobial disease challenge model in pigs identifies the importance of the major histocompatibility complex region

Infectious diseases cause tremendous financial losses in the pork industry, emphasizing the importance of disease resilience, which is the ability of an animal to maintain performance under disease. Previously, a natural polymicrobial disease challenge model was established, in which pigs were challenged in the late nursery phase by multiple pathogens to maximize expression of genetic differences in disease resilience. Genetic analysis found that performance traits in this model, including growth rate, feed and water intake, and carcass traits, as well as clinical disease phenotypes, were heritable and could be selected for to increase disease resilience of pigs. The objectives of the current study were to identify genomic regions that are associated with disease resilience in this model, using genome-wide association studies and fine-mapping methods, and to use gene set enrichment analyses to determine whether genomic regions associated with disease resilience are enriched for previously published quantitative trait loci, functional pathways, and differentially expressed genes subject to physiological states. Multiple quantitative trait loci were detected for all recorded performance and clinical disease traits. The major histocompatibility complex region was found to explain substantial genetic variance for multiple traits, including for growth rate in the late nursery (12.8%) and finisher (2.7%), for several clinical disease traits (up to 2.7%), and for several feeding and drinking traits (up to 4%). Further fine mapping identified 4 quantitative trait loci in the major histocompatibility complex region for growth rate in the late nursery that spanned the subregions for class I, II, and III, with 1 single-nucleotide polymorphism in the major histocompatibility complex class I subregion capturing the largest effects, explaining 0.8–27.1% of genetic variance for growth rate and for multiple clinical disease traits. This single-nucleotide polymorphism was located in the enhancer of TRIM39 gene, which is involved in innate immune response. The major histocompatibility complex region was pleiotropic for growth rate in the late nursery and finisher, and for treatment and mortality rates. Growth rate in the late nursery showed strong negative genetic correlations in the major histocompatibility complex region with treatment or mortality rates (−0.62 to −0.85) and a strong positive genetic correlation with growth rate in the finisher (0.79). Gene set enrichment analyses found genomic regions associated with resilience phenotypes to be enriched for previously identified disease susceptibility and immune capacity quantitative trait loci, for genes that were differentially expressed following bacterial or virus infection and immune response, and for gene ontology terms related to immune and inflammatory response. In conclusion, the major histocompatibility complex and other quantitative trait loci that harbor immune-related genes were identified to be associated with disease resilience traits in a large-scale natural polymicrobial disease challenge. The major histocompatibility complex region was pleiotropic for growth rate under challenge and for clinical disease traits. Four quantitative trait loci were identified across the class I, II, and III subregions of the major histocompatibility complex for nursery growth rate under challenge, with 1 single-nucleotide polymorphism in the major histocompatibility complex class I subregion capturing the largest effects. The major histocompatibility complex and other quantitative trait loci identified play an important role in host response to infectious diseases and can be incorporated in selection to improve disease resilience, in particular the identified single-nucleotide polymorphism in the major histocompatibility complex class I subregion.

Genetic analysis of disease resilience in wean-to-finish pigs from a natural disease challenge model

The objective was to estimate the genetic parameters of performance and resilience of growing pigs under disease. Data were from 3,139 Yorkshire × Landrace wean-to-finish pigs that were exposed to a natural polymicrobial disease challenge that was established by entering naturally infected animals into a nursery barn, targeting various viral and bacterial diseases. The challenge was maintained by entering batches of 60 or 75 healthy nursery pigs every 3 wk in a continuous flow system. Traits analyzed included average daily gain (ADG), feed intake (ADFI) and duration (ADFD); feed conversion ratio (FCR); residual feed intake (RFI); mortality (MOR); number of health treatments (TRT); health scores (HScore); carcass weight (CWT), back fat (CBF) and loin depth (CLD); dressing percentage (DRS); lean yield (LYLD); day-to-day variation in feed intake and duration (VARFI and VARDUR); and the proportion of off-feed days (OFFFI and OFFDUR). Analyses were performed by mixed linear models with genomic relationships. The resilience traits, such as TRT, MOR, and HScore, were lowly heritable (up to 0.15) but had high genetic correlations with each other. Performance traits, such as ADG, ADFI, ADFD, FCR, RFI, and carcass traits, were moderate to highly heritable (0.17 to 0.49). Heritabilities of resilience indicator traits such as OFF and VAR had low to moderate heritabilities (0.08 to 0.23) but were higher when based on duration vs. amount. ADFI had a low genetic correlation with ADFD (0.13). ADG in the challenge nursery had stronger negative genetic correlations with both TRT and MOR than ADG in the finisher (-0.37 to -0.74 vs. -0.15 to -0.56). ADFI and FCR had moderate negative (-0.21 to -0.39) and positive (0.34 to 0.49) genetic correlations, respectively, with TRT and MOR. ADFD and RFI had very low genetic correlations with TRT and MOR. CWT and DRS were moderately negatively correlated with TRT and MOR (-0.33 to -0.59). Resilience indicator traits based on feed intake or duration had moderate to high positive genetic correlations with TRT (0.18 to 0.81) and MOR (0.33 to 0.87). In conclusion, performance and resilience traits under a polymicrobial disease challenge are heritable and can be changed by selection. Phenotypes extracted from feed intake patterns can be used as genetic indicator traits for disease resilience. Most promising is day-to-day variation in intake duration, which had a sizeable heritability (0.23) and favorable genetic correlations with MOR (0.79) and treatment rate (0.20).

Exploring Phenotypes for Disease Resilience in Pigs Using Complete Blood Count Data From a Natural Disease Challenge Model

Disease resilience is a valuable trait to help manage infectious diseases in livestock. It is anticipated that improved disease resilience will sustainably increase production efficiency, as resilient animals maintain their performance in the face of infection. The objective of this study was to identify phenotypes related to disease resilience using complete blood count (CBC) data from a wean-to-finish natural disease challenge model, established to mimic the disease pressure caused by many common pathogens at the commercial level of pig production. In total, 2433 F1 crossbred (Landrace × Yorkshire) barrows that went through the natural disease challenge model were classified into four groups (resilient, average, susceptible, and dead) based on their divergent responses in terms of growth and individual treatment. Three sets of blood samples for CBC analysis were drawn at 2-weeks before, and at 2- and 6-weeks after the challenge: Blood 1, Blood 3, and Blood 4 respectively. CBC of Blood 1 taken from healthy pigs before challenge did not show differences between groups. However, resilient animals were found to be primed to initiate a faster adaptive immune response and recover earlier following infection, with greater increases of lymphocyte concentration from Blood 1 to Blood 3 and for hemoglobin concentration and hematocrit from Blood 3 to Blood 4, but a lower neutrophil concentration from Blood 3 to Blood 4 than in susceptible and dead animals (FDR < 0.05). The CBC traits in response to the challenge were found to be heritable and genetically correlated with growth and treatment, which may indicate the potential for developing CBC under disease or commercial conditions as a phenotype in commercial systems as part of developing predictions for disease resilience.

218 Quantifying resilience in growing pigs under a heavy disease challenge using daily individual feed intake records

Resilience is defined as the ability to maintain productivity through any number of stressors such as disease or heat stress. A total of 2273 animals, in groups of ~60–75 piglets, were sent through a natural disease challenge barn every three weeks that consisted of three phases: i) a healthy quarantine nursery to collect immune parameters, ii) a challenge nursery, and iii) a challenged finishing unit which was attached to the challenge nursery. Individual feed intake (FI) was collected in the finishing unit with IVOG® feeders and aggregated into daily totals. Three resilience phenotypes were extracted from the individual trends in feed intake over time, including the root mean square error (RMSE), the quantile regression (QR), and run of depression (ROD) phenotypes. The RMSE phenotype was calculated by fitting a simple linear regression of FI on age within animal and taking the square root of the average squared residual from the model. To calculate the QR phenotype, a 5% quantile regression was fitted across all daily feed intake records to set a lower bound for off-feed days. The QR phenotype was quantified as the proportion of days within animal that fell below the overall quantile regression line. The ROD phenotype was calculated by fitting a within animal linear regression line, flagging extended consecutive stretches of days below that regression line (i.e. a ROD), and calculating the percentage of days that fall within a ROD for each animal. Heritability estimates for the FI resilience phenotypes ranged from 0.10±0.04 to 0.17±0.04. Genetic correlations of the FI resilience phenotypes with mortality and treatment rate ranged from 0.66±20 to 0.94±0.20. This research demonstrates that resilience phenotypes can effectively quantify resilience and can add value to a breeding program to improve resilience to many stressors. Funded by Genome Canada, Genome Alberta, and PigGen Canada.

375 Identification of QTL associated with antibody response to common infectious diseases in commercial sows

The objective of this study was to perform genome-wide association studies (GWAS) to identify Quantitative Trait Loci (QTL) associated with antibody response to infectious diseases in commercial sows. A total of 2,848 Large White x Landrace replacement gilts were sourced from 17 high-health multipliers (7 breeding companies; BC) and introduced to 23 commercial farms with a history of common pig diseases, following the standard acclimation procedures with an average of 53 animals per entry group (CG). Serum was used to quantify antibody response to swine influenza virus (SIV), Mycoplasma hyopneumoniae (MH), porcine circovirus type-2 (PCV2), and 8 serotypes of Actinobacillus pleuropneumoniae (APP1-3, 5, 7, 10, 12, and 13) at entry (S/PEntry), following acclimation (S/PAcclimation), and during parities 1 (S/PParity1) and 2 (S/PParity2). All animals were genotyped for 38,191 SNPs. GWAS was performed using BayesB (pi=0.99), with the fixed effect of CG and the random effects of SNPs in the model. For APP, QTL were only identified at S/PAcclimation; on SSC14 (2Mb) for APP3, APP7, APP10, and APP13 that explained 5.6, 4.7, 2.8, and 3.6% of the genetic variance, respectively. A gene within this QTL region is SYK, involved in the control of immune-receptors. For APP5, a QTL that explained 4.2% of the genetic variance was identified on SSC4 (105Mb), which co-localizes with two genes associated with immune-response: SIKE1and NRAS. For SIV, no QTL was identified. A QTL on SSC7 (130-131Mb) was identified for MH (S/PParity1, 5.1%) and PCV2 (S/PEntry, 34%; S/PAcclimation, 43.4%). These results provide new information on the genetic basis of response to infectious diseases in sows. The identified QTL have the potential to be used to select for improved immune response. The authors thanks PigGen Canada, Genome Canada, and the Canadian Swine Health Board for financial support, and the late Dr. Stephen Bishop for his scientific contributions.

PSIII-5 Accuracy of genomic prediction of antibody response to common infectious diseases in commercial sows

Previous results indicated that antibody response to PRRSV has moderate genomic prediction accuracy; however, little is known about this for other common infectious diseases. Therefore, the objective of this study was to estimate the accuracy of genomic prediction for antibody response to infectious diseases in commercial sows. A total of 2,848 Large White x Landrace replacement gilts were sourced from 17 high-health multipliers (7 breeding companies; BC) and introduced to 23 commercial farms with a history of common diseases, following standard acclimation procedures. Serum was used to quantify antibody response to swine influenza virus (SIV), Mycoplasma hyopneumoniae (MH), porcine circovirus type 2 (PCV2), and 8 serotypes of Actinobacilluspleuropneumoniae(APP1-3, 5, 7, 10, 12, and 13) at entry (S/PEntry), following acclimation (S/PAcclimation), and during parities 1 (S/PParity1) and 2 (S/PParity2). All animals were genotyped for 38,191 SNPs. Genomic prediction was performed using BayesB (pi=0.99), with the fixed effect of CG and random effects of SNPs included in the model. Training and validation were performed using 7-fold cross-validation, with data from each BC used as the validation dataset in one-fold. In general, prediction accuracies were low: SIV, from 0.13 (S/PAcclimation) to 0.26 (S/PParity1); MH, -0.07 (S/PAcclimation) to 0.13 (S/PParity2); PCV2, 0.04 (S/PParity1) to 0.32 (S/PAcclimation); APP, -0.08 (S/PEntry, APP10) to 0.26 (S/PAcclimation, APP7). At each point, average accuracies were 0.06 for S/PEntry, 0.09 for S/PAcclimationand S/PParity1, and 0.08 for S/PParity2, showing small increases in accuracy after the acclimation period. Among diseases, average accuracies ranged from 0.01 (APP1) to 0.22 (PCV2). Results show that, overall, the accuracy of genomic prediction of antibody response to common infectious diseases in commercial gilts is limited. The authors thank PigGen Canada, Genome Canada, and the Canadian Swine Health Board for financial support, and the late Dr. Stephen Bishop for his scientific contributions.

PSIII-8 Genomic prediction of reproductive performance of commercial sows in health challenged herds

The objective of this study was to perform genomic predictions for reproductive performance of sows under natural health challenge. Reproductive performance (1 to 4 parities) and genotype (~40K SNPs) were available for 2,604 crossbred sows, for a total of 7,635 farrowing records. Animals from 17 high-health multipliers from 7 breeding companies (PigGen Canada) were shipped to 23 commercial farms with recent history of common infectious diseases. Gilts entered farms with an average of 53 animals per contemporary group (CG). Traits included total number of piglets: born (TB), born alive (NBA), stillborn (SB), mummified (MUM), born dead (NBD), and weaned (NW). Genomic predictions were performed using Bayes-B (pi=0.995) with a seven-fold cross-validation using each company in turn for validation and the others for training. The model included the effects of CG (fixed) and SNP (random), and net number of fosters (covariate) for NW. Genomic predictions were done for animal lifetime performance (sum performance of parities) for each trait and using first parity performance as the training set to predict subsequent parity performance. Accuracy was calculated as the weighted average correlation between GEBV and adjusted phenotype across validation sets divided by the square root of heritability. Lifetime performance accuracies were low to moderate, ranging from 0.11 (TB) to 0.45 (NBD). Accuracies using parity 1 to predict subsequent performance were low, ranging from -0.07 (SB in parity 3) to 0.19 (NBD in parity 2), with average accuracies per trait ranging from 0.04 (SB) to 0.16 (NBD).Although most accuracies were low, the moderately high accuracies for some lifetime performance shows that genomic prediction can be used to improve performance under natural health challenge in sows. We appreciate the financial support of PigGen Canada, Canadian Swine Health Board, Genome Alberta and Swine Innovation Porc, and the late Dr. Stephen Bishop for his scientific contributions.