Source: UNIVERSITY OF NEBRASKA submitted to
TRANSLATIONAL GENOMICS FOR IMPROVING SOW REPRODUCTIVE LONGEVITY
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
AFRI COMPETITIVE GRANT
Reporting Frequency
Annual
Accession No.
0230886
Grant No.
2013-68004-20370
Project No.
NEB-26-204
Proposal No.
2012-02112
Multistate No.
(N/A)
Program Code
A5102
Project Start Date
Dec 1, 2012
Project End Date
Nov 30, 2017
Grant Year
2013
Project Director
Ciobanu, D. C.
Recipient Organization
UNIVERSITY OF NEBRASKA
(N/A)
LINCOLN,NE 68583
Performing Department
Animal Science
Non Technical Summary
Reducing economic losses associated with reproductive inefficiency represents a high priority for pig producers and will contribute to increased profitability and sustainability of swine industry. Reproductive longevity is measured during the lifetime of an animal and is significantly influenced by many environmental factors. While early age at puberty has been correlated with reproductive longevity of sows, reliable predictors for the onset of puberty, the establishment of productive estrous cycles, and reproductive longevity of an animal remain elusive. Preliminary genome-wide association data indicate clear relationship between genetic variation that influence age at puberty and reproductive longevity. The goal of this proposal is to identify genetic markers that will predict at weaning, gilts with early age at puberty and superior reproductive longevity, which in turn will reduce culling rates and the cost associated with developing replacement sows. Our central hypothesis is that genetic sources that affect age at puberty explain substantial variation in sow reproductive longevity and productivity. To test the central hypothesis, genome-wide association studies will be coupled to transcriptome and genome sequencing, in order to establish the relationships between genetic variation and physiological mechanisms of puberty and reproductive longevity. A unique population resource that incorporates genetics from different commercial lines that expresses important differences in growth and reproductive traits will be used to carry out these studies. We expect that this research will generate knowledge that could be applied to predict and improve early puberty onset and reproductive potential of commercial populations through marker assisted selection.
Animal Health Component
(N/A)
Research Effort Categories
Basic
50%
Applied
50%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
30335101080100%
Knowledge Area
303 - Genetic Improvement of Animals;

Subject Of Investigation
3510 - Swine, live animal;

Field Of Science
1080 - Genetics;
Goals / Objectives
The long-term goal of this project is to develop a panel of genetic markers that will predict at weaning, gilts with superior propensity for reproductive longevity, which in turn will reduce culling rates and the cost associated with developing replacement sows. In order to test our central hypothesis and accomplish the overall objective of this application, the following objectives have been proposed. Objective 1. Molecular dissection of the genetic factors that explain differences in age at puberty and reproductive longevity. Objective 2. Assess the effect of the interaction between energy input and SNP variation on age at puberty and sow reproductive longevity. Objective 3. Identify procedures to apply and transfer marker-association results in selection programs to improve sow reproductive longevity. The data and resources generated by this project will be used for dissection of genetic factors associated with early age at puberty and reproductive longevity and basic understanding of the physiological variation in puberty onset. This project will provide knowledge, resources and tools that can be successfully applied in swine selection. In the short term we plan to analyze DNA variations associated with age at puberty and reproductive longevity by combining phenotypic and physiological data sets, genome-wide associations and gene expression profiling. We will use this information to select a customized panel of markers associated with early age at puberty and identify procedures to apply the information in molecular selection of commercial populations. This research will lower the production costs due to improved reproductive efficiency, it will reduce sow death losses, replacement rates and welfare issues, and increase international competitiveness of the US swine industry.
Project Methods
Objective 1. Activity 1.1. Refinement of QTL positions. This effort plans to expand the size of the phenotypic resource to approximately 1,500 females. The % of genetic variance for age at puberty, litter size traits and reproductive longevity and productivity will be estimated from high-density SNPs genotypes using Bayes analyses implemented through GenSel software. Activity 1.2. Statistical assessment of common sources of variation that influence age at puberty and reproductive longevity. Two multiple trait approaches will be used to identify SNP that explain variability in both age at puberty and reproductive longevity. The first approach will use a model based on the genomic relationship matrix G. The second approach will extend the Bayes Cπ model to handle estimation of multiple traits. Activity 1.3. Genome sequencing. Genome sequencing of 20 boars representative of both ends of the distribution for age at puberty and reproductive longevity will be employed. This information will be used to identify SNP genotypes outside the limited capability of Porcine SNP60K Bead Array and infer genotypes in the experimental gilts. Genome sequencing will use Proton sequencing technology. Activity 1.4. Assess gene expression profile in the hypothalamus. The gilts generated from the future replicates will be genotyped using Porcine SNP60K BeadArray following weaning. Genome predictors for age at puberty will be estimated based on the candidate SNPs located at QTL regions and based on the entire SNP chip set. Genome predictors will be used to classify the animals as early, late and animals that most likely will not express estrus before 240 days of age. Hypothalami will be collected from candidate animals at three different time points: 140, 170 and 240 d of age. RNA will be isolated and RNA sequencing will employ Solid technology. We will estimate the differences in expression between different time points and genetics groups using Cuffdiff package. Objective 2. A mixed model analysis of phenotypic variation including diet and replicate as fixed effects and sire, sire x diet and litter as random effects, will be employed to estimate the response of energy restriction during gilt development on age at puberty in the expanded population. In addition, a Genome Wide Association Study will be performed separately in gilts developed with ad libitum and energy restricted diets to uncover potential genes functionally triggered by a fluctuation in energy. Objective 3. The working hypothesis of this aim is that major genetic variants that explain significant variation in age at puberty and reproductive longevity will be expressed in multiple commercial populations, and can be used to generate a selection response and economic gain. To test this hypothesis, we will organize demonstration projects of the effect of a reduced DNA marker panel on sow longevity and productivity, we will identify and implement innovative solutions to disseminate the technology and knowledge to swine industry.

Progress 12/01/12 to 11/30/17

Outputs
Target Audience:Target audiences include scientists in the same research area, veterinarians, graduate and undergraduate students and swine producers. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project provided significant training opportunities to junior technicians, graduate and undergraduate students. During the time of the grant there were 4 graduate students (4 MS and 2 PhD) directly involved in this research project. How have the results been disseminated to communities of interest?Journal articles, presentations and posters were presented to regional, national and international conferences as well as meetings with producers. We collaborate closely with two commercial companies involved in this study. We plan to provide technical support for the transfer to commercial production of the knowledge and technologies developed during this research. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Our research made advancements in the understanding of the role of genetic variation in predicting puberty and reproductive success. Our research reached the proposed short- and long-term impact/outcomes outlined in the Logic Model. Specifically, our research "increased knowledge of applications of DNA technology in genetic improvement programs" and provided "preliminary associations between genotypes and sow fertility", "informed use of DNA technology by swine producers" and "transfer of research findings to the industry". Our research generated genomic and phenotypic information to try to understand the role of genetic and nutrition in sow fertility, specifically age at puberty and reproductive longevity. We integrated genome and RNA sequencing, high-density genotyping and Quantitative Trait Locus (QTL) mapping to uncover genes and polymorphisms that influence age at puberty and other sow fertility traits. All this information was integrated in an Affymetrix SNP chip (SowPro90) that included 103K SNPs that was used to validate our research findings in commercial populations. Two commercial population including 2,300 sows were genotyped with the SowPro90. The impact of this research could be substantial by predicting early in life sow with high fertility potential. One of the initial DNA markers identified improved lifetime productivity by 0.5 parities, total number born pigs per lifetime by 5.1 and number of pigs weaned per lifetime by 4.2. Any progress in the ability to designate early in life females with high fertility will have an important economic value. An increase in the average number of parities by only 0.10 will generate approximately $10 mil/year in US alone. Specific progresses in each of the objectives are described below. Objective 1. 1.1 Major activities completed -Expanded data for range of fertility and growth phenotypes. -QTL positions were refined using a multitude of genomic approaches using an expanded resource population (n~1,700). The statistical models evaluated included BayesC, BayesB and the Bayes Interval Mapping (BayesIM) recently introduced by Steve Kachman. -Hypothalamic nuclei from gilts with different puberty status were micro-dissected, RNA isolated, sequenced and data analyzed. Differentially expressed genes (DEG) between gilts with variation in the expression of puberty were identified. Ingenuity Pathway Analysis (IPA) was conducted and integrated with GWAS results to uncover functional genes, potential pathways and upstream regulators that could explain some of the differences in age at puberty. 1.2. Data collected -We collected genotypes (Porcine SNP60 BeadArray), fertility and growth phenotypes from over 800 sows. -20 sires with divergent genome-wide predictors of the daughter's age at puberty were sequenced using next generation sequencing. -RNA sequencing data was obtained from hypothalamic nuclei micro-dissected from 37 gilts. 1.3 Summary statistics and discussion of results -Phenotypes and high-density genotypes from UNL gilts (n = 1,644) were used in GWAS. The largest fraction of phenotypic variation explained by the Porcine SNP60 BeadArray was for age at puberty (27.3%). -Genomic sequencing revealed 11,896,069 SNP and 1,074,512 indels among 20 sires sampled. -Seventy DEG were identified in hypothalamic arcuate nucleus between early and late age at puberty gilts. Arcuate nucleus plays a major role in regulating the onset of puberty through controlled secretion of gonadotropins. Three of these genes overlapped with top 1% of the QTL for age at puberty. Genetic variants located upstream of transcription start site affecting potential cis- binding motifs were identified as possible sources of differential expression and variation in puberty. There were 363 upstream regulators of 70 DEG identified. Some of these upstream regulators overlapped with QTL for age at puberty. 1.4. Key outcomes or other accomplishments realized. -Genome-wide association analysis using BayesB uncovered major 1-Mb windows associated with age at puberty that explained from 0.32 to 0.61% of the genetic variation for age at puberty. One of the high-ranked windows detected (SSC2, 12-12.9 Mb) showed pleiotropic features, affecting both age at puberty and litter size traits. A candidate gene in this area, P2X3R, is involved in embryo implantation and maintenance of pregnancy. -Integration of GWAS, genome and RNA sequencing uncovered potential genes and functional polymorphism that explain some of the variation in age at puberty. -The development of BayesIM provided a more refined genome-wide associations, especially in genomic regions characterized by limited linkage disequilibrium. -Energy restriction during the gilt development period delayed age at puberty by 7 days, but increased the probability of a sow to produce up to 3 parities (P < 0.05). Objective 2. 2.1 Major activities completed -Different statistical assessments of common sources of variation that influence age at puberty and reproductive longevity were evaluated. 2.2. Data collected -QTL data was obtained for age at puberty and lifetime number of parities. 2.3 Summary statistics and discussion of results -BayesIM, single trait and multiple trait models incorporating the genomic relationship matrix were used to assess common sources of variation influencing lifetime number of parities and age at puberty 2.4. Key outcomes or other accomplishments realized. -Discrepancy in heritability estimates (high vs low) between traits such as age at puberty and lifetime number of parities targeted for assessment of pleiotropic effects could impact the ability to identify with confidence common sources of genetic variation. Objective 3. 3.1 Major activities completed / experiments conducted; -We integrated genome and RNA sequencing, high-density genotyping and QTL mapping and designed a custom Affymetrix SNP chip (SowPro90) that included 103K SNPs. This chip includes SNPs located in genes/QTL associated with age at puberty and SNPs located in genes involved in innate and adaptive immunity -The potential of genomic information to explain phenotypic variation for age at puberty was analyzed using UNL data set and the Porcine SNP60 BeadArray genotypes. -Commercial phenotypes were analyzed and a subset of commercial samples was selected for genotyping using SowPro90 (n = 2,300). -A sow longevity economic model was constructed reporting cost of production and reproductive efficiencies. Genomic information can be added into the model. 3.2. Data collected -Tissues, phenotypes and SowPro90 genotypes were obtained from two commercial populations (n ~ 2,300 sows) 3.3 Summary statistics and discussion of results -In an evaluation data set using Porcine SNP60 BeadArray, the predictive ability of all SNP from high-ranked 1-Mb windows (1 to 50%), based on genetic variance explained in training, was greater (12.3 to 36.8%) compared to the most informative/major SNP from these windows (6.5 to 23.7%). This finding is probably due to the fact that SNP identified as highest ranked in the training set are not functional variants, and the LD between these SNP and the actual functional polymorphisms is redefined in the evaluation set. -The quality of the data generated by SowPro90 was analyzed using Axiom Analysis Tool. The vast majority of samples passed the quality control (>98.5%). Also, 90.7% of the SNPs passed several quality control filtering. 3.4. Key outcomes or other accomplishments realized. -We expect that major genetic variants that explain important variation in age at puberty and reproductive longevity will be expressed in different populations, and can be used to predict a selection response and economic gain. The knowledge developed by this grant has the potential to improve sow fertility, reproductive longevity and profitability in the commercial herds

Publications

  • Type: Journal Articles Status: Published Year Published: 2017 Citation: Wijesena H. R., C. A. Lents, J.-J. Riethoven, M. D. Trenhaile-Grannemann, J. F. Thorson, B. N. Keel, P. S. Miller, M. L. Spangler, S. D. Kachman, D. C. Ciobanu (2017) Integration of Genomic Approaches to Uncover Sources of Variation in Age at Puberty and Reproductive Longevity in Sows, J Anim Sci. 2017 Sep;95(9):4196-4205.
  • Type: Journal Articles Status: Published Year Published: 2016 Citation: Trenhaile M.D., J. Petersen, S.D. Kachman, R. K. Johnson, D.C. Ciobanu, (2016) Long-term selection for litter size in swine results in shifts in allelic frequency in regions involved in reproductive processes, Anim Genet. 2016 May 26. doi: 10.1111/age.12448.
  • Type: Journal Articles Status: Published Year Published: 2015 Citation: Lucot K.L., M.L. Spangler, M.D. Trenhaile, S.D. Kachman, D.C. Ciobanu, (2015) Evaluation of reduced subsets of Single Nucleotide Polymorphisms for the prediction of age at puberty and reproductive longevity in sows, Animal Genetics, 46(4):403-9.
  • Type: Journal Articles Status: Published Year Published: 2015 Citation: Morota G., Pe�agaricano F., Petersen J.L., Ciobanu D.C., Tsuyuzaki K., Nikaido I. (2015), An application of MeSH enrichment analysis in livestock. Anim. Genet. Aug; 46(4):381-7.
  • Type: Journal Articles Status: Published Year Published: 2013 Citation: Tart J.K., R.K. Johnson, J.W. Bundy, N.N. Ferdinand, A.M. McKnite, J.R. Wood, P.S. Miller, M.F. Rothschild, M.L. Spangler, D.J. Garrick, S.D. Kachman, D.C. Ciobanu (2013) Genome-wide prediction of age at puberty and reproductive longevity in sows, Animal Genetics, Aug;44(4):387-97. doi: 10.1111/age.12028.


Progress 12/01/15 to 11/30/16

Outputs
Target Audience:Target audiences include scientists in the same research area, veterinarians, graduate and undergraduate students and swine producers. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Junior technicians, graduate and undergraduate students are involved in this project that provides great training opportunities. How have the results been disseminated to communities of interest?Journal articles, presentations and posters were presented to regional, national and international conferences as well as meetings with producers. What do you plan to do during the next reporting period to accomplish the goals?The goals for 2017 are: - Complete selection and filtering of potential functional polymorphisms to be included in the functional array. Validate the effect of these novel polymorphisms in commercial populations.

Impacts
What was accomplished under these goals? Our main progresses were generated in the areas of RNA and genome sequencing and evaluation of a novel approach for genome-wide associations. The integration of all this data has potential to lead to the discovery of functional mutations that could reduce age at puberty, improve fertility and reproductive longevity, leading to an increase in sow net values in the commercial herds. A paper was submitted and accepted in Animal Genetics. Another paper was submitted to Journal of Animal Science. Objective 1. Activity 1.1. Refinement of QTL positions. Progress: - Phenotype collection: batches B1-B13 completed, B14-B15 in progress. - GWAS updated for age at puberty (AP) (B1-B14, n~1,700) - A novel genome wide association approach called Bayes Interval Mapping (BayesIM) was evaluated. BayesIM introduced by Kachman (2014) fits haplotypes rather than individual SNP as is the case of other Bayesian approaches. Results: This year several improvements were introduced in BayesIM such as direct estimation of the haplotype block size and input of the parental haplotypes if available. The improved version was evaluated using different inputs options and across diverse traits. This statistical approach provides improved accuracy in QTL mapping. Activity 1.3. Genome sequencing. Progress: Genome sequencing and primary data analysis of 20 sires that represent both ends of the distribution for daughters' age at puberty was completed. Sequencing was performed using Proton sequencing technology to identify SNP genotypes outside the limited capability of the Porcine SNP60K Bead Array. Results: The obtained genomic sequences of the all sires varied from 16.2 to 26.7 fold (x) with an average of 22.2 x coverage. The average length of the sequencing reads after filtering was 164.9 bp. The uncovered genetic variants filtered for Phred quality score (≥ 20), coverage (≥ 10) and BAF score (≥ 0.5 for indels only) totaled 11,218,177 SNPs and 2,879,291 indels. The majority of the uncovered SNPs were intergenic (60.6%). Intronic SNPs were the most prevalent (96.4%) from all polymorphisms located in genes, followed by 5' and 3' UTR (2.0%) and the coding region (1.6%). A number of these polymorphisms could be potential sources of variation if they are located in the extended area of the major 1-Mb windows associated with phenotypic differences for the targeted traits. Activity 1.4. Assess gene expression profile in the hypothalamus. Progress: Hypothalamic, pituitary, and ovarian tissues were collected, microdissected, RNA isolated and sequenced from gilts with different puberty status (n=37). Results: Arcuate, paraventricular and ventromedial nucleus of the hypothalamus were micro-dissected from the hypothalamus of 37 gilts. RNA was extracted and sequenced from arcuate nucleus using Ion Proton technology. High throughput RNA sequencing reads were obtained from gilts representing pre- and post-pubertal time points, and early and late puberty gilts fed 3 different dietary treatments. On average, 55.3 million raw, single-end reads with an average length of 150 bp were obtained per each gilt. After trimming the reads based on quality, ~90% of the raw reads per gilt were available for transcriptome analysis. Using a 2-step alignment process we were able to map ~94% of the trimmed reads to the genome. Of these mapped reads, ~45% of the reads were mapped to actual genes. Differential expression between early and late pubertal gilts was found for 70 genes (Padj < 0.1) including genes involved in embryo implantation and oxytocin-signalling pathway. The expression of these genes was up regulated in gilts exhibiting puberty at later ages compared to gilts with early age at puberty. Objective 2. Assess the effect of interaction between energy input and sow genetics on age at puberty and sow reproductive longevity. Progress: - Hypothalamic, pituitary, and ovarian tissues were collected, microdissected, RNA isolated and sequenced from gilts fed either a standard or energy-restricted diet prior to breeding. Results: The arcuate nucleus is one of the major modulators of energy homeostasis and links nutrition with reproductive development in gilts. Gene expression of the arcuate nucleus was profiled using RNA sequencing to investigate the role of energy restriction in sow reproductive development. There were 42 differentially expressed genes between standard and energy restricted diets. Genes down-regulated by energy restriction include those involved in ion transport, insulin secretion, carbohydrate digestion and regulation of the response to lipopolysaccharides. Objective 3. Identify procedures to apply and transfer marker-association results in selection programs to improve sow reproductive longevity. Progress: The working hypothesis of this aim is that major genetic variants that explain significant variation in age at puberty and reproductive longevity will be expressed in different populations, and can be used to generate a selection response and economic gain. The technologies generated by the research component of the grant will be evaluated in three commercial populations. Results: Preliminary discussions regarding the design of the functional DNA marker chip were initiated. This chip will be evaluated in different populations. The chip will contain ~90K polymorphisms and will include functional polymorphisms discovered in this project but also SNPs represented in common arrays/chips. Impact Genome and RNA sequencing of representative samples uncovered a large number of genetic variants across genome and captured transcriptome diversity and gene expression differences between gilts with different puberty status and energy intake. Genome sequencing provided polymorphisms outside the limited capability of the commercial SNP array or chips. Some of these polymorphisms could be actual causative sources of phenotypic variation. Some of the differences in expression of arcuate nucleus could be a result of cis-modulation and if are located in the extended QTL regions they could be potential sources of the differences in age at puberty between gilts. In addition, a greater understanding of the molecular mechanisms involved in the interaction between sow genome and nutrition could be useful in designing efficient strategies to develop highly prolific sows.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2016 Citation: D.C. Ciobanu, E.R. Tosky, S. Olson, M.D. Trenhaile, C.A. Lents, T.P.L. Smith, D.J. Nonneman, G. Rohrer, J. Chin, P.S., Miller, T.E. Burkey, M.L. Spangler, J.J. Riethoven, G.S. Plastow, S.D. Kachman, Development of resources and tools for mapping genetic sources of phenotypic variation, Plant and Animal Genome Conference, January 9 - 13, 2016, San Diego, CA.
  • Type: Journal Articles Status: Published Year Published: 2016 Citation: Trenhaile MD, Petersen JL, Kachman SD, Johnson RK, Ciobanu DC., 2016, Long-term selection for litter size in swine results in shifts in allelic frequency in regions involved in reproductive processes. Anim Genet. 2016 Oct;47(5):534-42. doi: 10.1111/age.12448.
  • Type: Journal Articles Status: Submitted Year Published: 2017 Citation: Wijesena HR, CA Lents, J-J. Riethoven, MD Trenhaile-Grannemann, JF Thorson, BN Keel, PS Miller, ML Spangler, SD Kachman, DC Ciobanu, Integration of Genomic Approaches to Uncover Sources of Variation in Age at Puberty and Reproductive Longevity in Sows, Submitted to Journal of Animal Science.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2016 Citation: Daniel C. Ciobanu, Stephen D. Kachman, Sean Olson, Matthew L. Spangler, Melanie Trenhaile, Hiruni R. Wijesena, Phillip Miller, Jean-Jack Riethoven, Clay A. Lents, Jennifer Thorson, Raymond Massey, Timothy J. Safranski, 2016, Translational Genomics for Improving Sow Reproductive Longevity, ADSA ASAS Joint Annual Meetings, Salt Lake City, July 19-23, 2016.
  • Type: Conference Papers and Presentations Status: Submitted Year Published: 2017 Citation: Hiruni R. Wijesena, Clay A. Lents, Melanie D. Trenhaile - Grannemann, Jean-Jack Riethoven, Brittney N. Keel, Jennifer F. Thorson, Phillip S. Miller, Rodger K. Johnson, Matthew L. Spangler, Stephen D. Kachman, Daniel C. Ciobanu, 2017, The roles of age at puberty and energy restriction in sow reproductive longevity: a genomic perspective, Midwest ASAS Annual Meeting, March 13-15, 2017.
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2017 Citation: H.R. Wijesena, C.A. Lents, B.N. Keel, J.F. Thorson, G.A. Sullivan, M.L. Spangler, S.D. Kachman and D.C. Ciobanu, 2017, Variation In Gene Expression In The Hypothalamic Arcuate Nucleus of Gilts With Differences in Pubertal Status and Subjected To Dietary Energy Restriction, Plant and Animal Genome Conference, San Diego, January 14-18, 2017.
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2017 Citation: D.C. Ciobanu, H.R. Wijesena, C.A. Lents, M.D. Trenhaile - Grannemann, J.J. Riethoven, J.F. Thorson, B.N. Keel, P.S. Miller, M.L. Spangler, S.D. Kachman., 2017, Integration of genomic resources to uncover pleiotropic regions associated with age at puberty and reproductive longevity in sows. Plant and Animal Genome Conference, San Diego, January 14-18, 2017.


Progress 12/01/14 to 11/30/15

Outputs
Target Audience:Target audiences include scientists in the same research area, veterinarians, graduate and undergraduate students and swine producers. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Junior technicians, graduate and undergraduate students are involved in this project that provides great training opportunities. How have the results been disseminated to communities of interest?Journal articles, presentations and posters were presented to regional, national and international conferences as well as meetings with producers. What do you plan to do during the next reporting period to accomplish the goals?Areas to focus in 2016: - Generate a reduced SNP panel for age at puberty, litter size and reproductive longevity. - Evaluate the potential of genomic information to explain phenotypic variation for reproductive traits in commercial populations.

Impacts
What was accomplished under these goals? Our main progresses were generated in the area of development and evaluation of different genome-wide association methods, gene and genome annotation and dissection of QTL associated age at puberty, litter size and reproductive longevity. We also discovered DNA markers with predictive potential to be used across different populations of swine. These results have potential to reduce age at puberty, improve fertility and reproductive longevity, leading to an increase in sow net values in the commercial herds. A paper was submitted and accepted in Animal Genetics. Objective 1. Activity 1.1. Refinement of QTL positions. Progress: - Phenotype collection: batches B1-B11 completed, B12-B14 in progress. - GWAS updated for age at puberty (AP) (B1-B13, n~1550) and lifetime number of parities (NP) (B1-B11, n=1100) - Selection sweep was investigated in Nebraska Index line (NIL) based on changes in allelic, genotype and haplotype frequencies and compared to control line (CTRL) and a Duroc x Hampshire (D x H) cross. - Various genome-wide associations approaches were employed including Single DNA Marker Association Analysis, gBLUP, BayesB and Bayes Interval Mapping (BayesIM). Results: - A new GWAS approach called Bayes IM was developed. A hidden Markov model was used to generate haplotype clusters based on SNP genotypes. Age of puberty and longevity were then analyzed with a hierarchal Bayesian model. Haplotype loci were placed at designated distances across the genome. If a locus had an effect, haplotype effects for each cluster were modeled as independent normal random variables. This approach provides flexibility in QTL search and accuracy in QTL location. - Regions that displayed substantial changes in SNP heterozygosity and allelic frequency between populations and haplotypes that are likely under selection in NIL were identified. Some of these regions overlapped major QTL for litter size traits identified by genome-wide association studies and harbor candidate genes involved in reproductive processes. One of these regions is located on SSC2 (13-14 Mb) and includes gene P2X3R, which plays a role in implantation and sustained release of hormones associated with reproductive processes; SNPs that are fixed in NIL and polymorphic with equal frequencies in DxH were identified and initial associations underlined suggestive relationships between SNP genotypes in P2X3R and litter size traits. Activity 1.2. Statistical assessment of common sources of variation that influence age at puberty and reproductive longevity. Progress: - BayesIM approach was implemented for multiple traits analyses. Results: - BayesIM was used for multiple trait analyses to identify SNPs and regions that explain variability in both, age at puberty and lifetime number of parities. Activity 1.3. Genome sequencing. Progress: - 20 sires representative of both ends of the distribution for age at puberty were sequenced using Ion Torrent sequencing technology to identify SNP genotypes outside the limited capability of the Porcine SNP60K Bead Array. Results: - The obtained genomic sequence of the all sires varied from 16.2 to 26.7 folds with an average of 22.2 folds in genome coverage. The average length of the sequencing reads post quality filtering was 164.9 nucleotides. Post filtering average number of polymorphisms per sire was 4.21 mil SNPs and 1.28 mil short indels. Activity 1.4. Assess gene expression profile in the hypothalamus. Progress: - Hypothalamic, pituitary, and ovarian tissues were collected from pre- and post-pubertal gilts (total n=39) fed either a standard or energy-restricted diet prior to breeding. Results: - Arcuate, paraventricular and ventromedial nucleus of the hypothalamus were micro-dissected from 27 gilts. RNA was extracted and RNAseq data obtained from the first set of 12 hypothalamic arcuate nucleus RNA samples. Objective 2. Assess the effect of interaction between energy input and SNP variation on age at puberty and sow reproductive longevity. Progress and results: - A Generalized Linear Mixed Model was implemented to assess the effect of SNP x Diet on age at puberty and reproductive longevity across genome. Objective 3. Identify procedures to apply and transfer marker-association results in selection programs to improve sow reproductive longevity. Progress: - The working hypothesis of this aim is that major genetic variants that explain significant variation in age at puberty and reproductive longevity will be expressed in multiple commercial populations, and can be used to generate a selection response and economic gain. The technologies generated by the research component of the grant will be evaluated in commercial datasets from Pillen Family Farms, The Maschhoffs and Smithfield. - Assess the potential of genomic information to explain phenotypic variation for age at puberty. Results: - Tissues and phenotypic information was collected from ~12,000 gilts from Smithfield, ~3,000 gilts from The Maschhoffs and 1,200 from Pillen Family Farms. - Full set of SNPs and various subsets of SNPs from 1-Mb windows that explained the greatest proportions of variation in a training resource based on the first UNL batches (B1 - B7) were used to estimate genomic prediction values (GPVs) in the subsequent UNL batches (B8-B11). The phenotypic variance explained in the evaluation data sets varied from 12.3% to 36.8% when all SNPs from major windows were used compared to 6.5-23.7% explained by most informative SNPs. The correlation between phenotype and genomic prediction values based on SNP effects estimated in the training population was marginal compared to their effects retrained in the evaluation population for all (0.46-0.81) or most informative SNPs (0.30-0.65) from major windows. A manuscript reporting these results was submitted and published in Animal Genetics (Lucot et al., 2011). The same study estimated that an increase in genetic gain of 20.5% for age at puberty could be obtained if genomic selection included both sexes compared to females alone. - A sow longevity economic model was constructed. The model contains two components for weaned pig production, each reporting cost of production and reproductive efficiencies. The first component models the productivity and economics of individual females that reach parity 1 through 10. The second component models the productivity and economics of the herd given different parity achievements. The cost of production is reported on a dollars per pig weaned basis. Genomic information will be added into the model in the following version.

Publications

  • Type: Journal Articles Status: Published Year Published: 2015 Citation: Lucot KL, Spangler ML, Trenhaile MD, Kachman SD, Ciobanu DC., Evaluation of reduced subsets of single nucleotide polymorphisms for the prediction of age at puberty in sows. Anim Genet. 2015 Aug;46(4):403-9
  • Type: Journal Articles Status: Published Year Published: 2015 Citation: Morota G, Pe�agaricano F, Petersen JL, Ciobanu DC, Tsuyuzaki K, Nikaido I., An application of MeSH enrichment analysis in livestock. Anim Genet. 2015 Aug;46(4):381-7.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2015 Citation: M.D. Trenhaile, S.D. Kachman, P.S. Miller, R.K. Johnson, and D.C. Ciobanu, Genomic analysis of the interaction between energy intake and SNP genotypes on age at puberty, Plant and Animal Genome Conference, January 10 - 14, 2015, San Diego, CA.


Progress 12/01/13 to 11/30/14

Outputs
Target Audience: Target audiences include scientists in the same research area, veterinarians, graduate and undergraduate students and swine producers. Changes/Problems: The large Proton sequencing chip was not released yet and delayed some research activities. However, with the help of the Molecular Core from UTHSC we identified another option, at a similar cost with what was budgeted, that will privide the same sequencing coverage planned in the research proposal. As a result, genome sequencing of 7 sires (boars) was completed with additional 3 boars being in sequencing process. What opportunities for training and professional development has the project provided? Junior technicians, graduate and undergraduate students are involved in this project that provides great training opportunities. How have the results been disseminated to communities of interest? During 2014 we organized two meetings with the three industry organizations participating in the research and extension work. During that meeting we shared flyers with them that summarized progress on our research and extension activities. What do you plan to do during the next reporting period to accomplish the goals? The objective of this project for 2015 was to 1) continue the dissection of major QTLs and candidate genes, known for their pleiotropic effect on AP and RL, 2) evaluate hypothalamic gene expression differences in gilts that expressed early and late age at puberty, and 3) start to analyze sequencing data from sires that have important differences in genomic prediction values in age at puberty and reproductive longevity.

Impacts
What was accomplished under these goals? Advancements were made in generating data that contributed to the understanding of the role of genetic variation in predicting puberty and reproductive success. Predictive ability of the phenotype by subsets of DNA markers was assesed in the UNL resource population. We determined that subsets of DNA markers associated with age at puberty are able to predict the phenotype as well as large DNA marker panels. One of the main advantages of Marker Assisted Selection is that selection based on DNA markers can be applied even on traits limited by sex such as pubery and lifetime number of pigs farrowed. An increase in genetic gain of 20.5% for age at puberty could be obtained if genomic selection included both sexes compared to females alone. One of the DNA markers identified has a pleiotropic role, influencing both age at puberty and reproductive longevity. This DNA marker was found to reduce age at puberty and improve lifetime productivity by 0.5 parities, total number born pigs per lifetime by 5.1 and number of pigs weanned per lifetime by 4.2. The difference in probabilities between the favorable genotype and the other genotypes in generating the first parity increased with the delay in the expression of age at puberty. For example, the difference in probability to generate parity 1 between gilts with the favorable genotype and gilts with the alternate homozygote genotype was 5% higher in the group of gilts that expressed early estrus compared to 12.9% higher in the group that expressed estrus late. Any progress in the ability to designate early in life females with high fertility will have an important economic value. An increase in the average number of parities by only 0.10 will generate aproximately $10 mil/year in US alone. Specific progresses in each of the research and extension objectives are described below. Objective 1. Activity 1.1. Refinement of QTL positions. Progress: - Phenotype collection: batches B1-B10 completed, B11-B13 in progress. - GWAS updated for age at puberty (AP) (B1-B12, n=1,432) and lifetime number of parities (NP) (B1-B10, n=990) - Selection sweep was investigated in Nebraska Index line (NIL) based on allelic frequency and heterozigosity changes compared to control line (CTRL) and a Duroc x Hampshire (D x H) cross Results: - High density genotyping showed distinct clustering of the NIL, CTRL and D x H populations, indicating potential selective pressure in NIL. There were 64 SNPs (0.15%) with large differences in allelic frequency between NIL and CTRL (FST > 0.616), which corresponds to the 99.9 percentile of FST simulation data including multiple SNPs from SSC2 (12-14 Mb) that also overlapped a major QTL for litter size Activity 1.2. Statistical assessment of common sources of variation that influence age at puberty and reproductive longevity. Progress: - A student was recruited in fall in the Dept. of Statistics (Kachman PI) to tackle this objective. Results: - Initial ASReml statistical models were used for multiple trait analyses to identify SNPs that explain variability in both, age at puberty and lifetime number of parities. Activity 1.3. Genome sequencing. Progress: - Semen was collected and DNA was extracted from all sires that generated the gilts from the last 5 batches. - Genome sequencing of 7 sires is completed with additional 3 being in process - Sequences have been quality trimmed and short sequences (<30bp) removed - SNP analysis for the first 7 sires is in progress Results: - The obtained genomic sequence of the first set of sires (n=7) varies from 21.3 to 29.7 folds with an average of 25.2 folds in genome coverage. A slight increase in average read size and reduction of the standard deviation was observed across samples following filtering protocols to improve sequencing quality. The average length of the sequencing reads post quality filtering was 170.6 nucleotides. Activity 1.4. Assess gene expression profile in the hypothalamus. Progress: - Hypothalamic, pituitary and ovarian tissues were collected from pre and pubertal gilts (n=24) fed 4 months before breeding with either standard or energy restricted diet. - Dissection methods of hyphothalamic nuclei involved in puberty were adapted to swine. Results: - Evaluation of gene expression of the dissected tissues is underway. Melanie Trenhaile (UNL student) works in Lents's USMARC lab 1 day/week. Objective 2. A mixed model analysis of phenotypic variation was employed to estimate the response of energy restriction during gilt development and genotype on age at puberty and reproductive longevity. Progress: - New GWAS approach that could uncover SNP effects affected by diet were assessed - Validation of diet dependent SNP effects was performed by single marker association analyses using additive and dominance generalized linear mixed models. Results: - Reduced age at puberty and restricted energy intake before breeding had a positive effect on the probability that the sows would generate subsequent litters starting with parity 2 (P < 0.05). The restricted diet delayed AP by 7 days. - The proportion of the phenotypic variation of AP explained by SNPs was 28% with SNP x Diet explaining a limited proportion of phenotypic variation (0.3%). Diet-dependent QTL regions were identified on SSC6, SSC7, SSC8, SSC11, SSC14, and SSC16. - Individual genes of interest in these regions include genes involved in lipid metabolism, regulation of MAP kinase activity and embryonic development, and carbohydrate and sugar uptake and transport and intestinal absorption. All eight SNPs in these QTL regions had a significant additive SNP x diet effects on age at puberty. The SNPs either had an effect in only one of the dietary treatments or the direction of the effect was opposite between the two treatments. Objective 3. The working hypothesis of this aim is that major genetic variants that explain significant variation in age at puberty and reproductive longevity will be expressed in multiple commercial populations, and can be used to generate a selection response and economic gain. The technologies generated by the research component of the grant will be evaluated in demonstration projects tailored to major State and National producers, such as Pillen Family Farms, The Maschhoffs and Smithfield. Progress: - Industry sample collection was finalized. Tissues collected from ~12,000 gilts from Smithfield, ~3,000 gilts from The Maschhoffs and 1,200 from Pillen Family Farms were shipped and stored at UNL. - The full set of SNPs and subsets of major SNPs and 1 Mb windows based on a GWAS in a training set (batches B1-B7) were used to assess the potential of genomic information to explain phenotypic variation for age at puberty and reproductive longevity in an evaluation set (B8-B11). Results: - A GWAS using the full set of 56,424 SNPs explained 25.2% of the phenotypic variation in age at puberty in a training set (B1 - B7, n = 820). When the predictive potential of subsets of SNPs was evaluated in subsequent batches (B8 - B11, n = 412) the phenotypic variance explained varied from 12.3% to 36.8% when all SNPs from major windows were used compared to 6.5% to 23.7% explained by most informative SNPs.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2014 Citation: Trenhaile MD, JL Petersen,KL Lucot,SD Kachman,RK Johnson,DC Ciobanu (2014) Long-term selection for litter size results in significant shifts in allelic frequency in regions involved in reproductive processes, Proceedings of WCGALP, 17 - 22 August, 2014, Vancouver, BC.
  • Type: Conference Papers and Presentations Status: Submitted Year Published: 2015 Citation: M.D. Trenhaile, S.D. Kachman, P.S. Miller, R.K. Johnson, and D.C. Ciobanu, Genomic analysis of the interaction between energy intake and SNP genotypes on age at puberty, Plant and Animal Genome Conference, January 10 - 14, 2015, San Diego, CA.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2014 Citation: Trenhaile M.D., K.L. Lucot, J.K. Tart, J.W. Bundy, J. F. Thorson, E.M. Keuter, J.R. Wood, M.F. Rothschild, G.A. Rohrer, P.S. Miller, M.L. Spangler, C.A. Lents, R.K. Johnson, S.D. Kachman, and D.C. Ciobanu (2014) AVPR1A alleles are pleiotropic sources of the variation in age at puberty and reproductive longevity in sows, ASAS Midwest Meeting, March 17-19, 2014, Des Moines, IA.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2014 Citation: Lucot K.L, M.L. Spangler, S.D. Kachman, and D.C. Ciobanu (2014) Genomic prediction of age at puberty in sows using Bayesian methods, ASAS Midwest Meeting, March 17-19, 2014, Des Moines, IA.


Progress 12/01/12 to 11/30/13

Outputs
Target Audience: Target audiences include scientists in the same research area, veterinarians, graduate and undergraduate students and swine producers. Changes/Problems: The release of the large Proton sequencing chip that will allow us to sequence sires of both ends of the distribution for age at puberty and reproductive longevity at high density and low cost was delayed for Spring 2014. What opportunities for training and professional development has the project provided? Both graduate and undergraduate students are involved in this project that provides great training opportunities. How have the results been disseminated to communities of interest? Presentations and posters were presented to regional, national and international meetings. What do you plan to do during the next reporting period to accomplish the goals? The objective of this project for 2014 was to 1) continued the dissection of AVPR1A and other major genes, known for its pleiotropic effect on AP and RL, 2) analyze the effects of the major markers in other populations and 3) start to sequence the genome of sires that have important differences in genomic prediction values in age at puberty and reproductive longevity

Impacts
What was accomplished under these goals? Objective 1. Activity 1.1. Refinement of QTL positions. The goal of this activity was to refine the QTL position for age at puberty (AP) and reproductive longevity (RL) using a resource population developed at UNL. The gilts were produced in replicates of approximately 90-120 gilts per replicate. Females were maintained through four parities. Females that did not conceive or farrow a litter and those with structural problems were culled. The first eight replicates are completed and three replicates are in various stages of progress (R9-R12). In all replications, gilts received same diet and management until 123 d of age when the gilts are placed on experimental energy restricted diets (~80% of the ad libitum diet) until breeding. The DNA of the first 12 replicates was isolated and genotyped using Porcine 60K SNP BeadArray. In addition to previous collection routines, we collected plasma from all experimental sows, before boar exposure and during the boar exposure process. We conducted a genome-wide association study (GWAS) for AP and RL using a Bayes B algorithm with a pi value of 0.99 and the concatenation of diet and batch fitted as a fixed effect. A total of 56,424 SNPs explained 0.28 of the phenotypic variation for AP. The proportion of the phenotypic variation explained by all SNPs within the top 1%, 5%, 10% and 20% windows varied from 0.22 (1% windows; 645 SNPs) to 0.39 (10% windows; 19,362 SNPs). In contrast, the proportion of the phenotypic variation explained by the most informative SNP from these windows varied from 0.18 (1% windows; 24 SNPs) to 0.48 (20% windows; 259 SNPs). Different pi values (0, 0.25, 0.50, 0.75 and 0.99) had a limited effect on the proportion of phenotypic variation explained by the top 1% (0.20 to 0.23) and 10% (0.36 to 0.37) windows. The first 7 batches were used as training data (B1 - B7, n=822) to evaluate the ability of major SNPs and windows to predict AP in subsequent batches. The pooled simple correlation between genomic prediction values (GPV) and adjusted AP phenotypes was 0.18 in B8 - B11 (n=412) when 56,424 SNPs were used. When GPV were derived using the most informative SNP from each of the top 10% windows or all SNPs from the top 10% windows identified in training, rGPV,AP was 0.18 and 0.12, respectively. Weaker correlations were obtained when the most informative SNP or all of the SNPs from the top 1% windows were used for prediction (0.01 and 0.06, respectively). Activity 1.2. Statistical assessment of common sources of variation that influence age at puberty and reproductive longevity. GWAS for AP and lifetime number of parities (NP) uncovered regions that have pleiotropic effects and influence both traits. One of these regions includes arginine vasopressin recptor 1A (AVPR1A). Sequencing AVPR1A in crossbred gilts exhibiting AP early (134-149 d, n=8) and late (219-243 d, n=8) uncovered three non-synonymous SNPs (G31E, G256D and K377Q). SNP G256D is located in the third intracellular loop of AVPR1A and was in complete linkage disequilibrium with G31E, located in the extracellular NH2-terminus. The NH2-terminus has a role in agonist binding and intracellular signaling. The SNP K377Q is located at the C terminus known to be involved in coordinating protein interactions with AVPR1A. A 0.19 difference in allelic frequency was observed between gilts that expressed AP early and late for G31E and G256D compared to 0.06 for K377Q. The frequency of the favorable allele A from G31E increased from 0.42 in the group of gilts that were not able to generate a litter to 0.54 in the sows that generated 3 parities. The GG genotype of SNP G31E was associated with a 4.6 d earlier expression of AP compared with genotype AA (P < 0.05) and 3.0 d earlier expression compared with genotype AG (P < 0.08). The GG genotype was also associated with 0.51 more lifetime parities than AA genotype (P < 0.006) and 0.31 more parities than AG genotype (P < 0.06). Irrespective of the expression of AP the sows with the GG genotype had a higher probability to generate first and second parity compared with the sows with AA and AG genotype (P <0.05). Irrespective of the expression of first estrus, the sows with the GG genotype had a higher probability to generate first and second parity compared with the sows with AA and AG genotype (P <0.05). Activity 1.3. Genome sequencing. Semen was collected and DNA was extracted from all sires that generated the gilts from the last 4 replicates. The release of the large Proton sequencing chip that will allow us to sequence sires of both ends of the distribution for age at puberty and reproductive longevity at high density and low cost was delayed for Spring 2014. Activity 1.4. Assess gene expression profile in the hypothalamus. Hypothalamic, pituitary, ovarian and granulossa cells were collected from gilts that expressed and gilts that did not express estrus by 240 d of age. Gene expression of the main candidate gene, AVPR1A, was assessed by Real-Time qPCR. Several genome prediction approaches were tested to classify experimental gilts as early, late and gilts that most likely will not express estrus before 240 days of age. Objective 2. A mixed model analysis of phenotypic variation was employed to estimate the response of energy restriction during gilt development and genotype on age at puberty and reproductive longevity. Animals that expressed puberty early and under restricted diet tend to have higher probability to generate first parity (P<0.45), parity 2 (P <0.05) and parity 3 (P<0.05). The interaction between the genotypes of AVPR1A and diet was found to influence the probability of the gilts to generate parity 2 (P<0.10) and 3 (P<0.05). Objective 3. The working hypothesis of this aim is that major genetic variants that explain significant variation in age at puberty and reproductive longevity will be expressed in multiple commercial populations, and can be used to generate a selection response and economic gain. The technologies generated by the research component of the grant will be evaluated in demonstration projects tailored to major State and National producers, such as Pillen Family Farms, The Maschhoffs and Smithfield. Tissues were collected from ~12,000 gilts from Smithfield and ~3,000 gilts from The Maschhoffs. Collection of traditional commercial phenotypes with a focus on fertility traits is in progress. Construction of the website that will provide details about the research and extension components of the grant and disseminate the results is in progress (http://liferaydemo.unl.edu/web/anisci/sow-reproductive-longevity) and will be public soon.

Publications

  • Type: Journal Articles Status: Published Year Published: 2013 Citation: Tart J.K., R.K. Johnson, J.W. Bundy, N.N. Ferdinand, A.M. McKnite, J.R. Wood, P.S. Miller, M.F. Rothschild, M.L. Spangler, D.J. Garrick, S.D. Kachman, D.C. Ciobanu (2013) Genome-wide prediction of age at puberty and reproductive longevity in sows, Animal Genetics, Aug;44(4):387-97. doi: 10.1111/age.12028.
  • Type: Conference Papers and Presentations Status: Submitted Year Published: 2014 Citation: Lucot K.L, M.L. Spangler, S.D. Kachman, and D.C. Ciobanu (2014) Genomic prediction of age at puberty in sows using Bayesian methods, ASAS Midwest Meeting, March 17-19, 2014, Des Moines, IA.
  • Type: Conference Papers and Presentations Status: Submitted Year Published: 2014 Citation: Trenhaile M.D., K.L. Lucot, J.K. Tart, J.W. Bundy, J. F. Thorson, E.M. Keuter, J.R. Wood, M.F. Rothschild, G.A. Rohrer, P.S. Miller, M.L. Spangler, C.A. Lents, R.K. Johnson, S.D. Kachman, and D.C. Ciobanu (2014) AVPR1A alleles are pleiotropic sources of the variation in age at puberty and reproductive longevity in sows, ASAS Midwest Meeting, March 17-19, 2014, Des Moines, IA.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2013 Citation: Daniel C. Ciobanu, Stephen D. Kachman, Clay A. Lents, and Timothy J. Safranski (2013) Translational Genomics for Improving Sow Reproductive Longevity, SSR Annual Meeting, Montr�al, Qu�bec, 2226 July 2013.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2013 Citation: Tart, J. K., Lucot, K. L., Bundy, J. W., Miller, P., Rothschild, M. F., Spangler, M., Garrick, D. J., Johnson, R., Kachman, S., Ciobanu, D. (2013) Genome-wide prediction of pleiotropic regions that influence age at puberty and reproductive longevity in sows. Plant and Animal Genome XXI Conference, San Diego, January 12 - 16, 2013.