Source: TEXAS A&M UNIVERSITY-CORPUS CHRISTI submitted to
BUILDING FOUNDATIONS FOR GENETIC IMPROVEMENT OF EASTERN OYSTER IN TEXAS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1026296
Grant No.
2021-70007-34783
Cumulative Award Amt.
$320,000.00
Proposal No.
2020-06947
Multistate No.
(N/A)
Project Start Date
Sep 1, 2021
Project End Date
Aug 31, 2024
Grant Year
2021
Program Code
[AQUA]- Aquaculture Research
Project Director
Hollenbeck, C.
Recipient Organization
TEXAS A&M UNIVERSITY-CORPUS CHRISTI
6300 OCEAN DR STE NRC 2011
CORPUS CHRISTI,TX 78412-5503
Performing Department
Life Sciences
Non Technical Summary
Due to flattening trends in capture fisheries, aquaculture is expected to be the primary source of growth in seafood production over the next several decades. In an effort to relieve the strain on wild fisheries, the state of Texas recently approved the development of a legal framework for the aquaculture of oysters, which presents a considerable opportunity to expand domestic seafood production. A key factor in the success of the oyster aquaculture industry in Texas will be access for farmers to robust, fast-growing oysters that can thrive in a variety of local environmental conditions. Unlike most agricultural species, oysters available for aquaculture in Texas are undomesticated, meaning that their performance for key traits such as growth rate and meat yield are no different from wild stocks. Selective breeding, enhanced with cutting-edge genomic information, is a well-proven strategy for genetic improvement of aquaculture species. The overall objective of this project is to develop critical information to support the development of sustainable, long-term selective breeding programs for oyster aquaculture in Texas. Specifically, the objectives of the project are to 1) generate critical information, such as the heritability of economically-important traits such as growth and meat yield, for two distinct sub-populations of oysters native to the Texas coast, and 2) determine the feasibility of cutting-edge, genomics-based breeding strategies for oyster breeding programs. Results of the project will inform the design of breeding programs and elite oysters identified will form the basis of future breeding populations. The project will lead to efficient, long-term breeding programs that will ultimately provide farmers with oyster seed with improved production traits, leading to a reduction in risk and an increase in farm profitability. The further refinement of genomics-based breeding strategies will improve breeding of oysters and other bivalves in the wider United States and worldwide, resulting in more efficient and sustainable production of seafood.
Animal Health Component
100%
Research Effort Categories
Basic
0%
Applied
100%
Developmental
0%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
30337231080100%
Knowledge Area
303 - Genetic Improvement of Animals;

Subject Of Investigation
3723 - Oysters;

Field Of Science
1080 - Genetics;
Goals / Objectives
The project responds to the Program Area Priority "Genetics of commercial aquaculture species" through the development of genomic tools, information, and establishment of broodstock lines important for selective breeding of eastern oyster in Texas and elsewhere. The project will have a direct impact on commercial aquaculture in the United Stated by developing the foundations for breeding robust, fast-growing oysters to serve the newly initiated Texas oyster aquaculture industry. The overall goal of this project is to build a foundation for modern, genomics-based breeding of eastern oyster in Texas and the wider U.S. To that end, the primary objectives of the proposed research are to 1) measure in two sub-populations genetic parameters such as heritability and genetic correlations for key traits which critically factor into long-term breeding programs, and 2) assess the potential for incorporating genome-wide SNP information into the calculation of breeding values (genomic estimated breeding values; GEBV) to drive genetic gain in valuable traits.
Project Methods
Sampling and generation of experimental populationsApproximately 50-75 wild C. virginica will be sampled from each of two locations along the Texas coast: Lower Laguna Madre and Galveston Bay. These localities represent both sub-populations identified by Anderson et al. (2014) and are at either end of the known temperature/salinity gradient. The collected individuals will be taken to the TAMAR facility in Flour Bluff, TX, conditioned and strip-spawned to produce experimental F1 families. While strip-spawning will require sacrifice of the collected parents, it will allow more precise control over parental contribution as compared to mass spawning and will result in a more even distribution of family sizes. At least 30 families per locality will be created using a nested-cross design, whereby the sperm from one male will be mixed with eggs from three different females in separate containers. The nested-cross design will control for maternal effects in mixed model approaches used to estimate genetic parameters and predict breeding values. The methods of Scarpa and Allen (1992) and Allen et al. (1993) will be followed for the production oyster larvae. In brief, gametes from each oyster will be collected by dissection and placed in individual beakers; care will be taken not to cross contaminate gametes by rinsing instruments between uses. At this time, the sex of each parent will be recorded, and a tissue sample taken for genetic analyses. Egg number will be equalized for each female, and then divided into groups for fertilization by sperm from different males (Burford et al., 2014). This maximizes the potential genetic input from parents into offspring of each population. Sample collection and genotypingAfter six months, approximately 1000 progeny per locality will be measured for four traits: shell height, shell width, shell length, and meat yield, following standard procedures (Galtsoff, 1964), and a tissue sample from each individual will be taken and preserved in 20% DMSO buffer. High-quality DNA will be extracted from all parents and progeny, using a Mag-Bind Blood and Tissue DNA Extraction Kit (Omega Bio-tek). Genotyping will be conducted using two platforms (ddRAD and amplicon sequencing), that will allow cost-effective, high-density genotypes to be recorded on all individuals through genotype imputation. In this strategy, parents will be genotyped at 3000-4000 SNP loci with double-digest RAD sequencing (ddRAD) and sequenced on an Illumina HiSeq 4000 DNA sequencer. Bioinformatic processing and SNP filtering will be conducted with the dDocent pipeline (Puritz et al., 2014), a pipeline for processing ddRAD data which was developed in the PI's laboratory. All offspring (~2000; 1000 per locality) will be genotyped at low-density (200-300 SNP loci) using a high-throughput amplicon sequencing technique (GT-Seq; Campbell et al. 2015). The GT-Seq SNP panel is currently under development in the PIs laboratory and consists of a subset of the loci reliably genotyped by ddRAD. Using a combination of high-density parental genotypes from ddRAD and low-density GT-seq genotypes from offspring, genotype imputation techniques will be used to accurately predict the high-density genotypes in the offspring, allowing genomic analyses on all 2000+ individuals. The final set of filtered, imputed genotypes will be used for all downstream genetic analyses. Estimation of genetic parameters and evaluation of genomic predictionGenetic parameters (heritability and genetic correlations) will be calculated for the three traits using two methods. First, a pedigree-based method will utilize data from individuals measured for all traits. Parentage of these individuals will be established using standard genetic parentage assignment techniques. The pedigree-based approach will employ linear mixed models with fixed effects of spawn date and rearing environment and random effect of individual and dam (a standard animal model), as implemented in the program AIREMLF90 (Misztal et al., 2002). Heritability estimates will involve univariate mixed model equations, while estimates of genetic correlations will involve bivariate models for each combination of traits. A second method will utilize a genomic relationship matrix calculated using the high-density genotypes obtained by ddRAD. The genomic relationship matrix will be calculated following the method of VanRaden (2008), using the GenABEL package in R (Aulchenko et al., 2007), and genetic parameters will be calculated using the BLUPF90 suite of programs. Estimates of genetic parameters for each trait will be compared between the northern and southern sub-populations.Breeding values for each individual will be calculated using the mixed model procedure described above with both the pedigree-based method and the genomic method in BLUPF90. The accuracy of breeding value prediction for both methods will be calculated using a five-fold cross validation approach, following Yoshida et al., (2018). The relative accuracy of the genomic vs the pedigree-based approach can be used to enumerate the benefits of genomic prediction of breeding values over traditional approaches.Genome wide association analysisGenome-wide association (GWAS) analysis for discovery of SNP-containing loci associated with each trait will be performed using the family-based score test, as implemented in the mmscore function in the GenABEL R package. To account for potential bias caused by population structure, principal components analysis (PCA) will be conducted on the genotype data using the R package adegenet, and the top principal components will be included in the mixed model analysis. SNP-containing loci significantly associated with trait phenotypes will be mapped to the eastern oyster genome to identify candidate genes. The program SnpEff will be used to annotate SNPs into functional categories (within coding region, introduces premature stop codon) for prioritizing significant SNPs for future functional validation.

Progress 09/01/21 to 08/31/24

Outputs
Target Audience:The target audience for this project includes oyster producers, including growers and operators of nurseries and hatcheries, in Texas and the wider Gulf of Mexico, as well as researchers and industry professionals involved in oyster breeding. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has directly supported a research technician, which has contributed to hands-on training in oyster husbandry through all life stages. The project has also continued to provide opportunities for undergraduate students to participate in oyster hatchery operations. A major accomplishment of the project is the development of an overall strategy for implementing a cutting-edge oyster breeding program, which has resulted in documented best practices and protocols that will be adapted into training material for future students and breeding program personnel. Because of this, the training provided by the project has been incredibly valuable to the long-term development of the new oyster aquaculture industry in Texas. How have the results been disseminated to communities of interest?PI Hollenbeck has been in direct contact with current and prospective oyster growers and hatchery operators in the state throughout the course of the project. The Texas A&M AgriLife Mariculture research facility regularly hosts tours involving students, regulators, propective growers and hatchery/nursery operators, and other groups, in which overall goals of the project (selective breeding for trait improvement of Texas oysters) are communicated. Over the course of the project, PI Hollenbeck was an invited panelist and speaker for two years at the Texas Oyster Roundup, a food festival and oyster conservation fundraiser with participants from the oyster aquaculture industry, conservation groups, and the general public, and planned and spoke at a conference "Overcoming Barriers to Expanding Aquaculture in Texas", an event (>150 participants, including producers, researchers, students, and USDA/NOAA personnel) hosted by Texas A&M Univeristy, Texas A&M University - Corpus Christi, and USDA ARS to discuss the opportunities and challenges facing Texas aquaculture. Scientific results have been communicated at national conferences, research symposia, and through direct communication with collaborators at other institutions involved in oyster breeding, including USDA ARS, and will also be disseminated through peer-reviewed publications that are in preparation. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Objective 1: measure in two sub-populations genetic parameters such as heritability and genetic correlations for key traits which critically factor into long-term breeding programs Major activities The project focused on three primary traits: whole weight, meat weight, and the presence of a "hook", a shell deformity. To measure these values, test breeding populations were formed for two sub-populations, grown to market size, and subsets of individuals were measured for each trait and genotyped to reconstruct the pedigree. Parent oysters derived from a combination of wild oysters and first generation individuals (F1). Wild oysters were collected from Galveston Bay, Matagorda Bay, and the Laguna Madre. Progeny were produced in Nov. 2022 at the TAMU AgriLife Research Mariculture Laboratory in Corpus Christi, Texas. Northern groups were produced by crossing (via strip spawning) oysters derived from north Texas (Galveston Bay and Matagorda Bay), and southern groups by crossing oysters derived from south Texas (Laguna Madre). After hatchery and nursery rearing (6 mm size) each group was split and deployed to two grow-out sites: a TAMUCC research farm in Corpus Christi Bay, TX, and a commercial oyster farm in Copano Bay, TX. Oysters were deployed to both sites in Nov. of 2022 in 4 mm mesh bags contained in floating cages . Oysters were collected from both field sites in March of 2023 for physical tagging and non-lethal sampling of DNA and were then deployed back to their site of origin. In fall of 2023, oysters were collected from field sites for measurement. Oysters were first cleaned of any fouling organisms on their shell. They were then weighed to measure live weight (herein, weight), qualitatively assigned a rating on the shape of their umbo (hook), and then shucked for meat weight. The hinge was graded for presence or absence of bending backwards toward the dorsal shell. Oysters were genotyped using a custom GT-seq panel. The panel consists of primers targeting 420 informative microhaplotypes for C. virginica. Libraries were developed for hundreds of oysters at a time through Genotyping-in-Thousands by sequencing (GT-seq) methods (Campbell et al., 2015) at the Marine Genomics Lab at Texas A&M University - Corpus Christi. Samples were sequenced on an Illumina NovaSeq X, and sequences were processed using a custom bioinformatics pipeline to assign genotypes for each individual. A separate northern and southern pedigree was produced based on the origin of parents. First, candidate parents (those used to produce oysters in the experiment) were assigned to genetic background based on how they clustered on a PCA. Offspring were assigned to parents using the software Colony and APIS. Genetic parameters were estimated separately for northern and southern oysters via Best Linear Unbiased Prediction (BLUP) with the pedigrees, using linear mixed models in ASReml-R and the BLUPF90 suite of programs. Data collected and results Mean weight and meat weight were largest in northern oysters from Copano Bay. The proportion of oysters without a hinge bending backwards was lower in northern oysters than southern oysters. For oysters of northern origin, 5 unique dams, 13 unique sires, and 18 full-sib families were identified from oysters included in statistical analysis. For southern oysters, 26 unique dams, 16 unique sires, and 88 full-sib families were identified. For northern oysters, heritability of weight was 0.56 (0.34, SE), heritability of meat weight was 0.63 (0.47), and heritability of hook was 0.16 (0.33). For southern oysters, heritability of weight was 0.54 (0.15), heritability of meat weight was 0.04 (0.07), and heritability of hook was 0.29 (0.17). Differences in heritability of meat weight may be attributable to differences in the timing of spawning between the two populations relative to the time of assessment. Genetic correlations were more difficult to assess due to issues with model convergence that need further investigation, although initial estimates of the genetic correlation between total weight and meat weight were high for both populations (0.88 northern and 0.97 southern). Objective 2: assess the potential for incorporating genome-wide SNP information into the calculation of breeding values (genomic estimated breeding values; GEBV) to drive genetic gain in valuable traits. Major activities To assess the potential benefits of genomic selection, 68 individual broodstock parents of evaluation individuals were genotyped with whole-genome shotgun sequencing (WGS). Genotypes were called with GATK and filtered for quality score, depth, and minor allele frequency. To mimic a realistic high-density genotyping platform, a subset of 5K biallelic SNPs were randomly selected for assessment of genomic prediction. Progeny genotypes (GT-seq, described above) were imputed to the higher density (5K) using FImpute3. To assess the benefits of genomic prediction, genomic architecture for each trait was assessed with genome-wide association analysis (GWAS), conducted using the R package gaston, and breeding values (and associated genomic prediction accuracy) were calculated for each trait/population using a GBLUP-based cross-validation approach implemented in BLUPF90. Data collected and results GWAS revealed a polygenic genetic architecture for the traits assessed, suggesting that genomic selection is likely a more effective strategy than marker-assisted selection for these traits; however, a small number of candidate trait-associated SNPs have been identified for future investigation, and the whole-genome data collected will facilitate this work in the future. GBLUP was shown to be effective in improving breeding value accuracy for the northern population (7-23% increase compared to pedigree-based methods), but was less effective in the southern population. This may be the result of unaccounted-for admixture in the southern population and may be improved by fine-tuning genomic prediction models for this population. This will be the subject of future research for the breeding program. Key outcomes (Objective 1 and 2) The project has had significant impacts on the development of optimal strategies for a breeding program for eastern oysters in Texas. Genetic parameter estimation for key traits such as weight, meat weight, and shell deformity (hook) indicates that the traits can be reliably improved with selective breeding, and differences in heritability observed between populations, particularly for meat weight, highlight need for population-specific breeding criteria. The project also demonstrated the potential for integrating genomic selection into the breeding program. The strategy of combined low and high density genotyping with imputation was shown to be highly cost-effective and feasible from a technical standpoint, and it is clear that the use of genomic data for future genetic improvement of eastern oysters in Texas is a valuable approach. Various procedures necessary for genomics-informed breeding of oysters, from non-lethal DNA collection to sequencing to data analysis, were tested and will be used to improve standard operating procedures for the breeding program. An overall accomplishment of the project is development of a wealth of the technical and scientific knowledge necessary for cutting-edge oyster breeding, which will support the growing oyster aquaculture industry in the state and contribute to long-term increases in oyster aquaculture productivity.

Publications


    Progress 09/01/22 to 08/31/23

    Outputs
    Target Audience:The target audience for this project includes oyster producers, including growers and operators of nurseries and hatcheries, in Texas and the wider Gulf of Mexico, as well as researchers and industry professionals involved in oyster breeding. Changes/Problems:No major changes or problems were encountered during the project period. Minor challenges related to hatchery and field mortality, as well as loss of individual ID tags due to high levels of wave action at field sites during grow-out. The effect of these are expected to be minor, but may result in reduced sample size in the final dataset relative to initial expectations. What opportunities for training and professional development has the project provided?This project has directly supported a research technician, which has contributed to hands-on training in oyster husbandry through all life stages. The project has also continued to provide opportunities for undergraduate students to participate in oyster hatchery operations. The training provided by the project has been incredibly valuable to the development of the new oyster aquaculture industry in Texas. How have the results been disseminated to communities of interest?PI Hollenbeck has been in direct contact with current and prospective oyster growers and hatchery operators in the state throughout the course of this reporting period. The Texas A&M AgriLife Mariculture research facility regularly hosts tours involving students, regulators, propective growers and hatchery/nursery operators, and other groups, in which overall goals of the project (selective breeding for trait improvement of Texas oysters) are communicated. PI Hollenbeck also was an invited panelist at the first annual Texas Oyster Roundup, a food festival and oyster conservation fundraiser with participants from the oyster aquaculture industry, conservation groups, and the general public. What do you plan to do during the next reporting period to accomplish the goals?The next project period will be focused on finalizing low- and high-density genotyping for the experimental populations and for data analysis. Genotype data will be used to reconstruct the pedigree for traditional pedigree BLUP calculation of breeding values. The genotype data will also be used for the calculation of genomic estimated breeding values. Genome-wide association analysis will be conducted to characterize the genetic architecture of key traits measured in this project period. The next project period will also focus on outreach activities and preparing results for publication.

    Impacts
    What was accomplished under these goals? In this project period, major accomplishments include completing growout and collecting trait data (growth, shape, meat yield, and deformities) for groups composed of broodstock collected from both Texas sub-populations. Other accomplishments include finalizing a genotyping stategy and initiating genotyping, which will allow for genomic prediction of breeding values.

    Publications


      Progress 09/01/21 to 08/31/22

      Outputs
      Target Audience:The target audience for this project includes oyster farmers and nursery/hatchery operators, as well as public and private groups involved with oyster breeding in Texas and the wider Gulf of Mexico. p { margin-bottom: 0.1in; direction: ltr; line-height: 115%; text-align: left; orphans: 2; widows: Changes/Problems:No major changes or problems were encountered during the first project period. Minor challenges related to spawning and larval rearing included 1) issues with unexplained larval mortalityof experimental batches in Spring/Summer of 2022 and 2) challenges associated with simultaneously spawning two genetically distinct populations (North/South Texas) with different seasonal reproductive timing. Despite these challenges, experimental batches were produced in Spring and Fall of 2022, and it is anticipated that this delay will not prevent project objectives from being accomplished. What opportunities for training and professional development has the project provided?This project, through direct support of a research technician, has contributed to workforce development in the nascent oyster aquaculture industry in Texas. During this reporting period, the project has also allowed several undergraduate students to learn and directly participate in hatchery culture of oysters, including the processes of conditioning broodstock and larval and juvenile oyster husbandry. This training, as well as the awareness of aquaculture career paths, is incredibly valuable to the development of this new industry. How have the results been disseminated to communities of interest?PI Hollenbeck has been in direct contact with current and prospective oyster growers and hatchery operators in the state throughout the course of this reporting period, and progress towards the stated project objectives have been communicated in this way. PI Hollenbeck also attended a Texas SeaGrant-sponsored session entitled: Texas Cultured Oyster Mariculture Vision Planning in December 2021. This session involved stakeholders from across the oyster industry, including farmers, regulators, hatchery and nursery operators, and academics, to discuss the current state of the industry and to identify knowledge gaps that are needing to be addressed to advance the industry. What do you plan to do during the next reporting period to accomplish the goals?In the next reporting period, experimental animals (progeny) produced during this reporting period will be grown to the appropriate size, measured for growth, meat yield, and morphological traits, and will also be genotyped at 200-300 SNP loci. Parental genotypes will be obtained through high-density genotyping, and offspring genotypes will be imputed to the higher density genotypes of the parents. Genomic analyses will be conducted to estimate important genetic parameters such as trait heritability, as well as to evaulate the efficacy of genomic selection strategies.

      Impacts
      What was accomplished under these goals? For the first year of the project, the major accomplishment was producing the experimental animals for grow-out. Once these animals reach the size required for assessment of trait performance and genotyping, the primary objectives of the research project (measurement of genetic parameters and assessment of genomic selection) can be conducted.

      Publications