Even with the amazingly complex network of mechanisms plants use to defend themselves against pathogens, 15% of the world’s total crop potential is lost to microbes. Strong artificial selection for favorable crop traits like growth rate or flavor, and population bottlenecks in the breeding process, tend to greatly reduce the allelic diversity within crop cultivars and leave crops vulnerable to pathogens. The fields of plant genomics and breeding are rapidly advancing to address these issues, however, for plant species with long generation times, there is a unique need for strategies to accelerate breeding progress. This project will establish a new scientific approach for use with perennial crop plants to identify candidate loci for disease resistance using a model tree crop, Theobroma cacao (the chocolate tree). Whole genome re-sequencing and transcriptome sequencing of a core collection of highly diverse cacao genotypes will provide the genetic information necessary to drive the discovery of genes critical for pathogen resistance. High throughput functional analysis of these genes will test their role in resistance and set the stage for future translation of these basic findings to guide more efficient breeding programs utilizing a wider array of genetic diversity. Importantly, the methods, tools, and knowledge gained will be directly applicable to discovery of genes underlying important traits in other crops, especially heterozygous perennials such as trees and many grasses which are not particularly amenable to approaches developed for the major annual crops such as corn and soybean.
"Genotype Selections in the CATIE Collection and Tree Tagging" - Leaf samples were collected from trees at the La Montaña field site at the CATIE field station in Turrialba, Costa Rica. Plants used were tagged with barcodes and GPS coordinates were taken for each tree. We evaluated a subset of the cacao germplasm maintained at IC3 focused primarily on four genetic groups (Guiana, Iquitos, Marañon, and Nanay).
"SNP genotyping and identity verification of cacao clones" - Mature, stage E leaves [Plant Methods (2016) 12:1-13; Small Wonders Pept Dis Control (2012) 1905. doi: 10.1021/bk-2012-1095.ch018] were collected into plastic, resealable bags with desiccating silica gel beads and transported to the USDA-ARS in Beltsville, Maryland for genotyping analysis. DNA was extracted as previously described, and genotyping was performed using a set of 90 SNPs with a 96.96 Dynamic Array™ IFC (Fluidigm, San Francisco, CA). These SNPs were selected from a larger set of 1560 SNPs developed from EST sequences [BMC Genomics (2008) 9:512], and have been used previously to analyze cacao genetic ancestry [Genetic Resources and Crop Evolution (2013) 60:441-53; Journal of Agricultural and Food Chemistry (2014) 62:481-7; Tropical Plant Biology (2014) 7:133-43; Tropical Plant Biology (2018) 11:93-106].
Genotypic data was analyzed to validate identity and determine genetic membership of each sampled tree. First, the clonality (or intra-clone mislabelling) among the multiple individual trees was verified using pairwise multi-locus matching, as implemented in the computer program GenAlex 6.503 [Molecular ecology notes (2006) 6:288-95; Bioinformatics (2012) 28:2537-9]. Samples with SNP profiles fully matched at all genotyped SNP loci were declared the same genotype. For a subset of cacao accessions, reference genotype data is available, which was generated from the cacao trees maintained in Marper Farm, Trinidad. These trees have been used by cacao researchers as the original trees for most of the Upper Amazon Forastero germplasm [Cacao Diseases: Springer; 2016. pp. 3-31; Genetic Diversity of Cacao and Its Utilization 2005. pp. 1-341]. Genotype data of trees sampled at CATIE were compared to these reference data, in all cases where it was available. The genetic integrity of the experimental accessions was also assessed by checking their population memberships. An assignment test for the experimental clones was performed using model-based Bayesian cluster analysis software STRUCTURE v2.3.4 [Genetics (2000) 155:945-59]. The analysis included SNP sets representing 10 distinctive cacao germplasm groups, which served as references [PLoS ONE (2008) 3(10):e3311].
"Sample collection, library preparation, and sequencing" - We acquired whole-genome and transcriptome data from 31 individual plants representing four populations: Guiana (n= 8), Marañón (n = 8), Nanay (n= 7) and Iquitos (n= 8). Leaf tissue was collected from trees maintained in the International Cacao Collection at Centro Agronómico Tropical de Investigación y Enseñanza (CATIE), Costa Rica. Two healthy and mature (stage E) leaves were collected from each genotype, flash frozen, and shipped to Penn State University under permits and IP agreements involving Costa Rica, USDA, and Penn State University. Samples were ground to a fine powder under liquid nitrogen. DNA was extracted by taking ground tissue powder at low input (0.25 g / 15 ml lysis buffer) through a CTAB isolation based on [Anal. Biochem. (2003) 315:85–9] with the following modifications: the lysis buffer contained 4% PVP; the isopropanol precipitation was carried out at –20°C overnight; pellets from two isolations were combined during the second pellet wash (using a flamed Pasteur pipet to transfer pellets to a fresh 50 ml tube) and the day 2 clean up steps were carried out in 50 ml tubes; a proteinase K digest (20 ml / ml Qiagen #19131, 2 hours at 50°C) was added prior to the phenol:chloroform:isoamyl alcohol extraction; final pellets were resuspended in 150 ml TE. RNA was extracted from the ground tissue according to the PureLink Plant RNA Reagent protocol (Thermo Fisher Scientific), with minor modifications as previously described in [Plant Mol. Biol. (2019) 99:499–516]. After extraction, RNA samples were treated with RNase-free DNase (Thermo Fisher Scientific) according to manufacturer’s instructions, purified with an additional ethanol precipitation, and resuspended in RNase-free water. Initial measurements of concentration and quality were done with Qubit and Nanodrop spectrophotometers, respectively. Library preparation was performed at the Pennsylvania State University Genomics Core Facility. Illumina TruSeq DNA PCR-Free High Throughput kit and Illumina TruSeq Stranded mRNA Library kit were used to prepare the DNA and RNA libraries, respectively. Manufacturer’s protocols were followed in both cases (Illumina). The concentration of each library was determined by qPCR using a Kapa Library Quantification kit according to the manufacturer’s instructions (Kapa Biosystems). Pooled libraries were normalized and denatured with 0.2 N NaOH for a loading concentration of 375 pM on two Illumina NovaSeq S2(300) flowcells at the Pennsylvania State Hershey Medical Center. Cluster amplification of denatured templates and 150 bp paired-end sequencing was performed according to the standard Illumina NovaSeq S2 protocol. Less...