Multiplexed Fragaria Chloroplast Genome Sequencing

A method to sequence multiple chloroplast genomes using ultra high throughput sequencing technologies was recently described. Complete chloroplast genome sequences can resolve phylogenetic relationships at low taxonomic levels and identify informative point mutations and indels. The objective of this research was to sequence multiple Fragaria chloroplast genomes using the Illumina Genome Analyzer. Sixty-three PCR fragments from 22 species were sequenced in four multiplex sequencing runs. Plastome sequences were assembled using a combination of de novo and reference guided assembly with F. vesca ‘Hawaii 4’ providing the reference. De novo assembly resulted in plastome coverage of 43-74%, and reference guided assembly increased the plastome coverage to 76-82%. The alignment of sequenced chloroplast genomes was 96,438 bp and contained 319 parsimonyinformative sites which will be useful in identifying chloroplast genome regions of high sequence variation for testing evolutionary relationships. INTRODUCTION Strawberry, Fragaria L., is in the Rosoideae subfamily in Rosaceae (Potter et al., 2007). The genus Fragaria was classified in the Fragariinae subtribe, Potentilleae tribe in the Rosodae superfamily. Fragaria species include thirteen diploids (2n=2x=14), five tetraploids (2n=4x=28), one hexaploid (2n=6x=42), four octoploids (2n=8x=56) (Staudt, 2008) and one decaploid (2n=10x=70) (Hummer et al., 2009). F. iturupensis, whose octoploid forms were previously reported (Staudt, 1989; Staudt and Olbricht, 2008), is also represented by decaploid forms (Hummer et al., 2009), which constituted the first reported naturally occurring decaploid strawberry species. Resolution of Fragaria phylogeny is not only useful for identification of sources of useful genes (Bringhurst and Voth, 1984; Lawrence et al., 1990), but also for verification of current species designations by inference from predicted species relationships. The need to verify Fragaria species designations became evident following flow cytometry (Hummer and Bassil, 2008; Hummer et al., 2009), simple sequence repeat (SSR) data analysis (Njuguna and Bassil, 2008), and species introductions and revisions to the strawberry germplasm collection at the USDA/ARS/NCGR in Corvallis, Oregon by strawberry expert Günter Staudt (1928-2008). In Fragaria, phylogenetic analyses have used chloroplast (Harrison et al., 1997) and nuclear genome sequences (Potter et al., 2000; Rousseau-Gueutin et al., 2009), but relationships remain unclear. The suggested Fragaria octoploid genome models AAA’A’BBB’B’ (Bringhurst, 1990) and YYY’Y’ZZZZ/YYYYZZZZ (Rousseau-Gueutin et al., 2009), suggest the contribution of two to four diploids to the octoploid genome. The diploid donor species are still not known, but evidence based on grouping of octoploid Proc. IS on Molecular Markers in Horticulture Eds.: N.V. Bassil and R. Martin Acta Hort. 859, ISHS 2010 316 and diploid nuclear genes (ADH, alcohol dehydrogenase; GBSSI-2 or Waxy; and DHAR, dehydroascorbate reductase) points to F. vesca L., F. mandschurica Staudt sp. nova and F. iinumae Makino (Davis and DiMeglio, 2004; Harrison et al., 1997; Potter et al., 2000; Rousseau-Gueutin et al., 2009; Senanayake and Bringhurst, 1967) as possible contributors. Based on low copy nuclear gene sequences, Rousseau-Gueutin et al. (2009) grouped Fragaria diploids into three clades. However, the diploid species within clade X were poorly resolved and the placement of F. bucharica remains unclear. The use of nuclear genes for phylogenetic analysis is complicated by polyploidy and recombination (Nishikawa et al., 2002), making the chloroplast genome an attractive option for Fragaria. Further exploitation of the chloroplast genome in Fragaria for phylogenetic analysis is warranted due its small size, non-recombinant nature, and high sequence conservation, all factors that reduce the complexity of analysis and interpretation of results. For efficient use of limited chloroplast sequence divergence, large-scale sequencing is required, and is now possible with high throughput sequencing platforms. Sequencing of multiple small genomes to high coverage depth using high-throughput sequencing platforms was recently demonstrated (Cronn et al., 2008). In this study, we used multiplex sequencing to uncover sequence divergences in the chloroplast genomes of Fragaria species. MATERIALS AND METHODS Plant Material and DNA Extraction Twenty-one of 22 wild Fragaria species, and one Potentilla villosa accession, a close relative of Fragaria in the Rosaceae family, that have been preserved at the germplasm repository in Corvallis (Table 1) were included in the study. DNA was extracted from actively-growing leaves using a modified protocol based on the PUREGENE kit (Gentra Systems Inc. Minneapolis, MN). The DNA samples were cleaned further with Nucleon PhytoPure resin, a component found in the illustra Nucleon Phytopure kits for plant and fungal DNA extraction (GE Healthcare UK Limited, Buckinghamshire, UK). DNA quality and quantity was analyzed with a Wallac 1420 VictorV microplate reader (PerkinElmer, Waltham, MA). DNA concentration was adjusted to 3 ng/μl for PCR. Primer Selection, Design and PCR A total of 203 chloroplast primers (108 forward, 95 reverse) were screened in various logical combinations in four species to identify primer pairs that amplify >2500 bp fragments and provide maximum coverage of the chloroplast genome. Where possible, primer pairs that amplified single bands and that amplified in most or all of the species were chosen. Of the 203 primers, 141 had been used to amplify the Cucumis sativus L. chloroplast genome (Chung et al., 2007), 25 were designed from the complete genome sequence of Morus indica ‘K2’ (Ravi et al., 2006), and 36 were designed from the nearly complete chloroplast genome of F. vesca cv. Hawaii4 (http://strawberry.vbi.vt.edu/tikiindex.php). Sixty-three primer pairs were chosen to amplify the entire chloroplast genomes of 21 Fragaria species and one Potentilla accession (list available upon request from corresponding author). Long-range PCR was carried out using Phusion HighFidelity DNA polymerase (New England Biolabs, Ipswich, MA). Amplifications were performed in 10 μl total reaction volumes containing 5x Phusion GC buffer, 2.5 mM of each dNTP, 10 μM of each primer, 5 U of Phusion DNA polymerase, 0.05 μl of 3% DMSO and 5 ng of DNA template. PCR product quantification was carried out using the Quant-iT PicoGreen dsDNA quantification protocol (Molecular Probes, Inc. Eugene, OR) manufacturer’s specifications. Equimolar amounts of PCR products were pooled for each species to generate 1-5 μg of chloroplast DNA which was prepared for sequencing using the Illumina sample preparation kit (Illumina Inc., San Diego, CA) and the modifications of Cronn et al. (2008). Each PCR pool was sheared using nitrogen from a