Phylogenetic Relationships Among Low Ploidy Poa Species Using Chloroplast Sequences
Total Page:16
File Type:pdf, Size:1020Kb
Genome Phylogenetic relationships among low ploidy Poa species using chloroplast sequences Journal: Genome Manuscript ID gen-2016-0110.R1 Manuscript Type: Article Date Submitted by the Author: 17-Oct-2016 Complete List of Authors: Joshi, Alpana; USDA-ARS, FRRL Bushman, Shaun; USDA-ARS, FRRL Pickett, Brandon; USDA-ARS, FRRL Robbins, Matthew;Draft US Department of Agriculture, Agricultural Research Service, Forage and Range Research Laboratory, Staub, Jack; USDA ARS, Forage & Range Research Laboratory Johnson, Paul; Utah State University, Plant, Soils, and Climate Keyword: Poa, flow cytometry, phylogeny, Kentucky bluegrass https://mc06.manuscriptcentral.com/genome-pubs Page 1 of 62 Genome 1 Phylogenetic relationships among low ploidy Poa species using chloroplast sequences Alpana Joshi, B. Shaun Bushman *, Brandon Pickett, Matthew D. Robbins, Jack E. Staub, Paul G. Johnson Alpana Joshi, B. Shaun Bushman *, Brandon Pickett, Matthew D. Robbins, and Jack E. Staub. USDA-ARS Forage and Range Research Unit, 695 N 1100 E, Logan, UT 84322-6300. Paul G. Johnson . Department of Plants,Draft Soils, and Climate, Utah State University, 4820 Old Main Hill, Logan, UT 84322-4820. Corresponding author email: [email protected] Abbreviations: TLF is trnT-trnF chloroplast region Keywords: Poa , flow cytometry, phylogeny, Kentucky bluegrass https://mc06.manuscriptcentral.com/genome-pubs Genome Page 2 of 62 2 Abstract Species in the Poa genus are taxonomically and genetically difficult to delineate due to high and variable polyploidy, aneuploidy, and challenging breeding systems. Approximately 5% of the proposed species in Poa are considered to include or comprise diploids, but very few of those diploids are represented in seed collections. Recent phylogenetic studies of Poa have included some diploid species to elucidate Poa genome relationships. In this study we build upon that foundation of diploid Poa relationships with additional confirmed diploid species and accessions, and with additional chloroplast sequences. We also include a sample from P. pratensis and P. arachnifera to hone in on possible ancestral genomes in these two agronomic and highly polyploidy species. Relative to most of Draftthe Poa species, Poa section Dioicopoa (P. ligularis , P. iridifolia , and P. arachnifera ) contained relatively large chromosomes. Phylogenies were constructed using the TLF gene region and five additional chloroplast genes, and the placement of new species and accessions fit within chloroplast lineages reported in Soreng et al. (2010) better than by taxonomic subgenera and sections. Low ploidy species in the “P” chloroplast lineage, such as P. iberica and P. remota , grouped closest to P. pratensis . https://mc06.manuscriptcentral.com/genome-pubs Page 3 of 62 Genome 3 Introduction Poa is the largest genus of grasses, comprising 5 subgenera, 13 sections, and up to 500 species (Gillespie and Soreng 2005). The genus is taxonomically and genetically difficult to delineate, and interspecific hybridization, high and variable polyploidy, and facultative apomixis have been major sources of variation (Stebbins 1950; Clausen 1961; Tzvelev 1976; Hunziker and Stebbins 1987; Soreng 1990; Gillespie and Soreng 2005). Because of the high degree of polyploidy in the genus, it is believed that many Poa species originated through allopolyploidy (Stebbins 1950; Darmency and Gasquez 1997; Brysting et al. 2000, 2004; Patterson et al. 2005). Additionally, genomic redundancy and rampant aneuploidy among highly polyploid taxa suggests some degree of autopolyploidy. Many taxa within PoaDraft include a range of polyploid levels, from diploids to octaploids (Kelley et al. 2009) or tetraploids to duo-decaploids in the same species (Bowden 1961; Barkworth et al. 2003; Soreng 2005, 2007). With a basic chromosome number of x=7 (Gould 1968), Poa has been referred to as one large polyploid complex (Stebbins 1950). Given the challenging diversity in ploidy among and within species, flow cytometry has been used to quantify the DNA content in a several other Poa species (Eaton et al. 2004; Patterson et al. 2005; Kelly et al. 2009; Raggi et al. 2015). Sectional and infra-sectional taxonomy of Poa is complicated and has been the subject of major revisions by taxonomists over the years. (Tzvelev 1976; Hunziker and Stebbins 1987; Phillips 1989; Gillespie et al. 2007). Consideration of section Ochlopoa has switched back and forth between a subgenus and genus designation (Hylander 1953; Soreng 1990; Gillespie and Boles 2001; Bohling and Scholz 2003). Poa section Dioicopoa was initially considered as subgenus (Nicora 1977, 1978) but later it recognized as section within the Poa subgenus (Soreng 1998; Gillespie et al. 2007). Arctopoa was initially considered a section (Tzvelev 1964) or a https://mc06.manuscriptcentral.com/genome-pubs Genome Page 4 of 62 4 subgenus (Probatova 1971) of Poa , but then later recognized as a separate genus (Probatova 1974; Gillespie et al. 2008). When polyploid complexes within species are added to this taxonomic murkiness, which complexes can show substantial morphological differences (e.g., Speckmann and Van Dijk 1972), resolution of Poa taxa can in many cases be challenging. However, DNA sequence analysis, particularly from chloroplast regions, has been used to aid in delineation of Poa sections and species (Gillespie and Soreng 2005; Patterson et al. 2005; Gillespie et al. 2007; Raggi et al. 2015). Soreng et al. (2010) used the chloroplast TLF region (Taberlet et al. 1991) to classify diploid species into four of the Poa subgenera: Ochlopoa , Poa , Pseudopoa , and Stenopoa . The distribution of those diploids are mainly concentrated in Eurasia (Edmondson et al. 1980; Moore et al. 1982), with very few represented in seed collections. Some Poa species are widely cultivatedDraft as forage and turf grasses (Balasko et al. 1995; Weddin and Huff 1996; Huff 2003), with P. pratensis (Kentucky bluegrass) and P. arachnifera (Texas bluegrass) as two heavily utilized species for turf. Poa pratensis was examined cytologically to determine its somatic chromosome number, and 91% of the P. pratensis taxa had chromosome numbers ranging from 2n=24–124 (Bowden 1961; Love and Love 1975), with the most common ploidy levels of 56, 63, 70, and 77 (Speckmann and Van Dijk 1972). Similarly, polyploid P. arachnifera chromosome numbers have ranged from 2n=42-91 (Brown 1939; Gould 1958; Kelley et al. 2009), with the most common chromosome numbers of 2n=8x=56 (Hartung 1946; Patterson et al. 2005). Interspecific hybridization between P. arachnifera and P. pratensis have been made, taking advantage of the apomixis in P. pratensis and the dioecy in P. arachnifera , for the development of hybrid turfgrass cultivars (Read et al. 1999; Meyer et al. 2005; Rose-Fricker et al. 2007; Smith and Meyer 2009). However, despite an analysis done by Meeks and Chandra (2015) showing unique sequences in the thioredoxin region of P. https://mc06.manuscriptcentral.com/genome-pubs Page 5 of 62 Genome 5 arachnifera compared to P. pratensis , the genomic relationship responsible for their ability to form interspecific hybrids is not yet well understood. The purposes of this study were to confirm chloroplast lineage analyses of Soreng et al. (2010) in diploid Poa species, and build upon that framework with the addition of more chloroplast region sequences and further diploid species and accessions. Additionally, we include samples of high-polyploid P. pratensis and P. arachnifera with the aim to hone in on possible diploid progenitors of those species. Because of ploidy inconsistencies in published literature, we use chromosome counts and flow cytometry to assure ploidy levels, and further highlight relationships within sections of Poa that contain P. pratensis and P. arachnifera . Materials and methods Draft Plant materials and sampling. Twenty-one accessions of Poa were obtained from the National Plant Germplasm System (USA), the Margot Forde Germplasm Centre (New Zealand), and the IPK Genebank (Germany). The accessions included sampling of putative diploid Poa species, and several tetraploid Poa species for Poa subgenera where diploids were unavailable. The sampling included the three main Poa subgenera and eight sections (Table 1). Additionally, certified sod-quality seed of the P. pratensis cultivar ‘Midnight’ was sampled, as was a collection of P. arachnifera obtained from J. Goldman (USDA-ARS, Woodward, OK). Ten seeds of each accession were planted in Sunshine Mix #2 (Sun Gro Horticulture, Agawam, MA) and five healthy plants of each accession were selected and maintained in a greenhouse in Logan, UT. For a few accessions, less than five plants germinated such that only those plants were used in analyses (Table 1). https://mc06.manuscriptcentral.com/genome-pubs Genome Page 6 of 62 6 Flow cytometry and cytological analysis. Flow cytometry was performed on the plants from each accession. Young, fully expanded leaves weighing 100 mg were collected and finely chopped in Petri dishes containing 1 mL of freshly made chopping buffer (10 mM MgSO 4, 50 mM KCl, 5 mM Hepes). Leaf nuclei were filtered through 30 µm nylon mesh into test tubes, centrifuged at 200 × g for 5 min, and resuspended in 1 mL of fresh chopping buffer with the addition of 100 mg L −1 dithiothreitol, 16.5 mg L −1 of ribonuclease A and 100 mg L −1 propidium iodide. Tubes were incubated at 37°C for 15 min after which 3 µL of chicken erythrocyte nuclei (CEN) singlets (Biosure, Grass Valley, CA) were added to each tube, serving as an internal control for each sample. Samples were analyzed at 488 nm (FL2A filter) with a BD Accuri™ C6 Flow Cytometer (BD Biosciences, San Jose, CA). For each sample, the plant nuclear 2C DNA content, measured in picograms (pg), wasDraft determined by multiplying the relative 2C DNA content (plant sample peak mean/CEN peak mean) to the CEN 2C DNA content of 2.5 pg. The process was repeated to confirm original 2C values. Diploid plants (2x=14) were tested with and without the CEN standard to determine if their flow cytometry peak overlapped with the CEN peak. In those cases where the CEN and plant peak were inseparable, a genome size of 2.5 pg was assigned.