The bluegrasses, also commonly referred to as meadowgrasses, are one the most economically important genera of the Poaceae (Huff, 2010; Soreng and Barrie, 1999). Kentucky bluegrass (Poa pratensis L.) is the botanical-type species for the genus Poa (Soreng and Barrie, 1999) and is recognized as one of the most widely used temperate perennial grass species for both forage and amenity turf in the northern United States and Canada (Huff, 2003, 2010). Kentucky bluegrass is a facultative apomictic species, which has a highly variable chromosome number, creating a series of polyploidy and aneuploidy ranging from 2n = 28 to 154 (Akerberg, 1939; Grazi et al., 1961; Huff, 2003; Love and Love, 1975; Meyer and Funk, 1989; Muntzing, 1933; Nielsen, 1946; Tinney, 1940). Although this complex polyploidy may sometimes present a challenge to breeding efforts in this species, the ability of kentucky bluegrass to propagate diverse and odd ploidy levels through apomixis results in many genetically distinct phenotypes within the species (Huff, 2010). This wide range of diversity of cultivars and accessions of kentucky bluegrass has been previously characterized based on common turf performance or morphological characteristics (Bara et al., 1993; Bonos et al., 2000; Murphy et al., 1997; Shortell et al., 2009) as well as by random amplified polymorphic DNA markers (Curley and Jung, 2004; Huff, 2001; Johnson et al., 2002).

Although currently used in many plant variety protection (PVP) schemes, the use of morphological characteristics to determine genetic diversity and distinctness, uniformity, and stability (DUS) of cultivars has numerous drawbacks, including the maintenance of increasingly large reference collections for comparative analyses, a limited number of descriptors available to distinguish varieties, time-consuming field-based measurement of large numbers of samples and replicates, and the potential for the expression of morphological traits to be influenced by the environment (Giancola et al., 2002; Ibanez et al., 2009; Kwon et al., 2005; Lombard et al., 2000; Roldan-Ruiz et al., 2001). As a result of these drawbacks of using morphological characters for determining genetic diversity or for DUS testing, numerous researchers have proposed using molecular markers for PVP (Bonow et al., 2009; Borchert et al., 2008; Cooke and Reeves, 2003; Giancola et al., 2002; Gunjaca et al., 2008; Heckenberger et al., 2002, 2003, 2005a; Ibanez et al., 2009; Kwon et al., 2005; Roldan-Ruiz et al., 2001; Smith et al., 2009; Smykal et al., 2008; Tommasini et al., 2003; van Eeuwijk and Law, 2004; Vosman et al., 2004). In contrast to morphological characters, molecular markers offer a number of advantages, including a nearly unlimited number of characters, high degree of polymorphism, ease of scoring, and they are unaffected by the environment (Lombard et al., 2001; Smykal et al., 2008; Tommasini et al., 2003). In this report, we describe the development of the first polymorphic microsatellite markers for ongoing molecular genetic research in kentucky bluegrass.

Total genomic DNA was extracted from a single plant of the kentucky bluegrass cultivar Cabernet (Bonos et al., 2004) using the Sigma GenElute Plant Genomic DNA Miniprep Kit (St. Louis, MO) according to the manufacturer's instructions. DNA was sent to Genetic Identification Services Inc. (GIS, Chatsworth, CA) for construction of simple sequence repeat (SSR) libraries enriched for CA, GA, AAT, and CAG SSRs. Methods for DNA library construction and enrichment were developed following Jones et al. (2002). Briefly, genomic DNA was partially digested with a cocktail of seven blunt-end restriction enzymes (RsaI, HaeIII, BsrB1, PvuII, StuI, ScaI, EcoRV). Fragments in the size range of 300 to 750 bp were adapted with 20 bp oligonucleotides, which contained a HindIII site at the 5′ end and subjected to magnetic bead capture (CPG, Inc., Lincoln Park, NJ) using 5′-biotinylated CA₍₁₅₎, GA₍₁₅₎, AAT₍₁₂₎, and CAG₍₁₂₎ as capture molecules according to the manufacturer's instructions. Captured molecules were amplified using a primer complimentary to the adaptor, digested with HindIII to remove the adaptor sequences, and ligated into the HindIII site of pUC19. Recombinant plasmids were then electroporated into Escherichia coli DH5α. GIS delivered DNA libraries 50% enriched for CA and GA repeats and libraries 15% enriched for the trinucleotides AAT and CAG.

Several 50-μL aliquots of each SSR library were plated out onto LB agar plates containing ampicillin, IPTG, and Bluo-Gal. The plates were incubated at 37 °C overnight. Three thousand individual colonies were chosen from the CA- and GA-enriched libraries and grown in 6 mL LB broth containing ampicillin. Five hundred colonies were chosen from the AAT- and CAG-enriched libraries. DNA was isolated from the cultures using the Qiagen QIAprep Spin Miniprep kit (Valencia, CA). Samples were sequenced using an ABI 3130xl Genetic Analyzer (Applied Biosystems, Foster City, CA). A total of 2523 clones were sequenced. Sequence data from clones in the AAT and CAG libraries proved to be complex (long sequences of highly repetitive DNA), and the percentage appropriate for primer design was less than 1%. As a result, only the dinucleotide libraries were used for further analyses. Six hundred seventy-four dinucleotide clones contained no repeat, whereas 546 clones had other problems that precluded primer design (poor sequence, repeat too close to cloning site, etc.). One thousand seventy-one clones contained either CA or GA repeat motifs suitable for primer design.

Sequence data from the clones containing dinucleotide SSRs were analyzed for primer selection, and polymerase chain reaction (PCR) primers were designed to flank regions surrounding the SSR motif using Primer3 software (Rozen and Skaletsky, 2000). The forward primer in the pair was elongated at the 5′ end using the M13(-21) 18-bp sequence (5′−TGTAAAACGACGGCCAGT-3′) for economic fluorescent labeling (Schuelke, 2000) and synthesized by Integrated DNA Technologies (Coralville, IA). Of the 1071 clones containing dinucleotide repeats, PCR primer pairs were designed for 500 of these sequences, of which 88 (17.6%) produced distinct, repeatable polymorphic bands in a panel of 265 kentucky bluegrass cultivars, experimental selections, collections, and hybrids with texas bluegrass (Poa arachnifera Torr.).

For genotyping, PCR reactions were conducted in 13-μL reactions using 50 ng genomic DNA, 1X Ramp-Taq PCR buffer (Denville Scientific, Metuchen, NJ), 2 mm MgCl₂, 0.25 mm each dNTP (Denville Scientific), 0.5 U Ramp-Taq DNA polymerase (Denville Scientific), 0.5 pmol forward primer with M13(-21) addition, 1 pmol reverse primer, and 1 pmol forward florescent dye-labeled M13(-21) primer (FAM, NED, PET, or VIC). Thermal-cycling parameters were 94 °C for 5 min followed by 30 cycles of 94 °C for 30 s, 55 °C for 45 s, 72 °C for 45 s followed by 20 cycles of 94 °C for 30 s, 53 °C for 45 s, 72 °C for 45 s, ending with a final extension of 72 °C for 10 min. PCR reaction products were analyzed on an ABI 3130xl Genetic Analyzer and sized using Genemapper 3.7 software (Applied Biosystems) and LIZ 500(-250) size standard (Applied Biosystems).

Although microsatellites can generate codominant data, problems can arise during the identification of polyploid genotypes because it is difficult to determine the number of copies of an allele in heterozygotes (Liao et al., 2008; Markwith et al., 2006; Saltonstall, 2003). Thus, banding patterns observed at particular loci are referred to as “allele phenotypes” (Becher et al., 2000). The individual alleles of the SSR markers used in the current study of 265 polyploid kentucky bluegrasses and hybrids were treated as dominant markers, in which the banding phenotypes were scored as band absence (0) or presence (1). Polymorphism information content (PIC) of each individual SSR allele was calculated according to the formula described by Weir (1990): PIC = 1 – ∑ P_i², where P_i is the frequency of the ith allele in the genotypes examined. For dominant markers, this formula can be simplified to PIC = 2P_iQ_i where P_i is the frequency of presence and Q_i is the frequency of absence of a particular band (Tehrani et al., 2008). For dominant markers, the maximum PIC value is 0.50. PIC values for the 88 SSR markers in the current study are given as a range of PIC values for all alleles generated by a particular SSR marker (Table 1).

Table 1.Primer sequences and characteristics of 88 Poa pratensis (L.) microsatellite markers obtained from the cultivar Cabernet tested on 265 Poa pratensis cultivars, experimental selections, collections, and hybrids.

View Fullscreen

Characteristics of the SSR markers and their primer sequences are shown in Table 1. All SSR markers produced well-defined discrete alleles. The total number of alleles produced by the 88 SSR markers in the 265 cultivars and accessions was 3373. The number of alleles for individual SSR markers ranged from four to 81 with an average of 38.3 alleles per SSR. The high average number of alleles per SSR is likely the result of the combination of high genetic diversity and high ploidy levels in the genotypes under investigation. The PIC value ranges for the majority of the SSR markers approached the maximum PIC value for dominant markers, indicating that each SSR marker produced alleles that were highly polymorphic (Table 1). PIC values and expected heterozygosity of the 3373 individual alleles are available on request. The high level of polymorphism observed for the described microsatellite markers support their application in genetic studies of kentucky bluegrass. The potential use of these markers includes assessment of genetic diversity, creation of genetic linkage maps, assessment of levels of apomixis in cultivars and experimental varieties, and identification of aberrant progeny in apomictic kentucky bluegrass breeding programs. Additionally, the large number of microsatellite markers described in this report falls within a range (75 to 100 SSRs) that other researchers have suggested to be appropriate for PVP analyses in other crops (Heckenberger et al., 2005b; Smith et al., 2009).