In Vitro Cell.Dev.Biol.—Plant (2009) 45:659–666 DOI 10.1007/s11627-009-9230-x
EMBRYO CULTURE
Interploid St. Augustinegrass [Stenotaphrum secundatum (Walt.) Kuntze] hybrids recovered by embryo rescue Anthony D. Genovesi & Russell W. Jessup & Milton C. Engelke & Byron L. Burson
Received: 5 August 2008 / Accepted: 2 July 2009 / Published online: 24 July 2009 / Editor: Mark C. Jordan # The Society for In Vitro Biology 2009
Abstract St. Augustinegrass is one of the most important warm season turfgrasses in the southern United States because of its shade tolerance. Most cultivars are diploids (2n=2x=18) and are susceptible to various diseases and insects. Polyploid cultivars in the species have some resistance to pests, but most lack cold tolerance. In this study, eight polyploid genotypes were crossed with six diploid cultivars to transfer pest resistance to the diploids. Because interploid crosses often result in aborted seed, it was necessary to use in vitro techniques. Using embryo rescue, 268 plants were recovered from 2,463 emasculated and pollinated florets (10.88% crossability). Because of the heterogeneous nature of the species, these purported hybrids could not be verified by phenotype. DNA markers were used for hybrid identification. A subset of 25 plants from crosses between the aneuploid cultivar Floratam (2n= 4x=32) and five diploid cultivars were analyzed using 144 expressed sequence tags–simple sequence repeats (ESTSSRs) developed from buffelgrass cDNA sequence data. Chi-square tests for paternal-specific markers revealed that all analyzed progeny were true F1 hybrids and none originated from self-fertilization or unintended outcrossing. In addition to identifying DNA polymorphism, the ESTSSRs revealed that genetic variation exists among all A. D. Genovesi (*) : M. C. Engelke Texas AgriLife Research, Texas A&M System, 17360 Coit Road, Dallas, TX 75252, USA e-mail:
[email protected] R. W. Jessup : B. L. Burson USDA-ARS, Crop Germplasm Research Unit, Texas A&M University, 430 Heep Center, College Station, TX 77843, USA e-mail:
[email protected]
analyzed cultivars and is not partitioned between ploidy levels. The findings demonstrate that these embryo rescue techniques will enable the entire spectrum of St. Augustinegrass genetic variation to be better used through the recovery of interploid hybrids. Keywords Floratam . Polyploid . Diploid . Tissue culture . Turfgrass . EST-SSRs
Introduction St. Augustinegrass, Stenotaphrum secundatum (Walt.) Kuntze, is widely used as a turfgrass in warm, humid, tropical, and subtropical climates where its broad leaf blades produce a tight canopy from rapidly elongating stolons, resulting in a coarse-textured turf (Sauer 1972). Artificial propagation is usually vegetative, by stolon cuttings, plugs, or sod. The base chromosome number of St. Augustinegrass is x=9, with diploids (2n=2x=18), triploids (2n=3x=27), and tetraploids (2n=4x=36; Long and Bashaw 1961). Adaptive and morphological variations in St. Augustinegrass are associated with chromosome differences. The most conspicuous differences between ploidy levels are that diploids are lower growing and have narrower, translucent, bright green leaf blades while polyploids have coarser, thicker leaf blades that are blue/green and less saturated in color (Busey 1986). Most population improvement has been accomplished using diploids, while polyploid cultivars (e.g., ‘Bitterblue’ and ‘Floratam’) are often of unknown origin. The southern chinch bug, Blissus insularis Barber, is generally considered to be the most serious pest of St. Augustinegrass. The polyploid cultivar Floratam was released in 1973 because of its resistance to the southern chinch bug and the St. Augustine decline
660
GENOVESI ET AL.
(SAD) virus (Horn et al. 1973; Reinert and Dudeck 1974). The resistance of Floratam has been described as antibiosis because of the mortality and reduced oviposition of southern chinch bugs confined on Floratam. For St. Augustinegrass, it was initially thought that only polyploids were resistant (Reinert et al. 1986). One exception is the recent identification of a diploid with resistance to the southern chinch bug biotype that overcame the resistance in Floratam (Nagata and Cherry 2003). Various genetic resistances among the polyploids include (as has already been indicated) resistance to some biotypes of southern chinch bug (Reinert and Dudeck 1974); resistance to the SAD virus (Horn et al. 1973); and resistance genes to the sting nematode, Belonolaimus longicaudatus Pace (Busey et al. 1993). Polyploids are less preferred by Lepidoptera (Busey et al. 1982). While polyploids such as Floratam have good pest resistance and use water more efficiently than diploids, they do poorly in the shade and lack cold hardiness (Busey 2003). These are traits that might benefit from genetic exchange with diploid genotypes. Even though a large amount of genotypic variation has been demonstrated in germplasm screenings, variances in these studies are partitioned between ploidy levels (Busey 2003). The use of genetic variations is complicated by compatibility issues related to ploidy differences. Unlike interspecific crosses in bluegrass (Read et al. 1999) and Bermudagrass (Burton 1969) that produce viable seed without the use of embryo rescue, wide crosses in St. Augustinegrass, especially at the interploid level, have not been successful. St. Augustinegrass is a long day plant which makes it easy to manipulate for breeding purposes (Philley et al. 1993) because inflorescence initiation can be triggered by artificially extending the photoperiod (Dudeck 1974) and its florets are relatively large and easy to emasculate (Philley 1994). Ploidy level differences, however, impede the full use of germplasm resources since attempted crosses between plants with different ploidy levels have failed (Long and Bashaw 1961; Busey 1995). It would therefore be desirable to use embryo rescue in order to bridge the ploidy barrier. Doing so would provide opportunities to combine desirable traits from diploid and polyploidy St. Augustinegrass and develop cultivars with improved pest resistance, drought tolerance, shade tolerance, and winter hardiness. Embryo rescue has been used to recover progeny from a number of interspecific or interploid hybridizations where fertilization occurs but the endosperm fails to develop and the developing embryo eventually aborts because of the lack of nutrients (Bridgen 1994). Embryo rescue overcomes this problem by aseptically culturing the embryo on a nutrient medium and has been used to recover hybrids, which broadens the genetic base of various plant species by introducing genes with enhanced resistance to biotic and abiotic stresses. This approach has been used effectively in
rice (Brar and Khush 1997), Brassica (Inomatu 1993), eggplant (Kashyap et al. 2003), wheat (Bai et al. 1994), squash (Sisko et al. 2003), and seedless grape (Ramming et al. 2000) among others. A popular media formulation is one with Murashige and Skoog (MS; Murashige and Skoog 1962) basal salts amended with sugar, vitamins, and amino acids. Sucrose is the most commonly used carbon source for embryo culture. Sucrose is primarily an energy source, although it also plays an important role in maintaining a suitable osmotic potential in the nutrient medium. High osmolarity prevents precocious germination and keeps cells that are in the state of division from going into a state of elongation. Exogenous auxins do not seem to be required for plant embryo growth in vitro. Following embryo rescue, it is critical to determine if the resulting progeny are derived from somaclonal regeneration, self-pollination, outcrossing, or controlled crosspollination. Because of the highly heterogeneous nature of most St. Augustinegrass genotypes, it is essentially impossible to phenotypically determine if progeny from controlled crosses are actually hybrids or the result of outcrossing or self-pollination. If the parents of the purported hybrids differ in chromosome number, the actual hybrids can be identified by determining their chromosome numbers. However, if a large number of plants need to be analyzed cytologically, this is not an economically feasible approach. DNA markers can be a more efficient and practical approach. Random amplification of polymorphic DNAs (Ballester and Carmen de Vicente 1998), amplified fragment length polymorphisms (Pooler and Riedel 2002), and sequence characterized amplified regions (Abraham et al. 2005) have been used to identify hybrids, but these markers provide no information regarding the allelic nature of the DNA fragments that are amplified via polymerase chain reaction (PCR). Restriction fragment length polymorphisms (RFLPs) have also been used for hybrid verification (Wolff et al. 1994) and do provide allelic information; however, RFLPs are more time- and laborintensive than PCR-based marker systems. Expressed sequence tags–simple sequence repeats (EST-SSRs) combine the advantages of being PCR-based while also providing allelic information. We report the recovery of over 260 progeny via embryo rescue from hybridizations between polyploid maternal parents and diploid pollen donors, as well as the verification of such progeny as true hybrids using EST-SSR markers.
Materials and Methods Plant material. Several commercially available diploid St. Augustinegrass cultivars (e.g. ‘Raleigh,’ ‘DelMar,’
INTERPLOID ST. AUGUSTINEGRASS HYBRIDS RECOVERED BY EMBRYO RESCUE
‘Palmetto,’ ‘Mercedes,’ ‘Nortam,’ and ‘Texas Common’) were used as pollen parents in crosses with female polyploids. Commercial polyploids, used as maternal parents in crosses with the diploid pollen parents, were Floratam, Bitterblue, ‘FX-10,’ and ‘FHSA-115.’ In addition, exotic germplasm identified as S. secundatum and introduced from Africa was obtained from the Plant Genetic Resources Conservation Unit, USDA-ARS, Griffin, GA. These genotypes (PI 290888, PI 291594, PI 300129, PI 300130) are referred to as African polyploids which are fertile when crossed with each other, but are sterile when crossed with adapted US diploid germplasm (Busey 2003). Breeding methods. Plant breeding methods described by Philley et al. (1993) were used in making pollinations. Flowering was induced by extending the photoperiod with continuous lighting from fluorescent lights 24 h each day (Dudeck 1974). When terminal florets on the inflorescence exerted anthers in the greenhouse, the entire inflorescence, including an attached stolon with two nodes, was collected and taken to the laboratory where it was placed in a 2.54-cm diameter test tube containing tap water, sealed with Parafilm, and labeled according to genotype. The test tube with the inflorescence was placed under fluorescent lights (approximately 50 μmol m−2 s−1, 18-h photoperiod). At approximately 1700 hours each day, a rack of test tubes was placed in a translucent plastic box with moistened towels in the bottom and the lid was closed. The box was placed under the fluorescent lights overnight. The relative humidity inside the box delayed dehiscence of the newly exerted anthers until they could be removed the following morning. Male fertile diploid plants were placed in isolation at a different location the night before and the pollen was collected using a spore collector the day the pollinations were made. The spore collector consisted of an aspirator which allowed pollen to be sucked into a receptacle (Read and Bashaw 1969). A camel hair artist’s brush was used to apply the collected pollen on the stigmas of the emasculated floret. The florets initiated flowering at the top of the inflorescence and continued to the base during the course of a week with newly exerted stigmas treated daily as indicated above. Pollinations ceased after 1 wk, at which time the flower/test tube was allowed to remain under the fluorescent lights for an additional 2 wk following the last pollination. Embryo rescue. After 2–3 wk following pollination, spikelets were taken to the laminar flow hood for aseptic removal of the developing embryo. The inflorescence was removed from the peduncle and placed in a Petri dish containing enough 70% ethanol (ETOH) to cover the bottom. With the aid of a dissecting microscope and forceps, the florets were opened to reveal their contents. If an enlarged ovule was
661
found (Fig. 1a), it was dissected away from the floret into the surrounding bath of 70% ETOH. The ovule was transferred directly to a sterile Petri dish lid and left in the flow hood with a bead of 70% ETOH surrounding it and allowed to air dry. Immediately upon air drying but before the ovule began to desiccate, the ovule was either placed directly on the nutrient medium or aseptically opened with forceps and the developing embryo plus vestigial endosperm removed and placed on solidified nutrient medium in a 100-mm Petri dish. If the whole ovule was placed intact on the medium, it was necessary to open the ovule at some point to allow plantlet development. The nutrient source was a 1/2 MS (Murashige and Skoog 1962) basal medium with MS vitamins, 5% sucrose as a carbon source and osmoticum and solidified with 3.5 g/L Phytagel. Cultured embryos (Fig. 1b) per ovules were placed under an 18-h photoperiod at 25°C. After 2 to 3 wk, individual plants (Fig. 1c) were moved to a 2.54-cm diameter test tube with the same medium for another 3 to 4 wk, at which time they were ready to plant. Plants were taken into the greenhouse where they were potted in a soil mix and placed on a misting bench to acclimate for a wk before being moved to ambient air conditions. DNA isolation. Genomic DNA was isolated following a modified Aljanabi and Martinez (1997) protocol. Four hundred microliters of homogenizing buffer (0.4 M NaCl, 10 mM Tris–HCl, pH 8.0, 2 mM EDTA, pH 8.0) and 100 mg of fresh leaf tissue were added to 1.7 mL microtubes. Forty microliters of 20% sodium dodecyl sulfate and 8 μL of 20 mg/mL proteinase K were added and vortexed for 5 s. Following incubation in a water bath at 65°C for 1 h, 300 μL of NaCl saturated H2O was added and vortexed for 30 s. Samples were spun at 12,000 rpm for 10 min, supernatant was transferred to new tubes, samples were spun at 12,000 rpm for 20 min, and supernatant was transferred to new tubes without disturbing any remaining pellet. Following the addition of 800 μL of cold isopropanol and 20 gentle inversions, samples were incubated at −20°C for 1 h. Samples were spun at 10,000 rpm for 5 min, supernatant was removed, 500 μL of cold 70% ethanol was added, samples were spun at 10,000 rpm for 5 min, and supernatant was removed. Microtubes were inverted until dry, and DNA was resuspended in 100 μL of sterile ddH2O. EST-SSR markers. A total of 1,027 partial or full-length complementary DNA (cDNA) sequences from buffelgrass [Pennisetum ciliare (L.) Link syn. Cenchrus ciliaris L.] were downloaded from GenBank (National Center for Biotechnological Information; http://www.ncbi.nlm.nih. gov/Genbank/index.html). Buffelgrass was chosen because it is relatively close taxonomically to St. Augustinegrass. SSR identification and primer design were performed
662
GENOVESI ET AL.
Figure 1. (a) Enlarged ovule 2–3 wk after fertilization with an b immature embryo located in the area indicated (E L) with remnants of the stigma still present (R S). (b) Rescued embryo 4 d after isolation with bipolar elongating of the coleorhiza (Cr) and coleoptile (Co) along the embryo axis (E Ax). The embryo remains attached to the failing endosperm (En) by the scutellum (Sc). (c) Plantlet resulting from embryo germination 3 wk after isolation of an ovule with an embryo inside. Ovule was broken open to allow emergence of the plantlet with radicle (R), coleoptile (Co), shoot (S), and several pieces of ovule wall (OW) evident (4×).
a
←RS
←
EL
using the web-based ‘SSR Primer Discovery Tool’ (Plant Biotechnology Centre, La Trobe University; http://hornbill. cspp.latrobe.edu.au/ssrdiscovery.html). Selected SSRs contained at least ten dinucleotide or five tri-, tetra-, or pentanucleotide repeats. Primer design was based on the criteria of 50% GC content, minimum melting temperature of 50°C, absence of secondary structure, length of 20–27 nucleotides, and amplified product range of 100–400 bp. PCR conditions. PCR reactions were conducted in a total volume of 20 μL using 50 ng of DNA, 1X Promega MgCl2-free PCR buffer, 2.5 mM MgCl2, 0.2 mM dNTPs, 2 mM of each primer, and 1 U of Promega Taq polymerase. The PCR method included: (1) an initial denaturation at 95°C for 3 min; (2) ten touchdown decrement cycles at 95°C for 25 s, 64–55°C for 25 s, and 70°C for 45 s; (3) 36 cycles at 95°C for 25 s, 55°C for 25 s, and 70°C for 45 s; (4) an elongation cycle at 70°C for 10 min; and (5) a final hold at 4°C. Electrophoresis was run using a MEGA-GEL highthroughput vertical unit (C.B.S. Scientific, Del Mar, CA) as described by Wang et al. (2003). Allele bands were distinguished from minor bands that occur in nondenaturing polyacrylamide gel electrophoresis as described in Rodriguez et al. (2001) to prevent scoring errors. Data analysis. DNA markers were analyzed across five controlled hybridizations involving the aneuploid cultivar Floratam (2n=32) as the maternal parent and each of the diploid cultivars (2n=2x=18), DelMar, Mercedes, Nortam, Palmetto, and Raleigh as the pollen donors. For each cross, the female parent, putative male parent, and five purported hybrids were analyzed. A mixture of DNA from the parents was used as a control for possible interactions between genotypes. Chi-square goodness of fit values were determined and compared to χ20.05 =3.84 for one degree of freedom. Values were calculated for paternal-specific markers using both a 1:0 expected transmission ratio for homozygous markers and a 1:1 expected transmission ratio for heterozygous markers. Transmission of at least 50% of paternal-specific markers with acceptable chi-square values was considered evidence that a plant was a hybrid. Homozygous, maternal-specific markers for which alternative alleles could be separated by size were scored to
2 mm
b
Cr ↓
Sc →
← E Ax ↑
Co ← En
1 mm
c
←S
OW ↓
← Co
←OW
←R OW→
INTERPLOID ST. AUGUSTINEGRASS HYBRIDS RECOVERED BY EMBRYO RESCUE
663
Table 1. St. Augustinegrass progeny derived from crosses between polyploid females and diploid males in 2003–05 using embryo rescue Parents
No. of inflorescences
♀×♂
Each cross
Bitterblue Bitterblue Bitterblue Bitterblue Bitterblue
× × × × ×
DelMar Nortam Raleigh Texas Common Mercedes
4 3 1 1 3
Each ♀
No. of spikelets pollinated Each cross
245
1 1 0 0 0
540
0 1 0 3 7
27
13 54 119 106 248
460
13 1 8 11 9 13 3
94
1 17 5
DelMar Garrett’s 141 Mercedes Nortam Raleigh
1 3 5 6 12
Floratam Floratam Floratam Floratam Floratam Floratam Floratam
DelMar Garrett’s 141 Mercedes Nortam Palmetto Raleigh Texas Common
4 1 5 8 4 4 1
27
64 10 101 118 82 66 19
FX-10 × DelMar FX-10 × Nortam FX-10 × Raleigh
1 1 1
3
27 40 27
PI 290888 × DelMar PI 290888 × Nortam
2 1
PI PI PI PI PI
291594 291594 291594 291594 291594
× × × × ×
DelMar Mercedes Palmetto Raleigh Texas Common
1 1 4 4 1
PI PI PI PI PI
300129 300129 300129 300129 300129
× × × × ×
DelMar Mercedes Palmetto Raleigh Texas Common
1 1 1 2 1
PI PI PI PI PI
300130 300130 300130 300130 300130
× × × × ×
DelMar Mercedes Palmetto Raleigh Texas Common
2 3 2 1 1 98
Each cross
12
× × × × ×
Total
Each ♀
76 69 17 9 74
FHSA-115 FHSA-115 FHSA-115 FHSA-115 FHSA-115 × × × × × × ×
Progeny no.
3
82 45
11
48 36 146 153 46
6
34 53 37 84 52
9
78 87 76 44 23 2463
127
25 17
429
2 2 22 14 8
260
6 5 8 17 6
308
13 11 4 6 8 268
Progeny/No. spikelets pol. (%) Each ♀
Each cross
Each ♀
2
1.3 1.4 0 0 0
0.82
11
0 1.9 0 2.8 2.8
2.04
58
20.3 10.0 7.9 9.3 10.9 19.7 15.8
12.61
23
3.7 42.5 18.5
24.47
42
30.5 37.8
33.07
48
4.2 5.6 15.1 9.2 17.4
11.19
42
17.6 9.4 21.6 20.2 11.5
16.15
42
16.7 12.6 5.3 13.6 34.8
13.64
10.88
664
GENOVESI ET AL.
Table 2. Survey of buffelgrass EST-SSRs across St. Augustinegrass controlled hybridizations
Parental Hybridization
Floratam Floratam Floratam Floratam Floratam
× × × × ×
PCR amplification (%)
DelMar Mercedes Nortam Palmetto Raleigh
82 81 58 81 74
investigate the occurrence of progeny in which no alleles for a given EST-SSR were present. Cluster and ordination analysis were performed using NTSYS-pc version 2.0 (Rohlf 1997). Similarity coefficients were calculated using Jaccard’s coefficient, SJ =a/(a+u), where a is the number of bands in which the two operational taxonomic units (OTUs) agree and u the number of bands present in one OTU but absent in the other (Jaccard 1908) with the SIMQUAL function. Cluster analysis was performed using the unweighted pair group method with arithmetic mean (UPGMA) algorithm within the SAHN function.
Results A total of 268 progeny (Fig. 1b, c) were recovered from the interploid crosses between polyploid female and diploid male plants using embryo rescue methodology from 2003 to 2005 (Table 1). The most widely used cultivar of the polyploid females, Floratam, produced 58 progeny from crosses with diploid male cultivars for an efficiency of 12.6% of pollinated florets producing viable plants. Bitterblue, an outdated commercial line of limited use, was the least productive, resulting in two progeny from 245 pollinated florets, which is a crossability of only 0.82%. FHSA-115, a polyploid with moderate chinch bug resistance from Florida, produced 11 progeny out of 540 pollinated florets (2.04%). FX-10, a polyploid with stable chinch bug resistance but of limited commercial
Polymorphism (%) Maternal-specific
Paternal-specific
Codominant
Total
13 15 9 10 15
6 8 6 10 8
17 12 9 16 15
36 35 24 36 38
value due to production problems, produced 23 progeny from 94 florets (24.5%). Crosses with the exotic polyploid plant introductions from Africa were quite productive. PI 290888 (45-251 from South Africa) yielded 42 progeny from 127 florets (33.1%). PI 291594 (51-339 from Zimbabwe) yielded 48 progeny from 429 florets (11.2%). PI 300129 (626 from South Africa) yielded 42 progeny from 260 florets (16.2%). Finally, PI 300130 (668 from South Africa) yielded 42 progeny from 308 florets (13.6%). Progeny from these cultivars are of considerable interest due to the reported high levels of chinch bug resistance exhibited by the African plant introductions (Reinert et al. 1986). Bitterblue, FHSA-115, Floratam, and FX-10 typically do not produce any seed, but one could still question if embryo rescue might enable an apomictic event to occur or a self to survive. Progeny from the crosses using Floratam as the female parent were chosen to represent the group (based on Floratam’s wide commercial use) for verification of their hybrid origin using molecular markers. A total of 144 EST-SSRs were developed from buffelgrass cDNA sequence data. The buffelgrass EST-SSRs produced significant levels of PCR amplification (58% to 82%) and polymorphism (24% to 38%) across five selected
Floratam
DelMar
Palmetto
Table 3. Similarity matrix of five hybrid combinations of St. Augustinegrass cultivars Mercedes
Floratam DelMar Mercedes Nortam Palmetto Raleigh Floratam DelMar Mercedes Nortam Palmetto Raleigh
1.00 0.67 0.50 0.49 0.56 0.52
Nortam
1.00 0.64 0.62 0.65 0.62
1.00 0.66 0.58 0.64
Raleigh
1.00 0.57 0.74
1.00 0.62
0.50
1.00
0.63
0.75
0.88
1.00
Figure 2. Dendrogram of St. Augustinegrass cultivars obtained by UPGMA cluster analysis.
INTERPLOID ST. AUGUSTINEGRASS HYBRIDS RECOVERED BY EMBRYO RESCUE Table 4. Summary of markers tested across progeny, allele losses, and verified hybrids
Parental hybridization
Floratam Floratam Floratam Floratam Floratam
× × × × ×
665
No. heterozygous paternal-specific markers examined
No. homozygous paternal-specific markers examined
No. cases of allele loss/No. homozygous maternal-specific markers examined
No. hybrids confirmed/No. progeny tested
10 10 9 10 10
6 8 6 10 8
0/5 1/6 2/5 1/5 1/6
5/5 5/5 5/5 5/5 5/5
DelMar Mercedes Nortam Palmetto Raleigh
St. Augustinegrass hybridizations (Table 2). Pairwise genetic similarity values among six hybrid cultivars ranged from 0.49 to 0.74 with a mean of 0.61 (Table 3), and a dendrogram produced by clustering analysis clearly separated each cultivar from one another (Fig. 2). Chi-square tests for paternal-specific markers revealed that all of the plants analyzed were true F1 hybrids and none resulted from self-fertilization or outcrossing (Table 4). Individuals missing both alleles of homozygous, maternal-specific EST-SSRs were identified among progeny from all hybridizations except Floratam × DelMar (Table 4).
Discussion The number of progeny recovered in this study reveals the importance of embryo rescue techniques in circumventing ploidy barriers and associated sterility problems in St. Augustinegrass. The most notable of the recovered progeny had African polyploid genotypes as maternal parents. For example, FX-10 is a cultivar derived from African polyploid germplasm known to possess genes conferring the most stable form of resistance to chinch bug (Busey 2003; Nagata and Cherry 2003). Future breeding efforts should focus on crossing these exotic genetic resources with superior diploid cultivars to improve sod production, cold hardiness, color, texture, shade tolerance, and marketable turf quality in St. Augustinegrass. While possibly less important, embryo-rescued progeny obtained from hybridizations using Floratam as the maternal parent also have potential value to St. Augustinegrass improvement programs. Floratam was the industry standard for chinch bug resistance until being rendered obsolete by the polyploid-damaging population (PDP) biotype of the pest (Busey 2003). With the identification of a diploid cultivar (NUF-76 also known as ‘Captiva’) possessing resistance to the PDP chinch bug (Nagata and Cherry 2003), it is technically feasible to stack resistance genes by crossing Floratam or FX-10 with NUF-76. Turf quality characteristics of NUF-76, however, are unknown and warrant evaluation. The EST-SSRs in this study proved to be highly informative in St. Augustinegrass. This was expected due
to the relatively close taxonomic relationship between buffelgrass and St. Augustinegrass. Buffelgrass EST-SSRs therefore provide a marker resource for St. Augustinegrass and allow the use of the abundant comparative genomic tools available within rice, maize, sorghum, and other grass species. In addition to identifying DNA polymorphism, the buffelgrass EST-SSRs revealed that genetic variation was present in all analyzed cultivars and not partitioned between ploidy levels (Table 3 and Fig. 2). While known to produce more diffuse clusters than alternative similarity coefficients, the use of Jaccard’s coefficient was critical to omit undesired negative matches between polyploidization-induced chromosome loss in Floratam and natural mutations among the diploid St. Augustinegrass cultivars. EST-SSR results differed from measurements of genetic variance based on morphological characteristics (Busey 2003) and were likely due to the increased precision of DNA markers. This is important because it indicates that both diploid and polyploid genotypes possess significant genetic variation that can be used in St. Augustinegrass improvement programs. Because morphological verification is extremely difficult and counting the chromosome number of the offspring is also difficult and time-consuming, molecular markers are the best tools to identify the true hybrids. All 25 progeny evaluated in this study were true hybrids, confirming that embryo rescue techniques are efficient at overcoming fertility barriers in St. Augustinegrass. Progeny missing both alleles for given genes in all but the Floratam × DelMar hybrids provided strong evidence for the occurrence of either chromosomal or segmental aneuploidy. This was expected due to the ploidy differences between the polyploid, maternal parent Floratam (2n=32; Busey 1979) and the diploid paternal parents (2n=2x=18; Busey 2003) and the chromosomal imbalances that would occur in the progeny from these crosses. Since Floratam is an aneuploid with 32 chromosomes, a high level of aneuploidy would be expected in the hybrids from crosses involving this cultivar. However, the lack of aneuploidy found in the Floratam × Delmar hybrids was unexpected. Because of the lack of aneuploidy, this population is most suited for genetic mapping studies; however, the small population size makes this a tentative conclusion that requires confirmation in
666
GENOVESI ET AL.
additional progeny. Aneuploidy suggests that sterility may persist in embryo rescue-derived hybrids, resulting in little viable seed production. For a vegetatively propagated turfgrass such as St. Augustinegrass, this can be viewed as a positive trait because the risk of off-type plants originating from seed would be greatly diminished. In summary, the embryo rescue techniques described here will enable a better and more complete use of the entire spectrum of genetic variation in St. Augustinegrass through the recovery of interploid hybrids. The novel genome combinations produced will allow the selection and development of new turf cultivars with improved tolerance to biotic and abiotic stresses. Acknowledgments This study was funded in part by a grant from the Turfgrass Producers of Texas, Inc. The authors wish to thank Dan Litteer, Pitak Skulkaew, and Robin Taylor for their excellent technical assistance.
References Abraham E.; Aa M.; Honig J.; Kubik C.; Bonos S. The use of SCAR markers to identify Texas × Kentucky bluegrass hybrids. Int. Turfgrass. Soc. Res. J. 10: 495–500; 2005. Aljanabi S. M.; Martinez I. Universal and rapid salt-extraction of high quality genomic DNA for PCR based techniques. Nucleic Acids Res. 25: 4692–4693; 1997. Bai D.; Scoles G.; Knott D. Transfer of leaf rust and stem rust resistance genes from Triticum triaristatum to durum and bread wheats and their molecular cytogenetic localization. Genome 373: 410–418; 1994. Ballester J.; Carmen de Vicente M. Determination of F1 hybrid seed purity in pepper using PCR-based markers. Euphytica 103: 223–226; 1998. Brar D.; Khush G. Alien introgression in rice. Plant Mol. Biol. 35: 35– 47; 1997. Bridgen M. A review of plant embryo culture. Hortic. Sci. 2911: 1243–1246; 1994. Burton G. Improving turfgrasses. In: Hanson A.; Juska F. (eds) Turfgrass science, vol 14. American Society of Agronomy, Madison, WI, pp 410–424; 1969. Busey P. What is Floratam? Florida State Hort. Soc. Proc. 92: 228– 232; 1979. Busey P. Morphological identification of St. Augustinegrass cultivars. Crop Sci. 26: 28–32; 1986. Busey P. Genetic diversity and vulnerability of St. Augustinegrass. Crop Sci. 35: 322–327; 1995. Busey P. Chapter 20: St. Augustinegrass, Stenotaphrum secundatum (Watt.) Kuntze. In: Casler M.; Duncan R. (eds) Turfgrass biology, genetics and breeding. Wiley, Hoboken, pp 309–330; 2003. Busey P.; Broschat T.; Center B. Classification of St. Augustinegrass. Crop Sci. 22: 469–473; 1982. Busey P.; Giblin-Davis R.; Center B. Resistance in Stenotaphrum to sting nematode. Crop Sci. 33: 1066–1070; 1993. Dudeck A. Flowering in several selections of St. Augustinegrass. In: Roberts E. (ed) 2nd International Turfgrass Research Conference
Proceedings, Blacksburg VA, June 1973. ASA and CSSA, Madison, pp 74–78; 1974. Horn G.; Dudeck A.; Toler R. ‘Floratam’ St. Augustinegrass: A fast growing new variety of ornamental turf resistant to St. Augustine decline and chinch bugs. Fla. Agric. Exp. Stn. Circ. S-224; 1973. Inomatu N. Embryo rescue techniques for wide hybridization. Monogr. TAG 19: 94–107; 1993. Jaccard P. Nouvelles recherches sur la distribution florale. Bulletin de la Societe Vaudoise de Sciences. Naturelles 44: 223–270; 1908. Kashyap V.; Vinod Kumar S.; Collonnier C.; Fusari F.; Haicour R.; Rotino G.; Sihachakr D.; Rajam M. Biotechnology of eggplant. Sci. Hortic. 97: 1–25; 2003. Long J.; Bashaw E. Microsporogenesis and chromosome numbers in St. Augustinegrass. Crop Sci. 1: 41–43; 1961. Murashige T.; Skoog F. A revised medium for rapid growth and bioassays with tobacco tissue. Physiol. Plant. 15: 473–497; 1962. Nagata R.; Cherry R. New source of southern chinch bug (Hemiptera: Lygaeide) resistance in a diploid selection of St. Augustinegrass. J. Entomol. Sci. 384: 654–659; 2003. Philley H. Inheritance of cold tolerance in St. Augustinegrass. MS thesis, Mississippi State University, p 62; 1994. Philley H.; Krans J.; Goatley J. Jr.; Maddox V.; Watson C. The effects of nutrient solutions on seed set of excised St. Augustinegrass inflorescences. Agronomy Abstracts, American Society of Agronomy, Madison WI, p 163; 1993. Pooler M.; Riedel L. Molecular markers used to identify interspecific hybridization between hemlock (Tsuga) species. J. Am. Soc. Hortic. Sci. 127: 623–627; 2002. Ramming D.; Emershad R.; Tarailo R. A stenospermocarpic, seedless Vitis vinifera × Vitis rotundifolia hybrid developed by embryo rescue. Hortic. Sci. 35: 732–734; 2000. Read J.; Bashaw E. Cytotaxonomic relationships and the role of apomixis in speciation in buffelgrass and birdwoodgrass. Crop Sci. 9: 805–806; 1969. Read J.; Reinert J.; Colbaugh P.; Knoop W Registration of ‘Reveille’ hybrid bluegrass. Crop Sci. 39: 590; 1999. Reinert J.; Busey P.; Bilz F. Old world St. Augustinegrasses resistant to the southern chinch bug (Heteroptera: Lygaeidae). J. Econ. Entomol. 79: 1073–1075; 1986. Reinert J.; Dudeck A. Southern chinch bug resistance in St. Augustinegrass. J. Econ. Entomol. 67: 275–277; 1974. Rodriguez S.; Visedo G.; Zapata C. Detection of errors in dinucleotide repeats typing by nondenaturing electrophoresis. Electrophoresis 22: 2656–2664; 2001. Rohlf F. NTSYS-pc numerical taxonomy and multi-variate analysis system, version 2.1. Exeter Software, Setauket; 1997. Sauer J. Revision of Stenotaphrum (Gramineae: Paniceae) with attention to its historical geography. Brittonia 24: 202–222; 1972. Sisko M.; Iancic A.; Bohanec B. Genome size analysis in the genus Cucurbita and its use for determination of interspecific hybrids obtained using the embryo rescue technique. Plant Sci. 165: 663– 669; 2003. Wang D.; Shi J.; Carlson S.; Cregan P.; Ward R.; Diers B. A low-cost, high-throughput polyacrylamide gel electrophoresis system for genotyping with microsatellite DNA markers. Crop Sci. 43: 1828–1832; 2003. Wolff K.; Rijn J.; Hofstra H. RFLP analysis in chrysanthemum. I. Probe and primer development. Theor. Appl. Genet. 883–4: 472– 478; 1994.
