RAPD Analysis Reveals Genetic Variability Among Sexual and Apomictic Paspalum dilatatum Poiret Biotypes

Abstract

Paspalum dilatatum is a valuable forage grass in the subtropics. This species consists of several sexual (tetraploid) and apomict (penta- and hexaploid) biotypes. It has been proposed that the presence of a genome of unknown origin, the X genome, is responsible for apomixis in penta- and hexaploid biotypes. Here we evaluated the utility of random amplified polymorphic DNA (RAPD) markers for discriminating sexual and apomictic P. dilatatum biotypes. DNA samples from nine accessions, including P. intermedium, P. juergensii, and P. dilatatum (ssp. flavescens, and the common and Uruguayan biotypes) were analyzed with 86 RAPD primers. Three hundred sixty-two fragments were scored and genetic similarity estimates revealed that the penta- and hexaploid biotypes were highly similar (S_D ≥ 0.913). Forty RAPDs were unique to the penta- and hexaploid biotypes. Overall RAPD markers were useful for assessing genetic variation among closely related P. dilatatum genotypes as well as generating putative X genome markers.

Paspalum dilatatum (dallisgrass) is an important South American forage species. Seven polyploid biotypes, including sexual tetraploids (2n = 4x = 40) and apomictic penta- (2n = 5x = 50) and hexaploids (2n = 6x = 60), have been described (Bashaw and Forbes 1958; Burson 1991a,b, 1995; Burson et al. 1973). Common dallisgrass, an apomictic pentaploid, has excellent forage potential (Bashaw and Forbes 1958) but large-scale cultivation has been hampered, primarily because of its susceptibility to ergot (Claviceps paspali) (Bennett 1940). Because conventional breeding methods are not feasible, much effort has been expended to identify the progenitors of this biotype (Burson et al. 1973; Espinoza and Quarin 2000). Identification of progenitor species would provide both a means for reconstructing a pentaploid biotype, possibly with increased resistance to ergot, and offer insights into the origin and genetic mechanisms controlling apomixis in P. dilatatum.

It has been suggested that common dallisgrass (IIJJX genome) originated from a hybridization event between P. dilatatum ssp. flavescens (IIJJ) and a hexaploid biotype, Uruguayan (IIJJXX) (Burson 1991b). This hypothesis is supported both by the sympatric occurrence of these biotypes and by the phenotypic similarity of their hybrid progeny to common dallisgrass (Burson 1991a). Cytogenetic analyses have revealed that diploid species P. intermedium and either P. juergensii or P. paniculatum are the likely donors of the II and JJ genomes, respectively (Burson et al. 1973). The X genome progenitor, however, is unknown (Bennett et al. 1969). Because penta- and hexaploid P. dilatatum are apomictic, it has been proposed that apomixis is associated with the presence of the X genome in these biotypes (Burson 1995).

Random amplified polymorphic DNA (RAPD) markers have been used to characterize genetic diversity in Paspalum germplasm (Liu et al. 1994; M'Ribu and Hilu 1996). In P. notatum (bahiagrass), RAPDs have also been employed for discriminating apomictic and sexual accessions (Martinez et al. 1999; Ortiz et al. 1997). The primary objective of this study was to identify RAPDs unique to apomictic P. dilatatum biotypes. Because PCR-based techniques are less labor intensive than performing interspecific crosses, genome-specific molecular markers should expedite identification of the P. dilatatum X genome donor species.

Materials And Methods

Plant Material

Plant material assayed in this study was obtained from Embrapa Recursos Genéticos e Biotecnologia (Brasília, Brazil) and Facultad de Agronomia (Montevidéo, Uruguay). For each accession sampled, leaves from five different plants were pooled. Accessions included two diploid species, P. intermedium (BRA-012599) and P. juergensii (BRA-016225), and seven polyploid P. dilatatum accessions, including one ssp. flavescens (BRA-020885), five common dallisgrass (BRA-016764, BRA-017086, BRA-017116, BRA-017825, and BRA-019496), and one Uruguayan (BRA-000825). The common dallisgrass accessions represented the natural distribution of this biotype in Brazil.

Cytology

Root-tip chromosome counts and observations of meiotic behavior in dividing pollen mother cells were conducted to confirm the initial morphology-based identifications of all accessions. Cytological analyses were performed as previously described (Espinoza and Quarin 2000) and chromosome preparations were observed with a phase contrast microscope. One plant was analyzed from each accession.

DNA Analysis

Total genomic DNA was isolated from pooled fresh leaves (five plants per accession) according to Doyle and Doyle (1990). DNA samples were visualized on 0.8% agarose gels stained with ethidium bromide and quantified by comparison to known quantities of undigested bacteriophage λ DNA.

In total, 86 random 10-mer primers (Qiagen Operon, Alameda, CA) were tested in all DNA samples. PCRs were done in 20 μl volumes containing 1× PCR buffer, 1.6 μg of nonacetylated bovine serum albumin (BSA); 300 μM dNTPs, 25 ng primer, 1.5U Taq DNA polymerase (Invitrogen Life Technologies, Carlsbad, CA), and 10 ng DNA. Amplifications were performed using an MJ Research PT-100 thermal cycler with the following temperature profile: 92°C for 1 min, 35°C for 1 min, and 72°C for 2 min for 40 cycles, with a final extension of 7 min at 72°C. PCR products were resolved on 1.5% agarose gels stained with ethidium bromide and fragment sizes were estimated by comparison to a size standard (1 kb DNA ladder; Invitrogen Life Technologies).

Fragments were scored as either present (1) or absent (0) in each sample, and pairwise genetic similarities were calculated as follows:

\[S_{Di,j}\ {=}\ 2N_{ij}{/}\left[N_{i}\ {+}\ N_{j}\right],\]

where N_ij is the total number of fragments common to accessions i and j, and N_i and N_j are the total number of fragments present in i and j, respectively (Dice 1945). Dice similarity, S_D, is equivalent to the Nei and Li coefficient (Nei and Li 1979). Genetic similarities and UPGMA phenograms were both generated by using NTSYS-pc version 2.02i (Rohlf 1998).

Results and Discussion

Cytological Analyses

Results from root-tip chromosome counts (2n) were identical to previously reported values for all accessions (data not shown). Because some tetra- and hexaploid Paspalum spp. have the same chromosome number as the P. dilatatum ssp. flavescens and Uruguayan biotypes, analysis of meiotic chromosome pairing was also conducted. Diploid accessions P. intermedium and P. juergensii had the same meiotic chromosome associations with 10 bivalents. Among the polyploids, P. dilatatum ssp. flavescens behaved cytologically as an allotetraploid with 20 bivalents, the five common dallisgrass accessions (pentaploids) showed 20 bivalents and 10 univalents, while the hexaploid Uruguayan biotype primarily exhibited 30 bivalents (data not shown). Meiotic behavior observed in these P. dilatatum biotypes was consistent with previous cytological studies (Bashaw and Forbes 1958; Burson et al. 1991). Therefore both chromosome counts and meiotic pairing supported the morphology-based identifications of all accessions.

RAPD Analysis

Of 86 RAPD primers tested, 16 were discarded due to poor amplification. The remaining 70 primers (OPA10, OPB12–OPB20, OPG01–OPG19, OPH07, OPJ05, OPK10, OPN01–OPN17, OPN19, OPN20, OPR03, OPR04, OPR15, OPR16, OPU15, OPV02, OPW11, OPW20, OPX02, OPX04, OPX07, OPX12, OPX13, OPX17, OPY09, OPY15, OPY16, OPY20, OPZ12) amplified well, and DNA fragment data from these primers were used to estimate the genetic similarity among accessions.

Results from two polymorphic RAPD primers are shown in Figure 1. Invariably a larger number of DNA fragments was amplified in penta- and hexaploid (240–244) compared to diploid and tetraploid accessions (139–183). In total, 362 RAPDs were scored and 41 of these markers (11.3%) were uninformative (i.e., monomorphic across all accessions). Among the diploid accessions, 44 and 41 RAPDs were unique to P. intermedium (II) and P. juergensii (JJ), respectively. One hundred thirty-six DNA fragments were amplified only in the polyploid P. dilatatum biotypes. Among these, five markers were unique to ssp. flavescens (IIJJ) and 62 were present only in the common dallisgrass (IIJJX) and Uruguayan (IIJJXX) biotypes. Notably, 65% (40/62) of the markers that amplified exclusively in penta- and hexaploid accessions were fixed. Therefore, these fragments may represent X genome-specific markers (see Figure 1). RAPDs unique to apomictic biotypes were obtained with primers OPB12, OPB14, OPB15, OPB19, OPG01, OPG05, OPG10, OPG11, OPG15, OPG18, OPH07, OPN03, OPN05, OPN07, OPN10, OPN11, OPN12, OPN13, OPN14, OPN15, OPN20, OPR03, OPR04, OPU15, OPW11, OPX02, OPX04, OPX17, OPY16, and OPZ12. Primer sequences are available upon request (from A.M.C.).

Figure 1.

RAPD patterns obtained from Paspalum DNA with primers (A) OPG05 and (B) OPG18. Arrows denote RADPs unique to the penta- and hexaploid biotypes. Lane information: 1 kb ladder; (lane 1) P. intermedium (II); (lane 2) P. juergensii (JJ); (lane 3) ssp. flavescens (IIJJ); (lanes 4–8) common dallisgrass (IIJJX); (lane 9) Uruguayan (IIJJXX). Corresponding accession numbers follow the order listed in the text (see Materials and Methods).

Results from relationship analyses (Figure 2) indicated that the diploid species were distinct, both relative to each other and to the polyploids. Although this result is consistent with ploidy level, we should note that the small number of diploid accessions analyzed could have influenced the outcome. It is also possible that P. intermedium and P. juergensii are not the progenitors of P. dilatatum. Future studies should include other Paspalum species from the Quadrifaria and Paniculata groups as potential donors of the II and JJ genomes, respectively.

Figure 2.

Phenetic analysis of Paspalum accessions based on 362 RAPDs. For common dallisgrass, superscripts denote the following accessions: (1) BRA-016764, (2) BRA-017116, (3) BRA-017825, (4) BRA-019496, and (5) BRA-017086. See text for accession numbers of other species/biotypes.

Conversely, P. dilatatum accessions were much more closely related (S_D > 0.76) (Figure 2). Similarities were highest among apomictic biotypes (common dallisgrass and Uruguayan accessions) (S_D > 0.91), possibly due to their related genomic composition, asexual mode of reproduction, and/or local selective pressures favoring only a few very adapted genotypes with limited genetic variation. Most of the variation observed among P. dilatatum sexual and apomictic biotypes was due to the presence of markers exclusive to the penta- and hexaploids (i.e., the putative X genome-specific markers) (see Figure 1).

In this study we have shown that RAPD markers are useful for assessing genetic variability of closely related P. dilatatum genotypes as well as for generating candidate X genome-specific markers. Obviously the next step would be to confirm the genomic location of these RAPDs. Markers could then be converted to sequence characterized amplified regions (SCARs) and used in PCR or hybridization-based assays for screening diploid species. Identification of the X genome donor should provide insights into the origin and genetic control of apomixis in P. dilatatum.

Corresponding Editor: Brandon Gaut

Alexandra M. Casa is currently at the Cornell University Institute for Genomic Diversity, 153 Biotechnology Building, Ithaca, NY 14853. We thank Marcio Ferreira and Dario Grattapaglia for allowing A.M.C. to conduct molecular analyses in their laboratory. Thanks to Mercedes Rivas (Facultad de Agronomia, Montevideo, Uruguay) for providing ssp. flavescens and the Uruguayan biotype. We are also grateful to Stephen Kresovich, Byron Burson, Peggy Ozias-Akins, and two anonymous reviewers for their comments and suggestions. This work, funded by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, Brazil) (to A.M.C., C.R.L., and J.F.M.V.), was done in partial fulfillment of an MSc degree (A.M.C.).