Apomixis and sexuality in some Paspalum species.
The diversity within the genus is immense. It extends from the more obvious traits such as general morphology, growth habit, and areas of adaptation to the less conspicuous, such as mode of reproduction and cytological composition of the species. For example, chromosome numbers for the genus range from 2n = 2x = 12 to 2n = 16x = 160 (Burton, 1940; Quarin, 1974) with the most frequent number being 2n = 4x = 40 (Quarin and Norrmann, 1987). Cytotypes with different ploidy levels are common in some species, and depending on their genomic composition, meiotic chromosome pairing can also differ. For example bahiagrass, P. notatum Flugge, has cytotypes with 2n = 2x = 20, 3x = 30, 4x = 40, and 5x = 50 chromosomes (Tischler and Burson, 1995). Dallisgrass, P. dilatatum Poir., has cytotypes with 2n = 4x = 40, 5x = 50, and 6x = 60 chromosomes and some cytotypes with the same chromosome number have different meiotic pairing behaviors (Bashaw and Forbes, 1958; Burson et al., 1991).
Apomixis is prevalent within the genus and many species vary as to their reproductive behavior. Diploids are sexual and apomixis is expressed in the polyploids. However, some polyploids reproduce sexually. Many apomicts are facultative apomicts.
Even though progress has been made in cytologically evaluating Paspalum species, a large number remain in which little or no information is available regarding their basic cytological composition. During the past several years, considerable Paspalum germplasm has been introduced into the National Plant Germplasm System from plant collection trips to South America. This germplasm includes species that have never been studied cytologically. To use this germplasm in plant breeding programs, its basic cytology must be understood. Thus, the objectives of this investigation were to determine the chromosome number, meiotic behavior, method of reproduction and fertility of accessions of P. alcalinum Mez., P. falcatum Nees, P. modestum Mez., P. monostachyum Vasey, P. paucifolium Swallen, P. polyphyllum Nees, P. repens Bergius, and P. unispicatum (Scribn. et Merr.) Nash.
MATERIALS AND METHODS
The accessions, their identification numbers, and collection sites are in Table 1. I collected all of the accessions except for PI 337555 and PI 422022. The P. alcalinum accession PI 337555 was obtained from the USDA-ARS, Southern Regional Plant Introduction Station, Griffin, GA, and the P. monostachyum accession PI 422022 came from the USDA-NCRS Plant Materials Center, Knox City, TX.
Seeds of each accession were germinated in petri dishes inside a germinator and the seedlings were transplanted into small pots in a greenhouse. After the seedlings reached adequate size, they were planted into a field nursery spaced at a distance of 1 m between plants within rows and 1 m between rows. A minimum of 12 plants was planted for each accession.
Chromosome number of each accession was determined by counting the somatic chromosomes in root tips stained with Feulgen solution using a previously described technique (Burson, 1991). Inflorescences of plants growing in the field nursery were collected, fixed in Carnoy's solution (100% ethanol-chloroform-glacial acetic acid, 6:3:1) for approximately 1 h, and stored in 70% ethanol to study microsporogenesis. Pollen mother cells (PMCs) were stained in aceto-carmine and examined using phase contrast microscopy.
Inflorescences also were collected from plants growing in the field to determine their mode of reproduction. These were fixed in FAA (70% ethanol-glacial acetic acid-37% formaldehyde, 18:1:1) and subsequently used to study megasporogenesis and female gametophyte development. Spikelets with ovaries in different stages of development were removed from the inflorescences, dehydrated in a tertiary butyl alcohol series, embedded in paraffin, sectioned at 15 [[micro]meter], stained in a safranin-O fast green series, and examined using light microscopy. Additional information regarding the reproductive behavior of the polyploid accessions was obtained by growing 15 open-pollinated progeny of each accession in a field nursery and observing them for uniformity and variability.
Pollen stainability, an estimate of pollen viability, was determined using a 1% potassium iodide ([I.sub.2]-KI) solution. A minimum [TABULAR DATA FOR TABLE 1 OMITTED] of 500 pollen grains was observed at random using at least two plants of each accession. Seed set was determined (Burson and Bennett, 1971) from all of the spikelets on one inflorescence from three propagules of each accession.
RESULTS AND DISCUSSION
Paspalum modestum, P. monostachyum, P. repens, and one P. alcalinum accession (PI 404636) are diploids with 20 chromosomes that pair primarily as 10-ring bivalents at metaphase I (MI) of meiosis (Table 2). In each of these accessions, two univalents were observed in a limited number of cells usually at late MI, and probably resulted from the precocious separation of a bivalent. One PMC from the P. alcalinum accession had four univalents. Chromosome segregation was regular in all diploids. Laggards and micronuclei were not present at anaphase I (AI) or anaphase II (AII) and telophase I (TI) or telophase II (TII), respectively. Based on their regular meiotic behavior, pollen stainability was lower than expected in P. modestum and P. repens (Tables 2 and 3). This is the first report of the chromosome number for P. mnonostachyum and of a diploid P. alcalinum. The findings for the P. modestum and P. repens accessions are similar to those reported by Quarin and Hanna (1980) and Davidse and Pohl (1974).
Except for the P. alcalinum accession PI 404658, the remaining species and accessions were tetraploids with [TABULAR DATA FOR TABLE 2 OMITTED] 2n = 4x = 40 chromosomes (Table 2). Their meiotic chromosome pairing behavior was somewhat similar, with the most prevalent pairing configurations being bivalents, univalents and quadrivalents. The most frequent associations were bivalents and quadrivalents except for P. unispicatum in which univalents were common. However, a trivalent was seen in a limited number of P. falcatum and P. unispicatum PMCs (Table 2).
The two tetraploid P. alcalinum accessions differed, with PI 404638 having fewer univalents and more quadrivalents than PI 337555. This indicated more homology exists between members of the two genomes in PI 404638 than in PI 337555. As many as five laggards were observed at AI and AII in both accessions. One to three micronuclei were present in approximately 50% of the TI and TII cells. Pollen stainability was higher in PI 404638 than in PI 337555 (Table 3).
Most bivalents in the P. falcatum accession were well synapsed rings. Among the 75 PMCs, only two had two-rod bivalents. Chromosome segregation at AI was normal, and very few laggards and micronuclei were observed. However, pollen stainability was low (Table 3), which could be due to hot, dry growing conditions.
Chromosome stickiness was encountered in both P. paucifolium accessions, especially at MI. Consequently, much of the data in Table 2 for this species was obtained from cells at diakinesis. The overall average chromosome pairing behavior for both accessions was 0.24 univalents + 17.45 bivalents + 1.22 quadrivalents. One to three laggards were present in 30% of the AI cells, but micronuclei were absent at TI and TII. Pollen stainability (Table 3) was low for accession PI 404845, which also may be due to adverse environmental conditions.
Table 3. Pollen stainability and seed set of eight Paspalum species. Seed set Pollen Open Self Species stainability pollinated pollinated % P. modestum 43.6 0 0 P. monostachyum 80.0 0.6 0 P. repens 37.3 0 84.9 P. alcalinum PI 404636 70.9 11.5 27.4 PI 404638 61.5 53.4 20.6 PI 337555 35.9 20.8 - PI 404658 33.1 - - P. falcatum 29.2 23.4 - P. paucifolium PI 404845 3.2 50.5 1.2 PI 404847 42.4 10.2 10.0 P. polyphyllum PI 404872 0.4 6.0 - PI 404495 45.0 26.9 - PI 404498 9.3 6.0 4.3 P. unispicatum 22.0 25.4 8.6
Both P. polyphyllum accessions had a higher frequency of quadrivalents than any of the other species (Table 2). The overall average pairing for these two accessions was 0.28 univalents + 15.62 bivalents + 2.12 quadrivalents. Laggards were present in approximately 40% of the AI cells and micronuclei were present in 30% of the TI cells. Because of chromosome stickiness and clumping, the meiotic behavior of the third accession (PI 404872) was not determined. However, this accession also had 2n = 4x = 40 chromosomes.
Among the tetraploid species, meiosis was most irregular in P. unispicatum. It had the highest frequency of univalents (Table 2). Its chromosomes often clumped together at MI, making it difficult to interpret pairing. Unequal chromosome segregation at AI was common and some cells had as many as 21 laggards, but most had from 10 to 15. However, most telophase cells had no more than four micronuclei. Pollen stainability was higher than expected (Table 3), given the meiotic irregularities in this plant.
Based on the number and frequency of quadrivalents in all of the tetraploid accessions, both genomes in each species are apparently partially homologous. The degree of homology between members of the two genomes varies for each species, but apparently all are segmental allotetraploids.
The third P. alcalinum accession (PI 404658) was a pentaploid with 50 chromosomes. Meiotically, it was more irregular than the tetraploids because of its odd ploidy level. Its chromosomes associated as univalents, bivalents, trivalents and quadrivalents (Table 2). As many as 14 laggards and four micronuclei were observed at AI and TI, respectively. Micronuclei were present in about 20% of the spore tetrads. This plant probably originated from a cross between tetraploid and diploid cytotypes, where the female parent was an apomictic tetraploid and the male was a sexual diploid. The pentaploid hybrid resulted from the fertilization of an unreduced tetraploid egg by a haploid sperm from the diploid. In this plant, members of the extra haploid genome are present as either univalents or trivalents. In the trivalents, a haploid chromosome is synapsed with one of the members of a bivalent from the female parent. This synapsis represents homology between chromosomes of the haploid genome and members of one genome of the tetraploid parent.
This is the first report of chromosome numbers of 20, 40, and 50 for P. alcalinum. The only previously reported number for this species was 80 but this number is questionable because Saura (1941) counted 76 bodies and assumed the plant was an octoploid with 80 chromosomes. These findings demonstrated the cytological diversity of P. alcalinum. Including Saura's (1941) findings, only five accessions have been examined cytologically and four different ploidy levels were identified. If additional germplasm were available, other ploidy levels would undoubtedly be recovered. The 2n = 4x = 40 chromosomes in the P. falcatum accession is a new number for this species. The only previously reported number is 2n = 2x = 20 (Honfi et al., 1990). This is the first reporting of the chromosome number and meiotic behavior for P. paucifolium and P. polyphyllum. Saura (1941) reported 2x = 4x = 40 chromosomes for P. unispicatum, which agreed with the findings from this study.
Method of Reproduction
In all accessions of these eight species, a single nucellar cell in the micropylar end of the ovule enlarged and underwent meiosis to produce a linear tetrad of megaspores. The chalazal megaspore enlarged to become the functional megaspore and the three members nearest to the micropyle degenerated. In the diploid P. alcalinum, P. modestum, P. monostachyum, and P. repens accessions, the functional megaspore enlarged while its nucleus underwent three mitotic divisions to produce an eight-nucleate embryo sac of the Polygonum type. Antipodal cells at the chalazal end of the sac usually divided again producing a cluster of antipodals. However, in the diploid P. alcalinum accession, antipodal cells were not visible in several mature sacs. In the genus Paspalum, the absence of antipodal cells usually indicates aposporous sacs. Consequently, more than 100 young ovules in which megasporogenesis had occurred were examined and none exhibited nucellar activity. Apparently all of the mature sacs are probably of meiotic origin. Thus, the sacs in which antipodals were not visible were in early stages of deterioration and the antipodals had already degenerated. As expected because of their ploidy level, these four diploid accessions are sexual. The only previous information regarding the reproductive behavior of these species is that of Quarin and Hanna (1980). Using a different P. modestum accession, they reported that it was sexual and the female gametophyte deteriorated in 28% of the mature ovules. This is similar to the findings from this study (Table 4).
The method of reproduction in the tetraploid accessions [TABULAR DATA FOR TABLE 4 OMITTED] was more complex (Table 4). An eight-nucleate sac of the Polygonum classification, similar to those in the diploid species, developed in some ovules. However, in most ovules, shortly after megasporogenesis from one to seven nucellar cells in the area adjacent to the functional megaspore enlarged and their nuclei became prominent. One to three of these nucellar cells continued enlarging and their nuclei divided mitotically, producing aposporous sacs with an egg cell, two synergids cells and two polar nuclei. In some ovules, the functional megaspore deteriorated and only aposporous sacs developed, whereas in others, both the functional megaspore and a varying number of nucellar cells developed into mature female gametophytes (Table 4). Every tetraploid accession had some ovules in which the female gametophyte either failed to develop or deteriorated (Table 4). Cytologically, all tetraploids were facultative apomicts. The uniformity and variability observed in the progeny from each tetraploid accession supported the cytological classification of the reproductive behavior for each accession. Between 30 and 40% of the progeny from the P. falcatum, P. paucifolium, and P. polyphyllum accession were variable. However, all progeny from both tetraploid P. alcalinum accessions and the P. unispicatum accession were uniform.
The pentaploid 50-chromosome P. alcalinum accession behaved as an obligate apomict. Among the mature ovules, 88% had one or more aposporous sacs, whereas the female gametophyte had deteriorated in the other 12% (Table 4). Apparently the functional megaspore degenerated shortly after megasporogenesis in all of the ovules because no eight-nucleated Polygonum type sacs were observed. Results from the progeny test confirmed these findings because all the progeny were uniformed and appeared identical to the pentaploid accession.
Overall, seed set in these species was low (Table 3). Seed set in the sexual diploids was lower than expected since they were meiotically stable. The failure of the P. modestum accession to produce viable seed was associated with the high level of deterioration of the female gametophyte discussed earlier (Table 4). However, it will produce a few seed because seedlings have been recovered. Low seed set must be characteristic of this species because Quarin and Hanna (1980) reported similar findings for the P. modestum accession they investigated. Seed set in the P. monostachyum accession was surprisingly low. The data for the P. repens and diploid P. alcalinum accessions demonstrated that both are highly self-pollinated, which is normal for many sexual Paspalum species. The remaining species reproduced primarily by apomixis and these data demonstrated that apomicts do not necessarily have high seed set.
Findings from this study demonstrated the cytological diversity within the genus Paspalum. This is the first report of the chromosome number and meiotic behavior for P. monostachyum, P. paucifolium and P. polyphyllum. New chromosome numbers were found for P. falcatum and the three P. alcalinum accessions. Except for P. modestum, this is the first report of the reproductive behavior of these eight species. This study also revealed the reproductive diversity of these species, ranging from complete sexuality in the diploids to varying levels of facultative apomixis in the tetraploids to obligate apomixis in the pentaploid. These findings reaffirmed the importance of knowing the chromosome number, meiotic behavior and method of reproduction of warm-season grasses before they are used in plant improvement programs.
Abbreviations: PMCs, pollen mother cells: FAA, formalin-acetic acid-alcohol: AI, anaphase I; AII, anaphase II; TI, telophase I; TII, telophase II; [I.sub.2]KI, potassium iodide.
The authors expresses his appreciation to the following individuals for providing assistance and transportation while collecting germplasm in the following countries. Brazil: Prof. I. Barreto, Dr. A. Pott, Eng. H. Schreiner, and Dr. J.F.M. Valls; Paraguay: Ing. A. Borgognon and Mr. S.W. White; Uruguay: Ing. J.C. Millot and Prof. B. Rosengurtt. Appreciation is also expressed to Professors I. Barreto, C.L. Quarin, and B. Rosengurtt for their assistance in the identification of the germplasm.
Bashaw, E.C., and I. Forbes, Jr. 1958. Megasporgenesis, embryo sac development and embryogenesis in dallisgrass, Paspalum dilatatum Poir. Agron. J. 50:753-756.
Burson, B.L. 1991. Homology of chromosomes of the X genomes in common and Uruguayan dallisgrass, Paspalum dilatatum. Genome 34:950-953.
Burson, B.L., and H.W. Bennett. 1971. Chromosome numbers, microsporogenesis, and mode of reproduction of seven Paspalum species. Crop Sci. 11:292-294.
Burson, B.L., P.W. Voigt, and G.W. Evers. 1991. Cytology, reproductive behavior and forage potential of hexaploid dallisgrass biotypes. Crop Sci. 31:636-641.
Burton, G.W. 1940. A cytological study of some species in the genus Paspalum. J. Agric. Res. 60:193-197.
Chase, A. 1929. North American species of Paspalum. Contribution from the U.S. National Herbarium, vol. 28, pt. 1. U.S. Gov. Print. Office, Washington, DC.
Davidse, G., and R.W. Pohl. 1974. Chromosome numbers, meiotic behavior, and notes on tropical American grasses (Gramineae). Can. J. Bot. 52:317-328.
Honfi, A.I., C.L. Quarin, and J.F.M. Valls. 1990. Estudios cariologicos en gramineas Sudamericanas. Darwiniana 30:87-94.
Quarin, C.L. 1974. Relaciones citotaxonomicas entre Paspalum almum Chase y P. hexastachyum Parodi (Gramineae). Bonplandia 3:115-127.
Quarin C.L., and W.W. Hanna. 1980. Chromosome behavior, embryo sac development, and fertility of Paspalum modestum, P. bosciahum, and P. conspersum. J. Heredity 71:419-422.
Quarin, C.L., and G.A. Norrmann. 1987. Relaciones entre el numero cromosomas, su comportamiento en la meiosis y sistema reproductivo del genero Paspalum. Anales del IV Congreso Latinoamericano de Botanica (Bogota) 3:25-34.
Saura, F. 1941. Cariologia de algunas especies del genero Paspalum. U. Buenos Aires, Fac. Agron. y Vet., Inst. Genet. 2:41-48.
Tischler, C.R., and B.L. Burson. 1995. Evaluating different bahiagrass cytotypes for heat tolerance and leaf epicuticular wax content. Euphytica 84:229-235.
|Printer friendly Cite/link Email Feedback|
|Author:||Burson, Byron L.|
|Date:||Jul 1, 1997|
|Previous Article:||Genetic variation for traits related to temperature adaptation of rice cultivars.|
|Next Article:||The reproductive versatility of eastern gamagrass.|