| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a USDA-ARS, Crop Germplasm Research Unit, 430 Heep Center, Texas A&M Univ., College Station, TX 77843-2474
b Dep. of Soil & Crop Sciences, Texas A&M Univ., College Station, TX 77843-2474
c Yoder Bros., Inc., Chualar, CA 93925
* Corresponding author (b-burson{at}tamu.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Apomixis is an asexual form of reproduction in which seeds are produced without fertilization of the egg cell. In the aposporous form of apomixis, the female gametophyte (embryo sac) develops from an unreduced somatic nucellar cell in the ovule. Consequently, all cells, including the egg cell, in the mature female gametophyte of an apomict have an unreduced chromosome number (2n). In most apomicts, the unreduced egg cell initiates division near or at the time of anthesis and develops autonomously into an embryo without being fertilized by a sperm nucleus from the pollen; this is referred to as parthenogenesis.
Unreduced gametes occur in both sexual and apomictic plants. In sexual plants, most unreduced gametes result from the failure of chromosome reduction at the first meiotic division or the failure of cytokinesis at the second meiotic division during megasporogenesis and/or microsporogenesis. Regardless of the mechanism, the resultant unreduced "gametophyte" has the same chromosome number as the sporophyte. In apomicts, the development of an unreduced egg in the female gametophyte is normal; therefore, 2n + n fertilization should occur at a higher frequency in apomicts than in sexual plants.
In apomictic species where sexual genotypes are not available to use as the female parent, fertilization of an unreduced egg is a method of creating new genotypes. Intraspecific Kentucky bluegrass, Poa pratensis L., hybrids were produced by crossing facultative apomicts with highly apomictic ecotypes, and occasionally, an unreduced egg was fertilized by a reduced male gamete resulting in an apomictic 2n + n hybrid (Pepin and Funk, 1971). Some of these 2n + n hybrids were superior to the n + n hybrids as well as the existing apomictic ecotypes and were released as improved cultivars (Bashaw and Funk, 1987). Burson and Hussey (1996) reported that most 2n + n F1 hybrids between Pennisetum flaccidum Griseb. and P. mezianum Leeke were more winter hardy and produced more forage than the n + n F1 hybrids. These findings demonstrate that the fertilization of an unreduced egg has potential for producing superior apomictic germplasm. However, the unpredictability and low frequency of this event limits its usefulness for improving apomictic species.
The control of 2n + n fertilization is poorly understood. Pepin and Funk (1971) recovered high numbers of 2n + n intraspecific Kentucky bluegrass hybrids when pollinations were made before anthesis. They hypothesized that early pollination may have contributed to the fertilization of unreduced eggs in aposporous embryo sacs of the facultative apomictic female parent. Martinez et al. (1994) investigated the effect of pollination time on the recovery of 2n + n hybrids by pollinating stigmas of apomictic tetraploid bahiagrass, Paspalum notatum Flügge, from 1 to 5 d before anthesis. All 2n + n hybrids recovered were from florets pollinated 2 and 3 d before anthesis. They hypothesized that early pollination prevented parthenogenetic development of the unreduced egg cell and this permitted the egg to be fertilized. These findings suggest a relationship between time of pollination and the recovery of 2n + n hybrids, but neither study established definitively a relationship between these two events.
Buffelgrass is an important forage grass in the semiarid, subtropical areas of the world. It is an excellent species to investigate this purported relationship. The grass reproduces primarily by apomixis with the mechanism being apospory followed by pseudogamy (Fisher et al., 1954; Snyder et al., 1955). Most accessions are either obligate apomicts (Fisher et al., 1954; Snyder et al., 1955) or facultative apomicts (Bray, 1978; Sherwood et al., 1980); however, a few sexual plants exist (Bashaw, 1962). When apomictic buffelgrass is pollinated in controlled hybridization studies, 2n + n hybrids are sometimes recovered (Bashaw and Hignight, 1990; Bashaw et al., 1992; Hussey et al., 1993). Buffelgrass has a protogynous flowering behavior where its stigmas are exserted from the florets 1 to 3 d before anthesis depending on the genotype. The stigmas are receptive when exserted from the floret and remain receptive throughout the duration of the protogynous interval or until they are pollinated (Shafer et al., 2000). This study was undertaken to determine if the timing of pollination (days before anthesis) influenced the frequency at which 2n + n progeny are recovered in apomictic and sexual buffelgrass.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Two or three young leaves about 3 to 4 cm long were removed from each seedling and taken into the laboratory to determine the DNA content of each seedling. Common buffelgrass, 2n = 4x = 36, was used as an internal standard. A piece of leaf about 1 cm2 was cut from a leaf blade of each seedling and placed into a 15-cm petri dish along with a similar sample from common buffelgrass. A drop of a commercial buffer solution (Partec GmbH, Münster, Germany)1 was added to the dish, and both leaf samples (standard and unknown) were finely chopped with a double-edged razor blade until the tissue was thoroughly macerated. Approximately 0.5 mL of the buffer solution was added to the chopped tissue in the petri dish. This solution was resuspended several times with a pipette and filtered through a 30- µm filter to eliminate the debris. After 5 min, 2.5 mL of a commercial Partec DAPI (4'-6'-diamidino-2phenylindole) solution were added to the filtered buffer solution containing the nuclei and the solution was resuspended. After about 5 min, the sample was analyzed with a Partec CA II flow cytometer. A minimum of 600 cells was analyzed for each sample. Common buffelgrass was used as the internal standard for every seedling analyzed because its 2C DNA content (3.08 pg) and chromosome number (2n = 4x = 36) were known. It was possible to determine the ploidy level of the plant being analyzed by comparing the G1 peak of the unknown seedling with the G1 peak of the internal standard. Because PIs 409367, 409407, 295657, 315679, and S 2103 have 36 chromosomes, all offspring that resulted from regular apomictic reproduction (2n + 0) or sexual reproduction (n + n) would have 36 chromosomes. The DNA content of these tetraploid offspring was similar to that of the internal standard (common buffelgrass) and their G1 peaks were similar. With additional genomes, the 2n + n progeny would have a higher chromosome number (2n = 6x = 54) and their G1 peaks would differ from the standard (2n = 4x = 36).
Because PI 409704 is a pentaploid (2n = 5x = 45), common buffelgrass was used as an internal standard for all PI 409704 seedlings. Progeny resulting from apomictic reproduction (2n + 0) had a G1 peak associated with 45 chromosome plants that was higher than the G1 peak of the internal standard. Progeny with G1 peaks higher than the 45 chromosome plants were suspected 2n + n products. These plants were reanalyzed with Sorghum bicolor (L.) Moench as the internal standard which has a 2C DNA content of 1.74 pg.
The ploidy level of some of the predicted 2n + n plants was confirmed by counting the number of chromosomes in their root tip cells. Root tips were collected from plants growing in pots in a greenhouse between 0900 and 1000 h and prepared as previously described (Burson, 1991).
All data were analyzed with the PROC GLM and PROC REG procedures of the SAS Statistical Software Package (SAS, 1999). Regression analysis was used to determine the relationship between day of pollination and number of 2n + n hybrids recovered.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Nuclear DNA content of leaf cells was determined for 1962 self-pollinated plants of which 1766 had a mean 2C DNA content of 3.09 pg, nearly identical to that of the internal standard, common buffelgrass (3.08 pg). These plants were considered to be tetraploids with 36 chromosomes. One hundred sixty-eight plants were progeny from PI 409704 and their mean 2C DNA content was 4.15 pg which indicated they were pentaploids with 45 chromosomes. Of the remaining 28 plants, 25 had a 2C DNA content of 4.76 pg suggesting they were hexaploids with 54 chromosomes. Three plants from PI 409704 had a 2C DNA content of 5.52 pg, and they were thought to be aneuploids with approximately 67 chromosomes. These 28 plants are the products of 2n + n fertilization from self-pollination (Table 2).
|
Of the four apomictic accessions that produced 2n + n progeny, there was no significant difference in the overall frequency of 2n + n hybrids recovered. Across protogynous intervals, the frequency of 2n + n hybrids produced was 1.0, 1.8, 0.8, and 2.9% for PIs 409367, 409704, 295657, and 315679, respectively. When averaged across all genotypes, a complex polynomial relationship between the day of pollination and frequency of 2n + n hybrids recovered was observed (Fig. 1) . However, for the five apomictic accessions, the frequency of 2n + n hybrids recovered from the 0, 1, 2, and 3 d protogynous interval groups was 1.7, 0.7, 3.1, and 1.3%, respectively. The highest frequency of 2n + n fertilization was in those florets pollinated 2 d before anthesis (Fig. 1).
|
Cross-Pollinated Progeny
Percent germination of seed from the buffelgrass x birdwoodgrass crosses ranged from 23 to 91% with a mean of 60.9%. No relationship between germination and protogynous interval was observed.
The 2C DNA content was determined for 3072 cross-pollinated seedlings of which 2992 had a DNA content similar to their maternal parent (n + n and 2n + 0), while the remaining 80 plants had a higher DNA content (Table 3). Of the 3072 plants, 2721 had a 36-chromosome buffelgrass plant as their maternal parent. Of these 2721 plants, 2649 had a mean 2C DNA content of 3.11 pg compared with 3.08 pg for the common buffelgrass standard. The remaining 72 offspring had a mean DNA content of 4.77 pg suggesting these plants are hexaploids with 54 chromosomes (36 + 18 = 54) and are the products of 2n + n fertilization. The 45-chromosome accession, PI 409704, was the maternal parent of the remaining 279 cross-pollinated seedlings, and 271 of these had a mean DNA content of 4.12 pg, which is similar to that of the maternal parent (4.15 pg). The remaining eight plants had a mean DNA content of 5.41 pg indicating that they were septaploids with 63 chromosomes (45 + 18 = 63) resulting from fertilization of unreduced gametes (Table 3).
|
There was a difference among the five apomictic accessions for the number of 2n + n hybrids recovered (Table 3). If the protogynous intervals are ignored and only the total number of 2n + n hybrids from each cross are considered, the mean frequency of 2n + n fertilization was 0.5, 0.5, 2.9, 3.4, and 3.5% for PIs 409367 x 193444, 409407 x 193444, 409704 x 193444, 295657 x 193444, and 315679 x 193444 hybrids, respectively. These data demonstrate that a higher frequency of 2n + n hybrids was produced by the latter three crosses. Because accessions PIs 409704, 295657, and 315679 have longer protogynous intervals, data indicate the length of the protogynous interval influences the frequency of 2n + n fertilization when cross-pollinated. When the number of 2n + n hybrids recovered from the different protogynous intervals are examined across hybrids, this pattern is even more evident. Across all five apomictic accessions, the frequency of 2n + n fertilization at 0, 1, 2, and 3 d protogynous intervals was 1.0, 2.0, 4.3, and 4.0, respectively. This demonstrates that as the length of the protogynous interval increases up to 2 or 3 d, the number of 2n + n hybrids produced also increases (r2 = 0.86).
The highest frequency of 2n + n fertilization (8.2%) occurred when the stigmas were pollinated 2 d before anthesis in the PI 409704 x PI 193444 cross (Table 3). In comparing the number of 2n + n hybrids recovered to the different protogynous intervals when the pollinations were made for all crosses, the frequency of 2n + n fertilization was higher when pollinations were made 2 or 3 d before anthesis (Table 3; Fig. 1). The only exception was the PI 409367 x PI 193444 cross.
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The recovery of three 2n + n hybrids from the S 2103 x PI 193444 cross was unexpected (Table 3). An argument could be made that the unreduced eggs that were fertilized were not the products of sexual reproduction and that S 2103 is a facultative apomict. If that were true, then apomictic sacs could be produced and unreduced eggs in these sacs could have been fertilized to produce the 2n + n individuals. This is unlikely because more than 1000 mature ovules from this accession have been examined cytologically and no evidence of apomictic development has been observed. Another possibility is that the three hybrids did not result from the fertilization of an unreduced egg (2n + n) but are the products of the fertilization of a reduced egg by an unreduced sperm (n + 2n). However, unreduced male gametes have not been reported in PI 193444. Further, if this accession produced unreduced male gametes, some of the hybrid offspring from the five apomictic accessions should have been octoploids (2n = 8x = 72; 2n + 2n) rather than hexaploids (2n = 6x = 54; 2n + n), but all the offspring were either tetraploids (2n + 0) or hexaploids (2n + n) (Tables 2 and 3). Therefore, the three hexaploid S 2103 x PI 193444 hybrids probably originated from the fertilization of an unreduced egg, suggesting that unreduced female gametes occasionally develop in sexual genotypes of this species.
Another concern is the 2n + n plants recovered from both self- and cross-pollinations made on the day of anthesis (Tables 2 and 3). If early pollination is the primary criterion for the fertilization of an unreduced egg, how did these 2n + n plants originate since they were pollinated on 0 d? This behavior has been reported in other grass species (Harlan and de Wet, 1975; Espinoza and Quarin, 2000). However, depending on the accession, the frequency of this event in this study ranged from 0.9 to 3.8% (Tables 2 and 3) which is higher than what normally occurs in other species. An explanation for these higher frequencies is that the controlled pollinations (both self and cross) were made approximately 4 h before anthesis. This 4-h period may be enough time to allow the unreduced egg cell to be fertilized before the onset of parthenogenesis because the time required from pollination until the pollen tube enters the embryo sac is between 2 and 3 h for most buffelgrass accessions (Shafer et al., 2000).
Findings from this study demonstrate unequivocally that early pollination of apomictic buffelgrass by birdwoodgrass increases the frequency of the fertilization of an unreduced egg. Besides supporting Martinez's et al. (1994) findings in apomictic bahiagrass, this study provides credibility to their theory because a greater number of plants were examined in this study. Martinez et al. (1994) examined only 59 progeny, whereas we examined 5034 offspring. The frequency of 2n + n fertilization was higher in bahiagrass, but the large number of buffelgrass plants analyzed demonstrates this phenomenon is not an aberration.
Upon establishing a relationship between early pollination and an increased frequency of 2n + n fertilization in apomicts, close examination of the products from crosses where other protogynous, apomictic species were used as the female parent tend to support this hypothesis. For example, 2n + n hybrids were recovered from interspecific crosses with protogynous, apomictic accessions of Pennisetum orientale L. and P. flaccidum as the female parents (Bashaw et al., 1992; Hussey et al., 1993; Wipff, 1995). Kindiger and Dewald (1994) reported the frequency of 2n + n fertilization in some eastern gamagrass, Tripsacum dactyloides (L.) L., crosses as high as 27%. Eastern gamagrass is a monecious species with the female and male florets on the same inflorescence, and the stigmas of the female flowers normally are exserted several days before anther exsertion in the male flowers. This protogynous flowering behavior allows the female florets to be pollinated early and may account for the high number of 2n + n plants. As mentioned earlier, a higher frequency of 2n + n hybrids was recovered when apomictic Kentucky bluegrass plants were pollinated early (Pepin and Funk, 1971). However, early pollination does not guarantee 2n + n fertilization in an apomict. Espinoza and Quarin (2000) hand emasculated and pollinated common dallisgrass, Paspalum dilatatum Poir., florets 2 to 3 d before anthesis but no 2n + n hybrids were produced. The only 2n + n hybrids recovered were from crosses made on the day of anthesis. Apparently the success of 2n + n fertilization from early pollination is species dependent. There also is a difference among genotypes within a species and whether or not they are self- or cross-pollinated (Tables 2 and 3).
A plausible explanation for this phenomenon has not been presented. The only hypothesis that attempts to explain why early pollination increases the chance of the fertilization of an unreduced egg was proposed by Martinez et al. (1994). They hypothesized that early pollination prevented the parthenogenetic development of the unreduced egg cell and this allowed the unreduced egg cell to be fertilized. Using electron microscopy, Vielle et al. (1995) studied the ultrastructural changes that occur in the egg apparatus of sexual and apomictic embryo sacs of buffelgrass before and shortly after pollination. Their findings provide insight into why the frequency of the fertilization of unreduced eggs in apomictic embryo sacs increase when pollinated early. In aposporous sacs, a cell wall completely covered the plasma membrane of the unreduced egg cell several hours before a pollen tube entered the female gametophyte. Whereas, in sexual embryo sacs, a cell wall only covered the micropylar end of the reduced egg cell and the plasma membrane of the cell was exposed on the chalazal end. Even 4 h after pollination, the cell wall had not covered the chalazal end of the reduced egg cell. These findings provide a plausible explanation for this phenomenon in that a cell wall that forms around the unreduced aposporous egg cell provides a physical barrier that prevents the pollen tube from penetrating the cell. Since this occurs before anthesis, it prevents the unreduced egg cell from being fertilized. However, if the stigmas are pollinated before the cell wall has developed and before the initiation of parthenogenesis, the unreduced egg cell can be fertilized. This may explain how the frequency of 2n + n fertilization is increased when the stigmas are pollinated 2 or 3 d before anthesis.
Findings from this study demonstrate that when apomictic buffelgrass is pollinated with birdwoodgrass pollen, early pollination increases the frequency of 2n + n fertilization. There are differences between genotypes and an effect of the male parent on this event. However, in protogynous, apomictic buffelgrass, when pollinations are made 2 or 3 d before anthesis, the frequency of 2n + n fertilization is high enough that this approach could be used as a tool to create new apomictic genotypes. This reproductive phenomenon may have application as a breeding tool to improve other apomictic species where sexual types are not available to use as the maternal parent in a cross.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 Mention of a trade mark of a proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture and does not imply its approval to the exclusion of other products that may also be suitable. 
Received for publication February 21, 2001.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M.A. Hussey and B.L. Burson Registration of 'Frio' Buffelgrass Crop Sci., January 1, 2005; 45(1): 411 - 412. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Plant Registrations | Soil Science Society of America Journal | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
