| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Institute of Experimental Botany, Sokolovska 6, 77200 Olomouc, Czech Republic
b Dep. of Botany and Plant Sciences, Univ. of California, Riverside, CA 92521 USA
* Corresponding author (adam.lukaszewski{at}ucr.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Species in the genus Festuca are known for complementary characteristics to those of Lolium, such as good persistence and high stress tolerance (Thomas and Humphreys, 1991; Thomas et al., 2003). Members of the two genera can be intercrossed and fertile progeny can be obtained. In hybrids, the homeologous chromosomes are capable of surprisingly frequent pairing (Jauhar, 1993) and recombination (Zwierzykowski et al., 1998a, 1998b, 1999a). Genetically, the chromatin of the two genera seems fully interchangeable and there is little evidence for selection against chromosomes of one species over the other (Zwierzykowski et al., 1998b, 1999a). Hence, combining of the gene pools of the two genera for practical manipulation is possible. Breeders of forage grasses have been exploiting intergeneric Lolium–Festuca hybrids for close to a century and developed numerous cultivars with improved parameters. In turfgrasses, interspecific hybridization has been used sparingly (Brilman, 2001; Belanger et al., 2003). However, screening of individual plants in populations of Lolium–Festuca hybrids developed for forage revealed an enormous range of variation for every observable characteristic, including those of interest in the turf industry such as vigor, color, leaf length, and width. Many types were identified which seemed highly desirable for the development of turfgrasses.
For practical reasons, the early-stage turfgrass breeding with the Lolium–Festuca hybrids is being done at the tetraploid level. The original hybrids were made for forage grass breeding (Zwierzykowski, pers. comm.) where larger plants and enhanced vigor of tetraploids are positive characteristics. Tetraploid hybrids are easier to produce, are more fertile and tetraploidy may enhance intergeneric recombination as well as mitigate selection pressures, if any, that could operate against some combinations of chromatin from the two parental genera. In breeding of turf grasses, reduced vigor and plant size are more advantageous. It was, therefore, desirable to reduce the ploidy level of the hybrids to diploid, if an earnest germplasm enhancement effort for turf was to commence. The rapid intergeneric crossover rate in the Lolium x Festuca hybrids (Zwierzykowski et al., 1998b) made it likely that in the span of several generations at the tetraploid level the parental genomes were thoroughly recombined. The resulting diploids, if sufficiently fertile, could be subject to direct further breeding as turf grasses, or be used in crosses to the existing turf cultivars of the parental species, especially L. perenne.
Of several techniques for the reduction of ploidy levels in crop species, in the Lolium–Festuca complex only anther or microspore cultures appear available. In vitro anther culture has been used in similar wide hybrids of grasses (Lesniewska et al., 2001; Rose et al., 1987; Zare et al., 2002; Zwierzykowski et al., 1999b), yet it still has many problems in routine use. These include genotype-specific responses with a high frequency of recalcitrant genotypes, low induction rates of androgenesis and low regeneration capacity of the microspore-derived embryos, a high frequency of albino plants, and spontaneous polyploidization enhanced by long culture periods of embryos, especially when the callus stage is involved.
Of all published studies on androgenesis in interspecific and intergeneric hybrids of grasses, only one reported on successful regeneration of haploid green plants in L. perenne x F. pratensis hybrids and the regenerants were recovered through the callus stage (Rose et al., 1987). While their ploidy level has not been reported, the callus stage in tissue culture is generally associated with increased spontaneous chromosome doubling rates and with aneuploidy (Logue, 1996). In the context of this study, spontaneous chromosome doubling, usually considered a benefit, would be a clear detriment. The experiments were designed to produce haploid progeny of L. multiflorum x F. pratensis and F. pratensis x L. perenne through the microspore embryo stage, thus avoiding callus formation. In addition, given limited experience with androgenesis of these two hybrids, the experiments were designated to test such culture parameters as the effects of the pretreatment, of the induction medium and the density and position of cultured anthers on culture media. Finally, the ease of discrimination of the parental chromosome segments in these hybrids by the in situ probing with total genomic DNA (Humphreys et al., 1995) made it possible to verify the origin of all classes of regenerants and to assess the status of genome recombination in the hybrids.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In addition to the 23 clones described above, attempts were made to generate androgenic progeny from tetraploid L. multiflorum cv. Mitos, also obtained from Dr. Zwierzykowski, IPG, Poland, and from a diploid turf breeding line SRX4MO97 of L. perenne, kindly provided by Dr. L.A. Brilman, Seed Research of Oregon, Corvallis, OR, USA.
Pretreatment and Anther Culture
Inflorescences deemed to be at the appropriate stage of development were cut from the donor plants and stored in an incubator at 4°C in the dark for 7, 14, or 21 d. After cold storage, the developmental stage of microspores was determined on acetocarmine squashes of selected anthers. Spikelets with anthers containing uninucleate microspores were surface sterilized in 70% (v/v) ethanol for 30 s followed by sterilization in approximately 2.6% (v/v) aqueous solution of sodium hypochlorite for 10 min, and rinsed three times in sterile distilled water. Excised anthers were plated on the surface of the induction medium in 90-mm Petri dishes at a density of about 100 anthers per plate. One half of the anthers in each Petri dish were cultivated in a vertical position, where only one locule of each anther was in a direct contact with the medium, and these anthers were plated in high density, close to one another. The other half of the anthers were cultivated in a horizontal position where both locules were in contact with the medium, and at a low anther density. Plated anthers were incubated at 25°C in darkness.
Culture Media
The induction media were the C17 medium (Wang and Chen, 1983) modified with 12 g/L of maltose and 1.5 mg/L 2.4-D (2,4-dichlorophenoxyacetic acid), and 0.5 mg/L kinetin and a modified MN6 medium based on the N6 medium (Chu, 1978) supplemented with 8 g/L of maltose, 2 mg/L 2,4-D, 0.5 mg/L kinetin and 0.5 mg/L NAA (naphthaleneacetic acid). The differentiation medium was 190-2 (Zhuang and Jia, 1983) with 5 g/L maltose instead of sucrose. For the induction of root growth, the same 190-2 medium was used but with standard 3 g/L sucrose and free of hormones. All media were solidified by 3.5 g/L Gelrite (Monsanto Company, St. Louis) and autoclaved at 121°C for 20 min.
Plant Regeneration
After 5, 7, and 9 wk of incubation on the induction medium, observations were taken on the androgenic capacity of the cultures. Androgenic capacity was understood as the percentage of anthers with developing embryos and, rarely, with plantlets developed from the embryos.
Well-developed embryos were transferred to Petri dishes with the differentiation medium and incubated at 25°C under a 12-h day/night photoperiod. After about 30 d on the differentiation medium, observations were made on the numbers of albino and green plants present. The number of green plants per 100 plated anthers was taken as a parameter of the efficiency of anther culture. Developing plantlets were transferred from the differentiation medium to either test tubes or Erlenmeyer flasks with the rooting medium. Rooted plantlets were potted and transferred to the greenhouse.
During handling of the rooted plants, in transplantation from test tubes to small pots, or from small to larger pots, seemingly single plants separated in several instances in into two to five individual segments, each one with its own root system. To determine the nature and origin of such "sets" of plants, each member of a "set" was individually potted, labeled, and analyzed cytologically.
Cytology
All plants for cytological screening were transferred from pots to a hydroponic culture with aerated full strength Gromagnon 9-5-18 All-Purpose-Nutrient solution from American Hydroponics, Arcata, CA. After 5 to 7 d, root tips from actively growing plants were collected to ice water for 26 to 28 h, fixed in a 3:1 mixture of absolute ethanol and glacial acetic acid, and treated as recommended by Masoudi-Nejad et al. (2002).
Each cytological preparation was made from a single root tip or a fraction of a single root tip. For chromosome counts, root tips were squashed in a drop of 2% (v/v) acetocarmine and observed under a microscope, aided by phase contrast if needed. For GISH, the preparations, probes and blocks as well as all treatments were made according to Masoudi-Nejad et al. (2002). In all experiments, DNA of F. pratensis was labeled with digoxigenin and used as a probe; DNAs of the Lolium species were used as blocks. The standard probe to block ratio was 1:150 with minor deviation from experiment to experiment. The detection of the hybridization signal was with the Anti-DIG-FITC conjugate; counterstaining was with propidium iodide (PI) at a concentration of 2% in various antifade solutions. Observations were made under a Zeiss Axioscope 20 equipped with epi-fluorescence, recorded with a SPOT RT Color digital camera from Diagnostic Instruments Inc., and processed using the SPOT Advanced and Adobe Photoshop v. 6 software.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Cytological Analyses
Chromosome counts were made in 330 androgenic plants. Of these, 23 were from L. multiflorum, 66 from the LmFp hybrid, and 241 from the FpLp hybrids. The latter group included five sets, each one consisting of two to four plants, for a total of 15 individual plants. The chromosome numbers of the regenerants ranged from 13 to 56, with a majority being in the diploid (about 14 chromosomes, 85%) and tetraploid (about 28 chromosomes, 14.5%) range (Table 2). Aneuploids were present in both groups of regenerants; hyperploids (15 and 16 chromosomes) were more common among diploids while hypoploids (27 chromosomes) were more common among tetraploids. Among the 23 regenerants of L. multiflorum tested, 11 were diploid, 11 were tetraploid and one plant had 56 chromosomes, probably a result of two consecutive rounds of chromosome doubling. Among 66 regenerants of LmFp, the proportions of diploids to tetraploids were 49: 17 (25.7% tetraploids) while among the 241 regenerants of FpLp there were only 20 tetraploids (8.3%). One plant had 41 chromosomes; its origin is unclear. Of the five sets analyzed, three had 14 chromosomes, one had 15, and one had 28.
|
|
Among the entire sample of cytologically identified plants only a single instance of a somatic chimera for chromosome number or structure was observed. On a single preparation, cells were found with 14 and 28 chromosomes. The patterns of probe hybridization to the chromosomes were identical in the 14 and 28 chromosome cells, excluding sample mix up as a cause of variation.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Early experiments with androgenesis in grasses were not encouraging. Nitzsche (1970) obtained only one green plant from anther culture of L. multiflorum x F. arundinacea hybrid and little improvement was seen soon afterwards (Zenkteler and Misiura, 1974). The first success was reported by Rose et al. (1987) who produced small populations of green plants from L. multiflorum x F. pratensis cv. Elmet and L. perenne x F. pratensis cv. Prior hybrids on the potato extract medium (Chuang et al., 1978) supplemented with 2.4-D and kinetin. Good success rate with androgenesis of F. pratensis x L. multiflorum hybrids was reported by Lesniewska et al. (2001).
The experiments described here, while undertaken with a practical goal in mind, were designed to test several factors and conditions known to affect the efficiency of the androgenesis. The anther culture response is very genotype-dependent (Rose et al., 1987). This experiment was no different: L. perenne gave no green plants, whereas 25.5 green plants per 100 cultivated anthers were recovered in the LmFp. A failure of regeneration in L. perenne might have also been related to the ploidy level. Rose et al. (1987) discussed the relationship between the ploidy level of donor plants and androgenic response. Cold pretreatment of the donor material is generally believed to enhance the androgenic response (Rose et al., 1987; Bante et al., 1990). Sunderland (1978) suggested that the switch from the gametophytic to the sporophytic development taking place during stress pretreatment results from a reduction in the levels of endogenous hormones. Here, a 2- or 3-wk cold pretreatment clearly enhanced the regeneration efficiency. However, Zwierzykowski et al. (1999b) used no cold pretreatment in similar hybrids and their regeneration capacity in the best responding clone was still 28.8 green plants per 100 anthers plated.
The effect of the culture medium on the induction of the androgenic response and the regeneration of plants from embryos or calli is frequently discussed. Particular attention has been given to the concentration and the type of the carbon source and, especially, to the presence of growth regulators. The most commonly used induction media are modifications of PII (Chuang et al, 1978), but the best results reported were for modifications of C17 (Wang and Chen, 1983). Since the report of Rose et al. (1987), 1.5 to 2 mg/L of 2.4-D and eventually 0.5 mg/L of kinetin were used in most experiments (Bante et al., 1990; Halberg et al., 1990; Andersen et al., 1997; Zwierzykowski et al., 1999b). Although growth regulators may not be necessary for the induction of androgenesis, a combination of an auxin and a cytokinin dramatically increases the androgenic response.
A serious problem in the anther culture of cereals and grasses is the formation of albino plants. Day and Ellis (1985) detected deletions or alterations in a part of the chloroplast genome of albino plants of barley (Hordeum vulgare L.) and wheat (Triticum aestivum L.). The regeneration of green and albino plants seems to be affected by different genes localized on different chromosomes (de Buyser et al., 1992) and may be more affected by the environment than by genetics (Opsahl-Ferstad et al., 1994). This is consistent with the general concept of albinism, as being sustained by mutations induced by the in vitro conditions in addition to effect of nuclear genes. In this study, albinism was a problem only in the diploid control; its frequency in the critical material was negligible.
In most haploidization experiments, spontaneous doubling of chromosome numbers is perceived as a benefit, albeit an unpredictable one. It is a common phenomenon for plants regenerated from anther culture, including Poaceae. Polyploidization usually takes place during the callus phase of plant regeneration or during the subsequent plantlet development (Bante et al., 1990). The reported frequencies were in the range of 70% for regenerants from diploid L. multiflorum and about 50% for a tetraploid genotype (Bante et al., 1990) whereas Halberg et al. (1990) reported 50 to 60% diploids among regenerants from diploid L. perenne. In the most recent studies, the reported spontaneous doubling rates among regenerants from Festuca–Lolium hybrids were 14 to 25% in Zwierzykowski et al. (2001) and 12% in Lesniewska et al. (2001). In this study, spontaneous doubling of chromosome numbers was a highly undesirable consequence of anther culture. Fortuitously, its incidence was very low in the most critical hybrid here, F. pratensis x L. perenne, and most (about 91%) of progeny were haploid. According to Lesniewska et al. (2001), some of these haploids can be expected to be fertile and will be used directly in hybridization with turf cultivars of L. perenne.
In all plant regeneration experiments, the nature and the tissue origin of the regenerated progeny are of primary interest. While there is never much doubt that haploids from anther culture must originate from some products of meiosis, the origin of diploids (that is, plants with the same chromosome numbers as the somatic chromosome number of the donor) is never certain without additional tests. Such diploids can, at least in theory, regenerate from the somatic tissue of the anther and be, therefore, clones of the parental plants, or originate by fusion of two products of meiosis and, consequently, be no different from any sexually derived progeny. The origin of the diploid progeny can be tested by segregation or genetic markers studies (Andersen et al., 1997), always a time-consuming and expensive proposition. The unique features of the Lolium–Festuca hybrids, with their very high frequency of homeologous recombination (Canter et al., 1999; Zwierzykowski et al., 1998a, 1998b, 1999a) and easy discrimination of the parental chromatin by GISH (Humphreys et al., 1995) offer a unique opportunity to determine unequivocally the nature and origin of each individual plant. This is not a trivial matter. A cytological study of 17 anthers-derived progeny of F. arundinacea x L. multiflorum hybrids identified two with such unusual karyotypes that their origin could not be explained (Zwierzykowski et al., 1998a).
All haploids karyotyped in this study had unique patterns of parental chromatin even though some similarities of chromosomes from plant to plant were observed. Clearly, the haploids were recovered from microspores; karyotype similarities reflect the relationships of the donor plants. All analyzed tetraploids had clearly identifiable pairs of chromosomes (Fig. 1D,E). Therefore, they were doubled haploids originating from microspores, after a single round of chromosome doubling. Perfect disomy for every chromosome present excludes somatic tissue and diads as possible sources of these plants. The donor plants are cross pollinating and were produced by open pollination of selected individuals. As it has been shown earlier (Zwierzykowski et al., 1998b), advanced generation Lolium–Festuca hybrids show extensive recombination of the parental genomes and wide plant-to-plant variation in recombination patterns along their chromosomes. Regenerants from the somatic tissue of such plants would be expected to show wide polymorphism for their chromosome structure. No such variation was detected in any of the tetraploids recovered. Homozygosity for chromosome structure among recovered tetraploids also excludes the contribution of diads. Diads may still be present among cultured anthers. However, plants originating from diads would be expected to show polymorphism for their chromosome structure equivalent to the rate of homeologous recombination (recombined vs. non-recombined chromatids).
The presence of several clones among the regenerants was not entirely surprising even though routine steps were taken to avoid it. In the study of androgenic progeny of F. arundinacea x L. multiflorum hybrids (Zwierzykowski et al., 1998a), three clones comprising a total of 13 plants (essentially one half of all plants analyzed) were observed. In this study, the frequency of clones was considerably lower. Additionally, with one exception of a tetraploid, all clones had reduced vigor and are not expected to contribute to the next generation. Therefore, their presence did not significantly affect the efficiency of regeneration.
Among the recovered progeny, both in the diploid (14-chromosome) and the tetraploid (28-chromosome) range, aneuploids were present. Plants with 13 and 15 to 17 chromosomes probably represent aneuploid microspores generated by irregularities of meiosis. Despite the presence of two genomes each from two distinct genera, these plants behave meiotically as autotetraploids. While no good data exist for the aneuploid frequency among sexually derived progeny of such hybrids, aneuploid frequencies observed here do not appear excessive. On the other hand, 27-chromosome plants that clearly were derived by spontaneous doubling of chromosome numbers must represent instances of somatic loss of single chromosomes. This was surprisingly frequent in the analyzed sample (four 27-chromosome plants among 37 tetraploids of the Lolium–Festuca hybrids analyzed). It is, therefore, possible, that some aneuploids in the diploid range were also generated by a somatic and not by a meiotic chromosome missegregation.
In the process of GISH analysis of the origin of the regenerants, a survey was taken of the frequencies of complete parental and recombined chromosomes and the frequency distribution of the translocation breakpoints on the latter. As in the study of the F8 Festulolium hybrids developed for forage (Zwierzykowski et al., 1998b) and in the early generations of tetraploid F. pratensis x L. perenne hybrids (Canter et al., 1999; Zwierzykowski et al., 2003), also in this study, an approximate 50:50 ratio of chromatin of both parental species was deduced. The LmFp population had more translocation breakpoints and fewer complete chromosomes of the parents than the FpLp population (Fig. 1), but it is advanced by at least five generations over the FpLp hybrids, and all progeny were regenerated from a single clone hence the sample may not be representative. Among the FpLp hybrids, the average numbers of parental chromosomes per haploid (n = 2x = 14) plant were 4.16 for F. pratensis, 3.83 for L. perenne, and 6.0 for Lolium–Festuca translocations, with the average of 9.6 translocation breakpoints per genome or 1.6 per translocated chromosome. A comparison between the current and the previous generation of the FpLp hybrids (data not shown) indicated that the average number of translocated Lolium–Festuca chromosomes per plant increased by 2.3 and the average number of translocation breakpoints per translocated chromosome increased by 0.4. It is clear that genome recombination is still active in these plants and that even with extensive recombination and after several cycles of selection, most, if not all, chromatin of both parents is still present. This suggests that the germplasm enhancement of turf L. perenne is quite possible and a reduction of the ploidy level to diploid will permit a rapid progress of selection.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 15, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ców F1 i odmian Festulolium. Zesz. Probl. Postepow Nauk Roln. 474:47–54.This article has been cited by other articles:
|
|
 |
|
  M. R. Malik, F. Wang, J. M. Dirpaul, N. Zhou, J. Hammerlindl, W. Keller, S. R. Abrams, A. M. R. Ferrie, and J. E. Krochko Isolation of an embryogenic line from non-embryogenic Brassica napus cv. Westar through microspore embryogenesis J. Exp. Bot., July 1, 2008; 59(10): 2857 - 2873. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
