Journal of Experimental Botany, Vol. 53, No. 371, pp. 1119-1129,
May 2002
© 2002 Oxford University Press
Original Papers |
Antirrhinum majus microspore maturation and transient transformation in vitro
1Vienna Biocenter, Institute of Microbiology and Genetics, Vienna University, Dr Bohrgasse 9, A-1030 Vienna, Austria
2Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, 117871 Moscow, Russia
Received 16 October 2001; Accepted 21 December 2001
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The male gametophyte of higher plants represents an excellent system to study gene regulation, cell fate determination and cellular differentiation in plants because of its relative simplicity compared to the sporophyte and its accessibility for cytological and molecular analysis. Unicellular plant microspores are single haploid cells, which can be isolated in large amounts at a defined developmental stage. Microspores cultured in vitro in a rich medium develop into mature pollen grains, which are fertile upon pollination in vivo. It is reported here that isolated Antirrhinum majus microspores when cultured in an optimal medium develop to form mature, fertile pollen. Their development closely resembled that of pollen formed in vivo. Isolated microspores were bombarded with Aquorea victoria Green Fluorescent Protein (GFP), Discosoma Red Fluorescent Protein (dsRFP) and ß-glucuronidase (GUS) reporter genes under the control of various promoters and transient expression was observed throughout pollen development in vitro. Bombarded and not bombarded in vitro-matured pollen grains were able to germinate both in vitro and on receptive stigmas and to set seed. The protocol of maturation, transient transformation and germination of Antirrhinum majus pollen in vitro described here provides a valuable tool for basic and applied research.
Key words: Antirrhinum majus L., Green Fluorescent Protein (GFP), in vitro maturation, microspores, Red Fluorescent Protein (dsRFP), transformation.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The development of the plant male gametophyte follows a tightly controlled sequence of events within the anther, which can be divided into two major processes: microsporogenesis and microgametogenesis (Bedinger, 1992). Microsporogenesis begins with meiosis and ends with the formation of polarized haploid microspores. During microgametogenesis the unicellular microspore divides asymmetrically to give a young pollen grain with a vegetative and generative cell, which then differentiates to mature bicellular or tricellular pollen.
Isolated in vitro microspore cultures contain only single haploid cells in large amounts making the use of microbiological techniques and single cell selection possible (Dunwell, 1996). In addition, highly synchronous and homogenous microspore populations can be obtained using careful selection of flower buds or gradient centrifugation (Touraev et al., 1996). Since isolated microspores are separated from the surrounding anther wall tissues, it is possible to study the effect of medium components on microspore performance directly. When tobacco microspores are isolated from anthers and cultured in vitro in a rich medium, they develop into mature pollen grains, which are fertile upon pollination in situ (Benito Moreno et al., 1988). Their development closely resembles that of pollen formed in vivo.
Microspore maturation in vitro provides a number of research opportunities for basic and applied science (Heberle-Bors et al., 1996). The absence of the participation of the anther wall in pollen formation makes it possible to investigate the developmental events and requirements for external nutrients, and to determine the gametophytic or sporophytic origin of pollen components. In addition, in vitro pollen maturation has also been used for pollen selection (Touraev et al., 1995) and for the development of a new transformation system, MAGELITR (Touraev et al., 1997). Unicellular microspores were transformed with antibiotic resistance and reporter genes, matured in vitro until the formation of fully fertile pollen that was used for pollination, and the transgenic seeds obtained were selected for antibiotic resistance. The establishment of an efficient in vitro maturation protocol is the key element of male germ line transformation.
Antirrhinum majus is a well-known classical model species in plant biology (Carpenter and Coen, 1990). The large number of morphogenetic mutants and genes controlling plant architecture (squamata), floral organ formation (deficiens, plena), structure (deficiens), symmetry (cycloidea, centroradialis), transition from inflorescence to floral meristems and identity of the floral primordium (floricaula, squamata, squamosa), floral pigmentation (pallida) and anthocyanin biosynthesis genes (delila, eluta), were identified in snapdragon (Carpenter and Coen, 1990; Luo et al., 1995). Mutants affecting male reproductive organ formation have also been described (Davies et al., 1999; Doonan, personal communication) and an in vitro pollen maturation system would facilitate their detailed analyses. Unfortunately, not much is known about Antirrhinum male gametophyte development and only a few studies concerning the ultrastructure of the tapetum (Lombardo and Carraro, 1976), and analysis of the presence and content of plastid and mitochondrial DNA (Corriveau et al., 1990; Sangwan and Sangwan-Norreel, 1987) have been published.
One of the major limitations in the use of Antirrhinum majus as a model is the non-availability of an efficient, genotype-independent, simple method of transformation. The few reported protocols are all tissue culture-dependent and restricted to a few genotypes (Cui et al., 2001; Heidmann et al., 1998). Therefore, male germ line transformation would be a genotype-independent efficient alternative for Antirrhinum majus transformation. The microspore maturation system is also very attractive for testing the expression of promoters or genes using transient transformation (Ottenschlager et al., 1999).
This work reports the optimization of the in vitro maturation of isolated Antirrhinum microspores until the formation of fertile mature pollen, which facilitates a detailed analysis of pollen mutants and is the key step of a novel transformation method for Antirrhinum majus. Isolated microspores and pollen grains at various stages of development were transformed with GUS, GFP and dsRFP genes, and transient expression was observed throughout in vitro pollen development. In vitro-matured transformed pollen was shown to germinate in vitro and, after pipette pollination onto the surface of receptive stigmas, to enter the style and perform fertilization to produce seeds.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Growth of donor plants
Seeds of the Antirrhinum majus L. cv. 165 E (niv::Tam3) were kindly provided by Dr Z Schwarz-Sommer (MPI, Cologne, Germany). Plants were grown on soil (3400 cm3, 18 cm diameter) in a growth chamber at the temperature of 18 °C during the day, and 15 °C during the night, with a daylength of 16 h, 60% humidity, and a light intensity of 10000 lx.
FDA and DAPI staining of developing pollen
Microspores and developing pollen were isolated from flower buds of various sizes and stained in FDA (fluorescein diacetate, Sigma, USA) solution (5 µg µl-1) for 1 min and observed under a fluorescent microscope (Leitz-DIAPLAN), using the FITC filter channel to determine viability. The stage of pollen development was identified in microspore and pollen samples fixed in a solution of 96% EtOH:glacial acetic acid (3:1 v/v) for 10–30 min, washed once in 70% EtOH by centrifugation at 150 g for 2 min, stained in DAPI (4', 6-diamidino-2-phenylindole, Partec, Germany) and observed under the fluorescent microscope.
Determination of the duration of pollen development in planta
Flower buds with microspores at the unicellular stage were labelled in planta using a paper marker and the development of flowers was followed until the stage of anther dehiscence and opening in order to determine the exact time necessary for pollen maturation in planta. Samples of developing pollen were taken every day and the stage of pollen development was determined by staining with DAPI and correlated with the bud size.
In vitro maturation of isolated Antirrhinum microspores
Flower buds of Antirrhinum majus L. (6–7 mm) at the unicellular stage of pollen development were surface-sterilized for 5 min in 70% EtOH and the anthers from 10–15 flower buds were released into a glass vial (17 ml) containing approximately 3 ml of medium AT3 composed of KNO3 (13 mM), (NH4)2SO4 (8.6 mM), KH2PO4 (2.9 mM), CaCl2.2H2O (1.1 mM), MgSO4.7H2O (0.7 mM), MES (10 mM), glutamine (8.6 mM), maltose (300 mM), and Fe-EDTA, vitamins and microsalts according to Murashige and Skoog (Murashige and Skoog, 1962). The pH of the medium was adjusted to 6.5, unless stated otherwise, using 1 N KOH and the medium was filter-sterilized before use. The anthers were stirred with a magnetic bar for 2–3 min at a 500–750 rpm (Heidolph, MR3001, Germany) until the mixture of released microspores/anther debris and medium turned milky. The resulting suspension was filtered through a 40 µm nylon mesh in order to remove anther wall debris. The filtrate was washed three times in medium AT3 by centrifugation at 350 g for 2–3 min and resuspended in various media and used for maturation or transformation (see below for details). For in vitro maturation, 1.5 ml of the microspore/pollen suspension was cultured in 3 cm Petri dishes (Nunc, Denmark) in the dark at 25 °C for 7–9 d depending on the developmental stage of the initial microspore/pollen population. The in vitro-matured pollen grains were germinated in germination media or used for in situ pollination (see below for details).
Particle bombardment-mediated gene delivery into Antirrhinum gametophyte cells
A PDS 1000/He (Bio-Rad, USA) particle delivery system was used in all transient transformation experiments. Plasmid DNA was isolated using Qiagen Plasmid Mini Kit (Germany) columns, pretreated with RNase to get rid of RNA contamination, and dissolved in distilled water or TE buffer to give a final concentration of 1 µg µl-1. Precipitation of plasmid DNA onto gold particles and preparation of microcarriers were done essentially as described previously (Sanford et al., 1993) with slight modifications: the amount of gold particles per shot was decreased to 30 mg ml-1, sonication of gold particles before the coating reaction was performed to avoid clumps during the coating reaction, and the concentration of CaCl2 was decreased to 1 M.
Unicellular Antirrhinum majus microspores or pollen at different developmental stages were bombarded immediately after isolation in the respective culture media. The suspension (50 µl) containing 3x105 cells was dropped in the middle of a 3 cm wet Petri dish (Nunc, Denmark) to make a circle of evenly distributed cells on the bottom surface of the Petri dish without any support material. Bombardments were performed essentially as described earlier (Sanford et al., 1993). The bombarded microspores and developing pollen were cultured in the optimal maturation media for various time durations and analysed for the expression of the introduced genes.
In vitro germination of in vivo and in vitro-matured Antirrhinum majus pollen
Antirrhinum majus pollen collected from flowers at various stages of development and an aliquot of pollen at different time points during in vitro maturation were germinated in media with supplements. Standard media GV (stGV) (Touraev and Heberle-Bors, 1999) or PEG8000 (Read et al., 1993) were supplemented with different carbohydrates (sucrose, lactose, glucose, maltose) at pH 6.5 and filter-sterilized. In some experiments the diffusate from in vivo-matured pollen, flavonols or various oils were added to the germination medium. Approximately 1.5x105 pollen grains ml-1 of media were used for each in vitro germination assay.
In situ pollination with in vivo and in vitro-matured Antirrhinum pollen grains
The optimization of in situ pollination parameters were initially performed with in vivo pollen. Mature pollen grains isolated from open flowers were used dry or gently resuspended in culture medium with additional components and applied to Antirrhinum flowers emasculated by cutting off the corolla with a scalpel and removing the anthers with a forceps 1 d before pollination. In vitro-matured pollen grains were washed once in medium GV and then transferred onto stigmas of emasculated flowers in a droplet of 1 µl of GV medium at different concentrations, unless stated otherwise. Pollinated flowers were harvested after 10–48 h, stigma and styles were sectioned longitudinally and stained in aniline blue solution (0.1% aniline blue in 0.1 M K3PO4, pH=12.0) for 1–2 h to detect pollen tubes in the styles (Shivanna and Johri, 1985). Pollinated stigmas and stylar tissue were softened in some experiments in a solution of 8 N NaOH overnight and washed with fresh water before staining with aniline blue. Sections were mounted on microscopic slides in a mixture of glycerol and sterile water (1:1) and analysed under the Leitz-DIAPLAN fluorescence microscope equipped with FITC and UV filter sets.
Seed set was analysed 1.5–2.0 months after pollination by recording the number of fruits in relation to the number of pollinated flowers and by counting the number of seeds per pod in mature brown pods.
Detection of GFP and dsRFP expression
Expression of the GFP gene was visualized under a fluorescence microscope (Leitz-DIAPLAN) fitted with filter sets suitable for UV-light (Leitz A; excitation filter: BP340-380, dichroic mirror: RKP400, emission filter: LP430) and blue light (Leitz I2/3; excitation filter: BP450-490, dichroic mirror: RKP510, emission filter: LP515) excitation of GFP or an inverted microscope (Leitz-DIAVERT) with filter set H-3 (excitation filter: BP420-490, dichroic mirror: RKP510, emission filter: LP515). Expression of the dsRFP was detected using filter set N2 (excitation filter: BP530-560, dichroic mirror: RKP580, emission filter: LP580). Images of GFP and dsRFP positive microspores/pollen grains were taken using Kodak-Ektachrome film (ASA-400 or 1600).
ß-glucuronidase (GUS) assays
Histochemical assays for ß-glucuronidase (GUS) with transformed microspores and pollen were performed essentially as described earlier (Jefferson et al., 1987) with modifications (Stöger et al., 1992). Blue microspores and pollen grains were counted under an inverted microscope after 2 h of incubation of cells in X-Gluc solution at 37 °C. Under the conditions used, no endogenous GUS activity was observed in the control cultures with non-transformed pollen.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Antirrhinum majus pollen development in vivo
A good correlation was found between the size of flower bud and the stage of pollen development, which is essential to obtain sufficient numbers of microspores at defined developmental stages for in vitro culture (Fig. 1
). Flower buds of about 6 mm in size (from the base of the flower bud to the top of the emerging petals) contained early unicellular microspores (Fig. 1A, A1, A2). Early Antirrhinum majus microspores initially lack a vacuole and are 18 µm in diameter. The nucleus has a central position. Later, in flowers buds of approximately 7 mm in size, the vacuole occupies the whole volume of the microspore while the nucleus comes to lie at the periphery. Starch granules were apparent already at the late tetrad stage (data not shown) and were clearly present in early microspores (Fig. 1A2). The first, asymmetrical, haploid mitosis leads to the formation of a bicellular pollen grain with the vegetative and generative nuclei in flower buds with a size of 8 mm (Fig. 1B, B1, B2). The vacuole disappears soon after first pollen mitosis and the pollen grains are filled with cytoplasm and starch granules in flower buds with a size of 11 mm (Fig. 1C, C1, C2). A few days before anthesis the starch granules disappear and the mature snapdragon pollen grain is transparent when viewed under the light microscope (Fig. 1D, D1, D2). Mature snapdragon pollen is bicellular, with the generative cell suspended in the cytoplasm of the vegetative cell, has a size of 27–30 µm in diameter and a triangular shape (Fig. 1E1, E2). Approximately 100 000 pollen grains develop highly synchronously inside one anther of the snapdragon flower, which contains four anthers. The maturation of Antirrhinum pollen from early microspores lasted 8 d in plants grown at 18 °C.
|
In vitro maturation of Antirrhinum majus microspores
Media components were optimized to maintain the viability of isolated and cultured snapdragon microspores. Of the liquid basic media tested [MS (Murashige and Skoog, 1962), B5 (Gamborg et al., 1968), H (Nitsch and Nitsch, 1969), and AT3 (Touraev and Heberle-Bors, 1999)], microspore viability, determined by FDA staining, was highest in AT3 media after 8 d, reaching 50% (data not shown). Therefore the medium AT3 was chosen for further optimization experiments to evaluate the effect of various carbohydrates (sucrose, maltose, cellobiose, fructose, and mannitol) in the culture medium. The viability of microspores, the accumulation of starch, the formation of the vacuole, first haploid mitosis, and the germination of in vitro-matured pollen, were used as criteria for successful pollen development in vitro. The highest viability and optimal starch accumulation was observed in medium AT3 supplemented with maltose (Table 1). The germination of in vitro-matured pollen was observed in this medium only.
|
Furthermore, the effect of various concentrations of maltose at different pH values on the maturation of snapdragon microspores was tested (Fig. 2
). A rapid accumulation of starch and the enlargement of microspores led to death within 3 d when cultured in media with 0.4 M and 0.5 M maltose at pH 6.2 (Fig. 2). A slow accumulation of starch and further vacuolization of the microspores was characteristic for all media with a pH of 7.0. The highest frequency of maturation, optimal starch accumulation and germination of in vitro-developed Antirrhinum majus pollen was observed in medium with 0.4 M maltose at pH 6.5 (Fig. 2). The use of potassium-phosphate buffer or the absence of any buffer in the medium decreased the number of viable microspores. The highest percentage of viable microspores and in vitro-matured pollen (60%) was achieved in medium AT4 (0.4 M maltose and pH 6.5) with MES buffer. The development of microspores to fully mature pollen in vitro took 8 d, the same time as inside anthers in vivo (data not shown).
|
Light and fluorescent microscopical observation of pollen development in vitro under optimal conditions showed the disappearence of the vacuole, as in vivo (Fig. 3C
, E). The first pollen mitosis was asymmetric and bicellular pollen was formed at the second day of maturation, similar to the development inside the anther in vivo (Fig. 3B, D, F). A dramatic increase in starch accumulation was visible at the third day of culture (Fig. 3C). Bicellular pollen increased in size and formed the triangular shape typical of in vivo mature pollen at day 8 of in vitro maturation (Fig. 3E, F). Thus, cytological events during in vitro development of isolated Antirrhinum microspores closely resemble those of in vivo developing pollen.
|
In vitro germination of in vivo and in vitro-matured snapdragon pollen
Antirrhinum majus pollen did not readily germinate on the standard germination medium GV (Brewbaker and Kwack, 1963). The frequency of germination was usually below 20% for both in vitro and in vivo pollen (data not shown). Various media compositions and additions described in the literature (Read et al., 1993; Tupy and Rihova, 1984; Tupy et al., 1991) were therefore tested for the germination of Antirrhinum pollen matured in vivo and in vitro. Parameters which were varied included carbohydrates, pH of the media, organic substances such as oils, pollen diffusates, and flavonols (Lush et al., 1998; Touraev et al., 1997).
The presence and the type of carbohydrates in the medium was found to be essential for Antirrhinum pollen germination. Various concentrations of sucrose, glucose, maltose, lactose, and PEG8000 were tested and the highest frequency of germination of both in vivo and in vitro-matured pollen was observed in medium GV supplemented with 12.5% PEG8000 and 5% sucrose (Fig. 3G
), and slightly less with 10% sucrose alone (Table 2). Only thin or very short pollen tubes were observed when pollen grains were germinated in medium GV supplemented with lactose and glucose, respectively (data not shown). The frequency of germination of in vivo-matured pollen decreased with an increase in the pH of the media (Table 2). By contrast, the germination of in vitro-matured pollen was the same in all media containing PEG8000 with different pH values and lower in medium GV with pH values higher then 7.0.
|
The effect of a diffusate obtained from in vivo pollen and defined substances such as flavonols on the germination of Antirrhinum pollen was also analysed. The addition of pollen diffusate (50–100 µl) or the flavonol quercetin (0.1–100 µM) to the germination medium GV increased the germination of in vivo and in vitro-matured pollen 2-fold compared to the control GV medium with 10% sucrose (data not shown). However, concentrations of quercetin from 50–100 µM caused abnormal growth of pollen tubes. Pollen matured in vivo and in vitro could also germinate in GV-based media supplemented with several oils of plant origin including olive oil (Lush et al., 1998), but the frequency of germination was similar to or even significantly lower than the control medium with PEG or sucrose (data not shown).
It was found that Antirrhinum pollen is able to germinate in vitro when isolated from still closed flowers of 17 mm in size, but the maximum germination frequency is observed in pollen collected from fully open flowers of 30 mm in size (Fig. 1E). By contrast, the germination of in vitro-matured pollen was strongly dependent on the microspore stage from which the in vitro maturation was initiated. The highest germination frequency was observed in pollen matured 8 d starting from late microspores.
Pollen germination was also studied using in vitro-matured pollen applied to the snapdragon stigma. The growth of the pollen tube and its penetration through the style could be detected using aniline blue staining (Fig. 3H).
Transformation of Antirrhinum majus gametophytic cells
Antirrhinum majus microspores and pollen at different developmental stages were transformed with plasmids containing reporter genes (GUS, GFP and dsRFP) driven by different constitutive (CaMV 35S, UBI, ACTIN) or pollen-expressed (DC3, LAT52) promoters using particle bombardment. The majority of plasmids used for transformation were generated in house except the plasmid 35S
GFP, which was provided by Syngenta-MOGEN (Leiden, The Netherlands). Several parameters of bombardment, including coating reaction conditions such as particle type and amount, concentration and amount of DNA, precipitation time, and bombardment conditions such as gas pressure, particle travel distance, partial vacuum were optimized. The highest efficiency of transient expression of the transferred gene into Antirrhinum microspores and pollen was observed using 60 µg gold particles of 1.2 µm size, 10 min precipitation on ice, 5.5 cm target distance, helium pressure of 1000 psi, and 27 inch Hg vacuum. Efficiencies of transient expression detected in microspores were of 0.01–0.2% for the GFP and dsRFP, and of 5.0–7.0% for the GUS gene, in the best experiments (Table 3; Fig. 4). It should be noted, however, that the frequency of transient expression in microspores decreased up to 2-fold during in vitro maturation for 8 d. Significantly higher frequencies of transient expression were found when mature pollen was used as the recepient cell. Up to 30% of the pollen grains transiently expressed the GUS gene when transformed by the DC3GUS construct (Fig. 4A, B), and up to 0.5% when transformed by the DC3GFP5ER (Fig. 4C, D) and DC3dsRFP constructs (Fig. 4E, F). The transient transformation frequency was dependent on the promoters used to drive the reporter genes (Table 3). The highest frequency was observed when the GUS, GFP or dsRFP genes were expressed under the control of the DC3 and UBI promoters (Table 3).
|
|
Seed set after pollination with in vitro-matured Antirrhinum pollen
The final proof of the fertility of the in vitro-matured pollen is seed set after the pollination of normal in vivo flowers. To establish proper conditions for in situ pollination, the receptivity of Antirrhinum flowers was investigated. Flowers were emasculated 1 d before anthesis and hand pollination with cross pollen was performed at different time intervals using in vivo-matured, dry pollen collected from different plants. The number of seeds was counted after 1.5 months in mature, dry seed pods. The formation of fruits with seeds indicated that the Antirrhinum stigma is receptive for pollination during a 3 d period after flower opening (data not shown).
In vivo mature pollen of most species are dehydrated to various degrees (Heslop-Harrison, 1974). In a species with a wet stigma such as Antirrhinum majus, pollen hydrates in the stigmatic exudate when placed onto a stigma. Pollen matured in vitro is not dry since it is cultured in a liquid medium. To find out whether the lack of dehydration would affect germination, penetration and reproductive success, the ability of in vivo-matured pollen resuspended in 1 µl liquid GV or PEG8000 media at different numbers (
500, 1000, 2000, 3000) to germinate when placed onto the stigma and to set seeds was tested. Successful germination and seed set was observed in both media when more than 500 in vivo-matured pollen grains were used for pollination of one flower. Medium GV showed better seed set (Table 4).
|
For pollination with in vitro-matured pollen, Antirrhinum flowers were emasculated 1 d before anthesis and pipette pollinated with 1 µl droplets of suspensions containing different numbers (
500, 1000, 2000, 3000) of in vitro-matured pollen. The maturation of seed pods took 1.5 months, as with pollination by in vivo pollen. The minimal number of in vivo-matured pollen grains required for successful fruit set was found to be 1000 per receptive stigma. An increase in the number of pollen per flower led to an increase in seed set (Table 4). The addition of pollen diffusate, flavonols or various oils did not increase the frequency of seed set. Successful seed set was also obtained after pollination of receptive stigmas with transformed and in vitro-matured pollen. The number of seeds was lower compared to experiments with non-bombarded pollen, however (data not shown). No transgenic seeds have been obtained so far, although seeds were able to germinate readily in the control medium without selectable agents.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Detailed protocols are presented here for the successful maturation of isolated Antirrhinum majus microspores, transient transformation of pollen at different developmental stages, germination of in vivo and in vitro-matured pollen in vitro and onto the surface of receptive stigmas and successful seed set after pollination with in vitro-matured pollen (Fig. 5
).
|
Previous studies on Lilium (Clement et al., 1996; Tanaka and Ito, 1980), tulip (Tanaka and Ito, 1981), rapeseed (Custers et al., 1994), tobacco (Benito Moreno et al., 1988; Tupy et al., 1991), and wheat (Stauffer et al., 1991) have shown that microspores are developmentally independent from anther wall tissues soon after the release from the callose wall of tetrads. They develop cell-autonomously, requiring only nutrients from the sporophyte (Worrall et al., 1992) The nutrients can easily be replaced by culturing microspores in a defined medium, taking into consideration the changing nutritional requirements during development (Touraev and Heberle-Bors, 1999). The results obtained with in vitro-cultured-snapdragon microspores confirm these observations.
Several factors were found to play an important role during the maturation of Antirrhinum microspores, including carbohydrates, buffers, and pH of the media. One of the most important components of the in vitro culture media for maturation was the carbohydrate source (Touraev and Heberle-Bors, 1999). Sucrose is the most commonly used carbohydrate for in vitro culture of microspores. However, microspores of some species such as barley (Scott and Lyne, 1994) and snapdragon (present work) were not able to survive on sucrose-containing media, and died soon after isolation and culture. The optimal medium for successful maturation of late Antirrhinum microspores consisted of 0.4 M maltose, MES buffer, and a pH value of 6.5. Approximately 60% of the microspores matured when cultured in this medium and more than 40% of the matured pollen germinated when transferred to the best germination media or onto the stigma. The stages of maturation closely resembled those of in vivo development and lasted 8 d in both cases.
All stages, starting from microspores until fully mature pollen, were suitable targets for particle-mediated gene delivery using optimized in vitro culture conditions and bombardment parameters. The frequency of pollen grains transiently expressing the marker genes was, however, significantly higher with more advanced stages of pollen development. DNA on particles that entered the large vacuole, even when they did not destroy the microspore, was prevented from entering the nucleus. Apart from the larger effective target volume for gold particles, the increase in cytoplasm, ribosome number and general transcription activity in more advanced pollen grains probably increase the expression from any given transgene, which results in a higher transformation frequency. The majority of particles coated with DNA were indeed identified in the vacuole of microspores and no expression was observed in these cells. Using the conditions described here, approximately 0.02–7.6% of snapdragon microspores can be transformed by using particle bombardment depending on the construct used.
Unicellular microspores were transformed with foreign DNA and matured for 8 d in vitro, and a population of mature pollen was used for in situ pollination. In snapdragon the transformed microspores were able to continue their development in vitro until the formation of mature transgenic pollen. This is proof that particle bombardment of microspores and the introduction of foreign DNA into the cell does not change the normal development of transformed cells. Thus, in vitro maturation of microspores is a meaningful experimental system to study gene expression during pollen development.
Several constructs containing three modifications of GFP (GFP4, GFP4S65C and GFP5ER) and one dsRFP under the control of different promoters have been bombarded into developing Antirrhinum microspores. In all cases the frequency of the GFP or dsRFP-expressing pollen was lower than GUS-expressing pollen driven by the same promoter, and reached in the best cases only 0.2% of the total population of cells. Therefore, a very strong promoter should be used to drive the GFP or dsRFP genes. Initial attempts to sort GFP or dsRFP positive pollen grains by using the Partec (Germany) or FACS sorters failed (data not shown). Antirrhinum transformation will be dramatically improved when cell sorting can be used to select transgenic pollen. Manual picking of GFP or dsRFP positive pollen grains is possibly a more realistic approach at present and can be used to collect GFP-positive snapdragon pollen (I Barinova, E Heberle-Bors, A Touraev, unpublished observations).
In general, pollen grains from dicot species germinate more easily in vitro compared to monocots, such as cereals (Heslop-Harrison, 1979; Stauffer et al., 1991). Data from the literature and our own results clearly show the limiting effect of carbohydrate and osmotic sources on pollen germination (Balatkova and Tupy, 1973; Tupy et al., 1991; Nygaard, 1977). A medium supplemented with PEG8000 and sucrose has previously been reported to enhance germination of tobacco (Read et al., 1993), petunia (Zhang and Croes, 1982), and Brassica pollen (Shivanna and Sawhney, 1995). The present study shows that a PEG8000 medium is the best for the germination of both in vivo and in vitro-matured Antirrhinum pollen. However, PEG8000-based media slightly prevented the penetration of pollen tubes into the stigma and led to the lower efficiency of seed set. In situ pollination experiments, however, were successful using medium GV supplemented with sucrose. Among many other additives tested (pollen diffusate, stigmatic exudate, flavonols, oils, etc.) pollen diffusate and flavonols significantly enhanced in vitro germination which is consistent with data obtained earlier in tobacco and petunia (Read et al., 1993; Zhang and Croes, 1982). The optimal medium for in vitro germination of in vivo or in vitro-matured snapdragon pollen was thus designed based on medium PEG8000 (Read et al., 1993) and GV (Touraev and Heberle-Bors, 1999) with an addition of pollen diffusate or the flavonols quercetin or kaemferol (Ylstra et al., 1992). This medium is also best suited for transient expression studies of genes transferred into microspores.
Successful seed set after pollination with in vitro mature pollen depends on the reproductive biology of a given species, i.e. pollen–stigma interactions. All plant species can be divided into two groups depending on their stigmatic surface: dry and wet stigmas (Heslop-Harrison and Shivanna, 1977). Antirrhinum belongs to the group of plants with wet stigmas. Pollen grains from this group of species are highly dehydrated and the rehydration process takes a relatively long time on the surface of a stigma with participation of stigmatic exudate released in response to pollen loading (Heslop-Harrison, 1974). Usually an excess of pollen grains has to be applied per stigma in order to get full seed set in these species.
It was found that the optimal number of in vitro-matured pollen grains per stigma required for successful fruit set was 2000, which is 2–3 times higher than in vivo-matured, dry pollen which had been resuspended in liquid medium for in situ pollination. One possible reason could be that pollen matured in vitro independently from the surrounding anther tissue lacks compounds that are released from the tapetum onto the surface of the pollen, such as flavonols, tryphine, lipidic or oil compounds, and others which are necessary for germination and successful pollination (Taylor and Jorgensen, 1992; Mo et al., 1992; Lush et al., 1998). Another reason could be the distortion of pollen–stigma interaction and thus pollen tube directionality when in vitro pollen is applied onto the surface of the stigma in an aqueous solution. The addition of pollen diffusate or other pollen wall compounds and the application of artificially dried in vitro pollen onto stigmas could increase the efficiency of seed set after pollination with in vitro-matured pollen.
|
Acknowledgements |
|---|
The authors would like to thank Dr Z Schwarz-Sommer and Dr S Zachgo for continuous advice and support. Authors would like to thank Dr J Haseloff (MRC, Cambridge, UK) and Dr Ch Maas (Max-Planck Institute für Züchtungsforschung, Köln, Germany) for kindly providing the mGFP4, mGFP5ER and mGFP4S65C harbouring plasmids, respectively. Construct containing 35S
GFP was provided by Syngenta-MOGEN (Leiden, The Netherlands). The promoters used in these studies were a kind gift from Professor T Thomas (Texas A&M University, Texas, USA), Dr D Twell (Leicester University, Leicester, UK), Professor R Wu (Cornell Univeristy, Ithaca, USA) and Dr P Quail (Plant Gene Expression Center, USDA-ARS, Albany, New York, USA). We also thank Dr C Wilson and Dr S Zachgo for a critical reading of the manuscript and helpful discussions. This work was supported by the Austrian Ministry of Science and Traffic and the Ministry of Health and Consumer Protection. |
Footnotes |
|---|
3 To whom correspondence should be addressed. Fax: +43 1 4277 9546. E-mail: Alisher{at}gem.univie.ac.at
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Balatkova V, Tupy J. 1973. The significance of the stigmatal and placental pollination in vitro in Antirrhinum majus L.: seed and callus formation on placentae. Biologia Plantarum (Praha) 15, 102–106.
Bedinger P. 1992. The remarkable biology of pollen. The Plant Cell 4, 879–887.
Benito Moreno RM, Macke F, Hauser MT, Alwen A, Heberle-Bors E. 1988. Sporophytes and male gametophytes from in vitro-cultured, immature tobacco pollen. In: Cresti M, Jori P, Pacini E, eds. Sexual reproduction in higher plants. Berlin: Springer, 137–142.
Brewbaker JL, Kwack BH. 1963. The essential role of calcium ion in pollen germination and pollen tube growth. American Journal of Botany 50, 859–865.[Web of Science]
Carpenter R, Coen ES. 1990. Floral homeotic mutations produced by transposon-mutagenesis in Antirrhinum majus. Genes and Development 4, 1483–1493.
Clement C, Al-Award D, Audran J-C. 1996. In vivo and in vitro pollen maturation in Lilium: influence of carbohydrates. Acta Societatis Botanicorum Poloniae 65, 73–82.
Corriveau JL, Goff LJ, Coleman AW. 1990. Plastid DNA is not detectable in the male gametes and pollen tubes of an angiosperm (Antirrhinum majus) that is maternal for plastid inheritance. Current Genetics 17, 439–444.[Web of Science]
Cui ML, Takayanagi K, Kamada H, Nishimura S, Handa T. 2001. Efficient shoot regeneration from hairy roots of Antirrhinum majus L. transformed by the rol type MAT vector system. Plant Cell Reports 20, 55–59.
Custers JBM, Gordewener JHG, Nöllen Y, Dons JJM, Van Lookeren-Campagne MM. 1994. Temperature controls both gametophytic and sporophytic development in microspore cultures of Brassica napus. Plant Cell Reports 13, 267–271.[Web of Science]
Davies B, Motte P, Keck E, Saedler H, Sommer H, Schwarz-Sommer Z. 1999. PLENA and FARINELLI: redundancy and regulatory interactions between two Antirrhinum MADS-box factors controlling flower development. EMBO Journal 18, 4023–4034.[Web of Science][Medline]
Dunwell JM. 1996. Microspore cultures. In: Summerfield RJ, ed. In vitro haploid production in higher plants. Dordrecht, Boston, London: Kluwer Academic Publishers 1, 205–216.
Gamborg OL, Miller RA, Ojima K. 1968. Nutrient requirements of suspension cultures of soybean root cells. Experimental Cell Research 50, 151–158.[Web of Science][Medline]
Heberle-Bors E, Stöger E, Touraev A, Zarsky V, Vicente O. 1996. In vitro pollen cultures: progress and perspectives. In: Mohapatra SS, Knox RB, eds. Pollen biotechnology. Gene expression and allergen characterization. New York: Chapman and Hall, 85–100.
Heidmann I, Efremova N, Saedler H, Schrwarz-Sommer Z. 1998. A protocol for transformation and regeneration of Antirrhinum majus. The Plant Journal 13, 723–728.
Heslop-Harrison J. 1974. Pollen–stigma interaction in grasses: a brief review. New Zealand Journal of Botany 17, 537–546.
Heslop-Harrison J. 1979. Aspects of the structure, cytochemistry and germination of rye (Secale cereale). Annals of Botany Supplement 1 44, 1–47.
Heslop-Harrison Y, Shivanna KR. 1977. The receptive surface of the angiosperm stigma. Annals of Botany 41, 1233–1258.
Jefferson RA, Kavanagh TA, Bevan MW. 1987. GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO Journal 6, 3901–3907.[Web of Science][Medline]
Lombardo G, Carraro L. 1976. Tapetal ultrastructural changes during pollen development. 1. Studies on Antirrhinum majus. Caryologia 29, 113–125.
Luo D, Carpenter R, Vincent C, Copsey L, Coen E. 1995. Origin of floral asymmetry in Antirrhinum. Nature 383, 794–799.
Lush WM, Grieser F, Wolters-Arts M. 1998. Directional guidance of Nicotiana alata pollen tubes in vitro and on the stigma. Plant Physiology 118, 733–741.
Mo Y, Nagel C, Taylor LP. 1992. Biochemical complementation of chalcone synthase mutants defines a role for flavonols in functional pollen. Proceedings of the National Academy of Sciences, USA 89, 7213–7217.
Murashige T, Skoog F. 1962. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiologia Plantarum 15, 473–497.
Nitsch SP, Nitsch C. 1969. Haploid plants from pollen. Science 163, 85–87.
Nygaard P. 1977. Utilization of exogenous carbohydrates for tube growth and starch synthesis in pine pollen suspension cultures. Physiologia Plantarum 39, 206–210.
Ottenschlager I, Barinova I, Voronin V, Dahl M, Heberle-Bors E, Touraev A. 1999. Green fluorescent protein (GFP) as a marker during pollen development. Transgenic Research 8, 279–294.[Web of Science][Medline]
Read SM, Clarke AE, Bacic A. 1993. Stimulation of growth of cultured Nicotiana tabacum W 38 pollen tubes by polyethylene glycole and Cu (II) salts. Protoplasma 177, 1–14.
Sanford JC, Smith FD, Russell JA. 1993. Optimizing the biolistic process for different biological applications. Methods of Enzymology 217, 483–509.
Sangwan RS, Sangwan-Norreel BS. 1987. Ultrastructural cytology of plastids in pollen grains of certain androgenic and nonandrogenic plants. Protoplasma 138, 11–22.
Scott P, Lyne RL. 1994. Initiation of embryiogenesis from cultured barley microspores: a further investigation into the toxic effects of sucrose and glucose. Plant Cell Tissue and Organ Culture 37, 61–65.
Shivanna KR, Johri BM. 1985. The angiosperm pollen. Structure and function. New Delhi, Bangalore, Bombay, Calcutta, Madras, Hyderabad: Willey Eastern Limited.
Shivanna KR, Sawhney VK. 1995. Polyethilene glycol improves the in vitro growth of Brassica pollen tubes without loss in germination. Journal of Experimental Botany 46, 1771–1774.
Stauffer C, Benito Moreno RM, Heberle-Bors E. 1991. Seed set after pollination with in-vitro-matured, isolated pollen of Triticum aestivum. Theoretical and Applied Genetics 81, 576–580.
Stöger E, Benito Moreno R-M, Ylstra B, Vicente O, Heberle-Bors E. 1992. Comparison of different techniques for gene transfer into mature and immature tobacco pollen. Transgenic Research 1, 71–78.
Tanaka I, Ito M. 1980. Induction of typical cell division in isolated microspores of Lilium longiflorum and Tulipa gesneriana. Plant Science Letters 17, 279–285.
Tanaka I, Ito M. 1981. Control of division patterns in explanted microspores of Tulipa gesneriana. Protoplasma 108, 329–340.
Taylor LP, Jorgensen R. 1992. Conditional male fertility in chalcone synthase-deficient petunia. Journal of Heredity 83, 11–17.
Touraev A, Fink CS, Stöger E, Heberle-Bors E. 1995. Pollen selection: a transgenic reconstruction approach. Proceedings of the National Academy of Sciences, USA 92, 12165–12169.
Touraev A, Heberle-Bors E. 1999. Microspore embryogenesis and in vitro pollen maturation in tobacco. In: Hall R, ed. Plant cell culture protocols 111, Totowa, New Jersey: Humana Press, 281–291.
Touraev A, Indrianto A, Wratschko I, Vicente O, Heberle-Bors E. 1996. Efficient microspore embryogenesis in wheat (Triticum aestivum L.) induced by starvation at high temperatures. Sexual Plant Reproduction 9, 209–215.
Touraev A, Stöger E, Voronin V, Heberle-Bors E. 1997. Plant male germ line transformation. The Plant Journal 12, 949–956.
Tupy J, Rihova L. 1984. Changes and growth effect of pH in pollen tube culture. Journal of Plant Physiology 115, 1–10.
Tupy J, Rihova L, Zarsky V. 1991. Production of fertile tobacco pollen from microspores in suspension culture and its storage for in situ pollination. Sexual Plant Reproduction 4, 284–287.
Worrall D, Hird DL, Hodge R, Paul W, Draper J, Scott R. 1992. Premature dissolution of the microsporocyte callose wall causes male sterility in transgenic tobacco. The Plant Cell 4, 759–771.
Ylstra B, Touraev A, Benito Moreno RM, Stöger E, van Tunen AJ, Vicente O, Mol JNM, Heberle-Bors E. 1992. Flavonols stimulate development, germination and tube growth of tobacco pollen. Plant Physiology 100, 902–907.
Zhang HQ, Croes AF. 1982. A new medium for pollen germination in vitro. Acta Botanica Neerlandica 31, 113–119.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||