کارایی بالای کندم دابل هاپلوئئید در تولید
arelationship between microspore embryogenesis and chemical treatment was observed in our experiments and by others (Konzak et al., 2000; Bennett and Hughes, 1972; Rowell and Miller, 1971; Picard et al., 1987). Although Picard et al. (1987) described improvements in androgenesis with wheat (T. aestivum) anther cultures, their treatments were less effective than those in use for anther culture at that time (Zhou and Konzak, 1989). We envisioned that according to the signal system concept of Ryan and Balls (1962) and Constabel et al. (1995) some chemical formulations could effectively induce a large proportion of microspores to become embryogenic, if the correct formulations were developed. We recognized, however, that after embryogenesis was induced, the induced microspores require an optimal physiological environment to develop further into embryoids able to germinate and develop into green plants.
The objectives of this work were to develop a method for efficiently initiating microspore embryogenesis by a chemical inducer formulation, and for producing large quantities of micros pore-derived green plants from a wide spectrum of genotypes under optimal culture conditions.
MATERIALS AND METHODS Growing Wheat Plants
The spring wheat genotypes 'Chris', 'Pavon 76', WED 20216-2,'Yecora Rojo', 'Calorwa', 'Waldron', 'Wawawai', and winter wheats 'Capo' and 'Svilena', were used. These genotypes differed in responsiveness to embryogenesis and green plant regeneration on the basis of previous anther culture experiments. One to three plants per pot (20 by 25 ern in diameter) were grown in a greenhouse controlled at 27 :+: 2°C, at a light regime of 17 h light and 7 h dark. Winter wheat
Abbreviations: 2,4-D, 2,4-dichlorophenoxyacetic acid; 2-HNA, 2-hydroxynicotinic acid; BAP, 6-benzylaminopurine; DH, doubled haploid; NPB, Northwest Plant Breeding Company.
LIU ET AL.: DOUBLED-HAPLOID PRODUCTION VIA MICROSPORE EMBRYOGENESIS
seedlings were artificially vernalized at 6°C for 2 to 2.5 mo before growing in the greenhouse. Fertilizers (N, P, K) were premixed with soil at the time of sowing, and additional fertilizer was applied by daily watering, which contained liquid forms of nitrogen (N), phosphorus (P) and potassium (K).
Fresh tillers containing microspores at an appropriate developmental stage were cut from below two nodes, counting from the top of the tiller, and immediately placed in a clean container with distilled water. All leaves were removed by cutting at their bases. The time between the collection of tillers and their treatment was minimized to reduce the possibility of contamination by microorganisms. Microspores enclosed within the anthers in the middle section of a spike preferably were collected at the mid to late-uninucleate stage of development. Morphological features of tillers containing microspores at these stages were established for each cultivar via microscopic examination of microspores in an anther sample with acetocarmine stain.
Pretreatment of Spikes
After removing the lower nodal section, the collected tillers were placed in an autoclaved sterile flask, containing 50 mL of sterile (autoclave d) distilled water (the control treatment) or 50 mL of the desired amount of sterile (autoclave d) inducer formulations [0-1 g L -[2-HNA (Sigma-Aldrich Co., St. Louis, MO), with or without 10-6 mol L -[ 2,4-D and 10-6 mol L-[ BAP]. The open end of a plastic bag (thin-walled, grocery store bag) was placed over the spikes, wrapped around the neck of the flask, and sealed around the flask with masking tape to limit microbial contamination and excessive loss of water. The flask was placed in an incubator at 33°C for a desired period of time, ranging between about 48 h and about 72 h with different genotypes, until microspores enclosed within the anthers from the center section of a spike showed a characteristic embryogenic structure, i.e., the fibrillar-appearing cytoplasm of induced microspores (Fig. la).
After the tillers were pretreated, they were removed from the treatment flask in a laminar flow hood. All foliage beneath the first tiller node was removed, keeping only the boot containing the spike. Isolated boots were then placed on a paper towel and sprayed with 750 g L -[ ethanol to saturation. The boots were wrapped in the towel and placed in the hood for approximately 45 min, or until the ethanol fully evaporated. The spikes were aseptically removed from each disinfected boot and placed on top of a 125-mL kitchen blender cup. Awns (if present), and the upper spikelets were removed with sterile forceps and scissors. Florets were cut from their bases and allowed to drop into the open blender cup. Florets obtained from one to three spikes were used for each run of the blending process. Forty milliliters of a 0.3 mol L -[ mannitol solution (autoclave d) was added to the blender cup, and a sterilized cap was placed on the blender cup, which was assembled on the blender. The florets were blended for 20 s at 2200 rpm to release most microspores. The blended slurry was poured from the blender cup into a sterile filter (a container with 100-fLm stainless steel mesh at the bottom), and the blender top was rinsed twice with 5 mL of a 0.3 mol L-[ mannitol solution per rinse, and the mannitol solution was poured into the filter. Residue trapped on top of the filter was discarded, and the filtrate was pipetted into 15-mL sterile centrifuge tubes and centrifuged at 100 X g for 3 min. The supernatant was discarded from the tubes, and the pellets were combined and resuspended in 2 mL of 0.3 mol L -[ mannitol solution. The resuspended pellets were layered over 5 mL of a 0.58 mol L -[ maltose solution (sterile) and centrifuged at 100 X g for 3 min. Three milliliters of the upper band (containing microspores) was collected and resuspended in 10 mL of a 0.3 mol L -[ mannitol solution in a 15-mL centrifuge tube. The lower band (pellet) was resuspended (for counting purposes) in 12 mL water in a separate 15-mL centrifuge tube. Both centrifuge tubes were centrifuged at 100 X g for 3 min. The supernatant was discarded and the pellet resuspended in 3-mL culture medium for upper band microspores, or 3 mL water for lower band microspores. The number of microspores
Fig. I. Process of isolated microspore liquid culture for doubled-haploid production, genotype Chris. (a) Mid- to late-uninucleate microspores from the freshly harvested spikes. (b) Embryogeuic microspores with fibrillar cytoplasm induced by 2-HNA treatment at high temperature for 65 h. (c) Microspore derived developing embryoids cultured in liquid media for 18 d. (d). Continuous production of mature embryoids upon liquid culture from 21 d on. (e) Germination of embryoids on solid media in Petri dish at 10 d. (f) Thousands of doubled-haploid plants derived from microspores of a single spike.
in each band was counted with a hemocytometer, and after counting the lower band microspores were discarded. The total of microspores isolated was the sum of the microspores from both the upper band and the lower band. Only the microspores from the upper band were used for culture. The lower band microspores appeared to be those that were too young, or too old and containing starch; thus, they had not developed sufficiently or had developed beyond the stage of development useful for DH production. The upper band microspores were resuspended in 10 mL of culture medium in a 15-mL centrifuge tube and centrifuged at 100 X g for 3 min. The supernatant was discarded and the pellet resuspended in culture medium at a concentration of approximately 1 X 104 microspores mL -I.
Culture of Isolated Microspores
Isolated microspores were cultured as a suspension in liquid NPB 99 medium, which contains 232 mg L -[ (NH4)2S04, 1415 mg L -[ KN03, 83 mg L -[ CaCb-2H20, 200 mg L -[ KH2P04, 93 mg L -[ MgS04-7H20, 5 mg L -[ H3B03, 0.0125 mg L-[ CoCI-6H20, 0.0125 mg L -[ CuS04-5H20, 0.4 mg L -[ KI, 5 mg L -[ MnS04-4H20, 0.0125 mg L -[ Na2Mo04-2H20, 5 mg L -[ ZnS04-7H20, 37.3 mg L -[ Na2EDTA, 27.8 mg L -[ FeS047H20, 50 mg L -[ myo-inositol, 0.5 mg L -[ nicotinic acid, 0.5 mg L -[ pyridoxine-HCI, 5 mg L -[ thiamine-H'Cl, 500 mg L-[ glutamine, 0.2 mg L -[ 2,4-D, 0.2 mg L -[ kinetin, 1 mg L-[ phenylacetic acid (P AA), and 90 g L -[ maltose, adjusted to pH7 and filter sterilized. An aliquot of 2 mL media per 35 by 10 mm Petri dish, or 5 mL media per 60 by 15 mm Petri dish, at a density of approximately 1 X 104 microspores mL -[ was used. Immature ovaries were added to the culture at a density of one per milliliter, immediately preceding the incubation. Ovaries were aseptically dissected from fresh and disinfected spikes. The ovaries from the cultivar Chris were used as convenient sources for supporting embryogenesis of the nine wheat lines tested. The Petri dish was sealed with Parafilm (American Can Co., Greenwich, CT) and incubated in the dark at 27°e.
After embryoids had grown to 1 to 2 mm in diameter, they were transferred aseptically to solid 190-2 medium (Zhuang and Xu, 1983) at a density of 25 to 30 embryo ids in each 100- by 15-mm Petri dish for germinating into plants. The embryo ids were incubated under continuous fluorescent light at room temperature (22°C). In approximately 2 wk, green plants developed and subsequently transferred to soil and grown to maturity in the greenhouse. To avoid bias, the first available 200 embryoids from each Petri dish were transferred for evaluating plant germination rate and DH percentage. Green and albino plants with well-developed roots and shoots were counted at 14 dafter embryoids were transferred to germination culture media. Plant fertility was evaluated on
the basis of seed set. More than 20 plants per replication were evaluated for seed fertility.
All experiments were analyzed as completely randomized designs. There were two to six replications for each treatment. For the 2-HNA dose experiment, similar Chris tillers were assigned to each flask, and each treatment was randomly applied twice to the flasks. Microspores from each of the two flasks with the same treatment were separately isolated and cultured in the same Petri dish (replication), and each Petri dish was separately evaluated. For the experiments on osmolality or ovary source, microspores from six Pavon 76 or six Chris spikes were first isolated, and equally distributed to each Petri dish, and each treatment was randomly applied twice to the Petri dishes. Each of the two Petri dishes with the same treatment was considered as a replication and was evaluated separately. NPB99 media with different osmotic pressures were made by adjusting concentrations of maltose and mannitol. For the genotypic response experiment, the same pretreatment regime with 50 mL of the inducer formulation (0.1 g L -[ 2-HNA, 10-6 mol L -[ 2,4-D and 10-6 mol L -[ BAP) was applied to eight genotypes, and data were pooled means of two to six replications per genotype. General linear model (Lentner and Bishop, 1993) was used to analyze the data. Analysis of variance was conducted, followed by a 5% least square difference analysis for the three properties, i.e., number of embryoids, green plant percentage, and DH percentage.
RESULTS AND DISCUSSION Embryogenesis Triggered by a 2-Hydroxynicotinic Acid Formulation
Over 50% of the total microspores in a spike routinely can be induced to become embryogenic by treatment with a formulation, including the chemical 2-RNA at 33°C, leading to the development of thousands of green plants originating from the microspores of a single wheat spike (Table 1; Fig. 1). After the treatment, the embryogenic microspores typically have eight or more small vacuoles immediately enclosed by the cell wall (Fig. 1a and b). These vacuoles surround the condensed cytoplasm in the center, forming a fibrillar structure. The embryogenic microspores are usually, but not always, of a larger size (about 50 urn) than the average nontreated or noninduced microspores (25-45 urn).
The optimal concentration of 2-RNA in the formula-
LIU ET AL.: DOUBLED-HAPLOID PRODUCTION VIA MICROSPORE EMBRYOGENESIS
tion for treating micros pores to induce embryogenesis and form mature embryo ids was determined to be approximately 100 mg L -1 (Tab~e 2). The .number of induced embryo ids increased with mcreasmg co~\cent~ations of 2-RNA up to a threshold of 100 mg L , while the percentage of germinated green plants (expressed as a percentage of the number of e~b.r~oids tra~sferred to germination medium) did not significantly differ between different concentrations of 2-RNA. Spontaneous chromosome doubling percentage reached 65% with 2-RNA treatment at 100 mg L -1. A toxic level of2-RNA in the pretreatment formulation w~s observed. at the dose 1 g L -\ when the tiller stem tissues detenorated and microspores died. Microspores, when isolated fr?m tillers pretreated without chemical inducer formulation (in distilled R20), appeared to develop .to,:"ard ~ollen maturation, and died when cultured m induction medium.
The chemical, 2-RNA, was previously identified to increase the efficiency of androgenesis in anther cultures when applied to wheat spikes at a critical developmental stage (Konzak et al., 2000). It can be effectively and conveniently delivered to act on microspores by ~he described method. The chemical inducer formulation is absorbed by the vascular system of the stem and transported to the anthers and into the microspores, and the 33°C temperature speeds up the efficiency of chemical delivery to microspores. Lower temperatures may be employed, but adjustments must ?e made for the slower rates of chemical uptake and tiller growth. The optimal period of pretreatment appears to vary somewhat with the genotype and the treatment temperature, ranging between about 48 h and about 72 h at 33°C (Konzak et al., 2000). Tillers can be stored for convenience in a refrigerator at 4°C for up to 1 mo before subjecting the microspores to pretreatment with the chemical inducer formulation and temperature, and with nutrient stress. Because the microspore viability falls sharply, tillers should not be stored in a refrigerator at 4°C after the temperature-nutrient stress treatment. The influence of microspore developmental stages on androgenesis was very strong. The mid t? late uninu~leate microspores were the most responSIve to chemical induction of androgenesis. Since micros pores in wheat spikes are not synchronized in their developmental stage, one can only expect a portion of the microspores
from a given spike to be inducible to embryogenesis. The task is to synchronize the maximum number of microspores in a spike at the appropriate de:~lopm~ntal stage. Success can be determined by. staining mI~rospores of an anther in the middle section of the spike.
Isolation Methods and Isolated Microspore Purification
Once microspores are embryogenic, it is necessary to separate them from spikes and c~lture the~ in liquid nutrient media. Liquid culture of Isolated microsp~res provides many advantages over anther cultur.e .. FIrst, the entire process of microspore embryogenesis m the culture plate can be easily monitored under ~n inverted microscope and the process of embryogenesis followed over time, as desired. Microscopic examination provides an effective way to observe development. Sec?nd, all embryoids formed in the culture plate are certain t.o be micros pore derived, and plants regenerated are .eIther haploid or DR, because the only cells placed m the
culture plates are microspores (Fig 1 a and b). .
The isolation process should minimize damage to rmcrospores. Different microspore isolati?n.methods were tested, including use of a vortexer, stirring bars, glass bar grinding, and blending. While all methods seemed to work (Konzak et al., 2000), only isolation by use of a blender gave us repeatable results, and hi~h yields of viable and responsive microspores, especially when fixed mechanical conditions such as blend speed and time were monitored. Use of a blender is a rapid and efficient means for processing large numbers of samples. Despite these advantages, we have observed that using a blender damages over 50% of the embryogenic microspores, resulting in early abortion of developme~t toward mature embryo ids. Future research should arm to improve the isolation methods to reduce damage to the induced embryogenic microspores. Nevertheless, since as many as 50% of the micros pores in the spikes can be induced to be embryogenic, the current procedure already can produce large numbers of micr~sporederived green plants. In fact, so many embryoids are usually produced that the transfer o~ t~~ embryoids to germination media can be the factor limiting the number
of plants recovered. ..
Purifying embryogenic microspores IS another Important step for which results are repeat.able. T~e dead, nonembryogenic microspores or de~ns may m.terfere with developing embryo ids by releasing phenohcs and by changing media composition, such as pR and osmolality. Several purification methods w~re found. to wo.rk, but the combination of a simple gradient centrifugation by 0.58 mol L -1 maltose and mesh-filter filtration proved to be most effective and efficient.
Effect of Osmolality in Culture Media
With a large population of embryogenic mi~rospores isolated, the task is to provide a favorable envIro~ment to enable them to develop into mature embryoids. In the described method, embryogenic microspores began their first cell division after approximately 12 h in cul-
Table 3. Optimization of osmotic pressure in the induction media for androgenesis.
Osmolity, mOsmol kg-1"2°:1: