Plant regeneration from protoplasts of alfalfa 683 PLANT REGENERATION FROM PROTOPLASTS OF

Alfalfa is one of the most frequently studied species from the production of tissue culture-derived embryos point of view. In this study, five alfalfa cultivars were analyzed with reference to their ability to regenerate plants from protoplast cultures via somatic embryogenesis. Plant regeneration from leaf-derived protoplasts isolated from the cultivar Rangelander was achieved using a protocol defined for alfalfa cell suspension-derived embryogenesis. Because of its high efficiency, this procedure is recommended for protoplast electroporation-mediated genetic transformation of alfalfa.


INTRODUCTION
Alfalfa is a warm season perennial legume planted worldwide.Originated in the Middle-East near Iran, is now found in nearly all continents.It represents one of the oldest forage crops.Depending upon the cultivar it withstands a wide variety of climates and is highly drought resistant through dormancy for as much as two years.Having high minerals contents, vitamins and protein, alfalfa is one of the most nutritious crops that can be utilized as forage.Alfalfa is also used for silage, hay, green chop, erosion control, honey production, rotational grazing, and in some forms, for human consumption.It is a high-quality forage plant for all livestock types.Alfalfa is also a good nitrogen soil fixer, by symbiosis with bacteria of the genum Rhizobium (www.farmseeds.com/forages/alfalfa).
In Brazil, alfalfa cropping began in the South (Coelho Jr., 1994), but now it is possible to grow alfalfa in the Southeastern and Northeastern regions of the country.'Crioula' is the most accepted variety, as a result of its adaptability and productivity (28 t DM per ha per year).It is used feed for horses, cattle, dairy cattle, sheep, and also ostriches (Haddad & Castro, 1999).
Alfalfa is also one of the most frequently studied crops from the point of view of tissue culture-derived embryo production.It is also considered the species with the most advanced synthetic seed system, although its use for commercial propagation purposes is still under evaluation (Piccioni et al., 1997).Moreover, significant progress has been made using genetic engineering methods for improving its nutritional value (Schroeder et al., 1991), to enhance tolerance to abiotic stress (McKersei et al., 1993), and to use alfalfa as a source of value-added products.
Scientia Agricola, v.60, n.4, p.683-689, Oct./Dec.2003 Technologies are available to introduce genes into alfalfa via genetic transformation, but commercially important cultivars cannot yet be engineered.Regenerability capacity is the primary constraint to successfully transform elite genotypes.Preliminary studies indicate that alfalfa in vitro response is strongly genotype dependent (Bingham et al., 1975;Atanassov & Brown, 1984).This feature was also reported by Myers et al. (1989) working on Trifolium pratense L. Also, as a rule, growth regulators have a critical importance in inducing somatic embryogenesis in legumes.Atanassov & Brown (1984) reported the induction of callus from cotyledon, hypocotyl and leaf-derived alfalfa explants on B 5 h medium, which consists of B 5 medium (Gamborg et al., 1968) modified to contain 1 mg L -1 2,4-D (2,4-dichlorophenoxyacetic acid); 0.2 mg L -1 kinetin; 30,000 mg L -1 sucrose; 3,000 mg L -1 KNO 3 ; 895 mg L -1 CaCl 2 ; 0,800 mg L -1 L-glutamine; 500 mg L -1 MgSO 4 •7 H 2 O; 100 mg L -1 serine; 10 mg L -1 L-glutathione; 1.0 mg L -1 adenine and 9 g L -1 Bacto agar.Primary calli, three weeks after subculturing, were used to establish cell suspension cultures that were maintained by subculturing 2-5 mL of the topmost layer of settled cell clusters into 45 mL of fresh B 5 h medium every seven days.Under these conditions, cells had a characteristic growth curve, and various stages of differentiation were observed in the newly established cell suspensions.All of the 13 tested varieties produced embryos on the B 5 h → SHb → BOi2Y media sequence: a treatment on solid (8 g L -1 Bacto agar) SHb medium, which consisted of Shenk & Hildebrandt (1972) inorganic and organic components plus 11 mg L -1 2,4-D and 1mg L -1 kinetin, followed by culture on BOi2Y that consisted of modified Blaydes (Bingham et al., 1975) inorganic and organic constituents plus 100 mg L -1 meso-inositol, 2 g L -1 yeast extract and 10 g L -1 Bacto agar.Large green embryos were formed and transferred after 14 days to hormone-free SH or B 5 media, continued development and begin rooting after 14 days.The responses of 'Rangelander', 'Regen S' and 'Rambler' were 100, 67 and 53%, respectively, of individual calli forming embryos.The last two produced somatic embryos from which plants could be regenerated and successfully acclimatized.Plant regeneration was achieved.
Based on previous studies on forage plants, namely Trifolium rubens, it is possible to regenerate plants from leaf tissues and cell suspension-derived protoplasts (Grosser & Collins, 1984).Therefore the protocol of Atanassov & Brown (1984) was used to obtain somatic embryos and, subsequently, plants from M. sativa 'Rangelander' protoplasts.This system should allow microparticle bombardment and protoplast electroporation, the principal mechanisms for direct gene delivery into legume plant cells (Aragão et al., 1999;Potrykus, 1991;Quecini et al., 2002).Genetic engineer-ing methods are very useful for the transference of disease resistance genes into alfalfa genome in Brazil, namely to control the fungi Rhizoctonia solani and Colletotrichum trifolii.
The present paper describes plant regeneration from protoplasts isolated from five varieties of M. sativa L. Using histological techniques, the characteristics of the protoplast-derived somatic embryos were investigated.Phenotypic and molecular analyses were also carried out between the donors and corresponding regenerated plantlets.The RAPD technique was used for genomic DNA fingerprinting.

Plant material
Seeds of four alfalfa cultivars Crioula, Alfa 200, Valley Plus and Semit 921, cultivated in the subtropical regions of South America, and of the cultivar Rangelander (CSIRO, Division of Plant Industry, Australia), were disinfected by immersion in a 1.5% (v/v) NaOCl solution for 6 min, followed by 4 rinses in sterilized water.Seeds were scarified and placed on half-strength MS basal medium (Murashige & Skoog, 1962) to obtain aseptic material.Because of the small number of seeds available for cultivar Rangelander, nodal cuttings were used to produce whole plantlets within 6 to 7 weeks from the beginning of each subculture.Propagation was done through axillary branching.
Protoplasts were cultured at three densities, i.e. 0.5, 1.0 and 1.5 x 10 5 protoplasts per mL of agarose-solidified (0.6%, Sigma type VII) K8P medium (Kao, 1977; modified by Gilmour et al., 1989) for 5 d in the dark.Cultures were then kept at 25 ± 2°C, with a 16 h light regime, and 25 µM m -2 s -1 light radiation.A mixture of K8P: K8 (Kao, 1977;modified by Gilmour et al., 1989) was used for a progressive replacement of one third of the bathing medium after 7 (2:1), 14 (1:1) and 21 (0:1) days.At this time, plating efficiencies were calculated as the number of colonies in relation to the initial number of plated protoplasts by microscopic evaluation.Twenty fields (20 X objective) and 3 plates were used for each Scientia Agricola, v.60, n.4, p.683-689, Oct./Dec.2003 combination, i.e., genotype vs. protoplast source vs. plating density.Data were analyzed by current statistical analysis methods.

DNA fingerprinting protocols
Molecular analysis based on Random amplified polymorphic DNA (RAPD) markers was carried out to verify genetic identity, comparing the four explant-donors and the in vitro derived plantlets of 'Rangelander'.Total genomic DNA was extracted from leaf tissues (250 mg) using a modified CTAB method as described by Vieira et al. (1997).DNA samples were diluted in TE buffer and submitted to electrophoresis (3 V cm -1 ) in 0.8% agarose gels (w/v).DNA was stained by gel immersion into ethidium bromide solution (10 µg mL -1 ) for 30 min.To quantify the DNA, the fluorescence intensities of the samples were compared to those of l DNA standards.
RAPD reactions were performed in a final volume of 25 µL in PCR buffer (10 mM Tris-HCl pH 8.3, 50 mM KCl) containing 3.0 mM MgCl 2 , 200 µM primers (Operon Technologies, Kit A), 0.2 mM of each dNTP, and 1.5 units of Taq DNA polymerase (Gibco BRL).Ten, 20 and 30 ng of template DNA were used.Reactions were covered with mineral oil and submitted to the following PCR program (PCR Express Hybaid machine): preliminary DNA denaturation for 5 min at 94°C, followed by 45 cycles consisting of denaturation (1 min, 94°C), primer annealing (1 min, 35°C), and extension (2 min, 72°C).A final extension for 6 min at 72°C was included.The RAPD products were separated by electrophoresis (3 V cm -1 ) in 1.4% agarose gels, which run with 1 x TBE buffer.Photo documentation was performed under UV light using a video imaging system (Bio Rad).
RAPD data were analyzed using locus-to-locus gel readings and the rates of in vitro DNA polymorphism were calculated.Only polymorphisms present in all DNA concentrations were included in the analysis.The formula of Dice (1945) was used to calculate the similarity between each donor and its embryo-derived plantlets.

In vitro procedures
The enzymatic treatment was efficient for liberating cotyledon or leaf-derived protoplasts, and high viabilities were observed for all genotypes (Table 1).Cultured protoplasts grew very fast.After 24 h the cell-walls were regenerated, and first mitotic divisions were subsequently observed.Cell colonies were observed after 7 d (Figure 1 A-D).Microcalli were prominent in all cultures after 21 d, and cotyledon-derived protoplasts cultured at 0.5 x 10 5 led to superior plating efficiencies.'Crioula' cotyledonderived protoplast cultures developed calli at a plating efficiency ratio of 17.5% while from leaf-derived plates, the ratio was only 2.2% (Figure 2).Differences between densities, protoplast sources, and cultivars as well as the interaction between densities versus protoplast sources (explants) were all significant (Figure 2) (P < 0.05).At all densities, the response of cotyledon-derived cultures was statistically different in comparison to leaf-derived cultures.
Leaf-derived protoplasts isolated from 'Rangelander' were very competent for embryogenic callus formation when cultivated at 1.0 x 10 5 cells per mL of K8P protoplast culture medium.Approximately 100 calli developing from the five cultivars were transferred to solid medium (SHb) containing a high auxin/low cytokinin ratio i.e. 11 mg L -1 2,4-D and 1mg L -1 kinetin.After the transfer to BOi2Y medium, only 'Rangelander' cultures gave rise to somatic embryos.Although very promising at the cellular stage, no differentiation was observed on the 'Crioula', 'Alfa 200' and 'Valley Plus' plates.To induce embryo formation on the 'Crioula', 'Alfa 200' and 'Valley Plus' cultures (data not shown), colonies were transferred to UM (Uchimiya & Murashige, 1974) containing 2,4-D at 2, 4, 6, 8 and 10 mg L -1 , and also to B1INK (Parrot et al., 1995) media, but all attempts failed to induce embryos.As previously suggested, high influence of the genotype on the in vitro response of alfalfa, especially for somatic embryogenesis was observed (see Bingham et al., 1975, Kao & Michayluk, 1980, Chen et al., 1987;Atanassov & Brown, 1984;Du et al., 1994).
On average, 20.3 ± 12.7 embryoids per callus occurred in the 'Rangelander' cultures (Figure 1E).In addition secondary embryoids were easily proliferated.After 20 d, cotyledonary embryos were excised and transferred to half-strength MS (Figure 1F).Whole plantlets were formed in the subsequent 90 d (Figure 1G-H).Acclimatization to greenhouse conditions was successful.
Electroporation is an efficient technique to introduce exogenous DNA into plant cells (Potrykus, 1991).In Stylosanthes guianensis, a tropical forage legume, direct DNA uptake using an electroporation-mediated method (3.2 x 10 7 electroporated protoplasts produced 50 regenerants, from which 18 were transgenic) was superior to a biolistic approach (Quecini, 1999;Quecini et al., 2002).The protocol described herein could be used for genetic transformation of alfalfa, and somatic hybridization experiments involving Medicago species.

Anatomical studies
Embryo-like structures were observed after 60 d in the 'Rangelander' cultures (Figure 3a).Serial longitudinal sections revealed that these structures were bipolar, with a defined protoderm, ground meristem and procambium, typical of a heart stage embryo (Figure 3c).The somatic embryogenesis was as asynchronous as described for other legumes like Stylosanthes scabra and Glycine max by Dornelas et al. (1992) and Fernando et al. (2002), respectively.Some alfalfa embryos presented new embryoids of epidermal origin, close to their hypocotyledonary region.
On the Semit 921 plates, embryo-like structures, similar to those described by Téoulé & Dattée (1987),  were observed (Figure 3b).However, the histological analysis showed that they were friable calli with typical parenchyma cells.Root meristems were identified, characterizing the rhizogenesis process (Figure 3d).

RAPD analysis
Template DNA concentrations of 20 and 30 ng produced clear PCR profiles.RAPD fingerprints were always identical to that of the explant-donor plants, but in a leaf-protoplast derived plant of 'Rangelander', primers A01, A04 and A12 detected differences, i.e. among 19 loci, 6 (31.5%) were polymorphic (Figure 4).The similarity was 81.25% according to the Dice coefficient.Although in vitro-induced polymorphism at DNA level was observed, no morphological or even reproductive features of the adult plant were anomalous.
Several authors have used molecular markers to monitor genetic stability in in vitro culture-derived plants,  including legume species.Gesteira et al. (2002) used RAPD markers to evaluate genetic stability of regenerants of soybean plants, obtained through somatic embryogenesis using 180 µM 2,4-D.Twenty primers were used to screen 44 regenerants from two soybean cultivars.Three of them were polymorphic for two of the 'Spring'-derived regenerants, with somaclonal variants occurring at a frequency of 4.5%, and four primers were polymorphic for the 'CAC'-1-derived regenerant, with somaclonal variant frequency of 3.57%.Piccioni et al. (1997) reported the incidence of molecular variation in 9 out of 39 (23%) 'Rangelander' alfalfas derived from somatic embryogenesis.RAPDbased similarity was calculated, and ranged from 72 to 97%.These authors attributed the differences to somaclonal variation.RAPD markers are absolutely useful to detect molecular instability in primary regenerant plants derived via somatic embryogenesis.This is the first report on plant regeneration from 'Rangelander' alfalfa protoplasts using a protocol defined for alfalfa cell suspension-derived embryogenesis.In summary, the use of agarose-embedded protoplasts at 1 x 10 5 per mL of the culture medium (K8P), and the osmotic reduction protocol led to microcalli development in 28 days.Somatic embryo induction, development and conversion into plant were achieved by using the protocol described by Atanassov & Brown (1984) (Table 2).Because of its agronomical importance and highly morphogenic ability, the use this protoplast-to-plant technology may be useful for genetic manipulation of the cultivar Rangelander.5 d in the dark →25 µEm -2 s -1 1.0 x 10 5 (leaf) 0.5 x 10 5 (cotyledon) Droplets (leaf); Thin layers (cell suspensions) Bathing medium: Kao (1977) with 81 g L -1 glucose, 75 mg L -1 penicillin G, 5 mg L -1 streptomycin 7 µEm -2 s -1 for 24 h →25 µEm -2 s -1 5.0 x 10 3 (leaf) 1.0 x 10 5 (cell suspension) -Osmotic 1/3 of the bathing medium was substituted by a mixture of K8P: K8 after 7, 14 and 21 d as follows: 2:1, 1:1 and 0:1 At 17-21 d of culture, 300-500 µL of Kao12 were added i.e.Kao (1977) plus 36 g L -1 glucose, 5 g L -1 sucrose, 0.2 mg L-1 2,4-D, 1 mg L -1 NAA and 0. SH or B5 with 10 g L -1 sucrose (9 g L -1 agar)

Figure 2 -
Figure 2 -The influence of the source of protoplasts, plating density (x 10 5 ) and period of culture on the plating efficiencies (PE) evaluated by the number of microcolonies in relation to total number of protoplasts by microscope field in five alfalfa genotypes.

Table 1 -
Production, viability and plating efficiency (PE) in protoplast cultures of M. sativa obtained from cotyledon (C) explants at density of 0.5 x 10 5 , and leaf explants (L), at density of 1.0 x 10 5 , according to the cultivar.