Bedford were initiated in vitro from seedling ex- plants consisting of the ... and PI 438489B than from Bedford. These results coq- ... explants (ca. 2 mm long) ...
Plant Cell Reports
Plant Cell Reports (1989) 8:199-202
© Springer-Verlag1989
Somaclonal variation in soybean plants regenerated from tissue culture *' ** A. H. Freytag 1, A. P. Rao-Arelli 1, S. C. Anand l, j. A. Wrather 1, and L. D. Owens 2 1 Department of Agronomy and Department of Plant Pathology, University of Missouri, Delta Center, Portageville, Missouri, MO 63873. USA z Plant Molecular Biology Laboratory, USDA-ARS, Beltsville, Maryland, MD 20705. USA
ABSTRACT Callus cultures of soybean (Glycine max (L.) Merr.) genotypes PI 88788, PI 438489B, and cultivar Bedford were initiated in vitro from seedling explants consisting of the cotyledonary node plus epicotyl from germinated mature seed. Plants were regenerated from these callus cultures and subsequently evaluated for qualitative variation in three to four subsequent generations. Variant phenotypes observed that have not been previously reported from tissue culture include lanceolate leaves, leaf variegation (chimeral variegated plants), pod variegation on otherwise normal plants, and change in growth habit from indeterminate to determinate. The lanceolate leaf, chimeral variegated plant, and change from indeterminate to determinate growth habit characters were inherited through at least three generations (R0-R2), and segregation occurred in each generation. Pod variegation was inherited through the two generations tested thus far and segregation occurred in each generation. No variation was observed in control plants derived from normal seed. Variants appeared more frequently in regenerants from PI 88788 and PI 438489B than from Bedford. These results coqfirm and extend the finding that certain tissue culture techniques may be used to induce novel plant formation from somatic tissue of soybean. INTRODUCTION The phenomenon of novel variation or somaclonal variation observed in tissue culture derived plants is well recognized (Hughes 1986). Several explanations for this variation have been documented, including gross chromosomal abnormalities, cytoplasmically controlled mutations, nuclear genic mutations and transposable elements (Orton 1984, Peschke et al. 1987). Further, spontaneous mutations arising from tissue culture have been reported to occur with high frequency as compared to conventional mutation programs (Rice 1982). Recently, somaclonal variants for certain qualitative and quantitative genes have been reported from soybean (Kollipara et al. 1987, Barwale and Widholm 1987, Graybosch et al. 1987). However, the full range of somaclonal variation possible in soybeans has yet to be determined. Such variation may have value in soybean improvement programs. In this study we regenerated plants from cultured cotyledonary node plus epicotyl explants of soybean seedlings, compared the regenerated plants with controls for variants, and determined the in-
heritance
of the novel traits. MATERIALS AND METHODS
Plant Materials Soybean genotypes used in this study were plant introduction accessions PI 438489B, PI 88788 and the cultivar Bedford. Media The RV-5 medium used is described in Table i. Media RV-5 (Table i) and Gamborg's B5 (Gamborg et al. 1968) were adjusted to pH 5.8, solidified with agar (0.65%) and sterilized by autoclaving at 1.05 kg/cm 2 for 15 minutes.
Table i.
Composition
of Media Used
Constituents
RV-51
Inorganic salts Sucrose (g/l) Vitamins
MS 2 25.0
(mg/l):
p-Aminobenozic acid Ascorbic acid Niacin d-Pantothenic acid Pyridoxine HCI Riboflavin Thiamine HCI Amino Acid
0.2 0.4 0.5 0.4 0.5 0.015 1.0
(mg/l):
L-Arginine (free base) L-Aspargine (anhydrous) Glycine L-Glutamine Growth Regulators
(mg/1):
Indolebutyrie acid (IBA) 6-Benzyladenine (BA)
IA modification et al. 1988) 2Murashige
40.0 40.0 20.0 60.0
of RV medium
and Skoog
0.I 0.4
(Freytag
(1962)
* Missouri Agricultural Experiment Station, University of Missouri, Columbia, Missouri, USA ** Mention of tradenames does not constitute a guarantee or warranty of the product by University of Missouri or USDA-ARS and does not imply their approval to the exclusion of other products. Offprint requests to." A. H. Freytag
200 In Vitro Culture The cotyledonary node plus epicotyl explants were derived from 200 germinated mature field-grown seed harvested in 1986 and stored at 4 ° C until used. Seeds were surface-sterilized by immersing in 15% Clorox bleach (5.25% sodium hypochlorite) for 15 minutes. After rinsing twice in sterile distilled water the seeds were placed in Petri dishes (i00 x 15 mm, 5 seeds/dish) containing approximately 30 ml of RV-5 medium and incubated at 26 ° C in the dark. One week later, the cotyledons were excised (i mm from point of attachment) from healthy germinated seed, and 150 explants (ca. 2 mm long) consisting of the shoot tip from just below the cotyledonary node were plated (4 explants/plate) on fresh RV-5 medium. Explants were incubated at 26 ° C and illuminated 12 hours daily (25 ~E s-i m-2). Within six weeks, callus covers the shoot tip explant and multiple shoots have developed from within or below callus. Histology studies have shown that these shoots arise de nova from the swollen base of the excised explants (Hinchee 1988). Calli with shoots were cut into two pieces, and each was cultured (2/dish) on RV-5 with biw~e~lytransfers. In all cases the plates were randomly distributed in the growth chamber and cultured as above. When 1 cm long, shoots along with an attached callus were plated on B5 medium (Gamborg et al. 1968) containingIBA (5.0 mg/l) to initiate roots. These were incubated as above. Roots formed on the shoots within 4 weeks. The resulting plant, with 2 cm long shoots, were potted in sphagnum moss and eventually in soil as described by Freytag et al. (1987, 1988). Plants were maintained in a Phototron growth chamber (Pyraponic Industries) at 26 ° C with a 12 hour daily photoperiod (25 ~E s-i m-2). Each regenerated plant was assigned an identification code that included the parental genotype and a number, and seeds from these plants were kept separately according to generation. Regenerated plants (R0) were transferred to the greenhouse where they were randomly distributed along with control plants from seed and observed over several months. From 200 germinated seed of each soybean line, 150 explants of each were cultured. Callus developed on these explants and 131 R0 plants were generated (56 PI 438489B, 60 PI 88788, and 25 Bedford). R1 seed from all R0 plants were planted in progeny rows along with rows of control plants in the field the following May and allowed to self-pollinate. Distribution of progeny rows in the field was random. Similarly, R2 seeds from individual R1 variant plants were planted the following year in the field along with control seed to observe segregation of variant traits. Only seed from variant plants observed in the R0 and/or R1 generations were planted for further observation in the R2 generation. The number of control plants used for each generation included at least 5 of each genotype for the R0 generation, 25 for the RI, and 50 for each of the R2 and R 3 generations. RESULTS AND DISCUSSION Phenotypic Variation The R0 plants were selfed to produce 131 R1 families. Of the 131 R1 families, 13 produced variants in the R0 and/or R1 generations (Table 2), and these were selfed to produce R2 seeds. With one exception, only those variant traits observed in the R0 and R1 plants were observed in later generations. The exception was with PI 88788-1. This plant was initially selected for a change from indeterminate growth habit to determinate, but later a variant having dark green, large, thick leaves, resembling those of polyploid plants, was observed in the progeny row of the
R 3 generation (Table 2). Whether this trait is heritable remains to be determined. The frequencies with which several classes of variants were observed are presented in Table 2. The variation observed included; albino chimeras, chimeral variegated plants (Fig. IA), pod variegation (Fig. IB), lanceolate leaves (Fig. IC), large, thick leaves, and change from indeterminate to determinate growth habit (Fig. ID). One of these characteristics, lanceolate leaves, was inherited through three generations (R0-R2), and another, indeterminate to determinate growth habit, through four generations (R0-R3). Two other variant phenotypes, chimeral variegated plants, and chimeral albinos were inherited through three generations (R0-R2). Variegated pods only was inherited through two generations (R0-RI). Control plants from seed showed no variants of these or other types. To date this is as far as the above generations have been observed, but testing will continue. R0 plants from PI 438489B appeared normal in color but segregated in the R1 generation for albinism. The ratio was close to 13 normal green; 3 albino plants. The R2 plant ratio for albinism was approximately 7: 8 : 0 (normal green: segregating: albino). Since R1 albino plants did not survive to produce seed, homozygous albino plants were not recovered in R2 generation. Within the small populations evaluated, a single nuclear recessive gene in PI 438489B appeared to control the variegated pod character. The R1 plant ratio for variegated pods was approximately 27 normal and 12 variegated. The variant determinate growth habit was derived from PI 88788 and PI 438489B that are both indeaerminate in growth habit. The R1 population for PI 88788 segregated into 3 indeterminate and 1 determinate, which appeared to be controlled by a single recessive gene. A recessive gene also appears to control the determinate trait in PI 438489B which has segregated into 17 indeterminate and 3 determinate in the R1 generation. The lanceolate leaf variant in PI 88788 and Bedford has been very unstable and does not fit any logical ratio. Data from R2 generation did not fit into any known genetic ratio. Graybosch et al. (1987) using a cotyledonarynode tissue culture system of shoot regeneration observed two genetic variants of height difference and sterility. Barwale and Widholm (1986) reported five genetic variants non-lethal chlorophyll deficiency, complet~ sterility, wrinkled leaf, dwarf plants and multiple shoots. We have observed four different genetic variants which have not been reported previously:lanceolate leaves, change from indeterminate growth habit to determinate, variegated pods only (normal leaves), and chimeral variegated plants. None of the variant plants have been tested as yet for agronomic variability such as yield, oil content, maturity, etc. The production of a wider range of variants in our study than was reported by Graybosch et al. (1987) may reflect the number of cells in the original explant. Graybosch et al. (1987) used only the cotyledonary node tissue, which we included the entire shoot apex along with the cotyledonary node tissue. Shoots emerged from primary callus in our study, but whether they arose from organogenesis or from development of pre-existing meristems was not determined. Regardless of the mode of regeneration, whether by true organogenesis or shoot multiplication, the final outcome was the production of a range of somaclonal variants. These results confirm that certain tissue culture techniques may be used to enhance the induction of novel plant formation from somatic tissue of soybean and expand the range of phenotypes that have been obtained with this species through
201 tissue culture.
Table 2.
Genotype PI 88788
Soybean regenerant
families in different generations
showing vaiant phenotypes
Number of Plants in Observed Generation Variant Phenotype
Number of Plants Expressing Variation
Frequencyl (%)
Family Number
Generation
6 & 15
R0 R1 R2
2 54 1152
Lanceolate
leaves
2 5 26
i00 9 2
leaves
1 1 i0
i00 3 2
Bedford
1
R0 R1 R2
1 33 588
Lanceolate
PI 88788
1
R0 R1 R2 R3
1 28 451 549
Indeterminate growth to determinate
1 3 25 23
i00 i0 5 4
PI 438489B
21
R0 R1 R2 R3
1 20 400 510
Indeterminate growth to determinate
1 3 12 17
i00 2 3 3
PI 438489B
15
R0 R1
1 39
Varigated pods only, rest of plant normal
1 12
i00 15
I
R0 R1 R2 R3
1 28 451 549
3,6,8 & 9
R0 R1 R2
4 116 1877
Chimeral varigated
RO R1 R2
6 147 2490
Chimeral albino
PI 88788
PI 438489B
PI 438489B
1,2,3,4,8
i Number of variants
& 9
Large thick leaves dark green (polyploid?)
plants
divided by the total number of plants in that generation.
0 0 0 1
0 0 0 0.2
0 4 12
0 3.0 0.6
0 4 81
202
Fig. i. Phenotypic variation observed in the RI-R2 generations. A. Leaf variegation on an R2 segregant. B. Variegated seed pods on an otherwise normal R1 plant. C. Normal plant (left) and lanceolate leaves on an R1 plant (right). D. R2 progeny rows of indeterminate wild type (left) and determinate variants (right).
ACKNOWLEDGEMENT This work was supported by USDA-ARS cooperative agreement 58-3204-6-172 which is gratefully acknowledged. We also appreciate the gift of Phototron growth chambers from Pyraponic Industries, 7868 Silverton Suite B, San Diego, CA 92126. We gratefully appreciate the technical assistance of Mrs. Joyce Elrod and Mrs. Nancy Reed.
REFERENCES Barwale UB, Kerns JR, Widholm JM (1986) Planta 167: 473-481. Freytag, AH, Anand SC, Rao-Arelli AP, Owens LD (1988) Plant Cell Rep. 7:30-34. Freytag, AH, Anand, SC, Rao-Arelli AP (1987) Current Science 57:No.6:291-294. Gamborg OL, Miller RA, Ojima K (1968) Exp. Cell Res. 50:151-158. Graybosch RA, Edge ME, Delannay W (1987) Crop Sci. 27: 803-806. Hinchee M (1989) 19th Stadler Genetics Symposium University of MO, Columbia, MO Hughes KW (1986) In: Crowley JR (ed). Research for tomorrow. 1986 Yearbook of Agriculture, USDA Washington DC. pp134-138. Kollipara KP, Singh RJ, Hymowitz T (1987) Amer. Soc. Agronomy Abstract 150. Murashige T, Skoog F (1962) Physiol. Plant 15:473-497. Orton TJ (1984) Gustafson JP (ed) Plenum Press pp 427468, London, New York. Peschke VM, Phillips RL, Gengenbach BG (1987) Science 238:804-807. Rice TB (1982) Proc. 37 Ann. Corn and Sorghum Ind. Res. Conf. pp148-162 American Seed Trade Assoc., Washington, DC.
