Plant Cell Reports
Plant Cell Reports (1996) 15:381-385
Rapid and repetitive plant regeneration in sweetpotato via somatic embryogenesis Qi Zheng, Ananta Porobo Dessai, and Channapatna S. Prakash Plant Molecular and Cellular Genetics Lab, School of Agriculture and Home Economics, Milbank Hall, Tuskegee University, Tuskegee, AL 36088-1641, USA Received 29 August 1994/Revised version received 21 July 1995 - Communicated by G. C. Phillips
Abstract. An efficient in vitro plant regeneration system character-
1979, Liu and Cantliffe 1984, Jarret etal. i984, Desamero etaL 1994).
ized by rapid and continuous production of somatic embryos using leaf and petiole explants has been developed in sweetpotato [Ipomoea batatas L. (Lam.)]. The optimal somatic embryogenic response was
Factors influencing embryogenic callus production in solid and liquid media (Chee and Cantliffe 1988b, Chee and Cantliffe 1989), embryony patterns and plantlet regeneration (Chee and Cantliffe
obtained in the genotype PI 318846-3 with a two-step protocol: (1)
1988a, Chee et al. 1990) have been well studied in sweetpotato cv.
stage I-incubation of explants in the dark for 2 weeks on MurashigeSkoog (MS) medium containing 2,4-dichlorophenoxyacetic acid
'White Star' using the apical meristem as the explant. However, direct somatic embryo production from leaf and petiole explants in a
(2,4-D) (2.5 mg/1) and 6-benzylaminopurine (0.25 mg/1) and, (2) stage
repetitive fashion has not been reported in this crop. Availability of a
II-culture in the light on MS medium with abscisic acid (ABA) (2.5 mg/l). The addition of ABA was critical for enhanced production
reliable and reproducible somatic embryogenic system that is rapid,
of somatic embryos. Secondary somatic embryos were produced from the primary embryos cultured on MS medium with 2,4-D at 0.2 mg/1. The somatic embryos were converted into normal plantlets when cultured on basal MS medium. Upon transfer to soil, plants grew well and appeared normal with no mortality. The system of somatic embryogenesis described here will facilitate tissue culture, germplasm con-
repetitive and that provides a choice of direct and callus-mediated embryogenesis will be useful in sweetpotato micropropagation, germplasm conservation and gene transformation research. Here, we report results of studies aimed at developing a prolific and rapid system of somatic embryogenesis in sweetpotato, a crop of considerable importance as a source of protein and calories in the developing world.
servation and gene transfer research of sweetpotato due to its rapidity (6 to 10 weeks), prolific plant production by direct embryogenesis,
Materials and Methods
ease of secondary somatic embryo production and reproducibility.
Abbreviations: ABA=abscisic acid, BAP=6-benzylaminopurine, 2,4-D
The sweetpotato genotypes were obtained from the USDA Plant Introduction Center, Griffin, Georgia, as in vitro cultured shoot tips.
=2,4-dichlorophenoxyaceticacid, GA3=gibberellicacid, KIN=kinetin,
Explants were obtained from plants grown in vitro on a multiplica-
MS=medium of Murashige and Skoog (1962), NAA= 1-naph-
tion medium consisting of MS inorganic salts and vitamins (Murashige
thaleneacetic acid, PIC=picolinic acid, TDZ=thidiazuron
and Skoog 1962), myo-inositol 100 mg/1, gibberellic acid (GA3) 5 mg/1, sucrose 30 g/1 and Phytagel 3 g/1. The basal medium consisted of MS inorganic salts, myo-inositol 100 mg/l, thiamine-HC1
Introduction Regeneration of plants in vitro using somatic embryogenesis has some distinct features such as single-cell origin, the consequent low frequency of chimeras and the production of a high number of regenerates (Haccing 1978, Ammirato 1983, Sato et al. 1993). Thus, the somatic e m b ~ o system is being increasingly employed to produce transgenic plants and to develop synthetic seeds using encapsulation strategies (Ritchie and Hodgcs 1993, Redenbaugh et al. 1986). Somatic embryos have been produced in sweetpotato (Tsay and Tseng Correspondence to: C. S. Prakash
1.7 mg/1, nicotinic acid 1.2 mg/1, pyridoxine-HC1 1.0 mg/1, sucrose 30 g/1 and Phytagel 3 g/1. The stage I medium consisted of basal medium supplemented with potassium chloride (2.235 g/I) (Chee et al. 1992), while stage II medium consisted of basal medium with
ammonium nitrate reduced to half-strength (Chee et al. 1992). The medium pH was adjusted to 5.8 prior to the addition of Phytagel (Sigma), and autoclaved at 121~ 105 KPa for 20 min. The stage I and II media were further modified as noted for each study. The optimized stage I medium had additions of 2,4-D at 2.5 mg/1 and BAP at
382 125 lOO 100 7s
J 25 0 1 0
No. of Weeks Cultured on Stage I Medium
Fig. 3. Effect of duration of incubation of explants on stage I medium on the frequency of somatic embryogenesis in sweetpotato. The data (meansS:s.e) recorded after 8 week incubationon stage II medium.
fixed in 2% glutaraldehyde in 0.1 M cacodylate buffer for 4 h, washed using the same buffer and dehydrated with a graded ethanol series (Wetzstein and Baker 1993). Tissues were dried with a Tousimis
Growth Regulators (mg/i)
Fig. 2. Effect of 2,4-D and cytokinin (BAP or TDZ) levels in the stage I medium on the frequency of somatic embryogenesisin sweetpotato. The data (mear~.+s.e) recorded after 8 week incubation on stage II medium.
0.25 mg/l, while the optimized stage II medium contained additional
critical point dryer, using liquefied CO 2 as the transitional fluid. Samples were mounted on stubs, sputter-coated with gold/palladium, and viewed using a Philips 505 scanning electron microscope. Results and Discussion
ABA (+99% cis-trans; Sigma) at 2.5 mg/1. In stage III medium, ABA was deleted from stage II medium, and was used for germination and
Various combinations of auxins (NAA, PIC or 2,4-D) and cytokinins (BAP, TDZ or KIN) were tested with stage I medium (8 weeks) for
conversion of somatic embryos to plants. Cut pieces of lamina (10 x 5 mm) and petiole (5 mm) isolated from
their potential to induce somatic embryos in sweetpotato using geno-
the apical (first and second) nodes of the shoot and internodal seg-
types PI 318846-3 and PI 531143. Culture of explants on stage I
ments (5 mm) of in vitro grown plants of genotype PI 318846-3
medium containing 2,4-D at 1.0 or 2.5 mg/1 (with or without BAP or
(unless indicated otherwise) were used as explants. The explants were
TDZ at 0.25 mg/1), and followed by 8 week culture on stage II
placed in petri dishes (100 x 15 mm) containing 30 ml of stage I
medium, produced somatic embryos only in genotype PI 318846-3
medium with various concentrations of growth regulators. After 2 - 8
but not PI 531143. Culture of explants on media with NAA or PIC
weeks of incubation on stage I medium in the dark at 25~
did not result in the production of embryos (data not shown). The
were transferred to stage II medium with or without ABA. All
genotype PI 318846-3 was also highly regenerative in studies with
cultures were incubated in 16h/8h photoperiod at 25~ in a growth
adventitious shoot regeneration (Gosukonda et al. 1995).
chamber. The data were collected at periodic intervals for up to 10
Two major types of calli were observed on the leaf and petiole
weeks. Well developed somatic embryo clusters were separated and
explants toward the end of culture period on stage I medium: embry-
transferred to stage III medium. To produce secondary somatic
ogenic callus which appeared yellowish, compact and slower grow-
embryos, primary embryos at late cotyledonary stage were cultured
ing (Fig. 1A), and non-embryogenic callus which was white, friable
on stage I medium containing 2,4-D and BAP (various levels). Plantlets were transferred to pots containing a sterile mixture of perlite
and faster growing with loose cells (Fig. 1B). The auxin-cytokinin
and vermiculite (1:1), initially placed under high humidity in a growth
genic calli (loose, snow-white and yellow).
chamber and subsequently grown in the greenhouse. For scanning electron microscopy, embryogenic explants were
combinations tested without 2,4-D yielded a variety of non-embryoTo further optimize and evaluate somatic embryogenesis, 2,4-D was tested at 1.0, 2.5, or 5.0 mg/1 individually or in combination with
Fig. 1. Somatic embryos in the sweetpotato genotype PI 318846-3. A & B: Embryogenic(A) and non-embryogenic (B) callus developed on the leaf explants. C: Leaf explant showinggreen somatic embryos and embryogeniccallus mass. D and E: Somaticembryos at various stages of development.F: Secondarysomaticembryos (se) arising from the primary somaticembryo (pc). G: Secondaryembryos at cotyledonarystage. H: Sweetpotatoplantlet obtainedfrom germinatedsomaticembryosin stage III medium. I" Sweetpotato plants developed from somatic embryos growing in the soil.
384 effective treatment in inducing somatic embryos as all the explants developed embryos (Fig. 3). Those explants subjected to a 2-week incubation on stage I medium developed somatic embryos directly from the leaf with very little calli when cultured on stage II medium.
In contrast, the explants subjected to prolonged incubation on stage I medium exhibited greater callus proliferation and predominantly callus-mediated embryogenesis on stage II medium. We have rou-
tinely observed that, under optimal conditions, somatic embryos first
appeared after 2 weeks of culture of explants on stage II medium, and all the explants turned embryogenic after 6 weeks of culture. Earlier studies (Lin and Cantliffe 1984, Janet et al. 1984) have shown that 2,4-D at 0.1 to 3.0 mg/l was effective in inducing somatic embryos in sweetpotato. Chee and Cantliffe (1988a) have suggested
the use of 2,4-D at 10 gM (2.2 mg/1) for the induction of embryogenic callus from the apical meristem, and 2,4-D at 10 RM and B A P
Abscisic Acid (ABA)Conch. (mg/l) Fig. 4. Effect of abscisicacid (ABA)in the stage II mediumon somaticembryogenesis in sweetpotato(means_+s.e.).
at 1 pM for indefinite maintenance of the embryogenic potential of the callus. Recently, Desamero et aL (1994) have reported somatic embryogenesis in sweetpotato cv. Regal using PIC and KIN with apical meristem explants. While our results show that somatic embryos can
BAP, TDZ or KIN each at 0.25 mg/l in stage I medium. At 8 weeks of
be produced on a 2,4-D medium (2.5 rag/l) with or without BAP
incubation on stage II medium (following 8 weeks on stage I
(0.25 mg/l), the presence of BAP in stage I medium dramatically
medium), 78% of the leaf explants cultured on 2,4-D at 2.5 rag/1 and
enhanced the embryo induction (Fig. 2).
BAP at 0.25 mg/1 produced somatic embryos. All other treatments
After 2 weeks of incubation on stage I medium containing 2,4-D at
resulted in less than 40% of the explants producing somatic embryos
2.5 mg/l and BAP at 0.25 rag/l, leaf explants of PI 318846-3 were
(Fig. 2). Both leaf and petiole explant types produced somatic embryos
transferred to stage H medium containing ABA at 0, 2.5, 5.0 or 10.0 mg/l. The use of ABA at 2.5 mg/1 resulted in 100% of the explants producing somatic embryos (Fig. 4). Under optimal conditions, each explant produced more than 50 somatic embryos/cm2. The use of ABA promoted the production of a larger number of somatic embryos per explant, the rapid development of embryos and increased conversion of embryos into plantlets. Although the mode of action of ABA during embryo development is not well understood (Hetherington and Quatrano 1991), it has been widely employed to synchronize somatic embryogenesis and to improve developmental attributes and conversion of somatic embryos (Nickle and Yeung 1994). The prolonged exposure of somatic embryos to ABA in stage II medium in our study (3-4 weeks) may have stimulated embryo development and subsequent germination and conversion of embryos in stage III medium (without ABA). Short, rather than long, exposure to 2,4-D in
Fig. 5. Scanningelectronmicroscopeof somaticpro-embryosarisingdirectlyfrom sweetpotatoleaf explant
in comparable frequencies. When leaves isolated from various positions on the stem (differing in developmental maturity) were compared,
stage I medium may also be critical to the direct induction of somatic embryos from leaf explants. The typical developmental stages were observed in somatic embryos developing from both callus-mediated or those directly arising from the explants. Pro-embryonic masses with tightly packed
there were no consistent differences in the rate of somatic embryo; production (data not shown).
isodiametric cells developed from the leaf explants or calli (Fig. 1C) from which the globular, heart, torpedo and cotyledonary stages
In a time course study, leaf explants from PI 318846-3 were cttltured for 0, 1, 2, 4, 6 or 8 weeks on the 'optimized' stage I medium containing 2,4-D at 2.5 mg/1 and BAP at 0.25 rag/1 and subsequently transferred to stage II medium with ABA at 2.5 mg/1. The incubation of explants on stage I medium for 2 weeks was clearly the most
appeared successively (Fig. 1D and 1E). Somatic embryos appear distinctly green against the yellowish background of the callus or the explant. Scanning electron microscopic observations provided an evidence of the direct origin of the pro-embryo from the leaf without intervening callus (Fig. 5). No abnormalities were observed in the
385 somatic embryos, and the developmental stages were similar to those observed in the sweetpotato zygotic embryos (data not shown). Leaf explants initially cultured in the dark on the optimized stage I medium had an etiolated appearance but produced many somatic embryos when transferred to optimized stage I/medium. The elongated torpedo and cotyledonary embryos germinated readily on stage III medium and converted to plantlets (Fig. 1H). These plantlets, when
Acknowledgments. Contribution no. 236 of the George Washington Carver Agricultural Experiment Station. We are grateful to Dr. Kei Shimonishi for providing impetus to this study. We thank Mr. Arlington Weithers and Mr. Thomas Martin for help with photography and art work. We thank Ms. Carol Williams for assistance with electron microscopy study. Assistance of Mr. Matand Kanyand and Dr. Guohao He is also acknowledged. Research supported by grants from USDA under the 1890 Institution Capacity Building Grants Program (ALX-9201871), USDA/CSRS (ALX-TU-SP01) and NASA (NAGW -2940).
transferred to soil showed 100% survival, and grew into plants with normal morphological appearance ( Fig. 1I). Primary embryos developed on stage II medium were transferred back to stage II medium supplemented with various levels of 2,4-D
References Ammirato PV (1983) In: Evans DA, Sharp WR, Ammirato PV, Yamada Y (eds) Handbook of Plant Cell Culture, vol 1. Macmillan, New York, pp 82-123
(0.1, 0.2, 0.5 mg/l) or TDZ (0.1, 0.5, 2.0, 10.0 mg/l). After 2 - 4 weeks of incubation, secondary embryos developed on explants cultured on stage II medium with 2,4-D (Fig. 1F) but not with TDZ (data not shown). Primary somatic embryos cultured on 2,4-D at 0.2 mg/1 developed secondary somatic embryos rapidly and in higher frequen-
Chee RP, Canfliffe DJ (1988a) Plant Cell Tiss Org Cult 15:149-159 Chee RE Cantliffe DJ (1988b) In Vitro Cell Devel Biol 24:955-958 Chee RR Cantliffe DJ (1989) In Vitro Cell Devel Biol 25!757-760
cies compared to the other 2,4-D levels tested (Fig. IF). Secondary somatic embryos were in general more numerous and better synchronized than primary embryos (Fig. IG). This feature may be useful in genetic transformation methods as primary somatic embryos can be subjected to gene transfer using particle bombardment (Prakash and Varadarajan 1992), and transgenic sweetpotato plants may be devel-
Chee RP, Leskovar DI, Cantliffe DJ (1992) J Amer Soc Hort Sci 117:663-667 Chee RP, Schultheis JR, Canfliffe DJ (1990) HortScience 25:795-797 Desamero NV, Rhodes RB, Decoteau DR, Bridges WR (1994) Plant Cell Tiss Org Cult 37:103-111
oped from secondary somatic embryos through selection. Primary somatic embryos in the early torpedo stage produced numerous and faster-growing secondary somatic embryos than did elongated and mature primary somatic embryos. While secondary embryos arose
Gosukonda RM, Porobodessai A, Blay ET, Prakash CS, Peterson CM (1995) In Vitro Cell Devel Bio131:65-71 Haccing B (1978) Phytomorphology 28:74-81
over the surface of primary somatic embryos, the hypocotyl region of primary embryos developed embryos in higher frequencies. Low conversion rates of somatic embryos into complete planflets in sweetpotato have been reported (Padmanabhan et al. 1994). Although we made no attempt to quantify the germination and con-
Hetherington AM, Quatrano RS (1991) New Phytol 119:9-32 Jarret RL, Salazar S, Fernandez R (1984) HortScience 19:397-398 Liu JR, Cantliffe DJ (1984) Plant Cell Reports 3:112-115
version rates in our study, repeated experiments showed that several hundred somatic embryos could be produced directly or via callus
Murashige T, Skoog F (1962) Physiol Plant 15:473-497
from a single leaf explant, and most somatic embryos germinated readily into normal plants. This may be due to the genotype (PI 318846-3) that was employed, the shorter exposure of explants to 2,4-D (2 weeks), the inclusion of BAP, and/or the use of ABA in our
Niclde TC, Yeung EC (1994) In Vitro Cell Devel Biol 30P:96-103 Padmanabhan K, Canfliffe DJ, Harrell RC, McConnell DB (1994) In Vitro Cell Devel Biol 30A: 33
optimized protocol in contrast to other published methods, where explants are exposed to 2,4-D for 8 weeks (Chee and Cantliffe 1988a, 1989). When 10 additional sweetpotato genotypes were further tested, three genotypes-P1508507, 'Red Jewel' and 'White Jewel' produced
Prakash CS, Varadarajan U (1992) Plant Cell Reports 11:53-57 Redenbaugh K, Paasch BD, Nichol JW, Kossler ME, Viss PR Walker KA (1986) Bit/Technology 4:797-801
somatic embryos while P1531143, PI 508513, Hi Dry, Rojo Blanco, White Star, Jewel, and Regal were unresponsive. In summary, we have developed an expeditious protocol to induce somatic embryogenesis from leaf and petiole explants of sweetpotato. Secondary embryos can be produced efficiently in large numbers from
Ritchie SW, Hodges TK (1993) In: Kung S and Wu R (eds) Transgenic Plants, vol I Engineering and Utilization. Academic Press, San Diego, pp 147-194 Sato S, Newell C, Kolacz K, Tredo L, Finer J Hinchee M (1993) Plant Cell Reports 12:408-413
the primary somatic embryos. The plants developed by the protocol were transferred to soil with no loss and appeared normal and healthy. This system has recently been employed in our laboratory to successfully regenerate transgenic sweetpotato plants with various foreign genes (manuscript in preparation).
Tsay HS, Tseng MT (1979) Bot Bull Acad Sin 20:117-122 Wetzstein HY, Baker CM (1993) Plant Science 92:81~89