Plant Cell Reports (1986) 3: 178-181
© Springer-Verlag 1986
Characterization of somatic embryogenesis and plant regeneration in cotton (Gossypium hirsutum L.) R. C. Shoemaker *, L. J. Couche, and D. W. Galbraith University of Nebraska-Lincoln, School of Biological Sciences, 348 Manter Hall of Life Sciences, Lincoln, NE 68588-0118, USA Received July 1, 1985 / Revised version received March 17, 1986 - Communicated by P. Maliga
Summary Seventeen cultivars of cotton (Gossypium hirsutum L.) were evaluated for callus initiation and maintenance using 3 initiation media and 3 maintenance media. After a series of transfers of a 3% glucose media, c~lli were placed on a 3% sucrose medium. After several weeks calli were observed for the presence of embryo-like structures. Cultivars Coker 201 and Coker 315 were identified as embryogenic. Embryogenic callus has since been routinely obtained within 6 weeks by initiating callus on glucose media for 3-4 weeks followed by transfer to sucrose media. Histological examination has shown that embryos are derived from isodiametric, densely cytoplasmic cells and follow predictable patterns of development. Upon maturity, transfer to auxin-free media with reduced sucrose levels results in embryo germination. Regenerated plants can be transferred to greenhouse within 90 days of callus initiation. Key words: Gossypium hirsutum - Somatic embryos Regeneration - Morphogenesis - Genotype Introduction Plant regeneration is a critical step in the success of any crop improvement program entailing tissue culture techniques. Plant regeneration can be achieved in two ways: through organogenesis or through somatic embryogenesis. The latter is the preferred method, for two reasons. It is probable that plants derived from somatic embryogenesis are of singlecell origin (Haccius 1978). Thus, the plants will not be genetic chimeras, as is possible with those derived from organogenesis. Secondly, since nonzygotic embryos have no vascular connection with maternal tissue (llaccius 1978), in principle, they are more easily manipulated than plantlets derived through organogenesis. Attempts to induce somatic embryogenesis revolve around two philosophies: manipulation of a wide range of inductive media and culture conditions in an attempt to incite an embryogenic response from a specific genotype, or evaluation of a large number of genotypes on a narrow range of 'inductive' media. The latter approach has been extended towards the identification of embryogenic genotypes within a species (Bingham et al 1975; Sears and Deckard 1982; Oelck and Schieder 1983). In cotton (Gossypium hirsutum L.), plant regeneration has been achieved in specific callus lines (Davidonis and Hamilton 1983). The experimental conditions that led to the emergence of the regener-
able phenotype were not well defined and required an unusually long time frame (two years) (Davidonis and Hamilton 1983). Preliminary results from our laboratory suggested that the Gossypium sp. germplasm collection contains much genotypic variation for callus initiation, proliferation, morphology and regeneration capacity. The objectives of this study were therefore to evaluate a wide range of genotypes for callus initiation and callus maintenance responses to several different media, to identify cotton genotypes that possess strong regeneration potential, and to characterize regeneration through histological analysis. Materials and Methods Callus Initiation and Maintenance: Seed were aseptically germinated on half-strength MS medium (Murashige and Skoog 1962) in darkness at 32 ° C for 7 days. Germinated seeds were then transferred to a culture room at 27 ° C with a 16/8 hour (2500 ix) day/night cycle. Hypocotyl longitudinal half-sections were obtained from 10-14 day-old seedlings of 17 cotton cultivars (Table i). Eight to i0 hypocotyl sections, with 2 replicates, were placed on each of 3 initiation media. Medium iI contained the salts of MS medium with 1 mg/ L naphthalene acetic acid (NAA), i mg/L kinetin, 40 mg/L adenine, and 0.6% agar, pH 5.8 (Rani and Bhojwani 1976). Medium 2I comprised MS salts with 2 mg/L indole acetic acid (IAA), 1 mg/L kinetin, and 0.6% agar, pH 5.7 (Smith et al 1977), and Medium 3I contained the salts of LS medium (Linsmaier and Skoog 1965) with 2 mg/L NAA, 1 mg/L kinetin and 0.85% agar (pH 5.7) (Davidonis and Hamilton 1983). Each medium contained glucose (3%) as the carbohydrate source. Explants were cultured on initiation media without subculture for 30 days, at 27 ° C with 16/8 hour (2500 ix) day/night cycle. At the end of the callus initiation period, individual calli were weighed and transferred to one of three maintenance media. Medium IM contained MS salts with I0 mg/L N6"(isopentenyl)-adenine (2iP) and 1 mg/L NAA (Price et al 1977), Medium 2M contained MS salts with 1 mg/L NAA, 0.5 mg/L kinetin, and i0 mM glutamine, and Medium 3M contained LS salts (lacking NH4NO 3, but containing 3.8 g/L KNO3), with i mg/L NAA and 0.5 mg/L kinetin (Davidonis and Hamilton 1983). After 30 days on maintenance media calli were visually scored for appearance, growth and general vigor and then transferred to media containing MS salts, 2 mg/L NAA, 1 mg/L kinetin, 3% glucose, and 0.3% Gel-rite (pH 5.8). Transfers were made at 7-10
* The senior author is presently a Research Geneticist, USDA-ARS, and Assistant Professor Present address: Department of Genetics and Agronomy, Iowa State University, Ames IA 50011, USA
Offprint requests to: R. C. Shoemaker at his present address
179 Table i.
Responses of different cotton cultivars callus mass (wet weight) was determined
to callus initiation. The after four weeks in culture.
MEAN CALLUS MASS a (mg) ±S.D. Cultivar Acala 1517 Coker 201 Coker 208 Coker 310 Coker 315 Delcott 311 Deltapine 61 DES 56 GSA 71 Lankart 57 McNair 235 MoDel Paymaster 145 Quapaw RCI0-3 8troman 254 Tamcot CAMD-E
no induction 47.50-+ 14.95 41.63± 22.71 26.49± 7.82 41.78-+ 9.50 184.43± 79.32 50.29± 21.61 56.87± 26.62 35.45± 18.00 56.25± 28.40 15.52± 8.17 98.53± 49.32 42.22± 19.85 30.90± 21.63 19.08i 11.49 45.21± 20.09 38.78± 16.00
no induction 42.08-+ 18.23 40.90± 17.96 35.53± 19.79 49.00-+ 21.45 216.55±152.00(R) 80.26± 26.28 51.77± 16.44 67.19± 35.24 82.69± 34.22(R) 21.40-+ 12.27 144.07± 81.40 78.72± 42.22 51.90± 38.40(S) 36.50± 23.15 88.28-+ 24.90 60.65± 26.59
3I no induction 55.25-+ 28.75 48.82± 9.58 29.64± 9.43 36.58± 9.16 57.08-+ 27.61 56.96± 17.68 .... 25.13± 12.27 77.13-+ 26.93 22.40± 10.36 138.44± 53.27 76.72± 22.74 37.98± 24.74 36.42± 19.78 54.82± 22.22 49.40± 23.83(R)
R, root development; S, leaf or stem development The explants were equal In slze and their initial weights comprised 6.2 -+2.0 mg (for this estimate N=29). a
day intervals for 3-4 transfers. After this series of transfers, call± were placed onto media containing 3% sucrose instead of 3% glucose. Call± were observed weekly for signs of embryogenic potential. Embryogenic callus was subcultured onto the same media for increase and maintenance. Histology: Embryos and embryogenic callus Were fixed in formalin:acetic acid:ethanol:water (5:5:63:27) for 24 hours at room temperature. Fixed tissues were dehydrated in an ethanol/xylene series (Constabel 1982) and were embedded in paraffin (Paraplast TM 56-57 ° C). Ribbon sections of 8 um thickness were dried onto slides, dewaxed in xylene and rehydrated in a d e s c e n ding ethanol series. Sections were then triplestained in safranin 0 and aniline blue-orange G (McDaniel et al 1982). Embryo Germination and Plant Regeneration: Mature embryos were selected for germination on the basis of normal morphology. Germination was achieved by transferring to auxin-free MS media with 1 mg/L kinetin and 1.5% sucrose. This and subsequent manipulations were carried out at 27 ° with a 16/8 hour (25001x) day/night cycle. Once roots and leaves had developed the plantlets were transferred to Magenta boxes containing hormone-free 50% strength MS media with 1.5% sucrose (Fig IC). After i-2 weeks the plantlets were transferred into sterile vermiculite in a humid chamber, were gradually hardened-off, and were transferred to soil under normal greenhouse conditions (Fig ID). Results Callus Initiation and Maintenance: A wide range of callus initiation and proliferation r e s p o n s e s w e r e observed among the hypocotyl explants of all 17 cultivars (Table i). Acala 1517 represented one extreme in which no callus growth was apparent on any of the media that were tested. For the remaining 16 cultivars, cell proliferation was clearly visible under the light microscope within 3-5 days. On average, Medium 2I produced approximately 30% more callus mass than either Medium iI or 3I, although this figure varies widely from genotype to genotype. The texture of the call± ranged from very hard and compact to watery and friable. Some degree of
browning was observed in all cases. Regular and extensive root proliferation was observed in several cultivars (Table i). We believe that the single isolated instance of shoot proliferation observed with cultivar Quapaw resulted from shoot emergence from meristematic regions of the explant rather than as a result of shoot organogenesis from callus. Call± from each cultivar/initiation media combination were subcultured onto each of the three maintenance media. The morphology of the callus grown on each of the maintenance media varied widely, with no obvious callus 'type' correlated to any specific media. Because callus types varied widely, even within a single plate, no attempt has been made to describe each callus type found on each combination of media. However, it was possible to rate the general growth and vigor of each group. Call± placed on Medium 3M generally showed the least vigorous growth. Medium IM (2iP as the cytokinin) produced call± with an exceptionally vigorous and healthy appearance more often than either of the other maintenance media. Medium 2M produced consistently 'average' growth and vigor, with very little apparent effect from genotype. Somatic Embryogenesis: The embryogenic response was restricted to cultivars Coker 201 and 315. Embryogenie callus was first observed as small sectors of a pale grey, compact, densely cytoplasmic callus emerging from a soft, yellowish to brown, friable callus. Emergence of embryogenic callus could not be correlated to any specific induction/maintenance media combinations. Repeated experiments showed that subculturing onto maintenance media was not necessary for the induction of embryogenic callus. Therefore, 'frequencies of embryogenicity were determined from callus initiated directly on MS medium with 3% glucose, 2 mg/L NAA and 1 mg/L kinetin and switched to 3% sucrose medium, with no intervening maintenance medium. Approximately 73% of Coker 201 and 64% of Coker 315 seed tested produced embryogenic callus (Table 2). When subcultured onto media containing MS salts, 2 mg/L NAA, 1 mg/L kinetin, and 3% sucrose, embryogenic callus continued to proliferate and produce somatic embryos. Figure IA shows an example of typical embryogenic callus containing globular and heart-shaped embryos. These gradually develop into 'tulip-shaped' and mature embryos (several per plate)
180 Table 2. Frequencies of induction of embryogenic callus from responsive cotton cultivars: Coker 201 and Coker 315. Coker 201
Number of seed tested
Percent seed giving embryogenic callus
Number of explants per seedling Percent embryogenic explants from embryogenic seed
(Fig. IB) without need for transfer to other media. Histology: Histological examination shows a distinctive difference between embryogenic callus and nonembryogenic callus. Embryogenic callus appears as a tightly compact, densely cytoplasmic mass, whereas nonembryogenic callus is comprised of a loose, friable mass of cells that do not exhibit a strong cytoplasm staining reaction (Fig. 2A). This indicates that cells of nonembryogenic callus may be highly vacuolate. Somatic embryos first appear as proembryo-like globular structures (Fig 2B) possessing distinctive non-filliform, multicellular suspensors (Fig. 2C). These globular structures gradually elongate and begin to show signs of cotyledonary development (Fig. 2D). They go through a typical 'heart-stage', or more frequently, a 'cup-stage ~ (Fig. 2E), and finally develop into a 'mature' cotyledon-stage embryo (Fig. 2F). Full embryonic axis elongation occurs only after the cotyledons have reached the mature stage.
Fig. 2 A-F. Histological longitudinal sections of developmental stages observed during somatic embryogenesis. A: embryogenic callus containing an area of high meristematic activity (X70); B: globular proembryo (X200); C: globular pro-embryo showing nonfilliform suspensor development (X200); D: late globular-stage embryo (X90); E: heart-stage embryo showing early cotyledon development (X75); F: tulipstage embryo showing well-developed vascular traces and elongated cotyledons (X75). Embryo Germination and Plant Regeneration: Upon stringent selection and subculturing, calli could be expected to produce 5-6 mature embryos per plate per transfer. Less than 40% of mature-appearing embryos underwent normal germination, producing both roots and leaves. However, once roots and leaves had been established, transfer to greenhouse conditions was highly successful. Discussion
Fig. 1 A - D . Developmental progression of plantlet regeneration through somatic embryogenesis. A: embryogenic callus with globular and early heart-stage embryos (X30); B: tulip-stage and mature embryos (X33); C: regenerated plantlet growing under photoautotrophic conditions in a Magenta Box; D: cotton plants regenerated through somatic embryogenesis.
The effectiveness of the various callus initiation media for each of the cotton cultivars tested suggests that optimal cultivar/medium combinations do exist. It is probable that the wide range of callus initiation, proliferation, and maintenance responses observed in this study reflects the degree of genotypic diversity found within the cotton germplasm collection (Lee 1984). Genotype has been shown to play an important role in callus induction and plant regeneration in several species (Bingham et al 1975; Sears and Deckard 1982; Oelck and Schieder 1983). It has even been suggested that genetic improvement may prove more useful than manipulation of environmental variables in the establishment and optimization of culturing strategies (Lazar, Schaeffer and Baenziger 1984). Plant regeneration from callus-derived somatic embryoids has been previously reported from cotton cv. Coker 310 (Davidonis and Hamilton 1983). However, these were reported to have developed only after two
181 years in culture. This unusually prolonged culture period is prohibitive for crop improvement techniques involving tissue culture. In contrast, the regeneration protocol reported here is simple, straightforward and rapid, and therefore conducive to most cotton improvement programs requiring plant regeneration from tissue culture. Embryogenic callus can be obtained within 4-6 weeks of plating the initial explant, and can be found directly on the initiation media containing 3% glucose. Without exception however, embryogenicity has been lost in embryogenic callus subcultured on media containing 3% glucose, but only occasionally lost in calli subcultured on media containing 3% sucrose. We have no experimental evidence to explain the apparent selective advantage of sucrose for embryogenic callus of cotton. However, the physiological importance of sucrose in the process of differentiation in callus culture has been reported before (Wetmore and Rier 1963; Jeffs and Northcote 1966; Jells and Northcote 1967). Although the specific role of sucrose in cellular differentiation remains obscure, it has been suggested that the inductive action is a specific function of the intact disaccharide molecule (Jeffs and Northcote 1967) and that the mechanism of action may be comparable to that of inducer and repressor systems (Jeffs and Northcote 1966). We were unable to find conclusive histological evidence for the origin of somatic embryos from single-cell initials from embryogenic callus. However, the existence of non-filliform, multicellular suspensor-like structures suggests that embryos have arisen from single-cells (Haccius 1970). The developmental progression of our somatic embryos, morphologically follows the progression seen in cotton zygotic embryos (Pundir 1972) with the exception that cotyledons of somatic embryos are significantly underdeveloped relative to zygotic embryos; a not uncommon situation in vitro (Crouch 1982). Acknowledgments We thank Glen Drohman for excellent technical assistance in the greenhouse. This work was sponsored by Agrigenetics Research Associates. References Bingham ET, Hurley LV, Kaatz DM, Saunders JW (1975) Breeding alfalfa which regenerates from callus tissue in culture. Crop Sci 15:719-721. Constabel F (1982) Histological methods. In: Wetter LR, Constabel F (eds) Plant tissue culture methods. National Research Council of Canada, Saskatoon, Saskatchewan, pp 34-37. Crouch ML (1982) Non-zygotic embryos of Brassica napus L. contain embryo-specific storage proteins. Planta 156:520-524.
Davidonis GH, Hamilton RH (1983) Plant regeneration from callus tissues of Gossypium hirsutum L. Plant Sci Lett 32:89-93. Haccius B (1978) Questions of unicellular origin of non-zygotic embryos in callus cultures. Phytomorphology 28:74-81. Jeffs RA, Northcote DH (1966) Experimental induction of vascular tissue in an undifferentiated plant callus. Biochem J 101:146-152. Jeffs RA, Northcote DH (1967) The influence of indole-3yl acetic acid and sugar on the pattern of induced differentiation in plant tissue culture. J Cell Sci 2:77-88. Lazar MD, Schaeffer GW, Baenziger PS (1984) Cultivar and cultivar X environment effects on the development of callus and polyploid plants from anther cultures of wheat. Theor Appl Genet 67:273-277. Lee JA (1984) Cotton as a world crop. In: Kohel RJ, Lewis CF (eds) Cotton. Agronomy Monograph 24. American Society of Agronomy, Inc. Madison, WI pp 1-25. Linsmaier FM, Skoog F (1965) Organic growth factor requirements of tobacco tissue culture. Physiol Plant 18:100-127. McDaniel JK, Conger BV, Graham ET (1982) A histological study of tissue proliferation, embryogenesis, and organogenesis from tissue cultures of Dactylis glomerata L. Protoplasma 110:121-128. Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco cultures. Physiol Plant 15:473-497. Oelck MM, Schieder O (1983) Genotypic differences in some legume species affecting the redifferentiation ability from callus to plants. Z Pflanzenzuchtg 91:312-321. Price JH, Smith RH, Grumbles RM (1977) Callus cultures of six species of cotton (Gossypium L.) on defined media. Plant Sci Lett 10:115-119. Pundir NS (1972) Experimental embryology of Gossypium arboreum L. and G. hirsutum L. and their reciprocal crosses. Bot Gaz 133:7-26. Rani A, Bhojwani SS (1976) Establishment of tissue cultures of cotton. Plant Sci Lett 7:163-169. Sears RG, Deckard EL (1982) Tissue culture variability in wheat: callus induction and plant regeneration. Crop Sci 22:546-556. Smith RH, Price J, Thaxton JB (1977) Defined conditions for the initiation and growth of cotton callus in vitro I. Gossypium arboreum. In Vitro 13:329-334. Wetmore RH, Rier JP (1963) Experimental induction of vascular tissues in callus of angiosperms. Amer J Bot 50:418-430.