Plant Cell Reports

Plant Cell Reports (1994) 14:87-93

c~, Springer-Verlag 1994

In vitro regeneration of the tropical multipurpose leguminous tree Sesbania Grandiflora from cotyledon explants Catherine Detrez, Sai~r Ndiaye, and Bernard Dreyfus Laboratoire de Microbiologie des sols, [nstitut Franqais de Recherche Scientifique pour le D6veloppement en Coop+ration, ORSTOM, BP 1386, Dakar, S6n~gal Received 31 May 1993/Revised version received 8 June 1994 - Communicated by G.C. Phillips

S u m m a r y . A system using cotyledon pieces as explants and a BAP/NAA containing medium was developed for in vitro mass propagation of Sesbania grandiflora, a tropical nitrogen-fixing leguminous tree. The age and the lighting conditions of seedlings providing the explants were shown to be critical factors for both bud induction and bud elongation. Optimal choice for an efficient and reproducible bud induction process consisted of dark-grown seedlings, 24/36 h-old-post-imbibition, that yielded up to 96% of explants producing more than 30 buds each, after one week in culture. Bud development occurred throughout a direct organogenesis pathway, from the proximal and adaxial cut surface of the explants as proved by histological studies. Additional sites of regeneration were also obtained after wounding on the epidermal surface of explants, suggesting a large distribution of regenerative cells all along the explants. Bud elongation, i.e. stem differentiation and leaf growth, was improved by bud isolation from cotyledon explants and their further subculture in liquid bud elongation media for one week. Rooting was obtained on an auxin medium after 3 weeks and plants were established in soil with 92% success.

Abbreviations. BAP: 6-BenzyIAminoPurine; NAA: a-Naphthalene Acetic Acid; IBA: Indole-3-Butyrie Acid; 1AA: Indole-3-Acetic Acid; GA3: Gibberellic Acid: MS: Murashige and Skoog (1962) medium

Introduction Sesbania grandiflora (2n=24) belongs to the subfamily Papilionoideae of the Leguminosae and is placed in the botanical tribe Robinieae (Evans 1990). It is a small tree of the Sesbania family which includes more than 50 species wide-spread in tropical and subtropical areas; 13 species are recognized as woody biennials or perennials. To take advantage of its vigorous early growth, often Correspondence to: C. Detrez

reaching 8-10 m in height within 2-3 years, S. grandiflora is commonly managed for fodder production purposes as an annual crop. Because of its nitrogen fixing potential due to nodulation by Rhizobium (Harris et al. 1949; Ndoye et al. 1990), it is also planted as green manure or as a component in alley cropping systems to improve or to restore soil fertility in tropical and subtropical farming. Sesbania grandiflora is also used as an ornamental and a source of human food, firewood, gum, tannins and paper (Von Carlowitz 1990). Because of its multiple uses, S. grandiflora is widely recommended in social forestry programs in India (Shanker and Mohan Ram 1990). It has also recently received increased attention for the enhancement of the productivity, diversity and sustainability of ecosystems in Africa. Despite the ecological and agricultural importance of such a multipurpose tree, there has been little assessment of its genetic variability. The germplasm collections and the genetic improvement of Sesbania perennials have not been well developed (Brewbaker 1990). In this context, micropropagation of tree species could offer a rapid means of producing clones for afforestation, breeding systems and for conservation of elite or rare genotypes. Gene transfer technologies through tissue culture also offers ways of introducing desirable characters in plant genomes from cells which express cellular totipotency (for review see Draper et al. 1988; Gasser and Fraley 1989). Although the majority of tree species have consistently proved to be recalcitrant or difficult to achieve organ formation and plant regeneration in vitro (Bonga 1987; Franclet et aL 1987), some successes have been reported recently with the nitrogen-fixing tree Allocasuarina verticillata (Phelep et al. 1991) and some of the tropical leguminous tree species (Bajaj 1986; Dhawan 1989; Jaiwal and Gulati 1991; Ranga Rat and Prasad 1991; Mathur and Mukunthakumar 1992). In the Sesbania genus, in vitro culture experiments have been focussed

88 on S. rostrata (Vlachova et al. 1987; Pellegrineschi and Tepfer 1993), S. sesban (Khattar and Mohan Ram 1982, 1983; Harris and Moore 1989; Harris and Puddephat 1989; Harris et al. 1989; Dhir and Gupta 1990; Yan-Xiu et al. 1990), S. cannabina (Xu et al. 1984), S. bispinosa (Kapoor and Gupta 1986; Sinha and Mallick 1991), S. formosa and S. exalta (Harris and Puddephat 1989). In these species, micropropagation has been sometimes obtained by enhanced axillary branching from seedling apices or from nodal cuttings of greenhouse plants. However, shoot regeneration via adventitious bud organogenesis is more widely documented and the authors have reported effective and reproducible regeneration from seedling tissues. In S. grandiflora, Shanker and Mohan Ram (1990) have identified suitable treatments to promote and sustain callus growth, then bud formation, from cotyledon and hypocotyl explants sampled on 10/12-d-old seedlings. In this paper, we show how cotyledon explants are also effective tissue to achieve direct bud organogenesis (without a callus step) and how the lighting conditions and the age of seedling are critical factors to obtain, more quickly, higher percentages of bud inducing explants than those reported by the previous authors. Histological studies were also carried out to confirm the origin and ontogenesis of the multiple shoot formation as a prerequisite for future genetic transformation procedures.

Material and methods Seed germination and explant isolation. Seeds of Sesbania grandiflora (Inland and Foreign Trading Co, Singapore) were surface sterilized for 1 h in concentrated H2SO 4 and rinsed five times with sterile distilled water. Additionally, seeds were immersed in sterile water for 3 h to promote imbibition. They were finally germinated in 90 x 20 mm Petri dishes containing half-strength MS medium, then incubat~edlfor 24-72 h at 28 + 2~ either under continuous light (99 ktmol.m'Ls of Photosynthetically Active Radiations) or in the dark. Aseptically dark- or light-grown seedlings, 24- to 72-h-old after seed imbibition, were used as the source of cotyledon explants. Unless specified otherwise, a cotyledon explant consisted of a transversal half (0.4-0.6 cm in length) of a cotyledon, taking care to excise its quiescent axillary bud. After isolation from seedlings, the explants were then transferred to Petri dishes (20 explants per dish) and plated horizontally with the dorsal (abaxial) side down on the bud induction medium.

sources of cotyledon explants (i.e. dark- or light-grown seedlings, 24-, 36-, 48-, 60- or 72-h-old post-imbibition), the percentages of explants giving buds after 3 weeks of culture on MSI were recorded. The assays were statistically analyzed as an unbalanced design with at least 3 separate replications (i.e. 3 Petri dishes) of 20 explants each. Multiple comparisons of data were conducted using the Newman-Keuis test (at the 5% level), after arcsin transformation of the percentages and standard analysis of variance (AOV) (STAT-ITCF, IBM). Bud number per responsive explant was also recorded. Bud elongation - Elongation of adventitious buds (i. e. formation of shoots including a shoot apex, an elongated stem and expanded leaves with petiole) was first observed on the MS 1 bud induction medium, without any subculture step. The percentage of explants showing bud elongation were scored for each source of explants after 6 weeks of culture on MSI, and statistically analyzed as described above (AOV and Newman-Keuls test at the 5% level). Alternatively, attempts to increase the percentage of bud elongation were performed after isolation of 3-week-old buds from the explants to obtain "propagules". Such propagules generally consisted of a group of 1-3 buds and were subeultured on a fresh bud elongation medium (MS supplemented with NAA, BAP and/or GA3), containing agar or not. The percentages of propagules exhibiting bud elongation, as shoot length, were recorded 3 weeks after subculture, among at least 50 propagules per medium. Rooting - Rooting of elongated buds was performed on three auxincontaining MS media. The percentages of rooted shoots, as the root number per shoot and root length, were recorded after 4 weeks of culture, from at least 24 shoots per medium.

Cytological investigations. Cotyledon explants were sampled at daily intervals, from the first up to the seventh day of culture on the MS1 bud induction medium. They were fixed in a F.A.A. mixture (formaldehyde/acetic acid/ethanol, 5/5/90, V/V/V) for 24 h, dehydrated in an ethanol series, cleared in xylol and then embedded in paraplast (60~ Sections, 5p.m thick, were stained with paragon or methylene blue (1% aqueous solution) and observed via light microscopy.

Results and discussion Bud induction

,0'q80.* "2_

[

%_

100 80

- 60~ 40 =

9

20"

Nll

a_._

iiiililiiiii

~

iiiiiiiiii!!

ii

6(

iiiiii!iiiiiii ....-.......

4(

:!:i:i:!:i:i ii!i!iiiiiiiiiiiiiii :i:i:i:i:i:!

..-......... .....-.....

2a( ::::::::::::: i~iiiiiiiii

0' 24h 36h 48h 60h 72h Age of light-grown seedlings

ii;iiiiiiiili

24h 36h- 48h 601a 72h Age of dark-grown seedlings

1A: Percentage of explants showing bud induction

%

% 106

Media and in vitro growth conditions of the explants. Salts and vitamins of MS were used as a basal medium. Sucrose and the growth regulators (BAP, NAA, IBA, IAA, GA3) were added before autoclaving. All the media were adjusted to pH 5.9 with KOH or HCI, then autoclaved at I10~ for 30 rain. The agar media (0.8% Difco Bacto agar, W/V) were dispensed in Petri dishes (90 x 20 mm) or m culture tubes (20 x 150 mm). The liquid media were dispensed in 250 ml Edenmeyer flasks (40 ml per flask) and agitated at 20 rpm (Schiittelmaschine RO 20, Osi). The, MS medium, sup.l{lemented with sucrose (30 g.l't), NAA (1 mg.l") and BAP (1 mg.l" ) (referred to as MS1 in the text below), was chosen as the bud induction medium because this proved most effective in preliminary attempts (unpublished data). All the explant cultures were incubated at 28 + 2~ under continuous light conditions dispensed by a mixture of Daylight and Gro- fluorescen[ t~bes (Sylvania, QUE) (5 : 1, respectively), providing 99 lamol.m . s of Photosynthetically Active Radiations (PAR) at the Petri dish level.

Data scoring Bud induction - To compare bud induction abilities of different

8O" 60

80 t 60

40

40 -I

20t

2G 0

"I 24h 36h 48h 60h 72h Age of light-grown seedlings

0-1 24h 36h 48h 60h 72h Age of dark-grown seedlings

1B: Percentage of explants showing bud elongation

Fig. 1. Percentages of half-cotyledon explants cultured on MS1 for 3 weeks in the lighL showing: (1A) bud induction, and/or (1B) bud elongation, for 10 sources of explants: dark- or light- grown Sesbania grandiflora seedlings, 24-, 36-, 48-, 60-, or 72-h-old post-imbibition Percentages recorded from 60 explants per source of explants

89 For each of the 10 sources of cotyledon explants studied q.e. sampled from dark- or light-grown seedlings, 24-, 36-, 48-, 60-, 72-h-old post-imbibition), Figure 1A lists the percentages of explants exhibiting bud induction on the MS 1 medium. These percentages range from 23 to 96%. Analysis of variance of these data, arcsin transformed, shows a very significant effect of both the lighting conditions (referred to as factor 1) and the age (factor 2) of seedlings providing explants, whereas no interaction effect was observed between factors 1 and 2 (Table 1). Concomitantly, the Newman-Keuls test designates the dark-grown, and/or 24/36-h-old seedlings, as being the more efficient sources of bud inducing explants (Table 1); more than 30 buds were moreover routinely obtained from the entire cut surface of these explants. In contrast, adventitious buds on explants from light-grown seedlings, 48/60/72-h-old, occurred exclusively at the edge of their adaxial cut surface, and no more than 3 buds were recorded per explant. No budding was observed on explants sampled from 4-d-old or older seedlings (data not shown). Such a decline in the percentage of bud inducing explants with the seedling age (occuring from 48 h post-imbibition, when emergence of cotyledons from their seed coats and their early involvement in photosynthetic functions are observed), is in agreement with data reported by Gulati and Jaiwal (1990) in Vigna. With these authors we suggest that, at such advanced stages of germination, the cotyledon's reserves (already mobilized for cell metabolism of embryo axis growth) have limited availability in terms of prime sources of energy or growth factors to support bud organogenesis. The fact that dark grown cotyledons formed more buds than those developed under light also

rules out the possibility of photosynthates being critical regarding bud induction. The improvement of bud induction by the selection of the youngest explant may also have resulted from the persistance of cells not fully determined, activated by the exogenous and endogenous hormone ratio.

Morphogenesis pathways from cotyledon explants The localization of regenerative cells and ontogeny of adventitious buds were histologically examined for one culture week, among explants isolated from 36-h-old, dark-grown seedlings. No pre-existing quiescent bud was observed in explants before plating on bud induction medium. By day 2-3 on MS1 medium, cell divisions were observed from the cut surface (Fig. 2A). By day 5-6 cell divisions resulted in the organization of a meristematic structure (Fig. 2B), which in turn differentiated a bud meristem with overlapping leaf primordia by day 7-8. Buds became macroscopic by day 7-8 and at this stage, vascular bundles connecting the adventitious buds to the vascular tissues of the explants were observed (Fig. 2C, 2D). More than 30 buds shoots were obtained after 2 weeks, but as reported below full shoot differentiation was not frequently observed (Fig. 2E, 2F). The most of buds were arrested in their development and did not express stem differentiation (and a well defined apical dominance), unless subcultured (see also the "Elongation" section). The pattern of adventitious bud induction and growth, described above, exclusively occurred at the proximal and ventral (adaxial) cut surface of cotyledon explants (Fig. 2E), from dividing cells observed at zero to 400 Jam

Table 1. Statistical analysis of the percentages of explants showing bud induction or bud elongation on the MS1 medium, according to

the lighting conditions (factor 1) and the age (factor 2) of seedlings providing the cotyledon explants of Sesbaniagrandiflora:results from the analysis of va n"ance(F-test)O) and from the Newman-Keuls test. Factor studied

FACTOR 1

(lighting conditions ofseedlings) FACTOR 2

(ageof seedlings)

INTERACTION Factor 1/Factor 2

Statistical values calculated Probability value of the F-ratio Percentages of responsive explants (2) dark light Probability value of the F-ralio Percentages of responsive explants (2) 24hrs 36 hrs 48 hrs 60hrs 72 hrs Probability value of the F-ralio Coefficiant of variation

Morphogenesis step studied

Bud induction