In Vitro Cell. Dev. Biol.ÐPlant 37:462±465, July±August 2001 DOI:10.

1079/IVP2001202
q 2001 Society for In Vitro Biology
1054-5476/01 $10.0010.00

SYNCHRONIZED SOMATIC EMBRYO DEVELOPMENT IN EMBRYOGENIC SUSPENSIONS OF

FRAXINUS ANGUSTIFOLIA

G. TONON*, G. BERARDI, C. ROSSI, and U. BAGNARESI

Dipartimento di Colture Arboree, UniversitaÁ di Bologna, Via F. Re 6, 40126 Bologna, Italy

(Received 2 October 2000; accepted 29 March 2001; editor D. Ellis)

Summary
The synchronization of somatic embryo development in embryogenic suspension cultures is a crucial step in taking
advantage of somatic embryogenesis for high production potential and reduction of unit cost through automation. In the
present study, a synchronous somatic embryogenic system was developed for Fraxinus angustifolia suspension cultures.
High cell density, 6-benzyladenine (BA) and 2,4-dichlorophenoxyacetic acid proved essential for the establishment and
maintenance of suspension cultures. Low cell density, BA and 1-naphthaleneacetic acid enhanced somatic embryo
development. Cell and cell cluster fractionation by density gradient centrifugation in Ficoll solution proved useful for
separation of subpopulations with differing potentials for embryo development. A synchronous development of somatic
embryos at high frequency was achieved only from the heaviest cell population.
Key words: synchronization; Ficoll; somatic embryogenesis; ash.

Introduction by low efficiency, and its application on an industrial scale is still

far from practicable.
Fraxinus angustifolia is a species of increasing interest in the Therefore, the aim of the present study was to establish a
Mediterranean Basin for both quality wood production and urban synchronous F. angustifolia embryogenic system in liquid culture
landscaping; finding ways to improve it is therefore desirable. with increased efficiency.
Somatic embryogenesis represents an important tool in breeding
tree species because of its great production potential, the possibility
of cryo-conserving embryogenic cell lines and the reduction in unit
costs through process automation (Thorpe et al., 1995). Of Materials and Methods
particular interest is the integrated use of somatic embryogenesis
Plant material. Embryogenic calluses were obtained from immature
and cryo-conservation, which should make it possible to overcome embryos extracted from seeds of a selected tree of F. angustifolia (Tonon
such limiting factors as reduced rooting ability, which is correlated et al., unpublished data). Before suspension establishment, calluses were
with mother-plant age and adversely affects the improvement on a subcultured monthly for about a year on the basal medium solidified with
clonal basis of many woody species (Park et al., 1998). Indeed, 3 g l21 gellan gum (Gelrite, Duchefa) and supplemented with 4.4 mM 6-
through cryo-preservation, the clones obtained by somatic embryo- benzyladenine (BA) and 0.4 mM 2,4-dichlorophenoxyacetic acid (2,4-D).
The basal medium was half-strength MS salt formulation (Murashige and
genesis from young reacting tissues (e.g. mature and immature Skoog, 1962) supplemented with 20 g l21 sucrose; pH was adjusted to 5.8
zygotic embryos) can be stored until field test results are available, with KOH or HCl before autoclaving at 1208C for 20 min.
and only superior clones can be subsequently propagated (Cheliak Culture conditions and experimental procedures. All suspension cultures
and Rogers, 1990). were performed in Erlenmeyer flasks (250 ml), sealed with cotton plugs and
a double layer of aluminum foil and kept on a gyratory shaker at 90 rpm,
A high-frequency, synchronous embryogenic system in liquid 20 ^ 28C; under 1.5 W m22 photosynthetically active radiation (PAR),
culture is needed to take full advantage of somatic embryogenesis provided by 80 W cool-white fluorescent lamps, with a 16-h photoperiod. At
(Novak, 1992), as it is essential for automation and for investigating least three flasks per treatment were used for each experiment. Experiments
physiological, biochemical and molecular aspects of a process for were replicated twice, except that of embryo conversion. All quantitative
which there is still limited information concerning woody tree data are reported as mean value^standard deviation.
Establishment and maintenance of suspension cultures. Samples of
species. approximately 2 or 4 g (fresh weight) of embyogenic callus were suspended
In a previous study (Tonon et al., unpublished data), a somatic in 35 or 70 ml of the basal liquid medium supplemented with 2.2 mM BA
embryogenesis protocol on semi-solid medium, from induction to and 0.4 mM 2,4-D (maintenance medium). After 1 wk, only 35 ml
embryo conversion, was established. This system was still suspensions were diluted by adding an equal amount of the same fresh
medium. At day 14, all suspensions were filtered (stainless steel sieve, 2 mm
characterized by a high asynchronism in embryo development and pore size) to discard the largest aggregates. After centrifugation at 400  g
for 10 min, the packed cell volume (PCV) was adjusted to 2.5% (v/v) with
fresh medium. Cultures were subcultured every 2 wk by collecting the solid
*Author to whom correspondence should be addressed: Email gtonon@ fraction on a nylon sieve (160 mm pore size) and keeping the initial cell
agrsci.unibo.it inoculum at 2.5% (w/v). At the end of each subculture, the fresh weight of

462
 SUSPENSION CULTURES OF ASH 463

Fig. 1. A, PEMs after several subcultures in the maintenance medium …bar ˆ 1 mm†: B, Cell suspension with prevalence of embryo
globular stage after three subcultures (6 wk) in the PGR-free medium …bar ˆ 1 mm; arrow=green torpedo embryo). C, Somatic embryo
with fused cotyledons …bar ˆ 1 mm†: D, Asynchronous somatic embryo development from 500 to 1000 mm size fraction after five
subcultures in development medium …bar ˆ 1 cm†: E, Large yellowish aggregates from 14 to 16% Ficoll density gradient centrifugation
fraction after 4 wk of culture in the development medium …bar ˆ 1 cm†: F, Mature somatic embryos from 18% Ficoll density gradient
centrifugation fraction after 4 wk of culture in the development medium …bar ˆ 1 cm†:
464 TONON ET AL.

the collected solid fraction was recorded in order to check suspension growth TABLE 1
rate (final/initial fresh weight of cells and cell clusters).
Development of somatic embryos in liquid cultures. Suspension cultures, NUMBER OF MATURE SOMATIC EMBRYOS PER FLASK IN
grown for several months in the maintenance medium, were used for somatic DIFFERENT DIMENSIONAL FRACTIONS OF EMBRYOGENIC
embryo (SE) development in liquid culture. Two treatments, i.e. either a total FRAXINUS ANGUSTIFOLIA SUSPENSION CULTURES AFTER 10 WK
absence of plant growth regulators (PGR) or a combination of 2.2 mM BA OF CULTURE (FIVE SUBCULTURES) IN THE DEVELOPMENT
and 0.5 mM 1-naphthaleneacetic acid (NAA), were tested. The initial cell MEDIUM
inoculum was adjusted to 1% (w/v) for both treatments.
Synchronization of somatic embryo development. In order to synchronize Dimensional fraction (mM) Somatic embryo numbera
SE development in liquid medium and improve process efficiency, two 160±500 25.7 ^ 5.1
procedures were tested. In the first, cells and cell clusters from several 500±1000 23.7 ^ 0.6
culture cycles on the liquid maintenance medium were separated into four 1000±2000 26.7 ^ 1.5
size classes by four successive sievings through pore size: 160±500 mm; .2000 18.7 ^ 3.6
500±1000 mm; 1000±2000 mm; .2000 mm. Aggregates of each class size
were grown separately for 10 wk (five biweekly subcultures) in 50 ml of a
basal medium supplemented with 0.5 mM NAA and 2.2 mM BA (develop- Mean value of three samples^SD.
ment medium). Initial cell inoculum was adjusted to 1% (w/v). SE
development was periodically monitored and the number of completely
culture conditions SE development occurred in a completely
formed SEs was recorded after 10 wk, at the end of the fifth subculture. asynchronous manner and at low frequency.
In the second procedure, the fraction ranging from 160 to 1000 mm was Synchronization of embryo development. Fractionation of cells
taken from cells subcultured for several cycles on the liquid maintenance and cell clusters on a size basis, followed by several subcultures at
medium, successively raised for three subsequent cycles on the PGR-free low cell inoculum in the development medium, initially yielded
basal medium and separated by density gradient centrifugation in Ficoll
solution. Cells and cell clusters (2.5 g FW) were layered on 36 ml of Ficoll homogeneous cultures, but subsequently the SEs started to mature
discontinuous density gradient in a 50-ml tube and then centrifuged at asynchronously at the end of the third subculture regardless of size
150  g for 5 min. The density gradient was formed by five Ficoll classes of the aggregates. Moreover, the SEs were few and often
concentrations incrementally increasing from 10% to 18% by 2%. All marked by an abnormal shape, i.e. cotyledon fusion or lacking
Ficoll solutions contained 2% sucrose. After centrifugation, each cell
fraction was re-suspended in the basal medium, centrifuged at 100  g for
cotyledons (Fig. 1C), and tended to green precociously (Fig. 1D;
30 s to remove Ficoll residues and re-suspended again in 50 ml Table 1). No significant difference was observed among fractions
development medium with 1% (w/v) as initial cell inoculum. The medium with regard to growth rate during the course of the subculture
was replaced after 2 wk without further filtration. The number of SEs formed sequence (data not reported).
from each fraction was recorded after 2 and 4 wk of culture on the The fractionation by density gradient centrifugation in Ficoll
development medium.
Preliminary embryo conversion test. Forty-eight SEs of the 18% Ficoll solutions of cells and cell clusters previously grown for three cycles
fraction that were beginning to turn green and had elongated radicles were in the PGR-free medium enabled the cell population to be divided
removed from liquid culture and transferred to PGR-free solidified basal into two subpopulations in terms of SE development capacity.
medium dispensed into 25  150 mm culture tubes that were sealed with Indeed, while no embryos were formed from cells of the lightest
transparent plastic caps. The cultures were kept at the standard culture
conditions except for PAR at 2.0 W m22. Germination percentage was
three fractions, a high frequency of embryo formation was observed
recorded after 40 d. in the two heaviest fractions, especially in the fraction collected in
the 18% Ficoll layer (Table 2). After 4 wk of culture, `non-embryo
forming' fractions were distinguished by large yellowish cell
Results clusters with occasional abnormal somatic embryo structures inside
(Fig. 1E). By contrast, the heaviest fractions were composed of
Establishment and maintenance of suspension cultures. Suspen-
small brown cells and cell clusters that began to differentiate
sion cultures were successfully established when 4 g of callus were
normally shaped embryos after 2 wk, and after 4 wk only mature
inoculated in 35 ml of liquid medium and then diluted with another
SEs with elongated radicles were visible in culture (Fig. 1F). Even
35 ml after 1 wk. In these conditions, after the first filtration,
in these conditions, however, cells did not start differentiating
cultures proliferated without forming overly large aggregates, and
embryos simultaneously, some mature stages being already visible
embryo development was mostly blocked at the proembryogenic
masses (PEM) stage (Fig. 1A). By contrast, a lower cell inoculum
(2 g per 35 ml, 4 g per 70 ml, or 2 g per 70 ml) at the beginning of TABLE 2
suspension cultures induced the formation of larger aggregates and
NUMBER OF MATURE FRAXINUS ANGUSTIFOLIA SOMATIC
the sporadic development of later SE stages. The adopted EMBRYOS PER FLASK IN FRACTIONS OBTAINED BY FICOLL
maintenance medium and cell inoculum (2.5% w/v) proved to be DENSITY GRADIENT CENTRIFUGATION (EMBRYOS WERE
suitable for the maintenance of culture at the PEM stage, at least for COUNTED AFTER 2 AND 4 WK OF CULTURE IN LIQUID
the 6 mo. tested, with a stable growth rate ranging from 3.0 to 3.5. DEVELOPMENT MEDIUM)
Development of somatic embryos in liquid cultures. The lack of Somatic embryo number after
PGRs in the medium associated with low cell inoculum enabled
Fraction (Ficoll concentration %) 2 wka 4 wka
PEMs to reach the globular embryo stage. The latter prevailed on
PEM stage after two subcultures (Fig. 1B). Nevertheless, further 10±12 0 0
subcultures in these conditions seldom resulted in SE mature 12±14 0 0
14±16 0 0
stages. By contrast, low cell inoculum associated with the use of 16±18 27.7 ^ 6.9 162.3 ^ 20.0
NAA and BA promoted embryo development as evinced by the .18 43.0 ^ 6.2 237.6 ^ 27.4
presence of several SEs completely formed after 6 wk of culture (at
a
the end of the third subculture). However, even in these latter Mean value of three samples^SD.
 SUSPENSION CULTURES OF ASH 465

after one subculture (Table 2). Nevertheless, at the end of the fourth by cell fractionation through Ficoll density gradient centrifugation
week the homogeneity of mature embryos and their frequency were and by manipulation of PGR and cell density in the medium. Given
satisfactory. the high regeneration rate, the present study offers hope for large-
Embryo conversion into plants. Of the mature somatic embryos, scale propagation of F. angustifolia by suspension cultures and
39.6% were converted into normal plantlets, 16.7% showed establishes the steps for somatic embryo production in bioreactors.
abnormal germination and 43.7% did not develop at all (Table 2). That two cell fractions with opposite potential in terms of somatic
embryo development and differentiation can be distinguished
Discussion should prove useful for biochemical, physiological and molecular
studies of embryogenesis processes in woody plants.
As reported in many other woody (Merkle et al., 1993) and
herbaceous plant systems, F. angustifolia somatic embryo devel- Acknowledgment
opment was inhibited by 2,4-D and by high cell inoculum in the
liquid medium. On the other hand, these conditions proved to be The authors thank Dr. Grazia Marino for a critical reading of the
suitable for culture maintenance at the PEM stage by preventing SE manuscript.
maturation. The addition of NAA and BA appears to be essential for
reaching later stages of SE development, the lack of PGRs alone References
being insufficient. The absence of PGRs, however, was suitable to
Cheliak, W. M.; Rogers, D. L. Integrating biotechnology into tree
reach and maintain SE globular stage. A similar response was improvement programs. Can. J. For. Res. 20:452±463; 1990.
observed by Coutos-Thevenot et al. (1992) and Jayasankar et al. Coutos-Thevenot, P.; Gloebel-Tourand, I.; Mauro, M. C.; Jounneau, J. P.;
(1999) in some grape cultivars, where the lack of 2,4-D was Boulag, M.; Deloire, A.; Guern, J. Somatic embryogenesis in
ineffective in promoting SE development beyond the heart stage. grapevine cells. Improvement of embryo development by changes
The separation, through discontinuous density gradient centrifu- in culture conditions. Plant Cell Tiss. Organ Cult. 29:125±133;
1992.
gation in Ficoll solutions, of the cell population into two
Jayasankar, S.; Gray, D. J.; Litz, R. E. High-efficiency somatic embryogen-
subpopulations with opposite SE development capacity supports esis and plant regeneration from suspension cultures of grapevine.
the findings of Osuga and Komamine (1994) in Daucus carota and, Plant Cell Rep. 18:533±537; 1999.
in part, those of Jayasankar et al. (1999) in Vitis vinifera, where only Litz, R. E.; Moon, P. A.; Mathews, H.; Jayasankar, S.; Monsalud, M. J.;
the heaviest and cytoplasm-rich cells proved suitable for embryo Pliego-Alfaro, F. Somatic embryogenesis in mango. In: Gupta, P. K.;
Jain, S. M.; Newton, R. J., eds. Somatic embryogenesis in woody
development. The browning of these embryogenic cells, in contrast
perennials, vol. 2. Dordrecht: Kluwer Academic; 1995:341±356.
to the yellowish color of non-embryogenic ones, has been reported Merkle, S. A.; Hoey, M. T.; Watson-Pauley, B. A.; Schlarban, S. E.
also in Mangifera indica (Litz et al., 1995), Elaeis guineensis Propagation of Liriodendron hybrids via somatic embryogenesis.
(Teixeira et al., 1995) and Vitis vinifera (Jayasankar et al., 1999) as Plant Cell Tiss. Organ Cult. 34:191±198; 1993.
related to high amounts of phenolic compounds. Murashige, T.; Skoog, F. A revised medium for rapid growth and bioassay
On the other hand, the different response of the two fractionation with tobacco tissue cultures. Physiol. Plant. 15:473±497; 1962.
Novak, F. J. Musa (bananas and plantains). In: Hammerschlag, F. A.; Litz,
methods in terms of SE yield and culture synchronization might also
R. E., eds. Biotechnology of perennial fruit crops. Wallingford, UK:
be due to a beneficial effect of culture in the PGR-free medium that CAB International; 1992:449±458.
was adopted only prior to fractionation via discontinuous density Osuga, K.; Komamine, A. Synchronisation of somatic embryogenesis from
gradient centrifugation and not for separation on size basis. Indeed, carrot cells at high frequency as a basic for the mass production of
while the fractionation via density gradient centrifugation was embryos. Plant Cell Tiss. Organ Cult. 39:125±135; 1994.
performed when the suspension was primarily at the globular Park, Y. S.; Barrett, J. D.; Bonga, J. M. Application of somatic
embryogenesis in high-value clonal forestry ± deployment,
embryo stage, the separation on size basis was done when the genetic-control, and stability of cryopreserved clones. In Vitro
suspension was still at the PEM stage. Cell. Dev. Biol. Plant 34:231±239; 1998.
The percentage of conversion into normal plants was quite low Teixeira, J. B.; Sondahl, M. R.; Nakamura, T.; Kirby, E. G. Establishment of
and further studies are needed to improve this step of the protocol. oil palm suspensions and plant regeneration. Plant Cell Tiss. Organ
Moreover, ex vitro acclimization should be carried out for industrial Cult. 40:105±111; 1995.
Thorpe, T. A.; Harry, I. S.; Kumar, P. K. Application of micropropagation to
application. forestry. In: Debergh, P. C.; Zimmerman, R. H., eds. Micropropaga-
The results of the present study show that F. angustifolia SEs can tion. Technology and application. Dordrecht: Kluwer Academic
be produced successfully in suspension culture with high efficiency Publishers; 1995:311±336.