Home | Profile | Achievement | Programmes | Projects | Staffs | Publications | Journals |
Biotech Glossary | Bioinformatics | Lab Protocol | Notes | Malaysia University |

Recent Advances in the Tissue Culture of Cocoa from Somatic Embryos to Bentwood Gardens - a Short Review
Mark j Guilitinan and Siela Maximova
ACRI Molecular Biology Laboratory, Tyson Building, The Pennsylvania State University, University Park, PA, 16802 (,


An efficient in-vitro clonal propagation method for cocoa has been developed (somatic embryogenesis, SE) through the efforts of an international group of cocoa researchers. The intent of this manuscript is not to cover all aspects of the field or its history, but to review recent results. This review includes information from papers presented at the 3~ INGENIC Workshop and the l3~ International Cocoa Research Conference, both of which were held in Kota Kinabalu, Malaysia in October 2000. Recent progress has shown that while primary somatic embryos arise from a multi-cellular pathway, secondary somatic embryos arise predominantly from single cells, Secondary somatic embryos also exhibit a higher quality and conversion rate compared to primary somatic embryos as well as a higher production rate for mast genotypes. Variations in hormonal and other treatments were reported, and significant differences in genotypic response observed, It was estimated that about 100 different genotypes have been successfully propagated via somatic embryogenesis, in approximately nine laboratories worldwide, Several reports discussed the development of low cost, SE-coupled downstream propagation systems, including mini-cuttings and bentwood gardens, capable of multiplying the SE plants and producing orthotropic rooted cuttings. These systems hold great potential for research and production, however full validation of such plants awaits the results of long-term field-testing now underway.


Since the early work of cocoa researchers, vegetative propagation systems have been an important tool, enabling the multiplication of wild or selected genotypes for germplasm collection, distribution and for production of clonal materials for replicate performance trials. While traditional rooted cutting and grafting systems have been used throughout the world for propagation of cocoa, to date, a vast majority of production stock is grown from seed. Due to the high heterozygosity of most cocoa genotypes, this results in a large degree of yield and resistance variation, bringing the mean yield well below that of individual, high yielding and/or disease resistant individuals. For example, over a four-year survey period of yield data from Costa Rican hybrid varieties, Irizarry and Rivera (1998) showed that 3% of the plants produced over 60% of the yield. In the past, researchers and nurserymen turned to rooted cuttings and grafting techniques to develop clonal propagation systems. These systems use predominantly plagiotropic cuttings, since this type of material is much more abundant in the field, but these cuttings do not grow orthotropically, nor will they produce a taproot. In some cases, this may present problems with water deficit and anchorage during drought or in high winds, especially during plantlet establishment. In the 1950’s and 1960’s, plant tissue culture methods were developed for the propagation of a wide variety of species, but were not applied to cocoa until the late 1970’s and even then with only limited success. Recently, research conducted at a number of laboratories worldwide has led to the development of efficient methods for somatic embryogenesis (SE) of cocoa. The first report of cocoa SE by Esan (1975) described a method using immature zygotic embryo tissue explants, similar methods were henceforth reported by others (Pence, et a/. 1979, Villalobos and Aguilar 1990).

However, while a step forward, these methods do not reproduce clonal plants. Subsequent efforts were directed towards the development of tissue culture systems via different sporophytic tissues including leaves (Litz 1986), nucellus (Chatelet et a!. 1992; Figueira and Janick 1993; Sondahl et a!. 1993) and floral explants including petals and staminodes (Lopez-Baez eta!. 1993, Alemanno eta!. 1996a, Alemanno, et a!. 1996b, Alemanno eta!. 1997). However, these systems were limited by genotypic variability and low plant conversion rates. More recently an efficient SE system using staminode and petal explants, which is capable of propagating a wide variety of cocoa genotypes with high efficiency, was reported (Li eta!. 1998).

In addition to somatic embryogenesis, a number of related in vitro, and ox vitro propagation systems have also been developed, combining the power of tissue culture with the simplicity of rooted cuttings. The intent of this manuscript is not to cover all aspects of this field of research nor its history, but to review recent advances in cocoa tissue culture, ex vitro performance, and multiplication of plants produced by this method. This review includes information from papers presented at the 3~ INGENIC Workshop and at the 13°’ International Cocoa Research Conference held in October 2000, Kota Kinabalu, Malaysia. Apologies are extended to those left out, if any.

But first a precautionary note. While the propagation methods reviewed here are potentially very powerful, it should be noted that they are relatively new technologies that have yet to be fully evaluated in the field. Although early field and greenhouse tests have been very promising, the use of somatic embryo derived plants for large-scale propagation and production must clearly await full validation through multi-locational field trails with multiple genotypes. Such trials are now underway in several countries, but until fully evaluated, we must regard these new technologies as research tools.

Recent advances in cocoa embryo genesis
During the past two years, researchers at the Penn State Cocoa Molecular Biology Laboratory have continued to refine the somatic embryogenesis procedure developed in 1998, and to investigate the cellular origins and developmental pathways operative in this system. We have developed a secondary embryogenesis system, in which explants from primary somatic embryos are re-cultured on induction media, and secondary embryos are formed (Maximova et at 2000). The morphological and ultrastructural changes occurring over time during both primary and secondary cocoa somatic embryogenesis were studied by Dr. Siela Maximova of Penn State in collaboration with Dr. Laurence Alemanno of CIRAD, using a combination of electron and light microscopy for the genotype Scavina 6 (Alemanno of at 2000). This analysis showed that primary embryos arise predominantly from clusters of cells which cooperatively form embryonic nodules perhaps resulting from complex interactions of hundreds of cells. Interestingly, secondary embryos usually arise from the division of single epidermal cells, in a pathway reminiscent of zygotic embryogenesis. Thus, the two types of embryos originate through pathways that differ in the number and location of the cells contributing to embryo formation. One manifestation of these two developmental pathways is the resulting conformity of embryo morphology. In both systems, not all embryos formed are ‘normal” shaped with a single well-defined axis and balanced root/shoot symmetry. Sometimes fused embryos or embryos with other abnormalities are observed. However, as might be predicted in retrospect, secondary embryos, which arise from a uni-cellular origin, exhibit a higher rate of normal embryo conformity (approximately 85%) than do primary embryos (approximately 18%). Another striking difference reported was the number of embryos produced per explant in each of the systems. Secondary embryos were produced at a much higher rate (average 70 per cotyledon explant) than were primary embryos (average 17 per staminode explant). Secondary embryos also performed better than primary embryos during the conversion step, when embryos begin to grow into plantlets capable of autotrophic growth in soil, with approximately 55% success compared to a rate of 32% for primary embryos. The influence of genotype on efficiency and conformity was also examined, and shown to have a strong influence, with up to ten-fold differences in embryogenic potential between genotypes observed. However, this research has shown that nearly every genotype tested to date has produced at least a minimal number of somatic embryos, and that these can then be used for secondary embryogenesis to scale up production. Another contrast in the cellular morphology observed between the two pathways was the difference in plasmodesmatal frequency. Epiderrnal cells within staminodes (the tissue explants for primary embryogenesis) have more plasmodesmata than do epidermal cells of cotyledons from primary embryos (the source of tissue for secondary embryo production). It is possible that the higher level of intercellular connections seen in staminode tissues is in part responsible for the higher degree of cellular co-ordination seen in primary embryogenesis as compared to secondary embryogenesis, but this hypothesis awaits experimental testing. This study concluded that, despite the losses due to abnormal embryos and during the conversion procedure, it is theoretically possible to produce over 4,000 secondary embryo derived plants from a single flower in approximately one year (using Scavina 6).

This method of producing plants is currently estimated to cost approximately $10 per plant, using US based labour costs. One possibility to reduce this high cost would be to couple the tissue culture based systems to downstream, greenhouse or field based, low cost classical propagation systems. One such system presented by these authors utilises the concept of Bentwood Gardens’ (Guiltinan et S. 2000a; Guiltinan ot at 200Db). In this system, the stems of juvenile somatic embryo plants, approximately 3-4 feet in height, are bent and secured in a horizontal position. This releases the strong apical dominance normally exhibited by cocoa plants and results in the outgrowth of the previously dormant meristems in the lower portion of the plant, typically resulting in the growth of approximately five shoots per plant. After about two months, each shoot can be excised, and about ten nodal cuttings produced and rooted using well-known conventional methods in a greenhouse. The bent plant will then continue to produce more shoots, which can be harvested repeatedly for long periods of time. The rooted cutting will grow with the orthotropic architecture of a normal seedling plant. Importantly, these plants also exhibit a strong dominant root, or perhaps two, which grow straight down in a similar way to a taproot. This is quite different from rooted cuttings developed from plagiotropic or fan branch materials, which form shallow, fibrous root systems without a taproot. Using this system, nearly 250 plants can be produced from each somatic embryo plant per year, greatly reducing the cost per plant compared to use of tissue culture alone. The development of plants produced by this method is still under evaluation.

Smilija Lambert of Mars Inc. presented work performed at the Almirante Centre for Cocoa Studies in collaboration with researchers from Penn State (Lambert eta!. 2000). She described a system for downstream propagation of somatic embryo plants referred to as mini-cuttings. In these experiments, the team used small tissue cultured plantlets, acclimated to greenhouse conditions, as sources of apical cuttings which were then rooted and grown in the greenhouse for three to four months after which they were ready for transfer to the field. Very high success rates were obtained. Similar to the Bentwood technique described above, the original stock plants continue to produce orthotropic shoots from which additional cuttings can be made. This is another system that provides a rapid way to produce large numbers of orthotropic plantlets at a lower cost than tissue culture alone.

During the 3~ INGENIC workshop, Dr. Lopez Baez from the State University of Chiapas, Mexico presented the results of recent work using modified MS salts based me dia (Lopez-Baez et at 2000). Differing from the hormonal system used by the Penn State method, 2,4-D or 2,4,6-T (1 mg/I) and kinetin (0.26 mg/I) were used for induction of embryogenesis, and this system successfully produced embryos from twelve different genotypes with success rates varying between 20% and 41%. Glucose and sucrose concentrations used were between 50 mg/I and 80 mg/I. A cold shock treatment of the flowers prior to culture, consisting of exposure to 0°C for 15 or 20 mm and 5°C for 120 mm, was shown to enhance embryogenesis efficiency.

Dr. Douglas Furtek presented research conducted at the Malaysian Cocoa Board which also examined the different factors affecting somatic embryogenesis of cocoa (Furtek et at 2000; Tan et at 2000). This group tested interactions between different genotypes, carbon sources, basal media and plant growth regulators. Thirty clones were evaluated for their potential to produce somatic embryos. The percentage of calli producing embryos ranged from 0% to 18.2%. It was determined that different clones vary in their response to given carbohydrate sources, plant growth regulators or basal media. The results indicated that it may be necessary to optimise the procedure for the individual genotype of interest in order to obtain the highest success rates. Similar to the results reported by other researchers, normal and abnormal embryos were also observed using this procedure. Seven different clones produced 18 embryos of normal appearance that have been converted into plantlets and transferred to pots. The research group in the University of Yaounde, Cameroon, investigated the differences between embryogenic and non-embryogenic callus at the biochemical level (Niemenak ot at 2000). The results presented by Dr. Niemenak at the 13th International Cocoa Research Conference demonstrated levels of activities of peroxidase and IAA-oxidase three to five times higher in embryogenic callus compared to non-embryogenic callus. The highest peroxidase activity was detected in the soluble fractions and the highest IAA-oxidase activity was detected in the ionic enzymatic fraction. The tissue culture media used in the course of this study were: MS/2 and 2,4-Dfkinetin (2/0,5 mg/l)for callus induction with 57% success and MS/2 and IBA at 2 mg/I for embryo development, with 25% success.

Technology transfer and field test establishment

Beginning in March of 2000, a Penn State research associate, living in Africa, has worked to establish the cocoa embryogenesis system in Ghana and Ivory Coast. The work, although initially hindered by some difficult conditions and malfunctioning equipment, has been successful, and the first plants propagated via somatic embryogenesis and micropropagation have been produced and acclimated, and will soon be planted in the field in both countries. At the end of the first year, there were more than 800 somatic embryos produced in Ivory Coast and more than 400 embryos in Ghana. With support from Sustainable Tree Crop Program, USAID during 2001, the technology will also be also transferred to scientists in Cameroon and Nigeria. Field test sites have also been established in St. Lucia and Brazil, and are planned for other countries. Scientists from over eight different countries have participated in technology transfer workshops.


Systems for cocoa somatic embryogenesis have now been well developed in a number of laboratories worldwide, and technology transfer will ensure the spread of the technique to producing countries in the future. The two basic systems for embryogenesis induction in use (one 2,4-D based and the second TDZ based), result in embryo production but some differences in efficiencies are apparent. The DKW basal salts media appears to be superior to MS media for cocoa tissue culture, and although modifications of MS have given adequate results, most laboratories have nowswitched to DKW based media. Downstream variants of rooted cutting systems have been developed, which offer the possibility of cost reductions on a per plant basis. Field tests of plants produced by these methods are necessary to fully validate the systems for use in large-scale propagation systems. These methods hold great promise for the future of cocoa genetic improvement and production by contributing to the ability to propagate elite genotypes rapidly for research and production. The tissue culture system may also be further developed to include disease indexing, useful for production of clean stock, for international germplasm exchange, for cryopreservation and for germplasm conservation.


Alemanno L., M. Berthouly and N. Michaux-Ferriere. 1996a. Embryogenèse somatique du cacaoyer a partir de pièces florales, Plantations Recherche Developpement 3: 225-237.
Alemanno L., M. Berthouly and N. Michaux-Ferriere. 1996b. Histology of somatic embryogenesis from floral tissues cocoa. Plant Cell, Tissue and Organ Culture 46: 187-194.
Alemanno L., M. Berthouly and N. Michaux-Ferrmère. 1997. A comparison between The obroma cacao L. zygotic embryogenesis and somatic embryogenesis from floral explants. In Vitro Cell. Dev, Bio. Plant. 33: 163-172.
Alemanno L., S. Maximova, N. Ferrier and M.J. Guiltinan. 2000. Comparison de lembryogenese somatique primaire et secondaire: evaluation de lefficacité et ontogenEse des embryons somatiques. In Proceedings of 13th Int. Cocoa Research Conf., Kota Kinabalu, Sabah, Malaysia.
Chatelet P., N. Michaux-Ferriere and P. Dublin. 1992, Potentialites embryogenes du nucelle et du tegument mnterne de graines immatures de cacaoyer (Theobroma cacao L.). CR AcM Scm Paris 315: 55-62.
Esan ES. 1975. Tissue culture studies on cacao (Theobroma cacao L.). Pages 116-125 in Proceedings of 5th nt. Cocoa Research Conf. Ibadan, Nigeria.
Figueira A. and J. Janick. 1993. Development of nucellar somatic embryos of Theobroma cacao. Acta Horticulturae, In Vitro Culture 336: 231-238.
Furtek ID., R. Kasran, CL. Tan, L. Johnsiul, A. Mohammed, V. Hartney, MM. Yusof, K. Lamin and L.M. Tong. 2000. Biotechnology research by the Malaysian Cocoa Board. In Abstr. 3~ INGENIC Workshop on the New Technologies and Cocoa Breeding, Kota Kinabalu, Sabah, Malaysia.
Guiltinan M., C. Miller, A. Traore and S. N. Maximova. 2000a. Greenhouse and field evaluation of orthotropic cacao plants produced via somatic embryogenesis, micro and macro-propagation. In Proceedings of 13th nt. Cocoa Research Conf., Kota Kinabalu, Sabah, Malaysia.
Guiltinan M., SN. Maximova, A. Traore, C. Miller! M. Robinson, G. Antunez de Mayolo, E. Barreto, A. Young and S. Pishak. 2000b. Integrated systems for the propagation of cacao: tissue culture, bentwood gardens and rooted cuttings. In Abstr. 3~ INGENIC Workshop on the New Technologies and Cocoa Breeding, Kota Kinabaju, Sabah, Malaysia.
Irizarry H. and E. Rivera. 1998. Early yield of five cacao families at three ocations in Puerto Rico. J. Agric. Univ. P.R. 82(3-4): 163-171.
Lambert S.V., M. Guiltinan, S. Maximova and W.M. Aitken. 2000. Ex vitro propagation of acclimated somatic embryo derived cocoa plants. In Proceedings of 13tb nt. Cocoa Research Conf.. Kota Kmnabalu, Sabah, Malaysia.
Li Z., A. Traore, S. Maximova and M.J. Gumltmnan. 1998. Somatic embryogenesis and plant regeneration from floral explants of cacao (Theobroma cacao L.) using thidiazuron. In Vitro Cell Dev. Biol. Plant 34: 293-299.
Litz RE. 1986. Tissue culture studies with Theobroma cacao. Pages 111-120 in: (Dimick, P.S.. ed) Cacao biotechnology symposium, Tropical Research and Education Center: University of Florida, IFAS Department of Food Science, College of Agriculture, The Pennsylvania State University.
LOpez-Baez 0., H. BoRon, A.B. Eskes and V. Petiard. 1993. Embryogenese somatique de cacaoyer Theobroma cacao L., a partir de pièces florales. CR Acad Scm Paris 316: 579-584.
Lopez-Baez 0., J.L. Moreno-Martmnez and S. Pacheco-Rodas. 2000. Advances in cocoa Theobroma cacao propagation by somatic embryogenesis mn Mexico. In Abstr. 3’~ INGENIC Workshop on the New Technologies and Cocoa Breeding, Kota Kinabalu, Sabah, Malaysia.
Maximova SN., L. Alemanno, A. Young, A. Traore, N. M~chaux-Ferriere and M. Guiltinan. 2000. Efficiency, origin and quality of cacao somatic embryogenesis. In Proceedings of j3th Int. Cocoa Research Conf., Kota Kinabalu, Sabah, Malaysia.
Niemenak N., P.F. Djogoue, G. Tsala Ndzomo and D. Omokolo Ndoumou. 2000. Activites peroxydasique et auxine-oxidasique de l’embryogenese somatique in vitro chez The obroma cacao L. In Proceedings of 13th nt. Cocoa Research Conf., Kota Kinabalu, Sabah, Malaysia.
Pence V.C., P.M. Hasegawa and J. Janick. 1979. Asexual embiyogenesis in Theobroma cacao L.J. Amer. Soc. Hort. Sci. 104: 145-148.
Sondahl MR., S. Liu, C. Bellato and A. Bragmn. 1993. Cacao somatic embryogenesis. Acta Horticulturae 336: 245-248.
Tan CL., MR. Davey, K.C. Lowe, J.B. Power and D.B. Furtek. 2000. A study on the factors affecting somatic embryogenesis in Theobroma cacao L. In Proceedings of 13th Int. Cocoa Research Conf., Kota Kinabalu, Sabah, Malaysia.
Villalobos V. and ME. Aguilar. 1990. Plant regeneration of cacao (Theabrome cacao L.) by micro-grafting somatic embryos. In Abstr. Vllth Int. Congr Plant Tissue and Cell Culture, Amsterdam, Assoc, for Plant Tissue Culture 140,