753 HORTSCIENCE VOL. 41(3) JUNE 2006 HORTSCIENCE 41(3):753–758. 2006. Received for publication 17 Oct. 2005. Accepted for publication 8 Jan. 2006. The authors thank Siela Maximova and John Carlson for critical reading of the manuscript, Zhijan Li for his helpful advice and Sharon Pishak for her technical support. We also extend our thanks to Yo Tiemoko (Director of CNRA in Cote d’Ivoire), Sangare Abdourahamane (Director of the Central Biotechnology Laboratory at CNRA Cote d’Ivoire) and his technicians (Ale Emmanual, Toure Mah, Assovi Claude and Zran Emmanuel) for assisting with embryo maturation and conver- sion experiment. This research was supported by a grant to Mark Guiltinan from the American Cocoa Research Institute. 1 Author to whom correspondence should be directed; axt14@psu.edu. Effects of Carbon Source and Explant Type on Somatic Embryogenesis of Four Cacao Genotypes Abdoulaye Traore 1 School of Forest Resources, The Pennsylvania State University, University Park, PA 16802 Mark J. Guiltinan Department of Horticulture, The Pennsylvania State University, 422 Life Sciences Building, University Park, PA 16802-4200 Additional index words. petal, staminode, tissue culture Abstract. The effects of five carbon sources (glucose, fructose, maltose, sorbitol, and sucrose) and two explant types (petals and staminodes) on cacao somatic embryogenesis was studied. No growth was observed on both types of explants cultured on sorbitol containing media and slow growth was obtained on media supplemented with maltose. Depending on the genotype, the percentage of explants producing one or more embryos ranged from 6% to 99%, 18% to 98%, and 3% to 82% on media containing glucose, fructose and sucrose respectively. Explants cultured continuously on maltose or sorbitol-containing media failed to produce embryos. Staminode explants produced 3 to 10 times more somatic embryos than petals. A strong genotypic effect on somatic embryogenesis was observed. Staminode explants of the Forastero clones Laranja and PSUSca 6 produced 2 to 30 times more so- matic embryos than the Trinitarios UF 613 and ICS 16. During embryo maturation and conversion, no significant differences were observed among glucose, fructose, maltose, or sucrose for embryo weight, total shoot and root production. However, we found that all plantlets produced on glucose had shoots with normal cacao leaves while the other carbon sources sometimes produced plantlets with cotyledon-like leaves. In vitro propagation via somatic embryo- genesis (SE) provides a means to produce large numbers of genetically identical and often pathogen-free plants. It can also be used to develop a genetic transformation system or to preserve germplasm via cryopreservation of somatic embryos. However, for these ap- plications to be technically and economically feasible, it is essential to optimize the system variables to obtain high multiplication rates of quality embryos. In somatic embryogen- esis protocols, a carbon source is required to provide the necessary energy and building blocks for the process. Sucrose and glucose have been widely employed; however other carbon sources have also been reported to be effective in supporting somatic embryogenesis in plants (Brown et al., 1995). For example, the effectiveness of maltose as a carbon source for somatic embryogenesis has been demonstrated in petunia (Raquin, 1983), alfalfa (Strickland explants with sucrose and glucose, however the effects of various carbon sources on embryo production, maturation or conversion was not studied. In order to determine the relative efficacy of various carbon sources, and the most efficient explant type for cacao somatic embryogenesis, a comparative study of these parameters was performed. The objectives of this study were to identify the best combination of explant type and carbon source for efficient cacao somatic embryo production and to test the effects of various carbon sources on embryo maturation and conversion into plantlets. Material and Methods Plant materials. Staminodes and petal explants from immature flowers of four cacao genotypes (ICS 16, Laranja, PSUSca 6 and UF 613) were used. Cacao trees were grown in a greenhouse maintained at 70% to 80% relative humidity and 27 to 30 °C at the Pennsylvania State University, University Park, Pa. Flowers were surface sterilized for 20 min using 10% calcium hypochlorite solution, then dissected, and staminodes and petals cultured on various media in 100 × 15 mm plastic petri dishes. Media formulation. The DKW-based media protocol, according to Li et al. (1998), was used to test five different carbon sources: fructose, glucose, maltose, sorbitol and sucrose, each at a concentration of 100 mM. The other media components remained as reported by Li et al. (1998). Somatic embryogenesis. Embryogenic callus was induced from staminode and petal explants by culturing them on primary callus growth (PCG) medium (Li et al., 1998) at 26 °C in the dark for 14 d, after which they were transferred onto a secondary callus growth (SCG) medium (Li et al., 1998) and cultured at 26 °C in the dark for 14 d. Embryo development was initiated by a transfer of the explants from SCG medium to embryo development (ED) medium (Li et al., 1998). In the first experi- ment, each of the media contained glucose, fructose, maltose, sorbitol or sucrose, and the explants were consecutively transferred to the next media with the respective carbon source. Data for explant growth were collected every 2 weeks by weighing a set of five ran- domly selected staminode or petal explants. In a second experiment, the different carbon sources were tested only on PCG medium, then the explants were transferred onto SCG containing glucose followed by a transfer to ED containing sucrose and glucose according to Li et al. (1998). Two months after culture initiation, the percentage of embryogenic explants [(number of explants producing one or more embryos / total number of explant) × 100] and the aver- age number of embryos per responsive explant (number of embryos per number of explant producing one or more embryos) were recorded by counting the number of embryo-producing explants in each treatment and the number of embryos on each responsive explant. Each medium and carbon source combination was evaluated with four replicate plates contain- ing 20 staminodes and 20 petals each per et al., 1987), pine (Nagmani et al., 1993), pea (Loiseau et al., 1995), and apple (Daigny et al., 1996). In addition, maltose has been used to improve somatic embryo maturation in cacao (Lopez-Baez et al., 1993) and pine (Li et al., 1998). In maize, Swedlund and Locy (1993), indicated that sorbitol improved embryogenic calli production and plant regeneration. Fruc- tose has been shown to induce large numbers of embryos in cucumber (Ladyman and Girard, 1991), and to promote more embryo production than sucrose, glucose or maltose when used at concentrations higher than 252 mM in pea (Loiseau et al., 1995). In 1987, Elhag et al. reported a comparative study of sucrose, glucose and fructose effects on cacao somatic embryogenesis using calli induced from zygotic tissues. As a result of that study, the authors indicated that embryogen- esis increased when sucrose was replaced by glucose or fructose; however, several somatic embryogenesis protocols using tissue from maternal origin were subsequently developed using sucrose (Alemanno et al., 1997; Figueira and Janick, 1993; Lopez-Baez et al., 1993; Sondahl et al., 1989). Li et al. (1998) reported a successful somatic embryogenesis protocol using maternal tissue (staminodes and petals), glucose and sucrose. Maximova et al. (2002, 2005) have further reported on this system and a related secondary embryogenesis procedure. Traore et al. (2003) also reported a cacao micro- propagation system based on somatic embryo derived plantlet using glucose and sucrose. These studies described efficient production of cacao somatic embryos obtained from floral