Introduction
Coffee is one of the most important agriculture commodities in the international market. Its production is restricted to tropical countries including Indonesia. There are two commercially important coffee species: Coffee arabica L. (Arabica coffee) and C. canephora Pierre (Robusta coffee). Quality beverage is produced from Arabica coffee which represent 70% of the commercial coffee of the world (Sondahl et al., 1985). Those two important species are propagated via seeds. Propagation using tissue culture might provide a superior alternative to seed or vegetative propagation by cuttings. Recent advances of cell and tissue culture open the possibility of applying the improvement of economically important plants (Smith, 1974; Chaleff and Carlson, 1974). Coffee geneticists are interested in the application of this technology for the regeneration and selection of genetic variability for obtaining new cultivars.
Staritsky (1970) published the first work on coffee tissue culture in which callus cells were successfully cultivated using explants from soft internodes of young orthotropic shoots of Coffea arabica, C, canephora, and C, liberica. He also succeeded in producing somatic embryos and plantlets of C, canephora shoots. For these studies Staritsky used a modified medium of Linsmaier & Skoog with 2,4-Dichlorophenoxyacetic acid (2,4-D) 0.1 mg/l or Naphthalene Acetic Acid (NAA) 1.0 mg/l and Kinetin 0.1 mg/l. Herman and Haas (1975) have reported of obtaining organoids from leaf explants of C. arabica cultured on medium Linsmaier & Skoog (medium LS) supplemented with 0.46 µM Kinetin and 0.45 µM 2,4-D after 60 days in darkness. Many of these organoids showed abnormal leaf formation. Selected organoids characterized by green spots with normal leaves were subcultured onto medium Gresshoff and Dov no. 4 containing 0.54 μM NAA in the absence of Kinetin. Root formation occurred after 60 days and following 7 months the surviving plantlets were transferred to soil. Recently, Sondahl et al. (1985) obtained plantlets of coffee by somatic embryogenesis and auxillary bud culture. Leaf explants were grown on Murashige & Skoog's inorganic salts, 30 \(\mu\)M thiamine-HCl, 210 \(\mu\)M L-lysine, 550 \(\mu\)M meso-inositol, 117 \(\mu\)M sucrose, and 8 g/l Difco Bacto agar supplemented with 20 μM Kinetin and 5 μM 2,4-D. Somatic embryogenesis was observed on Murashige & Skoog's organic salts, 58.4 µM sucrose, 2.5 µM Kinetin, and \(0.5 \mu M\) NAA. In these studies of somatic embryogenesis, plant regeneration with coffee tissue has been accomplished from solid cultures using relatively high amount of 2.4-D or NAA.
In our laboratory at the Dept. of Biology, ITB, attempts have been made to evaluate the feasibility of developing various aseptic techniques for clonal multiplication and improvement of C. arabica. This paper reports the results of initial experiments designed to establish somatic embryogenesis using solid and liqr-lid cultures with relatively high antounts of Kinetin in the absence or with re latively low amount of NAA.
Materials and Methods
The original coffee plants were cultivated in the area of Pengalenganear Cibiana at an altitude of about 900 meter above sea level and obtained through the Perkebunan Rakyat Tanaman Pangan Ekspor in West Java. Yourrg leaves from plagjotropic branches of about 2 years old C. arabica plants were surface stenlized in 1.6i2 Naiypochlorite with 0-l% Tween l0 as surfactant for 20 nrinutes, then washed several times in sterile distilled water. Inteweinal segnrents about 5 mm2 were excised and transferred to basal medium for callus formation. The source of experimental n'!aterial was callus obtained from these leaf segments. Therefore, some parts of the callus were inoculated on solid rnediLrrn of Linsmaier & Skoog (1965 ) for multiplication of callus and for induction of somatic embryogencsis. Some other parts of the callus wr"-re transt'erred into liquid medium of Gamborg, Miller, and Ojima (1968) for sonratic embryogenesis. All the cultures were grown in diurnal condition at rooln temperature.
The basal nutrient lnediunr for all callus cultures is of Linsnraier and Skoog (1965) or LS medium containing the following substances (mg/liter):
NH4NO3 1650: KNO3 1900; CaClr.2HrO 440; MgSOo.THrO 370; KH2PO4 170: Na, EDTA 37.3; FeSOa.THrO 27.8; H3BO3 6.2; MnSOa.TH2O 22.3; ZnSO".7H.O 8.61 Kl 0.831 NarMoOa.2HrO 0.25; CuSOo.5H2O 0.025i CoCl2.6HrO 0.025; thiaminc-HCl 0.8; myo-inositol 100; sucrose 30 g/l; and agar 8 g/1. The pI{ is adjusted to 5.7 before sterilization in an autoclave at I20'C for 15 rninutes. The basal nutrient medium for liquid culture is of Camborg, Miller, and Ojirna 11968) or CA medium which contains the following substances (rng/liter): KNO, 2500: CaClr.2HrO 150; MgSOa.THrO 250; (NH4)2SO4 134; NaHrPOo.HrO 150; KI 0.75;H.BO3 3.0; MnSOo.HrO l0; ZnSOa.THrO 2.0; NarMoOa.2HrO 0.25; CuSOo.5llrO 0.025; CoClr.6HrO 0.025; NaTEDTA 37.3; FeSOa.THrO 27.8; myo-inositol 100; ttriaminc-HCl l0; pyridoxine-LlCI 1.0: nicotinic acid 1.0; sucrose 20 g/l; and pH 5.7.
Results and Discussion
Culture of leaf segments. Two auxins 2,4-Dichlorophenoxyacetic acid (2,21-D) or Naphthalene Acetic Acid (NAA) and one cytokirrin in the forrn of Kinetin werc used for callus initiation on leaf segments ol Coffea arcbica. Fdr Illis purpose different combination and concentrations of growth regulators were used. In cultures on LS medium with 2 gM 2,4D and 5 7.5 pM Kinetin organized erowth occurcd 18 weeks after inoculation (Fig. I A). In cultures containing 0.5 pM NAA compact calhrs were formed after two weeks. However these calli soon became brown and failed to be subcultured. The only organized growth occured on LS mcdiurr containing 0 or 0.05 pM NAA and 6 8 gM Kinctin. Whitc and friablc callus were fornted. Globular tissues and "gr('en spots" as signs of shoot formation appcared on the surface of the callus {Fig. I B). Somatic embryos will then bc fornred 5 weeks after callus initialions (Fig.2).
Fig. 1A Callus formation maier and Skoog with 2 irlV or) leaf segments grown on medium Lins-2,4-D and 5-7,5 llvl Kinetin.
Fig. 1 B Close-up view on globular types and "green spots" on the surface of the callus.
From the above results it is again clear that for callus formation both auxin and cytokinin are necessary. However, the amount of the respective growth regirlators rlight be different. For example, Staritsky (1970) and Herman & Haas (1975) or Sondahl and Sharp (1975, 1977) and Sondahl et al. (1985) needed relatively high amount of auxin and relatively low cytokinin for the induction of callus on leaf segments of C. arabica and for their induction of somatic embryos. In contrast we have found for this inductions relatively low amount of auxin and relatively high amount of cytokinin. The same phenomena of direct somatic embryogenesis were observed by Dublin (1981) and Yasuda et al. (in press). However, Dublin (1981) used medium of Murashige & Skoog (1962) containing high content of cytokinin in the form of Benzylanrinopurine (BAP). He studied leaf segments of Arabusta coffee and needed 0.01-l mell L,4-D and l-10 mg,/l BAP for somatic embryogenesis. Yasuda el41. (in press) studied somatic embryogenesis of C. arabica on medium Murashige and Skoog containing only 5 pM BAP.
Culture of leaf callus. Leaf calli were subcultured on two different kinds of media: a solid medium of LS and a liquid mediunr of GA. Callus which was fbnned on LS mediurn contxining 0 or 0.5 pM NAA and 6-8 gM Kinetin was sLrbcultLrrcd on a sin:ilar mediunr for multiplication and sornatic cmbryogenesis. However, plaDtlets were subcultured on LS medium without growth regulators (Fig. 3) or LS medium with Indole Butyric Acid (IBA) as much as 0.05 pM for root formation.
Callrrs segnrents originating from LS media containing 2,4-D and Kinetin were all transferred into GA )iquid medium with 0 0.25 1tM 2,4-D. Organized growth was observed 8-10 weeks after transferring. With varying length of time in culture random ccll division actively occured. These scattered regions of high mitotic activity led to the formation of aggregates, which became more friable and continued to divide. Continue division of the surface led to the fornration of small proturbances giving the tissue a nodular appearance which then produced embryos. These embryos were released when they reached the globular stage (Thorpe, 1982; Sondahl eral., 1985). This stage was followed by heart stal:le and torpedo stage (Fig. 4 A, B, C) and cotyledonary stage. At a later stage these embryos were subculturcd from the liquid medium (Fig 5 A) to GA solid medium containing 0.51l. suc-rose and 0-0.1 pM 2,4-D. Within 4-6 weeks after transfcrring these somatic embryos developed into plantlcts (Fig. 5 B). This pattern of il ril/o embryogenic developnrent could be considcred as Induced-Embryogenic Determined Cells (IEDC) as proposed by Sondahl and Sharp ( 1977).
Fig. 2 Spongy type of cailus tissue (left), somatic embryos, and plantlets (right) oI Coffea arcbica.
Fig.3 Older plantlets grown on LS medium with 3% sucrose without qrowth regulators.
Fig. 4 A and B Globular and heart stage embryos.
Fig. 4 C Terpedo stage embryos.
Fig. 5 A Somatic embryos formed in liquid culture.
Fig.5 B Plantlets grown on GB medium (see text).
Acknowledgement
The research reported and discusscd in this paper is one of tllc results of cooperative research progranrme betrveen the Japan Society for the Promotion of Scicncc' (JSPS) in Tokvo and the Directorate Gcneral of ieher Educntion (DGHE) in Jakarta. To all those institutes we all are thanlifLrl and ffateful.
