J. A mer . S oc . H ort . S ci . 114(3):511-515. 1989. Recovery of y-Linolenic Acid from Somatic Embryos of Borage James Quinn, James E. Simon, and Jules Janick Department of Horticulture, Purdue University, West Lafayette, IN 47907 Additional index words. Borago officinalis, y-linolenic acid, lipids, tissue culture Abstract. Somatic embryos of borage (Borago officinalis L., Boraginaceae) were induced directly from immature zygotic embryos and indirectly from callus. Embryogenic callus maintained on liquid basal medium supplemented with 4.5 pM 2,4-D and 10% (v/v) coconut water (CW) produced globular structures that became rhizogenic upon transfer to 2,4-D-free basal medium. Embryogenic callus maintained on semisolid basal medium supplemented with 4.5 jxm 2,4-D and 10% CW continued to produce somatic embryos, but development was abnormal. Globular structures often failed to develop cotyledons, and those that developed were small and fused; hypocotyls tended to be large and elongated. Root meristems appeared normal, but shoot meristems were not formed. Carbon source (sucrose, glucose, or maltose) and ABA did not normalize somatic embryogenesis. A highly embryogenic, non-browning clone produced various tissue types when 2,4-D was withdrawn that varied in total fatty acids: white nodular structures (12.6%), cotyledonary structures (22.5%), white callus (5.0%), green leafy growth (3.1%), and translucent globular growth (5.1%). y-Linolenic acid, as a percentage of total fatty acids, was highest in cotyledonaiy structures (19.9%) and lowest in white callus (10.2%). Chemical names used: D); coconut water (CW). Borage, native to Europe, North Africa, and Asia Minor (Si- mon et al., 1984), is an ancient herb mentioned by Theophrastus in the 3rd century bc (Hort, 1916) and noted by Dioscorides (Gunther, 1934) and Pliny the Elder (Bostock and Riley, 1855) for its “ mirth” -inducing qualities. Recent interest has focused on borage as a source of y-linolenic acid, a precursor to pros- taglandins (Bergstrom et al., 1964; Willis, 1981; Carter, 1988), and as a treatment of atopic eczema (Wright and Burton, 1982). Borage seed, the richest known plant source of y-linolenic acid, contains 28% to 38% lipid, of which 17% to 25% is y-linolenic acid (Wolf et al., 1983; Kleiman et al., 1964; Whipkey et al., 1988). Large-scale field production of borage is problematic, however, due to its indeterminate growth and seed-shattering characteristics (Beaubaire and Simon, 1987). The agricultural problems of borage prompted a search for an in vitro method to produce y-linolenic acid via the culture of somatic embryos (Janick et al., 1987; Whipkey et al., 1988). The present study on borage was undertaken to investigate the initiation and reg- ulation of somatic embryogenesis and to determine fatty acid accumulation by somatic embryos. Materials and Methods Plant material. Borage seed was obtained from PGE Tech- nology (Danvers, Mass.). Borage was grown in the field and the lathhouse (50% shade) in Summer 1985, or in the green- house (18C night and 24--27C day) with supplemental fluores- cent lighting providing between 110 to 140 (jumol-s ^m -2 of PAR. Zygotic embryo developmental stage was measured by the number of days after pollination (DAP). Tissue culture. Immature ovules were surface-sterilized with 0.5% sodium hypochlorite (10% Clorox) for 30 min and rinsed three to five times with sterile distilled water. Each excised embryo was cultured in 10 ml of basal medium either in plastic petri dishes (60 x 15 mm) sealed with Parafilm, or in glass Received for publication 31 May 1988. Journal paper no. 11,416 of the Purdue Univ. Agricultural Experiment Station, West Lafayette, IN 47907. The cost of publishing this paper was defrayed in part by the payment of page charges. Under postal regulations, this paper therefore must be hereby marked adver- tisement solely to indicate this fact. abscisic acid (ABA); 2,4-dichlorophenoxyacetic acid (2,4- culture tubes (22 mm diameter x 150 mm height) using plastic closures. The basal medium consisted of Murashige and Skoog salts (1962) supplemented with 0.3 p,M (0.1 mg-liter-1) thia- mine-HCl, 2.4 |xm (0.5 m g-liter1) pyridoxine-HCl, 0.55 mM (100 mg*liter_1) /-inositol, 4 \xm (0.5 mg-liter-1) nicotinic acid, 26.6 fiM (2 mg-liter-1) glycine, 1 g-liter-1 casein hydrolysate, and 87.6 mM (30 g-liter-1) sucrose with pH adjusted to 5.7. Medium was gelled with 6 g agar/liter (Sigma). Growth regu- lators, when used, were added before autoclaving, except for ABA, which was filter-sterilized and added after autoclaving. Coconut water (CW) was added before autoclaving at 10% or 20% (v/v). Cultures were grown at 26C under cool-white flu- orescent lamps (45 fimol*s-1-m~2) with a 16-hr photoperiod. Fatty acid analysis. Lipids were extracted from tissues using a procedure modified from Folch et al. (1957); triglycerides were saponified and soaps methylated using the procedure of Metcalfe and Schmitz (1961). These procedures are described by Whipkey et al. (1988). Fatty acid composition was analyzed with a Varían 3700 gas chromatograph using a 30 m x 0.32 mm fused silica capillary column with SP-2330 as the stationary phase. Initial column temperature was 150C for 2 min and increased by 4C/min to 220C. Peak areas of fatty acids were determined with a Hewlett- Packard 3390A integrator. The order of peak elution was: 15:0 (internal standard), 16:0, 18:0, 18:1, 18:2, y-18:3, 20:0, a- 18:3, 20:1, 22:1, and 24:1. Peaks were tentatively identified by comparing retention times with those of pure standards. The identity of each fatty acid constituent was confirmed by gas chromatography (Finnigan 9610) and mass spectroscopy (Fin- nigan 4000) using electron impact analysis. Results Induction of somatic embryogenesis. Immature zygotic em- bryos (10 DAP) of borage, obtained from field-grown plants, were cultured in a semisolid basal medium supplemented with factorial combinations of 2,4-D and CW for 1 month. Globular structures and somatic embryos (Fig. 1) were produced in all treatments, with the highest percent produced at 22.6 p,M 2,4- D and 10% or 20% CW (Table 1). The number of somatic embryos per embryogenic explant varied from one to three. J. Amer. Soc. Hort. Sci. 114(3):511-515. 1989. 511