Plant Cell Reports
Plant Cell Reports (1996) 15: 582-585
9 Springer-Verlag 1996
Agrobacterium tumefaciens-mediated transformation of Solanum gilo Raddi as influenced by explant type Essie Blay 1 and Janette V. Oakes 2 1 Crop Science Dept., University of Ghana, Legon, Accra, Ghana 2 Calgene Fresh, Inc., 1920 Fifth St., Davis, CA 95616 USA Received 7 April 1995/Revised version received 25 September 1995 - Communicated by G. C. Phillips
Abstract. An efficient system for Agrobacterium tumefaciens-mediated transformation of Solanum gilo was established. The marker genes for kanamycin resistance and 13-glucuronidase expression were introduced. A comparison between cotyledon and hypocotyl explants showed that while regeneration was better from hypocotyl explants, cotyledon explants gave better transformation efficiency (46% vs. 32%). Four levels of kanamycin selection (100, 150, 200 and 250 mg/1) were tested for effect on transformation efficiency with each type of explant. Lower levels of kanamycin worked better using cotyledon explants, while higher levels of kanamycin worked better for hypocotyl explants. All nine T O plants tested for expression of the kan r gene were positive. The progeny of three of these plants showed a pattern of classical Mendelian inheritance (3 to 1) for both the kan r and the 13glucuronidase genes. Key Words: Transformation
Garden egg - Solanum
gilo-
Abbreviations: MS, Murashige and Skoog (1962) medium; 2,4-D, 2,4-Dichlorophenoxyacetic acid; Nt~II, neomycin phosphotransferase; GUS, 13-glucuronidase.
Introduction
The garden egg (Solanum gilo), a member of the solanaceous family, is an important vegetable in East, West, and Central Africa. The crop has also been reported in Asia (Arora and Hardas 1976). The fruit is consumed as it turns from green to cream with the yellow, orange and red stages rejected by the consumer as overripe. The major production constraints on the crop are various insect pests, which attack the stems, leaves, flowers and fruits, and precocious fruit-ripening. The latter is frequently associated with infestation by fruitborer (Leucinoides orbonalis) and results in overripe fruit and a drastic reduction in consumer acceptance. Correspondence to: J. V. Oakes
Even though the closely related wild species, S. integrifolium, has been reported as a possible source of pest and disease resistance for improvement of eggplant (Solanum melongena), gene introgression into crop plants from wild species frequently involves difficult crosses and lengthy selection owing to deleterious gene linkages. Introduction of single gene traits for insect resistance (Delannay et al. 1989), or ethylene regulation in conjunction with antisense technology to control ripening (Hiatt et al. 1989; Klee et al. 1991; Oeller et al. 1991), could be used to achieve relatively quick improvements in garden egg. However, the first requirement is the development of an efficient in vitro transformation system for this crop. Related species of garden egg, including eggplant, have been successfully transformed with different chimeric genes (Fifippone and Lurquin 1989; Rotino and Gleddie 1990; Rotino et al. 1992). However, crop transformation is influenced by several parameters including genotype, explant source and type, Agrobacterium strain, type of plasmid vector, cultural conditions and composition of the media (Golds et al. 1991; Mukhopadhyay 1992; van Wordragen and Dons 1992). Hence, it is frequently necessary to develop a transformation system for each species and sometimes different genotypes within a species. This paper reports high frequency regeneration of transgenic garden egg plants from cotyledon and hypocotyl explants via Agrobacterium co-cultivation and discusses the effects of explant type and selection level on transformation efficiency. Transformed plants and their progeny were analyzed using root formation on kanamycin, the NPTII dot blot assay, the 13glucuronidase (GUS) histochemical assay, and Southern blot analysis. Materials and Methods Plant material and seed germination. Seeds of S. gilo were obtained
from the germplasm collection of Dr. Corned Bonsi, Tuskegee University. S e e d s w e r e surface sterilized by dipping in 95% ethanol for 1 min, followed by 30 min in 50% Clorox (commercial NaOC1) containing 2 drops of Tween 20 per 100 ml. Seed were subsequently
583 rinsed 5 times in sterile, deionized water. Fifty seeds per plate were germinated asceptically in 100 X 15 mm petri dishes containing 40 ml of seed germination medium composed of MS salts (Murashige and Skoog 1962) supplemented with 0.8 mg/l thiamine hydrochloride, 0.4 rag/1 g]ycine and 3.5 g/1 AGARGELTM (Sigma Chemical Co.). The pH of the medium was adjusted to 5.8, prior to autoclaving at 121~ for 20 vain. Seeds and all in vitro plant material were incubated at 23~ under a 16 h fight and 8 h dark photoperiod. Light was provided by a 2 to 1 mix of cool white and gro-lux fluorescent lamps with an intensity of 50 mmol m -2 s-I.
Transformation, selection and regeneration procedures. For both transformation and regeneration procedures explants were prepared as follows. Cotyledons from 6 d old seedlings were excised and both ends were trimmed to generate rectangular explants approximately 3 mm in length. The hypocotyls were similarly sectioned into approximately 3 nun pieces. The apical as well as lower 5 mm pieces of the hypocotyls were discarded. Fifty cotyledon or hypocotyl explants were precultured on sterile filter paper (Whatman Ltd. qualitative) overlaying 0.5 ml of a 1 week old Nicotiana tabacum nonregenerable suspension culture, which was overlaying the preculture medium. Preculture medium consisted of MS salts, t00 mg/1 inositol, 1.3 rag/1 thiamine-HC1, 200 rag/1 Kk-I2PO4, 3% sucrose (w/v), 0.2 mg/l 2,4-D, 0,1 mg/1 6-furfurylaminopurine (KIN) and 8 g/1 bactoagar. For transformation experiments, Agrobacterium tumefaciens strain LBA4404 containing binary vector pCGN7001 (Comai et al. 1990), which included the kan r and GUS genes, was used (Fig.l). EcoRI
EcoRI
Figure 1. Map of pCGN7001 T-DNA. (LB - left border;, 35S cauliflower mosaic vires 35S promoter; kan r - kanamycin resistance gene; tr7 3' -polyadenylation site of transcript 7 of Agrobacterium tumefaciens T-DNA; mrs - mannopine synthase of Agrobacterium tumefaciens; GUS - g-glucnronidase gene; RB - right border) Single bacterial colonies were selected and cultured at 30~ ovemight in 5 ml of a medium consisting of 50% Luria broth and 50% of a mannitol-glutamate-salts medium (MG/L), (Gaffinkel and Nester 1980) In an incubator using a rotary shaker at 200 rpm, and diluted with MG/L to 5x108 bacteria/ml just prior to use. The precultured explants were soaked in the diluted bacterium culture for 5 min, drained on sterile filter paper and returned to the same plates containing preculture media. After 48 h, the explants were plated on regeneration medium supplemented with 0, 100, 150, 200, or 250 mg/1 kanamycin. Regeneration medium was composed of MS salts, 2% sucrose, 100 rag/1 myo4nositol, Nitsch vitamins (Nitsch and Nitsch 1969), 500 mg/l carbenicillin, 2 mg/l zeatin and 5 g/1 AGARGELTM, Ten explants were cultured per petri plate and each treatment had 5 petri plates making a total of 50 explants/treatment. The explants were incubated on the selection medium for an initial 10 d period and subsequently transferred to similar fresh media every 3 weeks. After 5 weeks of culture on the selection media, the calli which had developed at the cut ends of the explains were excised and cultured on their respective regeneration media. Shoots (2-5 cm high) which developed from the calli were similarly harvested and placed into GA7 (Magenta Corp.) boxes containing a rooting medium composed of MS, supplemented with 2% sucrose, 1 mg/1 indole-3butyric acid (IBA), 50 rag/1 carbenicillin and 100 mg/1 kanamycin. Shoots which rooted readily on kanamycin supplemented rooting medium were regarded as putative transformants and further analyzed to confirm transformation. For regeneration experiments, explants were prepared as above but with the cocultivation step eliminated, and explants were placed on regeneration medium without kanamycin.
Plant growth. Rooted shoots were transferred to the greenhouse and grown m 2 gallon pots containing (3:2:1) peat moss, vermiculite and sand. Minimum temperature was 24~ and maximum temperature was 30~ Minimum daylength was 14 h. Flowers were allowed to self-pollinate for seed collection. GUS and NPTtl assays. Freshly harvested leaf tissue samples (50-100 rag) were assayed by the NPTII dot blot enzyme assay (Radke et aL 1988) to confirm transformation. Reaction samples which produced reactions that were darker in intensity than the control samples without kanarnycin were rated positive for the NPTII gene.
T 1 seedlings were germinated on kanamycin for progeny analysis (details presented below) and roots were assayed for GUS activ-Jty using the histochemical assay (Jefferson 1987).
DNA isolation and Southern hybridization. Genomic DNA was isolated from young leaves of T 1 seedlings. The method used was defiveA from Bernatsky and Tank~ley (t986). Five to 7 leaves from each plant were harvested, immediately frozen in liquid nitrogen and ground in a motar. Fifteen mls of extraction buffer (350 mM sorbitol, 100 mM tris, 5 mM ethylenediaminetetraacetic acid (EDTA), 32 mM sodium bisulfate, pH 7.5) was added, the tissue homgenized with a polytron and centrifuged for 15 rain at 10,000 rpm. The pellet was redissolved in 4 rnl of extraction buffer and the nuclei were lysed by the addition of 4 ml of nuclear lysis buffer (200 rnM tris(hydoxymethyl)aminomethane, 50 mM EDTA, 2 M sodium chloride, 55 mM cetyltrimethylammonium bromide, pH 7.5) and 1.6
ml of 5% N-lauroylsarcosinesodiumsalt, incubatingat 65~ for 20 min. The mixture was then extracted with chloroform/isoamyl alcohol (24:1) and the nucleic acids precipitated with an equal volume of cold isopropanol. The nucleic acids were redissolved Jn 10 mM tris(hydoxymethyl)aminomethane, 1 mM EDTA, pH 8.0 and the RNA removed by a 15 rain room temperature incubation with RNase at 0.2 mg/ml. DNA was further purified with a phenol/chloroform extraction and an ethanol precipitation. Ten mg of genornic DNA was digested with EcoRI and fragments were separated by electrophoresis on a 0.7% agarose gel. DNA was transferred to a nylon membrane (Southern 1975). The 1.0 kb fragment of the NPTII gene was labeled with 32p using a Random Prime kit (Stratagene).
Genetic analysis of transgenic progeny. Inheritance of the NPTII gene was determined by segregation of the kanamycin resistance trait. Seeds of three T O plants and a non-transformed control were weighed out and surface sterilized as described above. The sterilized seeds were placed in 15 X 100 mm petri plates lined with sterile Whatman filter paper (Whatman Ltd. qualitative) and moistened with 7 ml of 1/10 MS salts solution supplemented with 100 mg/1 kanamycin sulphate. The plates were sealed with parafihn and cultured at 23~ under a 16 h fight/8 h dark photoperiod. Seedlings were scored by root morphology after 14 d of incubation. Stunted, tan-colored roots with no lateral branches were seen on non-transformed seedlings, while transformed seedlings exhibited white roots with lateral branching that were comparable in length to the seedlings grown on media which did not contain kanamycin.
Results
and Discussion
Influence ofexplanttype on regeneragon Non-cocultivated explants of both hypocotyls and cotyledons showed expansion of tissue after 10 d of culture. Within 5 weeks of culture, calli and shoot primordia had developed on both types of explants. All of the hypocotyl explants produced shoot primordia that went on to develop into shoots. While all of the cotyledonary explants also developed calli, they were slower and less efficient for regeneration than hypocotyls, with only 30% of the cotyledonary explants producing shoot primordia at 5 weeks. At 8 weeks of culture, 100% of the hypoeotyl explants had produced fully developed shoots while only 60% of the cotyledon explants had produced shoots. When non-cocultivated hypocotyl and cotyledon explants were placed on regeneration media supplemented with various concentrations of kanamycin, response was again dependent on explant type. Both explant types showed tissue expansion and bleaching. None of the control cotyledon explants placed on regeneration media supplemented with kanamycin from 100 mg/1 to 250 mg]l produced any calli or shoot primordia. Control hypocotyt explants developed small calli and shoot primordia at the rate of 2% on 100 rag/l, 6% on 200 rag/1 and 2% on 250 rag/1 kanamycin. None of the shoots produced from these explants produced roots when
584 the shoots were placed on rooting media supplemented with 100mg/l kanamycin. Thus, the non-cocultivated hypocotyl explants produced shoots which were escapes on regeneration media supplemented with kanamycin, indicating the potential for very fast growing tissue to develop non-transformed shoots.
Table 2. Transformation frequency of S. gilo as influenced by explant type and level of kanamycin. Explant type
Kanamycin (mgd)
cotyledon
100 150 200 250
Regeneration of transformed shoots. Cocultivated explants of hypocotyls and cotyledons showed the same degree of tissue expansion as seen on non-cocultivated explants. However, less bleaching was seen in cocultivated explants than non-cocultivated explants. Differences in explant response between hypocotyls and cotyledons were apparent by 5 weeks in culture. Calli initiation from cotyledon explants was approximately 2X higher as compared to hypocotyls at all levels of kanamycin selection (Table 1). Table 1. Influence of cocultivated S. gilo explant type on frequency of calli and shoot primordia development using different levels of kanamycin for selection. Explant type
Kanamycin (rag/l)
No. of explants
No. of explants with calli @ 5 wks
Cotyledon
0 100 150 200 250
10 50 50 50 50
lO (loo)
0 100 150 200 250
10 50 50 50 50
(%)
Hypocotyl
No. of explants with shoot prirnodia @
5 wks (%) 3 (30)
46 (92)
40 (80)
45 (90)
41 (82)
38 (76) 29 (58)
34 (68) 22 (44)
10 19 13 16 15
10 (100) 13 (26) 9 (18) 11 (22) 11 (22)
(100) (38) (26) (32) (30)
At 6 weeks after cocultivation cotyledon explants had produced putatively transformed shoots that were harvested and placed on rooting media supplemented with 100 mg/l kanamycin. As mentioned earlier, no shoots from non-cocultivated explants rooted on 100 mg/l kanamycin. Shoot harvests for cocultivated hypocotyls started at 9 weeks after cocultivation. The number of shoot primordia and shoot production were generally better using cocultivated cotyledon explants than for hypocotyl explants. Hence, even though the noncocultivated hypocotyl explants regenerated faster than cotyledon explants, the latter appeared to be a better choice for cocultivation and production of transformed shoots. For each explant type, four levels of kanamycin selection were evaluated. For cotyledon explants, the 100 and 150 mg/1 kanamycin levels produced the highest transformation efficiencies [# explants producing transformed shoots / # explants cocultivated) X 100] at 46% and 42%, respectively (Table 2). Culture of cotyledon explants on 200 and 250 mg/1 kanamycin decreased transformation efficiency by half to 22%. Hypocotyl explants had the highest transformation efficiency on 200 and 250 mg/l kanamycin (32% and 30%). Lower concentrations of kanamycin decreased transformation efficiency of hypocotyl explants to 22% for 150 mg/1 kanamycin and 16% for 100 mg/l
No. of No. of explants explants cocultivated producing + shoots a 5O 23 50 21 50 11 50 11
Transformation frequency (%) 46 42 22 22
hypocotyl
100 50 8 150 50 11 200 50 16 250 50 15 a Shoots which rooted on 100 mg/1 kanamycin
16 22 32 30
kanamycin. Thus, the two explant types responded differently to the levels of kanamycin used for selection of transformed shoots.
Confirmation of transformed shoots A subsample of the shoots rooted on 100 mg/l kanamycin were subjected to a NPTII dot blot assay to gain evidence of transformation. All shoots tested that rooted on kanamycin were also positive for the NPTII activity (Fig. 2; darker reactions in row A compared to row B are positive). 1
2
3
4
5
6
7
8
9
10
A B Figure 2. Neomycin phosphotransferase (NPT II) dot blot assay of leaf samples from kanamycin selected Solanum gilo transformed with pCGN7001 (Lanes 2-10) and a negative control (Lane 1). Row A includes kananaycin in the reaction buffer and row B does not include kanamycin.
Four T O shoots (#1-4) that rooted on 100 mg/l kanamycin were transferred to the greenhouse, grown to maturity, and fruits were harvested at the red, ripe stage for selfed-seed. T1 seedlings germinated on kanamycin supplemented 1/10 MS salts exhibited two distinctive root morphologies. One set had stunted, tan-colored roots with no lateral branches, similar to the morphology of non-transformed seedlings germinated on kanmnycin. The other set had white roots with lateral branching that were comparable in length to the morphology seen by control seedlings germinated without kanamycin. This morphology was exhibited by the majority of the seedlings. It was concluded that seedlings exhibiting the stunted, non branching root morphology were kanamycin sensitive and the seedlings exhibiting the normal branching and length were kanamycin resistant. Ten samples of each root type were subjected to GUS histochemical analysis. The stunted roots, lacking lateral branches showed no expression of GUS, similar to conlrols, while the branched roots showed expression of GUS (Fig. 3). Thus, GUS expression was correlated with the root phenotype produced on kanamycin and provided further evidence of transformation.
585
A
B
Figure 3. Root morphology and GUS expression in seedlings germinated on 100 mg/1 kanamycin, a) Kanamycin sensitive seedling. b) Kanamycin resistant seedling.
For segregation analysis, seedlings were sorted according to root morphology (Table 3). The progeny of the three transformation events exhibited a 3 resistant: 1 sensitive segregation ratio, indicating one locus of insertion. Table 3. Segregation of selfed seedlings for kanamycin resistance in progeny of transgenic S. gilo.
Plant
rooting phenotype kanr kans
Segregation ratio
Chi square p values
(p:0.05)
a
114 30
3tol
1.33
b
122 39
3to 1
0.05
c
168 53
3 to 1
0.12
Southern Blot Analysis. For further confirmation of transformation, a Southern blot (Southern, 1975) was performed on three TI plants (#1-3) obtained from three different TO lines. Genomic DNA was isolated from leaf tissue and digested with EcoRI. An EcoRI digest of pCGN7001 yields an insert of 3.87 kb containing the kanR gene, the GUS gene and the mas 5' region. The Southern was probed with a 1.0kb kanR probe. As shown in Figure 4, a band approximately 3.9 kb, was present in the three T 1 plant DNA samples but not in control tissue. Subsequently, the Southern was stripped of the kan probe and reprobed with a 0.384 kb gentarnycin probe. The three T 1 plants samples each had different sized border fragments and all of the border fragments were greater than 4.03 kb (data not shown). These results indicate the T-DNA was integrated into different genomic sites in the three T1 lines. In conclusion, an efficient Agrobacterium system for production of transgenic plants from Solanum gilo, garden egg, has been established. For the purpose of regeneration of transformed plants, cotyledon explants were superior to hypocotyl explants. Kanarnycin selection levels needed to be varied when using different explant sources. Cotyledonary explants gave the best
Figure 4. Southern blot analysis of genomic DNA (10 mg~ from transformed plants of Solanum gilo probed with the kan ~ gene (EcoRI fragment of pCGN7001). Lane h EcoRI digest from a nontransformed plant; lanes 2-4: EcoRI digests from the T 1 progeny from 3 independent TOplants.
transformation efficiencies at 100 and 150 mg/1 kanamycin. This protocol can now be used for transformation of garden egg with genes for improved agronomic ~raits and fruit quality. Acknowledgments. We wish to thank Dr. Corned Bonsi of Tuskegee University for the kind donation of the Solanum gilo seed. We also would like to thank Dr. Christine K. Shewmaker and Dr. William R. Hiatt for helpful discussion and critical reading of the manuscript.
References Arora RK, Hardas MW (1976) J Bombay Nat Hist Soc 73:423-424 Bernatsky R, Tanksley SD (1986) Gene 19:327-336 Comai L, Moran P, Maslyar D (1990) Plant Mol Biol 15:373-381 Delannay X, La Valle BJ, Prokseh RK, Sims SR, Greenplate JT, Fisehoff DA (1989) Bio/Technology 7:1265-1269 Filipone E, Lurquin, PF (1989) Plant CellRep 8:370-373 Garfinkel DJ, Nester EW (1980) J Bacteriol 144:732-743 Golds TJ, Lee JY, Ghose, TK, Davey MR (1991) J Exp Bot 42:11471157 Hiatt WR, Kramer M, Sheehy RE (1989) In: Setlow J (ed) Genetic engineering vol 11. Plenum Publishing Corporation, pp 49-63 Jefferson RA, Kavanagh TA, Bevan MW (1987) EMBO J 16:29012907 Klee HJ, Hayford MB, Kretzmer KA, Barry GF, Kishore GM (1991) Hant Cell 3:1187-1193 Mukhopadhyay A, Arumugam N, Nandakumar PBA, Pradhan AK, Gupta V, Pental D (1992) Hant Cell Rep 11:506-513 Murashige T, Skoog F (1962) Physiol Hant 15:473-497 Nitseh JP, Nitsch C (1969) Science 169:85-87 Oeller PW, Min-Wong L, Taylor LP, Pike DA, Theologis A (1991) Science 254:43%439 Radke SE, Andrew BM, Moloney MM, Crouch ML, Kridil JC, Knanf VC (1988) Theor Appl Genet 75:685-694 Rotino GL, Gleddie S (1990) Plant Cell Rep 9:26-29 Rotino GL, Perrone D, Ajamone-Marsan P, Lupotto E (1992) Plant Cell Rep 11:11-15 Sanders RA, Sheehy RE, Martineau B (1992) Plant Mol Biol Rep 10:164-172 Southern EM (1975) J Mol Bio198:503-517 van-Wordragen MF, Dons HJM (1992) Plant Mol Biol Rep 10:12-36
Plant Cell Reports (1996) 15: 582-585
9 Springer-Verlag 1996
Agrobacterium tumefaciens-mediated transformation of Solanum gilo Raddi as influenced by explant type Essie Blay 1 and Janette V. Oakes 2 1 Crop Science Dept., University of Ghana, Legon, Accra, Ghana 2 Calgene Fresh, Inc., 1920 Fifth St., Davis, CA 95616 USA Received 7 April 1995/Revised version received 25 September 1995 - Communicated by G. C. Phillips
Abstract. An efficient system for Agrobacterium tumefaciens-mediated transformation of Solanum gilo was established. The marker genes for kanamycin resistance and 13-glucuronidase expression were introduced. A comparison between cotyledon and hypocotyl explants showed that while regeneration was better from hypocotyl explants, cotyledon explants gave better transformation efficiency (46% vs. 32%). Four levels of kanamycin selection (100, 150, 200 and 250 mg/1) were tested for effect on transformation efficiency with each type of explant. Lower levels of kanamycin worked better using cotyledon explants, while higher levels of kanamycin worked better for hypocotyl explants. All nine T O plants tested for expression of the kan r gene were positive. The progeny of three of these plants showed a pattern of classical Mendelian inheritance (3 to 1) for both the kan r and the 13glucuronidase genes. Key Words: Transformation
Garden egg - Solanum
gilo-
Abbreviations: MS, Murashige and Skoog (1962) medium; 2,4-D, 2,4-Dichlorophenoxyacetic acid; Nt~II, neomycin phosphotransferase; GUS, 13-glucuronidase.
Introduction
The garden egg (Solanum gilo), a member of the solanaceous family, is an important vegetable in East, West, and Central Africa. The crop has also been reported in Asia (Arora and Hardas 1976). The fruit is consumed as it turns from green to cream with the yellow, orange and red stages rejected by the consumer as overripe. The major production constraints on the crop are various insect pests, which attack the stems, leaves, flowers and fruits, and precocious fruit-ripening. The latter is frequently associated with infestation by fruitborer (Leucinoides orbonalis) and results in overripe fruit and a drastic reduction in consumer acceptance. Correspondence to: J. V. Oakes
Even though the closely related wild species, S. integrifolium, has been reported as a possible source of pest and disease resistance for improvement of eggplant (Solanum melongena), gene introgression into crop plants from wild species frequently involves difficult crosses and lengthy selection owing to deleterious gene linkages. Introduction of single gene traits for insect resistance (Delannay et al. 1989), or ethylene regulation in conjunction with antisense technology to control ripening (Hiatt et al. 1989; Klee et al. 1991; Oeller et al. 1991), could be used to achieve relatively quick improvements in garden egg. However, the first requirement is the development of an efficient in vitro transformation system for this crop. Related species of garden egg, including eggplant, have been successfully transformed with different chimeric genes (Fifippone and Lurquin 1989; Rotino and Gleddie 1990; Rotino et al. 1992). However, crop transformation is influenced by several parameters including genotype, explant source and type, Agrobacterium strain, type of plasmid vector, cultural conditions and composition of the media (Golds et al. 1991; Mukhopadhyay 1992; van Wordragen and Dons 1992). Hence, it is frequently necessary to develop a transformation system for each species and sometimes different genotypes within a species. This paper reports high frequency regeneration of transgenic garden egg plants from cotyledon and hypocotyl explants via Agrobacterium co-cultivation and discusses the effects of explant type and selection level on transformation efficiency. Transformed plants and their progeny were analyzed using root formation on kanamycin, the NPTII dot blot assay, the 13glucuronidase (GUS) histochemical assay, and Southern blot analysis. Materials and Methods Plant material and seed germination. Seeds of S. gilo were obtained
from the germplasm collection of Dr. Corned Bonsi, Tuskegee University. S e e d s w e r e surface sterilized by dipping in 95% ethanol for 1 min, followed by 30 min in 50% Clorox (commercial NaOC1) containing 2 drops of Tween 20 per 100 ml. Seed were subsequently
583 rinsed 5 times in sterile, deionized water. Fifty seeds per plate were germinated asceptically in 100 X 15 mm petri dishes containing 40 ml of seed germination medium composed of MS salts (Murashige and Skoog 1962) supplemented with 0.8 mg/l thiamine hydrochloride, 0.4 rag/1 g]ycine and 3.5 g/1 AGARGELTM (Sigma Chemical Co.). The pH of the medium was adjusted to 5.8, prior to autoclaving at 121~ for 20 vain. Seeds and all in vitro plant material were incubated at 23~ under a 16 h fight and 8 h dark photoperiod. Light was provided by a 2 to 1 mix of cool white and gro-lux fluorescent lamps with an intensity of 50 mmol m -2 s-I.
Transformation, selection and regeneration procedures. For both transformation and regeneration procedures explants were prepared as follows. Cotyledons from 6 d old seedlings were excised and both ends were trimmed to generate rectangular explants approximately 3 mm in length. The hypocotyls were similarly sectioned into approximately 3 nun pieces. The apical as well as lower 5 mm pieces of the hypocotyls were discarded. Fifty cotyledon or hypocotyl explants were precultured on sterile filter paper (Whatman Ltd. qualitative) overlaying 0.5 ml of a 1 week old Nicotiana tabacum nonregenerable suspension culture, which was overlaying the preculture medium. Preculture medium consisted of MS salts, t00 mg/1 inositol, 1.3 rag/1 thiamine-HC1, 200 rag/1 Kk-I2PO4, 3% sucrose (w/v), 0.2 mg/l 2,4-D, 0,1 mg/1 6-furfurylaminopurine (KIN) and 8 g/1 bactoagar. For transformation experiments, Agrobacterium tumefaciens strain LBA4404 containing binary vector pCGN7001 (Comai et al. 1990), which included the kan r and GUS genes, was used (Fig.l). EcoRI
EcoRI
Figure 1. Map of pCGN7001 T-DNA. (LB - left border;, 35S cauliflower mosaic vires 35S promoter; kan r - kanamycin resistance gene; tr7 3' -polyadenylation site of transcript 7 of Agrobacterium tumefaciens T-DNA; mrs - mannopine synthase of Agrobacterium tumefaciens; GUS - g-glucnronidase gene; RB - right border) Single bacterial colonies were selected and cultured at 30~ ovemight in 5 ml of a medium consisting of 50% Luria broth and 50% of a mannitol-glutamate-salts medium (MG/L), (Gaffinkel and Nester 1980) In an incubator using a rotary shaker at 200 rpm, and diluted with MG/L to 5x108 bacteria/ml just prior to use. The precultured explants were soaked in the diluted bacterium culture for 5 min, drained on sterile filter paper and returned to the same plates containing preculture media. After 48 h, the explants were plated on regeneration medium supplemented with 0, 100, 150, 200, or 250 mg/1 kanamycin. Regeneration medium was composed of MS salts, 2% sucrose, 100 rag/1 myo4nositol, Nitsch vitamins (Nitsch and Nitsch 1969), 500 mg/l carbenicillin, 2 mg/l zeatin and 5 g/1 AGARGELTM, Ten explants were cultured per petri plate and each treatment had 5 petri plates making a total of 50 explants/treatment. The explants were incubated on the selection medium for an initial 10 d period and subsequently transferred to similar fresh media every 3 weeks. After 5 weeks of culture on the selection media, the calli which had developed at the cut ends of the explains were excised and cultured on their respective regeneration media. Shoots (2-5 cm high) which developed from the calli were similarly harvested and placed into GA7 (Magenta Corp.) boxes containing a rooting medium composed of MS, supplemented with 2% sucrose, 1 mg/1 indole-3butyric acid (IBA), 50 rag/1 carbenicillin and 100 mg/1 kanamycin. Shoots which rooted readily on kanamycin supplemented rooting medium were regarded as putative transformants and further analyzed to confirm transformation. For regeneration experiments, explants were prepared as above but with the cocultivation step eliminated, and explants were placed on regeneration medium without kanamycin.
Plant growth. Rooted shoots were transferred to the greenhouse and grown m 2 gallon pots containing (3:2:1) peat moss, vermiculite and sand. Minimum temperature was 24~ and maximum temperature was 30~ Minimum daylength was 14 h. Flowers were allowed to self-pollinate for seed collection. GUS and NPTtl assays. Freshly harvested leaf tissue samples (50-100 rag) were assayed by the NPTII dot blot enzyme assay (Radke et aL 1988) to confirm transformation. Reaction samples which produced reactions that were darker in intensity than the control samples without kanarnycin were rated positive for the NPTII gene.
T 1 seedlings were germinated on kanamycin for progeny analysis (details presented below) and roots were assayed for GUS activ-Jty using the histochemical assay (Jefferson 1987).
DNA isolation and Southern hybridization. Genomic DNA was isolated from young leaves of T 1 seedlings. The method used was defiveA from Bernatsky and Tank~ley (t986). Five to 7 leaves from each plant were harvested, immediately frozen in liquid nitrogen and ground in a motar. Fifteen mls of extraction buffer (350 mM sorbitol, 100 mM tris, 5 mM ethylenediaminetetraacetic acid (EDTA), 32 mM sodium bisulfate, pH 7.5) was added, the tissue homgenized with a polytron and centrifuged for 15 rain at 10,000 rpm. The pellet was redissolved in 4 rnl of extraction buffer and the nuclei were lysed by the addition of 4 ml of nuclear lysis buffer (200 rnM tris(hydoxymethyl)aminomethane, 50 mM EDTA, 2 M sodium chloride, 55 mM cetyltrimethylammonium bromide, pH 7.5) and 1.6
ml of 5% N-lauroylsarcosinesodiumsalt, incubatingat 65~ for 20 min. The mixture was then extracted with chloroform/isoamyl alcohol (24:1) and the nucleic acids precipitated with an equal volume of cold isopropanol. The nucleic acids were redissolved Jn 10 mM tris(hydoxymethyl)aminomethane, 1 mM EDTA, pH 8.0 and the RNA removed by a 15 rain room temperature incubation with RNase at 0.2 mg/ml. DNA was further purified with a phenol/chloroform extraction and an ethanol precipitation. Ten mg of genornic DNA was digested with EcoRI and fragments were separated by electrophoresis on a 0.7% agarose gel. DNA was transferred to a nylon membrane (Southern 1975). The 1.0 kb fragment of the NPTII gene was labeled with 32p using a Random Prime kit (Stratagene).
Genetic analysis of transgenic progeny. Inheritance of the NPTII gene was determined by segregation of the kanamycin resistance trait. Seeds of three T O plants and a non-transformed control were weighed out and surface sterilized as described above. The sterilized seeds were placed in 15 X 100 mm petri plates lined with sterile Whatman filter paper (Whatman Ltd. qualitative) and moistened with 7 ml of 1/10 MS salts solution supplemented with 100 mg/1 kanamycin sulphate. The plates were sealed with parafihn and cultured at 23~ under a 16 h fight/8 h dark photoperiod. Seedlings were scored by root morphology after 14 d of incubation. Stunted, tan-colored roots with no lateral branches were seen on non-transformed seedlings, while transformed seedlings exhibited white roots with lateral branching that were comparable in length to the seedlings grown on media which did not contain kanamycin.
Results
and Discussion
Influence ofexplanttype on regeneragon Non-cocultivated explants of both hypocotyls and cotyledons showed expansion of tissue after 10 d of culture. Within 5 weeks of culture, calli and shoot primordia had developed on both types of explants. All of the hypocotyl explants produced shoot primordia that went on to develop into shoots. While all of the cotyledonary explants also developed calli, they were slower and less efficient for regeneration than hypocotyls, with only 30% of the cotyledonary explants producing shoot primordia at 5 weeks. At 8 weeks of culture, 100% of the hypoeotyl explants had produced fully developed shoots while only 60% of the cotyledon explants had produced shoots. When non-cocultivated hypocotyl and cotyledon explants were placed on regeneration media supplemented with various concentrations of kanamycin, response was again dependent on explant type. Both explant types showed tissue expansion and bleaching. None of the control cotyledon explants placed on regeneration media supplemented with kanamycin from 100 mg/1 to 250 mg]l produced any calli or shoot primordia. Control hypocotyt explants developed small calli and shoot primordia at the rate of 2% on 100 rag/l, 6% on 200 rag/1 and 2% on 250 rag/1 kanamycin. None of the shoots produced from these explants produced roots when
584 the shoots were placed on rooting media supplemented with 100mg/l kanamycin. Thus, the non-cocultivated hypocotyl explants produced shoots which were escapes on regeneration media supplemented with kanamycin, indicating the potential for very fast growing tissue to develop non-transformed shoots.
Table 2. Transformation frequency of S. gilo as influenced by explant type and level of kanamycin. Explant type
Kanamycin (mgd)
cotyledon
100 150 200 250
Regeneration of transformed shoots. Cocultivated explants of hypocotyls and cotyledons showed the same degree of tissue expansion as seen on non-cocultivated explants. However, less bleaching was seen in cocultivated explants than non-cocultivated explants. Differences in explant response between hypocotyls and cotyledons were apparent by 5 weeks in culture. Calli initiation from cotyledon explants was approximately 2X higher as compared to hypocotyls at all levels of kanamycin selection (Table 1). Table 1. Influence of cocultivated S. gilo explant type on frequency of calli and shoot primordia development using different levels of kanamycin for selection. Explant type
Kanamycin (rag/l)
No. of explants
No. of explants with calli @ 5 wks
Cotyledon
0 100 150 200 250
10 50 50 50 50
lO (loo)
0 100 150 200 250
10 50 50 50 50
(%)
Hypocotyl
No. of explants with shoot prirnodia @
5 wks (%) 3 (30)
46 (92)
40 (80)
45 (90)
41 (82)
38 (76) 29 (58)
34 (68) 22 (44)
10 19 13 16 15
10 (100) 13 (26) 9 (18) 11 (22) 11 (22)
(100) (38) (26) (32) (30)
At 6 weeks after cocultivation cotyledon explants had produced putatively transformed shoots that were harvested and placed on rooting media supplemented with 100 mg/l kanamycin. As mentioned earlier, no shoots from non-cocultivated explants rooted on 100 mg/l kanamycin. Shoot harvests for cocultivated hypocotyls started at 9 weeks after cocultivation. The number of shoot primordia and shoot production were generally better using cocultivated cotyledon explants than for hypocotyl explants. Hence, even though the noncocultivated hypocotyl explants regenerated faster than cotyledon explants, the latter appeared to be a better choice for cocultivation and production of transformed shoots. For each explant type, four levels of kanamycin selection were evaluated. For cotyledon explants, the 100 and 150 mg/1 kanamycin levels produced the highest transformation efficiencies [# explants producing transformed shoots / # explants cocultivated) X 100] at 46% and 42%, respectively (Table 2). Culture of cotyledon explants on 200 and 250 mg/1 kanamycin decreased transformation efficiency by half to 22%. Hypocotyl explants had the highest transformation efficiency on 200 and 250 mg/l kanamycin (32% and 30%). Lower concentrations of kanamycin decreased transformation efficiency of hypocotyl explants to 22% for 150 mg/1 kanamycin and 16% for 100 mg/l
No. of No. of explants explants cocultivated producing + shoots a 5O 23 50 21 50 11 50 11
Transformation frequency (%) 46 42 22 22
hypocotyl
100 50 8 150 50 11 200 50 16 250 50 15 a Shoots which rooted on 100 mg/1 kanamycin
16 22 32 30
kanamycin. Thus, the two explant types responded differently to the levels of kanamycin used for selection of transformed shoots.
Confirmation of transformed shoots A subsample of the shoots rooted on 100 mg/l kanamycin were subjected to a NPTII dot blot assay to gain evidence of transformation. All shoots tested that rooted on kanamycin were also positive for the NPTII activity (Fig. 2; darker reactions in row A compared to row B are positive). 1
2
3
4
5
6
7
8
9
10
A B Figure 2. Neomycin phosphotransferase (NPT II) dot blot assay of leaf samples from kanamycin selected Solanum gilo transformed with pCGN7001 (Lanes 2-10) and a negative control (Lane 1). Row A includes kananaycin in the reaction buffer and row B does not include kanamycin.
Four T O shoots (#1-4) that rooted on 100 mg/l kanamycin were transferred to the greenhouse, grown to maturity, and fruits were harvested at the red, ripe stage for selfed-seed. T1 seedlings germinated on kanamycin supplemented 1/10 MS salts exhibited two distinctive root morphologies. One set had stunted, tan-colored roots with no lateral branches, similar to the morphology of non-transformed seedlings germinated on kanmnycin. The other set had white roots with lateral branching that were comparable in length to the morphology seen by control seedlings germinated without kanamycin. This morphology was exhibited by the majority of the seedlings. It was concluded that seedlings exhibiting the stunted, non branching root morphology were kanamycin sensitive and the seedlings exhibiting the normal branching and length were kanamycin resistant. Ten samples of each root type were subjected to GUS histochemical analysis. The stunted roots, lacking lateral branches showed no expression of GUS, similar to conlrols, while the branched roots showed expression of GUS (Fig. 3). Thus, GUS expression was correlated with the root phenotype produced on kanamycin and provided further evidence of transformation.
585
A
B
Figure 3. Root morphology and GUS expression in seedlings germinated on 100 mg/1 kanamycin, a) Kanamycin sensitive seedling. b) Kanamycin resistant seedling.
For segregation analysis, seedlings were sorted according to root morphology (Table 3). The progeny of the three transformation events exhibited a 3 resistant: 1 sensitive segregation ratio, indicating one locus of insertion. Table 3. Segregation of selfed seedlings for kanamycin resistance in progeny of transgenic S. gilo.
Plant
rooting phenotype kanr kans
Segregation ratio
Chi square p values
(p:0.05)
a
114 30
3tol
1.33
b
122 39
3to 1
0.05
c
168 53
3 to 1
0.12
Southern Blot Analysis. For further confirmation of transformation, a Southern blot (Southern, 1975) was performed on three TI plants (#1-3) obtained from three different TO lines. Genomic DNA was isolated from leaf tissue and digested with EcoRI. An EcoRI digest of pCGN7001 yields an insert of 3.87 kb containing the kanR gene, the GUS gene and the mas 5' region. The Southern was probed with a 1.0kb kanR probe. As shown in Figure 4, a band approximately 3.9 kb, was present in the three T 1 plant DNA samples but not in control tissue. Subsequently, the Southern was stripped of the kan probe and reprobed with a 0.384 kb gentarnycin probe. The three T 1 plants samples each had different sized border fragments and all of the border fragments were greater than 4.03 kb (data not shown). These results indicate the T-DNA was integrated into different genomic sites in the three T1 lines. In conclusion, an efficient Agrobacterium system for production of transgenic plants from Solanum gilo, garden egg, has been established. For the purpose of regeneration of transformed plants, cotyledon explants were superior to hypocotyl explants. Kanarnycin selection levels needed to be varied when using different explant sources. Cotyledonary explants gave the best
Figure 4. Southern blot analysis of genomic DNA (10 mg~ from transformed plants of Solanum gilo probed with the kan ~ gene (EcoRI fragment of pCGN7001). Lane h EcoRI digest from a nontransformed plant; lanes 2-4: EcoRI digests from the T 1 progeny from 3 independent TOplants.
transformation efficiencies at 100 and 150 mg/1 kanamycin. This protocol can now be used for transformation of garden egg with genes for improved agronomic ~raits and fruit quality. Acknowledgments. We wish to thank Dr. Corned Bonsi of Tuskegee University for the kind donation of the Solanum gilo seed. We also would like to thank Dr. Christine K. Shewmaker and Dr. William R. Hiatt for helpful discussion and critical reading of the manuscript.
References Arora RK, Hardas MW (1976) J Bombay Nat Hist Soc 73:423-424 Bernatsky R, Tanksley SD (1986) Gene 19:327-336 Comai L, Moran P, Maslyar D (1990) Plant Mol Biol 15:373-381 Delannay X, La Valle BJ, Prokseh RK, Sims SR, Greenplate JT, Fisehoff DA (1989) Bio/Technology 7:1265-1269 Filipone E, Lurquin, PF (1989) Plant CellRep 8:370-373 Garfinkel DJ, Nester EW (1980) J Bacteriol 144:732-743 Golds TJ, Lee JY, Ghose, TK, Davey MR (1991) J Exp Bot 42:11471157 Hiatt WR, Kramer M, Sheehy RE (1989) In: Setlow J (ed) Genetic engineering vol 11. Plenum Publishing Corporation, pp 49-63 Jefferson RA, Kavanagh TA, Bevan MW (1987) EMBO J 16:29012907 Klee HJ, Hayford MB, Kretzmer KA, Barry GF, Kishore GM (1991) Hant Cell 3:1187-1193 Mukhopadhyay A, Arumugam N, Nandakumar PBA, Pradhan AK, Gupta V, Pental D (1992) Hant Cell Rep 11:506-513 Murashige T, Skoog F (1962) Physiol Hant 15:473-497 Nitseh JP, Nitsch C (1969) Science 169:85-87 Oeller PW, Min-Wong L, Taylor LP, Pike DA, Theologis A (1991) Science 254:43%439 Radke SE, Andrew BM, Moloney MM, Crouch ML, Kridil JC, Knanf VC (1988) Theor Appl Genet 75:685-694 Rotino GL, Gleddie S (1990) Plant Cell Rep 9:26-29 Rotino GL, Perrone D, Ajamone-Marsan P, Lupotto E (1992) Plant Cell Rep 11:11-15 Sanders RA, Sheehy RE, Martineau B (1992) Plant Mol Biol Rep 10:164-172 Southern EM (1975) J Mol Bio198:503-517 van-Wordragen MF, Dons HJM (1992) Plant Mol Biol Rep 10:12-36
