Plant Cell Reports

Plant Cell Reports (1995) 14:589-592

9 Springer-Verlag1995

Transformation of grape (Vitis vinifera L.) zygotic-derived somatic embryos and regeneration of transgenic plants R. Scorza ~, J.M. Cordts ~, D.W. Ramming 2, and R.L. Emershad z 1 USDA-ARS Appalachian Fruit Research Station, 45 Wiltshire Rd., Kearneysville,WV 25430, USA 2 USDA-ARS HCRL, 2021 South Peach Ave., Fresno, CA 93727, USA Received 9 September 1994/Revised version received 11 November 1994 - Communicatedby G.C. Phillips

Summary. Transgenic grape plants were regenerated from somatic embryos derived from immature zygotic embryos of seedless grape (Vitis vinifera L.) selections. Somatic embryos were bombarded twice with 1 ~tm gold particles using the Biolistic PDS-1000/He device (Bio-Rad Laboratories) and then exposed to Agrobacterium tumefaciens strain C58/Z707 containing the binary plasmid pGA482GG or pCGN7314. Following cocultivation, secondary embryos were allowed to proliferate on Emershad/Ramming proliferation (ERP) medium for 6 weeks before selection on ERP medium containing 20-40 ~lg/nd kanamycin (kan). Transgenie embryos were identified after 3-5 months under selection and allowed to germinate and develop into rooted plants on Woody Plato Medimn containing 1 l.tM 6-benzylaminopurine (BAP), 1.5% sucrose, 0.3% activated charcoal and 0.75% agar. Integration of the foreign genes into these grapevines was verified by growth in the presence of kan, positive GUS and PCR assays, and Southern analysis.

Intrmla~on In the United States, seedless grapes represent approximately 80 and 99% of the total table and raisin grape .l?roduction, respectively (California Table Grape tcommission 1993; California Raisin Advisory Board, personal comm.). A limited number of seedless cultivars make up this production, resulting in deficiencies in the supply o f seedless grapes at certain times. The development of new seedless grape culliw~rs is important not only to fill the seasomd production gaps, but also to provide cultivars with resistance to diseases and insects which cause economic losses and which require chemical control. The incorporation of beneficial genes into somatic embryos of zygotic origin in grapes and the development of transgenic plants could be used to etd, ance gemaplasm for grape breeding programs. Transgenic genotypes would be used as parents in developing new generations of elite seedlings carrying the transferred gene(s). Wlfile useful for developing improved grape breeding stock, gene transfer into zygote-derived explants would not allow breeders to

Correspondence to: R. Scorza

use gene transfer to improve established cultival~, wlfich is a goal of many research programs (Baribault et al. 1990, Clog et at. 1990, H~hert et al. 1993, Mullins et al. 1990, Stamp et at. 1990). The success of gene transfer in grape, whether to improve established cultivars or parental germplasm lines, is dependent upon efficient transformation and regeneration of transgenic plants. Using in-ovulo embryo culture, a regeneration system of proliferative somatic embryogenesis and plant development has been developed from zygotic embryos of stenospermic seedless grapes (Vitis vinifera L.) (Emershad and Ramming 1995). These somatic embryos have been shown to originate from individual epidemlal cells of larger embryos (Margosan et al. 1994). Directed gene transfer protocols targeting these cells could be used to recover transformed plants. There

have

been

several reports of transformation of cells, tissues, and organs of grape (Baribault et at. 1989, 1990, Guellec et al. 1990, Colby et at. 1991, Berres et al. 1992). Recently, Htbert et al. (1993) reported the application of particle bombardment to successfully transform embryogenic cell suspensions of 'Chancellor' (Vitis L., complex interspecific hybrid). Mullins et al. (1990) reported the first regeneration of transformed plants of the rootstock Rupestris St. George (Vitis rupestris Scheele).

Agrobacwrium-mediated

A

nov el

approach

to

promote

Agrobacterium-mediated transformation was reporled by Bidney et al. (1992). They used particle bombardment combined with Agrobacterium to transform tobacco leaves and sunflower meristems. Tlfis method of wounding to promote Agrobacterium-mediated transformation gave at least 100 fold more transformation events than the standard particle gun transformation protocol. To facilitate transformation and the recovery of transgenic plants in Vitis vinifera L., which has thus far proven to be a recalcitrant species, we combined our higldy proliferative system of somatic embryo regeneration (Emershad and Ramming 1995) with the lfighly effective paxticle-wounding/A, tumefaciens treatment (Bidney et al. 1992).

590

Figure 1. (A) Somatic embryos of grape on polycarbonate~membrane immediately prior to microprojectile bombardme.t; (B) grape somatic embryos after 6 weeks of culture on ERP medium with ka.amycin; (C) grape shoots ready for rooting produced from transgenic somatic embryos 12 weeks after microprojectile bombardment and cocultivation with engineered A. tumefaciens. Bars represent 1.0 cm. Matm:l.~ and

Methods

Plant Materials and Culture. Zygotic embryos (approximately 0.3-2.0 mm in length) from 3 genotypes of Vitis vinifera L. were used as the original explant sources for transformation experiments. These genotypes, 2-19-6, 72-659-2, attd 69-636-5M, are advanced lines of dark-fruited seedless table grapes from the USDA-Agricultural Research Service Horticultural Crops Research Laboratory grape breeding program. Zygotic embryos were cultured on Emershad/Ramming (ER) medium following the protocol of Emershad and Ramnting (1995). Briefly, ER medium consists of White's (1954) macroelements modified as follows: Ca(NO3)2"4H20 , 600 mr/I; KNO3, 160 mr/I; NH4NO3, 360 mr/! plus Norstog s (1973) microelements and vitamins plus glycine, 3 mr/I; casein hydrolysate, 50 mg/i; L-cysteine 1211.6 mg/I (10 raM) plus 0.3% activated charcoal, and 6% sucrose. The pH was adjusted to 6.0 with IN HC! or IN NaOH. ER medium stimulates somatic embryogenesis from zygotic embryos. Proliferation of the somatic embryos is carried out on ER proliferation (ERP) medium which differs from ER medium in the addition of I ~tM BAP and a lowering of the pH to 5.5. ERP medium induces secondary embryogene~sis providing a continuous supply of somatic embryos for experimental use. Cultures were grown at 27+1~ in the dark and subcultured every 3 weeks. Somatic embryo cultures used in the transformation experiments had been proliferated on ERP medi,nl for 5 months. Following cocultivation and selection on ERP medium, putatively transformed embryos were induced to germinate on Woody Plant Medium (Lloyd and McCown 1981) with !.5% (w/v) sucrose, I [.tM BAP, 0.3% (w/v) activated charcoal and 0.75% (w/v) Ultrapure agar (USB, Cleveland) following the protocol of Emershadand Ramming (1995).

Traraformation. Transformation experiments utilized the Biolistic PDS-1000/fle device (Bio-Rad Laboratories) in combination with exposure to engineered Agrobacterium tumefacienz. Approximately 100 embryos of each genotype were selected and placed onto a 25 mm polycarbonate membrane in the center of a 100 mm petri plate containing ERP medimn 24 h prior to bombardment (Fig. IA). Embryos were shot with gold particles (average diameter !.0 ~tm) following the general procedures of Sanford et al. (1991). Bombardment was carried out using 7585 kPa (!100 psi) rupture discs, 7 mm gap distance, 6 mm macrocarrier flight distance, 686-711 mm Hg (27-28 in H8) chamber vacuum, and 7.5 cm target distance (2 positions below the microcarrier launch assembly in the PDS-1000/He). All plates were bombarded twice. Within 2 h of bombardment, approximately 100 embryos of each genotype were cocultivated with A. tumefacienz containing either the engineered plasmid pGA482GG or pCGN7314 as described below and in Figure 2.

[[5-glucuronidase (GUS)] and ka/t [neomycin phosphotransferase II (NFI'II)] genes (Fig. 2).

Cocultivation and selection. A. tumefaciens cultures were grown overnight with vigorous agitation at 28~ in LB medium containing the selective antibiotics appropriate for each plasnfid. These cultures were ceutrlfuged (5000 x g, 10 rain) and resnspended in the same volume of a medium consisting of Murashige and Skoog (MS) salts (Murashige and Skoog 1962), 2% sucrose (w/v), 100 ttM acetosyrlngone and 1.0 mM betaine phosphate and shaken for approximately 6 It at 20~ before use. After bombardment, embryos were washed off the polycarbonate membrane with I/2-strength MS medium into a !15 ml 0.2 I.tm disposable filter unit. The wash medium was removed by applying a vacuum to the unit. Then l0 nd of the resuspended A. tumefaciens culture was added to immerse the embryos. After 15 to 20 min, the resuspension medium was removed by vacuum and the embryos, along with the filter membrane, were cut out of the unit and placed onto cocultivation medium (ERP medium containing 100 I.tM acetosyringone). Embryos were cocultlvated for 2 days and then washed at least 3 times each with 10 ml of liquid ERP medium (without charcoal) containing 300 [.tg/ml cefotaxime and 200 ~tg/ml carbenicillin. Enthryos were plated on seml-solld ERP medium (0.75% agar) with the stone levels of antibiotics. Cultures were allowed to proliferate for 2 passages (3 weeks each) before being placed onto selection medium. Selection was carried out on ERP medium with the addition of 20 ttg/ml kanamycin (kan) for the first 6 weeks and then 40 [.tg/ml kan for the next 6 weeks of proliferation (Fig. IB).

-

-

I

RB

o

I

I

I

I

I

Eco RI

Eco RI

I

I D3sa I o.s

I..3'1

Im"3' I II

N~T, .-.s'l

LB

)

2.0 kb

r.0 I~ pCOS7314

L__L I I BamlHI

Eco RJ

RB

I

"

I NPrP PCR fragment and probe

I ~~

3.

I

~us

I.o,

I

um

PCR fragment pGA482GG

Agrobaeterium strain and plasrnid deseriptions. The A. tumefaciens strain used was C58/Z707 (supplied by A. Hepburn) containing either plasmidpOA482GG (Slightom 1991, Slightom et al. 1991) (supplied by L Slightom) or pCGN7314 (Comai et al. 1990) (supplied by L. Comai). Both plasmids contained chimeric uidA

F i g u r e 2. Partial m a p s o f the T - D N A r e g i o n s o f p G A 4 8 2 G G and p C G N 7 3 1 4 . F o r c o m p l e t e m a p s see S l i g h t o m (1991) and C o m a i et al. (1990).

591Confirmation of transformation. Transformation of embryos and of shoots produced following embryo germination (Fig. IC) was assayed by growth and rooting on kan-containing medium and through a histological GUS assay (Jefferson 1987). Following rooting and transfer to the greenhouse, plants were subjected to PCR and Southern analyses. PCR amplification was conducted on DNA isolated front leaves of putatively transformed grape plants. Specific oligonucleetide printers for GUS and NOS/HPTII gone sequences were used to identify the presence of these genes in DNA from the different clones (Scorza et al. 1994). The primers for amplification of the Hlrl'll gone were designed to am~plify a !052 bp fragment extending from the NOS promoter st the 5 end of the gene to 150 bp beyond rite NOS 3' termination sequence (Chee et al. 1989). Since plasmid p ( ~ N 7 3 | 4 uses a MAS instead of a NOS promoter to control expression of the NPTII gone (Fig. 2), the primers were not expected to amplify the NVrlI gene in transformants developed using pCX3N7314. Primers for the gentandcin-(3)-N-acetyl transferase gene from the region of the plasmid outside of the border regions was used to verify the absence of A. tumefaciena on or within putatively transformed grape plants. Theprimer sequences for this gene were 5' primer, 5'-GGCAC'I~rGATGGGATACGCG-3' and 3' primer, 5'-AGAACGCTCGGAAGGCCAATC-3'. All PCR reactions were run using the GeneAmp kit components (Perkin-Elmer, Norwalk, CT) with the following cycle parameters: 1 min st 94~ 1.5 rain at 65~ and 2 min at 72~ The first cycle used an additional 3 rain melt at 95~ and the last 5 cycles had 4 rain extension times at 720C. After 35 amplification c~cles the PCR products were analyzed by agarose gel electrophoresm and stained with ethidium bromide.

Figure 3. PCR amplifed GUS and NOS/NPTII fragments from putative transgenic grapes. Plants G1-GI5 represent 8 of the 10 plants recovered following cocultivation with A. tumefacien$ containing the plasmid pGA482GG. Plants 73-1 and 73-2 were cocultivated with A. twnefaciens containing the plasmid pCGN7314. Lanes 1 and 2 are 1 kb and 123 bp ladders (BRL, Bethesda, MD), respectively.

Southern analysis was carried out using a PCR-generated !.! kb NOS/NVrll probe, Digestion with EcoRI was used to test for unique insertion events which would include segments of grape DNA in pGA482GG transforntants. Xbal was used to assay for the unique insertion events in pCGN7314 transformants. DNA extraction followed the procedures of Callahan et al. (1992). Conditions for Southem analysis were as described by Scorza et al. (1990). The 1.1 kb NOS/N]rFII probe was radioactively labeled using random primers according to the instructions with the BioRad Random Printer DNA Labeling Kit.

Results and Discn.,etlon Bombardment of grape embryos with gold particles followed by cocultivation with engineered A. tumefaciens produced transgenic plants. A total of 14 plants were produced which grew on selective medium. All were GUS positive. PCR analysis of these plants using GUS and NOS/NPT[I primers demonstrated that all contained the GUS gene. The NOS/NPTII sequences were amplified in the pGA482GG transformants. A portion of these assays are presented in Figure 3. As expected, plants transformed with pCGN7314 showed no amplification of NOS/NPTII sequences since the MAS instead of the NOS promoter is used in this plasmid. PCR amplification of the gentamicin-(3)-N-acetyl transferase gene gave negative results indicating no A. tumefaciens contamination of the regenerated plants. Of the 14 plants produced, 10 resulted from cocultivation with A. tumefaciens containing pGA482GG. Southern analysis showed identical banding patterns from all 10 plants following digestion with EcoRI. Digestion with HindHI gave the same results (data not presented). Tiffs suggests that these 10 plants arose from a single transformation event. This result is not surprising considering the highly proliferative nature of these embryos when cultured on ERP medium such that in a short time numerous secondary embryos are produced by each original (and in this case) transformed embryo. The 4 plants transformed with pCGN7314 all showed unique banding patterns (Fig. 4) when digested with Xbal, an enzyme with a single restriction site in the plasmid

Figure 4. Southem analysis of 7 putative transgenic grapes. Plants G1, GII and GI4 were transformed with A. tumefaciens containing plasmid pGA482GG. All show the same banding pattern when digested with EcoRI, which produces a single cut within the T-DNA. These plants were most likely the result of proliferation of secondary embryos from one original transformed embryo. Plants 73-1, 73-2, 73-3 and 73-4 were transformed with A. tumefaciens containing the plasmid pCGN7314 and show unique banding patterns when digested with Xbal, which produces a single cut within ttle T-DNA. The probe was the 1.1 kb NOS/NPTII fragment as indicated in Fig. 2. Controls are untransformed grape.

592 (Fig. 2). Plant 73-1 appears to have a single insertion event while plants 73-3 and 73-4 each have 2 bands indicating 2 insertions. Plant 73-2 has 3 clearly separate bands in a short exposure, 2 of which appear to blend together in the longer exposed blot presented in Figure 4. To date there have been few reports of transgenic grapevine production. Mullins et al. (1990) produced a transgenic V. rupestris plant after cocultivating 240 somatic euthryo hypocotyls with A. tumefaciens. Transformation of Cabemet Sauvignon and Chardotmay (240 explants each) w.as unsuccessful. Baribault et at. (1990) cocultivated fragmented shoot apices of Cabernet Sanvignon with engineered A. tumefaciens and obtained chimeric expression of the transgenes in 2-4% of the regenerated shoots. These shoots did not root in the presence of kanamycin even after 2 years of growth. Recently, Nakano et at. (1994) produced transgenic plants of V. vinifera cv Koshusanjaku through cocultivation of somatic embryos derived from leaf callus with A. rhizogenes. Although most plants exhibited an altered phenotype due to expression of the wild type Ri T-DNA or were otherwise shown to contain wild type Ri T-DNA, 2 transgenic plants were apparently free of this DNA. These 2 plants were selected from 45 somatic embryo clones which grew under kan selection. The total number of explants originally cultured was not stated. We feel tbat the method described in this report compares favorably with the few published reports of grape transformation. In fact, to our knowledge, it appears that there is only one report of transgenic Vitis vinifera plants having previously been produced (Nakano et al. 1994). Using the method described we produced transgenic plants from immature embryos of 2 of the 3 grape genotypes tested (2-19-6 and 72-659-2) and from both plasmids tested, which we found to be encouraging. Of approxhnately 300 somatic embryos bombarded (100 of each genotype) 5 distinct transgenic lines were produced, a transformation rate of 1.7%. Seven months following bombardment and A. tumefaciens infection, robust, rooted plants were transferred to the glasshouse and all have thrived. We produced transgenic grape plants when uncoated microprojectile bombardment was followed by cocultivation with A. tumefaciens. The effect of bombardment is unkalown since embryos were not exposed to A. tumefaciens without prior bombardment. In general, bombardment appears to provide a large number of micro-wounds that enhance the ability of the engineered A. tumefaciens to infect and transform tissues (Bidney et at. 1992). Comparisons of methods for transformation of grape were not undertaken in this study, rather, we selected what appeared to be a promising methodology based on published information (Bidney et at. 1992) and unpublished data from our laboratory to maximize the potential of obtaining transformants in what Ires been a recalcitrant species. Studies are continuing to further improve the rates of transformation in Vitis vinifera and to engineer agronomically useful genes into elite grape germplasm.

References Baribault TJ, Skene KGM, Scott NS (1989) Plant Cell Reports 8:137-140 Baribault TJ, Skene KGM, Cain PA, Scott INS (1990) J Expt But 41:1045-1049 Berres R, Often L, Tinland B, Malgarinin-ClogE, Walter B (1992) Plant Cell Rep 11:!92-195 Bidney D, Scelonge C, Martlch J, Burros M, Sims L, HuffmanG (1992) Plant Mol Biol 18:301-313 CaliforniaTable Grape Cormnission(1993) 1992-1993:The distribution and per capita consumption Of California table grapes by major varieties in the United States and Canada. PC_)Box 5498, Fresno, CA 93755 Callahan AM, MorgensPIt, Wright P, Nichols Jr ICE (1992) Plant Physiol 100:482-488. Chee PP, Fober KA, Slightom JL (1989) Plant Physiol 91:1212-1218 Clog E, Bass P, Walter B (1990) Plant Cell Pep 8:726-728 Colby SM, Juncosa AM, Meredith CP (1991) J Amer Soc Hurt Sci l !6:356-361 Comai L, Moran P, Maslyar D 0990) Plant Mol Biol 15:373-381 Emershad RL, Ramming DW (1994) Plant Cell Pep 14:6-12 Guellec V, David C, BranchardM, Tempe J (1990) Plant Cell Tiss Org Cult 20:211-215 Hdbert D, Kikkert JR, Smith FD, Reisch BI (1993) Plant Cell Rep 12:585-589 Jefferson RA (1987) Plant Mol Biol Pep 5:387-405 Lloyd G, MeCownBH (1981) proc Int Plant Prop Soc 30:421-427 Margosan DA, EmershadRL, Rammin~DW (1994) Inter Syrup Table Grape Production,Anahetm, Ca USA p 133-135 Mullins MG, 'FangFCA, Facciotti D (1990) Bio/l'echnology 8:1041-1045 Murashige T, Skoog G (1962) Physiol Plant 15:473-497 Nakano M, Hoshino Y, Mii M (! 994) J Expt But 45:649-656 Norstog K (1973) In Vitro 8:307-308 Sanford JC, Smith FD, Russell JA (1991) Meth Enzymol 217:483-509 Scorza R, MorgensPH, Cordts JM, Mante S, Callahan AM (1990) In Vitro Cell Devel Biol 26:829-834 Scorza R, ZimmermanTW, Cordts JM, Footen KJ, RavelonandroM (1994) J Amer Soc Hurt Sci 119:1091-1098 Sllghtom JL {1991)Gene 100:252-255 Slightom JL, Drong RF, Sieu LC, Chee PP (1991) In: SB Gelvin, RA Schilperoot and DPS Verma (Eds), Plant Molec Biol Man B I6, pp. 1-55, Kluwer Academic Pub, Dordrecht, The Netherlands Stamp JA, Colby SM, Meredith CP (1990) J Amer Soc HoLt Set 115:1038-1042 White PR (1954) The cultivation of animal and plant cells. Ronald Press, NY