Plant Cell Reports

Plant Cell Reports (1996) 15:311-316

9 Springer-Verlag1996

Transgenic plantlets of 'Chancellor' grapevine (Vitis sp.) from biolistic transformation of embryogenic cell suspensions Julie R. Kikkert 1, Dominique H6bert-Soul6 2, * Patricia G. Wallace 1, Michael J. Striem 1, and Bruce I. Reisch 1 1 Department of Horticultural Sciences, New York State Agricultural Experiment Station, Cornell University, Geneva, NY 14456, USA 2 Institut National de la Recherche Agronomique, Station de G~n&ique et d'Am~lioration des Plantes de Montpellier, Domaine de Melgueil, F-34130 Mauguio, France * Current address: Residence les Fleurs, bat. C, appt. 58, 63 Chemin de Pelleport, F-31500 Toulouse, France Received 10 August 1994/Revised version received 17 July 1995 - Communicated by G. C. Phillips

Abstract. Transgenic plantlets of 'Chancellor' grapevine (Vitis L. complex interspecific hybrid) were produced via biolistic transformation. Embryogenic cell suspensions were bombarded with 1 ~tm tungsten particles coated with pBI426 which encodes a fusion peptide between 13-glucuronidase (GUS) and neomycin phosphotransferase II (NPTII). The fusion peptide is under the control of a double 35S Cauliflower Mosaic Virus promoter and a leader sequence from Alfalfa Mosaic Virus. The cells were placed on kanamycin-containing media (I0, 25 or 50 mg/1) 2 d after bombardment. Activated charcoal reduced cell browning. Embryos were first observed on selective media 14-29 weeks after bombardment. More than 1600 clusters of embryos were germinated and/or assayed for GUS. Of 621 embryos assayed for GUS expression, 182 (29.3%) were positive. PCR confirmed the presence of the NPTII gene in all 5 GUS-positive and 2 GUS-negative (bombarded) embryos tested. In germination experiments, 15% of the embryo clusters produced at least one plant with normal shoot growth. Of 164 normal plants assayed for GUS expression, 37 (22.6%) were positive. The NPTII gene was amplified by PCR in 1 (of 1) GUS-positive and 4 (of 5) GUS-negative bombarded plants, but not in nonbombarded control plants. Southern blotting confirmed integration of the NPTII gene in all 3 of the GUS and PCR-NPTII positive plants tested. Biolistics is an efficient method for transformation of 'Chancellor' and should be applicable to other important grape cultivars. Abbreviations: AC, activated charcoal; GUS, 13-glucuronidase; 2,4-D, 2,4-dichlorophenoxyacetic acid; BA, 6-benzylaminopurine; NAA, ~t-naphthalene acetic acid; TDZ, thidiazuron; NPTII, neomycin phosphotransferase lI; Kin, kanamycin; MS, Murashige and Skoog (1962) medium; WPM, Woody Plant Medium of Lloyd and McCown (1980).

Introduction Grape is a major fruit crop throughout the temperate regions of the world and is subject to a number of diseases and pests (Bovey and Martelli 1986; Pearson and Correspondence to: J. R. Kikkert

Goheen 1988). Because grape is a vegetativelypropagated perennial crop with production and marketing rooted in traditional cultivars, genetic engineering is a very promising approach for the development of pest resistant cultivars. A wide array of potential resistance genes are available for use in grape including those that encode chitinases (Broglie et al. 1991; Lund and Dunsmuir 1992), lyric peptides (Jaynes et al. 1993), lysozymes (Trudel et al. 1992), viral coat proteins (Wilson 1993), and insect toxins (Vaeck et al. 1987). Genetic transformation of grapevines has been difficult, particularly for V. vinifera. Most attempts have utilized Agrobacterium-mediated transformation (Baribault et al. 1989, 1990; Berres et al. 1992; Colby et al. 1991; Guellec et al. 1990), however, recovery of whole plants has been rare. MuUins et al. (1990) obtained transgenic plants of the rootstock 'St. George' (V. rupestris Scheele) that expressed both Km-resistance and GUS. However, the same techniques applied to the scion cultivars (V. vinifera) 'Cabernet Sauvignon' and 'Chardonnay' did not yield plants. Scorza et al. (1995) have recovered GUSand Km-positive plants after A g r o b a c t e r i u m transformation of zygotic-derived somatic embryos of V. vinifera. Recently, there have been the following reports in the popular press of transgenic grape plants that express viral coat proteins: Bouquet (1993) recovered plants from embryogenic cultures of the 110 Richter rootstock (V. berlandieri X V. rupestris) that expressed the coat protein of Grape Chrome Mosaic Virus. The French wine company Mo~t et Chandon has transformed vines of the 41B rootstock (V. vinifera 'Chasselas' X V. berlandieri) (Deloire 1993), the SO4 rootstock (V. berlandieri X V. riparia), and the scion cultivar V. vinifera 'Chardonnay' (O. Goniak, 1994 press release) all with the coat protein from Grapevine Fanleaf Virus. W e have taken a different approach to grapevine transformation by using biolistics (the gene gun) to deliver genes into embryogenic suspension cultures. Previously, we reported long term GUS expression in calli of the interspecific hybrid scion cultivar 'Chancellor' (H6bert et al. 1993). Here, we report that such caUi can form GUS- and NPTII-positive embryos and plants.

312 Materials and Methods Plant material. Embryogenic suspension cultures of 'Chancellor' (Seibel 7053, a complex interspecific hybrid derived from V. vinifera, V. rupestris, V. lincccumii, V. riparia, and V. labrusca) were obtained from Dr. R.N. Goodman, University of Missouri-Columbia. Cultures were grown and prepared for bombardment as described by H6bert el al. (1993). The bombardment medium contained 0.125 M mannitol and 0.125 M sorbitol as om~xica (Russell el al. 1992). Piasmid DNA and microc~rrier coatin&. Plasmid pBI426 was obtained from Dr. W.L. Crosby, Plant Biotechnology Institute, Saskatchewan, Canada. This plasmid encodu a bifunotional fusictt peptide between Escherichia coli [3-glucuronidase ((]US) and neomycin phosphotransferase II (NPTI/), such that GUS expression and kanemycin (Kin) resistance can be simultaneously assayed (Datla et al. 1991). Bxpression of this fusion peptide is under the control of a double 35S Cauliflower Mosaic Vires promoter plus a leader sequen~ from Alfalfa Mosaic Virus. Plasmid DNA was CsCl purified and resuspended in TB buffer (Sambrcok el aL 1989). The DNA was complexed with 1 pan tungsten particles (MI0. Sylvania GTE Products Corp., Towanda, Pennsylvania, USA) using CaCl2/spermidine precipitation, and loaded onto macrocarrier disks, both as described by Kikkert (1993). Bombardmoa conditions. The biolistic device that was used is the prototype of the commercially available PDS-1000/He (BioRad, Hercules, California, USA) and is described by Sanford el al. (1991). The cells were placed 12 cm from the microcarrier launch point and were bombarded a single time using the following conditions: 1 can distance between the rapture disk and macrocarrier launch point, 1 cm macrocarrier flight distance, sample chamber evacuated to 0.1 arm, and gas acceleration tube pressurized with 1000 psi helium. Post-bom~rdm~nt cell handling. After bombardment. Petri plates (100 x 15 nun) containing the cells were placed, unsealed, in styrene boxes (Flambeau Products, Middiefield, Ohio, USA) and incubated at 24 :t_ 2 C in the dark. On the day after bombardment, the osmotic pressure of the medium was gradually reduced by overlaying the cells and the bombardment medium on top of medium without mannitol and sorbitol (H6bea el al. 1993; Russell el al. 1992). Selection for Kin-resistant cells began 2 d after bombardment. Km concentrations between 0 and 50 mg/l were tested. The medium was either ball-strength Murashige and Skoog (1/2 MS) with 2.0% (w/v) sucrose and without hormones to permit embryogenesis, or fullstrength MS medium with 3.0% (w/v) sucrose, 5.0 pM 2,4-D and 1.0 p.M BA to stimulate callus growth. Some media contained 0.3% (w/v) activated charcoal (AC). The initial selection media were solidified with 0.6% (w/v) Bactoagar (Difco, Detroit, Michigan). Approximately 3 months post-bombardment, 0.25% (w/v) Gelrite (sehweizeriudl, Inc., South Plainfield, New Jersey, USA) was used to solidify all selection media. Grape cells exhibit less browning on Gelrite-solidified medium than on agar-solidified medium (A. Perl, personal communication). During selection, plates were wrapped with Parafdm (American National Can, Greenwich, Connecticut, USA) and were incubated at 24 :t: 2 C in the dark. Embryos at different stages of develolanent were harvested individually or in dusters from the selection plates and were placed on 1/2 MS with 0.3% (w/v) AC, in 100 x 15 nun Petri plates. The germination medium was hormone-free and Kin-free. Germinated embryos and nomud plants were grown in baby food jars (72 nun tall, capped with Magenta B-ca~) containing MS or Woody Plant Medium (WPM) (Lloyd and McCown 1980) with different growth regulators as described in the Results and Discussion. These media were solidified with 0.3% (w/v) Bactoagar plus 0.2% (w/v) Geirite. Analysis of traasformants. Histochemical GUS assays were performed by incubating tissue in X-glue solution (McC_.abe el al. 1988) overnight at 37 C. GUS-positive tissues turned bright blue. There was no endogenous GUS activity in negative control tissue. Stable incorporation of the NFTII gene in embryo and plantlet tissue was assayed by DNA amplification using the Polymerase Chain Reaction (PCR) (Saiki et al. 1985), and by Southern Blot analysis (Soutbem 1975). DNA was extracted from grape tissues as described by Lodhi el al. (1994).

For PCR, 100 ng of plant DNA or 1 ng of pBI426 plasmid DNA was used per 25 FI reaction. The primen were expected to amplify a 757 bp fragment of the NPTR gene. The ampfification conditions consisted of an initial denaturation step of 94 C for 2 rain, followed by 30 cycles of 94 C for 30 s, 60 C for 2 rain, and 72 C for 1 min. The final extension was 72 C for 10 min. The amplification mixture was loaded directly onto 1% agarose and eleetrophoresed at 5.5 V/cm. Gels were stained with ethidium bromide. Southern analysis was performed on three GUS-, and PCR-NPTII positive plants, and one non-transformed plant. DNA was digested with 40-50 unitl of two endenucleases (EcoRI and HindIII, Promega Corp., Madison, Wisconsin, USA) at 37 C, and electrophoresed on 0.8% (w/v) agarose at 15mA. DNA was transferred to a MagnaGraph (MSI, Westborough, Massachusetts, USA) nylon membrane (Sambrook et al. 1989). The membrane was probed with a PCRamplified fragment of the NPTH gene that had been random-labelled with 32p (Feinberg and Vogelstein 1983) using a Prime-a-Came@ kit from Promega Corp. (Madison, Wisconsin, USA). Hybridization (65 C for 20 h) and washing were according to standard techniques (Sambrook el aL 1989). The membrane was subsequently exposed to Kodak (Rochester, New York, USA) XAR-5 X-ray f'dm with intensifying screens for 4 d at -70 C.

Results and Discussion

Km Inhibition of Embryogenesis To determine the appropriate concentration of Km for selection of transformants, non-bombarded cells were cultured on 1/2-MS with 0, 8, 10, 15, or 20 mg/l Km. The media were hormone-free to permit embryogenesis, and were either with or without AC. There were 3 replicate plates per treatment. Within 4 weeks, embryos began to emerge from ceils cultured on 0 and 8 mg/l Km (Fig. la). A final evaluation of the cells was performed 16 weeks after plating. Hundreds of embryos per plate formed on Km-free medium and less than 50 embryos per plate developed on 8 mg/l Kin. There were no embryos at Km concentrations of 10 mg/1 or higher. Callus growth was not inhibited at any of the Km concentrations tested. The results were the same with or without AC in the medium, indicating that AC does not interfere with selection.

Biolistic Experiments 1 and 2 The first two biolistic experiments were performed 1 week apart, but otherwise were identical in treatment and results. The initial selection medium was hormone-free 1/'2-MS to permit embryogenesis and contained 10 mg/l Km. Between the 2 experiments there was a total of 19 plates. Ten plates contained AC and 9 plates were ACfree. After 6-7 weeks of culture, embryos had not developed and the cells were transferred to a callus proliferation medium (MS with 5.0 ttM 2,4-D and 1.0 BA; with or without AC as before) to stimulate further growth of transformed cells. The Km concentration was increased to 25 and 50 mg/l (Table 1) to restrict growth of non-transformed calli. The cells were transferred to fresh medium every 6-8 weeks. Cells on medium without AC were maintained for 6 months post-bombardment, but embryos did not form (Table 1). Additionally, the cells turned dark brown/black, and there was a heavy accumulation of brown exudates on the f'dter papers supporting the cells

313

Fig. 1. Regeneration and GUS expression of embryos and plantlets from bombarded 'Chancellor' cultures: Embryos produced on non-selective medium (A); White globular embryos amongst brown calli on medium with 50 rag/1 kanamycin (B); GUS expression in germinated embryo with abnormal shoot growth (C); GUS expression in normal leaves: uncut leaf (D) showing poor substrate penetration, negative control sliced leaf (E), GUS-positive leaf sliced for better substrate penetration (F) - note: D and F are from the same plant; Normal rooted plantlet positive for GUS and NPTII (G). Scales for the photos are as follows: A) 1 cm=2 ram, B) I cra---0.5ram, C) 1 cm=41 ram, D) lcm=21 ram, E) 1 cm=43 nun, F) I cm--42 ram, G) I cm=10.4 nun.

314 and in the growth medium below. Thus, the AC-free plates were discarded. In contrast, cells on AC medium began producing embryos 14 to 15 weeks after bombardment. At this time there were a total of 27 embryos, from 4 of the 5 plates with 25 mg/1 Km and 1 of the 5 plates with 50 mg/l Km. The embryos were excised and germinated as described below. Even though the cells were on a callus proliferation medium, we believe that AC absorbed the hormones (Weatherhead et al. 1978) and permitted embryogenesis. AC did reduce the accumulation of brown exudates on the white filter paper supporting the cells, but the cells still turned brown (Fig. lb), although less so than on medium without AC. Thirty weeks after bombardment, all cells on AC medium were transferred to I/2-MS with AC and 50 mg/l Kin. All subsequent transfers (6-8 week intervals) were to this medium. Nine weeks later, 40 additional embryos were excised. At this time 4 of 5 plates previously on medium with AC and 25 mg/l Km and 5 of 5 plates on medium with AC and 50 mg/l Km had produced embryos. Additional embryos formed on all AC-containing plates over a 1 year period after bombardment. A total of 180 embryo clusters were obtained (Table 1). Embryos were germinated and assayed for gene expression as described below.

nature of the cultures and the stress of Km selection permitted embryogenesis. Cells on the original plates were subsequently transferred at 6-8 week intervals to 1/2MS with AC and 50 mg/1 Km. Embryos continued to form over a 1 year period after bombardment. A total of 1457 embryo clusters were obtained (Table 1), but from only 50% of the original bombarded plates from each treatment. Embryos were germinated as described below. Compared to experiments 1 and 2, embryo production in experiment 3 was initiated after a longer time on selective medium and in a smaller percentage of the plates. This is probably due to both the use of a callus proliferation medium and higher initial Km concentrations in experiment 3. Other experiments with very high Km concentrations (100-500 rag/l) that are more inhibitory to callus growth have not produced embryos after 1 year of culture even though calli are still alive (unpublished results). Additionally, experiment 3 demonstrates that cells on AC-free medium can form embryos, even though the calli were darker brown than those formed using AC.

Embryo Germination and Analysis Tables 1 and 2 summarize the results of embryo germination and assays for GUS expression and the NPTII gene. The initial group of 27 embryos from experiments 1 and 2 were transferred to Kin-free and hormone-free 1/2-MS with AC, and were incubated at 4 C for 9 d. Embryos were subsequently transferred to fresh medium and incubated at 24 C in the light. After 3 weeks, most embryos rooted but shoot growth was abnormal. All embryos from this batch were destructively assayed for GUS expression. Six embryos (22%) were solid blue (Fig. lc), 1 appeared chimeric, and 20 were not blue. A second group of embryos arising from experiments 1, 2, and 3 were placed on Km-free and hormone-free 1/2MS medium with AC, and incubated at 24 C in the light for 3 weeks. Germination was poor, so the embryos were transferred to Magenta GA7 vessels (Magenta Corp, Chicago, Illinois) containing 50 ml of hormone-free

Biolistic Experiment 3 Experiment 3 was designed to produce transformed calli that would later be transferred to embryogenesis medium. Because transformation and embryogenesis are both rare events, this strategy should provide a greater number of transformed cells from which embryos could form. The initial selection medium for experiment 3 was MS with 5.0 ILM 2,4-D and 1.0 I.tM BA, with or without AC, and in combination with 25 or 50 mg/l Km (Table 1). The cells were transferred to fresh medium of the same kind every 6-8 weeks. Embryos were first observed 29 weeks after bombardment, and in 50% of the plates from all treatments. Embryogenesis might have been expected on AC medium because of hormone absorption, but was not expected on AC-free medium. Perhaps the long-term

Table 1. The number of 'Chancello~ grapevine embryos and normal plants obtained after biolistic treatment and selection on Kin-containing medium. Results of histochemical GUS assays and PCR amplification of the NIYI'H gene are also indicated. Embryos GUS Assays Selection a Medium Experiments I and 2 c 0 AC/Km25 0 AC/KmS0 3 AC/Km25 3 AC/KmS0 Experiment 3 0 0 3 3

AC/Km25 AC./KmS0 AC/Km25 AC/Km50

No. Plates

No. b Obtained

~ Pos.

4 5

0 0

. .

5 5

67 113

7 4

7 5 5 5

911 295 29 222

30 . . 0 0

Normal Plants G U S Assay

Root rids Pcs. Nee.

Net.

. . . . . . . . . . . . . . . . . . . 18 2

.

.

5 . . 1 8

22 60

.

59 . . 0 0

. .

Pos.

302 . . 9 62

a 0 AC = without AC; 3 AC = 3 g/l AC; Km25=25 mg/l Km; Km50=50 rag/! Km b Includes clusters of embryos; each cluster counted as one embryo.

No. Obtained

. . . . . . . . . . . . . . . . . . .

31 1

.

Nee.

2 3

.

-----

6 15

...... . . ...... 2 ---

142 33 0 40

Pos. . .

Net.

pos.

Ne2.

. . . . . . . . . . . . . . . . . 1 7

. . . . . . . . 7 1 ---

13 91 1 --16 . . . . . . . . . . . . . . . . . . 0 29 3 1

c All plates initially on 10 mg/l Km

315 Table 2. Normal shoot formation from 'ChanceUof grapevine embryos cultured on different media following biolistic treatment.

Germination TI~ar-e.t

Embryo Single ~ Embryos

Normal % Embryos with Shoots Normal Shoots

Group 1 (Destructive GUS assay)

---

27

---

48

ACfttofmone-free 4~ for 2 weeks 1.0 p.M BA

45 55 ---

----18

0.1pM'rDz

---

Group 2 (Destructive GUS assay)

Group 2 a (WPM medium)

15

35 2 4 1

77.8% 3.6% 22.2% 6.7%

420 77 75 70

---------

71 27 8 I0

16.9% 35.1% 10.7% 14.3%

198 180 196 213

--------

8 10 21

4.0% 5.6% 10.7%

39

18.3%

Group 3 b (WPM medium) ACAIonnone-free 0.54 pN[ NAA 0.1 p.M TI)Z 1.0 lttM BA

Group 3 b (MS Medium) 0/a,.,VlBA 0.1 jJ.M BA 0.4 ~VI BA 1.0 tiM BA

aEmbryos were initially grown on hormone-free 1/2 MS with 3 g/1 AC at 24 C for 3 weeks; then transferred to hormone-free WPM with AC and incubated at 4 C for 1 week and subsequently at 24 C for 12 weeks. Embryos had roots but abnormal shoots after this initial treatment. bEmbryos were initially cultured on hormone-free 1/2 MS with 3 g/l AC at 4 C for 2 weeks and subsequently at 27 C for 1 week. Embryos were then transferred to the media listed in the table. Six weeks later all embryos were transferred to hormone-free media.

WPM with AC, and incubated in the dark at 4 C for 1 week. The embryos were subsequently returned to 24 C in the light. At this time, 48 embryos were destructively assayed for GUS expression. Thirty-four (70%) of the embryos were brilliant blue throughout. The remaining embryos not assayed for GUS were transferred to individual baby food jars with 30 ml of the above medium. Most embryos had normal roots, but abnormal shoot growth. GUS assays were performed on excised root tips (approx. 1 mm long) so that the remaining tissue could be saved. Five of 25 (20%) root tips assayed were GUS-positive. To further demonstrate the transgenic nature of the embryos, PCR was used to amplify DNA from root and hypocotyl tissue (Fig. 2). The NPTII gene was amplified in all 5 GUS-positive and 2 GUS-negative embryos tested, but not in nonbombarded controls. Because the plasmid encodes a fusion peptide, we expected both genes to be present and expressed equally. In the GUS- negative/NPTII-positive

Fig. 2. Ethidium bromide stained agarose gel of PCR amplified NPTII gene from 5 GUS-positive embryos (lanes 2 to 4 = root tissue, lanes 5 and 6 = cotyledons and hypocntyl tissue), non-bombarded tissue (lane 7), 2 GUS-negative embryos (lanes 8 and 9), and pBI426 DNA (lane 10). A zero DNA amplification control is in lane 11. A portion of a 1 kb ladder (Gibco BRL) is shown in lanes 1 and 12.

embryos, low levels of NPTII expression may have allowed for growth on Km medium, whereas low levels of GUS may have been below the threshold for histochemical detection. Four treatments were tested to induce normal shoot growth from the second group of germinated embryos: 1) an additional cold treatment at 4 C for 2 weeks, 2) continued growth on WPM + AC without hormones, 3) growth on WPM with 1.0 IJM BA, and 4) growth on WPM with 0.1 ~tM thidiazuron (TDZ). Embryos on medium with BA and TDZ soon enlarged and turned bright green. A f t e r 3-4 weeks, the cotyledons and hypocotyls on TDZ-containing medium turned red and were covered with meristematic buds. Small shoots formed on hypocotyl tissue from both the BA and TDZ treatments. A few grew into normal rooted shoots after transfer to hormone-free medium (Table 2). The embryo clusters that were maintained on hormone-free WPM with AC produced the highest percentage of normal shoots (Table 2), but shoot formation was sporadic over a 2-6 month period after germination. The additional chilling treatment did not stimulate shoot production (Table 2). Normal shoots were maintained on MS medium, hormone-free for elongation and rooting, or 1.0-4.0 lxM BA for multiplication. A third group of 1429 embryo clusters was harvested from all three biolistic experiments approximately 1 year after bombardment. The embryos were initially placed on 1/2-MS medium with AC (hormone-free) at 4 C for 2 weeks. The Petri plates were subsequently given new lids and wrapped with Vent tape (Scotch Brand no. 394, 3M Corp., St Paul, Minnesota, USA) to reduce condensation in the plates and vitrification of the tissue. Embryos were incubated at 27 C for 1 week and were then transferred to fresh medium in baby food jars with Bcaps. Different media were tested: hormone-free WPM with AC; WPM with 0.54 ~tM NAA, or 0.1 [tM TDZ, or 1.0 ~tM B A without AC; or MS with 0, 0.1, 0.4, or 1.0 gM BA without AC. Within 1 month, embryos on BA and TDZ had grown larger and turned brighter green than those on NAA or hormone-free media. Rooting occurred on all media, but there were thicker and more numerous roots on NAA-containing medium. Root tips from 405 embryo clusters were assayed for GUS expression and 136 (33.6%) were positive. After 6 weeks on the above media, the embryos were transferred to hormone-free media. Normal shoot formation occurred over a 6 month period in 13.4% of the embryo clusters (Table 2). BA and NAA in the medium appeared to aid in normal shoot formation, however, caution should be used in the interpretation of Table 2 because the initial embryos were in clusters and at different stages of development. Additionally, embryos that arose 1 year or more after bombardment appeared to have a lower capacity for germination and normal plant formation and, thus, a higher requirement for growth regulators in the germination medium. To date, a total of 236 plants with normal shoot growth have developed from 12 different original plates bombarded on 3 different days (Table 1). These plates represent all of the original selection treatments, except

316

To date only a few grape cultivars have been transformed by either biolistics or Agrobacterium. Biolistic transformation might be better suited for certain cultivars, but this remains to be tested. Future experiments will test biolistic transformation of other grape cultivars, as well as insertion of viticulturally important genes. Acknowledgments.

W e gratefully acknowledge the following

contributions to this project: R.N, Goodman for the gift of embryogenic 'Chancellor' cultures, W.L. Crosby for the use of pBI426, A.D. Blowers for the girl of P C R primers, J.C. Sanford for the use of research facilities,and G.N. Ye for technical assistance. Financial support was provided by the N e w York Wine and Grape

Foundation, the InstitutNational de la Recherche Agronomique (D. H~bert-Soul~), and Grant No. IS-2251-93 from BARD, the United States-lsraelBinational AgriculturalResearch & Development Fund. References

Fig. 3. Southern blot analysis of 1 non-bombarded (lane 1) and 3 GUS and PCR-NPTII positive (lanes 2-4) 'Chancellor' plants. The blot was probed with a 32p-labelled fragment of the NIY]~ gene.

the AC-free plates from experiments 1 and 2. Of 164 normal shoots that were assayed for GUS, 37 (22.6%) were positive (Table 1, Fig. ld-g). PCR analysis of DNA from leaves of 1 GUS-positive and 4 of 5 GUSnegative bombarded plants indicated the presence of the gene (not shown), whereas non-bombarded control plants were negative for both GUS and Ntr~I. The infial GUS and PCR tests were performed shortly after the normal plants developed. A selected group of plants have been cultured on Km-free medium for 1 year and have maintained the GUS and NPTII genes. Three of these plants, which were positive for GUS and NPTII according to the PCR assay, were selected for Southern blot analysis along with one non-transformed plant. Each of the 3 transformants showed strong positive signals and unique integration patterns (Fig. 3). The non-wansformed plant did not produce visible bands. The complex pattern observed in lane 4 is likely the result of multiple insertions of the vector DNA into the plant genome. In summary, transgenic plantlets of 'Chancellor' grapevine were produced via biolistic transformation of embryogenic suspension cultures. The basic bombardment methods are similar to those used for suspension cultures of other genera (Russell et al. 1992; Sanford et al. 1993). The selection scheme can likely be further optimized, and may be slightly different for other cultivars. Further improvements are also needed to obtain a higher percentage of normal shoots from the embryos. Grape researchers now have an additional method for genetic transformation of important cultivars.

Baribault TJ, Skene KGM, Scott NS (1989) Plant Cell Rep 8:137140 Baribault TJ, Skene KGM, Cain PA, Scott NS (1990) J Exp Bot 1:1045-1049 Berres R, Otten L, Tinland B, Malgarini-ClogE, Walter B (1992) Plant Cell Rep 11:192-195 Bouquet A (1993) Progr~s Agricole et Viticole 110:327-330 Bovey R, Martelli GP (1986) Vitis 25:227-275 Broglie K, Chet I, Holiday M. Cressman R. Biddle P, KnowltonS, Mauvais CJ, BrogUe R (1991) Science 254:1194-1197 Colby SM, Juncosa AM, Stamp JA, Meredith CP (1991) J Amer Soc Hort Sci 116:356-361 Datla RSS, HammerlindlJK, Pelcher LE, Crosby WL, Selvaraj G (1991) Gene 101:239-246 Deloire A (1993) Le Vigneron Champenois 4:25-26 Feinberg AP, Vogelstein B (1983) Anal Biochem 132:6-13 Guellec V, David C, Branchard M. Temp~ l (1990) Plant Cell Tissue Organ Cult 20:211-215 H~bert D, Kikkert JR, Smith FD, Reisch BI (1993) Plant Cell Rep 12:585-589 Jaynes JM, Nagpela P, Destefano-BeltranL, Huang JH, Kim J, Denny T, Cetiner S (1993) Plant Sci 89:43-53 Kikkert JR (1993) Plant Cell Tissue Organ Cult 33:221-226 Lloyd G, McCown B (1980) Proc Int Plant Prop Soc 30:421-427 Lodhi MA, Ye GN, Weeden NF. Reisch BI (1994) Plant Mol Biol Rep 12:6-13 Lend P, DunsmuirP (1992) Plant Mol Biol 18:47-53 McCabe DE, Swain WF, Marfiuell BJ, Christou P (1988) BioFl'echnology 6:923-926 Mullins MG, Tang FCA, Facciotti D (1990) Bio/Technology 8:1041-1045 Murashige T, Skoug F (1962) Physiol Plant 15:473-497 Pearson RC, Goheen AC (eds) (1988) Compenditunof grape diseases, APS Press, St. Paul, MN Russell JA, Roy MK, Sanford JC (1992) In Vitro Cell Devel Biol 28P:97-105 Saiki RK, Scharf S, Faloona F, Mullis KB. Horn GT, Edich HA, Amheim N (1985) Science 230:1350-1354 Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: A laboratory manual, 2rid ed, Cold Spring Harbor, NY Sanford JC, DeVit MJ, Russell JA, Smith FD, HarpendingPR, Roy MK. Johnston SA (1991) Technique 3:3-16 Sanford JC, Smith FD, Russell JA (1993) Methods Enzymol 217:483-509 Scorza R, Cordts JM, Ramming DW, Emershad RL (1995) Plant Cell Rep (In press) Southern EM (1975) J Mol Biol 98:503-517 Trudel J, Potvin C. Asselin A (1992) Plant Sci 87:55-67 Vaeck M. Reynaerts A, lttfte H. Jansens S, De BeuekeleerM, Dean C, Zabean M. Van M~ntagu M. Leemans J (1987) Natme 328:3337 Weatherhead MA, Burdon J, Hensbaw GG (1978) Z Pflanzenphysiol 89:141-147 Wilson TMA (1993) Proc Natl Acad Sci USA 90:3134-3141