Plant Physiol. (1990) 92, 334-3390032-0889/90/92/0334/06/$01 .00/0

Received for publication June 29, 1989and in revised form September 17, 1989

Transient Gene Expression in Maize, Rice, and Wheat CellsUsing an Airgun Apparatus'

James H. Oard*, David F. Paige, John A. Simmonds, and Thomas M. GradzielDepartment of Agronomy and Range Science, University of California, Davis 95616 (J.H.O., D.F.P.); AgricultureCanada, Plant Research Center, Ottawa, Ontario, Canada (J.A.S.); and Department of Pomology, University of

California, Davis, California 95616 (T.M.G.)

ABSTRACT

An airgun apparatus has been constructed for transient geneexpression studies of monocots. This device utilizes compressedair from a commercial airgun to propel macroprojectile and DNA-coated tungsten particles. The ,-glucuronidase (GUS) reportergene was used to monitor transient expression in three distinctcell types of maize (Zea mays), rice (Oryza sativa), and wheat(Triticum aestivum). The highest level of GUS activity in culturedmaize cells was observed when distance between stopping plateand target cells was adjusted to 4.3 centimeters. Efficiency oftransformation was estimated to be 4.4 x 10-3. In a partial vacuumof 700 millimeters Hg, velocity of macroprojectile was measuredat 520 meters per second with a 6% reduction in velocity atatmospheric pressure. A polyethylene film placed in the breechbefore firing contributed to a 12% increase in muzzle velocity. A700 millimeters Hg level of vacuum was necessary for maximumnumber of transfornants. GUS expression was also detected inwheat leaf base tissue of microdissected shoot apices. Highlevels of transient gene expression were also observed in hard,compact embryogenic callus of rice. These results show that theairgun apparatus is a convenient, safe, and low-cost device forrapid transient gene expression studies in cereals.

accelerate DNA-coated gold spheres for stable transformationof soybean callus (2) and adult plants (9). In the particle gunsystem, micron-size tungsten particles coated with DNA areaccelerated to velocities sufficient for the nonlethal penetra-tion of cells walls and membranes. This approach has allowedtesting oftransient expression ofchloramphenicol acetyltrans-ferase and GUS2 reporter genes in maize cultured cells andimmature embryos (7), rice, soybean, and einkorn wheatcultured cells (15) and onion leaf epidermal cells (8). Theparticle gun relies upon a chemical explosion from the firingof a .22 cartridge to propel the DNA coated microparticlestoward the target cells. To reduce target cell trauma resultingfrom gases produced by chemical propellants, an airgun de-vice was originally conceived and developed by one of theauthors (T.M.G). We show in this report that a version ofthisdevice can be used to test transient gene expression in differentcell types of maize, rice, and breadwheat. Cost of materialsused in construction of the airgun device was relatively inex-pensive, and the simple design should permit rapid transientgene expression studies in certain plant cells or tissues.

MATERIALS AND METHODS

Plasmids

The capability to introduce modified genes into plant modelsystems such as tobacco or petunia has proven invaluable fortissue specific and developmental gene expression studies (1,5, 1 1). In addition, gene transfer experiments have contributedsignificantly to the evaluation of cloned genes of agronomicimportance (14). This approach, however, normally reguirescocultivation with Aarobacterium tumefaciens, tissue culture,and plant regeneration steps which altogether can take severalmonths for completion of one experiment. Similar time pe-riods are required for DNA uptake and regeneration of pro-toplasts to produce stably transformed plants. MicroinjectionofDNA has been tested successfully in isolated cells culturedin vitro (3, 12), but this method may not be applicable toother cell types or cells located below the surface of certaintissues and organs.One alternative to A. tumefaciens, protoplasts, and microin-

jection for gene transfer studies has been the development ofa 'particle gun' (13) for delivery ofDNA or RNA into intactcells or tissues. Arc electric discharge has also been used to

' This research was funded in part by Sogetal. Inc.

pAI,GusN (15), constructed by M. Fromm, USDA-ARS,Albany, CA, is 6.5 kb and consists of maize Adh promoter,maize Adh intron 1, GUS and NOS sequences inserted intothe pUC8 plasmid. Plasmid pZO1016 obtained from SandozCorporation, Palo Alto, CA, is 5.7 kb long and consists of35S cauliflower mosaic virus promoter, maize Adh intron 6,GUS, and NOS sequences inserted into the pUC19 plasmid.Recombinant plasmids were amplified in liquid cultures ofEscherichia coli, isolated by alkaline lysis, and purified twiceby CsCl/ethidium bromide density centrifugation. Restrictionmaps and description of each construct are presented inFigure 1.

Plant Material

Maize Cultures

Cell suspensions of Zea mays cv Black Mexican Sweet weregrown in a solution containing MS salts (10), B5 vitamins (4),

2 Abbreviations: GUS, f-glucuronidase; kb, kilobase pairs; NOS,nopaline synthase terminator; MS, Murashige and Skoog.

334 www.plantphysiol.orgon June 22, 2018 - Published by Downloaded from

Copyright © 1990 American Society of Plant Biologists. All rights reserved.

TRANSIENT GENE EXPRESSION STUDIES OF MONOCOTS USING AN AIRGUN

1 mM asparagine, 2 mg L-l 2,4-D, 20 g L' sucrose pH 5.8.Cells were transferred once a week by a 1:5 dilution with freshmedium and passed through a 500 ,um nylon screen at everyother transfer. Cultures were grown with agitation at 150 rpmunder constant light (25 uE m-2 s-') and temperature (260C).Four to 5 d old subcultures were used in all experiments.

Rice Callus

Field grown mature seeds of an Oryza sativa, ssp. japonicaadvanced breeding line were dehusked, soaked 10 s in 90%ethanol, then soaked in 50% commercial bleach with onedrop 100 mL-' X-77 brand surfactant (Ortho) for 15 minunder vacuum. Surface-sterilized seeds were washed threetimes with sterile distilled water and placed in 6 cm plasticdishes containing 12 mL MS salts and vitamins, 1.5 mg-12,4-D, 0.2 mg L' kinetin, 10% coconut water (Gibco) 40 gL' sucrose, 0.8% Phytoagar (Gibco) (pH 5.7). Plated seedswere grown under constant light (60 ,uE m-2 s-') and temper-ature (26°C) for 14 to 21 d. Hard, compact calli that developedfrom the scutellum were used for particle bombardmentstudies.

Wheat Leaf Base and Apical Tissue

Mature seeds of Triticum aestivum cv Tadinia were surface-sterilized in 50% commercial bleach plus 1 drop 100 mL-'Tween 20 for 15 min and washed thoroughly with steriledistilled water. Seeds were then germinated and grown indi-vidually in 25 mL glass culture tubes containing MS salts andvitamins 100 mg L' cefotaxime antibiotic (Calbiochem) andsolidified with 0.7% Phytagar (Gibco) (pH 5.7). Seedlingswere maintained at 25°C in continuous light (60 ,E m-2 s-').When coleoptiles reached approximately 2 cm in length, leafbase segments and the dome region of the apex were excisedseparately and transferred to Petri dishes containing MS me-dium solidified with 0.7% Phytagar for bombardment. Theexplants were transferred after bombardment to Petri dishes

pZ01016(5.7 Kb)S

500 470

x

GUS1800

P H

240

pAI1GUSN(6.6 Kb)

E BC p

ADH All

1200 560

GUS1900

P BG

N1

250

Figure 1. Diagram of pZ01016 and pAl1GusN constructs. The con-struct pZ01016 is 5.7 kb long and consists of a 500 bp fragmentfrom the 355 promoter, a 470 bp fragment from intron 6 (Al6) ofmaize Adh 1 gene, the 1.8 kb GUS coding region, a 240 bp fragmentfrom the nopaline synthase (N) terminator region; the remaining 2.7kb fragment is from the pUC19 plasmid. The construct pAlGusN is6.5 kb long and has been described previously (15). Size of eachfragment given in base pairs. S = Sall, X = Xhol, P = Pstl, H =HindIll, E = EcoRI, Bc = Bclll, and Bg = Bglll.

containing MS salts and vitamins, 5 mg L' zeatin (Sigma),100 mg L' cefotaxime, 1% sucrose, and 0.7% Phytagar (pH5.7). Leaf base and apical domes were cultured for 48 h incontinuous light (60 ,E m-2 s-1) and then assayed for GUSactivity.

Construction of Airgun Apparatus

The apparatus (Fig. 2) incorporated the following elements:(a) airgun (model RI lazerized version without stock, BeemanFirearms, Inc., Santa Rosa, CA) for propelling DNA-coatedmicroprojectiles; (b) polycarbonate vacuum chamber enclos-ing the airgun muzzle and the target material; (c) stoppingplate to halt and contain the macroprojectile while allowingthe microprojectiles to continue toward the target; and (d)vacuum pump with hoses and valves for evacuating thechamber.The vacuum chamber was constructed of 1.27 cm thick

polycarbonate machined to size and assembled with 6-32machine screws, with all mating surfaces wiped with siliconsealant before assembly. The outside dimensions ofthe cham-ber were 16.5 cm high by 12.7 cm wide by 12.7 cm deep. Thefront door was provided with a gasket and snap closures forremoval. The chamber top was drilled and tapped at its centerfor a 3/4 inch male pipe thread, waterproof, electrical con-nector complete with a compression rubber gasket (7529K57,McMaster-Carr, Los Angeles, CA) that clamps and sealsaround the barrel (shortened to 24 cm length), leaving 1 cmof the muzzle protruding into the chamber. Polycarbonatestrips were fastened on each side of the chamber interior at 2cm intervals, 1.27 cm apart, to provide slots for two removable1.27 cm thick polycarbonate shelves. The upper shelf servesto support the stopping plate holding cup, and has a 1.27 cm

AIRGUN

BREECH

ALUMINUMSUPPORTS

BARREL

HOLDING SUPPORTCUP SHELF

VACUUMBCHAMBER\ 7 TARGET

SHELF

Figure 2. Diagram of the airgun apparatus used for bombardmentof maize, rice, and wheat cells.

i

335

www.plantphysiol.orgon June 22, 2018 - Published by Downloaded from Copyright © 1990 American Society of Plant Biologists. All rights reserved.

Plant Physiol. Vol. 92, 1990

diameter hole aligned with the muzzle to allow passage of themicroprojectiles. The lower shelf supports the target materialand is inserted into the slot which gives the desired distancefrom the muzzle. Aluminum angle pieces, 0.32 cm thick x3.81 cm x 3.81 cm, were used to construct a support for theairgun. A 1.27 x 45 x 60 cm plexiglass base plate supportsthe entire assembly which has a final height of 85 cm.A stopping plate (Fig. 3) for bombardment of maize cul-

tured cells and rice callus was machined from a 1.27 cmdiameter stainless steel disc 0.2 cm thick, having a 0.193 cmdiameter center hole, with one side back-relieved at a 560angle from the horizontal. A second stopping plate for bom-bardment of wheat explants was constructed as the first,except no back relief was provided for the center hole. Thestopping plate holding cup, 2 cm long, was machined from2.54 cm diameter nylon rod. The stopping plate was press-fitagainst a shoulder in the base ofthe cup, which was positionedover the muzzle of the airgun and allowed to rest against itssupport shelf. Four 0.635 cm diameter vent holes were drilledat right angles to each other 0.70 cm from the base of thecup. The distance between the muzzle and stopping plate was0.80 cm. This arrangement maintained alignment betweenthe airgun bore and the hole in the stopping plate duringfiring.Wool cleaning pads supplied by Beeman were used as

macroprojectiles. In addition, 0.318 cm diameter discs werepunched from 0.435 mm thick galvanized sheet metal andglued with rubber cement to the face of the pads. A metalpunch (Roper Co., Rockford, IL) was used to create a dimplein the center of the disc for placement of the tungsten micro-projectiles. The dimple also served to reduce movement ofthe tungsten slurry until impact occurred on the stoppingplate. Average combined weight of the pad and disc was 76.6± 0.55 mg.

Velocity Measurements Macroprojectile

A 3.5 cm long nylon cup was machined similar to thestopping plate holding cup and two slits were cut horizontallyto its center at two locations 2 cm apart (Fig. 4). A solderingpost was attached near each end of each slit. Very fine (42gauge) bare copper wires were stretched through the slits andsoldered to the posts. The result was two wires drawn tautacross the center of the cup, in line with the rifle bore, which

BARREL

STOPPING CUP

PLATE X -VENT HOLE

SHELF

Figure 3. Design of stopping plate used in airgun apparatus.

would be broken in succession by the passage of the macro-projectile. The distance from the muzzle to the top wire was1.0 cm. A frequency counter (Beckman model UC10) and afunction generator (Heath model 1274) set at 2.0 MHz wereconnected to the wires as shown. In this design, the top wiregrounded the signal from the generator to the counter andprevented counting until it was broken. The counter couldthen accumulate cycles until the second wire was broken, thusinterrupting the signal. The velocity was then calculated fromthe signal frequency, distance, and total number of cyclescounted. The velocity of macroprojectiles was measured withthree replications each for the following conditions: (a) 700mm Hg vacuum in the chamber and barrel with a polyethyl-ene film (which creates a seal between the air filled cylinderand evacuated bore) placed in the breech before firing, (b) novacuum in the chamber and with a polyethylene film, (c) novacuum and no film.

Bombardment of Target Cells

To evaluate the airgun apparatus, maize suspension cellswere distributed as a thin layer onto 47 mm diameter What-man No.4 paper or 47 mm diameter nylon membrane (GN-6 type, Gelman). The average number of cell aggregates perbombardment (3.82 x 10' ± 6.6 x 10-2) was estimated byplacing cells over four different nylon membranes and allow-ing cells to settle for five minutes. Each membrane was thenplaced in a separate Petri dish with 10 mL of water. Disheswere placed on a rotary shaker at 100 rpm for 1 h. Cellsdislodged from the membranes (>95%) were collected andaggregate density per 10 mL was determined using a hema-cytometer. Three different distances 4.3, 7.0, and 10.5 cmfrom stopping plate to target cells were tested at 700 mm Hgvacuum with four replications at each distance. In a separateexperiment the effect of vacuum was also evaluated at 0 and580 mm Hg, three replications at each level. Rice callus andwheat leaf bases and apical explants were placed on 0.7%Phytagar solidified medium for bombardment at 7.0 and 4.3cm target distance, respectively.

Circular plasmid DNA was precipitated onto 1.2 ,m tung-sten particles (GTE, size range = 0.5 to 3.0 Mm) essentially asdescribed by Klein et al. (6) with a 5.0 ,g:2.5 mg DNA totungsten ratio. Target cells were bombarded once for eachexperiment using 2.5 ML of DNA-tungsten preparation pershot. The back-relieved stopping plate was used during bom-bardment of the maize and rice cells while the 'straight hole'stopping plate was used in an attempt to 'focus' the distribu-tion of tungsten during bombardment of the wheat tissue.

Detection of GUS Expression

Immediately after bombardment, MS culture medium wasadded to maize cells on the filter paper and cultured for 24 to48 h (26°C temperature and 60 MuE m-2 s-' light intensity).Bombarded rice callus and wheat cultures were placed ontheir respective growth media and cultured under the sametemperature and light conditions as the maize cells. To detectGUS activity, 100 ML of the following solution was added toeach cell type: 0.1 M sodium phosphate buffer (pH 7.0), 5 mMpotassium ferrocyanide, S mM potassium ferricyanide, 1.0%

336 OARD ET AL.

www.plantphysiol.orgon June 22, 2018 - Published by Downloaded from Copyright © 1990 American Society of Plant Biologists. All rights reserved.

TRANSIENT GENE EXPRESSION STUDIES OF MONOCOTS USING AN AIRGUN

CUP -

TOP WIREFUNCTIONGENERATOR _F

BARREL FREQUENCYCOUNTER

5- - SHELF

Triton X-100, and 0.3% w/v GUS substrate 5-bromo-4-chloro-3-indoyl-f-D-glucuronic acid (Clonetech, Palo Alto,CA). Cells were incubated 12 to 18 h at 37°C and thenobserved under a dissecting microscope. A single isolated cellor an aggregate of cells exhibiting blue color was consideredas one 'blue cell.' The number of blue cells was counted andaveraged for each treatment.

RESULTS AND DISCUSSION

The airgun apparatus utilizes compressed air to acceleratethe macroprojectile and tungsten particles. To facilitate un-derstanding of this process, a brief description of the compo-nents and operation of the device is given. A powerful spring(Fig. 5) is manually cocked and held by the trigger mechanism.The spring is attached to a piston which is withdrawn duringcocking to create an air-filled cylinder behind the projectileto be loaded into the firing chamber. The piston plungesforward when the trigger is pulled to compress the air andpropel the projectile through the barrel. When a partial vac-uum is created in the chamber and barrel, a piece of 50 ,tmthick polyethylene film is placed flat across the open breechafter loading the macroprojectile. This film retains the air tobe compressed by the piston despite the barrel being evacu-ated. The film ruptures when the air is compressed by thepiston and the projectile is propelled in the usual manner.The macroprojectile velocity measurements show that the

POLYETHYLEN EFl LM

MICROPROJECTILE'

MACROPROJECTILE

Figure 5. Internal components of airgun that generate compressedair for particle acceleration.

presence of theachieved.

Figure 4. Diagram of setup used to measure ve-locity of macroprojectile. Distance between wiresis 2.0 cm. Distance between barrel of airgun andtop wire is 1.0 cm.

film actually increases the muzzle velocity

Velocity of Macroprojectile

Replicated velocity measurements were made using thesetup illustrated in Figure 4 to determine if the airgun devicecould generate sufficient air pressure to propel macroprojec-tiles for transient expression studies. Velocity of the macro-projectile was measured at 520 ± 10 m sec-' with a 700 mmHg vacuum in the chamber and barrel using a polyethylenefilm in the breech. There was only a 6% reduction in velocityto 490 ± 12 m sec-' with no vacuum and a polyethylene film.Further reduction of velocity was detected at 432 ± 4 m sec'with no polyethylene film and no vacuum. Velocity of thislast measurement is virtually identical to speed of micropro-jectile reported by Klein et al. (8). The use of a polyethylenefilm appeared to aid in increasing velocity by approximately12%. No measurements were made with a vacuum andwithout a polyethylene film due to potential damage to pistonseal during firing. Data from these measurements suggest thatthe airgun apparatus can create sufficient velocity to penetratevarious cells and tissues by tungsten, gold, or other highdensity microprojectiles. In addition, the setup to measuremuzzle velocity is easy to operate so that velocity can bequickly monitored for different experiments. Based on theseresults, all further experiments were conducted under 700mm Hg vacuum and with a polyethylene film unless otherwisenoted.

Delivery and Expression of GUS Gene in Maize Cells

Maize suspension cultures were first used to evaluate thepotential of the airgun system to deliver reporter genes intocells for transient expression. In these experiments the effectof distance between the stopping plate and target cells andlevel of vacuum were examined. The greatest number of bluecells was observed at 4.3 cm, the shortest distance tested (Figs.6 and 7). These values are similar to those reported forbombarded maize (7) and wheat (15) cells. GUS expressionwas not observed when cells were bombarded with nakedtungsten particles (no DNA) as a negative control (Fig. 7).Increasing the distance to 7.0 cm resulted in a slight decreasein average GUS expression, though with an increase in samplevariance. At 10.5 cm, the greatest distance tested, there was a35% reduction in blue cells detected as compared to the 4.3cm distance. In a separate experiment, maize cells were bom-barded under 0 and 500 mm Hg vacuum (both 7.0 cm

I~~ Il

337

www.plantphysiol.orgon June 22, 2018 - Published by Downloaded from Copyright © 1990 American Society of Plant Biologists. All rights reserved.

Plant Physiol. Vol. 92, 1990

400w

O XOo~z Xr-jz -J<: wWm.)

-j

4 6

DISTANCE

Figure 6. The effect of distance betwEon GUS gene expression in Black MlStandard errors of mean values are pre

transient expression was estimated at 4.4 x 10-3. This valueis similar to gene transfer efficiency reported for bombard-ment of maize (7) and rice (15) cultured cells.

GUS Expression in Rice Callus

After bombardment of maize suspension cells, the airgunapparatus was used in gene transfer studies of rice callusgrown on solid agar. Transient GUS expression was detectedin hard, compact, regenerable callus of Oryza sativa spp.japonica breeding line (Fig. 8). The optimal distance forbombardment was 7.0 cm from stopping plate to target cells.Callus clumps did not break up upon impact of tungsten

8 10 12 particles, and no background blue color was detected in

. IN CM negative controls when GUS substrate was added (Fig. 8).Both large and small blue spots were observed, indicating at

hen stopping plate and target least single cells were transformed over the top portion of the

)sented as vertical bars. callus. At the 4.3 cm distance, virtually all callus cells wereblown out of the Petri dish from impact of the tungstenparticles and/or compressed gases. GUS expression was notdetected at the 10.5 cm distance. These results suggest thatthe airgun device, when properly 'calibrated,' can be used intransient gene expression studies of rice callus cells.

Figure 7. Left, detection of gene expression in maize cells bom-barded with construct pZO1016 containing GUS coding sequence.Four d old subculture of Black Mexican Sweet maize suspensionwas bombarded once with pZO1 01 6 construct and incubated for 48h. Bombarded cells were exposed to GUS substrate solution for 16h at 370C and exhibit a number of blue cells (black cells as indicatedby arrows in this black and white print of the original color slide).Right, bombardment of maize cells with naked tungsten particles (noDNA). No blue cells were detected for this negative control treatment.Scale bar = 1.4 mm.

distance), which resulted in an average of 0.25 and 2.5 bluecells, respectively.The majority of transformants that stained blue tended to

occur in aggregates of 5 to 20 cells (Fig. 7). Less than 1% ofthe transformants was observed as single isolated cells, a resultthat is in contrast to those reported by Wang et al. (15). Thesedifferences in distribution of blue cells may be attributed tothe amount of tunsten used per bombardment, the degree ofclumping in the DNA-tungsten suspension, the distance fromstopping plate to target cells, or cell number per aggregate.An average of 3.82 x 103 cell aggregates were placed on 47mm diameter filter paper, and if we assume 20 cells peraggregate were available for bombardment, then efficiency of

GUS Expression in Wheat Leaf Base Tissue

Figure 9 shows GUS expression in leaf base tissue of wheatafter bombardment of the vegetative shoot apex region. Thenumber of detected blue spots was low (1-8) for each explant,but these results were reproducible over three different exper-iments. Thin parafilm sections of bombarded shoot apicaldomes revealed the tungsten particles penetrated as far downas the fourth cell layer (our unpublished observations). How-ever, GUS expression was never detected in the apical domeitself even after repeated bombardment experiments. Theseresults may be due to inefficient GUS substrate penetration,wrong promoter (35S) for GUS expression in the apical dome,degradation of construct DNA in dome cells, or damage to

Figure 8. Left, GUS expression in embryogenic rice callus afterbombardment with pAl1GusN construct. Fourteen d old callus wasbombarded once and cultured 48 h before incubation with GUSsubstrate for 16 h at 3700. As in Figure 7, transformed cells (arrows)appear black in this print. Right, bombardment of rice callus withnaked tungsten particles (no DNA). Scale bar = 0.10 cm.

338 OARD ET AL.

www.plantphysiol.orgon June 22, 2018 - Published by Downloaded from Copyright © 1990 American Society of Plant Biologists. All rights reserved.

TRANSIENT GENE EXPRESSION STUDIES OF MONOCOTS USING AN AIRGUN

crease penetration and transformation efficiency. The poten-tial of the airgun device for transient gene expression studiesin leaves, roots, and other plant parts will also be evaluatedin future experiments.

ACKNOWLEDGMENTS

We thank Irvin Mettler (Sandoz) and M. Fromm (USDA, Albany,CA) for their gifts of constructs pZO1016 and pAI1GusN, respectively.Thanks also to J. Dvorak for use of his laboratory facilities to conductthese experiments and to Sogetal, Inc. for financial support.

Figure 9. Bombardment of wheat leaf base tissue with pZO1016construct and detection of GUS expression (top leaf base). Explantswere cultured 48 h before incubating in GUS substrate for 16 h,370C. Arrows indicate site of GUS expression. Bombardment of leafbase with naked tungsten particles (no DNA) (bottom leaf base).Scale bar = 230 ,m.

cells caused by penetration of tungsten particles. Experimentsare currently underway to test these possibilities.

CONCLUSIONS

We have demonstrated that the airgun apparatus describedin this report can be used successfully to deliver and express

the GUS reporter gene in three different cell types of maize,rice, and breadwheat. The data clearly indicate that optimumconditions for each tissue or cell type must be determinedempirically. This apparatus is relatively inexpensive to build($500, plus labor costs) and to operate ($0.04 per shot formacroprojectile and steel disc). We are currently testing dif-ferent designs of stopping plate and macroprojectile to in-

LITERATURE CITED

1. Benfey PN, Chua NH (1989) Regulated genes in transgenicplants. Science 244: 174-181

2. Christou P, McCabe DE, SwainWF (1988) Stable transformationof soybean callus by DNA-coated gold particles. Plant Physiol87: 671-674

3. Crossway A, Oakes JV, Irvine JM, Ward B, Knauf VC, Shew-maker CK (1985) Integration of foreign DNA following mi-croinjection of tobacco mesophyll protoplasts. Mol Gen Genet202: 179-185

4. Gamborg OL, Miller RA, Ojima K (1968) Nutrient requirementsof suspension cultures of soybean root cells. Exp Cell Res 50:151-158

5. Harpster MN, Townsend JA, Jones JDG, Bedbrook J, DunsmuirP (1988) Relative strengths of the 355 cauliflower mosaic virus,1', 2', and nopaline synthetase promoters in transformedtobacco, sugarbeet, and oilseed rape callus tissue. Mol GenGenet 212: 182-190

6. Klein TM, Fromm M, Weissinger A, Tomes D, Schaaf S, SlettenM, Sanford JC (1988) Transfer of foreign genes into intactmaize cells with high-velocity microprojectiles. Proc Natl AcadSci USA 85: 4305-4309

7. Klein TM, Gradziel T, Fromm ME, Sanford JC (1988) Factorsinfluencing delivery into Zea mays cells by high-velocity mi-croprojectiles. Bio/Technology 6: 559-563

8. Klein TM, Wolf ED, Wu R, Sanford JC (1987) High-velocitymicroprojectiles for delivering nucleic acids into living cells.Nature 327: 70-73

9. McCabe DE, Swain WF, Martinell BJ, Christou P (1988) Stabletransformation of soybean (Glycine max) by particle accelera-tion. Bio/Technology 6: 923-926

10. Murashige T, Skoog F (1962) A revised medium for rapid growthand bioassays with tobacco tissue cultures. Physiol Plant 15:473-497

11. Nagy F, Kay SA, Boutry M, Hsu MY, Chua NH (1986) Phyto-chrome-controlled expression of a wheat Cab gene in trans-genic tobacco seedlings. EMBO J 5: 1119-1124

12. Neuhaus G, Spangenberg G, Mittelsten 0, Schweiger HG (1987)Transgenic rapeseed plants obtained by the microinjection ofDNA into microspore-derived embryoids. Theor Appl Genet75: 30-36

13. Sanford JC, Klein TM, Wolf ED, Allen N (1987) Delivery ofsubstances into cells and tissues using a particle bombardmentprocess. Part Sci Tech 5: 27-37

14. Schell J (1987) Transgenic plants as tools to study the molecularorganization of plant genes. Science 237: 1176-1183

15. Wang Y, Klein TM, Fromm M, Cao J, Sanford JC, Wu R (1988)Transient expression of foreign genes in rice, wheat and soy-bean cells following particle bombardment. Plant Mol Biol 1 1:433-439

339

www.plantphysiol.orgon June 22, 2018 - Published by Downloaded from Copyright © 1990 American Society of Plant Biologists. All rights reserved.