Plant Physiol. (1983) 72, 1110-11130032-0889/83/72/11 10/04/$00.$0/0

Fusion of Plant Protoplasts by Electric FieldsReceived for publication February 8, 1983 and in revised form April 29, 1983

GEORGE W. BATES', JOHN J. GAYNOR2, AND NARPAT S. SHEKHAWATDepartment of Biology, Yale University, New Haven, Connecticut 06511

ABSTRACT

The electrical fusion technique of Zimmermann and Scheurich (1981Planta 151: 26-32) has been used to fuse mesophyll protoplasts of Avena,Zea, Vigna, Petunia, and Amaranthus. Electrical fusion proves to be asimple, effective, and general fusion technique that can be controlled toform either dikaryons or large multinucleate fusion bodies. In addition, weshow that Vigna mesophyll protoplasts that are subjected to the electricalfields used in this technique are viable in culture. The construction of thefusion chambers, necessary electrical equipment, and the fusion protocolare described in sufficient detail for reproduction of the technique.

The fusion of protoplasts has become an important tool in plantsomatic cell genetics, especially for the production of interspecifichybrids (10). Recently, the list of methods available for protoplastfusion has been expanded by the development of techniques forfusing cells electrically (8, 14). The electrical technique uses ashort DC pulse, of sufficient voltage to cause reversible membranebreakdown, to fuse the protoplasts. Obviously a prerequisite forfusion is intimate contact between the membranes of adjacentcells. Senda et al. (8) provided membrane contact by pushing twoprotoplasts together with the microelectrodes used to give thefusogenic electrical pulse. Although this approach was effective,its usefulness is limited by the small number of cells that can befused at one time. An elegant improvement on this technique,developed by Zimmermann and coworkers (13, 14), now makespossible the simultaneous fusion of large batches of protoplasts byelectrical fields. Zimmermann's method appears to have severalimportant advantages over the standard PEG technique for pro-toplast fusion. (a) Fusion is rapid, highly synchronous, and usuallycomplete within 15 min. (b) The yield of fusion products can bevery high (50-80%). (c) Electrically induced protoplast fusion doesnot require any chemical treatment. (d) The electrical fusiontechnique affords a greater degree of control over the number ofcells fusing and possibly even the types of fusions produced (i.e.heterokaryons as opposed to homokaryons) than does PEG. De-spite the outstanding attributes of Zimmermann's technique, ithas not been shown that protoplasts are still capable of growthand cell division after being subjected to the strong electrical fieldsrequired for fusion. Clearly, this is an extremely important pointif the technique is to be applied to somatic cell genetics.We report here that the protoplast fusion technique of Zimmer-

man and Scheurich (13) is reproducible and relatively simple. Wealso show that Vigna protoplasts that have been subjected to theelectrical fusion technique remain capable of cell division inculture.

' Present address: Department of Biological Science, Florida StateUniversity, Tallahassee, FL 32306.

2 Present address: Laboratory of Plant Molecular Biology, RockefellerUniversity, New York, NY 10021.

MATERIALS AND METHODS

Fusion Chamber. We used the flow-through chamber dia-grammed in Figure 1. It was built of two pieces of glass gluedonto a glass microscope slide with epoxy cement. Two metal rods,which served as electrodes, were glued between these pieces ofglass so that an even slot 500 ,um wide remained in the center ofthe chamber. Care was taken to keep this central slot free of glue.The ends of the chamber were fitted with polyethylene tubing tomake an inlet and an outlet port. The entire chamber was coveredwith a cover slip and sealed with epoxy cement. Several differenttypes of metal were tried as electrodes; gold, platinum-iridium,and silver all worked equally well. The crucial factor in making asuccessful chamber proved to be getting the electrodes as nearlyparallel as possible.

Electronics. AC fields were provided by a Hewlett Packardoscillator model 200CD. A Grass Medical Instruments (Quincy,MA) stimulator model S4 was used to give the DC pulses. Theseinstruments were arranged in parallel with each other and thefusion chamber. The performance of the stimulator and oscillatorwere monitored on an oscilloscope.

Isolation of Oat and Corn Protoplasts. The upper epidermis ofleaves from light-grown, 6-d-old oat seedlings (A vena sativa, cvVictory) and 8-d-old corn seedlings (Zea mays, cv Bear Hybrid)were removed with fine forceps. The peeled leaves were floated,upside down, on a solution containing 2% (w/v) Cellulysin (Cal-biochem), 0.5 M mannitol, 3 mm CaCl2, 1 mm KCl, and 3 mm Mes(morpholinoethanesulfonic acid) at pH 5.6. Digestion was com-plete after 3 h at 30'C in the dark. The protoplasts were filteredthrough a nylon screen (pore diameter 80 ,um), layered onto a 17%(w/v) sucrose pad, and centrifuged 10 min at 100g. The protoplastsat the interface were collected, resuspended in 12 ml of 0.5 Mmannitol and centrifuged for 3 min at 70g. The pellet was washedonce with 0.5 M mannitol.

Isolation of Amaranthds Protoplasts. Primary leaves from 7-d-old plants of Amaranthus cruentus (strain R102/104) were cut into1-mm-thick strips, washed briefly with 0.6 M mannitol, and infil-trated 4 min under vacuum with 1% (w/v) Cellulysin, 0.5% (w/v)Pectolyase Y-23, 0.6 M mannitol, 5 mm CaCl2, and 5 mm Mes at

4- I.o

FIG. 1. Diagram of the fusion chamber; the drawing is not necessarilyto scale. The inter-electrode distance, which is the critical dimension, is500 pm. + and - designate the two electrodes, and the arrows indicate thepath of the protoplasts through the chamber.

1110 www.plantphysiol.orgon August 19, 2020 - Published by Downloaded from

Copyright © 1983 American Society of Plant Biologists. All rights reserved.

FUSION OF PROTOPLASTS BY ELECTRIC FIELDS

pH 5.8. Tissue slices were gently shaken, in the dark, at 30C for4 h. The protoplasts were passed through an 80-,tm nylon screenand purified on an isoosmotic sucrose gradient as described byHarms and Potrykus (2). The isolated protoplasts were resus-pended in 0.5 M mannitol.

Isolation of Petunia Protoplasts. The upper surfaces of leavesof Petunia axillaris were removed and the leaves were floated on2% (w/v) Cellulysin, 1% (w/v) Pectinol AC (Rohm and HaasCo.), 0.5 M mannitol, 3 mm CaCl2, 1 mm KCl, and 3 mm Mes atpH 5.7. Digestion took 4 h at 30'C. The Petunia protoplasts wererecovered and purified by the same methods used for oats andcorn. In early experiments, the pectinase was omitted from thedigestion medium.

Isolation and Culture of Vigna Protoplasts. Mesophyll proto-plasts were isolated from Vigna aconitifolia 'Jadia' as described byShekhawat and Galston (9). Briefly, primary leaves of 14-d-oldplants were surface sterilized for 5 min in 5% (w/v) commercialbleach containing a few drops of Tween-80. After 5 to 6 washingsin sterile distilled H20, the lower epidermis was removed and theleaves were floated on desalted 1% (w/v) Driselase (KyowaHakko, Co., Ltd., Japan), 9% (w/v) mannitol, salts (KH2PO4, 27.2;KNO3, 101; CaCl2 2H20, 148; MgSO4.7H20, 240; and KI, 0.16mg/l), and 3 mM Mes, at pH 5.8. This solution was filter sterilized.After a 5-h digestion at 30°C the protoplasts were passed througha 55-,um nylon mesh and centrifuged for 5 min at 100g. Theprotoplast pellet was resuspended in 9% mannitol and recentri-fuged. The final pellet was resuspended in 9% mannitol layeredover 2 ml of a 21% (w/v) sucrose solution and centrifuged at lOOgfor 5 min. The protoplasts at the interface were resuspended in 9%mannitol.

After fusion, the protoplasts were concentrated by centrifuga-tion and cultured in 25-,ul sitting drops on the bottoms of 60 x 15mm plastic Petri dishes. The culture dishes were sealed withparafilm and maintained at 26°C under day light fluorescent tubesat a Photon Flux Density of 30 to 50 ,iE/m2 -s. The culturemedium contained Murashige and Skoog (6) macro- and micro-nutrients with KM vitamins and organic acids (4); CaCl2.2H20,900 mg/l; sucrose, 30 g/l; L-glutamine and L-asparagine (2 mmeach); arginine, 10 mg/l; mannitol, 45 g/l; glucose, 30 g/l; ribose,500 mg/l; and xylose, 300 mg/l. Hormones were supplied as 0.5mg/l each of 2,4-D, naphthaleneacetic acid, BA, Zeatin, and GA3.The pH of the culture medium was adjusted to 5.8 before filtersterilization.

RESULTS

The electrical fusion of protoplasts is a two-step process. First,the protoplasts are subjected to a high-frequency AC field (>100kHz), which draws the protoplasts towards regions of higher fieldstrength, that is toward the electrodes. This phenomenon, calleddielectrophoresis, has been described in detail by Pohl (7). Dielec-trophoresis requires a medium of low conductivity. Thus, com-pared with the medium, the protoplasts are a highly conductiveelectrical path and the poles of the protoplasts also become localregions of high field strength. Consequently, the protoplasts areattracted to each other as well as to the electrodes and becomealigned in 'pearl chains' along the lines of force of the AC field(Fig. 2). Once cell contact has been established, the second step isto superimpose a DC pulse (duration 2-100 ,us) of sufficientmagnitude (about 1 kv/cm) to produce reversible breakdown ofthe cell membrane (1, 5). This procedure causes the fusion ofneighboring protoplasts within the pearl chains. Cell lysis ratherthan fusion results if the DC pulse is too long or too large (13).Oat-Oat Fusions. Initial experiments were carried out with oat

mesoph7ll protoplasts suspended in 0.5 M mannitol (conductivity,4 x 10 mho/cm). A small number of protoplasts were introducedinto the chamber and an AC field of 10 v (peak to peak) at 500kHz was applied. Chains of protoplasts form on each electrode in

CV:w

FIG. 2. Examples of pearl chains. Oat mesophyll protoplasts aligned inan AC field (500 kHz, 100 v/cm) at a low density. Under these conditions,pearl chains two to three cells long predominate.

about 10 s (Fig. 2). In all of the experiments discussed in thispaper, an AC frequency of 500 kHz was used. At lower frequen-cies, especially those below 100 kHz, dielectrophoresis still occurs,but many of the protoplasts also spin. This spinning motion breaksmembrane contact and prevents fusion.

Pearl chains were found to form in AC fields of 5 to 15 v (50-150 v/cm). The AC field strength and the number of cells allowedinto the chamber control the lengths of the pearl chains. Pearlchains two to three cells long predominate when the density ofcells and the AC field strength are kept low (Fig. 2). Fusion,however, is most efficient when the AC field strength is fairly highbecause of the increased area of contact between the protoplasts.Therefore, in order to develop and fuse short chains of protoplasts,a low-strength AC field was used to form the pearl chains, thenthe chamber was gently flushed with mannitol to remove anyunaligned cells, and the AC field was increased in strength. Wefound that dielectrophoresis was not prevented by the presence of0.1 mm CaCl2 in the fusion medium; however, higher concentra-tions of salt were inhibitory. Inclusion of 0.1 mm CaCl2 seemed tohave no effect on the fusion rate, so it was therefore generallyomitted from the medium. However, in an effort to keep the cellsviable and to reduce lysis, calcium was supplied during protoplastisolation.Once pearl chains had formed, the oat protoplasts were fused

by application of a DC square wave. We found that DC pulses of10 to 50 ,is duration and 35 v (700 v/cm) were effective. Theprecise DC voltage required for fusion varies somewhat betweenexperiments and appears to be a function of the number of cellsin the chamber. Our general procedure was to give a DC pulse of25 v, and if no fusion occurred the voltage was increased stepwiseby 5 v until fusion was observed. Because applying trains of DCpulses causes cell lysis and electrolysis, we allowed 30 s betweensuccessive pulses. When the critical voltage for fusion is reached,the cells elongate immediately following the pulse and flatten attheir points of contact (Fig. 3b). Fusion of the plasma membranesis probably immediate, but actual coalescence of the protoplaststakes 2 to 5 min for oats. After fusion, the AC field strength wasreduced because this procedure was found to shorten the timerequired for the fusion product to round up as a single protoplast(Fig. 3d).The percentage of protoplasts in the chamber that fuse depends

on the uniformity of the lengths of the pearl chains. Cells in longerchains undergo fusion at lower DC voltages than cells in shortchains. This problem can be largely overcome by giving severalDC pulses at 30- to 60-s intervals. However, as already mentionedthis procedure causes some cell lysis. We found that 2 or 3 DC

lilll

www.plantphysiol.orgon August 19, 2020 - Published by Downloaded from Copyright © 1983 American Society of Plant Biologists. All rights reserved.

BATES ET AL.

a bI.0

Plant Physiol. Vol. 72, 1983

I. ... P ...

tb.

a beI -tp *1.-

Zi I d

ic d

FIG. 3. Fusion sequence of two oat mesophyll protoplasts. These pro-

toplasts were aligned in an AC field of 70 v/cm at 500 kHz (a). Then a

DC square wave (15 Us, 800 v/cm) was applied. Photographs were takenat the following intervals after the DC pulse: (b) 5 s; (c) 75 s; (d) 135 s. Jusibefore the photograph in 'd' was taken, the AC field was disconnected.

pulses of sufficient voltage to cause fusion could be given safelyand maximized the number of protoplasts fused. We have nottried to quantify the percentage of cells fused, but the reports of50 to 80% fusion rates (13, 14) appear to be justified as long as thepearl chains are quite uniform in length.

If a large number of protoplasts are introduced into the cham-ber, the pearl chains can become long enough to bridge the gapbetween the electrodes. Under these conditions, electrical fusionresults in the production of giant cells. We have recovered oatfusion products with as many as 25 nuclei.

Following fusion, we disconnected the AC field and collectedthe protoplasts by flushing the chamber with fresh medium. Theunfused protoplasts do not stick to each other in the absence ofthe AC field. However, some protoplasts do adhere to the elec-trodes. Most ofthe fusion products can be recovered ifthe mediumis gently forced back and forth in the chamber a few times.Corn-Oat Fusions. Mesophyll protoplasts prepared from oat

and corn leaves were readily fused with each other. So that thetwo types of protoplasts could be distinguished, the oat protoplastswere stained with neutral red for 5 min (and washed) prior tofusion. Inasmuch as our chamber contains only a single inlet, thetwo types of protoplasts had to be mixed before they were intro-duced into the chamber. Under these conditions, we had nocontrol over whether homokaryons or heterokaryons were pro-duced. However, the experiment clearly shows that heterokaryonscan be produced by electrical fusion. Figure 4 shows a timesequence of electrically stimulated fusion of a corn and an oatprotoplast. The specific frequency and voltages required for die-lectrophoresis and fusion were exactly the same for corn and oatprotoplasts. Coalescence of the protoplasts took longer in thisexperiment than in that shown in Figure 2. This range of fusiontimes probably reflects some difference between the protoplastpreparations used in these experiments.

Fusion of Dicot Protoplasts. We have successfully fused proto-plasts from several species of dicots by the electrical technique,including those from Amaranthus, Vigna, and Petunia. All of thesewere homokaryotic fusions. Dielectrophoresis occurred readily

FIG. 4. Fusion sequence of corn and oat mesophyll protoplasts. Theoat protoplast is the darker of the two because of staining with neutral red.The corn protoplast was unstained. In 'a' the protoplasts are aligned in theAC field (500 kHz, 100 v/cm). b to f are at various times following the DCpulse (20 ps, 700 v/cm): (b) 5 s; (c) 30 s; (d) 2 min, (e) 5 min, (f) 10 min.Just before the photograph in 'e' was taken, the AC field was disconnected.

with the dicots with precisely the same settings and approachesthat have been described for oat protoplasts. However, the dicotprotoplasts were more resistant to electrical fusion than the mon-ocots were. DC pulses of 10 to 20 ps duration, which were veryefficient in fusing oat and corn protoplasts, rarely produced anydicot fusions. Moreover, a greater DC voltage was required forfusing the dicot protoplasts than for the monocot protoplasts. Theoptimal settings of the DC pulse for fusing the dicot protoplastsproved to be 50 ,is and 50 to 60 v (1.0-1.2 kv/cm). Even withthese settings the rates of dicot fusion were lower than those seenfor the monocots. We found, however, that the fusion of dicotprotoplasts was greatly enhanced if a pectinase was includedduring protoplast isolation. Possibly residual cell wall material orcell surface determinants were blocking fusion.

Vigna protoplasts were subjected to electrical fusion, recoveredfrom the fusion chamber and cultured in microdrops. Cell wallsynthesis and cell division were monitored by staining of theprotoplasts with Calcofluor White ST (American CyanamidCorp.). Immediately after recovery from the fusion chamber theprotoplasts lacked any fluorescence when stained with CalcofluorWhite. After 3 d in culture, the protoplasts had reformed cell wallsand some dividing cells were observed. After 5 d in culture, about50%1o of the cultured cells had undergone at least one division. Cellwall biosynthesis and the cell division rates of electrically treatedcells were identical to those of control cells, which were cultureddirectly (9) or were passed through the chamber (without electricalstimulation) and then cultured.

DISCUSSIONOur experiments support the observations of Zimmermann and

his co-workers to a large degree. We find that electrically inducedcell fusion is a rapid, simple, and controllable process, whichworks with a wide range ofplant species. Moreover, with monocotswe were able to obtain very high rates of fusion.The success we had with culturing the Vigna protoplasts indi-

cates that viable protoplasts can be recovered following applica-tion ofthe electrical fusion technique. Because no selection schemewas applied after fusion we do not know whether any of thedividing Vigna protoplasts were actually fusion products. How-ever, the dividing cells were subjected to all of the electrical fieldsand manipulations required for production, fusion, and recoveryof the protoplasts under sterile conditions. Therefore, it is highlylikely that some of the fusion products are also viable.We encountered two problems with the electrical fusion tech-

I1112

k.

I

Ii.j

em.i

-lw - V4-v

www.plantphysiol.orgon August 19, 2020 - Published by Downloaded from Copyright © 1983 American Society of Plant Biologists. All rights reserved.

FUSION OF PROTOPLASTS BY ELECTRIC FIELDS

nique: (a) the protoplasts often adhered to the electrodes and (b)dicots were more difficult to fuse than monocots.The protoplasts stuck to all of the metals we tried as electrodes

(i.e. gold, silver, and platinum-irridium). Moreover, the longer theprotoplasts were in contact with the electrodes the more stronglythey became attached. While most of the adhering protoplastscould be recovered if the medium was forced back and forth inthe chamber, this procedure does break some cells. Althoughflushing the chamber with 200 mm KCI did not release theprotoplasts, it is likely that the initial interaction of the cells withthe electrodes is electrostatic and is probably followed by aprogressive molding of the plasma membrane to the electrode'ssurface. It should be possible to reduce adhesion by placement ofan inert membrane filter between the protoplasts and the elec-trodes.The observation that adding a pectinase to the protoplast iso-

lation medium makes dicot protoplasts easier to fuse indicates thatpart of the difficulty in fusing these protoplasts is related toincomplete removal of their cell walls. A similar observation wasmade by Kao et al. (3) during PEG-induced fusion of dicotprotoplasts. Because cell fusion requires very close approach ofthe lipid phases of the two cell membranes, any exposed mem-brane proteins or residual wall material adhering to the membranecould prevent fusion. Consistent with this view is the report byZimmermann et al. (12) that pretreating animal cells with pronasegreatly increases the yield of electrically induced fusion products.Undoubtedly, the fusion rates of the dicot protoplasts can beimproved even further with similar treatments.

Future work needs to address the problem of identifying andrecovering heterokaryons. This is, of course, the same problemthat arises when any technique is used to fuse protoplasts. How-ever, electrically stimulated fusion may simplify this problem bymaking possible high yields of heterokaryons. Vienken and Zim-merman (11) reported that viable heterokaryons of lymphocytesand murine myeloma cells could be produced by electrical fusionwith yields of 60 to 80%o. This result required a fusion chamberwith multiple inlet ports to allow sequential introduction of the

two types of cells into the chamber. Use of this approach could beimportant to the field of plant somatic cell genetics as it mightmake possible the recovery oflarge numbers ofheterokaryons andthereby reduce the need for a stringent post-fusion selectionprocedure.

Acknowledgments-We wish to thank Drs. Philip Applewhite and Timoth Gold-smith for generously lending electrical equipment and Drs. Arthur Galston and MaryHelen Goldsmith for their advice and support. N. S. S. thanks the Ministry ofEducation and Culture, Government of India, New Delhi, for granting NationalScholarship for Study Abroad, 1980-81. G. W. B. thanks the Whitehall Foundationfor support.

LITERATURE CITED

1. BENz R, U ZIMMERMANN 1981 High electric field effects on the cell membranesof Halicystis parvula A charge pulse study. Planta 152: 314-318

2. HARMs CT, T POTRYKUS 1978 Fractionation of plant protoplast types by iso-osmotic density gradient centrifugation. Theor Appl Genet 53: 57-63

3. KAo KN, F CONSTABEL, MR MICHAYLUK, OL GAMBORG 1974 Plant protoplastfusion and growth of intergeneric hybrid cells. Planta 120: 215-227

4. KAo KN, MR MICHAYLUK 1975 Nutritional requirements for growth of Viciahajastana cells and protoplasts at a very low population density in liquid media.Planta 126: 105-109

5. KINOSITA K, TY TSONG 1977 Formation and resealing of pores of controlledsizes in human erythrocyte membrane. Nature 268: 438-440

6. MURASHIGE T, F SKOOG 1962 A revised medium for rapid growth and bio assayswith tobacco tissue cultures. Physiol Plant 15: 473-497

7. PoHL HA 1978 Dielectrophoresis. Cambridge University Press, Cambridge8. SENDA M, J TAKEDA, S ABa, T NAKAMURA 1979 Induction of cell fusion of plant

protoplasts by electrical stimulation. Plant Cell Physiol 20: 1441-14439. SHEKHAWAT NS, AW GALSTON 1983 Isolation, culture and regeneration ofmoth

bean ( Vigna aconitifolia) leaf protoplasts. Plant Sci Lett In press10. VASIL IK, V VASIL 1980 Isolation and culture of protoplasts. Int Rev Cytol I IB:

1-191 1. VIENKEN J, U ZIMMERMANN 1982 Electric field-induced fusion: electro-hydraulic

procedure for production of heterokaryon cells in high yield. FEBS Lett 137:11-13

12. ZIMMERMAN U, G PILWAT, H-P RICHTER 1981 Electric-field-stimulated fusion:increased field stability of cells induced by pronase. Naturwissenschaften 68:577-579

13. ZIMMERMAN U, P SCHEURICH 1981 High frequency fusion of plant protoplastsby electric fields. Planta 151: 26-32

14. ZIMMERMANN U, J VIENKEN 1982 Electric field-induced cell-to-cell fusion. JMembr Biol 67: 165-182

1113

www.plantphysiol.orgon August 19, 2020 - Published by Downloaded from Copyright © 1983 American Society of Plant Biologists. All rights reserved.