- Email: [email protected]
Use of a transient expression assay for the optimization of direct gene transfer into tobacco mesophyll protoplasts by electroporation
Biochimie, 69 (1987) 621 - 628 © Soci6t6 de Chimie biologique/Elsevier, Paris
621
Use of a transient expression assay for the optimization of direct gene transfer into tobacco mesophyll protoplasts by electroporation Philippe GUERCHE 1, Catherine BELLINI 1, Jean-Michel LE MOULLEC 2 and Michel CABOCHE 1 ILaboratoire de Biologic Cellulaire, Route de St-Cyr, INRA, F-78000 Versailles, and 2Soci~t~ Roussel UCLAF, F-93230 Romainville, France (Received 13-3-1987, accepted after revision 21-5-1987)
Summary - Plasmid DNA was transfected into tobacco mesophyll protoplasts by electroporation. Transfection efficiency was estimated, using a transient expression assay based on the measurement of chloramphenicol transacetylase activity or by scoring colonies expressing resistance to paromomycin, an aminoglycoside related to kanamycin. Under conditions of cell survival superior to 50% after electroporation, transient expression signals and transformation efficiencies were found to be proportional. Factors affecting the efficiency of transformation were studied. A clear-cut optimum voltage (250300 V/cm) was detected. Among various salts tested, potassium chloride was the best electrolyte. No improvement of electroporation efficiency was obtained by a h e a t - s h o c k (45°C/5 min) treatment prior to electroporation or by the presence of polyethylene glycol in the electroporation medium. The physiological state of plants used as the protoplast source significantly affected the transfection ability of the resulting protoplasts. These results are discussed and compared to previously published procedures. electroporation I tobacco / protoplast / gene transfer I transformation
R~sum~ - Usage d'un essai de l'expression transitoire pour optimiser la transfection dans des protoplastes de m~sophylle de tabac par Electropora~.ion. De l'ADNplasmidique a c~t6transfect6 darts des protoplast~,~ d: me'sophylle de tabac par ~lectroporation. L "efficacit6 a ~t6 mesur~e par dosage de l'activit6 transitoire de la chloramphdnicol transacdtylase ou par ddnombrement des colonies cellulaires transformdes exprimant une rdsistance ~ la paromomycine, aminoglycoside proche de la kanamycine. Dans des conditions de transfection oh la survie cellulaire restait supdrieure ~ 50%, les mesures d'expression transitoire et de frdquence de transformation se sont avdrdes proportionnelles. Divers facteurs susceptibles d'affecter l'efficacitO de transfection ont Otd ~tudids. Les transfections se sont avdrdes efficaces pour des voltages compris entre 250 et 300 V/cm. Le chlorure de potassium dtait le meilleur dlectrolyte parmi divers sels testds. L "efficacitd des dlectroporations n'a dtO amdlior~e ni en faisant subir un choc thermique ~ 45 °C pendant 5 min aux protoplastes avant leur dlectroporation, ni en ajoutant du polydthyibne glycol dans la chambre d'~iectroporation. L'Otat physiologique des plantes utilisdes comme sources de protoplastes semble par contre affecter de fa¢on importante l'efficacitd de I'dlectroporation. Les rdsultats obtenus sont discutds et compards ~ d'autres travaux. dlectroporation I tabac I protoplasW I transfert de gene / transformation
Abbreviations: APH3°: aminoglycoside phosphotransferase 3'; CAT: chloramphenicol transacetylase; EDTA: ¢thylenediaminetetraacetic acid; MOPS: 3-(N-morpholino) propanesulfonic acid; PEG: polyethylene glycol.
622
P. Guerche et al.
Introduction Improved methods to obtain plant tranformation m u s t be developed sc that this tool can be used to
study plant physiology arid, in the future, contribute to improve plant crops. Direct gene transfer into plant protoplasts can be achieved by various techniques. Polyethylene glycol (PEG) v, as first used successfully to transform plant protoplasts [1, 2]. In these early experiments, transformation efficiency was low (10- s _ 10- 6). Improvement of transformation efficiency was obtained either by liposome mediated transfection [3], electroporation [4] or a combination of P E G treatment and electroporation [5]. The efficiency of this last procedure is high, but seems to be affected by various parameters and requires therefore m a n y experiments to be optimized for a new type of protoplast. This observation led us to develop similar studies under our experimental conditions of electroporation and to try to simplify the procedure. Protoplasts were transfected with a plasmid carrying the chloramphenicol transacetylase (CAT) coding sequence put under the control o f the 35S C a M V promoter [4]. The transient expression of C A T was measured in cells transfected under various conditions. The relationship between transient expression signal and transformation efficiency was also studied in cotransfer experiments with another plasmid carrying the aminoglycoside phosphotransferase ( A P H 3 ' ) coding sequence put under the control of the 19S CaMV promoter.
Materials and methods Plasmids Two plasmids were used for transfection experiments. Plasmid pABD1 [2] carries the coding sequence of the aminoglycoside phosphotransferase expressed by the bacterial transposon Tn5 and conferring resistance to aminoglycosides, such as kanamycin. This coding sequence is put under the control of the 19S promoter of CaMV and polyadenylation signals derived from the nopaline synthase gene. Plasmid pCaMV [6] carries the coding sequence of chloramphenicol transacetylase derived from pBR328. This coding sequence is put under the control of the 35S promoter of the CaMV and polyadenylation signals derived from the nopaline synthase gene. Plasmids were extracted according to the procedure of Holmes [7] and purified by isopycnic centrifugation in a CsCl gradient. Calf thymus DNA (Boehringer, Mannhelm) was sonicated to produce fragments of average size 1 kb and used as carrier DNA. Plasmid DNA and calf thymus DNA were resuspended at a concentration
of 1 tzg/ml in the electroporation buffer (0.5 M mannitol, 5 mM KCI and 200 ~M MOPS, pH 7.2, sterilized by autoclaving).
Protoplasts Protoplasts were isolated essentially as described by Chupeau [8] from the leaves of Nicotiana tabacum (cv. Xanthi, diploid clone D8) grown in the greenhouse and washed twice in the electroporation buffer (0.5 M mannitol, 200/zM MOPS, pH 7.2, and 5 mM KCI). Protoplasts were resuspended in this buffer at a concentration of 2 x 106/ml for electroporation. Electroporation Electroporation conditions were adapted from previously published procedures [4, 5]. The plexiglass electroporation chambers were of two types: A (length 2 cm, width I cm, depth 0.5 cm), or B (length, width and depth I cm). Two stainless steel electrodes were put on the more distant sides of the cuvettes and connected to the electrical system. The generator was an ISCO 490 power supply coupled with one or two 16 ~F capacitors put in parallel. The charged capacitors were discharged into the electroporation cuvette. An oscilloscope was used to control the electrical discharge through the cuvette. A complete discharge of the capacitor through the cuvette lasted 200 ms. To I ml of protoplast suspension, 10ttg of the plasmid and 50 t~g of calf thymus DNA were added. Electric pulses were performed at 5 s intervals. Under standard conditions, 3 pulses (300 V/cm) were performed in the type B cell. Cvlture of electroporated protoplasts and selection of t r A,7 ~ t~ ~ A , i r m a ~ l ' r ~ lf ~
Electroporated protoplasts were diluted immediately after electroporation in the culture medium To [8] at a density of 6 x 104 protoplasts/ml and incubated in the dark for 6 days. The resulting colonies were diluted in medium C [9] at a density of 10 4 colonies/ml and incubated under light. Paromomycin (20 ~g/ml) was used as a substitute for kanamycin to select for aminoglycosideresistant colonies in liquid medium C. Resistant colonies were scored 1 month later.
Chloramphenicol acetyltransferase transient expression assay Cells derived from electroporated protoplasts incubated in medium To were harvested 3 days after electroporation. Cells from 2 dishes were washed twice in medium KL (0.3 M KCI, 5 mM CaCI2) and resuspended in 1 ml of the CAT buffer (1/.tM leupeptine, 30 mM ascorbic acid, 20 mM EDTA, 250 mM Tris-HCl, pH 8.0) containing 2.5% polyclar powder. Cells were broken by a brief sonication (Branson soaifier, power setting 3, 30 s). The resulting lysate was heated at 60°C for 5 min and clarified by centrifugation for 10 rain at 4°C in an Eppendorf centrifuge. Supernatants were collected and stored frozen at - 2 0 ° C . CAT activity was measured as described prevint+sly [10]. The reaction mixture contained 100/zl
Electroporation of tobacco mesophyll protoplasts
623
of cell extract, 20~1 of acetyl CoA (4 mM, pH 8.3, Boehringer) and 5 t~l of [14Clchloramphenicol (58.5 Ci/mol, 1.7 mM, New England Nuclear). After incubation for 1 h at 37°C, the reaction was stopped by adding 200/.tl of Tris base (2.2 M) and reaction products were extracted twice with 700/~l of benzene. The benzene extract was dried in vacuo in siliconized Eppendorf tubes in a speed vac. Samples were solubilized in 20/~l of ethylacetate and subjected to thin-layer chromatography on silica gel plastic plates (Merck) using a 1:9 (v/v) mixture of methanol and chloroform. Dried plates were exposed to X-ray film. Spots corresponding to acetylated derivatives of chloramphenicol were cut out of the plastic plates and their radioactivity measured by liquid scintillation.
tion, endogenous tobacco CAT activity is negligible and accounts for less than 1°70 of transgenic CAT activity. CAT expression was followed for a period of 6 days in a batch of mesophyll protoplasts transfected with pCaMV CAT. Maximal levels of CAT activity were detected on the third and fourth days of culture and CAT measurements were therefore routinely, taken on the third day after electroporation.
Results
Voltage was critical in obtaining a maximal transient expression. A clear-cut optimum was obtained for a voltage of 300 V (Fig. 1), although a s~gnificant expression was detected also for 250 V. Similar results were obtained when experiments were performed with a higher number of pulses (10, 20 or 30 pulses per voltage condition (results not shown). During electroporation, the electron flow through the electroporation chamber is delivered by connecting a charged capacitor to the two electrodes of the chamber. This results in an exponen-
Measurement o f the transient expression of CA T activity in protoplasts transfected with the plasmid p C a M V CA T The assay of E. coli CAT activity was chosen as a convenient way of measuring transient expression signals in plant cells. This technique, commonly used to study gene expression in eukaryotic cells [11], had to be adapted to plaut cells. Preliminary experiments had shown that E. coli CAT activity is partially destroyed when mixed with plant cell extracts. This difficulty was overcome by stabilizing extracts with reducing agents, such as ascorbic acid (mercaptoethanol interferes with the CAT assay) and by adding poiyciar, an insoluble matrix binding phenolic compounds efficiently. Polyphenol oxidases were destroyed by heat inactivation at 60°C and inactivation with EDTA metals, such as Ca 2+, known to inactivate these enzymes. Under these extraction conditions, E. coli CAT activity was stable in extracts (results not shown). Transient expression experiments performed with various plasmids carrying the CAT structural gene under the control of prokaryotic or retroviral promoters were unsuccessful. In the controls, a signal corresponding to acetylated derivatives of chloramphenicol was detected, using the protocol described under Materials and Methods. This signal could be assigned to an endogenous tobacco CAT activity, since the activity was destroyed by boiling the extract prior to the incubation, and since the reaction was dependent upon the presence of acetyl CoA in the reaction mixture (results not shown). E. coli CAT activity is thermostable and inactivated only by treatments at temperatures above 65°C. Unfortunately, the cellular CAT activity displayed a thermal stability comparable to E. coli CAT activity. Under optimized conditions of electropora-
Influence of voltage, capacitor size and the number o f pulses on the efficiency o f transfection
r/) wa
0~
m
0 O r.. @
o E e~ r.)
X v
°~ > °~ 4-1 ¢.)
F-.
!
0
200
400
V01ts/cm Fig. 1. Influence of pulse voltage on the transient expression of CAT activity in tobacco mesophyli protoplasts transfected by electroporation with the plasmid pCaMV CAT. Electroporations were performed in the electroporation cell B as described under Materials and Methods at varied voltages, using a single 16 ~F capacitor and a series of 5 pulses.
P. Guerche et al.
624
tial discharge of the capacitor through the cell. The half-life of this discharge, recorded with an oscilloscope, is approximately 120 ms for a single 16/~F capacitor. When two capacitors are connected in parallel and discharged into the cell, this results in a 2-fold increase in the duration of the pulse. A measurement of CAT activity in protoplast cultures electroporated with one or two capacitors and increasing pulse numbers is presented in Fig. 2. For one capacitor, CAT transient expression is roughly proportional to the number of pulses, reaches a maximum, and then decreases for pulse numbers higher than 20 for cell A and 10 for cell B. This decrease correlates with reduced viability of protoplasts in the cultures (results not shown). Using two instead of one capacitor results in an approximately 2-fold decrease in the number of pulses required for optimal CAT expression. For a given voltage, the efficiency of eleetroporation seems to be related to the total amount of charge migrating through the cell, rather than to the duration of pulses.
Influence of salts on the efficiency of electroporation It is generally assumed that divalent cations play a critical role in the transformation of various organisms. We have compared the efficiency of electroporation in the presence of K +, Li +, M g 2 +, Ca 2+ and Zn 2+ at a 5 mM concentration (Fig. 3). K + ion was found to be most suited for efficient
10 • ,
U'J
g
"~
10
¢o
"5
s
"-1
0
Number of pulses Fig. 3. Influence of the nature of different salts on the efficienc; of electroporation. Tobacco mesophyll protoplasts were transfected with the plasmid pCaMV CAT in the electroporation buffer 5 mM KCi (&) which was replaced by the same concentration of other salts: LiCI ( • ) . MgCl 2 ([] ~, ZnCI 2 (e) CaC~z (,k) according to the exper:,ment.
electroporation. When the Ca 2+ concentration was decreased 2-fold in the electroporation buffer in order to keep a conductivity equivalent to 5 mM KCI, this resulted in efficiencies reaching values obtained with 5 mM KCI. On the contrary, increasing the KCI concentration to 25 mM resulted in a 10-fold decrease of transient expression signals (results not shown). Conductivity is therefore probably the main parameter involved in salt effects on electroporation efficiency.
Comparison of transient expression signals and transformation efficiencies
o
"
X
v
0
lO 2o 3o Number of pulses
Fig. 2. Influence of capacitor characteristics on the transient expression of CAT activity in tobacco mesophyll protoplasts transfected by electroporation with the plasmid pCaMV CAT. Electroporations were performed in the electroporation cell A at 300 V using one ( 0 ) or two (<>) 16/~F capacitors and variable numbers of pulses.
In previous works [2-4, 12-14], protoplasts transfected with a plasmid conferring resistance to kanamycin were subsequently used to select transformants on the basis of their resistance to kanamycin. Under selective conditions, kanamycinsensitive cells can prevent the development of kanamycin-resistant colonies, presumably by releasing toxic substances into the culture medium. This could lead to inaccurate measurements of transformation efficiencies. This difficulty was overcome
Eiectroporation of tobacco mesophyll protoplasts by Shillito et al. [15] who used agarose embedded cultures incubated in selective liquid medium containing kanamycin and transferred repeatedly into fresh medium. By replacing kanamycin with paromomycin, an aminoglycoside also detoxified by the aminoglycoside phosphotransferase (APHY) coded by Tn5 and pABD1, we could select transformants efficiently at high col! densities as shown in a reconstruction experiment presented in Table I, without having to resort ~,o time-consuming replenishment of the selective medium. Transformants were selected on the basis of paromomycin resistance in protoplast cultures cotransfected with two plasmids, pABD1 and pCaMV CAT. Simultaneously, part of the electroporated cells were used for transient expression and plating efficiency measurements (Fig. 4). When cultures were transfected under conditions in which protoplast viability remained higher than 50070 of the control, a proportionality between the transient expression signal and transformation efficiency was observed. When viability was reduced, transformation efficiency dropped more rapidly than the transient expression signal. A comparison of the transient expression signal and transformation efficiency was performed at constant pulse number
Table I. Influence of cell density on the recovery of kanamycin-rcsistant colonies. No. o f cells plated in 5 ml o f m e d i u m C
No. o f colonies growing in media containing
clone D8
Ka4
kanamytin (70 t~g/hi)
paromomycin (20 tzg/ml)
100 100 100 I00 100 100 100 0
37 --7 39 36 23* 0 0 0
33 32 38 42 36 32 38 0
0 10000 25 000 50000 100000 200 000 500000
25000
Protoplasts were prepared from the leaves of wild type tobacco (D8) and from a kanamycin-resistant tobacco (Ka4) obtained by transfection of pLGV23 neo [3]. These protoplasts were induced to divide and 4 days later they were mixed and sujected to selection of paromomycin- or kanamycin-resistant colonies in medium C for I month. Resistant colonies were ~hen scored. The plating efficiency of D8 and Ka4 protoplast-Oerived cells grown in medium C in the absence of selective pressure was approximately 35%. * Very small colonies.
625
g g k 30%
;/ i-
~,
__
.%
|
o
I0
v
"z z CD
"1
)
5
"I Number of pulses
10
Fig. 4. Comparison of transient expression signals and transformation efficiencies in tobacco mesophyil protoplasts cotransfected with a 1:1 mixture of pABDI and pCaMV CAT: influence of a h e a t - shock treatment. Prior to electroporation, part of the protoplast suspensions was subjected to a 5 rain heat-shock at 45°C (&) or not ( • ) . Protoplasts were dectroporated in cell B at 300 V using one capacitor and variable numbers of pulses. Plating efficiency Le., the proportion of pro•,,v,-o~ able to divide was measured 6 days after protopiast plating ( - - - ) . Transient expression of CAT activity was measured 3 days after transfection (--) and paromomycinresistant colonies were selected and scored in the remaining cultures as described under Materials and Methods (----).
and various voltages and showed again a propor tionality between these two measurements of transfection efficiency (results not shown).
Influence o f other treatments on the efficiency o f transfection by electroporation PEG and h e a t - shock can affect cell competence for transformation [5]. Under our electroporation conditions, no significant effect of 10°70 PEG on the transient expression of CAT activity was detected in electroporated protoplasts and a higher PEG concentration (20070) was even deleterious (Fig. 5). A h e a t - shock (45°C, 5 rain) was performed on a batch of protoplasts and this batch as well as non-treated protoplasts were electroporated in parallel with a mixture of pCaMV CAT and pABD1 and used for transient expression ex-
P. Guerche et al.
626
T a b l e l l . Influence o f plasmid and carrier D N A concen-
trations on the efficiency of electroporation.
• PEG 0% v .. I0% [:] , , ZO% 0 0 e_ r~
pCaMV CAT (~g)
C a l f thymus DNA (/,tg)
C A T activity (cpm/105 protoplasts)
0 5 10 20 30
50 50 50 50 50
205 6 200 10800 25 400 50300
10 30
0 0
740 6 500
o
o
E 0
X
/
._~ U
m
I-0 o
'
"
~1
Number
"
• lb
" of
pulses
Fig. S. Effect of the polyethylene glycol concentration on the efficiency of electroporation. Tobacco mesophyll protoplasts were electroporated with the plasmid pCaMV CAT in cell A at a voltage of 300 V, using a single capacitor (16 I~F) and in the absence (11) or in the presence of 10% ( V ) or 20% ( [] ) PEG 6000. CAT transient expression signals were measured 3 days later as described under Materials and Methods.
periments and transformant selection experiments. The data presented in Fig. 4 does not show any significant effect of a heat - shock on transient expression signal as well as on transformation efficiency. The presence of carrier D N A significantly increased electroporation efficiency (Table II). Transient expression signals increased roughly with the m o u n t of pCaMV CAT plasmid added, in the protoplast suspension: for a same amount of plasmid, the transient expression was increased 10-fold when 50/zg of carrier D N A was added, contrary to experiments in which no carrier D N A was added.
Influence of the source o f protoplasts used for electroporation on the efficiency o f transfection When plants are grown at low light intensity, all their leaves can be used as sources of dividing protoplasts [8]. Leaves were collected separately from different parts of these plants and used as sources of protoplasts for electroporation. Differences were observed as regards the ability of these batches of protoplasts to express CAT activity after transfection with the plasmid pCaMV CAT (Table III). In two independent experiments, the highest expres-
(300 V and 16/~F, 10 pulses in cell B). Tobacco mesophyll protoplasts were electroporated as described under Materials and Methods with plasmid pCaMV CAT and calf thymus DNA. Transient expression of CAT activity was measured in transfected cultures 3 days later.
T a b l e HI. Influence o f leaf position on the mother plant
on the transfection ability of the corresponding protoplasts. L e a f level
C A T activity (cpm/10 s protoplasts)
16 11 6 1
11 800 24500 16000 4600
~ --, --, ~
apex 15 10 5
No electroporation
280
Plants were grown under light until they developed 20 expanded leaves. Leaves were collected at 4 levels on these plants and used for protoplast preparation. Protoplasts were transfected under standard conditions with pCaMV CAT (300 V and 16/,tF, 10 pulses in cell B) and the transient expression of CAT activity was measured in transfected cultures 3 days later.
sion was obtained with the fully-expanded leaves located below the rapidly growing leaves of the apex area. Discussion Electroporation is an efficient technique for the transfer of naked D N A in eukaryotic cells and has been successfully used with mammalian cells [16], yeast [17], and more recently with plant cells [ 4 - 6 , 18, 19]. We have reinvestigated the condi-
Electroporation o f tobacco mesophyil protoplasts
tions of transfer of plasmid DNA into tobacco mesophyll protoplasts using this technique. For the first time, the efficiency of electroporation was studie,.~ by measuring the transient expression of CAT activity encoded by the plasmid pCaMV CAT and by simultaneously scoring the number of paromomycin-resistant clones induced by the stable expression of APH3' activity encoded by the plasmid pABD1. It was found useful to optimize the conditions of extraction of CAT, since this enzyme was destroyed in crude extracts. The data obtained by these two optimized approaches agreed well, as long as electroporation conditions did not strongly affect the viability of treated protoplasts. From this correlation, it is concluded that transient activity measurements in transfected protoplast cultures can be used to rapidly optimize conditions of stable transformation of protoplasts from various plant species. Paromomycin was found to be more efficient than kanamycin for the efficient selection of transformants. Under standard transfection conditions, the frequency of transformants in electroporated cultures varied from 1 to 5 x l 0 -3 transformants per protoplast able to divide after electroporation. This frequency is high enough for most experiments requiring the transfer of a cloned gene, but still too low for routine transfection of the total genome of a higher plant and the subsequent recovery of transformants complementing a specific deficient allele. These experiments would require a population of l0 s - 109 protoplasts. Factor.~ a f f ~ c t l n o t r a n c f ~ t ; ~ n , ~ ¢ ¢ ; , ~ ; ~ n . . . . . . A , ~ . . our experimental conditions and previously published procedures showed similarities. A clear-cut optimum voltage was detected. Addition of carrier DNA significantly improved transfection efficiencies. However, our results differ in several respects from data previously published by Fromm et ai. [6] and Shillito et al. [5]. Our electroporation conditions are intermediate between these previously published procedures. The utilization of a 16/zF capacitor charged at 300 V resulted in a limited joule effect during electroporation, and therefore, no pre-cooling of the protoplas~ suspension was required to prevent the increased temperature resulting from the electroporation. This scaling down of the capacitor size was possible because the resistivity of the electroporation medium was simultaneously decreased. The efficiency of the transfection procedure described by Shillito et al. [5] is improved by heat - shock and PEG treatments of the protoplasts. We presume that these treatments affect the structure of the plasmalemma and increase the efficiency of an electroporation
627
performed at high voltage with a short pulse length (l 0/zs) which may not be as efficient alone to create holes in protoplasts as our conditions involving much longer pulses (100 ms). This discrepancy could explain the absence of effect of heat - shock and PEG treatments on the efficiency of electroporation under our experimental conditions. Other authors have previously shown the influence of the cell cycle on transformation efficiency [14]. In this report, we have shown for the first time that cell competence for transformation also depended upon the stem position of leaves collected for protoplast preparation. The growth conditions of the mother plant and its influence on the competence for transformation of the derived protoplasts are being studied.
Acknowledgments We wish to thank Georges Pelletier and Yves Chupeau for their suggestions and interest in this work. This study is funded by a grant from the E.E.C. (BAP program, contract 0085 F). Catherine Bellini is funded by a fellowship from the Roussel UCLAF Company.
References 1 Krens F.A., Molendkijk L., Wullems G.J. & Schilperoort R.A. (1982)Nature 296, 72-74 2 Paszkowski J., Shillito R.D., Saul M., Mandak V., Hohn T., Hohn B. & Potrykus I. (1984) EMBO J. 3, 2717-2722 3 Deshayes A., Herrera-Estrella L. & Caboche M. (1985) EMBO J. 4, 2731-2737 4 Fromm M., Taylor L.P. & Walbot V. (1986) Nature 319, 791-793 5 ShillitoR.D., Saul M.W., Paszkowski J., Miiller M. & Potrykus I. (1985) Biotechnology 3, 1099- 1103 6 Fromm M., Taylor L.P. & Walbot V. (1985) Proc. Natl. Acad. Sci. USA 82, 5824-5828 7 Holmes D.S. (1982)Anal. Biochem. 127, 428-433 8 Chupeau Y., Bourgin J.P., Missonier C., Dorion N. & Morel G. (1974) C.R. Acad. Sci. Paris Set. D 127, 428-433 9 Muller J.F. & Caboche M. (1983)Physiol. Plant. 57, 35-41 10 Shaw W.V. (1967) J. Biol. Chem. 242, 687-693 11 Cohen J.D., Eccleshall T.R., Needleman R.B., Federoff H., Bechferer B.A. & Marmur J. (1980) Proc. Natl. Acad. Sci. USA 77, 1078-1082 12 Hain R., Stabel P., Czernilofsky A.P., Steinbiss H.H., Herrera-Estrella L. & Schell J. (1985) Mol. Gen. Genet. 199, 161-168 13 L6rz H., Baker B. & Schell J. (1985) Mol. Gen. Genet. 199, 178-182
628
P. Guerche et al.
14 Meyer P., Walgenbach E., Bussmann K., Hombrecher G. & Saedler H. (1985) Mol. Gen. Genet. 201,513-518 15 Shillito R.D., Paszkowsld J. & Potrykus I. (1983) Plant Cell Rep. 2, 244-247 16 Wong T.K. & Neumann E. (1982) Biochem. Biophys. Res. Commun. 107, 584-587
17 Hashimoto H., Morikawa H., Yamada Y. & Kimura A. (1985)Appl. Microbiol. Biotechnol. 21, 336-339 18 Landgridge W.H.R., Li B.J. & Szalay A.A. (1985) Plant Cell Rep. 4, 355-359 19 Riggs C.D. & Bates G.W. (1986) Proc. Natl. Acad. Sci. USA 83, 5602-5606
621
Use of a transient expression assay for the optimization of direct gene transfer into tobacco mesophyll protoplasts by electroporation Philippe GUERCHE 1, Catherine BELLINI 1, Jean-Michel LE MOULLEC 2 and Michel CABOCHE 1 ILaboratoire de Biologic Cellulaire, Route de St-Cyr, INRA, F-78000 Versailles, and 2Soci~t~ Roussel UCLAF, F-93230 Romainville, France (Received 13-3-1987, accepted after revision 21-5-1987)
Summary - Plasmid DNA was transfected into tobacco mesophyll protoplasts by electroporation. Transfection efficiency was estimated, using a transient expression assay based on the measurement of chloramphenicol transacetylase activity or by scoring colonies expressing resistance to paromomycin, an aminoglycoside related to kanamycin. Under conditions of cell survival superior to 50% after electroporation, transient expression signals and transformation efficiencies were found to be proportional. Factors affecting the efficiency of transformation were studied. A clear-cut optimum voltage (250300 V/cm) was detected. Among various salts tested, potassium chloride was the best electrolyte. No improvement of electroporation efficiency was obtained by a h e a t - s h o c k (45°C/5 min) treatment prior to electroporation or by the presence of polyethylene glycol in the electroporation medium. The physiological state of plants used as the protoplast source significantly affected the transfection ability of the resulting protoplasts. These results are discussed and compared to previously published procedures. electroporation I tobacco / protoplast / gene transfer I transformation
R~sum~ - Usage d'un essai de l'expression transitoire pour optimiser la transfection dans des protoplastes de m~sophylle de tabac par Electropora~.ion. De l'ADNplasmidique a c~t6transfect6 darts des protoplast~,~ d: me'sophylle de tabac par ~lectroporation. L "efficacit6 a ~t6 mesur~e par dosage de l'activit6 transitoire de la chloramphdnicol transacdtylase ou par ddnombrement des colonies cellulaires transformdes exprimant une rdsistance ~ la paromomycine, aminoglycoside proche de la kanamycine. Dans des conditions de transfection oh la survie cellulaire restait supdrieure ~ 50%, les mesures d'expression transitoire et de frdquence de transformation se sont avdrdes proportionnelles. Divers facteurs susceptibles d'affecter l'efficacitO de transfection ont Otd ~tudids. Les transfections se sont avdrdes efficaces pour des voltages compris entre 250 et 300 V/cm. Le chlorure de potassium dtait le meilleur dlectrolyte parmi divers sels testds. L "efficacitd des dlectroporations n'a dtO amdlior~e ni en faisant subir un choc thermique ~ 45 °C pendant 5 min aux protoplastes avant leur dlectroporation, ni en ajoutant du polydthyibne glycol dans la chambre d'~iectroporation. L'Otat physiologique des plantes utilisdes comme sources de protoplastes semble par contre affecter de fa¢on importante l'efficacitd de I'dlectroporation. Les rdsultats obtenus sont discutds et compards ~ d'autres travaux. dlectroporation I tabac I protoplasW I transfert de gene / transformation
Abbreviations: APH3°: aminoglycoside phosphotransferase 3'; CAT: chloramphenicol transacetylase; EDTA: ¢thylenediaminetetraacetic acid; MOPS: 3-(N-morpholino) propanesulfonic acid; PEG: polyethylene glycol.
622
P. Guerche et al.
Introduction Improved methods to obtain plant tranformation m u s t be developed sc that this tool can be used to
study plant physiology arid, in the future, contribute to improve plant crops. Direct gene transfer into plant protoplasts can be achieved by various techniques. Polyethylene glycol (PEG) v, as first used successfully to transform plant protoplasts [1, 2]. In these early experiments, transformation efficiency was low (10- s _ 10- 6). Improvement of transformation efficiency was obtained either by liposome mediated transfection [3], electroporation [4] or a combination of P E G treatment and electroporation [5]. The efficiency of this last procedure is high, but seems to be affected by various parameters and requires therefore m a n y experiments to be optimized for a new type of protoplast. This observation led us to develop similar studies under our experimental conditions of electroporation and to try to simplify the procedure. Protoplasts were transfected with a plasmid carrying the chloramphenicol transacetylase (CAT) coding sequence put under the control o f the 35S C a M V promoter [4]. The transient expression of C A T was measured in cells transfected under various conditions. The relationship between transient expression signal and transformation efficiency was also studied in cotransfer experiments with another plasmid carrying the aminoglycoside phosphotransferase ( A P H 3 ' ) coding sequence put under the control of the 19S CaMV promoter.
Materials and methods Plasmids Two plasmids were used for transfection experiments. Plasmid pABD1 [2] carries the coding sequence of the aminoglycoside phosphotransferase expressed by the bacterial transposon Tn5 and conferring resistance to aminoglycosides, such as kanamycin. This coding sequence is put under the control of the 19S promoter of CaMV and polyadenylation signals derived from the nopaline synthase gene. Plasmid pCaMV [6] carries the coding sequence of chloramphenicol transacetylase derived from pBR328. This coding sequence is put under the control of the 35S promoter of the CaMV and polyadenylation signals derived from the nopaline synthase gene. Plasmids were extracted according to the procedure of Holmes [7] and purified by isopycnic centrifugation in a CsCl gradient. Calf thymus DNA (Boehringer, Mannhelm) was sonicated to produce fragments of average size 1 kb and used as carrier DNA. Plasmid DNA and calf thymus DNA were resuspended at a concentration
of 1 tzg/ml in the electroporation buffer (0.5 M mannitol, 5 mM KCI and 200 ~M MOPS, pH 7.2, sterilized by autoclaving).
Protoplasts Protoplasts were isolated essentially as described by Chupeau [8] from the leaves of Nicotiana tabacum (cv. Xanthi, diploid clone D8) grown in the greenhouse and washed twice in the electroporation buffer (0.5 M mannitol, 200/zM MOPS, pH 7.2, and 5 mM KCI). Protoplasts were resuspended in this buffer at a concentration of 2 x 106/ml for electroporation. Electroporation Electroporation conditions were adapted from previously published procedures [4, 5]. The plexiglass electroporation chambers were of two types: A (length 2 cm, width I cm, depth 0.5 cm), or B (length, width and depth I cm). Two stainless steel electrodes were put on the more distant sides of the cuvettes and connected to the electrical system. The generator was an ISCO 490 power supply coupled with one or two 16 ~F capacitors put in parallel. The charged capacitors were discharged into the electroporation cuvette. An oscilloscope was used to control the electrical discharge through the cuvette. A complete discharge of the capacitor through the cuvette lasted 200 ms. To I ml of protoplast suspension, 10ttg of the plasmid and 50 t~g of calf thymus DNA were added. Electric pulses were performed at 5 s intervals. Under standard conditions, 3 pulses (300 V/cm) were performed in the type B cell. Cvlture of electroporated protoplasts and selection of t r A,7 ~ t~ ~ A , i r m a ~ l ' r ~ lf ~
Electroporated protoplasts were diluted immediately after electroporation in the culture medium To [8] at a density of 6 x 104 protoplasts/ml and incubated in the dark for 6 days. The resulting colonies were diluted in medium C [9] at a density of 10 4 colonies/ml and incubated under light. Paromomycin (20 ~g/ml) was used as a substitute for kanamycin to select for aminoglycosideresistant colonies in liquid medium C. Resistant colonies were scored 1 month later.
Chloramphenicol acetyltransferase transient expression assay Cells derived from electroporated protoplasts incubated in medium To were harvested 3 days after electroporation. Cells from 2 dishes were washed twice in medium KL (0.3 M KCI, 5 mM CaCI2) and resuspended in 1 ml of the CAT buffer (1/.tM leupeptine, 30 mM ascorbic acid, 20 mM EDTA, 250 mM Tris-HCl, pH 8.0) containing 2.5% polyclar powder. Cells were broken by a brief sonication (Branson soaifier, power setting 3, 30 s). The resulting lysate was heated at 60°C for 5 min and clarified by centrifugation for 10 rain at 4°C in an Eppendorf centrifuge. Supernatants were collected and stored frozen at - 2 0 ° C . CAT activity was measured as described prevint+sly [10]. The reaction mixture contained 100/zl
Electroporation of tobacco mesophyll protoplasts
623
of cell extract, 20~1 of acetyl CoA (4 mM, pH 8.3, Boehringer) and 5 t~l of [14Clchloramphenicol (58.5 Ci/mol, 1.7 mM, New England Nuclear). After incubation for 1 h at 37°C, the reaction was stopped by adding 200/.tl of Tris base (2.2 M) and reaction products were extracted twice with 700/~l of benzene. The benzene extract was dried in vacuo in siliconized Eppendorf tubes in a speed vac. Samples were solubilized in 20/~l of ethylacetate and subjected to thin-layer chromatography on silica gel plastic plates (Merck) using a 1:9 (v/v) mixture of methanol and chloroform. Dried plates were exposed to X-ray film. Spots corresponding to acetylated derivatives of chloramphenicol were cut out of the plastic plates and their radioactivity measured by liquid scintillation.
tion, endogenous tobacco CAT activity is negligible and accounts for less than 1°70 of transgenic CAT activity. CAT expression was followed for a period of 6 days in a batch of mesophyll protoplasts transfected with pCaMV CAT. Maximal levels of CAT activity were detected on the third and fourth days of culture and CAT measurements were therefore routinely, taken on the third day after electroporation.
Results
Voltage was critical in obtaining a maximal transient expression. A clear-cut optimum was obtained for a voltage of 300 V (Fig. 1), although a s~gnificant expression was detected also for 250 V. Similar results were obtained when experiments were performed with a higher number of pulses (10, 20 or 30 pulses per voltage condition (results not shown). During electroporation, the electron flow through the electroporation chamber is delivered by connecting a charged capacitor to the two electrodes of the chamber. This results in an exponen-
Measurement o f the transient expression of CA T activity in protoplasts transfected with the plasmid p C a M V CA T The assay of E. coli CAT activity was chosen as a convenient way of measuring transient expression signals in plant cells. This technique, commonly used to study gene expression in eukaryotic cells [11], had to be adapted to plaut cells. Preliminary experiments had shown that E. coli CAT activity is partially destroyed when mixed with plant cell extracts. This difficulty was overcome by stabilizing extracts with reducing agents, such as ascorbic acid (mercaptoethanol interferes with the CAT assay) and by adding poiyciar, an insoluble matrix binding phenolic compounds efficiently. Polyphenol oxidases were destroyed by heat inactivation at 60°C and inactivation with EDTA metals, such as Ca 2+, known to inactivate these enzymes. Under these extraction conditions, E. coli CAT activity was stable in extracts (results not shown). Transient expression experiments performed with various plasmids carrying the CAT structural gene under the control of prokaryotic or retroviral promoters were unsuccessful. In the controls, a signal corresponding to acetylated derivatives of chloramphenicol was detected, using the protocol described under Materials and Methods. This signal could be assigned to an endogenous tobacco CAT activity, since the activity was destroyed by boiling the extract prior to the incubation, and since the reaction was dependent upon the presence of acetyl CoA in the reaction mixture (results not shown). E. coli CAT activity is thermostable and inactivated only by treatments at temperatures above 65°C. Unfortunately, the cellular CAT activity displayed a thermal stability comparable to E. coli CAT activity. Under optimized conditions of electropora-
Influence of voltage, capacitor size and the number o f pulses on the efficiency o f transfection
r/) wa
0~
m
0 O r.. @
o E e~ r.)
X v
°~ > °~ 4-1 ¢.)
F-.
!
0
200
400
V01ts/cm Fig. 1. Influence of pulse voltage on the transient expression of CAT activity in tobacco mesophyli protoplasts transfected by electroporation with the plasmid pCaMV CAT. Electroporations were performed in the electroporation cell B as described under Materials and Methods at varied voltages, using a single 16 ~F capacitor and a series of 5 pulses.
P. Guerche et al.
624
tial discharge of the capacitor through the cell. The half-life of this discharge, recorded with an oscilloscope, is approximately 120 ms for a single 16/~F capacitor. When two capacitors are connected in parallel and discharged into the cell, this results in a 2-fold increase in the duration of the pulse. A measurement of CAT activity in protoplast cultures electroporated with one or two capacitors and increasing pulse numbers is presented in Fig. 2. For one capacitor, CAT transient expression is roughly proportional to the number of pulses, reaches a maximum, and then decreases for pulse numbers higher than 20 for cell A and 10 for cell B. This decrease correlates with reduced viability of protoplasts in the cultures (results not shown). Using two instead of one capacitor results in an approximately 2-fold decrease in the number of pulses required for optimal CAT expression. For a given voltage, the efficiency of eleetroporation seems to be related to the total amount of charge migrating through the cell, rather than to the duration of pulses.
Influence of salts on the efficiency of electroporation It is generally assumed that divalent cations play a critical role in the transformation of various organisms. We have compared the efficiency of electroporation in the presence of K +, Li +, M g 2 +, Ca 2+ and Zn 2+ at a 5 mM concentration (Fig. 3). K + ion was found to be most suited for efficient
10 • ,
U'J
g
"~
10
¢o
"5
s
"-1
0
Number of pulses Fig. 3. Influence of the nature of different salts on the efficienc; of electroporation. Tobacco mesophyll protoplasts were transfected with the plasmid pCaMV CAT in the electroporation buffer 5 mM KCi (&) which was replaced by the same concentration of other salts: LiCI ( • ) . MgCl 2 ([] ~, ZnCI 2 (e) CaC~z (,k) according to the exper:,ment.
electroporation. When the Ca 2+ concentration was decreased 2-fold in the electroporation buffer in order to keep a conductivity equivalent to 5 mM KCI, this resulted in efficiencies reaching values obtained with 5 mM KCI. On the contrary, increasing the KCI concentration to 25 mM resulted in a 10-fold decrease of transient expression signals (results not shown). Conductivity is therefore probably the main parameter involved in salt effects on electroporation efficiency.
Comparison of transient expression signals and transformation efficiencies
o
"
X
v
0
lO 2o 3o Number of pulses
Fig. 2. Influence of capacitor characteristics on the transient expression of CAT activity in tobacco mesophyll protoplasts transfected by electroporation with the plasmid pCaMV CAT. Electroporations were performed in the electroporation cell A at 300 V using one ( 0 ) or two (<>) 16/~F capacitors and variable numbers of pulses.
In previous works [2-4, 12-14], protoplasts transfected with a plasmid conferring resistance to kanamycin were subsequently used to select transformants on the basis of their resistance to kanamycin. Under selective conditions, kanamycinsensitive cells can prevent the development of kanamycin-resistant colonies, presumably by releasing toxic substances into the culture medium. This could lead to inaccurate measurements of transformation efficiencies. This difficulty was overcome
Eiectroporation of tobacco mesophyll protoplasts by Shillito et al. [15] who used agarose embedded cultures incubated in selective liquid medium containing kanamycin and transferred repeatedly into fresh medium. By replacing kanamycin with paromomycin, an aminoglycoside also detoxified by the aminoglycoside phosphotransferase (APHY) coded by Tn5 and pABD1, we could select transformants efficiently at high col! densities as shown in a reconstruction experiment presented in Table I, without having to resort ~,o time-consuming replenishment of the selective medium. Transformants were selected on the basis of paromomycin resistance in protoplast cultures cotransfected with two plasmids, pABD1 and pCaMV CAT. Simultaneously, part of the electroporated cells were used for transient expression and plating efficiency measurements (Fig. 4). When cultures were transfected under conditions in which protoplast viability remained higher than 50070 of the control, a proportionality between the transient expression signal and transformation efficiency was observed. When viability was reduced, transformation efficiency dropped more rapidly than the transient expression signal. A comparison of the transient expression signal and transformation efficiency was performed at constant pulse number
Table I. Influence of cell density on the recovery of kanamycin-rcsistant colonies. No. o f cells plated in 5 ml o f m e d i u m C
No. o f colonies growing in media containing
clone D8
Ka4
kanamytin (70 t~g/hi)
paromomycin (20 tzg/ml)
100 100 100 I00 100 100 100 0
37 --7 39 36 23* 0 0 0
33 32 38 42 36 32 38 0
0 10000 25 000 50000 100000 200 000 500000
25000
Protoplasts were prepared from the leaves of wild type tobacco (D8) and from a kanamycin-resistant tobacco (Ka4) obtained by transfection of pLGV23 neo [3]. These protoplasts were induced to divide and 4 days later they were mixed and sujected to selection of paromomycin- or kanamycin-resistant colonies in medium C for I month. Resistant colonies were ~hen scored. The plating efficiency of D8 and Ka4 protoplast-Oerived cells grown in medium C in the absence of selective pressure was approximately 35%. * Very small colonies.
625
g g k 30%
;/ i-
~,
__
.%
|
o
I0
v
"z z CD
"1
)
5
"I Number of pulses
10
Fig. 4. Comparison of transient expression signals and transformation efficiencies in tobacco mesophyil protoplasts cotransfected with a 1:1 mixture of pABDI and pCaMV CAT: influence of a h e a t - shock treatment. Prior to electroporation, part of the protoplast suspensions was subjected to a 5 rain heat-shock at 45°C (&) or not ( • ) . Protoplasts were dectroporated in cell B at 300 V using one capacitor and variable numbers of pulses. Plating efficiency Le., the proportion of pro•,,v,-o~ able to divide was measured 6 days after protopiast plating ( - - - ) . Transient expression of CAT activity was measured 3 days after transfection (--) and paromomycinresistant colonies were selected and scored in the remaining cultures as described under Materials and Methods (----).
and various voltages and showed again a propor tionality between these two measurements of transfection efficiency (results not shown).
Influence o f other treatments on the efficiency o f transfection by electroporation PEG and h e a t - shock can affect cell competence for transformation [5]. Under our electroporation conditions, no significant effect of 10°70 PEG on the transient expression of CAT activity was detected in electroporated protoplasts and a higher PEG concentration (20070) was even deleterious (Fig. 5). A h e a t - shock (45°C, 5 rain) was performed on a batch of protoplasts and this batch as well as non-treated protoplasts were electroporated in parallel with a mixture of pCaMV CAT and pABD1 and used for transient expression ex-
P. Guerche et al.
626
T a b l e l l . Influence o f plasmid and carrier D N A concen-
trations on the efficiency of electroporation.
• PEG 0% v .. I0% [:] , , ZO% 0 0 e_ r~
pCaMV CAT (~g)
C a l f thymus DNA (/,tg)
C A T activity (cpm/105 protoplasts)
0 5 10 20 30
50 50 50 50 50
205 6 200 10800 25 400 50300
10 30
0 0
740 6 500
o
o
E 0
X
/
._~ U
m
I-0 o
'
"
~1
Number
"
• lb
" of
pulses
Fig. S. Effect of the polyethylene glycol concentration on the efficiency of electroporation. Tobacco mesophyll protoplasts were electroporated with the plasmid pCaMV CAT in cell A at a voltage of 300 V, using a single capacitor (16 I~F) and in the absence (11) or in the presence of 10% ( V ) or 20% ( [] ) PEG 6000. CAT transient expression signals were measured 3 days later as described under Materials and Methods.
periments and transformant selection experiments. The data presented in Fig. 4 does not show any significant effect of a heat - shock on transient expression signal as well as on transformation efficiency. The presence of carrier D N A significantly increased electroporation efficiency (Table II). Transient expression signals increased roughly with the m o u n t of pCaMV CAT plasmid added, in the protoplast suspension: for a same amount of plasmid, the transient expression was increased 10-fold when 50/zg of carrier D N A was added, contrary to experiments in which no carrier D N A was added.
Influence of the source o f protoplasts used for electroporation on the efficiency o f transfection When plants are grown at low light intensity, all their leaves can be used as sources of dividing protoplasts [8]. Leaves were collected separately from different parts of these plants and used as sources of protoplasts for electroporation. Differences were observed as regards the ability of these batches of protoplasts to express CAT activity after transfection with the plasmid pCaMV CAT (Table III). In two independent experiments, the highest expres-
(300 V and 16/~F, 10 pulses in cell B). Tobacco mesophyll protoplasts were electroporated as described under Materials and Methods with plasmid pCaMV CAT and calf thymus DNA. Transient expression of CAT activity was measured in transfected cultures 3 days later.
T a b l e HI. Influence o f leaf position on the mother plant
on the transfection ability of the corresponding protoplasts. L e a f level
C A T activity (cpm/10 s protoplasts)
16 11 6 1
11 800 24500 16000 4600
~ --, --, ~
apex 15 10 5
No electroporation
280
Plants were grown under light until they developed 20 expanded leaves. Leaves were collected at 4 levels on these plants and used for protoplast preparation. Protoplasts were transfected under standard conditions with pCaMV CAT (300 V and 16/,tF, 10 pulses in cell B) and the transient expression of CAT activity was measured in transfected cultures 3 days later.
sion was obtained with the fully-expanded leaves located below the rapidly growing leaves of the apex area. Discussion Electroporation is an efficient technique for the transfer of naked D N A in eukaryotic cells and has been successfully used with mammalian cells [16], yeast [17], and more recently with plant cells [ 4 - 6 , 18, 19]. We have reinvestigated the condi-
Electroporation o f tobacco mesophyil protoplasts
tions of transfer of plasmid DNA into tobacco mesophyll protoplasts using this technique. For the first time, the efficiency of electroporation was studie,.~ by measuring the transient expression of CAT activity encoded by the plasmid pCaMV CAT and by simultaneously scoring the number of paromomycin-resistant clones induced by the stable expression of APH3' activity encoded by the plasmid pABD1. It was found useful to optimize the conditions of extraction of CAT, since this enzyme was destroyed in crude extracts. The data obtained by these two optimized approaches agreed well, as long as electroporation conditions did not strongly affect the viability of treated protoplasts. From this correlation, it is concluded that transient activity measurements in transfected protoplast cultures can be used to rapidly optimize conditions of stable transformation of protoplasts from various plant species. Paromomycin was found to be more efficient than kanamycin for the efficient selection of transformants. Under standard transfection conditions, the frequency of transformants in electroporated cultures varied from 1 to 5 x l 0 -3 transformants per protoplast able to divide after electroporation. This frequency is high enough for most experiments requiring the transfer of a cloned gene, but still too low for routine transfection of the total genome of a higher plant and the subsequent recovery of transformants complementing a specific deficient allele. These experiments would require a population of l0 s - 109 protoplasts. Factor.~ a f f ~ c t l n o t r a n c f ~ t ; ~ n , ~ ¢ ¢ ; , ~ ; ~ n . . . . . . A , ~ . . our experimental conditions and previously published procedures showed similarities. A clear-cut optimum voltage was detected. Addition of carrier DNA significantly improved transfection efficiencies. However, our results differ in several respects from data previously published by Fromm et ai. [6] and Shillito et al. [5]. Our electroporation conditions are intermediate between these previously published procedures. The utilization of a 16/zF capacitor charged at 300 V resulted in a limited joule effect during electroporation, and therefore, no pre-cooling of the protoplas~ suspension was required to prevent the increased temperature resulting from the electroporation. This scaling down of the capacitor size was possible because the resistivity of the electroporation medium was simultaneously decreased. The efficiency of the transfection procedure described by Shillito et al. [5] is improved by heat - shock and PEG treatments of the protoplasts. We presume that these treatments affect the structure of the plasmalemma and increase the efficiency of an electroporation
627
performed at high voltage with a short pulse length (l 0/zs) which may not be as efficient alone to create holes in protoplasts as our conditions involving much longer pulses (100 ms). This discrepancy could explain the absence of effect of heat - shock and PEG treatments on the efficiency of electroporation under our experimental conditions. Other authors have previously shown the influence of the cell cycle on transformation efficiency [14]. In this report, we have shown for the first time that cell competence for transformation also depended upon the stem position of leaves collected for protoplast preparation. The growth conditions of the mother plant and its influence on the competence for transformation of the derived protoplasts are being studied.
Acknowledgments We wish to thank Georges Pelletier and Yves Chupeau for their suggestions and interest in this work. This study is funded by a grant from the E.E.C. (BAP program, contract 0085 F). Catherine Bellini is funded by a fellowship from the Roussel UCLAF Company.
References 1 Krens F.A., Molendkijk L., Wullems G.J. & Schilperoort R.A. (1982)Nature 296, 72-74 2 Paszkowski J., Shillito R.D., Saul M., Mandak V., Hohn T., Hohn B. & Potrykus I. (1984) EMBO J. 3, 2717-2722 3 Deshayes A., Herrera-Estrella L. & Caboche M. (1985) EMBO J. 4, 2731-2737 4 Fromm M., Taylor L.P. & Walbot V. (1986) Nature 319, 791-793 5 ShillitoR.D., Saul M.W., Paszkowski J., Miiller M. & Potrykus I. (1985) Biotechnology 3, 1099- 1103 6 Fromm M., Taylor L.P. & Walbot V. (1985) Proc. Natl. Acad. Sci. USA 82, 5824-5828 7 Holmes D.S. (1982)Anal. Biochem. 127, 428-433 8 Chupeau Y., Bourgin J.P., Missonier C., Dorion N. & Morel G. (1974) C.R. Acad. Sci. Paris Set. D 127, 428-433 9 Muller J.F. & Caboche M. (1983)Physiol. Plant. 57, 35-41 10 Shaw W.V. (1967) J. Biol. Chem. 242, 687-693 11 Cohen J.D., Eccleshall T.R., Needleman R.B., Federoff H., Bechferer B.A. & Marmur J. (1980) Proc. Natl. Acad. Sci. USA 77, 1078-1082 12 Hain R., Stabel P., Czernilofsky A.P., Steinbiss H.H., Herrera-Estrella L. & Schell J. (1985) Mol. Gen. Genet. 199, 161-168 13 L6rz H., Baker B. & Schell J. (1985) Mol. Gen. Genet. 199, 178-182
628
P. Guerche et al.
14 Meyer P., Walgenbach E., Bussmann K., Hombrecher G. & Saedler H. (1985) Mol. Gen. Genet. 201,513-518 15 Shillito R.D., Paszkowsld J. & Potrykus I. (1983) Plant Cell Rep. 2, 244-247 16 Wong T.K. & Neumann E. (1982) Biochem. Biophys. Res. Commun. 107, 584-587
17 Hashimoto H., Morikawa H., Yamada Y. & Kimura A. (1985)Appl. Microbiol. Biotechnol. 21, 336-339 18 Landgridge W.H.R., Li B.J. & Szalay A.A. (1985) Plant Cell Rep. 4, 355-359 19 Riggs C.D. & Bates G.W. (1986) Proc. Natl. Acad. Sci. USA 83, 5602-5606