PROTOCOL

A Simple Method for the Production of Highly Competent Cells of Agrobacteriumfor Transformation via Electroporation A. C. McCormac, M. C. Elliott, and D. F. Chen*

Abstract The introduction of binary plasmids into Agrobacterium hosts for Agrobacterium-medialed transformalion of plants is most readily achieved by electroporation. However, occasionally, no transformed colonies are recovered and the transformation program is delayed. Poor transformation rates are commonly associated with particular combinations of Agrobacterium strains and plasmid-selection markers. In order to avoid this problem, it is important for the bacteria to have a highly competent status for reception of plasmid DNA. It is also important to optimize the level of antibiotic for the selection of transformed colonies. In this article, we demonstrate that transformation competence is strongly related to the phase of growth at which a bacterial culture is prepared for elcctroporation, and we describe a simple procedure !ha! allows the level of transformation-competent cells to be maximized. We have observed that there is significant variation between transformed Agrobacterium strains in the levels of antibiotic tolerance: we define the antibiotic levels that are appropriate for selection of three Argohacterium tumefaciens (EHAI01, LBA4404. C58) and two Argobacterium rhi=ogenes(LBA9402, Ar2626) strains, transformed with three ahernative resistance markers (spectinomycin ~ . kanamycin re*, and genlamycinr~'). Index Entries: Agrobacterium; binary plasmids: electroporation;transformation; competence: selection.

encompasses three plasmid backbones to provide a choice of bacterial selection markers; namely kanamycin~% spectinomycin re~ or carbenicillin r~~ + gentamycin res. In general, the introduction of plasmids into Agrobacterium using electroporation is a highly effective technique. However, when attempting to introduce certain plasmids into the required Agrobacterium strains by electroporation, we have found that the frequency of obtaining transformed colonies was so low as to be unreliable. Such difficulties were often associated with p a r t i c u l a r Agrobacterium s t r a i n s a n d , for example, were especially obvious when selecting for plasmids carrying the spectinomycin-resistance marker (pBECKS400) within cells of A. tumefaciens strain LBA4404.This unreliability has been shown to relate, in large part, to variations between Agrobacterium strains in the following two aspects: their competence to receive

1. Introduction Following the creation of a T-DNA-containing plasmid, it is necessary to introduce the construct into an Agrobacterium host in order to provide a b i o l o g i c a l v e h i c l e for plant t r a n s f o r m a t i o n . Electroporation is currently the method of choice for introducing plasmids into Agrobacterium cells (e.g., 1). This technique is not only more rapid than mobilization of the plasmid from an Escherichia coli host via triparental mating, but also removes constraints on selection of the introduced plasmid, which may be imposed by chromosomaIly located resistance genes in the donor and helper bacteria. Modern vector stratagies exploit this by incorporating various selectable genes on the plasmid replicons, thereby aiding compatability with a wide-range of Agrobacterium genotypes. In this vogue, a new vector series, pBECKS, has been created (described in detail in ref. 2) which

*Authorto whomall correspondenceand reprintrequestsshouldbe addressed.The NormanBuildingInstitutefor Plant ScienceResearch, I De MontfortUniverisity,Scraptoft,Leicester,LE7 9SU, UK. E-mail:[email protected] Molecular Biotechnology 9

Mo[~(uLnn 810TECHNOLOGV

Humana Press Inc. All rights of any nature whatsoever reserved. 1073-6085/1998/9:2/155-159/$9.25

1 55

Volume 9, 1998

156

McCormac, Elliott, and Chen

plasmid DNA and/or express the plasmid resistance genes and heir sensitivity to levels of the antibiotic agents used to select for the binary plasmids. Low competence of a given batch of bacteria for transformation is also clearly related to the growth stage at the time of harvest and we have defined, therefore, a simple method for regulating the growth of Agrobacterium cultures in order to obtain consistently high levels of transformation-competent cells. We have also established the levels of antibiotics that are effective for the selection of each of three A. tumefaciens and two A. rhizogenes strains after electroporation with the pBECKS vectors carrying the spc/spm re~,kan "es or gent"~s selection markers.

2. Materials 1. Bacterial strains: E. coli (recA, IacZAM15): XL 1-blue (Stratagene, Cambridge, UK); DH5c~.

A.tumefaciens: EHA101 (3): LBA4404 (4): C58. A. rhizogenes: LBA9402; Ar2626. 2. Orbital incubator providingl00-200 rpm and a temperature of 28~ 3. Sterile centrifuge tubes (50 mL volume, disposable) with screw-caps (Elkay Products, MA). 4. Media for bacterial culture: LB-(for culture of E. coli and A. tumefaciens strains)-10 g/L Bactotryptone, 5 g/L Bacto-yeast extract, 5 g/L NaC1, agar added to 1.2 w/v % where appropriate (5). YMB-(for culture of A. rhizogenes strains)-10 g/L Mannitol, 0.4 g/L Yeast extract, 0.2 g/L MgSO4.7H20, 0.1g/1 NaC1, 0.5 g/L K~HPO4, agar added to 1.0 w/v % where appropriate (6). 5. Antibiotics: Kanamycin, gentamycin, spectinomycin (Sigma-Aldrich, Dorset, UK)-made to stock solutions of 50 mg/mL in water and filter sterilized; stored at -20~ 6. Plasmid DNA isolation: Alkaline lysis, phenol/ chloroform partitioning, and ethanol precipitation according to standard procedures (5). 7. Electroporation system: BIO-RAD Gene Pulser with pulse controller unit. Cuvets with 0.1 cm electrode gap. (Bio-Rad, Hercules, CA). Electroporation buffer: 10% v/v glycerol (sterile).

3. Methods 3.1. Plasmid Isolation Plasmid DNA was readily isolated from E. coli by standard alkaline-lysis techniques (5) (see

MOLECUtRABtOTECHNOtOGY

Note 1). The plasmids pBECKS 19, pBECKS400, and pBECKSgen cany, respectively, the genes for bacterial resistance to kanamycin, spectinomycin, or carbenicillin + gentamycin. These plasmids and the derivative series of constructs are described in (2).

3.2. Protocol for Transformation of Agrobacterium via Electroporation Bacterial cultures were prepared for electroporation by pelleting the ice-chilled cells for 5 rain at 3 Krpm and washing three times with ice-cold 10% v/v glycerol.The final pellet was resuspended in 0.01-0.02 volumes of 10% glycerol. Forty ~tL of these competent cells were mixed with approx 1 ~tg plasmid DNA (< 10 ~L volume) and electroporated in a 0.1-cm cuvet (ice-chilled) using a Bio-Rad Gene pulser set to parameters 12.5 Kv/cm, 25 ~tF, 400 ohms. Prior to plating onto agar medium, supplemented with antibiotics, cells were diluted with 5 mL liquid medium (see Note 2) and grown for 1 h at 28~ in an orbital incubator.

3.3. Problems Experienced in Obtaining Transformed Agrobacterium Strains Two factors were identified that could compound one another to impede recovery of transformed colonies, namely strain-dependent variations in the sensitivity of the Agrobacterium ceils (both native and transformed) to the selective agent and differences in the levels of viable/competent cells recovered in preparations of cultures. The latter effect showed notable differences between Agrobacterium strains and also between independent preparations of the same strain (data not shown). Low levels of viable cells were found to relate, in the main, not to losses during the pretreatment and electroporation processes, but rather to the growth-stage of the bacterial cultures at the time of preparation for electroporation.

3.4. Maximizing Cell-Competence for Transformation Competent cells to be transformed with binary plasmids were isolated from liquid cultures of the Agrobacterium, which had been inoculated from streak plates and incubated at 28~ in an orbital incubator for the stated periods. "Open" cultures

Volume 9, 1998

Argobacteri

um

Transformation

157

LBA4404

EHA101 were grown in 50 mL of medium contained within a 250-mL flask, which was sealed with a paper bung; such well-aerated cultures displayed typical rapid (approx l o g - l i n e a r ) and s t a t i o n a r y growth phases. Bacterial cultures were assessed at various stages of growth with respect to their optical density (ODe,00) (Fig. 1A), concentration of viable cells (Fig. 1B) and the number of transformed colonies which could be recovered following electroporation in the presence of the plasmid pBECKS400 (2); the latter was expressed according to the total n u m b e r of t r a n s f o r m e d colonies per lug of plasmid DNA (Fig. 1C) or as a proportion of the number of viable bacterial cells that were present during electroporation (Fig. ID). Cultures entering stationary phase displayed very poor competence for transformation (Fig. ! C,D-open 40h) compared to cultures still showing rapid growth (Fig. 1C,D-open 16h). Variations in absolute growth times, not only between different Agrobacterium strains, but also between independent inoculations of the same strain, made it very difficult to identify a particular incubation period as optimal for the recovery of m a x i m u m numbers of competent cells. Individual measurements of ODe,00 p r o v e d p o o r l y c o r r e l a t e d to viability levels (compare Figs. 1A and B) and to transformation competence (compare Figs. 1A

1.5-

(a)

0

1

~ x:: + +

1E+10-

(b) E IE+09-

IE+08

!! !!

1E+11

+ + (c)

,~

1E+06 -

"~

1E+05

"~

1E+04

1E+03 1E+02

~ 1g+02 1E+01

--

~

fill .~

1E-05

1E- 0

8=

. " . ~. . .

1E-08~

,.~ + +

+ +

Fig. 1. The competence of Agrobacterium cells for e l e c t r o p o r a t i o n - m e d i a t e d transformation with pBECKS400. A. tttmefaciens strains EHAI01 and LBA4404 were grown in liquid LB medium at 28~ in an orbital incubator. For 'open' cultures, 50 mL of inoculated medium was held in a 250-mL flask sealed with a paper bung; growth and transformation parameters were measured 16 and 40 h after initial inoculation (open: 16 h, 40 h). 'Closed' cultures consisted of 20 mL of inoculated medium held in a 50-mL screwcapped tube and parameters were measured from sealed tubes that had been incubated for the stated times (closed: 16 h, 40 h). Other closed cultures were opened after 16 h or 40 h and allowed to equilibrate with the air; these were subsequently resealed and returned to the incubator for a further 4 h, and growth and transformation parameters were then measured (closed: 16 + 4 h, 40 + 4 h). Optical density of the bacterial cultures (ODe,00) was measured using a colo-

MOtECUtnnBIOT~ZCHNOIOGV

|

| open

closed

open

cbsed

incubation conditions

rimeter with a 600nm filter. The number of viable cells was measured by counting the number of colonies arising from serial dilutions of the culture plated onto antibiotic-free medium. Transformation competence was assessed by counting the number of spectinomycin-resistant colonies, which resulted following electroporation in the presence of 1 lag DNA of pBECKS400. The data shown are from a single, representative experiment. The transformed cells of A. tmnefaciens EHA101 were selected on LB supplemented with 200 mg/L spectinomycin; LBA4404 was selected on LB containing I00 mg/L (see Table 1) (cells were allowed to grow at 28~ for 1 h prior to plating onto selection medium).

Volume 9, 1998

158

McCormac, Elliott, and Chen Table 1 Effective Levels of Antibiotic-Supplements(mg/L) for the Selection of the Binary Vectors pBECKS19, pBECKS400, pBECKSgen (and derivative plasmids, ref. 2) in five Agrobacterium Strains Plasmid progenitor (antibiotic selective agent)

pBECKS 19 (kanamycin)

pBECKS400 (spectinomycin)

pBECKSgen (gentamycin)

"N/A 50 100

200 100 200

60 30 60

50 50

100 100

30 30

Agrobacteriumtumefaciens: EHAI01 LBA4404 C58

Agrobacterium rhizogenes: LBA9402 Ar2626

"EHA101 carrieskanamycinresistanceon the disarmedTi-plasmid. and C). Furthermore, even viability levels could not be used to predict the competence of a given culture for transformation (compare Figs. 1B and C). Clearly, not all viable cells have the capacity to produce transformed colonies, and this capacity is dramatically reduced in cells isolated from stationary cultures (Fig. 1D-open 40h). 'Closed' cultures were grown in 50-mL screwcapped centrifuge tubes containing 20 mL of medium (and 30 mL of air); tubes were held horizontally in the orbital incubator to ensure full mixing of air and medium. Limiting the air-supply to a culture through such containment caused the growth to be restricted. Between approximately 16 and 40 h of incubation, closed cultures maintained a relatively constant, intermediate level of viable cell numbers; a constant OD600 value suggests that these cells had a low turnover rate (Fig. 1A and B) and so were probably in a semi-arrested state. Aeration of closed cultures was performed by removing the lids under sterile conditions for approx 30 s (with slight agitation) before resealing and returning to the orbital incubator. Such aeration of 16-40 h old cultures stimulated a phase of rapid growth comparable, in rate, to the log-linear phase of an open culture. Preparation of cells after 4 h of such rapid growth consistently allowed the recovery of high numbers of viable cells (Fig. 1Bclosed 16 + 4 h, 40 + 4 h), a large proportion of which were able to receive and express the plasmid DNA introduced by electroporation (Fig. 1Dclosed 16 + 4 h, 40 + 4 h). This procedure has MOLECUtRRBIOTECHNOLOOY

proved effective for the routine preparation of competent cells from all of the Agrobacterium strains tested so far in this laboratory, namely EHAI01, LBA4404, C58, LBA9402, and Ar2626. Furthermore, this technique obviates the inconvenience of precise long-term planning or monitoring of culture growth because closed cultures, inoculated anywhere between 16 or 40 h earlier, can be brought to near optimal levels within 4 h. This is contrasted with the situation for "open" cultures, where the window of opportunity is restricted to a few hours and is difficult to predict accurately.

3.5. Storing Competent Cells Once prepared for electroporation by washing in 10% glycerol, these competent cells can be stored at -70~ After l 0 wk of storage at -70~ aliquots of EHA101 retained 78% of their original levels of viable cells and the competence of these viable cells to produce transformed colonies, following electroporation with plasmid DNA, was undiminished by the storage process. Storage at -20~ was found to lead to large reductions in viability and, hence, transformation competence of a given aliquot after 10 wk at this temperature was typically less than 10% of original levels.

3.6. Appropriate Antibiotic Levels for Different Agrobacterium Strains Variations in the sensitivity of Agrobacterium strains to antibiotic selection were addressed by establishing, for each genotype, the level required to fully inhibit the growth of nontransformed

Volume 9, 1998

Argobacteriu m Transformation

colonies while allowing the recovery of maximum numbers of transformed colonies. The levels of kanamycin, spectinomycin and gentamycin that proved appropriate for the selection of A. tumefaciens strains E H A 101, LBA4404, and C58 and A. rhizogenes strains L B A 9 4 0 2 and Ar2626, following the introduction (via electroporation) of the binary vectors pBECKS19, p 400 and p__gen (and their derivatives), are shown in T a b l e 1.

3.7. Preparation of Piasmids from Transformed Agrobacterium Small scale plasmid preparations from transformed, antibiotic-resistant Agrobacterium colonies can be performed by alkaline lysis (5), as for E. coli. This can therefore be used to confirm the presence of the introduced binary plasmid and the products of endonuclease restriction digestion can be directly assessed on an agarose gel to check against recombination events (see Notes 3 and 4). We used this technique to screen representative numbers of colonies, recovered following electropotation, to confirm that they had been transf o r m e d with the p l a s m i d and h e n c e that the selection strategy was effective.

4. Notes 1. If plasmid DNA has been prepared using commercial kits such as the 'Wizard DNA Purification System' (Promega, Hants, UK), carry-over (presumably from the 'wash' solution) can interfere with electroporation; a very short time-pulse is often recorded. Therefore, it is advisable to precipitate the DNA in 2 vols. ethanol + 0. l vols. 3 M Sodium Acetate, pH 4.8 and resuspend the dried pellet in water, before attempting to introduce into A grobacterium cells via electroporation. 2. Agrobacterium cells can be diluted in only 0.5 mL (instead of 5 mL) of medium for the postelectroporation recovery period without adverse effect; this can be a means to concentrate the number of transformed colonies on each selection plate in cases of poor transformation freq u e n c y , such as found with L B A 4 4 0 4 + pBECKS.400. But note: such concentration of cell numbers can encourage high backgrounds on selection medium, especially, for example, with EHA 101 on 200 mg/L spectinomycin.

fV~OLECULRRBIOTECHIVOtOGV

159

3. Low yields of plasmid DNA are recovered from A. tumefaciens cells and it is difficult to obtain good purity. Partitioning with phenol and chloroform is particularly important to allow digestion with restriction endonucleases, and even then we have found that good results can only be obtained with certain enzymes; EcoRI is recommended if it is an option and PstI, HindIII, BamHI, BglII, and XbaI can also be effective. Restricted products can usually be viewed directly on an ethidium-stained agarose gel (without the need for Southern-blot analysis); if the DNA is not of adequate purity, smearing may result. 4. Plasmid D N A isolated from A. rhizogenes strains Ar2626 and LBA9402 appears to be particularly resistant to digestion with endonucleases. It has therefore been found necessary in these cases to use the isolated DNA to transform E. coli cells; the recovery of antibioticresistant E. coli colonies confirms the presence of the plasmid and good-quality plasmid DNA can then be isolated from these cells to check against recombination events.

References ]. Nagel, R., Blliott, A., Masel, A., Birch, R. G., and Manners, J. M. (1990) Electroporation of binary Ti plasmid vector into Agrobacterium tumefaciens and Agrobacterium rhizogenes. FEMS Microbiol. Lett. 67, 325. 2. McCormac, A. C., Elliott, M. C., and Chen D. F. (1997) pBECKS: A flexible series of binary vectors for Agrobacterium-mediated plant transformation. Mol. Biotechnol. (in press). 3. Hood, E. E., Helmer, G. L., Fraley, R. T., and Chilton, M.-D. (1986) The hypervirulence of Agrobacterium tumefaciens A281 is encoded in a region of pTiBo542 outside of T-DNA. J. Bacteriology 168, 1291-1301. 4. Hoekema, A., Hirsch, P, R., Hooykaas, P. J. J., and Schilperoort, R. A. (1983) A binary vector strategy based on separation of vir- and T-region of the Agrobacterium tumefaciens Ti-plasmid. Nature 303, 179,180. 5. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory, Manual, 2nd ed. Cold Spring Harbor Laboratory press. Cold Spring Harbor, New York. 6. Chilton, M.-D., Currier, T. C., Farrand, S. K., Bendich, A. J., Gordon, M. P., and Nester, E. W. (1974) Agrobacterium tumefaciens DNA and PS8 bacteriophage DNA not detected in crown gall tumours. Proc. Natl. Acad. Sci. USA 71, 3672-3676.

Volume 9, 1998