Electroporation-mediated transfection of mammalian cells with crude plasmid DNA preparations

NENETIC ALYSIS ELSEVIER

Genetic Analysis: BiomolecularEngineering 12 (1995) 113-117

Blomolecular Engineering

Electroporation-mediated transfection of mammalian cells with crude plasmid DNA preparations Masaaki Tatsuka *a, Nobuyuki Yamagishi a, Morimasa Wada a, Hiromi Mitsui a, Takahide Ota b, Shizuo Odashima h alnstitute of Molecular and Cellular Biology, Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto 607, Japan bDepartment of Pathology, Kanazawa Medical University, Uchinada, lshikawa 920-02, Japan

Received 15 March 1995; revision received22 May 1995;accepted 12 June 1995

Abstract

We designed a simple and reproducible electroporation-mediated transfection procedure with which to screen mammalian expression vector-constructed c'DNA libraries. Using a specific chamber composed of five parallel electrodes, the recipient cells can be electroporated separately with 40 plasmid DNA preparations in a single experiment. Over 300 crude plasmids prepared from E. coil (DH-5) carrying a pcD2neo-vector-derived cDNA library were tested. The efficiency of stable transfection by electroporation with crude plasmid D N A preparations was 10-times higher than with the CsCl-purified plasmid DNA. When the crude plasmids were digested with RNase, the efficiency of stable transfection markedly decreased, indicating that the contaminating bacterial RNA in the crude plasmid preparations has a strong carrier effect during the electroporation. Even when salmon sperm D N A or genomic DNA from the recipient cells was used as the carrier of the purified plasmid, the efficiency was not higher than that using the crude preparations. This procedure is useful not only for screening a number of cDNAs but also for routinely introducing biologically active foreign genes into cultured mammalian ceils. Keywords: Electroporation; Transfection; cDNA libraries

1. Introduction

To isolate and identify a gene which is functionally active in mammalian cells, expression cloning systems using eDNA libraries constructed with mammalian expression vectors are very useful [1]. In all of these systems, however, the biological activity of each plasmid recovered from the recipient cells must be reconfirmed. This includes the recovery of many candidate plasmids either from episomal fractions [2, 3], from the PCR products [4] or by plasmid rescue [51, so identifying the plasmid clone with the biological activity is highly laborintensive. Sib-selection [6] may overcome this problem. When searching for genes with recessive phenotype such as tumor suppressor genes, it is also necessary to * Corresponding author, Tel.: +81 762 628151 ext. 5292; Fax: +81 762 607840; E-mail: [email protected]. 1050-3862/95/$09.50 © 1995 ElsevierScience B.V. All rights reserved SSDI 1050-3862(95)00117-5

transfect candidate genes into mammalian cells and to assess individual phenotypes. The preparation of plasmid DNAs and their transfection into mammalian cells are critical and require simplification to allow a number of eDNA clones to be rapidly screened. We developed a simple and reproducible procedure and tested it by transfecting over 300 crude plasmid preparations into cultured mammalian cells. 2. Materials and methods 2.1. Electroporation apparatus

The electroporation chamber consisted of five parallel electrodes (Model 10503; D.E.P. systems, Troy, MI). The distance between the 30-cm long electrodes was 3 mm. The chamber was connected to an electric power supply (Gene Pulser Unit; Bio-rad Laboratories, Hercules, CA).

i14

M. Tatsuka et al./Genetic Analysis." BiomolecularEngineering 12 (1995) 113-117

2. 2. Cells and culture To simplify the procedure for screening purposes, we transfect cultured mammalian cells with crude plasmid DNA preparations from E. coli (DH5). Since neither the genotoxic nor the epigenetic effects to mammalian cells of crude extracts from E. coli are unknown, we were used BALB/c 3T3 A31-1-1 cells [7] as the recipient. This cell line is widely used as the target for estimating the carcinogenic effects of industrial and environmental agents [81. The cells were cultured in Eagle's MEM (GIBCO, Grand Island, NY) containing 10% fetal calf serum. 2.3. cDNA library A cDNA library was constructed with the pcD2neo mammalian expression vector using mRNA purified from 100 dishes of density-arrested, quiescent cultured BALB/c 3T3 A31-1-1 cells [2]. The library contained full-length cDNAs and a complexity of over 5 × 106. 2.4. Plasmid preparations The electroporation of crude plasmid DNAs prepared either as 'minipreps' [9] or the 'one-step minipreps' [10] was tested. We also compared DNA that was extensively purified through two preparative CsCl/ethidium bromide equilibrium gradients [111. 2.5. Carrier DNAs The carrier effects of salmon sperm DNA (Stratagene, La Jolla, CA) and genomic DNA purified from BALB/c 3T3 A31-1-1 cells were determined. 2.6. Electroporation The procedure was as follows: 1. E. coli carrying plasmids was grown overnight in 3 ml of LB and pelleted in a microfuge tube. 2. Plasmid DNAs were then extracted from E. coli by means of the 'miniprep' [91 or the 'one-step miniprep' [10] procedure. The upper aqueous phase after phenol:chloroform extraction (1:1, saturated with H20) was collected and plasmids were precipitated with ethanol and stored at -20°C. 3. The crude plasmid DNA was rinsed with 70% ethanol, dried and dissolved in 25 #1 of Dulbecco's phosphate-buffered saline without Mg 2+ and Ca 2+ (PBS(-)). 4. Prior to the electroporation, the chamber was sterilized by UV light for 30 min. To prevent crosscontamination between samples in an electrode, the chamber was waxed with silicon vacuum grease (Beckman, Palo Alto, CA) before use. 5. To electroporate, cultured mammalian cells were detached from the culture plates by trypsinization and resuspended in PBS(-) at a concentration of 2-5 × 107/ml. Cell suspension (25 #1) was mixed with the plasmid solution by pipetting. Usually, 40 microfuge

tubes contained 25 #1 of different sample plasmid preparations to which 25 t~l of the cell suspension was added. After 5 rain at room temperature, the mixtures (total volume: 50 ~1) were transferred into the chamber. Eight samples were placed into each electrode, then electroporated. Since there were five electrodes, 40 sample plasmids were electroporated. Generally, electric pulseconditions for the transfection differ among cell types. Our conditions (0.25 #F, 6.7 kV/cm) were optimized for BALB/c 3T3 A31-1-1 cells as described [121. 6. After electroporation, each mixture was recovered from the electrodes into microfuge tubes and left at room temperature for 5 rain. 7. The electroporated cells were diluted with growth medium and an appropriate number of cells were replated in plastic dishes. After the cells attached to the dishes (usually within 1 h), the medium was changed. After culturing overnight, the medium was changed again and, if necessary, the selectable drug was added. 3. R e s u l t s and d i s c u s s i o n

3.1. Cytotoxicity of crude plasmid DNA preparations Crude plasmid DNA preparations made by the 'miniprep' or the 'one-step miniprep' procedure were cytotoxic to BALB/c 3T3 cells (Fig. 1). In contrast, the DNA extensively purified through two preparative CsCl/ethidium bromide equilibrium gradients had no cytotoxic effects within 16 h (Fig. 1). To avoid the cytotoxicity of crude preparations, it was necessary to remove the medium containing the plasmid D N A from

100 . l ~ q l ~ i;li..#a~

.vM

;

,

_t

8

12

16

10'

1-

0 0

4

20

Incubation Time (Hours) Fig. 1. Cytotoxicityof crude or purifiedplasmidDNA preparationsin BALB/c 3T3 A31-1-1 cells. Electroporated (open symbols)or nonelectroporated(closedsymbols)cellswerecluturedwithcrudeplasmid DNAs prepared by the 'miniprep' (circles), the 'one-step miniprep' (squares), or with DNA extensivelypurifiedthrough two preparative CsCFethidiumbromideequilibriumgradients(triangles)for 0, 1, 2, 3, 4 and 16 h.

M. Tatsuka et al. / Genetic Analysis." Biomolecular Engineering 12 (1995) 113-117 To Power Supply

-

-

115

30 cm - -

A_ 3mm T Electrode Electrode

Electrode Electrode Electrode Fig. 2. Scheme of the electroporation chamber. The five parallel electrodes are 3 mm apart. Eight samples can be placed in single electrode and simultaneously electroporated. A total of 40 samples can be sequentially analyzed in an experiment. The numbers (1)-(40) are where the samples were placed.

A

the mammalian cell culture. Therefore, the electroporated cells were cultured for 1 h after electroporation, then the medium was replaced.

Hind HI BamH I Sal I I SV40 ori

3.2. Positional effects on transfection efficiency in the Amp R

chamber

To perform the electroporation with many samples, we used a specific chmnber composed of five parallel electrodes as schematically shown in Fig. 2. In this chamber, eight samples per electrode were simultaneously electroporated. A total 40 samples can be treated in a single experiment. The chamber is commercially available (see Materials and methods), but there are no data regarding tke positional effects of each electrode. We examined the transfection efficiency at 40 locations as shown in Fig. 2. BALB/c 3T3 cells were electroporated with the pcD2neo vector and G418 resistance colonies were counted. The average efficiency of stable transfection per colony-forming (surviving) cell was 3.5 ± 0.3% (S.D.), and no significant differences were observed among 40 locations.

3.3. Comparison of transfection efficiencies between crude and purified plasmid DNA preparations To compare the transfection efficiencies and to evaluate side effects such as genotoxicity and carcinogenicity upon cultured mammalmn cells when using crude plasmid DNA preparations, we tested the focus formation of BALB/c 3T3 A31-1-1 cells by electroporation with pcD vector carrying the, viral oncogene, v-src [13]. The DNA contents of crude or purified plasmid DNA preparations were calibrated by agarose gel electrophoresis (Fig. 3). DNA (3/~g) was used in the focus formation assay. As shown in Fig. 4, the frequency of focus formation was 10 times higher in experiments using crude plasmid DNA preparations (lanes 3 and 4) than in those using the purified preparation (lane 2). However, when BALB/c 3T3 A31-1-1 cells were transfected with the peD vector prepared by the 'miniprep' method, there were no transformed foci (Fig. 4, lane 1). There were no significant differences in the

pcDsrc

l-src pBR322 ori

poly-A

BamH I

B

12345678

CC

~23.1 --9.4 ~6.6 --4.4 ~2.3 --2.0 1.6

m0.5 m0.3 Fig. 3. (A) A map of the plasmid peDsrc DNA constructed by inserting a 2.9-kilobase EcoR l fragment of v-src into the peD vector [13]. (B) The pcDsrc plasmid isolated by the 'one-step miniprep' (lanes 1 and 4) and 'miniprep' methods (lanes 2 and 5), or by two consecutive CsCl/ethidium bromide equilibrium gradient centrifugations followed by phenol extraction and ethanol precipitation (lanes 3 and 6). Plasmid DNA was electrophoresed in a 0.8% agarose gel before (lanes 1-3) and after digestion with BomH I plus Hind III and RNase (lanes 4-6). Bands in lane 7 and 8 are H/rid III-digested lambda DNA plus H/nf I-digested pBR322 DNA and Sty I-digested lambda DNA, respectively. A position of the closed circle DNA was indicated as CC.

M. Tatsuka et al./Genetic Analysis: Biomolecular Engineering 12 (1995) 113-117

116

1

2

3

4

Colony formation

Focus formation

Fig. 4. Colony (top) and focus formation (bottom) of BALB/c 3T3 A31-1-1 cells after electroporation with 3/~g of peDsre DNA isolated b y the 'miniprep' (lane 3) and 'one-step miniprep' methods (lane 4), or by two preparative CsCI/ethidium bromide equilibrium gradient centrifugations followed by phenol extraction and ethanol precipitation (lane 2). Three micrograms of the pcD empty vector DNA isolated by the 'miniprep' method was also analyzed (lane 1). After electroporation, 5 x 102 cells wcr¢ seeded into 100-ram plastic dishes (Coming) and cultured in Eagle's MEM containing 10% fetal calf serum without G418 for 7 days to determine cell viability as the colony formation (top), and 5 x 104 cells were cultured for 14 days to determine focus forming ability (bottom). Three dishes were used and three independent experiments were done. The results were confirmed as shown in the picatre.

viable cell numbers among the experiments using three plasmid DNA preparations (Fig. 4, top). These findings suggest that this protocol can detect the biological activity of foreign genes without influence from crude plasmid preparations. 3.4. Examination of the carrier effect Bacterial RNA contamination in crude plasmid DNA preparations is one explanation as to why crude plasmids were more efficient in stable transfection than with the purified plasmid. To investigate this, we exposed the preparations to RNase prior to electroporation. The results showed that the transfection efficiency was decreased when crude plasmid DNA was digested with RNase (Table 1). Even when salmon sperm DNA or

genomic DNA from BALB/c 3T3 A31-1-1 cells were added to the purified plasmid DNA preparation, the transfection efficiency was not any higher than that using the crude preparation (Table 1). Thus, plasmid DNAs prepared by both 'miniprep' methods contain a large amount of bacterial RNA (Fig. 3, lanes 1 and 2), which seems to have a strong carrier effect during electroporation, although the possibility that the plasmid preparations are in a different state (probably nicked) cannot be excluded. Other explanations for a carrier effect of crude plasmid preparations are possible. For example, the RNase could act as an inhibitor for electroporation, or the CsCI plasmid DNA preparation could have an inhibitory effect to electroporation. We have no data to determine these possibilities, but the addition of yeast tRNA to the CsCI plasmid DNA preparation increased the transfection efficiency of the CsCI DNA over ten times (data not shown). 3.5. Running Test We tested the transfection efficiency of many samples using our protocol with eight cycles of electroporation (40 samples per each cycle). With crude plasmids prepared from 320 batches of E. coil pools from the cDNA library described in Materials and methods, about 2-4% of the colony-forming (surviving) cells became (3418 resistant. In contrast, with a pcD2-1ibrary DNA extensively purified through two CsCl/ethidium bromide equilibrium gradients, only 0.2% of the colonyforming (surviving) cells became G418 resistant. Finally, we tested whether a plasmid clone could isolate from cDNA library in our procedure. A small amount DNA (0, 10 and 100 pg) of pSV2neo-ras plasmid carrying the activated oncogene, c-Ha-rasl [14] were mixed into 1 ~,g of the pcD2-1ibrary DNA and each mixture was transformed into E. coil, DH-5. The transformed E. coil colonies were mixed and divided into 100 batches. The 'mini-prep' DNA preparations from each batch of E.

Table I Carrier effects on electroporation using crude or purified plasmid DNA preparations Plasmid DNA

Carrier DNA

RNase treatment -

'Mini-prep' 'CsCl-purified' Salmon sperm DNA Genomic DNA from BALB/c 3T3

+ -

Transfection efficiency (% ± S.D.) 3.5±

0.83 0.16 0.43 0.58

± ± ± ±

1.1 0.23 0.08 0.06 0.10

Method: BALB/c 3T3 A3 I-I-I cells were electroporated with 3 ~,g of peD2neo plasmid prepared by the 'miniprep' method or by two preparative CsCI/ethidium bromide equilibrium gradient centrifugations followed by phenol extraction and ethanol precipitation. Carrier DNA (10 ~,g) was added into 'CsCl-purified' plasmid DNA preparation before electroporation. 'Mini-prep' plasmid DNA preparation was first digested with 10 t~g/ml RNase A for 16 h. Electroporated cells were cultured with or without 400 ~g/ml G418. Colonies were stained and counted. The transfection efficiency was calculated as the number of (3418 resistant colonies per number of colonies appeared in the culture without G418. Three batches of plasmids were tested and the standard deviation (S.D.) of the data is indicated.

M. Tatsuka et al./ Genetic Analysis: Biomolecular Engineering 12 (1995) 113-117

coil were electroporated into BALB/c 3T3 A31-1-1 cells. The electroporated cells (105 cells) were seeded into 100 mm dishes and selected iin the (3418-containing medium. After 14 days, (3418 resistant colonies were appeared in all of the dishes (2-4 x 103 colonies per dish). Cell morphology of the colonies was observed after Giemsa staining. Two and 15 colonies containing c-Ha-raslinduced morphologically transformed cells were found in dishes from 10 pg/1 t~g mixed plasmid DNAtransfected cell populatiion and 100 pg/1 #g mixed plasmid DNA-transfected cell population, respectively. No morphologically transformed cells were found in 0 pg/1 pg mixed plasmid DNA-transfected cell population.

4. Summary A specific chamber composed of five parallel electrodes was used to electroporate mammalian cells with crude plasmid DNA preparations. Forty samples can be individually and simultaneously transfected into the cells. The stable transfo~tion efficiency of over 300 samples of crude plasmid DNA preparations were higher than that using extensively purified DNA. The considerable bacterial RNA contamination in crude plasmid DNA preparations seems to have a strong carrier effect upon electroporation, because RNase reduced the transfection efficiency. This procedure is particularly suited for the stable transfection assay, perhaps for the transient transfection assay of a large number of samples and to screen of many candidates when isolating a functional gene by expression cloning.

117

Acknowledgements We thank Yoshikazu lhara and Takahiko Iba for technical assistance, and Hiroto Okayama for helpful discussion.

References [1] Okayama H, Kawaichi M, Brownstein M, Lee F, Yokota T, Arai K. Methods Enzymol 1987; 154: 3-28. [2] Tatsuka M, Mitsui H, Wada M, Nagata A, Nojima H, Okayama H. Nature 1992; 359: 333-336. [3] Legerski R, Peterson R. Nature 1992; 359: 70-73. [4] Hayakawa H, Koike G, Sekiguchi M. J Mol Biol 1990; 213: 739-747. [5] Noda M, Kitayama H, Matsuzaki T, Sugimoto Y, Okayama H, Bassin RH, Ikawa Y. Proc Natl Acad Sci USA 1989; 86: 162-166. [6] Noda A, Ning Y, Venable S, Pereira-Smith OM, Smith JR. Exp Cell Res 1994; 211: 90-98. [7] Kakunaga T, Crow JD. Science 1980; 209: 505-507. [8] Kakunaga T. In Kakunaga T, Yamasaki H (eds). Transformation Assay for Established Cell Lines: Mechanisms and Application. New York, Oxford University, 1985, pp. 55-73. [9] Birnboim HC, Doly J. Nucleic Acids Res 1979; 13: 1513-1523. [10] Chowdhury K. Nucleic Acids Res 1991; 19: 2792. [11] Chen C, Okayama H. Mol Cell Biol 1987; 7: 2745-2752. [12] Tatsuka M, Orita S, Yagi T, Kakunaga T. Exp Cell Res 1988; 178: 154-162. [13] Kizaka S, Hakura A. Mol Cell Biol 1989; 9: 5669-5675. [14] Yagi T, Sasayama S, Sasai H, Kakunaga T. Mol Carcinog 1989; I: 222-228.