Letters in Applied Microbiology 1992, 14, 65-68

Electrotransformation of Xanthomonas campestris by RF DNA of filamentous phage 4Lf T Z E - W EW N A N G *& Y I - H S I U N GT S E N G * ?*Department ~ ofBotany and ?Institute of Molecular Biology, National Chung Hsing University, Taichung 400, Taiwan, Republic of China JRNI119: received 31 October 1991 and accepted 1 November I991 W A N G ,T Z E - W E N & T S E N GY, I - H S I U N G1992. Electrotransformation of Xanthornonas campestris by R F DNA of filamentous phage 6Lf. Letters in Applied Microbiology 14, 65-68. Conditions were optimized for electrotransfonnation of Xanthomonas campestris pv. campestris by the replicative form (RF) DNA of filamentous phase 6Lf. Early logarithmic cells were washed exhaustively with deionized water and subjected to a pulse at a field strength of 12.5 kV/cm with a 25 pF capacitor and a 400 R resistor. An efficiency of 5.1 x 10' pfu per pg R F DNA was obtained. Under the same conditions, the broad host range plasmid pLAFRl (21.6 kb) transformed X. campestris strains at eficiences around lo5 pfu per pg DNA prepared from XcP20H. The advantages of the protocol used in the present study are that the cells can be washed with water instead of complex buffer, and the DNA used can be prepared by the alkaline method of Birnboim & Doly without purification by ultracentrifugation.

Xanthomonas campestris pv. cumpestris is the phytopathogenic bacterium causing black rot in crucifers (Williams 1980) and the industrial microorganism for xanthan gum production (Kennedy & Bradshaw 1984; Sandford & Baird 1983). Presently, gene cloning in this bacterium is mainly carried out by a two-stage strategy, owing to lack of an efficient transformation method. A genomic bank of X . campesrris is first constructed with a mobilizable broad host range vector in Escherichia coli. Then the target genes are recovered by genetic complementation following conjugal transfer of the bank en m u s e into a mutant strain of X . campestris. This is time- and labour-consuming. Therefore, an eficient transformation protocol is needed for direct cloning of genes in X . campestris. Previously, efforts have been made in our laboratory to transform X . campestris by the filamentous phage 4Lf DNA. However, the eficiency obtained was not high enough for direct gene cloning (Tseng et a / . 1990). Recently electrotransformation of X . campestris pathovars $ Corresponding author.

has been reported (White & Gonzalez 1991; Yang et al. 1991). In these studies, variations in transformability between strains were noticed; and, since the best efficiency obtained was lo6 transformants per fig DNA, improvement in this regard seemed possible. Therefore, attempts were made to optimize the conditions for transfoming our working strains. This report describes a simple protocol to electrotransform X . campestris strains by unpurified RF DNA of filamentous phage 4Lf (Tseng et (11. 1990) and the broad host range cosmid pLAFRl (Friedman et a / . 1982). Materials and Methods BACTERIA, PHAGE, PLASMIDS A S D CULTIVATIONS

q5Lf is a filamentous phage (Tseng et a / . 1990) which specifically infects X . campestris. X . campestris strains Xc17, and XcllA have been described elsewhere (Yang and Tseng 1988). Xc17 is a virulent wild-type strain, resistant to 4Lf and producing large amount of xanthan polysaccharide IXPS). Xcl I A is a spontaneous

66

T.-W. Wang & Y . - H . Tseng

avirulent strain producing moderate amounts of XPS. P20H is a derivative of non-mucoid mutant P20 (Yang et al. 1988) that yields a high titre of #Lf when used as the propagation host. Broad host range mobilizable cosmid pLAFRl (21.6 kb), derived from RK2, was obtained from A.M. Friedman et al. (1982) and maintained in P20H and E. coli DH1. Plasmid pPAl (57-kb) (Ma0 1987), carrying protease gene@), was from a XcllA genomic bank constructed in pLAFR1. A final concentration of 15 pg of tetracycline per ml was used to select the transformants of these plasmids. TYG medium contained 10 g/l tryptone, 6 g/l yeast extract, 1 mM-MgSO, and 5 g/1 glucose. X. campestris was grown at 28°C and E. coli at 37°C; all liquid cultures were aerated with vigorous shaking. Double layer assay for infective centres was carried out as described previously (Tseng et al. 1990), using P20H as the indicator host. PREPARATION OF D N A

R F DNA of 4Lf was prepared by the alkaline method of Birnboim & Doly (1979), except that two additional phenol/chloroform extractions were performed to remove the contaminants. The DNA, then resuspended in deionized water (400 pl for 30 ml of original culture), was good for electroporation. The yield was 1.8-2.3 pg R F DNA from 1 ml of the infected P20H culture. DNA of plasmids pLAFRl and pPAl were also prepared by the alkaline method; however, the yields were about 0.5 pg from each ml of original plasmid-harbouring P20H cultures. P R E P A R A T I O N OF B A C T E R I A L C E L L S

A single colony of X. campestris was inoculated into 3.0 ml of TYG in a 15 ml vial and grown overnight. Then 2.0 ml of the overnight culture was transferred into a one litre flask containing 200 ml of fresh TYG and grown until an OD,,, of 0.3-0.4 was attained. The cells were harvested by centrifugation (8000 g) at 4°C and washed five times with 20 ml of cold sterile deionized water. Then the cells were resuspended in 200-300 p1 sterile deionized water to obtain a final concentration of approximately 1.1 x 10" cells/ml, kept on ice and used directly for electroporation. For storage at -7O"C, cells were dispensed in aliquots into microfuge tubes and glycerol was added to a

final concentration of 10%. The frozen cells, thawed on ice before use, were good for 1 month. ELECTROPORATION

A Gene-Pulser electroporation apparatus with pulse controller (Bio-Rad, CA, USA) was used throughout these studies. Forty pl of the washed cells (ca 4.5 x lo9 cells) were added to a prechilled electroporation cuvette (0.2 cm electrode gap), followed by the addition of 0.1 pg DNA in 1 pl. Then the mixture was exposed to a field strength of 2.5 kV with a 25 pF capacitor and a 400 Cl resistor. After the pulse, 1 ml of TYG medium was added to the mixture. For transformation by #Lf RF, the mixture was used to assay for infective centres on the double layer. When transformed by pLAFRl and pPA1, the mixture was incubated at 2 8 T , for 60 min, for expression of tetracycline resistance before plating out for colony counting. Results and Discussion The non-mucoid mutant strain P20H was used as the representative strain to conduct electroporation. As in initial trial, the method described for Pseudomonas putida (Fiedler & Wirth 1988) was adopted, with the rationale that Xanthomonads are phylogenetically close to Pseudomonads. Following this protocol, cells were harvested at 8 h, washed twice and resuspended in 7.0 mM sodium phosphate buffer (pH 7.4) containing 300 mM sucrose and 1 mM MgSO,. Then, 200 pl of the cell suspension and 0.2 pg of DNA were added to the cuvette (0.2 cm electrode gap) and pulsed at a field strength of 2.5 kV with a 25 p F capacitor and a 200 C2 resistor. The efficiency obtained was 3.6 x lo4 pfu pg DNA, which was 10-fold higher than that by chemical treatment (Tseng et al. 1990). However, efficiencies at this level are not sutlicient for direct cloning of genes into X. campestris, and further optimization was needed. With all the settings unchanged, further optimization of the conditions was first done by washing three times and resuspending the cells in deionized water instead of phosphate buffer. With this change an increase of one order of magnitude in efficiency (1.3 x lo5 pfu/pg DNA) was attained. Under these conditions, early logarithmic cells showed better transformability than the cells from later growth stages. When

Transjormation of X.campestris cells from 4, 8 and 16 h culture were transformed, efficiencies of 6.5 x lo6, 1.2 x lo5 and 2.0 x lo4 pfu/,ug DNA, respectively, were obtained. The volume of the reaction mixture and the amount of DNA added also affected the transformation efficiency. Reduction of the reaction volume to 40 pl resulted in an elevation of the efficiency by another order of magnitude. In addition increase of the resistance from 200 R to 400 R enhanced the efficiency, and DNA amounts over 0.2 fig caused arcing which reduced drastically the transformation efficiency. Therefore, in later experiments, 0.1 ,ug DNA per reaction was used. Under these optimized conditions, the pulse duration was near 4.2 ms with about 50% survival of cells, and the eficiencies of electrotransforming P20H by 4Lf RF DNA were between 1.7 x lo7 and 5.1 x lo7 pfu fig (Table 1). The efficiency of electrotransforming frozen cells (in 10% glycerol) was comparable to that of the freshly prepared cells. In other words, lo‘% glycerol had no effect on the transformation efficiency. Using Xc17 and Xcl 1A as recipients, transformation efficiencies similar to that of P20H were obtained, indicating that differences in the ability of XPS production had no effect on transformability. In addition, Xc17 was resistant to 4Lf infection, therefore, success in transfecting this strain by 4Lf R F indicates that Xcl? is defective in the initial steps of infection, such as adsorption and/or penetration. The broad host range cosmid pLAFRl (Friedman et a/. 1982) has been used widely for gene cloning in Gram-negative bacteria other than E . coli. In this study, pLAFRl DNA prepared from P20H was found to transform P20H, Xcl 1A and Xc17 with equal efficiency (ca lo5 transformants/pg DNA); however, the DNA prepared from E. coli transformed Xc17 at an efficiency about 30-fold lower than P20H (Table Table 1. Electrotransformation of Xanrhomonas campestris P20H by filamentous phage &Lf RF DNA* Trial no.

Time constant Ims)

Survival (‘Y”)

Transformation efficiency IafuiDE RF DNA)

~

1

2 3 4

4.6 4.2 3.9 4.1

* Electroporation conditions.

54 44 64 55

5.1 4.9

107 10’

1.9

107

3.2

107

performed following optimized

67

Table 2. Electrotransformation eficiency of Xanthomonas campestris strains by broad host range cosmid pLAFRl* Transformation efficiency Strain

DNA from E . coli

P20H XcllA Xc17

1 . 3 x 105 7.9 lo4 5.0 x to3

DNA from P20H 3.3 2.0 2.4

105

105 105

* Electroporation performed following optimized conditions. 2). These data are in good agreement with the previous results showing that pLAFRl conjugally transferred better to P20H than to Xc17 (Yang er a / . 1988). It is likely that a strong restriction system exists in Xcl7. The usefulness of the electrotransformation protocol was further demonstrated by transforming the 57 kb plasmid pPAl (Ma0 198?), the pLAFR1-derived plasmid carrying Xcl 1 A protease gene(s), into the isogenic P20H at an efficiency as high as 2.6 x lo4 transformants/pg DNA. Very recently, two protocols were reported for electrotransforming several X . campestris pathovars. White & Gonzalez (1991) were able to transform X. campestris by plasmid pUFR027 (9.3 kb) at an efficiency of 1.3 x lo6 transformants/,ug DNA, while Yang et nl. (1991) transfected X . campestris by the RF DNA of filamentous phage Xf at an efficiency of 9.6 x lo5 pfu/,ug DNA. Compared with these methods, the method developed in the present study has higher efficiency. Furthermore, our method has the advantages that the cells can be washed with deionized water instead of complex buffer systems, and the DNA extracted by miniprep can be used directly without purification by ultracentrifugation. All studies on X . carnpestris in this laboratory have been supported by The National Science Council, Republic of China. We thank Dr Ming-Te Yang for very helpful suggestions.

References BIRNBOIM, H.C. & DOLY,J. 1979 A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Research 7, 15131523. FIEDLER, S. & WIRTH,R. 1988 Transformation of bacteria with plamid DNA by electroporation. Analytical Biochemistry 170, 38-44. FRIEDMAN, A.M., LONG,S.R., BROWN,S.E., BUIKEMA, W. J. & AIJSUBEL,F.M. 1982 Construction of a

68

T.-W. Wang & Y.-H. Tseng

broad host range comsid cloning vector and its use in the genetic analysis of Rhizobium mutants. Gene 18,289-296. KENNEDY, J.F. & BRADSHAW, I.J. 1984 Production, properties and applications of xanthan. In Progress in Industrial Microbiology ed. Bushell, M.E. vol. 19. pp. 3 19-371. Amsterdam : Elsevier/North-Holland Publishing Co. MAO, J.R. 1987 Proteolytic activities and cloning of the genes from Xanthomonas campestris pv. campestris. MS Thesis, National Chung Hsing University, Taiwan. SANDFORD, P.A. & BAIRD,J. 1983 Industrial utilization of polysaccharides. In The polysaccharides ed. Aspinall, G.O. Vol. 2. pp. 411-490. New York: Academic Press. TSENG,Y. H., Lo, M.C., LIN, K.C., PAN, C.C. & CHANG,R.Y. 1990 Characterization of filamentous bacteriophage bLf from Xanthomonas campestris pv. campestris. Journal of General Virology 71, 1881-1 884.

WHITE,T.J. & GONZALEZ, C.F. 1991 Application of electroporation for efficient transformation of Xanthomonas campestris pv. oryzae. Phytopathology 81,521-524. WILLIAMS, P.H. 1980 Black rot: a continuing threat to world crucifers. Plant Disease 64,136-742. YANG, B.Y. & TSENG,Y.H. 1988 Production of exopolysaccharide and levels of protease and pectinase activity in pathogenic and non-pathogenic strains of Xanthomonas campestris pv. campestris. Botanical Bulletin of Academia Sinica 29,93-99. YANG,B.Y., TSAI, H.F. & TSENG,Y.H.1988 Broad host range cosmid pLAFRl and nonmucoid mutant XCPZO provide a suitable vector-host system for cloning genes in Xanthomonas campestris pv. campestris. Chinese Journal of Microbiology & Immunology 21,231-240. YANG, M.K., Su, W.C. & Kuo, T.T. 1991 Highly eficient transfection of Xanthomonas campestris by electroporation. Botanical Bulletin of Academia Sinica 32, 197-203.

c