Journal of General Microbiology (1980), 116, 545-547. Printed in Great Britain 545 SHORT COMMUNICATION Crude Lysates of Staphylococcus aureus Can Transform Bacillus subtilis By JERALD S. FEITELSON?* ano JOSHUA LEDERBERG#t Department of Genetics, Stanford University School of Medicine, Stanford, California 94305, U.S.A. (Received 16 July 1979; revised 15 October 1979) Plasmids can be transferred from Staphylococcus aureus to Bacillus subtilis by crude lysates prepared with penicillin or lysostaphin. These lysates mediate drug-resistance plasmid transformation in competent B. subtilis at an efficiency paralleling that of purified DNA. INTRODUCTION Purified plasmid DNA from Staphylococcus aureus coding for drug resistance can trans- form, replicate and express its genetic information in Bacillus subtilis (Ehrlich, 1977). The regimen employed in the laboratory involved extensive purification of the plasmid DNA. We show here that heterospecific transformation of B. subtilis can occur using crude lysates of plasmid-harbouring cells, a regimen much more likely to be encountered under natural conditions. Initially, a variety of mixed cultures of the two organisms were tested for plasmid transfer but without success. It is known that competent cells are biosynthetically latent and rela- tively insensitive to killing by penicillin G (Nester, 1963). Therefore, a penicillin-induced lysate of plasmid-harbouring S. aureus, when added to competent B. subtilis, should have minimal effect in killing the recipient bacteria but at the same time could promote plasmid transfer. Another lytic agent with selective killing effects on the donor S. aureus is lyso- staphin. Crude lysates made with lysostaphin were also mixed with competent cells and found to mediate plasmid transfer. METHODS Strains. Staphylococcus aureus donor strains bearing Cm" plasmids were as follows (Novick, 1976): RN154(pC223), RN1305(pC221), RN2438(pUB112) and SA231(pC194). The plasmids were all originally transduced by Novick into the same plasmid-free host strain, designated RN450, resulting in an isogenic host background. The B. subtilis recipient strain SB863, derived from SB168, had the Sm’ (s¢r-2) gene as well as the aroBl, trpC2, tyrA, hisA, cys-f and leu-1 markers (Stanford collection). Bacteria were grown in Luria broth (Miller, 1972). Resistant organisms were selected on Difco nutrient agar supplemented with anti- biotics [chloramphenicol (Cm) at 15 wg ml-'; streptomycin (Sm) at 30 #g ml). DNA preparations. Plasmids were purified from S. aureus by the low-salt lysis of stationary phase cultures using clear lysis followed by CsCl-ethidium bromide density gradient centrifugation (Novick, 1976), and from B. subtilis by a rapid screening method using 0-7°% (w/v) agarose gel electrophoresis (Meyers et al., 1976). Lysostaphin lysates of S. aureus were made by growing cultures to mid-exponential phase (about 2 108 ml-4), centrifuging, resuspending the cell pellet in 1 ml 20 mu-Tris/HCI containing 2 mmM-EDTA, pH 8-0 (TE buffer) and adding lysostaphin (Schwarz-Mann) to a final concentration of 50 7g mi}; lysis occurred after incubation at 37 °C for 15 min. Penicillin G lysates were made by adding 0-6 yg penicillin + Present address: Rockefeller University, 1230 York Avenue, New York, New York 10021, U.S.A. 0022-1287/80/0000-8852 $02.00 © 1980 SGM 546 Short communication G ml" to the culture and allowing growth to continue for 3 h; the culture was then divided into two and penicillinase was added to one half to inactivate the antibiotic. Competence procedure and transformation. Bacteria were made competent by using C1 medium at 37°C for growth and C2 medium at 30 °C for induction of competence (Stewart, 1969); 0-8°% of the viable cells could be transformed for a single marker using prototrophic DNA. For transformation, 0-5 ml of one of the S. aureus lysates was added to 5 ml of competent B. subtilis (approximately 10° organisms) and incubated at 37°C for 30 min. Then 2 vol. L broth was added and the bacteria were grown for a further 2 h before concentration and plating. RESULTS AND DISCUSSION To test for in vivo plasmid transfer from S. aureus to B. subtilis, three mixed culture experiments were performed. The B. subtilis recipients were chromosomally Sm’, while the S. aureus donors had plasmid-coded Cm". Growing the two cultures separately and mixing them in broth, or growing the two cultures together in liquid or on non-selective plates, prior to plating on selective plates, all gave uniformly negative results. The only colonies growing on doubly selective media were S. aureus Sm™ mutants, arising at a frequency of 10-8 to 10-1°. No plasmid transfer was detected. To test whether penicillin-lysed S. aureus could transform B. subtilis, four Cm™ S. aureus strains were grown separately to mid-exponential phase (about 2x 108 ml~!) and then divided into two portions for penicillin or lysostaphin lysis. Three preparations (lysostaphin Table 1. Plasmid transfer from crude lysates of S. aureus to B. subtilis Competent (0:8°%) and non-competent (late-exponential phase culture not treated by the com- petence-induction protocol) B. subtilis SB863 were used as recipients for S. aureus crude lysates pre- pared by treatment with (i) lysostaphin (Lyso), (ii) penicillin followed after 3h by penicillinase (Pen’ase) or (iii) penicillin alone (Pen). Cm! colonies / Competence 10-® x viable S. aureus strain and lysate of SB863* colonies Frequencyt SA231(pC194) Lyso + 0/3 0 Lyso _ 0/3 0 Pen’ase + 0/14 0 Pen’ase - 0/3-9 0 Pen + 1/0-03 3x 10-8 Pen _ 0/5:2 0 RN154(pC223) — Lyso + 0/2-6 0 Lyso _ 0/5-6 0 Pen’ase + 0/2-1 0 Pen’ase _ 0/48 0 Pen + 0/0-02 0 Pen - 0/4:9 0 RN1305(pC221) Lyso + 18/3-6 5x 10-9 Lyso - 0/6°4 0 Pen’ase + 70/2-5 2:8 x 10-8 Pen’ase - 0/5-9 0 Pen + 4/0-15 2:7 x 10-8 Pen - 0/49 0 RN2438(pUB112) Lyso + 0/1-5 0 Lyso - 0/5-8 0 Pen’ase + 0/1-4 0 Pen’ase — 0/4-9f 0 Pen + 0/0-01 0 Pen _ 0/2-9 0 SB863 control + 0/1-9 0 * +, Competent; —, non-competent. t (No. of transformants ml~1)/(no. of viable cells mI-}). { There was one Sm'Cn’ S. aureus on this plate. 547 lysate, penicillin plus penicillinase lysate, and penicillin lysate) were used in transformation experiments with both competent and non-competent B. subtilis. It is evident (Table 1) that only cells subjected to the competence regime were transformed. Strain RN1305, harbouring plasmid pC221 (Novick, 1976), gave a positive result, with transformation of competent B. subtilis to Cm* occurring at a frequency of 5 x 10~® to 3 x 10-*. This frequency is comparable to that using purified S. aureus plasmid DNA, about 5 x 10~* colonies per genome equivalent (Ehrlich, 1977). Since the donor S. aureus strains were isogenic except for the presence of plasmids, the transfer frequency into B. subtilis is a property of the plasmid and not of the donor genetic background. The molecular weights of the plasmids do not strongly influence the transforma- tion efficiency, as pC221, pC223 and pUB112 are all 3 x 10° daltons, while pC 194 is 1-8 x 108 daltons (Ehrlich, 1977). Plasmids were extracted from cultures grown from two randomly selected colonies from each B. subtilis transformant class, except for the unique transformant generated from the SA231 lysate. All the plasmids were found to have the same molecular weight as the S. aureus donor plasmids by agarose gel electrophoresis, within the resolution of the technique (Meyers et al., 1976). These experiments demonstrate that plasmids present in the lysostaphin- or penicillin- induced lysates of S. aureus can enter B. subtilis competent cells and donate their chloram- phenicol resistance phenotype to the sensitive recipients. Since any B. subtilis culture normally contains a small proportion of competent bacteria (Young & Spizizen, 1961), it is likely that plasmids from S. awreus or other bacteria can transform B. subtilis or other bacteria in nature at a low frequency and have substantial evolutionary importance. Two recent reports on high-frequency genetic exchange between strains of B. subtilis inoculated into sterile soil indicate that homospecific genetic exchanges can occur naturally in soil (Graham & Istock, 1978, 1979). We have shown that purified DNA is not necessary for plasmid transformation between widely divergent bacterial species and thus offer no basis to doubt that heterospecific gene transfers can occur in nature. Short communication This work was supported by NIH grants 5 RO] CA16896-19 and 5 TO] GM00295-17. We wish to thank Drs A. T. Ganesan and S. D. Ehrlich for helpful comments. REFERENCES Eur icu, S. D. (1977). Replication and expression of plasmids from Staphylococcus aureus in Bacillus subtilis. Proceedings of the National Academy of Sciences of the United States of America 74, 1680-1682. GranHamM, J.B. & Istocx, C. A. (1978). Genetic exchange in Bacillus subtilis in soil. Molecular and General Genetics 166, 287-290. GraHaM, J.B. & Istocx, C. A. (1979). Gene ex- change and natural selection cause Bacillus subtilis to evolve in soil culture. Science 204, 637-639, Meyers, J.A., SANCHEZ, D., ELWeELL, L.P. & Fa.kow, S. (1976). Simple agarose gel electro- phoretic method for the identification and characterization of plasmid deoxyribonucleic acid. Journal of Bacteriology 127, 1529-1537. MILLer, J.H. (1972). Experiments in Molecular Genetics, p. 433. Cold Spring Harbor, N-Y.: Cold Spring Harbor Laboratory. NEsTeR, E. W. (1963). Penicillin resistance of competent cells in deoxyribonucleic acid trans- formation of Bacillus subtilis. Journal of Bacteriology 87, 867-875. Novick, R. (1976). Plasmid—protein relaxation complexes in Staphylococcus aureus. Journal of Bacteriology 127, 1177-1187. STEwart, C. (1969). Physical heterogeneity among Bacillus subtilis deoxyribonucleic acid molecules carrying particular genetic markers. Journal of Bacteriology 98, 1239-1247. Youna, F.E. & Spizizen, J. (1961). Physiological and genetic factors affecting transformation of Bacillus subtilis. Journal of Bacteriology 8), 823-829.