We have developed a simple, one-step procedure for the preparation ofcompetentEscherichia coli that uses a transformation and storage solution [TSS; lx TSS is LB broth containing i0% (wt/vol) polyethylene glycol, 51% (vol/vol) (2,3). Since these initial studies, a number of factors have been elucidated that produced an increase in transformation efficiency. Such factors include prolonged incubation of bacteria with CaCl2 (4), addition of multiple cations into the transformation mixture (5) and treatment of bacteria with dimethyl sulfoxide (DMSO), hexaminecobalt, and dithiothreitol in the presence of both monovalent and divalent cations (6). The latter modification produced yields of >108 transformants per ,ug of DNA. Polyethylene glycol (PEG) has also been shown to mediate plasmid DNA uptake by protoplasts of a number of bacterial strains, but PEG-mediated transformation of E. coli by pBR322 DNA averaged 106 transformants per ,ug of DNA (7), a value 2 orders of magnitude lower than that obtainable by other methods (6).In this study, we describe an effective method using PEG for the preparation of competent bacterial cells. This procedure is convenient and rapid and routinely yields 10 -101 transformants per ,gg of plasmid DNA. In addition, bacteria prepared by this method can be frozen and stored for future use. Thus, this transformation system is advantageous because of its simplicity and dual use.MATERIALS AND METHODS Chemicals. DMSO, PEG, and PEG derivatives were purchased from Sigma. for 1 hr to allow expression of the antibiotic-resistance gene.In our experiments, transformants were selected by plating cells (in triplicate) on agar plates containing carbenicillin (30 mg/liter). Transformation efficiencies (expressed as the number of transformants per ,ug of DNA) were calculated after incubation of the plates at 37°C for 17-20 hr. To demonstrate that transformation of bacterial cells was mediated by the input pUC19 DNA, transformants were screened for the presence of plasmid DNA (by alkaline lysis of minipreps, restriction endonuclease digestion, and gel electrophoresis). The input pUC19 DNA was recovered in all cases. Changes were not noted in the restriction fragment sizes ofthe plasmid DNA, suggesting that TSS is not mutagenic for input DNA sequences.For long-term storage of competent cells, cells in TSS were frozen immediately in a dry ice/ethanol bath and stored at -70°C until needed. Frozen cells were thawed on ice and used immediately in the transformation assay.Statistical Analysis. Regression analysis was applied to the transformation efficiencies expressed as a function of several experimental conditions (pulse temperature, DNA concentration, pH, etc.). Regression functions were assumed to be polynomials of appropriate degree (linear, quadratic, cubic, quartic) and were fitted to the data by least squares (8). Maximum yields were then estimated from the regression functions. In addition, for some of the experimental conditions, one-way or two-way analyses of variance were applied to th...