Analytical Biochemistry 275, 15 (1999) Article ID abio.1999.4259, available online at http://www.idealibrary.

com on

A Method for Extraction of High-Quality and High-Quantity Genomic DNA Generally Applicable to Pathogenic Bacteria 1
Awdhesh Kalia,* ,2,3 Ashok Rattan, and Prem Chopra*
*Department of Pathology and Department of Microbiology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110 029, India

Received November 11, 1998

In this study, we report a modied procedure for extraction of high-quality genomic DNA that is rapid, simple, biologically nonhazardous, and generally applicable to pathogenic bacteria. Bacterial cells were pretreated with 70% ethanol prior to enzymatic digestion with lysozyme. Exposure of bacterial cells to 70% ethanol sterilized the cultures, making the process biologically safe and increased the susceptibility of the cells to lysozyme-induced lysis. Consistently high yields of genomic DNA (mean average yield, 0.52.5 mg/ml) were obtained from 465 isolates representing over 30 clinically important bacterial species. Genomic DNA obtained was determined to be suitable for further analysis, including bacterial ngerprinting techniques like restriction endonuclease analysis, Southern hybridization, and repetitive PCR. Availability of a generally applicable procedure for extraction of high-quality and high-quantity genomic DNA would be immensely benecial for laboratories engaged in molecular surveillance of nosocomial and community-based outbreaks. 1999 Academic Press Key Words: genomic DNA; pathogenic bacteria; 70% ethanol.

Extraction of genomic DNA in a reasonably intact and pure state forms the rst step in studies attempting to understand the molecular aspects of bacterial pathogenesis, physiology, and epidemiology. Chemical methods for extraction of genomic DNA rely primarily on the use of lysozyme in conjunction with other lipoPresented at the 8th International Congress on Infectious Diseases, Boston, MA, May 1518th 1998. 2 To whom correspondence should be addressed. 3 Present address: Department of Epidemiology and Public Health, 60 College Street, 925 LEPH, Yale University School of Medicine, New Haven, CT 06520. Fax: (203) 785-6130.
0003-2697/99 $30.00 Copyright 1999 by Academic Press All rights of reproduction in any form reserved.
1

lytic and proteolytic enzymes. However, many bacteria, such as Staphylococcus aureus, are resistant to lysozyme (1) and still others, such as Mycobacterium tuberculosis (MTB) 4 and Streptococcus pyogenes, show reduced susceptibility to lysozyme because of a complex cell wall. Consequently, a variety of DNA isolation procedures are now available depending on the nature of the bacterial cell wall. For example, lysostaphin is commonly and exclusively used for releasing DNA from Staphylococcus sp. (2). Likewise, hyaluronidase and mutanolysin have been found to be optimal for lysing streptococcal cells (3). Bacterial ngerprinting techniques like restriction endonuclease analysis (REA) and probe-based restriction fragment length polymorphisms (RFLPs) require a high quantity of intact and pure genomic DNA (ranging from 5 to 10 g per run). Use of puried DNA is also recommended for highly sensitive PCR-based ngerprinting methods such as repetitive PCR (rep-PCR) analysis and random polymorphic DNA analysis (RAPD) to minimize chances of artifactual polymorphism (4). Further, more than one or two methods are frequently used to determine the clonality of bacterial populations reliably. For laboratories engaged in surveillance of bacterial outbreaks, lack of a rapid, simple, and generally applicable procedure for extraction of genomic DNA may delay the outcome, and therefore the necessary corrective measure, of the investigation. Additionally, handling of concentrated suspensions of pathogenic bacteria is a potential health hazard. Thus, there exists a need for a procedure that is rapid, biologically safe, and applicable to a wide range of bacterial pathogens.
4 Abbreviations used: MTB, Myobacterium tuberculosis; REA, restriction endonuclease analysis; RFLP, restriction fragment length polymorphisms; rep-PCR, repetitive PCR; RAPD, random polymorphic DNA analysis.

KALIA, RATTAN, AND CHOPRA

MATERIALS AND METHODS

Bacterial strains. Bacterial isolates representing over 30 clinically important species were included in the study. Reference strains of S. aureus (ATCC

FIG. 1. (a) Agarose gel electrophoresis of the extracted genomic DNA. Lanes 13, MTB; lanes 4 and 5, S. pyogenes; lanes 6 8, S. aureus; lane 9, E. coli; lanes 10 12, K. pneumoniae; lane 13, H. pylori. One microliter of the DNA sample was loaded onto a 0.5% agarose gel. Genomic DNA was consistently obtained as an intact band. The presence of RNA did not affect further manipulations of DNA. (b) REA of S. pyogenes isolates. DNA (8 10 g) was digested with PvuII (Boerhinger Mannheim, GmBH, Germany), at 3 U/ g of DNA for 20 24 h. (c) In a similar manner, genomic DNA obtained from MTB was restricted with PvuII, transferred onto nylon membrane (Hybond, Amersham), and probed with 32P-labeled IS6110 DNA as described earlier (11, 13). DNA from clinical isolates in lanes 2 and 57 failed to light up with the IS6110 probe. Negative results were also obtained by PCR analysis for the presence of IS6110 in these isolates (data not shown), indicating the lack of the insertion sequence; lane 9, MTB H 37Rv and B indicate blank lane (water blank).

In this study, we describe a modied procedure for extraction of genomic DNA applicable to a wide range of pathogenic bacteria. We sterilized the bacterial cells at an early stage of DNA extraction by pretreatment with 70% ethanol, which is actively bactericidal (5, 6). A 70% ethanol treatment of the cells increased the susceptibility of cells to subsequent lysis as was inferred from recovery of a higher quantity of genomic DNA from cells treated with 70% ethanol in relation to cells not treated with 70% ethanol.

FIG. 2. Genomic DNA extracted was also amenable for PCR amplication. (a) PCR amplication of the 411-bp fragment of the rpoB gene from clinical isolates of MTB. Lane 2, negative control (water blank). PCRs were performed in a PTC-100 thermal cycler (MJ Research, U.S.A.) with 100 ng of puried DNA, 1.5 mM MgCl 2, 200 M dNTP mix, and 1 U of Taq DNA polymerase (Boehringer Mannheim, GmBH, Germany) in 1 PCR buffer with the following temperature prole: 94C for 5 min followed by 40 cycles of 94C for 30 s, 62C for 30 s, and 72C for 1 min, followed by extension at 72C for 10 min. The amplied products could be successfully used further for single-strand conformational polymorphism analysis (13). (b) repPCR analysis for MTB was performed with primer ERIC-2 (4) with the following temperature prole: 94C for 5 min followed by 4 cycles of 94C for 30 s, 40C for 90 s, and 72C for 8 min. These cycles were followed by 40 cycles at 94C for 30 s, 52C for 90 s, and 72C for 5 min. This was followed by a nal extension at 72C for 10 min. The products were resolved on 1.8% agarose gel at 10 V/cm. Lanes 1 4 depict ERIC ngerprints of epidemiologically related isolates of MTB. ERIC-PCR was repeated with fresh DNA extractions of the same isolates. Banding patterns (lanes 5 8) similar to the earlier patterns (i.e., lanes 1 4) were obtained, indicating reproducibility of ERIC-PCR and consistency of genomic DNA extraction. Lanes 9 and 10 are ngerprints from other epidemiologically related MTB isolates.

DNA EXTRACTION FROM PATHOGENIC BACTERIA TABLE 1

Reproducibility of the DNA Extraction Procedure

Organism S. pyogenes E. faecalis M. tuberculosis S. aureus N. gonorrheae P. aeruginosa A. baumanii H. pylori S. typhi Number of isolates 55 18 50 18 14 12 17 18 15 Mean DNA yield ( g/100 l) 77.2 82.3 55.0 56.0 108.0 102.4 140.0 125.3 152.0 4.0 4.7 3.0 2.9 6.4 4.8 6.0 3.4 3.4 Mean replicated yield ( g/100 l) 76.5 81.1 55.1 55.6 107.1 102.5 134.8 125.5 151.0 5.7 4.2 2.7 2.9 3.8 7.3 5.8 3.0 4.0

25923), Escherichia coli (ATCC 25922), and M. tuberculosis (H 37Rv and H 37Ra) were obtained from American Type Culture Collection and the Centers for Disease Control and Prevention, Atlanta. All other isolates used in this study were clinical isolates. Bacteria were grown on liquid or solid medium using standard microbiological procedures and recommendations (a complete listing of the bacterial species included in this study and the specic growth conditions employed can be obtained from the authors). DNA extraction procedure. The protocol described below is suitable for DNA extraction from bacterial growth from solid and liquid media. We preferred to use growths from solid media as this enhanced biosafety by reducing the chance of aerosol generation and laboratory spillover. Unless otherwise mentioned, all chemicals and reagents were procured from Sigma (St. Louis, MO). The following series of steps were performed for genomic DNA extraction: 1. Approximately 60 100 mg of bacterial cells (either scraped from growth on solid medium or harvested from liquid cultures) was transferred to Eppendorf tubes containing 1.5 ml TNE buffer (0.1 M Tris Cl, pH 8.0, 0.15 M NaCl, and 20 mM EDTA) and washed thoroughly by extensive vortexing. Glass beads (425 600 m) were included for clump-forming bacteria like MTB. 2. Cells were harvested by centrifugation at 15,000 rpm for 5 min. The pellet was resuspended in 2 ml ice-cold 70% ethanol, mixed thoroughly, and incubated on ice for 20 min. This treatment completely sterilized the cultures, including cultures of MTB (data not shown). 3. Cells were collected by centrifugation and resuspended in 480 l of TEST/LR buffer [0.1 M TrisCl, pH 8.0, 20 mM EDTA, 0.5 M sucrose, 1% (v/v) Triton X-100, 24 g of lysozyme, and, optionally, 0.8 g of RNase A; to 446 l of TEST buffer, 30 l of lysozyme (16 mg/ml) and 4 l of RNase A (10 mg/ml) are added separately] and incubated on ice for 1 h with occasional shaking. The tubes were then kept at 20C for 20 min.

4. The tubes were then immediately transferred to a water bath set at 68C and incubated for 10 min. Fiftythree microliters of 10% sodium lauryl sulfate was added and incubation was carried out for 15 min. After the addition of 87 l of 5 M NaCl, 69 l of CTAB/NaCl solution (1% N-cetyl-N,N,N-trimethylammonium bromide in 0.73 M NaCl) was added and the tube contents were mixed by inversion and rotation. The tubes were incubated further at 68C for 15 min and then kept at 20C for 30 min. 5. Subsequently, 1 vol chloroform:isoamyl alcohol (24:1, v/v) was added and the tubes were placed on a at rocking platform for 5 min. Phases were separated by centrifugation at 10000 rpm for 10 min. 6. DNA was puried rst with phenol:chloroform (1:1, v/v) extraction followed by another extraction with chloroform:isoamyl alcohol. 7. DNA was precipitated at 20C for 1 h after 2 vol ice-cold absolute ethanol was added. Ethanol was added slowly, drop by drop. Genomic DNA precipitated out forming clumps of intertwined threads. 8. DNA was recovered by centrifugation at 15,000 rpm for 15 min. The DNA pellet was washed in ice-cold 70% ethanol, dried in air, and suspended nally in 50 250 l of TE (10 mM TrisCl, pH 8.0, 1 mM EDTA) buffer. 9. Concentration and purity of the DNA isolated were determined spectrophotometrically (Hitachi, Tokyo, Japan) by taking the ratio of absorbance at 260 nm (A 260) and 280 nm (A 280).

FIG. 3. The method also demonstrated the potential to detect extrachromosomal DNA. Plasmid DNA could be observed among strains that were previously known to harbor the extrachromosomal elements. Plasmid DNA depicted in a and b was obtained from Shigella dysenteriae and N. gonorrheae, respectively.

4
TABLE 2

KALIA, RATTAN, AND CHOPRA

Pretreatment of Bacterial Cells with 70% Ethanol Resulted in Increased Recovery of DNA from Gram-Positive Bacteria
Mean DNA yield ( g/100 l) 77.2 82.3 55.0 56.0 108.0 102.5 140.0 125.3 4.0 4.7 3.0 2.9 6.4 4.8 6.0 3.4 DNA yield without 70% ethanol pretreatment ( g/100 l) 50.7 57.0 22.5 7.5 99.5 101.3 138.2 121.4 1.5 6.6 4.0 0.9 4.7 7.3 6.4 7.6

Organism S. pyogenes (55) E. faecalis (18) M. tuberculosis (50) S. aureus (18) N. gonorrheae (14) P. aeruginosa (12) A. baumanii (17) H. pylori (18)

RESULTS AND DISCUSSION

Genomic DNA was extracted from 465 isolates representing 32 bacterial species and was found to be intact by agarose gel electrophoresis (Fig. 1a; please note that Fig. 1a shows data from only a representative few of the 465 isolates used). The yield of DNA obtained varied from 50 to 250 g for different bacteria. However, replicate extractions performed on each isolate did not reveal any statistically signicant variations indicating excellent reproducibility of the method (Table 1). DNA obtained using the method consistently gave an A 260/A 280 ratio of 1.8 2.0 and was amenable to further molecular techniques, including REA (Fig. 1b), Southern hybridization (Fig. 1c), and PCR (Fig. 2). PCR-based techniques do not necessarily require highly pure DNA. However, pure DNA is recommended for PCR-based assays such as the rep-PCR and RAPD, which are susceptible to artifactual polymorphism, with the quality of DNA often affecting the banding pattern (4). The method also demonstrated its utility in detection of extrachromosomal DNA from bacterial cells that were known to harbor the plasmids (Fig. 3). The presence of RNA did not affect DNA integrity or purity nor interfere with subsequent manipulations (Figs. 1b, 1c, and 2). The TEST/LR, designed based on the observations of Irwin et al. (7), is unique in its capacity to effect membrane permabilization, solubilization, and removal of peptidoglycan and RNA in a single step. However, the use of RNase as described in step 3 is optional. Increased recovery of genomic DNA, in particular from gram-positive bacteria, was obtained when the cells were pretreated with 70% ethanol (Table 2). Similar variation was not observed in the DNA yields from gram-negative bacteria (Table 2). Pretreatment of bacterial cells with 70% ethanol did not result in any loss of RNA or DNA. However, intracellular proteins could be visualized in concentrated ethanolic supernates (Fig. 4), indicating that this treatment results in protein loss, culminating in bacterial death. Earlier studies of Bhaduri and Demchik demonstrated the release

of intracellular proteins and increased susceptibility of Staphylococcus sp. to lysozyme, subsequent to exposure with acetone (9). It is unlikely, primarily due to the actively bactericidal effects of 70% ethanol, that the proteins in the ethanolic supernates represent proteins secreted by active bacterial transport. It is probable that 70% ethanol exposure induces changes in the bacterial cell wall and membrane, thereby increasing cellular porosity that could accentuate subsequent bacterial lysis resulting in greater recovery of genomic DNA (Table 2). Bacterial lysis was further augmented by the inclusion of the freezethaw option, described in steps 3 and 5, that enhanced complete lysis of bacterial cells, particularly the gram-positive cells. In this study we attempted to rationalize the use of detergents. To achieve optimal detergent action, all detergents were used at their theoretical critical micellar concentration (cmc) and critical micellar temperature (cmt) (8). Both SDS [cmc 8.2 mM, Krafft temperature (T k) 16C] and CTAB/NaCl [cmc 0.92 mM, T k 22C] when used at the nal concentration of 1.0% (v/v) and 0.1% (v/v) achieve their cmc, theoretically, at 68C (8). In contrast, most methods recommend the use of CTAB/NaCl at a concentration of 1% (v/v). Under this condition, only a fraction of CTAB can

FIG. 4. Pretreatment of bacterial cells with 70% ethanol resulted in release of intracellular proteins while no loss of DNA was observed. Ethanolic supernates were concentrated in a speed-vac instrument (Savant Instruments, Farmingdale, U.S.A.). The concentrate was resolved in a 10% SDSPAGE and visualized by staining in Coomassie blue. Lane 1, molecular weight marker (Bangalore Genei Ltd., India); lane 2, proteins concentrated from ethanolic supernates of S. aureus; and lane 3, negative control (ethanol).

DNA EXTRACTION FROM PATHOGENIC BACTERIA TABLE 3

Comparison with Previously Available Methods a

Method Current study Nelson et al. Nelson and Selander Soolingen et al. Heath et al.
a b c

Time for extraction b (h) 4.0 2426 3 3.5 1824

Cost of extraction c 0.7 1.3 1.1 0.7 0.9

Mean average yield (mg/ml) 0.52.5 0.050.14 0.060.18 0.060.16 0.050.15

A 260/A 280 1.9 1.9 1.9 1.8 1.9

Reference (3) (10) (11) (12)

For comparison, DNA was extracted from 10 isolates simultaneously by the reported method and the method chosen for comparison. Up to 10 isolates could be processed simultaneously within the time specied. Cost was calculated on the basis of prices listed in Sigma catalogue (Sigma, St. Louis, MO).

be assumed to undergo micellization and thus may result in suboptimal activity, thereby necessitating longer incubation periods. Both SDS and CTAB/NaCl solutions were prewarmed to 68C prior to use. We evaluated the efcacy of our procedure, in terms of time required, yield and purity of DNA obtained, and the cost effectiveness, with some of the other methods published previously (Table 3). As indicated in Table 3, the reported procedure was more cost effective, particularly if time consumed and yield of DNA obtained are considered together. In summary, we have reported here a method for extraction of genomic DNA generally applicable to pathogenic bacteria. The method is amenable to scaling up or scaling down according to the amount of genomic DNA required for the study. The extracted DNA was pure and intact as demonstrated by REA, PCR, and Southern hybridization. Exposure of cells to 70% ethanol before digestion with lysozyme rendered the process biologically safe; induced lysis of S. aureus [generally resistant to lysozyme (1)]; induced the release of proteins; and improved recovery of genomic DNA, particularly from gram-positive bacteria. We presume that the availability of a generally applicable procedure for extraction of high-quality, high-quantity genomic DNA from pathogenic bacteria would be immensely benecial to laboratories engaged in molecular surveillance and control of nosocomial or community-based outbreaks.
ACKNOWLEDGMENTS
The authors are grateful to various colleagues for critical analysis of the reported method, especially Dr. A. G. Amin, Baylor College of Medicine, Texas, and Dr. Swathi Arur, AIIMS, New Delhi. Work presented was supported by grants from the Indian Council of Med-

ical Research (to P.C.) and Department of Biotechnology (to A.R.), Government of India. A.K. was supported by a National Research Fellowship from the Council of Scientic and Industrial Research, Government of India.

REFERENCES
1. Lachica, R. V. F., Hoeprich, P. D., and Genigeorgis, C. (1971) Appl. Microbiol. 21, 823 826. 2. Johnson, W. M., and Tyler, S. D. (1993) in Diagnostic Molecular Microbiology: Principles and Applications (Persing, D. H., Smith, T. F., Tenover, F. C., and White, T. J., Eds.), pp. 294 299, Am. Soc. Microbiol., Washington, DC. 3. Nelson, K., Schlievert, P. M., Selander, R. K., and Musser, J. M. (1991) J. Exp. Med. 174, 12711276. 4. Tyler, K. D., Wang, G., Tyler, S. D., and Johnson, W. M. (1997) J. Clin. Microbiol. 35, 339 346. 5. Koshiro, A., and Oie, S. (1984) Microbios 40, 33 40. 6. Price, P. B. (1950) Arch. Surg. 60, 492502. 7. Irwin, R. T., MacAllister, T. J., and Costerton, J. W. (1981) J. Bacteriol. 145, 13971403. 8. Neugebauer, J. M. (1990) in Methods in Enzymology (Deustcher, M. P., Ed.), Vol. 182, pp. 239 243, Academic Press, San Diego. 9. Bhaduri, S., and Demchik, P. H. (1983) Appl. Environ. Microbiol. 46, 941943. 10. Nelson, K., and Selander, R. K. (1994) in Methods in Enzymology (Clark, V. L., and Bavoil, P. M., Eds.), Vol. 235, pp. 174 183, Academic Press, San Diego. 11. Soolingen, D. V., DeHass, P. E. W., Hermans, P. W. M., and Van Embden, J. D. A. (1994) in Methods in Enzymology (Clark, V. L., and Bavoil, P. M., Eds.), Vol. 235, pp. 196 205, Academic Press, San Diego. 12. Heath, L. S., Sloan, G. L., and Heath, H. E. (1986) Appl. Environ. Microbiol. 51, 1138 1140. 13. Kalia, A., and Rattan, A. (1998) in Proceedings and Abstracts of the 98th General Meeting of the American Society for Microbiology, 17th May21st May, 1998, Atlanta, GA, pp. 180, Am. Soc. Microbiol.