Repetitive element PCR fingerprinting (rep-PCR) using

Letters in Applied Microbiology 1997, 25, 17–21

Repetitive element PCR fingerprinting (rep-PCR) using enterobacterial repetitive intergenic consensus (ERIC) primers is not necessarily directed at ERIC elements M. Gillings and M. Holley Key Centre for Biodiversity and Bioresources, School of Biological Sciences, Macquarie University, Sydney, NSW, Australia 1335/96: received 13 November 1996 and accepted 15 November 1996

We examined the use of enterobacterial repetitive intergenic consensus (ERIC) sequences in PCR on the DNAs of various bacteria, bacteriophage, invertebrates, fungi, plants and vertebrates and have shown that complex ERIC-PCR patterns can be readily produced from all of these target organisms. A range of annealing temperatures was tested, from 52°C (the commonly used annealing temperature) to 66°C (the approximate Tm of ERIC primers). At the higher temperatures, most bands failed to amplify, the exception being a subset of bands from enterobacterial targets. It was concluded that ERIC-PCR does not necessarily direct amplification from genuine ERIC sequences. M . G IL L IN GS A ND M. H OL LE Y . 1997.

INTRODUCTION

Families of short interspersed repetitive elements have been described in eubacteria, these being the repetitive extragenic palindromic (REP) elements, enterobacterial repetitive intergenic consensus (ERIC) sequences and the BOX element (Hulton et al. 1991; Martin et al. 1992). Sequence data for REP and ERIC elements have been described only from Gram-negative enteric bacteria and closely related phyla (Hulton et al. 1991; Versalovic et al. 1991), while the BOX element has been described only in the Gram-positive Streptococcus pneumoniae (Martin et al. 1992). Consensus primers to each of the elements have been used in polymerase chain reactions designed to amplify regions between neighbouring repetitive elements. This process, called rep-PCR, generates unambiguous DNA fingerprints that enable differentiation of eubacterial species and strains (Versalovic et al. 1991). Despite the fact the DNA sequence and hybridization data suggest that ERIC, BOX and REP elements have only restricted distribution amongst eubacterial species, rep-PCR is now being widely used for phylogenetic analysis and differentiation of bacterial strains. The method has been applied to members of diverse bacterial genera, including Rhizobium (de Bruijn 1992), Frankia (Murry et al. 1995), Staphylococcus (Del Vecchio et al. 1995), Legionella (Georghiou et al. 1994), Correspondence to: Dr Michael Gillings, Key Centre for Biodiversity and Bioresources, Macquarie University, Sydney, NSW 2109, Australia (e-mail: [email protected]). © 1997 The Society for Applied Bacteriology

Xanthomonas and Pseudomonas (Louws et al. 1994). Reports have also appeared using rep-PCR to identify strains of fungi in the genera Aspergillus (van Belkum et al. 1993) and Fusarium (Edel et al. 1995). These reports raise the question of how broad a range of organisms can be analysed with rep-PCR. In this paper we show that complex ERIC-PCR patterns can be readily generated from eukaryotes and bacteriophage, as well as eubacteria, when using the standard annealing temperatures. This observation, taken with the results of analyses performed at higher annealing temperatures, suggests that ERIC-PCR performed on non-enterobacterial targets may not necessarily be directed at genuine ERIC sequences, but rather, is a highly reproducible variant of the randomly amplified polymorphic DNA method (RAPD; Welsh and McLelland 1990; Williams et al. 1990).

METHODS Source of DNA

Bacterial, fungal and bacteriophage strains were obtained from the International Collection of Micro-organisms from Plants (Erwinia amylovora ICMP 1540), the Plant Pathology Herbarium, Rydalmere, NSW, Australia (Xanthomonas campestris pv. citri DAR 65982, Penicillium digitatum DAR 69705, Botrytis cinerea DAR 69763), Promega (Escherichia coli

18 M . G IL L IN GS A ND M. H OL LE Y

JM101, lambda bacteriophage) and Dr D. Veal, Macquarie University (Pseudomonas corrugata R117). Mites (Amblyseius sullivani) were a gift from J. Macdonald, NSW Agriculture. Centipedes (Cormacephalus westwoodi) and flatworms (Geoplana caerula) were collected on site (Macquarie University campus). Apple leaves (Malus x domestica cv. Red Delicious High Early) were obtained from Batlow Fruit Cooperative (Batlow, NSW, Australia). DNA was extracted from bacteria and invertebrates using the method of Gillings and Fahy (1993), from fungi with the method of Lee and Taylor (1990), and from plants with the method of Lohdi et al. (1994). DNA from aphids (Sitobion miscanthi) and rock wallabies (Petrogale spp.) were gifts from Dr P. Sunnucks and Dr M. Eldridge, respectively. DNA from herring sperm was from Boehringer Mannheim.

GAACCCGCGCCTGATCCAG and 5?ATCGGACTTGATGCGCAGGCCGTT) (Gillings et al. 1993), BCF and BCR (5?GCTACCTTCTCCGTCGTC and 5?GGTTGAGAACTCTGACGC) (Luck and Gillings 1995) and pUC/M13F and pUC/M13R (5?CGCCAGGGTTTTCCCAGTCACGAC and 5?TCACACAGGAAACAGCTATGA) (Messing 1983). Gel electrophoresis

An 8 ml aliquot of each amplification reaction was analysed using electrophoresis on 2% agarose gels cast and run in TBE buffer, pH 8·3 (Sambrook et al. 1987). Gels were stained with ethidium bromide and photographed using transmitted u.v. light and Polaroid film (Sambrook et al. 1987). A 100 base pair marker (Pharmacia LKB) was included on every gel.

Polymerase chain reaction

One ml of purified DNA (ca 50 ng) was mixed with 9 ml of GenereleaserTM (Bioventures Inc.) in a 0·5 ml tube and overlaid with two drops of sterile mineral oil. GenereleaserTM is a proprietary agent that sequesters inhibitors of PCR. Reaction tubes were heated on the high setting of a 650 W microwave oven for 7 min (4550 W min−1) in a microwave transparent rack (Bioventures Inc.). An Erlenmeyer flask containing 100 ml of water was included as a microwave sink. Tubes were incubated for 10 min in block 2 of a Stratagene Gradient 96 Robocycler (Stratagene Inc.) preheated to 80°C. PCR master mix (40 ml) was then added to each tube. The final concentrations of reagents in the PCR were as follows: 50 mmol l−1 KCl, 10 mmol l−1 Tris–HCl (pH 9·0), 0·1% Triton X-100, 4 mmol l−1 MgCl2, 0·2 mmol l−1 each of the deoxyribonucleotide triphosphates dATP, dCTP, dGTP and dTTP, 0·5 mmol l−1 of each primer, 20 mg ml−1 RNAse A, and 2 U of Taq polymerase (Promega). Dispensing of DNA, sterile water and master mixes was performed with aerosol pipette tips, and the master mixes were made up using dedicated pipettes. Negative controls, containing water only, and Genereleaser only, were included in every reaction set. The following thermal cycle was then performed: 94°C 3 min (1 cycle), 94°C 30 s, 52°C 90 s, 68°C 8 min (35 cycles), 68°C 8 min (1 cycle). The annealing step was lengthened from the standard ERIC-PCR protocol to compensate for the lack of ramping time in the block-transfer PCR cycler. Annealing gradients were set, where appropriate, in 2°C increments from 48°C to 68°C (for Escherichia coli) or 52°C to 66°C (for other samples). The Tm of ERIC primers (below) is ¼ 66°C, using a simple calculation based on length and GC content (Innis and Gelfand 1990). Primers tested in this paper included: ERIC1R and ERIC2 (5?ATGTAAGCTCCTGGGGATTCAC and 5?AAGTAAGTGACTGGGGTGAGCG) (Versalovic et al. 1991; Louws et al. 1994), pehA#3 and pehA#6 (5?CAGCA-

RESULTS

DNA from E. coli JM101 was subjected to ERIC-PCR using a gradient of annealing temperatures from 46°C to 68°C. At annealing temperatures below 56°C a typical ERIC-PCR profile of ¼20 bands was generated. As the annealing temperature was raised above 56°C, individual bands failed to amplify, until at 68°C only two major bands and five faint bands were still produced (Fig. 1). Using the standard annealing temperature for ERIC-PCR of 52°C (Versalovic et al. 1991), complex DNA fingerprints could be readily produced from a range of organisms includ-

Fig. 1 ERIC-PCR performed on Escherichia coli JM101 using a

gradient of annealing temperatures. Electrophoresis was on a 2% agarose gel stained with ethidium bromide. Primer annealing temperatures were varied in 2°C increments from 48°C to 68°C (right hand side of gel). Tracks (m) and (n) are 100 base pair ladders (AMRAD Pharmacia) and negative control, respectively

© 1997 The Society for Applied Bacteriology, Letters in Applied Microbiology 25, 17–21

R EP ET I TI VE E LE ME N T P CR F IN GE R PR IN T S 19

Fig. 3 PCR performed using pairs of long primers and ERICPCR thermal cycles. (a) M13 forward and reverse sequencing primers; (b) fungal B-tubulin primers; (c) polygalacturonase primers for Pseudomonas solanacearum (see Methods for details). Lanes: m, 100 base pair ladder; E, Escherichia coli; Er, Erwinia amylovora; X, Xanthomonas campestris pv. citri; P, Pseudomonas corrugata; n, negative control

Fig. 2 ERIC-PCR performed on a range of phyla using annealing

temperature gradients. Each set of reactions was performed using annealing temperatures of 52°C to 66°C in 2°C increments, with the lowest annealing temperature being on the left hand side of each panel. (a) Escherichia coli; (b) Erwinia amylovora; (c) lambda bacteriophage; (d) Pseudomonas corrugata; (e) Xanthomonas citri; (f) Penicillium digitatum; (g) Cormacephalus westwoodi (centipede); (h) Malus x domestica (apple). Tracks labelled (m) and (n) are molecular weight markers (100 bp ladder) and negative controls, respectively

ing eubacteria (Fig. 2a, b, d, e), a bacteriophage (Fig. 2c), fungi (Fig. 2f, evidence for Botrytis not presented), invertebrates (Fig. 2g, evidence for aphids, flatworms, mites not presented), a plant (apple, Fig. 2h) and vertebrates (herring sperm, rock wallaby not presented). The complexity of the banding patterns produced from these diverse phyla was similar, with 10–25 distinct bands generated from each DNA sample. These patterns were not generated from eubacterial contaminants in the samples, since the bacteriophage and fungi came from pure culture, and for other samples efforts were made, where possible, to avoid tissues (such as gut) that might contain significant bacterial contamination. While members of the enterobacteria did tend to generate band numbers at the high end of this range, so also did apple and Penicillium samples, and there was no obvious relationship between band number and phyletic position (Fig. 2).

As the primer annealing temperature was raised, the DNA fingerprints remained fairly constant up to 56°C, above which temperature bands progressively failed to amplify with each increment in annealing temperature. The only specimens generating any amplified products at the highest annealing temperature tested (66°C) were E. coli, Erw. amylovora (also a member of the Enterobacteriaceae) and X. citri. Even of these, the only specimen with strong, multiple bands was E. coli (Fig. 2). To test if primers other than those based on ERIC sequences could also generate complex fingerprints using ERIC cycling parameters, we used three other primer sets to amplify anonymous bands from bacterial DNAs. Primers based on M13 vector sequence, the Ps. solanacearum polygalacturonase gene or on a fungal B-tubulin gene, all produced complex DNA fingerprints from the various bacterial DNAs tested (Fig. 3). DISCUSSION

We have shown that complex DNA fingerprints can be generated from a wide range of phyla using standard ERICPCR protocols. The variety of DNAs tested here, and other unpublished results from this laboratory, suggest that ERICPCR patterns can be generated using DNA from any specimen. This raises the question as to whether the broad applicability of ERIC-PCR is due to the widespread occurrence of ERIC elements in living organisms. There are several lines of evidence that suggest this is not the case. First, no significant similarities were found between ERIC sequences and any phage or eukaryotic sequences in a search of GenBank and EMBL DNA databases (Hulton et al. 1991). Even so, both lambda bacteriophage and all the eukaryotic


20 M . G IL L IN GS A ND M. H OL LE Y

DNAs tested in this paper produced multiple DNA products when subjected to ERIC-PCR. However, when the primer annealing temperature was raised to near the Tm, these products failed to amplify. The conclusion must be that the ERIC primers are annealing to, and amplifying from, anonymous binding sites with accidental, and only partial homology to the primer sequences. Second, the majority of ERIC-positive species in oligonucleotide hybridization tests were Gram-negative enterics (Versalovic et al. 1991). The enterics tested in this paper exhibited few bands at annealing temperatures close to the Tm of the ERIC primers (66°C; Innis and Gelfand 1990). Given that there are less than 50 copies of the ERIC sequence in the E. coli genome (Hulton et al. 1991), and that the whole genome is 4·8 × 106 bp, there should be (on average) one ERIC element every 100 000 bp. It therefore seems unlikely that all 20 bands produced in ERIC-PCR on E. coli DNA could represent sequences lying between genuine ERIC elements, particularly as all 20 bands are less than 3000 bp. This suspicion is confirmed by the observation that the majority of PCR products fail to amplify at more stringent temperatures. The conclusion again must be, using standard ERIC-PCR annealing temperatures of 52°C, that the majority of amplified bands do not represent sequences lying between ERIC elements. Hence the production of amplified products from targets using the ERIC-PCR protocol should not be taken as proof of the existence of ERIC elements in the target organism, as is claimed in many papers using the method. This is not to say that ERIC-PCR is not useful as a means of distinguishing between organisms; indeed it seems that when it is applied to non-enterobacterial targets, ERIC-PCR is a highly reproducible and sensitive variant of the DNA amplification fingerprinting (DAF; Caetano-Anolles et al. 1991) or randomly amplified polymorphic DNA methods (RAPD; Welsh and McLelland 1990; Williams et al. 1990). These conclusions are given weight by our observation that complex DNA fingerprints can be generated from bacterial DNAs using pairs of long (18–24 base), arbitrary primers, and thermal cycles based on ERIC-PCR parameters. Primers originally designed to amplify fungal B-tubulin genes, the polygalacturonase gene from Ps. solanacearum or M13 sequencing primers all generated banding patterns of similar complexity to those generated using the ERIC-PCR method, yet clearly these target genes are not present in the bacterial species tested. Finally, the DNA sequences (unpublished) of some ERIC-PCR fragments generated from Sphingomonas paucimobilis by this laboratory show that the annealing sites for the ERIC primers occur within coding regions, whereas genuine ERIC sequences always occur between genes (Hulton et al. 1991). Furthermore, there is no significant homology to the ERIC consensus sequence downstream from the primer annealing sites.

ACKNOWLEDGEMENTS

Thanks are due to Vianney Brown for photography. This is publication number 226 from the Key Centre for Biodiversity and Bioresources, Macquarie University, NSW.

REFERENCES Caetano-Anolles, G., Bassam, B.J. and Gresshoff, P.M. (1991) DNA amplification fingerprinting: a strategy for genome analysis. Plant Molecular Biology Reporter 9, 294–307. de Bruijn, F.J. (1992) Use of repetitive (repetitive extragenic palindromic and enterobacterial repetitive intergenic consensus) sequences and the polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria. Applied and Environmental Microbiology 58, 2180–2187. Del Vecchio, V.G., Petroziello, J.M., Gress, M.J., McClesky, F.K., Melcher, G.P., Crouch, H.K. and Lupski, J.R. (1995) Molecular genotyping of methicillin-resistant Staphylococcus aureus via fluorophore enhanced repetitive sequence PCR. Journal of Clinical Microbiology 33, 2141–2144. Edel, V., Steinberg, C., Avelange, I. and Alabouvette, C. (1995) Comparison of three molecular methods for the characterization of Fusarium oxysporum strains. Phytopathology 85, 579–585. Georghiou, P.R., Doggett, A.M., Kielhofner, M.A., Stout, J.E., Watson, D.A., Lupski, J.R. and Hamill, R.J. (1994) Molecular fingerprinting of Legionella species with repetitive element PCR. Journal of Clinical Microbiology 32, 2989–2994. Gillings, M. and Fahy, P. (1993) Genetic diversity of Pseudomonas solanacearum biovars 2 and N2 assessed using restriction analysis of total genomic DNA. Plant Pathology 42, 744–753. Gillings, M.R., Fahy, P. and Davies, C.(1993) Restriction analysis of an amplified polygalacturonase gene differentiates strains of the phytopathogenic bacterium Pseudomonas solanacearum. Letters in Applied Microbiology 17, 44–48. Hulton, C.J.S., Higgins, C.F. and Sharp, P.M. (1991) ERIC sequences: a novel family of repetitive elements in the genomes of Escherichia coli, Salmonella typhimurium and other enterobacteria. Molecular Microbiology 5, 825–834. Innis, M.A. and Gelfand, D.H. (1990) Optimisation of PCRs. In PCR Protocols: A Guide to Methods and Applications ed. Innis, M.A., Gelfand, D.H., Sninsky, J.J. and White, T.J. pp. 282–287. London: Academic Press. Lee, S.B. and Taylor, J.W. (1990) Isolation of DNA from fungal spores and mycelium. In PCR Protocols: A Guide to Methods and Applications ed. Innis, M.A., Gelfand, D.H., Sninsky, J.J. and White, T.J. pp. 282–287. London: Academic Press. Lohdi, M.A., Ye, G., Weeden, N.F. and Reisch, B.I. (1994) A simple and efficient method for DNA extraction from grapevine cultivars and Vitis species. Plant Molecular Biology Reporter 12, 6–13. Louws, F.J., Fulbright, D.W., Stephens, C.T. and de Bruijn, F.J. (1994) Specific genomic fingerprints of phytopathogenic Xanthomonas and Pseudomonas pathovars and strains generated with repetitive sequences and PCR. Applied and Environmental Microbiology 60, 2286–2295. Luck, J.E. and Gillings, M.R. (1995) Rapid identification of benomyl


R EP ET I TI VE E LE ME N T P CR F IN GE R PR IN T S 21

resistant strains of Botrytis cinerea using the polymerase chain reaction. Mycological Research 99, 1438–1488. Martin, B., Humbert, O., Camara, M., Guenzi, E., Walker, J., Mitchell, T. et al. (1992) A highly repeated DNA element located in the chromosome of Streptococcus pneumoniae. Nucleic Acids Research 20, 3479–3483. Messing, J. (1983) New M13 vectors for cloning. Methods in Enzymology 101, 20–78. Murry, M.A., Zhang, D., Schneider, M. and de Bruijn, F.J. (1995) Use of repetitive sequences and the polymerase chain reaction (rep-PCR) to fingerprint the genomes of Frankia isolates. Symbiosis 19, 223–240. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1987) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, MA: Cold Spring Harbor Laboratory Press.

van Belkum, A., Quint, W.G.V., De Pauw, B.E., Melchers, W.J.G. and Meis, J.F. (1993) Typing of Aspergillus species and Aspergillus fumigatus isolates by interrepeat polymerase chain reaction. Journal of Clinical Microbiology 31, 2502–2505. Versalovic, J., Koeuth, T. and Lupski, J.R. (1991) Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Research 19, 6823–6831. Welsh, J. and McClelland, M. (1990) Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Research 18, 7213– 7218. Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A. and Tingey, S.V. (1990) DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18, 6531–6535.
