Borrelia burgdorferi is an etiologic agent of Lyme disease (Barbour and Fish, 1993; Steere, 1994). The bacterium, a member of the spirochete phylum (Baranton and Old, 1995), has a genome predominantly composed of linear DNA molecules (Hinnebusch and Tilly, 1993; Saint Girons et al., 1994). Formulating a medium in which B. burgdorferi grows in vitro was the first step toward a genetic understanding of the physiology and pathogenesis of the organism (Burgdorfer et al., 1982; Barbour, 1984). The growth of B. burgdorferi as single colonies in solid medium (Kurtti et al., 1987; Bundoc and Barbour, 1989; Rosa and Hogan, 1992) has facilitated mutant isolation by selection (Sadziene et al., 1992; Samuels et al., 1994b), although a defined medium for selection of auxotrophs is not currently available. The transformation system described at this web site will be useful for manipulating the spirochete on a molecular genetic level.
Electroporation is the use of an electric pulse to reversibly permeabilize cell membranes (Shigekawa and Dower, 1988) and is an extremely efficient method of genetically transforming bacteria (Nickoloff, 1995). Electrotransformation has been used to disrupt a hemolysin gene (ter Huurne et al., 1992) and flagellar genes (Rosey et al., 1995) in the spirochete Serpulina hyodysenteriae, an etiologic agent of swine dysentery, by homologous recombination. The effect of electroporation buffers and capacitance on the survival of B. burgdorferi has been reported (Sambri and Lovett, 1990). Electroporation has been employed to insert point mutations conferring antibiotic resistance into the gyrB gene of B. burgdorferi by homologous recombination using both PCR products (Samuels et al., 1994a) and oligonucleotides (Samuels and Garon, 1997) as transformation substrates. Electron micrographs of electroporated B. burgdorferi reveal darkly staining regions on the surface of the spirochetes that may represent pores through which DNA can enter the cell (Samuels and Garon, 1997).
Unfortunately, there are both biosafety and physiological limitations on the use of many antibiotics with B. burgdorferi. Despite the use of both kanamycin and chloramphenicol resistance as a genetic marker in S. hyodysenteriae (ter Huurne et al., 1992; Rosey et al., 1995), the successful use of these antibiotics in B. burgdorferi has not been reported and attempts to use them in our laboratory to date have not been fruitful. The coumermycin A1-resistant gyrB gene (Samuels et al., 1994a; Samuels et al., 1994b), currently the only available selectable genetic marker conferring antibiotic resistance in B. burgdorferi, has been developed into an expression cassette that can now be applied toward dissecting the physiology and pathogenesis of B. burgdorferi on a molecular genetic level. This cassette has been successfully inserted into a natural B. burgdorferi circular plasmid (Rosa et al., 1996) and recently used to disrupt a gene encoding an outer surface protein (Tilly et al., 1997 1997 ), demonstrating that the technology is feasible. However, recombination into the chromosomal gyrB locus generates a large background and only about 0.4% of the coumermycin A1-resistant transformants contain the targeted insertion (Tilly et al., 1997 1997 ; Rosa et al., 1996). This requires extensive screening, which is facilitated by picking colonies with a sterile toothpick directly to a PCR tube and using a 96-well thermal cycler (Rosa et al., 1996).
This web site, derived from a previously published book chapter (Samuels, 1995), provides detailed methods for introducing DNA into B. burgdorferi by electroporation and for the selection of transformants (or spontaneous mutants) on solid medium. We typically obtain transformation efficiencies of 103 to 104 transformants/ug of linear DNA. The protocol may work for other species of the genus Borrelia.
Please do not hesitate to contact me at firstname.lastname@example.org with questions, comments or reprint requests.
Although we routinely use high-passage B. burgdorferi strain B31 for genetic experiments, we have had success transforming B. burgdorferi strains HB19 and N40, although we have not confirmed the N40 transformants by sequence analysis. We have transformed low-passage infectious B31, but have not yet assayed the infectivity of the transformants.
BSK (Barbour-Stoenner-Kelly) medium can be purchased from Sigma, but it is expensive. BSK contains 8% (v/v) 10x CMRL-1066 (without l-glutamine and sodium bicarbonate; Life Technologies, Gaithersburg, MD), 4 g/L Neopeptone (Difco, Detroit, MI), 40 g/L bovine serum albumin (BSA; fraction V, Pentex; Miles, Kankakee, IL), 1.6 g/L Yeastolate (TC; Difco), 4.8 g/L N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES), 4 g/L glucose, 0.56 g/L sodium citrate, 0.64 g/L sodium pyruvate, 0.32 g/L N-acetyl-d-glucosamine, 1.76 g/L sodium bicarbonate, and 6.6% rabbit serum (trace hemolyzed; Pel-Freez, Rogers, AR). Adjust to pH 7.6 with 1 N NaOH, stir slowly for 2-3 h, and sterilize by filtration (successively through a prefilter, a 1.2 æm filter, a 0.45 æm filter, and a 0.22 æm filter). BSK-H from Sigma comes frozen, but then is stored at 4øC for up to 2 months (see Note 1).
dPBS (Dulbecco's phosphate buffered saline) contains 8 g/L NaCl, 0.2 g/L KCl, 1.15 g/L Na2HPO4, and 0.2 g/L KH2PO4. Sterilize by filtration and store at 4øC.
dPBS++ (dPBS with divalent cations) is made by adding 0.1 g/L CaCl2 and 0.2 g/L MgCl2.6H2O to dPBS.
EPS (electroporation solution) is 93 g/L sucrose and 15% (v/v) glycerol. Sterilize by filtration and store at 4øC.
P-BSK (plating-BSK, also known as 1.5x BSK) medium: add 83 g BSA, 8.3 g Neopeptone, 10 g HEPES, 1.2 g sodium citrate, 8.3 g glucose, 1.3 g sodium pyruvate, 0.7 g N-acetyl-d-glucosamine, 3.7 g sodium bicarbonate, and 4.2 g Yeastolate to 1 L of water (18 Mêcm). Adjust to pH 7.5 with 1 N NaOH, stir slowly for 2 to 3 h, and sterilize by filtration (successively through a prefilter and a 0.22 æm filter). Store at -20øC.
Coumermycin A1 (Sigma, St. Louis, MO) at 50 mg/mL in dimethyl sulfoxide (DMSO) (see Note 2). Dilute to 10 mg/mL or other convenient concentration in DMSO for a working solution. Store at -20øC. We currently use a BTX TransPorator Plus for electroporations, but we have successfully used a BioRad Gene Pulser with Pulse Controller.
Preparation of Competent Cells
1.Inoculate 500 mL of BSK medium in a 500 mL screw-top bottle with 1 mL of a late-log phase culture (see Note 3). Incubate at about 32øC (without agitation) until the culture reaches a density of about 5 x 107 cells/mL (see Note 4). This requires 36-60 h.
2.Transfer culture to two sterile 250 mL screw-top centrifuge bottles and cap.
3.Centrifuge at 4000 x g for 20 min at 4øC. Decant the supernatant fraction and resuspend each cell pellet in 30 mL of cold dPBS (see Note 5).
4.Transfer cells to two sterile 50 mL screw-top centrifuge tubes and cap.
5.Centrifuge at 3000 x g for 10 min at 4øC. Decant the supernatant fraction and resuspend each cell pellet in 30 mL of cold dPBS.
6.Centrifuge at 3000 x g for 10 min at 4øC. Decant the supernatant fraction and resuspend each cell pellet in 10 mL of cold EPS. (EPS wash 1.)
7.Transfer cells to two sterile 14 mL polypropylene tubes and cap.
8.Centrifuge at 2000 x g for 10 min at 4øC. Decant the supernatant fraction and resuspend each cell pellet in 10 mL of cold EPS. (EPS wash 2.)
9.Centrifuge at 2000 x g for 10 min at 4øC. Decant the supernatant fraction and resuspend each cell pellet in 10 mL of cold EPS. (EPS wash 3.)
10.Centrifuge at 2000 x g for 10 min at 4øC. Decant the supernatant fraction and pool the cell pellets in 0.6 mL of cold EPS (see Note 6).
11.Distribute 50 æl aliquot fractions of the cell suspension into sterile
1.7 mL tubes on ice (see Notes 7 and 8).
12.Unused competent cells can be transferred directly into a -70 to -80øC freezer and then thawed on ice 10 to 30 min prior to electroporation (see Note 8).
1.Cool electroporation cuvettes (0.2 cm electrode gap) to 4øC.
2.Transfer 1 to 5 uL of a solution containing 0.3 to 1 ug of DNA in water (see Notes 9 and 10) to the cell suspension, mix gently, and incubate on ice for about 1 min.
3.Transfer the cell/DNA mixture to a chilled electroporation cuvette. Cap the cuvette and shake the cell/DNA mixture to the bottom of the cuvette with a flick of the wrist so that the mixture spans the two electrodes.
4.Place the cuvette in the pulse generator and deliver a single exponential decay pulse of 2.5 kV, 25 æF, and 200 ê. This should produce a time constant of 4 to 5 msec (see Notes 10 and 11).
5.Immediately (within 1 min) add 1 mL of BSK medium (at room temperature) without antibiotics and mix the cell suspension by pipeting up and down.
6.Transfer the entire mixture to a sterile 14 mL tube that contains an additional 9 mL of BSK medium (at room temperature) and incubate (without agitation) at about 32øC for about 18 h.
Selection of Transformants
You will need 35 mL of molten medium for each plate and at least two plates for each transformation. This recipe gives a final volume of 510 mL. 1.Mix 240 mL of P-BSK medium, 38 mL of 10x CMRL-1066, and 12 mL of rabbit serum. Equilibrate the mixture at 55øC in a water bath. Autoclave 200 mL of 1.7% agarose (high strength analytical grade), equilibrate to 55øC, and combine with the medium mixture. Add 20 mL of fresh 5% sodium bicarbonate and the selective agent (see Notes 2 and 12).
2.Transfer 15 mL of the molten medium into 12 to 14 100 mm dishes and allow to solidify at room temperature. Equilibrate the remainder of the molten medium at 42øC.
3.Transfer 0.1 mL of BSK II medium containing the electroporated cells to a 50 mL tube. Add 20 mL of the molten medium (at 42øC) and mix by pipeting up and down once. Transfer the mixture to the plates containing the solidified bottom agarose medium and allow to solidify at room temperature.
4.Centrifuge the remaining 9.9 mL of culture at 8000 x g for 5 min, resuspend in 1 mL of supernatant fraction, and plate as above.
5.Incubate the plates at 32 to 34øC in a humidified 1 to 5% CO2 atmosphere. Colonies should appear in about 14 days.
6.Isolate single colonies by picking with a plugged 15 cm pasteur pipet (with bulb). Transfer to 10 mL of BSK II in the presence of antibiotics. Cultures should reach late-log phase in 6-9 days.
Screening of Transformants
1.Number colonies using a felt-tip marker on the bottom of the plate. Set up PCR using primers that flank the directed insertion site (see Note 13).
2.Pick colonies with a sterile toothpick and transfer directly into tubes containing a PCR mix (Taq DNA polymerase, buffer, dNTPs, and primers) (see Note 14). Plates are then returned to the incubator for 1 week to allow for expansion of picked colonies.
3.PCR conditions are typically 30 cycles of 94øC for 0.5 min, 45 to 50øC for 0.5 min, and 68øC for 2.5 min.
4.An equal volume (20 uL) of 2x gel-loading buffer is added to each reaction and 20 uL is analyzed by (1%) agarose gel electrophoresis (using a 30-well comb) and ethidium bromide staining. Colonies exhibiting the correct phenotype are picked as described above after 1 week of expansion.
1.The quality of BSA varies by source and lot. We have found Miles to be a reliable source. However, we reserve 5 or 10 kg batches and test samples for the ability to support the growth of B. burgdorferi. We currently purchase BSK-H containing rabbit serum from Sigma. We have stored BSK-H medium (without serum) at 4øC for up to 2 years and found that it can support the growth of highly passaged strain B31 upon the addition of fresh serum.
2.The only antibiotic that is not clinically useful and has been shown to be effective for selection of resistant mutants is coumermycin A1 (Samuels and Garon, 1993; Samuels et al., 1994b) However, bactericidal antibodies have been used to select for mutants of B. burgdorferi (Cinco, 1992; Coleman et al., 1992; Sadziene et al., 1992). Novobiocin, another coumarin antibiotic, which is 20-fold less expensive than coumermycin A1 per use, can be used at 5 ug/mL to select coumermycin A1-resistant transformants. Novobiocin is labile upon exposure to light or freeze-thaw cycles; we prepare it fresh.
3.B. burgdorferi is a class 2 human pathogen and therefore should be handled in a class II biological safety cabinet (laminar flow hood). In addition, BSK medium is rich and all procedures should be performed aseptically. Introduction of recombinant DNA into a class 2 pathogen requires permission from the Institutional Biosafety Committee before initiation of the experiments according to Section IIIB of the Guidelines for Research Involving Recombinant DNA Molecules (Federal Register).
4.The cell density (or growth phase) is a significant factor for successful electrotransformation, as is the case with other bacteria (Shigekawa and Dower, 1988; Dower et al., 1992; Nickoloff, 1995). The cells will either not transform efficiently or not plate efficiently if the cell density is too high (when the color of the medium changes). We have had success electrotransforming cultures harvested at 1 to 7 x 107 cells/mL, although a low cell density (1 to 2 x 107 cells/mL) requires pelleting the cells at a higher g force (up to 5000 x g) and adjusting the final volume of the cell suspension (see Note 6). Alternatively, higher cell densities can be used for preparing competent cells but then 5% or less of the transformation should be plated (Rosa et al., 1996). Cell density can be determined using a Petroff Hausser Counting Chamber (Hausser Scientific Partnership, Horsham, PA). Dilute 0.1 mL of the culture with 0.9 mL of cold dPBS++ and place in the counting chamber. Count cells over all 25 groups of 16 small squares in all planes using a dark-field microscope. Multiply the number of cells counted by 5 x 105 to calculate cells/mL. Alternatively, cell density can be determined by spectrophotometry. Centrifuge 10 mL of the culture at 5000 x g for 10 min. Decant the supernatant fraction and resuspend the cell pellet in 1 mL of dPBS++. Centrifuge at 8000 x g for 5 min. Decant the supernatant fraction, resuspend the cell pellet in 1 mL of dPBS++, and measure the A600. Multiply the A600 by 1.4 x 108 to calculate cells/mL in the culture.
5.Thorough washing is important to remove components of the medium (see Note 10). Cell pellets are resuspended in both dPBS and EPS by pipeting followed by vortex mixing. These treatments do not appear to affect cell viability. The two dPBS washes during the preparation of competent cells are not required for successful transformation, although the cells are harder to pellet in EPS without these washes. We have not quantitatively assessed the effect of omitting the dPBS washes, and we still routinely perform them.
6.The final cell concentration should be 1 to 5 x 1010 cells/mL (with a final volume of about 0.9 mL). The volume of EPS used to resuspend the final cell pellet may have to be adjusted to account for initial cell number and efficiency of decanting.
7.We find that use of presterilized aerosol-resistant pipet tips (with aerosol barriers) helps to maintain sterility when handling small volumes of liquid.
8.We have not examined the effect of temperature on transformation efficiency, but maintaining the competent cells at 4øC is generally considered to yield optimal efficiencies (Shigekawa and Dower, 1988; Dower et al., 1992; Nickoloff, 1995). Frozen competent cells have a transformation efficiency only about 15% lower than that of fresh competent cells, although we still prefer to use fresh cells whenever possible.
9.We routinely obtain 1000 or more transformants/ug of DNA with strain B31, although we have only used linear DNA generated by PCR as an electrotransformation substrate. Linear molecules are 1000-fold less efficient in electrotransformation of Escherichia coli than circular molecules (Shigekawa and Dower, 1988) and we are currently constructing circular replicons from broad-host-range vectors and natural spirochete plasmids for use in B. burgdorferi.
10.Electroporation in the presence of high ionic strength solutions causes arcing (and a lowered time constant). Two arcs will kill all of the B. burgdorferi cells. We use various commercial DNA purification kits or we ethanol precipitate the DNA and either elute or resuspend the DNA at a high concentration in water. Transformation efficiency generally increases with DNA concentration (Shigekawa and Dower, 1988; Dower et al., 1992; Nickoloff, 1995).
11.Preliminary studies suggest that one pulse effected higher transformation efficiencies than multiple pulses and that varying the resistance from 100 to 400 ê affected the time constant, but did not significantly alter the transformation efficiency. The BTX TransPorator Plus has only a variable voltage setting.
12.An antibiotic concentration that inhibits bacterial growth in liquid culture by 80 to 90% relative to growth in the absence of antibiotics has been used to select for spontaneous mutants and transformants in solid medium (Sambri and Lovett, 1990; Samuels and Garon, 1993; Samuels et al., 1994a; Samuels et al., 1994b; Samuels and Garon, 1997). We currently use 0.2 ug/mL coumermycin A1 for selection of single mutant gyrB genes and 0.5 ug/mL for selection of the triple mutant NGR gyrB gene.
13.PCR primers should be designed so that a negative result (recombination into the chromosomal gyrB locus) yields an amplification product of about 100 to 500 bp and a positive result (directed insertion and gene disruption) yields a product of 2.2 to 2.6 kb (which includes the 2.1 kb coumermycin A1-resistant cassette). Use of a 96-well thermal cycler and 20 uL reactions allows for large-scale screening and saves on reagents.
14.The transfer only requires sticking the numbered colonies with the toothpick and sticking the toothpick briefly into the PCR mix. Use a fresh toothpick for each colony.
I sincerely thank Claude Garon for support during my postdoctoral fellowship when this transformation protocol was first developed, Kit Tilly for genetic counseling, Patti Rosa for the plating protocol and the PCR screening protocol, and Iain Old for making this web site a reality.
Baranton, G., and Old, I.G. (1995) The Spirochaetes: a different way of life. Bull Inst Pasteur 93: 63-95.
Barbour, A.G. (1984) Isolation and cultivation of Lyme disease spirochetes. Yale J Biol Med 57: 521-525.
Barbour, A.G., and Fish, D. (1993) The biological and social phenomenon of Lyme disease. Science 260: 1610-1616.
Bundoc, V.G., and Barbour, A.G. (1989) Clonal polymorphisms of outer membrane protein OspB of Borrelia burgdorferi. Infect Immun 57: 2733-2741.
Burgdorfer, W., Barbour, A.G., Hayes, S.F., Benach, J.L., Grunwaldt, E., and Davis, J.P. (1982) Lyme disease--a tick-borne spirochetosis? Science 216: 1317-1319.
Cinco, M. (1992) Selection of a Borrelia burgdorferi antigenic variant by cultivation in the presence of increasing amounts of homologous immune serum. FEMS Microbiol Lett 92: 15-18.
Coleman, J.L., Rogers, R.C., and Benach, J.L. (1992) Selection of an escape variant of Borrelia burgdorferi by use of bactericidal monoclonal antibodies to OspB. Infect Immun 60: 3098-3104.
Dower, W.J., Chassy, B.M., Trevors, J.T., and Blaschek, H.P. (1992). Protocols for the transformation of bacteria by electroporation. Guide to electroporation and electrofusion. San Diego, Academic Press. 485-499.
Hinnebusch, J., and Tilly, K. (1993) Linear plasmids and chromosomes in bacteria. Mol Microbiol 10: 917-922.
Kurtti, T.J., Munderloh, U.G., Johnson, R.C., and Ahlstrand, G.G. (1987) Colony formation and morphology in Borrelia burgdorferi. J Clin Microbiol 25: 2054-2058.
Nickoloff, J.A., Ed. (1995). Electroporation Protocols for Microorganisms. Methods in Molecular Biology. Totowa, New Jersey, Humana Press.
Rosa, P., Samuels, D.S., Hogan, D., Stevenson, B., Casjens, S., and Tilly, K. (1996) Directed insertion of a selectable marker into acircular plasmid of Borrelia burgdorferi. J Bacteriol 178: 5946-5953.
Rosa, P.A., and Hogan, D.M. (1992). Colony formation by Borrelia burgdorferi in solid medium: clonal analysis of osp locus variants. First International Conference on Tick-Borne Pathogens at the Host-Vector Interface: An Agenda for Research. St. Paul, University of Minnesota. 95-103.
Rosey, E.L., Kennedy, M.J., Petrella, D.K., Ulrich, R.G., and Yancey, R.J., Jr (1995) Inactivation of Serpulina hyodysenteriae flaA1 and flaB1 periplasmic flagellar genes by electroporation-mediated allelic exchange. J Bacteriol 177: 5959-5970.
Sadziene, A., Rosa, P.A., Thompson, P.A., Hogan, D.M., and Barbour, A.G. (1992) Antibody-resistant mutants of Borrelia burgdorferi: in vitro selection and characterization. J Exp Med 176: 799-809.
Saint Girons, I., Old, I.G., and Davidson, B.E. (1994) Molecular biology of the Borrelia, bacteria with linear replicons. Microbiology 140: 1803-1816.
Sambri, V., and Lovett, M.A. (1990) Survival of Borrelia burgdorferi in different electroporation buffers. Microbiologica 13: 79-83.
Samuels, D.S. (1995). Electrotransformation of the spirochete Borrelia burgdorferi. Electroporation Protocols for Microorganisms. Totowa, New Jersey, Humana Press. 253-259.
Samuels, D.S., and Garon, C.F. (1993) Coumermycin A1 inhibits growth and induces relaxation of supercoiled plasmids in Borrelia burgdorferi, the Lyme disease agent. Antimicrob Agents Chemother 37: 46-50.
Samuels, D.S., and Garon, C.F. (1997) Oligonucleotide-mediated genetic transformation of Borrelia burgdorferi. Microbiology 143: 519-522.
Samuels, D.S., Mach, K.E., and Garon, C.F. (1994a) Genetic transformation of the Lyme disease agent Borrelia burgdorferi with coumarin-resistant gyrB. J Bacteriol 176: 6045-6049.
Samuels, D.S., Marconi, R.T., Huang, W.M., and Garon, C.F. (1994b) gyrB mutations in coumermycin A1-resistant Borrelia burgdorferi. J Bacteriol 176: 3072-3075. Shigekawa, K., and Dower, W.J. (1988) Electroporation of eukaryotes and prokaryotes: a general approach to the introduction of macromolecules into cells. BioTechniques 6: 742-751.
Steere, A.C. (1994) Lyme disease: a growing threat to urban populations. Proc Natl Acad Sci U S A 91: 2378-2383.
ter Huurne, A.A.H.M., van Houten, M., Muir, S., Kusters, J.G., van der Zeijst, B.A.M., and Gaastra, W. (1992) Inactivation of a Serpula (Treponema) hyodysenteriae hemolysin gene by homologous recombination: importance of this hemolysin in pathogenesis of S. hyodysenteriae in mice. FEMS Microbiol Lett 92: 109-114.
Tilly, K., Casjens, S., Stevenson, B., Bono, J., Samuels, D.S., Hogan, D. and Rosa, P. (1997) The Borrelia burgdorferi circular plasmid cp26: conservation of plasmid structure and targeted inactivation of the ospC gene. Mol. Microbiol. 25:361-373.