Novel antibiotic-resistance markers in pGK12-derived vectors for Borrelia burgdorferi.

Extension of molecular genetics studies in Borrelia burgdorferi has been hampered by a lack of a variety of antibiotic resistance selective markers. Such markers are critical for isolation of B. burgdorferi strains with multiple mutants, for complementation with different cloning vectors, and for methods such as negative selection and reporter genes. To remedy this lack, resistance to various antibiotics of non-infectious (B31, 297) and infectious (N40) B. burgdorferi strains was examined and vectors incorporating appropriate antibiotic resistance genes as selective markers were developed. Minimal inhibitory concentrations for growth of B. burgdorferi on plates and in liquid media for aminoglycosides (kanamycin, gentamycin, sisomycin, amikacin, spectinomycin, neomycin), macrolides-lincosamids (erythromycin, lincomycin), coumarin derivatives (coumermycin A(1), novobiocin), glycopeptides (vancomycin, ristocetin), peptides (bacitracin, cycloserine), and chloramphenicol were found to differ significantly. There were also striking differences in resistance to these antibiotics between non-infectious and infectious B. burgdorferi strains. Antibiotic-resistance genes aph(3')-IIIa from Streptococcus faecalis, aad9 from Staphylococcus aureus Tn554, linA' from Staphylococcus aureus, and aac(3)-VIa from Enterobacter cloacae (conferring resistance to kanamycin, spectinomycin, lincomycin, and gentamycin/sisomycin, respectively) were subcloned either with their own promoters or under the control of the B. burgdorferi flaB promoter into pGK12 or its derivative pED1 to develop new cloning vectors for B. burgdorferi with the rationale that the absence of homologous regions between derived recombinant plasmids lacking the flaB promoter and the B. burgdorferi genome would permit avoidance of possible recombination with target DNA. Resistance to the corresponding antibiotic was conferred by vectors containing aph(3')-IIIa, aad9, linA' or aac(3)-VIa whether under the control of their own promoters or under the control of the flaB promoter. We conclude that these markers can be used for genetic study of B. burgdorferi and suggest they will be an important addition to the previously used coumermycin A(1), erythromycin and kanamycin in these studies.