Borrelia burgdorferi, the causative agent of Lyme disease, is the most prevalent arthropod-borne infectious agent in the United States and, along with infections attributed to other related Borrelia spp. worldwide, contributes to a significant amount of morbidity. Much of what is known regarding pathogenic mechanisms of B. burgdorferi has been limited to biochemical analyses due to the genetic intractability of this spirochetal pathogen. Recently, advances by a number of investigators have made it possible to isolate and genetically complement isogenic mutants in infectious B. burgdorferi as a first step in determining the role of specific genes in Lyme borreliosis. The approach described here involves a customized transposon that contains an antibiotic resistance marker expressed from a strong borrelial promoter, which is used to mutagenize a borrelial gene target in vitro. The resulting transposon mutagenized candidates are then mapped and sequenced. The desired constructs, containing the inactivated gene, are transformed into a low passage but non-infectious B. burgdorferi strain lacking the 25 kb linear plasmid (lp25) and putative transformants selected with the appropriate antibiotic. The lp25 deficient strain used here is more amenable to transformation due to the absence of an lp25-specific restriction/modification system. Using this genetic background, several borrelial genes purported to be involved in B. burgdorferi pathogenesis have been genetically inactivated. The loss of infectivity associated with the lp25 deficient background can be restored back to wild-type levels by transformation with the borrelial shuttle vector pBBE22, which encodes the lp25 bbe22 locus. The bbe22 (pncA) gene of this strain encodes a nicotinamidase enzyme required for de novo biosynthesis of NAD only during infection. Thus, transformation with pBBE22 serves two purposes, as it: (i) restores infectivity in the parent background, and (ii) asks whether the resulting mutant exhibits an infectivity deficient phenotype relative to the parent background using the well-established mouse model of infection. Subsequent genetic complementation with an intact version of the targeted gene in pBBE22 is then carried out to ensure that the mutant phenotype observed can be attributed to the inactivated gene. The strategy presented here could be applied to the genetic analysis of other borrelial genes to determine their significance in the pathogenesis of Lyme borreliosis.