There are many examples of positive and negative interactions between different species of bacteria inhabiting the same ecosystem. This observation provides the basis for a novel approach to preventing microbial diseases called replacement therapy. In this approach, a harmless effector strain is permanently implanted in the host's microflora. Once established, the presence of the effector strain prevents the colonization or outgrowth of a particular pathogen. In the case of dental caries, replacement therapy has involved construction of an effector strain called BCS3-L1, which was derived from a clinical Streptococcus mutans isolate. Recombinant DNA technology was used to delete the gene encoding lactate dehydrogenase in BCS3-L1 making it entirely deficient in lactic acid production. This effector strain was also designed to produce elevated amounts of a novel peptide antibiotic called mutacin 1140 that gives it a strong selective advantage over most other strains of S. mutans. In laboratory and rodent model studies, BCS3-L1 was found to be genetically stable and to produce no apparent deleterious side effects during prolonged colonization. BCS3-L1 was significantly less cariogenic than wild-type S. mutans in gnotobiotic rats, and it did not contribute at all to the cariogenic potential of the indigenous flora of conventional Sprague-Dawley rats. And, its strong colonization properties indicated that a single application of the BCS3-L1 effector strain to human subjects should result in its permanent implantation and displacement over time of indigenous, disease-causing S. mutans strains. Thus, BCS3-L1 replacement therapy for the prevention of dental caries is an example of biofilm engineering that offers the potential for a highly efficient, cost effective augmentation of conventional prevention strategies. It is hoped that the eventual success of replacement therapy for the prevention of dental caries will stimulate the use of this approach in the prevention of other bacterial diseases.