As a first approach to establishing a three-dimensional culture infection model, we studied the growth behavior of the extracellular pathogen Yersinia enterocolitica in three-dimensional collagen gels (3D-CoG). Surprisingly, we observed that plasmidless Y. enterocolitica was motile in the 3D-CoG in contrast to its growth in traditional motility agar at 37°C. Motility at 37°C was abrogated in the presence of the virulence plasmid pYV or the exclusive expression of the pYV-located Yersinia adhesion gene yadA. YadA-producing yersiniae formed densely packed (dp) microcolonies, whereas pYV⌬yadA-carrying yersiniae formed loosely packed microcolonies at 37°C in 3D-CoG. Furthermore, we demonstrated that the packing density of the microcolonies was dependent on the head domain of YadA. Moreover, dp microcolony formation did not depend on the capacity of YadA to bind to collagen fibers, as demonstrated by the use of yersiniae producing collagen nonbinding YadA. By using a yopE-gfp reporter, we demonstrated Ca 2؉ -dependent expression of this pYVlocalized virulence gene by yersiniae in 3D-CoG. In conclusion, this study revealed unique plasmid-dependent growth behavior of yersiniae in a three-dimensional matrix environment that resembles the behavior of yersiniae (e.g., formation of microcolonies) in infected mouse tissue. Thus, this 3D-CoG model may be a first step to a more complex level of in vitro infection models that mimic living tissue, enabling us to study the dynamics of pathogen-host cell interactions.For decades, cell culture monolayers have successfully been used to study mechanisms of microbial adherence, invasion, and intracellular survival/multiplication. For in vitro infection studies, eukaryotic cells are grown on solid supports as monolayers and then are challenged with the respective pathogen. In spite of the liquid culture medium (third dimension) covering the adherent cell monolayer, this infection model can be considered a two-dimensional system which obviously does not reflect the environment of host tissue and dynamic events, such as cell migration, happening during infection. Host tissue (i) is vascularized, allowing the extravasation and migration of host immune cells toward the invading microbe, and (ii) consists of a network of extracellular matrix (ECM) proteins with diverse resident cells (e.g., fibroblasts, macrophages, etc.). Thus, there are many reasons to leave the two-dimensional system and establish an in vitro three-dimensional infection model by approaching in vivo conditions. In order to simulate a tissue-like environment, immunology and cell biology as well as tumor biology make use of three-dimensional collagen matrices to study cell migration, cell-cell interactions, and cell-matrix interactions (reviewed in reference 13). To develop and assess such a three-dimensional infection model, we started with the well-established, three-dimensional collagen gel (3D-CoG) on microscopic glass supports and with Yersinia enterocolitica as the prototype of an extracellular pathogen.Y. enterocoli...