A central aspect of Brucella pathogenicity is its ability to invade, survive, and replicate in diverse phagocytic and non-phagocytic cell types, leading to chronic infections and chronic inflammatory phenomena. Adhesion to the target cell is a critical first step in the invasion process. Several Brucella adhesins have been shown to mediate adhesion to cells, extracellular matrix components (ECM), or both. These include the sialic acid-binding proteins SP29 and SP41 (binding to erythrocytes and epithelial cells, respectively), the BigA and BigB proteins that contain an Ig-like domain (binding to cell adhesion molecules in epithelial cells), the monomeric autotransporters BmaA, BmaB, and BmaC (binding to ECM components, epithelial cells, osteoblasts, synoviocytes, and trophoblasts), the trimeric autotransporters BtaE and BtaF (binding to ECM components and epithelial cells) and Bp26 (binding to ECM components). An in vivo role has also been shown for the trimeric autotransporters, as deletion mutants display decreased colonization after oral and/or respiratory infection in mice, and it has also been suggested for BigA and BigB. Several adhesins have shown unipolar localization, suggesting that Brucella would express an adhesive pole. Adhesin-based vaccines may be useful to prevent brucellosis, as intranasal immunization in mice with BtaF conferred high levels of protection against oral challenge with B. suis.

Brucella infection is frequently acquired by inhalation, but the pathogen disseminates systemically from the lungs. However, little is known about the interaction of Brucella spp. with the airways. Using a 3D air-exposed organotypic human bronchial tissue model (polarized 16HBE14o- bronchial epithelial cells grown over a collagen matrix containing MRC-5 lung fibroblasts), we analyzed Brucella abortus replication, translocation and cytokine responses over prolonged post-infection times. Apically inoculated B. abortus invaded, replicated and persisted during the whole follow-up (16 days) within the bronchial tissue without inducing cytotoxicity. Viable bacteria were also detected in the conditioned medium (CM) since day five post-infection, indicating release from the basolateral side. In parallel experiments, no invasion or bacterial release was detected for Escherichia coli. The levels of IL-6, IL-8 and MCP-1 were increased in CM from Brucella-infected 3D cultures and in monocultures of polarized bronchial epithelial cells or lung fibroblasts. Collagenase/gelatinase activity was increased in 3D cultures and MRC-5 monocultures. Infection transference from bronchial cells to lung fibroblasts was documented using monocultures. An immune cross-talk was detected, as cytokine levels were increased in fibroblasts stimulated with bronchial CM, and vice versa. These results suggest that the bronchial mucosa can sustain B. abortus persistence, replication and dissemination, and that it induces a proinflammatory response to which both epithelial cells and fibroblasts contribute.