RESUMEN
A critical component in clinical trials for vaccines against pneumococcal disease is the establishment of robust preclinical models and clinical correlates of protection, which, in the case of the causative bacterial agent Streptococcus pneumoniae, include standard sepsis/pneumonia mouse models and opsonophagocytic activity (OPA), respectively. Despite broad usage, these gold-standard measures are ill equipped to evaluate nontraditional antigens that target virulence factors beyond capsular polysaccharides and/or proteins not associated with colonization or routine growth. These assays are further complicated by observed inconsistencies in the expression of target protein antigens and in the quantity of usable bacteria provided from respective growth processes. In an effort to overcome these issues, we performed an extensive optimization study of the critical steps in a bacterial biofilm dispersion model (termed "the biofilm model") to identify conditions that yield the greatest quantity of released pneumococci displaying a consistent virulence phenotype. Using this knowledge, we developed a secondary immune absorbance assay to provide immediate insight into the phenotypic state of bacteria conditioned using the biofilm model. Specifically, positive correlations between the expression of PncO (a key virulence-associated protein antigen) and immune absorbance (R2 = 0.96), capsule shedding, and OPA assay titers were translated into a predictive readout of virulence in sepsis and pneumonia challenge models. These results present a methodology for generating consistent lots of virulent bacteria to standardize inputs in preclinical and clinical models for testing vaccines against biofilm-associated bacteria.
