Disease-attenuated pneumococcal biosynthesis gene mutants invade the mucosal epithelium and induce innate immunity

ABSTRACT Nasopharyngeal colonisation by Streptococcus pneumoniae is characterised by bacterial adherence to epithelial cells, microinvasion and innate immune activation. Previously, we have shown two serotype 6B S. pneumoniae mutant strains affecting bacterial metabolism ( ΔproABC/pia and Δfhs/pia ) colonise humans and mice, but in a murine disease model do not cause invasive infection. Here, we explore whether S. pneumoniae epithelial microinvasion and the induction of innate immune responses persist despite disease attenuation. We show that under serum stress, these biosynthesis gene mutations had a broad but different impact on pneumococcal virulence gene expression, oxidative stress regulation, and purine and carbohydrate metabolism genes. However, although these mutations did not attenuate microinvasion in human challenge and epithelial models, there was less transmigration of Detroit 562 nasopharyngeal epithelial cells by the mutants compared to WT. Cellular reorganisation of primary human airway epithelium varied considerably between strains. Compared to WT, infection of Detroit 562 epithelial cells by the Δfhs/piaA strain, but not the ΔproABC/piaA strain was less pro-inflammatory, induced less caspase 8 production, and were associated with increased pneumococcal hydrogen peroxide and reduced pneumolysin secretion. These findings suggest that the observed differences in microinvasion and the epithelial response were driven by the differential expression of multiple bacterial virulence and metabolic pathways, rather than single genes or pathways of genes. These data highlight the complex impact of single gene mutations on bacterial virulence and suggest that the virulence determinants of pneumococcal epithelial colonisation, microinvasion and innate immunity are not necessarily directly linked to disease. Author Summary Streptococcus pneumoniae (the pneumococcus) commonly colonises the back of the human nose, and is a leading cause of pneumonia, meningitis, and sepsis. During colonisation, the pneumococcus adheres to the cells in the nose, invades these cells (so-called microinvasion), and activates them. Colonisation is a pre-requisite for disease, however, since disease is largely a dead end for S. pneumoniae , it remains unclear whether these processes are directly linked to disease progression. We have previously shown that if we introduce gene mutations into S. pneumoniae that affect key metabolic pathways, these bacteria retain their ability to colonize human and animal models without causing disease. We now show that these mutants retain their ability to microinvade epithelial cells in human and mouse models, and some may still cause inflammation, but are less able to pass through the epithelial barrier. However, although the attenuation of disease may be explained by the broad-ranging impact of these mutations on pneumococcal virulence, oxidative stress, and metabolism, they are not driven by a single determinant. Our findings suggest that pneumococcal microinvasion and immune activation are not necessarily pre-cursors to disease progression. This supports the idea that S. pneumoniae adapts and evolves to promote colonisation and ultimately transmission rather than cause disease. Graphical Abstract S. pneumoniae colonisation is characterised by mucus association, epithelial adherence, microcolony formation and microinvasion – where the pneumococcus invades the epithelial barrier without causing disease. Although mutations in S. pneumoniae biosynthesis genes ( ΔproABC and Δfhs ) attenuate disease in a murine model, they do not attenuate microinvasion in either experimental human pneumococcal challenge (EHPC), ex vivo or in vitro epithelial cells. Transmigration of the epithelial barrier is attenuated. These mutations show strain-dependent effects on both the epithelial and bacterial responses to infection. Factors such as epithelial cellular reorganisation, inflammation and caspase 8 activity alongside pneumococcal metabolic adaptation, virulence factor expression and response to stress are important components of these processes.