| Summary: | Neisseria meningitidis is considered as the main suspect behind the majority of meningitis and sepsis cases globally. The major component of the outer membrane in Neisseria, lipooligosaccharide (LOS), is recognized by Toll-like receptor 4 (TLR4), triggering a pro-inflammatory cytokine response. Particularly, TLR4 recognizes the lipid A moiety of the LOS molecule, and the recognition is determined by the chemical composition of lipid A. The meningococcal lipid A normally has six acyl chains, the optimal number of acyl chains for TLR4 activation. Acyl chain addition is regulated by the products of two genes (lpxL1 and lpxL2), which are present in a number of Gram-negative bacteria as well as N. meningitidis. Unlike the pathogenic B serogroup, the Y serogroup clonal complex 23 is pathogenic, in spite of being naturally LpxL1-deficient. We aimed to investigate how the changes in LOS structure could affect the recognition of Neisseria serogroups by the host, therefore, both B and Y serogroups variants have been used to infect human differentiated bronchial epithelial cells (Calu-3), cultured in an air-liquid interface. In addition, mutant serogroup B strains lacking lpxL1 have been generated, as well as a complemented mutant strains expressing LpxL1. LOS analysis by silver staining suggested the presence of changes in the LOS of the lpxL1 mutant strains that could not be rescued by the complemented lpxL1 gene. In addition, the expression status of selected Neisseria phase variant genes during the infection has been investigated but no changes were observed. Therefore the work was redirected to study bacterial behaviour during the infection of Calu-3. Cells analysis included bacterial growth, bacterial adhesion and invasion, damage to the epithelial barrier and TNF-α production. Differentiated Calu-3 cells were infected with N222.1 (Y serogroup, cc23) and two isolates, designated N459.3 and N459.6. These isolates were obtained as two colony picks from a single nasopharyngeal swab taken from the same individual as isolate N222.1 but six months later, and have identical strain designations (i.e. ST, PorA and FetA types) to N222.1. Furthermore, allelic variation was detected in only five genes for a whole genome comparison of N459.1 to N222.1. Both N222.1 and N459.6 increased the permeability of the epithelial barrier at 12 h post-infection and induced production of high levels of the pro-inflammatory cytokine TNF-α. In contrast, N459.3 did not affect the permeability of the epithelial cell layer or induce TNF-α production. Expression of Opa proteins, a family of Neisserial adhesins that interact with CEACAMs in host cells, varies among isolates and could contribute to differences in their phenotype. These observations contrast with those obtained with control serogroup B strains (MC58 and H44/76; cc32) and another MenY isolate, N59.1 (cc174). The latter isolate, in contrast to N222.1, readily adhered/invaded the epithelial layer, did not disrupt the epithelial barrier, but promoted TNF-α production. Further work demonstrated that blocking of TNF-α production during N222.1 infection does not protect the epithelial layer from infection-induced damage. Preliminary signalling studies showed that N222.1, N459.3 and N459.6 trigger distinct signaling pathways.
The distinct patterns of bacterial behaviour exhibited by isolates N459.3 and N459.6 (with identical strain designations and contemporaneously present in the Nasopharynx of the same carrier) in our in vitro infection model suggest that virulence in N. meningitidis is intrinsically linked to subtle genomic changes. The temporal stability of the core bacterial genome and our observation that tract lengths of phase-variable loci do not alter during in vitro infection assays could enable us to use this and other infection models as platforms to link particular bacterial genomic traits to specific and reproducible phenotypic traits, and hence to assess the risk of closely related isolates causing invasive disease.
|
|---|