Investigation of galectin-3 interaction with N. meningitidis and its dimerization with laminin receptor

• Main Author: Alqahtani, Fulwah Yahya Saleh
• Published: University of Nottingham 2012
• Subjects: 616
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.568705

Meningococcal meningitis from the causative organism Neisseria meningitidis is the leading cause of meningitis globally. This bacterium is among a limited number of pathogens that have the propensity to cross the blood brain barrier (BBB) vasculature causing meningitis. It has been recently demonstrated that Neisseria meningitidis targets the laminin receptor (37 LRP/67 LR) on the surface of human brain microvascular endothelial cells, and two meningococcal outer membrane proteins, PorA and PilQ, have been identified as bacterial ligands. Interestingly, this interaction is hypothesized to underlie meningococcal tropism for the central nervous system (CNS). There are two isoforms of laminin receptor; monomeric 37 kDa laminin receptor precursor (37 LRP) and mature 67 kDa laminin receptor (67 LR). The relationship between the 67 LR and its precursor 37 LRP is not completely understood, but previous observations have suggested that 37 LRP can undergo homo- and/or hetero- dimerization with Galectin-3 (Gal-3) to form mature 67 LR. Gal-3 is the only member of the chimera-type group of galectins, and has one C-terminal carbohydrate recognition domain (CRD) that is responsible for binding the ß-galactoside moieties of mono- or oligosaccharides on several host and bacterial molecules, including neisserial lipooligosaccharide (LOS). To identify the LOS-independent meningococcal ligands that bind Gal-3, binding of lactose liganded Gal-3 and CRD with meningococci was investigated using ELISA assay. Neisseria meningitidis bound lactose liganded Gal-3 significantly more than H. pylori, which is known to bind Gal-3 via LPS. This binding was not inhibited by increasing concentrations of lactose. Also the lactose liganded CRDof Gal-3 bound meningococci but to a lesser extent than full molecule. Importantly, binding of Gal-3 was conserved among 25 meningococcal clinical isolates tested in the current study. A meningococcal mutant lacking the glycosyltransferase required for chain elongation from the core lipid A-(KDO)2-Hep2 showed reduced binding to lactose-liganded Gal-3, but binding was not abolished indicating that the meningococcal-Gal-3 binding was not entirely LOS-dependant. Using a re-tagging approach, meningococcal PilQ and PilE proteins were identified as Gal-3 binding ligands. Mutation of the genes encoding either of these two molecules in strain MC58 led to a significant reduction in Gal-3 binding. PilQ is not known to be glycosylated, therefore its interaction with Gal-3 is likely to be protein-mediated. PilE is post-translationally glycosylated and deletion of the pilin glycosylation genes pglC and/or pglL dramatically reduced bacterial-Gal-3 binding. Given the binding of meningococcal PilQ to 37 LRP/67 LR and Gal-3, this study sought to investigate possible dimerization between 37 LRP and Gal-3 to form 67 LR. Double immunofluorescence staining of endogenous receptors revealed colocalization of 67 LR with its precursor and both of them with Gal-3 in HBMECs, astrocyte and COS7 cells. Moreover, co-expression of 37 LRP and Gal-3 fused to different fluorescent proteins indicated colocalization of these receptors in COS7 cells. Using bimolecular fluorescence complementation (BiFC) assays, the presence of 67 LR in homo- and hetero-dimer forms with Gal-3 has been confirmed in different cell lines. In addition, the recombinant laminin receptor bound Gal-3 and its CRD to comparable level. Further investigation for Gal-3 and 37 LRP dimerization mechanism revealed that the conserved cysteine (C173A) within the CRD of human Gal-3, which is known to abolish disulphide-mediated dimerization of murine Gal-3, is critical for Gal-3 homo- and hetero-dimerization with 37 LRP, whereas neither of the two cysteines on 37LR (cys148 and cys163) are required for dimerization. To examine the role of Gal-3 in meningococcal interaction with host cells, the adhesive and invasive capacities of meningococci were compared between Gal-3 transfected and non-transfected neuroblastoma cell line (N2a) cells. Transient expression of Gal-3 in mouse N2a cells significantly enhanced meningococcal invasion when compared with non-transfected cells. Moreover, infection of CD46-expressing transgenic mice with meningococcal strain MC58 significantly increased the expression of Gal-3 and 37 LRP in the brain. This work also attempts to study whether the 37 LRP/67 LR meningococcal ligands (rPorA, loop 4 of PorA and rPilQ) have any influence on the surface level of 67 LR and Gal-3. As indicated by flow cytometry analysis, recruitments of 67 LR and Gal-3 to the surface of HBMECs were increased in cells incubated with rPilQ, Loop 4 of PorA and more prominently rPorA. To examine these results in more detail, effect of each of these ligands on 37 LRP expression was investigated using qPCR. Loop4 of PorA and rPilQ induced 37 LRP expression significantly more than PBS. Although there was a trend for an increase in 37 LRP expression with treatment with rPorA, the difference was not statistically significant (p = 0.1507). Further investigation in future study for the effect of these bacterial adhesins on Gal-3 gene expression will be of great value. Collectively, these data revealed the capacity of Gal-3 to target meningococcal PilQ and PilE, as well as the previously known LOS and showed the importance of Gal-3 in the meningococcal-host cell interaction. This interaction may be part of host-cell defence against the organism, and/or, conversely, it may be part of a strategy adopted by the organism to modulate the host response and facilitate its invasion. Remarkably, the current findings also demonstrated the existence of 67 LR as homo- and hetero- dimer with Gal-3. This dimerization of two meningococcal host receptors may help to extend spectrum of their bacterial adhesins which may act cooperatively or synergistically at different stages of infection. Besides, the expression pattern of these receptors may suggest specific receptor repertoire in the BBB which might contribute in meningococcal tropism for the CNS.