Neisseria gonorrhoeae pilin glycan contributes to CR3 activation during challenge of primary cervical epithelial cells

Abstract

Expression of type IV pili by Neisseria gonorrhoeae plays a critical role in mediating adherence to human epithelial cells. Gonococcal pilin is modified with an O-linked glycan, which may be present as a di- or monosaccharide because of phase variation of select pilin glycosylation genes. It is accepted that bacterial proteins may be glycosylated; less clear is how the protein glycan may mediate virulence. Using primary, human, cervical epithelial (i.e. pex) cells, we now provide evidence to indicate that the pilin glycan mediates productive cervical infection. In this regard, pilin glycan-deficient mutant gonococci exhibited an early hyper-adhesive phenotype but were attenuated in their ability to invade pex cells. Our data further indicate that the pilin glycan was required for gonococci to bind to the I-domain region of complement receptor 3, which is naturally expressed by pex cells. Comparative, quantitative, infection assays revealed that mutant gonococci lacking the pilin glycan did not bind to the I-domain when it is in a closed, low-affinity conformation and cannot induce an active conformation to complement receptor 3 during pex cell challenge. To our knowledge, these are the first data to directly demonstrate how a protein-associated bacterial glycan may contribute to pathogenesis.

Fig. 1

Diagram denoting the contribution of PglA and PglD to the expression of the pilin glycan, [Gal (β1-4)] Gal (α1-3)-DATDH glycan. PglD, in conjunction with other pgl gene products, functions in the biosynthesis of DATDH, which is O-linked to the pilin monomer. PglA, functioning as a galactosyl transferase, then adds a galactose to the DATDH core sugar. An additional galactosyl transferase, PglE, may add a terminal galactose to Gal-DATDH to form a trisaccharide. However, in the majority of gonococcal strains, PglE is switched off (Power et al., 2007).

Fig. 2

Scanning electron microscopic analysis of pex cells. Scanning electron microscopy analysis of uninfected pex cells (A) or pex cells that were challenged for 30 min with (B) 1291 wild-type (C) 1291pglA or (D) 1291pglD was performed as indicated in the text. Lightened areas/blemishes (encircled) are readily evident across the surface of pgl mutant-infected cells that are not present on the surface of wild-type-infected cells and hint at a role for the pilin glycan in mediating a hyper-adhesive and/or hyper-invasive phenotype during cervical infection. Magnification: (A) ×1 K, (B) ×9 K, (C) ×7 K and (D) ×7 K.

Fig. 3

Glycan-deficient gonococci exhibit an early, pilus-dependent, hyper-adhesive phenotype during pex cell challenge. Pex cells were challenged with gonococci for various times as noted (A and B) or infections were limited to 30 min (C and D). The ability of wild-type and mutant bacteria to associate with (A, C, D) or to invade (B) pex cells was quantified as a percentage of the original inocula, as described in the Experimental procedures. (C) Comparative assays were performed using pgl mutant bacteria in the presence (+) and absence (−) of their respective (isogenic) PglA or PglD pili competimer (10 ng ml⁻¹) or (D) by using wild-type bacteria without (i.e. none) or with (10 ng ml⁻¹) wild-type, PglA or PglD pili competimers, as noted. *P ≥ 0.66 for comparisons of 1291pglA or 1291pglD bacteria with 1291 wild-type. †P = 0.052 for the comparison of 1291 bacteria in the presence of wild-type isogenic pili with pili isolated from 1291pglA.

Fig. 4

Glycan-deficient gonococci co-immunoprecitate with CR3. Pex cells were challenged with N. gonorrhoeae strains 1291 (A and B), 1291pglA (A) or 1291pglD (B) for various lengths of time, as indicated across the top of each Western blot image. The CR3 alpha subunit, CD11b, was immuno-captured from pex cell lysates using (rabbit) antibody, H-61. The presence of a single band of a size equivalent with gonococcal pili, after immuno-blotting with the (mouse) IE₈G₈ anti-pilin monoclonal antibody, is consistent with a CR3–gonococcal pilin interaction occurring during the course of pex cell challenge. UI, uninfected pex cells; Ab, the anti-CD11b capture antibody was omitted from the initial capture step; Pil, purified PglA (A) or PglD (B).

Fig. 5

pglD mutant gonococci cannot bind to the closed conformation of the CR3 I-domain. A. Purified gonococcal pili that were transferred to a solid support medium were subjected to Far-Western blot analysis using wild-type (left panel), locked-open (middle panel) or locked-closed (right panel) rI-domain, as outlined in the text. The presence of an approximate 18–20 kDa band, after immuno-blotting with the (mouse) LM2/1 I-domain-specific antibody, is indicative of rI-domain bound to immobilized gonococcal pilin. B. Microtitre plates were coated with purified 1291, 1291pglA or 1291pglD gonococcal pilus, as noted; incubated with wild-type, locked-open or locked-closed rI-domain (also as indicated); and subjected to ELISA analysis using the I-domain-specific (mouse) antibody, Bear 1, as described in the text. *P ≥ 0.35 when comparing the use of wild-type rI-domain with the use of locked-open rI-domain.

Fig. 6. The pilin glycan is required to bind to CR3 in a low-affinity conformation on pex cells

A. Pex cells were challenged with 1291, 1291pglA or 1291pglD gonococci for 30 min in the presence or absence of wild-type, locked-open or locked-closed rI-domain, as noted. B. Alternatively, gonococci pre-opsonized with rI-domain were used to challenge pex cell monolayers. The ability of gonococci to adhere to pex cells in the presence or absence of the I-domain competimer was measured by an ELISA in which (mouse) antibody 2C3, which is specific for the conserved H.8 outer membrane gonococcal protein was used. *P ≥ 0.36 for assays performed in the presence of the rI-domain competimer when compared with assays in which the rI-domain competimer was omitted.

Fig. 7

The pilin glycan is required to confer an active state to CR3 during pex cell challenge. Pex cell monolayers in 96-well plates were challenged for 30 min with wild-type (WT) and mutant N. gonorrhoeae, isolated gonococcal pili, ChoP or PEtN, or they were left unchallenged (UI), as indicated. The ability of each biological agent to induce an active, high-affinity, conformation to CR3 was determined using the CR3 active state-specific antibodies CBRM1/5 (A) or MEM48 (B and C) and the anti-CD11b (i.e. total CR3) antibodies H-61 (A) Ox42 (B) or LM2/1 (C) by a fluorometric ELISA, as described in the Experimental procedures. Data shown are the percentage of the amount of active CR3, relative to the amount of total CR3, present on the pex cell surface following challenge with (A) 1291 wild-type and pgl mutant bacteria and isolated pili, (B) 1291 and MS11 wild-type and pgl mutant bacteria or (C) pili isolated from 1291 or MS11 wild-type or pgl mutant bacteria. *P ≥ 0.15 for pex cells challenged as indicated when compared with unchallenged cells.