Attachment

As the negatively chlamydial elementary body approaches the negatively charged host cell surface, there is an electrostatic repulsive barrier to be overcome. From the DLVO theory of colloid physics (Derjaguin-Landau-Verwey-Overbeek theory)  the energy of interaction of two charged particles of like sign and magnitude is the sum of the electrostatic energy of repulsion and the energy of attraction provided by electromagnetic interactions due to London-van der Waal's forces [Watt & Ward, 1980]. Penetration through the electrostatic repulsive barrier will be minimized by a small radius of curvature between the approaching surfaces. For this reason, initial chlamydial contact is likely to be with microvillous appendages of the host cell, as is observed in practice (Figs 1 and 2).  

With respect to attachment, the two main biovars of C. trachomatis clearly behave differently. For the trachoma biovar, positively charged macromolecules such as DEAE dextran or poly-L-lysine, increase chlamydial infectivity for HeLa cells in tissue culture. This is not the case for the LGV biovar [Kuo et al., 1973]. This probably reflects the ability of positively charged molecules to neutralise the net negative charge at the host cell surface, coupled with the fact that LGV strains are less negatively charged. The dominant proteins in the outer membrane complex of C. trachomatis are MOMP and the cysteine-rich proteins of molecular weights 12-15 KDa and ~ 40 KDa respectively. These proteins are less negatively charged in the LGV biovar strains than in trachoma biovar strains [Hackstadt, 1999]. This would reduce the electrostatic barrier to attachment for the LGV biovar strains and might be one explanation as to why these strains do not require centrifugation to efficiently infect susceptible cells in laboratory culture. Figs 3 and 4 show the effect of different concentrations of negatively charged molecules on the infectivity of LGV and trachoma biovar strains of C. trachomatis:

While heparan sulphate undoubtedly has an effect on the infectivity of the LGV biovar, its actual role is controversial. 

One school of thought is that the effect is non-specific and charge mediated, which would explain why the more sulphated and thus more negatively charged heparan sulphate is more inhibitory to LGV biovar infectivity via DLVO forces [Watt & Ward, 1980] than the less sulphated members of the GAG family. 

A second view, proposed by Zhang & Stephens, 1992 is that chlamydiae synthesize their own heparan sulphate-like GAG, which acts as a trimolecular bridge between the chlamydial envelope, chlamydial heparan and host cell heparan receptors. The infectivity of heparitinase-treated elementary bodies could be restored with heparan sulphate [Zhang & Stephens, 1992]. A heparan sulphate-specific monoclonal antibody specifically delineated GAG on the surface of C. trachomatis and C. pneumoniae, neutralising their infectivity. GAG staining was also observed on chlamydiae within the intracellular inclusion [Rasmussen-Lathrop et al., 2000].  There was some evidence that C. trachomatis L2 might synthesize GAG [Zhang & Stephens, 1992; Rasmussen-Lathrop et al., 2000]. Heparin and heparan sulphate were also found to inhibit the attachment of C. pneumoniae to human epithelial cells. Reduction in infectivity resulted from the binding of heparin to the organism. Enzymatic removal of heparan sulphate moieties from the host cell surface led to a marked decrease in C. pneumoniae infectivity. Mutant CHO cell lines defective in heparan sulphate biosynthesis were less susceptible to C. pneumoniae infection than was the wild-type cell line and  preincubation of the GAG-deficient Chinese Hamster Ovary cells with exogenous heparin greatly increased infectivity [Wuppermann et al., 2002].

This evidence is not, in the view of this writer, definitive evidence of a key role of glycosaminoglycans in anything other than non specific [charge] interactions. If Chlamydia indeed synthesise glycosaminoglycans, which genes from the genome sequence form the biosynthetic pathway for its synthesis? This question needs to be addressed.

Using flow cytometry, Taraktchoglou et al., 2001 were unable to find evidence for chlamydial-specified GAG. Hackstadt, 1999 points out chlamydial synthesis of GAG would be a novel biosynthetic pathway for any bacteria. However, as chlamydiae interact closely with the host cell Golgi apparatus, it is conceivable that both host and chlamydial enzymes might contribute to "chlamydial" GAG synthesis [Hackstadt, 1999]. 

A third view is that host cell surface GAG alone is sufficient for attachment [Su et al., 1996]. However, given the differential effects of GAG on the attachment and infectivity of different Chlamydiaceae species, it seems likely there are both GAG-dependent and GAG-independent mechanisms operating at different stages of the infection process. Thus, Carabeo & Hackstadt, 2001 isolated a a mutant cell line of Chinese hamster ovary cells that was refractory to infection by C. trachomatis L2 serovar. The mutant cell line apparently lacked a temperature-dependent and heparin-resistant binding step that occurred subsequent to the engagement of cell surface heparan sulphate by L2 elementary bodies. Such differences may contribute to the different pathologies associated with the trachoma and LGV biovars of C. trachomatis. The role of heparan sulphate might also be less straight-forward. The chicken pox virus attaches to host cells via a heparan sulphate proteoglycan whose function is to stabilise the virus so that it can interact with cellular mannose-6-phosphate receptor to effect entry [Zhu et al., 1995]. In this context, the fact that MOMP has high mannan oligosaccharide side chains may be very relevant.  

However an unusual feature of C. trachomatis MOMP is that it is N-glycosylated with a high mannose oligosaccharide side chain. This is rare in the kingdom Eubacteria [Lechner & Wieland, 1989]. However as chlamydiae are intracellular bacteria, it is possible the N-glycosylation is performed by host cell enzymes. The oligomannan side chain, cleaved off MOMP with glycanase, itself binds to HeLa 229 cells and inhibits chlamydial attachment and infectivity [Swanson & Kuo, 1994]. The possible role of a chlamydial glycoprotein in attachment is also suggested by the differential ability of glycoprotein-binding plant lectins to block it. N-linked high mannose oligosaccharides competitively inhibit attachment to (and infectivity of chlamydiae in) HeLa cells.  C. trachomatis infects mannose-receptor positive J774E mouse macrophages better than the equivalent J774A mannose-receptor negative cells. In contrast, C. psittaci infected both mannose-receptor negative and positive cells equally well, while C. pneumoniae infected mannose-receptor negative cells better than mannose-receptor positive cells. Thus there are species differences in the attachment, entry and survival of chlamydiae in mouse macrophages [Kuo et al., 2002].

NEXT: Entry into host cells.