Assembly of capsular polysaccharides

Capsules are surface structures that cover the surfaces of many Gram-negative and Gram-positive bacteria. They are comprised of polysaccharides with molecular weights in the range of 10⁵-10⁶ daltons. Capsules play critical roles in bacterial physiology and these functions vary according to the organism and its habitat. For example, they are often virulence factors and required to protect the bacterium from phagocytosis. They can also promote colonization, and may be involved in biofilm formation.

Electron micrographs of encapsulated Escherichia coli (left)
and Klebsiella pneumoniae (right) with group 1 capsules

There are two pathways used for assembly of capsules in Gram-negative bacteria. The prototypes are the group 1 and group 2 capsules of E. coli (Whitfield, 2006). We currently focus on the mechanisms of group 1 capsule assembly. In the working model, we envision the biosynthesis proteins forming a multienzyme complex spanning the cell envelope. Synthesis and export are coupled in this molecular machine.

A model for the envelope-spanning enzyme complex involved in group 1 capsule assembly

The biosynthesis reaction sequence is as follows:

1. glycosyltransferase enzymes (represented in green) add the glycose residues to a lipid carrier to generate an undecaprenol diphosphate-linked polymer repeat unit.
2. a reaction involving the Wzx protein is involved in the transfer of the lipid-linked repeat units across the inner membrane.
3. polymerization of the repeat units occurs by a Wzy protein-dependent mechanism. The polymerization process may be modulated by the tyrosine autokinase, Wzc. Transphosphorylation occurs within Wzc oligomers and the products are dephosphorylated by Wzb.
4. a multimeric complex formed by the Wza and Wzc proteins is required to translocate the polymer to the cell surface.

---

Current projects

The Wza multimer complex - a polysaccharide translocon

The terminal stages in the assembly of capsular and exopolysaccharides require the translocation of a high-molecular-weight polymer across the outer membrane. This is complicated by the need to maintain the essential barrier properties of the outer membrane. An outer membrane lipoprotein known as Wza is involved in surface assembly of the group 1 capsule in E. coli serotype K30 (Drummelsmith and Whitfield, 2000). Mutants lacking Wza are unable to assemble capsule on the cell surface. Wza proteins form a detergent-resistant octameric complex (Nesper et al. 2003).

In collaboration with the labs of Dr. JH Naismith (University of St. Andrews, UK) and Dr. R. Ford and Dr. R Collins (University of Manchester, UK), we are resolving the structure of Wza through complementary experimental approaches including cryoelectron microscopy (with 3D-image reconstruction) and X-ray crystallography. Cryo-electron microscopy with negative stain gave the first view of the shape of the Wza octamer. (Beis et al. 2003). A more detailed view has now become available with the crystal structure, solved at 2.26 Å resolution (Dong et al. 2006). This unprecedented structure has significant implications for polysaccharide export in particular and for outer membrane function in general.

Structure of the Wza translocon at 2.26 Å resolution.

1. shows a Wza monomer and its contribution to four distinct domains in the complex.
2. shows the octamer with the central cavity identified by the brown space-filling shape. The C-termini of the monomers are exposed at the cell surface, placing the α-helical barrel in the outer membrane (marked) and the three rings (R1-R3) in the periplasm.
3. shows the view from outside the cell, revealing an open entry. Loops are evident that occlude the channel at the (periplasmic) base.
4. represents the surface-rendered view from the periplasmic face and shows a closed concave surface with a band rich in negative electrostatic charge (red). (from Dong et al. 2006)

The complex shows eight-fold symmetry with each monomer contributing to each of the four structural domains. Surprisingly, the majority of the complex resides outside the outer membrane, within the periplasm. The outer membrane domain is a novel α-helical barrel comprising the C-termini of the eight Wza monomers. This forms a channel that is open to the extracellular environment and represents the first example of an outer membrane channel that is not formed by a β-barrel. The remainder of the protein is comprised of three periplasmic rings (R1-R3), each with a novel fold. The centre of the bottom ring (R1) of the oligomer is filled by eight loops, one from each monomer. Thus, the purified Wza complex exists in a conformation where the periplasmic face is "closed". This may reflect the possibility that permanently open channels would compromise the barrier function of the outer membrane. We are testing this and other hypotheses by studying the structures of the translocon and its interacting proteins, particularly Wzc.

The critical function of phosphorylation of Wzc in group 1 capsule assembly

Mutants lacking Wzc cannot form a capsule structure but low level polymerization (forming oligomers of 1-5 repeat units) is still evident. The Wzc protein is located in the inner membrane. It autophosphorylates at several tyrosine residues located at the C-terminus in a mechanism that is dependent on a Walker box motif (Wugeditsch et al. 2001). The phosphorylation process then involves a transphosphorylation reaction between monomers within an oligomeric complex.

Organization and membrane topology of the Wzc protein

Capsule assembly is dependent on both the autophosphorylation of Wzc and its dephosphorylation by the cognate phosphatase, Wzb. This suggests that Wzc may cycle between a phosphorylated and dephosphorylated state. Up to seven tyrosine residues can be phosphorylated in Wzc and systematic removal of tyrosine residues through site-directed mutagenesis results in a reduction in (and finally elimination of) the capacity to support capsule production (Paiment et al. 2003). Attempts to solve the structure of Wzc are underway to shed light onto the mechanism of phosphorylation and its role in capsule assembly.

Wzc forms a complex with four-fold symmetry that is evident in electron microscopy images of cryo-negatively stained samples from single particle analysis (Collins et al. 2006). The structure of the tetramer exhibits an extensive periplasmic domain with four unconnected elongated domains extending from it into the cytoplasm. The overall complex resembles a molar tooth with a crown and four roots. The base of each root is proposed to contain a tyrosine autokinase domain.

Surface-rendered images of the Wzc tetramers revealed by electron microscopy with cryo-negative staining.
Resolution = 14Å. From Collins et al. 2006

Current information suggests that Wzc plays two roles in capsule assembly. The first involves maintenance of high-molecular-weight polymer synthesis. The underlying mechanism is unclear but it may involve interaction with other biosynthesis proteins. The second function of Wzc involves coordination of polymer biosynthesis and export. It does this by interacting with Wza (Nesper et al. 2003).

Wza and Wzc form a trans-envelope protein complex

Purified Wza and Wzc form a complex when mixed in vitro. The complex comprises one Wza octamer and one Wzc tetramer and has been visualized by electron microscopy with cryo-negative staining (Collins et al. manuscript submitted).

Structure and proposed organization of the Wza:Wzc complex in the bacterial envelope.
The 3-D structure of Wza-Wzc is shown in green wire-frame. The Wza octamer crystal structure (purple ribbon) and Wzc (yellow) have been fitted manually into the complex volume using CHIMERA. The orange densities indicate the position of Ni-NTA gold labels bound to the N-terminus of Wzc. The inferred locations of the inner and outer membranes are shown.

Comparison of the structure of the Wza:Wzc complex with those of the individual components suggests that, within the complex, Wza undergoes major conformation changes in the domain which interacts with Wzc. This minimally involves an opening of the bottom ring (R1) of the Wza structure. Genetic evidence indicates that there is specificity in interaction between Wza and Wzc (Reid et al. 2005) and this presumably reflects sequence variation in the interfaces.

The structure of the Wza:Wzc complex provides insight into possible mechanism(s) to coordinate elements in the cytoplasm and inner membrane involved in polymer biosynthesis, with those components that mediate export across the periplasm and through the outer membrane. Current studies are aimed at resolving the molecular details of these interactions, the mechanism by which Wzc contributes to the opening of the Wza channel, and the involvement of the tyrosine autokinase activity of Wzc in the overall process.

Homologs of Wza and Wzc are required for biosynthesis of many surface polysaccharides, so these studies will provide broad insight into polymer translocation in diverse Gram-negative bacteria. Also, the Wza multimeric complex resembles "secretins" involved in some protein secretion systems so the findings may have wider implications.

top