Content uploaded by Yaoqin Hong
Author content
All content in this area was uploaded by Yaoqin Hong on May 21, 2019
Content may be subject to copyright.
Diversity of O-antigen Wzx translocases and Wzy polymerases
Peter R. Reeves and Yaoqin Hong
School of Molecular Bioscience, The University of Sydney, NSW 2006
peter.reeves@sydney.edu.au
Background
The O antigen, which is the outermost component of lipopolysaccharide (LPS), is
composed of short repeat units (O units) that vary enormously within and
between species, with about 200 O-antigen forms in Escherichia coli and 54 in
Salmonella enterica
In S. enterica and E. coli, most O antigens are synthesized via the Wzx/Wzy-
dependant pathway, and biosynthesis can be divided into three stages (Fig 1): (1)
biosynthesis of nucleotide sugars (not shown); (2) sequential transfer of sugars to
the lipid carrier, Undecaprenol-pyrophosphate (Und-PP), to form complete O
units (Steps 1 and 2); (3) O-unit processing involving transposition of completed
O units to the periplasmic side of the inner membrane by Wzx (step 3), where it
is subsequently polymerized by Wzy (step 4), followed by ligation of the
polymerized O units to lipid A-core to form complete bacterial LPS which is
exported to the outer membrane (not shown).
Fig 1. O-antigen biosynthesis in S. enterica Group B (modified from Reeves
and Cunneen, 2009).
The absence of O-antigen ladders is due to Wzx specificity,
as the LPS phenotype in D2 strain producing O unit lacking
the DDH side branch can be fully restored by replacing the
chromosomal wzxDby wzxE(Fig 4). However, by restoring
the side branch, the O-antigen ladder was effectively
removed again (Fig 4) (Hong et al 2012).
Fig 2. The processing steps of
Wzx/Wzy-dependent O-antigen
synthesis
The Wzx/Wzy Proteins
The two proteins, Wzx and Wzy, that mediate the translocation of Und-PP-linked O units into the
periplasmic face of the IM and the subsequent polymerization step respectively, both have very diverse
amino acid sequences. For Wzy this relates to variation in the polymerase linkage. Wzx also processes
these diverse O units, but there is abody of evidence that Wzx flippases have specificity only for the first
sugar (Feldman et al., 1999, Marolda et al., 2004). However, we have shown recently that at least some
Wzx flippases have substantial specificity for their native O unit, but that this is masked if Wzx is over-
expressed. We have also looked at Wzy specificity and find that it can extend beyond that acceptor sugar
on the O unit. We summarise the published data on Wzx and present new data on Wzy
Fig 3. Mutants lacking the
abequose or tyvelose side-
branch sugar lack an O-
antigen ladder
Fig 4. WzxEand WzxDrespectively require absence or
presence of tyvelose
O-unit side branch and Wzx specificity Wzx specificity in the original E. coli model and the role of protein overexpression
S. enterica Group B, D1 and D2 strains with deletions of the
abe or prt-tyv genes (synthesis of abequose or tyvelose) or
wbaV gene (DDH transferase) do not have the side-branch
sugar, which produce the same (D2) or similar (B and D1) O
unit as group E.
The mutants all make LPS lacking the O-antigen ladder (Fig 3,
lanes 2, 4, 7 ,9, 12, 14). Complementation of these deletions
restored the O-antigen ladders confirming that the effect is due
to absence of the side-branch sugar.
We used K-12 strain MG1655 to generate strains with the only difference being the chromosomal wzx gene. We switch on O-
antigen expression by either adding a wbbL clone (O16)or a clone of the O111 gene cluster (O111) with wzx deleted. Thus both
substrate are using the Wzx encoded by the chromosomal O-antigen wzx gene for translocation. We replaced the wzxO16 gene in
the K-12 strain with the wzxO111 gene, to put the two genes (wzxO16 and wzxO111) under similar expression.
Reciprocal experiments shows that the Wzx translocases, when in the same location, has normal translocation ability for the
native O unit, but could not effectively process the alternative substrate (Fig 5). However, overexpression of Wzx by adding a
cloned gene can compensate for the reduced effectiveness (lane 5).
It appears that the original conclusion for the lack of specificity (Feldman et al., 1999, Marolda et al., 2004) was affected by
protein overexpression. However, our data does not disclaim the importance of the first sugar, but suggests that specificity can
extend well beyond the first sugar.
We observed persistence of modal phenotype (lane 2), despite the reduced supply of O units. This supports a model in which Wzz
interacts with the growing Und-PP linked chain, and prevents access by WaaL, which would otherwise terminate the
polymerization. Data from Hong and Reeves, 2014.
Data from Hong et al 2012
Wzy polymerization requires the correct Man-Rha configuration in closely related Salmonella
galactose-initiated groups
Fig 5. Reciprocal data suggests Wzx
translocases in E. coli also have very
high substrate requirement
Salmonella groups D1, D2 and D3 have closely related O-unit structures
(Fig 6), the only difference being in the Man-Rha (1-4) linkage, which has
either alpha or beta configuration
The Wzy proteins encoded are not related at sequence level. It appeared that
WzyD1 and WzyD3 can only polymerize O units with alpha (1-4) Man-Rha
linkage, and WzyD2 only those with beta (1-4) linkage (Fig 7).
Group D3 has been reported to produce O units with both alpha and beta
Man-Rha linkages. Our data show that the wzy gene in the gene cluster the
same as in D1, that functions only with O units that have the alpha Man-
Rha linkage.
We also knocked out the wzyD3 gene from the D3 gene cluster, and found
that the mutant still produces O-antigen polymer. (data not shown). This
confirms the earlier structural data that D3 produces 2 types of O units, and
implies that it encodes 2 different Wzy proteins.
# 1 # 2
# 3 # 4
# 5
Fig 6. The Salmonella groups D1, D2 and D3
O-unit structures
Fig 7. Correct Man-Rha (1-4) linkage is an intrinsic factor for Wzy polymerization
•We find that the WzxBand WzxDflippases have a strong preference for O units with aDDH side branch. In contrast theWzxEflippase has a strong preference for
O units without aDDH side branch (Hong et al 2012).
•We examined the set of Wzx proteins that were important for the “first sugar” E. coli model (Hong and Reeves 2014). We obtained reciprocal data such that
WzxO16 and WzxO111, while giving full efficiency for each of their native repeat unit, were much less effective for an alternative substrate.
•However the specificity is not absolute as excess Wzx protein can restore full function. This nuanced effect probably accounts for the finding in other studies that
Wzx flippases had rather limited specificity.
•Ineffective translocation results in reduced O-unit supply. However, despite the much reduced level of Und-PP linked O units available for polymerization, the
modal phenotype has always been maintain whenever long-chain O antigen can be detected. This suggests that Wzz must interact with the growing Und-PP
linked O-unit chain, and prevents WaaL from terminating polymerization.
•We also examined the cross activity for the three Wzy polymerases from the highly related Salmonella galactose-initiated groups, for two substrates which only
differ in the identity of a Man-Rha (1-4) linkage, alpha for D1 and D3, and beta for D2.We found that the native Man-Rha configuration is a critical factor for
successful polymerization, despite the substrates being otherwise identical.
•Diversity of Wzx translocases and Wzy polymerases reflect the diversity of the O-antigen repeat-unit substrates.
# 6 # 7 References
Reeves, P.R. & M.M. Cunneen, (2009) Biosynthesis of O-antigen chains and assembly. In: Microbial
Glycobiology: Structures, Relevance and Applications.A.P. Moran, O. Holst, P.J. Brennan & M. von Itzstein
(eds). Amsterdam: Elsevier, pp. 319-335.
Feldman, M. F., C. L. Marolda, M. A. Monteiro, M. B. Perry, A. J. Parodi, and M. A. Valvano. 1999. The activity
of a putative polyisoprenol-linked sugar translocase (Wzx) involved in Escherichia coli O antigen assembly is
independent of the chemical structure of the O repeat. J. Biol. Chem. 274:35129-38.
Hong, Y., M. M. Cunneen, and P. R. Reeves. 2012. The Wzx translocases for Salmonella enterica O-antigen
processing have unexpected serotype specificity. Mol. Microbiol. 84:620-630.
Hong, Y. & P.R. Reeves, (2014) Diversity of O-antigen repeat unit structures can account for the substantial
sequence variation of Wzx translocases. J. Bacteriol. 196:1713-1722.
Marolda, C. L., J. Vicarioli, and M. A. Valvano. 2004. Wzx proteins involved in biosynthesis of O antigen
function in association with the first sugar of the O-specific lipopolysaccharide subunit. Microbiology 150:4095-
105.
.
Conclusions