Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M501259200 on May 20, 2005

J. Biol. Chem., Vol. 280, Issue 27, 25936-25947, July 8, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
280/27/25936    most recent
M501259200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schild, S.
Right arrow Articles by Reidl, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schild, S.
Right arrow Articles by Reidl, J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Molecular and Functional Characterization of O Antigen Transfer in Vibrio cholerae*{boxs}

Stefan Schild, Anna-Karina Lamprecht, and Joachim Reidl{ddagger}

From the Institut für Hygiene und Mikrobiologie, Universität Würzburg, Josef Schneider Strasse 2, E1, Würzburg 97080, Germany

Received for publication, February 3, 2005 , and in revised form, May 19, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The majority of Gram-negative bacteria transfer O antigen polysaccharides onto the lipid A-core oligosaccharide via the action of surface polymer:lipid A-core ligases (WaaL). Here, we characterize the WaaL proteins of Vibrio cholerae with emphasis on structural and functional characterization of O antigen transfer and core oligosaccharide recognition. We demonstrate that the activity of two distantly related O antigen ligases is dependent on the presence of N-acetylglucosamine, and substitution of an additional sugar, i.e. galactose, alters the site specificity of the core oligosaccharide necessitating discriminative WaaL types. Protein topology analysis and a conserved domain search identified two distinct conserved motifs in the periplasmic domains of WaaL proteins. Site-directed mutagenesis of the two motifs, shown for WaaLs of V. cholerae and Salmonella enterica, caused a loss of O antigen transfer activity. Moreover, analogy of topology and motifs between WaaLs and O polysaccharide polymerases (Wzy) reveals a relationship between the two protein families, suggesting that the catalyzed reactions are related to each other.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The outer lipid leaflet of the outer membrane of Gram-negative bacteria (1) is composed almost exclusively of lipopolysaccharide (LPS)1 consisting of three complex portions, i.e. the lipid A, the core oligosaccharides (core OS), and the O polysaccharides. As a polyanionic lipid, the LPS leaflet acts as an effective barrier for the diffusion of lipophilic and hydrophobic compounds. This characteristic depends on the lipid A-core OS complex (for a recent summary, see Ref. 2). In pathogenic bacteria O polysaccharides serve important biological functions in disease. They protect against the host immune recognition system, complement attack, the immune response, are involved in mediating adherence to host surfaces, and they produce host mimicry (2–7).

For the Gram-negative bacterium Vibrio cholerae it is well established that about 200 serogroups exist and that the O polysaccharides play a major role in pathogenesis and epidemiology (8–12). The O polysaccharide structures of the cholera causing serogroups O1 and O139 (13, 14), as well as the underlying core OS, have been resolved (15, 16). Compared with O1, the core OS of two non-O1 isolates, H11 (17), and O22 (14, 18), revealed distinct structural differences. In addition, genetic analysis of the wav genes (for core OS biosynthesis) of various environmental isolates of V. cholerae shows distinct genetic variation (19). Thus, core OS variability does exist within V. cholerae as found in Escherichia coli (20).

Many aspects of LPS biosynthesis have been resolved, and it is clear that lipid A serves as an acceptor for core OS synthesis. Core OS synthesis is coordinately achieved by the interplay of highly specific heptosyl- and glycosyltransferases, which are associated with the inner side of the cytoplasmic membrane. After completion of synthesis, the lipid A-core OS molecules are subsequently transported to the periplasmic compartment via MsbA (for recent review, see Ref. 5). For the O polysaccharides, three different synthesis and transport systems are known, which are all based on the transfer of capped und-PP-linked O polysaccharides, delivered to the periplasm either by the Wzy, ABC transporters, or synthase-dependent pathways (5). The subsequent ligation of O polysaccharide and lipid A-core OS takes place in the periplasm (21). The key protein responsible for this reaction was identified as a surface polymer:lipid A-core ligase (O antigen ligase), encoded by the waaL gene as summarized recently (5, 22). Based on complementation studies in Enterobacteriaceae, it was suggested that the O antigen ligases (23, 24) are membrane-associated and seem to possess substrate specificity for lipid A-core OS, but they are relaxed in their recognition of the O polysaccharides (25, 26). However, since the first genetic evidence for WaaL function (27), no direct biochemical data have been produced which prove that the WaaL proteins are enzymes. Therefore, the important final step in the synthesis of the LPS molecule is still not fully understood. Using WaaL of V. cholerae and Salmonella enterica sv. Typhimurium, we investigated core OS-dependent WaaL specificity, ligase activity, and WaaL membrane topology.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Bacterial Strains, Plasmids, and Media—Except as noted otherwise, for all genetic manipulations E. coli strain XL1-blue (New England Biolabs, Inc.), Sm10{lambda}pir (28), and CC118 (29) were used; other bacterial strains and plasmids used are listed in Table I. Bacteria were grown in Luria broth (LB, BD Biosciences) at 37 °C with aeration. Antibiotics (Sigma) and other medium supplements were used in the following concentrations: 100 µg/ml streptomycin, 50 µg/ml kanamycin, and 50 or 100 µg/ml ampicillin for V. cholerae or E. coli, respectively. For TnphoA, TnlacZ mutagenesis in E. coli kanamycin was used at up to 300 µg/ml. For determining PhoA and LacZ activity 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and 5-bromo-4-chloro-3-indolyl-{beta}-D-galactopyranoside (X-gal) (40 µg/ml) were used, respectively. V. cholerae strain O1 and O139, referred to in this manuscript, are the O1 El Tor strain P27459 [GenBank] and the O139 strain MO10, respectively (see Table I).


View this table:
[in this window]
[in a new window]
  		TABLE I
Bacteria strains used in this study

 
PCR, DNA Purification, DNA Sequencing, and Southern Blot Analysis—PCR products or digested plasmid DNA was purified using the Qiaquick Gel Extraction or Qiaquick PCR Purification Kit (Qiagen). PCRs for sequencing and subcloning were carried out using the Triple Master system (Eppendorf). Automated DNA sequencing was performed with an ABI 377 using the dye terminator cycle method with AmpliTaq (Applied Biosystems). Chromosomal DNA was prepared as described by Grimberg et al. (30). Southern blot analysis was performed as described by Southern (31).

Construction of Suicide Plasmids—All insertion mutants using the suicide plasmid pGP704 were constructed in a similar manner. An internal fragment of the respective gene was amplified by PCR using oligonucleotides with SacI or XbaI restriction sites, as listed in Table II. For the construction of pGPwaaLV194, pGPwavMV194 and pGPwavLV194 chromosomal DNA of strain V194 served as template. In all other cases chromosomal DNA of strain O1 was utilized. SacI/XbaI-digested PCR fragments were ligated into pGP704 digested with SacI/XbaI and transformed into Sm10{lambda}pir. Ampr colonies were characterized by restriction analysis and PCR.


View this table:
[in this window]
[in a new window]
  		TABLE II
Plasmids used in this study

 
The suicide plasmids pKEK{Delta}wavL and pKEK{Delta}1wavL were constructed to introduce a 336-bp deletion for {Delta}wavL and 918 bp for {Delta}1wavL by generating two PCR fragments using the oligonucleotide pairs wavL-SacI-intern5' wavL-BamHI-intern5', and wavL-BamHI-intern3', wavL-XbaI-intern3' pKEK{Delta}wavL. pKEK{Delta}1wavL was constructed using the oligonucleotide pairs wavL-SacI-intern5' wavL-BamHI-intern5' and wavL-BamHI-intern1–3' wavL-XbaI-intern3'. PCR fragments were treated with BamHI and ligated together. The ligation product was amplified by PCR using oligonucleotides wavL-SacI-intern5' and wavL-XbaI-intern3', subsequently digested with SacI/XbaI, and ligated in a SacI/XbaI-digested pKEK229. After transformation of the ligation mix into Sm10{lambda}pir, ampr colonies were characterized by restriction analysis and PCR (data not shown).

Construction of Mutant Strains—The suicide plasmids were conjugated into V. cholerae O1, O139, or SV194, and integration onto the chromosome was selected by isolation of ampr/strepr colonies, as described before (32). For the construction of the deletion mutants O1{Delta}wavL and O1{Delta}1wavL sucrose selection was used to obtain amps colonies. The correct insertions or chromosomal deletions of the constructs were confirmed by PCR and Southern blot analysis (data not shown).

Construction of Expression Plasmids—For the construction of the plasmids pTrcwavL and pGEMwavL, the wavL gene was amplified by PCR using the oligonucleotides wavL-5'-EcoRI and wavL-3'-BamHI and chromosomal DNA of O1 as template. The purified PCR product was either ligated directly into pGEM-T easy (Promega) or digested with EcoRI/BamHI and ligated into an EcoRI/BamHI-digested pTrc99A. To construct pGEMwavLN the N-terminal part of the gene was amplified by PCR using the oligonucleotides wavL-5'-EcoRI and wavL-N-term-up and chromosomal DNA of O1 as template, then the purified PCR product was ligated into pGEM-T easy.

The expression plasmids pwaaLV194 and pwaaLwavMV194 were constructed by amplifying the genes using oligonucleotide pairs waaLV194-5'-SacI, waaL-V194-3'-KpnI, and wavM-V194-5'-SacI, waaLV194-3'-KpnI, using chromosomal DNA of V194 as template. The PCR products were treated with SacI/KpnI, then ligated into SacI/KpnI-digested pBAD18-Kan. Digestion of pwaaLwavMV194 with XbaI and religation resulted in the plasmid pwavMV194 lacking waaL of V194. After transformation of the ligation products into XL1-blue, ampr colonies for pTrcwavL, pGEMwavL, and pGEMwavLN or kanr colonies (in the other cases) were characterized by restriction analysis and PCR (data not shown). Complementation was observed by supplementing the media with 1 mM isopropyl-{beta}-D-thiogalactopyranoside, in the case of pTrcwavL, pGEMwavL, and pGEMwavLN, and 0.02% arabinose for pwaaLV194, pwaaLwavMV194, and pwavMV194.

To analyze the O antigen ligase activity, the hybrid plasmids of waaL, phybrid-waaLO1/V194, and phybrid-waaLV194/O1, consisting of N- and C-terminal parts of waaL derived from V194 and O1, were constructed. N-terminal portions (N-cassettes) of waaL were amplified by PCR using oligonucleotide pairs waaL-O1-5'-NheI and waaL-O1-up-NcoI in the case of O1, and oligonucleotide pairs waaL-V194-5'-NheI and waaL-V194-up-NcoI in the case of V194. C-terminal parts (C-cassettes) of waaL were amplified by PCR using oligonucleotide pairs waaL-O1-3'-KpnI and waaL-O1-down-NcoI in the case of O1, and oligonucleotide pairs waaL-V194-3'-KpnI and waaL-V194-down-NcoI in the case of V194. PCR products containing the N-terminal portions of waaL were treated with NcoI/NheI and ligated to C-terminal portions of waaL that had been digested with NcoI/KpnI, along with an NheI/KpnI-digested pBAD18-Kan in a three-body ligation reaction. The fusion construct of hybrid WaaLV194-WaaLO1 consists of the N-terminal 220 amino acids of WaaLV194 fused to 186 amino acids of the C-terminal portion of waaLO1 ligase, starting at position Val214. The connecting region of this construct also contains an additional Ala and Met. The other fusion construct of hybrid WaaLO1-WaaLV194 consists of the N-terminal 212 amino acids of WaaLO1 fused to 180 amino acids of the C-terminal portion of V194 ligase, starting at position Thr224. The connecting region of this construct harbors an additional Met and Val. After transformation of the ligation products into XL1-blue, kanr colonies were characterized by restriction analysis and PCR (data not shown). By constructing these hybrids between N- and C-cassettes of waaLs we also reconstructed waaLO1 and waaLV194 with the NcoI restriction site and subsequently showed that they were active in O antigen ligase complementation in the respective waaL mutants (data not shown).

Construction of His-tagged WavL and WavM Expression Systems— For the construction of the C-terminal His-tagged WavL and WavM plasmids pTOPOwavL and pTOPOwavM, the pCRT7 TOPO TA Expression Kit (Invitrogen) was used. Genes were amplified by PCR using the oligonucleotides wavL-topo-up and wavL-topo-down and chromosomal DNA of O1 as template in the case of wavL and wavM-topo-up and wavM-topo-down and chromosomal DNA of V194 as template in the case of wavM. Purified PCR products were ligated into pCRT7/CT-TOPO vector according to the kit manual and transformed into TOP10F' E. coli cells (Invitrogen). Ampr colonies were characterized by restriction and PCR analysis. Correct construction was verified by DNA sequencing (data not shown).

Construction of His-tagged WaaLO1—For the construction of the N-terminal His-tagged WaaLO1 plasmid pQE30waaLO1 the gene was amplified by PCR using the oligonucleotides waaL-start-KpnI and waaL-stop-KpnI and chromosomal DNA of O1 as template. The PCR product was treated with KpnI and ligated into the KpnI-digested pQE30 (Qiagen). After transformation of the ligation products into XL1-blue ampR colonies were obtained, isolated, and plasmid DNA was characterized by restriction analysis (data not shown). To decrease the expression of WaaLO1 in XL1-blue Glc (0.2%) was added into the LB medium. Correct construction was verified by sequencing (data not shown). To test the expression of WaaLO1, pQE30waaLO1 was transformed into MO10{Delta}waaL. Complementation was observed by supplementing the media with 1 mM isopropyl-{beta}-D-thiogalactopyranoside. Expression was verified by immunoblot using anti-His monoclonal antibodies (Invitrogen).

Construction of Amino Acid Exchanges in WaaLO1 and WaaLSARC6— Using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) defined amino acids were exchanged. According to the kit manual PCRs were carried out using pQE30waaLO1 as template and the oligonucleotides waaL-R186M and waaL-R186M-antisense to switch Arg186 to Met, waaL-H309A and waaL-H309A-antisense to switch His309 to Ala, waaL-H311A and waaL-H311A-antisense to switch His311 to Ala, and waaL-H309A/H311A and waaL-H309A/H311A-antisense to switch His309 and His311 to Ala, respectively. To lose the remaining intact plasmid template, the reaction mix was treated with DpnI, then the remaining PCR products were transformed into XL1-blue. Subsequently, ampr transformants were obtained, isolated, and plasmid DNA was prepared. The amino acids exchanges in waaL were verified by DNA sequencing. To test for expression and complementation of WaaLO1 -R186M, -H309A, -H311A, and -H309A/H311A, the respective plasmids pQE30waaLO1-R186M, -H309A, -H311A, and -H309A/H311A were transformed into MO10{Delta}waaL. For the construction of amino acids exchange in WaaLSARC6 for Arg208 -> Met and His321 -> Ala, we followed the same procedure as described above. There we were using plasmid pWQ322 as template and oligonucleotides waaLSal-R208M, waaLSal-R208M-antisense, and waaLSal-H321A, waaLSal-H321A-antisense for Arg208 -> Met and His321 -> A exchange (Table III), respectively. Expression was verified by immunoblot using anti-His monoclonal antibodies (Invitrogen), and LPS was prepared and analyzed as described below.


View this table:
[in this window]
[in a new window]
  		TABLE III
Oligonucleotides used in this study

 
Purification of His-tagged WavL and WavM—Recombinant Histagged WavL and WavM proteins were expressed in BL21 (DE3) pLysS as C-terminal V5 epitope/His6 fusion proteins. After transformation of the plasmids into BL21 (DE3) pLysS, the bacterial cells were grown to an absorbance at 600 nm of 0.8 at 37 °C under vigorous shaking. After decreasing the temperature to 20 °C the cultures were induced with 1 mM isopropyl-{beta}-D-thiogalactopyranoside and incubated overnight. For protein purification cells were harvested, resuspended in 1.5 ml of LEW buffer (50 mM NaH2PO4, 300 mM NaCl, pH 8) with protease inhibitor mix (Complete EDTA-free, Roche Applied Science). Cell extract was obtained by a cell shredder (FastPrep FP120, SAVANT, Holbrook, NY) using glass/plastic beads (Lysing Matrix B, Q BioGene) for shearing. Extracts were centrifuged at 5,000 rpm for 10 min at 4 °C to remove intact cells and glass/plastic beads. For purification of WavL the supernatant was loaded onto a Protino 1000 column (Macherey-Nagel, Dueren, Germany) according to the manufacturer's instructions. After washing the column twice with LEW buffer, WavL was eluted with LEW buffer (containing 250 mM imidazole). Because His-tagged WavM did not bind to the Protino 1000 column, after pelleting the cell extracts, the supernatant was dialyzed (4–6 h at 4 °C) against TMD buffer (50 mM Tris-HCl, 10 mM MgCl2, 1 mM dithiothreitol) (26). This procedure was also applied to avoid the de novo LPS synthesis capacity of the extract. The cell extract of BL21 (DE3) pLysS, used as a negative control, was prepared in the same manner. Dialyzed cell extracts and purified WavL were used for radioactive assays and analyzed by SDS-PAGE followed by Coomassie staining (33) and immunoblot procedure (34) utilizing anti-His monoclonal antibodies (Invitrogen) and horseradish peroxidase-conjugated anti-mouse as a secondary antibody (Dianova, Hamburg, Germany).



View larger version (5K):
[in this window]
[in a new window]
  		FIG. 1.
Genetic structure of the gene cluster wav of V. cholerae. Shown is the wav gene cluster of O1 and the corresponding waaL gene region encoded by environmental strain V194. Filled arrows in gray indicate wav genes that were targeted for knock-out mutation in this or previous studies (19). Black filled arrows indicate wav genes that were not targeted because they were assumed to be essential (19). Insertion mutations in wavH could not be obtained.

 
Analysis of Cell Surface Polysaccharides—Proteinase K-digested whole cell extracts were isolated as described by Hitchcock and Brown (35) and analyzed by electrophoresis on 15% SDS-polyacrylamide gels. LPS was visualized by silver staining (36). For the radioactive assay LPS was isolated in a different way. Cells from an overnight culture (50 ml) were harvested, washed, and resuspended in 1.5 ml of TMD buffer (26). Cells were lysed by a cell shredder using glass/plastic beads (see above). The extract was centrifuged in an Eppendorf tube centrifuge (5 min at 5,000 rpm and then 1 min at 13,000 rpm) to remove intact cells and glass/plastic beads. The supernatant was transferred into a fresh tube and centrifuged at high speed (TLA100 rotor, Beckmann, TL-100 ultracentrifuge, 60,000 rpm for 30 min). The pellet containing the LPS was resuspended in 500 µl of TMD buffer supplemented with 40 µg/ml proteinase K and incubated overnight at 55 °C. The next day, LPS was collected by ultracentrifugation (60,000 rpm for 30 min), washed twice, and finally resuspended in 500 µl of TMD containing protease inhibitor mix (Complete EDTA-free). Purified LPS was stored at -20 °C.

LPS Labeling Assay—Glycosyltransferase activities of WavM and WavL were investigated by a radioactive assay. The reaction mix consisted of 25 µl of LPS, 50 µl of 6 mg/ml cell extract, 8 µg/ml purified WavL, and 0.2 µCi of UDP-14C-labeled sugars (specific activity 304 mCi/mmol) (Amersham Biosciences). The reaction volume was always adjusted to the same volume (108 µl) using appropriate volumes of TMD buffer. After incubation overnight at 37 °C LPS was collected by centrifugation (30 min, 13,000 rpm) and resuspended in 50 µl of lysis buffer. Samples (25 µl) were used for electrophoresis on SDS-polyacrylamide gels (15%). Afterward, the SDS-polyacrylamide gel was dried and exposed to a film (Hyperfilm MP, Amersham Biosciences). For control, LPS samples used in the assays were loaded on the same SDS-polyacrylamide gel, and lanes were excised before drying and visualized by silver staining (36).

WaaL Topolgy—To prove the predicted transmembrane topology of WaaL of strain O1 derived from various transmembrane prediction programs (SOSUI, TOPPRED, TEMPRED, DAS, TMHMM), protein fusions of WaaL with either PhoA or LacZ were constructed using transposon mutagenesis. In the case of WaaL-PhoA fusions, pwaaLO1 was transformed into strain IS212 (37), possessing a TnphoA on phage {lambda}. Ampr colonies were grown in LB and plated on LB agar plates containing elevated kanamycin concentrations (300 µg/ml) in appropriate dilutions. Kanr colonies were pooled and used for plasmid preparation. Plasmids were then transformed into CC118 selecting for blue colonies on LB agar plates containing kanamycin, ampicillin, arabinose, and BCIP. Transposition of TnphoA into pwaaLO1 resulting in pwaaLO1::TnphoA fusions was verified by PCR using the oligonucleotides waaL-O1–5'-NheI and phoA-seq. Correct fusions between waaL and phoA were analyzed by DNA sequencing using phoA-seq. In addition, immunoblot analysis was performed with the WaaL::TnphoA hybrid fusion proteins. Whole cell extracts were prepared as follows. Overnight cultures (500 µl) were harvested and resuspended in 50 µl of Laemmli buffer, and subsequently the samples were incubated at 100 °C for 30 min. Samples were separated by SDS-PAGE (12%), and the gel was then subjected to immunoblot analysis (see above). PhoA protein bands were detected using a horseradish peroxidase-conjugated anti-PhoA antibody (BioTrend, Koeln, Germany).

In the case of WaaL-LacZ fusions, pwaaLO1 was transformed into CC313, which harbors pOxygen with TnlacZ (38). Ampr colonies were grown in LB and plated onto LB agar plates containing 300 µg/ml kanamycin in appropriate dilutions. Kanr colonies were pooled and used for plasmid preparation. Subsequently, the plasmids were transformed into CC118, selecting for blue colonies on LB agar plates containing kanamycin, ampicillin, arabinose, and X-gal. Transposition of TnlacZ into pwaaLO1, resulting in pwaaLO1::TnlacZ fusions, was verified by PCR using the oligonucleotides waaL-O1–5'-NheI and lacZ-seq. Correct fusions between waaL and lacZ were analyzed by DNA sequencing using lacZ-seq.

Alkaline Phosphatase and {beta}-Galactosidase Assays—To determine the PhoA and LacZ enzymatic activities for the WaaL-PhoA and WaaL-LacZ fusions, alkaline phosphatase and {beta}-galactosidase assays were performed as described earlier (38) using overnight cultures of CC118 harboring pwaaLO1::TnphoAs or pwaaLO1::TnlacZs grown in LB supplemented with arabinose (0.02%). The activities were expressed in Miller units: A405/(A6001 x 1 ml x 1 min).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Identification of wav Genes That Interfere with O Antigen Attachment: wavL, wavM, and waaL Mutations Result in O Antigen Deficiency—With the aim of identifying wav genes involved in the process of O antigen attachment, the wav gene cluster (Fig. 1) was mutated by insertional and deletion mutagenesis (see "Experimental Procedures"). Highly conserved and putatively essential core OS genes, i.e. waaA encoding the putative KDO-transferase, wavC, encoding the KDO-kinase, and waaC encoding the heptosyl-I-transferase (19), were excluded from the analysis. Attempts to disrupt the putative short open reading frame of wavH were not successful. LPS analysis by SDS-PAGE (Fig. 2A) revealed that in addition to mutation of waaL, mutations in wavA, encoding a putative heptosyl-III-transferase and wavL, encoding a putative glycosyltransferase, affected O antigen attachment. Mutants of wavDJ showed no O antigen deficiency, as described elsewhere (39).

For strain O1{Delta}wavL, we demonstrated that a wavL-encoding plasmid could complement O antigen attachment (Fig. 2A). To test whether wavL mutations also led to a deficient attachment of the O antigen in other V. cholerae strains, wavL knock-out insertions were generated in O139 via pGPwavL and in the environmental V. cholerae isolate V194 via pGPwavLV194. In both strains, an O antigen-deficient phenotype was observed in SDS-PAGE analysis (Fig. 2, C and D).

Computational analysis (RPS-BLAST 2.2.9. NCBI (40)) of WavL revealed homology to two different enzyme families (19). First, the N-terminal 342 amino acids are assigned to the glycosyltransferase family 1 of RfaG, and amino acids 13 to 260 to the MurG family of UDP-N-acetylglucosamine:LPS N-acetylglucosaminyltransferase. Second, the amino acids 405–543 of WavL are assigned to the family of pfam01522 polysaccharide deacetylases, and amino acids 348–544 to CDA1 predicted xylanase/chitin deacetylases. Synthesis of recombinant His-tagged WavL protein analyzed by SDS-PAGE and immunoblot demonstrated that WavL migrates as a protein with an apparent molecular size of 65 kDa (Fig. 3), which is close to the calculated size of 67 kDa.

To investigate the necessity of the putative glycosyltransferase and deacetylase activities of WavL for O antigen attachment, we constructed a V. cholerae O1 wavL mutant, which only maintained the N-terminal 176 amino acids and C-terminal 100 amino acids of wavL (see "Experimental Procedures"). This mutant, O1{Delta}1wavL, is O antigen-deficient (Fig. 2B). This phenotype was restored by full-length wavL, encoded by plasmid pGEMwavL, and also with pGEMwavLN encoding only the N-terminal part of wavL which harbors the glycosyltransferase domain.



View larger version (73K):
[in this window]
[in a new window]
  		FIG. 2.
Mutant analysis of the wav gene cluster of V. cholerae. Shown is a 16% SDS-PAGE (silver-stained) of LPS derived from wav mutants of serogroups O1, O139, and of the non-O1/O139 strain V194. Knock-out mutations of wavA, encoding a putative heptosyl-III-transferase, and wavL, encoding a putative glycosyltransferase, produced loss of O antigen attachment. Expression of plasmid-encoded wavL restored O antigen attachment, as did pGEMwavLN expressing only the N-terminal part of WavL, which harbors the glycosyltransferase domain. Mutations of wavL in the V. cholerae strains O139 and V194 also caused deficient O antigen attachment. Wt, wild type. The arrows indicate lipid A-core OS with O antigen (a) and lipid A-core OS (b).

 



View larger version (53K):
[in this window]
[in a new window]
  		FIG. 3.
In vitro reconstitution of UDP-[14C]GlcNAc labeling with purified O1 LPS and WavL protein. Shown is the reaction mix (left panel) containing UDP-[14C]GlcNAc, -[14C]Gal, or -[14C]Glc and core OS region, separated on 16% SDS-PAGE and exposed to radiosensitive x-ray film (see "Experimental Procedures"). Incorporation of the 14C label could be only observed in the presence of WavL, LPS of O1{Delta}wavL, and UDP-[14C]GlcNAc, whereas in all other combinations no incorporation could be detected. As a control for migration and quantity the middle panel shows the core OS region of the LPS samples used in the incorporation assay after silver staining of the same SDS-PAGE. The right panel shows an immunoblot of the purified WavL probed with anti-His monoclonal antibodies.

 



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 4.
In vitro reconstitution of UDP-[14C]Gal labeling with purified V194 LPS and dialyzed whole cell extracts of BL21 with or without WavM expression. Shown is the reaction mix (left panel) containing UDP-[14C]Gal, -[14C]Glc, or -[14C]GlcNAc and the core OS region, separated on SDS-PAGE (16%), and exposed to radiosensitive x-ray film (see "Experimental Procedures"). Incorporation of the 14C label could only be observed in the presence of WavM, UDP-[14C]Gal, and LPS of V194wavM, whereas in all other combinations no significant incorporation of [14C]Gal could be detected. As a control for migration and quantity the middle panels show the core OS region of the LPS samples used in the incorporation assay after silver staining of the same SDS-PAGE. The right panel shows an immunoblot of the dialyzed whole cell extracts of BL21 with or without WavM expression probed with anti-His monoclonal antibodies.

 
As we showed recently (19), a region of heterogeneity was found when the wav gene clusters of O1 and environmental non-O1 isolates were compared. For example, strain V194 contains an additional open reading frame, termed wavM, which is not found in O1 (Fig. 1). By analyzing the bioinformatic data by CAZy (carbohydrate-active enzymes server at afmb.cnrs-mrs.fr/~cazy/CAZY/index.html), WavM is assigned as a member of the glycosyltransferase family 25, specified as galactosyltransferase (41). Disruption of wavM led to a deficient O antigen attachment in strain V194, which could be complemented by plasmid-encoded wavM (see Fig. 5).

WavL and WavM Act as Distinct Specific Sugar Transferases—To identify the enzymatic activities and specificity of WavM and WavL, both genes were expressed in a His-tagged expression system (see "Experimental Procedures"). The synthesis of both gene products was monitored by immunoblot analysis (Figs. 3 and 4). Although the 65-kDa WavL His-tagged product could be purified by a standard procedure, the 32-kDa WavM His-tagged protein could only be obtained in the extract because it failed to bind to the Ni2+ matrix.

Purified WavL protein was used together with isolated LPS fractions of wild type O1, O1{Delta}wavL, and O1wavA{Delta}wavL mutants and incubated with UDP-activated [14C]GlcNAc, [14C]Glc, and [14C]Gal (see "Experimental Procedures"). The reaction samples were separated by SDS-PAGE, and the gel was exposed to x-ray film (Fig. 3). In the presence of WavL, LPS of O1{Delta}wavL, and UDP-[14C]GlcNAc, incorporation of the 14C label could be observed. This reaction resulted in the specific labeling of the lipid A-core OS portion, as can be seen by comparison with the same SDS-PAGE subjected to silver staining. Substrates such as UDP-[14C]Gal or UDP-[14C]Glc were not recognized by WavL. As shown in these assays, WavL acts specifically with UDP-GlcNAc and can therefore be considered a GlcNAc transferase. No incorporation of [14C]GlcNAc was observed in this assay using LPS of wild type O1 or O1wavA{Delta}wavL. This indicates that the appropriate acceptor site on the core OS molecule for the attachment of GlcNAc is missing in a wavA mutant encoding the putative heptosyl-III-transferase or is already occupied by GlcNAc in the wild type.



View larger version (69K):
[in this window]
[in a new window]
  		FIG. 5.
Complementation analysis with waaL and wavM expression in V194waaL and O1waaL mutants. Shown is a 16% SDS-PAGE (silver-stained) of LPS derived from O1 and V194 mutants. The experiments show that waaL complementation is strain-specific, i.e. pwaaLO1 only restores O antigen attachment in O1waaL, and pwaaLV194 in V194waaL. Only the combination of waaLV194 and wavM, encoded by pwaaLwavMV194, is able to complement O1waaL. wavM is essential for O antigen attachment in V194 but hinders O antigen attachment in O1. Wt, wild type.

 
To characterize the activity of WavM, similar experiments were performed with the cell extracts containing WavM, along with isolated LPS fractions of wild type V194, V194wavM, and incubation with UDP-activated [14C]GlcNAc, [14C]Glc, and [14C]Gal (Fig. 4). As can be observed in the presence of WavM only UDP-]14C[Gal, but not UDP-[14C]GlcNAc or UDP-[14C]Glc, allowed significant labeling of the lipid A-core OS moiety of V194wavM. Only a very low level of labeling could be obtained for [14C]Gal incorporation into wild type V194 LPS. However, no labeling was observed by using wild type O1 or various core OS mutant LPS substrates (data not shown). For strain V194, these results demonstrate that WavM acts specifically with UDP-Gal and can therefore be considered a Gal transferase.

WaaL Ligase Complementation Phenotypes—To test for the ability of WaaL from two different serogroups (O1 and non-typed strain V194) to complement WaaL function in the respective strains (Fig. 5, A and B), the waaL gene was disrupted in both strains. waaL of O1 and V194 were subcloned into expression plasmids. Complementation analysis in strain O1waaL and V194waaL with waaLO1- and waaLV194-encoding plasmids revealed complementation of each strain with its respective cognate WaaL (i.e. waaLO1 was able to complement O1waaL, and waaLV194 was able to complement V194waaL). However, no cross-complementation was observed with the non-cognate WaaL (i.e. waaLO1 failed to complement V194waaL, and waaLV194 failed to complement O1waaL). Interestingly, expression of both waaLV194 and wavM in an O1waaL mutant allowed the attachment of the O antigen, whereas expression of wavM alone was unable to complement O antigen attachment in O1waaL (data not shown). In addition, we found that expressing wavM in the wild type O1 strain even interfered with the attachment of O1 antigen by the native waaLO1, and expression of waaLO1 in the V194wavM mutant restored O antigen attachment.

Topology of WaaL—WaaL protein sequences show poor similarity among each other, even when derived from the same species (see supplemental data) (19, 26), providing no evidence for recent clonal origin. However, this class of transmembrane proteins shows predictable secondary structural features, as demonstrated by hydrophobicity blot analysis (26). So far, no experimental approach has been performed to unravel the membrane protein topology of WaaL. We decided to determine the topology of WaaLO1 using TnphoA and TnlacZ (38).

The TnphoA and TnlacZ gene fusion technology represents a general genetic approach, allowing determination of periplasmic and cytosolic domains of transmembrane proteins. The rationale is that a signal sequence-deleted PhoA, when fused correctly to the open reading frame of a target protein, is only active if the PhoA fusion junctions and therefore PhoA is located in the periplasm. To the contrary, LacZ is only active if fused to sections of the protein which are exposed to the cytoplasm (38). TnlacZ fusions to periplasmic domains are in most cases lethal for the cell because of membrane damage, and such fusions cannot be obtained. To the contrary, TnphoA fusions within cytosolic domains are possible, but alkaline phosphatase activity is less or not detectable there (38). Numerous TnphoA and TnlacZ insertions into waaL were screened for PhoA or LacZ activity. All TnlacZ and TnphoA insertions were verified by sequencing the fusion junctions, and only those with correct in-frame fusion to the waaL reading frame were included in the analysis. Each insertion was verified and quantified by measuring the LacZ and PhoA activities. With the TnlacZ insertion analysis (Table IV), fusion activities of 98–1,500-fold ratio over the background control were measured, indicating active LacZ fused to cytoplasmic orientated portions of WaaL. Also, as shown in Table IV for TnphoA insertions, 30–1,800-fold ratio PhoA activities over background were measured, and moreover immunoblot analysis allowed detection of PhoA fusions or degraded PhoA fusion products (Fig. 6B). One TnphoA fusion (Asp79), supposedly located in a periplasmic domain, was significantly less active and showed diminished expression, as observed by immunoblot analysis. We assume that this TnphoA insertion causes improper expression. Additionally, attempts were made to isolate nonactive phoA gene fusions to waaL. However, because of hot spot problems, after screening more than 90 TnphoA insertions in waaL, only two in-frame insertions (Ser116 and Phe360) could be identified with apparently reduced PhoA activity. These two TnphoA fusions were found to be inserted close to TnlacZ insertions, indicating that they occurred into cytosolic exposed domains of WaaL. An alignment of the distribution of the insertion sites to the WaaL sequence allowed the extraction of a secondary topology prediction model for WaaLO1 (Fig. 6), which fits well with a computational analysis. As a result of this analysis, the deduced WaaLO1 structure is composed of 10 transmembrane domains with five periplasmic and four cytoplasmic loops; accordingly, the N and C termini reside in the cytoplasm. Two of the periplasmic loops were rather large. One loop (amino acid position 31–86, loop I) of 55 amino acids is located at the immediate N-terminal portion of WaaL, and a second loop of 103 amino acids (amino acid position 224–327, loop IV) is found at the C-terminal portion of WaaL.


View this table:
[in this window]
[in a new window]
  		TABLE IV
Activities of waaL::TnlacZ {beta}-galactosidase and waaL::TnphoA alkaline phosphatase

 
Hybrid WaaL and Core OS Specificity—To determine further the functional domains, which are associated with substrate core OS recognition, we constructed hybrid fusion proteins between O antigen ligases WaaLO1 and WaaLV194. The fusion junction of these hybrids is located between periplasmic loops III and IV in the cytosolic loop from Ile201 to Phe215. These hybrids were expressed from inducible plasmids and tested for O antigen attachment in the waaLO1 mutant (see "Experimental Procedures"). As shown in Fig. 7, in O1waaL only the WaaLV194-WaaLO1 fusion exhibited O antigen ligase activity. Because we had no expression control of the hybrids, we focused on the hybrid WaaLV194-WaaLO1 because it was active in O1waaL, but not in V194waaL (latter data not shown). By expressing this hybrid in an O1waaLwavM+ background we could not observe O antigen attachment (Fig. 7B), indicating that the WavM-mediated modification of the core OS remains nonsupportive for ligase function for this hybrid. This is consistent with WavM expression in O1 leading to an O antigen-deficient phenotype (Fig. 5). From this experiment we conclude that ligase activity in strain O1 requires the C-terminal portion of WaaLO1 and that the hybrid WaaLV194-WaaLO1 still cannot operate on core OS modified by WavM.

Motif Analysis—No relevant motifs have been found associated with WaaL function. The mode of action and the residues of WaaL, which are involved in the O antigen transfer reaction, are unknown. To search for highly conserved amino acid residues within the quite diverse amino acid sequences of WaaLs a ClustalW analysis was performed (www.ch.embnet.org/software/ClustalW.html). This alignment analysis revealed two conserved motifs encompassing amino acid Arg186 with the motif RX3L (X for variable amino acids) in loop III, and a motif found at amino acid His311 with HX10G in loop IV (see supplemental data). In addition, a Wzy consensus sequence (pfam 04932), as defined by the conserved domain BLAST search (40), was found to align within the WaaL ligases in loop IV, including ligases of E. coli K-12, R2, and S. enterica sv. Typhimurium (see supplemental data). The HX10G motif can also be found within the Wzy consensus motif. Defined point mutations were introduced into His-tagged WaaLO1, altering Arg186 to Met and His311 to Ala. Mutated waaL gene products were expressed in an O139{Delta}waaL background rather than in O1waaL because of residual chromosomal waaL sequences remaining in O1waaL. WaaL derived from O1 and O139 are 100% identical, and WaaLO1 complements O139{Delta}waaL (42). O antigen ligase reaction with these mutants was assessed by SDS-PAGE analysis. The O antigen ligase complementation was achieved by expressing the N-terminal His-tagged waaLO1 (Fig. 8A). In contrast, the R186M mutation revealed a null mutation waaL phenotype, with no attached O antigen, and the H311A mutation showed a significantly reduced O antigen transferase activity. By observing residues close to the HX10G motif another H, at position 309, which is not conserved in all other ligases, was detected in WaaLO1. To investigate whether His309 contributes to WaaL function we conducted combined site directed mutagenesis of both His309 and His311 residues. The results show (Fig. 8A) that the single exchange H309A produces a defect similar to H311A, but a double replacement of both His residues to Ala produces a null phenotype. All four recombinant mutant forms of WaaLO1 and the control His-tagged WaaLO1 proteins were expressed correctly at comparable levels as shown in the immunoblot (Fig. 8B). To summarize, changing conserved residue Arg186 or His311 and the minor conserved His309 resulted in a loss of protein function.

As can be seen in the alignment (see supplemental data) the Arg186 and His311 are conserved amino acids and can be identified in the approximate positions, e.g. in WaaLSARC6 derived from S. enterica sv. ArizonaeIIIA (43). To challenge these residues in a remotely related ligase, the corresponding Arg208 and His321 residues of a WaaLSARC6 of S. enterica sv. ArizonaeIIIA were subjected to site-directed mutagenesis, resulting in R208M and H321A mutations. For complementation of WaaL activity we used these recombinant expressed Histagged WaaL mutant forms in strain CWG620 (SARC6waaL). Using this complementation system, we can demonstrate that the recombinant expressed WaaLSARC6 point mutants do not show any WaaL activity (Fig. 9A). Correct expression of the WaaL proteins was also demonstrated in corresponding immunoblot analysis (Fig. 9B).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
O polysaccharide attachment is a vital attribute of Gram-negative bacteria which confers protection to the organism in stressful environmental conditions. So far, no biochemical activity or conserved motifs were assigned to the O antigen ligases (5, 43). It was proposed that the O polysaccharide transfer mediated by WaaL comprises recognition and detachment of the O polysaccharide from the und-PP O polysaccharide complex as well as subsequent transfer of the O polysaccharide onto the acceptor lipid A-core OS. However, experimental evidence that corroborates this model has been lacking, and whether the O antigen ligase is a multifactorial complex is still an open question. Recently published data suggested that WaaL may not be the sole determinant for acceptor specificity (43). In this study, we have characterized some properties of O antigen ligases, which determine core OS specificity and define functional amino acid residues of the WaaL proteins, which most likely contribute to WaaL enzymatic activity.

Structure and Function of WaaL O Antigen Ligases—To define the topology of WaaL we employed a genetic approach based on LacZ and PhoA fusions to WaaLO1 ligase. Based on these results, WaaL traverses the inner membrane 10 times. Two large periplasmic domains were observed, loop I located close to the N terminus, and loop IV, located near the C terminus. Loop IV is also predicted by in silico analysis to exist in the WaaLs of other V. cholerae, E. coli and S. enterica sv. Typhimurium and sv. ArizonaeIIIA strains (26, 43). For another LPS assembly-associated transmembrane protein, Wzy of Shigella flexneri, the topology was resolved (44). Wzy type proteins encode O polysaccharide polymerase enzymes, which recognize and extend und-PP O polysaccharide-repeating units prior to the ligase reaction (5). The topology of Wzy revealed 12 transmembrane domains and a 54-amino acid periplasmic loop (44), located at a position similar to loop IV of WaaL. Thus, a comparison of both topology models suggests that WaaL and Wzy may share common structural features.



View larger version (52K):
[in this window]
[in a new window]
  		FIG. 6.
WaaL topology. A, secondary topology model of WaaL derived from O1 El Tor strain based on TnphoA and TnlacZ insertion determination and bioinformatics. The positions of reporter gene fusions to WaaL sequence of TnphoA are indicated by diamonds, and TnlacZ insertions are indicated by circles; in addition, the position number is provided. Amino acids that were targeted for site-directed mutagenesis are shown in bold and with indicated position (Arg186, His309, His311). Periplasmic loops (I–V) are indicated by bars. Detailed hybrid enzyme activities for LacZ and PhoA are provided in Table IV. B, immunoblot of alkaline phosphatase antibodies used to mark the waaL::TnphoA fusion products. An arrow indicates the degradation products, which corresponds to the apparent size of truncated PhoA protein. Accordingly, longer fusion proteins tend to decrease in stability.

 



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 7.
WaaL hybrid analysis. A, hybrid constructions and hydrophobicity plots, according to Kyte and Doolittle (50). Black and gray bars represent the WaaLO1 and WaaLV194 fragments, respectively. B, 16% SDS-PAGE silver-stained gel. The hybrid WaaLV194/O1 can be identified as an active O antigen ligase in O1waaL. This activity is not observed if WavM is also expressed in this background and demonstrates that the WaaL hybrid activity is influenced by WavM in the same way as observed for wild type (Wt) WaaLO1.

 
A conserved domain BLAST search with ligases of different bacterial species revealed two conserved motifs: RX3L and HX10G as located in periplasmic loops III and IV, respectively, for WaaLO1. A similar Arg motif (RX2L) can be found within the Wzy of S. flexneri on a corresponding periplasmic loop (44), and the HX10G motif can be found in a Wzy consensus sequence derived from a conserved BLAST search. Changing these motifs by site-directed mutagenesis, for WaaLO1 of V. cholerae and the corresponding amino acids of the distantly related WaaLSARC6 of S. enterica sv. ArizonaeIIIA, resulted in nonfunctional ligases. Hence, Arg and His represent conserved amino acid residues that are crucial for the O antigen ligase reaction. Similar functional requirements could be envisaged for Wzy because both types of enzymes require und-PP-capped O polysaccharides, and both are subsequently able to transfer O polysaccharide onto some sugar acceptors. The identification and characterization of the relevant motifs in WaaL may point to a catalytic reaction center and mode that may be deduced from already established reaction mechanisms of other glycosyltransferases, e.g. PBPb1 (45), or nucleases involved in resolving phosphodiester bonds. For ribonuclease A for instance (for overview, see Ref 46), histidine and lysine residues were characterized to represent the active site. Responsible for the activity of the ribonuclease are two histidine residues, which can act as either proton donor or proton acceptor, targeting the oxygen atom of the scissile phosphate of the ribonucleic acid chain, leading to nuclease activity. Whether or not Arg186 and His311 are acting in protonation/deprotonation on the phosphate bonds e.g. of the und-PP-linked O polysaccharide, needs further analysis. However, the motifs found in WaaL and their corresponding location in the periplasm can provide a first hint of a conserved reaction mode, which addresses the O polysaccharide transfer onto the core OS. Based on the fact that these motifs are conserved among the identified ligases, we exclude their role in core OS substrate recognition but rather claim that they are involved in the release of O antigen from the O polysaccharide transporter and transfer to the core OS molecules.



View larger version (53K):
[in this window]
[in a new window]
  		FIG. 8.
WaaLO1 point mutations and their effect on the O antigen ligase reaction. A, SDS-PAGE profile of the core OS and O antigen after complementation analysis of mutants with defined waaLO1 point mutations. Amino acid exchange R186M or H309A/H311A caused null mutant waaL phenotypes. The single exchange H309A or H311A caused a significant decrease in ligase activity. Wt, wild type. B, immunoblot probed with anti-His antibody of whole cell extracts, containing His-tagged WaaL proteins. Visible lanes show anti-His antibody-probed proteins corresponding to WaaL with an apparent size of 40 kDa. The molecular size standard (kDa, New England Biolabs, Inc.) is indicated.

 



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 9.
WaaLSARC6 point mutations and their effect on O antigen ligase reaction in S. enterica sv. ArizonaeIIIA. A, SDS-PAGE profile of the core OS and O antigen of a complementation analysis using defined point mutations of waaLSARC6. Amino acid exchanges R208M or H321A caused null mutant waaL phenotypes. Wt, wild type. B, immunoblot probed with anti-His antibody of whole cell extracts, containing His-tagged WaaL proteins. Visible lanes show anti-His antibody-probed proteins corresponding to WaaLSARC6 with an apparent size of 44 kDa. The molecular size standard (kDa, New England Biolabs, Inc.) is indicated.

 
To characterize WaaL further and to address core OS specificity, we constructed hybrid WaaL proteins, consisting of N- and C-terminal portion swaps from two different serogroups. We tested the hybrid WaaL constructs in the O1waaL background and found that the hybrid WaaLV194-WaaLO1 was able to complement for O antigen attachment in O1waaL, but not in V194waaL. Further analysis of the functional hybrid WaaLV194-WaaLO1 revealed that no O antigen attachment was observed if it was expressed in O1waaLwavM+. This is complementary with the observation that waaLO1 expressed in V194wavM mutant showed complementation. Taken together, these results point to the C-terminal portion of the ligase WaaLO1, harboring periplasmic loop IV, as being responsible for discriminating between core OS structures. In the case of WaaLO1, the C terminus is probably responsive to Gal substitution in the core OS (see below). Further details of O polysaccharide and core OS recognition by WaaL and subsequent O polysaccharide transfer reaction need to be resolved, which will not be trivial considering the difficulty in handling large transmembrane protein complexes.

Modification of Core OS and Recognition by O Antigen Ligases—We initially investigated whether other gene functions, besides WaaL, are involved in O antigen attachment, by systematically introducing knock-out mutations in the wav gene cluster of O1. This knock-out mutant series allowed the identification of one open reading frame, termed wavL, which if interrupted also produced a deficient O antigen attachment phenotype. Interestingly, WavL is present in all analyzed wav gene clusters of V. cholerae (19), indicating conserved function. WavL shows homology to two different enzyme families: the N-terminal half of the protein harbors a conserved domain corresponding to the glycosyltransferase family 1, whereas the C-terminal half of WavL shares homology with polysaccharide deacetylases. DNA sequencing and full-length protein expression confirmed wavL as a single annotated open reading frame, as reported originally (47). With the purified WavL His-tagged protein we were able to demonstrate glycosyltransferase activity and identified it as a GlcNAc transferase. According to the established core OS structure of the O1 El Tor strain (15), we can deduce that a wavL mutant is lacking a GlcNAc residue in the core OS, which was determined before to be linked 1–7 with heptosyl-III. In labeling experiments we are able to show qualitatively that purified WavL can transfer GlcNAc, but not Glc or Gal, into the core OS derived from O1wavL mutants. Furthermore, we can demonstrate that no incorporation occurs if wild type LPS is used, indicating that heptosyl-III was already replaced by GlcNAc, hence in vitro incorporation was blocked. In addition, and in agreement with the structural data (15), we suggest that heptosyl-III is the acceptor site for GlcNAc because wavA (with significant homology to heptosyl-III-transferases and effect on migration pattern) and wavL double mutant-derived LPS could not serve as an acceptor for GlcNAc incorporation. Additional mutation and complementation experiments in a {Delta}1wavL strain revealed that only the glycosyltransferase and not the putative deacetylase function is involved in O antigen attachment. Strains V194wavL and MO10wavL also have an O antigen attachment-deficient phenotype. It can be excluded that WavL contributes the linkage residue for O antigen attachment because as known for O139 strains, O antigen gets attached to the terminal heptosyl-III residue (14, 48). Therefore, we postulate that the presence of GlcNAc in the core OS constitutes a basic signature residue affecting O antigen ligase activity in V. cholerae, as shown here for the two highly diverse WaaLs of O1 and V194, which share only 23.7% identity (19). Interestingly, it has also been shown recently for E. coli R2 and some Salmonella isolates (26, 43) that a terminal GlcNAc residue is crucial for O antigen attachment, although it does not act as the acceptor site. Perhaps this amino sugar plays some conserved requirement in the O antigen transfer mechanism rather than in core OS specificity. It was reported recently by Heinrichs and colleagues (49) that, similar to V. cholerae WavL, in S. enterica sv. Typhimurium and E. coli R2 a glycosyltransferase, encoded by waaK, has an UDP-N-acetylglucosamine:(glycosyl) LPS {alpha}1,2-N-acetylglucosaminyltransferase activity, which is essential for O antigen attachment. It was proposed that the terminal GlcNAc residue may provide specificity for the WaaL O antigen ligases of E. coli R2 and S. enterica sv. Typhimurium. However, the same group showed that a waaK mutant phenotype could be restored for O antigen attachment if the corresponding WaaL was overexpressed from a plasmid (43). This finding also predicts that the GlcNAc residues may be involved in the transfer reaction rather than representing an essential attribute for ligase specificity.

The O antigen ligase encoding waaL locus of isolate V194 exhibits heterogeneity compared with the corresponding O1 locus (19). In V194, the presence of an additional glycosyltransferase-encoding gene, termed wavM, was observed (19). Neither wavM nor a homolog could be found in the published genome sequence of V. cholerae O1 El Tor strain N16961 (47). We showed that a wavM knock-out mutation caused O antigen attachment deficiency in strain V194. Therefore, in addition to the wavL knock-out phenotype we can assign two glycosyltransferases necessary to promote O antigen attachment in strain V194. In contrast to wavL, it was shown that WavM activity determines WaaL selectivity, which only allows the WaaLV194 to operate. This is supported by the following observations. First, the WaaLO1 fails to complement a V194waaL mutant, and WaaLV194 is not able to complement an O1waaL mutant. Second, expression of both waaLV194 and wavM in an O1waaL mutant results in O antigen attachment, whereas the single expression of wavM in an O1 strain interferes with O antigen attachment. Finally, expression of waaLO1 in a V194wavM mutant results in O antigen attachment. Taken together, these results suggest that indeed WavM, in contrast to WavL, is dictating WaaLV194 specificity.

WavM was characterized in further detail. According to the CAZy data base, WavM was classified to be a member of the glycosyltransferase family 25 (41). Members of this family branch were assigned to encode for {beta}-1,4-galactosyltransferases. By utilizing an in vitro assay it was demonstrated that WavM recognizes UDP-Gal as a substrate and transfers Gal into the core OS derived from the V194wavM mutant strain. In contrast, only minor Gal incorporation was observed in LPS derived from the wild type strain V194 (wavM+), indicating that most of the synthesized core OS molecules are already replaced by Gal. Also, WavM possesses no detectable Glc or GlcNAc transferase activity in these assays. Attempts to incorporate 14C-Gal into isolated O1 LPS and various O1wav mutants failed (data not shown). This may indicate that there is a difference in whether WavM is operating while de novo biosynthesis of core OS is taking place versus postsynthesis under in vitro conditions (e.g. competitive effects with other core OS enzymes).

To conclude, the presented results indicate that a GlcNAc residue attached to the core OS of O1 and V194 by WavL is a necessary requirement for both ligases, whereas the Gal residue attached by WavM in V194 represents an additional specificity determinant for WaaLV194. Future analysis will address the functional mode of WaaL and the participating structural requirements of the core OS acceptor in V. cholerae.


   FOOTNOTES
 
* This work was funded by Deutsche Forschungsgemeinschaft Grant DFG1561/2-1 (to J. R.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{boxs} The on-line version of this article (available at http://www.jbc.org) contains a supplemental figure. Back

{ddagger} To whom correspondence should be addressed. Tel.: 49-0-931-201-46159; Fax: 49-0-931-201-46445; E-mail: joachim.reidl{at}mail.uni-wuerzburg.de.

1 The abbreviations used are: LPS, lipopolysaccharide; ampr, ampicillin-resistant; BCIP, 5-bromo-4-chloro-3-indolyl phosphate; core OS, core oligosaccharide; kanr, kanamycin-resistant; LB, Luria broth; strepr, streptomycin-resistant; und-PP, undecaprenyl pyrophosphate; X-gal, 5-bromo-4-chloro-3-indolyl-{beta}-D-galactoyranoside. Back


   ACKNOWLEDGMENTS
 
We are grateful to R. Hengge, M. Herbert, and K. E. Klose for critically reading this manuscript and making suggestions for it. We thank C. Whitfield for providing plasmid pWQ322, Salmonella strains SARC6 waaL and wild type, and C. Manoil for sending plasmid pOxygen.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Kamio, Y., and Nikaido, H. (1976) Biochemistry 15,2561 -2570[CrossRef][Medline] [Order article via Infotrieve]
  2. Nikaido, H. (2003) Microb. Mol. Biol. Rev. 67,593 -656[Abstract/Free Full Text]
  3. Harvey, H. A., Swords, W. E., and Apicella, M. A. (2001) J. Autoimmun. 16, 257-262[CrossRef][Medline] [Order article via Infotrieve]
  4. Moran, A. P. (1999) J. Physiol. Pharmacol. 50,787 -805[Medline] [Order article via Infotrieve]
  5. Raetz, C. R., and Whitfield, C. (2002) Annu. Rev. Biochem. 71,635 -700[CrossRef][Medline] [Order article via Infotrieve]
  6. Paton, A. W., Voss, E., Manning, P. A., and Paton, J. C. (1998) Microb. Pathog. 24, 57-63[CrossRef][Medline] [Order article via Infotrieve]
  7. Edwards, N. J., Monteiro, M. A., Faller, G., Walsh, E. J., Moran, A. P., Roberts, I. S., and High, N. J. (2000) Mol. Microbiol. 35,1530 -1539[CrossRef][Medline] [Order article via Infotrieve]
  8. Reidl, J., and Klose, K. E. (2002) FEMS Microbiol. Rev. 26,125 -139[CrossRef][Medline] [Order article via Infotrieve]
  9. Faruque, S. M., Albert, M. J., and Mekalanos, J. J. (1998) Microbiol. Mol. Biol. Rev. 62,1301 -1314[Abstract/Free Full Text]
  10. Chiang, S. L., and Mekalanos, J. J. (1999) Infect. Immun. 67,976 -980[Abstract/Free Full Text]
  11. Waldor, M. K., Colwell, R., and Mekalanos, J. J. (1994) Proc. Natl. Acad. Sci. U. S. A. 91,11388 -11392[Abstract/Free Full Text]
  12. Iredell, J. R., Stroeher, U. H., Ward, H. M., and Manning, P. A. (1998) FEMS Immunol. Med. Microbiol. 20, 45-54[CrossRef][Medline] [Order article via Infotrieve]
  13. Kenne, L., Lindberg, B., Unger, P., Gustafsson, B., and Holme, T. (1982) Carbohydr. Res. 100,341 -349[CrossRef][Medline] [Order article via Infotrieve]
  14. Knirel, Y. A., Widmalm, G., Senchenkova, S. N., Jansson, P.-E., and Weintraub, A. (1997) Eur. J. Biochem. 247,402 -410[Medline] [Order article via Infotrieve]
  15. Vinogradov, E. V., Bock, K., Holst, O., and Brade, H. (1995) Eur. J. Biochem. 233,152 -158[Medline] [Order article via Infotrieve]
  16. Cox, A. D., and Perry, M. B. (1996) Carbohydr. Res. 290,59 -65[CrossRef][Medline] [Order article via Infotrieve]
  17. Bock, K., Vinegradov, E. V., Holst, O., and Brade, H. (1994) Eur. J. Biochem. 225,1029 -1039[Medline] [Order article via Infotrieve]
  18. Cox, A. D., Brisson, J. R., Thibault, P., and Perry, M. B. (1997) Carbohydr. Res. 304,191 -208[CrossRef][Medline] [Order article via Infotrieve]
  19. Nesper, J., Kraiss, A., Schild, S., Blass, J., Klose, K. E., Bockemühl, J., and Reidl, J. (2002) Infect. Immun. 70,2419 -2433[Abstract/Free Full Text]
  20. Amor, K., Heinrichs, D., Frirdich, E., Ziebell, K., Johnson, R., and Whitfield, C. (2000) Infect. Immun. 68,1116 -1124[Abstract/Free Full Text]
  21. Mulford, C. A., and Osborn, M. J. (1983) Proc. Natl. Acad. Sci. U. S. A. 80,1159 -1163[Abstract/Free Full Text]
  22. Yethon, J. A., and Whitfield, C. (2001) Curr. Drug Targets Infect. Disord. 1, 91-106[CrossRef][Medline] [Order article via Infotrieve]
  23. MacLachlan, P. R., Kadam, S. K., and Sanderson, K. E. (1991) J. Bacteriol. 173,7151 -7163[Abstract/Free Full Text]
  24. Schnaitman, C. A., and Klena, J. D. (1993) Microbiol. Rev. 57,655 -682[Abstract/Free Full Text]
  25. Whitfield, C., Amor, P. A., and Köplin, R. (1997) Mol. Microbiol. 23,629 -638[CrossRef][Medline] [Order article via Infotrieve]
  26. Heinrichs, D. E., Monteiro, M. A., Perry, M. B., and Whitfield, C. (1998) J. Biol. Chem. 273,8849 -8859[Abstract/Free Full Text]
  27. Schmidt, G., Jann, B., and Jann, K. (1970) Eur. J. Biochem. 16,382 -392[Medline] [Order article via Infotrieve]
  28. Miller, V. L., and Mekalanos, J. J. (1988) J. Bacteriol. 170,2575 -2583[Abstract/Free Full Text]
  29. Manoil, C., and Beckwith, J. (1985) Proc. Natl. Acad. Sci. U. S. A. 82,8129 -8133[Abstract/Free Full Text]
  30. Grimberg, J., Maguire, S., and Belluscio, L. (1989) Nucleic Acids Res. 17,8893[Free Full Text]
  31. Southern, E. M. (1975) J. Mol. Biol. 51,503 -517
  32. Nesper, J., Kapfhammer, D., Klose, K. E., Merkert, H., and Reidl, J. (2000) J. Bacteriol. 182,5097 -5104[Abstract/Free Full Text]
  33. Laemmli, U. K. (1970) Nature 227,680 -685[CrossRef][Medline] [Order article via Infotrieve]
  34. Towbin, H., Staehlin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. U. S. A. 76,4350 -4354[Abstract/Free Full Text]
  35. Hitchcock, P. J., and Brown, T. M. (1983) J. Bacteriol. 154,269 -277[Abstract/Free Full Text]
  36. Tsai, C. M., and Frasch, C. E. (1982) Anal. Biochem. 119,115 -119[CrossRef][Medline] [Order article via Infotrieve]
  37. Barondess, J., and Beckwith, J. (1990) Nature 346,871 -874[CrossRef][Medline] [Order article via Infotrieve]
  38. Manoil, C. (1991) Methods Cell Biol. 34,61 -75[Medline] [Order article via Infotrieve]
  39. Schild, S., Lamprecht, A. K., Fourestier, C., Lauriano, C. M., Klose, K. E., and Reidl, J. (2005) Int. J. Med. Microbiol., in press
  40. Marchler-Bauer, A., Anderson, J. B., DeWeese-Scott, C., Fedorova, N. D., Geer, L. Y., He, S., Hurwitz, D. I., Jackson, J. D., Jacobs, A. R., Lanczycki, C. J., Liebert, C. A., Liu, C., Madej, T., Marchler, G. H., Mazumder, R., Nikolskaya, A. N., Panchenko, A. R., Rao, B. S., Shoemaker, B. A., Simonyan, V., Song, J. S., Thiessen, P. A., Vasudevan, S., Wang, Y., Yamashita, R. A., Yin, J. J., and Bryant, S. H. (2003) Nucleic Acids Res. 31,383 -387[Abstract/Free Full Text]
  41. Coutinho, P. M., Deleury, E., Davies, G. J., and Henrissat, B. (2003) J. Mol. Biol. 328,307 -317[CrossRef][Medline] [Order article via Infotrieve]
  42. Nesper, J., Schild, S., Lauriano, C. M., Kraiss, A., Klose, K. E., and Reidl, J. (2002) Infect. Immun. 70,5990 -5996[Abstract/Free Full Text]
  43. Kaniuk, N. A., Vinogradov, E., and Whitfield, C. (2004) J. Biol. Chem. 279,36470 -36480[Abstract/Free Full Text]
  44. Daniels, C., Vindurampulle, C., and Morona, R. (1998) Mol. Microbiol. 28,1211 -1222[CrossRef][Medline] [Order article via Infotrieve]
  45. Terrak, M., Ghosh, T. K., van Heijenoort, J., Van Beeumen, J., Lampilas, M., Aszodi, J., Ayala, J. A., Ghuysen, J. M., and Nguyen-Disteche, M. (1999) Mol. Microbiol. 34, 350-364[CrossRef][Medline] [Order article via Infotrieve]
  46. Richards, F. M., and Wyckoff, H. (1971) inThe Enzymes, 3rd ed. (Boyer, P. D., ed) Vol.4 , pp. 647-806, Academic Press, New York
  47. Heidelberg, J. F., Eisen, J. A., Nelson, W. C., Clayton, R. A., Gwinn, M. L., Dodson, R. J., Haft, D. H., Hickey, E. K., Peterson, J. D., L., U., Gill, S., Nelson, K. E., Read, T. D., Tettelin, H., Richardson, D., Ermolaeva, M. D., Vamathevan, J., Bass, S., Qin, H., Dragoi, I., Sellers, P., McDonald, L., Utterback, T., Fleishman, R. D., Nierman, W. C., White, O., Salzberg, S. L., Smith, H. O., Colwell, R. R., Mekalanos, J. J., Venter, J. C., and Fraser, C. M. (2000) Nature 406,477 -483[CrossRef][Medline] [Order article via Infotrieve]
  48. Cox, A. D., Brisson, J. R., Varma, V., and Perry, M. B. (1996) Cabohydr. Res. 290, 43-58
  49. Heinrichs, D. E., Yethon, J. A., and Whitfield, C. (1998) Mol. Microbiol. 30, 221-232[CrossRef][Medline] [Order article via Infotrieve]
  50. Kyte, J., and Doolittle, R. F. (1982) J. Mol. Biol. 157,105 -132[CrossRef][Medline] [Order article via Infotrieve]
  51. Boyd, E. F., Wang, F. S., Whittam, T. S., and Selander, R. K. (1996) Appl. Environ. Microbiol. 62, 804-808[Abstract]
  52. Correa, N. E., Lauriano, C. M., McGee, R., and Klose, K. E. (2000) Mol. Microbiol. 35, 743-755[CrossRef][Medline] [Order article via Infotrieve]
  53. Guzman, L. M., Belin, D., Carson, M. J., and Beckwith, J. (1995) J. Bacteriol. 177,4121 -4130[Abstract/Free Full Text]
  54. Amann, E., Ochs, B., and Abel, K. J. (1988) Gene (Amst.) 69,301 -315[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Bacteriol.Home page
J. Zhang, W. Li, Q. Zhang, H. Wang, X. Xu, B. Diao, L. Zhang, and B. Kan
The Core Oligosaccharide and Thioredoxin of Vibrio cholerae Are Necessary for Binding and Propagation of Its Typing Phage VP3
J. Bacteriol., April 15, 2009; 191(8): 2622 - 2629.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Microbiol.Home page
Y.-J. Pan, H.-C. Fang, H.-C. Yang, T.-L. Lin, P.-F. Hsieh, F.-C. Tsai, Y. Keynan, and J.-T. Wang
Capsular Polysaccharide Synthesis Regions in Klebsiella pneumoniae Serotype K57 and a New Capsular Serotype
J. Clin. Microbiol., July 1, 2008; 46(7): 2231 - 2240.
[Abstract] [Full Text] [PDF]

Home page
 J. Bacteriol.Home page
N. Jimenez, R. Canals, M. T. Salo, S. Vilches, S. Merino, and J. M. Tomas
The Aeromonas hydrophila wb*O34 Gene Cluster: Genetics and Temperature Regulation
J. Bacteriol., June 15, 2008; 190(12): 4198 - 4209.
[Abstract] [Full Text] [PDF]

Home page
 Infect. Immun.Home page
A. Nakhamchik, C. Wilde, and D. A. Rowe-Magnus
Identification of a Wzy Polymerase Required for Group IV Capsular Polysaccharide and Lipopolysaccharide Biosynthesis in Vibrio vulnificus
Infect. Immun., December 1, 2007; 75(12): 5550 - 5558.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
280/27/25936    most recent
M501259200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schild, S.
Right arrow Articles by Reidl, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schild, S.
Right arrow Articles by Reidl, J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2005 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement