Department of Microbiology and Immunology, University of British Columbia, 300-6174 University Blvd, Vancouver B.C., CanadaV6T 1Z31
Author for correspondence: John Smit. Tel: +1 604 822 4417. Fax: +1 604 822 6041. e-mail: jsmit{at}interchange.ubc.ca
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Keywords: perosamine, dideoxyaminohexose, outer membrane, glycosyltransferase
Abbreviations: S-layer, surface layer; S-LPS, smooth LPS
The GenBank accession number for the NA1000 rsaADEF and gmd, per, wbqA, wbqR, wbqX and wbqZ sequences reported in this paper is AF06235.
a Present address: School of Biological Sciences, University of Auckland, Private Bag 92019, Auckland, New Zealand.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
The outer membranes of Gram-negative bacteria contain phospholipids, proteins and LPS (Nikaido & Vaara, 1985 ). In many cases, including C. crescentus strains, there is also an extracellular polysaccharide (Ravenscroft et al., 1991
). The S-layer is external to all of these molecules (although the extracellular polysaccharide may pass through the S-layer). Smooth LPS (S-LPS) is a major component of the outer membrane of Gram-negative bacteria and consists of three regions. The lipid A moiety is anchored in the outer leaflet of the outer membrane. The core, a branched chain oligosaccharide linked to ketodeoxyoctulosonic acid (KDO), is attached to the lipid A molecule. Extending from the core, when present, is the O antigen, which contains a repeating linkage of oligosaccharides (Schnaitman & Klena, 1993
), forming S-LPS.
Although there is only a limited number of instances identified so far, it appears that in Gram-negative bacteria S-LPS is required for attachment of the S-layer (Boot & Pouwels, 1996 ; Yang et al., 1992
). This is the case in C. crescentus where the S-LPS anchors the S-layer to the cell surface (Walker et al., 1994
). A region near the N terminus of RsaA appears to be responsible (Bingle et al., 1997b
). Immunolabelling showed that the S-LPS is completely occluded by the S-layer (Walker et al., 1994
). Isolation and characterization of the S-LPS showed that the core sugars and fatty acids are identical to those of the rough LPS and that the O antigen is of a homogeneous length, unlike the variable length O antigen found in many enteric bacteria. If the S-LPS in C. crescentus is disrupted or absent, the S-layer detaches from the membrane (Walker et al., 1994
).
The precise nature of the interaction between the RsaA S-layer subunit protein and the O antigen is not known, in large part because the exact chemical composition of the oligosaccharide is not yet clarified. Reported here is a genetic approach toward that definition. A number of potential S-LPS synthesis genes were identified by transposon mutagenesis events that resulted in an S-layer shedding phenotype. These genes suggest that the C. crescentus O antigen is composed of the 4,6-dideoxy-aminohexose perosamine. A number of glycosyltransferases were also identified, suggesting that they are involved in polymerization of the O antigen.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Recombinant DNA manipulations.
Standard methods of DNA manipulation and isolation were used (Sambrook et al., 1989 ). Electroporation of C. crescentus was performed as described previously (Gilchrist & Smit, 1991
). For cloning of chromosomal DNA adjacent to Tn5 insertions, chromosomal DNA of a Tn5 mutant was digested with BamHI, SalI or XmaI. BamHI fragments were cloned directly into the BamHI site of pTZ18 vectors selecting for kanamycin resistance. The SalI and XmaI fragments were cloned using the inverse PCR method described by Martin & Mohn (1999)
. The Tn5 library has been described previously (Awram & Smit, 1998
). Aliquots of the library were stored at -70 °C.
Southern blot analysis of Tn5 insertion points.
Southern blot hybridizations were done according to the membrane manufacturers instructions (Hybond-N; Amersham). Radiolabelled probes were made by nick translation using the DNase/DNA Pol manufacturers instructions (Gibco-BRL). Chromosomal DNA was isolated as described previously (Yun et al., 1994 ). Identical Tn5 insertions were determined by cutting chromosomal DNA from the Tn5 mutants with restriction enzymes that cut Tn5 (BamHI, BglII) and probing with Tn5. Identical banding patterns indicated that Tn5 had inserted in the same place in the chromosome. Linkage of Tn5 insertions was determined by cutting chromosomal DNA with enzymes that did not cut Tn5 (EcoRI, SstI) and probing with a fragment of Tn5. Identical banding patterns indicated that the Tn5 insertions were linked. To ensure that the shedding phenotype was not a result of an altered rsaA gene, the chromosomal DNA was cut with HindIII and probed with the rsaA gene. If the banding pattern differed from the wild-type pattern, the Tn5 mutant was discarded.
SDS-PAGE and Western blot analysis.
Proteins and S-LPS were isolated from C. crescentus as described previously (Walker et al., 1992 , 1994
). SDS-PAGE and Western immunoblot analysis was performed as described previously (Walker et al., 1992
). After transfer of proteins to nitrocellulose, the blots were probed with polyclonal antibody and antibody binding was visualized using goat anti-rabbit serum coupled to horseradish peroxidase and colour-forming reagents (Smit & Agabian, 1984
). Preparation of polyclonal primary antibodies against RsaA (Walker et al., 1992
) and S-LPS (Walker et al., 1994
) were described previously.
To detect C. crescentus whole cells synthesizing an S-layer, a colony immunoblot assay was used (Bingle et al., 1997b ). Briefly, cell material was transferred to nitrocellulose by pressing the membrane onto the surface of an agar plate containing bacterial colonies. The membrane was air-dried, washed in a blocking solution (3% skim milk powder, 20 mM Tris, pH 8·0, 0·9% NaCl) with vigorous agitation on a rotary shaker and then processed in the standard fashion for Western blots (Bingle et al., 1997a
).
Surface protein from C. crescentus cells was extracted using 100 mM HEPES at pH 2·0 as described previously (Walker et al., 1992 ). To compare the amounts of surface protein extracted from different mutants, equal amounts of cells growing at exponential phase were harvested and equal amounts of the protein extract were loaded on the protein gel. SDS-PAGE and Western blotting were performed according to standard procedures (Sambrook et al., 1989
).
Nucleotide sequencing and sequence analysis.
Sequencing was performed on a DNA sequencer (Applied Biosystems model 373). Sequencing of the DNA surrounding the RsaA transporter complex is described elsewhere (Awram & Smit, 1998 ; P. Awram & J. Smit, unpublished). The Tn5 insertion sites were determined by sequencing using the primer Tn5L32 (5'-AAACGGGAAAGGTTCCGTTCAGGA-3') which hybridizes 32 bp from the end of Tn5. Nucleotide and amino acid sequence data were analysed using Geneworks and MacVector software (Oxford Molecular Group) and the NCBI BLAST e-mail and web server using the BLASTX algorithm (Altschul et al., 1990
). Primers were designed with the assistance of MacVector 6.0 software (Oxford Molecular Group). Protein alignments were generated using the CLUSTAL W algorithm as implemented by the MacVector software using the default settings. The position of putative promoters was predicted using the C. crescentus promoter consensus sequence (Malakooti et al., 1995
). Preliminary sequence data of the C. crescentus genome were obtained from The Institute for Genomic Research through the website at http://www.tigr.org and were used to determine the ORFs interrupted by the Tn5 insertions.
![]() |
RESULTS AND DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
The S-LPS synthesis genes were genetically linked to the RsaA transport genes
Analysis of the DNA sequence around the rsaA transporter complex (Awram & Smit, 1998 ; P. Awram & J. Smit, unpublished) had previously revealed five ORFs with coding sequences having significant similarity to enzymes expected to be involved with perosamine or S-LPS synthesis. Most occurred between rsaE and rsaF; one ORF was found 3' of rsaF (Fig. 2
). Immediately 3' of rsaE is an ORF which encoded a protein with similarity to GDP-D-mannose dehydratase (Currie et al., 1995
; Stroeher et al., 1995
), the second ORF encoded a protein with similarity to UDP-N-acetylglucosamine acyltransferases (Canter Cremers et al., 1989
; Vuorio et al., 1994
) and the third protein had similarity to perosamine synthetase (Bik et al., 1996
; Stroeher et al., 1995
). The fourth and fifth proteins had similarities to mannosyltransferases (Drummelsmith & Whitfield, 1999
; Rocchetta et al., 1998
). These five ORFs have been designated gmd, wbqR, per, wbqZ and wbqY, respectively (Fig. 2
). Immediately 3' of wbqY, the gene for outer-membrane component of the RsaA transporter, rsaF, was found (P. Awram & J. Smit, unpublished), but further sequencing revealed another ORF, designated wbqA, which had similarity to glycosyltransferases (Kido et al., 1998
). Further 3' of wbqA an ORF was found with no similarity to known genes and no other possible LPS-related genes were found in this 3' region.
|
Analysis and proposed function of individual proteins involved in S-LPS synthesis
A total of 14 ORFs associated with the formation of the S-LPS were found (Table 1). Another ORF interrupted by two Tn5s, F10 and F22, was found, but has no similarity to known genes. All of the putative genes start with an ATG codon except manC and wbqL, where sequence similarity and codon preference indicate that the GTG is the most probable start codon. Using the C. crescentus promoter consensus for biosynthetic genes (Malakooti et al., 1995
), promoters were predicted with transcript start sites 31 and 99 bp 5' of manB, 52 bp 5' of manC, 204 bp 5' of lpsI, 154 bp 5' of wbqP and 63 bp 5' of wbqV.
Gmd resembles GDP-mannose 4,6-dehydratases
The start codon for gmd is 143 bp 3' of rsaE (Fig. 2). No promoter matching the consensus sequence was found upstream of gmd, as would be expected if there is a terminator after rsaE as predicted previously (Awram & Smit, 1998
). The Gmd sequence has high similarity over its entire length to GDP-mannose 4,6-dehydratases from Pseudomonas aeruginosa and E. coli. (Table 1
). These homologous enzymes convert GDP-mannose to GDP-4-keto-6-deoxymannose (Stevenson et al., 1996
) as part of polysaccharide biosynthetic pathways.
WbqR is similar to N-acetyltransferases
The gene wbqR follows gmd by 2 bp, suggesting that these genes are transcriptionally coupled. The protein encoded by the gene shows significant similarity to WlaI from Campylobacter jejuni and NeuD from E. coli (Table 1). WlaI is involved in the synthesis of the LPS O antigen (Fry et al., 1998
) while the function of NeuD is not clear, but is thought to be involved in N-acetylated neuraminic acid transfer (Annunziato et al., 1995
). These proteins also show some similarity to the LpxA genes from E. coli and Salmonella enterica. The LpxA proteins are UDP-N-acetylglucosamine O-acetyltransferases that are involved in the first step of Lipid A biosynthesis and have 2426 unique hexapeptide motifs, starting with an isoleucine, leucine or valine residue often followed by a glycine (Vaara, 1992
; Vuorio et al., 1994
). WbqR, WlaI and NeuD also contain several of these hexapeptide repeats. The protein WbdR from E. coli O157 also contains these hexapeptide repeats and has 72·2% sequence similarity to WbqR. WbdR is thought to encode an N-acetyltransferase which converts GDP-perosamine to GDP-N-acetylperosamine (Wang & Reeves, 1998
). Since the data presented here suggest that the genes involved in perosamine synthesis in E. coli O157 are also present in C. crescentus it is postulated that WbqR may acetylate GDP-perosamine, like WbdR. Preliminary proton NMR studies of the S-LPS have also indicated N-acetylation of the 4,6-dideoxy-4-aminohexose identified (W.-R. Abraham, unpublished).
Per appears to be a perosamine synthetase
The gene encoding Per starts 74 bp 3' of wbqR, but no promoter sequence was found between wbqR and per. Per has considerable identity over its entire length to the rfbE and per gene products that are thought to synthesize perosamine (Table 1). These proteins likely catalyse the conversion of GDP-4-keto-6-D-deoxymannose to GDP-perosamine (4-amino-4,6-dideoxymannose) in Vibrio cholerae and E. coli O157 (Stroeher et al., 1995
; Wang & Reeves, 1998
) and show similarity to two classes of pyridoxal-binding proteins involved in the synthesis of amino sugars similar to perosamine. Based on the similarity to these genes, it is likely that Per is a perosamine synthetase.
WbqZ and WbqY resemble glycosyltransferases
The gene for WbqZ followed per by 6 bp and the gene for WbqY followed wbqZ by 13 bp, suggesting that all three genes are part of a polycistron. Both WbqZ and WbqY have significant similarity to the WbaZ proteins (Table 1). These proteins also have similarity to the RfbU related proteins, but size and amino acid similarity suggest that the WbaZ-like proteins are a separate family. WbaZ is a mannosyltransferase in S. enterica (Liu et al., 1993
); as such it seems likely that WbqZ and WbqY function to link perosamine monomers (a derivative of mannose) to the O antigen with each providing a different form of linkage.
WbqA is similar to perosamine transferases
The gene for WbqA is separated from the other LPS synthesis genes by rsaF, the outer-membrane component gene required for the secretion of RsaA (P. Awram & J. Smit, unpublished), and is transcribed in the opposite orientation. WbqA, like WbqZ and WbqY, may be a mannosyltransferase, but has greater similarity to the RfbU family of mannosyltransferases. The similarity to mannosyltransferases is much less than that seen with WbqZ and WbqY, but it does have significant similarity to the C terminus of the E. coli mannosyltransferases WbdB and WbdA (Kido et al., 1998 ; Sugiyama et al., 1998
), and RfbU from V. cholerae (Wang & Reeves, 1998
). RfbU from V. cholerae is known to transfer a perosamine residue onto the growing O antigen chain. These proteins contain a signature motif that is also found in WbqA (Fig. 3
). This motif consists of the sequence EX[XF]GXXXXE[AG] with a serine preceding the motif by 35 residues (Geremia et al., 1996
; Rocchetta et al., 1998
). Again, like WbqZ and WbqY, it seems likely that WbqA acts to add perosamine residues onto the O antigen.
|
ManC may have a dual function as a phosphomannoisomerase and mannose-1-phosphate guanyltransferase
Two shedder mutants have Tn5 insertions within manC which result in loss of proper O antigen production. ManC has significant identity over its entire length to a large family of enzymes that have dual functions as a phosphomannoisomerase and mannose-1-phosphate guanyltransferase (Table 1). Both functions are required for the synthesis of perosamine (Stroeher et al., 1995
) and are predicted to also be performed by ManC in C. crescentus.
LpsI has similarity to the LacI repressor family
The Tn5 insertion in mutant F1 interrupts lpsI. Southern blot analysis indicated that this insertion was linked to the RSA locus. Mutant F1 has a phenotype different from every other shedder Tn5 insertion described here. Analysis of the S-LPS by SDS-PAGE and silver staining revealed that reduced amounts of O antigen were produced by this mutant. Analysis of LpsI indicated that the highest degree of identity was with CcpA, the catabolite control protein in Bacillus subtilis. CcpA represses carbohydrate utilization enzymes such as -amylase and acetyl coenzyme A synthetase and has a positive regulatory affect on proteins involved in excess carbon excretion such as acetate kinase (Henkin et al., 1991
). Lower sequence identity was found to a number of LacI repressor-like proteins (Table 1
). Analysis of the genes adjacent to lpsI revealed the presence of C. crescentus analogues of glucokinase, 6-phosphogluconate dehydratase and glucose-6-phosphate 1-dehydrogenase enzymes involved in the EntnerDoudoroff central metabolic pathway. From the position of lpsI we hypothesize that LpsI has a regulatory effect on the synthesis of glucokinase. Interruption of LpsI by the F1 insertion may alter the expression of glucokinase, which in turn would affect perosamine synthesis, resulting in the phenotype seen in the F1 mutant (i.e. less O antigen).
WbqP is similar to galactosyltransferases
The Tn5 insertion F24 interrupts a gene with sequence similarity to several galactosyltransferases. These enzymes appear to transfer the first sugar residue (usually a galactose) to undecaprenol phosphate, the lipid precursor. RfbW is one of these enzymes and its sequence is 47·2% identical and 79·8% similar to WbqP over 144 aa. RfbW is involved in the synthesis of the perosamine homopolymer making up the O antigen of V. cholerae O1 (Fallarino et al., 1997 ), suggesting that RfbW may transfer the first perosamine to the lipid precursor. In C. crescentus, WbqP may initiate the formation of the O antigen by attaching the first sugar residue (presumably perosamine) to the undecaprenol phosphate carrier lipid.
WbqV has sequence similarity to amino sugar synthesis enzymes
The mutant F3 had an interruption in wbqV. WbqV is similar to a number of large proteins, usually larger than 600 aa. There is considerable similarity, especially in the middle of the protein, to WlaL, RfbV and WlbL from C. jejuni, V. cholerae O1 and Bordetella pertussis. These proteins contain five hydrophobic, predicted transmembrane domains in the N terminus. The central portion contains an NAD-binding site and is homologous to UDP-glucose 4-epimerases. Two motifs have been implicated in binding of NAD in these proteins, GXGXXG and GAGGSIG (Fallarino et al., 1997 ). As seen in Fig. 4
, the second motif is found in all the proteins, but the first only occurs in RfbV and WlbL, suggesting that not all members of this family contain this motif. The C-terminal 300 aa of these proteins have identity with dTDP-glucose 4,6-hydratases (Bechthold et al., 1995
; Linton et al., 1995
). These proteins are usually associated with synthesizing amino-6-deoxy and dideoxy sugars involved in LPS synthesis or extracellular polysaccharides and probably perform multiple functions accounting for the three domains. WbqV was not found linked to the other O antigen synthesis genes. This may indicate that WbqV is involved in the synthesis of a core sugar, possibly the terminal core sugar. Interruption of this gene may prevent attachment of the O antigen to the core, resulting in the observed shedding phenotype. The presence of three domains may also account for the small amount of full length O antigen that is seen. The Tn5 insertion occurs very close to the N terminus of the ORF and would only interfere with the first domain. If a truncated product were formed it may still have some function that would result in a small amount of O antigen being produced.
|
The functions of some of the Tn5-interrupted genes are still unidentified
The Tn5 insertions F22 and F10 interrupt an ORF with no identity to any known protein. But 5' of this ORF are two ORFs corresponding to an ABC-2 transporter. These transporters are known to transport extracellular polysaccharides and O antigens through the cytoplasmic membranes (Whitfield, 1995 ). Unlike the ATP binding cassette (ABC) transporters of type I protein secretion systems, the ABC and transmembrane domains reside on separate proteins. Alternatively, ABC-2 transporters are often found adjacent to genes involved in polysaccharide synthesis. This appears to be the case here as wbqL was found only two ORFs removed from the F22 and F10 Tn5 insertions and three ORFs from the ABC-2 transporter; therefore it may be that the ORF interrupted by the F10 and F22 mutants is also involved in polysaccharide synthesis. Alternatively, the Tn5 insertion may interfere with the expression of the ABC-2 transporter and cause the loss of O antigen. This may be the more likely explanation since the Tn5 insertions are 135 and 382 bp from the start codon of a 945 bp ORF. This position suggests that any protein made from the gene is less likely to have function with most of its N terminus interrupted. As the Western immunoblot showed that some O antigen was still be made (Fig. 1b
), this suggests that the Tn5 insertions are affecting transcription of the ABC-2 transporter. O antigens are elongated at either the reducing terminus or the non-reducing terminus. If synthesis of the O antigen occurs at the non-reducing terminus, the chain elongates in the cytoplasm and an ABC-2 transporter is required to transport the O antigen chain across the cytoplasmic membrane (Whitfield, 1995
). If the ABC-2 transporter upstream of the F10 and F22 insertions is involved in the transport of the O antigen, this suggests that the O antigen is elongated by polymerization at the non-reducing terminus.
The Tn5 insertion F6 interrupts an ORF which has similarity to a chemotaxis receptor (Ward et al., 1995 ). Analysis of the surrounding genes does not suggest that any are related to LPS synthesis and this Tn5 insertion may simply be unrelated to the observed phenotype.
In both of these cases, because a direct connection to LPS synthesis is not obvious, it is possible that these ORFs have nothing to do with LPS synthesis and the Tn5 insertions may not cause the shedding phenotype. Instead, a second spontaneous mutation may cause the altered phenotype.
Homologues to all the genes required for perosamine synthesis were found in C. crescentus
The S-LPS of V. cholerae and E. coli O157 contain perosamine and a pathway has been proposed for its synthesis (Stroeher et al., 1995 ; Wang & Reeves, 1998
). All the corresponding homologues have been described here. The pathway functions by converting fructose 6-phosphate to perosamine and is shown in Fig. 5
. The corresponding C. crescentus homologues are described here as ManC, ManB, GMD and Per.
|
The S-LPS of C. crescentus is a complex linked polymer
Six of the Tn5 insertions occur in putative glycosyltransferases (wbqZ, wbqY, lpsF, wbqP and wbqL) (Table 1). This suggests there may be a number of different linkages between the sugars in the O antigen (and preliminary proton NMR of purified S-LPS also indicates this). A number of Tn5 insertions were found in genes with similarity to mannosyltransferases. Since perosamine is a derivative of mannose, the transferases are probably highly similar. This has been found to be the case with the perosamine transferase RfbV from E. coli O157. The large number of glycosyltransferases found suggest that the O antigen is a complex polymer, even though potentially composed of a single sugar, and the glycosyltransferases are important factors in the proper assembly of the S-LPS molecule.
The Tn5 insertions interrupt synthesis of the S-LPS
While it has not been proved that the ORFs listed here are required for O antigen synthesis, the presence of multiple Tn5 insertions in some of the ORFs indicates that Tn5 must be responsible for causing the defective S-LPS phenotype and the interrupted ORF is very likely a gene involved in S-LPS synthesis. In a number of cases it is also unlikely that the Tn5 insertion is having a polar effect and that Tn5 is interrupting the gene directly involved in S-LPS synthesis. For example, the transcript orientation of wbqA is opposite to rsaF, suggesting that there are no polar genes to affect downstream of wbqA. Another example is that a polar effect by Tn5 insertions in wbqY would interfere with production of RsaF and stop RsaA secretion, which does not happen, showing that insertion must only be affecting wbqY.
The genes per, gmd, lbqZ and lbqY all resulted in a shedding phenotype when interrupted with Tn5 and were found closely linked to the RsaA transport genes. LPS genes have also been implicated in the proper formation of the transport complex in some type I secretion signals and it is thought that LPS is required for proper insertion of the outer-membrane protein (OMP) into the outer membrane (Wandersman & Létoffé, 1993 ). The close physical proximity of S-LPS and RsaA transport genes may be a reflection of coordinated expression, perhaps not only for optimal S-layer attachment but also for optimal assembly of the secretion apparatus.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Annunziato, P. W., Wright, L. F., Vann, W. F. & Silver, R. P. (1995). Nucleotide sequence and genetic analysis of the neuD and neuB genes in region 2 of the polysialic acid gene cluster of Escherichia coli K1. J Bacteriol 177, 312-319.[Abstract]
Awram, P. & Smit, J. (1998). The Caulobacter crescentus paracrystalline S-layer protein is secreted by an ABC transporter (type I) secretion apparatus. J Bacteriol 180, 3062-3069.
Bechthold, A., Sohng, J. K., Smith, T. M., Chu, X. & Floss, H. G. (1995). Identification of Streptomyces violaceoruber Tu22 genes involved in the biosynthesis of granaticin. Mol Gen Genet 248, 610-620.[Medline]
Bik, E. M., Bunschoten, A. E., Willems, R. J., Chang, A. C. & Mooi, F. R. (1996). Genetic organization and functional analysis of the otn DNA essential for cell-wall polysaccharide synthesis in Vibrio cholerae O139. Mol Microbiol 20, 799-811.[Medline]
Bingle, W. H., Nomellini, J. F. & Smit, J. (1997a). Linker mutagenesis of the Caulobacter crescentus S-layer protein: toward a definition of an N-terminal anchoring region and a C-terminal secretion signal and the potential for heterologous protein secretion. J Bacteriol 179, 601-611.[Abstract]
Bingle, W. H., Nomellini, J. F. & Smit, J. (1997b). Cell surface display of a Pseudomonas aeruginosa PAK pilin peptide within the paracrystalline S-layer of Caulobacter crescentus. Mol Microbiol 26, 277-288.[Medline]
Boot, H. J. & Pouwels, P. H. (1996). Expression, secretion and antigenic variation of bacterial S-layer proteins. Mol Microbiol 21, 1117-1123.[Medline]
Canter Cremers, H., Spaink, H. P., Wijfjes, A. H., Pees, E., Wijffelman, C. A., Okker, R. J. & Lugtenberg, B. J. (1989). Additional nodulation genes on the Sym plasmid of Rhizobium leguminosarum biovar viciae. Plant Mol Biol 13, 163-174.[Medline]
Currie, H. L., Lightfoot, J. & Lam, J. S. (1995). Prevalence of gca, a gene involved in synthesis of A-band common antigen polysaccharide in Pseudomonas aeruginosa. Clin Diagn Lab Immunol 2, 554-562.[Abstract]
Drummelsmith, J. & Whitfield, C. (1999). Gene products required for surface expression of the capsular form of the group 1 K antigen in Escherichia coli (O9a:K30). Mol Microbiol 31, 1321-1332.[Medline]
Fallarino, A., Mavrangelos, C., Stroeher, U. H. & Manning, P. A. (1997). Identification of additional genes required for O-antigen biosynthesis in Vibrio cholerae O1. J Bacteriol 179, 2147-2153.[Abstract]
Fisher, J. A., Smit, J. & Agabian, N. (1988). Transcriptional analysis of the major surface array gene of Caulobacter crescentus. J Bacteriol 170, 4706-4713.[Medline]
Fry, B. N., Korolik, V., ten Brinke, J. A., Pennings, M. T., Zalm, R., Teunis, B. J., Coloe, P. J. & van der Zeijst, B. A. (1998). The lipopolysaccharide biosynthesis locus of Campylobacter jejuni 81116. Microbiology 144, 2049-2061.[Abstract]
Geremia, R. A., Petroni, E. A., Ielpi, L. & Henrissat, B. (1996). Towards a classification of glycosyltransferases based on amino acid sequence similarities: prokaryotic alpha-mannosyltransferases. Biochem J 318, 1333-1338.
Gilchrist, A. & Smit, J. (1991). Transformation of freshwater and marine caulobacters by electroporation. J Bacteriol 173, 921-925.[Medline]
Henkin, T. M., Grundy, F. J., Nicholson, W. L. & Chambliss, G. H. (1991). Catabolite repression of alpha-amylase gene expression in Bacillus subtilis involves a trans-acting gene product homologous to the Escherichia coli lacI and galR repressors. Mol Microbiol 5, 575-584.[Medline]
Kido, N., Sugiyama, T., Yokochi, T., Kobayashi, H. & Okawa, Y. (1998). Synthesis of Escherichia coli O9a polysaccharide requires the participation of two domains of WbdA, a mannosyltransferase encoded within the wb* gene cluster. Mol Microbiol 27, 1213-1221.[Medline]
Linton, K. J., Jarvis, B. W. & Hutchinson, C. R. (1995). Cloning of the genes encoding thymidine diphosphoglucose 4,6-dehydratase and thymidine diphospho-4-keto-6-deoxyglucose 3,5-epimerase from the erythromycin-producing Saccharopolyspora erythraea. Gene 153, 33-40.[Medline]
Liu, D., Haase, A. M., Lindqvist, L., Lindberg, A. A. & Reeves, P. R. (1993). Glycosyl transferases of O-antigen biosynthesis in Salmonella enterica: identification and characterization of transferase genes of groups B, C2, and E1. J Bacteriol 175, 3408-3413.[Abstract]
Malakooti, J., Wang, S. P. & Ely, B. (1995). A consensus promoter sequence for Caulobacter crescentus genes involved in biosynthetic and housekeeping functions. J Bacteriol 177, 4372-4376.[Abstract]
Martin, V. J. & Mohn, W. W. (1999). An alternative inverse PCR (IPCR) method to amplify DNA sequences flanking Tn5 transposon insertions. J Microbiol Methods 35, 163-166.[Medline]
Nikaido, H. & Vaara, M. (1985). Molecular basis of bacterial outer membrane permeability. Microbiol Rev 49, 1-32.
Poindexter, J. S. (1981). The caulobacters: ubiquitous unusual bacteria. Microbiol Rev 45, 123-179.
Ravenscroft, N., Walker, S. G., Dutton, G. G. & Smit, J. (1991). Identification, isolation, and structural studies of extracellular polysaccharides produced by Caulobacter crescentus. J Bacteriol 173, 5677-5684.[Medline]
Riley, R. G. & Kolodziej, B. J. (1976). Pathway of glucose catabolism in Caulobacter crescentus. Microbios 16, 219-226.[Medline]
Rocchetta, H. L., Burrows, L. L., Pacan, J. C. & Lam, J. S. (1998). Three rhamnosyltransferases responsible for assembly of the A-band D-rhamnan polysaccharide in Pseudomonas aeruginosa: a fourth transferase, WbpL, is required for the initiation of both A-band and B-band lipopolysaccharide synthesis. Mol Microbiol 28, 11031119 (erratum 30, 1131).[Medline]
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Schnaitman, C. A. & Klena, J. D. (1993). Genetics of lipopolysaccharide biosynthesis in enteric bacteria. Microbiol Rev 57, 655-682.[Abstract]
Smit, J. & Agabian, N. (1984). Cloning of the major protein of the Caulobacter crescentus periodic surface layer: detection and characterization of the cloned peptide by protein expression assays. J Bacteriol 160, 1137-1145.[Medline]
Smit, J., Grano, D. A., Glaeser, R. M. & Agabian, N. (1981). Periodic surface array in Caulobacter crescentus: fine structure and chemical analysis. J Bacteriol 146, 1135-1150.[Medline]
Stevenson, G., Andrianopoulos, K., Hobbs, M. & Reeves, P. R. (1996). Organization of the Escherichia coli K-12 gene cluster responsible for production of the extracellular polysaccharide colanic acid. J Bacteriol 178, 4885-4893.[Abstract]
Stroeher, U. H., Karageorgos, L. E., Brown, M. H., Morona, R. & Manning, P. A. (1995). A putative pathway for perosamine biosynthesis is the first function encoded within the rfb region of Vibrio cholerae O1. Gene 166, 33-42.[Medline]
Sugiyama, T., Kido, N., Kato, Y., Koide, N., Yoshida, T. & Yokochi, T. (1998). Generation of Escherichia coli O9a serotype, a subtype of E. coli O9, by transfer of the wb* gene cluster of Klebsiella O3 into E. coli via recombination. J Bacteriol 180, 2775-2778.
Vaara, M. (1992). Eight bacterial proteins, including UDP-N-acetylglucosamine acyltransferase (LpxA) and three other transferases of Escherichia coli, consist of a six-residue periodicity theme. FEMS Microbiol Lett 76, 249-254.[Medline]
Vuorio, R., Harkonen, T., Tolvanen, M. & Vaara, M. (1994). The novel hexapeptide motif found in the acyltransferases LpxA and LpxD of lipid A biosynthesis is conserved in various bacteria. FEBS Lett 337, 289-292.[Medline]
Walker, S. G., Smith, S. H. & Smit, J. (1992). Isolation and comparison of the paracrystalline surface layer proteins of freshwater caulobacters. J Bacteriol 174, 1783-1792.[Abstract]
Walker, S. G., Karunaratne, D. N., Ravenscroft, N. & Smit, J. (1994). Characterization of mutants of Caulobacter crescentus defective in surface attachment of the paracrystalline surface layer. J Bacteriol 176, 6312-6323.[Abstract]
Wandersman, C. & Létoffé, S. (1993). Involvement of lipopolysaccharide in the secretion of Escherichia coli alpha-haemolysin and Erwinia chrysanthemi proteases. Mol Microbiol 7, 141-150.[Medline]
Wang, L. & Reeves, P. R. (1998). Organization of Escherichia coli O157 O antigen gene cluster and identification of its specific genes. Infect Immun 66, 3545-3551.
Ward, M. J., Bell, A. W., Hamblin, P. A., Packer, H. L. & Armitage, J. P. (1995). Identification of a chemotaxis operon with two cheY genes in Rhodobacter sphaeroides. Mol Microbiol 17, 357-366.[Medline]
Whitfield, C. (1995). Biosynthesis of lipopolysaccharide O antigens. Trends Microbiol 3, 178-185.[Medline]
Yang, L. Y., Pei, Z. H., Fujimoto, S. & Blaser, M. J. (1992). Reattachment of surface array proteins to Campylobacter fetus cells. J Bacteriol 174, 1258-1267.[Abstract]
Yun, C., Ely, B. & Smit, J. (1994). Identification of genes affecting production of the adhesive holdfast of a marine caulobacter. J Bacteriol 176, 796-803.[Abstract]
Received 16 October 2000;
revised 28 February 2001;
accepted 7 March 2001.