MBE Advance Access originally published online on March 30, 2007
Molecular Biology and Evolution 2007 24(6):1355-1365; doi:10.1093/molbev/msm058
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Articles |
The Yersinia kristensenii O11 O-Antigen Gene Cluster was Acquired by Lateral Gene Transfer and Incorporated at a Novel Chromosomal Locus
School of Molecular and Microbial Biosciences, University of Sydney, Sydney, New South Wales, Australia
E-mail: reeves{at}angis.org.au.
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResults and DiscussionFinal RemarksAcknowledgementsReferences |
|---|
We have sequenced the O-antigen gene clusters for the Escherichia coli O98 and Yersinia kristensenii O11 O antigens. The basic structures of these O antigens are identical, and the sequence data indicate that Y. kristensenii O11 gained its O-antigen gene cluster by lateral gene transfer (LGT). Escherichia coli O98 has a typical O-antigen gene cluster between galF and gnd as is usual in E. coli. However, the O-antigen gene cluster of Y. kristensenii O11 is not located at the traditional Yersinia O-antigen gene cluster locus, between hemH and gsk, but at a novel chromosomal locus between aroA and cmk where it is flanked by remnant galF and gnd genes that indicate the probable source of the gene cluster. Phylogenetic analysis indicated that the source was not E. coli itself but a species in the Escherichia, Salmonella, and Klebsiella group of genera. Although other O-antigen studies imply LGT on the basis of the hypervariability of the loci and GC content, this report also identifies a potential donor and provides evidence for the mechanism involved. Remnant insertion sequence (IS) sequences flank the galF and gnd remnants and suggest that LGT of the gene cluster was IS mediated.
Key Words: O-antigen gene cluster • lateral gene transfer • Escherichia coli • Yersinia kristensenii
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionFinal RemarksAcknowledgementsReferences |
|---|
O antigens are a major component of the bacterial cell wall lipopolysaccharide, a known virulence factor for mammalian infection that also acts as a protective barrier during environmental stress (Zhang et al. 1997
; Lerouge and Vanderleyden 2002). O antigens exhibit extreme structural diversity, due to the hypervariability of the O-antigen gene cluster locus, with differences in sugar composition, arrangement, and glycosyl linkages within and between the repeat units that comprise this polysaccharide; for example, 186 O antigens have been identified in Escherichia coli (including Shigella strains) (Ewing et al. 1958; Brenner 1984; Ewing 1986; Lior 1994). Most reported O-antigen structures have been found in only one species but there are some exceptions. Three examples where sequence data are available are E. coli O157, Salmonella enterica O30, and Citrobacter freundii F90; E. coli O55 and S. enterica O50; and E. coli O111 and S. enterica O35 (Lindberg et al. 1981; Kenne et al. 1983; Wang and Reeves 1998, 2000; Samuel et al. 2004). In all the 3 cases, it is most probable that the gene cluster was present in a common ancestor, although lateral gene transfer (LGT) by homologous recombination between species remains a possibility as the O-antigen gene clusters for the three species involved are all located between galF and gnd (Samuel and Reeves 2003; Samuel et al. 2004).
Unlike E. coli and S. enterica, O-antigen gene clusters in Yersinia spp. are located between hemH and gsk (Skurnik et al. 1995; Zhang et al. 1996; Reeves et al. 2003; Skurnik 2003), so the situation where O antigens are shared in E. coli and Yersinia spp. must be more complex than in the above 3 cases, which involve the same locus. The O antigens of E. coli O98 and Yersinia kristensenii O11,23 and O11,24 have the same basic structure (fig 1) with the repeat unit composed of 1 residue each of N-acetyl-D-glucosamine (D-GlcNAc) and N-acetyl-D-galacturonic acid (D-GalNAcA) and 2 residues of N-acetyl-L-quinovosamine (L-QuiNAc), but in Y. kristensenii O11,23 the D-GalNAcA residue is acetylated (Marsden et al. 1994). As O-antigen acetylation is often a post O-unit synthesis modification (Clark et al. 1991; Slauch et al. 1996; Allison and Verma 2000), we have referred to Y. kristensenii O11,23 and O11,24 as Y. kristensenii O11 except for differentiation purposes. In this paper, we report the sequence of the gene clusters responsible for O antigen synthesis in E. coli O98 and Y. kristensenii O11 to better understand O-antigen gene cluster evolution and acquisition in these 2 organisms.
|
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionFinal RemarksAcknowledgementsReferences |
|---|
Bacterial Strains
Escherichia coli H510d, the E. coli O98 type strain (Ewing 1986
) (lab number M1277), was kindly provided by the Institute of Medical and Veterinary Sciences, Adelaide, Australia. Yersinia kristensenii IP105, the O11,23 type strain (Bercovier et al. 1980) (lab number M2655), IP841 (serotype O11,24; lab number M2675), IP490 (serotype O12,25; lab number M2676), and 11047 (serotype O28; lab number M2674) were kindly provided by Prof. Roy Robins-Browne (University of Melbourne, Melbourne, Australia). Bacterial strains were grown overnight in nutrient broth at 37 °C for E. coli and 25 °C for Y. kristensenii.
Molecular Techniques
Chromosomal DNA was extracted using the Wizard chromosomal DNA purification kit (Promega, Madison, WI). Long-range polymerase chain reaction (PCR) was carried out using the Long Expand Template System according to the manufacturer (Roche Diagnostics, Dee Why, New South Wales). Short-range PCR conditions generally consisted of 94 °C for 2 min followed by 30 cycles of 94 °C for 30 s, 54 °C for 30 s, and 72 °C for 1–3 min, depending on the expected product size, and a final cycle of 72 °C for 5 min, using high fidelity VENT Taq Polymerase (New England Biolabs, Ipswich, MA). Inverse PCR was carried out using the same conditions as for short-range PCR. Primers used in this study are described in table 1. Template DNA for inverse PCR was prepared by digesting 1 µg chromosomal DNA in single restriction enzyme reactions with 10U of AvaII, EcoRI, NdeI, NcoI, or XmnI (NEB) followed by self-ligation of 4 fmol digested DNA using T4 DNA ligase (NEB). DNA fragments for cloning were end repaired with a Novagen dA-tailing kit (Merck, Kilsyth, Victoria), cloned using the pGEM-T Easy kit (Promega), and propagated in E. coli JM109.
|
Sequencing and Analysis
Plasmid DNA templates for sequencing were prepared using the Wizard Plus Miniprep DNA Purification System (Promega). The pGEM-T Easy clones, containing E. coli O98 O-antigen gene cluster fragments, were sequenced using M13F and M13R primers (Promega). PCR templates were purified using the UltraClean Purification System (Mo Bio Laboratories, Solana Beach, CA). Sequencing was carried out at the Sydney University and Prince Alfred Macromolecular Analysis Centre in Sydney, Australia, using Applied Biosciences 377 automated DNA sequencers.
Sequences were analyzed and compiled using the PHRAP/PHRED package from the University of Washington Genome Centre and CONSED (Gordon et al. 1998). Both the forward and reverse strands were sequenced to minimize PCR and sequencing errors. Sequences were submitted to ORF Finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html) and potential open reading frames (Orfs) were submitted to Blast (Altschul et al. 1990). O-antigen genes were named in accordance with the Bacterial Polysaccharide Genes Database (Reeves et al. 1996). TMHMM version 2.0 (http://www.cbs.dtu.dk/services/TMHMM-2.0/) was used for transmembrane segment (TMS) prediction (Sonnhammer et al. 1998). Sequence alignments were constructed using ClustalW (Thompson et al. 1994). Sequences were screened for programmed ribosomal frameshift sites using FSFinder (http://wilab.inha.ac.kr/FSFinder/) (Moon et al. 2004).
LPS Extraction and Analysis
Lipopolysaccharide was extracted from 10 mL overnight cultures and visualized by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and silver staining as previously described (Hitchcock and Brown 1983; Lesse et al. 1990).
Sequences
DNA sequences derived from this study have been deposited into the GenBank database under accession numbers DQ180602
[GenBank]
, DQ180603
[GenBank]
, and DQ192109
[GenBank]
.
|
Results and Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionFinal RemarksAcknowledgementsReferences |
|---|
E. coli O98 Sequence
The E. coli O98 O-antigen gene cluster was PCR amplified using primers 1523 and 1524 (table 1), and the product was digested with NcoI. The 4 fragments produced,
6, 3.5, 2.9, and 2.4 kb, were end repaired and cloned into pGEM-T Easy. Primers were designed for walking out from the sequence obtained with one clone of each sequenced from both ends. PCR with chromosomal DNA template, was used for gap filling and verification of the complete sequence. Genes were identified on the basis of Blast results (table 2). The O-antigen gene cluster, containing 11 genes, is 11,709 bp in length (fig 2).
|
|
Y. kristensenii O11 Sequences
PCR was carried out targeting the adk–hemH (primers 5383/5249) and gsk (primers 5250/5251) regions in strain IP105, and the products were purified and sequenced. Long-range PCR between hemH and gsk, with primers designed from the new sequence, was unsuccessful, so primers based on the E. coli O98 O-antigen gene cluster were then used in short-range PCR to successfully amplify and sequence the Y. kristensenii O11 wbwU–wbwW region. Short-range PCR using hemH (5385) and gsk (5301) primers in combination with primers based on the Y. kristensenii O11 sequence already obtained was unsuccessful, with the exception of the gne–gsk combination. We concluded that the Y. kristensenii O11 O-antigen gene cluster does not reside between hemH and gsk and so inverse PCR was used to complete the sequence and extend it to the adjacent genes aroA and cmk. The O-antigen gene cluster, containing 10 genes, is 11,315 bp in length (fig 2).
Nucleotide Sugar Biosynthesis and Transferase Genes
The gene clusters contain the genes for the biosynthesis of UDP-D-GalNAcA (gna and gne) and UDP-L-QuiNAc (fnlA, qnlA, and qnlB) (Belanger et al. 1999; Creuzenet et al. 2000; Zhao et al. 2000; Kowal and Wang 2002; Kneidinger et al. 2003). The remaining nucleotide sugar, UDP-D-GlcNAc, is involved in housekeeping functions within the cell and the biosynthesis genes are located elsewhere on the chromosome (Dutka-Malen et al. 1988; Mengin-Lecreulx and van Heijenoort 1994, 1996).
Three putative glycosyltransferase genes (wbwU, wbwV, and wbwW) are present, but no initial sugar phosphate transferase gene was identified (fig 1). This is probably because WecA, the initial transferase for the enterobacterial common antigen (ECA) functions as the initial GlcNAc-P transferase for both ECA and O antigen. The 1st sugar of the ECA repeat unit is GlcNAc and in E. coli it has been shown that WecA can also initiate O antigen synthesis (Meier-Dieter et al. 1992; Klena and Schnaitman 1993; Alexander and Valvano 1994; Rick et al. 1994; Kido et al. 1995). It has also been shown that WecA can act as a GalNAc-P transferase to initiate synthesis of the Yersinia enterocolitica O8 O antigen (Zhang et al. 1997). In E. coli and Yersinia spp., gene clusters for O antigens that have GlcNAc or GalNAc in the structure always lack an initial transferase gene, and WecA is generally assumed to be the initial transferase in all such cases. It seems clear that WecA transfers D-GlcNAc-P to undecaprenol phosphate to initiate E. coli O98 and Y. kristensenii O11 O antigen synthesis, and the wbwU, wbwV, and wbwW genes must encode the 3 other glycosyltransferases required. E. coli O98 and Y. kristensenii O11 WbwW proteins have 66% and 62% identity, respectively, to E. coli O26 WbuB (table 2) and similar values for other E. coli WbuB proteins. WbuB is the presumed N-acetyl-L-fucosamine (L-FucNAc) transferase in the synthesis of the E. coli O26 O-antigen which contains L-FucNAc linked to D-GlcNAc via an 
1,3 linkage (Manca et al. 1996; D'Souza et al. 2002). A similar situation occurs in the Shigella boydii 13 and Vibrio cholerae O37 O-antigen gene clusters which contain the fnlA, qnlA, and qnlB gene set with a wbuB-like gene immediately downstream; wbwH and wbrP, respectively (Li et al. 2002; Feng et al. 2004). E. coli O98 WbwW has 44% and 40% identity, respectively, to these putative L-QuiNAc transferases. L-FucNAc and L-QuiNAc are similar sugars with similar biosynthetic pathways and these homologous genes are in the same order in all the above strains including E. coli O98 and Y. kristensenii O11. WbwW is, therefore, proposed to transfer L-QuiNAc onto D-GlcNAc via an 1,3 linkage, with WbwU and WbwV responsible for the other 2 glycosidic linkages (fig 1).
Frameshift Mutation in WbwW and Potential for Programmed Ribosomal Frameshifting
The Y. kristensenii O11,23 wbwW gene has a frameshift mutation, confirmed by sequencing both forward and reverse strands of this region twice, and the wbwW region is predicted to encode 2 Orfs (Orfs10 and 11; 240 and 167 amino acids, respectively). The Y. kristensenii O11,23 and O11,24 wbwW sequences are identical (data not shown), and for both strains O-antigen expression was confirmed by SDS-PAGE and silver staining of extracted LPS samples (fig 3). The wbwW is the predicted glycosyltransferase gene for one of the L-QuiNAc residues, and must be functional for O-antigen expression to occur. The Pfam-A domain present in both Y. kristensenii and E. coli WbwW is Glycos_trans_1, which is located in the C-terminal domain of E. coli WbwW, and other glycosyltransferases that contain it (http://www.sanger.ac.uk/cgi-bin/Pfam/getacc?PF00534), and is in the 2nd Orf of Y. kristensenii WbwW. It may be that only the downstream Orf is required for function in Y. kristensenii, although, the presence of an N-terminal region domain in all other proteins containing Glycos_trans_1 suggests a functional role for the upstream Orf. Also, there is no obvious Shine Dalgarno sequence to initiate translation of the downstream Orf although translational read-through may occur.
|
It is also possible that programmed -1 ribosomal frameshifting is involved in Y. kristensenii O11 WbwW expression. Programmed ribosomal frameshifting involves the ribosome stalling during translation and slipping out of frame, by either +1 or -1 base, to express 2 overlapping orfs as a single polypeptide (Ton-Hoang et al. 1998
; Plant et al. 2004; Makelainen and Makinen 2005; Manktelow et al. 2005). Successful -1 programmed ribosomal frameshifting requires several mRNA signals; a frameshift site, a suitable spacer region and a stable pseudoknot, or hairpin loop stem secondary structure (Blinkova et al. 1997; Hammell et al. 1999; Rettberg et al. 1999; Baranov et al. 2002b; Ferracci et al. 2004). Y. kristensenii O11 wbwW contains these frameshift signals, as detected by FSFinder (Moon et al. 2004), that would enable expression as a single polypeptide (fig 4). E. coli O98 wbwW and 12 other related genes, found by Blast that are also located downstream of fnl or qnl gene sets, were analyzed by FSFinder (fig 4) in addition to the other O antigen genes in E. coli O98 and Y. kristensenii O11. Of the 12 only the E. coli O172 gene has both frameshift signals in this region, and in a further test, of the 10 O-antigen genes shared between E. coli O98 and Y. kristensenii O11, only wbwW and gna contain both a conserved frameshift site and potential secondary structure with a spacing consistent with functional frameshift sites (Dinman et al. 1991; Baranov et al. 2002a; Bekaert et al. 2003). It seems unlikely that by chance wbwW has the appropriate set of signals for programmed -1 ribosomal frameshifting at the frameshift mutation site, but experimental work is required to determine if ribosomal frameshifting does occur.
|
It will be very interesting to see if programmed ribosomal frameshifting is involved in wbwW expression or regulation and if so, it would be the first case associated with O-antigen expression.
O-Unit Processing Genes
Most E. coli and Yersinia O-antigens are synthesized by the Wzy dependent pathway (Valvano 2003), as shown by the presence of wzy and wzx genes. E. coli O98 and Y. kristensenii O11 Orf3 have 56% and 60% identity, respectively, to Vibrio vulnificus MO6-24 Wzx, whereas Orf4 has low level identity (<26%) with putative Wzy proteins (table 2). Wzx and Wzy often have low levels of similarity to the closest homologue, but usually encode 12 or 9 potential TMS, respectively (Macpherson et al. 1995; Samuel et al. 2004), and that was found to be the case here using TMHMM (Sonnhammer et al. 1998). We assigned Orf3 and Orf4 as Wzx and Wzy, respectively.
E. coli O98 and Y. kristensenii O11 Gene Cluster Relationships
The orfs in both clusters were in the same transcriptional direction as the adjacent neighboring genes, with most genes in common and in the same order (fig 2). Both gene clusters have low G + C% content compared with representative genus chromosomes and potentially indicate acquisition from a low G + C% content bacterium via LGT (Reeves 1991, 1993). However, the best evidence for LGT is in the flanking genes. The E. coli O98 O-antigen gene cluster is flanked by galF and gnd as are almost all E. coli and S. enterica O-antigen gene clusters. Remnants of these genes flank the Y. kristensenii O11 O-antigen gene cluster, but only the first half of galF, usually distal from O-antigen genes, is present and the second, usually proximal, half is missing, although the promoter region for the gene cluster is not disrupted. It appears that the Y. kristensenii O11 gene cluster was acquired from a species in which the O-antigen genes are between galF and gnd. The gnd fragment at the end of the cluster is also truncated and lacks the sequence on which our primer 1524 is based, but nonetheless the galF–gnd primer pair, 1523/1524, when used on our Y. kristensenii O11 strain, gave a PCR product showing that it had adjacent galF and gnd genes, as in other Yersinia spp. (data not shown). The O-antigen associated galF and gnd remnants are clearly additional to these. They are also more closely related to genes from E. coli and S. enterica strains than to those from Yersinia spp. The polypeptide based on the galF remnant has 61% identity to E. coli, S. enterica, and Klebsiella pneumoniae GalF, being the closest hits in a Blast search. Phylogenetic analysis demonstrates its relationships within the Enterobacteriaceae (fig 5), which show that it is related to the Escherichia, Salmonella, and Klebsiella group of genera. The common gene order indicates that the 2 gene clusters have a common ancestor, and the relationships of the galF gene suggest that it was near the Escherichia, Salmonella, and Klebsiella group of genera. The sequence divergence of the O-antigen genes and housekeeping genes is quite similar; so on that basis, the ancestral gene cluster could have been in the Yersinia, Escherichia common ancestor, but that would not explain how the Y. kristensenii gene cluster is bracketed by E. coli-like galF and gnd genes. It is also surprising that these gene clusters have such low G + C% content if they have been in the Yersinia, Escherichia lineage for as long as under either of the above hypotheses. The alternative hypothesis that they diverged in a low G + C% content species also does not explain the presence of remnant galF and gnd genes that are related to the Escherichia, Salmonella, and Klebsiella group of genera. With regard to the low G + C% content, this seems to be so common that one has to question the usual explanation of LGT from a low G + C% content species, although we are not aware of a convincing alternative explanation.
|
Insertion sequence (IS) remnants are present in Y. kristensenii O11 at the junctions of aroA and cmk, with the remnant galF and gnd sequences, respectively. The partial IS upstream encodes a protein with 54% amino acid identity to the 3' end of Pseudomonas aeruginosa transposase IstB and that downstream of the cluster encodes a protein with 55% amino acid identity to a Yersinia pestis pCD1 transposase LcrS (Perry et al. 1986
; Larbig et al. 2002). It is likely that these IS were involved in the transfer of the O antigen gene cluster to Yersinia. The galF and gnd remnants transferred with the Y. kristensenii O11 cluster may not have shared sufficient identity to the galF–gnd chromosomal region for homologous recombination to occur at this site. The aroA and cmk genes are not thought to have been transferred with the gene cluster as they are closely related to those of other Yersinia spp. and mark the site of chromosomal integration.
WbuC
Directly downstream of E. coli O98 wbwW is wbuC, which is absent in Y. kristensenii O11. The block of genes fnlA, fnlB, fnlC, wbuB, and wbuC have been found in several gene clusters for O antigens that, where known, have L-FucNAc, or a derivative, as the 2nd sugar and GlcNAc as the initiating sugar, and a similar situation exists for L-QuiNAc with fnlA, qnlA, qnlB, wbwW, and wbuC genes. The first 3 genes encode synthesis of UDP-L-FucNAc or UDP-L-QuiNAc, respectively, that have similar biosynthetic pathways with fnlB and fnlC being homologues of qnlA and qnlB, respectively. The next gene is wbwW, or homologue, encoding the transferase believed responsible for the (1-3) linkage of L-QuiNAc or L-FucNac to GlcNAc. In E. coli there is usually an additional gene wbuC, that has no known function, but its conservation implies that it must have a function. It is interesting that E. coli O98 resembles other E. coli in having wbuC but it is absent in Y. kristensenii O11. It was presumably transferred to Y. kristensenii as part of the segment from galF to gnd, but there is no remnant present. It seems that even if imported into another species wbuC provides no useful function outside of E. coli.
A Related V. vulnificus Gene Cluster
Nine of the 10 genes in the Y. kristensenii O11 O-antigen gene cluster are also found in the group 1 capsule gene cluster of V. vulnificus MO6-24 (fig 2 and table 2), in the same order, but with an unrelated wzy gene in MO6-24 (GenBank accession number DQ360502) (fig 2). In V. vulnificus MO6-24 the wza, wzb, and wzc genes are upstream of gna and the wbfTUY-ugd genes are downstream of wbwW (Chatzidaki-Livanis et al. 2006).
The capsule structure of V. vulnificus MO6-24 is known (Reddy et al. 1992), and 3 of the 4 repeat unit sugar residues and 2 glycosyl linkages are shared with E. coli O98 and Y. kristensenii O11 (fig 1). The 3rd shared glycosyltransferase, WbwW, is presumably responsible for the related QuiNAc (1-3) GlcNAc and QuiNAc (1-3) GalNAc linkages in E. coli O98 and V. vulnificus MO6-24, respectively. Unfortunately the comparison does not help us allocate WbuU or WbuV to specific linkages.
The relationships between the V. vulnificus MO6-24 gene cluster and those sequenced in this study were analyzed by using housekeeping genes taken from the V. vulnificus YJ016 genome (GenBank accession number BA000037). The DNA identity levels observed suggest that E. coli O98 and Y. kristensenii O11 O-antigen genes are more closely related than either is to those in V. vulnificus MO6-24, and housekeeping gene divergence indicated a similar genera relationship (table 3). Furthermore, different wzy genes were identified that correlate with the different repeat unit linkages observed.
|
The V. vulnificus MO6-24 capsule and E. coli O98 and Y. kristensenii O11 O-antigen gene clusters are similar and share a conserved gene order. However, given the lower levels of DNA identity, additional capsule genes and a different wzy gene, the evolution of the V. vulnificus MO6-24 gene cluster is more complex with a distinct evolutionary history different from those of E. coli O98 and Y. kristensenii O11 although a common gene block is conserved.
Polysaccharide Gene Cluster Loci of E. coli and Yersinia spp.
Yersinia Wzy dependent O-antigen gene clusters generally map between hemH and gsk. This applies to 11 Y. pseudotuberculosis, Y. enterocolitica O8, and Y. aldovae O-antigens (GenBank accession number AJ871364) and also the Y. enterocolitica O3 and O9 outer core gene clusters that are effectively single O-units (Skurnik et al. 1995, 2000; Zhang et al. 1996; Reeves et al. 2003; Skurnik 2003). Analysis of a Y. mollaretii genome showed that putative O-antigen genes, including wzx and wzy, are also present at this locus (GenBank accession number AALD00000000), with an extensive 43-kb gene cluster that unusually also contains wzm–wzt genes. For Y. intermedia putative O antigen genes are present at this locus but no wzx–wzy or wzm–wzt genes (GenBank accession number AALF00000000).
The Y. kristensenii O11 O-antigen gene cluster is atypical in not being between hemH and gsk. As noted above there is another gne gene upstream of gsk. The Y. enterocolitica O3 outer core, Y. aldovae and Y. enterocolitica O8 O-antigen gene clusters all have gne near or adjacent to gsk and in these instances wbcQ is upstream of gne. PCR with Y. enterocolitica O3 wbcQ (5402) and gsk (5251) based primers was successful and long-range PCR, with hemH (5385) and wbcQ (5469) primers produced a 15-kb product. The 2.6-kb wbcQ–gsk PCR product was sequenced and the 2 orfs found had similarity to Y. enterocolitica O3 outer core wbcQ and gne genes in addition to gsk. The 17-kb segment between hemH and gsk, including wbcQ and gne genes, indicates the presence in Y. kristensenii O11,23 of another polysaccharide gene cluster in this region (fig 6), although structural studies (Marsden et al. 1994) did not indicate the presence of a 2nd polysaccharide structure. This suggests that the gene cluster between hemH and gsk is either nonfunctional or not expressed under laboratory conditions. It is quite likely that it is the native ancestral O-antigen gene cluster of this strain that has become redundant. In a comparable situation, E. coli Sonnei, with a plasmid-borne O-antigen gene cluster derived from Plesiomonas shigelloides, has lost its original chromosomal O-antigen gene cluster by deletion (Lai et al. 1998) presumably after gaining the plasmid-borne gene cluster.
|
It is interesting that ABC-transporter dependent O-antigen gene clusters in Yersinia spp. are usually in the galF–gnd region, for example, the Y. enterocolitica O9 gene cluster. There are also putative O-antigen gene clusters upstream of galF–gnd in Y. bercovieri and Y. frederiksenii (GenBank accession numbers NZ_AALC00000000 and NZ_AALE00000000, respectively) but not in Y. intermedia or Y. mollaretii (discussed above). Both gene clusters include wzm and wzt and thus also code for ABC-transporter dependent O antigens. The galF–gnd region is a hotspot for polysaccharide gene clusters in E. coli and close relatives with the colanic acid gene cluster upstream of galF, E. coli, and S. enterica Wzy dependent O-antigen and E. coli group 1 capsule gene clusters between galF and gnd, and the Klebsiella and E. coli (serotypes O8 and O9) ABC-transporter dependent O-antigen gene clusters downstream of gnd (Stevenson et al. 1996
; Samuel and Reeves 2003; Samuel et al. 2004; Whitfield 2006). Generally in Yersinia spp., Wzy dependent gene clusters are between hemH and gsk and ABC-transporter dependent gene clusters are upstream of galF-gnd, with only one of the sites being occupied in the available genome sequences.
Y. kristensenii O11, O12, and O28 may have related gene clusters as the O28 O antigen shares D-GalNAcA with O11 and the O12 O-antigen contains L-FucNAc, which is very similar to the L-QuiNAc of O11 (L'Vov et al. 1990, 1992; Perry and MacLean 2000). We examined the aroA-cmk region in these strains using long-range PCR, with aroA (5520) and cmk (5491) primers. This gave the expected 13-kb PCR product for Y. kristensenii O11 but revealed that for O28, O12,25, and O12,26 aroA and cmk were adjacent and thus not associated with an O-antigen gene cluster. The aroA and cmk are also adjacent in the genomes of Y. pseudotuberculosis, Y. bercovieri, Y. frederiksenii, Y. intermedia, Y. mollaretii, and Y. pestis (GenBank accession number AE009952) but separated by 1 gene, ycaL, in E. coli K12 and S. enterica LT2 (GenBank accession number AE008742). Thus there is no evidence for other O-antigen gene clusters between aroA and cmk. We conclude that the aroA–cmk region is a new O-antigen gene cluster locus, in which the presence of galF and gnd remnants allows one to see that it arose by insertion of a gene cluster from a species where these genes flank the O antigen gene cluster. This is the first reported instance of the establishment of a new O-antigen locus within a species. Yersinia spp and E. coli have extensive synteny (ENTERIX; http://globin.bx.psu.edu/enterix/enteric/enteric.html), but the Wzy dependent O-antigen gene clusters are at different loci. The situation in Y. kristensenii O11 shows how new loci can arise. Although it is not known why the Yersinia and Escherichia genera have different loci for Wzy dependent O-antigen gene clusters, this study indicates that new loci can emerge and may be mediated by IS activity during acquisition of new gene clusters in LGT events. Such events are significant as instances of O-antigen gene IS-related LGT events have been responsible, for example, for the emergence of the pathogenic V. cholerae O139 strain via a gene block transfer (Bik et al. 1995).
|
Final Remarks |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionFinal RemarksAcknowledgementsReferences |
|---|
The O-antigen gene clusters of E. coli O98 and Y. kristensenii O11 have provided information not only on the relationship between these O-antigen genes but also evidence in Y. kristensenii O11 for the establishment of a new and novel O-antigen gene cluster locus via LGT. Another gene cluster is also present in the traditional Yersinia O-antigen locus between hemH and gsk in Y. kristensenii O11 that probably relates to the ancestral O antigen gene cluster of this strain. It was not sequenced in this study as it does not appear to be expressed, but sequencing may provide information on the original O antigen of this strain.
The most probable explanation for a given species having a single locus for all or most of its O antigens of a given class is that recombination will generally lead to replacement of one O antigen with another, rather than gain of a new O antigen. There are well-documented cases where this has happened, for example, V. cholerae O139 Bengal and E. coli O157:H7 (Bik et al. 1995; Wick et al. 2005). The new O antigen perhaps confers advantages by being not affected by host immunity to the previous O antigen. This clearly applied to V. cholerae O139 Bengal, which could infect adults immune to V. cholerae O1. The benefit may be in loss of the old O antigen while retaining the benefits of a complete LPS, rather than the specific structure of the new O antigen. This concept is supported by the finding that E. coli Sonnei with its O antigen on a plasmid has a deletion inactivating the original O antigen at the galF/gnd site (Lai et al. 1998). There will be selection for such new O-antigens to translocate to the usual locus for that species, such that when it is transferred to another strain it too can replace the pre-existing O antigen. This is similar to the selection for co-location of genes in the selfish operon hypothesis (Lawrence and Roth 1996). Presumably after some such event in either the E. coli or Yersinia lineage, the new locus somehow gained ground eventually displacing the other locus completely.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionFinal RemarksAcknowledgementsReferences |
|---|
This work was supported by a grant from the Australian Research Council.
|
Footnotes |
|---|
Martin Embley, Associate Editor
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionFinal RemarksAcknowledgementsReferences |
|---|
Alexander DC, Valvano MA. Role of the rfe gene in the biosynthesis of the Escherichia coli O7-specific lipopolysaccharide and other O-specific polysaccharides containing N-acetylglucosamine. J Bacteriol (1994) 176:7079–7084.
Allison GE, Verma NK. Serotype-converting bacteriophages and O-antigen modification in Shigella flexneri. Trends Microbiol (2000) 8:17–23.[CrossRef][Web of Science][Medline]
Altschul AF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol (1990) 215:403–410.[CrossRef][Web of Science][Medline]
Baranov PV, Gesteland RF, Atkins JF. Recoding: translational bifurcations in gene expression. Gene (2002a) 286:187–201.[CrossRef][Web of Science][Medline]
Baranov PV, Gesteland RF, Atkins JF. Release factor 2 frameshifting sites in different bacteria. EMBO Rep (2002b) 3:373–377.[CrossRef][Web of Science][Medline]
Bekaert M, Bidou L, Denise A, Duchateau-Nguyen G, Forest JP, Froidevaux C, Hatin I, Rousset JP, Termier M. Towards a computational model for -1 eukaryotic frameshifting sites. Bioinformatics (2003) 19:327–335.
Belanger M, Burrows LL, Lam JS. Functional analysis of genes responsible for the synthesis of the B-band O antigen of Pseudomonas aeruginosa serotype O6 lipopolysaccharide. Microbiology (1999) 145:3505–3521.
Bercovier H, Ursing J, Brenner DJ, Sterigerwalt AG, Fanning GR, Carter GP, Mollaret HH. Yersinia kristensenii: a new species of Enterobacteriaceae composed of sucrose-negative strains (formerly called atypical Yersinia enterocolitica or Yersinia enterocolitica-like). Curr Microbiol (1980) 4:219–224.[CrossRef][Web of Science]
Bik EM, Bunschoten AE, Gouw RD, Mooi FR. Genesis of the novel epidemic Vibrio cholerae O139 strain: evidence for horizontal transfer of genes involved in polysaccharide synthesis. EMBO J (1995) 14:209–216.[Web of Science][Medline]
Blinkova A, Burkart MF, Owens TD, Walker JR. Conservation of the Escherichia coli dnaX programmed ribosomal frameshift signal in Salmonella typhimurium. J. Bacteriol (1997) 179:4438–4442.
Brenner DJ. Recommendations on recent proposals for the classification of shigellae. Int J Syst Bacteriol (1984) 34:87–88.
Chatzidaki-Livanis M, Jones MK, Wright AC. Genetic variation in the Vibrio vulnificus group 1 capsular polysaccharide operon. J Bacteriol (2006) 188:1987–1998.
Clark CA, Beltrame J, Manning PA. The oac gene encoding a lipopolysaccharide O-antigen acetylase maps adjacent to the integrase-encoding gene on the genome of Shigella flexneri bacteriophage Sf6. Gene (1991) 107:43–52.[CrossRef][Web of Science][Medline]
Creuzenet C, Belanger M, Wakarchuk WW, Lam JS. Expression, purification, and biochemical characterization of WbpP, a new UDP-GlcNAc C4 epimerase from Pseudomonas aeruginosa serotype O6. J Biol Chem (2000) 275:19060–19067.
Dinman JD, Icho T, Wickner RB. A -1 ribosomal frameshift in a double-stranded RNA virus of yeast forms a gag-pol fusion protein. Proc Natl Acad Sci USA (1991) 88:174–178.
D'Souza JM, Wang L, Reeves PR. Sequence of the Escherichia coli O26 antigen gene cluster and identification of O26 specific genes. Gene (2002) 297:123–127.[CrossRef][Web of Science][Medline]
Dutka-Malen S, Mazodier P, Badet B. Molecular cloning and overexpression of the glucosamine synthetase gene from Escherichia coli. Biochimie (1988) 70:287–290.[Medline]
Ewing WH. Edwards and Ewing's identification of the enterobacteriaceae (1986) Amsterdam (The Netherlands): Elsevier Science Publishers.
Ewing WH, Reavis RW, Davis BR. Provisional Shigella serotypes. Can J Microbiol (1958) 4:89–107.[Web of Science][Medline]
Feng L, Senchenkova SN, Yang J, et al. Structural and genetic characterization of the Shigella boydii Type 13 O antigen. J Bacteriol (2004) 186:383–392.
Ferracci F, Day JB, Ezelle HJ, Plano GV. Expression of a functional secreted YopN-TyeA hybrid protein in Yersinia pestis is the result of a +1 translational frameshift event. J Bacteriol (2004) 186:5160–5166.
Gordon D, Abajian C, Green P. CONSED—a graphical tool for sequence finishing. Genome Res (1998) 8:195–202.
Hammell AB, Taylor RC, Peltz SW, Dinman JD. Identification of putative programmed -1 ribosomal frameshift signals in large DNA databases. Genome Res (1999) 9:417–427. [see comment].
Hitchcock PJ, Brown TM. Morphological heterogeneity among Salmonella lipopolysaccharide chemotypes in silver-stained polyacrylamide gels. J Bacteriol (1983) 154:269–277.
Kenne L, Lindberg B, Söderholm E, Bundle DR, Griffith DW. Structural studies of the O-antigens from Salmonella greenside and Salmonella adelaide. Carbohydr Res (1983) 111:289–296.[CrossRef][Web of Science][Medline]
Kido N, Torgov VI, Sugiyama T, Uchiya K, Sugihara H, Komatsu T, Kato N, Jann K. Expression of the O9 polysaccharide of Escherichia coli: sequencing of the E. coli O9 rfb gene cluster, characterization of mannosyl transferases, and evidence for an ATP-binding cassette transport system. J Bacteriol (1995) 177:2178–2187.
Klena JD, Schnaitman CA. Function of the rfb gene cluster and the rfe gene in the synthesis of O antigen by Shigella dysenteriae 1. Mol Microbiol (1993) 9:393–402.[CrossRef][Web of Science][Medline]
Kneidinger B, Larocque S, Brisson JR, Cadotte N, Lam JS. Biosynthesis of 2-acetamido-2,6-dideoxy-L-hexoses in bacteria follows a pattern distinct from those of the pathways of 6-deoxy-L-hexoses. Biochem J (2003) 371:989–995.[CrossRef][Web of Science][Medline]
Kowal P, Wang PG. New UDP-GlcNAc C4 epimerase involved in the biosynthesis of 2-acetamino-2-deoxy-L-altruronic acid in the O-antigen repeating units of Plesiomonas shigelloides O17. Biochemistry (2002) 15410–15414.
Lai V, Wang L, Reeves PR. Escherichia coli clone Sonnei (Shigella sonnei) had a chromosomal O-antigen gene cluster prior to gaining its current plasmid-borne O-antigen genes. J Bacteriol (1998) 180:2983–2986.
Larbig KD, Christmann A, Johann A, Klockgether J, Hartsch T, Merkl R, Wiehlmann L, Fritz HJ, Tummler B. Gene islands integrated into tRNA(Gly) genes confer genome diversity on a Pseudomonas aeruginosa clone. J Bacteriol (2002) 184:6665–6680.
Lawrence JG, Roth JR. Selfish operons: horizontal transfer may drive the evolution of gene clusters. Genetics (1996) 143:1843–1860.[Abstract]
Lerouge I, Vanderleyden J. O-antigen structural variation: mechanisms and possible roles in animal/plant-microbe interactions. FEMS Microbiol Rev (2002) 26:17–47.[CrossRef][Web of Science][Medline]
Lesse AJ, Campagnari AA, Bittner WE, Apicella MA. Increased resolution of lipopolysaccharides and lipooligosaccharides utilizing tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis. J Immunol Methods (1990) 126:109–117.[CrossRef][Web of Science][Medline]
Li M, Shimada T, Morris JG Jr, Sulakvelidze A, Sozhamannan S. Evidence for the emergence of non-O1 and non-O139 Vibrio cholerae strains with pathogenic potential by exchange of O-antigen biosynthesis regions. Infect Immun (2002) 70:2441–2453. [erratum appears in Infect Immun. 2003; 71(1):588].
Lindberg B, Lindh F, Lönngren J, Lindberg AA, Svenson SB. Structural studies of the O-specific side-chain of the lipopolysaccharide from Escherichia coli O55. Carbohydr Res (1981) 97:105–112.[CrossRef][Web of Science][Medline]
Lior H. Classification of Escherichia coli. In: Escherichia coli in domestic animals and humans—Gyles CL, ed. (1994) Wallingford (UK): CAB International. 31–72.
L'Vov VL, Gur'ianova SV, Rodionov AV, Dmitriev BA, Shashkov AS, Ignatenko AV, Gorshkova RP, Ovodov Iu S. Struktura povtoriaiushchegosia zvena glitserofosfatsoderzhashchei O-spetsificheskoi polisakharidnoi tsepi iz lipopolisakharida Yersinia kristensenii, shtamm 103 (0:12,26). Bioorg Khim (1990) 16:379–389.[Web of Science][Medline]
L'Vov VL, Gur'yanova SV, Rodionov AV, Gorshkova RP. Structure of the repeating unit of the O-specific polysaccharide of the lipopolysaccharide of Yersinia kristensenii strain 490 (O:12,25). Carbohydr Res (1992) 228:415–422.[CrossRef][Web of Science][Medline]
Macpherson DF, Manning PA, Morona R. Genetic analysis of the rfbX gene of Shigella flexneri. Gene (1995) 155:9–17.[CrossRef][Web of Science][Medline]
Makelainen K, Makinen K. Factors affecting translation at the programmed -1 ribosomal frameshifting site of Cocksfoot mottle virus RNA in vivo. Nucleic Acids Res (2005) 33:2239–2247.
Manca MC, Weintraub A, Widmalm G. Structural studies of the Escherichia coli O26 O-antigen polysaccharide. Carbohydr. Res (1996) 281:155–160.[CrossRef][Web of Science][Medline]
Manktelow E, Shigemoto K, Brierley I. Characterization of the frameshift signal of Edr, a mammalian example of programmed -1 ribosomal frameshifting. Nucleic Acids Res (2005) 33:1553–1563.
Marsden BJ, Bundle DR, Perry MB. Serological and structural relationships between Escherichia coli O:98 and Yersinia enterocolitica O:11,23 and O:11, 24 lipopolysaccharide O-antigens. Biochem Cell Biol (1994) 72:163–168.[Web of Science][Medline]
Meier-Dieter U, Barr K, Starman R, Hatch L, Rick PD. Nucleotide sequence of the Escherichia coli rfe gene involved in the synthesis of enterobacterial common antigen. Molecular cloning of the rfe-rffi gene cluster. J Biol Chem (1992) 267:746–753.
Mengin-Lecreulx D, van Heijenoort J. Copurification of glucosamine-1-phosphate acetyltransferase and N-acetylglucosamine-1-phosphate uridyltransferase activities of Escherichia coli: characterization of the glmU gene product as a bifunctional enzyme catalyzing two subsequent steps in the pathway for UDP-N-acetylglucosamine synthesis. J Bacteriol (1994) 176:5788–5795.
Mengin-Lecreulx D, van Heijenoort J. Characterization of the essential gene glmM encoding phosphoglucosamine mutase in Escherichia coli. J Biol Chem (1996) 271:32–39.
Moon S, Byun Y, Kim HJ, Jeong S, Han K. Predicting genes expressed via -1 and +1 frameshifts. Nucleic Acids Res (2004) 32:4884–4892.
Perry RD, Harmon PA, Bowmer WS, Straley SC. A low-Ca2+ response operon encodes the V antigen of Yersinia pestis. Infect Immun (1986) 54:428–434.
Perry MB, MacLean LL. Structural identification of the lipopolysaccharide O-antigen produced by Yersinia enterocolitica serotype O:28. Eur J Biochem (2000) 267:2567–2572.[Web of Science][Medline]
Plant EP, Wang P, Jacobs JL, Dinman JD. A programmed -1 ribosomal frameshift signal can function as a cis-acting mRNA destabilizing element. Nucleic Acids Res (2004) 32:784–790.
Reddy GP, Hayat U, Abeygunawardana C, Fox C, Wright AC, Maneval DRJ, Bush CA, Morris JGJ. Purification and determination of the structure of capsular polysaccharide of Vibrio vulnificus M06-24. J Bacteriol (1992) 174:2620–2630.
Reeves P. The O antigen of Salmonella. Today's Life Sci (1991) 3:30–40.
Reeves P. Evolution of Salmonella O antigen variation by interspecific gene transfer on a large scale. Trends Genet (1993) 9:17–22.[CrossRef][Web of Science][Medline]
Reeves PR, Hobbs M, Valvano MA, et al. Bacterial polysaccharide synthesis and gene nomenclature. Trends Microbiol (1996) 4:495–503.[CrossRef][Web of Science][Medline]
Reeves PR, Pacinelli E, Wang L. O antigen gene clusters of Yersinia pseudotuberculosis. Adv Exp Med Biol (2003) 529:199–206.[Web of Science][Medline]
Rettberg CC, Prere MF, Gesteland RF, Atkins JF, Fayet O. A three-way junction and constituent stem-loops as the stimulator for programmed -1 frameshifting in bacterial insertion sequence IS911. J Mol Biol (1999) 286:1365–1378.[CrossRef][Web of Science][Medline]
Rick PD, Hubbard GL, Barr K. Role of the rfe gene in the synthesis of the O8 antigen in Escherichia coli K-12. J Bacteriol (1994) 176:2877–2884.
Samuel G, Hogbin JP, Wang L, Reeves PR. Relationships of the Escherichia coli O157, O111, and O55 O-antigen gene clusters with those of Salmonella enterica and Citrobacter freundii, which express identical O antigens. J Bacteriol (2004) 186:6536–6543.
Samuel G, Reeves P. Biosynthesis of O-antigens: genes and pathways involved in nucleotide sugar precursor synthesis and O-antigen assembly. Carbohydr Res (2003) 338:2503–2519.[CrossRef][Web of Science][Medline]
Skurnik M. Molecular genetics, biochemistry and biological role of Yersinia lipopolysaccharide. Adv Exp Med Biol (2003) 529:187–197.[Web of Science][Medline]
Skurnik M, Peippo A, Ervela E. Characterization of the O-antigen gene clusters of Yersinia pseudotuberculosis and the cryptic O-antigen gene cluster of Yersinia pestis shows that the plague bacillus is most closely related to and has evolved from Y. pseudotuberculosis serotype O:1b. Mol Microbiol (2000) 37:316–330.[CrossRef][Web of Science][Medline]
Skurnik M, Venho R, Toivanen P, Al-Hendy A. A novel locus of Yersinia enterocolitica serotype O:3 involved in lipopolysaccharide outer core biosynthesis. Mol Microbiol (1995) 17:575–594.[CrossRef][Web of Science][Medline]
Slauch JM, Lee AA, Mahan MJ, Mekalanos JJ. Molecular characterization of the oafA locus responsible for acetylation of Salmonella typhimurium O-antigen: oafA is a member of a family of integral membrane trans-acylases. J Bacteriol (1996) 178:5904–5909.
Sonnhammer EL, von Heijne G, Krogh A. A hidden Markov model for predicting transmembrane helices in protein sequences. Proc Int Conf Intell Syst Mol Biol (1998) 6:175–182.[Medline]
Stevenson G, Andrianopoulos K, Hobbs M, Reeves PR. Organization of the Escherichia coli K-12 gene cluster responsible for production of the extracellular polysaccharide colanic acid. J Bacteriol (1996) 178:4885–4893.
Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res (1994) 22:4673–4680.
Ton-Hoang B, Polard P, Chandler M. Efficient transposition of IS911 circles in vitro. EMBO J (1998) 17:1169–1181.[CrossRef][Web of Science][Medline]
Valvano MA. Export of O-specific lipopolysaccharide. Front Biosci (2003) 8:s452–s471.[Web of Science][Medline]
Wang L, Reeves PR. Organization of Escherichia coli O157 O antigen gene cluster and identification of its specific genes. Infect Immun (1998) 66:3545–3551.
Wang L, Reeves PR. The Escherichia coli O111 and Salmonella enterica O35 gene clusters: gene clusters encoding the same colitose-containing O antigen are highly conserved. J Bacteriol (2000) 182:5256–5261.
Whitfield C. Biosynthesis and assembly of capsular polysaccharides in Escherichia coli. Annu Rev Biochem (2006) 75:39–68.[CrossRef][Medline]
Wick LM, Qi WW, Lacher DW, Whittam TS. Evolution of genomic content in the stepwise emergence of Escherichia coli O157:H7. J Bacteriol (2005) 187:1783–1791.
Zhang L, Radziejewska-Lebrecht J, Krajewska-Pietrasik D, Toivanen P, Skurnik M. Molecular and chemical characterization of the lipopolysaccharide O-antigen and its role in the virulence of Yersinia enterocolitica serotype O:8. Mol Microbiol (1997) 23:63–76.[CrossRef][Web of Science][Medline]
Zhang L, Toivanen P, Skurnik M. The gene cluster directing O-antigen biosynthesis in Yersinia enterocolitica serotype O8: identification of the genes for mannose and galactose biosynthesis and the gene for the O-antigen polymerase. Microbiology (1996) 142:277–288.
Zhao X, Creuzenet C, Belanger M, Egboshima E, JL, Lam JS. WbpO, a UDP-N-acetyl-D-galactosamine dehydrogenase from Pseudomonas aeruginosa serotype O6. J Biol Chem (2000) 275:33252–33259.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  N. Mullane, P. O'Gaora, J. E. Nally, C. Iversen, P. Whyte, P. G. Wall, and S. Fanning Molecular Analysis of the Enterobacter sakazakii O-Antigen Gene Locus Appl. Envir. Microbiol., June 15, 2008; 74(12): 3783 - 3794. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Stevenson, M. Dieckelmann, and P. R. Reeves Determination of Glycosyltransferase Specificities for the Escherichia coli O111 O Antigen by a Generic Approach Appl. Envir. Microbiol., February 15, 2008; 74(4): 1294 - 1298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Iguchi, T. Ooka, Y. Ogura, Asadulghani, K. Nakayama, G. Frankel, and T. Hayashi Genomic comparison of the O-antigen biosynthesis gene clusters of Escherichia coli O55 strains belonging to three distinct lineages Microbiology, February 1, 2008; 154(2): 559 - 570. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||