Skip to main content
NCBI home page
Applied and Environmental Microbiology logoLink to Applied and Environmental Microbiology
. 2010 Jun 25;76(16):5471–5478. doi: 10.1128/AEM.02946-09

Molecular and Genetic Analyses of the Putative Proteus O Antigen Gene Locus▿ †

Quan Wang 1,2,3, Agnieszka Torzewska 4, Xiaojuan Ruan 1,2,3, Xiaoting Wang 1,2,3, Antoni Rozalski 4, Zhujun Shao 5, Xi Guo 1,2,3, Haijian Zhou 5, Lu Feng 1,6,7, Lei Wang 1,2,3,6,7,*
PMCID: PMC2918944  PMID: 20581173

Abstract

Proteus species are well-characterized opportunistic pathogens primarily associated with urinary tract infections (UTI) of humans. The Proteus O antigen is one of the most variable constituents of the cell surface, and O antigen heterogeneity is used for serological classification of Proteus isolates. Even though most Proteus O antigen structures have been identified, the O antigen locus has not been well characterized. In this study, we identified the putative Proteus O antigen locus and demonstrated this region's high degree of heterogeneity by comparing sequences of 40 Proteus isolates using PCR-restriction fragment length polymorphism (RFLP). This analysis identified five putative Proteus O antigen gene clusters, and the probable functions of these O antigen-related genes were proposed, based on their similarity to genes in the available databases. Finally, Proteus-specific genes from these five serogroups were identified by screening 79 strains belonging to the 68 Proteus O antigen serogroups. To our knowledge, this is the first molecular characterization of the putative Proteus O antigen locus, and we describe a novel molecular classification method for the identification of different Proteus serogroups.

Proteus species are usually found in soil, water, and sewage and are well-known opportunistic pathogens that most commonly cause urinary tract infections (UTIs) in persons with anatomical and physiological defects of their urinary tracts (15, 28). This genus includes the five named species P. mirabilis, P. vulgaris, P. myxofaciens, P. penneri, and P. hauseri and the three unnamed Proteus genomospecies 4, 5, and 6 (20, 21). Among these, P. mirabilis, P. vulgaris, and P. penneri are the most common human pathogens (28). Among Proteus species, P. mirabilis is most frequently associated with UTIs and is a common cause of catheter-associated UTIs (12).

Potential virulence factors and bacterial behaviors associated with the infection processes and disease, including swarming, growth rates, fimbria expression, flagella, and the production of hemolysins, ureases, proteases, and amino acid deaminases, in addition to the expression of lipopolysaccharide (LPS) antigens and capsular polysaccharides (CPSs), have been described in many studies (11, 18, 28). Both LPSs and CPSs have been considered to play an important role in the progression of UTIs, in addition to affecting both kidney and bladder stone formation (7, 25, 35). Furthermore, the LPS O antigen confers protection against serum-mediated bactericidal activity (13, 27), and bacterial LPS released from bacteria is a biologically active endotoxin that causes a broad spectrum of pathophysiological conditions, including septic shock (26). Recently, two additional virulence factors with cytotoxic and agglutination properties, the high-affinity phosphate transporter (Pst) and the autotransporter (Pta), have been described (1, 11).

The O antigen located on the cell surface of Gram-negative bacteria consists of oligosaccharide repeats (O unit) that normally contain 2 to 8 sugar residues. The O antigen is one of the most variable constituents on the cell surface, due to variations in the types of sugars present and their arrangements and respective linkages, and is subject to intense selection by the host immune system and bacteriophages. The serological classification scheme established by Kauffman and Perch defines 49 different P. mirabilis and P. vulgaris O serogroups (10), and an additional 11 serogroups were later proposed (23). In the case of P. penneri, an additional 15 O antigen serogroups were described (16, 42; Z. Sidorczyk, K. Zych, K. Kolodziejska, D. Drzewiecka, and A. Zablotni, presented at the Second German-Polish-Russian Meeting on Bacterial Carbohydrates, Moscow, Russia, 10 to 12 September 2002). To date, the O antigen structures of 78 Proteus species have been described (unpublished data), and uronic acid, which can sometimes be substituted for amino acids, was identified as a component of the Proteus O antigen. Although acidic O-specific polysaccharides have been identified in most Proteus O antigens, a study of the genetic locus associated with Proteus O antigens has never been carried out.

The genome sequence of P. mirabilis was published for the first time in 2008 (22). In this study, we characterized the putative O antigen locus by analyzing genomic sequences and confirming the locus heterogeneity by carrying out PCR-restriction fragment length polymorphism (RFLP) on 40 strains. Four putative O antigen gene clusters were sequenced and analyzed, and specific primers were identified for Proteus species by screening 79 Proteus strains, confirming that the identified loci were specific to particular serogroups.

MATERIALS AND METHODS

Bacterial strains.

All bacterial strains used in this study are described in Table 1. All Proteus strains were provided by the Institute of Microbiology, Biotechnology and Immunology, University of Lodz (Lodz, Poland).

TABLE 1.

Proteus strains screened

Serogroup Laboratory stock no. Original no. Pool
Proteus vulgaris O1 G2290 CCUG 4635 1
Proteus vulgaris O2 G2291 Prk 5/57 1
Proteus mirabilis O3a,3b G2292 CCUG 4637 1
Proteus mirabilis O3a,3c G2293 G1 1
Proteus mirabilis O10 G2294 Prk 20/57 1
Proteus mirabilis O23a G2295 CCUG 19016 1
Proteus mirabilis O23a,23c,23d G2296 Prk 42/57 1
Proteus vulgaris O23a,23c G2297 CCUG 19017 1
Proteus mirabilis O27 G2298 Prk 50/57 2
Proteus mirabilis O28 G2299 Prk 51/57 2
Proteus vulgaris O42 G2300 CCUG 4677 2
Proteus vulgaris O4 G2608 PrK 9/57 2
Proteus mirabilis O5 G2609 PrK 12/57 2
Proteus mirabilis O6 G2610 PrK 14/57 2
Proteus mirabilis O7 G2611 PrK 15/57 2
Proteus vulgaris O8 G2612 PrK 17/57 2
Proteus mirabilis O9 G2613 PrK 18/57 3
Proteus mirabilis O11 G2614 PrK 24/57 3
Proteus vulgaris O12 G2615 PrK 25/57 3
Proteus mirabilis O13 G2616 PrK 26/57 3
Proteus mirabilis O14 G2617 PrK 28/57 3
Proteus vulgaris O15 G2618 PrK 30/57 3
Proteus vulgaris O17 G2619 CCUG4652 3
Proteus penneri O17 G2620 16 3
Proteus mirabilis O18 G2621 PrK 34/57 4
Proteus vulgaris O19 G2622 PrK 37/57 4
Proteus penneri O19ab G2623 31 4
Proteus mirabilis O20 G2624 PrK 38/57 4
Proteus vulgaris O21 G2625 PrK 39/57 4
Proteus vulgaris O22 G2626 PrK 40/57 4
Proteus mirabilis O24 G2627 PrK 47/57 4
Proteus vulgaris O25 G2628 PrK 48/57 4
Proteus mirabilis O26 G2629 PrK 49/57 5
Proteus mirabilis O29 G2630 PrK 52/57 5
Proteus mirabilis O30 G2631 PrK 53/57 5
Proteus vulgaris O31 G2632 PrK 55/57 5
Proteus penneri O31a G2633 26 5
Proteus vulgaris O32 G2634 PrK 57/57 5
Proteus vulgaris O34 G2635 CCUG4669 5
Proteus vulgaris O37a,37b G2636 PrK 63/57 5
Proteus mirabilis O38 G2637 PrK 64/57 6
Proteus mirabilis O39 G2638 PrK 65/57 6
Proteus mirabilis O40 G2639 PrK 66/57 6
Proteus mirabilis O41 G2640 PrK 67/57 6
Proteus vulgaris O44 G2641 PrK 70/57 6
Proteus vulgaris O45 G2642 CCUG 4680 6
Proteus vulgaris O47 G2643 PrK 73/57 6
Proteus mirabilis O48 G2644 PrK 74/57 6
Proteus mirabilis O49 G2645 PrK 75/57 7
Proteus mirabilis O51 G2646 CCUG 19011 7
Proteus mirabilis O50 G2647 TG 332 7
Proteus vulgaris O53 G2648 TG 276-1 7
Proteus mirabilis O54a G2649 10704 7
Proteus vulgaris O54ab G2650 TG 103 7
Proteus vulgaris O55 G2651 TG 155 7
Proteus genomospecies 4 O56 G2652 7
Proteus mirabilis O57 G2653 TG 83 8
Proteus penneri O58 G2654 11 8
Proteus penneri O59 G2655 14 8
Proteus myxofaciens O60 G2656 8
Proteus penneri O61 G2657 52 8
Proteus penneri O62 G2658 41 8
Proteus penneri O63 G2659 22 8
Proteus penneri O64abc G2660 19 8
Proteus penneri O64abd G2661 62 9
Proteus penneri O64ace G2662 71 9
Proteus penneri O65 G2663 34 9
Proteus penneri O66 G2664 2 9
Proteus penneri O67 G2665 8 9
Proteus penneri O68 G2666 63 9
Proteus penneri O69 G2667 25 9
Proteus penneri O70 G2668 60 9
Proteus penneri O71 G2669 42 10
Proteus penneri O72a G2670 1 10
Proteus penneri O72ab G2671 4 10
Proteus penneri O73ab G2672 103 10
Proteus penneri O73ac G2673 75 10
Proteus penneri O74 G2674 10705 10
Proteus penneri O75 G2675 10702 10
Open in a new tab

Genomic-DNA extraction and O antigen gene cluster amplification.

Genomic DNA was prepared as previously described (3). The O antigen gene cluster was identified from the genome sequence of P. mirabilis strain H14320 (GenBank accession no. NC_010554). Primers wl_31262 (5′-GAGTTATTACGHGAAACBGTAAAAGC-3′) and wl_31263 (5′-GTTAACTTTGATGCGTTGTTTATGAACTA-3′) were designed based on the cpxA and secB gene sequences, respectively, and were used to amplify the O antigen gene cluster. PCR amplification was performed in 50-μl volumes containing 1× LA buffer (plus MgCl2), 0.4 mM (each) deoxynucleoside triphosphates, 0.4 μM each primer, 2.5 U TaKaRa LA Taq polymerase (Takara, Shiga, Japan), and 2 μl of template DNA (approximately 500 ng). The PCR cycles used were as follows: 30 cycles of denaturation at 95°C for 30 s, annealing at 55°C for 45 s, an extension at 68°C for 15 min, and a final extension at 68°C for 7 min.

Putative O antigen RFLP.

Eight microliters of purified O antigen PCR product (the sizes are about 14,000 to 20,000 bp) was digested with 7.5 U of HindIII and 7.5 U of EcoRI (Takara). The mixture was incubated at 37°C for 12 to 16 h and analyzed by subjecting the enzymatic digest reaction to 1.0% agarose gel electrophoresis. DNA fingerprints were analyzed by BioNumerics software (Applied Maths, Belgium), and dendrograms were created, using the Dice coefficient with the unweighted-pair group method with an arithmetic mean.

Sequencing and analysis of putative O antigen gene clusters.

The O antigen PCR products were digested with DNase I, and the resulting DNA fragments were cloned into pGEM-T Easy to produce a gene library, as described previously (41). Sequencing was carried out using an ABI 3730 automated DNA sequencer (Applied Biosystems, Foster City, CA). Sequencing data were assembled, using the Staden package (31), and were annotated by Artemis (30). BLAST was used to search available databases, including GenBank and the COG protein motif databases (33). The HMMER program was used to search the protein domain database Pfam (4). The TMHMM version 2.0 analysis program (http://www.cbs.dtu.dk/services/TMHMM/) was used to identify potential transmembrane motifs. Sequence alignment and comparisons were done using the ClustalW program (34).

PCR specificity testing.

A total of 10 DNA pools containing DNA from 7 to 8 Proteus strains (Table 1) were prepared. Pools were screened, using primers based on the wzx and wzy gene sequences of P. mirabilis O3a,3b, P. mirabilis O10, P. vulgaris O23a,23c, P. mirabilis O27, and P. vulgaris O47 (Table 2). The PCR cycles used were as follows: 30 cycles of denaturation at 95°C for 15 s, annealing for 30 s at different temperatures, and an extension at 72°C for 1 min, and a final extension at 72°C for 5 min. PCR was carried out in a total volume of 25 μl. Two microliters of the respective reaction was subjected to agarose gel electrophoresis for examination of amplified products.

TABLE 2.

Analysis of Proteus PCR specificity

Serogroup Gene Base positions of genes Forward primer/reverse primer (5′-3′) Predicted PCR fragment length (bp) Annealing temp (°C)
P. mirabilis O3a,3b wzx 4705-5925 wl_11292 AGGAGTTAGGCAAGCATCT/wl_11293 CAACCGCTAAGACTGAGAA 695 58
wl_11294 CAGTATTCCCTATCCTTTCA/wl_11295 GTTATACCTATGCACCCTCC 346 48.5
wzy 8591-9325 wl_11296 AGCATTTGAACAGCCTTTCT/wl_11297 TCCGGCATCTATTACATTTT 451 52
wl_11298 AGATATGATGCTGGAAATGA/wl_11299 GTGAAAGTATAGCCAGCAAC 367 58
P. mirabilis O10 wzx 4724-5944 wl_11300 TTCACTATCTGGGATCATTT/wl_11301 CACAAGCTATGAGTAAGGGA 701 48.5
wl_11302 CCCTTACTCATAGCTTGTGC/wl_11303 AAGGCTTGCTTAGACCAGTTT 271 58
wzy 7080-8270 wl_11304 TTCAATCATCCCTCTAGTCA/wl_11305 ATTAGACCTAGCAGCACCAA 481 58
wl_11306 TACCCTAAAGCAATGGGAATA/wl_11307 ATGGAGAATAAAAAAGTCCAC 586 58
P. vulgaris O23a,23c wzx 8203-9429 wl_11308 CATTGTATCGGCGTCAGT/wl_11309 AAAAGAGTTTGATAGGCTAAA 274 58
wl_11310 ATGGTATTTCCAAGCCACAT/wl_11311 AGATACTGACGCCGATACAA 397 58
wzy 10535-11731 wl_11312 TGTGGTGAAGTATTAGGTGGAT/wl_11313 TGAAATAGCAGCGATAGGAG 300 58
wl_11314 TATTGTTGCTACCCATCTGC/wl_11315 GTAAGGCTTCTAGTAATGGAGA 510 58
P. mirabilis O27 wzx 1719-2975 wl_11316 GGCTAACTCTAACGGTGCTT/wl_11317 GCTAATACATAAATTCCGACAG 528 58
wl_11318 GGACAATGGCAAGTAAGGAA/wl_11319 AAAGCTAATGCAGCACCAA 767 58
wzy 2983-4074 wl_11320 AATACCATTCGCCCTTATC/wl_11321 ATAAAGCCCGAATCTATCAG 472 58
wl_11322 AAACCCTGTAGTATCATTTCTA/wl_11323 AAATGATACAACACCAGCAA 673 58
P. vulgaris O47 wzx 2676-3800 wl_11324 TTTTCTCGCTTTCTCCTTTC/wl_11325 CTCGGCTACCATAGAAGTGTT 451 58
wl_11326 TCTATGGTAGCCGAGAACTT/wl_11327 CAGCATAATTTGAAGGGAAT 252 58
wzy 4376-5620 wl_11328 AGCCTAGCAACAGGTGTAAT/wl_11329 ATAGCCGACCTTGAACTGAT 748 58
wl_11330 AATAGGTGATTCCCTTTCAT/wl_11331 ATTCATCTATAATGCCTTGTG 701 58
Open in a new tab

Analysis of Proteus LPS.

Strains were incubated in 10 ml of Luria-Bertani medium overnight at 37°C. Cells were harvested and washed with 10 ml of Tris-HCl (30 mM, pH 8.1) and suspended in 400 μl of 20% sucrose (20 μl of 1 mg/ml of lysozyme was added), where they stayed for 30 min at 4°C and 30 min at −80°C. After that, 3 ml of EDTA was added (3 mM), and the mixture was sonicated. An equal volume of phenol was added, and the cells were incubated at 70°C for 10 min. The aqueous phase was then centrifuged (16,000 rpm; 60 min; 4°C) to collect the LPS. The LPS was separated by SDS-PAGE (15%). The gel was fixed, using 25% isopropanol and 7.5% acetic acid at 25°C for 20 min; oxidized with 0.42% periodic acid and 7.5% acetic acid at 25°C for 10 min; washed three times with distilled water; stained with freshly prepared straining solution containing 4 ml of 30% ammonium hydroxide, 5.6 ml 1 M sodium hydroxide, 200 ml water, and 14 ml of 20% silver nitrate; washed three times with distilled water; and reduced in 500 ml of water containing 25 mg of citric acid and 0.25 ml of 37% formaldehyde (40).

Nucleotide sequence accession numbers.

The DNA sequences of the P. mirabilis O3a,3b, P. vulgaris O23a,23c, P. mirabilis O27, and P. vulgaris O47 O antigen gene clusters have been deposited in GenBank under accession numbers GU254059, GU254061, GU254062, and GU254063, respectively.

RESULTS

Identification of a candidate gene cluster associated with P. mirabilis O10 O antigen synthesis.

The published (22) P. mirabilis H14320 (serogroup O10) genome was analyzed, and several putative gene clusters associated with polysaccharide synthesis were identified (data not shown). One gene cluster containing various glycosyltransferase genes, O antigen assembly genes, and nucleotide sugar synthesis genes, including ugd (encoding UDP-glucose dehydrogenase) and gla (encoding UDP-glucuronate 4′-epimerase), which are responsible for UDP-GalA synthesis (36, 37), is proposed to be involved in P. mirabilis O10 O antigen synthesis (32). Two housekeeping genes, cpxA (encoding two-component system sensor kinase) and secB (encoding a preprotein translocase subunit), were also found to the left and right ends, respectively, of the putative O antigen gene cluster.

PCR-RFLP analyses of the putative O antigen loci from various Proteus isolates.

The proposed O antigen gene cluster of 40 strains from 39 serogroups was amplified, using primers targeting the flanking cpxA and secB genes followed by digestion with HindIII and EcoRI. DNA fragments ranged from 200 to 5,000 bp (Fig. 1). The restriction patterns were analyzed by BioNumerics software and demonstrated that strains corresponding to different serogroups had distinct digestion patterns, different from those of strains of the same serogroup, which had similar patterns (Fig. 2). Reproducible restriction patterns were obtained for each strain. The PCR-RFLP results suggested that the region between cpxA and secB was involved in Proteus O antigen synthesis.

FIG. 1.

FIG. 1.

Open in a new tab

PCR-RFLP analyses of putative O antigen gene clusters following restriction with HindIII and EcoRI. Lane M, DNA marker; lane 1, O69 (G2667); lane 2, O71 (G2669); lane 3, O72a (G2670); lane 4, O73ac (G2673); lane 5, O74 (G2674); lane 6, O75 (G2675).

FIG. 2.

FIG. 2.

Open in a new tab

Dendrogram of Proteus O antigen gene clusters generated by using the BioNumerics software program following restriction digestion by HindIII and EcoRI.

Sequencing and analyses of the putative O antigen gene clusters of different Proteus species.

A 15,098-bp gene cluster sequence was obtained from the genome sequence of the P. mirabilis H14320 strain. The region between the cpxA and secB genes was sequenced, and 16,197-, 14,554-, 13,935-, and 11,063-bp sequences were obtained from P. mirabilis O3a,3b, P. vulgaris O23a,23c, P. mirabilis O27, and P. vulgaris O47, respectively. Open reading frames (ORFs) 16, 14, 14, 14, and 11, corresponding to the region between (but not including) the cpxA and secB genes of strains O3a,3b, O10, O23a,23c, O27, and O47, respectively, were identified (Fig. 3), and their functions were assigned based on similarity matches to ORFs from available databases (see Tables S1 to S5 in the supplemental material). The GC content of the putative O antigen synthesis ORFs ranged from 22.0 to 35.8%, which was lower than that of the rest of the Proteus genome (38.9%) (22). Genes associated with sugar synthesis, sugar transfer, O antigen processing, and other functions were found between cpxA and secB.

FIG. 3.

FIG. 3.

Open in a new tab

Structural organization of O antigen gene clusters of P. mirabilis O3a,3b, P. mirabilis O10, P. vulgaris O23a,23c, P. mirabilis O27, and P. vulgaris O47.

Nucleotide sugar biosynthesis genes.

Genes involved in the biosynthesis of common sugar nucleotide precursors (GlcNAc, Glc, and Gal) were not located between the cpxA and secB genes. A common structural feature of the Proteus O antigens was the presence of uronic acids, which at times can be substituted for amino acids during bacterial protein synthesis (28). The O antigen structures of strains O3a,3b, O10, O23a,23c, O27, and O47 have been defined (14, 24, 29, 32, 38) (Fig. 4). Uncommon sugars of GalA, GlcA, and GalNAc were found in O3a,3b; of GalA, AltA, and GalNAc in O10; of GalA and GalNAc in O23a,23c; of GlcA and GalA in O27; and of GalNAc and GlcA in O47. The ugd and gla genes have been shown to be involved in UDP-GlcA and UDP-GalA biosyntheses (the nucleotide-activated forms of GlcA and GalA, respectively) (36, 37) and were found to be part of the putative O antigen gene clusters of these five serogroups, except that only ugd (but not gla) was located in the O47 O antigen gene cluster. The gne gene responsible for UDP-GalNAc synthesis (the nucleotide-activated form of GalNAc) (5) was found only in the O antigen gene cluster of strain O3a,3b but not in strain O10, O23a,23c, or O47 (with GalNAc in their O antigen structures) gene clusters. The gne gene may be present outside of the O antigen cluster, as reported for Escherichia coli (39). The glf gene involved in the synthesis of UDP-Galf (19) was found in the O antigen gene cluster of strain O3a,3b, whose O antigen does not include Galf, suggesting that glf may not be responsible for O antigen biosynthesis in strain O3a,3b.

FIG. 4.

FIG. 4.

Open in a new tab

Respective O antigen compostions of P. mirabilis O3a,3b, P. mirabilis O10, P. vulgaris O23a,23c, P. mirabilis O27, and P. vulgaris O47.

Sugar transferase genes.

Five sugars were found in the O antigen cluster of strain O3a,3b, and four sugars were found in the O antigen gene clusters of strains O10, O23a,23c, O27, and O47. Five sugar transferase genes were found between cpxA and secB in strains O10 and O23a,23c, and four, six, and three sugar transferase genes were identified between cpxA and secB in strains O3a,3b, O27, and O47, respectively. Among the sugar transferase genes in this locus of the five serogoups, it was noted that three sugar transferase genes in O10, O23a,23c, and O47, four in O3a,3b, and six in O27 have the same promoter used for the transcription of other O antigen genes.

O antigen-processing genes.

Both wzx and wzy genes were found to be associated with the O3a,3b, O10, O23a,23c, O27, and O47 serogroups, respectively. The putative Wzx proteins from the five serogroups have 10 to 12 well-proportioned transmembrane segments characteristic of Wzx proteins (17). The putative Wzy proteins from these serogroups were found to have 9 to 11 predicted transmembrane segments with large periplasmic loops of 32- to 89-amino-acid residues typical of the topological character of Wzy proteins (8). These observations suggested that the process for synthesis and translocation of O antigens in these five Proteus serogroups was a Wzx/Wzy-dependent process.

Additional genes identified.

A putative methyltransferase, two serine acetyltransferases, and a glycerol-3-phosphate dehydrogenase gene were found between cpxA and secB of the five serogroups examined. The methyltransferase gene was found to share the same promoter as the other putative O antigen synthesis genes; however, this promoter was not associated with the other three genes identified, suggesting that these genes were not associated with Proteus O antigen synthesis. An acetyltransferase gene was found in the O23a,23c and O47 putative O antigen gene clusters, suggesting that the acetyltransferase genes may be responsible for the transfer of acetyl groups to O antigen sugar residues.

Identification of serogroup-specific genes.

For each of the Proteus strains examined, four distinct primer pairs were used (Table 2) to amplify wzx and wzy gene products from DNA isolated from the representative 79 Proteus strains (Table 1). Primer pairs designed to amplify specific DNA regions of each of the serogroups examined amplified only DNA corresponding to the respective strain, based on the identification of PCR products of the anticipated size (Fig. 5). The sequences of wzx and wzy genes amplified by four distinct primer sets in each of the five serogroups were aligned to generate a phylogenetic tree. It showed that the partial wzx/wzy sequences are serogroup specific (Fig. 6). Therefore, the wzx and wzy gene sequences were specific to particular Proteus serogroups, and the four primer pairs used to amplify these sequences in the respective serogroups could be used in the development of PCR assays adapted for the identification and detection of respective Proteus serogroups.

FIG. 5.

FIG. 5.

Open in a new tab

Agarose gel electrophoresis of PCR products generated from amplification of template DNA derived, respectively, from P. mirabilis O3a,3b, P. mirabilis O10, P. vulgaris O23a,23c, P. mirabilis O27, and P. vulgaris O47, using primer pairs wl_11292/11293 (lane 2), wl_11294/11295 (lane 3), wl_11296/11297 (lane 4), wl_11298/11299 (lane 5), wl_11300/11301 (lane 6), wl_11302/11303 (lane 7), wl_11304/11305 (lane 8), wl_11306/11307 (lane 9), wl_11308/11309 (lane 10), wl_11310/11311 (lane 11), wl_11312/11313 (lane 12), wl_11314/11315 (lane 13), wl_11316/11317 (lane 14), wl_11318/11319 (lane 15), wl_11320/11321 (lane 16), wl_11322/11323 (lane 17), wl_11324/11325 (lane 18), wl_11326/11327 (lane 19), wl_11328/11329 (lane 20), and wl_11330/11331 (lane 21). Lanes 1 and 22 correspond to the DNA ladder. Bands correspond to 100, 250, 500, and 750 1-kb and 2-kb pairs.

FIG. 6.

FIG. 6.

Open in a new tab

A phylogenetic tree based on the alignment of sequences from wzx and wzy genes amplified by four distinct primer sets in each of the five Proteus O serogroups. O3a,3b wzx1 represents the sequence amplified by the primer pair wl_11292 and wl_11293, O3a,3b wzx2 represents the sequence amplified by the primer pair wl_11294 and wl_11295, O3a,3b wzy1 represents the sequence amplified by the primer pair wl_11296 and wl_11297, O3a,3b wzy2 represents the sequence amplified by the primer pair wl_11298 and wl_11299, O10 wzx1 represents the sequence amplified by the primer pair wl_11300 and wl_11301, O10 wzx2 represents the sequence amplified by the primer pair wl_11302 and wl_11303, O10 wzy1 represents the sequence amplified by the primer pair wl_11304 and wl_11305, O10 wzy2 represents the sequence amplified by the primer pair wl_11306 and wl_11307, O23a,23c wzx1 represents the sequence amplified by the primer pair wl_11308 and wl_11309, O23a,23c wzx2 represents the sequence amplified by the primer pair wl_11310 and wl_11311, O23a,23c wzy1 represents the sequence amplified by the primer pair wl_11312 and wl_11313, O23a,23c wzy2 represents the sequence amplified by the primer pair wl_11314 and wl_11315, O27 wzx1 represents the sequence amplified by the primer pair wl_11316 and wl_11317, O27 wzx2 represents the sequence amplified by the primer pair wl_11318 and wl_11319, O27 wzy1 represents the sequence amplified by the primer pair wl_11320 and wl_11321, O27 wzy2 represents the sequence amplified by the primer pair wl_11322 and wl_11323, O47 wzx1 represents the sequence amplified by the primer pair wl_11324 and wl_11325, O47 wzx2 represents the sequence amplified by the primer pair wl_11326 and wl_11327, O47 wzy1 represents the sequence amplified by the primer pair wl_11328 and wl_11329, and O47 wzy2 represents the sequence amplified by the primer pair wl_11330 and wl_11331.

Proteus LPS analysis.

LPSs of 8 Proteus strains belonging to 5 serogroups, O3, O10, O23, O27, and O47, were extracted and subjected to SDS-PAGE. The corresponding banding patterns were visualized by silver staining, and the phenotypes of LPS molecules were analyzed by comparison with that of an E. coli strain. The structure of Proteus LPS is similar to that of E. coli, which has low-molecular-mass bands corresponding to the lipid A core oligosaccharide (OS) (a single O unit attached to the lipid A core OS) and higher-molecular-mass bands corresponding to a typical smooth LPS phenotype (Fig. 7). The LPS phenotypes of strains belonging to different serogroups are distinct, and strains belonging to the same serogroups, such as two strains of serogroup O3 (P. mirabilis O3a,3b and O3a,3c) and three strains of serogroup O23 (P. mirabilis O23a and 23a,23c,23d and P. vulgaris O23a,23c), have similar LPS phenotypes (Fig. 7).

FIG. 7.

FIG. 7.

Open in a new tab

SDS-PAGE analysis of Proteus and E. coli LPS. Lane 1, P. mirabilis O3a,3b (G2292); lane 2, P. mirabilis O3a,3c (G2293); lane 3, P. vulgaris O23a,23c,23d (G2296); lane 4, P. vulgaris O23a (G2295); lane 5, P. mirabilis O10 (G2294); lane 6, P. vulgaris O23a,23c (G2297); lane 7, P. mirabilis O27 (G2298); lane 8, P. vulgaris O47 (G2643), and lane 9, E. coli (G1216).

DISCUSSION

This report describes for the first time the molecular characterization of the putative gene cluster responsible for Proteus O antigen synthesis. The O antigen gene cluster is localized between galF and gnd in the E. coli and Salmonella genomes. In this report, by analyzing the genome sequence of Proteus mirabilis H14320, we predicted that a putative O antigen gene cluster is located between cpxA and sacB. PCR-RFLP analysis and comparison of restriction patterns of 40 Proteus strains made the location seem more probable. In addition, we sequenced the locus from four Proteus O antigen serogroups and tested the validity of diagnostic primer sets against 79 Proteus strains that included 68 distinct O serogroups.

Similar to the E. coli and Salmonella O antigen gene cluster, the Proteus cluster also contains three major gene classes: sugar biosynthetic pathway genes, sugar transferase genes, and O antigen-processing genes. As most O antigens of Proteus contain uronic acids, it was not surprising to find the ugd and gla genes among the putative O antigen gene clusters. As one fewer sugar transferase gene than the number of sugars in each O unit (except in strain O27) was found to have the same promoter used for the transcription of other putative O antigen synthesis genes of the five O serogroups, we propose that the transferase gene responsible for the first sugar (such as GlcNAc or GalNAc) may be located outside of this gene cluster, as seen in E. coli (2), and that the sugar transferase genes sharing the same promoter of other putative O antigen genes were responsible for the transfer of other sugars in the O unit. All of the putative O antigen-processing genes of the five serogroups were wzx and wzy, indicating that the assembly of Proteus O antigen was likely to be Wzx/Wzy dependent. Additional genes not necessary for O antigen synthesis were also found at the 3′ end of the gene cluster, and we hypothesized that these genes were not responsible for O antigen biosynthesis.

Comparison of Proteus LPS profiles to E. coli LPS profiles indicated that Proteus LPS was composed of O chains and lipid A and that LPS profiles of strains belonging to the same O serogroup had similar LPS patterns.

We also attempted to generate Proteus O antigen gene synthesis mutants; however, as the Proteus strains tested were resistant to various antibiotics, including ampicillin and kanamycin, the mutation and complementation procedures could not be carried out to generate a mutant strain.

Epidemiological studies have demonstrated that the O3, O10, O27, and O28 Proteus serogroups are most often associated with human infections. Therefore, techniques that can be used in the rapid serotyping of Proteus strains can provide useful diagnostic information that can be applied toward patient treatment. The traditional Proteus serotyping methods utilize antisera specific for respective O antigens. This is a slow, labor-intensive procedure with potential cross-reactivity concerns that may not lead to the correct diagnosis. PCR-based methods are rapid, cost-effective, accurate approaches that can be modified for diagnostic purposes. wzx and wzy usually displayed low levels of homogeneity between Proteus serogroups, and amplification of these gene sequences has proved successful in the identification of E. coli and Salmonella isolates. Since PCR assays targeting these genes for the detection of pathogenic E. coli strains have been reported for serogroups O157, O123, and O86 (6, 9, 41), the same strategy was applied to the characterization and identification of Proteus strains.

The data presented in this report describe the identification and molecular characterization of the putative Proteus O antigen synthesis gene cluster from 5 distinct serogroups and the development of a novel PCR-based method for the identification of respective Proteus isolates based on the amplification of unique wzx and wzy gene sequences.

Supplementary Material

[Supplemental material]
supp_76_16_5471__index.html (1KB, html)

Acknowledgments

This study was supported by funds from the National Science Foundation of China (30900255, 30788001, 30670038, and 30870070), the National 863 Program (2007AA02Z106, 2007AA021303, and 2010AA10A203), the Tianjin Research Program of Application Foundation and Advanced Technology (10JCYBJC10100), the National Key Program for Infectious Diseases of China (2009ZX10004-108), and the Ministry of Science and Higher Education, Poland (core funding for statutory research activities, grant 803 from the Department of Immunobiology of Bacteria, University of Lodz).

Footnotes

▿

Published ahead of print on 25 June 2010.

†

Supplemental material for this article may be found at http://aem.asm.org/.

REFERENCES

  • 1.Alamuri, P., and H. L. Mobley. 2008. A novel autotransporter of uropathogenic Proteus mirabilis is both a cytotoxin and an agglutinin. Mol. Microbiol. 68:997-1017. [DOI] [PubMed] [Google Scholar]
  • 2.Alexander, D. C., and M. A. Valvano. 1994. 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. 176:7079-7084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Bastin, D. A., and P. R. Reeves. 1995. Sequence and analysis of the O antigen gene (rfb) cluster of Escherichia coli O111. Gene 164:17-23. [DOI] [PubMed] [Google Scholar]
  • 4.Bateman, A., E. Birney, L. Cerruti, R. Durbin, L. Etwiller, S. R. Eddy, S. Griffiths-Jones, K. L. Howe, M. Marshall, and E. L. Sonnhammer. 2002. The Pfam protein families database. Nucleic Acids Res. 30:276-280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Bengoechea, J. A., E. Pinta, T. Salminen, C. Oertelt, O. Holst, J. Radziejewska-Lebrecht, Z. Piotrowska-Seget, R. Venho, and M. Skurnik. 2002. Functional characterization of Gne (UDP-N-acetylglucosamine-4-epimerase), Wzz (chain length determinant), and Wzy (O-antigen polymerase) of Yersinia enterocolitica serotype O:8. J. Bacteriol. 184:4277-4287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Beutin, L., Q. Wang, D. Naumann, W. Han, G. Krause, L. Leomil, L. Wang, and L. Feng. 2007. Relationship between O-antigen subtypes, bacterial surface structures and O-antigen gene clusters in Escherichia coli O123 strains carrying genes for Shiga toxins and intimin. J. Med. Microbiol. 56:177-184. [DOI] [PubMed] [Google Scholar]
  • 7.Beynon, L. M., A. J. Dumanski, R. J. McLean, L. L. MacLean, J. C. Richards, and M. B. Perry. 1992. Capsule structure of Proteus mirabilis (ATCC 49565). J. Bacteriol. 174:2172-2177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Daniels, C., C. Vindurampulle, and R. Morona. 1998. Overexpression and topology of the Shigella flexneri O-antigen polymerase (Rfc/Wzy). Mol. Microbiol. 28:1211-1222. [DOI] [PubMed] [Google Scholar]
  • 9.Feng, L., W. Han, Q. Wang, D. A. Bastin, and L. Wang. 2005. Characterization of Escherichia coli O86 O-antigen gene cluster and identification of O86-specific genes. Vet. Microbiol. 106:241-248. [DOI] [PubMed] [Google Scholar]
  • 10.Fritz, K. 1966. The bacteriology of Enterobacteriaceae. Williams & Wilkins, Baltimore, Md.
  • 11.Jacobsen, S. M., M. C. Lane, J. M. Harro, M. E. Shirtliff, and H. L. Mobley. 2008. The high-affinity phosphate transporter Pst is a virulence factor for Proteus mirabilis during complicated urinary tract infection. FEMS Immunol. Med. Microbiol. 52:180-193. [DOI] [PubMed] [Google Scholar]
  • 12.Jacobsen, S. M., D. J. Stickler, H. L. Mobley, and M. E. Shirtliff. 2008. Complicated catheter-associated urinary tract infections due to Escherichia coli and Proteus mirabilis. Clin. Microbiol. Rev. 21:26-59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kaca, W., M. Arabski, R. Fudala, E. Holmstrom, A. Sjoholm, A. Weintraub, B. Futoma-Koloch, G. Bugla-Ploskonska, and W. Doroszkiewicz. 2009. Human complement activation by smooth and rough Proteus mirabilis lipopolysaccharides. Arch. Immunol. Ther. Exp. 57:383-391. [DOI] [PubMed] [Google Scholar]
  • 14.Kaca, W., I. A. Knirel, E. V. Vinogradov, and K. Kotelko. 1987. Structure of the O-specific polysaccharide of Proteus mirabilis S1959. Arch. Immunol. Ther. Exp. (Warsz.) 35:431-437. [PubMed] [Google Scholar]
  • 15.Kim, B. N., N. J. Kim, M. N. Kim, Y. S. Kim, J. H. Woo, and J. Ryu. 2003. Bacteraemia due to tribe Proteeae: a review of 132 cases during a decade (1991-2000). Scand. J. Infect. Dis. 35:98-103. [DOI] [PubMed] [Google Scholar]
  • 16.Knirel, Y. A., W. Kaca, A. Rozalski, and Z. Sidorczyk. 1990. Structure of O-antigenic polysaccharides of Proteus bacilli. Pol. J. Chem. 73:859-907. [Google Scholar]
  • 17.Liu, D., R. A. Cole, and P. R. Reeves. 1996. An O-antigen processing function for Wzx(RfbX): a promising candidate for O-unit flippase. J. Bacteriol. 178:2102-2107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Mobley, H. L., and R. Belas. 1995. Swarming and pathogenicity of Proteus mirabilis in the urinary tract. Trends Microbiol. 3:280-284. [DOI] [PubMed] [Google Scholar]
  • 19.Nassau, P. M., S. L. Martin, R. E. Brown, A. Weston, D. Monsey, M. R. McNeil, and K. Duncan. 1996. Galactofuranose biosynthesis in Escherichia coli K-12: identification and cloning of UDP-galactopyranose mutase. J. Bacteriol. 178:1047-1052. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.O'Hara, C. M., F. W. Brenner, and J. M. Miller. 2000. Classification, identification, and clinical significance of Proteus, Providencia, and Morganella. Clin. Microbiol. Rev. 13:534-546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.O'Hara, C. M., F. W. Brenner, A. G. Steigerwalt, B. C. Hill, B. Holmes, P. A. Grimont, P. M. Hawkey, J. L. Penner, J. M. Miller, and D. J. Brenner. 2000. Classification of Proteus vulgaris biogroup 3 with recognition of Proteus hauseri sp. nov., nom. rev. and unnamed Proteus genomospecies 4, 5 and 6. Int. J. Syst. Evol. Microbiol. 50:1869-1875. [DOI] [PubMed] [Google Scholar]
  • 22.Pearson, M. M., M. Sebaihia, C. Churcher, M. A. Quail, A. S. Seshasayee, N. M. Luscombe, Z. Abdellah, C. Arrosmith, B. Atkin, T. Chillingworth, H. Hauser, K. Jagels, S. Moule, K. Mungall, H. Norbertczak, E. Rabbinowitsch, D. Walker, S. Whithead, N. R. Thomson, P. N. Rather, J. Parkhill, and H. L. Mobley. 2008. Complete genome sequence of uropathogenic Proteus mirabilis, a master of both adherence and motility. J. Bacteriol. 190:4027-4037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Penner, J. L., and J. N. Hennessy. 1980. Separate O-grouping schemes for serotyping clinical isolates of Proteus vulgaris and Proteus mirabilis. J. Clin. Microbiol. 12:304-309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Perepelov, A. V., A. S. Shashkov, D. Babichka, S. N. Senchenkova, B. Bartodziejska, A. Rozalski, and Y. A. Knirel. 2000. Structure of the O-specific polysaccharide of the bacterium Proteus vulgaris O23. Biochemistry (Mosc.) 65:1055-1059. [PubMed] [Google Scholar]
  • 25.Perry, M. B., and L. L. MacLean. 1994. The structure of the polysaccharide produced by Proteus vulgaris (ATCC 49990). Carbohydr. Res. 253:257-263. [DOI] [PubMed] [Google Scholar]
  • 26.Raetz, C. R., and C. Whitfield. 2002. Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 71:635-700. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Rozalski, A. 2008. Lipopolysaccharide (LPS, endotoxin) of Proteus bacteria—chemical structure, serological specificity and the role in the pathogenicity. Folia Biol. Oecol. 4:5-24. [Google Scholar]
  • 28.Rozalski, A., Z. Sidorczyk, and K. Kotelko. 1997. Potential virulence factors of Proteus bacilli. Microbiol. Mol. Biol. Rev. 61:65-89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Rozalski, A., A. Torzewska, B. Bartodziejska, D. Babichka, I. Kwil, A. V. Perepelov, A. N. Kondakova, S. N. Senchenkova, I. A. Knirel, and E. V. Vinogradov. 2002. Chemical structure, antigenic specificity and the role in the pathogenicity of lipopolysaccharide (LPS, endotoxin) on Proteus vulgaris bacteria's example. Wiad. Chem. 56:585-604. [Google Scholar]
  • 30.Rutherford, K., J. Parkhill, J. Crook, T. Horsnell, P. Rice, M. A. Rajandream, and B. Barrell. 2000. Artemis: sequence visualisation and annotation. Bioinformatics 16:944-945. [DOI] [PubMed] [Google Scholar]
  • 31.Staden, R. 1996. The Staden sequence analysis package. Mol. Biotechnol. 5:233-241. [DOI] [PubMed] [Google Scholar]
  • 32.St. Swierzko, A., A. S. Shashkov, S. N. Senchenkova, F. V. Toukach, A. Ziolkowski, M. Cedzynski, N. A. Paramonov, W. Kaca, and Y. A. Knirel. 1996. Structural and serological studies of the O-specific polysaccharide of the bacterium Proteus mirabilis O10 containing l-altruronic acid, a new component of O-antigens. FEBS Lett. 398:297-302. [DOI] [PubMed] [Google Scholar]
  • 33.Tatusov, R. L., D. A. Natale, I. V. Garkavtsev, T. A. Tatusova, U. T. Shankavaram, B. S. Rao, B. Kiryutin, M. Y. Galperin, N. D. Fedorova, and E. V. Koonin. 2001. The COG database: new developments in phylogenetic classification of proteins from complete genomes. Nucleic Acids Res. 29:22-28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Torzewska, A., P. Staczek, and A. Rozalski. 2003. Crystallization of urine mineral components may depend on the chemical nature of Proteus endotoxin polysaccharides. J. Med. Microbiol. 52:471-477. [DOI] [PubMed] [Google Scholar]
  • 36.Usadel, B., U. Schluter, M. Molhoj, M. Gipmans, R. Verma, J. Kossmann, W. D. Reiter, and M. Pauly. 2004. Identification and characterization of a UDP-d-glucuronate 4-epimerase in Arabidopsis. FEBS Lett. 569:327-331. [DOI] [PubMed] [Google Scholar]
  • 37.Vigetti, D., M. Ori, M. Viola, A. Genasetti, E. Karousou, M. Rizzi, F. Pallotti, I. Nardi, V. C. Hascall, G. De Luca, and A. Passi. 2006. Molecular cloning and characterization of UDP-glucose dehydrogenase from the amphibian Xenopus laevis and its involvement in hyaluronan synthesis. J. Biol. Chem. 281:8254-8263. [DOI] [PubMed] [Google Scholar]
  • 38.Vinogradov, E. V., D. Krajewska-Pietrasik, W. Kaca, A. S. Shashkov, Y. A. Knirel, and N. K. Kochetkov. 1989. Structure of Proteus mirabilis O27 O-specific polysaccharide containing amino acids and phosphoethanolamine. Eur. J. Biochem. 185:645-650. [DOI] [PubMed] [Google Scholar]
  • 39.Wang, L., S. Huskic, A. Cisterne, D. Rothemund, and P. R. Reeves. 2002. The O-antigen gene cluster of Escherichia coli O55:H7 and identification of a new UDP-GlcNAc C4 epimerase gene. J. Bacteriol. 184:2620-2625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Wang, L., and P. R. Reeves. 1994. Involvement of the galactosyl-1-phosphate transferase encoded by the Salmonella enterica rfbP gene in O-antigen subunit processing. J. Bacteriol. 176:4348-4356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Wang, L., and P. R. Reeves. 1998. Organization of Escherichia coli O157 O antigen gene cluster and identification of its specific genes. Infect. Immun. 66:3545-3551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Zych, K., M. Kowalczyk, Y. A. Knirel, and Z. Sidorczyk. 2000. New serogroups of the genus Proteus consisting of Proteus penneri strains only. Determination of some LPS epitopes responsible for specificity. Adv. Exp. Med. Biol. 485:339-344. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

[Supplemental material]
supp_76_16_5471__index.html (1KB, html)
supp_76_16_5471__supplymentary_material_091113.zip (56.5KB, zip)

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES