|Home | About | Journals | Submit | Contact Us | Français|
Escherichia coli is the primary cause of urinary tract infection (UTI) in the developed world. The major factors associated with virulence of uropathogenic E. coli (UPEC) are fimbrial adhesins, which mediate specific attachment to host receptors and trigger innate host responses. Another group of adhesins is represented by the autotransporter (AT) subgroup of proteins. In this study, we identified a new AT-encoding gene, termed upaH, present in a 6.5-kb unannotated intergenic region in the genome of the prototypic UPEC strain CFT073. Cloning and sequencing of the upaH gene from CFT073 revealed an intact 8.535-kb coding region, contrary to the published genome sequence. The upaH gene was widely distributed among a large collection of UPEC isolates as well as the E. coli Reference (ECOR) strain collection. Bioinformatic analyses suggest β-helix as the predominant structure in the large N-terminal passenger (α) domain and a 12-strand β-barrel for the C-terminal β-domain of UpaH. We demonstrated that UpaH is expressed at the cell surface of CFT073 and promotes biofilm formation. In the mouse UTI model, deletion of the upaH gene in CFT073 and in two other UPEC strains did not significantly affect colonization of the bladder in single-challenge experiments. However, in competitive colonization experiments, CFT073 significantly outcompeted its upaH isogenic mutant strain in urine and the bladder.
Urinary tract infections (UTIs) are among the most common infectious diseases of humans. It is estimated that 40 to 50% of adult healthy women have experienced at least one UTI episode in their lifetime, and there is a tendency for these infections to become chronic due to a high rate of recurrence. Almost all patients with an indwelling urinary catheter for 30 days or longer develop catheter-associated UTI, which accounts for 40% of all nosocomial infections (13).
Uropathogenic Escherichia coli (UPEC) is the cause of the majority (>80%) of UTIs in humans and one of the most common sources of Gram-negative bacteremia in hospitalized patients. The ability of UPEC to colonize the urinary tract and cause disease involves adhesins (e.g., type 1 and P fimbriae), toxins (e.g., hemolysin), and iron acquisition systems that utilize siderophores (e.g., enterobactin, salmochelin, and aerobactin) (31, 60). Adherence to the urinary tract epithelium enables bacteria to resist the hydrodynamic forces of urine flow, to trigger host and bacterial cell signaling pathways, and to establish infection. Among adhesins, P and type 1 fimbriae correlate strongly with uropathogenesis and mediate binding to specific receptors in the urinary tract (7, 39, 47, 54, 62, 63). Both P and type 1 fimbriae recognize their receptor targets by virtue of organelle tip-located adhesins, PapG and FimH, respectively (27, 33).
In addition to fimbrial adhesins, a number of autotransporter (AT) proteins associated with virulence have been characterized for UPEC. These include the secreted toxin Sat (17, 18), the phase-variable outer membrane protein antigen 43 (Ag43) (57), and the surface-located trimeric AT protein UpaG (59). AT proteins are unique in that their sequence is sufficient to direct their transport across the bacterial membrane system and final routing of a variable passenger (α) domain to the cell surface. AT proteins are generally highly similar with respect to the structure of the translocation (β) domain that assists the transport of the α-domain across the outer membrane, whereas they differ substantially in their α-domain, which determines the unique functional characteristics of AT proteins (23, 24). Once at the bacterial surface, the α-domain may be processed and released into the extracellular surroundings (e.g., Pet and EspP), or it may be cleaved but remain in contact with the cell surface via noncovalent interactions with the β-domain (e.g., Ag43 and AIDA). Thus, AT proteins have diverse functions, ranging from cell-associated adhesins to secreted toxins (24).
Ten putative AT-encoding genes have been identified in the sequenced genome of the prototype UPEC strain CFT073 (45). In this study, we identified and characterized a new AT-encoding gene from CFT073, which we have named upaH. The upaH gene was not identified in the original analysis of the CFT073 genome sequence due to a sequence misassembly within a highly repetitive region. We show that upaH encodes a large cell surface-located AT protein that contributes to biofilm formation and colonization of the mouse urinary tract in coinfection experiments.
The strains and plasmids used in this study are listed in Table Table1.1. E. coli strains used to study the prevalence of the upaH gene have been described previously (35) or were part of the E. coli Reference (ECOR) collection (42). Cells were routinely grown at 37°C on solid or in liquid Luria broth (LB) medium (5) supplemented with the appropriate antibiotics: kanamycin (Kan) (50 μg/ml), ampicillin (Amp) (100 μg/ml), or chloramphenicol (Cam) (30 μg/ml). For growth under defined conditions, M9 minimal medium supplemented with 0.2% glucose (49) was used as indicated. The plasmid pUpaH contains the upaH gene from CFT073 and was created by PCR amplification using the primers 520 (5′-CGCGCTCGAGATAATAAGGAATTATTATGCAAAGGAAAACTCTATTGTCG) and 521 (5′-CGCGCAAGCTTTTAGAATGTTTTCTTGAAGT), digestion with XhoI-HindIII, and insertion into XhoI-HindIII-digested pBAD/Myc-HisA. A final modification of the plasmid was made by inserting a kanamycin resistance-encoding gene into the HindIII site. Resistance to kanamycin was required to facilitate transformation of these plasmids into the flu-negative gfp-positive K-12 strain OS56. In pUpaH, expression of upaH is under the control of the arabinose-inducible araBAD promoter (19). Mutants containing deletions in the upaH gene were constructed by λ-red-mediated homologous recombination as previously described (9). Primers with 50-bp homology extensions (718, 5′-GCTTACCCTCTGAATGCTCCTCTGTAGTCGAACAGAACCTGATGCCATGGGTGTAGGCTGGAGCTG CTTC; and 719, 5′-CTGTCCACCGCCGTTTCGCTGCCGCTGTTCATGCTCAGGTTAAAGTCACCCATATGAATATCCTCCTTAG) were employed to amplify the kanamycin resistance cassette used to replace upaH. Mutants were confirmed via PCR and sequencing using the primers 716 (5′-GCAAAGGAAAACTCTATTGTCG) and 717 (5′-GGCTGTTGAAACCACCACC). The upaH promoter insertion mutant (CFT073PcLupaH) was constructed using a strategy described previously (8, 57) and the following primers: upstream region (587, 5′-AGAACATTACTTCTTTGTGC; and 586, 5′-GTGAGAATTACTAACTTGAGCGAATCAAATGCGTCACATAGATATTTAT), downstream region (584, 5′-CATTAAAGAGGAGAAAGGTACCGCATGCAAAGGAAAACTCTATT GTCGGCCT; and 585, 5′-TCGGTATAAGTAAAAACTGA), PcL promoter and kanamycin resistance cassette (588, 5′-CTCTGGCAAGCGCCTCGATTACTG; and 589, 5′-CATCACCTTCACCCTCTCCACTGAC); verification (592, 5′-GCTTGAGACCCGTAGCGATC; 591, 5′-CTTCCTCGTGCTTTACGGTATCG; 593, 5′-AGAGAAATAGTCCCGTCAGC; and 590, 5′-CAGAGCAGCCGATTGTCTGTTG) (8, 57). CFT073PcLupaH was confirmed by PCR amplification and subsequent DNA sequencing. CFT073cam was constructed by inserting the chloramphenicol resistance gene cassette from pKD3 at the lacZ locus using primers with 50-bp homology extensions (1547, 5′-TGATCATTAACTATCCGCTGGATGACCAGGATGCCATTGCTGTGGAAGCTTAGGGCGAATTGACATTGTG; and 1606, 5′-CCGCGCCAATAACAGAATTATCCCCGATGGTGACGCCTGGATTAATAACCCTGGCAGTTCCCTACTCTCG). The mutant was confirmed via PCR and sequencing with the following primers: upstream region (1551, 5′-CCGGTGTCTCTTATCAGACC; and 1177, 5′-TCACCTTCACCCTCTCCACT), downstream region (1638, 5′-CTGGAGTGAATACCACGACG; and 1635, 5′-TCCCGGTCGTTTATTTCGCG).
DNA techniques were performed as previously described (49). Isolation of plasmid DNA was carried out using the QIAprep spin miniprep kit (Qiagen). Restriction endonucleases were used according to the manufacturer's specifications (New England Biolabs). Chromosomal DNA purification was made using the GenomicPrep cell and tissue DNA isolation kit (Amersham Pharmacia Biotech Inc.). Oligonucleotides were purchased from Sigma, Australia. The upaH gene was amplified with the Expand long-template polymerase system (Roche) as described by the manufacturer. DNA sequencing was performed using the ABI Big Dye ver3.1 kit (ABI) by the Australian Genome Research Facility, Brisbane, Australia. Prevalence studies for the upaH gene used NovaTaq DNA polymerase, as described by the manufacturer (Novagen), with the primers 524 (5′-AGTGAAGGGGCAAAAACCTT) and 525 (5′-TGAAACCACCACCATTCTGA). The following program was used: 94°C for 15 s, 50°C for 15 s, and 72°C for 1 min (25×). The orientation of the type 1 fimbrial switch promoter was assessed as described previously (55).
In vitro transposon mutagenesis of the upaH gene cloned into pUpaH was performed with the Epicentre EZ::Tn5<KAN-2> transposon mutagenesis kit as described by the manufacturer. A library of overlapping transposon mutants was generated by mapping the transposon insertion site relative to the start site of upaH. A primer targeted to the end of the transposon was used to sequence into upaH.
For the detection of upaH mRNA by reverse transcriptase PCR (RT-PCR), urine from six infected mice was collected just prior to euthanasia, pooled, and added directly to a two-times volume of RNAprotect (Qiagen). Total bacterial RNA was isolated and purified using an RNeasy minikit (Qiagen). After purification, RNA was treated with RNase-free DNase I (Novagen) to remove contaminating DNA and repurified using Qiagen RNeasy columns. RNA samples were quantified spectrophotometrically at 260 nm and additionally checked by gel electrophoresis. RNA was converted to cDNA using Superscript III first-strand synthesis supermix as described by the manufacturer (Invitrogen Life Technologies). cDNA was used directly as a template for PCR using primers specific for upaH (524 and 525). A negative control using the original RNA was run in parallel to confirm the absence of contaminating DNA.
A 747-bp segment from the passenger-encoding domain of upaH was amplified with the primers UpaHTF (5′-GGCCATGGAAATCGCGGGTAATAACGCGGTGGTGAACCAGGA) and UpaHTR (5′-GGAAGCTTGGCAAATATACTGATGAGTT), digested with NcoI/HindIII, and ligated into NcoI/HindIII-digested pBAD/Myc-HisB. The resultant plasmid (pUpaHTruncated) contained the base pairs 5050 to 5796 of upaH fused to a 6×His-encoding sequence. E. coli BL21 was transformed with the plasmid pUpaHTruncated, induced with arabinose, and the resultant 6×His-tagged protein was purified using the Qiagen nickel-nitrilotriacetic acid (Ni-NTA) spin kit according to the manufacturer's instructions. Protein purity was assessed by SDS-PAGE analysis as previously described (58). Polyclonal anti-UpaH serum was raised in rabbits by the Institute of Medical and Veterinary Sciences (South Australia), following a standard protocol (http://www.imvs.sa.gov.au/). For immunoblotting, whole-cell lysates were subjected to SDS-PAGE using NuPAGE Novex 3 to 8% Tris-acetate precast gels with NuPAGE Tris-acetate running buffer and subsequently transferred to polyvinylidene difluoride (PVDF) microporous membrane filters using the iBlot dry blotting system as described by the manufacturer (Invitrogen). Serum raised against the passenger subunit of UpaH was used as primary serum, and the secondary antibody was alkaline phosphatase-conjugated anti-rabbit IgG. Sigma Fast 5-bromo-4-chloro-3-indolylphosphate-nitroblue tetrazolium (BCIP/NBT) was used as the substrate in the detection process.
Biofilm formation on polystyrene surfaces was monitored by using 96-well microtiter plates (BD Falcon) essentially as previously described (52). Briefly, cells were grown for 24 h in glucose M9 minimal medium (containing 0.1% arabinose for induction of upaH gene expression) at 37°C, washed to remove unbound cells, and stained with crystal violet. Quantification of bound cells was performed by addition of acetone-ethanol (20:80) and measurement of the dissolved crystal violet at an optical density of 595 nm. Flow chamber experiments were performed as previously described (30, 51). Briefly, biofilms were allowed to form on glass surfaces in a multichannel flow system (4 ml/h) that permitted online monitoring of community structures. Flow cells were inoculated with optical density at 600 nm (OD600)-standardized cultures pregrown overnight in glucose M9 minimal medium. For E. coli OS56 biofilms, kanamycin was added as required for maintenance of plasmids. CFT073, CFT073upaH, and CFT073upaHPcL strains were tagged by introduction of the gfp-containing plasmid pSG10, and biofilms were grown as described above but in the presence of chloramphenicol. Biofilm development was monitored by confocal scanning laser microscopy at 24 h or 40 h after inoculation. For analysis of the flow cell biofilms, 16 z-stacks were collected for each strain and analyzed by using the COMSTAT software program (26). All experiments were performed in triplicate.
An anti-UpaH specific serum was used for immunofluorescence microscopy and immunogold electron microscopy as previously described (56, 59). Microscopic observation of biofilms and image acquisition were performed on a scanning confocal laser microscope (LSM510 Meta; Zeiss) equipped with detectors and filters for monitoring of green fluorescent protein (GFP). Vertical cross sections through the biofilms were visualized by using the Zeiss LSM image examiner. Images were further processed for display by using Photoshop software (Adobe, Mountain view, CA).
Female C57BL/6 mice (8 to 10 weeks) were purchased from the University of Queensland Animal Facility and housed in sterile cages with ad libitum access to sterile water. The mouse model of UTI described previously was used for this study (48). Briefly, an inoculum of 20 μl, containing 5 × 108 CFU of bacteria in phosphate-buffered saline (PBS) containing 0.1% India ink, was instilled directly into the bladder using a 1-ml tuberculin syringe attached to the catheter. Urine was collected from each mouse at 18 h after inoculation for quantitative colony counts. Mice were euthanized at 18 h after challenge by cervical dislocation; bladders and kidneys were then excised aseptically, weighed, and homogenized in PBS. Bladder and kidney homogenates were serially diluted in PBS and plated on LB agar for colony counts. Data are expressed as the median number of CFU per 0.1 g of bladder tissue or ml of urine. All animal experiments were repeated at least twice. Competitive mixed-infection assays were performed as described above except that mice were inoculated with a mixture of 2.5 × 108 CFU of CFT073cam and 2.5 × 108 CFU of CFT073upaH grown in the absence of antibiotics. The two strains were differentiated by their resistance to chloramphenicol (CFT073cam) and kanamycin (CFT073upaH), respectively, to enable differential colony counts of the recovered bacteria. Median bacterial CFU were compared using either the nonparametric Mann-Whitney test for the independent challenges or the nonparametric Wilcoxon matched-pairs test for cochallenge infections within the InStat software package (Graphpad, San Diego, CA). P values of <0.05 were considered significant.
Amino acid sequence searches were performed using the BlastP and PSI-BLAST software programs (1) as implemented on the NCBI (National Center for Biotechnology Information) website. Sequence repeats were analyzed using the RADAR (22), REPRO (16), and XSTREAM (41) software programs. Secondary structure was predicted using PSI-PRED software program (v2.6) (6). Fold recognition was performed using the I-TASSER (65) and mGenThreader software (6). The three-dimensional structures were modeled using the program Modeler 9v5 (12). The BetaWrapPro (37) server was used for the detection of beta-helix motifs. Structure figures were prepared using the PyMol software program (DeLano Scientific LLC).
The complete nucleotide sequence of upaH has been deposited in the NCBI database (Bankit number 1182914).
We performed a bioinformatic analysis of the E. coli CFT073 genome to search for novel putative adhesins and identified a 6.5-kb unannotated intergenic region that contains a disrupted copy of a previously unidentified AT-encoding gene (Fig. (Fig.1A).1A). At the corresponding position on the E. coli K-12 MG1655 genome, there is a large pseudogene (8622 nucleotides), called ydbA, that has been truncated by insertion sequences (Fig. (Fig.1A).1A). Were it not for these insertion sequences, ydbA would encode a full-length AT protein. An intact positional orthologue is also present in the genome sequence of UPEC strain UTI89, although it has been misannotated as a major facilitator superfamily-type transporter (Fig. (Fig.1A).1A). In the published genome sequence of E. coli CFT073, a frameshift is reported in the ydbA gene; however, this may be a sequencing artifact since there is a section of uncertainty (CAAGGCGGTYGTSAAYAAA) in this region. Close examination of the E. coli CFT073 genome suggests that the repeat region of ydbA was misassembled during genome closure and an intact AT-encoding gene may actually be present. In accordance with the terminology adopted for other AT-encoding genes of UPEC (45), we have termed the ydbA gene upaH. PCR amplification of the entire upaH gene from E. coli CFT073 revealed a product that was approximately 2 kb larger than that expected based on the published genome sequence, confirming an error in genome assembly. In order to sequence the upaH gene, the upaH PCR product was cloned as a transcriptional fusion downstream of the tightly regulated araBAD promoter in the pBAD/Myc-HisA expression vector to generate the plasmid pUpaH. In vitro transposon mutagenesis was employed to generate a library of overlapping transposon mutants, which enabled sequencing across the repetitive region of the gene. This revealed that, in contrast to the published genome sequence, the E. coli CFT073 upaH gene is an intact coding region of 8.535 kb in length.
Analysis of the translated upaH sequence using the Conserved Domain Database (CDD) search (36) detected the AT β-domain (translocation domain) in the C-terminal region of the protein (Fig. (Fig.1B).1B). Sequence searches against the Protein Data Bank revealed no significant similarities even when searching with a position-specific scoring matrix (PSSM) based on PSI-BLAST searches of the nonredundant protein sequences database. The known crystal structures of AT β-domains include NalP from Neisseria meningitidis (43) and EspP from E. coli (3). The similarity of both of these structures with the β-domain of UpaH could be detected by the fold recognition algorithm I-TASSER (65). Alignment of the UpaH and NalP β-domain sequences with ClustalW (34) shows 15% identity. We modeled the structure of this UpaH domain (residues 2528 to 2845) with the software program Modeler (12) using the NalP structure as a template (Fig. (Fig.1C).1C). The model predicts a central helix (residues 2530 to 2556) winding through a 12-strand β-barrel. The periplasmic β-barrel loops are short and well defined in the model; on the other hand, the extracellular loops are longer and show more variability in different models created by Modeler, suggesting they are less well defined in the model (in particular the loops connecting β-barrel strands 1 and 2 [residues 2573 to 2579], 3 and 4 [residues 2616 to 2624], 5 and 6 [residues 2673 to 2679], and 9 and 10 [residues 2783 to 2802]). The large passenger domain of UpaH contains ~50 imperfect tandem sequence repeats (Fig. (Fig.1D).1D). The smallest repeating unit alternates in length between 24 and 25 amino acids, and larger-length repeats are evident that are multiples of 24 to 25 (98, 147, 196, etc). It appears that the sequence of this domain may have evolved through a number of sequential duplications involving fragments containing variable numbers of repeats. The predominant secondary structure predicted in this domain is β-structure, and a repeating pattern of three β-sheets per sequence repeat can be observed.
The widespread occurrence of AT-encoding genes in E. coli prompted us to investigate the prevalence of upaH in UPEC isolates from our laboratory collection, as well as the 72 strains from the E. coli reference (ECOR) collection. For this purpose, we employed primers designed to amplify the region encompassing nucleotides 8083 to 8410 of upaH. We found that the correct-size base pair product was amplified from 90/118 (76%) of the UPEC strains (15/21 cystitis strains, 28/31 pyelonephritis strains, and 47/66 asymptomatic bacteriuria [ABU] strains), as well as 65/72 (90%) of the ECOR strains. The presence of upaH as assessed by PCR was also determined for these strains according to their phylogenetic group and revealed the following prevalence pattern: group A, 30/35 (86%); B1, 17/20 (85%); B2, 80/95 (84%); and D, 23/33 (70%). These results suggest that the upaH gene is widely distributed among E. coli strains.
To demonstrate expression of the UpaH protein, plasmid pUpaH was transformed into the previously described E. coli K-12 flu mutant strain MS427 (53). This strain is unable to mediate the classical cell aggregation and biofilm phenotypes associated with Ag43 expression (46). UpaH expression was detected using a rabbit polyclonal antiserum targeting a region of UpaH comprising amino acids 1424 to 1673 of the N-terminal passenger domain. Western blot analysis employing anti-UpaH antibodies detected the UpaH protein in whole-cell lysates prepared from E. coli MS427(pUpaH) following induction with arabinose. The size of the UpaH protein was approximately 460 kDa, significantly larger than its expected molecular mass of 290 kDa (Fig. (Fig.2A).2A). Such a disparity in mass has been observed previously for proteins containing repeated acidic sequences, including the AT protein ShdA from Salmonella enterica (29). UpaH is also highly acidic, containing 376/2,845 (13.2%) acidic amino acids (glutamate and aspartate). The overexpression of UpaH also resulted in multiple smaller bands presumed to be breakdown products (Fig. (Fig.2A2A).
Our bioinformatic analysis suggested that upaH encodes an AT protein and thus UpaH may localize to the outer membrane. To test this tenet, we performed immunofluorescence microscopy (Fig. (Fig.2B).2B). UpaH antiserum readily reacted with E. coli MS427(pUpaH) following induction with arabinose during growth in LB broth, demonstrating that the N-terminal region of UpaH was effectively translocated to the cell surface. No reaction was seen with E. coli MS427(pBAD) cells. The surface location of UpaH was confirmed using immunogold labeling and electron microscopy (Fig. (Fig.2C).2C). E. coli MS427(pUpaH) cells displayed abundant surface labeling with gold particles following incubation with UpaH antiserum compared to results for E. coli MS427(pBAD) cells. Thus, the upaH gene is functional and encodes a large protein located at the cell surface in E. coli.
The presence of multiple repeats within the predicted passenger-encoding domain of UpaH suggested it may possess some functional properties similar to those of the well-characterized Ag43 and AIDA-I AT proteins. However, we saw no evidence of UpaH-mediated cell aggregation in E. coli MS427 harboring pUpaH (data not shown). UpaH overexpression also failed to mediate adherence to T24 bladder epithelial cells or to the extracellular matrix proteins collagen I, collagen III, collagen IV, fibronectin, and laminin (data not shown). To determine whether the UpaH protein promotes biofilm formation, we examined the effect of UpaH overexpression in E. coli MS427 (or the gfp-tagged derivative strain OS56) using static and dynamic biofilm assays. First, we employed a static nontreated polystyrene microtiter plate model and showed that UpaH promotes biofilm formation after growth in M9 minimal medium (Fig. (Fig.3A).3A). Next, we tested the ability of UpaH to promote biofilm formation under dynamic conditions using the continuous-flow-chamber model, which permits monitoring of the bacterial distribution within an evolving biofilm at the single-cell level due to the combination of GFP-tagged cells and scanning confocal laser microscopy. UpaH promoted strong biofilm growth under these conditions and produced a structure with a depth of approximately 10 μm (Fig. (Fig.3B3B).
To determine whether UpaH is expressed in E. coli CFT073, we constructed a upaH deletion mutant employing λ-red-mediated homologous recombination of linear DNA (referred to as CFT073upaH). Examination of whole-cell lysates prepared from these isogenic strains following growth in LB broth by SDS-PAGE and Western blotting demonstrated expression of UpaH in E. coli CFT073 but not in CFT073upaH (Fig. (Fig.4A).4A). Furthermore, complementation of CFT073upaH with plasmid pUpaH restored the ability of this strain to produce UpaH (albeit with evidence of protein degradation) (Fig. (Fig.4A).4A). We also randomly selected 16 UPEC strains positive in the upaH PCR screen and tested them for the ability to express the UpaH protein following growth in LB by Western blot analysis employing UpaH-specific antiserum. A band corresponding to the UpaH protein was identified in all of these strains (data not shown). To confirm that this was in fact UpaH, we deleted the upaH gene in two strains (i.e., M161 and M357) using λ-red-mediated homologous recombination. Whole-cell lysates prepared from E. coli M161upaH and M357upaH failed to react with the UpaH-specific antiserum (Fig. (Fig.4B).4B). Thus, UpaH is a newly identified AT protein expressed by E. coli CFT073 and other UPEC isolates. We also tested for surface localization of UpaH on E. coli CFT073 using immunofluorescence microscopy; however, we could not detect any expression. This was not caused by a defect in the ability to transport UpaH to the outer membrane, since surface localization of UpaH was detected in CFT073upaH(pUpaH) and a UpaH overexpression strain constructed by inserting a chromosomally located constitutive promoter upstream of the upaH coding region (CFT073PcLupaH) (Fig. (Fig.4C4C).
The contribution of UpaH to biofilm formation by E. coli CFT073 was assessed, employing both microtiter plate and flow-cell-based assays. In the microtiter plate assay, E. coli CFT073 produced a significantly increased biofilm compared to E. coli CFT073upaH (Fig. (Fig.5A).5A). We also examined biofilm formation by the E. coli M161/M161upaH and M357/M357upaH isogenic strain sets. E. coli M161 and M357 did not form a strong biofilm in this assay; there was no significant difference in the biofilm formed by E. coli M161upaH and M357upaH (Fig. (Fig.5A).5A). To characterize the function of UpaH in E. coli CFT073 further, we tested the ability of E. coli CFT073, CFT073upaH, and CFT073PcLupaH to promote biofilm formation in a continuous-flow-chamber system. In this assay, CFT073upaH formed a biofilm with a lower biovolume, substratum coverage, and mean thickness than those of the biofilms formed by E. coli CFT073 and CFT073PcLupaH (P = 0.008, P = 0.008, and P = 0.012, respectively) (Fig. 5B and C). Taken together, these results demonstrate that UpaH contributes to biofilm formation by CFT073.
To study the role of UpaH in virulence, we examined the ability of CFT073 and CFT073upaH to survive in the mouse urinary tract. There was no significant difference in the number of CFT073 or CFT073upaH cells recovered from the bladders and urine of mice following infection (data not shown). No significant colonization of the kidneys was observed for either strain; this is consistent with previous data from our laboratory using C57BL/6 mice (57, 59). To examine if this was a strain-specific effect, we also compared E. coli M161/M161upaH and M357/M357upaH for their ability to colonize the mouse urinary tract as described above. There was no significant difference in colonization of the bladder or recovery from urine for the parent and upaH mutant of both isogenic strain sets (data not shown); no significant colonization of the kidneys was observed for any of the strains.
The above result prompted us to examine whether upaH is transcribed during infection of C57BL/6 mice by CFT073 using RT-PCR. RNA was extracted from pooled mouse urine collected 18 h postinoculation and converted to cDNA. A PCR using upaH-specific primers revealed the presence of an amplification product from the cDNA samples, while no PCR product was obtained from the RNA sample prior to cDNA synthesis (Fig. (Fig.6A).6A). Thus, the upaH gene located on the E. coli CFT073 chromosome is transcribed during colonization of the mouse urinary tract.
For a more sensitive method to test the role of upaH in colonization of the mouse urinary tract, we performed competitive-colonization experiments where we coinoculated mice with the CFT073 and CFT073upaH strains in a 1:1 ratio. These experiments were performed using a derivative of CFT073 tagged at the lacZ locus with a chloramphenicol resistance gene cassette (i.e., CFT073cam) and CFT073upaH (which contains a kanamycin resistance gene cassette in upaH). The two strains had an identical growth rate in pooled human urine (and LB broth) and did not display any difference with respect to type 1 fimbria production (as assessed by yeast cell agglutination and a fim switch orientation PCR), hemolysis of red blood cells, and motility (data not shown). We found that CFT073cam significantly outcompeted CFT073upaH at 18 h in urine (P < 0.001) and the bladder (P < 0.001) (Fig. (Fig.6B).6B). In contrast, we did not observe a significant difference in colonization using mixed-infection experiments employing M161/M161upaH or M357/M357upaH (data not shown). These results suggest that the ability to express UpaH confers a competitive advantage for the survival of CFT073 in the mouse urinary tract but that this effect is not specific to all strains of UPEC.
UPEC produces a range of fimbrial and nonfimbrial adhesins that play a role in virulence and contribute to persistent infection of the urinary tract. Some UPEC fimbrial adhesins, including type 1 and P fimbriae, have been well characterized with respect to their expression, regulation, receptor-binding target, and role in virulence. Of the nonfimbrial adhesins, the AT family of proteins represents a novel group of virulence factors because of their role in adhesion, invasion, and biofilm formation in other organisms. Although UPEC strains possess multiple AT-encoding genes, very little is known about their function, and only Ag43, UpaG, and Sat have been shown to play a role in virulence. Here we identify and characterize UpaH as a new AT protein of UPEC and demonstrate that it contributes to biofilm formation and colonization of the mouse bladder in a mixed-infection assay.
The upaH gene is not annotated in the genome sequence of E. coli CFT073 due to a misassembly within a region of highly repetitive DNA. In E. coli K-12, a large pseudogene (8,622 bp), called ydbA, that has been truncated by insertion sequences is located at the corresponding position on the chromosome. Positional orthologues of ydbA are also discernible in several other pathogenic and nonpathogenic E. coli strains, although they are not always intact (e.g., E. coli O157:H7 EDL933 z2322/z2323 represent a split-gene version of ydbA). In the UPEC strain UTI89 genome, ydbA has been misannotated as encoding the “EntS/YbdA major facilitator superfamily (MFS) transporter.” This unfortunate error likely arose because the EntS MFS enterobactin transporter described by Furrer et al. (14) was encoded by the gene ybdA. Similarity searches with the true EntS sequence (gi|2506613) clearly demonstrate that the UTI89 EntS protein and the original EntS/YbdA protein are not related proteins. The combination of events that led to the misannotation of upaH emphasizes the need for caution when assembling highly repetitive regions of genome sequence data, especially since in some cases these repeats may occur within genes that encode cell surface proteins. The “genome rot” caused by the UTI89 EntS error (already propagated to three recent E. coli genome annotations) also highlights the need for caution when predicting function based on annotation alone, even when the annotation has been manually curated.
The passenger domain of UpaH contains 50 imperfect repeats, each of which comprises 24 to 25 amino acid residues. The solved structures of passenger domains from other AT proteins, including pertactin from Bordetella pertussis (10), hemoglobin protease from E. coli (44), and pore-forming toxin from Helicobacter pylori (15), all contain the β-helix motif. The β-helix structure contains a superhelical arrangement of structural repeats, each consisting of three β-strands in a triangular arrangement (64). These repeats pack into a cylindrical solenoid structure, with the β-strands forming parallel β-sheets along the cylinder. The repeating sequence pattern in the UpaH passenger domain is consistent with structural repeats found in a β-helix. Despite the lack of detectable sequence similarities, we therefore propose that the bulk of the passenger domain would correspond to a β-helix structure. Several other pieces of evidence support this prediction: (i) the program BetaWrapPro (37), which specializes in identifying β-helix motifs, identifies several such motifs in UpaH; (ii) the fold recognition programs, such as I-TASSER (65) and mGenThreader (6), suggest the sequence of the UpaH passenger domain is most compatible with the structures of AT passenger domains and other β-helix proteins; and (iii) most AT passenger domains have been suggested previously to contain the β-helix structure, presumably to facilitate efficient secretion across the outer membrane and subsequent folding and to present the functional determinant at a distance from the bacterial surface (28).
Short repeat sequences have been described within the passenger domain of two other surface-located AT proteins from E. coli, namely, AIDA-I (4) and Ag43 (25). AIDA-I contains 21 repeats built up of 19 to 20 amino acid residues followed, after a short interruption, by an additional 10 repeats. The repeat structure of Ag43 is less well defined and more variable in length (12 to 20 amino acids). Repeat sequences often form solenoid structures with frequent roles in protein-protein interactions (32) and possibly ligand binding (61), both of which are consistent with the surface location of UpaH (and AIDA-I and Ag43) and the ability of these proteins to mediate biofilm formation. Other possible functions of short repeat sequences may be to increase flexibility, alter the length, incorporate repetitive domain architecture, and mediate antigenic variation via duplication or deletion. The precise biological role of these short repeat sequences in UpaH and their level of conservation in UPEC remain to be determined.
In the murine UTI model, UPEC colonization of the bladder is associated with the formation of intracellular bacterial communities (IBCs) within superficial umbrella cells (2). IBCs possess biofilm-like properties and are comprised of compact cell clusters encased in a polysaccharide matrix. IBC formation creates a quiescent state that may be associated with long-term persistence in the bladder (11, 40). Two cell surface factors associated with biofilm formation, type 1 fimbriae, and Ag43 are expressed by cells within IBCs (2). We observed a significant difference in biofilm formation between CFT073 and CFT073upaH. However, this difference did not translate into a significant role for UpaH in the mouse urinary tract in single-infection experiments. To investigate this further, we also constructed a upaH deletion in two other clinical isolates from our laboratory collection that expressed UpaH (i.e., M161 and M357). There was no significant difference in bladder colonization or biofilm growth for M161/M161upaH or M357/M357upaH. Despite these results, we did observe a competitive advantage for CFT073 for colonization of the urinary tract in mixed-infection experiments with CFT073upaH. It is possible that UpaH and other AT proteins (e.g., Ag43 and UpaG) possess a high degree of functional redundancy with respect to the biofilm phenotype. The apparent strain-specific effect of UpaH in CFT073 versus M161/M357 may be associated with their different biofilm formation properties. Other surface factors differentially expressed between CFT073, M161, and M357 (e.g., fimbriae, flagella, and capsule) could also interfere with the function of UpaH, as has been reported previously for Ag43 (21, 50, 58). Thus, our in vivo analysis of the function of UpaH suggests it may play a modulatory role in colonization of the mouse urinary tract by CFT073 and that this contribution could occur in concert with other AT proteins (e.g., Ag43 and UpaG) that possess similar biofilm formation properties.
In a recent study, Hagan and Mobley identified 23 outer membrane proteins from E. coli CFT073 that elicit a humoral response in the mouse UTI model (20). This included porins, transporters, and receptors (including siderophore receptors). There are several possible reasons for the failure to detect UpaH in this screen. First, it is possible that as a result of its very large size, the UpaH protein did not migrate properly into the gel. We note that despite an apparent molecular mass of 290 kDa, UpaH migrated at approximately 460 kDa. Second, UpaH may be expressed in vivo at low levels. Although we were able to detect the presence of a upaH transcript by RT-PCR, the exact level of UpaH expression in vivo remains to be determined. Finally, it is possible that UpaH does not elicit a strong humoral response. However, the highly repetitive sequence of the passenger domain and its similarity to other AT proteins would suggest otherwise. The UpaH homologues present in the published genomes exhibit some variation in size, which may reflect the natural variation in the number of repeats. Alternatively, these genomes may contain incorrect assemblies due to the challenges of sequencing repetitive regions.
This work was supported by grants from the National Health and Medical Research Council (455914 and 631654). S.A.B. is supported by an ARC Australian Research Fellowship (DP0881247).
We thank Ruth Lee and Chelsea Stewart for expert technical assistance and Sharon Goh for providing plasmid pSG10.
Editor: J. B. Bliska
Published ahead of print on 9 February 2010.