J Med Microbiol Track the topics, authors and articles important to you
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lima, A.
Right arrow Articles by Piccini, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lima, A.
Right arrow Articles by Piccini, C.
Agricola
Right arrow Articles by Lima, A.
Right arrow Articles by Piccini, C.
J Med Microbiol 56 (2007), 1600-1607; DOI: 10.1099/jmm.0.47320-0
© 2007 Society for General Microbiology
ISSN 1473-5644

An iron-regulated outer-membrane protein of Proteus mirabilis is a haem receptor that plays an important role in urinary tract infection and in in vivo growth

Analía Lima, Pablo Zunino, Bruno D'Alessandro and Claudia Piccini

Laboratorio de Microbiología, Instituto de Investigaciones Biológicas Clemente Estable (IIBCE), Av. Italia 3318, CP 11600, Montevideo, Uruguay

Correspondence
Claudia Piccini
piccini{at}iibce.edu.uy

Received 29 March 2007
Accepted 15 August 2007

Proteus mirabilis, a common cause of urinary tract infections, expresses iron-regulated outer-membrane proteins (OMPs) in response to iron restriction. It has been suggested that a 64 kDa OMP is involved in haemoprotein uptake and that this might have a role in pathogenesis. In order to confirm this hypothesis, this study generated a P. mirabilis mutant strain (P7) that did not express the 64 kDa OMP, by insertion of the TnphoA transposon. The nucleotide sequence of the interrupted gene revealed that it corresponded to a haemin receptor precursor. Moreover, in vitro growth assays showed that the mutant was unable to grow using haemoglobin and haemin as unique iron sources. The authors also carried out in vivo growth and infectivity assays and demonstrated that P7 was not able to survive in an in vivo model and was less efficient than wild-type strain Pr 6515 in colonizing the urinary tract. These results confirmed that the P. mirabilis 64 kDa iron-regulated OMP is a haem receptor that has an important role for survival and multiplication of these bacteria in the mammalian host and in the development of urinary tract infection.

Abbreviations: Amp, ampicillin; EDDHA, di-o-hydroxyphenylacetic acid; GaPPIX, gallium protoporphyrin IX; Hb, haemoglobin; Hm, haemin; IPC, intraperitoneal chamber; Km, kanamycin; OMP, outer-membrane protein; Tet, tetracycline; UTI, urinary tract infection.


   INTRODUCTION
TOP
 INTRODUCTION
METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Iron plays a central role in bacterial metabolism and pathogenesis. In the mammalian host, iron is unavailable for bacterial uptake (Litwin & Calderwood, 1993) as most of it is located intracellularly, stored in ferritin or linked to haem or haem-containing compounds, whilst extracellular iron is bound to transferrin, lactoferrin, haemopexin and haptoglobin (Braun, 2001).

To overcome this problem, Gram-negative pathogenic bacteria have evolved iron-regulated systems to acquire this element in the host. These include the direct binding of iron-containing proteins to outer-membrane receptors and the secretion of siderophores or haemophores (Wandersman & Stojiljkovic, 2000).

Transport of haem, iron-binding proteins and Fe3+–siderophore complexes towards the cytoplasm represents an active process. Bacteria have developed a system formed by an outer-membrane receptor and a cytoplasmic membrane complex composed of ExbB, ExbD and TonB proteins that transduces the proton-motive force of the cytoplasmic membrane to the high-affinity outer-membrane receptors, known as TonB-dependent receptors (Olczak et al., 2005; Llamas & Bitter, 2006).

Proteus mirabilis is an important cause of urinary tract infections (UTIs) in humans, especially in patients with urinary catheters and structural abnormalities. It shows a predilection for the upper urinary tract where it can cause serious renal damage (Stamm et al., 1977; Senior, 1978). Several potential virulence factors have been described for P. mirabilis, including haemolysin production and iron-uptake systems (Ròzalski et al., 1997).

Piccini et al. (1998) studied the effect of iron deprivation on the expression of outer-membrane proteins (OMPs) in uropathogenic isolates of P. mirabilis. They found that under low-iron conditions, expression of three OMPs with molecular masses ranging from 66 to 75 kDa increased and that a 64 kDa OMP was synthesized de novo. They also suggested that the 64 and 66 kDa OMPs were involved in haemin (Hm) and haemoglobin (Hb) uptake. Moreover, Burall et al. (2004) identified four P. mirabilis ORFs that are involved in iron uptake as potential in vivo-expressed virulence determinants, three of which are potential OMPs.

Here, we describe the generation and characterization of a P. mirabilis insertion mutant lacking the 64 kDa OMP, and the role of this protein in bacterial growth, in vivo survival and infectivity. Growth assays using different iron sources were performed to evaluate the ability of the mutant strain to grow using Hm or Hb as the sole iron source. Moreover, we studied the role of the 64 kDa OMP in the pathogenicity of P. mirabilis through in vivo experimental UTI models in mice and during in vivo growth using an intraperitoneal chamber (IPC) model.


   METHODS
TOP
 INTRODUCTION
 METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Bacterial strains, plasmids, transposon and growth conditions. Bacterial strains, plasmids and the transposon used in this study are listed in Table 1Go. Bacterial strains were grown in Luria–Bertani (LB) broth, LB agar, nutrient broth (NB) or nutrient agar (NA). When necessary, the media were supplemented with 100 µg ampicillin (Amp) ml–1, 15 µg tetracycline (Tet) ml–1 and 40 µg kanamycin (Km) ml–1. Iron-limited conditions were obtained by adding di-o-hydroxyphenylacetic acid (EDDHA) to the medium at a final concentration of 100 µM.


View this table:
[in this window]
[in a new window]

  		Table 1. Bacterial strains, plasmids and transposon used in this study

 
Animals. Female CD-1 outbred mice weighing 20–25 g and Sprague–Dawley outbred rats weighing 220–250 g from the breeding facilities at IIBCE, Uruguay, were provided with food pellets and tap water ad libitum. All animal experiments were conducted in accordance with procedures authorized by IIBCE.

Transposon mutagenesis and isolation of mutant strains. TnphoA was introduced into the P. mirabilis strain Pr 6515 chromosome by conjugation. To this end, Escherichia coli SM10{lambda}pir was mated with Pr 6515 for 6 h at 37 °C on LB agar plates supplemented with 100 mM MgCl2. Bacteria were scraped from the plate, suspended in PBS and plated on LB agar (lacking NaCl and solidified with 2 % agar to inhibit swarming motility) supplemented with Tet, Km, 40 µg 5-bromo-4-chloro-3-indolyl phosphate ml–1 and 2 µg gallium protoporphyrin IX (GaPPIX) ml–1. Plates were incubated overnight at 37 °C and selected clones were restreaked on Amp plates to discard clones containing plasmid co-integrates.

OMP extraction. P. mirabilis Pr 6515 and the mutant strain P7 were inoculated into NB and NB-EDDHA and incubated at 37 °C for 24 and 48 h, respectively. OMP extraction was performed as described by Piccini et al. (1998).

SDS-PAGE. OMP-enriched pellets were suspended in electrophoresis loading buffer and heated at 100 °C for 5 min. The samples were analysed by SDS-PAGE using a 10 % acrylamide resolving gel according to the method described by Sambrook et al. (1989) and stained with 0.1 % Coomassie brilliant blue.

Chromosomal DNA extraction and Southern blotting. Clones P3, P7, P14 and wild-type strain Pr 6515 were analysed by Southern blot hybridization using the entire transposon as a probe. Chromosomal DNA from each strain was obtained using an agarose-embedded DNA extraction procedure, as described by Maskell et al. (1988), and was subjected to simple digestion with the restriction enzyme BamHI or HindIII and double digestion with both enzymes. The restriction reactions were incubated overnight at 37 °C. Southern blot analysis was carried out according to Sambrook et al. (1989). Hybridization was carried out overnight at 55 °C using AlkPhos Direct hybridization buffer (Amersham) and the TnphoA transposon was labelled with AlkPhos Direct (Amersham) when used as a probe. The expected hybridization pattern for a single TnphoA insertion was two bands for simple digestion with BamHI and three bands for simple digestion with HindIII (one of about 3.5 kb, the expected size of the internal fragment). In the double digestion, the expected hybridization pattern was five bands, two having approximate sizes of 1.5 and 2 kb.

DNA amplification, sequencing and analysis. In order to identify the mutated gene, TnphoA flanking regions were amplified by two rounds of PCR according to the protocol described by Gallagher & Manoil (2001). Briefly, P. mirabilis Pr 6515 and P7 chromosomal DNA was obtained and a first-round PCR was performed using the primer pair Tn1 (5'-GTGCAGTAATATCGCCCTGAG-3', nt 93–73 of TnphoA) and CEKG2C (5'-GGCCACGCGTCGACTAGTACNNNNNNNNNNGATAT-3', a partially random primer that hybridizes at different sites in the P. mirabilis genome) for the 5' end. For amplification of the 3' end, Tn2 (5'-CGATGAAGAGCAGAAGTTATC-3', nt 7620–7640 of the TnphoA gene) and CEKG2C were used. The annealing temperature was 37 °C for both pairs of primers. PCR products obtained from both strains were compared in an agarose gel, and a band present in P7 but not in Pr 6515 was excised, purified with a Concert Rapid Gel Extraction System kit (Life Technologies) and used as a template for the second-round PCR. For the second round, Tn1 or Tn2 was used in combination with CEKG4 (5'-GGCCACGCGTCGACTAGTAC-3'), which hybridizes to the sequence introduced at the 5' end of primer CEKG2C, using an annealing temperature of 52 °C. The obtained PCR product was purified using a Concert Rapid Gel Extraction System kit. DNA sequencing was performed at the Centro Técnico de Análisis Genéticos, Facultad de Ciencias, Universidad de la República, Uruguay, using an Applied Biosystems Model 377 Automated DNA sequencer. The obtained transposon flanking sequences were searched in the genome sequence from P. mirabilis strain HI4320. These sequence data were produced by the P. mirabilis Sequencing Group at the Sanger Institute and are available at ftp://ftp.sanger.ac.uk/pub/pathogens/pm. Sequence analysis was carried out using Artemis software (Rutherford et al., 2000). The DNA sequence was also searched in the protein sequences derived from the patent division of GenBank (PAT) database at NCBI using the BLASTN tool.

Growth curves. Pr 6515 and P7 were incubated overnight in NB-EDDHA at 37 °C in order to promote the consumption of intracellular iron. Cells were then suspended in PBS and inoculated at a final concentration of 1x105 c.f.u. ml–1 into the following media: NB, NB-EDDHA, NB-EDDHA supplemented with 5 mg Hm ml–1 and NB-EDDHA supplemented with 5 mg Hb ml–1. These cultures were incubated at 37 °C with vigorous shaking. Samples were taken at different times and the number of c.f.u. ml–1 was determined by bacterial plate counts on NA at 37 °C. Each growth assay was repeated at least three times.

Growth assays with different iron sources. The ability of Pr 6515 and P7 to grow using different iron sources was assessed. Thus 1x105 c.f.u Pr 6515 or P7 ml–1 from NB-EDDHA overnight cultures was seeded onto NA-EDDHA. Sterile paper discs were placed onto the plates and soaked with 15 µl of one of the following iron sources: 5 mg Hb ml–1, 5 mg Hm ml–1 or 50 mM ferric citrate. A disc soaked in sterile water was used as a negative control. Plates were incubated at 37 °C for 48 h and the diameters of the growth zones around the discs were determined. Each growth assay was repeated at least three times.

Bacterial growth in urine. To evaluate the ability of the strains to grow in human urine, urine samples from five healthy female donors were pooled and filter-sterilized. P. mirabilis strains Pr 6515 and P7 from overnight cultures on NB-EDDHA at 37 °C were inoculated in 20 ml fresh urine at a final density of 1x104 c.f.u. ml–1. Samples were taken at different times and cells were counted after overnight culture on NA (Zunino et al., 1999). Each growth assay was repeated at least three times.

Haemolysis assays. Evaluation of the haemolytic activity of Pr 6515 and P7 was performed as described previously (Mobley & Chippendale, 1990; Zunino et al., 1999). The haemolytic titre was defined as the highest dilution in which no visible pellet was observed.

Bacterial growth in IPCs in rats. Growth of Pr 6515 and P7 in IPCs was monitored. Diffusion chambers were prepared as described previously (Zunino et al., 1999). Two chambers were introduced through a 1 cm longitudinal incision into the peritoneal cavity of rats anaesthetized previously by intramuscular injection of ketamine hydrochloride (88 mg kg–1) and xylazine (4 mg kg–1). The abdomen was sutured with 0000 vicryl and the skin with 000 nylon (Quessy et al., 1994). After 12, 24, 48 and 72 h, the rats were killed (three rats per sampling point) and the chambers were removed aseptically. Bacteria were removed from the chamber using a syringe and viable bacterial counts were determined by plating serial dilutions on NA.

P. mirabilis UTI models in mice. The ascending UTI mouse model was performed as described previously by Zunino et al. (2001). Briefly, Pr 6515 and P7 bacteria were suspended in PBS at a concentration of 1x108 c.f.u ml–1. Mice were anaesthetized with xylazine (10 mg kg–1) and ketamine (50 mg kg–1) and their bladders were emptied by gentle abdominal massage. A soft polyethylene catheter (outer diameter 0.61 mm) was inserted through the urethra and 0.05 ml bacterial suspension (5x106 c.f.u. per mouse) was slowly introduced into the bladder over a period of not less than 30 s to avoid vesicourethral reflux. Moreover, in order to observe the ability of both strains to compete in vivo, a co-challenge experiment was performed. For these assays, mice were treated as described above and a group of mice was challenged with a 1 : 1 mixture of wild-type Pr 6515 and mutant P7.

When the haematogenous model was used, bacterial suspensions (0.1 ml, 1x107 c.f.u. per mouse) were injected into a tail vein (Zunino et al., 1994).

In every experiment, mice were sacrificed 7 days after infection by cervical dislocation. Kidneys and bladder were removed aseptically and homogenized in 10 ml PBS, and viable bacterial counts were determined on NA. In transurethral co-challenge experiments, recovery of the mutant strain from the bladder and kidneys of infected mice was determined by viable bacterial counts on NA-Km, whereas recovery of the wild-type strain was determined as the difference between the counts on NA and NA-Km. In haematogenous infection assays, animals were recorded as infected when bacteria were recovered from one or both kidneys.

Statistical analysis. Mann–Whitney one-tailed non-parametric analysis was used to compare bacterial colonization levels in kidney and bladder. A {chi}2 test with Yates correction for small sample sizes was used to compare the number of animals infected with Pr 6515 and P7. Differences were considered significant for P <0.05.


   RESULTS AND DISCUSSION
TOP
 INTRODUCTION
 METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Generation and selection of a P. mirabilis mutant strain deficient in a haem receptor

The mechanism of Hm or Hb utilization as an iron source by P. mirabilis, the protein receptors involved in iron uptake and their participation in pathogenesis are poorly understood. In a previous study, Piccini et al. (1998) demonstrated that P. mirabilis could use Hm and Hb as unique iron sources and they suggested that the 64 kDa OMP was involved in Hm and Hb uptake. In order to confirm the role of the 64 kDa OMP as the Hm or Hb receptor, in the present study we used the clinical P. mirabilis strain Pr 6515 to generate a mutant strain that did not express the 64 kDa OMP.

Mutant strains were generated by insertion of the TnphoA transposon, which was transferred by conjugation through a suicide plasmid. TnphoA was randomly inserted in the chromosome generating fusions between target genes and the E. coli gene for alkaline phosphatase. This genetic tool was chosen as it allowed simple selection and enrichment of mutant strains of genes for secreted, transmembrane and periplasmic proteins by selecting those clones that showed alkaline phosphatase activity (Manoil & Beckwith, 1985).

Selection of transconjugant clones was also based on their resistance to GaPPIX, a structural analogue of the haem group but with gallium instead of iron conjugated to the protoporphyrin group. Gallium is not able to substitute for iron in biological reactions and for that reason this compound is toxic to bacteria that use haem or haemoprotein uptake systems, causing cell death by the production of reactive oxygen radicals. Therefore, mutant strains lacking haem or haemoprotein receptors would be resistant to GaPPIX (Stojiljkovic et al., 1999). After conjugation, those insertion mutants that were able to grow on the selective media were chosen. This strategy allowed the isolation of 15 GaPPIX-resistant clones, but just three (P3, P7 and P14) were sensitive to Amp. These results showed a high frequency of plasmid co-integration at the transposon insertion site. This phenomenon has been reported by other authors (Ali et al., 2000) and is one of the obstacles described for the use of this Tn5-derived transposon (Taylor et al., 1989; Ahmed, 1991).

Among the three Amp-sensitive selected clones, only P7 showed blue colonies when 5-bromo-4-chloro-3-indolyl phosphate was used as the alkaline phosphatase substrate. Moreover, when the OMP patterns of Pr 6515 and mutant strain P7 grown on iron-restricted media were analysed by SDS-PAGE (Fig. 1Go), it was observed that the selected clone lacked the band corresponding to the 64 kDa OMP, indicating that P7 had the transposon inserted in the structural gene encoding the 64 kDa OMP.


Figure 1
View larger version (50K):
[in this window]
[in a new window]

  		Fig. 1. OMP profile of Pr 6515 (a) and the 64 kDa OMP-deleted mutant P7 (b). OMPs were obtained from bacteria grown in NB (lane 1) and in NB supplemented with 100 µM EDDHA (lane 2). IROMPs, iron-regulated outer-membrane proteins.

 
Genetic analysis of the P. mirabilis mutant lacking the 64 kDa OMP

Clones P3, P7 and P14 were analysed by Southern blot hybridization using the entire transposon as a probe. The results showed that P7 exhibited the expected hybridization pattern for a single TnphoA insertion, whilst the hybridization pattern of P3 and P14 mutant strains did not show a unique insertion of the transposon (data not shown). The wells corresponding to Pr 6515 did not show any signal. For these reasons, P7 was selected for further characterization studies.

The DNA sequence of the P7 mutated gene was obtained by semi-random PCR of chromosomal DNA (Gallagher & Manoil, 2001). When the 5' end of the transposon sequence was analysed, we always found TnphoA-related sequences. It has been shown that Tn5, from which TnphoA derives, transposes by a replicative pathway and promotes IS50 insertions frequently accompanied by inversions of adjacent DNA (Ahmed, 1991). This particular behaviour of Tn5 and derivatives could explain why DNA sequences obtained from the 5' end of the transposon corresponded to TnphoA sequences.

The DNA sequence obtained from the transposon 3' end was compared with the P. mirabilis genome sequence and was found to match the predicted gene PMI1426, which encodes a Hm receptor (M. Sebaihia, personal communication). The 3' TnphoA flanking sequence was also compared with the PAT database at NCBI using the BLASTN tool, which recovered the sequence 310 from patent US 6605709 (GenBank accession no. AR375304.1; E-value 2x10–5, identity 100 %), described as ‘nucleic acid and amino acid sequences related to Proteus mirabilis for diagnostics and therapeutics’. The sequences of PMI1426 and US 6605709 were analysed using the BLASTX tool from NCBI and both retrieved various Hm receptor precursors, outer-membrane receptors and TonB-dependent outer-membrane receptors from various bacteria. The protein with the highest amino acid identity (58 %, E-value 0.0) was a Hm receptor precursor from Photorhabdus luminescens (GenBank accession no. NP_929867).

In addition, the P. mirabilis ORF was subjected to a search for conserved domains employing the CDART tool (NCBI), which determined that the sequence represented a TonB-dependent/ligand-gated channel domain. It is well documented that Hm, Hb, siderophore and haemophore receptors are TonB-dependent proteins (Braun & Braun, 2002) and that in most Gram-negative pathogenic bacteria that have been examined, a functional TonB protein is required for haem and Hb utilization (Braun, 1995).

Taken together, these data strongly indicated that P7 had a single in-frame transposon insertion in the structural gene encoding the 64 kDa OMP. The nucleotide sequence of the interrupted gene revealed that it corresponded to a Hm receptor precursor.

In vitro growth assays

When Pr 6515 and P7 in vitro growth curves were determined in NB (Fig. 2Go), P7 showed a lag phase slightly longer than that of Pr 6515 and reached a lower final bacterial count (about tenfold). As NB is a complex medium containing meat extract, which is a source of haem, the slower growth of P7 could be due to its inability to grow using haem as its iron source. In contrast, Pr 6515 can use the inorganic iron present in the medium as well as the haem, giving it a growth advantage. It was noticed that neither Pr 6515 nor P7 were able to sustain growth in iron-restricted medium (NB-EDDHA). However, P7 was still unable to grow in NB-EDDHA supplemented with Hb or Hm as the sole iron source, whilst Pr 6515 presented a growth curve similar to that observed in NB when grown in NB-EDDHA supplemented with Hb or Hm. Moreover, when we carried out in vitro growth assays on NA-EDDHA plates supplemented with different iron sources spotted onto sterile discs placed on the medium, it was observed that P7 was unable to grow around the Hb-supplemented discs (diameter of growth zone: 0.0±0.0 mm) and grew weakly around the Hm discs (diameter of growth zone: 2.7±0.47 mm), but grew normally using ferric citrate as the only iron source (diameter of growth zone: 13.3±1.53 mm). Pr 6515 could grow using all three different iron sources (diameter of growth zones: 11.3±1.15, 14.3±1.53 and 16.3±0.58 mm, respectively). These results supported our experimental findings indicating that P7 has a TnphoA insertion in an OMP gene related to haem uptake and that its insertion prevents Hm or Hb utilization.


Figure 2
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 2. Representative growth curves of P. mirabilis strain Pr 6515 (filled symbols) and mutant strain P7 (open symbols) in NB ({blacksquare}, {square}), NB-EDDHA (bullet, {circ}), NB-EDDHA with Hm (5 µg ml–1) ({blacktriangleup}, {triangleup}) and NB-EDDHA with Hb (5 µg ml–1) ({blacktriangledown}, {triangledown}). Growth was monitored by plate counts on NA.

 
Growth of Pr 6515 and P7 in urine was also tested (Fig. 3aGo). The results showed that, in contrast to the growth observed in NB, both strains were able to grow in urine to similar levels. The different growth patterns of P7 observed in NB and urine, when compared with Pr 6515, could be attributed to the higher levels of iron sources present in the culture medium. Relatively high levels of iron are excreted in normal urine, although it is not available as it is associated with transferrin or lactoferrin (Shand et al., 1985). It has been shown that P. mirabilis is not able to utilize transferrin or lactoferrin as an iron source (Piccini et al., 1998) and that this pathogen does not produce conventional siderophores (Coker et al., 2000). Moreover, it was reported that the enzyme amino acid deaminase may play a role in iron acquisition (Massad et al., 1995).


Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 3. Representative growth curves of P. mirabilis Pr 6515 ({blacksquare}) and mutant P7 ({blacktriangleup}) in human urine (a) and in IPCs in rats (b). Growth was monitored by plate counts on NA.

 
Therefore, the iron acquisition from urine by P. mirabilis could be mediated by the action of the enzyme amino acid deaminase or another uncharacterized system that could involve other P. mirabilis iron-regulated OMPs. Uropathogens have functional redundancy in their iron-uptake systems and there are general hypotheses that may explain the existence of such redundancy, such as the maximization of possibilities to obtain iron from different sources and increases in efficacy of iron acquisition in different media (Russo et al., 2002).

In vivo assays

In a previous study, it was observed that sera obtained from animals infected with P. mirabilis Pr 6515 reacted with the 64 kDa OMP, suggesting that this protein is expressed in vivo during infection (Piccini et al., 1998). It was reported that P. mirabilis is able to grow and survive in vivo in an IPC model, and it was also suggested that P. mirabilis OMPs are potential specialized virulence factors for survival and may play an important role in UTIs (Moayeri et al., 1991; Zunino et al., 1999).

In this work, in vivo bacterial growth was studied using IPCs filled with a suspension of Pr 6515 or P7 and implanted into the peritoneal cavity of rats. Viable counts of bacteria were performed at different periods of time (Fig. 3bGo). It was observed that, whereas Pr 6515 was able to multiply 1000-fold after 24 h and to maintain levels of 108 c.f.u. ml–1 over the period of time evaluated, the number of P7 c.f.u. drastically decreased after 12 h. As iron is not available in the mammalian host as a free nutrient but is linked to carrier proteins (Braun, 2001), it is possible that the absence of the 64 kDa OMP in P7 prevented its growth in vivo, confirming the importance of this OMP for P. mirabilis survival in this in vivo environment.

The role of the 64 kDa OMP in urovirulence was assessed using different models of ascending and haematogenous UTI in the mouse. The ascending UTI model (Hagberg et al., 1983) resembles the natural route of infection in humans and allowed us to evaluate the colonization of the urinary tract. For transurethral infection, bacteria recovered from the kidneys and bladder of mice challenged with strain Pr 6515 or P7 were enumerated (Fig. 4aGo). In the simple experimental transurethral infection, no significant differences were observed between kidney and bladder viable counts of P7 and Pr 6515 (P >0.05).


Figure 4
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 4. Virulence of wild-type strain Pr 6515 and mutant strain P7 in a simple ascending UTI model (a) and a co-infection ascending UTI model (b) in mice. The median (horizontal bar) is indicated for each group. The detection limit was 1x102 c.f.u. per kidney or bladder. P values were determined by the Mann–Whitney non-parametric test.

 
However, when the ability of Pr 6515 and P7 to compete directly in vivo was assessed by performing an ascending co-infection UTI assay, P7 was significantly outcompeted by Pr 6515 and colonization of the kidneys and bladder of mice was significantly affected (P <0.05) (Fig. 4bGo). Co-infection assays provide a model to compare the colonization ability of different strains, as they avoid individual host differences and permit the detection of subtle differences between competing strains (Nagy et al., 2002). In our co-infection experiments, P7 was probably impaired in its ability to colonize the urinary tract, an aspect that was not apparent in the simple infection model. From our results, it could be hypothesized that the role of bacterial haem/Hb uptake mechanisms, particularly those mediated by OMP receptors, is more evident in later steps of infection, contributing to bacteria survival and invasion (Warren et al., 1987; Zunino et al., 1999).

When we carried out intravenous infection assays, we observed that 63 % (10/16) of mice inoculated with Pr 6515 and 20 % (3/15) of mice inoculated with P7 were infected. Therefore, a significant difference between infectivity in mice inoculated with Pr 6515 and P7 was observed (P <0.05). Considering that, in this experimental infection route, bacteria reach the urinary tract through the bloodstream, where haem is the main iron source, and that it was proposed that haem is the dominant iron source in kidneys (Burall et al., 2004), this result suggested that the absence of the 64 kDa OMP in P7 impaired the development of infection by avoiding the access of this strain to haem. Moreover, considering that the production of functional haemolysin in P. mirabilis is regulated at the transcriptional level by iron concentration (Lukomski et al., 1991), we suggest that the utilization of haem delivered from eukaryotic cells by the action of haemolysin could be an important strategy for obtaining iron during an infection. We were able to determine that Pr 6515 and P7 presented similar haemolytic activity (haemolytic titres of 1 : 4 for both bacteria after 3 h of incubation). This could indicate that the attenuation of the mutant strain in intravenous infection was exclusively due to the lack of the 64 kDa OMP.

In conclusion, we confirmed that the 64 kDa iron-regulated OMP is a haem receptor of P. mirabilis and has an important role in the growth of these bacteria in the mammal host and in UTIs. Further work assessing the amino acid sequence of this OMP by mass spectrometry is currently under way.


   ACKNOWLEDGEMENTS
 
This work was supported by Third World Academy of Science (TWAS) and Comisión Sectorial de Investigación Científica (CSIC), Universidad de la República, Uruguay. The authors want to thank Marcelo Fernández for his assistance in procedures involving animal work and Nicholas Thompson and Mohammed Sebaihia, from The Wellcome Trust Sanger Institute, Genome Campus, Hinxton, Cambridge, UK, who kindly responded to our P. mirabilis sequence enquiries.


   REFERENCES
TOP
 INTRODUCTION
 METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Ahmed, A. (1991). A comparison of intramolecular rearrangements promoted by transposons Tn5 and Tn10. Proc Biol Sci 244, 1–9.[Abstract/Free Full Text]

Ali, A., Johnson, J. A., Franco, A. A., Metzger, D. J., Connel, T. D., Morris, G. & Sozhmannan, S. (2000). Mutations in the extracellular protein secretion pathway genes (eps) interfere with rugose polysaccharide production in and motility of Vibrio cholerae. Infect Immun 68, 1967–1974.[Abstract/Free Full Text]

Braun, V. (1995). Energy coupled transport and signal transduction through the Gram-negative outer membrane via TonB-ExbB-ExbD-dependent receptor proteins. FEMS Microbiol Rev 16, 295–307.[CrossRef][Medline]

Braun, V. (2001). Iron uptake mechanisms and their regulation in pathogenic bacteria. Int J Med Microbiol 291, 67–79.[CrossRef][Medline]

Braun, V. & Braun, M. (2002). Iron transport and signaling in Escherichia coli. FEBS Lett 529, 78–85.[CrossRef][Medline]

Bullock, W. O., Fernández, J. M. & Short, J. M. (1987). XL-1 Blue: a high efficiency plasmid transforming recA Escherichia coli strain with β-galactosidase selection. Biotechniques 5, 376–379.

Burall, L. S., Harro, J. M., Li, X., Lockatell, C. V., Himpsl, S. D., Hebel, J. R., Johnson, D. E. & Mobley, H. L. T. (2004). Proteus mirabilis genes that contribute to pathogenesis of urinary tract infection: identification of 25 signature-tagged mutants attenuated at least 100-fold. Infect Immun 72, 2922–2938.[Abstract/Free Full Text]

Coker, C., Poore, C. A., Li, X. & Mobley, H. L. T. (2000). Pathogenesis of Proteus mirabilis urinary tract infection. Microbes Infect 2, 1497–1505.[CrossRef][Medline]

Gallagher, L. A. & Manoil, C. (2001). Pseudomonas aeruginosa PAO1 kills Caenorhabditis elegans by cyanide poisoning. J Bacteriol 183, 6207–6214.[Abstract/Free Full Text]

Hagberg, L., Engberg, I., Freter, R., Lane, J., Olling, S. & Svanborg-Eden, C. (1983). Ascending unobstructed urinary tract infection in mice caused by pyelonephritogenic Escherichia coli of human origin. Infect Immun 40, 273–283.[Abstract/Free Full Text]

Litwin, C. M. & Calderwood, S. B. (1993). Role of iron in regulation of virulence genes. Clin Microbiol Rev 6, 137–149.[Abstract/Free Full Text]

Llamas, M. A. & Bitter, W. (2006). Iron gate: the translocation system. J Bacteriol 188, 3172–3174.[Free Full Text]

Lukomski, S., Serwecinska, L., Rozalski, A., Dziadek, J., Staczek, P. & Jaworski, A. (1991). Cell-free and cell-bound hemolytic activities of Proteus penneri determined by different Hly determinants. Can J Microbiol 37, 419–424.[Medline]

Manoil, C. & Beckwith, J. (1985). TnphoA: a transposon probe for protein export signal. Proc Natl Acad Sci U S A 82, 8129–8133.[Abstract/Free Full Text]

Maskell, D. J., Morrisey, P. & Dougan, G. (1988). Cloning and nucleotide sequence of the aroA gene of Bordetella pertussis. J Bacteriol 170, 2467–2471.[Abstract/Free Full Text]

Massad, G., Zhao, H. & Mobley, H. L. T. (1995). Proteus mirabilis amino acid deaminase: cloning, nucleotide sequence, and characterization of aad. J Bacteriol 177, 5878–5883.[Abstract/Free Full Text]

Moayeri, N., Collins, C. M. & O'Hanley, P. (1991). Efficacy of a Proteus mirabilis outer membrane protein vaccine in preventing experimental pyelonephritis in a BALB/c mouse model. Infect Immun 59, 3778–3786.[Abstract/Free Full Text]

Mobley, H. L. T. & Chippendale, G. R. (1990). Hemagglutinin, urease and hemolysin production by Proteus mirabilis from clinical sources. J Infect Dis 161, 525–530.[Medline]

Nagy, G., Dobrindt, U., Schneider, G., Salam, K. A., Hacker, J. & Emody, L. (2002). Loss of regulatory protein RfaH attenuates virulence of uropathogenic Escherichia coli. Infect Immun 70, 4406–4413.[Abstract/Free Full Text]

Olczak, T., Simpson, W., Liu, X. & Genco, C. A. (2005). Iron and heme utilization in Porphyromonas gingivalis. FEMS Microbiol Rev 29, 119–144.[CrossRef][Medline]

Piccini, C. D., Barbé, F. M. & Legnani-Fajardo, C. L. (1998). Identification of iron-regulated outer membrane proteins in uropathogenic Proteus mirabilis and its relationship with heme uptake. FEMS Microbiol Lett 166, 243–248.[CrossRef][Medline]

Quessy, S., Dubreuil, J. D., Jacques, M., Malouin, F. & Higgins, R. (1994). Increase of capsular material thickness following in vivo growth of virulent Streptococcus suis serotype 2 strains. FEMS Microbiol Lett 115, 19–26.[CrossRef][Medline]

Ròzalski, A., Sidorxsyk, Z. & Kotelko, K. (1997). Potential virulence factors of Proteus bacilli. Microbiol Mol Biol Rev 61, 65–89.[Abstract]

Russo, T. A., McFadden, C. D., Carlino-MacDonald, U. B., Beanan, J. M., Barnard, T. J. & Johnson, J. R. (2002). IroN functions as a siderophore receptor and is a urovirulence factor in an extraintestinal pathogenic isolate of Escherichia coli. Infect Immun 70, 7156–7160.[Abstract/Free Full Text]

Rutherford, K., Parkhill, J., Crook, J., Horsnell, T., Rice, P., Rajandream, M. A. & Barrell, B. (2000). Artemis: sequence visualization and annotation. Bioinformatics 16, 944–945.[Abstract/Free Full Text]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Senior, B. W. (1978). The special affinity of particular types of Proteus mirabilis for the urinary tract. J Med Microbiol 12, 1–8.[Medline]

Shand, G. H., Anwar, H., Kadurugamuya, J., Brown, M. R. W., Silverman, S. H. & Melling, J. (1985). In vivo evidence that bacteria in urinary tract infection grow under iron-restricted conditions. Infect Immun 48, 35–39.[Abstract/Free Full Text]

Simon, R., Priefer, U. & Puhler, A. (1983). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Biotechnology 1, 784–791.[CrossRef]

Stamm, W. E., Martin, S. M. & Bennett, J. V. (1977). Epidemiology of nosocomial infection due to Gram-negative bacilli: aspects relevant to development and use of vaccines. J Infect Dis 136, S151–S160.[Medline]

Stojiljkovic, I., Kumar, V. & Srinivasan, N. (1999). Non-iron metalloporphyrins: potent antibacterial compounds that exploit haem/Hb uptake systems of pathogenic bacteria. Mol Microbiol 31, 429–442.[CrossRef][Medline]

Taylor, R. K., Manoil, C. & Mekalanos, J. J. (1989). Broad-host-range vectors for delivery of TnphoA: use in genetic analysis of secreted virulence determinants of Vibrio cholerae. J Bacteriol 171, 1870–1878.[Abstract/Free Full Text]

Wandersman, C. & Stojiljkovic, I. (2000). Bacterial heme sources: the role of heme, hemoprotein receptors and hemophores. Curr Opin Microbiol 3, 215–220.[CrossRef][Medline]

Warren, J. W., Damron, D., Tenney, J. H., Hoopes, J. M., Deforge, B. & Munice, H. L. (1987). Fever, bacteremia, and death as complications of bacteriuria in women with long-term catheters. J Infect Dis 155, 1151–1158.[Medline]

Yanisch-Perron, C., Vieira, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strain: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103–119.[CrossRef][Medline]

Zunino, P., Piccini, C. & Legnani-Fajardo, C. (1994). Flagellate and non-flagellate Proteus mirabilis in the development of experimental urinary tract infection. Microb Pathog 16, 379–385.[CrossRef][Medline]

Zunino, P., Piccini, C. & Legnani-Fajardo, C. (1999). Growth, cellular differentiation and virulence factor expression by Proteus mirabilis in vitro and in vivo. J Med Microbiol 48, 527–534.[Abstract/Free Full Text]

Zunino, P., Geymonat, L., Allen, A., Preston, A., Sosa, V. & Maskell, D. (2001). New aspects of the role of MR/P fimbriae in Proteus mirabilis urinary tract infection. FEMS Immunol Med Microbiol 31, 113–120.[CrossRef][Medline]




This article has been cited by other articles:


Home page
 Infect. Immun.Home page
G. R. Nielubowicz, S. N. Smith, and H. L. T. Mobley
Outer Membrane Antigens of the Uropathogen Proteus mirabilis Recognized by the Humoral Response during Experimental Murine Urinary Tract Infection
Infect. Immun., September 1, 2008; 76(9): 4222 - 4231.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lima, A.
Right arrow Articles by Piccini, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lima, A.
Right arrow Articles by Piccini, C.
Agricola
Right arrow Articles by Lima, A.
Right arrow Articles by Piccini, C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL J MED MICROBIOL MICROBIOLOGY J GEN VIROL ALL SGM JOURNALS