| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Cardiff School of Biosciences, Cardiff University, Cardiff, Wales, UK
Correspondence D. J. Stickler stickler{at}cardiff.ac.uk
Received April 18, 2005
Accepted June 14, 2005
The care of many patients undergoing long-term bladder catheterization is frequently complicated by infection with Proteus mirabilis. These organisms colonize the catheter, forming surface biofilm communities, and their urease activity generates alkaline conditions under which crystals of magnesium ammonium phosphate and calcium phosphate are formed and become trapped in the biofilm. As the biofilm develops it obstructs the flow of urine through the catheter, causing either incontinence due to leakage of urine around the catheter or retention of urine in the bladder. The aim of this study was to investigate the role of the surface-associated swarming motility of P. mirabilis in the initiation and development of these crystalline catheter biofilms. A set of stable transposon mutants with a range of swimming and swarming abilities were tested for their ability to colonize silicone surfaces in a parallel-plate flow cell. A laboratory model of the catheterized bladder was then used to examine their ability to form crystalline, catheter-blocking biofilms. The results showed that neither swarming nor swimming motility was required for the attachment of P. mirabilis to silicone. Mutants deficient in swarming and swimming were also capable of forming crystalline biofilms and blocking catheters more rapidly than the wild-type strain.
Present address: Alimentary Pharmabiotic Centre, Department of Microbiology, University College Cork, Cork, Ireland.  |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The well-known ability of P. mirabilis to transform when it contacts a surface from small swimming bacilli into elongated, highly flagellated swarmer cells is accompanied by a substantial increase in the production of urease (Allison et al., 1992; Falkinham & Hoffman, 1984). Swarming in the presence of urine could therefore accelerate the generation of alkaline conditions that cause the deposition of crystalline material on the catheters. Swarmer cells have been shown to move rapidly over catheter surfaces (Stickler & Hughes, 1999). It is possible therefore that swarming facilitates the initiation of infection by mediating the migration of the organism from the peri-urethral skin along the catheter into the bladder.
The aim of this study was to investigate the contribution of swarming to the development of the crystalline catheter biofilms. A set of stable transposon mutants with a range of swimming and swarming abilities were tested for their ability to colonize silicone surfaces in a parallel-plate flow cell (Gottenbos et al., 1999). A laboratory model of the catheterized bladder (Stickler et al., 1999) was then used to examine their ability to form crystalline, catheter-blocking biofilms.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
Urease production.
Urease activity of swarming-deficient mutants was measured using a protocol modified from Creno & Wenk (1970). P. mirabilis strains were cultured for 4 h in 5 ml LB broth supplemented with urea (0.1 % w/v). Cells were harvested by centrifugation (1600 g for 10 min), the supernatant discarded and pellets resuspended in 2.5 ml ice-cold sodium phosphate buffer (0.1 M sodium phosphate, 10 mM EDTA, pH 7.3). Total protein in cell suspensions was determined using a Micro Protein Determination kit (Sigma Diagnostics) according to the manufacturer's instructions. To measure urease activity, 200 µl of cell suspension was added to 800 µl reaction buffer (50 mM urea, 0.1 M sodium phosphate), mixed thoroughly and incubated at 37 °C for 10 min. The reaction was terminated by addition of 2 ml phenol sodium nitroprusside solution (0.5 % w/v phenol, 0.025 % w/v sodium nitroprusside). Colour development was initiated by addition of 2 ml sodium hypochlorite solution (0.2 % w/v sodium hydroxide, 0.21 % w/v sodium hypochlorite).
These mixtures were incubated at 56 °C for 5 min. Colour change was measured against a blank containing 200 µl sodium phosphate buffer instead of cell suspension at 626 nm using a Helios
UV-Vis spectrophotometer (Unicam). The intensity of the blue colour that developed was compared to that given by standard solutions of ammonium chloride ranging in concentration from 0.01 mM to 10 mM. Urease activity was expressed as mmol urea hydrolysed min–1 (mg protein)–1.
Parallel-plate flow chambers.
Flow chambers (Gottenbos et al., 1999) were purchased from the Department of Bio-Medical Engineering, University of Groningen, Groningen, The Netherlands. Cell deposition onto a silicone-coated glass plate was monitored directly using a phase-contrast microscope with a long working distance objective lens and equipped with a VC600 CCD video camera (Olympus). Images were viewed using a Vista KVM14 monitor (Norbain SDL) and captured in TIF format using a monochrome frame grabber (Mach Series DT3155, Data Translation) installed in an IBM-compatible Pentium II PC. All flow-chamber experiments were carried out at 37 °C. The glass top plate (uncoated) and teflon spacers were immersed in a solution of Decon 90 (2 % v/v; Decon Laboratories) and sonicated for 5 min at 60 kHz. Plates and spacers were then washed thoroughly in hot water and rinsed in methanol followed by deionized water. A glass bottom plate coated with silicone was positioned in the base of the flow chamber and the flow cell assembled and positioned on the microscope stage. Phosphate buffer (di-sodium hydrogen orthophosphate 11.357 g l–1, potassium di-hydrogen orthophosphate 2.722 g l–1, pH 7.4) was flushed through the system for 15 min prior to use.
Assessment of P. mirabilis adherence to silicone in the flow chambers.
Test strains were grown overnight in LB broth (15 ml). Cells were harvested by centrifugation and resuspended in 15 ml phosphate buffer. The suspension was sonicated gently for 20 s at 60 kHz to break up cell aggregates. Cell density was adjusted to 3 x 108 cells ml–1 in a final volume of 250 ml. The flow rate through the chamber was set to 1 ml min–1 and a stable flow established. To monitor the rate of cell deposition, images were recorded at 10 min, 30 min and 1 h, and then at hourly intervals for a total of 5 h. At each time point Vision XXL software was used to record a series of 30 images over a 1 s period and collate these to produce a final image (this allowed only stationary cells to be recorded). The overall ability of the wild-type and mutant strains to adhere to silicone was assessed by calculating the number of cells attached per cm2 after 5 h. The numbers of cells attached in 18 separate fields of view (six fields of view in each of three replicate flow-chamber experiments) were used to calculate the number of cells per cm2.
In vitro bladder models.
In vitro bladder models were assembled and run as described previously (Stickler et al., 1999). Bladder models consisted of glass chambers surrounded by a water jacket, with an outlet from the interior chamber of the vessel. Water jackets were connected to a circulating water bath to maintain the incubation chamber at 37 °C. A 10 cm section of silicone tubing attached to the outlet of the model served as the urethra. Size 14 all-silicone Foley catheters (Bard) were inserted into models and balloons inflated with 10 ml sterile deionized water, holding the catheters in position and creating a seal over the bladder model outlet. A residual volume (20 ml) of artificial urine was allowed to collect in models below the level of the catheter eye-hole. The models were then inoculated with 10 ml of 4 h culture of the test strain grown in artificial urine. After inoculation cultures were allowed to establish themselves in the models for 1 h before the urine was supplied at a constant flow rate of 0.5 ml min–1. The models were run until the catheters became completely obstructed. Each strain was tested in triplicate, and the time taken for catheters to block was recorded. The pH of the media in the models and the number of viable cells in the residual urine at the time of activation and the time of blockage were also measured.
Scanning electron microscopy.
Catheters were removed from the bladder models at the end of the experimental periods. Sections (1 cm in length) were cut from each catheter, from immediately below the eye-hole. These sections were viewed directly in a JEOL 5200 scanning electron microscope (JEOL) using the low vacuum setting.
Statistical analysis.
Data were analysed using either one-way analysis of variance, or two-sample t-test as appropriate. All calculations were performed using Minitab 10.51 (Minitab).
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Ability of P. mirabilis to adhere to silicone in parallel-plate flow chambers
The ability of the wild-type and mutant strains to adhere to silicone was assessed in the flow cell over 5 h periods. The attachment rates of the wild-type strain and representative strains from each category of mutant are shown in Fig. 1. Mutants with increased swarming ability were attenuated in their ability to attach to the silicone surface, while mutants deficient in swarming or swimming were able to attach at rates greater than the wild-type. The numbers of cells attached per cm2 of silicone surface after 5 h was also calculated for each of the test strains and these data are presented in Table 2. Overall these data suggest that adhesion to silicone is enhanced by loss or impairment of the ability to swarm. For statistical analysis strains were divided into two broad groups, those with normal or increased motility (wild-type and group 1 mutants) and those with reduced motility (groups 2, 3 and 4). The analysis of variance revealed a highly significant difference (P < 0.001) in adherence between strains possessing normal or increased swarming ability and those exhibiting a reduction in swarming ability.
|
|
Ability to encrust all-silicone catheters
The role of swarming in the development of crystalline biofilms was assessed using an in vitro model of the catheterized bladder (Stickler et al., 1999). The ability of the mutants to block all-silicone catheters was compared to that of the wild-type strain B4.
In general, strains exhibiting a reduction in swarming ability (groups 2, 3 and 4) were found to produce crystalline biofilms and block catheters more quickly than the wild-type strain B4. In contrast, the mutants that exhibit swarming abilities greater than that of the wild-type (group 1) showed an increase in the time taken to block the catheters (Table 3). When strains were divided into the two broad groups of normal or increased motility (wild-type and group 1 mutants) and reduced motility (groups 2, 3 and 4) as above, the mean blockage times of these two groups were significantly different (P = 0.028, analysis of variance). The scanning electron micrographs (Fig. 2) show examples of crystalline biofilms produced by the wild-type strain, a non-swarming mutant (G93) and the non-swimming, non-swarming mutant (NS63).
|
|
The pH of the urine in the bladder chamber was measured at the beginning and end of each experiment, and the mean values for each test strain are shown in Table 3. The urine used in these experiments simulates that produced by elderly patients undergoing long-term catheterization. These patients have low fluid intakes and concentrated urines from which calcium and magnesium phosphates crystallize at around pH 7.0. It can be seen that throughout the experimental period therefore, in all the models, conditions under which catheters can encrust were established. No significant differences were observed in the pH values of media in models after blockage of catheters (P = 0.23 or greater). All strains grew well in the models, producing viable cell counts in excess of 109 c.f.u. ml–1. Analysis of variance revealed no significant differences in the number of viable cells of the strains present in the urine after the blockage of catheters (P = 0.67 or greater).
Swarming and the pathogenesis of P. mirabilis in the urinary tract
There is uncertainty about the role of swarming in the pathogenicity of P. mirabilis in the urinary tract. Studies by Allison et al. (1992) concluded that the differentiated swarm cell, rather than the swimmer cell, was associated with the ability of P. mirabilis to invade uroepithelial cells in vitro. Subsequently, Belas et al. (1995) proposed that the ability of P. mirabilis to swarm may allow it to overcome some of the intrinsic defence mechanisms of the urinary tract. For example, the highly viscous layer of mucous covering the epithelial surfaces traps many motile bacteria and prevents them ascending the tract. In the case of P. mirabilis, however, this would inhibit the flagella rotation of trapped swimmer cells and induce swarm cell differentiation. Swarming motility may then facilitate the migration of this organism through the mucous layer. It was also suggested that the increased expression of several virulence factors (urease, metalloprotease and haemolysin) that occurs during swarming helps the bacteria overcome host defences.
The recent report of Jansen et al. (2003), however, has raised some doubts about this attractive hypothesis. In an animal model of ascending infection, they found that the predominant morphotype of P. mirabilis was the short rod-like swimmer cell, and the elongated swarmer cells were rarely observed. Additionally, Zunino et al. (1994) reported the isolation of a non-motile mutant of P. mirabilis from patients with symptomatic urinary tract infections. Subsequent studies of this strain, and a laboratory-generated mutant unable to synthesize flagella, showed that both non-motile strains were as capable as the wild-type strain of causing urinary tract infections in mice (Legnani-Fajardo et al., 1996).
In the case of the catheterized urinary tract the situation could be quite different and it is remarkable that while P. mirabilis is not a major pathogen of the normal urinary tract, this species becomes responsible for over 40 % of infections when a long-term indwelling catheter is introduced into the bladder (Mobley, 1996). Previous experiments in a simple laboratory model demonstrated that P. mirabilis was capable of migrating over the surfaces of all currently available types of catheter (Stickler & Hughes, 1999). A later study in the same model demonstrated that P. mirabilis was able to migrate over all the basic types of catheter more effectively than any of the other common urinary tract pathogens tested. This ability, however, was impeded under conditions in which swarming did not occur (Sabbuba et al., 2002). Mutants lacking the ability to swarm also failed to migrate over catheters (Jones et al., 2004). There is thus evidence to suggest that swarming has a role in the initiation of catheter-associated urinary tract infections by facilitating the migration of the P. mirabilis from the skin at the insertion site, along the catheter and into the bladder.
The results from the flow-cell experiments confirm the findings of earlier studies that suspensions of P. mirabilis cells in buffer at pH 7.4 will adhere well to silicone surfaces (Downer et al., 2003). However, a reduction in swarming ability was found not to impair attachment to silicone. Mutants with poor or non-swarming phenotypes showed a significantly enhanced ability to adhere to silicone under these conditions (Fig. 1, Table 2). Similar results were reported by Mireles et al. (2001) in a study of the role of swarming in biofilm formation by Salmonella enterica. They reported that non-swarming mutants defective in LPS production were generally more proficient than the wild-type strain in forming biofilms on PVC. It is possible that on contact with a surface, the non-swarmers are more likely to remain at the site of attachment, divide and produce the microcolonies that initiate biofilm formation. The non-swarming mutants produce either very few or defective flagella, or, in the case of NS63, no flagella at all (Table 1). The fact that these strains adhere to silicone in significantly greater numbers compared to strains that produce normal flagella strongly suggests that flagella do not have a role in the attachment of P. mirabilis to this biomaterial.
The flow-cell experiments were performed with suspensions of cells in buffer to achieve standardized conditions throughout the test. However, in the catheterized urinary tract P. mirabilis cells would be actively growing in urine, producing urease and driving the formation of crystals of calcium and magnesium phosphates, which accumulate in the urine. The experiments performed in the bladder model allowed us to examine the role of swarming under conditions in which these processes take place. Sabbuba et al. (2002) suggested that swarming might facilitate the development and spread of the crystalline biofilm over the catheter surfaces, thus contributing to the eventual obstruction of urine flow from the bladder. It has also been suggested that the increased urease expression exhibited by swarmer cells accelerates mineralization of the biofilm and enhances blockage (Stickler & Hughes, 1999).
The results presented in Table 3 and Fig. 2 demonstrate quite clearly that the ability to swarm is not important in the development of catheter-blocking crystalline biofilms, and that increased urease production by swarmer cells is not an important factor in this process. All strains tested were capable of forming crystalline biofilms and blocking silicone catheters. However, strains with increased or normal swarming ability (group 1 and wild-type) were found to take significantly longer (P = 0.028) to block catheters than strains with reduced swarming ability (groups 2, 3 and 4). As all these strains produced equivalent levels of urease (Table 1), this suggests that mutants with reduced swarming ability are more proficient in generating catheter-blocking crystalline biofilms on all-silicone catheters.
In the case of P. mirabilis in the catheterized urinary tract, once the organisms have reached the bladder and infected the urine, the subsequent flow of urine through the catheter into the drainage bag will ensure the distribution of the cells over its surface. In addition, the amorphous crystals of calcium phosphate that come out of solution under alkaline conditions form macroscopic aggregates with the cells. The simple deposition of this co-aggregated material under gravity could serve to initiate biofilm formation on the catheters. Therefore, as long as the infecting organism is able to produce sufficient urease to generate alkaline urine it will be able to form crystalline biofilms and block catheters.
In conclusion, the results of the flow-chamber and bladder-model experiments indicate that neither swimming nor swarming motility are required for the adhesion of P. mirabilis to silicone or for the subsequent development of catheter-blocking crystalline biofilms.
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. M. Pearson, M. Sebaihia, C. Churcher, M. A. Quail, A. S. Seshasayee, N. M. Luscombe, Z. Abdellah, C. Arrosmith, B. Atkin, T. Chillingworth, et al. Complete Genome Sequence of Uropathogenic Proteus mirabilis, a Master of both Adherence and Motility J. Bacteriol., June 1, 2008; 190(11): 4027 - 4037. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Jacobsen, D. J. Stickler, H. L. T. Mobley, and M. E. Shirtliff Complicated Catheter-Associated Urinary Tract Infections Due to Escherichia coli and Proteus mirabilis Clin. Microbiol. Rev., January 1, 2008; 21(1): 26 - 59. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
, B4 (wild-type); 
, BVJ14 (group 1 – increased swarming ability); •, G78 (group 2 – poor-swarming); 
, G93 (group 3 – non-swarming); 
, NS63 (group 4 – non-swimming, non-swarming). The data show the mean number of cells attached in one field of view (0.014 mm) at each time point from three replicate experiments. Error bars show the SEM.