TY - JOUR
AU - Ruzicka, Filip
AB - Abstract More than 40% of nosocomial infections are those of the urinary tract, most of these occurring in catheterized patients. Bacterial colonization of the urinary tract and catheters results not only in infection, but also various complications, such as blockage of catheters with crystalline deposits of bacterial origin, generation of gravels and pyelonephritis. The diversity of the biofilm microbial community increases with duration of catheter emplacement. One of the most important pathogens in this regard is Proteus mirabilis. The aims of this study were to identify and assess particular virulence factors present in catheter-associated urinary tract infection (CAUTI) isolates, their correlation and linkages: three types of motility (swarming, swimming and twitching), the ability to swarm over urinary catheters, biofilm production in two types of media, urease production and adherence of bacterial cells to various types of urinary tract catheters. We examined 102 CAUTI isolates and 50 isolates taken from stool samples of healthy people. Among the microorganisms isolated from urinary catheters, significant differences were found in biofilm-forming ability and the swarming motility. In comparison with the control group, the microorganisms isolated from urinary catheters showed a wider spectrum of virulence factors. The virulence factors (twitching motility, swimming motility, swarming over various types of catheters and biofilm formation) were also more intensively expressed. biofilm, CAUTI, virulence factors, motility, catheter Introduction The number of the catheter-associated urinary tract infections (CAUTIs) increases every year, reflecting the fact that urinary catheters are the second most often used foreign objects inserted into the human body (Trautner & Darouiche, 2004). Over 40% of nosocomial infections are those of the urinary tract, found especially in catheterized patients (Bryers, 2008). Despite good aseptic management, about half of all patients experience bacteriuria in the first 10–14 days of catheterization (Morris & Stickler, 1998). The risk of urinary tract infections is significantly higher in patients with long-term inserts (28 days); the percentage of infected catheters in these patients approaches 100% (Morris & Stickler, 1998). Catheterization means the bacteria can more easily attack the urinary tract and urinary bladder. There are also other complications that are linked to bacterial colonization of the urinary tract and catheters, for example blockage of catheters with crystalline deposits of bacterial origin, generation of gravels and pyelonephritis. These complications are mainly caused by a few bacterial species: those from the genera Proteus (and close related Morganella) and Pseudomonas (Gorman & Tunney, 1997). Adhesion of bacteria to the catheter depends on many factors, for example surface charge, hydrophobicity or hydrophilicity of the catheter and bacterial cell, and specific genes for adhesion (Liedl, 2001). The risk of infection depends on the duration of catheterization and catheter management. The large number of CAUTIs is associated with the biofilm mode of growth of the microorganisms. The artificial surface of the implants facilitates adhesion of bacteria, which can then form biofilm. Biofilm-positive bacteria use also other virulence factors for their survival, for example those increasing their invasiveness, contagiousness and toxicity (Rózalski et al., 1997; Stickler, 2008). Bacteria of the genus Proteus are widely distributed in the environment and the intestine of many mammals, including humans (O'May et al., 2008). Their contagiousness is linked to their resistance to antimicrobial compounds. This resistance enables them to survive in the hospital environment. Their invasiveness is related to many virulence factors such as surface adherence, and penetration into and spread throughout the host's body. Three types of movement have been identified in the genus Proteus: swimming, swarming and twitching. Proteus bacteria use swimming motility (SM) in liquid media. Six to 10 flagella are present on the bacterial surface for this type of movement. Swarming motility (SW) requires an increased number of flagella and this type of motility is used for movement on solid surfaces. SW is the typical mode of movement of Proteus grown on solid agar, the Rauss phenomenon. With twitching motility (TM), the bacteria twitch between two solid surfaces (Rózalski et al., 1997; Rashid & Kornberg, 2000). Another important virulence factor is biofilm formation: an attached structure with microbial cells and populations embedded in a polysaccharide layer. The biofilm facilitates, for example, survival, enables better adaptation to the conditions of the external environment and more effective use of nutrition (Costerton, 1999). For effective treatment of CAUTIs, it is necessary to understand virulence factors and possible linkages among their production. We focus here on the detection of selected virulence factors and the linkages among their production in bacteria of the genus Proteus, including biofilm formation under various conditions. Methods A sonication technique was used for isolation of Proteus strains from polymicrobial biofilms of urinary catheters as previously described (Hola et al., 2010). Briefly, all catheters were sonicated in 5 mL of Brian-Heart Infusion (BHI) broth for 5 min, vortexed for 2 min and sonicated for another 5 min. The suspension was subsequently inoculated on solid media to isolate individual strains. All isolated strains were identified to the species level using conventional biochemical tests (Micro-LA-tests, Lachema, Czech Republic; API, bioMérieux, France). We isolated 102 strains of the genus Proteus (90 strains of Proteus mirabilis, 12 strains of Proteus vulgaris). As a control group we also included in this study 50 strains of the genus Proteus (43 strains of P. mirabilis, seven strains of P. vulgaris) isolated from stools of healthy individuals. We performed the following tests for virulence factors with isolated strains: biofilm formation, three tests for motility (swarming, swimming and twitching), the ability to attach to various types of catheters (silicon, silicon-coated, latex, hydrogel-silver-coated, etc.), the ability to swarm over various types of catheters, production of urease and susceptibility/resistance to selected antibiotics. To assess the influence of diabetes mellitus on production of virulence factors we also assessed the effect of glucose supplementation on the strength of biofilm formation and its correlation with other virulence factors. The relationships among the virulence factors were assessed. Biofilm formation tests and the effect of glucose supplementation Culture conditions corresponded to those described in Stepanović et al. (2007); all strains were cultivated in microtitre tissue culture plates at 37 °C for 24 h in BHI and in BHI plus 4% glucose (BHI-g). Wells of the microtitre plates were then washed three times and the biofilm layer was fixed by air-drying. The fixed biofilm layer was stained with crystal violet and the biofilm positivity was assessed quantitatively based on assessment of OD595 nm. All strains were examined in triplicate. Tests for SW, SM and TM Luria–Bertani (LB) medium with various agar content appropriate for the type of motility being tested was used. Swarming medium contained 0.7% agar and strains were inoculated by picking on the agar surface. Swimming medium contained 0.3% agar and strains were inoculated by puncture into the medium. Twitching medium contained 1.0% agar and strains were inoculated with a micropipette under the agar layer (10 µL). After cultivation for 24 h at 37 °C, the diameters of produced zones were measured (SM and SW). The medium from the TM plates was tilted, the TM zone was fixed by air-drying, stained by crystal violet and measured (Rashid & Kornberg, 2000; Jones et al., 2004) (see Fig. 1). Figure 1 View largeDownload slide (a) Swarming motility on agar; (b) Swimming motility within the agar layer; (c) Twitching motility on the Petri dish (crystal violet staining). Figure 1 View largeDownload slide (a) Swarming motility on agar; (b) Swimming motility within the agar layer; (c) Twitching motility on the Petri dish (crystal violet staining). Biofilm formation on various catheters Various urinary tract catheters were used. Four were silicon-coated: Unomedical Latex Foley Catheter (26Fr; Well Lead Medical, China), Meditec Foley Catheter (26Fr; Medilab, Czech Republic), Kendall Curity Catheter (24Fr; Tyco Healthcare, UK) and Berocath Foley Ballon Catheter (24Fr; Beromed, Germany); one catheter was made of latex: MPI Foley Catheter (24Fr; Medicoplast, Germany); and two catheters were whole-silicon: Kendall Argyle (26Fr; Tyco Healthcare, UK) and Kendall Dover Silver (18Fr; Tyco Healthcare, UK). The Kendal Dover Silver catheter was coated with silver-hydrogel. The catheters were aseptically cut into approx. 2-cm pieces and submerged in 2 mL BHI-g, which was inoculated with 200 µL suspension of the tested strain. After 24 h cultivation at 37 °C, the catheters were aseptically removed from the medium, washed three times with sterile phosphate-buffered saline to wash out nonadherent cells and transferred into fresh phosphate-buffered saline. The biofilm of adherent cells was disrupted by sonication (5 min sonication, 2 min vortexing, 5 min further sonication) in an ultrasonic bath at 40 kHz (Cole-Parmer, IL), and the number of CFU was assessed using the agar plate method. XLD Agar was used for assessment CFU to obtain separate colonies and prevent their swarming. The cut pieces of the catheters were then measured and the number of CFU was converted to the number of CFU per 1 cm2 (Compère et al., 2009). Test for urease production Strains were inoculated using a pick on Christensen medium (HiMedia, Mumbai, India) and cultivated for 24 h at 37 °C. The diameter of the formed pink zone was assessed (Stankowska et al., 2008). Test for swarming over the various catheters LB Swarming Medium was used for this test. From the agar plate, a strip of the medium was aseptically cut away. The prepared gap was bridged with a piece of the tested catheter. Strains were inoculated using a pick on one half of the agar. After 24 h cultivation at 37 °C, the ability to swarm over the tested catheter was assessed. The strains were divided into three groups: (0) non-swarming strains, (1) swarming strains not able to bridge the catheter, and (2) strains able to bridge the catheter (Sabbuba et al., 2002; Jones et al., 2004) (see Fig. 2). Figure 2 View largeDownload slide Test for swarming over the catheter: (a) strain able to swarm over the catheter; (b) strain not able to swarm over the catheter. Figure 2 View largeDownload slide Test for swarming over the catheter: (a) strain able to swarm over the catheter; (b) strain not able to swarm over the catheter. Antibiotic susceptibility test Susceptibility/resistance of the UTI isolates was assessed qualitatively using the Kirby–Bauer disc diffusion test. The antibiotics tested were cephalothin, co-trimoxazole, ciprofloxacin, gentamicin, cefotaxime, ceftazidime and amoxicillin/clavulanic acid. The zones of inhibition were measured and isolates were assessed as susceptible, intermediate or resistant. Statistical analysis Differences between the group of CAUTI isolates and the control group were assessed by unpaired t-tests and differences in swarming over the catheter within the tested groups were assessed by Kruskal–Wallis anova in the program statistica for Windows. For assessment of relationships and correlations, redundancy analysis (RDA) from the group of generalized linear models was used. RDA, as a method of direct linear analysis, searches in the data set for one or more (independent) gradients that will be optimal predictors of regression models of the response of the strains examined, so that they best express relationships and linkages within species data and among species data and explanatory variables. The statistical significance of ordination models was tested with Monte-Carlo Permutation Tests. These tests refer to the zero hypothesis that primary (species) data are independent of explanatory variables. All permutation tests were calculated on the level of 499 permutations. RDA was performed in the statistical program canoco 4.02 (Ter Braak & Šmilauer, 1999). All tests were examined at the significance level of P = 0.05. Results and discussion Differences between the group of strains isolated from CAUTIs and the control group were found for several virulence factors: TM (P = 0.012, t = 2.548), biofilm formation in BHI with and without glucose supplementation (P << 0.001, t = 5.105 and P << 0.001, t = 6.859, respectively), and swarming over one type of catheter, the Argyle Catheter (P = 0.023, t = 2.297). In all the above differences, the CAUTI isolates showed a higher occurrence of virulence factors. All strains formed biofilm in BHI and BHI-g and the CAUTI isolates formed a higher mass of biofilm in both tested media. Our results of higher biofilm formation in BHI-g medium reflect a relationship between the presence of saccharides in urine and higher incidence of UTIs in patients with diabetes mellitus (Chen et al., 2009). Although bacteria of the genus Proteus can move by means of TM, and this virulence factor plays an important role in biofilm formation in Pseudomonas (Mattick, 2002), it has not previously been explored in detail in Proteus. This virulence factor was more expressed in CAUTI isolates, but none of the isolates moved over the whole plate. Most CAUTI isolates (33.3%) formed a zone of 2–3 cm; most (40%) of the stool isolates formed a zone of 1–2 cm. Differences were also seen between the group of strains isolated from CAUTIs and the control group in adherence to the catheter surface (P << 0.001, t = 5.229), but in this case the control group showed higher expression — greater CFU numbers were present on the catheter. Most of the CAUTI isolates (52%) formed only 0.1–10 CFU cm−2 of the catheter in comparison with 36% of control group isolates, which formed 21–30 CFU cm−2 (Fig. 3). These differences may be caused not only by different growth rates of the particular isolates, but also by differences in the ability to adhere to a particular type of catheter. Figure 3 View largeDownload slide Relative number of CFU attached to the catheter. Light blue, CAUTI isolates; dark blue, control group. Figure 3 View largeDownload slide Relative number of CFU attached to the catheter. Light blue, CAUTI isolates; dark blue, control group. The difference in SM was not statistically significant (P = 0.074, t = 1.8). A large proportion (88.2%) of strains from CAUTIs swam over the whole plate, as compared with only 66% of stool strains. There were no differences in other virulence factors between these two groups of strains (SW, urease production and swarming over all other types of catheters): 89.2% of strains from UTIs swarmed over the whole plate, and 90% of strains from stool samples swarmed over the whole plate. The ability to swarm in CAUTI as well as in stool isolates is in agreement with data from other studies (Müller, 1986; Paerson et al., 2010) and the similar numbers may be explained by the fact that most UTIs brought about by bacteria from the genus Proteus come from the intestinal microbial community. Although SW is typical for the genus Proteus, about 10% of isolates of both groups did not swarm. From all strains, 98 strains from CAUTI and 50 strains from stool samples showed all types of movement; 101 and 50 isolates, respectively, showed SM and TM. Figures 4 and 5 show the results of SW over various types of catheters in the CAUTI and control isolates, respectively. The ability of the isolates to swarm over particular catheters was different in both tested groups of isolates [H6 (n = 714) = 2337 864, P << 0.01; H6 (n = 350) = 8107 407, P << 0.01, respectively]. These differences may be caused by different hydrophobicity of the catheters and type of material. Surface hydrophobicity varies even among catheters made from the same material, but provided by different suppliers (Hola et al., 2009). Our results correspond with data published previously (Sabbuba et al., 2002; Jones et al., 2004). The CAUTI isolates showed greatest SW over Kendall Catheter-type Curity catheters (91.2%), and worst SW over Meditec Foley catheters (2%). The strains from stool samples swarmed worst over latter (10%). Stool isolates showed the best SW over Unomedical Latex Foley catheters (88%). A summary of the results is shown in Figures 4 and 5. Regarding differences between the group of CAUTI isolates and control group, there were no statistically significant differences in most of the catheters. The only catheters where the CAUTI isolates swarmed more easily than the control group isolates were Kendall Argyle catheters, where the difference was statistically significant (P = 0.023, t = 2.297), and Kendall Curity catheters, where the difference was not significant (P = 0.087, t = 1.724). Figure 4 View largeDownload slide Determination of SW over various types of urinary catheters in CAUTI isolates. 0, nonmotile isolates; 1, strains swarmed on the inoculated half of the agar, but did not cross the catheter bridge; 2, strains swarmed on the inoculated half of the agar, and crossed the catheter bridge and swarmed over the second half of the agar. Figure 4 View largeDownload slide Determination of SW over various types of urinary catheters in CAUTI isolates. 0, nonmotile isolates; 1, strains swarmed on the inoculated half of the agar, but did not cross the catheter bridge; 2, strains swarmed on the inoculated half of the agar, and crossed the catheter bridge and swarmed over the second half of the agar. Figure 5 View largeDownload slide Determination of SW over various types of urinary catheters in the control group. 0, nonmotile isolates; 1, strains swarmed on the inoculated half of the agar, but did not cross the catheter bridge; 2, strains swarmed on the inoculated half of the agar, and crossed the catheter bridge and swarmed over the second half of the agar. Figure 5 View largeDownload slide Determination of SW over various types of urinary catheters in the control group. 0, nonmotile isolates; 1, strains swarmed on the inoculated half of the agar, but did not cross the catheter bridge; 2, strains swarmed on the inoculated half of the agar, and crossed the catheter bridge and swarmed over the second half of the agar. Antobiotic resistance was found more often in strong biofilm formers than in weak biofilm formers. This trend was strongest for co-trimoxazole and ciprofloxacine. Strong urease production was present in all isolates, and only a small proportion of the isolates showed moderate urease production (5.9% of CAUTI isolates and 4% of stool isolates). Figure 6 shows relationships and correlations within the data of CAUTI isolates based on RDA. The close or similar direction of the arrows reflects correlation of the data; the contrasting orientation of the arrows reflects a negative correlation. The length of the arrow reflects the strength of impact of a given variable. SW correlated very well with the ability to swarm over the various catheters (range of Pearson correlation coefficient: r = |0.69; 0.83|) and with SM (r = 0.65). Biofilm production in the two media, BHI and BHI-g, correlated very well (r = 0.87), but a higher mass of formed biofilm was found in BHI-g for all isolates, CAUTI as well as the control group. Catheter type correlated very well with each other (r = |0.76; 1.0|), except for the Meditec Foley Catheter (highest correlation with MPI Foley Catheter; r = 0.72). Biofilm production in BHI and BHI-g also correlated well with SM (r = 0.81 and r = 0.73, respectively). Variability within the strains of P. vulgaris appeared to be dependent on other factors, which are not included in this analysis and should be evaluated in detail. Figure 6 View largeDownload slide RDA of relationships and correlations within the data of CAUTI isolates. BHI-g, brain heart infusion + 4% glucose. Figure 6 View largeDownload slide RDA of relationships and correlations within the data of CAUTI isolates. BHI-g, brain heart infusion + 4% glucose. It appears that the catheter material does not play a role in prevention of swarming over the catheter, but does plays a role in preventing attachment of biofilm-positive strains. The silicon-coated Meditec Foley Catheter showed the best properties as regards prevention of swarming. The silver-hydrogel coating does not seem to have any effect and is thus not cost-effective. The data presented show differences among individual isolates as well as between a group of CAUTI isolates and control group isolates, especially regarding swarming over various types of catheters. This may play an important role in choice of appropriate catheters and the management of CAUTIs in practice. Choice of the appropriate catheter could decrease the number of CAUTIs, not only reducing the number of nosocomial infections but also reducing the cost of their treatment. Conclusions Because catheters represent a suitable surface for bacterial attachment and enable easier colonization of the urinary tract, nosocomial infections of the urinary tract are becoming increasingly common. This is demonstrated even more strongly in long-term catheterized patients (Bryers, 2008). The results of this study show that the genus Proteus is well adapted to survival in the host body. Identifying and understanding particular virulence factors is necessary for prevention and treatment of CAUTIs. Choice of the catheter based on knowledge of its properties, especially regarding bacterial adherence and swarming ability, could decrease the risk of CAUTIs in patients. This is an important aim as the lower number of nosocomial infections will reduce stress for patients, and also save money connected with treatment of these infections. Acknowledgements The research was funded by project IGA MZ NS-9678. References Bryers J.D. ( 2008) Medical biofilms. Biotechnol Bioeng  100: 1– 18. Google Scholar CrossRef Search ADS PubMed  Chen S.L. Jackson S.L. Boyko E.J. ( 2009) Diabetes mellitus and urinary tract infection: epidemiology, pathogenesis and proposed studies in animal models. J Urol  182: S51– S56. Google Scholar CrossRef Search ADS PubMed  Compère V. Legrand J.F. Guitard P.G. Azougagh K. Baert O. Ouennich A. Fourdrinier V. Frebourg N. Dureuil B. ( 2009) Bacterial colonization after tunneling in 402 perineural catheters: a prospective study. Anesth Analg  108: 1326– 1330. Google Scholar CrossRef Search ADS PubMed  Costerton J.W. ( 1999) Introduction to biofilm. Int J Antimicrob Agents  11: 217– 221. Google Scholar CrossRef Search ADS PubMed  Gorman S.P. Tunney M.M. ( 1997) Assessment of encrustation behaviour on urinary tract biomaterials. J Biomater Appl  12: 136– 166. Google Scholar CrossRef Search ADS PubMed  Hola V. Ruzicka F. Tejkalova R. Kadlec R. ( 2009) The surface charge of the urinary catheters and biofilm adhesion. ASM Conference Biofilms , Abstract book. American Society for Microbiology, Cancún, Mexico, pp. 151. Hola V. Ruzicka F. Horka M. ( 2010) Microbial diversity in biofilm infections of the urinary tract with the use of sonication techniques. FEMS Immunol Med Microbiol  59: 525– 528. Google Scholar CrossRef Search ADS PubMed  Jones B.V. Young R. Mahenthiralingam E. Stickler D.J. ( 2004) Ultrastructure of Proteus mirabilis swarmer cell rafts and role of swarming in catheter-associated urinary tract infections. Infect Immun  72: 3941– 3950. Google Scholar CrossRef Search ADS PubMed  Liedl B. ( 2001) Catheter-associated urinary tract infections. Curr Opin Urol  11: 75– 79. Google Scholar CrossRef Search ADS PubMed  Mattick J.S. ( 2002) Type IV pili and twitching motility. Annu Rev Microbiol  56: 289– 314. Google Scholar CrossRef Search ADS PubMed  Morris N.S. Stickler D.J. ( 1998) Encrustation of indwelling urethral catheters by Proteus mirabilis biofilms growing in human urine. J Hosp Infect  39: 227– 234. Google Scholar CrossRef Search ADS PubMed  Müller H.E. ( 1986) Occurence and pathogenic role of Morganella–Proteus–Providencia group bacteria in human feces. J Clin Microbiol  23: 404– 405. Google Scholar PubMed  O'May G.A. Jacobsen S.M. Stickler D.J. Mobley HLT Shirtliff M.E. ( 2008) Complicated urinary tract infections due to catheters. The Role of Biofilms in Device-Related Infections  ( Shirtliff M. Leid J.G., eds), pp. 123– 165. Spinger-Verlag, Berlin. Google Scholar CrossRef Search ADS   Paerson M.M. Rasko D.A. Smith S.N. Mobley HLT ( 2010) Transcriptome of swarming Proteus mirabilis. Infect Immun  78: 2834– 2845. Google Scholar CrossRef Search ADS PubMed  Rashid M.H. Kornberg A. ( 2000) Inorganic polyphosphate is needed for swimming, swarming and twitching motilities of Pseudomonas aeruginosa. P Natl Acad Sci USA  97: 4885– 4890. Google Scholar CrossRef Search ADS   Rózalski A. Sidorczyk Z. Kotelko K. ( 1997) Potential virulence factors of Proteus bacilli. Microbiol Mol Biol Rev  61: 65– 89. Google Scholar PubMed  Sabbuba N. Hughes G. Stickler D.J. ( 2002) The migration of Proteus mirabilis and other urinary tract pathogens over Foley catheters. BJU Int  89: 55– 60. Google Scholar CrossRef Search ADS PubMed  Stankowska D. Kwinkowski M. Kaca W. ( 2008) Quantification of Proteus mirabilis virulence factors and modulation by acylated homoserine lactones. J Microbiol Immunol Infect  41: 243– 253. Google Scholar PubMed  Stepanović S. Vuković D. Hola V. Di Bonaventura G. Djukić S. Cirković I. Ruzicka F. ( 2007) Quantification of biofilm in microtiter plates: overview of testing conditions and practical recommendations for assessment of biofilm production by staphylococci. APMIS  115: 891– 899. Google Scholar CrossRef Search ADS PubMed  Stickler D.J. ( 2008) Bacterial biofilms in patiens with indwelling urinary catheters. Nat Clin Pract Urol  5: 598– 608. Google Scholar CrossRef Search ADS PubMed  Ter Braak CJF Šmilauer P. ( 1999) Canoco for Windows; [CD-ROM]; Version 4.02 . Centre for Biometry Wageningen, Wageningen. Trautner B.W. Darouiche R.O. ( 2004) Catheter-associated infections: pathogenesis affects prevention. Arch Intern Med  164: 842– 850. Google Scholar CrossRef Search ADS PubMed  © 2012 Federation of European Microbiological Societies.
TI - Virulence factors in Proteus bacteria from biofilm communities of catheter-associated urinary tract infections
JF - Journal of the Endocrine Society
DO - 10.1111/j.1574-695X.2012.00976.x
DA - 2012-07-01
UR - https://www.deepdyve.com/lp/oxford-university-press/virulence-factors-in-proteus-bacteria-from-biofilm-communities-of-psLOSK5Ewv
SP - 343
EP - 349
VL - 65
IS - 2
DP - DeepDyve
ER -