OCCURRENCE AND CHARACTERISATION OF UROPATHOGENIC ESCHERICHIA COLI IN URINARY TRACT INFECTIONS

Transcription

1 Indian Journal of Medical Microbiology, (2003) 21 (2): Original Article OCCURRENCE AND CHARACTERISATION OF UROPATHOGENIC ESCHERICHIA COLI IN URINARY TRACT INFECTIONS R Raksha, *H Srinivasa, RS Macaden Abstract Purpose: To study the occurrence of Uropathogenic Escherichia coli (UPEC) in cases with urinary tract infections. Methods: A total of 220 cases from urinary tract infections and 50 stool samples from apparently healthy individuals were included. The colonies identified as Escherichia coli were screened for virulence factors namely haemolysin, Mannose Resistant and Mannose Sensitive Haemagglutination (MRHA, MSHA), Cell surface hydrophobicity and Serum resistance by recommended methods. Results: Among 220 cases 91(41.36%) were haemolytic, 68(30.9%) showed MRHA, 58(26.36%) were cell surface hydrophobicity positive and 72(32.72%) were serum resistant. In 50 controls 3(6%) were haemolytic, 6(12%) showed MRHA, 9(18%) showed cell surface hydrophobicity and 12(24%) were serum resistant. The difference between cases and controls for haemolysis and MRHA were significant (p<0.001 and p<0.01 respectively). A total of 14 atypical Escherichia coli were isolated from urine and all showed the presence of one or the other virulence markers.out of 18 mucoid Escherichia coli isolated 10 were serum resistant. Interestingly among the 15 Escherichia coli isolated from patients with pyelonephritis 8 were UPEC. Conclusions: Out of 220 urinary isolates, 151 could be labelled as UPEC. Key words: Uropathogenic Escherichia coli, mannose resistant haemagglutination, mannose sensitive haemagglutination, haemolysin, fimbriae Urinary tract infection is one of the most important causes of morbidity and mortality. E.coli is the most frequent urinary pathogen isolated from 50% - 90% of all uncomplicated urinary tract infections. 1 E.coli present in the gastrointestinal tract as commensals provide the pool for initiation of UTI. It has been traditionally described that certain serotypes of E.coli were consistently associated with uropathogenicity and were designated as Uropathogenic E.coli (UPEC). These isolates express chromosomally encoded virulence markers. These markers of UPEC are expressed with different frequencies in different disease states ranging from asymptomatic bacteriuria to chronic pyelonephritis. In the late 1970s it was recognized for the first time that E.coli strains causing urinary tract infections typically agglutinate human erythrocytes despite the presence of Mannose 2 and this was mediated mainly by fimbriae. Subsequently an array of virulence factors have been proposed as virulence markers for uropathogenic isolates of E.coli. It is now recognized that there are a subset of faecal E.coli having the above mentioned factors which can colonise periurethral area, *Corresponding author Department of Microbiology, St. John s Medical College, Bangalore , Karnataka, India. Received: Accepted: enter urinary tract and cause symptomatic disease. These are currently defined as UPEC. 3 However, most of these studies have been carried out on patient isolates and no studies have been carried out on commensal /gut isolates. In addition the relative importance of these bacterial factors has not been validated. In the present study an attempt has been made to answer these two questions with respect to four proposed virulence factors namely haemolysin, haemagglutination, cell surface hydrophobicity and serum resistance. Materials and Methods Cases The study was conducted in a tertiary care hospital for a period of two years from November 1998 to October Urine samples from a total of 220 cases of UTI belonging to the Departments of Urology, Nephrology, Medicine, Paediatrics and few other departments of the hospital were studied for the detection of virulence markers of E.coli. A total of 50 stool samples from apparently healthy individuals who had come for routine health check up were screened for E.coli and studied further for virulence markers. E.coli were identified as described by Bailey and

2 April, 2003 Raksha et al - Characterisation of Uropathogenic E.coli 103 Scott 4 with modifications. Care was taken to collect clean catch midstream urine, which was immediately processed or if refrigerated, was processed within one hour. Considering the importance of quantitation, bacteriological loop was calibrated using Evan s blue dye solution. 5 A battery of biochemical tests were used for the identification of E.coli. 6 The isolates were maintained by inoculating nutrient agar butts and stored at room temperature. Standard uropathogenic E.coli serotypes O4 and O6 and E.coli ATCC were used as controls for detection of virulence markers. Colony Count All urine samples having growth of single morphotype of colony with counts 10 5 colonies/ml were considered significant. A count of colonies/ml was considered probably significant and a count of <10 4 colonies/ml as insignificant growth of perineal or urethral flora. 7 The additional features noticed in terms of colony morphology were, whether they were mucoid/ nonmucoid and biochemical reactions, whether showed hemolysis on sheep blood agar after overnight incubation, whether they were typical and atypical isolates. An isolate was considered as typical if it was a lactose fermenter and aerogenic and atypical if it was nonlactose fermenter and anaerogenic. Atypical isolates were further tested for sorbitol, cellobiose and adonitol fermentation. E.coli thus obtained from cases with significant or probably significant counts were screened for virulence markers. Detection of Virulence Factors Haemolysin The cytolytic protein toxin secreted by most haemolytic E.coli isolates is known as alpha haemolysin. 8 Haemolysin was detected by determining a zone of lysis around each colony on 5% sheep blood agar plates after overnight incubation. Haemagglutination The haemagglutination was detected by clumping of erythrocytes by fimbriae of bacteria in presence of D-mannose. The method followed was according to Siegfred et al. 9 Bacteria grown on TSI agar medium were subcultured onto MacConkey s plates (MA) and incubated at 37 C overnight. E.coli grown on MA plates were inoculated into 5 ml of phosphate buffered saline ph 7.4 (PBS) and incubated for 5 days at 37 C to get fimbriae enriched E.coli. The pellicle formed on the surface was noted and subcultured onto colonization Factor Antigen (CFA) agar and incubated overnight at 37 C. Five millilitre of group A positive venous blood was collected using disposable syringe from a voluntary donor and added to an equal amount of Alsever s solution. This was washed three times and 3% erythrocyte suspension was made with PBS. Controls used were ATCC E.coli for mannose sensitive haemagglutination, UPEC serotypes 06 and 011 as MRHA positive controls. The procedure of Siegfred et al 9 was modified and carried out on VDRL slides instead of microtitre plates. Interpretation Haemagglutination was considered to be mannose resistant when it occurred in presence of D-mannose and mannose sensitive when it was inhibited by D-mannose. The slides in which isolates gave a weak reaction were placed at 4 C for five minutes and enhancement of weak reactions were checked and considered positive if obvious. Isolates which gave doubtful reactions were retested. Cell surface Hydrophobicity Salt aggregation test: Bacteria were tested for their hydrophobic property by using different molar concentrations of ammonium sulphate. 9,10 Those which aggregated with salt particles and formed clumps were considered hydrophobic. A strain of E.coli which was haemolytic, MRHA positive and consistently positive for cell surface hydrophobicity was used as positive control. A strain which was nonlytic, negative for MRHA and consistently negative for cell surface hydrophobicity was used as negative control. E.coli grown on MA plates were inoculated into 1mL of PBS ph 6.8 and turbidity were matched with McFarland tubes 6 and 7 to get a colony count of 5x10 9 colonies/ml. Different molar concentrations of ammonium sulphate namely 1M, 1.4M and 2M concentrations were prepared. Forty microlitre of 0.2 M PBS ph 6.8 was taken in first column of VDRL slide. 40 l of 1M, 1.4M and 2M concentrations of ammonium sulphate were taken in each well of other columns of VDRL slide. Forty microlitre of E.coli suspension was added to each of these wells. The clumps formed in different molar concentrations of ammonium sulphate was observed under Leights Wetzlar inverted microscope - binocular (0.32) at 20x magnification and scored positive on one to four scale (1+ to 4+). Strains were considered hydrophobic if they aggregated in concentrations of 1.4M. Isolates which gave doubtful results were retested. Serum Bactericidal Assay This assay was done according to Siegfried et al 9

3 104 Indian Journal of Medical Microbiology Vol.21, No.2 with modifications, such as Mueller Hinton agar plates were used in place of blood agar, and a shaker water bath was used in placed of roller angle shaker. According to method of Benge 10 strains were termed serum sensitive if viable count 11 dropped to 1% of initial value and resistant if 90% of organisms survived after 180 minutes. Statistical Analysis Chi square test was used to compare the occurrence of virulence markers in cases and controls. P value less than 0.05 was considered significant. Results E.coli isolates from bacteriuria Of the 220 urine samples in which E.coli was isolated, 191 had significant bacteriuria with counts 10 5 colonies/ml and 29 had probably significant bacteriuria with counts between colonies/ml. Phenotypic and biochemical characteristics of E.coli A total of 18 Esch.coli isolates were mucoid lactose fermenting colonies. All controls were non mucoid lactose fermenters. Out of 220 isolates 14 were non lactose fermenters and anaerogenic and considered as atypical E.coli. Virulence markers of Uropathogenic E.coli (UPEC) obtained from cases and controls The performance of standard uropathogenic E.coli isolates were satisfactory in all the 4 assays. Occurrence of UPEC with virulence markers in different combinations among cases and controls is shown in table. Haemolysin Among the 220 cases 91(41.36%) were haemolytic. Among 50 controls, 3(6%) were haemolytic. The difference between cases and controls for haemolysin production was highly significant (p < 0.001). Haemagglutination A total of 68 (30.9%) among 220 cases and 6 (12%) among 50 controls showed mannose resistant haemagglutination. There was a significant (p<0.01) difference in MRHA between cases and controls. Table : Occurrence of uropathogenic E.coli with virulence markers among cases and controls Virulence markers Cases Controls One marker Lytic 16 2 MRHA 7 1 CSH 10 1 SR 35 4 Total 68 8 Two markers Lytic + MRHA 15 0 Lytic + SR 6 1 Lytic + CSH 7 0 MRHA + CSH 1 2 MRHA + SR 2 1 CSH + SR 5 4 Total 36 8 Three markers L+MRHA+CSH 23 0 L+MRHA+SR 12 0 L+CSH+SR 4 0 MRHA+CSH+SR 0 2 Total 39 2 L - Lytic, MRHA - Mannose resistant haemagglutination, CSH - Cell surface hydrophobicity, SR-Serum resistant Cell surface Hydrophobicity A total of 58 (26.36%) among 220 cases and 9(18%) among 50 controls were cell surface hydrophobic. There was no significant difference (p > 0.05) between cases and controls, for cell surface hydrophobicity. An E.coli isolate U56 served as a positive control and gave 1+, 3+, 4+ grade clumping in1m, 1.4M and 2M ammonium sulphate concentrations respectively. An isolate U 3055 from a case served as a negative control and gave negative results in all 3 molar concentrations. Serum resistance Among 220 cases, 72 (32.72%) E.coli were serum resistant and among 50 controls, 12 (24%) E.coli were serum resistant. There was no significant difference (p>0.05) in cases and controls for serum resistance. The results of virulence markers among E.coli isolates from cases and controls are shown graphically in Fig. 1 and occurrence of multiple virulence factors are shown in Fig. 2.

4 April, 2003 Raksha et al - Characterisation of Uropathogenic E.coli 105 Figure 1 : Comparison of individual virulence markers of Escherichia coli from cases and controls % HLY haemolysin, HA Mannose Resistant haemagglutination, CSH Hydrophobicity, SBA Serum Resistance Figure 2 : Comparison of positivity for virulence markers of Escherichia coli among cases and controls % UPEC and pyelonephritis Out of 15 pyelonephritis cases eight were positive for E.coli having one or other virulence markers. Among the eight cases of uropathogenic E.coli, six were acute pyelonephritis and two were chronic pyelonephritis cases. Atypical / Mucoid E.coli and virulence Of the 18 mucoid lactose fermenters, 10 were serum resistant and three were haemolytic, two were MRHA, one was MSHA. Among the 14 non lactose fermenters all had one or more virulence markers. Discussion HLY HA CSH SBA All 4 Any 3 Any 2 Any 1 CASES Cases CONTROLS Controls Considering the high degree of morbidity and mortality of UTIs the subject of uropathogenic E.coli is receiving increasing attention. Cell morphology and molecular biology studies have revealed that uropathogenic E.coli express several surface structures and secrete protein molecules some of them cytotoxic, peculiar to the strains of E.coli causing UTI. 1 Hence it is important to identify UPEC from non UPEC isolates in the urinary samples. All -ve CAS Cases E S CONT Controls R OL S Results in our study showed that a large number of urinary isolates from cases had more than one virulence markers. In this case control study we conclude that UPEC strains are definitely associated with the aetiopathogenesis of UTI. The occurrence of multiple virulence factors in UPEC strains further strengthens the concept of association of UPEC with urinary pathogenicity. It was interesting to note that UPEC with multiple virulence factors were significantly more in cases than in controls. Phenotypic and biochemical characterization revealed interesting findings. A total of 18 urinary isolates were mucoid and all these were obtained from cases only. No mucoid strains were isolated from control group. Further, 10 of these 18 were serum resistant. Mucoid strains are capsulated 12 and capsule confers serum and phagocyte resistance to some E.coli strains. 3 This property is attributed to content of sialic acid which reduces ability of bacterial surface to activate complement by alternative path way. However, 8 mucoid strains were susceptible to serum indicating non capsular factors had a role in serum resistance. Thus the conventional phenotypic marker such as capsule is also an important virulence factor for UPEC. In the present study, 14 (6.36%) of 220 isolates were atypical E.coli. Altered phenotype could be due to an altered genetic make up. Bhat et al 13 studied 210 E.coli strains isolated from urine and found 26 (12.4%) to be atypical. In their study, of the 26 atypical urinary isolates 12% were haemolytic and 3 had hydrophobicity (< 0.156) and were MRHA positive and all the atypical E.coli had one or other virulence marker indicating that atypical phenotype probably contributes to their virulence. The cytolytic protein toxin secreted by most haemolytic E.coli strains is a-haemolysin. E.coli also produces cell associated lysin on blood agar plates. and haemolysin cause a clear zone of lysis. 14 In the present study, though the nature of haemolysin was not further characterized it can be considered as cytotoxic necrotising factor ( haemolysin). The difference between cases and controls for production of haemolysin was highly significant (p< 0.001). This was similar to Johnson et al 2 study where haemolysin was produced by 38% of urinary and 12% of faecal isolates. Haemagglutination is mediated by fimbriae. 15 MRHA can be mediated by P fimbriae and also X, FIC, Dr fimbriae. Thus MRHA positive strains can be considered as UPEC most likely having P fimbriae. 2 In the present study there was a significant difference for

5 106 Indian Journal of Medical Microbiology Vol.21, No.2 MRHA between cases and controls (p <0.01). This was similar to a study by Johnson 2 et al where 58% of urinary isolates and 19% of faecal isolates showed MRHA. The expression of type 1 fimbriae is indicated by MSHA. MSHA were more in faecal strains than urinary isolates in our study. More work is required to assess role of MSHA in pathogenecity. The role of cell surface hydrophobicity (CSH) in mediating bacterial adherence to mammalian cells was conceived by Mudd and Mudd. 16 Crystalline surface layers S layer present on both gram negative and gram positive organisms, play a role in this. 17 Hydrophobicity is a recently described novel virulence mechanism by E.coli. In the present study, though there was no significant difference for CSH between cases and controls (p>0.05) more isolates from cases were hydrophobic. Taylor 18 reviewed that bacteria are killed by normal human serum through lytic activity of alternative complement system. Bacterial resistance to killing by serum results from individual or combined effects of capsular polysaccharide, O polysaccharide and surface proteins. 19 Although more E.coli isolates from cases were serum resistant compared to controls the difference was not statistically significant (p>0.05). The O group designation reveals little about a strain. The apparent virulence associated with certain groups may be mediated through other virulence factors like P fimbriae, haemolysin, serum resistance, which are commonly associated with UTI associated strains 2. Therefore serotyping was not attempted in the present study. In this study, eight out of 15 cases of pyelonephritis were caused by uropathogenic E.coli with one or more virulence markers. Further studies need to be done on pyelonephritonegic isolates. In this study, 151 out of 220 isolates had one or more virulence factors. When a comparison was done between urinary and faecal E.coli isolates, haemolysin production, presence of capsule and capacity to cause MRHA emerged as important virulence factors. The haemolysin, especially a-haemolysin, also known as cytotoxic necrotising factor, is strongly proinflammatory leading to secretion of IL - 6 and chemotaxins which sets pace for pathogenesis of renal disease. The capacity to cause MRHA is due to various adhesins mainly P fimbriae, P associated fimbriae, and FIC fimbriae seen in pyelonephritis cases. These adhere to fibronectin on uroepithelial cells contributing to persistence. We believe that the methods of detection of the above mentioned virulence markers is reasonably easy and screening them in a clinical microbiology laboratory is a worthwhile exercise. References 1. Steadman R, Topley N. The virulence of Escherichia coli in urinary tract, Chapter 3 In:Urinary tract infections. 1 st Ed. Brumfitt W, Jeremy MT, Hamilton Miller. Eds (Chapman and Hall publication, London) 1998: Johnson JR., Virulence factors in Escherichia coli urinary tract infection. Clin Microbiol Rev 1991;4: Warren JW. Host parasite interactions and host defence mechanisms, Chapter In:Diseases of the kidney, 6 th ed, Vol 1, Schrier RW, Gottschalk CW.Eds (Little Brown, London) 1997: Baron EJ, Finegold SM, Eds. Microorganisms encountered in urinary tract, Chapter 18. In Bailey and Scotts. Diagnostic Microbiology, 8 th ed. (Mosby Company, St. Louis) 1990: Pezzlo MD Aerobic bacteriology. In:Clinical Microbiology procedure handbook volume I Editor- Henry D Eisenberg. (American Society of Microbiology, Washington DC) 1992: Cowan and Steel s Manual for identification of Medical bacteria, 3 rd ed. Eds. Barrow Gl, Feltham RKA. Eds. (University of Cambridge) Campus MJ, McNamara AM, Howard BJ. Specimen collection and processing, Chapter 11. In: Clinical and pathogenic Microbiology, 2 nd ed. Howard BJ, Kaiser JF, Smith T. (Mosby publication, Philadelphia) 1994: Cavalieri SJ, Bohach GA, Snyder IS. Escherichia coli alpha haemolysin: characteristics and probable role in pathogenicity. Microbiol Rev 1984;48: Siegfried L, Marta Kmetova, Hana Puzova, et al. Virulence associated factors in Escherichia coli strains isolated from children with urinary tract infections. J Med Microbiol 1994;41: Benge GR. Bactericidal activity of human serum against strains of Klebsiella from different sources J Med Microbiol 1988;27:11-15.

6 April, 2003 Raksha et al - Characterisation of Uropathogenic E.coli Cruickshank R, Duguid JP, Swain RHA Medical Microbiology: A guide to laboratory diagnosis and control of infection 11 th ed.( English Language Book society, Churchill Livingstone, Edinburgh) Altwegg and Bockemuhl J. Escherichia and Shigella. Chapter 40. In Topley & Wilsons Microbiology & Microbial infections. Vol -2 Systematic bacteriology. 9 th ed. Leslie Collier, Albert Balows, Eds. (Edward Arnold,London). 1998: Bhat GK, Bhat GM. Atypical Escherichia coli in urinary tract infections.tropical Doctor : Smith HW. The haemolysins of Escherichia coli. J Pathol Bacteriol 1963;85: Duguid JP, Clegg S, Wilson ML. The fimbrial and non fimbrial haemagglutinins of Escherichia coli. J Med Microbiol 1979;12: Mudd S, Mudd EBH. The penetration of bacteria through capillary spaces IV. A kinetic mechanism in interfaces J Exp Med 1924;40: Sleytr B, Messner P. Crystalline surface layers of bacteria. Annual review of Microbiol 1983;37: Taylor PW. Bactericidal and bacteriolytic activity of serum against gram negative bacteria. Microbiol Rev 1983;47: Montenegro MA, Bittersuermann D, Jimmis JK et al. Serum resistance and pathogenicity related factors in clinical isolates of Escherichia coli and other gram negative bacteria. J Gen Microbiol 1985;131: