Epidemiological Link Between Wastewater and Human
Malihe TalebiÆ Fateh RahimiÆMohammad KatouliÆ Roland Mo¨llbyÆMohammad R. Pourshafie
Received: 21 September 2007 / Accepted: 21 November 2007 / Published online: 22 February 2008
ÓSpringer Science+Business Media, LLC 2008
Abstract We investigated the prevalence of vancomycin-resistant enterococci (VRE) isolated from wastewater (n =593) and clinical (n =450) samples, and the genetic linkage between the isolates was compared. Out of the total samples, 38 Enterococcus faecium (3.6%) from sewage (n=19) and clinical (n=19) isolates were found to be highly resistant to vancomycin. The majority of the VRE isolates from the two sources showed distinct phenotyping and genotyping patterns. At the same time, one common pulsed-field gel electrophoresis pattern was found among the VRE obtained from wastewater and human clinical isolates, suggestive of an epidemiological link.
Enterococci are members of the normal intestinal micro-flora in humans and animals . They are major nosocomial pathogens with the ability to acquire high-level resistance to antimicrobial agents and to exchange genetic information [12,16]. It has been suggested that the envi-ronmental enterococcal strains could serve as a reservoir of
the antimicrobial resistance genes, which can be transferred to other potentially pathogenic bacteria . Growing evidence is supporting the fact that the horizontal transfer of antibiotic resistance genes is occurring in nature between the clinical and nonclinical microorganisms . Such an event from the standpoint of community health is of great importance.
The isolation of nosocomial enterococcus resistant to multiple antibiotics, especially vancomycin, has become worrisome worldwide . Moreover, enterococcus in ani-mal feces, food of aniani-mal origin, municipal and hospital sewage treatment plants (STPs), which could persist in the ecosystem for a long time, is under intensive investigations , especially when the multiple-drug-resistant entero-cocci could disperse through STPs to surface waters and may eventually find a way of returning to humans .
This study was undertaken to determine the genetic relatedness between the VRE isolates of sewage and human urinary tract infection isolates.
Materials and Methods
Collection and Identification of Sewage Sample
Sampling was carried out during 2005. Sewage samples were collected from three urban STPs located at different parts of Tehran. Six samplings were done from three urban STPs (two each). Wastewater samples were collected into 250-mL sterile bottles. All samples were refrigerated and transported to the microbiology laboratory. Samples were diluted five-fold with phosphate-buffered saline before filtration on a 0.45-lm membrane (Millipore Corporation, Bedford, MA) and were cultured in mEnterococcusagar (Becton Dickinson and Co., Sparks, MD) plates as described before . M. TalebiF. RahimiM. R. Pourshafie (&)
Department of Microbiology, Pasteur Institute of Iran, Tehran, Iran
Faculty of Science, Health and Education, University of the Sunshine Coast, Maroochydore, Queensland 4558, Australia
Microbiology, Tumor and Cell Biology Centre, Karolinska Institutet, S-171 77 Stockholm, Sweden
Collection of Clinical Sample
Clinical samples (n =450) included in this study com-prised specimens from patients admitted to three major hospitals in Tehran during 2005. Vancomycin-resistant enterococci (VRE) isolation from the clinical specimens was carried out by standard microbiological techniques. Briefly, clinical samples were inoculated on blood agar. The suspicious colonies from these cultures were Gram stained. Gram-positive cocci were tentatively identified as enterococcus on the basis of negative catalase test, growth in 6.5% sodium chloride, hydrolysis of esculin, and PYR test.
DNA Extraction and Polymerase Chain Reaction
DNA for polymerase chain reaction (PCR) was extracted by the boiling method as described previously . PCR assay for species identification and vancomycin resistance genes (vanA, vanB) was performed with the specific primers as reported by others [6, 8]. Primer sequences (vanA: 50-CATGAATAGAATAAAAGTTGCAATA-30, 50 -CCCCTTTAACGCTAATACGATCAA-30 vanB : 50-GT GACAAACCGGAGGCGAGGA-30, 50-CCGCCATCCTC CTGCAAAAAA-30 ) were from the published sequences of the genes . PCR assay was performed in a total vol-ume of 25lL containing 10 mMTris-HCl (pH 8.3), 1.5 mM MgCl2, 0.2 mM each dNTPs), 0.5 U of Taq DNA
poly-merase (HT Biotechnology, Cambridge, UK) and each primer (40 pmol). The PCR cycle was done as follows: initial denaturation at 94°C for 3 min, 30 cycles of dena-turation at 94°C for 1 min, annealing at 54°C for 1 min, and extension at 72°C for 1 min and a final extension at 72°C for 7 min.E. faecalisV583,E. faeciumBM4147, and E. faecalisATCC29212 were used as the reference strains.
Antimicrobial Susceptibility Test
The isolates were primarily identified as enterococci by biochemical tests and PCR and were subsequently tested for resistance to vancomycin (30lg) by the disk diffusion agar method. VRE were also tested with teicoplanin (30 lg), gentamicin (120 lg), erythromycin (15 lg), cipro-floxacin (5lg), tetracycline (30lg), chloramphenicol (30 lg), and ampicillin (10 lg) (BioRad, Hercules, CA). Minimum inhibitory concentrations (MICs) of vancomycin and teicoplanin were determined by the Etest (AB Biodisk, Solna, Sweden) method on Mueller-Hinton agar according to the manufacturer’s instructions. MIC results were interpreted according to Clinical and Laboratory Standards Institute guidelines . Both antibiotics were tested in the range 0.25–256lg/mL.
The isolates were typed using the PhP-RF plates (PhPlate AB, Stockholm, Sweden). These are microplates with eight sets of dehydrated reagents that are used to measure the kinetics of bacterial metabolism of 11 substrates specifically chosen to differentiate between strains of enterococci . For each bacterial isolate, it yields a biochemical fingerprint made of 11 quantitative data, which were used with the PhPlate software to calculate the level of similarity between the tested isolates and to identify different phenotypes. Preparation and inoculation of the plates were done according to the manufacturer’s instructions. The inoculated microplates were incubated at 37°C and the absorbance value (A620) of each well was
measured at 16, 40, and 64 h . The mean value of all three readings was calculated and the similarity between the strains was calculated as the correlation coefficient after pairwise comparison of the strains. The similarity matrix was then clustered according to the unweighted pair group with arithmetic averages (UPGMA) to obtain a dendrogram. An identity level (ID level =0.965) was set up for the system after testing five isolates in duplicate. The mean similarity between the compared isolated minus 2SD was taken as the ID level of the system. Isolates showing similarity to each other above this level were considered as identical (Common Biochemical Pheno-types: C-BPT). The diversity of the bacterial populations was calculated as Simpson’s index of diversity (Di) .
The optical readings, calculation of correlation coeffi-cients, diversity indexes, and clustering were performed using the PhPWin software (PhPlate Microplates Tech-niques AB, Sweden).
Pulsed-Field Gel Electrophoresis
were placed in the wells of 1% agarose in 0.5%9TBE and electrophoresed with switch times ramped from 5 s to 35 s at 6 V with a run time of 27 h at 16°C in the Bio-Rad CHEF-DRIII system. DNA from Salmonella choleraesuis serotype Branderup H9812 (Pulsenet, www.cdc.gov/ pulsenet) was included as the molecular size marker. The gels were then stained with ethidium bromide and the restricted DNA was visualized with ultraviolet light . The banding patterns were clustered by the UPGMA method using the software Gelcompar II version 4.0 (Applied Maths, Sint-Matens-latem, Belgium) and inter-preted using the guidelines set out by Tenover et al.. Isolates having the same banding patterns or differing by one to three bands were regarded as identical and were assigned the same type. The isolates that differed by more than three bands were not considered to be related and were regarded as different types .
Prevalence of Enterococci Species and VRE Isolates
A total of 593 enterococcal isolates were collected from three STPs in Tehran. Among the samples, 315 (53%)E. faecium and 72 (12.5%)E. faecalis were found. The rest were other species of enterococcus.
Nineteen VRE were isolated from the STPs and they all wereE. faecium. Of the total 450 clinical isolates, 365 and 85 were E. faecalis and E. faecium, respectively. The clinical samples included 380 (85%) from urine, 25 (5.5%) wounds, 20 (4.5%) blood cultures, 10 (2%) body fluid, 10 (2%) sputum, and 5 (1%) abscess. The 19 clinical VRE were E. faecium, which were obtained from the urinary tract infections. The vancomycin and teicoplanin MIC for all VRE isolated from humans and wastewater wasC128 andC48, respectively.
Comparison of Antimicrobial Resistance of Human and Wastewater VRE
A high level of resistance to ciprofloxacin (95%), genta-micin (95%), and erythromycin (95%) was found among the clinical VRE isolates. On the contrary, the level of resistance to tetracycline (37%) and chloramphenicol (5.2%) was low. About 89% of the human VRE isolates were concomitantly resistant to Am/Gm/Cip/E. Similarly, the number of VRE isolated from wastewater that was concomitantly resistant to these four antibiotics was also the same (Table1). The most significant difference between the levels of antibiotic resistance was observed with chloramphenicol. The level of resistance in the wastewater isolates was 37% as compared to 5.2% in the
human clinical isolates. All clinical and sewage VRE iso-lates were resistant to ampicillin and teicoplanin.
PhP Typing of Total VRE Isolates
The results of PhP typing for the clinical and sewage VRE isolates showed the presence of diverse (diversity index, Di=0.97) PhP types among the isolates from both origins. The diversity index (Di) for clinical and STPs isolates were 0.95 and 0.96, respectively. PhP typing discriminated the 38 isolates into 22 types with 12 single types (32%) and the remaining 10 common types (CTs) constituting 68% of the VRE isolates (Fig. 1). The results showed that each com-mon BPT (C-BPT) comprised two to three strains.
Some of the isolates collected from the clinical and STPs were found to have the same C-BPT (i.e., CT5, CT6). In general, when isolates showed the same pulsed-field gel electrophoresis (PFGE) pattern, they also showed the same PhP types (i.e., CT4, CT9). However, in some cases exceptions could be seen in the isolates with the same PFGE types but different PhP types (i.e., ST9, ST11) (Fig.1).
All 38 human and wastewater VRE isolates were ana-lyzed by PFGE. The analysis showed 23 PFGE patterns including 17 and 6 pulsotypes for VRE isolated from humans and wastewater, respectively. The 19 VRE from the wastewater showed 4 common pulsotypes with a predominant type (M) that comprised 8 isolates. These eight VRE isolates were obtained from different STPs at different locations and times. On the other hand, the 19 human VRE contained only one common pulsotype, J, comprising 2 isolates. The rest of the human VRE iso-lates showed unique PFGE patterns. Type J was found to be the only pattern to be shared by the human and Table 1 Patterns of antibiotic resistance of 38 vancomycin-resistant enterococci isolated from sewage treatment plants (STP) and clinical isolates in Tehran
Pattern of antibiotic resistance
Number (%) of isolates
Am/Gm/Cip/E/C/Te 0 4 (21)
Am/Gm /Cip/E /C 1 (5.3) 3 (15.8)
Am/Gm/Cip/E/Te 7 (37.1) 1 (5.3)
Am/Gm/Cip/E 9 (47) 9 (47.4)
Am/Cip /E 1(5.3) 2 (10.5)
Am/Gm 1 (5.3) 0
sewage VRE isolates (Fig.2). The five isolates of human and wastewater, within the pulsotype J, had the same antimicrobial resistance pattern, which only was sensitive to chloramphenicol.
The prevalence of VRE isolates from patients and the environment is different in Europe and the United States. In the United States, VRE isolates are restricted to hospitalized patients, whereas in European countries VRE have been isolated from different environment conditions . As of now, there is no information available about the interrelation of the VRE in waste-water and human clinical isolates in Iran. In our study,
the prevalence of VRE among the enterococcal isolates in sewage (3%) was almost similar to that in patients with urinary tract infections (4%). Moreover, in both isolation sources, only E. faecium were found to be highly resistant to vancomycin. Other investigators have also indicated that VR E. faecium were the predominant isolates in STPs, while other VRE species such as E. faecalis and E. hirae in sewage samples were detected [1, 10].
A high rate of concomitant resistance to erythromycin and gentamicin (89%) was observed in our VRE isolates. It has been reported that gentamicin- and erythromycin-resistant determinants can be co-transferred at a high frequency rate along with vancomycin-resistant genes [2,
20]. Simultaneously, a high rate of antimicrobial resis-tance was observed with ampicillin/ciprofloxacin and Fig. 1 A unweighted pair
group with arithmetic averages dendrogram showing the vancomycin-resistant Enterococcus faeciumstrains isolated from different sewage treatment plants and clinical samples in Tehran during 2005. PFGE, pulsed-field gel electrophoresis; Gm,
gentamicin; Cip, ciprofloxacin; Te, tetracycline; E,
ampicillin/quinolone in both clinical and STPs types, which is in accordance with the reports published by others [11, 21].
The results of PFGE and PhP typing showed that some isolates were found at different sampling occasions and in different STPs, indicating a high prevalence of certainE. faecium(i.e., Pulsotype J, K, and PhP type C5, C9) in the municipal STPs.
PFGE analysis revealed an extensive genetic diversity among the isolates. The isolates from the STPs were more homogeneous than the isolates from clinical. The absence of a dominant VRE pulsotype among the human isolates was consistent with the fact that there had been no VRE outbreaks in Tehran and consequently no predominant bacterial clone dissemination.
In general, the majority of the VRE isolates showed distinct pulsotypes as an indication of lack of a close genetic relationship between the isolates from the human and STP. Moreover, the majority of isolates from the two sources may have acquired vancomycin resistance ele-ments independently, possibly by horizontal transfer. Nevertheless, J pulsotype was the only pattern shared by the clinical and STP isolates, which may imply that this pulsotype was genetically stable, even in the harsh
condition of STP. The presence of one common pulsotype with five isolates may be noteworthy, considering the fact that only a total of 38 VRE isolates from the clinical sources and STPs were studied. Furthermore, this pulso-type was found in two out three STPs, which further supports the spread of this clonal type in Tehran.
Although these five isolates had the same PFGE pattern but a different PhP and antibiotic resistance pattern (human isolates were sensitive, whereas the sewage isolates were resistant to chloramphenicol) by the two wastewater VRE isolates may suggest the acquisition of new traits by these isolates surviving in the wastewater. Obtaining these characteristics may, in turn, allow better fitness and increase the chance of passing of this clone of bacteria to other milieu such as surface water, as indicated by others .
In conclusion, the presence of VR E. faecium with PFGE type J in 13% of VRE isolates may provide evidence for the epidemiological link between the isolates from the STP reservoirs and human urinary tract infections. More-over, a possible transmission route for this clone from hospital patients via urban sewage to surface water may occur. The presence of this clone in surface water is under investigation.
Acknowledgments This work was supported in part by World Health Organization, Eastern Mediterranean Regional Office grant no. R6/18/3, Swedish International Development Cooperation Agency (Sida) grant no. 6342, and Ministry of Health of Iran, under-secre-tariat of research. We are also thankful to Dr. Mohammad Rahbar from National Reference Laboratory for collecting the samples.
1. Blanch AR, Caplin JL, Iversen A, Ku¨hn I, Manero A, Taylor HD, Vilanova X (2003) Comparison of enterococcal populations related to urban and hospital wastewater in various climatic and geographic European regions. J Appl Microbiol 94:994–1002 2. Borgen K, Sorum M, Wasteson Y, Kruse H, Oppegaard H (2002)
Genetic linkage betweenerm(B) andvanAinEnterococcus hirae of poultry origin. Microb Drug Resist 8:363–368
3. Cetinkaya Y, Falk P, Mayhali G (2000) Vancomycin resistant Enterococci. Clin Microbiol Rev 13:686–707
4. Iversen A, Ku¨hn I, Franklin A, Mo¨llby R (2002) High prevalence of vancomycin-resistant Enterococciin Swedish sewage. Appl Environ Microbiol 68:2838–2842
5. Iversen A, Ku¨hn I, Rahman M, Franklin A, Burman LG, Olsson-Liljequist B, Torell E, Mo¨llby R (2004) Evidence for transmis-sion between humans and the environment of a nosocomial strain ofEnterococcus faecium. Environ Microb 6:55–59
6. Jackson CR, Fedorka-Cray PJ, Barrett JB (2004) Use of genus-and species-specific Multiplex PCR for identification of entero-cocci. J Clin Microbiol 42:3558–3565
7. Jarnheimer P, Ottoson J, Lindberg R (2004) Fluoroquinolone antibiotics in a hospital sewage line, occurrence, distribution and impact on bacterial resistance. Scand J infect Dis 36:752–755 8. Kariyama R, Mitsuhata R, Chow JW, Clewell DB, Kumon H
(2000) Simple and reliable multiplex PCR assay for surveillance isolates of vancomycin-resistant enterococci. J Clin Microbiol 38:3092–3095
9. Ku¨hn I, Iversen A, Mo¨llby R (2003) The PhenePlate system for studies of the diversity of enterococcal populations from the food chain and the environment. Intern J Food Microbiol 88:189–196 10. Ku¨hn I, Iversen I, Finn M, Greko C, Burman LG, Blanch AR, Vilanova X, Manero A, Taylor H, Caplin J, Domingues L, Mo¨llby R (2005) Occurrence and relatedness of vancomycin-resistantEnterococciin animals, humans, and the environment in different european regions. Appl Environ Microbiol 71:5383– 5390
11. Leavisa HL, Bontena MJM, Willemsa RJL (2006) Identification of high-risk enterococcal clonal complexes: global dispersion and antibiotic resistance. Curr Opin Microb 9:5454–5460
12. Manero A, Vilanova X, Cerda`-Cue´llar M, Blanch AR (2006) Vancomycin- and erythromycin-resistant enterococci in a pig farm and its environment. Environ Microb 8:667–674
13. Manson JM, Smith JMB, Cook GM (2004) Persistence of van-comycin-resistant Enterococci in New Zealand broilers after discontinuation of avoparcin use. Appl Environ Microbiol 70:5764–5768
14. Nallapareddy SR, Wenxiang H, Weinstock GM, Murray E (2005) Molecular characterization of a widespread, pathogenic, and antimicrobial resistance receptiveEnterococcus faecalislineage and dissemination of its putative pathogenicity island. J Bacteriol 187:5709–5718
15. National Committee for clinical Laboratory Standards (2000) Methods for dilution antimicrobial susceptibility tests for bacteria that aerobically, 5th edn. Approved Standards M7-A5, vol. 20, no. 2. Wayne, Pennsylvania: National Committee for Clinical Laboratory Standards
16. Oprea SF, Zaidi N (2004) Molecular and clinical epidemiology of vancomycin resistantEnterococcus faecalis. J Antimicrob Che-mother 53:626–630
17. Schwartz DC, Cantor CR (1984) Separation of yeast chromosome sized DNAs by pulsed field gradient gel electrophoresis. Cell 37:67–75
18. Sneath PHA, Sokal RR (1973) Numerical taxonomy. San Fran-cisco: The W. H. Freeman Co
19. Tenover FC, Arbeit RD, Goering RV, Mickelsen PA, Murray BM, Persing DH (1995) Interpreting chromosomal DNA restriction patterns produced by pulsed field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 33:2233–2239 20. Tomita H, Pierson C, Lim SK, Clewell DB, Ike Y (2002) Possible connection between a widely disseminated conjugative genta-micin resistance (pMG1-like) plasmid and the emergence of vancomycin resistance inEnterococcus faecium. J Clin Microbiol 40:3326–3333
21. Torell E, Kuhn I, Olsson-Liljequist B, Haeggman S, Hoffman BM, Lindahl C, Burman LG (2003) Clonality among ampicillin-resistantEnterococcus faeciumisolates in Sweden and relation-ship with ciprofloxacin resistance. Clin Microbiol Infect 9:1011– 1019