- Email: [email protected]
A rapid two dot filter assay for the detection of E. coli O157 in water samples
Journal of Immunological Methods 336 (2008) 159–165
Contents lists available at ScienceDirect
Journal of Immunological Methods j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j i m
Research paper
A rapid two dot filter assay for the detection of E. coli O157 in water samples Sujatha Kamma a, Lily Tang a, Kelvin Leung a, Edie Ashton c, Norman Newman d, Mavanur R. Suresh a,⁎ a b c d
Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, Alberta, Canada Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada Provincial Lab of Alberta, University of Alberta, Edmonton, Alberta, Canada Provincial Lab of Alberta, Calgary, Alberta, Canada
a r t i c l e
i n f o
Article history: Received 13 November 2007 Received in revised form 10 April 2008 Accepted 15 April 2008 Available online 15 May 2008 Keywords: E. coli O157 Filter dot assay Colloidal gold
a b s t r a c t E. coli O157:H7 is an enterohemorrhagic bacteria that cause deadly water-borne infections implicated in outbreaks of a wide spectrum of human gastrointestinal diseases. It is therefore important to have a rapid convenient, simple and sensitive range of detection of E. coli O157:H7. A new E. coli O157 MAb designated P124 was developed for ultrasensitive detection of E. coli O157 in water, apple juice and beef for routine use. A prototype filter dot assay was designed with anti-E. coli O157 MAb bound to 0.2 µm nitrocellulose filter disk as the capture antibody. A 100 ml water sample spiked with 1–50 CFU of E. coli O157 either in the presence or absence of other non-specific bacteria were filtered for capture of the pathogen on the antibody coated nitrocellulose disk. The detection of the pathogen was successfully accomplished by the same antibody both as a capture and detecting antibody as a homosandwich. In a non-enriched format, detection of E. coli was possible with a sensitivity of 2500 CFU/100 ml. Ultrasensitive detection of ~1 CFU/100 ml sample could be achieved by a prior pathogen enrichment step before the addition of the labeled antibody. The design of this diagnostic test is based on the common architecture of all bacteria, viruses and spores, namely the manifestation of repeat lipopolysaccharide epitopes on the surface. We have developed an easy-to-use two dot visual filter assay for translation into current water testing in public health laboratories to detect E. coli O157:H7. In a 5 h assay ~1 CFU and ~5 CFU of E. coli O157 could be detected in 100 ml of water or juice and lake samples respectively. This simple homosandwich enrichment strategy can also be used to detect low levels of other water-borne pathogens. © 2008 Elsevier B.V. All rights reserved.
1. Introduction E. coli O157:H7 causes a wide spectrum of human diseases, including bloody and non-bloody diarrhea, hemorrhagic colitis, occasional kidney failure, hemolytic uremic syndrome (HUS) and death at times (DeCludt et al., 2000; Shelton and Karns, 2001) due to ingestion of meat (Willshaw et al., 1994), water, and uncooked fruits and vegetables (Pebody et al., 1999). Outbreak of E. coli O157:H7 infections through drinking water was first reported in the USA in 1989 (Swerdlow et al., 1992). O157 contamination of drinking and recreational water has emerged as important cause of human disease (Ackman et al., 1997; Armstrong et al., 1996; ⁎ Corresponding author. Tel.: +780 492 9233; fax: +780 492 1217. E-mail address: [email protected] (M.R. Suresh). 0022-1759/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2008.04.011
Chalmers et al., 2000; Friedman et al., 1999). E. coli O157 present in the drinking water offered to livestock contributed to the prevalence of infection in animals leading to the contamination of meat products and the environment (Elder et al., 2000). An extreme example of the dangers this bacteria poses was seen in the Walkerton, Ontario and in the neighboring Maritimes, described as Canada's worst-ever outbreak of E. coli contamination (Hrudrey et al., 2003). E. coli O157: H7 is a top disease concern of the multi-billion dollar North American cattle industry. In Canada, Alberta has the greatest population of cattle and hence live stock generated manure contamination of water from surface drainage channels has precipitated one of the highest levels of gastroenteritis resulting from E. coli O157:H7 and Salmonella species (Khakhria et al., 1996). E. coli O157:H7 is the most common strain of Shiga toxin producing enterohemorrhagic
160
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
E. coli (STEC) in United States, Canada and United Kingdom (Kaper, 1998). Several recent outbreaks of gastrointestinal diseases caused by STEC has highlighted the threat these organisms pose to public health and are also considered as potential biowarfare agents (Trochimchuk et al., 2003). Many detection methods have been employed to rapidly detect low levels of pathogens in food, beverages and water. Current techniques include traditional enrichment and plating methods with selective media such as Sorbitol MacConkey agar and Rainbow agar (Manafi and Kremsmaier, 2001; Meng et al., 2001; Novicki et al., 2000). E. coli O157 is particularly difficult to confirm from enrichment cultures, even with highly selective media due to the problem of high background levels of competing microorganisms including other type of serotypes of E. coli. A variety of immunological methods have been developed for the detection and enumeration of E. coli O157 whole bacteria (Chapman et al., 1991, 1997; Park and Durst, 1999; Todd et al., 1988). The common denominator among all methods was the use of monoclonal or polyclonal anti-E. coli O157 antibodies to selectively capture, or capture and label by sandwich assay of E. coli O157 whole bacteria. Enrichments and plating often take 24–48 h to identify the organism. It is therefore important that rapid sensitive methods are developed to detect E. coli O157 during outbreaks, surveillance and quality control to prevent costly errors and fatalities due to delayed detection. Hence we have developed a simple 5 h two dot assay on nitrocellulose filter disks using a growth medium. This enrichment allows viable bacteria to grow for a short period prior to detection in a visual immunoassay (Fig. 1).
2. Materials and methods 2.1. Bacterial strains Strains used for the development of two dot filter assay were obtained from Dr. Roger Johnson's group (Health Canada Labs, Winnipeg, Canada) and from Dr. Newman & Edie Ashton (Provincial lab of Alberta). The strains of bacteria include E. coli O157:H7 (ATCC 43895), E. coli non-O157: non-H7 (O6: H34, O26:H11), E. coli non-O157: H7 (O18:H7, O91:H7), Salmonella urbana, S. typhimurium, Pseudomonas aeruginosa and 5 other laboratory strains of E. coli namely, JM87, Top 10 F, BL21DE3, JM109 and MKH14. All the strains used for the experiments were grown on Trypticase soy agar plates at 37 °C overnight followed by growth in Trypticase soy broth (TSB, pH 7.2–7.3) for 18 h with gentle agitation at 200 rpm. The classical dilution and colony counting methods were supplemented with direct estimation of fluorescent bacterial counts for accurate bacterial concentration measurements as described previously by our group (Guttikonda et al., 2004). 2.2. P124 MAb conjugation with colloidal gold A 10 ml aliquot of 20 nm colloidal gold (Sigma–Aldrich St. Louis, MO, USA) was adjusted to pH 9.0 with 0.1 M K2CO3. The P124 MAb at 10 µg/ml was adjusted to pH 9.0 with 100 nM K2CO3 and 10 ml of the antibody solution was added to 10 ml of colloidal gold. The solutions were shaken gently for 15 min at RT. Subsequently, 1 ml of 10% BSA pH 9.0 was
Fig. 1. Diagrammatic representation of our two dot filter assay for the detection of E. coli O157:H7.
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
added and kept at RT for 10 min and then centrifuged at 12,000 g for 45 min at 4 °C. The supernatant was discarded and the pellet was re-suspended in 2 ml of 0.1% BSA with 0.1% NaN3 in PBS (Hermanson, 1996). 2.3. Preparation of filter disks Sterile 0.45 µm nitrocellulose filter disks of 25 mm diameter (Fisher Scientific, Canada) were spotted with approximately 4 µg of P124 MAb and 4 µg of goat anti-mouse IgG spacially separated from each other (for two dot filter assay). The disks were left to air dry for 30 min at room temperature and blocked with 3% BSA for 2 h at room temperature. The disks were given a gentle wash with PBS buffer pH 7.2 and stored in the refrigerator for further use. 2.4. Two dot filter assay using colloidal gold antibodies This is an simple alternative non-enzymatic method of detection. The spiked water samples with 0–50 CFU/100 ml were filtered employing the Millipore manifold vacuum filtration unit using a nitrocellulose filter disk spotted with P124 MAb and goat anti-mouse IgG as separate spots. The filtered disks were incubated for 5 h or 24 h in TSB in a sterile plastic dish. Following a wash step, 250 µl of the P124-GC (colloidal gold) was added to the filter disks and gently shaken for 10 min at RT. The results were read visually for the appearance of one or two pink dots on the filter paper. 2.5. Detection of E. coli O157:H7 in apple juice and ground beef We also conducted limited comparisons assays with E. coli O157 spiked into apple juice or ground beef which are other sources that often gets contaminated with this pathogen. E. coli O157 was mixed with excess of the non-pathogenic and other pathogenic strains of E. coli in ratios of 1:5 CFU, 1:25 CFU, 1:50 CFU, 1:250 CFU and 1:500 CFU. Each of the above mixture of E. coli was spiked in 100 ml of apple juice and filtered on nitrocellulose disks spotted with P124 MAb and goat anti-mouse IgG. Following amplification of E. coli for 5 h and 24 h respectively, the bacteria were detected by colloidal gold conjugated to anti-E. coli O157 P124 MAb. Similarly, various concentrations of E. coli O157 were spiked into 25 g of ground beef and were enriched with 250 ml of TSB for 5 and 24 h respectively. Subsequently the meat was carefully separated from the broth and the entire enriched broth (~240 ml) was filtered on to the P124 and Goat anti-mouse IgG coated filter disks. The food enrichments were filtered with a steel mesh with pores to prevent clogging of the nitrocellulose test filters. The filter disks were incubated for 1 h by gentle shaking to allow the binding of E. coli to the antibody spot. Following a wash, P124-Colloidal gold was added and incubated for 10 min at RT. 2.6. Detection of E. coli O157 in lake water using two dot filter assay Sterile 0.45 µm nitrocellulose filter disk of 47 mm diameter (Millipore Canada Ltd) used in our local water testing labs were coated with approximately 10 µg (instead of 4 µg) of P124 MAb (anti-E. coli O157) and 10 µg of goat anti-mouse IgG in two separate spots. The disks were left to air dry for 30 min
161
at RT and blocked with 3% skim milk for 2 h at RT. The disks were given a gentle wash with PBS pH 7.2 and stored in the refrigerator for further use. A 10 ml, 25 ml and 100 ml each of a lake water sample collected from various regions of Northern Alberta were filtered onto the coated nitrocellulose disks using a filtration manifold and magnetic funnels obtained from the Provincial Lab of Alberta–Edmonton branch. The microbes on the nitrocellulose disks were amplified using 3 to 3.5 ml of the tryptic soy broth pH 7.2 for a minimum of 5 h and a maximum of 18 h (overnight). Following amplification, the disks were washed three times with PBS pH 7.2 and 5 µg of colloidal gold particles conjugated to P124 MAb per filter was added to the disks and given a gentle shake until the pink spot was developed. Cross-correlation of the two dot filter assay with other standard methods: The amplified TS broths were retained in the refrigerator for additional tests with established methods. Approximately 10–50 µl of TS broth was plated on the CT-SMAC plates and incubated at 37 °C overnight. Plates were examined for the presence of colorless colonies (non-sorbitol-fermenters). A representative number of bacteria from each colony type (approximately 10–15 colonies from each plate) were transferred to non-selective agar (such as TSB agar), spread for isolation to ensure purity and incubated at 37 °C overnight. The colonies were then screened by the oxidase test. (The oxidase test involved dipping a sterile swab into oxidase reagent (tetramethyl-pphenylenediamine dihydrochloride) and then touching the growth from each colony on the TSB agar plate. No more than 10–15 s were allowed for a reaction to develop. Purple color indicated oxidase positive and no color were scored oxidase negative. E. coli O157 are all oxidase negative, therefore any colonies which developed a purple coloration were discarded. All the reagents were tested with a positive and negative control before using them for screening the plated colonies. The oxidase reagent was tested against Pseudomonas aeruginosa as a positive control organism and E. coli 25922 as a negative control. Those colonies which yielded a negative oxidase test were screened by a slide agglutination test using the E. coli O157 antisera (BD Biosciences, USA). The E. coli O157 antiserum was tested against E. coli O157 and E. coli 25922 which are the positive and negative controls respectively. E. coli O157 produces a very fine agglutination reaction. 2.7. Cross-correlation of two dot filter assay using magnetic beads The amplified TS broth was further tested with the immunomagnetic bead separation procedure (Dynal ASA, Oslo, Norway). A 20 µL of anti-E coli O157 Dynal beads of 1 µm diameter were dispersed into an Eppendorf tube. Using a sterile pipette, 1 ml of amplified TS broth was added into each of the tubes containing anti-E. coli O157 beads. The samples were mixed thoroughly and incubated at RT for 10–30 min on a rocking platform. Following a very thorough wash (3 times) the beads were resuspended in 100 µl of PBS and inoculated to selective media CT-SMAC (Cefixime-Tellurite Sorbitol MacConkey Agar) and/or CHROMagar. (Cefiximine, Tellurite were from Dynal ASA, Oslo, Norway. SMAC was from Sigma– Aldrich St. Louis, MO, USA and CHROMagar was from Dynal ASA, Oslo, Norway). The colonies were subjected to the same degree of cross-checks as above. The positive and the negative
162
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
controls were 10–20 CFU of E. coli O157:H7 and E. coli 25922 in 1 ml of the TS broth respectively.
device for use in farms, wells, streams and lakes where portability is desirable when sample transportation to the laboratory is not practical.
3. Results We have recently developed a monoclonal and bispecific antibody against the E. coli O157:H7 by traditional hybridoma techniques (Guttikonda et al., 2007). The P124 monoclonal antibody was shown to bind to several different strains of E. coli O157 as well as purified LPS. Employing this newly developed anti-E. coli O157:H7 P124 MAb we explored two designs suitable for adaptation to routine water testing laboratories. Both these designs utilized the nitrocellulose filters currently used in water testing labs. 3.1. Two dot filter assay using colloidal gold In the two dot filter assay a single pink dot indicated the absence of the pathogen (binding of the P124-GC to the control goat anti-mouse IgG) and two pink dots indicated positive confirmation for E. coli O157:H7 (Figs. 2 and 3). The second pink dot is due to the sandwich formation wherein the pathogen was captured by P124 MAb and detected by P124-GC. Our results in two dot filter assay format with 5 h and 24 h amplification followed by the detection with MAb P124 conjugated to colloidal gold showed the detection of ~1 CFU of bacteria even at shorter periods of amplification step (Fig. 2). The background was clean and scoring was unambiguous. The overnight assay also gave similar results (data omitted). The assay using colloidal gold in both the 5 and 24 h assays had a detection limit of ~1 CFU/100 ml of water with clear visualization of end point from control. High sensitivity, shorter time and direct adaptation to water testing labs are the major advantages of this simple method. Additional advantages include the formulation of this assay into a unitized
3.2. Detection of E. coli O157:H7 strain in mixed cultures with excess of non-pathogenic and pathogenic strains of E. coli In nature the pathogenic strain could exist among several other microbes including the benign coliforms. Evaluation of mixed cultures of E. coli to investigate the detection of E. coli O157 was carried out as a part of our further prototype validation studies. The experiments were carried out in a two dot filter format. E. coli O157 that belong to H7 type was mixed with excess of 2 strains of E. coli non-O157:non-H7, 2 strains of E. coli non-O157: H7, Salmonella typhimurium species and Pseudomonas strain and 5 other laboratory strains of E. coli in 1:1 (E. coli O157:H7: Non-O157 strains), 1:5, 1:10, 1:50, 1:100 and 1:500 ratios. The above mixture of E. coli was spiked in 100 ml of water and filtered on nitrocellulose disks. Following amplification of E. coli for 5 h, the bacteria were detected by tracer P124-colloidal gold conjugate. The results show a significant detection of O157 even in a 1:500 fold excess of other strains (Fig. 3). 3.3. Detection of E. coli O157:H7 in apple juice and ground beef Since some of the E. coli O157 outbreaks were found associated in unpasteurized apple juice and ground beef (Willshaw et al., 1994), we conducted the filter dot assay by spiking the pathogen in commercial apple juice and ground beef. Filter dot assay was able to detect ~ 1 E. coli O157:H7 in the presence of 500 CFU of other bacteria in apple juice even after 5 h incubation. However, the detection limit of the E. coli O157 in 25 g of beef samples was at a higher level of 25 CFU
Fig. 2. E. coli O157 detection in water samples using our two dot filter assay following a 5 h enrichment. One to 50 CFU of E. coli O157 was filtered on P124 coated filter disk. Following the amplification of bacteria for 5 h, O157 was detected with tracer P124-colloidal gold.
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
163
Fig. 3. E. coli O157 detection simulating mixed microbes natural conditions: 1 CFU of E. coli O157 was mixed with various ratios of 2 strains of E. coli non-O157:nonH7, 2 E. coli non-O157: H7, 1 Salmonella species, one Pseudomonas strain and 5 other laboratory strains of E. coli strains of bacteria and filtered on P124 coated filter disk. Following the amplification of bacteria for 5 h, O157 was detected with tracer P124-colloidal gold.
(Fig. 4). We proposed testing in other media for increasing the detection sensitivity. 3.4. Detection of E. coli O157 in lake water using two dot filter assay Based on the promising results with the spiked water samples we evaluated several Northern Alberta lake water
samples for E. coli O157 in a small scale pilot experiment. The pH of the lake water samples ranged from ~ pH 6.5 to ~pH 10.5. None of the lake water samples tested in our limited sampling was positive for E. coli O157. However the pathogen spiked lake water samples gave the expected two dot confirmation. The two dot filter assay showed concordance with the CT-SMAC and Dynabeads cross-correlation results.
Fig. 4. E. coli O157 detection in 25 g of ground beef. E. coli O157 was added to the 25 g ground beef sample and amplified for 5 h at 37 °C. The meat was removed from the broth and filtered on P124 coated filter disk. Following further amplification of bacteria for 1 h, O157 was detected with tracer P124-colloidal gold.
164
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
3.5. Cross-correlation of the two dot filter assay with other standard methods and magnetic beads Cross-correlation using standard methods and magnetic beads showed that our two dot visual assay yielded consistent results and was a reliable method for E. coli O157 detection. 4. Discussion Infection with E. coli O157 can be asymptomatic but more commonly is characterized by watery diarrhoea or haemorrhagic colitis (HC). Approximately 5 to 30% of cases go on to develop Haemolytic Uremic Syndrome (HUS), characterized by haemolytic anaemia, low platelet counts and renal impairment. E. coli O157 is recognized as the most common cause of renal failure in children in the Europe and North America UK. The severity of the symptoms and high mortality rates (10–30%) make this pathogen an important public health concern (Bolton and Aird, 1998). The infectious dose of this organism is thought to be as few as 10 viable organisms and food contaminated with very low numbers of organisms can initiate disease. In this report we described a rapid, sensitive simple visual assay for routine detection of this pathogen in water, fruit juice and meat samples. This simple assay has a sensitivity below 10 CFU and can potentially detect 1 CFU/100 ml. Minimally trained personnel can perform the assay and score the endpoints for quality control and quality assurance. In all our experiments the E. coli O157:H7 sample preparations at various dilutions were estimated using the epifluorescence counting for cross-correlation of culture plate count estimates (Guttikonda et al., 2004). This is critical to ensure that the number of E. coli O157 that we were spiking into 100 ml of tap water, juice and meat broth was indeed comparable by the two methods. Correlation results indicated a non-significant variance in the bacterial counts between the plating and the epifluorescence counts. We acknowledge that it is difficult to ensure that one E. coli particle was indeed a single cell and not a clump of two or three cells and hence we have used the term ~1 cell to indicate very low levels of detection. In order to investigate the utility of the immunoconjugate of P124 MAb for the detection of E. coli O157 bacterium we developed a variant of the homosandwich assay to include an amplification step using growth medium to allow the viable bacteria to divide. In the non-amplified approach, the bacteria were directly filtered on to the nitrocellulose disks spotted with specific and non-specific antibodies. A comparison of amplified and non-amplified two dot filter assays was initially performed and the results showed the improved ultrasensitivity of the amplified two dot filter assay approach. At 5 h and longer amplification periods, E. coli detection limit was found to be ~1 bacterium. In the non-amplified assay the detection sensitivity of the assay was ~ 2500 CFU/100 ml (data not shown). Nitrocellulose filter disks are routinely used in the water testing laboratories for detection of total coliforms. In our provincial water testing lab ~100,000 water samples are processed each year from N. Alberta alone. Extrapolating this to all states and provinces in Northern America, nearly 10 million tested are likely done on an annual basis. Hence, a non-enzymatic end point was designed to eliminate the additional enzymatic step. Ease and simplicity in detecting O157 in water was achieved with the colloidal gold which does not require any enzymatic
reactions and extra steps of incubations. The two dot filter assay is a one-step detection method following bacterial amplification and requires only 5 h incubation for the E. coli O157:H7 to be detected. Thus the assay can be completed in few hours from sample collection if a unitized device is developed based on the current format. Based on our studies the two dot filter assay has sensitivity, simplicity and rapidity. When detecting human pathogens, faster detection is very much an asset for determining the immediate course of treatment. Most of the present commercial detection methods take 16–24 h for enrichment to detect the E. coli O157 bacteria, whereas our amplified two dot filter disk assay takes only 5 h with ultrasensitive detection of ~1 CFU of E. coli O157 in 100 ml of water sample. If higher levels are suspected, the assay could potentially be completed in less than 5 h. In our experiments with mixed cultures of other E. coli and bacteria, the two dot filter disk assay format was shown to be sensitive even with 500 fold higher concentrations of other E. coli and bacterial strains. The reliability of the testing kits depends on the complexity of procedure, sensitivity, specificity and duration of assay. Previous methods of isolation of E. coli O157 from food was based on enrichment culture, concentration using the immunomagnetic separation (IMS) technique (Wright et al., 1994) and sub-culture on sorbitol MacConkey agar containing cefixime and potassium tellurite. The Dynal kit requires their specific instruments for performing the assay. In addition the protocol for immunomagnetic separation requires an 18–24 h amplification step preceeded by the separation of E. coli O157, and visualization on specialized plates (CT-SMAC or CHROMagar). Our two dot visual assay is an ultrasensitive detection method that could be adopted for routine water testing of E. coli O157:H7. Samples with algal interferences gave us inconclusive results when water samples are processed without filtering. However filtering out algae gave consistent results and we plan to investigate this further. However most of the other lake water samples gave us two dots when negative lake samples were spiked with known number of E. coli O157:H7. Our results show that following enrichment, two dot filter assay can be used to detect the E. coli O157:H7 pathogen with good sensitivity. Our two dot filter assay could detect 25 CFU O157 in 25 g of ground beef and can be improved by further optimization. The detection limit in beef was less sensitive than in water. We plan further studies to determine if the bacteria adhere to the sedimented bulk of the meat sample and reduce detection sensitivity. The key feature of our simple visual system is that it is based on scoring simple end points as one dot (−ve) or two dot (+ve) (Fig. 1). Like ELISA, the visual system can process a large number of samples per day. Our two dot filter assay system is simple to perform and interpret, inexpensive and relatively rapid (~5 h). It requires neither costly special equipment nor highly qualified operators. It is also highly reproducible. Compared to other commercial visual detection formats, the manufacturing cost per test could be low and practical for routine use. We propose that this simple format is an ideal test for implementation as routine test for water samples in both developed and developing countries to manage E. coli O157. It is also possible that the above filter dot concept could be extended to other pathogens if convenient amplification broth is available. Alternatively a macro array of different
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
antibody spots could be constructed for multiple pathogen testing in water and food samples. Acknowledgements MRS would like to thank NCE-CBDN for the grant support and the CIHR-Industry chair award for salary support. Thanks are due to Dr. Roger Johnson and Erika Lingohr from Health Canada Lab for the E. coli and Salmonella strains. References Ackman, D.S., Marks, S., Mack, P., Caldwell, P., Root, T., Birkhead, G., 1997. Swimming associated hemorrhagic colitis due to Escherichia coli O157: H7 infection: evidence of prolonged contamination of a fresh water lake. Epidemiol. Infect. 119, 1. Armstrong, G.L., Hollingsworth, J., Morris Jr, J.G., 1996. Emerging foodborne pathogens: Escherichia coli O157:H7 as a model of entry of a new pathogen into the food supply of the developed world. Epidemiol. Rev. 18, 29. Bolton, F.J.H., Aird, H., 1998. Verocytotoxin producing Escherichia coli O157: public health and microbiological significance. Br. J. Biomed. Sci. 55, 127. Chalmers, R.M., Aird, H., Bolton, F.J., 2000. Waterborne Escherichia coli O157. J. Appl. Microbiol. 88, 124S (Suppl). Chapman, P.A., Siddons, C.A., Zadik, P.M., Jewes, L., 1991. An improved selective medium for the isolation of Escherichia coli O157. J. Med. Microbiol. 35, 107. Chapman, P.A., Cerdan, A.T., Malo, P.M., Siddons, C.A., Harkins, M., 1997. Use of commercial enzyme immunoassays and immunomagnetic separation system for detection of E. coli O157:H7 in bovine fecal sample. Appl. Environ. Microbiol. 63, 2549. DeCludt, B., Bouvet, P., Mariani-Kurkdjian, P., Grimont, F., Grimont, P.A., Hubert, B., Loirat, C., 2000. Haemolytic uraemic syndrome and Shiga toxin producing E. coli infection in children in France. Epidemiol. Infect. 124, 215. Elder, R.O., Keen, J.E., Siragusa, E.R., Barkocy-Gallagher, G.A., Koohmaraie, M., Laegreid, W.W., 2000. Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing. Proc. Natl. Acad. Sci. U. S. A. 97, 2999. Friedman, M.S., Roels, T., Koehler, J.E., Feldman, L., Bibb, W.F., Blake, P., 1999. Escherichia coli O157:H7 outbreak associated with an improperly chlorinated swimming pool. Clin. Infect. Dis. 29, 298. Guttikonda, S., Wang, W.W., Suresh, M.R., 2004. Molecular zipper assays: a simple homosandwich with the sensitivity of pcr. J. Pharm. Pharmaceut. Sci. 7 (S1), 7. Guttikonda, S., Tang, X.L., Yang, M.B., Armstrong, G.D., Suresh, M.R., 2007. Monospecific and bispecific antibodies against E. coli O157 for diagnostics. J. Immunol. Methods 327 (1–2), 1.
165
Hermanson, T.J., 1996. Preparation of colloidal gold labeled proteins. Bioconjugate Techniques. Academic press, San Diego, California, USA. Hrudrey, S.E., Payment, P., Hych, P.M., Gillham, R.W., Hrudey, E.J., 2003. A fatal waterborne disease epidemic in Walkerton, Ontario: comparison with other waterborne outbreaks in the developed world. Water Sci. Technol. 47, 7. Kaper, J.B., 1998. Enterohemorrhagic E. coli. Curr. Opin. Microbiol. 1, 103. Khakhria, R., Woodward, D., Johnson, W., 1996. Salmonella, Shigellae, pathogenic E. coli, Campylobacter and Aeromonas identified in Canada: annual summary 1994. National Laboratory for Enteric Pathogens. Manafi, M., Kremsmaier, B., 2001. Comparative evaluation of different chromogenic/flurogenic media for detecting Escherichia coli O157:H7 in food. Int. J. Food. Microbiol. 71, 257. Meng, J., Doyle, M.P., Zhao, T., Zhao, S., 2001. Enterohemorrhagic Escherichia coli. In: Doyle, M.P. (Ed.), Food Microbiology:Fundamentals. ASM Press, Washington, DC, p. 193. Novicki, T.J., Daily, J.A., Mottice, S.L., Carroll, K.C., 2000. Comparison of MacConkey Agar and a two step method which utilizes ELISA toxin testing and a chromogenic agar to detect and isolate E.coli. J. Clin. Microbiol. 38, 547. Park, S., Durst, R.A., 1999. Immunoliposome sandwich assay for the detection of Escherichia coli O157:H7. Anal. Biochem. 280, 151. Pebody, R.G., Furtado, C., Rojas, A., McCarthy, N., Nylen, G., Ruutu, P., Leino, P., Chalmers, R., De Jong, B., Donnelly, M., Fisher, I., Gilham, C., Graverson, L., Cheasty, T., Willshaw, G., Navarro, M., Salmon, R., Leinikki, P., Wall, P., Bartlett, C., 1999. An international outbreak of Vero cytotoxin producing E.coli O157 infection amongst tourist: a challenge for the European infectious disease surveillance network. Epidemiol. Infect. 123, 217. Shelton, D.R., Karns, J.S., 2001. Qualitative detection of E.coli O157 in surface waters by using immunomagnetic electrochemiluminescence. Appl. Environ. Microbiol. 67 (7), 2908. Swerdlow, D.L., Woodruff, B.A., Brady, R.C., Griffin, P.M., Tippen, S., Donnell Jr, H.D., Geldreich, E., Payne, B.J., Meyer Jr., A., Wells, J.G., Greene, K.D., Bright, M., Bean, N.H., Blake, P.A., 1992. A waterborne outbreak in Missouri of Escherichia coli O157:H7 associated with bloody diarrhea and death. Ann. Intern. Med. 117, 812. Todd, E.C.D., Szabo, R.A., Peterkin, P., Sharpe, A.N., Parrington, L., Bundle, D., Gidney, M.A.J., Perry, M.B., 1988. Rapid hydrophobic grid membrane filterenzyme-labeled antibody procedure for identification and enumeration of Escherichia coli 0157 in foods. Appl. Environ. Microbiol. 54, 2536. Trochimchuk, T., Fotheringham, J., Topp, E., Schraft, H., Leung, K.T., 2003. A comparison of DNA extraction and purification methods to detect Escherichia coli O157:H7 in cattle manure. J. Microbiol. Methods 54, 165. Willshaw, G.A., Thirlwel, J., Jones, A.P., Rowe, B., Hickey, M., 1994. Vero cytotoxin-producing Escherichia coli O157 in beefburgers linked to an outbreak of diarrhoea, haemorrhagic colitis and haemolytic uraemic syndrome in Britain. Lett. Appl. Microbiol. 19 (5), 304. Wright, D.J., Chapman, P.A., Siddons, C.A., 1994. Immunomagnetic separation as a sensitive method for isolating Escherichia coli from food samples. Epidemiol. Infect. 113, 31.
Contents lists available at ScienceDirect
Journal of Immunological Methods j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j i m
Research paper
A rapid two dot filter assay for the detection of E. coli O157 in water samples Sujatha Kamma a, Lily Tang a, Kelvin Leung a, Edie Ashton c, Norman Newman d, Mavanur R. Suresh a,⁎ a b c d
Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, Alberta, Canada Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada Provincial Lab of Alberta, University of Alberta, Edmonton, Alberta, Canada Provincial Lab of Alberta, Calgary, Alberta, Canada
a r t i c l e
i n f o
Article history: Received 13 November 2007 Received in revised form 10 April 2008 Accepted 15 April 2008 Available online 15 May 2008 Keywords: E. coli O157 Filter dot assay Colloidal gold
a b s t r a c t E. coli O157:H7 is an enterohemorrhagic bacteria that cause deadly water-borne infections implicated in outbreaks of a wide spectrum of human gastrointestinal diseases. It is therefore important to have a rapid convenient, simple and sensitive range of detection of E. coli O157:H7. A new E. coli O157 MAb designated P124 was developed for ultrasensitive detection of E. coli O157 in water, apple juice and beef for routine use. A prototype filter dot assay was designed with anti-E. coli O157 MAb bound to 0.2 µm nitrocellulose filter disk as the capture antibody. A 100 ml water sample spiked with 1–50 CFU of E. coli O157 either in the presence or absence of other non-specific bacteria were filtered for capture of the pathogen on the antibody coated nitrocellulose disk. The detection of the pathogen was successfully accomplished by the same antibody both as a capture and detecting antibody as a homosandwich. In a non-enriched format, detection of E. coli was possible with a sensitivity of 2500 CFU/100 ml. Ultrasensitive detection of ~1 CFU/100 ml sample could be achieved by a prior pathogen enrichment step before the addition of the labeled antibody. The design of this diagnostic test is based on the common architecture of all bacteria, viruses and spores, namely the manifestation of repeat lipopolysaccharide epitopes on the surface. We have developed an easy-to-use two dot visual filter assay for translation into current water testing in public health laboratories to detect E. coli O157:H7. In a 5 h assay ~1 CFU and ~5 CFU of E. coli O157 could be detected in 100 ml of water or juice and lake samples respectively. This simple homosandwich enrichment strategy can also be used to detect low levels of other water-borne pathogens. © 2008 Elsevier B.V. All rights reserved.
1. Introduction E. coli O157:H7 causes a wide spectrum of human diseases, including bloody and non-bloody diarrhea, hemorrhagic colitis, occasional kidney failure, hemolytic uremic syndrome (HUS) and death at times (DeCludt et al., 2000; Shelton and Karns, 2001) due to ingestion of meat (Willshaw et al., 1994), water, and uncooked fruits and vegetables (Pebody et al., 1999). Outbreak of E. coli O157:H7 infections through drinking water was first reported in the USA in 1989 (Swerdlow et al., 1992). O157 contamination of drinking and recreational water has emerged as important cause of human disease (Ackman et al., 1997; Armstrong et al., 1996; ⁎ Corresponding author. Tel.: +780 492 9233; fax: +780 492 1217. E-mail address: [email protected] (M.R. Suresh). 0022-1759/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2008.04.011
Chalmers et al., 2000; Friedman et al., 1999). E. coli O157 present in the drinking water offered to livestock contributed to the prevalence of infection in animals leading to the contamination of meat products and the environment (Elder et al., 2000). An extreme example of the dangers this bacteria poses was seen in the Walkerton, Ontario and in the neighboring Maritimes, described as Canada's worst-ever outbreak of E. coli contamination (Hrudrey et al., 2003). E. coli O157: H7 is a top disease concern of the multi-billion dollar North American cattle industry. In Canada, Alberta has the greatest population of cattle and hence live stock generated manure contamination of water from surface drainage channels has precipitated one of the highest levels of gastroenteritis resulting from E. coli O157:H7 and Salmonella species (Khakhria et al., 1996). E. coli O157:H7 is the most common strain of Shiga toxin producing enterohemorrhagic
160
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
E. coli (STEC) in United States, Canada and United Kingdom (Kaper, 1998). Several recent outbreaks of gastrointestinal diseases caused by STEC has highlighted the threat these organisms pose to public health and are also considered as potential biowarfare agents (Trochimchuk et al., 2003). Many detection methods have been employed to rapidly detect low levels of pathogens in food, beverages and water. Current techniques include traditional enrichment and plating methods with selective media such as Sorbitol MacConkey agar and Rainbow agar (Manafi and Kremsmaier, 2001; Meng et al., 2001; Novicki et al., 2000). E. coli O157 is particularly difficult to confirm from enrichment cultures, even with highly selective media due to the problem of high background levels of competing microorganisms including other type of serotypes of E. coli. A variety of immunological methods have been developed for the detection and enumeration of E. coli O157 whole bacteria (Chapman et al., 1991, 1997; Park and Durst, 1999; Todd et al., 1988). The common denominator among all methods was the use of monoclonal or polyclonal anti-E. coli O157 antibodies to selectively capture, or capture and label by sandwich assay of E. coli O157 whole bacteria. Enrichments and plating often take 24–48 h to identify the organism. It is therefore important that rapid sensitive methods are developed to detect E. coli O157 during outbreaks, surveillance and quality control to prevent costly errors and fatalities due to delayed detection. Hence we have developed a simple 5 h two dot assay on nitrocellulose filter disks using a growth medium. This enrichment allows viable bacteria to grow for a short period prior to detection in a visual immunoassay (Fig. 1).
2. Materials and methods 2.1. Bacterial strains Strains used for the development of two dot filter assay were obtained from Dr. Roger Johnson's group (Health Canada Labs, Winnipeg, Canada) and from Dr. Newman & Edie Ashton (Provincial lab of Alberta). The strains of bacteria include E. coli O157:H7 (ATCC 43895), E. coli non-O157: non-H7 (O6: H34, O26:H11), E. coli non-O157: H7 (O18:H7, O91:H7), Salmonella urbana, S. typhimurium, Pseudomonas aeruginosa and 5 other laboratory strains of E. coli namely, JM87, Top 10 F, BL21DE3, JM109 and MKH14. All the strains used for the experiments were grown on Trypticase soy agar plates at 37 °C overnight followed by growth in Trypticase soy broth (TSB, pH 7.2–7.3) for 18 h with gentle agitation at 200 rpm. The classical dilution and colony counting methods were supplemented with direct estimation of fluorescent bacterial counts for accurate bacterial concentration measurements as described previously by our group (Guttikonda et al., 2004). 2.2. P124 MAb conjugation with colloidal gold A 10 ml aliquot of 20 nm colloidal gold (Sigma–Aldrich St. Louis, MO, USA) was adjusted to pH 9.0 with 0.1 M K2CO3. The P124 MAb at 10 µg/ml was adjusted to pH 9.0 with 100 nM K2CO3 and 10 ml of the antibody solution was added to 10 ml of colloidal gold. The solutions were shaken gently for 15 min at RT. Subsequently, 1 ml of 10% BSA pH 9.0 was
Fig. 1. Diagrammatic representation of our two dot filter assay for the detection of E. coli O157:H7.
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
added and kept at RT for 10 min and then centrifuged at 12,000 g for 45 min at 4 °C. The supernatant was discarded and the pellet was re-suspended in 2 ml of 0.1% BSA with 0.1% NaN3 in PBS (Hermanson, 1996). 2.3. Preparation of filter disks Sterile 0.45 µm nitrocellulose filter disks of 25 mm diameter (Fisher Scientific, Canada) were spotted with approximately 4 µg of P124 MAb and 4 µg of goat anti-mouse IgG spacially separated from each other (for two dot filter assay). The disks were left to air dry for 30 min at room temperature and blocked with 3% BSA for 2 h at room temperature. The disks were given a gentle wash with PBS buffer pH 7.2 and stored in the refrigerator for further use. 2.4. Two dot filter assay using colloidal gold antibodies This is an simple alternative non-enzymatic method of detection. The spiked water samples with 0–50 CFU/100 ml were filtered employing the Millipore manifold vacuum filtration unit using a nitrocellulose filter disk spotted with P124 MAb and goat anti-mouse IgG as separate spots. The filtered disks were incubated for 5 h or 24 h in TSB in a sterile plastic dish. Following a wash step, 250 µl of the P124-GC (colloidal gold) was added to the filter disks and gently shaken for 10 min at RT. The results were read visually for the appearance of one or two pink dots on the filter paper. 2.5. Detection of E. coli O157:H7 in apple juice and ground beef We also conducted limited comparisons assays with E. coli O157 spiked into apple juice or ground beef which are other sources that often gets contaminated with this pathogen. E. coli O157 was mixed with excess of the non-pathogenic and other pathogenic strains of E. coli in ratios of 1:5 CFU, 1:25 CFU, 1:50 CFU, 1:250 CFU and 1:500 CFU. Each of the above mixture of E. coli was spiked in 100 ml of apple juice and filtered on nitrocellulose disks spotted with P124 MAb and goat anti-mouse IgG. Following amplification of E. coli for 5 h and 24 h respectively, the bacteria were detected by colloidal gold conjugated to anti-E. coli O157 P124 MAb. Similarly, various concentrations of E. coli O157 were spiked into 25 g of ground beef and were enriched with 250 ml of TSB for 5 and 24 h respectively. Subsequently the meat was carefully separated from the broth and the entire enriched broth (~240 ml) was filtered on to the P124 and Goat anti-mouse IgG coated filter disks. The food enrichments were filtered with a steel mesh with pores to prevent clogging of the nitrocellulose test filters. The filter disks were incubated for 1 h by gentle shaking to allow the binding of E. coli to the antibody spot. Following a wash, P124-Colloidal gold was added and incubated for 10 min at RT. 2.6. Detection of E. coli O157 in lake water using two dot filter assay Sterile 0.45 µm nitrocellulose filter disk of 47 mm diameter (Millipore Canada Ltd) used in our local water testing labs were coated with approximately 10 µg (instead of 4 µg) of P124 MAb (anti-E. coli O157) and 10 µg of goat anti-mouse IgG in two separate spots. The disks were left to air dry for 30 min
161
at RT and blocked with 3% skim milk for 2 h at RT. The disks were given a gentle wash with PBS pH 7.2 and stored in the refrigerator for further use. A 10 ml, 25 ml and 100 ml each of a lake water sample collected from various regions of Northern Alberta were filtered onto the coated nitrocellulose disks using a filtration manifold and magnetic funnels obtained from the Provincial Lab of Alberta–Edmonton branch. The microbes on the nitrocellulose disks were amplified using 3 to 3.5 ml of the tryptic soy broth pH 7.2 for a minimum of 5 h and a maximum of 18 h (overnight). Following amplification, the disks were washed three times with PBS pH 7.2 and 5 µg of colloidal gold particles conjugated to P124 MAb per filter was added to the disks and given a gentle shake until the pink spot was developed. Cross-correlation of the two dot filter assay with other standard methods: The amplified TS broths were retained in the refrigerator for additional tests with established methods. Approximately 10–50 µl of TS broth was plated on the CT-SMAC plates and incubated at 37 °C overnight. Plates were examined for the presence of colorless colonies (non-sorbitol-fermenters). A representative number of bacteria from each colony type (approximately 10–15 colonies from each plate) were transferred to non-selective agar (such as TSB agar), spread for isolation to ensure purity and incubated at 37 °C overnight. The colonies were then screened by the oxidase test. (The oxidase test involved dipping a sterile swab into oxidase reagent (tetramethyl-pphenylenediamine dihydrochloride) and then touching the growth from each colony on the TSB agar plate. No more than 10–15 s were allowed for a reaction to develop. Purple color indicated oxidase positive and no color were scored oxidase negative. E. coli O157 are all oxidase negative, therefore any colonies which developed a purple coloration were discarded. All the reagents were tested with a positive and negative control before using them for screening the plated colonies. The oxidase reagent was tested against Pseudomonas aeruginosa as a positive control organism and E. coli 25922 as a negative control. Those colonies which yielded a negative oxidase test were screened by a slide agglutination test using the E. coli O157 antisera (BD Biosciences, USA). The E. coli O157 antiserum was tested against E. coli O157 and E. coli 25922 which are the positive and negative controls respectively. E. coli O157 produces a very fine agglutination reaction. 2.7. Cross-correlation of two dot filter assay using magnetic beads The amplified TS broth was further tested with the immunomagnetic bead separation procedure (Dynal ASA, Oslo, Norway). A 20 µL of anti-E coli O157 Dynal beads of 1 µm diameter were dispersed into an Eppendorf tube. Using a sterile pipette, 1 ml of amplified TS broth was added into each of the tubes containing anti-E. coli O157 beads. The samples were mixed thoroughly and incubated at RT for 10–30 min on a rocking platform. Following a very thorough wash (3 times) the beads were resuspended in 100 µl of PBS and inoculated to selective media CT-SMAC (Cefixime-Tellurite Sorbitol MacConkey Agar) and/or CHROMagar. (Cefiximine, Tellurite were from Dynal ASA, Oslo, Norway. SMAC was from Sigma– Aldrich St. Louis, MO, USA and CHROMagar was from Dynal ASA, Oslo, Norway). The colonies were subjected to the same degree of cross-checks as above. The positive and the negative
162
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
controls were 10–20 CFU of E. coli O157:H7 and E. coli 25922 in 1 ml of the TS broth respectively.
device for use in farms, wells, streams and lakes where portability is desirable when sample transportation to the laboratory is not practical.
3. Results We have recently developed a monoclonal and bispecific antibody against the E. coli O157:H7 by traditional hybridoma techniques (Guttikonda et al., 2007). The P124 monoclonal antibody was shown to bind to several different strains of E. coli O157 as well as purified LPS. Employing this newly developed anti-E. coli O157:H7 P124 MAb we explored two designs suitable for adaptation to routine water testing laboratories. Both these designs utilized the nitrocellulose filters currently used in water testing labs. 3.1. Two dot filter assay using colloidal gold In the two dot filter assay a single pink dot indicated the absence of the pathogen (binding of the P124-GC to the control goat anti-mouse IgG) and two pink dots indicated positive confirmation for E. coli O157:H7 (Figs. 2 and 3). The second pink dot is due to the sandwich formation wherein the pathogen was captured by P124 MAb and detected by P124-GC. Our results in two dot filter assay format with 5 h and 24 h amplification followed by the detection with MAb P124 conjugated to colloidal gold showed the detection of ~1 CFU of bacteria even at shorter periods of amplification step (Fig. 2). The background was clean and scoring was unambiguous. The overnight assay also gave similar results (data omitted). The assay using colloidal gold in both the 5 and 24 h assays had a detection limit of ~1 CFU/100 ml of water with clear visualization of end point from control. High sensitivity, shorter time and direct adaptation to water testing labs are the major advantages of this simple method. Additional advantages include the formulation of this assay into a unitized
3.2. Detection of E. coli O157:H7 strain in mixed cultures with excess of non-pathogenic and pathogenic strains of E. coli In nature the pathogenic strain could exist among several other microbes including the benign coliforms. Evaluation of mixed cultures of E. coli to investigate the detection of E. coli O157 was carried out as a part of our further prototype validation studies. The experiments were carried out in a two dot filter format. E. coli O157 that belong to H7 type was mixed with excess of 2 strains of E. coli non-O157:non-H7, 2 strains of E. coli non-O157: H7, Salmonella typhimurium species and Pseudomonas strain and 5 other laboratory strains of E. coli in 1:1 (E. coli O157:H7: Non-O157 strains), 1:5, 1:10, 1:50, 1:100 and 1:500 ratios. The above mixture of E. coli was spiked in 100 ml of water and filtered on nitrocellulose disks. Following amplification of E. coli for 5 h, the bacteria were detected by tracer P124-colloidal gold conjugate. The results show a significant detection of O157 even in a 1:500 fold excess of other strains (Fig. 3). 3.3. Detection of E. coli O157:H7 in apple juice and ground beef Since some of the E. coli O157 outbreaks were found associated in unpasteurized apple juice and ground beef (Willshaw et al., 1994), we conducted the filter dot assay by spiking the pathogen in commercial apple juice and ground beef. Filter dot assay was able to detect ~ 1 E. coli O157:H7 in the presence of 500 CFU of other bacteria in apple juice even after 5 h incubation. However, the detection limit of the E. coli O157 in 25 g of beef samples was at a higher level of 25 CFU
Fig. 2. E. coli O157 detection in water samples using our two dot filter assay following a 5 h enrichment. One to 50 CFU of E. coli O157 was filtered on P124 coated filter disk. Following the amplification of bacteria for 5 h, O157 was detected with tracer P124-colloidal gold.
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
163
Fig. 3. E. coli O157 detection simulating mixed microbes natural conditions: 1 CFU of E. coli O157 was mixed with various ratios of 2 strains of E. coli non-O157:nonH7, 2 E. coli non-O157: H7, 1 Salmonella species, one Pseudomonas strain and 5 other laboratory strains of E. coli strains of bacteria and filtered on P124 coated filter disk. Following the amplification of bacteria for 5 h, O157 was detected with tracer P124-colloidal gold.
(Fig. 4). We proposed testing in other media for increasing the detection sensitivity. 3.4. Detection of E. coli O157 in lake water using two dot filter assay Based on the promising results with the spiked water samples we evaluated several Northern Alberta lake water
samples for E. coli O157 in a small scale pilot experiment. The pH of the lake water samples ranged from ~ pH 6.5 to ~pH 10.5. None of the lake water samples tested in our limited sampling was positive for E. coli O157. However the pathogen spiked lake water samples gave the expected two dot confirmation. The two dot filter assay showed concordance with the CT-SMAC and Dynabeads cross-correlation results.
Fig. 4. E. coli O157 detection in 25 g of ground beef. E. coli O157 was added to the 25 g ground beef sample and amplified for 5 h at 37 °C. The meat was removed from the broth and filtered on P124 coated filter disk. Following further amplification of bacteria for 1 h, O157 was detected with tracer P124-colloidal gold.
164
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
3.5. Cross-correlation of the two dot filter assay with other standard methods and magnetic beads Cross-correlation using standard methods and magnetic beads showed that our two dot visual assay yielded consistent results and was a reliable method for E. coli O157 detection. 4. Discussion Infection with E. coli O157 can be asymptomatic but more commonly is characterized by watery diarrhoea or haemorrhagic colitis (HC). Approximately 5 to 30% of cases go on to develop Haemolytic Uremic Syndrome (HUS), characterized by haemolytic anaemia, low platelet counts and renal impairment. E. coli O157 is recognized as the most common cause of renal failure in children in the Europe and North America UK. The severity of the symptoms and high mortality rates (10–30%) make this pathogen an important public health concern (Bolton and Aird, 1998). The infectious dose of this organism is thought to be as few as 10 viable organisms and food contaminated with very low numbers of organisms can initiate disease. In this report we described a rapid, sensitive simple visual assay for routine detection of this pathogen in water, fruit juice and meat samples. This simple assay has a sensitivity below 10 CFU and can potentially detect 1 CFU/100 ml. Minimally trained personnel can perform the assay and score the endpoints for quality control and quality assurance. In all our experiments the E. coli O157:H7 sample preparations at various dilutions were estimated using the epifluorescence counting for cross-correlation of culture plate count estimates (Guttikonda et al., 2004). This is critical to ensure that the number of E. coli O157 that we were spiking into 100 ml of tap water, juice and meat broth was indeed comparable by the two methods. Correlation results indicated a non-significant variance in the bacterial counts between the plating and the epifluorescence counts. We acknowledge that it is difficult to ensure that one E. coli particle was indeed a single cell and not a clump of two or three cells and hence we have used the term ~1 cell to indicate very low levels of detection. In order to investigate the utility of the immunoconjugate of P124 MAb for the detection of E. coli O157 bacterium we developed a variant of the homosandwich assay to include an amplification step using growth medium to allow the viable bacteria to divide. In the non-amplified approach, the bacteria were directly filtered on to the nitrocellulose disks spotted with specific and non-specific antibodies. A comparison of amplified and non-amplified two dot filter assays was initially performed and the results showed the improved ultrasensitivity of the amplified two dot filter assay approach. At 5 h and longer amplification periods, E. coli detection limit was found to be ~1 bacterium. In the non-amplified assay the detection sensitivity of the assay was ~ 2500 CFU/100 ml (data not shown). Nitrocellulose filter disks are routinely used in the water testing laboratories for detection of total coliforms. In our provincial water testing lab ~100,000 water samples are processed each year from N. Alberta alone. Extrapolating this to all states and provinces in Northern America, nearly 10 million tested are likely done on an annual basis. Hence, a non-enzymatic end point was designed to eliminate the additional enzymatic step. Ease and simplicity in detecting O157 in water was achieved with the colloidal gold which does not require any enzymatic
reactions and extra steps of incubations. The two dot filter assay is a one-step detection method following bacterial amplification and requires only 5 h incubation for the E. coli O157:H7 to be detected. Thus the assay can be completed in few hours from sample collection if a unitized device is developed based on the current format. Based on our studies the two dot filter assay has sensitivity, simplicity and rapidity. When detecting human pathogens, faster detection is very much an asset for determining the immediate course of treatment. Most of the present commercial detection methods take 16–24 h for enrichment to detect the E. coli O157 bacteria, whereas our amplified two dot filter disk assay takes only 5 h with ultrasensitive detection of ~1 CFU of E. coli O157 in 100 ml of water sample. If higher levels are suspected, the assay could potentially be completed in less than 5 h. In our experiments with mixed cultures of other E. coli and bacteria, the two dot filter disk assay format was shown to be sensitive even with 500 fold higher concentrations of other E. coli and bacterial strains. The reliability of the testing kits depends on the complexity of procedure, sensitivity, specificity and duration of assay. Previous methods of isolation of E. coli O157 from food was based on enrichment culture, concentration using the immunomagnetic separation (IMS) technique (Wright et al., 1994) and sub-culture on sorbitol MacConkey agar containing cefixime and potassium tellurite. The Dynal kit requires their specific instruments for performing the assay. In addition the protocol for immunomagnetic separation requires an 18–24 h amplification step preceeded by the separation of E. coli O157, and visualization on specialized plates (CT-SMAC or CHROMagar). Our two dot visual assay is an ultrasensitive detection method that could be adopted for routine water testing of E. coli O157:H7. Samples with algal interferences gave us inconclusive results when water samples are processed without filtering. However filtering out algae gave consistent results and we plan to investigate this further. However most of the other lake water samples gave us two dots when negative lake samples were spiked with known number of E. coli O157:H7. Our results show that following enrichment, two dot filter assay can be used to detect the E. coli O157:H7 pathogen with good sensitivity. Our two dot filter assay could detect 25 CFU O157 in 25 g of ground beef and can be improved by further optimization. The detection limit in beef was less sensitive than in water. We plan further studies to determine if the bacteria adhere to the sedimented bulk of the meat sample and reduce detection sensitivity. The key feature of our simple visual system is that it is based on scoring simple end points as one dot (−ve) or two dot (+ve) (Fig. 1). Like ELISA, the visual system can process a large number of samples per day. Our two dot filter assay system is simple to perform and interpret, inexpensive and relatively rapid (~5 h). It requires neither costly special equipment nor highly qualified operators. It is also highly reproducible. Compared to other commercial visual detection formats, the manufacturing cost per test could be low and practical for routine use. We propose that this simple format is an ideal test for implementation as routine test for water samples in both developed and developing countries to manage E. coli O157. It is also possible that the above filter dot concept could be extended to other pathogens if convenient amplification broth is available. Alternatively a macro array of different
S. Kamma et al. / Journal of Immunological Methods 336 (2008) 159–165
antibody spots could be constructed for multiple pathogen testing in water and food samples. Acknowledgements MRS would like to thank NCE-CBDN for the grant support and the CIHR-Industry chair award for salary support. Thanks are due to Dr. Roger Johnson and Erika Lingohr from Health Canada Lab for the E. coli and Salmonella strains. References Ackman, D.S., Marks, S., Mack, P., Caldwell, P., Root, T., Birkhead, G., 1997. Swimming associated hemorrhagic colitis due to Escherichia coli O157: H7 infection: evidence of prolonged contamination of a fresh water lake. Epidemiol. Infect. 119, 1. Armstrong, G.L., Hollingsworth, J., Morris Jr, J.G., 1996. Emerging foodborne pathogens: Escherichia coli O157:H7 as a model of entry of a new pathogen into the food supply of the developed world. Epidemiol. Rev. 18, 29. Bolton, F.J.H., Aird, H., 1998. Verocytotoxin producing Escherichia coli O157: public health and microbiological significance. Br. J. Biomed. Sci. 55, 127. Chalmers, R.M., Aird, H., Bolton, F.J., 2000. Waterborne Escherichia coli O157. J. Appl. Microbiol. 88, 124S (Suppl). Chapman, P.A., Siddons, C.A., Zadik, P.M., Jewes, L., 1991. An improved selective medium for the isolation of Escherichia coli O157. J. Med. Microbiol. 35, 107. Chapman, P.A., Cerdan, A.T., Malo, P.M., Siddons, C.A., Harkins, M., 1997. Use of commercial enzyme immunoassays and immunomagnetic separation system for detection of E. coli O157:H7 in bovine fecal sample. Appl. Environ. Microbiol. 63, 2549. DeCludt, B., Bouvet, P., Mariani-Kurkdjian, P., Grimont, F., Grimont, P.A., Hubert, B., Loirat, C., 2000. Haemolytic uraemic syndrome and Shiga toxin producing E. coli infection in children in France. Epidemiol. Infect. 124, 215. Elder, R.O., Keen, J.E., Siragusa, E.R., Barkocy-Gallagher, G.A., Koohmaraie, M., Laegreid, W.W., 2000. Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing. Proc. Natl. Acad. Sci. U. S. A. 97, 2999. Friedman, M.S., Roels, T., Koehler, J.E., Feldman, L., Bibb, W.F., Blake, P., 1999. Escherichia coli O157:H7 outbreak associated with an improperly chlorinated swimming pool. Clin. Infect. Dis. 29, 298. Guttikonda, S., Wang, W.W., Suresh, M.R., 2004. Molecular zipper assays: a simple homosandwich with the sensitivity of pcr. J. Pharm. Pharmaceut. Sci. 7 (S1), 7. Guttikonda, S., Tang, X.L., Yang, M.B., Armstrong, G.D., Suresh, M.R., 2007. Monospecific and bispecific antibodies against E. coli O157 for diagnostics. J. Immunol. Methods 327 (1–2), 1.
165
Hermanson, T.J., 1996. Preparation of colloidal gold labeled proteins. Bioconjugate Techniques. Academic press, San Diego, California, USA. Hrudrey, S.E., Payment, P., Hych, P.M., Gillham, R.W., Hrudey, E.J., 2003. A fatal waterborne disease epidemic in Walkerton, Ontario: comparison with other waterborne outbreaks in the developed world. Water Sci. Technol. 47, 7. Kaper, J.B., 1998. Enterohemorrhagic E. coli. Curr. Opin. Microbiol. 1, 103. Khakhria, R., Woodward, D., Johnson, W., 1996. Salmonella, Shigellae, pathogenic E. coli, Campylobacter and Aeromonas identified in Canada: annual summary 1994. National Laboratory for Enteric Pathogens. Manafi, M., Kremsmaier, B., 2001. Comparative evaluation of different chromogenic/flurogenic media for detecting Escherichia coli O157:H7 in food. Int. J. Food. Microbiol. 71, 257. Meng, J., Doyle, M.P., Zhao, T., Zhao, S., 2001. Enterohemorrhagic Escherichia coli. In: Doyle, M.P. (Ed.), Food Microbiology:Fundamentals. ASM Press, Washington, DC, p. 193. Novicki, T.J., Daily, J.A., Mottice, S.L., Carroll, K.C., 2000. Comparison of MacConkey Agar and a two step method which utilizes ELISA toxin testing and a chromogenic agar to detect and isolate E.coli. J. Clin. Microbiol. 38, 547. Park, S., Durst, R.A., 1999. Immunoliposome sandwich assay for the detection of Escherichia coli O157:H7. Anal. Biochem. 280, 151. Pebody, R.G., Furtado, C., Rojas, A., McCarthy, N., Nylen, G., Ruutu, P., Leino, P., Chalmers, R., De Jong, B., Donnelly, M., Fisher, I., Gilham, C., Graverson, L., Cheasty, T., Willshaw, G., Navarro, M., Salmon, R., Leinikki, P., Wall, P., Bartlett, C., 1999. An international outbreak of Vero cytotoxin producing E.coli O157 infection amongst tourist: a challenge for the European infectious disease surveillance network. Epidemiol. Infect. 123, 217. Shelton, D.R., Karns, J.S., 2001. Qualitative detection of E.coli O157 in surface waters by using immunomagnetic electrochemiluminescence. Appl. Environ. Microbiol. 67 (7), 2908. Swerdlow, D.L., Woodruff, B.A., Brady, R.C., Griffin, P.M., Tippen, S., Donnell Jr, H.D., Geldreich, E., Payne, B.J., Meyer Jr., A., Wells, J.G., Greene, K.D., Bright, M., Bean, N.H., Blake, P.A., 1992. A waterborne outbreak in Missouri of Escherichia coli O157:H7 associated with bloody diarrhea and death. Ann. Intern. Med. 117, 812. Todd, E.C.D., Szabo, R.A., Peterkin, P., Sharpe, A.N., Parrington, L., Bundle, D., Gidney, M.A.J., Perry, M.B., 1988. Rapid hydrophobic grid membrane filterenzyme-labeled antibody procedure for identification and enumeration of Escherichia coli 0157 in foods. Appl. Environ. Microbiol. 54, 2536. Trochimchuk, T., Fotheringham, J., Topp, E., Schraft, H., Leung, K.T., 2003. A comparison of DNA extraction and purification methods to detect Escherichia coli O157:H7 in cattle manure. J. Microbiol. Methods 54, 165. Willshaw, G.A., Thirlwel, J., Jones, A.P., Rowe, B., Hickey, M., 1994. Vero cytotoxin-producing Escherichia coli O157 in beefburgers linked to an outbreak of diarrhoea, haemorrhagic colitis and haemolytic uraemic syndrome in Britain. Lett. Appl. Microbiol. 19 (5), 304. Wright, D.J., Chapman, P.A., Siddons, C.A., 1994. Immunomagnetic separation as a sensitive method for isolating Escherichia coli from food samples. Epidemiol. Infect. 113, 31.