Evaluation of Colilert-18 for the detection of coliforms and Escherichia coli in tropical fresh water

Department of Microbiology, Soochow University, Shih Lin, Taipei, Taiwan.
Letters in Applied Microbiology (Impact Factor: 1.66). 03/2006; 42(2):115-20. DOI: 10.1111/j.1472-765X.2005.01814.x
Source: PubMed


To evaluate the suitability of Colilert-18 in detecting Escherichia coli and total coliforms in tropical freshwater samples.
Target organisms were isolated from yellow-fluorescent and yellow wells of Colilert-18/Quanti-Tray using m-TEC agar and m-ENDO LES agar respectively. All the selected isolates were first identified based on their fatty acid methyl ester profile. Isolates showing contradictory results to that of the Colilert-18 procedure were re-identified using API 20E strips. A total of 357 isolates, 177 from yellow-fluorescent wells and 180 from yellow wells, were identified.
The false-positive and -negative rates for E. coli detection using Colilert-18 were 36.4% and 11%, respectively, while for coliform detection the false-positive rate was 10.3%.
The high false-positive rate of Colilert-18, tempers its value for E. coli detection when used for tropical freshwater samples.

Full-text preview

Available from:
  • Source
    • "E. coli was isolated from the Idexx Quanti-Trays as described by Chao (2006). A 10 μl aliquot of liquid culture from each sampled fluorescent well was streaked onto modified mTEC agar (Becton Dickinson, Sparks, "
    [Show abstract] [Hide abstract]
    ABSTRACT: Lyngbya wollei is a filamentous cyanobacterium which forms large nuisance mats and has infested eastern and southeastern U.S. Lakes and reservoirs for over 100 years. Lyngbya was recently identified in the Great Lakes system in the St. Lawrence River, and Western Lake Erie. Here we report on large deposits of L. wollei washing onshore at a popular recreational beach in Lake Saint Clair, part of the Great Lakes system. The amount of L. wollei deposited on shore was quantified and evaluated for the presence of fecal indicator bacteria (FIB). High concentrations of Escherichia coli, enterococci and Clostridium perfringens were found in the L. wollei in nearshore waters. The densities of E. coli (MPN), enterococci (MPN) and C. perfringens (CFU) attached to L. wollei averaged 3.5, 3.2 and 3.2 log/g, respectively. In contrast, nearshore waters contained nearly 10 times less FIB, averaging 2.6, 2.4 and 2.6 log/100 ml of E. coli (MPN), enterococci (MPN) and C. perfringens (CFU), respectively. DNA fingerprint analysis was used to examine the population structure of E. coli isolates obtained from L. wollei mats. The L. wollei-borne E. coli strains were genetically diverse, suggesting a causal relationship between E. coli and L. wollei. Results from this study indicate that in addition to the macroalga such as Cladophora, cyanobacteria like L. wollei also harbor FIB, potentially impacting water quality and human health in the Great Lakes.
    Full-text · Article · Dec 2013 · Journal of Great Lakes Research
  • Source
    • "For the purposes of this study, and in accordance with ISO methods comparison protocols (ISO 17994), the test methods were considered to be equivalent to the reference method if the mean difference (MD) in recovery of target organisms was not significantly different from zero. Several equivalence studies have been published based on the comparison protocol described in ISO 17994 (2004) (Niemelä et al. 2003; Bernasconi et al. 2006; Bonadonna et al. 2006; Chao 2006). In recent years Colilert-18/Quantitray has been widely used by many laboratories worldwide (Fricker et al. 1997; Bonadonna et al. 2006). "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study ten laboratories in Greece compared the performance of reference method TTC Tergitol 7 Agar (with the additional test of beta-glucuronidase production) with five alternative methods, to detect E. coli in water, in line with European Water Directive recommendations. The samples were prepared by spiking drinking water with sewage effluent following a standard protocol. Chlorinated and non-chlorinated samples were used. The statistical analysis was based on the mean relative difference of confirmed counts and was performed in line with ISO 17994. The results showed that in total, three of the alternative methods (Chromocult Coliform agar, Membrane Lauryl Sulfate agar and Trypton Bilex-glucuronidase medium) were not different from TTC Tergitol 7 agar (TTC Tergitol 7 agar vs Chromocult Coliform agar, 294 samples, mean RD% 5.55; vs MLSA, 302 samples, mean RD% 1; vs TBX, 297 samples, mean RD% -2.78). The other two alternative methods (Membrane Faecal coliform medium and Colilert 18/ Quantitray) gave significantly higher counts than TTC Tergitol 7 agar (TTC Tergitol 7 agar vs MFc, 303 samples, mean RD% 8.81; vs Colilert-18/Quantitray, 76 samples, mean RD% 18.91). In other words, the alternative methods generated performance that was as reliable as, or even better than, the reference method. This study will help laboratories in Greece overcome culture and counting problems deriving from the EU reference method for E. coli counts in water samples.
    Full-text · Article · Jan 2010 · Water Science & Technology
  • Source
    • "However, other studies have disputed the degree of accuracy for E. coli detection owing to the risk of obtaining false-positive (or negative ) results, but the accuracy varies extensively according to the types and locations of water tested (Pisciotta et al. 2002; Chao et al. 2004). Notably, Chao (2006) reported a high false-positive rate (36AE4%) for E. coli detection using Colil- ert-18 with tropical freshwater. Similarly, a large percentage (27AE3%) of the isolates obtained from the yellow and fluorescent wells, inoculated with marine waters from Florida, were not E. coli (Pisciotta et al. 2002). "
    [Show abstract] [Hide abstract]
    ABSTRACT: To assess levels of faecal contamination in the Qu'Appelle River (Saskatchewan, Canada) and its suitability for irrigation, by using the Colilert-18/Quanti-Tray technology. Various sites located along the Qu'Appelle River were sampled weekly from May to August 2005-2007. A total of 594 freshwater samples were collected and analysed for enumeration of Escherichia coli using the Colilert-18. The false-positive rate for E. coli detection using Colilert-18 was at most 1.5%. Throughout the irrigation period (June to August), up to 85% of the water samples collected from one of the irrigation water-pumping sites exceeded the recommended limit of 100 CFU per 100 ml. Spikes in E. coli counts were generally concomitant with the sudden rise in river flows. A sub-sample of confirmed E. coli isolates were typed by randomly amplified polymorphic DNA (RAPD). RAPD analysis revealed a high degree of genetic diversity among E. coli isolates. A significant association between RAPD patterns and the month of E. coli isolation was demonstrated. Colilert-18 provides an effective means for assessing microbial quality of irrigation water. Qu'Appelle River is subject to variability of faecal contamination during irrigation times and monitoring throughout irrigation season is important for ensuring safe production practices.
    Full-text · Article · Dec 2008 · Journal of Applied Microbiology
Show more