Detection of Escherichia coli colonies on confluent plates of chromogenic media used in membrane filtration Andrée F. Maheux a,b, ⁎, Vanessa Dion-Dupont a,c , Marc-Antoine Bisson a,c , Sébastien Bouchard a,c , Manuel J. Rodriguez a,b a Chaire de recherche en eau potable de l'Université Laval, Québec City, Quebec, Canada b École supérieure d'aménagement du territoire et de développement régional, Université Laval, Québec City, Quebec, Canada c Centre de recherche en aménagement et développement de l'Université Laval, Québec City, Quebec, Canada abstract article info Article history: Received 4 December 2013 Received in revised form 11 December 2013 Accepted 11 December 2013 Available online 21 December 2013 Keywords: Chromocult coliform ES agar DC with BCIG agar E. coli MI agar Total coliforms MI agar (MI), Chromocult® Coliform agar ES (Chromocult), and DC with BCIG agar (DC) are chromogenic membrane filtration culture-based methods used to assess microbiological water quality. In this study, their ability to detect Escherichia coli colonies on confluent growth plates was determined by testing water samples containing increasing concentrations of a non-E. coli growing bacterium, Citrobacter youngae. Then, their ability to inhibit the growth of non-coliform bacteria was determined by simultaneously testing 603 well water samples. Results were compared with those obtained with mFC and Colilert® methods. Results showed that the E. coli count was difficult to determine on mFC, Chromocult and DC when non-E. coli colonies reached levels of 10 4 colony forming units (CFU)/100 mL. However, the E. coli count did not interfere with Colilert until non-E. coli colonies reached concentrations of 10 7 CFU/100 mL. No inhibition was observed with MI as E. coli colonies could be easily detected in the presence of at least 10 7 CFU/100 mL of C. youngae. Using well water samples, confluent growth plates were observed for 144, 177, and 185 of the 603 well water samples tested with the MI, Chromocult and DC methods, respectively. Among these confluent growth plates, E. coli colonies were not detected for 10, 20, and 31 water samples. However, they were detected by the mFC and/or Colilert methods. Thus, among the three methods tested, the MI method presented the lowest grow rate of atypic colonies and was the only one that presents no interference in the E. coli count. © 2013 Elsevier B.V. All rights reserved. 1. Introduction In the Province of Quebec (Canada), the Programme d'accréditation des laboratoires d'analyse (“Accreditation programme of analytic labora- tories”; PALA) is administered by the Centre d'expertise en analyse environnementale du Québec (CEAEQ) which accredits private, municipal and institutional laboratories. The Regulation respecting the quality of drinking water (RRQDW) states that when the number of colonies ex- ceeds 200 on a filtration membrane, the growth of target bacteria may be inhibited or their typical appearance may be modified causing a false-negative result (Government of Quebec, 2013). This regulation was established following a series of studies showing that the detection of faecal coliform on plates may be negatively influenced by the presence of more than 1000 colony forming units (CFU) of heterotrophic bacteria per 100 mL (Geldreich et al., 1972). Indeed, the presence of high numbers of background heterotrophic bacteria has been shown to decrease coliform recovery by membrane filtration techniques using mEndo LES agar (Clark, 1980; Burlingame et al., 1984). Furthermore, excessive crowding of colonies on mEndo LES plates has also been asso- ciated with a reduction of coliform colonies producing metallic sheen (Hsu and Williams, 1982; Burlingame et al., 1984; Prescott et al., 1995; Clesceri et al., 1998). Consequently, when more than 200 atypical colo- nies are present on a plate during the analysis of filter results of mem- brane filtration of a water sample, appropriate corrective measures must be taken immediately and another sampling carried out (APHA et al., 2005; United States Environmental Protection Agency, 2013; Gov- ernment of Quebec, 2013). Since 2001, thermotolerant coliform detection by the mFC membrane filtration method has been recommended by the CEAEQ for the assessment of the microbiological quality of drinking water (Centre d'expertise en analyse environnementale du Québec, 2011). Indeed, in the presence of these microorganisms, drinking water is potentially contaminated with pathogens and improper for human consumption. The high specificity of the mFC agar method, attributable primarily to the high incubation temperature (44.5 °C), limits the growth of non-targeted microorganisms. Thereafter, the concentration of atypical colonies on mFC agar rarely exceeds 200 CFU/100 mL of drinking water. However, the mFC method is not totally specific and microorganisms of non-faecal origin can be detected with this method as false-positive results. Consequently, in the fall of 2010, the CEAEQ Journal of Microbiological Methods 97 (2014) 51–55 ⁎ Corresponding author at: ÉSAD, 1616 Pavillon Savard, Université Laval, Québec City, Quebec G1K 7P4, Canada. Tel.: +1 418 656 2131x8933; fax: +1 418 656 2018. E-mail address: andree.maheux.3@ulaval.ca (A.F. Maheux). 0167-7012/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mimet.2013.12.008 Contents lists available at ScienceDirect Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth