Improved membrane filtration media for enumeration of total coliforms and Escherichia coli from sewage and surface waters

Abstract

Two media were developed that allowed both a total coliform count and an Escherichia coli count to be determined on the same medium after 24 h of incubation at 35°C. The new media were tested along with two standard media on 10 surface water and 7 sewage samples. The experimental media yielded equivalent or higher counts relative to the standard media and recovered more specifically the desired indicator groups as determined by colony identification. Current membrane filtration methods have been criticized for their many shortcomings (for a review, see reference 15). Several investigators have suggested that Escherichia coli be used as an indicator because it is specific for fecal pollution and provides a more accurate indicator-to-pathogen ratio than total or fecal coliform enumeration methods (4, 7, 9). Increased attention is being given to the recovery of injured indicator bacteria (3, 5, 6, 19, 20, 22, 23). One of the major factors causing reduced recoveries of injured bacteria is the selective agent(s) that is used. Monensin is an ionophore that selectively inhibits all gram-positive bacteria while it allows growth of injured gram-negative bacteria (25). Tergitol 7, a surfactant, has a similar effect (17, 19). Recently, a very sensitive and specific method for the detection of P-glucuronidase production by E. coli was developed (10, 15, 25). Over 95% of E. coli hydrolyze a nonfluorescent substrate, 4-methylumbelliferyl-p-D-glucuronide (MUG), to produce 4-methylumbelliferone, which fluoresces under long-wave UV light. Approximately 50% of Shigella spp., some Salmonella spp., and a few strains of Yersinia enterocolitica are the only other members of the families Enterobacteriaceae and Vibrionaceae that produce this enzyme (12, 14, 18, 21, 25, 29). The purpose of this study was to develop an improved membrane filtration method by using media that contain monensin, Tergitol 7, and MUG. The goal was the enumeration of total coliforms and E. coli on a single medium incubated at 35°C with improved recovery of injured cells and improved selective specificity when compared with standard membrane filtration media and methods. MATERIALS AND METHODS Media. MUG was obtained from Hach Co. (Loveland, Colo.). Monensin (90 to 95% pure) was obtained from Sigma. All media were prepared with 1.5% agar, sterilized, tempered to 47°C, and poured in approximately 3-ml volumes into petri plates (diameter, 47 mm). Peptone-Tergitol-glucuronide (PTG) agar was prepared as described previously (5). Peptone glucuronide (PG) agar was prepared exactly as PTG agar, except that the Tergitol 7 (J. T. Baker Chemical Co., Phillipsburg, N.J.) was omitted. PG agar was poured into standard petri plates (diameter, 100 mm) for the isolation and purification of colonies. Enriched

Poor specificity of m-Endo and m-FC culture media for the enumeration of coliform bacteria in sea water

“…These percentages are similar to those obtained by Avila et al (1989) from sea water samples in the Spanish coastal area. It is also interesting to point out the higher incidence, compared with other studies (Freier and Hartman 1987;Avila et a!. 1989) of Serratia isolates.…”

“…(1989) (14.1%). We did not detect the presence of Aeromonas hydrophila, which is usually identified from typical colonies up to 50% on the same medium (Avila et al 1989;Freier and Hartman 1987), because in our case, the colonies for identification were selected according to their negative response to the cytochrome oxidase test. It is evident that the confirmatory tests are not useful for the elimination of certain frequent non-coliform genera, such as Serratia, Hafnia, Kluyvera and Shigella, but they help to avoid other common non-coliform bacteria, such as those belonging to the genera Aeromonas and Edwardsiella.…”

“…Other authors consider that the determination of E. colt directly provides a more accurate indication of faecal pollution (Dufour and Cabelli 1975;Capleans and Kanarek 1984;Freier and Hartman 1987;Hernandez et al 1991) and in this sense new techniques have been described for the rapid and easy detection of E. coli in recreational waters, based on the p-glucuronidase activity, e.g. MUG, Autoanalysis Colilert, microtitration plates, etc.…”

“…MUG, Autoanalysis Colilert, microtitration plates, etc. (Freier and Hartman 1987;Mates andSchaffer 1988, 1989;Covert et al 1989;Gauthier et al 1991;Hernandez et al 1991). However, some other, mainly American authors, consider that E. coli is a poor indicator in sea water (Cabelli et al 1982;Dufour 1984).…”

“…Therefore, the biotypes of E. coli, unable to ferment Eactose, including those responsible for diarrhoea1 illnesses, will grow with nontypical colonial morphology giving false-negative results (Sakazaki et al 1967;Rychert and Stephenson 1986). Recently 8-glucuronidase activity has been recommended for the specific enumeration of E. coli, since most other coliform bacteria lack this enzymatic activity (Freier and Hartman 1987;Mates andSchaffer 1988, 1989;Gauthier et af. 1991 ;Hernindez et al 1991).…”