New fluorescein precursors for live bacteria detection.

Abstract

Swiftness, reliability, and sensitivity of live bacteria detection in drinking water are key issues for human safety. The most widespread used indicator of live bacteria is a caged form of carboxyfluorescein in which 3' and 6' hydroxyl groups are masked as acetate esters (CFDA). This derivatization altogether abolishes fluorescein fluorescence and renders the molecule prone to passive diffusion through bacterial membranes. Once in the cytoplasm, acetate groups from CFDA are removed by bacterial hydrolases and fluorescence develops, rendering live but not dead cells detectable. Yet the reagent, carboxyfluorescein diacetate, still possesses a free carboxyl group whose ionization constant is such that the majority of the probe is charged at physiological pH. This unfavors probe permeation through membranes. Here, we prepare several chemical modifications of the carboxyl moiety of CFDA, in order to neutralize its charge and improve its passive diffusion through membranes. We show that the ethylamido derivative of the 5-carboxyl group from 5-carboxy-fluorescein diacetate or from Oregon green diacetate or from Oregon green diacetoxymethylester are stable molecules in biological media, penetrate into bacterial cells and are metabolized into fluorescent species. Only live bacteria are revealed since bleached samples are not labeled. Other derivatives with modification of the 5-carboxyl group with an ester group or with a thiourea-based moiety were almost inefficient probes. The most interesting probe, triembarine (5-ethylaminocarboxy-oregon green, 3',6'diacetoxymethyl ester) leads to 6-10 times more sensitive detection of bacteria as compared to CFDA. Addition of contrast agents (trypan blue or brilliant blue R) improve the signal-to-noise ratio by quenching extracellular fluorescence while bromophenol blue quenches both intracellular and extracellular fluorescence, allowing standardization of detections.