Practical coliforms and Escherichia coli detection and enumeration for industrial food samples using low-cost digital microscopy

Abstract

This research explored several alternative coliforms and E. coli detection strategies proposed for industrial application especially low-resource settings and less advanced food manufacturers. The colony count using 2 industrial favourites (i.e., PetrifilmTM by 3 M and regular pour plate techniques) were contrasted to 4 alternative lowcost strategies. Two modified conventional protocol (i.e., pour and spread plate techniques) in standard mini Petri dishes and two drop plate techniques in microtiter plate formats (i.e., 24 - and 96-well plate) were applied to count industrial frozen food samples. The colony detection in all treatments was visually facilitated by low-cost digital microscopy technique comparing colony count, the detection time and the colony area in pixels. All experiments except for the PetrifilmTM E. coli/Coliform (EC) Plate utilized Chromocult® Coliform Agar (CCA). The inoculum sizes were varied depending on the cell count technique used; 10 μl and 5 μl for the 24- and 96-well microtiter plates, 50 μl for the mini-plate (both pour and spread plate techniques), and 1000 μl for the full-size Petri dishes and PetrifilmTME. coli/Coliform (EC) Plate. The incubation temperature was fixed at 35±2°C. The number of colonies from the conventional pour plate technique was plotted against those of the other techniques. In all treatments, the relationship plots showed highly linearity from the lower detection limit (100 CFU/ml depending on technique used) to the upper detection limit (10,000 CFU/ml). The slopes of all regression lines were close to unity showing very high correlation of the values of colony counts from different techniques. All techniques were applied to evaluate actual swap samples from the production line and returned highly consistent colony count numbers and good separation of E. coli from coliforms. The colony counts from the ready-to-eat product samples showed similar and comparable results to the two routine factory techniques. The results indicated that all six colony count methods were simply interchangeable to perform colony enumeration in the low-resource industrial setting.