Development and validation of an alternative parameter for quantification of signals emitted by fluorescently labelled bacteria in microscopic images

Abstract

In this study, an alternative parameter for quantifying the signals of fluorescently labelled bacteria (e.g. propidium iodide, Cyanine 3, etc.) in microscopic images was investigated. Three common parameters (mean grey value (MGV), mean grey value which is corrected for the background (MGVcwB) and the signal to background ratio (SBR) per bacterial cell) are used as reference parameters. As an alternative, the coefficient of variation (CV) is defined as the ratio of the logarithm of the standard deviation and the logarithm of the mean grey value of a bacterial cell in a microscopic image. The actual fluorescence value was safeguarded by measuring commercially available fluorescence latex microspheres at regular time intervals within our study. The precision and the correlation of the respective values of MGV, MGVcwB, SBR and CV taken from identical images were measured and subsequently normalized in order to enhance the inter-parameter comparability. The average precision of CV was the highest (89% ± 14) with decreasing numbers for MGVcwB, SBR, and MGV (78% ± 25, 71% ± 32, and, 52% ± 22, respectively). Changes in operational parameters, e.g., microscope settings, protocol steps, etc., yielded good results for the CV but less precise results for MGV, MGVcwB, and SBR in the analyses of identical images. In conclusion, using the alternative parameter CV, changes in the composition of microbial ecosystems may thus be investigated at the highest precision level.