Testing a dual-fluorescence assay to monitor the viability of filamentous cyanobacteria

Filamentous cyanobacteria are currently being engineered to produce long-chain organic compounds, including 3rd generation biofuels. Because of their filamentous morphology, standard methods to quantify viability (e.g., plate counts) are not possible. This study investigated a dual-fluorescence assay based upon the LIVE/DEAD® BacLight™ Bacterial Viability Kit to quantify the percent viability of filamentous cyanobacteria using a microplate reader in a high throughput 96-well plate format. The manufacturer's protocol calls for an optical density normalization step to equalize the numbers of viable and non-viable cells used to generate calibration curves. Unfortunately, the isopropanol treatment used to generate non-viable cells released a blue pigment that altered absorbance readings of the non-viable cell solution, resulting in an inaccurate calibration curve. Thus we omitted this optical density normalization step, and carefully divided cell cultures into two equal fractions before the isopropanol treatment. While the resulting calibration curves had relatively high correlation coefficients, their use in various experiments resulted in viability estimates ranging from below 0% to far above 100%. We traced this to the apparent inaccuracy of the propidium iodide (PI) dye that was to stain only non-viable cells. Through further analysis via microplate reader, as well as confocal and wide-field epi-fluorescence microscopy, we observed non-specific binding of PI in viable filamentous cyanobacteria. While PI will not work for filamentous cyanobacteria, it is possible that other fluorochrome dyes could be used to selectively stain non-viable cells. This will be essential in future studies for screening mutants and optimizing photobioreactor system performance for filamentous cyanobacteria.