Quantification methods of determining brewer’s and pharmaceutical yeast cell viability: accuracy and impact of nanoparticles

Abstract

AbstractFor industrial processes, a fast, precise, and reliable method of determining the physiological state of yeast cells, especially viability, is essential. However, an increasing number of processes use magnetic nanoparticles (MNPs) for yeast cell manipulation, but their impact on yeast cell viability and the assay itself is unclear. This study tested the viability of Saccharomyces pastorianus ssp. carlsbergensis and Pichia pastoris by comparing traditional colourimetric, high-throughput, and growth assays with membrane fluidity. Results showed that methylene blue staining is only reliable for S. pastorianus cells with good viability, being erroneous in low viability (R2 = 0.945; $$\widehat{\sigma }$$ σ ^ = 5.78%). In comparison, the fluorescence microscopy–based assay of S. pastorianus demonstrated a coefficient of determination of R2 = 0.991 at $$\alpha =0$$ α = 0 ($$\widehat{\sigma }$$ σ ^ = 2.50%) and flow cytometric viability determination using carboxyfluorescein diacetate (CFDA), enabling high-throughput analysis of representative cell numbers; R2 = 0.972 ($$\alpha =0$$ α = 0 ; $$\widehat{\sigma }$$ σ ^ = 3.89%). Membrane fluidity resulted in a non-linear relationship with the viability of the yeast cells ($$\alpha \ne 0$$ α ≠ 0 ). We also determined similar results using P. pastoris yeast. In addition, we demonstrated that MNPs affected methylene blue staining by overestimating viability. The random forest model has been shown to be a precise method for classifying nanoparticles and yeast cells and viability differentiation in flow cytometry by using CFDA. Moreover, CFDA and membrane fluidity revealed precise results for both yeasts, also in the presence of nanoparticles, enabling fast and reliable determination of viability in many experiments using MNPs for yeast cell manipulation or separation.