Time-lapse imaging of live cells using multiple fluorescent reporters is an essential tool to study molecular processes in single cells. However, exposure to even moderate doses of visible excitation light can disturb cellular physiology and alter the quantitative behavior of the cells under study. Here, we set out to develop guidelines to avoid the confounding effects of excitation light in multi-color long-term imaging. We use widefield fluorescence microscopy to measure the effect of the administered excitation light on growth rate (here called photomorbidity) in yeast. We find that photomorbidity is determined by the cumulative light dose at each wavelength, but independent of the way excitation light is applied. Importantly, photomorbidity possesses a threshold light dose below which no effect is detectable (NOEL). We found, that the suitability of fluorescent proteins for live-cell imaging at the respective excitation light NOEL is equally determined by the cellular autofluorescence and the fluorescent protein brightness. Last, we show that photomorbidity of multiple wavelengths is additive and imaging conditions absent of photomorbidity can be predicted. Our findings enable researchers to find imaging conditions with minimal impact on physiology and can provide framework for how to approach photomorbidity in other organisms.

使用多个荧光报告基因对活细胞进行延时成像是研究单细胞分子过程的重要工具。然而，即使暴露于中等剂量的可见激发光也会扰乱细胞生理学并改变所研究细胞的定量行为。在这里，我们着手制定指南，以避免多色长期成像中激发光的混杂影响。我们使用宽场荧光显微镜来测量所给予的激发光对酵母生长速率（此处称为光发病率）的影响。我们发现光发病率由每个波长的累积光剂量决定，但与激发光的施加方式无关。重要的是，光发病具有一个阈值光剂量，低于该阈值则无法检测到任何影响 (NOEL)。我们发现，荧光蛋白在相应激发光 NOEL 下对活细胞成像的适用性同样由细胞自发荧光和荧光蛋白亮度决定。最后，我们证明多个波长的光发病率是相加的，并且可以预测没有光发病率的成像条件。我们的研究结果使研究人员能够找到对生理学影响最小的成像条件，并可以为如何处理其他生物体的光发病率提供框架。