Multi-color imaging of the bacterial nucleoid and division proteins with blue, orange, and near-infrared fluorescent proteins.

Studies of the spatiotemporal protein dynamics within live bacterial cells impose a strong demand for multi-color imaging. Despite the increasingly large collection of fluorescent protein (FP) variants engineered to date, only a few of these were successfully applied in bacteria. Here, we explore the performance of recently engineered variants with the blue (TagBFP), orange (TagRFP-T, mKO2), and far-red (mKate2) spectral colors by tagging HU, LacI, MinD, and FtsZ for visualizing the nucleoid and the cell division process. We find that, these FPs outperformed previous versions in terms of brightness and photostability at their respective spectral range, both when expressed as cytosolic label and when fused to native proteins. As this indicates that their folding is sufficiently fast, these proteins thus successfully expand the applicable spectra for multi-color imaging in bacteria. A near-infrared protein (eqFP670) is found to be the most red-shifted protein applicable to bacteria so far, with brightness and photostability that are advantageous for cell-body imaging, such as in microfluidic devices. Despite the multiple advantages, we also report the alarming observation that TagBFP directly interacts with TagRFP-T, causing interference of localization patterns between their fusion proteins. Our application of diverse FPs for endogenous tagging provides guidelines for future engineering of fluorescent fusions in bacteria, specifically: (1) The performance of newly developed FPs should be quantified in vivo for their introduction into bacteria; (2) spectral crosstalk and inter-variant interactions between FPs should be carefully examined for multi-color imaging; and (3) successful genomic fusion to the 5(')-end of a gene strongly depends on the translational read-through of the inserted coding sequence.