Cyan variants of green fluorescent protein are widely used as donors in Förster resonance energy transfer experiments. The popular, but modestly bright, Enhanced Cyan Fluorescent Protein (ECFP) was sequentially improved into the brighter variants Super Cyan Fluorescent Protein 3A (SCFP3A) and mTurquoise, the latter exhibiting a high-fluorescence quantum yield and a long mono-exponential fluorescence lifetime. Here we combine X-ray crystallography and excited-state calculations to rationalize these stepwise improvements. The enhancement originates from stabilization of the seventh β-strand and the strengthening of the sole chromophore-stabilizing hydrogen bond. The structural analysis highlighted one suboptimal internal residue, which was subjected to saturation mutagenesis combined with fluorescence lifetime-based screening. This resulted in mTurquoise2, a brighter variant with faster maturation, high photostability, longer mono-exponential lifetime and the highest quantum yield measured for a monomeric fluorescent protein. Together, these properties make mTurquoise2 the preferable cyan variant of green fluorescent protein for long-term imaging and as donor for Förster resonance energy transfer to a yellow fluorescent protein.

Fluorescent proteins have become extremely popular tools for in vivo imaging and especially for the study of localization, motility and interaction of proteins in living cells. Here we report TagRFP, a monomeric red fluorescent protein, which is characterized by high brightness, complete chromophore maturation, prolonged fluorescence lifetime and high pH-stability. These properties make TagRFP an excellent tag for protein localization studies and fluorescence resonance energy transfer (FRET) applications.

Enhanced cyan and yellow fluorescent proteins are widely used for dual color imaging and protein-protein interaction studies based on fluorescence resonance energy transfer. Use of these fluorescent proteins can be limited by their thermosensitivity, dim fluorescence, and tendency for aggregation. Here we report the results of a site-directed mutagenesis approach to improve these fluorescent proteins. We created monomeric optimized variants of ECFP and EYFP, which fold faster and more efficiently at 37 degrees C and have superior solubility and brightness. Bacteria expressing SCFP3A were 9-fold brighter than those expressing ECFP and 1.2-fold brighter than bacteria expressing Cerulean. SCFP3A has an increased quantum yield (0.56) and fluorescence lifetime. Bacteria expressing SYFP2 were 12 times brighter than those expressing EYFP(Q69K) and almost 2-fold brighter than bacteria expressing Venus. In HeLa cells, the improvements were less pronounced; nonetheless, cells expressing SCFP3A and SYFP2 were both 1.5-fold brighter than cells expressing ECFP and EYFP(Q69K), respectively. The enhancements of SCFP3A and SYFP2 are most probably due to an increased intrinsic brightness (1.7-fold and 1.3-fold for purified recombinant proteins, compared to ECFP & EYFP(Q69K), respectively) and due to enhanced protein folding and maturation. The latter enhancements most significantly contribute to the increased fluorescent yield in bacteria whereas they appear less significant for mammalian cell systems. SCFP3A and SYFP2 make a superior donor-acceptor pair for fluorescence resonance energy transfer, because of the high quantum yield and increased lifetime of SCFP3A and the high extinction coefficient of SYFP2. Furthermore, SCFP1, a CFP variant with a short fluorescence lifetime but identical spectra compared to ECFP and SCFP3A, was characterized. Using the large lifetime difference between SCFP1 and SCFP3A enabled us to perform for the first time dual-lifetime imaging of spectrally identical fluorescent species in living cells.

Epac-based FRET sensors have been widely used for the detection of cAMP concentrations in living cells. Originally developed by us as well as others, we have since then reported several important optimizations that make these sensors favourite among many cell biologists. We here report cloning and characterization of our fourth generation of cAMP sensors, which feature outstanding photostability, dynamic range and signal-to-noise ratio. The design is based on mTurquoise2, currently the brightest and most bleaching-resistant donor, and a new acceptor cassette that consists of a tandem of two cp173Venus fluorophores. We also report variants with a single point mutation, Q270E, in the Epac moiety, which decreases the dissociation constant of cAMP from 9.5 to 4 μM, and thus increases the affinity ~ 2.5-fold. Finally, we also prepared and characterized dedicated variants with non-emitting (dark) acceptors for single-wavelength FLIM acquisition that display an exceptional near-doubling of fluorescence lifetime upon saturation of cAMP levels. We believe this generation of cAMP outperforms all other sensors and therefore recommend these sensors for all future studies.

Comparative genome- and proteome-wide screens yield large amounts of data. To efficiently present such datasets and to simplify the identification of hits, the results are often presented in a type of scatterplot known as a volcano plot, which shows a measure of effect size versus a measure of significance. The data points with the largest effect size and a statistical significance beyond a user-defined threshold are considered as hits. Such hits are usually annotated in the plot by a label with their name. Volcano plots can represent ten thousands of data points, of which typically only a handful is annotated. The information of data that is not annotated is hardly or not accessible. To simplify access to the data and enable its re-use, we have developed an open source and online web tool with R/Shiny. The web app is named VolcaNoseR and it can be used to create, explore, label and share volcano plots (https://huygens.science.uva.nl/VolcaNoseR). When the data is stored in an online data repository, the web app can retrieve that data together with user-defined settings to generate a customized, interactive volcano plot. Users can interact with the data, adjust the plot and share their modified plot together with the underlying data. Therefore, VolcaNoseR increases the transparency and re-use of large comparative genome- and proteome-wide datasets.

Reporting of the actual data in graphs and plots increases transparency and enables independent evaluation. On the other hand, data summaries are often used in graphs because they aid interpretation. To democratize state-of-the-art data visualization of raw data with a selection of statistical summaries, a freely available, open-source web app was written using R/shiny that uses the ggplot2 package for generating plots. Users can to choose how to display the data and which of the data summaries to add. In addition, the 95% confidence intervals (95CIs) can be added for visual inferences. By adjusting the visibility of the layers, the visualization of the raw data and their summaries can be tuned for optimal presentation and interpretation. The app is dubbed PlotsOfData and is available at https://huygens.science.uva.nl/PlotsOfData/ .

Optimization of autofluorescent proteins by intensity-based screening of bacteria does not necessarily identify the brightest variant for eukaryotes. We report a strategy to screen excited state lifetimes, which identified cyan fluorescent proteins with long fluorescence lifetimes (>3.7 ns) and high quantum yields (>0.8). One variant, mTurquoise, was 1.5-fold brighter than mCerulean in mammalian cells and decayed mono-exponentially, making it an excellent fluorescence resonance energy transfer (FRET) donor.

SummaryPolyphosphoinositides represent a minor group of phospholipids, accounting for less than 1% of the total. Despite their low abundance, these molecules have been implicated in various signalling and membrane trafficking events. Phosphatidylinositol 4-phosphate (PtdIns4P ) is the most abundant polyphosphoinositide. 32 Pi-labelling studies have shown that the turnover of PtdIns4P is rapid, but little is known about where in the cell or plant this occurs. Here, we describe the use of a lipid biosensor that monitors PtdIns4P dynamics in living plant cells. The biosensor consists of a fusion between a fluorescent protein and a lipid-binding domain that specifically binds PtdIns4P, i.e. the pleckstrin homology domain of the human protein phosphatidylinositol-4-phosphate adaptor protein-1 (FAPP1). YFP-PH FAPP1 was expressed in four plant systems: transiently in cowpea protoplasts, and stably in tobacco BY-2 cells, Medicago truncatula roots and Arabidopsis thaliana seedlings. All systems allowed YFP-PH FAPP1 expression without detrimental effects. Two distinct fluorescence patterns were observed: labelling of motile punctate structures and the plasma membrane. Co-expression studies with organelle markers revealed strong co-labelling with the Golgi marker STtmd-CFP, but not with the endocytic/pre-vacuolar marker GFP-AtRABF2b. Co-expression with the Ptdins3P biosensor YFP-2 · FYVE revealed totally different localization patterns. During cell division, YFP-PH FAPP1 showed strong labelling of the cell plate, but PtdIns3P was completely absent from the newly formed cell membrane. In root hairs of M. truncatula and A. thaliana, a clear PtdIns4P gradient was apparent in the plasma membrane, with the highest concentration in the tip. This only occurred in growing root hairs, indicating a role for PtdIns4P in tip growth.