Structural Characterization of Fluorescent Proteins Using Tunable Femtosecond Stimulated Raman Spectroscopy.

The versatile functions of fluorescent proteins (FPs) as fluorescence biomarkers depend on their intrinsic chromophores interacting with the protein environment. Besides X-ray crystallography, vibrational spectroscopy represents a highly valuable tool for characterizing the chromophore structure and revealing the roles of chromophore-environment interactions. In this work, we aim to benchmark the ground-state vibrational signatures of a series of FPs with emission colors spanning from green, yellow, orange, to red, as well as the solvated model chromophores for some of these FPs, using wavelength-tunable femtosecond stimulated Raman spectroscopy (FSRS) in conjunction with quantum calculations. We systematically analyzed and discussed four factors underlying the vibrational properties of FP chromophores: sidechain structure, conjugation structure, chromophore conformation, and the protein environment. A prominent bond-stretching mode characteristic of the quinoidal resonance structure is found to be conserved in most FPs and model chromophores investigated, which can be used as a vibrational marker to interpret chromophore-environment interactions and structural effects on the electronic properties of the chromophore. The fundamental insights gained for these light-sensing units (e.g., protein active sites) substantiate the unique and powerful capability of wavelength-tunable FSRS in delineating FP chromophore properties with high sensitivity and resolution in solution and protein matrices. The comprehensive characterization for various FPs across a colorful palette could also serve as a solid foundation for future spectroscopic studies and the rational engineering of FPs with diverse and improved functions.

A Single Fluorescent Protein-Based Indicator with a Time-Resolved Fluorescence Readout for Precise pH Measurements in the Alkaline Range.

The real-time monitoring of the intracellular pH in live cells with high precision represents an important methodological challenge. Although genetically encoded fluorescent indicators can be considered as a probe of choice for such measurements, they are hindered mostly by the inability to determine an absolute pH value and/or a narrow dynamic range of the signal, making them inefficient for recording the small pH changes that typically occur within cellular organelles. Here, we study the pH sensitivity of a green-fluorescence-protein (GFP)-based emitter (EGFP-Y145L/S205V) with the alkaline-shifted chromophore's pKa and demonstrate that, in the pH range of 7.5-9.0, its fluorescence lifetime changes by a factor of ~3.5 in a quasi-linear manner in mammalian cells. Considering the relatively strong lifetime response in a narrow pH range, we proposed the mitochondria, which are known to have a weakly alkaline milieu, as a target for live-cell pH measurements. Using fluorescence lifetime imaging microscopy (FLIM) to visualize the HEK293T cells expressing mitochondrially targeted EGFP-Y145L/S205V, we succeeded in determining the absolute pH value of the mitochondria and recorded the ETC-uncoupler-stimulated pH shift with a precision of 0.1 unit. We thus show that a single GFP with alkaline-shifted pKa can act as a high-precision indicator that can be used in a specific pH range.

Amino acid residue at the 165th position tunes EYFP chromophore maturation. A structure-based design.

For the whole GFP family, a few cases, when a single mutation in the chromophore environment strongly inhibits maturation, were described. Here we study EYFP-F165G - a variant of the enhanced yellow fluorescent protein - obtained by a single F165G replacement, and demonstrated multiple fluorescent states represented by the minor emission peaks in blue and yellow ranges (~470 and ~530 nm), and the major peak at ~330 nm. The latter has been assigned to tryptophan fluorescence, quenched due to excitation energy transfer to the mature chromophore in the parental EYFP protein. EYFP-F165G crystal structure revealed two general independent routes of post-translational chemistry, resulting in two main states of the polypeptide chain with the intact chromophore forming triad (~85%) and mature chromophore (~15%). Our experiments thus highlighted important stereochemical role of the 165th position strongly affecting spectral characteristics of the protein. On the basis of the determined EYFP-F165G three-dimensional structure, new variants with ~ 2-fold improved brightness were engineered.

Impacts of OrX and cAMP-insensitive Orco to the insect olfactory heteromer activity.

Insect odorant receptors (ORs) have been suggested to function as ligand-gated cation channels, with OrX/Orco heteromers combining ionotropic and metabotropic activity. The latter is mediated by different G proteins and results in Orco self-activation by cyclic nucleotide binding. In this contribution, we co-express the odor-specific subunits DmOr49b and DmOr59b with either wild-type Orco or an Orco-PKC mutant lacking cAMP activation heterologously in mammalian cells. We show that the characteristics of heteromers strongly depend on both the OrX type and the coreceptor variant. Thus, methyl acetate-sensitive Or59b/Orco demonstrated 25-fold faster response kinetics over o-cresol-specific Or49b/Orco, while the latter required a 10-100 times lower ligand concentration to evoke a similar electrical response. Compared to wild-type Orco, Orco-PKC decreased odorant sensitivity in both heteromers, and blocked an outward current rectification intrinsic to the Or49b/Orco pair. Our observations thus provide an insight into insect OrX/Orco functioning, highlighting their natural and artificial tuning features and laying the groundwork for their application in chemogenetics, drug screening, and repellent design.

Chromophore reduction plus reversible photobleaching: how the mKate2 "photoconversion" works.

mKate red-to-green photoconversion is a non-canonical type of phototransformation in fluorescent proteins, with a poorly understood mechanism. We have hypothesized that the daughter mKate2 protein may also be photoconvertible, and that this phenomenon would be connected with mKate(2) chromophore photoreduction. Indeed, upon the intense irradiation of the protein sample supplemented by sodium dithionite, the accumulation of green as well as blue spectral forms is enhanced. The reaction was shown to be reversible upon the reductant's removal. However, an analysis of the fluorescence microscopy data, absorption spectra, kinetics and time-resolved fluorescence spectroscopy revealed that the short-wavelength spectral forms of mKate(2) exist before photoactivation, that their fractions increase light-independently after dithionite addition, and that the conversion is facilitated by the photobleaching of the red chromophore form.

Developing Bright Green Fluorescent Protein (GFP)-like Fluorogens for Live-Cell Imaging with Nonpolar Protein-Chromophore Interactions.

Fluorescence-activating proteins (FAPs) that bind a chromophore and activate its fluorescence have gained popularity in bioimaging. The fluorescence-activating and absorption-shifting tag (FAST) is a light-weight FAP that enables fast reversible fluorogen binding, thus advancing multiplex and super-resolution imaging. However, the rational design of FAST-specific fluorogens with large fluorescence enhancement (FE) remains challenging. Herein, a new fluorogen directly engineered from green fluorescent protein (GFP) chromophore by a unique double-donor-one-acceptor strategy, which exhibits an over 550-fold FE upon FAST binding and a high extinction coefficient of approximately 100,000 M-1 cm-1 , is reported. Correlation analysis of the excited state nonradiative decay rates and environmental factors reveal that the large FE is caused by nonpolar protein-fluorogen interactions. Our deep insights into structure-function relationships could guide the rational design of bright fluorogens for live-cell imaging with extended spectral properties such as redder emissions.

Deciphering the Role of Positions 145 and 165 in Fluorescence Lifetime Shortening in the EGFP Variants.

The bright ultimately short lifetime enhanced emitter (BrUSLEE) green fluorescent protein, which differs from the enhanced green fluorescent protein (EGFP) in three mutations, exhibits an extremely short fluorescence lifetime at a relatively high brightness. An important contribution to shortening the BrUSLEE fluorescence lifetime compared to EGFP is provided by the T65G substitution of chromophore-forming residue and the Y145M mutation touching the chromophore environment. Although the influence of the T65G mutation was studied previously, the role of the 145th position in determining the GFPs physicochemical characteristics remains unclear. In this work, we show that the Y145M substitution, both alone and in combination with the F165Y mutation, does not shorten the fluorescence lifetime of EGFP-derived mutants. Thus, the unlocking of Y145M as an important determinant of lifetime tuning is possible only cooperatively with mutations at position 65. We also show here that the introduction of a T65G substitution into EGFP causes complex photobehavior of the respective mutants in the lifetime domain, namely, the appearance of two fluorescent states with different lifetimes, preserved in any combination with the Y145M and F165Y substitutions.

A General Mechanism of Green-to-Red Photoconversions of GFP.

Here we dissect the phenomena of oxidative and reductive green-to-red photoconversion of the Green Fluorescent Protein. We characterize distinct orange- and red-emitting forms (λabs/λem = 490/565 nm; λabs/λem = 535/600 nm) arising during the Enhanced Green Fluorescent Protein (EGFP) photoconversion under low-oxygen conditions in the presence of reductants. These forms spectroscopically differ from that observed previously in oxidative redding (λabs/λem = 575/607 nm). We also report on a new green-emitting state (λabs/λem = 405/525 nm), which is formed upon photoconversion under the low-oxygen conditions. Based on the spectral properties of these forms, their light-independent time evolution, and the high-level computational studies, we provide a structural basis for various photoproducts. Under the low-oxygen conditions, the neutral quinoid-like structure formed via a two-electron oxidation process is found to be a key intermediate and a most likely candidate for the novel green-emitting state of the chromophore. The observed large Stokes shift is traced to the formation of the zwitterionic form of the chromophore in the excited state. Subsequently, this form undergoes two types of cyclization reactions, resulting in the formation of either the orange-emitting state (λabs/λem = 490/565 nm) or the red-emitting form (λabs/λem = 535/600 nm). The T65G mutant lacks one of the proposed cyclization pathways and, indeed, the photoconverted T65G EGFP exhibits a single orange-emitting state. In oxidative redding, the red-emitting state resembles the structure of the chromophore from asFP595 (λabs/λem = 572/595 nm), which is directly formed upon two-electron oxidation and deprotonation bypassing the formation of the quinoid-like structure. Our results disclose a general "oxidative" mechanism of various green-to-red photoconversions of EGFP, providing a link between oxidative redding and the photoconversion under low-oxygen conditions.

Two independent routes of post-translational chemistry in fluorescent protein FusionRed.

The crystal structure of monomeric red fluorescent protein FusionRed (λex/λem 580/608 mn) has been determined at 1.09 Å resolution and revealed two alternative routes of post-translational chemistry, resulting in distinctly different products. The refinement occupancies suggest the 60:40 ratio of the mature Met63-Tyr64-Gly65 chromophore and uncyclized chromophore-forming tripeptide with the protein backbone cleaved between Met63 and the preceding Phe62 and oxidized Cα-Cß bond of Tyr64. We analyzed the structures of FusionRed and several related red fluorescent proteins, identified structural elements causing hydrolysis of the peptide bond, and verified their impact by single point mutagenesis. These findings advance the understanding of the post-translational chemistry of GFP-like fluorescent proteins beyond the canonical cyclization-dehydration-oxidation mechanism. They also show that impaired cyclization does not prevent chromophore-forming tripeptide from further transformations enabled by the same set of catalytic residues. Our mutagenesis efforts resulted in inhibition of the peptide backbone cleavage, and a FusionRed variant with ~30% improved effective brightness.

Plasmon-Enhanced Fluorescence of EGFP on Short-Range Ordered Ag Nanohole Arrays.

Fluorescence of organic molecules can be enhanced by plasmonic nanostructures through coupling to their locally amplified electromagnetic field, resulting in higher brightness and better photostability of fluorophores, which is particularly important for bioimaging applications involving fluorescent proteins as genetically encoded biomarkers. Here, we show that a hybrid bionanosystem comprised of a monolayer of Enhanced Green Fluorescent Protein (EGFP) covalently linked to optically thin Ag films with short-range ordered nanohole arrays can exhibit up to 6-fold increased brightness. The largest enhancement factor is observed for nanohole arrays with a propagating surface plasmon mode, tuned to overlap with both excitation and emission of EGFP. The fluorescence lifetime measurements in combination with FDTD simulations provide in-depth insight into the origin of the fluorescence enhancement, showing that the effect is due to the local amplification of the optical field near the edges of the nanoholes. Our results pave the way to improving the photophysical properties of hybrid bionanosystems based on fluorescent proteins at the interface with easily fabricated and tunable plasmonic nanostructures.

Influence of the First Chromophore-Forming Residue on Photobleaching and Oxidative Photoconversion of EGFP and EYFP.

Enhanced green fluorescent protein (EGFP)-one of the most widely applied genetically encoded fluorescent probes-carries the threonine-tyrosine-glycine (TYG) chromophore. EGFP efficiently undergoes green-to-red oxidative photoconversion ("redding") with electron acceptors. Enhanced yellow fluorescent protein (EYFP), a close EGFP homologue (five amino acid substitutions), has a glycine-tyrosine-glycine (GYG) chromophore and is much less susceptible to redding, requiring halide ions in addition to the oxidants. In this contribution we aim to clarify the role of the first chromophore-forming amino acid in photoinduced behavior of these fluorescent proteins. To that end, we compared photobleaching and redding kinetics of EGFP, EYFP, and their mutants with reciprocally substituted chromophore residues, EGFP-T65G and EYFP-G65T. Measurements showed that T65G mutation significantly increases EGFP photostability and inhibits its excited-state oxidation efficiency. Remarkably, while EYFP-G65T demonstrated highly increased spectral sensitivity to chloride, it is also able to undergo redding chloride-independently. Atomistic calculations reveal that the GYG chromophore has an increased flexibility, which facilitates radiationless relaxation leading to the reduced fluorescence quantum yield in the T65G mutant. The GYG chromophore also has larger oscillator strength as compared to TYG, which leads to a shorter radiative lifetime (i.e., a faster rate of fluorescence). The faster fluorescence rate partially compensates for the loss of quantum efficiency due to radiationless relaxation. The shorter excited-state lifetime of the GYG chromophore is responsible for its increased photostability and resistance to redding. In EYFP and EYFP-G65T, the chromophore is stabilized by π-stacking with Tyr203, which suppresses its twisting motions relative to EGFP.

Red Fluorescent Genetically Encoded Voltage Indicators with Millisecond Responsiveness.

Genetically encoded fluorescent indicators typically consist of the sensitive and reporter protein domains connected with the amino acid linkers. The final performance of a particular indicator may depend on the linker length and composition as strong as it depends on the both domains nature. Here we aimed to optimize interdomain linkers in VSD-FR189-188-a recently described red fluorescent protein-based voltage indicator. We have tested 13 shortened linker versions and monitored the dynamic range, response speed and polarity of the corresponding voltage indicator variants. While the new indicators didn't show a contrast enhancement, some of them carrying very short interdomain linkers responded 25-fold faster than the parental VSD-FR189-188. Also we found the critical linker length at which fluorescence response to voltage shift changes its polarity from negative to positive slope. Our observations thus make an important contribution to the designing principles of the fluorescent protein-derived voltage indicators.

Bright GFP with subnanosecond fluorescence lifetime.

Fluorescence lifetime imaging microscopy (FLIM) measures fluorescence decay rate at every pixel of an image. FLIM can separate probes of the same color but different fluorescence lifetimes (FL), thus it is a promising approach for multiparameter imaging. However, available GFP-like fluorescent proteins (FP) possess a narrow range of FLs (commonly, 2.3-3.5 ns) which limits their applicability for multiparameter FLIM. Here we report a new FP probe showing both subnanosecond fluorescence lifetime and exceptional fluorescence brightness (80% of EGFP). To design this probe we applied semi-rational amino acid substitutions selection. Critical positions (Thr65, Tyr145, Phe165) were altered based on previously reported effect on FL or excited state electron transfer. The resulting EGFP triple mutant, BrUSLEE (Bright Ultimately Short Lifetime Enhanced Emitter), allows for both reliable detection of the probe and recording FL signal clearly distinguishable from that of the spectrally similar commonly used GFPs. We demonstrated high performance of this probe in multiparameter FLIM experiment. We suggest that amino acid substitutions described here lead to a significant shift in radiative and non-radiative excited state processes equilibrium.

Insertion of the voltage-sensitive domain into circularly permuted red fluorescent protein as a design for genetically encoded voltage sensor.

Visualization of electrical activity in living cells represents an important challenge in context of basic neurophysiological studies. Here we report a new voltage sensitive fluorescent indicator which response could be detected by fluorescence monitoring in a single red channel. To the best of our knowledge, this is the first fluorescent protein-based voltage sensor which uses insertion-into-circular permutant topology to provide an efficient interaction between sensitive and reporter domains. Its fluorescent core originates from red fluorescent protein (FP) FusionRed, which has optimal spectral characteristics to be used in whole body imaging techniques. Indicators using the same domain topology could become a new perspective for the FP-based voltage sensors that are traditionally based on Förster resonance energy transfer (FRET).

Functioning of Fluorescent Proteins in Aggregates in Anthozoa Species and in Recombinant Artificial Models.

Despite great advances in practical applications of fluorescent proteins (FPs), their natural function is poorly understood. FPs display complex spatio-temporal expression patterns in living Anthozoa coral polyps. Here we applied confocal microscopy, specifically, the fluorescence recovery after photobleaching (FRAP) technique to analyze intracellular localization and mobility of endogenous FPs in live tissues. We observed three distinct types of protein distributions in living tissues. One type of distribution, characteristic for Anemonia, Discosoma and Zoanthus, is free, highly mobile cytoplasmic localization. Another pattern is seen in FPs localized to numerous intracellular vesicles, observed in Clavularia. The third most intriguing type of intracellular localization is with respect to the spindle-shaped aggregates and lozenge crystals several micrometers in size observed in Zoanthus samples. No protein mobility within those structures was detected by FRAP. This finding encouraged us to develop artificial aggregating FPs. We constructed "trio-FPs" consisting of three tandem copies of tetrameric FPs and demonstrated that they form multiple bright foci upon expression in mammalian cells. High brightness of the aggregates is advantageous for early detection of weak promoter activities. Simultaneously, larger aggregates can induce significant cytostatic and cytotoxic effects and thus such tags are not suitable for long-term and high-level expression.

Photoinduced Chemistry in Fluorescent Proteins: Curse or Blessing?

Photoinduced reactions play an important role in the photocycle of fluorescent proteins from the green fluorescent protein (GFP) family. Among such processes are photoisomerization, photooxidation/photoreduction, breaking and making of covalent bonds, and excited-state proton transfer (ESPT). Many of these transformations are initiated by electron transfer (ET). The quantum yields of these processes vary significantly, from nearly 1 for ESPT to 10-4-10-6 for ET. Importantly, even when quantum yields are relatively small, at the conditions of repeated illumination the overall effect is significant. Depending on the task at hand, fluorescent protein photochemistry is regarded either as an asset facilitating new applications or as a nuisance leading to the loss of optical output. The phenomena arising due to phototransformations include (i) large Stokes shifts, (ii) photoconversions, photoactivation, and photoswitching, (iii) phototoxicity, (iv) blinking, (v) permanent bleaching, and (vi) formation of long-lived intermediates. The focus of this review is on the most recent experimental and theoretical work on photoinduced transformations in fluorescent proteins. We also provide an overview of the photophysics of fluorescent proteins, highlighting the interplay between photochemistry and other channels (fluorescence, radiationless relaxation, and intersystem crossing). The similarities and differences with photochemical processes in other biological systems and in dyes are also discussed.

Turning On and Off Photoinduced Electron Transfer in Fluorescent Proteins by π-Stacking, Halide Binding, and Tyr145 Mutations.

Photoinduced electron transfer in fluorescent proteins from the GFP family can be regarded either as an asset facilitating new applications or as a nuisance leading to the loss of optical output. Photooxidation commonly results in green-to-red photoconversion called oxidative redding. We discovered that yellow FPs do not undergo redding; however, the redding is restored upon halide binding. Calculations of the energetics of one-electron oxidation and possible electron transfer (ET) pathways suggested that excited-state ET proceeds through a hopping mechanism via Tyr145. In YFPs, the π-stacking of the chromophore with Tyr203 reduces its electron-donating ability, which can be restored by halide binding. Point mutations confirmed that Tyr145 is a key residue controlling ET. Substitution of Tyr145 by less-efficient electron acceptors resulted in highly photostable mutants. This strategy (i.e., calculation and disruption of ET pathways by mutations) may represent a new approach toward enhancing photostability of FPs.

Influence of cell growth conditions and medium composition on EGFP photostability in live cells.

Photostability is a key characteristic of fluorescent proteins. It was recently demonstrated that green fluorescent protein (GFP) photobleaching in live cells can be suppressed by changes in medium composition. Here we show that Ham's F12 medium provides very high enhanced GFP (EGFP) photostability during fluorescence microscopy of live cells. This property of Ham's F12 medium is associated with decreased concentrations of riboflavin and pyridoxine, and increased concentrations of FeSO4, cyanocobalamine, lipoic acid, hypoxanthine, and thymidine compared with DMEM. We also found that the rate of EGFP photobleaching strongly depends on cell growth conditions such as cell density and the concentration of serum. We conclude that both imaging medium composition and the physiological state of the cells can strongly affect the photostability of fluorescent proteins. Thus, accurate comparison of the photostabilities of fluorescent proteins should be performed only in side-by-side analysis in identical cell growth conditions and media.

Anti-fading media for live cell GFP imaging.

Photostability is one of the most important characteristic of a dye for fluorescence microscopy. Recently we demonstrated that vitamins present in imaging media dramatically accelerate photobleaching of Enhanced Green Fluorescent Protein (EGFP) and many other green fluorescent and photoactivatable proteins. Here we tested all vitamins of commonly used media (such as Dulbecco's Modified Eagle Medium, DMEM) one-by-one and found that only two vitamins, riboflavin and pyridoxal, decrease photostability of EGFP. Thus, DMEM without riboflavin and pyridoxal can be used as an imaging medium, which ensures high photostability of GFPs at the expense of minimal biochemical disturbance. Then, we tested some antioxidants and found that a plant flavonoid rutin greatly enhances photostability of EGFP during live cell microscopy. In complete DMEM, rutin increased EGFP photostability up to the level of vitamin-depleted DMEM. Moreover, being added to vitamin-depleted DMEM, rutin was able to further suppress EGFP photobleaching. Potentially, new medium formulations can be widely used for fluorescence microscopy of GFP-expressing cells and model multicellular organisms in a variety of imaging applications, where photostability represents a challenge.