Fluorescent Probes Based on Charge and Proton Transfer for Probing Biomolecular Environment

Fluorescent probes for sensing fundamental properties of biomolecular environment, such as polarity and hydration, help to study assembly of lipids into biomembranes, sensing interactions of biomolecules and imaging physiological state of the cells. Here, we summarize major efforts in the development of probes based on two photophysical mechanisms: (i) an excited‐state intramolecular charge transfer (ICT), which is represented by fluorescent solvatochromic dyes that shift their emission band maximum as a function of environment polarity and hydration; (ii) excited‐state intramolecular proton transfer (ESIPT), with particular focus on 5‐membered cyclic systems, represented by 3‐hydroxyflavones, because they exhibit dual emission sensitive to the environment. For both ICT and ESIPT dyes, the design of the probes and their biological applications are summarized. Thus, dyes bearing amphiphilic anchors target lipid membranes and report their lipid organization, while targeting ligands direct them to specific organelles for sensing their local environment. The labels, amino acid and nucleic acid analogues inserted into biomolecules enable monitoring their interactions with membranes, proteins and nucleic acids. While ICT probes are relatively simple and robust environment‐sensitive probes, ESIPT probes feature high information content due their dual emission. They constitute a powerful toolbox for addressing multitude of biological questions.


Introduction
Molecular fluorescent probes are increasingly popular tools for sensing and visualizing biological [1] and chemical [2] systems.This research direction is accelerated by rapid developments of optical instruments.The state-of-the art devices enable the acquisition of single photons in the femtosecond time scale from the volumes less than a femtoliter as well as detect single molecules and image cells with nanoscopic resolution.On the other hand, the rapid development of organic chemistry methods has now made available new quite complex molecules, including fluorescent probes, which are essential to provide the information about structure, composition and interactions in a biological system in form of a fluorescence signal.Let us first define general principles for construction a molecular fluorescent probe.
Despite small size of fluorescent molecular probes, they are complex devices consisting of several functional elements (Figure 1A,B).They contain a fluorescent dye, capable to both absorb and emit light.In case of dyes undergoing Forster Resonance Energy Transfer (FRET) or aggregation-based dyes, two or more dyes can be used within the same probe.The probe should also contain a recognition unit responsible for the selective binding of studying analyte in a complex medium containing other competing molecules.A signal transduction unit ensures coupling of the recognition unit with the fluorophore and, thus, switches the color or intensity of fluorescence in the presence of an analyte.In some systems, for example probes sensitive to physical properties of their environment, like polarity and viscosity (environment-sensitive probes), the recognition and signal transduction units are directly integrated within the same fluorescent dye.The probes that target specifically a biomolecule or cellular compartment contain two additional elements, an anchor, and frequently -a linker.The purpose of the anchor is to immobilize the probe covalently (also called label) or noncovalently at the desired site of a studied object.The linker (or spacer) regulates the distance and probe orientation relatively to the object for an adequate function of the probe.In addition, the probe is frequently equipped with hydrophilic groups (e. g., carboxylates, sulfonates, polyethylene glycol or peptide chains etc.), which prevent uncontrolled aggregation, non-specific interactions with components of a biological system or fluorescence quenching by water.There are also specific requirements for the shape, size and charge of the molecular fluorescent probe.For example, a strict adequacy of these parameters should be respected for probes mimicking nucleotides and lipids.
Let us discuss specific examples of fluorescent probes.In the first example, indicator Fluo-3 is used to determine Ca 2 + concentration in cells (Figure 1A). [3]It is composed of Ca 2 + recognition unit, linker and fluorescent dye units.The recognition unit, based on BAPTA derivative, is highly selective for binding Ca 2 + ions.Both nitrogen atoms in the recognition unit serve as the signal transduction units of cation recognition.Being free of cation, their lone electron pairs switch off the fluorescence of the dye as a result of photoinduced electron transfer (PET) to the fluorophore unit (see below).Such probe shows a single emission band of calcium complex in the fluorescence spectrum, whose intensity reflects the concentration of Ca 2 + in solution in a range of 2 • 10 À 8 -2 • 10 À 6 M.This type of probe provides so-called intensiometric response to the analyte.To shift the range of Ca 2 + detection to higher concentrations, the electron withdrawing groups (e. g., fluorine atom or nitro group) were installed into aromatic rings of the recognition unit. [4] more advanced example is a calcium indicator working on a lipid membrane interface (Figure 1B). [5]3a] The linker ensures a ~0.5 nm distance from the Ca 2 + recognition unit to the membrane surface.The recognition unit, also based on BAPTA derivative, is designed for selective binding of Ca 2 + .The role of signal modulation unit is played by the nitrogen atom conjugated with the fluorophore unit.Its lone electron pair can be withdrawn by the fluorophore or Ca 2 + cation.So, binding of Ca 2 + cation induces a shift of absorption, excitation and emission spectra of the probe to shorter wavelengths.This type of probe provides ratiometric response to the analyte.
It is important to compare intensiometric and ratiometric methods of detection for biological applications.Intensity variations are easy to acquire using a single detection channel at a fixed wavelength of a fluorometer or a microscope.However, the intensity is not an absolute value and would depend on multiple factors, such a volume of the analytical cell, power density of the excitation light and probe local concentration, local viscosity or a presence of different fluorescence quenchers.The intensiometric probes transmit information about a single parameter of the environment only, while it is often important to obtain information about two or more of its parameters.In this respect, the ratiometric Vasyl G. Pivovarenko is a professor at the Faculty of Chemistry, Taras Shevchenko National University of Kyiv.He has more than 30 years' experience in design and synthesis of fluorescent probes and was the team leader of numerous projects in this direction.Prof. Pivovarenko is an expert in fluorescence spectroscopy of organic dyes and other fluorophores, in probing local parameters of complex microheterogeneous systems like peptides, proteins, micelles, lipid vesicles and their complexes.In 2020 he was awarded the "Excellence in Education" medal of the Ministry of Education and Science of Ukraine for his outstanding contribution to chemistry education.detection method has multiple advantages.First, the intensity ratio is an absolute value that is independent of certain experimental settings, like light excitation power and probe concentration.Second, ratiometric probes provide additional channels, such as intensity of individual bands, and their ratio.
In the probe design, one should distinguish "classical", environment-sensitive fluorophores, [6] and aggregation-induced emission (AIE) dyes. [7]The classical dyes are represented by well-known commonly used dye families, such as xanthene (e. g. rhodamines), cyanine and BODIPY series (Figure 2A).These are very bright dyes because of their high extinction coefficient and fluorescence quantum yield. [8]Their fluorophore is generally planar with highly developed conjugation and a delocalized charge that ensure high oscillator strength.These dyes are particularly suitable for designing intensiometric probes operating by PET or aggregation-caused quenching mechanism.The latter would require at least two fluorophores that provide ON/OFF response because of change in the dyes aggregated state (Figure 2B). [9]To design ratiometric probes, the "classical" dyes should be combined in FRET pairs, where the change in the donor-acceptor distance would switch emission from donor to acceptor. [10]n the other hand, environment-sensitive dyes change their emission intensity and/or color as a function of their environment, such as polarity, viscosity, etc. [6] Their chemical structure features an asymmetric electronic distribution and/or incomplete planarization of their fluorophore or reaction centers.These dyes undergo photophysical processes that ensures their sensitivity to the environment. [6]Thus, push-pull dyes, composed of electron donor and acceptor connected by a conjugated unit (Figure 2C).They undergo excited state intramolecular charge transfer (ICT) and, thus, exhibit significant variation of their emission color as a function of polarity (see below).Molecular rotors are non-planar structures with rotating conjugated units (Figure 2C).The molecular rotation provides effective quenching of the dye in the excited

R e v i e w T H E C H E M I C A L R E C O R D
state by the non-viscous environment, whereas high viscosity inhibits intramolecular rotation leading to fluorescence light up. [11]Ground-state rotation in sterically hindered conjugated structures of so-called fluorescent "flippers" [12] (Figure 2C) was also exploited to generate probes sensitive to the environment rigidity.The dyes bearing proton donor and acceptor groups can undergo excited-state intramolecular proton transfer (ESIPT), [13] which generates dual emission sensitive to multiple characteristics of the environment (Figure 2C).
In comparison to classical dyes, environment-sensitive dyes are characterized by a significantly lower brightness.Indeed, push-pull fluorophores and non-planar dyes present lower oscillator strength and thus lower extinction coefficient, while their quantum yields strongly depend on their environment.These shortcomings are a price to pay for their unique capacity to sense their molecular environment.Moreover, an important advantage of some environment-sensitive push-pull ICT dyes and ESIPT dyes is capacity to change their emission color, which provides the ratiometric response. [6]ne should also mention a rapidly developing AIE dyes (Figure 2C). [7]They are considered apart because, on the one hand, similarly to some environment sensitive dyes they are twisted fluorophores and some of them can present push-pull architecture.On the other hand, they present a unique feature to light up in the solid state, generally due to restriction of intramolecular rotation.They gave rise to a variety of probes where response to the analyte is triggered by molecular aggregation, which was described in several comprehensive reviews. [7,14]Aggregation-induced emission enhancement (AIEE) dyes is additional subclass of dyes with environmentsensitive behavior. [15]he present review will focus on push-pull ICT and ESIPT dyes, which are particularly suitable for sensing their local molecular environment, such as polarity, H-bonding and hydration.These are fundamental properties that enable monitoring interactions of biomolecules and deciphering their organization.Thus, interaction of a biomolecule with another biomolecule or biomembrane is associated with exclusion of water and decrease in the local polarity, which can be monitored as a switch in the emission of color of the probes located in the interaction site. [6]On the other hand, change in the lipid organization within the biological membrane, for example between ordered and disordered phases, can drastically change the water penetration, which can be readily monitored for ICT and ESIPT probes.Here, we will present principles of molecular design of these probes and present multitude of applications, ranging from biomolecular sensing up to super-resolution cellular imaging.Given the theme of the special issue, some accent is done on the efforts of Ukrainian researchers in this field.

Fluorophore Design
The dyes undergoing excited state intramolecular charge transfer (ICT) constitute the large class of environmentsensitive dyes.These are essentially push-pull dyes, presenting electron donor and acceptor groups connected by an electronic conjugation through aromatic rings and double bonds.They increase the dipole moment in the excited state due to ICT from the donor to the acceptor group.The stronger influence of polar environment on its excited state compared to the ground state leads to decrease in the energy gap of HOMO-LUMO, which is reflected as significant red shift in the emission.Therefore, these dyes present strong red shifts in the emission, called fluorescence solvatochromism, or solvatofluorochromism.These dyes are also called fluorescent solvatochromic dyes.The primary examples known in literature are Prodan, NBD, dimethylaminonitrostilbene and Nile Red derivatives.With the exception of the two last, vast majority of early examples of fluorescent solvatochromic dyes operate in the blue-green regions.This region is not the most optimal for fluorescence imaging, because of phototoxicity of blue light excitation, strong light scattering and insufficient tissue penetration.
The blue-green operating range contrasts with the "classical" fluorophores like cyanines and rhodamines, where the charge is delocalized within the long electronic conjugation, allowing absorption and emission colors ranging from blue to near-infrared.In contrast, in the push-pull dyes, the donor and acceptor groups transfer relatively small charge in the ground state, but substantially higher in the excited one.This process is directly connected with two fluorescence quenching mechanisms, such as PET and internal conversion, so the design of the red shifted push-pull dyes is problematic.The process is also characterized by low oscillator strength, which means relatively low extinction coefficient for majority of push-pull dyes.
To start the systematic design of improved push-pull fluorophores, we considered development of Prodan analogues with improved properties (Figure 3).It features absorption in UV, low extinction coefficient (~14,000 M À 1 cm À 1 ) and low two-photon absorption cross-section. [16]At that time, the only example was anthracene analogue of Prodan, which showed some red shift in the absorption, but relatively low extinction coefficient. [17]Therefore, we considered using fluorene core, which is attractive for fabrication of large range of optically active materials as well as dyes with strong two-photon absorption cross-section.The obtained derivatives FR0 and FR8 (Figure 3) showed red shifted absorption and emission, > 2-fold higher extinction coefficient (43000 M À 1 cm À 1 ) and > 5-fold higher two-photon cross-section (400 GM).Moreover, larger electronic conjugation resulted in > 2-fold larger transition dipole moment and thus improved significantly its sensitivity to polarity.Later on, we developed another close analogue by conjugating diethylaminophenyl group with furaldehyde (DAF, Figure 3), yielding solvatochromic dye absorbing around 390 nm with good extinction coefficient (up to 44,000 M À 1 cm À 1 ) as well as good solvatochromism. [18]iko et al developed another analogue of Prodan, but based on pyrene (PA, Figure 3).It also showed remarkably improved spectroscopic properties, notably red shifted absorption and emission and dramatically improved fluorescence quantum yield and photostability. [19]High fluorescence quantum yield observed even in apolar solvents was explained by blocking of n-pi transition state due to more favorable pi-pi transition in this dye.To further extend the pi-pi conjugation we coupled benzofuran and fluorene to an acceptor heterocycle of 3-methoxychromone (3MC-2, Figure 3).As a result, we could shift the absorption further to 440 nm, in case of benzofuran derivative, although one should admit that the shift was relatively small.A significant improvement in the red shifted absorption was achieved by Lord et al in case of DCDHF family (Figure 3), which contained remarkably strong acceptor and polarizable dialkylamino-furanyl donor moieties. [20]A remarkable example is recently reported family of push-pull thiazolothiazoles with strong nitrophenyl acceptor (Bu2N-TTz-NO2, Figure 3).The long conjugation and planar structure ensured absorption around 450 nm and remarkably high transition dipole moment. [21]o further red shift the absorption and emission, we designed push-pull dye with a dioxaborine acceptor unit (DXB-Red, Figure 3).This latter unit is particularly interesting because it is stronger acceptor than classical carbonyl unit and at the same time it fixes the rotation of the carbonyl unit.In the first example, dioxaborine cycle was coupled with dialkylamino-styryl unit. [22]The obtained dye showed absorption above 550 nm, which shifted at higher polarity above 600 nm.The emission of this dye also showed good solvatochromism, with emission shifting from 556 nm in hexane to 659 nm in DMSO.As both absorption and emission shifted strongly to the red, we could conclude that strong acceptor properties of dioxaborine unit produced a significant ground state charge transfer.This phenomenon could explain strongly red shifted absorption and unusually high extinction coefficient of this dye: 120,000 M À 1 cm À 1 .On the other hand, this dye also showed significant dependence of its quantum yield of viscosity especially in polar solvents.This was clearly connected with twisted intramolecular charge transfer (TICT) behavior of this probe, which was favored in polar solvents but inhibited at higher viscosity.Therefore, this dye DXB-Red exhibits properties of a solvatochromic dye and a molecular rotor.
Later on, we replaced benzene ring in the aromatic linker with longer and more polarizable benzofuryl. [23]Remarkably, we obtained a near-infrared solvatochromic dye (DXB-NIR, Figure 3) with absorption ranging from 602 to 678 nm and high extinction coefficient reaching 128,000 M À 1 cm À 1 .Moreover, it showed remarkably high two-photon absorption crosssection: 13800 GM at 930 nm.Its solvatochromism in emission showed red shift from 602 nm in hexane till 778 nm in DMSO, while its quantum yield did now show significant dependence on viscosity. [23]DXB-NIR can be considered as unique example of a NIR solvatochromic dye, featuring both strong solvatochromism and high absorption coefficient (single and two-photon) as well as good fluorescence quantum yields.

Plasma Membranes
Cell plasma membrane is an ideal target for solvatochromic dyes because it is highly lipophilic compartment with strong polarity gradient.As solvatochromic dyes are generally quenched by water, their partitioning into lipid membrane leads to fluorescence light up (fluorogenic response) and characteristic spectral shift that reflects local membrane properties. [24]In particular, local polarity correlates directly with other properties of lipid membranes such as lipid composition and lipid order. [25]n the early works, solvatochromic dye Laurdan, which is a more lipophilic analogue of Prodan, showed blue shifted emission in liquid ordered (Lo) phase composed of saturated lipids and cholesterol, in comparison liquid disordered (Ld) phase, composed mainly of unsaturated lipids. [26]Later on, it was found that this blue shift is a generic behavior of solvatochromic fluorescent dyes in lipid membrane, including styryl pyridinium dyes, [27] Nile Red, [28] push-pull fluorene, [29] pyrene, [30] diaoxaborine, [23] etc.
Among biological membranes, the primary target of solvatochromic probes is the plasma membrane, which is the frontier between the cell and the external environment.It is the first cellular compartment that fluorescent molecules encounter when added to the cells.This does not make plasma membrane an easy target for fluorescent probes. [24]Indeed, one of the first examples of solvatochromic membrane probes, Laurdan, showed significant internalization inside the cells.Its highly lipophilic nature allows easy transfer through the plasma membranes towards intracellular lipid compartments, which leads to a strong non-specific staining of intracellular lipid compartments.
To solve this problem Loew et al, proposed to use zwitterionic group in the styryl pyridinium dyes (ANEPPS family). [31]These dyes showed strong charge transfer behavior and found application for detection of transmembrane potential as well as monitoring lipid order in live cells.Later on, we introduced a membrane anchor based on zwitterionic group with dodecyl chain and validated it in case of 3hydroxyflavone (see below). [32]It was found to be universal, as it was successfully applied to different solvatochromic dyes, such as Nile Red, [28] push-pull fluorene, [29] 3-methoxychromone, [33] as well as classical dyes such as BODIPY [34] and cyanines (Figure 4). [35]n particular, one should highlight the Nile Red derivative bearing this anchor group: NR12S (Figure 4A). [28]In model membranes it showed strong blue shift in the emission when the Lo phase was compared to the Ld phase.Moreover, experiments with lipid membranes and sodium dithionite that bleaches Nile Red, revealed only very slow flip-flop process on the timescale of hours, while parent Nile Red underwent flipflop on instantaneously.In cells, NR12S showed highly specific staining of cell plasma membrane, while patent Nile Red internalized, in line with our flip-flop studies. [28]To image spectral shifts in the emission, we analyzed the ratio of the blue to red emission spectrum and presented it in pseudo-color in fluorescence ratiometric images.These results revealed that the outer leaflet of plasma membrane is characterized by highly ordered lipid membrane close to Lo phase, which is supported by its high content in sphingomyelin and cholesterol.NR12S found numerous applications worldwide to study cholesterol, [36] lipid order during apoptosis, [37] eryptosis, [38] membrane asymmetry, [39] maturation of endosomes, [40] internalization of nanoparticles, [41] etc.
Further improvement to the lipid anchor was done by replacing it with anionic sulfonate group with dodecyl chain.The obtained analogue NR12A (Figure 4A) showed improved brightness due to more effective partitioning into the bilayer and higher sensitivity to lipid phases in model membranes (Figure 4B,C). [42]Moreover, alternative substitution of Nile Red (at dialkylamino side vs phenolic side) significantly improved its photostability.Importantly, sulphonate anchor with short butyl chain was found to bind reversibly to plasma membranes, which was accompanied by ON-OFF switching.The latter enabled super-resolution PAINT imaging of plasma membrane lipid organization of live cells, revealing formation of nanoscopic domains of Ld phase of curved structures on the cell surface (figure 4D,E). [42]The anionic anchor with dodecyl chain has been recently applied to Prodan fluorophore to obtain Pro12A (Figure 4A), plasma-membrane specific analogue of Laurdan. [43]

Intracellular Membranes
Lipophilic fluorescent dyes without plasma membrane anchor group can penetrate inside the cells and stain non-specifically a

R e v i e w T H E C H E M I C A L R E C O R D
variety of intracellular biomembranes.Indeed, inside the cells biomembranes delimit specific organelles, such as mitochondria, lysosomes, endoplasmic reticulum, Golgi Apparatus, etc.
In this respect, it is interesting to label non-specifically all these membranes by a solvatochromic dye and monitor distribution of polarity all over the cell.Here, we made use of push-pull pyrene probe PA (Figure 5). [30]This probe showed strong color change in response to lipid order, similarly to Laurdan, but it operated the red-shifted region, providing higher signal to noise ratio.In live cells, it stained multiple biomembranes and provided evidence for a gradient of membrane polarity between cell plasma membranes and intracellular membranes (Figure 5B).This gradient is probably controlled by highly asymmetric distribution of cholesterol.Later on, this probe was improved by replacing aldehyde acceptor group with ketone (PK, Figure 5A), which prevented it from making Schiff-bases with amino groups inside the cells. [44]This probe was applied for polarity mapping in live cells and zebrafish embryo, revealing remarkable contrast within different newly forming organs (Figure 5C,D). [44]Later Adapted with permission from. [42]gure 5. Push-pull pyrene probes PK and PA (A) and their applications for imaging polarity distribution in cells (PA) and zebrafish embryo (PK, C,D).(B) Two-color ratiometric imaging of cells (left) and model lipid vesicles (right), presenting Ld (DOPC) and Lo (SM/Chol) phases using PA probe.Adapted with permission from. [30](C,D) Two-color ratiometric confocal imaging of a zebrafish embryo at different development stages: 6 h (C) and 32 h (eye part, D) postfertilization, using PK probe.Adapted with permission from. [44]n, other teams made use of Nile Red dye to image at high resolution polarity of different biomembranes, and were able to recognize multiple intracellular membrane compartments.[45] DXB-NIR dye was also tested to monitor lipid organization in cells.[23] First, owing to strong solvatochromism, it showed record-breaking blue shift of emission band from by 80 nm from Ld to Lo phase.In cells, two-color ratiometric imaging showed strong variation of colors within different intracellular compartments, such as plasma membranes, endoplasmic reticulum, and lipid droplets.Remarkably, DXB-NIR revealed that starvation and oxidative stress increased the local polarity, where the stress response varied within different studied cell compartments.[23]

Specific Organelles
Alternative approach is to target specifically biomembranes of organelles by solvatochromic dyes.To this end, specially designed ligands are used, which exploits cell biochemistry to accumulate in a specific compartment. [46]49b,d,50] Recently, we exploited the ligand targeting strategy for multiple organelles using Nile Red fluorophore (Figure 6). [51]Thus, the dye was functionalized with different ligands: triphenylphosphonium, N-alkylmorpholin, propyl chloride and pentafluorophenyl, myristic acid and two cyclohexyl groups, that enabled specific targeting of mitochondria, lysosomes, ER, Golgi, and lipid droplets, respectively.The obtained probes enabled two-color ratiometric imaging of the local polarity of organelles in live cells (Figure 6). [51]Each organelle was characterized by its unique local polarity profile and lipid order, which is probably related to its specific lipid composition.Experiments with cholesterol extraction/enrichment suggested that the probes are sensitive to the lipid order in multiple organelles.Interestingly, organelles poor in cholesterol were particularly affected by its enrichment, while those reach in cholesterol were rather sensitive to cholesterol extraction.These probes revealed that each organelle responded differently to the external oxidative and mechanical stresses, where particularly stress-sensitive were mitochondria and lysosomes.
Lipid droplets are particularly interesting compartments, which raised growing attention, and new evidences show that biological role of LDs goes far beyond simple storage of nutrients. [52]On the other hand, it is a relatively simple target for organic dyes.LDs are the most hydrophobic lipid compartment of the cells and, therefore, attracts most of lipophilic dyes with sufficiently high LogP. [53]Thus, parent Nile Red is commonly used as a lipid droplet marker, [54] even though it shows non-negligible fluorescence in other lipid compartments.A vast number of dyes have been developed to target LDs. [53]Among them, solvatochromic dyes can provide additional information about functional states of LDs.
Thus, we reported a push-pull DAF, which stained intracellular lipids compartments with particular preference to LDs. [18] Owing to solvatochromism, it revealed heterogeneity of LDs within the same cell as well as within the same droplet.On the other hand, DXB-NIR mentioned above showed that polarity of LDs can drastically increase on the oxidative stress. [23]Nile Red derivative bearing two cyclohexyl groups showed better selectivity to LDs than parent Nile Red and showed that it is the most apolar organelle within the cell (Figure 6). [51]gure 6.Nile Red-based solvatochromic probes for specific organelles (A) and their application for ratiometric confocal imaging of KB cells (B).Adapted with permission from. [51]

Biomolecular Interactions
Solvatochromic dyes are particularly suitable for detection of interactions between biomolecules, because this interaction results in exclusion of water from the site of interaction.In a seminal works of Imperialli and co-workers, fluorescent amino acids based on solvatochromic dyes 4DMP [55] and 6DMN (Figure 3) were reported. [56]When incorporated into peptides, they enabled monitoring their interaction with MHC proteins, responsible for cell-cell recognition.We conjugated solvatochromic Nile Red with a cyclic peptide carbetocin, which is a ligand for oxytocin GPCR. [57] Similar results were obtained with conjugate of carbetocin with DXB-Red, presetting both solvatochromic and molecular rotor properties. [22]he solvatochromic dyes were grafted also to specific proteins in live cells using Tag technology.Thus, Dumat and co-workers used rhodanine-based solvatochromic dye to obtain fluorogenic response on HaloTag based labelling. [59]Particularly popular for the Tag-based labelling was Nile Red with multiple examples: (1) it was used in combination with SnapTag to achieve fluorogenic response to protein labelling; [60] (2) it was grafted to membrane protein receptors to achieve sensitivity to transmembrane potential; [61] (3) it was used in combination with HaloTag method to monitor protein aggregation in live cells. [62]ecently, we developed a series of Nile Red derivatives for HaloTag labelling (Figure 7). [63]We found that the length of alkyl chains is crucial to achieve both efficient cell penetration and specific protein labelling without non-specific accumulation in the membrane structures.It was targeted to proteins located in different cell compartments, revealing strong variation in their local nano-environment (Figure 7).As a general trend, membrane proteins exhibited much less polar nano-environment compared to cytosolic proteins, which was clearly related to proximity of low polar lipid membrane.
Among solvatochromic dyes, Nile Red derivatives turned out to be particularly successful as environment-sensitive probes. [24]On the one hand, synthetic availability allowed synthesis of differently substituted Nile Red probes.On the other hand, unlike many other push-pull dyes, Nile Red is characterized by highly rigid architecture and long conjugation, similar to that of rhodamine dyes.As a result, Nile Red combines key attractive features, such as high fluorescence quantum yield (due to minimized TICT) and extinction coefficient, good photostability and operation in the red spectral region.
Solvatochromic dyes were also coupled with nucleic acids to study their interactions. [64]In this case, one should use fluorophores of smallest possible size in order to minimize its direct effect of hybridization of DNA.Therefore, the most commonly used example is 2-AP derivative, but it has relatively weak solvatochromism. [65]Dziuba et al designed fluorescent nucleoside based on push-pull fluorene. [66]Its high sensitivity to polarity, enabled detection of DNA-protein (p53) or DNA-lipid interactions, showing strong emission color changes visible even to the naked eye.Adapted with permission from. [63]

ESIPT Dyes with 6-Membered H-Bonding Cycle
Dyes in the excited state may change acidity of their groups, which can lead to proton transfer processes.When the proton is H-bonded with a neighboring basic H-bond acceptor, the intramolecular proton transfer (ESIPT) may take place. [13,67]sually, in the excited state the increase in the acidity of the proton donor is accompanied by the increase in the basicity of the proton acceptor. Some selected examples of these systems are presented in Figure 8B.
The prominent systems are derivatives of salicylic aldehyde, ketone of acid (SAÀ R), where the proton is transferred from phenolic hydroxyl to carbonyl oxygen.The other large class of ESIPT dyes are 2-hydroxyphenyl derivatives of five-membered nitrogen-containing heterocycles, like benzoxazole, benzothiazole and benzimidazole (HBX, Figure 8B).In this case, proton is transferred from hydroxyl to nitrogen of the heterocycle.The proton can be also transferred from proton linked to nitrogen to another nitrogen, as in case of HBX2 (Figure 8B), but these are rarer cases.
Overall, the 6-membered H-bonded cycles present two key features.First, the intramolecular H-boding in this system is very strong in the ground and excited state.Second, excited state leads to dramatic increase in the acidity of the proton donor and increase in the basicity of the proton acceptor.These two features make their ESIPT highly efficient and generally irreversible processes.Therefore, in these compounds, the way to control ESIPT for sensing is to substitute the proton donor with a cleavable bond (example HBX1) [69] or to force disruption of this bond by formation of metal complexes, as shown in case of Zn probe HBX2 [70] (Figure 8B).The reader is directed to a review dedicated to design of this type of ESIPT probes. [71]On the other hand, controlling ESIPT process without breaking the H-bond is challenging in the 6membered H-bonding cycles.
Nevertheless, resent examples showed that modification of ESIPT core system with strong donors, exemplified by SA1, [72] SA2 [73] and HBX3 or acceptors, exemplified by HBX4 [74] and HBX5, [75] can result in the inhibition of ESIPT (Figure 8B).In particular, presence of these groups and extension of conjugation chain length may favor formation of excited state charge transfer states, which could stabilize phenolic (normal form), leading to appearance of the dual emission (both normal and tautomeric forms, Figure 8A).This dual emission is generally highly sensitive to the environment polarity and H-bonding, because these factors could tune charge transfer of the dyes (and thus acidity and basicity of corresponding groups) and the intramolecular H-bonding within these dyes.A review dedicated to ESIPT dyes coupled to a charge transfer and sensitivity of ESIPT dyes to the environment was published earlier. [13]verall, one should stress that for six-membered Hbonding cycles the dual emission related to inhibited ESIPT is relative rare and not easy to control.A dedicated review on this issue was recently published. [76]In sharp contrast, dyes with 5-membered H-bonded cycles are naturally prone to exhibit dual emission.The prominent examples are 3hydroxyflavones (3HFs) and their analogues, which will be described in the next chapters.

ESIPT Dyes with 5-Membered H-Bonding Cycle
ESIPT in dyes with 5-membered H-bonding cycles was first described by Kasha and co-workers in their seminal work on 3HFs. [77]The four-state model of ESIPT process in 3HF is shown in Figure 9A, which explains appearance of dual emission, belonging to the normal (N*) and tautomer (T*) states.The simplest analogue of 3HF corresponds to dye 1 (Figure 9B), while its analogues without 2-phenyl group (e. g. 2, 5-7, Figure 9B) are 3-hydroxychromones (3HC).For simplicity we will call both these families as 3-hydroxyflavones (3HFs).Their aza-analogues are called 3-hydroxyquinolones (3HQs).3HFs and 3HQs overcome typical limitations of ESIPT systems with 6-membered H-bonded cycles.Here, 5-member ring forms a reaction site for the proton transfer with optimal H-bond distance fixed in a conformationally rigid cycle, so that ESIPT and the dual emission of the dye can be readily tuned by chemical substituents and the environment.Thus, the lateral phenyl unit in such structures creates a spatial shielding of the ESIPT system, reducing the influence of surrounding molecules on it. [78]In addition, electron donating substituents can be readily introduced in the side phenyl ring and the conjugation system of their fluorophore can be lengthened. [79] Thus, in 3HFs and 3HQs, the ESIPT and ICT are strongly coupled, which drives sensing properties of these dyes. [13,79] valuable advantage of 3HFs and their analogues is the synthetic availability of compounds with a tailor-made structure.Their synthesis usually goes through the condensation of arylaldehydes with ortho-hydroxyacetophenones, followed by the transformation of the obtained chalcones into target compounds by oxidative cyclization by Algar-Flynn-Oyamada reaction. [79,81]In case of 3HQs, the aza-analogues of 3HF, several synthesis routes were reported. [82]The synthetic availability of 3HFs and 3HQs made it possible to prepare a broad range of fluorescent dyes and functional probes.

Tuning the Fluorescence Response of 3HF Dyes to Polarity of Molecular Surrounding
Kasha and coworkers were the first to recognize the high potential of 3HFs in the studies of complex systems, including biological ones.They were applied for detection of traces of water in hexane by the increased ratio of the N* and T* band intensities (N*/T), and by the appearance of a third fluorescence band. [83]Their reports on the intermolecular Hbonding in rigid solvent matrix [84] and on binding of 3HFs to proteins with information about the nature of the binding site environment [85] can be considered as the first examples of applications of 3HFs as the fluorescent probes. The latter is accompanied by the red shift of the N* band (Figure 9A), which reflects classical solvatochromic behavior of highly dipolar N* state, especially in substituted 3HFs.
The synthetic availability of 3HF and their analogues made it possible to generate a wide range of dyes with different lengths of chromophores, containing electron donor substituents or linkers, or analyte recognition unit. [79,87]It was shown that increasing the length of the chromophore and the strength of the electron donor group at the 2-aryl group leads to higher sensitivity of the probe to polarity of surroundings, as well as to a shift of the absorption and fluorescence bands to the longwavelength spectral region (Figure 9B-D). [79]The effect was explained by the enhanced excited state charge transfer of the N* state and thus its higher excited state dipole moment. [80]he effect could be further enhanced by the electron acceptor group at 7-position (compound 7, Figure 9B) [88] or reduced by the donor group. [79]Remarkably, increase in the ICT character of the N* state reduced efficiency of ESIPT in the same solvent (Figure 9D), making it energetically less favorable.As a result, more dipolar 3HF dyes with higher ICT character showed two-band response to polarity in more apolar media, corresponding to lipid membranes, micelles or aprotic apolar solvents. [79]In more polar and H-bonding environments, such as aqueous or alcoholic solutions they worked as traditional ICT-probes showing a solvatochromic shift of the N* band.
The uncovered structure-sensitivity relationship in 3HF dyes facilitated the development of probes designed to work in a given environment of the bio-object.It revealed that for the experiments in highly polar cytosol and other aqueous solutions (e. g., with peptides, proteins, nucleic acids and carbohydrates) the fluorescent cores 1-4 should be chosen, while for the studies of lipid membranes -the ones from 4-7, as well as the other red-emitting fluorophores with an increased length of conjugated chain [91] are more appropriate.In the following sections the use of these core fluorophores as building blocks for fluorescent probes in various biological applications will be shown.

Fluorescence Sensing of Polarity, H-Bonding and Local Water Content in Biomolecular Complexes
The problem of measuring the local polarity of the molecular environment by a fluorescent probe remains challenging because both universal dipole-dipole and H-bonding interactions affect emission fluorescent probes.Thus, increase in solvent polarity and H-binding ability of a solvent produces red shift in the emission of classical push-pull dyes, which are difficult to distinguish.To resolve such problem, empirical polarity scales E T and E T (30) [92] were developed, in which the reference dyes reported the total effect of these parameters by shifting the band maximum in the absorption spectrum.Due to the dominant influence of H-bonding, proton-donating solvents are the most polar in these scales.
The problem of separation of the effects of dipole-dipole and H-bonding interactions on the probe response needs application of more advanced multi-parametric probes, as shown for 3HF dye 4 (Figure 10B). [93]In neat aprotic solvents a linear dependence (y = ax + b) for the logarithm of the ratio I N* /I T* of band intensities of dye 4 versus the function of the dielectric constant of the medium (f = (ɛ-1)/(2ɛ + 1)) was first described (Figure 10C).For the neat protic solvents another fit with similar coefficient a, but with higher b, was obtained (Figure 10C), showing additional stabilization of N* form of the dye by H-bonding with a solvent.The logarithm of I N* /I T* corresponds to the logarithm of the ratio of two populations of the fluorescent probe, so it is proportional to the difference in the energy of these states (see Maxwell-Boltzmann statistics) and, similarly to the E T scale, reflects the ability of the medium to solvate the probe. [93]he structure of 3HFs possesses a unique possibility for the lone electron pair of the oxygen atom of the carbonyl group to be spatially shielded from H-bonding with solvent by the installed into neihbour position substituent.87d] Spectral and chromatographic data confirmed that in the BFE molecule, the hydrogen atom of benzene ring and the 3-OH group prevent the formation of intermolecular H-bonds with the C=O group.The oxygen atom of the 3-OH group is also blocked by lateral phenyl from intermolecular H-bonding.

R e v i e w T H E C H E M I C A L R E C O R D
Steady-state fluorescence spectra showed only the sensitivity of BFE probe to dielectric properties of the solution and almost no sensitivity to presence of H-bond donor solvents.For example, the BFE probe in ethyl acetate and 2-methylbutanol-2, the solvents of equal dielectric constant exhibited a similar bands intensity ratio parameter I N* /I T* equal to 0.25 and 0.28, respectively (Figure 10A-B).Meantime, the parent dye 4 has an increased value of intensity ratio I N* /I T* parameter in protic solvent: 1.77 in 2-methylbutanol-2 vs 0.26 in aprotic ethyl acetate.Insensitivity of BFE to H-bonding also takes place in solvent mixtures (Figure 10B).For BFE the spectral parameter Log(I N* /I T* ) shows a linear dependence versus physical polarity function (ɛ-1)/(2ɛ + 1) in all types of solvents, [87d] while for the dye 4 two separate curves for H-bonding and aprotic media are observed (Figure 10C).Thus, BFE offers a unique opportunity to measure physical polarity expressed as dielectric constant function in its molecular environment, including the protic solvents and polymer films. [94]

Applications of 3HF Probes in the Study of Micelles
Micelles are one of the most accessible objects to study by fluorescent probes.Micelle formation usually causes an increase in intensity and anisotropy of the probe fluorescence due to an increase in the local viscosity and a decrease in local polarity and water content in probe surrounding.The sensitivity of 3HFs to polarity and hydration makes it possible to further monitor these parameters in both regular and reverse micelles.In previous studies, the fluorescent properties of several 3HFs were studied in media composed from neutral, cationic and anionic surfactants in regular micelles formation in water, [95] and in reverse micelles composed from sodium bis(2-ethylhexyl) sulfosuccinate (AOT) formation in hexane. [96]harp effects of charge of polar headgroups and their level of hydration on the fluorescence of 3HF probes were registered in all cases.95b] 3.6.3HF Probes for Lipid Membranes

Sensing Hydration and Polarity in Lipid Membranes
Targeting 3HF dye to lipid membrane requires its functionalization with an anchor, containing charged headgroups and alkyl chains.The core dye 4 in this case turned out to be particularly appropriate, allowing attachment of different functional units on both sides of the fluorophore. [97]Further, by the study of the properties of lipid membranes of different composition, a way of measuring the polarity and hydration of the probe localization sites by steady-state fluorescence spectroscopy was elaborated.Such possibility was found after the analysis of the N* band shape, which was deconvoluted into two components, namely on the free N* and H-bound HÀ N* forms of the probe (Figure 11). [98]he distribution between the hydrated and non-hydrated forms of the probes in the lipid bilayer can be further analyzed by calculating the relative integral intensities of HÀ N*, N* and T* bands obtained by deconvolution of the experimental spectra.Since the full width at half maximums of the N* and HÀ N* bands are close to each other and are 2.5 times as wide as that of the T* band, it is possible to calculate the approximate partition of probe F2N8 between hydrated and non-hydrated forms from the following equation:  [98]

R e v i e w T H E C H E M I C A L R E C O R D
where [B] and [F] are the concentrations of the probe in Hbonded and H-bond-free forms, respectively, C is a constant corresponding to the ratio of molar absorption coefficients of the H-bonded and H-bond-free forms of the probe at the excitation wavelength, and Q(F) and Q(B) are their respective fluorescence quantum yields.If we assume that the ratio of quantum yields does not change substantially from one type of lipid vesicles to another, the intensity ratio I H-N* (I N* + 0.4I T* ) could serve as a relative parameter for the probe hydration and, accordingly, the hydration of the site of probe location in bilayer.On the other hand, the I N* /I T* ratio reflects local polarity and electric fields in the membrane, independently of the hydration parameter. [98]This multiparametric approach was successfully used in the study of the properties of lipid membranes, [99] including the study of membrane leaflets properties during the apoptosis and cholesterol depletion, [100] peptide aggregation [101] and lipid peroxidation. [102]

Sensing Transmembrane, Surface and Dipole Electric Potentials
The sensitivity of 3HF probes to electric fields in solutions was studied in detail by the series of dyes, where the positively charged group is located on the opposite sides of the ESIPT system of the molecule (Figure 12A). [103]In this work, it was unequivocally confirmed that the local electric field promotes the proton transfer when it is directed along the ESIPT coordinate, and inhibits ESIPT when the field is directed against the proton transfer.The amplitude of response by the I N* /I T* intensity ratio exceeded one order of magnitude for the two cases of opposite charge locations in the probe molecule.Such a high sensitivity of ESIPT to the electric fields and fast electrochromic nature of the response prompted us to create a fluorescent probe for transmembrane potential V m measurements in cell plasma membranes.In this direction the probe di-SFA was synthesized, in which the extended fluorophore type 5 was used for better sensitivity. [104]luorescence spectra of the new dye in lipid vesicles and cell membranes confirmed the fluorophore location in the hydrophobic region of the membranes (Figure 12B).Variation of V m in lipid vesicles and cell plasma membranes resulted in a change of the intensity ratio of the two emission bands of the probe.The ratiometric response of the dye in cells was � 15 % per 100 mV, that is larger in comparison with most singlefluorophore, fast-response probes reported that time.Com-Figure 12. Fluorescence emission spectra in chloroform and structures of F2N8, 4 and PPZ dyes, which demonstrate an influence of cationic charge on ESIPT reaction (A).Adapted with permission from. [103]Fluorescence emission spectra of di-SFA in phosphatydyl choline lipid vesicles at transmembrane potentials 0 mV and À 118 mV.The spectra are normalized at the peak maxima (B).Plot of the I 527 /I 605 ratio as a function of the potential value (C).Structure of di-SFA probe (D).Adapted with permission from. [104]

R e v i e w T H E C H E M I C A L R E C O R D
bined patch-clamp/fluorescence data further showed that the ratiometric response of the dye in cells was faster than 1 ms.Thus, for the first time the ESIPT reaction has been successfully applied as a sensing principle for detection of transmembrane potential, allowing to couple electrochromic band shifts in excitation/emission spectra with changes in the relative intensities of the two well-separated emission bands.High sensitivity of 3HF probes to electric fields made it possible to develop the probes for sensing the surface [105] and dipole potentials [106] of lipid membranes.

Sensing Lipid Membrane Composition and Phase State
Since all 3HFs are highly sensitive to polarity and hydration of surrounding, a comparison of these parameters was made for the most common phases of the lipid membranes, such as gel, liquid ordered (raft), liquid crystalline, and liquid disordered states.Using 3HF probe F2N8, it was found that the Lo phase exhibited the lowest hydration, [99] which is probably linked to the presence of cholesterol.Liquid crystalline phase was found to be the most hydrated among all studied ones, while presence of cholesterol decreased the hydration parameter.
In fluorescence microscopy of lipid vesicles and living cells it is important to design a probe with high affinity to lipid membranes.To this end, 3HF core 4 was substituted with an anchor containing a dodecyl chain and a zwitterionic group to yield F2N12S (Figure 13).High sensitivity of dual emission of the probe to phase state allowed us to visualize the coexistence of Ld and Lo phases in giant unilamellar vesicles. [107]Then, using F2N12S, the polarity and hydration of plasma membranes of living cells and model lipid vesicles were compared. [100]The hydration parameter value in cell membranes closer to the Lo phase rather than to the Ld phase in model membranes.This suggested that cell plasma membranes exhibit a significant fraction of Lo phase in their outer leaflet.Two-photon fluorescence microscopy experiments showed that cell membranes labeled with F2N12S probe exhibit a homogeneous lipid distribution, that may be a result of low size of the putative domains of Lo phase and their homogeneous distribution all over the membrane and/or high dynamics of Lo-Ld conversion.Cholesterol depletion affected dramatically the dual emission of the probe suggesting the disappearance of the Lo phase in the cell membranes.It was also found that cell apoptosis resulted in a similar loss of Lo phase, which could be attributed to a flip of sphingomyelin from the outer to the inner leaflet of the plasma membrane and to an opposite flip of the anionic phosphatydylglycerol due to apoptosis-driven lipid scrambling. [100]Thus, the increased information content of F2N12S probe allowed monitoring by laser scanning confocal microscopy the degree of apoptosis and spatial distribution of the apoptotic changes at the cell plasma membranes, a feature that can be hardly achieved with the commonly used fluorescently labeled annexin V assay. [32]o further improve the ratiometric probe for apoptosis, we designed three new 3HF analogues with expected vertical orientation in the biomembrane (Figure 13A). [108]They differ by the length of alkyl chains at nitrogen atom from 4 to 8 carbons.The probe F66NS with medium chain length showed the optimal membrane binding properties, while the F86NS with the longest alkyl chains did not efficiently stain the cell membranes, probably due to too low water solubility.The two new probes were found to be more sensitive than F2N12S to differentiate Lo/Ld phases, to detect the surface charge changes in lipid vesicles and to sense the loss of lipid order in cell plasma membranes after cholesterol extraction.The F46NS probe, containing the shortest (butyl) chains was found to be the most sensitive to apoptosis (Figure 13B), while the F66NS probe was the brightest.Important observation was that apoptosis, induced by different agents, led to similar spectroscopic effects to those produced by the loss of lipid order and change in the surface charge, confirming that apoptosis decreases the lipid order and increases the negative surface charge in the outer leaflet of cell membranes.Thus, fluorescent probes based on the 3HF core have shown to be effective in the study of lipid membranes, namely their properties such as transmembrane, surface and dipole electric potentials, as well as polarity and hydration of the regions of probe location.They enabled detection of changes in the composition or phase state of the model and cell membranes.These capabilities result from the fact that these probes, owing to the multi-band emission, provide increased number of information channels as well as exhibit high sensitivity of ESIPT reaction to the environment.

3HF Labels and Amino Acids in Peptide Research
Labeling of proteins by 3HF dyes were initially developed based on coupling with the thiol group of cysteine and the amino group of lysine. [109]This enabled studying the behavior of α-crystallin, a chaperone protein under conditions of varied temperature. [110]The labeling by cysteine then was improved and applied in the study of α-synuclein protein aggregation [111] and its conformational state changes. [112]The N-terminal labeling of peptide by 3HF probe was successfully applied to the detection of proteins in solutions by peptide-protein interaction. [113]N-terminal labeling was especially effective in the studies of peptide conformational state and peptide-nucleic acid interactions because of its experimental simplicity.
The choice of fluorophore for peptide labelling depends on the application.In case of peptide-membrane interactions, one should use dye 4 (Figure 9), which operates in medium polar environment.On the other hand, for studying interactions of peptides with nucleic acids, the dye should operate in the polar ration, which corresponds to dyes 2 and 3 (Figure 9).

Sensing Peptide-Lipid Membrane Interactions
Environment-sensitive labels of peptides are suitable for studying peptide interaction with lipid membranes, because they respond to decrease in the local polarity.The 3HF probes give the most detailed information about peptide surroundings in such experiments, since the analysis of their spectral parameters can differentiate the dipole-dipole interactions from the Hbonding ones, [87d] and to measure the level of hydration of different areas of lipid membrane. [98]o study peptide-lipid interactions, it is suitable to label peptide by N-terminus or to install a fluorescent amino acid into peptide chain.The both operations were performed for various biologically important peptides, as linear or cyclic.For example, to study the membrane binding of melittin, magainin and poly-L-lysine, a simple procedure of peptide labeling at the N-terminus was applied. [114]The binding to lipid membrane was easily controlled by the increase of fluorescence intensity (Figure 14).It was found that melittin and magainin bind to zwitterionic and negatively charged lipid bilayers, while poly-L-lysine binds only to negatively charged lipids.Deconvolution of fluorescence spectra into three components demonstrated that the hydration at the peptide N-terminus was approximately the same for all peptides in all tested membranes, while the polarity differed.The highest value of environment polarity was observed for poly-L-lysine in DOPC/DOPS (8 : 2) lipid vesicles and the lowest one -for melittin in DOPC vesicles.A lower polarity means a deeper penetration of the peptide N-terminus into the lipid bilayer.These data were confirmed by parallax quenching experiments.The orientation of the label in giant unilamellar vesicles was It was evidenced that poly-L-lysine displays an orientation of the label orthogonal to the membrane surface, while melittin displays both parallel and orthogonal orientations. [114]This work represents a good example where a 3HF probe grafted to peptide simultaneously gave several different types of data, about binding, location and orientation of a peptide domain, as well as about polarity and hydration of location site, which would not be possible with other types of probes.113a]

3HF α-L-Amino Acid for Studying Peptide Interaction with Lipids
The insertion of a fluorescent label into the peptide chain imposes strict requirements on its spatial parameters and lipophilicity, and also requires dedicated synthetic procedures in the creation of the α-L center of chirality.This challenge was realized for a 3HF dye [117] by the synthesis of a fluorescent α-L-amino acid AFaa from tyrosine.The amino acid was introduced in two melittin peptide variants by solid phase synthesis instead of leucine-9 (L9) or tryptophan-19 (W19, Figure 15).In contrast to covalent labeling approaches via a flexible linker to N-terminus, here the fluorophore was closely attached to the peptide backbone in desirable position, with a well-defined orientation.This feature enabled a more precise determination of the peptide insertion and orientation in lipid membranes.
Due to high sensitivity of AFaa emission towards the hydration of environment, it was possible to register the kinetics of melittin tetramer dissociation after dilution of the solution.The dissociation was monitored by the fluorescence intensity decrease, and also by the I N* /I T* ratio and shifts in their maxima.Basing on the I N* /I T* ratio, it was concluded that the lowest hydration corresponds to the L9 position of melittin tetramer, while in the monomer, melittin spatial structure in buffer exhibits approximately the same hydration for all studied peptide sites.
The insertion of the labeled peptide into the membrane bilayer of DOPC large unilamellar vesicles resulted in a strong (up to 20-fold) increase in the fluorescence emission intensity and in the appearance of two emission bands, suggesting that the 3HF probes were embedded within the viscous hydrophobic region of the bilayer.The label at position L9 showed the lowest I N* /I T* ratio corresponding to the deepest insertion of the peptide segment into the membrane.The emission spectrum observed in living cells was quite similar to those in DOPC vesicles, suggesting similar insertion of the peptide in these two membrane systems.
Then, fluorescence microscopy experiments with giant vesicles and live cells using a polarized light excitation showed drastic differences in the orientation of AFaa label at L9 and W19 positions.This implies that in cell membranes, as in lipid bilayers, melittin is preferentially oriented parallel to the membrane surface.Thus, the rational probe design and the combined utilization of several fluorescence parameters provided insights into peptide location and orientation in lipid membranes.
Recently the amino acid AFaa was applied in the study of antimicrobial cyclic peptide antibiotic gramicidin S. [118] Label- Adapted with permission from. [117]

R e v i e w T H E C H E M I C A L R E C O R D
ing biomolecules with fluorescent probes is an established tool for biophysical studies.However, it remains underused for small peptides, as there is a serious threat in such case to change affinity and selectivity of the peptide interaction with its target in the object.In a recent work, the fluorescent amino acid 3HCaa was incorporated into a known antimicrobial peptide, cyclo[RRRWFW] (cWFW), in place of three aromatic residues W, F and W sequentially. [119] It was shown that the installation of the probe at the W 1 site did not significantly reduce the antibacterial activity, although it slightly changed the conformation of the peptide in model solutions.The equilibrium between two conformations of the peptide in lipid membranes, micelles, trifluoroethanol and buffer solution was suggested by the data on the position and intensity of bands in the fluorescence and absorption spectra.
Comparing the results obtained with other probes on cWFW peptide modification, 3HCaa can be considered an advantageous label for studying small peptides.It provided insights into conformational equilibria of the labeled peptides localized in cells, and the level of polarity of the local environment around the label by confocal fluorescence microscopy.It was found that the labeled peptides efficiently penetrated cancerous cells and localized mainly in nonpolar subcellular compartments.In the zebrafish embryo, the peptides remained in the bloodstream upon injection into the cardinal vein, presumably adhering to lipoproteins and/or microvesicles.They did not diffuse into any tissue to a significant extent during the first 3 h after administration. [119]his study demonstrated the advantages of fluorescent labeling by multi-channel probes to elucidate the sites and mechanism of action of peptides in organism.

Labeling of Peptides to Study Peptide-Nucleic Acid Interactions
On the first steps, the covalent conjugates of 3HC type 2 (Figure 9B) with polycationic spermine were synthesized and studied. [120]After validating the capacity of 3HC type 2 probe to distinguish single-and double-stranded DNA, its analogue probe FC (Figure 16) was applied to the study of peptide interaction with DNAs. [121]The peptide sequence corresponding to the zinc finger domain of the HIV-1 nucleocapsid protein (NC) was labeled at N-terminus by FC probe.On interaction with target oligodeoxynucleotides (ODNs), the labeled peptide showed strong changes in the I N* /I T* intensity ratio, indicating an enhanced screening of the FC fluorophore from the bulk water by the ODN bases.This two-color response depended on the ODN sequence and correlated with the 3D structure of the corresponding complexes, suggesting that the FC label monitors the peptide-ODN interactions sitespecifically.By measuring the two-color ratio, the peptide-ODN-binding parameters were determined, and multiple binding sites in ODNs were distinguished, [121] which is rather difficult to do using other fluorescence methods.This method was found to be more sensitive than the commonly used methods of steady-state fluorescence anisotropy, especially in the case of small ODNs.Structures of 3HFs for labling the peptide N-terminus.A) N*/T* intensity ratios for free FC, F6A, F6C and F4O labels (red) and for their conjugates with Tat(44-61) peptide in the absence (green) or in the presence of poly-dT (blue) or CT-DNA (cyan).B) Fluorescence ratiometric response of F4O-labeled Tat(44-61) peptide on interaction with various oligonucleotides.Concentration of peptide was 0.5 μM in 10 mM phosphate buffer pH 7.0 and 30 mM NaCl.The ODNs were added at a ratio of 10 or 40 bases or base pairs per peptide molecule.In all cases, the full complexation of the peptide by the ODNs was achieved, as no change in the N*/T* ratio was observed after a two-fold increase in the concentration of ODNs.Adapted with permission from. [122]he level of probe hydration was related to its ability to stack with DNA bases.Owing to this consideration, the best resolution was obtained with F4O label, presenting higher hydrophobicity and the most compact geometry. [122]Thus, using the I N* /I T* ratio of the F4O label, it was possible to distinguish the interaction of the peptide with four types of DNAs, namely small single-stranded (ss) ODNs, ss DNAs, partially double-stranded (ds) DNAs and fully ds DNAs (Figure 16B).As a consequence of the high sensitivity of the F4O label to the nature of the bound nucleic acid, it was possible to compare the relative affinities of this peptide to ss and ds DNAs.Thus, this new generation of hydration-sensitive probes appears as a highly sensitive ratiometric tool to siteselectively monitor the binding of peptides to different types of nucleic acids.
Based on fluorophore FC (type 2, Figure 9), we synthesized α-L-amino-acids starting from L-tyrosine, e. g., 3HCaa probe, which was the first reported ESIPT-based amino acid analogue at that time. [123]3HCaa was introduced by solid-phase synthesis either in position A30 or W37 of the 11-55 fragment of nucleocapsid protein (NC peptide), instead of alanine and tryptophan residues, respectively (Figure 15A).Then, both peptide variants were tested in binding experiments with the ODN sequences known to have increased affinities for NC (11-55) peptide, namely SL2, SL3 and ΔP(À )PBS.
The 3HCaa label gave very different responses on binding of the peptide with ODNs, depending whether it is located in A30 or W37 position.The binding of A30 variant with SL2 and SL3 RNAs resulted only by the T* band red shift, while its binding with ΔP(À )PBS sequence additionally resulted in the decrease of the intensity ratio I N* /I T* , showing a stronger contact of A30 site of peptide with ΔP(À )PBS.In the case of W37 variant, on interaction with ODNs SL2, SL3 and ΔP(À )PBS, a strong change in the fluorescence spectrum was observed, with notably a large drop in the ratio I N* /I T* , which can be assigned to a decrease in the local hydration of the probe environment caused by the close distance to RNA segment.
These results can be rationalized by a larger distance between A30 residue and the nucleic bases within the complexes and consequently a higher exposition to water.These data correlate well with the NMR data for the NC (11-55) peptide and for the whole NC sequence (Figure 17B,C).In both cases, the water access W A parameter was calculated from the calibration procedure [124] for the interaction sites of A30 and W37 residues with ODNs.W A parameter drops to 0.32-0.41for W37 variant, but only to 0.49-0.54for the A30 variant, to be compared to 0.57 for the free labels.Thus, a fluorescent L-amino acid based on a 3HC probe, incorporated at different positions into a peptide, enabled both quantitative characterization of peptide interactions with nucleic acid targets and determine the level of water access at the specific peptide sites.
In the following work, the new α-L-amino acid bearing the 4'-methoxy-3-hydroxyflavone fluorophore (M3HFaa) was synthesized, incorporated into NC(11-55) peptide at the same A and W sites and studied in an experiment on binding with oligodeoxynucleotides (ODNs) and lipid vesicles (LUVs). [125]he dual emission of M3HFaa was found to be substantially more sensitive to hydration as compared to 3HCaa (Figure 18).It was found that M3HFaa preserves the peptide structure and functions.Binding of the labeled peptides with nucleic acids and lipid vesicles produced a strong switch in their dual emission, explained as the strong decrease of local hydration.This switch was also associated with the appearance of long-lived fluorescence of the T* form, as a consequence of the rigid environment in the complexes that restricted the relative motions of the M3HFaa aromatic moieties.The strongest restriction and thus the longest fluorescence lifetimes were observed at position 37 in complexes with nucleic acids, where the probe likely stacks with the nucleobases.Based on the dependence of the lifetime values on the nature of the ligand and the labeled position, two-photon fluorescence lifetime imaging was used to identify the binding partners of the labelled peptides microinjected into cells.Thus, M3HFaa appears as the highly sensitive tool for monitoring siteselectively peptide interactions in solution and living cells.

Quantification of Local Water Concentration of Probe Environment in Peptide-Peptide and Peptide-Nucleic Acid Complexes
All biomolecules function in a hydrated environment, where the concentration of water varies from mM (middle of lipid membranes), up to 25 M (polar heads of lipids and proteins) and further, up to 55.56 M, in pure water.In this connection, H-bond sensitive probes allow measuring the actual concentration of water in a certain site of interaction of biomolecules, the active center of an enzyme, or a certain site of proteinlipid, peptide-protein, protein-DNA complexes, etc.In contrast to lipid membranes, where deconvolution of the N* band into two components was applied to extract hydration parameter (see above), such measurements in highly hydrated biomolecules was addressed in simpler way using 3HF dyes with weak ICT character (type 2 and 3, Figure 9B).Thus, the probe F4O and its analogues with lower sensitivity to dipole-dipole interactions and better separation of fluorescence bands of the N* and T* forms were applied there (Figure 19A). [124]It was shown that the I N* /I T* ratio of the probe is linearly dependent on the molar water concentration in an organic solvent, in a range 10-55 M of water (Figure 19B).The mechanism of probe response resulted from the interaction of the atoms involved in ESIPT with water molecules. [124]Since H-bonding occurs via the 4-carbonyl group of 3HF, water weakens the intramolecular H-bond and thus slows down or even blocks ESIPT.
Local concentration of water in the label surrounding was determined for the label alone, in the labeled peptide alone and for labeled peptide complexes with single-stranded and double-stranded ODNs.The level of hydration was then converted into the water access coefficient W A , a more convenient unit on the molecular scale:  (11-55) peptides labelled at position 37 by M3HFaa or 3HCaa.Adapted with permission from. [125]gure 19.A) Fluorescence spectra of F4O probe in water-dioxane binary mixture of varying compositions.Probe concentration is 3 μM.B) Dependence of Log(I N* /I T* ) versus molar concentration of hydrogen bond donor in water-solvent mixtures for FC probe and linear fits for all probes.C) Dependence of the water access on the nature of peptide and N-terminal amino acid.The W A coefficients are given for FC-labeled (Gly) 5 , peptide (4-17), peptide (7-17), peptide (9-17), Tat (44-61) and NC peptide.The nature of the N-terminal amino acid is indicated over the bars.Adapted with permission from. [124]

R e v i e w T H E C H E M I C A L R E C O R D
where 55.5 corresponds to the molar concentration of neat water.W A coefficient appears more adequate for describing hydration level at nanometric scale.The method was applied for measuring hydration at the N-terminus of labeled peptides (Figure 19C) and their complexes with oligodeoxynucleotides (ODNs), more particularly to labeled-Tat(44-61) peptide.It was found that water access mainly depends on the structure of the neighboring amino acid residue, but not on the length of peptide, since shorter peptides with a hydrophobic residue at the N-terminus gave lower hydration of the label (compare the data for 4-17 and 9-17 peptides, where the lipophilic phenylalanine strongly decreases the W A value).Besides the N-terminal, other amino acid residues had lower influence on the probe hydration.Interestingly, the W A value for the F4O label in the peptide-DNA complex is almost zero, suggesting that in this case the probe has no access to water, which could be explained by a deeper intercalation of F4O molecule into the base pairs. [124]hus, the elaborated method of quantitative evaluation of the hydration at the peptide N-terminus gives better insights about the structure and interactions of biopolymers.

3HF as Fluorescent Nucleic Acid Analogues
The first synthesis of fluorescent nucleosides bearing 3HC moiety was reported in 2009. [126]Then, an improved analogue compatible with solid-phase DNA synthesis was synthesized, incorporated into oligonucleotides and further used as twocolor probe for sensing DNA interactions. [127]Later work on 3HF nucleosides enabled probing DNA/DNA from DNA/ RNA duplexes, [128] which can be considered as an successful application of the hydration evaluation method to nucleic acids.In the mentioned work, the fluorescent deoxyribonucleotide with uridine-3-hydroxychromone base was synthesized and inserted into the series of ODNs.Since the fluorescent base almost did not disturb the spatial structure of DNA/DNA and DNA/RNA duplexes, for the first time the local water concentration in different sites around the complexes was estimated.Based on the I N* /I T* ratio and on the position of T* band, the method provided the highest sensitivity and allowed discriminating between matched and mismatched dsDNA, as well as between B-and A-forms of DNA/DNA and DNA/ RNA duplexes.Detailed review on design and application of nucleotides presenting multi-color fluorescence was published earlier. [129]

Operation in Aqueous Media. High-Selectivity Detection of Adenosine Triphosphate (ATP) Anion
Selective detection of ATP anion in different buffer solutions and in mitochondria was achieved by applying the 4'-dimethylamino-3-hydroxyflavone, FME probe (Figure 20). [130]The interaction with ATP in an aqueous medium induced a new band in the fluorescence excitation spectrum of FME, showing a large red shift.Such a shift was assigned to the appearance of anionic form of FME in a complex with ATP, where FME was stabilized by the electric field of the tetra-charged ATP anion.Previously, FME anion formation was reported at neutral pH in presence of N-(2-hydroxyethyl)piperasine-N'-ethylsulfonate (HEPES) buffer when labelling anionic egg yolk phosphatidylglycerol (EYPG) vesicles, which also present strongly negatively charged interface. [131]Here, the fluorescence intensity increased with ATP concentration (Figure 20B).Most surprising was the unusually high selectivity of FME probe in respect to ATP as compared to the other nucleotides (Figure 20B).The other nucleotide anions also induce some increase of FME fluorescence intensity, but in a much lower extent than ATP anion and without formation of the red-shifted band.Using the spectroscopic response to ATP, FME was used successfully to monitor the succinate-induced production of endogenous ATP in mitochondria.130b] As a consequence, the probe FME allows the selective detection of ATP anion and thus can be considered as a starting molecule to design even more efficient ATP sensors.
In the further studies, an uracil-linked 3-hydroxyflavone [132] and 33 probes from 3-hydroxyflavone and 3-hydroxyquinolone series have been shown their good multi-band response in fluorescence excitation and emission modes to the presence of ATP. [133]It was found that 30 compounds from the set of 33 gave a new band in the spectrum in the presence of ATP in the concentration range from 2 • 10 À 6 to 5 • 10 À 2 mol L À 1 .The increased range of ATP detection was explained by the formation of two consecutive probe-ATP complexes with stoichiometry 1 : 1 and 1 : 2. This work has shown that the 3HF core is the scaffold that binds to ATP within its physiological concentrations, and the substituents only partially change the affinity of the probe to ATP.

Bioorthogonal Chemical and Enzymatic Reactions in Detection of Small Molecules and Enzymes
There is a significant number of works, where high-selectivity chemical or enzymatic reactions are used for fluorescence detection by 3HF probes of the analytes, such as metal cations, thiols, nitroxides, peroxides, etc., or enzymes and live bacteria.In this case, the molecular recognition results in restoration or disruption of ESIPT in 3HF, which serve as an analytical signal for the analyte detection.This approach has gained widespread applications due to numerous aspects of its effectiveness, such as universality, relative simplicity of probe design and synthesis, and ease of its use.In such cases the brightest ESIPT fluorophores are strongly preferred, [134] where OH or NH groups chemically modified with analyte-sensitive groups. [71,135]In Table 1 the structures of probes and their corresponding analytes are listed.

Aza-and Thia-Analogues of 3HFs. Sensing the H-Bond Accepting Ability of Solutions
Currently known isosteres of 3-hydroxyflavones include a large family of 3-hydroxyquinolones (3HQs, Figure 21A) and some examples of thiachromones.The latters were synthesized recently and still have not found practical applications because of low synthetic availability, low fluorescence quantum yields and low stability in presence of oxygen. [155]In contrary, 3HQs found applications as fluorescent probes in biochemistry and pharmacology [156] or in other fields, e. g. in detection of metal ions, [157] ATP anion, [133] viscosity, [158] polarity [159] or H-bond accepting ability of a medium. [160]Let us discuss the last mentioned unique property of such fluorescent probes.
Dielectric properties (polarity), H-bond donating and Hbond accepting ability are the driving forces that affect the dyes in biosystems at neutral pH.160a-b] What is important, the sensitivity to polarity and H-bond donating ability in 3HQs were suppressed because of their very poor ICT character, confirmed by quantum chemistry calculations.Among 3HQ dyes, 2-benzofuryl derivatives were found the most promising for H-bond basicity sensing (Figure 21).It was found that the intensity ratio I N* /I T* of the probe fluorescence was linearly dependent on the Abraham's basicity parameter β of a solvent, which is a quantitative index of the H-bond accepting ability (Figure 21C).160a]

R e v i e w T H E C H E M I C A L R E C O R D
160b] This probe opens a possibility to explore the importance H-bond basicity in biological systems.

Conclusions and Perspectives
Fluorescent probes for sensing fundamental properties of biomolecular environment, such as polarity and hydration, help to study assembly of lipids into biomembranes, sensing  * Additional data on 3HF probes and their analytes, see review. [135]** Bearing 6-Methyl-N-morpholyl substituent.Appropriate design of the ligand ensures targeting these dyes to plasma membranes or intracellular membranes of specific organelles.Importantly, response of these to local polarity helps to understand lipid organization and formation of lipid microdomains, as well as they shed light on biologically important phenomena, such as apoptosis, oxidative and mechanical stress, etc.So, far Nile Red fluorophore remains among the most successful solvatochromic probes for these applications, but we expect that rapidly emerging new solvatochromic probes with improved characteristics will open new possibilities in polarity-based sensing in cells.Another promising application is labelling of biomolecules, which concerns both proteins and nucleic acids.Here, it is important to preserve strong emission of these dyes in polar environments of these biomolecules, which is one of the challenges in the design of solvatochromic dyes.Overall, the optical properties of solvatochromic ICT dyes should be further improved.In particular, it concerns relatively low extinction coefficient and photostability of these dyes.In this respect, the examples of photostable push-pull pyrene and fluorene and strongly absorbing push-pull dioxaborine (see above) provide some insights on where the molecular design should be oriented.

R e v i e w T H E C H E M I C A L R E C O R D
On the other hand, ESIPT dyes offer an alternative way to sense environment, where dyes with 5-membered intramolecular H-bonds are particularly suitable, with prominent example of 3-hydroxyflavones.3-Hydroxyflavones as fluorescent dyes do not compete in brightness with classical dyes, because they have 2-4 times lower molar extinction coefficients and ~2 times lower fluorescence quantum yields.However, the increasing sensitivity of modern photodetectors puts them on the first positions due to high information content of their emission.In contrast to classical fluorophores and push-pull dyes, 3HF dyes exhibit dual emission sensitive to the nature of their environment, namely its polarity, hydration, viscosity, as well as ionic analytes and enzymes.The dual emission enables extraction of parameters such as maxima of the two emission bands and their intensity ratio, thus proving additional spectroscopic parameters for the analysis of the environment.Undoubtedly, greater information content obtained in the spectral data increases its value, because it enables monitoring multiple characteristics of the molecular surrounding.
In the design of fluorescent probes, an important feature of 3HF is their relatively easy synthetic availability.The latter made it possible to expand the series of fluorophores with brighter fluorescence and to establish regularities in the structure-fluorescence relationship, and thus to achieve desired optical properties and plan the areas of application.In particular, extension of the fluorophore and increase in the dipolar nature, shifts the emission of the dyes to the red, increases their brightness and sensitivity to polarity and inhibits ESIPT reaction.The latter enables preparation of ESIPT probes operating in different ranges of solvent polarity.Two main principles should be highlighted to achieve response of an ESIPT probe.In the first one, the analyte inhibits the ESIPT reaction, which is realized by H-bond donors, H-bond acceptors or some ionic species.In the second one, the ESIPT system is formed under the conditions of a chemical or biochemical reaction with the analyte.
The fundamental properties of solutions, which are the most relevant to biosystems, turned out to be the most difficult to study by fluorescent methods.These are the dipole-dipole interactions (polarity), the strength and concentration of Hbond donors and H-bond acceptors in a given location of a microheterogeneous system, including biosystems.ESIPT probes based on 3HFs and their structural analogues provided a unique solution to address these problems.In particular, 3HF dye enabled simultaneous characterization of the levels of polarity and hydration in specific sites of biological systems.
Therefore, the examples of application of 3HF probes are vast.3HF derivatives bearing amphiphilic anchor groups enabled probing lipid composition, organization and electric fields in model and biological membranes.3HF labels and amino acid analogues allowed monitoring interactions of peptides with lipid membranes, proteins and nucleic acids.Fluorescent nucleic acid analogues provided information about DNA/RNA transformations and interactions.3HF dyes can also detect small analytes such as ATP.Finally, the derivatives with blocked ESIPT yielded a bunch of probes for sensing variety of analytes, such as metal ions, reactive oxygen species, enzymes, etc.Nevertheless, it is also worth paying attention to problems in the application of ESIPT probes.In particular, multi-band fluorescence requires dedicated spectral analysis, such as band deconvolution, and sometimes additional control experiments.The development of new probes with better separation of the fluorescence bands also contributes to addressing this issue.
The so far synthesized ESIPT fluorophores and corresponding fluorescent probes indicate the directions and limits of their application.Here, for sensing lipid membranes, the most convenient are 4'-dialkylamino-derivatives of type 4 (Figure 9), in particular -the probes F2N12S and F46NS

R e v i e w T H E C H E M I C A L R E C O R D
(Figure 13).MFL or AFaa are more suitable for monitoring interactions of peptides/proteins with lipid membranes.These dyes represent an optimal balance between ESIPT and ICT effects, which ensures response in the medium polar environments.For measurements of electric fields in nonpolar lipid media, it is worth using probes based on fluorophores type 5-7 (Figure 9).As for aqueous environments, when studying interactions like peptide-peptide, peptide-protein, proteinprotein, peptide-DNA/RNA, fluorophores 2 and 3 (Figure 9) are the most effective.Their weak ICT character ensures sensitivity to polarity in the rather polar environments.In case of nucleic acids sensing, compact derivatives of 2 (Figure 9) are the most effective.As for the detection of nucleoside triphosphates, multiple 3HF dyes could be suitable, but the remaining challenges is to increase the sensitivity and selectivity to the specific targets and compatibility with cellular imaging.
Since proton is the lightest ion and its transfer leads to striking changes in electronic structure of dyes, the phenomenon of ESIPT is among the most attractive for creation of new fluorescent probes.Primary challenge here will be improving optical properties of ESIPT dyes.Here, we expect the following main directions.First is further shifting the excitation and emission wavelength to the red.Vast majoring of ESIPT dyes absorb in UV/violet region, which remains their key limitation.Moreover, 3HF dyes and ESIPT dyes in general exhibit limited photostability, so the efforts to improve it will be highly welcome.Enhancing separation between the N* and T* emission bands is another important challenge, which could improve probe sensitivity and simplify the data analysis.A logical step to increase the information content in the environment sensing is design of dyes with double ESIPT. [161]Further progress in this direction requires improved efforts in organic synthesis, control of molecular symmetry and electronic interactions in these di-fluorophore systems.Another direction is the combination of ESIPT with ICT, which provides additional tuning of the ESIPT processes and ensures multiparametric response of the probe to the environment.The probes that couple chemical reaction, such as nucleophilic addition, substitution, or redox reactions, with ESIPT, also gain significant interest, as depicted in Table 1.As this approach is quite generic, we expect rapid development of multiple reactive probes for sensing enzymes, metal ions, ROS, etc.However, a special effort should be made to address reaction kinetics and sensitivity of this method.
Overall, fluorescent probes based on ICT and ESIPT provide increased information content due to emission band shifts and/or additional emission bands.The appropriate molecular design of these probes provides researchers with unique chemical tools to study multiple characteristics of the biomolecular environment in solutions and live cells.This opens the way to new discoveries at the interface of chemistry, biology and photonics.

Andrey
Klymchenko was born in Kherson, Ukraine.He received his Ph.D. degree in 2003 from Kyiv Taras Shevchenko University.After postdoctoral work at the University of Strasbourg and the Catholic University of Leuven, he joined CNRS at the University of Strasbourg in 2006.He got promoted to CNRS Research Director in 2014, and he was an ERC Consolidator fellow in 2015-2020.He is a leader of "Nanochemistry and Bioimaging" group and co-founder of BrightSens Diagnostics SAS.His research interests include functional fluorescent molecules and nanomaterials for biosensing, imaging and theranostics.

Figure 1 .
Figure 1.Functional elements of fluorescent probes.A) Indicator Fluo-3 used to determine Ca 2 + concentration in cells.B) Example of a membrane interface calcium indicator.

Figure 2 .
Figure 2. Different types of dyes.(A) Examples of classical dyes.(B) Classical dyes with environment-sensitivity. (C) Environment-sensitive and AIE dyes.

Figure 4 .
Figure 4. Examples of push-pull solvatochromic membrane probes.(B) Principle of their color response to lipid organization and fluorescence spectra of NR12A in an aqueous buffer and in liposomes of different lipid compositions, presenting Ld (DOPC) and Lo (Sm/Chol) phases.(C) Ratiometric confocal microscopy of giant vesicles composed of DOPC/SM/Chol, 1/1/0.7,stained with NR12A.(GD) 3D PAINT superresolution images of COS-7 cells with NR4A.(E) Superresolution mapping of lipid order in cell membranes with NR4A.Color represents the single-molecule spectral mean (color scale bar).Adapted with permission from.[42] A L R E C O R D Chem.Rec.2024, 24, e202300321 (8 of 29) © 2023 The Authors.The Chemical Record published by The Chemical Society of Japan and Wiley-VCH GmbH

Figure 7 .
Figure 7. Halo-Tag solvatochromic probe based on Nile Red (NR12-Halo) for probing nanoscale environment of proteins in live cells.Polarity imaging of KB cells expressing HaloTagged proteins at various localizations by ratiometric confocal microscopy.Scale bars: 20 μm.White arrows point at inner cell structures.Adapted with permission from.[63] A L R E C O R D Chem.Rec.2024, 24, e202300321 (9 of 29) © 2023 The Authors.The Chemical Record published by The Chemical Society of Japan and Wiley-VCH GmbH

Figure 8 .
Figure 8. Example of ESIPT dyes with six members cycle forming intramolecular H-bond.

Figure 9 .
Figure 9. (A) Scheme of ESIPT in a 3HF dye and typical two-color response of dye 4 to polarity changes.(B) A series of 3HFs synthesized, which are arranged in order of increasing sensitivity to the polarity of liquid medium.For properties of 1 and 2, see Ref [89] 3 -Ref.[90] 4 -Ref.[80] 5 and 6 -Ref.[79] 7 -Ref.[88] (C) Absorption and (D) emission spectra of some 3HF dyes in toluene.

Figure 10 .
Figure 10.Structure of BFE dye showing no abilities to H-bonding with protic solvent of their atoms belonging to fluorophore unit.A) Fluorescence spectra of 4 and BFE dyes in ethyl acetate (solid lines) and 2-methylbutanol-2 (dash-dot lines), the solvents of an equal dielectric constant.B) Logarithm of fluorescence intensity ratio I N* /I T* as a function of volume content of 2-methylbutanol-2 in ethyl acetate for FE and BFE dyes.C) Logarithm of the fluorescence intensity ratio I N* /I T* versus solvent polarity function f(ɛ) for 4 (*) and BFE (~) dyes.Adapted with permission from. [87d]

Figure 11 .
Figure 11.H-bonded and H-bond free forms of F2N8 and their estimated location in a PC lipid bilayer using Parallax quenching method.The location of nitroxide paramagnetic quenchers is shown as five or six-membered rings on the left side.(B) Fluorescence spectrum of F2N8 probe in EYPC vesicles and their deconvolution into the N*, HÀ N* and T* bands.Adapted with permission from.[98]

Figure 13 .
Figure 13.(A) Structure of probes and their hypothetic location with respect to lipids in membranes.(B) Two channels to control the object: fluorescence intensity ratio (top) and integral fluorescence intensity per cell (bottom) of 3HF probes in two populations of actinomycin D-treated HeLa cells corresponding to normal and apoptotic cells based on flow cytometry data.Data from.[108]

Figure 14 .
Figure 14.A) Fluorescence modifications on binding of MFL-labeled peptides to DOPC vesicles.A): Emission spectra of the peptides in absence (dash) and in presence of LUVs composed of neutral DOPC lipids.B): Dependence of poly-L-lysine/LUV binding versus the percentage of negatively charged DOPS lipid in DOPC-DOPS vesicles.Structure of MFL label and amino acid sequences of studied peptides are presented on the right side of the figure.Concentration of peptide and lipids was 0.3 and 100 μM, respectively.Adapted with permission from.[114]

Figure 15 .
Figure 15.Spatial structure of melittin showing the orientations of AFaa fluorophore depending on the positions of substituted aminoacids.Five unstructured residues at basic C-terminus are not shown.Middle: the structure of AFaa amino acid fmoc-derivative applied in the peptide synthesis.Bottom: Sequence and labeling scheme of melittin.The alternative directions of the transition dipole moments of the fluorophore are shown by arrows.Cationic amino acids of melittin are marked in gray.Adapted with permission from.[117]

Figure 16 .
Figure16.Structures of 3HFs for labling the peptide N-terminus.A) N*/T* intensity ratios for free FC, F6A, F6C and F4O labels (red) and for their conjugates with Tat(44-61) peptide in the absence (green) or in the presence of poly-dT (blue) or CT-DNA (cyan).B) Fluorescence ratiometric response of F4O-labeled Tat(44-61) peptide on interaction with various oligonucleotides.Concentration of peptide was 0.5 μM in 10 mM phosphate buffer pH 7.0 and 30 mM NaCl.The ODNs were added at a ratio of 10 or 40 bases or base pairs per peptide molecule.In all cases, the full complexation of the peptide by the ODNs was achieved, as no change in the N*/T* ratio was observed after a two-fold increase in the concentration of ODNs.Adapted with permission from.[122]

Figure 18 .
Figure 18.Fluorescence lifetime imaging data of solutions of W37-M3HFaa and A30-M3HFaa, either as free peptides or in complexes with ODNs or LUVs Comparison of N*/T* ratio (B) and fluorescence quantum yield (C) values for the NC(11-55) peptides labelled at position 37 by M3HFaa or 3HCaa.Adapted with permission from.[125]

Figure 20 .
Figure 20.Fluorescence excitation (A) spectra of FME probe in aqueous solution of sucrose and HEPES-carbonate buffer with increasing ATP concentration.Emission wavelength: 554 nm for all excitation spectra.Concentrations are: 250 mM (sucrose), 10 mM (HEPES), 40 mM (carbonate); pH 7.4.(B) Fluorescence excitation spectra of FME probe in the presence of different nucleoside phosphates.Adapted with permission from.[130b]

Figure 21 .
Figure 21.A) Structures of dyes 1c and 2c.B) Normalized fluorescence emission spectra of 1c in aprotic solvents.Abbreviations: DMSO -dimethylsulfoxide, HMPA -hexamethylphosphotriamide, DMF -dimethylformamide, NMP -N-methylpyrrolidone, TMU -tetramethylurea, Ac -acetone, An -acetonitrile, Bb -bromobenzene.C) Dependence of log(I N* /I T* ) on the Abraham's hydrogen bond acceptor basicity parameter for the probe 2c.Numbering of solvents: 1toluene, 2 -ethyl acetate, 3 -dichloromethane, 4 -acetonitrile, 5 -methanol, 6 -ethanol, 7 -n-butanol, 8 -dioxane, 9 -THF, 10 -DMF and 11 -DMSO.Adapted with permission from.[160a] interactions of biomolecules and imaging physiological state of the cells.Here, we summarize major efforts in the development of probes based two photophysical mechanisms.The first one is based on excited-state intramolecular charge transfer (ICT).It is represented by push-pull dyes containing electron donor and acceptor group connected by a pi-conjugated system.They are characterized by fluorescence solvatochromism, where their emission band shifts to the red with increase in the environment polarity and hydration.the most common examples of solvatochromic fluorescent dyes and their spectroscopic properties are discussed.Compared to well-established push-pull naphthalene

Table 1 .
Fluorescent detection of analytes by chemical or enzymatic transformation of 3HF probe*.