Catalytic and selective red light-triggered photodegradation of a G-quadruplex DNA by a zinc (II) phthalocyanine.

Water-soluble phthalocyanine (Pc) derivatives have been regarded as potential G-quadruplex (G4) nucleic acid-targeting ligands for anticancer therapy and have been extensively studied as effective photosensitizers for photodynamic therapy (PDT). Understanding how photosensitizers interact with nucleic acids and the subsequent photolytic reactions is essential for deciphering the initial steps of PDT, thereby aiding in the development of new photosensitizing agents. In this study, we found that red-light irradiation of a mixture of a Zn(II) Pc derivative and an all-parallel G4 DNA leads to catalytic and selective photodegradation of the DNA by reactive oxygen species (ROS) generated from the Zn(II) Pc derivative bound to DNA through a reaction mechanism similar to that of an enzyme reaction. This finding provides a novel insight into the molecular design of a photosensitizer to enhance its PDT efficacy.

Hydrogen Evolution of a Unique DNAzyme Composed of Cobalt-Protoporphyrin IX and G-Quadruplex DNA.

Molecular hydrogen (H2 ) is a clean and renewable fuel that has garnered significant interest in the search for alternatives to fossil fuels. Here, we constructed an artificial DNAzyme composed of cobalt-protoporphyrin IX (CoPP) and G-quadruplex DNA, possessing a unique H2 Oint ligand between the CoPP and G-quartet planes. We show for the first time that CoPP-DNAzyme catalyzes photo-induced H2 production under anaerobic conditions with a turnover number (TON) of 1229 ± 51 over 12 h at pH 6.05 and 10 °C. Compared with free-CoPP, complexation with G-quadruplex DNA resulted in a 4.7-fold increase in H2 production activity. The TON of the CoPP-DNAzyme revealed an optimal acid-base equilibrium with a pKa value of 7.60 ± 0.05, apparently originating from the equilibrium between Co(III)-H- and Co(I) states. Our results demonstrate that the H2 Oint ligand can augment and modulate the intrinsic catalytic activity of H2 production catalysts. These systems pave the way to using DNAzymes for H2 evolution in the direct conversion of solar energy to H2 from water.

Photosensitized Generation of Singlet Oxygen (1O2) from Pyropheophorbide a Specifically Bound to All Parallel-Stranded G-Quadruplex DNAs.

Pyropheophorbide a (Pyro-a), a chlorophyll metabolite, is a potential photosensitizer for photodynamic therapy (PDT), but little is known about its interaction with the target molecules for PDT, e.g., nucleic acids. Elucidation of the interaction between photosensitizers and nucleic acids will help us understand the initial process of PDT at the molecular level and hence to develop photosensitizing agents. We found that pyro-a forms a 1:1 complex with an all parallel-stranded G-quadruplex DNA and that pyro-a in the complex exhibits a quantum yield for singlet oxygen generation, with excitation at 664 nm, higher by a factor of ∼10 than that of pyro-a in an aqueous solution. These findings provided novel insights into molecular design of pyro-a-based photosensitizers to enhance their PDT efficacy.

A Photochemical Study of Photo-Induced Electron Transfer from DNAs to a Cationic Phthalocyanine Derivative.

Water-soluble cationic gallium(III)-Pc complex (GaPc) is capable of photogenerating ROSs but does not exhibit photocytotoxicity in vivo. GaPc binds selectively, through a π-π stacking interaction, to the 5'-terminal G-quartet of a G-quadruplex DNA. The photo-excited state of GaPc of the complex is effectively quenched through electron transfer (ET) from the ground state of DNA guanine (G) bases to the photo-excited state of GaPc (ET(G-GaPc)). Hence the loss of the photocytotoxicity of GaPc in vivo is most likely to be due to the effective quenching of its photo-excited state through ET(G-GaPc). In this study, we investigated the photochemical properties of GaPc in the presence of duplex DNAs formed from a series of sequences to elucidate the nature of ET(G-GaPc). We found that ET(G-GaPc) is allowed in electrostatic complexes between GaPc and G-containing duplex DNAs and that the rate of ET(G-GaPc) (kET(G-GaPc)) can be reasonably interpreted in terms of the distance between Pc moiety of GaPc and DNA G base in the complex. We also found that the quantum yields of singlet oxygen (1O2) generation (ΦΔs) determined for the GaPc-duplex DNA complexes were similar to the value reported for free GaPc (Fujishiro R, Sonoyama H, Ide Y, et al (2019) J Inorg Biochem 192:7-16), indicating that ET(G-GaPc) in the complex is rather limited. These results clearly demonstrated that photocytotoxicity of GaPc is crucially affected by ET(G-GaPc). Thus elucidation of interaction of a photosensitizer with biomolecules, i.e., an initial process in PDT, would be helpful to understand its subsequent photochemical processes.

Nature of a H2O Molecule Confined in the Hydrophobic Interface between the Heme and G-Quartet Planes in a Heme-DNA Complex.

Heme binds selectively to the 3'-terminal G-quartet of all parallel G-quadruplex DNAs to form stable heme-DNA complexes. Interestingly, the heme-DNA complexes exhibit various spectroscopic and functional properties similar to those of hemoproteins. Since the nature of the axial ligands is crucial in determining the physicochemical properties of heme, identification and characterization of the axial ligands in a heme-DNA complex are essential to elucidate the structure-function relationship in the complex. NMR studies of a complex possessing a low-spin ferric heme with a water molecule (H2O) and cyanide ion (CN-) as the axial ligands allowed detailed characterization of the physicochemical nature of the axial H2O ligand. We found that the in-plane asymmetry of the heme electronic structure of the complex is not largely affected by the axial H2O coordination, indicating that the H2O confined in the hydrophobic interface between the heme and G-quartet planes of the complex rotates about the coordination bond with respect to the heme. The effect of the hydrogen(H)/deuterium(D) isotope replacement of the axial H2O on the heme electronic structure was manifested in the isotope shifts of paramagnetically shifted heme methyl proton signals of the complex in such a manner that three resolved peaks associated with axial H2O, HDO, and D2O were observed for each of the heme methyl proton signals. These findings provide not only the basis for an understanding of the nature of the unique axial H2O but also an insight into the molecular mechanism responsible for the control of the heme reactivity in the heme-DNA complex.

A cationic copolymer as a cocatalyst for a peroxidase-mimicking heme-DNAzyme.

Heme binds to a parallel-stranded G-quadruplex DNA to form a peroxidase-mimicking heme-DNAzyme. An interpolyelectrolyte complex between the heme-DNAzyme and a cationic copolymer possessing protonated amino groups was characterized and the peroxidase activity of the complex was evaluated to elucidate the effect of the polymer on the catalytic activity of the heme-DNAzyme. We found that the catalytic activity of the heme-DNAzyme is enhanced through the formation of the interpolyelectrolyte complex due to the general acid catalysis of protonated amino groups of the polymer, enhancing the formation of the iron(IV)oxo porphyrin π-cation radical intermediate known as Compound I. This finding indicates that the polymer with protonated amino groups can act as a cocatalyst for the heme-DNAzyme in the oxidation catalysis. We also found that the enhancement of the activity of the heme-DNAzyme by the polymer depends on the local heme environment such as the negative charge density in the proximity of the heme and substrate accessibility to the heme. These findings provide novel insights as to molecular design of the heme-DNAzyme for enhancing its catalytic activity.

Effects of Heme Electronic Structure and Local Heme Environment on Catalytic Activity of a Peroxidase-Mimicking Heme-DNAzyme.

The catalytic cycle of a peroxidase-mimicking heme-DNAzyme involves an iron(IV)oxo porphyrin π-cation radical intermediate known as compound I formed through heterolytic O-O bond cleavage of an Fe3+-bound hydroperoxo ligand (Fe-OOH) in compound 0, like that of a heme enzyme such as horseradish peroxidase (HRP). Peroxidase assaying of complexes composed of chemically modified hemes possessing various electron densities of the heme iron atom (ρFe) and parallel-stranded tetrameric G-quadruplex DNAs of oligonucleotides d(TTAGGG), d(TTAGGGT), and d(TTAGGGA) was performed to elucidate the effects of the heme electronic structure and local heme environment on the catalytic activity of the heme-DNAzyme. The study revealed that the DNAzyme activity is enhanced through an increase in the ρFe and general base catalysis of the adenine base adjacent to the heme, which are reminiscent of the "push" and "pull" mechanisms in the catalytic cycle of HRP, respectively, and that the activity of the heme-DNAzyme can be independently controlled through the heme electronic structure and local heme environment. These findings allow a deeper understanding of the structure-function relationship of the peroxidase-mimicking heme-DNAzyme.

Structural and functional characterization of complexes between heme and dimeric parallel G-quadruplex DNAs.

Heme has been receiving considerable interest as a prosthetic group of ribozymes and deoxyribozymes, because heme-bound nucleic acids exhibit peroxidase-like catalytic activities (Travascio, P., Li, Y., and Sen, D. (1998) Chem. Biol, 5, 505-517). The interaction of heme with dimeric G-quadruplexes formed from d(TAGGGTTAGGGT) and d(TAGGGTTAGGGA) has been characterized to gain a deeper understanding of the molecular recognition of G-quadruplex DNAs by heme. We found that heme binds selectively to the 3'-terminal G-quartet of a dimeric parallel G-quadruplex of d(TAGGGTTAGGGT), whereas binding of heme to a dimeric antiparallel G-quadruplex of d(TAGGGTTAGGGA) does not occur, suggesting that an orderly arrangement of the constituent guanine deoxyribose rings, with respect to the G-quartet plane, is crucial for the binding of heme to the DNA. The preferential binding of heme to the 3'-terminal G-quartet of parallel G-quadruplex DNAs allowed a systematic modification of the heme environment in the complex through the DNA sequence. The activity of the complexes was found to increase with increasing number of adenine bases adjacent to the heme in the complexes, possibly due to improvement of the accessibility of aromatic substrate, i.e., 10-acetyl-3,7-dihydroxyphenoxazine, to the heme, and an increase in the frequency of appearance of a specific orientation of the adenine bases, with respect to the heme, optimized for its activity as an acid-base catalyst to enhance the peroxidase activity of the complex.

Effect of the Electron Density of the Heme Fe Atom on the Nature of Fe-O2 Bonding in Oxy Myoglobin.

Mössbauer spectroscopy has been used to characterize oxygenated myoglobins (oxy Mbs) reconstituted with native and chemically modified 57Fe-enriched heme cofactors with different electron densities of the heme Fe atom (ρFe) and to elucidate the effect of a change in the ρFe on the nature of the bond between heme Fe and oxygen (O2), i.e., the Fe-O2 bond, in the protein. Quadrupole splitting (ΔEQ) was found to decrease with decreasing ρFe, and the observed ρFe-dependent ΔEQ confirmed an increase in the contribution of the ferric-superoxide (Fe3+-O2-) form to the resonance hybrid of the Fe-O2 fragment with decreasing ρFe. These observations explicitly accounted for the lowering of O2 affinity of the protein due to an increase in the O2 dissociation rate and a decrease in the autoxidation reaction rate of oxy Mb through decreasing H+ affinity of the bound ligand with decreasing ρFe. Therefore, the present study demonstrated the mechanism underlying the electronic control of O2 affinity and the autoxidation of the protein through the heme electronic structure. Carbon monoxide (CO) adducts of reconstituted Mbs (CO-Mbs) were similarly characterized, and we found that the resonance between the two canonical forms of the Fe-CO fragment was also affected by a change in ρFe. Thus, the nature of the Fe-ligand bond in the protein was found to be affected by the ρFe.

Stepwise binding of a cationic phthalocyanine derivative to an all parallel-stranded tetrameric G-quadruplex DNA.

Interaction between an Ga(III) phthalocyanine (Pc) derivative bearing eight N-methylpyridinium groups at peripheral ß-positions (2,3,6,7,10,11,14,15-octakis-[N-methyl-(4-methylpyridinium-3-yloxy)phthalocyaninato] chloro gallium(III) iodide (GaPc)) and an all parallel-stranded tetrameric G-quadruplex formed from a heptanucleotide d(TTAGGGT) ([d(TTAGGGT)]4) has been investigated to elucidate the molecular recognition of G-quadruplex DNA by the Pc derivative, which provides a useful insight as to the design of G-quadruplex ligands suitable for various in vitro and in vivo applications. We found that GaPc binds to the A3G4 and G6T7 steps of [d(TTAGGGT)]4, with binding constants of (21 ± 2) × 106 and (0.09 ± 0.06) × 106 M-1, respectively, to form a 2:1 complex. Obviously, upon the binding of GaPc to each of the sites, the π-π stacking and electrostatic interactions of the Pc moiety and positively-charged side chains of GaPc with a G-quartet and the negatively-charged phosphate groups in nearby phosphodiester bonds of the DNA, respectively, are major driving forces for the complexation. Considering the similarity in the local structural environment between the A3G4 and G6T7 steps of [d(TTAGGGT)]4, the remarkably large difference in the GaPc-binding affinity between them is most likely accounted for by the effect of the polarity of the GaPc-binding site on the intermolecular electrostatic interaction. This finding provides valuable insights as to the design of Pc derivatives as G-quadruplex ligands.

Multimodal Multiphoton Imaging of the Lipid Bilayer by Dye-Based Sum-Frequency Generation and Coherent Anti-Stokes Raman Scattering.

Coherent anti-Stokes Raman scattering (CARS) imaging is widely used for imaging molecular vibrations inside cells and tissues. Lipid bilayers are potential analytes for CARS imaging due to their abundant CH2 vibrational bonds. However, identifying the plasma membrane is challenging since it possesses a thin structure and is closely apposed to lipid structures inside the cells. Since the plasma membrane provides the most prominent asymmetric location within cells, orientation sensitive sum-frequency generation (SFG) imaging is a promising technique for selective visualization of the plasma membrane labeled by a nonfluorescent and SFG-specific dye, Ap3, when using a CARS microscope system. In this study, we closely compare the characteristics of lipid bilayer imaging by dye-based SFG and CARS using giant vesicles (GVs) and N27 rat dopaminergic neural cells. As a result, we show that CARS imaging can be exploited for the visualization of whole lipid structures inside GVs and cells but is insufficient for identification of the plasma membrane, which instead can be achieved using dye-based SFG imaging. In addition, we demonstrate that these unique properties can be combined and applied to the live-cell tracking of intracellular lipid structures such as lipid droplets beneath the plasma membrane. Thus, multimodal multiphoton imaging through a combination of dye-based SFG and CARS can serve as a powerful chemical imaging tool to investigate lipid bilayers in GVs and living cells.

Monitoring the morphological evolution of giant vesicles by azo dye-based sum-frequency generation (SFG) microscopy.

In the present work, dye-based sum-frequency generation (SFG) imaging using sodium 4-[4-(dibutylamino)phenylazo]benzenesulfonate (butyl orange, BO) as a new non-fluorescent specific azo dye is employed to monitor the morphological evolution of giant vesicles (GVs). After loading BO to the membrane of 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) single-component GVs, the outermost membranes were clearly visualized using SFG microscopy, which provided images of the distinct outer and inner faces of the lipid bilayers. In addition, SFG-active vesicles were detected also inside the GVs, depending on the dye concentrations. The dye-based SFG imaging technique provided experimental evidence that these oligolamellar vesicles containing an SFG-active interior had been formed after BO loading. The formation process of the oligolamellar vesicles with inner SFG-active vesicles was successfully monitored, and their formation mechanism was discussed.

Synergistic Effect of Distal Polar Interactions in Myoglobin and Their Structural Consequences.

In the L29F variant of myoglobin (Mb), the coordination of oxygen (O2) to the heme Fe atom is stabilized by favorable electrostatic interactions between the polar Fe-O2 moiety and the multipole of the phenyl ring of the Phe29 side chain (Phe29 interaction), in addition to the well-known hydrogen bond (H-bond) between the Fe-bound O2 and the 64th residue (distal H-bond; Carver, T. E.; Brantley, R. E., Jr.; Singleton, E. W.; Arduini, R. M.; Quillin, M. L.; Phillips, G. N., Jr.; Olson, J. S. J. Biol. Chem. 1992, 267, 14443-14450). The O2 and carbon monoxide (CO) binding properties and autoxidation of the L29F/H64L and L29F/H64Q variants reconstituted with a series of chemically modified heme cofactors were analyzed and then compared with those of native Mb, and the L29F, H64Q, and H64L variants similarly reconstituted with the chemically modified heme cofactors in order to elucidate the relationship between the Phe29 interaction and the distal H-bond that critically contributes to stabilization of Fe-bound O2. We found that the Phe29 interaction and distal H-bond act cooperatively to stabilize the Fe-bound O2 in such a manner that the Phe29 interaction strengthens with increasing strength of the distal H-bond. Comparison of the functional properties between the L29F and H64L variants indicated that the synergistic effect of the two interactions decreases the O2 dissociation and autoxidation rate constants of the protein by factors of ∼1/2000 and ∼1/400, respectively. Although the CO binding properties of the proteins were not greatly affected by the distal polar interactions, their synergistic effects were clearly and sharply manifested in the vibrational frequencies of the Fe-bound C-O stretching of the proteins.

Structures and Catalytic Activities of Complexes between Heme and All Parallel-Stranded Monomeric G-Quadruplex DNAs.

Heme in its ferrous and ferric states [heme(Fe2+) and heme(Fe3+), respectively] binds selectively to the 3'-terminal G-quartet of all parallel-stranded monomeric G-quadruplex DNAs formed from inosine(I)-containing sequences, i.e., d(TAGGGTGGGTTGGGTGIG) DNA(18mer) and d(TAGGGTGGGTTGGGTGIGA) DNA(18mer/A), through a π-π stacking interaction between the porphyrin moiety of the heme and the G-quartet, to form 1:1 complexes [heme-DNA(18mer) and heme-DNA(18mer/A) complexes, respectively]. These complexes exhibited enhanced peroxidase activities, compared with that of heme(Fe3+) alone, and the activity of the heme(Fe3+)-DNA(18mer/A) complex was greater than that of the heme(Fe3+)-DNA(18mer) one, indicating that the 3'-terminal A of the DNA sequence acts as an acid-base catalyst that promotes the catalytic reaction. In the complexes, a water molecule (H2O) at the interface between the heme and G-quartet is coordinated to the heme Fe atom as an axial ligand and possibly acts as an electron-donating ligand that promotes heterolytic peroxide bond cleavage of hydrogen peroxide bound to the heme Fe atom, trans to the H2O, for the generation of an active species. The intermolecular nuclear Overhauser effects observed among heme, DNA, and Fe-bound H2O indicated that the H2O rotates about the H2O-Fe coordination bond with respect to both the heme and DNA in the complex. Thus, the H2O in the complex is unique in terms of not only its electronic properties but also its dynamic ones. These findings provide novel insights into the design of heme-deoxyribozymes and -ribozymes.

Characterization of Catalytic Activities and Heme Coordination Structures of Heme-DNA Complexes Composed of Some Chemically Modified Hemes and an All Parallel-Stranded Tetrameric G-Quadruplex DNA Formed from d(TTAGGG).

Heme binds selectively to the 3'-terminal G-quartet (G6 G-quartet) of an all parallel-stranded tetrameric G-quadruplex DNA, [d(TTAGGG)]4, to form a heme-DNA complex. Complexes between [d(TTAGGG)]4 and a series of chemically modified hemes possessing a heme Fe atom with a variety of electron densities were characterized in terms of their peroxidase activities to evaluate the effect of a change in the electron density of the heme Fe atom (ρFe) on their activities. The peroxidase activity of a complex decreased with a decreasing ρFe, supporting the idea that the activity of the complex is elicited through a reaction mechanism similar to that of a peroxidase. In the ferrous heme-DNA complex, carbon monoxide (CO) can bind to the heme Fe atom on the side of the heme opposite the G6 G-quartet, and a water molecule (H2O) is coordinated to the Fe atom as another axial ligand, trans to the CO. The stretching frequencies of Fe-bound CO (νCO) and the Fe-C bond (νFe-C) of CO adducts of the heme-DNA complexes were determined to investigate the structural and electronic natures of the axial ligands coordinated to the heme Fe atom. Comparison of the νCO and νFe-C values of the heme-DNA complexes with those of myoglobin (Mb) revealed that the donor strength of the axial ligation trans to the CO in a complex is considerably weaker than that of the proximal histidine in Mb, as expected from the coordination of H2O trans to the CO in the complex.

Multimodal two-photon imaging using a second harmonic generation-specific dye.

Second harmonic generation (SHG) imaging can be used to visualize unique biological phenomena, but currently available dyes limit its application owing to the strong fluorescent signals that they generate together with SHG. Here we report the first non-fluorescent and membrane potential-sensitive SHG-active organic dye Ap3. Ap3 is photostable and generates SH signals at the plasma membrane with virtually no fluorescent signals, in sharp contrast to the previously used fluorescent dye FM4-64. When tested in neurons, Ap3-SHG shows linear membrane potential sensitivity and fast responses to action potentials, and also shows significantly reduced photodamage compared with FM4-64. The SHG-specific nature of Ap3 allows simultaneous and completely independent imaging of SHG signals and fluorescent signals from various reporter molecules, including markers of cellular organelles and intracellular calcium. Therefore, this SHG-specific dye enables true multimodal two-photon imaging in biological samples.

Producing a dual-fluorescent molecule by tuning the energetics of excited-state intramolecular proton transfer.

We report herein the selective preparation of normal, tautomeric, and dual-fluorescent molecules with a common ESIPT core. 2'-Hydroxyacetophenone (OHAP) is known as a typical molecule that undergoes excited-state intramolecular hydrogen transfer (ESIPT) to display fluorescence emission from the excited state of the tautomer. In this study, a series of ten OHAP-cored fluorescent molecules were prepared and their excited state properties have been explored. The bathochromic shift of the π-π* absorption band with π-extensions of substituents of these molecules indicates that the excitation energy of the normal form of the OHAP unit was reduced due to the substituents, while the energy of the excited tautomer appeared to be independent of the π-extension of the substituents. When pyrene or anthracene was connected at the end (molecules 4 and 5), only normal fluorescence appeared, and the tautomer fluorescence disappeared. An anthracene derivative (molecule 10) displayed dual fluorescence, indicating that the normal and the tautomer excited states were energetically "balanced". A fluorescence lifetime analysis revealed the ESIPT reaction rate of 10 to be much slower than those of other derivatives and that the normal and tautomer forms were in equilibrium in the excited state.

Self-assembly of a poly(glutamate) dendrimer: solvent-dependent expression of molecular chirality and a [2 + 2] photocrosslinking reaction.

Upon photoirradiation of a stilbene-cored poly(glutamate) dendrimer, the photocrosslinking reaction preferentially occurred to produce the [2 + 2] cycloadduct in benzene. In addition, blue-shifted absorption spectra and a bisignate circular dichroism (CD) signal were observed, indicating that the aggregates of the stilbene core were chiral.

Synthetic localized calcium transients directly probe signalling mechanisms in skeletal muscle.

The contribution of Ca2+-induced Ca2+ release (CICR) to trigger muscle contraction is controversial. It was studied on isolated muscle fibres using synthetic localized increases in Ca2+ concentration, SLICs, generated by two-photon photorelease from nitrodibenzofuran (NDBF)-EGTA just outside the permeabilized plasma membrane. SLICs provided a way to increase cytosolic [Ca2+] rapidly and reversibly, up to 8 µM, levels similar to those reached during physiological activity. They improve over previous paradigms in rate of rise, locality and reproducibility. Use of NDBF-EGTA allowed for the separate modification of resting [Ca2+], trigger [Ca2+] and resting [Mg2+]. In frog muscle, SLICs elicited propagated responses that had the characteristics of CICR. The threshold [Ca2+] for triggering a response was 0.5 µM or less. As this value is much lower than concentrations prevailing near channels during normal activity, the result supports participation of CICR in the physiological control of contraction in amphibian muscle. As SLICs were applied outside cells, the primary stimulus was Ca2+, rather than the radiation or subproducts of photorelease. Therefore the responses qualify as 'classic' CICR. By contrast, mouse muscle fibres did not respond unless channel-opening drugs were present at substantial concentrations, an observation contrary to the physiological involvement of CICR in mammalian excitation­contraction coupling. In mouse muscle, the propagating wave had a substantially lower release flux, which together with a much higher threshold justified the absence of response when drugs were not present. The differences in flux and threshold may be ascribed to the absence of ryanodine receptor 3 (RyR3) isoforms in adult mammalian muscle.

Environmental polarity estimation in living cells by use of quinoxaline-based full-colored solvatochromic fluorophore PQX and its derivatives.

Fluorosolvatochromic probes have recently attracted interest for live cell imaging. We recently tested our quinoxaline-based fluorosolvatochromic probes, PQX and MPQX, for in vivo imaging applications. PQX and MPQX are characterized by donor (pyrrole site)-acceptor (quinoxaline site) structure, and the peak emission wavelength of these compounds varies over the visible wavelength range depending on the polarity of the solvent used, for a variety of solvents from hexane to water. A linear relationship was obtained between peak emission wavenumber and E(T)(N) (normalized solvent polarity). When tested on cultured HEp-2 cells, the results showed that the PQX and MPQX were not harmful at the applied concentrations (10(-5) M), and site-dependent fluorescence spectra in living cells were observed with PQX treatment, which indicates that PQX penetrates into the plasma membrane, followed by delocalization throughout the cells. MPQX was also able to penetrate the cell membrane, distribute throughout the cells and emit fluorescence, but did not show site-dependent intracellular fluorescence spectra. These results suggest that PQX is a remarkable tool for investigating the local polarity environment in cells after appropriate conjugation to biomolecules.