Turn-on Fluorene Push–Pull Probes with High Brightness and

Oct 20, 2017 - The rational design of environmentally sensitive dyes with superior properties is critical for elucidating the fundamental biological p...
0 downloads 0 Views 4MB Size
Articles pubs.acs.org/acschemicalbiology

Turn-on Fluorene Push−Pull Probes with High Brightness and Photostability for Visualizing Lipid Order in Biomembranes Janah Shaya,† Mayeul Collot,‡ Frédéric Bénailly,† Najiba Mahmoud,† Yves Mély,‡ Benoît Y. Michel,† Andrey S. Klymchenko,‡ and Alain Burger*,† †

Université Côte d’Azur, CNRS, Institut de Chimie de Nice, UMR 7272 − Parc Valrose, 06108 Nice cedex 2, France Laboratoire de Biophotonique et Pharmacologie, UMR 7213, Faculté de Pharmacie, Université de Strasbourg, CNRS, 74 Route du Rhin, 67401 Illkirch, France



S Supporting Information *

ABSTRACT: The rational design of environmentally sensitive dyes with superior properties is critical for elucidating the fundamental biological processes and understanding the biophysical behavior of cell membranes. In this study, a novel group of fluorene-based push−pull probes was developed for imaging membrane lipids. The design of these fluorogenic conjugates is based on a propioloyl linker to preserve the required spectroscopic features of the core dye. This versatile linker allowed the introduction of a polar deoxyribosyl head, a lipophilic chain, and an amphiphilic/anchoring group to tune the cell membrane binding and internalization. It was found that the deoxyribosyl head favored cell internalization and staining of intracellular membranes, whereas an amphiphilic anchor group ensured specific plasma membrane staining. The optimized fluorene probes presented a set of improvements as compared to commonly used environmentally sensitive membrane probe Laurdan such as red-shifted absorption matching the 405 nm diode laser excitation, a blue-green emission range complementary to the red fluorescent proteins, enhanced brightness and photostability, as well as preserved sensitivity to lipid order, as shown in model membranes and living cells.

F

frontier between the cell and its extracellular environment. Such characterization is challenging because the composition and the nature of membrane lipids are highly diverse, and they differ from cell to cell and over time. These variations have profound effects on the membrane fluidity, permeability, and dynamics, and thus they impact and modulate various cellular processes. Another consequence of the heterogeneity and dynamics of membrane lipids is the formation of transient but highly ordered lipid domains, often referred to as membrane “rafts.”15 Searching for probes responding specifically to lipid domains can inevitably help in deciphering this lateral heterogeneity of cell membranes described by the lipid raft hypothesis.16,17 Lipid rafts are hitherto under debate and are believed to be at the origin of fundamental biological processes such as signal transduction and neurodegenerative diseases.18,19 Recent years have seen significant progress in the development of fluorescent probes that can specifically address the membrane properties, especially in relation to the phenomenon of lipid rafts.16 One major class of probes is composed of solvatochromic push−pull dyes that sense membrane polarity, hydration, as well as water relaxation dynamics.13,20 Mechano-

luorescence imaging has revolutionized the way to observe cellular structures and to understand their functions.1,2 The use of fluorescent probes and labels remains undoubtedly the hallmark of the success of these techniques. According to their size, the employed fluorophores can be classified into three different families: fluorescent proteins, nanoparticles, and smallmolecule organic dyes.3−9 The latter has attracted major attention in advanced applications since an organic dye is a relatively small label that might be less disrupting than fluorescent proteins and nanoparticles. Furthermore, their chemistry and photophysics can be finely tuned.10−12 Among organic labels, the fluorescent probes that turn “on” their fluorescence intensity (fluorogenic dyes) and change their color (solvatochromic dyes) in response to interactions with their biomolecular targets are highly attractive for imaging.13 These dyes provide a target-specific emission signal and ensure background-free imaging using simple and convenient staining protocols without the need to wash out the excess reagent.14 Such emissive dyes should display high brightness, photostability, and selectivity to the target in order to maximize the spatiotemporal imaging resolution as well as compatibility for excitation with the available laser lines. As a matter of fact, organic probes that fulfill all these requirements are rare. Investigation of membranes, and especially of the cytoplasmic membrane, is of key importance because it determines the © 2017 American Chemical Society

Received: August 2, 2017 Accepted: October 20, 2017 Published: October 20, 2017 3022

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030

Articles

ACS Chemical Biology

Chart 1. Structuresa and Selected Photophysical Properties of Common Push−Pull Dyes and Membrane Probes13,30,35−38

a

Donor and acceptor groups are highlighted in green and red colors, respectively.

sensitive fluorescent probes belong to another important class of membrane probes.21,22 The key representatives are molecular rotors that change their intensity and lifetime in response to the viscosity of the lipid membrane.23,24 One should also mention the “flipper” fluorescent probes that change their ground state conformation (i.e., excitation spectra) as a function of lipid order.25,26 Thus, particularly important are membrane probes based on push−pull dyes (e.g., Prodan/ Laurdan, Chart 1). They present donor and acceptor groups and undergo an excited-state intramolecular charge transfer (ICT) resulting in strong sensitivity to polarity and hydration. They are characterized by red-shifted emissions (color changes) as the solvent becomes more polar or hydrated and switchedoff emission in water. Further, they allow ratiometric detection, and thus quantitative fluorescence microscopy, when recording the fluorescence signal as the ratio of intensities at two wavelengths. These peculiar properties were exploited to probe hydrophobic environments and as such to investigate membrane properties.20 Fluorescence-based techniques using push−pull probes (Chart 1) have shed light on the properties and the functions of membranes.16 Laurdan, a lipophilic analog of Prodan, is one of the first examples.27,28 It inserts and stains the plasma membranes but also the intracellular membranes due to its fast diffusion across the plasma membrane and accumulation inside the cells. It changes its emission maximum as a function of lipid order, and therefore it was employed to study the lateral organization of membranes and dynamics. However, its absorption in the ultraviolet region, nonspecific binding, and poor photostability have limited its use. Poor photostability is a common drawback of push−pull probes.13 Several types of probes were developed to address the drawbacks of Laurdan. Introduction of polar headgroup carboxylate in C-Laurdan improved the cell membrane localization.27 Likewise, the absorption and emission properties were shifted to the red with the introduction of styrylpyridinium probe di-4-ANEPPDHQ, which was also characterized by plasma membrane specific staining.29 Further improvements in push−pull probes were achieved with the development of NR12S.30,31 This dye is built of the environment-sensitive Nile Red dye and a dodecyl chain bearing a zwitterionic group. Compared to Laurdan and the

previously reported probes, NR12S exhibits a red-shifted operating range and superior brightness. The introduction of the polar zwitterionic group affords exclusive staining of the plasma membrane at the outer leaflet with negligible probe internalization and flip-flop. NR12S was also applied for imaging of endosome maturation, conformational changes of membrane proteins, internalization of nanostructures, and detection of apoptosis.13 Recently, this probe as well as the styryl-pyridinium were successfully employed for superresolution imaging using Stimulated Emission Depletion (STED) microscopy.32 Nevertheless, NR12S is not an ideal probe since it shows limited photostability,33 and its absorption and emission bands are broad, thus limiting the possibility for multicolor imaging. In a more recent work, the push−pull pyrene dye PA, a red-shifted analog of Laurdan, was proposed as a cell penetrating probe for studying lipid order in biomembranes with superior photostability and brightness;34 nevertheless, its derivatization for specific staining of the cell plasma membrane was not reported. Moreover, its emission spectrum is rather broad (between 500−600 nm), and therefore its use in combination with green and red fluorescent proteins is limited. Therefore, an advanced plasma membrane probe should present the following rigorous properties: excitation in the visible range, high brightness, high photostability, exclusive staining of the plasma membrane with high contrast, and distinction of the lipid domains by changing its fluorescence color. Additionally, the probe should show minimal internalization and interference from cell autofluorescence and should operate in the visible range compatible for multicolor imaging.20 Finally, tuning the photophysical properties of the probe and its synthesis should be accessible and avoid demanding and lengthy multistep routes. The previously reported push−pull fluorene derivative (FR0) is a promising building block for the preparation of efficient membrane probes. FR0 presents a number of advantages as compared to Laurdan: red-shifted absorption and emission, a higher extinction coefficient and fluorescence quantum yield, as well as higher photostability.37 However, its application to biomembrane research has not been explored. In the continuity of our work to find advanced fluorescent probes 3023

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030

Articles

ACS Chemical Biology Scheme 1. Synthesis of Probes FR3−5 via CuAAC



for biological applications,39−41 we have developed concise routes to access an array of push−pull fluorene-based fluorophores derived from the reference FR0 dye and screened different types of linkers for conjugation (FR1 and FR2 as examples, Chart 1).38,42 These dyes consist of a fluorene πscaffold functionalized with an electron donor and acceptor favorable to intramolecular charge transfer (ICT) and Hbonding interactions.43 Analysis of the structure−photophysics relationships of the library allows identifying the most prospective push−pull combination in terms of visible absorption, laser excitation, accessibility, brightness, Stokes shift, and solvatofluorochromism.38 The propioloyl linker was found to be particularly attractive because the conjugated product (FR1) was obtained conveniently via 1,3-dipolar “click” cycloaddition and exhibited desirable optical properties (compare the absorption and emission maxima, and the extinction coefficients of FR1 with FR0 and FR2).38,44 Indeed, FR1 absorbs in the violet (400 nm) with a high extinction coefficient (38 400 M−1 cm−1 vs 18 400 M−1 cm−1 for Laurdan), which is useful for selective excitation with the common 405 nm laser. It demonstrates ICT and H-bonding acceptance characters (the carbonyl acting as an H-bond acceptor) among the highest within the fluorene series. FR1 exhibits high quantum yield and red-shifted emission (Δλ ∼ 80−190 nm) in aprotic solvents. The largest displacement of its emission to near-infrared (Δλ > 220 nm) is shown in polar protic solvents where the dye turns almost nonemissive in bulk water. The turn-on ability upon shielding this dye from water and its prominent anchoring point for biolabeling render the fluorene family a promising candidate for sensing biomembranes and probing their lipid domains. In this paper, we describe the synthesis of three derivatives of push−pull fluorene dyes bearing (a) a polar deoxyribosyl head (FR3), (b) a lipophilic long chain (FR4), and (c) an amphiphilic anchoring group (FR5) to fix the probe at the membrane interface and prevent its internalization.30,38,45−47 Moreover, we characterize the photophysics and photodegradation studies of the three dyes FR3−5 as well as their ability to sense the lipid order of large and giant unilamellar vesicles (LUVs and GUVs) of different lipid compositions and to image membranes of HeLa cells.

RESULTS AND DISCUSSION Synthetic Strategy. Scheme 1 depicts the synthesis of the membrane probes FR3−5. Initially, the propiolyl intermediate 1 and the azido derivatives 2−4 were prepared as previously done and are detailed (Supporting Information).38,45−47 Next, the clickable fluorene 1 was converted via copper-assisted azide−alkyne Huisgen cycloaddition (CuAAC) into the ketotriazolyl derivatives: (FR3, 80% yield) with 2-deoxy-β-Derythro-pentofuranosyl azide 2 (FR3, 75%), with n-dodecyl azide 3, and (FR5, 61%) with the azido amphiphilic chain 4. The optimized click reaction conditions consisted of CuSO4· 5H2O (0.5 equiv) and sodium ascorbate (0.7 equiv) in a mixture of DMF/H2O (3:1) at RT for 3 h. Photophysical Properties. We first investigated the impact of the triazole substituent on the UV−vis absorption and fluorescence properties in a set of aprotic and protic solvents with a large difference in polarities (Table 1). For analyzing the absorption and emission of FR3−5, we used Reichardt’s polarity scale ET(30), an empirical scale commonly used in the analysis of the solvatochromism of dyes.48 The emission spectra of FR3 are depicted in Figure 1 to illustrate the solvatofluorochromism of these dyes (FR4,5, Supporting Information). The three dyes absorbed strongly in the violet region around 400 nm (35−38 × 103 M−1·cm−1 in dioxane) similarly to the reference compound FR1. Their absorption maxima demonstrated little dependence on the polarity change (ΔλAbs < 20 nm), displaying almost no solvatochromism. This evidences the absence of ICT character in their ground states contrary to Nile Red, which shows solvent-dependent variation of its absorption maximum. In sharp contrast to absorption, their emission maxima were proportionally red-shifted along the range of polarity, demonstrating their high sensitivity to the polarity and H-bond donating ability of the solvent (Figure 1A). Noticeably, the dyes showed already large Stokes shifts (e.g., FR3 = 100 nm in toluene, the solvent of the lowest polarity, and 256 nm in MeOH). FR3−5 are strongly emissive in aprotic solvents. They are much less emissive in alcohols, and they become nonfluorescent in water. The photophysical properties of FR3−5 parallel those reported for the parent compound FR1. Effect of Hydration on FR3. We next explored the effect of hydration on the emission of dye FR3 in mixtures of dioxane 3024

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030

Articles

ACS Chemical Biology

Photophysical Characterization in Large Unilamellar Vesicles of Different Lipid Compositions. LUVs are the simplest models to study biomembranes. Sphingomyelin (SM) and cholesterol (CL) were used to produce a liquid-ordered membrane phase (Lo), while a liquid-disordered membrane phase (Ld) was obtained with the unsaturated phospholipid, dioleoylphosphatidylcholine (DOPC).16 First, the optimal probe/lipid ratio was determined with FR3 by varying the concentration of SM/CL LUVs. As shown in Figure 1C,D, the fluorescence intensity of FR3 increases upon addition of LUVs and saturates at a probe/lipid ratio about 1:500. This suggests that nearly all the dye molecules are bound to lipids at this ratio and, thus, that the probe binds efficiently to the lipid membranes.30 It is worth mentioning that the binding effect on fluorescence was instantaneous, as the same spectra were reproduced after the samples were left for 5, 10, and 15 min at RT or at 60 °C. Next, the ability of dyes FR3−5 for probing LUVs of different compositions was tested. The main spectroscopic properties are given in Table 2 and Figure 2 (Supporting Information). The absorption maxima of FR3−5 in different lipid vesicles were around 400 nm. Their emission maxima exhibited some dependence on the lipid composition and liquid phases of the LUVs (Figure 2, Table 2). Binding of FR3−5 to lipid vesicles resulted in a considerable increase of the fluorescence intensities. Depending on the lipid composition, the quantum yields varied from 51 to 64, 24 to 63, and 59 to 99% for FR3, FR4, and FR5, respectively (Table 2). The quantum yields of FR3 and FR5 are generally larger than those of Laurdan and NR12S.30 Noticeably, the quantum yield of FR5 in the liquidordered phase is larger than that of Laurdan and NR12S by 5 and 2 times, respectively. Further, FR3−5 showed a significant hypsochromic shift (e.g., from 549 to 488 nm for FR4) when the lipid phase changed from Ld (DOPC) to Lo (SM/CL). This is consistent with the fact that the Lo phase is less hydrated as a consequence of the higher level of lipid packing as compared to the Ld phase.16 The differences in the emission (ΔλEm) between the two extreme states were 28, 61, and 21 nm for FR3−5, respectively. Thus, FR4 demonstrated the highest sensitivity among the three conjugates. FR4 was also more responsive than Laurdan (ΔλEm = 48 nm) and Nile Red (ΔλEm = 36 nm).27,28,30 The lower sensitivity of FR3 and FR5 to phase changes is likely related to the functional groups that anchor the dye at the interface. Indeed, the response of the environmentally sensitive probes to phase changes is frequently associated with changes in the location and orientation of the fluorophore in membranes.20 Therefore, the precise positioning of FR3 and FR5 due to the functional groups may limit the movement of fluorophores and/or their contact with the water molecules at the bilayer interface, thus decreasing their sensitivity. Photodegradation Study. Next, we investigated the photodegradation of FR3−5 in their free forms and bound to lipid vesicles (DOPC) under continuous illumination of the samples (Figure 3 and Supporting Information). Xylene was chosen as a representative high boiling point apolar solvent mimicking a hydrophobic environment. The photostability of solvatochromic fluorescent dyes in apolar media is a major concern in imaging of lipid membranes since commonly used dyes (Prodan, Laurdan, and Nile Red) photobleach rapidly in this apolar environment.16,34,52 Prodan and the fluorenyl carbaldehyde (FR0) were used as references.

Table 1. Spectroscopic Properties of FR0 and FR3−5 in Various Solventsa εmax (103 M−1·cm−1)b

37.6

38.0

37.1

35.0

solvent (ET(30))

λ, Δλ [nm]c,d Φ [%]e

FR0

FR3

FR4

FR5

PBSf (63.1) MeOH (55.4)

λAbs λAbs λEm Δλ Φ λAbs λEm Δλ Φ λAbs λEm Δλ Φ λAbs λEm Δλ Φ λAbs λEm Δλ Φ λAbs λEm Δλ Φ λAbs λEm Δλ Φ λAbs λEm Δλ Φ

ndg nd nd nd nd 396 622 226 6 389 578 189 77 402 582 180 85 399 532 133 72 nd nd nd nd 391 488 97 52 394 473 79 86

390 395 651 256 1.5 403 630 227 6 395 586 191 42 405 592 187 53 403 549 146 74 393 550 157 72 394 494 100 82 391 491 100 60

401 400 647 247 3 400 629 229 8 391 578 187 51 402 584 182 75 404 539 135 56 390 527 137 61 393 481 88 96 399 472 73 85

409 399 662 263 2 397 636 239 8 390 587 197 61 401 592 191 93 402 559 157 66 392 547 155 69 397 517 120 72 402 518 116 54

EtOH (51.9)

CH3CN (45.6)

DMSO (45.1)

CHCl3 (39.1)

EtOAc (38.1)

dioxane (36)

toluene (33.9)

a

Reported values are the average of two or more independent and reproducible measurements; ±1 nm for wavelengths. bThe molar extinction coefficient was determined in dioxane and methanol; relative standard deviations are lower or equal to 5%. cExcitation wavelength was at the corresponding absorption maximum. dFor convenience, Stokes shifts (Δλ = λEm − λAbs) are expressed in nm rather than in cm−1. eFluorescence quantum yields Φ were determined using the excitation at the corresponding absorption maximum of each compound in the considered solvent. Quinine sulfate in 0.5 M H2SO4 solution (λEx = 350 nm, Φ = 55%)49 and p-diMethylAminoFlavone (dMAF) in EtOH (λEx = 404 nm, Φ = 27%)50 were used as references, ±10% mean standard deviation. f20 mM, pH 7.4. gnd = not determined.

and water. As evidenced in Figure 1B, hydration dramatically shifts the position of the emission band from 494 nm to the near-infrared and results in a gradual quenching of the fluorescence, which is common for dyes with strong ICT character.48,51 As a result, the change from a polar aqueous environment to apolar media results in a dramatic turn-on of the fluorescence intensity (1200 folds for a fixed wavelength at 494 nm). Altogether, our data underline the high sensitivity of the fluorene push−pull dyes to hydration and polarity. These properties appear ideal for sensing structural changes in lipid membranes involving modifications of water concentration in the bilayer. 3025

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030

Articles

ACS Chemical Biology

Figure 1. (A) Absorption (dashed line) and emission (solid lines) of FR3 in toluene (black), dioxane (blue), chloroform (light green), ethyl acetate (magenta), acetonitrile (khaki), DMSO (orange), ethanol (dark green), and methanol (red) as representatives of aprotic apolar, aprotic polar, and protic polar solvents. Excitation wavelength was at the corresponding absorption maximum. Absorption and emission of FR3 were normalized on their respective maximum. As the absorption is weakly solvatochromic, one representative spectrum in toluene was displayed. (B) Fluorescence wavelength-shifting responses of conjugate FR3 to hydration in water−dioxane mixtures. Growing water percentage results in a bathochromic shift of the emission maximum. Excitation wavelength was at 395 nm. (C) Fluorescence of FR3 at different concentration ratios of SM/CL LUVs (20 mM phosphate buffer, pH 7.4). Excitation wavelength was 390 nm. Concentration of FR3 was 0.4 μM. (D) Nonlinear fit curve of FR3 fluorescence intensities as a function of SM/CL LUV concentration ratios.

FR3 and FR5 are more photostable than FR4 and FR0 probably because they bear polar head groups, which increase the environment polarity of the dyes in xylene. The latter should decrease photodegradation, because a more polar environment decreases the probability of triplet formation, which is a common source of photobleaching of solvatochromic dyes.1,53 In addition, FR4 is less photostable than FR0, FR3, and FR5 in lipid vesicles. As FR4 is the only dye bearing a hydrophobic chain from the carbonyl size of fluorophore, its location and orientation in the lipid bilayer could be very different from those of the three other dyes, thus favoring photodegradation. In polar media like ethanol (Supporting Information), the dyes demonstrated high photostability in line with an increase of the singlet−triplet energy gap, which creates a disadvantage for the intersystem crossing and production of free oxygen radical species.34,52 As the previously reported dye PA, FR3 and FR5 demonstrate superior photostability with respect to Laurdan. Imaging GUVs with FR5. The significant shifts in emission of the new probes in Lo and Ld phases open the possibility for ratiometric imaging of lipid order in membranes by recording the emission intensities at two different wavelengths. FR5 was selected to image GUVs made of DOPC and DOPC/SM/CL since it showed the highest specificity to plasma membranes in cells (vide inf ra). GUVs with a ternary mixture of lipids is a simple model for imaging the coexistence of lipid domains. FR5

Table 2. Photophysical Properties of Probes FR3−5 in LUVs of Different Compositionsa composition DOPC LUVs

DOPC/CL LUVs (1:0.9)

SM/CL LUVs (1:0.9)

a

λ, Δλ [nm]a Φ[%]a

FR3

FR4

FR5

λAbs λEmb Δλ Φ λAbs λEmb Δλ Φ λAbs λEmb Δλ Φ

390 510 120 57 390 500 110 64 384 482 98 51

410 549 139 24 406 537 131 28 408 488 80 63

399 514 115 59 397 506 109 67 403 493 90 99

As in Table 1. bExcitation wavelength was 390 nm.

In xylene, the emission decreased by less than 10% for FR3 and FR5, by about 30% for FR4 and FR0, and by 50% for Prodan after 1 h of illumination. Once bound to DOPC vesicles, the intensities decreased by less than 10% for FR3, FR5, and FR0 but 55% for FR4 and Prodan after 1 h. These results confirm the superior photostability of the fluorene dyes as compared to Prodan. Noticeably, FR3 and FR5 demonstrate the highest photostability in both tested conditions. In xylene, 3026

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030

Articles

ACS Chemical Biology

Figure 2. Fluorescence spectra of FR3−5 (150 nM) in different LUVs (100 μM) in 20 mM phosphate buffer (pH 7.4). Black (DOPC), red (DOPC/CL), blue (SM/CL), green (DOPC/SM/CL). Excitation wavelength was 390 nm. All the spectra were normalized to their emission maximum.

Figure 3. Photodegradation curves of FR0, FR3−5, and Prodan as a function of time in (A) xylene and in the presence of (B) LUVs (DOPC). Emission and excitation slits set to 8 nm. Photodegradation was performed at 400 nm with the xenon lamp of the spectrofluorometer.

0.1 μM but gave also high quality images at concentrations as low as 25 nM, which is at least 80 times lower than the concentration used for Laurdan.54 Imaging Cells with FR3−5. Fluorescence microscopy was then used to examine the staining of cells with the probes. Adherent HeLa cells were incubated with FR3−5 probes at a concentration of 0.1 μM and imaged after 5−10 min of incubation. FR3−5 presented noticeable differences in their internalization and intracellular distribution as evidenced in Figure 5. FR5 displayed bright fluorescence exclusively in plasma membranes (colored in magenta, Figure 5E) without internalization of the dye. The ratiometric fluorescence image showed a uniform magenta pseudocolor corresponding to the Lo phase in GUVs that confirms the domination of the Lo phase in the cell plasma membranes as previously observed with Laurdan, NR12S, and other probes.30,55,56 In contrast, FR3 and FR4 dyes penetrate into the cells and mainly stain the cytoplasm (Figure 5A−D). It should be noted that the fluorescence intensity of FR4 in cells was relatively low, in line with its lower photostability and brightness. Another reason for low brightness could be its high hydrophobicity, so that its aggregation/precipitation was faster than binding to cell membranes. The ratio images further suggest that the intracellular compartments stained in green pseudocolor correspond to less ordered membrane phases of the endoplasmic reticulum and other membrane-containing organelles by contrast to the magenta pseudocolor of the cell plasma membranes (Figure 5A). The results are in line with those obtained recently with internalizing solvatochromic Laurdan and push−pull pyrene probe PA.34,54,55 Altogether, our results support FR5 as a highly useful probe for the exclusive staining of the plasma membrane at the outer leaflet. On the other hand, FR3 can stain both cell

was added to adherent GUVs at a concentration of 0.1 μM and studied without washing by confocal microscopy to evaluate its capability to visualize Ld and Lo phases. The images were processed by collecting the emission of the blue and green regions separated at 500 nm (Figures 4A,B). Ratiometric

Figure 4. Confocal fluorescence images of DOPC/SM/CL GUVs stained with FR5. Images were obtained by collecting the emission of the (A) blue channel and (B) green channel. (C) Ratiometric green/ blue imaging of GUVs. The ratios of intensities are displayed by using the color code given in the right scale. Scale bar = 20 μm.

images were obtained by dividing the images of the green channel by the blue ones (Figure 4C). FR5 distinguishes well the Lo and Ld phases through variations in the green/blue ratio signal. Thus, separated Lo (magenta) and Ld (yellow-green) domains were clearly observed. A control experiment with DOPC vesicles confirms this interpretation (Supporting Information) as the membranes appear in green pseudocolor corresponding to the disordered phase observed in the ternary mixture (Figure 4C). FR5 stains only the membrane, providing bright images with low background fluorescence (Supporting Information). Thus, the FR5 fluorene dye performs comparably to Laurdan to distinguish lipid domains in model LUVs but with superior brightness.30 Indeed, FR5 was employed at a concentration of 3027

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030

Articles

ACS Chemical Biology

and PA). These probes will enable effective excitation with a common violet laser (405 nm) and ratiometric imaging in the blue-green spectral region, which was not possible before for the common membrane probes (UV excitation for Laurdan and green-red emission of PA). This means that they should be compatible with red fluorescent proteins (e.g., mCherry, mStrawberry, mRoseberry, mKate, mPlum, etc.).57 As the parent push−pull fluorene dye displays a high two-photon absorption cross-section (e.g., 400 GM at 800 nm),37 we expect that these new probes will be attractive candidates for twophoton excitation and microscopy as well. Moreover, a relatively high photostability and high quantum yield (generally >60%) and extinction coefficient (>35 000 M−1cm−1) makes the new fluorene-based probes especially attractive for emerging spectrally resolved super-resolution techniques, such as points accumulation for imaging in nanoscale topography (PAINT) and stochastic optical reconstruction microscopy (STORM).58,59

■ ■

METHODS

See Supporting Information for a detailed description of the experimental methods.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acschembio.7b00658. Photophysical, photodegradation, and imaging investigations in addition to the experimental protocols for the synthesis of intermediates 1−4 and FR3−5 conjugates; 1 H NMR, 13C NMR and (in part) 1H−13C COSY, 1 H−13C HMQC, and 1H−13C HMBC spectra of FR3−5 conjugates (PDF)



Figure 5. Confocal fluorescence images of HeLa cells stained with FR3 (A and B), FR4 (C and D), and FR5 (E and F). Concentration of the probe was 0.1 μM. Left column: ratiometric green/blue imaging. The intensity ratios are displayed by using the color code given in the right scale. Right column: images obtained by collecting the fluorescence intensity recorded for both channels. Scale bar = 20 μm.

AUTHOR INFORMATION

Corresponding Author

*Tel.: (+33) 492 076 152. Fax: (+33) 492 076 151. E-mail: [email protected]. ORCID

Mayeul Collot: 0000-0002-8673-1730 Yves Mély: 0000-0001-7328-8269 Andrey S. Klymchenko: 0000-0002-2423-830X Alain Burger: 0000-0001-5709-8086

plasma membranes and cytoplasm, showing a clear contrast in their lipid organization.



CONCLUSION To conclude, we have demonstrated the ability of the push− pull fluorene dyes to operate as environmentally sensitive probes for monitoring the biophysical properties of cell membranes. Based on a rational design, three novel fluorescent conjugates were synthesized using a straightforward and practical strategy. This work validated the versatility of the introduced propiolyl linker in bio-orthogonal labeling. It allowed the introduction of different substituents to tune the cell binding and internalization and gave products without compromising the spectroscopic features for biological applications. It was found that the deoxyribosyl head favored cell internalization and staining of intracellular membranes, whereas the amphiphilic anchor group ensured specific plasma membrane staining. The first unique properties of these probes are their blue spectral range and large Stokes shift, while conserving sensitivity to lipid phases (as for Laurdan, NR12S,

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This research was supported by the French Government through the Ph.D. grant of J.S., the ANR (ANR-12-BS08-000302), PACA région (DNAfix-2014-02862 and 2014-07199), and the FRM (DCM20111223038). We would like to acknowledge J.-M. Guigonis for the mass spectroscopy, R. Vauchelles from PIQ platform for help with confocal imaging, and A. Demchenko for critical reading.



REFERENCES

(1) Ishikawa-Ankerhold, H. C., Ankerhold, R., and Drummen, G. P. C. (2012) Advanced Fluorescence Microscopy TechniquesFRAP, FLIP, FLAP, FRET and FLIM. Molecules 17, 4047−4132.

3028

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030

Articles

ACS Chemical Biology (2) Drummen, G. (2012) Fluorescent Probes and Fluorescence (Microscopy) Techniques  Illuminating Biological and Biomedical Research. Molecules 17, 14067−14090. (3) Tsien, R. Y. (2009) Constructing and Exploiting the Fluorescent Protein Paintbox (Nobel Lecture). Angew. Chem., Int. Ed. 48, 5612− 5626. (4) Resch-Genger, U., Grabolle, M., Cavaliere-Jaricot, S., Nitschke, R., and Nann, T. (2008) Quantum dots versus organic dyes as fluorescent labels. Nat. Methods 5, 763−775. (5) Wysocki, L. M., and Lavis, L. D. (2011) Advances in the chemistry of small molecule fluorescent probes. Curr. Opin. Chem. Biol. 15, 752−759. (6) Yang, Y., Zhao, Q., Feng, W., and Li, F. (2013) Luminescent Chemodosimeters for Bioimaging. Chem. Rev. 113, 192−270. (7) Kubota, R., and Hamachi, I. (2015) Protein recognition using synthetic small-molecular binders toward optical protein sensing in vitro and in live cells. Chem. Soc. Rev. 44, 4454−4471. (8) Demchenko, A. P. (2013) Nanoparticles and nanocomposites for fluorescence sensing and imaging. Methods Appl. Fluoresc. 1, 022001. (9) Wu, C., and Chiu, D. T. (2013) Highly Fluorescent Semiconducting Polymer Dots for Biology and Medicine. Angew. Chem., Int. Ed. 52, 3086−3109. (10) Kobayashi, H., Ogawa, M., Alford, R., Choyke, P. L., and Urano, Y. (2010) New Strategies for Fluorescent Probe Design in Medical Diagnostic Imaging. Chem. Rev. 110, 2620−2640. (11) Yuan, L., Lin, W., Zheng, K., and Zhu, S. (2013) FRET-Based Small-Molecule Fluorescent Probes: Rational Design and Bioimaging Applications. Acc. Chem. Res. 46, 1462−1473. (12) Demchenko, A. P. (2005) Optimization of fluorescence response in the design of molecular biosensors. Anal. Biochem. 343, 1−22. (13) Klymchenko, A. S. (2017) Solvatochromic and Fluorogenic Dyes as Environment-Sensitive Probes: Design and Biological Applications. Acc. Chem. Res. 50, 366−375. (14) Karpenko, I. A., Collot, M., Richert, L., Valencia, C., Villa, P., Mély, Y., Hibert, M., Bonnet, D., and Klymchenko, A. S. (2015) Fluorogenic Squaraine Dimers with Polarity-Sensitive Folding As Bright Far-Red Probes for Background-Free Bioimaging. J. Am. Chem. Soc. 137, 405−412. (15) Lingwood, D., and Simons, K. (2010) Lipid rafts as a membrane-organizing principle. Science 327, 46−50. (16) Klymchenko, A. S., and Kreder, R. (2014) Fluorescent probes for lipid rafts: from model membranes to living cells. Chem. Biol. 21, 97−113. (17) Bagatolli, L. A. (2006) To see or not to see: lateral organization of biological membranes and fluorescence microscopy. Biochim. Biophys. Acta, Biomembr. 1758, 1541−1556. (18) Simons, K., and Ikonen, E. (1997) Functional rafts in cell membranes. Nature 387, 569−572. (19) Pike, L. J. (2006) Rafts defined: a report on the Keystone symposium on lipid rafts and cell function. J. Lipid Res. 47, 1597− 1598. (20) Demchenko, A. P., Mély, Y., Duportail, G., and Klymchenko, A. S. (2009) Monitoring biophysical properties of lipid membranes by environment-sensitive fluorescent probes. Biophys. J. 96, 3461−3470. (21) Dal Molin, M., Verolet, Q., Soleimanpour, S., and Matile, S. (2015) Mechanosensitive Membrane Probes. Chem. - Eur. J. 21, 6012− 6021. (22) Su, D., Teoh, C. L., Wang, L., Liu, X., and Chang, Y.-T. (2017) Motion-induced change in emission (MICE) for developing fluorescent probes. Chem. Soc. Rev. 46, 4833−4844. (23) Kuimova, M. K., Yahioglu, G., Levitt, J. A., and Suhling, K. (2008) Molecular Rotor Measures Viscosity of Live Cells via Fluorescence Lifetime Imaging. J. Am. Chem. Soc. 130, 6672−6673. (24) López-Duarte, I., Vu, T. T., Izquierdo, M. A., Bull, J. A., and Kuimova, M. K. (2014) A molecular rotor for measuring viscosity in plasma membranes of live cells. Chem. Commun. 50, 5282−5284. (25) Dal Molin, M., Verolet, Q., Colom, A., Letrun, R., Derivery, E., Gonzalez-Gaitan, M., Vauthey, E., Roux, A., Sakai, N., and Matile, S.

(2015) Fluorescent Flippers for Mechanosensitive Membrane Probes. J. Am. Chem. Soc. 137, 568−571. (26) Soleimanpour, S., Colom, A., Derivery, E., Gonzalez-Gaitan, M., Roux, A., Sakai, N., and Matile, S. (2016) Headgroup engineering in mechanosensitive membrane probes. Chem. Commun. 52, 14450− 14453. (27) Kim, H. M., Choo, H.-J., Jung, S.-Y., Ko, Y.-G., Park, W.-H., Jeon, S.-J., Kim, C. H., Joo, T., and Cho, B. R. (2007) A Two-Photon Fluorescent Probe for Lipid Raft Imaging: C-Laurdan. ChemBioChem 8, 553−559. (28) Parasassi, T., Gratton, E., Yu, W. M., Wilson, P., and Levi, M. (1997) Two-photon fluorescence microscopy of laurdan generalized polarization domains in model and natural membranes. Biophys. J. 72, 2413−2429. (29) Jin, L., Millard, A. C., Wuskell, J. P., Dong, X., Wu, D., Clark, H. A., and Loew, L. M. (2006) Characterization and Application of a New Optical Probe for Membrane Lipid Domains. Biophys. J. 90, 2563− 2575. (30) Kucherak, O. A., Oncul, S., Darwich, Z., Yushchenko, D. A., Arntz, Y., Didier, P., Mély, Y., and Klymchenko, A. S. (2010) Switchable Nile Red-Based Probe for Cholesterol and Lipid Order at the Outer Leaflet of Biomembranes. J. Am. Chem. Soc. 132, 4907− 4916. (31) Kreder, R., Pyrshev, K. A., Darwich, Z., Kucherak, O. A., Mély, Y., and Klymchenko, A. S. (2015) Solvatochromic Nile Red Probes with FRET Quencher Reveal Lipid Order Heterogeneity in Living and Apoptotic Cells. ACS Chem. Biol. 10, 1435−1442. (32) Sezgin, E., Schneider, F., Zilles, V., Urbančič, I., Garcia, E., Waithe, D., Klymchenko, A. S., and Eggeling, C. (2017) PolaritySensitive Probes for Superresolution Stimulated Emission Depletion Microscopy. Biophys. J. 113, 1321−1330. (33) Darwich, Z., Klymchenko, A. S., Kucherak, O. A., Richert, L., and Mély, Y. (2012) Detection of apoptosis through the lipid order of the outer plasma membrane leaflet. Biochim. Biophys. Acta, Biomembr. 1818, 3048−3054. (34) Niko, Y., Didier, P., Mély, Y., Konishi, G.-I., and Klymchenko, A. S. (2016) Bright and photostable push-pull pyrene dye visualizes lipid order variation between plasma and intracellular membranes. Sci. Rep. 6, 18870. (35) Weber, G., and Farris, F. J. (1979) Synthesis and spectral properties of a hydrophobic fluorescent probe: 6-propionyl-2(dimethylamino) naphthalene. Biochemistry 18, 3075−3078. (36) Niko, Y., Kawauchi, S., and Konishi, G.-I. (2013) Solvatochromic Pyrene Analogues of Prodan Exhibiting Extremely High Fluorescence Quantum Yields in Apolar and Polar Solvents. Chem. Eur. J. 19, 9760−9765. (37) Kucherak, O. A., Didier, P., Mély, Y., and Klymchenko, A. S. (2010) Fluorene Analogues of Prodan with Superior Fluorescence Brightness and Solvatochromism. J. Phys. Chem. Lett. 1, 616−620. (38) Shaya, J., Fontaine-Vive, F., Michel, B. Y., and Burger, A. (2016) Rational Design of Push-Pull Fluorene Dyes: Synthesis and StructurePhotophysics Relationship. Chem. - Eur. J. 22, 10627−10637. (39) Dziuba, D., Postupalenko, V. Y., Spadafora, M., Klymchenko, A. S., Guérineau, V., Mély, Y., Benhida, R., and Burger, A. (2012) A Universal Nucleoside with Strong Two-Band Switchable Fluorescence and Sensitivity to the Environment for Investigating DNA Interactions. J. Am. Chem. Soc. 134, 10209−10213. (40) Kuznetsova, A. A., Kuznetsov, N. A., Vorobjev, Y. N., Barthes, N. P. F., Michel, B. Y., Burger, A., and Fedorova, O. S. (2014) New environment-sensitive multichannel DNA fluorescent label for investigation of the protein-DNA interactions. PLoS One 9, e100007. (41) Kilin, V., Gavvala, K., Barthes, N. P. F., Michel, B. Y., Shin, D., Boudier, C., Mauffret, O., Yashchuk, V., Mousli, M., Ruff, M., Granger, F., Eiler, S., Bronner, C., Tor, Y., Burger, A., and Mély, Y. (2017) Dynamics of Methylated Cytosine Flipping by UHRF1. J. Am. Chem. Soc. 139, 2520−2528. (42) Shaya, J., Deschamps, M.-A., Michel, B. Y., and Burger, A. (2016) Air-Stable Pd Catalytic Systems for Sequential One-Pot 3029

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030

Articles

ACS Chemical Biology Synthesis of Challenging Unsymmetrical Aminoaromatics. J. Org. Chem. 81, 7566−7573. (43) Bureš, F. (2014) Fundamental aspects of property tuning in push−pull molecules. RSC Adv. 4, 58826−58851. (44) Shvadchak, V. V., Kucherak, O., Afitska, K., Dziuba, D., and Yushchenko, D. A. (2017) Environmentally sensitive probes for monitoring protein-membrane interactions at nanomolar concentrations. Biochim. Biophys. Acta, Biomembr. 1859, 852−859. (45) Rolland, V., Kotera, M., and Lhomme, J. (1997) Convenient Preparation of 2-Deoxy-3,5-di-O- p-toluoyl-α-D- erythro-pentofuranosyl Chloride. Synth. Commun. 27, 3505−3511. (46) Maity, A., Choi, J.-S., Teets, T. S., Deligonul, N., Berdis, A. J., and Gray, T. G. (2013) Cyclometalated Iridium(III) Complexes with Deoxyribose Substituents. Chem. - Eur. J. 19, 15924−15932. (47) Collot, M., Kreder, R., Tatarets, A. L., Patsenker, L. D., Mély, Y., and Klymchenko, A. S. (2015) Bright fluorogenic squaraines with tuned cell entry for selective imaging of plasma membrane vs. endoplasmic reticulum. Chem. Commun. 51, 17136−17139. (48) Reichardt, C. (1994) Solvatochromic dyes as solvent polarity indicators. Chem. Rev. 94, 2319−2358. (49) Melhuish, W. H. (1961) Quantum Efficiencies of Fluorescence of Organic Substances: Effect of Solvent and Concentration of the Fluorescent Solute. J. Phys. Chem. 65, 229−235. (50) Ormson, S. M., Brown, R. G., Vollmer, F., and Rettig, W. (1994) Switching between charge-and proton-transfer emission in the excited state of a substituted 3-hydroxyflavone. J. Photochem. Photobiol., A 81, 65−72. (51) Marini, A., Muñoz-Losa, A., Biancardi, A., and Mennucci, B. (2010) What is Solvatochromism? J. Phys. Chem. B 114, 17128−17135. (52) Sezgin, E., Sadowski, T., and Simons, K. (2014) Measuring Lipid Packing of Model and Cellular Membranes with Environment Sensitive Probes. Langmuir 30, 8160−8166. (53) Zheng, Q., Juette, M. F., Jockusch, S., Wasserman, M. R., Zhou, Z., Altman, R. B., and Blanchard, S. C. (2014) Ultra-stable organic fluorophores for single-molecule research. Chem. Soc. Rev. 43, 1044− 1056. (54) Golfetto, O., Hinde, E., and Gratton, E. (2013) Laurdan Fluorescence Lifetime Discriminates Cholesterol Content from Changes in Fluidity in Living Cell Membranes. Biophys. J. 104, 1238−1247. (55) Owen, D. M., Rentero, C., Magenau, A., Abu-Siniyeh, A., and Gaus, K. (2011) Quantitative imaging of membrane lipid order in cells and organisms. Nat. Protoc. 7, 24−35. (56) Kilin, V., Kilin, V., Glushonkov, O., Glushonkov, O., Herdly, L., Herdly, L., Klymchenko, A., Klymchenko, A., Richert, L., and Mély, Y. (2015) Fluorescence lifetime imaging of membrane lipid order with a ratiometric fluorescent probe. Biophys. J. 108, 2521−2531. (57) Shaner, N. C., Steinbach, P. A., and Tsien, R. Y. (2005) A guide to choosing fluorescent proteins. Nat. Methods 2, 905−909. (58) Bongiovanni, M. N., Godet, J., Horrocks, M. H., Tosatto, L., Carr, A. R., Wirthensohn, D. C., Ranasinghe, R. T., Lee, J. E., Ponjavic, A., Fritz, J. V., Dobson, C. M., Klenerman, D., and Lee, S. F. (2016) Multidimensional super-resolution imaging enables surface hydrophobicity mapping. Nat. Commun. 7, 13544. (59) Moon, S., Yan, R., Kenny, S. J., Shyu, Y., Xiang, L., Li, W., and Xu, K. (2017) Functional Super-Resolution Microscopy Reveals Nanoscale Compositional Heterogeneity in Live-Cell Membranes. J. Am. Chem. Soc. 139, 10944−10947.

3030

DOI: 10.1021/acschembio.7b00658 ACS Chem. Biol. 2017, 12, 3022−3030