Cy5 Dye Cassettes Exhibit Through-Bond Energy Transfer and Enable Ratiometric Fluorescence Sensing

Abstract

The chemosensor literature contains many reports of fluorescence sensing using polyaromatic hydrocarbon fluorophores such as pyrene, tetraphenylethylene, or polyaryl(ethynylene), where the fluorophore is excited with ultraviolet light (<400 nm) and emits in the visible region of 400–500 nm. There is a need for general methods that convert these “turn-on” hydrocarbon fluorescent sensors into ratiometric sensing paradigms. One simple strategy is to mix the responsive hydrocarbon sensor with a second non-responsive dye that is excited by ultraviolet light but emits at a distinctly longer wavelength and thus acts as a reference signal. Five new cyanine dye cassettes were created by covalently attaching a pyrene, tetraphenylethylene, or biphenyl(ethynylene) component as ultraviolet-absorbing energy donor directly to the pentamethine chain of a deep-red cyanine (Cy5) energy acceptor. Fluorescence emission studies showed that these Cy5-cassettes exhibited large pseudo-Stokes shifts and high through bond energy transfer efficiencies upon excitation with ultraviolet light. Practical potential was demonstrated with two examples of ratiometric fluorescence sensing using a single ultraviolet excitation wavelength. One example mixed a Cy5-cassette with a pyrene-based fluorescent indicator that responded to changes in Cu²⁺ concentration, and the other example mixed a Cy5-cassette with the fluorescent pH sensing dye, pyranine.

Graphical Abstract

INTRODUCTION

The last two decades has seen major and ongoing development of intensiometric fluorescent sensors that respond to the presence of specific analytes with ‘turn on” fluorescence.^{1 2 3} Many of the classic examples of fluorescence sensing involve polyaromatic hydrocarbon fluorophores such as pyrene,^{4 5 6} tetraphenylethylene (TPE),^{7 8} and polyaryl(ethynylene),^{9 10} and in each case the fluorophore is excited with ultraviolet light (<400 nm), with subsequent emission in the visible region of 400–500 nm. It is well established that many quantitative sensing procedures are more reliable if they monitor a change in the ratio of two emission signals (ratiometric sensing),^{11 12} and thus there is a need for general methods that convert the classic “turn-on” hydrocarbon fluorescent sensors into ratiometric sensing paradigms. One simple strategy is to mix the responsive hydrocarbon sensor with a second non-responsive dye that emits at a distinctly longer wavelength and thus it acts as a reference signal (Scheme 1a).^13,14 In addition, it is often a technical advantage if the two fluorescent signals for this dye mixture can be produced by exciting the sample at a single wavelength (Scheme 1b).¹⁵ This requires the second non-responsive, long wavelength dye to be excited with ultraviolet light which means the dye must exhibit a reasonably intense ultraviolet absorption band and also a very large Stokes shift^{16 17 18} (or pseudo-Stokes shift^{19 20}) which is a very unusual combination.

Scheme 1.

Schematic illustration of (a) intensiometric (turn on) fluorescence sensing using a sample containing a responsive aromatic hydrocarbon fluorescent sensor (e.g., pyrene, TPE or biphenyl(ethynylene)), and (b) how ratiometric sensing is generated by including a non-responsive, deep-red Cy5-cassette dye in the same sample. (c) Chemical structures of Cy5-cassettes 1P, 1T, 1B, 2P, and 2T, along with control Cy5 dyes 1H and 2H.

One way to produce a fluorescent dye with a large pseudo-Stokes shift is to design a fluorescent dye cassette, that is, a two-component molecule that has a short wavelength excitation component (energy donor) covalently attached to a long wavelength emission component (energy acceptor). After excitation of the donor, there is intramolecular energy transfer to the emissive acceptor. The mechanism for energy transfer can be through space (Förster resonance energy transfer, FRET)^21,22 or through bonds (through bond energy transfer, TBET). Unlike FRET, the TBET process does not require donor/acceptor spectral overlap, but efficient TBET dye cassettes must be a conjugated in way that connects the π-electrons of the donor and acceptor components.^23–25

The deep-red, pentamethine cyanine (Cy5) dye with fluorescence emission in the wavelength range 660–720 nm matches a common filter setting on many fluorescence detectors and cameras, and therefore is a good choice as the energy acceptor in a dye cassette. But examples of TBET using cyanine cassette dyes are rare, and the very few examples have attached the donor component to the terminal heterocycles of the cyanine acceptor.^{19 20 26 27} Here we describe a different molecular design strategy that attaches the polyaromatic hydrocarbon donor directly to the center of a Cy5 pentamethine chain. Specifically, we report three new classes of cyanine TBET cassettes, the pyrene-Cy5 analogues 1P and 2P, the TPE-Cy5 analogues 1T and 2T, and the biphenyl(ethynylene)-Cy5 1B (Scheme 1c). We provide evidence for efficient TBET, and demonstrate potential utility with two examples of ratiometric fluorescence sensing using a method that employs a single excitation wavelength.

RESULTS AND DISCUSSION

Fluorophore Synthesis

The Cy5-cassettes (1P, 1T, 2P, and 2T) were prepared by conducting palladium mediated Suzuki coupling reactions that attached each polyaromatic donor chromophore to the central 3’ position of a cyanine pentamethine dye,²⁸ while 1B was prepared by conducting palladium mediated Sonogashira coupling reaction (Scheme 2). The control Cy5 dyes 1H and 2H were commercially available.

Scheme 2.

Synthesis of Cy5 cassettes 1P, 2P, 1T, 2T, and 1B.

Spectral Studies in Methanol

Initial spectral studies focused on the pyrene and TPE Cy5-cassettes (1P, 1T, 2P, and 2T) and control Cy5 dyes 1H and 2H in methanol, where the compounds are readily soluble, chemically stable, and there is no self-aggregation (see Table 1 for a list of spectral data). The absorption spectra exhibit an intense deep-red absorption band corresponding the Cy5 component and a weak ultraviolet absorption band corresponding to the indole or benzoindole units at the ends of the pentamethine chain, and also the appended pyrene or TPE components on the Cy5-cassettes (Figure 1).

Table 1.

Spectra data in methanol at room temperature

Cy5-Cassette	Absorption			Emission
	${\mathit{\lambda }}_{\mathit{m}\mathit{a}\mathit{x}}^{\mathit{a}\mathit{b}\mathit{s}}$ [nm][a]	${\mathit{\lambda }}_{\mathit{m}\mathit{a}\mathit{x}}^{\mathit{a}\mathit{b}\mathit{s}}$ [nm][b]	ε [M−1cm−1]	${\mathit{\lambda }}_{\mathit{m}\mathit{a}\mathit{x}}^{\mathit{e}\mathit{m}}$ [nm][c]	ΦF [%][d]	ETE [%][e]	FEF[f]	Δλ [nm][g]
1P	342	645	159 000	664	20.3±0.9	97	9.0	322
2P	374	682	188 000	703	20.0±0.7	84	1.6	329
1T	339	647	142 000	668	18.4±0.4	93	4.6	329
2T	373	684	173 000	707	23.8±0.5	78	1.5	334
1B	417	652	149 000	673	27.2±0.9	-[h]	21.6	256

^[a]Absorption maxima of donor.

^[b]Absorption maxima of acceptor.

^[c]Emission maxima of acceptor.

^[d]Absolute fluorescence quantum yield of the Cy5 component measured directly by photon counting.

^[e]Energy transfer efficiency = [1-(I_DA/I_D)]x100. I_DA is the integrated emission of the donor; I_D is the integrated emission of the donor as free molecule.²⁹

^[f]Fluorescence enhancement factor.

^[g]Pseudo-Stokes shift of Cy5-cassette.

^[h]ETE of 1B cannot be obtained by this method.

Figure 1.

Absorption spectra of each dye (4 μM) in methanol.

Clear evidence of TBET was gained by comparing separate methanol solutions of 1P, 1H, or pyrene (Figure 2a); 2P, 2H, or pyrene (Figure 2b); 1T, 2H, or TPE (Figure 2c); 2T, 2H, or TPE (Figure 2d). As expected, ultraviolet excitation of free pyrene or TPE did not produce any deep-red fluorescence. In the case of control Cy5 dyes 1H and 2H, ultraviolet excitation produced moderate deep-red emission, presumably due to excitation of the indole or benzoindole units at the ends of the pentamethine chains. A comparison of the dyes with indole termini, that is 1P or 1T with 1H (Figures 2a,c) produced a fluorescence enhancement factor (FEF) of 9.0 and 4.6, respectively. A similar comparison of the dyes with benzoindole termini, that is 2P or 2T with 2H (Figures 2b,d) produced a FEF of 1.6 and 1.5, respectively. These smaller FEF values reflect increased TBET to the Cy5 acceptor from the terminal benzoindole units. The energy transfer efficiency (ETE) for each Cy5 cassette was quantified by measuring the integrated fluorescence emission of the donor component as a free molecule (i.e., free pyrene or TPE) and when the donor was part of a Cy5-cassette (i.e., 1P, 1T, 2P, or 2T). In all cases the ETE values in methanol were very high (Table 1).

Figure 2.

Evidence for TBET. Emission spectra of (a) 1P, 1H, and pyrene (λ_ex = 350 nm), (b) 2P, 2H, and pyrene (λ_ex = 370 nm), (c) 1T, 1H, and TPE (λ_ex = 350 nm), and (d) 2T, 2H, and TPE (λ_ex = 370 nm). Each dye is 4 μM in methanol.

A manifestation of the strong absorption of Cy5 dyes is the appearance of fluorescence inner filter effects even at relatively low, micromolar dye concentrations.^{30 31} However, the dual excitation of Cy5 cassettes provides a way to mitigate the inner effect. As shown in Figure 3, the absorption spectra for 1T and 2T in methanol exhibit linear Beer-Lambert relationships for concentration up to 10 μM, but the emission spectra with 630 nm excitation exhibit classic inner filter effects. That is, the Cy5 fluorescence intensity is a maximum at 4 μM and decreases with higher concentration (attenuation of the excitation beam due to the primary inner filter effect) along with a noticeable red-shift in the peak maxima wavelength (reabsorption of the deep-red emission due secondary inner filter effect). However, these inner filter effects are avoided when the Cy5 emission spectra for 1T or 2T are generated with 350 nm where there is much less absorption of the ultraviolet excitation light by other Cy5-cassette molecules in the solution. The same trends were observed with solutions of 1P and 2P in methanol (Figure S1).

Figure 3.

(a) Absorption spectra of 1T (Insert: Beer-Lambert plot of absorption over 1–10 μM) in MeOH. Emission spectra of 1T in MeOH: (b) λ_ex = 630 nm, and (c) λ_ex = 350 nm. (d) Absorption spectra of 2T (Insert: Beer-Lambert plot of absorption over 1–10 μM) in MeOH. Emission spectra of 2T in MeOH: (e) λ_ex = 660 nm, and (f) λ_ex = 370 nm.

Spectral Studies in Aqueous Solution

Deep-red cyanine dyes are known to self-aggregate in aqueous solution and form non-fluorescent aggregates. On the other hand, pyrene dyes can form excimers in self-aggregation conditions,^{4 5 6 32–36} and TPE is also well-known for aggregation induced enhancement (AIE).^{7 8} To probe the change in fluorescent properties upon dye cassette self-aggregation, we acquired absorption (Figure 4a–d) and emission (Figure S2–S5) spectra for each dye in water/methanol mixtures. In the case of the relatively hydrophobic dyes, 1P or 1T, the transition from 100% methanol to 99.5% water induced minor spectral changes indicating little formation of dye self-aggregates. But with the more hydrophobic dyes, 2P or 2T, the transition from 100% methanol to 99.5% water induced large changes in the Cy5 absorption profiles, namely, conversion of an emissive Cy5 monomer absorption band at ~680 nm to a self-quenched H-aggregate band at ~620 or ~570 nm (Figure S6–S7). The photographs of illuminated vials in Figure 4e visually illustrate the contrast in AIE exhibited by TPE, where fluorescence is enhanced when there is ≤50% methanol, to the aggregation-caused-quenching (ACQ) exhibited by 2T where fluorescence is decreased when there is ≤30% methanol. Additional studies that compared absorption and emission spectra for the Cy5-cassettes in water and PBS solution consistently observed increased self-aggregation and loss of fluorescence in PBS solution (Figure S8). It is notable that these Cy5-cassettes exhibit ACQ and not AIE. The lack of AIE for self-aggregated 1T and 2T, each with an appended TPE component, is rare but matches the findings of other literature studies.^{37 38 39} In these cases, we assume there is energy transfer from the excited state TPE component to the appended Cy5 component which is non-emissive due to its intrinsic ACQ property.

Figure 4.

Absorption spectra for (a) 1P, (b) 2P, (c) 1T, and (d) 2T in binary solvent mixtures ranging from 100% methanol to 99.5% water. Each dye is 5 μM. (e) Photographs of vials containing 2T or tetraphenylethylene (TPE) in different methanol (MeOH)/water mixtures with illumination by ultraviolet light (365 nm).

Ratiometric Sensing of Copper(II) in Methanol

The propensity of pyrene to form excimers with visible emission is often exploited for fluorescence sensing.^{4 5 6} For example, the pyrene-based probe L is known to form an excimer complex (L₂•Cu²⁺) in the presence of Cu²⁺ with “turn-on” fluorescence at 430 nm (Figure 5a).³² By simply mixing “responsive” probe L with non-responsive cassette 2P, we achieved ratiometric sensing of Cu²⁺ with a single excitation wavelength of 380 nm. Shown in Figure 5b are emission spectra for a binary mixture of 2P and L in methanol treated with different concentrations of Cu²⁺. The fluorescence intensity ratio (F₄₃₀/F₇₁₀) increased over 24 hours by nearly 3.5-fold upon addition of the Cu²⁺ (Figure 5c). The 24-hour time period needed for the ratiometric signal to reach equilibrium is a technical drawback for some applications but not all, such as automated environmental testing protocols that can easily accommodate an overnight waiting period. This simple result is proof of concept that ratiometric fluorescence sensing with a single excitation wavelength is achieved when the non-responsive cassette 2P is mixed with a responsive aromatic hydrocarbon fluorescent sensor.

Figure 5.

(a) Responsive probe L forms an excimer complex with Cu²⁺ that “turns-on” fluorescence at 430 nm. (b) Emission spectra of non-responsive cassette 2P (10 μM) and responsive probe L (4 μM) mixed with different concentrations of Cu²⁺ (0−2 μM). The 1 mM of Cu²⁺ stock solution was prepared by dissolving CuSO₄·5H₂O in methanol at room temperature. Emission spectra was acquired 24 h after sample preparation to ensure the sample is at equilibrim. λ_ex = 380 nm. (c) Plot of fluorescence intensity ratio (F₄₃₀/F₇₁₀) versus [Cu²⁺]. Error bars represent standard deviation for triplicate experiments.

Ratiometric Sensing of Aqueous pH

To better demonstrate the scope and impact of these Cy5 cassettes we decided to create a ratiometric sensing system for measuring pH in a completely aqueous solution. Fluorescent pH sensors are broadly useful in many technology sectors such as food production, environmental science, and clinical diagnostics, in part because many biochemical assays are designed to produce a change in pH as an observable signal for assay read out.^{40 41 42} To ensure proper operation in water, we modified the sensing components in two ways. First, we decided to employ the sulfonated pyrene dye, pyranine (hydroxypyrene-1,3,6-trisulfonic acid) as a pH responsive sensor whose emission intensity at 510 nm (425 nm excitation) increases with pH.^{43 44} Second, we needed to ensure that the non-responsive, long wavelength reference dye was sufficiently soluble in water and that efficient TBET transfer could be achieved upon excitation with 425 nm light. To meet both criteria, we synthesize the Cy5-cassette 1B with an appended biphenyl(ethynylene) as the donor component and found that it exhibited a very high fluorescence enhancement factor in methanol reflecting efficient TEBT (Table 1, Figure S10). The absorption spectrum of 1B in water indicated some H-aggregation (Figure S9), and so we assessed the capacity of various water-soluble cyclodextrin host compounds to capture 1B as a more emissive monomer. We found that 2-hydroxyisopropyl β-cyclodextrin was quite effective at reducing 1B self-aggregation (Figure S11) and enabling efficient TBET with 425 nm excitation. As shown in Figure 6, the mixture of 1B, 2-hydroxyisopropyl β-cyclodextrin, and pyranine acted as a ratiometric fluorescent pH sensor with a near-linear change in output signal over the pH range of 6–8. Looking to future applications, it should be feasible to incorporate the binary mixture of 1B and pyranine within suitable nanoparticles and create robust nanoscale sensing probes for ratiometric diagnostics and multiplex microscopy using a single excitation wavelength.^{13,14 45 46}

Figure 6.

(a) Pyranine is a pH responsive probe that “turns-on” fluorescence at 510 nm (λ_ex= 425 nm) with increasing pH. (b) Fluorescence spectra of non-responsive cassette 1P (10 μM), pH responsive probe pyranine (1 μM), and 2-hydroxypropyl β-cyclodextrin (800 μM) in citrate buffer at different pH values. λ_ex = 425 nm, λ_em = 460–900 nm. (c) Plot of fluorescence intensity ratio (F₅₁₀/F₆₆₈) versus pH. Error bars represent standard deviation for triplicate experiments.

Conclusions

Five new Cy5-cassettes were created by directly attaching a pyrene, TPE, or biphenyl(ethynylene) component, as an ultraviolet-absorbing energy donor, to the pentamethine chain of a deep-red Cy5 acceptor. The fluorescent Cy5-cassettes exhibited large pseudo-Stokes shifts and high TBET efficiencies upon excitation with ultraviolet light. In predominantly aqueous solutions, the Cy5-cassettes exhibited aggregation-caused-quenching, but in the case of biphenyl(ethynylene) derivative 1B, this was mitigated by supramolecular complexation of the Cy5-cassette. Potential utility was demonstrated with two examples of ratiometric fluorescence sensing using a single excitation wavelength. One example mixed a Cy5-cassette (2P) with a pyrene-based fluorescent probe (L) that responded to changes in Cu²⁺ concentration, and the other example mixed a Cy5-cassette (1B) with the fluorescent pH sensor pyranine. To the best of our knowledge, these two examples are among the first demonstrations of ratiometric fluorescence sensing based on the TBET principle using a single excitation wavelength.^{14 15} The enabling chemical advance is the large pseudo-Stokes shifts generated by the Cy5-cassettes. This TBET approach is a more general than other literature ways to excite two dyes with a single wavelength, such as a binary mixture of one and two photon excitation dyes,⁴⁷ or a mixture that exploits energy upconversion within a nanoparticle.⁴⁸ In the long term, it should be possible to combine an analyte-responsive pyrene, TPE, or polyaryl(ethynylene) fluorescent indicator with a non-responsive Cy5-cassette within solid matrices or nanoscale assemblies and create a broad platform of ratiometric fluorescent sensors with single wavelength excitation.

Experimental Section

General materials and methods.

Organic reactions were performed under an atmosphere of argon using anhydrous solvents unless otherwise noted. Reagents and solvents including 1H and 2H were purchased from Sigma-Aldrich, VWR, Oakwood, Alfa Aesar, Ambeed, TCI America, Alfa Chemistry, and used without further purification. Column Chromatography was performed using Biotage Sfär C18 Duo columns (part # FSUD-0401). Reverse phase thin layer chromatography (TLC) experiments were performed on C18 TLC plates with F254s indicator (MilliporeSigma, part # 1.15683). ¹H NMR spectra and ¹³C NMR were recorded on a Bruker 500 NMR spectrometer installed with TopSpin^™ software. Multiplicities were given as singlet (s), doublet (d), doublet of doublets (dd), triplet (t) or multiplet (m). Spectra were visualized and analyzed using MestReNova^™ software (version 14.2). Chemical shifts are presented in ppm and referenced by residual solvent peaks. High-resolution mass spectrometry (HRMS) was performed using a time-of-flight (TOF) analyzer with electrospray ionization (ESI). Absorption spectra were recorded on an Evolution 201 UV/vis spectrometer with Thermo Insight software. Emission spectra were collected on a Horiba Fluoromax-4 fluorometer with FluoroEssence^™ software. Analyte solutions were prepared in HPLC grade water (Sigma-Aldrich), HPLC grade methanol (Sigma-Aldrich), or phosphate buffered saline (Thermo Fisher). All absorption and emission spectra were collected using quartz cuvettes at room temperature (1 mL, 1 cm path length; for emission spectra, slit width = 2 nm, unless stated otherwise).

Absolute quantum yields measurement.

Absolute quantum yields of 1B, 1P, 1T, 2P, and 2T were measured on a Horiba FluoromaxPlus spectrometer with an integrating sphere. Samples were excited at 640 nm for 1B, 1P and 1T, and 680 nm for 2T and 2P with optical density ≤ 0.05. First, all the photons were recorded with an integrating sphere after the excitation of a blank solvent reference, then the reference was replaced by a sample solution, and the spectrum (630–800 nm for 1B, 1P and 1T, and 670–800 nm for 2T and 2P) was acquired again. The quantum yield was calculated by the equation below:

$${\Phi }_F=\frac{P_{em}}{P_{abs}}=\frac{{\int }_{650}^{800}\left(F_{\mathit{\text{sample}}}-F_{\mathit{\text{blank}}}\right)\text{d}\lambda }{{\int }_{630}^{650}\left(E_{\mathit{\text{blank}}}-E_{\mathit{\text{sample}}}\right)\text{d}\lambda }\text{ for }\mathbf{1}\mathbf{B},\mathbf{1}\mathbf{P}\text{ and }\mathbf{1}\mathbf{T}$$

$${\Phi }_F=\frac{P_{em}}{P_{abs}}=\frac{{\int }_{690}^{800}\left(F_{\mathit{\text{sample}}}-F_{\mathit{\text{blank}}}\right)\text{d}\lambda }{{\int }_{670}^{690}\left(E_{\mathit{\text{blank}}}-E_{\mathit{\text{sample}}}\right)\text{d}\lambda }\text{ for }\mathbf{2}\mathbf{T}\text{ and }\mathbf{2}\mathbf{P}$$

where P is the number of photons, F is the fluorescence intensity and E is the intensity at the excitation wavelength. Experiments were conducted in triplicates with the reported absolute quantum yields corresponding to the mean value ± standard deviation.

General procedure for bromo-substituted pentamethine dyes 3 and 4 (Scheme S1).⁴⁹

A mixture of indolium salt (5 or 6) (2.633 mmol, 2.5 equiv.), N-((1E,2Z)-2-Bromo-3-(phenylamino)allylidene) benzenaminium bromide (1.053 mmol, 1.0 equiv.), sodium acetate (3.159 mmol, 3.0 equiv.), and acetic anhydride (5 mL, cat.) in EtOH (10 mL) was stirred in an oil bath at 70 °C for 16 h under Ar in the dark. The solvent was evaporated from the reaction mixture under reduced pressure and the residue was purified by reverse phase column chromatography (C18, 0–60% methanol in H₂O). The products were obtained as blue solids.

Purification by reverse phase column chromatography (C18, 0–60% methanol in H₂O) gave 3 (0.74 g, 98%). ¹H NMR (500 MHz, DMSO-d₆) δ 8.50 (d, J = 13.4 Hz, 2H), 7.67 (d, J = 6.1 Hz, 2H), 7.55 (d, J = 8.2 Hz, 2H), 7.43 (td, J = 7.7, 1.3 Hz, 2H), 7.30 (d, J = 7.4 Hz, 2H), 6.33 (d, J = 13.4 Hz, 1H), 4.18 (t, J = 7.4 Hz, 4H), 2.52–2.47 (m, 4H), 1.89–1.79 (m, 4H), 1.75–1.65 (m, 16H). ¹³C{¹H} NMR (125 MHz, DMSO-d₆) δ 174.2, 149.4, 141.9, 141.4, 128.6, 125.5, 122.5, 119.0, 115.8, 112.0, 102.2, 50.8, 49.5, 44.0, 26.7, 26.1, 22.6. HRMS (ESI-TOF) m/z: [M]⁻ Calcd for C₃₃H₄₀BrN₂O₆S₂⁻ 703.1517; Found 703.1508.

Purification by reverse phase column chromatography (C18, 0–60% methanol in H₂O) gave 4 (0.65 g, 74%). ¹H NMR (500 MHz, DMSO-d₆) δ 8.65 (d, J = 13.4 Hz, 2H), 8.26 (d, J = 8.6 Hz, 2H), 8.09 (t, J = 8.9 Hz, 4H), 7.88 (d, J = 9.0 Hz, 2H), 7.70 (t, J = 8.0 Hz, 2H), 7.54 (t, J = 7.6 Hz, 2H), 6.40 (d, J = 13.4 Hz, 2H), 4.31 (t, J = 7.4 Hz, 4H), 2.50–2.46 (m, 4H), 1.98 (s, 12H), 1.94–1.87 (m, 4H), 1.77–1.69 (m, 4H). ¹³C{¹H} NMR (125 MHz, DMSO-d₆) δ 175.2, 148.5, 139.6, 133.7, 131.6, 130.4, 130.0, 127.8, 127.4, 125.2, 122.2, 115.9, 112.1, 101.8, 51.2, 50.7, 44.2, 26.4, 26.3, 22.6. HRMS (ESI-TOF) m/z: [M]⁻ Calcd for C₄₁H₄₄BrN₂O₆S₂⁻ 803.1830; Found 803.1812.

General procedure for Cy5-cassettes.²⁸

To a solution of bromo-substituted Cy5 dyes (3 or 4) (0.110 mmol, 1.0 equiv.), commercially available boronic acids (pyren-2-ylboronic acid or (4-(1,2,2-triphenylvinyl)phenyl) boronic acid) (0.220 mmol, 2.0 equiv.), potassium fluoride (0.242 mmol, 2.2 equiv.) and cesium carbonate (0.242 mmol, 2.2 equiv.) in a mixture of EtOH (2 mL) and water (2 mL) under Ar at room temperature, was added palladium acetate (0.011 mmol, 0.1 equiv.) and triphenylphosphine (0.044 mmol, 0.4 equiv.). The reaction mixture was stirred in an oil bath at 60 °C for 16 h in the dark. Then boronic acid (0.033 mmol, 0.3 equiv.), palladium acetate (0.011 mmol, 0.1 equiv.) and triphenylphosphine (0.044 mmol, 0.4 equiv.) were added to the reaction mixture. It was stirred in an oil bath at 60 °C for additional 16 h in the dark. The reaction progress was monitored by reverse phase TLC. The solvent was then evaporated from the reaction mixture under vacuum. The residue was purified by reverse phase column chromatography (C18, 0–80% methanol in H₂O). The products were obtained as blue solids.

Purification by reverse phase column chromatography (C18, 0–80% methanol in H₂O) gave 1P (24 mg, 26%) m.p. > 150 °C (dec). ¹H NMR (500 MHz, DMSO-d₆) δ 8.68 (d, J = 14.1 Hz, 2H), 8.45 (d, J = 8.9 Hz, 2H), 8.37–8.31 (m, 4H), 8.24 (d, J = 9.0 Hz, 2H), 8.12 (t, J = 7.7 Hz, 0H), 7.65 (d, J = 7.4 Hz, 2H), 7.38–7.30 (m, 4H), 7.24 (ddd, J = 8.1, 6.4, 2.0 Hz, 2H), 5.89 (d, J = 14.1 Hz, 2H), 3.59 (t, J = 6.9 Hz, 2H), 2.36 (t, J = 7.4 Hz, 2H), 1.82 (s, 12H), 1.65–1.56 (m, 4H), 1.42–1.32 (m, 4H). ¹³C{¹H} NMR (125 MHz, DMSO-d₆) δ 173.0, 152.8, 142.0, 141.2, 134.4, 131.3, 130.8, 128.4, 127.7, 127.6, 126.8, 126.3, 125.2, 124.9, 123.7, 123.1, 122.4, 111.4, 101.4, 50.8, 49.1, 43.3, 27.1, 26.0, 22.5. HRMS (ESI-TOF) m/z: [M]⁻ Calcd for C₄₉H₄₉N₂O₆S₂⁻ 825.3038; Found 825.3034.

Purification by reverse phase column chromatography (C18, 0–80% methanol in H₂O) gave 1T (52 mg, 48%) m.p. > 150 °C (dec). ¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 8.43 (d, J = 14.1 Hz, 2H), 7.63 (d, J = 7.4 Hz, 2H), 7.44–7.35 (m, 4H), 7.28–7.07 (m, 17H), 7.07–7.00 (m, 4H), 5.62 (d, J = 14.1 Hz, 2H), 3.77 (t, J = 6.9 Hz, 2H), 2.33 (t, J = 7.3 Hz, 4H), 1.72 (s, 12H), 1.66–1.59 (m, 4H), 1.49–1.41 (m, 4H). ¹³C{¹H} NMR (125 MHz, DMSO-d₆) δ 172.7, 152.4, 143.4, 143.2, 143.1, 143.1, 143.0, 142.1, 141.1, 140.8, 140.6, 140.5, 131.5, 130.9, 130.8, 130.7, 130.6, 129.8, 128.5, 128.0, 127.9, 127.8, 127.8, 127.8, 126.8, 126.6, 124.9, 122.5, 111.4, 101.3, 50.8, 49.0, 43.5, 27.9, 27.1, 26.1, 22.6. HRMS (ESI-TOF) m/z: [M]⁻ Calcd for C₅₉H₅₉N₂O₆S₂⁻ 955.3820; Found 955.3818.

Purification by reverse phase column chromatography (C18, 0–80% methanol in H₂O) gave 2P (51 mg, 49%) m.p. > 150 °C (dec). ¹H NMR (500 MHz, DMSO-d₆) δ 8.81 (d, J = 14.3 Hz, 2H), 8.47 (d, J = 9.0 Hz, 2H), 8.40–8.33 (m, 4H), 8.31–8.23 (m, 4H), 8.13 (t, J = 7.7 Hz, 1H), 8.06–7.97 (m, 4H), 7.72–7.64 (m, 4H), 7.51 (t, J = 7.6 Hz, 2H), 5.95 (d, J = 14.1 Hz, 2H), 3.81–3.62 (m, 4H), 2.37 (t, J = 7.3 Hz, 4H), 2.09 (s, 12H), 1.73–1.63 (m, 4H), 1.46–1.36 (m, 4H). ¹³C{¹H} NMR (125 MHz, DMSO-d₆) δ 174.0, 151.8, 139.8, 133.2, 131.4, 130.8, 130.3, 129.9, 127.7, 127.7, 127.6, 127.5, 126.8, 126.4, 125.2, 124.8, 123.7, 123.2, 122.1, 111.7, 101.1, 50.9, 50.7, 43.5, 26.8, 26.2, 22.5. HRMS (ESI-TOF) m/z: [M]⁻ Calcd for C₅₇H₅₃N₂O₆S₂⁻ 925.3351; Found 925.3290.

Purification by reverse phase column chromatography (C18, 0–80% methanol in H₂O) gave 2T (71 mg, 60%) m.p. > 150 °C (dec).¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 8.56 (d, J = 14.1 Hz, 2H), 8.25 (d, J = 8.5 Hz, 2H), 8.08–8.03 (m, 4H), 7.77 (d, J = 8.8 Hz, 2H), 7.68 (t, J = 7.6 Hz, 2H), 7.51 (t, J = 7.5 Hz, 2H), 7.28–7.10 (m, 15 H), 7.06 (t, J = 7.6 Hz, 4H), 5.69 (d, J = 14.0 Hz, 2H), 3.91 (d, J = 7.3 Hz, 2H), 2.38 (t, J = 7.3 Hz, 4H), 1.99 (s, 12H), 1.75–1.64 (m, 4H), 1.55–1.45 (m, 4H). ¹³C{¹H} NMR (125 MHz, DMSO-d₆) δ 173.6, 151.4, 143.4, 143.1, 143.1, 143.0, 140.8, 140.5, 139.8, 133.1, 131.4, 131.4, 130.9, 130.7, 130.6, 130.3, 129.9, 129.8, 127.9, 127.8, 127.8, 127.6, 126.8, 126.6, 124.8, 122.1, 111.8, 101.0, 50.8, 50.7, 26.7, 26.2, 22.5. HRMS (ESI-TOF) m/z: [M]⁻ Calcd for C₆₇H₆₃N₂O₆S₂⁻ 1055.4133; Found 1055.4161.

Synthesis of 1B.

To a solution of bromo-substituted Cy5 dye 3 (59 mg, 0.081 mmol, 1.0 equiv.), 4-ethynyl-1,1′-biphenyl (43 mg, 0.243 mmol, 3.0 equiv.), and DIPEA (28 μL, 0.162 mmol, 2.0 equiv.) in degassed DMF (4 mL), was added CuI (3.1 mg, 0.016 mmol, 20 mol%), and [Pd(PPh₃)₂Cl₂] (5.7 mg, 0.008 mmol, 10 mol%) at room temperature. The reaction mixture was stirred in an oil bath at 100 °C for 4 h in the dark. After cooling to room temperature, the solvent was evaporated under vacum and the residue purified by reverse phase column chromatography (C18, 0–80% methanol in H₂O). The product 1B was obtained as a blue solid (25 mg, 37%) m.p. > 150 °C (dec). ¹H NMR (500 MHz, DMSO-d₆) δ 8.50 (d, J = 14.0 Hz, 2H), 7.93 (d, J = 8.6 Hz, 2H), 7.78 (d, J = 7.1 Hz, 2H), 7.69 (dd, J = 7.7, 5.3 Hz, 4H), 7.65–7.58 (m, 6H), 7.57 – 7.52 (m, 1H), 7.49 (t, J = 7.7 Hz, 2H), 7.45 (t, J = 7.8 Hz, 2H), 7.40 (t, J = 7.3 Hz, 1H), 7.32 (t, J = 7.4 Hz, 2H), 6.62 (d, J = 14.1 Hz, 2H), 4.23 (t, J = 7.6 Hz, 4H), 2.56 (t, J = 7.4 Hz, 4H), 1.95–1.87 (m, 4H), 1.87–1.79 (m, 4H), 1.74 (s, 12H). ¹³C{¹H} NMR (125 MHz, DMSO-d₆) δ 174.0, 153.5, 141.9, 141.4, 140.0, 139.2, 133.1, 132.3, 132.1, 132.0, 131.6, 131.5, 131.4, 129.0, 128.8, 128.7, 128.6, 127.8, 127.5, 126.7, 125.4, 122.5, 121.3, 112.7, 111.9, 102.0, 100.5, 83.7, 50.9, 49.5, 44.3, 26.9, 26.2, 23.0. HRMS (ESI-TOF) m/z: [M]⁻ Calcd for C₄₇H₄₉N₂O₆S₂⁻ 801.3038; Found 801.3043.

Data Availability

The data underlying this study are available in the published article and its Supporting Information.