Deoxygenation Increases Photoluminescence Lifetime of Protein

Article pubs.acs.org/JPCB

Deoxygenation Increases Photoluminescence Lifetime of ProteinResponsive Organic Probes with Triplet−Singlet Resonant Energy Transfer Kadri Ligi, Erki Enkvist, and Asko Uri* Institute of Chemistry, University of Tartu, 14a Ravila Street, 50411 Tartu, Estonia S Supporting Information *

ABSTRACT: Cells and bodily fluids possess strong nanosecond-lifetime autofluorescence, therefore photoluminescent probes with microsecond-scale luminescence decay time would be useful for analysis of biological samples, as they allow the performance of measurements in time-resolved (TR) format in a time gate (time window) where the nonspecific background fluorescence has ceased. We have previously disclosed bindingresponsive luminescent probes for protein kinases (PKs), ARC-Lum(Fluo) probes. High brightness of the probes is achieved through intramolecular Förster-type resonant energy transfer (FRET) from excited triplet state of a thiophene- or selenophene-comprising phosphor (3D*) to singlet acceptor dye (1A) leading to amplified emission from the dye. Here, we determined quantum yields (QYs) and oxygen sensitivity of separate phosphorescent donor and fluorescent acceptor and compared these with those of the corresponding ARC-Lum(Fluo) probes both in nonbound and PK-bound states. The microsecond-scale luminescence of free and of PK-bound probes was quenched with different efficiency by molecular oxygen and the luminescence intensity of the probes was substantially increased upon deoxygenation. The brightness of an ARC-Lum(Fluo) probe in PK-bound state was more than 50-fold higher than that of the phosphorescent donor alone. The findings of the study can be used for the construction of bright long-lifetime organic tandem probes.

■

been focused to the phosphorescence in solid matrixes.5 Moreover, small-molecule responsive probes that selectively respond to the presence of the analyte have been mostly described for ions and small molecules (pH, oxygen, metal ions, and so forth.) and probes whose selective binding to target proteins leads to a substantial change in TR signal intensity have been described in rare cases.6−8 We have reported novel protein binding-responsive organic photoluminescent probes (ARC-Lum probes) whose association with a protein kinase (PK) led to green, yellow, orange, or red luminescence with long (microsecond-scale) decay upon illumination with a pulse of near-UV radiation.9−13 ARC-Lum probes can be divided into two categories. First, ARC-Lum(−) probes are structurally conjugates of adenosine analogues and peptides (ARCs) that incorporate a heteroaromatic bicyclic or a tricyclic fragment with a sulfur or a selenium atom in one of the aromatic rings (Figure 1). The heteroaromatic fragment of an ARC-Lum(−) probe targets the ATP-binding pocket of the PK where it possesses phosphorescence emission at 500−750 nm upon excitation with a pulse of near-UV radiation. Additionally, ARC-Lum(−) probes possess weak fluorescence emission at

INTRODUCTION Protein binding-responsive photoluminescence probes that acquire increased intensity of the luminescence emission in complex with the target protein are in great need for measurement of protein concentration both in biochemical studies and in biomedical research. Because of nanosecondlifetime autofluorescence of cells and bodily fluids, probes possessing long-lived (microsecond-scale) luminescence emission would allow performance of the measurements in timeresolved (TR) format in a time gate where the fluorescence has ceased.1,2 Also, by introducing a time-delay into luminescence measurements, signals from probes with microsecond-scale lifetime can be easily distinguished from fluorescent smallmolecule compounds and fluorescent proteins used in cellular experiments.3 Unfortunately, responsive photoluminescence probes possessing long-lifetime photoluminescence (phosphorescence or delayed fluorescence) in water solution at room temperature are not common, being mostly restricted to lanthanide and noble metal complexes.4 Although significant improvement has been made in the development of purely organic roomtemperature phosphorescent materials within last 5 years, there is still not much known about the application of small-molecule organic phosphors in water solution as most of the research about organic room-temperature phosphorescent materials has © XXXX American Chemical Society

Received: April 1, 2016 Revised: May 15, 2016

A

DOI: 10.1021/acs.jpcb.6b03342 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B

This energy transfer leads to the excited singlet state of the acceptor (1A*), which thereafter emits light, resulting in powerful sensitization of the phosphorescence emission.9 Long lifetime of the emission mediated by the short-lifetime fluorescent dye is due to slow resonant energy transfer from 3 D* to 1A leading to excitation of the dye to 1A* and relaxation of 3D* to the ground singlet state (3D* + 1A → 1D + 1A*). The FRET from 3D* to 1A can be efficient only if the energy transfer is faster than other processes that depopulate 3D*. In case the efficiency of FRET is 100%, the QY of microsecondscale luminescence is that of the acceptor.14−16 The amplification of a microsecond-lifetime luminescence signal of the donor luminophore by a conjugated organic fluorophore leading to long-wavelength long-lifetime tandem dyes has been described for metal−ligand complexes conjugated with organic dyes (MLCs).14−17 The amplification of phosphorescence signal of an organic phosphor by a conjugated fluorescent dye resulting from triplet−singlet energy transfer at room temperature in aqueous solution was first described by us for ARC-Lum(Fluo) probes.9 On the other hand, the theoretical possibility of nonradiative energy transfer from the triplet excited state of the donor to a singlet-state acceptor was considered in Förster’s original paper in 195918 and the practical realization of triplet−singlet energy transfer by FRET mechanism was shown for several pairs of organic luminophores already in the 1960s.19−22 For example, occurrence of triplet−singlet FRET was reported for a pair of small molecules phenanthrene and rhodamine B,21 for DNA− acridine dye complexes,20 and for chymotrypsin−proflavin complexes.19 All these demonstrations concerned experiments performed at very low temperatures (usually at 77 K) and in solid matrixes and the amplification of the long-lifetime luminescence by a nearby fluorophore was not reported in these publications. The discovered binding-responsiveness of luminescence intensity of ARC-Lum probes could be reasoned by restricted molecular movements of the heteroaromatic phosphor if the probe is associated with the ATP-binding pocket of the PK, but also with decreased quenching rates of the heteroaromatic fragment by dissolved molecular oxygen and buffer components. The current study was performed to establish the impact

Figure 1. Structures of the compounds applied for the study.

wavelength range from 400 to 600 nm. Second, ARCLum(Fluo) probes are tandem probes that in addition to the phosphor of ARC-Lum(−) probes incorporate a fluorescent dye (e.g., HiLyte 488, TAMRA, Cy3B, TexasRed, Alexa Fluor 647) whose absorption spectrum at least partly overlaps with phosphorescence emission spectrum of the sulfur or selenium atom-comprising heteroaromatic fragment (Figure 2B). Upon binding to a PK and excitation with a pulse of near-UV radiation, ARC-Lum(Fluo) probes emit light with microsecond-scale luminescence decay time (τ = 20−250 μs in the presence of dissolved molecular oxygen), therefore the emission spectrum of the probe coincides with the fluorescence emission spectrum of the attached dye.9,11 The signal intensity of the microsecond-lifetime luminescence of ARC-Lum(Fluo) probes in complex with a PK is 20- to 1000-fold higher than that of the phosphorescence signal of their ARC-Lum(−) counterparts.9 This enhancement is bigger for thiophenecontaining compounds than for selenophene-containing compounds whereas the selenophene-containing ARC-Lum probes are brighter in complex with PKs than their thiophenecontaining counterparts.9 A cascade mechanism of energy transfer (Figure 3) that was proposed for the discovered phenomenon included efficient Förster-type resonant energy transfer (FRET) from the excited triplet state of the low-QY donor phosphor (3D*) to the singlet state of the acceptor fluorophore (1A), possessing high QY.9

Figure 2. Overlap of absorption spectra of Se(Cy3B), Se(PF647) and Cy3B-r9-NH2 with fluorescence (A) and phosphorescence spectra (B) of nonbound probe Se(−). Time-delayed phosphorescence spectrum of Se(−) was measured on the FS900 instrument described in Materials and Methods (exc. 350 nm, delay 0.2 ms, integration window of 2.8 ms) and the sample was purged with argon gas. Absorption and fluorescence spectra were recorded with 1 nm recording steps, and phosphorescence spectra were recorded with 5 nm steps. The graphs representing emission of Se(−) have been smoothed. Abs., absorption; Fluor., fluorescence; Phosph., phosphorescence. B


Article


Figure 3. Jablonski diagram of an ARC-Lum(Fluo) probe. The donor-luminophore (1D) is excited with near-UV radiation. The excited donorluminophore (1D*) can dissipate its energy via radiative (vertical solid lines) and nonradiative processes (curved lines), such as internal conversion (IC) and vibrational relaxation (VR). The presence of the acceptor-fluorophore (1A) in close proximity to 1D* enables dipole−dipole interaction between the luminophores, leading to the relaxation of the donor to the ground singlet state (1D) and the formation of excited singlet state of the acceptor (1A*) that thereafter emits photons and relaxes to the ground singlet state (1A). 1D* can also change its spin by intersystem crossing (ISC) and go into excited tripet state 3D*. In the presence of quenchers (Q), such as dissolved molecular oxygen, the energy of 3D* is dissipated mainly by nonradiative processes and no phosphorescence occurs (the dashed vertical lines). If the concentration of the quencher is greatly reduced, for example, inside the catalytic pocket of a PK, the lifetime of 3D* increases. The increased lifetime of 3D* enables triplet−singlet energy transfer from 3 D* to the conjugated acceptor-luminophore that leads to the relaxation of the donor to the ground singlet state 1D and formation of 1A* leading to emission of light from the acceptor dye.

resolution mass spectra of all new compounds were recorded on Thermo Electron LTQ Orbitrap mass spectrometer using electrospray ionization. Structures of compounds that were used in the study are depicted in Figure 1 and in Table S1 in SI. TR Measurement of Luminescence. TR measurement of luminescence was performed with four different instruments, which all were operated in time-domain. The time-delayed photoluminescence spectra were recorded at room temperature in quartz cuvettes with 10.00 ± 0.01 mm optical path length (Starna Inc.) on a spectrofluorometer (FS900, Edinburgh Instruments) equipped with a photon counting photomultiplier tube (R2658P, Hamamatsu) and using a xenon flash lamp as the excitation source. The solution of probe Se(−) in 50 mM phosphate buffer (pH = 7.4) was deoxygenated by bubbling the solution with argon gas (BOC Gases, grade 5.5). The deoxygenation was monitored on an in-house constructed phosphorometer26 modified for time-domain operation and equipped with an avalanche photodiode (Hamamatsu), a 365 nm LED, and quartz fibers for the transfer of the excitation radiation to the cuvette. The third instrument applied for the TR measurement of luminescence was a plate reader (PHERAstar, BMG Labtech.) with appropriate optical modules HTRF802D1 [ex. 330(60), em. 675(50)] or TRF904B1 [ex. 330(60), em. 590(50)]. The measurements were performed on black, low-volume, 384-well, nonbonding-surface microplates (Corning, code 3676) at 30 °C. The assays were performed in phosphate buffer (100 mM sodium phosphate, pH = 7.4) or Hepes buffer (100 mM Hepes, pH = 7.4) containing Tween-20 (0.005%, Sigma, P9416) and bovine serum albumin (BSA, 0.5 mg/mL, Sigma, A4503). The plates containing samples (15 μL per well) with glucose oxidase (GO, final activity 2 U/mL, Sigma, G6125) and catalase (final concentration 0.05 mg/mL, Sigma, C1345) were preincubated at 30 °C for 15 min and thereafter the enzymatic deoxygenation was started by addition of 5 μL of D-glucose (Sigma, G7021) solution (final concentration 25 mM). Two negative controls (one with

of dissolved molecular oxygen on the microsecond-scale luminescence decay times of both free and PK-bound probes. It was proposed that the dissolved molecular oxygen rapidly quenches the 3D* of free ARC-Lum probes whereas the diffusion of oxygen is slower into the catalytic pocket of a PK, thereby increasing the QY of 3D* upon binding to a PK. The second question that was addressed within this study, was the possibility of energy transfer from the excited singlet state of the donor (1D*) to 1A. If there is energy transfer from 1D* to 1 A, this would contribute to the depopulation of 1D* and may decrease the transition to 3D* and finally the microsecond-scale luminescence intensity of ARC-Lum(Fluo) probes. Förster distances were calculated for both (tripet-singlet and singlet− singlet) types of energy transfer and for different donor− acceptor pairs. We now propose a more detailed insight into the photoluminescence phenomenon of ARC-Lum(Fluo) probes. The obtained knowledge can be applied for further improvement of the probes for analysis of different proteins in biological samples as we have shown to be possible for protein kinases.9,12

■

MATERIALS AND METHODS Materials. All solvents, buffer components, and referent compounds were obtained from commercial sources and were used as received. The synthesis of ARC-Lum probes is described below. The chemicals used for synthesis were purchased from Sigma-Aldrich, if not stated differently. NHS ester of the fluorescent dye Cy3B was purchased from GE Healthcare Life Sciences. The full-length catalytic subunit of human PKA (PKAc) was produced as described previously.23 Chemical Synthesis. ARC-Lum(−) probes were synthesized according to conventional Fmoc-strategy of solid-phase peptide synthesis procedures as described earlier24 and purified with reverse-phase HPLC. Labeling of ARCs was performed through the side chain of the D-lysine residue by using Nhydroxysuccinimide esters of the fluorescent dyes.25 HighC


Article

The Journal of Physical Chemistry B Table 1. Luminescence Decay Times (τ, μs), Normalized Luminescence Intensity (I0), and Normalized Integrated Luminescence Intensity (I0*τ) of ARC-Lum Probes in PK-Bound Statea Se(−) components added to the probe + PKAc + G + PKAc + GO + C + PKAc + GO + C + G

Se(Cy3B)

Se(PF647)

S(PF647)

τ (μs)

I0

I0*τ

τ (μs)

I0

I0*τ

τ (μs)

I0

I0*τ

τ (μs)

I0

I0*τ

111 ± 7 108 ± 5

1.0 ± 0.1 1.0 ± 0.1

1.0 ± 0.1 1.0 ± 0.1

77 ± 1 76 ± 2

34 ± 7 50 ± 14

12 ± 3 18 ± 5

36 ± 3 33 ± 1

240 ± 50 280 ± 10

52 ± 12 57 ± 3

101 ± 4 109 ± 5

6.8 ± 0.4 6.6 ± 0.4

4.3 ± 0.3 4.4 ± 0.4

447 ± 5

1.2 ± 0.1

4.7 ± 0.4

157 ± 4

45 ± 8

33 ± 6

47 ± 1

270 ± 30

76 ± 9

556 ± 5

7.5 ± 1.0

25 ± 3

The data are based on measurements performed with the plate reader PHERAstar at 30 °C in the presence of dissolved molecular oxygen and in enzymatically deoxygenated conditions. The transmittance of the optical filters as well as the concentration of the probe applied for the measurements has been considered for normalization of I0 of ARC-Lum/PKAc complexes. The normalization of I0 and integrated luminescence intensity (I0*τ) have been done in relation to Se(−)/PKAc complex in the presence of dissolved molecular oxygen. Error bars represent SEM (N = 2 to N = 5). a

recorded on a spectrophotometer (Lambda-35 UV−vis, PerkinElmer) and steady-state fluorescence was measured with the FS900 spectrofluorometer described above. The maximal number of counts did not exceed 2 × 105 countsper-second (cps). The excitation energy was corrected for the excitation light intensity measured by the reference detector and the emission signal intensity was corrected for the wavelength-dependency of the detection system. The emission QYs were determined relative to that of 9,10-diphenylanthracene (DPA) in ethanol (QY = 95%)29 or Rhodamine 6G (R6G) in ethanol (QY = 95%).30 During measurements, the optical density of DPA and R6G was kept below 0.05 units at the absorption maximum. The refraction indices used for phosphate buffer, ethanol, and dichloromethane (DCM) were 1.333, 1.361, and 1.424, respectively. The data were processed and the QYs were calculated using the software Origin 7.0 (OriginLab Corporation). All the dilutions of ARC-Lum probes were made into 50 mM phosphate buffer (pH = 7.4). The exact concentration of the probes in samples was calculated by optical absorption of the solution according to the Lambert−Beer law. The molar extinction coefficients of ARC-Lum probes are known from the previous studies9 and have been determined either by the weighing method or in ratio to the molar extinction coefficient of the C-terminal fluorescent dye. The molar extinction coefficient ε of probe S(−) in a water-based buffer at pH = 7.4 is 15 000 M−1 cm−1. For conjugates comprising the fluorescent dye Cy3B (GE Healthcare Life Sciences), the ε value of 130 000 M−1 cm−1 was used at the absorption maximum.

glucose only and the other with GO and catalase) were always monitored if enzymatic deoxygenation was performed. The luminescence lifetimes were not altered by addition of glucose only. The samples with glucose, but no GO or catalase have not been represented on graphs (except for Figures S2 and S5 in SI) because they cannot be distinguished from the corresponding samples containing glucose and catalase. PKAc was always added in small excess to ensure full complexation of the ARCLum probe by the PK. The concentration of the active form of PKAc was determined by titration of the probe using fluorescence anisotropy (FA) or TR luminescence intensity read-out as described previously.11 The activity of GO was determined with a porphyrin-based optrode27 and its performance was measured on the plate reader by monitoring the luminescence lifetime of the oxygen-sensitive probe MX-400 (final dilution 600x, Luxcel Biosciences). The data were analyzed with GraphPad Prism software version 5.0 (GraphPad Software, Inc.). Calculation of Fö rster Distances. Förster distances of ARC-Lum(Fluo) probes were calculated on the basis of the overlap of the emission spectrum of the donor luminophore Se(−) and the absorption spectrum of the acceptor fluorophore Cy3B or PromoFluor647 (PF647) by applying eqs 1 and 228 ⎛ κ 2ϕ 0J(λ) ⎞1/6 D ⎟⎟ R 0 = 0.211⎜⎜ 4 n ⎝ ⎠

(1)

∞

J (λ ) =

∫0 FD(λ)εA (λ)λ 4 dλ ∞

∫0 FD(λ)dλ

■

(2)

where R0 is the Förster distance; κ is the orientation factor (κ2 value of 0.667 was used); φ0D is QY of the donor phosphorescence in the absence of the acceptor (0.5%); n is the refractive index of the solution (1.33 for aqueous solutions); εA(λ) is the molar extinction coefficient of the acceptor at wavelength λ; J(λ) is the spectral overlap integral calculated by eq 2, and FD(λ) is the luminescence emission intensity of the donor in the wavelength range λ + Δλ. The coefficient 0.211 was used because the wavelength was expressed in nanometres. Measurement of QYs. For the determination of QYs, quartz fluorometric cells with 10.00 ± 0.01 mm optical path length were used (Starna Inc.). The spectra were recorded at room temperature. The one-photon absorption spectra were

RESULTS AND DISCUSSION

Structure and Characterization of ARC-Lum Probes. ARC-Lum probes used in the current study comprise either a thiophene- or a selenophene-containing heteroaromatic luminophore (Figure 1 and Table S1 in SI) that is the source of both fluorescence and phosphorescence if the probe is illuminated with near-UV radiation. The phosphorescence decay time of thiophene-comprising organic phosphors is longer than that of selenophene-comprising organic phosphors.31 Furthermore, a clear heavy-atom effect was shown recently for two families of organic bis-benzoazole luminophores doped with oxygen, sulfur, selenium, or tellurium.32 We have detected significantly higher luminescence intensities for selenophene-containing ARC-Lum probes in complex with a D


Article

The Journal of Physical Chemistry B PK compared to their thiophene-containing counterparts.9 However, the phosphorescence intensity of thiophene-containing ARC-Lum(−) probes in complex with a PK has been difficult to quantify and requires the application of high concentration of both the probe and the PK, which may lead to undesired effects, such as self-quenching and high viscosity. Therefore, we chose two selenophene-containing probes Se(−) and Se(Cy3B) to establish and quantify the luminescence enhancement of ARC-Lum probes upon binding to a PK both in deoxygenated conditions and in the presence of dissolved molecular oxygen. Se(−) is an ARC-Lum(−) probe that incorporates a single selenophene-comprising luminophore, a heteroaromatic fragment that binds to the ATP-pocket of the PK. ARC-Lum(Fluo) probe Se(Cy3B) incorporates besides the selenophene-comprising heteroaromatic luminophore a fluorescent dye Cy3B, whose absorption spectrum well overlaps with the phosphorescence emission spectrum of the selenophene-containing donor-luminophore (Figure 2B). For the control of the TR luminescence measurement equipment and for comparison to the probe Se(Cy3B) we also determined the QY and recorded the luminescence spectra of nona-Darginine peptide labeled with the fluorescent dye Cy3B (Cy3Br9-NH2, see Table S1 in SI). In addition, we determined the luminescence lifetime of ARC-Lum(Fluo) probes that incorporate the fluorescent dye PF647 as the acceptor-luminophore. Se(PF647) incorporates a selenophene-containing donorluminophore and in complex with PKAc it possesses about three-fold shorter luminescence lifetime (in the presence of dissolved molecular oxygen) than the thiophene-comprising probe S(PF647) (Table 1). The structures of the studied compounds are depicted in Figure 1 and in Table S1 in SI. Deoxygenation of Aqueous Solutions. One critical factor in case of phosphorescence measurements is the quenching of phosphorescence by dissolved molecular oxygen. Thus, the relatively low popularity of phosphorescence techniques in the studies on protein structure and dynamics is to a certain extent caused by the existing literature discrepancy regarding photophysical parameters of the indole luminophore, such as triplet state lifetime, triplet QY and the rate constant for intersystem crossing (ISC).33 Long history of measurement of protein phosphorescence has shown that improvement of deoxygenation methods leads to increased values of phosphorescence decay time.34 The most efficient deoxygenation methods require the application of sealed systems, big measurement volumes, extra-purified inert gases, freeze−thaw cycles, and other time-consuming procedures that make them not suitable for high-throughput studies. Therefore, we looked for a deoxygenation technology that would allow the performance of high-throughput measurements in a microtiter plate format with a luminescence platereader. A two-enzyme method, based on the application of GO, catalase, and glucose, that effectively generates oxygen deficiency in aqueous solutions has been in active use for decades. GO catalyzes the oxidation of glucose by molecular oxygen and catalase depletes hydrogen peroxide that is produced during oxidation of glucose. Baumann et al.35 disclosed a simple and efficient variant of the two-enzyme method for generation of oxygen deficiency in cell culture. GO/ catalase method enabled quick reduction of molecular oxygen concentration in unsealed vessels. In their unsealed model system without cells, the oxygen concentration was decreased by almost 200-fold (down to 1 μM at 37 °C). One drawback of the GO/catalase method is that the added proteins and buffer

components can also perform as quenchers for phoshphors under investigation.34 Here, the two-enzyme deoxygenation method was applied by us to study the oxygen sensitivity of luminescence of ARCLum/PKAc complexes and free ARC-Lum probes. With such deoxygenation system we could perform the measurements reliably in a volume of 20 μL in wells of a microtiter plate. The efficiency of deoxygenation was measured with a commercial homogeneous oxygen-sensitive probe MX-400. Deoxygenation was very fast and efficient (Figures S1 and S2 in SI). The deoxygenation results obtained for measurements with a 384well microtiter plate (using plate reader PHERAstar) were compared to the results of measurements performed in a sealed quartz cuvette using the spectrofluorometer Fluoromax-4. Oxygen Sensitivity of Luminescence of ARC-Lum(Fluo)/PKAc Complexes. Photoluminescence properties of ARC-Lum(Fluo) probes Se(Cy3B), Se(PF647), and S(PF647) in complex with PKAc reveal different sensitivity to the deoxygenation of the solution (Figures 4 and 5). Thiophene-

Figure 4. Decay of luminescence intensity of Se(Cy3B) (2 nM) in PKbound state in nondeoxygenated and in enzymatically deoxygenated solutions. The assay was performed on plate reader PHERAstar. GO, glucose oxidase; C, catalase; G, glucose.

comprising ARC-Lum(Fluo) probe S(PF647) possesses the greatest sensitivity; its luminescence lifetime increases from 109 to 556 μs (Table 1 and Figure 5). In the case of probe Se(PF647), a selenophene-comprising counterpart of probe S(PF647), the enzymatic deoxygenation increased the

Figure 5. Comparison of luminescence lifetimes of S(PF647)/PKAc and Se(PF647)/PKAc complexes in nondeoxygenated and in enzymatically deoxygenated solutions. The luminescence intensity has been expressed for 1 nM complex. The data are shown for a representative experiment with three independent measurements. The assay was performed on a plate reader PHERAstar as described in Materials and Methods. The error bars represent SEM (N = 3). GO, glucose oxidase; C, catalase; G, glucose. E


Article


Figure 6. Luminescence brightness (I0τ) of Se(−) and Se(Cy3B) in PK-bound and nonbound state in phosphate-based buffer (A). Luminescence intensity (gate 150−650 μs) of Se(Cy3B) in PK-bound and nonbound state in Hepes-based buffer (B). The data in panel A have been presented for 1 nM compound/complex. The transmittance of the optical filter applied for the measurement was determined and thereafter this knowledge was used for comparing luminescence brightness of Se(Cy3B) and Se(−) (A). The assays were performed on a plate reader PHERAstar as described in Materials and Methods. The error bars represent SEM (N = 3 to N = 5). GO, glucose oxidase; C, catalase, G, glucose.

after a delay of 45 μs whether excited with near-UV light or at 540(20) nm (data not shown), whereas the luminescence of Se(Cy3B) was proportional to the concentration within the concentration range from 3 to 100 nM (Figure 7). This means

luminescence decay time less than 1.5-fold, from 33 to 47 μs. The enzymatic deoxygenation of the sample with Se(Cy3B)/ PKAc complex led to an increase of the luminescence lifetime of Se(Cy3B) to about 2-fold, from 76 to 157 μs if measured on the plate reader (Table 1 and Figure 4) and from 69 to 161 μs if measured on the spectrofluorometer (data not shown). These results point to the tendency that ARC-Lum(Fluo) probes with longer decay time possess higher sensitivity to deoxygenation, which is an expected result. In enzymatically deoxygenated conditions the normalized integrated luminescence intensity, that is, normalized luminescence brightness, I0τ (I0, luminescence intensity extrapolated to t = 0 μs, multiplied with luminescence lifetime τ) of Se(Cy3B)/ PKAc complex is 30-fold higher than the luminescence brightness of the nonbound probe Se(Cy3B) in phosphatebased buffer (Figure 6A). This enhancement of luminescence brightness depends on the acceptor-fluorophore conjugated to the donor-luminophore as well as the origin of the PK.9,13 The binding-responsiveness of ARC-Lum(Fluo) probes can be further increased with the application of the appropriate delay and gate time (Figure 6B). As studies with proteins are usually performed within the temperature range from room temperature to 37 °C, the effect of temperature on the microsecond-scale luminescence lifetime of ARC-Lum probes within this temperature range was studied: slight increase of the luminescence lifetime of Se(Cy3B)/PKAc complex was observed at lower temperatures (Figure S3 in SI). This change was smaller than for the free donor-compound (Figure S4 in SI), however, it shows that the temperature has to be tightly fixed within an assay with ARC-Lum probes. Oxygen Sensitivity of Luminescence of Free ARCLum(Fluo) Probes. The luminescence of free Se(Cy3B) was studied in enzymatically deoxygenated conditions and compared with luminescence of the labeled peptide Cy3B-r9NH2, which does not comprise a phosphorescent donor. Additional experiments were performed to show that the signal of free Se(Cy3B) is not caused by unspecific binding of the probe to nontarget proteins GO, catalase, and BSA. The luminescence intensity of the labeled peptide Cy3B-r9-NH2 could not be distinguished from that of the buffer neither in enzymatically deoxygenated or nondeoxygenated conditions

Figure 7. Oxygen sensitivity of TR luminescence intensity of free Se(Cy3B). The concentration series of probe Se(Cy3B) was made in the presence of BSA (0.5 mg/mL). GO, glucose oxidase; C, catalase; G, glucose.

that the long-lifetime luminescence signal detected for free Se(Cy3B) was not caused by the after-glow of the xenon flash lamp. The addition of BSA or GO and catalase to the buffer solution did not change the luminescence lifetime of the probe (Figure S5 in SI). The value of I0 of free Se(Cy3B) in deoxygenated conditions was 3.9 ± 0.5 times higher than that of Se(−), considering the transmittance of the emission filter, which passes 36% and 52% of the emission of Se(−) and Se(Cy3B), respectively, showing that the energy transfer from 3 D* to A1 also occurs in free Se(Cy3B) (see discussion about the luminescence of free Se(Cy3B) and Figure S5 in SI). Oxygen Sensitivity of Luminescence of ARC-Lum(−) Probe. The phosphorescence emission spectrum of free Se(−) in argon-purged buffer solution coincided with the phosphorescence emission spectrum of Se(−)/PKAc complex if the latter was measured in nondeoxygenated buffer solution (Figure S6 in SI). The phosphorescence lifetime of free F


Article


Figure 8. Comparison of luminescence emission spectra of Se(Cy3B) and Cy3B-r9-NH2. The steady-state emission spectra were recorded with excitation at 353 nm (A) and 495 nm (B). The luminescence was calculated for 1 μM solution and then normalized by the shared maximal and minimal values.

Se(−) at room temperature in argon-purged buffer solution (τ = 290 ± 2 μs) was longer than the phosphorescence lifetime of Se(−) in complex with PKAc in nondeoxygenated buffer solution at room temperature (τ = 163 ± 7 μs). Enzymatic deoxygenation of the sample containing Se(−)/PKAc complex at 30 °C increased the luminescence lifetime of Se(−) from 108 to 447 μs (Table 1). The luminescence lifetime of free Se(−) in Ar-purged solutions at room temperature (τ = 290 ± 2 μs) was 2.5-fold longer than in enzymatically deoxygenated conditions at 30 °C (τ = 109.8 ± 3.4 μs). This difference in luminescence lifetimes of free Se(−) was not caused by temperature (Figure S4 in SI). Although very small content of oxygen in enzymatically deoxygenated solution can still cause luminescence quenching of the free probe,36 additional quenching of phosphorescence by components of the enzymatic deoxygenation buffer may also be possible.34 Efficiency of Energy Transfer. Both the fluorescence and phosphorescence emission spectra of Se(−) [that is the luminescence donor phosphor of the ARC-Lum(Fluo) probe Se(Cy3B)] partially overlap with the absorption spectrum of the fluorescent dye Cy3B (Figure 2). Therefore, in the case of Se(Cy3B) preconditions are fulfilled for realization of two types of energy transfer: first, for energy transfer from 1D* to 1A (1D* + 1A → 1D + 1A*) and, second, for energy transfer from 3 D* to 1A (3D* + 1A → 1D + 1A*) (Figure 3). Although longrange triplet−singlet transfer is a spin-forbidden process, the slow rate of such a transfer is compensated by the long lifetime of 3D* that makes possible energy transfer to the fluorescent acceptor if other modes of triplet deactivation are less competitive.37,38 The Förster distances (eqs 1 and 2) for the luminophores of the ARC-Lum(Fluo) probe Se(Cy3B) as calculated for energy transfer from the 1D* and 3D* states of the donor to Cy3B were rather similar, 2.2 and 2.6 nm, respectively. The QY of phosphorescence (0.5%) that was applied for the calculation of the Förster distance was determined for free Se(−) in 50 mM phosphate buffer at room temperature in argon-purged solution. The luminescence lifetime of Se(−)/PKAc complex in the same conditions was 1.8-fold shorter, which would result in slightly smaller QY and Förster distance of 2.1 nm for triplet−singlet energy transfer, provided that the I0 and radiative decay rate do not change upon binding to PKAc. The efficiency of energy transfer from 1D* to 1A was evaluated by comparison of the normalized steady-state emission spectra of Se(Cy3B) and Cy3B-r9-NH2 (Figure 8). Equation 3 was used for the calculation of the efficiency of energy transfer39

E=

⎤ εA (λDex ) ⎡ FAD(λAem) − 1⎥ ex ⎢ em εD(λD ) ⎣ FA(λA ) ⎦

(3)

ex where E is efficiency of the energy transfer; εA(λex D ) and εD(λD ) are the molar extinction coefficients of the acceptor and donor at the excitation wavelength of the donor, respectively; FAD(λAem) and FA(λAem) are the fluorescence intensities of the acceptor in the absence and presence of the donor at the emission wavelength of the acceptor, respectively. The normalized excitation spectrum (rather than the absorption spectrum) was applied for the calculation of molar extinction coefficients at the applied excitation wavelengths. The obtained efficiency of energy transfer from 1D* to 1A is very high, approximately 90%. The strong interaction between luminophores may be caused by stacking of aromatic moieties in an aqueous buffer solution. The measurement of QYs, which is discussed in the following paragraph, showed that the efficiency of singlet−singlet energy transfer between luminophores in Se(Cy3B) decreased upon binding to PKAc (Table 2), which shows that the interaction of luminophores is interfered upon binding.

Table 2. Quantum Yields of Studied Compounds compound a

S(−) Se(−)a deox Se(−)a,b Cy3B-r9-NH2c Se(Cy3B)c Se(Cy3B)a Se(Cy3B) + PKAca,d Se(Cy3B) + PKAcc,d

concentration (μM)

quantum yield (%)

2.0 7.7 7.7 0.62 0.76 0.76 0.40 0.40

5.5FL 0.5FL 0.9FL+PH 71FL 56FL 56FL 29FL+LL 50FL

Excitation in near-UV. bThe phosphorescence lifetime was 250 μs. Excitation at 495 nm. dPKAc was added in small excess. FL, fluorescence; PH, phosphorescence; LL, long-lifetime luminescence. a c

The efficiency of energy transfer from 3D* to 1A was evaluated on the basis of the luminescence lifetimes of the probes according to eq 4 τ E = 1 − DA τD (4) where E is the FRET efficiency; τDA is the luminescence lifetime of the donor in the presence of the acceptor and τD is the luminescence lifetime of the donor in absence of the acceptor. The calculated FRET efficiencies from 3D* to 1A for Se(Cy3B) G


Article


from two processes: first, excitation of the donor (1D → 1D*) followed by nonradiative energy transfer from 1D* to 1A and emission of light from 1A*, and second, direct excitation of 1A (1A → 1A*) followed by emission of light from 1A*. The QY of Se(Cy3B), if excited at 495 nm, represents the QY of fluorescence from directly excited acceptor dye only. The two QYs determined at different excitation wavelengths were equal for the nonbound Se(Cy3B), 56%, which points to strong interaction between the two luminophores and is in accord with the calculated efficiency for the intramolecular singlet−singlet energy transfer (90%) in nonbound Se(Cy3B). In the case of PKAc-bound Se(Cy3B), these two QYs differ; if excited in the near-UV region, the QY is 29%, and if excited at 495 nm the QY is 50%. In the case of illumination with near-UV radiation, the decrease of QY of Se(Cy3B) upon binding to PKAc points to less efficient energy transfer from 1D* to 1A in PK-bound state than in nonbound state, which could result from increased distance between the donor and that of the acceptor. This means that more of the excited donor-luminophores in Se(Cy3B) can enter ISC in complex with PKAc if compared to the free probe Se(Cy3B), resulting in increased QY of microsecond-scale luminescence of Se(Cy3B) in complex with PKAc. Taking into account the FRET efficiency for triplet−singlet energy transfer in PK-bound Se(Cy3B) as determined by eq 4, the molar extinction coefficients of the donor and that of the acceptor at 350 nm, and the phosphorescence QY of the donor (0.5%) and the QY of the acceptor (50%), it was possible to calculate the QY of donor-sensitized emission (including both singlet−singlet and triplet−singlet energy transfer) for PKbound Se(Cy3B) in nondeoxygenated conditions, which was 26%. On the basis of this value, it was estimated that the maximal efficiency of ISC and the corresponding QY of microsecond-scale luminescence of PK-bound Se(Cy3B) are 68% and 10%, respectively (SI). The calculated Förster distances and FRET efficiencies suggest higher proportion of triplet−singlet energy transfer and higher efficiency of ISC for PK-bound Se(PF647) if compared to PK-bound Se(Cy3B). It was estimated that if efficiency of ISC is 80%, the QYs of microsecond-scale luminescence of PK-bound Se(PF647) are 17% and 22% in nondeoxygenated and in enzymatically deoxygenated conditions, respectively (SI). In comparison, the QYs of the Eu-based probes for biological assays ranged between 24 and 31% in water at 22 °C.40 However, in a practical application the luminescence intensity of a different Eu-cryptate was smaller than that of a PK-bound ARCLum(Fluo) probe with a thiophene-comprising donor moiety.41 Deoxygenation of the solution containing the ARC-Lum(−) probe Se(−) by argon-purge, increased the QY of luminescence (including both phosphorescence and fluorescence) by almost 2-fold from 0.5% to 0.9% and resulted in approximately 1:1 ratio of fluorescence and phosphorescence (Figure S7 in SI). This result showed that the overall enhancement of QY was caused by decreased quenching of 3D* by dissolved molecular oxygen and that the QY of emission from 1D* remained almost unaffected after argon-purge. Deoxygenation of the solution of the thiophene-comprising probe S(−) (c = 5.9 μM) was also performed by argon-purge, but no phosphorescence could be detected, which points to the absence or very low phosphorescence QY of the thiophenecomprising ARC-Lum(−) probe. The measurements on plate reader showed that the luminescence brightness of S(−) in

in complex with PKAc were 29.4 ± 5.3% and 64.5 ± 2.5% in nondeoxygenated and enzymatically deoxygenated solution, respectively. These results are in accordance with the theory and show that if other modes of triplet deactivation are less competitive, the efficiency of the energy transfer from 3D* to 1 A increases as the luminescence lifetime of 3D* increases.38 For comparison, we also characterized the luminescence of Se(PF647)/PKAc complex in enzymatically deoxygenated conditions. Se(PF647) is a selenophene-comprising ARCLum(Fluo) probe conjugated with the fluorescent dye PF647 instead of the dye Cy3B [probe Se(Cy3B)]. The absorption spectrum of PF647 well overlaps with the phosphorescence emission spectrum of 3D* in Se(−) (Figure 2B). Also, the molar extinction coefficient at the absorption maximum of PF647 (ε = 250 00 M−1cm−1) is almost 2-fold higher than that of Cy3B at its absorption maximum (ε = 130 000 M−1cm−1). The Förster distances calculated for singlet−singlet and triplet− singlet energy transfer in Se(PF647) were 2.0 and 3.2 nm, respectively. As a result, the energy transfer from 3D* to 1A is more efficient for Se(PF647)/PKAc complex (69.4 ± 5.8% in the presence of dissolved molecular oxygen) than for Se(Cy3B)/PKAc complex. Because of the efficient FRET, rather fast depopulation of the 3D* takes place in the presence of PF647 and deoxygenation of the solution has smaller impact on the luminescence lifetime of Se(PF647) in complex with PKAc: the luminescence lifetime of Se(PF647)/PKAc complex increased only by 14 μs upon deoxygenation from 33 to 47 μs (Figure 5 and Table 1). The calculated FRET efficiency for Se(PF647)/PKAc complex in deoxygenated solution was 89.5 ± 4.4%. This result shows that PF647 is a very good acceptor dye for 3D* in Se(PF647). A thiophene-comprising ARC-Lum(Fluo) probe S(PF647) showed more than five-fold increase of luminescence lifetime in complex with PKAc after enzymatic deoxygenation from 109 to 556 μs (Figure 5 and Table 1). The greater increase of luminescence lifetime of S(PF647)/PKAc complex upon deoxygenation if compared to that of Se(PF647)/PKAc complex shows that the selenophene-containing 3D* is more appropriate donor for the dye PF647 than the thiophenecontaining 3D*. On the other hand, the long luminescence lifetime of S(PF647)/PKAc allows for the application of a long delay after pulse excitation without significant loss in the luminescence signal intensity (for example, in the case of using a xenon flash-lamp with long after-glow for pulsed excitation of the probe). Quantum Yields. The QYs of the luminophores were determined by application of the relative method. The chosen reference compounds possessed good overlap of absorption as well as luminescence emission spectra with those of the luminophores in ARC-Lum probes. The determined QYs are listed in Table 2. If Se(Cy3B) is illuminated with near-UV radiation, both the donor- and acceptor-luminophores are directly excited {ε350[Se(−)] = 14 500 M−1 cm−1, ε350[Cy3B] = 2300 M−1 cm−1} and produce luminescence, whereas if Se(Cy3B) is excited at wavelengths above 450 nm, direct excitation of the acceptorfluorophore Cy3B occurs {ε495[Se(−)] < 200 M−1 cm−1, ε495[Cy3B] > 20 000 M−1 cm−1}, leading to fluorescence emission with short decay time. If illuminated with near-UV radiation, some of the 1D* can enter ISC and form 3D*, which is almost entirely quenched by oxygen in nondeoxygenated solution in case of free Se(Cy3B) molecule. Hence, the QY of free Se(Cy3B), if illuminated with near-UV radiation, results H



■

complex with PKAc is roughly 20 times smaller than that of Se(−) in complex with PKAc (Figure S8 in SI). On the other hand, fluorescence QY of the probe S(−) was higher than that of its selenophene-comprising counterpart, probe Se(−). As a rule, the compounds with higher phosphorescence reveal weaker fluorescence, and therefore it is impossible to construct highly phosphorescent probes by reference to fluorescent properties of the compound.

Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.6b03342. Chemical synthesis of Cy3B-r9-NH2 and the full structures of the probes; the kinetics of the enzymatic deoxygenation in the unsealed system; the temperaturedependence of ARC-Lum probes; the decay curves of long-lifetime luminescence of the nonbound ARCLum(Fluo) probe Se(Cy3B); the comparison of phosphorescence spectra of samples containing nonbound probe Se(−) in deoxygenated conditions or PKbound probe Se(−); the steady-state emission spectrum of the donor-compound Se(−) in the absence of dissolved molecular oxygen, which illustrates the ratio of phosphorescence and fluorescence signal of Se(−); the comparison of luminescence brightness of ARCLum(−) probes S(−) and Se(−) in complex with PKAc; the calculation of the proportions of the luminescence resulting from the direct excitation and the donorsensitized emission of the acceptor Cy3B in PK-bound Se(Cy3B) in nondeoxygenated conditions and the estimation of QYs of PK-bound ARC-Lum(Fluo) probes. (PDF)

■

CONCLUSIONS This work further discloses photoluminescent properties of unique PK binding-responsive organic probes, ARC-Lum(Fluo) probes, that possess luminescence decay time in the microsecond range. An ARC-Lum(Fluo) probe is a tandemluminophore that incorporates a low-QY donor phosphor linked to a long-wavelength short-lifetime acceptor with high QY. The long-lifetime donor induces slow acceptor decay, which is due to FRET from the exited triplet state of the donor (3D*) to the singlet state of the acceptor. The energy is stored in 3D* and released gradually via FRET to the acceptor fluorescent dye, leading to 1A* of the latter and the following light emission from the dye. Therefore, the emission spectrum of an ARC-Lum(Fluo) probe coincides with the fluorescence spectrum of the dye it incorporates, but the emission decay follows the persistence of the excited triplet state of the donor phosphor. The current study showed that additionally ARC-Lum(Fluo) probes possess singlet−singlet energy transfer from 1D* to 1A. These two types of energy transfer (triplet−singlet and singlet− singlet, respectively) occur on a different time scale and it was discovered here that their proportion and efficiency depends on the environment surrounding the luminophores. If the probe is not bound to a PK, singlet−singlet energy transfer is efficient and 3D* is rapidly quenched by dissolved molecular oxygen and buffer components. Upon binding to a protein kinase, energy transfer from 1D* to 1A decreases and the probability of ISC increases. Furthermore, inside the active site of a PK 3D* is hindered from quenchers, such as dissolved molecular oxygen, leading to longer lifetime of 3D* and enhanced efficiency of FRET from 3 D* to 1 A. However, the long-lifetime luminescence decay time and thereby the luminescence of the ARC-Lum/PK complex can be further increased by deoxygenation of the buffer solution. Luminescence lifetimes as long as half a millisecond were determined for ARC-Lum(Fluo) probes in complex with a PK in enzymatically deoxygenated conditions on unsealed microplates, making the probes well suitable for high-throughput studies with commonly used xenon flash lamp as the excitation source. This study points to a possibility of how to circumvent the low QY of organic phosphors in aqueous solution at room temperature. An efficient FRET from the excited triplet state 3 D* of the donor phosphor to a bright fluorescent acceptor dye leads to substantially increased effective QY (here more than 50-fold increase was obtained) of the donor−acceptor pair, resulting in bright tandem probes possessing long-lived (microsecond-scale) emission, large pseudo-Stokes shift, and emission in orange or red spectral region. Probes with microsecond-scale luminescence lifetime can be applied with TR luminescence measurement techniques, which effectively reduce the strong autofluorescence of biological samples that have mainly nanosecond-scale lifetime, leading to increased sensitivity of the assay.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +372 737 5275. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Professor Sergei A. Vinogradov for hosting K. Ligi at the University of Pennsylvania, where she performed the photophysical measurements. We appreciate Professor S. Vinogradov’s interest in and support to the study of photoluminescence of ARC-Lum probes. We would also like to thank Toonika Rinken at the University of Tartu, who helped us to determine the enzymatic activity of glucose oxidase, thereby enabling us to choose sufficient amount of the enzyme to perform enzymatic deoxygenation of samples with protein kinases. This research has been supported by grants from the Estonian Ministry of Education and Research (SF0180121s08) and Estonian Research Council (former Estonian Science Foundation) (ETF8230, ETF8419, IUT2017); by the Graduate School “Functional Materials and Technologies” receiving funding from the European Social Fund under project 1.2.0401.09-0079; and European Social Fund’s Doctoral Studies and Internationalisation Programme DoRa, which is carried out by Foundation Archimedes.

■

REFERENCES

(1) Amoroso, A. J.; Pope, S. J. A. Using Lanthanide Ions in Molecular Bioimaging. Chem. Soc. Rev. 2015, 44 (14), 4723−4742. (2) Rajendran, M.; Miller, L. W. Evaluating the Performance of Time-Gated Live-Cell Microscopy with Lanthanide Probes. Biophys. J. 2015, 109 (2), 240−248. (3) Murphy, L.; Congreve, A.; Pålsson, L.-O.; Williams, J. a G. The Time Domain in Co-Stained Cell Imaging: Time-Resolved Emission Imaging Microscopy Using a Protonatable Luminescent Iridium Complex. Chem. Commun. (Cambridge, U. K.) 2009, 46, 8743−8745.

I


Article

The Journal of Physical Chemistry B (4) Thibon, A.; Pierre, V. C. Principles of Responsive LanthanideBased Luminescent Probes for Cellular Imaging. Anal. Bioanal. Chem. 2009, 394 (1), 107−120. (5) Mukherjee, S.; Thilagar, P. Recent Advances in Purely Organic Phosphorescent Materials. Chem. Commun. 2015, 51 (55), 10988− 11003. (6) Allain, C.; Faulkner, S. Photophysical Approaches to Responsive Optical Probes. Future Med. Chem. 2010, 2 (3), 339−350. (7) Piszczek, G. Luminescent Metal-Ligand Complexes as Probes of Macromolecular Interactions and Biopolymer Dynamics. Arch. Biochem. Biophys. 2006, 453 (1), 54−62. (8) Papkovsky, D. B.; O’Riordan, T. C. Emerging Applications of Phosphorescent Metalloporphyrins. J. Fluoresc. 2005, 15 (4), 569−584. (9) Enkvist, E.; Vaasa, A.; Kasari, M.; Kriisa, M.; Ivan, T.; Ligi, K.; Raidaru, G.; Uri, A. Protein-Induced Long Lifetime Luminescence of Nonmetal Probes. ACS Chem. Biol. 2011, 6 (10), 1052−1062. (10) Ekambaram, R.; Enkvist, E.; Manoharan, G. B.; Ugandi, M.; Kasari, M.; Viht, K.; Knapp, S.; Issinger, O.-G.; Uri, A. Benzoselenadiazole-Based Responsive Long-Lifetime Photoluminescent Probes for Protein Kinases. Chem. Commun. (Cambridge, U. K.) 2014, 50, 4096−4098. (11) Kasari, M.; Ligi, K.; Williams, J. A. G.; Vaasa, A.; Enkvist, E.; Viht, K.; Pålsson, L. O.; Uri, A. Responsive Microsecond-Lifetime Photoluminescent Probes for Analysis of Protein Kinases and Their Inhibitors. Biochim. Biophys. Acta, Proteins Proteomics 2013, 1834 (7), 1330−1335. (12) Vaasa, A.; Ligi, K.; Mohandessi, S.; Enkvist, E.; Uri, A.; Miller, L. W. Time-Gated Luminescence Microscopy with Responsive Nonmetal Probes for Mapping Activity of Protein Kinases in Living Cells. Chem. Commun. 2012, 48 (68), 8595−8597. (13) Ekambaram, R.; Manoharan, G. B; Enkvist, E.; Ligi, K.; Knapp, S.; Uri, A. PIM Kinase-Responsive Microsecond-Lifetime Photoluminescent Probes Based on Selenium-Containing Heteroaromatic Tricycle. RSC Adv. 2015, 5 (117), 96750−96757. (14) Lakowicz, J. R.; Piszczek, G.; Kang, J. S. On the Possibility of Long-Wavelength Long-Lifetime High-Quantum-Yield Luminophores. Anal. Biochem. 2001, 288 (1), 62−75. (15) Maliwal, B. P.; Gryczynski, Z.; Lakowicz, J. R. Long-Wavelength Long-Lifetime Luminophores. Anal. Chem. 2001, 73 (17), 4277−4285. (16) Kang, J. S.; Piszczek, G.; Lakowicz, J. R. Enhanced Emission Induced by FRET from a Long-Lifetime, Low Quantum Yield Donor to a Long-Wavelength, High Quantum Yield Acceptor. J. Fluoresc. 2002, 12 (1), 97−103. (17) Gryczynski, Z.; Maliwal, B. P.; Gryczynski, I.; Lakowicz, J. R. Long-Wavelength Long-Lifetime Luminophores for Cellular and Tissue Imaging. Proc. SPIE 2004, 5323, 88−98. (18) Förster, T. 10th Spiers Memorial Lecture. Transfer Mechanisms of Electronic Excitation. Discuss. Faraday Soc. 1959, 27, 7−17. (19) Galley, W. C.; Stryer, L. triplet−singlet Energy Transfer in Proteins. Biochemistry 1969, 8 (5), 1831−1838. (20) Isenberg, I.; Leslie, R. B.; Baird, S. L. J.; Rosenbluth, R.; Bersohn, R. Delayed Fluorescence in DNA-Acridine Dye Complexes. Proc. Natl. Acad. Sci. U. S. A. 1964, 52, 379−387. (21) Bennett, R. G.; Schwenker, R. P.; Kellogg, R. E. Radiationless Intermolecular Energy Transfer. II. Triplet→Singlet Transfer. J. Chem. Phys. 1964, 41 (10), 3040−3041. (22) Kellogg, R. E.; Bennett, R. G. Radiationless Intermolecular Energy Transfer. III. Determination of Phosphorescence Efficiencies. J. Chem. Phys. 1964, 41 (10), 3042−3045. (23) Pflug, A.; Rogozina, J.; Lavogina, D.; Enkvist, E.; Uri, A.; Engh, R. A.; Bossemeyer, D. Diversity of Bisubstrate Binding Modes of Adenosine Analogue-Oligoarginine Conjugates in Protein Kinase a and Implications for Protein Substrate Interactions. J. Mol. Biol. 2010, 403 (1), 66−77. (24) Enkvist, E.; Kriisa, M.; Roben, M.; Kadak, G.; Raidaru, G.; Uri, A. Effect of the Structure of Adenosine Mimic of Bisubstrate-Analog Inhibitors on Their Activity Towards Basophilic Protein Kinases. Bioorg. Med. Chem. Lett. 2009, 19 (21), 6098−6101.

(25) Vaasa, A.; Viil, I.; Enkvist, E.; Viht, K.; Raidaru, G.; Lavogina, D.; Uri, A. High-Affinity Bisubstrate Probe for Fluorescence Anisotropy Binding/Displacement Assays with Protein Kinases PKA and ROCK. Anal. Biochem. 2009, 385 (1), 85−93. (26) Vinogradov, S. A.; Fernandez-Searra, M. A.; Dugan, B. W.; Wilson, D. F. Frequency Domain Instrument for Measuring Phosphorescence Lifetime Distributions in Heterogeneous Samples. Rev. Sci. Instrum. 2001, 72 (8), 3396−3406. (27) Kagan, M.; Kivirand, K.; Rinken, T. Modulation of Enzyme Catalytic Properties and Biosensor Calibration Parameters with Chlorides: Studies with Glucose Oxidase. Enzyme Microb. Technol. 2013, 53 (4), 278−282. (28) Lakowicz, J. R. Theory of Energy Transfer for a Donor-Acceptor Pair. In Principles of Fluorescence Spectroscopy; Lakowicz, J. R., Ed.; Springer: Boston, MA, 2006; pp 445−451. (29) Morris, J. V.; Mahaney, M. A.; Huber, J. R. Fluorescence Quantum Yield Determinations. 9,10-Diphenylanthracene as a Reference Standard in Different Solvents. J. Phys. Chem. 1976, 80 (9), 969−974. (30) Kubin, R.; Fletcher, A. Fluorescence Quantum Yields of Some Rhodamine Dyes. J. Lumin. 1982, 27 (4), 455−462. (31) Zander, M.; Kirsch, G. On the Phosphorescence of Benzologs of Furan, Thiophene, Selenophene, and Tellurophene - a Systematic Study of the Intra-Annular Internal Heavy-Atom Effect. Z. Naturforsch., A: Phys. Sci. 1989, 44 (3), 205−209. (32) Kremer, A.; Aurisicchio, C.; De Leo, F.; Ventura, B.; Wouters, J.; Armaroli, N.; Barbieri, A.; Bonifazi, D. Walking Down the Chalcogenic Group of the Periodic Table: From Singlet to Triplet Organic Emitters. Chem. - Eur. J. 2015, 21 (43), 15377−15387. (33) Kowalska-Baron, A.; Gałęcki, K.; Wysocki, S. Photophysics of Indole-2-Carboxylic Acid (I2C) and Indole-5-Carboxylic acid (I5C): Heavy Atom Effect. Spectrochim. Acta, Part A 2013, 116, 183−195. (34) Banks, D. D.; Kerwin, B. A. A Deoxygenation System for Measuring Protein Phosphorescence. Anal. Biochem. 2004, 324 (1), 106−114. (35) Baumann, R. P.; Penketh, P. G.; Seow, H. A.; Shyam, K.; Sartorelli, A. C. Generation of Oxygen Deficiency in Cell Culture Using a Two-Enzyme System to Evaluate Agents Targeting Hypoxic Tumor Cells. Radiat. Res. 2008, 170 (5), 651−660. (36) Englander, S. W.; Calhoun, D. B.; Englander, J. J. Biochemistry without Oxygen. Anal. Biochem. 1987, 161 (2), 300−306. (37) Lower, S. K.; El-Sayed, M. A. The Triplet State and Molecular Electronic Processes in Organic Molecules. Chem. Rev. 1966, 66 (2), 199−241. (38) Rohatgi-Mukherjee, K. K. Fundamentals of Photochemistry; New Age International: New Delhi, 1978; p 197. (39) Lakowicz, J. R. Energy Transfer Efficiency From Enhanced Acceptor Fluorescence. In Principles of Fluorescence Spectroscopy; Springer: Boston, MA, 2006; pp 461−462. (40) Butler, S. J.; Delbianco, M.; Lamarque, L.; McMahon, B. K.; Neil, E. R.; Pal, R.; Parker, D.; Walton, J. W.; Zwier, J. M. EuroTracker® Dyes: Design, Synthesis, Structure and Photophysical Properties of Very Bright Europium Complexes and Their Use in Bioassays and Cellular Optical Imaging. Dalton Trans. 2015, 44 (11), 4791−4803. (41) Kasari, M.; Padrik, P.; Vaasa, A.; Saar, K.; Leppik, K.; Soplepmann, J.; Uri, A. Time-Gated Luminescence Assay Using Nonmetal Probes for Determination of Protein Kinase Activity-Based Disease Markers. Anal. Biochem. 2012, 422 (2), 79−88.

J


Deoxygenation Increases Photoluminescence Lifetime of Protein

Recommend Documents