- Email: [email protected]
Singlet oxygen produced by quasi-continuous photo-excitation of hypericin in dimethyl-sulfoxide
Journal of Luminescence 177 (2016) 17–21
Contents lists available at ScienceDirect
Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin
Full Length Article
Singlet oxygen produced by quasi-continuous photo-excitation of hypericin in dimethyl-sulfoxide J. Varchola a, K. Želonková a, D. Chorvat Jr b, D. Jancura a,c, P. Miskovsky a,b,c, G. Bánó a,c,n a
Department of Biophysics, Faculty of Science, P. J. Šafárik University, Jesenná 5, 041 54 Košice, Slovak Republic International Laser Centre, Ilkovicova 3, 841 05 Bratislava, Slovak Republic c Center for Interdisciplinary Biosciences, Faculty of Science, P. J. Šafárik University, Jesenná 5, 041 54 Košice, Slovak Republic b
art ic l e i nf o
a b s t r a c t
Article history: Received 24 November 2015 Received in revised form 8 April 2016 Accepted 11 April 2016 Available online 14 April 2016
Singlet oxygen (O2(1Δg)) production by photo-excited hypericin (Hyp) dissolved in dimethyl-sulfoxide (DMSO) was studied by means of time-resolved phosphorescence measurements. In order to minimize photo-bleaching, the samples were excited in quasi-continuous mode using long-pulse (35 μs) laser excitation. The measured lifetime of singlet oxygen is τΔ ¼5.5 70.3 μs. This result helps to resolve the discrepancy existing in the literature concerning singlet oxygen lifetime in DMSO. The obtained quantum yield of singlet oxygen photosensitized by Hyp in air-saturated DMSO is ΦΔ ¼ 0.47 0.03. The rate constant for Hyp triplet state depopulation in reaction with ground state molecular oxygen is measured to be kq ¼ 1.67 0.3 109 M 1 s 1. & 2016 Published by Elsevier B.V.
Keywords: Singlet oxygen Hypericin Photosensitizer Dimethyl-sulfoxide
1. Introduction Singlet oxygen is a highly reactive molecule which causes oxidation of many biologically important compounds [1]. From medicinal point of view, singlet oxygen has been studied intensively because of its essential role in photodynamic therapy (PDT) of cancer. During PDT, the oxidative stress induced by reactive oxygen species (in most cases by singlet oxygen in the O2(1Δg) state) initiates processes leading to cell death [2,3]. Singlet oxygen is mostly produced by energy transfer between light-activated (triplet state) photosensitizer molecules and ground state (triplet, O2(3Σg )) oxygen molecules [1]. The rate of singlet oxygen deactivation shows significant variation depending on the environment [4]. Singlet oxygen lifetime values ranging from a few microseconds (e.g. 3-4 μs in water) to milliseconds (e.g. 59 ms in airsaturated CCl4) were reported for different pure solvents [4,5]. Dimethyl-sulfoxide (DMSO) is clinically used as a penetration enhancer to carry drugs into tissues [6,7] and is known as an efficient solvent for many hydrophobic compounds. Due to these reasons, the information on singlet oxygen properties in DMSO is highly important. In spite of the large number of singlet oxygen lifetime data published during the last four decades, relatively few papers deal with the deactivation rate of singlet oxygen in DMSO. n Corresponding author at: Center for Interdisciplinary Biosciences, Faculty of Science, P. J. Šafárik University, Jesenná 5, 041 54 Košice, Slovak Republic. E-mail address: [email protected] (G. Bánó).
http://dx.doi.org/10.1016/j.jlumin.2016.04.020 0022-2313/& 2016 Published by Elsevier B.V.
Moreover, the available data scatter significantly. Lifetime values of 1.8 μs [8], 5.6 μs [9] and 30 μs [10] were published by other authors, while a lifetime value of 19 μs can be calculated from the quenching rate reported in [11]. One of the objectives of the present work is to provide additional experimental data that help to clarify the contradiction between the previous reports. Singlet oxygen in the O2(1Δg) state can be detected through its phosphorescence around 1270 nm. The lifetime of singlet oxygen is readily determined by time-resolved phosphorescence decay measurements following pulsed photo-activation of photosensitizer molecules [12]. Most of the previous experiments were carried out with short-pulse laser excitation using laser pulses in the nanosecond or sub-nanosecond range (see e.g. [1,12–17]). An alternative excitation scheme using longer (microsecond) laser pulses was applied by Lee and co-workers [18,19]. In our previous work, the duration of the laser pulses was set long enough so that steady-state concentrations of the photosensitizer's triplet state and O2(1Δg) were reached at the end of the laser pulse [20]. The major advantage of this so-called quasi-continuous excitation scheme was that the peak laser power was reduced significantly when compared to laser pulses of nanosecond duration, which resulted in a lower bleaching rate of the used photosensitizers. Moreover, steady-state conditions at the end of the laser pulse allowed for analytical solution of the differential equations describing the time-evolution of O2(1Δg) concentration after the excitation pulse [20].
18
J. Varchola et al. / Journal of Luminescence 177 (2016) 17–21
Photo-excited hypericin (Hyp) was used as a source of the singlet oxygen in this work. Hyp is a naturally occurring potent photosensitizer that is being studied extensively for its possible application in PDT [21–25]. Under light illumination Hyp causes anti-proliferative and cytotoxic effects in many tumor cell lines. Singlet oxygen is supposed to play the major role in the photoactivity of Hyp at aerobic conditions. O2(1Δg) generation via energy transfer from Hyp triplet state (3Hyp) to ground state oxygen molecules was theoretically predicted [26] as well as experimentally confirmed in several organic solvents and lipid environments [27–29]. The corresponding quantum yields for singlet oxygen formation by Hyp vary between 0.25 and 0.45 [27– 29]. Stock solutions of Hyp used in PDT-related experiments are often prepared in DMSO [30–33]. On the other hand, Hyp forms non-fluorescence aggregates in aqueous environment [21,34]. These aggregates are not able to produce singlet oxygen [15,35]. In the present work, besides the determination of singlet oxygen lifetime in DMSO, we report also on the rate of 3Hyp deactivation by oxygen in DMSO and the quantum yield of singlet oxygen production by Hyp in this solvent.
2. Experimental Hypericin, Rose Bengal, dimethyl-sulfoxide (4 99.9%), RuPhen (dichlorotris(1,10-phenantroline) ruthenium(II) hydrate) and protoporphyrin IX disodium salt (PpIX) were purchased from SigmaAldrich. The purity of DMSO was checked by measuring its freezing point before the experiments. The results (17.570.5 °C) indicated water content of about 0.5 wt%. According to the work of Staicu et al. [9], the effect of this water content on singlet oxygen lifetime in DMSO is below the error range of the present experiments. The experimental apparatus used for time-resolved phosphorescence measurements has been described in details previously [20]. Briefly, the excitation laser beam (532 nm, Cobolt Samba) passed vertically through a cuvette (1 1 4 cm) filled with 2 ml of the sample solution. The cuvette holder was thermostated at 22 °C. The sample was continuously flushed with a nitrogen/oxygen mixture (at 25 standard cubic centimeter per minute, sccm) and was stirred by a magnetic stirrer. An acousto-optic modulator (AOM, Isomet 1205C) was used to switch the excitation laser beam (on/off) creating 35 μs long laser pulses with a repetition rate of 8 kHz. The average laser power on the sample was 1.4 mW. The switching time of the AOM was approx. 400 ns. The emitted phosphorescence light was passing through a 1250–1350 nm bandpass filter and was detected with a Hamamatsu H10330A-75 photomultiplier tube (PMT) operated in photon counting mode. The time evolution of the photon rate was acquired with a multichannel scaler (Becker & Hickl, MSA-300). Typically, a signal of 3.106 pulses was collected at each experimental condition. The obtained time-dependences were corrected for the phosphorescence background of the optical components as measured in pure DMSO. In case of RuPhen measurements the sample was excited by a 476 nm (Coherent 90 C FreD) laser. The repetition rate of 20 μs long laser pulses was set to 2 kHz and the average laser power at the sample was 0.7 mW. The concentration of oxygen in the sample was calculated from the gas-phase oxygen pressure [36]. Rose Bengal (RB) was used to calibrate the experimental apparatus for quantum yield measurements of singlet oxygen production (see below). The calibration was carried out with a RB solution (in DMSO) having the same absorbance (0.022 per cm at 532 nm) as the used sample of Hyp. The quantum yield of singlet oxygen production by RB in DMSO
was estimated to be 0.76 (see also [37]) for air-saturated conditions. The lifetime data obtained with quasi-continuous excitation were compared with those measured by conventional short-pulse experiments. The apparatus was equipped with a Q-switched Nd: YAG laser (Spectra-Physics Quanta-Ray INDI-40-10) pumping an OPO (optical parametric oscillator, Spectra-Physics basiScan) that was tuned to 532 nm. The repetition rate of the 5–7 ns long laser pulses was 10 Hz producing an average beam power of 1.2 mW at the sample. The phosphorescence signal of 3000 pulses was collected at each experimental condition.
3. Data analysis The decay of the O2(1Δg) concentration [Δ] after long pulse (quasi-continuous) excitation is described by [20]: t t S τΔ ½Δ ¼ 0 ð1Þ τ e τΔ τ T e τT ; τΔ τT Δ where S0 is the rate of O2(1Δg) production at steady-state conditions, and τΔ and τT are the lifetimes of O2(1Δg) and the photosensitizer's triplet state, respectively. The O2(1Δg) phosphorescence intensity detected at 1270 nm is proportional to [Δ] through an equipment specific factor c and the O2(1Δg) phosphorescence emission rate constant kΔe: t τΔ τtΔ e ð2Þ τ e τ T e τT ; I 1270 meas ¼ ckΔ ½Δ ¼ F τΔ τT Δ where F¼ S0kΔec. The given time dependence differs from the kinetic curves used to analyze short-pulse (nanosecond) experiments [12], where (following the short excitation pulse) the phosphorescence signal first increases, then reaches its maximum and finally decays. By contrast, dependence (2) gives a monotonous decrease of the phosphorescence intensity with its time derivative being zero at t¼0 (right after the excitation pulse). Fig. 1 shows a typical experimental O2(1Δg) phosphorescence decay curve obtained with quasi-continuous excitation of Rose Bengal dissolved in DMSO using 35 μs excitation pulses. Rose Bengal was chosen as a photosensitizer here because (unlike Hyp) it has a low level of fluorescence at 1270 nm. Right after turning the laser on (at t ¼ 35 μs in Fig. 1) the fluorescence of RB appears instantaneously. The onset of the fluorescence is mostly limited by the switching time of the AOM. During the laser pulse the signal increases due to a gradually increasing contribution from singlet oxygen phosphorescence, which reaches steady-state intensity at the end of the pulse. When the laser is switched off (at t ¼0) the
Fig. 1. Typical time-dependence of the emission signal detected around 1270 nm during long-pulse excitation of Rose Bengal (not all points are shown). The red curve represents data fit according to (2). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
J. Varchola et al. / Journal of Luminescence 177 (2016) 17–21
19
Fig. 2. Testing the fitting procedure on calculated (model) phosphorescence decay data. Panels a) and b): Model decay curves (open circles) calculated according to (2) for two different τΔ, τT pairs as indicated. The colored lines in a) and b) represent the results of data fits using (2) with fixed values of τΔfixed. Panels c) and d): Pairs of τΔfixed and the corresponding τTfitted obtained by fitting the model curves of a) and b), respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
fluorescence disappears quickly and the phosphorescence of singlet oxygen decays according to (2) – see the red curve in Fig. 1. It can be shown, that the shape of (2) is identical to the decay phase of phosphorescence kinetic curves after short-pulse excitation [12]. It follows that the same general rules apply when fitting experimental data with (2). Namely, as emphasized in previous works [12,38,39], one has to be cautious about determining τΔ and τT by simply fitting the phosphorescence decay with (2). The related problem is twofold. First, exchanging τΔ and τT in (2) does not change the curve shape, which may lead to confusion between these two values. Due to this fact it is usually recommended to determine the photosensitizer's decay time τT from an independent experiment [38,39]. Second, when τΔ and τT are about the same (τΔ E τT) the decay curve (2) can be well approximated by different combinations of τΔ, τT pairs. This is demonstrated in Fig. 2 showing decay data calculated according to (2) for two different τΔ, τT pairs. The open circles in Fig. 2a belong to τΔ ¼5 and τT ¼4 (in arbitrary time units), while τΔ ¼ 5 and τT ¼1 were used to calculate the data points in Fig. 2b. The two data sets are normalized to unity at t¼0 and both are shown decreasing by 1.5 orders of magnitude. The lines plotted in Fig. 2a and b represent the results of data fits. During the fitting procedure τΔ was fixed (at different τΔfixed values ranging from 3 to 6.5) and only τT was varied. The combinations of τΔfixed and τTfitted obtained this way are shown in Fig. 2c and d as they belong to the fitted curves in Fig. 2a and b, respectively. It can be seen that for τΔ ¼5 and τT ¼4 (Fig. 2a) all the different data fits reproduce the original data points well. Only four curves are shown here (out of seven), the remaining ones being identical to them. Taking into account the usual experimental noise, it is concluded that the uncertainty of determining τΔ and τT by fitting experimental data with (2) can be high for τΔ E τT. By contrast, for τΔ ¼5 and τT ¼ 1 (Fig. 2b) the modeled decay curve is only reproduced well when τΔfixed and τTfitted stay close to their original value (fit #4 in Fig. 2b and d). When τΔ c τT the late phosphorescence decay is governed by τΔ,
which is the optimal condition for reliable singlet oxygen lifetimes measurements. Fitting experimental phosphorescence decay curves with (2) one obtains τΔ, τT and a value F that is proportional to the steadystate rate of O2(1Δg) production, S0. The fitted F values can be used to measure the quantum yield of O2(1Δg) production ΦΔ. For that the apparatus is to be calibrated with a photosensitizer of known quantum yield (for details see [20] and references cited therein). As mentioned in the Section 2 RB dissolved in DMSO was used for calibration in this work. The reciprocal value of τT can be written as: 1
τT
¼ k0 þkq ½O2 ;
ð3Þ
where k0 is the rate constant of 3Hyp quenching in absence of oxygen, kq represents the rate constant for 3Hyp depopulation in reactions with ground state molecular oxygen and [O2] denotes the molecular oxygen concentration in the sample. Based on (3) k0 and kq can be determined by measuring τT at different oxygen concentrations. Knowing k0 and kq the fraction of 3Hyp quenched in reaction with oxygen can be calculated as: 2 PO T ¼
kq ½O2 k0 þ kq ½O2
ð4Þ
Finally, the quantum yield of singlet oxygen production ΦΔ is 3 2 connected to P O T through the quantum yield of Hyp photo-excitation, ΦT, and the fraction of 3Hyp quenched by oxygen yielding T singlet oxygen, f Δ , as [20]:
ΦΔ ¼ ΦT P OT 2 f TΔ :
ð5Þ
Assuming ΦT and f Δ being independent of oxygen concentration, substitution of (4) into (5) gives the oxygen concentration dependence of ΦΔ. T
20
J. Varchola et al. / Journal of Luminescence 177 (2016) 17–21
4. Results and discussion The evolution of the near infrared ( 1270 nm) emission signal from the DMSO solution of Hyp (5.10 6 M) during and after the 35 ms long laser pulses was detected for gas-phase oxygen concentrations of 5%, 10%, 20% and 40% in the O2/N2 mixture (see Fig. 3a). The rapid decrease of the emission intensity immediately after the laser is switched-off is attributed to the decrease of Hyp fluorescence. Based on the fluorescence intensity, the extent of Hyp photo-bleaching was estimated to be less than 3% during the experiments, as it was determined from two consecutive measurements on each sample (not shown). Both the 3Hyp and O2(1Δg) lifetimes were determined by fitting singlet oxygen phosphorescence decays with Eq. (2). As shown in the Data analysis section 3, reliable lifetime values can only be obtained at conditions where τΔ and τT differ significantly. The lifetime of 3Hyp is sensitive to oxygen concentration, being smaller at higher oxygenation level. By contrast, deactivation of O2(1Δg) by ground state oxygen is negligible. It was shown previously that the effect of O2(1Δg) quenching by O2 is only of high importance in solvents with long O2(1Δg) lifetimes (typically in the millisecond range) [4,40]. It can be seen from Fig. 3a that at high oxygen concentration the late decay of O2(1Δg) phosphorescence became mono-exponential. Moreover, the rate of the exponential decay (the slope of the curves on logarithmic scale) did not change between 20% and 40% (gas-phase percentage) of oxygen. All this gave us an evidence of reaching conditions, where τΔ»τT was valid, and the course of the late decay was determined solely by the lifetime of O2(1Δg). Fitting the experimental decay curves at 20 and 40% of oxygen with Eq. (2), the lifetime of singlet oxygen in DMSO was determined to be τΔ ¼5.570.3 μs. The indicated experimental error is mostly due to the uncertainty introduced by the background phosphorescence of the
Fig. 3. a) Experimental phosphorescence decay curves detected around 1270 nm after quasi-continuous excitation of Hyp (5.10 6 M) in DMSO. Excitation: 532 nm, pulse length 35 μs, repetition rate 8 kHz, average laser power 1.4 mW. b) Time course of the phosphorescence signal after short-pulse excitation of Hyp (5.10 6 M) in DMSO. The red curve represents the fitting results (see [12]) with parameters τΔ ¼5.5 μs, τT ¼1.36 μs. Excitation: 532 nm, pulse length 5–7 ns, repetition rate: 10 Hz, average laser power: 1.2 mW. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
optical components that is to be subtracted from the measured signal. A set of control measurements was carried out to verify the obtained lifetime value. First, the Hyp concentration was lowered (to 2.10 6 M) to estimate the rate of O2(1Δg) deactivation by Hyp in the sample. It was shown that the effect of Hyp on τΔ was within the range of the experimental error. Second, the lifetime of singlet oxygen was measured at lower laser power (0.7 mW average) to exclude the effect of possible photoproducts on τΔ. Third, the measurements were repeated with two other photosensitizers, Rose Bengal and PpIX, dissolved in DMSO. Finally, the lifetime of singlet oxygen was evaluated from our short-pulse experiments using Hyp dissolved in DMSO (see Fig. 3b). All these measurements confirmed the given value of singlet oxygen lifetime in DMSO (τΔ ¼5.5 μs), which is in good agreement with the result (τΔ ¼ 5.6 μs) of Staicu and co-workers [9]. To establish credibility in the singlet oxygen lifetime data obtained with the quasi-continuous excitation scheme the technique was tested on water solutions, in which singlet oxygen has a similarly short lifetime as in DMSO. Singlet oxygen production by photo-excited Ruphen (1.10 5 M) was measured in 0.9% aqueous NaCl solution (see also Ref. [20]). There is an indication that water salinity does not affect the lifetime of singlet oxygen [41]. Indeed, as shown on Fig. 4, the singlet oxygen lifetime determined by the long-pulse technique at different temperatures and gas-phase oxygen concentrations (10%, 20% and 40%) is in a good agreement with the values published by Jensen and co-workers for pure water [42]. All this gave us confidence, that the quasi-continuous excitation scheme works fine in various solvents. The phosphorescence decays of singlet oxygen measured for low oxygen concentrations in DMSO (5% and 10%, see Fig. 3a) were fitted with Eq. (2) using the above determined τΔ value. Reciprocal values of 3Hyp lifetimes are shown in Fig. 5a as a function of oxygen concentration. The data were fitted by a linear [O2] dependence (3). The rate constant for 3Hyp depopulation in reaction with ground state molecular oxygen was determined to be kq ¼ (1.670.3) 109 M 1 s 1. This value can be compared to that reported for Hyp dissolved in acetonitrile (kq ¼1.4 109 M 1 s 1) [43]. The experimental error for determining the rate constant of 3Hyp quenching in absence of oxygen is relatively large; however, the fitted value k0 ¼2.104 s 1 is in good agreement with that obtained by Darmanyan and co-workers for Hyp dissolved in argon-flushed DMSO (k0 ¼ 1.4 104 s 1) [43]. The phosphorescence intensity of singlet oxygen at t¼0 (which is proportional to steady-state O2(1Δg) concentration) showed little variance with changing oxygen concentration (see Fig. 3a). The corresponding F values obtained by fitting the phosphorescence decay curves with Eq. (2) were used to evaluate the
Fig. 4. Comparison of singlet oxygen lifetime data measured in aqueous environment. Solid symbols: experimental results obtained with quasi-continuous excitation of Ruphen at different gas-phase oxygen concentrations (10%, 20% and 40%). Open squares: reference data taken from Ref. 42.
J. Varchola et al. / Journal of Luminescence 177 (2016) 17–21
21
be kq ¼ (1.670.3) 109 M 1 s 1) has not been published. The quantum yield of singlet oxygen production by Hyp in air-saturated DMSO was determined to be 0.470.03.
Acknowledgments This work was supported by the APVV-0242-11 grant of the Slovak Ministry of Education and the FP7 EU Projects: CELIM 316310 and LASERLAB-EUROPE 284464. This work was also supported by the projects SEPO-II (26220120039) and NanoBioSens (26220220107) of the Operation Program Research and Development funded by the European Regional Development Fund.
References
Fig. 5. a) Reciprocal lifetime of 3Hyp plotted against the oxygen concentration in DMSO fitted with Eq. (3). b) The quantum yield of O2(1Δg) production by Hyp in DMSO. The data points are fitted with the oxygen concentration dependence of (4) substituted to Eq. (5) using k0 and kq values determined by the linear fit of panel a.
quantum yield of O2(1Δg) photosensitized by Hyp in DMSO. The results are shown in Fig. 5b. The theoretical curve (solid line) shown in Fig. 5b was constructed by substituting the measured k0 T and kq to (4) and (5). The product of ΦT f Δ was assumed to be constant and was adjusted to reproduce the measured data points. It is apparent that the oxygen concentrations used in our experi2 ment fell into the saturation region of P O T (see (4)), which explains the weak oxygen dependence of the observed quantum yield data. The measured quantum yield of singlet oxygen production by Hyp in DMSO (at air saturated conditions – or 21% of gas-phase oxygen) is ΦΔ ¼0.4 70.03. The indicated error reflects the uncertainty of the reference quantum yield of singlet oxygen photosensitized by RB in DMSO [37]. The quantum yield obtained for Hyp in DMSO agrees well with that reported for Hyp dissolved in methanol (0.39) [27] and is higher than those measured for Hyp in ethanol (0.35) or acetonitrile (0.25) [29].
5. Conclusions The quasi-continuous excitation scheme was applied successfully for quantitative singlet oxygen measurements providing both excited state lifetime and quantum yield data. Singlet oxygen was produced by photo-excitation of Hyp in DMSO at different oxygen concentrations. The major advantage of working with long laser pulses is that bleaching of the photosensitizer can be suppressed significantly as compared to short-pulse (nanosecond) excitation. It was shown, that special care is to be taken when singlet oxygen lifetime is determined by fitting phosphorescence decay curves with the corresponding theoretical timedependence. Reliable data can be obtained, when the photosensitizer's triplet state lifetime is significantly smaller than that of singlet oxygen. The experiments carried out at high oxygen concentrations (where τΔ c τT) yielded O2(1Δg) lifetime of τΔ ¼5.570.3 μs, which is in good agreement with the result (τΔ ¼5.6 μs) of Staicu and co-workers [9]. To the best of our knowledge, the rate constant for 3Hyp depopulation in reactions with ground state molecular oxygen in DMSO (measured to
[1] P.R. Ogilby, Chem. Soc. Rev. 39 (2010) 3181. [2] D.E.J.G.J. Dolmans, D. Fukumura, R.K. Jain, Nat. Rev. Cancer 3 (2003) 380. [3] C.A. Robertson, D. Hawkins Evans, H. Abrahamse, J. Photochem. Photobiol. B 96 (2009) 1. [4] C. Schweitzer, R. Schmidt, Chem. Rev. 103 (2003) 1685. [5] F. Wilkinson, W.P. Helman, A.B. Ross, J. Phys. Chem. Ref. Data 24 (1995) 663. [6] A.C. Williams, B.W. Barry, Adv. Drug Deliv. Rev. 56 (2004) 603. [7] R. Notman, W.K. den Otter, M.G. Noro, W.J. Briels, J. Anwar, Biophys. J. 93 (2007) 2056. [8] M. Korínek, R. Dědic, A. Molnár, A. Svoboda, J. Hála, J. Mol. Struct. 744–747 (2005) 727. [9] A. Staicu, M. Pascu, M.L. Boni, M. Pascu, J. Mol. Struct. 1044 (2013) 188. [10] R. Nilsson, D.R. Kearns, J. Phys. Chem. 78 (1974) 1681. [11] H.J. Guiraud, C.S. Foote, J. Am. Chem. Soc. 98 (1976) 1984. [12] A.A. Krasnovsky Jr., J. Photochem. Photobiol. A 196 (2008) 210. [13] F. Shafii, R. Schmidt, J. Phys. Chem. A 105 (2001) 1805. [14] J. Schlothauer, S. Hackbarth, B. Röder, Laser Phys. Lett. 6 (2009) 216. [15] D. Petrovajova, D. Jancura, P. Miskovsky, D. Chorvat Jr, A. Chorvatova, X. Ragas, M. Garcia-Diaz, S. Nonell, Z. Nadova, Laser Phys. Lett. 10 (2013) 075609. [16] P. Gbur, R. Dědic, D. Jancura, P. Miškovský, J. Hála, J. Lumin. 128 (2008) 765. [17] X. Ragas, A. Gallardo, Y. Zhang, W. Massad, C.D. Geddes, S. Nonell, J. Phys. Chem. C 115 (2011) 16275. [18] S. Lee, L. Zhu, A.M. Minhaj, M.F. Hinds, D.H. Vu, D.I. Rosen, S.J. Davis, J. Biomed. Opt. 13 (2008) 034010. [19] S. Lee, D.H. Vu, M.F. Hinds, S.J. Davis, A. Liang, T. Hasan, J. Biomed. Opt. 13 (2008) 064035. [20] J. Varchola, V. Huntosova, D. Jancura, G. Wagnières, P. Miskovsky, G. Bánó, Photochem. Photobiol. Sci. 13 (2014) 1781. [21] H. Falk, Angew. Chem. Int. Ed. 38 (1999) 3116. [22] P. Agostinis, A. Vantieghem, W. Merlevede, P.A. De Witte, Int. J. Biochem. Cell Biol. 34 (2002) 221. [23] P. Miskovsky, Curr. Drug Targets 3 (2002) 55. [24] T. Kiesslich, B. Krammer, K. Plaetzer, Curr. Med. Chem. 13 (2006) 2189. [25] C.L.L. Saw, M. Olivo, K.C. Soo, P.W.S. Heng, Cancer Lett. 241 (2006) 23. [26] R.C. Guedes, L.A. Eriksson, J. Photochem. Photobiol. A 172 (2005) 293. [27] M. Roslaniec, H. Weitman, D. Freeman, Y. Mazur, B. Ehrenberg, J. Photochem. Photobiol. B 57 (2000) 149. [28] B. Ehrenberg, J.L. Anderson, C.S. Foote, Photochem. Photobiol. 68 (1998) 135. [29] S. Dumas, J.C. Leprêtre, A. Lepellec, A. Darmanyan, P. Jardon, J. Photochem. Photobiol. B 163 (2004) 297. [30] V. Huntosova, Z. Nadova, L. Dzurova, V. Jakusova, F. Sureau, P. Miskovsky, Photochem. Photobiol. Sci. 11 (2012) 1428. [31] L. Dzurová, D. Petrovajova, Z. Nadova, V. Huntosova, P. Miskovsky, K. Stroffekova, Photodiagn. Photodyn. 11 (2014) 213. [32] B. Kuchárová, J. Mikeš, R. Jendželovský, J. Vargová, L. Mikešová, Z. Jendželovská, J. Kovaľ, P. Fedoročko, Photodiagn. Photodyn. 12 (2015) 490. [33] M.P. Paz-Cristobal, D. Royo, A. Rezusta, E. Andrés-Ciriano, M.C. Alejandre, J. F. Meis, M.J. Revillo, C. Aspiroz, S. Nonell, Y. Gilaberte, Mycoses 57 (2014) 35. [34] G. Bánó, J. Staničová, D. Jancura, J. Marek, M. Bánó, J. Uličný, A. Strejčková, P. Miškovský, J. Phys. Chem. B 115 (2011) 2417. [35] P. Gbur, R. Dedic, D. Chorvat Jr, P. Miskovsky, J. Hala, D. Jancura, Photochem. Photobiol. 85 (2009) 816. [36] R.L. Arudi, A.O. Allen, B.H.J. Bielski, FEBS Lett. 135 (1981) 265. [37] A.E. Alegria, C.M. Krishna, R.K. Elespuru, P. Riesz, Photochem. Photobiol. 49 (1989) 257. [38] S. Hatz, L. Poulsen, P.R. Ogilby, Photochem. Photobiol. 84 (2008) 1284. [39] M.K. Kuimova, G. Yahioglu, P.R. Ogilby, J. Am. Chem. Soc. 131 (2009) 332. [40] E. Afshari, R. Schmidt, Chem. Phys. Lett. 184 (1991) 128. [41] L.P.F. Aggarwal, M.S. Baptista, L.E. Borissevitch, J. Photochem. Photobiol. A 186 (2007) 187. [42] R.L. Jensen, J. Arnbjerg, P.R. Ogilby, J. Am. Chem. Soc. 132 (2010) 8098. [43] A.P. Darmanyan, W.S. Jenks, D. Eloy, P. Jardon, J. Phys. Chem. B 103 (1999) 3323.
Contents lists available at ScienceDirect
Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin
Full Length Article
Singlet oxygen produced by quasi-continuous photo-excitation of hypericin in dimethyl-sulfoxide J. Varchola a, K. Želonková a, D. Chorvat Jr b, D. Jancura a,c, P. Miskovsky a,b,c, G. Bánó a,c,n a
Department of Biophysics, Faculty of Science, P. J. Šafárik University, Jesenná 5, 041 54 Košice, Slovak Republic International Laser Centre, Ilkovicova 3, 841 05 Bratislava, Slovak Republic c Center for Interdisciplinary Biosciences, Faculty of Science, P. J. Šafárik University, Jesenná 5, 041 54 Košice, Slovak Republic b
art ic l e i nf o
a b s t r a c t
Article history: Received 24 November 2015 Received in revised form 8 April 2016 Accepted 11 April 2016 Available online 14 April 2016
Singlet oxygen (O2(1Δg)) production by photo-excited hypericin (Hyp) dissolved in dimethyl-sulfoxide (DMSO) was studied by means of time-resolved phosphorescence measurements. In order to minimize photo-bleaching, the samples were excited in quasi-continuous mode using long-pulse (35 μs) laser excitation. The measured lifetime of singlet oxygen is τΔ ¼5.5 70.3 μs. This result helps to resolve the discrepancy existing in the literature concerning singlet oxygen lifetime in DMSO. The obtained quantum yield of singlet oxygen photosensitized by Hyp in air-saturated DMSO is ΦΔ ¼ 0.47 0.03. The rate constant for Hyp triplet state depopulation in reaction with ground state molecular oxygen is measured to be kq ¼ 1.67 0.3 109 M 1 s 1. & 2016 Published by Elsevier B.V.
Keywords: Singlet oxygen Hypericin Photosensitizer Dimethyl-sulfoxide
1. Introduction Singlet oxygen is a highly reactive molecule which causes oxidation of many biologically important compounds [1]. From medicinal point of view, singlet oxygen has been studied intensively because of its essential role in photodynamic therapy (PDT) of cancer. During PDT, the oxidative stress induced by reactive oxygen species (in most cases by singlet oxygen in the O2(1Δg) state) initiates processes leading to cell death [2,3]. Singlet oxygen is mostly produced by energy transfer between light-activated (triplet state) photosensitizer molecules and ground state (triplet, O2(3Σg )) oxygen molecules [1]. The rate of singlet oxygen deactivation shows significant variation depending on the environment [4]. Singlet oxygen lifetime values ranging from a few microseconds (e.g. 3-4 μs in water) to milliseconds (e.g. 59 ms in airsaturated CCl4) were reported for different pure solvents [4,5]. Dimethyl-sulfoxide (DMSO) is clinically used as a penetration enhancer to carry drugs into tissues [6,7] and is known as an efficient solvent for many hydrophobic compounds. Due to these reasons, the information on singlet oxygen properties in DMSO is highly important. In spite of the large number of singlet oxygen lifetime data published during the last four decades, relatively few papers deal with the deactivation rate of singlet oxygen in DMSO. n Corresponding author at: Center for Interdisciplinary Biosciences, Faculty of Science, P. J. Šafárik University, Jesenná 5, 041 54 Košice, Slovak Republic. E-mail address: [email protected] (G. Bánó).
http://dx.doi.org/10.1016/j.jlumin.2016.04.020 0022-2313/& 2016 Published by Elsevier B.V.
Moreover, the available data scatter significantly. Lifetime values of 1.8 μs [8], 5.6 μs [9] and 30 μs [10] were published by other authors, while a lifetime value of 19 μs can be calculated from the quenching rate reported in [11]. One of the objectives of the present work is to provide additional experimental data that help to clarify the contradiction between the previous reports. Singlet oxygen in the O2(1Δg) state can be detected through its phosphorescence around 1270 nm. The lifetime of singlet oxygen is readily determined by time-resolved phosphorescence decay measurements following pulsed photo-activation of photosensitizer molecules [12]. Most of the previous experiments were carried out with short-pulse laser excitation using laser pulses in the nanosecond or sub-nanosecond range (see e.g. [1,12–17]). An alternative excitation scheme using longer (microsecond) laser pulses was applied by Lee and co-workers [18,19]. In our previous work, the duration of the laser pulses was set long enough so that steady-state concentrations of the photosensitizer's triplet state and O2(1Δg) were reached at the end of the laser pulse [20]. The major advantage of this so-called quasi-continuous excitation scheme was that the peak laser power was reduced significantly when compared to laser pulses of nanosecond duration, which resulted in a lower bleaching rate of the used photosensitizers. Moreover, steady-state conditions at the end of the laser pulse allowed for analytical solution of the differential equations describing the time-evolution of O2(1Δg) concentration after the excitation pulse [20].
18
J. Varchola et al. / Journal of Luminescence 177 (2016) 17–21
Photo-excited hypericin (Hyp) was used as a source of the singlet oxygen in this work. Hyp is a naturally occurring potent photosensitizer that is being studied extensively for its possible application in PDT [21–25]. Under light illumination Hyp causes anti-proliferative and cytotoxic effects in many tumor cell lines. Singlet oxygen is supposed to play the major role in the photoactivity of Hyp at aerobic conditions. O2(1Δg) generation via energy transfer from Hyp triplet state (3Hyp) to ground state oxygen molecules was theoretically predicted [26] as well as experimentally confirmed in several organic solvents and lipid environments [27–29]. The corresponding quantum yields for singlet oxygen formation by Hyp vary between 0.25 and 0.45 [27– 29]. Stock solutions of Hyp used in PDT-related experiments are often prepared in DMSO [30–33]. On the other hand, Hyp forms non-fluorescence aggregates in aqueous environment [21,34]. These aggregates are not able to produce singlet oxygen [15,35]. In the present work, besides the determination of singlet oxygen lifetime in DMSO, we report also on the rate of 3Hyp deactivation by oxygen in DMSO and the quantum yield of singlet oxygen production by Hyp in this solvent.
2. Experimental Hypericin, Rose Bengal, dimethyl-sulfoxide (4 99.9%), RuPhen (dichlorotris(1,10-phenantroline) ruthenium(II) hydrate) and protoporphyrin IX disodium salt (PpIX) were purchased from SigmaAldrich. The purity of DMSO was checked by measuring its freezing point before the experiments. The results (17.570.5 °C) indicated water content of about 0.5 wt%. According to the work of Staicu et al. [9], the effect of this water content on singlet oxygen lifetime in DMSO is below the error range of the present experiments. The experimental apparatus used for time-resolved phosphorescence measurements has been described in details previously [20]. Briefly, the excitation laser beam (532 nm, Cobolt Samba) passed vertically through a cuvette (1 1 4 cm) filled with 2 ml of the sample solution. The cuvette holder was thermostated at 22 °C. The sample was continuously flushed with a nitrogen/oxygen mixture (at 25 standard cubic centimeter per minute, sccm) and was stirred by a magnetic stirrer. An acousto-optic modulator (AOM, Isomet 1205C) was used to switch the excitation laser beam (on/off) creating 35 μs long laser pulses with a repetition rate of 8 kHz. The average laser power on the sample was 1.4 mW. The switching time of the AOM was approx. 400 ns. The emitted phosphorescence light was passing through a 1250–1350 nm bandpass filter and was detected with a Hamamatsu H10330A-75 photomultiplier tube (PMT) operated in photon counting mode. The time evolution of the photon rate was acquired with a multichannel scaler (Becker & Hickl, MSA-300). Typically, a signal of 3.106 pulses was collected at each experimental condition. The obtained time-dependences were corrected for the phosphorescence background of the optical components as measured in pure DMSO. In case of RuPhen measurements the sample was excited by a 476 nm (Coherent 90 C FreD) laser. The repetition rate of 20 μs long laser pulses was set to 2 kHz and the average laser power at the sample was 0.7 mW. The concentration of oxygen in the sample was calculated from the gas-phase oxygen pressure [36]. Rose Bengal (RB) was used to calibrate the experimental apparatus for quantum yield measurements of singlet oxygen production (see below). The calibration was carried out with a RB solution (in DMSO) having the same absorbance (0.022 per cm at 532 nm) as the used sample of Hyp. The quantum yield of singlet oxygen production by RB in DMSO
was estimated to be 0.76 (see also [37]) for air-saturated conditions. The lifetime data obtained with quasi-continuous excitation were compared with those measured by conventional short-pulse experiments. The apparatus was equipped with a Q-switched Nd: YAG laser (Spectra-Physics Quanta-Ray INDI-40-10) pumping an OPO (optical parametric oscillator, Spectra-Physics basiScan) that was tuned to 532 nm. The repetition rate of the 5–7 ns long laser pulses was 10 Hz producing an average beam power of 1.2 mW at the sample. The phosphorescence signal of 3000 pulses was collected at each experimental condition.
3. Data analysis The decay of the O2(1Δg) concentration [Δ] after long pulse (quasi-continuous) excitation is described by [20]: t t S τΔ ½Δ ¼ 0 ð1Þ τ e τΔ τ T e τT ; τΔ τT Δ where S0 is the rate of O2(1Δg) production at steady-state conditions, and τΔ and τT are the lifetimes of O2(1Δg) and the photosensitizer's triplet state, respectively. The O2(1Δg) phosphorescence intensity detected at 1270 nm is proportional to [Δ] through an equipment specific factor c and the O2(1Δg) phosphorescence emission rate constant kΔe: t τΔ τtΔ e ð2Þ τ e τ T e τT ; I 1270 meas ¼ ckΔ ½Δ ¼ F τΔ τT Δ where F¼ S0kΔec. The given time dependence differs from the kinetic curves used to analyze short-pulse (nanosecond) experiments [12], where (following the short excitation pulse) the phosphorescence signal first increases, then reaches its maximum and finally decays. By contrast, dependence (2) gives a monotonous decrease of the phosphorescence intensity with its time derivative being zero at t¼0 (right after the excitation pulse). Fig. 1 shows a typical experimental O2(1Δg) phosphorescence decay curve obtained with quasi-continuous excitation of Rose Bengal dissolved in DMSO using 35 μs excitation pulses. Rose Bengal was chosen as a photosensitizer here because (unlike Hyp) it has a low level of fluorescence at 1270 nm. Right after turning the laser on (at t ¼ 35 μs in Fig. 1) the fluorescence of RB appears instantaneously. The onset of the fluorescence is mostly limited by the switching time of the AOM. During the laser pulse the signal increases due to a gradually increasing contribution from singlet oxygen phosphorescence, which reaches steady-state intensity at the end of the pulse. When the laser is switched off (at t ¼0) the
Fig. 1. Typical time-dependence of the emission signal detected around 1270 nm during long-pulse excitation of Rose Bengal (not all points are shown). The red curve represents data fit according to (2). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
J. Varchola et al. / Journal of Luminescence 177 (2016) 17–21
19
Fig. 2. Testing the fitting procedure on calculated (model) phosphorescence decay data. Panels a) and b): Model decay curves (open circles) calculated according to (2) for two different τΔ, τT pairs as indicated. The colored lines in a) and b) represent the results of data fits using (2) with fixed values of τΔfixed. Panels c) and d): Pairs of τΔfixed and the corresponding τTfitted obtained by fitting the model curves of a) and b), respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
fluorescence disappears quickly and the phosphorescence of singlet oxygen decays according to (2) – see the red curve in Fig. 1. It can be shown, that the shape of (2) is identical to the decay phase of phosphorescence kinetic curves after short-pulse excitation [12]. It follows that the same general rules apply when fitting experimental data with (2). Namely, as emphasized in previous works [12,38,39], one has to be cautious about determining τΔ and τT by simply fitting the phosphorescence decay with (2). The related problem is twofold. First, exchanging τΔ and τT in (2) does not change the curve shape, which may lead to confusion between these two values. Due to this fact it is usually recommended to determine the photosensitizer's decay time τT from an independent experiment [38,39]. Second, when τΔ and τT are about the same (τΔ E τT) the decay curve (2) can be well approximated by different combinations of τΔ, τT pairs. This is demonstrated in Fig. 2 showing decay data calculated according to (2) for two different τΔ, τT pairs. The open circles in Fig. 2a belong to τΔ ¼5 and τT ¼4 (in arbitrary time units), while τΔ ¼ 5 and τT ¼1 were used to calculate the data points in Fig. 2b. The two data sets are normalized to unity at t¼0 and both are shown decreasing by 1.5 orders of magnitude. The lines plotted in Fig. 2a and b represent the results of data fits. During the fitting procedure τΔ was fixed (at different τΔfixed values ranging from 3 to 6.5) and only τT was varied. The combinations of τΔfixed and τTfitted obtained this way are shown in Fig. 2c and d as they belong to the fitted curves in Fig. 2a and b, respectively. It can be seen that for τΔ ¼5 and τT ¼4 (Fig. 2a) all the different data fits reproduce the original data points well. Only four curves are shown here (out of seven), the remaining ones being identical to them. Taking into account the usual experimental noise, it is concluded that the uncertainty of determining τΔ and τT by fitting experimental data with (2) can be high for τΔ E τT. By contrast, for τΔ ¼5 and τT ¼ 1 (Fig. 2b) the modeled decay curve is only reproduced well when τΔfixed and τTfitted stay close to their original value (fit #4 in Fig. 2b and d). When τΔ c τT the late phosphorescence decay is governed by τΔ,
which is the optimal condition for reliable singlet oxygen lifetimes measurements. Fitting experimental phosphorescence decay curves with (2) one obtains τΔ, τT and a value F that is proportional to the steadystate rate of O2(1Δg) production, S0. The fitted F values can be used to measure the quantum yield of O2(1Δg) production ΦΔ. For that the apparatus is to be calibrated with a photosensitizer of known quantum yield (for details see [20] and references cited therein). As mentioned in the Section 2 RB dissolved in DMSO was used for calibration in this work. The reciprocal value of τT can be written as: 1
τT
¼ k0 þkq ½O2 ;
ð3Þ
where k0 is the rate constant of 3Hyp quenching in absence of oxygen, kq represents the rate constant for 3Hyp depopulation in reactions with ground state molecular oxygen and [O2] denotes the molecular oxygen concentration in the sample. Based on (3) k0 and kq can be determined by measuring τT at different oxygen concentrations. Knowing k0 and kq the fraction of 3Hyp quenched in reaction with oxygen can be calculated as: 2 PO T ¼
kq ½O2 k0 þ kq ½O2
ð4Þ
Finally, the quantum yield of singlet oxygen production ΦΔ is 3 2 connected to P O T through the quantum yield of Hyp photo-excitation, ΦT, and the fraction of 3Hyp quenched by oxygen yielding T singlet oxygen, f Δ , as [20]:
ΦΔ ¼ ΦT P OT 2 f TΔ :
ð5Þ
Assuming ΦT and f Δ being independent of oxygen concentration, substitution of (4) into (5) gives the oxygen concentration dependence of ΦΔ. T
20
J. Varchola et al. / Journal of Luminescence 177 (2016) 17–21
4. Results and discussion The evolution of the near infrared ( 1270 nm) emission signal from the DMSO solution of Hyp (5.10 6 M) during and after the 35 ms long laser pulses was detected for gas-phase oxygen concentrations of 5%, 10%, 20% and 40% in the O2/N2 mixture (see Fig. 3a). The rapid decrease of the emission intensity immediately after the laser is switched-off is attributed to the decrease of Hyp fluorescence. Based on the fluorescence intensity, the extent of Hyp photo-bleaching was estimated to be less than 3% during the experiments, as it was determined from two consecutive measurements on each sample (not shown). Both the 3Hyp and O2(1Δg) lifetimes were determined by fitting singlet oxygen phosphorescence decays with Eq. (2). As shown in the Data analysis section 3, reliable lifetime values can only be obtained at conditions where τΔ and τT differ significantly. The lifetime of 3Hyp is sensitive to oxygen concentration, being smaller at higher oxygenation level. By contrast, deactivation of O2(1Δg) by ground state oxygen is negligible. It was shown previously that the effect of O2(1Δg) quenching by O2 is only of high importance in solvents with long O2(1Δg) lifetimes (typically in the millisecond range) [4,40]. It can be seen from Fig. 3a that at high oxygen concentration the late decay of O2(1Δg) phosphorescence became mono-exponential. Moreover, the rate of the exponential decay (the slope of the curves on logarithmic scale) did not change between 20% and 40% (gas-phase percentage) of oxygen. All this gave us an evidence of reaching conditions, where τΔ»τT was valid, and the course of the late decay was determined solely by the lifetime of O2(1Δg). Fitting the experimental decay curves at 20 and 40% of oxygen with Eq. (2), the lifetime of singlet oxygen in DMSO was determined to be τΔ ¼5.570.3 μs. The indicated experimental error is mostly due to the uncertainty introduced by the background phosphorescence of the
Fig. 3. a) Experimental phosphorescence decay curves detected around 1270 nm after quasi-continuous excitation of Hyp (5.10 6 M) in DMSO. Excitation: 532 nm, pulse length 35 μs, repetition rate 8 kHz, average laser power 1.4 mW. b) Time course of the phosphorescence signal after short-pulse excitation of Hyp (5.10 6 M) in DMSO. The red curve represents the fitting results (see [12]) with parameters τΔ ¼5.5 μs, τT ¼1.36 μs. Excitation: 532 nm, pulse length 5–7 ns, repetition rate: 10 Hz, average laser power: 1.2 mW. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
optical components that is to be subtracted from the measured signal. A set of control measurements was carried out to verify the obtained lifetime value. First, the Hyp concentration was lowered (to 2.10 6 M) to estimate the rate of O2(1Δg) deactivation by Hyp in the sample. It was shown that the effect of Hyp on τΔ was within the range of the experimental error. Second, the lifetime of singlet oxygen was measured at lower laser power (0.7 mW average) to exclude the effect of possible photoproducts on τΔ. Third, the measurements were repeated with two other photosensitizers, Rose Bengal and PpIX, dissolved in DMSO. Finally, the lifetime of singlet oxygen was evaluated from our short-pulse experiments using Hyp dissolved in DMSO (see Fig. 3b). All these measurements confirmed the given value of singlet oxygen lifetime in DMSO (τΔ ¼5.5 μs), which is in good agreement with the result (τΔ ¼ 5.6 μs) of Staicu and co-workers [9]. To establish credibility in the singlet oxygen lifetime data obtained with the quasi-continuous excitation scheme the technique was tested on water solutions, in which singlet oxygen has a similarly short lifetime as in DMSO. Singlet oxygen production by photo-excited Ruphen (1.10 5 M) was measured in 0.9% aqueous NaCl solution (see also Ref. [20]). There is an indication that water salinity does not affect the lifetime of singlet oxygen [41]. Indeed, as shown on Fig. 4, the singlet oxygen lifetime determined by the long-pulse technique at different temperatures and gas-phase oxygen concentrations (10%, 20% and 40%) is in a good agreement with the values published by Jensen and co-workers for pure water [42]. All this gave us confidence, that the quasi-continuous excitation scheme works fine in various solvents. The phosphorescence decays of singlet oxygen measured for low oxygen concentrations in DMSO (5% and 10%, see Fig. 3a) were fitted with Eq. (2) using the above determined τΔ value. Reciprocal values of 3Hyp lifetimes are shown in Fig. 5a as a function of oxygen concentration. The data were fitted by a linear [O2] dependence (3). The rate constant for 3Hyp depopulation in reaction with ground state molecular oxygen was determined to be kq ¼ (1.670.3) 109 M 1 s 1. This value can be compared to that reported for Hyp dissolved in acetonitrile (kq ¼1.4 109 M 1 s 1) [43]. The experimental error for determining the rate constant of 3Hyp quenching in absence of oxygen is relatively large; however, the fitted value k0 ¼2.104 s 1 is in good agreement with that obtained by Darmanyan and co-workers for Hyp dissolved in argon-flushed DMSO (k0 ¼ 1.4 104 s 1) [43]. The phosphorescence intensity of singlet oxygen at t¼0 (which is proportional to steady-state O2(1Δg) concentration) showed little variance with changing oxygen concentration (see Fig. 3a). The corresponding F values obtained by fitting the phosphorescence decay curves with Eq. (2) were used to evaluate the
Fig. 4. Comparison of singlet oxygen lifetime data measured in aqueous environment. Solid symbols: experimental results obtained with quasi-continuous excitation of Ruphen at different gas-phase oxygen concentrations (10%, 20% and 40%). Open squares: reference data taken from Ref. 42.
J. Varchola et al. / Journal of Luminescence 177 (2016) 17–21
21
be kq ¼ (1.670.3) 109 M 1 s 1) has not been published. The quantum yield of singlet oxygen production by Hyp in air-saturated DMSO was determined to be 0.470.03.
Acknowledgments This work was supported by the APVV-0242-11 grant of the Slovak Ministry of Education and the FP7 EU Projects: CELIM 316310 and LASERLAB-EUROPE 284464. This work was also supported by the projects SEPO-II (26220120039) and NanoBioSens (26220220107) of the Operation Program Research and Development funded by the European Regional Development Fund.
References
Fig. 5. a) Reciprocal lifetime of 3Hyp plotted against the oxygen concentration in DMSO fitted with Eq. (3). b) The quantum yield of O2(1Δg) production by Hyp in DMSO. The data points are fitted with the oxygen concentration dependence of (4) substituted to Eq. (5) using k0 and kq values determined by the linear fit of panel a.
quantum yield of O2(1Δg) photosensitized by Hyp in DMSO. The results are shown in Fig. 5b. The theoretical curve (solid line) shown in Fig. 5b was constructed by substituting the measured k0 T and kq to (4) and (5). The product of ΦT f Δ was assumed to be constant and was adjusted to reproduce the measured data points. It is apparent that the oxygen concentrations used in our experi2 ment fell into the saturation region of P O T (see (4)), which explains the weak oxygen dependence of the observed quantum yield data. The measured quantum yield of singlet oxygen production by Hyp in DMSO (at air saturated conditions – or 21% of gas-phase oxygen) is ΦΔ ¼0.4 70.03. The indicated error reflects the uncertainty of the reference quantum yield of singlet oxygen photosensitized by RB in DMSO [37]. The quantum yield obtained for Hyp in DMSO agrees well with that reported for Hyp dissolved in methanol (0.39) [27] and is higher than those measured for Hyp in ethanol (0.35) or acetonitrile (0.25) [29].
5. Conclusions The quasi-continuous excitation scheme was applied successfully for quantitative singlet oxygen measurements providing both excited state lifetime and quantum yield data. Singlet oxygen was produced by photo-excitation of Hyp in DMSO at different oxygen concentrations. The major advantage of working with long laser pulses is that bleaching of the photosensitizer can be suppressed significantly as compared to short-pulse (nanosecond) excitation. It was shown, that special care is to be taken when singlet oxygen lifetime is determined by fitting phosphorescence decay curves with the corresponding theoretical timedependence. Reliable data can be obtained, when the photosensitizer's triplet state lifetime is significantly smaller than that of singlet oxygen. The experiments carried out at high oxygen concentrations (where τΔ c τT) yielded O2(1Δg) lifetime of τΔ ¼5.570.3 μs, which is in good agreement with the result (τΔ ¼5.6 μs) of Staicu and co-workers [9]. To the best of our knowledge, the rate constant for 3Hyp depopulation in reactions with ground state molecular oxygen in DMSO (measured to
[1] P.R. Ogilby, Chem. Soc. Rev. 39 (2010) 3181. [2] D.E.J.G.J. Dolmans, D. Fukumura, R.K. Jain, Nat. Rev. Cancer 3 (2003) 380. [3] C.A. Robertson, D. Hawkins Evans, H. Abrahamse, J. Photochem. Photobiol. B 96 (2009) 1. [4] C. Schweitzer, R. Schmidt, Chem. Rev. 103 (2003) 1685. [5] F. Wilkinson, W.P. Helman, A.B. Ross, J. Phys. Chem. Ref. Data 24 (1995) 663. [6] A.C. Williams, B.W. Barry, Adv. Drug Deliv. Rev. 56 (2004) 603. [7] R. Notman, W.K. den Otter, M.G. Noro, W.J. Briels, J. Anwar, Biophys. J. 93 (2007) 2056. [8] M. Korínek, R. Dědic, A. Molnár, A. Svoboda, J. Hála, J. Mol. Struct. 744–747 (2005) 727. [9] A. Staicu, M. Pascu, M.L. Boni, M. Pascu, J. Mol. Struct. 1044 (2013) 188. [10] R. Nilsson, D.R. Kearns, J. Phys. Chem. 78 (1974) 1681. [11] H.J. Guiraud, C.S. Foote, J. Am. Chem. Soc. 98 (1976) 1984. [12] A.A. Krasnovsky Jr., J. Photochem. Photobiol. A 196 (2008) 210. [13] F. Shafii, R. Schmidt, J. Phys. Chem. A 105 (2001) 1805. [14] J. Schlothauer, S. Hackbarth, B. Röder, Laser Phys. Lett. 6 (2009) 216. [15] D. Petrovajova, D. Jancura, P. Miskovsky, D. Chorvat Jr, A. Chorvatova, X. Ragas, M. Garcia-Diaz, S. Nonell, Z. Nadova, Laser Phys. Lett. 10 (2013) 075609. [16] P. Gbur, R. Dědic, D. Jancura, P. Miškovský, J. Hála, J. Lumin. 128 (2008) 765. [17] X. Ragas, A. Gallardo, Y. Zhang, W. Massad, C.D. Geddes, S. Nonell, J. Phys. Chem. C 115 (2011) 16275. [18] S. Lee, L. Zhu, A.M. Minhaj, M.F. Hinds, D.H. Vu, D.I. Rosen, S.J. Davis, J. Biomed. Opt. 13 (2008) 034010. [19] S. Lee, D.H. Vu, M.F. Hinds, S.J. Davis, A. Liang, T. Hasan, J. Biomed. Opt. 13 (2008) 064035. [20] J. Varchola, V. Huntosova, D. Jancura, G. Wagnières, P. Miskovsky, G. Bánó, Photochem. Photobiol. Sci. 13 (2014) 1781. [21] H. Falk, Angew. Chem. Int. Ed. 38 (1999) 3116. [22] P. Agostinis, A. Vantieghem, W. Merlevede, P.A. De Witte, Int. J. Biochem. Cell Biol. 34 (2002) 221. [23] P. Miskovsky, Curr. Drug Targets 3 (2002) 55. [24] T. Kiesslich, B. Krammer, K. Plaetzer, Curr. Med. Chem. 13 (2006) 2189. [25] C.L.L. Saw, M. Olivo, K.C. Soo, P.W.S. Heng, Cancer Lett. 241 (2006) 23. [26] R.C. Guedes, L.A. Eriksson, J. Photochem. Photobiol. A 172 (2005) 293. [27] M. Roslaniec, H. Weitman, D. Freeman, Y. Mazur, B. Ehrenberg, J. Photochem. Photobiol. B 57 (2000) 149. [28] B. Ehrenberg, J.L. Anderson, C.S. Foote, Photochem. Photobiol. 68 (1998) 135. [29] S. Dumas, J.C. Leprêtre, A. Lepellec, A. Darmanyan, P. Jardon, J. Photochem. Photobiol. B 163 (2004) 297. [30] V. Huntosova, Z. Nadova, L. Dzurova, V. Jakusova, F. Sureau, P. Miskovsky, Photochem. Photobiol. Sci. 11 (2012) 1428. [31] L. Dzurová, D. Petrovajova, Z. Nadova, V. Huntosova, P. Miskovsky, K. Stroffekova, Photodiagn. Photodyn. 11 (2014) 213. [32] B. Kuchárová, J. Mikeš, R. Jendželovský, J. Vargová, L. Mikešová, Z. Jendželovská, J. Kovaľ, P. Fedoročko, Photodiagn. Photodyn. 12 (2015) 490. [33] M.P. Paz-Cristobal, D. Royo, A. Rezusta, E. Andrés-Ciriano, M.C. Alejandre, J. F. Meis, M.J. Revillo, C. Aspiroz, S. Nonell, Y. Gilaberte, Mycoses 57 (2014) 35. [34] G. Bánó, J. Staničová, D. Jancura, J. Marek, M. Bánó, J. Uličný, A. Strejčková, P. Miškovský, J. Phys. Chem. B 115 (2011) 2417. [35] P. Gbur, R. Dedic, D. Chorvat Jr, P. Miskovsky, J. Hala, D. Jancura, Photochem. Photobiol. 85 (2009) 816. [36] R.L. Arudi, A.O. Allen, B.H.J. Bielski, FEBS Lett. 135 (1981) 265. [37] A.E. Alegria, C.M. Krishna, R.K. Elespuru, P. Riesz, Photochem. Photobiol. 49 (1989) 257. [38] S. Hatz, L. Poulsen, P.R. Ogilby, Photochem. Photobiol. 84 (2008) 1284. [39] M.K. Kuimova, G. Yahioglu, P.R. Ogilby, J. Am. Chem. Soc. 131 (2009) 332. [40] E. Afshari, R. Schmidt, Chem. Phys. Lett. 184 (1991) 128. [41] L.P.F. Aggarwal, M.S. Baptista, L.E. Borissevitch, J. Photochem. Photobiol. A 186 (2007) 187. [42] R.L. Jensen, J. Arnbjerg, P.R. Ogilby, J. Am. Chem. Soc. 132 (2010) 8098. [43] A.P. Darmanyan, W.S. Jenks, D. Eloy, P. Jardon, J. Phys. Chem. B 103 (1999) 3323.