- Email: [email protected]
Effect of temperature on the photobehavior of Rose Bengal associated with dipalmitoylphosphatidyl choline liposomes
Journal of Luminescence 131 (2011) 2468–2472
Contents lists available at ScienceDirect
Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin
Effect of temperature on the photobehavior of Rose Bengal associated with dipalmitoylphosphatidyl choline liposomes Estefanı´a Hugo a, Elsa Abuin a, Eduardo Lissi a,n, Emilio Alarco´n b, Ana M. Edwards b a b
Facultad de Quı´mica y Biologı´a, Universidad de Santiago de Chile, USACH, Casilla 40-Correo 33. Santiago, Chile ´lica de Chile, PUC, Santiago, Chile Facultad de Quı´mica, Pontificia Universidad Cato
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 August 2010 Received in revised form 8 June 2011 Accepted 13 June 2011 Available online 20 June 2011
The association and photobehavior of Rose Bengal (RB) in the presence of dipalmitoylphosphatidyl choline (DPPC) small unilamellar liposomes is determined by the temperature. At temperatures above the main phase transition of the bilayer, the incorporation of the dye is ca. 2.5 times more efficient than that taking place when the bilayer is in the gel state. In both temperature ranges, adsorption isotherms show a noticeable anti-cooperativity that can be related to electrostatic repulsion between bound molecules. The photophysics and the photochemistry of the bound dye molecules also depend on the bilayer status. In particular, in the liquid crystalline state the surrounding of the dye is more polar and production of singlet oxygen is less efficient (F 0.1). This reduced singlet oxygen production is partially due to a low triplet yield (FT ¼0.35) and triplet self-quenching due to a high local RB concentration. In spite of these, tryptophan is efficiently photobleached when RB is associated to liposomes in the liquid crystalline state, probably due to a Type I mechanism favored by its high local concentration in the sensitized surroundings. & 2011 Elsevier B.V. All rights reserved.
Keywords: Rose Bengal Dipalmitoylphosphatidyl choline Photosensitization Singlet oxygen Liposomes Dyes
1. Introduction Dyes are extensively employed in photodynamic therapy [1–3]. The efficiency of this procedure, aimed to the selective killing of undesired cells, relies on the localization of the photosensitizer and its capacity to generate singlet oxygen, a reactive oxygen species considered as the main damaging agent [2]. It is relevant then to evaluate how the dye is distributed between aqueous environments and relevant biological targets, such as proteins, DNA, and membranes. Rose Bengal (RB; Scheme 1) is frequently employed as a singlet oxygen source, mostly due to the high quantum yield of the process (ca. 0.76 in water) [4]. Its photophysics in homogeneous solvents and bound to proteins has been extensively studied [5]. In particular, several studies have been aimed to the evaluation of its binding to albumins and the photochemistry and photophysics of the bound molecules [6–10]. On the other hand, the binding of RB to membrane-like structures has been less investigated considerably [11,12]. As a first step to get an insight on the factors determining the association of the dye to biological membranes, we report in this communication the extent of RB binding to dipalmitoylphosphatidyl choline (DPPC) small unilamellar liposomes (SUVs) over a temperature range that
n
Corresponding author. Tel.: þ56 2 7181132; fax: þ56 2 6812108. E-mail address: [email protected] (E. Lissi).
0022-2313/$ - see front matter & 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jlumin.2011.06.021
covers the gel and liquid crystalline states. Furthermore, the microenvironment of the adsorbed dye and its capacity to generate singlet oxygen are briefly discussed.
2. Experimental section 2.1. Chemicals Rose Bengal (RB), L-Tryptophan (Trp), dipalmitoyphosphatidylcholine (DPPC), deuterium oxide, glycerol, NaCl, NaHPO4, and Na2HPO4 (499%) were Sigma–Aldrich products. Chloroform, HPLC quality, was obtained from Merck. Water employed was purified through a Milli-Q purification system. All the experiments were carried out in 100 mM phosphate buffer saline (PBS) pH 7.4 or in deuterated buffer, pD 7.4. 2.2. Small unilamellar liposomes (SUVs) preparation and characterization SUVs were prepared by sonication of a suspension of multilamellar vesicles, according to the standard procedure [13]. Multilamellar liposomes were prepared by solvent (chloroform) evaporation of a DPPC solution, followed by re-suspension in PBS. The multilamellar-formed liposomes were sonicated (sonicator Mixomix 3000) employing 10 cycles (1 min on, 3 min off). After
E. Hugo et al. / Journal of Luminescence 131 (2011) 2468–2472
2469
Excitation was performed employing a Surelite II, Nd-YAG laser (532 nm, ca. 10 ns, 10 mJ/pulse). The primary data were acquired and processed with a customized Luzchem Research LFP-112 system, and the singlet oxygen quantum yields were calculated according to the equation:(2)
FRB-Liposome ¼ FRB-D2O
ðIntensity at1270nmRB-Liposome =A532nm-Liposome Þ ðIntensity at1270 nmRB-D2O =A532nm-Water Þ
ð2Þ The intensity at 1270nm corresponds to that obtained immediately after the laser pulse. Singlet oxygen quantum yield of Rose Bengal in buffer (pH¼0.8) was employed as Ref. [18].
Scheme 1. Chemical structure of Rose Bengal (RB).
sonication, samples were centrifuged to eliminate titanium particles coming from the tip of the sonicator. Lipid concentration in the SUVs suspensions was determined by titration with ammonium ferrothiocyanate, according to the procedure described by Marshall [14]. Finally, SUVs sizes were determined by dynamic light scattering in a Malvern nano-sizer S90 equipment. The mean diameter of the number-averaged distribution was 30 nm, with a mean width of 78 nm. 2.3. RB binding to SUVs Addition of SUVs to a RB solution in PBS produces a red shift of the maximum of the visible absorption spectrum, initially located at 545 nm. The absorbance of the sample at 567 nm, measured as a function of the lipid concentration, was employed to quantify the amount of dye associated to the liposomal pseudophase [6–7,11,15]. The visible spectrum at high DPPC concentrations, where most of the dye is bound to the liposomes, was considered to estimate the microenvironment of bound RB molecules [7–8,16]. 2.4. Tryptophan consumption photosensitized by RB molecules
3. Results and discussion Addition of DPPC SUVs to an RB solution produces noticeable changes in its absorption spectra. Typical results obtained when, at a fixed temperature, aliquots of a concentrated SUVs suspension are added to a RB solution in buffer (pH¼7.4) are shown in Fig. 1. These changes indicate adsorption of the dye by the liposomes [8,19]. This type of data allows an evaluation of the dye distribution (from the change in absorbance at a given wavelength with DPPC concentration). This procedure has been previously employed to evaluate the incorporation of RB to macromolecules [6,7,20], micelles [21–24], and liposomes [11]. Furthermore, since the absorption of RB is determined by the characteristics of the solvent [7,8,16], the microenvironment of the bound dye can be estimated from the wavelength of maxima absorbance when most of the dye is bound to the dispersed microphase [7,8,25,26]. 3.1. Adsorption isotherms
The consumption of Trp sensitized by RB was evaluated in the absence and in the presence of SUVs. Solutions comprising RB (6 mM) and Trp (300 mM) were irradiated with visible light from a W-filament lamp at temperatures below (25 1C) and above (52 1C) the main phase transition for DPPC–SUVs [17]. Trp consumption was followed by the decrease of its fluorescence intensity (excitation 290 nm, and emission 330 nm) as a function of irradiation time. Experiments were performed in the absence and presence of SUVs (0.2 mM DPPC).
Fig. 2 shows the changes in the RB absorbance at 567 nm as a function of DPPC concentration at temperatures below and above DPPC SUVs main phase transition (42 1C) [17]. The data given in Fig. 2 show that, at both temperatures, the absorbance tends to reach a plateau, indicative of total association of the dye to DPPC SUVs [11]. In order to obtain the value of the absorbance of the bound species (AN), the data of Fig. 2 were fitted to a sigmoidal function and extrapolated to infinite lipid concentration. This allows a representation of the adsorption isotherm as [RB bound]/[lipid] vs. [RBfree]. The free and bound concentrations were obtained considering that
2.5. Photophysics of liposome-bound RB molecules
½RBbound ¼ ½RB2½RBfree
ð3Þ
Transient Rose Bengal triplet absorption decay at 620 nm was recorded in an m-LFP 111 laser-flash photolysis system (Luzchem Inc., Ottawa, Canada), employing the second harmonic from a Surelite II, Nd-YAG laser (532 nm, ca. 10 ns, 10 mJ/pulse) as excitation source. Less than 10% of photodegradation was observed in all the experiments. The triplet quantum yield was evaluated employing the following Equation: (1)
FRB-Liposome ¼ FRB-Water
ðDODRB-Liposome =A532nm-Liposome Þ ðDODRB-Water =A532nm-Water Þ
ð1Þ
where DOD values correspond to changes in the optical density measured immediately after the excitation laser pulse. These values were obtained by extrapolating to zero time the transient absorption, fitted to a monoexponential decay. Rose Bengal triplet quantum yield in buffer (0.9) was taken as Ref. [18]. Singlet oxygen quantum yields were evaluated in deuterated buffer solution by recording its phosphorescence emission decay at 1270 nm with a Hamamatsu NIR detector (peltier cooled at 62.8 1C operating at 900 V, coupled to a grating monochromator).
Fig. 1. Absorption spectra of RB (6 mM) in PBS buffer (pH 7.4) as a function of the added DPPC concentration (up to 0.47 mM). Data obtained at 53 1C. The arrow indicates the increasing DPPC concentration.
2470
E. Hugo et al. / Journal of Luminescence 131 (2011) 2468–2472
40000 35000
0.6
[DPPC]-1 (M-1)
Absorbance (567 nm)
0.7
0.5 0.4
30000 25000 20000 15000
0.3
10000
0.2
20 0.1 0.0
0.1
0.2 0.3 [DPPC] (mM)
0.4
0.5
25
30
35 40 45 50 Temperature (°C)
55
60
Fig. 4. Reciprocal of the DPPC concentration required to bind half of the RB present in the solution as a function of temperature. Analytical RB concentration: 6 mM.
Fig. 2. Changes on the 6 mM RB absorbance at 567 nm, as a function of DPPC concentration at 25 1C (K) or 53 1C (m). The lines shown correspond to the best fitting of the data to a sigmoidal function. Table 1 Values of the lipid concentrations (mM) needed to bind 50% of the Rose Bengal at different dye concentrations. Data obtained at 49 1C.
[RBBound]/[DPPC]
0.16 0.12 0.08
Rose Bengal (mM)
Lipid for 50% binding
na
1.88 6.8 14.9
24 38 55
0.04 0.09 0.13
a
0.04 0.00
0
2
4
6
8
10
12
14
[RBfree] (µM) Fig. 3. Adsorption isotherm of RB on DPPC SUVs. Data obtained at 53 1C in three independent experiments.
n¼ [RB]bound/[DPPC].
The anti-cooperativity evidenced in the data given in Fig. 3 implies that the lipid concentration needed to bind 50% of the analytical RB concentration increases when the dye concentration increases. This is evidenced by the data collected in Table 1. These data indicate that the surface potential (determined by n) at half-binding increases with the analytical RB concentration, leading to a decrease in the apparent binding constant.
and A ¼ A0 ½RBfree =½RB þA1 ½RBbound =½RB
ð4Þ
hence, ½RBfree ¼ ½RBðAA1 Þ=ðA0 A1 Þ
ð5Þ
where A is the absorbance of the sample and A0 is the absorbance in absence of liposomes. A typical plot of the number of RB molecules bound per lipid ([RB bound]/[lipid]) as a function of the free RB concentration is shown in Fig. 3. This plot shows a strong downward curvature, which implies that the data cannot be fitted to an ideal RB partitioning in accordance with the two pseudophases model. The most likely explanation of this decreased partitioning towards the liposomes at high free concentrations is based on the charge (–) of RB molecules at the working pH. Binding of RB to the liposomes would lead to a negative charge of the interface, leading to anti-cooperative binding of further charged substrate molecules. In order to characterize the binding efficiency of RB to the liposomes, we employed the lipid concentration needed to bind half of the RB molecules when the analytical concentration of the dye is 6 mM. These values are shown as a function of temperature in Fig. 4. The data of Fig. 4 clearly show that the incorporation of the dye is determined by the bilayer fluidity, being ca. 2.5 times more efficient when the liposomes are in the liquid crystalline state. This result, similar to those obtained for other solutes [27], is due to the more ordered and compact packing of the lipids at temperatures below the main phase transition.
3.2. Microenvironment of bound RB molecules The position of the absorption band of a dye is generally strongly dependent on the solvent polarity (solvatochromism) [28]. This property has been employed to characterize the micropolarity sensed by a dye molecule when it is incorporated to microphases such as liposomes [15]. Rose Bengal absorption spectrum depends on the solvent polarity [7,8,16,29] and hence can be considered as a reporter of the surrounding properties when associated to liposomes [8], proteins [7,20], or surfactants [21]. The shift in the position of the visible absorption band of RB in presence of liposomes, shown in Fig. 1, implies that RB, in spite of its hydrophilicity, interacts with the liposome and that the microenvironment of the bound dye is different than that sensed in bulk water. These changes can be employed to estimate the microenvironment sensed by the dye molecules bound to the liposomes [8,19]. The red shift observed in presence of the liposomes is indicative of a less polar environment. The micropolarity sensed by bound RB was quantified in terms of ET30 values [7]. The wavelengths of maxima absorbance, when most of the dye is bound to the liposomes, were 567 and 564 nm, at temperatures below and above the liposome main phase transition, respectively. On the other hand, in buffer, the absorption band position is not significantly modified by the temperature, remaining, giving the wavelength of maximum absorption at 550 nm over all the temperature range considered. Interpolation of the values obtained in liposomes in a plot of maximum absorbance vs. ET30 values for a series of solvents renders values of 37 and 42 kcal mol 1 for liposomes in the gel and liquid crystalline states, respectively. This would
E. Hugo et al. / Journal of Luminescence 131 (2011) 2468–2472
imply that the micropolarity sensed by the dye bound to fluid liposomes is similar to that provided by a solvent of moderate polarity, such as acetone, while for more rigid liposomes their polarity is similar to that of a solvent of relatively low polarity such as THF [7]. These results indicate that RB molecules are associated to the bilayer, almost independently of the bilayer fluidity. The difference observed between the two states could indicate a deeper water penetration in the liquid crystalline bilayer. It is interesting to note that in DMPC vesicles maximum wavelength absorption of 563.9 nm at 25 1C has ben reported [8]. This value is very close to that obtained here for DPPC vesicles in the liquid crystalline state. However, it has to be considered that the value in DMPC liposomes has been obtained at a much higher RB/lipid ratio and under conditions that do not warrant total association of the dye to the liposomes. Similarly, Lambert and Kochevar [19] have reported maxima absorptions for RB at 548 and 562 nm in buffer and in the presence of liposomes, respectively. 3.3. Reversibility of the binding process In order to test the reversibility of RB binding to the liposomes, samples containing RB and vesicles were diluted with buffer. The results obtained show that the addition of more buffer decreases the sample absorbance (due to its dilution) accompanied by a hypsochromic shift from 564 nm (indicative of RB partially associated to the liposomes) to 554 nm (close to the value measured in buffer). This shift implies that RB binding to the liposomes is a reversible process, justifying to the treatment of the data performed in the previous sections. 3.4. Tryptophan consumption photosensitized by bound RB molecules In order to assess the efficiency of bound RB molecules as sensitizers for the bleaching of hydrophilic substrates, we have measured the bleaching of Trp in the absence and in the presence of SUV’s, at temperatures below and above the main phase transition of the bilayer. The results are shown in Fig. 5. No differences on the Trp rate consumption in PBS were observed at 25 or 55 1C (data not shown). On the other hand, the data obtained in the presence of liposomes show remarkable differences between the two temperatures employed. In fact, at temperatures higher than that of the main phase transition, the rate of Trp fluorescence bleaching is faster than that in buffer and in liposome at 25 1C. On the other hand, slower bleaching rates are observed in presence of liposomes at lower temperatures. In particular, if it is
2471
considered that under the conditions employed (25 1C [DPPC]¼ 0.2 mM) approximately 50% of the RB is free, the results obtained would imply a very low capacity of bound RB molecules to generate free singlet oxygen able to react with Trp present in the external medium. This could be due, at least partially, to a reduced interaction between the bound dye and the oxygen dissolved in the external solvent and/or to self quenching of the triplets favored by the high local RB concentration. Given the rather high singlet oxygen quantum yield from RB in aqueous solution (ca. 0.76), the enhanced rate in the presence of liposomes in the liquid crystalline state cannot be ascribed to a more efficient 1O2 production. Since it has been reported that Trp could be incorporated to fluid liposomes [30], a likely explanation for the increased bleaching rate could be a contribution of a Type I mechanism to the observed bleaching. This Type I contribution could be promoted by the increased local concentration of Trp molecules in the surrounding of RB bound triplets. In agreement with these considerations, the remaining Trp fluorescence bleaching in presence of 10 mM azide at 53 1C is considerably larger (34%) in the presence of liposomes than in buffer solution (17%). The rather high Trp consumption rate remaining in presence of azide can be attributed to a significant contribution of Type I processes.
3.5. Photophysics of liposome-bound RB molecules In order to evaluate the changes in RB photobehavior elicited by dye association to DPPC SUVs, we have measured RB triplet quantum yield (FT) and singlet oxygen quantum yield (FD) in the presence and absence of liposomes. Also, triplet lifetimes (tT), singlet oxygen lifetime (tD) and the rate constant of RB triplet quenching by oxygen (kair) were also measured. The data obtained are shown in Table 2. The data obtained at 251 in the presence of SUVs are very close to those measured in PBS, a result partly due to incomplete association of RB to the liposomes under our experimental conditions. If it is considered that nearly 50% of the dye is bound, the singlet oxygen quantum yield of the bound RB is ca. 0.15. This low yield, together with the presence of a fast component in the triplet decay could be due to partial self-quenching of the dye, favored by the relatively high local RB concentration. Furthermore, it has to be considered that the presence of high local RB concentrations in the surroundings of the singlet oxygen generation locus could lead to a fast singlet oxygen decay not detectable under our experimental conditions, reducing the measurable singlet oxygen yield. Measured lifetimes are then those of the ‘‘free’’ singlet oxygen [30], a proposal compatible with the similar values estimated in absence of liposomes (Table 2).
1.0 F/F0 (360 nm)
0.9
Table 2 Rose Bengal selected photophysical parameters in PBS buffer and in DPPC suspensions.
0.8 0.7 0.6
Parameter
PBS at 25 1C
PBS at 55 1C
DPPC at 25 1C
DPPC at 55 1C
FT
0.9 1307 1
0.9 57 71
450 0.75 607 1
651 0.63 507 1
0.8 10 71 (22)a 140 7 2 (78)a 400c 0.41 60 71
0.35 9 7 1 (49)a 200 7 10 (51)a 1900c E 0.1 46 7 2
tT (ms)
0.5
kair (106 s 1)b
0.4 0
50
150 100 time (s)
200
250
Fig. 5. Photobleaching of tryptophan fluorescence sensitized by RB ([RB] ¼6 mM); [Tryptophan] ¼300 mM; (’) in PBS buffer; (K) in liposomal solution at 25 1C; (m) in liposomal solution (0,2 mM) at 53 1C.
FD
tD (ms) a
Weight of the component in the fitting of the bi-exponential decay. Quenching rate constant of triplet state of RB in solutions air equilibrated. c Obtained for the slow component of the triplet lifetime. b
2472
E. Hugo et al. / Journal of Luminescence 131 (2011) 2468–2472
Remarkable features of the data obtained in the presence of liposomes in the liquid crystalline state are as follows: i) A low triplet quantum yield ii) A bi-exponential decay of the excited triplets with a very long (200 ms) component. iii) A very low quantum yield of free singlet oxygen. iv) A very fast triplet quenching by oxygen. All these data point to the relevance of intra-liposome process following the excitation of liposome associated RB molecules and suggest the occurrence of self-quenching and/or intra-particle quenching of singlet oxygen by the dye [31]. These processes reduce the free singlet oxygen and hence minimize the rate of Type II processes, as mentioned in the previous section. The very fast quenching by oxygen can be related to its enhanced concentration in the bilayer and is similar to that reported in other liposomes in the liquid crystalline state [32].
Acknowledgments Thanks are given to Fondecyt (Grants nos.1070285 and AT-24080017), Dicyt (USACH), and VRAID (PUC) Grant (03/2008). E. Alarcon thanks to Becas Chile for a postdoctoral fellowship. The authors would like to acknowledge Dr. Juan (Tito) Scaiano for the facilities to perform time resolved measurements and to Michel Grenier for his help in these measurements. References [1] C. Giulivi, M. Sarcansky, E. Rosenfeld, A. Boveris, Photochem. Photobiol. 52 (1990) 745.
[2] K. Plactzer, B. Krammer, J. Berlanda, F. Berr, T. Kiesslich, Lasers Med. Sci. 24 (2009) 259. [3] D.E. Dolmans, D. Fukumura, R.K. Jain, Nat. Rev. Cancer 3 (2003) 380. [4] D.C. Neckers, J. Photochem. Photobiol. 47 (1989) 1. [5] P.F. Aramendia, K. Krieg, C. Nitsch, E. Bittersmann, S.E. Braslavsky, Photochem. Photobiol. 48 (1988) 187. [6] E. Abuin, A. Aspe´e, El Lissi, L. Leo´n, J. Chil. Chem. Soc. 52 (2007) 1196. [7] E. Alarco´n, A.M. Edwards, A. Aspe´e, C.D. Borsarelli, E.A. Lissi, Photochem. Photobiol. Sci. 8 (2009) 933. [8] P.M. Saikia, R.K. Dutta, J. Surf. Sci. Technol. 23 (2007) 195. [9] K. Naeem, S.S. Shah., S.W.H. Shah, G.M. Laghari, Chem. Mat. Sci. 131 (2000) 761. [10] A.M. Khan, S.S. Shah, J. Disper Sci. Technol. 29 (2008) 1401. [11] A. Posadaz, N.M. Correa, M.A. Biasutti, N.A. Garcı´a, Photochem. Photobiol. 86 (2010) 96. [12] H. De Lima Santos, C. Fortes Rigos, A.C. Tedesco, P. Ciancaglini, Biochim. Biophys. Acta 1715 (2005) 96. [13] M.A. Cubillos, E. Lissi, E. Abuin, Prog. React. Kinet. Mech. 26 (2001) 277. [14] J.C. Marshall, Anal. Biochem. 140 (1980) 10. [15] S.K. Gosh, S.C. Battacharya, Chem. Phys. Lipids 131 (2004) 151. [16] M.A. Rauf, J.P. Graham, S.B. Bukallah, M.A.S. A-Saedi, Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 72 (2009) 133. [17] A.S.L. Derycke, P.A.M. de Witte, Adv. Drug Delivery Rev. 56 (2004) 17. [18] S.L. Murov, I. Carmichael, G.L. Hug, Handbook of Photochemistry, Mercel Decker Inc., New York, 1993. [19] C.R. Lambert, I.E. Kochevar, Photochem. Photobiol. 66 (1997) 15. [20] Ch-Wen Wu, F.Y.H. Wu, Biochemistry 12 (1973) 4348. [21] R.C. Kapoor, M.K. Jian, V.N. Mishra, J. Lumin. 22 (1081) 435. [22] P.M. Saikia, R.K. Dutta, J. Surf. Sci. Technol. 23 (2007) 195. [23] K. Naeem, S.S. Shah., S.W.H. Shah, G.M. Laghari, Chem. Mater. Sci. 131 (2000) 761. [24] A.M. Khan, S.S. Shah, J. Disper. Sci. Technol. 29 (2008) 1401. [25] P. Bilksi, C. Chignelli, J. Photochem. Photobiol. A: Chem. 77 (1994) 49. [26] T. Handa, M. Nakagaki, K. Miyajima, J. Colloid Interface Sci. 137 (1990) 253. [27] E. Abuin, E. Lissi., D. Aravena, A. Zanocco, M. Macuer, J. Colloid Interface Sci. 122 (1988) 201. [28] C. Reichardt, Chem. Rev. 94 (1994) 2319. [29] A.Y. Nooraldeen, M. Palanichant, P.K. Palanisamy, Int. J. Nonlinear Sci. 7 (2009) 290. [30] V. Viti, S. Agostini, M. Ceccarini, M. Minetti, Physiol. Chem. Phys. Med. NMR 17 (1985) 413. [31] E.A. Lissi, M.V. Encinas, E. Lemp, M.A. Rubio, Chem. Rev. 93 (1993) 699. [32] E.B. Abuin, E.A. Lissi, Prog. React. Kinet. 16 (1991) 1.
Contents lists available at ScienceDirect
Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin
Effect of temperature on the photobehavior of Rose Bengal associated with dipalmitoylphosphatidyl choline liposomes Estefanı´a Hugo a, Elsa Abuin a, Eduardo Lissi a,n, Emilio Alarco´n b, Ana M. Edwards b a b
Facultad de Quı´mica y Biologı´a, Universidad de Santiago de Chile, USACH, Casilla 40-Correo 33. Santiago, Chile ´lica de Chile, PUC, Santiago, Chile Facultad de Quı´mica, Pontificia Universidad Cato
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 August 2010 Received in revised form 8 June 2011 Accepted 13 June 2011 Available online 20 June 2011
The association and photobehavior of Rose Bengal (RB) in the presence of dipalmitoylphosphatidyl choline (DPPC) small unilamellar liposomes is determined by the temperature. At temperatures above the main phase transition of the bilayer, the incorporation of the dye is ca. 2.5 times more efficient than that taking place when the bilayer is in the gel state. In both temperature ranges, adsorption isotherms show a noticeable anti-cooperativity that can be related to electrostatic repulsion between bound molecules. The photophysics and the photochemistry of the bound dye molecules also depend on the bilayer status. In particular, in the liquid crystalline state the surrounding of the dye is more polar and production of singlet oxygen is less efficient (F 0.1). This reduced singlet oxygen production is partially due to a low triplet yield (FT ¼0.35) and triplet self-quenching due to a high local RB concentration. In spite of these, tryptophan is efficiently photobleached when RB is associated to liposomes in the liquid crystalline state, probably due to a Type I mechanism favored by its high local concentration in the sensitized surroundings. & 2011 Elsevier B.V. All rights reserved.
Keywords: Rose Bengal Dipalmitoylphosphatidyl choline Photosensitization Singlet oxygen Liposomes Dyes
1. Introduction Dyes are extensively employed in photodynamic therapy [1–3]. The efficiency of this procedure, aimed to the selective killing of undesired cells, relies on the localization of the photosensitizer and its capacity to generate singlet oxygen, a reactive oxygen species considered as the main damaging agent [2]. It is relevant then to evaluate how the dye is distributed between aqueous environments and relevant biological targets, such as proteins, DNA, and membranes. Rose Bengal (RB; Scheme 1) is frequently employed as a singlet oxygen source, mostly due to the high quantum yield of the process (ca. 0.76 in water) [4]. Its photophysics in homogeneous solvents and bound to proteins has been extensively studied [5]. In particular, several studies have been aimed to the evaluation of its binding to albumins and the photochemistry and photophysics of the bound molecules [6–10]. On the other hand, the binding of RB to membrane-like structures has been less investigated considerably [11,12]. As a first step to get an insight on the factors determining the association of the dye to biological membranes, we report in this communication the extent of RB binding to dipalmitoylphosphatidyl choline (DPPC) small unilamellar liposomes (SUVs) over a temperature range that
n
Corresponding author. Tel.: þ56 2 7181132; fax: þ56 2 6812108. E-mail address: [email protected] (E. Lissi).
0022-2313/$ - see front matter & 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jlumin.2011.06.021
covers the gel and liquid crystalline states. Furthermore, the microenvironment of the adsorbed dye and its capacity to generate singlet oxygen are briefly discussed.
2. Experimental section 2.1. Chemicals Rose Bengal (RB), L-Tryptophan (Trp), dipalmitoyphosphatidylcholine (DPPC), deuterium oxide, glycerol, NaCl, NaHPO4, and Na2HPO4 (499%) were Sigma–Aldrich products. Chloroform, HPLC quality, was obtained from Merck. Water employed was purified through a Milli-Q purification system. All the experiments were carried out in 100 mM phosphate buffer saline (PBS) pH 7.4 or in deuterated buffer, pD 7.4. 2.2. Small unilamellar liposomes (SUVs) preparation and characterization SUVs were prepared by sonication of a suspension of multilamellar vesicles, according to the standard procedure [13]. Multilamellar liposomes were prepared by solvent (chloroform) evaporation of a DPPC solution, followed by re-suspension in PBS. The multilamellar-formed liposomes were sonicated (sonicator Mixomix 3000) employing 10 cycles (1 min on, 3 min off). After
E. Hugo et al. / Journal of Luminescence 131 (2011) 2468–2472
2469
Excitation was performed employing a Surelite II, Nd-YAG laser (532 nm, ca. 10 ns, 10 mJ/pulse). The primary data were acquired and processed with a customized Luzchem Research LFP-112 system, and the singlet oxygen quantum yields were calculated according to the equation:(2)
FRB-Liposome ¼ FRB-D2O
ðIntensity at1270nmRB-Liposome =A532nm-Liposome Þ ðIntensity at1270 nmRB-D2O =A532nm-Water Þ
ð2Þ The intensity at 1270nm corresponds to that obtained immediately after the laser pulse. Singlet oxygen quantum yield of Rose Bengal in buffer (pH¼0.8) was employed as Ref. [18].
Scheme 1. Chemical structure of Rose Bengal (RB).
sonication, samples were centrifuged to eliminate titanium particles coming from the tip of the sonicator. Lipid concentration in the SUVs suspensions was determined by titration with ammonium ferrothiocyanate, according to the procedure described by Marshall [14]. Finally, SUVs sizes were determined by dynamic light scattering in a Malvern nano-sizer S90 equipment. The mean diameter of the number-averaged distribution was 30 nm, with a mean width of 78 nm. 2.3. RB binding to SUVs Addition of SUVs to a RB solution in PBS produces a red shift of the maximum of the visible absorption spectrum, initially located at 545 nm. The absorbance of the sample at 567 nm, measured as a function of the lipid concentration, was employed to quantify the amount of dye associated to the liposomal pseudophase [6–7,11,15]. The visible spectrum at high DPPC concentrations, where most of the dye is bound to the liposomes, was considered to estimate the microenvironment of bound RB molecules [7–8,16]. 2.4. Tryptophan consumption photosensitized by RB molecules
3. Results and discussion Addition of DPPC SUVs to an RB solution produces noticeable changes in its absorption spectra. Typical results obtained when, at a fixed temperature, aliquots of a concentrated SUVs suspension are added to a RB solution in buffer (pH¼7.4) are shown in Fig. 1. These changes indicate adsorption of the dye by the liposomes [8,19]. This type of data allows an evaluation of the dye distribution (from the change in absorbance at a given wavelength with DPPC concentration). This procedure has been previously employed to evaluate the incorporation of RB to macromolecules [6,7,20], micelles [21–24], and liposomes [11]. Furthermore, since the absorption of RB is determined by the characteristics of the solvent [7,8,16], the microenvironment of the bound dye can be estimated from the wavelength of maxima absorbance when most of the dye is bound to the dispersed microphase [7,8,25,26]. 3.1. Adsorption isotherms
The consumption of Trp sensitized by RB was evaluated in the absence and in the presence of SUVs. Solutions comprising RB (6 mM) and Trp (300 mM) were irradiated with visible light from a W-filament lamp at temperatures below (25 1C) and above (52 1C) the main phase transition for DPPC–SUVs [17]. Trp consumption was followed by the decrease of its fluorescence intensity (excitation 290 nm, and emission 330 nm) as a function of irradiation time. Experiments were performed in the absence and presence of SUVs (0.2 mM DPPC).
Fig. 2 shows the changes in the RB absorbance at 567 nm as a function of DPPC concentration at temperatures below and above DPPC SUVs main phase transition (42 1C) [17]. The data given in Fig. 2 show that, at both temperatures, the absorbance tends to reach a plateau, indicative of total association of the dye to DPPC SUVs [11]. In order to obtain the value of the absorbance of the bound species (AN), the data of Fig. 2 were fitted to a sigmoidal function and extrapolated to infinite lipid concentration. This allows a representation of the adsorption isotherm as [RB bound]/[lipid] vs. [RBfree]. The free and bound concentrations were obtained considering that
2.5. Photophysics of liposome-bound RB molecules
½RBbound ¼ ½RB2½RBfree
ð3Þ
Transient Rose Bengal triplet absorption decay at 620 nm was recorded in an m-LFP 111 laser-flash photolysis system (Luzchem Inc., Ottawa, Canada), employing the second harmonic from a Surelite II, Nd-YAG laser (532 nm, ca. 10 ns, 10 mJ/pulse) as excitation source. Less than 10% of photodegradation was observed in all the experiments. The triplet quantum yield was evaluated employing the following Equation: (1)
FRB-Liposome ¼ FRB-Water
ðDODRB-Liposome =A532nm-Liposome Þ ðDODRB-Water =A532nm-Water Þ
ð1Þ
where DOD values correspond to changes in the optical density measured immediately after the excitation laser pulse. These values were obtained by extrapolating to zero time the transient absorption, fitted to a monoexponential decay. Rose Bengal triplet quantum yield in buffer (0.9) was taken as Ref. [18]. Singlet oxygen quantum yields were evaluated in deuterated buffer solution by recording its phosphorescence emission decay at 1270 nm with a Hamamatsu NIR detector (peltier cooled at 62.8 1C operating at 900 V, coupled to a grating monochromator).
Fig. 1. Absorption spectra of RB (6 mM) in PBS buffer (pH 7.4) as a function of the added DPPC concentration (up to 0.47 mM). Data obtained at 53 1C. The arrow indicates the increasing DPPC concentration.
2470
E. Hugo et al. / Journal of Luminescence 131 (2011) 2468–2472
40000 35000
0.6
[DPPC]-1 (M-1)
Absorbance (567 nm)
0.7
0.5 0.4
30000 25000 20000 15000
0.3
10000
0.2
20 0.1 0.0
0.1
0.2 0.3 [DPPC] (mM)
0.4
0.5
25
30
35 40 45 50 Temperature (°C)
55
60
Fig. 4. Reciprocal of the DPPC concentration required to bind half of the RB present in the solution as a function of temperature. Analytical RB concentration: 6 mM.
Fig. 2. Changes on the 6 mM RB absorbance at 567 nm, as a function of DPPC concentration at 25 1C (K) or 53 1C (m). The lines shown correspond to the best fitting of the data to a sigmoidal function. Table 1 Values of the lipid concentrations (mM) needed to bind 50% of the Rose Bengal at different dye concentrations. Data obtained at 49 1C.
[RBBound]/[DPPC]
0.16 0.12 0.08
Rose Bengal (mM)
Lipid for 50% binding
na
1.88 6.8 14.9
24 38 55
0.04 0.09 0.13
a
0.04 0.00
0
2
4
6
8
10
12
14
[RBfree] (µM) Fig. 3. Adsorption isotherm of RB on DPPC SUVs. Data obtained at 53 1C in three independent experiments.
n¼ [RB]bound/[DPPC].
The anti-cooperativity evidenced in the data given in Fig. 3 implies that the lipid concentration needed to bind 50% of the analytical RB concentration increases when the dye concentration increases. This is evidenced by the data collected in Table 1. These data indicate that the surface potential (determined by n) at half-binding increases with the analytical RB concentration, leading to a decrease in the apparent binding constant.
and A ¼ A0 ½RBfree =½RB þA1 ½RBbound =½RB
ð4Þ
hence, ½RBfree ¼ ½RBðAA1 Þ=ðA0 A1 Þ
ð5Þ
where A is the absorbance of the sample and A0 is the absorbance in absence of liposomes. A typical plot of the number of RB molecules bound per lipid ([RB bound]/[lipid]) as a function of the free RB concentration is shown in Fig. 3. This plot shows a strong downward curvature, which implies that the data cannot be fitted to an ideal RB partitioning in accordance with the two pseudophases model. The most likely explanation of this decreased partitioning towards the liposomes at high free concentrations is based on the charge (–) of RB molecules at the working pH. Binding of RB to the liposomes would lead to a negative charge of the interface, leading to anti-cooperative binding of further charged substrate molecules. In order to characterize the binding efficiency of RB to the liposomes, we employed the lipid concentration needed to bind half of the RB molecules when the analytical concentration of the dye is 6 mM. These values are shown as a function of temperature in Fig. 4. The data of Fig. 4 clearly show that the incorporation of the dye is determined by the bilayer fluidity, being ca. 2.5 times more efficient when the liposomes are in the liquid crystalline state. This result, similar to those obtained for other solutes [27], is due to the more ordered and compact packing of the lipids at temperatures below the main phase transition.
3.2. Microenvironment of bound RB molecules The position of the absorption band of a dye is generally strongly dependent on the solvent polarity (solvatochromism) [28]. This property has been employed to characterize the micropolarity sensed by a dye molecule when it is incorporated to microphases such as liposomes [15]. Rose Bengal absorption spectrum depends on the solvent polarity [7,8,16,29] and hence can be considered as a reporter of the surrounding properties when associated to liposomes [8], proteins [7,20], or surfactants [21]. The shift in the position of the visible absorption band of RB in presence of liposomes, shown in Fig. 1, implies that RB, in spite of its hydrophilicity, interacts with the liposome and that the microenvironment of the bound dye is different than that sensed in bulk water. These changes can be employed to estimate the microenvironment sensed by the dye molecules bound to the liposomes [8,19]. The red shift observed in presence of the liposomes is indicative of a less polar environment. The micropolarity sensed by bound RB was quantified in terms of ET30 values [7]. The wavelengths of maxima absorbance, when most of the dye is bound to the liposomes, were 567 and 564 nm, at temperatures below and above the liposome main phase transition, respectively. On the other hand, in buffer, the absorption band position is not significantly modified by the temperature, remaining, giving the wavelength of maximum absorption at 550 nm over all the temperature range considered. Interpolation of the values obtained in liposomes in a plot of maximum absorbance vs. ET30 values for a series of solvents renders values of 37 and 42 kcal mol 1 for liposomes in the gel and liquid crystalline states, respectively. This would
E. Hugo et al. / Journal of Luminescence 131 (2011) 2468–2472
imply that the micropolarity sensed by the dye bound to fluid liposomes is similar to that provided by a solvent of moderate polarity, such as acetone, while for more rigid liposomes their polarity is similar to that of a solvent of relatively low polarity such as THF [7]. These results indicate that RB molecules are associated to the bilayer, almost independently of the bilayer fluidity. The difference observed between the two states could indicate a deeper water penetration in the liquid crystalline bilayer. It is interesting to note that in DMPC vesicles maximum wavelength absorption of 563.9 nm at 25 1C has ben reported [8]. This value is very close to that obtained here for DPPC vesicles in the liquid crystalline state. However, it has to be considered that the value in DMPC liposomes has been obtained at a much higher RB/lipid ratio and under conditions that do not warrant total association of the dye to the liposomes. Similarly, Lambert and Kochevar [19] have reported maxima absorptions for RB at 548 and 562 nm in buffer and in the presence of liposomes, respectively. 3.3. Reversibility of the binding process In order to test the reversibility of RB binding to the liposomes, samples containing RB and vesicles were diluted with buffer. The results obtained show that the addition of more buffer decreases the sample absorbance (due to its dilution) accompanied by a hypsochromic shift from 564 nm (indicative of RB partially associated to the liposomes) to 554 nm (close to the value measured in buffer). This shift implies that RB binding to the liposomes is a reversible process, justifying to the treatment of the data performed in the previous sections. 3.4. Tryptophan consumption photosensitized by bound RB molecules In order to assess the efficiency of bound RB molecules as sensitizers for the bleaching of hydrophilic substrates, we have measured the bleaching of Trp in the absence and in the presence of SUV’s, at temperatures below and above the main phase transition of the bilayer. The results are shown in Fig. 5. No differences on the Trp rate consumption in PBS were observed at 25 or 55 1C (data not shown). On the other hand, the data obtained in the presence of liposomes show remarkable differences between the two temperatures employed. In fact, at temperatures higher than that of the main phase transition, the rate of Trp fluorescence bleaching is faster than that in buffer and in liposome at 25 1C. On the other hand, slower bleaching rates are observed in presence of liposomes at lower temperatures. In particular, if it is
2471
considered that under the conditions employed (25 1C [DPPC]¼ 0.2 mM) approximately 50% of the RB is free, the results obtained would imply a very low capacity of bound RB molecules to generate free singlet oxygen able to react with Trp present in the external medium. This could be due, at least partially, to a reduced interaction between the bound dye and the oxygen dissolved in the external solvent and/or to self quenching of the triplets favored by the high local RB concentration. Given the rather high singlet oxygen quantum yield from RB in aqueous solution (ca. 0.76), the enhanced rate in the presence of liposomes in the liquid crystalline state cannot be ascribed to a more efficient 1O2 production. Since it has been reported that Trp could be incorporated to fluid liposomes [30], a likely explanation for the increased bleaching rate could be a contribution of a Type I mechanism to the observed bleaching. This Type I contribution could be promoted by the increased local concentration of Trp molecules in the surrounding of RB bound triplets. In agreement with these considerations, the remaining Trp fluorescence bleaching in presence of 10 mM azide at 53 1C is considerably larger (34%) in the presence of liposomes than in buffer solution (17%). The rather high Trp consumption rate remaining in presence of azide can be attributed to a significant contribution of Type I processes.
3.5. Photophysics of liposome-bound RB molecules In order to evaluate the changes in RB photobehavior elicited by dye association to DPPC SUVs, we have measured RB triplet quantum yield (FT) and singlet oxygen quantum yield (FD) in the presence and absence of liposomes. Also, triplet lifetimes (tT), singlet oxygen lifetime (tD) and the rate constant of RB triplet quenching by oxygen (kair) were also measured. The data obtained are shown in Table 2. The data obtained at 251 in the presence of SUVs are very close to those measured in PBS, a result partly due to incomplete association of RB to the liposomes under our experimental conditions. If it is considered that nearly 50% of the dye is bound, the singlet oxygen quantum yield of the bound RB is ca. 0.15. This low yield, together with the presence of a fast component in the triplet decay could be due to partial self-quenching of the dye, favored by the relatively high local RB concentration. Furthermore, it has to be considered that the presence of high local RB concentrations in the surroundings of the singlet oxygen generation locus could lead to a fast singlet oxygen decay not detectable under our experimental conditions, reducing the measurable singlet oxygen yield. Measured lifetimes are then those of the ‘‘free’’ singlet oxygen [30], a proposal compatible with the similar values estimated in absence of liposomes (Table 2).
1.0 F/F0 (360 nm)
0.9
Table 2 Rose Bengal selected photophysical parameters in PBS buffer and in DPPC suspensions.
0.8 0.7 0.6
Parameter
PBS at 25 1C
PBS at 55 1C
DPPC at 25 1C
DPPC at 55 1C
FT
0.9 1307 1
0.9 57 71
450 0.75 607 1
651 0.63 507 1
0.8 10 71 (22)a 140 7 2 (78)a 400c 0.41 60 71
0.35 9 7 1 (49)a 200 7 10 (51)a 1900c E 0.1 46 7 2
tT (ms)
0.5
kair (106 s 1)b
0.4 0
50
150 100 time (s)
200
250
Fig. 5. Photobleaching of tryptophan fluorescence sensitized by RB ([RB] ¼6 mM); [Tryptophan] ¼300 mM; (’) in PBS buffer; (K) in liposomal solution at 25 1C; (m) in liposomal solution (0,2 mM) at 53 1C.
FD
tD (ms) a
Weight of the component in the fitting of the bi-exponential decay. Quenching rate constant of triplet state of RB in solutions air equilibrated. c Obtained for the slow component of the triplet lifetime. b
2472
E. Hugo et al. / Journal of Luminescence 131 (2011) 2468–2472
Remarkable features of the data obtained in the presence of liposomes in the liquid crystalline state are as follows: i) A low triplet quantum yield ii) A bi-exponential decay of the excited triplets with a very long (200 ms) component. iii) A very low quantum yield of free singlet oxygen. iv) A very fast triplet quenching by oxygen. All these data point to the relevance of intra-liposome process following the excitation of liposome associated RB molecules and suggest the occurrence of self-quenching and/or intra-particle quenching of singlet oxygen by the dye [31]. These processes reduce the free singlet oxygen and hence minimize the rate of Type II processes, as mentioned in the previous section. The very fast quenching by oxygen can be related to its enhanced concentration in the bilayer and is similar to that reported in other liposomes in the liquid crystalline state [32].
Acknowledgments Thanks are given to Fondecyt (Grants nos.1070285 and AT-24080017), Dicyt (USACH), and VRAID (PUC) Grant (03/2008). E. Alarcon thanks to Becas Chile for a postdoctoral fellowship. The authors would like to acknowledge Dr. Juan (Tito) Scaiano for the facilities to perform time resolved measurements and to Michel Grenier for his help in these measurements. References [1] C. Giulivi, M. Sarcansky, E. Rosenfeld, A. Boveris, Photochem. Photobiol. 52 (1990) 745.
[2] K. Plactzer, B. Krammer, J. Berlanda, F. Berr, T. Kiesslich, Lasers Med. Sci. 24 (2009) 259. [3] D.E. Dolmans, D. Fukumura, R.K. Jain, Nat. Rev. Cancer 3 (2003) 380. [4] D.C. Neckers, J. Photochem. Photobiol. 47 (1989) 1. [5] P.F. Aramendia, K. Krieg, C. Nitsch, E. Bittersmann, S.E. Braslavsky, Photochem. Photobiol. 48 (1988) 187. [6] E. Abuin, A. Aspe´e, El Lissi, L. Leo´n, J. Chil. Chem. Soc. 52 (2007) 1196. [7] E. Alarco´n, A.M. Edwards, A. Aspe´e, C.D. Borsarelli, E.A. Lissi, Photochem. Photobiol. Sci. 8 (2009) 933. [8] P.M. Saikia, R.K. Dutta, J. Surf. Sci. Technol. 23 (2007) 195. [9] K. Naeem, S.S. Shah., S.W.H. Shah, G.M. Laghari, Chem. Mat. Sci. 131 (2000) 761. [10] A.M. Khan, S.S. Shah, J. Disper Sci. Technol. 29 (2008) 1401. [11] A. Posadaz, N.M. Correa, M.A. Biasutti, N.A. Garcı´a, Photochem. Photobiol. 86 (2010) 96. [12] H. De Lima Santos, C. Fortes Rigos, A.C. Tedesco, P. Ciancaglini, Biochim. Biophys. Acta 1715 (2005) 96. [13] M.A. Cubillos, E. Lissi, E. Abuin, Prog. React. Kinet. Mech. 26 (2001) 277. [14] J.C. Marshall, Anal. Biochem. 140 (1980) 10. [15] S.K. Gosh, S.C. Battacharya, Chem. Phys. Lipids 131 (2004) 151. [16] M.A. Rauf, J.P. Graham, S.B. Bukallah, M.A.S. A-Saedi, Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 72 (2009) 133. [17] A.S.L. Derycke, P.A.M. de Witte, Adv. Drug Delivery Rev. 56 (2004) 17. [18] S.L. Murov, I. Carmichael, G.L. Hug, Handbook of Photochemistry, Mercel Decker Inc., New York, 1993. [19] C.R. Lambert, I.E. Kochevar, Photochem. Photobiol. 66 (1997) 15. [20] Ch-Wen Wu, F.Y.H. Wu, Biochemistry 12 (1973) 4348. [21] R.C. Kapoor, M.K. Jian, V.N. Mishra, J. Lumin. 22 (1081) 435. [22] P.M. Saikia, R.K. Dutta, J. Surf. Sci. Technol. 23 (2007) 195. [23] K. Naeem, S.S. Shah., S.W.H. Shah, G.M. Laghari, Chem. Mater. Sci. 131 (2000) 761. [24] A.M. Khan, S.S. Shah, J. Disper. Sci. Technol. 29 (2008) 1401. [25] P. Bilksi, C. Chignelli, J. Photochem. Photobiol. A: Chem. 77 (1994) 49. [26] T. Handa, M. Nakagaki, K. Miyajima, J. Colloid Interface Sci. 137 (1990) 253. [27] E. Abuin, E. Lissi., D. Aravena, A. Zanocco, M. Macuer, J. Colloid Interface Sci. 122 (1988) 201. [28] C. Reichardt, Chem. Rev. 94 (1994) 2319. [29] A.Y. Nooraldeen, M. Palanichant, P.K. Palanisamy, Int. J. Nonlinear Sci. 7 (2009) 290. [30] V. Viti, S. Agostini, M. Ceccarini, M. Minetti, Physiol. Chem. Phys. Med. NMR 17 (1985) 413. [31] E.A. Lissi, M.V. Encinas, E. Lemp, M.A. Rubio, Chem. Rev. 93 (1993) 699. [32] E.B. Abuin, E.A. Lissi, Prog. React. Kinet. 16 (1991) 1.