Jmmnal of Photochemistry
and Photobidogy,
B: Biology,
7 (1990)
43-56
43
CHLORIN e,-LIPOSOME INTERACTION. INVESTIGATION BY THE METHODS OF FLUORESCENCE SPECTROSCOPY AND INDUCTIVE RESONANCE ENERGY TRANSFER k
A. F’ROLOV, E. I. ZENKEVICH,
Institute of Physics, (U.S.S.R.) (Received
G. P. GURINOVICH
BSSR Academy
June 19, 1989; accepted
of Sciences,
and G. A. KOCHUBEYEV
L+eninsky Prospekt
70, 220602 Minsk
January 24, 1990)
K~~~~TTJ!.s. Chlorin e6, 1,6-diphenyl-1,3,5_hexatriene, liposomes, localization, partition, fluorescence quenching, inductive resonance energy transfer.
On the basis of spectral fluorescence and polarization measurements and results obtained on the luminescence quenching of the membrane fluorescent probe 1,6-diphenyl-1,3,5hexatriene (DPH) by incorporated chIorin e6 (chl es) molecules, it is shown that the interaction of the water-soluble pigment with smaller unilamellar lipid vesicles occurs by a mechanism of partition between the aqueous and lipid phases (partition coefficient Kp = 6.7 X 103) and provides rigid fixing of chl e6 monomers at the boundary between the polar and non-polar parts of the lipid membrane. In terms of inductive resonance electronic excitation energy transfer between DPH and chl e6 (I&,=36.2 A), we have analysed data on DPH fluorescence quenching under different conditions of chl eg localization in the lipid bilayer and have concluded that the incorporation of the pigment molecules into the vesicles from the aqueous phase occurs mainly into the external monolayer.
1. Introduction In recent years, the use of the photodynamic effect of porphyrin compounds for selective and effective destruction of human malignant tumours has become increasingly apparent [ 1, 21. At the present time, investigations in this field are mainly focused on the solution of two problems: optimization of the methods and regimes of phototherapeutic action using the well-known compounds haematoporphyrin derivative and Photofrin II [3, 41 and the search for new effective photosensitizers which exhibit optimal photophysical and physicochemical properties [5, 61. In this context, chlorin e6 (chl e6) and its derivatives are a promising new class of compounds for the photochemotherapy of tumours [ 7, 91.
loll-1344/90/$3.50
0 Elsevier
Sequoia/Printed
in The Netherlands
44
However, it should be noted that the choice and optimization of new tetrapyrrole-based photosensitizers from the results of investigations of their photophysical parameters in homogeneous solutions are not always consistent with their photodynamic activity in biological systems which are characterized by a significant heterogeneity of composition and properties. The elucidation of the detailed mechanisms of photodestruction of cellular systems requires a stepwise study of the interactions between sensitizers and components of biological membranes (proteins and lipids). The data available in the literature clearly demonstrate the usefulness of such an approach. For example, investigation of tryptophan fluorescence quenching by porphyrins [ 10, 1 l] and chl e, [ 121 in pro+.eins has permitted the determination of the nature of the photosensitizer-protein interaction, the localization of the conjugate macrocycles on the protein carrier and the correlation of the structural organization of pigment-protein complexes with the results obtained on the photosensitized destruction of various amino acid residues in proteins [ 131. In addition to determining the influence of the protein component on the photophysical and photochemical properties of photosensitizers, data are available in the literature on the effect of the lipid environment on the photodynamic activity of porphyrin compounds [ 141; the results indicate a disaggregating action of lipid vesicles on water-insoluble porphyrins. It is interesting to analyse the behaviour of water-soluble sensitizers, especially compounds of the chlorin series, during interaction with bilayer lipid membranes, since studies with tetra(p-sulphonatophenyl)porphin reveal significant differences between the photophysical parameters of this compound and its photochemical activity in solutions, membrane systems (lipid vesicles) [ 151 and erythrocyte ghosts [16]. On the basis of experimental data (spectral fluorescence and polarization measurements) and theoretical calculations (analysis of the results of the luminescence quenching of a membrane fluorescent probe by incorporated chl e6 molecules), we have investigated the interaction of the water-soluble pigment with smaller unilamellar lipidvesicles (SUV) in an attempt to determine its behaviour, properties and affinity to the lipid component. Thus the fluorescence quenching of 1,6-diphenyl-1,3,5-hexatriene (DPH) due to energy transfer to chl e6 molecules incorporated into the lipid bilayer is used as a methodological approach to investigate the biophysical aspects of interaction between the new type of photosensitizer chl e, and the lipid component of biological membranes.
2. Materials
and methods
2.1. Materials Chl e, and cNorophyll a (chl a) were obtained and identified at the Institute of Physics of the BSSR Academy of Sciences (Minsk, U.S.S.R.) by M. V. Sarzhevskaya. Serva-produced 1,6-diphenyl-1,3,5-hexatriene (DPH) and
45
sodium cholate were used without further purification. The preparation of chromatographically pure egg yolk lecithin was kindly supplied by A. F. Mironov (Institute of Fine Chemical Technology, Moscow, U.S.S.R.). All Reachim (U.S.S.R.) analytical grade organic solvents were subjected to additional purihcation by conventional methods [ 171. Aqueous solutions were prepared using double-distilled water. 2.2. Methods 2.2.1. fiepUrUtiOn Of @OSOmal CO?npkXeS Of DPH Und chl e6 DPH was incorporated into a lipid bilayer containing SUV by the method described in ref. 18. For this purpose, a small quantj’;y of DPH stock solution in tetrahydrofuran was added to an egg yoke lecithin alcoholic solution and evaporated to dryness on a rotary evaporator. The lipid film obtained was then dispersed in an isotonic phosphate buffer (pH 7.4), containing sodium cholate. The dispersion was gel chromatographed on a column (1 cm x 37 cm) packed with Sephadex G-25 (middle) (Pharmacia). The final concentration of lipid in the solution was determined from the phosphorus content using atom emission spectroscopy on a Plasma-100 instrument. The value obtained was corrected by taking into account the phosphorus content in the buffer solution. The molar lipid to DPH ratio in the experiments was 4OO:l. The concentration of DPH incorporated into the SW was determined spectrophotometrically using the molar extinction coefficient 8 x lo4 dm3 mol-’ cm- ’ at a wavelength A of 350 nm [ 191. The incorporation of chl e6 into the SUV bilayer was carried out in a similar manner, except that various quantities of chl e6 in diethyl ether were added to the lipid solution. The molar lipid to pigment ratio was varied in the experiment from 4OO:l to 1500: 1. The molar decimal extinction coefficient of chl e6 at the main long-wavelength Q,, (0,O) band maximum (E664= 5.5 X lo4 dm3 mol-’ cm-‘) was determined by titration of a buffer solution with a known quantity of pigment by increasing quantities of lipid vesicles in the region of changes in intensity and the reaching of A,, of this band onto the plateau. 2.2.2. Spectral masurmts Electronic absorption spectra were recorded on a Hitachi 150-20 spectrophotometer equipped with a microprocessor. Corrected fluorescence spectra were recorded on an SLM-4800 spectrophotometer. The pigment fluorescence lifetime TVwas measured using a PRA-3000 pulse fluorometer operating in the single-photon counting mode. The quantum yields 4 of chl e6 fluorescence were determined by the relative method [20] to an accuracy of+lO%. Chl a in diethyl ether (&=0.32) was used as a standard [21]. Fluorescence excitation polarization spectra of chl e, were recorded by the standard static photoelectrical method (with an absolute accuracy of + 2%); the angle between the direction of excitation and recording was 90”. The accuracy of the determination of the relative quantum yields of DPH fluorescence was 5% at low quencher concentrations in the solution and did not exceed 15% at its maximum concentration taking into consideration the
46
biexponential kinetics of DPH luminescence decay in SUV. All measurements were taken at room temperature. 3. Results and discussion 3.1. Special properties of chl e6 and its state in lipid vesicles It is known [22, 231 that in weak alkaline solutions (pH 7.4) chl e6 is in the monomer state as shown by its spectral and energetic parameters. In a lipid environment, various effects are observed. (i) A bathochromic shift of all bands in the electronic absorption and fluorescence spectra (Ah = 6 nm for the main long-wavelength QY(0,O) band); this may possibly be associated with the orientational effects caused by the change in the dielectric constant of the microenvironment of the sensitizer. Similar effects are observed when complexes between chl e6 and human serum albumin (HSA) or bovine serum albumin (BSA) are formed; the causes of these changes have been studied in detail experimentally [ 12 1. (ii) A slight increase in the fluorescent capacity of chl e6 (79=5.4 ns, 7t’$/ 7000 A) provides their spatial isolation in time, 7,). Estimates made on the basis of the inductive resonance theory [26] for chl e, (spectral overlapping integralJ= 3.2 X lo-l3 cm6 mol-‘; critical transfer distance Rbheor = 45 A; formulae for calculation given below) show that with uniform distribution of chl e6 molecules on WV with an external radius Reti= 150 A (the localization of chl e, molecules is described in more detail in Section 3.3.), the fluorescence depolarization (P/P,&&r=O.9 is realized even in the presence of six pigment molecules on the SUV surface (MUp:Mp= 900: 1). This decrease in the degree of chl e, emission polarization at M&kf,=900:1 is detected in our experiments: (P/P,&_= 0.85-0.9 at A,,= 425 nm and hreg=670 run. Prom this it can be concluded that the results of polarization measurements of the emission of chlorin-type compounds cannot be used to elucidate the character and parameters of their interaction with lipid vesicles. In this context, the conclusions of ref. 27 based on such measurements cannot be considered as convincing. 3.2. Determination with
of the character and constants of chl e6 association liposomes The interaction of molecules with lipid membranes is usually analysed
and described in terms of non-covalent binding or partition of the molecules between the aqueous and membrane phases. These types of interaction should be strictly differentiated, since they are based on quite different thermodynamic processes [ 281. A convenient tool for this purpose is the method proposed by Blatt et al. [28] which permits the determination of the character of the interaction of different types of molecules with lipid structures (micelIes and liposomes) and its major parameters. To this end, the fluorescence quenching of a membrane-bound probe (in our case DPH) by the incorporated quencher (chl e,J molecules is investigated as a function of chl e, concentration for a series of solutions with known molar concentrations of lipid vesicles with ratio. The results obtained are then represented as a the same MLip:Mprobe series of experimental curves with the coordinates (F,JF- l), [QIT where [Q]= is the total molar concentration of chl e, in the solution and F,, and F are the DPH fluorescence intensities in the absence and presence of quencher respectively (see Pig. 2). It is therefore assumed that the degree
48
0
5
i0
15
~Q1,407moLe.dm
-3
Fig. 2. DPH fluorescence quenching in a liposome bilayer (M,,,,:Mr,,=400:1) by incorporated chl e6 molecules (A,,=370 run, Ares=432 nm) at various liposome concentrations in solution [Lip] (10’ mol drn3): 0.5 (1); 1.1 (2); 2.2 (3); 4.9 (4); 9.4 (5); 13.4 (6). F. and Fare the DPH fluorescence intensities in the absence and presence of the quencher respectively; [Q]r is the total molar quencher concentration in the solution.
of quenching F,,/F is independent of the quenching mechanism and the statistical distribution of quenchers between the liposomes and is only determined by the average number (Q) of quencher per lipid vesicle. According to ref. 28 the following relation holds true
IQl,=$$) + (Q)[Lipl
(1)
as
where [Lip] is the molar concentration of liposomes in the suspension and K, is the generalized association parameter describing both the binding and the distribution K, = V,K, +
PKI, 1 +&[QIA
where V, is the molar volume of liposomes, Kr is the partition coefficient, K,, is the binding constant, p is the number of binding sites and [Q]* is the molar concentration of the non-associated quencher. Further analysis of the data given in Fig. 2 can be performed by plotting [QIT VS. [Lip] at tied values of (F,/F- 1). This procedure permits the determination of the parameters K, and (Q) which form the only pair of quantities for each value of F,,IF. For DPH, we must take into consideration, when carrying out this analysis, the non-exponentiality of the probe fluorescence decay kinetics; in our experiments we obtained I(t) =A, e-t’r’ +A2 emtln
(3)
where A, = 0.912, TV= 8.8 ns, A2 = 0.088 and 72= 2.6 ns. A similar situation has already been reported for DPH in synthetic and native membranes and is attributed to the conformational inhomogeneity of the probe molecules [291. Therefore when the components are not equally quenched with increasing [Q]r, the measured ratios of the intensities F,JF of the solutions may not correspond to the true quenching dynamics of each component. Direct
49
measurements of the decay kinetics of solutions with increasing [QIT have shown that the contribution of the short component is greater than 15% at [QIT> 2 X 10e6 mol dme3 when the quenching of the total DPH luminescence is F,JF>2.5 (curve 2, Fig. 2). Since, in the absence of quencher, the long component of DPH fluorescence is dominant the true change in the intensity of this component is distorted at values of F,/F> 2.5 due to the increasing contribution of the less quenched component. Accordingly, under these conditions, the measurement error at [Q]=> 2 x lo-’ mol dmw3 is significant for all the curves in Pig. 2. Therefore we have restricted ourselves to the quenching values F,/F< 2.5. The values are indicated in Pig. 2 by the broken line. [QIT VS. [Lip] curves are shown in Pig. 3(a) for the systems investigated. The rectilinear character according to eqn. (1) makes it possible to determine the generalized parameter of the association and the average number of chl e6 molecules on one vesicle using the slope and the y intercept for each curve. From the values of the parameters K, and (Q), the dependence K,=f((Q)) given in Fig. 3(b) was obtained. As can be seen from Fig. 3(b), the value of K, is independent of (Q) within experimental error. According to ref. 28, this indicates that the interaction of chl e6 with liposomes occurs only by a partition mechanism. We can easily determine the partition coefficient, which is related to the parameter K, by K,=K,V, [28]. The value obtained (I-&= 6.7 X 103) for chl e6 is about two orders of magnitude less than the analogous value for dihaematoporphyrin ether which has no polar group (interaction with liposomes prepared from dimyristoylphosphatidylcholine at 31°C [30]) and about one order of magnitude less than the value for haematoporphyrin (HP) which has two carboxyl groups (interaction with the 9.5
20
15
10 0.25
5
1
0
5
10
5
(0
LLipl~vJ-lnoc~di?
5 3
‘
and Photobidogy,
B: Biology,
7 (1990)
43-56
43
CHLORIN e,-LIPOSOME INTERACTION. INVESTIGATION BY THE METHODS OF FLUORESCENCE SPECTROSCOPY AND INDUCTIVE RESONANCE ENERGY TRANSFER k
A. F’ROLOV, E. I. ZENKEVICH,
Institute of Physics, (U.S.S.R.) (Received
G. P. GURINOVICH
BSSR Academy
June 19, 1989; accepted
of Sciences,
and G. A. KOCHUBEYEV
L+eninsky Prospekt
70, 220602 Minsk
January 24, 1990)
K~~~~TTJ!.s. Chlorin e6, 1,6-diphenyl-1,3,5_hexatriene, liposomes, localization, partition, fluorescence quenching, inductive resonance energy transfer.
On the basis of spectral fluorescence and polarization measurements and results obtained on the luminescence quenching of the membrane fluorescent probe 1,6-diphenyl-1,3,5hexatriene (DPH) by incorporated chIorin e6 (chl es) molecules, it is shown that the interaction of the water-soluble pigment with smaller unilamellar lipid vesicles occurs by a mechanism of partition between the aqueous and lipid phases (partition coefficient Kp = 6.7 X 103) and provides rigid fixing of chl e6 monomers at the boundary between the polar and non-polar parts of the lipid membrane. In terms of inductive resonance electronic excitation energy transfer between DPH and chl e6 (I&,=36.2 A), we have analysed data on DPH fluorescence quenching under different conditions of chl eg localization in the lipid bilayer and have concluded that the incorporation of the pigment molecules into the vesicles from the aqueous phase occurs mainly into the external monolayer.
1. Introduction In recent years, the use of the photodynamic effect of porphyrin compounds for selective and effective destruction of human malignant tumours has become increasingly apparent [ 1, 21. At the present time, investigations in this field are mainly focused on the solution of two problems: optimization of the methods and regimes of phototherapeutic action using the well-known compounds haematoporphyrin derivative and Photofrin II [3, 41 and the search for new effective photosensitizers which exhibit optimal photophysical and physicochemical properties [5, 61. In this context, chlorin e6 (chl e6) and its derivatives are a promising new class of compounds for the photochemotherapy of tumours [ 7, 91.
loll-1344/90/$3.50
0 Elsevier
Sequoia/Printed
in The Netherlands
44
However, it should be noted that the choice and optimization of new tetrapyrrole-based photosensitizers from the results of investigations of their photophysical parameters in homogeneous solutions are not always consistent with their photodynamic activity in biological systems which are characterized by a significant heterogeneity of composition and properties. The elucidation of the detailed mechanisms of photodestruction of cellular systems requires a stepwise study of the interactions between sensitizers and components of biological membranes (proteins and lipids). The data available in the literature clearly demonstrate the usefulness of such an approach. For example, investigation of tryptophan fluorescence quenching by porphyrins [ 10, 1 l] and chl e, [ 121 in pro+.eins has permitted the determination of the nature of the photosensitizer-protein interaction, the localization of the conjugate macrocycles on the protein carrier and the correlation of the structural organization of pigment-protein complexes with the results obtained on the photosensitized destruction of various amino acid residues in proteins [ 131. In addition to determining the influence of the protein component on the photophysical and photochemical properties of photosensitizers, data are available in the literature on the effect of the lipid environment on the photodynamic activity of porphyrin compounds [ 141; the results indicate a disaggregating action of lipid vesicles on water-insoluble porphyrins. It is interesting to analyse the behaviour of water-soluble sensitizers, especially compounds of the chlorin series, during interaction with bilayer lipid membranes, since studies with tetra(p-sulphonatophenyl)porphin reveal significant differences between the photophysical parameters of this compound and its photochemical activity in solutions, membrane systems (lipid vesicles) [ 151 and erythrocyte ghosts [16]. On the basis of experimental data (spectral fluorescence and polarization measurements) and theoretical calculations (analysis of the results of the luminescence quenching of a membrane fluorescent probe by incorporated chl e6 molecules), we have investigated the interaction of the water-soluble pigment with smaller unilamellar lipidvesicles (SUV) in an attempt to determine its behaviour, properties and affinity to the lipid component. Thus the fluorescence quenching of 1,6-diphenyl-1,3,5-hexatriene (DPH) due to energy transfer to chl e6 molecules incorporated into the lipid bilayer is used as a methodological approach to investigate the biophysical aspects of interaction between the new type of photosensitizer chl e, and the lipid component of biological membranes.
2. Materials
and methods
2.1. Materials Chl e, and cNorophyll a (chl a) were obtained and identified at the Institute of Physics of the BSSR Academy of Sciences (Minsk, U.S.S.R.) by M. V. Sarzhevskaya. Serva-produced 1,6-diphenyl-1,3,5-hexatriene (DPH) and
45
sodium cholate were used without further purification. The preparation of chromatographically pure egg yolk lecithin was kindly supplied by A. F. Mironov (Institute of Fine Chemical Technology, Moscow, U.S.S.R.). All Reachim (U.S.S.R.) analytical grade organic solvents were subjected to additional purihcation by conventional methods [ 171. Aqueous solutions were prepared using double-distilled water. 2.2. Methods 2.2.1. fiepUrUtiOn Of @OSOmal CO?npkXeS Of DPH Und chl e6 DPH was incorporated into a lipid bilayer containing SUV by the method described in ref. 18. For this purpose, a small quantj’;y of DPH stock solution in tetrahydrofuran was added to an egg yoke lecithin alcoholic solution and evaporated to dryness on a rotary evaporator. The lipid film obtained was then dispersed in an isotonic phosphate buffer (pH 7.4), containing sodium cholate. The dispersion was gel chromatographed on a column (1 cm x 37 cm) packed with Sephadex G-25 (middle) (Pharmacia). The final concentration of lipid in the solution was determined from the phosphorus content using atom emission spectroscopy on a Plasma-100 instrument. The value obtained was corrected by taking into account the phosphorus content in the buffer solution. The molar lipid to DPH ratio in the experiments was 4OO:l. The concentration of DPH incorporated into the SW was determined spectrophotometrically using the molar extinction coefficient 8 x lo4 dm3 mol-’ cm- ’ at a wavelength A of 350 nm [ 191. The incorporation of chl e6 into the SUV bilayer was carried out in a similar manner, except that various quantities of chl e6 in diethyl ether were added to the lipid solution. The molar lipid to pigment ratio was varied in the experiment from 4OO:l to 1500: 1. The molar decimal extinction coefficient of chl e6 at the main long-wavelength Q,, (0,O) band maximum (E664= 5.5 X lo4 dm3 mol-’ cm-‘) was determined by titration of a buffer solution with a known quantity of pigment by increasing quantities of lipid vesicles in the region of changes in intensity and the reaching of A,, of this band onto the plateau. 2.2.2. Spectral masurmts Electronic absorption spectra were recorded on a Hitachi 150-20 spectrophotometer equipped with a microprocessor. Corrected fluorescence spectra were recorded on an SLM-4800 spectrophotometer. The pigment fluorescence lifetime TVwas measured using a PRA-3000 pulse fluorometer operating in the single-photon counting mode. The quantum yields 4 of chl e6 fluorescence were determined by the relative method [20] to an accuracy of+lO%. Chl a in diethyl ether (&=0.32) was used as a standard [21]. Fluorescence excitation polarization spectra of chl e, were recorded by the standard static photoelectrical method (with an absolute accuracy of + 2%); the angle between the direction of excitation and recording was 90”. The accuracy of the determination of the relative quantum yields of DPH fluorescence was 5% at low quencher concentrations in the solution and did not exceed 15% at its maximum concentration taking into consideration the
46
biexponential kinetics of DPH luminescence decay in SUV. All measurements were taken at room temperature. 3. Results and discussion 3.1. Special properties of chl e6 and its state in lipid vesicles It is known [22, 231 that in weak alkaline solutions (pH 7.4) chl e6 is in the monomer state as shown by its spectral and energetic parameters. In a lipid environment, various effects are observed. (i) A bathochromic shift of all bands in the electronic absorption and fluorescence spectra (Ah = 6 nm for the main long-wavelength QY(0,O) band); this may possibly be associated with the orientational effects caused by the change in the dielectric constant of the microenvironment of the sensitizer. Similar effects are observed when complexes between chl e6 and human serum albumin (HSA) or bovine serum albumin (BSA) are formed; the causes of these changes have been studied in detail experimentally [ 12 1. (ii) A slight increase in the fluorescent capacity of chl e6 (79=5.4 ns, 7t’$/ 7000 A) provides their spatial isolation in time, 7,). Estimates made on the basis of the inductive resonance theory [26] for chl e, (spectral overlapping integralJ= 3.2 X lo-l3 cm6 mol-‘; critical transfer distance Rbheor = 45 A; formulae for calculation given below) show that with uniform distribution of chl e6 molecules on WV with an external radius Reti= 150 A (the localization of chl e, molecules is described in more detail in Section 3.3.), the fluorescence depolarization (P/P,&&r=O.9 is realized even in the presence of six pigment molecules on the SUV surface (MUp:Mp= 900: 1). This decrease in the degree of chl e, emission polarization at M&kf,=900:1 is detected in our experiments: (P/P,&_= 0.85-0.9 at A,,= 425 nm and hreg=670 run. Prom this it can be concluded that the results of polarization measurements of the emission of chlorin-type compounds cannot be used to elucidate the character and parameters of their interaction with lipid vesicles. In this context, the conclusions of ref. 27 based on such measurements cannot be considered as convincing. 3.2. Determination with
of the character and constants of chl e6 association liposomes The interaction of molecules with lipid membranes is usually analysed
and described in terms of non-covalent binding or partition of the molecules between the aqueous and membrane phases. These types of interaction should be strictly differentiated, since they are based on quite different thermodynamic processes [ 281. A convenient tool for this purpose is the method proposed by Blatt et al. [28] which permits the determination of the character of the interaction of different types of molecules with lipid structures (micelIes and liposomes) and its major parameters. To this end, the fluorescence quenching of a membrane-bound probe (in our case DPH) by the incorporated quencher (chl e,J molecules is investigated as a function of chl e, concentration for a series of solutions with known molar concentrations of lipid vesicles with ratio. The results obtained are then represented as a the same MLip:Mprobe series of experimental curves with the coordinates (F,JF- l), [QIT where [Q]= is the total molar concentration of chl e, in the solution and F,, and F are the DPH fluorescence intensities in the absence and presence of quencher respectively (see Pig. 2). It is therefore assumed that the degree
48
0
5
i0
15
~Q1,407moLe.dm
-3
Fig. 2. DPH fluorescence quenching in a liposome bilayer (M,,,,:Mr,,=400:1) by incorporated chl e6 molecules (A,,=370 run, Ares=432 nm) at various liposome concentrations in solution [Lip] (10’ mol drn3): 0.5 (1); 1.1 (2); 2.2 (3); 4.9 (4); 9.4 (5); 13.4 (6). F. and Fare the DPH fluorescence intensities in the absence and presence of the quencher respectively; [Q]r is the total molar quencher concentration in the solution.
of quenching F,,/F is independent of the quenching mechanism and the statistical distribution of quenchers between the liposomes and is only determined by the average number (Q) of quencher per lipid vesicle. According to ref. 28 the following relation holds true
IQl,=$$) + (Q)[Lipl
(1)
as
where [Lip] is the molar concentration of liposomes in the suspension and K, is the generalized association parameter describing both the binding and the distribution K, = V,K, +
PKI, 1 +&[QIA
where V, is the molar volume of liposomes, Kr is the partition coefficient, K,, is the binding constant, p is the number of binding sites and [Q]* is the molar concentration of the non-associated quencher. Further analysis of the data given in Fig. 2 can be performed by plotting [QIT VS. [Lip] at tied values of (F,/F- 1). This procedure permits the determination of the parameters K, and (Q) which form the only pair of quantities for each value of F,,IF. For DPH, we must take into consideration, when carrying out this analysis, the non-exponentiality of the probe fluorescence decay kinetics; in our experiments we obtained I(t) =A, e-t’r’ +A2 emtln
(3)
where A, = 0.912, TV= 8.8 ns, A2 = 0.088 and 72= 2.6 ns. A similar situation has already been reported for DPH in synthetic and native membranes and is attributed to the conformational inhomogeneity of the probe molecules [291. Therefore when the components are not equally quenched with increasing [Q]r, the measured ratios of the intensities F,JF of the solutions may not correspond to the true quenching dynamics of each component. Direct
49
measurements of the decay kinetics of solutions with increasing [QIT have shown that the contribution of the short component is greater than 15% at [QIT> 2 X 10e6 mol dme3 when the quenching of the total DPH luminescence is F,JF>2.5 (curve 2, Fig. 2). Since, in the absence of quencher, the long component of DPH fluorescence is dominant the true change in the intensity of this component is distorted at values of F,/F> 2.5 due to the increasing contribution of the less quenched component. Accordingly, under these conditions, the measurement error at [Q]=> 2 x lo-’ mol dmw3 is significant for all the curves in Pig. 2. Therefore we have restricted ourselves to the quenching values F,/F< 2.5. The values are indicated in Pig. 2 by the broken line. [QIT VS. [Lip] curves are shown in Pig. 3(a) for the systems investigated. The rectilinear character according to eqn. (1) makes it possible to determine the generalized parameter of the association and the average number of chl e6 molecules on one vesicle using the slope and the y intercept for each curve. From the values of the parameters K, and (Q), the dependence K,=f((Q)) given in Fig. 3(b) was obtained. As can be seen from Fig. 3(b), the value of K, is independent of (Q) within experimental error. According to ref. 28, this indicates that the interaction of chl e6 with liposomes occurs only by a partition mechanism. We can easily determine the partition coefficient, which is related to the parameter K, by K,=K,V, [28]. The value obtained (I-&= 6.7 X 103) for chl e6 is about two orders of magnitude less than the analogous value for dihaematoporphyrin ether which has no polar group (interaction with liposomes prepared from dimyristoylphosphatidylcholine at 31°C [30]) and about one order of magnitude less than the value for haematoporphyrin (HP) which has two carboxyl groups (interaction with the 9.5
20
15
10 0.25
5
1
0
5
10
5
(0
LLipl~vJ-lnoc~di?
5 3
‘
