J. Photo&em.
Photobiol.
B: Biol.,
Laser picosecond haematoporphyrin
10 (1991)
239-243
microspectrofluorometry in cells and liposomes
E. B. Chernyaeva, A. G. Vardanyan, and 0. V. Lobanov MoscowState University, Physics Department,
N. I. Koroteev, MGU,
Lenin
Hills,
of
V. F. Kamalov Moscow
119899
(U.S.S.R.)
A. F. Mironov Moscow Institute (U.S.S.R.] (Received
and V. D. Rumyanzeva of Fine
June 7, 1990;
Chemical
accepted
Technology,
February
Vemodsky
au. 86, Moscow
117571
7, 1991)
Keywords. Porphyrins, liposomes, time-resolved fluorescence microscopy.
Abstract The results of a laser picosecond microspectrofluorometric study of the spectral and kinetic characteristics of haematoporphyrin (Hp) fluorescence at various sites in cultured SPEV cells and phosphatidylcholine liposomes are presented. The computer-controlled detection system is based on the single-photon counting method with picosecond time resolution. In aqueous medium, the Hp fluorescence spectrum is characterized by two bands at 615 and 675 run. In living cells and llposomes, Hp fluorescence is red shifted to 630 and 690 run. In addition a new band at 665 nm is detected. The dependence of this band on the incubation time and Hp concentration was investigated. The fluorescence decay kinetics of Hp in a culture medium, liposome and a cell nuclear membrane were measured. Possible Hp aggregate formation in the lipid bilayer and its implications are discussed.
1. Introduction
The photodynamic therapy (PDT) of cancer has been developed over the past 10 years and is now effectively used for the treatment of a variety of tumours [ 1, 21. The method is based on a cytotoxic effect caused by the excited molecules of a photosensitizer which has a greater affinity or malignant tissue than normal tissue. Notwithstanding the increase in the clinical applications of PDT, the mechanism of the interaction of the photosensitizer with cells and tissues has not been fully established. Haematoporphyrin derivative (HpD) and its more selectively accumulating hydrophobic fraction 1991
-
Elsevier
Sequoia,
Lausanne
240
Photofrin II are usually used as photosensitizers [3]. Recently, a reaction has been obtained called dihaematoporphyrinether (DHE) which consists of a complex mixture of ether and ester Hp oligomers (41. To our notion, the correct terminology has been suggested by Kessel [5] who rejects the term “DHE”. However, both terms are used in this paper. HpD, Photofrin II and DHE are well aggregated in aqueous solutions (61 due to the hydrophobic interactions of the molecules. Binding to lipid bilayers has been reported to monomerize these aggregates [6] and to unfold the oligomers [ 71. Singlet oxygen (‘0,) has been identified as the main cytotoxic agent in vivo (31. It should be noted that Hp aggregates are poor singlet oxygen generators [ 2 1. Evidence has been obtained [3, 81 that the plasma membrane and membranes of the intracellular organelles are the main targets of photodynamic damage to HpD-sensitized cells. Binding to membranes, liposomes and proteins shifts the Hp and HpD fluorescence spectra to the red by about 15 nm [91. Thus optical spectroscopy can be used to investigate the intracellular dynamics of Hp and its complexes. Roeder and Wabnitz [ 10 ] investigated Hp aggregation in aqueous solution using laser fluorescence spectroscopy with temporal resolution. A new fluorescent band was found at 640 nm and attributed to the photoproduct formed from the Hp aggregates. Valuable information on photosensitizer microenvironment and state of aggregation can be obtained from fluorescence anisotropy measurements [ 111. Yamashita et al. [ 121 measured the spatial Hp distribution in cells using a streak camera. The decay kinetics of Hp fluorescence in single cells were obtained by Doccio et al. [ 131 and Schneckenburger et al. [ 141. Picosecond fluorescence spectroscopy has been applied to study the decay kinetics of Hp fluorescence in cell suspensions [7]. In this work, the appearance of a new Hp intracellular fluorescence band at 665 run was established. This new emission band also observed for HpD and DHE in micelles when the concentration was below the critical concentration of micelle formation [ 15, 161. Its contribution decreased at higher surfactant concentrations. In this paper, the fluorescence spectrum and decay kinetics of Hp in subcellular structures and liposomes are reported. The dependence of the intracellular Hp fluorescence spectrum on the incubation time, irradiation dose and Hp concentration in the outer medium is studied.
2. Materials
and methods
2.1. Porphyrin HP-IX (purity, 95%), produced at the Moscow Institute of Fine Chemical Technology (MIFCT), was used as received. To obtain the stock solution (10m2 M), Hp was diluted in 1% NaOH and the pH was adjusted to 7.2 with
241
1% HCl. The stock solution was kept at 0 “C in the dark. In experiments with cells, a solution of 10m4 M or 10m5 M Hp in medium 199 was used (incubation medium). 2.2. Liposomes Multilayer and monolayer liposomes from egg phosphatidylcholine (PC) were obtained from MIFCT as a suspension (30 mg PC ml-‘) in Tris buffer (0.05 M; pH 7.4). For experiments, the suspension was diluted in the same buffer to a concentration of 1O-4 M and Hp stock solution was added to a fmal concentration of 10m4 M, 10m5 M or 5X lO-‘j M. After the necessary incubation time, a small drop of liposome suspension with Hp was placed between the subject and cover glasses and examined under a laser microscope. Measurements were made on the liposomes adhered to the cover glass. 2.3. Cell lines The SPEV cell line was used; the cultivating procedure is described elsewhere [ 171. This cell line originates from the kidneys of pig embryo’s and was first described in the U.S.S.R. [ 1’71. Monolayers of cells grown on the cover glass were used on the third day of growth. The cells were washed twice with medium 199 and were then incubated in Hp solution in the dark at room temperature. Just before measurement, the sample was placed into a specially designed experimental chamber filled with medium 199 and was then examined under the laser microscope. 2.4. Choice of excitation parameters Prior to the spectral and kinetics studies of Hp fluorescence in cells, the laser radiation intensity was adjusted such, that it did not result in cell damage. Plasma membrane integrity was tested using trypan blue. By variation of the laser radiation intensity and irradiation time, it was shown that irradiation of a cell site of 1 pm in diameter with 0.01 PW (A-532 nm) for 10 min caused no membrane damage; no photobleaching of Hp was observed.
3. Instrumentation The experimental set-up of the laser picosecond microspectrofluorometer is shown in Fig. 1. A continuous wave (CW) pumped, acousto-optically & = switched and mode-locked Nd:YAG laser was used as the master generator. The laser output at 1.064 pm consists of a sequence of trains (repetition rate, 5 kHz), each containing 25-30 pulses of 80 ps duration. For sample excitation, a single pulse was selected from the train using an electro-optical modulator (EOM). Cell fluorescence was obtained using pulses of the second harmonic (SH) of the Nd:YAG laser of the second harmonic of a picosecond continuum, generated by stimulated Raman scattering in a single optical fibre (OF). A
242
YAG:h'd
Fig. 1. Laser picosecond microspectrofluorometer: SH, second harmonic crystal; EOM, electrooptical modulator; OF, optical fibre; 0, objective; M, monochromator; PMT, photomultiplier tube; PD, photodiode; Ll, L2, L3, lenses; SPC, single-photon counting system.
high quality objective (“Reichert”, 160 X , NA= 1.25) (0) was used to focus the exciting laser beam onto the sample, resulting in a beam width smaller than 1 pm. It should be noted that, due to the high spatial resolution and high Hp concentration in the lipid bilayer, there is practically no contribution of Hp fluorescence from the aqueous medium, in contrast with the results obtained with the cell suspensions [9]. The fluorescence signal from the selected subcellular structure or liposome was focused by a lens (L3) onto the input slit of a monochromator (M) (spectral resolution, l-2 nm) and detected by a fast photomultiplier tube (PMT). The detection system based on the time-correlated single-photon counting technique, was computer controlled. The output from the time-toamplitude converter, which has two time gates (O-30 ns and O-300 ns), was fed directly to a microcomputer. The full width at half-maximum of the detection system response function (Pig. 5, broken line, see later) was about 800 ps. The specially designed software enabled us to obtain either the fluorescence decay kinetics at the selected emission wavelengths, or the time-resolved fluorescence spectra detected at various time delays after the excitation pulse. A deconvolution procedure was used for data processing [ 131. The experimental fluorescence decay curves at different wavelengths were fitted by a convolution of the system response function with a polyexponential fluorescence intensity decay function: F(t) = C”ai exp( - t/~~), n = 1,2,3. However, in our case (limited data collection time), the contribution of the middle component was negligible in comparison with the other two; consideration of the middle component did not produce a better fitting. The quality of approximation was estimated by the x2 value, the residual autocorrelation function and visual fitting were considered to be satisfactory for x2 < 1.5 and when an increase in n produced no more than a 1% decrease in x2.
243
4. Results and discussion The spectra and decay kinetics of Hp fluorescence were measured in medium 199 (10v4 M). The fluorescence spectra were characterized by two bands at 615 and 675 nm. The fluorescence decay at 615 and 675 nm was found to be monoexponential with a lifetime 7 of 13 ns (corresponding to Hp monomer fluorescence [ 181). In order to study the Hp fluorescence in a lipid environment, the spectral and kinetic characteristics of Hp in PC liposomes (model of a membrane lipid bilayer) were measured. Relative to the fluorescence spectrum of Hp in aqueous solution, the fluorescence bands in the PC liposomes were shifted by 15 nm to the red. The dependence of the fluorescence spectrum of Hp in PC liposomes on the incubation time is presented in Pig. 2. The fluorescence spectrum displays two bands at 630 and 690 run. With an increase in incubation time from 1 to 4 h, a new fluorescence band at 665 nm appears. The ratio of the fluorescence intensity of the band at 665 nm to that of the band at 630 nm increases from 0.3 at 2 h incubation to 0.4 at 4 h to 0.5 at 6 h. A decrease in Hp concentration to 10T5 M and 5X lo-’ M results in a decrease in the contribution of the new band: the 1665/1830ratio is only 0.2 for 6 h incubation for 5 X lop6 M Hp. The Hp fluorescence decay kinetics from PC Iiposomes were recorded at A= 635 nm and A= 665 run for two Hp concentrations (C= 10m4 M and C= 10e5 M) and two incubation times (1 h and 6 h). These decay kinetics were found to be biexponential with F(t) = a, exp( - t /TV) + a2 exp( - t/T2). The data are presented in Table 1. At h= 665 run, the contribution of the fast component increases with an increase in Hp concentration and incubation time, whereas at h= 635 nm the pattern is quite different. In particular, a decrease in the contribution of the fast component is observed at increasing concentrations. The fluorescence spectra and decay kinetics of Hp were measured at the nuclear and cytoplasmic membrane sites of cultured SPEV cells. The
600
655
Fig. 2. Dependence ([HP]= lo+’ M).
710
X,nm
of the Hp fluorescence
spectrum
in PC liposomes
on incubation
time
244 TABLE 1 Fitting of the HP fluorescence decay curve for the liposomes Parameter
n 6-1
Wavelength (nm)
C=10-4 lh
6h
lh
6h
635
0.5 23.8 0.17
0.6 19.9 0.16
0.20 15.4 0.07
0.3 15.0 0.05
665
0.4 15.2 0.1
0.4 10.1 0.06
0.4 16.0 0.33
0.3 14.3 0.1
72 (ns) a2la,
71 (ns) 72 a2fal
(ns>
M
C=10-6
M
Fig. 3. Dependence of the Hp fluorescence spectrum from the cytoplasmic membrane of SPEV cells on incubation time ([HP] = 10m6 M).
dependence of the Hp fluorescence spectrum from the cytoplasmic membrane on the incubation time is shown in Fig. 3. It is quite similar to that observed in liposomes. For incubation times longer than 2 h, the band at 665 run is detected, its relative contribution increasing with an increase in incubation time. The Hp fluorescence spectrum at the site of the nuclear membrane (C= 10m4 M) for an incubation time of 4 h is shown in Fig. 4. It can be seen that this spectrum is identical with that observed in liposomes after prolonged incubation. The lifetimes of the three fluorescence bands were measured. The data obtained from the fitting of the fluorescence decay curve for the nuclear membrane are presented in Table 2. The fluorescence decay is monoexponential at 690 nm with a lifetime of 11 ns and biexponential at 630 and 665 run
245
01
600
650
700 A,nm
Fig. 4. Hp fluorescence spectrum at the site of the nuclear incubation time of 4 h ([HP] = 10m4 M).
TABLE
2
Fitting
of the Hp fluorescence
Parameter
decay
(ns)
72
cm>
al
Ia2
or the nuclear
membrane
of SPEV cells for an
of a SPEV cell
J+l (nm) 635
71
curve
membrane
665
0.5
0.6
12.2 2.5
13.4 5
690 11.0 -
with lifetimes of r1= 0.5 ns, 7-a= 12.2 ns (aI /czz= 5) and T, = 0.6 ns, T~= 13.4 ns (a, /uZ = 1.3) respectively. The amplitude of the fast component is larger at 665 nm. The apparatus function, fluorescence decay kinetics and fluorescence decay curve for Hp fluorescence at the nuclear membrane of the cell at 630 nm are presented in Fig. 5 together with the residual autocorrelation function. The similar spectra and decay kinetics of Hp fluorescence observed in liposomes and the cytoplasmic and nuclear membranes of the cell indicate that the Hp localized in the lipid bilayer of the membrane makes an overwhelming contribution to the fluorescence signal from the membrane. The new band at 665-670 nm with a short lifetime (observed after prolonged incubation and at high Hp concentrations (sufficiently high porphyrin to lipid molar ratio)) may be due to Hp aggregates or to the binding of Hp to sites inside or at the surface of the lipid bilayer. The possible formation of Hp photoproducts must also be considered. To clarify this, we studied the dependence of the Hp fluorescence intensity at 665-670 nm on the irradiation dose. No change in I665 was observed for
246
Ol
II
I
-\
0
I
10
20 T,ns
Fig.5. Hp fluorescence decay kinetics recorded at the site of the nuclear membrane ([HP] = 10-6 M): 9, experimental data; ---, apparatus function; -, autocorrelation function is given in the inset.
fluorescence decay curve. The residual
irradiation doses in the range lo4 to 6 X lo6 J rnm2.No Hp photobleaching was detected during our measurements. Other data in favour of the formation of Hp aggregates in membranes are as follows: (i) the significant decrease in the photodynamic damage of the cell plasma membrane on prolonged incubation [ 81, which can be explained by the low quantum yield of singlet oxygen generation by Hp aggregates [ 21; (ii) the decrease in Hp fluorescence intensity at the cell plasma membrane after 4-8 h incubation time [ 191, which can be explained by the very low fluorescence yield of aggregates; (iii) the distinct fluorescence of Hp aggregates in the cell plasma membrane and membranes of intracellular structures [20]; (iv) the fluorescence of DHE in liposomes which suggests self-association in a lipid environment [21]. The possible formation of Hp aggregates in the lipid bilayer of membranes, leading to a decrease in the extent of photodynamic damage, must be considered when choosing the PDT regimen.
Acknowledgments The authors are grateful to Professor S. A. Akhmanov for support of this work, Dr. L. L. Litinskaya and Dr. M. I. Leikina for supplying the cell culture and for fruitful discussions and I. Deruzhenko for preparation of the samples.
247
References
1 T. J. Dougherty, J. E. Kaufman, A. Goldfarb, K. R. Welshaupt, D. Boyle and A. Mittleman, Photoradiation therapy for treatment of malignant tumors, Cancer Res., 38 (1986) 2628-2635. 2 D. Kessel, Hp and HpD: photophysics, photochemistry and phototherapy, Pholocha.
Photobiol., 39 (1984) 851-859. 3 T. J. Dougherty, Photosensitlzers: Photobiol., 45 (1987) 879-889.
therapy and detection of malignant tumors, Photo&em.
4 T. J. Dougherty, Studies on the structure of porphyrins contained in Photofrln II, Photo&em. Photobiol., 46 (1987) 569-573. 5 D. Kessel, HpD-terminology, Photo&em. Photobiol., 50 (1989) in Guest editorial. 6 R. Margalit and S. Cohen, Studies of hematoporphyrln derivative equilibria in heterogeneous systems. Porphyrin-liposome binding and porphyrin aqueous dlmerization, Biochim; Biophys. Acta, 736 (1983) 163-170. 7 M. Yamashita, T. Tomono, S. Kobayashi, K. Torlzuka and K. Aizawa, Picosecond fluorescence spectroscopy on incorporation process of HpD into malignant tumor cells in vitro,
Photo&em. Photobiol., 47 (1988) 189-192. 8 D. Kessel, Sites of photosensitization by derivatives 9
10
11
12 13 14
15
16 17
18
19
20
of Hp, Photo&em. Photobiol., 44 (1986) 489-493. B. Ehrenberg, Z. Mallk and Y. Nitzan, Fluorescence spectral changes of hematoporphyrin derivative upon binding to lipid vesicles, Staphylococcus aureus and Escherichia coli cells, Photochem. Photobiol., 41 (1985) 429-437. B. Roedcr and H. Wabnitz, Time-resolved fluorescence spectroscopy of hematoporphyrin, mesoporphyrln, pheophorbide a and chlorine e a ln ethanol and aqueous solution, J. Photo&em. Photobiol. B: Biol., I (1987) 103-113. A. G. Vardanyan, V. F. Kamalov, A. F. Mlronov, I. A. Struganova and B. N. Toleutaev, Picosecond time-resolved anisotropy decay kinetics of hematoporphyrin fluorescence in solution Opt. Spektrosk., 66 (1989) 346-348 (in Russian). M. Yamashita, M. Nomura, S. Kobayashi, T. Sato and K. Aizawa, Picosecond time-resolved fluorescence spectroscopy of HpD, J. Quantum Electron., 20 (1984) 1363-1367. F. Doccio, R. Ramponi, C. A. Sacchi, G. Bottiroli and I. Freitas, Fluorescence studies of biological molecules by laser h-radiation, Lasers Surg. Med., 2 (1982) 21-28. H. Schneckenburger, F. Pauker, E. Unsold and D. Jocham, Intracellular distribution and retention of the fluorescent components of Photofrln II, Photobiochem, Photobiophys., 10 (1985) 61-67. N. Miyoshl, H. Hlsazuml, 0. Ueki, K. Nakajima and M. Fukuda, Photodynamic inactivation of cultivated human bladder cancer cells (KK-47) sensitized by HpD and argon-dye laser light, Photobiochem. Photobiophys., 9 (1985) 241-252. R. Cubeddu, A. Andreoni, C. N. Knox and T. G. Truscott, Fluorescence lifetimes of angular furocoumarlns, Photo&em. Photobiol., 46 (1987) 169-173. Z. D. Sultanova, K. S. Kulikova and E. V. Gurvich, Comparative morphological, cultivating and vlrusological characteristics of cell cultures, Vopr. Virusol., 3 (1968) 291-298 (in Russian). A. Andreoni, R. Cubeddu, S. De Sllvestry, G. Jori, P. Laporta and E. Reddi, Time-resolved fluorescence studies of hematoporphyrin ln diierent solvent systems, 2. Naturforsch., Teil C, 38 (1983) 83-89. Z. A. Kchurshilova, M. A. Lelkiia, L. L. Lltlnskaya and E. B. Chemyaeva, Microfluorimetric study of porphyrin type photosensitizer accumulation in normal and malignant cultured cells, Vestn. Mosk. Univ., Fiz Asronomiya, 27 (1986) (in Russian). H. K. Scidlitz, H. Schneckenburger and K. Stettmaier, Time-resolved polarization measurements of porphyrin fluorescence ln solution and ln single cells, J. Photo&em. Photobiol.
B: Biol., 5 (1990)
391-400.
248 21 S. E. Kendrick, J. Persky, M. L. E&l-rawer, A. L. Imbembo, J. D. Fleischmann and K. A.
Koehler, Dihematoporphyrin-ether-phospholipid interactions. The roles of surface charge and lateral phase separations. Photo&em. Photobiol., 46 (1987) 669-673. 22 L. Eidus, Yu. N. Korystov, L. L. Litlnskaya, M. A. Morosov and A. M. Vexler, The study of dye accumulation in granules on vital dyeing of Chinese hamster fibroblasts in culture, Stud. Biophys., 62 (1977) 85-92. 23 E. B. Chernyaeva, E. A. Shleina, Z. A. Kchurshilova, L. L. Litinskaya and M. I. Leikina, Intracellular dynamics of hematoporphyrln: action of light, Stud. Biophys., I25 (1988) 203-210. 24 D. V. O’Connor and D. Phillips, Time-Correlated
London, 1984.
Single Photon
Counting,
Academic Press.
Photobiol.
B: Biol.,
Laser picosecond haematoporphyrin
10 (1991)
239-243
microspectrofluorometry in cells and liposomes
E. B. Chernyaeva, A. G. Vardanyan, and 0. V. Lobanov MoscowState University, Physics Department,
N. I. Koroteev, MGU,
Lenin
Hills,
of
V. F. Kamalov Moscow
119899
(U.S.S.R.)
A. F. Mironov Moscow Institute (U.S.S.R.] (Received
and V. D. Rumyanzeva of Fine
June 7, 1990;
Chemical
accepted
Technology,
February
Vemodsky
au. 86, Moscow
117571
7, 1991)
Keywords. Porphyrins, liposomes, time-resolved fluorescence microscopy.
Abstract The results of a laser picosecond microspectrofluorometric study of the spectral and kinetic characteristics of haematoporphyrin (Hp) fluorescence at various sites in cultured SPEV cells and phosphatidylcholine liposomes are presented. The computer-controlled detection system is based on the single-photon counting method with picosecond time resolution. In aqueous medium, the Hp fluorescence spectrum is characterized by two bands at 615 and 675 run. In living cells and llposomes, Hp fluorescence is red shifted to 630 and 690 run. In addition a new band at 665 nm is detected. The dependence of this band on the incubation time and Hp concentration was investigated. The fluorescence decay kinetics of Hp in a culture medium, liposome and a cell nuclear membrane were measured. Possible Hp aggregate formation in the lipid bilayer and its implications are discussed.
1. Introduction
The photodynamic therapy (PDT) of cancer has been developed over the past 10 years and is now effectively used for the treatment of a variety of tumours [ 1, 21. The method is based on a cytotoxic effect caused by the excited molecules of a photosensitizer which has a greater affinity or malignant tissue than normal tissue. Notwithstanding the increase in the clinical applications of PDT, the mechanism of the interaction of the photosensitizer with cells and tissues has not been fully established. Haematoporphyrin derivative (HpD) and its more selectively accumulating hydrophobic fraction 1991
-
Elsevier
Sequoia,
Lausanne
240
Photofrin II are usually used as photosensitizers [3]. Recently, a reaction has been obtained called dihaematoporphyrinether (DHE) which consists of a complex mixture of ether and ester Hp oligomers (41. To our notion, the correct terminology has been suggested by Kessel [5] who rejects the term “DHE”. However, both terms are used in this paper. HpD, Photofrin II and DHE are well aggregated in aqueous solutions (61 due to the hydrophobic interactions of the molecules. Binding to lipid bilayers has been reported to monomerize these aggregates [6] and to unfold the oligomers [ 71. Singlet oxygen (‘0,) has been identified as the main cytotoxic agent in vivo (31. It should be noted that Hp aggregates are poor singlet oxygen generators [ 2 1. Evidence has been obtained [3, 81 that the plasma membrane and membranes of the intracellular organelles are the main targets of photodynamic damage to HpD-sensitized cells. Binding to membranes, liposomes and proteins shifts the Hp and HpD fluorescence spectra to the red by about 15 nm [91. Thus optical spectroscopy can be used to investigate the intracellular dynamics of Hp and its complexes. Roeder and Wabnitz [ 10 ] investigated Hp aggregation in aqueous solution using laser fluorescence spectroscopy with temporal resolution. A new fluorescent band was found at 640 nm and attributed to the photoproduct formed from the Hp aggregates. Valuable information on photosensitizer microenvironment and state of aggregation can be obtained from fluorescence anisotropy measurements [ 111. Yamashita et al. [ 121 measured the spatial Hp distribution in cells using a streak camera. The decay kinetics of Hp fluorescence in single cells were obtained by Doccio et al. [ 131 and Schneckenburger et al. [ 141. Picosecond fluorescence spectroscopy has been applied to study the decay kinetics of Hp fluorescence in cell suspensions [7]. In this work, the appearance of a new Hp intracellular fluorescence band at 665 run was established. This new emission band also observed for HpD and DHE in micelles when the concentration was below the critical concentration of micelle formation [ 15, 161. Its contribution decreased at higher surfactant concentrations. In this paper, the fluorescence spectrum and decay kinetics of Hp in subcellular structures and liposomes are reported. The dependence of the intracellular Hp fluorescence spectrum on the incubation time, irradiation dose and Hp concentration in the outer medium is studied.
2. Materials
and methods
2.1. Porphyrin HP-IX (purity, 95%), produced at the Moscow Institute of Fine Chemical Technology (MIFCT), was used as received. To obtain the stock solution (10m2 M), Hp was diluted in 1% NaOH and the pH was adjusted to 7.2 with
241
1% HCl. The stock solution was kept at 0 “C in the dark. In experiments with cells, a solution of 10m4 M or 10m5 M Hp in medium 199 was used (incubation medium). 2.2. Liposomes Multilayer and monolayer liposomes from egg phosphatidylcholine (PC) were obtained from MIFCT as a suspension (30 mg PC ml-‘) in Tris buffer (0.05 M; pH 7.4). For experiments, the suspension was diluted in the same buffer to a concentration of 1O-4 M and Hp stock solution was added to a fmal concentration of 10m4 M, 10m5 M or 5X lO-‘j M. After the necessary incubation time, a small drop of liposome suspension with Hp was placed between the subject and cover glasses and examined under a laser microscope. Measurements were made on the liposomes adhered to the cover glass. 2.3. Cell lines The SPEV cell line was used; the cultivating procedure is described elsewhere [ 171. This cell line originates from the kidneys of pig embryo’s and was first described in the U.S.S.R. [ 1’71. Monolayers of cells grown on the cover glass were used on the third day of growth. The cells were washed twice with medium 199 and were then incubated in Hp solution in the dark at room temperature. Just before measurement, the sample was placed into a specially designed experimental chamber filled with medium 199 and was then examined under the laser microscope. 2.4. Choice of excitation parameters Prior to the spectral and kinetics studies of Hp fluorescence in cells, the laser radiation intensity was adjusted such, that it did not result in cell damage. Plasma membrane integrity was tested using trypan blue. By variation of the laser radiation intensity and irradiation time, it was shown that irradiation of a cell site of 1 pm in diameter with 0.01 PW (A-532 nm) for 10 min caused no membrane damage; no photobleaching of Hp was observed.
3. Instrumentation The experimental set-up of the laser picosecond microspectrofluorometer is shown in Fig. 1. A continuous wave (CW) pumped, acousto-optically & = switched and mode-locked Nd:YAG laser was used as the master generator. The laser output at 1.064 pm consists of a sequence of trains (repetition rate, 5 kHz), each containing 25-30 pulses of 80 ps duration. For sample excitation, a single pulse was selected from the train using an electro-optical modulator (EOM). Cell fluorescence was obtained using pulses of the second harmonic (SH) of the Nd:YAG laser of the second harmonic of a picosecond continuum, generated by stimulated Raman scattering in a single optical fibre (OF). A
242
YAG:h'd
Fig. 1. Laser picosecond microspectrofluorometer: SH, second harmonic crystal; EOM, electrooptical modulator; OF, optical fibre; 0, objective; M, monochromator; PMT, photomultiplier tube; PD, photodiode; Ll, L2, L3, lenses; SPC, single-photon counting system.
high quality objective (“Reichert”, 160 X , NA= 1.25) (0) was used to focus the exciting laser beam onto the sample, resulting in a beam width smaller than 1 pm. It should be noted that, due to the high spatial resolution and high Hp concentration in the lipid bilayer, there is practically no contribution of Hp fluorescence from the aqueous medium, in contrast with the results obtained with the cell suspensions [9]. The fluorescence signal from the selected subcellular structure or liposome was focused by a lens (L3) onto the input slit of a monochromator (M) (spectral resolution, l-2 nm) and detected by a fast photomultiplier tube (PMT). The detection system based on the time-correlated single-photon counting technique, was computer controlled. The output from the time-toamplitude converter, which has two time gates (O-30 ns and O-300 ns), was fed directly to a microcomputer. The full width at half-maximum of the detection system response function (Pig. 5, broken line, see later) was about 800 ps. The specially designed software enabled us to obtain either the fluorescence decay kinetics at the selected emission wavelengths, or the time-resolved fluorescence spectra detected at various time delays after the excitation pulse. A deconvolution procedure was used for data processing [ 131. The experimental fluorescence decay curves at different wavelengths were fitted by a convolution of the system response function with a polyexponential fluorescence intensity decay function: F(t) = C”ai exp( - t/~~), n = 1,2,3. However, in our case (limited data collection time), the contribution of the middle component was negligible in comparison with the other two; consideration of the middle component did not produce a better fitting. The quality of approximation was estimated by the x2 value, the residual autocorrelation function and visual fitting were considered to be satisfactory for x2 < 1.5 and when an increase in n produced no more than a 1% decrease in x2.
243
4. Results and discussion The spectra and decay kinetics of Hp fluorescence were measured in medium 199 (10v4 M). The fluorescence spectra were characterized by two bands at 615 and 675 nm. The fluorescence decay at 615 and 675 nm was found to be monoexponential with a lifetime 7 of 13 ns (corresponding to Hp monomer fluorescence [ 181). In order to study the Hp fluorescence in a lipid environment, the spectral and kinetic characteristics of Hp in PC liposomes (model of a membrane lipid bilayer) were measured. Relative to the fluorescence spectrum of Hp in aqueous solution, the fluorescence bands in the PC liposomes were shifted by 15 nm to the red. The dependence of the fluorescence spectrum of Hp in PC liposomes on the incubation time is presented in Pig. 2. The fluorescence spectrum displays two bands at 630 and 690 run. With an increase in incubation time from 1 to 4 h, a new fluorescence band at 665 nm appears. The ratio of the fluorescence intensity of the band at 665 nm to that of the band at 630 nm increases from 0.3 at 2 h incubation to 0.4 at 4 h to 0.5 at 6 h. A decrease in Hp concentration to 10T5 M and 5X lo-’ M results in a decrease in the contribution of the new band: the 1665/1830ratio is only 0.2 for 6 h incubation for 5 X lop6 M Hp. The Hp fluorescence decay kinetics from PC Iiposomes were recorded at A= 635 nm and A= 665 run for two Hp concentrations (C= 10m4 M and C= 10e5 M) and two incubation times (1 h and 6 h). These decay kinetics were found to be biexponential with F(t) = a, exp( - t /TV) + a2 exp( - t/T2). The data are presented in Table 1. At h= 665 run, the contribution of the fast component increases with an increase in Hp concentration and incubation time, whereas at h= 635 nm the pattern is quite different. In particular, a decrease in the contribution of the fast component is observed at increasing concentrations. The fluorescence spectra and decay kinetics of Hp were measured at the nuclear and cytoplasmic membrane sites of cultured SPEV cells. The
600
655
Fig. 2. Dependence ([HP]= lo+’ M).
710
X,nm
of the Hp fluorescence
spectrum
in PC liposomes
on incubation
time
244 TABLE 1 Fitting of the HP fluorescence decay curve for the liposomes Parameter
n 6-1
Wavelength (nm)
C=10-4 lh
6h
lh
6h
635
0.5 23.8 0.17
0.6 19.9 0.16
0.20 15.4 0.07
0.3 15.0 0.05
665
0.4 15.2 0.1
0.4 10.1 0.06
0.4 16.0 0.33
0.3 14.3 0.1
72 (ns) a2la,
71 (ns) 72 a2fal
(ns>
M
C=10-6
M
Fig. 3. Dependence of the Hp fluorescence spectrum from the cytoplasmic membrane of SPEV cells on incubation time ([HP] = 10m6 M).
dependence of the Hp fluorescence spectrum from the cytoplasmic membrane on the incubation time is shown in Fig. 3. It is quite similar to that observed in liposomes. For incubation times longer than 2 h, the band at 665 run is detected, its relative contribution increasing with an increase in incubation time. The Hp fluorescence spectrum at the site of the nuclear membrane (C= 10m4 M) for an incubation time of 4 h is shown in Fig. 4. It can be seen that this spectrum is identical with that observed in liposomes after prolonged incubation. The lifetimes of the three fluorescence bands were measured. The data obtained from the fitting of the fluorescence decay curve for the nuclear membrane are presented in Table 2. The fluorescence decay is monoexponential at 690 nm with a lifetime of 11 ns and biexponential at 630 and 665 run
245
01
600
650
700 A,nm
Fig. 4. Hp fluorescence spectrum at the site of the nuclear incubation time of 4 h ([HP] = 10m4 M).
TABLE
2
Fitting
of the Hp fluorescence
Parameter
decay
(ns)
72
cm>
al
Ia2
or the nuclear
membrane
of SPEV cells for an
of a SPEV cell
J+l (nm) 635
71
curve
membrane
665
0.5
0.6
12.2 2.5
13.4 5
690 11.0 -
with lifetimes of r1= 0.5 ns, 7-a= 12.2 ns (aI /czz= 5) and T, = 0.6 ns, T~= 13.4 ns (a, /uZ = 1.3) respectively. The amplitude of the fast component is larger at 665 nm. The apparatus function, fluorescence decay kinetics and fluorescence decay curve for Hp fluorescence at the nuclear membrane of the cell at 630 nm are presented in Fig. 5 together with the residual autocorrelation function. The similar spectra and decay kinetics of Hp fluorescence observed in liposomes and the cytoplasmic and nuclear membranes of the cell indicate that the Hp localized in the lipid bilayer of the membrane makes an overwhelming contribution to the fluorescence signal from the membrane. The new band at 665-670 nm with a short lifetime (observed after prolonged incubation and at high Hp concentrations (sufficiently high porphyrin to lipid molar ratio)) may be due to Hp aggregates or to the binding of Hp to sites inside or at the surface of the lipid bilayer. The possible formation of Hp photoproducts must also be considered. To clarify this, we studied the dependence of the Hp fluorescence intensity at 665-670 nm on the irradiation dose. No change in I665 was observed for
246
Ol
II
I
-\
0
I
10
20 T,ns
Fig.5. Hp fluorescence decay kinetics recorded at the site of the nuclear membrane ([HP] = 10-6 M): 9, experimental data; ---, apparatus function; -, autocorrelation function is given in the inset.
fluorescence decay curve. The residual
irradiation doses in the range lo4 to 6 X lo6 J rnm2.No Hp photobleaching was detected during our measurements. Other data in favour of the formation of Hp aggregates in membranes are as follows: (i) the significant decrease in the photodynamic damage of the cell plasma membrane on prolonged incubation [ 81, which can be explained by the low quantum yield of singlet oxygen generation by Hp aggregates [ 21; (ii) the decrease in Hp fluorescence intensity at the cell plasma membrane after 4-8 h incubation time [ 191, which can be explained by the very low fluorescence yield of aggregates; (iii) the distinct fluorescence of Hp aggregates in the cell plasma membrane and membranes of intracellular structures [20]; (iv) the fluorescence of DHE in liposomes which suggests self-association in a lipid environment [21]. The possible formation of Hp aggregates in the lipid bilayer of membranes, leading to a decrease in the extent of photodynamic damage, must be considered when choosing the PDT regimen.
Acknowledgments The authors are grateful to Professor S. A. Akhmanov for support of this work, Dr. L. L. Litinskaya and Dr. M. I. Leikina for supplying the cell culture and for fruitful discussions and I. Deruzhenko for preparation of the samples.
247
References
1 T. J. Dougherty, J. E. Kaufman, A. Goldfarb, K. R. Welshaupt, D. Boyle and A. Mittleman, Photoradiation therapy for treatment of malignant tumors, Cancer Res., 38 (1986) 2628-2635. 2 D. Kessel, Hp and HpD: photophysics, photochemistry and phototherapy, Pholocha.
Photobiol., 39 (1984) 851-859. 3 T. J. Dougherty, Photosensitlzers: Photobiol., 45 (1987) 879-889.
therapy and detection of malignant tumors, Photo&em.
4 T. J. Dougherty, Studies on the structure of porphyrins contained in Photofrln II, Photo&em. Photobiol., 46 (1987) 569-573. 5 D. Kessel, HpD-terminology, Photo&em. Photobiol., 50 (1989) in Guest editorial. 6 R. Margalit and S. Cohen, Studies of hematoporphyrln derivative equilibria in heterogeneous systems. Porphyrin-liposome binding and porphyrin aqueous dlmerization, Biochim; Biophys. Acta, 736 (1983) 163-170. 7 M. Yamashita, T. Tomono, S. Kobayashi, K. Torlzuka and K. Aizawa, Picosecond fluorescence spectroscopy on incorporation process of HpD into malignant tumor cells in vitro,
Photo&em. Photobiol., 47 (1988) 189-192. 8 D. Kessel, Sites of photosensitization by derivatives 9
10
11
12 13 14
15
16 17
18
19
20
of Hp, Photo&em. Photobiol., 44 (1986) 489-493. B. Ehrenberg, Z. Mallk and Y. Nitzan, Fluorescence spectral changes of hematoporphyrin derivative upon binding to lipid vesicles, Staphylococcus aureus and Escherichia coli cells, Photochem. Photobiol., 41 (1985) 429-437. B. Roedcr and H. Wabnitz, Time-resolved fluorescence spectroscopy of hematoporphyrin, mesoporphyrln, pheophorbide a and chlorine e a ln ethanol and aqueous solution, J. Photo&em. Photobiol. B: Biol., I (1987) 103-113. A. G. Vardanyan, V. F. Kamalov, A. F. Mlronov, I. A. Struganova and B. N. Toleutaev, Picosecond time-resolved anisotropy decay kinetics of hematoporphyrin fluorescence in solution Opt. Spektrosk., 66 (1989) 346-348 (in Russian). M. Yamashita, M. Nomura, S. Kobayashi, T. Sato and K. Aizawa, Picosecond time-resolved fluorescence spectroscopy of HpD, J. Quantum Electron., 20 (1984) 1363-1367. F. Doccio, R. Ramponi, C. A. Sacchi, G. Bottiroli and I. Freitas, Fluorescence studies of biological molecules by laser h-radiation, Lasers Surg. Med., 2 (1982) 21-28. H. Schneckenburger, F. Pauker, E. Unsold and D. Jocham, Intracellular distribution and retention of the fluorescent components of Photofrln II, Photobiochem, Photobiophys., 10 (1985) 61-67. N. Miyoshl, H. Hlsazuml, 0. Ueki, K. Nakajima and M. Fukuda, Photodynamic inactivation of cultivated human bladder cancer cells (KK-47) sensitized by HpD and argon-dye laser light, Photobiochem. Photobiophys., 9 (1985) 241-252. R. Cubeddu, A. Andreoni, C. N. Knox and T. G. Truscott, Fluorescence lifetimes of angular furocoumarlns, Photo&em. Photobiol., 46 (1987) 169-173. Z. D. Sultanova, K. S. Kulikova and E. V. Gurvich, Comparative morphological, cultivating and vlrusological characteristics of cell cultures, Vopr. Virusol., 3 (1968) 291-298 (in Russian). A. Andreoni, R. Cubeddu, S. De Sllvestry, G. Jori, P. Laporta and E. Reddi, Time-resolved fluorescence studies of hematoporphyrin ln diierent solvent systems, 2. Naturforsch., Teil C, 38 (1983) 83-89. Z. A. Kchurshilova, M. A. Lelkiia, L. L. Lltlnskaya and E. B. Chemyaeva, Microfluorimetric study of porphyrin type photosensitizer accumulation in normal and malignant cultured cells, Vestn. Mosk. Univ., Fiz Asronomiya, 27 (1986) (in Russian). H. K. Scidlitz, H. Schneckenburger and K. Stettmaier, Time-resolved polarization measurements of porphyrin fluorescence ln solution and ln single cells, J. Photo&em. Photobiol.
B: Biol., 5 (1990)
391-400.
248 21 S. E. Kendrick, J. Persky, M. L. E&l-rawer, A. L. Imbembo, J. D. Fleischmann and K. A.
Koehler, Dihematoporphyrin-ether-phospholipid interactions. The roles of surface charge and lateral phase separations. Photo&em. Photobiol., 46 (1987) 669-673. 22 L. Eidus, Yu. N. Korystov, L. L. Litlnskaya, M. A. Morosov and A. M. Vexler, The study of dye accumulation in granules on vital dyeing of Chinese hamster fibroblasts in culture, Stud. Biophys., 62 (1977) 85-92. 23 E. B. Chernyaeva, E. A. Shleina, Z. A. Kchurshilova, L. L. Litinskaya and M. I. Leikina, Intracellular dynamics of hematoporphyrln: action of light, Stud. Biophys., I25 (1988) 203-210. 24 D. V. O’Connor and D. Phillips, Time-Correlated
London, 1984.
Single Photon
Counting,
Academic Press.
