- Email: [email protected]
Delivery of the tumour photosensitizer zinc(II)-phthalocyanine to serum proteins by different liposomes: studies in vitro and in vivo
Cancer Letters, 49 (1990) 59-65 Elsevier Scientific Publishers Ireland
59 Ltd.
Delivery of the tumour photosensitizer zinc(II)-phthalocyanine to serum proteins by different liposomes: studies in vitro and in vivo F. Ginevra”,
S. Biffantib, A. Pagnanb,
R. Bioloa, E. Reddi” and G. Jori”
‘Department of Biology and bClinica Medica I, University of Padoua (ItalyJ (Received 17 April 1989) (Revision received 31 July 1989) (Accepted 21 August 1989)
Summary
Introduction
Zn-phthalocyanine (Zn-Pc) incorporated into liposomes of different phospholipids has been incubated in vitro with human serum and administered i.v. to rabbits. In both cases, chromatographic and density gradient ultracenttijugation studies indicate that Zn-Pc is almost exclusioely bound by the 3 major hpoprotein components of the plasma (VLDL, LDL and HDL). The amounts of Zn-Pc recovered from the different lipoprotein jractions rejlect their relative concentration in the serum. The presence of 20 5%moles of cholesterol in liposomes of dipalmitoyl phosphatidylcholine (DPPC) optimizes the release of Zn-Pc to LDL. This fact is important for enhancing the selectiuity of drug delioery to tumors since LDL display a preferential interaction with neoplastic cells.
There appear to be two main modes of photosensitizer localization in tumour tissues, depending on the nature of the protein of the dye; while albumin mainly delivers the bound dyes to the vascular stroma, some lipoproteins (LDL) mainly release the dye to malignant cells [l]. The LDL pathway is especially for important hydrophobic photosensitizers, which are often preferentially or selectively transported by lipoproteins in the serum [2,3]; this fact enhances the specificity of tumour targetting by the photosensitizer since one class of lipoproteins, the low-density lipoproteins (LDL) , develops a very active interaction with neoplastic cells through receptor-mediated endocytosis [4,5]. As a consequence, it appears important to devise some delivery methods for photosensitizers in vivo which favour their association with LDL. Recently, we showed [6] that Zn-Pc, a promising new photosensitizer for tumours, can be incorporated in a stable form into the phospholipid bilayer of small unilamellar vesides of dipalmitoyl-phosphatidylcholine and, once injected into the bloodstream, is
Keywords: phthalocyanine; lipoproteins; liposomes; photodynamic therapy; tumours. Correspondence to: Loredan 10, l-35131
G. Jori, Padova,
Dipartimento Italy.
di Biologia,
via
0304-3835/90/$03.50 0 1990 Elsevier Scientific Publishers Published and Printed in Ireland
Ireland Ltd.
60
quantitatively transferred to the In the present investigation,
Materials and methods Sera
For in vitro experiments blood samples were given by healthy volunteers (male and female). Erythrocytes were removed by centrifugation at 3000 rev./min for 15 min at room temperature. The serum (9 ml) was then incubated at 20 f l°C with 1 ml of a Iiposomal dispersion in 0.1 M phosphate buffer at pH 7.4. After 30 min of gentle magnetic stirring in the dark, the samples were subjected to chromatographic analysis and/or ultracentrifuge separation. For in vivo experiments New Zealand male rabbits were injected i.v. with 0.1 mg of Iiposome-bound Zn-Pc per kg body weight. At 2-4 h and 48 h after injection, blood samples (25 ml) were carefully taken from the ear vein, centrifuged to remove the erythrocytes and used for chromatography and ultracentrifuge studies. Zn-PC-containing liposomes Stock Zn-Pc solutions (l- 1.5 mM) were prepared in pyridine and stored at 4OC. The into Iiposomal of Zn-Pc incorporation vesicles was performed according to an injection procedure [7]. Typically, 0.53 ml of the stock Zn-Pc solution was added to 1.47 ml of the phospholipid solution in absolute ethanol; 0.75 ml of this mixture was injected into 10 ml of pH 7.4 phosphate-buffered (0.1 M KH,PO,-Na,HPO,) injection was solution. The aqueous performed at a speed of 1 $/s with magnetic stirring and at a temperature higher than the phase transition temperature of the The details of the given phospholipid. experimental protocols adopted for each
Table 1. Experimental conditions used in the preparation of various kinds of liposomes. Phospholipid Phospholipid concentration in ethanol (mM)
Injection temperature foe)
Liposome diameter (nm)
DPPC DPPC DPPC-Ch DMPC DSPC
55 55 55 35 65
52.4 (SW 99.2 (LUV) n.d. 30.0 63.7
9.60 38.90 7.68-1.19 2.88 9.60
n.d. = not determined; DMPC = dimiristoyl-phosphatidylcholine; DSPC = distearoyl-phosphatidylcholine; Ch = cholesterol; SUV, LUV = small and large unilamellar vesicles.
case are given in Table 1. At the end, the liposome aqueous dispersion was dialyzed at room temperature for 24 h against 250 ml of the phosphate buffer with one change after the first hour. Chromatographic analyses of sera In a typical experiment, 2 ml of Zn-Pcdoped human serum were applied to a column (1.7 x 140 cm) of Sephacryl S-300, which had been previously equilibrated with 0.1 M phosphate buffer at pH 7.4. The column was eIuted with the same solvent at room temperature and at a flow rate of 26 ml/h. In general, 2.5 ml fractions were collected. For each fraction the protein absorbance was monitored at 280 nm while the of Zn-Pc was fluorescence emission measured at 675 nm upon excitation with measure600-nm light. All fluorescence ments were carried out with a Perkin Elmer MPF-4 spectrophotofluorimeter equipped with a red-sensitive phototube. Care was taken to keep the absorbance of the solutions lower than 0.05 at both the excitation and observation wavelength in order to minimize optical artifacts. Discontinuous density gradient ultracentri~ugation of sera All ultracentrifugation studies were per-
61
formed at 39,000 rev./min with a Beckman apparatus equipped with a rotor type 40.3. The density gradient was obtained by controlled addition of aqueous KBr to the [8]. Four density regions were serum obtained and the proteins isolated from the various regions were classified as follows: very low density (VLDL, d < 1.006); low density (LDL, 1.006 < d < 1.063); high density (HDL, 1.063 < d < 1.21) lipoproteins and a fraction at d > 1.21 containing the other serum proteins (bottom). The purity of each fraction was assayed by agarose gel-electrophoresis [9]. For each fraction, the content of proteins, triglycerides, phospholipid and cholesterol was estimated in order to calculate the total lipoprotein mass (holoprotein). The amount of Zn-Pc bound with each protein fraction was evaluated by spectrophotofluorimetric analysis = 600 emission range: nm) after of the with 2% SDS to absorbance lower 0.05 at 600 and nm. Before tion with the LDL, and bottom dialyzed overnight 250 ml phosphate buffer order to the KBrprecipitation of surfactant. The intensity was into Znconcentration by with a bration curve with known of Zn-Pc. Results In vitro studies A typical chromatogram obtained by column fractionation of human serum incubated with liposome-bound Zn-Pc is shown in Fig. 1. Clearly, 3 main protein peaks are observed. The most rapidly eluted peak, corresponding to high molecular weight proteins, is essentially constituted by lipoproteins, as shown by phospholipid analysis [2]. The second and third peaks mainly contain globulins and albumins, respectively [2]. Control studies show that under our experimental conditions liposomes are not eluted
110
130
150
Effluent
170
190
210
(ml)
Fig. 1. Elution profile of human serum which has been chromatographed through a column of Sephacryl S-300 after a 30-min incubation with liposome-incorporated Zn-Pc. The protein concentration was followed by the absorbance at 280 nm (O0). The concentration of Zn-Pc added in DMPC ( 0 - - - - 0) and DPPC (SW) ( l- - - -‘) liposomes was foflowed by the fluorescence emission at 680 nm.
being retained at the top of the Sephacryl S-300 column. This indicates that the Zn-Pc originally incorporated into DPPC liposomes is quantitatively and selectively transferred to serum lipoproteins. The long tailing edge of the Zn-Pc peak (Fig. 1) reflects the contribution of some light scattering to the phthalocyanine fluorescence spectrum determined in the single fractions collect:: from the column. Moreover, a limited Zn-Pc binding to other serum proteins can also occur (Table 2). Essentially identical chromatograms were obtained with all other types of liposomes studied by us; only in the case of DMPC liposomes, Zn-Pc becomes associated with several protein fractions (Fig. 1). The distribution of Zn-Pc among the various components of the lipoprotein family, and the remaining serum proteins (bottom) after delivery of the drug by the different types of liposomes is reported in Tables 2 and 3 and referred to the content of holoand apo-protein, respectively. Only for DMPC liposomes, a significant (22%) frac-
62
Recovery of Zn-Pc (ng/mg of holoprotein) from lipoprotein fractions isolated by ultracentrifugation after Table 2. 30 min incubation of human serum with the drug incorporated into different liposomes. The values in brackets represent the percentage recovery of Zn-Pc from the various protein fractions. Liposomes
VLDL
DPPC (LUV) DPPC (WV) DPPC-Ch DMPC DSPC
15.5 14.3 15.1 16.8 18.3
LDL (2.8) (2.1) (3.3) (3.5) (3.9)
16.1 23.1 67.0 32.1 7.8
(12.9) (12.7) (31.1) (16.9) (4.6)
tion of the total recovered Zn-Pc is present in the bottom, in agreement with the results of chromatographic analyses. In all cases, the largest amount of Zn-Pc @O-90% of the totally recovered phthalocyanine) is associated with HDL, while remarkably lower amounts are bound with LDL (approx. 10-158) and VLDL (3-4%). Such a distribution reflects the relative abundance of the different lipoprotein classes in human serum [lo]. A notable exception is represented by the results obtained in the presence of DPPC-Ch liposomes, where the aliquot of LDL-bound Zn-Pc is as high as 31%. In uiuo studies
Four kinds of liposomes (DMPC, DPPCSW, DPPC-Ch and DSPC) have been tested as delivery systems for Zn-Pc in vivo.
Recovery of Zn-Pc (ng/mg of apoprotein) Table 3. from lipoprotein fractions isolated by ultracentrifugation after 30 min incubation of human serum with the drug incorporated into different liposomes. Liposomes
VLDL
LDL
HDL
Other proteins
DPPC (LUV) DPPC (SW) DPPC-Ch DMPC DSPC
52.5 45.6 67.3 71.1 79.3
54.3 67.0 223.7 96.1 25.9
129.8 194.9 189.0 114.6 174.7
0.07 0.16 0.18 1.38 0.08
HDL
Other proteins
732 (82.9) 110.7 (83.2) 107.0 (62.6) 63.6 (57.3) 94.1 (90.0)
0.07 0.15 0.17 1.38 0.08
(1.4) (2.0) (3.0) (22.3) (1.5)
Once again, chromatographic analyses of rabbit sera taken at 2-4 h and 48 h after i.v.-injection of the drug indicate that Zn-Pc is exclusively carried by lipoproteins, with the exception of DMPC liposomes which release Zn-Pc to a variety of serum proteins. Tables 4 and 5 show the amount of Zn-Pc associated with the protein fractions isolated by density gradient ultracentrifugation of the same serum samples as analyzed by column chromatography. Apparently, HDL still accumulate the greatest amount of phthalocyanine, while approximately 1520% of the totally recovered Zn-Pc is bound to LDL. The latter percentage raises to 27% in the case of DPPC-Ch liposomes. In all cases, VLDL accumulate and retain a much larger percentage of the recovered Zn-Pc as compared with in vitro results. VLDL are remarkably more abundant in rabbit than in human serum [ 111. Discussion Our results clearly demonstrate that the distribution of a hydrophobic drug such as Zn-Pc among serum proteins can be modulated to some extent by the nature of the lip;>somal vesicle used as a drug-delivery system. A major role is probably played by the physical state of the phospholipid used for the preparation of the liposome. In the case of DMPC, which is in a quasi-liquid state at body temperature, Zn-Pc is released to several serum proteins. Possibly,
63 Distribution of Zn-Pc among the lipoprotein fractions isolated by ultracentrifugation from rabbit serum at Table 4. 2 or 4 h and 48 h after intravenous injection of Zn-Pc incorporated into different liposomes. Values expressed as ng/mg of holoprotein. The values in brackets represent the percentage recovery of Zn-Pc from the various protein fractions. Liposomes
(h)
VLDL
LDL
DPPC (WV)
2 48 2 48 2 48 4 48
21.1 15.9 19.6 15.0 18.4 14.2 32.9 21.9
22.9 11.4 19.8 8.6 21.1 9.8 24.3 26.6
DMPC DSPC DPPC-Ch
(10.7) (35.7) (22.6) (62.2) (21.3) (40.5) (39.7) (20.1)
the less tight packing and enhanced motility of the hydrocarbon chains favours both the interaction of the DMPC vesicle with the protein matrices and the leakage of Zn-Pc from the bilayer into the external medium [12]. On the other hand, all the liposomes whose phase transition temperature is above 37OC, hence exist in a quasi-solid state in the serum, specifically release Zn-Pc to lipoproteins. In vitro studies suggest that, in the presence of serum, such liposomes have a lifetime shorter than 30 min. In fact chromatographic analysis showed that after this period of time all Zn-Pc had been transferred from liposomes to lipoproteins. It is likely that the phospholipid vesicle efficiently
Other proteins
HDL (14.3) (21.5) (19.4) (15.1) (16.8) (17.1) (27.2) (45.8)
27.6 11.6 14.0 6.9 23.2 9.5 22.7 15.9
(66.4) (36.4) (44.0) (15.6) (56.5) (34.6) (24.6) (22.2)
0.12 0.08 0.15 0.08 0.08 0.09 0.19 0.17
(8.6) (6.4) (14.0) (6.7) (5.4) (7.8) (8.5) (11.9)
interacts with the lipid moiety of lipoproteins, independently of their chemical composition. This would explain the correlation between the amount of Zn-Pc recovered from the various lipoproteins and their relative abundance in the serum. The selectivity of the interaction between lipoproteins and solid-type liposomes is not affected by parameters as the length of the such hydrocarbon chain of the phospholipid (DPPC vs. DSPC) or the size of the vesicle (SW vs. LUV). Actually, for all these kinds of liposomes, rather similar patterns of ZnPC binding to the various lipoproteins have been found both in vitro and in vivo. The larger amount of Zn-Pc recovered from
Table 5. Distribution of Zn-Pc among the lipoprotein fractions isolated by uitracentrifugation from rabbit serum at 2 h or 4 h and 48 h after intravenous injection of Zn-Pc incorporated into different liposomes. Values expressed as ng/mg of apoprotein . Proteins
VLDL LDL HDL Other proteins
Liposomes DPPC (WV)
DMPC
2h
48 h
2h
48 h
2h
48 h
4h
48 h
53.6 65.8 47.7 0.12
70.6 37.0 20.4 0.08
78.6 61.2 25.5 0.15
90.0 28.4 12.2 0.08
68.0 65.6 50.5 0.08
60.4 32.0 16.5 0.09
175.4 85.7 37.1 0.19
82.9 102.7 24.3 0.17
DSPC
DPPC-Ch
64
VLDL in vivo almost certainly reflects the larger concentration of VLDL in rabbit as compared with human serum. This finding further supports the hypothesis that the interaction of solid liposomes with the different lipoproteins is controlled by the relative abundance of the latter in the serum. Actually, a comparative analysis of the data reported in Tables 2 and 3 indicates that the in vitro distribution of Zn-Pc among the different lipoprotein fractions, as obtained with the various liposomes, is very similar whether it is referred to the holoproteins or the apoproteins. Previous in vitro studies [13] showed that both those constituents of lipoproteins possess a discrete number of binding sites for porphyrin-type drugs. The in vivo situation is more complex. The data on Zn-Pc recovery shown in Tables 4 and 5 indicate that the phthalocyanine is eliminated at very similar rates from the holo- and the apo-lipoproteins suggestpools that both phthalocyanine ing contribute to the accumulation of the dye by tissues. Moreover, the apoprotein moiety of VLDL appears to retain larger amounts of the dye as compared with the other apolipoproteins. This finding is probably related with the different metabolic fate of the single lipoproteins and cannot be interpreted on the basis of our data owing to the complex nature of the metabolic cycles of these proteins [ 131. Only the presence of cholesterol in the phospholipid bilayer causes an alteration of the aforesaid pattern of Zn-Pc distribution, favouring the release of the phthalocyanine to LDL and VLDL. Moreover, LDL exhibit a prolonged retention of Zn-Pc both in vitro and in vivo. This observation may indicate that the presence of cholesterol determines a different mode of Zn-Pc localization in the LDL. The importance of cholesterol for the stability [ 141 and pharmacokinetic behaviour [15] of liposomes in vivo has been observed also by other investigators. The factors responsible for the preferential interaction of
cholesterol-containing liposomes with LDL are not fully understood. However, this fact is of utmost importance for the application of PDT in the treatment of tumours. Preliminary findings from our laboratory show that the administration of Zn-Pc to tumour-bearing mice via DPPC-Ch liposomes brings about a more selective targetting of the tumour by the photosensitizing drug as well as an enhanced necrotic response of the tumour to PDT especially for irradiations performed at short times (612 h) after systemic injection. References 1
Kessel, D., Thompson, P., Saatio, K. and Nantwi, K.D. (1987) Tumor localization and photosensitization by sulfonated derivatives of tetraphenylporphine. Photochem. Photobiol., 45, 787-790.
2
Barel, A., Jori, G., Perin, A., Romandini, P., Pagnan, A. and Biffanti, S. (1986) Role of high-, low- and very low-density lipoproteins in the transport and tumordelivery of hematoporphyrin in vivo. Cancer Lett., 32, 145-150. Kessel, D. (1986) Porphyrin-lipoprotein association as a factor in porphyrin localization. Cancer Lett., 33, 183188. Goldstein, J.L., Anderson, R.G.W. and Brown, M.S. (1979) Coated pits, coated vesicles and receptormediated endocytosis. Nature, 279, 679-685. Brown, M.S., Kovanen, P.T. and Goldstein, J.L. (1980) Evolution of the LDL receptor concept from cultured animal cells to intact animals. Ann. N.Y. Acad. Sci., 348, 48-68. Reddi, E., Lo Castro, G., Biolo, R. and Jori, G.
3
4
5
6
(1987) Pharmacokinetic anine in tumour-bearing 7
8
9
10 11
studies with zinc(Il)-phthalocymice. Br. J. Cancer, 56,
597-600. Kremer, J.M.H., van der Esker, W .M. J., Pathomamanoharan, C. and Wiersema, P.H. (1977) Vesicles of variable diameter prepared by a modified injection method. Biochemistry, 16, 3932-3935. Havel, R.J., Eder, r: A. and Bragdon, J.H. (1955) Distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J. Clin. Invest., 34, 1345- 1353. Noble, R.P. (1968) Electrophoretic separation of plasma lipoproteins in agarose gel. J. Lipid. Res., 9, 693700. Giesow, M. (1980) Pathways of endocytosis. Nature, 288, 434-436. Rudel, L.L., Marzetta, C.A. and Johnson, F.L. (1986) Separation and analysis of lipoproteins by gel filtration.
65
12
13
14
In: Methods in Enzymology, pp. 45-57. Editors: J.J. Albens and J.P. Segrest. Academic Press, London. Gupta, C.M. and Dhawan, S. (1981) Modification of phospholipid structure results in greater stability of liposomes in serum. B&him. Biophys. Acta, 648, 192198. Beltramini, M., Firey, P.A., Ricchelli, F., Rodgers, M.A.J. and Jori, G. (1987) Steady-state and timeresolved spectroscopic studies on the hematoporphyrinlipoprotein complex. Biochemistry, 26, 6852-6858. Eisenberg, S. (1986) Plasma lipoprotein conversion. In:
15
16
Methods in Enzymology, pp. 347-366. Editors: J.J. Albens and J.P. Segrest. Academic Press, London. Kirby, C., Clarke, J. and Gregoriadis, G. (1980) Effect of the cholesterol content of small unilamellar liposomes on their stability in vitro and in vivo. Biochem. J., 186, 591-598. Patel, H.M., Tuzel, N.S. and Ryman, B.E. (1983) Inhibitory effect of cholesterol on the uptake by liver and spleen. Biochim. Biophys. 142-151.
of liposomes Acta, 761,
59 Ltd.
Delivery of the tumour photosensitizer zinc(II)-phthalocyanine to serum proteins by different liposomes: studies in vitro and in vivo F. Ginevra”,
S. Biffantib, A. Pagnanb,
R. Bioloa, E. Reddi” and G. Jori”
‘Department of Biology and bClinica Medica I, University of Padoua (ItalyJ (Received 17 April 1989) (Revision received 31 July 1989) (Accepted 21 August 1989)
Summary
Introduction
Zn-phthalocyanine (Zn-Pc) incorporated into liposomes of different phospholipids has been incubated in vitro with human serum and administered i.v. to rabbits. In both cases, chromatographic and density gradient ultracenttijugation studies indicate that Zn-Pc is almost exclusioely bound by the 3 major hpoprotein components of the plasma (VLDL, LDL and HDL). The amounts of Zn-Pc recovered from the different lipoprotein jractions rejlect their relative concentration in the serum. The presence of 20 5%moles of cholesterol in liposomes of dipalmitoyl phosphatidylcholine (DPPC) optimizes the release of Zn-Pc to LDL. This fact is important for enhancing the selectiuity of drug delioery to tumors since LDL display a preferential interaction with neoplastic cells.
There appear to be two main modes of photosensitizer localization in tumour tissues, depending on the nature of the protein of the dye; while albumin mainly delivers the bound dyes to the vascular stroma, some lipoproteins (LDL) mainly release the dye to malignant cells [l]. The LDL pathway is especially for important hydrophobic photosensitizers, which are often preferentially or selectively transported by lipoproteins in the serum [2,3]; this fact enhances the specificity of tumour targetting by the photosensitizer since one class of lipoproteins, the low-density lipoproteins (LDL) , develops a very active interaction with neoplastic cells through receptor-mediated endocytosis [4,5]. As a consequence, it appears important to devise some delivery methods for photosensitizers in vivo which favour their association with LDL. Recently, we showed [6] that Zn-Pc, a promising new photosensitizer for tumours, can be incorporated in a stable form into the phospholipid bilayer of small unilamellar vesides of dipalmitoyl-phosphatidylcholine and, once injected into the bloodstream, is
Keywords: phthalocyanine; lipoproteins; liposomes; photodynamic therapy; tumours. Correspondence to: Loredan 10, l-35131
G. Jori, Padova,
Dipartimento Italy.
di Biologia,
via
0304-3835/90/$03.50 0 1990 Elsevier Scientific Publishers Published and Printed in Ireland
Ireland Ltd.
60
quantitatively transferred to the In the present investigation,
Materials and methods Sera
For in vitro experiments blood samples were given by healthy volunteers (male and female). Erythrocytes were removed by centrifugation at 3000 rev./min for 15 min at room temperature. The serum (9 ml) was then incubated at 20 f l°C with 1 ml of a Iiposomal dispersion in 0.1 M phosphate buffer at pH 7.4. After 30 min of gentle magnetic stirring in the dark, the samples were subjected to chromatographic analysis and/or ultracentrifuge separation. For in vivo experiments New Zealand male rabbits were injected i.v. with 0.1 mg of Iiposome-bound Zn-Pc per kg body weight. At 2-4 h and 48 h after injection, blood samples (25 ml) were carefully taken from the ear vein, centrifuged to remove the erythrocytes and used for chromatography and ultracentrifuge studies. Zn-PC-containing liposomes Stock Zn-Pc solutions (l- 1.5 mM) were prepared in pyridine and stored at 4OC. The into Iiposomal of Zn-Pc incorporation vesicles was performed according to an injection procedure [7]. Typically, 0.53 ml of the stock Zn-Pc solution was added to 1.47 ml of the phospholipid solution in absolute ethanol; 0.75 ml of this mixture was injected into 10 ml of pH 7.4 phosphate-buffered (0.1 M KH,PO,-Na,HPO,) injection was solution. The aqueous performed at a speed of 1 $/s with magnetic stirring and at a temperature higher than the phase transition temperature of the The details of the given phospholipid. experimental protocols adopted for each
Table 1. Experimental conditions used in the preparation of various kinds of liposomes. Phospholipid Phospholipid concentration in ethanol (mM)
Injection temperature foe)
Liposome diameter (nm)
DPPC DPPC DPPC-Ch DMPC DSPC
55 55 55 35 65
52.4 (SW 99.2 (LUV) n.d. 30.0 63.7
9.60 38.90 7.68-1.19 2.88 9.60
n.d. = not determined; DMPC = dimiristoyl-phosphatidylcholine; DSPC = distearoyl-phosphatidylcholine; Ch = cholesterol; SUV, LUV = small and large unilamellar vesicles.
case are given in Table 1. At the end, the liposome aqueous dispersion was dialyzed at room temperature for 24 h against 250 ml of the phosphate buffer with one change after the first hour. Chromatographic analyses of sera In a typical experiment, 2 ml of Zn-Pcdoped human serum were applied to a column (1.7 x 140 cm) of Sephacryl S-300, which had been previously equilibrated with 0.1 M phosphate buffer at pH 7.4. The column was eIuted with the same solvent at room temperature and at a flow rate of 26 ml/h. In general, 2.5 ml fractions were collected. For each fraction the protein absorbance was monitored at 280 nm while the of Zn-Pc was fluorescence emission measured at 675 nm upon excitation with measure600-nm light. All fluorescence ments were carried out with a Perkin Elmer MPF-4 spectrophotofluorimeter equipped with a red-sensitive phototube. Care was taken to keep the absorbance of the solutions lower than 0.05 at both the excitation and observation wavelength in order to minimize optical artifacts. Discontinuous density gradient ultracentri~ugation of sera All ultracentrifugation studies were per-
61
formed at 39,000 rev./min with a Beckman apparatus equipped with a rotor type 40.3. The density gradient was obtained by controlled addition of aqueous KBr to the [8]. Four density regions were serum obtained and the proteins isolated from the various regions were classified as follows: very low density (VLDL, d < 1.006); low density (LDL, 1.006 < d < 1.063); high density (HDL, 1.063 < d < 1.21) lipoproteins and a fraction at d > 1.21 containing the other serum proteins (bottom). The purity of each fraction was assayed by agarose gel-electrophoresis [9]. For each fraction, the content of proteins, triglycerides, phospholipid and cholesterol was estimated in order to calculate the total lipoprotein mass (holoprotein). The amount of Zn-Pc bound with each protein fraction was evaluated by spectrophotofluorimetric analysis = 600 emission range: nm) after of the with 2% SDS to absorbance lower 0.05 at 600 and nm. Before tion with the LDL, and bottom dialyzed overnight 250 ml phosphate buffer order to the KBrprecipitation of surfactant. The intensity was into Znconcentration by with a bration curve with known of Zn-Pc. Results In vitro studies A typical chromatogram obtained by column fractionation of human serum incubated with liposome-bound Zn-Pc is shown in Fig. 1. Clearly, 3 main protein peaks are observed. The most rapidly eluted peak, corresponding to high molecular weight proteins, is essentially constituted by lipoproteins, as shown by phospholipid analysis [2]. The second and third peaks mainly contain globulins and albumins, respectively [2]. Control studies show that under our experimental conditions liposomes are not eluted
110
130
150
Effluent
170
190
210
(ml)
Fig. 1. Elution profile of human serum which has been chromatographed through a column of Sephacryl S-300 after a 30-min incubation with liposome-incorporated Zn-Pc. The protein concentration was followed by the absorbance at 280 nm (O0). The concentration of Zn-Pc added in DMPC ( 0 - - - - 0) and DPPC (SW) ( l- - - -‘) liposomes was foflowed by the fluorescence emission at 680 nm.
being retained at the top of the Sephacryl S-300 column. This indicates that the Zn-Pc originally incorporated into DPPC liposomes is quantitatively and selectively transferred to serum lipoproteins. The long tailing edge of the Zn-Pc peak (Fig. 1) reflects the contribution of some light scattering to the phthalocyanine fluorescence spectrum determined in the single fractions collect:: from the column. Moreover, a limited Zn-Pc binding to other serum proteins can also occur (Table 2). Essentially identical chromatograms were obtained with all other types of liposomes studied by us; only in the case of DMPC liposomes, Zn-Pc becomes associated with several protein fractions (Fig. 1). The distribution of Zn-Pc among the various components of the lipoprotein family, and the remaining serum proteins (bottom) after delivery of the drug by the different types of liposomes is reported in Tables 2 and 3 and referred to the content of holoand apo-protein, respectively. Only for DMPC liposomes, a significant (22%) frac-
62
Recovery of Zn-Pc (ng/mg of holoprotein) from lipoprotein fractions isolated by ultracentrifugation after Table 2. 30 min incubation of human serum with the drug incorporated into different liposomes. The values in brackets represent the percentage recovery of Zn-Pc from the various protein fractions. Liposomes
VLDL
DPPC (LUV) DPPC (WV) DPPC-Ch DMPC DSPC
15.5 14.3 15.1 16.8 18.3
LDL (2.8) (2.1) (3.3) (3.5) (3.9)
16.1 23.1 67.0 32.1 7.8
(12.9) (12.7) (31.1) (16.9) (4.6)
tion of the total recovered Zn-Pc is present in the bottom, in agreement with the results of chromatographic analyses. In all cases, the largest amount of Zn-Pc @O-90% of the totally recovered phthalocyanine) is associated with HDL, while remarkably lower amounts are bound with LDL (approx. 10-158) and VLDL (3-4%). Such a distribution reflects the relative abundance of the different lipoprotein classes in human serum [lo]. A notable exception is represented by the results obtained in the presence of DPPC-Ch liposomes, where the aliquot of LDL-bound Zn-Pc is as high as 31%. In uiuo studies
Four kinds of liposomes (DMPC, DPPCSW, DPPC-Ch and DSPC) have been tested as delivery systems for Zn-Pc in vivo.
Recovery of Zn-Pc (ng/mg of apoprotein) Table 3. from lipoprotein fractions isolated by ultracentrifugation after 30 min incubation of human serum with the drug incorporated into different liposomes. Liposomes
VLDL
LDL
HDL
Other proteins
DPPC (LUV) DPPC (SW) DPPC-Ch DMPC DSPC
52.5 45.6 67.3 71.1 79.3
54.3 67.0 223.7 96.1 25.9
129.8 194.9 189.0 114.6 174.7
0.07 0.16 0.18 1.38 0.08
HDL
Other proteins
732 (82.9) 110.7 (83.2) 107.0 (62.6) 63.6 (57.3) 94.1 (90.0)
0.07 0.15 0.17 1.38 0.08
(1.4) (2.0) (3.0) (22.3) (1.5)
Once again, chromatographic analyses of rabbit sera taken at 2-4 h and 48 h after i.v.-injection of the drug indicate that Zn-Pc is exclusively carried by lipoproteins, with the exception of DMPC liposomes which release Zn-Pc to a variety of serum proteins. Tables 4 and 5 show the amount of Zn-Pc associated with the protein fractions isolated by density gradient ultracentrifugation of the same serum samples as analyzed by column chromatography. Apparently, HDL still accumulate the greatest amount of phthalocyanine, while approximately 1520% of the totally recovered Zn-Pc is bound to LDL. The latter percentage raises to 27% in the case of DPPC-Ch liposomes. In all cases, VLDL accumulate and retain a much larger percentage of the recovered Zn-Pc as compared with in vitro results. VLDL are remarkably more abundant in rabbit than in human serum [ 111. Discussion Our results clearly demonstrate that the distribution of a hydrophobic drug such as Zn-Pc among serum proteins can be modulated to some extent by the nature of the lip;>somal vesicle used as a drug-delivery system. A major role is probably played by the physical state of the phospholipid used for the preparation of the liposome. In the case of DMPC, which is in a quasi-liquid state at body temperature, Zn-Pc is released to several serum proteins. Possibly,
63 Distribution of Zn-Pc among the lipoprotein fractions isolated by ultracentrifugation from rabbit serum at Table 4. 2 or 4 h and 48 h after intravenous injection of Zn-Pc incorporated into different liposomes. Values expressed as ng/mg of holoprotein. The values in brackets represent the percentage recovery of Zn-Pc from the various protein fractions. Liposomes
(h)
VLDL
LDL
DPPC (WV)
2 48 2 48 2 48 4 48
21.1 15.9 19.6 15.0 18.4 14.2 32.9 21.9
22.9 11.4 19.8 8.6 21.1 9.8 24.3 26.6
DMPC DSPC DPPC-Ch
(10.7) (35.7) (22.6) (62.2) (21.3) (40.5) (39.7) (20.1)
the less tight packing and enhanced motility of the hydrocarbon chains favours both the interaction of the DMPC vesicle with the protein matrices and the leakage of Zn-Pc from the bilayer into the external medium [12]. On the other hand, all the liposomes whose phase transition temperature is above 37OC, hence exist in a quasi-solid state in the serum, specifically release Zn-Pc to lipoproteins. In vitro studies suggest that, in the presence of serum, such liposomes have a lifetime shorter than 30 min. In fact chromatographic analysis showed that after this period of time all Zn-Pc had been transferred from liposomes to lipoproteins. It is likely that the phospholipid vesicle efficiently
Other proteins
HDL (14.3) (21.5) (19.4) (15.1) (16.8) (17.1) (27.2) (45.8)
27.6 11.6 14.0 6.9 23.2 9.5 22.7 15.9
(66.4) (36.4) (44.0) (15.6) (56.5) (34.6) (24.6) (22.2)
0.12 0.08 0.15 0.08 0.08 0.09 0.19 0.17
(8.6) (6.4) (14.0) (6.7) (5.4) (7.8) (8.5) (11.9)
interacts with the lipid moiety of lipoproteins, independently of their chemical composition. This would explain the correlation between the amount of Zn-Pc recovered from the various lipoproteins and their relative abundance in the serum. The selectivity of the interaction between lipoproteins and solid-type liposomes is not affected by parameters as the length of the such hydrocarbon chain of the phospholipid (DPPC vs. DSPC) or the size of the vesicle (SW vs. LUV). Actually, for all these kinds of liposomes, rather similar patterns of ZnPC binding to the various lipoproteins have been found both in vitro and in vivo. The larger amount of Zn-Pc recovered from
Table 5. Distribution of Zn-Pc among the lipoprotein fractions isolated by uitracentrifugation from rabbit serum at 2 h or 4 h and 48 h after intravenous injection of Zn-Pc incorporated into different liposomes. Values expressed as ng/mg of apoprotein . Proteins
VLDL LDL HDL Other proteins
Liposomes DPPC (WV)
DMPC
2h
48 h
2h
48 h
2h
48 h
4h
48 h
53.6 65.8 47.7 0.12
70.6 37.0 20.4 0.08
78.6 61.2 25.5 0.15
90.0 28.4 12.2 0.08
68.0 65.6 50.5 0.08
60.4 32.0 16.5 0.09
175.4 85.7 37.1 0.19
82.9 102.7 24.3 0.17
DSPC
DPPC-Ch
64
VLDL in vivo almost certainly reflects the larger concentration of VLDL in rabbit as compared with human serum. This finding further supports the hypothesis that the interaction of solid liposomes with the different lipoproteins is controlled by the relative abundance of the latter in the serum. Actually, a comparative analysis of the data reported in Tables 2 and 3 indicates that the in vitro distribution of Zn-Pc among the different lipoprotein fractions, as obtained with the various liposomes, is very similar whether it is referred to the holoproteins or the apoproteins. Previous in vitro studies [13] showed that both those constituents of lipoproteins possess a discrete number of binding sites for porphyrin-type drugs. The in vivo situation is more complex. The data on Zn-Pc recovery shown in Tables 4 and 5 indicate that the phthalocyanine is eliminated at very similar rates from the holo- and the apo-lipoproteins suggestpools that both phthalocyanine ing contribute to the accumulation of the dye by tissues. Moreover, the apoprotein moiety of VLDL appears to retain larger amounts of the dye as compared with the other apolipoproteins. This finding is probably related with the different metabolic fate of the single lipoproteins and cannot be interpreted on the basis of our data owing to the complex nature of the metabolic cycles of these proteins [ 131. Only the presence of cholesterol in the phospholipid bilayer causes an alteration of the aforesaid pattern of Zn-Pc distribution, favouring the release of the phthalocyanine to LDL and VLDL. Moreover, LDL exhibit a prolonged retention of Zn-Pc both in vitro and in vivo. This observation may indicate that the presence of cholesterol determines a different mode of Zn-Pc localization in the LDL. The importance of cholesterol for the stability [ 141 and pharmacokinetic behaviour [15] of liposomes in vivo has been observed also by other investigators. The factors responsible for the preferential interaction of
cholesterol-containing liposomes with LDL are not fully understood. However, this fact is of utmost importance for the application of PDT in the treatment of tumours. Preliminary findings from our laboratory show that the administration of Zn-Pc to tumour-bearing mice via DPPC-Ch liposomes brings about a more selective targetting of the tumour by the photosensitizing drug as well as an enhanced necrotic response of the tumour to PDT especially for irradiations performed at short times (612 h) after systemic injection. References 1
Kessel, D., Thompson, P., Saatio, K. and Nantwi, K.D. (1987) Tumor localization and photosensitization by sulfonated derivatives of tetraphenylporphine. Photochem. Photobiol., 45, 787-790.
2
Barel, A., Jori, G., Perin, A., Romandini, P., Pagnan, A. and Biffanti, S. (1986) Role of high-, low- and very low-density lipoproteins in the transport and tumordelivery of hematoporphyrin in vivo. Cancer Lett., 32, 145-150. Kessel, D. (1986) Porphyrin-lipoprotein association as a factor in porphyrin localization. Cancer Lett., 33, 183188. Goldstein, J.L., Anderson, R.G.W. and Brown, M.S. (1979) Coated pits, coated vesicles and receptormediated endocytosis. Nature, 279, 679-685. Brown, M.S., Kovanen, P.T. and Goldstein, J.L. (1980) Evolution of the LDL receptor concept from cultured animal cells to intact animals. Ann. N.Y. Acad. Sci., 348, 48-68. Reddi, E., Lo Castro, G., Biolo, R. and Jori, G.
3
4
5
6
(1987) Pharmacokinetic anine in tumour-bearing 7
8
9
10 11
studies with zinc(Il)-phthalocymice. Br. J. Cancer, 56,
597-600. Kremer, J.M.H., van der Esker, W .M. J., Pathomamanoharan, C. and Wiersema, P.H. (1977) Vesicles of variable diameter prepared by a modified injection method. Biochemistry, 16, 3932-3935. Havel, R.J., Eder, r: A. and Bragdon, J.H. (1955) Distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J. Clin. Invest., 34, 1345- 1353. Noble, R.P. (1968) Electrophoretic separation of plasma lipoproteins in agarose gel. J. Lipid. Res., 9, 693700. Giesow, M. (1980) Pathways of endocytosis. Nature, 288, 434-436. Rudel, L.L., Marzetta, C.A. and Johnson, F.L. (1986) Separation and analysis of lipoproteins by gel filtration.
65
12
13
14
In: Methods in Enzymology, pp. 45-57. Editors: J.J. Albens and J.P. Segrest. Academic Press, London. Gupta, C.M. and Dhawan, S. (1981) Modification of phospholipid structure results in greater stability of liposomes in serum. B&him. Biophys. Acta, 648, 192198. Beltramini, M., Firey, P.A., Ricchelli, F., Rodgers, M.A.J. and Jori, G. (1987) Steady-state and timeresolved spectroscopic studies on the hematoporphyrinlipoprotein complex. Biochemistry, 26, 6852-6858. Eisenberg, S. (1986) Plasma lipoprotein conversion. In:
15
16
Methods in Enzymology, pp. 347-366. Editors: J.J. Albens and J.P. Segrest. Academic Press, London. Kirby, C., Clarke, J. and Gregoriadis, G. (1980) Effect of the cholesterol content of small unilamellar liposomes on their stability in vitro and in vivo. Biochem. J., 186, 591-598. Patel, H.M., Tuzel, N.S. and Ryman, B.E. (1983) Inhibitory effect of cholesterol on the uptake by liver and spleen. Biochim. Biophys. 142-151.
of liposomes Acta, 761,