MAY/JUNE 1998 Volume 9, Number 3 © Copyright 1998 by the American Chemical Society
COMMUNICATIONS Cascade Liposomal Triggering: Light-Induced Ca2+ Release from Diplasmenylcholine Liposomes Triggers PLA2-Catalyzed Hydrolysis and Contents Leakage from DPPC Liposomes Nathan J. Wymer, Oleg V. Gerasimov, and David H. Thompson* Department of Chemistry, Purdue University, West Lafayette, Indiana 47907-1393. Received January 7, 1998; Revised Manuscript Received February 19, 1998
We have previously reported a direct triggering approach [Thompson, D. H., et al. (1996) Biochim. Biophys. Acta 1279, 25-34; Gerasimov, O. V., et al. (1997) Biochim. Biophys. Acta 1324, 200-214] based on the facile degradation of plasmenylcholine and diplasmenylcholine vinyl ether linkages by either photooxidation or low-pH environments. This report describes a novel, cascade-type triggering technique that utilizes liposome photooxidation and contents release to activate an enzyme capable of destabilizing conventional phosphatidylcholine liposomes. Our application of this concept employs a mixture of two different liposome populations, one composed of synthetic diplasmenylcholine (1,2dihexadec-1′-enyl-sn-glycero-3-phosphocholine, DPPlsCho) containing Ca2+ as a signaling agent for phospholipase A2 (PLA2) and the second composed of 1,2-dihexadecanoyl-sn-glycero-3-phosphocholine (DPPC) with encapsulated calcein as the reporter molecule. Bacteriochlorophyll (BChl)-sensitized photorelease of Ca2+ from PLA2-resistant DPPlsCho liposomes activates extravesicular PLA2, thereby promoting catalyzed DPPC hydrolysis in a secondary triggering reaction, leading to calcein release. BChl/DPPlsCho/DHC/DPPE-PEG5000/Ca2+IN (0.5:85:10:5) liposomes can be phototriggered using 800 nm excitation, resulting in Ca2+ release (t50% release ) 15 min) that cocatalyzes the release of calcein (t50% release ) 40 min) from DPPC liposomes (1.5 mM total lipid in DPPlsCho liposomes, 0.18 mM DPPC, 210 µM final Ca2+ concentration, 90 units of PLA2/ml, 50 mM calcein, and 36 µM EDTA). No appreciable calcein release occurs in the absence of either PLA2 or BChl/DPPlsCho/DHC/DPPEPEG5000/CaIN liposomes. The implications of this cascade triggering technique on drug delivery approaches are briefly discussed.
Triggered release from liposomal carriers has been the subject of numerous studies (1, 2). We have previously reported a triggering approach (3), based on the facile degradation of plasmenylcholine vinyl ether linkages by either photooxidation (3a-d) or low environmental pH (3a,d,e), that utilizes plasma-stable liposomes having long blood circulation times in a murine model (1b) and * Corresponding author. Fax:
[email protected].
(765) 496-2592. E-mail:
efficiently delivers water-soluble drugs to the cytoplasm of target cells (3e). Plasmenylcholines and diplasmenylcholines, however, suffer from higher costs compared with egg or soy lecithins and from liposome processing restrictions due to the involvement of vinyl ether degradation reactions. In an attempt to mitigate these limitations, we investigated a novel, cascade-type triggering technique that utilizes liposome photorelease to activate an enzyme capable of destabilizing conventional phosphatidylcholine liposomes. Inexpensive phosphatidylcholine
S1043-1802(98)00003-2 CCC: $15.00 © 1998 American Chemical Society Published on Web 04/16/1998
306 Bioconjugate Chem., Vol. 9, No. 3, 1998
Figure 1. Schematic view of cascade triggering in a binary liposomal system. Calcium ions are shown as shaded circles, and calcein is shown as diamonds. Hydrolyzed/photolyzed lipid molecules are shown as one-tail species, as opposed to intact two-tail molecules.
(lecithin) liposomes are intended to serve as a main payload carrier, while phototriggerable liposomes containing a signaling substance will be ideally present in a small proportion. To demonstrate the feasibility of this approach, we present a system (Figure 1) which employs a mixture of two different populations of liposomes, one composed of synthetic diplasmenylcholine (4) (1,2-dihexadec-1′-enyl-sn-glycero-3-phosphocholine, DPPlsCho1) containing calcium ions as a signaling agent and the second composed of 1,2-dihexadecanoyl-sn-glycero-3-phosphocholine (DPPC) with encapsulated calcein as the reporter molecule. Photorelease of Ca2+ from diplasmenylcholine liposomes, therefore, should activate extravesicular calcium-dependent phospholipase A2 (PLA2) and promote catalyzed DPPC hydrolysis in a secondary triggering reaction, leading to calcein release (5). Bacteriochlorophyll (BChl)-sensitized photooxidation of semisynthetic plasmenylcholine (1-alk-1′-enyl-2-hexadecanoyl-sn-glycero-3-phosphocholine) liposomes has previously been used as a phototriggering mechanism (3c,d); however, due to the presence of an sn-2 acyl substituent in this phospholipid, liposomes formed from it can also serve as a PLA2 substrate. DPPlsCho, containing 1′hexadecenyl ether linkages in both the sn-1 and sn-2 positions, was chosen to form the photosensitive liposome fraction since its resistance to PLA2-catalyzed hydrolysis would help us avoid complications that could arise from PLA2-induced background calcium leakage before illumination. In addition, a protective, steric-stabilization layer of poly(ethylene oxide) was added to the DPPlsCho liposomes, in the form of 5 mol % DPPE-PEG5000, to 1 Abbreviations: BChl, bacteriochlorophyll (Porphyrin Products); DHC, (+)-dihydrocholesterol (Aldrich); DPPC, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (Avanti Polar Lipids); DPPlsCho, 1,2-dihexadec-1′-enyl-sn-glycero-3-phosphocholine; DPPE-PEG5000, 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-[poly(ethylene glycol) 5000] (Avanti Polar Lipids); EDTA, ethylenediaminetetraacetic acid, disodium salt (Sigma); HBS, HEPES-buffered saline, where [HEPES] ) 20 mM and [NaCl] ) 150 mM at pH 7.4; PLA2, calcium-dependent bee venom phospholipase A2 (Fluka); BChl/DPPlsCho/DHC/DPPEPEG5000/Ca2+IN, BChl/DPPlsCho/DHC/DPPE-PEG5000 liposomes (0.5:85:10:5 mole ratio) with encapsulated CaCl2.
Wymer et al.
Figure 2. Photorelease of Ca2+ from DPPlsCho liposomes. BChl/DPPlsCho/DHC/DPPE-PEG5000 liposomes (0.5:85:10:5 mole ratio, 10 µmol of total lipid) were prepared by extrusion (9). Briefly, lipids and BChl were dissolved in CHCl3 in a cryovial and after chloroform removal (N2 flow followed by e100 µm vacuum for g6 h) were hydrated in 0.1 M CaCl2 by multiple freeze-thaw cycles and extruded through two stacked 0.1 µm Poretics polycarbonate filters. The extraliposomal Ca2+ was then replaced by 150 mM NaCl on a 1 × 20 cm Sephadex G50M column. Illumination was carried out with a Spectra Diode Labs SDL820 laser emitting at 800 nm (incident light power of 1 W/cm2, 1 cm path length) in an unstoppered 3 mL (1 × 1 cm) glass cuvette thermostated at 37 °C, the contents of which were continuously stirred to aerate the liposome solution. Illumination was ceased after 15 min due to complete photobleaching of BChl. The release of Ca2+ was monitored by taking a 100 µL aliquot and diluting it in 2 mL of 0.2 mM Arsenazo III solution in HBS. The absorbance of the solution at 656 nm was then measured before and after the addition of 1 drop of a 10% Triton X-100 solution. Concentrations used in the experiments presented here were as follows: lipid (total) ) 1.2 mM and Ca2+ ) 120 µM (bulk concentration when completely released) in HBS.
prevent PLA2 contact with them (6), since PLA2 adsorption might, in principle, destabilize the DPPlsCho bilayer and induce contents release even in the absence of lipid hydrolysis (7). Dihydrocholesterol (DHC, 10 mol %) was also included in the DPPlsCho liposome formulation (3e) to further suppress the dark leakage rate of calcium ions (8). The results shown in Figure 2 indicate that 0.5:85:10:5 (mole ratio) BChl/DPPlsCho/DHC/DPPE-5000/Ca2+IN liposomes can be phototriggered using 800 nm excitation, yielding Ca2+ release rates that were comparable to those reported for calcein release from semisynthetic plasmalogen liposomes under similar conditions (3c). Subsequent control experiments demonstrated that pure DPPC liposomes loaded with 50 mM calcein are stable at 37 °C in the presence of PLA2 for at least 20 min, yet they rapidly release calcein in the presence of 50 µM Ca2+ (Figure 3). These results suggested that the individual liposomes in mixtures of BChl/DPPlsCho/DHC/DPPE-5000/Ca2+IN, DPPC/calcein, and PLA2 would retain their contents until the solution was illuminated. Liposome-compartmentalized Ca2+ and calcein are, in fact, retained until 800 nm irradiation induces the onset of calcein leakage (Figure 4), presumably via a cascade mechanism involving calcium ion photorelease and PLA2-catalyzed hydrolysis. Control experiments showed that appreciable calcein release from DPPC liposomes was not observed in the absence of either PLA2 or BChl/DPPlsCho/DHC/DPPE5000/CaIN liposomes. The two coupled chemical reactions of the cascade triggering mechanism (Figure 1) are DPPlsCho photooxidation and DPPC hydrolysis catalyzed by Ca2+-activated
Communications
Figure 3. PLA2-induced calcein release from DPPC liposomes in the presence of 50 µM EDTA and varying concentrations of Ca2+ at 37 °C. The DPPC liposomes were prepared by extrusion (9) (after hydrating dry DPPC in a 50 mM calcein solution), followed by gel filtration on a 1 × 40 cm Sephadex G50M column, equilibrated with 150 mM NaCl. Calcein release was monitored by measuring the solution luminescence intensity at 520 nm (λex ) 475 nm) (10). Concentrations (in HBS) were as follows: [DPPC] ) 120 µM, [PLA2] ) 15 units/mL, [EDTA] ) 50 µM, and [Ca2+]added ) (b) 50 µM, (9) 60 µM, and (2) 100 µM, resulting in 0, 10, and 50 µM free Ca2+, respectively.
PLA2. BChl-sensitized photooxidation of DPPlsCho, which initiates the entire triggering sequence, is a welldocumented reaction for the case of semisynthetic plasmenylcholine (3c,d). By analogy, singlet oxygen-mediated cleavage of the DPPlsCho vinyl ether bonds would produce a water-soluble bisformyl-sn-glycerophosphocholine fragment and 2 equiv of the single-chain surfactant, pentadecanal:
Accumulation of these surfactant photolysis products in the liposomal membrane leads to increased membrane permeability and Ca2+ release (11). Once the photoreleased Ca2+ concentration exceeds the EDTA concentration, PLA2 activation occurs, leading to the catalytic degradation of DPPC and the production of the singlechain surfactants, lysophosphatidylcholine and hexadecanoic acid, in a subsequent consecutive reaction. DPPC hydrolysis, therefore, increases the liposome bilayer permeability (12) and initiates calcein leakage in a manner analogous to that of photooxidized DPPlsCho liposomes. It should be noted that bee venom PLA2 activation requires the presence of bulk calcium ion concentrations in the micromolar range (13). We achieved these relatively high concentrations of photoreleased Ca2+ by using concentrated solutions of DPPlsCho liposomes. Therefore, in its present form, the DPPlsCho content required to elicit a detectable cascade triggering response from DPPC liposomes is approximately 8:1; presumably, this may be improved by increasing the Ca2+ concentration within the DPPlsCho liposome population.
Bioconjugate Chem., Vol. 9, No. 3, 1998 307
Figure 4. Kinetics of cascade-triggered calcein photorelease in the system comprised of two different liposomal populations described in Figures 2 and 3. Concentrations used were as follows: [BChl/DPPlsCho/DHC/DPPE-PEG5000 (0.5:85:10:5 mole ratio)] ) 1.5 mM (total lipid in the diplasmenylcholine liposome population), [Ca2+] ) 210 µM (total Ca2+ concentration), [PLA2] ) 90 units, [EDTA] ) 36 µM, and [DPPC] ) 0.18 mM.
While we have not yet reached our goal of reducing the amount of DPPlsCho required for photorelease of water-soluble materials from liposomal carriers (i.e. cascade amplification), these results clearly indicate new possibilities for cascade liposomal triggering. Incorporation of novel photochemical and cascade triggering properties into standard liposomal formulations that previously released their contents via passive diffusion is now feasible (14). Although this work focused on PLA2mediated release from DPPC liposomes, in principle, calcium ion photorelease from DPPlsCho liposomes may be used to activate any Ca2+-dependent enzyme or other calcium-requiring agent (15). It is also conceivable that this system could be used with minor modifications (i.e. encapsulation of small phototriggerable, Ca2+-containing DPPlsCho liposomes within larger PLA2- and drugcontaining phosphatidylcholine liposomes) to assemble a complete autonomous vesicular drug delivery vehicle that is photoresponsive. Structures such as these (vesicles inside vesicles with external diameters of ∼1 µm) have been described recently by Zasadzinski and co-workers (16). Another system, based on lecithin liposomes microencapsulated in much larger particles (>100 µm) of an alginate poly(L-lysine)-containing gel in which contents release rates on a week time scale was modulated by the presence of co-encapsulated PLA2, has been reported by Langer et al. (17). These and related enzyme-activated systems (18), in principle, can now be actively triggered using this cascade triggering approach. ACKNOWLEDGMENT
We thank Dr. Y. Rui for the synthesis of diplasmenylcholine and Mr. J. Boomer for assistance with manuscript preparation. Financial support from the Purdue Research Foundation and INEX Pharmaceuticals, Inc., is gratefully acknowledged. LITERATURE CITED (1) (a) O’Brien, D. F., and Tirrell, D. A. (1993) Photoinduced reorganization of bilayer membranes. In Bioorganic Photochemistry (H. Morrison, Ed.) pp 111-167, Wiley, New York. (b) Gerasimov, O. V., Rui, Y., and Thompson, D. H. (1996) Triggered release from liposomes mediated by physically and chemically induced phase transitions. In Vesicles (M. Rosoff,
308 Bioconjugate Chem., Vol. 9, No. 3, 1998 Ed.) pp 679-746, Marcel Dekker, New York, and references therein. (c) Gerasimov, O. V., Boomer, J. A., Qualls, M., and Thompson, D. H. (1998) Cytosolic drug delivery using pHand light-sensitive liposomes. Adv. Drug Delivery Rev. (in press). (d) A review of other triggerable materials has been reported: Hoffman, A. S. (1995) Intelligent polymers in medicine and biotechnology. Macromol. Symp. 98, 645-664. (2) (a) Vogel, K., Wang, S., Lee, R. J., Chmielewski, J., and Low, P. S. (1996) Peptide-mediated release of folate-targeted liposome contents from endosomal compartments. J. Am. Chem. Soc. 118, 1581-1586. (b) Miller, C. R., Bennett, D. E., Chang, D. Y., and O’Brien, D. F. (1996) Effect of liposomal composition on photoactivated liposome fusion. Biochemistry 35, 11782-11790. (c) Armitage, B. A., Bennett, D. E., Lamparski, H. G., and O’Brien, D. F. (1996) Polymerization and domain formation in lipid assemblies. Adv. Polym. Sci. 126, 53-84. (d) Song, X., Perlstein, J., and Whitten, D. G. (1995) Photoreactive supramolecular assemblies: aggregation and photoisomerization of azobenzene phospholipids in aqueous bilayers. J. Am. Chem. Soc. 117, 7816-7817. (e) Chang, C., Niblack, B., Walker, B., and Bayley, H. (1995) A photogenerated pore-forming protein. Chem. Biol. 2, 391-400. (f) Lee, R. J., and Huang, L. (1996) Lipidic vector systems for gene transfer. J. Biol. Chem. 271, 8481-8487. (g) Nicol, F., Nir, S., and Szoka, F. C. (1996) Effect of cholesterol and charge on pore formation in bilayer vesicles by a pH-sensitive peptide Biophys. J. 71, 3288-3301. (h) Bailey, A. L., and Cullis, P. R. (1997) Membrane fusion with cationic liposomes: Effects of target membrane lipid composition. Biochemistry 36, 16281634. (i) Bailey, A. L., Monck, M. A., and Cullis, P. R. (1997) pH-induced destabilization of lipid bilayers by a lipopeptide derived from influenza hemagglutinin. Biochim. Biophys. Acta 1342, 232-244. (j) Kono, K., Igawa, T., and Takagishi, T. (1997) Cytoplasmic delivery of calcein mediated by liposomes modified with a pH-sensitive poly(ethylene glycol) derivative. Biochim. Biophys. Acta 1325, 143-154. (3) (a) Anderson, V. C., and Thompson, D. H. (1992) Triggered release of hydrophilic agents from plasmalogen liposomes using visible-light or acid. Biochim. Biophys. Acta 1109, 3342. (b) Anderson, V. C., and Thompson, D. H. (1992) Photoinduced morphological-changes in plasmalogen liposomes using visible-light. In Macromolecular Assemblies (P. Stroeve and A. Balazs, Eds.) ACS Symposium Series 493 pp 154-170, American Chemical Society, Washington, DC. (c) Thompson, D. H., Gerasimov, O. V., Wheeler, J. J., Rui, Y., and Anderson, V. C. (1996) Triggerable plasmalogen liposomes: Improvement of system efficiency. Biochim. Biophys. Acta 1279, 25-34. (d) Gerasimov, O. V., Schwan, A., and Thompson, D. H. (1997) Acid-catalyzed plasmenylcholine hydrolysis and its effect on bilayer permeability: a quantitative study. Biochim. Biophys. Acta 1324, 200-214. (e) Rui, Y., Wang, S., Low, P. S., and Thompson, D. H., (1998) J. Am. Chem. Soc. (submitted for publication). (4) Rui, Y., and Thompson, D. H. (1994) Stereocontrolled synthesis of plasmalogen-type lipids from glyceryl ester precursors. J. Org. Chem. 59, 5758-5762. (5) This reaction sequence is analogous to the cascade amplification pathways involving intracellular Ca2+ release and calcium-dependent PLA2 activation in mammalian cell signal transduction cascades [Dennis, E. A. (1994) J. Biol. Chem. 269, 13057-13060; Clark, J. D., Schievella, A. R., Nalefski, E. A., and Lin, L. L. (1995) J. Lipid Mediators Cell Signalling 12, 83-117]. (6) Addition of PEG-conjugated lipids is a common strategy for reducing protein adsorption onto the liposome surface (see ref 14a). (7) It has been reported that PLA2 isozymes, while lacking PLA2 activity, can cause leakage of carboxyfluorescein from phosphatidylcholine liposomes [Shimohigashi, Y., et al. (1995) J. Biochem. 118, 1037-1044]. This means that PLA2 can, in some cases, cause leakage simply by disturbing bilayer structure upon contact. We found that the presence of PEG reduced dark leakage of Ca2+ from DPPlsCho liposomes but did not affect their photosensitivity. (8) EDTA was present in the solution (∼0.25 equiv of EDTA/ equiv of Ca2+) to chelate any Ca2+ that may have leaked
Wymer et al. passively before the illumination started. EDTA was included as a precautionary measure even though appreciable dark calcium leakage was not observed from any batches of DPPlsCho/DHC/DPPE-PEG5000/Ca2+IN liposomes (cf. Figure 2). Its presence did not affect the performance of either the phototriggering or PLA2 hydrolysis reactions once the free (i.e. photoreleased) calcium ion concentration exceeded the EDTA concentration. (9) Mayer, L. D., Hope, M. J., and Cullis, P. R. (1986) Vesicles of variable sizes produced by a rapid extrusion procedure. Biochim. Biophys. Acta 858, 161-168. (10) Allen, T. M., and Cleland, L. G. (1980) Serum-induced leakage of liposome contents. Biochim. Biophys. Acta 597, 418-426. (11) The details of this release mechanism and its quantitative correlation to the accumulation of single chain hydrolysis fragments within the liposome bilayer are described in ref 3d. In particular, it was shown that 3-5 mol % plasmenylcholine cleavage was sufficient to promote liposomal contents release within minutes in the absence of cholesterol. (12) (a) Fugman, D. A., Shirai, K., Jackson, R. L., and Johnson, J. D. (1984) Lipoprotein lipase- and phospholipase A2catalyzed hydrolysis of phospholipid vesicles with an encapsulated fluorescent dye. Effects of apolipoproteins. Biochim. Biophys. Acta 795, 191-195. (b) Eriksson, O., and Saris, N. E. (1989) The phospholipase A2-induced increase in the permeability of phospholipid membranes to Ca2+ and H+ ions. Biol. Chem. Hoppe Seyler 370, 1315-1320. (c) Kaszuba, M., and Hunt, G. R. A. (1990) 31P- and 1H-NMR investigations of the effect of n-alcohols on the hydrolysis by phospholipase A2 of phospholipid vesicular membranes. Biochim. Biophys. Acta 1030, 88-93. (13) When completely released, Ca2+ concentrations were typically 150-200 µM. (14) Practical methods for active liposome triggering are still sought for site-specific drug and gene delivery applications since the passive drug release-exchange mechanisms involved in standard liposome formulations (14a-e) deliver water-soluble agents inefficiently. (a) (1995) Lasic, D., and Martin, F., Eds. Stealth Liposomes, CRC Press Inc., Boca Raton, FL. (b) Gregoriadis, G. (1995) Engineering liposomes for drug delivery: progress and problems. Trends Biotechnol. 13, 527-537. (c) Chonn, A., and Cullis, P. (1995) Recent advances in liposomal drug-delivery systems. Curr. Opin. Biotechn. 6, 698-708. (d) Allen, T. M. (1996) Liposomal drug delivery. Curr. Opin. Colloid Interface Sci. 1, 645-651. (e) Lasic, D. D., and Papahadjopoulos, D. (1996) Liposomes and biopolymers in drug and gene delivery. Curr. Opin. Solid State Mater. Sci. 1, 392-400. (15) Calcium-containing photosensitive plasmenylcholine liposomes are a higher payload analogue of stoichiometric caged calcium reagents. (16) Walker, S. A., Kennedy, M. T., and Zasadzinski, J. A. (1997) Encapsulation of bilayer vesicles by self-assembly. Nature 387, 61-64. (17) Kibat, P. G., Igari, Y., Wheatley, M. A., Eisen, H. N., and Langer, R. (1990) Enzymatically activated microencapsulated liposomes can provide pulsatile drug release. FASEB J. 4, 2533-2539. (18) (a) PLA2 has also been used to activate the degradation of phospholipid tubules: Carlson, P. A., Gelb, M. H., and Yager, P. (1997) Zero-order interfacial enzymatic degradation of phospholipid tubules. Biophys. J. 73, 230-238. (b) Menger, F. M., and Johnston, D. E. (1991) Specific enzyme-induced decapsulation. J. Am. Chem. Soc. 113, 5467-5468. (c) Russo, M. J., Bayley, H., and Toner, M. (1997) Reversible permeabilization of plasma membranes with an engineered switchable pore. Nat. Biotechnol. 15, 278-282. (d) Pinnaduwage, P., and Huang, L. (1988) A homogeneous, liposome-based signal amplification for assays involving enzymes. Clin. Chem. 34, 268-272. (e) Liu, D. (1994) Enzyme-controlled release with liposomes. J. Liposome Res. 4, 413-425.
BC980003P
