LIPOSOMES

Abstract

Liposomes with covalently bound PEG moieties on the external surface which demonstrate improved serum half-life following intravenous administration are provided.

Description

Summary of Invention

[0008] The present invention therefore provides liposomes having covalently bound PEG moieties on the external surface.

[0009] Preferably the PEG moieties are linked to amino groups in the head group of at least one phospholipid species forming the liposome. Suitable phosholipids having amino groups in the head group include phosphatidylethanolamine (PE) and phosphatidyl serine (PS).

[0010] The liposomes may be formed of any suitable phospholipid or phospholipid mixture, of which a great many are already known in the literature, provided that at least one of the phospholipid species has a suitable head group for binding PEG. The space within the liposomes may contain any conventional aqueous phase and the liposomes may be presented as an aqueous suspension or as any other conventional formulation, for instance as pharmaceutical formulations also comprising a pharmaceutically acceptable carrier or diluent, for instance as formulations for intravenous administration. Preferred carriers include sterile water for injection with optional accessory ingredients such as buffers, preservatives, antioxidants and isotonic salts.

[0011] Preferably the liposomes are large unilamellar vesicles prepared by extrusion (LUVettes), more preferably lipid bilayers consist of a 7:3 to 5:5 molar ratio of dioeylphosphatidyl choline and dioleylphosphatidyl ethanolamine and most preferably the liposomes contain aqueous Gd-DTPA.

[0012] The invention further provides a process comprising treating liposomes with a reactive derivative of polyethylene glycol, preferably 2,2,2-trifluoroethanesulphonyl (tresyl) monomethoxy PEG. Tresyl monomethoxy PEG (TMPEG) and its production is described in our co-pending British application no. 8824591.5.

[0013] Preferably the reaction between the reactive PEG derivative and the liposomes is conducted in aqueous solution at ambient or physiological temperatures. The reaction occurs at near neutral pH, for instance in physiological buffer but is faster and more extensive at pH9-10. By controlling the ratio of reactive PEG derivative to liposomes, the number of PEG moieties linked to the liposomes may be controlled.

[0014] Poly(ethylene glycol) (PEG) is a linear, water-soluble polymer of ethylene oxide repeating units with two terminal hydroxyl groups:HO(CH2CH2O)nCH2CH2OHPEG's are classified by their molecular weights, thus PEG 6000, for example, has a molecular weight of about 6000 and n is approximately 135.

[0015] PEG's can be covalently linked to proteins by a variety of chemical methods. We have used tresyl chloride (2,2,2-trifluoroethane sulphonyl chloride) to activate the single free hydroxyl group of monomethoxy PEG 5000 (MPEG) but other tresyl halides and other reactive derivatives of MPEG may be used. By having the other hydroxyl group of PEG "blocked" as the unreactive methyl ether, the possibility of producing PEG activated at both ends, which would give rise to cross-linked lipids in the coupling stage, is avoided.

[0016] The phospholipids phosphatidylethanolamine (PE) and phosphatidyl serine (PS) have a free amino group in the polar head group. In aqueous solutions phospholipids show lyotropic mesomorphism; most phospholipids adopt closed vesicle structures comprising lipid bilayers (liposomes). PE on its own adopts the HII phase, but in mixtures with phosphatidylcholine (PC) adopts bilayer organizations. We have prepared liposomes from PE/PC mixtures to provide lipid vesicles with the amino groups of PE exposed at both the outer and inner surface. Only the outer PE molecules are accessible to the tresyl-PEG, so the modification is asymmetric.

[0017] The amount of PEG linked to the liposome surface can be controlled by varying the lipid composition, the ratio of the reactive derivative of polyethylglycol to the phospholipid having an amino group-containing head group, the duration of the reaction and the pH. The production process may be optimised by systematic studies using, for instance, release of entrapped dye as a marker for disruption of the integrity of the lipid bilayer and by monitoring half-life of treated liposomes in, for instance, the blood stream of mice following intravenous administration.

[0018] The major fate of untreated liposomes injected in to the circulation, regardless of size, is uptake by the Kupfer cells of the liver and by fixed macrophages in the spleen. Such uptake by the reticulo-endothelial system (RES) limits the applicability of liposomes in applications such as the formation of reservoirs for the slow release of biologically active molecules and for treatment of tissues other than RES tissues. Treatment of the liposomes according to the present invention, in order to introduce PEG moieties on the external surface surprisingly reduces the interaction between serum and the liposome and surprisingly increases the circulation life-time following intravenous administration.

[0019] A particularly preferred use of the PEG-bearing liposomes of the present invention is in the delivery of MR imaging agents such as Gd:diethylenetriaminepentacedic acid chelate.

[0020] The invention further provides the use of liposomes having PEG moieties bound to their external surfaces in therapeutic and diagnostic methods practised on the human or animal body, for instance as delivery means for drugs and for contrast agents for Magnetic resonance (MR) imaging. The invention provides a therapeutic or diagnostic process comprising intravenous administration of an effective, non toxic amount of a PEG-bearing liposomes as hereinbefore described containing a diagnostic or therapeutic agent to a human or non-human animal in need thereof.

Brief Description of Drawings

[0021] The invention will now be illustrated by the figures of the accompanying drawings which:

[0022] FIG. 1. shows a comparison of the clearance of PEGylated SUV's and unPEGylated SUVs from the circulation in mice.

[0023] FIG. 1A: SUVs of composition DSPC:PE:Cholesterol (molar ratio 0.4:0.1.5) either PEGylated ▪ or untreated (Circlesolid) were injected iv into mice (0.4 mg/25 g mouse). Blood levels of CF (dose+se, 5 animals) are shown; 3H phospholipid clearance was similar (not shown).

[0024] FIGS. 1B: and 1C: Identical conditions to FIG. 1A except that the SUV preparation had been centrifuged to 100,000 g for 1 hr to remove larger vesicles and the injected dose was 0.8 mg/25 g mouse. Both CF FIG. 1B clearance and 3H phospholipid clearance FIG. 1C are shown for PEGylated (&Circlesolid;) and unPEGylated (compfn) vesicles.

Detailed Description

[0025] The invention will now be illustrated by the following Examples:

[0026] EXAMPLES 1-10

[0027] PREPARATION OF PEGYLATED LIPID VESICLES

[0028] A. Preparation of Activated Tresyl

[0029] MPEG Tresylated monomethoxy PEG (TMPEG) was obtained by treating dry monomethoxy PEG 5000, which is available from Union Carbide, in dichloromethane, with tresyl chloride (2,2,2-trifluoroethane-sulphonyl chloride) which is available from Fluka, at room temperature, using pyridine as a base catalyst. Dichloromethane was removed under reduced pressure and the solid obtained dissolved in methanol-HCl mixture (0.3 ml conc HCl per 1000 ml) and reprecipitated at between -20 and 00. The solid was isolated by centrifugation, the process repeated until the sample was free of pyridine (detected at 255 nm), and then the solid was reprecipitated from methanol until acid free.

[0030] B. PEGylation of Lipid Vesicle Surfaces

[0031] The resulting TMPEG was reacted with lipid vesicles at room temperature in buffered solutions (see below). The MPEG covalent attachment of the MPEG to the outer surface of the vesicles was demonstrated by the alteration in the partitioning behaviour of the vesicles in aqueous two-phase systems of PEG and dextran, by a method similar to that of Tilcock et al., Biochim. Biophys. Acta 979:208-214 (1989). The composition of the phase system was adjusted so that the vesicles showed a low partition in the top PEG-rich phase; vesicles were at the interface or in the MPEG bottom dextran-rich phase. Attachment of MPEG to the vesicle surface makes them more "PEG-like" (increases their wetting by the PEG-rich phase) and they partition to the top phase.

[0032] Example 1

[0033] PEGylation of MLVs (Multilamellar Vesicles)

[0034] Multilamellar vesicles containing 20% (w/w) egg phosphatidylethanolamine (EPE) and 80% (w/w) egg phosphatidylcholine (EPE) and 14C EPC were prepared in 0.125M NaCl containing 0.05M sodium phosphate buffer, pH 7.5 (PBS) at 10 mg total lipid/ml. 0.1 ml samples of vesicles were incubated with solutions of TMPEG prepared in PBS (final concentrations 0-170 mg/ml) for 2 hours at room temperature. Samples were partitioned by adding samples (0.05 ml) to a biphasic system (1 ml of top phase and 1 ml of bottom phase of a phase system of 5% (w/w) PEG 6000 and 5% (w/w) Dextran T500 in 0.15M NaCl containing 0.01M sodium phosphate, pH 6.8, mixing the systems and measuring the radioactivity in samples taken from the mixture immediately after mixing (total) and from the top and bottom phases after phase separation was completed (20 min).

[0035] The results in Table 1 show that exposure of the liposome to TMPEG increases their partition into the PEG-rich top phase. This indicates that PEG has become attached to the liposome, presumably by the covalent attachment to the amino group of the EPE.

[0036]

FINAL TMPEG	PARTITION (%)	PARTITION (%)	PARTITION (%)	n
(mg/ml)	Top Phase	Interface	Bottom Phase
0.0	9.1 ±4.7	84.5 ±4.1	6.4 ±2.4	9
2.0	14.5 ±5.4	80.2 ±4.2	5.3 ±1.6	3
8.0	44.9 ±6.3	50.8 ±6.5	4.3 ±0.4	3
12.5	74.7 ±9.5	20.1 ±10.5	5.2 ±1.4	3
25.0	96.3 ±7.8	3.1 ±3.6	4.6 ±0.8	4
50.0	89.3	6.5	4.5
100.0	88.8	5.1	6.1
170.0	89.3	6.5	4.2	1

[0037] The presence of PE in the vesicle is required for TMPEG to have any effect. When MLVs of 100% EPC were treated with TMPEG for two hours and then partitioned in a 5%/5% PEG 6000-Dextran T500 systems in 0.15M NaCl buffered with 0.01M sodium phosphate, pH 6.8 there was no difference compared to MLVs treated with buffer (Table 2).

FINAL TMPEG	PARTITION (%)	PARTITION (%)	PARTITION (%)	n
(mg/ml)	Top Phase	Interface	Bottom Phase
0	22.5 ±13.0	71.6 ±12.0	5.9 ±1.0	5
25	25.8 ±13.0	67.8 ±14.0	6.4 ±1.0	5

[0038] The activity of TMPEG declines on storage. Samples that had lost their ability to PEGylate proteins were found to have no effect on the partitioning of liposomes containing EPE. This observation, taken together with the inablity of TMPEG to effect non-PE containing vesicles supports the conclusion that TMPEG attaches to PE specifically, and that altered partitioning does not arise from adsorption of TMPEG to vesicle surfaces.

[0039] Example 2

[0040] PEGylation of SUVs (Small Unilamellar Vesicles)

[0041] SUVs composed of distearoylphosphatidylcholine (DSPC), dipalmitoylphosphatidylethanolamine (DPPE) and cholesterol in molar ratio 0.8:0.2:1 were prepared by the method of Senior et al., Biochim. Biophys. Acta. 839: 1-8 (1985), with tracer 3H-DPPC (6x106 dpm per 30 mg phospholipid): 25 mg DSPC, 5.5 mg DPPE and 15 mg cholesterol were hydrated in 2 ml PBS (0.125M NaCl buffered with 0.05M Naphosphate buffer, pH 8.5). To measure liposomal retention of water-soluble molecules during the coupling reaction and subsequent procedures, Carboxyfluorescein was partially purified and entrapped at 0.15M as described by Senior et al., Biochim. Biophys. Acta 839: 1-8 (1985). 0.5 ml SUV were incubated with an equal volume of TMPEG, prepared in PBS (0.125M NaCl buffered with 0.05M Naphosphate buffer, pH 8.5) at 125 mg/ml. for 2 hours at room temperature (Ratio of TMPEG to total DPPE is 6.25). The vesicles were then separated from unreacted TMPEG by gel filtration on Sepharose 4B-CL and partitioned as in Example 1 in a phase system of 5% PEG 8000 (Union Carbide) and 5% Dextran T500 (Pharmacia) in 0.15M NaCl containing 0.01M sodium phosphate, pH 6.8. The results in Table 3 show that exposure of the liposomes to TMPEG increases their partition into the PEG-rich top phase compared with vesicles treated only with buffer (control). This suggests that PEF has been covalently linked to the amino group of the DPPE. PEGylation proceeded without the loss of the entrapped CF.

VESICLES	PARTITION (%)	PARTITION (%)	PARTITION (%)
Phase	Top Phase	Interface	Bottom Phase
Untreated	1.4 ± 0.2	36.0 ± 5.0	62.5 ±5.1
TMPEG-treated	96.5 ± 1.0	1.4 ± 1.1	2.1 ±0.4
1mean ±n = 6

[0042] Example 3

[0043] The SUVs, as used in Example 2, were treated with TMPEG (125 mg/ml) and their partitioning compared with SUVs treated with MPEG (125 mg/ml) or buffer: the TMPEG treated vesicles were completely (100%) partitioned into the top phase, whereas the MPEG-treated vesicles and buffer-treated vesicles showed no top phase partitioning, and similar even distributions between the interface and the bottom phase. This provides additional support for the suggestion that TMPEG acts by covalent attachment to the vesicle surface, and not by adsorption.

[0044] Example 4

[0045] PEGylation of LUVettes (Large UnilamellarVesicles Prepared by Extrusion) of Defined Size

[0046] LUVettes were prepared as described by Tilcock et al., Biochim. Biophys. Acta (979:208-214 (1989).

[0047] LUVettes of 100 nm diameter were prepared at a final concentration of 10 mg/ml. Mixtures of dioleylphosphatidylcholine (DOPC) and dioleylphosphatidyl ethanolamine (DOPE) in chloroform at various molar ratios (total 20 mmoles) were combined with 2uC of 33H DPPC and the solvent removed by evaporation under reduced pressure (<0.1 mn Hg) for 2 hours. The lipid was dispersed by vortex mixing at room temperature in 1.55 ml of 50 mM Hepes, 100 mM NaCl pH 7-9 to give a final lipid concentration of 10 mg/ml. Large unilamellar vesicles were then produced by repeated extrusion (10 times) of the lipid dispersion MLVs through two stacked 100 nm polycarbonate filters using the Extruder device (Lipex Biomembranes, Canada) by the method of Hope et al., Biochim. Biophys. Acta 812: 55-65 (1985). Diameters determined by QEL using a Nicomp model 270 particle analyzer.

[0048] The vesicles were PEGylated by incubation with 40 ul of buffer containing TMPEG at room temperature. At intervals 20 ul samples were removed and partitioned in a phase system of 1.5 ml top phase and 1.5 ml bottom phase of a 5% PEG 8000 (Union Carbide) and 5% Dextran T500 (Pharmacia) system prepared 0.15M NaCl buffered with sodium phosphate pH 6.8 at room temperature. Samples of top and bottom phase were removed for counting 20 min after the phase had been mixed and allowed to separate. This phase system was selected so that the partitioning of the untreated vesicles into the top phase was extremely low (>5%); the majority of the vesicles were approximately equally divided between the bottom phase and the bulk interface.

[0049] Example 5

[0050] The time course and pH dependency of the PEGylation reaction using a two-fold excess of TMPEG to the DOPE present at the outer surface of LUVettes are used in Example 4. At pH 8-9 incubation with TMPEG rapidly caused a time dependent transfer of vesicles to the top phase. At pH 7.5 the reaction was considerably slower and at pH 7.0 there was virtually no transfer to the top phase. In a separate experiment in which the bottom phase and interface partitioning was also measured it is seen that at pH 7.2, although top phase partitioning does not alter there was decrease in bottom phase partitioning with an increase in interface partitioning, indicating that PEGylation proceed at pH 7.2 albeit more slowly than at higher pHs. At pH 8 the partitioning moves from the bottom phase to the interface and then to the top phase; at pH 9 and 10 vesicles are moved rapidly from the interface and bottom phase to the top phase. Thus the PEGylation reaction is very sensitive to pH and appropriate choice of conditions of time and pH can determine the degree of PEGylation. The extent of PEGylation can also be controlled by the amount of TMPEG used. Treating 100 nm Luvettes of DOPE/DOPC (0.2:0.8) at pH 9.0 with varying molar ratios of TMPEG increased partitioning into the top phase consistent with increasing PEGylation. There was a marked increase in top phase partitioning between the molar ratios 1.0 and 1.3 from 20% to 90%. When the partitioning in the bottom phase and at the interface is also measured (Table 4) it can be seen that PEGylation at the lower ratios of TMPEG:outerDOPE molar ratio causes a progressive change in the partition from the bottom phase to the interface and subsequently to the top phase demonstrating gradations in the degree of PEGylation.

[0051] It is clear from the time course of the partitioning that reaction at pH 9 is virtually complete by 1 hour. Thus defined degrees of PEGylation are obtained by control of the TMPEG:DOPE ratio.

Molar ratioTMPEG:DOPE	Partitioning (%)	Partitioning (%)	Partitioning (%)
at outer surface	Bottom	Interface	Top
0	50	40	10
0.2	56	41	3
1.0	28	58	1
1.3	1	9	89

[0052] Measurement of the fraction of amino groups (from PE) exposed at the outer surface of the LUVettes, made by the method of Hope, M. J. and Cullis. P. R. J. Biol. Chem. 262: 4360-4366 (1987) in 0.05M TNBS in borate buffer at pH 8.5, gave values of 47% for DOPC:DOPE vesicles (8:2), close to the theoretical value of 50% for equal distribution of the PE between the inner and outer surfaces. PEGylation caused a decrease in the PE content detectable by this assay, suggesting covalent attachment of the MPEG to the free NH2 group of PE. For example, when a 3-fold mole excess of TMPEG to outer PE was added to DOPC:DOPE vesicles of 7:3 molar ratio for 1 hour, the percentage of outer PE PEGylated was 36%; when a 6-fold molar excess was added, this percentage PEGylation increased to 45%.

[0053] Example 6

[0054] Stability of Lipid Vesicles to PEGylation

[0055] The stability of lipid vesicles was measured by the extent of efflux of 6CF (6-carboxyfluorescein) as described by Senior and Gregoriadis in "Liposome Technology." (G Gregoriadis ed) vol 3, p. 263 (1984) CRC Press. LUVettes of 100 nm composed of DOPC:DOPE were prepared with entrapped 50mM 6CF (6-carboxyfluorescein) in 100 mM NaCl at pH 8.5, external 6CF was removed by column chromatography on Sephadex G-25 using 50 mM Hepes, 100 mM NaCl, pH 8.5 as eluant. Samples for latency measurement were added to 4 ml of buffer (100 mM NaCl, 50 mM HEPES pH 9) and fluorescence measured (dye released), and to 4 ml of buffer containing 25 mM octylglucoside, incubated for 30 mon at 37.degree. to ensure complete disruption of the vesicles and fluorescence measured (total dye). Fluorescence was measured at 490 nm excitation and 520 nm emission.

[0056] LUVettes of 100 nm were PEGylated with TMPEG without any loss of latency. Vesicles of DOPC: DOPE 8:2 were treated with a 3 fold molar ratio of TMPEG to DOPE present in the outer vesicle surface at pH 8.5 to ensure extensive PEGylation (demonstrated by phase partitioning). There was no leakage of 6CF out of the vesicles over a period of 2 hours demonstrating that PEGylation occurs without disruption of the lipid bilayer.

[0057] Example 7

[0058] Interaction of SUVs with Serum

[0059] 0.1 ml of SUVs of composition DSPC:PE:Cholesterol (molar ratio 0.4:0.1:0.5), with or without coupled PEG (see above) were incubated at 376° with 0.5 ml of fresh plasma (mouse) or buffer. Samples were removed at intervals and partitioned as in Example 2 above. SUVs partitioned about 20% top phase, 60% interface and 20% bottom phase. Treatment with serum caused an immediate (within 1 min) alteration in the vesicle surface properties indicated by their partition: 0% top phase, 40% interface and 60% bottom phase. The plasma proteins alone partitioned mainly to the bottom phase (68% bottom, 32% top; Partition coefficient=0.47±0.02, n=4). Thus it appears that the SUVs are immediately coated with serum proteins which then cause the vesicles to partition with similar characteristics to the proteins. PEGylation of the SUVs increased their partition into the top phase (almost 100%); on exposure to serum there was a change in their partition towards the interface and the bottom phase, but importantly this process was very slow compared with the virtually instantaneous effect of serum on unPEGylated SUVs. Since the partitioning behaviour relates to the sum of the forces imposed by the PEGylation and serum binding, and with the former is not a linear function, it is not simple to determine whether the effect of serum on partition is equal for the PEGylated and for the unPEGylated liposomes. This could, however, be determined with a detailed dose response analysis of the effect of PEGylation on the partition coefficient so that the influence of serum could be determined at various parts of the dose response curve in "PEG-equivalents". This would establish whether serum had different effects on the PEGylated and unPEGylated liposomes. The order of magnitude differences in partition behaviour suggests that PEGylation slows down the adsorption of serum components onto the vesicles.

[0060] Separation of the SUVs exposed to serum by gel chromatography gave vesicles which showed partitioning behaviour close to that of the vesicles before exposure. Thus the interaction between vesicles and serum is reversed by reisolation of the vesicles.

[0061] These experiments also demonstrate that the altered surface properties of the SUVs imposed by PEGyliition are not substantially reversed by serum protein adsorption.

[0062] Example 8

[0063] Stability of LUVettes to Serum is Increased by PEGylation

[0064] To determine the stability of LUVettes to serum vesicles containing entrapped 6CF (50 ul) were incubated at 37° with 0.5 ml serum (freshly hydrated lyophilised human serum, Monitrol-ES, Dade Diagnostics) to provide a final lipid concentration of approx 1 mg/ml, a concentration corresponding to the maximum in vivo serum concentrations expected on the basis of the imaging experiments of Unger et al Radiology 171: 81-85. Samples were removed at intervals and the 6CF released was measured fluorimetrically. Vesicles were PEGYlated with a 3-fold excess of TMPEG to outer surface DOPE overnight at room temperature, after which time there had been loss of latency.

[0065] 50 nm vesicles of DOPC:DOPE at 8:2 molar ratio showed considerable loss of latency in the presence of serum (eg only 10% latency remained after 2 hrs) which PeGylation did not decrease; 100 nm vesicles showed a latency of 35% after 2 hrs which was unaffected by PEGylation; 200 nm vesicles showed a smaller loss of latency (eg 65% latency remained after 2 hrs), which also was not inhibited by PEGylation. However, for 100 nm vesicles of 7:3 molar ratio DOPC:DOPE, PEGylation decreased serum induced loss of latency by a factor of 2. Increasing the DOPE content to 40 mole % and 50 mole % increased the stability of the vesicles to serum; nevertheless PEGylation produced additional stabilisation. Table 5 summarises these data.

DOPC:DOPE	Latency (%)	Latency (%)
molar ratio	UnPEGYLATED	PEGylated
8:2	35	35
7:3	55	83
6:4	90	9
5:5	92	99

[0066] Example 9

[0067] PEGylation Does not Alter the Relativity of Encapsulated Gd-DTPA

[0068] Gd-DTPA was encapsulated in LUVettes composed of DOPC:DOPE 7:3 by the method of Tilcock et al Radiology 171: 77-80 (1989).

[0069] Half of the sample was PEGylated with TMPEG (molar ratio of TMPEG: PE on outer surface of 3:1). Both control and PEGylated samples were diluted in saline bilffer (139 mM NaCl, 10 m Hepes, 6 mM KCl, pH 8.5) to give four samples with effective Gd concentrations of 2, 1, 0.5, and 0.25 mM (calculated as described by Tilcock et al., Radiology 171: 77-80 (1989) given the known trap volume of the vesicles, the lipid concentration and assuming the concentration of entrapped Gd-DTPA was 0.67M. ) Samples of 10-12 ml were imaged with a Toshiba 0.5T MRT-50A whole body scanner. Relaxivites are obtained from linear regressions of 1/T1 (spin lattice relaxation time constant) against the effective Gd-DTPA concentration. These were unaffected by PEGylation of the vesicles.

[0070] Example 10

[0071] PEGylation of SUVs Decreases Their In Vivo Clearance

[0072] SUVs of composition DSPC:PE:Cholesterol (molar ratio 0.4:0.1:0.4) (0.2 ml containing 0.4 mg phosphpholipid) were injected intravenously into the tail vein of male TO mice (5 in each group). Clearance of PEGylated and unPEGylated vesicles was assessed from entrapped CF and 3H-radiolabelled phospholipid measured in blood samples (25 ul) withdrawn at intervals in the method of Senior and Gregoriadis in "Liposome Technology" vol 3 pp 263-282 (1984), CRC Press. In another experiment an 0.8 mg dose of phospholipid was given as the supernatant from ultracentrifugation at 100,000 g for 1 hour, which contains small vesicles of 20-100 nm (average 50 nm) as described by Senior et al Biochim Biophys Acta 839: 1-8 (1985).

[0073] FIG. 1A shows the clearance of SUVs after intravenous administration of a sonicated, uncentrifuged preparation. This preparation contains, presumably, some larger vesicles which are cleared rapidly, in both the PEGylated and unPEGylated samples. However the slower clearance phase corresponds to about 50-60% of the lipid dose and showed a marked difference in the half life of the PEGylated sample (10 hr) compared with the unPEGylated preparation (7 hr). In the preparation in which the larger vesicle had been removed (FIG. 1B and FIG. 1C) the PEGylated vesicles had half life of 14 hr compared with untreated vesicles of 12 hr.

Claims

1.Liposomes having PEG moieties covalently bound to phospholipids on the external surface, wherein said liposomes are selected from large unilamellar vesicles (LUV's), small unilamellar vesicles (SUV's) and multilamellar vesicles (MLV's).

2.Liposomes according to claim 1, wherein said liposomes comprise a mixture of lipids.

3.Liposomes according to claim 2 wherein the lipid bilayers comprise a 7:3 to 5:5 molar ratio of DOPC to DOPE.

4.Liposomes according to claim 2 wherein the lipid bilayers comprise a mixture of dioleylphosphatidylcholine (DOPC) and dioleylphosphatidylethanolamine (DOPE).

5.A pharmaceutical composition comprising an aqueous suspension of liposomes according to claim 1 and a pharmaceutically acceptable carrier or diluent.

6.A process for producing a liposome according to claim 1 comprising treating liposomes with a polyethylene glycol having at least one activating group capable of coupling said polyethylene glycol to said liposome.

7.A process according to claim 6 wherein the reactive derivatave is 2,2,2-trifluoroethane sulphonyl-monomethoxy-polyethylene glycol.

8.Liposomes according to claim 1, obtained by reacting 2,2,2-trifluoroethane sulfonyl-monomethoxy PEG derivatives with liposomes.

9.Liposomes according to claim 1, wherein essentially all said PEG moieties are bound on the external surface of the liposome.

10.Liposomes according to claim 1, wherein said liposomes display an enhanced partition to the PEG-rich (upper) phase of a PEG:dextran aqueous two phase system in which liposomes not having PEG moieties covalently bound to phospholipids on the external surface separate predominantly to the interface or bottom phase.

11.Liposomes according to claim 1, wherein said liposomes display a decreased adsorption of serum proteins than liposomes not having PEG moieties covalently bound to phospholipids on the external surface.

12.Modified liposomes having a reduced rate of removal from in vivo circulation, characterized in that said liposomes comprise an aqueous interior compartment enclosed by a lipid bilayer comprising phospholipid species having covalently linked PEG moieties, wherein those PEG moieties that are on an exterior surface of the liposomes reduce the rate of removal of the liposomes from in vivo circulation.

13.Liposomes according to claim 12, wherein said liposomes comprise a mixture of lipids.

14. Liposomes according to claim 12, wherein said liposomes display an enhanced partition to the PEG-rich (upper) phase of a PEG:dextran aqueous two phase system in which liposomes not having covalently linked PEG moieties separate predominately to the interface or bottom phase.

15.Liposomes according to claim 13 wherein the lipid bilayers comprise a mixture of dioleylphosphatidylcholine (DOPC) and dioleylphosphatidylethanolamine (DOPE).

16.A plurality of liposomes each comprising an aqueous compartment contained by a lipid bilayer, the lipid bilayer comprising:

17.A plurality of liposomes each comprising an aqueous compartment contained by a lipid bilayer, the lipid bilayer comprising:

18.The liposomes of claims 16 or 17, wherein the liposomes show partitioning primarily to the PEG-rich phase of a PEG:dextran aqueous two phase system.

19.The liposomes of claims 16 or 17, wherein the aqueous compartment further comprises a therapeutic or diagnostic agent.

20.The liposomes of claims 16 or 17, wherein the PEG moieties covalently linked to phosphatidylethanolamine are asymmetrically disposed to the outer surface of the liposomes.

21.The liposomes of claim 20, wherein essentially all the PEG moieties covalently linked to phosphatidylethanolamine are on the outer surface of the liposomes.

22.A method of increasing the circulation half-life of a therapeutic or diagnostic agent comprising encapsulating the therapeutic or diagnostic agent in the aqueous compartment of the liposomes of claims 16 or 17.

23.The liposomes of claims 16 or 17, wherein the covalent linkage of PEG moieties does not affect the permeability barrier of the lipid bilayer.

24.A method of increasing the circulation half-life of a therapeutic or diagnostic agent comprising encapsulating the therapeutic or diagnostic agent in an aqueous compartment of a PEG-bearing liposome, wherein said liposome comprises two or more phospholipids, at least one of which is a phosphatidylethanolamine or phosphatidyl serine covalently attached to PEG.

25.A method of increasing the circulation half-life of a therapeutic or diagnostic agent comprising encapsulating the therapeutic or diagnostic agent in an aqueous compartment of a PEG modified liposome, said PEG-modified liposome comprising a PEG moiety covalently linked to an amino group in a head group of at least one phospholipid species forming the liposome.

26. The method of claim 25, wherein the PEG-modified liposome comprises two or more phospholipids, at least one of which is a phosphatidylethanolamine or phosphatidyl serine covalently linked to PEG.

27.The method of claim 24, wherein the PEG is covalently linked to phosphatidylethanolamine.

28.The method of claim 27, wherein the PEG covalently linked to phosphatidylethanolamine is asymmetrically disposed to the outer surface of the liposome.

29.The method of claim 28, wherein essentially all the PEG covalently linked to phosphatidylethanolamine is on the outer surface of the liposome.

30.The liposomes of claim 12, wherein the phospholipid species having covalently linked PEG moieties comprise phosphatidylethanolamine or phosphatidyl serine.

31.The liposomes of claim 30, wherein PEG moieties are covalently linked to phosphatidylethanolamine.

32.The liposomes of claim 31, wherein PEG moieties covalently linked to phosphatidylethanolamine are asymmetrically disposed to the outer surface of the liposomes.

33.The liposomes of claim 32, wherein essentially all the PEG moieties covalently linked to phosphatidylethanolamine are on the outer surface of the liposomes.