Carrier comprising a vitamin E derivative

Abstract

A pharmaceutical composition is provided comprising a vitamin E derivative, an anionic phopholipid-binding protein, an anionic phospholipid and a sterol. Also provided is a method for preparing a composition for delivering a cargo to a subject comprising a) forming liposomes from a vitamin E derivative, an anionic phospholipid-binding protein such as annexin, an anionic phospholipid such as phosphatidylserine and a sterol; and b) encapsulating the cargo in the liposomal composition.

Claims

1. A pharmaceutical composition comprising a vitamin E derivative, an anionic phospholipid-binding protein, an anionic phospholipid and a sterol, wherein the vitamin E derivative, anionic phospholipid and sterol form a lipid membrane and wherein the anionic phospholipid-binding protein is bound to the surface of the lipid membrane.

2. The composition according to claim 1, further comprising cargo to be transported.

3. The composition according to claim 1, wherein the vitamin E derivative is tocopherol.

4. The composition according to claim 1, wherein the anionic phospholipid-binding protein is an annexin.

5. The composition according to claim 1, wherein the anionic phospholipid is phosphatidylserine.

6. The composition according to claim 1, wherein the sterol is cholesterol or 6-ketocholestanol.

7. The composition according to claim 1, wherein the vitamin E derivative is 0.1-20% of the lipid membrane components.

8. The composition according to claim 1, wherein the anionic phospholipid is 5-20% of the lipid membrane components.

9. The composition according to claim 1, wherein the sterol is 15-30% of the lipid membrane components.

10. The composition according to claim 1 further comprising an additional phospholipid, and wherein the additional phospholipid is 30-80% of the lipid membrane components.

11. The composition according to claim 1 further comprising an additional phospholipid, and wherein the relative amount of each component is: vitamin E derivative: 0.1-20%; anionic phospholipid: 5-20%; sterol: 15-30%; additional phospholipid: 30-80% respectively, of the lipid membrane components.

12. The composition according to claim 2, wherein the cargo is alpha-bungarotoxin.

13. A method for preparing a composition for delivering a cargo to a subject comprising a) forming liposomes from a vitamin E derivative, an anionic phospholipid-binding protein, an anionic phospholipid and a sterol, wherein the anionic phospholipid-binding protein is bound to the surface of the lipid membrane; and b) encapsulating the cargo in the liposomal composition.

14. The method according to claim 13, wherein the phospholipid concentration is 20 mg/ml or more when the cargo is encapsulated.

15. The method according to claim 13, wherein the cargo is encapsulated by being present at the time of liposome formation, or by electroporation, freeze-thawing, sonication or vortexing.

16. A method of delivering an agent across the blood brain barrier or across the blood retinal barrier, comprising administering the agent to a patient in need thereof, in combination with the pharmaceutical composition according to claim 1.

17. A method of delivery of an agent to the posterior region of the eye comprising administering the agent to an eye in combination with the pharmaceutical composition according to claim 1.

18. A method of delivery of an agent to the central nervous system, comprising administering the agent to a patient in need thereof, in combination with the pharmaceutical composition according to claim 1.

19. A method of labeling the retinal vasculature of a subject, comprising administering alpha-bungarotoxin (BT) to the eye of a subject.

20. The composition according to claim 1, wherein the vitamin E derivative is tocopherol, the anionic phospholipid-binding protein is an annexin, the anionic phospholipid is phosphatidylserine, and the sterol is cholesterol or 6-ketocholestanol.

Description

DESCRIPTION OF THE DRAWINGS

(1) FIG. 1. Imaging of Lucentis 488. Imaging at baseline (marked, left) shows that there is weak inherent 488 autofluorescence, with the Argon laser set at 95% intensity. Imaging of the same eye 40 minutes after topical 488-labelled Lucentis is shown above (marked, right), at the same laser intensity as the baseline recording. Note that much greater fluorescence is seenthe whole picture is much brighter. This change in fluorescence activity reflects an increase in 488 fluorophores at the retinawhich can only be attributed to the passage of 488-labelled Lucentis to the back of the eye. This demonstrates that the 488-labelled Lucentis has been transported by the universal drug delivery system (UDDS) carrier, through the cornea, anterior & posterior chambers, lens and vitreous structures to reach the retinal nerve fibre layer of the eye.

(2) FIG. 2. shows imaging of Cy3-labelled IgG. Imaging at baseline (marked, left) shows that there is weak inherent 488 autofluorescence, with the Argon laser set at 95% intensity. Imaging of the same eye 1 hour after topical Cy3-labelled IgG is shown above (marked, right), at the same laser intensity as the baseline recording. Note that much greater fluorescence is seenthe whole picture is much brighter. Furthermore, focal areas of hyperfluorescence (white spots) are seen indicating that IgG has bound to specific structures on the retina. This change in fluorescence activity can only be attributed to the passage of the Cy3-labelled IgG to the back of the eye.

(3) FIG. 3 is a chart showing the fluorescence associated with various samples for ultracentrifugation.

(4) FIG. 4 is a chart showing the fluorescence associated with various samples for size exclusion.

(5) FIG. 5 is a chart showing the level of fluorescence associated with various samples for encapsulation.

(6) FIG. 6 shows the elution profiles of liposomes and FITC-aBT using a Superdex 75 column.

(7) FIG. 7 is a graph showing the fluorescence associated with the elution profiles of FIG. 6.

(8) FIG. 8 is a graph showing the phosphate concentration in relation to the absorbance.

(9) FIG. 9 is a graph showing the phosphatidylcholine concentration in relation to the absorbance.

(10) FIG. 10 is a graph showing the aborbance associated with 1.5 ml fractions eluted from FPLC superdex 75 and superdex 200 columns.

(11) FIG. 11 shows the elution profiles of liposomes fluorescein and -BT run down a G-50 coarse column. The elution profiles were measured by turbidity at 600 nm and fluorescence at 488 nm respectively.

(12) FIG. 12 is a chart showing the amount of total protein associated with various samples for ultracentrifugation.

(13) FIG. 13 is a chart showing the effect on encapsulation of varying the lipid concentration.

(14) FIG. 14 is a chart showing the effect of light scattering due to increasing lipid concentration.

(15) FIG. 15 is a chart showing fluorescence for different fluorescein levels with a number of detergents.

(16) FIG. 16 is images of the eye which show BT specifically labels the arteries in the retina.

(17) FIG. 17 show the successful release of BT and Annexin V-776. (A) demonstrates the release of Annexin in the presence of liposomes and (B) shows the release of encapsulated BT, whilst (C) and (D) show the effect of 1 mg/ml Annexin V-776 and 50 g/ml BT respectively, injected without the presence of liposomes.

(18) FIG. 18 is a gel showing high expression levels of Annexin V-GST.

(19) FIG. 19 shows fluorescein is unable to cross an intact blood retinal barrier and remains isolated to the vessels, as shown in figure D, E and F. It is important to highlight that the retinal blood vessels were not seen with the encapsulated material (A-C) as they are without (D-F), although there is a hint of arterial wall fluorescence in the encapsulated treated eye at 34 minutes (B). However, the encapsulated material treated eyes showed after 7 minutes focal patches or spots of fluorescence in the retina, as highlighted by the circles in figure A. The frequency of these fluorescent patches appeared to peak after 30 minutes shown in figure B, before starting to disperse after 74 minutes as shown in figure C.

(20) FIG. 20. Sodium flourescein (NaFl) leakage as a marker of BBB disruption in RT and HAT hypoxic animals. Sections from control rats injected with NaFl show no leakage in the brain tissue of the frontal (top) and parietal (bottom) cortexes. In contrast, sections from the brain of 1 or 2 days of RT-hypoxic rats show moderate monofocal leakages in the frontal and parietal cortexes. Sections from the brain of 1 or 2 days of HAT plus hypoxia show multiple focal regions of NaFl in the brain tissue. Scale bar, 100 m.

(21) FIG. 21. In order to assess the therapeutic release of cargo from encapsulated liposomes eyes a VEGF model using 500 ng per eye of recombinant human VEGF (rhVEGF165) was developed, which displayed increased blood vessel tortuosity (A, reflective image), leakage and thickening of the blood vessels, seen after a fluorescein angiography taken 24 hours after VEGF injections (B, fluorescent image). Reversal of these effects were achieved using intravitreal injections of 2.5 l of therapeutic grade Lucentis, (C-D). The Lucentis treated eye shows reduced vessel size to normal (C, reflective images), no leakage and normal vascular anatomy (D, fluorescent image) in relation to the untreated VEGF model.

(22) FIG. 22 is a number of images which are fluroescein angiograms taken at 24 (A,C) and 48 (B,D) hours from the left intravitreal VEGF-treated eyes. Firstly, it is obvious that simultaneous lucentis treatment with VEGF induced at 24 hours great inflammatory changes including severe leakage, tortuosity as vessel thickening (A,C). However, at 48 hours there is a big difference with the encapsulated lucentis treated animal (B) showing no areas of focal leakage, but a vasculitis (phlebitis), compared to massive leakage, neovascularisation and haemorrhage in the intravenous clinical grade lucentis animal.

(23) FIG. 23. The encapsulated lucentis treated eye shows reduced tortuosity, reduced vessel size to normal (A, reflective image), no leakage and more normal vascular anatomy (B, fluorescent image).

(24) FIG. 24 is a number of images which come from similar regions in the brain from an animal with intravenous encapsulated fluorescein sodium (A-B) compared to one with plain intravenous fluorescein sodium (C-D). Both fluorescence (A, C) and transmission (B, D) images are shown. Obviously, only the encapsulated material shows fluorescence (A) with no signal at all from the plain fluorescein sodium intravenous brain (C).

DETAILED DESCRIPTION OF THE INVENTION

Examples

Mechanism of Transport

(25) Physical barriers such as the corneal epithelium and the endothelium of the blood brain and blood retinal barriers are comprised of sheets of tightly adhered cells. The junctions between cells (either corneal or BBB) are regulated by protein complexes which form tight junctions restricting the movement of compounds. The selectivity of these junctions depends on the specific type of epithelial tissue but generally only hydrophilic drugs or ions of small molecular weight (100-200 Da) are able to utilise the paracellular pathway. Lipophilic drugs are able to penetrate through the endothelium of the BBB but their transit is still restricted to molecules <400 Da. In the cornea, the stroma offers an additional barrier to lipophilic drugs, preventing their penetration into the posterior of the eye.

(26) Transcytosis allows the selective transport of molecules across the epithelium. It is proposed to be the mechanism by which large molecular complexes and even viruses, which display similarities (100 nm phospholipid enclosed structures) to the liposomes used, are able to cross the epithelium.

(27) Transcytosis proceeds via either clathrin-coated pits or flask-like invaginations present on the surface of epithelial tissue, called caveolae. Clathrin-coated pits rely on receptor-mediated endocytosis, whilst caveolae are mediated by cholesterol and the membrane protein, Caveolin. The essential nature of cholesterol in caveolae correlates well with the observed sterol dependence in the carrier system.

(28) Without being bound by a particular theory, it is the inventors' belief that the composition of the invention could be stimulating/utilising the caveolar transcytosis pathway in order to cross epithelial and endothelial cells. The presence of oxidised cholesterol has also been proposed to oppose the function of normal cholesterol and inhibit caveolae function, whilst the presence of vitamin E derivatives neutralises the inhibitory effect of oxidised cholesterol. The hypothesis is further supported by recent reports identifying Annexin V as a caveolae associated protein.

(29) Crossing the BBB

(30) The global brain drug market is an underdeveloped field due mainly to the fact that the great majority of drugs do not cross the BBB in vivo, and those that do, are limited to small molecules with high lipid solubility and a low molecular mass of <400-500 Daltons (Da). The number of small molecules that have these chemical properties is <2% of all drugs. Other drugs do not cross the BBB. Without BBB solutions, >98% of all drugs that could potentially treat the brain are not developed. This has been a setback in serious disorders such as Alzheimer's disease, amyotrophic lateral sclerosis, Huntington's disease, Parkinson's disease, stroke, brain and spinal cord injury, brain cancer, and multiple sclerosis, with a paucity of new drugs including neuroprotective agents. The ability to transport large-molecules would also allow the delivery of recombinant proteins, antibodies and RNA/DNA molecules to target sites within the CNS.

(31) In recognizing that the BBB is a problem, many strategies have been used to deliver a compound from the blood to the brain. These include neurosurgery-based strategiesvery invasive, via intracerebroventricular infusion or intracerebral implants, and a temporary disruption of the tight junctions of the vascular endothelium by infusing hypertonic solutions or biologically active agents such as bradykinin has also been used, but are problematicwith risks of infection and neuropathological changes. It has been suggested that to enable transport across the BBB, lipidation may be used through conjugation of polar functional groups on a water-soluble drug with lipid-soluble moieties, or a lipid-soluble drug carrier, or drug reformulation to create a lipid soluble prodrug.

(32) The composition of the invention can be used to transport neuroprotective and other active agents across the BBB.

(33) Method of Producing the Composition.

(34) Liposome Preparation

(35) Liposomes were prepared by the lipid film rehydration method. Lipids of interest (phosphatidylcholine (PC), phosphatidylserine (PS), Cholesterol and -Tocopherol), dissolved in Chloroform:Methanol (5:1 v/v), were mixed in appropriate quantities before the solvent was removed under steady nitrogen flow. The working range was shown to be 15-30% cholesterol (with 15% preferred), 5-20% PS (with 10% preferred) and 0.1-20% vitamin E derivative (with 1% preferred), with the remainder being comprised of PC. The subsequent lipid film was rehydrated in the appropriate volume of buffer (PBSfluorophores/cargo). A stock concentration of 52 mM liposomes was prepared which was subsequently diluted to 6.5 mM for use in the UDDS.

(36) Fluorescent Labelling of Lucentis or any Protein

(37) Lucentis (ranibizumab) is a humanized monoclonal antibody fragment targeted against vascular endothelial growth factor A (VEGF-A). The protein nature of the antibody allows for conjugation of a fluorescent dye to specific amino acids. The Lucentis (10 mg/ml) was washed in 0.1 M sodium bicarbonate buffer, pH 8.3 and concentrated to approximately 2 mg/ml. The labelling reaction was performed at room temperature using 100 g alexafluor-488 (succinimidyl ester, dissolved in DMSO) labelling the protein at any primary amine sites (lysine residues and N-terminus). As the drug is still under patent and the amino acid sequence of the protein unavailable in the public domain the number of primary amine groups available for labelling is unknown. The unconjugated dye was removed by gel filtration and the labelled Lucentis eluted into 1PBS pH 7.4. Analysis of the protein by spectroscopy yielded the following results.

(38) The peaks at 280 nm (0.215 AU) and 488 nm (0.128 AU) indicated an adequate level of labelling. However, the actual protein concentration and the stoichiometry of labelling were unable to be established due to the lack of sequence information.

(39) Encapsulation by Electroporation (Preferred Method)

(40) To determine whether the drug can be delivered to the back of the eye using the composition, excess Lucentis-488 was encapsulated by the composition using the process of electroporation. The normal method of encapsulation, freeze/thawing had previously led to the denaturation of the protein. 15 l of Lucentis was added to 50 l of composition and transferred to a 0.1 cm gap Gene Pulser Cuvette, usually used for the transformation of bacterial and mammalian cells. The cuvette was subjected to an electric pulse using a BioRad Gene Pulser under the following conditions: 2.5 kV, 25 F and 200. In normal use, the electric pulse allows small molecules (commonly plasmid DNA) to cross the phospholipid membrane of cells. The method was judged to be likely to be a successful alternative encapsulation method to freeze/thawing as liposomes have structural similarities to cell membranes.

(41) Encapsulation by Freeze-Thawing

(42) Encapsulation via freeze/thawing was performed by freezing the carrier, in the presence of the cargo, in liquid nitrogen and then rapidly thawing the sample under a steady flow of hot water. The process was repeated 10 times to provide optimal encapsulation within a reasonable time scale. The process works well for simple molecules but can prove denaturing to complex molecules such as proteins.

(43) Removal of Unincorporated Fluorophores

(44) Unincorporated fluorophores were removed by ultracentrifugation for 45 minutes at 60,000g. The supernatant was discarded and the pellet resuspended in fresh PBS.

(45) Addition of Anionic Phospholipid Binding Proteins

(46) Prior to administration the loaded liposomes were mixed with the anionic phospholipid binding protein annexin V so that a final concentration of 5 mg/ml liposomes and 300 g/ml annexin V was achieved. Variation of the concentrations of the liposomes and annexin can be used to achieve differing levels of delivery. A maximal concentration of 2 mg/ml of annexin can be used due to stability issues, whilst the doses of liposomes above 20 mg/ml would increase the embolisms.

(47) Imaging In Vivo

(48) Anaesthetized animals (Dark Agouti rats) were imaged using a confocal scanning laser ophthalmolscope (cSLO). A fluorescence baseline image was taken of each animal. Briefly the animal is positioned before the cSLO so that the interior of the eye is imaged. An Argon laser wavelength of 488 nm is focused into a small spot and scanned across the retina by a pair of mirrors to excite fluorophores at that wavelength. The resulting fluorescence is optically focused onto a confocal aperture which has the effect of excluding unwanted fluorescence in planes above or below the depth plane of interest. Imaging was performed at the plane of the retinal nerve fibre layer. The baseline image therefore records only inherent 488 autofluorescence, with the Argon laser set at 95% intensity.

(49) After the baseline image was recorded, 10 l of the encapsulated 488-labelled Lucentis, was administered topically onto the rat eye. 40 minutes following topical treatment the eye was imaged again at the same settings as the baselinei.e. argon laser wavelength of 488 nm 95% intensity. The image taken then records inherent 488 autofluorescence (as the baseline) PLUS any 488-labelled Lucentis reaching the plane of the retinal nerve fibre layer.

(50) Imaging Cy3-Labelled IgG

(51) After a baseline image was recorded, 10 l of the invention encapsulated Cy3-labelled IgG, was administered topically onto rat eye. Cy3 is a well known fluorescent dye which can be detected after argon laser excitation. One hour following topical treatment the eye was imaged again at the same settings as the baselinei.e. argon laser wavelength of 488 nm 95% intensity. The image taken then records inherent 488 autofluorescence (as the baseline) PLUS any Cy3-labelled IgG reaching the plane of the retinal nerve fibre layer.

(52) This result illustrates the potential of the system to deliver functional antibodies to the posterior of the eye, thereby demonstrating potential in the delivery of antibody derived therapeutic agents.

(53) A. Development of Universal Drug Delivery System (UDDS) Technology

(54) 1. Refinement of UDDS for Delivering Cargo

(55) Components for successful cargo delivery include vitamin E derivatives, Phosphatidylcholine, Phosphatidylserine, Cholesterol and Annexin. In the refinement of the process to allow different cargo/constituent ratios, methods of purification and encapsulation were optimised as described below:

(56) (i) Removal of Unincorporated Material

(57) The primary challenge in assessing successful encapsulation is producing reliable and efficient methods for the removal of unincorporated material. Several methods, outlined below, have been screened.

(58) a) Ultracentrifugation

(59) Ultracentrifugation of Liposomes at 60,000 g will cause liposomes to pellet, drawing down any encapsulated material with them. Individual molecules however are unable to pellet at this speed and remain in the supernatant. The supernatant is removed and the pellet resuspended in fresh buffer. This process of washing the sample is repeated until the supernatant no longer contains significant quantities of unencapsulated material.

(60) Each wash is a 45 minute step and even after 5 washes there are noticeable quantities of unencapsulated material being removed (see FIG. 3). The in FIG. 3 signifies where the fluorescence reading is saturated.

(61) b) Size Exclusion

(62) Columns

(63) The sample is loaded onto a column which separates molecules based on their size. Macromolecular structures such as liposomes pass rapidly down the column whilst the progress of smaller molecules is retarded. For Fluorescein, for example, disposable Pd-10 columns were used which have a molecular cut off of 5,000 da.

(64) Optimisation of these methods have shown that whilst one pass down the column removes the majority of the unencapsulated material, a second pass is recommended for improved accuracy (see FIG. 4).

(65) Dialysis

(66) Samples were loaded into dialysis tubing containing pores which correspond to an appropriate molecular weight. Particles larger than this cut off, such as liposomes, are retained within the tubing, whilst particles smaller than the pores are able to diffuse freely into a larger bulk solution. Dialysis was performed over night and even after two dialysis steps there was considerable background fluorescence.

(67) b) Encapsulation

(68) Having decided the most appropriate method for purification, different methods for the encapsulation of low molecular weight compounds were also investigated. These methods included: (i) Making the liposomes in the presence of the compound; (ii) Vortexing; (iii) Sonication; (iv) Freeze-Thawing; and (v) Electroporation.

(69) Due to the highly ionic nature of the fluorescein it proved incompatible with electroporation which relies on a low salt environment.

(70) The vortexed sample produced very little encapsulation, whilst sonication, freeze-thawing and preformed liposomes all produced noticeable encapsulation (see FIG. 5).

(71) 2. Optimizing Composition of Matter for Different Cargos

(72) Grading cargo by molecular weight has identified 3 bands: low molecular weight fluorescent compound, e.g. Sodium fluorescein (Mwt: 376.28 da), medium MW, e.g. BT (8,000 da) and Lucentis (48,000 da), and high MW, e.g. Avastin (149,000 da.). One advantage of using the size exclusion column system is that it is possible to use different resins with specific molecular weight cut offs.

(73) For the separation of unencapsulated BT, the inventors opted to trial a Superdex 75 column which is able to separate proteins in the range of 3,000 to 70,000 da. Initially fluorescein encapsulated liposomes were used to spike the column and identify the fraction in which the liposomes exit the column. The liposome peak is indicated by the box in FIG. 6 (top). The column was then loaded with a sample of FITC-aBT. The box in FIG. 6 (bottom) indicates the fraction in which the BT exits the column.

(74) The results in FIG. 7 show that a Superdex 75 column provides sufficient resolution for the separation of encapsulated and unencapsulated BT.

(75) Of special interest, will be the ability to separate liposomes from large molecular weight proteins such as antibodies, particularly with regards to therapeutic agents such as Lucentis (48,000 da) and Avastin (149,000 da). This should be achievable using a Superdex 200 column which operates over the range of 10,000 to 600,000 da.

(76) In adapting the size exclusion column principal from the gravity fed Pd10 columns to the FPLC run Superdex columns, the inventors encountered some problems with sample loss. This could be due to a variety of reasons such as liposomal rupture. To be able to investigate this, the inventors adapted an assay for monitoring phosphate levels to identify Phospholipid levels.

(77) Phospholipid Quantification

(78) Samples can be processed by 5M HCl and incubated with a mixture of Rhodamine B and Ammoniumheptamolybdate in order to provide an absorbance value corresponding to the phosphate concentration. The amount of phosphate in the solution directly reflects the levels of phospholipid present.

(79) The inventors optimised the phospholipid assay. FIG. 8 shows the absorbance at 555 nm and the corresponding phosphate concentration.

(80) This method has been adapted to detect the phosphate groups within the head of phospholipid molecules shown below. This will enable us to identify which fractions exiting the size exclusion column contain the encapsulated liposomes. It can thus be used to detect encapsulation efficiency (see FIG. 9).

(81) FPLC Superdex-75 and Superdex-200 Columns

(82) Liposomal samples were run down the FPLC superdex 75 and superdex 200 columns and the elute collected in 1.5 ml fractions. None of the fractions produced the expected liposomal pellet when ultracentrifuged and the phospholipid assay showed neglible amounts of phosphate in comparison to a buffer control as shown in FIG. 10.

(83) The absorbance values represented in FIG. 10 are below the lower limit of sensitivity and confirm that the FPLC column is incompatible with liposomal purification.

(84) G50 Coarse Resin

(85) The limitation with the gravity fed PD-10 columns was the molecular weight cut off (MWCO) of 5,000 Da, which meant that is would not be able to resolve the difference between any compounds greater than 5,000 Da. Other resins such as the G-50 coarse resin which have a higher MWCO of 30,000 Da and therefore should be able to separate small proteins from liposomes without causing liposomal rupturing.

(86) A more accurate and less involved method for determining the lipid concentration was devised based upon the turbidity of the solution, whereby the ability of liposomes to scatter light at 600 nm was used to determine their concentration.

(87) Samples of Liposomes, Fluorescein and -BT were passed down the G-50 coarse column individually and there elution profile measured by turbidity at 600 nm and fluorescence at 488 nm respectively.

(88) The ability of the column to separate Fluorescein, as shown in FIG. 11, indicates that the column is correctly packed and not leaking, however the resolution between liposomes and BT does not provide sufficient resolution. This problem could potentially be resolved by using a longer column. Due to the expensive nature of large chromatography columns and the subsequent amounts of expensive resin, it was decided to revisit ultracentrifugation as a method of separation.

(89) Ultracentrifugation

(90) The problems encountered when trying to use ultracentrifugation with fluorescein were predominantly concentration dependent. The sheer quantity of dye present restricted the efficiency of the process but, as proteins such as -BT and Lucentis are to be used in much lower concentrations, fewer washes should be required.

(91) Initial encapsulation protocols were preformed using sonication and the fluorescence of the pellet corrected for final volumes to give an amount of total encapsulated protein.

(92) It was found that ultracentrifugation was able to remove the free protein and that a value of around 10% encapsulation was being achieved with approximately 1 g of the 10 g initially added residing in the pellet (see FIG. 12).

(93) Based on the results of our in vivo -BT titration, 5 l of 50 g/ml was the optimal dose, representing an addition of 0.25 g. From these encapsulation results we have 1 g in 200 l and therefore would require 100% transmission of the 50 l eyedrop.

(94) Encapsulation Efficiency

(95) In order to produce a more realistic transmission target, the inventors attempted to increase the encapsulation efficiency further. There are three possible methods for doing this: Increase the initial -BT concentration Increase the Lipid concentration Apply alternative methods of encapsulation such as Freeze thawing and Electroporation.

(96) Increasing the concentration of the -BT only showed a slight enhancement in encapsulation.

(97) In FIG. 13, the difference moving from A&B to C&D to E&F is more pronounce than the differences between the two bars of A&B, C&D or E&F indicating that increasing the lipid concentration resulted in clear increases to the degree of encapsulation.

(98) Light Scattering

(99) A precise value for the degree of encapsulation is difficult to calculate, however, as the increased light scattering due to the turbidity of the high lipid solution interfers with the fluorescence readings, reducing the fluorescent signal (see FIG. 14).

(100) The results in FIG. 14 show that even at the lower lipid values light scattering plays a significant effect of the fluorescence, suggesting that the encapulsation efficiencies could be greater than the 10% first suggested.

(101) Liposomal Disruption

(102) In order to acurately determine the encapsulation values Fluorescein was incubated with both buffer and liposomes alone and in the presence of a variety of detergents including SDS, NP40, Tween-20 and Triton X-100.

(103) In FIG. 15, the presence of detergent was shown to reduce the overal fluorescent signal in samples containing buffer and fluorescein, an effect which was most pronounced an higher fluorescent intensities. The presence of detergent, SDS in particular reduced the loss of fluorescent signal due to lipid light scattering. By using 1% SDS and restricting the fluorescent yields measured to below 200000 AU, the inventors were able to accurately measure the encapsulation efficiency of the cargo.

(104) B. In Vivo Testing of UDDS

(105) 1. Novel Method for Screening Successful Delivery to Retina

(106) In order to assess the efficacy of different UDDS compositions, the inventors identified a novel label Alpha-bungarotoxin (BT) to assess retinal delivery. BT (obtained from Sigma (T9641)) specifically labels the arteries in the retina, as shown in FIG. 16, and thereby represents an excellent marker.

(107) 2. Release of Cargo

(108) To access the release of encapsulated material, liposomes containing encapsulated BT and annexin V-776 were administered via Intravitreal injection and the eyes were imaged after 3 hours.

(109) The results in FIG. 17 show the successful release of BT and Annexin V-776. (A) demonstrates the release of Annexin in the presence of liposomes and (B) shows the release of encapsulated BT, whilst (C) and (D) show the effect of 1 mg/ml Annexin V-776 and 50 g/ml BT respectively, injected without the presence of liposomes. The signal strength of the encapsulated BT is reduced in comparison to the free BT but sufficient release is seen to elicit specific labelling of the arteries.

(110) 3. Annexin Production

(111) The inventors obtained a codon optimised sequence for annexin. As FIG. 18 shows, the new system expresses high levels of Annexin V-GST and allows the generation of sufficient quantities of Annexin for intravenous delivery.

(112) 4. Intravenous DeliveryImplications for Blood Brain Barrier (BBB) & Blood Retinal Barrier (BBB)

(113) A. Blood Retinal Barrier (BRB)

(114) a) Intravenous Delivery of Fluorescein Sodium to the Retina

(115) The primary aim was to show the successful delivery of encapsulated material across the blood-brain and blood-retinal barriers. Having established the optimised encapsulation and release for the UDDS, the ability of UDDS to deliver fluorescein sodium to the retina and brain were assessed.

(116) 300 l of 5% fluorescein sodium or 1 ml of 0.2% encapsulated fluorescein sodium, were administered to the rat through the tail vain and the retina monitored. Free fluorescein was removed by use of ultracentrifugation and PD-10 columns as outlined earlier.

(117) Fluorescein is unable to cross an intact blood retinal barrier and remains isolated to the vessels, as shown above in FIG. 19 (D, E and F). It is important to highlight that the retinal blood vessels were not seen with the encapsulated material (A-C) as they are without (D-F), although there is a hint of arterial wall fluorescence in the encapsulated treated eye at 34 minutes (B). However, the encapsulated material treated eyes showed after 7 minutes focal patches or spots of fluorescence in the retina, as highlighted by the circles in figure A. The frequency of these fluorescent patches appeared to peak after 30 minutes shown in figure B, before starting to disperse after 74 minutes as shown in figure C. They were absent from the non-encapsulated eyes. These focal patches of fluorescence are believed to be intact liposomes containing fluorescein. The localisation of these patches of fluorescence suggests the delivery of intact liposomes to the retina. It is also confirms that only the encapsulated fluorescein crossed the BRB. The appearance, of fluorescent patches or monofocal leakages of sodium flourescein is similar to that that has been described when BBB disruption occurs in the brain (see FIG. 20 from Natah, S. S. et al. J Appl Physiol 107: 1348-1356 2009).

(118) b) Development of a VEGF Model to Assess Therapeutic Release

(119) In order to assess the therapeutic release of cargo from encapsulated liposomes in eyes, a VEGF model using 500 ng per eye of recombinant human VEGF (rhVEGF165) was developed. FIG. 21 shows an eye with increased blood vessel tortuosity (A, reflective image), leakage and thickening of the blood vessels, seen after a fluorescein angiography taken 24 hours after VEGF injections (B, fluorescent image). Reversal of these effects were achieved using intravitreal injections of 2.5 l of therapeutic grade Lucentis, (C-D). The Lucentis treated eye shows reduced vessel size to normal (C, reflective images), no leakage and normal vascular anatomy (D, fluorescent image) in relation to the untreated VEGF model.

(120) c) Intravenous Delivery Lucentis to Retina

(121) Due to the large molecular weight of Lucentis, it is unable to cross the blood retinal barrier making it unsuitable for intravenous administration. This study was designed to see if encapsulated lucentis (lucentis UDDS) could be delivered successfully intravenously to the retina, through the BRB. Animals were prepared with one eye VEGF treated and the contra lateral eye left untreated. At the time of VEGF treatment the rats also received an intravenous injection of either 1 ml of encapsulated Lucentis (A,B) or 300 l of clinical grade Lucentis (C,D). A larger volume of the encapsulated material was administered to allow for the reduced Lucentis concentration due to the encapsulation efficiency.

(122) FIG. 22 shows fluroescein angiograms taken at 24 (A,C) and 48 (B,D) hours from the left intravitreal VEGF-treated eyes. Firstly, it is obvious that simultaneous lucentis treatment with VEGF induced at 24 hours great inflammatory changes including severe leakage, tortuosity as vessel thickening (A,C). However, at 48 hours there is a big difference with the encapsulated lucentis treated animal (B) showing no areas of focal leakage, but a vasculitis (phlebitis), compared to massive leakage, neovascularisation and haemorrhage in the intravenous clinical grade lucentis animal. This strongly suggests that encapsulated lucentis by 48 hours has been released, with a therapeutic effect of reversing dramatic changes. This would imply a sustained/delayed release of lucentis over the period, with succesful BRB passage.

(123) The effect of giving encapsulated lucentis intravenously 24 hours after intravitreal VEGF was also assessed (see FIG. 23). After a fluorescein angiography taken 28 hours after VEGF injections (B, fluorescent image). reversal of VEGF effects were achieved. The encapsulated lucentis treated eye shows reduced tortuosity, reduced vessel size to normal (A, reflective images), no leakage and more normal vascular anatomy (B, fluorescent image).

(124) This all suggests that encapsulated Lucentis has crossed BRB and has been therapeutic.

(125) B. Blood Brain Barrier

(126) Intravenous Delivery Fluorescein to the Brain

(127) 300 l of 5% fluorescein sodium or 1 ml of 0.2% encapsulated fluorescein sodium, were administered to the rat through the tail vain and the retina monitored. Free fluorescein was removed by use of ultracentrifugation and PD-10; columns as outlined earlier. After 2 hours; under general anaesthesia, the animals underwent fixation perfusion with 4% paraformaldehyde. Brains were stored in 4% paraformaldehyde overnight, following which 800 um slices were cut and examined under fluorescent confocal microscopy.

(128) Fluorescein sodium is unable to cross an intact blood brain barrier; this is the basis of several experiments, including the study mentioned previously by Natah, S. S. et al. J Appl Physiol 107: 1348-1356 2009.

(129) The images in FIG. 24 come from similar regions in the brain from an animal with intravenous encapsulated fluorescein sodium (A-B) compared to one with plain intravenous fluorescein sodium (C-D). Both fluorescence (A, C) and transmission (B, D) images are shown. Obviously, only the encapsulated material shows fluorescence (A) with no signal at all from the plain fluorescein sodium intravenous brain (C). This confirms the fact that encapsulation allows the fluorescein sodium to cross the BBB.

(130) Furthermore, only the encapsulated material had patches of fluorescence evident in the substance of the brain as shown in representative fluorescent images (E,G) with corresponding transmission images (F, H respectively). The appearance of fluorescent patches or monofocal leakages of sodium flourescein is similar to that that has been described when BBB disruption occurs in the brain, as discussed earlier (Natah, S. S. et al. J Appl Physiol 107: 1348-1356 2009).