WO2004032970A2

WO2004032970A2 - Carriers attached to blood cells

Info

Publication number: WO2004032970A2
Application number: PCT/US2003/032502
Authority: WO
Inventors: Samir Mitragotri; Elizabeth Chambers
Original assignee: Individual
Current assignee: Individual
Priority date: 2002-10-10
Filing date: 2003-10-10
Publication date: 2004-04-22
Anticipated expiration: 2005-04-10
Also published as: WO2004032970A3; AU2003300043A1; AU2003300043A8

Abstract

The present invention is a method by which carriers can remain in vascular circulation for extended periods of time by binding to the outer membrane of red blood cells (RBCs). Freely circulating carriers are cleared from circulation within a few minutes to hours, depending on their size and surface characteristics. The adhesion between the RBC and carriers protects the carrier from vascular clearance by phagocytic cells. Sustained circulation of carriers is used for drug delivery. Alternatively, sustained carriers may also be used as contrast agents that circulate for extended periods of time and used for diagnostics.

Description

CARRIERS ATTACHED TOBLOOD CELLS

[0001] This application claims the benefit of provisional application No. 60/417,576 filed on October 10, 2002, which is incorporated herein by reference.

FIELD OF THE INVENTION

[0002] The invention relates generally to the field of drug delivery, diagnostics, and imaging and more specifically to intravascular administration of carriers.

BACKGROUND OF THE INVENTION

Description of the Related Art

[0003] A carrier is defined as a container or device that is used to store and deliver an agent of interest. It can be made up of polymers, lipids, proteins, and/or polysaccharides. In certain cases, the carrier may be a therapeutic agent or the contrast agent.

[0004] Carriers containing suitable therapeutic agents can be injected into vascular circulation to achieve sustained release of therapeutic agents in the bloodstream. Carriers offer several advantages over traditional injections, including sustained drug concentrations with less frequent injections. Alternatively, carriers can be used as contrast agents for vascular imaging. Traditional imaging agents remain in circulation for only a few minutes making it difficult to image organs other than the liver, the organ where imaging agents are sequestered.

[0005] Carriers • in vascular circulation- are rapidly cleared by the reticulo-endothelial system (RES) of the body. The RES system, includes the liver, spleen, bone marrow, and the pulmonary intravascular macrophages [Brain, J.D.M., Ramon M.; Decamp, Malcom M.; Warner, Angeline E., Pulmonary intravascular macrophages: their contribution to the mononuclear phagocyte system in 13 species. American journal of physiology-lung cellular and molecular physiology, 1999. 276(1): p. 1,146-1,154.]. The speed of clearance of carriers has been related to their size, surface properties, and opsonization. Opsonization refers to the absorption of proteins that trigger particle recognition and removal [Storm, G.B., Sheila O.; Daemen, Toos; Danilo, D. Lasic, Surface modification of nanoparticels to oppose uptake by the mononuclear phagocyte system. Advances Drug Delivery Reviews, 1995. 17: p. 31-48; Roser, M.F., Dagmar; Kissel, Thomas, Surface- modified biodegradable albumin nano- and microspheres. II: effect of surface charges on in vitro phagocytosis and biodistribution in rats. European Journal of Pharmaceutics and Biopharmaceutics, 1998. 46: p. 255-263; Ogawara, K.-i.F., Kentaro; Takakura, Yoshinobu; Hashido, Mitsuru; Higaki, Kazutaka; Kimura, Toshikiro, Surface hydrophobicity of particles is not necessarily the most important determinant in their in vivo disposition after intravenous administration in rats. Journal of Controlled Release, 2001. 77: p. 191-198; Ogawara, K.-i.Y., Minoru ; Kubo, Jun-ichi; Nishikawa, Makiya; and Y.H. Talcakura, Mitsuru; Higaki, Kazutaka ; Kimura, Toshildro, Mechanisms of hepatic disposition of polystyrene microspheres in rats: Effects of serum depend on the sizes of microspheres. Journal of Controlled Release, 1999. 61: p. 241-250.].

[0006] Polymeric particles, in the context of intravenous carriers, have been extensively studied for sustained drug release, vehicles for site specific targeting, or agents for blood- pool imaging. Particles offer, many advantages over traditional methods of drug delivery. Particles can protect sensitive therapeutic agents from degradation and clearance in the vasculature while maintaining a steady plasma concentration. Furthermore, sustained drug concentrations reduce the dose required of the drug, thus reducing the possibility of side effects. However, applications of polymeric particles have been limited due to short circulation times. The particles are rapidly removed from circulation after intravenous injection [Moghimi, S.M.H., A. Christy; Murray, J. Clifford, supra].

[0007] In order to improve circulation lifetimes of polymeric particles, a wide range of sizes, materials, and surface characteristics have been explored. It has been shown that particle diameter is of primary importance in determining in vivo biodistribution. Particles greater than 5μ,m will predominantly be trapped in the capillary beds of the lungs due to physical entrapment. Particles smaller than 5μm will be cleared by phagocytosis in the organs of the RES, also known as the mononuclear phagocyte system (JMPS) [Simon, B.H.A., Howard, Y.; Gupta, Parddp K., Circulation Time and Body Distribution of 14C-Labeled Amino-Modified Polystyrene Nanoparticles , in Mice. Journal of Pharmaceutical Sciences, 1995. 84(10): p. 1249-1253.].

[0008] The specific distribution of particles in the MPS further depends on the size and surface characteristics of the particles. Particles larger than 200nm can be physically filtered and removed by phagocytosis within the spleen. The spleen is composed of a meshwork of filaments, which form slits possessing widths around 200-500nm. Thus, smaller particles are better for avoiding spleen clearance. However, particles smaller than lOOnm tend to be trapped in the fenestrations in the hepatic sinusoidal endothelium, which are between 100-150nm. Therefore, particles should be between 100-200nm in maximum dimension to achieve long circulation. However, even particles m this size range fail to exhibit prolonged circulation times [Moghimi, S.M.H., A. Christy; Murray, J. Clifford, supra.].

[0009] Particle size also partially determines the particle opsonization and dysopsonization. Dysopsonization refers to the absorption of proteins that prevent particle recognition in the RES. Smaller polystyrene particles (50nm) have shown decreased in vitro hepatic uptake in the presence of serum, where as larger particle (500nm) have shown increased clearance in the presence of serum. The results suggest that smaller particles attract dysopsonins while larger particles attract opsonins. This was verified by the addition of bovine serum albumin (BSA) to the medium, a dysopsonin, and anti-C3 antibody, where C3 antibody is an opsonin. The smaller particles showed decreased clearance with only 0.5% BSA in solution. Larger particle required at least 7% BSA in solution and only showed a moderate decrease in clearance. On the other hand the addition of anti-C3 antibody decreased the clearance of larger particles by 48% [Ogawara, K.-i.Y._; supra]. However, particle size has little effect when plasma is used instead of serum. Plasma causes an increase in clearance of both the 50nm and 500nm particles. This suggests that blood coagulation factor, which is only present in plasma, is an important factor for particle clearance. In support, it was shown that fibrinogen, a blood coagulation factor, increased clearance of both small and large particles in the absence of serum or plasma [Ogawara, K.-i.Y., Minoru; Takakura, Yoshinobu; Hashida, Mitsuru; Higaki, Kazutaka; Toshikiro, Kimura, Interaction of polysytrene microspheres with liver cells: roles of membrane receptors and serum proteins. Biochimica et Biophysica Acta, 1999. 1472: p. 165-172.]. Particle size alone cannot control opsonization. [0010] Surface modification is the main tool in prolonging circulation times. A wide variety of polymers and proteins have been used to modify the surface of particles through physical or chemical attachment. It has been shown that surface hydrophobicity, charge, and structure play an important role in determining the circulation life of particles [Moghimi, S.M.H., A. Christy; Murray, J. Clifford, supra; Storm, G.B., Sheila O.; Daemen, Toos; Danilo, D. Lasic, supra; and Ogawara, K.-i.F., Kentaro; Takakura, Yoshinobu; Hashido, Mitsuru; Higaki, Kazutaka; Kimura, Toshikiro, supra]. The surface characteristics of the particle determine opsonin binding. Surface bound polymers or proteins provide a repulsive steric barrier to the flocculation of particles and the adsorption of biological components. The effectiveness of the steric barrier depends on the chemical nature of the polymer or protein, the polymer length and structure, the polymer surface coverage on the particle, and the strength of adhesion between the particle and polymer or protein [Storm, G.B., Sheila O.; Daemen, Toos; Danilo, D. Lasic, supra].

[0011] Proteins are used to coat the surface of particles in hopes of mimicking the protein coat on the body's own cells and thus avoid recognition in the RES. Thus, glycoproteins, glycolipids, or polysaccharides, which are naturally present on cells or pathogens, are chosen for surface modification. Sialic acid is an essential component of both cell membranes and pathogenic envelopes [Moghimi, S.M.H., A. Christy; Murray, J. Clifford, supra]. Sialilated proteins bind factor H, which is a cofactor for the cleavage of C3b and a blocker for the formation of C3 convertase. This promotes the inactivation of the alternative complement system [Goldsby, R.A;K., Thomas J.; Osbome, Barbara A., The Complement System, in Kuby Immunology, N. Folchetti, Editor. 2000, W. H. Freeman and Company: New York. p. 329-350.]. Polysialic acid has been used to increase the circulation time and reduce the anitgenicity of asparaginase. The extent of shielding is directly related to the amount of sialilation; the greater the sialilation, the better the screening [Fernandes, A.LG., Gregory, JJTbe effect of polysialylation on the immunogenicity and antigenicity of asparaginase: implication in its pharmacokinetics. International Journal of Pharmaceutics, 2001. 217: p. 215-224.]. However, the use of polysialic acid has not been tested for increasing the circulation life of particles. This may be attributed to low surface coverage achieved due to the extremely hydrophilic nature of polysialic acid. Other proteins, which prevent complement activation, have shown improved circulation time for particles. Heparin has been shown to increase circulation times and reduce complement activation [Passirani, C.B., Gillian; Devissaguet, Jean-Philipe; Labarre, Denis, Interactions of nanop articles bearing heparin of dextran covalently bound to poly(methylmethacrylate) with the complement system. Life Sciences, 1998. 62(8): p. 775-785.]. However, prolonged circulation of particles has not been achieved through the addition of proteins or peptides that mimic cell surfaces. This is most likely due to low surface coverage, desorption, or conformational change of the protein [Moghimi, S.M.H., A. Christy; Murray, J. Clifford, supra.].

[0012] The use of synthetic polymers, such as polyoxamines, polyoxamers, and polyethylene glycol, has been shown to increase the circulation life of particles. These polymers are attached to the surface of particles by passive adsorption or covalent attachment. Passive adsorption is used for particles that have block hydrophobic and lydrophilic- regions. The common formulation for polyoxamines and polyoxamers is an _; -B-A formulation in which the central block is a hydrophobic polypropylene region and the A blocks are hydrophilic polyoxyethylene regions [Moghimi, S.D., SS, Innovations in avoiding particle clearance from blood by kupffer cells: Cause for reflection. Critical Reviews in Therapeutic Drug Carrier Systems, 1994. 11(1): p. 31-59.]. PEG di-block polymers can also be absorbed onto the surface of particles by using an appropriate hydrophobic block. PEG is commonly covalently attached to the surface of particles. The effectiveness of the polymer coating depends strongly on the surface density and length of hydrophobic region. Longer hydrophobic regions and denser surface coverage lead to better circulation half-lives. However, reported half-lives of modified particles are generally on the order of several hours [Moghimi, S.M.H., A. Christy; Murray, J. Clifford, supra.]. Ultimately the particles are recognized and cleared. Additionally, coated particles elicit some form of an immune response. Repeated administration of coated particles over a short period of time leads to increased opsonic activity in the plasma and faster clearance of the particles [Moghimi, S.D., SS, supra].

Red Blood Cells: RBCs are terminally differentiated cells, which are responsible for O₂ and CO₂ transport. They are approximately 5-7 μm in diameter, 1-2 μm in height and are discoid in shape. RBCs carry a net negative charge at physiological pH. They have a flexible membrane skeleton made primarily of spectrin and actin proteins. The JRBC membrane is anchored to the membrane skeleton via band HI and glycophorin A proteins [Agre, P.P., John C, ed. Red Blood Cell Membranes: Structure, Function, Clinical Implications. Hematology, ed. K.M.S. Brinkhous, Sanford A. Vol. 11. 1989, Marcel Dekker: New York.].

[0013] The RBC membrane is composed of 20% phospholipids, 16% cholesterol, 4% σlycoli idSi and various proteins, glycoproteins, and carbohydrates. There are four primary phospholipids; 28% phosphatidylcholine, 26% phosphatidylethanolamine, 25% sphingomyelin, and 13% phosphatidylserine, which show asymmetry across the bilayer plane. The majority of phosphatidylcholine and sphingomyelin (65-75% and 85+%, respectively) are located on the extracellular side of the membrane, whereas 80-85% and 96+% of the phosphatidylethanolamine and phosphatidylserine are located on the cytosolic side [Alberts, B.B., Dennis; Lewis, Julian; Raff, Martin; Roberts, Keith; Watson, James D, Molecular Biology of the cell. Vol. 3. 1994, New York: Garland Publishing, Inc.]. The asymmetry of phosphatidylserine is due to its interaction with spectrin. Translocation of phosphatidylserine must therefore be catalyzed by flippases [Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D; Montreuil, J, Cellular and molecular mechanisms of senescent erythrocyte phagocytosis by machrophages . Biochimie, 1998. 80: p. 173-195.].

[0014] Phospholipids are not the only membrane components that demonstrate asymmetry. Glycolipids and proteins are also specific to the extracellular or cytoloic side. Glycolipids are located entirely on the extracellular side of the membrane. They are composed of a ceramide base that is glycated by various carbohydrates. The glycan extends into the aqueous phase and carries some of the RBC antigents such as A, B, H, P^k, Pi, Ii, and Le^a. The glycan structure varies considerably but can be divided into two main subgroups, neutral and sialylated glycolipids [Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D; Montreuil, J, supra]. The proteins of the RBC membrane are transmembrane, integral, or peripheral in nature. There are several hundreds .of membrane proteins but only 10-42 of great abundance [Alberts, B B., et. al., supra]. The most abundant membrane proteins are glycophorin A, band IH, and band 4.5.

[0015] Glycophorin A is a 31 Da transmembrane protein present at 5x10 to 1x10 copies per cell. Glycophorin A is heavily glycosilated, with carbohydrates making up 60% of it mass. Glycophorin A exclusively carries the MN-blood group on five N- terminal amino acids [Niitala, J.J., J, The red cell surface revisited. Trends in Biochemical Science, 1985. 10: p. 392-395.]. Glycophorin A has 32 sialic acid groups on 16 oligosaccharide side chains, this accounts for 80% of the total 20-40 million sialic acid residues of the RBC membrane [Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D; Montreuil, J, supra&nd Donath, E., "Hairy surface layer" concept of electrophoresis combined with local fixed surface charge density isotherms: application to human erythrocyte electrophoretic fingerprinting. Langmuir, 1996. 12: p. 4832-4839.]. Glycophorin A's function is unknown. It is believed to be responsible for preventing non-specific cell adhesion through charge repulsion. However, studies of individuals lacking glycophorin A did not reveal any observable adverse affects [Juliano, R., The proteins of the erythrocyte membrane. Biochimica et Biophysica Acta, 1973. 300: p. 341-378.].

[0016] Band JUH is an integral protein anchored to the cytosckeleton via ankyrin. Band IE functions as an anion transporter for the passive exchange of HCO₃ ^" for Cl^" and thus increase the CO carrying capacity of the RBC [Bratosin, D.M., J,; Tissier, JP; Estaquier,

Huart, JJ; Ameisen, JC; Aminoff, D; Montreuil, J, supra]. Band JUJuJ is present at 1.2 minon copies per ceπ ana is tnougnt to exist as aimers ana tetramers. ϋan ill is heavily glycosylated and carries determinants for the ABO blood group [Niitala, J.J., J, supra]. Aggragation of band JJI is believed to play a role in senescent cell removal.

[0017] Band 4.5 is a poorly characterized protein which is present at 0.7 million copies per cell [Viitala, J.J., J, supra.]. It is responsible for glucose transport into the RBC through an insulin-insensitive water-filled channel. Band 4.5 is structurally very similar to band HI with similar glycans [Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D; Montreuil, J, supra].

[0018] RBCs circulate for 110-120 days before they are removed from circulation at a rate of 5 million per second primarily by macrophage endocytosis. The exact mechanism of senescent cell removal is unknown [Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D; Montreuil, J, supra]. RBCs exhibit minor changes over their life span. Their volume decreases and their density increases. The protein and lipid content is altered with age and the asymmetry of the phospholipid membrane is disrupted. A decrease in the sialic acid residues is observed and there is an increased appearance of phosphatidylserine in the extracellular side of the membrane. There are several different theories for the cause of RBC removal.

[0019] It has been suggested that a change in protein distribution is responsible for senescent cell removal. It has been shown that band Jul aggregates upon destruction of its a binding to ankyrin, which is caused by the denaturation, aggregation and binding of hemoglobin to the cytoplasmic side of the membrane. Band JJJJH aggregation also results in increased rigidity, of the cell, and fragility of the .membrane. It has been demonstrated that hemoglobin denatures and aggregates in aged and oxidatively damaged cells. However, it has not been shown that the aggregation of membrane proteins results in clearance. It has been suggested that this aggregation results in autoantibody binding but it has not been experimentally verified [Agre, P.P., John C, supra]. The existence of an autoantibody, which binds to band HI and elicits phagocytosis, has been demonstrated. The autoantibody is only present on aged cells but is not related to band IH clustering. However, erythrophagocytosis can occur in the absence of antibodies [Brain, J.D.M., Ramon M.; Decamp, Malcom M.; Warner, Angeline E., supra].

[0020] Another possible mechanism is the loss of sialic acid residues. It has been demonstrated that aged erythrocytes contain 10% fewer sialic acid residues due to enzymatic cleavage and glycoprotien and glycolipid loss during budding. The loss of sialic acid results in increased macrophage clearance due to the exposure of β-galactosyl residues. The mechanism of sialic acid loss is unknown.

[0021] The loss of membrane asymmetry has also been proposed as a marker for senescent cell removal. The appearance of phosphatidylserine in the outer leaflet of the membrane has been shown to cause an increase in phagocytosis. It has also been demonstrated that only the densest or oldest RBCs have significantly increased phosphatidylserine in the outer leaflet. This population also showed the greatest phagocytosis in vitro [Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D supra]. Again, the reason for the translocation of phosphatidylserine is unknown. There is not one clear mechanism responsible for senescent cell removal. However, the abundance of literature on the subject demonstrates the sensitivity of erythrophagocytotis to the chemical and physical structure of the RBC. [0022] Erythrocyte Modification: There are several instances in which the RBC membrane is altered without resulting in cell lysis, phagocytosis, or filtering. This includes both pathogens and intentional modifications [Duarte, M.O., MS; Shikanai- Yasuda, MA; Mariano, ON; Takakura, CFH; Pagliari, C; Corbett, CEP, Haemobartonella-like microorganism infection in AIDS patients: ultrastructural pathology. Journal of Infectious Diseases, 1992. 165: p. 976-977.]. [Kallick, C.L., S; Reddi KT; Landau, WL, Systemic Lupus Erythematosus associated with Haemobartonella-like organisms. Nature New Biology, 1972. 236: p. 145-146.]. [Archer, G.L.C., Philip H; Cole, Roger M; Duma, Richard J; Johnston, Charles L, Human infection from an unidentified erythocyte-associated bacterium.

[0023] Intentional modification of the RBC membrane has also been carried out with no apparent effect on circulation. Biotin has been covalently attached to the RBC to monitor RBC circulation half-lives [Suzuki, T.D., George L., Biotinylated erythrocytes: In vivo survival and in vitro recovery. Blood, 1987. 70(3): p. 791-795.]. It has been proposed that the bound biotin can be used to attach drug molecules to the RBC membrane and improve their circulation half-life and bioavailability [JJKrantz, A., Red cell-mediated therapy: Opportunities and Challenges. Blood Cells, Molecules and Diseases, 1997. 23(3): p. 58-68.]. However, cell lysis was observed when the bound biotin groups were subsequently attached to avidin or streptavidin. Streptavidin cross-linked complement inhibitors, present on the surface of the RBC, and thus induced cell lysis. The amount of cell lysis can be reduced by decreasing the number of biotin sites on the surface of the RBC or-by reducing the number of . binding sites, on streptavidin. [Muzykantov, N.R.M., Jaun C; Taylor, Ronald P; Atochina, Elena Ν; Herraez, Angel, Regulation of the complement-mediated elimination of red blood cells modified with biotin and streptavidin. Analystical Biochemistry, 1996. 241: p. 109-119.Aoki, H.F., Kaoru; Miyajima, Koichiro, Effects of blood on the uptake of charges liposomes by perfused rat liver.' cationic glucosamine-modified lipososmes interact with erythrocyte and escape phagocytosis by macrophages. International Journal of Pharmaceutics, 1997. 149: p. 15-23.].

SUMMARY OF THE INVENTION

[0024] The present invention comprises a method by which therapeutic carriers remain in vascular circulation for extended periods of time by binding to the outer membrane of blood cells, specifically red blood cells (RBCs). Freely circulating polymeric particles are cleared from circulation within a few minutes to hours depending on their size and surface characteristics. The adhesion between the RBC and particle protects the particle from vascular clearance by phagocytic cells. Sustained circulation of particles is used for drug delivery and prolonged circulation of contrast agents. Specifically, drugs that are rapidly eliminated from the blood can be delivered effectively in a sustained manner using the method of the present invention. At the same time, contract agents that are rapidly eliminated from the blood can be forced to circulate for a long time and used for diagnostics.

[0025] These and other features, aspects, and advantages of the present invention will become better understood with regard to the following detailed description and .accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

FIGURES 1 is an environmental scanning electron micrograph (ESEM) images of the particle-bound to a red blood cell.

FIGURE 2: The figure shows circulation time of 220 nm polystyrene nanoparticles bound to RBCs (closed circles) and not bound to RBC (open circles). RBC-bound nanoparticles exhibit >100 fold increase in circulation time.

FIGURE 3: Dependence of circulation time constant of RBC-bound nanoparticles on particle diameter. The time constant was determined by fitting exponential curves to plasma clearance curves similar to those in Figure 2.

FIGURE 4: Comparison of circulation times of RBC-bound nanoparticles (closed triangles) with poloxamine-908 coated nanoparticles (closed squares). Circulation of uncoated nanoparticles (open circles) is also shown. All nanoparticles are 220 nm in diameter. RBC-bound nanoparticles offer superior circulation compared to poloxamine- coated nanoparticles.

FIGURE 5: Binding of PLGA nanoparticles to RBCs. Bright spots correspond to PLGA nanoparticles.

FIGURE 6: A bifunctional particle (40) attached to a red blood cell (10). The particle has two function groups,_..an erythrocyte-adhesive.grqμp. (20) and ste.alth group (30). DETAILED DESCRIPTION OF THE INVENTION

[0026] Therapeutic carriers can be comprised of polymeric particles, lipid vesicles, micelles, or condensed DNA. These carriers can then be attached to the surface of cell, specifically RBC. Cell attachment can be carried out in a variety of manners primarily separated into passive and active binding. Binding allows the particle to stay attached to the cell in its in vivo environment without affecting the behavior and functionality of the cell.

[0027] Passive binding: Particles can be attached to RBC membranes through a passive, non-covalent electrostatic and hydrophobic interaction. Passive binding (that is binding based on physical adsorption), which has been demonstrated in EXAJMPJLE 1, provides an easy and efficient system to attach particles.

[0028] Active Bonding through modification of the particle surface: There are several routes to achieve such active binding. It has been shown that the N-hydroxy-succinimide ester of biotin binds to the surface proteins of RBCs [Krantz, A., supra]. It may also be possible to directly bind a particle to the surface of amine groups of an RBC using an N- hydroxy-succinimide ester linker. Alternatively, it is possible to utilize antibodies specific to RBC cell surfaces. RBCs are covered with a variety of blood group antigens. Many of RBC-antibodies are readily available, including anti-glycophorin A, anti- glycophorin C, anti-blood group antigen A and O. Peptides or proteins specific to RBC membranes may also be used for particle attachment. There are several proteins that have been identified as specific, for various surface groups of RJJBC membranes.-. This includes wheat germ agglutinin, WGA, which exhibits specificity for sialic acid [Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D; Montreuil, J, supra].

[0029] Peptide sequences from the malaria parasite with high binding affinity to the RBC membrane have also been identified [Puentes, A.G., Javier; Nera, Ricardo; Lopes, Q Ramses; Urquiza, Mauricio; Vanegas, Magnolia; Salazar, Luz Mary; Patarroyo, Manuel ElJkin, Serine repeat antigen peptides which bind specifically to red blood cells. Parasitology International, 2000. 49: p. 105-117.]. Any of these groups can be covalently or non-covalently attached to the particle surface to achieve specific binding.

[0030] Attaching particles to the membrane of RBCs using biotin offers a simple method of attachment. Ν-hydroxysuccinimidobiotin (JΝHS-biotin) has been shown to have over 120,000 binding sites on each RBC. The immobilized biotin remains active and is able to bind to avidin. Biotin-labeled RBCs do not have altered circulation half-lives if the bound-biotin levels are less than 30,000 molecules per cell. The biotin marker is stable in vivo [Suzuki, T.D., George L., supra]. In this method, ΝHS-biotin is first added to blood in vivo or in vitro. And avidin-coated carriers are then attached to the biotin end of ΝHS- biotin.

[0031] Wheat Germ Agglutinin (WGA) has been shown to have 40,000 binding sites per RBC [Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D; Montreuil, J,]. WGA alters the RBC phospholipid metabolism and increases the membrane rigidity up to 30 times [Kelm, S.P., James C; Rose, Ursula; Brossmer, Rejnhard; Schmid, .Walther;, Brandgar, Babasaheb P; . .Schreiηer, Erwin Harfmann, _. Michael; Zbiral,. Erich, Use of sialic acid analouges to define functional groups involved in binding to influenze virus hamagglutinin. European Journal of Biochemisty, 1992. 205: p. 147-153., and Dale, G.L.S., Takashige, Erythrocyte attached to a wheat germ agglutinin coated surface display an altered phospholipid metabolism. Journal of Cellular Biochemistry, 1988. 38: p. 1-11.]. Increased membrane rigidity was observed for free WGA concentrations above O.Olμg/ml. At lower concentrations the elasticity was not altered but the surface viscosity was changed. WGA can be placed on the surface of polystyrene surfaces by physical adsorption or covalent linkage using the amine groups on the protein. Physical adsorption is an easy technique, but it may not result in permanent or dense coverage. Covalent linkage can simply be obtained using an amine reactive linker, such as an aldehyde.

[0032] Peptide sequences from malaria parasites could be used to attach particles to RBC surfaces. Peptide sequences with up to 120,000 binding cites per cell have been identified [AoJki, H.F., Kaoru; Miyajima, Koichiro, supra]. These peptide sequences can be placed on the surface of particles and used to bind to the RBC membrane.

EXAMPLE 1

[0033] The passive binding between erythrocytes and polystyrene particles was obtained in vitro in the absence of plasma. Fresh human and rat blood was used for the experiments. Human RBCs were obtained by finger puncture or venous draw into heparinized microtainers. Rat RBCs were obtained by cardiac puncture or jugular vein draw in anesthetized animals. The procedures performed were the same for human and rat.RBCs. [0034] The RBCs were diluted 20 to 100 times in isotonic PBS at a pH of interest in the range of 5 to 10. Because RBCs were used immediately, a balanced salt medium containing glucose was not necessary. The RBCs were washed four times by centrifugation (1000-1500rpm) for about 1 to 2 min. This was done to remove the serum, primarily the albumin, which coated the outer membrane of the RBC. The cells were counted under an inverted microscope with 400x magnification using a hemocytometer. The cell concentration was then adjusted to 2.5xl0⁷ cells/ml.

[0035] Polystyrene particles were obtained from Interfacial Dynamics Corperation, Bangs Labs, Spherotech, Polysciences, and/or Molecular Probes. All experiments were performed with polystyrene particles of various sizes and surface chemistries. The

particles ranged in size from 0.1 to 1.2 μm. The surface chemistries present on the

particles were amine, carboxyl, sulfate, streptavidin, or NuetrAvidin (Molecular Probes) groups. The particles arrived sterile and packed in distilled water, with or without preservative, at concentrations from 1-10% w/v. The particles were diluted into the medium of choice to a concentration of 0.1 % w/v. The particles were vortexed and sonicated to ensure dispersion. This was checked using an inverted microscope with 400x magnification. The particles were then added drop wise to the RBC suspension to obtain the desired particle to cell ratio, between about 1:1 and 10:1. The particles were allowed to mix for about 10 min. The sample was then washed by centrifugation at 1000-1500 rpm for about 1 to 2 minutes to remove unbound particles. The binding was then quantified by optical microscopy or flow cytometry. For optical quantification, a 10 μ.1 drop was spread on a glass cover slip pretreated- with PBS-.- The cells- were viewed at 400x-1000x magnification on an inverted microscope. Four pictures were taken- of the sample. These pictures were analyzed to determine binding. Flow cytometry quantification was also used for fluorescently labeled particles. The particles were dyed with nile red fluorescent dye, which excites at 520nm and emits at 580nm, or yellow- green dye, which excites at 505nm and emits at 515nm. Using a green laser or argon laser, the forward and side scatter was used to view cells, cell/particle groups, and

particles, if the particles were over 0.8 μm. The signal in the particle fluorescence

emission range allowed the number of RBCs, with particles bound, to be determined. If the particles were greater than 0.45 μm, the number of particle bound to each RBC was also determined.

[0036] The effect of particle surface charge, size, and medium on binding efficiency was investigated. The surface chemistry of the particle and the ionic strength of the medium were found to have a dramatic effect on binding. One micron diameter polystyrene particles with different surface chemistries were tested for binding affinity to RBC membranes in phosphate buffered saline at a range of pH values from 5 to 10. The Particles had either amine, carboxyl, sulfate groups, and carboxyl modified groups (CJML), which have an excess of carboxyl groups, on their surface.

[0037] The pH has a significant effect on the binding of the amine, carboxyl, and plain particles but little effect on the binding of the CJML particles. This is due to the fact that CML particles have an abundance of carboxyl groups on their surface rendering the particle hydrophilic. The other particles have larger spacing between their groups, thus exposing the hydrophobic polystyrene below. These results show the importance of hydrophobic interactions. The sulfate and carboxyl functionalized particles are anionic in nature with pKa's around 2 and 5 respectively. They show similar binding trends. The amine particles are cationic in nature, pKa 13-15, and show an opposite trend compared to the sulfate and carboxyl particles. These results show the importance of surface charge and electrostatic interactions.

[0038] The RBC outer membrane is heavily glycosilated. The glycocalyx is hydrophilic with a net negative charge carried primarily by the sialic acid residues present on glycophorin A. There are positive charges associated with glycoprotein amino acid residues. Thus, the RBC goes through a charge reversal in the pH range of 3-5 [Donath, E., supra]. Electrostatic interactions are important for binding.

[0039] The effect of particle size on binding efficiency was also investigated. Carboxyl particles ranging from 0.1 to 1.2μm were bound to RBCs at a constant particle to cell ratio of about 3: 1. The binding decreased with decreasing particle diameter. This may be due to a decrease in surface area available for binding or a decrease in the frequency of collisions due to a decrease in the volume of particles in the system. The results show that binding increases with increased surface area of the particle.

[0040] Images of passive binding were obtain using an a Philips JXL-30 ESEM-EEG microscope. The cells were fixed in a 2% glutaraldehyde solution for 12 Jhrs and then washed with deionized water. The salt must be removed from solution because it will crystallize and occlude the image. The cells were imaged in a hydrated state to avoid artifacts due to volume loss, which occurs when particles are dehydrated for standard

SEM imaging. An ESEM image of a lμm particle attached to a human RBC is shown in

Figure L.

EJXAMPLE 2 [0041] The ability of RBC-bound carriers to remain in circulation was measured in vivo using rats as an experimental model. All experiments were performed in non-anesthetized animals. In these experiments, 500 μl of rat blood was withdrawn from jugular vein into a heparinized syringe and washed twice by centrifugation to remove plasma. The supernatant was removed and RBCs were reconstituted in saline containing therapeutic carriers. Polystyrene nanoparticles with diameters in the range of 100 nm-1100 nm were added to RBCs at a ratios ranging from 3:1 to 50:1 (particles:cells). Particles were allowed to remain in contact with RBCs for 5 minutes to allow physical adsorption. Unbound particles were removed by centrifugation. The suspension of RBCs (with particles attached on them) was re-mjected in the tail vein. Blood samples were collected from the tail vein over a period of 12 hours and particles remaining in circulation were counted using a flow cytometer.

[0042] Figure 2 shows the percentage of the 220 nm particles dose remaining in circulation at various times. Closed circles correspond to nanoparticles bound to RBCs. Open circles show circulation of 220 nm nanoparticles not bound to RBCs. More than 99% of unbound particles were removed in less than 1 minute. However, nanoaprticles bound to RBCs remained in circulation for prolonged periods. The circulation time of nanoaprticles (time required to remove 99% particles) was increased by >500-fold by allowing the particles to adhere to RBCs.

[0043] Figure 3 shows a plot of circulation half-life as a function of particle diameter when the particles are physically adsorbed on RBCs. The half-life, initially increases with particle diameter, after which it decreases with a further increase in particle diameter. The short half-life of 100 nm particles is due to weak binding of these particles on RBC. As the particle diameter increases, adsorption becomes stronger due to decreased particle curvature and results in increased half-life. With a further increase in particle diameter, the decrease in half-life may originate from recognition by the RES system.

[0044] Figure 4 compares the efficacy of RBC-adhesion in prolonging circulation compared to that achieved by the use of stealth polymeric coating. Closed triangles show circulation of RBC-bound nanoparticles (220 nm diameter). Open circles show circulation of unbound particles (220 nm diameter). Closed squares show circulation of 220 nm diameter particles coated with poloxamine 908. Poloxamine was adsorbed on nanoaprticles by overnight incubation as described in the literature [23]. Poloxamine coating had some effect on particle circulation, however, the effect of RBC-binding on circulation times was far superior compared to that of poloxamine.

EXAMPLE 3

[0045] Attachment of Wlieat Germ Agglutinin (WGA) on Particles.: Attachment of WGA on particle surface was achieved using 200 nm polystyrene particles (NuetrAvidin fluorescent yellow green (excitation-505 nm/emission-515 nm) from Molecular Probes) and Streptavidin fluorescent dragon green (480 nm/520 nm) polystyrene particles (530nm diameter) from Bangs Laboratories. Particles were diluted to a 0.05 wt% (0.5mg/ml) concentration in 0.1M PBS pH 7.4 and were dialyzed against PBS using a fast spin ^• dialyzer ' with 30Θ JkDa membranes from Harvard apparatus with 3 changes of medium, ^• This procedure removed the surfactant- and anti-microbial agent in particle suspension. Biotinylated WGA (Nector Laboratories) was then added to the particle suspension for a final concentration above the binding concentration of the particles. Reaction was allowed to occur for 2 hours. Excess biotin-WGA was removed by dialyzing against PBS using the fast spin dialyliser with 300 kDA membranes. Dialysis medium was changed at least 4 times and dialysis was allowed to proceed for 24 hours. Binding of WGA was confirmed using the Bio-Rad Protein Assay JMicroassay. The concentration of WGA in the particle solution is known at the beginning. A small amount of the particles suspension was removed at the end of binding. It was diluted in a known amount such that the starting WGA concentration would be in the linear range of the protein assay and

filtered through a O.lμm low binding millipore syringe filter to remove particles. 800μl

of this suspension was mixed with 200μl of Bio Rad Protein Assay and incubated for 10

min. The concentration was then read at 595 nm using a spectrophotometer. This difference in the concentrations is the amount bound to the particles. These particles exhibited strong adhesion to RBCs. Binding of WGA coated particles was significantly greater, up to 11 -fold, than binding of polystyrene or streptavidin coated particles.

EXAMPLE 4

[^"00461 Binding of PLGA nanoparticles to RBCS PLGA nanoaprticles were synthesized and adsrobed on rat RBCs through physical adsorption. PLGA was dissolved in acetone at a concentration of 0.25-0.5 mg/ml. Cumarin-6 was added to acetone to label the nanoparticles. 5 ml of PLGA solution was added to 60 ml of water and stirred. Acetone was evaporated to form solid PLGA nanoaprticles. This procedure yielded PLGA nanoaprticles with a mean diameter of 150 nm (as determined by dynamic light scattering). PLGA nanoparticles were added to a suspension of RBCs (serum free) at a concentration 10:1 (particles: cells). Unbound particles were removed by centrifugation. Attachment of nanoaprticles to RBCs was confirmed by visual observations and flow cytometry (Figure 5).

[0047] Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity and understanding, it will be obvious that various modifications and changes which are within the knowledge of those skilled in the art are considered to fall within the scope of the invention.

[0047] The following references are incorporated herein by reference:

Moghimi, S.M.H., A. Christy; Murray, J. Clifford, Long-Circulating and Target-Specific

Nanoparticles: Theory to Practice. Pharmacological Reviews, 2001. 53(2): p. 283-318.

Storm, G.B., Sheila O.; Daemen, Toos; Danilo, D. Lasic, Surface modification of nanoparticels to oppose uptake by the mononuclear phagocyte system. Advances Drug

Delivery Reviews, 1995. 17: p. 31-48.

Roser, M.F., Dagmar; Kissel, Thomas, Surface-modified biodegradable albumin nano- and microspheres. II: effect of surface charges on in vitro phagocytosis and biodistribution in rats. European Journal of Pharmaceutics and Biopharmaceutics, 1998.

46: p. 255-263.

Ogawara, K.-i.F., Kentaro; Takakura, Yoshinobu; Hashido, Mitsuru; Higaki, Kazutaka;

Kimura, ToshiJkiro, Surface hydrophobicity of particles is not necessarily the most important determinant in their in vivo disposition after intravenous administration in rats. Journal of Controlled Release, 2001. 77: p. 191-198.

Ogawara, K.-i.Y., Minoru ^'; Kubo, Jun-ichi; Nishikawa, Makiya; and Y.H. Takakura,

Mitsuru; Higaki, Kazutaka ; JKimura, Toshikiro, Mechanisms of hepatic disposition of polystyrene microspheres in rats: Effects of serum depend on the sizes of microspheres.

Journal of Controlled Release, 1999. 61: p. 241-250.

Simon, B.H.A., Howard, Y.; Gupta, Parddp K., Circulation Time and Body Distribution of 14C-Labeled Amino-Modified Polystyrene Nanoparticles in Mice. Journal of

Pharmaceutical Sciences, 1995. 84(10): p. 1249-1253.

Brain, J.D.M., Ramon M.; Decamp, Malcom M.; Warner, Angeline E., Pulmonary intravascular macrophages: their contribution to the mononuclear phagocyte system in 13 species. American journal of physiology-lung cellular and molecular physiology,

1999. 276(1): p. 1,146-1,154.

Ogawara, K.-i.Y., Minoru; Takakura, Yoshinobu; Hashida, Mitsuru; JHigaki, Kazutaka;

Toshikiro, Kimura, Interaction of polysytrene microspheres with liver cells: roles of membrane receptors and serum proteins. Biochimica et Biophysica Acta, 1999. 1472: p.

165-172.

Goldsby, R.A.K., Thomas J.; Osborne, Barbara A., The Complement System, in Kuby

Immunology, N. Folchetti, Editor. 2000, W. H. Freeman and Company: New York. p.

329-350.

Fernandes, A.LG., Gregory, The effect of polysialylation on the immunogenicity and antigenicity of asparaginase: implication in its pharmacokinetics. International Journal of Pharmaceutics, 2001. 217: p. 215-224.

Passirani, C.B., Gillian; Devissaguet, Jean-Philipe; Labarre, Denis, Interactions of nanoparticles bearing heparin of dextran covalently bound to poly(methylmethacrylate) with the complement system. Life Sciences, 1998. 62(8): p. 775-785.

Moghimi, S.D., SS, Innovations in avoiding particle clearance from blood by kupffer cells: Cause for reflection. Critical Reviews in Therapeutic Drug Carrier Systems, 1994.

11(1): p. 31-59.

Agre, P.P., John C, ed. Red Blood Cell Membranes: Structure, Function, Clinical

Implications. Hematology, ed. K.M.S. Brinkhous, Sanford A. Vol. 11. 1989, Marcel

Dekker: New York.

Alberts, B.B., Dennis; JLewis, Julian; Raff, Martin; Roberts, Keith; Watson, James D,

Molecular Biology of the cell. Vol. 3. 1994, New York: Garland Publishing, Inc. Bratosin, D.M., J,; Tissier, JP; Estaquier, J; Huart, JJ; Ameisen, JC; Aminoff, D;

Montreuil, J, Cellular and molecular mechanisms of senescent erythrocyte phagocytosis by machrophages. Biochimie, 1998. 80: p. 173-195.

Niitala, J.J., J, The red cell surface revisited. Trends in Biochemical Science, 1985. 10: p. 392-395.

Donath, E., "Hairy surface layer" concept of electrophoresis combined with local fixed surface charge density isotherms: application to human erythrocyte electrophoretic fingerprinting. Langmuir, 1996. 12: p. 4832-4839.

Juliano, R., The proteins of the erythrocyte membrane. Biochimica et Biophysica Acta,

1973. 300: p. 341-378.

Duarte, M.O., MS; Shikanai-Yasuda, MA; Mariano, ON; Takakura, CFH; Pagliari, C;

Corbett, CEP, Haemobartonella-like microorganism infection in AIDS patients: ultrastructural pathology. Journal of Infectious Diseases, 1992. 165: p. 976-977.

Kallick, C.L., S; Reddi KT; Landau, WL, Systemic Lupus Erythematosus associated with

Haemobartonella-like organisms. Nature New Biology, 1972. 236: p. 145-146.

Archer, G.L.C., Philip H; Cole, Roger M; Duma, Richard J; Johnston, Charles L, Human infection from an unidentified erythocyte-associated bacterium. The New England

Journal of Medicine, 1979. 301(17): p. 897-900.

Suzuki, T.D., George L., Biotinylated erythrocytes: In vivo survival and in vitro recovery.

Blood, 1987. 70(3): p. 791-795.

Krantz, A., Red cell-mediated therapy: Opportunities and Challenges. Blood Cells,

Molecules and Diseases, 1997. 23(3): p. 58-68. Muzykantov, N.R.M., Jaun C; Taylor, Ronald P; Atochina, Elena Ν; Herraez, Angel,

Regulation of the complement-mediated elimination of red blood cells modified with biotin and streptavidin. Analystical Biochemistry, 1996. 241: p. 109-119.

Νowakowski, R.L., Paul; Winlove, Peter, Imaging erythrocytes under physiological conations by atomic force microscopy. Biochimica et Biophysica Acta, 2001. 1514: p.

170-176.

Puentes, A.G., Javier; Nera, Ricardo; Lopes, Q Ramses; Urquiza, Mauricio; Nanegas,

Magnolia; Salazar, Luz Mary; Patarroyo, Manuel Elkin, Serine repeat antigen peptides which bind specifically to red blood cells. Parasitology International, 2000. 49: p.

105-117.

Evans, E.L., A, Adhesivity and rigidity of erhtocyte membrane in relation to wheat germ agglutinin binding. The Journal of Cell Biology, 1984. 98: p. 1201 1208.

Dale, G.L.S., Takashige, Erythrocyte attached to a wheat germ agglutinin coated surface display an altered phospholipid metabolism. Journal of Cellular Biochemistry, 1988. 38: p. 1-11.

Fujimoto, K.Ν., Kazumichi; Shidara, Miwako; Haruma, Kawaguchi, Preparation of unsymmetrical microspheres at the interfaces. Langmuir, 1999. 15: p. 4630463 5.

Claims

ClaimsWhat is claimed is:

1. A method of administering carriers in the blood of a subject wherein the carriers are attached to at least one type of cells in the blood.

2. A method of claim 1, wherein the carriers are attached to red blood cells.

3. A method of claim 1, wherein the carriers are attached to platelets.

4. A method of claim 1, wherein the carriers are attached to leukocytes.

5. A method of claim 1, the carriers are attached to blood cells in vivo.

6. A method of claim 1 wherein only a fraction of injected carriers are attached to blood cells.

7. A method of claim 1 wherein the attached carriers are slowly released from the blood cells.

8. A method of claim 1 wherein the release of carriers occurs in a period ranging from one minute to one month, more preferably in a period ranging from 1 hour to 1 day.

9. A method of claim 1, wherein the attachment of carriers to cells is performed ex vivo via steps comprising:

(a) removal of a sample of blood from a subject.

(b) preparation of cells for the attachment of carriers.

(c) addition of carriers to prepared cells for the purpose of attachment.

(d) removal of earners that are not attached to cells

(e) injection of cells in the subject.

10. A method of claim 9, wherein step (b) may be omitted.

11. A methods of claim 9, wherein step (d) may be omitted.

12. A method of claim 9 wherein the preparation step includes washing.

13. A method of claim 1, wherein a carrier is selected from a group of polymeric nanoparticles, polymeric microspheres, lipid vesicles, micelles, and condensed DNA.

14. A method of claim 13, wherein the carrier diameter is between 5 nm and 2 micron and preferably between 100 nm and 500 nm.

15. A method of claim 13, wherein the carrier is prepared from a biocompatible polymer.

16. A method of claim 13, wherein the carrier is prepared from a biodegradable polymer.

17. A method of claim 1 wherein the carrier contains a therapeutic agent.

18. A method of claim 17, wherein the said therapeutic agent is a thrombolytic agent.

19. A method of claim 17, wherein the said therapeutic is an anti-coagulating agent.

20 A method of claim 17, wherein the said therapeutic agent is a vaccine.

21. A method of claim 1, wherein the carrier is attached to a cell using physical adsorption, chemical binding, electrostatic attraction, or a combination thereof.

22. A method of claim 21, wherein the carrier is attached to a cell using wheat germ agglutinin.

23. A method of claim 21, wherein the carrier is attached to a cell using a peptide.

24. A method of claim 1, wherein the carrier is made up of a semiconductor material.

25. A method of claim 1, wherein the carrier is a contrast agent for ultrasound, x-ray, or JMRI imaging.

26. A method of claim 1, wherein the carrier has more than one functional group.

27. A method of claim 26, wherein at least one functional group is selected from poloxamer, polyoxamine, and polyethylene glycol.

28. A method of administering a therapeutic agent into body by steps comprising:

(a) encapsulation of a therapeutic agent in a carrier.

(b) attaching the carrier to at least one type of cells in blood.

29. A method of claim 28, wherein the attachment of carriers to cells is performed ex vivo.

30. A method of claim 28, wherein the attachment of carriers to cells is performed in vivo.

31. A method claim 28, wherein the therapeutic agent is itself a carrier.

32. A method of administering a contrast agent into a subject by attaching the contrast agent to at least one type of cells in the blood.

33. A method of claim 32, wherein the contrast agent may be chosen from a list of stabilized air bubbles, hollow polymeric particles, magnetic nanoparticles, and quantum dots.

34. A method of claim 1 wherein the carrier displays a therapeutic agent on its surface.

35. A method of claim 34, wherein the therapeutic agent is an enzyme.

36. A carrier designed to attach to at least one type of cells in the blood.