MIMOTOPE RECEPTORS AND INHIBITORS FOR PLATELET- PLATELET AND PLATELET-ENDOTHELIUM INTERACTIONS

TECHNICAL FIELD

The present invention relates generally to mimotopes and, in particular, to mimotopes for mimicking the receptor and inhibitor functionality of platelets.

BACKGROUND OF THE INVENTION

Mimotopes [mimetics or mimics) are molecules that mimic the function of other, naturally-occurring molecules by virtue of having the same shape (topography) and size as the naturally-occurring molecules that they are mimicking. A method for determining mimotopes is described in U.S. Patent 4,833,092 (Geysen) .

As shown in Figure Ia, a natural ligand has a particular shape and size that enables it to bind to a natural receptor. A mimotope ligand is a molecule that mimics the shape of the natural ligand and thus mimics its functional ability to bind to a natural receptor, as shown in Figure Ib. In other words, a mimotope ligand is a molecule that is the topographical equivalent of a natural ligand (at least in terms of their binding surfaces) so as to be complementary to a particular receptor of interest.

A variety of ligand mimics are known in the art, which are used primarily as inhibitors or blockers, e.g. U.S. Patent 4,550,163 (Voss et al.) entitled "Ligand analog- irreversible enzyme inhibitor conjugates" and U.S. Patent 6,139,832 (Li et al.) entitled "Leukocyte adhesion inhibitor-1 (LAI-I) Polypeptides". Small peptides are also known as protein mimetics (see, e.g. Wrighton et al.,

"Small Peptides as Potent Mimetics of the Protein Hormone Erythropoietin" in Science (1996 JuI 26; 273 (5274 ): 458-64 ). Mimetics of polypeptides used to detect antibodies are described in U.S. Patent 6,858,210 (Marquis et al.). Peptide mimics for backbone-to-backbone or backbone-to- chain cyclizations are described in U.S. Patent 6,706,862 (Hornik) .

In the context of platelets, mimotopes are also known as inhibitors of platelet adhesion and aggregation, such as described in U.S. Patent 5,114,842 (Plow et al.) entitled "Peptides and Antibodies that Inhibit Platelet Adhesion". Specifically, Plow et al. teach a polypeptide analog capable of immunologically mimicking a linear hGPIIb antigenic determinant expressed when platelet-associated GPIIb-IIIa binds fibrinogen. Both U.S. Patent 5,817,748 (Miller et al.) and its Continuation-in-Part U.S. 5,877,155 describe mimotopes and anti-mimotopes of human platelet glycoprotein IbIX as well as a method for modulating platelet adhesion, aggregation or agglutination by exposing the platelets to an anti-mimotope in order to inhibit von Willebrand factor interaction with platelets through the glycoprotein IbIX complex receptor. However, these ligand mimics only perform an inhibitory (antithrombotic) function.

Although the foregoing represent useful advances in the art, further advances in platelet and antithrombotic technology remain highly desirable.

SUMMARY OF THE INVENTION

Accordingly, one object of the present invention is to provide more pharmacologically compatible mimotope inhibitors for a new class of antithrombotic drugs.

Another object of the present invention is to provide mimotope receptors, which would function either as inhibitors or which would be attached to a suitable carrier to constitute a synthetic or artificial platelet.

This invention relates to the creation of peptide mimics of platelet integrins and their ligands . Short peptides, usually between 10-20 mer, are designed to provide shapes complementary to either the receptor or the ligand. A shape that mimics an integrin receptor's binding surface can be used to mimic the integrin receptor's binding function. Attached to a supporting surface of a carrier, such a peptide can behave as a receptor. As a free molecule, such a peptide can attach to the ligand, preventing it from accessing the receptor, thus acting as an inhibitor of the receptor-ligand interaction. Similarly, a peptide that mimics the ligand' s binding surface for the receptor will compete with the ligand and reduce its access to the receptor, thus also acting an inhibitor of receptor-ligand interaction. Such peptides may have, but are not obligated to have, sequence similarities to their parent proteins: they just need to have a complementary shape with sufficient binding affinity to attach to their counterpart in the receptor-ligand pair. Consequently, such peptides may be composed of L or D amino acids, although the D amino acids are preferred as these resist proteolytic degradation.

Accordingly, one aspect of the present invention provides a mimotope receptor comprising a peptide that mimics the shape and function of a natural receptor, thus providing a synthetic binding site for ligands. As a free molecule, the mimotope receptor inhibits ligand-receptor interaction, e.g. acts as an antithrombotic in the context

of platelet-platelet or platelet-endothelium interactions. If attached to a carrier, the mimotope receptor acts as a synthetic binding site, e.g. the carrier and mimotope receptor together function as a synthetic platelet.

Another aspect of the present invention provides a mimotope ligand comprising a peptide that mimics a natural ligand capable of binding to a receptor to thus inhibit ligand-receptor interaction, wherein the peptide is a D- peptide. Since the peptide is dextrorotary, it resists proteolytic degradation and thus forms the basis for a new class of antithrombotic drugs.

Yet another aspect of the present invention provides a mimotope ligand comprising a peptide that mimics a natural ligand capable of binding to a receptor to thus inhibit ligand-receptor interaction, wherein the peptide is attached to a carrier. Since the peptide is attached to a carrier, it resists excretion, again forming the basis for a new class of antithrombotic drugs. In one embodiment, the peptide is also dextrorotary to resist proteolytic degradation.

Yet a further aspect of the present invention provides a synthetic platelet comprising a carrier and a receptor mimic attached to the carrier, the receptor mimic mimicking a shape and size of a binding site of a natural receptor on a natural platelet. A synthetic or artificial platelet (or "platelet substitute") would have virtually limitless shelf life and would not require disease screening prior to transfusion, thereby providing a solution to the perpetual platelet shortages, as well as the safety and storage issues associated with natural blood platelets.

BRIEF DESCRIPTION OF THE DRAWINGS

Further features and advantages of the present invention will become apparent from the following detailed description, taken in combination with the appended drawings, in which:

Figure Ia is a schematic illustration of a ligand- receptor interaction between a natural ligand and a natural receptor;

Figure Ib is a schematic illustration of a ligand mimic binding to a natural receptor, thus acting as an inhibitor of the ligand-receptor interaction, as is known in the art;

Figure Ic is a schematic illustration of a peptide- based material that mimics the function of a receptor such as, for example, an integrin receptor on the surface of a platelet and further showing a natural ligand binding to the receptor mimic-

Figure 2a is a schematic illustration of a peptide- based material that, by binding to the ligand like a receptor, can inhibit receptor-ligand interactions;

Figure 2b is a schematic illustration of a peptide- based material that, when attached to a large carrier at low coupling ratios, binds to the ligand to thus mimic a receptor, thereby providing a specific, quasi-monovalent inhibitory function such as, for example, functioning as an antithrombotic in the case of platelet-endothelium and platelet-platelet interactions;

Figure 2c is a schematic illustration of a peptide- based material that, when coupled to a large carrier at high coupling ratios, provides specific multivalent

attachment possibilities, thus mimicking a receptor that is capable of binding multiple ligands;

Figure 3a is a schematic illustration of a peptide- based material comprising D-amino acids that can bind into an integrin receptor to thereby inhibit its ligand-binding function;

Figure 3b is a schematic illustration of a peptide- based material that, when attached to a large carrier at a low coupling ratio, binds to the receptor, mimicking a ligand, and thus providing a specific, quasi-monovalent inhibitory function such as, for example, functioning as an antithrombotic in the case of platelet-endothelium or platelet-platelet interactions;

Figure 4 shows a 3D computer model of a parent protein used for finding positions of particular sequences to enable the position to be related to potential vWf-GPIb interaction sites;

Figure 5 shows four cellulose membranes to which peptides were attached and which were then probed with purified GPIb in order to identify sequences of D-amino acids which potentially inhibit the GPIb-vWf interaction;

Figure 6 shows the confirmatory structural results of 3D computer modeling of the interaction between a D-peptide and vWf;

Figure 7 shows schematically how surface plasmon resonance in a Biacore machine can be used to validate that the peptides can act as receptors/binding partners; and

Figure 8 shows a Langmuir binding analysis used to determine the KD of the binding interaction between the peptide and fibrinogen.

It will be noted that throughout the appended drawings, like features are identified by like reference numerals .

DESCRIPTION OF PREFERRED EMBODIMENTS

In general, and as will be elaborated below, embodiments of the present invention provide mimotope receptors and inhibitors that employ peptide mimics for mimicking the shape and function of natural receptors and ligands, thus providing synthetic binding sites for ligands and receptors. Receptor mimics can be attached to carriers, such as liposomes, to act as synthetic platelets, for example, by providing binding sites for binding to other (natural or synthetic) platelets or to the endothelium. Mimotope inhibitors (either free-molecule receptors or ligands) can act as antithrombotics by inhibiting platelet-platelet and/or platelet-endothelium interactions.

As shown in Figure Ic, a peptide-based material can be used as a mimotope to mimic the form/shape (and thus the function) of a receptor. In one embodiment, the mimotope receptor (receptor mimic) can bind to a ligand to inhibit binding of the ligand to a natural receptor. In another embodiment, the mimotope receptor can be a peptide-based material that mimics an adhesion receptor or integrin on the surface of a platelet-like carrier like a liposome, preferably a cross-linked liposome.

In the context of platelets, an integrin, integrin receptor or (simply) receptor shall be used synonymously in

the present specification to mean a molecule, such as a peptide or protein, on the surface of the platelet or carrier that selectively binds a specific molecule known as a ligand.

As illustrated in Figure 2a, a peptide-based material can be used as a receptor mimetic to bind to the ligand like a receptor, thus inhibiting receptor-ligand interactions. As shown in Figure 2a, the mimotope receptor can be a "free" (unattached) peptide that has a shape/topology like that of a natural receptor so that it binds "preemptively" to ligands, thus preventing the ligands from binding to their natural receptors. These unattached, "free" receptor mimics thus act as inhibitors or blockers of the natural receptor-ligand interactions. In one embodiment, these mimotope receptors can be made of peptides that mimic the adhesion receptors or integrins of platelets. In the context of platelets, therefore, these unattached, "free" peptides would have an antithrombotic effect by binding to ligands and/or other factors, thus inhibiting normal platelet-platelet or platelet-endothelium adhesion.

As noted above, the mimotope receptor shown in Figure 2a could be a peptide that mimics an integrin of a platelet. For example, the peptide mimic could be shaped to bind to a ligand such as one of the active sites of a von Willebrand factor (vWf) protein. In a vWF monomer (which is a -2050 amino acid protein) , a number of specific domains are known to have specific functions. The Al domain, for example, binds to the platelet GPIB receptor. The Cl domain binds to platelet integrin αn _b β ₃ when activated. Therefore, in this example, the mimotope receptor could be a peptide that mimics the shape and

structure of the binding site of platelet GPIb-receptor by binding preemptively to the Al domain of the vWf monomer. Similarly, and again by way of example only, the mimotope receptor could be a peptide that mimics the shape and structure of the binding site of platelet integrin αnbβ3-

The mimotope receptor shown in Figure 2a could also be used to inhibit platelet-endothelium interaction by binding to the corresponding natural ligand that normally promotes adhesion of platelets to the vascular endothelial cells such as, for example, von Willebrand factor. As is known in the art, circulating platelets do not adhere to normal endothelium because platelet adhesion requires endothelial cell secretion of von Willebrand factor, which is found in the vessel wall and in plasma. The vWf protein binds during platelet adhesion to a glycoprotein receptor of the platelet surface membrane (glycoprotein Ib) . Thus, in this example, platelet-endothelium interaction can be inhibited by a mimotope receptor (peptide mimic) that binds preemptively to one of the active sites of the vWf protein to thus obstruct subsequent binding to that particular site on the vWf protein.

As illustrated in Figure 2b, a peptide-based material can also be attached to a large carrier at low coupling ratios for providing monovalent or quasi-monovalent inhibitory functions. This mimotope is thus a monovalent receptor mimic which, whether attached to a carrier or not, can bind to a corresponding ligand, thus inhibiting receptor-ligand interactions. By mimicking a receptor, this mimotope provides a specific, quasi-monovalent inhibitory function that can be used, for example, as an inhibitor of platelet-platelet and platelet-endothelium

interactions . This mimotope could thus be used as an antithrombotic .

As illustrated in Figure 2c, a peptide-based material can be coupled to a large carrier at high coupling ratios to provide specific, multivalent attachment possibilities, i.e. the synthetic receptor can simultaneously bind a plurality of ligands. In this case, the mimotope mimics a multivalent receptor and thus can form the basis of a synthetic platelet substitute.

As is known in the art, platelets (or "thrombocytes") are anuclear and discoid spherules ("flattened ellipsoids") that measure approximately 1.3-3.0 microns in diameter. Platelets adhere to each other via adhesion receptors or integrins that bind their specific ligands, which in turn facilitate adhesion to the endothelial cells of blood vessel walls. Platelets form haemostatic plugs with fibrin, a clotting protein derived from fibrinogen.

A synthetic platelet thus includes a carrier, such as a cross-linked liposome, that is manufactured to emulate some of the key physical characteristics of platelets (approximate size and shape, and resistance to liposome- cell fusion) . The synthetic platelet also includes at least one receptor mimic attached to the carrier (i.e. the outer surface of the liposome) . The receptor mimic includes a peptide that mimics a shape and size of a binding site of a natural receptor on a natural platelet. Preferably, the cross-linked liposome (or other equivalent carrier) includes a plurality of peptides attached to its outer surface, each one functioning as a receptor mimic to thus provide a "multivalent" synthetic platelet with multiple binding sites. In other words, each of the peptides is a

mimotope that mimics a natural adhesion receptor or integrin found on a natural platelet.

As shown in Figure 3a, a peptide-based material comprising D-amino acids can be used to bind into an integrin receptor to thus inhibit its ligand-binding function. Although some L-peptides (levorotatory peptides) are known in the art, D-peptides (dextrorotary peptides) are preferred because they resist proteolytic degradation.

As shown in Figure 3b, a peptide-based material can be attached to a large carrier (e.g. a liposome, vesicle or other body) at a low coupling ratio for binding to the receptor, thus mimicking a ligand and thus providing a specific, quasi-monovalent inhibition function. For example, the monovalent ligand mimic interferes with ligand-receptor interaction and thus can serve as an antithrombotic in the case of platelet-platelet interactions or platelet-endothelium interactions. The peptide attached to the carrier can be levorotary (L) or dextrorotary (D) . Attachment to the large carrier would resist excretion through the kidneys. In other words, the carrier (preferably a PEG, polyglycidol, or cross-linked liposome) provides circulatory resistance and physical blocking or obstruction of the binding site(s).

A peptide-based material in accordance with one of the foregoing embodiments would have great utility in the context of an artificial platelet substitute or as an antithrombotic drug.

A peptide-based antithrombotic drug would resist proteolytic degradation (proteolysis) because it is made of D-amino acids which form peptide bonds that natural enzymes cannot break down. Furthermore, a peptide drug where the

peptide is attached to a large carrier structure would resist excretion through the kidneys.

As platelets are routinely in short supply, it would be highly desirable to develop artificial platelets (also known as platelet substitutes) . The advantages of artificial platelets are numerous, namely virtually indefinite shelf-life and easy storage. Moreover, artificial platelets would not require infectious disease testing or assessment to determine whether the platelets are still viable for transfusion. The technology described in the foregoing paragraphs would thus provide the "specificity" component for artificial platelets. In other words, the peptide mimotopes could be attached to a liposome or other (synthetic) platelet-like structure to form an artificial platelet capable of binding to other platelets, either real (natural) platelets or other artificial (synthetic platelets). Furthermore, the peptide mimotopes could be coupled to a carrier at low density

(e.g. a quasi-monovalent interaction) to enable these peptides to function as platelet-inhibitors, thus giving rise to a new class of antithrombotic drugs.

Validation and Proof of Concept

The von Willebrand factor (vWf) amino acid sequence and available literature were used to select the potential vWf binding site for the integrin, glycoprotein Ib (GPIb) . As is known in the art, von Willebrand factor (vWf) is a large multimeric blood glycoprotein present in blood plasma that plays a significant role in platelet thrombus formation. The vWf is produced in the Weibel-Palade bodies of the endothelium, in megakaryocytes (stored in α-granules of platelets), and in subendothethial connective tissue. The primary function of von Willebrand factor is binding to

other proteins, such as Factor VIII, binding to collagen, binding to platelet GPIb, and binding to other platelet receptors when activated, e.g. by thrombin.

The vWf amino acid sequence was used to generate 10- mer L-amino acid overlapping peptides, shifted by two (2), according to the following pattern:

ACDFGHIKWER

DFGHIKWERAL

GHIKWERALND etc.

These peptides were synthesized and remained attached on the cellulose membrane. The membranes were probed by purified GPIb which was detected by anti-GPIb coupled to horseradish peroxidase (HRP) . A number of positive spots were found whose sequences were derived from their positions on the membrane.

The sequences were analyzed in silico by (a) finding their positions in a 3D model of the parent protein (see Figure 4) and then (b) relating that position to the potential vWf-GPIb interactive site. This suggested that the peptides colored black and brown (identified in Figure 4 as "+ve peptides") were in the interactive region and thus, as free peptides, could serve as competitive inhibitors of the interaction.

A similar study was conducted using overlapping peptides of the GPIb molecule, but the positive peptides identified by colours (in Figure 4) contributed relatively little to the interactive site.

This series of experiments identified a number of native sequences of L-amino acids with potential inhibitory activity for the GPIb-vWf interaction.

Random D-amino acid peptides (15 mer) were synthesized and probed with vWf to detect random sequences capable of binding vWf . Figure 5 shows the membranes from which four positive sequences were derived.

To determine whether these peptides were complementary to the binding surface defined by the GPIb molecule, they were analyzed in silico by (a) comparing them to known sequences in PDB. A. Fasta search provided homologues/decoys of known structure, (b) then the structures were docked onto the vWf molecule to check for 3D fit. Figure 6 shows the confirmatory structural results of this analysis for one of the three functional peptides identified.

Thus, the structural analysis by computer confirms the physical findings that random D-amino acid peptides that are structurally complementary (in this case to vWf) are also those that can be demonstrated experimentally to bind in vitro.

To confirm that peptides can act as receptors/binding partners, not just as inhibitors, real-time binding was demonstrated by surface plasmon resonance in a Biacore machine. In this case, peptides known to interfere with fibrinogen-GPIIbllla interaction were synthesized, and coupled to the end of a long (3400MW) PEG molecule whose other end was attached -to biotin, as illustrated schematically in Figure 7. (As is known in the art, fibrinogen is a soluble protein in the blood plasma essential for clotting of blood which the enzyme thrombin converts into the insoluble protein fibrin.) As shown

schematically in Figure 7, the biotin molecule was used to tether down the peptide-PEG onto a streptavidin-modified Biacore chip. This allowed the GPIIbIIa mimicking peptide to be hanging off the free end of the PEG.

By allowing free fibrinogen to flow past the peptide, the binding kinetics (i.e., the "on/off rate") between fibrinogen and the peptides were measured. Then, the fibrinogen was released from the peptide. Using several fibrinogen concentrations, it was possible to measure the KD of the binding interaction between the peptide and the fibrinogen. The Langmuir binding analysis is shown in Figure 8.

This showed that a peptide can generate binding kinetics/affinities similar to that of the parent protein and thus confirms the concept that peptides can act as synthetic receptor molecules.

The novel concept of using a peptide as a receptor mimic rather than only as an inhibitor opens a whole new potential field in the realm of peptide array and drug delivery.

A synthetic receptor bestows a number of significant advantages. First, since the receptor is synthetic, it does not have to be extracted, or made out of living material, purified, cleaned, etc. Second, it can be made (designed) to carry out any receptor function as long as the three dimensional shape of the receptor is mimicked. Third, the future production of synthetic cells (or cell- replacing materials) would require synthetic receptor functionality and thus a synthetic receptor would be a very significant first step in creating synthetic cells or synthetic platelets.

Potential uses of a synthetic receptor are numerous. As mentioned above, a synthetic receptor can be used on a platelet substitute (i.e. a synthetic or artificial platelet) . Furthermore, the synthetic receptor can be used to offer a specific binding capacity for isolating and analyzing ligand molecules without the need for monoclonal antibodies. These synthetic receptors could thus replace monoclonal antibodies in assay systems currently relying on monoclonal antibody technology. This would thus potentially eliminate the need for culturing and maintaining specific antibody-producing clones.

Moreover, the synthetic receptors can be tailored to obtain defined kinetics and binding affinities. The synthetic receptors could also be made from D- amino acids, thereby preventing proteolysis.

It is obvious for those skilled in the art that as the technology develops the basic idea of the invention can be implemented in various ways . The invention and the embodiments thereof are thus not restricted to the examples described above, but they may vary within the scope of the claims .