FIELD OF THE INVENTION

This invention relates to aluminum oxide surfaces for attaching biomolecules such as proteins for use in biomedical or industrial applications. The aluminum oxide surfaces are useful for applications such as diagnostic testing, affinity assays, and isolating biomolecules of interest from complex mixtures of biomolecules.

BACKGROUND

Over the last 50 years the need for immobilized biomolecules has grown significantly. Early efforts to immobilize biomolecules were for diagnostics such as ELISA assays. In these assays antibodies were added to polymer surfaces such as microtiter dishes. These surfaces were often polystyrene. The immobilized antibodies were only 5 to 7% active as the antibodies were randomly adsorbed to the surface. It was recognized that sensitivity of ELISA assays could be increased by an order magnitude if the antibodies were oriented on the surface such that the Fab fragments were directed away from the surface and the actual binding to the surface occurred through the Fc fragment. The need for oriented immobilized biomolecules has increased with time. Microfluidic devices, gene and protein microarrays and immunodiagnostics in general all benefit from oriented binding. Further analytical techniques such as quartz microbalances, surface plasmon resonance and capillary electrophoresis all benefit from improved oriented binding. Industrial processes such as affinity chromatography and affinity separations have improved efficiency if all the proteins were ordered on the surface properly. Equally important in all of these fields is the need to prevent nonspecific protein binding.

Protein adsorption on oxides is very important in the area of bioceramics. Titania, alumina, and zirconia are some of the more common oxide ceramics used for medical applications. Due to their excellent material properties, such as corrosion resistance, wear resistance, biocompatibility and mechanical strength, these materials are often used in medicine and dentistry as biomaterials. Protein adsorption to oxides must take into account many material and environmental aspects, such as pH, microstructure, zeta-potential, and surface reactivity.

Proteins and other biomolecules such as DNA and RNA readily adsorb onto surfaces, particularly if they are charged. Attempts to improve the adsorption process have included activating the surface chemically or physically to increase the number of charged sites. When materials such as alumina are considered, which are highly catalytic, the proteins adsorbed may be denatured. There is ample evidence in the literature to show that this is a common phenomena (Murray and Laband, 1979; Murray, 1980; Thurman and Gerba, 1988; Bowen and Gan, 1992). Since the adsorption process tends to be random it can also lead to inappropriate binding which may lead to steric hindrance. If molecules are oriented inappropriately or if they pack too tightly, steric hindrance occurs. If the desired proteins do not cover the surface fully nonspecific binding can become a significant problem.

Polymer networks, such as hydrogels, will hold water in which other molecules can be dissolved. These molecules are then available for reactions. The drawbacks to these techniques include the thickness of the thin-film and diffusion processes which may generate slow responses. Others have attempted to use thin polymer films to coat surfaces such as alumina. The polymers have exposed pendant groups which are then used to covalently bond to the biomolecules. This technique helps solve the problem of the overly thick thin-film but can lead to nonspecific binding and poor orientation.

On glass and silica surfaces the most common technique for covalently bonding biomolecules to the surface is through a silanation process. In these processes triethoxysilane derivatives are added to the surface where the ethoxy-groups react with the silicon hydroxide groups on the surface with the release of ethanol to form a silated surface. The silicon in the tri-ethoxysilane derivative is linked to a group that determines the surface properties. If this group is a long chain alkane the surface becomes hydrophobic, whereas if the group is a methyl group the surface becomes hydrophilic. These groups are then used to link covalently to biomolecules. In these cases the biomolecules may be an antibody, an antigen, DNA, RNA, or a molecule like avidin. Mao et al. (U.S. Patent No. 8,178,602) have shown such a system using R-PEG-silane as a preferred way to functionalize a surface. The R group in their example is either biotin or methoxy. When they use a blend of the two surface modifying molecules the methoxy component is a neutral spacer which reduces steric hindrance and allows the biotin to bind more avidin. Wagner et al. (U.S. Patent 6,596,545) show that a silane can be used to bind a linker molecule to glass or a thiol can be used to bind a sulphur- containing linker molecule to a gold surface. These linkers can then be attached to a biomolecule of interest. Coyne et al. (U.S. Patent No. 6,589,799) demonstrated activating surface hydroxyl groups on a support matrix material and reacting the activated hydroxyl groups with an aldehydic alkoxy silane. The derivatized aldehydic support matrix material was then useful for immobilizing biomolecules and biological applications.

Other covalent techniques include the use of calixarene. If the metal oxide surface is first reacted with SiCl ₄ to create a silicon chloride surface then that silicon chloride can be reacted with the calixarene to form a stable structure on the surface. This structure can have biomolecules covalently attached to it. (Katz et al. U.S. Patent No. 6,951,690)

SUMMARY

The inventor discovered that an aluminum oxide surface can be reliably re- functionalized for biomedical applications with an interface molecule having a carboxy rich domain. In particular, the interface molecule includes a polypeptide having at least one carboxy rich domain providing at least 5 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , the free carboxyl groups being provided by amino acids containing two or more carboxyl groups, through with the interface molecule is immobilized to the aluminum oxide surface. The interface molecule is immobilized on the aluminum oxide surface, and a biomolecule or a cross linking agent is attached to the interface molecule. This aluminum oxide/interface molecule structure or device provides a base, for example, for a diagnostic device or an affinity assay, and in a manner that the biomolecule retains its biological identity, and remains attached to the interface molecule in a stable manner.

In one broad aspect, the invention provides a device for use in binding to an analyte of interest. The device includes an aluminum oxide surface and an interface molecule immobilized on the aluminum oxide surface. The interface molecule includes a polypeptide having at least one carboxy rich domain providing at least 5 tree carboxyl groups within a molecular volume of 2.2-25 nm ³ , the tree carboxyl groups being provided by amino acids containing two or more carboxyl groups, through which the interface molecule is

immobilized to the aluminum oxide surface. One of the following is attached to the interface molecule:

i) a cross linking agent for binding to the analyte;

ii) a biomolecule attached to the interface molecule through one or more covalent bonds, the biomolecule being specific to the analyte;

iii) a biomolecule in the form of an engineered antibody attached to the interface molecule through a first antigenic determinant specific to the interface molecule, and having a second antigenic determinant specific to the analyte; and

iv) a biomolecule in the form of an engineered antibody attached to the interface molecule though an Fc fragment of the antibody and having Fab fragments specific to the analyte.

The invention also extends to an interface complex for use in binding to an analyte of interest and capable of being immobilized on an aluminum oxide surface. The interface complex includes an interface molecule comprising a polypeptide having at least one carboxy rich domain providing at least 5 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , the free carboxyl groups being provided by amino acids containing two or more carboxyl groups, through which the interface molecule is capable of being immobilized to the aluminum oxide surface. The interface molecule further includes one of:

i) a cross linking agent attached to the interface molecule for binding to an analyte of interest;

ii) a biomolecule attached to the interface molecule through one or more covalent bonds, the biomolecule being specific to an analyte of interest;

iii) a biomolecule in the form of an engineered antibody attached to the interface molecule through a first antigenic determinant specific to the interface molecule, and having a second antigenic determinant specific to an analyte of interest; and

iv) a biomolecule in the form of an engineered antibody attached to the interface molecule though an Fc fragment of the antibody and having Fab fragments specific to the analyte of interest.

ln another broad aspect, the invention provides a diagnostic system or a kit for testing whether binding has occurred to an analyte of interest. The diagnostic system or kit includes: 1) an aluminum oxide surface and an interface molecule including a polypeptide having at least one carboxy rich domain providing at least 5 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , the free carboxyl groups being provided by amino acids containing two or more carboxyl groups, through which the interface molecule is capable of being immobilized to the aluminum oxide surface, and 2) one of the following:

i) a cross linking agent attached to the interface molecule for binding to the analyte; ii) a biomolecule attached to the interface molecule through one or more covalent bonds, the biomolecule being specific to the analyte;

iii) a biomolecule in the form of an engineered antibody attached to the interface molecule through a first antigenic determinant specific to the interface molecule, and having a second antigenic determinant specific to the analyte; and iv) a biomolecule in the form of an engineered antibody attached to the interface molecule though an Fc fragment of the antibody and having Fab fragments specific to the analyte.

In some embodiments of the above, the interface molecule and one of (i), (ii), (iii) and (iv) are immobilized on the aluminum oxide surface. In other embodiments the interface molecule and one of (i), (ii), (iii) and (iv) are provided as an interface complex for contact with the analyte prior to immobilizing on the aluminum oxide surface.

Some embodiments of the diagnostic system or kit provide a visual diagnostic device. The aluminum oxide surface is provided on a reflective metal capable of generating a colour when covered by a porous layer of aluminum oxide, and the aluminum oxide surface is a porous anodized surface. In this manner, the visual diagnostic device, when contacted with a sample to test for the analyte, a colour change is detected denoting the presence of the analyte upon binding of the analyte either to the cross linking agent if (i) is present, or to the biomolecule if (ii), (iii) or (iv) is present.

The invention also broadly extends to a method of testing whether binding has occurred to an analyte of interest. The method includes:

a) providing an aluminum oxide surface having an interface molecule immobilized thereon, the interface molecule comprising a polypeptide having at least one carboxy rich domain providing at least 5 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , the free carboxyl groups being provided by amino acids containing two or more carboxyl groups, through which the interface molecule is immobilized to the aluminum oxide surface, the interface molecule being attached to one of:

i) a cross linking agent for binding to the analyte;

ii) a biomolecule specific to an analyte;

iii) a biomolecule in the form of an engineered antibody having a first antigenic determinant specific to the interface molecule, and having a second antigenic determinant specific to the analyte of interest;

iv) a biomolecule in the form of an engineered antibody attached to the interface molecule though an Fc fragment of the antibody and having Fab fragments specific to the analyte of interest;

b) contacting the surface of a) with a sample to test for the analyte; and

c) detecting the presence of the analyte upon binding of the analyte to the surface of a).

In some embodiments, the invention broadly extends to a method of testing whether binding has occurred to an analyte of interest. The method includes:

a) providing an aluminum oxide surface;

b) providing an interface complex capable of binding to the aluminum oxide surface, wherein the interface complex includes an interface molecule comprising a polypeptide having at least one carboxy rich domain providing at least 5 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , the free carboxyl groups being provided by amino acids containing two or more carboxyl groups, through which the interface molecule is capable of being immobilized on the aluminum oxide surface, and wherein the interface molecule is attached to one of:

i) a cross linking agent for binding to the analyte;

ii) a biomolecule specific to an analyte;

iii) a biomolecule in the form of an engineered antibody having a first antigenic determinant specific to the interface molecule, and having a second antigenic determinant specific to the analyte of interest; and

iv) a biomolecule in the form of an engineered antibody attached to the interface molecule though an Fc fragment of the antibody and having Fab fragments specific to the analyte of interest;

b) contacting the interface complex of a) with a sample to test for the analyte;

c) contacting the sample and the interface complex of b) with the aluminum oxide surface; and

d) detecting the presence of the analyte upon binding of the analyte and the interface complex to the aluminum oxide surface.

In some embodiments of the above methods, the aluminum oxide surface is provided on a reflective metal capable of generating a colour when covered by a porous layer of aluminum oxide, and the aluminum oxide surface is a porous anodized surface. In this manner, when contacted with a sample to test for the analyte, a colour change is detected denoting the presence of the analyte upon binding of the analyte either to the cross linking agent if (i) is present, or to the biomolecule if (ii), (iii) or (iv) is present.

In some embodiments of the above device, diagnostic system, kit or method the carboxy rich domain provides: at least 10 free carboxyl groups within a molecular volume of 2.2-25 nm ³ ; or at least 20 free carboxyl groups within a molecular volume of 2.2-25 nm ³ ; or at least 10 free carboxyl groups within a molecular volume of 2.2-17 nm ³ , or at least 20 free carboxyl groups within a molecular volume of 2.2-17 nm ³ ; or at least 10 free carboxyl groups within a molecular volume of 7.0-17 nm ³ , or at least 20 free carboxyl groups within a molecular volume of 7.0-17 nm ³ .

In some embodiments of the above device, diagnostic system, kit or method:

the cross linking agent is covalently bonded to the interface molecule; or

the biomolecule is covalently bonded to the interface molecule through a cross-linking agent; or

the interface molecule and the biomolecule are engineered as an amino acid sequence such that the interface molecule and biomolecule are attached through peptide bonds; or the interface molecule is an engineered or synthetic protein, polypeptide or antibody incorporating the carboxy rich domain; or

the biomolecule is an engineered antibody having a first antigenic determinant specific to the interface molecule, and having a second antigenic determinant specific to the analyte of interest; or

the biomolecule is an engineered antibody attached to the interface molecule though an Fc fragment of the antibody and having Fab fragments specific to the analyte of interest. ln some embodiments, the cross linking agent is covalently bonded to the interface molecule. In other embodiments, the biomolecule is covalently bonded to the interface molecule through a cross-linking agent. In other embodiments, the interface molecule and the biomolecule are engineered as an amino acid sequence such that the interface molecule and biomolecule are attached through peptide bonds. In still further embodiments, the interface molecule is an engineered protein, polypeptide or antibody incorporating the carboxy rich domain.

In some embodiments the interface molecule is provided on the aluminum oxide surface as a continuous or discontinuous coating, as delineated spots or lines, or as an array.

In some embodiments the interface molecule is one of different types of interface molecules spaced on the aluminum oxide surface, with biomolecules attached to one type of the interface molecules.

In some embodiments, the aluminum oxide surface is provided on a substrate in the form of a particle, powder, thin film, slide, strip, bead, magnetic bead, magnetic particle or coating.

In some embodiments, the aluminum oxide surface is provided by sputtering, evaporating, casting or extruding aluminum metal or an aluminum alloy, which is further anodized to provide a porous anodized aluminum oxide surface.

In some embodiments, the aluminum oxide surface is provided by RF sputtering, reactive sputtering or chemical vapour depositing aluminum oxide onto a substrate.

In some embodiments, the interface molecule includes one or more of the amino acids selected from the group consisting of aspartic acid (Asp), glutamic acid (Glu), and gamma- carboxyglutamic acid (Gla).

In some embodiments, the interface molecule is a Vitamin K dependent protein, a fragment thereof containing a Gla domain, or a fragment thereof containing a modified Gla domain.

In some embodiments, the interface molecule and the biomolecule are formed as an engineered molecule such that the carboxy rich domain is included in a protein, a polypeptide, an antigen, an antibody, a carbohydrate, an aptamer or a lipid.

In some embodiments, the carboxy rich domain includes a Gla domain of a Vitamin K depend protein, or a fragment or deriviative thereof, and the engineered molecule includes protein A, fragment B of protein A, or an IgG molecule.

In some embodiments, the interface molecule is a protein, a Gla domain of a protein, or a modified Gla domain of a protein in which one or more of the Gla residues are substituted with Glu, Asp, Glu-Glu, Glu-Asp, Asp-Glu or Asp-Asp, wherein the protein is selected from the group consisting of prothrombin, Fragment 1 of prothrombin, protein S, coagulation Factor IX, Factor X, Factor VII, protein C, matrix Gla protein, and bone Gla protein.

In some embodiments, the biomolecule is a member of a binding pair selected from the group consisting of antibody-antigen, antibody-hapten, enzyme-substrate, enzyme- receptor, toxin-receptor, protein-protein, avidin-biotin, aptamer-aptamer target, and drug receptor-drug.

The invention also extends to use of an interface molecule to refunctionalize an aluminum oxide surface for use in biomedical applications, wherein the interface molecule comprises a polypeptide having at least one carboxy rich domain providing at least 5 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , the free carboxyl groups being provided by amino acids containing two or more carboxyl groups, through which the interface molecule is capable of being immobilized to the aluminum oxide surface.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 depicts the structural domains of vitamin K-dependent proteins prothrombin (PT), factor IX (FIX), factor X (FX), factor VII (FVII), protein C (PC), protein S (PS), matrix Gla protein (MGP), and bone Gla protein (BGP). The legend identifies protein components. Proteolytic cleavage sites are shown with thin arrows when cleaving occurs to create the mature protein structure, and cleavage sites are indicated by thick arrows when linked to enzymatic activation. Image from Furie B and Furie BC (Furie and Furie, 1988).

Figure 2 shows the 3D structure of prothrombin. There are 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein (UnitProt Consortium, 20l7a). This forms an 8.3 nm ³ carboxyglutamic domain (carboxy rich Gla domain) at the base of the structure where binding to the alumina surface occurs.

Figure 3 shows the 3D structure of human coagulation factor IX. There are 12 Gla residues located at sites 7, 8, 15, 17, 20, 21, 26, 27, 30, 33, 36 and 40 on the protein.

Additional metal binding sites are 1, 2, 47, 48, 50, 64, 65, 235, 237, 240, 242, and 245 (UnitProt Consortium, 20l7b). This forms a 17.0 nm ³ carboxyglutamic domain at the base of the structure where binding to the alumina surface occurs.

Figure 4 shows the 3D structure of protein S. There are 11 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, 32 and 36 (UnitProt Consortium, 20l7c). This forms a 12.3 nm ³ carboxyglutamic domain at the base of the structure where binding to the alumina surface occurs.

Figure 5 shows the 3D structure of a modified Fragment 1 of Factor P. There are 10 Asp residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms an 8.7 nm ³ carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 6 shows the surface charge map of the 3D structure of a modified Fragment 1 of Factor P. There are 10 Asp residues in place of 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 7 shows the 3D structure of a modified Fragment 1 of Factor P. There are 10 (Asp-Asp) residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 7.7 nm ³ carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 8 shows the surface charge map of the 3D structure of a modified Fragment 1 of Factor P. There are 10 (Asp-Asp) residues in place of 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 9 shows the 3D structure of a modified Fragment 1 of Factor P. There are 10 (Asp-Glu) residues replacing Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 8.5 nm ³ carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 10 shows the surface charge map of the 3D structure of a modified Fragment 1 of Factor P. There are 10 (Asp-Glu) residues replacing 10 Gla residues located at sites 6, 7,

14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 11 shows the 3D structure of a modified Fragment 1 of Factor P. There are 10 (Glu-Asp) residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 12.5 nm ³ carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 12 shows the surface charge map of the 3D structure of a modified Fragment 1 of Factor P. There are 10 (Glu-Asp) residues replacing 10 Gla residues located at sites 6, 7,

14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 13 shows the 3D structure of a modified Fragment 1 of Factor P. There are 10 (Glu-Glu) residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 11.8 nm ³ carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 14 shows the surface charge map of the 3D structure of a modified Fragment 1 of Factor P. There are 10 (Glu-Glu) residues replacing 10 Gla residues located at sites 6, 7,

14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 15 shows the 3D structure of Fragment 1 of Bovine Factor P. There are 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 7.0 nm ³ carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 16 shows the surface charge map of the 3D structure of Fragment 1 of Bovine Factor P. There are 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a carboxy rich domain at the base of the structure where binding to the alumina surface occurs.

Figure 17 shows the 3D structure of a modified (Asp in place of Gla) Fragment 1 of human Factor P added to the carboxyl terminus of an IgG heavy chain. There are 10 Asp residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 3.6 nm ³ carboxy rich domain at the base of the IgG structure allowing binding to an alumina surface.

Figure 18 shows the surface charge map of the 3D structure of a modified (Asp in place of Gla) Fragment 1 of human Factor P added to the carboxyl terminus of an IgG heavy chain. There are 10 Asp residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a carboxy rich domain at the base of the IgG structure allowing binding to an alumina surface.

Figure 19 shows the 3D structure of a modified (Glu in place of Gla) Fragment 1 of human Factor P added to the carboxyl terminus of an IgG heavy chain. There are 10 GLU residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 3.7 nm ³ carboxy rich domain at the base of the IgG structure allowing binding to an alumina surface.

Figure 20 shows the surface charge map of the 3D structure of a modified (Glu in place of Gla) Fragment 1 of human Factor P added to the carboxyl terminus of an IgG heavy chain. There are 10 Glu residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a carboxy rich domain at the base of the IgG structure allowing binding to an alumina surface.

Figure 21 shows the 3D structure of Fragment B of Protein A linked to the first 35 amino acids of Fragment 1 of human Factor P. There are 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29 and 32 on the protein. This forms a 2.2 nm ³ carboxy rich domain at the base of the Fragment B structure allowing binding to an alumina surface.

Figure 22 shows the surface charge map of the 3D structure of Fragment B of Protein A linked to the first 35 amino acids of Fragment 1 of human Factor P. There are 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29 and 32 on the protein. This forms a carboxy rich domain at the base of the Fragment B structure allowing binding to an alumina surface.

Figure 23 is a schematic drawing showing an aluminum oxide surface with a prothrombin (FII) interface molecule immobilized to the surface through the Gla (carboxy rich) domain. The FIT is modified with a cross linker such as glutaraldehyde for bonding a biomolecule such as an antigen. The antigen is shown bound to its specific antibody. Example 1 is representative of this figure. In Ex. 1 , the linker is glutaraldehyde, the antigen is from Influenza B and the target antibody is anti-influenza B virus.

Figure 24 is a schematic drawing showing an aluminum oxide surface with a

Fragment 1 of prothrombin (Frag 1) interface molecule immobilized to the surface through the Gla domain. The Frag 1 is modified with a cross linker for bonding a biomolecule, shown as an antigen, which is bound to its specific antibody. Example 5 is representative of this figure, in which the linker is glutaraldehyde, the antigen is from Hepatitis B vims and the target antibody is anti-Hep B antibody.

Figure 25 is a schematic drawing showing an aluminum oxide surface with a Factor IX (FIX) interface molecule immobilized to the surface through the Gla domain. The FIX is modified with a cross linker for bonding a biomolecule, shown as an antigen, which is bound to its specific antibody. For example, the linker can be glutaraldehyde, the immobilized antigen can be influenza B, and the antibody can be anti-influenza B.

Figure 26 is a schematic drawing showing an aluminum oxide surface having immobilized thereon, a Fragment B of Protein A engineered to have a Frag 1 of FII incorporated in its structure, such that the Gla domain binds to the aluminum oxide surface. The engineered Fragment B of Protein A provides an interface molecule that is bound to a biomolecule, shown as an antibody, which is bound to its specific antigen. For example, the immobilized antibody can be from anti-influenza B and the antigen can be influenza B.

Figure 27 is a schematic drawing showing an aluminum oxide surface with an antibody (IgG) engineered to have the Gla domain of Fragment 1 of FII incorporated at the carboxy terminus of its Fc structure, bound to the surface through the Gla domain. The engineered antibody is shown bound to its specific antigen. For example, the Fab can be anti influenza B and the target antigen can be from influenza B. Figure 28 is a schematic drawing showing an aluminum oxide surface with an antibody (IgG) engineered to have the Gla domain of Fragment 1 of FIT incorporated at the carboxy terminus of its Fc structure, bound to the surface through the Gla domain. The engineered antibody is shown bound to its specific antigen. A second antibody is shown binding to the target antigen, which can be used to amplify the signal. The second antibody may be unmodified or it may be modified with a linked enzyme or radiolabel in order to amplify the signal from the overall device. For example, the Fab may be anti-testosterone, the target antigen can be testosterone, and the amplifying antibody can be anti-testosterone.

Figure 29 is a schematic drawing showing an aluminum oxide surface with both Fragment 1 of prothrombin and prothrombin (FIT) as two types of interface molecules bound to the surface through the Gla domain. A cross linking agent such as glutaraldehyde is added for binding a biomolecule, such as an antigen. This device uses one type of interface molecule (Frag 1) to act as a spacer between the other interface molecule (FIT) to reduce factors such as steric hindrance. The antigen, and the antibody specific to the antigen (not shown), have less steric hinderance for attaching to the interface molecule (FIT) most distant from the surface. For example, the immobilized antigen can be the Hep B vims antigen, and a linker capping agent (ex. lysine) can be added after binding of the antigen to prevent non-specific protein binding.

Figures 30A-30D are schematic drawings showing the use of an interface complex including an interface molecule and an antibody. The complex is an engineered antibody with a carboxy rich domain on the Fc terminus, or an antibody which is Fc conjugated, via the carboxy terminus, to a carboxy rich domain, used to capture antigens as analytes in a test sample. An example of an engineered antibody of this type is the Gla domain of a vitamin K dependent protein linked to the Fc terminus of an IgG antibody. The antigens are separated from the solution by contacting with an aluminum oxide surface, such that the carboxyl rich domain of the engineered antibody, bound to the antigens, binds to the alumina surface. Figure 30A shows the engineered antibody in a solution with a mixture of proteins; Figure 3 OB shows the binding of the engineered antibody to its specific antigen; Figure 30C shows the binding of the engineered antibody-antigen complex through the carboxyl rich domain to the aluminum oxide surface; and Figure 30D shows the separation of the protein of interest from the mixture by removing the aluminum oxide surface from the solution. If the aluminum oxide surface is formed as a visual diagnostic device, the binding of the antigen is detected by a colour change in Figure 30D. As an example, the Fab can be anti-albumin and the target protein can be albumin.

Figures 31A-31D are schematic drawings showing an aluminum oxide surface having immobilized thereon an interface molecule such as prothrombin (FII). The FII molecule is modified with a cross linking agent such as glutaraldehyde, as shown in Figure 31 A. The surface is immersed in an sample such urine in Figure 31B. If there is analyte protein in the sample, it binds to the glutaraldehyde on the interface molecule, as shown in Figure 31 C. Removing the surface from the sample removes protein bound to the interface molecule through the cross linking agent, as shown in Figure 31D. If the aluminum oxide surface is formed as a visual diagnostic device, the binding of the protein is detected by a colour change in Figure 31D.

Figures 32A-32C are schematic drawings showing an interface complex including an interface molecule and an antibody as an engineered antibody. The interface molecule, such as Fragment 1 of Factor P is immobilized on the aluminum oxide surface. An antibody is engineered with two distinct Fab antigenic determinants, Fab _j and Fab ₂ , as shown in Figure 32A. The engineered antibody can be made through genetic engineering or wet chemistry, by techniques known in the art. Fab is specific to the interface molecule in order to bind to the interface molecule (for example the antithrombin component of antiprothrombin), while Fab ₂ is specific to an analyte of interest such as an antigen. Figure 32B shows the interface molecule on the aluminum oxide surface and the engineered antibody attached to the interface molecule through Fab, . Figure 32C shows the engineered antibody binding to an analyte antigen through Fab ₂ . Alternatively, the Fab ₂ can be attached to a further biomolecule that functions as part of a binding pair such as an antibody or aptamer. For example Fab can be anti-thrombin and Fab ₂ can be anti-prostate specific antigen, while the analyte is prostate specific antigen.

Figures 33A and 33b are schematic drawings showing the simultaneous use of two different sized interface molecules (high and low) bound to an aluminum oxide surface through a carboxy rich domain. This configuration allows two different biomolecules to be linked to the interface molecule, for example through a cross linking agent such as an aptamer (Figure 33A). The linker can be added in bulk solution for one interface molecule, and in a separate solution of the other interface molecule. For example the high interface molecule can be prothrombin, and the low interface molecule can be Fragment 1 of Factor II. In Figure 33B, biomolecules are atached through the cross linking agent. The biomolecules are different antigens which are recognized by the different antibodies in a sample. For example, target molecules may be two different hormones, such as hCG and LH. If the underlying surface is aluminum oxide on a reflective layer to function as a visual assay, binding of the different antibodies generates multiple colour combinations depending on which antibodies are present in the sample. For example, 0 denotes no colour change for no antibody binding, 1 or 2 denoting colour change for binding of either antibody 1 or antibody 2, and 3 denoting colour change for binding of both antibody 1 and antibody 1. In other embodiments, the biomolecules may be antibodies, aptamers or a components of alternate binding pairs.

Figures 34A and 34B are schematic drawings showing binding of polyclonal engineered IgG to the aluminum oxide surface. The IgG molecules are engineered with a carboxy rich domain formed at the carboxyl terminus of the Fc fragment, for binding to the aluminum oxide surface. The Fab fragments are illustrated as Fab and Fab ₂ for binding to different epitopes of the antigen of interest (orientation 1 or orientation 2 being shown in Figure 34B). In this manner, the Fab fragments are able to bind specifically to the antigen they were developed for. By using a polyclonal antibody, the binding is faster as the orientation of the antibody is not a significant factor. For example, Fab can be monoclonal- 1 for prostate specific antigen (specific site/orientation) and Fab ₂ can be monoclonal-2 for prostate specific antigen (specific site/orientation), while the analyte is prostate specific antigen.

Figure 35A and 35B are schematic drawings showing two embodiments for paterning the interface molecules on an aluminum oxide surface. In Figure 35A, aluminum is sputered onto a surface in a patern of lines using masking techniques known in the art. The aluminum is sputered onto a thin film of tantalum on an underlying support, such as a slide. The aluminum can then be anodized to form a layer of aluminum oxide on a layer of tantalum oxide on a tantalum surface. Once anodizing is complete, the aluminum oxide strips are capable of producing interference colours, which change upon each addition of interface molecules, biomolecules and analytes binding to the biomolecules. Detection of the change in interface colours can be detected by the eye in the visible spectrum, and/or with suitable detectors, cameras etc. in the visible spectrum or outside the visible spectrum. In some embodiments the paterns such as these can be achieved by printing the interface molecules as lines or paterns on a continuous aluminum oxide film on an underlying reflective layer. Figures 36A and 36B show embodiments of diagnostic devices in accordance with the invention in which test sites are printed in lines or arrays, as set out for Figures 35 A, 35B, together with barcode or QR codes containing information relating to the array, the assay, the patient etc. for reading by a barcode reader or a smartphone.

DETAILED DESCRIPTION

Certain terms used herein and in the claims are defined and clarified hereinbelow.

The term“biomolecule” includes molecules that interact with a biological system. In general a biomolecule (or a biological molecule) is a term for molecules or ions that are present in organisms. Biomolecules include large macromolecules (or polyanions) such as proteins, carbohydrates, lipids and nucleic acids (aptamers), or derivatives or fragments thereof. Biomolecules also include small molecules such as primary metabolites, secondary metabolites, natural products and their derivatives. Biomolecules are usually endogenous, but may be exogenous. For example, pharmaceutical drugs may be natural products or semisynthetic (biopharmaceuticals) or they may be totally synthetic, and as such are included in the term biomolecules. Biomolecules also extends to synthetic proteins, polypeptides, and peptides, synthetic DNA or RNA, synthetic lipids and synthetic carbohydrates.

The terms“alumina” and“aluminum oxide” are used synonymously herein, and includes an oxide formed on the surface of aluminum or an aluminum metal alloy, whether the oxide is native or formed, for example by sputtering or by anodizing to provide an aluminum oxide. The aluminum oxide surface may be formed on a substrate or a support, such as on a particle, powder, bead, magnetic bead or particle, thin film, slide, strip, filter, or coating. The aluminum oxide surface may be formed by sputtering, RF sputtering, reactive sputtering, chemical vapour deposition, evaporating, casting, or extruding aluminum metal or an aluminum alloy, and may be further anodized to provide a porous anodized aluminum oxide surface.

The terms“bioidentity” and“biological identity” are used synonymously to refer to the structural and chemical property of a biomolecule that make it recognizable or specific to binding to other biomolecules, cells or tissues. For example, when an antigen is immobilized it is said to have retained its bioidentity if an antibody that was developed to it in an animal model, in vitro system, or computational model, binds to it with the same specificity and sensitivity as it would if it were free in solution. As used herein, bioidentity does not necessarily require retained biological activity in a biomolecule, for example in an enzyme, it is sufficient that the biomolecule is still recognized.

The term“domain” refers to a group of amino acids within a protein or polypeptide which is identifiable by function, properties or structure from other parts of the protein or polypeptide.

The terms“immobilized” or“immobilize” as used herein refers to the attachment or adherence of one or more interface molecules to the aluminum oxide surface, whether or not through chemical bonding.

The terms“covalent binding”,“covalent bonding” and“cross-linking” are used herein to refer to the formation of covalent bonds between the interface molecule and the biomolecule. Typically, when the biomolecule is a polypeptide, the interface molecule is covalently bonded through a cross-linking agent. Alternatively, when the interface molecule is engineered to include the biomolecule, the covalent binding is through peptide bonds.

The term“polypeptide” refers to chains of amino acids held together by peptide bonds, and as used herein includes proteins, whether natural or engineered, fragments of such proteins, and amino acid sequences derived from such proteins or fragments.

The term“sample” as used herein includes any biological or environmental material suspected of containing one or more analytes of interest, and includes a sample which lacks the analyte such that the test for the analyte is negative. Biological samples include, for example, bodily fluids and organic materials such as foodstuffs. A bodily fluid includes, for example, whole blood, plasma, serum, sputum, cerebrospinal fluid, pleural fluid, tissue, fecal material and the like. Environmental samples include, for example, soil, sludge, water and the like. The sample can be processed, for example centrifuged, extracted, and/or lysed if cells are present. Alternatively, the sample can be directly placed in contact with the diagnostic device.

ln accordance with the invention, an aluminum oxide surface is refunctionalized with an interface molecule capable of being immobilized on the surface, for use in biomedical applications ln some embodiments, the interface molecule can be activated by attaching to a cross linking agent, which can than be used to bind to analytes of interest in test sample ln other embodiments, the interface molecule can be attached to a biomolecule, which can then be used (directly or indirectly) to bind to analytes of interest in a test sample. In some embodiments, the biomolecule is covalently bonded to the interface molecule, either with cross linking agents, or by covalent bonds such as peptide bonds. The interface molecule and the biomolecule may be engineered together as an interface complex. In other embodiments, the biomolecule may be an engineered antibody, for example having one antigenic determinant specific to the interface molecule and a second antigenic determinant specific to an analyte of interest in a test sample, or to another biomolecule. In other embodiments, the biomolecule may be an engineered antibody attached to the interface molecule though an Fc fragment of the antibody and having Fab fragments specific to the analyte of interest.

The interface molecule includes a polypeptide having at least one carboxy rich domain providing at least 5 free carboxyl groups within a molecular volume of 2.2-25 nm ³ . The free carboxyl groups are provided by amino acids containing two or more carboxyl groups, through which the interface molecule is immobilized to the aluminum oxide surface. The attached biomolecule retains its biological identity when attached to the interface molecule.

For clarity, as defined herein and in the claims, the complex resulting for an interface molecule attached to a biomolecule excludes an end product which is a vitamin K dependent protein such as prothrombin.

In some embodiments, the carboxy rich domain provides at least 10 free carboxyl groups within a molecular volume of 2.2-25 nm ³ ; or at least 20 free carboxyl groups within a molecular volume of 2.2-25 nm ³ ; or at least 10 free carboxyl groups within a molecular volume of 7.0-20 nm ³ , or at least 20 free carboxyl groups within a molecular volume of 7.0- 17 nm ³ .

i) Interface molecules or Interface Complexes with Biomolecules

In accordance with the invention, interface molecules having a particularly high affinity to the aluminum oxide surface through one or more carboxy rich domains, are immobilized on an aluminum oxide surface, for instance as a continuous or discontinuous coating, as delineated spots or lines, or as an array. The interface molecule can be immobilized as a coating on the aluminum oxide surface from a solution of the interface molecule in a suitable solvent, followed by removing the solvent. In general, the interface molecule includes one or more of the amino acids aspartic acid (Asp), glutamic acid (Glu) and gamma-carboxyglutamic acid (Gla) to provide the carboxy rich domain(s). Examples of interface molecules are the Vitamin K dependent proteins, fragments thereof containing the carboxy rich Gla domain, fragments thereof containing a modified Gla domain, or synthetic peptides providing one or more carboxy rich domains. The interface molecule and the biomolecule may be formed as an engineered molecule such that the carboxy rich domain is included in a protein, a polypeptide, an antigen, an antibody, a carbohydrate, an aptamer or a lipid. For example, the carboxy rich domain may include one or more of the Gla domains of a Vitamin K dependent protein, or a fragment or a derivative thereof, and the engineered molecule may include a biomolecule such as protein A, fragment B of protein A, or an IgG molecule.

In embodiments in which the one or more carboxy rich domains are provided in the interface molecule as a synthetic peptide, the carboxyl group-containing amino acids are positioned and spaced in the synthetic peptide to ensure that sufficient free carboxyl groups are exposed on the surface of the folded synthetic peptide such that they are available for immobilizing to the aluminum oxide surface. The ability to fold in a manner to expose the free carboxyl groups on the surface of the folded synthetic peptide can readily be confirmed by molecular models, as demonstrated in the examples which follow.

The size of the interface molecule varies with the particular application for the immobilized interface and biomolecule. For instance, when the biomolecule is an engineered antibody having an interface molecule incorporated in the Fc fragment, the interface molecule is preferably sized to limit steric hindrance. For example, for an antibody having about a 50 kD sized heavy chain, the interface molecule preferably has a smaller size, such as less than about 30kD, and each of one or more carboxy rich domain portions, Gla domain portions, or modified Gla domain portions of the interface molecule provides at least 5 free carboxyl groups within about 50 consecutive amino acids. Examples of this interface molecule sizing with IgG biomolecules is shown in Examples 14 and 16-19 which follow.

The distance from the aluminum oxide surface can be controlled by varying the size of the interface molecule. Prothrombin holds the immobilized entity about 1 Onm off the surface while the Fragment 1 domain of prothrombin holds hold the immobilized entity about 3 nm off the surface. Suitable linkers can be used to vary the distance the molecules are held off the surface from about lnm to about 4 Onm.

The interface molecule can include different types of interface molecules spaced on the aluminum oxide surface, for example as shown in Figure 29, to reduce steric hindrance.

As shown in Figure 29, prothrombin is spaced apart from Fragment 1 of prothrombin. The prothrombin is bound to a biomolecule such as an antigen.

The interface molecule immobilized on the aluminum oxide surface can be activated with a cross linking agent to attach to an analyte of interest (as in Figure 31), or for attaching to a biomolecule. Alternatively, the biomolecule and the interface molecule are preassembled as a complex using cross linking agents, and the complex is then immobilized on the aluminum oxide surface. Alternatively, the interface molecule and the biomolecule are formed through genetic engineering or wet chemistry, for example by forming antigens or antibodies in large scale with appropriate binding sequences of a carboxy rich domain (e.g the first 32 amino acids of human prothrombin: Ala-Asn-Thr-Phe-Leu-Gla-Gla-Val-Arg-Lys- Gly-Asn-Leu-Gla-Arg-Gla-Cys-Val-Gla-Thr-Cys-Ser-Tyr-Gla-Gla- Ala-Phe-Gla- Ala-Leu- Gla-XXX ) attached to the end of the molecule.

In some embodiments, the interface molecule is provided on the aluminum oxide surface as a continuous or discontinuous coating, as delineated spots or lines, or as an array.

In some embodiments the aluminum oxide surface is provided on a substrate in the form of a particle, powder, thin film, slide, strip, filter, bead, magnetic bead, magnetic particle, or coating.

There are several Gla containing proteins including the Vitamin K dependent proteins: Factor P, Factor VII, Factor IX, Factor X, protein C, protein S, protein Z, osteocalcin, matrix Gla protein (MGP), bone Gla protein (BGP), GAS6 periostin, two transmembrane Gla proteins (TMGPs), and two proline-rich Gla-proteins (PRGPs). The structures of some of these Gla domain proteins are shown in Figure 1. Sequence listings are provided for Factor P (prothrombin) in SEQ ID NO. 1, Factor IX in SEQ ID NO. 2, and Protein S in SEQ ID NO. 3. The charge to mass ratios of each protein plays a key role in how it diffuses to the surface of a substrate at low concentrations. The inventor has determined that the number, location and density of the free carboxyl groups is important to the binding of the proteins to the aluminum oxide surface. In particular, as set out below, the inventor established that the volume of the carboxy rich domain and the density of free carboxy groups (-COOH) in the carboxy rich domain of the interface molecule are factors that determine the binding of the interface molecule to the aluminum oxide surface.

The volume of the Gla domain of Fragment 1 of prothrombin was established using computer software. I-Tasser generated models were opened in Molsoft ICM-Pro. The protein region of interest was selected and the other regions hidden. Under the display settings, the option to measure the distance between two atoms was selected. Two points were then selected to generate length, width, and height dimensions for the area of interest. The product of these dimensions provided the volume of the Gla domain or charged region, depending on the molecule. The same software was used to then establish the volumes and density of the free carboxy groups for the carboxy rich domains of naturally occurring proteins, protein S, FIX and Fragment 1 of bovine prothrombin, as shown in Table 1 below. Likewise, the volumes and free carboxy group densities were established for a range of synthetic carboxy rich domains based on the base structure of the Gla domain of prothrombin, as shown in Table 1, and as more fully described hereinbelow. These volumes ranged from 2.2 to 17 nm ³ with free carboxy group densities from 1.15 to 9.09/nm ³ . The volume range was found to be bracketed by the naturally occurring FIX (17 nm ³ ) and the fusion protein of two naturally occurring domains, the Gla domain of prothrombin and the Fragment B domain from protein A (2.2 nm ³ ). The synthetic carboxy rich domain volumes fell between these two proteins.

The density range is bracketed by a synthetic carboxy rich domain (FII-fragl-Asp described below, at 1.15 COOH/nm ³ ) and the fusion protein of two naturally occurring domains, the Gla domain of prothrombin and the Fragment B domain from protein A (9.09 COOH/nm ³ ), with other carboxy rich domain examples falling between these proteins. Density multiplied by volume in Table 1 provides the number of free carboxy groups in the carboxy rich domain. This ranges from 10-24 free carboxyl groups. A comparison of the synthetic protein constructs (see Figures 5-14) to Fragment 1 of Factor P (Figure 2 and Figures 15-16) shows that the basic structure is maintained with the carboxy groups exposed in a general cluster. This indicates that the synthetic constructs provide a carboxy rich domain for binding to aluminum oxide surfaces in much the same manner as do the natural vitamin K dependent molecules.

Based on examples to follow, and as further established including the molecular modelling work, the carboxy rich domain for binding to aluminum oxide surfaces provides at least 5 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , or at least 10 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , or at least 20 free carboxyl groups within a molecular volume of 2.2-25 nm ³ , or at least 10 free carboxyl groups within a molecular volume of 2.2-17 nm ³ , or at least 20 free carboxyl groups within a molecular volume of 2.2-17 nm ³ , or at least 10 free carboxyl groups within a molecular volume of 7.0- 17 nm ³ , or at least 20 free carboxyl groups within a molecular volume of 7.0-17 nm ³ .

Human prothrombin contains 10 g-carboxyglutamic acid (Gla) residues in the NH ₂ - terminal domain. The Gla residues occur in adjunct pairs or in close proximity to each other (residues 6, 7, 16, 19, 20, 25, 26, 29, 32) (Walz DA et ah, 1977; UnitProt Consortium, 20l7a). These 10 Gla residues, which have 20 free carboxyl groups, occupy a space of 8.3 nm ³ . This gives a density of 2.4 carboxyl groups per nm ³ . Prothrombin has high affinity metal binding sites for Ca ²⁺ .These binding sites are formed by two Gla residues in the polypeptide chain, which share a single bound metal ion. An intermolecular bridge forms, which stabilizes the tertiary structure of the protein. The high concentration of Gla residues at the NH ₂ - terminal domain in prothrombin, when used as an interface molecule with alumina surfaces, provides consistent binding to aluminum oxide with a specific orientation.

Figure 2 shows the 3-dimensional structure of prothrombin, which illustrates the localization of the Gla domains to a confined (8.3 nm ³ ) portion of the protein structure. As an interface molecule herein, prothrombin is found to bind to an aluminum oxide surface at the Gla domain because Gla has a high affinity to the alumina surface relative to other amino acids. Prothrombin has a molecular weight of 72,000 Da, an isoelectric point in the range of 4.7-4.9, and a dissociation constant of ~lpmol/L (Bajaj et al., 1975; Kotkow et al., 1993; Mann, 1976; UnitProt Consortium, 20l7a).

Human prothrombin is a three-domain protein consisting of Fragment 1, Fragment 2 and thrombin. The Gla residues (10) are all located in Fragment 1. It consists of 155 amino acid residues (human). The Gla residues are all located in the first 32 amino acid residues. This moiety thus has a high affinity for alumina substrates. These 10 Gla residues, which have 20 free carboxyl groups, occupy a space of 8.3 nm ³ . This gives a density of 2.4 carboxyl groups per nm ³ .

The carboxy rich domain of the interface molecule can be provided by modifying the Gla domain of a Vitamin K dependent protein, or of a fragment of a Vitamin K dependent protein, such as set out below and in the Examples which follow. The modification can be achieved by genetic engineering techniques well known to persons skilled in the art, or by wet chemistry.

Fragment 1 of Factor P can be modified with a dicarboxyl amino acids such as aspartic acid. This is shown in Figure 5 where there are 10 Asp residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 in the protein. This forms an 8.7 nm ³ carboxy rich domain at the base of the structure where binding to the alumina substrate occurs. These 10 Asp residues, which have 10 free carboxyl groups, occupy a space of 8.7 nm ³ . This gives a density of 1.15 carboxyl groups per nm ³ . The surface charge map of the 3D structure of a modified Fragment 1 of Factor P (Figure 6) shows the concentration of surface charge at the carboxy rich domain that facilitates binding to alumina.

Fragment 1 of Factor P can be modified with a pair of dicarboxyl amino acids such as aspartic-aspartic acid. In Figure 7 there are 10 (Asp-Asp) residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 7.7 nm ³ carboxy rich domain at the base of the structure where binding to the alumina substrate occurs. These 20 Asp residues, which have 20 free carboxyl groups, occupy a space of 8.7 nm ³ with a carboxyl density of 2.60 carboxyl groups per nm ³ . The surface charge map of the 3D structure of a modified Fragment 1 of Factor II (Figure 8) shows the concentration of surface charge at the carboxy rich domain which facilitates binding to alumina.

Fragment 1 of Factor P can be modified with a pair of dicarboxyl amino acids such as aspartic-glutamic acid. In Figure 9 there are 10 (Asp-Glu) residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 8.5 nm ³ carboxy rich domain at the base of the structure where binding to the alumina substrate occurs. These 10 Asp-Glu residues, which have 20 free carboxyl groups, occupy a space of

8.5 nm ³ with a carboxyl density of 2.35 carboxyl groups per nm ³ . The surface charge map of the 3D structure of a modified Fragment 1 of Factor II (Figure 10) shows the concentration of surface charge at the carboxy rich domain which facilitates binding to alumina.

Fragment 1 of Factor P can be modified with a pair of dicarboxyl amino acids such as glutamic-aspartic acid. In Figure 11 there are 10 (Glu-Asp) residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 12.5 nm ³ carboxy rich domain at the base of the structure where binding to the alumina substrate occurs. These 10 Glu-Asp residues, which have 20 free carboxyl groups, occupy a space of

12.5 nm ³ with a carboxyl density of 1.60 carboxyl groups per nm ³ . The surface charge map of the 3D structure of a modified Fragment 1 of Factor II (Figure 12) shows the concentration of surface charge at the carboxy rich domain that facilitates binding to alumina.

Fragment 1 of Factor P can be modified with a pair of dicarboxyl amino acids such as glutamic-aspartic acid. In Figure 13, there are 10 (Glu-Glu) residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 11.8 nm ³ carboxy rich domain at the base of the structure where binding to the alumina substrate occurs. These 10 Glu-Glu residues, which have 20 free carboxyl groups, occupy a space of 11.8 nm ³ with a carboxyl density of 1.69 carboxyl groups per nm ³ . The surface charge map of the 3D structure of a modified Fragment 1 of Factor II (Figure 14) shows the concentration of surface charge at the carboxy rich domain that facilitates binding to alumina.

Bovine prothrombin is a three-domain protein consisting of Fragment 1, Fragment 2 and thrombin. Figure 15 shows that the Gla residues (10) are all located in Fragment 1. It consists of 156 amino acids. The Gla residues are all located in the first 33 amino acid residues. This moiety has a high affinity for alumina substrates. These 10 Gla residues, which have 20 free carboxyl groups, occupy a space of 7.0 nm ³ . This gives a density of 2.86 carboxyl groups per nm ³ . The surface charge map of the 3D structure of bovine Fragment 1 of Factor P (Figure 16) shows the concentration of surface charge at the carboxyl rich domain that facilitates binding to alumina.

The interface molecule and the biomolecule can be formed as an engineered molecule, by genetic engineering techniques well known to persons skilled in the art, in a manner such that the carboxy rich domain is included in a protein, a polypeptide, an antigen, an antibody, a carbohydrate, an aptamer or a lipid. For example, Fragment 1 of human Factor 11, or a fragment of a Vitamin K dependent protein containing the Gla domain, can be added to the carboxyl terminus of an IgG heavy chain to provide a carboxyl rich domain at the base of the IgG structure, allowing binding of the engineered antibody to an alumina substrate. Similarly, synthetic antigens can be produced for immunoassays by adding a carboxy rich domain to a sequence of an antigen of interest. For example, there are 5-7 serotypes of Dengue vims, and synthetic antigens specific to each serotype can be engineered with the carboxy rich domain. When placed in a multiplexed pattern on the aluminum oxide surface, the disease can be rapidly diagnosed.

A modified (Asp in place of Gla) Fragment 1 of human Factor P can be added to the carboxyl terminus of an IgG heavy chain as shown in Figure 17. There are 10 Asp residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the modified Fragment 1 protein. This forms a 3.6 nm ³ carboxyl rich domain at the base of the IgG structure allowing binding to an alumina substrate. This results in a domain with a density of 3.03 carboxyl groups per nm ³ . The surface charge map of the 3D structure of a modified (Asp in place of Gla) Fragment 1 of human Factor II added to the carboxyl terminus of an IgG heavy chain (Figure 18) shows the concentration of surface charge at the carboxyl rich domain that facilitates binding to alumina.

A modified (Glu in place of Gla) Fragment 1 of human Factor P can be added to the carboxyl terminus of an IgG heavy chain as shown in Figure 19. There are 10 Glu residues in place of Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 3.7 nm ³ carboxy rich domain at the base of the IgG structure allowing binding to an alumina substrate. This results in a domain with a density of 2.70 carboxyl groups per nm ³ . The surface charge map of the 3D structure of a modified (Glu in place of Gla) Fragment 1 of human Factor P added to the carboxyl terminus of an IgG heavy chain (Figure 20) shows the concentration of surface charge at the carboxy rich domain that facilitates binding to alumina.

The Fragment B domain of Protein A can be linked to the first 32 amino acids of Fragment 1 of human Factor P. There are 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein. This forms a 2.2 nm ³ carboxy rich domain at the base of the Fragment B structure allowing binding to an alumina substrate. This results in a domain with 9.09 carboxy groups per nm ³ . The surface charge map of the 3D structure of Fragment B of Protein A linked to the first 32 amino acids of Fragment 1 of human Factor 11 (Figure 22) shows the concentration of surface charge at the carboxy rich domain at the base of the Fragment B structure allowing binding to an alumina substrate.

Human factor IX (hFIX) is a vitamin K dependent plasma serine protease that is involved in the intrinsic pathway of blood coagulation. In its pre-pro zymogen form, it consists of 461 amino acids. During biosynthesis to its mature structure, the pre- and pro peptides are removed. Mature hFIX contains 12 Gla residues near the NH ₂ -terminus, resulting in a Gla rich domain. This region of the protein is a membrane-anchoring domain with affinity to metal ions (Furie & Furie, 1988; 1992; Furie et ah, 1979). There are also twelve other metal binding sites that may have an affinity to the alumina substrate surface (UnitProt Consortium, 2017b). Figure 3 shows the 3-dimensional structure of hFIX. The image identifies the Gla rich domain that has a high affinity to metal. This domain has a volume of 17.0 nm ³ and is located at the base of the structure where binding to the alumina substrate occurs. There are 24 carboxyl groups in this domain that gives a density of 1.4 carboxyl groups per nm ³ .

The binding orientation of hFIX to an alumina substrate is at the Gla domain. At this orientation, the protein has a wide base and so it is structurally stable. This suggests that the binding orientation is be consistent because hFIX proteins are unlikely to shift orientation. This is desirable because a predictable and consistent binding orientation results in consistent properties of the immobilized protein. Factor IX has a molecular weight of 55,000 Da, an isoelectric point in the range of 4.2-4.5, and a dissociation constant of < 1.0 pmol/L (Amphlett et al., 1978; Nelsestuen et al., 1978; Thompson, 1986; UnitProt Consortium,

2017b).

Protein S is a glycoprotein that acts as a non-enzymatic cofactor to activated protein C (APC) in the degradation of coagulation factor V/V a and factor Villa. It has 11 Gla residues near the NH ₂ -terminal, which are important for membrane binding and are the site of binding onto alumina substrates. Protein S has a molecular weight of 69,000 Da, an isoelectric point in the range of 5.0-5.5 and a dissociation constant is 0.005 pmol/L (Lundwall et al., 1986; Walker, 1981; Sugo et al 1986; UnitProt Consortium, 2017c). Figure 4 shows the 3- dimensional structure of protein S and the general location of the Gla domain (residues 42- 87). The Gla domain is located in near proximity to the NH ₂ -terminal that suggests that it is exposed to the outside of the structure and is the binding site of the protein to the alumina substrate. This forms a 12.3 nm ³ carboxyglutamic domain at the base of the structure where binding to the alumina substrate occurs. There are 22 carboxyl groups in the domain that gives a density of 1.79 carboxyl groups per nm ³ .

Table 1 - Domain Volume/Density of Free Carboxyl Groups of Interface Molecules

ii) Aluminum Oxide Surface

The aluminum oxide surface may take the form of aluminum oxides and hydroxides including calcined aluminas, the various transition aluminas (e.g. gamma, eta, delta), aluminum oxide hydroxides (e.g. boehmite), and amorphous aluminas (e.g. native oxide on aluminum, anodized aluminum and aluminum alloys). In general, binding to the interface molecule occurs at cationic surface sites through ligand exchange with protonated hydroxyl surface groups or through electrostatic attraction between positively charged surface sites, such as exposed aluminum ions, and the polyanionic groups.

The aluminum oxide surface may be formed on a substrate, such as on a particle, powder, magnetic bead, thin film, slide or coating. The aluminum oxide surface may be formed by sputtering, RF sputtering, reactive sputtering, chemical vapour deposition, evaporating, casting, or extruding aluminum metal or an aluminum alloy, and may be further anodized to provide a porous anodized aluminum oxide surface.

iii) Biomolecules

The biomolecule is broadly defined above. For affinity binding assays, the

biomolecule may be a member of a binding pair: antibody-antigen, antibody-hapten, enzyme- substrate, enzyme-receptor, hapten-hormone, toxin-receptor, protein-protein, avidin-biotin, aptamer-aptamer target, protein-drug, and drug receptor-drug.

As used herein, the binding partner or bioconjugate of the biomolecule bound to the interface molecule, when used in a diagnostic test for the binding pair, is referred to as an analyte of interest.

ln general, binding between binding pairs is referred to as molecular binding.

Depending on the binding pairs, the molecular binding may be non-covalent, reversible covalent or irreversible covalent binding. Biomolecules that participate in molecular binding generally include proteins, nucleic acids, carbohydrates, lipids and small organic molecules such as drugs, whether natural, biosynthetic, synthetic or derivatized.

iv) Cross-linking agents

The interface molecule, or modified/engineered molecules of the interface molecule with biomolecule, can be activated using any number of homo- or heterobifunctional cross linking agents including: imidoesters, glutaraldehyde, carbodiimides, maleimides, haloacetyles, hydrazides, as well as any others commonly known to those skilled in the art.

An example of a homobifunctional cross linking agent is pentane- 1, 5-dial. An example of a heterobifunctional cross linking agent is 3-[2-pyridyldithio]propionyl hydrazide. The interface can then be bound through the cross linking agent to a biomolecule, such as proteins/peptides, carbohydrates, nucleic acids and lipids. The proteins/peptides, carbohydrates, nucleic acids and lipids thus bound are held at the surface, retain their bioidentity, and remain structurally intact, such that they are recognizable as the original material and will thus interact with antigens, antibodies, aptamers etc. The cross linking agent will vary depending on the chemistries of the interface molecule and the biomolecule, as is well known in the art. A list of common cross linking agents is found in

https://tools.thermofisher.com/content/sfs/brochures/l602 l63-Crosslinking-Reagents-

Handbook.pdf.

v) Anodization

Aluminum or anodizable aluminum alloy materials can be used for anodization to create a barrier or porous alumina layer. The main governing factors for the resulting anodic alumina films, include the electrolyte type (Ono, Masuko 2003, Abd-Elnaiem, Gaber 2013), strength (i.e., concentration or pH) (Belwalkar et al. 2008, Araoyinbo et al. 2010, Yim et al. 2007), and temperature (Li, Zhang & Metzger 1998, Abd-Elnaiem, Gaber 2013, Yim et al. 2007) as well as the voltage (Zhu et al. 2011, Ono, Masuko 2003, Belwalkar et al. 2008, Rahman et al. 2012, Abd-Elnaiem, Gaber 2013, Yim et al. 2007). These parameters control the porosity through pore diameter and pore wall thickness (Van Overmeere et al. 2010), along with the dissolution and etch rates of layers. Further treatments during oxidation can include pretreatment (Zhu et al. 2011) and post treatment.

When alumina is soluble in the electrolyte (e.g., oxalic, sulfuric, and phosphoric acids), dissolution of Al ³⁺ occurs and a porous alumina layer forms. During the initial period of anodization, a highly resistant Al ₂ 0 ₃ barrier film is created on the aluminum layer. Further anodization results in the propagation of individual paths through the barrier film, which are precursors to pore formation. Next, a breakdown of the barrier film and formation of the porous structure occurs. Once porous oxide formation is complete in the aluminum layer, anodization of an underlying layer occurs (e.g. Ta). When the barrier oxide layer has completely formed current density is approximately zero (Eftekhari 2008).

Of particular interest are porous anodized surfaces formed on a reflective metal for use in a visual assay. The aluminum oxide surface is provided on a reflective metal capable of generating a colour when covered by a porous layer of aluminum oxide. The aluminum oxide surface is anodized, as above, to provide a porous anodized surface, such that, when contacted with a sample to test for analyte specific to the biomolecule, a colour change is detected denoting the presence of the analyte upon binding of the biomolecule and the analyte. vi) Solvents

Suitable solvents for use in immobilizing the interface molecule on the aluminum oxide surface are typically aqueous solutions, preferably with a low salt content and devoid of anions with a high affinity for the alumina surface (e.g., phosphates, carboxylates, sulfates). Binding time can vary depending on the surface coverage, the solvent, and the pH of the solution.

vii) Signal Amplification

When the aluminum oxide surface is part of a colour generating device, for example for a visual assay, some applications may benefit from signal amplification, that is in providing a bound layer on the device that is sufficient to generate a detectable colour change. For example, if an antibody is attached to, or engineered with, the interface molecule, and the matching antigen is very small (for example less than 1.5 nm), there may not be a significant change in colour upon antigen binding as the change in the film thickness may not be large enough (generally greater than 1.5 nm). To amplify the signal (i.e., detectable colour change), a second antibody can be added to the surface after the surface has been exposed to a sample containing the antigen analyte of interest. The second antibody (specific to the antigen), binds to the antigen bound to the first antibody. This second antibody binding occurs only if the desired antigen is bound to the first antibody, but increases the film thickness, for example by about 7 nm. This increase in thickness results in a dramatic change in interference colours so as to be detectable.

One example of signal amplification is depicted in Figure 28. Figure 28 shows an aluminum oxide surface as a thin layer of aluminum oxide on an underlying reflective layer, prepared as set out above. An antibody (lgG) is engineered to have the Gla domain of Fragment 1 of Fll incorporated at the carboxy terminus of its Fc structure. The lgG is bound to the aluminum oxide surface through the Gla domain. The engineered antibody is shown bound to its specific antigen. A second antibody is shown binding to the target antigen, which can be used to amplify the signal. The second antibody can be unmodified or it can be modified with a linked enzyme or radiolabel in order to amplify the signal from the overall device.

For example, signal amplification can be used to detect small molecules such as hormone molecules. Testosterone (C ₁₉ H ₂₈ 0 ₂ ) has a molecular weight of 288 and a 19 carbon backbone ft is approximately 2 nm in length and less than 1 nm in width, putting it at the lower limit of detection from an interference colour based assay. To amplify the signal a second antibody that recognizes a testosterone-antibody complex can be added to the system. Where ever the testosterone is bound to the antibody, the second antibody binds. This increases the optical path length by about 7-10 nm, which effectively amplifies the signal and allows detection of the hormone of interest

viii) Kits and Diagnostic Applications

In some embodiments, such as for diagnostic testing, the invention extends to diagnostic systems or kits containing a diagnostic device providing an aluminum oxide surface on an appropriate substrate such as a slide. For visual assays or visual diagnostic devices, the diagnostic device includes an aluminum oxide surface on a reflective metal such as tantalum capable of generating a colour when covered by a porous layer of aluminum oxide. In such applications, the aluminum oxide surface is provided as a porous anodized surface. In some embodiments of a visual assay or visual diagnostic device, the interface molecule is provided on the aluminum oxide surface, and a specific antigen is attached to the interface molecule. Using a specific antigen, this kit detects the presence of an antibody specific to the antigen (analyte) when contacted with a test solution such as a bodily fluid, tissue or the like, since upon binding of the antigen and antibody a visible colour change occurs on the device.

In some embodiments of a visual assay or visual diagnostic device, the kit includes a diagnostic device as set out above, but conjugated with a specific antibody. In some applications, the specific antibody is engineered to include the carboxy rich domain as the interface molecule.This kit allows for the detection of specific antigens in a test solution, since upon binding of the antigen, a visible colour change occurs on the device.

In some embodiments of a visual assay or visual diagnostic device, the kit includes a diagnostic device as set out above, but attached to a bispecific engineered antibody having a first antigenic determinant (Fa ) specific to the interface molecule, and a second antigenic determinant (Fab ₂ ) specific to an analyte of interest in a test sample, such as an antigen. This kit allows for the detection of specific antigens in a test solution, since on binding of the antigen to the engineered antibody, a visible colour change occurs. Bispecific antibody synthetic techniques are well known (see for example Brinkmann, Ulrich et ah).

In some embodiments of a visual assay or visual diagnostic device, the kit includes a diagnostic device as set out above, but conjugated with an aptamer. The aptamer binds to its specific target molecule with a visible colour change.

In some embodiments of a visual assay or visual diagnostic device, the kit includes a diagnostic device as set out above, but conjugated to an engineered protein A which includes the carboxy rich domain as the interface molecule. This protein A binds to the FC region of an antibody, ensuring proper orientation of the antibody for detecting its corresponding specific antigen, with a visible colour change.

In some embodiments of a visual assay or visual diagnostic device, the kit includes a diagnostic device as set out above, but the interface molecule is modified or activated by attachment to a cross linking agent. The cross linking agent binds to a specific analyte of interest in a test sample (such as protein in urine) with a visible colour change.

In some embodiments of a visual assay or visual diagnostic device, the kit includes the aluminum oxide surface as set out above, but the interface molecule and the cross linking agent or biomolecule, are provided separately from the aluminum oxide surface, such as by providing one of the following:

a cross linking agent attached, or for attachment to, the interface molecule, the cross linking agent being capable of binding to the analyte of interest; or

a biomolecule specific to the analyte of interest, attached to, or for attachment to, the interface molecule; or

a biomolecule in the form of an engineered antibody attached to, or for attachment to, the interface molecule through a first antigenic determinant and having a second antigenic determinant specific to the analyte of interest.

The components of the kit can be included separately from the interface molecule for complexing prior to the test, or the interface molecule and the cross linking agent or biomolecule can be formed as a interface complex in the kit. The interface complex is contacted with a sample to test for the analyte, and the sample and the interface complex are then contacted with the aluminum oxide surface. The kit detects the presence of the analyte upon binding of the analyte and the interface complex to the aluminum oxide surface with a detectable colour change.

In addition to the diagnostic device, such kits typically include a container housing the device, and one or more other components. The kit can include pharmaceutical or diagnostic grade components in one or more containers, such as cross linking agents, assay standards, testing components etc. The kit can include instructions or labels promoting or describing the use of the device or components. Instructions can involve written instruction on or associated with packaging of the components. Instructions can also include any oral or electronic instructions provided in any manner, for example for mixing one or more components of the kit, and/or for isolating and analyzing a sample.

To facilitate quantification of distinct analytes in a sample, the assay component can employ standards for the analytes, where the standard is a predetermined amount of an analyte being detected, provided on the device to allow for the quantification of the analyte. Standards can be analyzed to produce working curves equating analyte amounts with the amount of analyte present in the sample. For visual assays, standards can include sample colours charts denoting the diagnostic device pre-testing, for a negative result, and for a positive result.

Techniques for protein printing are well known and can be used to print interface molecules, biomolecules and or complexes of the interface molecule and biomolecule (see for example, Delaney, Joseph T et ah, McWilliam I, et ah, and Li, J et ah). Protein printing, for example with an inkjet printer can be used to deposit arrays of different proteins to multiplex the tests. Each protein generates a colour that changes if their target molecules (ex. Antibody, antigen, aptamer, DNA strand or R A strand) are present. For example immunoassays can be multiplexed by placing different proteins on the aluminum oxide surface in a specific pattern. These patterns can be in the form of straight lines, curves, circles, dots, or complex patterns. They provide information on a variety of analytes in the test sample. The results are determined by changes in the colours of various portions of the printed patterns. For example, a variety of antigens can be bound to the diagnostic surface that are characteristic of specific viruses such as those that cause Chikungunya, Dengue Fever, Yellow Fever and Zika. The diagnostic technique identifies not only the disease, but also the serotype present, based upon colour change in specific regions. This multiplexed test determines cases in which a patient has been infected with one or more than one vims or serotype from a single sample.

In some applications, analyte-specific binding to the biomolecule can be detected using any suitable detector, and will depend on the type of test or assay being conducted. In general, the detector includes an illumination source and detection electronics. The light source can be daylight, for example for a point-of-care diagnostic, or other light sources such as LEDs, lasers and filament lamps. These sources can be used in conjunction with optical filters, polarizers, diffraction gratings and other optical components to provide a specified spectral component of light. Other forms of radiation such as bioluminescence, fluorescence, and others can be used. Excitation wavelengths my be in the visible portion of the spectrum (300-700 nm wavelength), or other wavelengths such as infrared and ultraviolet. The absorbed, reflected, or re-emitted light can then be observed and/or detected using the eye (for visible wavelengths), or using photosensitive detectors such as photodiodes or

photomulipliers, in combination with spectral and/or spatial filtering.

If colour changes are below the visible detection limit and not detectable by eye, sensitive methods for detection are available. Digital image analysis, spectrophotometric and other photon counting detectors allow for the analysis of shifts in reflected wavelengths. The capture of high resolution images and digital processing is commonly used in biological studies to quantify and analyse colour paterns, so are techniques that are well understood in the art. Quantification of visible colours can be achieved with digital processing. To distinguish colours, plots can be generated using the International Commission on

Illumination or Commission Internationale de I’Ecla (CIE) colour space that allows for 2D plots of chromaticity coordinates. Spectrophotometers can provide analysis with full spectrum measurements on reflectance properties, beyond what is detectable by the human eye. For very low levels of light intensity, photon counting detector assemblies (ex. photomultipliers can be used to measure the number of photons by multiplying the signal prior to detection.

One example of a suitable detector is a reflective spectrometer which measures reflectance of reflecting surfaces. Alternatively the detector can be a camera or imaging device. The detection can be at the point-of-care site, such as for a visual assay, or can be remote from the sample collection or patient site.

The methods of the present invention can extend to a computer for data integration, analysis, storage and transmission in order to integrate the detected analyte-specific binding with the data acquired by any on of the data acquisition components. The binding and data thus integrated can then be analyzed and stored by the computer for subsequent access. The computer typically includes an operating system that accesses one or more algorithms and/or software to analyze data from the assay component to determine the presence and/or quantity of analytes that are being tested, for example by comparing to compiled or standard curves. The raw data and/or integrated and analyzed data are displayed on a display screen, such as a computer display or cell phone, PDA and the like.

A barcode reader can be provided for automatically entering information about an assay component. The barcode reader can be combined with a barcode activation system which identifies the test to be analyzed and automatically initiates one-point assay calibration of that particular test to reduce user errors. Each individual assay component can contain a unique barcode to be read and used to initialize the apparatus such that the appropriate algorithms are employed. Exemplary embodiments of assay components in accordance with the invention are shown in Figures 36A (barcode) and 36B (QR code).

The diagnostic device can be adapted as a sample apparatus, such as a hand held device or kit, including one or more of the components such as described above

ix) Applications

The aluminum oxide surface on which an interface providing carboxy rich domain is immobilized in accordance with this invention has far reaching applications, including for example: protein coatings on various surfaces for a variety of products such as in biomedical applications; coatings on anodic thin films for medical diagnostic devices as visual immunoassays, coatings on anodic thin films for environmental monitoring; coatings on anodic thin films for food safety diagnostic devices; coatings on anodic films in microtitre plates for EL1SA testing and the like; coatings on MEMS/NEMS for diagnostics, coatings on nanoparticles for targeted therapeutic delivery; coatings on nanoparticles for contrast imaging; alumina particles for column based affinity separation applications; magnetic particles for affinity recovery of biomolecules and cells; research assay to verify veracity of antibodies before testing; coatings for plastic sample cups and lids to facilitate testing, and coating on medical devices or implants.

The present invention is also illustrated by the following non-limiting examples.

EXAMPLES

(i) Aluminum Oxide Surfaces and Interface Molecules as Affinity Device

Example 1

This example demonstrates the bonding of a protein (specifically a viral antigen) to an alumina substrate on which prothrombin is immobilized as an interface molecule. Alumina substrates were made by first sputtering tantalum to a thickness of 200nm onto an amorphous support, followed by sputtering aluminum to a thickness of 120nm on the Ta. Aluminum was converted to alumina through electrochemical oxidation (anodization) which was carried out using phosphoric acid anodization at 8V with an electrolyte of 0.4 M phosphoric acid until the current decayed to near 0 mA. This produced a layer of Al ₂ 0 ₃ with a thickness of approximately l90nm on top of a Ta^ layer of thickness 14nm. The colour of the resulting device, when observed through a polarizer with white light at 15°, was gold. Prothrombin was placed on the alumina surface (1 OpL of a 1 mg/mL solution) for 30 minutes in a 100% RH environment and then rinsed thoroughly with deionized water. After allowing the device to air dry the colour had changed to rust. The device was then immersed in a 0.5% (v/v) glutaraldehyde solution with a 25mM phosphate buffer at pH of 6.9 for 60 minutes.

Following another rinse with deionized water, an antigen for the Influenza B virus (1 OpL of a 1.75 mg/mL solution), strain Hong Kong 5/72, was placed on the surface for 60 minutes. The device was then rinsed in a 25mM phosphate buffer for 60 minutes and allowed to air dry, upon which the colour had shifted to lavender. The shift in colour was caused by the increase in optical path length which resulted from the immobilization of the antigen.

Example 2

This example illustrates that proteins such as IgG do not bind to the anodic alumina surface. Devices were made as in example 1 and prothrombin was immobilized on the surface in two of three delineated spots with the same procedure as described in Example 1.

A rust colour was observed after immobilization of prothrombin. Antiprothrombin was added to one spot with prothrombin and the remaining blank spot for 15 minutes (1 OpL of a 100 pg/mL solution) and rinsed off with deionized water. The device was allowed to air dry and then observed at 15°, as in Example 1. A burgundy spot was observed on the spot that contained both the antigen (prothrombin) and the protein antibody (antiprothrombin). The colour shift indicated that the proteins immobilized were recognized by the antibody which bound to them and changed the overall thickness of the protein layer. There was no observable colour shift on the bare anodic oxide surface that had antiprothrombin added to it. This indicates that there was not enough residual protein on the surface to alter the optical path length of the light, so the protein did not bind to the anodic surface.

Example 3

This example shows how an immobilized protein can be made specific by capping the glutaraldehyde with an amino protein, specifically tris(hydroxymethyl)aminomethane), such than no bonding occurs with the immobilized prothrombin.

A device was made, prothrombin was immobilized on the surface, and a

glutaraldehyde solution was added, as described in Example 1. The device was then submersed in a lx TBS buffer to cap all glutaraldehyde open for binding. A layer of protein (specifically an IgG antibody, goat raised anti-influenza B) was placed on the surface for 30 minutes and then rinsed off with deionized water. The device was allowed to air dry and then was observed at 15°. There was no visible colour change noticed. This lack of colour change indicated that the optical path length was not altered by the exposure to the nonspecific antibody that was added. This is important to show that the device functions as an affinity substrate without false positives from non-specific proteins.

Example 4

This example illustrates a visual assay for an antibody-antigen complex on the surface by binding the antigen to a prothrombin interface. Devices were made, prothrombin was immobilized on the surface, and the cross-linker was added as described in Example 1. A synthetic surface antigen for Hepatitis B (Hep B) was then bonded to the prothrombin using the same procedure as in Example 1. A rust colour was observed after bonding of the antigen.

The remaining glutaraldehyde was then capped with a 0.1M L-lysine solution in 0.3M phosphate buffer of pH 7.0 for 60 minutes. Finally, serum samples (n=20) were placed on the surface for 15 minutes and rinsed off with deionized water. The device was allowed to air dry and then observed at 15° where a magenta spot was observed. The colour shift indicated that the bonded proteins were recognized by the antibody which bound to them and changed the overall thickness of the protein layer.

The above examples use the naturally occurring protein prothrombin as an interface coating on alumina. Prothrombin was then activated with a homo- or hetero-bifunctional molecule to link the protein of interest to the prothrombin. If a shorter interface molecule than prothrombin (~l0nm) is desired, (e.g. ~3nm) then Fragment 1 of the prothrombin molecule can be used in place of prothrombin, as shown in the examples below. Glutaraldehyde or any other homo- or hetero-bifunctional molecule can then be used to link biomolecules of interest to the Fragment 1.

Example 5

This example illustrates the ability to form an antibody-antigen complex on the surface by binding the antigen to Fragment 1 of FII. Devices were made as in Example 1 , and Fragment 1 was immobilized on the alumina surface instead of prothrombin. Fragment 1 was placed on the device surface ( l iJpL of a 0.3 mg/mF solution) for 30 minutes in a 100% RH environment and then rinsed thoroughly with deionized water. After allowing the device to air dry the colour had changed to light rust. An antigen to the Hepatitis B vims was then bonded on the surface using the same cross-linking procedure as in Example 1. A rust colour was observed after immobilization of the Hepatitis B antigen.

The remaining glutaraldehyde was then capped with a 0.1M L-lysine solution in 0.3M phosphate buffer of pH 7.0 for 60 minutes. Finally, the anti-Hep B antibody was placed on the surface for 15 minutes (l OpL of a 100 pg/mL solution) and rinsed off with deionized water. The device was allowed to air dry and then observed at 15° where a purple spot was observed. The colour shift indicated that the proteins immobilized were recognized by the antibody which bound to them and changed the overall thickness of the protein layer.

Example 6

This example illustrates an interface molecule immobilized on a aluminum oxide surface and activated by attaching the interface molecule to a cross linking agent, for use in binding to an analyte in a test sample. Devices were made as in Example 1 and prothrombin was immobilized on the aluminum oxide surface in two spots. On one spot, the prothrombin was activated by attaching a glutaraldehyde cross linking agent, as described in Example 1. The other spot was not modified. Both spots were rust coloured. A solution of nonsense IgG (i.e., not specific for prothrombin) was added to both spots and allowed to sit for 60 minutes followed by a washing step and air drying. The spot containing the activated prothrombin bound the nonsense IgG, as indicated by a colour change to purple. The second spot did not change colour indicating that no binding occurred. This example is schematically illustrated in Figure 31, and is illustrative of an application of a diagnostic test to detect protein in urine, as a point of care device.

(ii) Engineered Biomolecules with Carboxy Rich Interface

Example 7

This example illustrates an interface molecule engineered as an anchor protein with a carboxy rich amino acid sequence for binding to the alumina surface. Here, the first 32 amino acids from Fragment 1 from human Factor P are modified and coupled directly to an aluminum oxide surfaces in a stable form. The modified Fragment 1 is bound to the alumina surface through the cluster of carboxyl groups associated with the dicarboxyl amino acid Asp. Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein are substituted for with Asp. This halves the number of carboxylic groups available for binding to the surface relative to native Fragment 1 in Example 6. This substitution results in a carboxy rich domain at the amino terminus of Fragment 1 as shown in the computer model of the structure (Fig. 5), generated by protein folding software. The sequence listing for the modified Gla domain of the Fl fragment is provided in SEQ ID NO. 4. The computer model of the surface charge on the protein (Fig. 6) supports this, showing the carboxy rich domain at the base of the engineered interface molecule where binding to the alumina substrate occurs.

Example 8

This example illustrates an interface molecule engineered as an anchor protein with a carboxy rich amino acid sequence for binding to the alumina surface. Here, the first 32 amino acids from Fragment 1 from human Factor P are modified and coupled directly to an aluminum oxide surfaces in a stable form. The modified Fragment 1 is bound to the alumina surface through the cluster of carboxyl groups associated with the dicarboxyl amino acids Asp (a dicarboxylic amino acid). Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein are substituted for with Asp-Asp. This retains the number of carboxylic groups available for binding to the surface relative to native Fragment 1 in Example 6. This substitution results in a carboxy rich domain at the amino terminus of Fragment 1 as shown in the computer model of the structure (Fig 7). The sequence listing for the modified Gla domain of the Fl fragment is provided in SEQ ID NO. 5. The computer model of the surface charge on the protein (Fig 8) supports this., showing the carboxy rich domain at the base of the engineered interface molecule.

Example 9

This example illustrates an interface molecule engineered with a carboxy rich amino acid sequence for binding to the alumina surface. The first 32 amino acids from Fragment 1 from human Factor P are modified for direct coupling to an aluminum oxide surface in a stable form. The modified Fragment 1 is bound to the alumina surface through the cluster of carboxyl groups associated with the dicarboxyl amino acids Asp-Glu. Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein are substituted for with Asp-Glu. This retains the number of carboxylic groups available for binding to the surface relative to native Fragment 1 in Example 6. This substitution results in a carboxy rich domain at the amino terminus of Fragment 1 as shown in the computer model of the structure (Fig. 9). The sequence listing for the modified Gla domain of the Fl fragment is provided in SEQ ID NO.

6. The computer model of the surface charge on the protein (Fig. 10) supports this, showing the carboxy rich domain at the base of the engineered interface molecule. Example 10

This example illustrates an engineered interface molecule, in which the first 32 amino acids from Fragment 1 from human Factor P are modified for coupling directly to an aluminum oxide surfaces in a stable form. The modified Fragment 1 is bound to the alumina surface through the cluster of carboxyl groups associated with the dicarboxyl amino acids Glu-Asp. Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein are substituted for with Glu-Asp. This retains the number of carboxylic groups available for binding to the alumina surface relative to native Fragment 1 in Example 6. This substitution results in a carboxy rich domain at the amino terminus of Fragment 1 as shown in the computer model of the structure (Fig. 11). The sequence listing for the modified Gla domain of the Fl fragment is provided in SEQ 1D NO. 7. The computer model of the surface charge on the protein (Fig. 12) supports this, showing the carboxy rich domain at the base of the engineered interface molecule.

Example 11

This example illustrates a further engineered interface molecule. The first 32 amino acids from Fragment 1 from human Factor P are modified for coupling directly to an aluminum oxide surfaces in a stable form. The modified Fragment 1 is bound to the alumina surface through the cluster of carboxyl groups associated with the dicarboxyl amino acids Glu-Glu. Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein are substituted for with Glu-Glu. This retains the number of carboxylic groups available for binding to the surface relative to native Fragment 1 in Example 6. This substitution results in a carboxy rich domain at the amino terminus of Fragment 1 as shown in the computer model of the structure (Fig. 13). The sequence listing for the modified Gla domain of the Fl fragment is provided in SEQ 1D NO. 8. The computer model of the surface charge on the protein (Fig. 14) supports this, showing a carboxy rich domain at the base of the engineered molecule.

Example 12

This example illustrates an interface molecule from Fragment 1 from bovine Factor P. Bovine Fragment 1 has an insertion of an extra amino acid at position 4 relative to human Fragment 1. The use of bovine Fragment 1 results in a carboxy rich domain at the amino terminus of Fragment 1 as shown in the computer model of the structure (Fig. 15). The sequence listing for the modified Gla domain of the Fl fragment is provided in SEQ ID NO. 9. The computer model of the surface charge on the protein (Fig. 16) supports this, showing the carboxy rich domain at the base of the molecule.

(iii) Genetically engineered or chemically derivatized protein/antibody with carboxy rich domain

Example 13

This example illustrates an antibody (here an IgG heavy chain) that is chemically altered or engineered with a carboxy rich amino acid sequence from Fragment 1 of human Factor P, for coupling directly to an aluminum oxide surface in a stable form. This engineered molecule is bound or incorporated into the carboxyl terminus of the Fc fragment of the antibody. Gla located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein are substituted for with Asp. This halves the number of carboxylic groups available for binding to the surface relative to natural Fragment 1. This substitution results in a carboxy rich domain at the amino terminus of modified Fragment 1 as shown in the computer model of the structure (Fig. 17). The sequence listing for the modified Gla domain of the Fl fragment is provided in SEQ ID NO. 10. The computer model of the surface charge on the protein (Fig.

18) supports this, showing a carboxy rich domain at the base of the engineered molecule for binding to alumina.

Example 14

This example illustrates an IgG heavy chain antibody chemically altered or engineered with a carboxy rich amino acid sequence from Fragment 1 of human Factor P, for coupling directly to an aluminum oxide surfaces in a stable form. The engineered molecule is bound or incorporated into the carboxyl terminus of the Fc fragment of an antibody. Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the protein are substituted for with Glu, which halves the number of carboxylic groups available for binding to the surface relative to natural Fragment 1. This substitution results in a carboxy rich domain at the amino terminus of Fragment 1 as shown in the computer model of the structure (Fig. 19). The sequence listing for the modified Gla domain of the F 1 fragment with the heavy chain IgG antibody is provided in SEQ ID NO. 11. The computer model of the surface charge on the protein (Fig. 20) supports this, showing a carboxy rich domain at the base of the engineered molecule for binding to the alumina surface.

Example 15

This example illustrates another engineered interface-biomolecule. The Fragment B domain of Protein A is linked to the first 32 amino acids of Fragment 1 of human Factor II. There are 10 Gla residues located at sites 6, 7, 14, 16, 19, 20, 25, 26, 29, and 32 on the Fragment 1 protein. The engineered molecule is shown in Fig. 21, as generated in the computer model. The sequence listing for the engineered molecule is provided in SEQ ID NO. 12. This provides a 2.2nm ³ carboxy rich domain at the base of the Fragment 1 -Fragment B structure, allowing binding to an alumina substrate. This results in a carboxy rich domain with 9.09 carboxy groups per nm ³ . The surface charge map of the 3D structure of Fragment B of Protein A linked to the first 32 amino acids of Fragment 1 of human Factor P (Fig. 22) shows the concentration of surface charge at the base of the engineered molecule for binding to the alumina substrate.

Example 16

Gla sequences containing one or more of the Gla domain from Factor II, Factor VII, Factor X, Factor P with a Kringle domain, Protein C, Protein Z, and a synthetic peptide sequence having a carboxy rich domain (see US Patent No. 9,694,048 to Bauzon et al.), were cloned onto each of the Fc fragments of an IgG (anti Gaussia luciferase). The cloning techniques used were standard Fc engineering techniques at the Centre for the

Commercialization of Antibodies and Biologies at the University of Toronto. (Liu et al.). The Gla domains from these proteins or from the synthetic peptide were cloned as units from 1 to 3, such that one, two or three Gla domains from the protein were present on each of the Fc fragments of the IgG biomolecule. These are designated as Gla lx, Gla 2x or Gla 3x in Table 2. The cloning techniques used a flexible linker sequence (Gly-Gly-Gly-Gly-Ser, as shown in SEQ ID NO. 20) and an Open Reading Frame (SEQ ID NO. 21) to attach the Gla domain to the Fc fragment. The flexible linker sequence was also used to link the Gla domains when more than one was attached per heavy chain of the Fc fragment. Other linkers and ORF are well known in the art and may be used. The cloning technique used subclone 4275 hGl, where hGl refers to the Ab isotype that is expressed as (i.e. human IgGl), and 4275 is the antibody that specifically binds to Gaussia luciferase and is used as the standard negative control for this mammalian in vivo work (Liu et al.). The cloning added the Gla sequence to the carboxy terminus of both heavy chains of the Fc fragment. Thus, a Glalx, as listed in Table 2 had two Gla domains, a Gla2x had 4 Gla domains, and a Gla3x had 6 Gla domains cloned on the Fc fragment of the IgG. All clones were expressed in 10 ml volumes. These expressed forms utilized the native form of the Gla domain sequence in each case. Table 2

Source of GLA SEQ ID NO.

Factor VII Glalx SEQ ID NO. 16

Factor VII Gla2x SEQ ID NO. 16

Factor VII Gla3x SEQ ID NO. 16

Factor X Glalx SEQ ID NO. 13

Protein C Glalx SEQ ID NO. 15

Protein C Gla2x SEQ ID NO. 15

Protein C Gla3x SEQ ID NO. 15

Factor P Glalx SEQ ID NO. 14

Factor P Gla3x SEQ ID NO. 14

Factor P Gla with Kringle lx SEQ ID NO. 19

Factor P Gla with Kringle 2x SEQ ID NO. 19

Synthetic Peptide Glalx SEQ ID NO. 18

Synthetic Peptide Gla2x SEQ ID NO. 18

Synthetic Peptide Gla3x SEQ ID NO. 18

Protein Z Glalx SEQ ID NO. 17

Protein Z Gla2x SEQ ID NO. 17

Protein Z Gla3x SEQ ID NO. 17

Example 17

This example illustrates the strong colour shifts that resulted with immobilizing engineered antibodies from Example 16 on a range of initial aluminum sputtered thicknesses. The thin film device was formed by sputtering a tantalum layer 200 nm thick onto substrates, followed by sputtering an aluminum layer ranging between 90 to 140 nm thick. The thin metallic films were then anodized in a mixed electrolyte containing 0.4 M phosphoric acid and 0.1 M oxalic acid with an applied constant potential of 4 V. Upon removal from the electrolyte bath the surface was thoroughly rinsed with distilled water. The device surface was subsequently exposed to protein solutions of native prothrombin (0.004 mg) and recombinant IgGs modified with Gla domains on the Fc region, as described in Example 16 (Table 2). Proteins were exposed for 30 minutes, after which the solution was removed, rinsed with distilled water and air dried. Regardless of the alumina thickness, the engineered IgGs (Gla lx) (-150 kDa) generated a strong colour shift on the device surface when viewed at 75 degrees from normal as did the native prothrombin. As the number of cloned Gla units increased from 1-3, the ability to generate a colour shift declined such that, for the Gla3x engineered antibodies, the colour shift was only slightly visible to the human eye, although other detection techniques may be used, as indicated above. These colour changes indicate that engineered proteins can be successfully immobilized to tailored alumina thicknesses and generate varying colour shifts within the first and second order colour regions.

Example 18

This example illustrates that the engineered antibodies of Example 16 were immobilized to surface through the Gla modification of Example 16, while the IgG molecules on their own did not lead to a visible colour shift. Surfaces coated with engineered human antibodies of Example 16 were subsequently coated with a solution of either goat anti-human (GAH) IgG (0.02 mg) or goat anti-mouse (GAM) IgG (0.02 mg). When GAH IgG (pos. control) was exposed to the spot with immobilized engineered antibody an increase in the colour shift resulted, whereas when GAM IgG (neg. control) was exposed to the spot with immobilized engineered human antibody no visible colour shift resulted. Both GAH and GAM solutions were also exposed to the bare device surface for 30 minutes and no colour shift resulted from either solution. This clearly demonstrated a colour shift due to the formation of a secondary IgG layer when GAH detected the adsorbed recombinant antibody on the surface.

Example 19

This example illustrates the strong colour shifts that resulted when the immobilized engineered antibodies of Example 16 were exposed to their specific antigen, Gaussia luciferase. The thin film devices were formed by sputtering a tantalum layer 200 nm thick onto silicon substrates, followed by sputtering an aluminum layer that was 110 nm thick. The thin metallic films were then anodized in a mixed electrolyte containing 0.4 M phosphoric acid and 0.1 M oxalic acid with an applied constant potential of 4 V. Upon removal from the electrolyte bath the surface was thoroughly rinsed with distilled water. The device surface was subsequently exposed to protein solutions of native prothrombin (0.004 mg) and a recombinant IgG (anti-Gaussia luciferase) modified with Gla domains (Factor VII Glalx) on the Fc region as described in Example 16. Proteins were exposed for 30 minutes, after which the solution was removed, rinsed with distilled water and air-dried. The engineered IgGs (-150 kDa) generated a strong colour shift (light purple) on the device surface when viewed at 75 degrees from normal as did the native prothrombin. The surfaces were then exposed to a 0.2 mg/ml solution of Fragment 1 from prothrombin (Factor P) to cap any active sites on the surface. No further colour change was observed after rinsing in distilled water and drying. Solutions of Gaussia luciferase (0.2mg/ml) or procalcitonin (0.2mg/ml) were then exposed to the surface for 30 minutes followed by a rinse and drying. The surfaces with the engineered IgG that were exposed to Gaussia luciferase changed to violet while the surfaces exposed to procalcitonin did not change colour. The prothrombin treated surfaces did not change colour with either Gaussia luciferase or procalcitonin. These colour changes indicate that engineered proteins can be successfully immobilized to alumina surfaces and generate colour shifts when exposed to their specific antigens. The engineered and immobilized antibodies retain their functionality.

INCORPORATION BY REFERENCE AND VARIATIONS

All references throughout this application, for example patent documents including issued or granted patents or equivalents; patent application publications; and non-patent literature documents or other source material; are hereby incorporated by reference herein in their entireties, as though individually incorporated by reference, to the extent each reference is at least partially not inconsistent with the disclosure in this application (for example, a reference that is partially inconsistent is incorporated by reference except for the partially inconsistent portion of the reference).

The terms and expressions which have been employed herein are used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by preferred embodiments, exemplary embodiments and optional features, modification and variation of the concepts herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention as defined by the appended claims. The specific embodiments provided herein are examples of useful embodiments of the present invention and it will be apparent to one skilled in the art that the present invention can be carried out using a large number of variations of the devices, device components, methods steps set forth in the present description. As will be obvious to one of skill in the art, methods and devices useful for the present methods can include a large number of optional composition and processing elements and steps.

Whenever a range is given in the specification, for example, a temperature range, a time range, a size range, or a composition or concentration range, all intermediate ranges and subranges, as well as all individual values included in the ranges given are intended to be included in the disclosure. It will be understood that any subranges or individual values in a range or subrange that are included in the description herein can be excluded from the claims herein.

All patents and publications mentioned in the specification are indicative of the levels of skill of those skilled in the art to which the invention pertains. References cited herein are incorporated by reference herein in their entirety to indicate the state of the art as of their publication or filing date and it is intended that this information can be employed herein, if needed, to exclude specific embodiments that are in the prior art. For example, when composition of matter are claimed, it should be understood that compounds known and available in the art prior to Applicant's invention, including compounds for which an enabling disclosure is provided in the references cited herein, are not intended to be included in the composition of matter claims herein.

As used herein,“comprising” is synonymous with "including," "containing," or "characterized by," and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps. As used herein, "consisting of' excludes any element, step, or ingredient not specified in the claim element. As used herein, "consisting essentially of' does not exclude materials or steps that do not materially affect the basic and novel characteristics of the claim. In each instance herein any of the terms "comprising", "consisting essentially of' and "consisting of' may be replaced with either of the other two terms. The invention illustratively described herein suitably may be practiced in the absence of any element or elements, limitation or limitations which is not specifically disclosed herein.

One of ordinary skill in the art will appreciate that starting materials, biological materials, reagents, synthetic methods, purification methods, analytical methods, assay methods, and biological methods other than those specifically exemplified can be employed in the practice of the invention without resort to undue experimentation. All art-known functional equivalents, of any such materials and methods are intended to be included in this invention. The terms and expressions which have been employed are used as terms of description and not of limitation, and there is no intention that in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by preferred embodiments and optional features, modification and variation of the concepts herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention as defined by the appended claims.

REFERENCES

Abd-Elnaiem, A. & Gaber, A. 2013, "Parametric study on the anodization of pure aluminum thin film used in fabricating nano-pores template", lntemational Journal of Electrochemical Science, vol. 8, no. 7, pp. 9741-9751.

Amphlett, G. W., Byrne, R., & Castellino, F. J. (1978). The binding of metal ions to bovine factor IX. The Journal of Biological Chemistry, 253(19), 6774-6779.

Araoyinbo, A.O., Ahmad Fauzi, M.N., Sreekantan, S. & Aziz, A. 2010, "A novel process to produce nanoporous aluminum oxide using titration technique to prepare the neutral electrolyte", Journal ofNon-Crystalline Solids, vol. 356, no. 20-22, pp. 1057-1060.

Bajaj, S. P., Butkowski, R. J., & Mann, K. G. (1975). Prothrombin fragments. Ca2+ binding and activation kinetics. The Journal of Biological Chemistry, 250(6), 2150-2156.

Belwalkar, A., Grasing, E., Van Geertruyden, W., Huang, Z. & Misiolek, W.Z. 2008, Effect of processing parameters on pore structure and thickness of anodic aluminum oxide (AAO) tubular membranes.

Bowen, W. R., & Gan, Q. (1992). Properties of Micro filtration Membranes: The Effects of Adsorption and Shear on the Recovery of an Enzyme. Biotechnology and

Bioengineering, 40, 491-497.

Brinkmann, U., Kontermann, R., The making of bispecific antibodies. MABS 2017, 9, 2, 182-212 http://dx.doe.org/l0.l080/l9420862.20l6. l268307.

Eftekhari, A. 2008, Nanostructured materials in electrochemistry, Weinheim: Wiley- VCH, Chichester.

Furie, B., & Furie, B. C. (1988). The molecular basis of blood coagulation. Cell,

53(4), 505-518.

Furie, B. C., Blumenstein, M., & Furie, B. (1979). Metal binding sites of a gamma- carboxyglutamic acid-rich fragment of bovine prothrombin. The Journal of Biological Chemistry, 254(24), 12521-12530.

Delaney, J., Smith, P., Schubert U., lnkjet printing of proteins, SoftMater, 2009, 5, 4866-4877. D0l: l0. l039/b909878j. Kotkow, K. J., Furie, B., & Furie, B. C. (1993). The interaction of prothrombin with phospholipid membranes is independent of either kringle domain. The Journal of Biological Chemistry, 268(21), 15633-15639.

Li, F., Zhang, L. & Metzger, R.M. 1998,“On the growth of highly ordered pores in anodized aluminum oxide”, Chemistry of Materials, vol. 10, no. 9, pp. 2470-2480.

Li, J., Rossignol, F., and Macdonald, J. Inkjet printing for biosensor fabrication; combining chemistry and technology for advanced manufacturing, Lab Chip, 2015, 15, 2538- 2558 DOI: l0. l039/c5 lc00235d.

H. Liu, A. Saxena, S. S. Sidhu, and D. Wu, Fc Engineering for Developing

Therapeutic Bispecific Antibodies and Novel Scaffolds, Front Immunol. 2017; 8: 38.

Lundwall, A, Dackowski, W., Cohen, E., Shaffer, M., Mahr, A., Dahlback, B., et al. (1986). Isolation and sequence of the cDNA for human protein S, a regulator of blood coagulation. Proceedings of the National Academy of Sciences of the United States of America, 83(18), 6716-6720.

Mann, K. G. (1976). Prothrombin. Methods Enzymol, 45(pt. B), 123-156.

McWilliam L, Chong Kwan M., Hall D., Inkjet printing for the production of protein microarrays. Methods Mol Biol. 2011;785:345-51 DOI: l0. l007/978-l-6l779-286-l_23.

Murray, J. P. (1980). Physical Chemistry of Virus Adsorption and Degradation on Inorganic Surfaces: Its relation to wastewater treatment. Cincinnati, OH: United States Environmental Protection Agency.

Murray, J. P., & Laband, S. J. (1979). Degradation of Poliovirus by Adsorption on Inorganic Surfaces. Applied and Environmental Microbiology, 37(3), 480-486.

Nelsestuen, G. L., Kisiel, W., & Di Scipio,R.G. (1978). Interaction of vitamin K dependent proteins with membranes. Biochemistry, 17(11), 2134-2138.

Ono, S. & Masuko, N. 2003, "Evaluation of pore diameter of anodic porous films formed on aluminum", Surface and Coatings Technology, vol. 169-170, pp. 139-142.

Rahman, M.M., Garcia-Caurel, E., Santos, A., Marsal, L.F., Pallares, J. & Ferre- Borrull, J. 2012, "Effect of the anodization voltage on the pore-widening rate of nanoporous anodic alumina", Nanoscale Research Letters, vol. 7, no. 1, pp. 474-474.

Sugo, T., Dahlback, B., Holmgren, A., & Stenflo, J. (1986). Calcium binding of bovine protein S. effect of thrombin cleavage and removal of the gamma-carboxyglutamic acid-containing region. The Journal of Biological Chemistry, 261(11), 5116-5120. Thompson, A. R. (1986). Structure, function, and molecular defects of factor IX. Blood, 67(3), 565-572.

Thurman, R. B., & Gerba, C. P. (1988). Characterization of the effect of aluminum metal on poliovirus. Journal of Industrial Microbiology, 3, 33-38.

UnitProt Consortium. (2017a). UniProtKB - P00734 (thrb_human). Retrieved 02/04, 2017, from http://www.uniprot.org/uniprot/P00734

UnitProt Consortium. (2017b). UniProtKB - P00740 (fa9_human). Retrieved 04/04, 2017, from http://www.uniprot.org/uniprot/P00740

UnitProt Consortium. (20l7c). UniProtKB - P07225 (pros_human). Retrieved 03/18, 2017, from http://www.uniprot.org/uniprot/P07225

Van Overmeere, Q., Blaffart, F. & Proost, J. 2010, "What controls the pore spacing in porous anodic oxides?", Electrochemistry Communications, vol. 12, no. 9, pp. 1174-1176.

Walker, F. J. (1981). Regulation of activated protein C by protein S. the role of phospholipid in factor va inactivation. The Journal of Biological Chemistry, 256(21), 11128- 11131.

Walz DA, Hewett-Emmett D, Seegers WH. (1977). Amino acid sequence of human prothrombin fragments 1 and 2. Proceedings of the National Academy of Sciences of the United States of America, 74(5), 1969.

Wefers, K., & Misra, C. (1987). Oxides and Hydroxides of Aluminum: Alcoa Technical Paper No. 19, Revised. Alcoa Laboratories.

Yim, S., Bonhote, C., Lille, J. & Wu, T. 2007, "Parameter effects for the growth of thin porous anodic aluminum oxides", ECS Transactions, vol. 6, no. 8, pp. 267-274.

Zhu, Z., Garcia-Gancedo, L., Qiang, L., Flewitt, A., Milne, W.I. & Moussy, F. 2011, "Size-tunable porous anodic alumina nano-structure for biosensing", Soft Nanoscience Letters, vol. 1, no. 3, pp. 55-60.