WO1997005489A1 - Soluble and polymeric metal complexing materials for measurement of sugars and related molecules in solution - Google Patents

Abstract

A sensor for use in detecting the presence of sugars and related molecules in aqueous solutions. The sensor is composed of a metal complex that binds to the targeted sugar or other molecule in aqueous solution with the result being the release of a proton. Measurement of proton release from the sensor provides an indirect indication of target molecule concentration. The metal complexes may be attached to support structures to provide both anchoring and positioning of the metal ions to increase selectivity of sugar/metal complex interactions. Detection systems in which pH is used as an indication of proton release are disclosed. Methods for monitoring the concentrations of sugars and related molecules using the metal complex sensors are also disclosed.

Description

SOLUBLE AND POLYMERIC METAL COMPLEXING MATERIALS FOR MEASUREMENT OF SUGARS AND RELATED MOLECULES IN SOLUTION

BACKGROUND OF THE INVENTION

1 . Field of the Invention

The present invention relates generally to devices and methods used to test for and monitor the presence of sugars and chemically related compounds in solutions. More particularly, the present invention is directed to sensors and methods which rely on metal coordination interactions between electronegative functionalities on the targeted compounds and metal ion complexes to provide detection and/or measurement of samples of sugars or related compounds in aqueous solutions.

2. Description of Related Art

Many different devices and methods are presently being used to measure the concentration of various sugars in aqueous solutions. Many industrial manufacturing and food processing systems require that the level of one or more sugars be carefully monitored at various stages to insure desired quality of final products. For example, the varying glucose concentrations during fermentation processes are important process control parameters, and their continuous monitor¬ ing can improve the yield and quality of the fermentation product. In addition, there are a large number of situations where the amount of sugar in finished food stuffs and other sugar containing products must be determined. On line, continu¬ ous measurement is important to reduce the risk of contamination, labor costs and delays associated with off-line measurements. To be able to perform on-line measurements, there is a need for sterilizable sensors with rapid response times and high sensitivity, yet which also require minimum maintenance and calibration.

One of the most important uses for sugar analysis techniques is in the medical field where monitoring of sugar levels in biological fluids is critical to proper diagnosis and treatment of diabetes and other diseases. With respect to medical applications, glucose is by far the most important sugar, and diabetes is the most common disease for which glucose determinations are routinely conducted. Diabetes is a disease of the metabolic system that affects more than 14 million people in the United States and over 100 million people worldwide. It is characterized by an elevated blood-glucose concentration which is caused by a lack of the hormone insulin. Sugars are the primary source of metabolic energy, and the inability to self-regulate the levels of sugar metabolized by the body leads to many other medical problems, including but not limited to blindness, heart disease and kidney failure. Treatment of diabetes involves monitoring of the patient's blood-glucose levels, with insulin injections being given when the glucose concentration rises above normal levels. A simple and accurate method for measuring blood-glucose concentrations is an essential cornerstone of any diabetes treatment protocol, since excessively high blood-glucose levels in diabetes patients can result in coma and even death. Frequent testing and insulin administration can significantly reduce long-term complications of diabetes. The vast majority of sensors which are used currently for glucose monitoring are simple devices based on enzymes such as glucose oxidase or glucose dehydrogenase. These enzyme-based sensors are simple to use and have relatively high sensing selectivity. They are widely used for one-time measurement of blood-glucose concentrations in vitro.

However, among the many drawbacks of enzyme-based sensors are that they are costly and have a short life time. The inherently unstable enzyme must be protected from extreme conditions during manufacturing and storage in order to preserve its catalytic activity. In addition, there have been a number of problems associated with the use of enzymes in implantable sensors used in systems for continuously monitoring blood-glucose levels in vivo, among them the fact that enzymes can elicit an immune response and are not stable to most sterilization methods.

The one-time or 'spot' measurement of blood-glucose concentration in vitro is also not optimal, as it requires collection of a blood sample, usually obtained by pricking the finger, which must be done at least several times each day. To avoid the need to subject the patient to this painful process for glucose monitoring, much effort has gone into identifying alternative sources for samples, such as subcutaneous tissue fluid or saliva instead of blood, and for identifying less-painful and more efficient ways to obtain these samples, such as by transdermal extraction or using very thin needles. The development of new, highly sensitive and miniaturizable glucose monitoring technology will make some of these alternative methods more feasible for use at home by patients. Continuous real time measurement of glucose concentrations is most desirable because it can be used for close monitoring and treatment. Studies indicate that medical outcomes are improved by more frequent, smaller insulin pulses. Continuous monitoring of glucose can also form part of a fully automatic insulin delivery system. A number of different sensor configurations have been proposed for use in either in vitro or in vivo monitoring systems. Continuous glucose monitoring systems are described in a number of publications, for example by E. Wilkins and M.G. Wilkins (J. Biomed. Eng. 1983, Vol. 5, October, pp. 309-315) and J. Pickup (TIBTECH, July, 1993, Vol. 1 1 , pp. 285-291).

While nearly all glucose sensors utilize an enzyme for glucose recognition, sensors described in the literature vary widely in their signal transduction mechanisms. Electrodes, optical methods and solid-state electronic devices such as field-effect transistors (FETs) are among the configurations that have been reported. The enzyme electrodes used in these enzyme sensor systems have been amperometric (mediated and non-mediated) and potentiometric. Although some of the above-described glucose sensors have shown promise, none have been found to be entirely satisfactory. There is a continuing need to develop robust sensor systems which can be used to accurately and simply measure the amount of a particular sugar which is present in aqueous solutions and other environments. The need is particularly apparent with respect to in vitro and in vivo glucose determinations which are critical in medical diagnosis and treatment of various metabolic disorders, including diabetes. There is also a strong need to develop robust sensor technology for real-time monitoring of sugars in manufacturing process environments.

SUMMARY OF THE INVENTION

In accordance with the present invention, sensors and systems are presented which are well-suited for use in measuring the presence of a variety of sugars and related compounds (target molecule) in aqueous solutions. The invention is based in part on the discovery that certain metal complexes will chelate sugars and other molecules in alkaline media and can either be used alone in solution, attached to a suitable support surface, or embedded in a polymer to provide a selective sensor which releases a proton when the target molecule binds to the metal complex. The release of protons from the sensor provides a simple, accurate and easily detected indirect measurement of target molecule concentra¬ tion in the solution.

As one feature of the present invention, a sensor is provided that includes metal complexes that bind a given target molecule in aqueous solution wherein binding of said metal complex to the target molecule results in the release of a proton. The metal complexes are preferably attached or embedded in a solid support to provide both anchoring of the complexes and positioning of the metal ions to increase selectivity of the target binding interactions. Attachment or incorporation of the metal complexes in a porous solid support can also reduce fouling of the sensor from other components in the biological sample (e.g. proteins) and provide selectivity in terms of molecular size. The sensor may be used to measure the target molecule concentration in a wide variety of aqueous solutions so long as the solutions are initially alkaline, are treated with a suitable base to make them alkaline prior to contact with the sensor, or the local environment of the metal complex is maintained in an alkaline state. The sensor is especially well-suited for use in measuring the concentration of glucose in blood or serum and other bodily fluids. However, the sensor, when appropriately configured, is suitable for measuring the concentration of a wide variety of other sugars and chemically-related compounds that also chelate the metal complexes and release protons. As another feature of the present invention, the metal complex contains a polymerizable functionality, which allows it to be copolymerized with other monomers and crosslinking agents to provide incorporation of the metal complex into a polymer support matrix. The copolymerization may further be conducted with the target molecule or another molecule (referred to as the template) bound to the polymerizable metal complex to improve the polymerization process and to provide imprinting of the resulting polymer matrix. If the template binds only to one metal complex, polymerization in the presence of the template and subsequent removal of the bound template molecules can result in a polymer with greater shape and size selectivity for that molecule or its structural analogs. Alternatively, if the template binds more than one metal complex, the resulting imprinted polymer matrix holds and positions the metal complexes in a spatial orientation which increases the selectivity of the sensor for the template molecule or its structural analogs. In both these cases, the template serves to direct the formation of a selective polymer. It has also been discovered that the template serves to improve the extent of polymerization, providing more rigid polymeric materials that can be obtained in the absence of the target molecule or other templates which bind the metal ions.

As a further feature of the present invention, a target molecule detection system is provided in which a signal transduction system is used to detect the protons which are released from the sensor metal complexes as a result of target molecule binding. Although any number of detection devices may be used to transduce the target molecule binding into an electrical or optical signal, it was found that detection systems based on changes in the solution pH caused by the proton release provided many advantages. Proton release can be detected simply and accurately by measuring changes in the pH of the solution when it is exposed to the sensor material. Altematively, the solution may be titrated with a known basic solution after exposure to the sensor to maintain a constant pH.

As another feature of the present invention, specific metal ion complexes are provided which are designed for interaction with sugars and related molecules.

Modification of the complexes allows them to be copolymerized with suitable monomers to form imprinted polymer sensors. The metal ion complexes are designed to 1 ) hold the metal ion tightly, 2) allow at least two coordination sites to be or become available for chelation to the target sugar or other molecule. Further design features are that the chelating ligand from which the metal complexes are formed can be chemically modified to 1 ) have a polymerizable functionality for copolymerization, or 2) have functional groups appropriate for covalent attachment to a solid surface, and 3) provide additional favorable interactions (electrostatic, hydrogen bonding, hydrophobic, etc.) with the target molecule. The metal ion for a particular metal ion complex is chosen such that the target molecule chelates the complex formed by the metal ion and chelating ligand rapidly and reversibly in the presence of the sample solution and that a proton is released upon chelation. The sensors in accordance with the present invention may be used in both the in vitro and in vivo environments. The sensors can be used in spot monitoring of target molecule concentrations, for example, as in monitoring glucose concen¬ tration in the blood, serum or subcutaneous tissue fluid sample of a diabetic patient. Alternatively, the sensor can be formulated into a continuous monitoring device for continuous measurement of target molecule levels ex vivo or in vivo, for example, as a subcutaneous implant or as part of an ex vivo continuous monitoring system for glucose using subcutaneous tissue fluid, serum or blood. The continuous sensors may be used as part of a feedback device for providing automatic dosing of insulin and other drugs which affect blood-sugar levels. The sensors in accordance with the present invention may also be used to monitor (on a spot or continuous basis) levels of sugars in manufacturing processes such as fermentations, purifications, and for product quality control, where it is desirable to know specific carbohydrate concentrations levels. Sensors in accordance with the present invention can be formulated for additional clinical and manufacturing applications, such as diagnosis of diseases or pathogens characterized by specific carbohydrates or other molecules that will chelate the metal complexes and produce proton signals. Additionally, such sensors can be formulated for monitoring important sugar or other chemical concentrations in other applications in clinical medicine, forensic science, drug testing, and manufacturing where it is important to rapidly obtain accurate measures of such concentrations.

The above described features and attendant advantages of the present invention will become better understood by reference to the following detailed description when taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS FIGS. 1a-h are diagrammatic representations showing exemplary ligands in their free state, in the form of metal complexes and as bound to glucose. The state of the complexes in alkaline aqueous is illustrated for Cu(TACN).

FIG. 2 is a diagrammatic representation of an exemplary synthesis of a preferred sugar binding polymer sensor wherein the sensor is a macroporous solid which is formed by co-polymerization of a polymerizable copper complex and crosslinking agent. FIG. 3 is a diagrammatic representation of an exemplary synthesis of a preferred sugar binding polymer sensor wherein the selectivity of the sensor is increased by imprinting the polymer with the sugar (or analog) that is to be detected.

FIG. 4 is a schematic representation of an exemplary sugar detection system in accordance with the present invention.

FIG. 5 is a schematic representation of a field effect transistor (FET) detector utilizing an imprinted matrix in accordance with the present invention.

FIG. 6 is a schematic representation of an exemplary implantable (subdermal) continuous glucose monitoring system utilizing fluorescence detection and a microporous sensor material incorporating pH-sensitive fluorescent probe molecules.

FIG. 7. is a graph showing change in pH resulting from sequential addition of c/s-diol (1 ,4-anhydroerythritol) to solutions of four different metal complexes at initial pH = 12.0. In accordance with the present invention, the concentration of an unknown sample of the c/s-diol can be determined by adding a known quantity to one of the metal complex solutions, measuring the resulting depression in pH and comparing that value to these calibration curves prepared using known quantities.

FIG. 8 is a graph showing change in pH resulting from sequential addition of Me- ?-D-Glc to solutions of four different metal complexes at initial pH = 12.0. In accordance with the present invention, the concentration of an unknown sample of Me- ?-D-Glc can be determined by adding a known quantity to one of the metal complex solutions, measuring the resulting depression in pH and comparing that value to these calibration curves prepared with known quantities.

FIG. 9 is a graph depicting the results of a pH static titration with c s-diol (1 ,4-anhydroerythritol). The graph shows the amount of NaOH required to maintain the pH of the Cu(TACN) metal complex solution at four initial pH values. In accordance with the present invention, the concentration of an unknown sample of c/s-diol can be determined by adding a known quantity to Cu(TACN), measuring the amount of NaOH required to maintain constant pH and comparison to these calibration curves prepared using known quantities.

FIG. 10 is a graph depicting the results of a pH static titration with Me- -D- Glc, showing amount of NaOH required to maintain the pH of the Cu(TACN) metal complex solution at four initial pH values. The concentration of an unknown sample of Me- ?-D-Glc can be determined by adding a known quantity to Cu(TACN), measuring the amount of NaOH required to maintain constant pH and comparison to these calibration curves prepared using known quantities.

FIG. 1 1 is a graph showing the results of a titration of an exemplary polymer sensor (Polymer I) with glucose at pH 1 1.50, 10.50 and 10.25. The graph shows total released proton concentration vs total glucose concentration.

FIG. 12 is a graph showing the results of static titration of exemplary polymer sensors (Polymers II, III and IV) at a pH of 10.25. The graph show total released proton concentration vs total glucose concentration.

FIG. 13 is a graph showing a comparison of protons released from exemplary polymer sensors II and Ile. The graph shows total released proton concentration vs total glucose concentration.

FIG. 14 is a graph showing a comparison of protons released from exemplary polymer sensors IV and IVe. The graph shows total released proton concentration vs total glucose concentration.

FIG. 15 is a graph showing a comparison of glucose and glucosamine binding to polymer IV at pH 10.25.

DETAILED DESCRIPTION OF THE INVENTION The sensors in accordance with the present invention may be used to detect and measure the presence of a wide variety of target molecules (sugars and related compounds) in solution. The sensors are suitable for detecting sugars such as glucose, mannose, sialic acid, aminosugars such as glucosamine, disac¬ charides, trisaccharides, oligosaccharides, sugar-amino acids, sugar-peptides and giycoproteins. Other target molecules can also be measured using this sensor. Exemplary target molecules include glycerol, dopamine, catechols, ascorbic acid, polyols, diols such as 1 ,4-anhydroerythritol, ethyleneglycol and 3-hydroxy-L- tyrosine. The general characteristics of or requirement for a suitable target molecule is that it must be able to chelate the metal complex and release a proton when used at an appropriate pH. Sensors which are capable of measuring the concentration of glucose in biological samples are preferred because of the importance of glucose in the diagnosis and treatment of diabetes and other disorders. The concentration range which is typically of interest in biological samples is 0-25 mM.

The sensors may be used to detect the presence of target molecules in a wide variety of different aqueous solutions. The only requirement is that the immediate environment of the metal complexes be at a pH that will provide ade¬ quate binding between the metal complexes in the sensor and the target molecule of interest and will release a proton(s). The pH of the environment can be tailored either by adjusting the sample pH or through the choice of polymer support, co- monomers or crosslinking agents for copolymerization, or by the addition of appropriate counterions. Alkaline solutions are those having a pH of above 7.0. The upper limit for solution pH is not particularly critical provided that the solution is not so alkaline that the target molecules or sensor components are adversely affected. In general, it is preferred that the immediate environment around the metal complex have a pH of between 9 and 13. The optimum values of pH will depend on the buffer capacity of the sample and the sensor's response at that pH, among other factors. Changes in the local pH resulting from binding to the target molecule are easier to measure at pH values where the sample has little buffering capacity. Similarly, the sensor will be more sensitive to target molecule concen- tration at pH values where the metal complex binds the target molecule most tightly. The optimal pH will also depend on the particular metal complex and target molecule.

The ability of sugars and other molecules with multiple hydroxyl groups to form chelate complexes with metal ions in aqueous solution are well known (general review by: Whitfield, D.M. et al., "Metal coordination to carbohydrates. Structures and Function," Coord. Chem. Reviews 122, 171 -225 (1993) and Angya, S.J. Complexes of Metal Cations with Carbohydrates in Solution, in "Advances in Carbohydrate Chemistry and Biochemistry" , Academic Press, Inc. 1989, pp. 1-4.). The complexation of Cu(ll) with various sugar σ-amino acids is described by (M. Angeles Diaz-Diez et al., Transition Met. Chem. 20, 402-405 (1995)). Sugar-σ-amino acid will also form complexes with Co(ll), Ni(ll), Zn(ll) and Cd(ll) (M. Angeles Diaz-Diez et al., J. Inorg. Biochem. 56, 243-247 (1994)). Klufers and Schuhmacher describe complex formation between sugar alcohols and Cu(ll) (Klufers, P. and J. Schuhmacher, Angew. Chem. Intl. Ed. Engl. 33, 1742-

1744 (1994)). Cu(ll) complexes with various amino sugars are described by (Kozlowski, H. et al., Carbohydrate Res. 197, 109-1 17 (1990)). Weaker complexes are formed with Ni(ll) and Co(lll).

Vanadium complexes of various sugars are described by (Sreedhara, A. et al., Carbohydrate Res. 264, 227-235 (1994)). Complex formation between D- xylo 5-hexulosonic acid and borate, molybdate and tungstate is described by (Caldeira, M.M. et al., Inorg. Chim. Acta 221 , 69-77 (1994)). The ability of Al(lll) to form complexes with glucose and glucoronic acid is reported by (Tonkovic, M. and Bilinski, H., Polyhedron 14, 1025-1030 (1995)). Fe(lll) complexes with eighteen different sugar-type molecules are reported by (Nagy, L. et al., Inorg.

Chim. Acta 124, 55-59 (1986). D-Glucuronic acid forms complexes with uranium (Tajmir-Riahi, H.A., Inorg. Chim. Acta 1 19, 227-232 (1986)), Ba(ll), Mg(ll) and Sr(ll) (Tajmir-Riahi, H.A., J. Inorg. Biochem., 24, 127-136 (1985)).

Glucose complexation to a variety of metals has been described: Cu(ll), Ba(lll) (Sharareh, S. and E. Wilkins, J. Environ. Sci. Health A26(7), 1021 -1032

(1991 ), Mg(ll), Ca(ll) (Tajmir-Riahi, H.-A., Carbohydrate Res. 183, 35-46 (1988), Fe(lll) (Geetha, K. et al.. Carbohydrate Res. 271 , 163-175 (1995)).

Furthermore, it is known that carbohydrates will bind reversibly to metal complexes immobilized on surfaces during ligand-exchange chromatographic separations. For example, carbohydrates form complexes with electrostatically immobilized rare earth (Tb(IID) and uranyl ions at alkaline pH (Stefannson, M., D. Wsterlund, "Ligand exchange chromatography of carbohydrates at alkaline pH: effects of mobile phase additives and temperature," J. Chromatogr. Sci. 32, 46-49 (1994)). Chromatographic separation of hexoses, pentoses and corres- ponding polyols has been reported on adsorbents containing Ca^{2 +} , Sr^{2 +} , Ba^{2 +} , Pb^{2 +} , Y^{3 +} , La^{3 +} and Pr^{3 +} (Caruel, H. et al.. Carbohydrate separation by ligand- exchange liquid chromatography, J. Chromatogr. 558, 89-104 (1991 )).

The sensor includes specific types of metal ion complexes that bind to sugars and other molecules in alkaline aqueous solution. Metal ion complexes which may be used in accordance with the present invention should: 1 ) hold the metal ion tightly; and 2) allow at least two coordination sites to be or become available for chelation to the target sugar or other molecule. Further useful features are that the chelating ligand from which the metal complexes are formed may be chemically modified to 1 ) have a polymerizable functionality for copoly¬ merization, or 2) have functional groups appropriate for covalent attachment to a solid surface, and 3) provide additional favorable interactions (e.g. electrostatic, hydrogen bonding and hydrophobic) with the target molecule. The metal ion for a particular metal ion complex is chosen such that the target molecule chelates the complex formed by the metal ion and chelating ligand rapidly and reversibly in the presence of the sample solution and that a proton is released upon chelation.

Four exemplary metal complexes are shown in FIG. 1 a. On the left side of FIG. 1 a, the metal complexes are shown in an unbound state in alkaline aque- ous solution. On the right side of FIG. 1 a, the configurations of the metal com¬ plexes when they are bound to glucose are shown. Exemplary metal complexes include copper(ll)-triazacyclononane (Cu(TACN)), copper(ll)-1 ,4-dimethyl triaza- cyclononane (Cu(1 ,4-dimethyl-TACN)) (not shown in Fig. 1 a), copper{ll)-ethyl- enediamine (Cu(En)), copper(ll)-propylenediamine (Cu(ll)-PDN)), copper(ll)- iminodiacetate (Cu(IDA)), andcopper(ll)-diethylenetriamine (Cu(Dien)). Derivatives of the above compounds may also be used including the alkylated derivatives. Preferred exemplary metal complexes are those which have the formula

wherein M is copper or iron; X is H or a functional group selected from styrene, methacrylate, acrylate, vinyl ether, vinyl acetate, trialkoxysilane, dialkyl- chlorosilane, epoxy and alkylhydroxyl or alkylamine groups having from 1 to 3 carbon atoms. Y is H or a functional group selected from styrene, methacrylate, acrylate, vinyl ether, vinyl, vinyl acetate, trialkoxysilane, dialkylchlorosilane, epoxy and alkyl, alkylhydroxyl or alkylamine groups having from 1 to 3 carbon atoms. Z is H or a functional group selected from styrene, methacrylate, acrylate, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane, epoxy and alkyl, alkylhydroxyl or alkylamine groups having from 1 to 3 carbon atoms. Additional exemplary metal complexes are listed below and classified by their coordination number and the donor atoms. Polymerizable functional groups for these complexes are also selected from styrene, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane and epoxy.

Class I (Nitrogen-based bidentate ligands)

Ethylenediamine (En), which can form a five-membered ring with metal ions by chelation, shows relatively strong complexation with metal ions (M) such as Cu ⁺ , Cu^{2 +} , Co^{3 +} , Ni^{2 +} , Fe^{3 +} and Hg^{2 +} . The geometry of the complexes is either square planar, tetrahedral or octahedral. Even when there is no vacant coordination site left in the complex, the deproteinated sugar is able to replace one of ethylene ligands at the equatorial positions under strongly basic conditions. The coordinating amine can be either primary or secondary or even tertiary, as long as steric strain is minimized. Polymerizable functional groups can be attached either through nitrogen or carbon atoms. Other diamines including linear and macrocyclic forms can serve the same purpose, as long as they can chelate the metal ion to form stable metal complexes with the features listed above. Exemplary chemical formulas for this class of complexes are set forth in FIG. 1 b where R is the polymerizable functional group. The binding of the complexes to glucose is also shown wherein " «»»»« " represents glucose.

Class II (Nitrogen-based linear tridentate ligands)

Linear triamines bind metal ions more tightly than the bidentate open chain diamines. The ligands which are able to form either five-membered ring or six- membered rings are suitable. Facial and meridional coordination stereoisomers are commonly observed for the octahedronal complexes for most metal complexes. The negatively-charged oxygens from sugar molecules are able to displace one of the ligands to a bind metal ion at its equatorial positions. Again, polymerizable functional groups can be incorporated at the nitrogen or carbon atoms. In the following example, only derivatization on the nitrogen is shown.

Exemplary metal complexes belonging to this class are shown in FIG. 1 c. The chemical formula for the ligand alone is shown along with exemplary formulas for the ligand bound to a metal ion to form the metal complex. An exemplary binding of the metal complex to glucose is also shown. Derivatization on the nitrogen atom only is shown. As mentioned above, derivatization on the carbon atom is also possible. In FIG. 1 c, M includes the same metal ions as the Class I complexes and R', R" is H, CH₃ or another other functional group that confers greater stability to the complex. For a polymerizable complex, one or more of the R groups is selected from styrene, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane or epoxy.

Class III (Nitrogen-based linear tetradentate ligands)

The structure of exemplary ligand from this class is shown in FIG. 1 d. The ligand structure is shown as well as the structure of the ligand after binding to a metal ion (M) and subsequent binding to glucose ( w ). The four nitrogen atoms can be either primary or secondary or tertiary amines. Ethylene units or propyl units between coordinating nitrogens are suitable. R, R', R", R" is H, CH₃ or another other functional group that confers greater stability to the complex. For a polymerizable complex, one or more of the R groups is selected from styrene, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane or epoxy.

Class IV (Nitrogen-based tridentate macrocycles)

Ligands used in the metal complexes can be improved by making them more rigid, with the coordinating groups correctly preoriented for coordination to the metal ions. Macrocyclic rings provide this advantageous feature. The tridentate macrocyclic ligands display strong complexation with most transition metal ions, chelating in a facial arrangement with two equatorial sites open for sugar binding. As before, structural variations on the ring can be made through the carbon or nitrogen atoms. A preferred ligand for the metal complex is triazacyclononane. This ligand tightly complexes metal ions such as copper(ll), leaving coordination sites appropriately positioned for sugar chelation. The binding of this type of ligand to a metal ion is shown in FIG. 1 a and in more detail in FIG. 1 e. The stability of triazacyclononane under alkaline conditions can be improved by alkylating the nitrogens. A preferred metal ion complex is therefore a polymerizable derivative of di-alkylated triazacyclononane. The triazacyclononane can be methylated, for example, at the 1 and 4 nitrogens (synthesis shown in Example 2), while a polymerizable group can be added at the remaining unalkylated nitrogen (using the procedures of Example 3).

Referring to FIG. 1e, M is a metal ion as described above and R, R', R" is H, CH₃ or another other functional group that confers greater stability to the complex. For a polymerizable complex, one or more of the R groups is selected from styrene, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxy- silane, dialkylchlorosilane or epoxy.

Class V (Nitrogen-based tridentate with pendant arm(s) able to form additional interactions with sugar molecules)

This class of ligands is very similar to the ligands in Class III. The only difference is that the functional groups with hydrogen bonding capability, such as hydroxyl, phosphinic, amide and carboxylic acid groups, are introduced to the nitrogen atoms. When the sugar molecule binds to the metal complex, the close proximity of hydrogen donors (acceptors) on the pendant arms can provide additional hydrogen bonding interactions besides metal-hydroxyl interactions. The chemical structure of this type of ligand and the resulting metal complex as bound to glucose is shown is FIG. 1 f.

In FIG. 1 f, one of the R groups can be hydroxyl, phosphinic, amide or carboxylic acid group. Alternatively, aromatic moieties or alkyl chains are incorporated on the ligands to provide hydrophobic interactions with sugar molecules. Other favorable interactions (e.g. electrostatic, etc.) can be envisioned, based on the target molecule to be recognized. For a polymerizable complex, one or more of the R groups is selected from styrene, methacrylate, acrylate, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane or epoxy. Class VI (Nitrogen-based tetradentate macrocyclic ring)

The macrocyclic ring is bigger than the tridentate macrocyclic rings. Most metals can sit in the plane consisting of four nitrogens atoms, and the conforma¬ tion of these macrocyclic rings is important for the binding geometry. If four nitrogens adopt square planar geometry around the metal ion, this metal complex cannot interact with a sugar molecule effectively. If two nitrogen atoms occupy two equatorial sites and the other two atoms coordinate to metal from axial positions and leave two equatorial coordination sites available, the resulting metal complex can bind sugar molecules more efficiently. The chemical structure of this class of ligand is shown in FIG. 1 g. Noncoordinating functional groups can be introduced on the rings to provide additional interactions similar to class V discussed earlier.

Class VII (Tridentate ring ligands) The tridentate ligands consisting of pyridine or pyrazole or imidazole rings are also suitable for forming metal complexes which may be used in detector systems in accordance with the present invention and are illustrated in FIG 1 h. The coordination geometry is similar to the saturated nitrogen donors set forth above. Other functional groups which can provide additional interactions with the target molecules can be introduced, as described previously. In FIG. 1h,

" ^{«ΛΛ \} " is a polymerizable functional group which can be the same as the other classes. B is boron.

Complexes which utilize copper as the metal ion are preferred. However, other metal ions which may be used in certain metal complexes include lead, vanadium, iron, mercury, nickel, cobalt, aluminum, uranium, borate, calcium, barium, Y^{3 +}, La^{3 +} .

Referring to FIGS. 1 a-h, upon binding of glucose to the metal complex, a proton is released into solution. The release of protons provides a direct indication of the amount of sugar which is available to bind to the metal complexes. The released protons may be measured directly using proton detec¬ tion equipment. It is preferred that the released protons be measured by monitoring their effect on solution pH. As is well known, release of protons into solution results in lowering of the solution pH. Accordingly, measurement of pH provides a relatively simple and straightforward method for qualitatively observing the binding reaction between sugar or other target molecule and the metal complex. This can be done, for example, using a pH electrode, field effect transistor (FET), light-addressable potentiometric sensor (LAPS), or other device which provides an electrical signal or, alternatively, using an chemical probe whose optical properties (fluorescence, absorption, etc.) are sensitive to solution pH. The probe then provides an optical signal which can be read to determine the sugar concentration.

When relatively large concentrations of sugar are being measured, care must be taken to ensure that the pH does not drop significantly. Significant drops in pH may cause the binding constant between the metal complex and sugar to decrease beyond measurement. A smaller sample size may be appropriate for very high glucose concentrations. Alternatively, static pH titration techniques can be employed, but they introduce additional complexity. Static titration involves addition of a suitable base, such as sodium hydroxide, to the solution to maintain a constant pH level. The amount of base which must be added to the solution to maintain a constant pH provides a direct indication of the amount of protons which are released due to sugar binding. The advantage of static titration is that the pH level remains the same so that the apparent binding constant between the target molecule and metal complex will also remain unchanged. It is preferred that the metal complexes of the present invention be anchored to a support surface. However, the present invention does cover the use of metal complexes which are in solution and not anchored to a support. The free metal complexes may be used as a liquid sensor to detect the presence of sugar in those situations where solid sensors are either undesirable or not practical. As will be discussed in detail below, attachment of the metal complex to ligand support structures is preferred since the support can provide the sensor with additional selectivity for the target molecule and can be used to connect the metal complexes with the appropriate proton detection system (e.g. field effect transistor (FET), light addressable potentiometric sensor (LAPS), pH-sensitive chromophore with optical wave guide).

It is preferred that the metal complex be attached to a ligand support structure. Any number of methods may be used to attach the metal complexes, and any number of different support materials and physical forms may be utilized. Exemplary ligand support materials include silicon, glass, quartz, ceramics, organic or inorganic polymers, and zeolites and other inorganic materials. The ligand support structure can be almost any solid form which provides a surface to which the metal complex may be attached. Exemplary forms include beads, porous polymer beads, particles or membranes, plates, threads, fibers and solid-state electronic devices such as FETs or LAPS devices and the like. The metal complexes may be attached to these materials and forms covalently or noncova- lently, using methods well known to those practiced in the art.

In a preferred embodiment, the metal complexes include a polymerizable moiety which allows the complex to be copolymerized with monomers and crosslinking agents to form porous polymeric materials. Polymerizable metal complexes are preferred since they can be incorporated via co-polymerization directly into the support structure. The polymer can be formulated into appropriate forms or configurations (membrane, beads, etc.) using methods well known to those practiced in the art. In addition, appropriate pH-sensitive chromophores or fluorophores can be incorporated during polymerization for optical detection of target molecule binding. The resulting polymer, whether it be in the form of a powder, micro beads or a larger structure, can be used directly as a sensor in conjunction with a pH monitoring system. Exemplary functional groups which are attached to the metal complex to form polymerizable metal complexes are set forth in detail above.

Exemplary monomers and cross-linkers which may be co-polymerized with the polymerizable metal complex include styrene, methyl methacrylate, 2- hydroxyethyl methacrylate, 2-hydroxyethyl acrylate, methyl acrylate, acrylamide, vinyl ether, vinyl acetate, divinylbenzene, ethylene glycol dimethacrylate, ethylene glycol diacrylate, pentaerythritol dimethacrylate, pentaerythritol diacrylate, N,N'- methylenebisacrylamide, N,N'-ethylenebisacrylamide, N,N'-(1 ,2-dihydroxyethy- lene)bis-acrylamide, trimethylolpropane trimethacrylate, etc. The choice of co¬ monomer and cross-linker will be dictated by the chemical (hydrophilicity, local pH, chemical stability, degree of crosslinking, ability to graft to other surfaces, interactions with other molecules, etc.) and physical (porosity, morphology, mechanical stability, etc.) properties desired for the polymeric sensor material. An exemplary synthesis is shown in FIG. 2 for forming a sugar sensing polymer by co-polymerization of a polymerizable copper(ll) complex with a cross¬ linking agent to form a macroporous polymer solid. The cross-linker and polymerizable metal complex are initially polymerized (step 1 ) to form the macroporous polymer sensor wherein the metal complexes are exposed for interaction with glucose molecules in solution. When the polymer sensor is exposed to glucose in the sample solution (step 2), the glucose binds to the metal complex which is in turn bound by polymerization to the polymer support structure which is represented by the shaded region in FIG. 2.

The procedures and conditions which are use to copolymerize the polymerizable metal complex and cross-linking agent are conventional. The relative amounts of each monomer can be varied to achieve desired concentra- tions of metal complexes in the polymer support structure. Typically, the amount of chelating monomer will be on the order of 5-10% weight percent of the cross¬ linking monomer. The solvent, temperature and means of polymerization (e.g. free radical initiation, 7-radiation) can be varied in order to obtain polymeric materials of optimal physical or chemical features, for example, porosity, stability, and hydrophilicity. The solvent will also be chosen based on its ability to solubilize all the various components of the reaction mixture. In addition, pH- sensitive indicators such as fluorescent probes or other optical probes can be incorporated into these polymers for detection of glucose binding by optical methods. Polymerizations are generally conducted in bulk solution of a polymerizable metal complex, a template molecule (if used), and a cross-linker by the free radical method. Similar methodology can be applied to surface grafting and particle coating with the polymer, as described in "Surface Grafting of Functional Polymers to Macroporous PolyCTrimethylolpropane Trimethacrylate," P. K. Dhal, S. Vidyasankar and F. H. Arnold, Chemistry of Materials 7, 154-162 (1995) and

"Molecularly-lmprinted Polymers on Silica: Selective Supports for High Performance Ligand-Exchange Chromatography," S.D. Plunkett and F.H. Arnold, J. Chromatogr. A 708, 19-29 (1995)).

For bulk polymerization, usually, about 5-10 weight percent of the polymerizable metal complex, 95-90 weight percent cross-linker, and 1 % of a free radical initiator such as azo-bis(isobutyronitrile) (AIBN) are dissolved in an aqueous/organic mixed solvent which serves as a porogen to create a porous structure. The solution is heated at 60 - 70° C for 24 hrs. The resulting cross- linked polymer is cut into pieces or ground into a powder and washed thoroughly with the same solvent used for the polymerization.

Polymerizations can be carried out by a sol-gel process when an alkoxy- silane type of polymerizable metal complex is used. In this case, the alkoxysilane metal chelating monomer is mixed with tetramethoxysilane or tetraethoxysilane in aqueous solution. The sol-gel condensation can be conducted in acidic or basic conditions using procedures well known to those practiced in the art.

In a further preferred embodiment, the selectivity of the resulting metal complexing sensor polymer is enhanced by utilizing a polymerizable metal complex which is initially bound to the target molecule (or an analog of the target molecule) to form an imprinted polymer matrix. The target molecule (or its analog) serves as a template to assist in the formation of selective binding cavities in the polymer. Imprinting of polymers using metal complexes is a well-known technique which is described in detail in United States Patent No. 5,310,648. The imprinting process is shown schematically in FIGS. 3a and b. Referring to

FIG. 3a, initially, the polymerizable copper metal complex is allowed to bind a sugar (methyl-^-D-glucopyranoside) to form an imprinting polymerizable monomer/ template complex which is composed of one metal complex attached to the glucose analog. This imprinting monomer/template complex is then co-polγmer- ized with a suitable crosslinking agent to form a porous polymer structure which is schematically shown at 10 in FIG. 3a. The polymer 10 includes the sugar analog and metal complex which is fixed in a cavity formed by the surrounding polymer support structure. The polymer support structure is shown as the shaded region in FIG. 3a. In the final step of sensor formation, the template molecule, i.e. the glucose analog, is removed. Removal of the template molecule leaves a macroporous polymer with complementary molecular cavities which include metal complexes that have metal ions which are positioned for specific binding with the template sugar.

Referring to FIG. 3b, the polymerizable copper metal complex is allowed to bind a sugar (glucose) to form an imprinting polymerizable monomer/template complex which is composed this time of two metal complexes attached to the glucose. This imprinting complex is then co-polymerized with a suitable crosslinking agent to form the polymer structure which is schematically shown at 1 1 in FIG. 3b. The polymer 1 1 includes the sugar and the polymerizable metal complexes which are fixed in a three-dimensional spatial distribution within the surrounding polymer support structure. The polymer support structure is shown as the shaded region in FIG. 3b. As before, the glucose template is removed in the final step of sensor formation. Removal of the template molecule leaves a macroporous polymer (13) with complementary molecular cavities which include metal complexes which are positioned for specific binding with the template sugar or its analogs.

The imprinted polymers (12 and 13) can be prepared in a wide variety of forms ranging from powders to beads to macro structures such as plates, rods, membranes or coatings on other materials. As before, pH-sensitive indicators such as fluorescent probes or other optical probes can be incorporated into these polymers for detection of glucose binding by optical methods. The same polymerizable metal complexes and cross-linking monomers described previously may be used. As examples of other imprinting polymerizations, the same polymerization conditions described above can be used, except a sugar template molecule such as methyl-σ-D-glucopyranoside, σ-D-glucose, a disaccharide or oligosaccharide, is first equilibrated with the polymerizable metal complex in aqueous solution at pH > 9 to form the polymerizable template complex. Alternatively, other non-sugar template molecules, such as a diol or dopamine, can be used.

The morphology and selectivity of the polymer for binding the target molecule may be improved by altering the solvent, polymerization temperature, choice of crosslinking agent, as described by Sellergren, B., Shea, K.J., "Influence of polymer morphology on the ability of imprinted network polymers to resolve enantiomers," . Chromatogr. A 1993, 635:31 -49). In particular, photoinitiation at low temperature should promote high selectivity and strong binding by materials that rely on temperature sensitive interactions for recognition.

An exemplary detection system for measuring the presence of sugar in an alkaline aqueous solution is shown in FIG. 4. The solution which is being measured for sugar content is shown at 20. The solution is introduced into the testing vessel 22 via sample introduction line 24. The sample may be blood, urine or other biological fluid. In non-medical applications, the sample solution introduced through inlet 24 can be any aqueous solution such as a fermentation broth or process stream. As mentioned previously, an important aspect of the present invention is that either the test solution 20 be alkaline or that the sensor material 32 be able to provide a local alkaline environment. If necessary, an inlet 26 is provided for allowing input of either acid or basic solutions into the sample solution 24 in order to pre-adjust it to desired pH levels. PH levels of 9 to 13 are preferred. The pH in the test solution 20 is monitored by way of pH measuring probe 28 which is connected to pH meter 30. The probe (field effect transistor, electrode) and metering equipment utilized to measure and monitor pH are well known. A sugar sensor in accordance with the present invention is shown at 32. The sugar sensor material can be in the form of a plate, block, beads or other macro structure. A preferred sensor type is a macroporous polymer which has been formed by the polymer imprinting process shown in FIGS. 3a or b. The pH of the test solution 20 is continuously monitored while it is present in the test cell 22.

One way to measure the glucose concentration is to monitor the drop in pH which occurs as the test solution 20 is allowed to interact with sensor 32. The drop in pH provides an accurate measurement of sugar concentration in the solution when compared to a calibration curve, provided the levels are below saturation in the pH response. In a preferred system, a titrator shown schematically at 34 is provided which adds known amounts of basic solution to the test vessel 22 in order to maintain the reaction solution 20 a constant pH.

This type of static-titration is well known wherein the amount of known base which must be added in order to maintain a constant pH is directly related to the sugar concentration. Feedback loop 36 is provided between the pH meter 30 and titrator 34 to provide automated operation of the system by providing constant pH level input into the titrator to allow control of the amount of basic solution being added.

A similar device can be used for continuous monitoring of sugar levels in a continuous sample stream 24. In this case, an effluent stream 38 removes sample continuously at a rate equal to streams 24 plus 26, such that a constant sample level is maintained in vessel 22. As before, the pH, or amount of basic solution required to maintain constant pH, can be monitored continuously in order to obtain a measure of the glucose concentration in the inflowing stream 24.

Another exemplary sensor device utilizing a field effect transistor (FET) is shown generally at 50 in FIG. 5. Chemical sensors based on pH detection by FETs are well known (Saito, A. et al., "An ISFET glucose sensor with a silicone rubber membrane for undiluted serum monitoring," Sensors and Actuators, B 20, 125-129 (1994); Reinhoudt, D. and E. J. R. Sudholter, "The transduction of Host- Guest interactions into electronic signals by molecular systems," Advanced Materials, 2,23-32 (1990)). As shown in FIG. 5, the imprinted polymer membrane 52 is attached to the gate oxide surface 54 of the semiconductor chip 56. The semiconductor chip 56 is preferably p-type silicon. The FET 50 also includes an encapsulant 58 and n- type source/drain 60 as is well-known in the art. An appropriate reference device can be made without the imprinted polymer membrane.

Alternatively, the protons released upon glucose binding to the sensor material could be measured using a light addressable potentiometric sensor (LAPS), as described by McConnell, H. M. et al., "The cytosensor microphysio¬ meter: biological applications of silicon technology," Science, 257, 1906-1912 (1992). The sensor material would be applied such that it would be in diffusive contact with the pH-sensitive surface of the LAPS chip.

The simplified schematic systems set forth in FIG. 4 and FIG. 5 will be understood by those of ordinary skill in the art as being merely representative of the type of detection system for which the metal complex sensors are well suited. A wide range of devices may be prepared, and the type of device will depend on the conditions of use (e.g. spot monitoring, continuous monitoring, implantable sensor, process monitoring, whether used in a hospital or at home, disposable, etc.) Another type of device appropriate for spot monitoring of glucose concentrations in clinical or process samples would be a simple colorimetric stick device or paper, in which the protons released by glucose binding would result in a color change of a pH-sensitive dye incorporated into the sensor material. When contacted with the biological sample, an estimate of the glucose concentration can be obtained from the resulting color of the indicator.

Additional exemplary systems include micro detection devices which can be implanted into a patient to provide continual in vivo glucose monitoring. An example of such a system is shown in FIG. 6 in which a biocompatible sugar sensor 70 is implanted under the skin 72. The sugar sensor material 74 is made such that the local environment of the metal complexes embedded in the material is alkaline. Variations in the local glucose concentration in the tissue will result in the release or uptake of protons in the immediate vicinity of the sugar sensor device. The material can be made so that it provides an optical signal (e.g. by incorporating pH-sensitive fluorophores 76 inside the sensor material) that varies with pH. The sensing material (i.e. metal complex) is preferably embedded in a biocompatible membrane 78. This optical signal can be detected through the skin by interrogation with red light (which readily penetrates skin) and using an appropriate detector to measure either the steady state fluorescence or fluorescence lifetimes, as has been described previously (Bambot, S.B. et al. "Potential applications of lifetime-based phase-modulation fluorimetry in bioprocess and clinical monitoring," TIBTECH 13, 106-1 15 (1995)). The light source is shown in FIG. 6 as laser 80 and the detector is shown at 82. With the use of long-lived fluorophores covalently embedded in the sensor material, subdermal sensing can be carried out with simple and inexpensive instrumentation. Alternatively, a device can be used ex vivo, connected to a continuous method for drawing a glucose sample from the patient, for example by microdialysis of subcutaneous tissue fluid (Meyerhoff, C. et al. 'Use of the microdialysis technique in the monitoring of subcutaneous tissue glucose concentration, Int'l J. of Art. Organs 16, 268-275 (1993)), by using a needle to continuously draw subcutaneous tissue fluid, or by transdermal extraction

(Tamada et al.. Measurement of glucose in diabetic subjects using noninvasive transdermal extraction. Nature Medicine 1 , 1 1 , 1 198-1201 (1995)).

These continuous monitors may be connected to insulin delivery devices by appropriate feed back systems to provide automatic insulin delivery to maintain desired blood glucose levels.

Examples of practice are as follows:

EXAMPLE 1 Use Of Soluble Metal Complexes In Measuring The Concentration

Of Sugars And Other Molecules In Aqueous Solution.

This example shows the use of several metal complexes in accordance with the present invention in measuring the concentration of sugars and other related molecules in aqueous solution. This example demonstrates that a c/s-diol (1 ,4- anhydroerythritol) and a sugar molecule (methyl- ?-D-glucopyranoside, Me- ?-D-Glc) will chelate to several complexes of the general formula CudDL (where L = propylenediamine (PDN), iminodiacetate (IDA), diethylenetriamine (DIEN), or triazacyclononane (TACN)) in alkaline aqueous solution. Chelation of the target molecule to the metal complex to form a ternary complex results in the release of a proton, which can be used to measure the concentration of the bound molecule either by the change in solution pH or by titration and comparison to a calibration curve. The corresponding trans-d\o\ does not chelate and therefore does not result in a change in solution pH, demonstrating the selectivity of these complexes. This example also demonstrates how a metal complex is evaluated for its suitability to recognize a particular target molecule in the current invention.

Measurement by pH Depression. 0.10 M solutions of target molecule (c s-diol or Me-^-D-Glc) were prepared at pH = 1 1.0 or 12.0. 10.0 mM solutions of CudDL (where L is either PDN, DIEN, IDA, or TACN) at pH = 1 1.0 or 12.0 were also prepared. The target molecule solution was titrated slowly into 5.0 mL of the CuL solution, and the pH was monitored until saturation was nearly reached (as indicated by a very small changes in pH). FIG. 7 shows the pH curves for titration of the various metal complexes CuL by the c/s-diol. The change in pH is most dramatic for Cu(TACN), and the relative order in pH change is in agreement with the order of binding affinities obtained by isothermal titration calorimetry. In contrast, the corresponding trans-d o\ does not result in a pH change, because its diol geometry differs from that of the cis compound, such that it is not capable of chelating the Cu(ll) complex. Thus Cu(TACN) is specific for the c s-diol. FIG. 8 shows the pH titration of the various metal complexes CuL using Me- ?-D-Glc.

Again, the pH change is most dramatic for the Cu(TACN), indicating that Cu(TACN) is the best of the four Cu(ll) complexes for chelating sugars or the related c s-diol compounds.

The concentration of an unknown sample of c/s-diol or Me-yff-D-Glc can be determined by adding a known quantity to one of the metal complex solutions, measuring the resulting depression in pH and comparing that value to these calibration curves prepared using known quantities. Measurement by Solution pH Titration. A solution of Cu(TACN) with concentra¬ tion around 10.0 mM was prepared without adjusting the pH. Since TACN is very hygroscopic and weighing errors are unavoidable, the concentration of the solution was calibrated to be 8.9 mM by its UV/vis absorption at 656 nm M_max = 656 nm, e = 47 M^"1 cm^"1). Target molecule solutions (c/s-diol and Me-^-D-

Glc) of concentration 0.10 M and pH of 9.0, 10.0, 1 1.0, or 12.0 were prepared. 5.00 mL of the Cu(TACN) solution was introduced into the titration vessel of a Brinkman pH titrator. The vessel was sealed, purged thoroughly with N₂, and equilibrated to 25 °C by a constant temperature water bath. The pH of the Cu(TACN) solution was adjusted to a desired value (9.0, 10.0, 1 1.0, or 12.0) with the addition of 0.10 N sodium hydroxide solution. Then a target molecule solution of the same pH as the Cu(TACN) solution in the titration vessel was titrated into the Cu(TACN) solution. After each injection of target molecule solution (c s-diol or Me- ?-D-Glc), the pH of the solution decreased, and the automatic pH titrator automatically added 0.10 N sodium hydroxide to bring the solution pH back to the original value. The volumes of the solution injected and the volumes of the 0.10 N sodium hydroxide added to maintain constant pH were recorded.

The amount of sodium hydroxide added during the titration is theoretically equal to the amount of protons released during the sugar binding reaction. FIG. 9 shows the titration data for the Cu(TACN)/c/s-diol system at various values of pH. The apparent binding strength greatly depends on the pH of the solution. At pH 9, the binding is very weak and saturation conditions are approached only at higher target molecule concentrations. However, at pH 1 1 or 12, saturation occurs at lower concentrations of target molecule, corresponding to a molar ratio of target molecule to Cu(TACN) of approximately 2. Thus there is a trade-off between the range of easily measurable concentrations and sensitivity of the sensor. At lower pH, the concentration range is greatest, but the sensitivity in terms of protons released is lowest. FIG. 10 shows the pH static titration data for Cu(TACN)/Me- ?-D-Glc system at various values of pH. Again, the apparent binding strength increases at higher pH.

The concentration of an unknown sample of c/s-diol or Me- ?-D-Glc can be determined by adding a known quantity to one of the metal complex solutions, measuring the resulting depression in pH and comparing that value to these calibration curves prepared with known quantities.

EXAMPLE 2 Synthesis Of A Novel Soluble Metal Complexing Agent,

1 ,4-dimethyl-1 ,4,7-triazacyclononane. For Sugar Sensing

1 ,4,7-Triazacyclononane-N, N', N"-tritosylate (100g) was suspended in a mixture of 200 mL of glacial acetic acid (100%) and 300 mL of HBr (48%) and the reaction mixture was refluxed under vigorous stirring for 15-18 hours. After cooling, the reaction mixture was filtered. To the filtrate were added 1 L of ethanol and 1 L of diethylether. 1 ,4,7-Triazacyclononane-N-monotosylated-dihy- drobromide (ca 45.4 g) was crystallized and the crystals were washed with acetone and diethylether and dried in the air. For further purification, the product can be stirred in refluxing acetone for 30 minutes. 1 ,4,7-Triazacyclononane-N-monotosylated-dihydrobromide (42.0 g) was dissolved in minimum amount of H₂0 (ca. 50 mL) and to this solution was added NaOH carefully to pH 7. 75 mL of an aqueous solution of formaldehyde (37 % CH₂0) and 75 mL of concentrated formic acid (HCOOH) were added to the solution and the resultant mixture was brought to reflux for 15 hours. The reaction was cooled to room temperature and 30 mL of concentration. HCl added and the excess solvent was removed on a rotovapor to yield a solid. This solid was suspended in minimum of water, neutralized with NaOH and the resultant solution was stood for 12 hours at room temperature. 1 ,4-Dimethyl-1 ,4,7- triazacyclononane-7-tosylate was crystallized slowly as a white solid, which was collected by filtration (ca.17 g) and was dried over CaO in a desiccator. This intermediate was then suspended in 300 mL of HBr (48%) and brought to reflux with stirring for 48 hours. The solution was concentrated to dryness under vacuum. The residue was then washed with dry acetone and about 54 g of 1 ,4- dimethyl-1 ,4,7-triazacyclononane tribromide was isolated. This salt was dissolved in 200 mL of water and to this solution was added 20 g of NaOH and 500 mL of toluene. The mixture was then refluxed with dean stark apparatus to remove all the water. The resultant solution was then filtered and the filtrate was concentrated on a rotovap to yield a yellow oil which can be stored under Ar at 0°C indefinitely. The complex is made polymerizable and loaded with copper(ll) by the procedures outlined in Example 3.

EXAMPLE 3 Synthesis Of Polymerizable Metal Complexes For

Use In The Preparation Of Selective Sensors

This example shows the synthesis of polymerizable metal complexes containing triazacyclononane in accordance with the present invention. Three polymerizable metal-chelating ligands, Msty-TACN (1 -(4'-vinylbenzyl)-1 ,4,7- triazacyclononane), Dsty-TACN(1 ,4-bis(4'-vinylbenzyl)-1 ,4,7-triazacyclononane), and Tsty-TACN (1 ,4,7-tris(4'-vinylbenzyl)-1 ,4,7-triazacyclononane), were prepared and loaded with copper ions according to the following procedures. The resulting polymerizable metal complexes were used to prepare imprinted polymeric sensors, as described in further examples.

1 -(4'-vinylbenzyl)-1 ,4,7-triazacyclononane (Msty-TACN)

Triazacyclononane (2 g, 15.5 mmol), powdered NaOH (0.62 g, 15.5 mmol) and dry acetonitrile (35 mL) were placed in a 100 mL 2-neck round-bottomed flask charged with a reflux condenser. The reaction mixture was brought to reflux under argon for an hour. To this suspension was added dropwise a solution of chloromethyl styrene (1.84 g, 12.4 mmol) in 10 mL of acetonitrile over a period of 8 hours, during which the consumption of the starting material was closely monitored by thin layer chromatography on a neutral alumina oxide gel plate with 20% MeOH/CHCI₃ as eluent. White solid was then removed by filtration and washed with acetonitrile (20 mL x 2). The filtrate was concentrated on a rotovap to yield a yellowish oil which was taken up in CHCI₃ (100 mL) and washed with water (30 mL x 2) and brine (30 mL x 2). The organic layer was dried over Na₂S0₄, filtered and concentrated in vacuo to obtain a blondish oil, which was chromatographed on neutral alumina oxide gel with 10 % MeOH/CHCI₃ as eluent (R_f of the product, 0.25). The product obtained was a yellowish solid, which weighed 1.68 g (44%): mp 56-59°C; ¹HNMR (CDCI₃) δ 2.72 (t, 4H, J = 4.8 Hz), 2.876 (t, 4 H, J = 4.8 Hz), 3.085 (s, 4H), 3.769 (s, 2H), 5.239 (d, 1 H, J = 10.8 Hz), 5.741 (d, 1 H, J = 17.4 Hz), 6.202 (s, 2H), 6.692 (dd, 1 H, J = 10.8 Hz, 17.4 Hz), 7.259 (d, 2H, J = 7.8 Hz), 7.376 (d, 2H, J = 7.8 Hz); ¹³C{¹ H} NMR (CDCI₃, 75 MHz) δ 44.07, 45.07, 50.68, 60.76, 1 13.83, 126.30, 129.18, 136.30, 136.77, 137.75; HRMS (FAB ⁺ ) cacld for C_{1 5}H₂₃N₃ 245.1892, found 246.1970 (MH ⁺); IR (KBr): u 3453, 3050, 2907,1 633, 1 517, 1455, 1291 , 1 105, 904.

1 ,4-bis(4'-vinylbenzyl)-1 ,4,7-triazacyclononane (Dsty-TACN)

Dsty-TACN was obtained as another product from the same reaction and chromatography (R_f of the product, 0.54) as described above. Yellowish solid product was obtained, which weighed 1.51 g (27%): mp 162-1 64°C; ¹H NMR (CDCI₃) δ 2.521 (s, 4H), 2.913 (t, 4H, J = 5.7 Hz), 3.010 (s, 1 H), 3.104 (t, 4H,

J = 5.7 Hz), 3.734 (s, 4H), 5.319 (d, 2H, J = 10.8 Hz), 5.817 (d, 2H, J = 17.4 Hz), 6.764 (dd, 2H, J = 10.8 Hz, 17.4 Hz), 7.267 (d, 4H, J = 8 Hz), 7.432 (d, 4H, J = 8 Hz); ¹³C{¹H} NMR (CDCI₃, 75 MHz) (J 43.92, 47.50, 50.33, 60.66, 1 14.21 , 126.47, 129.35, 136.15, 136.66, 137.17 HRMS (FAB⁺) cacld for C₂₄H₃₁ N 362.2596, found 362.2591 ; IR (KBr) v 3472, 3450, 2920, 2741 , 1627,

1560, 1394, 1207, 923, 860, 736 cm^"1.

1 ,4,7-tris(4'-vinylbenzyl)-1 ,4,7-triazacyclononane (Tsty-TACN),

Tsty-TACN was obtained as the third product from the same reaction and chromatography (R_f of the product, 0.73) as described above. Yellowish solid product was obtained, which weighed 1 .33 g (18%): mp 71 -73°C; ¹ H NMR (CDCI₃) δ 2.963 (s, 12H), 3.739 (s, 6H), 5.353 (d, 3H, J = 10.8 Hz), 5.861 (d, 3H, J = 17.4 Hz), 6.828 (dd, 4H, J = 10.8 Hz, 17.4 Hz), 7.381 (d, 6H, J = 7.8 Hz), 7.475 (d, 6H, J = 7.8 Hz); ¹³C{¹ H} NMR (CDCI₃, 75 MHz) δ 55.36, 62.73, 1 13.17, 125.94, 129.19, 136.04, 136.66, 140.09; HRMS (FAB⁺) cacld for

C₃₃H₃₉N₃ 477.3144, found 477.3139; IR (KBr) υ 3049, 2904, 1627, 1508, 1451 , 1295, 909.7, 828 cm^"1.

[Cu(Msty-TACN)]^{2 +} S 0₄ ^2" To a solution of Msty-TACN (1.71 g, 6.98 mmol) in methanol (100 mL) was added CuS0 - 5H₂0 (1.7 g, 0.98 mmol) portionwise at room temperature. The solution turned turbid when about half amount of CuS0₄ 5H₂0 was added. After vigorous stirring for a few minutes, the solution turned clear sky blue. The solution was kept stirring for an additional hour after the remaining CuS0₄-5H₂0 was added. Solvent was then removed on a rotavap to yield a blue solid, which was recrystallized from methanol/acetonitrile to give 1.35 g (48%) of a blue crystalline solid: mp 177-181 °C; Anal. Calcd for C₁₅H₂₃N₃0₄SCu: C, 44.55; H, 5.74; N, 10.40; Cu, 15.57. Found: C, 44.90; H, 5.86; N, 10.13; Cu, 15.80; IR (KBr): υ 3260, 2924, 1627, 1458, 1097 cm^"1; UV-vis (in CH₃OH) e: 75 at 657 nm.

[Cu(Dsty-TACN)]²⁺S0₄ ²-

To a solution of Dsty-TACN (0.56 g, 1.55 mmol) in methanol (50 mL) was added CuS0₄ ^* 5H₂0 (0.387 g, 1 .55 mmol) as a solid portion wise at room temperature. The solution turned green first and then cloudy. Blue precipitate started forming after a few more minutes' vigorous stirring. The resulting mixture was kept stirring for an additional hour. The precipitate was collected by filtration, and was further purified by recrystallization with methanol/water to give 0.45 g (53%) of a blue crystalline solid: mp 276-280 (decomp.); Anal. Calcd for

C₂₅H₃₅N₃O_εSCu (M + CH₃OH): C, 54.33; H, 6.39; N, 7.61 ; Cu, 1 1 .40. Found: C, 53.99; H, 5.99; N, 7.78; Cu, 1 1.78; IR (KBr): υ 3500-3100, 2977, 1629, 1055, 830, 746 cm^'1-

[Cu(Tsty-TACN)]²⁺S0₄ ^2"

To a solution of Dsty-TACN (1.33 g, 2.79 mmol) in CH₃OH/CH₃CI (50 mL/30 mL) was slowly added a solution of CuS0₄" 5H₂0 (0.692 g, 2.79 mmol) in 20 mL of methanol at room temperature. The solution turned blue and cloudy.

The resulting mixture was kept stirring for an additional hour. The precipitate was collected by filtration, and was further purified by recrystallization with DMF/H₂0 to give 1.34 g (75%) of a blue crystalline solid: mp 280-284 (decomp.); Anal.

Calcd for C₃₃H₃₉N₃0₄SCu: C, 62.24; H, 6.18; N, 6.60; Cu, 9.89. Found: C,

60.16; H, 6.20; N, 6.33; Cu, 9.89; IR (KBr): υ 3000 (br), 1508, 1458, 12228,

1 143 858, 830 cm-¹-.

EXAMPLE 4 Preparation Of A Polymeric Sensor For Methyl-σ-D-glucopyranoside This example shows the preparation of a sensor in accordance with the present invention which includes a polymer support structure imprinted with a glucopyranoside. The example also demonstrates the ability of the sensor to rebind with the template molecule and produce protons upon rebinding. This example uses Cu(ll)-[1 ,4-bis(4'-vinylbenzyl)-1 ,4,7-triazacyclononane] (Dsty-TACN- Cu ⁺) as the polymerizable metal complex.

Imprinted polvmer preparation:

30 mg (0.058 mmol) of [Dsty-TACN-Cu^{2 +}]S0₄ ²\ 33.58 mg (0.165 mmol) of Me-σ-D-GIc were dissolved in a mixed solvent composed of 1.1 1 mL water and 2.22 mL methanol. The pH of the solution was adjusted to 1 1.50 with addition of 1 N sodium hydroxide. After equilibrating at room temperature for half an hour, 1.0 g (6.49 mmol) of N,N'-methylene bisacrylamide and 10 mg of 2,2'- azobisisobutyronitrile (AIBN) were added into the solution. After purging the solution with argon, the flask was sealed and heated at 60 °C. The solution was polymerized at 60°C for 7 hrs. After polymerization was complete, the polymer was ground into a fine powder using a mortar and washed three times with 50/50 water/methanol. The resin was equilibrated with 100 mM EDTA solution at 60°C for several hrs to strip off most of the Cu(ll) ion and the template sugar molecule. The polymer resin was then washed with water to remove extra EDTA, reloaded Cu(ll) with 20 mM CuS0₄ solution, washed again with water to remove free Cu(ll), and finally frozen and lyophilized.

Rebinding of methyl-σ-D-αlucopyranoside to the imprinted polvmer:

1 g of the imprinted polymer resin and 20 mL of water were placed into the pH titration vessel. The vessel was equilibrated to 60°C. The pH of the suspension was adjusted to 1 1.00 by addition of 0.1 N sodium hydroxide. A solution of methyl-σ-D-glucopyranoside with a concentration of 0.10 M and pH of 1 1.00 was prepared. The sugar solution was titrated into the polymer suspension sequentially. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system at 1 1.00. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the sugar solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The data from this static titration are set forth in Table 1. TABLE 1 Titration of sensor with methyl-σ-D-glucopyranoside (pH 11.00)

titration No.# ^Vαlucose<^m > NaθH< ^L>

0 0.00 0.000

1 0.50 0.027

2 1.00 0.073

3 1.50 0.100

4 2.00 0.137

5 2.50 0.144

6 3.00 0.144

7 3.50 0.144

8 4.00 0.144

As can be seen from Table 1 , the amount of sugar added to the solution is directly related to the amount of NaOH which must be added in order to maintain a constant pH. Accordingly, the sensor provided an indirect measure of the amount of sugar added to the solution.

EXAMPLE 5

Preparation Of A Sensor For Methyl-σ-D-glucopyranoside Which Includes A Non-Imprinted Ligand Support Polymer This example shows the preparation of a sensor which includes a polymer that is not imprinted with a template molecule. This example also demonstrates the sensor's ability to bind methyl-σ-D-glucopyranoside and produce protons.

Polvmer preparation: 30 mg (0.058 mmol) of [Dsty-TACN-Cu^{2 +}]S0₄ ^2" was dissolved in a mixed solvent composed of 1.1 1 mL water and 2.22 mL methanol. The pH of the solu¬ tion was adjusted to 1 1.50 with addition of 1 N sodium hydroxide. After equi¬ librating at room temperature for half an hour, 1 .0 g (6.49 mmol) of N,N'- methylene bisacrylamide and 10 mg of 2,2'-azobisisobutyronitrile (AIBN) were added into the solution. After purging the solution with argon, the flask was sealed and heated at 60°C. The solution was polymerized at 60°C for 7 hours. After polymerization was complete, the polymer was ground into a fine powder using a mortar and washed three times with 50/50 water/methanol. The resin was equilibrated with 100 mM EDTA solution at 60°C for several hours to strip off most Cu(ll) ion. The polymer resin was then washed with water to remove extra EDTA, reloaded Cu(ll) with 20 mM CuS0₄ solution, washed again with water to remove free Cu(ll), and finally frozen and lyophilized.

Binding of methyl-σ-D-αlucopyranoside to the polvmer:

1 g of the polymer resin and 20 mL of water were placed into the pH titration vessel. The vessel was equilibrated to 60°C. The pH of the suspension was adjusted to 1 1.00 by addition of 0.1 N sodium hydroxide. A solution of methyl-σ-D-glucopyranoside with a concentration of 0.10 M and pH of 1 1.00 was prepared. The sugar solution was titrated into the polymer suspension sequentially. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system at 1 1.00. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the sugar solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The data from this static titration are set forth in Table 2.

TABLE 2

Titration of sensor with methyl-σ-D- glucopyranoside at pH 11.00 and 60°C

titration No.# ^Vαlucosβ «^m ) VNaOH <^m >

0 0.00 0.000

1 0.50 0.031

2 1 1..0000 0.068

3 1.50 0.093

4 2.00 0.134

5 3.00 0.202

6 4.00 0.303

7 5 5..0000 0.321

8 6.00 0.321

As can be seen from Table 2, the amount of sugar added to the solution is directly related to the amount of NaOH which must be added in order to maintain a constant pH. Accordingly, as in Example 4, the sensor provided an indirect measure of the amount of sugar added to the solution. EXAMPLE 6

Preparation Of An Imprinted Polymer Sensor Which Includes Methyl- ?-D-glucopyranoside As The Template Molecule ^' 5 And Cu(ll)-[N-(4-vinyl)imino)diacetic acid]

(Sty-IDA-Cu^{2 +} As The Polymerizable Metal Complex

This example demonstrates the ability of the imprinted metal-complexing polymer to bind glucose and produce protons. This example also shows that the system can be configured such that the proton release is linearly proportional to

10 the glucose concentration. Finally, this example demonstrates the use of Styrene-

IDA-Cu^{2 +} as the polymerizable metal complex.

Imprinted polvmer preparation:

0.75 g (2.29 mmol) of Styrene-IDA-Cu^{2 +} (prepared according to the

15 procedures outlined in United States Patent No. 5,310,648) and 0.465 g (2.29 mmol) of Me- -D-Glc were dissolved in a mixed solvent composed of 16.67 mL water and 33.33 mL methanol. The pH of the solution was adjusted to 1 1.50 with addition of 1 N sodium hydroxide. After equilibrating at room temperature for half an hour, 14.25 g (92.4 mmol) of N,N'-methylenebisacrylamide and 150 0 mg of 2,2'-azobisisobutyronitrile (AIBN) were added into the solution. After purging the solution with argon, the flask was sealed and heated at 60 °C. The solution was polymerized at 55 °C for 21 hours and at 70 °C for additional 4 hours. After polymerization was complete, the polymer was ground into a fine powder using a mortar and washed three times with 50/50 water/methanol. The

25 resin was equilibrated with 100 mM EDTA solution at 60 °C for several hours to strip off most of the Cu(II) ion and the template sugar molecule. The polymer resin was then washed with water to remove extra EDTA, reloaded Cu(ll) with 20 mM CuS0 solution, washed again with water to remove free Cu(ll), and finally frozen and lyophilized. 0

Rebinding of glucose to the polvmer:

3.0 g of the imprinted polymer resin and 30 mL of pH 1 1 .50 NaOH solution were placed into the pH titration vessel. The vessel was equilibrated to 60°C. The pH of the suspension was adjusted to 1 1.52 by addition of 0.1 N sodium

35 hydroxide. A solution of D-( + )-glucose with a concentration of 0.50 M and pH of 1 1.52 was titrated into the polymer suspension. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system at 1 1.52. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the sugar solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The data for the static titration are tabulated in Table 3. As can be seen from Table 3, the amount of base required to maintain the pH of the solution is linearly proportional to the concentration of glucose in the solution.

TABLE 3 Titration of sensor with glucose at pH 11.52 and 60°C

αlucosβ' ) ^Cαlucosβ(^mM) NaOH<mD

0.100 0.133 1.56

0.200 0.188 3.12

0.300 0.231 4.69

0.400 0.262 6.24

0.500 0.301 7.81

0.600 0.336 9.37

0.700 0.373 10.9

0.800 0.412 12.4

0.900 0.442 14.1

1.00 0.475 15.6

1.10 0.508 17.2

1.20 0.543 18.7

EXAMPLE 7

Preparation Of An Imprinted Glucose Sensor Which Includes

Methyl- ?-D-glucopyranoside As The Template And

Msty-TACN-Cu²⁺ As The Polymerizable Metal Complex This example demonstrates the ability of the metal-complexing polymer to bind glucose and produce protons. This example also shows that the system can be configured such that the proton release is proportional to glucose concentra¬ tion. This example further demonstrates the use of Cu(ll)-[1-(4'-vinylbenzyU- 1 ,4,7-triazacyclononane] (Msty-TACN-Cu^{2 +} ) as the polymerizable metal complex. Imprinted polvmer preparation:

0.4045 g (1.0 mmol) of [Msty-TACN-Cu^{2 +} ]S0₄ ^2_ and 0.2032 g (1.0 mmol) of Me-yff-D-Glc were dissolved in a mixed solvent composed of 4.0 mL water and 8 mL methanol. The pH of the solution was adjusted to 1 1.70 with addition of 1 N sodium hydroxide. After equilibrating at room temperature for half an hour,

3.13 g (20 mmol) of N,N'-methylene bisacrylamide was added into the solution. After freeze-thawing for 4 cycles to degas the solution, the flask was sealed and heated at 65 °C. As all solid became completely dissolved, 35 mg AIBN dissolved in 0.5 mL degassed methanol was syringed into the solution. The solution was polymerized at 65°C overnight and at 70°C for 12 hours. After polymerization was complete, the polymer was ground into a fine powder using a mortar and washed three times with 50/50 water/methanol. The resin was loaded into a glass column and was washed with 100 mM EDTA solution continuously overnight to strip off all the Cu(ll) ion and the template sugar molecule. The polymer resin was then washed with water to remove extra EDTA, reloaded Cu(ll) with 20 mM CuS0₄ solution, washed again with water to remove free Cu(ll), and finally frozen and lyophilized.

Rebinding of glucose to the imprinted polvmer: 0.5 g of polymer resin was suspended in 5.0 mL pH 1 1.50 NaOH solution in the pH titration vessel, which was kept at 25 °C by a constant temperature water bath. The pH of the suspension was adjusted to 1 1.50 by addition of 6N sodium hydroxide solution. A solution of D-( + )-glucose with concentration of 0.208 M and pH of 1 1.50 was prepared. The glucose solution was titrated into the polymer suspension sequentially. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system at 1 1 .50. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the glucose solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The data for the static titration are given in Table 4. TABLE 4 Titration of sensor with glucose at pH 11.52.

^Valucose<^mL> ^Calucose<^mM> NaOH (mL)

0.020 0.743 0.028

0.040 1.49 0.056

0.060 2.23 0.077

0.080 2.97 0.096

0 0..110000 3 3..7711 0.114

0.120 4.46 0.128

0.140 5.20 0.141

0.160 5.94 0.158

0.180 6.69 0.170

0 0..220000 7 7..4433 0.180

0.220 8.17 0.189

0.240 8.91 0.202

0.260 9.66 0.210

0.280 10.4 0.217

0 0..330000 1 111..11 0.224

0.305 13.0 0.239

0.400 14.9 0.249

0.450 16.7 0.259

0.500 18.6 0.269

0 0..555500 2 200..44 0.275

0.600 22.3 0.281

As can be seen from Table 4, the proton release from the sensor (the volume of NaOH added to maintain constant pH) is proportional to the glucose concentration.

EXAMPLE 8

Preparation Of Polymers Using Methyl- ?-D-glucopyranoside As The Template And Msty-TACN-Cu²⁺ As The Polymerizable

Metal Complex And Glucose Rebinding To The Polymers

This example demonstrates that polymers containing various concentrations of functional monomer (Cu^{2 +} sites) can be prepared. It also shows that glucose rebinding to the materials depends on the solution pH. Polymers I to IV are rigid polymers prepared using four different ratios of functional monomer Msty-TACN-Cu^{2 +} (Cu(ll)-[1-(4'-vinylbenzyl)-1 ,4,7-triazacyclo- nonane]) to crosslinker MBA (N,N'-methylenebisacrylamide), as listed in Table 5.

The synthesis of each polymer is described below, as is their behavior in glucose rebinding studies. Polymer IV, with the highest Cu^{2 +} density, gives the most linear response over the widest range of solution glucose concentration.

Synthesis of polymers I-IV Four polymers of varying molar ratio Msty-TACN-Cu^{2 +} monomer to crosslinker MBA (ratios shown in Table 5) were prepared as described below. Following polymerization, the material was ground into a fine powder using a mortar and washed with 50/50 water/methanol. The powder was loaded into a glass column and washed extensively with 100 mM EDTA solution to strip off all the Cu(ll) ion and the template sugar molecules. The polymer resin was then washed with water to remove residual EDTA, reloaded with 20 mM CuS0₄ solution, washed again with water to remove free Cu(ll), frozen and lyophilized.

TABLE 5. Molar ratios of Msty-TACN-Cu^{2 +} to crosslinker MBA in Polymers I-IV.

Polymer I Polymer II Polymer III Polymer IV

Mole ratio of Msty-

TACN-CU^{2 +} to MBA 1 :20.32 1 :16.14 1 :13 1 :10

Polymer I. [Msty-TACN-Cu^{2 +} ]S0₄ ^2" (0.4045 g, 1 .0 mmol) and Me-^-D-Glc (0.2032 g, 1.0 mmol) were dissolved in a mixture of 4.0 mL pH 1 1.50 aqueous NaOH solution and 8 mL methanol. The pH of the solution was adjusted to 1 1.50 with addition of 1 N sodium hydroxide. After equilibrating at room temperature for half an hour, 3.13 g (20.30 mmol) of MBA was added into the solution. After freeze-thawing for 4 cycles to degas the solution, the flask was sealed and heated at 65 °C. As all solid became completely dissolved, 35 mg AIBN dissolved in 0.5 mL degassed methanol was syringed into the solution. The solution was polymerized at 65 °C overnight and at 70°C for 12 hrs.

Polymer II. [Msty-TACN-Cu^{2 +}]S0₄ ²- (0.101 1 g, 0.25 mmol) and Me-^-D-Glc (0.0508 g, 0.25 mmol) were dissolved in a mixture of 2.0 mL of methanol and 1 mL of pH 1 1.50 aqueous NaOH solution in a vial. The pH of the solution was further adjusted to 1 1.50 with addition of 1 N sodium hydroxide. After equilibrating at room temperature for half an hour, 0.621 g (4.028 mmol) of MBA and 9 mg of AIBN were added into the solution. After freeze-thawing for 4 cycles to degas the solution, the vial was sealed and heated at 65 °C. The solution was polymerized at 65°C overnight and at 70°C for 12 hours.

Polymer III. [Msty-TACN-Cu^{2 +}]S0₄ ^2' (0.101 1 g, 0.25 mmol) and Me-0-D-Glc (0.0508 g, 0.25 mmol) were dissolved in a mixture of 1.7 mL of methanol and 0.8 mL of pH 1 1.50 aqueous NaOH solution in a vial. The pH of the solution was further adjusted to 1 1.50 with addition of 1 N sodium hydroxide. After equilibrating at room temperature for half an hour, 0.501 g (3.25 mmol) of MBA and 9 mg of AIBN were added into the solution. After freeze-thawing for 4 cycles to degas the solution, the vial was sealed and heated at 65 °C. The solution was polymerized at 65°C ovemight and at 70°C for 36 hours.

Polymer IV. [Msty-TACN-Cu^{2 +} ]S0₄ ²- (0.101 1 g, 0.25 mmol) and Me-yff-D-Glc (0.0508 g, 0.25 mmol) were dissolved in a mixture of 1.5 mL of methanol and 0.7 mL of pH 1 1.50 aqueous NaOH solution in a vial. The pH of the solution was further adjusted to 1 1.50 with addition of 1 N sodium hydroxide. After equilibrating at room temperature for half an hour, 0.385 g (2.50 mmol) of MBA and 9 mg of AIBN were added. After freeze-thawing for 4 cycles to degas the solution, the vial was sealed and heated at 65 °C. The solution was polymerized at 65 °C overnight and at 70 °C for 36 hrs.

Glucose rebinding studies a) Glucose rebinding to polymer I at pH 11.50, 10.50 and 10.25 by pH static titration:

Polymer I (0.5 g) was suspended in 5 mL of pH 1 1.50, 10.50 or 10.25 NaOH solution in the pH titration vessel maintained at 25 °C by a constant temperature water bath. The pH of the suspension was adjusted as necessary by addition of 6N sodium hydroxide solution to maintain the starting pH. A solution of D-( + )-glucose with concentration of 0.208 M and pH of 1 1.50, 10.50 or 10.25 was titrated into the polymer suspension sequentially. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system constant. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the glucose solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The data in terms of the total concentration of released protons versus glucose concentration are plotted in FIG. 1 1 for examples performed at the three values of pH.

b) Glucose rebinding to polymer II at pH 10.25 by pH static titration:

Polymer II (0.100 g) was suspended in 0.980 mL of pH 10.25 NaOH solution in a pH titration vessel, which was kept at 25 °C by a constant temperature water bath. The pH of the suspension was adjusted to 10.25 by addition of 6N and 0.1 N sodium hydroxide solution and pH 10.25 NaOH solution (0.020 mL total). A solution of D-( + )-glucose with concentration of 0.500 M and pH of 10.25 was titrated into the polymer suspension sequentially. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system at 10.25. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the glucose solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The titration data are plotted in FIG. 12.

c) Glucose rebinding to polymer III at pH 10.25 by pH static titration:

Polymer III (0.100 g) was suspended in 0.980 mL of pH 10.25 NaOH solution in a pH titration vessel, which was kept at 25 °C by a constant temperature water bath. The pH of the suspension was adjusted to 10.25 by addition of 6N and 0.1 N sodium hydroxide solution and pH 10.24 NaOH solution (0.020 mL total). A solution of D-( + )-glucose with concentration of 0.500 M and pH of 10.25 was titrated into the polymer suspension sequentially. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system at 10.25. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the glucose solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The titration data are plotted in FIG. 12 d) Glucose rebinding to polymer IV at pH 10.25 by pH static titration:

Polymer IV (0.100 g) was suspended in 0.980 mL of pH 10.25 NaOH solution in a pH titration vessel, which was kept at 25 °C by a constant temperature water bath. The pH of the suspension was adjusted to 10.25 by addition of 6N and 0.1 N NaOH and pH 10.25 NaOH solution (0.020 mL total).

A solution of D-( + )-glucose with concentration of 0.500 M and pH of 10.25 was titrated into the polymer suspension sequentially. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system at 10.25. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the glucose solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The titration data are plotted in FIG. 12.

FIG. 12 compares the titration results for polymers II to IV at pH 10.25 in terms of protons released versus glucose concentration. The largest total signal per unit weight of polymer is provided by polymer IV, made with the largest amount of polymerizable metal complex. This polymer also gives the most linear response, with respect to protons released versus glucose concentration.

EXAMPLE 9 Preparation Of Polymers Using Ethylene Glycol As The

Template And Msty-TACN-Cu²⁺ As The Polymerizable Metal Complex And Glucose Rebinding To The Polymers

This example demonstrates that polymers prepared using ethylene glycol as a template do not release as many protons in the presence of glucose as the analogous polymers prepared using the sugar analog methyl-^-D-glucopyranoside as the template (Example 8). Thus while different molecules can serve as a template during synthesis of these selective sensor materials, structural analogs of the target molecule (here, glucose) are likely to yield the best materials.

Polymers Ile and IVe are rigid polymers prepared using the same polymerization conditions and ratios of functional monomer Msty-TACN-Cu^{2 +}

(Cu(ll)-[1-(4'-vinylbenzyl)-1 ,4,7-triazacyclononane]) to crosslinker MBA (N,N'- methylenebisacrylamide) as for polymers II and IV listed in Table 5 of Example 8. Thus the performance of these materials can be compared directly to that of polymers II and IV of Example 8. The synthesis and behavior of the polymers in glucose rebinding studies is presented. Synthesis of polymers He and IVe.

Polymers of molar ratio Msty-TACN-Cu^{2 +} monomer to crosslinker MBA were prepared as described in Example 8 for polymers II and IV. The only differences was that ethylene glycol was used as the template instead of methyl- 5 7-D-glucopyranoside. Polymer Ile was prepared using 0.101 1 g of Msty-TACN-

Cu^{2 +} , 21.0 mL (0.375 mmol )of ethylene glycol, 0.621 g MBA and 9 mg AIBN. Polymer IVe was prepared using 0.101 1 g Msty-TACN-Cu^{2 +} , 21.0 mL (0.375 mmol) of ethylene glycol, 0.3854 g MBA and 9 mg AIBN.

10 Glucose rebinding studies a) Glucose rebinding to polymer II at pH 10.25 by pH static titration:

Glucose rebinding was measured as described in Example 8. The titration data are plotted in FIGS. 13 and 14. Also plotted, for comparison, are the titration data for polymers II and IV, made using methyl- 7-D-glucopyranoside as

15 the template (Example 8).

EXAMPLE 10

Binding of other substrates to polymer prepared using methyl- ?-D-glucopyranoside as the template

20 This example shows the binding of several substrates to glucopyranoside- imprinted polymer IV of Example 8 and the protons released upon binding. The substrates investigated are D-glucosamine, urea, lactic acid and the amino acid alanine. These are compounds commonly found in serum and other biological samples that might be monitored for glucose by the current invention. Thus this

25 example illustrates a method by which potential competing signals from compo¬ nents in the sample other than glucose can be evaluated under different operating conditions and with different polymers. Furthermore, this example demonstrates the potential application of the current invention to the detection and measure¬ ment of different compounds.

,30

D-Glucosamine binding studies on polymer IV at pH 10.25 by pH static titration:

Polymer IV (0.100 g) from Example 8 was suspended in 0.980 mL of pH 10.25 NaOH solution in a pH titration vessel, which was kept at 25 °C by a constant temperature water bath. The pH of the suspension was adjusted to 10.25 by addition of 6N and 0.1 N sodium hydroxide solution, and pH 10.25 NaOH solution (0.020mL total). A solution of D-glucosamine with concentration of 0.500 M and pH of 10.25 was titrated into the polymer suspension sequen¬ tially. After each addition, 0.10 N sodium hydroxide solution was added to maintain pH of the system at 10.25. The system was kept stirring for a few minutes until equilibrium was reached. The volumes of the glucose solution injected and of the 0.10 N sodium hydroxide solution added were recorded for data analysis. The titration data are plotted in FIG. 15, along with the data for glucose binding. D-Glucosamine binds well to polymer IV (2) at pH 10.25. However, it releases fewer protons than glucose does at the same concentration (FIG. 15).

Since glucosamine can release protons upon binding to the polymer, this method is useful for monitoring its concentration in solution, provided large concentrations of other, competing compounds such as glucose are not present. Other conditions of pH or the use of other metal complexes selective for glucosamine can be investigated in order to enhance the proton release.

Urea binding

Urea caused very little change in pH when titrated into a suspension of polymer IV under the same conditions as used for D-glucosamine titration.

Lactic acid binding

Lactic Acid (6) should be able to bind to polymer IV (2) well at pH 10.25. However, at high pH, its binding to Cu(TACN) is not expected to release proton (see Scheme 1 below). Titration of 0.5M, pH 10.25 lactic acid solution into a suspension of 100 mg of polymer IV in 1.00 mL water (pH = 10.25) did not change the pH of the polymer mixture very much. At the beginning of the titration, the pH slightly increased (Table 6). As more lactic acid solution was added, the pH went down a small amount. Overall, the pH changes were very small. Thus lactic acid should not interfere with the signal generated by glucose during glucose monitoring of samples containing small amounts of lactic acid. Scheme 1

TABLE 6 pH depression titration of polymer IV with lactic acid

titration V_acid <U PH

No.#

0 0.0 10.25

1 2.5 10.28

2 5.0 10.28

3 10.0 10.27

4 20.0 10.27

5 25.0 10.26

6 30.0 10.26

7 32.5 10.25

8 35.0 10.24

9 40.0 10.24

10 60.0 10.22

Having thus described exemplary embodiments of the present invention, it should be noted by those skilled in the art that the disclosures herein are exemplary only and that various other alternations, adaptations and modifications may be made within the scope of the present invention. Accordingly, the present invention is not limited to the specific embodiments as illustrated herein.

The following list of references provides additional background information relating to the present invention. The references cited in the following list and other references and patents referred to in this specification are hereby incorpo¬ rated by reference. REFERENCES

1. Pickup, J. C. and Williams, G. (eds) Textbook of Diabetes, Blackwell Scientific Publications, 1991.

2. Jannson, P.A., Fowelin, J., Smith, U. and Lonnroth, P., "Characterization by microdialysis of intercellular glucose level in subcutaneous tissue in humans," Am. J. Physiol., 255, 218-220, 1988.

3. Tamada, J.A., Bohannon, N.J.V., Potts, R.O., "Measurement of glucose in diabetic subjects using noninvasive transdermal extraction," Nature Medicine, 1 , 1 198-1201 , 1995.

4. Turner, A.D.F.; Kraube, I. and Wilson, G.S. (eds) Biosensors: Fundamental and Applications, Oxford University Press, 1987.

5. Wilkins, E. and Wilkins, M.G., "Implantable Glucose Sensor," J. Biomed. Eng. 5, 309-315, 1983.

6. Shaw, G.W., Claremont, D.J. and Pickup, J.C, "In vitro testing of a simply constructed, highly stable glucose sensor suitable for implantation in diabetic patients," Biosensors & Bioelectronics, 6, 401-406, 1991.

7. Pickup, J., "Developing glucose sensors for in vivo use," TIBTECH. 1 1 , 285-291 , 1993.

8. Pfeiffer, E., "Artificial pancreas, glucose sensors and the impact upon diabetology," Int'/ J. Artifi. Org. 16, 636-644, 1993.

9. Kawagoe, J.L., Niehaus, D.E. and Wrightman, R.M., "Enzyme-modified organic conducting salt microelectrode," Anal. Chem., 63, 2961 -2965, 1991.

10. Meyerhoff, C, Bischol, F., Mennel, F.J., Sternberg, F., Pfeiffer, E.F., "Use of the microdialysis technique in the monitoring of subcutaneous tissue glucose concentration," Int'I J. Artif. Org., 16, 268-275, 1993.

1 1 . Moussy, F., Harrison, D.J., O'Brien, D.W., Rajotte, R.V., "Performance of subcutaneously implanted needle-type glucose sensors employing a novel trilayer coating," Anal. Chem., 65, 2072-2077, 1993.

12. Bambot, S.B. et al.. Biosensors & Bioelectronics, 10, 643-652 (1995).

13. Kullick, T. et al., Analyt. Chim. Acta., 1994, 296, 263-26.

14. Vering, T. et al., Electroanalusis, 1994, 6, 953-956.

15. Bambot, S.B., Lakowicz, J.R. and Rao, G., "Potential applications of lifetime-based, phase-modulation fluorimetry in bioprocess and clinical monitoring," TIBTECH. 13, 106-1 15, 1995. 16. Wolfbeis, O.S., "Optical sensing based on analyte recognition by enzymes, carriers and molecular interactions, Analytica Chimica Acta, 250, 181 -210, 1991.

17. Trettnak, W., Leiner, M. J.P. and Wolfbeis, O.S., "Fibre-optic glucose sensor with a pH optrode as the transducer," Biosensors, 4, 1 5-26 (1988).

18. Bambot, S.B., Sipior, J., Lakowicz, J.R. and Rao, G., "Lifetime-based optical sensing of pH using resonance energy transfer in sol-gel films," Sensors and Actuators B, 11, 181 -188, 1994.

19. Szmacinski, H. and Lakowicz, J.R., "Optical measurements of pH using fluorescence lifetimes and phase-modulation fluorometry," Anal. Chem., 65, 1668-1674, 1993.

20. Lakowicz, J.R., Szmacinski, H. and Karakelle, M., "Optical sensing of pH and pC0 phase-modulation fluorimetry and resonance energy transfer," Analytica Chimica Acta, 272, 179-186, 1993.

21. Bambot, S.B., Rao, G., Romauld, M., Carter, M., Carter, G.M., Sipior, J., Terpetchnig, E. and Lakowicz, J.R., "Sensing oxygen through skin using a red diode laser and fluorescence lifetimes," Biosensors and Bioelectronics, 10, 643-652, 1995.

22. Werner, T. and Wolfbeis, O.S., "Optical sensor for the pH 10-13 range using a new support material," Fresenius J. Anal. Chem., 346, 564-568, 1993.

23. Lippitsch, M.E., Pusterhofer, J., Leiner, M.J. P. and Wolfbeis, O.S., Analytica Chimica Acta, 205, 1 -6, 1988.

24. Weigl, B.H. and Holobar, A., Rodriguez, N.V. and Wolfbeis, O.S., "Chemically and mechanically resistant carbon dioxide optrode based on a covalently immobilized pH indicator," Analytica Chimica Acta, 181, 335- 343, 1993.

25. Hitzmann, B. and Kullick, T., "Evaluation of pH field effect transistor measurement signals by neural networks," Analytica Chimica Acta, 294, 243-249, 1994.

26. Saito, A., Ito, N., Kumura, J. and Kuriyama, T., Sensors and Actuators B, 20, 125-129, 1994.

27. Bergveld, P., van Hal, R.E.G. and Eijkel, J.C.T., "The remarkable similarity between the acid-base properties of ISFETs and proteins and the consequences for the design of ISFET biosensors," Biosensors and Bioelectronics, 10, 405-414, 1995.

28. Wilkinson, S.G. (ed) Comprehensive Coordination Chemistry, Vol. 2, Pergamon Press, 1987, pp. 47. 29. a) Dung, N-H., Viossat, B., Busnot, A et al., Inorganica Chimica Acta, 1990, 1 69, 9-12; b) ibid, 1990, 174, 145-148.

30. Bereman, R.D., Churchill, M.R., Schaber, P.M. and Winkler, M.E., Inorg. Chem., 1979, 18, 3122.

Claims

CLAIMS What is claimed is:

1. A sensor for use in measuring the presence of a target molecule in aqueous solution, said sensor comprising: a metal complex which binds to said target molecule in aqueous solution wherein binding of said metal complex to said molecules releases a proton.

2. A sensor according to claim 1 wherein said metal complex is attached to a support structure.

3. A sensor according to claim 1 wherein said metal complex contains a polymerizable moiety and is embedded by copolymerization with a monomer and crosslinking agent into a polymeric matrix.

4. A sensor according to claim 1 wherein said metal complex contains a polymerizable moiety and is preorganized with a template molecule and is embedded by copolymerization with a monomer and crosslinking agent into an imprinted polymeric matrix, where said template molecule is the target molecule or an analog of the target molecule.

5. A sensor according to claim 1 wherein said metal complex contains a metal ion selected from the group of metals consisting of copper, lead, cobalt, nickel, vanadium, mercury, iron, aluminum, barium, borate, calcium and Y^{3 +} , La^{3 +} .

6. A sensor according to ciaim 1 wherein said metal complex is a nitrogen-based bidentate ligand, nitrogen-based linear tridentate ligand, nitrogen- based linear tetradentate ligand, nitrogen-based tridentate macrocycle ligand, nitrogen-based tridentate ligand with pendant arms, nitrogen-based tetradentate macrocyclic ring, ligand or tridentate ligand comprising pyridine, pyrazole or imidazole rings.

7. A sensor according to claim 1 wherein said metal complex is selected from the group consisting of copperdD-triazacyclononane (Cu(TACN)), copper (ll)-1 ,4-dialkyl triazacyclononane, copper(ll)-1 ,4-dimethyl triazacyclononane (Cu(1 ,4-dimethyl-TACN)), copper(ll)-ethylenediamine (Cu(En)), copper(ll)- propylenediamine (Cu(ll)-PDN)), copper(ll)-iminodiacetate (Cu(IDA)), copper(ll)- diethγlenetriamine (Cu(Dien)) and derivatives thereof.

8. A sensor according to claim 1 wherein said metal complex is copper D-triazacyclononane (Cu(TACN)) and derivatives thereof.

9. A sensor system according to claim 8 wherein said derivative of CudD-triazacyclononane comprises a polymerizable group selected from the group consisting of styryl, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane and epoxy.

10. A sensor according to claim 7 wherein said derivative of Cu(ll)-1 ,4- dialkyl triazacyclononane comprises a polymerizable group selected from the group consisting of styryl, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane and epoxy, wherein said alkyl group is methyl, ethyl or propyl.

1 1. A sensor according to claim 7 wherein said derivative of Cu(ll)-1 ,4- dimethyl triazacyclononane comprises a polymerizable group selected from the group consisting of styryl, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane or epoxy.

12. A sensor according to claim 9 wherein said polymerizable metal complex is styryl-Cu(ll)-triazacyclononane (Cu(Msty-TACN)).

13. A sensor according to claim 9 wherein said polymerizable metal complex is styryl-Cu(ll)-1 ,4-dialkyl triazacyclononane.

14. A sensor according to claim 13 wherein said polymerizable metal complex is styryl-Cu(ll)-1 ,4-dimethyl triazacyclononane.

15. A sensor according to claim 9 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-triazacyclononane (Cu(Dsty-TACN)).

16. A sensor according to claim 9 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-alkyl triazacyclononane.

17. A sensor according to claim 16 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-methyl triazacyclononane.

18. A sensor according to claim 9 wherein said polymerizable metal complex is tris-styryl-Cu(ll)-triazacyclononane (Cu(Tsty-TACN)).

19. A sensor according to claim 1 wherein said target molecule to which said sensor binds is selected from the group consisting of glucose, mannose, sialic acid, aminosugars, disaccharides, trisaccharides, oligosaccharides, sugar-amino acids, sugar peptides, giycoproteins, glycerol, dopamine, catechols, ascorbic acid, polyols, diols, ethyleneglycol, and 3-hydroxy-L-tyrosine.

20. A sensor according to claim 2 wherein said support structure comprises a polymer.

21 . A sensor according to claim 2 wherein said support structure is an optically transparent surface such as quartz, glass, or a fibre optic cable.

22. A sensor according to claim 1 wherein said metal complex contains a polymerizable moiety and is preorganized with a template molecule and is embedded by copolymerization with a monomer and crosslinking agent into an imprinted polymeric matrix, where said template molecule is the target molecule or an analog of the target molecule.

23. A sensor according to claim 1 wherein said target molecule is a sugar.

24. A sensor according to claim 23 wherein said sugar is glucose.

25. A detection system for measuring the presence of a target molecule in an aqueous solution, said sensor system comprising: an aqueous solution which is to be tested to measure the amount of said target molecule present in said solution; a sensor comprising: a metal complex which binds to said target molecule in aqueous solution wherein binding of said metal complex to said target molecule releases a proton; a detector for measuring the release of protons from said metal complex to thereby provide measurement of target molecules present in said aqueous solution.

26. A detection system according to claim 19 wherein said metal complex is attached to a support structure.

27. A detection system according to claim 25 wherein said metal complex contains a polymerizable moiety and is embedded by copolymerization with a monomer and crosslinking agent into a polymeric matrix.

28. A detection system according to claim 25 wherein said metal complex contains a metal ion selected from the group of metals consisting of copper, lead, cobalt, nickel, vanadium, mercury and iron.

29. A detection system according to claim 25 wherein said metal complex is a nitrogen-based bidentate ligand, nitrogen-based linear tridentate ligand, nitrogen-based linear tetradentate ligand, nitrogen-based tridentate macrocycle ligand, nitrogen-based tridentate ligand with pendant arms, nitrogen- based tetradentate macrocyclic ring, ligand or tridentate ligand comprising pyridine, pyrazole or imidazole rings.

30. A detection system according to claim 25 wherein said metal complex is selected from the group consisting of copperdD-triazacyclononane (Cu(TACN)), copper (ll)-1 ,4-dialkyl triazacyclononane, copper(ll)-1 ,4-dimethyl triazacyclononane (Cud ,4-dimethyl-TACN)), copperdD-ethylenediamine (Cu(En)), copperdD-propylenediamine (Cu(ll)-PDN)), copper(ll)-iminodiacetate (Cu(IDA)), copperdD-diethylenetriamine (Cu(Dien)) and derivatives thereof.

31. A detection system according to claim 30 wherein said metal complex is copperdD-triazacyclononane (Cu(TACN)) and derivatives thereof.

32. A detection system according to claim 25 wherein said target molecule is selected from the group consisting of glucose, mannose, sialic acid, aminosugars, disaccharides, trisaccharides, oligosaccharides, sugar-amino acids, sugar-peptides, giycoproteins, glycerol, dopamine, catechols, ascorbic acid, polyols, diols, ethyleneglycol, and 3-hydroxy-L-tyrosine.

33. A detection system according to claim 26 wherein said support structure comprises a polymer.

34. A detection system according to claim 26 wherein said support structure is an optically transparent surface such as quartz, glass, or a fibre optic cable.

35. A detection system according to claim 25 wherein said metal complex contains a polymerizable moiety and is preorganized with a template molecule and is embedded by copolymerization with a monomer and crosslinking agent into an imprinted polymeric matrix, where said template molecule is the target molecule or an analog of the target molecule.

36. A detection system according to claim 25 wherein said target molecule is a sugar.

37. A detection system according to claim 36 wherein said sugar is glucose.

38. A detection system according to claim 32 wherein said aqueous solution is mammalian bodily fluid.

39. A detection system according to claim 33 wherein said metal complex comprises a monomer which is co-polymerized with said support structure polymer.

40. A detection system according to claim 25 wherein said detector measures the release of protons from said metal complex by measuring the pH of said solution.

41. A detection system according to claim 40 wherein said detector signals the release of protons from said metal complex optically, by means of a pH-sensitive chromophore and a spectrophotometer.

42. A detection system according to claim 40 wherein said detector signals the release of protons from said metal complex optically, by means of changes in the fluorescence lifetime of a pH-sensitive chromophore.

43. A detection system according to claim 25 wherein said detector measures the release of protons from said metal complex by using a light addressable potentiometric sensor.

44. A detection system according to claim 25 wherein said detector measures the release of protons from said metal complex using a field effect transistor.

45. A detection system according to claim 30 wherein said derivative of triazacyclononane comprises a polymerizable group selected from the group consisting of styryl, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane and epoxy.

46. A detection system according to claim 30 wherein said derivative of CuOD-1 ,4-dialkyl triazacyclononane comprisesa polymerizable groupselected from the group consisting of styryl, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane, epoxy, alkylhydroxyl, alkylamine wherein said alkyl group is methyl, ethyl or propyl.

47. A detection system according to claim 30 wherein said derivative of Cu(ll)-1 ,4-dimethyl triazacyclononane comprises a polymerizable group selected from the group consisting of styryl, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane and epoxy.

48. A detection system according to claim 45 wherein said polymerizable metal complex is styryl-Cu(ll)-triazacyclononane.

49. A detection system according to claim 45 wherein said polymerizable metal complex is styryl-Cu(ll)-1 ,4-dialkyl triazacyclononane.

50. A detection system according to claim 49 wherein said polymerizable metal complex is styryl-Cu(ll)-1 ,4-dimethyl triazacyclononane.

51. A detection system according to claim 45 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-triazacyclononane (Cu(Dsty-TACN)).

52. A detection system according to claim 45 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-alkyl triazacyclononane.

53. A detection system according to claim 52 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-methyl triazacyclononane.

54. A detection system according to claim 45 wherein said polymerizable metal complex is tris-styryl-Cu(ll)-triazacyclononane (Cu(Tsty-TACN)).

55. A method for measuring the presence of a target molecule in an aqueous solution, said method comprising the steps of: placing said aqueous solution in contact with a sensor in an alkaline environment which comprises: a metal complex which binds to said target molecule in said aqueous solution wherein binding of said metal complex to said target molecule releases a proton; and measuring the release of protons from said metal complex in order to provide an indirect measurement of target molecules present in said aqueous solution.

56. A method according to claim 55 wherein said aqueous solution is a mammalian bodily fluid and said sugar is glucose.

57. A method according to claim 55 wherein said metal complex contains a polymerizable moiety and is embedded by copolymerization with a monomer and crosslinking agent into a polymeric matrix.

58. A method according to claim 55 wherein said metal complex contains a polymerizable moiety and is preorganized with a template molecule and is embedded by copolymerization with a monomer and crosslinking agent into an imprinted polymeric matrix, where said template molecule is the target molecule or an analog of the target molecule.

59. A method according to claim 55 wherein said metal complex contains a metal ion selected from the group of metals consisting of copper, lead, cobalt, nickel, vanadium, mercury, iron and aluminum.

60. A method according to claim 55 wherein said metal complex is a nitrogen-based bidentate ligand, nitrogen-based linear tridentate ligand, nitrogen- based linear tetradentate ligand, nitrogen-based tridentate macrocycle ligand, nitrogen-based tridentate ligand with pendant arms, nitrogen-based tetradentate macrocyclic ring, ligand or tridentate ligand comprising pyridine, pyrazole or imidazole rings.

61. A method according to claim 55 wherein said metal complex is selected from the group consisting of copperdD-triazacyclononane (Cu(TACN)), copper (II)- 1 ,4-dialkyl triazacyclononane, copper(ll)-1 ,4-dimethyl triazacyclononane (Cud ,4-dimethyl-TACN)), copperdD-ethylenediamine (Cu(En)), copper(ll)- propylenediamine <Cu(ll)-PDN)), copperdD-iminodiacetate (Cu(IDA)), copper(ll)- diethylenetriamine (Cu(Dien)) and derivatives thereof.

62. A method according to claim 61 wherein said metal complex is copperdD-triazacyclononane (Cu(TACN)) and derivatives thereof.

63. A method system according to claim 62 wherein said derivative of Cu(ll)-triazacyclononane comprises a polymerizable group selected from the group consisting of styrene, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane and epoxy.

64. A method according to claim 61 wherein said derivative of Cu(ll)- 1 ,4-dialkyl triazacyclononane comprises a polymerizable group selected from the group consisting of styryl, methacrylate, acrylate, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane, epoxy, alkylhydroxyl, alkylamine wherein said alkyl groups are methyl, ethyl and propyl.

65. A method according to claim 61 wherein said derivative of Cu(ll)- 1 ,4-dimethyl triazacyclononane comprises a polymerizable group selected from the group consisting of styryl, methacrylate, acrylate, vinyl, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane and epoxy.

66. A method according to claim 63 wherein said polymerizable metal complex is styryl-Cu(ll)-triazacyclononane.

67. A method according to claim 64 wherein said polymerizable metal complex is styryl-Cu{ll)-1 ,4-dialkyl triazacyclononane.

68. A method according to claim 67 wherein said polymerizable metal complex is styryl-Cu(ll)-1 ,4-dimethyl triazacyclononane.

69. A method according to claim 63 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-triazacyclononane (Cu(Dsty-TACN)).

70. A method according to claim 63 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-alkyl triazacyclononane.

71 . A method according to claim 70 wherein said polymerizable metal complex is bis-styryl-Cu(ll)-methyl triazacyclononane.

72. A method according to claim 63 wherein said polymerizable metal complex is tris-styryl-Cu(ll)-triazacyclononane (Cu(Tsty-TACN)).

73. A method according to claim 55 wherein said target molecule to which said metal complex binds is selected from the group consisting of glucose, mannose, sialic acid, aminosugars, disaccharides, trisaccharides, oligosaccharides, sugar-amino acids, sugar-peptides, giycoproteins, glycerol, dopamine, catechols, ascorbic acid, polyols, diols, ethyleneglycol, and 3-hydroxy-L-tyrosine.

74. A method according to claim 55 wherein said support structure comprises a polymer.

75. A method according to claim 55 wherein said support structure is an optically transparent surface such as quartz, glass, or a fibre optic cable.

76. A method according to claim 55 wherein said metal complex contains a polymerizable moiety and is preorganized with a template molecule and is embedded by copolymerization with a monomer and crosslinking agent into an imprinted polymeric matrix, where said template molecule is the target molecule or an analog of the target molecule.

77. A method according to claim 73 wherein said target molecule is a sugar.

78. A method according to claim 77 wherein said sugar is glucose.

79. A metal complex comprising a compound having the formula

wherein M is copper or iron; X is H or a functional group selected from styrene, methacrylate, acrylate, vinyl ether, vinyl acetate, trialkoxysilane, dialkyl¬ chlorosilane, epoxy and alkylhydroxyl or alkylamine groups having from 1 to 3 carbon atoms. Y is H or a functional group selected from styrene, methacrylate, acrylate, vinyl ether, vinyl, vinyl acetate, trialkoxysilane, dialkylchlorosilane, epoxy and alkyl, alkylhydroxyl or alkylamine groups having from 1 to 3 carbon atoms. Z is H or a functional group selected from styrene, methacrylate, acrylate, vinyl ether, vinyl acetate, trialkoxysilane, dialkylchlorosilane, epoxy and alkyl, alkylhydroxyl or alkylamine groups having from 1 to 3 carbon atoms.

80. A metal complex according to claim 79 wherein M is Cu; X is H; Y is H and Z is styrene.

81. A metal complex according to claim 79 wherein M is Cu; X is H; Y is styrene and Z is styryl.

82. A metal complex according to claim 79 wherein M is Cu; X, Y and Z are styryl.