Description

The present invention relates to an electrode system for measuring the concentration of an analyte under in-vivo conditions, comprising an electrode with immobilized enzyme molecules and an improved diffusion barrier that controls diffusion of the analyte from body fluid surrounding the electrode system to the enzyme molecules.

Sensors with implantable or insertable electrode systems facilitate measurements of physiologically significant analytes such as, for example, lactate or glucose in a patient's body. The working electrodes of systems of this type have electrically conductive enzyme layers in which enzyme molecules are bound which release charge carriers by catalytic conversion of analyte molecules. In the process, an electrical current is generated as measuring signal whose amplitude correlates to the analyte concentration.

Such electrode systems are e.g. known from WO 2007/147475 and WO 2010/028708, the contents of which are herein incorporated by reference.

The working electrodes of the electrode system are provided with a diffusion barrier that controls the diffusion of the analyte to be determined from the body fluid or tissue surrounding the electrode system to the enzyme molecules that are immobilized in the enzyme layer. According to WO 2010/028708, the diffusion barrier of the electrode system is a solid solution of at least two different polymers, preferably of acryiates. The polymers may be copolymers, e.g. copolymers of methyl methacrylate and hydroxyethyl methacrylate or copolymers of butyl methacrylate and hydroxyethyl methacrylate.

WO 2007/147475 discloses a diffusion barrier made from a polymer having a zwitterionic structure. An example of such a polymer is poly(2- methacryloyloxyethyl phosphorylcholine-co-n-butylmethacrylate). The zwitterionic polymer may be mixed with another polymer, for example polyurethane.

The use of polymer or copolymer mixtures, however, has drawbacks in that the preparation of the mixture and its application to the sensor is tedious and potentially problematic. Usually, the polymers to be mixed are individually dissolved and the resulting solutions are thereafter mixed in the desired ratio. This, however, may result in precipitation of one of the components and consequently in workability problems, e.g. in a spraying process. Increased difficulties occur when the mixture comprises a polymer with ionic characteristics, i.e. when one of the polymers to be mixed comprises a monomer having anionic or cationic groups. The presence of such charged groups, however, has a strong effect on the solubility, making it difficult to find a solvent suitable for both the charged polymer and an uncharged polymer.

WO 2006/058779 discloses an enzyme-based sensor with a combined diffusion and enzyme layer comprising at least one polymer material, and particles carry an enzyme, wherein the particles are dispersed in the polymer material. The polymer may comprise hydrophilic as well as hydrophobic polymer chain sequences, for example, the polymer may be a high or low water uptake polyether-polyurethane copolymer. The use of block copolymers having at least one hydrophilic block and at least one hydrophobic block as a diffusion layer is not disclosed.

EP-A-2 163 190 describes an electrode system for the measurement of an analyte concentration in-vivo comprising a counterelectrode with an electric conductor, and a working electrode with an electric conductor on which an enzyme layer comprising immobilized enzyme molecules is localized. A diffusion barrier controls the diffusion of the analyte from surrounding body fluids to the enzyme molecules. The diffusion barrier may comprise hydrophilized polyurethanes obtainable by polycondensation of 4,4'- methylene-bis-(cyclohexylisocyanate) and diol mixtures which may be polyethyleneglycol and polypropyleneglycol. The hydrophilic polyurethane layer may be covered with a spacer, e.g. a copolymer of butyl methacrylate and 2-methacryloyloxyethyl-phosphoryl choline. The use of block copolymers having at least one hydrophilic block and at least one hydrophobic block as a diffusion layer is not disclosed.

It is an object of the present invention to provide a diffusion barrier on an electrode system of an enzymatic in-vivo sensor which provides desirable physico-chemical characteristics and which can be manufactured easily.

This object is met by providing a diffusion barrier consisting of a single block copolymer having at least one hydrophilic block and at least one hydrophobic block. The hydrophilic and hydrophobic blocks are covalently linked to each other. Preferably the blocks are (meth)acrylate polymer blocks.

The block-copolymer based diffusion barrier provides excellent physico- chemical characteristics as follows:

(i) permeability of the diffusion barrier for the analyte to be determined,

(ii) permeability characteristics of the diffusion barrier which are suitable for the short-term behaviour (wettability) and the long-term behaviour (sensor drift) of the electrode,

(iii) mechanical flexibility of the diffusion barrier.which allows manufacture of in-vivo sensors with extended multiple electrodes;

(iv) efficient incorporation of ionic groups into the diffusion layer, i.e. the

density of cationic or anionic charges within the polymer can be efficiently adjusted, this is relevant for repulsion or attraction of charged analytes, and/or control of cell adhesion, e.g. of monocytes from the surrounding body fluid or tissue.

A subject-matter of the present invention is an electrode system for measuring the concentration of an analyte under in-vivo conditions, comprising an electrode with immobilized enzyme molecules and a diffusion barrier that controls diffusion of the analyte from the exterior of the electrode system to the enzyme molecules, characterized in that the diffusion barrier comprises a block copolymer having at least one hydrophilic block and at least one hydrophobic block.

Preferably, the diffusion barrier comprises a single, i.e. only one block copolymer having at least one hydrophilic block and at least one hydrophobic block, i.e. further polymers or copolymers are absent. More preferably, the diffusion barrier consists of a single block copolymer having at least one hydrophilic block and at least one hydrophobic block.

The electrode system of the present invention is suitable for insertion or implantation into a body, e.g. a mammalian body such as a human body. The electrode system is adapted for measuring a desired analyte in body fluid and/or body tissue, e.g. in the extracellular space (interstitium), in blood or lymph vessels or in the transcellular space.

The inserted or implanted electrode system is suitable for short-term application, e.g. 3-14 days, or for long-term application, e.g. 6-12 months. During the insertion or implantation period a desired analyte may be determined by continuous or discontinuous measurements.

The electrode system of the invention is preferably part of an enzymatic, non-fluidic (ENF) sensor, wherein enzymatic conversion of the analyte is determined. Preferably, the sensor comprises a working electrode with immobilized enzyme molecules for the conversion of the analyte which results in the generation of an electrical signal. The enzymes may be present in a layer covering the electrode. Additionally, redox mediators and/or electro-catalysts as well as conductive particles and pore formers may be present. This type of electrode is described e.g. in WO 2007/147475, the content of which is herein incorporated by reference.

The area of the working electrode is the sensitive area of the sensor. This sensitive area is provided with a diffusion barrier that controls diffusion of the analyte from the exterior, e.g. body fluid and/or tissue surrounding the electrode system to the enzyme molecules. The diffusion barrier can, for example, be a cover layer covering the enzyme layer, i.e. an enzyme-free layer. However, it is feasible just as well that diffusion-controlling particles are incorporated into the enzyme layer to serve as a diffusion barrier. For example, pores of the enzyme layer can be filled with the polymer which controls the diffusion of analyte molecules. The thickness of the diffusion barrier is usually from about 2-20 μιτι, e.g. from about 2-15 μιτι, or from about 5-20 pm, particularly from about 5-10 μητι or from about 10-15 pm (in dry state).

The diffusion barrier of the electrode system of the present invention comprises a block copolymer, preferably a single block copolymer having at least one hydrophilic block and at least one hydrophobic block. The block copolymer may comprise an alternating sequence of blocks, i.e. a hydrophilic block is linked to a hydrophobic block. The hydrophilic and hydrophobic blocks are covalently linked to each other within a polymer molecule. The average molecular weight of the polymer (by weight) is usually from 20-70 kD, particularly from 25-60 kD and more particularly from 30-50 kD. The molar ratio of the hydrophilic to hydrophobic portions in the block copolymer is usually in the range from about 75% (hydrophilic) : 25% (hydrophobic) to about 25% (hydrophilic) : 75% (hydrophobic), in the range from about 65% (hydrophilic) : 35% (hydrophobic) to about 35% (hydrophilic) : 65% (hydrophobic) or in the range from about 60% (hydrophilic) : 40% (hydrophobic) to about 40% (hydrophilic) : 60% (hydrophobic). A hydrophilic block of the block copolymer consists of at least 90%, at least 95% and particularly completely of hydrophilic monomeric units. It usually has a length of from 50-400, e.g. from 50-200, or from 150-300 particularly from 100-150, or from 200-250 monomeric molecules. A hydrophobic block of the copolymer consists of at least 90%, more particularly at least 95% and even more particularly completely of hydrophobic monomeric units. It has usually a length of from 50-300, e.g. from 50-200, or from 150-250, particularly from 80-150, or from 170-200 monomeric units.

The hydrophilic blocks and/or the hydrophobic blocks preferably consist of (meth)acrylic-based units. More preferably, both the hydrophilic blocks and the hydrophobic blocks consist of (meth)acrylic-based monomeric units.

The hydrophilic monomeric units of the hydrophilic block are preferably selected from hydrophilic (meth)acryl esters, i.e. esters with a polar, i.e. OH, OCH ₃ or OC2H5 group within the alcohol portion of the ester, hydrophilic (meth)acrylamides with an amide (NH ₂ ) or an N-alkyl- or N,N-dialkylamide group, wherein the a Iky I group comprises 1-3 C-atoms and optionally hydrophilic groups such as OH, OCH ₃ or OC ₂ H ₅ , and suitable (meth)acrylic units having a charged, e.g. an anionic or cationic group, such as acrylic acid (acrylate) or methacrylic acid (methacrylate). Further, combinations of monomeric units may be employed.

Specific examples of preferred monomeric units for the hydrophilic block are selected from:

2-hydroxyethyl acrylate,

2-hydroxyethyl methacrylate (HEMA),

2-methoxyethyl acrylate,

2-methoxyethyl methacrylate,

2-ethoxyethyl acrylate,

2-ethoxyethyl methacrylate, 2- or 3-hydroxypropyl acrylate,

2- or 3-hydroxypropyl methacrylate (2- or 3-HPMA),

2- or 3-methoxypropyl acrylate,

2- or 3-methoxypropyl methacrylate,

2- or 3-ethoxypropyl acrylate,

2- or 3-ethoxypropyl methacrylate,

1 - or 2-glycerol acrylate,

1 - or 2-glycerol methacrylate,

acrylamide,

methacrylamide,

an N-alkyl- or Ν,Ν-dialkyl acrylamide, and

an N-alkyl- or Ν,Ν-dialkyl methylamide, wherein a Iky I comprises

1 -3 C-atoms such as methyl, ethyl or propyl,

acrylic acid (acrylate),

methacrylic acid (methacrylate)

and combinations thereof.

Preferred hydrophilic monomers are 2-hydroxyethyl methacrylate (HEMA) and/or 2- or 3-hydroxypropyl methacrylate (2- or 3-HPMA). More preferably, the hydrophilic block consists of at least two different hydrophilic monomeric units. For example, it may be a random copolymer of at least two different hydrophilic monomeric units such as HEMA and 2-HPMA.

In order to introduce ionic groups into the monomer, charged monomeric units such as acrylic acid (acrylate) and/or methacrylic acid (methacrylate) may be incorporated into the hydrophilic block. Thus, in a particular embodiment of the present invention, the hydrophilic block can be made from at least one non-ionic hydrophilic monomeric unit (e.g. as described above) and from at least one ionic hydrophilic monomeric unit, wherein the ionic monomeric unit is present in a molar amount of preferably 1 -20 mole- %. In case the hydrophilic block comprises an ionic monomeric unit such as acrylic acid or methacrylic acid, copolymerization with a hydrophilic monomer selected from the group of (meth)acryiamides, particularly Ν,Ν-dialkyl acryl- or methacrylamides is preferred.

The hydrophobic monomeric units of the hydrophobic block are preferably selected from hydrophobic acrylic and/or methacrylic units, styrene-based monomeric units or combinations thereof. Preferably the hydrophobic monomeric units are selected from hydrophobic (meth)acryl esters, e.g. esters having an alcohol portion with 1 -3 C-atoms without hydrophilic group. Specific examples of monomeric units for the hydrophobic block are selected from: methyl acrylate,

methyl methacrylate (MMA),

ethyl acrylate

ethyl methacrylate (EMA),

n- or i-propyl acrylate,

n- or i-propyl methacrylate,

n-butyl acrylate,

n-butyl methacrylate (BUMA),

neopentyl acrylate,

neopentyl methacrylate,

and combinations thereof.

The hydrophobic block preferably comprises at least two different hydrophobic monomeric units, which are e.g. present as a random copolymer. In a preferred embodiment, the hydrophobic block comprises methyl methacrylate (MMA) and n-butyl methacrylate (BUMA). In an especially preferred embodiment, the hydrophobic block is a random copolymer of MMA and BUMA. The molar ratio between MMA and BUMA is preferably about 60% (MMA) : 40% (BUMA) to about 40% (MMA) : 60% (BUMA), e.g. about 50% (MMA) : 50% (BUMA). The glass transition temperature of the hydrophobic block is preferably 100°C or less, 90°C or less or 80°C or less, e.g. about 40-80°C. In an alternative embodiment, the hydrophobic block may consist of styrenic units, e.g. of polystyrene having a glass transition temperature of about 95°C.

The block copolymers used in the present invention may be manufactured according to known methods (Boker et al., Macromolecules 34 (2001 ), 7477- 7488).

The block copolymers may be applied to the electrode system by usual techniques, e.g. by providing a solution of the block copolymer in a suitable solvent or solvent mixture, e.g. an organic solvent, such as ether, which is applied to the prefabricated electrode system and dried thereon.

When the block copolymer is contacted with water, it shows a water uptake of preferably about 15%-30% by weight (based on the polymer dry weight) at a temperature of 37°C and a pH of 7.4 (aqueous phosphate buffer 10 mM KH ₂ P0 ₄ , 10 mM NaH ₂ P0 ₄ and 147 mM NaCI).

In addition to the block copolymer the diffusion barrier may also comprise further components, particularly non-polymeric components, which may be dispersed and/or dissolved in the polymer. These further compounds include plasticizers, particularly biocompatible plasticizers, such as tri-(2-ethylhexyl) trimellitate and/or glycerol.

The diffusion barrier of the invention has a high effective diffusion coefficient Deff for glucose which is preferably > 10 ^"10 cm ² /s, more preferably > 5 0 ^"10 cm ² /s, and even more preferably > 10 ^~9 cm ² /s, and e.g. up to 10 ^~7 or 10 ^~8 cm ² /s at a temperature of 37°C and a pH of 7.4. The effective diffusion coefficient is preferably determined as described in Example 4 according to the equation:

D _e ff=SE _m /F-L _m -5182-10- ¹ wherein SE _m is the sensitivity of the working electrode, F is the area of the working electrode, and L _m is the layer thickness of the diffusion barrier. SE _m and L _m may be determined as described in the Examples.

The electrode system of the present invention is suitable for measuring the concentration of an analyte under in-vivo conditions, i.e. when inserted or implanted into a body. The analyte may be any molecule or ion present in tissue or body fluid, for example oxygen, carbon dioxide, salts (cations and/or anions), fats or fat components, carbohydrates or carbohydrate components, proteins or protein components, or other type of biomolecules. Especially preferred is the determination of analytes which can be efficiently transferred between body fluid, e.g. blood and tissue such as oxygen, carbon dioxide, sodium cations, chloride anions, glucose, urea, glycerol, lactate and pyruvate.

The electrode system comprises an enzyme immobilized on an electrode. The enzyme is suitable for the determination of a desired analyte. Preferably, the enzyme is capable of catalytically converting the analyte and thereby generating an electric signal detectable by the electric conductor of the working electrode. The enzyme for measuring the analyte is preferably an oxidase, for example glucose oxidase or lactate oxidase or a dehydrogenase, for example a glucose dehydrogenase or a lactate dehydrogenase. In addition to the enzyme, the enzyme layer may also comprise an electrocatalyst or a redox mediator which favours the transfer of electrons to conductive components of the working electrode, e. g. graphite particles. Suitable electro-catalysts are metal oxides such as manganese dioxide or organo-metallic compounds such as cobalt phthalo-cyanine. In a preferred embodiment the redox mediator is capable of degrading hydrogen peroxide thereby counteracting depletion of oxygen in the surroundings of the working electrode. In a different embodiment, a redox mediator may be covalently bound to the enzyme and thereby effect direct electron transfer to the working electrode. Suitable redox mediators for direct electron transfer are prosthetic groups, such as pyrrolo quinoline quinone (PQQ), flavine adenine dinucleotide (FAD) or other known prosthetic groups. Enzymes immobilized on electrodes are e.g. described in WO 2007/147475, the content of which is herein incorporated by reference.

A preferred embodiment of the electrode system comprises a counterelectrode with an electrical conductor and a working electrode with an electrical conductor on which an enzyme layer and the diffusion barrier are arranged. The enzyme layer is preferably designed in the form of multiple fields that are arranged on the conductor of the working electrode at a distance, e.g. at least 0.3 mm or at least 0.5 mm from each other. The individual fields of the working electrode may form a series of individual working electrodes. Between these fields, the conductor of the working electrode may be covered by an insulation layer. By arranging the fields of the enzyme layer on the top of openings of an electrically insulating layer, the signal-to-noise ratio can be improved. Such an arrangement is disclosed in WO 2010/028708, the content of which is herein incorporated by reference.

The electrode system of the invention may additionally comprise a reference electrode capable of supplying a reference potential for the working electrode, e.g. an Ag/Ag-CI reference electrode. Moreover, an electrode system according to the invention can have additional counter- and/or working electrodes.

The electrode system may be part of a sensor, e.g. by being connected to a potentiostat and an amplifier for amplification of measuring signals of the electrode system. The sensor is preferably an enzymatic non-fluidic (ENF) sensor, more preferably an electrochemical ENF sensor The electrodes of the electrode system may be arranged on a substrate that carries the potentiostat or be attached to a circuit board that carries the potentiostat. A further subject-matter of the invention is related to the use of a block copolymer having at least one hydrophilic block and at least one hydrophobic block as a diffusion barrier for an enzymatic electrode. The block copolymer is preferably as described above, e.g. a single block- copolymer. The diffusion barrier and the enzymatic electrode are preferably also as described above.

Further details and advantages of the invention are explained based on an exemplary embodiment making reference to the appended drawings.

Fig. 1 shows an exemplary embodiment of an electrode system according to the invention.

Fig. 2 shows a detail view of Fig. 1.

Fig. 3 shows another detail view of Fig. 1.

Fig. 4 shows a section along the section line CC of Fig. 2.

Fig. 5 shows the sensitivity (with standard deviation) of four glucose sensors (at 10 mM glucose) provided with different block polymers (C, F, D, B) as barrier layers.

Fig. 6 shows the sensor drift of four glucose sensors provided with different block copolymers (A, C, D, F) as barrier layers.

Fig. 7 shows the conductivity of block copolymer A dependent on time (2 experiments).

Fig. 8 shows the conductivity of block copolymer F dependent on time (3 experiments). Fig. 9 shows the conductivity of block copolymer H dependent on time for a layer thickness of 2.77 μιτι or 4.43 μιη, respectively.

Figure 1 shows an exemplary embodiment of an electrode system for insertion into body tissue of a human or animal, for example into cutis or subcutaneous fatty tissue. A magnification of detail view A is shown in Figure 2, a magnification of detail view B is shown in Figure 3. Figure 4 shows a corresponding sectional view along the section line, CC, of Figure 2.

The electrode system shown has a working electrode 1 , a counterelectrode 2, and a reference electrode 3. Electrical conductors of the electrodes 1 a, 2a, 3a are arranged in the form of metallic conductor paths, preferably made of palladium or gold, on a substrate 4. In the exemplary embodiment shown, the substrate 4 is a flexible plastic plate, for example made of polyester. The substrate 4 is less than 0.5 mm thick, for example 100 to 300 micrometers, and is therefore easy to bend such that it can adapt to movements of surrounding body tissue after its insertion. The substrate 4 has a narrow shaft for insertion into body tissue of a patient and a wide head for connection to an electronic system that is arranged outside the body. The shaft of the substrate 4 preferably is at least 1 cm in length, in particular 2 cm to 5 cm.

In the exemplary embodiment shown, one part of the measuring facility, namely the head of the substrate, projects from the body of a patient during use. Alternatively, it is feasible just as well, though, to implant the entire measuring facility and transmit measuring data in a wireless fashion to a receiver that is arranged outside the body.

The working electrode 1 carries an enzyme layer 5 that contains immobilized enzyme molecules for catalytic conversion of the analyte. The enzyme layer 5 can be applied, for example, in the form of a curing paste of carbon particles, a polymeric binding agent, a redox mediator or an electro-catalyst, and enzyme molecules. Details of the production of an enzyme layer 5 of this type are disclosed, for example, in WO 2007/147475, reference to which is be made in this context.

In the exemplary embodiment shown, the enzyme layer 5 is not applied continuously on the conductor 1 a of the working electrode 1 , but rather in the form of individual fields that are arranged at a distance from each other. The individual fields of the enzyme layer 5 in the exemplary embodiment shown are arranged in a series.

The conductor 1 a of the working electrode 1 has narrow sites between the enzyme layer fields that are seen particularly well in Figure 2. The conductor 2a of the counterelectrode 2 has a contour that follows the course of the conductor 1 a of the working electrode 1. This means results in an intercalating or interdigitated arrangement of working electrode 1 and counterelectrode 2 with advantageously short current paths and low current density.

In order to increase its effective surface, the counterelectrode 2 can be provided with a porous electrically conductive layer 6 that is situated in the form of individual fields on the conductor 2a of the counterelectrode 2. Like the enzyme layer 5 of the working electrode 1 , this layer 6 can be applied in the form of a curing paste of carbon particles and a polymeric binding agent. The fields of the layer 6 preferably have the same dimensions as the fields of the enzyme layer 5, although this is not obligatory. However, measures for increasing the surface of the counterelectrode can just as well be foregone and the counterelectrode 2 can just as well be designed to be a linear conductor path with no coatings of any kind, or with a coating made from the described block copolymer and optionally a spacer.

The reference electrode 3 is arranged between the conductor 1 a of the working electrode 1 and the conductor 2a of the counterelectrode 2. The reference electrode shown in Figure 3 consists of a conductor 3a on which a field 3b of conductive silver/silver chloride paste is arranged.

Figure 4 shows a schematic sectional view along the section line, CC, of Figure 2. The section line, CC, extends through one of the enzyme layer fields 5 of the working electrode 1 and between the fields of the conductive layer 6 of the counterelectrode 2. Between the fields of enzyme layer 5, the conductor 1 a of the working electrode 1 can be covered with an electrically insulating layer 7, like the conductor 2a of the counterelectrode 2 between the fields of the conductive layers 6, in order to prevent interfering reactions which may otherwise be catalyzed by the metal of the conductor paths 1 a, 2a. The fields of the enzyme layer 5 are therefore situated in openings of the insulation layer 7. Likewise, the fields of the conductive layer 6 of the counterelectrode 2 may also be placed on top of openings of the insulation layer 7.

The enzyme layer 5 is covered by a cover layer 8 which presents a diffusion resistance to the analyte to be measured and therefore acts as a diffusion barrier. The diffusion barrier 8 consists of a single copolymer with alternating hydrophilic and hydrophobic blocks as described above.

A favourable thickness of the cover layer 8 is, for example, 3 to 30 μιη, particularly from about 5-10 pm or from about 10-15 pm. Because of its diffusion resistance, the cover layer 8 causes fewer analyte molecules to reach the enzyme layer 5 per unit of time. Accordingly, the cover layer 8 reduces the rate at which analyte molecules are converted, and therefore counteracts a depletion of the analyte concentration in surroundings of the working electrode.

The cover layer 8 extends continuously essentially over the entire area of the conductor 1 a of the working electrode 1 . On the cover layer 8, a biocompatible membrane may be arranged as spacer 9 that establishes a minimal distance between the enzyme layer 5 and cells of surrounding body tissue. This means advantageously generates a reservoir for analyte molecules from which analyte molecules can get to the corresponding enzyme layer field 5 in case of a transient disturbance of the fluid exchange in the surroundings of an enzyme layer field 5. If the exchange of body fluid in the surroundings of the electrode system is transiently limited or even prevented, the analyte molecules stored in the spacer 9 keep diffusing to the enzyme layer 5 of the working electrode 1 where they are converted. The spacer 9 therefore causes a notable depletion of the analyte concentration and corresponding falsification of the measuring results to occur only after a significantly longer period of time. In the exemplary embodiment shown, the membrane forming the spacer 9 also covers the counterelectrode 2 and the reference electrode 3.

The spacer membrane 9 can, for example, be a dialysis membrane. In this context, a dialysis membrane is understood to be a membrane that is impermeable for molecules larger than a maximal size. The dialysis membrane can be prefabricated in a separate manufacturing process and may then be applied during the fabrication of the electrode system. The maximal size of the molecules for which the dialysis membrane is permeable is selected such that analyte molecules can pass, while larger molecules are retained.

Alternatively, instead of a dialysis membrane, a coating made of a polymer that is highly permeable for the analyte and water, for example on the basis of polyurethane or of acrylate, can be applied over the electrode system as spacer membrane 9.

Preferably, the spacer is made from a copolymer of (meth)acrylates. Preferably, the spacer membrane is a copolymer from at least 2 or 3 (meth)acrylates. More preferably, the spacer membrane comprises more than 50 mol-%, at least 60 mol-% or at least 70 mol-% hydrophilic monomer units, e.g. HEMA and/or 2-HPMA, and up to 40 mol-% or up to 30 mol-% hydrophiiic units, e.g. BUMA and/or MMA. The spacer may be a random or block copolymer. An especially preferred spacer membrane comprises MMA or BUMA as hydrophobic moieties and 2-HEMA and/or 2-HPMA as hydrophiiic moieties. The spacer membrane is highly permeable for the analyte, i.e. it does significantly lower the sensitivity per area of the working electrode, for example 20% or less, or 5% or less with a layer thickness of less than about 20 pm, preferably less than about 5 pm. An especially preferred thickness of the spacer membrane is from about 1 to about 3 pm.

The enzyme layer 5 of the electrode system can contain metal oxide particles, preferably manganese dioxide particles, as catalytic redox mediator. Manganese dioxide catalytically converts hydrogen peroxide that is formed, for example, by enzymatic oxidation of glucose and other bioanalytes. During the degradation of hydrogen peroxide, the manganese di-oxide particles transfer electrons to conductive components of the working electrode 1 , for example to graphite particles in the enzyme layer 5. The catalytic degradation of hydrogen peroxide counteracts any decrease of the oxygen concentration in the enzyme layer 5. Advantageously this allows the conversion of the analyte to be detected in the enzyme layer 5 to not be limited by the local oxygen concentration. The use of the catalytic redox mediator therefore counteracts a falsification of the measuring result by the oxygen concentration being low. Another advantage of a catalytic redox mediator is that it prevents the generation of cell-damaging concentrations of hydrogen peroxide.

The preferred spacer membrane polymer described herein may be used as an outer coating for a diffusion barrier of the present invention, but also as an outer coating of an electrode system in general, particularly of an electrode system for measuring the concentration of an analyte under in-vivo conditions, comprising an electrode with immobilized enzyme molecules and a diffusion barrier that controls diffusion of the analyte from the exterior of the electrode system to the enzyme molecules.

Thus, it is a further object of the present invention to provide an electrode system for measuring the concentration of an analyte under in-vivo conditions, comprising an electrode with immobilized enzyme molecules and preferably a diffusion barrier that controls diffusion of the analyte form the exterior of the electrode system to the enzyme molecules, characterised in that a spacer membrane forms at least a portion of the outer layer of the electrode system, wherein the spacer membrane comprises a hydrophilic copolymer of acrylic and/or methacrylic monomers, wherein the polymer comprises more than 50 mol-% hydrophilic monomers.

The features of this embodiment particularly with regard to the structure of the electrode system, the analyte and the enzyme molecules are as described herein. The diffusion barrier is preferably as described herein, it may however also have a different composition or may be absent. The preferred acrylic and methacrylic monomers of the spacer membrane copolymer are as described herein. The outer spacer membrane preferably covers at least the working electrode portion comprising the enzyme molecules and optionally also other portions, e.g. the counter electrode.

Example 1 Permeability of an enzymatic non-fluidic (ENR glucose sensor with distributed electrodes for transcutaneous implantation having a diffusion layer consisting of one single block copolymer.

The sensor was built on a prefabricated palladium strip conductor structure on a polyester substrate having a thickness of 250 μιτι. Working electrode (WE) and counterelectrode (CE) were arranged distributedly (as shown in Figs 1 -2).

The fields of the CE were overprinted with carbon paste, the rest of the strip conductor was insulated. The fields of the WE were overprinted with a mixture of cross-linked glucose oxidase (enzyme), conductive polymer paste and electric catalyst, here manganese(IV)-oxide (Technipur). The remaining paths of the strip conductor were again insulated. The reference electrode (RE) consists of Ag/AgCI paste. The electrodes cover about 1 cm of the sensor shaft.

The WE-fields were coated with a block copolymer diffusion layer consisting of a HEMA block and a BUMA block. The thickness of the layer is 7 pm.

Four sensor batches were produced, each provided with a specific block copolymer as diffusion layer (see list hereinbelow). All block copolymers were obtained from Polymer Source, Montreal and are listed in the following Table 1 .

The respective block copolymer was dissolved in organic solvent (25% concentration) and the sensors were coated therewith. After drying by means of belt driers (2 min, 30 to 50°C), the coated sensors were tested invito in glucose solutions of different concentrations. Of each sensor batch 10 sensors were measured as random sample. As a measure for the in-vitro sensitivity, the signal was calculated by the difference of the measured currents at 10 mM and 0 mM glucose concentration, which then was divided by 10 mM (cf. Example 4).

All sensors were operated at a polarisation voltage of 350 mV versus Ag/AgCI, the measured temperature was kept constant at 37°C. The sensors used for this measurement series did not comprise the spacer described in WO 2010/028708, which, however, did not make any difference in view of the tested signal level. Fig. 5 shows the sensor sensitivity with standard deviations for the four different diffusion layers.

Concerning block copolymers C, D and F, there is a clear connection between in-vitro sensitivity and molar ratio of hydrophobic block compared to hydrophilic block. At about identical total chain length of the copolymer, the sensitivity increases as the amount of hydrophilic block (HEMA) increases.

The sensors having a diffusion layer of block copolymer B are an exception. Even though polymer B has a relative ratio of hydrophobic to hydrophilic amount similar to polymer F, the sensitivity and thus the permeability for glucose is reduced. Empirically it can be stated that in case of polymer B the total chain length - corresponding to the molecular weight (total) of the copolymer molecule - is so large that the permeability of the layer is reduced. This may also be seen in the gravimetrically determined water uptake of block copolymer B as compared to the remaining polymers. Polymer B has a water uptake of 10.6%±1 .5% (weight percent referred to the polymer dry weight). Polymer C lies at 15.6%±0.0%, polymer F at 16.5±3.1 % and polymer D at 27%±1.7%.

Example 2 Mechanic flexibility of the diffusion layer of an ENF glucose sensor

The sensor was manufactured as described in WO2010/028708, however having a diffusion layer according to the present invention. It was assumed that the glass transition temperature (Tg) is a substitute parameter for the mechanic flexibility. Further, it was assumed that the glass transition temperature, which may be allocated to the hydrophobic block, determines the mechanic flexibility in in-vivo applications. It should be noted that several Tgs may be identified for one block copolymer, corresponding to the number of blocks. The sensors were coated with the same electrode pastes as in Example 1. Then, some of the sensors were coated with a copolymer selected from MMA-HEMA (produced by Polymer Source, Montreal). This polymer (called E) has a total molecular weight of 41 kD, the molar ratio of MMA (hydrophobic amount) to HEMA is 60%:40%. The glass transition temperature of the hydrophobic block is 1 1 1 °C, determined by DSC and a heating rate of 10°C/min.

Besides, other sensors were provided with a diffusion layer of a block copolymer of the invention (called A). The hydrophobic block of said copolymer A contains MMA and BUMA at equal molar amounts in a randomised sequence. Again, the molar ratio of the hydrophobic part to the hydrophilic part is 60%:40%. The molecular weight is 36 kD. The Tg of the hydrophobic block decreases, due to the randomized sequence of MMA and BUMA (Tg about 45°C), to 73°C.

Both diffusion layers were generated from the respective solution (25%) of the copolymers in ether and dried as in Example 1. The thickness of the diffusion layers was 7 μιτι. A spacer layer was applied subsequently via dip coating and dried 24 h at room temperature. The spacer layer was made of Lipidure CM 5206, produced by NOF Japan.

After explantation from the tissue, sensors having a copolymer E diffusion layer show sporadic cracks in the diffusion layer. This is taken as an effect of the mechanic load. In contrast thereto, sensors having a copolymer A diffusion layer, do not show any cracks under identical load. This is obviously due to the reduction of Tg, which increases the mechanic stability of the copolymer. A physical mixture of two copolymers, as disclosed in WO2010/028708, is no longer required.

Example 3 Optimized permeation behaviour of an ENF glucose sensor with distributed electrode and diffusion layer according to the invention. A sensor was manufactured as described in Example 1 , but with an additional spacer layer on the total of the sensor shaft. Sensors with respective diffusion layer were produced for copolymers A, C, D and F of Examples 1 and 2. For this purpose, a 24% etheric solution of the copolymer was generated. Each solution was applied onto a set of sensors (N=10) and then dried in a band drier. Thereby, diffusion layers having a thickness of 7μηη, were obtained.

Afterwards, the sensors were provided with a spacer layer as described in Example 2.

The sensor was connected with a measuring system on the sensor head, which transfers the measured data to a data store. The in-vitro measurements were carried out as in Example 1 , however over a measuring period of 7 days. From the measured data, the sensitivity drift was calculated over the respective measuring period for each sensor. Figure 6 shows for each sensor variant, i.e. sensors of a variant of the diffusion layer, the mean value of the in-vitro drift value for the group. The initial phase of the measurement - the first 6h, the so-called startup phase - was excluded from the calculation.

For all copolymers C, D and F having a hydrophobic block of BUMA, there is a positive drift, i.e. the sensitivity increases according to time. Contrary thereto, copolymer A with the hydrophobic block of a random copolymer of MMA and BUMA, leads to a very low, slightly negative, drift.

These differences may be explained by the long-time permeability response of the respective diffusion layers, which was measured in additional experiments. Palladium sensors without WE-paste, but with a defined active surface, i.e. also without an enzyme layer - excluding the influence of its swelling behaviour on the results - were coated with the above polymer solutions, and after drying, the thickness of the layer was measured. Subsequently, conductivity was measured in sodium- and chloride- containing buffer solution

Fig. 7 shows that the conductivity of copolymer A remained nearly constant after a short startup phase.

This is not the case for copolymer F, even under identical measurement conditions, as may be seen in Fig. 8. In this case, a long-term and strong permeability response of the diffusion layer of copolymer F was observed, which was practically independent of the layer thickness. For copolymer F - and also copolymers C and D (not shown) - with a hydrophobic block of BUMA, an increase of permeability results even over a long time period. When measured, this leads to a continuous increase of sensitivity if the diffusion layer is applied onto the sensor with distributed enzyme layer. This explains the observed positive sensor drift.

Vice versa, a sensor having block copolymer A, shows a negligible drift, which is due to a very low permeability alteration in the conductivity measurement. Directly after starting measurement (until about 1 h afterwards), however, a strong increase of conductivity is observed in copolymer A. Here, a very fast startup is observed, which is terminated after about 1 hour. At this time, the diffusion layer is completely wetted and has terminated its structural reorganisation due to water uptake. The extent of the structural change presumably depends on the Tg. It seems plausible that a copolymer having an increased Tg passes a reorganisation, which is limited in time and amplitude, as compared to a copolymer having a Tg in the range of the ambient temperature.

In addition, it has to be stated that sensors with copolymer A show a comparatively high sensitivity at the start of measurements as compared to sensors having a copolymer F diffusion layer. This is to be expected due to the identical relative ratios between hydrophobic and hydrophilic blocks. The achieved sensitivity range of 1 to 1 .5 nA/mM (see Example 1 ) is deemed ideal. This sensitivity is likewise obtained for sensors having a diffusion layer consisting of copolymer A.

Regarding the sum of the three physico-chemical characteristics - permeability, mechanic stability and permeability response - an optimal sensor may preferably be obtained with a diffusion layer of a block copolymer, having a hydrophobic block with at least two different randomly arranged hydrophobic monomeric units, such as block copolymer A. None of the other block copolymers, whose hydrophobic blocks only consist of a single monomeric unit reaches a quality, which could be compared in all three parameters with copolymer A.

Example 4 Characterization of block copolymers

A multiple field sensor (10 fields of working electrodes and counterelectrodes, respectively) for the continuous measurement of the glucose was produced and characterized in-vitro.

The sensor was provided with a diffusion layer consisting of a block copolymer comprising a hydrophobic block of random copolymerized methyl methacrylate (MMA) and n-butyl methacrylate (BUMA) and a hydrophilic block of 2-hydroxyethyl methacrylate (HEMA). These polymers (specified G and H) had been produced by Polymer Source, Montreal, and are more permeable than polymer A from Examples 1-3, which is included herein by reference.

In the following Table 2, the copolymers are described: Polymer G H A

Molecular weights 23.5-b-29 21-b-20.5 21-D-15

Mn [kD]

Weight-% HEM A 55.2 49.4 41 .6

Mol-% HEM A 53.5 47.4 40

(stoichiometrically)

Mol-% HEMA 51 46 32.6

(measured by

1H, ¹³ C NMR)

Tg [°C] hydropho65 68 86

bic block

HEMA monomeric 223 157 1 15

units

MMA monomeric 194 174 174

units

The molecular weights Mn of each block are separately indicated in the above Table 2 and represent average values, as polymers are known to have distributions of molecular chain lengths around a specified mean value. This also applies to the derived quantities in Table 2.

The indicated glass transition temperatures of the hydrophobic block are within the desired range in order to guarantee mechanical flexibility.

The decisive parameter with regard to the permeability of the diffusion barrier for the analyte is the sensitivity per area unit of the working electrode (i.e. the geometric area). The sensitivity SE was calculated from current (I) measurements at 10 mM and at 0 mM glucose concentration in phosphate- buffered solution (pH 7.4) in nA mM:

SE = [1(10 mM) - 1(0 mM)]/10 for each of the analyzed sensors. From the individual measurement values (N=8) the mean sensitivity SE _m was determined. The obtained sensitivity values were divided by the microscopically measured geometric total area F of all working electrode spots on the multi-field sensor. Thereby, a sensitivity density SE _m /F was obtained.

The linearity Y of the in-vitro function curve is an indication of the diffusion control functionality of the polymer cover layer on the working electrode. It was calculated from current measurements at 20 mM, 10 mM and 0 mM glucose concentration in %:

Y ^20mM = 50-[l(20mM) - I(0mM)]/[l(10mM) - l(0mM)] for each of the analyzed sensors. From the individual measurement values the mean linearity value and its standard deviation were determined (cf. Table 3).

Finally, the layer thickness L of the diffusion barrier of the sensors was determined by optical measurement for each of the polymers. The corresponding mean values were computed for a sample of > 23 sensors with the same polymer. Therefrom, the effective diffusion coefficient D _eff of the cover layer may be calculated: in cm ² /s, wherein SE _m and L _m are the respective mean values for the sensitivity and the layer thickness, and F is the total area of all working electrode spots.

The sensor drift was calculated from repetitions of the glucose concentration stages over 7 days of in-vitro measurements. The results for polymer H showing a substantially constant conductivity are depicted in Figure 9.

The following Table 3 shows the results of the functional characterization: Polymer G H

SEJF 1.85 1.25

[nA/mM*mm ² )]

Drift [%d] -1.5±0.2 0.3±0.1

Y ^20mM [%] 88.2±0.7 88.6±0.3

layer thickness L _m 1 1.61 12.69

[pm]

Deff [cm ² /s] 1.1 1305*10 ^"9 8.22019*10 ^{" 0}

For the more hydrophilic polymer G (which is more permeable for glucose) the diffusion coefficient was also determined with an alternative method, e.g. permeation of glucose from a chamber with a glucose solution into a chamber with a glucose-free buffer through a film of the polymer. According to this method, a similar value for the diffusion coefficient was obtained (1 .17-10- ⁹ cm ² /s).