DE102022000897A1 - Implantable biosensor - Google Patents

Abstract

The invention relates to an implantable biosensor and a method for automated continuous glucose measurement in concentration equilibrium, comprising a microelectrode and a biocatalytically active film localized thereon containing a redox polymer which adheres to the surface of the microelectrode by means of a crosslinker and an oxygen-insensitive enzyme as a catalytic center, as well the use of such a sensor for minimally invasive implantation in the field of endocrinology.

Description

The invention relates to an implantable biosensor and a method for automated continuous analyte measurement with minimal disruption of the concentration balance, comprising a microelectrode and a biocatalytically active film localized thereon, which adheres to the surface of the microelectrode by means of a crosslinker, the biocatalytic film being a redox polymer and a includes an oxygen-insensitive enzyme that has a catalytic center, as well as the use of such a sensor for minimally invasive implantation in the field of endocrinology.

Diabetes mellitus is a disease that is widespread worldwide and affects more than 420 million people worldwide, with an increasing incidence. It is based on a metabolic disorder in the regulation of blood sugar levels, which is due to an absolute or relative lack of insulin or insulin resistance. In order to be able to detect and treat high and low blood sugar levels in a timely manner, those affected depend on their blood sugar levels being monitored as closely as possible. Although glucose self-monitoring (GSM) is well accepted and represents a major improvement in the quality of life for diabetes patients, it is significantly limited by the usual number of tests that have to be carried out manually every day. Taking small amounts of blood by pricking the finger, for example, and measuring blood sugar levels with disposable test strips is neither possible at night nor does it provide reliable data about trends, peaks or patterns in blood sugar levels. The need for optimized sensor technologies for automated continuous glucose monitoring (CGM) is obviously great. This not only offers the patient the greatest possible comfort and an improvement in quality of life, but also secondary diseases of diabetes and their high treatment costs can be prevented or reduced at an early stage.

A large number of commercial CGM systems currently on the market have needle-shaped, minimally invasive sensors that are inserted into the arm or abdomen and measure glucose in the interstitial fluid (ISF). Due to a concentration- and time-dependent difference between blood glucose and ISF glucose, the sensor values obtained are corrected using specific algorithms. Almost all currently available CGM sensor systems are based on enzymatic glucose conversion and electrochemical detection. Enzymatic glucose sensors have very high selectivity due to the specificity of the enzyme for its substrate, which generally results in very high measurement accuracy. In addition, the general measuring principle of electrochemical, catalytic detection is a robust method of measuring substrates such as glucose in proportion to their concentration in the respective environment. There are products on the market that, after an initial manual calibration of the sensor using the finger stick method, are operated continuously and without further adjustment for up to 14 days. Newer generations of these sensors are even handed over to the patient calibrated at the factory. In contrast to enzymatic sensors, a fully implantable system based on an affinity interaction principle on modified hydrogels has so far been approved. Sensors used in this way can remain under the skin in a patient's upper arm for up to 90 days (recent studies also show a possible usability of up to 180 days), but must be calibrated manually every day in order to be able to use the results for any therapy decisions as is the case with approved enzymatic sensors without calibration.

A biosensor to be implanted for a long time primarily requires a high level of stability of all its components as well as a high level of accuracy, accompanied by a very simple and, if possible, infrequent manual calibration of the sensor. These requirements are influenced by the overall biocompatibility of the designed sensor. In addition, the human body has a complex and strong self-defense mechanism when a foreign body is introduced. As a result of the implantation process, the body tissue suffers initial damage, which triggers a host reaction, which, in addition to wound healing, includes, among other things, the adsorption of plasma proteins to the foreign body. Followed by a cellular reaction and attack by lymphocytes and macrophages, this irreversibly leads to fibrous encapsulation of the object. This foreign body response (FBR) persists for days to weeks and leads to a drastic change in the conditions in the immediate vicinity of the sensor. This leads to an increasing hindrance of substrate diffusion towards and - in the case of a catalytically converting sensor - product diffusion away from the actual sensor layer. The FBR therefore has a direct effect on the measured sensor signal. Furthermore, the partial prevention of product diffusion away from the sensor layer limits the overall stability of the sensor, as substrate conversion potentially produces by-products such as, in the case of glucose sensors, gluconic acid or reactive peroxide species, which lead to local pH changes or can lead to radical reactions. In order to counteract the process of fibrous encapsulation, coatings with a wide variety of materials (e.g. polymers decorated with functional groups) were investigated, with the result that the process could be partially slowed down. However, since encapsulation cannot be completely prevented over the time of implantation, it is of great importance to develop new sensor architectures and measurement strategies in which the sensor response is not or only slightly affected by changes in the surrounding tissue.

The CGM systems currently approved for use on the market use two different general measurement methods, enzyme-based catalytic sensors and those based on affinity interactions on hydrogels. Both types have different advantages and disadvantages for the (long-term) determination of glucose, which are briefly explained below.

For glucose sensors that are based on affinity interactions between substrate and sensor film, for example, hydrogels modified with boronic acids are used as a detection matrix. Glucose binds reversibly to the boronates and modulates the fluorescence of the surface matrix. Using an LED source, this modulation can be recorded via comparative measurements using band-filtered photodiodes and evaluated electronically in code.

The reversible binding with no change in the glucose molecule has the great advantage that the concentration inside the sensor matrix is in equilibrium with the external substrate concentration. Changes in the concentration in the area surrounding the sensor are compensated for by diffusion after a short time. Assuming that all components built into the sensor are generally sufficiently stable, the glucose-dependent sensor signal is only slightly influenced even after the sensor has been encapsulated for a long period of time. This means that such a sensor is in principle robust and can be used for reliable glucose determination for a long time.

However, the reversible type of glucose measurement can also have significant disadvantages. The glucose binding constant of the detection matrix must be precisely tailored to the prevailing conditions. If it is too strong, the substrate molecule only detaches very slowly from the detection matrix and changes in the glucose concentration in the sensor environment can only be detected very slowly by the sensor. However, if the binding constant is too weak, the accuracy of the detection is impaired. Furthermore, molecules that are structurally very similar to glucose, such as other sugar molecules or vicinal dialcohols, can also bind to the boronic acids because, unlike enzymes, they do not have any specific substrate recognition features. This means that the sensor can deliver distorted results, for example when certain medications are administered. Food additives such as mannitol or sorbitol, if they enter the blood, can also influence the measured sensor signal due to non-specific substrate recognition.

In contrast to the sensors described above, which are based purely on affinity interactions, catalytic sensors chemically convert the substrate molecules that diffuse into the sensor layer into products. In the case of electrochemical glucose sensors, this happens, for example, by oxidizing the glucose by the corresponding enzyme and then regenerating it with the help of an electrical potential applied to the sensor electrode. During the enzyme-catalytic conversion, the substrate binds precisely to the active site of an enzyme and forms an enzyme-substrate complex. From this, the reaction takes place with the formation of the product, which then leaves the binding pocket of the enzyme, since the binding strength to the substrate is significantly higher than to the product of the reaction. The enzyme molecule is therefore available unchanged for a further catalytic reaction. Through the chemical conversion and release of the product, the enzyme essentially cleans the sensor matrix itself and is therefore quickly ready for further conversion reactions. Changes in glucose concentrations in the area surrounding such a sensor can be recorded much more quickly than with sensors based on affinity interactions.

Another advantage of enzyme-based sensors over those using boronic acids is the enzymes' unsurpassed substrate specificity. By appropriately modifying the enzymes, the specificity and reactivity for certain substrates can be further adjusted and undesirable side reaction pathways, such as with oxygen dissolved in the blood, can also be prevented. Enzymatic sensors are therefore faster, more precise and more sensitive than most competing technologies. However, the advantage of catalytic conversion and the associated self-cleaning of the sensor matrix is also the most significant disadvantage of conventional electrochemical biosensors when they are to be used in implantable systems. Due to the fact that substrate is constantly being converted and thus “consumed” in the active sensor film, the system is not in the Balance with its surroundings. The large differences in concentration between the sensor and the environment create a steep diffusion gradient. The current response of an electrochemical sensor is limited, among other things, by the diffusion of the substrate to the electrode. If, in the case of continuous fibrous encapsulation, a growing diffusion barrier is formed around the sensor, the signal from the sensor can change significantly at the same external substrate concentration. This means that such a sensor would have to be constantly calibrated to the changing situation. Furthermore, the formation of a high diffusion barrier can result in a reduction in the long-term stability of the sensor, since the products and by-products of the enzymatic conversion accumulate in the sensor film and, for example, can cause local pH value changes.

Some of the previous sensor types have already been described in the prior art.

That's how it is in the EP 0778897 A1 described an implantable glucose biosensor in which enzymes immobilized on a redox polymer electrode are present, the signals of which are periodically recalibrated. Further analogous systems with immobilized enzymes are in the publications of EP 1230249 A1 and the EP 0958495 A1 called.

Furthermore, the shutdown of the enzyme activity in the EP 2478350 A2 , the EP 3303574 A1 , the US 2020072823 A1 , the DE 19814761 A1 and the EP 0958351 B1 described. However, especially in the EP 2478350 A2 revealed that glucose measurement depends on the rate of substrate diffusion into the enzyme film. The measurement therefore does not take place in “quasi” equilibrium with the environment. There is also no reversible inactivation of an enzyme, which is of a chemical and not electrical nature.

The dependence of the measurement accuracy of the biosensors on the diffusion rate is described in the publications EP 2401390 B1 , EP 1691679 B1 , as well as the EP 1119637 B2 disclosed. This is particularly the case in the EP 1119637 B2 pointed out the different diffusion coefficients in different media. The concentration of the analyte is determined using a coulometric method in which the charge is directly correlated with the amount of analyte in a defined volume. However, the time required for such a measurement to be carried out is relatively long and it is difficult to calculate defined volumes in implantable systems.

Finally, the article, Advances in organic transistor-based biosensors: From organic electrochemical transistors to electrolyte-gated organic field-effect transistors, September 2011 in Analytical and Bioanalytical Chemistry 402(5):1813-26, describes how a voltage-induced shutdown of a Enzyme is carried out by so-called “enzyme transistors”. There are several ways to prevent an enzyme immobilized on an electrode from converting its substrate. However, the measurement times are relatively long and it is not possible to keep the sensor in quasi-concentration equilibrium with its surroundings (and therefore independent of diffusion barriers).

The object of the invention is therefore to develop a catalytic biosensor for analyte measurements with minimal disruption of the concentration balance of the analyte in order to achieve the main advantages (sensitive, accurate, substrate diffusion-independent and rapid analyte measurements over a long period of time) of the technologies on the market to use and at the same time reduce the disadvantages of the state of the art to a minimum.

The object is achieved according to the invention by the features of independent claims 1 and 6.

The (re)adjustment of a substrate concentration balance between the sensor and the environment can only take place if the enzymatic reaction on the sensor can be stopped quickly, completely and reversibly. Only then will there be no permanent diffusion gradient between the compartments. This on and off process is achieved by choosing a suitable sensor architecture in which the enzyme ensemble is efficiently electrically connected to the electrode via a redox polymer.

On the one hand, the redox polymer takes on the function of a suitable, stable immobilization matrix for the enzymes and, on the other hand, it is responsible for the electron transfer between the enzyme and the electrode, which takes place via the mediators bound to the redox polymer. The spatial proximity between the mediator complexes and the enzymes in the immobilized film makes it possible to quickly control the processes that occur within the catalytically active layer. This means that the enzymatic conversion process can be switched between “on” and “off” states quickly and reliably.

This leads to two limit states of a biosensor modified with a redox enzyme, in which the enzyme is modified into an Os complex Redox polymer is embedded and is switched between conversion and non-conversion states by applying different potentials to the electrode. The enzymatic reaction can only continue if a sufficient overpotential is applied to the electrode, which enables the reoxidation of the electron-transferring Os complex. With an applied potential that is sufficiently more negative than the midpoint potential of the Os ²⁺ /Os ³⁺ pair, the mediator and ultimately also the enzyme itself is forced into a reduced state so that no (further) substrate turnover takes place; this prevents the establishment of a substrate gradient. In order to achieve a sufficiently small number of oxidized Os complexes in the film, for example an Os ²⁺ /Os ³⁺ ratio of 10000:1, the applied potential according to the Nernst equation must be 236 mV more negative than the potential of the complex used under standard conditions .

In order to reliably switch off the enzyme reaction, all side reactions that lead to reoxidation of the Os complex or the enzyme itself must be prevented, as this would in turn result in substrate consumption and thus post-diffusion of substrate into the film, which would hinder the establishment of an equilibrium concentration would. In a comparative study of eight redox polymers with different standard potentials, it was shown that molecular oxygen can be reduced by osmium complexes with a redox potential more negative than +0.07 V against Ag/AgCl/3 M KCl (pH 7) to form hydrogen peroxide, part of which was further reduced to water by the same Os complexes. The direct reaction of oxygen with Os complexes leads to two effects that must be suppressed: During glucose turnover, the generated current signal would be reduced because the reaction of oxygen with the metal complexes partially prevents efficient mediated electron transfer between enzyme and electrode. Furthermore, the presence of oxygen in the resting state leads to the oxidation of Os ²⁺ , which can then react with the surrounding enzymes and thereby enable undesirable substrate turnover. Therefore, the redox potential of the polymer used should be more positive than the aforementioned value, but at the same time low enough to avoid co-oxidation of interfering compounds, such as ascorbic acid or uric acid, which may be present in the analyte. In addition, the enzyme itself must be insensitive to oxygen. Since native glucose oxidase from Aspergillus Niger uses oxygen as a natural electron acceptor, it can only be used for such measurements if oxygen is reliably excluded. However, this is practically impossible to implement for use as an implant in the human body. Enzymes such as PQQ-GDH, CDH and FAD-GDH have low sensitivity to oxygen and are therefore suitable candidates. Furthermore, oxygen sensitivity and glucose affinity as well as general enzymatic stability can be modulated by genetic mutation in desired regions.

The actual analyte measurement, i.e. the time in which the sensor catalytically converts the substrate before it is switched back to the “off” state, must be extremely short in order to use as little substrate as possible and thus disturb the concentration balance as little as possible. Compared to macroscopic electrodes with radii of several millimeters and more, electrochemical probes in the micrometer and nanometer range offer several advantages due to their small size, such as lower capacitance, fast spherical diffusion and negligible ohmic loss effects. These properties are of great importance for electrochemical measurements on very short time scales. In addition to choosing electrodes with small dimensions, the redox hydrogel layer immobilized on the electrode surface must also be sufficiently thin, since electron transport in such films can only take place efficiently and without interference over short distances. Polymer films that are too thick and have inefficient electron transport can lead to inhomogeneity of charges, i.e. gradients between reduced and oxidized mediator complexes, within the film and would therefore impair or slow down the measurement.

The following description of an enzymatic glucose sensor shows the feasibility of the developed sensor concept. This is a combined measurement sequence that can be repeated in cycles and consists of a potentiostatic control of the sensor state between “on” and “off” and data acquisition during a short potentiometric analyte measurement. The universal measurement strategy presented here can also be used for substrates other than glucose (e.g. for determining glutamate or lactate) by replacing the redox-active enzymes immobilized in the hydrogel film.

All measurement results presented here were obtained with sensors based on carbon microelectrodes with a disc-shaped electrode surface. The production of carbon fiber microelectrodes with a diameter of 6-8 µm was carried out using an established process. For long-term dynamic measurement of the analyte, carbon microelectrodes with an enlarged diameter of 18-20 μm were used. The electrodes produced were independent Depending on their size, they were modified as follows: First, the carbon surface was covalently decorated electrochemically with 1,7-diaminoheptane (DAH) using cyclic voltammetry. For this purpose, 2 voltammetric cycles were carried out with a voltage feed rate of 25 mV/s in a potential range of 0.2 to 1.6 V vs. Ag/AgCl/3M KCl in a solution with 4.7 mM DAH in 0.1 M LiCLO ₄ /EtOH _abs . The electrodes were then freed from residues of the DAH solution alternately by dipping them twice in ethanol and distilled water. This process enables covalent bonds between the polymer backbone to be immobilized and the modified electrode surface and thus leads to increased stability of the hydrogel film on the electrode surface. The amino-modified electrodes were then immersed in an aqueous solution of the bifunctional diepoxy crosslinker poly(ethylene glycol) diglycidyl ether (PEGDGE) (5 mg/mL) for 10 min to form an epoxy-terminated surface on which the free amino groups of the redox polymer backbone (see structure). 1 , left - predominantly the N atom of the imidazole) can bind covalently.

Subsequently, the electrodes were coated alternately by immersion in redox polymer and a mixture of the same polymer and the oxygen-insensitive enzyme GcFAD-GDH, expressed in Pichia Pastoris, in a ratio of 1:1 (all solutions 2.5 mg/mL in water). For this purpose, the electrodes were immersed in the pure polymer solution for 20 s, then dried in air for 30 s with the modified tip pointing downwards and then immersed in the polymer/enzyme solution for 60 s and air-dried again for 30 s. Depending on the desired total film thickness, the immersion procedures were repeated several times, and the electrode was finally coated with a layer of pure polymer for 20 s. To further support the stability of the biocatalytically active films, the electrodes can optionally be covered with a further protective layer, for example consisting of the polymer P-(SS-BA-GMA) (structure 1 right).

The developed sensor concept is based on potentiostatic control of the potential in order to keep the sensor in substrate concentration equilibrium by switching off and preventing enzymatic glucose conversion. For this purpose, a sufficiently negative potential of -250 mV is applied compared to the redox potential of the Os ²⁺ /Os3 ⁺ _transition of the mediator used. This means that the charge balance is strongly on the side of the reduced Os ²⁺ complex, which cannot accept electrons from the reduced enzyme.

A short potentiostatic charging pulse leads to an excess of Os ³⁺ complexes in the hydrogel film by applying a potential that is 250 mV higher than the Os ²⁺ /Os ³⁺ transition. This enables the transfer of electrons from the enzyme to the mediators and thus the enzymatic conversion of the glucose, which is located within the film in the immediate vicinity of the enzymes. This pulse must last long enough to sufficiently charge the film. At the same time, the charging process must be short enough so as not to use up too much glucose in the film. Otherwise, the substrate would diffuse from outside the film into it and the undesirable diffusion gradient would form.

The actual measurement of the analyte and data acquisition takes place directly after the potentiostatic pulse, whereby the circuit to the working electrode is opened and the flow of electrons is thus prevented. A potentiometer with a high-resistance connected in parallel is used to measure a change in potential at the working electrode ( 2 ). The measurement sequence presented here can be carried out both analogue and on digital potentiostats with correspondingly fast data acquisition. Many modern (digital) potentiostats combine the possibility of potentiometric and potentiostatic control by “internally” connecting these components in one device. When the circuit is open, the measured change in potential (if a suitable enzyme/polymer combination is selected) can be attributed exclusively to the enzymatic glucose oxidation and the associated reduction of the previously externally charged Os ³⁺ complexes. This change occurs the faster the higher the glucose concentration in the film at the time of the pulse. Therefore, this type of measurement can be used as a reliable detection method of glucose concentration.

Following data acquisition, the circuit is closed again and at the same time the output potential of 250 mV negative from the Os ²⁺ /Os ³⁺ half-cell potential is applied to quickly and reversibly deactivate the enzymatic reaction and thus allow the substrate concentration equilibrium to be re-established. This measurement cycle shown in 3 can be repeated as often as desired and can therefore also be used for time-dependent observation of changing glucose concentrations.

The measurement concept developed was experimentally implemented and tested using previously manufactured microbiosensors of various sizes. The sensors were each placed in a measuring cell with a connected inflow and outflow control and in a three-electrode system with a Ag/AgCI/3M KCI reference and a platinum counter electrode connected to a digital potentiostat. PBS with a pH of 7.4 was used as the measuring electrolyte. The glucose concentrations to be examined were adjusted manually via the inlet and outlet openings on the measuring cell by introducing and removing glucose solutions of different concentrations in PBS. An isolated measurement involves potentiostatic charging of the redox polymer film for one second, followed by measurement data acquisition for 2.5 s, during which no potential was applied to the working electrode.

Between measurements, the sensor was switched to the “off” state over a period of 90 s by applying a potential of -50 mV ( 4 Left). The curves shown show a change in potential as a function of glucose concentration, which is in line with the model predictions described in detail previously. The potential drops sharply at the beginning of data acquisition and then weakens significantly to a quasi-linear course with a small slope, which is mainly determined by the diffusion of glucose to the sensor and into the hydrogel film. If there is no substrate in the measurement solution, the potential drop after potentiostatic charging of the polymer film is significantly lower, both at the beginning of the measurement curve and in the later part, than if a larger glucose concentration is set in the measurement solution, since the oxidation of glucose finally exceeds that before The electron transfer pathway described leads to a reduction in the Os ³⁺ complexes bound to the polymer. A comparison of the individual measurement curves for the same substrate concentrations shows practically congruence ( 4 right). The present results confirm both the successful implementation of the measurement concept described above and the high reliability of the sensors produced.

One of the central aspects for a “substrate-consuming” implantable sensor that functions reliably over a long period of time is its behavior when the diffusion barrier is increased. As described above, the sensor signal should, if possible, not be dependent on a fibrous protein capsule that continuously forms around the sensor compartment in the human body. In a control experiment, a sensor was modified with an additional thickened polymer layer made of P(SS-GMA-BA) so that the diffusion of glucose into the redox-active hydrogel film was severely restricted, similar to the FBR effect.

With this sensor, comparative measurements were carried out with short regeneration times (t) between the charging pulses (sensor switched off for 2 s 5a) and longer regeneration times (sensor switched off for 10 s or 30 s, 5b or. 5c ). In this example, after waiting only two seconds between the respective measurements, a strong depletion of the glucose concentration in the film was observed, which is manifested by the shift of the diffusion-dependent potentials towards higher values compared to the first measurement, in which the hydrogel film was in glucose concentration equilibrium with its surroundings was standing. With a waiting time of 30 s ( 5c ), the data curves obtained are again the same as could be shown for the sensor without an additional diffusion barrier. In this case, turning off the sensor for a longer period of time gives the sensor enough time to regenerate to achieve equilibrium. In practice, such a sensor could be operated with a measuring interval of approx. 30 s. With such a short measurement interval, for example, trends during rapidly changing glucose concentrations in the human body can be quickly identified and thus treated.

The overall goal was to establish a catalytic measurement of substrate (glucose) under quasi-equilibrium conditions between the sensor film and the solution surrounding the sensor. The formation of an extended linear range from a time of 0.5 s of data acquisition in the sensor shown above suggests that the sensor signal is mainly controlled by substrate diffusion to the sensor at this point in the measurement at the latest. This would not have created any advantage over an amperometric, diffusion-limited glucose measurement with the same sensor. This means that the time for data collection in our experiments still had to be significantly shortened. In order to be able to measure as far as possible without the influence of substrate diffusion into the redox polymer film, our approach deals with determining the different slopes of the measurement curves within the first maximum 75 ms directly after the charging pulse. To suppress noise, the respective measurement data were processed with a digital fast Fourier transformation (FFT) low-pass filter (cutoff frequency ~ 50 Hz) and the potential curve was plotted versus time compared to the initial value after potentiostatic charging ( 6 Left). The slopes of the linear measurement data curve in the range of 20-45 ms show clearly recognizable differences between a small drop in potential in the presence of no or little substrate, to a larger time-dependent drop in potential at higher glucose concentrations. In order to validate the direct dependence of the slope of the change in potential on the glucose concentration in the measurement solution, the sensors were tested with regard to three parameters (current, potential, slope of the potential trans senten) and the sensor response of these parameters is plotted against the set glucose concentration in a calibration curve ( 6 right). The values of the diffusion-controlled measurements (current and potential, read after a few seconds of measurement time) correlate very closely with the determined slopes of the potential curves at the beginning of the potential transients for the same sensor under the same conditions. This observation confirms our claim for the measurement concept developed to achieve the same quality of valid measurement data within an extremely short time of a few milliseconds for data acquisition as with standard measurement methods based on diffusion control. Due to the very short measurement data acquisition, the re-establishment of the concentration equilibrium between the sensor and the environment is significantly accelerated. This enables the measurement interval to be further drastically shortened.

After eating, the glucose level in a diabetic patient can quickly rise sharply or drop to very low levels after insulin is administered. A potential implantable glucose sensor should provide rapid and reliable results during these changes, helping to detect such trends early. In order to simulate and graphically display such glucose changes, a dynamic glucose profile was measured with a sensor over a period of approximately 4 hours and the magnitudes of the slopes for each measurement were measured with constantly changing glucose concentration ( 7 ) applied. Meanwhile, the glucose concentration was adjusted between 1 and 7.5 mM by manually opening the inlet and outlet valves in the measuring cell. The line drawn in the graph represents the running average of four pulsed measurements. After every 20 pulse measurements (each with 60 s regeneration time; 1 s charging pulse; determination of the linear slopes between 10 and 20 ms after the pulse), an additional amperometric measurement was carried out a potential of 450 mV compared to an Ag/AgCl/3 M KCl reference electrode for 120 s (■). As with the previous calibration curve, the correlated measured values of the amperometry also fit smoothly into the measurement sequence of the pulse measurements in the dynamic measurement. This showed that the measuring principle developed is a robust, reliable and extremely fast method, even for longer-lasting series of measurements under continuously changing conditions, and that the quality of the measured values obtained is in no way inferior to that of the standard methods.

The evaluation of the slopes of the potential transients at different glucose concentrations showed a very good correlation between these measurements and comparative controls with the same sensors. It was also shown that the manufactured sensors deliver the same measurement data quality as unmodified sensors after a short regeneration time, even with an artificially enlarged diffusion barrier (simulation of the FBR effect). The experimental results obtained made it possible for the first time to combine the advantages of the competing glucose sensor technologies currently dominating the market (affinity sensors and electrochemical sensors) and to minimize the previous disadvantages (slow adjustment of equilibrium, influence by interference, substrate diffusion limitation after fibrous encapsulation). to reduce.

• 1 zeigt chemische Strukturen der Redoxpolymere (4) P(VI-AH)-Os und P(SS-GMA-BA). Nominelle Zusammensetzung (in mol %) der funktionellen Gruppen im Polymerrückgrat für P(VI-AH)-Os: k = 79, I = 20, m-I < 1 und für P(SS-GMA-BA): k = 50, I = 30, m = 20.
• 2 zeigt den schematischen Messaufbau für ein analoges System bestehend aus einem separatem Potentiostaten und einem Potentiometer. Der Stromkreis wird extern über ein computergesteuertes digitales in-/out Interface zwischen Potentiometer und Potentiostat geschaltet.
• 3 offenbart die graphische Darstellung des entwickelten Messprinzips. Das An- bzw. Ausschalten des Sensors (1) erfolgt über das Anlegen eines um 250 mV positives oder negatives Potential gegenüber des E_1/2 des verwendeten Os-Komplexes am Redoxpolymer (4). Ein übergeordneter Stromkreis zur Mikroelektrode (2) des Sensors (1) ermöglicht das Schalten zwischen potentiostatischer Kontrolle und potentiometrischer Messung.
• 4 zeigt Links: Isolierte Potentialtransienten eines Sensors (1) unter sich ändernden Glukosekonzentrationen von 0 mM (PBS, pH 7.4), 1 mM, 3 mM, 5 mM und 7.5 mM (von oben nach unten). Rechts: Je gemessene Konzentration 0 mM bzw. 3 mM wurden 4 individuelle Potentialtransienten aufgetragen, welche sich nahezu vollständig überlagern.
• 5 zeigt vergleichende Messungen eines Sensors (1), der mit einer hohen Diffusionsbarriere für die Eindiffusion des Analyten (8), hier Glukose, in den aktiven Polymerfilm (3) ausgestattet wurde. Während kurzer Regenerationszeiten von 2 s zwischen den individuellen gepulsten Messungen des Analyten (1. - 4.) a) ist eine Verarmung des Analyten (8), hier Glukose im Film (3), sukzessiv zwischen der ersten Messung (1. - unterste Kurve mit stärkstem Potentialabfall) und der letzten Messung (4. - oberste Kurve mit geringstem Potentialabfall) erkennbar - durch eine schwächere Veränderung des Potentials während der Messung. b) Je länger die Regenerationszeit, desto geringer ist die Verarmung des Analyten (8) im Film (3). c) Ab einer Regenerationszeit von 30 s ist kein Unterschied zwischen den individuellen Messungen mehr erkennbar.
• 6 zeigt Links: Auftragung der Steigungen individueller Potentialtransienten gegenüber ihres Anfangspotentials bei steigender Glukosekonzentration (0 mM, 1 mM, 3 mM, 5 mM und 7.5 mM - von oben nach unten) mit Bestimmung der jeweiligen Geradensteigung im Bereich zwischen 20 und 45 ms. Rechts: Kalibrierkurve bestehend aus korrelierten Werten des Stroms (■), Potentials (Δ) und der Steigung der Potentialtransienten (•) für denselben Sensor (1) abhängig von der Glukosekonzentration.
• 7 beschreibt die dynamische Messreihe bei sich ständig ändernder Glukosekonzentration zwischen 1 und 7.5 mM. Die einzelnen Messpunkte (x) wurden aus der Steigung ihrer Potentialtransienten zwischen 10 und 20 ms nach dem Aufladepuls ermittelt. Die Quadrate (■) zeigen unabhängig durchgeführte amperometrische Messungen in Korrelation mit den Pulspotentialtransienten. Die eingezeichnete Spur wurde aus dem laufenden Durchschnitt von 4 individuellen Pulsmessungen ermittelt.

The invention is based on the following explanations for the 1 until 7 explained again in detail.

• 1 shows chemical structures of the redox polymers (4) P(VI-AH)-Os and P(SS-GMA-BA). Nominal composition (in mol%) of the functional groups in the polymer backbone for P(VI-AH)-Os: k = 79, I = 20, mI < 1 and for P(SS-GMA-BA): k = 50, I = 30, m = 20.
• 2 shows the schematic measurement setup for an analog system consisting of a separate potentiostat and a potentiometer. The circuit is switched externally via a computer-controlled digital in/out interface between potentiometer and potentiostat.
• 3 reveals the graphical representation of the developed measuring principle. The sensor (1) is switched on or off by applying a potential of 250 mV positive or negative compared to the E _1/2 of the Os complex used on the redox polymer (4). A higher-level circuit to the microelectrode (2) of the sensor (1) enables switching between potentiostatic control and potentiometric measurement.
• 4 shows Left: Isolated potential transients of a sensor (1) under changing glucose concentrations of 0 mM (PBS, pH 7.4), 1 mM, 3 mM, 5 mM and 7.5 mM (from top to bottom). Right: For each measured concentration of 0 mM or 3 mM, 4 individual potential transients were applied, which almost completely overlap each other.
• 5 shows comparative measurements of a sensor (1) that was equipped with a high diffusion barrier for the diffusion of the analyte (8), here glucose, into the active polymer film (3). During short regeneration times of 2 s between the individual pulsed measurements of the analyte (1st - 4th) a) a depletion of the analyte (8), here glucose in the film (3), can be seen successively between the first measurement (1st - lowest curve with the greatest drop in potential) and the last measurement (4th - top curve with the lowest drop in potential) - by a weaker change in potential during measurement. b) The longer the regeneration time, the lower the depletion of the analyte (8) in the film (3). c) From a regeneration time of 30 s, there is no longer any difference between the individual measurements.
• 6 shows left: Plot of the slopes of individual potential transients compared to their initial potential with increasing glucose concentration (0 mM, 1 mM, 3 mM, 5 mM and 7.5 mM - from top to bottom) with determination of the respective straight line slope in the range between 20 and 45 ms. Right: Calibration curve consisting of correlated values of the current (■), potential (Δ) and the slope of the potential transients (•) for the same sensor (1) depending on the glucose concentration.
• 7 describes the dynamic series of measurements with constantly changing glucose concentrations between 1 and 7.5 mM. The individual measuring points (x) were determined from the slope of their potential transients between 10 and 20 ms after the charging pulse. The squares (■) show independently performed amperometric measurements in correlation with the pulse potential transients. The trace drawn was determined from the running average of 4 individual pulse measurements.

List of reference symbols:

1

Biosensor

2

microelectrode

3

Biocalytic film

4

Redox polymer

5

Epoxy crosslinker

6

enzyme

7

Catalytic center

8th

Analyt

QUOTES INCLUDED IN THE DESCRIPTION

This list of documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• EP 0778897 A1 [0012]
• EP 1230249 A1 [0012]
• EP 0958495 A1 [0012]
• EP 2478350 A2 [0013]
• EP 3303574 A1 [0013]
• US 2020072823 A1 [0013]
• DE 19814761 A1 [0013]
• EP 0958351 B1 [0013]
• EP 2401390 B1 [0014]
• EP 1691679 B1 [0014]
• EP 1119637 B2 [0014]

Claims (10)

Implantable biosensor (1) for automated continuous analyte measurement in concentration equilibrium, comprising a microelectrode (2) and a biocatalytically active film (3) localized thereon, which adheres to the surface of the microelectrode (2) by means of a crosslinker (5), the biocatalytic Film (3) comprises a redox polymer (4) and an oxygen-insensitive enzyme (6) and has a catalytic center (7) and the reaction of the analyte (8) in the catalytic center of the enzyme (7) is reversible, characterized in that an electrical Potential that is more negative than the potential of the redox polymer (4) used under standard conditions is permanently applied to the biosensor (1) and is only interrupted by the analyte measurement, with the application of a potential in a time interval of up to 2s that is more positive than that Potential of the redox polymer (4) used is under standard conditions, triggers the analyte measurement and the change in potential in the open circuit enables the concentration of the analyte (8) to be determined on the biosensor (1). Biosensor (1) after Claim 1 , characterized in that the analyte (8) is selected from the group comprising glucose, lactate, cholesterol, ethanol, methanol, glutamate, lactose and their metabolites. Biosensor (1) after Claim 1 or 2 , characterized in that the microelectrode (2) comprises carbon fibers and has a diameter of 1 µm to 100 µm. Biosensor (1) according to one of the Claims 1 until 3 , characterized in that the crosslinker (5) is an epoxy crosslinker containing PEGDGE and the redox polymer (4) comprises an Os ²⁺ /Os ³⁺ complex. Biosensor (1) according to one of the Claims 1 until 4 , characterized in that the oxygen-insensitive enzyme (6) comprises GcFAD-GDH or PQQ-GDH or CDH. Method for continuous analyte measurement in implantable biosensors (1) in concentration equilibrium, comprising a microelectrode (2) and a biocatalytically active film (3) localized thereon, which adheres to the surface of the microelectrode (2) by means of a crosslinker (5), the biocatalytic Film (3) comprises a redox polymer (4) and an oxygen-insensitive enzyme (6) and has a catalytic center (7) and the reaction of the analyte (8) in the catalytic center of the enzyme (7) is reversible, characterized in that an electrical Potential that is more negative than the potential of the redox polymer (4) used under standard conditions is permanently applied to the biosensor (1) and is only interrupted by the analyte measurement, the analyte measurement being carried out in a time interval of up to 2 s by applying a potential, which is more positive than the potential of the redox polymer (4) used under standard conditions, is triggered and the concentration of the analyte (8) is then determined on the biosensor (1) by measuring the change in potential in the open circuit. Procedure according to Claim 6 , characterized in that the potential for triggering the analyte measurement is in the range of +250 mV more positive than the potential of the redox polymer (4) used under standard conditions. Procedure according to Claim 6 or 7 , characterized in that the disturbance in the concentration balance caused by the analyte measurement is compensated for within a period of 10 s to 300 s. Procedure according to one of the Claims 6 until 8th , characterized in that the time range from which the electrical potential is applied until the measurement of the analyte (8) begins is 0.5 s to 2 s. Use of a biosensor (1) according to one of the Claims 1 until 6 for minimally invasive implantation in the field of endocrinology.

Images