EP0245396A1

EP0245396A1 - Sensor for determining the concentration of a biochemical species

Info

Publication number: EP0245396A1
Application number: EP86906797A
Authority: EP
Inventors: Anders Weber
Original assignee: Radiometer AS
Current assignee: Radiometer AS
Priority date: 1985-11-18
Filing date: 1986-11-17
Publication date: 1987-11-19
Also published as: DK531185A; DK531185D0; WO1987003093A1

Abstract

Le capteur (1) ci-décrit permettant de déterminer la concentration d'une espèce biochimique fonctionne d'après un principe conpétitif et la partie active (2) du capteur donnant la réponse comprend les deux partenaires dans une paire d'affinité immobilisée sur deux parties d'un substrat. L'un de ces partenaires est constitué de l'espèce biochimique ou d'un compétiteur de celle-ci.The sensor (1) described above making it possible to determine the concentration of a biochemical species operates according to a conpetitive principle and the active part (2) of the sensor giving the response comprises the two partners in an immobilized affinity pair out of two parts of a substrate. One of these partners consists of the biochemical species or a competitor thereof.

Description

SENSOR FOR DETERMINING THE CONCENTRATION OF A BIOCHEMICAL SPECIES

The present invention relates to a sensor for determining the concentration of a biochemical species, which may be one of the constituents in an affinity pair, and with a response-giving active component reacting with the response capable of being converted into a sensor signal to the concentration of the respective biochemical species in an ambient medium, whereby a physical characteristic of the response-giving active component is modified as a function of the said concentration.

Sensors of this type - so-called immunosensors - have been described in a number of patents and articles and can be subdivided into two main groups:

The first group comprises sensors the action of which is based on the fact that with every determination of concentration carried out reagent is added, containing an analog of the species, the concentration of which is to be determined, marked with a fluorescent dye, an enzyme or a similar substance. This group of sensors, which cannot be used for continuous determinations of concentration, are in the present context not discussed in detail.

The second group comprises sensors which, without any reagent, react direct to the concentration of a biochemical species, hereinafter also referred to as an analyte, in an ambient medium.

By far the majority of the sensors pertaining to the hitherto known sensors of the latter type are sensors in which one of the constituents of an affinity pair, also referred to as a receptor, is immobilised on a surface. If such a sensor and the said surface is brought into contact with a solution containing an analyte, which may be the second constituent of the affinity pair, hereinafter also referred to as a ligand, a certain amount of the ligand is adsorbed to the immobilised first constituent, until a state of thermodynamic equilibrium has been reached.

By measuring the rate of adsorption or the adsorbed amount of ligand it is possible to determine the concentration of the latter in the ambient solution. The adsorbed amount of ligand can be determined according to a number of fundamentally different methods, including inter alia the following:

Measurement of adsorbed protein by electric impedance measurement or ellipsometry (Stenberg et al., 1979; J. Colloid and Interface Sci.

72(2)255).

Measurement of adsorbed protein on a semiconductor substrate using field effect type transistors (U.S. Patent 4.238.757).

Measurement of adsorbed protein according to the surface acoustic wave effect (Roederer & Bastiaans, 1983; Analytical Chemistry 55,

2333).

Measurement of adsorbed protein on light conductors, e.g. by utilising the evanescent wave effect (EP 75353 and Sutherland et al., 1984; Clin. Chem. 30/9, 1533). Measurement of adsorbed protein on the reverse side of a metal reflector utilising the surface plasmon resonance effect (Pettigrew, R.M. et al., 1984; PCT-application no. WO 84/02578; Comtech Research Unit Ltd.).

Measurement of adsorbed antibody by field variation owing to the effect of the K⁺- ion selectivity with a PYC membrane with hapteneconjugated dibenzo-18-crown-6 (Connel, G.R. & Sanders, K.M., 1983; Biophysical Journal 44(1983)).

The surface sensors are all subject to the following fundamental limitations: (1) The surface sensors are in practice irreversible and are therefore unsuitable for continuous determination of concentration. It has therefore always been necessary to regenerate the surfaces for every determination.

It is assumed that the lack of reversibility is due to the fact that the diffusion of adsorbed ligand from the surface to the ambient fluid is impeded since the immobilised constituent possesses unsaturated binding sites.

(2) The range of dynamic response is determined only by the affinity between the two constituents of the affinity pair. This is explained in greater detail below in connection with the sensor according to the USA patent no. 4,344,438 (Schultz).

In an attempt to circumvent these limitations of the surface sensors and to produce a sensor suitable for continuous determination of the concentration of a biochemical species, such as in vivo monitoring or monitoring of production processes, it is proposed to use, by way of a first solution, of surface adsorption of affinity pairs of low affinity, so that the dwell time of a ligand on a binding site becomes small, even at high ligand concentrations. In this case it is necessary to be able to measure the binding event itself instead of the bound mass, see e.g. North, J., 1985; Trends in Biotechnology 3 (7) 180. However, so far no method has been proposed according to which such a measurement can be carried out.

By way of another solution it is proposed to instal a competititve principle in the sensor itself. A competitive immunosensor is described by Hirschfeld in the specification of PCT-application no. WO 84/00817. The Hirschfeld sensor is built about an optical fibre to which one of the constituents is immobilised. The sensor comprises in addition a certain amount of the other constituent marked with fluorophore and bound to the first constituent by affinity bonds.

An electrochemical competitive immunosensor is described by Hill, H.A.O. in Analytical Proceedings 22 (1985) 201. With the Hill sensor the analyte-analogous constituent (ligand) is linked to a mediator molecule, and the electrode current is a measure of the analyte concentration.

A competitive sensor is also described by Schultz et al., 1982 in Diabetes Care 5.(3)245 and in the specification to the USA patent no. 4, 344, 438. In the detailed part of the specification to USA patent no. 4.344,438 it is stated that the basic idea of the sensor therein described is for a specific receptor site to be enclosed in a restricted chamber together with a so-called competing ligand capable of being bound reversibly to the specific receptor site. The competing ligand competes with the analyte for the said receptor sites . Installing a competitive principle in the sensor has the advantage that it enables a certain modulation of the sensor's response level and sensitivity, whereas with the above surface sensors or direct binding sensors such modulation is not possible. This becomes understandable if the binding mechanisms of the two types of sensor are considered in greater detail.

With a direct binding sensor the response occurs when receptor R and ligand L are bound; the binding process is

R + L RL

With equilibrium we have:

K=

Since R = [R] + [RL] L = [L] + [RL]

we obtain the fallowing equation for the relative response:

( 1 )

This equation shows that with a given ligand concentration the relative response depends solely on the magnitude of the equilibrium constant K.

Equation (1) also shows that a given relative response for a given analyte concentration can only be achieved by making use of a receptor the affinity of which for the analyte can be described by the K value calculated when solving equation (1).

In practice the concentration of the analyte when determining the concentration with the aid of a direct binding sensor must therefore generally be matched to the sensor by appropriate dilution. In a competitive sensor two equilibrium reactions occur:

R + L RL R + M RM where R and L again designate the receptor and the analyte ligand respectively, and M designates the competing ligand.

If we assume that the equilibrium constant is the same for the two equilibrium reactions we obtain

K

By using the following terms:

R: Total concentration of receptor sites (number of receptor sites/sensor volume); Total concentration of competing ligand; K': 1/K;

[L] : Concentration of free analyte ligand; [M] : Concentration of free competing ligand; we can deduce the following equation in respect of the response:

(2) M is an expression of the response magnitude.

The connection between M/K' and L/K' with different values of R/K' and M/K' is detected in fig. 1.

It can be seen from fig. 1 that if the concentration of competing ligand is increased whereas the concentration of receptor sites is maintained (II - III), the absolute response becomes higher and the sensitivity (change of response/change of analyte ligand concentration) greater. If the number of receptor sites is increased whereas the concentration of competing ligand is maintained (i - II), the absolute response will be lower and the sensitivity smaller. It follows that the response characteristic for the active response-giving part of a competitive sensor can be modulated within certain limits depending on the concentration of receptor sites, equilibrium constants for the actual affinity pair and the total concentration of the competing ligand. In the competitive sensor described by Schultz the active part is separated from the ambient medium by a dialysis membrane. Within the chamber there are both constituents of an affinity pair. One of them is immobilised on a restricted area, possibly a surface (receptor sites), whereas the other (the ligand) is enclosed within the active part in dissolved form. The ligand is marked with fluorochrome. AJCL analyte, which on the one hand can pass through the membrane and on the other hand can compete for receptor sites, will be able to affect the ratio between bound and free ligand within the chamber. A fluorometer with a localised field of vision can be used to measure either bound or free ligand marked with fluorochrome. A disadvantage of this embodiment consists in the fact that the response time of the sensor is relatively long. This is due to the fact that the field of view of the fluorometer must have a certain extent to achieve a given sensitivity. Since the entire area within the field of view shall reach thermodynamic equilibrium before the sensor is in equilibrium the diffusion ways of analyte and competing ligand become relatively long. Since at the same time the diffusion coefficient of the competing ligand is low, owing to the relatively high molecular weight of the latter, see below, the response time of the sensor is bound to be relatively long. It is also a disadvantage that the separation between the active part and the ambient medium is brought about by a dialysis barrier. Since the marked ligand is to be effectively retained by the membrane with a view to long time stability, the marked ligand must have a certain size to be retained in the sensor, whereas the substances capable of analysis must be considerably smaller so as to enable them to diffuse into the chamber. Accordingly, Schultz also describes only the use of the sensor for the determination of low-molecular compounds. Lastly, the use of fluorochrome compounds in a sensor as used with long time measuring processes is less practical inasmuch as these compounds dissociate under prolonged action of light. As a result of this instability the sensor response can impair the measuring process.

It is the object of the invention to produce an immunosensor of the type mentioned initially, which is well suited for continuous monitoring of the concentration of a biochemical species in a medium and is not subject to the disadvantages associated with the sensors hitherto known.

This is achieved by the sensor in accordance with the invention which is characterised in that the response-giving active part comprises a substrate part to which a certain amount of a substance, which may be the other constituent in the affinity pair, is immobilised and a substrate part to which a certain amount of the biochemical species or a competitor thereof is immobilised, whereby the substrate parts are linked by affinity bonds between the two immobilised constituents, with a degree of linkage which changes reversibly as a function of the concentration of the biochemical species in the medium, in that the response reflects a change in the degree of linkage between the substrate parts or a physical change resulting from such a change in the active part of the sensor, and in that the sensor comprises elements for converting the response from the response-giving active part into a sensor signal. The substrate parts consist of an appropriate polymer or other macromolecular substance, cf. the examples below.

According to the invention a sensor is thus proposed, which acts in accordance with a competitive principle and, as a result, possesses the advantages associated therewith, i.e. special reversibility and an improved facility for adapting the sensor to the actual concentration range , but which at the same time differs from the Schultz sensor in several very advantageous ways.

With the sensor according to the invention it is for instance possible to process responses according to other principles than optical fluorescence detection. When used in connection with low-molecular analytes it will be possible to reduce the response time by comparison with the Schultz sensor. This is due to the fact that with the sensor according to the invention the diffusion necessary to achieve equilibrium is limited to analyte diffusion. With the Schultz sensor, as mentioned above, diffusion takes place of the high-molecular competitive ligand, the diffusion coefficient of which is far lower than that of the analyte, and this results in slower attainment of the equilibrium response.

Lastly, there is no dialysis membrane in the sensor according to the invention, due to the fact that the competitor of the analyte is immobilised by bonding to a substrate instead of being isolated from the ambient medium with the aid of a dialysis membrane. As mentioned above, the use of a non-immobilised competitor of the analyte means that the sensor must comprise a dialysis membrane so that the sensor can only be used in connection with relatively lowmolecular analytes.

By comparison with the other known competitive immunosensors, all of which like the Schultz sensor -are based on binding of a molecule, the inherent characteristics of which are utilised, to the analyte analog, the sensor according to the invention is based on binding the analyte analog to a macromolecular substrate. This enables the use of other detection principles, e.g. detection of light attenuation or detection of forces or pressures.

The biochemical species, the concentration of which is determined by means of the sensor according to the invention, may be a constituent of a body fluid such as blood, plasma, serum, spinal fluid, saliva, urine etc.. Representative examples of such constituents are polypeptides; proteins, including conjugated proteins, enzymes and hormons; antigenic polysaccharides of microbial origin; and antigenic low-molecular substances such as body fluids, agents for the treatment of plants, small peptides, aminoacids, low-molecular hormons and metabolites of these substances.

The biochemical species can also consist of a substance derived from plant or animal tissue cultures or by fermentation, e.g. a monoclonal antibody, or it can be a microorganism or a virus.

The other constituent of the affinity pair may be any substance whatsoever capable of taking part in a specific binding reaction with the biochemical species by recognising and linking up with a closely defined specific binding site or determinant in the biochemical species while giving rise to affinity binding. In the present context the term "a competitor of the biochemical species" signifies a substance competing with the biochemical species and the same binding site on the other constituent of the affinity pair, thus constituting in the same way as the biochemical species a partner for this other constituent. Such a competitor may e.g. be a fragment of the biochemical species comprising the same specific binding site as the biochemical species.

Frequently used synonyms of the expression "an affinity pair" are "a ligand-antiligand pair", "a ligand-receptor pair" and "a specific binding pair". Also the expressions "binding site", "receptor site", "binding location", "specific binding site", "determinant" and "epitop site" are used synonymously in the majority of contexts.

Examples of affinity pairs are: antigen-antibody; hapteneantibody; immunoglobulin- protein A; carbohydrate-lectin; biotinavidin; hormon-hormon recptor protein and complementary nucleotide chains.

In this connection reference is also made to EP 46004 (Syva), which contains a more comprehensive survey of analytically interesting biochemical species capable of being constituents of an affinity pair. With the preferred embodiment of the sensor according to the invention the two substrate parts have two directly adjacent surfaces to which the two immobilised constituents are immobilised. Immobilisation of the two constituents can be brought about by any suitable immobilisation technique whatsoever.

In order to achieve an adequate degree of linkage between the two substrate parts it is regarded as advantageous for each of the two surfaces to consist of a material with a large specific surface area, e.g. a polymer network such as a polyamide network. The degree of linkage between the two surfaces can be appropriately determined by mechanically measuring the adhesion force between the two surfaces.

With another preferred embodiment of the sensor according to the invention one of the substrate parts consists of an optical wave conductor, e.g. an optical fibre or a capillary tube made of a suitable material, the other substrate part consisting of a number of small particles, e.g. latex spheres. Once an analyte or an analyte analog is immobilised on the particles and the other partner is immobilised on the optical wave conductor, the degree of linkage between the optical wave conductor and the particles will depend on the analyte concentration in an ambient medium. The attenuation of light transmitted through the optical wave conductor will depend on the surface tightness of the particles on the wave conductor and, accordingly, on the degree of linkage between the two substrates. With a further preferred embodiment of the sensor according to the invention the substrate parts, to which the two partners are immobilised, consist of polymer chains of a type which, given an appropriate degree of cross-linkage, can form a gel, and between the polymer chains there is a sufficient number of cross-linkages to produce gel, these linkages consisting wholly or partly of affinity bonds between the two immobilised partners.

The cross-linkages in the gel can thus consist of a combination of cross-linkages in the form of affinity bonds between the immobilised constituents as well as other cross-linkages, e.g. of covalent type.

The polymer chains may e.g. be such as occur in synthetic or natural hydro-gels, and immobilisation of the two constituents to the polymer chains can be brought about by means of covalent binding via e . g. glutaraldehyde or an oxirane. If a sensor of this type, where the polymer gel in the active response-giving part of the sensor comprises cross-linkages in the form of affinity bonds, is applied in a solution containing the biochemical species, the concentration of which is to be determined with the aid of the sensor, the number of affinity bonds in the polymer gel will decline owing to competitive inhibition.

Changes in the degree of cross-linkage will affect a large number of physical characteristics of the gel, e.g. the mechanical characteristics, the dielectric characteristics, the conductivity and the volume of the gel, and in principle there is a number of ways in which the response can be detected and thereafter converted to a sensor signal.

Changes of the mechanical characteristics of a gel lead to changes in ultrasound attenuation and it will be possible to measure them with the aid of piezoelectric transmitters/receivers. Changes of the degree of cross-linkage are reflected also by a change of elasticity module. In the literature a considerable number of methods are described for measuring the elasticity module of solid substances. It is possible, for instance, to make use of an aluminium tube embedded in the gel. The tube is caused to oscillate electromagnetically, and one measures changes in the electrical impedance of the oscillator coil (Roscoe, 19^9; Eheol. Acta 8, 195).

Changes of the degree of cross-linkage will also affect the volume of the gel. A change in volume can be transformed into a mechanical application of force, suitable for direct measurement with the aid of a pressure transducer.

Furthermore, a change in volume owing to a change in the gel's dry substance content per unit of volume will bring about a change in refraction index. Such a change can be detected by interferometry, ellipsometry or plasmon resonance. It may be assumed that a gel comprising affinity bonds can be produced in several ways.

A first method of production involves the preparation of a first solution of polymer chains, the introduction of reactive groups originating from, for instance, glutaraldehyde or an oxirane on the polymer chains and immobilisation of one of the constituents by linkage to the reactive groups on the polymer chains. Correspondingly, another solution of polymer chains is produced, to which the other constituent is immobilised.

In the presence of kaotropic ions or extremely high or low pH-values, the affinity between the two types of polymer chain is eliminated, and it is then possible to mix the two solutions. Thereafter covalent cross-linkages are brought about with the aid of a bifunctional reagent, whereupon neutral buffer of low ion strength is dialysed in and affinity bonds are established between the two constituents. With another method of production one of the constituents if first linked with monomer molecules or low-molecular polymer chains , whereupon polymerisation to more highly molecular, gel-forming polymer chains is carried out. Again it is possible to produce, in a corresponding manner, polymer chains to which the other constituent is immobilised. Linkage of the polymer chains in the course of gel formation is then brought about as described above.

The invention is now discussed in details with reference to the drawing, where fig. 1 is a sketch of the response characteristics of a sensor with integral competitive principle; and fig. 2 is a sketch of an embodiment of the sensor according to the invention.

Fig. 3 is a sketch of a measuring rig, including a different embodiment of the sensor according to the invention.

Fig. 1 is discussed in the preamble of the present specification.

Fig. 2 is a schematic drawing of the above mentioned embodiment of the invention, where the active response-giving part of the sensor contains a cross-linked gel comprising affinity bonds between the two immobilised constituents.

The sensor bearing the general designation 1 is immersed in a medium 3 containing the biochemical species, the concentration of which is to be determined with the aid of the sensor. In a chamber within the response-giving active part there is a gel 2 which is in contact with medium 5 through a perforated plate 1. Gel 2 comprises affinity bonds, the concentration of which and, accordingly, the gel's degree of linkage, is acted upon by the analyte concentration. A transducer element 4, possibly a pressure transducer, detects changes in the degree of linkage and converts them to a sensor signal, which through a lead 5 connected with a sensor cable 6 is conducted away and processed within a measuring instrument not illustrated.

Fig. 3 shows in schematic form a sensor 1 according to the invention in the form of an optical fibre treated with the one half of an affinity pair and with a coating of particles not illustrated, to which the other half of the affinity pair is immobilised. The particles are held to the fibre surface by affinity bonds. The coated fibre and the particles are produced as described in the following example. The fibre is mounted centrally within a measuring chamber 2, which constitutes a borehole within a perspex block 3. A supply channel 4 leads to one end of the measuring chamber, the other end of the measuring chamber being provided with an outlet 5 from which issues an outlet channel 6.

A light-emitting diode 7 is connected to one end of the fibre and constitutes a source of light, and a light-sensitive diode 11 is connected to the other end of the fibre and constitutes a light detector. The light-emitting diode 7 is an IR light-emitting diode (SFH 450 from Siemens, West Germany), which emits light with a wavelength of 950 nm, and is supplied with current with a frequency of 1.4 kHz by a square-wave generator 8. The light-sensitive diode 11 is a PIN diode with a high zero slope. The flow of light therefrom is lock-in detected in an averaging circuit with an averaging time constant of 1 second. The detector circuit, in its entirety designated 9, consists of an electrometer amplifier with a bandwidth of 10 kHz. The signal therefrom is conducted via a transmission capacitor alternately to earth (when the light-emitting diode 7 is extinguished) and to an averaging operation integrator (when the light-emitting diode 7 is alight). The alternation is brought about by an analog multiplexer controlled by the supply voltage of the light-emitting diode. From the integrator the signal is conducted to a logarithmic amplifier, and the signal therefrom is transmitted to the computation and reading system 10.

When the measuring chamber 2 is filled with solutions having various contents of analyte competing with the half of the affinity pair immobilised to the particles, it is seen that the light attenuation depends on the analyte concentration. This is due to the fact that the exchange between the analyte and the competitor bound to the particles - and accordingly the degree of linkage between the fibre and the particles - depends on the concentration of analyte in the ambient medium. EXAMPLE 1 PEODUCTION OF A SENSOR ACCORDING TO THE INVENTION

1 A Production of an evanescent wave light conductor

The fibre of the evanescent wave light conductor is produced from a 600 urn thick commercial quartz light conductor with polymer cladding (HCP-606; 600/630 Ensign-Bickford Optics Co., Cincinnati, USA). The cladding is removed with a sharp knife and the remaining binding layer is removed by burning off in a gas flame. Then the two end faces are ground flat. After cleaning with 3 ml ammonia (25%) and 3 ml H₂O₂(30%) in 15 ml distilled water at 80° for 5 minutes the fibre is mounted in a rotating holder within a vacuum chamber, exposing 2 centimetres at each fibre end. A thin layer of gold is applied to the two fibre ends by vacuum metallisation. The end faces are protected by rubber caps subject to vapour metallisation.

1B Preparation

Silanisation

A number of fibres are mounted in a glassholder which supports the fibre ends and leaves the 6 cm at the centre free. The glassholder with the fibres is treated with 3 nil cone. HCl and 3 ml H₂O₂ in 18 ml distilled water at 80° for 5 minutes.

The holder with the fibres is cleaned in distilled water and transferred to dry toluene via dry acetone. The holder is suspended in a reaction vessel with mechanical agitator and reflux cooler. The ambient air is displaced by means of dry nitrogen, and 8 ml 100% 2-glycidcxy-trimethoxysilane is added. Then 160 μl triethylamine is added and the mixture boiled for 3 hours. The holder is transferred to distilled water via relatively warm toluene, acetone and diethylether.

Activation The silanised quartz is now activated by treatment for 30 minutes at 80° using HCl in water at pH 1.5. Then it is transferred to dry acetone and treated with 0.1 ml tresyl chloride and 0.2 ml pyridine in 20 ml dry acetone. The treatment is effected on ice for 10 - 20 minutes stirring slowly. The quartz is now washed in a 1:1 mixture of acetone and 1 mM HCl and then in 1 mM HCl alone and finally in distilled water.

Immobilisation

The activated quartz fibres are now brought into contact with an antibody, e.g. a monoclonal antibody orientated towards mono-component porcine insulin. The concentration is about 1 mg/ml in a 0.2 M phosphate buffer with a pH-value of 7.5. The reaction proceeds throughout the night at 10°. It is followed by washing with 0.5 M

NaCl and treatment with 1 M ethanolamine adjusted to pH 8.0, for one hour at 5°.

1 C Production of antigen-coated polymer particles

Spherical polymer particles with a diameter of 2.5 μm (acrylic with hydroxylic groups, Dynospheres^(R) XP6501 from Dyno Particles A/S,

Lillestrom, Norway) are washed in distilled water subject to repeated centrifugings and resuspensions. 1 ml epichlorohydrin and 3 ml NaOH are added to about 500 mg particles in a volume of 5 ml. The treatment is carried out throughout the night within a rotary mixer. Then the particles are washed in water until the supernatant after centrifuging is neutral. Linkage of antigen in the form of insulin to the particles is effected in 0.2 M carbonate buffer at pH 9. In 1 ml buffer 100 mg activated particles are suspended and 100 μg insulin in 100 μl 0.01 M HCl is added. The linkage reaction takes place throughout the night in a rotary mixer. Then the particles are washed in linkage buffer and suspended in 5 ml 0.1 M ethanolamine at pH 8. Deactivation in ethanolamine proceeds for three hours. The particles are washed in PBS (phosphate buffer salt water and 0.01 M Na₂HPO₄/NaH₂PO₄, pH 7.5, 140 mM NaCl) with 0.3 M NaCl alternating with linkage buffer. 1D Linkage between antibody-treated light conductor and antigen-treated particles

Use is made of the rig shown in fig. 3 . Chamber 2 is filled with a suspension of 100 μg particles/ml. In the course of time the absorbance increases towards a constant value A_o. Once the absorbance has stabilised, the chamber is emptied for suspension and briefly rinsed with PBS.

The fibre 1 is now converted into a competitive sensor according to the invention.

Claims

PATENT CLAIMS

1. Sensor for determining the concentration of a biochemical species constituting one of the constituents of an affinity pair and with a response-giving active part responding to the concentration of the respective biochemical species in an ambient medium with a response capable of being converted to a sensor signal, whereby the physical characteristics of the response-giving active part change as a function of the said concentration c h ar a c t e r i s e d in that the response-giving active part comprises a substrate part to which a certain amount of substance constituting the other constituent of the affinity pair is immobilised and a substrate part to which a certain amount of the biochemical species or a competitor thereof is immobilised, the substrate parts being linked together by affinity bonds between the two immobilised constituents with a degree of linkage which changes reversibly as a function of the concentration of the biochemical species and the medium, in that the response reflects a change in the degree of linkage between the substrate parts or a physical change in the active part of the sensor derived from such a change, and in that the sensor ccmprises elements for converting the response from the response-giving active part to a sensor signal.

2. Sensor in accordance with claim 1 c h a r a c t e r i s e d in that the two substrate parts have two directly adjacent surfaces to which the two immobilised partners are immobilised.

3- Sensor in accordance with claims 1-2 c h a r a c t e r i s e d in that each of the two surfaces consists of a material with a large specific surface area.

4. Sensor in accordance with claims 1-3 c h a r a c t e r i s e d in that the material with large specific surface area is a polymeric network, e.g. a polyamide network.

5. Sensor in accordance with claims 2-4 c h a r a c t e r i s e d in that the degree of linkage between the two substrate parts is determined by determining the adhesion force between the two surface parts.

6. Sensor in accordance with claim 1 c h a r a c t e r i s e d in that the substrate parts to which the two immobilised constituents are immobilised are polymer chains of a type which given a suitable degree of cross-linkage can form a gel and that a sufficient number of cross-linkages for the formation of gel is brought about between the polymer chains, the said cross-linkages consisting wholly or partly of affinity bonds between the two immobilised constituents.

7. Sensor in accordance with claim 1 c h a r a c t e r i s e d in that one of the substrate parts is an optical wave conductor, the other substrate part consisting of a number of particles.

8. Sensor in accordance with claim 7 c h a r a c t e r i s e d in that the optical wave conductor is an optical fibre.

9. Sensor in accordance with claim 7 c h a r a c t e r i s e d in that the optical wave conductor is a capillary tube.