Electronic nose or tongue sensors

Abstract

The present invention relates to a sensor for an electronic tongue or nose for analysing a sample or detecting a target. The sensor comprises a support, on one surface of which a plurality of sensitive areas are located, each sensitive area comprising at least one receptor and being capable of transmitting a measurable signal generated by the interaction of at least one constituent of the sample or one target with at least one receptor. The sensor is characterised in that it comprises at least three sensitive areas that differ from one another in terms of their respective receptor compositions, at least one of the sensitive areas comprising a mixture of at least two different receptors, while the two other sensitive areas each comprise at least one of the two receptors.

Claims

1. An electronic tongue or nose sensor for analyzing a sample or detecting at least one target, comprising a support on one surface of which contains at least three sensitive zones, each comprising at least one receptor, each sensitive zone being configured to emit a measurable signal generated by the interaction of at least one constituent of the sample or of at least one target with at least one receptor; the electronic tongue or nose sensor is characterized in that said at least three sensitive zones are different from one another by virtue of their respective receptor compositions, with at least one of the at least three sensitive zones comprising a mixture of at least two different receptors, the other two sensitive zones each comprising at least one of the at least two different receptors, wherein the electronic tongue or nose sensor has k number of sensitive zones having said at least one of the at least two different receptors, and the percentage proportion of said at least one of the at least two different receptors in the sensitive zones change in increments of 100/(k-1); and said at least three sensitive zones are different from one another in terms of the respective proportions of said at least two different receptors that said at least three sensitive zones contain.

2. The electronic tongue or nose sensor as claimed in claim 1, comprising a plurality of sensitive zones which each comprise a mixture of at least two different receptors, the different receptors being the same in each mixture.

3. The electronic tongue or nose sensor as claimed in claim 1, wherein at least one of the sensitive zones comprises a mixture of at least three different receptors.

4. The electronic tongue or nose sensor as claimed in claim 1, comprising a plurality of sensitive zones which each comprise a mixture of at least three different receptors, the different receptors being the same in each mixture, wherein said sensitive zones differ from one another in terms of the respective proportions of the different receptors that they comprise.

5. The electronic tongue or nose sensor as claimed in claim 1, comprising at least as many sensitive zones as a number of types of receptors in the sensor, each of the types of receptors in the sensor being respectively included in each of the sensitive zones.

6. The electronic tongue or nose sensor as claimed in claim 1, comprising at least one signal for each of the sensitive zones, for forming profile of the signals emitted by all the zones to identify the presence of at least one defective sensitive zone.

7. The electronic tongue or nose sensor as claimed in claim 1, wherein at least one of the other two sensitive zones comprises only one of the at least two different receptors.

8. The electronic tongue or nose sensor as claimed in claim 1, wherein said at least three sensitive zones have said at least one of the at least two different receptors at percentage proportions with a constant increment.

9. The electronic tongue or nose sensor as claimed in claim 1, wherein the at least two different receptors are not nucleotides.

10. The electronic tongue or nose sensor as claimed in claim 1, wherein the at least two different receptors are proteins.

11. An electronic tongue or nose sensor for analyzing a sample or detecting at least one target, comprising a support on one surface of which contains at least three sensitive zones and less than 10,000 sensitive zones, each comprising at least one receptor, each sensitive zone being joined to at least one other sensitive zone, wherein the at least two different receptors are protein; the electronic tongue or nose sensor is characterized in that said at least three sensitive zones are different from one another by virtue of their respective receptor compositions, with at least one of the at least three sensitive zones comprising a mixture of at least two different receptors, the other two sensitive zones each comprising at least one of the at least two different receptors, wherein said at least three sensitive zones have said at least one of the at least two different receptors at constantly incremented percentage proportions in order, wherein the electronic tongue or nose sensor has k number of sensitive zones having said at least one of the at least two different receptors, and the percentage proportion of said at least one of the at least two different receptors in the sensitive zones change in increments of 100/(k-1); and said at least three sensitive zones are different from one another in terms of the respective proportions of said at least two different receptors that said at least three sensitive zones contain.

12. A process for analyzing a sample, in which: the electronic tongue or nose sensor as claimed in claim 1 is brought into contact with the sample; signals emitted by the sensitive zones of the electronic tongue or nose sensor are measured; the signals measured in the previous step are compared with the signals generated independently by the sensitive zones of the same sensor subsequent to bringing the electronic tongue or nose sensor into contact with at least one control sample; and the sample is analyzed with respect to at least one control sample.

13. The process as claimed in claim 12, in which: the electronic tongue or nose sensor of claim 1 is brought into contact with a plurality of samples; the signals emitted by the sensitive zones of the sensor(s) are measured for each sample in the plurality of samples; the signals measured for each sample in the plurality of samples are compared with one another; and each sample is categorized with respect to all the samples analyzed.

14. The process as claimed in claim 12, further comprising a step of verifying the functionality of the sensitive zones of the electronic tongue or nose sensor by comparison of the signal emitted by a sensitive zone with the signals emitted by at least two other sensitive zones.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1 describes the obtaining and the use of a sensor according to the invention. Briefly, deposits of nine mixtures of lactose (BB 1) and of sulfated lactose (BB 2) in different proportions (0, 10, 20, 30, 50, 70, 80, 90 and 100% of BB 1) are made on portions of a support consisting of a prism covered with gold so as to form sensitive zones. The sensor thus formed is placed in a cell which makes it possible to circulate various solutions to be brought into contact with the sensor. The reflectivity of the sensor is measured in the form of sensograms obtained by surface plasmon resonance imaging (SPRi) using an emitting diode and a CCD camera.

(2) FIG. 2 represents the evolution of the reflectivity of the sensor of the invention (y-axis, as %) measured by surface plasmon resonance imaging as a function of the concentration of Erythrina cristagalli lectin (ECL) (x-axis, in nM) placed in the presence of the sensor.

(3) FIG. 3 represents the various continuous evolution profiles obtained respectively for ECL (200 nM), CXCL12-α (100 nM), CXCL12-γ (100 nM) and IFN-γ (25 nM) with the sensor of the invention. These profiles represent the reflectivity (x-axis, as %) as a function of the proportion of BB 1 contained in the portions of support described in FIG. 1.

(4) FIG. 4 represents the continuous evolution profile of an ECL+CXCL12-α mixture and those corresponding respectively to ECL and CXCL12-α alone.

(5) FIG. 5 represents 3D models of the continuous evolution profiles, as a function of the contact time with the sensor, obtained respectively for ECL alone (top left), for CXCL12-α alone (top right) and for an ECL+CXCL12-α mixture.

(6) FIG. 6 represents continuous evolution profiles obtained for food mixtures: namely soya milk (FIG. 6A), cow's milk (FIG. 6B) and rice milk (FIG. 6C).

(7) FIG. 7 represents the reflectivity measured by surface plasmon resonance imaging (y-axis, as %) for the sensitive zones of a sensor according to the invention, each representative of a mixture of lactose (BB 1) and of sulfated lactose (BB 2) in different proportions (0, 10, 20, 30, 50, 70, 80, 90 and 100% of BB 1) (x-axis) in the presence of interferon gamma D136 (IFNgD136).

(8) FIG. 8 represents a nanoparticle covered with a mixture of 10% BB 1/90% BB2 (on the left) and a nanoparticle covered with a mixture of 90% BB 1/10% BB2 (on the right).

(9) FIG. 9 represents the result of the SPRi measurement on the 10% BB 1 plot of the interaction of IFNgD136 placed in the presence of the gold nanoparticles functionalized with two types of coatings: 10% BB 1 or 90% BB 1. This effect is dependent on the nanoparticle concentration.

(10) FIG. 10 represents the assaying of IFNgD136 by ELISA analysis on an anti-interferon antibody. The IFNgD136 is preincubated with gold nanoparticles functionalized with two types of coatings: 10% BB 1 or 90% BB 1. The IFNgD136 concentration of the supernatant is then evaluated by means of an ELISA assay.

EXAMPLES

Example 1

Sensor of which the Sensitive Zones Consist of Mixtures of 2 Receptors

(11) 1. Fabrication of the Sensor

(12) Two simple molecules were used as building bricks (BB) to construct a sensor comprising a support accommodating several sensitive zones combining variable proportions of receptors: lactose (BB 1) and sulfated lactose (BB 2). Nine mixtures with [BB1]/([BB1]+[BB2]) ratios of 0, 10, 20, 30, 50, 70, 80, 90 and 100% were prepared, the total concentration being constant at 20 μM. The mixtures were then deposited on a support formed by the gold surface of a prism usable for surface plasmon resonance imaging (SPRi) and were kept in contact with the surface of the prism overnight. The chip thus formed was then washed with ethanol and then dried under a stream of N.sub.2.

(13) Consequently, after self-assembly, a sensor comprising nine sensitive zones, each having different proportions of BB 1 and BB 2, could be obtained, as is indicated in FIG. 1. This sensor was then used to analyze media comprising a single protein by SPRi.

(14) 2. Analysis of Control Media Containing a Single Protein

(15) The analysis of the protein media was carried out in a 10 μl Teflon cell, connected to a degasser and to a peristaltic pump. 500 μl of protein medium were injected. The working buffer solution used comprised 10 mM HEPES, 150 mM NaCl, 0.005% Tween 20, 2 mM MgCl.sub.2, pH 7.4, and was filtered and degassed before use. All the experiments were carried out at ambient temperature with a flow of 100 μl/min. Before each protein injection, bovine serum albumin was used to block the bare gold surface, in order to avoid nonspecific interactions.

(16) Four proteins which bind to sugars were tested: the Erythrina cristagalli lectin (ECL), the α and γ isoforms of the CXCL12 chemokine and the pro-inflammatory cytokine interferon-γ (IFN-γ). These proteins can be classified in two groups according to their ability to bind to saccharide chains: ECL is a lectin which binds to galactose, whereas the two forms of CXCL12 and also IFN-γ bind to heparan sulfate.

(17) In practice, ECL was used first to test the sensor at several concentrations (200 nM, 400 nM, 800 nM and 1.6 μM). A calibration curve (see FIG. 2) was established. This curve has a Langmuir type adsorption profile, which confirms that the system operates correctly as a sensor with a K.sub.D=300+/−150 nM.

(18) During preliminary tests, two different concentrations were used for the other proteins (100 and 200 nM for the CXCL12 isoforms, and 25 and 50 nM for the IFN-γ), in order to choose the concentrations at which the signals emitted were comparable. Finally, the concentrations that were chosen are the following: 200 nM ECL, 100 nM CXCL12-α, 100 nM CXCL12-γ and 25 nM IFN-γ. It should also be noted that, after each protein analysis, the sensor was regenerated using appropriate solutions. To this end, various solutions were tested in order to identify the solution suitable for each protein. Solutions at 0.02 M NaOH for ECL, 1 M NaCl for CXCL12-α and 1% SDS for CXCL12-γ and IFN-γ were thus retained since they allow complete regeneration of the sensor without causing damage.

(19) The pure proteins at the selected concentrations were successively injected onto the sensors, the deposits of mixtures on the support of the sensor then lighting up at various levels of gray. These images were recorded and then converted into series of sensograms.

(20) Surprisingly, the inventors observed that, for a given protein, the reflectivity was dependent on the BB composition of the sensitive zones, as is visible in FIG. 3. It is in particular observed that the response of the various BB mixtures is not the simple linear addition of the response of the pure receptors. This nonlinear behavior therefore justifies, a posteriori, the use of various sensitive zones with variable proportions of each of the receptors, since the response of each sensitive zone bears an additional piece of information. Furthermore, it is observed that, for a given sensitive zone, the intensity of the response depends on the protein injected, which indicates that the sensor responds differently to each protein.

(21) In order to illustrate the individual behavior of each protein, the reflectivity values measured one minute before the end of the protein injection were represented as a function of the proportion of BB 1 in the mixtures (FIG. 3). Interestingly, a distinctive profile in the form of a continuous evolution profile could be interpolated for each protein. The inventors were able to demonstrate that, for a given protein, the profile was essentially maintained regardless of the concentration of the protein, the intensity of the signal being, for its part, of course variable as a function of the concentration.

(22) Consequently, such sensors could be used not only for differentiating and identifying proteins, but also for quantification purposes. Furthermore, the continuous behavior of the response of these sensors to a protein provides a significant and important advantage compared with the sets of data obtained with the conventional electronic nose and tongue sensors. This is because, since, for each of the continuous evolution profiles, the signals of one receptor correlate with the others, the abnormal signals can be excluded, thereby making it possible to carry out more precise and more correct analyte identifications.

(23) In a detailed manner, the ECL profile, with a maximum signal at 70% of BB 1, is completely different than the others with maximum signals at 10% of BB 1, indicating that, as expected, ECL has a greater affinity for the sensitive zones of a support that are rich in lactose. Conversely, the proteins which bind to heparan sulfate have a greater affinity for the sensitive zones rich in sulfated lactose BB 2. Consequently, ECL can be easily distinguished from the others. More advantageously, the α isoform of CXCL12 gives a continuous evolution profile which is relatively different than those obtained for CXCL12-γ or IFN-γ. Indeed, for CXCL12-α, the reflectivity is virtually zero when the proportion of BB 1 is 50% or more, whereas, for CXCL12-γ or IFN-γ, it is much higher. A more thorough analysis of the continuous evolution profile represented in FIG. 3 reveals that, at the same concentration, CXCL12-γ has a much higher affinity for the sensitive zones than CXCL12-α. Thus, whereas with CXCL12-α at 100 nM no sensitive zone of the sensor exhibits a reflectivity greater than 1.35%, a signal of 2.30% is achieved with CXCL12-γ.

(24) While it is easy to understand that ECL can be distinguished from the other proteins using the sensor, it should be noted that the two isoforms of CXCL12, which are identical in their regions 1-68, are distinguished better than CXCL12-γ with respect to IFN-γ.

(25) However, with this first generation of sensor comprising 9 sensitive zones having different proportions of 2 receptors, it is not possible to distinguish CXCL12-γ and IFN-γ despite the difference in signal intensity. In this regard, the inventors predict that sensors prepared from additional receptors in order to generate greater diversity at the level of the sensitive zones should make it possible to distinguish heparan sulfate-binding proteins with similar but non-identical charge topologies.

(26) 3. Application to the Analysis of Complex Media

(27) The main applications of the electronic nose and tongue technology lie in the testing and analysis of complex media. In order to test the effectiveness of the above sensor in medium analysis, the inventors tested it with a mixture of two proteins: ECL (200 nM) and CXCL12-α (100 nM) (mixture 1) and also food mixtures such as soya milk, cow's milk or rice milk.

(28) The continuous evolution profile obtained is presented in FIG. 4, accompanied by the profiles obtained for the two pure proteins for comparison. This initial result demonstrates that the sensor is sensitive to the mixture and that it is capable of distinguishing the mixture from the pure proteins. In fact, it is even observed that the profile of a mixture of two proteins is close to a simple addition of the profiles of the pure proteins. This suggests that there is virtually no cooperative interaction between the proteins adsorbed on the sensitive zones of the sensor. The main advantage of this property is that it makes it possible to detect and quantify mixtures on the basis of the respective profiles of the individual components of the mixture by simple linear decomposition.

(29) Moreover, it will be advantageous to take advantage of the real-time adsorption and desorption kinetics obtained by SPRi. This additional information could add another way to distinguish the various proteins, in addition to the continuous evolution profiles. By way of illustration, FIG. 5 presents the temporal evolution of the profile of recognition of the ECL+CXCL12-α mixture in three-dimensions.

(30) In addition, in order to make a better comparison of the response of the electronic tongue with respect to three types of food samples, soya milk, cow's milk and rice milk, a profile was produced for each sample. This profile represents the reflectivity after 6 min of rinsing as a function of the lactose/sulfated lactose (L/SL) ratios of the sensitive zones (FIG. 6). It can very easily be confirmed that, for soya milk (FIG. 6A), there is a strong affinity with several sensitive zones rich in sulfated lactose, including pure sulfated lactose, 90% SL, 80% SL, 70% SL, 60% SL and 50% SL. On the other hand, for UHT milk (FIG. 6B), there is a strong affinity only with 2 sensitive zones, pure sulfated lactose and 90% SL. This shows that the electronic tongue is capable of differentiating these two products. As regards the rice milk (FIG. 6C), the profile is completely different since the strongest affinity is obtained with the sensitive zone composed of pure lactose. These results show that the electronic tongue is effective for analyzing and differentiating complex samples. Furthermore, the profile obtained for each sample can be considered to be a signature for their identification.

Example 2

Method for Selecting Appropriate Combinatorial Surfaces

(31) The use of decorated nanoparticles (NPs) as a therapeutic agent is a widely explored subject. Several authors have thus described systems which have a certain effectiveness.

(32) Among the coatings used, mention may be made of: (i) specific ligands, such as antibodies for example, which will interact directly with their target, molecule to molecule, or, (ii) developed more recently, coatings composed of a small molecule (Bowman et al. (2008) J. Am. Chem. Soc. 130:6896-6897; Baram-Pinto et al. (2009) Bioconjugate Chem. 20:1497-1502; Baram-Pinto et al. (2010) Small 6:1044-1050; Rele et al. (2005) J. Am. Chem. Soc. 127:10132-10133) or of an assembly of small molecules which in themselves individually do not have very defined biological properties (Ojeda et al. (2007) Carbohydrate Research 342:448-459; Di Gianvincenzo et al. (2010) Bioorganic & Medicinal Chemistry Letters 20:2718-2721; Bresee et al. (2010) Chem. Commun. 46:7516-7518; Bresee et al. (2011) Small 7:2027-2031; or else Wolfenden & Cloninger, (2006) Bioconjugate Chem. 17:958-966), but which generate specific properties when deposited on the surface of nanoparticles.

(33) These latter assemblies are conventionally produced combinatorially: defined sets of base bricks are mixed and then assembled with the nanoparticles. The choice of these mixtures is not necessarily rational since the appropriate structures are finally identified by screening the activity of each type of NP.

(34) This therefore means that, in order to find active coatings on the basis of libraries of compounds not necessarily having a biological activity, it will be necessary to fabricate a large number of NPs which have coatings resulting from combinatorial mixtures. The screening of the biological activity of these NP will optionally make it possible to detect which are the active mixtures. As a result of this there is a very low ratio between the number of different NPs prepared and the number of active NPs.

(35) In order to more successfully target these mixtures and therefore to reduce the number of nonactive NPs fabricated, it would be advantageous to be able to rapidly evaluate the activity of the mixtures of the basic bricks forming a receptor in a format more favorable to rapid high-throughput analysis.

(36) In this context, the inventors have shown, surprisingly, that the activity of a surface composed of a mixture of receptors on a 2-dimensional (2D) support remains very similar to that of a nanoparticle coated with the same mixture.

(37) Thus, the inventors propose to evaluate the activity of nanoparticles, or more generally three-dimensional structures, bearing at their surface a mixture of compounds forming a receptor by means of a test on a 2D sensor bearing a certain number of sensitive zones, for example in the form of spots, which are each representative of a mixture in two particular proportions of the initial compounds. The target, which may be a protein or a micro-organism for example, is then brought into contact with these sensors and the interactions are measured, by SPRi for example. The intensity of the interactions measured between the target and the various sensitive zones are indicative of the compositions of mixture to be favored for the production of the coating of the three-dimensional structures, such as NPs. Likewise, if it is preferred to avoid an interaction, a mixture composition similar to that of a sensitive zone exhibiting few or no interactions with the defined target will preferably be chosen.

(38) The inventors therefore propose to combine with the combinatorial production of coatings of three-dimensional structures, in particular of the nanoobject or microobject type, such as NPs, dendrimers or liposomes, a 2D screening step using a sensor according to the invention, very suitable for the rapid evaluation of the surface properties. The 2D-to-3D property preservation thus makes it possible to preselect active surfaces, thereby making the development of future medicaments much faster and less expensive.

(39) Description of the Principle: 1—A sensor is first of all constructed as indicated in Example 1 by preparing sensitive zones each consisting of a mixture with a specific proportion of the receptor-forming basic bricks, namely nine mixtures with [BB1]/([BB1]+[BB2]) ratios of 0, 10, 20, 30, 50, 70, 80, 90 and 100%, the total concentration being constant at 20 μM, BB 1 representing lactose and BB 2 representing sulfated lactose. 2—The affinity of the sensitive zones for the target, in this case interferon gamma D136 (IFNg D136) is then measured, in the present case by SPRi according to the modes of Example 1. 3—The sensitive zone(s) offering the best affinity is (are) selected. In the present case, the relevant sensitive zone for the IFNg D136 protein consists of a 10% BB 1/90% BB 2 mixture (FIG. 7). The sensitive zone corresponding to the 90% BB1/10% BB 2 mixture was chosen as a negative control. 4—NPs, which are gold particles coated with mixtures similar to those identified, are prepared (FIG. 8). The size of the nanoparticles is 20 nm. 5—The affinity of the NPs prepared, for the target, is verified by means of conventional biochemical tests, in this case by SPRi (FIG. 9) or by ELISA (FIG. 10).

(40) By SPRi, it is clearly observed that the NPs with the relevant coating (10% BB1) cause a decrease in signal which is dependent on the NP concentration; on the other hand, the negative control NPs (90% BB1) do not cause any variation in signal at all.

(41) The results of the ELISA confirm that the affinity for a target of a sensitive zone, consisting of a mixture of receptors on a 2-dimensional support, remains very similar to that of a nanoparticle coated with the same mixture.

(42) Moreover, the same type of process, from step 1 to 5, can be carried out on several targets; for example if it is desired for the particle to bind a first target A without binding a second target B.