NL2002868C2

NL2002868C2 - Method for the determination of competing interactants binding to a compound, use, combination and apparatus.

Info

Publication number: NL2002868C2
Application number: NL2002868A
Authority: NL
Inventors: Richard Bernardus Maria Schasfoort
Original assignee: Ibis Technologies B V
Priority date: 2009-05-11
Filing date: 2009-05-11
Publication date: 2010-11-15

Description

Method for the determination of competing interactants binding to a compound, use, combination and apparatus

Epitope-antibody interaction analysis is a method for determining whether several TgG’s 5 (antibodies) can bind simultaneously to an antigen. An antigen may contain different binding sites or epitopes. For several reasons it is important to find antibodies which are binding to different epitopes and are non-competing. If a first antibody is preventing the binding of a second antibody both antibodies can be assigned to the same group or cluster. Antibodies that can bind simultaneously to the same antigen are considered to 10 belong to different groups or clusters. The classical procedure is based on epitope-antibody interaction analysis or clustering of competing antibodies wherein, for example, in order to find competing antibodies of 14 antibodies for clustering in groups 14x14= 196 samples (antibodies) are injected. A drawback of the known epitope-antibody interaction analysis is that it is time consuming because only a single 15 competition (also referred to as interference or inhibition) can be observed in a sensorgram.

It is an object of the invention to provide a method suitable for determining the interaction of a compound with a plurality of interactants. Such a method is preferably 20 used in an epitope-antibody interaction analysis and can be used to assign antibodies to clusters of antibodies.

The invention provides a method comprising: (i) providing a compound, (ii) exposing the compound to a first interactant and allowing the compound and the first interactant 25 to interact in a first interaction, forming a primarily interacted compound, (iii) exposing the primary interacted compound to at least one second interactant and allowing the at least one second interactant to interact with said primarily interacted compound in a second interaction, forming a secondary interacted compound, and (iv) determining a parameter value representative for the second interaction. In the method according to 30 the invention a compound is tested for its interaction with at least a first and second interactant. The method is suitable to determine whether or not the molecular interaction of the first interactant with the compound influences the molecular interaction of the second interactant with the compound or vice versa. If the molecular interaction is a binding phenomenon the method may be used to study competitive binding of the first 2 and second interactant with the compound. This is suitable for use in an epitope-antibody interaction analysis. Using the IBIS-iSPR, a single injection of one of the interactants will show a response (negative or positive) to all primarily interacted compounds simultaneously. The interactant which is equal to the primarily interacted 5 interactant will be used as the inhibited reference signal. Mapping of multiple interactants in at least multiple clusters can be performed in a single run which is substantially faster then with respect to the classical epitope-antibody interaction analysis, which is considered as a major improvement in the technical field of clustering of interactants using label free biosensing instruments.

10

The compound that may be tested in the method according to the invention for its interaction with the interactant may be any compound of interest. For example the compound may be selected from a protein, (e.g. an enzyme, a structural protein, an antibody, a receptor), a carbohydrate, a lipid, a nucleic acid (e.g. a DNA molecule or an 15 RNA molecule), viruses, a low molecular weight organic molecule, e.g. a drug, etcetera. The compound may also comprise complex constituents such as a biological sample for example serum (e.g., blood) or a cell lysate from either a cell culture or tissue. Thus the term compound should be construed in a broad meaning and is not restricted to single compounds having specific and defined molecular structures. Although in some 20 preferred embodiments such specific and well defined compounds may be used in the method of the invention.

Preferably, the compound is immobilized on a solid carrier (substrate). For example solid glass carriers may be selected. Alternative solid carriers may be selected from 25 metals, inorganic and organic materials. The surfaces will contain linking layers which can either be 2D (flat) or 3D (hydrogel) layers. In Chapter 6 of the “Handbook of Surface Plasmon Resonance”, R.B.M. Schasfoort and A.J. Tudos, Royal Society of Chemistry, London March 2008, specific details of these alternative surfaces are given. A well known coupling method is the EDC-NHS coupling method, which in detail is 30 also described in Chapter 6 of said Handbook of Surface Plasmon Resonance, wherein it is shown that compounds can be immobilized in many different manners. A compound can also be immobilized/captured indirectly by an immobilized ligand. Examples are a solid carrier with immobilized streptavidine. A biotinylated compound will be captured by the streptavidin. An Fc tag of a protein/compound can be captured by Proteïne A/G

3 immobilized to the substrate or an antiFc, Flag tag, His tag, Myc tag, Universal linkage system (ULS) etcetera. An elucidation of this matter is given in Example 1 below. Selection of suitable carriers is within the skill of the skilled person. Methods suitable for immobilization of compounds of the type mentioned above are known in the art.

5

It is advantageous to deposit the same compound on a solid carrier, preferably substantially homogeneously, or a plurality of different compounds on separate areas of a solid carrier. In this way an array, such as a micro-array may be formed. It is known in the art that the use of arrays and in particular the use of micro-arrays is advantageous 10 for the analysis of a large number of samples. In certain embodiments of the invention it is preferred that the same compound is immobilized on separate areas of the solid carrier. Then a plurality of different interactants are formed in a microarray.

In the method of the invention the compound is exposed to a first interactant. The first 15 interactant may be selected from the same group of specific binding interactants to the compound. The skilled person will understand that the combination of the compound and the interactant should have a potency of interaction. In view of their potency to interact with a wide range of molecules it is preferred that either the compound or any of the interactants is selected from a protein, preferably an antibody, a receptor an 20 enzyme or nucleic acid. The use of antibodies as either the compound or any of the interactants is particularly preferred. Most preferably the first, second and any optional higher order interactants are selected from specific binding antibodies. Antibodies may be derived from natural sources such as animal blood (serum) or may be produced in cell culture with techniques known to the skilled person. The term antibody may also 25 refer to antibody fragments, nanobodies, affibodies, genetically engineered VHH’s, etcetera, or any library of interacting molecules.

During the exposure of the compound to the first interactant an interaction between the compound and the first interactant may occur. Depending on the properties of the 30 compound and the interactant this interaction may or may not result in some form of association, e.g. binding, between the compound and the interactant. The primary interacted compound resulting from the interaction of the compound with the interactant thus may or may not be an association of the compound with the interactant. Depending on the selection of the type of compound and the type of interactant selected the skilled 4 person will be able to select suitable conditions for the exposure of the compound to the interactant. In a preferred embodiment such conditions are selected so that they will prefer any possible association, e.g. binding, between the compound and the first interactant.

5

According to a further preferred embodiment a plurality of different compounds is immobilized directly or indirectly in an array or the same compound is immobilized substantially homogeneously to the surface by direct or indirect coupling methods. According to a further preferred embodiment the exposure of spots of the immobilized 10 compound(s) to the first multiple interactants is performed by means of a microspotter or microarraycr, preferably a microsporter in which the first intcractant is interacted with the immobilized compound(s) until saturation has been reached. Preferably microsporters that use conditions of flow, for instance a continuous flow microsporter is suitable to saturate the interaction of the interactants with the compound. In this 15 embodiment it is especially preferred if a single compound is immobilized to the solid carrier. The use of a continuous flow microsporter is beneficial to provide maximal interaction between the immobilized compound(s) and the first interactant. It is known that with the use of such a continuous flow microsporter binding processes may be performed up to the saturation level.

20

In the method according to the invention the primary interacted compounds, resulting from the interaction of the compound with a plurality of first interactants, is exposed to the second interactant. The second interactant may be selected from a group of specific binding interactants to the compound as the first interactant and again the skilled person 25 will be able to select suitable conditions for the exposure of the primary interacted compound to the second interactant. In case the primary interacted compound is an association of the compound with the first interactant, this association of the first interactant with the compound may or may not influences the interaction of the second interactant with the compound in comparison to the situation wherein the first 30 interactant was not in association with the compound. Tn order to determine whether or not the interaction of the second interactant with the compound is influenced by the previous interaction of the compound with the first interactant, a parameter value being representative for the second interaction is measured. In this context it is noted that the interaction with the first interactant can be performed blindly in the microsporter, 5 wherein saturating conditions are forced within the microspotter to be able to observe maximal blocking of during a subsequent measurement of a (binding of a) subsequent interactant.

5 It may be noted that a regeneration step which also removes the first interactant from the compound between the exposure to the first interactant and the exposure to the second interactant is preferably not performed in the method according to the invention. A regeneration step has several disadvantages, because a compound can sometimes not withstand the harsh conditions of a regeneration step (e.g. acid pH=2). If the compound 10 is captured by a ligand by indirect means then usually regeneration cannot be performed. Washing while maintaining binding conditions is preferred to remove any non-specific binding of impurities.

The parameter representative for the second and higher order interaction may be 15 selected from refractive index measurements, mass changes, or charge distribution changes. Labelling of the second and higher order interactant using e.g. fluorescent, radioactive, enzyme, luminescent, combination thereof, etcetera, labels for visualization of the second interactant can also be applied in the method according to the invention. Preferably refractive index measurements using the SPR principle can be carried out by 20 SPR (Surface Plasmon Resonance) angle shift measurements, wave length shift measurements or reflectivity measurements or any parameter which reflects the intrinsic refractive index shift at the surface. Other label free biosensing instruments as QCM (Quartz Crystal Microbalance) or any other acoustic device: e.g. vibrating cantilevers apply mass changes as a result of the binding event. It is preferred to select a parameter 25 that may be measured by means of a technique not requiring a label for determination of the parameter value such as surface plasmon resonance, quartz crystal microbalance, interferomety etcetera. According to a preferred embodiment of the invention the SPR dip shift which is correlated to refractive index and molecular mass of the binding interactant as the parameter representative for the second interaction is selected and 30 Surface Plasmon Resonance is used to detect this parameter. Other SPR embodiments use the wavelength shift or interference pattern shift as parameter for refractive index/mass changes to the surface.

6

According to a preferred embodiment of the invention interactions with additional higher order interactants e.g. a third interactant, a fourth interactant, a fifth interactant, a sixth interactant, a seventh interactant and even more interactants to an unlimited number may be included. For these steps (iii)-(iv) may be repeated for each additional 5 higher order interactant. This repeating exercise is indicated as step (v) in claim 1. Thus for forming a tertiary interacted compound step (v) would comprise repetition of steps (iii) and (iv), and more in particular step (iii) exposing the secondary interacted compound to at least one third interactant and allowing the at least one third interactant to interact with said secondary interacted compound in a third interaction, forming a 10 third interacted compound, and step (iv) determining a parameter value representative for the third interaction.

From this the process sequences of steps (iii) and (iv) for the optional fourth interactant, fifth interactant, sixth interactant, seventh interactant or higher interactant may easily be 15 derived.

In a preferred embodiment, identical compounds are immobilized on the solid carrier either direct or indirect (via capturing by an immobilized ligand) and are exposed to different first interactants. In this way the interaction of a second interactant interacted 20 with a large number of different primary interacted compounds may be tested. For a person skilled in the art it would also be conceivable that the immobilized compounds on distinct locations are exposed to the same first interactants.

It is further preferred if the immobilized compound with first interactant in an array is 25 exposed to multiple different second interactants in a single step. In this way even a larger amount of second interactants may be tested. Preferably, the immobilized primary interacted compound is exposed to multiple different second interactants simultaneously by presenting the multiple different second interactants in a mixture.

30 The value representative for the second interaction (and any optional higher order interaction) in a preferred embodiment is compared to a reference value. The reference value is obtained by measuring a parameter representative for the interaction of the compound (or a higher order interacted compound) with the second or any optional higher order interactant being equal to the interactant of the primarily interacted 7 compound. Preferably, the interaction values are clustered into data related to combinations of first and second interactants. Advantageously, the difference of response between the reference primarily interacted compound and other primarily interacted compound(s) to the second interaction (or an interactant of an order following 5 the certain higher order) is determined, and if the difference is below a predetermined value, the combination of first and second interactants is registered to show a mutually inhibitive binding with respect to the immobilized compound. This procedure provides a convenient way to qualitative and/or quantitatively express the extent to which certain interactants influence their mutual interaction to the compound.

10

The invention also provides the use of the method according to the invention for determining mutually inhibitive binding of combinations of first and second interactants and if applicable between one or more of second and third, third and fourth, fourth and fifth etcetera interactants with respect to the compound. Determination of such mutually 15 inhibitive binding of combinations of interactants to a certain compound may provide relevant information with respect to for example epitope-antibody interaction analysis.

The invention also provides a combination of first and second interactants and optionally one or more of second and third, third and fourth, fourth and fifth etcetera 20 intcractants as determined by the method according to the invention as having or not having mutually inhibitive binding to the immobilized compound. It is known that noncompetitive interactants, in particular antibodies, are more effective than competitive interactants during e.g. immunotherapy, colouring of tissues, and in mixtures of monoclonal antibodies.

25 The invention further provides a system designed to perform the method according to the invention. The system comprises at least one apparatus comprising at least one holder for holding a compound, preferably an immobilized compound, dispensing means for adding a predetermined amount of an interactant to the compound, means for determining a value of a parameter representative for interaction of the compound with 30 the interactant, and processing means for processing the measured parameter value. The system may comprise a single apparatus. Alternatively the system may comprise a plurality of apparatuses each performing a separate task. For example a microarray may be provided by a first apparatus, e.g. a continuous flow microspotter and determination 8 of a value representative for the interaction of the compound and interactants may be performed in a second apparatus.

The invention will now be further elucidated in detail by the following non-limiting 5 examples. These examples are directed at the use of antibodies as the interactants. However, it will be clear from the description above that the invention has broader utility.

Example 1: Epitope-antibody interaction analysis - the classical method 10

In the Bio Interaction Lab of IBIS Technologies, a study was performed for finding whether several IgG’s can bind simultaneously in combinations of three lgG’s to an antigen. First some experiments were performed according to the ‘accumulation method’ (acc) as described in J. de Kruif, A.B.H. Bakker, W.E. Marissen, A. Kramer, 15 M. Throsby E. Rupprecht and J. Goudsmit, “A Human Monoclonal Antibody Cocktail as a Novel Component of Rabies Postexposure Prophylaxis”, Annu. Rev. Med. 2007. 58:1-10. However competition according to this ‘accumulation method’ was hampered by technical issues for some of the combinations of three IgG’s tested and only reliable data could be obtained when very long exposure times to reach saturation are used. An 20 ‘inhibition-map’ (inh) method was developed, which has several advantages over an accumulation test of IgG’s.

Materials and methods

For the immobilization of the immobilized compound (antigen) on the surface of an S-25 EasyöSpot G-AE sensor (IBIS Technologies, lot H408-117) the IBIS easy2spot immobilization protocol and standard operating procedure was applied (available from IBIS Technologies). First concentration measurements of the IgG’s were carried out according to a mass transport limitation protocol followed by dilution of the IgG samples to 40 nM. For the accumulation experiment three IgG’s were injected in series 30 without regeneration and each association in the series was 600 seconds with a concentration of 40 nM for all three IgG. After the third association of 600 seconds a long dissociation of one hour (with 3 intermediate buffer refreshments) was performed in order to determine the dissociation rate constant of the complex. Subsequently the sensor was regenerated to clean the surface from the three analytes. In figure 1 a typical 9 3-series result of IgG binding to the antigen is shown. The three interactants, referred to as Al, A2, A6, were injected in a sequence; A1 at t = 120 s; A2 at t=2600 s and A6 at t= 4600 s. The injection of the sample was followed by injecting of the binding buffer at t= 1020 s, t=3200 s and t=5300 s respectively. Independent binding of the antibodies to the 5 antigen is observed.

Instead of accumulation of the IgG’s by injecting them in random order an epitope-antibody interaction analysis experiment was performed according to a different protocol. In this method, a single IgG is first injected followed by injection of a mixture 10 of two other IgG’s.

Results and discussion

Assuming an equal molecular weight for all the IgG’s and if every IgG binds independently until saturation then the total response of 3 associations of IgG should be 15 equal. It has been observed that the total response of the combinations of different injections deviate largely with each other. The reason for this is a basis of speculation but a variation in the number of available epitopes, different levels of saturation for each IgG and incomplete regeneration may lead to this observations. It should be noted that full saturation is not achieved, which could cause a problem as epitopes that should be 20 blocked arc still available. Another way around, regeneration of the 3-scrics was not always to the complete baseline, which can cause a problem because epitopes that should be available are still blocked. This incomplete regeneration may result in irreproducible results, the so-called “history effect”.

It was shown that the saturation levels of different IgG’s binding to antigen are not 25 equal or not reached to the same saturation level. This observation supports the idea that not only the regeneration conditions or saturation levels are an issue, but also that antigen coated at the surface of the sensor could have more or less epitopes for one or the other IgG.

30 In figure 2 it is shown that three distinguished levels of the second interaction can be observed . After the association of a IgG, the responses were zeroed at the baseline.

Then injection of the mixture IgG’s was performed. The highest level relates to the antibodies with total independency to the two IgG’s (A1/A2) from the mixture. The second level is inhibition to one of the IgG’s from the mixture and the lowest signal is a 10 control or reference signal. The highest level is the response of the mixture after injecting PBS at first followed by the mixture. The procedure is as follows. The antibodies A8, A12, A10, A6, A5, A2 and A1 are exposed to the antigen until saturation (not shown in this figure). Then the responses were zeroed at the baseline of the second 5 interaction. Then injection of the mixture(Al/A2) was performed and this second response is shown here. As mentioned three distinguished levels can be observed, wherein the highest level is the full response of the mixture with non-competing antibodies. The second level(half) is the response of the mixture to one of the interactants in the mixture (either A1 or A2). The lowest level is the interaction of the 10 mixture( A1/A2) while the antigen was blocked by the first interactant which was the same mixture A1/A2). This is the reference signal which is also important to correct for common mode effects by subtracting the response of a reference spot. As a rule a reference signal is from a spot with the first interactant is the same as the injected interactant. Per definition then full inhibition occurs as shown in this graph.

15

For comparison of the inhibition effect, the second interaction response value in a series of these measurements should always be the same when there is independency. This can be shown clearly if the signals after the first interactions are zeroed at the second baseline. When there is total independency, the response level should not be influenced. 20 In other words: an influence of the first binding to the second response is observed when IgG’s interfere with each other. Because injections were carried out in strict order the second response is independent of 1) the exact concentration of the first IgG, 2) the interaction time and total saturation of the first IgG 3) the total regeneration efficiency of the complex and 4) the amount of available epitopes of the immobilized antigen. The 25 data show that 3 IgG’s can bind the antigen independently. This information was a prelude to the invention as described in this patent application. In another example the compound was first captured by an human anti-Fc. In figure 3 and 4 the method of epitope antibody interaction analysis of revealing competition between antibodies to a captured compound is shown as an overlay plot and each line was measured separately. 30 Regeneration of the surface was necessary to build the complexes from the surface.

First the compound with Fc receptor was exposed to the anti-Fc sensor surface. Almost saturation occurs during the time that the receptor (compound) was captured by the anti-Fc. One can observe that after regeneration at the end of complex formation the binding of the compound to the anti-Fc was decreased showing a real problem of harsh 11 regeneration. After loading the antigen to the immobilized ligand injection of the first antibody was carried out followed by the injection of the second antibody. Zeroing was carried out between the two green and red rulers for determining the inhibition effect by the first antibody which binds to the receptor. Tn figure 4 the relevant sensorgrams of 5 the second interaction are shown revealing the inhibition of the antibody responses. The non-competing signals including a buffer injection give a real binding curve showing the independent binding of antibody 10. Als the inhibited /competing responses of antibody 10 are shown in this overlay plot. Referencing and zeroing to the baseline of the secondary interacted antibody is relevant to classify the antibodies and cluster them 10 into groups.

Conclusions

An IgG that binds specific to an antigen followed by injection of an IgG or a mixture of two IgG’s to the same antigen was used to develop the ‘epitope-antibody interaction 15 analysis’ for inhibition mapping IgG’s that should bind to the antigen simultaneously. Epitope-antibody interaction analysis by observing the response of the final interaction including comparing with the reference signal is independent of 1) the exact concentration of the first IgG, 2) the interaction time and total saturation of the first IgG, 3) the total regeneration efficiency of the complex and 4) the amount of available 20 epitopes of the immobilized antigen. Epitope-antibody interaction analysis by inhibition method is superior for obtaining reliable conclusions with respect to the accumulation method. A drawback of the method is the time consuming procedure of building up complexes for determining the inhibition of one antibody with the other. Important features and benefits for epitope-antibody interaction analysis of the IBIS-iSPR is a 25 combination of: • SPR imaging allows inspection of the sensor surface and real-time ‘visual’ control of the analysis.

• Small sample volume (80 μΐ) and micro litre plate operation and unlimited exposure times to get (almost) saturation of the sample to the surface by back and forth 30 mixing.

The results presented in this patent application lead to the conclusion that for epitope-antibody interaction analysis purposes, the IBIS-iSPR is highly suitable.

Example 2: Clustering of antibodies using the continuous flow microspotter 12 A new method according to the invention is described for clustering or epitope-antibody interaction analysis of antibodies with increase through-put. An antibody will bind to an epitope of a protein and for many applications it is relevant to cluster the antibodies that 5 bind to similar epitopes. Antibodies in the same cluster will compete with each other for that epitope. Clustering of antibodies can be performed using the IBIS-iSPR in a known way either by the accumulation or the inhibition mapping method as described in example 1. In the inhibition mapping method the second interactant to the primary interacted compound is always the same and will be measured. However different 10 primary interacted compounds were first built by first regenerating the surface and injecting a primary antibody followed by injection of the second intcractant. This procedure as shown in example 1 is time consuming and has disadvantages as need for regeneration for building up complexes automatically. The invention relates to create multiple primary interacted compounds by using a microarray where several primary 15 antibodies are first exposed to the compound. Any microarrayer can be used to create antibody interactions, however if a microarrayer is using flow then saturation of the antibody and blocking of the epitopes are preferably performed. In this way building up complexes with first interacting antibodies is parallellized in combination with the continuous flow microspotter (CFM). An immobilized compound (antigen QQ) was 20 coupled covalently and homogenously to the sensor surface. In the CFM spotter various antibodies against QQ are exposed to this layer until saturation and on the spots the various antibodies were captured. The CFM spotted chip was inserted in the IBIS-iSPR and several injections of antibodies in series were carried out without regeneration. The response of binding of the secondary (and third, fourth etcetera) antibodies will be 25 inhibited by the antibodies that bound during the CFM spotting process to an identical binding site.

Materials and methods

The Continuous Flow Microspotter™ (CFM) is a technology from Wasatch 30 Microfluidics http://www.microfl.com that uses flow to deposit biomolecules on biosensor surfaces. Unlike other systems which deposit droplets onto a surface, the CFM uses flow to cycle molecules back and forth over the surface, yielding optimal binding from crude or dilute solutions. Exposure of the analytes can be performed until saturation of binding to the spots. In order to get full inhibition of the secondary 13 response the saturation of antibody binding is crucial. A CFM consisting of 48 flow cells with individual in- and outlets has been applied to create figure 5.

For the immobilization of the immobilized compound (QQ) on the surface an S-5 Easy2Spot G-AE sensor (IBIS Technologies, lot H408-117), the IBIS easy2spot immobilization protocol and standard operating procedures were applied. In total 14 different antibodies and controls were exposed during 40 minutes in duplo to the sensorchip and 32 spots can be found in the SPR image (due to constructive restrictions of the device used). A typical SPR-image of a spotted sensor chip (microarray) is shown 10 in figure 5, wherein 32 spots are visible. Different antibodies were exposed to the spots. The weak spots arc reference or control spots. Clearly an air bubble (artefact) at the bottom can be observed.

Results and discussion 15 In example 1 it was shown that the saturation levels of different IgG’s binding to the antigen QQ are not equal or not reached to the same saturation level. This observation supports the idea that antigen coated at the surface of the sensor could have more or less epitopes for one or the other IgG. However the response of a single IgG to different spots can reveal inhibition. This should be observed in an association process of 20 injection of several antibodies in series. The binding of the antibodies either inhibited or not to the primarily interacted compounds should be revealed. In figure 6 a sensorgram (serial plot) of IgG interactions (unprocessed data) of 6 different antibodies (A1 - A6) to 14 IgG captured spots as obtained from the microarray of figure 5. Antibody A3 shows no interaction (flat lines) for all spots. Because the antigen which was 25 homogeneously immobilized to the sensor surface was already saturated with 14 antibodies in the CFM spotter the second, third etc response is different. Inhibition can be clearly observed in the next figures including tiled sensorgrams when the right reference spot was subtracted from the result. In the sensorgram of figure 6 it is difficult to observe directly the inhibition effect. By means of software, for example SPRint 30 software (IBIS Technologies B.V.), the individual injections can be zeroed and referenced. E.g. for injection of Antibody Al, the spot with already A1 is used as the reference signal and subtracted from the Rol’s. In the figures 6 to 10 the results are shown. In figure 7 in a single sensorgram the responses to 13 spots of antibody A1 with reference Al are shown. Just before injection the baselines of all spots were zeroed. The 14 spot with already A1 was used as a reference and the reference signal was subtracted from the responses of all the spots inhibition to spots containing the antibodies A4, A7, Al 1 and A1. The other spots containing A2, A3, A5, A6, A8, A9, AIO, A12 A13 and A14 showed no inhibition with A1 indicated by the higher responses. The levels deviate 5 because the immobilization of QQ is not equal for all the spots. Besides no region of interest (spot) sensitivity correction and calibration was carried out, A more clear presentation of this result is shown in the tiled graph of figure 8 which can be obtained e.g. by SPRint software of IBIS Technologies. Again, from this tiled graph, easily the inhibition (flat line) of the responses to the spots containing the antibodies A4, A7, Al 1 10 and Al(ref) can be observed and these antibodies can be grouped in the same cluster. This is a convenient representation where single responses arc shown in tiled sensorgrams obtained from responses of the spots. Figure 8 in tile format represents the same data as shown in figure 6 (first association) and figure 7 (overlay plot). Each sensorgram represents the interaction of antibody A1 with the spots with the antibodies 15 as indicated in the figure. All responses to the spots are shown including duplo measurements.

In figure 9 the second injection obtained from figure 6 is shown. In this presentation easy observation of the inhibited response from spots with A3, A9 and A2 (ref) are 20 shown as flat lines. The injection of A2 was followed by the injection of A3 (see figure 10). But because the spots were already saturated with A2 and with A1 no response to A3 was shown which proves that A3 are in the same cluster as A2. This is because inhibition is not caused by the spotted antibody but by the previous injection. A3 is assigned in the same group as A2 as was already shown in figure 9. Since this A3 25 injection can be regarded as being equal to A2, further injections may be performed. This shows that inhibition of the signal is not only caused by the primarily interacted antibody but can also be attributed by previous injections. In figure 11 all antibodies are grouped in clusters, which is the final result file of the measurement of figure 6. Antibodies belonging to a specific cluster have the same pattern, wherein antibody A7 30 is double patterned. As can be seen in figure 11, cluster I (derived from figure 8) is formed by A1, A4, A7, Al 1; cluster 2 (derived from figure 9) is formed by A2, A3, A9; cluster 3 (derived from figure 6) is formed by A5, A7, A12; cluster 4 is formed by A6; and cluster 5 is formed by A8, AIO, A13, A14.

15

Inhibited (but still weak) responses remain because during building up the complexes dissociation is observed which allows a weak binding of an antibody of the same cluster. During injection and building the complexes the reliability of clustering decreases definitely. In this test 6 antibodies were injected in series however many more 5 injections can be performed when there is a restricted amount of clusters.

An injection of A4 (not shown in a tiled new figure but it can be subtracted from figure 6) showed inhibition with spots which contain initially Al, A7, Al 1 and A4. An injection of A5 revealed that antibody A7 and A12 showed interference, (not shown in a 10 tiled figure but it can be subtracted from figure 6). Because this is the fifth injection in series the response to spot A12 is significant. However it is regarded as inhibited because also reference signals increased. An injection of A6 revealed that no interference to all spots occurred (except A6 as reference). All spots showed no inhibition. From this we could conclude that the following clusters could be identified 15 from this run using a single chip with prespotted primarily interacted compounds.

Cluster 1: Al, A4, A7, All; Cluster 2: A2, A3, A9; Cluster 3: A5, A7 and A12; Cluster 4: A6 and cluster 5: A8, AIO, A13, A14 (non inhibited). This result is finally presented in figure 11. From this analysis run, antibodies A8, AIO, A13 and A14 can not be classified in one of first 4 clusters because these antibodies showed no inhibition to Al-20 A6. A7 is double in cluster 1 and cluster 3. It is remarkable that antibodies can be found that are overlapping in two clusters.

The results presented in this example 2 lead to the conclusion that for antibody clustering purposes, the IBIS-iSPR in combination with the CFM spotter is a perfect 25 combination. Using the CFM spotter saturation of the binding sites for that antibody can be reached by exposing the antibodies at higher concentration for a longer period of time. From 14 antibodies captured to an antigen in a single run and in less then 3 hours at least 4 clusters could be identified, which is dramatically faster than existing methods, where all interactions are performed in series.

30

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. Use of the verb “comprise” and its conjugations does not exclude the presence of elements or steps 16 other than those stated in a claim. The article “a” or “an” preceding an element does not exclude the presence of a plurality of such elements. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

5

Claims

A method for determining the interaction of a compound with a plurality of reactants, comprising the steps of: (i) providing at least one compound, preferably at least one compound immobilized on a solid support, (ii) the at least one exposing at least one first reactant to the compound and reacting the at least one compound with the at least one first reactant in a first interaction, thereby forming at least one primary reacted compound, (iii) reacting to at least one second reactant exposing the primary reacted compound and allowing the at least one second reactant to react with the primary reacted compound in a second interaction, thereby forming a secondary reacted compound, (iv) determining a parameter value representative of the second interaction, and (v) optionally, performing additional subsequent higher-order interactions, e.g. a third int eraction, a fourth interaction, a fifth interaction, a sixth interaction, a seventh interaction, by continuing the steps (iii) - (iv) for the at least one secondary reacted compound and subsequent reacted higher order compounds. 20

A method according to claim 1, wherein the at least one compound is immobilized and a plurality of primarily reacting reactants for the compound have reacted at different locations of the support, preferably arranged in a matrix, more preferably a micro matrix, more preferably a micro matrix formed by capture.

The method of claim 2, wherein the immobilized compounds are exposed to different first reactants at different locations of the support. 30

The method of any one of the preceding claims, wherein the primarily reacted compound is exposed to a plurality of different second reactants.

The method of claim 4, wherein the plurality of different second reactants is in a mixture.

A method according to any one of the preceding claims, comprising a higher-order interaction, for example a third interaction, a fourth interaction, a fifth interaction, et cetera.

7. A method according to any one of the preceding claims, wherein there is no regeneration step between the exposure to the first reactant and the exposure to the subsequent reactants.

The method of any one of the preceding claims, wherein the value representative of the second interaction and each of the optional higher interactions is compared with a reference value. 15

A method according to claim 8, characterized in that the reference value is obtained by measuring a parameter representative of the interaction of a reacted compound with a higher order reacted compound, wherein the reactants used are substantially identical. 20

The method of any one of the preceding claims, wherein the value representative of the second interaction and each of the optional higher order interactions are clustered into data related to combinations of first and second, and optionally one or more of second and third, third and fourth, fourth and fifth etc. reactants.

11. A method according to any one of the preceding claims, wherein the difference between the first and second reactants, and optionally between one or more of second and third, third and fourth, fourth and fifth etc. reactants interaction values is determined, and the competitive binding of the reactants are recorded if the difference is less than a predetermined value.

A method according to any one of the preceding claims, wherein the interaction value is measured using a label-free technique, preferably Surface Plasmon Resonance.

Use of the method according to one of the preceding claims for determining mutual inhibitory binding of combinations of first and second reactants, and optionally between one or more of second and third, third and fourth, fourth and fifth etc. reactants with respect to of the connection.

A combination of at least two reactants as determined by the method of any one of claims 1 to 12 with or without mutual inhibitory binding to the compound.

System for carrying out the method according to one of claims 1 to 12.

16. System according to claim 15, wherein a micro matrix is applied and captured by a first device, and determination of a value representative of the second interaction or optionally a higher order interaction is performed in a second device.

The system of claim 16, wherein the first device comprises at least one microspotter, wherein interactions preferably occur under flow conditions, and preferably a continuous-current microspotter.