WO2004077054A1 - Quality control of binding measurements on microarrays - Google Patents

Abstract

The present invention relates to methods for controlling the quality of binding measurements on microarrays, microarrays with control ligands and control ligands per se for carrying out said methods of quality control and methods for detecting said type of control ligands.

Description

 Quality control of binding measurements on microarrays

The present invention relates to methods for quality control of binding measurements on microarrays, microarrays with control ligands and control ligands themselves for carrying out methods for quality control and methods for finding such control ligands.

Microarrays have become increasingly important in recent times. They enable the presentation of an enormous variety of chemical or biological materials for the most diverse areas of application.

Polynucleotide (e.g. DNA or RNA) or oligonucleotide arrays are the most common. These are often based on a glass support, which, chemically modified, enables the synthetic structure or the connection of the desired sample. Sometimes such modified carriers are also called chips or biochips. These can then include can be used for hybridization purposes. Microarrays with immobilized proteins have carried out a comparable development. However, there is a certain incompatibility with at least some proteins with the solid support, so that, for example, signs of denaturation can be observed. Chemical microarrays are less commonly described in the literature, i.e. Microarrays on which organic molecules mostly with a low molecular weight (so-called "small organic molecules"; in English also "small molecules" or "small organic molecules") are immobilized. This is due to the very different physical and chemical properties of the compounds of this heterogeneous class of substances, which must be taken into account in the synthesis of the samples and the production of the microarrays.

The advantages of such microarrays lie in their easy handling, their potential for miniaturization and the possible automation in the manufacture and use in instruments, such as measuring devices. On the one hand, this is due to microsystem technology, which enables structuring down to the nanometer range, for example. On the other hand, the progressive allow Developments in the handling of liquids with the help of micropipettes or pins (pins) the depositing of ever smaller amounts of liquid drops on the microarrays, trying to reduce the variance of the volumes from drop to drop.

An important area of application, in particular for chemical microarrays, is drug discovery, the advantages mentioned above being clearly evident. Here, a large number of different compounds are brought into contact almost simultaneously with a target molecule, for example a protein which is associated with a disease or a therapeutic area of application, and the interaction between the immobilized compound and the target molecule, e.g. as a binding assay, detected (screening). The information obtained from this then serves to further develop interesting structures up to the active ingredient. Arrayscreening proves to be faster and more cost-effective than traditional high-throughput (HTS) screening, which often works with microtiter plates (in soluble form).

The quality control of the microarrays is of great importance for all areas of application. Numerous controls relating to the quality and success of certain processes are generally known and are also used in chip or microarray technology. This includes, for example, making several copies, for example duplicates, both of samples on an array and of the array itself. In addition, so-called positive and negative controls are used.

For example, WO 01/6236 A1 describes a surface plasmon resonance (SPR) chip on which individual control spots are intended to serve as positive or negative controls. Such QC controls are intended to ensure that the chip production (in particular spotting) as a whole and the assay for which the chip can be used, for example, have worked as such. This is based on the assumption that a certain number of control samples, which are placed evenly on the chip, can be predicted when the expected result occurs (signal detection of the positive control or missing or a signal comparable to the background noise in a negative control) and after statistical evaluation a statement regarding the Allow the test to be successful for all sensor fields. However, WO 01/6236 A1 does not disclose anything about the property and chemical structure of such positive or negative controls.

DE 102 45 651.8 discloses the use of control ligands on sensor fields of a microarray, which serve to check how completely a sensor field is occupied by a ligand. The specific bond between this control ligand and a mobile interaction partner known per se is used here. An interaction of the control ligand with the mobile interaction partner actually to be examined for the ligands is not used for (quality) control purposes.

MacBeath et al. describe in J. Am. Chem. Soc. 1999, 121, 7967-7968 the production of microarrays and the detection of the protein-ligand interaction. Several sensor fields are covered with a derivative of tetramethylrhodamine as reference points and this fluorescence-active compound is detected with the aid of a fluorescence scanner. This can be used to demonstrate that the ligand immobilization step was successful. However, it is not possible to make statements about the general success (in the form of a quality control) of the actual protein-ligand interaction.

However, there is a need in the art to use control ligands on microarrays which enable control of binding measurements between ligands and their mobile interaction partners (also called targets).

It has now been found that such a control is possible if non-specific binding universal binders are used as control ligands, which bind with numerous targets with sufficient reproducibility. These properties make it possible to use such universal binders for a large number of different targets. This means that a special control ligand that specifically binds to the target does not have to be used for each target. Surprisingly, it has been shown that, despite the non-specific interaction of the universal binders, statements about quality control are possible. Furthermore, it was surprisingly found that when at least two oppositely charged control ligands are used on microarrays, the number of targets which can be used for binding measurement with ligands increases again strongly. So far, the inventors have not been able to find targets in which such oppositely charged control ligands were not applicable.

It has also surprisingly been found that it is possible to increase the binding capacity of a universal binder with a large number of targets by repeatedly incorporating one or more particularly suitable structural motifs (for example charge-carrying or charge-stabilizing) into a universal binder which is used as a control ligand.

The object is accordingly achieved by a method according to the invention for checking binding measurements on microarrays, the steps comprising (a) providing a microarray for ligands to be tested on sensor fields;

(b) applying at least one control ligand to at least one sensor field;

(c) Measuring the binding at least between the control ligand and a mobile binding partner for the ligands to be tested, at least one control ligand being a non-specific universal binder.

Advantageous features of the method are specified in the subclaims.

In the context of the invention, the term “universal binder” is to be understood as a chemical compound which, owing to its chemical structure, is capable of binding to numerous targets in a measurable manner. In this case, the binding must be demonstrably reproducible within the usual error tolerances. Such a universal binder preferably binds to at least 10 different targets, which differ as much as possible in size, hydrophobicity and isoelectric point.

In the context of the present invention, the term “non-specific” or “non-specific binding” is understood to mean a non-covalent interaction between ligand and target. This should not be based on the known principles of molecular recognition, such as the binding of the control ligand as a substrate in a binding pocket of a protein as a target, hybridization, etc. In the context of the present invention, the term “target” is to be understood as a mobile binding partner or a potential mobile binding partner for children. Targets can represent, for example, RNA, DNA, and natural or synthetic proteins. Here, "mobile" means that the binding partner is not fixed in any way on the microarray.

In the present invention, the term “sample” is applied to molecules which the person skilled in the art would consider as potential binding partners for such a target due to their spatial and / or electronic structure.

In the context of the present invention, the term “ligand” encompasses any sample that can be immobilized, for example, adsorptively, ionically, complex-chemically or preferably covalently on a sensor field, or any sample fragment of a biological, chemical or other nature. In connection with the molecular interaction of binding partners, the term "ligand" determines the partner that is to be present on the array. The other partner (s) is / are referred to as a "mobile binding partner".

A microarray for use in the method according to the invention preferably carries at least two different types of ligands. On the one hand, there are one or more control ligands, the structure of which has been optimized with regard to non-specific binding with the ligand. On the other hand, there are ligands to be tested, whose ability to interact, preferably specifically, with the target is to be checked.

Unless otherwise specified, a reference to several or to different control ligands or ligands to be tested within the scope of the present invention is to be understood in such a way that not only individual ligands of the respective structure are applied to the array, but that different types of ligands in each case in multiples Copies are available on the array.

In the context of the present invention, the term “microarray” encompasses any objective configuration which has at least one surface on which sensor fields are present. There are no restrictions on the Shape or dimensions of a microarray or the shape, number or dimensions of the sensor fields or the density of the sensor fields on a microarray. The suffix "micro" only serves to clarify that miniaturization is preferred wherever possible.

Different measurements can be compared with one another using the control method according to the invention. The comparison of the measured value of a control ligand can include with a previous measured value of this control ligand and / or with the measured value of the control ligand at another point in the microarray and / or with the measured value of the control ligand on another microarray and / or with another control ligand on the same and / or another array and / or with ligands on the same and / or a different array.

The character of the method according to the invention as a control measurement therefore generally results from a comparison of the binding values obtained for the

Control ligands with values that are in a further, i.e. previous and / or parallel

Use of the same control ligand was obtained. These measurements can be carried out spatially and / or chronologically on one or more arrays. Due to the "universal binding" property of the control ligand, values can be compared that were obtained with identical targets (e.g. with multiple targets)

Use of a control ligand in combination with changing ligands to be tested with the same target), but also values that were obtained with different targets (e.g. when using a control ligand in combination with constant ligands to be tested with different targets). Of course, the comparison used for the control can also be carried out with information on the binding ability, which the person skilled in the art e.g. are known from patent and technical literature for such ligands that can be used as universal binders.

Based on this comparison, statements in qualitative, semi-quantitative or quantitative terms are, for. B. possible for calibration purposes. In the case of a qualitative statement, a "yes-no statement" results as proof of quality. Here, for example, the fact that the control ligand shows a signal and thus a measured value can be determined is used to determine the success of the binding measurement as such. This is where the (qualitative) comparison is made with previous measured values for the control ligand, which indicate the occurrence of binding anzeigten. In another embodiment, a qualitative control statement is possible, for example, in which it is checked whether a certain control ligand shows a signal on a microarray (and thus a measurement value which should not be regarded as noise) and whether this signal can also be detected on a duplicate of the array is.

Different microarrays can therefore be compared with each other (array to array control) on which the same ligands are present (checking or scaling of duplicates). Different microarrays, each with different ligands, can also be compared with one another. In general, the method can also be used for scaling or a semi-quantitative comparison of signals (measured values) from ligands using the control ligand. However, measuring devices for binding measurement via the control ligands can also be compared with one another. If one or more control ligands are occupied multiple times on a microarray, homogeneity checks of the array (field to field control) can be carried out.

For the person skilled in the art there are obviously a number of further possibilities by means of which the individual data (measured values) obtained can be processed for quality control purposes.

In general, the success of the binding measurement (s) can be checked on a microarray using the control method. Causes of the failure of the experiment can be found, for example, in the microarray production, the method for applying the ligands / control ligands, the state of the target or in the measuring device.

In a preferred embodiment of the control method, at least two different control ligands are used on one or more sensor fields on a microarray. It is particularly preferred that there are exactly two control ligands. It is also advantageous if these control ligands are charged positively and negatively. Such control ligands preferably each have additional hydrophobic regions, for example aromatics. In the context of the present invention, the term “charged” is to be understood as the electronic state of a connection which allows the formation of a positive or negative, localized or delocalized charge under the conditions of the bond measurement. This also includes, for example, compounds that are not present in the charged state over the entire pH range, but that take on a charge by taking up or releasing protons.

It has been shown that despite unspecific interaction with numerous targets, the binding of like-loaded control ligands and targets can lead to a comparatively lower signal. The combination of complementarily loaded control ligands therefore means that almost all targets can be addressed using this control ligand combination.

Due to the influence of the charge on the property of the control ligands as non-specific universal binders, dyes or their derivatives are particularly suitable as control ligands, since these are often capable of forming charges

In the context of the present invention, “derivatives” are to be understood as compounds which result from the original compound by formal substitution, addition or elimination.

Eosin and its derivatives are particularly preferred as control ligands. The connection to the sensor array of the microarray can take place, for example, via the carboxyl function. The nonspecific binding of eosin to proteins is known and is used to determine protein concentrations photometrically (A. Waheed et al., Anal. Biochem. 287, 2000, 73-79).

Preferred eosin derivatives are compounds of the general formula (I)

wobei A¹ bis A⁴ jeweils unabhängig voneinander H oder Halogen, vorzugsweise Br oder I, und Ar eine durch eine Carboxylgruppe substituierte Phenylgruppe, die gegebenenfalls durch 1 bis 4 Halogenatome substituiert sein kann, ist.

where A ¹ to A ⁴ each independently of one another are H or halogen, preferably Br or I, and Ar is a phenyl group substituted by a carboxyl group, which may optionally be substituted by 1 to 4 halogen atoms.

In this context, particularly suitable control ligands are erythrosin, rose bengal, phloxin, cyanosin, daphinin, eosin B, eosin Y, eosin G or acid red 51.

Other suitable control ligands include, for example, naphthol blue, black and Coomassie, which are known to be used for staining protein gels.

The use of eosin or eosin derivatives as control ligands to control binding measurements on microarrays is also in accordance with the invention.

In a further particularly preferred embodiment the control ligand is a compound which comprises several i.e. contains at least two, preferably at least three, more preferably at least four structural motifs, which is responsible for the non-specific binding to numerous targets. These are preferably structural motifs which allow the formation of a positive or negative, localized or delocalized charge on the ligand. This makes it possible to increase the strength of the non-specific binding and thus to use the control ligand more universally. This is surprising insofar as one had to assume that, due to the spatial proximity of the structural motifs, these mutually hinder each other from forming bonds to the ligand and the effect of signal amplification is therefore absent. These structural motifs can be the same or different within a ligand. It has proven to be particularly advantageous to incorporate a single structural motif into the control ligands at least twice, preferably at least three times, more preferably at least four times.

On. the preferred structural motif in this context is the guanidinium group. In order to further increase its ability to form an unspecific bond, it is preferably covalently linked to a hydrophobic group, for example an arylene radical, in particular a phenylene radical. Guanidiniumphenylalanine (GuaPhe) or a guanidiniumphenylalanine residue is preferred here Use, particularly preferably the guanidino group should be in the para position. The guanidiniumphenylalanine residue can be coupled to a framework structure via the carboxyl function of the phenylalanine. This scaffold bears the other structural motifs and, together with them, forms the control ligand. Other structural motifs that tend to bind proteins unspecifically are derivatives of 2,6-diiodophenol and 2,6-dibromophenol.

Such structural motifs are preferably linked by a framework, for example a natural amino acid such as lysine or a peptide consisting of natural amino acids such as lysine. For example, it is possible to link one or both of the lysine amino functions to a GuaPhe residue using a peptide bond. In addition, two, three or more amino acid units connected via peptide bonds can be used. In the case of lysine, the amino residues which are alpha and / or omega-stable to the carboxylic acid group can be used to attach the GuaPhe. Three lysine units, which are linked to each other via amide bonds, can thus be linked with one, two, three or four structural motifs. Compounds with three or more functional groups, such as diaminocarboxylic acids, dicarboxyamines or glycols, are generally to be mentioned as further examples of suitable framework structures.

In the context of the present invention, particular preference is given to using compounds of the following structure as control ligands

where X = -OH, -NH ₂ , -NHA, where A is alkyl or acyl, or a bond for direct or indirect attachment to a solid phase, and their tautomers, isomers, enatiomers, mixtures or salts.

In the context of the present invention, the term “alkyl” means a linear or branched alkyl chain radical of the length given in each case, which can be saturated or unsaturated and in which up to five CH ₂ groups can be replaced by oxygen, sulfur or nitrogen atoms. The heteroatoms are separated from one another by at least two carbon atoms, for example d- ₄ -alkyl means, for example, methyl, hydroxymethyl, ethyl, hydroxyethyl, vinyl, 1-propyl, 2-propyl, allyl, 2-methyl-2-propyl, 2-methyl-l-propyl, 1-butyl, l-but-2-enyl, 2-butyl, C _1-6 alkyl e.g. C _1-4 alkyl, pentyl, 1-pentyl, 2-pentyl , 3-pentyl, 1-hexyl, 2-hexyl, 3-hexyl, 4-methyl-l-pentyl or 3,3-dimethyl-butyl. Cι _-8 -alkyl in addition to the indicated z for C ^ alkyl radicals B. C 1-4 alkyl, heptyl, 2- (2-methoxyethoxy) ethyl or octyl.

The term "acyl" in the context of the present compound means a radical -C (O) - alkyl. Sensor fields and microarrays can be created in different ways. For example, the microarray can be in one or more pieces and, for example, comprise several layers. A large number of sensor fields can be located on a surface of the microarray. The sensor fields can be separated from one another by intermediate sensor field regions which differ chemically, physically or in some other way from the sensor fields.

The fixation, attachment or immobilization of the ligands or control ligands can be achieved in various ways directly or indirectly on the sensor field. (Control) ligands can be directly adsorptively, ionically, complex-chemically, covalently or otherwise attached to the surface of the sensor field. A binding matrix of mostly organic nature can be used for the indirect connection. Each sensor field can contain an organic layer as a binding matrix. The binding matrix serves for the adsorptive, ionic, complex chemical, covalent or other connection of the ligand or the control ligand to the sensor field.

The materials that a microarray contains can be of many different types and depend, among other things. on the nature of the ligands and the measurement method necessary for the method according to the invention. Suitable materials e.g. are permeable substances such as cellulose or impermeable substances such as plastics, glasses, metals, metal oxides or alloys.

The microarray can be structured or unstructured on the surface on the sensor field side

For example, a carrier as part of the microarray can have a planar surface on which there is a uniform binding matrix, for example for covalent attachment of the ligand. The sensor fields then only arise when the ligands are applied, for example by spotting. The position and the shape of the sensor fields are then determined by the spotting device. There are numerous, mostly software-based methods for measuring field detection in the prior art which are familiar to the person skilled in the art. In this case, the intermediate sensor field areas differ chemically from those Sensor fields due to the lack of ligands. However, these areas also have a binding matrix.

However, structured microarrays are preferred. The structuring can be carried out, for example, chemically by photostructuring treatment of a binding matrix applied to a planar carrier. The sensor fields are thus defined by the nature or modification of the binding matrix. Here, as in the case of an unstructured microarray, the sensor fields differ only chemically from the sensor field intermediate regions. However, this distinction is recognizable before the ligands are immobilized and the sensor field position is thus determined before they are applied. An example of such an array is given in WO 00/67028 A1.

The structuring can also be achieved by means of elevations or depressions. A microtiter plate can serve as an example of a one-piece microarray. The cavity floors (recesses) and walls can represent the sensor fields and the plate body can represent the intermediate sensor field areas. In such a case, a binding matrix can only extend over the sensor fields or over the entire surface. For example, US-A-6,063,338 discloses a microtiter plate which contains a cycloolefin for spectroscopic purposes and is said to be suitable for solid phase synthesis. Among other things, it is proposed that the interior walls and floors of the cavities be functionalized in order to immobilize components for solid phase synthesis.

A microarray can also be constructed from several layers. In the context of the present invention, it is preferred that the microarray comprises a structuring layer on a planar carrier on its surface facing the sensor surfaces. The structuring layer can, for example, be applied to the carrier with the aid of a mask. The locations at which no layer was applied then represent the sensor fields. Likewise, the carrier or a further intermediate layer can additionally have depressions at these locations, as a result of which the cavities formed are designed to be deeper. The microarray can further preferably contain a metal layer, in particular a gold layer, on the structuring layer and / or on the carrier. The gold layer would be advantageous on both Sensor fields that are formed by the exposed carrier surface, as well as the structuring layer that covers the carrier. The binding matrix can also be located only in the area of the sensor fields or over the entire array on the metal layer to the outside. Such multilayer microarrays are described in WO 01/92883 A2, the disclosure content of which is fully referred to in this regard.

The carrier of such a multilayer microarray particularly preferably consists of a light-permeable material (such as glass) and the structuring layer additionally has suitable optical properties which form a microarray as a surface plasmon resonance (SPR) sensor, as described in WO 01/92883 A1 is. Reference is hereby made in full to the disclosure content of WO 01/92883 A1 in this regard.

There are no restrictions on the dimensions of the microarray. A commonly used dimension corresponds to that of a commercially available microscope slide with the dimensions 76 x 26 mm. Other dimensions, such as 125 x 83 mm, are also conceivable. The sensor field density is advantageously more than 50, preferably more than 100 sensor fields per cm ² . The microarray preferably contains at least 96, more preferably at least 4608 and most preferably at least 9216 sensor fields. The sensor fields are preferably present in a number adapted to commercially available microtiter plate formats (same number or an integral multiple thereof). The sensor fields can be round, elliptical, rectangular or square or have another shape. Preferred sensor fields are round and have a diameter of at most 700 μm, preferably 40 μm.

The sensor fields used in the method according to the invention for checking binding measurements preferably contain a binding matrix. This binding matrix serves to bind a ligand preferably (chemically) covalently. The binding matrix can be created, for example, as a result of the modification of existing parts of the microarray or by adding additional parts. For example, a solid support on one of its surfaces can be exposed to chemicals that release reactive chemical groups, which then make it possible to bind the ligand. For example, the alkaline treatment of glass surfaces sets Si-OH groups free, which can enable a covalent bond. The surface treated in this way is understood in the context of the present invention as a binding matrix.

However, the ligand / control ligand is preferably not immobilized directly on a solid support, but an additional organic layer is applied, which is bound to the support, for example, via the chemical modification mentioned above. However, the organic layer functioning as a binding matrix does not have to be covalently bound to a solid support. Rather, an ionic, adsorptive, or complex chemical or other type of connection is also possible. Bioorganic systems that combine hydrophobic interactions with hydrogen bonds and ionic interactions can also ensure the binding of the ligand.

In a particularly preferred embodiment, the binding matrix comprises a self-assembling monolayer (SAM) as an organic layer. The self-assembly of a SAM to form a dense film usually takes place through the hydrophobic interaction of long-chain hydrocarbons at one end of which there is a functional group which enables attachment to the support and at the other end of which a functional group enables the ligand to be immobilized. Compounds that comprise these functional building blocks (head group, foot group, hydrophobic part) are also called anchors or anchor molecules. Furthermore, the anchor can have a spacer portion, which preferably contains ethylene glycol units.

Advantageously, together with the anchor molecules mentioned, so-called

Thinner molecules are applied to the support surface in order to structure the self-assembling monolayer in a suitable manner and to control the concentration of the binding sites on the surface. One too dense

Surface concentration can be disadvantageous due to steric hindrance. Thinner molecules are structurally adapted to the anchor molecules, but they do not have a head group for the attachment of the ligand. It must also be ensured that the control ligand does not bind to the diluent molecules.

For example, SAMs can be generated by chemisorption of alkylthiols on a metal surface (eg gold). The long-chain molecules pack up as SAM on the solid phase, whereby the gold atoms are complexed by the sulfur functions. A further example is the silanization of glass or silicon with reactive silanes containing epoxy or amino groups and the subsequent acylation of the amino groups, for example with nucleoside derivatives (Maskos and Southern, Nucl. Acids Res. 20 (1992) 1679-84).

For the synthesis and preferred chemical structures of anchors and diluent molecules, reference is made to WO 00/73796 A2 and WO 01/92883 A2, the contents of which are hereby incorporated by reference.

An anchor molecule of the general formula Z-R-Y is preferred.

Z is an atom or an atomic grouping, which guarantees the connection to a solid support. The choice of group Z depends on the chemical nature of the solid support. Suitable functional groups or reactive compounds are known to the person skilled in the art and can easily be determined once the carrier is known. Z is particularly preferably an element of V. or VI. Main group, whereby combinations of identical or different elements can also be used. Combinations such as -S-Se- or -Se-Se- are advantageous here. It is also advantageous, depending on the nature of the surface, to use groups that are ionized at neutral pH, such as sulfonate. Sulfur is preferably used as Z, for example in the form of the disulfide function (-SS-), the thiol function (-SH) or the sulfide function (-S-). The elements used are characterized by the fact that they either have a high affinity for metals, in particular noble metals (gold, silver etc.), and thus enable immobilization of the anchor molecules, for example on a gold, silver or platinum surface, or if so is an ionic group, can bind to a metal oxide surface such as Al ₂ O ₃ . If Z has two free valences, both can be connected to the same or different residues - RY.

The radical R represents a branched or unbranched hydrocarbon chain of more than 10 atoms in length, optionally interrupted by heteroatoms such as S, N or O, amide or ester bonds. Y enables the ligands to be covalently bound directly or indirectly with or without prior modification of the ligands. Suitable functional groups or reactive compounds are known to the person skilled in the art and can easily be determined with knowledge of the ligands. Particularly preferred for Y are acetals, ketals, acylals, acyl halides, alcohols, aldehydes, alkenes, halides, alkynes, allenes, amides, amidines, aminals, amines, anhydrides, azides, azines, aziridines, azo compounds, boranes, carbamates, carbodiimides, carboxylic acids , Carboxylic acid esters, cyanamides, cyanates, diazo compounds, diazonium salts, epoxies, ethers, hydrazides, hydrazines, hydrazones, hydroxamic acids, hydroxamic acid esters, hydroxylamines, imides, imines, inorganic esters, isocyanates, isocyanides, isothiocyanates, ketenes, ketones, nitriles, nitro compounds, , Oximes, phenols, phosphines, phosphonates, ammonium salts, phosphonium salts, sulfonamides, sulfones, sulfonic acids, sulfonic esters, sulfonium salts, sulfonyl azides, sulfonyl halides, sulfoxides, thioamides, thiocarbamates, thiocyanates, triazenes, ureas or isoureas.

Most preferred are reactive head groups, which in principle enable an almost quantitative (> 99%) reaction of group Y with the ligand. An example is the addition of thiols to a maleimidyl group. Further examples are described in WO 01/92883 A2, the full disclosure of which is hereby incorporated by reference.

Furthermore, the organic layer can comprise polymeric molecules such as biomolecules or synthetic macromolecules. Examples include hydrogels as disclosed in WO 90/5303 A1. These form a three-dimensional framework that allows the ligands to be immobilized in three dimensions if necessary. The method for quality control according to the invention can also be used for arrays with such modified surfaces.

The use of the terms layer, surface or surface concentration etc. are not to be interpreted restrictively to the effect that a two-dimensionality of the presentation of the locations of a microarray present for the connection of the ligands or control ligands would be a prerequisite.

In a preferred embodiment, the binding matrix extends over the entire microarray surface on the sensor field side. Proteins, peptides, oligonucleotides, carbohydrates (glycosides), isoprenoids, enzymes, lipid structures and organic molecules (chemical microarrays) can preferably be used as ligands.

Organic molecules can preferably be used as control ligands.

In particular, small organic molecules (small molecules or small molecular weight molecules or small organic molecule) are preferred for the ligands and the control ligands or their structural motifs, which occur once or more than once in control ligands, in the literature. The molecular weight usually provides the basis for the definition of such small molecules WO 89/03041 and WO 89/03042 describe molecules with molecular weights of up to 7000 g / mol as small molecules. Usually, however, molecular weights between 50 and 3000 g / mol are given, but more often between 75 and 2000 g / mol and mostly in the range between 100-1000 g / mol. Examples are the documents WO 86/02736, WO 97/31269, US-A-5928868, US-A-5242902, US-A-5468651, US-A-5547853, US-A-5616562, US-A- 5641690, US-A-4956303 and US-A-5928643.

In the context of the present invention, organic molecules with a molecular weight below 3000, preferably below 1000, most preferably below 750 g / mol are small organic molecules.

If a binding matrix is used, chemical functional groups present in the ligand can be used to immobilize the ligands and control ligands on the binding matrix (eg N-terminal amino groups of peptides). Existing functional groups can also be chemically modified (eg cleavage of disulfide bridges to thiols in proteins). In the production of chemical microarrays, preference is given to the use of additional linker molecules (additional molecules or “tag”) which, on the one hand, enable binding to the ligand / control ligand and, on the other hand, immobilization on the binding matrix and thereby perform an additional spacer function. Preferred for everyone Ligands and control ligands use the same linker molecule, which makes the ligands "more chemically similar". Usable linker molecules ("ligand tag") are described in PCT / EP 02/01184, the content of which is hereby incorporated by reference.

The contacting of the control ligand with the binding matrix — if present — preferably takes place under conditions under which a preferably covalent attachment of the control ligand to the binding matrix is possible. In general, even if no binding matrix is present, a liquid containing the control ligand is applied directly to the binding matrix or the sensor field.

The control ligand can be applied to the sensor fields as such or in liquid by spotting or by incubating the entire microarray surface on the sensor field or by incubating the entire microarray.

The mobile interaction partner is preferably applied over the entire surface to the surface of the microarray on the sensor field in a suitable solution (e.g. buffer solution). The concentration of the mobile binding partner should be chosen so that sufficient interaction is possible at least with the control ligand.

The binding measurement can be carried out sequentially from sensor field to sensor field or sensor field row to sensor field row, or preferably in parallel, in that all fields are measured simultaneously. The measuring method is advantageously based on a radioactive, optical or electrical method. For example, radioactive fluorescence or luminescence based detection (e.g. RIA, ELISA, etc.) are possible. However, the measurement is preferably carried out without any markings, since this simplifies implementation. The use of an optical reflection method, such as SPR spectroscopy, is particularly preferred.

According to the invention there is also a microarray for binding measurements of ligands on sensor fields with at least one control ligand on at least one sensor field, the control ligand being a compound which contains a structural motif several times. With regard to the control ligand, the same statements apply as for the control method according to the invention.

According to the invention there is also a microarray for binding measurements of ligands on sensor fields with at least one positively charged and at least one negatively charged control ligand on at least one sensor field. Such control ligands preferably each have additional hydrophobic regions, for example aromatics.

Also according to the invention is a method for finding control ligands as non-specific universal binders, comprising the steps

(a) providing potential control ligands on sensor fields;

(b) measuring the binding between the potential control ligands and several mobile binding partners; (c) Selection of control ligands based on their binding values (measured values).

Furthermore, the method according to the invention for finding control ligands can comprise a further step (d) chemically linking a plurality of control ligands to form a control ligand with a plurality of structural motifs.

For this purpose, preferably at least 10 targets (mobile interaction partners) are tested. The targets are preferably selected from the group consisting of RNA, DNA and natural or synthetic proteins, which differ as far as possible in their size as well as their charge and hydrophobicity properties.

Examples

Manufacture of a chemical microarray

Example 1 Synthesis of a Diluent ("Diluent Molecule")

a) Immobilization of N- (N ⁵ -Fmoc-5-aminopentyl) -11-mercaptoundecanamide on chlortrityl resin

600 mg (1.15 mmol) of N- (N ⁵ -Fmoc-aminopentyl) -11-mercaptoundecanamide, obtainable from S-protected 11-mercaptoundecanamide and Fmoc-1,5-diaminopentane hydrochloride, were dissolved in 15 ml of DMF and mixed with 2 g Methoxytrityl chloride resin (1.6 mmol) (Novabiochem) added. The suspension was shaken gently for 1 hour. 500 μl of pyridine were then added and the suspension was shaken for a further 3 h. The resin was then washed once with N, N-dimethyiformamide (DMF), once with 5% water in DMF, four times with DMF, three times with dichloromethane and twice with hexane and dried in vacuo. The resin was loaded with N- (N ⁵ -Fmoc-5-aminopentyl) -11-mercaptoundecanamide determined by Fmoc analysis (GB Fields, RL Noble, Int. J. Peptide Protein Res. 1990, 35, 161-214) to 0.35 mmol / g (yield 60% of theory).

b) General instructions for the coupling of Fmoc-8-amino-3,6-dioxa-octanoic acid (Fmoc-Ado)

To remove the Fmoc protective group, 1 g of the loaded resin (0.35 mmol) was carefully stirred in 15 ml 1/3 (v / v) piperidine / DMF for 20 min and then washed 6 times with DMF. Fmoc-8-amino-3,6-dioxa-octanoic acid was coupled by incubating the resin for 4 h with a solution of 270 mg (0.70 mmol) of Fmoc-8-amino-3,6-dioxa-octanoic acid, 270 mg (0.71 mmol) O- (7-azabenzotriazol-1-yl) - N, N, N ', N'-tetramethyluronium hexafluorophosphate (HATU) and 250 μl (1.44 mmol) ethyl diisopropylamine (DIEA) in 7 ml DMF. The resin was then washed 5 times with DMF, 3 times with dichloromethane and 2 times with hexane and dried.

c) Further coupling of Fmoc-Ado and acetylation

Fmoc-8-amino-3,6-dioxa-octanoic acid was coupled again to 500 mg of resin from b) (0.175 mmol) as described under b) and the Fmoc protective group was then cleaved off as described under b). Then the free amino groups acetylated by incubating the resin for 30 min with 10 ml 1/1/2 (v / v / v) acetic anhydride / pyridine / DMF. Then the resin was washed 5 times with DMF and 3 times with dichloromethane. The product was split off from the resin with 2/18/1 (v / v / v) trifluoroacetic acid / dichloromethane / triethylsilane. The product was purified by preparative RP-HPLC and analyzed by LC / MS. LC-MS (calc.): [M + H] ⁺ 635.5 (635.4), [M + Naf 657.5 (657.4)

Example 2 Synthesis of an Anchor ("Anchor Molecule")

Fmoc-8-amino-3,6-dioxa-octanoic acid was coupled twice to 100 mg of the resin shown in Example 1b) as described under 1b), and the Fmoc protective group was then cleaved off. Then 3-maleimidopropanoic acid was incubated for 1 h with 4 eq. 3-maleimidopropanoic acid and 4 eq. Coupled diisopropylcarbodiimide in DMF (c = 0.15 M).

Then the resin was washed 5 times with DMF and 3 times with dichloromethane. The product was split off from the resin with 2/18/1 (v / v / v) trifluoroacetic acid / dichloromethane / triethylsilane. The product was purified by preparative RP-HPLC and analyzed by LC / MS. LC-MS (calc.): [M + H] ⁺ 889.2 (889.5) [M + Na] ⁺ 911, 1 (911, 5)

Example 3 Production of a microarray (gold chip) with a binding matrix

a) microarray

A sensor plate (40 nm gold, 2 nm titanium as adhesion promoter,

50 μm structuring layer on a planar glass plate with the external dimensions in microtiter format (approx.12.5x8.3 cm) as part of an SPR sensor arrangement in WO

01/63256 A1 is described. The structuring of the sensor plate was done with Ormocer ^® carried out, the graphite was mixed, so that there were 4608 almost circular sensor fields with a sensor field diameter of about 700 microns.

b) Coating with the binding matrix The binding matrix is composed of the anchor from example 2 and the thinner from example 1.

First, stock solutions of the individual components are prepared by dissolving the solids in ethylene glycol / 1% TFA with the desired molarity, as described in the Ellman test (GL Ellman, Arch. Biochem. Biophys. 82 (1959), 70-77) or via the extinction coefficient 295 nm is checked. Subsequently, the stock solutions of anchor and thinner are mixed in the desired ratio (e.g. 1/24, v / v) and the entire surface of the gold is incubated with this 100μM to 1 mM solution for 1 h at room temperature, then with methanol / 0.1% TFA and several times washed with water / acetic acid (2ppm) and dried in a stream of nitrogen.

II Synthesis and immobilization of control ligands

Eosin and compound 1 ("GuaPhe-4", see below), which have a carboxyl group, serve as control ligands. Before immobilization, the control ligands are reacted with a "tag" to introduce a thiol group which can then react in an addition reaction with the maleimide unit of the anchor head group. The tag is linked to the carboxyl function of the ligand via an amide bond. The resulting control ligand-tag conjugate has the formula HS- (CH ₂ ) ₂ -NH-C (O) -CH ₂ -O- (CH ₂ ) ₂ -O-CH ₂ -C (O) -NH- ( CH ₂ ) ₂ -O- (CH ₂ ) ₂ -O- (CH ₂ ) ₂ -NH-CO control ligand (for synthesis see WO-A-02/063299).

Example 4 Synthesis of the Eosin-Tag Conjugate

20 mg (8.6 μmol) of tentagel-trityl immobilized "Tag" are mixed in with a solution of 16.7 mg (25.8 μmol) of eosin, 3.5 mg (25.8 μmol) of 1-hydroxybenzotriazole (HOBT) and 4 μl (51.6 μmol) of diisopropylcarbodiimide (DIC) 100 μl DMF shaken for 2 hours and then washed 6 times with DMF. The product is cleaved from the resin using 48/48/4 (v / v / v) trifluoroacetic acid / dichloromethane / triethylsilane, which is subsequently purified by RP-HPLC and analyzed by LC / MS. LC-MS (calc.): [M + H] ⁺ 996 (998)

1 TFA DCM / Tnet ylsilan

Example 5 Synthesis of the conjugate from tag and compound 1

20 mg (8.6 μmol) of tentagel-trityl immobilized "Tag" are mixed with a solution of 15.2 mg (25.8 μmol) of Fmoc-L-Lys (Fmoc) -OH, 9.8 mg (25.8 μmol) of O- (7-azabenzotriazole-1- yl) -N, N, N \ N4etramethyluroniumhexafluorophosphate (HATU) and 9 μl (51.6 μmol) ethyl diisopropylamine (DIEA) shaken in 100 μl DMF for 2 h and then washed 6 times with DMF. To remove the Fmoc protective group, the resin is carefully stirred in 1 ml 1/3 (v / v) piperidine / DMF for 10 min and then washed 6 times with DMF.

The product is mixed with 20.3 mg (34.4 μmol) Fmoc-Lys (Fmoc), 13 mg (34.4 μmol) O- (7-azabenzotriazol-1 -yl) -N, N, N, N'-tetramethyluronium hexafluorophosphate (HATU) and 12 μl (68.8 μmol) ethyl-diisopropylamine (DIEA) shaken in 100 μl DMF for 2 h and then washed 6 times with DMF. To remove the Fmoc protective group, the resin is carefully stirred in 1 ml 1/3 (v / v) piperidine / DMF for 10 min and then washed 6 times with DMF. This is followed by the coupling of 33.3 mg (51.6 μmol) Fmoc-L-phenylalanine (4-guanidino-Boc ₂ ) -OH, 19.6 mg (51.6 μmol) O- (7-azabenzotriazol-1-yl) -N, N, N ', N ^, -tetramethyluronium hexafluorophosphate (HATU)) and 18 μl (103.2 μmol) ethyl-diisopropylamine (DIEA) in 200 μl DMF for 2 h. To remove the Fmoc protective group, the resin is carefully stirred in 1 ml 1/3 (v / v) piperidine / DMF for 10 min and then washed 6 times with DMF. The product is cleaved from the resin using 48/48/4 (v / v / v) trifluoroacetic acid / dichloromethane / triethylsilane, which is subsequently purified by RP-HPLC and analyzed by LC / MS. LC-MS (calc.): [M + 3H] ³⁺ 1570 (1572)

GuaPhe-4-Day

Beispiel 6 Belegung der Sensorfelder mit dem Kontrollligand-Tag-Konjugat

Example 6 Assignment of the sensor fields with the control ligand-tag conjugate

To cover the sensor fields of the microarray with the control ligand-tag conjugate, this is dissolved in a mixture of ethylene glycol / water / acetonitrile in the desired concentration and adjusted to pH 7 using a phosphate buffer. Liquid drops are transferred from this solution to the sensor fields with the aid of a spotting robot using steel pins. The entire array is then washed with methanol and dried.

III bond measurements

Example 7 General bond measurement

The binding between the control ligand and the mobile binding partner was carried out with the aid of an SPR sensor arrangement as described in WO 01/63256 A1 (cf. Example 3a). By imaging the microarray, the intensities of all sensor fields can be detected and evaluated simultaneously. The SPR shifts (SPR shift) determined in this way result from subtracting the determined minima from the wavelength-dependent intensity curves before and after bringing the sensor fields occupied by the control ligands into contact with the corresponding target.

Example 8 Review of Eosin and GuaPhe-4 as universal binders

The control ligands were examined with a selection of proteins which were characterized by different properties with regard to size (molecular weight, MW), charge (isoelectric point, pl) and hydrophobicity (GRAVY, Kyte, J .; J. Mol. Biol. 157, 1982, 105- 132) are characterized.

Figure 1 shows the SPR shifts obtained.

The SPR results show that the control ligands investigated bind very frequently and are characterized by complementary binding behavior. Proteins that are strongly positively charged under the measurement conditions, such as lysozyme and avidin, tend to bind to the negatively charged eosin control ligand. Proteins with a negative charge (fibrinogen, thrombin) bind more strongly to the guanidiniumphenylalanine derivative.

Example 9 Array to array and field to field control

To reproduce the non-specific signals of the two control ligands, the data of a target (t-RNA) asst is summarized below, which was measured on two different arrays.

Shifts in nm mean

Array 1 GuaPhe 3.2 4.2 3.8 3.9 3.0 3.5 3.6

Array 2 GuaPhe 3.9 4.0 4.0 4.0 3.8 3.5 3.8

Array 1 eosin 0.2 0.1 0.1 0.2 0.2 0.1 0.1

Array 2 eosin 0.6 0.0 -0.1 0.0 0.1 0.1 0.1

The array to array control is carried out by comparing the averaged measurement signals of the control ligands examined. The data show that the mean values only deviate from each other by a maximum of 0.2 nm, which is within the measurement accuracy. The results obtained in the various measurements on the various arrays can thus be compared without restriction. The comparison of the measurement signals of the control ligands within a row provides a measure of the homogeneity of the measurement conditions across a single array. The control ligand eosin is not informative in this measurement because it does not interact sufficiently with the target. At GuaPhe, fluctuations of 15% around the mean can be observed. If one takes into account that these are individual measurements and only six measuring points are available via the array, the data presented can be used to infer homogeneous measuring conditions. A higher sensitivity can be achieved, for example, if the control ligands are present in a higher number of copies on the array.

Claims

Expectations 1. A method of controlling binding measurements on microarrays comprising the steps (a) providing a microarray for ligands to be tested on sensor fields; (b) applying at least one control ligand to at least one sensor field; (c) Measuring the binding at least between the control ligand and a mobile binding partner for the ligands to be tested, at least one control ligand being a non-specific universal binder. 2. The method according to claim 1, characterized in that at least two different control ligands are used. 3. The method according to claim 2, characterized in that at least one positively charged in combination with at least one negatively charged control ligand is used, these preferably each additionally having hydrophobic regions. 4. The method according to one or more of claims 1 to 3, characterized in that at least one control ligand is a dye or a dye derivative. 5. The method according to claim 4, characterized in that the control ligand is eosin or an eosin derivative. 6. The method according to one or more of claims 1 to 4, characterized in that at least one control ligand is a compound which contains several structural motifs for binding to the target. 7. The method according to claim 6, characterized in that there are several identical structural motifs. 8. The method according to claim 7, characterized in that the structural motifs are guanidinium groups. 9. The method according to claim 7 or 8, characterized in that the structural motifs are guanidiniumphenylalanine residues 10. The method according to any one of claims 6 to 9, characterized in that the structural motifs are interconnected by a framework. 11. The method according to claim 10, characterized in that the framework is an amino acid or a peptide. 12. The method according to claim 11, characterized in that the amino acid is lysine or the peptide is composed of lysine. 13. The method according to one or more of claims 1 to 10, characterized in that the sensor fields contain an organic layer as a binding matrix. 14. The method according to claim 13, characterized in that the organic layer can form a self-assembling monolayer (SAM). 15. The method according to claim 13 or 14, characterized in that the SAM comprises anchor molecules. 16. The method according to claim 15, characterized in that the SAM further comprises diluent molecules. 17. The method according to one or more of claims 13 to 16, characterized in that the binding matrix extends over the entire microarray surface on the sensor field side. 18. The method according to one or more of claims 1 to 17, characterized in that the microarray is structured on its surface facing the sensor surfaces. 19. The method according to one or more of claims 1 to 18, characterized in that the microarray comprises a structuring layer on a planar support. 20. The method according to one or more of claims 1 to 19, characterized in that the microarray contains a metal layer on the structuring layer and / or the carrier, on which the binding matrix is located on the outside. 21. The method according to claim 20, characterized in that the metal layer contains gold. 22. The method according to one or more of claims 1 to 21, characterized in that the microarray contains at least 96, preferably at least 4608, most preferably at least 9216 sensor fields. 23. The method according to one or more of claims 1 to 22, characterized in that the ligands and possibly the recurring structural motif are small organic molecules. 24. The method according to one or more of claims 1 to 23, characterized in that the ligands and / or the control ligands all have the same linker molecule (tag). 25. The method according to one or more of claims 1 to 24, characterized in that the bond measurement runs in parallel on several or all sensor fields. 26. The method according to one or more of claims 1 to 25, characterized in that the binding measurement is based on a radioactive, optical or electrical method. 27. The method according to one or more of claims 1 to 26, characterized in that the binding measurement takes place without markings. 28. The method according to claim 24 or 27, characterized in that an optical reflection method is used. 29. The method according to claim 28, characterized in that the surface plasmon resonance (SPR) is used for binding measurement. 30. Microarray for binding measurements of ligands on sensor fields with at least one control ligand on at least one sensor field, the control ligand being a compound which contains multiple structural motifs. 31. Microarray according to claim 30, characterized in that the structural motifs are connected to one another by a framework. 32. Microarray according to claim 30 or 31, characterized in that the structural motif is guanidiniumphenylalanine. 33. Microarray according to one or more of claims 30 to 32, characterized in that the framework is an amino acid or a peptide. 34. Microarray according to one or more of claims 30 to 33, characterized in that the amino acid is lysine or the peptide is composed of lysine. 35. Microarray for binding measurements of ligands on sensor fields with at least one positively charged and at least one negatively charged control ligand, these preferably each additionally having hydrophobic regions, on at least one sensor field in each case. 36. Procedure for finding control ligands as non-specific universal binders, comprising the steps (a) providing potential control ligands on sensor fields; (b) measuring the binding between the potential control ligands and several mobile binding partners; (c) Selection of control ligands based on their binding values. 37. The method according to claim 36, characterized in that step (c) follows (d) chemical linkage of several control ligands to a control ligand with several structural motifs. 38. The method according to claim 36 or 37, characterized in that at least 10 mobile binding partners are tested. 39. The method according to one or more of claims 36 to 38, characterized in that the mobile binding partners are selected from the group consisting of RNA, DNA and proteins, which differ as far as possible in their size and their charge and hydrophobicity properties. 40. Use of eosin or eosin derivatives as control ligands to control binding measurements on microarrays. 41. Compounds of the formula

where X = -OH, -NH ₂ , -NHA, where A is alkyl or acyl, or a bond for direct or indirect attachment to a solid phase, and their tautomers, isomers, enatiomers, mixtures or salts. 42. Use of compounds according to claim 41 as control ligands for controlling binding measurements on microarrays.