RU2182708C2 - Method of multiple parallel screening of binding specificity of biologically active compounds with nucleic acids using biochip (versions) - Google Patents

Abstract

FIELD: molecular biology, medicine, pharmacology, environment protection. SUBSTANCE: biochip with immobilized oligonucleotides is prepared and hybridization of these nucleotides with a mixture of nonself-complementary oligonucleotides labeled with fluorescent label is carried out. Double-stranded oligonucleotides are formed on biochip that's are subjected for melting recording data and biochip is washed out. The repeated hybridization is carried out with the same mixture of oligonucleotides labeled with fluorescent label followed by incubation of biochip with the compound to be studied. Double-stranded oligonucleotides are melted again on biochip being these nucleotides are in complex with biologically active compound to be studied. Data are recorded and melting points of double-stranded oligonucleotides are determined in the presence and absence of compound to be studied and difference of melting point is measured. Based on total data obtained the specificity of binding of compound to be studied is determined. The universal biochip where in its units all possible hexanucleotides are immobilized is used preferably. Fluorescent dye can be Texas red (Texas Red). Oligonucleotides are melted using a thermotable. The mass of experimental data is treated using the computer program preferably. Dye Hoechst 33258 or protein HU can be used as compound to be studied. By the second variant method involves incubation of biochip with fluorescent compound to be studied immediately after preparing biochip with oligonucleotides immobilized on its. Method ensures to carry out the comparative experimental analysis of relationship degree of chemical compound to all possible sequences of nucleic acid in the range of binding site. EFFECT: improved method of screening. 20 cl, 8 dwg, 3 ex

Description

 The invention relates to molecular biology, pharmacology, medicine and environmental protection and considers a method for determining the relative affinity for binding natural and chemical biologically active compounds with nucleic acids to all possible nucleotide sequences within a region of a given nucleic acid size. The specificity of the binding of chemical compounds to different nucleic acid sequences is analyzed using a universal oligonucleotide biochip, in whose cells all possible sequences of synthetic oligonucleotides of a certain length are immobilized.

 The invention also includes an original method for preparing a universal oligonucleotide biochip for the analysis of the specificity of binding of chemical compounds with different sequences, a method for recording, analyzing and interpreting the results.

State of the art
Many compounds are tested and analyzed for the ability to specifically bind to a specific DNA base pair sequence, and among these compounds there are replication, transcription and translation regulators - potential drugs for the treatment of certain diseases. It is also of interest to identify all nucleic acid sequences with which this or that compound entering the human body can specifically interact, since it is necessary to anticipate the possible negative consequences of such interactions if the corresponding sequences are part of the genes or regulatory regions of the human genome.

 Such screening involves testing compounds for binding specificity and the stability of their complexes with nucleic acids. The most comprehensive results were achieved during experiments on binding and melting, followed by comparative thermodynamic analysis of ligand complexes with nucleic acids of various sequences. However, such an analysis of complexes in solution requires a lot of effort and time.

Alternative, simpler and more practical practical procedures for analyzing the selectivity of binding of various compounds to specific DNA base pair sequences include two main approaches:
I. Nucleic acid protection as part of a complex with a chemical or natural compound from chemical modification, the action of nucleases, footprinting;
Maxam A., Mirzabekov A. and Gilbert W. (1976). In: Control of Ribosome Synthesis (Alfred Benzon Symp. IX), p. 139.

 Fox K.R. (1997) Methods Mol. Biol., 90, 1-22.

 Petri V. and Brenowitz M. (1997) Curr. Opin. Biotechnol., 8, 36-44.

II. Separation of complexes by electrophoresis;
Hamdan II, Skellern GG and Waigh RD (1998) Nucleic Acids Res., 26, 3053-3058.

 The principal drawback of the above approaches is a significant limitation on the number of analyzed sequences. These methods make it impossible to carry out a comparative experimental analysis of the degree of affinity of the chemical compound for all possible nucleic acid sequences within the binding site.

 These disadvantages are eliminated in the present invention.

SUMMARY OF THE INVENTION
The essence of the invention lies in the development of a method for multiple parallel screening of the specificity of binding of biologically active compounds to double-stranded nucleic acids using a biochip, including the production of a biochip with oligonucleotides immobilized on it, hybridization of oligonucleotides immobilized on a biochip with a mixture of fluorescently labeled oligonucleotides for simultaneous data recording and after washing the biochip repeat polar hybridizes with the same oligonucleotide mixture followed by incubation with the test compound. After repeated melting on a biochip, the differences in the melting points of duplexes in complex with the studied compound and in the free state are determined, based on which the specificity of the interaction of the studied compound with double-stranded nucleic acids is determined.

The method involves the use of a universal biochip, in the cells of which all possible hexanucleotides in an amount of 4 ⁶ = 4096 and a mixture of fluorescently labeled non-self-complementary oligonucleotides, each of which is complementary to an oligonucleotide immobilized on a universal biochip, are immobilized. As a fluorescent dye, for example, Texas Red can be used. The oligonucleotide duplexes immobilized on the biochip are melted using a special thermostol, and the melting is recorded using a fluorescence microscope equipped with a highly sensitive camera connected to a computer. An array of experimental data obtained during melting is processed using a special computer program that calculates the melting temperature T _{PL of} double-stranded oligonucleotides in the presence and absence of the studied compound and calculates the shift values of T _PL for all double-stranded oligonucleotides immobilized on a biochip, based on which it is determined specificity of interaction. Using this method, various compounds can be analyzed, in particular low molecular weight or high molecular weight chemical, natural or recombinant.

 A method for multiple parallel screening of the specificity of binding of biologically active compounds to single-stranded nucleic acids using a biochip involves the manufacture of a biochip with oligonucleotides immobilized thereon; incubation of a biochip with a fluorescent test compound; melting on a biochip of complexes of oligonucleotides with a fluorescent test compound with determining the melting temperature and determining the specificity of binding of the test compound to nucleic acids.

 The method also involves the use of a universal biochip. The test compound may be naturally fluorescent or carry a fluorescent label. FITC dye can be used as a fluorescent label. The oligonucleotide complexes with the studied substance are melted using a special thermostol, and the melting is recorded using a fluorescence microscope equipped with a highly sensitive camera connected to a computer. An array of experimental data obtained during melting is processed using a special computer program that further determines the melting points of the complexes of the studied compound with all oligonucleotides immobilized on a biochip, on the basis of which the specificity of the interaction is determined. Using this method, it is possible to analyze various compounds, in particular low molecular weight or high molecular weight chemical, natural or recombinant.

List of figures
Figure 1. Nonequilibrium melting curves of DNA duplexes on biochips, taken in the presence of (A) and in the absence of (B) Hoechst. The double-stranded oligonucleotide gel-MTTTTSGM-3 '/ 5'-MSGAAAAM-TR-3' was formed by hybridization of a 5'-MM (A / G) MM (A / G) MM-TexasRed oligonucleotide mixture on a universal biochip. The nonequilibrium temperature of the decrease in Tm was determined as the temperature at which the hybridization signal decreases to 1/10 of the initial value, measured at 0 ^o C. The degree of affinity of Hoechst to the duplex was determined as ΔT _PL = T _PLV -T _PLA .

Figure 2. The increase in the melting temperature ΔT _PL caused by the binding of Hoechst with duplexes formed on the microchip. A universal 6-dimensional biochip was hybridized sequentially with two mixtures of fluorescently labeled oligonucleotides 5'-MM (A / G) MM (A / G) MM-TR and 5'-MM (A / C) MM (A / C) MM-TR in the absence and in the presence of Hoechst, after which the melting curves of the duplexes were recorded.

Duplexes on the biochip are arranged as a two-dimensional matrix in such a way that the 5'-halves of the 6-mers that make up the middle of the oligonucleotide are arranged line by line, and the 3'-halves are arranged in columns. The value ΔT _PL for duplexes is shown by the value of the black sector inside the circle for the corresponding matrix element.

Figure 3. A histogram showing the number of duplexes N with the indicated value ΔT _pl . About 300 duplexes have a strong affinity for Hoechst, the corresponding values of ΔT _PL > 1.5 ^o C (indicated by arrow).

 Figure 4. Nonequilibrium melting curves of a double-stranded oligonucleotide on a universal biochip in the absence and presence of HU protein.

 The sequence of the oligonucleotide immobilized on a biochip: 5'-MAGTCTGM-3 '.

Figure 5. A histogram showing the number of duplexes showing a certain ΔT _pl .

Figure 6. Average ΔT _PL values for duplexes containing different sequences from three DNA bases.

 Figure 7. Nonequilibrium melting curves of complexes of HU protein labeled with FITC with immobilized oligonucleotides on a universal biochip. Oligonucleotide sequences are indicated in the figure.

 Figure 8. The dependence of the melting temperature of the complexes of protein HU - DNA from the A / T composition of the immobilized single-stranded oligonucleotide.

Disclosure of invention
The invention consists in conducting a massive screening of the specificity of binding of natural and chemical compounds to DNA by analyzing the melting points or shifts in the melting points of the complexes of these compounds with all possible sequences of oligonucleotide duplexes or single-stranded oligonucleotides immobilized in gel cells of a universal biochip.

In different cells of the biochip, oligonucleotides having the same length but differing in base sequences are immobilized. Moreover, in each cell of the biochip, an oligonucleotide with a certain base sequence is immobilized, and the number of cells of the biochip is equal to the number of all possible base sequences for a nucleotide of a given length. Using this biochip, one can quantitatively analyze the specificity of the binding of natural or chemical compounds to all the nucleotide sequences of both single-stranded and part of the DNA double helix represented in its cells:
1. Determination of the specificity of binding of natural or chemical compounds to all sequences of oligonucleotide duplexes formed in biochip cells.

 When analyzing the specificity of binding to oligonucleotide duplexes, a solution containing a mixture of non-complementary fluorescently labeled oligonucleotides, each of which is complementary to one of the nucleotides immobilized in the cells of the biochip, is first added to the hybridization chamber of the universal biochip and forms a double-stranded complex with it. The formation of complexes or their melting is recorded by increasing or decreasing intensity of the fluorescent signal simultaneously in all cells of the biochip.

 Hybridization is carried out at a low temperature, after which the biochip is gradually heated, registering the melting curves in descending order of the fluorescent signals, and the melting temperatures of all duplexes formed by this hybridization mixture of oligonucleotides are determined.

 After that, repeated hybridization is carried out at low temperature and a solution containing a chemical compound that is tested for specificity of binding with nucleic acids is added to the hybridization chamber. The test compound penetrates the biochip cells and forms complexes with oligonucleotide duplexes, characterized in the general case by different binding energies. Next, the biochip heats up at the same speed as in the case of the melting of free duplexes, and the melting curves of all duplexes are re-recorded.

 If a chemical compound interacts with some double-stranded DNA sequences, then the melting curves recorded by decreasing fluorescence signal in the biochip cells containing oligonucleotides with the corresponding sequences will be shifted to higher or lower temperatures. Such shifts of the melting curves indicate the presence of interaction of the test compound with the corresponding DNA base sequences: a positive shift indicates an interaction that stabilizes the structure of the double helix, and a negative shift indicates interaction that destabilizes the structure of the double helix.

The shifts of the melting curves determine the shifts of the melting temperatures ΔT _PL , which are thermodynamic parameters characterizing the specificity of the binding of a chemical compound to the corresponding sequences of base pairs in the DNA double helix. The sets of ΔТ _pl values determined for different mixtures of non-complementary fluorescently labeled oligonucleotides are combined into an array of thermodynamic parameters characterizing the affinity of the test compound for different sequences of base pairs in the DNA double helix.

 2. Determination of the specificity of binding of natural or chemical compounds to all sequences of single-stranded oligodeoxyribonucleotides immobilized in biochip cells.

 When analyzing the specificity of binding to single-stranded oligonucleotides, a solution containing the molecules of the test compound pre-labeled with a fluorescent label is added to the hybridization chamber of the universal biochip described above. At a sufficiently low hybridization temperature, the molecules of the test compound penetrate into all cells of the biochip and form complexes with those oligonucleotides to which binding affinity is detected. As a result, fluorescent signals appear in the corresponding cells, the intensity distribution of which can serve as a complex characteristic of the affinity of the test compound for the presented sequences of oligonucleotides.

Next, a gradual increase in the temperature of the biochip is carried out, simultaneously recording the dissociation of the complexes of the test compound with all oligonucleotides by the decrease in the fluorescence intensity in each cell. The melting curves obtained determine the melting temperature T _{PL the} resulting complexes of the test compound with oligonucleotides.

 The objective of the present invention is to create an accelerated method that allows for mass screening of the specificity of binding of natural and chemical compounds to DNA. It is proposed that screening for binding specificity be performed using a universal biochip. In different cells of the biochip, oligonucleotides having the same length but differing in base sequences are immobilized. Moreover, in each cell of the biochip, an oligonucleotide with a certain base sequence is immobilized, and the number of cells of the biochip is equal to the number of all possible base sequences for a nucleotide of a given length.

 Using such a biochip, it is proposed to quantitatively analyze the specificity of the binding of natural or chemical compounds to all nucleotide sequences represented in its cells, both single-stranded and included in the DNA double helix. The stereospecificity or affinity of the tested compounds for specific base sequences is determined by the melting points of the complexes of these compounds with all possible sequences of oligonucleotide duplexes or single-stranded oligonucleotides immobilized in gel cells of a universal biochip. An additional method for assessing the stereospecificity of binding to single-stranded oligonucleotides is based on measuring the relative amounts of bound molecules of the test compounds in all cells of the biochip.

A distinctive feature of the proposed method is a large number of thermodynamic parameters (shifts in melting temperatures ΔT _pl or melting temperatures T _pl ) characterizing the interaction of the test compound with a complete enumeration of all possible sequences of a universal biochip.

 For a multiple parallel analysis of the specificity of binding of biologically active compounds with nucleic acids, a universal hexadeoxyribonucleotide biochip has been developed. The use of a universal biochip was demonstrated by the analysis of binding of Hoechst 33258 to various 6-dimensional duplexes, analysis of binding of HU protein to various hexameric duplexes, and analysis of binding of HU protein to various hexameric single-stranded oligonucleotides.

 In the case of using a universal hexadeoxyribonucleotide biochip for multiple parallel analysis of the binding of Hoechst 33258 to various 6-dimensional duplexes, a thermodynamic analysis of a number of such complexes was performed and the sequences responsible for the specific interaction were determined. Their quantitative comparison was carried out using a statistical analysis of a large number of melting curves of duplexes formed on a biochip in the presence and absence of dye. This approach may be extended to screen for binding specificity relative to the sequence of base pairs of other ligands, including proteins.

 In the case of using the universal hexadeoxyribonucleotide biochip for multiple parallel analysis of the binding of the HU protein to various hexamer duplexes, 1024 experimental melting curves of double-stranded oligonucleotides in the presence and absence of the HU protein were obtained. Previously, it was suggested that this protein does not have strong and pronounced binding specificity with certain double-stranded DNA sequences, however, a statistical analysis of 1024 melting curves revealed a number of binding patterns and detected DNA sequences with which HU binds preferentially.

 In the case of using the universal hexadeoxyribonucleotide biochip for multiple parallel analysis of the binding of the HU protein to various hexameric single-stranded oligonucleotides, the universal oligonucleotide biochip was not converted to a double-stranded state by adding a mixture of fluorescently labeled oligonucleotides, but fluorescently fluorescent. Then, direct binding of the fluorescently labeled protein to oligonucleotides immobilized on a biochip was measured, after which the HU protein complexes with single-stranded oligonucleotides were melted. The data obtained allowed us to determine the specificity of the binding of the HU protein directly to single-stranded DNA sequences.

Examples
Example 1. Multiple parallel analysis of the specificity of binding of Hoechst 33258 to DNA using a universal oligonucleotide biochip
Chemical compounds
Hoechst 33258 (E ₃₄₅ = 41000) was provided by H. Loeve (Hoechst, Germany) and was used without further purification.

4096 octadeoxyribonucleotides (CyberSyn, USA) were synthesized using standard procedures and the 3'-C (7) CPG amino modifier (Glen Research, USA) and used to prepare universal biochips. 8-measures had the following structure: 5'-NH ₂ -MNNNNNNM-3 ', where M is a mixture of four bases in a ratio of 1: 1: 1: 1 in both the 3'- and 5'-terminal positions; N is one of four bases representing in the middle part all 4096 possible 6-measures, the NH ₂ -aminolinker used to immobilize 8-measures in polyacrylamide gel cells on a biochip. Two types of mixtures of 8-mers 5'-MM (A / G) MM (A / G) MM-NH ₂ -3 'and 5'-MM (A / C) MM (A / C) MM-3'-NH ₂ were synthesized using a DNA / RNA synthesizer (Applied Biosystems, model 394) using standard procedures and a 3'-C (7) CPG amine modifier (Glen Research, USA). Mixtures of 8-mers were fluorescently labeled with Texas Red sulfonyl chloride dye (TexRed, Molecular Probes, USA) according to the manufacturer's protocol.

Universal biochip
Universal biochip was prepared in two stages. First, using the photopolymerization procedure (Guschin D., Yershov G., Zaslavsky A., Gemmel A., Shick V., Proudnikov D., Arenkov P. and Mirzabekov A. (1997) Anal. Biochem., 250, 203-211 ) prepared a lattice of 4200 cells of 5% polyacrylamide gel (60x70 cells 100x100x20 μm in size) separated by 200 μm hydrophobic glass strips. Then, drops of 1 mM aqueous solution of an oligonucleotide of 1 nL volume were applied to each well (Yershov G., Barsky V., Belgovskiy A., Kirillov E., Kreindlin E., Ivanov I., Parinov S., Guschin D., Drobishev A. , Dubiley S. and Mirzabekov A. (1996) Proc. Nat. Acad. Sci. USA, 93, 4913-4918), the oligonucleotides were immobilized by reductive amination of their amino groups with aldehyde gel groups (Timofeev E., Kochetkova SV, Mirzabekov AD and Florentiev VL (1996) Nucleic Acids Res., 24, 3142-3148).

Hybridization and removal of melting curves
Hybridization of the universal biochip with a mixture of fluorescently labeled 8-mers was carried out in a 200 μl hybridization cell at 0 ^° C for 24 hours. Hybridization solutions contained 200 mM oligonucleotides, 1 M NaCl, 10 mM NaHP04, 5 mM EDTA, pH 6.8, 0.1 % Tween-20. After hybridization, the biochip along with the cuvette was placed on a thermostatic stage of the microscope, where the nonequilibrium melting curves were recorded for all elements of the biochip at a temperature increase from 0 ^o C to 50 ^o C at a speed of 2 ^o C / min (Fotin AV, Drobyshev AL, Proudnikov DY, Perov AN and Mirzabekov AD (1998) Nucleic Acids Res., 26, 1515-1521). After taking the melting curves of duplexes in the absence of Hoechst 33258, the biochip was washed with water from fluorescently labeled oligonucleotides. The second round of hybridization-melting was carried out under the same conditions, but in the presence of 10 mm Hoechst 33258.

 Hybridization of the 8-dimensional biochip with a 1 mM fluorescently labeled 20-mer solution was also carried out under the same conditions both in the absence and in the presence of 2 mM Hoechst 33258. Equilibrium melting curves were recorded on the biochip as described (Fotin AV, Drobyshev AL Proudnikov DY, Perov AN and Mirzabekov AD (1998) Nucleic Acids Res. 26, 1515-1521).

 All measurements when taking the melting curves were carried out using an automatic epifluorescence microscope with a field of 3.5x3.5 mm, equipped with a CCD camera, a Peltier thermocouple and a temperature sensor connected to a computer equipped with an automatic data processing program (Fotin AV, Drobyshev AL, Proudnikov DY, Perov AN and Mirzabekov AD (1998) Nucleic Acids Res., 26, 1515-1521). The fluorescence intensity was recorded for each temperature when scanning a universal biochip in several fields containing 100 cells of the gel. The scanning system consisted of a two-coordinate table with guides, a micromotor and a controller (Newport, USA). To control the experiment and data processing, special software was developed using the Lab VIEW virtual instrumentation interface (National Instruments, USA).

Multiple parallel measurements of the binding of Hoechst 33258 to DNA duplexes on a universal biochip
The universal 6-dimensional biochip contained all possible 4096 (4 ⁶ = 4096) single-stranded hexadeoxyribonucleotides NNNNNN (N is one of four bases) immobilized in individual gel cells. To these 6-measures, as a core, two 8-measures are attached from both 3'- and 5'-ends, having a 5'-gel-MNNNNNNM-3 'type structure and composed of a mixture of four M bases in the ratio 1: 1: 1: 1.

Hoechst 33258 does not bind to single-stranded DNA and does not show significant specificity when strongly bound to double-stranded DNA. DNA duplexes in the complexes are stabilized by a dye and have a higher melting point T _pl . The specificity of dye binding can be assessed by the increase in T _PL in duplexes upon ligand binding. For these measurements, 4096 single-stranded oligonucleotides on a universal biochip must be completed in double-helix form. This can be achieved by hybridization of the biochip with a mixture of 4096 fluorescently labeled 8-mers having the structure 5'-MNNNNNM-TR-3 '.

 Symmetrical duplexes having the same base pairs on the gel side and on the dye side have been shown, for example 5'-gel-AGTCTCGA-3 '/ 3'-TR-TCAGAGCT-5' and 5'-AGTCTCGA-TR-3 '/ 3'-TCAGAGCT-gel-5 ', have similar melting curves both in the absence and in the presence of a ligand (data not shown). 6-dimensional universal biochips can contain 2080 different 6-dimensional duplexes. Among them, there are 64 palindromic self-complementary 6-measures that cannot be obtained on the biochip due to their self-hybridization in solution. In order to avoid competitive self-hybridization of oligonucleotides in solution and on a biochip, 6 mixtures containing non-self-complementary oligonucleotides can be selected: MMM (A / G) (A / G) MMM, MMM (A / C) (A / C) MMM, MM ( A / G) MM (A / G) MM, MM (A / C) MM (A / C) MM, M (A / G) MMMM (A / G) M, M (A / C) MMMM (A / CM. Such a set of mixtures takes into account all possible (2016) non-self-complementary duplexes.

 The most informative data can be obtained with two non-self-complementary mixtures: 5'-MM (A / G) MM (A / G) MM-TR-3 'and 5'-MM (A / C) MM (A / C) MM- TR-3 '. Each of these 8-mer mixtures contains 1,024 different central 6-mers, however together they combine 1,536 central 6-mers, since structures of the MAMAMMAM type are present in both mixtures. These two mixtures include the AT-rich duplexes needed to detect most of the sequences specific for Hoechst binding.

 The biochip was hybridized with each of these mixtures of fluorescently labeled 8-mer. The nonequilibrium melting curves of all duplexes formed on the biochip were recorded in the process of increasing temperature. The hybridization and melting processes were repeated on the same biochip in the presence of Hoechst in the hybridization buffer.

In FIG. 1 shows two typical melting curves taken for a duplex in the presence and absence of Hoechst. The temperature at which the hybridization signal decreased by 10 times compared with the initial level determined at 0 ^o C, was considered as a nonequilibrium melting temperature T _pl . The presence of a ligand shifts the melting curves toward higher temperatures, giving an increase in the melting temperature ΔТ _pl , depending on the structure of the duplex, up to 13 ^o С. Thus, the value ΔТ _pl was used to assess the degree of affinity of Hoechst for this duplex.

Figure 2 shows the ΔT _pl values for 1680 oligonucleotides that formed duplexes on a universal biochip during its hybridization with two types of mixtures of fluorescently labeled oligonucleotides. 112 oligonucleotides were excluded from consideration due to weak hybridization signal.

Hoechst binding specific sequence analysis
The data in FIG. 2 shows that Hoechst 33258 has a stronger affinity for AT-rich sequences compared to GC-rich ones. Characterization of the types of sequences specific for ligand binding requires a more detailed analysis. The following computer-based approach was developed to analyze large amounts of experimental data obtained using a universal biochip.

A histogram of a large number of duplexes having characteristic values of ΔT _{PL is} shown in FIG. 3. The histogram contains a sharp peak at about 0 ^° C, corresponding to duplexes that do not stabilize upon Hoechst binding, and a wide peak starting at 1.5 ^° C with a long arm of up to 13 ^° C. Thus, the excess of ΔТ _pl exceeds the level of 1.5 ^o C can be considered as a sign of binding of Hoechst to this duplex. 300 such duplexes were detected.

 Example 2. The study of the specificity of binding of HU protein to double-stranded DNA by multiple parallel analysis on a universal oligonucleotide biochip.

 Chemical compounds.

 Octadeoxyribonucleotides for the universal biochip were also manufactured as described in Example 1.

A mixture of 8-mers 5'-MM (A / C) MM (A / C) MM-NH ₂ -3 'was synthesized as described in Example 1, and was labeled with Texan Red (TexRed) in the same manner.

 A universal biochip was also prepared as described in example 1.

 Protein HU.

HU protein was isolated from E. coli strain JRY1 according to the previously described procedure (Rouviere-Yaniv, J. & Kjeldgaard, NO (1979) FEBS Letters, 106, 297-300). Protein concentration was determined by absorbance at 230 nm using an extinction coefficient A of ₂₃₀ = 2.3, corresponding to a HU protein concentration of 1 mg / ml.

 Hybridization and removal of melting curves.

 Hybridization of the universal biochip with a mixture of fluorescently labeled 8-mers was carried out as in Example 1. The hybridization solution contained 200 μM oligonucleotides, 0.1 M NaCI, 20 mM Tris (pH 7.2), 5 mM EDTA, and 0.1% Tween 20.

 Nonequilibrium melting curves were also taken as described in example 1.

After measuring the melting curves of duplexes in the absence of HU protein, the biochip was washed with water from fluorescently labeled oligonucleotides. The second round of hybridization and melting was carried out under the same conditions, but after hybridization, the oligonucleotide solution was replaced with a buffer containing HU protein (0.55 mg / ml) and incubated for 12 hours at 0 ^° C. Then, melting was performed.

 All measurements on a fluorescence microscope were also performed as described in example 1.

 Studying the specificity of binding of HU protein to double-stranded DNA on a universal biochip.

It is known that HU protein does not exhibit noticeable specificity upon binding to double-stranded DNA. The specificity of the HU-DNA protein complexes was evaluated using statistical analysis of multiple data containing duplex melting curves in the absence and presence of HU protein. To conduct such experiments, 1024 octamers immobilized on a biochip were converted into a double-stranded form as described in Example 1. To convert oligonucleotides into a double-stranded form, a non-self-complementary mixture of oligonucleotides with the general formula: 5'-MM (A / C) MM ( A / C) MM-NH ₂ -3'-TR containing 1024 oligonucleotides labeled with Texas Red.

 The biochip was hybridized with a mixture of fluorescently labeled octamers in a buffer, as described above. Then the mixture of octamers in the hybridization chamber was replaced with a clean buffer. Nonequilibrium melting curves for all duplexes on the biochip were obtained during melting of the biochip on a thermostol.

 After washing the biochip, hybridization and melting were repeated again. But this time, after hybridization, the oligonucleotide mixture was replaced with a solution of HU protein (0.55 mg / ml) in the same buffer and incubated for 12 hours. After this, the melting was carried out in the same way as for the first time.

где: f(T) - измеряемый сигнал,
А+В - начальный сигнал,
В - конечный сигнал,
Т - температура (К),
Т_пл - температура плавления,
N - фактор неоперативности.Thus, 1024 melting curves of octamers in the absence of HU protein and 1024 melting curves in the presence of HU protein were obtained. Figure 4 shows two such melting curves for one AGTCTG oligonucleotide. For the analysis of the curves used a special computer program. All experimental melting curves were approximated by the least squares method using the following formula:

where: f (T) is the measured signal,
A + B is the initial signal,
B is the final signal
T is the temperature (K),
T _PL - melting point,
N is the factor of inoperability.

After approximation, 1024 melting points of To were obtained for all duplexes in the presence and absence of the HU protein. Then an array was created of 1024 values ΔT _pl = T _pl (protein) - T _pl (without protein), where ΔT _pl is the shift in the melting temperature of the oligonucleotide in the presence of protein. Data for 14 duplexes were excluded from consideration due to a weak signal. An array of 1010 ΔT ₀ values was subjected to statistical analysis.

The values of ΔT _PL were located in the form of a histogram presented in figure 5. The histogram clearly indicates the presence of two types of complexes of the HU protein with oligonucleotides. The first, the main complex has a positive value of the shift in the melting temperature ΔT _pl , the average shift is about +3 ^o C. The second, minor complex is described by about 150 values of ΔT _pl and has an average negative shift of about -3 ^o C.

 To find the difference between the major and minor complexes, a special analysis was carried out. It turned out that the minor complex contains long A / T sequences consisting of 4, 5 and 6 bases. In the main A / T complex, the sequences are shorter (2, 3, and 4 bases).

The presence of a large data array of 1010 ΔТ ₀ values allowed us to study the specificity of the binding of the HU protein to double-stranded DNA. It was previously thought that this protein binds to DNA non-specifically.

 The melting temperature shifts of the HU complexes with DNA were calculated for all sequences from three bases on the biochip. The data are presented in Fig.6.

 Statistical analysis showed that there are three sequences: AAG, AGA and TAA, the binding of which the HU protein is preferred. This result was obtained solely through the use of a universal oligonucleotide biochip and the ability to obtain and process more than 1000 melting curves simultaneously.

 Example 3. The study of the specificity of binding of HU protein to single-stranded DNA fragments by multiple parallel analysis on a universal oligonucleotide biochip.

Universal biochip
To study the specificity of the binding of the HU protein to single-stranded DNA fragments, the same universal biochip was used as described in Examples 1 and 2, and the fluorescently labeled oligonucleotides described in Examples 1 and 2 were not used.

HU protein
The HU protein described in Example 2 was used. To measure binding to single-stranded DNA sequences, the protein was labeled with FITC fluorescent dye in accordance with standard procedures (Hermanson GT (1996) Bioconjugate Techniques, Academic Press, Amsterdam).

Hybridization and Melting
Hybridization and melting of the protein was carried out in the same hybridization chamber as described in Example 2. The hybridization solution contained HU protein (0.55 mg / ml), 0.020 M NaCl, 20 mM Tris (pH 7.2), 5 mM EDTA and 0.1% Tween 20. Hybridization carried out at 0 ^o C for 24 hours.

 All melting experiments were carried out as described in examples 1 and 2, but a special filter was used to measure the fluorescence intensity of FITS.

Studying the specificity of binding of HU protein to single-stranded DNA sequences on a universal biochip
The HU protein is known to bind to single-stranded DNA fragments. In previous work on the binding of a protein to single-stranded DNA, DNA fragments of 30-40 nucleotides in length were used. Such fragments may already have a secondary structure and, accordingly, the binding constant of a protein other than single-stranded DNA will be measured.

 Only in the case of immobilization of octamers on the biochip can it be guaranteed that the protein will bind to truly single-stranded DNA.

 The HU protein was hybridized with a universal biochip for 24 hours in the buffer described above. The melting of the protein-DNA complex was carried out in the same manner as described in example 2. However, in this case, during heating, not the melting of double-stranded oligonucleotides was recorded, but the thermal decomposition of the protein-DNA complex. The process of such melting was recorded by the fluorescence of FITS, and not by the fluorescence of Texas Red, as in the case of oligonucleotides. In FIG. Figure 7 shows some experimental melting curves of HU protein complexes with single-stranded oligonucleotides.

 Melting points were calculated for all 4096 protein-DNA complexes using the method described in example 2. The resulting data array was examined to find the specificity of the binding of the HU protein to different sequences of single-stranded oligonucleotides. On Fig presents the result of statistical processing of these data.

 The data shown in FIG. 8 show that among all possible hexamers, the HU protein preferably binds to those where 6 or 5 of the 6 bases are G or C. All other sequences have approximately the same HU binding constants. Thus, the use of a universal biochip revealed a specific binding specificity of the HU protein with single-stranded DNA. Further analysis showed that the most preferred DNA binding sequence for this protein is GCGC.

Claims (20)

1. A method for multiple parallel screening of the specificity of binding of biologically active compounds to double-stranded nucleic acids using a biochip, including: (a) manufacturing a biochip with oligonucleotides immobilized on it; (b) hybridization of oligonucleotides immobilized on a biochip with a mixture of fluorescently labeled non-self-complementary oligonucleotides, leading to the formation of double-stranded oligonucleotides on the biochip; (c) melting double-stranded oligonucleotides on a biochip with simultaneous data recording; (d) washing the biochip; (e) re-hybridization with the same mixture of fluorescently labeled oligonucleotides as in step (b); (e) incubation of the biochip with the test compound; (g) re-melting on a biochip of double-stranded oligonucleotides in combination with the biologically active compound under study, while simultaneously recording data similar to stage (c); (h) determining the melting temperature T _{PL of} double-stranded oligonucleotides in the presence and absence of the studied compound and calculating the difference in melting temperatures ΔT _pl for all double-stranded oligonucleotides immobilized on a biochip using a computer program; (g) determining the binding specificity of the test compound with a nucleic acid-based analysis of the data array comprising the difference ΔT _mp melting temperatures of oligonucleotides complexed with the test compound in the free form. 2. The method according to p. 1, characterized in that they use a universal biochip, in the cells of which all possible hexanucleotides are immobilized in an amount of 4 ⁶ = 4096.  3. The method according to p. 1 or 2, characterized in that at the stage (b) or (d) for hybridization use a mixture of fluorescently labeled non-self-complementary oligonucleotides, each of which is complementary to an oligonucleotide immobilized on a universal biochip.  4. The method according to any one of paragraphs. 1-3, characterized in that the fluorescent dye is Texas Red (Texas Red).  5. The method according to any one of paragraphs. 1-4, characterized in that the melting of oligonucleotides immobilized on a biochip is carried out using a thermostol.  6. The method according to any one of paragraphs. 1-5, characterized in that the registration of melting is carried out using a fluorescence microscope equipped with a highly sensitive television camera connected to a computer.  7. The method according to p. 6, characterized in that the array of experimental data obtained during melting is processed using a computer program.  8. The method according to any one of paragraphs. 1-7, characterized in that the studied compound includes a low molecular weight or high molecular weight chemical, natural or recombinant compound.  9. The method according to any one of paragraphs. 1-8, characterized in that the test compound is a dye Hoechst 33258.  10. The method according to any one of paragraphs. 1-8, characterized in that the test compound is an HU protein. 11. A method for multiple parallel screening of the specificity of binding of biologically active compounds to single-stranded nucleic acids using a biochip, including: (a) manufacturing a biochip with oligonucleotides immobilized on it; (b) incubation of the biochip with the fluorescent test compound; (c) melting on a biochip of complexes of oligonucleotides with a fluorescent test compound with simultaneous data recording; (d) determination of the melting temperature T _{pl of the} complexes of the studied compound with all oligonucleotides immobilized on a biochip using a computer program; (e) determination of the binding specificity of the studied compound with nucleic acids based on the analysis of a data array containing the melting temperature T _{pl of the} studied compound with all oligonucleotides immobilized on a biochip. 12. The method according to p. 11, characterized in that they use a universal biochip, in the cells of which all possible hexanucleotides are immobilized in an amount of 4 ⁶ = 4096.  13. The method according to p. 11 or 12, characterized in that a fluorescent label is previously introduced into the test compound.  14. The method according to p. 13, wherein the fluorescent label is FITC.  15. The method according to p. 11 or 12, characterized in that the studied compound is naturally fluorescent.  16. The method according to any one of paragraphs. 11-15, characterized in that the melting of oligonucleotides immobilized on a biochip is carried out using a thermostol.  17. The method according to any one of paragraphs. 11-16, characterized in that the registration of melting is carried out using a fluorescence microscope equipped with a highly sensitive camera connected to a computer.  18. The method according to any one of paragraphs. 11-17, characterized in that the array of experimental data obtained during melting is processed using a computer program.  19. The method according to any one of paragraphs. 11-18, characterized in that the studied compound includes a low molecular weight or high molecular weight chemical, natural or recombinant compound.  20. The method according to any one of paragraphs. 11-19, characterized in that the test compound is an HU protein.

Images