WO2003066890A2

WO2003066890A2 - Method for determining the methylation pattern of dna

Info

Publication number: WO2003066890A2
Application number: PCT/EP2003/001035
Authority: WO
Inventors: Jochen Muth; Andreas Kappel; Heike Behrensdorf
Original assignee: Nanogen Recognomics Gmbh
Priority date: 2002-02-04
Filing date: 2003-02-03
Publication date: 2003-08-14
Also published as: WO2003066890A3; DE10204566A1; AU2003210200A1

Abstract

The invention relates to a method for determining the methylation pattern in DNA-sequences whereby non-methylated cytosine in the DNA-sequence is chemically converted to uracil and the thus treated DNA-sequence is hybridised by a detector-DNA-sequence. The hybrids are fixed in a locally dissolved manner on a surface. Subsequently, the detection of guanine/uracil-mispairing occurs by using substances detected by said mispairings.

Description

Method for determining the methylation pattern of DNA

description

The present invention relates to a method for determining the methylation pattern of DNA, in particular using electronically addressable biochips.

The methylation of DNA plays an important role in nature. In particular, the activity of genes is regulated with the help of the methylation of individual cytosines in the DNA. In order to be able to better assess the activity and regulation of individual genes, there is great interest in methods with which such methylations can be easily detected. For this reason, several methods have now been developed which are suitable for determining the methylation rate and the methylation pattern in DNA.

One method uses restriction enzymes that cannot recognize methylated sites as substrates (e.g. described in Analysis of DNA methylation and DNAse I sensitivity in gene-specific regions of chromatin; Ginder, Gσrdon; Methods Hematol. Vol. 20 1989 p. 111 -123). With these methods, however, only methylations that occur at the restriction sites can be detected.

In addition, there are other methods in which the methylation of DNA sequences is determined using PCR (MethylLight: a high-througput assay to measure DNA methylation; Cindy A. Eads, Kathleen D. Danenberg, Kazuyuki Kawakami, et. Al .; Nucleic Acids Rearch 2000 Vol. 28. No 8.). With this method, however, a corresponding dye-labeled sample must be synthesized for each methylation position, which is technically very complex. In addition to the biotechnological approaches described, the methylation of DNA can also be determined quantitatively by chemical reactions with chloroacetaldehyde. The detection of methylation is carried out in this method using a spectrometer in the range between 300 and 400 nm. (Quantification of 5-methylcytosine in DNA by the Chloroacetaidehyde Reaction; Biotechniques 27, 744-752 October 1999, Oakeley EJ, Schmitt F, Jost JP )

In an alternative method, the methylation pattern is determined using bisulfite SSCP, whereby unmethylated cytosine is converted to uracil. The sequence of the DNA sample examined can then be determined with the aid of a DNA sequencer (Quantitative DNA Methylation analysis by fluorescent ploymerase chain reaction Single Strand conformation polymorphism using an automated DNA sequencer; Hiromu Suzuki, Fumio Itoh, Minoru Toyota Tekefumi KiKuchi Electrophoresis 200021, 904-908).

The invention is therefore based on the object of providing a simple method for determining the methylation pattern of DNA sequences.

The task is solved by a process that comprises the following process steps:

ä) Chemical conversion of unmethylated cytosine contained in starting DNA to uracil, the sequence of the starting DNA being known, b) hybridization of single-stranded DNA treated according to step a) with single-stranded detector DNA sequences which are complementary to the sequence the starting DNA or partial sequences thereof, whereby the guanine bases can be exchanged for adenine bases, c) fixation of single-stranded detector DNA or of single-stranded DNA sequences treated after step a) before or during hybridization or of heteroduplicates from detector DNA and DNA treated after step a) after or during the hybridization, the fixation being spatially resolved is carried out on a support so that each detector DNA sequence can be assigned a specific location, d) incubation with a guanine / uracil or adenine / methylcytosine

Mismatch-detecting substrate and detection of the substrate bonds.

For the purposes of the present invention, the term “starting DNA” means a DNA or a DNA fragment with a known sequence, the methylation pattern of which is to be determined.

The term “detector DNA” includes DNA polynucleotides that are complementary to partial sequences of the starting DNA sequence, all guanine bases preferably being replaced by adenine bases.

The term “spatially resolved” fixation means that a specific detector DNA sequence can be assigned a defined location on a carrier. The carrier can be loaded passively, for example using pipetting devices or actively, via electronic addressing Methods for the production of such carriers are known to the person skilled in the art, and corresponding methods for producing electronically addressable carrier surfaces are described, for example, in Radtkey et al., Nucl. Acids Res. 28, 2000, e17; RG Sonowsky et al., Proc. Natl. Acad. Sci. USA, 1119-1123; 1994 PN Gilles et al., Nature Biotechnol. 17, 365-370, 1999, the production itself also being able to be carried out by means of electronic addressing.

The starting DNA to be examined can e.g. B. a gene or gene fragment isolated from the cell nucleus, the nucleotide sequence of which is known. The starting DNA can be further fragmented for better handling. The fragmentation can be unspecific e.g. B. by shear forces or specifically z. B. done by a restriction digest. The DNA fragments thus obtained can then be denatured for chemical conversion.

The chemical conversion of unmethylated cytosine in the starting DNA sequences is preferably carried out using bisulfites, whereby Methylcytosine is not implemented in the reaction. The DNA sequence treated in this way can now be hybridized with a detector DNA sequence, it being possible for the detector DNA sequence to be complementary to the chemically untreated starting sequence. The hybridization of the chemically treated DNA sequence with the detector DNA sequence consequently results in guanine / uracil mismatches at the sequence positions at which an unmethylated cytosine was present in the starting sequence.

In a preferred embodiment, however, detector DNA sequences are used whose sequence is complementary to the chemically untreated starting sequence, the guanine bases being replaced by adenine bases. Adenine / methylcytosine mismatches are thus obtained when the detector DNA sequence is hybridized with the chemically treated DNA sequence. This variant of the method is particularly suitable for determining the methylation pattern of low-methylated DNA sequences, as are usually found in mammals or plants. Mammalian DNA has a methylation rate of 3 to 10%, that of plants up to 50%. Viruses, on the other hand, have a degree of methylation of up to 100%. Here, the direct detection of the non-methylated cytosines via the corresponding guanine / uracil mismatches can also be advantageous. It is also possible to carry out both process variants in parallel.

As detector DNA sequences z. B. synthetically accessible polynucleotides, preferably with a length between 5 and 100, particularly preferably between 12 and 35 nucleotides, can be used. For this purpose, the detector DNA sequences can be derived directly from the known starting DNA sequence. The individual detector DNA sequences can overlap, so that ultimately a mismatch that occurs can always be assigned to a specific location within the starting DNA sequence.

The mismatches can finally be identified with mismatching substances, such as. B. mutS can be demonstrated. The chemical conversion of non-methylated cytosine bases in DNA sequences is preferably carried out on single strands which, for. B. can be obtained by denaturing double-stranded DNA samples in alkaline. The converted DNA can then, if necessary, be worked up further, ie separated, denatured, neutralized, precipitated and desalted.

Methods for the chemical conversion of cytosine to uracil are e.g. B. described in "Bisulfite methylation analsis of Single DNA-Strand", Tech. Tips Online 1998 BioMednet; Kimberely D Tremblay, Deaprtment of Biology, University of Pennsylvania, Philadelphia, PA 19104 USA and in "Promoter methylation analysis on microdissected paraffin-embedeed tissues using bisulfite treatment and PCR-SSCP ", Biotechniques 30: 66-72 (January 2001) pp. 66-72. The DNA to be converted containing methylated cytosine bases is dissolved in water and then in the alkaline solution by adding a sodium bisulfite hydroquinone solution elevated temperature with exclusion of light after the reaction is complete, the DNA obtained can be purified by chromatography.

The hybridization of the chemically converted DNA nucleotides with the detector DNA can take place on any surface to which the hybrids can be bound in a spatially resolved manner. There are known options such. For example, the binding of nucleic acids via disulfide bridges to gold surfaces or the binding of biotinylated nucleic acids to surfaces covered with avidin are available. The hybridization itself can take place passively or actively electronically accelerated, as z. B. is possible with a chip from Nanogen Inc. (San Diego / USA). The electronic addressing is carried out by applying an electrical field, preferably between 1.5 V and 2.5 V with an addressing time between 1 and 3 minutes. Due to the electrical charge of the nucleotide sequences to be addressed, their migration is greatly accelerated by applying an electric field. The addressing can take place in a spatially resolved manner, addressing being carried out successively at different zones of the chip surface. You can send to the different addressing and hybridization conditions can be set in individual locations.

After hybridization, both the guanine / uracil and the adenine / methylcytosine mismatches can be detected with substrates that recognize mismatches. Suitable substrates are e.g. B. base mismatch binding proteins such as mutS or mutY, preferably from E.coli, T. thermophilus or T.aquaticus, MSH 1 to 6, preferably from S.cerevisiae, S1 nuclease, T4 endonuclease, thymine cosylase, cleavase or fusion proteins containing one Domain of these base mismatch binding proteins.

In addition to dye-bearing, luminescent and fluorescent groups, the protein which recognizes mismatches can also contain polymeric markers (J. Biotechnol. 35, 165-189, 1994), metal markers, enzymatic or radioactive markings or quantum dots (Science Vol 281, 2016, Sep 25, 1998 ) contain. The enzyme label can z. B. be a direct enzyme coupling or an enzyme substrate transfer or an enzyme complementation. Chloramphenicol acetyl transferase, alkaline phosphatase, luciferase or peroxidase are particularly suitable for enzymatic labeling.

In particular, substrate marking with dyes which absorb or emit light in the range between 400 and 800 nm is preferred. Among the fluorescent dyes suitable for labeling are especially Cy ™ 3, Cy ™ 5 (from Amersham Pharmacia), Oregon Green 488, Alexa Fluor 488, Alexa Fluor 532, Alexa Fluor 546, Alexa Fluor 594, Alexa Fluor 647, Bodipy 558/568 , Bodipy 650/665, Bodipy 564/570 (e.g. from Mobitec, Germany), S 0535, S 0536 (e.g. from FEW, Germany), Dy-630-NHS, Dy-635 -NHS, EVOblue30-NHS (e.g. from Dynomics, Germany), FAR-BIue, FAR-Fuchsia (e.g. from Medway, Switzerland), Atto 650 (from Atto Tech, Germany) , FITC or Texas Red. In addition to the directly labeled substrates, labeled antibodies can also be used directly against mutS or against a fused peptide domain, e.g. B. MBP are directed.

Since a defined nucleotide sequence can be assigned to each location on the support surface, the fluorescence signals obtained can be directly translated into the corresponding methylation pattern in the cases in which exactly one possible mismatch can occur at each location on the support. As a rule, however, several potential mismatch positions (cytosines in the corresponding partial sequence of the starting DNA) are contained at every location on the carrier surface. To determine the methylation pattern on the basis of the fluorescence signals, it is then expedient to use detector DNA sequences which overlap in their sequence. The evaluation is then carried out according to the “clustering analysis” method (described in Eisen, MB, Spellman, PT, Brown PO, Botstein D 1998, Cluster analysis and display of genome wide expression pattern, Proc. Natl. Acad Sei USA95, 14863- 14868; Review: Discovering Pattern in Microarray Dat Molecular Diagnosis Vol. 5 No. 4 2000 Harry Burke .S349) or the "Principal Components analysis" method (described in Hilsenbeck SG, Friederichs WE, Schiff R., Statistical analysis of array expression data as applied to the problem of tamoxifen resistance J. Natl Cancer Institut 1999 91 453-459).

The specificity of the measurement of mismatches with mutS, a preferred mismatching substrate, can be further increased by the addition of surfactants or by the addition of BSA that blocks non-specific binding sites on the hybrids. MutS also has the advantage that measurements can also be carried out under low salt conditions up to a salt concentration of 10 mM.

In order to further improve the reliability of the method, any single-stranded nucleotide sequences fixed on the support after the hybridization step can be broken down by adding a nuclease, preferably a mung bean nuclease or S1 nuclease. With the method according to the invention, the methylation pattern of DNA fragments with known sequences can be determined simply and quickly. In particular, by using active, electronically addressable chips, the measuring time can be reduced further and the reliability of the measurement can be improved by the very good electrical controllability of the hybridization. The carriers once covered with detector DNA are also reusable, the measured DNA can simply be dehybridized and washed off the surface. This enables a high sample throughput with an occupied carrier.

To explain the claimed method, the individual method steps are shown schematically in FIG. 1 as an example. Step I) shows the chemical conversion of starting DNA oligonucleotides. The oligonucleotides are shown in dashed lines, the left oligonucleotide containing a methylcytosine ( ^5m C), the right oligo in the same place an unmethylated cytosine (C). During the reaction, the unmethylated cytosine bases are converted into uracil. The oligonucleotides are then applied to a chip surface which contains detector DNA with an adenine base at the position on the detector DNA oligonucleotide which is complementary to the ^5m C and C positions (step II)). Step III) shows the hybridization of the applied oligonucleotides with the detector DNA sequence, wherein in the presence of a methylcytosine at this position a mismatch occurs which can be detected with fluorescent-labeled mutS.

EXAMPLES

Detection of methylated C building blocks in DNA on an active electrical chip and the enzyme MutS:

In a first experiment it is shown that the detection of adenine / methylcytosine mismatches with the aid of the claimed method fluorescent-labeled mutS succeeds as a substrate that recognizes mismatches. For hybridization.

For this, the oligonucleotides Seq. ID No. 1, 2 and 3 (0.05 μmol scale) diluted with water so that a final concentration of 10 picomoles / 1 μl is obtained. The aqueous solution was then dissolved in 50 mM histidine buffer 1: 100.

Oligonucleotides used:

1) Detector DNA with biotin at the 5 ^' end (sense + 2A (5' ~ biotinylated)) AAACATACAA AAATTAAAAT ACAAATCATC TCTAACCA (Seq. ID No. 1)

2) Model compounds to evaluate MutS binding (5 ^Λ -CY3 marked) TGGTTAGAGA TGATTTGTA ^5m C TTTAATTTTT GTATGT (Seq. ID No. 2) TGGTTAGAGA TGATTTGTAT TTTAATTTTT GTATGT (Seq. ID No. 3)

The methyl modified bases are marked as ^5m C.

Addressing on an electrically addressable chip with a hydrogel layer: The biotinylated oligo was addressed and fixed at 2 V for 60 s on the electrodes of the chip (Nanogen Inc.) at two separate locations on the surface of the chip. The complementary DNA oligos were hybridized in a spatially resolved manner at 2 V for 120 s in histidine buffer. The chip was washed for 10 min with buffer A (20 mM Tris / HCl pH, 7.6, 50 mM KCI, 5 mM Mg CI2, 0.01% Tween / 20, 1% BSA).

0.5 μg Alexa Fluor 647 (from Mobitec, Germany), labeled mutS were taken up in 100 μl buffer A. The chip is incubated with the dissolved mutS for 45 minutes. After 45 minutes, the chip is rinsed with 10 ml of buffer A without BSA and the fluorescence is then determined using a scanner. The result is shown in Fig. 2. The fluorescence of two measurements taken independently of one another was at the location where the hybridization of Seq. ID No. 1 and 2 led to an adenine / methylcytosine mismatch (bar 1 in Fig. 2) at 60.3 (spot 1) and 55, 7 (spot 2), the fluorescence was at the locations where there was no mismatch at 47.9 (spot 3 ) and 29.5 (Spot 4). The corresponding mean values of the two measurements are shown in FIG.

Chemical modification of cytosine building blocks to uracil building blocks:

10 μg or 5 μg or 1 μg of an oligonucleotide Seq. ID No. 4

TGGCTAGAGA TGATCCGCA ^5m C TTTAACTTCC GTATGC (Seq. ID No. 4) (starting DNA oligo that has been chemically treated (5 -CY3 labeled))

with a methylcytosine building block were dissolved in 100 ul water. Then 10 μl of 3 N aqueous NaOH solution were added. The 110 ul solution was 30 min. heated to 42 ° C and then taken up in 1020 μl 40.5% sodium bisulfite (pH 5) and 60 μl 10 mM hydroquinone solution. After adding a further 10 μl of water, the mixture was heated at 55 ° C. for 16-18 hours with the exclusion of light. 500 μl of the corresponding solution were then pipetted onto a NAP5 column (Pharmacia), which was conditioned according to the manufacturer's instructions. The corresponding oligonucleotide was eluted with 800 ul water. The reaction is completed by adding 3 M NaOH to the oligonucleotide fraction obtained to a final NaOH concentration of 0.3 M at 37 ° C. The chemically converted DNA oligonucleotides obtained are then purified again by chromatography on a NAP5 column and desalted. 50 ul of the desalted solution were diluted with 50 ul 100 mM histidine solution. DNA oligonucleotides are obtained whose unmethylated cytosine bases have been converted into uracil. Preparation of the 40.5% sodium bisulfite solution (Sigma, S-8890, 8.1 g / 20 ml): For this purpose, the solid substance was dissolved in 17 ml of cold water and adjusted to a pH of 5 with 10 N NaOH solution and then to Replenished 20 ml. A 10 mM (Sigma H-9003) solution was used as the hydroquinone solution. (The chemical conversion of cytosine to uracil is described, for example, in "Bisulfite methylation analsis of Single DNA-Strand", Tech. Tips Online 1998 BioMednet; Kimberely D Tremblay, Deaprtment of Biology, University of Pennsylvania, Philadelphia, PA 19104 USA and in "Promoter methylation analysis on microdissected paraffin-embedeed tissues using bisulfite treatment and PCR-SSCP", Biotechniques 30: 66-72 (January 2001) pp. 66-72)

Non-specific labeling of mutS with Alexa Fluor 647

For conjugation, the corresponding Alexa Fluor 647 succinimidyl ester is non-specifically covalently linked to protein lysine residues via a nucleophilic substitution reaction in accordance with the FluoroLink ™ product specification protocol, Amersham Pharmacia Biotech, Little Chalfont, UK. The fluorescence labeling of mutS is carried out by incubation with a large excess of fluorophore under strongly basic conditions (0.1 M Na ₂ C03, pH 9.3). The fluorescence-labeled mutS obtained is purified by gel permeation chromatography.

Claims

claims:

1. A method for determining the methylation pattern of DNA comprising the following process steps: a) chemical conversion of the starting DNA with a known sequence, the unmethylated cytosine contained therein being converted into uracil, b) hybridization of single-stranded DNA treated according to step a) with single-stranded DNA Detector DNA sequences which are complementary to the starting DNA sequence or partial sequences thereof, it being possible for the guanine bases to be replaced by adenine, c) fixation of single-stranded detector DNA sequences or of single-stranded DNA sequences treated according to step a) before or during the hybridization or of heteroduplexes from the detector sequences and DNA sequences treated according to step a) after or during the hybridization, the fixation being spatially resolved on a support, so that each detector DNA sequence can be assigned a specific location, d) incubation with a guanine / uracil or adenine / methylcytosine mismatch ends substrate and detection of the substrate bonds.

2. The method according to claim 1, characterized in that the hybridization takes place electronically accelerated.

3. The method according to any one of claims 1 or 2, characterized in that the fixation of single-stranded detector DNA sequences or single-stranded DNA sequences treated after step a) before or during the hybridization or of heteroduplexes from the detector and after step a) treated DNA sequences after or during the Hybridization takes place in a spatially resolved manner on an electronically addressable surface.

4. The method according to any one of claims 1 to 3, characterized in that the chemical conversion of the non-methylated cytosine is carried out with a bisulfite.

5. The method according to any one of claims 1 to 4, characterized in that the single-stranded nucleotide sequences fixed on the support, which are not hybridized, are broken down by adding a nuclease.

6. The method according to any one of claims 1 to 5, characterized in that DNA sequences with a length of 5 to 100 nucleotides are used as detector DNA.

7. The method according to any one of claims 1 to 6, characterized in that overlapping DNA sequences are used as detector DNA.