WO2013102290A1

WO2013102290A1 - Method for specifically recognizing dna containing 5-methylated cytosine

Info

Publication number: WO2013102290A1
Application number: PCT/CN2012/001718
Authority: WO
Inventors: 施一公; 颜宁; 邓东; 闫创业; 潘孝敬
Original assignee: 清华大学
Priority date: 2012-01-04
Filing date: 2012-12-21
Publication date: 2013-07-11
Also published as: CN103987860B; CN103987860A

Abstract

Disclosed is a method for specifically recognizing DNA containing 5-methylated cytosine. The method comprises recognizing 5-methylated cytosine in DNA by using TALE protein.

Description

Method for specifically recognizing DNA containing 5-mercapto-cytosine

The present invention relates to the field of biotechnology, and more particularly to a method for specifically recognizing DNA containing 5-methylcytosine. Background technique

TALE ( Transcription Activator Like Effectors) is a protein in the cell of the phytopathogenic genus Xanthomonas; : 0 When a pathogen infects a plant, the pathogen passes its own III The type of secretory system injects a series of effector molecules, including TALE, into plant cells. These effector molecules help the bacteria to further expand by affecting host cell signaling, gene expression, etc. TALE is the largest of these effectors. In one class, it functions like a plant's own transcriptional activator.

: TALE family proteins generally consist of three major functional domains, the N-terminal domain and

TALE is involved in secretion transport; the C-terminus has a transcriptional activation domain and a nuclear signal peptide fragment; the region located in the middle of TALE is a DNA-binding domain, but its DNA-binding domain is different from other known DNA-binding domains. It consists of a series of repeating units. In most cases, each repeat unit consists of 34 amino acids, and individual repeat units consist of 33 0 or 35 amino acid residues. Among the 34 amino acids, except for the amino acid changes at positions 12 and 13, the other amino acids are highly conserved. These two non-conservative amino acids are named RVD (repeat variable diresidue, repeated variable double residues). J. Boch et al. and MJ Moscou et al. (see J. Boch, H. Scholze, S. Schornack, A. Landgraf, S. Hahn, S. Kay, T. Lahaye, A. Nickstadt, U. Bonas, Breaking the code Of 5 DNA binding specificity of TAL-type III effectors, Science, 326 (2009) 1 509- 1 5 12 and .J. Moscou, AJ Bogdanove, A simple cipher governs DNA recognition by TAL effectors, Science, 326 (2009) 1501 In 2009, through experiments and bioinformatics studies, it was found that the amino acids (RVD) at positions 12 and 13 in each repeat unit have a specific correspondence with the identified nucleotide species, for example: Table 1 Correspondence between partial RVD and DNA base sequences

RVD amino acid sequence DNA base sequence

HD C

NG T

N1 A

NN G/A

NS A/G/C/T

The specific DNA sequence recognition and flexible assemblability of TALE proteins provide great promise for their application in molecular biology. Scientists can design and assemble arbitrary TALE units to recognize arbitrary DNA double helix sequences. This feature has been used to construct DNase (TALE nuclease, TALE nuclease), which cleaves a specific double-stranded DNA sequence, for introducing site-directed mutagenesis, site-specific knockout, etc. (AJ Bogdanove, DF Voytas, TAL Effectors: customizable proteins for DNA targeting, Science, 333 (201 1) 1843-1846. ). In all current known reports, TALE recognizes unmodified double-stranded DNA. Summary of the invention

In one aspect, the invention relates to a method for detecting cytosine thiolation in DNA, comprising the use of TALE protein and a derivative thereof to specifically recognize 5-mercapto-cytosine in DNA.

In a preferred embodiment, two different TALE proteins are used to specifically recognize cytosine and 5-mercapto-cytosine in the target sequence.

In a further preferred embodiment, the method is for detecting guanidation of a CpG island. In one aspect, the invention relates to the use of a TALE protein and a protein thereof for the specific recognition of 5-nonylated cytosine in DNA.

In another aspect, the invention relates to the use of a TALE protein and a derivative thereof for the preparation of a reagent for the specific recognition of 5-mercaptocytosine in DNA.

In another aspect, the invention relates to the use of a TALE protein and a derivative thereof for the preparation of a medicament for the diagnosis or treatment of cancer.

In a preferred embodiment, the diagnosis or treatment is carried out by specifically recognizing 5-mercapto-cytosine in the DNA.

The invention further relates to TALE proteins and derived proteins thereof for the specific recognition of 5- Amidoxime-modified DNA.

The invention also relates to TALE proteins and derived proteins thereof for use in the diagnosis or treatment of cancer. The TALE protein can be a natural TALE protein and a TALE-derived protein that retains or enhances the specific recognition of 5-methylcytosine in DNA by genetic modification, modification, and assembly. The TALE-derived protein further comprises a recombinant protein having a TALE protein DNA binding domain. DRAWINGS

Figure 1 is a schematic representation of the high-resolution crystal structure (1.85 angstrom) of the DNA binding domain of dHax3 (dHax3 truncated, labeled dHax3-A) and double-stranded DNA. 1-10 in the left panel shows each repeat unit of the DNA binding domain of dHax3, which recognizes the corresponding DNA sequence on the right side. Each repeating unit consists of two o helices, and the two helices are and 1>, respectively. The structure has been uploaded to the PDB database with the code: 3V6T. Where dHax3 (designed Hax3 ) refers to the modified TALE protein Hax3.

Figure 2 shows the interaction between dHax3 and DNA bases. A, the side chain of RVD in dHax3 is pointed, the first amino acid in RVD does not extend into the DNA major groove, and the second amino acid extends the amino acid side chain to the DNA major groove; B, the first amino acid in RVD passes Hydrogen bond stabilizes the loop region conformation. When the first amino acid of the DNA binding domain repeat unit is asparagine (N) or histidine (H), they are the eighth amino acid backbone of the repeat sequence. The carbonyl oxygen atom forms a hydrogen bond interaction, which acts to stabilize the loop conformation of the entire RVD; C, the second amino acid in RVD directly interacts with the DNA base, when the amino acid residue is aspartic acid (D) When the carboxyl oxygen of aspartic acid directly forms a hydrogen bond with the amino group of cytosine in DNA through hydrogen bonding; when the amino acid residue is serine (S), the hydroxyl group in serine forms direct hydrogen with N7 in adenine. Bond interaction; when the amino acid residue is glycine (G), there is a van der Waals interaction between it and the thymine methyl group, but D, as in the molecule shown in Figure A, RVD is the loop conformation of NG; E. In the molecule shown in Figure B, RVD is the loop conformation of NG.

Figure 3 is a comparison of the structure of thymine (left) and 5-mercapto-cytosine (right). It is clear from the comparison in the figure that the only difference between thymine (left) and 5-mercapto-cytosine (right) is the amino group at the six position and the carbonyl oxygen atom. Both the amino group and the carbonyl oxygen atom may interact with the amino acid residues of the protein by van der Waals forces. Figure 4 shows that biochemical experiments and crystal structure analysis revealed that TALE protein recognizes 5-mercapto-cytosine through NG. a, dHax3-recognized DNA sequence containing 5-mercapto-cytosine (5mC) (this sequence is called dHax3-5mC, contains three 5mC, only shows the base recognized by RVD of dHax3, the specific sequence is shown in the examples) And the corresponding RVD in the dHax3 protein; b. EMSA detects the dHax3 pair of DNA sequences without 5mC (called dHax3 box, which is identical to the dHax3-5mC sequence except for 5mC) and the dHax3 pair of 5mC DNA sequences (dHax3) -5mC) binding capacity, about 4 nM nucleic acid probe was added to each lane; and gradient concentrations of dHax3 protein were added to the lanes 0 to 10, respectively, concentration 0, 8nM, 16nM, 31.5nM, 62.5nM , 125nM, 250nM, 500nM, Ι ΟΟΟηΜ,

: G 2000nM , 4000nM; c, the crystal structure of the DNA binding domain of dHax3 ( dHax3-A ) and the DNA sequence containing 5mC (dHax3-5mC ), showing that the base of the side chain is 5-methylcytosine, glycine The van der Waals interaction with 5-methylcytosine interacts with glycine and thymine.

Figure 5 is an electropherogram showing the purification results of the dHax3 full-length protein. Lane marking

: Ming: 1. Whole bacteria breaking liquid; 2. Whole bacteria crushing and centrifugation; 3. Whole bacteria crushing centrifugal supernatant; 4. Nickel column culture waste liquid; 5. Nickel column cleaning liquid; 6. Nickel column elution recovery Liquid; 7. Nickel column; 8. Molecular weight marker.

Figure 6 is an electropherogram showing the purification results of the dHax3 truncated body protein (dHax3-A). Lane markings: A. Whole bacterial crushing solution; P. Whole bacterial crushing and centrifugation; S. Whole bacterial disruption 0 Centrifugal supernatant; F. Nickel column penetrating solution; W1. Nickel column cleaning solution 1; W1. Cleaning solution 2;

E. Nickel column elution recovery solution; R. Nickel column column; M. Molecular weight marker.

Figure 7 shows that DNA binding experiments demonstrate that NG can specifically recognize thiolated cytosine. a, different DNA probes for detecting DNA binding (only the bases recognized by the RVD of dHax3 are shown, see the examples for details). 6T-6C indicates that 6 thymines (T) in dHax3-box were replaced with 6 cytosines (C); 6T-6mC means 6 thymines (T) in dHax3-box with 6 methyl groups Cytosine (5mC) substitution; 5C-5mC means replacing 5 cytosines (C) in dHax3-box with 5 methylated cytosines (5mC); 5C-5mC means 5 of dHax3-box Cytosine (C) was replaced with 5 methylated cytosines (5mC); 5C-5T was substituted for 5 cytosines (C) in dHax3-box with 5 thymine (5T) 0; 5C-5A Five cytosines (C) in dHax3-box were replaced with five adenines (A); 5C-5G indicated that five cytosines (C) in dHax3-box were replaced with five guanines (G). b, dHax3 has a DNA sequence containing six thiolation modifications (6T-6mC) Similar binding ability to the control experiment (dHax3-box). c, an RVD in dHax3, NG, cannot bind to cytosine (C) without thiolation. d, an RVD in dHax3 - HD - is specifically recognized for cytosine (C), and methylation modification affects the recognition of HD and cytosine. In the EMSA experiment, gradient concentrations of dHax3 full-length protein were added to lanes 5, 6 to 10, 1 1 -15, and 16 to 20 at concentrations of 0, 146 nM, 440 nM, 1330 nM, and 4000 nM, respectively.

Figure 8 is the DNA binding domain of dHax3-NN variant (dHax3-NN-A, that is, the RVD (NS) in the seventh repeat unit of the DNA binding domain of dHax3 is changed to NN by point mutation technique and the ninth RVD (HD) in the repeat unit is changed to NN by point mutation technique to form recognition of two thiolated CpG islands, and the specific recognition sequence thereof is shown in the examples.) The crystal structure of the DNA containing two methylated CpG islands is combined. . detailed description

The inventors successfully analyzed the crystal structure of the complex of the DNA binding domain of the modified TALE protein Hax3 (referred to herein as dHax3 (designed Hax3)) and dsDNA. This structure reveals that RVD specifically recognizes the molecular basis of each DNA base. NG in RVD relies on van der Waals force and 5-methyl interaction of thymine, and other thymine groups do not participate in the reaction. This finding suggests that the TALE protein may specifically recognize 5-methylcytosine in the DNA Han chain through NG because 5-mercapto-cytosine has a similar structure to thymine. The inventors also successfully resolved the crystal structure of the complex of the DNA binding domain of dHax3 and the dsDNA having 5-mercaptocytosine.

This discovery provides a novel method for detecting and interfering with cytosine thiolation and can be used in the following ways:

1. Detection of cancer cells CpG island

Because 5-mercapto-cytosine appears in epigenetics, an important modification of DNA thiolation. DNA thiolation refers to the covalent bond of a methyl group at the 5' carbon position of the cytosine of the genomic CpG dinucleotide under the action of a DNA thiotransferase. Due to the close relationship between DNA thiolation and human development and tumor diseases, especially the transcriptional inactivation of tumor suppressor genes caused by thiolation of CpG islands,

Some methylation regions appear in the genome of cancer cells, and these thiolation phenomena do not occur in normal cells. Since the method of the present invention can effectively distinguish whether methylation occurs at a specific genomic locus, it can be used as a new cancer cell. testing method.

2. New ways to treat cancer

DNA methylation of cancer cells inhibits the expression of many tumor suppressor genes. Since the method of the present invention specifically reopens the expression of these genes in cancer cells, it can promote apoptosis of cancer cells. TALE itself has the function of activating transcription. By designing the RVD on the TALE repeat sequence, it specifically binds to the upstream promoter sequence of the thiolated modified tumor suppressor gene, specifically opening up a large number of tumor suppressor genes in cancer cells. , to achieve the purpose of killing cancer cells.

Unless otherwise defined herein, the relevant scientific and technical terms used in the present invention have the skill

: G Domain A common understanding of what ordinary technicians understand. Moreover, unless the context dictates otherwise, the singular term should include the plural and the plural term should include the singular. In general, the nomenclature and techniques of molecular biology, biochemistry, structural biology, and related uses described herein are those well known and commonly employed in the art. Unless otherwise stated, the following terms should be understood to have the following meanings:

: The term "TALE protein" as used herein refers to Transcription Activator Like

Effectors, ie transcriptional activator-like effectors. The TALE protein can be a natural TALE protein and a TALE-derived protein which retains or enhances the DNA, or DNA-RNA hybrid chain binding ability obtained by genetic modification, modification, and assembly.

The term "Hax3" as used herein refers to one of the members of the TALE protein family. The full name of Hax 0 is "Homolog of avrBs3 in Xanthomonas", t3⁄4 Hax3 ^Aff >^ 3⁄4 ^" One of the three homologous proteins identified by Armor aciae ( Xanthomonas campestris pv. Armoraciae ). One of its members, its function is similar to that of other known TALE proteins such as avrBs3 (see S. Kay, J. Boch, U. Bonas, Characterization of AvrBs3-like effectors from a Brassicaceae pathogen reveals virulence and avirulence activities and a protein with a novel repeat architecture, Molecular plant-microbe interactions : MPMI, 1 8 (2005 ) 838-848. ).

The term "dHax3" as used herein refers to an artificially engineered Hax3 (designed Hax3) whose nucleotide sequence is SEQ ID NO: 1 and the amino acid sequence can be found in SEQ ID NO: 20 (with a 6XHis tag inserted therein) MM Mahfouz et al. designed dHax3 to have the ability to specifically recognize the following DNA sequences: TCCCTTTATCTCT (MM Mahfouz, L. Li, M. Shamimuzzaman, A. Wibowo, X. Fang, JK Zhu, De Novo-engineered transcription activator-like effector (TALE) hybrid nuclease with novel DNA binding specificity creates double-strand breaks,

Proceedings of the National Academy of Sciences of the United States of

America, 108 (201 1 ) 2623-2628. ).

The term "dHax3 truncated body protein" ("dHax3-A") as used herein refers to a dHax3 truncated protein that has a N-terminal domain and a C-terminal domain removed, which is the dHax3 protein sequence.

230-721 with 1 1.5 repeating units.

The term "dHax3-NN variant" as used herein refers to a variant of dHax3 in which the RVD (NS) in the seventh repeat unit of the DNA binding domain of dHax3 is converted by a point mutation: G technology becomes NN and ninth RVD (HD) in a repeating unit becomes a point mutation technique

NN to form an identification of two two thiolated CpG islands, dHax3-NN following the DNA sequence: TCCCTTTATCTCT.

The term "dHax3-NN-A" as used herein refers to the protein sequence of the dHax3-NN variant.

A truncated body of 230-721, that is, a DNA binding domain is retained.

5 Because the molecular mechanism of RVD recognition DNA bases in all TALE proteins is the same, although there are certain sequence differences in different TALE proteins, RVD-NG, which involves dHax3 in the examples, specifically recognizes cytosine thiolation. The ability to apply equally to other TALE proteins other than the dHax3 sequence of the examples is also within the scope of this patent.

?.ο The various reagents used in the examples, including buffers, enzymes, vectors, kits, etc., are commercially available or in accordance with the third edition of the Guide to Molecular Cloning: Huang Peitang, Science Press , 2002) Prepared by the recommended method. Example

25 Example 1 : Construction and purification of several TALE proteins

1. The experimental methods for molecular cloning and expression vector construction are as follows:

• PCR amplification of target gene fragments

The composition of the 50 μΐ standard PCR reaction system is shown in the following table, and the system can be amplified according to the ratio if necessary;

Component volume (μΐ)

Ex Taq 0.25

Ι ΟχΕχ Tag Buffer 5

dNTP 4

DNA template 2.5 ng

5' primer 1

3 ' Primer 1

ddH ₂ 0 was added to 50 μΐ After the target fragment was successfully amplified, the amplified target gene fragment was directly recovered using a common DNA recovery kit. Note that if the amplified gene fragment is a point mutation, the DNA template is first removed by agarose gel electrophoresis, and then the target gene is recovered using an agarose gel DNA recovery kit.

• Restriction enzyme treatment of amplified fragments and vectors

The amplified fragment and vector were treated with the same restriction endonuclease to generate the same DNA cohesive ends. The composition of the 50 μΐ double digestion reaction system is shown in the following table:

50 μΐ standard double digestion reaction system

Component volume (μΐ)

PCR amplified fragment or plasmid 42 10χ digestion buffer (NEB buffer 4) 5 Ndel 1.2 Xhol 1.8

After 37 °C warm bath 30~180 min, after the reaction is completed, the DNA fragment is recovered by gel electrophoresis of the lipogel gel DNA recovery kit.

: Xin DNA connection

The digested target gene fragment was ligated into the vector at room temperature for 30 to 120 min using T4 DNA ligase. The connection system is shown in the following table:

Component volume (μΐ)

Target gene fragment after digestion 7

After digestion, vector 1

10χΤ4 ligase buffer 1

Τ4 DNA ligase 1 climbing transformation

The ligation product was transferred into DH5a competent cells according to the following method, and the positive clones were prepared for screening: 50~100μ1 DH5 competent cells were added to the ligation product, placed on ice for 30 min; heat shocked at 42 °C for 90 s; placed on water for 2 min; All products were applied to ampicillin-resistant agar plates, spread with a coating bar, and cultured in an inverted 14-16 hours.

筛选Use colony PCR to screen positive clones

Mark 4 to 8 colonies on the plate obtained in the previous step, and test positive using the following system.

:0 clone:

Colony PCR system

Component volume (μΐ)

Taq 0.2

Ι ΟχΕχ Tag Buffer 3

dNTP 2

DNA template

5 ' Primer 0.3

3 ' Primer 0.3

ddH ₂ 0 was added to 30 μΐ The results were confirmed by gel electrophoresis, and positive clones were picked and cultured in ampicillin-resistant LB medium at 37 ° C and 220 rpm overnight.

: 5 * Plasmid extraction

The plasmid was extracted using a common plasmid mini-kit, and sequencing was performed by Genewiz Biotech Co., Ltd.

Induced expression of punch recombinant protein

过量 In order to obtain a large amount of purified protein, overexpression is required. Existing overexpression systems are Escherichia coli (£. / ), yeast, insect cells, and the like. Different proteins may be suitable for expression in different systems. The target protein is a protein in Gram-negative bacteria, so E. coli was selected as an expression system for protein expression purification.

Purification of high-quality, high-purity proteins is a prerequisite for biochemical experiments and crystallization experiments. The purification of recombinantly expressed proteins from E. coli has been quite mature. For easy purification using affinity chromatography, recombinant proteins with various tags were constructed. After comparison, recombinant proteins with histidine tag were used for subsequent experiments. A histidine tag composed of six histidines may be bonded to a column having a metal atom such as nickel in the form of a coordinate bond. Proteins with a purity of about 95% or more can be obtained by nickel column affinity chromatography and heparin affinity chromatography.

The specific purification steps are as follows: 50 ml of LB medium containing ampicillin or ampicillin/chloramphenicol antibody was added and incubated overnight at 37 °C on a shaker.

b. Transfer 5-10 ml vial culture to 1 L of antibiotic-containing LB medium and incubate at 37 °C for about 3 hours. When 0D600 = 0.8 to: 1.0, the expression was induced by adding 0.2 mM final concentration of IPTG at 22 °C for 14 to 16 hours.

c The induced E. coli was centrifuged at 4400 rpm for 4 min at 10 ° C, and the supernatant was discarded. The wet bacteria collected by centrifugation per liter of culture medium were resuspended in 20 ml of lytic solution (25 11 1 ^ butyl 1^-1^1 1 8.0 8.0, 500 mM NaCl).

d. After ultrasonic disruption, centrifuge at 14000 rpm for 50 min, and take the supernatant for subsequent purification. e. Slowly add the supernatant to a nickel column that has been previously equilibrated with a lysis solution (25 mM Tris-HCl pH 8.0, 500 mM NaCl). Repeat the above operation 1 to 2 times with the passing solution.

5 f. Add 10 ml of Wash Buffer I (25 mM Tris-HCl pH 8.0, 1000 mM NaCl) to remove some impurities. Repeat the above operation 3 times.

g. Add 10 ml of Wash Buffer II (25 mM Tris-HCl pH 8.0; 100 mM NaCl; 1 OmM Imidazole) to further remove the heteroprotein.

h. Add 10 ml of elution buffer (25 mM Tris-HCl pH 8.0, 50 mM NaCl, 300 mM Imidazole) to elute the protein of interest from the nickel column. Use Coomassie Brilliant Blue G-250 to check for cleanliness. If the elution is incomplete, repeat the above procedure.

i. Slowly add the eluted protein to the previously used buffer (25 mM Tris-HCl) PH 8.0, 50 mM NaCl) Heparin sepharose 6 Fast Flow. The above operation was repeated 1 to 2 times with the passing liquid.

j. Add 10 ml of Wash Buffer I (25 mM Tris-HCl pH 8.0, 100 mM NaCl) to remove impurities. Repeat the above operation 3 times.

k. Add 10 ml of elution buffer (25 mM Tris-HCl pH 8.0, 1000 mM NaCl, 10 mM DTT) to elute the protein of interest from the heparin column. Use Coomassie Brilliant Blue G-250 to check for cleanliness. If the elution is not complete, repeat the above procedure. Protein purity was identified using SDS-PAGE.

1. The protein purified by the above two-step affinity chromatography was concentrated to ~10 mg/ml using an ultrafiltration concentrating tube. Finally, the protein was further purified using a molecular sieve (Superdax 200) and the protein was used. The buffer used for the molecular sieve was 25 mM Tris-HCl pH 8.0, 150 mM NaCl, 10 mM DTT. The buffer in the desalting column (Hiprep 26/10) dHax3 (231-720) protein was replaced with 25 mM MES pH 6.0, 50 mM NaCl, 5 mM MgCl ₂ , 10 mM DTT.

2. Construction and expression of dHax3 and dHax3-A

The dHax3 (designed Hax3) gene is obtained by whole gene synthesis and the sequence is as follows (SEQ

TGGTTAATGGAACTTCTACCGCAATGA

The synthesized gene was directly ligated into the pET300 (invitrogen) plasmid. The expressed full-length protein has six histidine tags at the N-terminus and is used for affinity purification of the nickel column by protein purification. The full-length protein sequence is as follows (SEQ ID NO: 2):

EELAWLMELLPQ

The purification map of dHax3 full-length protein is shown in Figure 5 (purified by nickel column affinity chromatography using a 6 χ histidine tag, and Coomassie blue is developed by SDS-PAGE electrophoresis).

Through protein secondary structure prediction, the inventors found that both the apical and C-terminal ends of the protein have a large number of regions without secondary structure. These regions are not suitable for protein crystallization, and the inventors have therefore designed a truncated body protein (dHax3 truncation, labeled dHax3-A) containing the protein sequence 230-721) to obtain a more stable protein. The dHax3 truncation was cloned into the pET21 (Novagen) expression vector. The expressed dHax3 truncated protein sequence is as follows, wherein the C-terminus contains a His ₆ tag for affinity purification by nickel column for protein purification (SEQ ID

The purification map of the dHax3 truncated body protein is shown in Figure 6 (purified by nickel column affinity chromatography using a Histidine ₆ tag, and subjected to Coomassie blue development by SDS-PAGE electrophoresis). 3. Construction and expression of dHax3-NN-A

The inventors also constructed and expressed the dHax3-NN-A protein for use with CpG islands.

Co-crystallization experiments of DNA sequences. Table 2 shows the RVD of the TALE repeat unit involved in the experiment and its identified DNA:

dHax3 0 1 2 3 4 5 6 7 8 9 10 11 11.5

RVD HD HD NG NG NG NS NG HD NG HD NG

dHax3-box τ CCCTTTATCTCT dHax3-5mC τ CCCT mC TA mC CTC mC dHax3-N 0 1 2 3 4 5 6 7 8 9 10 11 11.5 RVD HD HD HD NG NG NG NG NN NG HD NG dHax3-CpG τ CCCTT mC G mC GTCT Example 2: Obtaining the crystal structure of dHax3 and the crystal structure of the complex of dHax3-A and double-stranded DNA

Obtaining single-double-stranded DNA

In order to examine the ability of dHax3 to bind to single-stranded DNA and to obtain crystals of protein and dsDNA complexes, the inventors obtained single-stranded DNA (17 nt) by chemical synthesis: (Invitrogen & Takara)

5' TG TCCCTTTATCTCT CT 3' (SEQ ID NO: 4)

3' AC AGGGAAATAGAGA GA 5' (SEQ ID NO: 5)

The synthesized single-stranded DNA was dissolved to 1 mM, the two single-stranded DNAs were mixed in an equimolar ratio, and the bath was heated at 85 ° C for more than 3 min, and slowly cooled to 22 ° C, which was not less than 3 hours. For long-term preservation of the annealed double-stranded DNA, lyophilization and cryopreservation can be performed.

Obtainment of ruthenium complex crystal

The purified dHax3 truncated body protein (231-721 in the full-length sequence) was adjusted to a protein concentration of 6 to 7 mg/ml, and the double-stranded DNA after annealing at a molar ratio of 1.5:1 was added and incubated at 4 °C. 30 min.

The previous crystallization conditions were mainly based on the commercial Screen Kit, including: SaltRX from Hampton, Natrix, PEG/Ion, Crystal Screen, Index; Wizard I, II, III from Emerald; ProPlex from Molecular dimension.

The conditions for protein crystallization were screened from the above Kit, and the crystallization conditions were optimized by adjusting the concentration of the precipitant, the species, the concentration and type of the salt ions, and the concentration and type of the buffer. The crystal was optimized using the Addtive Screen and the Detergent Screen Kit. At the same time, the crystal is dehydrated, annealed, etc. to improve the diffraction quality of the crystal.

There is no regularity in the use of protein crystallization, so it is still a piece of art ': 0 surgery so far. Sparse matrix screen is commonly used in the initial stage, that is, the crystallization conditions of each company's configuration are purchased for screening. In most cases, crystals with high diffraction quality cannot be grown in the crystallization conditions obtained by the initial screening. In the following experiments, the inventors further refined the initial crystallization conditions, including adjusting the precipitant, pH. Buffer, salt, addition of reducing agent, detergent or alcohol; adjust the temperature, time, etc. of the crystallization experiment. The final crystallization conditions used are

1 The following crystallization mother liquid and the incubated protein nucleic acid complex were mixed by a volume ratio of 1:1, and cultured at 18 ° C for two days by a hanging drop vapor diffusion method, whereby crystals were obtained.

Crystallization mother liquor: 8-10% PEG3350 (w/v), 12% ethanol, 0.1 M MES pH 6.0. Climbing data collection and processing

?.o use Shanghai Synchrotron Radiation Center (SSRF) BL17U harness station or transcript

SPRING-8 BL41XU harness station for data collection. All collected diffraction data were integrated using H L2000 software, and further data processing was performed by CCP4 software. Using dHax3, which does not bind DNA, as a mode of substitution, the structure of dHax3 and DNA complexes was analyzed by molecular replacement. Finally use Phenix and COOT two soft

?. 5 pieces complete the correction processing of the structure. After data processing and structural analysis and modification, the structural resolution of dHax3 protein reached 2.4A, and the structure of dHax3 protein and dsDNA complex reached 1.85A. The statistics of data collection and structural correction are shown in the following table: Table 3 The crystal structure of dHax3 and the crystal structure of the complex of dHax3 -△ and double-stranded DNA

30 Statistics on collection and structural corrections

Data dHax3 ( 270-703 ) DNA ( dHax3 box ) combined with dHax3-A Integration Package HKL2000 HKL2000

Space Group C222! P2,

Unit Cell (A) 74.76, 95.51 , 153.21 81.719, 87.679, 88.494 Unit Cell (°) 90, 90, 90 90.00, 103.04, 90.00 Wavelength (A) 0.97915 1.00000

Resolution (A) 40-2.4 (2.49-2.4) 40-1.85 (1.92-1.85)

Rmerge (%) 4.9 (35.0) 6.1 (60.8)

I/sigma 24.1 (4.4) 22.5 (2.6)

Completeness (%) 95.6 (98.2) 99.7 (99.9)

Number of measured 84,417 391 ,380

Reflections

Number of unique 20,832 103,239

Reflections

Redundancy 4.1 (4.1) 3.8 (3.7)

Wilson B factor (A ² ) 60.9 24.6

R I free (%) 21. 1 1/ 26.36 19.07 / 21.99

No. atoms

Overall 2760 9579

Protein 271 1 7066

DNA 0 1383

Water 49 1 130

Other entities 0 0

Average B value (A ² )

Overall 63.86 33.26

Protein 63.89 31.94

DNA 0.0 33.98

Water 62.47 40.58

Other entities 0.0 0.0

R.m.s. deviations

Bonds (A) 0.009

Angle (.) 1.301 1.184

Ramachandran plot

Statistics (%)

Most favourable 92.7 93.5

Width allowed 7.3 6.5 Generously allowed 0.0 0.0

Disallowed 0.0 0.0 The inventors analyzed the high resolution crystal structure (1.8 angstroms) of dHax3-A and double-stranded DNA (dsDNA). This structure clearly demonstrates that dHax3 exhibits a right-handed helical structure that wraps dsDNA in the middle of the entire complex. The protein is entangled outside the DNA and embedded in the large 5 groove of DNA (see Figure 1).

The structure shows that the 12th amino acid (histidine/asparagine) located in each repeat does not directly interact with DNA, but instead they will be the 8th amino acid (alanine) of the repeat sequence in which they are located. The main chain oxygen atom forms a hydrogen bond, which acts to fix the ring in which the entire RVD is located.

: o The 13th amino acid in each repeat, if it is serine/aspartate, then they directly interact with the bases in the DNA to form hydrogen bonds; if it is glycine, then it is thiol with thymine The van der Waals interaction is formed (see Figure 2). Example 3. The crystal structure of the complex of dHax3-A and dHax3-5mC and the crystal structure of the complex of 5dHax3-NN-A and dHax3-CpG were obtained.

As shown in Figure 3, thymine (T) and 5-methylcytosine (5mC) indicate that 5-mercapto-cytosine has a sulfhydryl group in the fifth position, and this thiol group is the only group recognized by NG. Therefore, NG may recognize 5mC. Accordingly, the inventors designed the DNA sequence dHax-5mC (Fig. 4a)

?.ο 5' TCCT5mCTA5mCCTC5mC 3' (SEQ ID NO: 6)

3' AGGA GAT GGAG G 5' (SEQ ID NO: 7)

In order to study the recognition ability of the dHax3-NN variant CpG island, the inventor designed the DNA sequence dHax3-CpG.

5' TG TCCCTT(mC)G(mC)GTCTCT 3, (SEQ ID NO: 8)

2 y AC AGGGAA GC GCAGAGA 5' (SEQ ID NO: 9)

Using the method described in Example 2, the inventors obtained and analyzed the crystal structures of the two composites, and the statistical data of data collection and structural modification are shown in Table 4. Table 4: Crystal structure of the complex of dHax3-A and dHax3-5mC and statistical data of data collection and structural correction of the crystal structure of the complex of dHax3-NN-A and :0 dHax3-CpG Data DNA (dHax3-5mC) binds DNA (dHax3-CpG) to bind dHax3-A's dHax3-NN-A

Data collection

Space Group P21 P21

Cell dimensions

a, b, c (A) 81.84, 87.63, 88.46 81.20, 87.1 1, 88.15

β, γ, (° )——————....―—―.― 90, 102.85, 90 90, 102.85, 90

Resolution (A) 40-1.85 (1.92-1.85) 40-1.95 (2.02-1.95) Rmerge (%) 6.4 (64.1) 6.2 (55.7)

1 1 I 21.6 (2.8) 22.3 (2.8)

Completeness (%) 99.7 (100.0) 99.7 (99.5)

Redundancy 3.7 (3.7) 3.7 (3.8)

Refinement

Resolution (A) 40-1.85 40-1.95

No. reflections 103,273 87,970

Rwork 1 Rfree (%) 19.79/ 22.97 20.05/22.45

No. atoms

Protein 7121 7123

DNA/RNA 1387 1328

Water 832 753

Ion 0 2

B-factors

Protein 33.96 39.53

DNA/RNA 34.55 42.19

Water 39.94 47.88

Ion - 59.56

R. m. s. deviations

Bond lengths (A) 0.007 0.009

Bond angles (.) 1.158 1.344

Ramachandran plot

Statistics (%)

Most favoured 93.7 93.2

Additional allowed 6.3 6.6 Generously allowed 0.0 0.2

Disallowed 0.0 0.0 The inventors analyzed the complex structure of dHax3 protein with three 5mC DNA with a resolution of 1.85 angstroms. The high-resolution structure clearly reveals the molecular mechanism by which the dHax3 protein recognizes mC (Fig. 4c).

Figure 8 shows a schematic diagram of the DNA binding domain of the dHax3-NN variant and the crystal structure of the DNA containing two thiolated CpG islands, which confirmed that the dHax3-NN-A binding contains two thiolated CpG island DNAs. In mammalian cells, DNA thiolation occurs only on C in CpG islands. The applicant analyzed the crystal structure of TALE and the DNA sequence containing two CpG islands, further demonstrating that TALE has a specific recognition ability for thiolated DNA. This is very important for the expansion of TALE applications. Example 4. Gel retardation assay demonstrates the ability of dHax3 to bind to DNA duplexes with 5-mercaptocytosine (5mC)

• EMS A (electrophoretic mobility shift assay, also known as gel retardation assay)

The gel retardation assay is a special gel electrophoresis technique that studies the interaction of DNA/RNA with proteins in vitro. The basic principle is: In gel electrophoresis, due to the action of the electric field, the nucleic acid fragment of a small molecule moves faster toward the anode than the nucleic acid fragment to which the protein is bound. Therefore, a short nucleic acid fragment can be labeled, mixed with a protein, and the mixture can be subjected to gel electrophoresis. If the target DNA binds to a specific protein, the speed of movement is retarded, and autoradiography of the gel can be found. Nucleic acid binding protein. At the same time, by statistically comparing the amount of DNA bound to the protein and the amount of DNA of the unbound protein, a more accurate fit calculation can be made to the binding affinity of the protein to the nucleic acid.

• DNA/DNA oligo

Fragments of DNA/DNA oligo used in gel retardation experiments are shown in Table 5 below: Table 5 Fragment sequence of DNA/DNA oligo for gel retardation experiments

dHax3-bo 5'-CCACATATGTCATACGTGTCCCTTTATCTCTCTCCAGCTCGAG x GAATTC (SEQ ID NO: 10)

5'-GAATTCCTCGAGCTGGAGAGAGATAAAGGGACACGTATGACA TATGTGG (SEQ ID NO: 1 1)

dHax3-5m 5 '-CC AC ATATGTCAT ACGTGTCCCT TA I CTC 1 CTCCAGCTCGAGG C AATTC (SEQ ID NO: 12)

5 -GAATTCCTCGAGCTGGAGGGAGGTAGAGGGACACGTATGACA TATGTGG (SEQ ID NO: 13)

5C 5mC 5'-CCACATATGTCATACGTGTl l lTTTATlTlTCTCCAGCTCGAGG

AATTC (SEQ ID NO: 14)

TATGTGG (SEQ ID NO: 15)

5C-5G 5 -CCACATATGTCATACGTGTGGGTTTATGTGTCTCCAGCTCGAG

GAATTC (SEQ ID NO: 16)

5 -GAATTCCTCGAGCTGl

TATGTGG (SEQ ID NO: 17)

5C-5T 5 -CCACATATGTCATACGTGTTTTTTTATTTTTCTCCAGCTCGAGG

AATTC (SEQ ID NO: 18)

5,-(

TATGTGG (SEQ ID NO: 19)

5C-5A 5'-CCACATATGTCATACGTGTAAATTTATATATCTCCAGCTCGAG

GAATTC (SEQ ID NO: 20)

5'-(

TATGTGG (SEQ ID NO; 21)

6T-6mC 5 '-CC AC AT ATGTC AT ACGTGTCCC l l l AlClCl CTCCAGCTCGAGG

AATTC (SEQ ID NO: 22)

5,-GAATTCCTCGAGCTGGAGGGGGGTGGGGGGACACGTATGACA TATGTGG (SEQ ID NO: 23)

6T 6C 5 -CCACATATGTCATACGTGTCCCCCCACCCCCCTCCAGCTCGAG

GAATTC (SEQ ID NO: 24)

5 -GAATTCCTCGAGCTGGAGGGGGGTGGGGGGACACGTATGACA TATGTGG (SEQ ID NO: 25)

1 represents thiolated cytosine

The recognition sequence is highlighted. • DNA/RNA end labeling

To be phosphorylated DNA 1 ~ 20 pmol (5' end) Reaction buffer A (10X) 2 μΐ

[γ- ² Ρ]-ΑΤΡ (3,000 Ci/mmol) 20 pmol

Replenish nuclease-free deionized water to 19 μΐ

T4 polynucleotide kinase (lOU/μΙ) 1 μΐ After setting up the reaction system according to the above table, gently mix and incubate at 37 °C for 30 min; use G25 pre-installed desalting column to remove excess [γ] - ³² Ρ]-ΑΤΡ, adding an excess of unlabeled complementary strands, annealing to generate double-stranded DNA or DNA-RNA hybrid double strands.

• DNA/RNA and protein interaction systems

Full length protein (different concentrations) 5 ul

DNA /RNA 2 ul

5 Χ buffer 2 ul

ddH20 1 ul The reaction components were added to the reaction system in the above ratio, and mixed for 4 min at 4 ° C; the reacted sample was run 6 % non-denaturing gel;

After running the glue, dry the glue and put it on the phosphor screen for exposure overnight;

Image data was read with a Typhoon 9400 varible scanner.

The interaction of dHax3 protein with DNA with 5-methylcytosine (5mC) was detected by EMS A. The binding capacity is not significantly weakened (see Figure 4b for details). Figure 7 shows an RVD in dHax3 - NG - unable to bind to a cytosine without thiolation modification; and an RVD - HD in dHax3 - is specifically recognized for cytosine (C), and Methylation of cytosine affects the recognition of HD and cytosine.

Although the invention has been described in detail herein with reference to exemplary embodiments thereof, it is understood that the invention is not limited to the embodiments. Other variations, modifications, and embodiments within the scope of the invention will be apparent to those of ordinary skill in the art. Therefore, the invention should be construed broadly in accordance with the appended claims.

Claims

Rights request

1. A method for detecting cytosine methylation in DNA, comprising specifically recognizing 5-mercapto-cytosine in DNA using TALE protein and its derivative protein.

2. The method of claim 1, wherein two different TALE proteins and recombinant proteins thereof are used to specifically recognize cytosine and 5-mercapto-cytosine in the target sequence, respectively.

3. The method of claim 1 or 2, wherein the method is for detecting methylation of a CpG island.

4. The use of the TALE protein and its derived protein for the specific recognition of 5-mercapto-cytosine in DNA.

5. Use of a TALE protein and a derivative thereof for the preparation of a reagent for specifically recognizing 5-mercaptocytosine in DNA.

6. Use of a TALE protein and a protein thereof for the preparation of a medicament for the diagnosis or treatment of cancer, which is carried out by specifically recognizing 5-methylcytosine in DNA.

7. A method of diagnosing or treating cancer, wherein the TALE protein and its derived protein specifically recognize 5-methylcytosine in the DNA.

8. TALE protein and its derived protein, which are used to specifically recognize 5-mercapto-cytosine in DNA.

9. A TALE protein and a derivative thereof for use in the diagnosis or treatment of cancer by specifically recognizing 5-mercapto-cytosine in DNA.