Method for determining base sequence of dna or rna and dna sequencer

Description: TECHNICAL FIELD [Detailed description of the invention]

【0001】

[Technical field to which the invention belongs]

The present invention relates to a method and an apparatus for determining a base sequence of DNA or the like derived from a living organism or a non-living body. It is a system for DNA base sequence determination including sample preparation, electron microscope imaging, and image analysis centering on, and is related to a so-called ultra-high-speed DNA sequencer.

【0002】

[Conventional technology]

The conventional DNA sequencing method is wet chemistry with the core technology of separating DNA lengths by electrophoresis, and sequencing is based on reading DNA breaks, fluorescent labels, and separation / electrophoresis patterns. All of this has been developed around this core technology. Recently, the US company Cellular has 3x10 in one year.¹⁰It is said that the DNA base sequence of one (human genome for one person) was determined, and this was completed by operating 200 DNA sequencers with dozens of electrophoresis lanes in parallel. This is the result of massively parallelization of the same machine, but the analysis of the migration lane itself is not so fast due to the low migration separation efficiency. Currently 10 per lane per day⁴The base sequencing rate is considered to be the upper limit. The efficiency of DNA electrophoresis has been improved by shortening the migration distance and introducing microscopic observation, which is 10 times the current value, that is, 10⁵We are aiming for bases / day / lane, which is considered to be the limit of this method. This speed determines the cost of determining the base sequence as it is, and is currently estimated to be about 10 yen per base. Therefore, the cost of sequencing one human genome is currently 3x10.¹⁰Piece x 10 yen / piece = 3 x 10¹¹Yen = 30 billion yen is estimated.

【0003】

Under these circumstances, this current cost is too high for genomic information to become an industry. It is essential to develop a method to reduce the cost from 1/1000 to 1 / 10,000. Various methods have been proposed in this direction. For example, a method of targeting a specific gene DNA such as a DNA chip and probabilistically knowing its existence (already on the market), a scanning probe microscope, especially a scanning tunneling microscope, is used to magnify a DNA molecule and use it as an image. Methods for reading the base sequence are known (for example, Hiroyuki Tanaka et al., Biophysics, Vol. 40, No. 5, pp. 336-340 (2000)). The former has problems with low accuracy and is not a sequencing method for unknown DNA in the first place, and the latter has problems with slow data throughput, which is the limit of the scanning method, both exceeding the current electrophoresis method. It's not a thing.

【0004】

On the other hand, historically, various methods were tried in the 1960s before the invention of the current DNA sequencer by Glibert and Sanger. Among them, the electron microscope is considered to be the most promising because of its high spatial resolution, and research has been conducted in a wide range of fields from living organisms to physics. As was and is the case today, it is difficult to sequence double-lasen DNA using an electron microscope. According to computer simulation of electron microscopic images of DNA, it is not possible to distinguish between A (adenine), T (thymine), G (guanine), and C (cytosine) bases even with a spatial resolution of 0.05 nm. Compared to this, the use of single-stranded DNA has many advantages. That is, i) the spacing between bases when elongated is about twice (about 0.7 nm) as compared to double-racen DNA, ii) it is easy to label each base with a specific heavy atom, and , Iii) It can be applied not only to DNA sequencing but also to RNA sequencing.

【0005】

As a pioneering study that takes advantage of these advantages, an organic substance (diazotized 2-amino-p-benzenedisulfonic acid) is selectively bonded to a G (guanine) base and used as an active group (two sulfonic acids) of the organic substance. Attempts have been made to label the heavy atom uranium (Evangelos N. Moudrianakis et al., Proc. Natl. Acad. Sci. USA, Vol.53, pp.564-571 (1965)). By using an electron microscope capable of decomposing one uranium atom, the base G can be identified as a series of two uranium atoms.

【0006】

However, this method did not work as expected. This was due to the fact that there was a problem with the guanine-specific bond of organic matter, and that the electron microscope at that time did not have the ability to identify one uranium atom. The latter cause is considered to be that one atom could not be distinguished from the noise coming from the carbon film or polymer film, which is the background sample support substrate, because of the low contrast of the image, not the problem of spatial resolution. It seems that all were artificial errors.

【0007】

[Problems to be Solved by the Invention]

As described above, the DNA sequencing rate will be 10 in the future in the DNA sequencer that is the core of the conventional electrophoresis method.⁵Do not exceed base / day / lane. With the current method, it takes about one year to read the human genome of one person, and the cost is about 30 billion yen. It is required to raise it. For that purpose, the emergence of an epoch-making ultra-high-speed DNA sequencer is indispensable.

【0008】

The present invention provides a method for determining a base sequence of DNA or RNA and a novel DNA sequencer using this method, which enables discrimination of bases by transmission electron microscopy, not by conventional electrophoresis. The purpose.

【0009】

[Means for solving problems]

In order to determine the DNA base sequence at ultra-high speed, it is essential to develop a measurement method that extracts a large amount of primary sequence data at ultra-high speed. Since the electron microscope has a magnification of 1 million times and is a two-dimensional image, it is considered to be the method that most meets this requirement at present. However, considering the history of failure 30 years ago, it is necessary to rely on new means in some respects. They include expansion and extension of single-stranded DNA, specific heavy element labeling of bases on single-stranded DNA, high-resolution electron microscopy that can discriminate between heavy element types, and image analysis systems for DNA sequencing.

【0010】

The present inventor has completed a new system of nucleotide sequence determination method based on direct observation with an electron microscope without using electrophoresis by examining various points above and integrating these methods and devices. ..

【0011】

That is, the method for determining the base sequence of DNA or RNA (hereinafter, simply referred to as "DNA") of the present invention is a base-specific heavy element on DNA using a magnified image of extended single-stranded DNA by a transmission electron microscope. This is performed by discriminating the label, and the elongated single-stranded DNA is obtained by denatured the double-stranded DNA collected from a cell or the like to obtain a single-stranded DNA. On the other hand, RNA is originally single-stranded and can be used as it is, but cDNA prepared using RNA as a template can also be used.

【0012】

Specifically, the method comprises (a) forming a support film holding the elongated single-stranded DNA on the electron microscope grid, and (b) the electron microscope grid. A step of treating each base of the extended single-stranded DNA on the support film formed above with a base derivative containing a heavy element to form a base pair to form a base-specific heavy element label, and (c). It is characterized by including a step of discriminating the base-specific heavy element label using the magnified image obtained by observing with a transmission electron microscope, and further, in this series of steps, (d) was obtained. It is characterized by including a step of reading a magnified image and performing DNA sequence analysis by image analysis by a computer. At this time, it is appropriate to use a recording medium having a high density, low noise, and a high dynamic range as a recording medium for recording a magnified image by an electron microscope. As such a recording medium, for example, YAG Examples thereof include (yttrium-aluminum-garnet crystal), IP (imaging plate), and CCD camera, and among these, IP (imaging plate) is a particularly preferable recording medium in that it has a high dynamic range.

【0013】

High-speed reading is achieved by parallel processing of the recorded image. For example, the recorded image may be read by an IP (imaging plate) reader in which about 100 reading sensors, preferably about 1000 reading sensors are parallelized.

【0014】

In image analysis, in order to remove the overlap of base-specific heavy element labels at the intersections that enable sequencing even with intersecting single-stranded DNA, a magnified image is taken again with the same sample tilted. However, by separately providing a step of obtaining information in the depth direction and obtaining an image in which the overlap is removed, more accurate DNA sequence analysis becomes possible.

【0015】

As the transmission electron microscope used in the present invention, a retardation electron microscope or a complex electron microscope is preferable from the viewpoint of contrast, and the support film holding the single-stranded DNA is a film on which the DNA is fixed, and this support A sample for electron microscope measurement is prepared by transferring the film to the electron microscope grid. As such a support membrane, a protein-denatured membrane matrix is particularly preferable, and as a protein, a natural protein such as albumin and casein, and an artificial protein such as polylysine are preferable.

【0016】

Such a support film holding the elongated single-stranded DNA is obtained by being elongated by using the rapid expansion of the DNA solution on the water surface, and specifically, LB used on the surface of clean water. Examples include a membrane method, a vacuum spraying method in which an aqueous DNA solution is sprayed into a vacuum, and a glucose solution method in which a mixed aqueous solution of DNA and an adhesive protein is injected into a glucose solution. The glucose solution method using an aqueous solution is a preferable preparation method because the DNA has good retention due to the effect of surface denaturation of the protein as well as the elongation of the DNA.

【0017】

In the present invention, the base-specific heavy element label on the single-stranded DNA or RNA to be sequenced is the base on the single-stranded DNA, namely adenine (A), timine (T), guanine (G) cytosine ( C) and the bases on RNA, namely adenine (A), uracil (U), guanine (G), cytosine (C), and each base labeled with heavy elements (labeling with heavy elements with "*" (Show) is a heavy element label base, ie adenine^*(A^*), Uracil^*(U^*), Guanine^*(G^*), Cytosine^*(C^*) Is a base pair bond with AU^*, TA^*, GC^*, CG^*, UA^*It is formed by forming any of the base pairs of.

【0018】

As the base derivative containing a heavy element for forming such a base-specific heavy element label, a heavy element labeled base containing a heavy element that can be identified by an electron microscope is preferable, and this heavy element label is preferable. As for the base, it is preferable that the heavy element portion in the heavy element label base is composed of a heavy element complex, and this is bonded to the base. The heavy element complex may be a complex containing an element having a large atomic weight, and generally includes a complex containing a metal atom such as a heavy metal element.

【0019】

In the case of adenine (A) and guanine (G), the binding site between the substituent containing such a heavy element complex and the base is an imine group at the 7-position and / or 9-position, and uracil (U). ) And cytosine (C), the imine group at the 1-position is preferable in that it does not affect the specific binding property between the bases.

【0020】

The heavy element used in the heavy element complex of the present invention may be any element having a large atomic number that can be identified by an electron microscope, and a metal element having an atomic number of 25 or more is generally preferable. Then, in order to enable identification by an electron microscope, the metal elements to be adopted are preferably those whose atomic numbers are separated from each other by 15 or more, and in order to obtain four kinds of heavy element complexes, the above-mentioned Four kinds of metal elements that meet the conditions will be used in combination. Examples of such a combination of metal elements include "₇₈Pt,₆₃EU,₄₆Pd,₂₇Co ","₉₂U,₇₆Os,₄₆Pd,₂₆Fe ","₈₀Hg,₆₄Gd,₄₈Cd,₃₀Zn ","₈₉Ac,₇₄W,₄₂Mo,₂₅"Mn" etc. (subscripts indicate atomic numbers). It should be noted that there is no particular limitation on which metal is used for which base, but in general, combining a high-frequency base with a lighter element and a low-frequency base with a heavier element reduces mutual interference. Such examples include the combination of "adenine-Pd, guanine-Pt, cytosine-Eu, uracil-Co" and "adenine-Zn, guanine-Hg, cytosine-Gd, uracil-Cd". , Etc. are exemplified.

【0021】

Further, as the heavy element complex used in the present invention, not only a heavy element complex containing one metal in the molecule but also a complex containing a plurality of metal atoms can be used, and these are used for element discrimination and signal-noise discrimination. More favorable results can be obtained in this respect. Such complexes containing a plurality of metal atoms include disubstituted compounds of heavy element complexes containing two metal elements in one molecule, trisubstituted compounds containing three metal elements, and four. Examples include iron-sulfur cluster metal complexes containing metal elements. The plurality of metal elements contained in the complex may all be the same type of metal element, or may contain different metal elements.

【0022】

In addition, a heavy element label base (A), which is a heavy element complex used for this labeling.^*, U^*, G^*, C^*) Substitutes a part of the heavy element label base with at least one substituent selected from an alkyl group, a cyclohexyl group, a halogen group, a phenyl group, and a phenol group in order to ensure higher solubility in an organic solvent. These substituents are substituted at either the 2- or 8-position in the case of adenine (A), or both at the 2- and 8-positions, and at the 8-position in the case of guanine (G). In the case of uracil (U) and cytosine (C), it may be substituted with either the 5-position or the 6-position, or both the 5-position and the 6-position. When two substituents are bonded, these substituents may be the same or may be different from each other.

【0023】

In addition, these heavy element label bases (A)^*, U^*, G^*, C^*) Is at least one substituent selected from a halogen group and an amino group, in which a part of the heavy element label base is selected from a halogen group and an amino group so as to enhance the selectivity of the pair bond with the base and form a more specific bond. Substituted, these substituents are substituted at either the 2- or 8-position for adenine (A), or both at the 2- and 8-positions, and at the 8-position for guanine (G). It has been substituted, and in the case of uracil (U) and cytosine (C), it may be substituted at either the 5- or 6-position, or both the 5- and 6-positions. When two substituents are bonded, these substituents may be the same or may be different from each other.

【0024】

In the present invention, the formation of a base pair with a heavy element label base, which is a base derivative containing the above heavy element, is preferably carried out in an organic solvent, preferably in an organic solvent having a dielectric constant of 10 or less, and such dielectric Among organic solvents with a ratio of 10 or less, especially chloroform, heptane, cyclohexane, carbon tetrachloride, aniline, ethylamine, cresol, acetic acid, trichloroacetic acid, dimethyl ether, diethyl ether, toluene, toluidine, benzylamine, phenol, decanol, benzene. , Kinolin, morpholine, dimethylamine, chlorobenzene, dichloromethane, dichloroethylene, trichloroethylene, dichlorobenzene, fluorobenzene, bromobenzene, pentanol, siloxane, glyceride, etc. are preferable, and at least one organic solvent selected from these organic solvents is used. The DNA on the support membrane is treated with a solution in which a base derivative such as a heavy element label base is dissolved in a solvent containing the base to form a base pair to obtain a base-specific heavy element label, thereby ensuring specific binding of the base. It is preferable in that it does.

【0025】

The transmission electron microscope, which is a basic device used in the base sequence determination method of the present invention, is capable of sequencing even single-stranded DNA or RNA intersected on an image during DNA sequence analysis. In order to obtain an image with no overlap, it is preferable to provide an inclined image imaging system, and the phase difference electron microscope or the complex electron microscope transmission electron microscope used in the present invention is an electron that identifies heavy elements. It is preferable to have analysis software for a microscope, which performs element identification on the basis of quantitative measurement of signal intensity.

【0026】

Further, the present invention comprises a transmission electron microscope for discriminating base-specific heavy element labels and an analyzer for determining the base sequence of DNA or RNA by image analysis of a magnified image captured by the transmission electron microscope. It is a DNA sequencer that has, and is visualized as DNA in a state of being paired with a base on the target DNA, and the type of each paired base is discriminated by changing the type of heavy element to be labeled for each base. It is characterized by directly reading the base sequence of the target DNA.

【0027】

Hereinafter, the present invention will be described in detail.

【0028】

BEST MODE FOR CARRYING OUT THE INVENTION

According to the present invention, the base sequence of DNA or the like can be determined at high speed by discriminating the base-specific heavy element label using a magnified image of extended single-stranded DNA or RNA by a transmission electron microscope. A series of so-called DNA sequencer systems including expansion and extension of double-stranded DNA, specific heavy element labeling of bases on single-stranded DNA, high-resolution electron microscopy that can discriminate between heavy elements, and image analysis system for DNA sequence determination. It is configured as.

【0029】

First, a conceptual diagram of the entire DNA sequencer system of the present invention is shown in FIG. In Fig. 1, for each experimental step, the experimental concept and the corresponding concept of the DNA molecule are summarized. Hereinafter, with respect to the details of the contents of each part of FIG. 1, the matters to be improved will be explained and the embodiments will be described in detail. The DNA collection and the DNA thawing step of thawing and denaturing the DNA in FIG. 1 to obtain a single-stranded DNA are achieved by a known technique.

【0030】

1) Improvement of the electron microscope itself It is essential that the electron microscope used in the present invention has contrast and resolution capable of quantitatively performing elemental analysis of one atom for a heavy element (atomic number 25 or more). Since the distinction between elements will eventually utilize the difference in electron scattering intensity depending on the atomic number, quantitative detection of signal intensity that can distinguish two elements with atomic numbers separated by at least 15 or more on an electron microscope image is required. Be done.

【0031】

In consideration of the above points, the following electron microscope was used in the present invention. That is, an electron microscope capable of element identification of an atomic cluster consisting of 1 atom or 3 to 5 atoms can be achieved by improving a commercially available transmission electron microscope as follows, and such a transmission electron microscope is used.

【0032】

i) Phase difference electron microscope The most powerful method for making a quantitative electron microscope capable of discriminating elemental atomic numbers by recovering contrast without reducing the resolution is a phase difference electron microscope. The phase difference electron microscope is achieved by inserting a Zernike phase plate near the focal plane behind the objective lens (Japanese Patent Application No. 2000-085493). However, in order to obtain a resolution (0.2 to 0.3 nm) that enables discrimination of heavy elements, an electron microscope with a high resolution information limit is indispensable because spherical aberration and chromatic aberration are small. A high-resolution electron microscope capable of inserting a phase plate and discriminating an element of one atom requires a high voltage of 300 kV or more. However, in the case of discrimination of 3 to 5 element clusters, a 100 kV electron microscope is also possible.

【0033】

ii) Complex electron microscope Even in the case of a phase-difference electron microscope that cannot achieve a resolution of 0.2 to 0.3 nm, the resolution can be improved by using a complex electron microscope. The complex microscope is a method of combining a phase difference microscope and a normal microscope (Japanese Patent Laid-Open No. 11-258057). For example, a real number component microscopic image of a sample and an imaginary number component microscopic image obtained by π / 2 phase shifting only the transmitted light of the sample Is detected and the complex sum of the real number component microscopic image and the imaginary number component microscopic image is taken to obtain a complex microscopic image consisting of the real number component signal and the imaginary number component signal, but the analysis is a little more complicated than the phase difference microscope. is there.

【0034】

iii) Cryogenic sample stage Another way to improve the contrast is to use a cryogenic sample stage and increase the allowable electron beam irradiation. This method can be performed independently of the methods i) and ii) above. As the electron dose increases, the signal strength becomes stronger than the noise and the contrast is improved.

【0035】

As described above, by using a phase difference electron microscope or a complex electron microscope as a transmission electron microscope and keeping the sample stage at an extremely low temperature, if necessary, a magnified image having high resolution and contrast can be obtained.

【0036】

2) Robust thin film for retaining single-stranded DNA Since the distance between the bases when elongated is as large as 0.7 nm, it becomes easy to perform various chemical modifications to the bases, and the requirement for the resolution of the microscope is relaxed. However, a unique method is required for the method of making the target DNA a single strand and extending it to hold it on the sample support (electron microscope grid) for an electron microscope.

【0037】

Even if the electron microscope has the ability to identify one heavy element atom under ideal conditions, the following two noise problems must be solved in order to actually demonstrate its performance. That is, i) background shot noise and ii) fog from the sample support substrate. The former problem can be relatively reduced by increasing the allowable electron beam irradiation amount. Further, the latter problem may be solved by reducing the thickness of the substrate as long as it is not destroyed by electron beam irradiation.

【0038】

Therefore, the above i) and ii) can be solved if there is a thin film made of a light element that holds DNA firmly, has the strength to withstand strong electron beam irradiation, and has a small electron beam scattering ability. To.

【0039】

On the other hand, in the case of morphological studies using an electron microscope, an electron beam irradiation dose that does not significantly destroy the structure of the test object has been defined as an allowable amount. However, since the method of the present invention observes only labeled heavy elements, the structure of the organic substance (DNA, etc.) itself is not important. Rather, in order to obtain a strong contrast, it is preferable to use a strong electron beam irradiation exceeding the irradiation allowance of the organic substance. This is because the purpose can be achieved if the relative arrangement of heavy elements required for sequencing is maintained even if the structure is broken. Therefore, strong DNA retention and robustness of the retention membrane itself are required. Carbon thin films and aluminum thin films are already known as strong DNA supporting films. However, they tend to give a strong background fog and reduce the contrast of the base paired label heavy elements. Here, more preferable results can be obtained by using an electron beam scattering ability, that is, a protein film having a small fog.

【0040】

By the way, as a method of extending a polymer such as DNA and holding it on a support substrate, the LB film method using a water surface has been conventionally used. This is a method of expanding and extending single-stranded DNA on the surface of water and transferring it to a carbon film or an aluminum film on a grid. However, this conventional method has some disadvantages in that DNA elongation and transfer to a substrate are separate and independent processes. That is, i) the thin support substrate is fragile during transfer to the substrate, ii) the structure of the elongated DNA is liable to change during transfer, and iii) adsorption to the substrate is generally weak. As a method for solving all three points at once, a glucose solution method using a denatured protein membrane (Japanese Patent Laid-Open No. 8-157497) has been proposed. This method utilizes surface denaturation of proteins on the surface of water.

【0041】

Specifically, a mixed aqueous solution of single-stranded DNA and polyveptide (for example, polylysine), casein, and adhesive proteins such as albumin is prepared, and a small amount (about 1 μl) is prepared in a 2 to 5% glucose solution with an injection needle. ) Is injected. Then, the protein solution containing DNA floats due to the difference in density between the protein solution and the glucose solution, and rapidly develops on the surface of the glucose solution. At this time, two things proceed at the same time. One is the elongation of DNA accompanying the surface expansion of the protein solution, and the other is the surface denaturation of the protein itself. The denatured proteins are entangled with each other as polymer chains to form a tough protein thin film. Then, the elongated DNA is automatically retained in the thin film matrix.

【0042】

Another reason to use a protein film is to prevent image fogging on the support substrate. Since the protein has a lower density than the carbon film and other supporting substrates, its contribution to the image is small.

【0043】

Next, the protein thin film holding the DNA obtained as described above is transferred to a holy carbon film having a large number of holes. The protein membrane is attached so as to cover the holes on the carbon membrane and acts as a supporting substrate for DNA.

【0044】

Immobilization of elongated DNA on a protein thin film by this glucose solution method is extremely flexible in terms of control, and has the following advantages.

【0045】

i) The thickness of the protein-denatured membrane is adjusted by the protein concentration.

【0046】

ii) The entanglement and density of the denatured membrane, and therefore the strength, varies the pH, salt concentration, denaturant temperature, stabilizer temperature of the infusion solution, and some protein denaturation states (complete denaturation, molten globule, incomplete denaturation, etc. It can be adjusted by controlling (there is a state of).

【0047】

iii) The interaction between DNA and denatured protein, and therefore the strength of retention, depends on the protein selection, the pH of the infusion solution, and the concentration of divalent salt (Mg).²⁺, Ca²⁺Etc.) is controlled by adjusting.

【0048】

iv) The structure of elongated DNA in the membrane, especially the orientation of bases, is controlled by adjusting the pH and salt concentration of the glucose solution.

【0049】

Of course, the preparation of the protein thin film holding the single-stranded DNA is performed by independently creating the protein membrane and adsorbing the DNA, injecting the single-stranded DNA aqueous solution into the glucose solution, and adsorbing the protein membrane on the already completed protein membrane. Is also possible.

【0050】

3) Base-specific heavy element labels using specific base pair bonds As already mentioned, the method of directly labeling a base on a single-stranded DNA using a heavy element that can be identified by an electron microscope has a problem in base selectivity. In this labeling method, the problem was solved by using a specific base pair bond in nature. The function of DNA in nature is the transmission of genetic information by specific base pairs of AT and GC.

【0051】

It is generally thought that the specificity of this pair bond is due to complementary hydrogen bonds, but even if A, T, G, and C monomers are mixed, such a specific and selective pair bond is formed in an aqueous solution. It will not be done. Since a large amount of water molecules compete with hydrogen bonds, the base pair bonds are weakened and the specificity is lost. However, when the environment separates from water, the specificity of base pairs due to hydrogen bonds is restored. Specific pairing is expressed, especially in less polar organic solvents (eg, Kyogoku Yoshimasa, Chemistry, Vol. 22, No. 4, pp. 364-374 (1968)). This fact has been known for more than 30 years, but has not been used as a base-specific labeling method on DNA.

【0052】

This labeling method utilizes this specific base pair exhibited in an organic solvent. Then, in order to increase the efficiency, the base is further subjected to the following three types of chemical modifications. This includes i) modification with a heavy element complex to facilitate discrimination by electron microscopy, ii) solubilization modification to an organic solvent to facilitate the formation of base pair bonds, and iii) the base to be formed. It is a modification to increase the specificity of pair binding.

【0053】

In the present invention, a dry protein thin film containing single-stranded DNA is immersed in an organic solvent containing heavy element label bases to form complementary base pairs with each base on the single-stranded DNA. In this method, the chemical species of the heavy element-labeled base monomer and the type of organic solvent are important.

【0054】

i) Selection of base type Several are known as base species that selectively form strong pairing with A, T, G, and C of DNA (Kyogoku Yoshimasa, Chemistry, Vol. 22, No. 4, pp. 364-374). 1968)). Of these, four types, adenine (A), uracil (U), guanine (G), and cytosine (C), were selected as the basic base types of the high-speed DNA sequencer of the present invention because of their ease of modification and complementary base pair selectivity. Adopted as a monomer). Then, U is paired with A on the single-stranded DNA, A is paired with T, G is paired with C, and C is paired with G. In the present invention, uracil (U) in which the methyl group is not substituted was adopted because of the versatility of modification as compared with thymine (T).

【0055】

[Chemical 1]

<img file="JP2002153271A_D0001.tif" />The following Watson-Crick type is assumed as the base pair, but various measures are required to improve the selectivity as described below.

【0056】

[Chemical 2]

<img file="JP2002153271A_D0002.tif" />ii) Improvement of base pair selectivity For example, G makes incorrect base pairs such as GG, GA, and GU as well as C. Improving the selectivity of base pair AU and GC is a chemical design of how to reduce the probability of such errors. Furthermore, as shown below, there is a base pairing mode (Hoogsteen type) that is different from the Watson-Crick type even for AU and GC base pairs.

【0057】

[Chemical 3]

<img file="JP2002153271A_D0003.tif" />To prevent this, four types of bases A, U, G and C are modified with halogens such as Br and I, methyl groups such as methyl groups, alkyl chains such as ethyl groups and cyclohexyl groups, and amino groups. A can be 2nd and 8th, G can be 8th, and U and C can be 5th and 6th. Halogenation at the 8-position of A and G, or amination at the 2-position of A, and halogenation at the 5-position of U and C strengthen the pair bond and increase selectivity.

【0058】

On the other hand, in order to prevent Hoogsteen type pair binding, alkylation at the 8-position is preferable for A and G. C originally has a weak Hoogsteen-type bond with G, so it does not matter. However, there is currently no way for U to prevent the Hoogsteen type from base modification alone.

【0059】

Examples of such a compound include 8-brompurine, 2,6-diaminopurine, 5-bromuuracil, 5-iodouracil, and 5-alkyluracil, in the case of adenine and uracil as an example. Although uracil is used in the present invention, thymine itself is the one in which a methyl group is introduced at the 5-position of uracil in order to improve the selectivity of pairing, and therefore, the uracil of the present invention is also used in the case of a thymine derivative. It is contained in the derivative.

【0060】

iii) Organic solvent solubilization Solubilization in organic solvents is also a necessary matter, and as with the improvement of selectivity, A can be modified to the 2nd and 8th positions, G to the 8th position, and U and C to the 5th and 6th positions. .. Halogenation and alkylation to improve pair bond selectivity are themselves favorable for solubilization of organic solvents. It is preferable to modify the site that does not interfere with the base pair bond with a highly hydrophobic alkyl group. 8th place is good for A and G, and 6th place is good for U and C. Examples of the modifying group include an ethyl group, a propyl group and a cyclohexyl group. Such a compound is a compound similar to the compound exemplified in ii) except for the amino group, and the halogen or alkyl group improves the solubility in an organic solvent and the selectivity of pair bond formation. It can be seen that it contributes to both. In addition to the above substitutions, the improvement of solubility in organic solvents can also be achieved by introducing an alkyl group or the like into nitrogen at positions 7 and 9 of A and G, and nitrogen at positions 1 of U and C. .. Furthermore, the heavy atom complex, which is a labeling agent, also promotes solubilization in an organic solvent depending on the design method.

【0061】

iv) Selection of heavy element labels As the heavy element label used for labeling the base derivative forming the base-specific heavy element label of the present invention with a heavy element, a heavy element complex is preferable in that it has a high affinity for an organic solvent. These heavy element complexes are covalently anchored to the nitrogen at the 7th or 9th position in A and G and to the nitrogen at the 1st position in U and C to give the heavy element label base. As such a heavy element complex, any complex may be used as long as it exists stably, and many complexes such as amine complexes, benzene complexes, metallocene complexes, and olefin complexes of various metals can be exemplified. Further, when the heavy element complex is bonded to a base, it may be a heavy element complex obtained by directly coordinating a metal element with nitrogen in the base, and for example, a polymethylene chain, a polyoxyalkylene chain, or the like. , A heavy element complex may be bonded to nitrogen in the base via a so-called linker or adapter.

【0062】

On the other hand, the metal element used as a heavy element complex needs to have an electron scattering intensity that can be identified as one metal element by an electron microscope. At present, the metal element has an atomic number of 25 or more, preferably 30 or more. Corresponds to this. Since the distinction between metal elements is also performed by the electron scattering intensity from the metal, each metal element used preferably has an atomic number of at least 15 or more so that the difference in electron scattering intensity becomes clear. It is preferable that the metal elements are separated by 20 or more, and the four kinds of metal elements selected in this manner are used as heavy element labels for each base (A, U, G, C). The combination of metal elements used in this way is arbitrary, but if a preferable combination is shown as an example, "Pt, Eu, Pd, Co", "U, Os, Pd, Fe", "Hg, Gd," "Cd, Zn", "Ac, W, Mo, Mn" and the like can be mentioned, and among these, the combination of "Pt, Eu, Pd, Co" and "U, Os, Pd, Fe" is preferable. Also, the labeled base A^*, U^*, G^*, C^*Which heavy element is used as the label for which base is arbitrary.

【0063】

The heavy element label base obtained as described above is a base derivative containing a heavy element having a chemical structure in which a heavy element complex containing four kinds of metals is bonded to each base, and is on the target DNA. It is used as a treatment agent to form a base-specific heavy element label. Specific examples of such a heavy element label base containing a complex coordinated with one heavy element atom are as follows. In addition, these compounds are shown in a state of being paired with each base on DNA.

【0064】

[Chemical 4]

<img file="JP2002153271A_D0004.tif" />According to this example, A for each base T, G, C, A on DNA^*(In this case, labeled with palladium (atomic number 46)), C^*(In this case, labeled with Europium (atomic number 63)), G^*(In this case, labeled with platinum (atomic number 78)), U^*(In this case, labeling with cobalt (atomic number 27)) selectively creates a pair bond, and each heavy element on it acts as a reporter that can be discriminated by an electron microscope. When such a heavy element label base is used, if the spot of the platinum atom is discriminated by the electron microscope image, the base of the corresponding single-stranded DNA is identified as C. The other three types of bases are similarly determined by discriminating and identifying the spots of metal atoms. The example shown above is the case of the simplest base-metal complex bond, base-organometallic complex bond, and contains one heavy element. The anchor position of the heavy element complex is nitrogen, but in the case of A and G, there is an arbitrary 7th or 9th position, but the structural formula shows only the 7th position.

【0065】

Of course, it is also possible to combine at both the 7th and 9th positions instead of either the 7th or 9th position. In this case, the number of heavy elements increases, and the resolution and contrast of the electron microscope image increase accordingly. However, better discrimination is possible.

【0066】

On the other hand, if element clusters can be used for heavy element labels, the image contrast will be high, and the sensitivity and reliability of element discrimination will be significantly improved. Also, since the clusters are large, the required resolution may be low, and a low voltage (for example, 100 kV) phase-contrast microscope visualizes the DNA sequence to a sufficient extent.

【0067】

Naturally occurring iron-sulfur clusters or their metal substituents are the most promising candidates for such elemental clusters. The iron-sulfur cluster has a cube (cubane structure), and iron and sulfur are coordinated alternately at eight vertices. Therefore, it becomes a cluster label of 4 heavy elements. It should be noted that such clusters of metal elements also exist as complexes, and in the present invention, those composed of these clusters are also referred to as "heavy element complexes" or simply "complexes".

【0068】

In nature, it is usually incorporated into a protein and anchored to the protein's backbone via cysteine (Cys), as shown below.

【0069】

[Chemical 5]

<img file="JP2002153271A_D0005.tif" />Although this cluster is originally unstable in water, it is protected by proteins and exists stably in the living body. Many iron-sulfur cluster-containing proteins such as ferredoxin and hydrogenase are known. Also, in organic solvents, the clusters remain naked and stable, even if they are not protected by proteins. Therefore, if this cluster can be anchored to a base directly or via an appropriate adapter or linker, a strong heavy element cluster label base can be produced.

【0070】

The heavy elements constituting these clusters include various heavy elements in addition to iron (atomic number 26), and for example, tungsten (atomic number 74) and molybdenum (atomic number 42) are stably incorporated. By combining these metals, for example, a metal cluster containing two iron atoms and two molybdenum atoms can be formed. In this way, heavy element-sulfur clusters with different total atomic numbers (the sum of the atomic numbers of all the constituent metal elements, which is 26 × 2 + 42 × 2 = 136 in the above case) are formed and are base-specific. By labeling with, a heavy element cluster label base that can be discriminated from the base can be obtained.

【0071】

v) Choice of organic solvent Further, in order to treat the elongated DNA on the support membrane with the above-mentioned base derivative containing a heavy element to form a base-specific heavy element label, the organic solvent used in the treatment has a great influence. .. That is, from the viewpoint of base pair selectivity, a non-polar or low-polarity organic solvent such as carbon tetrachloride, cyclohexane, or benzene is generally preferable, but from the viewpoint of solubility, it has polarity such as chloroform, toluene, aniline, or pentanol. The solvent is preferable, and the solvent to be used is selected in consideration of these points. The solvent used in the present invention varies depending on the type of base derivative containing a heavy element used for pairing, and when the above-mentioned heavy element label base is used, chloroform, toluene, carbon tetrachloride are exemplified. ..

【0072】

4) Electron microscope imaging and image analysis system The image of the single-stranded DNA labeled with heavy elements obtained by the improved electron microscope is taken into a computer as digital data and image-processed at high speed to determine the base sequence. Dedicated software for DNA sequencing is installed on a general-purpose graphic computer. Hereinafter, an image analysis system of an image taken by an electron microscope and an enlarged image obtained will be described.

【0073】

1) Imaging and recording media The recording medium used for image analysis needs to be compatible with the magnified image by an electron microscope.

【0074】

The total number of bases contained in the extended single-stranded DNA that is magnified and visualized as an electron microscope image is 10.⁵Adjust to the extent. A microscope image can be taken in about 10 seconds. If this is the throughput of nucleotide sequence determination, 10 of 1 lane of electrophoresis⁵This means that the efficiency is about 10,000 times higher than the base / day separation speed.

【0075】

The area occupied by one base in the image is equivalent to 0.7 nm x 0.7 nm (the base spacing on the single-stranded DNA is a guide). If the screen has a margin of 10 times the occupied area for highly reliable base identification, 0.7 x 0.7 x 10 x 10⁵nm²=(0.7×10³)²nm²One screen will have the actual area of. That is, a space of about 1 μm square becomes the field of view of the microscope. If the pixels are set to 1/2 of the resolution, for example 1.5 nm, the required number of pixels is (0.7 x 10).³nm / 0.15nm)²≒2×10⁷It turns out that. Current high-speed digitized recording media with the equivalent number of pixels are CCD, imaging plate (IP), and YAG (yttrium-aluminum-garnet crystal). However, the size of each pixel is preferably about 3 μm per pixel of high-resolution photographic film. Especially in the case of low electron beam irradiation, IP is excellent in terms of noise characteristics, and the current 25 μm pixel size should be reduced to 10 μm. In that case, the IP size is about 5 cm square.

【0076】

ii) IP high-speed reader In order to perform image analysis processing at high speed, a device that reads recorded images on IP at high speed is required. For this purpose, for example, in order to read an IP of 5 cm × 5 cm in 10 seconds, 100 laser irradiation fluorescence detection sensors of the current reading device are used in parallel.

【0077】

iil) DNA sequence analysis of crossed elongated DNA The main point of this analysis system is that it is possible to analyze not only a completely straight single-stranded DNA but also a bent and slightly crossed single-stranded DNA sequence. If the base spacing between the strands near the intersection is wide, the base can be discriminated from the atomic number-dependent intensities of the spots observed while tracing on a single strand, and the sequence is determined. If there is no intersection, commercially available analysis software can be used.

【0078】

The problem is the distinction between base discrimination and depth direction at intersections. In order to solve this, another image is taken with the same sample tilted by about 30 °, information in the depth direction is obtained, and an image with no overlap is obtained. This makes it possible to distinguish between two overlapping elements. The feature of this system is the analysis that performs such a series of operations.

【0079】

As described above, the base-specific heavy element label obtained by pair-binding the DNA extended on the denatured protein with the heavy element label base indicated by "Chemical Formula 4" is subjected to complex processing using a complex electron microscope. Fig. 2 schematically shows the case of application and visualization.

【0080】

In the magnified image of an electron microscope, each heavy element paired with each base on DNA is discriminated by the size of the image by the difference in electron scattering intensity corresponding to its atomic number, that is, mass.

【0081】

According to Fig. 2, for the sake of conceptual clarity, the difference in image size is not shown (in the figure, the large black circle is A).^*, The small black circle is U^*, The big white circle is G^*And the small white circle is C^*The base sequence of the target DNA can be determined by discriminating the paired heavy element label bases. That is, in FIG. 2, from the upper left, the heavy element label base is A along the DNA.^*A^*U^*C^*G^*A^*..., and therefore the base sequence of the original DNA to be sequenced is determined to be TTAGCT ...

【0082】

[Effect of the invention]

The base sequence of DNA can be determined by selectively modifying the base on the single-stranded DNA with a heavy element label vs. base and discriminatingly reading the heavy element on the electron microscope image, and this DNA base sequence analysis speed. Compared with the current electrophoresis sequencing method,³~10⁴It will be twice as fast.

[Simple explanation of drawings]

[Figure 1]

It is a figure which shows the conceptual diagram of the whole DNA sequencer system of this invention.

[Figure 2]

It is a figure which showed typically the state which visualized the single difference DNA using the method and system of this invention. In the figure, the large black circle is A^*, The small black circle is U^*, The big white circle is G^*And the small white circle is C^*Are schematically shown.