DE10120797A1 - Methods for the analysis of nucleic acid chains - Google Patents

Abstract

The invention relates to a method for analyzing nucleic acid chains. The method is based on the detection of fluorescence signals from individual nucleotide molecules labeled with dyes, which are incorporated into growing nucleic acid chains by a polymerase. The reaction takes place on a flat surface. Many individual nucleic acid molecules are immobilized on this surface. All of these nucleic acid molecules are exposed to the same conditions, so that a build-up reaction can take place simultaneously on all nucleic acid molecules.

Description

The invention relates to a method for analyzing nucleic acid reketten. The basis of the method is the detection of fluo Resence signals of individual nucleotides labeled with dyes molecules created by a polymerase in a growing nucleus acid chain can be installed. The reaction is one plan surface. There are many individuals on this surface Immobilized nucleic acid molecules. All of these nucleic acid Molecules are exposed to the same conditions, so that at all Nucleic acid molecules undergo a building reaction at the same time can.  

The process essentially comprises the following steps:

• 1. Immobilize the nucleic acid chains on a plan Surface.
• 2. Perform a cyclical build-up reaction, each cycle consisting of the following steps:
  • a) adding a solution with labeled nucleotides (NTs ^* ) and polymerase to immobilized nucleic acid chains,
  • b) incubation of the immobilized nucleic acid chains with this solution under conditions which are suitable for extending the complementary strands by one NT,
  • c) washing,
  • d) detection of signals from individual molecules,
  • e) removing the marker from the built-in nucleotides,
  • f) washing.
  If necessary, the cycle is repeated several times.
• 3. Analysis of the detected signals of the individual molecules.
• 4. Reconstruction of the sequences from the individual data.

1. Abbreviations and explanations of terms

DNA - deoxyribonucleic acid of various origins and of different lengths: (genomic DNA, cDNA, ssDNA, dsDNA)
RNA - ribonucleic acid (mostly mRNA)
Polymerases - Enzymes that can incorporate complementary nucleotides into a growing strand of DNA or RNA (e.g. DNA polymerases, reverse transcriptases, RNA polymerases)
dNTP - 2'-deoxi nucleoside triphosphates for DNA polymerases and reverse transcriptases
NTP - nucleoside triphosphates for RNA polymerases
NT - natural nucleotide, usually dNTP, unless expressly stated otherwise.

Abbreviation "NT" is also used when specifying the length of a nucleic acid resequence used, e.g. B. 1000 NT. In this case "NT" stands for Nucleoside monophosphate.

The majority of abbreviations in the text are formed by using the suffix "s", for example "NT" stands for "nucleotide", "NTs" stands for several nucleotides.
NT * - modified nucleotide, usually dNTP, unless expressly stated otherwise. NTs * means: modified nucleotides
NSK - nucleic acid chain. DNA or RNA in their original length
NSKF - nucleic acid chain fragment, NSKFs - nucleic acid chain fragments. Fragments of DNA or RNA that arise after a fragmentation step.

Flat surface

Surface that preferably has the following characteristics comprises: 1) It allows several individual molecules, preferably more than 100, better still more than 1000, with each given lens-surface distance at a lens position to detect simultaneously. 2) The immobilized individual Molecules are in the same focal plane that reproduce bar can be set.

Steric obstacle

Sterically demanding group, whose chemical structure changes the properties of the NTs ^* coupled with this group in such a way that they cannot be successively incorporated into one extension reaction by a polymerase.

Definition of termination

As a termination in this An message the reversible stop of the installation of the modified ununge labeled NTs *.  

This term is from the usual use of the word "termi nation "by dideoxy-NTP in conventional sequencing to separate.

The termination comes after the installation of a modified NT ^* . The modified built-in NT * carries a steric group which is reversibly coupled to the base and which prevents the incorporation of a next complementary NT * into the growing strand by a polymerase.

Gene products - The gene products are the primary gene products of the genes. Essentially, these are RNA transcripts of the genes mentioned, which are also referred to as target sequences (or target nucleic acid sequences). In addition to mRNA, these target sequences also include single-stranded and double-stranded cDNA derived therefrom, RNA derived from cDNA or DNA amplified from cDNA.
SNP - single nucleotide polymorphism

2. State of the art

Nucleic acid chain sequence analysis is in many areas science, medicine and industry to an important work become a witness. Several methods have been developed for analysis.

The best known methods are the chain termination sequences ornamentation according to Sanger (F. Sanger et al. PNAS 1977 v. 74 p. 5463), which is based on the installation of chain terminators, and the Maxam- Gilbert method based on base-specific modification and Cleavage of nucleic acid chains is based (A.M. Maxam and W. Gilbert PNAS 1977, v. 74 p. 560). Both methods provide a number of Nucleic acid chain fragments of various lengths. This Fragments are separated lengthwise in a gel. there all disadvantages of electrophoresis (such as long Runtime, relatively short stretches of sequences in one Approach can be determined, limited number of parallel  Approaches and relatively large amounts of DNA) are accepted become. These methods are very labor intensive and slow.

Another method for sequencing is based on the Hybri dosing of nucleic acid chains with short oligonucleotides. Mathematical methods are used to calculate how many Oligonucleotides of a certain length must be present in order to to determine a complete sequence (Z. T. Strezoska et al. PNAS 1991 v. 88 p. 10089, R. S. Drmanac et al. Science 1993 v. 260 p. 1649). This method also has problems: it can only a sequence can be determined in one approach, secondary structure or interfere with hybridization and repeat repetitions the correct analysis.

Working groups have another option for sequencing by (Dower U.S. Patent 5,547,839, Canard et al. U.S. Patent 5,798,210, Rasolonjatovo Nucleosides & Nucleotides 1999, v. 18 P. 1021, Metzker et al. NAR 1994, v. 22, p. 4259, Welch et al. Nucleosides & Nucleotides 1999, v. 18, p. 197). This Method is abbreviated as BASS (Base Addition Sequencing Scheme) designated. A large number of identical single strands are used pieces of DNA at a defined location on a surface fixed and the signal from the entirety of these many iden table pieces of DNA analyzed. This fixed DNA becomes a Solution with polymerase and nucleotides added, so that a complementary strand can be synthesized. Thereby the Work polymerase step by step: in each step there is only one only nucleotide incorporated. This is detected, whereupon the Polymerase in a next cycle the next nucleotide installs.

The method has the following disadvantages: When building the complementary ren desynchronization of the syn occurs very quickly thesis so that the errors accumulate with each step. Therefore only very short fragments can be sequenced. It It should be emphasized that all described BASS methods are not based on based on the detection of individual molecules. The signal will instead of a large number of identical to a defined one  Location of immobilized molecules registered. The one in these Methods usual use of the terms "single molecules" and "Molecules" is not aimed at individual, mutually different separated molecules but on a population made up of many identical molecules. In this case, identical means that the molecules have the same sequence.

The object of the present invention is therefore a To provide methods for sequence analysis of nucleic acid chains len, which does not have the disadvantages of the methods mentioned above and above all a cheaper, faster and more efficient analysis of nucleic acid sequences. In particular, that should Procedures to be able to determine many sequences in parallel Men, preferably it should analyze very long nucleic acid enable chains (several Mb) in one approach.

3. Brief description

The object is achieved according to the invention by a method for sequence analysis of nucleic acid sequences (nucleic acid chains, NSKs), in which
Fragments (NSKFs) of single-stranded NSKs with a length of about 50 to 1000 nucleotides are generated, which represent overlapping partial sequences of the total sequences
the NSKFs immobilized on a reaction surface in a density of 10 to 100 NSKFs per 100 µm ² (stationary phase), one
cyclically build the complementary strand of NSKFs using one or more primers and one or more polymerases by

• a) adds a solution to the immobilized NSKFs that one or more polymerases and one to four modifi graced nucleotides (NTs *) containing fluorescence dyes are marked, with the simultaneous Use of at least two NTs * each on the NTs *  located fluorescent dyes are chosen so that the NTs * used differ by measurement cher distinguish fluorescence signals from each other let, the NTs * are structurally modified so that the polymerase after incorporation of such an NT * in a growing complementary strand not in the Is able to insert another NT * in the same strand build, the fluorescent dye is cleavable and the structural modification is a cleavable ste is a demanding ligand, one
• b) the stationary phase obtained in stage a) below Conditions incubated to extend the com complementary strands are suitable, the complemen tary strands are extended by one NT * each, one
• c) the stationary phase obtained in stage b) below Washing conditions that are not in one for removal complementary strand of built-in NTs * are suitable, you
• d) the individual NTs built into complementary strands * by measuring that for the respective fluorescent color Characteristic signal detected, where one at the same time the relative position of the individual fluo determined resence signals on the reaction surface, you
• e) to generate unlabeled (NTs or) NSKFs the fluo resence dyes and the sterically demanding Ligands from those attached to the complementary strand Split off NTs *, one
• f) the stationary phase obtained in stage e) below Conditions washes to remove the fluorescence dyes and the ligand are suitable, man

stages a) to f) are repeated several times, if necessary,  where the relative position of individual NSKFs on the reac tion surface and the sequence of these NSKFs by specific Assignment of the in stage d) in successive cycles to the respective positions detected fluorescence signals to the NTs determined and one from the determined overlapping partial sequences the overall sequence of NSKs is determined.

The population of overlapping partial sequences obtained leaves with commercially available programs for the overall sequence of the NSK (Huang et al. Genom Res. 1999 v. 9 p. 868, Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 P. 4992, Miller et al. J. Comput. Biol. 1994 v. 1 p. 257).

According to a particular embodiment of the invention, the process can be carried out by repeating steps a) to f) of the cyclic build-up reaction several times, with

• a) only one marked NT * in each cycle,
• b) in each cycle two differently marked NTs ^* or
• c) four differently marked in each cycle NTs *

starts.

If the NSKs can be variants of a known reference sequence the process can also be carried out by following steps a) to f) the cyclic build-up reaction is repeated several times, wherein you alternate two different cycles marked NTs * and two unmarked NTs and the Total sequences by comparison with the reference sequence averages.

The present invention furthermore relates to the nucleotides shown in FIGS . 7e, 7f and 7g and the corresponding labeled nucleotides which, for example, have fluorescent dyes attached to the terminal amino function, or the labeled nucleotides shown in FIGS . 7h, 7i or 7j.

The invention further relates to a kit for carrying out the method which contains a reaction surface, reaction solutions required for carrying out the method, one or more polymerases, and nucleotides (NTs), one to four of which are labeled with fluorescent dyes, the NTs also structurally so are modified (NT ^* or NTs *) that the polymerase after installing such an NT * in a growing complementary strand is not able to install another NT * in the same strand, the fluorescent dye being removable and the structural Modification is a cleavable, sterically demanding ligand. The nucleotides are preferably the above-mentioned nucleotides according to the invention.

According to a particular embodiment of the invention, this contains Kit also for producing single strands from double strands required reagents, single-stranded nucleic acid molecules, which are introduced as PBS in the KSKFs, oligonucleotide Primer, for splitting off the fluorescent dyes and sterically demanding ligands required reagents and / or washing solutions.

The method according to the invention is used to determine the nucleic acid resequences and can be found in different areas of genetics be used. In particular, this includes the determination unknown long sequences, analysis of sequence polymor phisms and point mutations as well as the parallel analysis of a large number of gene sequences.

The preparation of the material to be analyzed (single and double-stranded nucleic acid sequences) depends on the place of delivery lung and aims to create a long nucleic acid chain Population of relatively small, single-stranded nucleic acid chains fragments (NSKFs) to form these fragments with a for  suitable primers at the start of the sequencing reaction provided and fixed on a flat surface.

Individual NSKFs are placed on a flat surface in one fixed in such a way that an enzymatic reaction to this Molecules can run off. In principle, there are different types of Immobilization possible, depending on the objective, the type of NSK and depend on the polymerase used for the reaction. The NSKFs become random on the surface during immobilization distributed, d. H. So it doesn't have to be an exact positioning of the individual chains. The nucleic acid chains must be fixed in such a density on the surface be that an unambiguous assignment of those later detected Signals to individual NSKFs is guaranteed.

After preparing the NSKFs, you start with everyone at the Surface immobilized molecules the sequencing reaction. The synthesis of the comple serves as the basis for sequencing mental strand to every single immobilized NSKF. there marked NTs * are inserted into the newly synthesized strand builds. The polymerase only builds a single labeled NT * into the growing chain. This is due to the reversible coupling a sterically demanding group leading to the termination reached to the NTs *. The installation of another marked NT * this makes it impossible. This sterically sophisticated Group is preferably a fluorescent dye.

The sequencing reaction proceeds in several cycles. A cycle comprises the following steps:

• a) adding a solution with labeled nucleotides (NTs ^* ) and polymerase to immobilized nucleic acid chains,
• b) Incubation of the immobilized nucleic acid chains with them Solution under conditions that extend the com complementary strands around an NT are suitable,
• c) washing,
• d) detection of signals from individual molecules,
• e) removing the label from the incorporated nucleotides,  
• f) washing.

If necessary, the cycle is repeated several times (a- f).

The reaction conditions of step (b) in one cycle chosen so that the polymerases on more than 50% of those on the Sequencing reaction involved NSKFs in one cycle marked NT * can be installed, preferably on more than 90%.

The number of cycles to be carried out depends on each the task in hand is not theoretically limited and is preferably between 20 and 5000.

Then, for each fixed NSKF, its specific sequence is determined from the order of the installed NTs *. The original NSK sequence can be reconstructed from the overlapping NSKF sequences ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v. 9 p . 868, Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 p. 4992, Miller et al. J. Comput. Biol. 1994 v. 1 p. 257). The entire population of NSKF sequences is searched for matches / overlaps in the sequences of NSKFs. Through these matches / overlaps one can bring the NSKF in a row, e.g. B .:

In practice, there has been a sequencing of unknowns Sequences proven, a length of the sequenced pieces of more to reach 300 bp. This allows the sequencing of Eukaryotic genomes in shotgun process.

The method errors can be done by various means be recorded and corrected. All steps of the process can be largely automated.  

Working with individual molecules has several advantages over the BASS method described earlier:

• 1. Since the molecules are detected individually, there is none Risk of the signal being desynchronized in the Population becomes faulty. For each fixed NSKF created its own sequence. So it doesn't matter whether synthesis continues on a neighboring molecule advanced or lagged.
• 2. It is not necessary to arrange molecules in a defined order fixation on the surface because the signal from single molecules and not spatially defined population (which is necessary for BASS methods is).

4. Detailed description 4.1 General principles of the reaction

In the following, the general principles of the reaction are to be illustrated by way of example using the sequencing of a piece of DNA of several Mb length ( FIG. 1). The sequencing and reconstruction of nucleic acid sequences is based on the shotgun principle ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v. 9 p. 868 , Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 p. 4992, Miller et al. J. Comput. Biol. 1994 v. 1 p. 257). The sequence of a long piece of DNA is determined by sequencing small DNA fragments and subsequent reconstruction. The material to be analyzed (1) is prepared for the sequencing reaction by breaking it down into fragments of preferably 50 to 1000 bp in length (2). Each fragment is then provided with a primer binding site and a primer (3). This mixture of different DNA fragments is now fixed on a flat surface (4). The unbound DNA fragments are removed by a washing step. The sequencing reaction is then carried out on the entire reaction surface. This reaction is cyclical. In the first step of the cycle, an NT * labeled with a fluorescent dye is incorporated into the growing strand: the reaction is controlled so that only one labeled NT ^* from a polymerase can be incorporated into the growing strand in each cycle. This is achieved through the use of NTs * which carry a reversibly coupled group leading to termination. This makes it impossible to install another marked NT *. The polymerase and the labeled NTs * are used simultaneously in the reaction (5). The reaction mixture is then removed and the surface is washed in a suitable manner (6). Now there is a detection step (7): The surface is scanned with a device suitable for single-molecule detection (consisting of light source, microscope, camera, scanning table, computer with control and image recognition or image processing software) and the signals of the individual, built-in are marked NTs * identified. After the detection step, the marking and the group leading to the termination are removed from all installed NTs * (8). After a subsequent washing step, a new cycle can begin. For the reconstruction of a larger original DNA sequence (eg a piece of DNA that is several Mb long), the DNA fragments should be a few hundred NT long if the reconstruction is carried out according to the shotgun principle ("Automated DNA sequencing and analysis" p . 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v. 9 p. 868, Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 p. 4992, Miller et al. J. Comput. Biol. 1994 v. 1 p. 257). Since only one marked NT * is installed per cycle, at least 300 cycles are required for sequencing.

4.2 Selection of the material

With the help of the method according to the invention it is possible to both preselected DNA sequences (e.g. in YAC, PAC, or BAC Vectors (R. Anand et al. NAR 1989 v. 17 p. 3425, H. Shizuya et al. PNAS 1992 v. 89 p. 8794, "Construction of bacterial artificial chromosome libraries using the modified PAC system "in" Current  Protocols in Human Genetics "1996 John Wiley & Sons Inc.) cloned sections of a genome) and not pre-selection to analyze nated DNA (e.g. genomic DNA, cDNA mixtures). A pre-selection makes it possible to find relevant ones in advance Information such as B. sequence sections from a genome or Populations of gene products, from the large amount of genetic Filtering out information and hence the amount of restrict analyzing sequences.

4.3 Preparation of the material

The aim of the material preparation is to obtain immobilized single-stranded NSKFs with a length of preferably 50-1000 NTs, a single primer binding site and a hybridized primer. These fragments preferably have the structure shown in FIG. 2. In particular, very variable constructions can be derived from this general structure. To improve the clarity, here are some examples, whereby the methods listed can be used individually or in combination.

4.3.1 Generation of Short Nucleic Acid Chain Fragments (50-1000 NTs)

It is important that the NSKs are fragmented in such a way that Fragments are obtained, the overlapping partial sequences of the Display overall sequences. This is accomplished through procedures in which fragments of different lengths as fission products in random distribution arise.

According to the invention, the generation of the nucleic acid chains fragments (NSKFs) are made by several methods, e.g. B. by the Fragmentation of the starting material with ultrasound or by Endonucleases ("Molecular cloning" 1989 J. Sambrook et al. Cold Spring Harbor Laborotary Press), e.g. B. by unspecific Endonukleasegemische. According to the ultrasound question mentation preferred. You can set the conditions so that Fragments with an average length of 100 bp to 1 kb arise. These fragments are then terminated  by the Klenow fragment (E. coli polymerase I) or by the T4 DNA polymerase filled in ("Molecular cloning" 1989 J. Sam brook et al. Cold Spring Harbor Laborotary Press).

It can also randomize from a long NSK using complemented short NSKFs are synthesized primer. This method is particularly preferred when analyzing the genetic Sequences. Single-stranded DNA fragments are thereby on the mRNA with randomized primers and a reverse transcriptase formed (Zhang-J et al. Biochem. J. 1999 v. 337 p. 231, Ledbetter et al. J. Biol. Chem. 1994 v. 269 p. 31544, Kolls et al. Anal. Bio chem. 1993 v. 208 p. 264, Decraene et al. Biotechniques 1999 v. 27 P. 962).

4.3.2 Introduction of a primer binding site in the NSKF

The primer binding site is a sequence section which is a to enable selective binding of the primer to the NSKF.

For the sake of simplifying the analysis, it is convenient if a primer binding site that is as uniform as possible in all NSKFs is available. According to a preferred embodiment of the The primer binding sites are therefore invented in the NSKFs specially introduced. This allows primers with more uniform Structure can be used for the reaction.

The composition of the primer binding site is not specified limits. Their length is preferably between 20 and 50 NTs. The primer binding site can be a functional group Wear immobilization of the NSKF. This functional group can z. B. be a biotin group.

As an example of the introduction of a primer binding site in the following the ligation and nucleotide tailing of DNA Fragments described.  

a) Ligation

A double-stranded oligonucleotide complex with a primer binding site is used ( FIG. 3a). This is ligated to the DNA fragments using commercially available ligases ("molecular cloning" 1989 J. Sambrook et al. Cold Spring Harbor Laborotary Press). It is important that only a single primer binding site be ligated to the DNA fragment. This is achieved e.g. B. by a modification of one side of the Oligonucleotidkom complex on both strands ( Fig. 3b). The results after the ligation or after subsequent denaturation are shown in FIGS. 3c and 3d. The modifying groups on the oligonucleotide complex can serve for immobilization. The synthesis and modification of such an oligonucleotide complex can be carried out according to standardized regulations. For synthesis, e.g. B. the DNA synthesizer 380 A Applied Biosystems can be used. Oligonucleotides with a certain composition with or without modifications are also commercially available as custom synthesis, e.g. B. from MWG-Biotech GmbH, Germany.

b) Nucleotide tailing

Instead of ligation with an oligonucleotide, a terminal deoxynucleotidyl transferase can be used to attach several (eg between 10 and 20) nucleoside monophosphates to the 3 'end of an ss-DNA fragment ("Molecular cloning" 1989 J. Sambrook et al Cold Spring Harbor Laborotary Press, "Method in Enzymology" 1999 v. 303, pp. 37-38) ( Fig. 4), e.g. B. several guanosine monophosphates (called (G) n-tailing). The resulting fragment is used to bind the primer, in this example a (C) n primer.

4.3.3 Single strand preparation

Single-stranded NSKFs are used for the sequencing reaction needed. If the starting material is in double-stranded form there are several ways to do it from double-stranded DNA create a single-stranded shape (e.g. heat denaturation or alkali denaturation) ("Molecular cloning" 1989 J. Sambrook et al. Cold Spring Harbor Laborotary Press).  

4.3.4 Primers for the sequencing reaction

This has the function of starting at a single point on the To enable NSKF. It preferably binds to the primer application office in the NSKF. The composition and length of the Pri mers are not restricted. In addition to the start function, the Primers also take on other functions, such as. B. a verb to create the reaction surface. Primers should look like this adapted the length and composition of the primer binding site be that the primer starts the sequencing reaction with of the respective polymerase.

The length of the primer is preferably between 6 and 100 NTs, ideally between 15 and 30 NTs. The primer can Carry a functional group that serves to immobilize the NSKF, such a functional group is preferably a biotin group (see section Immobilization). It is not supposed to be sequencing to disturb. The synthesis of such a primer can e.g. B. with the DNA Synthesizer 380 A Applied Biosystems can be executed or as custom synthesis at a commercial provider, e.g. B. MWG- Biotech GmbH, Germany).

The primer can be analyzed prior to hybridization Fragments fixed on the surface using various techniques or synthesized directly on the surface (McGall et al. U.S. Patent 5412087, Barrett et al. US Patent 5482867, Mirzabekov et al. US Patent 5981734, "Microarray biochip technology "2000 M. Schena Eaton Publishing," DNA Microarrays " 1999 M. Schena Oxford University Press, Fodor et al. Science 1991 v. 285 p. 767, Timofeev et al. Nucleic Acid Research (NAR) 1996, v. 24 p. 3142, Ghosh et al. NAR 1987 v. 15 p. 5353, Gingeras et al. NAR 1987 v. 15 p. 5373, Maskos et al. NAR 1992 v. 20 p. 1679).

The primer or the primer mixture is mixed with NSKFs under Hybridi incubation conditions, which selectively attached it to the primerbin have the NSKF binding agency tied up. This primer hybridization can be before, during or after the immobilization of NSKFs respectively. The optimization of the hybridization conditions depends  the exact structure of the primer binding site and the primer and can be according to Rychlik et al. NAR 1990 v. 18 p. 6409 to calculate. The following are these hybridization conditions referred to as standardized hybridization conditions.

If there is a primer binding site common to all NSKFs known structure introduced by ligation, primers with a uniform structure. The primer at the 3 'end can be a functional group gene, the z. B. is used for immobilization. This is preferably Group a biotin group. The primer has one to the primer complementary structure.

An example of a primer binding site and primer is shown below.

4.3.5 Immobilization

The aim of immobilization is to put NSKFs on a suitable plan to fix surface in a way that a cyclic enzymatic sequencing reaction can take place. Surface and reaction surface are in this application as to understand equivalent terms, unless explicitly referring to one other meaning is pointed out. Serves as the reaction surface the surface of a solid phase of any material. This material is preferably against enzymatic reactions over inert and does not interfere with detection. Silicon, Glass, ceramic, plastic (e.g. polycarbonate or polystyrene) or any other material that makes this functional Requirements can be used. Preferably, the Surface not deformable, otherwise there is a distortion of the signals to be expected in the case of repeated detection.

If a gel-like solid phase is used, this gel can e.g. B. be an agarose or polyacrylamide gel. The gel is preferably freely passable for molecules with a molecular mass below 5000 Da (for example a 1 to 2% agarose gel or 10 to 15% polyacrylamide gel can be used). Such a gel surface has the advantage over other solid surfaces that the marked NSKFs, in contrast to free NTs *, are held on the surface. The immobilization of the NSKFs on the surface makes it possible to detect the fluorescence signals from built-in NTs *. The signals from free NTs * are not detected because they do not bind to the material of the gel and are therefore not immobilized. The gel is preferably attached to a solid base ( Fig. 5a). The thickness of the gel is preferably not more than 0.1 mm. However, the gel thickness must be greater than the simple depth of field of the lens, so that NTs * bound non-specifically to the solid base do not reach the focal plane and are therefore detected. If the depth of field e.g. B. is 0.3 microns, the Gelicke preference is between 1 micron and 100 microns. The surface can be produced as a continuous surface or as a discontinuous surface composed of individual small components (e.g. agarose beads) ( FIG. 5b). The reaction surface must be large enough to be able to immobilize the necessary number of NSKFs with the appropriate density. The reaction surface should preferably not be larger than 20 cm ² .

The different cycle steps require an exchange of the different reaction solutions above the surface. The reaction surface is preferably part of a reaction vessel. The reaction vessel is in turn preferably part of a reaction apparatus with a flow device. The flow device allows an exchange of the solutions in the reaction vessel. The exchange can take place with a pump device controlled by a computer or manually. It is important that the surface does not dry out. The volume of the reaction vessel is preferably less than 50 μl. Ideally, its volume is less than 1 µl. An example of such a flow system is given in FIG. 6.

The NSKFs are preferably immobilized on one of the two chain ends. This can be done by appropriate covalent, affine or other bonds can be achieved. There are many Examples of immobilization of nucleic acids are known (McGall et al. U.S. Patent 5412087, Nikiforov et al. US patent 5610287, Barrett et al. U.S. Patent 5482867, Mirzabekov et al. U.S. patent 5981734, "Microarray biochip technology" 2000 M. Schena Eaton Publishing, "DNA Microarrays" 1999 M. Schena Oxford University Press, Rasmussen et al. Analytical Biochemistry v. 198, p. 138, Allemand et al. Biophysical Journal 1997, v. 73, p. 2064, Trabesin ger et al. Analytical Chemistry 1999, v. 71, p. 279, Osborne et al. Analytical Chemistry 2000, v. 72, p. 3678, Timofeev et al. Nucleic Acid Research (NAR) 1996, v. 24 p. 3142, Ghosh et al. NAR 1987 v. 15 P. 5353, Gingeras et al. NAR 1987 v. 15 p. 5373, Maskos et al. NAR 1992 v. 20 p. 1679).

The NSKFs are immobilized on the surface preferably in a density between 10 and 100 NSKFs per 100 µm ² . Two methods of immobilization are described in more detail below by way of example: In a preferred embodiment, the NSKFs are immobilized via biotin-avidin or biotin-streptavidin binding. Avidin or streptavidin is covalently bound on the surface, the 5 'end of the primer contains biotin. After hybridization of the labeled primers with the NSKFs, they are fixed on the surface modified with avidin / streptavidin. The concentration of the hybridization products labeled with biotin and the time of incubation of this solution with the surface is selected so that a density suitable for sequencing is achieved.

In another preferred embodiment, the primers suitable for the sequencing reaction are fixed on the surface using suitable methods before the sequencing reaction (see above). The single-stranded NSKFs, each with one primer binding site per NSKF, are thus incubated under hybridization conditions. They bind to the fixed primers and are thereby immobilized. The concentration of the single-stranded NSKFs is chosen so that an immobilization density of 10 to 100 NSKFs per 100 µm ² suitable for sequencing is achieved. After immobilization, unbound NSKFs are removed by a washing step. If the surface of a solid phase (e.g. silicone or glass) is used for immobilization, a blocking solution is preferably applied to the surface before step (a) in each cycle in order to avoid non-specific adsorption of NTs * on the surface serves. For example, an albumin solution (BSA) with a pH between 8 and 10 fulfills these conditions for a blocking solution.

4.4 Choice of polymerase

In principle, all DNA-dependent DNA- Polymerases without 3'-5 'exonuclease activity (DNA replication " 1992 Ed. A. Kornberg, Freeman and company NY), e.g. B. modified Sequenase Version 2 type T7 polymerase (Amersham Pharmacia Biotech), 3'-5 'exonuclease-free Klenow fragment of the DNA poly merase I (Amersham Pharmacia Biotech), polymerase beta ver of various origins (Animal Cell DNA Polymerases "1983, Fry M., CRC Press Inc., commercially available from Chimerx) thermostable Polymerases such as Taq polymerase (GibcoBRL), proHATM polymerase (Euro GM).

Polymerases with 3'-5 'exonuclease activity can be used (e.g. Klenow fragment of E. coli polymerase I), provided that reaction conditions are selected, suppress the existing 3'-5' exonuclease activity, e.g. , B. a low pH (pH 6.5) in the Klenow fragment (Lehman and Richardson, J. Biol. Chem. 1964 v. 239 p. 233). Another possibility is to use NTs ^* with a phosphorothioate compound (Kunkel et al. PNAS 1981, v. 78 p. 6734). Built-in NTs * are not attacked by the 3'-5 'exonuclease activity of the polymerase. All these types of polymerases are referred to below as "polymerase".

4.5 chemistry 4.5.1 General principle

For the sequencing reaction, it is important that each is built in NT * is identified. A prerequisite for this is that only one only NT * is inserted into the nucleic acid chain per cycle. If a polymerase has several NTs * in succession in the same cycle installed, this leads to an error in the sequence determination. For this reason you have to control the installation of the NTs *. In this There are various ways of controlling the context enzymatic reaction conceivable, such as. B.

• 1. Single addition of NT so that the polymerase only one single step and stop (substrate limitation),
• 2. Changes in external conditions (temperature, pH, salt etc.), so that the polymerases perform their function differently (see optimization of reaction conditions),
• 3. by certain competitive and non-competitive inhibitors (antiviral drugs are an example)
• 4. Through certain mutations and other changes to the Polymerase as such,
• 5. Due to the structure of the NT. These structural changes can be:
  • a) irreversible (such as with dideoxyNTPs)
  • b) reversible.

The control is preferably carried out by a certain structure of the NTs *, with reversible structural changes especially are preferred. The reversible structural changes take place for example by attaching a sterically sophisticated Ligands, i.e. H. of a ligand that prevents the polymerase from incorporation another NT holds. The control due to a steric Blocking of the enzymatic reaction by reversible semi terminators are described in more detail below. According to one preferred embodiment of the invention, the bases by Coupling of a fluorescent dye structurally changed or modified.  

There are several basic variants of the NT * structure that are used for Disability or stop of the extension reaction leads to: B. were reversible 3'-OH modified NTs in the BASS method (Dower U.S. Patent 5,547,839, Canard et al. U.S. Patent 5,798,210, Rasolonjatovo Nucleosides & Nucleotides 1999, v. 18 P. 1021, Metzker et al. NAR 1994, v. 22, p. 4259, Welch et al. Nucleosides & Nucleotides 1999, v. 18, p. 197). The split is supposed to thereby photochemically under mild conditions (Dower US Patent 5,547,839, Welch et al. Nucleosides & Nucleotides 1999, v. 18 P. 197) or chemical (Canard et al. U.S. Patent 5,798,210, Rasolonjatovo Nucleosides & Nucleotides 1999, v. 18 p. 1021) respectively.

The synthesis of the 3'-OH-modified photochemically cleavable NTs ^* is very complex. The polymerases have a very different affinity for these nucleotide analogs, so that the nucleic acid synthesis is very uneven or does not take place at some DNA sites (Metzker et al. NAR 1994 v. 22 p. 4259, Welch et al. Nucleosides & Nucleotides 1999, v. 18 p. 197). For this reason, these analogs are only of limited suitability for the sequencing reaction. A cleavable 3'-ester linkage (Canard et al. US Pat. No. 5,798,210) is also not considered as the basis for a reversible termination of the synthesis. Most polymerases cleave 3'-OH ester compounds when a next complementary NT is available, so that the label bound to the 3'-OH group is released into the solution and can no longer act as a terminator (Rasolonjatovo et al. Nucleosides & Nucleotides 1999, v. 18 p. 1021, Canard et al. PNAS 1995 v. 92 p. 10859). In positions where a polymerase can insert several identical NTs * one after the other, this leads to an incorrect signal.

According to the invention, NTs ^* with a sterically demanding group on the base are used for sequencing. A sterically demanding group coupled to the base can hinder further synthesis. Biotin, digoxigenin and fluorescent dyes such as fluorescein, tetramethylrhodamine, Cy3 dye are examples of such a sterically demanding group (Zhu et al. Cytometry 1997, v. 28, p. 206, Zhu et al. NAR 1994, v. 22, p. 3418, Gebeyehu et al., NAR 1987, v. 15, p. 4513, Wiemann et al. Analytical Biochemistry 1996, v. 234, p. 166, Heer et al. BioTechniques 1994 v. 16 p. 54). The polymerases can incorporate such NTs * at larger intervals, but have difficulty in integrating these NTs ^* one after the other.

In the sequencing reaction in the method according to the invention are labeled NTs * with a polymerase and nucleic acid chains incubated. The NTs * are reversibly attached to the base coupled sterically demanding group. If a reaction mixture containing only modified NTs * in the reaction then the polymerase can only use a single NT * Install. The installation of a next NT * is sterically inhibited. These NTs * thus act as terminators of the synthesis. To removing the sterically demanding group can do that next complementary NT * will be installed. Because these NTs * no represent absolute obstacle to further synthesis, but only for the installation of another marked NT *, they are called Termed semiterminators.

The difference to the 3'-OH terminator method is that not a blockade of the 3'-OH group necessary for the synthesis is aimed for, but a group linked to the base as steric obstacle is used for further installation. The 3'-OH group remains free all the time.

4.5.2 General structure of the NT *

The semiterminators can have different structures. Their common features are shown in Fig. 7a. This structure is characterized in that a steric group (D) and the fluorescent marker (F) are bound to the base via a cleavable linker (AE).

Deoxynucleoside triphosphates also serve as the basis for the NTs * Adenosine (A), guanosine (G), cytidine (C) and thymidine (T) as Nucleoside residue. Instead of thymidine, uridine is preferred as  Nucleoside residue used. Instead of guanosine, inosine can be used be applied.

4.5.3 Markers, fluorophores

Each base is marked with a marker (F) which is characteristic of it ( FIG. 7). The marker is a fluorescent dye. Several factors influence the choice of the fluorescent dye. The choice is not restricted if the dye meets the following requirements:

• a) The detection equipment used must be this marker as only molecule bound to DNA under mild conditions (before can also identify reaction conditions). The color fabrics preferably have great photostability. Your fluorescence is preferably not or only insignificantly from the DNA quenches.
• b) The dye bound to the NT must not be irreversible Cause disturbance of the enzymatic reaction.
• c) NTs * marked with the dye must be from the polymerase in the nucleic acid chain are installed.
• d) When marking with different dyes these dyes have no significant overlap in theirs Have emission spectra.

Fluorescent color that can be used in the context of the present invention fabrics are in "Handbook of Fluorescent Probes and Research Chemicals "6th ed. 1996, R. Haugland, Molecular Probes Structural formulas compiled. According to the invention preferably the following dye classes are used as markers: Cyanine dyes and their descendants (e.g. Cy2, Cy3, Cy5, Cy7 Amersham Pharmacia Biotech, Waggoner U.S. Patent 5,268,486), Rhodamine and its descendants (e.g. TAMRA, TRITC, RG6, R110, ROX, Molecular Probes, see Manual), xanthene derivatives (e.g.  Alexa 568, Alexa 594, Molecular Probes, Mao et al. US Patent 6130101). These dyes are commercially available.

You can do this depending on the spectral properties and available Select appropriate dyes for the equipment. The dyes are sent to the linker z. B. via thiocyanate or ester bond coupled ("Handbook of Fluorescent Probes and Research Chemi cals "6th ed. 1996, R. Haugland, Molecular Probes, Jameson et al. Methods in Enzymology 1997 v. 278 p. 363, Waggoner Methods in Enzymology 1995 v. 246 p. 362), p. Examples 1 and 2.

4.5.4 Nature of the sterically demanding group

Group (D) ( FIG. 7) represents an obstacle to the incorporation of a further complementary labeled NT * by a polymerase. Biotin, digoxigenin and fluorescent dyes are examples of such a sterically demanding group (Zhu et al. Cytometry 1997, v. 28, p. 206, Zhu et al. NAR 1994, v. 22, p. 3418, Gebeyehu et al., NAR 1987, v. 15, p. 4513, Wiemann et al. Analyti cal Biochemistry 1996, v. 234, P. 166, Heer et al. BioTechniques 1994 v. 16 p. 54). The chemical structure of this group is not restricted provided that it does not significantly interfere with the incorporation of the labeled NT * to which it is attached and does not cause an irreversible disturbance in the enzymatic reaction.

This group can appear as an independent part in the linker ( 7 a) or can be identical to the dye ( 7 b) or the cleavable group ( 7 d). By cleaving the linker, this sterically demanding group (D) is removed after detection of the signal, so that the polymerase can incorporate another labeled NT *. With a structure as in Figure 7d, the steric group is removed by the cleavage.

In a preferred embodiment, the fluorescent dye takes over the function of such a sterically demanding group, so that a labeled nucleotide has a structure shown in FIG. 7b.

In another preferred embodiment, the photolabile cleavable group takes over the function of such a sterically demanding group ( FIG. 7d).

4.5.5 Left

The marker is preferably bound to the base via a spacer of different lengths, a so-called linker. Examples of linkers are given in Fig. 7e, f, h, i, j. This linker is preferably attached to the base at one of the sites which does not participate in the base pairing. In the preferred case, the positions to which the linker is attached are: the 5-position in the pyrimidine ring and the 7-position in the purine ring. Examples of coupling a linker to the base can be found in the following sources (Hobbs et al. US Patent 5,047,519, Khan et al. US Patent 5,821,356, Hanna M. Method in Enzymology 1996 v. 274, p. 403, Zhu et al. NAR 1994 v. 22 p. 3418, Herman et al. Methods in Enzymology 1990 v. 184 p. 584).

The total length of the linker can vary. It corresponds to the number of carbon atoms in sections A, C, E and is preferably between 3 and 20. Optimally, it is between 4 and 10 atoms. The chemical composition of the linker (sections A, C, E in FIG. 7a) is not restricted, provided that it remains stable under reaction conditions and does not cause any disturbance in the enzymatic reaction.

4.5.6 Fissile connection, splitting

The linker carries a fissile compound or fissile group (B). This fissile connection enables the removal of the Markers and the steric obstacle at the end of each cycle. Your choice is not restricted, provided it is under the conditions the enzymatic sequencing reaction remains stable, no irreversible polymerase disruption caused and under mild conditions can be split off. Under "mild Conditions "are to be understood as conditions that the NSKF Do not destroy the primer complex. B. the pH is preferred  is between 3 and 11, the temperature between 0 ° C and a temperature value (x). This temperature value (x) depends on the Tm of the NSKF primer complex (Tm is "melting point") and will calculated as Tm (NSKF primer complex) minus 5 ° C (e.g. Tm 47 ° C, then the maximum temperature is 42 ° C; Under these Conditions are particularly suitable for disulfide compounds and photolabile compounds as cleavable compounds).

The compound mentioned preferably belongs to chemical or enzymatically cleavable or photolabile compounds. As Examples of chemically cleavable groups are ester and Disulfide compounds preferred (Herman et al. Method in Enzymology 1990 v. 184 p. 584, Lomant et al. J. Mol. Biol. 1976 v. 104 243, "Chemistry of carboxylic acid and esters" S. Patai 1969 Interscience Publ.). Examples of photolabile compounds can be found in the following references: "Protective groups in organic synthesis "1991 John Wiley & Sons, Inc., V. Pillai Synthesis 1980 p. 1, V. Pillai Org. Photochem. 1987 v. 9 P. 225.

The position of the fissile link / group in the linker is preferably no more than 10 atoms from the base, more preferably no more than 3 atoms. Particularly preferred the cleavable compound or group lies directly on the base.

The cleavage and removal step is before each cycle act and must run under mild conditions (see above), so that the nucleic acids are not damaged or modified.

The cleavage is preferably chemical (e.g. in mild acidic or basic environment for an ester compound or by adding a reducing agent, e.g. B. Dithiothreitol (Sigma) at the Cleavage of a disulfide compound), see Example 1, or physically (e.g. by illuminating the surface with light a certain wavelength for the cleavage of a photolabile Group).  

After the cleavage, a linker residue (A) remains on the base ( FIG. 7c).

The synthesis of a cleavable linker is shown using examples (see Examples 1 and 2).

4.5.7 Combination of polymerase and NT *

Overall, the size, the charge and the chemical structure of the marker, the length of the cleavable linker and the linker residue and also the choice of polymerase play an important role. Together they determine whether the labeled NT * is incorporated into the growing nucleic acid chain by the polymerase and whether this prevents the insertion of the next labeled NT *. Two conditions are particularly important:
On the one hand, it is important that the polymerase can further extend the nucleic acid chain with the built-in modified NT * after the linker has been cleaved. It is therefore important that the linker residue "A" ( FIG. 7c) after cleavage does not represent a significant disturbance for the further synthesis. On the other hand, built-in, non-split NTs * must be an obstacle. Many NTs * suitable for the reaction can be synthesized. In particular, a preliminary test series must be carried out for each combination of polymerase and NTs ^* , in which the suitability of a specific NT * type for sequencing is tested.

The buffer conditions are chosen according to the polymerase manufacturer. The reaction temperature is chosen for non-thermostable polymerases according to the manufacturer (e.g. 37 ° C for Sequenase Version 2 ), for thermostable polymerases (e.g. Taq polymerase) the reaction temperature is at most the temperature value (x). This temperature value (x) depends on the Tm of the NSKF primer complex and is calculated as Tm (NSKF primer complex) minus 5 ° C (e.g. Tm is 47 ° C, then the maximum reaction temperature is 42 ° C). These buffer conditions and reaction temperature are further referred to as "optimal buffer and temperature conditions".

The reaction time (corresponds to the duration of the installation step in one cycle) is preferably less than one hour, ideally the reaction time is between 10 sec and 10 min.

The following combinations may be mentioned as examples of suitable combinations between NT * and polymerase:

• a) NT * with a short linker residue (synthesis see Example 2, Fig. 7e, h, i): dNTP-SS-TRITC (L7), dNTP-SS-Cy3 (L11) in combination with Sequenase Version 2, Taq- Polymerase (GibcoBRL), ProHA-DNA polymerase (Eurogentec) or DNA polymerase beta from mammals (Chimerx).
• b) NT * with a long linker residue (synthesis see Example 1, Fig. 7f, g, j): dNTP-SS-TRITC (L14) in combination with Sequenase version 2 or or ProHATM polymerase (Eurogentech).

The suitability of a link remnant at base (A) for the reak tion can be checked in a test system. This will split NTs * built into a nucleic acid chain one after the other. you uses z. B. dUTP * with the desired split linker residue, poly-dA as a template, oligo-dT20 primer, the desired polymerase and leads under optimal for the respective polymerase Buffer and temperature conditions cause a reaction. The NT * Concentration is preferably between 5 µM and 200 µM. To the reaction is the number of in the nucleic acid chain built-in NTs * analyzed, e.g. B. by separating the length after in a gel. For conclusions about the suitability of the Leftovers can use the following information: If the Polymerase can incorporate more than 20 NTs *, this is it Linker residue suitable for a sequencing reaction. During installation of less than 20 split NTs * this combination of NT * and polymerase not optimal for the sequencing reaction.  

If a suitable link remnant has been determined, another test system must be used to check whether the marked, non-split NTs ^* function as semiterminators. This is checked by incubating the labeled NTs * with the polymerase and a template under optimal buffer and temperature conditions suitable for the reaction. The NT * concentration is preferably between 5 µM and 200 µM. The matrix must be selected so that the installation of several NTs * would be expected one after the other, e.g. B. for dUTP * you can use polydA, as in the example above ver. Ideally, the polymerase incorporates only a single NT *.

If several NTs * are installed one after the other under given optimal buffer and temperature conditions, the reaction parameters (e.g. NT * concentration, reaction temperature) can be changed and adapted to the respective combination of polymerase and NT ^* . The most important thing is that the polymerase does not incorporate a second NT * in the specified time (preferably between 10 sec and 10 min).

According to the invention, this adaptation takes place in one embodiment by changing the reaction temperature. The other para reaction meters are kept constant.

As polymerases, for example, polymerases are used that Polymerases without 3'-5'-endonuclease activity are preferred from the group consisting of viral polymerases from sequenase Type, beta polymerases from eukaryotes, and thermostable Polymerases. In particular come Sequenase Version 2, Polymerase Beta from mammals, Taq polymerase or ProHATM polymerase in Consideration.

The NT * concentration is usually in these experiments between 5 µM and 200 µN, preferably between 10 and 100 µM. The concentration of the polymerase and the buffer conditions will selected according to the manufacturer. The duration of the reaction can vary and is preferably between 10 sec and 10 min. What correspond to the duration of the installation step (a) in one cycle  would. With non-thermostable polymerases such. B. Sequenase Version 2 (Amersham Pharmacia Biotech), exonuclease free Klenow- Fragment of DNA polymerase I (Amersham Pharmacia Biotech) is the reaction temperature of conventional 37 ° C is preferred reduced to 20 ° C to 30 ° C. With thermostable polymerases such as z. B. Taq polymerase (GibcoBRL), ProHATM polymerase (Eurogentech) the reaction temperature of conventional 70-75 ° C preferably reduced to values between 30 ° C and Temperature value (x). This temperature value (x) depends on the Tm of the NSKF primer complex and is called Tm (NSKF primer Complex) minus 5 ° C calculated (e.g. Tm is 47 ° C, then the Temperature value (x) at 42 ° C).

In another preferred embodiment of the invention, the reaction conditions are adjusted by reducing the NT * concentration to below 5 μM, the other parameters of the reaction (buffer conditions, temperature conditions) are kept constant. With this adaptation, the concentration of the NT * is preferably between 0.5 and 5 μM. The duration of the reaction is between 10 sec and 10 min. The most important thing when choosing the NT ^* concentration is that the polymerase does not incorporate a second NT * in the specified time (preferably between 10 sec and 10 min).

After optimizing the reaction conditions for the installation of a single NT * you have to react with split NTs * again to fetch. Under correspondingly changed reaction parameters Polymerase can incorporate the cleaved NTs * one after the other.

The optimization reaction correlates with the installation step, Step (b) in one cycle. The one for the optimization reaction determined conditions, the temperature, the concentration NT *, the buffer conditions, the duration of the reaction are for the Reaction on the surface adopted.

Under these reaction conditions, the incorporation of NT ^* into the NSKFs is preferably such that a labeled NT * is incorporated into more than 50% of the NSKFs involved in the sequencing reaction, preferably over 90%. This is due to the fact that the reaction takes place very slowly on some nucleic acid chains. An installation of the NTs * at every complementary position in each cycle is aimed for, but is not necessary because only the successful installation reactions are detected and evaluated; a delayed reaction in the subsequent cycle does not lead to a sequencing error.

The same polymerase is preferably used for all NTs *. It can also use different polymerases for different NTs * be used.

4.5.8 Color coding scheme, number of dyes

A cycle can be carried out with:

• a) four differently marked NT * s
• b) two differently marked NT * s
• c) a marked NT *
• d) two differently marked NT * s and two unmarked NTs

d. H.

• a) You can use all 4 NTs with different dyes Mark and use all 4 in the reaction at the same time. The sequencing of a nucleic acid chain is also achieved a minimum number of cycles. This variant of the invention however, places high demands on the detection system: 4 different dyes have to be identified in each cycle become.
• b) To simplify the detection, a label with two dyes can be selected. Two pairs of NTs * are formed, each marked differently, e.g. B. A and G are marked "X", C and U are marked "Y". Two differently marked NTs * are used simultaneously in the reaction in one cycle (s), e.g. B. C * in combination with A ^* , and in the subsequent cycle (n + 1) then U * and G * are added.
• c) You can also use only a single dye for marking Use all 4 NTs * and use only one NTs * per cycle.
• d) In a technically simplified embodiment used two differently marked NT * s per cycle and two unmarked NTs (so-called 2NT * s / 2NTs method). This Embodiment can be used to designate variants (e.g. Muta tions, or alternatively spliced genes) of an already known To determine the sequence.

4.6 Detection equipment

Individual molecules on a surface can be used with different Examine methods. Several methods are known: e.g. B. AtomForce microscopy, electron microscopy, near-field fluo rescence microscopy, wide field fluorescence microscopy, TIR Microscopy etc. (Science 1999 v. 283 1667, Unger et al. BioTech niques 1999 v. 27 p. 1008, Ishijaima et al. Cell 1998 v. 92 p. 161, Dickson et al. Science 1996 v. 274 p. 966, Xie et al. Science 1994 v. 265 p. 361, Nie et al. Science 1994 v. 266 p. 1018, Betzig et al. Science 1993 v. 262 p. 1422).

According to the invention, fluorescence signals are transmitted individually into the Nucleic acid chain of incorporated NTs * preferably with one Wide-field fluorescence microscope or a laser scanning Microscope or a TIRF microscope (Total Internal Reflection Fluorescence Microscope).

Various variants of the construction of such an apparatus are possible (Weston et al. J. Chem. Phys. 1998 v. 109 p. 7474, Trabesinger et al. Anal. Chem. 1999 v. 71 p. 279, Adachi et al. Journal of Microscopy 1999 v. 195 p. 125, Unger et al. BioTech niques 1999 v. 27 p. 1008, Ishijaima et al. Cell 1998 v. 92 p. 161, Dickson et al. Science 1996 v. 274 p. 966, Tokunaga et al. Bichem. Biophys. Res. Com. 1997 v. 235 p. 47, "Confocal Laser Scanning Microscopy" 1997 Ed. Sheppard, BIOS Scientific Publishers, "New Techniques of optical microscopy and microspectroscopy" 1991 Ed. R. Cherry CRC Press, Inc., "Fluorescence microscopy" 1998 2nd ed. Herman BIOS Scientific Publishers, "Handbook of biological confocal microscopy" 1995 J. Pawley Plenum Press). Differences in their concrete structure result from the variation of their individual parts. The device for the excitation light can, for. B. on the basis of a laser, a lamp or diodes function. For the detection device, both CCD cameras and PMT can be used. For other examples of technical details see ("Confocal Laser Scanning Microscopy" 1997 Ed. Sheppard, BIOS Scientific Publishers, "New Techniques of optical microscopy and microspectroscopy" 1991 Ed. R. Cherry CRC Press, Inc., "Fluorescence microscopy" 1998 2 ed. Herman BIOS Scientific Publishers, "Handbook of biological confocal microscopy" 1995 J. Pawley Plenum Press). It is not the object of this invention to list all possible technical variants of a detection device. The basic structure of a suitable apparatus is explained in a diagram in FIG. 8. It consists of the following elements: 1 light source to excite fluorescence
2 Light guiding part
3 scanning table
4 Spectra selection device
5 detection device
6 computers with control and analysis functions

These elements of the apparatus can be purchased commercially (Microscope companies: Zeiss, Leica, Nikon).

In the following, for example, a combination of these elements suitable for the detection of individual molecules will be presented:
Axioplan 2 (Zeiss) wide-field fluorescence microscope with mercury vapor lamp
Lens Planneofluar 100x, NA 1.4 (Zeiss)
Photometrix camera
Computer with software for control and analysis

The procedure for detection is explained below become. Note the general rules of the fluo Resence microscopy ("Confocal Laser Scanning Microscopy" 1997 Ed. Sheppard, BIOS Scientific Publishers, "New Techniques of optical microscopy and microspectroscopy "1991 Ed. R. Cherry CRC Press, Inc., "Fluorescence microscopy" 1998 2nd ed. Herman BIOS Scientific Publishers, "Handbook of biological confocal microscopy" 1995 J. Pawley Plenum Press).

The detection comprises the following phases:

• 1. Preparation for detection
• 2. Execution of a detection step in each cycle, each detection step running as a scanning process and comprising the following operations:
  • a) Setting the position of the lens (X, Y axis),
  • b) setting the focal plane (Z axis),
  • c) detection of the signals of individual molecules, assignment of the signal to NT * and assignment of the signal to the respective NSKF,
  • d) Shift to the next position on the surface.

The signals from NTs * built into the NSKFs are registered by scanning the surface. The scanning process can be carried out in various ways ("Confocal Laser Scanning Microscopy" 1997 Ed. Sheppard, BIOS Scientific Publishers, "New Techniques of optical microscopy and microspectroscopy" 1991 Ed. R. Cherry CRC Press, Inc., "Fluorescence microscopy" 1998 2. ed. Herman BIOS Scientific Publishers, "Handbook of biological confocal microscopy" 1995 J. Pawley Plenum Press). A discontinuous scanning process is preferably selected. The lens is moved step by step over the surface ( FIG. 8a), so that a two-dimensional image (2D image) is produced from each surface position ( FIG. 8b, c).

This 2D image can be created using various methods: z. B. by the laser scan of a position of the microscope field (Laser scanning microscopy) or through a camera recording a position (cf. microscopy manuals). As an an example is the detection of individual molecules with a CCD camera described.

The detection is based on the example of sequencing a 1 Mb long DNA piece explained:

1) Preparation for detection

At the beginning it is determined how many NSKFs to reconstruct the original sequence must be analyzed. In the case of one Reconstruction according to the shotgun method ("Automated DNA sequencing and analysis "p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v. 9 p. 868, Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 p. 4992, Miller et al. J. Comput. Biol. 1994 v. 1 p. 257) the following factors play a role Role: 1) Each NSKF becomes a sequence when sequencing of approximately 300-500 NTs. 2) The total length of the analysis sequence is important. 3) When sequencing a certain level of redundancy can be achieved to the accuracy to increase and correct any errors. Overall is to reconstruct most of the original sequence about 10 to 100 times the amount of raw sequences required, d. H. in this example with one Mb, 10 to 100 Mb Raw sequence data needed. With an average sequence Lengths of 400 bp per NSKF are required from 25,000 to 250,000 DNA fragments.

2) Performing a detection step in every cycle

The positions of the NSKFs must be determined for sequencing, so that one has a basis for the assignment of the signals. Knowing these positions allows a statement to be made as to whether the signals of individual molecules come from built-in NTs * or from randomly bound NTs ^* . These positions can be identified using various methods.

In a preferred embodiment, the positions are immo bilized NSKF identified during sequencing. there the fact is used that the signals from the Nu Small acid chain built NTs * always the same coordinates to have. This is due to the fixation of the nucleic acid chains guaranteed. The non-specifically bound NTs * bind at random different places on the surface.

To identify the positions of fixed NSKFs, the Signals matching their coordinates from several consecutive cycles checked. That can e.g. B. at the beginning sequencing. The matching coordinates are evaluated and saved as coordinates of the DNA fragments chert.

The scan system must be reproducible over several cycles can scan the surface. X, Y and Z axis settings on each Surface position can be controlled by a computer the. Stability and reproducibility of the setting of Ob Lens positions in each scan process determine the quality lity of detection and thus the identification of the Si single molecules.

a) Setting the position of the lens (X, Y axis)

The mechanical instability of the commercially available scan tables and the low reproducibility of repeated inputs Positioning the same X, Y positions make a precise analysis of single molecule signals over multiple cycles difficult. It there are many ways to match the Koor dinaten to improve with repeated settings or poss control deviations. As an example, one Control option mentioned. After a rough mechanical Setting the lens position becomes a control image of a made patterns firmly attached to the surface. Even if the mechanical adjustment not exactly the same coordinates exhibits (deviations up to 10 µm are quite possible), can make a correction using optical control. The  The control image from the sample serves as the coordinate system for the image with signals from built-in NTs *. A prerequisite for one such correction is that no further movements of the waiter area between these two shots. signals of individual molecules are placed in relation to the pattern, so that an X, Y deviation in the pattern position is equal to X, Y- Deviation in the position of the signals of individual molecules is important tet. The control image of the pattern can be before, during or after the Detection of individual molecules can be made. Such a con trollbild must accordingly with each setting on a new one Surface position can be made.

b) Setting the focus plane (Z axis)

The surface is not absolutely flat and has different features Bumps on. This changes the surface lens Distance when scanning neighboring places. Those differences at a distance, individual molecules can cause the Leave the focal plane and thus avoid detection.

For this reason it is important that when scanning the upper a reproducible setting of the focus level on each Lens position is reached.

There are various ways to make the focus level reproducible adjust. For example, the following method can be used : Because the excitation of individual molecules to extinguish their Fluorescence can result in a marker on the surface applied, which is used to adjust the focus level. After that the signals of individual molecules are detected. The marker can be of any nature (e.g. dye or pattern), may but do not interfere with detection and response.

c) detection of the signals of individual molecules, assignment of the Signals to NT * and assignment of the signal to the respective NSKF

The two-dimensional image of the reaction surface generated with the aid of the detection system contains the signal information from NT * s built into the NSKFs. Before further processing, these must be extracted from the total amount of image information using suitable methods. The algorithms required for scaling, transforming and filtering the image information are part of the standard repertoire of digital image processing and pattern recognition (Haberäcker P. "Practice of digital image processing and pattern recognition". Hanser-Verlag, Munich, Vienna, 1995; Galbiati LJ "Machine vision and digital image processing fundamentals ". Prentice Hall, Englewood Cliffs, New Jersey, 1990). The signal extraction is preferably carried out via a gray value image, which depicts the brightness distribution of the reaction surface for the respective fluorescence channel. If several nucleotides with different fluorescent dyes are used in the sequencing reaction, a separate gray value image can first be generated for each fluorescence-labeled nucleotide (A, T, C, G or U). In principle, two methods can be used for this:

• 1. By using suitable filters (Zeiss filter sets) a gray-scale image is generated for each fluorescence channel.
• 2. Using a captured multi-channel color image a suitable algorithm by an image processing program the relevant color channels extracted and each as a gray scale image processed individually. A channel extraction is used for the respective channel-specific color threshold algorithm used. This is how a multi-channel color image is created individual gray-scale images 1 to N. These images are defined as follows:

GB _N

= (s (x, y)) single-channel gray value image
N {1,. , ., Number of fluorescence channels}.
M = {0, 1,. , ., 255} Gray value set
S = (s (x, y)) image matrix of the gray scale image
x = 0, 1,. , ., L-1 image lines
y = 0, 1,. , ., R-1 image columns
(x, y) location coordinates of a pixel
s (x, y) ∈ M gray value of the pixel.

This amount of data is then converted into a suitable program relevant image information extracted. Such a program should implement the following steps:

For GB ₁ to GB _N :

• A) preprocessing the image, for example if necessary Reduction of the by digitizing the image information resulting image noise, for example by gray value smoothing.
• B) Checking each pixel (x, y) of the gray value image, whether this Point related to the immediate and surrounding further away neighboring pixels the properties of a Fluorescence point met. These properties depend on other of the detection equipment used and the resolution of the gray scale image. For example, you can use a typical Distribution pattern of brightness intensity values over one represent the matrix surrounding the pixel. The usable ones Image segmentation methods range from simple threshold process up to the use of neural networks.

If a pixel (x, y) fulfills these requirements, then follows Comparison with the coordinates of previously performed in Sequencing cycles identified NSKFs. With an over the signal is matched with that from the respective fluorescence channel emerging nucleotide to this NACF. Signals with mismatched coordinates are shown as Background signals evaluated and rejected. The analysis of the Signals can be made in parallel with the scanning process.

In an exemplary embodiment, an 8-bit gray scale image was created with a resolution of 1317 × 1035 pixels. To the through digitization has brought about changes in the image reduce, the whole was preprocessed picture: Each pixel was the mean of the brightness assigned to its 8 neighbors. At the chosen resolution this creates a pattern typical of a fluorescence spot a central pixel with the greatest brightness value and  Neighbor pixels with falling on all sides Brightness. Did a pixel meet these criteria and The centrifugal drop in brightness exceeded a certain one Threshold (to exclude weak fluorescence spots), then this central pixel became the coordinate of one Fluorescence point scored.

• a) Moving the lens to the next position on the Surface. After the detection of the signals of individual molecules the lens will be over a different position of the surface positioned.

Overall, for example, a sequence of recordings with the Control of the X, Y position, the adjustment of the focus plane and with the detection of individual molecules for each new object active position. These steps can be done by one Computer controlled.

4.7 Timing of the procedural steps

The scanning process and the biochemical reaction take a certain amount of time. If you switch these processes one after the other, you can achieve optimal performance of the equipment. In a preferred embodiment, the reaction is carried out on two separate surfaces ( FIG. 9).

As an example, a surface with immobilized NSKFs in 2 spatially isolated parts are separated so that reactions to these two parts can run independently. In In another example, NSKFs can also be set to 2 from the start separate surfaces can be immobilized.

The reaction is then started. The principle here is that while on part of the surface the reaction and washing steps, the second part is scanned. This can to achieve a continuous flow of analysis and the Increase the speed of sequencing.  

The number of surfaces on which the reaction takes place can also be larger than 2. That makes sense if the Reaction occurs as a time-limiting step, d. H. the Detection of signals on the surface faster than the reak tion and washing steps. To the total duration of the reaction Each step can be adapted to the detection duration the reaction on a single surface with a temporal Delay compared to the next surface expire.

The invention is illustrated below using examples light.

Examples example 1

Modified dUTP with a long cleavable linker ( Fig. 7f-1) 5- (3-aminoallyl) -2'-deoxyuridine- 5'-triphosphate, AA-dUTP, (Sigma), 3,3'-dithio- are used as starting substances. bis (propionic acid-N-hydroxysuccinimide ester), DTBP-NHS, (Sigma), 2-mercaproethanol amine, MEA, (Sigma). 3 equivalents of DTBP-NHS in DMF (25 µl 0.4M solution) are added to 100 µl 50 mM solution of AA-dUTP in 100 mM borate buffer pH 8.5. The reaction mixture is 4 hours at RT. incubated. Then conc. Ammonium acetate solution (pH 9) is added until the total concentration of CH ₃ COONH ₄ in the reaction solution is 100 mM, and the reaction is incubated for an additional hour. Then 200 μl of 1M MEA solution, pH 9, are added to this mixture and incubated for one hour at RT. A saturated solution of I ₂ in 0.3M KI solution is then added dropwise to this mixture until the iodine color remains. The modified nucleotides are separated from other reaction products on a DEAE cellulose column in an ammonium carbonate gradient (pH 8.5). The nucleotide with the cleavable linker is isolated on RP-HPLC. Dyes can now be coupled to this linker using various methods ("Handbook of Fluorescent Probes and Research Chemicals" 6th ed. 1996, R. Haugland, Molecular Probes, Waggoner Method in Enzymology 1995 v. 246, p. 362, Jameson et al. Method in Enzymology 1997, v. 278, p. 363).

Other nucleotide analogs (e.g. according to Hobbs et al. US Patent 5,047,519, Khan et al. US Patent 5,821,356) can also be used in the reaction, so that nucleotide analogs with structures in FIGS . 7f-2, 3, 4 and 7g -1, 2 can be generated.

The coupling of TRITC (tetramethylrhodamine isothiocyanate) is given as an example of the coupling of a dye to the linker (NT ^* structure Fig. 7j)

The dNTP (300 nmol) modified with the cleavable linker becomes dissolved in 30 ul 100 mM sodium borate buffer pH 9 (10 mM NT *). 10 µl of 10 mM TRITC in dimethylformamide (DMF) are added and incubated at RT for 4 h. Cleaning the with the dye modified NT * takes place via RP-HPLC in a methanol-water Gradient.

Example of cleavage of the disulphide compound in the modified NT ^* . The cleavage is carried out by adding 20 to 50 mM dithiothreitol solution DTT (Sigma) solution pH 8 to the reaction surface. The surface is 10 min. incubated with this solution, then the solution is removed and the surface washed with a buffer solution to remove DTT residues.

Example 2

Modified dUTP (dUTP-SS-CH ₂ CH ₂ NH ₂ ) with a short cleavable linker ( Fig. 7e-1). The starting substances are: bis-dUTP, synthesized according to Hanna (Method in Enzymology 1989, v. 180, p. 383), 2-mercaptoethanolamine (MEA) (Sigma).

100 μl of 100 mM MEA solution pH 8.5 in H ₂ O are added to 400 μl of 100 mM bis-dUTP in 40 mM borate buffer pH 8.5 and incubated for 1 hour at RT. A saturated solution of I ₂ in 0.3M KI solution is then added dropwise to this mixture until the iodine color remains. The nucleotides (bis-dUTP and dUTP-SS-CH ₂ CH ₂ NH ₂ ) can e.g. B. by ethanol precipitation or on a DEAE cellulose column in an ammonium carbonate gradient (pH 8.5) from other reaction products. Bis-dUTP does not interfere with the subsequent coupling of a dye to the amino group of the linker, so that the dUTP-SS-CH ₂ CH ₂ NH _{2 can be separated} from bis-dUTP in the final cleaning step.

DCTP ( Fig. 7-e2) can also be modified in a similar manner, bis-dCTP serving as the starting substance (synthesized according to Hanna et al. Nucleic Acid Research 1993, v. 21, p. 2073).

Dyes can now be attached to the linker using various methods be coupled ("Handbook of Fluorescent Probes and Research Chemicals "6th ed. 1996, R. Haugland, Molecular Probes, Waggoner Method in Enzymology 1995 v. 246, p. 362, Jameson et al. Method in Enzymology 1997, v. 278, p. 363).

The coupling of the FluoroLinkTM Cy3 monofunctional dye (Amersham Pharmacia biotech) (NT * structure Fig. 71) is given as an example of coupling a dye to the linker. It is a monofunctional NHS ester fluorescent dye. The reaction is carried out according to the manufacturer:
The dNTP (300 nmol) modified with the cleavable linker is dissolved in 300 ul 100 mM sodium borate buffer pH 8.5. For this, dye (300 nmol) is added and incubated for 1 h at RT. The NT * modified with the dye is cleaned by RP-HPLC in a methanol-water gradient.

Another example of the coupling of a dye to the linker is the coupling of TRITC (tetramethylrhodamine-5-isothiocyanate, Molecular Probes) (dUTP-SS-TRITC Fig. 7h).

The dNTP (300 nmol) modified with the cleavable linker becomes dissolved in 30 ul 100 mM sodium borate buffer pH 9 (10 mM NT *). 10 µl of 10 mM TRITC are added to DMF and 4 h at RT incubated. Cleaning the NT * modified with the dye takes place via RP-HPLC in a methanol-water gradient.  

Example 3 Sequence analysis with 4 marked NTs *

In a preferred embodiment of the invention, all four NTs * used in the reaction with fluorescent dyes marked.

3A. Reconstruction of the original sequences after the shot shooting principle ("Automated DNA sequencing and analysis" S. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al.   16523 00070 552 001000280000000200012000285911641200040 0002010120797 00004 16404Genom Res. 1999 v. 9 p. 868, Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 p. 4992, Miller et al. J. Com put. Biol. 1994 v. 1 p. 257). (This principle is special suitable for the analysis of new, unknown sequences.) 3A-1 sequencing of a long piece of DNA ( Fig. 1)

In the following, the sequencing of long nucleic acid chains will be shown schematically based on the sequencing of a 1 Mb long piece of DNA ( FIG. 1). The sequencing is based on the shotgun principle ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v. 9 p. 868, Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 p. 4992, Miller et al. J. Comput. Biol. 1994 v. 1 p. 257). The material to be analyzed is prepared for the sequencing reaction by breaking it down into fragments of preferably 50 to 1000 bp in length. Each fragment is then provided with a primer binding site and a primer. This mixture of different DNA fragments is now fixed on a flat surface. The unbound DNA fragments are removed by a washing step. The sequencing reaction is then carried out on the entire reaction surface. To reconstruct a 1 Mb long DNA sequence, the sequences of NSKFs should preferably be longer than 300 NTs, on average approx. 400 bp.

Since only one marked NT * is installed per cycle, at least 400 cycles are required for sequencing.

Overall, the original sequence is to be reconstructed which requires approximately 10 to 100 times the amount of raw sequences Lich, d. H. 10 to 100 Mb. At an average A sequence length of approx. 400 bp per NSKF is required accordingly 25,000 to 250,000 DNA fragments by more than 99.995% to cover the entire sequence.

The NSKF sequences determined represent a population of overlapping partial sequences that deal with commercial available programs to the total sequence of the NSK together have added ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. genome Res. 1999 v. 9 p. 868, Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 p. 4992, Miller et al. J. Comput. Biol. 1994 v. 1 p. 257).

3A-2 sequencing of the gene products using the example of the cDNA sequencing

In a preferred embodiment, instead of one Sequence multiple sequences can be analyzed in one approach. The original sequences can be obtained from the Raw data e.g. B. reconstructed according to the shotgun principle become.

First, NSKFs are created. You can e.g. B. mRNA in a Transfer double-stranded cDNA and this cDNA with Ul fragmenting sonic. Then these NSKFs with provided a primer binding site, denatured, immobilized and hybridized with a primer. Please note at this variant of sample preparation that the cDNA Molecules can represent incomplete mRNA sequences (Method in Enzymology 1999, v. 303, p. 19 and other articles in this volume, "cDNA library protocols" 1997 Humana Press).  

Another possibility when generating single-stranded NSKFs of mRNA consist in the reverse transcription of the mRNA with randomized primers. Many become relative short antisense DNA fragments formed (Zhang-J et al. Biochem. J. 1999 v. 337 p. 231, Ledbetter et al. J. Biol. Chem. 1994 v. 269 p. 31544, Kolls et al. Anal. Biochem. 1993 v. 208 P. 264, Decraene et al. Biotechniques 1999 v. 27 p. 962). This Fragments can then be linked to a primer be provided (see above). Further steps are the same as above described operations. With this method you can complete mRNA sequences (from the 5 'to the 3' end) are analyzed, since the randomized primers span the entire length of the mRNA tie.

Immobilized NSKFs are listed with one of the above Embodiments of sequencing analyzed. Because mRNA Sequences have significantly fewer repetitive sequences as z. B. genomic DNA, the number of detected Signals of the built-in NTs * from an NSKF less than 300 and is preferably between 20 and 1000. The number the NSKFs that need to be analyzed are calculated the same principles as for a shotgun reconstruction tion of a long sequence.

NSKF sequences are made according to the principles of the shot shot method to reconstruct the original gene sequences struiert.

This method allows the sequencing of many mRNAs without prior cloning.

3B. Analysis of sequence variants

Confirmation of an already known sequence or the Detection of variants of this sequence represents a great deal lower demands on the length and redundancy of the determined ten NSKFs. Sequence editing is also in this case easier. The full sequence does not need to be reconstructed  to become. The NSKF sequences are rather using a commercially available full sequence program assigned and any deviations detected. a such a program can e.g. B. BLAST or FASTA algorithm underlying ("Introduction to computational Biology" 1995 M. S. Waterman Chapman & Hall).

The sequence to be analyzed is in NSKFs with one of the shared above method. These NSKFs are with the sequenced inventive methods. Then be the sequences of NSKFs not determined after the shot shot method, but with the reference sequence and in this way their positions in assigned to the full sequence. This can be genomic or cDNA sequences.

In contrast to a reconstruction after the shotgun Procedures are needed for the analysis of a sequence variant significantly less raw sequence data. So the 5- to 10- times the raw sequence quantity is sufficient for the restoration a new variant of a full sequence. With the shot Shot procedure will be a 10- to restoration 100 times the amount of raw sequences required ("Automated DNA sequencing and analysis "p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v. 9 p. 868, Huang Genomics 1996 v. 33 p. 21, Bonfield et al. NAR 1995 v. 23 P. 4992, Miller et al. J. Comput. Biol. 1994 v. 1 p. 257).

The length of the NSKF sequences determined should be for one unique assignment to a specific position in the Reference sequence be sufficient, z. B. already Sequences with a length of 20 NTs (e.g. from not repetitive sections in the human genome) be identified. For the comparative analysis of the repet Itative sections require longer sequences. The Exact length of the sequences depends on the task position. The length of the determined is preferably NSKF sequences when analyzing non-repetitive  Sections more than 20 NTs. For the analysis of repetitive Sections it is preferably over 500 NTs.

The objectives when sequencing new variants an already known full sequence can be very under be different. A comparison of the new ones is usually made th sequence with the known full sequence / reference sequence sought. The two sequences from evolution species differing widely come. Different parameters of the composition of these the two sequences can be compared. As examples of such an analysis serve: mutation or polymorphism analyze and the analysis of alternatively spliced gene products.

The following is an example of a comparison of the below searching sequence with a reference sequence without previous Reconstruction of the sequence to be analyzed become. Such a comparison can e.g. B. for mutation or Serve SNP analysis.

3B-1

A long sequence to be analyzed, e.g. B. 1 Mb, is in NSKFs shared using one of the above methods. This NSKFs are sequenced using the method according to the invention ed. The sequences determined from each individual NSKF are compared directly with the reference sequence. The The reference sequence serves as the basis for the assignment determined NSKF sequences, so that the complex recon Structure according to the shotgun method is omitted. virtue the length of the NSKF sequences determined is the analysis of non-repetitive sections more than 20 NTs. It lies for the analysis of the repetitive sections preferably over 500 NTs. The number of to analyze NSKFs are based on the total length of the investigating sequence, the average length of the NSKF sequences and the necessary accuracy of sequencing.  With an average length of the determined NSKF sequence of 100 NTs, a total length of under Seeking sequence of 1 Mb and an accuracy that the Raw sequence determination corresponds (i.e. each position should be sequenced only once if possible) B. about 5 times the amount of raw sequences, d. H. 5 Mb because that Random distribution of NSKFs over the entire sequence he follows. A total of 50,000 NSKFs must be analyzed in order to cover more than 99% of the total distance.

Then the determined NSKF sequences with the help a commercially available full sequence program assigned and any deviations detected. a such a program can e.g. B. BLAST or FASTA algorithm underlying ("Introduction to computational Biology" 1995 M. S. Waterman Chapman & Hall).

Example 4 Sequence analysis with 2 marked NTs * and 2 unmarked NTs (2NTs ^* / 2NTs method)

In another embodiment, the analysis of the Sequences 2 modified NTs * and 2 unmodified NTs inserted puts.

This method is particularly suitable for analyzing the sequence va rianten (e.g. SNP or mutation analysis) and sets the knowledge ahead of a reference sequence. The full sequence is not reconstructed, but the identified sequences are created using a program assigned to the reference sequence and any Deviations registered. Such a program can e.g. B. the BLAST or FASTA algorithm are based ("Introduction to computational biology "1995 M. S. Waterman Chapman & Hall).

This embodiment is based on the principle that a sequence of 2 signals (marked NT * s) can contain enough information to identify a sequence. The determined sequence is compared with the reference sequence and assigned to a specific position, e.g. B .:

The unknown variant of the reference sequence to be analyzed is prepared for sequencing as described above (NSK will transferred to NSKFs, these are ligated with PBS, then hybridized with a primer and on reaction surface immobilized). NSKFs prepared in this way are included 2NTs * / 2NTs method sequenced. NSKF sequences are obtained, whereby each NSKF sequence represents a sequence of 2NTs *. To one clear assignment of the determined sequence to a known one To enable the reference sequence, this sequence must be long enough his. The length of the NSKF- Sequences more than 40 NT * s. Since 2 marked NTs * only a part represent the sequence is the total length of the synthesized complementary strand about twice as long as the sequence of detected NTs * (with 40 detected NTs * the total is length z. B. on average 80 NTs).

4 nucleotides are used to synthesize a complementary strand needed. Since the NTs marked with a fluorescent dye * appear as semiterminators in the present invention, d. H. the termination is only modifiable when available ornamental NTs * must occur, unmodified NTs in one too added additional step in each cycle in the reaction become. The exact position of this step in the cycle can vary. It is important that the marked NTs * and the unmodified NTs can be used separately.  

A cycle in this embodiment may look like this, for example:

• a) Add a solution with modified NTs * and polymerases on the surface with the NSKFs provided
• b) Incubation of the immobilized nucleic acid chains with them Solution under conditions that extend the com complementary strands around an NT are suitable
• c) washing
• d) detection of signals from individual, modified and in the newly synthesized strands complementary to the NSKFs built-in NTs * molecules
• e) removal of the marker and the terminating group at the built-in nucleotides
• f) washing
• g) addition of 2 unmodified NTs and polymerases
• h) washing.

The concentration of the NTs is preferably below 1 mM, more ideal less than 10 µM.

This 2NT * s / 2NTs method is suitable, for example, for the SNP analysis of a genomic stretch of a gene or for double-stranded cDNA analysis. It is based on the following principles:

• 1. The genetic information in each of the two complementary Ren DNA strands are identical, so missing information in one strand by the information from the other strand can be completed.
• 2. By means of certain combinations of pairs of labeled NTs *, the complete information can be obtained from a double-stranded DNA with only 2NTs *. Permissible combinations of NT * s in this embodiment are: A * C *; A * G *; C * T * / C * U *; G * T * / G ^* U *. The combination C * and U ^* is preferred
• 3. An already known Refe serves as the basis for the analysis Conference sequence.  
• 4. The NSKFs come from both strands of the one to be analyzed NSK and the NSKF sequences determined cover the whole Length of the sequence to be analyzed.

The following example explains how the information is obtained from a double-stranded DNA fragment with only 2 labeled NTs ^* and how the differences to the original or non-mutated sequence (reference sequence / comparison sequence) can be determined. Sequences under (1) and (2) are identical except for one position (underlined). A * and C * are marked.

1) Sequence to be checked

The sequence to be checked is sequenced using the 2NT * s / 2NTs method, so that a population of NSKF sequences (determined NSKF sequences (n)) is formed. The NSKF sequences determined contain information from each strand:

2) comparison sequence

A comparison sequence (reference sequence) is required for analysis:

3) Comparison sequence with adapted NSKF sequences determined

With the help of a program, determined NSKF sequences are assigned to specific points in the comparison sequence and possible deviations are detected:

With this embodiment, a double-stranded nucleic acid can be examined for SNP or mutations. The NSKF sequences determined are compared with a reference sequence. The basic rules of comparing a partial sequence and a complete sequence in the analysis with only 2 marked NTs do not differ fundamentally from those that apply when comparing the sequences using all 4 marked NTs *. For more information, see Sequence comparison in mutation analysis and SNP analysis with 4NTs ^* (Example 3B).

SEQUENCE LISTING

Claims (42)

1. Method for sequence analysis of nucleic acid sequences (Nu small acid chains, NSKs), in which one
Fragments (NSKFs) of single-stranded NSKs with a length of about 50 to 1000 nucleotides are generated, which represent overlapping partial sequences of the total sequences
the NSKFs immobilized on a reaction surface in a density of 10 to 100 NSKFs per 100 µm ² (stationary phase), one
cyclically build the complementary strand of NSKFs using one or more primers and one or more polymerases by

• a) a solution is added to the immobilized NSKFs which contains one or more polymerases and one to four modified nucleotides (NTs *) which are labeled with fluorescent dyes, the simultaneous use of at least two NTs * each on the NTs * located fluorescent dyes are selected so that the NTs * used can be distinguished from one another by measuring different fluorescence signals, the NTs * being structurally modified such that the polymerase is unable to incorporate such an NT * into a growing complementary strand, to install another NT * in the same strand, the fluorescent dye being cleavable and the structural modification being a cleavable sterically demanding ligand, one
• b) the stationary phase obtained in stage a) is incubated under conditions which are suitable for the extension of the complementary strands, the complementary strands being extended by one NT * each, one
• c) the stationary phase obtained in stage b) is washed under conditions which are not suitable for the removal of NTs * which are not incorporated into a complementary strand
• d) the individual NTs * built into complementary strands are detected by measuring the signal characteristic of the respective fluorescent dye, at the same time determining the relative position of the individual fluorescent signals on the reaction surface, one
• e) to generate unlabelled (NTs or) NSKFs, the fluorescent dyes and the sterically demanding ligands are split off from the NTs * attached to the complementary strand, one
• f) the stationary phase obtained in step e) is washed under conditions which are suitable for removing the fluorescent dyes and the ligands, one

stages a) to f) are repeated several times, if necessary,
wherein the relative position of individual NSKFs on the reaction surface and the sequence of these NSKFs are determined by specific assignment of the fluorescence signals detected in step d) in successive cycles at the respective positions to the NTs, and the total sequence of the NSKs is determined from the overlapping partial sequences determined , 2. The method according to claim 1, characterized in that one stages a) to f) of the cyclic build-up reaction several times repeated, with only one in each cycle marked NT *.   3. The method according to claim 1, characterized in that one stages a) to f) of the cyclic build-up reaction several times repeated, with two under each cycle uses differently marked NTs *. 4. The method according to claim 1, characterized in that stages a) to f) of the cyclic build-up reaction several times repeated, with four under each cycle uses differently marked NTs *. 5. The method according to claim 1, characterized in that the NSKs are variants of a known reference sequence stages a) to f) of the cyclic build-up reaction several times repeated, alternating two cycles each differently marked NTs * and two unmarked NTs and the total sequences are compared by comparison with the Reference sequence determined. 6. The method according to claims 1 to 5, characterized records that in the NSKFs one primer introduction site (PBS), whereby with double-stranded NSKs on each of two complementary single strands Introduces PBS and where the primer binding sites for everyone NSKFs each have the same or different sequences. 7. The method according to claims 1 to 6, characterized records that the NSKFs with primers under conditions in Brings contact to the hybridization of the primer to the Primer binding sites (PBSs) of the NSKFs are suitable, where the primers mutually identical or different sequences have, and then the NSKFs on the reaction immobilized surface. 8. The method according to claims 1 to 6, characterized records that the NSKFs first on the reaction surface immobilized and only then with primers under Contact conditions that lead to the hybridization of the Primer to the primer binding sites (PBSs) of the NSKFs  are suitable, the primers being the same as or have different sequences. 9. The method according to claims 1 to 6, characterized records that the primers are first on the reaction surface immobilized and only then with NSKFs under Contact conditions that lead to the hybridization of the Primer to the primer binding sites (PBSs) of the NSKFs are suitable, the primers being the same as or have different sequences. 10. The method according to claims 1 to 9, characterized records that the one used for structural modification sterically demanding ligand the one used for labeling Is fluorescent dye. 11. The method according to claims 1 to 10, characterized records that the reaction surface from the group consisting of silicone, glass, ceramics, plastics, gels is selected. 12. The method according to claim 11, characterized in that the Plastics are polycarbonates or polystyrenes. 13. The method according to claim 11, characterized in that the Gels are agarose or polyacrylamide gels. 14. The method according to claim 13, characterized in that the Gels 1 to 2% agarose gels or 10 to 15% polyacrylamide Gels are. 15. The method according to claims 1 to 14, characterized in net that the DNA polymerase is a DNA polymerase without 3'-5'- Endonuclease activity is. 16. The method according to claim 15, characterized in that the Polymerase from the group consisting of viral DNA polymerases  of the sequenase type, beta polymerases from eukaryotes and thermostable polymerases is selected. 17. The method according to claim 16, characterized in that the DNA polymerase Sequenase version 2, polymerase beta from Mammals, Taq polymerase or ProHA DNA polymerase. 18. The method according to claim 16, characterized in that the DNA polymerase Sequenase version 2 is. 19. The method according to claim 16, characterized in that the DNA Folymerase is Taq polymerase. 20. The method according to claim 16, characterized in that the DNA polymerase is ProHa DNA polymerase. 21. The method according to claim 16, characterized in that the DNA polymerase is DNA polymerase beta from mammals. 22. The method according to claims 1 to 14, characterized in net that the NTs * are alpha-phosphorothioate NTs * and the Polymerase the Klenow fragment of E. coli polymerase I or Is T4 DNA polymerase. 23. The method according to claims 1 to 18, characterized in net that the fluorescent dyes from the group from cyanine dyes, rhodamines, xanthenes and their Derivatives are selected. 24. Kit for performing the method according to claims 1 to 23, characterized in that it is a reaction upper surface (a solid support) for carrying out the procedure rens required reaction solutions, one or more poly merasen, and contains nucleotides (NTs), one to four are marked with fluorescent dyes, the marked NTs are also structurally modified (NT * or NTs *) that the polymerase after installation of such an NT * unable in a growing complementary strand  is to install another NT * in the same string, where the fluorescent dye is removable and the structural Modification of a cleavable sterically demanding Li gand is. 25. Kit according to claim 20, characterized in that the for structural modification used sterically demanding full ligand of the fluorescent color used for labeling is fabric. 26. Kit according to claim 24 or 25, characterized in that it further contains components:

• a) reagents required for producing single strands from double strands,
• b) nucleic acid molecules which are introduced as PBS into the NSKFs,
• c) oligonucleotide primers,
• d) reagents required for cleaving off the fluorescent dyes and bulky ligands,

and or

• a) Wash solutions.

27. Kit according to claims 24 to 26, characterized in that that the reaction surface from the group consisting of Silicon, glass, ceramics, plastics, gels is selected. 28. Kit according to claim 27, characterized in that the art substances are polycarbonates or polystyrenes. 29. Kit according to claim 27, characterized in that the gels Are agarose or polyacrylamide gels. 30. Kit according to claim 29, characterized in that the gels 1 to 2% agarose gels or 10 to 15% polyacrylamide gels are.   31. Kit according to claims 24 to 30, characterized in that the DNA polymerase is a DNA polymerase without 3'-5'- Endonuclease activity is. 32. Kit according to claim 31, characterized in that the poly merase from the group consisting of viral DNA polymerases of the sequenase type, beta polymerases from eukaryotes and thermostable DNA polymerases is selected. 33. Kit according to claim 32, characterized in that the DNA Polymerase Sequenase Version 2, polymerase beta from mammal animals, Taq polymerase or ProHa DNA polymerase. 34. Kit according to claims 24 to 30, characterized in that the NT * are alpha-phosphorothioate NTs * and the polymer sase the Klenow fragment of E. coli polymerase I or T4 Is DNA polymerase. 35. Kit according to claims 24 to 34, characterized in that the fluorescent dyes from the group consisting of Cyanine dyes, rhodamines, xanthenes and their derivatives are selected. 36. Nucleotide of the formula
37. Nucleotide of the formula
38. Nucleotide of the formula
39. Nucleotide according to claim 36 to 38, characterized in that at the terminal amino group a fluorescent dye is pinned.   40. Nucleotide of the formula
41. Nucleotide of the formula
42. Nucleotide of the formula