WO2003080861A1 - Single molecule sequencing using phosphate labeled nucleotides - Google Patents

Abstract

The present invention relates to a process for sequencing nucleic acids wherein the nucleic acid molecule to be sequenced is sequentially degraded in the presence of a fluorescence labeled reagent wherein a fluorescence labeled nucleoside is formed having nucleobase-specific fluorescence characteristics. Preferably the method is a single molecule sequencing procedure comprising a spatially resolved detection step, e.g. confocal detection. Preferably the method comprises a confocal detection step. Further, novel nucleobase-specific fluorescence dyes and reagents containing said dyes are provided.

Description

Single Molecule Sequencing Using Phosphate Labeled Nucleotides

Description

The present invention relates to a process for sequencing nucleic acids wherein the nucleic acid molecule to be sequenced is sequentially degraded in the presence of a fluorescence labeled reagent wherein a fluorescence labeled nucleoside is formed having nucleobase-specific fluorescence ©h-aracteristics. Preferably the method is a single molecule sequencing procedure comprising a spatially resolved detection step, e.g. confocal detection. Preferably the method comprises a confocal detection step. Further, novel nucleobase-specific fluorescence dyes and reagents containing said dyes are provided.

The milestone in genomic sequencing, the determination of the complete human genome, has been reached. Nevertheless, the sequencing of the DNA is still important for example in the context of establishing predispositions for hereditary diseases, finding mutations causing defects or disease, or seeking for genotypes from specific phenotypes. Especially the last task requires even larger capacities for sequencing. Therefore faster techniques for DNA sequencing combined with a reduction of the required amount of material are desirable.

A broad range of new techniques [1 -3] was developed in the past years to offer alternatives for the traditional (gel-based) sequencing methods. One very promising and successful technique here is the use of laser-induced fluorescence analysis since only a single fluorescent molecule is required for detection [4-8].

The concept of DNA sequencing using fluorescence spectroscopy was first proposed by Keller and co-workers [9, 10] and also investigated by Eigen and co-workers [1 1 , 1 2] . Both approaches are based on initial labeling of a complementary DNA strand with fluorescent dyes, followed by sequential degradation of the labeled DNA molecule by an exonuclease and the detection of the released mononucleotides by fluorescence spectroscopy. This approach requires the labeling of the nucleobases with four different dyes (or several reactions with base specific labeling) . Until now 100% labeling has not been shown [1 3, 1 4]. The incorporation of fluorescent dye labeled nucleotides in the DNA is not easy since most polymerases do not accept the fluorescent labeled nucleotides as substrates. In addition, after incorporation of several labeled nucleotides the DNA structure changes, which could introduce errors in the sequence analysis [1 3] .

Thus, the object underlying the present invention was to provide a method wherein the disadvantages of previous single molecule sequencing protocols are avoided.

According to the present application a method is provided wherein the nucleic acid molecule to be sequenced need not be labeled previously. The label is introduced during the sequential degradation wherein labeled monomeric degradation products, e.g. nucleoside diphosphates or triphosphates, particularly deoxyribonucleoside triphosphates (dNTPs) are released during a degradation step. By using nucleobase-specific labels, e.g. fluorescence labels a determination of the nucleic acid sequence may be carried out.

A nucleic acid elongation reaction comprises providing a starting nucleic acid molecule (A) and a nucleotide monomer, oligomer or polymer (B) optionally in the presence of a catalyst (Cat) . Reaction of (A) and (B) results in the formation of an elongated product nucleic acid molecule (C) and a phosphate or pyrophosphate group (D) . The basic concept of the invention relies on the principle that the equilibrium of this reaction, which facilitates the formation of energetically more favourable product molecules (C) and (D), can be influenced by addition of a large excess of the product molecule (D) in such a way, that the product molecule (C) is completely reverted into the educt molecule (A) .

(Cat.) A + B - C + D

Thus, a particular advantage of the present invention is the coupling of the backward reaction with the introduction of a label, e.g. a fluorescent label in the formed educt molecule (B) . Thus, an additional laborious and difficult labeling reaction is not needed anymore.

A specific application of the process described above comprises the sequencing of DNA, wherein the catalyst is an enzyme with one of the following properties:

(I .) a template-dependent polymerase having pyrophosphorolytic activity such as a DNA- or RNA-polymerase or a reverse transcriptase; (II .) a template-independent enzyme having pyrophosphorolytic activity such as a terminal transferase, e.g. Terminal Deoxynucleotidyltransferase (TdT); or (III .) a template-independent enzyme having phosphorolytic activity such as a polynucleotide phosphorylase.

If the reaction is a pyrophosphorolytic reaction, the labeled reagent is selected from pyrophosphates or pyrophosphate analogs which carry a labeling group and which are capable of taking part in the degradation reaction. If the reaction is a phosphorolytic reaction, the labeled reagent is preferably a labeled phosphate or phosphate analog, which is capable of taking part in the degradation reaction. In a preferred embodiment an approach is provided that incorporates the fluorescent dye during the enzyme catalyzed sequential degradation of the DNA into dNTP ^' s (Figure 1 ) . The reaction is based on the p y r o p h o p h o r o l y s i s o f D N A c a t a l y z e d b y Te rm in a l Deoxynucleotidyltransferase (TdT) [15, 1 6]. TdT is a unique DNA polymerase that catalyzes the extension of single strand oligonucleotides in the presence of deoxyribonucleotide 5 '-triphosphates (dNTP 's) in a template independent manner [15-18]. In the reverse reaction - the pyrophosphorolysis of the DNA using pyrophosphate as substrate - the 3 '-terminal dNMP is cleaved and a dNTP is formed. The ability of TdT to catalyze the reverse reaction allows us to utilize this reaction for the sequential degradation of DNA while incorporating fluorescent dyes via the pyrophosphorolysis. When fluorescent dye labeled pyrophosphate is used instead of inorganic pyrophosphate in the pyrophosphorolysis reaction, the cleaved 3 '-terminal nucleotide forms a deoxynucleotide 5 '-triphosphate that carries the fluorescent dye at the y-phosphate. It is obvious that only one dye is incorporated into the released nucleotides. For characterization and differentiation of the released fluorescent dye labeled dNTP 's the usually undesired excited state interaction between the fluorescent dye and the nucleobase can be employed. By incorporating an "intelligent" dye, a one-lane/one-dye concept for DNA sequencing may be realised . A preferred example of an "intelligent" dye used is the fluorescence dye JF4 that has been shown to have different fluorescence lifetimes in the presence of the four nucleobases [1 9] .

Thus, a first aspect of the present invention relates to a process for sequencing of nucleic acids comprising:

(a) providing a nucleic acid molecule to be sequenced,

(b) providing a reagent labeled with a dye, e.g. a fluorescence dye wherein said dye, e.g. fluorescence dye is capable to distinguish between nucleobases when bound thereto, (c) sequential degrading of said nucleic molecule in the presence of said labeled reagent wherein monomeric nucleobase-containing degradation products labeled with said dye, e.g. fluorescence dye are formed, and (d) determining the nucleic acid sequence by sequential measurement of nucleobase-specific signals, e.g. fluorescence signals from the monomeric degradation products formed in (c) .

Preferably, the degradation of the nucleic acid molecule of the invention comprises an enzymatic reaction, more preferably an enzymatic reaction wherein a sequential release of monomeric degradation products such as nucleotides occurs and wherein a labeled reagent, e.g. a labeled phosphate or pyrophosphate, or at least a labeled part thereof is incorporated into the liberated degradation product, e.g. a nucleoside diphosphate or triphosphate.

In an especially preferred embodiment the reagent is a pyrophosphate and the sequential degradation is catalyzed by an enzyme having pyrophosphorolytic activity such as Terminal Deoxynucleotidyltransferase (TdT), which is a commercially available enzyme. The cleavage of the nucleic acid to be sequenced in the presence of pyrophosphate preferably results in the formation of y-P labeled nucleoside triphosphates.

In the above embodiment an enzyme was used as a catalyst, which has the selectivity and property of an exonuclease, i.e. the DNA is digested from the 3' or 5' end to produce a labeled nucleobase-containing monomeric degradation product. In a further preferred embodiment the present invention is based on a cleavage reaction in the presence of a catalyst having a lower specificity, i.e. the DNA strand may be cleaved unspecifically to produce a labeled oligomeric or polymeric nucleic acid fragment. To obtain under these conditions a selectively working system for nucleic acid sequencing, the cleavage reaction is performed under spatially resolved conditions, i.e. that only the last base of the nucleic acid molecule is in contact with the catalyst. Thus, the invention also relates to a unidirectional cleavage reaction comprising:

(a) providing a nucleic acid molecule to be sequenced, wherein said nucleic acid molecule is carried in an extended form by a unidirectional transport system,

(b) providing an immobilized catalyst, which is capable to cleave said nucleic acid molecule,

(c) providing a reagent labeled with a dye, e.g. a fluorescent dye wherein said fluorescent dye is capable to distinguish between nucleobases when bound thereto,

(d) sequential degrading of said nucleic acid molecule by stepwise contacting the end of the extended nucleic acid molecule with said catalyst in a such a way that only the last nucleotide of the nucleic acid molecule is in spatial contact and can react with the catalyst in the presence of said labeled reagent, which allows forming monomeric nucleobase-containing degration products labeled with said dye, and

(e) determining the nucleic acid sequence by sequential measurement of nucleobase-specific signals, e.g. fluorescence signals from the monomeric degradation products formed in (d) .

Preferably, the unidirectional transport system comprises a flow and/or an electric field gradient. The extended form of the nucleic acid molecule may be provided by terminal immobilization to a support. There are multiple ways (e.g. via biotin-streptavidin) to immobilize one end of the nucleic acid strand on a coated surface for subsequent manipulation in the spatially resolved cleavage reaction. Possible tools for micromanipulation of the immobilized nucleic acid are: a AFM (atomic force microscopy) tip, a fiber, a microcapillary, a microcantilever or a bead, which is trapped by a electric, magnetic, optic field or by a combination field. Here the nucleic acid molecule is immobilized at one end and the other end - to be degraded - is stretched freely in the medium, so that one can achieve a stepwise cleavage.

The immobilzed catalyst is contacted with the nucleic acid molecule under spatially resolved conditions and cleaves the nucleic acid molecule at a first given position. Then, the positions of the nucleic acid and/or of the catalyst have to be moved from step to step in that sense, that the degradable end of the molecule is always at the same position nearby the catalyst. In this way, spatially resolved cleavage and detection can be achieved. Natural, synthetic and semisynthetic catalysts such as enzyme conjugates, imidazoles, amines and metal- or guanidinium ions [20] are known to cleave nucleic acids by hydrolysis and thus are suitable catalysts for this embodiment of the invention.

After degradation, the labeled product can be detected and determined in a detection volume element by its nucleobase-specific signal, e.g. fluorescence characteristics. The location of the detection volume element need not to be identical with the place where the cleavage occurs, but is preferably close to it in the downstream direction.

In a preferred embodiment of the present invention the dye, e.g . the fluorescence dye which is used for labeling the reagent is a so-called "intelligent" dye, i.e. the dye is capable of distinguishing between different nucleobases, e.g. 2, 3 or 4 different nucleobases when bound thereto. Further, the dye is capable of distinguishing between an unbound state, e.g . the free reagent and a bound state, e.g. the degradation product. The distinction between different nucleobases may be accomplished by different degrees of quenching when the dye is incorporated into a nucleoside triphosphate, e.g. when it is linked to the y-P of the released nucleoside triphosphate. Suitable "intelligent" fluorescence dye molecules are known in the art. Preferably the fluorescence dyes are selected from the group consisting of fluoresceines, rhodamines, oxazines, coumarines, carbostyrenes, oxadiazoles and derivatives thereof. More preferably, the fluorescence dye molecule is fluoresceine, rhodamine-6G or JF4.

The distinction between different nucleobases, i.e. the nucleobase-specifity in fluorescence is preferably accomplished by nucleobase-specific alterations in at least one fluorescence parameter selected from fluorescence intensity, lifetime of fluorescence, anisotropy of fluorescence and/or quantum efficiency of fluorescence or any combination thereof. It should be noted in this context that the term "fluorescence" as used in the present application encompasses any process wherein by absorption of light an excited state in a molecule is generated, from which a light quantum is emitted, the so-called "fluorescence radiation" .

The nucleic acid molecule to be sequenced is e.g. a DNA molecule, e.g. a genomic DNA molecule, a cDNA molecule, a synthetic DNA molecule or any combination thereof. The method of the present invention is, however, also suitable for the sequencing of RNA molecules such as mRNA molecules. The nucleic acid molecule is preferably single-stranded. When sequencing double stranded molecules it should be observed that only one strand is degraded, e.g. by protecting the 3' terminus of the second strand against degradation.

In a preferred embodiment the present invention comprises a single molecule sequencing procedure wherein the nucleotide sequence of a single nucleic acid molecule is determined. A complete sequence determination may be accomplished by only one reaction, e.g. when an "intelligent" fluorescence dye is used which distinguishes between all four nucleobases. In other cases, wherein fluorescence dyes are used which only distinguish between less than four bases, e.g. two or three bases or even which are specific for a single base, the single molecule sequencing reaction has to be performed in several parallel batches wherein the results of these parallel batches are combined, e.g. via an electronic device, in order to obtain the complete sequence. In some cases, however, a complete sequence determination is, however, not necessary, e.g. when only a partial sequence information has to be obtained, e.g. in the determination of single nucleotide polymorphisms.

The determination of fluorescence-labeled monomeric degradation products, e.g. nucleoside triphosphates may be accomplished by any suitable measurement method, e.g. using a space- and/or time-resolved fluorescence spectroscopy method which is preferably capable of determining fluorescence signals which originate from a small number of molecules, e.g. from a single molecule in a small detection element.

For example the measurement may comprise confocal single molecule detection, e.g. by fluorescence correlation spectroscopy wherein a small, preferably confocal volume element is provided, having a volume of e.g. 0.1 x 1 0^"15 - 20 x 1 0^"12 I. The fluorescent molecules which are located in this volume element may be subjected to the excitation light, e.g. from a laser wherein the fluorescent molecules are excited and emit fluorescent light and wherein the emitted fluorescent light originating from the volume element is measured by a photodetector. For further details it is referred to EP-B-0 679 251 , wherein single molecule determination by means of confocal spectroscopy is described in detail.

In a preferred embodiment the invention relates to a single molecule sequencing method comprising the following steps:

(i) introducing the nucleic acid molecule to be sequenced into a sequencing device, (ii) capturing the nucleic acid molecule at a predetermined position in the sequencing device, (iii) sequential degrading of the nucleic molecule, wherein labeled monomeric nucleobase-containing degradation products are formed,

(iv) passing the labeled monomeric degradation products to a detection element in the sequencing device, and (v) determining the nucleic acid sequence by sequential measurement of nucleobase-specific fluorescence in the detection element.

The sequencing device preferably comprises a system of microchannels, e.g. having a diameter of from 1 -1 00 μm, particularly from 1 0-50 μm. The nucleic acid molecule is captured at a predetermined position in the sequencing device. The capturing may comprise the use of a sealed reaction compartment, which comprises the nucleic acid to be sequenced and preferably a degradation enzyme. The compartment may be sealed by a membrane which on the one hand retains the nucleic acid molecule to be sequenced, and on the other hand is permeable for released labeled products, e.g. nucleoside triphosphates. For example, the membrane is a size exclusion membrane having a cut-off value in the range of about 1 000 Da.

In this embodiment of the invention the sequencing device preferably comprises a flow reactor, wherein after the capturing of the nucleic acid the fluorescence dye labeled reagent is introduced to the sequencing device, e.g. via a continuous flow thereby starting the degradation reaction. The labeled degradation products are then passed by the flow to the detection element. Thus, the fluorescent measurement takes place downstream of the capturing position.

In a further embodiment the nucleic acid molecule to be sequenced may be introduced in a carrier-bound form into the sequencing device. Preferably the nucleic acid molecule is bound on a carrier particle having a diameter of preferably 0.5-1 O μm, particularly of from 1 -3 /m. The carrier particle may be comprised of synthetic material such as polystyrene, glass, metals or semimetals such as silicon, metal oxides such as silica or composite materials.

Carrier particles containing single nucleic acid molecules may be captured by using a capturing laser, e.g. an IR laser as described in Ashkin et al. [21 ] and Chu [22]. After capturing, the labeled reagent and optionally a degradation enzyme are passed to the captured particle thereby starting the degradation. The detection may be carried out as described above.

In the method of the present invention fluorescence labeled reagents, particluarly fluorescence labeled phosphates or pyrophosphates are used for introducing a label into the degradation products, e.g. nucleoside diphosphates or triphosphates. Thus, a further aspect of the invention relates to novel fluorescence dye labeled phosphates or pyrophosphates wherein the fluorescence molecule is coupled to the phosphate or pyrophosphate moiety via a covalent bond and preferably via a linker group. The linker group preferably comprises a linear organic molecule having a chain length from at least 2 atoms, e.g. carbon atoms. More preferably, the linker group has a chain length of at least 3 atoms selected from carbon atoms and at least one heteroatom selected from O, P, S and/or N. The fluorescence dye molecule is preferably an "intelligent" dye capable of distinguishing between different nucleobases when bound thereto, which may be selected from fluoresceines, rhodamines, oxazines, coumarines, carbostyrenes, oxadiazoles, cyanines, carbopyranines, perylenes, pyrenes, pyronines, Bodipy-dyes and derivatives thereof.

Particulary the fluorescence labeled phosphate or pyrophosphate is of the general formula (I)

wherein FI is a fluorescence dye, X is a bond or a linker group, contains C, O„ S and/or N atoms and n is an integer, preferably from 1 -30, Y is in each occurence independently R, S^", O^", OR or SR; Z is S or O and R is a monovalent ligand selected from C,-C₂o hydrocarbon groups, e.g. aliphatic and/or cyclic alkyl, alkanyl, aralkyl or aryl groups, which may optionally contain at least one heteroatom such as halo, O, S, N, P or a salt thereof.

The fluorescence dye labeled phosphates or pyrophosphates may be manufactured by a process comprising the steps of

(i) optionally coupling a linker group to a phosphate or pyrophosphate moiety, (ii) optionally purifying the product obtained in step (i), (iii) coupling a fluorescence dye to the phosphate or pyrophosphate moiety, preferably via a linker group and (iv) isolating the fluorescence dye labeled phosphate or pyrophosphate.

Still a further aspect of the invention is a fluorescence dye labeled nucleoside diphosphate or triphosphate, particularly a deoxyribonucleoside triphosphate wherein the fluorescence dye molecule is covalently coupled to the γ-P of the nucleoside triphosphate via a linker group. The linker group and the fluorescence dye are as described above.

Particularly, the fluorescence labeled diphosphate or triphosphate is of the formula (II)

wherein FI is a fluorescence dye, X is a bond or a linker group, contains C, O, S and/or N atoms and n is an integer, preferably from 1 -30, Y is in each occurence independently R, S^", O^", OR or SR; Z is S or O and R is a monovalent ligand selected from C C₂₀ hydrocarbon groups which may optionally contain at least one heteroatom such as halo, O, S, N, P and B is a nucleobase or a salt thereof.

The fluorescence dye labeled nucleoside diphosphate or triphosphate may be manufactured by a process comprising a phosphorolytic or pyrophosphorylytic degradation of a nucleic acid molecule in the presence of a fluorescence dye labeled phosphate or pyrophosphate as described above. The degradation is preferably catalyzed by an enzyme having pyrophosphorylytic activity such as TdT.

Alternatively, the fluorescence dye labeled nucleoside diphosphate or triphosphate may be manufactured by a method comprising the steps of (i) reacting the fluorescence dye to a fluorescence dye labeled monophosphate or diphosphate, (ii) reacting the fluorescence dye labeled monophosphate obtained in step (i) with a nucleoside monophosphate or diphosphate, and (iii) isolating the fluorescence dye labeled nucleoside diphosphate or triphosphate.

Further, the present invention shall be explained in more detail by the following figures and examples.

Example

1 . General remarks

The TdT catalyzed pyrophosphorolysis of DNA and the detection of the reaction products are described in the following sections. First, the synthesis of the fluorescent dye labeled pyrophosphates is described. Then th e fluorescent dye labeled pyrophosphates are investigated with regard to their feasibility as substrates for TdT in the pyrophosphorolysis of DNA. Third, we simulate the single molecule sequencing conditions in a flow reactor. Then, we describe the synthesis of the γ-P fluorescent dye labeled dNTP 's that were used as model compounds for the fluorescence analysis. Finally, we report about the differentiation of the γ-P fluorescent dye labeled dNTP 's by confocal fluorescence microscopy.

The results prove the feasibility for the approach for single molecule DNA sequencing.

2. Materials and Methods

2.1 Materials

Fluorescent dyes (Rhodamine-6G and Fluorescein) were purchased from Molecular Probes (Eugene, USA) or synthesized as described elsewhere [19]. Tris(tetra-/7-butylammonium) hydrogen pyrophosphate was purchased from Fluka. All dNDP's were purchased from Aldrich. All oligonucleotides were synthesized by Eurogentec (Belgium). [ ^_32P] ATP was purchased from New England Biolabs. All other chemicals were purchased from Fluka or Aldrich. Calf thymus TdT and T4 polynucleotide kinase were purchased from Roche. QAE-Sephadex A25 was purchased from Pharmacia. Dowex 50W-X8 (H ⁺ ) was purchased from Fluka. Ultrafiltration membranes and Microcon centrifugal filter devices were purchased from Millipore.

¹H NMR and ³¹P spectra were recorded on a Bruker AMX400 spectrometer. Mass spectra were obtained on a VG-Platform/II (Fisons) spectrometer. HPLC analysis was performed using a L-3000 Photo Diode Array Detector (Merck-Hitachi) with a L-6200 pump (Merck-Hitachi). HPLC purification was performed with UVD, L-4000 (Merck-Hitachi), pump: LC21 (Bruker).

2.2 Synthesis of the fluorescent dye labeled pyrophosphates

Λ -{Phenylmethoxycarbonyl)-2-(2-aminoethoxy)ethanol 2. 2-(2-Aminoethoxy)-ethanol 1, 1.99 ml (20 mmol), was dissolved in 8 ml of Dioxan, and 10 ml of 1.5 M aqueous NaHCO₃ was added. The solution was stirred at 0°C, and 3.1 ml (22 mmol) of benzylchloroformate was slowly added. The reaction mixture was stirred for 24 h at room temperature. The water phase was separated and washed with 20 ml of ethyl acetate. The organic phase was dried over MgSO₄, filtered, and evaporated to yield 4.6 g (95% yield) of 2 as yellow oil.¹H NMR (CDCI₃, δ) : 3.42 (t, 2H), 3.52 (m, 4H), 3.68 (q, 2H), 5.1 0 (s, 2H), 7.35 (m, 5H) . ESI-MS: m/e (%) 240.1 [M + H]⁺ (100).

MethanesuIfonyl-[/V-phenylmethoxycarbonyl-2-(2-aminoethoxy)-ethyI]-ester 3. 4.6 g (1 9.1 mmol) of 2 was dissolved in 60 ml dry CH₂CI₂, 4 ml (28.7 mmol) of triethylamine was added, and the solution was cooled to 0 ° C. 1 .8 ml (23 mmol) of methanesulfonyl chloride was slowly added to the cooled solution of 2. The solution was stirred for 1 h at 0 ° C and washed with saturated aqueous NaHCO₃. The water phase was separated and washed with ice-cold^' CH₂CI₂. The organic phase was washed with satd aqueous NaCl and dried over MgSO₄ filtered, and evaporated. The product was dissolved in CH₂CI₂, loaded onto a 40 x 5 cm silicagel column and then eluted with 20: 1 CH₂CI₂ /MeOH. Pure fractions (0.58 R_fl 20: 1 CH₂CI₂ /MeOH) were pooled and evaporated to give 3.73 g (63% yield) of 3. ¹H-NMR (CDCI₃, δ): 3.01 (s, 3H), 3.41 (t, 2H), 3.57 (t, 2H), 3.72 (t, 2H), 4.34 (t, 2H), 5.1 0 (s, 2H), 7.35 (m, 5H) . ESI-MS m/e (%) 31 8.2 [M + H] ⁺ ( 1 00) .

[/V-Phenylmethoxycarbonyl-2-(2-amino-ethoxy)-ethyI] pyrophosphate 4. Compound 3 (1 .1 5 g, 3.6 mmol) was dissolved in 1 ml dry CH₂CI₂ under argon. Tris(tetra-/?-butylammonium)-hydrogen pyrophosphate (6.5 g, 7.2 mmol) dissolved in dry acetonitrile was slowly added and the reaction mixture was stirred at room temperature for 18 h. The reaction mixture was reduced to an oil by rotary evaporation. A column 20 x 2.5 cm with QAE-Sephadex A-25 was packed in 20 mM TEABC buffer, pH 9. The product was dissolved in 7.5 ml of 20 mM TEABC buffer, pH 9, and loaded onto the column. A gradient of 1 M TEABC buffer, pH 9 was run (buffer A: 20 mM TEABC, buffer B: 1 mM TEABC, 0 to 40% B over 200 ml of mobile phase, 40 to 70% B over 1000 ml, and 70 to 1 00% B over 1 1 00 mi, flow rate: 2.5 ml/min), and the fractions were detected by UV at 260 nm The UV active fractions were checked by ³¹P NMR. Those containing pure product were pooled and evaporated to give 2.1 g (62% yield) of triethylammonium salt of 4. ¹ H-NMR (D₂O, pH 8, δ) : 3.31 (t, 2H), 3.61 (t, 2H), 3.70 (t, 2H), 5.09 (s, 2H), 7.41 (m, 5H) . ³¹ P-NMR (D₂O, pH 8, δ) : -5,62 (d, β-P, ²J(P,P) = 22 Hz), -9,53 (d, a-P, ²J(P,P) = 22 Hz) . ESI-MS m/e (%) 398.2 [M-H]^' (1 00) .

[2-(2-Aminoethoxy)-ethyI]-pyrophosphate 5. The triethylammonium salt of compound 4 (2.1 g, 2.2 mmol) was dissolved in water and 0.7 g of Pd(C) powder was added. The solution was saturated with H₂ by stirring for 3 h whereupon TLC showed complete conversion (no UV activity was detected) . The solution was filtered and lyophilized to give 1 .72 g (90% yield) triethylammonium salt of 5 as colorless oil. H-NMR (D₂O, pH 8, δ) 3.31 (t, 2H), 3.61 (t, 2H), 3.70 (t, 2H), 4,05 (q, 2H) . ³¹P-NMR (D₂0, pH 8, δ) -5.62 (d, β-P, ²J(P,P) = 22 Hz), -9.53 (d, a-P, ²J(P,P) = 22 Hz) . ESI-MS m/e (%) 263.9 [M-H]^" (100) .

2-[2-(Fluoresceϊn-5-carboxamido)-ethoxy]-ethyl pyrophosphate 6a. The following reactions were performed under an atmosphere of argon. Compound 5 was transferred in a tri-π-butylammonium salt as described by Hoard and Ott [26] and dissolved in 1 ml of dry DMF. The 5-carboxyfluorescein (20 mg, 53 /mol) was placed in a flame-dried flask sealed with a silicone septum and dissolved in dry DMF (4 ml) . A solution of TBTU (1 8.5 mg, 58 /mol) in dry DMF and an aliquot of DIEA (14 //I, 80 //mol) were then added. After 1 5 minutes, the solution of tri-t?-butylammonium salt 5 (1 06 /mol) was added and the reaction mixture was stirred at room temperature. After 4 days the solvent was removed by rotary evaporation at reduced pressure. The Fluorescein pyrophosphate was purified by anion-exchange HPLC (PEI 4000, 1 0 x 250 mm; eluant: 20 mM triethylammonium bicarbonate pH 7.6 (A), 1 M triethylammonium bicarbonate pH 7.6 (B), a gradient of 0 to 1 00% B in 50 min, flow rate: 3 ml/min, t_R - 30 min) . The collected fractions were lyophilized to yield 1 7.4 mg (52%) of 6a. ¹H-NMR (D₂O, pH 8, δ) : 3.63 (t, 2H), 3.77 (t, 2H), 3.78 (t, 2H), 4.04 (q, 2H), 6.53 (s, 2H), 6,55 (d, 2H, ³J(H,H) = 9.23 Hz), 7.1 1 (d, 2H,³J(H,H) = 9.23 Hz), 7.23 (d, 1H,³J(H,H) = 7.95 Hz), 7.97 (d, 1H,

J(H,H) = 7.95 Hz), 8.19 (s, 1H).³¹P-NMR (400 MHz, D 2₂<O, pH 8, δ): -5.5

(d, β-P, ²J(P,P) = 22 Hz), -9.5 (d, σ-P, ²J(P,P) = 22 Hz). UV/VIS: /.max (H₂O) = 492 nm. ESI-MS m/e (%) = 622.19 [M-H]^" (36.4), 310.74 [M-2H]^2' (100), 206.56 [M-3H]^3" (18.6).

2-[2-(JF4)-ethoxy] ethyl pyrophosphate 6b. To a solution of 5-carboxy-JF4 (5 mg, 10 //mol) in 4 ml dry DMF a solution of TBTU (11.3 mg, 35 //mol) in dry DMF and DIEA (6 μ\, 30 /mol) were added. After 15 minutes, a solution of tri-n-butyiammonium salt 5 (30 //mol) in DMF was added and the reaction mixture was stirred at room temperature. After 4 days the solvent was removed by rotary evaporation at reduced pressure. The JF4 pyrophosphate was purified by reverse phase HPLC (RP-18, 8 x 250 mm; eluant: H₂O (A), acetonitrile (B), a gradient of 0 to 60% B in 60 min, flow rate: 3 ml/min, t_R 33.5 min). The collected fractions were lyophilized to yield 1 mg (13%) of 6b.¹H-NMR (D₂O, pH 8, δ): 3.55 (t, 2H), 3.75 (t, 2H), 3.82 (t, 2H),4.1 (q, 2H), 7.20 (s/ 2H), 7.44 (d, 1H, ³J(H,H) = 8 Hz), 7.95 (d, 1H, ³J(H,H) = 8 Hz), 8.20 (s, 1H).³¹P-NMR (400 MHz, D₂O, pH 8, δ): -5.6 (d, β-P, J(P,P) = 22 Hz), -9.6 (d, σ-P, ²J(P,P) = 22 Hz). UV/VIS: Λmax (H₂O, 1% TFA) = 515 nm. ESI-MS m/e (%) = 677.8 [M-H]^' (100).

2-[2-((Rhodamine-6G)-5-carboxamido)-ethoxy]-ethyl pyrophosphate 6c.

5-carboxy-(Rhodamine-6G), succinimidyl ester (5 mg, 9 //mol) was dissolved in 1 ml of dry DMF and a solution of tri-/7-butylammonium salt of 5 (11.9 mg, 45 //mol) in DMF was added and the reaction mixture was stirred at room temperature. After 2 days the solvent was removed by rotary evaporation at reduced pressure. The dark red precipitate was purified by reverse phase HPLC (RP-18, 8 x 250 mm; eluant: 0.1 M TEABC, pH 7.6 (A), acetonitrile (B), a gradient of 10 to 50% B in 50 min, flow rate: 3 ml/min, t_R 22.6 min). The collected fractions were lyophilised to yield 2 mg (31%) of 6c.¹H-NMR (D₂O, pH 8, a): 1.26 (t, 6H), 2.10 (s, 6H), 3.22 (q, 4H), 3.63 (t, 2H), 3.74 (t, 2H), 3.89 (t, 2H), 4.11 (q, 2H), 6.52 (s, 2H), 6.98 (s, 2H), 7.72 (d, 1 H, ³J(H,H) = 8 Hz), 8.1 (d, 1 H, ³J(H,H) = 8 Hz), 8.20 (s, 1 H). ³¹P-NMR (400 MHz, D₂O, pH 8, δ): -5.3 (d, β-P, ²J(P,P) = 22 Hz), -9.5 (d, σ-P, ²J(P,P) = 22 Hz). ESI-MS: m/e (%) = 704.5 [M-H]^" (100).

[2-{2-Aminoethoxy)-ethyI]-phosphate 7. 996 ml ^■ (10 mmol) of 2-(2-Aminoethoxy)-ethanol 1 was placed in a flask and 980 mg (10 mmol) of phosphoric acid was then added. The reaction mixture was heated at 1 50 ° in vacuum for 1 8 h and then cooled down to room temperature. The obtained amber oil was dissolved in 1 5 ml water and the solution was treat with 5 M LiOH up to pH 10.5. The precipitated lithium salt of phosphoric acid was then removed by filtration. The filtrate was treated with 1 M HCI, loaded onto a cation exchange resin (Dowex 50W-X8, pyridinium form) and eluted with water. The solution was lyophilised to give 1 .24 g (67% yield) of 7 as a colourless oil. The compound 7 was then converted in the DMF soluble tri-/7-butylammonium salt. ¹H-NMR (D₂O, pH 8, δ): 3.1 5 (t, 2H), 3.67 (t, 2H), 3.68 (t, 2H), 4.02 (q, 2H) . ³¹ P-NMR (D₂O, pH 8, δ) : 4.42 (³J (H,P) = 7 Hz) . ESI-MS m/e (%) = 1 84.0 [M-H]^" (1 00) .

2-[2-((Rhodamine-6G)-5-carboxamido)-ethoxy] -ethyl-phosphate 8.

5-carboxy-(Rhodamine-6G), succinimidyl ester (5 mg, 9 //mol) was dissolved in 1 ml of dry DMF and a solution of tri-/7-butylammonium salt of 7 (45 / mol) in DMF was added and the reaction mixture was stirred at room temperature. After 4 days the solvent was removed by rotary evaporation at reduced pressure. The dark red precipitate was purified by reverse phase HPLC (Eurospher RP-1 8, 8 x 250 mm, 5 //m; gradient: 10-40% B in 40 min; eluant: 0.1 M TEABC buffer, pH 7,6 (A), acetonitrile (B); flow rate: 3 ml/min; t_R = 22.6 min). The collected fractions were lyophilised to yield 2 mg (37%) of 8. ¹H-NMR (D₂O, pH 8, δ) : 1 .26 (t, 6H), 2.1 0 (s, 6H), 3.22 (q, 4H), 3.63 (t, 2H), 3.74 (t, 2H), 3.89 (t, 2H), 4.1 1 (q, 2H), 6.52 (s, 2H), 6.98 (s, 2H), 7.72 (d, 1 H, ³J(H,H) = 8 Hz), 8.1 (d, 1 H, ³J(H,H) = 8 Hz), 8.20 (s, 1 H) . ³¹P-NMR (D₂O, pH 8, δ): 4,3. ESI-MS m/e (%) = 624.5 [M-H]^" ( 100) .

2-[2-(JF4)-ethoxy]-ethyl-phosphate 9. 4 mg (7.8 //mol) of 5-carboxy-(JF4) was dissolved in 1 ml of dry DMF and a solution of TBTU (9.9 mg, 31 //mol) in dry DMF and DIEA (5 / I, 31 //mol) were added. After 1 5 minutes, a solution of tri-/?-butylammonium salt of 7 (23.4 //mol) in DMF was added and the reaction mixture was stirred at room temperature. After 4 days the solvent was removed by rotary evaporation at reduced pressure. The dark red precipitate was dissolved in acetonitrile with addition of TFA and purified by reverse phase HPLC (Eurospher RP-1 8, 8 x 250 mm, 5 μm; gradient: 20-40% B in 20 min, 50% B in 40 min, 1 00% B in 55 min; eluent: H₂O + 0.1 % TFA (A), acetonitrile + 0.1 % TFA (B); flow rate: 3 ml/min; t_R = 32 min). The collected fractions were lyophilized to yield 1 .6 mg (30%) of 9. ¹H-NMR (D₂O, pH 8, δ) : 3.55 (t, 2H), 3.75 (t, 2H), 3.82 (t, 2H), 4.1 (q, 2H), 7.20 (s, 2H), 7,44 (d, 1 H, ³J(H,H) = 8 Hz), 7.95 (d, 1 H, ³J(H,H) = 8 Hz), 8.20 (s, 1 H). ³¹P-NMR (D₂O, pH 8, δ): 4.42. ESI-MS m/e (%) = 677.8 [M-H]^" ( 1 00) .

2.3 Preparation of dye labeled dNTP ^'s. General procedure

The reactions were performed under argon. The fluorescent dye labeled monophosphates were transferred in a tri-/7-butylammonium salt as described by Hoard and Ott [26]. The fluorescent dye . labeled monophosphate tri-n-butylammonium salt (0.35 //mol, 1 eq) was dissolved in 0.5 ml DMF and placed in a flame-dried flask sealed with a silicone septum. The solution of 1 .75 //mol (5eq) 1 , 1 '-carbonyldiimidazole in 0.5 ml DMF was slowly added and the reaction mixture was stirred at room temperature for 1 day. 2.6 //mol (8eq) MeOH was then added and after 45 min a solution 0.7 //mol (2 eq) of bis-(tri-t?-butylammonium) dNDP ^'s in 0.5 ml DMF was added while stirring. After 48 h the solvent was removed by rotary evaporation. The resulting precipitate was dissolved in water or in the appropriate starting buffer and purified by HPLC. The reaction products were identified by mass spectrometry.

(2-[2-(Rhodamine-6G-5-carboxamido)-ethoxy]-ethyl)-2'-desoxyadenosine- 5 '-triphosphate 10 was synthesised from 8 to yield 0.07 /mol (19%). HPLC purification was performed on Nucleosil 4000-7 PEI, 4 x 125 mm, 7 μm; gradient: 20% B in 15 min, 40% B in 45 min; eluant: H₂O (A), 2 M LiCI + 10% Acetonitrile (B); flow rate: 1.3 ml/min; t_R = 17 min. ESI-MS m/e (%) = 1017,6 [M-H]^" (100).

(2-[2-(Rhodamine-6G-5-carboxamido)-ethoxy]-ethyl)-2^'-desoxycytidine-5 '-triphosphate 11 was synthesized from 8 to yield 0.13 //mol (38%). Eurospher RP-18, 8 x 250 mm, 5//m; gradient: 10-40% B in 30 min; eluant: 0.1 M TEABC buffer pH 7.6 (A), Acetonitrile (B); flow rate: 3 ml/min; t_R = 20.6 min. MALDI-MS m/e (%) = 993.37 (100).

(2-[2-(Rhodamine-6G-5-carboxamido)-ethoxy]-ethyI)-2'-desoxγguanosine~ 5 '-triphosphate 12 was synthesized from 8 to yield 0.08 //mol (24%). HPLC purification was performed on Eurospher RP-18, 8 x 250 mm, 5//m; gradient: 10-50% B in 40 min; eluant: 0,1% TEABC buffer pH 7.6 (A), acetonitrile (B); flow rate: 3 ml/min; t_R = 19.6 min. MALDI-MS m/e (%) = 1033,8 [M-H]^" (100).

(2-[2-(Rhodamine-6G-5-carboxamido)-ethoxy]-ethyl)-2^'-desoxythymidine- 5 '-triphosphate 13 was synthesized from 8 to yield 0.09 //mol (26%). HPLC purification was performed on Eurospher RP-18, 8 x 250 mm, 5 μm; gradient: 10-40% B in 30 min; eluant: 0.1 M TEABC buffer pH 7.6 (A), acetonitrile (B); flow rate: 3 ml/min; t_R = 21.2 min. MALDI-MS m/e (%) = 1009.8 (100).

(2-[2-(JF4)-ethoxy]-ethyl)-2 '-desoxyadenosine-5 '-triphosphate 14 was synthesized from 9 to yield 0.16 /mol (8%). HPLC purification was performed on Eurospher RP-1 8, 8 x 250 mm, 5 μm; gradient: 0-60% B in 60 min; eluant: H₂O + 0.1 % TFA (A), acetonitrile + 0.1 % TFA (B); flow rate : 3 ml/min; t_R = 35.7 min. ESI-MS m/e (%) = 1069.2 [M-H]^" (80), 1 071 .2 [M-H]^" ( 1 00), 1 072.4 [M-H]^' (52) .

(2-[2-{ JF4)-ethoxy]-ethyl)-2 ^' -desoxycytidine-5 ' -triphosphate 15 was synthesized from 9 to yield 0.06 //mol (3%) . HPLC purification was performed on Eurospher RP-1 8, 8 x 250 mm, 5 μm; gradient: 0-60% B in 60 min; eluant: H₂O + 0.1 % TFA (A), acetonitrile + 0.1 % TFA (B); flow rate: 3 ml/min; t_R = 36 min. ESI-MS: m/e (%) = 1 046.8 [M-H]^' (1 4.4), 522.4 [M-2H]^2" ( 100) .

(2-[2-( JF4)-ethoxy]-ethyl)-2 ^' -desoxyguanosine-5 ^' -triphosphate 16 was synthesized from 9 to yield 0.2 / mol ( 1 0%) . HPLC purification was performed on Eurospher RP-1 8, 8 x 250 mm, 5 //m; gradient: 0-60% B in 60 min; eluant: H₂O + 0.1 % TFA (A), acetonitrile + 0.1 % TFA (B); flow rate: 3 ml/min; t_R = 33.8 min. ESI-MS m/e (%) = 1087.3 [M-H]^" (20), 543,2 [M-2H]^2' (100) .

(2-[2-(JF4)-ethoxy]-ethyl)-2 ^'-desoxythymidine-5 ^'-triphosphate 17 was synthesized from 9 to yield 0.08 //mol (1 0%) . HPLC purification was performed on Eurospher RP-1 8, 8 x 250 mm, 5 μm; gradient: 0-60% B in 60 min; eluant: H₂O + 0.1 % TFA (A), acetonitrile + 0.1 % TFA (B); flow rate : 3 ml/min; t_R = 35.9 min. ESI-MS: m/e (%) = 1062.2 [M-H]^" (26), 530.4 [M-2H]^2" (1 00).

2.4 5 '-³²P-labeling and purification of oligonucleotides

5 '-³²P-Iabeled oligonucleotides were prepared using T4 polynucleotide kinase and [ ~³²P]ATP as the phosphate source. The reaction mixture contained, in a final volume of 50 //I, 50 mM Tris buffer, pH 8.2, 1 0 mM

MgCI₂, 0.1 mM EDTA, 5 mM dithiothreitol, 0.1 mM Spermidine, 60 μC\ [ -³²P]ATP, 30U T4 polynucleotide kinase and 300 μM of oligonucleotide (24mer: d(5 ^'-GCCTCGCAGCCGTCCAACCAACTC-3 ' ) . The reaction was carried out at 37 ° C for 1 h and quenched by keeping the reaction mixture at 1 00 ° C for 3 min. The radioactive labeled oligonucleotide was separated from unincorporated [ -³²P]ATP by using a Sephadex G-25 column and sterile water as eluent. The oligonucleotide (24mer) was purified from shorter and longer nucleotides by urea-PAGE and crush and soak to recover the oligonucleotide. Bands were located by UV-shadowing, excised and extracted into 0.1 % SDS, 0.5 M ammonium acetate and 1 0 mM MgCI₂ overnight. The oligoirucleotide was desalted on a Microcon YM-3 filter using water for washing, followed by a sterile Sephadex G-25 column with water as eluent. Radioactive fractions were collected in 1 .5 ml microcentrifuge tubes and monitored using a Geiger counter. Only the radioactive fractions were kept for further experiments. The final concentration of the radioactive labeled oligonucleotide was 50-100 / M.

2.5 Pyrophosphorolysis reaction

Four separate pyrophosphorolysis reactions were performed in sterile 1 .5 ml caps. Each reaction mixture contained, in a final volume of 20 μ\ at pH 7.8, 50 mM Hepes buffer, 2 mM MgCI₂, 20 mM KCI, 5 μU 5 '-[³²P] oligonucleotide and 50 U TdT. An aliquot of the respective pyrophosphate solution was added to yield the final concentration of 1 mM. The reaction was carried out at 37 ^° C. After 0, 30, 60 and 120 min, 5 μ\ were taken and added to 5 μ\ of stop solution (95% formamide/0.05% bromophenol blue) . The mixture was kept at 100 ° C for 3 min immediately prior to loading 2 μ\ onto the gel. To examine the exonuclease activity of TdT, one reaction without pyrophosphate and one reaction without pyrophosphate and TdT were performed under the same conditions just described. 2.6 Flow reactor

The set-up of the flow reactor is shown in Figure 5 and described in detail in the figure legend.

2.7 Flow reaction

In a final volume of 500 μ\ at pH 7.8, 50 mM Hepes buffer, 2 mM MgCI₂, 20 mM KCI, 2 / M 5 '-[³²P] oligonucleotide (24mer) and 1 000 U TdT were put into the reaction chamber of the flow reactor. The supply tube of the reactor was filled with buffer to exclude air bubbles in the reactor. The pyrophosphorolysis buffer containing 50 mM Hepes buffer, 2 mM MgCI₂, 20 mM KCI and 1 mM inorganic pyrophosphate at pH 7.8 was pumped into the reactor with a flow rate of 0.040 ml/min over 5 hours. Then the flow was stopped and the reaction mixture was kept at 100 ° C for 3 min to denature the TdT and then terminate the reaction. The reaction volume was then concentrated on a sterile Microcon YM-3 filter to 10 μ\. 1 μ\ was loaded onto a 1 9% polyacrylamide gel containing 8M urea and submitted to electrophores is at 2500 V.

The reaction with Fluorescein pyrophosphate 6a was performed under the same conditions in the reactor; the only difference being that the pyrophosphorolysis buffer contained only 0.2 mM Fluorescein pyrophosphate and the reaction was stopped after 2 hours.

2.8 Fluorescence measurements

The fluorescence properties (see definitions below) of fluorescent dyes can be influenced by the environment either by unspecific effects determined by the polarity of the surrounding medium or by specific effects caused by molecular interactions with neighboring molecules or groups. For example, many fluorescent dyes, such as xanthene or coumarine dyes, can be quenched by neighboring nucleobases. A dye that shows a significant different fluorescence behavior with different nucleobases is called an "intelligent" dye. A differently strong quenching effect from a nucleobase to a fluorophore enables the identification of the nucleobase. Coumarin 1 20 is such a fluorophore, but is less preferred for single molecule detection because of its low photostability. Rhodamines are more preferred for single-molecule detection, but the fluorescence of the fluorophores rhodamine 6G (Rh6G) and tetramethylrhodamine (TMR) is only quenched by guanine and not by the other three nucleobases.

Single molecule spectroscopy enables us to detect simultaneously several fluorescence parameters, i.e. fluorescence intensity, fluorescence lifetime and anisotropy. This can be realized with the method of multiparameter fluorescence detection (MFD), explained in detail below. MFD allows one to resolve inhomogeneities and sub-states in a sample containing molecules with fluorescent properties. In addition, the sequencing method on single molecule level allows the sorting out of each labeled nucleotide and to carry out a stepwise MFD analysis.

Fluorescence parameters

Fluorescence is the deactivation process from the lowest vibrational level of an electronically excited state of a fluorophore to the ground state after excitation with light and under emission of a photon . This process can be described by different parameters: quantum yield, intensity, lifetime, anisotropy and spectral (energetic) properties of excitation and emission.

The fluorescence quantum yield Φ_F is the ratio of the number of emitted photons to the number of absorbed photons. _ _F = number o 1f emitted i p. —hot .ons < _^ ,1 number of emitted photons Eq . 1 a The estimation of fluorescence quantum yield of a sample dye relative to a reference molecule with a known fluorescence quantum yield follows the Eq . 1 b

Φ l j \ sample ^ reference jr. f r \ _p (sample) = Φ_F (reference)

^Fr r_en_e - OD_sample ■ Eq . 1 b

F stands for the fluorescence intensity respectively the area under the fluorescence spectrum and optical density, OD, measured by an absorption spectrometer.

Considering the number of excited molecules at time t after excitation, N(t), Eq. 2 relates it to the number of excited molecules at time 0, N₀, and the rate constant of radiative deactivation k_F:

N(t) = N₀ - exp-* -' _{Eq < 2}

The fluorescence lifetime r_F is defined as time t, during that the number of excited molecules has dropped to Ν₀/e.

^{Tp =}Ε Eq. 3

The lifetime is inverse proportional to the fluorescence rate constant and is the time of the molecule being in the excited state until it fluoresces and returns to the ground state. r_F is called the "radiation lifetime", because for the experimentally obtained lifetime additional processes have to be taken into account such as internal conversion, IC, a radiationless relaxation process, and intersystem crossing, ISC, a process between triplet and singlet states of a molecule.

Eq. 4 So the quantum yield can be ascribed as

^φ^ = Eq. 5

The parameter anisotropy r gives an estimate for the local mobility of the dye in the sample. The values vary between -0.2 and 0.4. The higher r, the less mobile the dye. The anisotropy is related to the fluorescence lifetime over the Perrin equation

Calculation of relative fluorescence requires a reference sample, normally the pure dye with a known quantum yield φ_F and fluorescence lifetime τ, i.e. the xanthene dye rhodamine 6G with φ_F = 0.95 and r = 4 ns.

Fluorescence quenching

The electronic energy of the excited state can be transferred from the fluorophore to a quencher. One distinguishes between dynamic and static quenching mechanisms. Static quenching is caused by an ground state and excited state equilibrium between non-fluorescent complexes, DQ, formed between the dye, D, and the quencher, Q. The complex DQ can emit photons with another wavelength compared to the free fluorophore or can return into the ground state without emission of light.

Dynamic quenching is caused by collisions between the excited fluorophore and the quencher. Thus, dynamic quenching is also called collisional quenching. The decrease in fluorescence lifetime follows the equation

^•

with [Q] as the quencher concentration. This equations is valid for the case, where the quencher concentration is considerably higher than the fluorophore concentration and [Q] can be assumed to be constant in time. The 2^nd order reaction changes into a reaction of pseudo-1 ^st order.

The fluorescence quantum yield and therefore the fluorescence lifetime are proportional to the intensity F of a fluorescence band. The ratio of fluorescence lifetime and intensity in the presence (r and F) and absence (r₀ and F₀) of quencher follows Eq. 8.

The depiction of (F₀/F)-1 and (r₀/r)-1 versus quencher concentration (Stern-Volmer plot) gives the quenching constant K_sv. The higher the value for K_sv, the stronger the fluorescence quenching by the quencher.

According to Eq. 8, there are two ways to measure the nucleobase-specific quenching :

1 . Use of a conventional steady-state fluorescence spectrometer to determine fluorescence intensities.

2. Use of time-resolved fluorescence spectroscopy to measure fluorescence lifetimes.

The conditions of the single molecule measurements with MFD were as follows. Pulsed excitation with an argon ion laser at a wavelength λ =■ 51 4.5 nm with an irridiance of l₀ = 86 kW cm^'2 with an water immersion objective (60x) . The pinhole had a diameter of 1 00 //m. The dichroic mirror was 510DCLP and fluorescence filters HQ575/70 and HQ730/140 were used. The samples were measured in phosphate buffer (1 0 mM Na₂HP0₄/NaH₂PO₄, 1 80 mM NaCl, pH 7.46) at a temperature between 20 and 25 ° C.

The concentration of each sample was between 50 and 100 pM. Individual molecules passing through the open volume element are readily detected by their brief fluorescence bursts and selected from the background signal. The fluorescence signal is divided into its parallel and perpendicular components with respect to the linear polarized excitation beam by a polarizing beam splitter cube, which is then subsequently divided into "green" (wavelength λ < 595 nm) and "red" (λ > 595 nm) fluorescence components by dichroic mirrors resulting in four signal paths. Each burst is analyzed to determine a single set of dye-specific fluorescence parameters.

3. Results and discussion

3.1 Synthesis of the fluorescent dye labeled pyrophosphates

The synthesis of the fluorescent labeled pyrophosphates is summarized in Figure 2. The route is flexible with respect to the usage of several fluorescent dyes, employing simple well established coupling techniques based on the formation of amide bonds and allowing for adjusting the nature of the linker. An aminoethoxyethyl linker 1 was used to connect the fluorescent dyes with the pyrophosphate group. This mobile, water-soluble linker warrants better acceptance of the dye labeled pyrophosphate by the enzyme for sterical reasons. Similarly important, good stacking of the base with the fluorophore at the stage of the labeled triphosphate is necessary to allow differentiation of the four nucleobases based on different interaction with the fluorophore. The functional groups of 1 , amino group and hydroxyl group, can easily be substituted by activated fluorescent dyes and pyrophosphates. This is a versatile approach to synthesise different dye labeled pyrophosphates because one can easily vary the length of the linker and the dye.

The pyrophosphate ester 5 is the key reagent in the synthesis of fluorescent labeled pyrophosphates, and is prepared from the aminoethoxyethanol 1 in five steps. In the first step the amino group has been protected using the benzyloxycarbonyl protecting group. This prevents possible cyclisation, when the carbonyl group is activated by mesylate during the next step. Furthermore, the benzyloxycarbonyl group can be detected by UV light which simplifies the purification of the pyrophosphate. The carbonyl group of 2 was activated with methanesulfonyl chloride and 3 was pyrophosphorylated using twofold excess of tris(tetra-π-butylammonium) hydrogen pyrophosphate [23, 24] . At this stage a complete removal of inorganic pyrophosphate from the pyrophosphate ester 4 is performed, since the inorganic pyrophosphate would later disturb the enzymatic reaction with the dye labeled pyrophosphate. Purification of the pyrophosphate ester 4 is advantageous because of the UV activity of the amino protecting group. ³¹ P-NMR of 4 proves that the inorganic pyrophosphate was completely removed (Figure 3).

After the purification the amino group of 4 was hydrolytically deprotected using Pd/C catalyst. The reaction was complete within 3 hours (TLC) . The resulting compound 5 was used as the building block for all subsequent couplings with fluorescent dyes.

For the labeling of pyrophosphate 5 with the fluorescent dye Fluorescein we used its succinimidyl ester. For the dyes Rhodamine-6G and JF4, their carboxyl group was activated by the method described by Gillessen and co-workers [25] . This reaction requires a water free solvent that prevents hydrolysis of the intermediates. DMF was most suited due to the high solubility of fluorescent dyes in this polar, aprotic solvent. The pyrophosphate 5, however, is not soluble in DMF, whereas its tri-/7-butylammonium salt that is obtained by exchange of triethylammonium by tri-/7-butylammonium [26] . The resulting tri-/7-butylammonium salt of pyrophosphate 5 was reacted with the activated fluorescent dyes to give fluorescent dye labeled pyrophosphates 6a-c.

3.2 Pyrophosphorolysis reaction

The fluorescent dye labeled pyrophosphates were used in the pyrophosphorolysis reaction of DNA. The purity of the fluorescent dye labeled pyrophosphates was ensured by HPLC and ³¹ P NMR to exclude the contamination with inorganic pyrophosphate and/or with dNTP ^'s. We used DNA oligonucleotides of different length, 24mer (23mer and 20mer are not shown here) as substrates for pyrophosphorolysis.

The evaluation of the pyrophosphorolysis reaction can in principle be done by detection of the dye labeled dNTP ^'s or the modified DNA. We focused on a technique that is robust and simple. Since the amount of dNTP 's formed during TdT catalyzed pyrophosphorolysis reaction did not allow detection with a conventional fluorescence detector we used radioactive labeling of the DNA oligonucleotides at their 5 ^'-end with ³²P and analysis of the DNA length using sequencing gel. Fluorescent dye labeling of the DNA oligonucleotides is not suitable due to the use of an excess of fluorescent labeled pyrophosphates in the reaction. Radioactive labeling of the oligonucleotides was successfully used previously [1 6] to show the pyrophosphorolytic activity of TdT with inorganic pyrophosphates using homooligonucleotides.

The radioactively labeled DNA was purified before use in the pyrophosphorolysis reaction by denaturing gel electrophoresis. The DNA was visualised by UV absorption, excised and retrieved from the gel using "crush and soak" techniques as described in materials and methods. The purified DNA was desalted twice on Sephadex G-25 columns to completely remove the SDS contained in the elution buffer.

The purified, ³²P-5 '- labeled DNA was incubated for 1 h with TdT and the various pyrophosphate derivatives in the presence of 1 x TdT buffer containing 2 mM Mg²⁺, 20 mM KCI and 50 mM HEPES buffer (pH 7.8) . The concentration of inorganic pyrophosphate, Fluorescein-pyrophosphate 6a and Rhodamine-6G-pyrophosphate 6b was 1 mM. The concentration of JF4-pyrophosphate 6c was 0.2 mM due to its low solubility. The reaction products were analyzed in a 61 cm gel electrophoresis apparatus with 1 9% polyacrylamide gel containing 8 M urea. The results of the pyrophosphorolysis reactions using the 24mer d (5 '-GCCTCGCAGCCGTCCAACCAACTC-3 ' ) are shown in Figure 4. The results of the .reactions with Fluorescein-pyrophosphate 6a (lane 2), Rhodamine-6G-pyrophosphate 6c (lane 3) and JF4-pyrophosphate 6b (lane 4) are shown respectively. In addition, we have performed a reaction with inorganic pyrophosphate (Figure 4, lane 1 ) and in the absence of pyrophosphate (lane 5) . As expected, there are no new DNA bands in lane 5. In the presence of pyrophosphates a ladder of oligonucleotides that are shorter and longer than the initial 24mer is observed. The results show that TdT accepts, all three fluorescent dye labeled pyrophosphates as substrates. The distribution of oligonucleotides of different lengths around the 24mer puts the equilibrium nature of the pyrophosphorolysis reaction catalyzed by TdT into evidence. Due to the exclusive presence of fluorescent dye labeled pyrophosphates, occurrence of truncated DNA proves indirectly the formation of fluorescent dye labeled dNTP 's. The elongated oligonucleotides require the presence of dNTP ^'s which can only have been formed by pyrophosphorolysis and reincorporation at the 3 '-end of oligonucleotides. The equilibrium between the degradation versus incorporation reaction depends on the nature of the dye. While for JF4 the ratio is approximately 1 : 1 , despite the smaller pyrophosphate concentration, the polymerisation products dominate for Fluorescein pyrophosphate 6a and even more for the Rhodamine-6G substituted pyrophosphate 6c. With inorganic pyrophosphate both the pyrophosphorolysis activity of TdT and the overall rates are maximal.

3.3 Pyrophosphorolysis reaction with a flow reactor

In order to proceed towards single molecule DNA sequencing with TdT, we performed the reaction in a flow reactor with continuous addition of pyrophosphates. This should lead to further degradation of the initial oligonucleotides and less polymerisation products. The flow reactor consists of a cylindrical reaction chamber and two covers with holes for the supply and drain of the pyrophosphate containing 1 x TdT buffer (see Figure 5). On the top and the bottom of the reaction chamber two membranes with cut-off volumes of 1 000 Da are placed that are permeable only for low molecular weight compounds like pyrophosphates and dNTP 's but not for DNA and TdT. The reaction chamber was fasten with screws, after addition TdT, DNA and 1 x TdT buffer without inorganic pyrophosphate. A solution of inorganic pyrophosphate in 1 x TdT buffer is then pumped through the reactor using a HPLC-pump at flow-rates up to 40 //l/min.

The pyrophosphorolysis reaction with the flow reactor was carried out using inorganic pyrophosphate, Fluorescein-pyrophosphate 6a and no pyrophosphate as control. For comparison, the same reactions were performed in regular 1 .5 ml reaction caps. The reactions were analyzed using urea-PAGE as described previously with radioactively detection. The results are presented in Figure 6. For the control reactions no length change of oligonucleotides is observed, neither in the cap (Figure 6, lane 4) nor in the flow reactor (Figure 6, lane 3). In the reaction carried out with inorganic pyrophosphate in the reaction cap an almost equal distribution between polymerisation and degradation can be observed (lane 1 ), whereas in the flow reactor the truncated oligonucleotides prevail as expected (lane 2). This effect is also found using Fluoresceine labeled pyrophosphate and inorganic pyrophosphate. Thus we can extrapolate that the pyrophosphorolysis reaction can lead to complete degradation of the DNA even with fluorescent dye labeled pyrophosphates.

Conditions for the complete degradation of DNA without competitive polymerisation are not reached with this experimental setup. This is due to too slow removal of the formed dNTP 's. Under our conditions 1 2 ml of the

1 mM inorganic pyrophosphate solution are pumped through the reaction chamber within 5 hours leaving on average every dNTP 1 2.5 min in the reaction chamber during which it can be reincorporated into DNA.

3.4 Synthesis of y-P-fluorescent dye labeled dNTP 's

To investigate the fluorescence properties of the fluorescent dye labeled dNTP 's that are formed in the pyrophosphorolysis reaction we synthesised these compounds from fluorescent dye labeled monophosphates.

First, phosphoric acid was melted with 2-(2-aminoethoxy)-ethanol 1 in vacuo to give phosphate ester 7 (Figure 7). Unreacted phosphoric acid was precipitated as lithium salt and separated by filtration. The lithium salt of phosphate ester 7 was transformed to the tri-/7-butylammonium salt and reacted with the activated fluorescent dye in the same way as for the pyrophosphates (Figure 2). The purification and the handling of the dye labeled phosphates 8 and 9 is significantly easier than the purification of the dye labeled pyrophosphates, since they cannot hydrolyse.

The fluorescent dye labeled phosphates 8 and 9 were then converted to the y-P-fluorescent-dye-labeled dNTP ^'s as shown in Figure 7 using the phosphoester-forming reaction as described by Cramer [27] . The fluorescent dye labeled phosphates 8 and 9 were activated using the

1 , 1 '-carbonyldiimidazoie. This is a mild reaction and does not require protection and purification steps [26, 28] . Both activating and coupling (Figure 7) are performed in DMF and under argon because of the high tendency of P-N-bounds to hydrolyse. Similar to what was described previously, the solubility of fluorescent-dye-labeled monophosphates 8 and

9 and dNDP 's in DMF was increased by exchange of tri-π-ethylammonium salt by tri-n-butylammonium salt. The activated fluorescent dye labeled monophosphates 8 and 9 have been coupled with all four dNDP 's to yield the corresponding y-P-labeled dNTP ^'s of Rhodamin-6G and JF4. 3.5 Fluorescence measurements

Steady-state measurements

Table 1 summarizes the Stern-Volmer quenching constants from different fluorophores (i.e. rhodamine and oxazine dyes) with the four nucleotides dAMP, dCMP, dGMP and TMP. The measurements are taken in acetate buffer (50 mM sodium acetate, pH 5.5) .

All dyes show the known behavior of quenching by the nucleobase guanine. The mononucleotides were used because of their better water solubility. The structure formulas of dyes in Table 1 can be found in [29].

Table 1 : Stern-Volmer quenching constant K_sv (from steady-state measurements) . "lncrease" :F₀/F < 1 , " + "- the curve is tilted towards the y-axis, "-"- the curve is tilted towards the x-axis, "L"- from 0 to 50 mM quencher concentration the curve has linear dependence; if the dependency is not linear, the constant was measured in the region at very low concentrations of the quencher.

dAMP quenches the fluorescence from cyanorhodamine, from the listed oxazines and from rhodamine 1 1 0 and 1 23. In general, the quenching effect by dAMP is smaller than by dGMP, which corresponds to the more positive oxidation potential of adenosine. The non-alkylated dyes at the xanthene skeleton, i.e. rhodamine 1 10, rhodamine 1 23 and JF4 E, show the best capacity of an intelligent fluorophore.

3.6 Single molecule experiments

The fluorescence parameters shown here in the example are: fluorescence intensity in the green detection channels, S_G, the intensity ratio from the green and the red detection channels, S_G/S_R, fluorescence lifetime, T, and anisotropy, r. The frequency of parameter pairs, equal to the number (#) of fluorescence bursts, are counted in 2D histograms with the corresponding 1 D histograms given as projections (figures 8 to 1 0) .

In this series the fluorescence lifetime r is plotted versus the intensity ratio S_G/S_R in figures A. In figures B, the rotational correlation time of p = 0.3 ns and 0.8 ns are given by the overlaid, black curve in the 2D histograms by using the anisotropy value at time zero for rhodamine dyes of r₀ = 0.37.

Here we present two examples of -P-fluorescent dye labeled dNTP's depicted in comparison to the pure fluorescent dye: dGTP and dATP labeled with rhodamine 6G (Rh6G) . The powder was solved in buffer and measured immediately after dilution.

The sample Rh6G-dGTP contains two species: According to the results shown in figure 8, one belongs to the pure dye. The other compound is the dye with linked nucleoside-triphosphate. The fluorescence lifetime of the dye is shortened with a concomitant decrease of the fluorescence intensity. The third parameter, the anisotropy, shows a better resolution of the two species. Therefore, the differentiation of the two species is possible, avoiding further purification of the sample.

The sample containing Rh6G-dATP seems to be mostly homogeneous and can be distinguished from the sample with Rh6G-dGTP. The fluorescence lifetime of the species is increased compared to the lifetime of the pure dye. The line with p = 0.8 ns does not cross the center of the histogram peak leading to the assumption of a mixture of pure dye and complete molecule.

Through MFD one is able to recognize inhomogeneities in the sample due to a mixture of different substances such as pure dye and labeled dNTP. These results are more obvious than the multiexponential fit of a fluorescence decay yielding several lifetimes and amplitudes.

Figure legends:

Figure 1 : The concept of DNA sequencing by TdT catalyzed pyrophosphorolysis. A single DNA strand is incubated with TdT and pyrophosphate labeled with "intelligent" dye. The DNA is degraded sequentially by the pyrophosphorolysis reaction and deoxynucleotide 5 '-triphosphates labeled at y-P with the intelligent dye are formed during the degradation. The formed γ-P fluorescent dye labeled dNTP ^'s can be detected and identified by confocal fluorescence microscopy by virtue of their different fluorescence properties.

Figure 2: Synthesis of the fluorescent dye labeled pyrophosphates 6a-c. i: N-(Phenylmethoxy-carbonyl)-2-(2-aminoethoxy)ethanol, dioxan, 0 ° C, RT, 24 h; ii: methanesulfonyl chloride, Et₃N, CH₂CI₂, 0 ° C, 30 min; iii: (Bu₄N)₃HP₂0₇, acetonitrile, RT, 1 8 h; iv: QAE-Sephadex A-25, 1 M TEABC buffer (pH 9), 20 mM TEABC buffer (pH 9); v: H₂/Pd(C), H₂0, RT, 3 h; vi: R-COOH, TBTU/DIEA or R - succinimidyl ester, DMF; HPLC purification. Figure 3: Purification of pyrophosphate 4 as determinate by NMR spectroscopy (400 MHz), 300 K in D₂0 (pH 8). (a) Decoupled ³¹P spectrum of 4 before the chromatographic purification; the singlet at δ = -6.5 ppm correspond to inorganic pyrophosphate; (b) - Decoupled ³¹P spectrum of 4 after purification on QAE-Sephadex A-25 with a gradient of TEABC buffer (pH 9) .

Figure 4: Pyrophosphorolysis of 24mer DNA with inorganic pyrophosphate and fluorescent dye labeled pyrophosphates. TdT was incubated with the 24mer at 37 ° C and pH 7.8 in a 20 μ\ volume containing 50 mM Hepes buffer, 2 mM MgCI₂, 20 mM KCI, 5 μ 5 ^'-[³²P]-labeled oligonucleotide and different pyrophosphates. A: Each lane represents a 1 μ\ aliquot removed after 1 h and subjected to electrophoresis at 2500 V on a 1 9% polyacrylamide gel containing 8 M urea. Lane 1 shows the oligonucleotide products after the pyrophosphorolysis with 1 mM inorganic PPi; lane 2: with 1 mM Fluorescein-PPi (6a); lane 3 with 0.2 mM JF4-PP (6b); lane 4: with 1 mM Rhodamine-6G-PPi (6c); lane 5 is a control lane in the absence of pyrophosphate. B: slices trough the corresponding lanes.

Figure 5: Detailed view of the flow reactor. The reactor consists of a cylindrical reaction chamber (500 //I) made up of two steel blocks (1 and 2) with openings on the top and bottom for supply and release. The reactor chamber was sealed by O-seals and fasten with screws. The supply and release tubes were connected to the reactor using screw-in joints. The TdT and DNA are confined in the reaction chamber by two membranes with cut-off volumes of 1000 Da. The membranes are placed on the support between the opening and the chamber on both sides of supply and release. The cylindrical chamber and membrane supports were coated with Teflon to avoid contact of TdT and DNA with metal. The flow was controlled by HPLC pump. Figure 6: Demonstration of the pyrophosphorolysis in the flow reactor (lanes 3 and 6) in comparison with the pyrophosphorolysis reaction carried out without flow in 1 .5 ml cap (lanes 1 , 4, 5). Lane 3 and 4 are the results from the control reactions carried out in the absence of pyrophosphate in the flow reactor and in the 1 .5 ml cap, respectively. Lane 1 represents the results in the presence of 1 mM PPi after 5h, and lane 5 in the presence of 0.2 mM Fluorescein-PPi (6a) after 2 h, both in 1 .5 ml caps. Lane 2 stems from the reaction with 1 mM PPi after 5h and lane 6 with 0.2 mM Fluorescein-PPi after 2 hours both in flow reactor. From the two lanes stemming from the reaction in the flow reactor it is obvious that a full degradation of DNA with dye labeled pyrophosphate can be achieved.

Figure 7. Chemical synthesis of fluorescent dye labeled dNTP ^'s using the 1 , 1 '-carbonyldiimidazole method, i: The reaction was kept in vacuum at 1 50 ° C for 1 8 h; ii: 5 M LiOH solution was added up to pH 1 0.5; iii: Dowex 50W-X8 (pyridinium); tri-n-butylamine, DMF, 67%; iv: Rhodamine-6G succinimidyl ester, DMF, RT, 4 days; HPLC purification; Dowex 50W-X8 (pyridinium form), Bu₃N, DMF; v: 5-carboxy-JF4, TBTU/DIEA, RT, 4 days; HPLC purification; Dowex 50W-X8 (pyridinium form), Bu₃N, DMF; vi: 1 , 1 '- carbonyldiimidazole, DMF, RT, 1 day; vii: MeOH, RT, 45 min; viii: dNDP tri-/7-butylammonium salt, DMF; ix: HPLC purification.

Figure 8. Two-dimensional histograms of the fluorescence parameters of the fluorophore rhodamine 6G (Rh6G) at the single molecule level. (A) Fluorescence lifetime τ vs. intensity ratio F_D/F_A. (B) Fluorescence lifetime r vs. anisotropy r_G. (C) Fluorescence liftetime t vs. fluorescence intensity S_G.

Figure 9. Two-dimensional histograms of the fluorescence parameters of Rh6G-dGTP at the single molecule level. (A) Fluorescence lifetime r vs. intensity ratio F_D/F_A. (B) Fluorescence lifetime r vs. anisotropy r_G. (C) Fluorescence lifetime r vs. fluorescence intensity S_G.

Figure 10. Two-dimensional histograms of the fluorescence parameters of Rh6G-dATP at the single molecule level.

(A) Fluorescence lifetime r vs. intensity ratio F_D/F_A. (B) Fluorescence lifetime r vs. anisotropy r_G. (C) Fluorescence lifetime τ vs. fluorescence intensity S_G.

References

1 . Hillenkamp F., Karas M., Beavis R.C., Chait B.T. (1 991 ) Matrix-assisted laser desorption/ionization mass spektrometry of biopolymers. Anal. Chem., 63, 1 1 93A-1 203A.

2. Kδster H, van den Boom, D., Braun, A., Jacob, A., Jurinke, C, Little, D.P., Tang, K. ( 1 997) DNA analysis by mass spektrometry: applications in DNA sequencing and DNA diagnostics. Nucleosides & Nucleotides., 1 6, 563-571 . 3. Khrapko K.R, Lysov, Y.P., Khorlyn, A. A., Shik, V.V., Florentiev, V.L., Mirzabekov, A.D. (1 989) An oligonucleotide hybridization approach to DNA sequencing. FEBS Lett., 256, 1 1 8-1 22.

4. Schaffer J, Volkmer, A., Eggeling, C, Subramaniam, V., Striker, G., Seidel, C.A.M. ( 1 999) Identification of single molecules in aqueous solution by time-resolved fluorescence spectroscopy. J. Phys.

Chem., A 1 03, 331 -336.

5. Fries JR, Brand, L., Eggeling, C, Kollner, M ., Seidel, C.A.M. (1 998) Quantitative identification of different single-molecules by selective fluorescence spectroscopy. J. Phys. Chem., A 1 02, 6601 -661 3. 6. Brand L., Eggeling, C, Zander, C, Drexhage, K.H., Seidel, C.A.M. ( 1 997) Single-molecule identification of coumarin-1 20 by time-resolved fluorescence detection - comparison of one- and two-photon exitation in solution. J. Phys. Chem., A 101 , 431 3-4321 . 7. Brand L. ( 1 998) PhD thesis, Georg-August-Universitat, Gδttingen. 8. Zander C, Sauer M., Drexhage K.H., Ko D.S., Schulz A., Wolfrum J., Brand L., Eggeling C, Seidel, C.A.M. ( 1 996) Detection and characterisation of single molecules in aqueous solution. Appl. Phys., B 63, 51 7-523. 9. Ambrose W.P, Goodwin P.M, Jett JH, Jonson ME, Martin JC, Marrone BL, Schecker CW, Keller RA. (1 993) Application of single molecule detection to DNA sequencing and sizing. Ber. Bunsenges. Phys. Chem., 97, 1535-1542.

10. Keller RA, Ambrose WP, Goodwin PM, Jett JH, Martin JC, Wu M. (1996) Single-Molecule Fluorescence Analysis in Solution. Appl. Spectroscopy, 50, 12A-32A.

11. Stephan J, Dόrre, K., Brakmann, S., Winkler, T., Wetzel, T., Lapczyna, M., Stuke, M., Angerer, B., Ankenbauer, W., Fόldes-Papp, Z., Rigler, R., Eigen, M. (2001) Towards a general procedure for sequencing single DNA molecules. J. Biotec, 86, 255-267.

12. Dδrre K, Brakmann S, Brinkmeier M, Han K-T, Ribeseel K, Schwille P, Stephan J, Wetzel T, Lapczyna M, Stuke M, Bader R, Hinz M, Seliger H, Holm J, Eigen M, Rigler R. (1997) Techniques for single molecul detektion. Bioimaging, 5, 139-152. 13. Brakmann S, Lόbermann, S. (2001) Hochgradige Markierung von DNA: Herstellung und Charakterisierung des Substrats fur eine Einzelmolekϋl-Sequenzierung. Angew. Chem., 113, 1473-1476.

14. Fδldes-Papp Z, Angerer, B., Ankenbauer, W., Rigler, R. (2001) Fluorescent high-density labeling of DNA: error-free substitution for a normal nucleotide. J. Biotechnol., 86, 237-253.

15. Srivastava A, Modak MJ. (1980) Biochemistry of terminal deoxytransferase: identifikation, charakterisation, requirements, and active-site involvement in the catalysis of associated pyrophosphate exchange and pyrophosphorolytic activity. Biochemistry, 19, 3275-3280.

16. Anderson RS, Bollum, F.J., Beattle, K.L. (1999) Pyrophosphorolytic dismutation of oligodeoxynucleotides by terminal deoxynucleotidyltransferase. Nucleic Acids Research, 27, 3190-3196. 17. Pandey V, Modak MJ. (1988) Biochemistry of Deoxynucleotidyltransferase. J. Biol. Chem., 263, 3744-3751. 1 8. Fersht A. ( 1 995) Enzyme Structure and Mechanism. W.H. Freeman and Company, New York.

1 9. Fries JR. (1 998) Charakterisierung einzelner Molekϋle in Losung mit Rhodamin-Farbstoffen. In: PhDthesis, Universitat-Gesamthochschule Siegen.

20. Kurz K, Gδbel MW. (1 996) Hydrolytical Cleavage of TAR-RNA, the trans-Activation Responsive Region of HIV-1 , by a Bis(guanidinium) Catalyst Attached to Arginine, Helv. Chimica Acta, 79, 1 967-1 979.

21 . Ashkin A, Dziedzic JM, Yamane T. (1 987) Optical trapping and manipulation of^'single cells using infrared laser beams. Nature, 330,

769-71 .

22. Chu S. ( 1 991 ) Laser manipulation of atoms and particles. Science, 253, 861 -866.

23. Davisson VJ, Woodside AB, Neal TR, Stremler KE, Muehlbacher M, Poulter CD. (1 986) Phosphorylation of isoprenoid alcohols. J. Org.

Chem., 51 , 4768-4779.

24. Steiger A, Pyun H-J, Coates M. ( 1 992) Synthesis and characterization of aza analogue inhibitors of squalene and geranylgeranyl diphosphate syntheses. J. Org. Chem., 57, 3444-3449.

25. Knorr R, Trzeciak A, Bannwarth W, Gillessen D. (1 989) Tetrahedron Lett., 30, 1 927-1 930.

26. Hoard DE, Ott DG. ( 1 965) Conversion of mono- and oligodeoxyribonucleotides to 5'-triphosphates. J. Am. Chem. Soc, 87, 1 785-8.

27. Cramer F, Schaller H, Staab H. (1 961 ) Darstellung von Imidazoliden der Phosphorsaure. Ber., 94, 1 61 2-1 633.

28. Simon ES, Grabowski S, Whitesides GM. (1 990) Convenient syntheses of cytidine 5'-triphosphate, guanosine 5 ^'-triphosphate, and uridine 5'- triphosphate and their use in the preparation of

UDP-glucose, UDP- glucuronic acid, and GDP-mannose. J. Org. Chem., 55, 1 834-41 . 9. Brakman U. (1 997) Lambdachrome Laser Dyes. Lambda Physik. Gόttingen.

Claims

Claims

1 . A process for sequencing of nucleic acids comprising: (a) providing a nucleic acid molecule to be sequenced,

(b) providing a reagent labeled with a fluorescence dye wherein said fluorescence dye is capable to distinguish between nucleobases when bound thereto,

(c) sequential degrading of said nucleic molecule in the presence of s a i d l a b el ed rea g ent w h ere i n mo n o meric nucleobase-containing degradation products labeled with said fluorescence dye are formed, and

(d) determining the nucleic acid sequence by sequential measurement of nucleobase-specific fluorescence from the monomeric degradation products formed in (c) .

2. The process according to claim 1 , wherein the degradation comprises an enzymatic reaction.

3. The process according to claim 1 , wherein the degradation comprises a spatially resolved cleavage reaction.

4. The process according to claim 1 and 3, wherein the nucleic acid to be sequenced is carried in an extended form by a unidirectional flow system comprising a stepwise movement of the end of the extended nucleic acid molecule to the position of an immobilized catalyst in a such a way that only the last nucleotide of the nucleic acid is in spatial contact and can react with the catalyst.

5. The process according to claim 3 or 4, wherein the catalyst is a natural, synthetic or semisynthetic catalytic system.

6. The process according to claim 5, wherein the catalytic system is selected from an enzyme conjugate, an imidazole, an amine and a metal- or a guanidinium ion or a combination thereof capable of cleaving nucleic acids by a reagent labeled with the fluorescent dye.

7. The process according to claim 1 or 2, wherein the labeled reagent is selected from phosphates, pyrophosphates or analogues thereof.

8. A process according to claim 7, wherein the sequential degradation is catalyzed by an enzyme having pyrophosphorolytic or phosphorolytic activity.

9. The process according to claim 8, wherein the enzyme is selected from template-dependent polymerases, such as DNA- or RNA-polymerases, or reverse transcriptases, or template-independent enzymes such as terminal transferases e.g. terminal deoxynucleotidyltransferase (TdT) or polynucleotide phosphorylases.

10. The process according to any one of claims 1 to 9, wherein in step (c) γ-P-\abe\ed nucleoside triphosphates or β-P-labeled nucleoside diphosphates are formed.

11. The process according to any one of claims 1 to 10, wherein the fluorescence dye distinguishes between nucleobases by different degrees of quenching when incorporated into the released monomeric degradation products.

12. The process according to any one of claims 1 to 11, wherein the fluorescence dye distinguishes between 2, 3 or 4 nucleobases.

3. The process according to any one of claims 1 to 1 2, wherein the fluorescence dye molecule is selected from the group comprising fluoresceines, rhodamines, oxazines, coumarines, carbostyrenes, oxadiazoles, cyanines, carbopyranines, perylenes, pyrenes, pyronines, Bodipy-dyes and derivatives thereof.

4. The process according to claim 1 3, wherein the fluorescence dye molecule is fluoresceine, rhodamine-6G, JF4, rhodamine 1 1 0, oxazin 1 1 8 or tetramethyl-rhodamine.

5. The process according to any one of claims 1 to 1 4, wherein the nucleic acid molecule is a DNA molecule.

6. The process of any one of claims 1 to 1 5 which is a single molecule sequencing procedure.

7. The process according to any one of claims 1 to 1 6, wherein the fluorescence is measured fluorescence microscopy and/or spectroscopy.

8. The process according to any one of claims 1 to 1 6, wherein the n ucleobase-specificity in fluorescence is caused by nucleobase-specific alterations in at least one fluorescence parameter selected from fluorescence intensity, lifetime of fluorescence, anisotropy of fluorescence and/or quantum efficiency of fluorescence in one or several spectrally specified wavelength ranges or any combination thereof.

9. The process according to any one of claims 1 to 1 8, comprising the following steps:

(i) introducing the nucleic acid molecule to be sequenced into a sequencing device, (ii) capturing the nucleic acid molecule at a predetermined position in the sequencing device, (iii) sequential degrading of the nucleic molecule, wherein labeled monomeric nucleobase-containing degradation products are formed,

(iv) passing the labeled monomeric degradation products to a detection element in the sequencing device, and (v) determining the nucleic acid sequence by sequential measurement of nucleobase-specific fluorescence in the detection element.

20. The process according to claim 1 9, wherein the sequencing device comprises a flow reactor.

21 . The process according to claim 20 comprising a spatially resolved detection of the degradation product.

22. The process according to claim 1 9 or 20, wherein after capturing in step (ii) the fluorescence dye labeled reagents and the means for degrading the nucleic acid molecule are continually introduced into the sequencing device.

23. The process according to any one of claims 21 to 22, wherein the nucleic acid molecule is captured in a reaction compartment.

24. The process according to claim 23, wherein the reaction compartment is sealed with a membrane which is impermeable for the nucleic acid molecule and permeable for released labeled monomeric degradation products.

25. A fluorescence dye labeled phosphate or pyrophosphate, wherein a fluorescence dye molecule is coupled to the phosphate or pyrophosphate via a suitable functional group.

26. The compound according to any one of claim 25, wherein the fluorescence dye molecule is selected from fluoresceines, rhodamines, oxazines, coumarines, carbostyrenes, oxadiazoles, cyanines, carbopyranines, perylenes, pyrenes, pyronines, Bodipy-dyes and derivatives thereof.

27. The compound according to any one of claims 25 to 26 having a structure of the general formula (I)

wherein FI is a fluorescence dye, X is a bond or a linker group, which contains C, O, S and/or N-atoms and n is an integer, Y is in each occurence independently R, S^", O^", OR or SR; Z is S or O and R is a monovalent ligand selected from ^-C^ hydrocarbon groups which may optionally contain at least one heteroatom such as halo,

O, S, N, P or a salt thereof.

28. A process for manufacturing a fluorescence dye labeled phosphate or pyrophosphate according to any one of claims 25 to 27 comprising the steps of: (i) optionally coupling a linker group to a phosphate or pyrophosphate moiety, (ii) optionally purifying the product obtained in step (i), (iii) coupling a fluorescence dye to the phosphate or pyrophosphate moiety and

^■ (iv) isolating the fluorescence dye labeled phosphate or pyrophosphate.

29. A fluorescence dye labeled nucleoside triphosphate or diphosphate wherein the fluorescence dye molecule is covalently coupled to the γ-P of . the nurcleoside triphosphate or β-P of the nucleoside diphosphate.

30. The compound of claim 29 having the structure of the general formula (II)

wherein FI is a fluorescence dye, X is a bond or a linker group, which contains C, O, S and/or N-atoms and n is an integer, Y is in each occurence independently R, S^", O^", OR or SR; Z is S or O and R is a monovalent ligand selected from C,-^ hydrocarbon groups which may optionally contain at least one heteroatom such as halo,

O, S, N, P and B is a nucleobase or a salt thereof.

31 . A process for the manufacture of a fluorescence dye labeled nucleoside triphosphate or diphosphate according to claim 29 or 30, comprising a pyrophosphorolytic or phosphorolytic degradation of a nucleic acid molecule in the presence of a fluorescence dye labeled labeled pyrophosphate or phosphate according to any one of claims 25-27.

32. The process according to claim 31 , wherein the pyrophosphorolytic or phosphorolytic degradation is catalyzed by an enzyme having pyrophosphorolytic or phosphorolytic activity.

33. A process for the manufacture of a fluorescent dye labeled nucleoside triphosphates or diphosphates according to claim 29 or, comprising the steps of:

(i) reacting the fluorescence dye to a fluorescence dye labeled monophosphate or diphosphate, (ii) reacting the fluorescence dye labeled monophosphate obtained in step (i) with a nucleoside diphosphate or monophosphate, and

(iii) isolating the fluorescence dye labeled nucleoside triphosphate or diphosphate.