Incorporation of reporter molecule-labeled nucleotides by DNA polymerases. I. Chemical synthesis of various reporter group-labeled 2′-deoxyribonucleoside-5′-triphosphates

Abstract

Fluorescent-labeled DNA is generated through enzymatic incorporation of fluorophore-linked 2′-deoxyribonucleoside-5′-triphosphates (dNTPs) by DNA polymerases. We describe the synthesis of a variety of dye-labeled dNTPs. Amino-linker-modified 5′-triphosphates of all four naturally occurring nucleobases were used as precursors. Commercially available dyes were coupled to the amino function of the side chain. In addition, we attached novel fluorophore derivatives. The labeled products were obtained in at least 96% purity after HPLC purification. Enzymatic incorporation into DNA and subsequent extension of the modified DNA chain were studied. Vent^R exo^– DNA polymerase and a defined template–primer system were used to analyze each dye-labeled dNTP derivative. Our data suggest that the incorporation efficiency depends on the selected dye, the nucleobase or a combination of both.

Introduction

The life sciences owe their rapid growth in large part to the development of fluorescence methodologies. Important methods such as DNA microarray gene expression analysis (1–3), fluorescence in situ hybridization (4,5), DNA sequencing (6,7) and single nucleotide polymorphism analysis (8) are based on the detection of fluorescent-labeled nucleic acid probes. These probes are generated commonly by nick translation, primer extension, reverse transcription or PCR using fluorophore-modified nucleotides (9–12). The labeling procedure is straightforward, but does have some drawbacks. DNA polymerases tolerate very well biotin and small fluorophores such as aminomethyl-coumarins (13), but they have problems in incorporating bulky dye molecules attached to nucleotides into DNA or RNA (10,14). As a result, consistent and accurate generation of labeled DNA samples is difficult to achieve. This variation in degree of labeling influences the evaluation of gene expression profiling based on hybridization of labeled cDNA on microarrays. To circumvent these particular problems, it is possible to incorporate a less bulky, amine-modified nucleotide enzymatically and then to label the amine-modified DNA/RNA afterwards using an amine-reactive reagent (15,16). Although 5-(3-aminoallyl)-dUTP can be incorporated to extremely high and consistent levels, complete labeling at each amino modification cannot be guaranteed.

Due to these various limitations, a complete replacement of more than one of the four bases by the fluorescent analogs could not be achieved to date. New DNA sequencing approaches based on the detection of fluorescent dye-tagged nucleotides cleaved from a single fluorescence-labeled DNA molecule require the labeling at every position in the target sequence (17,18). In order to accomplish the synthesis of fully labeled DNA, a number of difficulties remain to be solved. Steric hindrance caused by the bulkiness of the fluorescent dye groups is one major problem, but the hydrophobic nature of the dye also has to be considered. Further, the linking moieties between dye and nucleobase with regard to chemical structure may affect the incorporation. In addition, potential dye–dye or dye–enzyme interactions have to be addressed. In this first paper, we describe the synthesis and purification of various dye-labeled 2′-deoxyribonucleoside-5′-triphosphates (dNTPs) covering all four naturally occurring bases, with a focus on fluorophores with absorption in the green (480–510 nm) or the red range (620–650 nm), respectively. These dyes are of particular interest for several applications. Furthermore, we examined the substrate specificity in primer extension reactions with Vent^R exo^– DNA polymerase.

MATERIALS AND METHODS

Materials

The N-hydroxysuccinimide (NHS) esters of the fluorophores were purchased from the following suppliers: Rhodamine Green™-X, Oregon Green™ 488-X, Alexa Fluor® 488 and Alexa Fluor® 633 from Molecular Probes (Eugene, OR); Dy630, Dy635, Evoblue 90 and Evoblue 30 from Dyomics GmbH (Jena, Germany); Atto655 from ATTO-TEC (Siegen, Germany); biotin (EZ-Link) from Pierce (Rockford, USA); and cyanine 5 from Amersham Pharmacia Biotech. 5-(3-Phthalimido-propynyl)-2′-deoxycytidine 3a, 7-(3-phthalimido- propynyl)-7-deaza-2′-deoxy-adenosine 3b and 7-(3- phthalimido-propynyl)-7-deaza-2′-deoxy-guanosine 3c were purchased from Chembiotech (Münster, Germany). The phenoxazine derivatives 18, 19 and 20 were synthesized by Dyomics GmbH (Jena, Germany) and were defined as Gnothis blue 1 (GB1), Gnothis blue 2 (GB2) and Gnothis blue 3 (GB3), respectively. All other chemicals were of the highest purity commercially available. Vent^R exo^– DNA polymerase was purchased from New England Biolabs.

General methods

Ion exchange HPLC purifications were performed on Mono Q HR 5/5 (analytical) and HighLoad 16/10 Q Sepharose (preparative) columns (Amersham Pharmacia Biotech). The following eluents were used: buffer A, 10% acetonitrile in H₂O; and buffer B, 1.5 M LiCl in H₂O. TLC analyses on Kieselgel 60 were done with the solvent system n-propanol/aqueous concentrated NH₃/H₂O 11:7:2. UV/VIS spectra were obtained with a Heλios alpha spectrophotometer (Thermo Spectronic, Cambridge, UK). ³¹P-NMR were recorded on an Advance DPX-250 NMR spectrometer (Bruker, Rheinstetten, Germany) with an operation frequency of 101.07 MHz for ³¹P nuclei. δ-Values were relative to external 85% H₃PO₄. The spectra were measured in 0.1 M Tris–HCl pH 8.0, 0.1 M EDTA/D₂O. Determination of the concentration of dye-labeled triphosphate solutions was performed by UV/VIS measurement using the extinction coefficient of the corresponding dye.

Synthesis of 2′-deoxyribonucleoside-5′-triphosphate derivatives

5-Aminoallyl-2′-deoxyuridine-5′-triphosphate (AA-dUTP) (1). AA-dUTP was prepared as described (19) with identical data.

5-(3-Amino-1-propynyl)–2′-deoxycytidine-5′-triphosphate (AP- dCTP) (2a). 5-(3-Phthalimido-1-propynyl)-2′-deoxycytidine 3a (100 mg, 0.24 mmol) was suspended in dry trimethylphosphate (0.8 ml). The suspension was cooled to 0°C. POCl₃ (30 µl, 0.25 mmol) was added and the reaction mixture was kept at 0°C for 3 h, resulting in a clear solution. Next, a pre-mixed solution of tributylamine (0.5 ml) and tributylammonium pyrophosphate [2.0 ml, 0.5 M solution in dimethylformamide (DMF)] was added. The reaction was stopped by addition of triethylammonium bicarbonate buffer (20 ml, 0.1 M) after 20 min. This mixture was stirred for another 45 min before concentrated aqueous ammonia (20 ml) was added and kept at room temperature for 3 h. The solution was concentrated and applied to a HighLoad 16/10 Sepharose column with the following HPLC program: flow rate 5 ml/min; 0–5 min: 100% buffer A; 5–30 min increase of buffer B to 35%. Sodium phosphate buffer (1 ml, 20 mM, pH 7.2) was added to the eluate and the solution was concentrated and poured into acetone/ethanol (60 ml, 2:1). After centrifugation, the precipitate was washed with ethanol, dried in a vacuum and finally redissolved in sodium phosphate buffer (20 mM, pH 7.2). The resulting triphosphate 2a (71 µmol, 30.5%) was kept in solution at –20°C until further use. UV (0.1 M phosphate buffer, pH 7.0, 25°C): λ_max = 295 nm, ε = 12 000. ³¹P-NMR (D₂O): –5.2 (d, J = 19.2 Hz, γP), –9.9 (d, J = 19.2 Hz, αP), –20.4 (t, J = 19.3 Hz, βP).

7-(3-Amino-1-propynyl)-7-deaza-2′-deoxyadenosine (AP-7-deaza-dATP) (2b). 7-(3-Phthalimido-1-propynyl)-7-deaza-2′-deoxyadenosine (3b) (100 mg, 0.23 mmol) was used as starting material. The synthesis and purification of the 5′-triphosphate 2b were performed according to the procedure of the corresponding dCTP derivative 2a. 2b (45 µmol, 20.1%) was kept in solution at –20°C until further use. UV (0.1 M phosphate buffer, pH 7.0, 25°C): λ_max = 280 nm. ³¹P-NMR (D₂O): –5.0 (d, J = 19.2 Hz, γP), –10.1 (d, J = 19.2 Hz, αP), –20.4 (t, J = 19.3 Hz, βP).

7-(3-Amino-1-propynyl)-7-deaza-2′-deoxyguanosine (AP-7-deaza-dGTP) (2c). 7-(3-Phthalimido-1-propynyl)-7-deaza-2′-deoxyguanosine (3c) (100 mg, 0.22 mmol) was used as starting material. The synthesis of the triphosphate 2c was performed as described for AP-dCTP 2a. After HPLC purification, the appropriate fraction was concentrated and the triphosphate was precipitated in acetone/ethanol 2:1 as described above. The precipitate was redissolved in 80% acetic acid (10 ml) and stirred at room temperature for 1 h. The acetic acid was removed in a vacuum and the compound re-purified using the standard HPLC program followed by the precipitation process. 2b (62 µmol, 28.1%) was kept in solution at –20°C until further use. UV (water, 25°C) λ_max = 274 and 292 nm. ³¹P-NMR (D₂O): –5.5 (d, J = 19.2 Hz, γP), –9.8 (d, J = 19.2 Hz, αP), –20.7 (t, J = 19.3 Hz, βP).

Rhodamine Green™-5(6)-carboxyamido-ε-aminocaproyl-[5-(3-aminoallyl)-2′-deoxyuridine-5′-triphosphate] (7d). AA-dUTP 1 (3 µmol) was dissolved in sodium borate buffer (1 ml, 0.1 M, pH 8.5). Rhodamine Green-X-N-hydroxysuccinimide ester hydrochloride (3.2 µmol) was dissolved in amine-free DMF (100 µl). Both solutions were combined and the reaction mixture was kept at room temperature for 3 h. The solution was filtrated and applied to a HighLoad 16/10 Sepharose column. The labeled triphosphate was purified using the following HPLC program: flow rate 4 ml/min; 0–5 min 100% buffer A; 5–30 min increase of buffer B from 0 to 28%. The product-containing fractions were combined and sodium phosphate buffer (1 ml, 20 mM, pH 7.2) was added. The buffered solution was concentrated and poured into acetone/ethanol 3:1. After centrifugation, the precipitate was washed, dried in a vacuum and finally redissolved in sodium phosphate buffer (20 mM, pH 7.2). The labeled triphosphate 7d (1.5 µmol, 50%) was kept in solution at –20°C. UV/VIS (0.1 M phosphate buffer, pH 7.0, 25°C): λ_max = 504 nm, ε = 74 000. ³¹P- NMR (D₂O): –5.5 (d, J = 19.2 Hz, γP), –10.3 (d, J = 19.2 Hz, αP), –20.7 (t, J = 19.3 Hz, βP).

The corresponding dCTP, dATP and dGTP derivatives 7a–c as well as all other labeled triphosphates were synthesized and purified according to the procedure described above. Some dye-labeled 5′-triphosphates had to be repurified several times in order to remove the starting material completely (1 and 2a–c). 6a, λ_max = 559 nm, ε = 100 000; 6b, λ_max = 659 nm, ε = 100 000; 6c, λ_max = 658 nm, ε = 100 000; 6d, λ_max = 658 nm, ε = 100 000; 7a, λ_max = 505 nm, ε = 74 000; 7b, λ_max = 503 nm, ε = 74 000; 8, λ_max = 498 nm, ε = 80 000; 9, λ_max = 493 nm, ε = 71 000; 11a, λ_max = 648 nm, ε = 250 000; 11b, λ_max = 650 nm, ε = 250 000; 11c, λ_max = 650 nm, ε = 250 000; 11d, λ_max = 649 nm, ε = 250 000. For the following dye-labeled 5′-triphosphates, the HPLC purification buffers had to be modified slightly. Due to the hydrophobic character of the dyes, the amount of acetonitrile in buffer A was increased from 10 to 20%. 12b, λ_max = 630 nm, ε = 120 000; 12d, λ_max = 630 nm, ε = 120 000; 13, λ_max = 669 nm, ε = 120 000; 14, λ_max = 657 nm, ε = 100 000; 15a, λ_max = 668 nm, ε = 100 000; 15b, λ_max = 669 nm, ε = 100 000; 15c, λ_max = 669 nm, ε = 100 000.

Biotin-EZ-Link™-[-5-(3-aminoallyl)-2′-deoxyuridine 5′-triphosphate] (10d). AA-dUTP 1 (5.5 µmol) was dissolved in sodium borate buffer (1 ml, 0.1 M, pH 8.5). Biotin-EZ-Link™ NHS ester (2 mg, 5.8 µmol) was dissolved in amine-free DMF (100 µl). Both solutions were combined and the reaction mixture was kept at room temperature for 16 h. The solution was filtrated and applied to a HighLoad 16/10 Sepharose column. The labeled triphosphate was purified using the following HPLC program: flow rate 4 ml/min; 0–1 min 100% buffer A; 1–30 min increase of buffer B from 0 to 22%. The product-containing fractions were combined and sodium phosphate buffer (1 ml, 20 mM, pH 7.2) was added. The buffered solution was concentrated and poured into acetone: ethanol 3:1. After centrifugation, the precipitate was washed, dried in a vacuum and redissolved in sodium phosphate buffer (20 mM, pH 7.2). The biotinylated compound 10d (4.0 µmol, 71%) was kept in solution at –20°C. Analytical data were identical to those described (19). Compounds 10a–c were synthesized and purified as described above.

{3-[N,N-Di-(3-sulfopropyl)-amino]-7-(N-carboxyethyl-N-methyl- amino)-phenoxazine}-[5-(3-aminoallyl)-2′-deoxyuridine-5′-triphosphate] (21d). The phenoxazine chromophores 18–20 were coupled to amino-linked triphosphates (1 and 2a–c). GB3 20 (2.5 mg, 4.4 µmol) was dissolved in dimethylsulfoxide (DMSO) (200 µl). Diisopropyl-ethylamine (2 µl, 11 µmol) and N,N,N′,N′-tetramethyl-(succinimido)-uronium-tetrafluoroborate (TSTU, 2 mg, 6.6 µmol) were added. The reaction mixture was stirred at room temperature for 10 min. AA-dUTP 1 (5 µmol) was dissolved in H₂O (0.5 ml) and added in two portions within 5 min. After stirring for another 10 min, the solution was filtrated. The filtrate was applied to a HighLoad 16/10 Q Sepharose column. The following HPLC program was used: flow rate, 5 ml/min, 0–5 min 100% buffer A; 5–35 min, increase of buffer B from 0 to 25%. The product-containing fractions (retention time: 26.5 min) were combined and sodium phosphate buffer (1 ml, 20 mM, pH 7.2) was added. The buffered solution was concentrated to a total volume of 3 ml and poured into a mixture of acetone/ethanol (50 ml, 3:1). After centrifugation, the precipitate was washed, dried in a vacuum and finally redissolved in sodium phosphate buffer (1 ml, 20 mM, pH 7.2). The labeled triphosphate 21d (1.1 µmol, 25%) was kept in solution at –20°C. UV (0.1 M phosphate buffer, pH 7.0, 25°C): λ_max = 655 nm, ε = 91 000. Compounds 21a–c, 22 and 23a–d were synthesized and purified according to the procedure described above. 21a, λ_max = 656 nm, ε = 91 000; 21b, λ_max = 655 nm, ε = 91 000; 21c, λ_max = 654 nm, ε = 91 000; 22, λ_max = 655 nm, ε = 80 000; 23a, λ_max = 655 nm, ε = 95 000; 23b, λ_max = 654 nm, ε = 95 000; 23c, λ_max = 655 nm, ε = 95 000; 23d, λ_max = 654 nm, ε = 95 000.

Modified nucleotide incorporation analysis

Enzymatic incorporation of various modified dNTPs by Vent^R exo^– DNA polymerase was investigated in template-directed primer extension assays. Homopolymer templates of dA₁₈, dT₁₈ and dC₁₈, respectively, with a specific 22 nt primer-binding site were used to analyze each group of modified dNTPs (for details see the accompanying manuscript by Tasara et al.).

RESULTS AND DISCUSSION

Synthesis of dye-labeled 2′-deoxyribonucleoside-5′-triphosphates

In order to test a variety of differently labeled dNTPs, the appropriate 5′-triphosphates were synthesized as precursors, which could easily be conjugated to fluorescent dyes (Scheme 1). This attachment should be independent of the selected heterocyclic base. The majority of commercially available dyes for labeling procedures are NHS ester derivatives. These activated esters react readily with aliphatic amino functions. The C5 position of the pyrimidine ring is an ideal site for the modification as it is located in the major groove of double-stranded DNA and does not disturb the A:T base pairing. Therefore, functional groups have been attached to C5 via an amino-linker (10). The most attractive position for modifications of the purine nucleobases appears to be at position 7 since it is also located within the funnel-shaped major groove (20). Modification of the naturally occurring purine nucleoside at N7 would give rise to a positively charged derivative with a highly labile glycosidic bond. Since it has been reported that 7-deazapurine nucleoside-5′-triphosphates are accepted by DNA polymerases (21), these analogs were used to attach the dye via an amino-linker at the C7 carbon. The synthesis as well as the enzymatic incorporation of 7-deazapurine-dNTPs containing a functional group at C7 have been described previously (22–24).

Scheme 1. Modified 2′-deoxyribonucleoside-5′-triphosphates.

AA-dUTP 1 was synthesized according to Langer et al. (19). However, this chemistry is of no practical value for the derivatization of the other nucleobases cytidine, adenine and guanine. Therefore, an alternative approach had to be chosen. The synthesis of C5-modified 2′-deoxycytidine as well as of 7-alkyl-substituted 7-deazapurine nucleosides was accomplished using a Pd-catalyzed C–C cross-coupling reaction (24–27). The phthalimido-protected aminopropynyl derivatives of 2′-deoxycytidine (2a), 7-deaza-2′-deoxyguanosine (2c) and 7-deaza-2′-deoxyadenosine (2b) were used as starting materials for the synthesis of the required amino-linker 5′-triphosphates. The aminopropynyl function is protected in order to avoid side reactions. The two-step 5′-triphosphorylation was performed according to Yoshikawa et al. (28) and Hoard and Ott (29). The amino-protected triphosphates 3a–c were not isolated but immediately treated with ammonia in order to remove the phthalimido group. Initially, the phthalimido-protected compounds were purified but the formation of a side product was observed during the purification procedure. These side products showed a similar HPLC mobility but slightly different UV spectra in the 220–250 nm range. The phthalimido group in this compound could not be removed anymore although treated at 60°C in concentrated aqueous ammonia. In the case of the guanosine derivative 3c, the formation of a second product in addition to 2c was observed. This second side-product showed a similar mobility in HPLC but a faster mobility on TLC. It is known that phosphorylating agents can attack the O6 position of guanine nucleosides (28,30). As these by-products are sensitive to acid treatment, the product mixture was treated with 80% acetic acid for 30 min. Under these conditions, the faster moving compound (on TLC) was transformed into 2c within 30 min. After HPLC purification, pouring into a mixture of acetone and ethanol desalted the concentrated solutions of the amino-linked 5′-triphosphates. The precipitate was filtered, dried in a vacuum and finally redissolved in 20 mM sodium phosphate buffer pH 7.2. The triphosphates were kept in solution at –20°C until further use. The amino-linked derivatives were obtained in at least 96% purity. The enzymatic incorporation of the amino-linked triphosphates 1 and 2a–c into DNA proved the presence of a 5′-triphosphate moiety. All four triphosphates were substrates for Vent^R exo^– DNA polymerase (see below and the accompanying manuscript by Tasara et al.) and could fully replace its natural counterpart in PCRs.

The dye-labeled 5′-triphosphates were all synthesized using a standardized procedure. The coupling reaction was performed in 0.1 M sodium borate buffer pH 8.5, and Evoblue 30-NHS ester 5 was coupled to AA-dUTP 1 (Scheme 2). The reaction mixture was kept at room temperature for 2 h. Finally, the product was purified by HPLC using a HighLoad 16/10 Sepharose column. Elution was performed with a linear gradient of 0–0.3 M LiCl in H₂O. For each dye–base combination, the elution gradient had to be adapted. Due to the limited solubility of certain dyes and their corresponding triphosphates, acetonitrile concentrations between 2 and 15% were used. To evaluate the incorporation efficiency of each dye-labeled triphosphate, it was important that the final product did not contain any unlabeled triphosphate. AA-dUTP 1 and Evoblue 30-dUTP 6 showed different mobilities in HPLC (Fig. 1). The purity of the labeled triphosphates was between 95 and 98%. Tables 1 and 2 summarize the synthesized labeled triphosphates.

Scheme 2. Coupling of Evoblue 30-NHS ester 5 to AA-dUTP 1.

Figure 1.

HPLC profile of modified dUTP derivatives. Purification on HighLoad 16/10 Q Sepharose as described in Materials and Methods. (A) AA-dUTP 1; (B) Evoblue 30-dUTP 6.

Table 1. 2′-Deoxyribonucleoside-5′-triphosphates labeled with ‘green’ dyes (absorption range: 480–520 nm) or a biotin group^a.

Base	Linkerb	Rhodamine Green	Oregon Green	Alexa 488	Biotin
Uracil	AA 1	7d	8	9	10d
Cytosine	AP 2a	7a			10a
Adenine	AP 2b	7b			10b
Guanine	AP 2c	7c			10c

^aThe numbers refer to the compound number as described in Materials and Methods.

^bAA, aminoallyl; AP, 3-amino-1-propynyl.

Table 2. 2′-Deoxyribonucleoside-5′-triphosphates labeled with ‘red’ dyes (absorption range: 630–660 nm)^a.

Base	Cy5b	Dy630	Dy635	Evo30b	Evo90b	Atto655	Alexa 633
Uracil	11d	12d	13	6d	14	15d	16
Cytosine	11a			6a		15a
Adenine	11b	12b		6b		15b
Guanine	11c			6c

^aThe numbers refer to the compound number as described in Materials and Methods.

^bCy5, cyanine 5; Evo30, Evoblue 30; Evo90, Evoblue 90.

As described below, the Evoblue 30-labeled triphosphates 6a–c showed good substrate properties. This is of particular interest as Evoblue 30 is a fluorophore that absorbs in the red range. Based on the phenoxazine chromophore system, new derivatives were designed and synthesized (Scheme 3) with the aim of improving the hydrophilic character. This is achieved commonly, for example, by introducing sulfonic acid functions into the dye molecule (i.e. GB3 20). On the other hand, a highly negative charged dye might interfere with the active site of DNA polymerases. Hence, we were seeking alternatives to the negatively charged sulfo groups retaining hydrophilicity.

Scheme 3. New phenoxazine chromophores: Gnothis blue 1 (GB1) 18, Gnothis blue 2 (GB2) 19 and Gnothis blue 3 (GB3) 20.

Hydroxyl groups (GB1 18) or a short peptide linker (GB2 19) might increase the water solubility of the dye. As each of these three derivatives had a different overall charge (18, +1; 19, 0; 20, –1), it could be determined if the charge of the fluorophore has an impact on the DNA polymerase activity. Since such compounds were not commercially available in the form of the activated NHS ester, the attachment chemistry had to be adjusted. Carboxylic functions could be transferred into their corresponding hydroxysuccinimido esters using TSTU (31,32). Furthermore, the following reaction with an amino group could be performed in situ even in the presence of water. GB3 20 was activated with TSTU and diisopropyl-ethylamine in DMSO. AA-dUTP 1 was added to the reaction mixture after 10 min. The obtained labeled triphosphate 21 was purified using the standard procedure as described before. The phenoxazine-labeled triphosphates are listed in Table 3.

Table 3. 2′-Deoxyribonucleoside-5′-triphosphates labeled with new phenoxazine derivatives (absorption range: 645–655 nm)^a.

Base	GB1b 18	GB2b 19	GB3b 20
Uracil	22	23d	21d
Cytosine		23a	21a
Adenine		23b	21b
Guanine		23c	21c

^aThe numbers refer to the compound number as described in Materials and Methods.

^bGB1, Gnothis blue 1, compound 18; GB2, Gnothis blue 2, compound 19; GB3, Gnothis blue 3, compound 20.

Incorporation of modified nucleoside 5′-triphosphates by DNA polymerase

The template-directed analytical method enabled the analysis of the incorporation of each modified dNTP by a specific DNA polymerase at up to 18 adjacent positions (for details, see accompanying manuscript by Tasara et al.). The results are summarized in Table 4. The various modified 5′-triphosphates tested could all be incorporated into nascent DNA by Vent^R exo^– DNA polymerase, confirming that they were suitable substrates for the DNA polymerase. However, there were clear distinctive incorporation threshold differences between the various modified 5′-triphosphates (summarized in Table 4). The incorporation efficiency of the modified dNTPs varied not only between different reporter groups but also within the bases to which the reporter groups were conjugated. Generally, modified pyrimidine nucleotides were incorporated more efficiently compared with purine nucleotides bearing the same modifying group. Furthermore, there were also some poorly tolerated reporter groups attached to the nucleobases. Once incorporated into the nascent DNA chain, they obviously affect the following extension by the DNA polymerase (e.g. Alexa 633-dUTP and Cy5-7-deaza-dATP). In summary, a wide range of these new reporter group-conjugated dNTPs were accepted as substrates and incorporated into DNA by Vent^R exo^– DNA polymerase. However, some chemical properties of both the reporter groups and their linker moieties need further optimization for efficient and high-density introduction of modified dNTPs into DNA using DNA polymerases. These issues are addressed further in the accompanying manuscript by Tasara et al.

Table 4. Incorporation performance of various modified 2′-deoxyribonucleoside-5′-triphosphates.

Reporter groupb	Linker length	dUTP	dATP	dGTP	dCTP
None	None	18a	>18c	18	18
Biotin	12	>18c	>18c	>18c	>18c
RhG	12	14	8d	17–18	11
Cy5	8	10–11	4	n.a.e	5
Evo30	8	14	13d	8	4
GB3	11	13	13d	13d	5
Atto655	8	4d	16	n.a.	8

^aNumber of documented individual incorporation steps out of a possible 18 in the homopolymeric DNA templates, as described in the DNA polymerase assay in Tasara et al. (accompanying manuscript).

^bRhG, Rhodamine Green; Cy5, cyanine 5; Evo30, Evoblue 30; GB3, Gnothis blue 3.

^cTemplate slippage or terminal transferase activity of Vent^R exo^– DNA polymerase observed.

^dLonger incorporation observed, but bands could not be resolved.

^en.a., not available.