Organic & Biomolecular Chemistry View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

PAPER

Cite this: Org. Biomol. Chem., 2013, 11, 6357

View Journal | View Issue

Exploring the role of the 5-substituent for the intrinsic fluorescence of 5-aryl and 5-heteroaryl uracil nucleotides: a systematic study†‡ Thomas Pesnot,a Lauren M. Tedaldi,b Pablo G. Jambrina,c Edina Rostac and Gerd K. Wagner*b,c Derivatives of UMP (uridine monophosphate) with a fluorogenic substituent in position 5 represent a small but unique class of fluorophores, which has found important applications in chemical biology and biomolecular chemistry. In this study, we have synthesised a series of derivatives of the uracil nucleotides UMP, UDP and UTP with different aromatic and heteroaromatic substituents in position 5, in order to systematically investigate the influence of the 5-substituent on fluorescence emission. We have determined relevant photophysical parameters for all derivatives in this series, including quantum yields for the best fluorophores. The strongest fluorescence emission was observed with a 5-formylthien-2-yl substituent in position 5 of the uracil base, while the corresponding 3-formylthien-2-yl-substituted regioisomer was significantly less fluorescent. The 5-(5-formylthien-2-yl) uracil fluorophore was studied further in solvents of different polarity and proticity. In conjunction with results from a conformational analysis based on NMR data and computational experiments, these findings provide insights into the steric and

Received 8th March 2013, Accepted 5th August 2013

electronic factors that govern fluorescence emission in this class of fluorophores. In particular, they highlight the interplay between fluorescence emission and conformation in this series. Finally, we carried out ligand-binding experiments with the 5-(5-formylthien-2-yl) uracil fluorophore and a UDP-sugardependent glycosyltransferase, demonstrating its utility for biological applications. The results from

DOI: 10.1039/c3ob40485d

our photophysical and biological studies suggest, for the first time, a structural explanation for the

www.rsc.org/obc

fluorescence quenching effect that is observed upon binding of these fluorophores to a target protein.

Introduction Fluorescent nucleotides are highly useful as chemical tools for numerous biological applications.1 While the natural nucleotide building blocks of nucleic acids are largely nonfluorescent, the incorporation of fluorescent derivatives has been used to study, for example, DNA and RNA structure2 and to detect oxidative lesions.3 Fluorescent nucleotides and nucleotide conjugates have also been used successfully as probes in enzymological studies, e.g. in biochemical assays for NAD-dependent enzymes4 and in ligand-displacement assays for inhibitor screening against glycosyltransferases.5 Three a

School of Pharmacy, University of East Anglia, Norwich, UK Institute of Pharmaceutical Science, School of Biomedical Sciences, King’s College London, UK c Department of Chemistry, School of Biomedical Sciences, King’s College London, Franklin-Wilkins Building, 150 Stamford Street, London, SE1 9NH, UK. E-mail: [email protected]; Fax: +44 (0)20 7848 4045; Tel: +44 (0)20 7848 4747 † Dedicated to my father on the occasion of his 65th birthday (GKW). ‡ Electronic supplementary information (ESI) available: Additional Fig. S1–S3; NMR spectra for 2a, 2b and 2l–q. See DOI: 10.1039/c3ob40485d b

This journal is © The Royal Society of Chemistry 2013

main strategies have been pursued to generate such fluorescent nucleotides: (i) the replacement of the nucleobase with a fluorescent isostere,6 (ii) the conjugation of a given nucleotide with an extrinsic fluorophore,5b,c,7 and (iii) the installation of a fluorogenic substituent at the nucleobase.5a,8,9 In principle, the latter approach is attractive for several reasons. Fluorescence can be generated with a relatively small molecular change, minimising the potential for undesired interference with the biological target. In addition, the installation of the fluorogenic substituent can often be achieved synthetically in a relatively straightforward manner. This approach is therefore finding increasing application for the development of fluorescent derivatives of both purine and pyrimidine nucleotides and nucleotide conjugates.5a,8,9 An important example for intrinsically fluorescent nucleotides are derivatives of uracil with a suitable, fluorogenic substituent in position 5 (Fig. 1). Fluorescent nucleotides and nucleotide conjugates incorporating such a 5-substituted uracil fragment have been used successfully as probes for nucleic acid structure9c–e and as fluorophores for glycosyltransferase ligand-displacement assays.5a To date, 5-substituents

Org. Biomol. Chem., 2013, 11, 6357–6371 | 6357

View Article Online

Paper

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Fig. 1

Organic & Biomolecular Chemistry

Fluorescent 5-substituted uracil nucleotides and nucleotide conjugates.

that have been explored for the development of intrinsically fluorescent uracil nucleosides and nucleotides have been limited mainly to small heterocycles5a,9c–e or phenol and anisol derivatives connected directly or via an alkenyl or alkynyl linker to the uracil ring.9a,b While 5-substituted uracil nucleotides, in which a known fluorophore is connected to the uracil base via an electronically non-conjugating linker, usually retain the photophysical features of the parent fluorophore,1a,10 appending an additional aromatic or heteroaromatic moiety to the uracil base typically generates a new fluorophore with unique and somewhat unpredictable photophysical characteristics.1a While it appears clear that the fluorescence of 5-substituted uracil nucleotides derives from the extended electron delocalisation across the uracil base and the 5-substituent, the precise structural basis for the unique fluorescence of these derivatives has not been systematically investigated. To address this question, we have prepared a series of uracil nucleotides with different aromatic and heteroaromatic

Table 1

substituents in position 5 and systematically investigated the effect of the 5-substituent on their photophysical properties. Specific goals of this study included the identification of structural parameters that determine the quantum yield and excitation/emission wavelength in this class of fluorophores in aqueous media. The highest quantum yields were determined for the thien-2-yl substituted derivatives 2p and 2q (Table 1), which are 10-fold more fluorescent than any other nucleotide analogues with an aryl- or heteroaryl substituent in position 5 that have been reported to date.1a An attractive feature of 2p and 2q is their absorbance and fluorescence emission at relatively long wavelengths, which makes these fluorophores suitable for analyses in biological environments. To illustrate such applications, we have carried out ligand-binding experiments with selected fluorophores and a UDP-sugar-dependent glycosyltransferase. In conjunction with results from solvatochromic experiments, the results from these biological experiments offer, for the first time, a structural explanation for the fluorescence quenching effect that is observed upon binding of these fluorophores to a target protein.

Results and discussion Synthesis To investigate the effect of the additional substituent on the photophysical properties of 5-substituted uracil nucleotides, we prepared a series of uracil nucleotides with a broad range

Yields and photophysical properties of 5-substituted UMP derivatives 2a–q

R

Code

Yield (%)

λmaxa (nm)

λemb (nm)

Stokes shiftc (cm−1)

Fluorescence intensityb (a.u.)

Phenyl 4-Chlorophenyl 4-Methyl-3-nitrophenyl 3-N-Boc-aminomethylphenyl 3-Hydroxyphenyl 2-Hydroxyphenyl 2-Naphthyl 4-Carboxyphenyl 4-Trifluoromethylphenyl 5-Methoxypyridin-3-yl 3-Mesylphenyl 4-Methoxyphenyl Furan-2-yl 5-Formylfuran-2-yl 3-Formylthien-2-yl 5-Acetylthien-2-yl 5-Formylthien-2-yl 5-Oximinothien-2-yl 5-Oximinothien-2-yl

2a 2b 2c 2d 2e 2f 2g 2h 2i 2j 2k 2l 2m 2n 2o 2p 2q Syn-2r Anti-2r

71 66 79 69 52 33 47 60 63 45 58 57 57 56 25 31 61 21 19

278 281 279 281 285 273 221 283 280 289 278 279 314 348 267 348 351 341 341

403 398 434 427 —d 415 450 411 383 414 382 444 437 431 453 433 434 459e 451e

11 157 10 462 12 801 12 168 —d 12 534 23 027 11 005 9605 10 447 9793 13 320 8964 5534 15 378 5641 5449 7539 7153

40 78 6.7 11.6 —d 0.9 32.2 17.6 5.3 74.8 5.3 603 497 640 387 >650 >650 78e 105e

Absorbance spectra recorded in H2O at 10 μM. b Fluorescence spectra recorded in H2O at 100 μM, unless otherwise stated. c Stokes shift = (1/λexcitation − 1/λemission). d Not fluorescent. e In H2O at 20 μM.

a

6358 | Org. Biomol. Chem., 2013, 11, 6357–6371

This journal is © The Royal Society of Chemistry 2013

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Organic & Biomolecular Chemistry of electronically and sterically different 5-substituents. We decided to target primarily the water-soluble uracil nucleotides (i.e. UMP, UDP and UTP derivatives) rather than the corresponding nucleosides in order to allow the investigation of the target fluorophores in aqueous media, the relevant environment for any potential biological application. First, we prepared 5-iodo UMP 1 either by direct iodination of UMP11 or by phosphorylation of 5-iodouridine under Yoshikawa conditions.12 The target 5-substituted UMP derivatives 2a–q were obtained by cross-coupling of 1 to a set of aromatic and heteroaromatic boronic acids under our previously reported aqueous Suzuki–Miyaura conditions (Scheme 1).5a,9b,11,13 Irrespective of their steric and electronic properties, most boronic acids were coupled readily to 5-iodo-UMP 1. This provided us with a series of 17 different UMP analogues 2a–q bearing sterically and electronically diverse 5-substituents as an ideal test set for photophysical studies (Table 1). Further structural variation was achieved by condensation of the 5-(5-formylthien-2-yl) substituted UMP derivative 2q with hydroxylamine (Scheme 2). This transformation furnished the corresponding oximino derivative 2r as a mixture of geometrical isomers, which were readily separated by column chromatography. The configuration of each isomer could be unambiguously assigned from the chemical shift of the oxime proton in the 1H NMR spectrum (anti-2r: 7.85 ppm; syn-2r: 8.35 ppm). 5-Iodo-UMP 1 also provided an excellent starting point for the synthesis of a selection of 5-substituted UDP and UTP

Paper Table 2

Yields for 5-substituted UDP and UTP derivatives

Compound

Code

Yield (%)

5-Iodo UDP 5-Iodo UTP 5-Phenyl UDP 5-(4-Methoxyphenyl) UDP 5-(Furan-2-yl) UDP 5-(5-Formylthien-2-yl) UDP 5-Phenyl UTP 5-(4-Methoxyphenyl) UTP 5-(Furan-2-yl) UTP 5-(5-Formylthien-2-yl) UTP

3 4 5a 5b 5c 5d 6a 6b 6c 6d

66 60 73 50 60 77 85 53 74 67

derivatives (Scheme 1). Thus, 1 was converted into the corresponding phosphoromorpholidate via a Mukaiyama redox activation,14 followed by the reaction of the resulting morpholidate with a tributylammonium phosphate or pyrophosphate salt to afford, respectively, 5-iodo-UDP 3 and 5-iodo-UTP 4. Conveniently, our cross-coupling conditions could also be applied to 3 and 4, as well as the nucleoside substrate 5-iodouridine. Thus, Suzuki–Miyaura cross-coupling of 3 and 4 to a selection of boronic acids provided 5-substituted derivatives of UDP (5a–d) and UTP (6a–d) in moderate to high yields (Table 2). The cross-coupling of 5-iodouridine with 5-formylthien-2-ylboronic acid gave 5-(5-formylthien-2-yl) uridine 7, the nucleoside analogue of UMP derivative 2q (Scheme 1). Photophysical characterisation

Scheme 1 Synthesis of 5-substituted derivatives of UMP (2a–q), UDP (5a–d), UTP (6a–d) and uridine (7). Reagents and conditions: (i) proton sponge, POCl3, MeCN, −5 °C; (ii) R-B(OH)2, TPPTS, Na2Cl4Pd, Cs2CO3, H2O, 60 °C, 1 h; (iii) morpholine, 2,2’-dipyridyldisulfide, PPh3, DMSO, rt, 1 h; (iv) KH2PO4 (3) or K4P2O7 (4), tributylamine, tetrazole, MeCN, rt, 5 h. For substituents R and individual yields see Tables 1 and 2.

Scheme 2 anti-2r.

Synthesis of 5-(5-oximinothien-2-yl) UMP isomers syn-2r and

This journal is © The Royal Society of Chemistry 2013

With a collection of 5-substituted uracil nucleotides in hand, we set out to determine their photophysical properties. First, we recorded the absorbance and fluorescence spectra of UMP derivatives 2a–r at a defined concentration in water (Table 1). This initial study showed that in this series, both the absorbance and emission maxima vary considerably, depending on the nature of the 5-substituent. While most analogues carrying a phenyl or substituted phenyl ring (2a–l) have an absorbance maximum at around 280 nm, the absorbance maxima of most analogues bearing a 5-membered heteroaromatic ring was, as a general trend, significantly red-shifted (2m–r). This red-shift is about 35 nm for the derivative with a simple, unsubstituted furan substituent (2m), and about 70 nm for derivatives with an additional formyl or acetyl group attached to a furan or thiophene ring (2n, 2p, 2q). The one exception to this general trend is the 5-(3-formylthien-2-yl) substituted derivative 2o, whose absorbance maximum is slightly blue-shifted compared to the phenyl-substituted derivatives. Interestingly, the UV absorbance profile of 2o is in general very different from those of its closest structural analogues 2n and 2q (Fig. 2). UMP analogues with a six-membered 5-substituent (2a–l) were relatively poor fluorescence emitters, with the exception of 5-(4-methoxyphenyl)-UMP 2l, which was strongly fluorescent under these conditions (Table 1). This is in agreement with the pronounced fluorescence emission that has previously been reported both for the corresponding, 5-methoxyphenylsubstituted derivatives of uridine9a and UDP-galactose.5a In

Org. Biomol. Chem., 2013, 11, 6357–6371 | 6359

View Article Online

Paper

Organic & Biomolecular Chemistry Table 3

Quantum yields for selected 5-substituted UMP fluorophores in water

R

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

a

H 4-Chlorophenyl 4-Methoxyphenyl Furan-2-yl 5-Formylfuran-2-yl 3-Formylthien-2-ylb 5-Acetylthien-2-ylb 5-Formylthien-2-ylb 5-Oximinothien-2-yl 5-Oximinothien-2-yl

Code

Q. Y.

UMP 2b 2l 2m 2n 2o 2p 2q Syn-2r Anti-2r

5 × 10–5 0.001 0.02 0.04 0.04 0.02 0.24 0.26 0.017 0.031

a From ref. 29. b See ref. 9e for comparison with the unsubstituted parent 5-(thien-2-yl) uridine: QY (H2O) 0.01, λex 314 nm, λem 434 nm.

Fig. 2 Absorbance (top) and fluorescence (bottom) spectra of selected UMP derivatives in water at 10 μM (absorbance) and 100 μM (fluorescence). Fluorescence emission was measured after excitation at λmax.

contrast to most phenyl-substituted UMP derivatives, analogues with a 5-membered heteroaromatic substituent (2m–q) consistently displayed pronounced fluorescence emission. 5-Substituted uracil nucleosides with a simple thiophene or furan substituent in position 5 have previously been reported as fluorescence emitters.9d,e In our series, the strongest fluorescence signal was observed for UMP derivatives with a formyl- or acetyl-substituted thiophene or furan ring in position 5 (2n, 2p, 2q). Indeed, two of these derivatives (2p, 2q) showed saturating fluorescence at 100 μM (Table 1). These three fluorophores also displayed the most pronounced bathochromic shifts for their absorbance maxima, with λmax at around 350 nm (Table 1). While the 5-formyl- or 5-acetyl-substituted furan and thiophene derivatives 2n, 2p, 2q all showed qualitatively similar absorbance and fluorescence spectra, stronger absorbance and weaker fluorescence emission was observed for the furan derivative 2n than for the thiophene derivatives 2p and 2q (Fig. 2). The intense fluorescence emission of 5-(5-formylthien-2-yl) UMP 2q was also preserved for the corresponding 5-(5-formylthien-2-yl) derivative of UDP (5d) and UTP (6d), indicating that, as expected, the presence of the additional phosphate groups has little influence on the absorbance and fluorescence emission of these 5-(5-formylthien-2-yl) uracil nucleotides (Fig. S1†). In order to obtain a more quantitative measure of fluorescence emission in this series, we next determined the quantum yields for selected UMP derivatives (Table 3). These measurements were performed in water following the procedure reported by Nighswander-Remper.15 Quantum yield

6360 | Org. Biomol. Chem., 2013, 11, 6357–6371

measurements confirmed that 2p and 2q are the strongest fluorescence emitters in this series. In particular, the 4-methoxyphenyl-substituted derivative 2l, which had shown similar fluorescence intensity in the initial screen, had a more than 10-fold lower quantum yield than 2p and 2q. To the best of our knowledge, the quantum yields of 2p and 2q are the highest for any 5-substituted uracil nucleotide and nucleoside fluorophores that have been reported to date.9 Particularly noteworthy is the 10-fold greater quantum yield for these substituted thiophene derivatives compared to the parent compound bearing an unsubstituted thiophene ring in position 5 (Table 3). Interestingly, the same effect was not observed in the furan series, where the same quantum yield was determined for 2m, bearing an unsubstituted furan-2-yl ring in position 5, and the 5-(5-formylfuran-2-yl) derivative 2n. Conversion of the formyl group in 2q into the corresponding oximino group in 2r led to a significant drop in quantum yield. Interestingly, however, the quantum yield of the anti-aldoxime (anti-2r) was about double than that of its geometrical isomer (syn-2r). Conformational analysis In order to better understand the interplay between structure and fluorescence in this series, we next carried out an NMRbased conformational analysis of those UMP derivatives, for which quantum yields had been determined. Important conformational parameters in nucleosides and nucleotides are the sugar pucker of the ribose (S-type vs. N-type) and the syn- or anti-orientation of the nucleobase relative to the ribose ring (Table 4). Both of these parameters can be extracted from 1 H NMR data using established empirical approaches: the coupling constant J1′,2′ is indicative of the ribose conformation, while the chemical shift of the H-2′ signal, relative to the 5-unsubstituted parent UMP, is a good indicator for the position of the syn/anti equilibrium.16 In aqueous solution, all 5-substituted UMP derivatives that were included in this analysis displayed a slight preference for the S-type conformation of the ribose, similar to UMP itself (Table 4). In addition, the chemical shifts for H-2′ did not deviate substantially from that of 5-unsubstituted UMP. This suggests that, like the parent nucleotide UMP, all of these 5-substituted derivatives adopt preferentially the anti conformation, in which the 5-substituent is oriented

This journal is © The Royal Society of Chemistry 2013

View Article Online

Organic & Biomolecular Chemistry

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Table 4

Paper

Conformational analysis of selected 5-substituted UMP derivatives

R

Code

S-type conformer (%) (10× J1′,2′)

H-2′ (ppm)

ΔH-2′ (ppm)

Syn/antia

H-6 (ppm)

H 4-Chlorophenyl 4-Methoxyphenyl Furan-2-yl 5-Formylfuran-2-yl 3-Formylthien-2-yl 5-Acetylthien-2-yl 5-Formylthien-2-yl 5-Oximinothien-2-yl 5-Oximinothien-2-yl

UMP 2b 2l 2m 2n 2o 2p 2q Syn-2ra Anti-2ra

51 55 56 53 53 n.d.b 52 50 57 57

4.41 4.45 4.42 4.45 4.51 4.42 4.46 4.45 4.46 4.46

0.00 0.04 0.01 0.04 0.1 0.01 0.05 0.04 0.05 0.05

anti anti anti anti anti anti anti anti anti anti

8.13 7.90 7.76 8.15 8.40 8.13 8.35 8.38 8.17 8.17

a

The syn/anti nomenclature of oximes is not to be confused with the syn/anti equilibrium about the glycosidic bond. b Not determined due to overlapping peaks in the 1H NMR spectrum.

Table 5

Calculated dihedral angle α for the relative orientation of the 5-substituent and the uracil ring in UMP derivatives 2o–2r

N-type R

Code

Energy difference: N-type vs. S-type (kcal mol−1)

3-Formyl 5-Acetyl 5-Formyl 5-Oximino 5-Oximino

2o 2p 2q Syn-2r Anti-2r

−5.73 −2.80 −2.69 −2.70 −1.89

towards the ribose ring. These conformational preferences are in agreement with those previously observed by X-ray crystallography for related 5-(hetero)aryl uridine derivatives.9d,e,17 To gain further insights into the conformational preferences of these fluorophores, we calculated energy-minimized geometries for the UMP derivatives 2o–2r. The absorbance spectra obtained from these calculations were generally in good agreement with experimental results (ESI, Fig. S2†). In agreement with the results from our NMR studies, which suggested an almost 1 : 1 ratio of N-type and S-type conformations, our calculations showed only very small energy differences between N-type and S-type conformations (Table 5). The marginally higher stability of the N-type conformers in the implicit solvent simulations may be due to an intramolecular hydrogen bond between the phosphate group and the C3′ hydroxyl group of the ribose. Interestingly, for 2o an additional intramolecular hydrogen bond was observed between the phosphate group and the 3-formyl substituent. Most importantly,

This journal is © The Royal Society of Chemistry 2013

S-type

α

“Out-of-plane”

α

“Out-of-plane”

−135.3° −161.4° −162.2° −159.7° −165.3°

ca. 45° ca. 20° ca. 18° ca. 20° ca. 15°

−125.3° 162.7° 162.7° 161.5° 160.9°

ca. 55° ca. 17° ca. 17° ca. 18° ca. 19°

however, the dihedral angle α, which describes the relative orientation of the 5-substituent and the uracil ring to each other, was very similar for both the N-type and S-type conformation of each derivative. Thus, 2p, 2q and 2r show an almost coplanar orientation of both rings, in both conformations (Table 5). In contrast, the calculated conformations of 2o suggest that in this case, the thiophene and uracil rings are oriented “out-of-plane” (Fig. 3). Solvatochromism Next, we investigated the influence of different solvents on the absorbance and fluorescence of the 5-(5-formylthien-2-yl) uracil fluorophore, which gave the strongest emission in this series. As the 5-(5-formylthien-2-yl)-substituted nucleotide 2q has only limited solubility in organic solvents, we chose 5-(5formylthien-2-yl) uridine 7, the nucleoside analogue of 2q, for these investigations. We prepared 200 nM solutions of 7 in three solvents with a different polarity index (P): water (P = 9),

Org. Biomol. Chem., 2013, 11, 6357–6371 | 6361

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Paper

Fig. 3 Chemical structures and calculated energy-minimised S-type conformers of 2o (left) and 2q (right). The different orientations of the formylthiophene substituent relative to the uracil ring in both fluorophores are clearly visible (2o: out-of-plane, 2q: coplanar).

Organic & Biomolecular Chemistry

Fig. 5

Fluorescence emission of 7 in different solvent mixtures (λem 450 nm).

containing less than 70% of water, both in the case of H2O–IPA and H2O–MeCN, whereas a steep, non-linear decrease in fluorescence was observed for solvent mixtures containing between 85%–70% water. Such sudden variations in fluorescence intensity are characteristic for non-specific solvent effects mediated by hydrogen-bond interactions.18 The observation that fluorescence emission was strongest in the two protic solvents used in this study (H2O, IPA), and weaker in the aprotic MeCN, may therefore suggest a possible role for hydrogen bonding in the fluorescence of the 5-(5-formylthien2-yl) uracil fluorophore in solution.

Fig. 4 Absorbance (dashed lines) and fluorescence emission (full lines) of 5-(5formylthien-2-yl) uridine 7 at 200 nM in water (blue), isopropanol (red) and acetonitrile (green).

acetonitrile (MeCN, P = 5.8) and isopropanol (IPA, P = 3.9) and recorded their absorbance and fluorescence spectra (Fig. 4). While the absorbance of 7 was largely unaffected by the nature of the solvent, its fluorescence was strongly solventdependent, with fluorescence emission about 20-fold stronger in water than in MeCN. Interestingly, fluorescence and solvent polarity were not directly correlated, since we observed higher fluorescence in IPA than in MeCN. Moreover, if polarity was the source of the variation in fluorescence, we would also have expected to see a correlation with a hypsochromic shift in the fluorescence spectra, which was not the case. In order to better understand these variations in fluorescence, we carried out a second set of experiments, recording the fluorescence emission of 7 in different solvent mixtures (H2O–IPA and H2O– MeCN) at different ratios H2O : organic solvent (Fig. 5). These experiments confirmed a complex correlation between solvent polarity and fluorescence intensity for this fluorophore. As expected from the preceding experiments, fluorescence emission is strongest in water and solvent mixtures containing >85% water. An increase in the concentration of the organic solvent led to a decrease in fluorescence intensity. However, this decrease in fluorescence is linear only for solvent mixtures

6362 | Org. Biomol. Chem., 2013, 11, 6357–6371

Discussion of photophysical properties In principle, various explanations are conceivable for the differences in fluorescence emission that were observed for the different 5-substituted UMP derivatives in this study, including charge transfer contributions such as the twisted intramolecular charge transfer (TICT) model,19 and conformational effects.9c We reasoned that a key to understanding the interplay of structure and fluorescence emission in this series may be provided by the regioisomeric 5-formylthien-2-yl fluorophores 2o and 2q, which differ markedly in their photophysical and conformational properties. Thus, 5-(3-formylthien-2-yl) UMP 2o had a quantum yield more than 10-fold lower than 5-(5-formylthien-2-yl) UMP 2q, the strongest emitter in this series, while its λmax was blue-shifted by 84 nm compared to 2q. In this context, it appeared significant that 2o and 2q display different conformational preferences with regard to the relative orientation of the formylthiophene and uracil rings, the two constituent parts of the formylthien-2-yl uracil fluorophore. While the uracil and thiophene rings adopt a nearly coplanar orientation in 2q, they are forced out of plane in the corresponding 5-(3-formylthien-2-yl) derivative 2o (Fig. 3). As efficient electron delocalisation along π-bonds is limited to planar systems, the non-coplanar orientation of the two rings in 2o directly affects electron delocalisation across the formylthien-2-yl uracil system. Experimentally, this is reflected in the different chemical shifts for the H-6 proton, which is shifted 0.25 ppm upfield for 5-(3-formylthien-2-yl)

This journal is © The Royal Society of Chemistry 2013

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Organic & Biomolecular Chemistry UMP 2o compared to its regioisomer 2q (Table 4), indicating that the formyl group in 2q exerts a stronger electron-withdrawing effect on the uracil ring than in 2o. Taken together, these results suggest that electron delocalisation, possibly via a push–pull system involving the nitrogen in position N1, may play an important role for the strong fluorescence emission of fluorophore 2q and closely related derivatives (e.g. 2p). Thus, the fluorescence intensity of these fluorophores may be attributed to effective electron delocalisation across the entire 5-thienyl uracil fragment, resulting from both the presence of the additional carbonyl group at the thiophene ring, and the coplanar orientation of the substituted thiophene and uracil rings. Intriguingly, this interpretation would also offer an explanation for the effects we have observed in our solvatochromic experiments, which show increased fluorescence emission for the 5-formylthien-2-yl fluorophore in protic solvents. Our computational results indicate that hydrogen bonding plays a role for the stabilisation of the preferred conformations of individual fluorophores. It can therefore be speculated that in the case of strong emitters (2p, 2q), hydrogen bonding in protic solvents may stabilise the fluorescent, coplanar conformation e.g. through a bi- or multi-molecular network of solvent molecules and fluorophore (Fig. S3, ESI†). This hypothesis, while speculative at this point, is supported by the observation that in the case of 2q and anti-2r, closely related regioisomers (2o, syn-2r), for which this hydrogen bonding stabilisation is not possible, due to their different geometries, are weaker fluorophores. Application of fluorophores in ligand-binding experiments Previously, we have shown that the fluorescent, 5-(5-formylthien-2-yl)-substituted UDP-galactose derivative 8 (Fig. 6) can be used as a fluorescent probe in ligand displacement assays with bacterial and mammalian galactosyltransferases (GalTs), including B. taurus α-1,3-GalT.5a This application was based on the quenching of the fluorescence of sugar-nucleotide 8 upon the specific binding to the respective target enzyme. With the corresponding 5-(5-formylthien-2-yl) uridine (7), UMP (2q) and UDP (5d) derivatives in hand, we investigated if a similar quenching effect would be observed with these alternative, structurally simplified ligands for B. taurus α-1,3-GalT. Interestingly, in these experiments, efficient quenching was still observed with 5-(5-formylthien-2-yl) UDP 5d, while the effect was less pronounced with 5-(5-formylthien2-yl) UMP 2q, and disappeared altogether with 5-(5-formylthien-2-yl) uridine 7 (Fig. 6). This rank order corresponds to the known binding affinities of the 5-unsubstituted parent molecules UDP, UMP and uridine for UDP-galactose-dependent GalTs.20 While UDP-galactose and UDP often have similar binding constants for a given GalT, removal of one or both phosphates invariably leads to a drop in affinity.20 The same correlation holds for our series of fluorophores. Thus, the lack of fluorescence quenching in the case of uridine 7 can very likely be attributed to the lack of binding affinity of this fluorophore for B. taurus α-1,3-GalT.

This journal is © The Royal Society of Chemistry 2013

Paper

Fig. 6 (a) Fluorescence quenching upon incubation of α-1,3-GalT with, respectively, 5-(5-formylthien-2-yl) UDP-galactose 8, 5d, 2q and 7. (b) Chemical structures of fluorophores 8, 5d, 2q and 7.

On the other hand, results from our photophysical studies provide, for the first time, a structural explanation for the strong fluorescence quenching observed for 5d and 8 upon binding at B. taurus α-1,3-GalT. Our photophysical studies indicate that a coplanar orientation of the 5-formylthien-2-yl and uracil rings, possibly stabilised by hydrogen bonding, is a prerequisite for fluorescence emission in this series. The strong fluorescence quenching observed for 5d and 8 may therefore be caused by the binding of these GalT ligands in an orientation that forces the fluorogenic 5-substituent out of the plane of the uracil ring and/or disrupts hydrogen bonding to O4 and the formyl group. Based on the results from our photophysical studies, such a non-coplanar orientation would be expected to lead to a drop in fluorescence, as observed in our experiments.

Conclusion In this study, we have developed a rapid and convenient synthesis of 5-aryl and 5-heteroaryl-substituted UMP, UDP and UTP derivatives. This has allowed the systematic investigation of the photophysical properties of these non-natural nucleotides. Two novel UMP derivatives bearing, respectively, a 5-acetylthien-2yl (2p) and 5-formylthien-2yl (2q) group in position 5 stood out as the strongest fluorescence emitters in this

Org. Biomol. Chem., 2013, 11, 6357–6371 | 6363

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Paper series, with quantum yields of around 0.25. This represents a 10-fold improvement compared to previously reported 5-substituted uracil fluorophores. The superiority of a 5-formyl- or 5-acetyl thiophene ring as a fluorogenic substituent, compared to both 5-unsubstituted heterocycles and various substituted phenyl substituents, appears to be due to two factors: first, the extended delocalised π electron system in the case of the 5-formyl- or 5-acetylthienyl uracil fluorophore, and second the coplanar orientation of the fluorogenic 5-substituent and the uracil ring. Importantly, the conformational preferences of 5-(5-formylthien-2-yl) UMP 2q, the strongest fluorescence emitter in this series, allow such a coplanar orientation, which is conducive to efficient electron delocalisation across the two ring systems. In contrast, in the regioisomer 5-(3-formylthien2-yl) UMP 2o, the presence of the formyl group in position 3 of the thiophene ring prevents the coplanar orientation of the fluorogenic 5-substituent and the uracil ring and, possibly as a direct consequence, leads to a markedly reduced quantum yield. This interpretation may also provide an explanation both for the solvent dependency of fluorescence emission in this series of fluorophores, and for the fluorescence quenching which we have observed in ligand binding experiments. The strongest fluorescence emission of the 5-(5-formylthien-2-yl) uracil fluorophores was observed in water. This finding raises the possibility that in protic solvents the fluorescent, coplanar orientation of the 5-formylthien-2-yl and uracil rings may be stabilised through hydrogen bonding. The role of hydrogen bonding for fluorescence in this series is further illustrated by results for the oximino-substituted fluorophore 2r, where the anti-isomer (which can engage in hydrogen bonding) has a greater quantum yield than the syn-isomer (which cannot). Finally, it can be speculated that this stabilising interaction, and consequently the coplanar orientation of the fluorogenic substituent and the uracil ring, may be disrupted upon binding of these fluorophores to a target protein, leading to the strong fluorescence quenching observed in our ligand binding experiments. Taken together, these findings provide new insights into the structural basis for fluorescence and fluorescence quenching in this series of 5-substituted uracil nucleotides. The results from this study therefore provide a basis for the further rational optimisation of this important class of fluorophores, and for their application as sensors in biological assays.

Experimental section All chemicals were obtained commercially and used as received unless stated otherwise. Fluorophore 8 was synthesised as previously reported.5a TLC was performed on precoated aluminium plates (Silica Gel 60 F254, Merck). Compounds were visualized by exposure to UV light (254 and 365 nm). NMR spectra were recorded at 298 K on a Varian VXR 400 S spectrometer, a Bruker Avance DRX-300 or a Bruker Avance DPX-400 spectrometer. Chemical shifts (δ) are reported

6364 | Org. Biomol. Chem., 2013, 11, 6357–6371

Organic & Biomolecular Chemistry in ppm and referenced to methanol (δH 3.34, δC 49.50) for solutions in D2O, to DMSO (δH 2.50, δC 39.52) or to CDCl3 (δH 7.26, δC 77.16). Coupling constants (J) are reported in Hz. Resonance allocations were made with the aid of COSY and HSQC experiments. NOESY measurements were carried out with a relaxation delay of 1.000 s and a mixing time of 0.200 s. Accurate electrospray ionisation mass spectra (HR ESI-MS) were obtained on a Finnigan MAT 900 XLT mass spectrometer at the EPSRC National Mass Spectrometry Service Centre in Swansea. Preparative chromatography was performed on a Biologic LP chromatography system equipped with a peristaltic pump and a 254 nm UV Optics Module under the following conditions. Purification method 1 Ion-pair chromatography was performed using Lichroprep RP-18 resin gradient 0–10% acetonitrile (or methanol) against 0.05 M TEAB (triethylammonium bicarbonate: prepared by bubbling CO2 gas through a mixture of Et3N in water until saturation was achieved) over 480 mL, flow rate 5 mL min−1. Product-containing fractions were combined and reduced to dryness. The residue was co-evaporated repeatedly with methanol to remove residual TEAB. Purification method 2 Anion-exchange chromatography was performed using a Macro prep 25Q resin, gradient 0–100% 1 M TEAB ( pH 7.3) against H2O over 480 mL, flow rate 5 mL min−1. Productcontaining fractions were combined and reduced to dryness. The residue was co-evaporated repeatedly with methanol to remove residual TEAB. General method for the Suzuki–Miyaura cross-coupling of 5-iodouridine, 5-iodo UMP (1), 5-iodo UDP (3) and 5-iodo UTP (4). A 2-necked round bottom flask with 5-iodouridine or the respective 5-iodouridine nucleotide 1, 3 or 4 (1 equiv.), Cs2CO3 (2 equiv.) and the requisite boronic acid (1.5 equiv.) was purged with N2. TPPTS (0.0625 equiv.), Na2Cl4Pd (0.025 equiv.) and degassed H2O (4 mL) were added, and the reaction was stirred under N2 for 1 h at 60 °C. Upon completion, the reaction was cooled to room temperature. The black suspension was concentrated under reduced pressure, and the residue was taken up in MeOH. The methanolic suspension was filtered through celite, and the residue was washed with methanol. The combined filtrates were evaporated under reduced pressure and the residue was purified using purification method 1. 5-Iodouridine-5′-monophosphate (1).21 A suspension of 5-iodouridine (480 mg, 1.3 mmol)22 and proton sponge (1.7 g, 7.8 mmol) in dry acetonitrile (20 mL) was cooled to −5 °C and stirred under N2. POCl3 (58 μL, 0.6 mmol) was added dropwise to this suspension. The orange-coloured reaction was stirred at −5 °C for 4 h, at which time TLC indicated near complete conversion (IPA–H2O–NH3 6 : 3 : 1; Rf 0.31; RfSM 0.71). The reaction was quenched with 250 mL of ice cold 0.2 M TEAB buffer. The pale orange solution was stirred for 1 h at 0 °C. After allowing to reach 25 °C, the aqueous layer was washed with ethyl

This journal is © The Royal Society of Chemistry 2013

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Organic & Biomolecular Chemistry acetate (3×) and concentrated under reduced pressure. The crude residue was purified sequentially by purification methods 1 and 2 to provide 1 in its triethylammonium salt form (1.7 equiv.) as a colourless, glassy solid in 53% yield (472 mg). δH (400 MHz, D2O) 3.96–4.10 (2H, m, H-5′), 4.22–4.26 (1H, m, H-4′), 4.31 (1H, t, J = 4.7 Hz, H-3′), 4.38 (1H, t, J = 5.3 Hz, H-2′), 5.93 (1H, d, J = 5.3 Hz, H-1′), 8.27 (1H, s, H-6); δC (75.5 MHz, D2O) 64.8 (d, JC,P = 4.6 Hz, C-5′), 69.6 (C-5), 70.8 (C-3′), 74.6 (s, C-2′), 84.8 (d, JC,P = 8.8 Hz, C-4′), 89.6 (C-1′), 147.1 (C-6), 152.8 (C-2), 164.3 (C-4); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 448.9255 [M − H]−, C9H11IN2O9P requires 448.9252. 5-Phenyluridine-5′-monophosphate (2a). The triethylammonium salt of the title compound was obtained as a glassy solid in 71% yield (13.5 mg) from 1 (16.5 mg, 36.7 μmol) and phenylboronic acid according to the general method. δH (400 MHz, D2O) 4.01–4.15 (2H, m, H-5′), 4.24–4.30 (1H, m, H-4′), 4.32–4.40 (1H, m, H-3′), 4.45 (1H, t, J = 5.5 Hz, H-2′), 5.98 (1H, d, J = 5.6 Hz, H-1′), 7.38–7.51 (5H, m, Ph), 7.82 (1H, s, H-6); δC (75.5 MHz, D2O) 65.0 (d, JC,P = 4.5 Hz, C-5′), 70.6 (C-3′), 74.1 (C-2′), 84.3 (d, JC,P = 8.3 Hz, C-4′), 89.2 (C-1′), 116.9 (C-5), 129.1, 129.4, 129.5, 132.4 (C-Ph), 139.2 (C-6), 152.3 (C-2), 165.5 (C-4); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 399.0593 [M − H]−, C15H16N2O9P requires 399.0599. 5-(4-Chlorophenyl)-uridine-5′-monophosphate (2b). The triethylammonium salt of the title compound was obtained as a glassy solid in 20% yield (5.5 mg) from 1 (20 mg, 49.6 μmol) and 4-chlorophenylboronic acid according to the general method. δH (400 MHz, D2O) 4.00–4.12 (2H, m, H-5′), 4.22–4.29 (1H, m, H-4′), 4.30–4.344 (1H, m, H-3′), 4.45 (1H, d, J = 5.4 Hz, H-2′), 6.01 (1H, d, J = 5.5 Hz, H-1′), 7.46–7.58 (4H, m, Ph), 7.90 (1H, s, H-6); δC (75.5 MHz, D2O), 65.0, 70.9, 74.3, 82.5, 89.6, 111.6, 129.6, 129.9, 131.2, 131.4, 139.6, 152.4, 165.8; δP (121.5 MHz, D2O) 10.4. m/z (ESI) 433.0210 [M − H]−, C15H15Cl35N2O9P requires 433.0209. 5-(3-Nitro-(4-methylphenyl))-uridine-5′-monophosphate (2c). The triethylammonium salt of the title compound was obtained as a glassy solid in 79% yield (10.9 mg) from 1 (10 mg, 22.2 μmol) and 4-methyl-3-nitrophenylboronic acid according to the general method. δH (400 MHz, D2O) 2.57 (3H, s, Me), 4.01–4.12 (2H, m, H-5′), 4.26–4.28 (1H, m, H-4′), 4.33 (1H, t, J = 4.7 Hz, H-3′), 4.45 (1H, t, J = 5.4 Hz, H-2′), 6.00 (1H, d, J = 5.5 Hz, H-1′), 7.46–7.71 (2H, 2d, J = 7.9 and 8.0 Hz, Ph), 7.99 (1H, s, H-6) 8.16 (1H, s, Ph); δC (75.5 MHz, D2O) 19.8 (Me), 64.9 (C-5′), 70.5 (C-3′), 74.3 (C-2′), 84.3 (d, J = 9.1 Hz, C-4′), 89.4 (C-1′), 114.4, 115.8, 125.2, 131.3, 133.9, 134.0 (C-5 + C-Ph), 139.7 (C-6), 149.2 (C-Ph), 152.0 (C-2), 164.9 (C-4); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 458.0607 [M − H]−, C16H17N3O11P requires 458.0606. 5-(3-(N-Boc-methylamine)-phenyl)-uridine-5′-monophosphate (2d). The triethylammonium salt of the title compound was obtained as a glassy solid in 69% yield (10.2 mg) from 1 (10 mg, 22.2 μmol) and 3-N-Boc-aminomethylphenylboronic acid according to the general method. δH (400 MHz, D2O) 1.42 (9H, s, tBu), 4.00–4.10 (2H, m, H-5′), 4.20–4.35 (4H, m, H-4′, H-3′, CH2), 4.47 (1H, t, J = 5.5 Hz, H-2′), 6.01 (1H, d, J = 5.7 Hz,

This journal is © The Royal Society of Chemistry 2013

Paper H-1′), 7.30–7.48 (4H, m, Ph), 7.86 (1H, s, H-6); δC (75.5 MHz, D2O) 20.4 (CH2), 28.2 (tBu), 65.1 (d, JC,P = 6.0 Hz, C-5′), 70.7 (C-3′), 74.1 (C-2′), 84.3 (d, JC,P = 8.3 Hz, C-4′), 84.9 (tBu), 89.3 (C-1′), 101.4 (C-Ph), 116.7 (C-5), 127.4, 127.7, 128.2, 129.7, 132.7 (C-Ph), 139.2 (C-6), 152.2 (C-2), 159.1 (CHO), 165.4 (C-4); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 528.1384 [M − H]−, C21H27N3O11P requires 528.1389. 5-(3-Hydroxyphenyl)-uridine-5′-monophosphate (2e). The triethylammonium salt of the title compound was obtained as a glassy solid in 52% yield (6.8 mg) from 1 (10 mg, 22.2 μmol) and 3-hydroxyphenylboronic acid according to the general method. δH (400 MHz, D2O) 4.05–4.12 (2H, m, H-5′), 4.27–4.31 (1H, m, H-4′), 4.40 (1H, t, J = 4.7 Hz, H-3′), 4.47 (1H, t, J = 4.9 Hz, H-2′), 6.01 (1H, d, J = 4.9 Hz, H-1′), 6.86 (1H, d, J = 7.3 Hz, Ph), 7.23 (1H, s, Ph), 7.25–7.27 (1H, m, Ph), 7.33 (1H, t, J = 7.8 Hz, Ph), 8.09 (1H, s, H-6); δC (75.5 MHz, D2O) 64.3 (d, JC, P = 3.8 Hz, C-5′), 70.2 (C-3′), 74.8 (C-2′), 84.5 (d, JC,P = 9.8 Hz, C-4′), 89.4 (C-1′), 115.4, 115.6, 116.0, 120.5, 120.7, 130.7, 133.9 (C-5 + C-Ph), 139.3 (C-6), 152.0 (C-2), 156.9 (C-4); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 415.0540 [M − H]−, C15H16N2O10P requires 415.0548. 5-(2-Hydroxyphenyl)-uridine-5′-monophosphate (2f ). The triethylammonium salt of the title compound was obtained as a glassy solid in 33% yield (4.4 mg) from 1 (10 mg, 22.2 μmol) and 2-hydroxyphenylboronic acid according to the general method. δH (400 MHz, D2O) 3.98–4.01 (2H, m, H-5′), 4.23–4.27 (1H, m, H-4′), 4.30–4.34 (1H, m, H-3′), 4.42 (1H, t, J = 5.4 Hz, H-2′), 6.04 (1H, d, J = 5.5 Hz, H-1′), 7.02 (2H, dd, J = 7.6 and 11.5 Hz, Ph), 7.30–7.35 (2H, m, Ph), 7.98 (1H, s, H-6); δC (150 MHz, D2O) 64.3, 70.6, 74.3, 84.4 (d, JC,P = 8.6 Hz), 89.0, 112.7, 117.7, 120.2, 121.4, 130.7, 132.0, 141.2, 152.1, 154.0, 165.4; δP (121.5 MHz, D2O) 7.6. m/z (ESI) 415.0546 [M − H]−, C15H16N2O10P requires 415.0548. 5-(2-Naphthyl)-uridine-5′-monophosphate (2g). The triethylammonium salt of the title compound was obtained as a glassy solid in 47% yield (6.3 mg) from 1 (10 mg, 22.2 μmol) and 1-naphthylboronic acid according to the general method. δH (400 MHz, D2O) 3.90–4.03 (2H, m, H-5′), 4.22–4.49 (4H, m, H-2′, H-3′, H-4′), 6.07 (1H, d, J = 5.6 Hz, H-1′), 7.49–7.65 (4H, m, napht), 7.72–7.78 (1H, m, napht), 7.93 (1H, s, H-6), 8.01 (2H, m, napht); δC (75.5 MHz, D2O) 64.6 (C-5′), 70.9 (C-3′), 73.8 (C-2′), 83.7 (C-4′), 89.3 (C-1′), 113.2 (C-5), 117.6, 125.8, 126.5, 127.1, 127.4, 129.1, 129.5, 130.0, 130.1, 133.9 (C-naphthyl), 140.9 (C-6), 154.6 (C-2), 166.0 (C-4); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 449.0760 [M − H]−, C19H18N2O9P requires 449.0755. 5-(4-Carboxyphenyl)-uridine-5′-monophosphate (2h). The triethylammonium salt of the title compound was obtained as a glassy solid in 60% yield (9.6 mg) from 1 (10 mg, 22.2 μmol) and 4-carboxyphenylboronic acid according to the general method. δH (400 MHz, D2O) 4.00–4.08 (2H, m, H-5′), 4.25–4.30 (1H, m, H-4′), 4.31–4.35 (1H, m, H-3′), 4.48 (1H, t, J = 5.6 Hz, H-2′), 6.00 (1H, d, J = 5.6 Hz, H-1′), 7.38, 7.69 (4H, 2d, J = 8.1 Hz, Ph), 7.73 (1H, s, H-6); δC (75.5 MHz, D2O) 64.7 (d, JC,P = 5.3 Hz, C-5′), 70.6 (C-3′), 73.8 (C-2′), 84.3 (d, JC,P = 8.3 Hz, C-4′), 89.7 (C-1′), 116.3 (C-5), 129.2, 129.7, 135.2, 136.7 (C-Ph), 139.8 (C-6), 152.2 (C-2), 165.3 (C-4), 171.1 (COOH); δP (121.5 MHz,

Org. Biomol. Chem., 2013, 11, 6357–6371 | 6365

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Paper D2O) 7.6. m/z (ESI) 443.0498 [M − H]−, C16H16N2O11P requires 443.0497. 5-(4-Trifluoromethylphenyl)-uridine-5′-monophosphate (2i). The triethylammonium salt of the title compound was obtained as a glassy solid in 63% yield (9.2 mg) from 1 (10 mg, 22.2 μmol) and 4-trifluoromethylphenylboronic acid according to the general method. δH (400 MHz, D2O) 4.01–4.06 (2H, m, H-5′), 4.24–4.28 (1H, m, H-4′), 4.30–4.36 (1H, m, H-3′), 4.46 (1H, t, J = 5.6 Hz, H-2′), 6.02 (1H, d, J = 5.7 Hz, H-1′), 7.68–7.80 (4H, 2d, J = 8.3 Hz, Ph), 7.98 (1H, s, H-6); δC (75.5 MHz, D2O) 64.6 (d, JC,P = 3.8 Hz, C-5′), 70.6 (C-3′), 74.1 (C-2′), 84.5 (d, JC,P = 8.3 Hz, C-4′), 89.4 (C-1′), 115.6 (C-5), 126.1 (q, J = 3.8 Hz, C-F3), 129.8, 130.2, 136.3 (C-Ph), 140.1 (C-6), 152.2 (C-2), 157.7 (C-Ph), 165.1 (C-4); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 467.0473 [M − H]−, C16H15F3N2O9P requires 467.0473. 5-(5-Methoxy-(3-pyridyl))-uridine-5′-monophosphate (2j). The triethylammonium salt of the title compound was obtained as a glassy solid in 45% yield (6.6 mg) from 1 (10 mg, 22.2 μmol) and 5-methoxypyridine-3-boronic acid according to the general method. δH (400 MHz, D2O) 3.92 (3H, s, MeO), 4.01–4.07 (2H, m, H-5′), 4.26–4.28 (1H, m, H-4′), 4.33 (1H, t, J = 4.7 Hz, H-3′), 4.46 (1H, t, J = 5.4 Hz, H-2′), 6.01 (1H, d, J = 5.5 Hz, H-1′), 7.59 (1H, s, pyr), 7.68–8.12 (2H, m, pyr), 8.00 (1H, s, H-6); δC (150 MHz, D2O) 56.5, 64.7, 70.5, 74.1, 84.3, 89.5, 129.2, 131.1, 130.8, 135.4, 136.5, 140.1, 144.2, 152.0, 164.9; δP (121.5 MHz, D2O) 7.6. m/z (ESI) 430.0658 [M − H]−, C15H17N3O10P requires 430.0657. 5-(3-Methanesulfonylphenyl)-uridine-5′-monophosphate (2k). The triethylammonium salt of the title compound was obtained as a glassy solid in 58% yield (8.8 mg) from 1 (10 mg, 22.2 μmol) and 3-(methylsulfonyl)phenylboronic acid according to the general method. δH (400 MHz, D2O) 3.30 (3H, s, Me), 4.00–4.06 (2H, m, H-5′), 4.25–4.27 (1H, m, H-4′), 4.32 (1H, t, J = 4.8 Hz, H-3′), 4.48 (1H, t, J = 5.5 Hz, H-2′), 6.02 (1H, d, J = 5.6 Hz, H-1′), 7.70–8.13 (4H, m, Ph), 8.03 (1H, s, H-6); δC (75.5 MHz, D2O) 43.8 (Me), 64.6 (C-5′), 70.5 (C-3′), 74.1 (C-2′), 84.4 (d, JC,P = 7.5 Hz, C-4′), 89.5 (C-1′), 115.0 (C-5), 127.3, 127.9, 130.9, 134.0, 139.4, 139.5 (C-Ph), 140.2 (C-6), 152.1 (C-2), 165.1 (C-4); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 477.0380 [M − H]−, C16H18N2O11PS requires 477.0374. 5-(4-Methoxyphenyl)-uridine-5′-monophosphate (2l). The triethylammonium salt of the title compound was obtained as a glassy solid in 57% yield (15.3 mg) from 1 (20 mg, 49.6 μmol) and 4-methoxyphenylboronic acid according to the general method. δH (300 MHz, D2O) 4.02–4.14 (2H, m, H-5′), 4.23–4.28 (1H, m, H-4′), 4.32 (1H, t, J = 4.6, H-3′), 4.42 (1H, t, J = 5.1, H-2′), 5.98 (1H, d, J = 5.6 Hz, H-1′), 7.02, 7.44 (4H, 2d, J = 8.0 and 8.0 Hz, Ph), 7.76 (1H, s, H-6); δC (100.6 MHz, D2O) 55.9, 65.1 (d, JC,P = 4.7 Hz), 70.6, 74.1, 84.1 (d, JC,P = 8.6 Hz), 89.2, 114.6, 116.1, 124.9, 130.5, 138.2, 151.9, 159.3, 165.2; δP (121.5 MHz, D2O) 3.9. m/z (ESI) 429.0701 [M − H]−, C16H18N2O10P requires 429.0705. 5-(Furan-2-yl)-uridine-5′-monophosphate (2m). The triethylammonium salt of the title compound was obtained as a glassy solid in 57% yield (14.6 mg) from 1 (20 mg, 49.6 μmol) and 2-furanboronic acid according to the general method. δH

6366 | Org. Biomol. Chem., 2013, 11, 6357–6371

Organic & Biomolecular Chemistry (400 MHz, D2O) 4.05–4.15 (2H, m, H-5′), 4.25–4.31 (1H, m, H-4′), 4.31–4.36 (1H, m, H-3′), 4.45 (1H, t, J = 5.3 Hz, H-2′), 6.00 (1H, d, J = 5.3 Hz, H-1′), 6.52 (1H, dd, J = 1.8 and 3.4 Hz, H-fur), 6.85 (1H, d, J = 3.4 Hz, H-fur), 7.55 (1H, d, J = 1.7 Hz, H-fur), 8.15 (1H, s, H-6); δC (100.6 MHz, D2O) 65.0 (d, JC,P = 4.3 Hz), 70.5, 74.3, 84.1 (d, JC,P = 8.4 Hz), 89.4, 108.0, 109.5, 112.0, 136.0, 143.1, 145.8, 151.3, 162.8; δP (121.5 MHz, D2O) 7.6. m/z (ESI) 389.0397 [M − H]−, C13H14N2O10P requires 389.0392. 5-(5-Formylfuran-2-yl)-uridine-5′-monophosphate (2n). The triethylammonium salt of the title compound was obtained as a glassy solid in 56% yield (8.8 mg) from 1 (10 mg, 22.2 μmol) and 5-formyl-2-furanboronic acid according to the general method. δH (400 MHz, D2O) 4.00–4.15 (2H, m, H-5′), 4.24–4.26 (1H, m, H-4′), 4.30–4.38 (1H, m, H-3′), 4.49–4.53 (1H, m, H-2′), 5.97 (1H, d, J = 5.3 Hz, H-1′), 7.16 (1H, d, J = 3.7 Hz, fur), 7.58 (1H, d, J = 3.7 Hz, fur), 8.40 (1H, s, H-6) 9.47 (1H, s, CHO); δC (75.5 MHz, D2O) 64.9 (C-5′), 70.4 (C-3′), 74.0 (C-2′), 84.3 (d, JC,P = 8.5 Hz, C-4′), 90.1 (C-1′), 106.4, 112.6, 119.0, 138.7 (C-5, C-fur), 140.1 (C-6), 151.4 (C-2), 153.7 (C-fur), 172.3 (C-4), 181.1 (CHO); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 417.0336 [M − H]−, C14H14N2O11P requires 417.0341. 5-(3-Formylthien-2-yl)-uridine-5′-monophosphate (2o). The triethylammonium salt of the title compound was obtained as a glassy solid in 25% yield (3.5 mg) from 1 (10 mg, 22.2 μmol) and 3-formyl-2-thiopheneboronic acid according to the general method. δH (400 MHz, D2O) 3.98–4.06 (2H, m, H-5′), 4.24–4.27 (1H, m, H-4′), 4.27–4.32 (1H, m, H-3′), 4.40–4.45 (1H, m, H-2′), 6.00 (1H, m, H-1′), 7.55 (2H, m, Th), 8.13 (1H, s, H-6) 9.67 (1H, s, CHO); δC (150 MHz, D2O) 64.4 (C-5′), 70.4 (C-3′), 74.2 (C-2′), 84.3 (C-4′), 89.6 (C-1′), 107.6 (C-5), 127.0, 128.5, 138.9, 142.3 (C-Th), 144.7 (C-6), 151.8 (C-2), 164.7 (C-4), 188.9 (CHO); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 433.0117 [M − H]−, C14H14N2O10PS requires 433.0112. 5-(5-Acetylthien-2-yl)-uridine-5′-monophosphate (2p). The triethylammonium salt of the title compound was obtained as a glassy solid in 31% yield (4.1 mg) from 1 (10 mg, 22.2 μmol) and 5-acetyl-2-thiopheneboronic acid according to the general method. δH (400 MHz, D2O) 2.60 (3H, s, Me), 4.10–4.18 (2H, m, H-5′), 4.29–4.34 (1H, m, H-4′), 4.38 (1H, t, J = 5.7 Hz, H-3′), 4.46 (1H, t, J = 5.2 Hz, H-2′), 6.00 (1H, d, J = 5.2 Hz, H-1′), 7.60 (1H, d, J = 4.2 Hz, Th), 7.90 (1H, d, J = 4.2 Hz, Th), 8.35 (1H, s, H-6); δC (150 MHz, D2O) 26.5, 64.5, 70.6, 74.7, 84.7, 89.5, 109.7, 126.0, 136.1, 138.8, 142.4, 143.1, 151.2, 163.5, 196.7; δP (121.5 MHz, D2O) 7.6. m/z (ESI) 447.0263 [M − H]−, C15H16N2O10PS requires 447.0269. 5-(5-Formylthien-2-yl)-uridine-5′-monophosphate (2q). The triethylammonium salt of the title compound was obtained as a glassy solid in 61% yield (8.6 mg) from 1 (10 mg, 22.2 μmol) and 5-formyl-2-thiopheneboronic acid according to the general method. δH (400 MHz, D2O) 4.12–4.20 (2H, m, H-5′), 4.29–4.31 (1H, m, H-4′), 4.38 (1H, t, J = 4.7 Hz, H-3′), 4.45 (1H, t, J = 5.0 Hz, H-2′), 5.98 (1H, d, J = 5.0 Hz, H-1′), 7.66 (1H, d, J = 4.0 Hz, Th), 7.94 (1H, d, J = 4.0 Hz, Th), 8.38 (1H, s, H-6) 9.75 (1H, s, CHO); δC (75.5 MHz, D2O) 64.7 (d, JC,P = 4.5 Hz, C-5′), 70.4 (C-3′), 74.9 (C-2′), 84.4 (C-4′), 89.7 (C-1′), 110.0, 125.9, 139.1, 140.1, 142.0 (C5 + C-Th), 144.9 (C-6), 152.7 (C-2), 163.5

This journal is © The Royal Society of Chemistry 2013

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Organic & Biomolecular Chemistry (C-4), 187.8 (CHO); δP (121.5 MHz, D2O) 7.6. m/z (ESI) 433.0107 [M − H]−, C14H14N2O10PS requires 433.0112. 5-[(5-Formaldoximyl)-thien-2-yl]-uridine-5′-monophosphate (2r). To an aqueous solution (1 mL) of 2q (7 mg, 13 μmol) and hydroxylamine hydrochloride (1.1 mg, 16 μmol), a solution of sodium bicarbonate (1.4 mg, 0.16 mmol) in water (1 mL) was added dropwise over 1 minute with stirring. The reaction was stirred at room temperature for 3 h, and the solvent was removed in vacuo. The residue was purified by ion-pair chromatography on a Lichroprep RP-18 resin, using a gradient of aqueous TEAB (0.05 M, pH 7.3) against 0–25% methanol over a total volume of 400 mL (flow rate: 2 mL min−1). Anti-2r eluted first, followed closely by syn-2r. The respective productcontaining fractions were combined and freeze-dried at least twice to remove excess TEAB. Thus, each isomer was obtained as a yellow residue in yields of, respectively, 19% (1.5 mg, 2 μmol, anti-2r) and 21% (3.4 mg, 3 μmol, syn-2r). Anti-2r: δH (400 MHz, D2O) 3.98 (2H, t, J = 4.6 Hz, H-5′); 4.20–4.23 (1H, m, H-4′), 4.33 (1H, dd, J = 4.4 and 5.3 Hz, H-3′), 4.46 (1H, t, J = 5.7 Hz, H-2′), 6.04 (1H, d, J = 5.8 Hz, H-1′), 7.52 (1H, d, J = 4.1 Hz, Th), 7.54 (1H, d, J = 4.1 Hz, Th), 7.85 (1H, s, H-oxime), 8.17 (1H, s, H-6); δP (162 MHz, D2O) 0.45. Syn-2r: δH (400 MHz, D2O) 3.98 (2H, t, J = 4.6 Hz, H-5′); 4.20–4.23 (1H, m, H-4′), 4.33 (1H, dd, J = 4.4 and 5.3 Hz, H-3′), 4.46 (1H, t, J = 5.7 Hz, H-2′), 6.03 (1H, d, J = 5.7 Hz, H-1′), 7.31 (1H, d, J = 4.0 Hz, Th), 7.49 (1H, d, J = 4.0 Hz, Th), 8.17 (1H, s, H-6), 8.35 (1H, s, H-oxime); δP (162 MHz, D2O) 0.45. Anti-2r/syn-2r: δC (100 MHz, D2O) 59.0 (CH2), 63.7 (CH2), 70.2 (CH), 73.5 (CH), 83.6 (CH), 83.7 (CH), 88.9 (CH), 110.3, 118.4, 123.3, 130.9, 133.4, 136.8, 142.6, 147.2, 161.2, 163.0, 163.5, 186.3, 190.3, 198.4. m/z (ESI) 448.0222 [M − H]−, C14H15N3O10PS requires 448.0221. 5-Iodouridine-5′-diphosphate (3).23 1 (292 mg, 0.65 mmol) was converted into the corresponding phosphoromorpholidate as previously described,11 affording 354 mg of 5-iodo UMP phosphoromorpholidate as a colourless powder (99% yield). δH (400 MHz, D2O) 3.04–3.16 (4H, m, morph), 3.63–3.73 (4H; m, morph), 3.98–4.15 (2H, m, H-5′), 4.24–4.26 (1H, m, H-4′), 4.27–4.32 (1H, m, H-3′), 4.37 (1H, t, J = 5.3 Hz, H-2′), 5.94 (1H, d, J = 5.3 Hz, H-1′), 8.18 (1H, s, H-6); δC (75.5 MHz, D2O) 44.2 (morph), 63.4 (d, JC,P = 5.3 Hz, C-5′), 66.3 (morph), 68.0 (C-5), 68.6 (C-3′), 73.2 (C-2′), 83.0 (d, JC,P = 8.5 Hz, C-4′), 88.2 (C-1′), 145.0 (C-6), 153.0 (C-2), 161.9 (C-4); δP (121.5 MHz, D2O) 11.0. m/z (ESI) 519.9968 [M + H]+, C13H19IN3O9P requires 519.9976. 5-Iodo UMP phosphoromorpholidate (50 mg, 96 µmol, sodium salt form) was repeatedly (3×) co-evaporated with pyridine. KH2PO4 (34 mg, 191 µmol) and tributylamine (92 µL, 382 mmol) were added, and the mixture was repeatedly (3×) co-evaporated with pyridine to remove excess tributylamine. Tetrazole (33 mg, 470 µmol) and dry DMF (5 mL) were added to the dry mixture. The reaction was left stirring for 5 h at room temperature. The crude mixture was concentrated under reduced pressure and isolated using purification method 1. The triethylammonium salt of the title compound (1.0 equiv.) was obtained as a glassy solid in 66% yield (40.2 mg). δH (400 MHz, D2O) 4.00–4.05 (2H, m, H-5′), 4.10–4.15 (1H, m, H-4′), 4.20–4.25 (2H, m, H-3′,H-2′), 5.77 (1H,

This journal is © The Royal Society of Chemistry 2013

Paper d, J = 4.6 Hz, H-1′), 8.08 (1H, s, H-6); δC (75.5 MHz, D2O) 65.5 (C-5′), 69.2 (C-5), 70.3 (C-3′), 74.4 (s, C-2′), 84.1 (d, JC,P = 8.5 Hz, C-4′), 89.3 (C-1′), 146.7 (C-6), 152.4 (C-2), 163.9 (C-4); δp (121.5 MHz, D2O) −6.6 (d, JP,P = 23.1 Hz), −11.2 (d, JP,P = 23.1 Hz). m/z (ESI) 528.8912 [M − H]−, C9H12IN2O12P2 requires 528.8916. 5-Iodouridine-5′-triphosphate (4).24 5-Iodouridine-5′-monophosphomorpholidate (81 mg, 0.156 mmol) was prepared as described for 3 and repeatedly (3×) co-evaporated with pyridine. To the white solid was added potassium pyrophosphate (280 mg, 0.628 mmol) and tributylamine (303 µL, 1.26 mmol) and the mixture was further co-evaporated (3×) with pyridine. Tetrazole (55 mg, 0.785 mmol) and dry DMF (5 mL) were added to the dry mixture and the reaction was left stirring for 5 h at room temperature. The crude mixture was concentrated under reduced pressure and isolated using purification method 1. The triethylammonium salt of the title compound (3.2 equiv.) was obtained as a glassy solid in 60% yield (86.8 mg). δH (400 MHz, D2O) 4.20–4.26 (2H, m, H-5′), 4.26–4.30 (1H, m, H-4′), 4.37–4.44 (2H, m, H-3′, H-2′), 5.94 (1H, d, J = 4.9 Hz, H-1′), 8.27 (1H, s, H-6); δC (75.5 MHz, D2O) 65.7 (d, JC,P = 6.1 Hz, C-5′), 69.2 (C-5), 70.4 (C-3′), 74.4 (s, C-2′), 84.3 (d, J = 9.1 Hz, C-4′), 89.1 (C-1′), 146.7 (C-6), 152.4 (C-2), 164.0 (C-4); δp (121.5 MHz, D2O) −6.5 (d, JP,P = 20.7 Hz), −11.6 (d, JP,P = 19.4 Hz), −22.6 (d, JP,P = 20.7 Hz). m/z (ESI) 608.8588 [M − H]−, C9H13IN2O15P3 requires 608.8579. 5-Phenyluridine-5′-diphosphate (5a). The triethylammonium salt of the title compound (1.9 equiv.) was obtained as a glassy solid in 73% yield (10.6 mg) from 3 (11.6 mg, 20.8 μmol) and phenylboronic acid according to the general method. δH (400 MHz, D2O) 4.15–4.19 (2H, m, H-5′), 4.27–4.31 (1H, m, H-4′), 4.38–4.43 (1H, m, H-3′), 4.47 (1H, t, J = 5.6 Hz, H-2′), 6.05 (1H, d, J = 5.8 Hz, H-1′), 7.40–7.58 (5H, m, phenyl), 7.88 (1H, s, H-6); δC (75.4 MHz, D2O) 65.7 (d, JC,P = 3.7 Hz, C-5′), 70.5 (C-3′), 73.9 (C-2′), 84.1 (d, JC,P = 5.0 Hz, C-4′), 89.1 (C-1′), 118.9 (C-5), 129.1, 129.4, 129.5, 133.8 (C-Ph), 139.2 (C-6), 152.3 (C-2), 165.5 (C-4); δP (121 MHz, D2O) −6.6 (d, JP,P = 23.1 Hz), −11.1 (d, JP,P = 23.1 Hz). m/z (ESI) 239.0096 [M − 2H]2−, C15H16N2O12P2 requires 239.0095. 5-(4-Methoxyphenyl)uridine-5′-diphosphate (5b). The triethylammonium salt of the title compound (1.9 equiv.) was obtained as a glassy solid in 50% yield (7.2 mg) from 3 (11.0 mg, 20.8 μmol) and 4-methoxyphenylboronic acid according to the general method. δH (400 MHz, D2O) 3.87 (3H, s, MeO), 4.16–4.22 (2H, m, H-5′), 4.26–4.31 (1H, m, H-4′), 4.41 (1H, m, H-3′), 4.47 (1H, t, J = 5.6 Hz, H-2′), 6.04 (1H, d, J = 5.8 Hz, H-1′), 7.07 (2H, d, J = 8.8 Hz, Ph), 7.49 (2H, d, J = 8.8 Hz, Ph), 7.83 (1H, s, H-6); δC (125 MHz, D2O) 55.9 (MeO), 65.7 (d, JC,P = 3.0 Hz, C-5′), 70.5 (C-3′), 73.9 (C-2′), 82.6 (C-4′), 89.0 (C-1′), 114.9, 116.5, 125.1, 130.8 (C-5 + C-Ph), 138.5 (C-6), 152.3 (C-2), 159.6 (C-Ph), 166.1 (C-4); δP (121 MHz, D2O) −6.6 (d, JP,P = 23.1 Hz), −11.1 (d, JP,P = 23.1 Hz). m/z (ESI) 254.0149 [M − 2H]2−, C16H18N2O13P2 requires 254.0148. 5-(Furan-2-yl)uridine-5′-diphosphate (5c). The triethylammonium salt of the title compound (1.9 equiv.) was obtained as a glassy solid in 60% yield (8.6 mg) from 3 (11.0 mg,

Org. Biomol. Chem., 2013, 11, 6357–6371 | 6367

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Paper 20.8 μmol) and 2-furanboronic acid according to the general method. δH (400 MHz, D2O) 4.20–4.26 (2H, m, H-5′), 4.29–4.33 (1H, m, H-4′), 4.44 (1H, t, J = 5.4 Hz, H-3′), 4.49 (1H, t, J = 5.4 Hz, H-2′), 6.04 (1H, d, J = 5.5 Hz, H-1′), 6.54 (1H, dd, J = 1.8 and 3.4 Hz, fur), 6.89 (1H, d, J = 3.3 Hz, fur), 7.59 (1H, d, J = 1.1 Hz, fur), 8.21 (1H, s, H-6); δC (75.5 MHz, D2O) 65.9 (C-5′), 70.7 (C-3′), 74.5 (C-2′), 84.4 (C-4′), 89.5 (C-1′), 108.4 (C-5), 109.9 (fur4), 112.4 (fur3), 136.6 (C-6), 143.7 (fur2), 146.3 (fur1), 152.0 (C-2), 163.6 (C-4); δP (121 MHz, D2O) −6.5 (d, JP,P = 23.1 Hz), −11.0 (d, JP,P = 23.1 Hz). m/z (ESI) 233.9991 [M − 2H]2−, C13H14N2O13P2 requires 233.9991. 5-(5-Formylthien-2-yl)uridine-5′-diphosphate (5d). The triethylammonium salt of the title compound (1.6 equiv.) was obtained as a glassy solid in 77% yield (8.2 mg) from 3 (14.0 mg, 26.4 μmol) and 5-formyl-2-thiopheneboronic acid according to the general method. δH (400 MHz, D2O) 4.26–4.32 (2H, m, H-5′), 4.32–4.35 (1H, m, H-4′), 4.43–4.50 (2H, m, H-2′ and H-3′), 6.01–6.07 (1H, m, H-1′), 7.71–7.75 (1H, m, Th), 7.99–8.01 (1H, m, Th), 8.45 (1H, s, H-6); 9.78 (1H, s, CHO); δC (75.5 MHz, D2O) 65.5 (C-5′), 70.2 (C-3′), 74.9 (C-2′), 84.2 (C-4′), 89.7 (C-1′), 109.3 (C-5), 125.8, 138.9, 140.3, 142.0, 144.8 (C-6 + C-Th), 151.1 (C-2), 163.3 (C-4), 187.8 (CHO); δP (121 MHz, D2O) −6.5 (d, JP,P = 23.1 Hz), −11.0 (d, JP,P = 23.1 Hz). m/z (ESI) 512.9776 [M − H]−, C14H15N2O13P2S1 requires 512.9776. 5-Phenyluridine-5′-triphosphate (6a). The triethylammonium salt of the title compound (3.2 equiv.) was obtained as a glassy solid in 85% yield (34.2 mg) from 4 (20 mg, 33.8 μmol) and phenylboronic acid according to the general method. δH (400 MHz, D2O) 4.17–4.27 (2H, m, H-5′), 4.29–4.31 (1H, m, H-4′), 4.43–4.52 (2H, m, H-2′ and H-3′), 6.06 (1H, d, J = 5.5 Hz, H-1′), 7.40–7.57 (5H, m, Ph), 7.92 (1H, s, H-6); δC (125 MHz, D2O) 66.0 (C-5′), 70.7 (C-3′), 74.1 (C-2′), 84.3 (C-4′), 88.9 (C-1′), 117.0 (C-5), 129.2, 129.5, 129.6, 132.4, (C-Ph), 139.2 (C-6), 152.4 (C-2), 165.5 (C-4) ; δP (121 MHz, D2O) −6.5 (d, JP,P = 20.7 Hz), −11.1 (d, JP,P = 19.7 Hz), −22.6 (d, JP,P = 20.7 Hz). m/z (ESI) 558.9935 [M − H]−, C15H18N2O15P3 requires 558.9926. 5-(4-Methoxyphenyl)uridine-5′-triphosphate (6b). The triethylammonium salt of the title compound (3.0 equiv.) was obtained as a glassy solid in 53% yield (12.1 mg) from 4 (10.9 mg, 17.9 μmol) and 4-methoxyphenylboronic acid according to the general method. δH (400 MHz, D2O) 3.87 (3H, s, MeO), 4.16–4.28 (2H, m, H-5′), 4.29–4.32 (1H, m, H-4′), 4.44 (1H, t, J = 5.4 Hz, H-3′), 4.49 (1H, t, J = 5.4, H-2′), 6.06 (1H, d, J = 6.0 Hz, H-1′), 7.07 (2H, d, J = 8.9 Hz, Ph), 7.50 (2H, d, J = 8.8 Hz, Ph), 7.86 (1H, s, H-6); δC (75.5 MHz, D2O) 55.9 (MeO), 65.9 (C-5′), 70.6 (C-3′), 73.9 (C-2′), 84.2 (d, JC,P = 8.8 Hz, C-4′), 88.7 (C-1′), 114.9, 116.6, 125.1, 130.8, 138.4, 152.3, 159.6, 165.6 (C-2, C-4, C-6, C-5, C-Ph); δP (121 MHz, D2O) −6.5 (d, JP,P = 20.7 Hz), −11.5 (d, JP,P = 19.7 Hz), −22.7 (d, JP,P = 20.1 Hz). m/z (ESI) 589.0040 [M − H]−, C16H20N2O16P3 requires 589.0031. 5-(Furan-2-yl)uridine-5′-triphosphate (6c). The triethylammonium salt of the title compound (3.0 equiv.) was obtained as a glassy solid in 74% yield (11.4 mg) from 4 (10.9 mg, 17.9 μmol) and 2-furanboronic acid according to general

6368 | Org. Biomol. Chem., 2013, 11, 6357–6371

Organic & Biomolecular Chemistry method A. δH (400 MHz, D2O) 4.23–4.30 (2H, m, H-5′), 4.31–4.33 (1H, m, H-4′), 4.45–4.52 (2H, m, H-2′ and H-3′), 6.06 (1H, d, J = 5.2 Hz, H-1′), 6.54 (1H, d, J = 1.5 Hz, fur), 6.89 (1H, d, J = 3.1 Hz, fur), 7.59 (1H, s, fur), 8.23 (1H, s, H-6); δC (125 MHz, D2O) 65.9 (d, JC,P = 4.3 Hz, C-5′), 70.4 (C-3′), 74.2 (C-2′), 84.4 (d, JC,P = 7.6 Hz, C-4′), 89.0 (C-1′), 108.2 (C-5), 109.6 (fur4), 112.1 (fur3), 136.4 (C-6), 143.5 (fur2), 146.0 (fur1), 151.7 (C-2), 163.3 (C-4); δP (121 MHz, D2O) −6.5 (d, JP,P = 20.7 Hz), −11.5 (d, JP,P = 19.7 Hz), −22.6 (d, JP,P = 20.1 Hz). m/z (ESI) 548.9728 [M − H]−, C13H16N2O16P3 requires 548.9718. 5-(5-Formylthien-2-yl)uridine-5′-triphosphate (6d). The triethylammonium salt of the title compound (3.4 equiv.) was obtained as a glassy solid in 67% yield (16.9 mg) from 4 (16 mg, 26.3 μmol) and 5-formyl-2-thiopheneboronic acid according to the general method. δH (400 MHz, D2O) 4.30–4.44 (3H, m, H-5′, H-4′), 4.44–4.51 (2H, m, H-2′ and H-3′), 6.04 (1H, s, H-1′), 7.73 (1H, s, Th), 8.01 (1H, s, Th), 8.45 (1H, s, H-6); 9.78 (1H, s, CHO); δC (75.5 MHz, D2O) 65.9 (C-5′), 70.4 (C-3′), 75.0 (C-2′), 84.5 (C-4′), 89.5 (C-1′), 101.5, 109.7, 126.1, 139.2, 140.6, 142.2, 145.0 (C-Th + C6 + C-5 + C-2), 163.6 (C-4), 188.0 (CHO); δP (121 MHz, D2O) −6.5 (d, JP,P = 20.7 Hz), −11.5 (d, JP,P = 19.7 Hz), −22.6 (d, JP,P = 20.1 Hz). m/z (ESI) 592.9448 [M − H]−, C14H16N2O16P3 requires 592.9439. 5-(5-Formylthien-2-yl)-uridine (7). The title compound was obtained from 5-iodouridine (100 mg, 270 μmol)22 and 5-formyl-2-thiopheneboronic acid according to the general method in 23% yield (16 mg). δH (400 MHz, DMSO-d6) 3.65–3.85 (2H, m, H-5′), 3.90–3.95 (1H, m, H-4′), 4.02–4.17 (2H, m, H-2′ and H-3′), 5.11 (1H, d, J = 5.9 Hz, OH-3′), 5.54 (1H, d, J = 5.0 Hz, OH-2′), 5.60 (1H, t, J = 4.3 Hz, OH-5′), 5.80 (1H, d, J = 3.0 Hz, H-1′), 7.56 (1H, d, J = 4.1 Hz, Th), 7.93 (1H, d, J = 4.0 Hz, Th), 9.02 (1H, s, H-6), 9.87 (1H, s, CHO), 11.9 (1H, s, NH); δC (75.5 MHz, DMSO-d6) 59.6 (C-5′), 68.7 (C-3′), 74.5 (C-2′), 84.4 (C-4′), 89.5 (C-1′), 106.9 (C-5), 122.9, 137.3, 138.7, 141.6, 144.4, 149.4 (C-2 + C-6 + C-Th), 161.4 (C-4), 184.3 (CHO). m/z (ESI) 353.0448 [M − H]−, C14H13N2O7S requires 353.0449. Photophysical characterisation Absorbance and fluorescence spectra. UV absorbance spectra were recorded in H2O on a PerkinElmer Lambda 25 UV-Vis spectrometer at ambient temperature in FarUV quartz cells ( path length 1.0 cm). Fluorescence spectra were recorded in H2O on a PerkinElmer LS-45 spectrometer at ambient temperature in a quartz micro fluorescence cell ( path length 1.0 cm). Quantum yields. Nucleotide derivatives were serially diluted in H2O (10, 20, 30, 40 and 50 μM for absorbance measurements, and 0.2, 0.4, 0.6, 0.8, 1 μM for fluorescence measurements), and UV absorbance and fluorescence emission (with λmax absorbance = λem fluorescence) were recorded for all samples. To determine quantum yields, for each absorbance and fluorescence spectrum the area under the curve (AUC) was calculated by numerical integration, applying the mid-point rule. For each compound, AUCabs and AUCem were then plotted over compound concentration according to AUCabs =

This journal is © The Royal Society of Chemistry 2013

View Article Online

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

Organic & Biomolecular Chemistry A × [conc] + B and AUCem = A′ × [conc] + B′. From these linear plots, the gradients A and A′ were extracted, and for each compound the specific quantum yield Φs, under these experimental conditions, was calculated as the ratio A′/A. Quantum yields determined with this protocol for two reference compounds, 2-aminopyridine (Φs 0.60) and L-tryptophan (Φs 0.14) were in agreement with literature values.25 The quantum yields for reference compounds were used to calculate the general quantum yield Φg for each nucleotide analogue, according to Φg = Φref × (A′/A)/(A′/A)ref. Solvatochromism. Fluorescence intensity measurements with 7 in different solvents were performed in NUNC F96 MicroWell polystyrene plates on a BMG Labtech PolarStar plate reader equipped with a 350 ± 5 nm absorbance filter and with a 430 ± 5 nm emission filter. Solutions of 7 (10 μM) were prepared in water (HPLC grade), acetonitrile and isopropanol. Sample assays with various solvent mixtures were incubated for 2 minutes at 30 °C and the fluorescence emission was recorded. Data were analysed with GraFit (version 5.0.10). Computational experiments Calculations were carried out with the Gaussian09 suite of programs26 on the monoanionic form of the molecules. The geometry minimizations were carried out using the DFT/B3LYP method27 and a 6-31+G(d,p) basis set. The solvation effects were included using the Polarizable Continuum Model28 with the standard dielectric constant of 78.3553 for water. The absorbance spectrum for each monoanion was calculated using the TDDFT method with the same functional and basis aforementioned. The theoretical absorbance spectra were renormalized to match the largest peak of their experimental counterparts. Ligand displacement experiments Bovine α-1,3-GalT was expressed and purified as previously reported.5a Individual wells on a black polystyrene NUNC F96 MicroWell plate were incubated with Tris–HCl buffer (“buffer B”, 50 mM, pH 7, 40 µL), MnCl2 (10 mM in buffer B, 80 µL), the respective ligand (7, 2q, 5d or 8, 200 nM in buffer B, 40 µL of ) and α-1,3-GalT (at decreasing concentrations in buffer B, 40 µL). Microplates were incubated for 10 minutes at 30 °C. Fluorescence emission was measured in triplicate with a BMG labtech PolarStar microplate reader equipped with a 350 ± 5 nm absorbance filter and a 430 nm ± 5 nm emission filter. The number of flashes were set at 50 per well, gain at 2240 and position delay at 0.2 s. Shaking of the plate following a double orbital with a 4 mm shaking width was performed for 10 seconds prior to reading.

Acknowledgements We thank the EPSRC (First Grant EP/D059186/1) and the MRC (Grant no. 0901746), for funding and the EPSRC National Mass Spectrometry Facility, Swansea, for the recording of mass spectra. The computational resources for the G09 calculations

This journal is © The Royal Society of Chemistry 2013

Paper were provided by the Imperial College High Performance Computing Service. We are grateful to Prof. Monica Palcic (Copenhagen) for making the α-1,3-GalT clone available, and for help with enzyme expression.

References 1 For recent reviews see: (a) R. W. Sinkeldam, N. J. Greco and Y. Tor, Fluorescent Analogs of Biomolecular Building Blocks: Design, Properties, and Applications, Chem. Rev., 2010, 110, 2579–2619; (b) K. Phelps, A. Morris and P. A. Beal, Novel Modifications in RNA, ACS Chem. Biol., 2012, 7, 100–109. 2 (a) Y. Xie, T. Maxson and Y. Tor, Fluorescent nucleoside analogue displays enhanced emission upon pairing with guanine, Org. Biomol. Chem., 2010, 8, 5053–5055; (b) N. J. Greco, R. W. Sinkeldam and Y. Tor, An emissive C analog distinguishes between G, 8-oxoG and T, Org. Lett., 2009, 11, 1115–1118; (c) Y. Xie, A. V. Dix and Y. Tor, FRET Enabled Real Time Detection of RNA-Small Molecule Binding, J. Am. Chem. Soc., 2009, 131, 17605– 17614. 3 W. Hirose, K. Sato and A. Matsuda, Selective Detection of 5-Formyl-2′-deoxyuridine, an Oxidative Lesion of Thymidine, in DNA by a Fluorogenic Reagent, Angew. Chem., Int. Ed., 2010, 49, 8392–8394. 4 (a) G. Pergolizzi, J. Butt, R. Bowater and G. K. Wagner, A novel fluorophore for NAD-consuming enzymes, Chem. Commun., 2011, 47, 12655–12657; (b) J. R. Barrio, J. A. Secrist and N. J. Leonard, A Fluorescent Analog of Nicotinamide Adenine Dinucleotide, Proc. Natl. Acad. Sci. U. S. A., 1972, 69, 2039–2042; (c) B. A. Gruber and J. L. Nelson, Dynamic and static quenching of 1,N6-ethenoadenine fluorescence in nicotinamide 1,N6-ethenoadenine dinucleotide and in 1,N6-etheno-9-(3-(indol-3-yl) propyl) adenine, Proc. Natl. Acad. Sci. U. S. A., 1975, 72, 3966–3969; (d) N. J. Leonard and J. R. Barrio, EthenoSubstituted Nucleotides and Coenzymes: Fluorescence and Biological Activity, Crit. Rev. Biochem. Mol. Biol., 1984, 15, 125–199. 5 (a) T. Pesnot, M. M. Palcic and G. K. Wagner, A novel fluorescent probe for retaining galactosyltransferases, ChemBioChem, 2010, 11, 1392–1398; (b) B. J. Gross, B. C. Kraybill and S. Walker, Discovery of O-GlcNAc Transferase Inhibitors, J. Am. Chem. Soc., 2005, 127, 14588– 14589; (c) J. S. Helm, Y. N. Hu, L. Chen, B. Gross and S. Walker, Identification of active-site inhibitors of MurG using a generalizable, high-throughput glycosyltransferase screen, J. Am. Chem. Soc., 2003, 125, 11168–11169. 6 (a) D. Shin, R. W. Sinkeldam and Y. Tor, Emissive RNA Alphabet, J. Am. Chem. Soc., 2011, 133, 14912–14915; (b) S. G. Srivatsan, N. J. Greco and Y. Tor, A Highly Emissive Fluorescent Nucleoside that Signals the Activity of Toxic Ribosome-Inactivating Proteins, Angew. Chem., Int. Ed., 2008, 47, 6661–6665; (c) R. W. Sinkeldam, N. J. Greco and

Org. Biomol. Chem., 2013, 11, 6357–6371 | 6369

View Article Online

Paper

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

7

8

9

10

11

12

13

Y. Tor, Polarity of major grooves explored by using an isosteric emissive nucleoside, ChemBioChem, 2008, 9, 706–709; (d) Y. Tor, S. Del Valle, D. Jaramillo, S. G. Srivatsan, A. Rios and H. Weizman, Designing new isomorphic fluorescent nucleobase analogues: the thieno[3,2-d]pyrimidine core, Tetrahedron, 2007, 63, 3608–3614. (a) H. Jiang, J. H. Kim, K. M. Frizzell, W. L. Kraus and H. Lin, Clickable NAD Analogues for Labeling Substrate Proteins of Poly(ADP-ribose) Polymerases, J. Am. Chem. Soc., 2010, 132, 9363–9372; (b) D. Topalis, H. Kumamoto, M.-F. Amaya Velasco, L. Dugue, A. Haouz, J. A. C. Alexandre, S. Gallois-Montbrun, P. M. Alzari, S. Pochet, L. A. Agrofoglio and D. Deville-Bonne, Nucleotide binding to human UMP-CMP kinase using fluorescent derivatives – a screening based on affinity for the UMP-CMP binding site, FEBS J., 2007, 274, 3704–3714; (c) J. Bhat, R. Rane, S. M. Solapure, D. Sarkar, U. Sharma, M. N. Harish, S. Lamb, D. Plant, P. Alcock, S. Peters, S. Barde and R. K. Roy, High-Throughput Screening of RNA Polymerase Inhibitors Using a Fluorescent UTP Analog, J. Biomol. Screen., 2006, 11, 968–976. A. Collier and G. K. Wagner, A fast synthetic route to GDPsugars modified at the nucleobase, Chem. Commun., 2008, 178–180. (a) M. Segal and B. Fischer, Analogues of uracil nucleosides with intrinsic fluorescence (NIF-analogues): synthesis and photophysical properties, Org. Biomol. Chem., 2012, 10, 1571–1580; (b) T. Pesnot and G. K. Wagner, Novel derivatives of UDP-glucose: Concise synthesis and fluorescent properties, Org. Biomol. Chem., 2008, 6, 2884–2891; (c) R. W. Sinkeldam, A. J. Wheat, H. Boyaci and Y. Tor, Emissive Nucleosides as Molecular Rotors, ChemPhysChem, 2011, 12, 567–570; (d) N. J. Greco and Y. Tor, Furan decorated nucleoside analogues as fluorescent probes: synthesis, photophysical evaluation, and site-specific incorporation, Tetrahedron, 2007, 63, 3515–3527; (e) N. J. Greco and Y. Tor, Simple Fluorescent Pyrimidine Analogues Detect the Presence of DNA Abasic Sites, J. Am. Chem. Soc., 2005, 127, 10784–10785. (a) N. Amann and H. A. Wagenknecht, Preparation of pyrenyl-modified nucleosides via Suzuki-Miyaura cross-coupling reactions, Synlett, 2002, 687–691; (b) L. H. Thoresen, G. S. Jiao, W. C. Haaland, M. L. Metzker and K. Burgess, Rigid, conjugated, fluoresceinated thymidine triphosphates: Syntheses and polymerase mediated incorporation into DNA analogues, Chem.–Eur. J., 2003, 9, 4603–4610. K. Descroix, T. Pesnot, Y. Yoshimura, S. S. Gehrke, W. Wakarchuk, M. M. Palcic and G. K. Wagner, Inhibition of galactosyltransferases by a novel class of donor analogues, J. Med. Chem., 2012, 55, 2015–2024. M. Yoshikawa, T. Kato and T. Takenishi, A novel method for phosphorylation of nucleosides to 5′-nucleotides, Tetrahedron Lett., 1967, 50, 5065–5068. T. Pesnot, R. Jørgensen, M. M. Palcic and G. K. Wagner, Structural and mechanistic basis for a new mode of

6370 | Org. Biomol. Chem., 2013, 11, 6357–6371

Organic & Biomolecular Chemistry

14

15

16

17

18 19

20

21

22

23

24

25

26

glycosyltransferase inhibition, Nat. Chem. Biol., 2010, 6, 321–323. T. Mukaiyama and M. Hashimoto, Phosphorylation by Oxidation-Reduction Condensation. Preparation of Active Phosphorylating Reagents, Bull. Chem. Soc. Jpn., 1971, 44, 2284. S. P. Nighswander-Rempel, Quantum yield calculations for strongly absorbing chromophores, J. Fluoresc., 2006, 16, 483–485. (a) C. Altona and M. Sundaralingam, Conformational analysis of the sugar ring in nucleosides and nucleotides. Improved method for the interpretation of proton magnetic resonance coupling constants, J. Am. Chem. Soc., 1973, 95, 2333–2344; (b) B. Ancian, NMR Studies for Mapping Structure and Dynamics of Nucleosides in Water, in: Annual Reports on NMR Spectroscopy (Chapter 2), volume 69, 2010, pp. 39–143. T. Pesnot, D. L. Hughes and G. K. Wagner, 5-Phenyluridine· 3H2O, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2008, 64, o44–o46. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer, New York, 3rd edn, 2006. K. Rotkiewicz, K. H. Grellmann and Z. R. Grabowski, Reinterpretation of the anomalous fluorescence of paraNN-dimethylaminobenzonitrile, Chem. Phys. Lett., 1973, 19, 315–318. F. Daligault, S. Rahuel-Clermont, S. Gulberti, M.-T. Cung, G. Branlant, P. Netter, J. Magdalou and V. Lattard, Thermodynamic insights into the structural basis governing the donor substrate recognition by human β-1,4-galactosyltransferase 7, Biochem. J., 2009, 418, 605–614. R. O. Rahn and H. G. Sellin, Action spectra for the photolysis of 5-iododeoxyuridine in DNA and related model systems: evidence for short-range energy transfer, Photochem. Photobiol., 1982, 35, 459–465. K. Shah, H. Wu and T. M. Rana, Synthesis of Uridine Phosphoramidite Analogs: Reagents for Site-Specific Incorporation of Photoreactive Sites into RNA Sequences, Bioconjugate Chem., 1994, 5, 508–512. P. Besada, D. H. Shin, S. Costanzi, H. Ko, C. Mathe, J. Gagneron, G. Gosselin, S. Maddileti, T. Kendall Harden and K. A. Jacobson, Structure–Activity Relationships of Uridine 5′-Diphosphate Analogues at the Human P2Y6 Receptor, J. Med. Chem., 2006, 49, 5532–5543. D. Lipkin, F. B. Howard, D. Nowotny and M. Sano, The Iodination of Nucleosides and Nucleotides, J. Biol. Chem., 1963, 238, 2249–2251. (a) R. Rusakowicz and A. C. Testa, 2-Aminopyridine as a standard for low-wavelength spectrofluorimetry, J. Phys. Chem., 1968, 72, 2680–2681; (b) E. P. Kirby and R. F. Steiner, Influence of solvent and temperature upon the fluorescence of indole derivatives, J. Phys. Chem., 1970, 74, 4480–4490. M. J. Frisch, G. W. Trucks, H. B. Schlegel, et al., GAUSSIAN 09 (Revision C.1), Gaussian, Inc., Wallingford, CT, 2009.

This journal is © The Royal Society of Chemistry 2013

View Article Online

Organic & Biomolecular Chemistry

28 J. Tomasi, B. Mennucci and R. Cammi, Quantum mechanical continuum solvation models, Chem. Rev., 2005, 105, 2999–3093. 29 E. Nir, K. Kleinermanns, L. Grace and M. S. deVries, On the Photochemistry of Purine Nucleobases, J. Phys. Chem. A, 2001, 105, 5106–5110.

Published on 14 August 2013. Downloaded by Old Dominion University on 29/08/2013 11:12:13.

27 (a) C. Lee, W. Yang and R. G. Parr, Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density, Phys. Rev. B: Condens. Matter, 1988, 37, 785–789; (b) D. A. Becke, Density-functional thermochemistry. III. The role of exact exchange, J. Chem. Phys., 1993, 98, 5648–6352.

Paper

This journal is © The Royal Society of Chemistry 2013

Org. Biomol. Chem., 2013, 11, 6357–6371 | 6371