Base-modified NAD and AMP derivatives and their activity against bacterial DNA ligases

Organic &

Biomolecular Chemistry

View Article Online

View Journal

PAPER

Base-modiﬁed NAD and AMP derivatives and their

activity against bacterial DNA ligases†

Cite this: DOI: 10.1039/c5ob00294j

Giulia Pergolizzi,^a,bMarco M. D. Cominetti,^aJulea N. Butt,^c,dRobert A. Field,^b

Richard P. Bowater^dand Gerd K. Wagner*^a,e

We report the chemical synthesis and conformational analysis of a collection of 2-, 6- and 8-substituted

derivatives of β-NAD⁺and AMP, and their biochemical evaluation against NAD⁺-dependent DNA ligases

from Escherichia coli and Mycobacterium tuberculosis. Bacterial DNA ligases are validated anti-microbial

targets, and new strategies for their inhibition are therefore of considerable scientiﬁc and practical inter-

est. Our study includes several pairs of β-NAD⁺and AMP derivatives with the same substitution pattern at

the adenine base. This has enabled the ﬁrst direct comparison of co-substrate and inhibitor behaviour

against bacterial DNA ligases. Our results suggest that an additional substituent in position 6 or 8 of the

adenine base in β-NAD⁺is detrimental for activity as either co-substrate or inhibitor. In contrast, substitu-

ents in position 2 are not only tolerated, but appear to give rise to a new mode of inhibition, which targets

the conformational changes these DNA ligases undergo during catalysis. Using a molecular modelling

approach, we highlight that these ﬁndings have important implications for our understanding of ligase

mechanism and inhibition, and may provide a promising starting point for the rational design of a new

class of inhibitors against NAD⁺-dependent DNA ligases.

Received 10th February 2015,

Accepted 7th May 2015

DOI: 10.1039/c5ob00294j

www.rsc.org/obc

been identified in any eukaryotic organisms to date.^1,2Because

of their essentiality for bacterial survival and their absence

Introduction

DNA ligases are essential enzymes found in all living cells due from the human host, NAD⁺-dependent DNA ligases represent

to their involvement in DNA replication, repair and recombina- an attractive target for the development of novel antibacterial

tion.¹The biochemical activity of DNA ligases consists of the drugs,⁵and several pertinent inhibitor discovery eﬀorts have

2−

sealing of DNA strand breaks between 5′-PO₄and 3′-OH been reported.⁶

termini. These enzymes have been grouped into two classes

An obvious starting point for the development of inhibitors

depending on their co-substrate specificity for this reaction. and probes for bacterial NAD⁺-dependent DNA ligases is the

All eukaryotic DNA ligases are ATP-dependent, while the essen- co-substrate β-NAD⁺itself (Fig. 1). Recently, the crystal struc-

tial bacterial DNA ligases use β-NAD⁺as their co-substrate.^2,3tures of several NAD⁺-dependent DNA ligases have been

Some ATP-dependent DNA ligases do occur in bacteria but resolved.^7,8Several of these structures show the existence of a

they are expressed only under particular conditions and do not hydrophobic tunnel leading from the protein surface to the

seem to participate in essential DNA replication functions.^2,4adenylation domain.^1,7dβ-NAD⁺binds in an orientation in

In contrast, no functional NAD⁺-dependent DNA ligases have which positions N-1, C-2 and N-3 of the adenine ring are

located adjacent to the entry of this tunnel, with position C-2

pointing exactly in the direction of the tunnel.^1,7dAs a conse-

^aSchool of Pharmacy, University of East Anglia, Norwich Research Park, Norwich,

NR4 7TJ, UK

quence, this tunnel could, in principle, accommodate

additional substituents introduced on to the adenine base,

and thus be exploited for the development of NAD⁺-based

inhibitors. Indeed, 2-substituted derivatives of adenosine, ADP

and ATP have previously been reported as inhibitors of NAD⁺-

dependent DNA ligase.⁹However, base-modified derivatives of

the complete co-substrate β-NAD⁺have, to the best of our

knowledge, not been investigated against bacterial NAD⁺-

dependent DNA ligases. In contrast to simple adenine nucleo-

sides and nucleotides, such NAD⁺derivatives may be recog-

nized by NAD⁺-dependent DNA ligases not only as inhibitors,

^bDepartment of Biological Chemistry, John Innes Centre, Norwich Research Park,

Norwich, NR4 7UH, UK

^cSchool of Chemistry, University of East Anglia, Norwich Research Park, Norwich,

NR4 7TJ, UK

^dSchool of Biological Sciences, University of East Anglia, Norwich Research Park,

Norwich, NR4 7TJ, UK

^eDepartment of Chemistry, Faculty of Natural & Mathematical Sciences,

King’s College London, Britannia House, 7 Trinity Street, London, SE1 1DB, UK.

E-mail: gerd.wagner@kcl.ac.uk; Tel: +44 (0)20 7848 1926

†Electronic supplementary information (ESI) available. See DOI: 10.1039/

c5ob00294j

This journal is © The Royal Society of Chemistry 2015

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

Fig. 1 β-NAD⁺, and base-modiﬁed β-NAD⁺derivatives 1a–b, 2a–b and 3a–e investigated in this study.

but also as non-natural co-substrates. They could, therefore, be for the binding of β-NAD⁺and its derivatives to NAD⁺-depen-

highly useful as molecular probes to investigate the structural dent DNA ligase. In addition, they provide insights into the

requirements for co-substrate utilization versus inhibition.

structural and mechanistic factors that govern co-substrate

Herein, we describe the synthesis of NAD⁺derivatives modi- utilization by, and inhibition of, bacterial NAD⁺-dependent

fied in position 2 or 6 of the adenine ring by addition of aryl/ DNA ligases. Our findings therefore provide new insight to

heteroaryl substituents by Suzuki–Miyaura cross-coupling (1 & underpin the development of improved inhibitors against

2, Fig. 1). We report the biochemical characterisation of these these promising antimicrobial targets.

new NAD⁺derivatives, as well as a series of previously reported

8-substituted NAD⁺derivatives 3 (Fig. 1),¹⁰as inhibitors and/or

non-natural co-substrates toward two bacterial NAD⁺-depen-

dent DNA ligases from Escherichia coli (EcLigA) and Myco-

Results

Chemical synthesis

bacterium tuberculosis (MtLigA). To assess the importance of the

entire NAD⁺scaﬀold for biochemical activity, we also tested

Due to its labile N-glycosidic and pyrophosphate bonds and,

therefore, its limited chemical stability, the chemical modifi-

cation of β-NAD⁺is not trivial. Indeed, the most common

pathway toward adenine-modified NAD⁺derivatives is multi-

step synthesis, where the substitution of the adenine ring is

conducted on stable precursors of β-NAD⁺, such as adenosine

or AMP.¹¹Our general strategy for the introduction of aryl/

heteroaryl substituents in position 2, 6 or 8 of the adenine

ring in β-NAD⁺therefore started with the Suzuki–Miyaura

cross-coupling of suitably halogenated purine nucleosides or

nucleotides. These cross-coupled building blocks were then

converted into the corresponding NAD⁺derivatives 1–3 (Fig. 1)

via phosphorylation of the 5′-OH group (where required), acti-

vation of the phosphate in the form of its phosphoromorpholi-

date, and coupling of the activated AMP derivative to β-NMN

(nicotinamide adenine mononucleotide).

the corresponding 2- and 6-substituted AMP derivatives in our

assays (6 & 10, Fig. 2). Our results highlight key interactions

In order to synthesize AMP derivatives substituted at C-2 of

the adenine ring, the commercially available 2-iodo adenosine

Fig. 2 AMP and AMP derivatives 6a–b and 10a–b investigated in this

study.

Org. Biomol. Chem.

This journal is © The Royal Society of Chemistry 2015

View Article Online

Organic & Biomolecular Chemistry

Paper

was chosen as starting material. The synthetic route proceeded Conformational analysis

through the protection of the 2′,3′-OH of the ribose by O-iso-

propylidene¹²to obtain 4a, which then underwent Suzuki– Nucleosides and nucleotides can assume diﬀerent confor-

Miyaura cross-coupling for the introduction of the phenyl mations in solution depending on the free or restricted

group (4b).¹³The phosphorylation of 4a–b on 5′-OH was rotation about several bonds.¹⁹The nucleobase is oriented per-

carried out using phosphoramidite chemistry¹⁴to give 5a–b, pendicularly to the plane of the ribose ring and can assume,

which were used without further purification in the next oxi- by rotation about the N-glycosidic bond, two diﬀerent confor-

dation–deprotection steps¹⁴to aﬀord the 2-substituted AMP mations (Table 2): in the syn conformation, the nucleobase is

derivatives 6a–b (Scheme 1).

pointing towards the ribose, and away from it in the anti con-

The synthetic pathway to AMP derivatives substituted at C-6 formation. The syn/anti equilibrium can, in turn, influence the

started with the Suzuki–Miyaura cross-coupling of the com- pucker of the ribose ring, which exists in an equilibrium

mercially available 6-chloro purine nucleoside to provide 7a– between the North (N, or C-3′ endo) and South (S, or C-2′

b,¹⁵which were protected on the 2′,3′-OH of the ribose by endo) conformations (Table 2).^19aThe preference of a given

O-isopropylidene^12ato give 8a–b. These intermediates were adenine nucleoside or nucleotide for the syn or anti confor-

phosphorylated on the 5′-OH with POCl₃to give 9a–b,¹⁶which mation can be deduced experimentally from diagnostic chemi-

were partially deprotected from the O-isopropylidene protect- cal shift diﬀerences in the ¹H and ¹³C NMR spectra. In

1

ing group under acidic reaction conditions; their complete particular, a downfield shift for H-2′ in the H NMR spectrum

deprotection^13aaﬀorded the 6-substituted AMP derivatives of a given derivative of AMP/β-NAD⁺(ΔH-2′), relative to the

10a–b (Scheme 2). The corresponding 8-substituted AMP respective parent molecule, is usually indicative of a change

derivatives 11a–e were obtained by Suzuki–Miyaura cross- from the anti to the syn orientation,^19b,cas is an upfield shift

coupling from 8-bromo AMP, as previously reported.^10,17

for the C-2′ signal in the ¹³C NMR spectrum.^19dWith the

In the final step, all AMP derivatives were subjected to nucleobase in the syn orientation, the ribose tends to favour

oxidation–reduction condensation under Mukaiyama’s con- the less hindered S conformation, and the percentage of the

ditions^10,18to aﬀord the phosphoromorpholidate inter- ribose population in the S conformation can be estimated

19e

mediates 12–14, which were then coupled to β-NMN under experimentally from the coupling constant J_1′,2′

.

modified Khorana–Moﬀatt conditions^10,11to give the corres-

ponding NAD⁺derivatives 1–3 (Scheme 3 & Table 1).

Drawing on these diagnostic parameters, a conformational

analysis was performed on the various substituted AMP and

Scheme 1 Synthetic route to 2-substituted AMP derivatives. (i) Acetone, 70% HClO₄, rt, 1 h; then, NH₄OH, rt, 3 h; (ii) phenyl boronic acid, Pd(OAc)₂,

(2-biphenyl)dicyclohexylphosphine, K₃PO₄in dry dioxane, 100 °C, 24 h; (iii) 2-cyanoethyl N,N-diisopropylchlorophosphoramidite, DIPEA, dry

CH₂Cl₂, rt, 1 h (5a–b); (iv) 5–6 M ^tBuOOH in nonane, rt, 1 h; NH₄OH, rt, 3 h; Dowex 88 (H⁺), rt, 12 h; 19% (6a), 17% (6b). Overall yields: 18% (6a), 7% (6b).

Scheme 2 Synthetic route to 6-substituted AMP derivatives. (i) Phenyl or N-Boc-2-pyrrole boronic acid, Na₂PdCl₄, TPPTS, K₂CO₃, H₂O, 100 °C,

1–24 h, 50% (7a), 37% (7b); (ii) acetone, 70% HClO₄, rt, 1 h; then, NH₄OH, rt, 3 h, 85% (8a), 39% (8b); (iii) POCl₃, dry acetonitrile, 4 °C, 12–48 h;

(iv) Dowex 88 (H⁺), rt, 12 h (10a), or AcOH/H₂O (1 : 9), 90 °C, 3 h (10b); overall yield for steps (iii) and (iv): 28% (10a), 47% (10b). Overall yields: 12%

(10a), 7% (10b).

This journal is © The Royal Society of Chemistry 2015

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

Scheme 3 Synthesis of base-modiﬁed NAD⁺derivatives. (i) Morpholine, 2,2’-dithiopyridine, PPh₃, dry DMSO, rt, 1 h; (ii) β-NMN, anhydrous MgSO₄,

0.2 M MnCl₂in formamide, rt, 12 h. For substituents X and individual yields see Fig. 1 and Table 1.

tuted derivatives of both AMP and NAD⁺exist predominantly

Table 1 Base-modiﬁed NAD⁺derivatives 1a–b, 2a–b and 3a–e ^a

in the anti conformation.

Position of

substitution

Yield^b

(%)

Cmpd

X (substituent)

Expression, purification and activity of Escherichia coli and

Mycobacterium tuberculosis DNA ligases

1a

1b

2a

2b

3a

3b

3c

3d

3e

2

6

8

Iodo

Phenyl

Pyrrol-2-yl

53

28

56

E. coli DNA ligase (EcLigA) is a protein consisting of 671 amino

acids, with a molecular weight of 74 kDa. M. tuberculosis DNA

ligase (MtLigA) is a protein consisting of 691 amino acids,

with a molecular weight of 75 kDa. The protocol for the over-

expression and purification of recombinant forms of EcLigA

and MtLigA, containing a 10-His-tag at the N-terminus of the

protein, is well established in the literature²¹and it was fol-

lowed with minor changes to obtain highly pure EcLigA with

24 μM concentration and MtLigA with 1.7 μM concentration.

The activity of EcLigA and MtLigA was tested using a gel elec-

87

Phenyl

41^c

61^c

24^c

43^c

30^c

3-(N-Boc-aminomethyl)phenyl

Pyridin-3-yl

2,4-DMT-pyrimidin-5-yl

Pyrrol-2-yl

^aSee Scheme 3 for structures. ^bIsolated yields. ^cRef. 15.

β-NAD⁺derivatives prepared in this study (Table 2). Natural trophoresis assay of the ligation of nicked DNA.²¹Preliminary

AMP and β-NAD⁺predominantly adopt the anti orientation for experiments were undertaken to establish the optimal enzyme

the adenine base, and the S-type conformation for the concentration and, consequently, the time sampling, in order

ribose.²⁰AMP derivatives substituted at C-2 (6a & 6b) or C-6 to achieve 40–50% ligation in the control reaction without

(10a & 10b) displayed the same conformational preferences. In inhibitor; under these conditions, varying from batch to batch

contrast, AMP derivatives substituted at C-8 (11a–e) clearly pre- of enzyme, it was possible to study the ligation reaction in an

ferred the syn conformation for the adenine base, as previously initial state. To validate the gel electrophoresis assay for inhi-

observed for other 8-substituted adenine nucleotides,¹⁷while bition studies, quinacrine, a well characterized inhibitor of

the ribose predominantly adopted the S-type conformation. In EcLigA,^3,22was tested at diﬀerent concentrations (ESI): an IC₅₀

general, NAD⁺derivatives substituted at position 8 (3a–e) fol- value of 3.8 1.5 μM was determined, which is comparable to

lowed approximately the same trend as the corresponding the literature value of 1.5 0.2 μM,³indicating the suitability

AMP derivatives. Interestingly, for NAD⁺derivatives substituted of this protocol for studying inhibitors.

at position C-2 (1a & 1b) a small downfield shift for the H-2′

Biochemical results: 2-substituted NAD⁺and AMP derivatives

signal was observed in the ¹H NMR spectrum (Table 2b), in

contrast to the corresponding 2-substituted AMP derivatives

(6a & 6b). The values for ΔH-2′ are, however, notably smaller

for the 2-substituted NAD⁺derivatives (0.03–0.13) than for the

8-substituted NAD⁺derivatives (0.37–0.62), which unambigu-

ously exist in the syn conformation. Also, the characteristic

upfield shift for the C-2′ signal that is usually strongly indica-

tive of a change from the anti to the syn conformation,^19dwas

not observed in the ¹³C NMR spectra of either the 2-substi-

tuted AMP or NAD⁺series (δ_C-2′(ppm): NAD⁺74.66, (1a) 75.49,

AMP 74.43, (6a) 75.49). It is therefore likely that the 2-substi-

First, we investigated the eﬀect of an additional 2-substituent

at the NAD⁺scaﬀold on co-substrate utilization. We therefore

measured the enzymatic activity of EcLigA and MtLigA in the

presence of the 2-substituted NAD⁺derivatives 1a and 1b (at

200 µM) instead of the natural co-substrate β-NAD⁺. Against

EcLigA, both NAD⁺derivatives at 200 µM showed co-substrate

activity that was similar to (1a), or higher (1b) than that of the

natural co-substrate β-NAD⁺at 26 µM (Fig. 3). Interestingly,

when we repeated these experiments with a range of diﬀerent

concentrations of 1a or 1b, from 50 to 250 μM, there was no

Org. Biomol. Chem.

This journal is © The Royal Society of Chemistry 2015

View Article Online

Organic & Biomolecular Chemistry

Paper

Table 2 Conformational analysis of AMP and NAD⁺derivatives

Cmpd

Position of substitution

X (substituent)

S-type conformer^a(%)

H-2′ (ppm)

ΔH-2′^b(ppm)

syn/anti

(a) AMP derivatives

AMP

6a

—

2

6

8

H

Iodo

Phenyl

52

57

49

53

55

62

58

n/a

62

4.75

4.72

4.68

4.75

5.21

5.51

5.22

5.19

—

anti

syn

−0.03

−0.07

0

0.46

n/a

0.47

0.44

6b

10a

10b

11a

11b

11c

11d

11e

Pyrrol-2-yl

Phenyl

3-(N-Boc-aminomethyl)phenyl^c

Pyridin-3-yl

2,4-DMT-pyrimidin-5-yl

Pyrrol-2-yl

Cmpd

Position of substitution

X (substituent)

S-type conformer^d(%)

H-2′ (ppm)

ΔH-2′^e(ppm)

syn/anti

(b) NAD⁺derivatives

NAD⁺

1a

—

2

H

Iodo

Phenyl

55

53

n/a

59

n/a

57

55

48

4.71

4.74

4.84

5.14

5.16

5.24

5.08

5.33

—

anti

anti^f

syn

0.03

0.13

0.43

0.45

0.53

0.37

0.62

1b

3a

3b

8

3-(N-Boc-aminomethyl)phenyl

Pyridin-3-yl

2,4-DMT-pyrimidin-5-yl

Pyrrol-2-yl

3c

3d

8

syn

3e

8

^aCalculated as 10 × J_1′,2′

.

^bΔH-2′ has been calculated as the diﬀerence between the chemical shift of H-2′ of the adenine-substituted AMP

derivative and H-2′ of AMP (4.75 ppm) in D₂O (referenced for D₂O, δ = 4.79 ppm). ^{c 1}H NMR spectrum for 11b recorded in CD₃OD. n/a = not

available. ^dCalculated as 10 × J_1′,2′^eΔH-2′ has been calculated as the diﬀerence between the chemical shift of H-2′ of the adenine-substituted

.

NAD⁺derivative and H-2′ of β-NAD⁺(4.71 ppm) in D₂O (referenced for D₂O, δ = 4.79 ppm). ^fanti assignment based on ¹³C NMR analysis (see

main text for details).

Fig. 3 Non-natural co-substrate activity for 2-substituted NAD⁺derivatives 1a & 1b at a single concentration. Conditions: DNA substrate (0.5 µM),

EcLigA (0.07 µM), MtLigA (0.17 µM), 1a (200 µM) or 1b (200 µM) in 1 × buﬀer (30 mM Tris/HCl pH 8, 4 mM MgCl₂, 1 mM DTT, 50 µg mL⁻¹BSA). All

concentrations are ﬁnal concentrations. Reactions were incubated at 30 °C under shaking and sampled at 5 minutes. Control reactions were carried

out under the same conditions with β-NAD⁺(26 μM) instead of a 2-substituted NAD⁺derivative. Ligase activity is shown relative to full ligation of the

nicked DNA substrate (= 100% activity). Bars indicate mean values S.D. of triplicate experiments.

This journal is © The Royal Society of Chemistry 2015

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

Fig. 4 Inhibitory activity of 2-substituted AMP/NAD⁺derivatives at 200 μM: (a) with EcLigA (0.07 μM) and sampled at 5 and 15 minutes; (b) with

MtLigA (0.17 μM) and sampled at 5 and 20 minutes. Conditions: DNA substrate (0.5 µM), enzyme at noted concentration, β-NAD⁺(26 μM), 2-substi-

tuted derivative (200 μM) in 1 × buﬀer (30 mM Tris/HCl pH 8, 4 mM MgCl₂, 1 mM DTT, 50 μg mL⁻¹BSA). All concentrations are ﬁnal concentrations.

Reactions were incubated at 30 °C under shaking. Control reactions were carried out under the same conditions without any 2-substituted deriva-

tive. Ligase activity is shown relative to full ligation of the nicked DNA substrate (= 100% activity). Bars indicate mean values S.D. of triplicate

experiments.

obvious dependency of EcLigA activity on co-substrate concen- 2-iodo derivatives 1a and 6a after 5 minutes, although the

tration (data not shown). Against MtLigA, the co-substrate eﬀect was less pronounced than in the case of EcLigA (Fig. 4b).

activities of both NAD⁺derivatives (at 200 µM) were signifi- Again, DNA ligase activity increased significantly with longer

cantly lower than that of the unsubstituted parent dinucleo- incubation time. Interestingly, and in contrast to their behav-

tide, despite the more than 7-fold increased concentration of iour against EcLigA, the presence of either 2-phenyl derivatives

the co-substrate (Fig. 3). The corresponding 2-substituted AMP reduced MtLigA activity. 2-Phenyl NAD⁺1b in particular exhibi-

derivatives 6a and 6b were also tested under these conditions. ted strong inhibition, which persisted at 20 minutes (Fig. 4b).

As expected, they did not show any co-substrate activity with Finally, we determined full IC₅₀values for those 2-substituted

either enzyme (data not shown).

derivatives that showed greater than 50% inhibition at 200 µM

Next, we investigated if these 2-substituted NAD⁺and AMP (Table 3 and ESI†).

derivatives may act as inhibitors of the two bacterial DNA

ligases. For these inhibition studies, EcLigA and MtLigA were

Biochemical results: 6-substituted NAD⁺and AMP derivatives

co-incubated with β-NAD⁺at 26 μM and each of the NAD⁺(1a &

1b) and AMP (6a & 6b) derivatives at 200 μM, and ligase activity

was determined at two diﬀerent time points (Fig. 4). In the

presence of the 2-iodo derivatives of either NAD⁺(1a) or AMP

(6a), the activity of EcLigA was substantially reduced after

The 6-substituted NAD⁺(2a & 2b) and AMP (10a & 10b) deriva-

tives were tested as non-natural co-substrates (in the absence

of β-NAD⁺) or inhibitors (in the presence of β-NAD⁺at 26 μM)

of both bacterial DNA ligases, using the protocols described

above. Under these conditions, none of the NAD⁺or AMP

derivatives showed any activity as either co-substrates (data not

shown) or inhibitors (Fig. 5). These results suggest that the

presence of an additional substituent in position 6 is detri-

mental for binding to both EcLigA and MtLigA.

5

minutes, but returned to almost normal levels after

15 minutes (Fig. 4a). In contrast, no such drop in EcLigA

activity was observed with either of the 2-phenyl derivatives

(NAD⁺: 1b, AMP: 6b) after either 5 or 15 minutes (Fig. 4a).

MtLigA also showed reduced activity in the presence of the

Org. Biomol. Chem.

This journal is © The Royal Society of Chemistry 2015

View Article Online

Organic & Biomolecular Chemistry

Paper

two of these NAD⁺derivatives, 3c and 3e, led to moderately

reduced ligase activity at 10 minutes (Fig. 6). One of these

derivatives, 8-pyrrol-2-yl NAD⁺3e, was also tested against

MtLigA at 200 µM, but did not act as either inhibitor or co-sub-

strate of this DNA ligase (data not shown).

Table 3 IC₅₀values for the 2-substituted derivatives 1a, 1b and 6a

against EcLigA and MtLigA

IC₅₀(μM)

Cmpd

Scaﬀold

2-Substituent

EcLigA

MtLigA

1a

1b

6a

NAD⁺

AMP

Iodo

Phenyl

Iodo

138 18

n.d.^b

120 23

>200

73 15^c

16 8^a

Discussion

^aResidual enzyme activity ca. 30%. ^bNot determined. ^cResidual

enzyme activity ca. 20%.

The diﬀerent bioactivities of the 2-, 6- and 8-substituted NAD⁺

and AMP derivatives investigated in this study suggest that the

nature and position of the additional substituent at the

adenine base is critical for the interaction of these co-substrate

derivatives with the structure and mechanism of their target

DNA ligases. Mechanistically, NAD⁺-dependent DNA ligases

follow a three-step ping-pong mechanism (Fig. 7).^1,3,9d,23In the

first step, β-NAD⁺binds to the enzyme active site, where a con-

served nucleophilic lysine residue (Lys115 in EcLigA)^7dattacks

the α-phosphate, leading to the release of β-NMN and the for-

mation of an adenylated-ligase intermediate (E-AMP). In the

second step, the DNA nick is positioned in the proximity of the

adenylated-ligase intermediate and its nucleophilic 5′ phos-

Biochemical results: 8-substituted NAD⁺derivatives

The 8-substituted NAD⁺derivatives 3a–e were evaluated as

non-natural co-substrates (in the absence of β-NAD⁺) or inhibi-

tors (in the presence of β-NAD⁺at 26 μM) of EcLigA, under the

conditions described above. At 250 µM, none of these deriva-

tives showed any co-substrate activity, even after 1 h of incu-

bation (data not shown). In the co-incubation experiments,

Fig. 5 Inhibitory activity of 6-substituted AMP/NAD⁺derivatives: (a) with EcLigA (0.07 μM), 6-substituted derivative (250 μM) and sampled at 7 and

15 minutes; (b) with MtLigA (0.17 μM), 6-substituted derivative (200 μM) and sampled at 5 minutes. Conditions: DNA substrate (0.5 µM), enzyme at

noted concentration, β-NAD⁺(26 μM), 6-substituted derivative at noted concentration in 1 × buﬀer (30 mM Tris/HCl pH 8, 4 mM MgCl₂, 1 mM DTT,

50 μg mL⁻¹BSA). All concentrations are ﬁnal concentrations. Reactions were incubated at 30 °C under shaking. Control reactions were carried out

under the same conditions without any 6-substituted derivative. Ligase activity is shown relative to full ligation of the nicked DNA substrate (= 100%

activity). Bars indicate mean values S.D. of triplicate experiments.

This journal is © The Royal Society of Chemistry 2015

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

Fig. 6 Inhibitory activity of 8-substituted NAD⁺derivatives at 250 μM against EcLigA. Conditions: DNA substrate (0.125 µM), EcLigA (0.07 μM),

β-NAD⁺(26 μM), 8-substituted derivative (250 μM) in 1 × buﬀer (30 mM Tris/HCl pH 8, 4 mM MgCl₂, 1 mM DTT, 50 μg mL⁻¹BSA). All concentrations

are ﬁnal concentrations. Reactions were incubated at 30 °C under shaking and sampled at 10 minutes. Control reactions were carried out under the

same conditions without any 8-substituted derivative. Data are normalized to the control reaction. Bars indicate mean values S.D. of triplicate

experiments.

Fig. 7 General reaction pathway of NAD⁺-dependent DNA ligases

(abbreviations: E: enzyme; β-NAD⁺: nicotinamide adenine dinucleotide;

β-NMN: nicotinamide mononucleotide; AMP: adenosine monophosphate.

phate attacks the phosphoramidate bond in the adenylated-

ligase intermediate to form an adenylated-DNA intermediate.

In the final step, the nucleophilic hydroxyl group in the 3′ posi-

tion of the DNA nick attacks the new pyrophosphate bond,

resulting in the formation of a phosphodiester bond between

Fig. 8 Overlay of EcLigA (green, in complex with nicked DNA-adenylate,

the 5′ and 3′ positions of DNA and the release of AMP.

PDB 2OWO)^7dand the homologous DNA ligase from Enterococcus faecalis

Throughout these steps, both the DNA ligase itself and the

(purple, in complex with β-NAD⁺, PDB 1TAE),^7billustrating the diﬀerent

positions of subdomain 1a in the “open” and “closed” DNA ligase confor-

bound β-NAD⁺co-substrate undergo significant conformation-

al changes. The N-terminal adenylation domain of DNA ligases

is composed of two subdomains 1a and 1b, which include

binding sites for, respectively, the NMN (subdomain 1a) and

AMP (subdomain 1b) moieties of β-NAD⁺(Fig. 8). The binding

of the β-NAD⁺co-substrate triggers the closure of subdomain

1a onto subdomain 1b, and the transition from an “open” to a

“closed” conformation, which is critical for DNA ligase activity

(Fig. 8).^7b,8Recent studies with Haemophilus influenzae LigA

mations. Proteins are shown in cartoon representation, ligands as space-

ﬁlled models (green: AMP; purple: β-NAD⁺). The box indicates the β-NMN

binding site. Residues 294–308 (2OWO) have been removed for clarity.

suggest that the initial recognition of β-NAD⁺very likely occurs

at the NMN binding site on subdomain 1a, providing an expla-

nation for the selectivity of bacterial DNA ligases for β-NAD⁺

Org. Biomol. Chem.

This journal is © The Royal Society of Chemistry 2015

View Article Online

Organic & Biomolecular Chemistry

Paper

In this context, it is interesting to note that some of these

8-substituted NAD⁺derivatives, although inactive against DNA

ligases, show activity against other NAD⁺-dependent enzymes.

8-Phenyl NAD⁺(3a), for example, which is neither an inhibitor

nor a co-substrate for either of the bacterial DNA ligases, is

recognised as an inhibitor, but not a substrate by the NAD⁺-

dependent histone deacetylase sirtuin 2.¹⁷Similarly, 8-(pyrrol-

2-yl) NAD⁺(3e), which is a modest inhibitor, but not a co-sub-

strate for either DNA ligase, is recognised as a substrate by

various NAD⁺-consuming enzymes, including a nucleotide

pyrophosphatase, a NAD⁺-glycohydrolase and an ADP-ribosyl

cyclase.¹⁰These diﬀerential activities suggest that the modifi-

cation of the adenine base, especially in position 8, oﬀers an

opportunity for the development of NAD⁺derivatives with

selectivity for individual NAD⁺-dependent enzymes or enzyme

classes. This is particularly important given the ubiquitous use

of NAD⁺as a substrate, co-substrate or co-factor by many

diﬀerent enzymes.

Fig. 9 Overlay of 2-iodo AMP 6a (green) and 2-phenyl AMP 6b

(orange), docked into the active site of EcLigA (PDB 2OWO),^7dand AMP

(yellow) from the crystal structure. Hydrogen bonding interactions

between the adenine base of AMP and residues Glu113 and Lys290 are

indicated with broken lines.

Against the two bacterial DNA ligases, the most pro-

nounced activities, in both the AMP and NAD⁺series, were

observed with 2-substituted derivatives. Thus, 2-iodo AMP 6a

was a relatively good inhibitor of EcLigA (IC₅₀120 μM), and a

modest inhibitor of MtLigA (Fig. 4). In contrast, 2-phenyl

AMP 6b was less active as an inhibitor. This result indicates

that the 2-substituent may be involved in direct interactions

with the target DNA ligase. This hypothesis is supported by

results from molecular docking experiments with 6a and 6b

and EcLigA (PDB 2OWO),^7dwhich suggest that these 2-substi-

tuted AMP derivatives can be accommodated at the active site

in an orientation that is nearly identical to that of natural AMP,

with the adenine and ribose moieties of all three ligands over-

lapping almost perfectly (Fig. 9). In this pose, the adenine ring

of 6a and 6b maintains all key interactions with active site resi-

dues (Fig. 9), while the substituent in position 2 fits into the

hydrophobic tunnel adjacent to the adenine binding site

(Fig. 10). However, while the 2-iodo substituent in 6a may

electrostatically interact with Lys290 at the entrance of the

tunnel, the phenyl substituent in 6b does not appear to make

any specific contacts with residues in the tunnel (Fig. 10). This

diﬀerence may provide an explanation for the superior inhibi-

tory activity of 6a. Importantly, such a binding mode at the

active site is also compatible with the presence of a dynamic

equilibrium between the two states of the ligase bound and

unbound to 6a. Indeed, AMP derivatives (in this particular case,

6a) lack the pyrophosphate bond necessary to catalyse the lig-

ation activity and, thus, any ligation observed in these experi-

ments must be promoted exclusively by β-NAD⁺, following the

displacement of 6a.

over ATP.⁸In the closed conformation, the adenine of β-NAD⁺

is bound tightly to the ligase active site by several key inter-

actions with conserved residues, such as Lys290 (H-bonding to

N1), Glu113 (H-bonding to N6) and Tyr225 (π–π stacking) in

EcLigA (Fig. 9); in particular, the interaction with Lys290 is

critical for enzymatic activity.^7dAll together, these interactions

hold the adenine ring in place, while the conformation of the

AMP moiety changes, through rotation of the ribose about the

N-glycosidic bond, from syn in non-covalently bound β-NAD⁺,

to anti in the adenylate-ligase intermediate, and back to syn in

the adenylated-DNA intermediate.^7d

In our experiments, 6-substituted derivatives were neither

co-substrates (in the case of the NAD⁺derivatives) nor inhibi-

tors (in the case of both NAD⁺and AMP derivatives). These

results suggest that the presence of an additional substituent

in position 6 is detrimental for co-substrate binding at both

EcLigA and MtLigA. While none of the 8-substituted NAD⁺

derivatives were utilised as co-substrates by EcLigA either, two

derivatives in this series (3c, 3e) were weak inhibitors. This

pattern of behaviour (no co-substrate activity, weak inhibition)

suggests that 8-substituted NAD⁺derivatives can, in principle,

bind at the EcLigA active site, but do so in a non-productive

orientation. Due to the proximity of the protein backbone and

position 8 of the adenine ring, there is not enough space

within the active site of EcLigA to accommodate an additional

substituent at this position without changing the original posi-

tion of the adenine (Fig. 9). Therefore, the binding of 8-substi-

tuted NAD⁺derivatives at the active site probably occurs in a

diﬀerent orientation from that of the natural co-substrate

β-NAD⁺, precluding the establishment of key interactions

within the active site and abolishing co-substrate activity. This

interpretation is supported by molecular docking experiments

with 6-pyrrolyl AMP (10b) and 8-pyrrolyl AMP (11e), which

suggest that these compounds bind in a non-productive con-

formation within the active site of EcLigA (data not shown).

The arguably most intriguing results were observed with

the 2-substituted NAD⁺derivatives 1a and 1b. In the absence

of β-NAD⁺, 2-iodo NAD⁺1a at 200 µM behaved as a co-substrate

of EcLigA, with similar DNA ligase activity to that of unmodi-

fied β-NAD⁺at 26 µM. However, in the presence of both 1a (at

200 μM) and β-NAD⁺(at 26 μM), the activity of EcLigA was

lower than for either 1a or β-NAD⁺alone. Indeed, in the pres-

ence of β-NAD⁺, 1a displayed comparable inhibitory activity

This journal is © The Royal Society of Chemistry 2015

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

Fig. 10 Overlay of docking solutions for 2-iodo AMP 6a (green) and 2-phenyl AMP 6b (orange) in the NAD⁺-binding site of EcLigA (PDB 2OWO).^7d

EcLigA is shown in surface representation. Residues in the foreground have been removed for clarity. The hydrophobic tunnel extends to the left of

the ligands. Lys290 sits at the entrance to this tunnel and is shown in red. Fragments of DNA (salmon) are shown in the DNA binding site.

against EcLigA as the corresponding AMP derivative 6a alone these interactions simultaneously, in the way the natural co-

(IC₅₀: 138 vs. 120 μM). A possible explanation for this un- substrate NAD⁺does in the “closed” conformation. Such an

expected result may be found in the conformational changes inability to stabilize the classical “closed” conformation, due

required for full DNA ligase activity, and the involvement of to the presence of the 2-iodo substituent on the adenine ring,

two distinct nucleotide binding sites (Fig. 8 and 11). As dis- would explain the experimentally observed behaviour of 1a

cussed, the transition of NAD⁺-dependent DNA ligases from both in the absence and presence of β-NAD⁺. In the absence of

the “open” to the “closed” conformation involves the binding β-NAD⁺, additional conformational rearrangements may be

of β-NAD⁺, first with its NMN half to subdomain 1a, and then required for the closure of subdomain 1a, containing NAD⁺

with its AMP half to subdomain 1b.^7b,d,8In principle, NAD⁺derivative 1a, onto subdomain 1b, which would explain the

derivative 1a can also bind with its NMN moiety to subdomain slower co-substrate activity that was observed for 1a (Fig. 11).

1a, like β-NAD⁺, and with its AMP half to subdomain 1b, like In the presence of β-NAD⁺, the slow closure of subdomain 1a

the corresponding 2-iodo AMP derivative 6a. However, our onto subdomain 1b could result in the simultaneous binding

results suggest, that 1a may not be able to engage in both of of two distinct dinucleotides at, respectively, the AMP and the

Fig. 11 Hypothetical model for the binding and utilisation of 2-iodo NAD⁺1a by DNA ligases. (a) Binding of the natural co-substrate β-NAD⁺at sub-

domain 1a triggers the formation of the “closed” ligase conformation, which has full catalytic activity. (b) Binding of NAD⁺derivative 1a also leads to

a productive ligase conformation, but more slowly. In the presence of β-NAD⁺, the separate binding of the two dinucleotides to, respectively, sub-

domains 1a and 1b may occur, abolishing catalytic activity.

Org. Biomol. Chem.

This journal is © The Royal Society of Chemistry 2015

View Article Online

Organic & Biomolecular Chemistry

Paper

Fig. 12 Overlay of EcLigA (full surface, PDB 2OWO)^7dand MtLigA (mesh surface, PDB 3SGI). Left: view into the hydrophobic tunnel, view point

rotated by ca. 90° from Fig. 10. Right: same view as in Fig. 10. Residues in the foreground have been removed for clarity. The AMP molecule in the

active site of each ligase is shown in stick representation (EcLigA: green; MtLigA: blue). The lysine residue at the entrance to the hydrophobic tunnel

is shown in red for EcLigA (Lys290, full surface), and in yellow for MtLigA (Lys300, mesh surface).

NMN binding site (Fig. 11). With both binding sites occupied orientation of 2-substituted NAD⁺derivatives in the active site

by two distinct dinucleotides (e.g. β-NAD⁺at the AMP of the respective DNA ligase.

binding site, and 1a at the NMN binding site), neither would

be used eﬃciently as a co-substrate, and hence the overall

DNA ligase activity would be lower than for either β-NAD⁺or

1a alone, as observed. Interestingly, this inhibitory profile

was dependent on the nature of the 2-substituent. In contrast

Conclusions

to 1a, 2-phenyl NAD⁺1b had pronounced co-substrate activity In this study, we have evaluated a collection of 2-, 6- or 8-sub-

for EcLigA, but did not act as an inhibitor, possibly because stituted derivatives of NAD⁺and AMP as potential co-sub-

NAD⁺derivative 1b binds as well as β-NAD⁺to the enzyme, strates or inhibitors of two bacterial DNA ligases, EcLigA and

allowing a faster closure of the two enzyme subdomains. In the MtLigA. The diﬀerent substitution patterns had a pronounced

context of our inhibition model, these diﬀerences suggest that eﬀect on biochemical activity: while the 6- and 8-substituted

the 2-substituent has an important role for orienting the NAD⁺derivatives were largely inactive, interesting co-substrate and/

scaﬀold in the active site of EcLigA, and consequently for sub- or inhibitory activities were observed in the series of 2-substi-

domain closure.

tuted derivatives. These results indicate that an additional sub-

Against MtLigA, the co-substrate activity of 1a and 1b at stituent at the 2-position of NAD⁺/AMP can be accommodated,

200 μM was lower than that of β-NAD⁺at 26 μM (Fig. 3), while modifications in other positions are less well tolerated,

suggesting once again that due to the presence of the 2-substi- most likely for steric and/or conformational reasons. The avail-

tuent, these NAD⁺derivatives are bound at the ligase active site ability of two pairs of NAD⁺/AMP derivatives with the same

in a non-optimal orientation. Compared to EcLigA, the general 2-substituent (2-iodo: 1a & 6a; 2-phenyl: 1b & 6b) has enabled,

trend for inhibition (2-iodo > 2-phenyl) was the same in the for the first time, an assessment of the role of the NMN frag-

AMP series (6a > 6b), but reversed in the NAD⁺series (1b > 1a) ment for DNA ligase inhibition by base-modified adenine

(Fig. 4b). Indeed, the MtLigA inhibitory activity of 2-phenyl nucleotides. Our results show that the presence of the com-

NAD⁺1b was significantly greater than that of the corres- plete NAD⁺scaﬀold, while not an absolute prerequisite for

ponding AMP derivative 6b. More interestingly, the MtLigA inhibitory activity, can alter the profile of inhibition, leading

inhibition of 1b was maintained even after prolonged incu- e.g. to synergistic eﬀects with β-NAD⁺(in the case of 1a) or

bation times. This profile is unique amongst the NAD⁺and overcoming the time-dependence of inhibition (1b against

AMP derivatives tested, and suggests that 1b may have an allo- MtLigA). The comparative analysis of our data from co-sub-

steric mode of inhibition, which would be largely independent strate and inhibition experiments raises the intriguing possi-

from the presence of β-NAD⁺. Taken together, these results bility that these unique inhibition profiles may result from an

suggest that subtle diﬀerences in the shape of the hydrophobic unexpected mode of inhibition, which targets the confor-

tunnel between EcLigA and MtLigA (Fig. 12) critically influence mational changes of these DNA ligases during the catalytic

the binding of diﬀerent 2-substituents and, consequently, the cycle. Our findings may, therefore, provide a promising start-

This journal is © The Royal Society of Chemistry 2015

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

ing point for the rational design of a new class of inhibitors 70% HClO₄(2.5 equiv.) was added to the mixture. The reaction

against bacterial DNA ligases.

was stirred at room temperature for 1.5 h. Then, NH₄OH 25%

(5.6 equiv.) was added to the reaction, and the solution stirred

for 3 h. The mixture was diluted with EtOAc and extracted with

brine (3×). Then, the organic phase was dried over anhydrous

MgSO₄and evaporated to dryness in vacuo. The crude product

was purified as indicated for each compound.

Experimental

Chemical synthesis: general conditions

All synthetic reagents were obtained commercially and used as

received, including anhydrous solvents over molecular sieves,

unless otherwise stated. Milli-Q water was used for the prepa-

ration of aqueous solutions and as solvent in the reactions.

The 8-substituted β-NAD⁺derivatives 3a–e were prepared as

previously reported.¹⁰For the synthesis of phosphoramidites,

CH₂Cl₂was distilled over calcium hydride, and DIPEA was

dried over molecular sieves (4 Å) for four days before the use.

For the preparation of dinucleotides, MnCl₂solution in forma-

mide was dried over molecular sieves (4 Å) and stored under

nitrogen atmosphere. All reagents and solvents used for

analytical applications were of analytical quality. TLC was per-

formed on precoated slides of Silica Gel 60 F254 (Merck).

Spots were visualized under UV light (254/365 nm). Reaction

products were characterized by high-resolution mass spec-

trometry (HR-MS), ¹H, ¹³C and ³¹P NMR spectroscopy, and,

Suzuki–Miyaura cross-coupling on C-6.¹⁵6-Chloro purine

nucleoside (1 equiv.), Na₂PdCl₄(0.03 equiv.), TPPTS (2.5 equiv.

to Pd), K₂CO₃(2 equiv.) and aryl/heteroaryl boronic acid (1.2

equiv.) were placed in a flask and purged with nitrogen for

15 min. Degassed water was added by syringe to the reaction

vessel and the mixture was stirred at 100 °C under nitrogen for

the given time. When the reaction was complete, the mixture

was evaporated in vacuo, and the crude product purified as

indicated for each compound.

Phosphorylation of 5′-OH by POCl₃.¹⁶The respective 2′,3′-

O-isopropylidene protected nucleoside (1 equiv.) was dissolved

in anhydrous acetonitrile and the solution cooled down to

2 °C in ice-bath. Then, the appropriate amount of POCl₃was

added drop-wise to the reaction mixture, which was kept at

4 °C for the given time. The reaction was quenched adding

drop-wise 1 M TEAB buﬀer. Then, the aqueous solution was

evaporated to dryness in vacuo, and the crude product was pur-

ified by purification method 1, as indicated for each com-

pound, to aﬀord a mixture of 2′,3′-O-isopropylidene protected

and unprotected nucleotides. The 2′,3′-O-isopropylidene pro-

tected nucleotides were deprotected and purified as indicated

for each compound.

1

where applicable, by HPLC. H NMR spectra were recorded at

298 K on a Bruker BioSpin GmbH spectrometer at 400 MHz.

¹³C NMR spectra were recorded at 300 K on a Bruker BioSpin

GmbH spectrometer at 101 MHz. ³¹P NMR spectra were

recorded at 298 K on a Bruker BioSpin GmbH spectrometer at

161.98 MHz. Chemical shifts (δ) are reported in ppm (parts

per million) and referenced to D₂O, CD₃OD or CDCl₃(respect-

ively, δ_H4.79, 3.31, 7.26) for ¹H NMR, and to CH₃of TEA,

CD₃OD or CDCl₃(respectively, δ_C9.07, 49.00, 77.16) for ¹³C

NMR. Coupling constants (J) are reported in Hz. Accurate elec-

trospray ionization mass spectra were obtained on a Finnigan

MAT 900 XLT mass spectrometer at the EPSRC National Mass

Spectrometry Service Centre, Swansea. Ion-pair and ion-

exchange 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:

Preparative chromatography – purification method 1: Ion-

pair chromatography was performed using Lichroprep RP-18

resin and a variable gradient of methanol against 0.05 M TEAB

(triethylammonium bicarbonate buﬀer). Product containing

fractions were combined and reduced to dryness. The residue

was co-evaporated repeatedly with methanol to remove

residual TEAB.

Synthesis of AMP-morpholidate derivatives.^10,18The respect-

ive nucleotide (1 equiv.) was dissolved in DMSO, and then co-

evaporated with DMF (3×). The residue was dissolved in anhy-

drous DMSO and morpholine (6 equiv.) was added. After

5 min, 2,2′-dithiopyridine (3 equiv.) was added and, sub-

sequently, triphenylphosphine (3 equiv.). The yellow solution

was stirred at room temperature under nitrogen until TLC

showed complete conversion. The product was isolated by

purification method 1, or by precipitation as sodium salt with

NaI in acetone (0.1 M), as indicated for each compound. The

NaI solution was added drop-wise to the reaction mixture until

a solid precipitate formed; then, the precipitate was washed

repeatedly with acetone until the solution was colourless. The

crude product was dried and used in the next reaction step

without further purification.

Synthesis of NAD⁺derivatives.^10,11The appropriate amounts

of AMP-morpholidate and β-NMN were co-evaporated with pyr-

idine (3×). The flask with the residual solid was purged with

nitrogen for 15 min. Anhydrous MgSO₄(2 equiv.) and MnCl₂

(0.2 M in formamide, 1.5 equiv.) were added and the reaction

was stirred at room temperature under nitrogen until TLC

showed complete conversion. The reaction was stopped by

drop-wise addition of acetonitrile until a white precipitate

formed. The supernatant was removed and the solid dissolved

in water and evaporated to dryness in vacuo. The crude

Preparative chromatography – purification method 2: Anion

exchange chromatography was performed using Bioscale™

Mini Macro-Prep High Q cartridges and a gradient of 0–100%

1 M TEAB (pH 7.3) against Milli-Q water. Product containing

fractions were combined and reduced to dryness. The residue

was co-evaporated repeatedly with methanol to remove

residual TEAB.

General synthetic procedures

2′,3′-O-Isopropylidene protection of nucleosides.^12aThe product was purified by chromatography, as indicated for each

respective nucleoside (1 equiv.) was suspended in acetone and compound.

Org. Biomol. Chem.

This journal is © The Royal Society of Chemistry 2015

View Article Online

Organic & Biomolecular Chemistry

Paper

2-Iodo-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-adenine ¹³C NMR (101 MHz; D₂O) δ: 156.10, 150.29, 145.45, 140.40,

(4a).^12bThe title compound was prepared according to the 120.35, 87.93, 85.06, 75.49, 71.26, 64.78, 47.49, 9.07; ³¹P NMR

general synthetic procedure 1, from a suspension of the com- (162 MHz; D₂O) δ: 1.65; m/z (ESI) 473.9666 [M + H]⁺,

mercially available 2-iodo adenosine (0.148 g, 1 equiv.) in C₁₀H₁₄IN₅O₇P⁺requires 473.9670.

acetone (10 mL). The crude product was purified by silica gel

Phosphoric

acid

mono-5′-[2-phenyl-9β-^D-ribofuranosyl-

chromatography with an eluent 98 : 2 CH₂Cl₂/MeOH. 4a was adenine] (6b). 4b (0.032 g, 1 equiv.) was dissolved in anhy-

obtained as colourless oil (0.160 g, 98% yield). ¹H NMR drous CH₂Cl₂and co-evaporated twice to dryness. Then,

(400 MHz; CD₃OD) δ: 8.20 (1H, s, 8-H), 6.10 (1H, d, J_1′,2′

=

it was dissolved again in anhydrous CH₂Cl₂(2.5 mL),

3.2 Hz, 1′-H), 5.24 (1H, m, 2′-H), 5.01 (1H, m, 3′-H), 4.34 (1H, and DIPEA (2.5 equiv.) and 2-cyanoethyl N,N-diisopropyl-

m, 4′-H), 3.79 (1H, dd, J_5′a,4′= 3.7 Hz, J_5′a,b= 12.0 Hz, 5′a-H), chlorophosphoramidite (2 equiv.) were added via syringe. The

3.72 (1H, dd, J_5′b,4′= 4.5 Hz, J_5′b,a= 12.0 Hz, 5′b-H), 1.61 (3H, s, reaction was stirred under nitrogen at room temperature for

isoprop.), 1.38 (3H, s, isoprop.); m/z (ESI) 432.0167 [M − H]⁻, 2 h. The intermediate 5b was isolated as a mixture of diastereo-

C₁₃H₁₅IN₅O₄⁻requires 432.0174.

mers by silica gel chromatography, with an eluent 80 : 20

2-Phenyl-9β-(2′,3′-O-isopropylidene-^D-ribofuranosyl)-adenine CH₂Cl₂/EtOAc with 3% TEA. ¹H NMR (400 MHz; CDCl₃)

(4b).^13a4a (0.083 g, 1 equiv.), phenyl boronic acid (1.7 equiv.), δ: 8.42–8.34 (2H, m, Ph), 7.98 (1H, s, 8-H), 7.49–7.41 (3H, m,

Pd(OAc)₂(0.12 equiv.), (2-biphenyl)dicyclohexylphosphine Ph), 6.23 (0.5H, d, J_1′,2′= 2.1 Hz, 1′-H), 6.20 (0.5H, d, J_1′,2′

=

(0.16 equiv.) and K₃PO₄(2.2 equiv.) were added in a sealed 2.1 Hz, 1′-H), 5.58 (0.5H, dd, J = 6.2, 2.1 Hz, 2′-H), 5.55 (0.5H,

tube and purged with nitrogen for 10 min; after addition of dd, J = 6.2, 2.1 Hz, 2′-H), 5.21–5.15 (1H, m, 3′-H), 4.50–4.45

anhydrous 1,4-dioxane (3 mL), the tube was sealed and heated (1H, m, 4′-H), 3.96–3.66 (4H, m, 5′-H2, P-O-CH₂), 3.51 (2H, m,

at 100 °C for 24 h. The reaction mixture was evaporated to N-CH), 2.66 (2H, m, CH₂-CN), 1.65 (3H, s, isoprop.), 1.42 (3H, s,

dryness in vacuo and the crude product was purified by silica isoprop.), 1.14–1.05 (6H, m, N-isoprop.); ³¹P NMR (162 MHz;

gel chromatography with a gradient 0–6% MeOH in CH₂Cl₂. CDCl₃) δ: 148.98. Then, 5.0–6.0 M tBuOOH in nonane solution

4b was obtained as white crystals (0.0318 g, 43% yield). (4 equiv.) was added to the reaction and stirred for 1 h. The

¹H NMR (400 MHz; CDCl₃) δ: 8.33–8.08 (2H, m, Ph), 7.83 (1H, mixture was diluted with EtOAc and extracted with a saturated

s, 8-H), 7.46 (3H, m, Ph), 5.90 (1H, d, J_1′,2′= 4.6 Hz, 1′-H), sodium bicarbonate solution (2×) and brine (1×). The organic

5.64–5.41 (1H, m, 2′-H), 5.19 (1H, m, 3′-H), 4.48 (1H, m, 4′-H), phase was dried over anhydrous MgSO₄and evaporated to

3.81 (2H, m, 5′-H2), 1.66 (3H, s, isoprop.), 1.41 (3H, s, isoprop.); dryness in vacuo; then, the residue was suspended in NH₄OH

¹³C NMR (101 MHz; CDCl₃) δ: 160.95, 155.71, 150.15, 140.86, 25% (30 equiv.) under stirring at room temperature overnight.

138.12, 130.27, 128.64, 128.47, 119.71, 114.30, 93.45, 85.92, The solution was evaporated to dryness in vacuo, and the

82.60, 81.71, 63.34, 27.76, 25.48; m/z (ESI) 384.1670 [M + H]⁺, crude product suspended in water and stirred with Dowex 88

C₁₉H₂₂N₅O₄⁺requires 384.1666.

(H⁺) (ca. 0.1 g) overnight. After filtration, the solution was evap-

Phosphoric acid mono-5′-[2-iodo-9β-^D-ribofuranosyl-adenine] orated to dryness in vacuo and the crude product purified by

(6a). 4a (0.0176 g, 1 equiv.) was dissolved in anhydrous CH₂Cl₂purification method 1 (0–50% MeOH against 0.05 M TEAB

and co-evaporated twice to dryness. Then, it was dissolved buﬀer over 350 mL, then isocratic 60% over 50 mL, flow: 3 mL

again in anhydrous CH₂Cl₂(2 mL), and DIPEA (2.5 equiv.) and min⁻¹, fraction size: 5 mL), to obtain 6b as a colourless oil

2-cyanoethyl N,N-diisopropylchlorophosphoramidite (1 equiv.) (0.0075 g, 0.92 equiv. TEA, 17% yield). ¹H NMR (400 MHz;

were added via syringe. The reaction was stirred under nitro- D₂O) δ: 8.35 (1H, s, 8-H), 7.90 (2H, d, J = 7.0 Hz, Ph), 7.40 (3H,

gen at room temperature for 2 h. Then, 5.0–6.0 M tBuOOH in m, Ph), 6.11 (1H, d, J_1′,2′= 5.2 Hz, 1′-H), 4.68 (1H, apparent t,

nonane solution (4 equiv.) was added to the reaction and 2′-H), 4.48 (1H, m, 3′-H), 4.32 (1H, br s, 4′-H), 4.17–3.98 (2H,

stirred for 1 h. The mixture was diluted with EtOAc and m, 5′-H2), 3.13 (5.5H, q, J = 7.3 Hz, CH₂TEA), 1.22 (9.6H, t, J =

extracted with a saturated sodium bicarbonate solution (2×) 7.3 Hz, CH₃TEA); ¹³C NMR (101 MHz; D₂O) δ: 156.37, 152.78,

and brine (1×). The organic phase was dried over anhydrous 152.38, 145.28, 134.60, 132.37, 131.31, 130.25, 129.66, 99.29,

MgSO₄and evaporated to dryness in vacuo; then, the residue 88.30, 84.94, 75.33, 71.21, 65.19, 59.67, 47.51, 9.07; ³¹P NMR

was suspended in NH₄OH 25% (30 equiv.) under stirring at (162 MHz; D₂O) δ: 0.94; m/z (ESI) 422.0870 [M − H]⁻,

room temperature overnight. The solution was evaporated to C₁₆H₁₇N₅O₇P⁻requires 422.0871.

dryness in vacuo, and the crude product suspended in water

6-Phenyl-9β-^D-ribofuranosyl-purine (7a).¹⁵The title com-

and stirred with Dowex 88 (H⁺) (ca. 0.1 g) overnight. After fil- pound was prepared according to the general synthetic pro-

tration, the solution was evaporated to dryness in vacuo and cedure 2, from the commercially available 6-chloro purine

the crude product purified by purification method 1 (0–50% nucleoside (0.2 g, 1 equiv.), using Na₂PdCl₄, TPPTS, phenyl

MeOH against 0.05 M TEAB buﬀer over 350 mL, then isocratic boronic acid and K₂CO₃in water (6 mL) at 100 °C. After 1 h,

50% over 50 mL, flow: 3 mL min⁻¹, fraction size: 5 mL), to the reaction was cooled down and extracted with EtOAc; the

obtain 6a as colourless oil (0.005 g, 1.68 equiv. TEA, 19% organic phase was dried over anhydrous MgSO₄and evapor-

1

yield). H NMR (400 MHz; D₂O) δ: 8.37 (1H, s, 8-H), 5.99 (1H, ated to dryness in vacuo to obtain 7a as white powder

d, J_1′,2′= 5.7 Hz, 1′-H), 4.68 (1H, apparent t, 2′-H), 4.46 (1H, m, (0.1136 g, 50% yield). ¹H NMR (400 MHz, DMSO-d₆) δ: 9.02

3′-H), 4.33 (1H, m, 4′-H), 4.07–3.89 (2H, m, 5′-H2), 3.15 (9.5H, (1H, s, H-2), 8.93 (1H, s, H-8), 8.84 (2H, d, J = 8.0 Hz, Ph),

q, J = 7.3 Hz, CH₂TEA), 1.24 (15.1H, t, J = 7.3 Hz, CH₃TEA); 7.55–7.65 (3H, m, Ph), 6.11 (1H, d, J_1′,2′= 5.5 Hz, H-1′), 5.56

This journal is © The Royal Society of Chemistry 2015

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

(1H, d, J = 5.7 Hz, OH-2′), 5.25 (1H, d, J = 5.0 Hz, OH-3′), 5.13 50 mL, isocratic 60% over 50 mL, flow: 3 mL min⁻¹, fraction

(1H, t, J = 5.4 Hz, OH-5′), 4.67 (1H, m, H-2′), 4.23 (1H, m, H-3′), size: 5 mL), to give a mixture of 2′,3′-O-isopropylidene pro-

4.02 (1H, m, H-4′), 3.76–3.59 (2H, m, 2H-5′).

tected 9a (0.024 g, 2.4 equiv. TEA, 12% yield) and unprotected

6-(Pyrrol-2-yl)-9β-^D-ribofuranosyl-purine (7b). The title com- 10a (0.0448 g, 2.18 equiv. TEA, 24% yield) as colourless oils. 9a

pound was prepared according to the general synthetic pro- (0.024 g, 1 equiv.) was dissolved in water and stirred with

cedure 2, from the commercially available 6-chloro purine Dowex 88 (H⁺) (ca. 0.1 g) overnight. After filtration, the solution

nucleoside (0.2 g, 1 equiv.), using Na₂PdCl₄, TPPTS, N-Boc-2- was evaporated to dryness in vacuo and the crude product puri-

pyrrole boronic acid and K₂CO₃in water (6 mL) at 100 °C. After fied by purification method 1 (0–50% MeOH against 0.05 M

24 h, the reaction was cooled down and evaporated to dryness TEAB buﬀer over 300 mL, then isocratic 60% over 100 mL,

in vacuo. The crude was purified by purification method 1 flow: 3 mL min⁻¹, fraction size: 5 mL), to aﬀord 10a as colour-

(0–50% MeOH against 0.05 M TEAB buﬀer over 350 mL, then less oil (0.0064 g, 2.1 equiv. TEA, 30% yield). ¹H NMR

isocratic 60% over 50 mL, flow: 2 mL min⁻¹, fraction size: (400 MHz, D₂O) δ: 8.86 (1H, s, H-2), 8.78 (1H, s, H-8), 8.23–8.08

5 mL) to obtain 7b as a yellow oil (0.0825 g, 37% yield). (2H, m, Ph), 7.58 (3H, m, Ph), 6.25 (1H, d, J_1′,2′= 5.3 Hz, H-1′),

¹H NMR (400 MHz, CD₃OD) δ: 8.69 (1H, s, H-2), 8.59 (1H, s, 4.68 (1H, m, H-2′), 4.51 (1H, m, H-3′), 4.40 (1H, m, H-4′), 4.14

H-8), 7.51 (1H, dd, J_5,4= 3.8 Hz, J_5,3= 1.4 Hz, H-5 pyrrole), 7.16 (2H, m, 2H-5′), 3.17 (8.2H, q, J = 7.3 Hz, CH₂TEA), 1.25 (13.9H,

(1H, dd, J_3,4= 2.5 Hz, J_3,5= 1.4 Hz, H-3 pyrrole), 6.36 (1H, dd, t, J = 7.3 Hz, CH₃TEA); ¹³C NMR (101 MHz, D₂O) δ: 156.04,

J_4,5= 3.7 Hz, J_4,3= 2.6 Hz, H-4 pyrrole), 6.09 (1H, d, J_1′,2′

=

152.66, 152.25, 145.22, 134.40, 132.36, 131.13, 130.14, 129.56,

6.0 Hz, H-1′), 4.79 (1H, apparent t, H-2′), 4.37 (1H, m, H-3′), 88.31, 84.92, 75.37, 71.21, 59.70, 47.50, 9.07; ³¹P NMR

4.19 (1H, m, H-4′), 3.91 (1H, dd, J_5′a,b= 12.4 Hz, J_5′a,4′= 2.7 Hz, (162 MHz, D₂O) δ: 2.98; m/z (ESI) 407.0754 [M − H]⁻,

H-5′a), 3.78 (1H, dd, J_5′b,a= 12.4 Hz, J_5′b,4′= 2.9 Hz, H-5′b), 2.80 C₁₆H₁₆N₄O₇P⁻requires 407.0762.

(4H, q, J = 7.2 Hz, CH₂TEA), 1.14 (5.9H, t, J = 7.3 Hz, CH₃

Phosphoric acid mono-5′-[6-(pyrrol-2-yl)-9β-^D-ribofuranosyl-

TEA); ¹³C NMR (101 MHz, CD₃OD) δ: 153.04, 151.97, 149.51, purine] (10b). The title compound was prepared according to

145.16, 129.65, 128.79, 125.20, 116.57, 111.86, 90.92, 87.88, the general synthetic procedure 3, from 8b (0.0179 g, 1 equiv.)

75.54, 72.46, 63.29, 47.30, 10.36; m/z (ESI) 316.1046 [M − H]⁻, and POCl₃(19 μL, 4 equiv.) in anhydrous acetonitrile (0.8 mL).

C₁₄H₁₄N₅O₄⁻requires 316.1051.

The reaction mixture was kept at 4 °C for 48 h. The crude was

6-Phenyl-9β-(2′,3′-O-isopropylidene-^D-ribofuranosyl)-purine purified by purification method 1 (0–50% MeOH against

(8a).²⁴The title compound was prepared according to the 0.05 M TEAB buﬀer over 350 mL, then isocratic 60% over

general synthetic procedure 1, from a suspension of 7a 50 mL, flow: 3 mL min⁻¹, fraction size: 5 mL) to give a mixture

(0.1136 g, 1 equiv.) in acetone (8 mL). The organic phase of 2′,3′-O-isopropylidene protected 9b (0.0102 g, 1.8 equiv.

aﬀorded 8a as colourless oil (0.109 g, 85% yield). ¹H NMR TEA, 33% yield) and unprotected 10b (0.0017 g, 1.8 equiv. TEA,

(400 MHz, CD₃OD) δ: 8.95 (1H, s, H-2), 8.73 (1H, s, H-8), 6% yield) as yellow oils. 9b (0.0102 g, 1 equiv.) was dissolved in

8.70–8.61 (2H, m, Ph), 7.62–7.50 (3H, m, Ph), 6.35 (1H, d, J_1′,2′

=

acetic acid solution (AcOH/H₂O, 1 : 9) at 90 °C for 3 h. After co-

3.0 Hz, H-1′), 5.42 (1H, m, H-2′), 5.09 (1H, m, H-3′), 4.40 (1H, evaporation with EtOH (3×) to remove AcOH traces, the crude

m, H-4′), 3.80 (1H, dd, J_5′a,b= 12.0 Hz, J_5′a,4′= 3.8 Hz, H-5′a), was purified by purification method 1 (0–50% MeOH against

3.73 (1H, dd, J_5′b,a= 12.0 Hz, J_5′b,4′= 4.4 Hz, H-5′b), 1.64 (3H, s, 0.05 M TEAB buﬀer over 300 mL, then isocratic 60% over

isoprop.), 1.40 (3H, s, isoprop.).

100 mL, flow: 3 mL min⁻¹, fraction size: 5 mL), to aﬀord 10b

6-(Pyrrol-2-yl)-9β-(2′,3′-O-isopropylidene-^D-ribofuranosyl)-purine as colourless oil (0.0084 g, 1.8 equiv. TEA, 88% yield). ¹H NMR

(8b). The title compound was prepared according to the (400 MHz, D₂O) δ: 8.66 (1H, s, H-2), 8.58 (1H, s, H-8), 7.26 (1H,

general synthetic procedure 1, from a suspension of 7b dd, J_5,4= 3.8 Hz, J_5,3= 1.4 Hz, H-5 pyrrole), 7.21 (1H, dd, J_3,4

=

(0.0825 g, 1 equiv.) in acetone (8 mL). The organic phase 2.5 Hz, J_5,3= 1.4 Hz, H-3 pyrrole), 6.42 (1H, dd, J_5,4= 3.8 Hz,

aﬀorded 8b as yellow oil (0.0367 g, 39% yield). ¹H NMR J_3,4= 2.6 Hz, H-4 pyrrole), 6.16 (1H, d, J_1′,2′= 5.5 Hz, H-1′), 4.75

(400 MHz, CD₃OD) δ: 8.71 (1H, s, H-2), 8.59 (1H, s, H-8), 7.51 (1H, m, H-2′), 4.52–4.47 (1H, m, H-3′), 4.40–4.35 (1H, m, H-4′),

(1H, dd, J_5,4= 3.8 Hz, J_5,3= 1.4 Hz, H-5 pyrrole), 7.15 (1H, dd, 4.18–4.06 (2H, m, 2H-5′), 3.16 (10.7H, q, J = 7.3 Hz, CH₂TEA),

J_3,4= 2.5 Hz, J_3,5= 1.4 Hz, H-3 pyrrole), 6.36 (1H, dd, J_4,5

=

1.25 (16.3H, t, J = 7.3 Hz, CH₃TEA); ¹³C NMR (101 MHz, D₂O)

3.8 Hz, J_4,3= 2.6 Hz, H-4 pyrrole), 6.27 (1H, d, J_1′,2′= 3.2 Hz, δ: 152.94, 151.51, 148.03, 144.01, 128.03, 127.34, 126.06,

H-1′), 5.38 (1H, m, H-2′), 5.08 (1H, m, H-3′), 4.39 (1H, m, H-4′), 115.89, 111.94, 88.04, 75.28, 71.24; ³¹P NMR (162 MHz, D₂O) δ:

3.80 (1H, dd, J_5′a,b= 12.0 Hz, J_5′a,4′= 3.7 Hz, H-5′a), 3.73 (1H, 0.63; m/z (ESI) 396.0714 [M − H]⁻, C₁₄H₁₅N₅O₇P⁻requires

dd, J_5′b,a= 12.0 Hz, J_5′b,4′= 4.3 Hz, H-5′b), 1.63 (3H, s, isoprop.), 396.0715.

1.40 (3H, s, isoprop.).

Phosphoric acid

Morpholin-4-yl-phosphonic acid mono-5′-[2-iodo-9β-^D-ribo-

mono-5′-[6-phenyl-9β-^D-ribofuranosyl- furanosyl-adenine] (12a). The title compound was prepared

purine] (10a). The title compound was prepared according to according to the general synthetic procedure 4, from 6a

the general synthetic procedure 3, from 8a (0.109 g, 1 equiv.) (0.009 g, 1 equiv.). The crude was purified by purification

and POCl₃(0.1 mL, 3.6 equiv.) in anhydrous acetonitrile method 1 (100% 0.05 M TEAB over 60 mL, 0–60% MeOH

(3 mL). The reaction mixture was kept at 4 °C for 12 h. The against 0.05 M TEAB buﬀer over 300 mL, isocratic 60% over

crude was purified by purification method 1 (0–50% MeOH 100 mL, flow: 3 mL min⁻¹, fraction size: 5 mL) to give 12a as

against 0.05 M TEAB buﬀer over 300 mL, isocratic 50% over colourless oil (0.0089 g, 1.1 equiv. TEA, 97% yield). ¹H NMR

Org. Biomol. Chem.

View Article Online

Organic & Biomolecular Chemistry

Paper

(400 MHz, D₂O) δ: 8.30 (1H, s, H-8), 6.01 (1H, d, J_1′,2′= 4.7 Hz, The reaction was stirred overnight. The crude was purified by

H-1′), 4.76 (1H, m, H-2′), 4.51 (1H, t, J = 4.9 Hz, H-3′), 4.31 (1H, purification method 1 (0–50% MeOH against 0.05 M TEAB

m, H-4′), 4.08–3.93 (2H, m, 2H-5′), 3.53 (4H, m, CH₂morpho- buﬀer over 360 mL, isocratic 50% over 40 mL, flow: 3 mL

line), 3.17 (6.8H, q, J = 7.3 Hz, CH₂TEA), 2.88 (4H, m, CH₂mor- min⁻¹, fraction size: 5 mL) and treated with Chelex to give 1a

pholine), 1.24 (9.8H, t, CH₃TEA); ³¹P NMR (162 MHz, D₂O) δ: as colourless oil (0.0093 g, 0.2 equiv. TEA, 53% yield). ¹H NMR

7.37; m/z (ESI) 541.0105 [M − H]⁻, C₁₄H₁₉IN₆O₇P⁻requires (400 MHz, D₂O) δ: 9.31 (1H, s, H-2N), 9.06 (1H, s, H-6N), 8.87

541.0103.

(1H, m, H-4N), 8.31 (1H, s, H-8), 8.21 (1H, m, H-5N), 6.02 (1H,

Morpholin-4-yl-phosphonic acid mono-5′-[2-phenyl-9β-^D- d, J_1″,2″= 5.0 Hz, H-1″), 5.91 (1H, d, J_1′,2′= 5.3 Hz, H-1′), 4.74

ribofuranosyl-adenine] (12b). The title compound was pre- (1H, br s, H-2′), 4.54–3.98 (9H, m, H-2″, H-3′, H-3″, H-4′, H-4″,

pared according to the general synthetic procedure 4, from 6b 2H-5′, 2H-5″), 1.24 (1.8H, t, J = 7.3 Hz, CH₃TEA); ¹³C NMR

(0.0075 g, 1 equiv.). The crude was purified by purification (126 MHz, D₂O) δ: 156.87, 147.44, 143.83, 141.65, 138.06,

method 1 (100% 0.05 M TEAB over 100 mL, 0–50% MeOH 130.41, 121.21, 101.62, 98.90, 88.71, 88.27, 85.50, 79.19, 75.49,

against 0.05 M TEAB buﬀer over 250 mL, isocratic 60% over 72.29, 71.99, 67.03, 66.57, 60.58, 9.07; ³¹P NMR (162 MHz,

50 mL, flow: 2.5 mL min⁻¹, fraction size: 5 mL) to give 12b as D₂O) δ: −11.4, −11.2; m/z (ESI) 787.9967 [M

−

2H]⁻,

colourless oil (0.0086 g, 1.1 equiv. TEA, 98% yield). ¹H NMR C₂₁H₂₅IN₇O₁₄P₂⁻requires 787.9985.

(400 MHz, D₂O) δ: 8.28 (1H, s, H-8), 8.01 (2H, m, Ph), 7.45 (3H,

P1-(2-Phenyl-adenine-9β-^D-ribofuranos-5′-yl)-P2-(nicotina-

m, Ph), 6.10 (1H, d, J_1′,2′= 4.2 Hz, H-1′), 4.87 (1H, apparent t, mide-1β-^D-ribofuranos-5′-yl) pyrophosphate (1b). The title

H-2′), 4.63 (1H, apparent t, H-3′), 4.30 (1H, m, H-4′), 4.00 (2H, compound was prepared according to the general synthetic

m, 2H-5′), 3.37 (4H, m, CH₂morpholine), 2.98 (6.6H, q, J = procedure 5, from 12b (1.7 equiv.) and β-NMN (0.0234 g,

7.3 Hz, CH₂TEA), 2.76 (4H, m, CH₂morpholine), 1.16 (10.1H, t, 1 equiv.). The reaction was stirred overnight. The crude was

J = 7.3 Hz, CH₃TEA); ³¹P NMR (162 MHz, D₂O) δ: 7.42; m/z purified by purification method 1 (0–50% MeOH against

(ESI) 491.1452 [M − H]⁻, C₂₀H₂₄N₆O₇P⁻requires 491.1450.

0.05 M TEAB buﬀer over 400 mL, flow: 3 mL min⁻¹, fraction

Morpholin-4-yl-phosphonic acid mono-5′-[6-phenyl-9β-^D- size: 5 mL) to give 1b as colourless oil (0.017 g, 1.18 equiv.

ribofuranosyl-purine] (13a). The title compound was prepared TEA, 28% yield). ¹H NMR (400 MHz, D₂O) δ: 9.04 (1H, br s,

according to the general synthetic procedure 4, from 10a H-2N), 8.83 (1H, br s, H-6N), 8.61 (1H, m, H-4N), 8.09–7.86

(0.0191 g, 1 equiv.). The crude was purified by purification (3H, m, H-5N, H-8, Ph), 7.45 (4H, m, Ph), 6.08 (1H, br s, H-1″),

method 1 (100% 0.05 M TEAB over 100 mL, 0–50% MeOH 5.76 (1H, br s, H-1′), 4.84 (1H, br s, H-2′), 4.56 (1H, br s, H-2″),

against 0.05 M TEAB buﬀer over 250 mL, isocratic 60% over 4.43–4.01 (6H, m, H-3′, H-3″, H-4′, H-4″, 2H-5′, 2H-5″), 3.16

50 mL, flow: 3 mL min⁻¹, fraction size: 5 mL) to give 13a as (7H, q, J = 7.2 Hz, CH₂TEA), 1.22 (10.6H, t, J = 7.2 Hz, CH₃

colourless oil (0.0172 g, 1.2 equiv. TEA, 82% yield). ¹H NMR TEA); ¹³C NMR (101 MHz, D₂O) δ: 146.30, 142.68, 140.55,

(400 MHz, D₂O) δ: 8.78 (1H, s, H-2), 8.68 (1H, s, H-8), 8.05 (2H, 138.29, 137.55, 137.23, 135.59, 131.55, 131.52, 129.56, 129.49,

m, Ph), 7.58–7.45 (3H, m, Ph), 6.16 (1H, d, J_1′,2′= 4.6 Hz, H-1′), 128.80, 100.61, 93.30, 87.75, 84.39, 78.37, 74.32, 71.26, 65.04,

4.53 (1H, t, J = 4.9 Hz, H-3′), 4.35 (1H, m, H-4′), 4.12–3.96 (2H, 53.17, 52.17, 47.51, 9.07; ³¹P NMR (162 MHz, D₂O) δ: −11.5,

−

m, 2H-5′), 3.48 (4H, m, CH₂morpholine), 3.16 (7.4H, q, J = −11.7; m/z (ESI) 738.1324 [M − H]⁻, C₂₇H₃₀N₇O₁₄P₂requires

7.3 Hz, CH₂TEA), 2.87 (4H, m, CH₂morpholine), 1.24 (11.6H, t, 738.1331.

J = 7.2 Hz, CH₃TEA); ³¹P NMR (162 MHz, D₂O) δ: 7.40; m/z

P1-(6-Phenyl-purine-9β-^D-ribofuranos-5′-yl)-P2-(nicotinamide-

1β-^D-ribofuranos-5′-yl) pyrophosphate (2a). The title com-

(ESI) 476.1339 [M − H]⁻, C₂₀H₂₃N₅O₇P⁻requires 476.1341.

Morpholin-4-yl-phosphonic acid mono-5′-[6-(pyrrol-2-yl)- pound was prepared according to the general synthetic pro-

9β-^D-ribofuranosyl-purine] (13b). The title compound was pre- cedure 5, from 13a (0.019 g, 1 equiv.) and β-NMN (1.8 equiv.).

pared according to the general synthetic procedure 4, from 10b The reaction was stirred overnight. The crude was purified

(0.015 g, 1 equiv.). The crude was purified by purification twice by purification method 1 (0–50% MeOH against 0.05 M

method 1 (100% 0.05 M TEAB over 100 mL, 0–50% MeOH TEAB buﬀer over 400 mL, flow: 3 mL min⁻¹, fraction size:

against 0.05 M TEAB buﬀer over 250 mL, isocratic 60% over 5 mL) to give 2a as colourless oil (0.016 g, 1.5 equiv. TEA, 56%

50 mL, flow: 3 mL min⁻¹, fraction size: 5 mL) to give 13b as yield). ¹H NMR (400 MHz, D₂O) δ: 9.20 (1H, s, H-2N), 9.03 (1H,

colourless oil (0.0132 g, 1.1 equiv. TEA, 88% yield). ¹H NMR d, J_6,5= 6.3 Hz, H-6N), 8.82 (1H, s, H-2), 8.79 (1H, s, H-8), 8.67

(400 MHz, D₂O) δ: 8.54 (1H, s, H-2), 8.52 (1H, s, H-8), 7.19 (2H, (1H, d, J_4,5= 8.2 Hz, H-4N), 8.11–8.09 (2H, m, Ph), 8.05 (1H, m,

m, H-5, H-3 pyrrole), 6.39 (1H, s, H-4 pyrrole), 6.09 (1H, d, H-5N), 7.57 (3H, m, Ph), 6.22 (1H, d, J_1″,2″= 5.7 Hz, H-1″), 5.90

J_1′,2′= 4.6 Hz, H-1′), 4.52 (1H, t, J = 4.9 Hz, H-3′), 4.34 (1H, m, (1H, d, J = 4.7 Hz, H-1′), 4.55–4.09 (10H, m, H-2′, H-2″, H-3′,

H-4′), 4.11–3.92 (2H, m, 2H-5′), 3.49 (4H, m, CH₂morpholine), H-3″, H-4′, H-4″, 2H-5′, 2H-5″), 3.13 (8.7H, q, J = 7.3 Hz, CH₂

3.15 (6.3H, q, J = 7.3 Hz, CH₂TEA), 2.85 (4H, m, CH₂morpho- TEA), 1.24 (13.3H, t, J = 7.2 Hz, CH₃TEA); ¹³C NMR (126 MHz,

line), 1.23 (10.1H, t, J = 7.3 Hz, CH₃TEA); ³¹P NMR (162 MHz, D₂O) δ: 162.12, 156.90, 153.67, 153.43, 147.40, 146.20, 144.11,

D₂O) δ: 7.41.

141.19, 138.05, 135.24, 133.34, 132.02, 131.14, 131.00, 130.54,

P1-(2-Iodo-adenine-9β-^D-ribofuranos-5′-yl)-P2-(nicotinamide- 130.10, 101.56, 98.89, 88.75, 85.62, 79.13, 75.74, 72.09, 67.00,

1β-^D-ribofuranos-5′-yl) pyrophosphate (1a). The title com- 66.49, 60.58, 48.24, 9.07; ³¹P NMR (162 MHz, D₂O) δ: −11.4 (d,

pound was prepared according to the general synthetic pro- J_P,P= 20.3 Hz), −11.7 (d, J_P,P= 20.7 Hz); m/z (ESI) 723.1210

cedure 5, from 12a (0.0134 g, 1 equiv.) and β-NMN (2.4 equiv.). [M − 2H]⁻, C₂₇H₂₉N₆O₁₄P₂⁻requires 723.1222.

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

P1-[6-(Pyrrol-2-yl)-purine-9β-D-ribofuranos-5′-yl]-P2-[nicotina- centrations). For the inhibition and the non-natural co-sub-

mide-1β-^D-ribofuranos-5′-yl] pyrophosphate (2b). The title strate activity experiments of C-8 substituted derivatives, DNA

compound was prepared according to the general synthetic substrate (0.125 μM), C-8 substituted derivatives (250 μM),

procedure 5, from 13b (0.0084 g, 1 equiv.) and β-NMN with/without β-NAD⁺(26 μM), were mixed with an appropriate

(2.1 equiv.). The reaction was stirred overnight. The crude was volume of ligation buﬀer (30 mM Tris/HCl pH 8, 4 mM MgCl₂,

purified by purification method 1 (0–50% MeOH against 0.05 1 mM DTT, 50 μg mL⁻¹BSA); the reactions were started with

M TEAB buﬀer over 400 mL, flow: 3 mL min⁻¹, fraction size: enzyme addition (EcLigA at 0.07 μM, or MtLigA at 0.17 μM)

5 mL) to give 2b as colourless oil (0.0132 g, 1.3 equiv. TEA, and incubated at 30 °C under shaking (15 μL final volume, all

87% yield). ¹H NMR (400 MHz, D₂O) δ: 9.11 (1H, s, H-2N), 8.94 concentrations are final concentrations). In all the set of

(1H, d, J_6,5= 5.1 Hz, H-6N), 8.69 (1H, s, H-2), 8.61 (1H, s, H-8), experiments, a control reaction was included to test enzyme

8.57 (1H, d, J = 7.7 Hz, H-4N), 7.90 (1H, m, H-5N), 7.23 (1H, s, activity in the presence of only β-NAD⁺, not containing any

H-5 pyrrole), 7.20 (1H, d, J_3,4= 3.1 Hz, H-3 pyrrole), 6.43 (1H, tested compound. Samples were drawn at diﬀerent time points

m, H-4 pyrrole), 6.17 (1H, d, J_1″,2″= 5.8 Hz, H-1″), 5.91 (1H, d, and mixed with an equal volume of 1× stop solution (deionized

J_1′,2′= 4.4 Hz, H-1′), 4.60–4.12 (10H, m, H-2′, H-2″, H-3′, H-3″, formamide, 0.5 M EDTA, 1 M NaOH, 20 mg bromophenol

H-4′, H-4″, 2H-5′, 2H-5″), 3.18 (6.2H, q, J = 7.3 Hz, CH₂TEA), blue), then heated at 95 °C for 5 min. Afterwards, samples

1.25 (12H, t, J = 7.3 Hz, CH₃TEA); ¹³C NMR (126 MHz, D₂O) δ: were run on a pre-warmed 15% polyacrylamide/urea denatur-

153.21, 151.85, 147.85, 146.39, 144.16, 143.33, 143.31, 140.35, ing gel; a potential of 400 V was applied for 20 min in 1 × TBE

129.02, 127.81, 127.11, 126.18, 116.04, 112.16, 101.03, 98.14, buﬀer (45 mM Tris/borate pH 8.3, 1.25 mM EDTA). The gels

87.89, 87.59, 85.07, 78.47, 74.90, 71.57, 71.37, 59.83, 47.50, were visualised using

a

Molecular Dynamics Storm

9.07; ³¹P NMR (162 MHz, D₂O) δ: −11.4, −11. 6; m/z (ESI) phosphorimager.

714.1311 [M]⁺, C₂₅H₃₀N₇O₁₄P₂⁺requires 714.1320.

Molecular docking

Enzymology: general conditions

All calculations were performed on an Intel Core™ i5 @ 2.30

All reagents used for biochemical applications were of analyti- GHz, Sony Vaio. The crystal structure of EcLigA in complex

cal quality and solutions prepared with Milli-Q water. EcLigA with nicked DNA-adenylate was obtained from the RCSB

from E. coli and MtLigA from M. tuberculosis were expressed Protein Data Bank (PDB code 2OWO). The protein was pre-

and purified as described with minor modifications.²¹The pared using UCSF Chimera.²⁵Explicit hydrogens were intro-

His-tag binding column (IDA-based His-Bind Resin Ni²⁺pre- duced on the protein, and protonation of the residues within

charged) and its buﬀer kit were purchased from Novagen. the active site was evaluated for consistency with physiological

Protein desalting was performed using a PD-10 Column pH. Hydrogen positions were minimized using Amber ﬀ12SB²⁶

(Sephadex G-25 Medium) and buﬀer (20 mM Tris, 200 mM (100 steps of steepest descent, 100 steps of conjugate gradient,

NaCl, pH 7.5). After this step glycerol was added to eluted step size 0.02 A). Crystallographic water molecules and ions

sample to 20% and the purified protein was stored at −80 °C. were removed. Pyrophosphate bond between DNA and AMP

To quantify the protein concentration according to the Brad- was broken, and DNA was removed. AMP residue was saved as

ford method, the Bio-Rad Protein Assay, consisting of Coomas- PDB file for subsequent modelling and reference and even-

sie Brilliant Blue G-250 in acidic solution, was used with tually removed from the model. The final, clean, protein was

absorbance reading at 595 nm. Absorbance readings were per- exported as pdb file. AutoDockTools4 ²⁷was used to process

formed in NUNC 96 plates on a BMG labtech PolarStar micro- the protein model and conversion to pdbqt for use in Auto-

plate reader equipped with suitable absorbance filters. The Dock Vina.²⁸Ligands were prepared by modification of the

oligonucleotides for the ligation assay were supplied by Euro- crystallographic AMP structure: MarvinSketch²⁹was used to

gentec. For fluorescence imaging, a Molecular Dynamics introduce the required structural modification and the modi-

Storm Phosphorimager was used.

fied part was minimized in UCSF Chimera (GAFF force

field,^30,31Gasteiger charges,³¹to convergence – 0.01 kJ mol⁻¹

nm⁻¹). AutoDockTools4 was used to convert and prepare the

Ligation assay

Separate stock solutions of EcLigA, MtLigA, DNA substrate, flexible ligands. The bonds around the ribose unit were

β-NAD⁺and the synthesized compounds were prepared in allowed to rotate to accommodate repositioning of nucleobase,

Milli-Q water. For the inhibition and the non-natural co-sub- alkylphosphate and hydrogens of the hydroxyl groups. Where

strate activity experiments of C-2 and C-6 substituted deriva- the substituent includes a rotable bond, rotations were

tives, DNA substrate (0.5 μM), C-2 substituted derivatives allowed. The ligands were exported as pdbqt files. Dockings

(200 μM), or C-6 substituted derivatives (250 μM with EcLigA, were performed within a cube centred on x = 27.445, y =

200 μM with MtLigA), with/without β-NAD⁺(26 μM), were 7.4025, z = 6.273 and having sides of 17.5 Å. Exhaustiveness

mixed with an appropriate volume of ligation buﬀer (30 mM parameter was set to 100. For validation, AMP was re-docked

Tris/HCl pH 8, 4 mM MgCl₂, 1 mM DTT, 50 μg mL⁻¹BSA); the after randomization of rotable bond’s angles. After docking,

reactions were started with enzyme addition (EcLigA at the resulting poses are evaluated in respect of key interactions

0.07 μM, or MtLigA at 0.17 μM) and incubated at 30 °C under and orientation within the active site. RMSD of re-docked AMP

shaking (15 μL final volume, all concentrations are final con- was 0.702 Å.

Org. Biomol. Chem.

View Article Online

Organic & Biomolecular Chemistry

Paper

C. Schmitt, G. Chapoux, E. Ilhan, N. Ekambaram,

A. Athanasiou, A. Knezevic, D. Sabato, A. Chambovey,

M. Gaertner, M. Enderlin, M. Boehme, V. Sippel and

P. Wyss, Structure-guided design, synthesis and biological

evaluation of novel DNA ligase inhibitors with in vitro and

in vivo anti-staphylococcal activity, Bioorg. Med. Chem. Lett.,

2012, 22(21), 6705–6711; (d) W. Gu, T. Wang, F. Maltais,

B. Ledford, J. Kennedy, Y. Wei, C. H. Gross, J. Parsons,

L. Duncan, S. J. Ryan Arends, C. Moody, E. Perola, J. Green

and P. S. Charifson, Design, synthesis and biological evalu-

ation of potent NAD⁺-dependent DNA ligase inhibitors as

potential antibacterial agents. Part I: Aminoalkoxypyrimi-

dine carboxamides, Bioorg. Med. Chem. Lett., 2012, 22(11),

3693–3698.

7 (a) J. Y. Lee, C. Chang, H. K. Song, J. Moon, J. K. Yang,

H. K. Kim, S. T. Kwon and S. W. Suh, Crystal structure of

NAD⁺-dependent DNA ligase: modular architecture and

functional implications, EMBO J., 2000, 19(5), 1119–1129;

(b) K. S. Gajiwala and C. Pinko, Structural Rearrangement

Accompanying NAD⁺Synthesis within a Bacterial DNA

Ligase Crystal, Structure, 2004, 12(8), 1449–1459;

(c) S. K. Srivastava, R. P. Tripathi and R. Ramachandran,

NAD⁺-dependent DNA Ligase (Rv3014c) from Mycobacter-

ium tuberculosis: crystal structure of the adenylation

domain and identification of novel inhibitors, J. Biol.

Chem., 2005, 280(34), 30273–30281; (d) J. Nandakumar,

P. A. Nair and S. Shuman, Last Stop on the Road to Repair:

Structure of E. coli DNA Ligase Bound to Nicked

DNA-Adenylate, Mol. Cell, 2007, 26(2), 257–271.

Acknowledgements

We thank the UEA for a PhD studentship to Giulia Pergolizzi,

the BBSRC Institute Strategic Programme Grant on Under-

standing and Exploiting Metabolism (MET) [BB/J004561/1] and

the John Innes Foundation for funding. The EPSRC National

Mass Spectrometry Centre Swansea is acknowledged for the

recording of mass spectra. NMR experiments were produced in

part using the facilities of the Centre for Biomolecular Spec-

troscopy, King’s College London, established with a Capital

Award from the Wellcome Trust. We thank Dr R.A. Atkinson

for assistance with these experiments, and Hannah Charlton

for the purification of MtLigA. Molecular graphics and ana-

lyses were performed with the UCSF Chimera package.

Chimera is developed by the Resource for Biocomputing, Visu-

alization, and Informatics at the University of California,

San Francisco (supported by NIGMS P41-GM103311).

Notes and references

1 S. Shuman, DNA Ligases: Progress and Prospects, J. Biol.

Chem., 2009, 284(26), 17365–17369.

2 A. Wilkinson, J. Day and R. Bowater, Bacterial DNA ligases,

Mol. Microbiol., 2001, 40(6), 1241–1248.

3 N. Dwivedi, D. Dube, J. Pandey, B. Singh, V. Kukshal,

R. Ramachandran and R. P. Tripathi, NAD⁺-Dependent

DNA Ligase: A novel target waiting for the right inhibitor,

Med. Res. Rev., 2008, 28(4), 545–568.

4 M. Korycka-Machala, E. Rychta, A. Brzostek, H. R. Sayer,

A. Rumijowska-Galewicz, R. P. Bowater and J. Dziadek,

Evaluation of NAD⁺-Dependent DNA Ligase of Mycobacteria

as a Potential Target for Antibiotics, Antimicrob. Agents

Chemother., 2007, 51(8), 2888–2897.

5 (a) S. D. Podos, J. A. Thanassi and M. J. Pucci, Mechanistic

Assessment of DNA Ligase as an Antibacterial Target in Sta-

phylococcus aureus, Antimicrob. Agents Chemother., 2012,

56(8), 4095–4102; (b) G. Sanyal and P. Doig, Bacterial DNA

replication enzymes as targets for antibacterial drug

discovery, Expert Opin. Drug Discovery, 2012, 7(4), 327–

339.

6 (a) K. Murphy-Benenato, H. Wang, H. M. McGuire,

H. E. Davis, N. Gao, D. B. Prince, H. Jahic, S. S. Stokes and

P. A. Boriack-Sjodin, Identification through structure-based

methods of a bacterial NAD⁺-dependent DNA ligase inhibi-

tor that avoids known resistance mutations, Bioorg. Med.

Chem. Lett., 2014, 24(1), 360–366; (b) S. Howard, N. Amin,

A. B. Benowitz, E. Chiarparin, H. Cui, X. Deng,

T. D. Heightman, D. J. Holmes, A. Hopkins, J. Huang,

Q. Jin, C. Kreatsoulas, A. C. L. Martin, F. Massey,

L. McCloskey, P. N. Mortenson, P. Pathuri, D. Tisi and

P. A. Williams, Fragment-Based Discovery of 6-Azaindazoles

As Inhibitors of Bacterial DNA Ligase, ACS Med. Chem.

Lett., 2013, 4(12), 1208–1212; (c) J.-P. Surivet, R. Lange,

C. Hubschwerlen, W. Keck, J.-L. Specklin, D. Ritz, D. Bur,

H. Locher, P. Seiler, D. S. Strasser, L. Prade, C. Kohl,

8 S. D. Lahiri, R.-F. Gu, N. Gao, I. Karantzeni, G. K. Walkup

and S. D. Mills, Structure Guided Understanding of NAD⁺

Recognition in Bacterial DNA Ligases, ACS Chem. Biol.,

2012, 7(3), 571–580.

9 (a) S. D. Mills, A. E. Eakin, E. T. Buurman, J. V. Newman,

N. Gao, H. Huynh, K. D. Johnson, S. Lahiri, A. B. Shapiro,

G. K. Walkup, W. Yang and S. S. Stokes, Novel Bacterial

NAD⁺-Dependent DNA Ligase Inhibitors with Broad-Spec-

trum Activity and Antibacterial Eﬃcacy In Vivo, Antimicrob.

Agents Chemother., 2011, 55(3), 1088–1096; (b) S. S. Stokes,

H. Huynh, M. Gowravaram, R. Albert, M. Cavero-Tomas,

B. Chen, J. Harang, J. T. Loch III, M. Lu, G. B. Mullen,

S. Zhao, C.-F. Liu and S. D. Mills, Discovery of bacterial

NAD⁺-dependent DNA ligase inhibitors: Optimization of

antibacterial activity, Bioorg. Med. Chem. Lett., 2011, 21(15),

4556–4560; (c) S. S. Stokes, M. Gowravaram, H. Huynh,

M. Lu, G. B. Mullen, B. Chen, R. Albert, T. J. O’Shea,

M. T. Rooney, H. Hu, J. V. Newman and S. D. Mills, Discov-

ery of bacterial NAD⁺-dependent DNA ligase inhibitors:

Improvements in clearance of adenosine series, Bioorg.

Med. Chem. Lett., 2012, 22(1), 85–89; (d) L. Miesel,

C. Kravec, A.-T. Xin, P. McMonagle, S. Ma, J. Pichardo,

B. Feld, E. Barrabee and R. Palermo, A high-throughput

assay for the adenylation reaction of bacterial DNA ligase,

Anal. Biochem., 2007, 366(1), 9–17; (e) E. T. Buurman,

V. A. Laganas, C. F. Liu and J. I. Manchester, Antimicrobial

Activity of Adenine-Based Inhibitors of NAD⁺-Dependent

Org. Biomol. Chem.

View Article Online

Paper

Organic & Biomolecular Chemistry

DNA Ligase, ACS Med. Chem. Lett., 2012, 3(8), 663–667;

Bull. Chem. Soc. Jpn., 1975, 48(7), 2084–2090; (b) A. Collier

and G. Wagner, A facile two-step synthesis of 8-arylated

guanosine mono- and triphosphates (8-aryl GXPs), Org.

Biomol. Chem., 2006, 4(24), 4526–4532.

(f) H. Jahić, C. F. Liu, J. Thresher, S. Livchak, H. Wang and

D. E. Ehmann, The kinetic mechanism of S. pneumoniae

DNA ligase and inhibition by adenosine-based antibacterial

compounds, Biochem. Pharmacol., 2012, 84(5), 654– 17 T. Pesnot, J. Kempter, J. r. Schemies, G. Pergolizzi,

660.

U. Uciechowska, T. Rumpf, W. Sippl, M. Jung and

G. K. Wagner, Two-Step Synthesis of Novel, Bioactive

Derivatives of the Ubiquitous Cofactor Nicotinamide

Adenine Dinucleotide (NAD), J. Med. Chem., 2011, 54(10),

3492–3499.

10 G. Pergolizzi, J. N. Butt, R. P. Bowater and G. K. Wagner, A

novel fluorescent probe for NAD-consuming enzymes,

Chem. Commun., 2011, 47(47), 12655–12657.

11 (a) J. Lee, H. Churchil, W.-B. Choi, J. E. Lynch,

F. E. Roberts, R. P. Volante and P. J. Reider, A chemical syn- 18 T. Mukaiyama and M. Hashimoto, Phosphorylation by oxi-

thesis of nicotinamide adenine dinucleotide (NAD⁺), Chem.

Commun., 1999, 729–730; (b) N. E. Batoux, F. Paradisi,

P. C. Engel and M. E. Migaud, Novel nicotinamide adenine

dation-reduction condensation: Preparation and reaction

of S-(2-pyridyl) phosphorothioates, Tetrahedron Lett., 1971,

12(26), 2425–2428.

dinucleotide analogues as selective inhibitors of NAD⁺- 19 (a) C. D. Jardetzky, Proton Magnetic Resonance Studies on

dependent enzymes, Tetrahedron, 2004, 60(31), 6609–6617;

(c) B. Zhang, G. K. Wagner, K. Weber, C. Garnham,

A. J. Morgan, A. Galione, A. H. Guse and B. V. L. Potter,

2′-Deoxy Cyclic Adenosine 5′-Diphosphate Ribose

Derivatives: Importance of the 2′-Hydroxyl Motif for the

Antagonistic Activity of 8-Substituted cADPR Derivatives,

J. Med. Chem., 2008, 51(6), 1623–1636.

Purines, Pyrimidines, Ribose Nucleosides and Nucleotides.

III. Ribose Conformation1, J. Am. Chem. Soc., 1960, 82(1),

229–233; (b) M. Ikehara, S. Uesugi and K. Yoshida, Nucleo-

sides and nucleotides. XLVII. Conformation of purine

nucleosides and their 50-phosphates, Biochemistry, 1972,

11, 830–836; (c) R. H. Sarma, H. C. Lee, F. E. Evans,

N. Yathindra and M. Sundaralingam, Probing the inter-

relation between the glycosyl torsion, sugar pucker, and the

backbone conformation in C(8) substituted adenine

nucleotides by proton and proton_phosphorus-31 fast

Fourier transfer nuclear magnetic resonance methods

and conformational energy calculations, J. Am. Chem. Soc.,

1974, 96, 7337–7348; (d) S. Uesugi and M. Ikehara, Carbon-

13 magnetic resonance spectra of 8-substituted purine

nucleosides. Characteristic shifts for the syn conformation,

J. Am. Chem. Soc., 1977, 99, 3250–3253; (e) C. Altona and

M. Sundaralingam, Conformational analysis of the

sugar ring in nucleosides and nucleotides. Improved

method for the interpretation of proton magnetic reson-

ance coupling constants, J. Am. Chem. Soc., 1973, 95, 2333–

2344.

12 (a) Y. Xu, H. Jin, Z. Yang, L. Zhang and L. Zhang, Synthesis

and biological evaluation of novel neamine–nucleoside

conjugates potentially targeting to RNAs, Tetrahedron,

2009, 65(27), 5228–5239; (b) H. Homma, Y. Watanabe,

T. Abiru, T. Murayama, Y. Nomura and A. Matsuda, Nucleo-

sides and nucleotides. 112. 2-(1-Hexyn-1-yl)adenosine-5′-

uronamides: a new entry of selective A2 adenosine receptor

agonists with potent antihypertensive activity, J. Med.

Chem., 1992, 35(15), 2881–2890.

13 (a) L. Chen, G. Gao, K. Felczak, L. Bonnac, S. E. Patterson,

D. Wilson, E. M. Bennett, H. N. Jayaram and L. Hedstrom,

Probing Binding Requirements of Type I and Type II Iso-

forms of Inosine Monophosphate Dehydrogenase with

Adenine-Modified Nicotinamide Adenine Dinucleotide

Analogues, J. Med. Chem., 2007, 50(23), 5743–5751; 20 (a) C. F. G. C. Geraldes and H. Santos, Solvent eﬀects on

(b) J. Neres, N. P. Labello, R. V. Somu, H. I. Boshoﬀ,

D. J. Wilson, J. Vannada, L. Chen, C. E. Barry,

E. M. Bennett and C. C. Aldrich, Inhibition of Siderophore

Biosynthesis in Mycobacterium tuberculosis with Nucleo-

side Bisubstrate Analogues: Structure–Activity Relation-

ships of the Nucleobase Domain of 5′-O-[N-(Salicyl)-

sulfamoyl]adenosine, J. Med. Chem., 2008, 51(17), 5349–

5370.

the conformation of nucleotides. Part 1. The conformation

of 5′-adenosine monophosphate in water-dimethyl sulph-

oxide using nuclear Overhauser eﬀects and lanthanide

relaxation probes, J. Chem. Soc., Perkin Trans. 2, 1983, 9,

1693–1697; (b) R. Stolarski, L. Dudycz and D. Shugar, NMR

Studies on the syn-anti Dynamic Equilibrium in Purine

Nucleosides and Nucleotides, Eur. J. Biochem., 1980, 108(1),

111–121.

14 H. Gold, P. van Delft, N. Meeuwenoord, J. D. C. Codée, 21 (a) A. Wilkinson, A. Smith, D. Bullard, M. Lavesa-Curto,

D. V. Filippov, G. Eggink, H. S. Overkleeft and G. A. van der

Marel, Synthesis of Sugar Nucleotides by Application of

Phosphoramidites, J. Org. Chem., 2008, 73(23), 9458–9460.

15 A. Collier and G. K. Wagner, Suzuki–Miyaura Cross-Coup-

ling of Unprotected Halopurine Nucleosides in Water—

Influence of Catalyst and Cosolvent, Synth. Commun., 2006,

36(24), 3713–3721.

H. Sayer, A. Bonner, A. Hemmings and R. Bowater, Analysis

of ligation and DNA binding by Escherichia coli DNA ligase

(LigA), Biochim. Biophys. Acta, Proteins Proteomics, 2005,

1749(1), 113–122; (b) A. Wilkinson, H. Sayer, D. Bullard,

A. Smith, J. Day, T. Kieser and R. Bowater, NAD⁺-dependent

DNA ligases of Mycobacterium tuberculosis and Strepto-

myces coelicolor, Proteins: Struct., Funct., Bioinf., 2003, 51

(3), 321–326.

16 (a) T. Sowa and S. Ouchi, The Facile Synthesis of 5′-Nucleo-

tides by the Selective Phosphorylation of

a Primary 22 B. O. S. Scott, M. Lavesa-Curto, D. R. Bullard, J. N. Butt and

Hydroxyl Group of Nucleosides with Phosphoryl Chloride,

R. P. Bowater, Immobilized DNA hairpins for assay of

Org. Biomol. Chem.

View Article Online

Organic & Biomolecular Chemistry

Paper

sequential breaking and joining of DNA backbones, Anal.

Biochem., 2006, 358(1), 90–98.

23 J. M. Pascal, DNA and RNA ligases: structural variations

and shared mechanisms, Curr. Opin. Struct. Biol., 2008,

18(1), 96–105.

Ferrer, G. Seabra, C. L. Simmerling, W. Smith, J. Swails,

R. C. Walker, J. Wang, R. M. Wolf, X. Wu and

P. A. Kollman, AMBER 14, University of California,

San Francisco, 2014.

27 G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner,

R. K. Belew, D. S. Goodsell and A. J. Olson, AutoDock4 and

AutoDockTools4: Automated docking with selective recep-

tor flexibility, J. Comput. Chem., 2009, 30(16), 2785–2791.

24 M. Martín-Ortíz, M. Gómez-Gallego, C. Ramírez de Arellano

and M. A. Sierra, The Selective Synthesis of Metallanucleo-

sides and Metallanucleotides:

A New Tool for the

Functionalization of Nucleic Acids, Chem. – Eur. J., 2012, 28 O. Trott and A. J. Olson, AutoDock Vina: improving the

18(40), 12603–12608.

25 E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch,

D. M. Greenblatt, E. C. Meng and T. E. Ferrin, UCSF

speed and accuracy of docking with a new scoring function,

eﬃcient optimization and multithreading, J. Comput.

Chem., 2010, 31(2), 455–461.

Chimera—A visualization system for exploratory research 29 http://www.chemaxon.com.

and analysis, J. Comput. Chem., 2004, 25(13), 1605–1612. 30 J. Wang, W. Wang, P. A. Kollman and D. A. Case, Automatic

http://www.cgl.ucsf.edu/chimera.

26 D. A. Case, V. Babin, J. T. Berryman, R. M. Betz, Q. Cai,

D. S. Cerutti, T. E. Cheatham III, T. A. Darden, R. E. Duke,

atom type and bond type perception in molecular mechan-

ical calculations, J. Mol. Graphics Modell., 2006, 25(2), 247–

260.

H. Gohlke, A. W. Goetz, S. Gusarov, N. Homeyer, 31 J. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and

P. Janowski, J. Kaus, I. Kolossváry, A. Kovalenko, T. S. Lee,

S. LeGrand, T. Luchko, R. Luo, B. Madej, K. M. Merz,

F. Paesani, D. R. Roe, A. Roitberg, C. Sagui, R. Salomon-

D. A. Case, Development and testing of a general

amber force field, J. Comput. Chem., 2004, 25(9), 1157–

1174.

Org. Biomol. Chem.

Article Doi

Base-modified NAD and AMP derivatives and their activity against bacterial DNA ligases

DOI: 10.1039/c5ob00294j

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c5ob00294j

Full text of DOI:10.1039/c5ob00294j

Products guided by the article

R&D Labs maybe for 51554-46-2

Relevant to this article

Hot Product