Design of a fused triazolyl 2-quinolinone unnatural nucleoside via tandem CuAAC-Ullmann coupling reaction and study of photophysical property

Article abstract of DOI:10.1016/j.tet.2018.03.037

This report presents the design and synthesis of a novel fused triazolyl 2-quinolinone (FTQuon) nucleoside as a new generation of angularly widened unnatural nucleobase surrogate with two possible H-bonding faces-one H-bond acceptor and another donor. The

Full text of DOI:10.1016/j.tet.2018.03.037

Accepted Manuscript
Design of a fused triazolyl 2-quinolinone unnatural nucleoside via tandem CuAAC-
Ullmann coupling reaction and study of photophysical property
Subhendu Sekhar Bag, Suman Kalyan Das, Hiranya Gogoi
PII:
S0040-4020(18)30302-8
10.1016/j.tet.2018.03.037
TET 29377
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 26 January 2018
Revised Date: 15 March 2018
Accepted Date: 17 March 2018
Please cite this article as: Bag SS, Das SK, Gogoi H, Design of a fused triazolyl 2-quinolinone unnatural
nucleoside via tandem CuAAC-Ullmann coupling reaction and study of photophysical property,
Tetrahedron (2018), doi: 10.1016/j.tet.2018.03.037.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Design of a Fused Triazolyl 2-Quinolinone Unnatural Nucleoside
via Tandem CuAAC-Ullmann Coupling Reaction and Study of
Photophysical Property
Subhendu Sekhar Bag,* Suman Kalyan Das and Hiranya Gogoi
†Dedicated to Professor Isao Saito on his 77th Birth day
Bioorganic Chemistry Laboratory, Department of Chemistry, Indian Institute of Technology
Guwahati,North Guwhati-781039, Assam, India. Fax: +91-361-258-2349;Tel: +91-361-258-
2324.E-mail:ssbag75@iitg.ernet.in
Abstract: This report presents the design and synthesis of a novel fused triazolyl 2-
quinolinone (FTQuon) nucleoside as a new generation of angularly widened unnatural
nucleobase surrogate with two possible H-bonding faces-one H-bond acceptor and
another donor. The synthesis via a tandem CuAAC-Ullmann coupling, the study of
photophysical properties and theoretical calculation in the context of DNA are the main
contents of this report. The newly designed nucleoside shows interesting photphysical
property with slight blue shifted solvatochromicity. It also shows pH sensitive emission.
All the theoretical DNA duplexes containing the FTQuon show right−handed B-form
1

ACCEPTED MANUSCRIPT
helicity as revealed from a molecular dynamics simulation using Schrodinger
Macromodel. A theoretical (DFT) study indicates a good stabilizing property of
FTQuon via pairing with natural pyrimidine bases. It also shows good interaction
property with BSA protein signalled via a switch on fluorescence response.
Keywords: Fused triazolylquinolinone nucleoside• Fluorescent •Tandem CuAAC-Ullmann
coupling• widen nucleo base• BSA interaction• spectroscopy•
Introduction
The Nature’s power of delegation for encoding the complex information necessary
for life is given to DNA orchestrated only by H-bonded two base pairs.1a However, the
motivation toward enhancement of functional ability has spurred the research to
expand the genetic alphabets via the design of unnatural nucleobases by mimicking the
natural H-bonded base pairing.1 The pioneering work by Professor Alex Rich and later
on by Prof. Steven A. Benner have led to expand the genetic alphabet from four to six
letters featuring orthogonal H-bonding complementarity.1 Since then an increasing
amount of research works has resulted in the acceleration of progress towards
expanding the genetic alphabets via the design and synthesis of unnatural H-bonded
base pairs.2-3
The creation of non-H-bonded unnatural nucleobase surrogates by Kool et al., has
opened a new dimension in the design of hydrophobic DNA base analogues.4a-c Since
then, much efforts have been put forth to develop non-natural, stable, hydrophobic
nucleosides.4d-j Many of such hydrophobic base pairs are found to be recognized by
polymerase enzymes creating an increased numbers of genetic alphabets. In recent
time, the development of size expanded DNA bases as well as widened base pairs has
impacted greatly in the field of design of hydrophobic as well as H-bonded
2

ACCEPTED MANUSCRIPT
nucleosides.5 Furthermore, the lack of practical fluorescence property of the natural
nucleobases and harnessing for the nucleic acid based diagnostics and sensing
materials, have greatly drawn interest, in recent time, toward the design of unnatural
DNA base pairs with tuned photophysical properties.6a-c Toward this end, several
unnatural nucleosides have appeared in the literatures for the development of
functional nucleic acids.6d-f
Therefore, the attractive forces like H-bonding, aromatic stacking, hydrophobic or
CH-π interactions have well been explored in the context of design of isomorphic or
other unnatural DNA bases.1-5 In general, in the design of isomorphic DNA bases, the
structural modification of purine in the five-membered ring is considered mostly.
Several studies have reflected that such modifications, especially for 2’-
deoxyisoguanosine, are tolerated as long as the Watson-Crick face of the nucleoside is
not perturbed.7 As for an example, 8
hydrogen bonds with cytosine, similar to natural G-C pair.8 Studies also show that the
-azaguanine d(8aza)G is capable of forming three
replacement of
G by d(8aza)G does not alter the thermal stability and the helicity of a
DNA duplex.7d In this context several other analogues have been reported utilizing
glycosylation strategy and their photophysical and biophysical properties in the
context of DNA have been evaluated.7d
The Design Concept
However, till the date, in the design of isomorphic purines, the concept of size
expansion or widening the shape in a similar fashion as has been shown by Kool et al.
,
were not considered.5b-c, 9-10 This might provide several advantages of more effective
π-stacking or arresting the inherent tautomerisation.1a,9-10 However, in such a design,
one has to sacrifice one H-bond stabilization energy. Even though, replacing one H-
3

ACCEPTED MANUSCRIPT
bond by hydrophobic benzene would be beneficial as non-H-bonding /hydrophobic/π-
π stacking forces (4-15 kcal/mol/base pair) are stronger than H-bonding force (2-3
kcal/mol/base pair). Furthermore, unnatural bases with the above ability and novel
chemical functionalities expectedly would expand the functional potential of DNA. As
a part of our ongoing research effort toward generation of molecules with tuned
photophysical properties via click reaction and the recent development of click
chemistry derived triazolyl-/tetrazolyl- nucleosides containing donor/acceptor
aromatics (P Q,
, Fig. 1)11a and widened aryl fused hydrophobic triazolyl nucleosides (
Fig. 1),11b we thought that it would be worthwhile to synthesize aryl fused triazolyl-N-
nucleoside with H-bonding ability and widened shape with strong stacking ability.
This design would expectedly produce modulated photophysical response into the
unnatural nucleosides (UNNs) and render strong
duplex if incorporated. With this view we wanted to report the design and
synthesis of a fused triazolyl 2-quinolinone ( FTQuon, Fig. 1) nucleoside with two
π-stacking interaction inside a DNA
1
,
H-bonding faces-one H-bond acceptor and another donor via click chemistry and
study the photophysical properties. We also wanted to study the property inside a
DNA duplex theoretically.
The logic behind our choice of triazole fused H-bonded nucleobase was: (a) the
nucleoside would be the widened version of isomorphic base analogue of 8-
azaguanosine with ability to form two H-bond; (b) the vertically (60o) widened (by 2.4
A) fused benzene expectedly would provide more stacking interaction with possibly
widening the groove size; (c) it could form H-bonding pair with dT/dU or dC as well
as render good stacking stabilization of DNA if incorporated; (d) the presence of the
tautomeric >CO-NH would possibly offer interesting photophysical property such as
4

ACCEPTED MANUSCRIPT
pH and/or microenvironment sensitive fluorescence emission and interaction ability
with biomolecule.
Figure 1.The design concept. Chemical structures of our earlier reported bases (P and Q),
present design (1), Kool’s dyG, natural dG and d(aza)G.
Recently, we reported the bi-conically expanded fused triazolylphenanthrene
nucleoside with no H-bonding faces.11b Now, we wanted to report a similar analogue
having H-banding faces which could be treated as conically widen version of
d(8aza
)G with a concept earlier shown in the design of dyA and dyG by Kool’s
group.5b-c, 9a The design concept is explained in Fig.1. The expansion is only possible
via deletion of one H-bonding face i.e. by replacing the amine group of d
d(8aza with a hydrophobic benzene moiety. In doing so, the 2-H-bonding and
hydrophobic/stacking interaction from additional phenyl ring would allow the base to
pair with natural d /d /d either in a monomeric state or in a DNA if incorporated.
This molecular design would lead the vertical dimension of the base 2.4 Å wider than
G or
)
G
T
U C
5

ACCEPTED MANUSCRIPT
the d
G
or d(8aza)G conically.5b-c, 9a Thus, the base would be an widened version of 8-
azaguanosine with two H-bonding faces.7d,8
The Synthesis
Before going to synthesise, we drew a retrosynthetic path which suggested that
click chemistry between β-azido-deoxyriboside
methylphenyl)propiolamide followed by an intramolecular C-C coupling are the two
key steps to achieve our target nucleoside FTQuon ). Thus, the β-azido-
was synthesized following our earlier
was, next, started with iodination of
followed by DCC coupling with propiolic
3 and N-benzyl-N-(2-iodo-4-
4
(
1
deoxyriboside in bis-toluoyl protected form
work.11 The synthesis of alkyne partner
5
4
p-
toluidine
6 to afford o-iodo-p-toluidine 7
acid and subsequent benzyl protection as shown in the Scheme 1. Finally, a Cu(I)
catalyzed azide-alkyne cycloaddition (CuAAC) reaction between β-azido-
deoxyriboside
5 and the alkyne 4 at room temperature in DMSO for 12 h afforded the
protected nucleoside
9 in 83 % yield via the intermediacy of 8 which underwent a C-C
coupling via Ulmann C-C coupling reaction.12
6

ACCEPTED MANUSCRIPT
Scheme 1. (a) Retrosynthetic analysis and (b) synthesis of FTQuon
.
Based on experimental observation we considered the reaction to proceed following
the proposed mechanism depicted in Scheme 2.12 Thus, in presence of K2CO3 and CuI
the acetylide I was formed which underwent a CuAAC reaction with the sugar azide 5
to afford the reactive organocopper containing triazolyl intermediate 8. This reactive
C-Cu bond containing intermediate then followed an oxidative addition at Cu which
was then inserted into the closely spaced aryl halide bond to afford intermediate II
.
Finally, the catalytic cycle became complete via a reductive elimination of copper
7

ACCEPTED MANUSCRIPT
from the intermediate II leading to the generation of fused triazolylquinolinone
nucleoside
9
via the formation of aryl C–C bond.
O
Bn
N
N
N
N
N
N
O
TolO
O
N Bn
Reductive
elimination
9 (83%)
I
4
OTol
CuI
O
Bn
K2CO3
N
N
N
O
II
Cu
I
Cu
TolO
N Bn
I
O
I
II
OTol
O
I
Oxidative
addition
Bn
N
N
N
TolO
Cu
N3
5
TolO
O
O
8
OTol
OTol
Scheme 2. Plaussible mechanism for the synthesis of FTQuon
.
Toluoyl deprotection by K2CO3 in MeOH:DCM (2:1) solvent afforded the
nucleoside 10 which was subsequently treated for the deprotection of N-benzyl with
o
Pd/C-MeOH/HCO2NH4 in methanol at 90 C to yield the desired fused triazolyl
nucleoside FTQuon (1) in 72% yield (Scheme 1). All the isolated intermediate
compounds and the final product were purified by silica-gel column chromatography
and characterized by NMR and mass spectrometry.
Study of Physical Property
After having the nucleoside in pure form in hand, we, next, studied the
configuration of FTQuon which was found to be β-configuration as was revealed
from the NOESY spectra (SI, Fig. S1). Thus, the presence of interaction between 3'H-
8

ACCEPTED MANUSCRIPT
5'H
,
3'H-2'H and 1'H-Ar-H and absence of any interaction between 1'H-3'H and
1'H-5'H suggested the β-configuration of FTQuon. A theoretical calculation was
carried out, based on B3LYP/6-31G* functional using G09 program package,13 to
ascertain the sugar and N-glycosidic conformation. Thus, the calculation suggested the
sugar puckers in nucleoside FTQuon was North (N) (3'-endo/C2’-exo) and the fused
triazolyl quinolinone base displayed an anti-conformation with respect to the glycosyl
torsion angle.11,14 Calculation of phase angle also supported the same with P = -3.320
and θ0= 33.850 and natural purine like value for -angle (-134.890).14 The anti-glycosyl
χ
conformation is probably due to the less repulsion between the lone pair of triazole-N
and O4'. This repulsive force restrict the rotation of the glycosyl bond and thus, fixed
the base in the anti-disposition. Therefore, the nucleoside FTQuon behaved in a
similar way as that had been shown by the bases we reported earlier11a-b and
expectedly, it would show similar property as the natural purine nucleosides if
incorporated into short oligonucleotide sequence.14a
Figure 2. (a) DFT-optimised geometry and (b)electrostatic potential surface of the FTQuon
nucleoside obtained at B3LYP/6-31G* level of theory using G09 package. (c) Molinspiration
property engine v2016.10 calculated CPK model.
It is expected that the widened base if incorporated into DNA, the groove depth would
differ from that of natural DNA helix. Moreover, one way vertical expansion in the present
9

ACCEPTED MANUSCRIPT
design is expected to have a good pairing selectivity with oppositely placed dT/dC natural
bases with enhanced intrastrand stacking interaction. The close shape relationship of the
present design with natural dG, d(aza)G,8 Kool’s widened bases9-10 and our previouly
reported base, FTPhen,11b revealed from a space-filling model supported our expectation
(Fig. 2). DFT-optimised geometry and the electrostatic potentials surface mapping showed
the negative charge density onto the triazole ring and amide moiety indicating hydrophobic as
well as moderately polar nucleoside (Fig. 2a-b).9a The triazolylquinolinone moiety in
FTQuon is planar similar to our earlier reported base FTPhen.11b We also calculated the
parameters, such as, surface area and hydrophobicity which are indicative of stacking
propensity of DNA bases and analogues, using Molinspiration property engine v2016.10
calculator and found topological polar surface area (TPSA) = 113.3 Å2, volume = 267.9 Å3
and log P (octanol-water) value as 0.01 indicating polar character of FTQuon (Fig. 2c).15
Only base part without sugar unit, the triazolyl quinolinone base exhibited TPSA = 63.6 Å2,
volume = 186.2 Å3 and log P (octanol-water) value of 1.2.11b, 15 All the results suggested that
FTQuon would be well-suited with the natural genetic system if incorporated. Therefore,
such type of bases would be useful in studying the structure and stability of DNA and in
various biotechnological applications and generation of functional DNA based materials with
interesting photophysics.
Study of Photophysical Property
Based on the studied conformation and shape, we expected that the nucleoside
FTQuon might possess favorable photophysical properties useful for analytical and
structural studies in DNA. Therefore, the photophysical property of the nucleoside in
various organic solvents was, next, evaluated, prior to study its property in a
theoretical B-DNA duplex. The nucleoside, FTQuon, showed interesting UV-Visible
10

ACCEPTED MANUSCRIPT
absorption spectra as the solvent polarity increased from dioxane to buffer. In dioxane,
chloroform and ethyl acetate it showed structureless absorption maxima at around 315
nm. With increase in solvent polarity, the structurless absorption band changed to
structured band with blue shifted pattern. In buffer, two bands, at 300 nm and 313 nm,
appeared with slight hyperchromicity and blue shifted by 14 nm from dioxane (Fig.
2a). The acid sensitivity on the photophysical property was also reflected. Thus, the
absorption spectra of FTQuon in acidic environment (acidic ACN) exhibited a
structurless absorption band at 315 nm, while in basic ACN solvent, a red shift of 10
nm along with hypochromic effect was observed (Fig. 3a). Upon excitation at
absorption maxima of each solvent (at 250-320 nm) FTQuon showed structureless
emission. While in dioxane, structureless emission band was appeared at 374 nm,
strong hyperchromicity with a slight (4 nm) blue shift was a result of emission in
methanol (Fig. 3b). On the other hand, in buffer, a hypochromic effect was observed at
428 nm. In acidic ACN, structured emission appeared at 358 nm and 426 nm, while a
red shifted (by 22 nm) structureless emission with decreased intensity at 448 nm was a
result in basic medium (Fig. 3b). All these observations are indicative of chronological
improvement of photophysical property, especially in terms of solvatochromicity,
from our previously reported11a-b bases. The present base can be utilised for sensing of
pH of an environment. Furthermore, it would be more efficient in sensing the DNA
microenvironment in a better way compared to our earlier bases.
11

ACCEPTED MANUSCRIPT
Figure 3. (a-b) UV-visible and fluorescence spectra of FTQuon in various organic
solvents and (c-d) in dioxane/water mixed solvent system (concentration = 10 ꢀM).
The solvatochromicity was also tested by titration of a dioxane solution of FTQuon with
water. The UV-visible absorptions at 243 and 316 nm in dioxane did not experience much
perturbation except slight hyperchromic shift of short wavelength was observed as the
increasing amount of water was added (Fig. 3c). However, upon excitation at 330 nm, the
emission band in dioxane at 374 nm experienced strong hypochromicity along with red sift as
the percentage of water increased and a maximum red shift of 42 nm was observed in 100%
water (Fig. 3d).
12

ACCEPTED MANUSCRIPT
Figure 4. UV-visible (a), its expansion spectra (b), the tautomeric forms (c) and
fluorescence spectra (d) of FTQuon in solutions of various pH (concentration = 10
ꢀM).
Next, pH sensitivity was tested. Thus, the FTQuon exhibited a structured short
wavelength absorption band at 250 nm and a long wavelength absorption band at 320
nm in high to low acidic pH. However, in strong basic pH of 12, FTQuon exhibited
absorption at 290 nm and 242 nm possibly indicating a tautomerization (Fig. 4a-c).
Upon excitation at 335 nm, FTQuon showed structureless emission at 423 nm in the
pH range of 2.2-10.4. Beyond pH 5 the emission bands exhibited a hypochromic
effect. At highly basic pH 12, FTQuon showed blue shifted structured emission bands
at 383 and 404 nm with decreased intensity indicating a tautomerisation leading to
deprotonation from acidic lactim form (Fig. 4c-d).16
13

ACCEPTED MANUSCRIPT
Theoretical (TDDFT) Calculation
To get further insight into the photophysical property, we carried out TDDFT
calculation using G09 programme package at B3LYP/6-31g (d,p)level of theory.13
Thus, the LUMO coefficient, calculated based on time dependent density functional
(TD-DFT) theory at B3LYP/6-31G∗ level, indicated a significant orbital overlap of the
triazole and quinolinone moiety reflecting a strong electronic coupling between them.
For an obvious reason, therefore, the nucleoside FTQuon showed modulated optical
properties which were already supported from experimental results. Thus, the
nucleoside would be utilised as a probe of DNA microenvironment if incorporated into
a short oligonucleotide sequence. All possible transitions from TDDFT calculation
were, then, analysed (SI, Fig. S2). It was clear from the HOMO-LUMO overlap and
transition oscillator strength that the S0→S1 transition is fully allowed (Fig. 4)
indicating the reverse transition, i.e., S0←S1, as fully allowed. Redistribution of
electronic charge density among the methylquinolinone and the triazole unit was also
revealed from an overlap in HOMO-LUMO. This supported the solvatochromicity of
the nucleoside (Fig. 3).17 The TD-DFT calculations also indicated the HOMO-LUMO
transition with high configuration interaction (CI) as the dominant orbital electronic
transition in the low-lying singlet excited state of the nucleoside. The calculated
excitation energy for the dominant transition, S0→S1, was found to be at 311 nm (E1 =
3.99 eV; f = 0.11; CI = 0.68) (in vacuum) which was closest to the experimental
results of 315 nm (dioxane). Other prominent transitions were found at 267 nm [(E2 =
4.65 eV; f = 0.02; CI = 0.51(HOMO-1
0.22; CI = 0.32 (HOMO-1 LUMO) /0.38 (HOMO
[E4 = 5.31 eV; f = 0.15; CI = 0.26 (HOMO-1
ꢀ
LUMO)]; at 243 nm [(E4 = 5.10 eV; f =
LUMO + 1)] and at 233 nm
LUMO)/ 0.12 (HOMO-1
LUMO+2)] (Fig. 5).These
ꢀ
ꢀ
ꢀ
ꢀ
LUMO+2)/0.21 (HOMO
ꢀ
LUMO+1)/0.47 (HOMO
14
ꢀ

ACCEPTED MANUSCRIPT
transitions in vacuum were close to the experimental values of 258, 237 and 225 nm in
dioxane. All the results rationalized the solvent polarity dependent emission of the
nucleoside.17 It is to be noted that the HOMO-LUMO transitions lie at 3.99, 5.12 and
4.88 ev for FTQuon, azadG and natural dG, respectively (See SI, Section 3),
indicating more efficient fluorophoric nature (with long wavelength transition at 311
nm) of FTQuon compared to others.
Figure 5. The possible transitions from TD-DFT calculation for the nucleoside, FTQuon.
15

ACCEPTED MANUSCRIPT
Macromodel Study
A macromodel calculation was carried our using Schrodinger Macromodel18 to
evaluate the property of a theoretical DNA duplex containing our synthesized
nucleoside. The AMBER* force field was utilized to optimize the structure of the
theoretical DNA duplexes [(5'-CGCAAT-FTQuon-TAACGC-3'/3'-GCGTTA-N-
ATTGCG-5'); (N = A, G, C, T)] wherein the FTQuon was placed at the middle and
opposite to all four natural bases to test its interaction behaviour. It was clear from the
structural investigation that in all the duplexes, FTQuon resided deeply into the
interior of the helix optimizing the helicity and the groove properties whenever present
opposite to thymine and cytosine (Fig. 6). The base FTQuon was able to maintain a
B-form DNA conformation and held both via interstand and intrastand stacking
interaction with the natural bases having an average base-base rise comparable to the
natural bases. Most interestingly, FTQuon formed hydrogen bonds with d
was within 1.81-1.88 Å comparable with the distance of 1.8 Å in natural dA:
G: pair. However, it did not fit well as an opposite pairing partner with d and d
C which
d
T
or
d
d
C
A
G
due to its mismatched hydrogen bonding sites as revealed by AMBER* optimized
geometry as well as steric bulk. However, it an unperturbed B-form DNA duplexes
were the result when placed opposite to d
A or dG due to the high stacking propensity
of the angularly expanded base that stabilized the duplex DNA by π-π stacking
interaction (SI, Fig. S4).
16

ACCEPTED MANUSCRIPT
Figure 6. AMBER* optimized theoretical duplex DNAs wherein FTQuon was placed
opposite to dC (a) and dT (b). (c-d) Corresponding top views.
DFT Calculation in the Context of DNA
To get a better idea about the pairing efficiency of FTQuon with d
C and dT in the
theoretical base pairs, we carried out DFT calculation using G09 program package.13
Optimization of the various base pairs at B3LYP/6-31g (d,p) level of theory and basis
set superposition error (BSSE) corrections using counterpoise correction procedure
(CP)19 revealed a good stacking energy offered by FTQuon when placed opposite to
d
d
C
and/or d
T. The pairing energy calculated, thus, followed the order-
G:dC>FTQuon:dC>>dA:dT ≥ dFTQuon:dT>FTQuon:dG>FTQuon:dA. It is
interesting to note that the stacking energy of FTQuon :
kcal/mol suggesting it to be more stable than natural d :d
G: pair (Table 1). As both the dA: and FTQuon:
d
C
pair was found to be -22.0
A
T pair and less stable than a
d
d
C
d
T
d
C
pairs possessed two
hydrogen bonds, the higher interaction energy compared to d
17
A:dT pair suggested extra

ACCEPTED MANUSCRIPT
stabilization through π-stacking interaction in FTQuon:
d
C
pair (Table 1, Figure 7a).
was found to be -14.3
kcal/mol which was close to the interaction energy of natural dA: pair (Table 1,
Figure 7b). The BSSE corrected interaction energies of other two pairs-FTQuon:
and FTQuon: were found to be much lower than any of the natural pair reflecting
On the other hand, the interaction energy of FTQuon:dT
d
T
d
A
d
G
the steric destabilization and structurally mismatched hydrogen bonding in these two
pairs.11 Our preliminary calculation based on azadG showed that, the azadG:C pair
exhibits comparable stability as that of a natural G:C pair and more stable than
FTQuon:C pair (See SI Section 5). However, it is worth mentioning to the fact that
the calculated pairing energies for FTQuon:C and FTQuon:T are found to be higher
than those observed with our previously reported11a-b bases which is a reflection of
combined effect of both H-bonding and stacking interactions. Further, study is needed
to get a more clear picture of stabilising property in a DNA with all our newly
developed bases (TPhenBDo
; FTPhen and FTQuon) which is our future target.
Table 1. BSSE corrected pairing energies of base pairs
Base pair
BSSE corrected stacking
Hydrogen bonding
distance(Å)
energy(kcal/mol)
FTQuon:A
FTQuon:G
FTQuon:C
FTQuon:T
A:T
-7.1
-8.1
-22.0
-14.3
-15.5
-31.5
1.9, 2.0
1.7, 2.0
1.9, 1.7
G:C
1.7, 1.9, 1.8
18

ACCEPTED MANUSCRIPT
Figure 7. (a) FTQuon:C base pair. (b) FTQuon:T base pair.
Study of Interaction with BSA
Next, we were interested in studying the interaction of our new unnatural
fluorescent nucleoside, FTQuon ) with bovine serum albumin (BSA) as a model
(
1
protein. Understanding the mode of interactions between a drug and serum proteins is
an essential to assess the therapeutic effectiveness of a drug because any drug entered
into blood plasma has to come upon the most abundant protein, serum albumin.20 The
binding of a drug to serum proteins correlates with its bioavailability, metabolism,
excretion and interaction with the target tissues and thereby its biological activity.21
However, the poor knowledge of such interactions for nucleoside is a major cause of
why only 40 nucleoside analogues have been approved for clinical use.22 As for an
example, the clinical use of AzU nucleoside is limited because of its poor cell
permeability.23 Therefore, immense biological and clinical importance of
nucleosides/nucleotides/their analogues enforce to accumulate knowledge how they
are being recognized and transported into the cell by nucleoside transporter membrane
protein. In this respect, we thought that it would be worthwhile to study the
preliminary interaction behaviour of our new unnatural nucleoside, FTQuon with
bovine serum albumin (BSA) which has 76% sequence identity to human serum
19

ACCEPTED MANUSCRIPT
albumin (HAS). BSA is widely used as a model protein in investigating protein-drug
interaction.24 It contains three linearly arranged domains (I–III) with two major drug
binding sites present in the subdomains IIA (site I) and IIIA (site II).25 The two
tryptophan (Trp) residues-Trp134 and Trp212 are located in the outer hydrophilic
environment of site II and in the hydrophobic binding pocket of site I, respectively.20,
26 These residues are, therefore, being widely used as an intrinsic fluorescent probe to
study drug binding interactions to BSA. Thus, we studied the interaction of FTQuon
with model protein BSA using UV-visible and fluorescence spectroscopy.
Figure 8.(a) UV-visible, (b) fluorescence excitation (λem= 420 nm), (c) fluorescence
emission (λex= 320 nm) and (d) steady-state anisotropy and polarisation plot of probe
nucleoside in absence and presence of increasing amount of BSA. [FTQuon]= 10 µM.
20

ACCEPTED MANUSCRIPT
The absorption band at 320 nm experienced slight hyperchromicity as the gradually
increased concentration of BSA was added to the probe’s solution in sodium
phosphate buffer of pH 7 (Fig. 8a). Excitation at 320 nm the emission spectra of
FTQuon exhibited structureless broad emission band at 420 nm which showed
hyperchromic effect upon gradual increase of BSA concentration (Fig. 8b). All these
observations indicated an interaction between FTQuon and the hydrophobic pocket of
BSA.27 The interaction was also evident from prominent hyperchromic effect of the
excitation spectra at 355 nm upon addition of increasing concentration of BSA (Fig.
8c). Furthermore, gradual increase in steady state anisotropy with increase in BSA
concentration also supported the binding of FTQuon inside the hydrophobic pocket of
BSA (Fig. 8d).27c-d
Figure 9.(a) Docking pose and (b) various interactions of FTQuon inside the
hydrophobic pocket of BSA.
Finally, to support binding of the probe nucleoside in the hydrophobic pocket of
BSA, we carried out molecular docking calculation using Autodock 4.28 The docking
pose suggested that the probe nucleoside was surrounded by the amino acid residues-
122 (Leu), 136 (Lys), 137 (Tyr), 160 (Tyr)-involving hydrophobic interaction inside
17b,
the hydrophobic pocket.
27cThe 5’-OH, 3’-OH of sugar and -HN/>CO of
21

ACCEPTED MANUSCRIPT
quinolinone unit involved in H-bonding interactions with ASP (118), Glu (125) and
Tyr (160), respectively, of BSA. The tolyl-ring of the nucleobase involved in π-
stacking interaction with Phe (133) of BSA (Fig. 9a-b).17b, 27cThe free energy of
binding was found to be -5.82 kcal/mol.
Conclusion
In conclusion, we successfully synthesised 1'-(8-methyl-1H-[1,2,3]triazolo[4,5-
c]quinolin-4(5H)-one)-2'-deoxyriboside via click chemistry. The nucleoside, we,
designated as fused triazolyl 2-quinolinone (FTQuon) nucleoside having two H-
bonding faces and a conically expanded benzene ring capable of rendering extra
stabilization via π-stacking interaction. Theoretical study revealed its very good DNA
duplex stabilization property. Furthermore, pairing selectivity with pyrimidine bases,
especially, the cytosine selectivity with pairing energy much higher than a natural A:T
pair, would impact greatly for utilizing this nucleoside as a possible probe for DNA
analysis. The vertically widened base would expectedly show good H-bonding as well
as π-stacking propensity if incorporated into DNA. Interesting microenvironment and
pH sensitive fluorescence and good interaction with BSA with enhancement of
fluorescence intensity would allow this nucleoside for further exploration in
biophysical research in the context of DNA, DNA-protein interactions and to study the
mode of interaction with the enzyme.
Experimental Section
General Experimental
All reactions were carried out under a nitrogen atmosphere using dry organic solvents.
Organic extracts were dried over anhydrous sodium sulphate and evaporated to dryness using
22

ACCEPTED MANUSCRIPT
a rotary evaporator under reduced pressure. Silica gel (60-120 mesh) was used for the column
1
chromatography. Reactions were monitored by TLC on silica gel 60 F254 (0.25mm). H
NMR spectra were recorded with Bruker 600 (600 MHz)/or Varian 400 MHz (400 MHz)
13
NMR machine and C NMR spectra were measured with Bruker 600 (150 MHz)/or Varian
400 (100 MHz) spectrometer. Coupling constants (J values) were reported in hertz. The
chemical shifts were shown in ppm downfield from tetramethylsilane, using residual
chloroform (δ = 7.24 in 1H NMR, δ = 77.23 in 13C NMR), methanol (δ = 3.34 in 1H NMR, δ
13
1
13
= 49.2 in C NMR), dimethyl sulfoxide (δ = 2.48 in H NMR, δ = 39.5 in C NMR), as
internal standards. Mass data were recorded using WATERS MS system, Q-tof premier and
data analyzed using Mass Lynx 4.1. IR spectra were recorded in KBr or neat on a Perkin
Elmer Spectrum one FT-IR spectrometer.
Synthesis of 2-iodo-4-methylaniline (7):29 This was prepared following a modified
literature reported procedure.29 Thus, p-toluidine (6, 2g, 18.7 mmol), NaHCO3 (2.35 g, 28
mmol) and water (20 ml) were introduced in a100 ml round bottom flask. The reaction
mixture was allowed to cool to 0 oC with stirring and iodine (0.96 g) was added in portion at
an interval of 15 min. After completion of the addition of iodine the reaction mixture was
allowed to stir at room temperature for 12 h. Completion of the reaction was monitored by
TLC. The reaction mixture was extracted with ethyl acetate and the organic fractions were
washed with Na2S2O3. The organic fractions were collected and evaporated to dryness to get
the crude reaction mixture which was purified by silica gel (60-120 mesh) column
chromatography using 15:1 Hex:EtOAc as eluent affording the product 7 as blackish brown
solid. Yield = 75% (3.21 g). 1H NMR (600 MHz, CDCl3) δ 7.48 (s, 1H), 6.95 (d, J = 7.7 Hz,
13
1H), 6.66 (d, J = 8.1 Hz, 1H), 3.95 (s, 2H), 2.21 (s, 3H); C NMR (150 MHz, CDCl3) δ
144.4, 139.1, 130.1, 129.6, 114.7, 84.4, 77.1, 19.9.
23

ACCEPTED MANUSCRIPT
Synthesis of N-benzyl-N-(2-iodo-4-methylphenyl)propiolamide (4):This was prepared
following a modified literature reported procedure.30 Thus, a 2-neck round bottom flask fitted
with a septum, adapter and magnetic stirring bar was evacuated and backfilled with nitrogen
and the process was repeated for 3 times. In nitrogen atmosphere DCM (10 ml) was injected
and DCC (412 mg, 2.0 mmol) was added while degassing by N2 was continued. Then, the
o
reaction mixture was allowed to come to 0 C. At that temperature propiolic acid (70 mg, 1
mmol) was added and the reaction mixture was allowed to stir for half an hour ensuring the
formation of the white precipitate in the transparent reaction mixture. After that 2-iodo-4-
methyl aniline (7, 280 mg, 1.2 mmol) was added to it and the reaction mixture was allowed to
stir for overnight. Completion of the reaction was monitored by TLC. The reaction mixture
was filtered through celite and washed with 3M HCl and saturated NaHCO3. Evaporation of
the organic layer gave crude product which was purified by column chromatography using
10:1 Hex:EtOAc as eluent to afford pure product 4 as brown gummy material. Yield = 46%
(0.46 mmol, 124 mg). This material then was treated with benzyl iodide (0.55 mmol, 120 mg)
o
and K2CO3 (0.92 mmol, 127 mg) in acetonitrile at 80 C for 12 h and completion of the
reaction was checked by TLC. The reaction mixture was then evaporated to dryness and
diluted with DCM and washed with water and brine. Evaporation of the organic residue
followed by purification through column chromatography afforded the pure product 4. Yield
= 42% (0.19 mmol, 71 mg). 1H NMR (600 MHz, CDCl3) δ 7.75 (s, 1H), 7.28 – 7.25 (m, 3H),
7.22 – 7.19 (m, 2H), 6.99 (d, J = 7.8 Hz, 1H), 6.63 (d, J = 8.0 Hz, 1H), 5.59 (d, J = 14.3 Hz,
13
1H), 4.04 (d, J = 14.3 Hz, 1H), 2.76 (s, 1H), 2.31 (s, 3H); C NMR (150 MHz, CDCl3) δ
153.4, 140.8, 140.4, 140.3, 136.0, 130.9, 129.7, 129.6, 128.6, 128.0, 100.0, 79.4, 77.1, 76.3,
51.4, 20.7.
Synthesis of ((2R,3S,5R)-5-(5-benzyl-8-methyl-4-oxo-4,5-dihydro-1H-[1,2,3]triazolo[4,5-
c]quinolin-1-yl)-2-(((4-methylbenzoyl)oxy)methyl)tetrahydrofuran-3-yl 4-methylbenzoate
24

ACCEPTED MANUSCRIPT
(9):In a 2-neck round bottom flask fitted with a magnetic stirring bar and a septum, 10 ml of
dry DMSO was added to it followed by N-benzyl-N-(2-iodo-4-methylphenyl)propiolamide (4,
190 mg, 0.506 mmol), the azido sugar 5 (200 mg, 0.506 mmol)11 and K2CO3 (140 mg, 1.012
mmol). Next, CuI (10 mg, 0.0506 mmol) was added to the reaction mixture and the reaction
was allowed to stir at ambient temperature for 12 h. The completion of the reaction was
monitored by TLC. The reaction mixture was diluted with ethyl acetate and washed with
water and brine. Evaporation of the organic solvent followed by purification of the crude
reaction mixture through column chromatography using 2:1 Hexane: Ethyl acetate as eluent
afforded the target nucleoside 9 (possibly via the intermediacy of 8) as a white solid. Yield =
83% (280 mg). 1H NMR (400 MHz, CDCl3) δ 8.03 (s, 1H), 7.98 – 7.93 (m, 2H), 7.61 – 7.57
(m, 2H), 7.25 – 7.17 (m, 6H), 7.16 – 7.07 (m, 4H), 6.79 (dd, J = 6.5, 4.4 Hz, 1H), 5.98 – 5.92
(m, 1H), 5.64 (bd, 1H), 5.48 (bd, 1H), 4.75 (q, J = 4.5 Hz, 1H), 4.45 (dd, J = 12.1, 4.4 Hz,
1H), 4.32 (dd, J = 12.2, 5.2 Hz, 1H), 4.29 – 4.23 (m, 1H), 2.97 (dt, J = 14.4, 5.9 Hz, 1H),
2.40 (s, 3H), 2.38 (s, 4H), 2.31 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 21.1, 21.8, 21.9, 36.2,
45.9, 63.4, 74.9, 77.2, 84.9, 88.6, 110.5, 116.9, 126.5, 126.6, 127.4, 128.9, 129.1, 129.3,
129.4, 129.6, 129.9, 130.3, 132.4, 132.9, 135.6, 136.3, 136.9, 137.4, 144.0, 144.6, 156.5,
166.1; IR (KBr) 2923, 1712, 1681, 1610, 1454, 1411, 1384, 1271, 1178, 1093, 1020, 805,
752, 691 cm-1; HRMS calculated for C38H35N4O6[M+H]+ 643.2551, found 643.2558.
Synthesis of 5-benzyl-1-((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-
8-methyl-1H-[1,2,3]triazolo[4,5-c]quinolin-4(5H)-one (10):The compound 9 (100 mg, 0.16
mmol) was taken in a round bottom flask fitted with magnetic stirring bar and a septum.
Methanol (10 ml) was added to it followed by DCM (5 ml) and K2CO3 (86 mg, 0.63 mmol).
The reaction mixture was allowed to stir at room temperature for 6 h. Completion of the
reaction was monitored by TLC. The organic part was evaporated to dryness and the crude
reaction mixture was purified by column chromatography using 10:1 CHCl3: MeOH as eluent
25

ACCEPTED MANUSCRIPT
to yield the target nucleoside 10 as a white solid. Yield = 98% (62 mg). 1H NMR (400 MHz,
CD3OD) δ 8.23 (s, 1H), 7.45 – 7.38 (m, 2H), 7.28 (d, J = 7.2 Hz, 2H), 7.20 (d, J = 7.9 Hz,
3H), 6.93 (dd, J = 6.9, 3.7 Hz, 1H), 5.65 (bd, J = 16.2 Hz, 2H), 4.79 (q, J = 5.8 Hz, 1H), 4.16
(q, J = 4.8 Hz, 1H), 3.69 (ddt, J = 14.0, 7.8, 3.6 Hz, 1H), 3.58 (dd, J = 12.1, 4.3 Hz, 1H),
3.43 (dd, J = 12.0, 5.4 Hz, 1H), 2.68 (dt, J = 13.3, 6.4 Hz, 1H), 2.46 (s, 3H), 2.40 (s, 1H); 13C
NMR (100 MHz, CD3OD) δ 20.8, 30.7, 39.3, 46.6, 49.0, 63.1, 72.3, 89.4, 90.1, 118.3, 126.1,
127.5, 128.4, 129.8, 130.0, 130.7, 133.4, 134.7, 137.6; IR (KBr): 2924, 2853, 1676, 1657,
1566, 1520, 1456, 1384, 1369, 1265, 1206, 1158, 1087, 1048, 1011, 993, 807, 696; HRMS
calculated for C22H23N4O4[M+H] + 407.1714, found 407.1701.
Synthesis
of
1-((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-8-
[Fused triazolyl quinolinone
methyl-1H-[1,2,3]triazolo[4,5-c]quinolin-4(5H)-one
nucleoside, FTQuon] (1):The compound nucleoside 10 (50 mg, 0.13 mmol) was taken in a
two neck round bottom flask fitted with magnetic stirring bar and septum. The reaction vessel
was evacuated and back filled with nitrogen. The process was repeated for 3 times to make
ensure that there was no tress of air present in the reaction vessel. Methanol was added to the
reaction mixture while degassing was continued with N2 gas. Next, Pd/C was added 5 times
of the weight of the N-benzyl-nucleoside 10 followed by ammonium formate (34 mg, 0.52
o
mmol). The reaction mixture was then allowed to heat at 90 C for about 4-5 h. The
completion of the reaction was monitored by TLC. The reaction mixture was then filtered
through Whatman 42 filter paper. Evaporation of the solvent followed by column
chromatographic purification (silica gel, 60-120) using 10:1 CHCl3:MeOH as eluent afforded
1
the final target nucleoside 1 as white solid compound. Yield 72% (30 mg). H NMR (400
MHz, D6-DMSO) δ 8.03 (s, 1H), 7.44 (s, 3H), 6.93 (dd, J = 6.7, 3.8 Hz, 1H), 5.57 (d, J = 4.9
Hz, 1H), 4.68 (t, J = 5.3 Hz, 1H), 4.61 – 4.53 (m, 1H), 3.99 (q, J = 4.9 Hz, 1H), 3.44 (s, 1H),
3.43 (s, 1H), 3.33 (t, J = 4.7 Hz, 1H), 3.20 – 3.12 (m, 1H), 2.43 (s, 3H);13C NMR (150 MHz,
26

ACCEPTED MANUSCRIPT
D6-DMSO) δ 155.7, 136.9, 136.1, 135.7, 131.9, 131.6, 123.8, 116.5, 108.7, 88.7, 87.4, 70.5,
61.4, 39.5, 38.0, 20.7; DEPT-135 (150 MHz, D6-DMSO) δ 131.7, 123.5, 88.5, 87.2, 70.2,
61.2, 39.5, 37.8, 20.5; IR (KBr): 3434, 3198, 3145, 1687, 1633, 1401, 1119, 1084, 1059,
1025, 948, 889, 859, 706; HRMS calculated for C15H17N4O4[M+H]+ 317.1244, found
317.1239.
Study ofphotophysical property of FTQuon
UV-visible measurements:All the UV–visible spectra of the nucleoside monomer (10 µM)
were measured in different organic solvents using a UV-visible spectrophotometer with a cell
of 1 cm path length at 25 °C. The measurements were taken in absorbance mode. The
absorbance values of the sample solutions were measured in the wavelength regime of 200–
550 nm. All the sample solutions were prepared just before doing the experiment.
Fluorescence experiments: were prepared as described in UV measurement experiments.
Fluorescence spectra for the nucleoside monomer (10 µM) in various solvents were obtained
using a fluorescence spectrophotometer at 25 °C using 1 cm path length cell. The excitation
max
abs
λ
wavelengths for each solvent were set at their corresponding absorption maxima (
), and
emission spectra were measured in the wavelength regime of 315–600 nm with an integration
time of 0.1 sec.
Studies On the Interaction of FTQuonwith BSA
Preparation of BSA solution:Phosphate buffer of pH 7.02 was used to prepare the
solution of BSA (Merck). A 1200 ꢀM of stock BSA solution was prepared by dissolving 0.08
gm of BSA in 1 mL phosphate buffer (20 mM) of pH 7.0. From that stock solution sub stock
of 1000 ꢀM BSA was prepared. The compound stock solution was prepared in DMF because
27

ACCEPTED MANUSCRIPT
of the poor solubility in water. 1.0 mg of nucleoside FTQuonwas dissolved in 1.5 mL DMF
to make a stock probe solution of concentration 910.3 ꢀM.
General experimental on interaction study of BSA by photophysical study:All the
spectral measurements were carried out at room temperature. To study the interaction of
compound with BSA, an aqueous solution of nucleoside FTQuon(10 ꢀM) was titrated with
different concentrations of BSA (ranging from 0, 0.25, 0.5, 7.5, 1.0, 1.5, 2, 2.5, 3.0, 4, 5, 6,
10, 14, 18, 20 equivalent). The total volume of the final solution for each sample was 3 mL.
The % of DMF content was 2%. The presence of 2% DMF does not induce structural
changes to biomolecules. Each sample solution was mixed well before spectral
measurements.
UV-Visible study with BSA:The UV–Visible absorbance measurements were performed
using UV-visible spectrophotometer with a cell of 1 cm path length at 298 K. All the UV-
Visible studies were carried out in 20 mM phosphate buffer of pH 7.02 containing solution at
298 K. 2 % DMF was used to solubilize the probe. The measurements were taken in
absorbance mode and the absorbance values of the sample solutions were measured in the
wavelength regime of 200– 700 nm. All the experiments were carried out with freshly
prepared sample solutions.
Fluorescence study with BSA:All fluorescence and steady state anisotropy experiments
were performed using a steady state fluorescence spectrophotometer with a cell of 1 cm path
length at 298 K. The excitation wavelength for probe nucleoside FTQuonwas set at 220 nm,
420 nm and emission spectra were measured in the wavelength regime of 290–650 nm.
Steady state anisotropy of the solutions was measured using steady state fluorescence
spectrophotometer. The fluorescence lifetime experiment was carried out as described above.
28

ACCEPTED MANUSCRIPT
DFT Calculation
The DFT calculations were carried out using Guassian 09 programme package.13All
the geometries of natural and unnatural bases were optimized by B3LYP/6-31g(d,p) and
stacked bases were optimized using the M05-2X density functional developed by Truhlar and
Zhao.31 In the present preliminary calculation, the initial starting geometries of stacked
dimers natural/unnatural hybrids in B-DNA conformation were generated using the
Schrodinger Macromodel program. From the generated structures in B-DNA conformation,
we removed the sugar, phosphate backbone for simplicity and optimized each individual pair.
Basis set superposition error (BSSE) corrections were carried out to obtain the corrected
interaction energy by using the counterpoise correction procedure (CP).19From the
calculation, a considerable amount of stacking interactions were observed between FTQuon
and natural nucleosides from DFT calculation using M05-2X/6-31G(d,p)31 level of theory.
Macromodel Study
Molecular modeling study of the theoretical DNA duplexes were carried out using
Schrodinger Macro Model18 (Maestro vs. 9.0)32 software with MMFFs force field in water.
After optimization we carried out conformational search using MMFFs force field at constant
dielectric in water with “large scale low-frequency-mode conformational search” (Mixed
torsional/Largescale low-mode sampling = MCMM/LMCS) method using Schrodinger
32
Macromodel (Maestro vs. 9.0) software package.18,
A total of 500 structures were
processed with 500 maximum no. of steps iteration. A global search analysis eliminates
redundant conformers using RMS deviation for all compared atoms exceed the threshold
Cutoff of 0.5 Å. An optimal minimization method was chosen for minimizing the generating
conformers. The geometry minimization, molecular dynamics calculation using MMFFs
force field supported B-DNA conformations of the theoretical DNAs.
29