Construction of 3D Antioxidants with Nucleosides as the Core: Inhibition of DNA Oxidation

Article abstract of DOI:10.1021/acs.joc.9b02104

We herein attach ferulic and caffeic acids to -OH and -NH₂ in cytidine, uridine, adenosine, or guanosine for achieving antioxidative hybrids with three-dimensional (3D) configuration. In the case of molecular docking computation, the nucleoside

Full text of DOI:10.1021/acs.joc.9b02104

Article
pubs.acs.org/joc
Cite This: J. Org. Chem. XXXX, XXX, XXX−XXX
Construction of 3D Antioxidants with Nucleosides as the Core:
Inhibition of DNA Oxidation
Peng-Fei Zhao,† An Liu,‡ Ming-Guang Wei,‡ and Zai-Qun Liu*,†
†Department of Organic Chemistry, College of Chemistry, Jilin University, Changchun 130021, People’s Republic of China
‡The Second Affiliated Hospital of the Air Force Medical University, Xi’an 710032, People’s Republic of China
S
* Supporting Information
ABSTRACT: We herein attach ferulic and caffeic acids to −OH and −NH2 in cytidine, uridine, adenosine, or guanosine for
achieving antioxidative hybrids with three-dimensional (3D) configuration. In the case of molecular docking computation, the
nucleoside antioxidants with 3D configuration facilitate to bind with the groove of DNA and to cover the surface of a DNA
helix. Experimentally, the antioxidative effects of nucleoside hybrids are measured in the inhibition of DNA oxidation caused by
2,2′-azobis(2-amidinopropane dihydrochloride) (AAPH), and the stoichiometric factor (n, the number of free radical
propagations terminated by one molecule of antioxidant) can be selected as a quantitative index for expressing antioxidative
effects. It is found that the effect of cytidine tetraferulate against AAPH-induced DNA oxidation is seven times better than that
of ferulic acid, even though four ferulic acid moieties are involved and cytidine itself does not exhibit activity. Moreover, the
antioxidative effect of cytidine tetracaffeate is almost 20 times higher than that of caffeic acid. The n values of nucleoside
antioxidants against AAPH-induced DNA oxidation are found to correlate proportionally with the rate constants for quenching
2,2′-diphenyl-1-picrylhydrazyl and galvinoxyl radicals. Therefore, nucleoside linking with antioxidative carboxylic acid might be
a promising way for constructing antioxidants against peroxyl radical-induced oxidation of DNA.
antioxidative effect herein.11 For this purpose, the DNA
oxidation mediated by peroxyl radical is selected to be a
INTRODUCTION
■
Ever since the correlation of pathogenesis of some diseases
biological target. The peroxyl radical is generated from the
with DNA damages has been discovered, a great deal of
decomposition of 2,2′-azobis(2-amidinopropane dihydrochlor-
speculation has focused on how the integrity of a DNA
ide) (AAPH, R−NN−R, R = −CMe2C(NH)NH2)12 and
structure can be artificially preserved in the unbalanced redox
surrounding.1 It has been found that some signaling pathways
is able to abstract a H atom from the 4-position of deoxyribose
can be regulated by endogenous antioxidants2 even though low
in DNA, leading to the oxidation of DNA. The oxidative
products can be detected after colorizing by reacting with 2-
concentration of reactive oxygen species might play active roles
thiobarbituric acid (TBA).13
in inhibiting cancer,3 whereas oversupplementation of
antioxidants might activate the cancer gene to spur the growth
of tumors.4 Nevertheless, convincing evidence has demon-
strated that the pathogeny of some diseases is definitely related
to DNA damage5 mediated by the oxygenation in vivo.6
Hence, usages of exogenous antioxidants are still promising
ways to avoid, at least, to rectify the unbalanced redox status in
vivo.7 These backgrounds motivate researchers to contribute
their efforts to the investigation on antioxidants, aiming at
preventing the diseases via inhibition of DNA damages.8 In
addition to antioxidative proteins,9 the major objective in this
field is to design reliable and effective antioxidants with a
specific configuration being recognized by biological micro-
environments readily.10 On the other hand, it is essential to
find an appropriate in vitro biological oxidation model for
mimicking DNA undergoing oxidation and for evaluating the
Nucleosides have been widely applied in the design of drugs
owing to their special affinities toward DNA strands via
multiple hydrogen bonds.14 This view guides numerous efforts
to synthesize nucleoside derivatives as potential therapeutic
agents,15 and a large number of structural styles are produced
to meet with biological requirements.16 However, few reports
have focused on the role of a nucleoside in the construction of
antioxidants even in the case of some reports proving the
benefit of a three-dimensional (3D) configuration for the drug
design.17 We have addressed dendritic antioxidants by applying
pentaerythritol as the core to achieve tetraferulate with
tetrahedral configuration, resulting in the inhibitory effect
Received: August 7, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 1. Nucleoside antioxidants synthesized by cytidine, uridine, adenosine, and guanosine as the core and p-hydroxybenzoic, ferulic, and caffeic
acids as antioxidative features.
against AAPH-induced oxidation of DNA being more than
four times better than ferulic acid itself.18 This result implies
that multiple antioxidative ligands with 3D configuration rather
than a planar configuration in the traditional view of
antioxidant might generate additional effects against DNA
oxidation.
As depicted in Figure 1, we herein adopt cytidine, uridine,
adenosine, or guanosine as the core for linking with ferulic or
caffeic acids,19 aiming at testing whether nucleoside as the core
is still capable of generating additional contribution to the
antioxidative effect against AAPH-induced DNA oxidation.
Besides, antioxidative nucleoside hybrids are envisioned to
incorporate with DNA strand more tightly than with other
dendritic antioxidants.20 Finally, applications of these nucleo-
side hybrids for scavenging 2,2′-diphenyl-1-picrylhydrazyl
(DPPH) and the galvinoxyl radical lead to an explanation for
understanding antioxidative effects of these compounds.
uration offers an additional antioxidative effect for the
compound.18 Herein, we envision that ferulic acid bridging
with nucleoside could overcome pentaerythrityl tetraferulate to
exhibit much higher antioxidative effect. There are several
commonly used methods for attaching functional substituents
to −OHs and −NH2 in nucleosides.21 As a first step, we try to
perform the reactions of both −OHs and −NH2 in cytidine
with acetylferuloyl chloride in the media of pyridine (condition
1 in eq 1 of Figure 2), followed by the deacetylation with
CH3COONH4 in the mixture of THF/CH3OH/H2O
(condition 2 in eq 1 of Figure 2). This protocol allows
preparing 1, 2, 3, 4, 15, 16, and 19 (see Figure 1) with
cytidine, uridine, adenosine, or guanosine being the cores.
Further investigation commences with attaching different
numbers of ferulic acid moieties to the cytidine core (the
antioxidative effect of ferulic acid is moderate, and thus, is
selected to be the antioxidative moiety, whereas the usage of
caffeic acid might lead to an unobvious difference on the
comparison of antioxidant effect during the increase of caffeic
acid moiety). We thereby focus on the protection of −OH or
−NH2, aiming at esterifying −OH or amidating −NH2 in
RESULTS AND DISCUSSION
■
Synthesis. We have employed ferulic acid to construct
pentaerythrityl tetraferulate and found that the 3D config-
B
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 2. Synthetic routines of nucleoside antioxidants which are numbered, whereas reagents as well as intermediates are identified by their
nomenclatures.
cytidine selectively. Now that 3 −OHs and a −NH2 in cytidine
exert similar reactivity toward acetylferuloyl chloride, the
−NH2 in cytidine is protected with di-tert-butyl dicarbonate to
afford 17. Three −OHs in 17 react with acetylferuloyl chloride,
followed by the deacetylation to give 6 as shown in eq 2 of
Figure 2. The Boc group in 6 can be removed by mixing with
silica and refluxing in toluene, and the afforded 5 is a cytidine
antioxidant with triferulate. So, the comparison of 5 with 6 is
able to reveal the contribution of −NH2 to the antioxidative
effect of a cytidine antioxidant. Moreover, double secondary
−OHs in 17 are ketalized by reacting with triethyl
orthoformate to give 18, and the primary −OH in 18 is
esterified with acetylferuloyl chloride, and then, the deacety-
lation allows to access 7 (eq 3 of Figure 2). The acetonide
moiety in 7 is removed under acidic condition to afford 9,
followed by the deprotection from −NH2 to give 10 as the
cytidine antioxidant with only one ferulic acid moiety attached.
Or, the same operation is performed on 7 to produce 8.
Despite the ribose moiety, the effects of N-fused hetero-
cycles such as cytosine and adenine are explored to connect
with ferulic acid moiety. The −NH2 in cytosine or adenine is
amidated with acetylferuloyl chloride, and then, the
deacetylation leads to the formations of 11 and 12 (13 is
achieved by using diacetylcaffeoyl chloride as the reagent).
Also, α-aminopyrinde is used to prepare 14, aiming at
comparing the antioxidative effect with that of 11.
Effect of Nucleoside. Activities of nucleoside antioxidants
are tested in the experimental system of peroxyl radical-
mediated DNA oxidation, and AAPH acts as the peroxyl
radical resource because it can provide peroxyl radicals at a
stable rate, Rg = 1.40 × 10−6 [AAPH] s−1 (37 °C).22 Oxidative
products from DNA can be determined spectrometrically after
colorizing with TBA, and thus, they are also called
C
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 3. Stoichiometric factors (n) of nucleoside antioxidants with four or three ferulic acid moieties, and molecular docking results from
interaction of pentaerythrityl tetraferulate or 1 with a DNA strand.
thiobarbituric acid reactive species (TBARS, λmax = 535 nm).23
In the blank experiment, the formation of TBARS is found to
be dependent linearly upon the reaction period (see Figure S1
in the Supporting Information), but the addition of nucleoside
antioxidant can hinder the formation of TBARS for a period.
The produced inhibition period (tinh) correlates proportionally
with the concentration of the nucleoside antioxidant as shown
in Figure S1 of the Supporting Information. Taking compound
1 as the example, eq 1 outlines the quantitative relationship of
tinh with the concentration of 1, and equations of tinh(min) ∼
[other nucleoside antioxidants (μM)] are collected in Table S1
of the Supporting Information.
ferulic acid (n of ferulic acid is 1.75,18 and ratio of n between 1
and ferulic acid is 12.16/1.75 = 6.95). As depicted in Figure 3,
n of pentaerythrityl tetraferulate (8.55) is only ∼5 times better
than that of ferulic acid (8.55/1.75 = 4.89) although
pentaerythrityl tetraferulate is also composed of four ferulic
acid moieties. The reason why the antioxidative effect of 1 (a
cytidine as the core bearing four ferulic moieties) is higher than
that of pentaerythrityl tetraferulate (pentaerythritol as the core
bearing the same number of ferulic acid moieties) encourages
us to explore the interaction between an antioxidative molecule
with a DNA strand in the manner of molecular docking,25
during which free energy (ΔG) is used to be the index for
predicting the extent of the intercalation of 1 or pentaerythrityl
tetraferulate with a DNA strand.26 Based on the molecular
mechanics-generalized Born surface area,27 −19.93 kcal/mol of
ΔG is generated by pentaerythrityl tetraferulate binding the
groove of a DNA strand, but a much lower ΔG (−40.35 kcal/
mol) is produced by 1 intercalating with the DtNA groove.
Hence, 1 interacts with the DNA groove more tightly than
pentaerythrityl tetraferulate. It is also found that 1 can cover
the surface of DNA, avoiding the attacks from radicals (see
Figure 3). The similarity of the n value of 1 (12.16) with that
of 3 (11.75) indicates that both cytidine and guanosine play
the same role in the generation of antioxidative effects for 1
and 3. Alternatively, the n value of 2 (13.37), higher than that
of 3 (11.75), is attributable to different positions of −NH2 in
adenosine and guanosine for bearing the ferulic acid moiety.
Finally, the low n value of 4 (7.34) is due to uridine bearing
three ferulic acid moieties only.
t
inh(min) = 3.62 min μM−1 [1(μM)]98.4
(1)
We next set out to reveal the meaning of eq 1 by comparing
it with eq 2, which is obtained from AAPH-induced oxidation
of polyunsaturated fatty acid in the manner of chemical
kinetics. Here, Ri is the initiation rate of the radical-induced
oxidation, and n refers to the stoichiometric factor, viz., the
number of the radical-propagation terminated by one molecule
of the antioxidant.22
t
inh(min) = (n/Ri) [antioxidant (μM)]
(2)
Eq 2 allows us to calculate the stoichiometric factor (n) of
the antioxidant if Ri can be estimated. Owing to DNA sodium
salt and AAPH both being dissolved in the water phase, the
initiation rate of DNA oxidation (Ri) is safely assumed to be
equal to the radical-generation rate as reported in our previous
work, viz., Ri = Rg = 1.40 × 10−6 s−1 × 40.0 mM = 5.60 × 10−5
mM s−1 = 3.36 μM min−1 (three significant digits are used
because the concentration of AAPH solution is accurate up to
three digits, 40.0 mM).24 The product of the coefficient in eq 1
(3.62 min μM−1, meaning that tinh can be increased to 3.62 min
by keeping on adding 1 μM of 1) and Ri (3.36 μM min−1)
gives 12.16 as the stoichiometric factor (n) of 1. This result
means that 1 is able to terminate ∼12 radical-propagations in
the protection of DNA against AAPH-induced oxidation, and
the antioxidative effect of 1 is ∼7 times better than that of
Effect of the Number of Ferulic Acid Moieties. Next,
we explore the influence of the number of antioxidative
moieties on the stoichiometric factor (n) of the nucleoside
antioxidant, during which cytidine bearing different numbers of
ferulic acid moieties acts as the model compounds. First, after
−NH2 in cytidine is protected with the t-butyoxycarbonyl
group (Boc), all of the −OHs in 17 are esterified with
acetylferuloyl chloride. Then, the deacetylation and removal of
the Boc group afford 5 as the representative cytidine
D
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 4. Influence of the number of ferulic acid moieties on the stoichiometric factor (n) of cytidine antioxidants, and molecular docking results
from the interaction of ferulic acid, 10, or 5 with a DNA strand.
Figure 5. Influence of protection of hydroxyl and amino groups in cytidine on the stoichiometric factor (n), and molecular docking results from
intercalation of 11 or 12 with a DNA strand.
antioxidant with three ferulic acid moieties. Subsequently,
double secondary −OHs in 17 are protected with ketalization,
and the primary −OH is left to be esterified with acetylferuloyl
chloride, followed by the deacetylation to give 10 as the
representative cytidine antioxidant with only one ferulic acid
moiety. As depicted in Figure 4, with the stoichiometric factors
(n) of 10 and 5 (see Figure 4) in hand, the number of ferulic
acid moieties in cytidine ({feruloyl}) is assigned to be an
independent variable, and the stoichiometric factor is defined
as the function to set up the relationship between n and
{feruloyl}. When the {feruloyl} of 1, 5, and 10 are assigned to
be 4, 3, and 1 for referring cytidine bearing 4, 3, and 1 ferulic
acid moieties, respectively, the n values of 1, 5, and 10
correlate with the {feruloyl} as an exponential growth function
shown in eq 3.
of n value from ferulic acid to 10 with eq 3 can wholly express
the influence of the number of ferulic acid moieties on the n
value. It can be assumed that the {feruloyl} of ferulic acid is 0
because of no cytidine being attached to ferulic acid.
Consequently, as depicted in the panel of n ∼ {feruloyl} of
Figure 4, the variation of n value from ferulic acid, 10, 5, to 1 is
fitted for the exponential growth function as shown in eq 4.
n = 2.03 e{feruloyl}/2.23
(R2 = 0.9982)
(4)
We have reported that the n value correlates linearly with the
{feruloyl} when alcohols are used to be the core (pentaery-
thritol bearing four ferulic acid moieties, glycerol bearing three
ferulic acid moieties, and ethylene glycol bearing two ferulic
acid moieties).18 Herein, in order to compare the contribution
of cytidine with that of alcohol as the core to the antioxidative
effect, we try to retreat the n values of alcoholic antioxidant
hybrids with {feruloyl} by the exponential growth function,
during which {feruloyl} of ferulic acid is also assigned to 0
(data are outlined in the panel of n ∼ {feruloyl} of Figure 4).
As a result, eq 5 is obtained.
n = 2.41 e{feruloyl}/2.43 − 0.33
(R2 = 1.0000)
(3)
Interestingly, the n of ferulic acid is 1.75, but the n value
increases to 3.31 of 10, in which a ferulic acid moiety attaches
to cytidine (cytidine does not exhibit any antioxidative
function in this case). Thus, the combination of the variation
E
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 6. Influence of carboxylic acids on the stoichiometric factor (n) of cytidine and pentaerythrityl antioxidants, and molecular docking results
from the interaction of 16 with a DNA strand (n value of pentaerythrityl tetraferulate is 8.55 as cited from ref 18).
n = 1.59 e{feruloyl}/2.40 + 0.12
(R2 = 0.9985)
with that of 9 (2.53) along with 8 (2.72) versus 10 (3.31) also
(5)
reveals that the contributions from double secondary −OHs to
the antioxidative effect are not as high as that from −NH2
because the differences of n value between the aforementioned
couples are not as large as those from the couples of 8 versus 7,
6 versus 5, and 10 versus 9.
By the comparison of the coefficients in eq 4 (2.03) and eq 5
(1.59), one can find that the n value in the case of cytidine as
the core is more sensitive toward the variation of the number
of ferulic acid moieties than alcohol as the core. Thus, cytidine
is a more efficient core than alcohol for organizing
antioxidative moieties. Furthermore, the molecular docking
computation could provide an explanation for the high
antioxidative efficacy of cytidine as the core although the
salvation in the real experimental system is not taken into
consideration in the process of nucleoside antioxidants docking
with the DNA groove. The intercalation of a free ferulic acid
with DNA strand generates −14.01 kcal/mol of ΔG, whereas
one ferulic acid moiety bridging with cytidine lowers the ΔG to
−32.96 kcal/mol in the case of 10 intercalating with DNA
strand. Additionally, 5 containing three ferulic acid moieties
and 1 containing four ferulic acid moieties further decrease ΔG
to −33.98 and −40.35 kcal/mol, respectively. Lower ΔG
implicates that cytidine enables ferulic acid moiety to combine
with a DNA strand more tightly, and thus, to protect DNA
against AAPH-induced oxidation more efficiently.
Effect of Functional Group in Nucleosides. To further
clarify the contribution of substituents in cytidine to the
antioxidative effect, we select 17 as the starting material to
prepare 7. As shown in the column panel of Figure 5, the n
value of 7 (1.74) is similar to that of ferulic acid (1.75),
indicating that the antioxidative effect of 7 is contributed from
ferulic acid moiety only. Contrarily, the n value of 10 (3.31) is
higher than that of 7 (1.74), implying that both −NH2 and
three −OHs in cytidine play an active role for enhancing the
antioxidative effect of 10 although cytidine itself does not exert
any antioxidative effect. This conclusion is also supported by
the comparison of the n value of 8 (2.72) with that of 7 (1.74),
5 (7.95) versus 6 (6.89), and 10 (3.31) versus 9 (2.53),
respectively. Additionally, comparing the n value of 7 (1.74)
We next employ cytosine and adenine as the starting
materials to link with ferulic acid, accessing 11 and 12 for
exploring the influence of the heterocyclic moiety of
nucleoside (avoiding ribose moiety) on the antioxidative
effect. Of interest to us is that the n values of 11 (4.83) and 12
(5.71) are higher than that of 10 (3.31), and 13 exhibits a high
n value (8.13) when a caffeic acid moiety bridges with cytosine
as shown in Figure 5. It seems that ribose moiety in nucleoside
is not necessary in the construction of antioxidants. However,
the molecular docking computation reveals that 10 (composed
of a ribose moiety) exerts a relatively lower ΔG (−32.96 kcal/
mol) than those of 11 and 12 (−31.63 and −28.62 kcal/mol,
respectively). Thus, the ribose moiety might provide additional
affinity sites for 10 toward a DNA strand. In addition, 14
represents a model structure with fewer nitrogen atoms, and its
n value (1.72) is similar to that of ferulic acid (1.75). This
result implies that the multiple N atoms in cytosine are
beneficial for cytosine antioxidant intercalating with the groove
of a DNA strand.28
Effect of Carboxylic Acid. Considering the antioxidative
property of a nucleoside antioxidant resulting from the
carboxylic acid moiety, we next set out to change the kind of
carboxylic acid around the cytidine core. As shown in Figure 6,
four benzoyl groups attaching to cytidine cannot generate any
antioxidative effect for 19, whereas adding a −OH to the para-
position of the benzene ring in 19 affords 15 (structure in
Figure 1), whose n value increases to 4.74. So, it is still
phenolic −OH that provides the antioxidative effect herein.
The most notable change is found in the comparison of the
n value of 1 (12.16) with that of 16 (38.71, almost 20 times
F
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
better than caffeic acid because the n value of caffeic acid is
1.9518). The remarkable increase of the n value of 16 is
attributable to catechol moiety in caffeic acid instead of
guaiacol moiety in ferulic acid. Also, one caffeic acid moiety
linking with cytosine enhances the n value of 13 to 8.13,
whereas one ferulic acid moiety linking with cytosine makes
the n value of 10 only 3.31. These results are in agreement with
a previous report on catechol moiety beneficial for trapping
radicals.29 In addition, the molecular docking computation
reveals that the lowest ΔG (−45.11 kcal/mol) is resulted from
the intercalation of 16 with a DNA strand, and part of caffeic
acid moieties in 16 can cover the surface of a DNA strand,
avoiding DNA being attacked by radicals. Besides, we measure
the binding constants (K) of 1 and 16 with a DNA strand by
using UV titration (see the Supporting Information).30 It is
found that 16 exerts slightly higher K (1.54 × 105 M−1) than 1
(1.26 × 105 M−1) in the formation of the antioxidant−DNA
complex.
By the comparison of the n value of pentaerythrityl
tetracaffeate (10.42, see Figure 6)18 with that of 16 (38.71),
it can be concluded that cytidine linking with four caffeic acid
moieties exerts higher efficacy against DNA oxidation than
pentaerythritol bearing the same number of caffeic acid
moieties. Furthermore, we connect the n value of pentaery-
thrityl tetra(p-hydroxybenzoate) (5.68) with those of pentaer-
ythrityl tetraferulate (8.55) and pentaerythrityl tetracaffeate
(10.42) to obtain a straight line as shown in Figure 6. On the
other hand, when points of the n values of 15, 1, and 16 are
connected, an exponentially increasing line is displayed. Thus,
the nucleoside acts as the core to organize antioxidative
carboxylic acid moieties for achieving a high antioxidative
effect. This finding might be used to guide the design for
antioxidative drugs when they are required to intercalate with a
DNA strand.31
Scavenging DPPH and Galvinoxyl Radicals. Reactions
of antioxidants with DPPH or galvinoxyl radicals involve a
process of a hydrogen atom transfer or proton transfer
followed by electron (PCET) from the phenolic moieties, or
from −OHs in ribose or −NH2 in cytosine, to the N-centered
(DPPH) or O-centered (galvinoxyl) radicals. Both the
hydrogen transfer processes are dependent on enthalpic effects
related to the bond dissociation energies of the antioxidant and
on the number of abstractable hydrogen atoms.32 The rate
constants in quenching DPPH (kDPPH) and galvinoxyl
(kgalvinoxyl) radicals are measured following a method as
reported in our previous work (see the Supporting
Information).33 Owing to the potential influence of protonic
solvents on the hydrogen transfer between antioxidant and
DPPH or galvinoxyl radicals,29,34 CH3CN is used as the
solvent for dissolving both radicals and antioxidants.35,36
As shown in panel A of Figure 7, the values of kDPPH of 7−12
and 14 are close to that of ferulic acid, implying that the
properties of these compounds in trapping DPPH are
attributable to the single ferulic acid moiety. Similarly, as
shown in panel B of Figure 7, the values of kDPPH of 1−6 range
from 6 to 7 × 102 M−1 s−1, 3−4 times better than that of ferulic
acid. This is due to three or four ferulic acid moieties involved
in these compounds. However, 15 is not able to scavenge
either DPPH or galvinoxyl radicals, indicating that the radical-
scavenging property of the nucleoside antioxidant is dependent
upon the antioxidative structure (phenolic −OH with ortho-
OCH3, or double adjacent phenolic −OH) other than the total
number of phenolic −OH in the nucleoside antioxidant. On
Figure 7. Rate constants of nucleoside antioxidants for trapping
DPPH and galvinoxyl radical, and the correlation of the
stoichiometric factor (n) with rate constant, kDPPH and kgalvinoxyl
.
the other hand, the kDPPH values of 13 (2.06 × 103 M−1 s−1, ∼4
times better than that of caffeic acid, 5.62 × 102 M−1 s−1) and
16 (4.17 × 104 M−1 s−1, ∼2 magnitudes higher than that of
caffeic acid) reveal that catechol moiety in caffeic acid provides
high activity to quench radicals, and thus, to enhance the
antioxidative property of the nucleoside antioxidant. Scaveng-
ing DPPH is related to the dissociation of O−H in the
phenolic −OH and the formation of N−H in DPPH,37
whereas quenching galvinoxyl radical just correlates with
exchanging the H atom between the phenolic −OH and O-
centered radical in galvinoxyl. With the increase of the number
of ferulic acid moieties, many more phenolic −OHs are
involved in the nucleoside antioxidants, leading to the increase
of kgalvinoxyl. Even the −OHs in ribose or −NH2 in cytosine play
active roles in this case (kgalvinoxyl of 7 is 1.58 × 102 M−1 s−1,
whereas kgalvinoxyl of 10 increases to 4.78 × 102 M−1 s−1). In
particular, the combination of a caffeic acid moiety with
cytosine increases kgalvinoxyl of 13 to 9.48 × 103 M−1 s−1 and
kgalvinoxyl of 16 to 3.15 × 105 M−1 s−1. Alternatively, the kDPPH is
dependent upon the bond dissociation energy (BDE) of O−H.
The increase of the number of ferulic acid moieties in
nucleoside antioxidants cannot vary the BDE of the −OH in
ferulic acid; thus, the variation of kDPPH is not as remarkable as
that of kgalvinoxyl. However, catechol moiety in caffeic acid still
increases kDPPH of 16 to 4.17 × 104 M−1 s−1.
The aim of the measurements of kDPPH and kgalvinoxyl is to
explore the correlation of the radical-scavenging property with
antioxidative effect against DNA oxidation. Hence, we assign
pkDPPH and pkgalvinoxyl (−log kDPPH and −log kgalvinoxyl) as the
independent variables and the stoichiometric factors (n) of
nucleoside antioxidants as the function and perform the linear
regression analysis on n ∼ pkDPPH or pkgalvinoxyl. As shown in
G
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
in 12 mL of CH3OH/H2O, v/v = 5:1) was added dropwise. The
reaction mixture was allowed to be stirred at 50 °C for 12 h. After
completion of the reaction as monitored by TLC, the organic solvent
was evaporated under vacuum. The residue was purified by silica gel
chromatography to afford nucleoside antioxidants.
Figure 7, it is noteworthy that linear regression coefficients
(R2) for both n ∼ pkDPPH (R2 = 0.9355) and n ∼ pkgalvinoxyl (R2
= 0.9143) are higher than 0.90, suggesting that the n value of
the nucleoside antioxidant is linearly related to pkDPPH and
pkgalvinoxyl. The effect of the nucleoside antioxidant against
DNA oxidation is demonstrated to correlate with the radical-
quenching property of these compounds.
Compound 1 2′,3′,5′-Tri-O-feruloyl-4-N-feruloylcytidine.
CH2Cl2/CH3COOC2H5 (v/v = 1:5) as eluent to afford 601 mg of
1
a yellow solid, yield 83%, and total yield 63%. H NMR (400 MHz,
DMSO-d6): δ 11.06 (s, 1H), 9.75 (s, 4H), 8.27 (d, J = 7.4 Hz, 1H),
7.68−7.59 (m, 3H), 7.57 (d, J = 5.6 Hz, 1H), 7.46 (d, J = 7.2 Hz,
1H), 7.33 (s, 1H), 7.28 (s, 2H), 7.20 (s, 1H), 7.17−7.00 (m, 4H),
6.88−6.70 (m, 5H), 6.59−6.47 (m, 3H), 6.11 (d, J = 2.4 Hz, 1H),
5.77 (d, J = 2.2 Hz, 1H), 5.67−5.61 (m, 1H), 4.56 (d, J = 8.9 Hz,
2H), 4.48 (dd, J = 11.4, 6.7 Hz, 1H), 3.83 (s, 3H), 3.81 (s, 3H), 3.74
(s, 6H). 13C NMR (101 MHz, DMSO-d6): δ 166.7, 165.9, 163.8,
154.8, 150.8, 148.5, 147.2, 146.3, 144.4, 125.7, 124.2, 123.8, 117.4,
116.3, 115.9, 113.9, 111.9, 96.8, 91.2, 80.0, 73.4, 70.5, 63.5, 56.0, 55.9.
HRMS (ESI-TOF) m/z: [M + H]+ calcd for C49H45N3O17, 948.2822;
found, 948.2736.
CONCLUSIONS
■
Taken collectively, we herein show that nucleosides are capable
of organizing antioxidative carboxylic acids to form a steric
configuration for fitting with the DNA groove, and thus, to
inhibit peroxyl radical-induced DNA oxidation efficiently. It
must be emphasized that the measurement of stoichiometric
factor (n) is only a part of the antioxidant property, and an
antioxidant is required to rapidly suppress as many kinds of in
vivo radicals as possible. Based on the present results, it is
reasonably believed that nucleoside antioxidants will play a key
role in the protection of DNA against radical-mediated
oxidation, and the 3D configuration of antioxidant might
bring a breakthrough to overcome the planar configuration in
the traditional view on the antioxidant structure.
Compound 2 2′,3′,5′-Tri-O-feruloyl-6-N-feruloyladenosine.
CH2Cl2/CH3COOC2H5 (v/v = 1:9) as eluent to afford 431 mg of
1
a white solid, yield 74%, and total yield 43%. H NMR (400 MHz,
DMSO-d6): δ 10.88 (s, 1H), 9.69 (s, 2H), 9.65 (s, 1H), 9.62 (s, 1H),
8.79 (s, 1H), 8.71 (s, 1H), 7.70−7.53 (m, 4H), 7.32 (d, J = 3.2 Hz,
2H), 7.25 (d, J = 7.4 Hz, 2H), 7.15−7.04 (m, 5H), 6.85 (d, J = 8.1
Hz, 1H), 6.80 (d, J = 8.1 Hz, 1H), 6.76 (d, J = 8.1 Hz, 1H), 6.72 (d, J
= 8.2 Hz, 1H), 6.61 (d, J = 15.9 Hz, 1H), 6.54−6.49 (m, 2H), 6.47 (d,
J = 4.1 Hz, 1H), 6.31 (t, J = 5.6 Hz, 1H), 5.97−5.92 (m, 1H), 4.64 (d,
J = 7.4 Hz, 2H), 4.50 (dd, J = 12.4, 6.4 Hz, 1H), 3.84 (s, 3H), 3.82 (s,
3H), 3.77 (s, 3H), 3.73 (s, 3H). 13C NMR (101 MHz, DMSO-d6): δ
166.7, 165.7, 164.3, 151.9, 150.5, 148.6, 147.4, 146.3, 143.5, 139.8,
126.2, 125.5, 124.3, 123.8, 118.3, 116.2, 113.8, 112.5, 111.8, 86.5,
80.4, 72.8, 70.8, 63.4, 56.1. HRMS (ESI) m/z: [M + H]+ calcd for
C50H45N5O16, 972.2934; found, 972.2912.
EXPERIMENTAL SECTION
■
Materials and Instrumentation. AAPH, naked DNA sodium
salt, DPPH, and galvinoxyl radical were products of Acros Organics,
Geel, Belgium, and used as received. Solvents and reagents applied in
the synthesis were obtained commercially and used as such unless
noted otherwise, and the heating in the synthesis was performed in a
water bath. Products were identified by 1H and 13C NMR
spectroscopy (Bruker AVANCE III 400 MHz spectrometer), and
molecular weight was detected by high-resolution mass spectra
(HRMS) equipped with ESI as the ionization mode (Agilent 1290-
micrOTOF Q II). All of the spectra were outlined in the Supporting
Information, and the molecular docking computation was carried out
Compound 3 2′,3′,5′-Tri-O-feruloyl-2-N-feruloylguaosine.
CH2Cl2/CH3COOC2H5 (v/v = 1:9) as eluent to afford 456 mg of
1
a white solid, yield 77%, and total yield 47%. H NMR (400 MHz,
DMSO-d6): δ 12.39 (s, 1H), 11.68 (s, 1H), 9.77 (s, 1H), 9.69 (s, 3H),
8.34 (s, 1H), 7.72 (d, J = 15.6 Hz, 1H), 7.64 (t, J = 15.4 Hz, 2H), 7.57
(d, J = 15.8 Hz, 1H), 7.33 (s, 2H), 7.25 (d, J = 5.7 Hz, 2H), 7.14 (dd,
J = 16.3, 8.3 Hz, 3H), 7.06 (d, J = 8.3 Hz, 1H), 6.87 (d, J = 8.3 Hz,
1H), 6.82−6.73 (m, 3H), 6.71 (d, J = 8.1 Hz, 1H), 6.59 (d, J = 15.9
Hz, 1H), 6.54 (d, J = 15.9 Hz, 1H), 6.47 (d, J = 16.0 Hz, 1H), 6.27
(d, J = 5.6 Hz, 1H), 6.07 (t, J = 5.8 Hz, 1H), 5.82−5.78 (m, 1H),
4.69−4.59 (m, 2H), 4.52 (d, J = 6.0 Hz, 1H), 3.83 (s, 3H), 3.82 (s,
3H), 3.78 (s, 3H), 3.72 (s, 3H). 13C NMR (101 MHz, DMSO-d6): δ
168.2, 167.6, 166.7, 165.9, 155.2, 150.3, 149.2, 148.4, 147.4, 146.3,
140.0, 126.8, 125.9, 124.1, 123.6, 122.1, 121.0, 119.4, 116.1, 115.9,
114.2, 113.4, 112.1, 111.9, 85.9, 80.5, 73.1, 71.1, 63.6, 56.1. HRMS
(ESI) m/z: [M + H]+ calcd for C50H45N5O17, 988.2883; found,
988.2840.
̈
by using the software of Schrodinger Suites 2018.
Synthesis of Nucleoside Antioxidants with Carboxylic Acids
Attaching to Both Amino and Hydroxyl Groups. The acetylation
of −OH in ferulic, caffeic, or p-hydroxybenzoic acids was performed
following a reported method,18 and then, the −COOH was
chloridated in refluxing SOCl2 with an oil bath for 3 h. After the
surplus SOCl2 was removed under vacuum, the afforded acyl chloride
was used without further purification. Compounds 1, 2, 3, 4, 15, and
16 were achieved by the reaction of nucleoside with acyl chloride as
per the following description. To 20 mL of pyridine (dried over CaH2
in advance) was added 1.5 mmol nucleoside (cytidine, uridine,
adenosine, or guanosine), and 7.5 mmol acyl chloride [acetylferuloyl
chloride, acetylbenzoyl chloride, or diacetylcaffeoyl chloride, dissolved
in 10 mL of dried tetrahydrofuran (THF)] was added dropwise at 0
°C under nitrogen atmosphere. The reaction mixture was allowed to
be stirred at room temperature for 10 h. After completion of the
reaction as indicated by TLC, THF and pyridine were removed under
vacuum. The residue was dissolved in 100 mL of ethyl acetate, and
the organic phase was washed with saturated NaHCO3 solution (50
mL × 2) and diluted HCl solution (50 mL × 2), respectively, and
then, dried over MgSO4. After the organic solvent was evaporated
under vacuum, the residue was purified by flash column
chromatography with CH2Cl2/CH3COOC2H5 (v/v = 2:1) being
the eluent to afford 1.27 g of acetylated 1 as a pale yellow solid, yield
76%. Similarly, 1.01 g of acetylated 2 was achieved as a white solid,
yield 58%; 1.06 g of acetylated 3 as a white solid, yield 61%; 1.23 g of
acetylated 4 as a white solid, yield 91% (6.0 mmol acetylferuloyl
chloride as regent); 1.14 g of acetylated 15 as a white solid, yield 85%;
and 1.34 g of acetylated 16 as a yellow solid, yield 73%.
Compound
4 2′,3′,5′-Tri-O-feruloyluridine. CH2Cl2/
CH3COOC2H5 (v/v = 1:5) as eluent to afford 710 mg of a white
solid, yield 92%, and total yield 83%. H NMR (400 MHz, DMSO-
1
d6): δ 11.52 (d, J = 2.1 Hz, 1H), 9.70 (s, 1H), 9.69 (s, 1H), 9.67 (s,
1H), 7.84 (d, J = 8.2 Hz, 1H), 7.62 (d, J = 15.9 Hz, 2H), 7.56 (d, J =
15.8 Hz, 1H), 7.33 (d, J = 1.8 Hz, 1H), 7.31 (d, J = 1.8 Hz, 1H), 7.27
(d, J = 1.8 Hz, 1H), 7.15 (dd, J = 8.2, 1.7 Hz, 1H), 7.09 (dd, J = 8.2,
1.8 Hz, 1H), 7.04 (dd, J = 8.3, 1.8 Hz, 1H), 6.80 (d, J = 8.2 Hz, 1H),
6.74 (d, J = 8.1 Hz, 1H), 6.71 (d, J = 8.2 Hz, 1H), 6.58 (d, J = 15.9
Hz, 1H), 6.53 (d, J = 16.0 Hz, 1H), 6.48 (d, J = 15.9 Hz, 1H), 6.08
(d, J = 5.2 Hz, 1H), 5.73 (dd, J = 8.1, 2.1 Hz, 1H), 5.71−5.67 (m,
1H), 5.62−5.55 (m, 1H), 4.54 (dd, J = 10.5, 2.7 Hz, 1H), 4.51−4.42
(m, 2H), 3.81 (s, 3H), 3.75 (s, 3H), 3.72 (s, 3H). 13C NMR (101
MHz, DMSO-d6): δ 166.7, 165.9, 163.5, 150.8, 149.5, 148.4, 147.2,
146.3, 141.9, 125.9, 124.1, 123.7, 116.0, 114.2, 113.5, 111.7, 102.9,
88.4, 79.7, 72.6, 70.5, 63.6, 56.1. HRMS (ESI) m/z: [M + H]+ calcd
for C39H36N2O15, 773.2188; found, 773.2174.
The deacetylation was carried out by using CH3COONH4. In 20
mL of THF was dissolved 1.0 mmol acetylated 1, 4, or 15 (0.6 mmol
acetylated 2, 3, or 16), and 10 times mmol CH3COONH4 (dissolved
Compound 15 2′,3′,5′-Tri-O-p-hydroxybenzoyl-4-N-p-hydroxy-
benzoylcytidine. CH2Cl2/CH3COOC2H5 (v/v = 1:5) as eluent to
H
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
1
afford 636 mg of a white solid, yield 88%, and total yield 75%. H
NMR (400 MHz, DMSO-d6): δ 11.12 (s, 1H), 10.40 (s, 4H), 8.25 (d,
J = 7.4 Hz, 1H), 7.94 (d, J = 8.8 Hz, 2H), 7.87 (d, J = 8.7 Hz, 2H),
7.76 (d, J = 8.7 Hz, 4H), 7.37 (d, J = 7.3 Hz, 1H), 6.87−6.81 (m,
4H), 6.78 (dd, J = 8.6, 1.1 Hz, 4H), 6.15 (d, J = 3.1 Hz, 1H), 5.88
(dd, J = 6.2, 3.2 Hz, 1H), 5.86−5.82 (m, 1H), 4.71 (dd, J = 9.9, 5.7
Hz, 1H), 4.65 (dd, J = 12.0, 3.4 Hz, 1H), 4.58 (dd, J = 12.0, 5.4 Hz,
1H). 13C NMR (101 MHz, DMSO-d6): δ 165.7, 164.8, 162.9, 154.7,
132.2, 131.4, 123.9, 120.2, 119.5, 115.8, 97.1, 92.0, 79.7, 73.9, 70.7,
63.7. HRMS (ESI) m/z: [M + H]+ calcd for C37H29N3O13, 724.1773;
found, 1447.3343 (detected as a protonated dimer).
Compound 7 2′,3′-O-Isopropylidene-5′-O-feruloyl-4-N-tert-bu-
toxycarbonylcytidine. With p-toluenesulfonic acid being the catalyst,
17 was allowed to be stirred in acetone solution of triethyl
orthoformate at room temperature for 24 h to afford 18, which was
employed to prepare 7. Briefly, 575 mg of 17 (1.5 mmol) and 508 mg
of acetylferuloyl chloride (2 mmol) were applied to prepare acetylated
7 following the synthetic procedure of acetylated 1, in which CH2Cl2/
CH3COOC2H5 (v/v = 5:1) was used as the eluent to afford 843 mg
of acetylated 7 as a white solid, yield 93%. Then, 601 mg of acetylated
7 (1.0 mmol) was utilized to perform the deacetylation with
CH3COONH4 (10 mmol), and 481 mg of 7 was obtained as a
white solid, yield 86%, with CH2Cl2/CH3COOC2H5 (v/v = 1:1) as
the eluent. 1H NMR (400 MHz, DMSO-d6): δ 10.43 (s, 1H), 9.64 (s,
1H), 8.10 (d, J = 7.2 Hz, 1H), 7.51 (d, J = 15.7 Hz, 1H), 7.30 (s, 1H),
7.09 (d, J = 8.0 Hz, 1H), 6.99 (d, J = 6.3 Hz, 1H), 6.79 (d, J = 8.3 Hz,
1H), 6.40 (d, J = 15.9 Hz, 1H), 5.82 (s, 1H), 5.07 (d, J = 5.7 Hz, 1H),
4.87 (d, J = 5.6 Hz, 1H), 4.41 (d, J = 8.2 Hz, 2H), 4.36−4.26 (m,
1H), 3.82 (s, 3H), 1.51 (s, 3H), 1.42 (s, 9H), 1.31 (s, 3H). 13C NMR
(101 MHz, CDCl3): δ 166.5, 163.3, 154.7, 151.1, 148.4, 146.9, 145.9,
126.4, 123.3, 115.0, 114.0, 113.9, 109.5, 96.8, 94.8, 86.5, 85.4, 82.8,
81.5, 64.4, 55.9, 27.9, 25.2. HRMS (ESI) m/z: [M + H]+ calcd for
C27H33N3O10, 560.2239; found, 560.2239.
Compound 16 2′,3′,5′-Tri-O-caffeoyl-4-N-caffeoylcytidine.
CH3COOC2H5/CH3OH (v/v = 100:1) as eluent to afford 417 mg
of a brown solid, yield 78%, and total yield 57%. 1H NMR (400 MHz,
DMSO-d6): δ 11.11 (s, 1H), 9.68 (s, 2H), 9.65 (s, 1H), 9.62 (s, 1H),
9.31 (s, 1H), 9.17 (s, 1H), 9.16 (s, 2H), 8.24 (d, J = 7.5 Hz, 1H),
7.59−7.50 (m, 4H), 7.45 (d, J = 7.4 Hz, 1H), 7.05 (dd, J = 21.7, 9.1
Hz, 5H), 6.94 (d, J = 4.8 Hz, 3H), 6.78 (dd, J = 11.2, 8.3 Hz, 2H),
6.74−6.66 (m, 3H), 6.34 (d, J = 6.0 Hz, 1H), 6.28 (d, J = 15.9 Hz,
2H), 6.08 (d, J = 3.3 Hz, 1H), 5.75−5.70 (m, 1H), 5.63 (t, J = 6.2 Hz,
1H), 4.53 (d, J = 13.0 Hz, 1H), 4.49−4.41 (m, 1H), 4.37 (t, J = 4.8
Hz, 1H). 13C NMR (101 MHz, DMSO-d6): δ 166.7, 165.8, 163.8,
161.9, 154.7, 150.1, 149.8, 147.4, 146.4, 144.7, 125.7, 122.6, 116.8,
115.1, 114.9, 113.3, 112.5, 96.7, 91.4, 80.0, 73.4, 70.4, 63.5. HRMS
(ESI) m/z: [M + H]+ calcd for C45H37N3O17, 892.2196; found,
892.2136.
Compound 8 2′,3′-O-Isopropylidene-5′-O-feruloylcytidine. Sim-
ilar to the preparation of 5 from 6, 279 mg of 7 (0.5 mmol) was
refluxed in the mixture of 5.0 equiv (w/w) silica gel and 20 mL of
toluene for 10 h under nitrogen atmosphere, and CH3COOC2H5/
CH3OH (v/v = 30:1) was used as the eluent to give 80 mg of 8 as a
Synthesis of Nucleoside Antioxidants with Different
Numbers of Ferulic Acid. In order to achieve cytidine antioxidants
composed of <4 carboxylic acid moieties, hydroxyl and amino groups
in cytidine were selectively protected. In brief, cytidine was allowed to
be stirred in N,N-dimethylformamide (DMF) solution of di-tert-butyl
dicarbonate at 45 °C for 48 h to give 17 (N-Boccytidine). Then, 17
(1.5 mmol) and acetylferuloyl chloride (6.0 mmol) were performed
the synthetic routine as the preparation of acetylated 1, and 1.26 g of
white solid (yield 84%) was obtained after the purification by silica gel
chromatography with CH2Cl2/CH3COOC2H5 (v/v = 1:1) being the
eluent. In 20 mL of THF was dissolved the aforementioned white
solid (997 mg, 1.0 mmol) and CH3COONH4 (10 mmol, dissolved in
12 mL of CH3OH/H2O, v/v = 5:1) to carry out the deacetylation.
Compound 6 2′,3′,5′-Tri-O-feruloyl-4-N-tert-butoxycarbonylcy-
tidine. CH2Cl2/CH3COOC2H5 (v/v = 1:3) as the eluent to afford
810 mg of a white solid, yield 93%. 1H NMR (400 MHz, DMSO-d6):
δ 10.52 (s, 1H), 9.67 (s, 3H), 8.19 (d, J = 7.6 Hz, 1H), 7.65−7.54 (m,
3H), 7.33 (s, 1H), 7.28 (s, 2H), 7.14 (d, J = 8.4 Hz, 1H), 7.11−7.03
(m, 3H), 6.79 (d, J = 8.1 Hz, 1H), 6.73 (d, J = 8.1 Hz, 2H), 6.57−
6.46 (m, 3H), 6.09 (d, J = 3.8 Hz, 1H), 5.74−5.68 (m, 1H), 5.62 (t, J
= 6.0 Hz, 1H), 4.54 (t, J = 7.8 Hz, 2H), 4.50−4.40 (m, 1H), 3.81 (s,
3H), 3.74 (s, 6H), 1.46 (s, 9H). 13C NMR (101 MHz, CDCl3): δ
166.5, 165.8, 162.9, 148.6, 146.9, 144.0, 126.4, 123.4, 114.9, 113.7,
109.8, 95.9, 88.7, 83.2, 80.7, 74.1, 70.5, 63.1, 55.9, 27.9. HRMS (ESI)
m/z: [M + H]+ calcd for C44H45N3O16, 872.2873; found, 872.3145.
Compound 5 2′,3′,5′-Tri-O-feruloylcytidine. The Boc group at
amino group in 6 was removed to afford 5. In 20 mL of toluene was
dissolved 348 mg of 6 (0.4 mmol) and to it was added 5.0 equiv (w/
w) silica gel. The reaction mixture was allowed to be refluxed for 10 h
under stirring. After completion of the reaction as indicated by TLC,
the silica gel was collected and washed with petroleum ether, and then
loaded on column chromatography with CH3COOC2H5/CH3OH (v/
v = 20:1) being the eluent to afford 211 mg of 5 as a white solid, yield
68%. 1H NMR (400 MHz, DMSO-d6): δ 9.68 (s, 3H), 7.74 (d, J = 7.5
Hz, 1H), 7.59 (m, 3H), 7.38 (s, 2H), 7.34 (d, J = 1.7 Hz, 1H), 7.28
(dd, J = 10.7, 1.6 Hz, 2H), 7.15 (dd, J = 8.2, 1.8 Hz, 1H), 7.08−7.03
(m, 2H), 6.79 (d, J = 8.1 Hz, 1H), 6.72 (dd, J = 8.1, 3.9 Hz, 2H),
6.59−6.44 (m, 3H), 6.05 (d, J = 4.6 Hz, 1H), 5.77 (d, J = 7.4 Hz,
1H), 5.70−5.65 (m, 1H), 5.64−5.58 (m, 1H), 4.52 (dd, J = 10.4, 2.1
Hz, 1H), 4.47−4.38 (m, 2H), 3.81 (s, 3H), 3.74 (s, 3H), 3.73 (s, 3H).
13C NMR (101 MHz, DMSO-d6): δ 167.8, 166.8, 165.9, 155.3, 148.7,
1
white solid, yield 35%. H NMR (400 MHz, DMSO-d6): δ 9.66 (s,
1H), 7.67 (d, J = 6.8 Hz, 1H), 7.56 (d, J = 15.8 Hz, 1H), 7.34 (s, 1H),
7.30 (s, 2H), 7.13 (d, J = 8.0 Hz, 1H), 6.80 (d, J = 8.0 Hz, 1H), 6.49
(d, J = 15.9 Hz, 1H), 5.80−5.65 (m, 2H), 5.04 (d, J = 6.0 Hz, 1H),
4.87 (s, 1H), 4.40 (d, J = 8.9 Hz, 1H), 4.28 (d, J = 9.5 Hz, 2H), 3.82
(s, 3H), 1.49 (s, 3H), 1.30 (s, 3H). 13C NMR (101 MHz, DMSO-d6):
δ 166.8, 155.2, 149.9, 148.3, 146.0, 144.4, 125.9, 123.7, 115.9, 114.3,
113.3, 111.6, 94.8, 85.2, 84.6, 81.9, 64.7, 56.1, 27.4, 25.5. HRMS
(ESI) m/z: [M + H]+ calcd for C22H25N3O8, 460.1714; found,
460.1774.
Compound 9 5′-O-Feruloyl-4-N-tert-butoxycarbonylcytidine.
Compound 9 was also achieved from 7 with a hydrolysis of
acetonide. In 15 mL of THF was dissolved 279 mg of 7 (0.5 mmol),
and 11 mL of THF solution of HCl (1 mL of 37% hydrochloric acid
was mixed with 10 mL of THF) was added. The reaction mixture was
allowed to be stirred at room temperature for 10 h. After completion
of the reaction as indicated by TLC, the solvent was evaporated under
vacuum. The residue was purified by silica gel chromatography with
CH3COOC2H5/CH3OH (v/v = 50:1) being the eluent to afford 184
mg of 9 as a white solid, yield 71%. 1H NMR (400 MHz, DMSO-d6):
δ 10.40 (s, 1H), 9.65 (s, 1H), 8.09 (d, J = 7.6 Hz, 1H), 7.60 (d, J =
15.9 Hz, 1H), 7.34 (d, J = 1.8 Hz, 1H), 7.15 (dd, J = 8.2, 1.8 Hz, 1H),
7.04 (d, J = 7.6 Hz, 1H), 6.80 (d, J = 8.1 Hz, 1H), 6.54 (d, J = 15.9
Hz, 1H), 5.79 (d, J = 2.9 Hz, 1H), 5.61 (d, J = 5.0 Hz, 1H), 5.31 (d, J
= 6.2 Hz, 1H), 4.47 (dd, J = 12.3, 2.8 Hz, 1H), 4.36 (dd, J = 12.3, 5.4
Hz, 1H), 4.17−4.10 (m, 1H), 4.07 (dd, J = 8.0, 5.0 Hz, 1H), 3.99 (dd,
J = 11.8, 6.4 Hz, 1H), 3.83 (s, 3H), 1.44 (s, 9H). 13C NMR (101
MHz, DMSO-d6): δ 166.8, 163.4, 154.8, 152.5, 149.9, 148.3, 146.1,
144.9, 125.8, 123.7, 115.9, 114.3, 111.6, 95.0, 91.4, 81.4, 74.3, 69.7,
63.5, 56.0, 28.1. HRMS (ESI) m/z: [M + H]+ calcd for C24H29N3O10,
520.1926; found, 520.1924.
Compound 10 5′-O-Feruloylcytidine. The Boc group at the amino
group in 9 was removed to afford 10, and 156 mg of 9 (0.3 mmol)
was employed as the reagent. The operation was similar to the
synthesis of 5 and 8, and CH3COOC2H5/CH3OH (v/v = 20:1) was
used as the eluent to afford 37 mg of 10 as a white solid, yield 30%.
1H NMR (400 MHz, DMSO-d6): δ 9.66 (s, 1H), 7.62 (d, J = 6.2 Hz,
1H), 7.59 (d, J = 14.5 Hz, 1H), 7.35 (d, J = 1.6 Hz, 1H), 7.15 (dd, J =
8.2, 1.6 Hz, 1H), 6.81 (d, J = 8.1 Hz, 1H), 6.53 (d, J = 15.9 Hz, 1H),
5.98 (s, 2H), 5.80 (d, J = 3.9 Hz, 1H), 5.71 (d, J = 7.4 Hz, 1H), 5.43
(d, J = 2.9 Hz, 1H), 5.26 (d, J = 2.2 Hz, 1H), 4.43 (dd, J = 12.1, 2.9
Hz, 1H), 4.27 (dd, J = 12.1, 5.9 Hz, 1H), 4.06−4.02 (m, 2H), 3.99 (d,
J = 4.3 Hz, 1H), 3.83 (s, 3H). 13C NMR (101 MHz, DMSO-d6): δ
147.3, 146.5, 142.9, 124.8, 116.2, 113.3, 112.5, 111.6, 95.3, 90.0, 79.5,
72.9, 70.7, 63.7, 56.0, 55.9. HRMS (ESI) m/z: [M + H]+ calcd for
C39H37N3O14, 772.2348; found, 772.2346.
I
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
167.9, 166.8, 155.6, 148.4, 146.0, 141.8, 125.9, 123.6, 115.9, 114.4,
94.6, 90.4, 81.0, 73.8, 70.3, 64.1, 56.1. HRMS (ESI) m/z: [M + H]+
calcd for C19H21N3O8, 420.1401; found, 420.1401.
AAPH-Induced Oxidation of DNA. DNA and AAPH were
dissolved in phosphate buffered solution (PBS: 8.1 mM Na2HPO4,
1.9 mM NaH2PO4, 10.0 μM EDTA, pH = 7.4) to reach 2.24 mg/mL
and 400 mM as the stock solution, respectively. Solutions of DNA
(13.4 mL) and AAPH (1.5 mL) were mixed with a certain volume of
DMSO solution of nucleoside antioxidant. This mixture was aliquoted
into test tubes, with each containing 2.0 mL (final concentrations of
DNA and AAPH were 2.00 mg/mL and 40.0 mM, respectively, and
final concentrations of the nucleoside antioxidant are shown in the
corresponding chart in the Supporting Information. The same volume
of DMSO was contained in the control experiment in order to
eliminate the influence from DMSO). The test tubes were incubated
at 37 °C for initiating the oxidation of DNA, and three of them were
taken out every 2 hours and cooled immediately in ice-water. Then,
1.0 mL of PBS solution of 2-thiobarbituric acid (1.00 g of 2-
thiobarbituric acid and 0.40 g of NaOH were dissolved in 100 mL of
PBS) and 1.0 mL of 3.0% aqueous solution of trichloroacetic acid
were added, and the mixture was heated in boiling water for 15 min.
After the test tubes were cooled in ice-water, 1.5 mL of n-butanol was
added and shaken vigorously to extract colorful TBARS. The
absorbance of TBARS (λmax = 535 nm) was the average value from
three independent measurements with the experimental error less
than 10%. Equations of tinh ∼ [nucleoside antioxidant] were analyzed
by one-way ANOVA in Origin 7.0 professional Software, and p <
0.001 indicated a significance difference.18
Synthesis of Antioxidants with Heterocycles Being At-
tached by a Single Ferulic Acid. Compounds 11, 12, 13, and 14
were prepared by the following procedure. To a flask was added 222
mg of cytosine (2.0 mmol) and 635 mg of acetylferuloyl chloride (2.5
mmol), and then, 15 mL of pyridine (dried by CaH2 in advance) was
added at 0 °C under nitrogen atmosphere. The reaction mixture was
allowed to be stirred at room temperature for 10 h. After completion
of the reaction as indicated by TLC, 20 mL of water (cooled to 0 °C
in advance) was added and stirred for 1 h. After the precipitate was
collected and washed with diluted hydrochloric acid and water, the
crude product was recrystallized in the mixture of DMF and ethanol
(v/v = 1:4−5). Then, 682 mg of acetylated 11 was obtained as a pale
yellow solid, yield 83%. Similarly, 2 mmol adenine and 2.5 mmol
acetylferuloyl chloride were used to carry out the aforementioned
procedure to afford 637 mg of acetylated 12 as a white solid, yield
72%. Or, 2 mmol cytosine reacted with 2.5 mmol diacetylcaffeoyl
chloride to give 606 mg of acetylated 13 as a brown solid, yield 68%.
Subsequently, 2.0 mmol o-aminopyridine and 2.5 mmol acetylferuloyl
chloride were used to perform the aforementioned routine, and
CH2Cl2/CH3COOC2H5 (v/v = 6:1) was used as the eluent in silica
gel chromatography to afford 710 mg of acetylated 14 as a white solid,
yield 91%.
The deacetylation was performed as per the following procedure.
To 20 mL of DMF was added 329 mg of acetylated 11 (1.0 mmol)
and aqueous solution of CH3COONH4 (10 mmol, dissolved in 10
mL of water). The reaction mixture was allowed to be stirred at 50 °C
for 12 h. After completion of the reaction as monitored by TLC, the
reaction mixture was poured into 30 mL of water (cooled to 0 °C in
advance). The precipitate was collected and recrystallized with
ethanol to afford 258 mg of 11 as a yellow solid, yield 77%. Similarly,
230 mg of 12 (white solid, yield 74%) and 177 mg of 13 (brown solid,
yield 65%) were achieved.
Scavenging DPPH and Galvinoxyl Radicals. In 50 mL of
CH3CN (refluxing in P2O5 and distillation in advance) was dissolved
2.0 mg of DPPH (the absorbance at 517 nm was close to 1.0 with
εDPPH = 4.09 × 103 M−1 cm−1). The absorbance of the mixture of 1.9
mL of DPPH and 0.1 mL of CH3CN was recorded as the basic value.
Then, to 1.9 mL of CH3CN solution of DPPH was added 0.1 mL of
CH3CN solution of a nucleoside antioxidant and mixed promptly (the
final concentration of the nucleoside antioxidant in the mixture was
outlined in the corresponding chart in the Supporting Information).
The absorbance of the mixture was recorded successively till the value
reached equilibrium (the variation of absorbance less than 10%). The
concentration of DPPH at every time point was the average value
from three independent measurements with the experimental error
being less than 10%. The decay of the concentration of DPPH was
plotted versus the reaction period. Similarly, 0.85 mg of the galvinoxyl
radical was dissolved in 100 mL of CH3CN (the absorbance at 428
nm was close to 1.0 with εgalvinoxyl = 1.45 × 105 M−1 cm−1). To 1.9 mL
of CH3CN solution of the galvinoxyl radical was added 0.1 mL of
CH3CN solution of a nucleoside antioxidant (the final concentration
of the nucleoside antioxidant in the mixture was outlined in the
corresponding chart in the Supporting Information). The following
operation was the same as that in quenching DPPH. Equations
collected in the Supporting Information were analyzed by one-way
ANOVA in Origin 7.0 professional Software, and p < 0.001 indicated
a significance difference.33
Compound 11 4-N-feruloylcytosine. 1H NMR (400 MHz,
DMSO-d6): δ 11.48 (s, 1H), 10.84 (s, 1H), 9.70 (s, 1H), 7.85 (s,
1H), 7.59 (d, J = 14.1 Hz, 1H), 7.40−6.95 (m, 3H), 6.84 (s, 2H),
3.82 (s, 3H). 13C NMR (101 MHz, DMSO-d6): δ 166.1, 164.0, 156.6,
149.7, 148.3, 147.6, 143.8, 126.2, 122.7, 118.0, 116.2, 111.9, 95.2,
55.9. HRMS (ESI) m/z: [M + H]+ calcd for C14H13N3O4, 288.0979;
found, 288.0973.
Compound 12 6-N-feruloyladenine. 1H NMR (400 MHz,
DMSO-d6): δ 9.44 (s, 1H), 7.40 (s, 1H), 7.32 (d, J = 15.8 Hz,
1H), 7.13 (d, J = 1.5 Hz, 1H), 6.99 (dd, J = 8.2, 1.4 Hz, 2H), 6.79 (d,
J = 8.1 Hz, 1H), 6.43 (d, J = 15.8 Hz, 1H), 3.81 (s, 3H). 13C NMR
(101 MHz, DMSO-d6): δ 168.8, 162.9, 156.9, 154.3, 151.7, 149.1,
147.6, 142.1, 139.4, 133.5, 124.1, 121.3, 120.2, 119.0, 114.0, 56.4.
HRMS (ESI) m/z: [M + H]+ calcd for C15H13N5O3, 312.1091; found,
312.1072.
Compound 13 4-N-caffeoylcytosine. 1H NMR (400 MHz,
DMSO-d6): δ 11.51 (s, 1H), 10.86 (s, 1H), 9.43 (s, 1H), 7.84 (d, J
= 6.9 Hz, 1H), 7.51 (d, J = 15.6 Hz, 1H), 7.28 (d, J = 6.9 Hz, 1H),
7.02 (s, 1H), 6.93 (d, J = 8.0 Hz, 1H), 6.79 (d, J = 8.0 Hz, 2H), 6.69
(d, J = 15.6 Hz, 1H). 13C NMR (101 MHz, DMSO-d6): δ 166.1,
164.1, 156.6, 148.9, 147.5, 146.1, 144.0, 126.2, 122.1, 117.4, 116.2,
114.5, 95.2. HRMS (ESI) m/z: [M + H]+ calcd for C13H11N3O4,
274.0822; found, 274.0824.
Compound 14 N-(Pyridin-2-yl)ferulic Amide. The deacetylation of
acetylated 14 was similar to that for acetylated 1, and CH2Cl2/
CH3COOC2H5 (v/v = 1:1) was used as the eluent in silica gel
chromatography to afford 243 mg of 14 as a white solid, yield 90%.
1H NMR (400 MHz, DMSO-d6): δ 10.50 (s, 1H), 9.56 (s, 1H), 8.33
(dd, J = 4.8, 1.0 Hz, 1H), 8.24 (d, J = 8.4 Hz, 1H), 7.84−7.75 (m,
1H), 7.54 (d, J = 15.6 Hz, 1H), 7.19 (d, J = 1.7 Hz, 1H), 7.13−7.04
(m, 2H), 6.89 (d, J = 15.6 Hz, 1H), 6.84 (d, J = 8.1 Hz, 1H), 3.83 (s,
3H). 13C NMR (101 MHz, DMSO-d6): δ 165.0, 152.8, 149.2, 148.5,
141.9, 138.5, 126.6, 122.3, 119.6, 116.2, 114.0, 111.6, 55.9. HRMS
(ESI) m/z: [M + H]+ calcd for C15H14N2O3, 271.1077; found,
271.1085.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.joc.9b02104.
NMR and HRMS spectra of the synthetic compounds,
the variation of TBARS absorbance via the reaction
period, the relationship between the inhibition period
(tinh) and the concentration of antioxidant, measurement
of rate constant for trapping DPPH and galvinoxyl
radicals, and UV titration for the estimation of binding
constants of 1 and 16 with a DNA strand (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: zaiqun-liu@jlu.edu.cn.
J
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
LNA and Carba-ENA and the application of their modified
oligonucleotides as potential RNA targeted therapeutics. Chem. Rev.
2012, 112, 3808−3832.
(22) Bowry, V. W.; Stocker, R. Tocopherol-mediated peroxidation.
The prooxidant effect of vitamin E on the radical-initiated oxidation
of human low-density lipoprotein. J. Am. Chem. Soc. 1993, 115, 6029−
6044.
(23) Sugden, K. D.; Wetterhahn, K. E. Direct and hydrogen
peroxide-induced chromium(V) oxidation of deoxyribose in single-
stranded and double-stranded calf thymus DNA. Chem. Res. Toxicol.
1997, 10, 1397−1406.
(24) Tang, Y.-Z.; Liu, Z.-Q. Chemical kinetic behavior of
chlorogenic acid in protecting erythrocyte and DNA against radical-
induced oxidation. J. Agric. Food Chem. 2008, 56, 11025−11029.
(25) Lee, J. K.; Bosnakovski, D.; Toso, E. A.; Dinh, T.; Banerjee, S.;
Bohl, T. E.; Shi, K.; Orellana, K.; Kyba, M.; Aihara, H. Crystal
structure of the double homeodomain of DUX4 in complex with
DNA. Cell Rep. 2018, 25, 2955−2962.
ORCID
Zai-Qun Liu: 0000-0001-9573-0724
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) Biswas, P. K.; Chakraborty, S. Targeted DNA oxidation and
trajectory of radical DNA using DFT based QM/MM dynamics.
Nucleic Acids Res. 2019, 47, 2757−2765.
(2) Satoh, T.; Okamoto, S.-I.; Cui, J.; Watanabe, Y.; Furuta, K.;
Suzuki, M.; Tohyama, K.; Lipton, S. A. Activation of the Keap1/Nrf2
pathway for neuroprotection by electrophillic phase II inducers. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 768−773.
(3) Harris, I. S.; Brugge, J. S. The enemy of my enemy is my friend.
Nature 2015, 527, 170−171.
(4) Kaiser, J. Antioxidants could spur tumors by acting on cancer
gene. Science 2014, 343, 477.
(5) Alnajjar, K. S.; Sweasy, J. B. A new perspective on oxidation of
DNA repair proteins and cancer. DNA Repair 2019, 76, 60−69.
(6) Kojo, S. Oxygen is the key factor associated with the difference
between in vivo and in vitro effects of antioxidants. Proc. Natl. Acad.
Sci. U.S.A. 2012, 109, E2028.
(7) Martelli, G.; Giacomini, D. Antibacterial and antioxidant
activities for natural and synthetic dual-active compounds. Eur. J.
Med. Chem. 2018, 158, 91−105.
(8) Wu, J.; Sturla, S. J.; Burrows, C. J.; Fleming, A. M. Impact of
DNA oxidation on toxicology: From quantification to genomics.
Chem. Res. Toxicol. 2019, 32, 345−347.
(9) Arjunan, P.; Lin, X.; Tang, Z.; Du, Y.; Kumar, A.; Liu, L.; Yin, X.;
Huang, L.; Chen, W.; Chen, Q.; Ye, Z.; Wang, S.; Kuang, H.; Zhou,
L.; Xu, K.; Chen, X.; Zeng, H.; Lu, W.; Cao, Y.; Liu, Y.; Zhao, C.; Li,
X. VEGF-B is a potent antioxidant. Proc. Natl. Acad. Sci. U.S.A. 2018,
115, 10351−10356.
(10) Service, R. F. Molecular CT scan could speed drug discovery.
Science 2018, 362, 389.
(11) Lundgren, S. Focusing on relevance: CETSA-guided medicinal
chemistry and lead generation. ACS Med. Chem. Lett. 2019, 10, 690−
693.
(12) Liu, Z.-Q. Chemical methods to evaluate antioxidant ability.
Chem. Rev. 2010, 110, 5675−5691.
(13) Shao, J.; Geacintov, N. E.; Shafirovich, V. Oxidative
modification of guanine bases initiated by oxyl radicals derived from
photolysis of azo compounds. J. Phys. Chem. B 2010, 114, 6685−
6692.
(14) Egli, M.; Manoharan, M. Re-engineering RNA molecules into
therapeutic agents. Acc. Chem. Res. 2019, 52, 1036−1047.
(15) Ushimaru, R.; Liu, H.-w. Biosynthetic origin of the atypical
stereochemistry in the thioheptose core of albomycin nucleoside
antibiotics. J. Am. Chem. Soc. 2019, 141, 2211−2214.
(16) Yan, M.; Baran, P. S. Fighting evolution with chemical
synthesis. Nature 2016, 533, 326−327.
(26) Wang, L.; Wu, Y.; Deng, Y.; Kim, B.; Pierce, L.; Krilov, G.;
Lupyan, D.; Robinson, S.; Dahlgren, M. K.; Greenwood, J.; Romero,
D. L.; Masse, C.; Knight, J. L.; Steinbrecher, T.; Beuming, T.; Damm,
W.; Harder, E.; Sherman, W.; Brewer, M.; Wester, R.; Murcko, M.;
Frye, L.; Farid, R.; Lin, T.; Mobley, D. L.; Jorgensen, W. L.; Berne, B.
J.; Friesner, R. A.; Abel, R. Accurate and reliable prediction of relative
ligand binding potency in prospective drug discovery by way of a
modern free-energy calculation protocol and force field. J. Am. Chem.
Soc. 2015, 137, 2695−2703.
̈
(27) Ahinko, M.; Niinivehmas, S.; Jokinen, E.; Pentikainen, O. T.
Suitability of MMGBSA for the selection of correct ligand binding
modes from docking results. Chem. Biol. Drug Des. 2019, 93, 522−
538.
(28) Wang, S.-R.; Wang, J.-Q.; Fu, B.-S.; Chen, K.; Xiong, W.; Wei,
L.; Qing, G.; Tian, T.; Zhou, X. Supramolecular coordination-directed
reversible regulation of protein activities at epigenetic DNA marks. J.
Am. Chem. Soc. 2018, 140, 15842−15849.
(29) Foti, M. C.; Daquino, C.; Geraci, C. Electron-transfer reaction
of cinnamic acids and their methyl esters with the DPPH radical in
alcoholic solutions. J. Org. Chem. 2004, 69, 2309−2314.
(30) Bera, R.; Sahoo, B. K.; Ghosh, K. S.; Dasgupta, S. Studies on
the interaction of isoxazolcurcumin with calf thymus DNA. Int. J. Biol.
Macromol. 2008, 42, 14−21.
(31) Crespo, R. A.; Dang, Q.; Zhou, N. E.; Guthrie, L. M.; Snavely,
T. C.; Dong, W.; Loesch, K. A.; Suzuki, T.; You, L.; Wang, W.;
O’Malley, T.; Parish, T.; Olsen, D. B.; Sacchettini, J. C. Structure-
guided drug design of 6-substituted adenosine analogues as potent
inhibitors of Mycobacterium tuberculosis adenosine kinase. J. Med.
Chem. 2019, 62, 4483−4499.
(32) Ingold, K. U.; Pratt, D. A. Antioxidant chemistry in the 21st
century: A kinetics and mechanisms perspective. Chem. Rev. 2014,
114, 9022−9046.
(33) Wang, R.; Liu, Z.-Q. Ugi multicomponent reaction product:
The inhibitive effect on DNA oxidation depends upon the isocyanide
moiety. J. Org. Chem. 2013, 78, 8696−8704.
(17) Ye, W.; Li, A. Reaction combination opens up 3D molecular
diversity for drug discovery. Nature 2018, 560, 314−316.
(18) Zhou, R.; Liu, Z.-Q. Tetramer as efficient structural mode for
organizing antioxidative carboxylic acids: The case in inhibiting DNA
oxidation. Arch. Biochem. Biophys. 2017, 631, 1−10.
(34) Litwinienko, G.; Ingold, K. U. Solvent effects on the rates and
mechanisms of reaction of phenols with free radicals. Acc. Chem. Res.
2007, 40, 222−230.
(35) Bietti, M.; Cucinotta, E.; DiLabio, G. A.; Lanzalunga, O.; Lapi,
A.; Mazzonna, M.; Romero-Montalvo, E.; Salamone, M. Evaluation of
polar effects in hydrogen atom transfer reactions from activated
phenols. J. Org. Chem. 2019, 84, 1778−1786.
́
́
(19) Fasi, L.; Meo, D. F.; Kuo, C.-Y.; Buric, S. S.; Martins, A.; Kusz,
́
́
́
N.; Beni, Z.; Dekany, M.; Balogh, G. T.; Pesic, M.; Wang, H.-C.;
Trouillas, P.; Hunyadi, A. Antioxidant-inspired drug discovery:
Antitumor metabolite is formed in situ from a hydroxycinnamic
acid derivative upon free-radical scavenging. J. Med. Chem. 2019, 62,
1657−1668.
(36) Haidasz, E. A.; Meng, D.; Amorati, R.; Baschieri, A.; Ingold, K.
U.; Valgimigli, L.; Pratt, D. A. Acid is key to the radical-trapping
antioxidant activity of nitroxides. J. Am. Chem. Soc. 2016, 138, 5290−
5298.
(37) Foti, M. C. Use and abuse of the DPPH radical. J. Agric. Food
Chem. 2015, 63, 8765−8776.
(20) Wagner, J. R.; Lee, C. T.; Durrant, J. D.; Malmstrom, R. D.;
Feher, V. A.; Amaro, R. E. Emerging computational methods for the
rational discovery of allosteric drugs. Chem. Rev. 2016, 116, 6370−
6390.
(21) Zhou, C.; Chattopadhyaya, J. Intramolecular free-radical
cyclization reactions on pentose sugars for the synthesis of Carba-
K
DOI: 10.1021/acs.joc.9b02104
J. Org. Chem. XXXX, XXX, XXX−XXX