Inhibition of radical-induced DNA strand breaks by water-soluble constituents of coffee: Phenolics and caffeine metabolites

Article abstract of DOI:10.3109/10715762.2013.788167

Epidemiological studies have associated coffee consumption with an inverse risk of developing Parkinson's disease, hepatocellular carcinoma and cirrhosis. The molecular mechanisms by which low concentrations of the constituents of coffee measured in human plasma can reduce the incidence of such diseases are not clear. Using an in vitro plasmid DNA system and radiolytically generated reactive oxygen species under constant radical scavenging conditions, we have shown that coffee chlorogenic acid, its derivatives and certain metabolites of caffeine reduce some of the free radical damage sustained to the DNA. A reduction in the amount of prompt DNA single-strand breaks (SSBs) was observed for all compounds whose radical one-electron reduction potential is < 1.0 V. However, except for chlorogenic acid, the compounds were found to be inactive in reducing the amount of radical damage to the DNA bases. These results support a limited antioxidant role for such compounds in their interaction with DNA radicals.

Full text of DOI:10.3109/10715762.2013.788167

Free Radical Research, June–July 2013; 47(6–7): 480–487
© 2013 Informa UK, Ltd.
ISSN 1071-5762 print/ISSN 1029-2470 online
DOI: 10.3109/10715762.2013.788167
ORIGINAL ARTICLE
Inhibition of radical-induced DNA strand breaks by water-soluble constituents of
coffee: phenolics and caffeine metabolites
M. A. Rathod1, D. Patel1, A. Das1, S. R. Tipparaju2, S. S. Shinde2 & R. F. Anderson1,2
1School of Chemical Sciences, Faculty of Science, The University of Auckland, Auckland, New Zealand, and 2Auckland Cancer Society
Research Centre, Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand
Abstract
Epidemiological studies have associated coffee consumption with an inverse risk of developing Parkinson’s disease, hepatocellular carci-
noma and cirrhosis. The molecular mechanisms by which low concentrations of the constituents of coffee measured in human plasma can
reduce the incidence of such diseases are not clear. Using an in vitro plasmid DNA system and radiolytically generated reactive oxygen
species under constant radical scavenging conditions, we have shown that coffee chlorogenic acid, its derivatives and certain metabolites
of caffeine reduce some of the free radical damage sustained to the DNA. A reduction in the amount of prompt DNA single-strand breaks
(SSBs) was observed for all compounds whose radical one-electron reduction potential is ꢀ1.0 V. However, except for chlorogenic acid,
the compounds were found to be inactive in reducing the amount of radical damage to the DNA bases. These results support a limited anti-
oxidant role for such compounds in their interaction with DNA radicals.
Keywords: coffee, antioxidants, DNA strand breaks, fast chemical repair, reduction potentials
Introduction
and chocolate-based foods. Caffeine rapidly penetrates the
biological membranes and is efficiently absorbed from the
gastrointestinal tract and distributed throughout the body.
Caffeine does have some antioxidant activity in protecting
membranes against oxidative damage induced by reactive
oxygen species (ROS) [9], as do some of its metabolites
[10]. However, it is oxidative DNA damage which is con-
sidered to be a pathogenic event in the induction of many
cancers [11], and a reduction in the rate of formation of
such damage through an antioxidant reaction may lead to
a reduced risk of cancer. Increased levels of oxidative
damage to cytoplasmic DNA and RNA [12] are observed
in the brain cells of Parkinson’s disease patients. Mito-
chondrial DNA in the stratum of Parkinson’s disease
The prevalence of age-related diseases, such as cancer and
neurological disorders, increases as the population ages.
Hence, there is a considerable public awareness in the pos-
sible long-term health benefits derived from certain dietary
substances which appear to delay or prevent the onset of
such diseases. The perceived health benefits of coffee and
caffeine consumption have gained much current interest,
especially the reports of inverse dose relationships with
the onset in humans of hepatocellular carcinoma [1], cir-
rhosis [2,3] and Parkinson’s disease [4,5]. There are also
reports that ingestion of decaffeinated coffee decreases the
incidence of colorectal cancer [6]. The results from animal
experiments indicate that the reduced incidence of certain
cancers associated with coffee consumption could arise in
part to the effects of the major constituents of coffee (caf-
feine, chlorogenic acids and diterpenes) on the increase in
the activity of phase II biotransformation enzymes [7].
However, coffee consumption did not increase the activity
of colorectal, phase II enzyme, glutathione-S-transferase
in human volunteers, although glutathione levels were
increased in the colorectal mucosa and plasma [8].
⋅
patients is extremely susceptible to hydroxyl radical ( OH)
attack, resulting in multiple fragmentations that could be
a link between oxidative stress and apotopic neuronal cell-
death [13]. Other major constituents of coffee beans are
chlorogenic acid and the hydroxycinnamic acids such as
ferulic acid and caffeic acid derived from it. These com-
pounds exhibit a wide range of antioxidant activities, as
well as tyrosinase inhibition, in vitro [14]. It is unclear as
to how much in vivo antioxidant activity can be associated
with the hydroxycinnamic acids, due to their extensive
metabolism. The diterpene constituents of coffee, kahweol
and cafestol are known to be chemoprotective through
their inhibition of cytochrome P450 enzymes, thus reduc-
ing the activation of carcinogens [15] and by up-regulating
protective pathways such as the synthesis of glutathione
There may well be similarities in the preventive mech-
anisms against the induction of certain diseases and cell
destruction from coffee ingestion, due to the antioxidant
activity of coffee constituents. Caffeine is a major con-
stituent of coffee and is probably the most widely used
drug, being also one of the active ingredients of tea, cola
Correspondence: Robert F. Anderson, School of Chemical Sciences, Faculty of Science, The University of Auckland, Private Bag 92019,
Auckland 1142, New Zealand. Tel: ꢁ64 9 9238315. Fax: ꢁ64 9 3737422. E-mail: r.anderson@auckland.ac.nz
(Received date: 23 February 2013; Accepted date: 18 March 2013; Published Online: 18 April 2013)

Reduction in radical-induced DNA breaks by coffee 481
ꢁ
ꢁ
ꢁ
⋅
⋅
⋅
[16]. While there is a report that the water-soluble acetate
derivatives of these diterpenes protect DNA against H2O2-
induced cytotoxicity [17], such an antioxidant activity has
not been demonstrated for the water-insoluble compounds
themselves.
R
ꢁA⇆RꢁA (A , H )
(1)
⋅
The rate constants for OH radical scavenging by the
test compounds (reaction 2) were determined using com-
⋅
petition plots of the decrease in the absorption of the OH-
radical adduct in the presence of increasing concentrations
We are studying how dietary antioxidants may play a
role in ameliorating ROS damage to DNA. As the concen-
trations of dietary antioxidants reaching the plasma are
extremely low, it is most unlikely that an antioxidant effect
will arise from direct scavenging of ROS. We [18–20] and
others [21] have reported that certain antioxidants are
capable of undergoing fast electron transfer (as well as
H-atom donation) to DNA radicals bringing about partial
restoration of the DNA. This process is made possible for
low concentrations of these antioxidants to act by the fol-
lowing: (i) the DNA radicals have lifetimes of up to sec-
onds [22], and (ii) the active antioxidants, for example,
flavonoids, possess one-electron reduction potentials of
of DMSO, as previously described [18], taking the sec-
⋅
ond-order rate constant for the reaction of OH radicals
with DMSO (reaction 3) as 1.0ꢄ1 0 10
M
s
ꢃ1 [31].
ꢃ1
⋅
⋅
OHꢁA→A(OH)
OHꢁ(CH3)2SO→(CH3)2S O(OH)
(2)
(3)
⋅
⋅
Irradiation procedures
Solutions of supercoiled plasmid (8 ng μLꢃ1 , 2.82 nM in
base pairs) with and without the test compounds were irra-
diated under aerobic conditions at room temperature
(22ꢅ2 °C) using a 1.0ꢄ1 0 4 GBq 60Co γ-ray source at a
ꢁ
⋅
ꢁ
⋅
ꢃ1
their radical forms, E(A (A , H )/A), (i.e. E(1)R), lower
than those of a range of DNA radicals [18,23–25]. The in
vitro test system of choice is supercoiled plasmid DNA,
where gel electrophoresis combined with pre-incubation
of radical-damaged DNA with endonucleases facilitates
quantification of prompt strand breakage and base damage
[26,27]. In this study, we report the capacity of the major
water-soluble coffee constituents, caffeine and its metabo-
lites, chlorogenic acid, and the hydroxycinnamic acids
such as ferulic acid and caffeic acid, in effecting the fast
chemical repair of DNA radicals produced on ROS
attack.
dose rate of ca. 5 Gy min
determined using Fricke
dosimetry. Irradiated samples were serially withdrawn and
kept on ice until subsequent analysis. Under these condi-
tions, the radiation breaks down the water into quantified
ꢃ1
amounts of radicals (yields expressed in μM·Gy given
⋅
in parenthesis, reaction 4) of which the OH radical is the
⋅
main oxidising species with the reducing H atoms and
ꢃ
e
aq being scavenged by oxygen to form superoxide (reac-
tion 5), which, in the absence of bound metal ions, does
not damage DNA to any significant extent [32].
ꢃ
aq
⋅
⋅
H2O ^^^^^→ OH (0.28), H (0.06), e
(0.28)
ꢁ
ꢁH3O , H2O2, H2
(4)
(5)
ꢃ
ꢃ
⋅ ⋅ ⋅
aq/H ꢁO2 →O2 /(HO2 )
e
Materials and methods
The chemical effects of the test compounds (A) on the
produced DNA radicals were studied by maintaining a
All chemicals, sourced from Sigma-Aldrich, were of the
highest analytical grade available and used as received.
Plasmid DNA, pBR322, was prepared and purified as pre-
viously described [18]. Endonucleases were purchased
from New England Bio laboratories. Plasmid DNA Forms
I (supercoiled) and II (relaxed coiled) were separated on
1% agarose gels using electrophoresis in chilled, circu-
lated TAE buffer. Quantification of the ethidium bromide-
stained bands (with a 1.3 enhancement applied to Form I)
was done using the NIH Image software ImageJ.
⋅
constant OH radical scavenging capacity, k, (the sum of
the pseudo-first-order rate constants for the reactions of
⋅
the OH radicals with A and the TRIS buffer), as previ-
ously described [18]. The concentration of the TRIS buf-
fer (in reaction 7) was reduced proportionately as more A
was added to the test solution (in reaction 2) to maintain
kꢂ1.5ꢄ1 0 7 s based on the rate constant for TRIS,
ꢃ1
1.5ꢄ1 0 9 Mꢃ1 sꢃ1 [31]. This system produces a constant
⋅
initial amount of DNA damage formed upon OH radical
attack (reaction 6). By maintaining such a low k value and
Pulse radiolysis and steady-state radiolysis studies were
carried out in the University of Auckland facility [28]. The
E(1)R values of the test compounds (A) at pH 7 were
determined using the selenite radical as the oxidant in
DNA concentration, strand breakage (and base damage)
⋅
almost exclusively arises from reactions of the OH radi-
cals that are formed in close proximity to the DNA, (reac-
tion 6) [33] rather than from direct ionisation of the DNA
by the γ-rays. Hence, the explicit effects of A reacting with
the DNAꢁ. radicals (and their oxygen adducts) are studied
(reaction 8).
establishing redox equilibria for three mixtures of A and
ꢁ
⋅
reference compounds, R of known E(R /R) value [29],
and in calculating the ΔE from the equilibrium constant,
ꢁ
ꢁ
⋅
⋅
K,whereKꢂ[R]/[A]{(Abseq ꢃAbsR )/(AbsA ꢃAbseq)}.
ꢁ
⋅
ꢁ
⋅
⋅
OHꢁDNA→H2OꢁDNA → →
base damage/strand break
AbsR , AbsA and Abseq are the observed absorbances
(6)
(7)
(8)
normalised for radiation dose for the reference compound,
antioxidant and at equilibrium, respectively. As
⋅
⋅
OHꢁTRIS→H2OꢁTRIS
ꢁ
⋅
ΔEꢂꢃ(RT/F)ln K, the E(1)R values for E(A /A) are
ꢁ
ꢁ
⋅
⋅
obtained [30].
AꢁDNA ꢁH → A ꢁDNA

482 M. A. Rathod et al.
irradiated plasmid with each of three DNA glycosylases
(FPG, formidopyrimidine-DNA glycosylase; Endo III,
endonuclease III; and Exo III, exonulease III) as previ-
ously described [18,20]. While all three proteins recognise
sites of base loss with unmodified deoxyribose moiety and
deoxyribose oxidised in the 4ꢆ position, Exo III addition-
ally recognises deoxyribose oxidised in the 1ꢆ position,
FPG recognises 7,8-dihydro-8-oxoguanine and formami-
dopyrimidines, and Endo III recognises 5,6-dihydro-
pyrimidines [34]. The difference in the G values between
the control and enzyme-treated cases gives the G value
representing the increase in strand breakage arising from
DNA damage assays
The radiation doses required to produce one SSB, Do (in
Gy), were obtained from the slopes of the regression lines
for the loss of supercoiled DNA, as a percentage of the
total concentration of DNA, with increasing radiation dose
using Do ꢂ{(log1037)-2}/slope. The Do values were used
to calculate the G values for ssb production (in pM.Gyꢃ1 ),
where G(ssb)ꢂ[DNA] (nM in base pairs)/Do (in Gy) ꢄ
ρ (kg Lꢃ1 , assumed to be unity).
DNA base damage was evaluated as the increased
amount of strand breakage formed upon incubating
Figure 1. Caffeine, chlorogenic acid and their metabolic derivatives.

Reduction in radical-induced DNA breaks by coffee 483
the action of each protein. The differences in these G
values when each of the compounds are present at the time
of irradiation gives the compound-induced reduction in
strand breaks and base damage which is recognised by the
proteins. Subtracting this second calculated value from the
control values obtained in the presence of the enzymes
alone yields the G values for base damage (recognised by
the proteins) that is influenced by the compounds.
be added to solutions of the test compounds to maintain a
constant radical scavenging capacity. While k2 values are
known for caffeine and the hydroxycinnamic acids, this is
not the case for the metabolites of caffeine. We have deter-
mined these values by competition kinetics, as described
above, and these values are given in Table I.
DNA strand breakage in the plasmid following serial
irradiations in the presence and absence of caffeine and
a range of its metabolites, as well as the phenolic com-
pounds, was followed by gel electrophoresis. For exam-
ple, radiation dose–response curves obtained for the loss
of supercoiled form of the DNA with increasing radia-
tion dose is given in Figure 2 for caffeine (200 μM) and
ferulic acid (20 μM), both with and without post-irradi-
ation incubation with Endo III. Do values, determined
from such plots, together with the Do values for controls
in the absence of the compounds, were obtained to cal-
culate the yields of prompt strand breaks and specific
base damages. Small variations in controls between day-
to-day experiments were normalised to the average
response and used to adjust G values obtained in the
presence of added test compounds accordingly. Average
G values from duplicate experiments, for both prompt
and endonuclease-induced strand-breaks, are presented
in Table II.
Results
Pulse radiolysis studies were carried out to determine the
E(1)R values for caffeine and its metabolites which do not
have values in the literature. Caffeine is primarily dem-
ethylated by the hepatic cytochrome P450 enzyme
CYP1A2, to initially 1,7-dimethylxanthine (paraxanthine,
80–85%), 3,7-dimethylxanthine (theobromine, 12%) and
1,3-dimethylxanthine (7%) [35]. Further demethylation of
these intermediates yields 1-, 3- and 7-methylxanthine,
and other pathways produce 1,3- and 1,7-dimethyluric
acid, and both 1-methyluric acid and 7-methyluric acid
(Figure 1). Redox equilibria were established for caffeine
and five of its metabolites (Supplementary material, Table
I available online at http://informahealthcare.com/doi/
abs/10.3109/10715762.2013.788167) to obtain their
E(1)R values. These measured values, together with lit-
erature values for other metabolites of caffeine, are tabu-
lated in Table I. The E(1)R values decrease from 1.39 V
for caffeine to ca. 0.74 V for the mono-methyl-substituted
uric acids. Literature values for E(1)R values of the
hydroxycinnamic acids, caffeic acid, ferulic and chloro-
genic acids are also given in Table I. All three compounds
possess E(1)R values lower than those of the metabolites
of caffeine. Knowledge of the second-order rate constants
for scavenging of the .OH radicals, k2, is necessary for
calculating the required concentrations of TRIS buffer to
No reduction in strand breaks under the constant radical
scavenging conditions, compared to controls, was observed
in the presence of added caffeine, the di-methylxanthines
or the mono-methylxanthines (i.e. compounds with E(1)
Rꢇ1.05 V), even at high concentrations. Concentration-
dependent reductions in prompt strand breaks were
observed for the mono-methyluric acids (E(1)R ca. 0.74 V)
and the lower potential phenolic compounds. No signifi-
cant reduction was observed in the range of DNA-base
damages (except for purine damage in the presence of 100
μM chlorogenic acid), which were tested for by treatment
with the endonucleases (Table II).
⋅
Table I. One-electron reduction potentials, OH radical scavenging rate constants and human plasma levels.
ꢃ9
ꢃ1 ꢃ1
⋅
Compounds
E(1)R pH 7/V
10 k( OHꢁA)/M
s
Plasma level after caffeine admin./mM
Caffeine and metabolites
1,3,7-Trimethylxanthine, caffeine
1,3-Dimethylxanthine, theophylline
1,7-Dimethylxanthine, paraxanthine
3,7-Dimethylxanthine, theobromine
3-Methylxanthine
1.36a
1.29b
1.26b
1.38a
1.28b
1.27b
1.07b
0.74a
0.74a
0.74a
0.72a
7.3d, 6.8e, 8.5f
6.3e, 5.0f
9.2ꢅ0.1a
5.8e, 5.5f
4.0f
19i, 30.7j, 23.0k, 35.6l
78i, 7.6k, 5.3l
6.6i, 10.9k, 15.6l, 6.57m
3.9i, 4.16j
4.8i
1.6i
0.4g
0.6i
6.7i
r
8.7ꢅ0.3a
8.9ꢅ0.4a
8.2ꢅ0.2a
8.6ꢅ0.2a
9.9ꢅ0.1a
4.1ꢅ0.2a
7-Methylxanthine
1-Methylxanthine
1,7-Dimethyluric acid
1,3-Dimethyluric acid
7-methyluric acid
1-Methyluric acid
r
ꢃ9
ꢃ1 ꢃ1
⋅
Phenolics
E(1)R pH 7/V
10 k( OHꢁA)/M
s
Plasma level after coffee intake/mM
Ferulic acid
0.60c
0.53c
0.55c
7.6g
7.4g
9.0h
0.139n, 0.8o,
0.081n, 1.1o, 0.98p
14.8o
Caffeic acid
Chlorogenic acid
athis work, see Supplementary Table I available online at http://informahealthcare.com/doi/abs/10.3109/10715762.2013.788167, b[38], c[39] d[9], e[49], f[50]
g[41], h[40], i[42], j[43], k[35], l[44], m[45], n[46], o[47], p[48], rundetermined, due to fast excretion.

484 M. A. Rathod et al.
100
100
10
1
10
1
0
100
200
300
400
500
600
0
100
200
300
400
500
600
Dose /Gy
Dose /Gy
Figure 2. Effect of caffeine (200 μM), LHS and ferulic acid (20 μM), RHS, on the radiation dose-dependent loss of super-coiled plasmid
DNA (8 ng/μL) upon γ-irradiation of aerobic solutions and subsequent incubation with and without endonuclease III (Endo III). Tris buffer
controls, ○; Tris buffer ꢁ compounds, ●; Tris buffer ꢁ post-irradiation incubation with Endo III, □; Tris bufferꢁcompoundsꢁpost-
irradiation incubation EndoIII, ▪. Data points are the average of two separate experiments with error bars representing the standard deviation.
Solid lines fitted by regression analysis.
Table II. Yields of DNA strand breaks.
fpg activity
Endo-activity
Exo-activity
ꢃ1
ꢃ1
ꢃ1
ꢃ1
Compounds
Test conc./mM SSB/pM Gy
(Purine)/pM Gy
(Pyrim.)/pM Gy
(AP sites)/pM Gy
ꢃ1
Control breaks
/pM Gy
8.0ꢅ0.2
21ꢅ 2
21ꢅ4
1 2 ꢅ 2
8 ꢅ 2
7ꢅ2
Caffeine and metabolites
1,3,7-Trimethylxanthine, caffeine
50
100
200
400
200
400
100
200
400
100
200
100
100
50
7.8ꢅ0 .5
7.4ꢅ0 .2
7.2ꢅ0 .7
6.9ꢅ1 .0
7.8ꢅ0 .2
8.0ꢅ0 .2
8.0ꢅ0 .1
7.8ꢅ0 .1
7.4ꢅ0 .2
8.2ꢅ0 .2
8.0ꢅ0 .2
8.0ꢅ0 .2
8.2ꢅ0 .2
8.0ꢅ0 .1
8.1ꢅ0 .2
7.8ꢅ0 .4
8.1ꢅ0 .1
8.0ꢅ0 .1
7.8ꢅ0 .2
7.2ꢅ1 .1
5.4ꢅ0 .6
5.3ꢅ0 .4
6.0ꢅ0 .8
6.1ꢅ0 .8
4.6ꢅ0 .7
6.5ꢅ0 .6
5.1ꢅ0 .2
4.5ꢅ0 .6
11 ꢅ2
10ꢅ3
1,3-Dimethylxanthine, theophylline
1,7-Dimethylxanthine, paraxanthine
3,7-dimethylxanthine, theobromine
3-Methylxanthine
7-Methylxanthine
1-Methylxanthine
17ꢅ2
21ꢅ2
23ꢅ2
21ꢅ1
20ꢅ2
1 1 ꢅ1
1 2 ꢅ2
1 0 ꢅ1
8 ꢅ1
7 ꢅ2
8 ꢅ2
8 ꢅ2
5 ꢅ1
5 ꢅ2
100
200
50
1,7-Dimethyluric acid
1,3-Dimethyluric acid
7-Methyluric acid
100
200
50
100
200
50
100
200
50
1-Methyluric acid
9 ꢅ2
100
200
Phenolics
Ferulic acid
20
100
20
6.3ꢅ0 .6
6.4ꢅ0 .1
6.6ꢅ0 .2
6.0ꢅ0 .5
6.6ꢅ0 .2
5.2ꢅ0 .4
4.6ꢅ0 .2
20ꢅ2
21ꢅ2
16ꢅ1
8 ꢅ1
10ꢅ0 .1
1 0 ꢅ1
6ꢅ1
7 ꢅ1
6ꢅ1
Caffeic acid
100
20
Chlorogenic acid
1 1 ꢅ2
13ꢅ2
100
200

Reduction in radical-induced DNA breaks by coffee 485
Discussion
base damage precursor radicals, even when tested at
the high concentrations reached under physiological
conditions [19].
The results from this in vitro study show that water-soluble
constituents of coffee of radical one-electron reduction
potential ꢀ1 V are active in reducing the amount of rad-
ical-induced DNA strand breaks, most likely by an anti-
oxidant mechanism between such active compounds and
DNA radicals. Prompt strand breaks are associated with
both the direct radical formation on the deoxyribose moi-
ety of DNA and the intramolecular radical transfer from
base radicals to the ribose [36]. Under aerobic irradiation
conditions, such carbon-centred radicals rapidly add oxy-
gen to form peroxyl radicals as the precursors to strand
breaks. The E(1)R of peroxyl radicals are in the region of
1 V [37], so it is understandable, in thermodynamic terms,
for a compound to possess an E(1)R value less than this
to act as an antioxidant in undergoing a chemical reaction
with a peroxyl radical to form the hydroperoxide. Forma-
tion of the hydroperoxide can be through either electron
transfer followed by rapid protonation or H-atom abstrac-
tion from the antioxidant. Relating such a consideration of
redox potentials to caffeine, its metabolites and the phe-
nolics of coffee, it is seen that for compounds with E(1)R
values ꢈ1 V, they indeed do not function as antioxidants
in this way, whereas compounds with E(1)R values ꢀ1 V
do. Clearly, thermodynamic considerations alone are not
sufficient to predict whether a compound can act as an
antioxidant towards DNA radicals or not. The DNA base
radicals possess E(1)R values in the region of 1.2–1.3 V
(purines) and 1.6–1.7 V (pyrimidines) [24], and while caf-
feine and its methylxanthine metabolites have relatively
high E(1)R values, the methyluric acids and the phenolics
have much lower E(1)R values but are inactive in effecting
the chemical repair of base radicals. A possible reason for
this is that the methyluric acids and the phenolic acids all
carry negative charges at pH 7, most likely resulting in
them being repelled by the negatively charged ribose-
phosphate backbone of DNA inhibiting close association
with the inner core of DNA bases in double-stranded
DNA. Some activity was observed for chlorogenic acid in
ameliorating damage to purine bases, where the negative
charge is positioned on the quinic acid, remote from the
poly-hydroxyphenol moiety.
It can be concluded from this study that the di-methy-
luric acids, mono-methyluric acids, hydroxycinnamic
acids and chlorogenic acid are all active in reducing the
amount of DNA strand breaks arising from free radical
attack. However, only chlorogenic acid reaches compara-
ble test levels in human plasma (Table I). It is possible
that intracellular levels of the other active compounds are
higher than those measured in plasma, due to the uptake
and metabolism of caffeine and chlorogenic acid inside
cells. Nevertheless, of all the constituents of coffee and
the metabolites of caffeine tested in this study, the anti-
oxidant which is likely to be the most active compound in
undergoing the fast chemical repair of DNA radicals with
possible activity in reducing the incidence of certain can-
cers and neurological disease is chlorogenic acid. How-
ever, the antioxidant mechanism investigated in this study
is but one of several mechanisms of how antioxidants may
ameliorate radical damage sustained to cellular structures.
The relative importance of these mechanisms to human
health is the subject of much research.
Declaration of interest
The authors report no declarations of interest. The authors
alone are responsible for the content and writing of the
paper.
This work was supported by the Faculty of Medical and
Health Sciences, University of Auckland grant 3626660
(to R.F.A.)
References
[1] Bravi F, Bosetti C, Tavani A, Bagnardi V, Gallus S, Negri E,
et al. Coffee drinking and hepatocellular carcinoma risk: a
meta-analysis. Hepatol 2007;46:430–435.
[2] Corrao G, Zambon A, Bagnardi V, D’Aicis A, Klatsky A;
Collaborative SIDECIR Group. Coffee, caffeine, and the risk
of liver cirrhosis. Ann Epidemiol 2001;11:458–465.
[3] Tverdal A, Skurtveit S. Coffee intake and mortality from liver
cirrhosis. Ann Epidemiol 2003;12:419–423.
The antioxidant activities of the compounds in the pres-
ent study can be compared with the those of the catechin
compounds found in green tea in the previous study [20].
The catechins possess E(1)R values similar to those of the
phenolics of coffee. While the catechins effect similar pro-
tection to the coffee phenolics against prompt DNA strand
breaks, they are also effective in reducing a wide range of
DNA base damages, as assayed for by FPG, EndoIII and
Exo III endonucleases. The catechins are uncharged com-
pounds at physiological pH, and this property may well
be important to enable localisation in the vicinity of the
DNA bases in comparison with the negatively charged
phenolic acids of coffee, as suggested above. Some other
antioxidants carrying negative charges, such as ascorbate
and glutathione, do not undergo a fast chemical repair of
DNA radicals in reducing either prompt strand breaks or
[4] Hernán MA, Takkouche B, Caamaño-Isorna F, Gestal-Otero
JJ. A meta-analysis of coffee drinking, cigarette smoking, and
the risk of Parkinson’s disease. Ann Neurol 2002;52:276–284.
[5] Costa J, Lunet N, Santos C, Santos J, Vaz-Carneiro A. Caffeine
exposure and the risk of Parkinson’s disease: a systematic
review and meta-analysis of observational studies. J Alzhe-
imer’s Disease 2010;20:S221–S238.
[6] Michels KB, Willett WC, Fuchs CS, Giovannucci E. Coffee,
tea, and caffeine consumption and incidence of colon and rec-
tal cancer. J Natl Cancer Inst 2005;97:282–292.
[7] José S, Vicente V, Ishimoto EY, Cruz RJ, Dias C, Pereira S,
Da Silva Torres EAF. Increase of the activity of phase II anti-
oxidant enzymes in rats after a single dose of coffee. J Agric
Food Chem 2011;59:10887–10892.
[8] Grubben MJAL, van den Braak CCM, Broekhuizen R, de Jong
R, Van Rijt L, de Ruijter E, et al. The effect of unfiltered cof-
fee on potential biomarkers for colonic cancer risk in healthy
volunteers: a randomised trial. Aliment Pharmacol Ther
2000;14:1181–1190.

486 M. A. Rathod et al.
[9] Devasagayam TPA, Kamat JP, Mohan H, Kesavan PC. Caf-
feine as an antioxidant: inhibition of lipid peroxidation induced
by reactive oxygen species. Biochimica et Biophysica Acta
1996;1282:63–70.
higher cells: techniques and preliminary results. Radiat Res
1985;101:356–372.
[28] Anderson RF, Denny WA, Li W, Packer JE, Tercel M,
Wilson WR. Pulse radiolysis studies on the fragmentation of
arylmethyl quaternary nitrogen mustards by one-electron reduc-
tion in aqueous solution. J Phys Chem A 1997;101:9704–9709.
[29] Wardman P. Reduction potentials of one-electron couples
involving free radicals in aqueous solution. J Phys Chem Ref
Data 1989;18:1637–1755.
[10] Lee C. Antioxidant ability of caffeine and its metabolites
based on the study of oxygen radical absorbing capacity and
inhibition of LDL peroxidation. Clin Chem Acta 2000;295:
141–154.
[11] Ames BN, Shigenaga MK, Hagen TM: Oxidants, antioxidants,
and the degenerative diseases of aging. Proc Natl Acad Sci
USA 1993;90:7915–7922.
[30] Anderson RF, Shinde SS, Hay MP, Gamage SA, Denny WA.
Radical properties governing the hypoxia-selective cytotoxic-
ity of antitumor 3-amino-1,2,4-benzotriazine 1,4-dioxides.
Org Biomol Chem 2005;3:2167–2174.
[12] Zhang J, Perry G, Smith MA, Robertson D, Olson SJ,
Graham DG, Montine TJ. Parkinson’s disease is associated
with oxidative damage to cytoplasmic DNA and RNA in sub-
stantia nigra neurons. Am J Pathology 1999;154:1423–1429.
[13] Ozawa T, Hayakawa M, Katsumata K, Yoneda M, Ikebe S,
Mizuno Y. Fragile mitochondrial DNA: The missing link in
the apoptotic neuronal cell death in parkinson’s disease. Bio-
chem Biophys Res Commun 1997;235:158–161.
[31] Buxton GV, Greenstock CL, Helman WP, Ross AB. Critical
review of rate constants of hydrated electrons, hydrogen atoms
ꢃ
⋅
⋅
and hydroxyl radicals ( OH/ O ) in aqueous solution. J Phys
Chem Ref Data 1988;17:513–886.
[32] van Hemmen JJ, Meuling WJA. Inactivation of biologically
active DNA by ꢉ-ray-induced superoxide radicals and their
dismutation products singlet oxygen and hydrogen peroxide.
Biochim Biophys Acta 1975;402:133–141.
[14] Iwai K, Kishimoto N, Kaakino Y, Mochida K, Fujita T. In vitro
antioxidative effects and tyrosinase inhibitory activities of
seven hydroxycinnamoyl derivatives in green coffee beans.
J Agric Food Chem 2004;52:4893–4898.
[33] Hodgkins PS, Fairman MP, O’Neill P. Rejoining of gamma-
radiation-induced single-strand breaks in plasmid DNA by
human cell extracts: dependence on the concentration of the
hydroxyl radical scavenger, tris. Radiat Res 1996;145:24–30.
[34] Wallace SS. Enzymatic processing of radiation-induced free
radical damage in DNA. Radiat Res 1998;150:S60–S69.
[35] Lelo A, Miners JO, Roson R, Birkett DJ. Assessment of caf-
feine exposure: Caffeine content of beverages, caffeine intake,
and plasma concentrations of methylxanthines. Clin Pharma-
col Ther 1986;39:54–59.
[15] Cavin C, Mace K, Offord EA, Schilter B. Protective effects of
coffee diterpenes against aflatoxin B1-induced genotoxicity:
mechanisms in rat and human cells. Food Chem Toxicol
2001;39:549–556.
[16] Huber WW, Scharf G, Rossmanith W, Prustomersky S,
GraslKraupp B, Peter B, et al. The coffee components, kah-
weol and cafestol induce ꢉ-gluamylcysteine synthetase, the rate
limiting enzyme of chemoprotective glutathione synthetase, in
several organs of the rat. Arch Toxicol 2002;75:685–694.
[17] Lee KJ, Jeong HG. Protective effects of kahweol and cafestol
against hydrogen peroxide-induced oxidative stress and DNA
damage. Toxicol Lett 2007;173:80–87.
[36] von Sonntag C. Free-Radical-Induced DNA Damage and Its
Repair. Berlin Heidelberg: Springer-Verlag; 2006.
[37] Buettner GR. The pecking order of free radicals and antioxi-
dants: lipid peroxidation, ꢊ-tocopherol, and ascorbate. Arch
Biochem Biophys 1993;300:535–543.
[18] Anderson RF, Amarasinghe C, Fisher LJ, Mak WB,
Packer JE. Reduction in free-radical-induced DNA strand
breaks and base damage through fast chemical repair by
flavonoids. Free Radic Res 2000;33:91–103.
[38] Stockert AL, Shinde SS, Anderson RF, Hille R. The reaction
mechanism of xanthine oxidase: Evidence for two-electron
chemistry rather than sequential one-electron steps. J Am
Chem Soc 2002;124:14554–14555.
[19] Anderson RF, Fisher LJ, Harris T. The pivotal antioxidant
hypothesis for the role of flavonoids in their reduction of HO.
radical-induced damage on DNA. Redox Rep 2001;6:
197–199.
[39] Foley S, Navaratnam S, McGarvey DJ, Land EJ, Truscott TG,
Rice-Evans CA. Singlet oxygen quenching and the redox prop-
erties of hydroxycinnamic acids. Free Radic Biol Med
1999;26:1202–1208.
[20] Anderson RF, Fisher LJ, Hara Y, Harris T, Mak WB,
Melton LD, Packer JE. Green tea catechins partially protect
DNA from .OH radical-induced strand breaks and base dam-
age through fast chemical repair of DNA radicals. Carcinogen-
esis 2001;22:1189–1193.
[40] Mahal HS, Mukherjee T. Radical scavenging reactions of
chlorogenic acid: a pulse radiolysis study. Res Chem Intermed
2006;32:671–682.
[41] Jiang Y, Lin WZ, Yao SD, Lin NY, Zhu DY. Fast repair of
oxidizing OH adducts of DNA by hydroxycinnamic acid
derivatives. A pulse radiolytic study. Radiat Phys Chem
1999;54:355–359.
[21] Zheng R, Shi Y, Jia Z, Zhao C, Zhang Q, Tan X. Fast repair
of DNA radicals. Chem Soc Rev 2010;39:2827–2834.
[22] Schulte-Frohlinde D, Behrens G, Onal A. Lifetime of peroxyl
radicals of poly(U), poly(A) and single-and double-stranded
DNA and the rate of their reaction with thiols. Int J Radiat
Biol 1986;50:103–110.
[42] Leakey TE. Simultaneous analysis of theophylline, caffeine
and eight of their metabolic products in human plasma by
gradient high-performance liquid chromatography. J Chroma-
tog 1990;507:199–220.
[23] Jovanovic SV, Steenken S, Tosic M, Marjanovic B, Simic MG.
Flavonoids as antioxidants. J Am Chem Soc 1994;116:
4846–4851.
[43] Soto J, Sacristan JA, Alsar MJ. Cerebrospinal fluid concentra-
tions of caffeine following oral drug administration: correla-
tion with salivary and plasma concentrations. Ther Drug
Monit 1994;16:108–110.
[24] Steenken S, Jovanovic SV. How easily oxidizable is DNA?
One-electron reduction potentials of adenosine and guanosine
radicals in aqueous solution. J Am Chem Soc 1997;119:
617–618.
[44] Muir KT, Kunitani M, Riegelman S. Improved high-performance
liquid chromatography assay for theophylline in plasma and
saliva in the presence of caffeine and its metabolites and com-
parisons with three other assays. J Chromatog 1982;231:73–82.
[45] Graham TE, Spriet LL. Metabolic, catecholamine, and exer-
cise performance responses to various doses of caffeine.
J Appl Physiol 1995;78:867–874.
[25] Shinde SS, Maroz A, Hay MP, Anderson RF. One-electron
reduction potential of the neutral guanyl radical in the GC
base pair of duplex DNA. J Am Chem Soc 2009;131:
5203–5207.
[26] Epe B, Pflaum M, Haring M, Hegler J, Rudiger H. Use of
repair endonucleases to characterize DNA damage induced by
reactive oxygen species in cellular and cell-free systems.
Toxicology Lett 1993;67:57–72.
[46] Renouf M, Guy PA, Marmet C, Fraering A-L, Longet K,
Moulin J, et al. Measurement of caffeic and ferulic acid
equivalents in plasma after coffee consumption: small intes-
tine and colon are key sites for coffee metabolism. Mol Nutr
Food Res 2010;54:760–766.
[27] Krisch RE, Tan PM, Flick MB. A sensitive SV40 viral probe
assay for DNA strand breaks and their biological repair in

Reduction in radical-induced DNA breaks by coffee 487
[47] Farah A, Monteiro M, Donangelo CM, Lafay S. Chlorogenic
acids from green coffee extract are highly bioavailable in
humans. J Nutr 2008, 138:2309–2315.
[49] Powers EL. Purine and its analogues and radiation damage
in Bacillus megaterum spores. Int J Radiat Biol 1986;50:
983–993.
[48] Monteiro M, Farah A, Perrone D, Trugo LC, Donangelo C.
Chlorogenic acid compounds from coffee are differentially
absorbed and metabolized in humans. J Nutr 2007;137:
2196–2201.
[50] Vinchurkar MS, Rao BSM, Mohan H, Mittal JP. Kinetics,
spectral and redox behaviour of OH adducts of methylxan-
thines: a radiation chemical study. J Chem Soc, Perkin Trans
1999;609–617.
Supplementary material available online
Supplementary Table I.