Antioxidant properties of 4-quinolones and structurally related flavones

Article abstract of DOI:10.1016/j.bmc.2011.11.068

Neurodegenerative disorders are frequently associated with increased oxidative damage to the brain as a result of free radicals produced by cellular respiration. The onset and progression of neurodegeneration may therefore be curbed by exogenous hydrogen-

Full text of DOI:10.1016/j.bmc.2011.11.068

Bioorganic & Medicinal Chemistry 20 (2012) 809–818
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Antioxidant properties of 4-quinolones and structurally related flavones
Jane Greeff a, Jacques Joubert b, Sarel F. Malan a,b, Sandra van Dyk a,
⇑
a Department of Pharmaceutical Chemistry, School of Pharmacy, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa
b School of Pharmacy, University of the Western Cape, Private Bag X17, Bellville 7535, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
Neurodegenerative disorders are frequently associated with increased oxidative damage to the brain as a
result of free radicals produced by cellular respiration. The onset and progression of neurodegeneration
may therefore be curbed by exogenous hydrogen-donating antioxidant moieties such as the naturally
occurring flavonoids. A series of 2-phenylquinolin-4(1H)-ones was synthesised and displayed moderate
to high antioxidant activity when compared to structurally related flavones and quinolines. Activity of
the hydroxy-2-phenylquinolin-4(1H)-ones (8–10) was established in reducing ferrous ions and diminish-
ing hydrogen peroxide and hydroxyl radical production, in the FRAP (1.41–97.71% TroloxÒ equivalents),
Received 23 September 2011
Revised 25 November 2011
Accepted 30 November 2011
Available online 8 December 2011
Keywords:
Quinolines
4-Quinolones
Flavones
Antioxidant
Neuroprotection
ORAC (9.18–15.27 l
M TroloxÒ equivalents at 0.001 mM) and TBARS (0.05–0.72 nmol MDA/mg tissue)
assays, respectively. The results indicated that the additional hydrogen donating groups on the
synthesised 2-phenylquinolin-4(1H)-one series increased antioxidant activity.
Ó 2011 Elsevier Ltd. All rights reserved.
Fe2þ ! H2O2 ! Fe3þ þ HOꢀ þ OHꢁ
ð2Þ
ð3Þ
1. Introduction
2ꢁ þ H2O2 ! O2 þ OHꢁ þ HOꢀ
Oxidative stress is a common occurrence in neurodegenerative
disorders, such as Alzheimer’s1 and Parkinson’s disease2 and takes
place prior to the onset of neurodegeneration, leading to, or exac-
erbating deterioration.3–5 The cellular oxidative status is regulated
by antioxidant enzymes responsible for neutralising free radicals.
With increasing age the enzymes are overwhelmed by the amount
of radicals requiring deactivation,6 leading to decreased protection
by the body’s antioxidant systems. In addition, the mitochondria
produce more reactive oxygen species at the cost of producing less
ATP.7–9 These natural by-products of cellular respiration act to
injure the mitochondria and cell structures containing lipids, pro-
teins and DNA,8,10 and serve to decrease the lifespan of the cell.11,12
Cellular respiration is at the origin of this cascade of events, yield-
ing, amongst others, destructive reactive oxygen species as by-
products of energy production by the mitochondria.7 Molecular
oxygen (O2) is reduced to water (Eq. 1) during respiration and
yields superoxide anions, hydrogen peroxide and hydroxyl radicals
as a result.13
O
With increased age the expression of antioxidant enzymes de-
crease6,14 and the ability to maintain the antioxidant/pro-oxidant
equilibrium diminishes, causing decreased antioxidant activity
and increased neuronal damage. The brain contains high concen-
trations of oxidisable substrate and catalysts that enable oxidation
as well as low concentrations of antioxidant enzymes.10 Therefore
free radical damage and apoptosis8 in the brain can contribute to
irreversible neurodegeneration. This deterioration is only diag-
nosed when damage to the brain is sufficient to induce disability
and irreversible neuronal damage.
Enhancing safe metabolism of oxygen to water by introducing
exogenous antioxidants could therefore prolong neuronal lifespan
and antioxidants may be useful in decreasing brain damage
caused by free radicals, curbing the progression of neurodegener-
ative disease and prolonging the lifespan and quality of life of the
patient.
Antioxidants may act according to one of two mechanisms: pre-
vention of initiation of oxidation, or as chain breaking antioxidants.
Prevention of initiation of oxidation occurs by inhibiting superoxide
anion production, degrading hydrogen peroxide and chelating or
reducing metal ions, while chain breaking antioxidants act by
scavenging radicals, mostly hydroxyl radicals, thereby inhibiting
the chain of oxidative events that leads to damage of lipid
membranes, proteins and DNA.15 In order to scavenge radicals the
antioxidant must act as a hydrogendonor able to reduce the radical
and thereby quench it. The antioxidant should form a stable radical
in order not to oxidise biological material itself.
O2 ! HO2 ! H2O2 ! HOꢀ þ H2O
ð1Þ
ꢀ
The most detrimental of the reactive oxygen species is the
hydroxyl radical,11 generated in the Fenton reaction (Eq. 2), when
hydrogen peroxide is catalytically oxidised by ferrous iron
(Fe2+).1 The ferric ions (Fe3+) produced in the Fenton reaction fur-
ther act as catalyst in the Haber–Weiss reaction (Eq. 3), producing
more hydroxyl radicals that may induce lipid peroxidation.1
⇑
Corresponding author. Tel.: +27 18 299 2267; fax: +27 18 299 4243.
E-mail address: Sandra.vandyk@nwu.ac.za (S. van Dyk).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.11.068

810
J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
Polyphenols are chain-breaking antioxidants16 that function by
R1
N
scavenging radicals. The phenol moiety17 of hydroxyl substituted
aromatic compounds (ArOH), act as electron donors driving radical
reduction, while the aromatic ring acts to stabilise the phenolic
oxygen radical produced (ArOÅ), thereby increasing reducing ability
of the antioxidant.18 Flavonoids (Fig. 1) are a well-known class of
natural compounds that possess radical scavenging properties19,20
due to the ability to form stable radicals.21 Qin et al. indicated that
flavonoids with three hydroxyl substitutions presented increased
antioxidant activity,22 while Khlebnikov et al. indicated that
hydroxyl groups on either positions 3, 7 or 8 enhanced scavenging
activity.23 It was therefore hypothesised that structurally similar
synthetic compounds containing extra hydrogen donating func-
tionalities, such as an amine in position 1 of the quinolone
structure, might improve the antioxidant activity compared to
flavonoids.
R2
Quinoline
(1) R1, R2 = H
(2) R1 = OH; R2 = H
(3) R1 = H; R2 = OH
O
O
O
R6
R7
R4
R5
N
H
Fluoroquinolones (Fig. 1) are antimicrobial drugs that are able
to infiltrate the cerebrospinal fluid and are used to treat infections
such as meningitis. Park et al. indicated that ciprofloxacin reduced
ischemia and increased cell survival in a rat focal cerebral ischemic
animal model and that a modified structure of ciprofloxacin im-
proved neuroprotection while decreasing antimicrobial activity.24
Naturally occurring 4-quinolones are structurally similar to the
flavonoids and have been found to show antioxidant activ-
ity.252-Phenylquinolin-4(1H)-one might therefore possess similar
or improved antioxidant activity to flavones when the structure–
activity relationships of Park et al. are applied.24 The aim of the
study was therefore to investigate the in vitro antioxidant proper-
ties of a series of compounds including quinolines, flavones and
2-phenylquinolin-4(1H)-ones structurally related to the selected
flavones (Fig. 2). Prior to these studies the oral bioavailability,
blood–brain barrier permeability and toxicity profiles of the pro-
posed series we re-evaluated in silico, utilising predictive compu-
tational software to determine the feasibility of these compounds
as neuroprotective agents.
Since polyphenols are reported to act via the hydrogen donor
mechanism of action,26 the ability of the selected compounds to
scavenge the peroxyl radical in the oxygen radical absorbance
capacity (ORAC) assay and the hydroxyl radical in the lipid perox-
idation thiobarbituric acid reactive substances (TBARS) assay, was
experimentally evaluated. The ferric reducing antioxidant power
(FRAP) of the test series was also assessed.
Quinoline structurally resembles the core of the 2-phenylquin-
olin-4(1H)-one structure and derivatives thereof were included in
the test series (Fig. 2) to establish the activity of the basic structure,
while the flavones were included as reference antioxidants and to
establish the effect of the oxygen and the protonated amine groups
on hydrogen donating ability. As 8-hydroxyflavone was not
commercially available, 8-hydroxyquinoline (3) was included in
the series to compare to the structurally related8-hydroxy-
2-phenylquinolin-4(1H)-one (10). These chemical and biological
evaluations provided a wide range of data enabling determination
of the key antioxidative reactions involved.
R8
Flavone
(4) R4, R5 = H
2-Phenylquinolin-4(1H)-one
(7) R6, R7, R8 = H
(5) R4 = OH; R5 = H
(6) R4 = H; R5 = OH
(8) R6 = OH; R7, R8 = H
(9) R6, R8 = H; R7 = OH
(10) R6, R7 = H; R8 = OH
Figure 2. The quinoline, flavone and synthesised 2-phenylquinolin-4(1H)-one test
series.
1.1. In silico bioavailability studies
The test compounds were evaluated for their oral bioavailabil-
ity, physicochemical properties and theoretic ability to cross
the blood–brain barrier byutilising Osiris Property Explorer,
Chemsilico Property Predictor for Drug Discovery, Accelrys DS
ViewerPro 5.0 and Advanced Chemistry Development Chemsketch
computational software to elaborate on the potential of the deriv-
atives as potential neuroprotective drug candidates (Table 1).
The test serieswas found to comply with the oral bioavailability
standards as set by Lipinski’s rule of five27,28 since the cLogP
values, the amount of H-bond donors and rotatable bonds were
less than 5, the molecular weights were less than 500 g/mol and
H-bond acceptors less than 10. This indicated a strong probability
that the compounds would display favourable oral bioavailability.
Table 1 collates several computed physicochemical properties
for prediction of brain permeation. An examination of the
calculated properties indicated that effective blood–brain barrier
permeability is to be expected for all test compounds as their
cLogP values are between 2 and 5,29 the sum of nitrogen and oxy-
gen (N+O) atoms are less than 530and according to the ‘Norinder
rule’, cLogP minus (N+O)-values are greater than 0, except for
8-hydroxyquinoline (3).30 The molecular weights are below
450 g/mol31and pKa values were found to be between 4 and
10.32–34 The in silico predicted LogBB values were greater than
0,35,36with several compounds (1, 4, 5, 6, 8, 9 and 10) displaying
values near the optimal level for effective permeation, at 0.3.42
Taking the above physicochemical properties into consideration,
we predict that the test compounds would allow for effective oral
absorption, distribution and permeation of the brain.
O
O
O
5
8
5
8
4
4
F
6
3
Predictive toxicities were studied using Osiris Property Explorer
software and indicated that the hydroxyflavones (5, 6) and 2-phe-
nylquinolin-4(1H)-ones (7, 8, 9, 10) would be free of mutagenicity,
tumorigenicity, reproductive side effects and irritation, while the
quinolines (1, 2, 3) and flavone 4 had some potential toxicities
(Table 1). These findings suggest that the toxicity decreased with
the addition of the 2-phenyl moiety, substitution of the oxo in
position 1 of the flavone with the secondary amine of the 2-phe-
nylquinolin-4(1H)-one, and the addition of hydroxyl groups, with
3
6
7
OH
7
2
2
O
N
N1
1
HN
Ciprofloxacin
Figure 1. Basic flavonoid structure21 and a fluoroquinolone antibiotic.24
Flavonoid

J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
811
Table 1
In silico physicochemical properties for oral bioavailability and blood–brain barrier permeability of the test compounds as evaluated utilising computational predictive software
a
Test
compound
cLogPa,d H-bond Molecular H-bond
donorsb weighta acceptorsb bondsb
Rotatable pKa
LogBBa N+O
count P- (N+O) likelinessc scorec
cLog
Drug
Drug
Toxicityc
1
2
3
4
5
6
7
8
9
10
2.08
2.45
1.87
3.56
3.72
3.32
4.70
4.04
4.35
4.42
0
1
1
0
1
1
1
2
2
2
129
145
145
222
238
238
221
237
237
237
1
2
2
2
3
3
1
2
2
2
0
0
0
1
1
1
1
1
1
1
4.67
5.28
5.24
7.59
8.15
8.14
0.29
0.08
0.09
0.48
0.27
0.27
1
2
2
2
3
3
2
3
3
3
1.08
0.45
ꢁ0.13
1.56
0.72
0.32
2.70
1.04
1.35
1.42
ꢁ1.62
ꢁ1.87
1.55
1.85
1.35
ꢁ0.6
2.52
1.88
2.20
2.43
0.12
0.26
0.12
0.44
0.74
0.56
0.78
0.79
0.80
0.81
Mutagenic tumorigenic irritant
Mutagenic tumorigenic
Mutagenic tumorigenic irritant
Mutagenic
None
None
None
None
None
7.54* 0.49
8.30* 0.26
8.31* 0.26
8.76* 0.32
None
Values obtained using the following computational software:
a
Chemsilico Property Predictor for Drug Discovery.37
Accelrys DS ViewerPro 5.0.38
b
c
Osiris Property Explorer.39
Advanced Chemistry Development Chemsketch.40,41
d
pKa Value of the basic nitrogen (NH+).
*
the following sequence of decreasing toxicity: quinolines >
flavones > 2-phenylquinolin-4(1H)-ones (Table 1). The 2-phenyl-
quinolin-4(1H)-oneswere also indicated to be the most likely of
the test compounds to illustrate drug activity when compared to
a comprehensive database of known drugs, according to their drug
score and drug likeliness.39
This data has significance in that it indicate that the 2-phenyl-
quinolin-4(1H)-ones should have improved biological activity with
regards to interaction on enzyme systems, penetration through cell
membranes and the blood–brain barrier as well as favourable
properties during drug metabolism.43
contributors to lipid peroxidation. The assay was performed
according to Ou et al. using fluorescein to measure fluorescent
decay caused by the presence of peroxyl radicals, induced by
AAPH (2,20-azobi(2-amidinopropane) dihydrochloride).26Inhibition
of fluorescent decay therefore indicated the ability of the test com-
pounds to scavenge peroxyl radicals, protecting against fluorescent
decay of fluorescein.
The underlying reaction in the ORAC assay is a hydrogen transfer
reaction,26 causing a phenolic antioxidant to donate a hydrogen
atom,46 terminating the lipid peroxidation chain reaction.16
A
TroloxÒ (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid)
standard curve was generated, allowing for the expression of values
as micromoles TroloxÒ equivalents utilising the formulae described
by Cao et al.47
1.2. Chemistry
The general synthetic route used to synthesise the 2-phenyl-
quinolin-4(1H)-ones (7–10) was the traditional one pot
Conrad–Limpach synthesis described by Somanathan and Smith,
where a b-keto-ester was cyclisised with an aniline derivative
using para-toluene sulfonic acid as a catalyst (Scheme 1).44 Dehy-
dration under Dean–Stark conditions for an extended period of
1.3.2. FRAP assay
The chemical ability of the test compounds to reduce ferric
iron was assessed in the ferric reducing/antioxidant (FRAP) as-
say to indicate a measure of inhibition of hydroxyl radical pro-
duction through the Haber–Weiss reaction (Eq. 3, thereby
reducing oxidative stress. The assay determined the ability of
19 h, using benzene as
a solvent, maintained conversion of
reactants to the intermediate45 as the reaction was halted in the
presence of the by-product ethanol. Cyclisation of the generated
intermediate to the product was accomplished by addition of
diphenyl ether after removal of benzene, to enable reflux at a
temperature of 265 °C for 30 min. Identification of synthesised
compounds where performed using NMR, MS and IR.
the test compounds to reduce ferric-tripyridyltriazine (Fe3+
-
TPTZ) to its blue, ferrous form (Fe2+-TPTZ). This reaction is a
one electron exchange transfer reaction leading to a second
electron transfer and producing
a stable antioxidant radical
(ArOÅ).16
The method described by Benzie and Strain, was used with the
addition of a TroloxÒ standard to enable expression of data as
percentage of 0.1 mM TroloxÒ equivalents, as this concentration
of TroloxÒ yielded the highest ferrous-TPTZ absorbance value at
the fixed end time.48 The assay was performed over a period of
33 min,49 to allow the reaction to reach an endpoint, since not all
antioxidants are equally rapid reductants.48
1.3. Biological assays
1.3.1. ORAC assay
The oxygen radical absorbance capacity (ORAC) assay assessed
the ability of compounds to scavenge peroxyl radicals, one of the
H
3C
O
O
O
O
5
O
Benzene,
p-TosOH
CH3
4
R1
R2
R1
R2
R1
6
7
9
3
O
+
2'
5'
2
1) -H2O
2) DPE
1'
10
3'
4'
N
H
NH2
R2
NH
8
R3
R3
R3
6'
(7-10)
R1,2,3 = H, OH
Scheme 1. General scheme for Conrad–Limpach synthetic method with p-TosOH (para-toluene sulfonic acid) and DPE (diphenyl ether).

812
J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
1.3.3. TBARS assay
hydrogen donor mechanism of action was investigated in this
assay.55
Lipid peroxidation of polyunsaturated fatty acids produces alde-
hydes50 that lead to the breakdown of lipid membranes and in-
creased levels of the bio-marker malondialdehyde (MDA).1,2,51,52
The aldehydes produced are thiobarbituric acid-reactive sub-
stances (TBARS) able to form complexes with thiobarbituric acid
(TBA) to produce a pink-coloured complex, spectrophotometrically
measured at a wavelength of 532 nm.53TBARS results are ex-
pressed as the amount of free MDA equivalents51 per 1 mg tissue53
through the calibration curve generated from 1,1,3,3-tetrameth-
oxypropane.51 The TBARS method was adapted from Ottino and
Duncan and evaluated the extent of lipid peroxidation53 by mea-
suring the amount of free malondialdehyde equivalents produced
after incubation with the toxin-solution, consisting of hydrogen
peroxide, iron(III)chloride and vitamin C.54 The hypothesised
2. Results and discussion
Evaluation of the antioxidant activity of the synthesised 2-phe-
nylquinolin-4(1H)-ones compared to structurally related flavones
was done using chemical and in vitro antioxidant evaluations
(Table 2). Whole rat brain homogenate and DMSO, as solvent for
all test compounds, was used in these assessments and TroloxÒ,
a water soluble derivative of Vitamin E was used as a positive
control. According to Prior et al., DMSO may act as an antioxidant
and was therefore included in the blank to correct for this effect.56
Test compound concentrations of 1 mM were found to be overly
concentrated in the ORAC and FRAP assays and the compounds
Table 2
Data obtained in the ORAC, FRAP and TBARS assays
Test compound (mM)
ORAC-valuee
equivalents) SEM
(
l
M TroloxÒ
FRAP-valuef (% TroloxÒ
equivalents) S.D.
TBARS-value (nmol MDA/
mg tissue) SEM
Blank homogenatea
Toxinb
—
—
—
—
—
—
—
—
—
—
—
—
0.27 0.08
1.09 0.07
0.93 0.09
0.17 0.01
0.04 0.01
0.03 0.004
—
0.80 0.04
0.77 0.05
0.73 0.03
—
DMSOc
TroloxÒd
0.01
0.1
1
0.001
0.01
0.1
1
2
1.40 0.09
N/A
N/A
—
1.97 0.21
2.42 0.57
0.98 0.11
—
2.82 0.36
1.69 0.53
0.23 0.61
—
1
0.001
0.01
0.1
1.50 0.26
N/A
N/A
—
0.93 0.03
0.92 0.05
0.77 0.04
1
3
0.001
0.01
0.1
10.23 2.44
30.46 0.03
55.42 2.17
—
2.65 0.04
3.68 0.78
15.41 0.25
—
0.20 0.14
3.73 0.36
2.72 0.14
—
1.84 0.32
2.47 0.07
0.61 0.36
—
2.04 0.04
2.52 0.43
15.13 1.14
—
0.53 0.68
1.06 0.07
4.79 1.00
—
2.14 0.32
9.53 0.07
58.88 0.39
—
ꢁ3.48 0.57
ꢁ1.26 2.28
1.41 0.21
—
3.56 1.11
25.49 0.04
97.71 0.39
—
—
0.93 0.06
0.83 0.07
0.07 0.02
—
0.78 0.02
0.83 0.04
0.87 0.04
—
0.82 0.07
0.79 0.05
0.63 0.07
—
0.42 0.07
0.48 0.09
0.91 0.04
—
0.91 0.04
0.87 0.05
0.87 0.06
—
0.48 0.08
0.31 0.07
0.05 0.02
—
0.69 0.09
0.71 0.09
0.72 0.07
—
0.58 0.12
0.49 0.09
0.27 0.04
1
4
0.001
0.01
1
N/A
1.51 0.39
2.40 1.33
—
21.44 1.04
37.07 1.07
59.02 4.14
—
N/A
N/A
N/A
—
0.1
5
0.001
0.01
0.1
1
6
0.001
0.01
0.1
1
7
0.001
0.01
0.1
N/A
N/A
1.91 1.26
—
12.28 0.57
29.01 1.29
58.29 1.27
—
15.27 0.91
26.41 0.83
57.51 3.34
—
9.18 0.45
26.07 0.56
65.39 0.91
—
1
8
0.001
0.01
0.1
1
9
0.001
0.01
0.1
1
10
0.001
0.01
0.1
1
a
b
c
Rat brain homogenate was used in TBARS.
Toxin solution induced lipid peroxidation in the TBARS assay.
10% DMSO blank was used in the TBARS assay. Blank DMSO values were deducted from ORAC and FRAP values.
d
e
TroloxÒ Standard Curves were used in the FRAP and ORAC assays, whereas TroloxÒ was assessed as a compound in the TBARS assay.
Negative ORAC-values indicated the absence of peroxyl scavenging activity and were obtained when the blank AUC had a greater value than that of the sample AUC. These
values were omitted from further calculations (Section 2).
f
Negative FRAP-values indicated the absence of ferric reduction activity.

J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
813
were therefore assessed at concentrations of 0.1 mM, 0.01 mM and
0.001 mM.
lin-4(1H)-ones, with the 1 mM concentration performing better
than 0.01 mM TroloxÒ. The compound displaying the best activity
in the TBARS assay was 4-hydroxyquinoline (2).
The ORAC assay was performed to establish the ability of the
test series to scavenge peroxyl radicals (Fig. 3). The 2-phenylquin-
olin-4(1H)-ones demonstrated moderate activity at 0.001 mM,
with 7-hydroxy-2-phenylquinolin-4(1H)-one (9) observed to be
the best of this group. 6-Hydroxyflavone (5) however, performed
the best of the complete test series. In the FRAP assay the chemical
ability to reduce ferric iron was evaluated (Fig. 4). In this case the
2-phenylquinolin-4(1H)-ones showed the best activity of the test
compounds, with the performance of 8-hydroxy-2-phenylquino-
lin-4(1H)-one (10) comparable to that of TroloxÒ, followed by
6-hydroxy-2-phenylquinolin-4(1H)-one (8) at 0.1 mM. In the
TBARS assay the ability of the compounds to scavenge the hydroxyl
radical was assessed (Fig. 5). In this assay 6-hydroxy-2-phenyl-
quinolin-4(1H)-one (8) performed the best of the 2-phenylquino-
In the ORAC assay the 0.01 mM and 0.1 mM concentrations of
8-hydroxyquinoline (3), 6-hydroxyflavone (5) and 6-, 7- and
8-hydroxy-2-phenylquinolin-4(1H)-one (8–10) did not mathemat-
ically allow the necessary standard decrease in fluorescence to 5%
of the initial value (Fig. 3)—causing a calculation47 error of the
ORAC-value at 0.01 mM and 0.1 mM—indicating that these
concentrations of the test compounds were overly concentrated
and that AAPH, the radical generator was overpowered. This
caused a near continuous protection of fluorescein and suggested
potent peroxyl scavenging activity. This observation led to the
conclusion that these test compounds were superior antioxidants
compared to the other compounds, which demonstrated normal
decay at the same concentrations. It was however possible to
Figure 3. ORAC-values obtained for all the test compounds at three concentrations expressed as TroloxÒ equivalents per litre sample. Each bar represents the mean S.D.;
⁄R.S.D. <5%; N = 3.
Figure 4. FRAP-values obtained for all test compounds in three concentrations at t = 33 min. Each bar represents the mean S.D.; ⁄R.S.D.<5%; N = 3; ⁄⁄⁄p < 0.0001 versus
0.1 mM TroloxÒ #p = 0.001 versus 0.1 mM 8-hydroxyquinoline (3); àp = 0.0035 versus 0.1 mM 7-hydroxyflavone (6); §p = 0.0002 versus 0.1 mM 6-hydroxy-2-phenylquinolin-
;
4(1H)-one (8); ¥p < 0.0001 versus 0.1 mM 8-hydroxy-2-phenylquinolin-4(1H)-one (10) (Paired t-test).

814
J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
Figure 5. Lipid peroxidation inhibition of all test compounds at three concentrations. Each bar represents the mean S.E.M.; N = 10; #p < 0.0001 versus blank; ⁄⁄⁄p < 0.0002
versus Toxin; ⁄⁄p = 0.0014 versus Toxin; §p = 0.0217 versus blank; ⁄p = 0.1902 versus Toxin; ¥p = 0.0104 versus blank; àp = 0.0013 versus Toxin.
compare the 0.001 mM concentration of all test compounds
(Fig. 3), as well as the 0.1 mM concentration of flavone (4) and 2-
phenylquinolin-4(1H)-one (7) as they allowed appropriate decay.
At 0.001 mM concentration, better protection of fluorescein was
observed
for the 2-phenylquinolin-4(1H)-ones than for the quinolines.
6-Hydroxyflavone (5), however showed the best activity. Flavone
(4) also performed slightly better than 2-phenylquinolin-4(1H)-
one (7) at a concentration of 0.1 mM. The peroxyl radical quench-
ing ability was increased in a concentration dependent manner for
most of the compounds, illustrating increased activity at higher
concentrations.
In general, hydroxyl substitution increased peroxyl scavenging
activity at 0.001 mM in this experiment (Fig. 3, Table 2). The
increased activity of the 8-hydroxyl substituted quinoline (3)
indicated increased scavenging activity in comparison to the other
quinolines (1 and 2). For substitution of 2-phenylquinolin-4(1H)-
one, the activity increased from 6-hydroxyl (8) to 7-hydroxyl (9)
substitution, with the 7-hydroxyl (9) performing second best of
all the test compounds. This trend was ascribed to the stabilising
effect of the C-4 carbonyl of 7-hydroxy-2-phenylquinolin-4(1H)-
one (9), which enhanced the proton donating ability of the
hydroxyl group. Weaker activity was observed for the quinolines
and 2-phenylquinolin-4(1H)-ones compared to 6-hydroxyflavone
(5) at 0.001 mM, possibly due to the pKa of the amine groups in
the basic environment of this assay. It is possible that the phenolic
compounds were hindered from donating protons in this hydro-
philic assay, since hydrogen bonds might have interfered with
hydrogen atom transfer.57,58
In the FRAP assay the 2-phenylquinolin-4(1H)-ones showed
prominent activity when compared to TroloxÒ while the quinolines
and flavones had low activity in reducing ferric-TPTZ to the blue
ferrous complex (Fig. 4). The highest activity was observed for 8-
and 6-hydroxy-2-phenylquinolin-4(1H)-one (10 and 8), followed
by 8-hydroxyquinoline (3) and 7-hydroxyflavone (6). Activity com-
parable to that of 0.1 mM TroloxÒ was observed for 8-hydroxy-2-
phenylquinolin-4(1H)-one (10) at the same concentration at the
fixed end time.
7-Hydroxy-2-phenylquinolin-4(1H)-one (9) and 6-hydroxyflavone
(5) were the exceptions in this case with very modest activity,
the exact opposite of what is observed in the ORAC assay. However,
the 8-hydroxyl substituted compounds (3 and 10) had the highest
scavenging activity in their structural groups in both the ORAC and
FRAP assays (Figs. 3 and 4).
In general the TBARS assay showed that hydroxyl substitution
of compounds increased their hydroxyl radical scavenging activity
by lowering the absorbance values of MDA (Fig. 5). The 6-hydroxy-
2-phenylquinolin-4(1H)-one (8) performed best of the 2-phenyl-
quinolin-4(1H)-ones, followed by the 8-hydroxyl (10) derivative,
with the 7-substituted compound (9) demonstrating the lowest
scavenging activity, as was observed in the FRAP assay. Of the flav-
ones, 7-hydroxyflavone (6) had better activity than 6-hydroxyflav-
one (5), as in the FRAP assay, indicating that the pH of the assay
environment as well as the type of radical affected the hydrogen
donating ability of the compounds in solution.
MDA equivalent production was reduced to below that of
the blank brain homogenate by 8-hydroxyquinoline (3) and the
6- and 8-hydroxy-2-phenylquinolin-4(1H)-ones (8 and 10),
indicating complete inhibition of the toxin-induced peroxidation
at 1 mM. Inhibition of hydroxyl radical production caused by the
Haber–Weiss reaction, as evaluated in the FRAP assay, decreased
lipid peroxidation further down the oxidative chain, as simulated
in the TBARS assay. As postulated from the FRAP assay results,
the 6- and 8-hydroxy-2-phenylquinolin-4(1H)-ones (8 and 10)
displayed high activity in the lipid peroxidation assay, indicating
that the FRAP assay correlated well with the data obtained in this
biological assay. Not all test compounds illustrated dose dependant
activity, indicating that other factors might be involved in the func-
tioning of these compounds in living systems, at different concen-
trations. Further studies need to be conducted to elaborate on this
finding.
3. Conclusion
Free radicals contribute to neurodegenerative disorders by
causing damage to cell membranes, proteins and DNA. Antioxi-
dants may be useful in preventative therapy and may slow the
onset and progression of neurodegeneration. The data obtained
For all three structural groups, the best reducing activity was
generally observed for the hydroxyl substituted structures.

J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
815
in this study confirmed the antioxidant activity of a series of
4-quinolones and demonstrated the potential for employing these
structures in the development of antioxidant approaches to
neurodegenerative disease.
respectively with 1H NMR signal multiplicity denoted as s (singlet),
br s (broad singlet) d (doublet), dd (doublet of doublets), t (triplet),
q (quartet) or m (multiplet). Appropriate deuterated solvents were
used. The uncorrected melting points were established using a Stu-
art melting point SMP10 apparatus with glass capillary tubes.
Infrared spectra (IR) were obtained at 4000–450 cmꢁ1 from a Nico-
let Nexus 470-FT-IR Spectrometer, using KBr pellets and the diffuse
reflectance method. A Thermo Electron LXQ ion trap mass spec-
trometer (MS) with atmospheric pressure chemical ionisation
(APCI) source set at 300 °C was used to obtain low resolution APCI,
From the biological evaluations it is possible to establish that
2-phenylquinolin-4(1H)-ones act as chain breaking antioxidants,
with a hydrogen donor mechanism of action. It is clear that hydro-
xyl substitution leads to an increase in antioxidant activity, with
the 8- and 6-hydroxyl substitution of the 2-phenylquinolin-
4(1H)-ones (10 and 8) enhancing antioxidant activity in the FRAP
and TBARS assays and the 7-substitution (9) in the ORAC assay.
This can be attributed to the effect of the C4-carbonyl group,
stabilising the anion of the 7-substituted compound (9), leading
to increased hydrogen donation and antioxidant activity in the
ORAC assay. The addition of an 8-hydroxyl substitution in the 2-
phenylquinolin-4(1H)-one (FRAP) and in particular the quinoline
series (ORAC and TBARS) indicated an increase in antioxidant
activity in their separate structural groups.
with capillary voltage at 7.0 V and Corona discharge at 10 lA,
while a Thermo Electron DFS magnetic sector mass spectrometer
at 70 eV and 250 °C was used to obtain high resolution electron
ionisation (HREI) spectra. Samples were introduced by a heated
probe and perfluorokerosene was used as reference compound.
4.3. Synthesis
Overall, the acidity of the protonated amine in the synthesised
series when in solution, proved to be advantageous compared to
the basic amine of the quinolines, especially in the acidic FRAP as-
say. The hydroxyl substituted 2-phenylquinolin-4(1H)-ones (8–10)
outperformed the flavones, with exception of 6-hydroxyflavone (5)
in the ORAC assay, indicating that under certain conditions the
hydroxyl substituted 2-phenylquinolin-4(1H)-ones may inhibit
radical mediated damage better than the flavones and therefore
show promise as possible neuroprotective agents. This indicates
that the acidic, protonated amine was able to donate a proton
while this was not possible in the flavone series. The type of radical
present in the assay was also found to affect the ability of the test
compound to donate hydrogen to neutralise or scavenge it.
Further studies into the iron chelating potential and in vitro and
in vivo blood–brain barrier permeability and anti-apoptotic
activity of these compounds need to be conducted. Compounds
displaying favourable activity might prove useful as possible lead
compounds or therapeutic agents in combating neurodegenerative
disease resulting from oxidative damage.
The Conrad–Limpach method for the synthesis of 4-quinolones
was followed as described by Somanathan and Smith (1981).44 The
reaction was stirred in benzene at reflux temperature for 19 hours,
with para-toluene sulfonic acid (50 mg, 2.629 mol) as catalyst. The
removal of diphenyl ether in the work-up was achieved with a sil-
ica column, eluting over several hours with petroleum ether until
all diphenyl ether was eluted from the product mixture.
4.3.1. 2-Phenylquinolin-4(1H)-one (7)
Aniline (279.39 mg, 0.003 moles) was cyclisised with ethyl ben-
zoylacetate (576.66 mg, 0.003 moles). In vacuo removal of benzene
yielded a yellowish oil, which was refluxed with 15 ml diphenyl
ether for 30 min. The product was purified by column chromatog-
raphy, initially eluting with petroleum ether, to remove diphenyl
ether, then with 97% DCM, 3% MeOH (Rf = 0.115) to obtain the
crude product, which was suspended in toluene and filtered to ob-
tain the pure product, as an off-white amorphous solid (Yield:
203 mg, 0.917 mmol, 30.58%). Rf = 0.115 (97% DCM, 3% MeOH);
C
15H11NO; Mp = 259 °C; APCI-MS (300 °C, 7.0 V, 10 lA) m/z:
221.64 (M+); HREI-MS: calcd. 221.08352, exp. 221.08371; mmax
(KBr, cmꢁ1): 3200.0, 3150.0, 3120.0, 3100.0, 3080.0, 3052.7,
1733.8, 1683.6, 1635.6, 1582.2, 1547.7, 1501.3, 1473.5, 1449.9,
1432.8, 2971.4, 800.5; dH (600 MHz, DMSO-d6): 6.34 (s; 1H; H-3),
7.331 (m; 1H; H-6), 7.576 (s; 3H; H-8; H-30; H50), 7.664 (m; 1H;
H-7), 7.761 (d; 1H; J = 7.96 Hz; H-40), 7.83 (d; 2H; J = 2.9 Hz; H-20,
H-60), 8.094 (d; 1H; J = 7.47 Hz; H-5), 11.732 (br s; 1H; NH-1); dC
(150 MHz, DMSO-d6): 107.24 (C-3), 118.80 (C-40), 123.25 (C-6),
124.65 (C-5, C-10), 127.37 (C-20, C-60), 128.96 (C-30, C-50), 130.39
(C-8), 131.74 (C-7), 134.25 (C-2), 140.54 (C-9), 150.05 (C-10),
176.90 (C-4).
4. Experimental
4.1. In silico physicochemical studies
The physicochemical properties of the test compounds, that is,
molecular weight, cLogP, H-bond acceptor, H-bond donor, pKa,
ionisation potential, drug score, drug likeness and LogBB values
were calculated using the online Osiris Property Explorer
(http://www.organic-chemistry.org/prog/peo/),39Chemsilico Prop-
erty Predictor for Drug Discovery (http://www.chemsilico.com),37
Advanced Chemistry Development ChemsketchÒ40,41 and Accelrys
DS ViewerPro 5.038 software for prediction of drug oral bioavail-
ability, toxicity and blood–brain barrier permeability.
4.3.2. 6-Hydroxyl-2-phenylquinolin-4(1H)-one (8)
Ethyl benzoylacetate (576.66 mg, 0.003 mole) was refluxed
with 4-aminophenol (327.39 mg, 0.003 mol). Removal of benzene
after 19 h yielded a dark yellow oil, which was refluxed with di-
phenyl ether. The petroleum ether column was flushed with ethyl
acetate after removal of diphenyl ether and the crude product pre-
cipitated from toluene. The crude was further diluted in ethyl ace-
tate (EtOAc) and extracted with water at pH 12. The water phase
was acidified to pH 3 and extracted with EtOAc. The deep brown
crude product was dried and resuspended in toluene and filtered.
Further purification was performed with a Discovery Solid Phase
Extraction Tube, from SUPELCO, Sigma–Aldrich, with positive pres-
sure (97% DCM, 3% MeOH). The retained product was flushed from
the tube with methanol to yield a dark brown amorphous solid
(Yield: 83 mg, 0.35 mmole, 11.66%). Rf = 0.71 (EtOH); C15H11NO2;
4.2. Chemistry: General procedures
All reagents were obtained commercially and used without
further purification. TLC Silicagel 60 F254 aluminium sheets from
Merck, Darmstadt, Germany and MerckÒSilica 60 (0.063
0.200 mm) were used as stationary phases for, respectively, thin
layer and column chromatography, with the appropriate mobile
phase constituted volumetrically immediately before use. Rf-val-
ues were observed under UV-light at 254 nm and 366 nm wave-
lengths.1H, 13C, HSQC and COSY NMR spectra were obtained from
a Bruker Advance 600 Spectrometer, in a 14.09 Tesla magnetic
field, using tetramethylsilane as a point of reference and measuring
all chemical shifts in parts per million (ppm). 1H NMR and 13C NMR
were recorded at frequencies of 600 MHz and 150 MHz,
mp = 151 °C; APCI-MS (300 °C, 7.0 V, 10
HR-MS: calcd. 237.07843, exp. 237.07833; mmax (KBr, cmꢁ1):
l
A) m/z: 237.67 (M+);

816
J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
3300.0, 3230.0, 3200.0, 3130.0, 3100.0, 3070.0, 2972.1, 2950.0,
1704.6, 1592.6, 1580.0, 1560.0, 1510.6, 1490.0, 1449.7, 832.8; dH
(600 MHz, DMSO-d6): 6.22 (s; 1H; H-3), 6.40 (d; 1H; J = 8.03 Hz;
H-7), 6.45 (d; 1H; J = 8.03 Hz; H-8), 6.66 (d; 2H; J = 8.08 Hz; H-20,
H-60), 7.17 (m; 2H; J = 8.08 Hz; H30; H50), 7.41 (m; 1H; H-40), 7.55
(s; 1H; H-5), 9.71 (s; 1H; OH-6); 11.61 (br s; 1H; NH-1); dC
(150 MHz, DMSO-d6): 105.77 (C-3), 107.23 (C-40), 115.14 (C-20, C-
60), 115.22 (C-7), 115.52 (C-8), 120.17 (C-30), 120.60 (C-6), 121.98
(C-50), 128.21 (C-10), 128.98 (C-5), 131.47 (C-2), 140.67 (C-9),
148.21 (C-10), 186.23 (C-4).
in the FRAP and ORAC assays, save for DMSO and the test com-
pounds, were generously donated by the Department of Biochem-
istry, NWU, Potchefstroom Campus.
4.4.2. ORAC assay
4.4.2.1. Reagents. Phosphate buffer, 75 mM, was prepared by
diluting 1 M K2HPO4 with 1 M KH2PO4 aqueous solutions in a
61.6:38.9 v/v ratio and further diluting with Milli-Q water to
75 mM. The pH was adjusted to 7.4 and the buffer stored in the
fridge. A 265 mM fluorescein stock solution was prepared and fur-
ther diluted with buffer before each assay to yield a 112 nM work-
ing solution. The 72 mM AAPH solution was prepared with buffer
immediately before use and kept on ice, while a sufficient quantity
4.3.3. 7-Hydroxyl-2-phenylquinolin-4(1H)-one (9)
Ethyl benzoylacetate (576.66 mg, 0.003 mol) was refluxed with
3-aminophenol (327.39 mg, 0.003 mol). Evaporation of benzene
yielded an amber coloured oil, which was refluxed with diphenyl
ether for 30 min. After removal of diphenyl ether the product
was purified by extraction with basic water from EtOAc. The water
phase was brought to pH 3 and extracted with EtOAc to yield the
crude product, which was further purified using silica column
chromatography (97% DCM, 3% MeOH; Rf = 0.61, EtOH). Suspend-
ing the product firstly in toluene, then in ethanol yielded the pure
product by filtration as a light brown crystalline solid (Yield:
178 mg, 0.75 mmol, 25.0%). Rf = 0.61 (97% DCM, 3% MeOH);
of the 250 l
M TroloxÒ standard solution was prepared before each
assay, dissolving TroloxÒ in DMSO by tip-sonication for 2 min. The
water insoluble test compounds were dissolved in DMSO to yield
final concentrations of 0.1 mM, 0.01 mM and 0.001 mM in the
well-plate in 10% DMSO. All reagents were commercially available.
4.4.2.2. Instrumentation. The ORAC assay was performed at
485 nm excitation and 530 nm emission wavelengths in static
mode on a BioTek FL600 plate reader, using KC4 software.
C15H11NO2; mp = 267 °C; APCI-MS (300 °C, 7.0 V, 10
lA) m/z:
4.4.2.3. Method. The TroloxÒ standard curve was generated by
237.62 (M+); HR-MS: calcd. 237.07843, exp 237.07854; mmax (KBr,
cmꢁ1): 3200.0, 3100.0, 3086.0, 3050.0, 3010.0, 2960.0, 2925.8,
2900.0, 1598.6, 1578.5, 1529.7, 1510.0, 1500.0, 1480.0, 1449.9,
1436.1, 1417.2, 852.6; dH (600 MHz, DMSO-d6): 6.38 (s; 1H; H-3),
6.57 (d; 1H; J = 5.01 Hz; H-6), 7.151 (d; 1H; J = 5.01 Hz; H-5),
7.50 (s; 1H; H-8), 7.58 (m; 3H; H-20, H-40, H-60), 7.83 (m; 2H;
H30; H50), 12.16 (s; 1H; OH-7), 14.52 (br s; 1H; NH-1); dC
(150 MHz, DMSO-d6): 105.86 (C-3), 107.50 (C-5), 108.02 (C-6),
127.34 (C-10), 127.83 (C-30, C-50), 129.17 (C-20, C-60),131.02 (C-
14), 133.58 (C-2), 133.99 (C-8), 141.39 (C-9), 151.76 (C-10), 161.0
(C-7), 182.15 (C-4).
diluting the TroloxÒ working solution with DMSO to yield 20
ll
standard solutions in a final 10% DMSO concentration in the first
row of an opaque 96-well plate. Concentrations of 0, 2.5, 5, 10,
15 and 20
To each well 80
l
M were tested in duplicate, in wells next to each other.
l of fluorescein solution (112 nM) was added with
l
a multi-channel pipette. The sensitivity of the well plate was deter-
mined before initiation of the reaction, in order to ascertain that
emission values fell within the measurable range of the plate read-
er. The sensitivity was set at 146 and the relative fluorescence
units (RFU) established to be approximately 65,000. The reaction
was started on adding 100 ll of 72 mM AAPH solution, yielding a
final 36 mM concentration in the well. The fluorescent decay was
measured at 5 min intervals over a period of 3 h. A second order
polynomial slope was obtained for the standard curve, enabling
the use of the area under the curve (AUC) to determine the TroloxÒ
equivalents of the assayed compounds. The test compounds were
diluted to give 20 ll volumes with final concentrations of
0.1 mM, 0.01 mM and 0.001 mM in 10% DMSO, and were placed
in the opaque well plate in triplicate, in wells next to each other.
4.3.4. 8-Hydroxyl-2-phenylquinolin-4(1H)-one (10)
Ethyl benzoylacetate (576.66 mg, 0.003 mol) was refluxed with
2-aminophenol (327.39 mg, 0.003 mol) for 19 h, the solvent evap-
orated and the reaction heated at reflux with diphenyl ether. Re-
moval of diphenyl ether was done as described above and the
brown crude eluted from the column with ethanol. After acid-base
extraction, as described above, precipitation from ethanol yielded
the pure product as a brown amorphous solid (Yield: 87 mg,
0.37 mmol, 12.2%). Rf = 0.767 (EtOH); C15H11NO2; mp = 272 °C;
4.4.2.4. Data collection. Theformulae utilised by Cao and Prior,47
were applied to obtain the ORAC-values as micromoles TroloxÒ
equivalents. The ORAC-values indicated the relative protection of
fluorescein by the sample antioxidants compared to that afforded
by TroloxÒ, by integrating into the TroloxÒ regression equation
(Fig 3).
APCI-MS (300 °C, 7.0 V, 10 l
A) m/z: 237.64 (M+), HR-MS: calcd.
237.07843, exp. 237.07957; mmax (KBr, cmꢁ1): 3293.7, 3110.0,
3100.0, 3050.0, 2970.0, 2900.0, 2550, 1781.6, 1620.0, 1573.4,
1514.4, 1482.7, 1447.8, 1409.3, 1350, 849.9; dH (600.17 MHz,
DMSO-d6): 6.54 (s; 1H; H-3), 7.10 (dd; 1H; J = 7.25 Hz; H-7), 7.19
(t; 1H; J = 7.25 Hz; H-6), 7.51 (m; 3H; H-20, H-40, H-60), 7.56 (dd;
1H; J = 7.25 Hz; H-5), 7.88 (m; 2H; H30; H50), 10.57 (s; 1H; OH-
8); dC (150 MHz, DMSO-d6): 106.54 (C-3), 114.02 (C-5), 114.24
(C-7), 123.94 (C-6), 125.06 (C-10), 127.54 (C-30, C-50), 128.18 (C-
8), 128.78 (C-20, C-60), 130.06 (C-40), 135.36 (C-2), 148.43 (C-9),
150.87 (C-10), 181.08 (C-4).
4.4.2.5. Statistical analysis. Themean of three values was ex-
pressed as the ORAC-value S.E.M. Statistical acceptability was
determined at a relative standard deviation of less than 5%. Further
statistical analysis was performed using a Paired Student–Newman
Keuls t-test on GraphPad Prism.
4.4. Biological evaluation
4.4.3. FRAP assay
4.4.3.1. Reagents. Acetate buffer (300 mM) was prepared by dis-
solving 1.55 g NaAcꢀ3H2O in 8 ml glacial acetic acid and diluting
to 500 ml with double distilled water, adjusting the pH to 3.6. A
40 mM HCl solution was prepared with double distilled water
and used to prepare a 10 mM 2,4,6-tripyridyl-S-triazine (TPTZ)
solution. A 20 mM FeCl3ꢀ6H2O solution was prepared with acetate
buffer and the FRAP reagent subsequently constituted of 25 ml ace-
tate buffer, 2.5 ml TPTZ solution and 2.5 ml FeCl3ꢀ6H2O solution,
4.4.1. Materials
All chemicals were purchased commercially from Sigma–Al-
drich (St. Louis, MO, USA and Steinham, Germany), Aldrich Chem-
ical Company, (Milwaukee, WIS, USA), BHD (Midrand, South Africa
and Darmstadt, Germany), Saarchem (PTY) Ltd (Wadeville and
Muldersdrift, South Africa), Merck (Darmstadt, Germany) and Al-
pha Pharmaceuticals (Durban, South Africa). All the chemicals used

J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
817
directly before addition to the well-plate, given that crystallisation
takes place rapidly. The TroloxÒ standard, as well as the test com-
pounds was prepared in DMSO to yield final concentrations of
0.1 mM, 0.01 mM and 0.001 mM, in a final 10% DMSO concentra-
tion. All reagents were purchased commercially.
and 0.01 mM in 10% DMSO. To establish the effect of the solvent
10% DMSO was assessed in parallel.
4.4.4.4. Method. A calibration curve was generated by plotting
the absorbance of the TBA/MDA-complex against the concentra-
tion MDA. A standard solution of 50 nmol/l 1,1,3,3-tetramethoxy-
propane (TMP) in double distilled water was used in the
calibration curve as a standard, and diluted with phosphate buffer
solution to achieve TMP concentrations ranging between 0 and
25 nmol/l with 5 nmol/l intervals. Before incubation at 60 °C, 2,6-
di-tert-butyl-4-methylphenol, trichloroacetic acid, and thiobarbi-
turic acid (TBA), solutions were added. After cooling on ice, a buta-
nol extraction was completed by centrifugation. From the top
4.4.3.2. Instrumentation. Absorbance values were measured at
595 nm with a BioTek Synergy HT Reader, using Gen5 1.05
software.
4.4.3.3. Method. The 0.1 mM, 0.01 mM and 0.001 mM TroloxÒ
standards were prepared in a 1:9 ratio with double distilled water
to yield a volume of 100 ll in each well of a transparent 96-well
plate. The wells containing the same concentration were placed next
butanol layer, 200 ll was transferred into a 96-well plate and the
to each other in triplicate. The FRAP reagent was constituted and
absorbance measured at a wavelength of 530 nm. Butanol was
used as a blank. The assay was performed as for the TMP standard
curve. The aqueous lipid peroxidation-inducing toxin solution
consisted of 0.5 mM hydrogen peroxide solution, 0.488 mM
iron(III)chloride solution and 0.14 mM ascorbic acid solution in a
ratio of 2:1:1. The final concentration of the positive control was
1 mM TroloxÒ in 10% DMSO. A series of 1 mM, 0.1 mM and
0.01 mM drug concentrations in 10% DMSO solutions were as-
sayed. To ascertain its effect on inhibition of lipid peroxidation,
10% DMSO was tested in parallel and treated exactly as the test
compounds. The absorbance values were converted to nanomoles
MDA equivalents produced per 1 mg tissue.
added with a multi-channel pipette; 125
l
l was added to all wells,
l reagent
followed by a second 125 l addition, yielding a final 250
l
l
per well. Care was taken to start the Gen5 1.05 computer protocol at
exactly 1.5 min after the first row of the well plate was filled with the
first 125 ll of FRAP reagent. Time dependence was implemented to
enhance the reproducibility of the assay. A Lag time of 3 min at 37 °C
ensued before the first absorbance measurement was taken at
595 nm, followed by measurements at 6 min intervals for a period
of 33 min to illustrate the reduction kinetics of the test compounds.
The absorbance mean of 0.1 mM TroloxÒ at 33 min was converted to
the 100% TroloxÒ equivalent FRAP-value since a three to fourfold
increase in activity was seen for the 0.1 mM series (⁄⁄⁄p < 0.0001)
to 0.01 mM and 0.001 mM.
4.4.4.5. Statistical analysis. This experiment was repeated five
times each on two rat brains. A toxin group was included in every
assay and used as a point of reference. Graphpad Prism was used to
statistically analyse data in the Unpaired Student–Newman Keuls
t-test and to determine the standard error of the means (S.E.M.).
Results were expressed as the mean value S.E.M of the 10 runs.
Significant differences were obtained when p < 0.05.
4.4.3.4. Data collection. The absorbance values for all test
compounds were expressed as percentage TroloxÒ equivalents,
employing the 0.1 mM TroloxÒ standard value at the fixed time
of 33 min as a 100% reference. The blank corrected for the absor-
bance of the well plate, solvent and all other reagents.
Acknowledgments
4.4.3.5. Statistical analysis. The assay was performed in triplicate
over a period of six time intervals and was analysed statistically
against the 0.1 mM TroloxÒ standard series in a Paired Student–
Newman Keuls t-test on GraphPad Prism with statistically signifi-
cant difference indicated at p < 0.05. Standard deviation (S.D.) and
relative standard deviation (R.S.D.) were calculated and were
statistically acceptable at R.S.D. <5%.
We are grateful to the National Research Foundation and North-
West University for financial support and the chemicals and equip-
ment provided by the Department of Biochemistry, North-West
University.
References and notes
4.4.4. TBARS assay
1. Markesbery, W. R.; Montine, T. M.; Lovell, M. A. In Pathogenesis of
Neurodegenerative Disorders; Mattson, M. P., Ed.; Humana Press: New Jersey,
2001; pp 21–52.
2. Coyle, J. T.; Puttfarcken, P. Science 1993, 262, 689.
3. Pratico, D.; Delanty, N. Am. J. Med. 2000, 109, 577.
4. Hirai, K. M.; Smith, M. A.; Wade, R.; Perry, G. J. Neuropathol. Exp. Neurol. 1998,
57, 511.
4.4.4.1. Animals. The rats used in the TBARS assay were bred and
maintained in a controlled environment in the North-West Univer-
sity Laboratory Animal Centre and the assays approved by the
North-West University Ethics Committee. Adult Sprague Dawley
rats, weighing between 200 g and 250 g, were decapitated with a
guillotine and the brains removed into phosphate buffer solution
and held on ice. The 10% (w/v) rat brain homogenate was formu-
lated with PBS with the aid of a manual glass–Teflon homogeniser.
5. Pratico, D. Biochem. Pharmacol. 2002, 63, 563.
6. Packer, L.; Tritschler, H. J.; Wessel, K. Free Radical Biol. Med. 1997, 22, 359.
7. Balaban, R. S.; Nemoto, S.; Finkel, T. Cell 2005, 120, 483.
8. Silva, M. T.; Schapira, A. H. V. In Pathogenesis of Neurodegenerative Disorders;
Mattson, M. P., Ed.; Humana Press: New Jersey, 2001; pp 53–79.
9. Shigenaga, M. K.; Hagen, T. M.; Ames, B. N. Proc. Natl. Acad. Sci. U.S.A. 1994, 91,
10771.
4.4.4.2. Instrumentation. The absorbance values were measured
with a Labsystems Original Multiscan RC plate-reader and Genesis
software.
10. McCord, J. M. N. Eng. J. Med. 1985, 312, 159.
11. Fu, W.; Luo, H.; Parthsarathy, S.; Mattson, M. P. Neurobiol. Dis. 1998, 5, 229.
12. Boveris, A.; Chance, B. Biochem. J. 1973, 134, 707.
13. Imlay, J. A.; Chin, S. M.; Linn, S. Science 1988, 240, 640.
14. Radi, R.; Turrens, J. F.; Chang, L. Y.; Bush, K. M.; Crapo, J. D.; Freeman, B. A. J. Biol.
Chem. 1991, 266, 22028.
15. Halliwell, B.; Gutteridge, J. M. C. Free Radical Biol. Med. 1989, 1, 331.
16. Ou, B.; Huang, D.; Hampsch-Woodill, M.; Flanagan, J. A.; Deemer, E. K. J. Agric.
Food Chem. 2002, 50, 3122.
17. Thomas, T. D. J. Chem. Soc., Perkin Trans. II 1994, 9, 1945.
18. Silva, P. J. J. Org. Chem. 2009, 74, 914.
19. Robak, J.; Gryglewski, J. Biochem. Pharmacol. 1988, 37, 837.
20. Bors, W.; Heller, W.; Michel, C.; Saran, M. Methods Enzymol. 1990, 186, 343.
21. Cotelle, N.; Bernier, J. L.; Catteau, J. P.; Pommery, J.; Wallet, J. C.; Gaydou, E. M.
Free Radical Biol. Med. 1996, 20, 35.
4.4.4.3. Reagents. A phosphate buffer solution (PBS) was prepared
and stored in the fridge with 137 mM NaCl, 2.7 mM KCl, 10 mM
Na2HPO4 and 2 mM KH2PO4 in double distilled water and the pH
adjusted to 7.4. A 0.5% solution of BHT in methanol and 10%
solution of trichloroacetic acid (TCA) in double distilled water
was made and stored in the refrigerator. The 0.33% thiobarbituric
acid (TBA) solution was made afresh every day, dissolved in double
distilled water, in an amber bottle. The test compounds were dis-
solved in DMSO to achieve final concentrations of 1 mM, 0.1 mM

818
J. Greeff et al. / Bioorg. Med. Chem. 20 (2012) 809–818
22. Qin, C. X.; Chen, X. C.; Hughes, R. A.; Williams, S. J.; Woodman, O. L. J. Med.
Chem. 2008, 51, 1874.
23. Khlebnikov, A. I.; Schepetkin, I. A.; Quinn, M. T. Bioorg. Med. Chem. 2006, 16,
2791.
40. ACD/ChemSketchÒ 12.01; Advanced Chemistry Development Inc.: Toronto,
Canada, 2010.
41. ACD/LogPÒ 12.01; Advanced Chemistry Development Inc.: Toronto, Canada,
2010.
24. Park, C.; Lee, J.; Jung, H. Y.; Kim, M. J.; Lim, S. H.; Yeo, H. T.; Choi, E. C.; Yoon, E.
J.; Kim, K. W.; Cha, J. H.; Kim, S.; Chang, D.; Kwon, D.; Li, F.; Suh, Y. Bioorg. Med.
Chem. 2007, 15, 6517.
25. Chung, H. S.; Shin, J. C. Food Chem. 2007, 104, 1670.
26. Ou, B.; Hampsch-Woodill, M.; Prior, R. L. J. Agric. Food Chem. 2001, 49, 4619.
27. Lipinski, C. A. Drug Discovery Today 2004, 1, 337.
28. Moncada, S. J.; Palmer, R. M.; Higgs, E. A. Pharmacol. Rev. 1991, 43, 109.
29. Silverman, R. B. In The Organic Chemistry of Drug Design and Drug Action, 2nd
ed.; Elsevier: London, 2004; pp 1–617.
30. Norinder, J. A.; Haeberlein, M. Adv. Drug Delivery Rev. 2002, 54, 291.
31. Van der Waterbeemd, H.; Camenisch, G.; Folker, G.; Chretien, J. R.; Raevsky, O.
A. J. Drug Target 1998, 6, 151.
42. Clark, D. E. J. Pharm. Sci. 1999, 88, 815.
43. Stanchev, S.; Momekov, G.; Jensen, F. Eur. J. Med. Chem. 2008, 43, 694.
44. Somanathan, R.; Smith, K. M. J. Heterocycl. Chem. 1981, 18, 1077.
45. Lange, J. H. M.; Verveer, P. C.; Osnabrug, S. J. M.; Visser, G. M. Tetrahedron Lett.
2001, 42, 1367.
46. Huang, D.; Ou, B.; Hampsch-Woodill, M.; Flanagan, J.; Deemer, E. J. Agric. Food
Chem. 2002, 50, 1815.
47. Cao, G.; Prior, R. L. Methods Enzymol. 1999, 299, 50.
48. Benzie, I. F. F.; Strain, J. J. Anal. Biochem. 1996, 239, 70.
49. Pulido, R.; Bravo, L.; Saura-Calixto, F. J. Agric. Food Chem. 2000, 48, 3396.
50. Faber, J. L. In Pathology of Environmental and Occupational Disease; Craighead, J.
E., Ed.; Mosby: St. Louis, 1995; pp 287–302.
32. Fischer, H.; Gottschlich, R.; Seelig, A. J. Membr. Biol. 1998, 165, 201.
33. Palm, K.; Luthman, K.; Ros, J.; Grasjo, J.; Artursson, P. J. Pharmacol. Exp. Ther.
1999, 291, 435.
34. Austin, R. P.; Davis, A. M.; Manners, C. N. J. Pharm. Sci. 1995, 84, 1180.
35. Seelig, A.; Gottschlich, R.; Devant, R. M. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 68.
36. Engkvist, O.; Wrede, P.; Rester, U. J. Chem. Inf. Comput. Sci. 2003, 43, 155.
37. Chemsilico. Property Predictors for Drug Discovery. Retrieved 14 May 2011,
from http://www.chemsilico.com.
51. Draper, H. H.; Hadley, M. Methods Enzymol. 1990, 186, 421.
52. Esterbauer, H.; Schaur, R. J.; Zollner, H. Free Radical Biol. Med. 1991, 11, 81.
53. Ottino, P.; Duncan, J. R. Free Radical Biol. Med. 1997, 22, 1145.
54. Carr, A.; Frei, B. FASEB J. 1999, 13, 1007.
55. Tan, D.-X.; Reiter, R. J.; Manchester, L. C.; Yan, M.-T.; El-Sawi, M.; Sainz, R. M.;
Mayo, J. C.; Kohen, R.; Allegra, M. C.; Hardelan, R. Curr. Top. Med. Chem. 2002, 2,
181.
56. Prior, R. L.; Hoang, H.; Gu, L.; Wu, X.; Bacchiocca, M.; Howard, L.; Hampsch-
Woodill, M.; Huang, D.; Ou, B.; Jacob, R. J. Agric. Food Chem. 2003, 51, 3273.
57. Pedrielli, P.; Pedulci, G. F.; Skibsted, L. H. J. Agric. Food Chem. 2001, 4, 3034.
58. Pinelo, M.; Manzocco, L.; Nuñez, M. J.; Nicolo, M. C. Food Chem. 2004, 88, 201.
38. Accelrys DS ViewerPro 5.0. Accelrys Inc. 2002.
39. Organic Chemistry Portal. Osiris Property Explorer. Retrieved 3 March 2011,
from http://www.organic-chemistry.org/prog/peo/.