Radical scavenging capacity of 2,4-dihydroxy-9-phenyl-1 H -phenalen-1-one: A functional group exclusion approach

Article abstract of DOI:10.1021/ol400384z

2,4-Dihydroxy-9-phenyl-1H-phenalen-1-one (4-hydroxyanigorufone, 1), a compound isolated from Anigozanthos flavidus and Monochoria elata, displayed a high radical scavenging capacity in the ORAC assay. A systematic approach was adopted in order to explore

Full text of DOI:10.1021/ol400384z

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Radical Scavenging Capacity of
2,4-Dihydroxy-9-phenyl‑1H‑phenalen-1-one:
A Functional Group Exclusion Approach
†
Luisa Duque, Carolina Zapata, Benjamın Rojano, Bernd Schneider, and
†
‡
§
´
,†
ꢀ
Felipe Otalvaro*
´
´
´
Instituto de Quımica, Sıntesis y Biosıntesis de Metabolitos Naturales, Universidad de
Antioquia, A.A 1226, Medellın, Colombia, Escuela de Quımica, Laboratorio Ciencia de
´
´
´
´
los Alimentos, Universidad Nacional de Colombia Sede Medellın, A.A 3840, Medellın,
€
Colombia, and Max-Planck Institut fu€r Chemische Okologie, Beutenberg Campus,
€
Hans-Knoll-Strasse 8, 07745, Jena, Germany
pipelion@quimica.udea.edu.co
Received May 7, 2013
ABSTRACT
2,4-Dihydroxy-9-phenyl-1H-phenalen-1-one (4-hydroxyanigorufone, 1), a compound isolated from Anigozanthos flavidus and Monochoria elata,
displayed a high radical scavenging capacity in the ORAC assay. A systematic approach was adopted in order to explore the effect of each
functional group. H-Atom transfer from the phenolic hydroxyl, a captodative effect from the hydroxy ketone, and the presumed involvement of the
phenyl ring in the termination step of the radical reaction were disclosed as relevant features of this type of antioxidant.
The oxidative degradation of organic molecules consti-
tutes animportant processwithstrong implicationsin both
life and material sciences.1 Generally, it is recognized that
these processes involve the participation of free radicals in
chain reactions from which the oxygen-mediated perox-
idation stands out as an important example:1,2
(AntioxH) that reacts rapidly with the peroxy radical derived
•
from the substrate (RO2 ) to form a relatively stable free
radical (Antiox•) that should react slowly with the molecule
to be protected (RH). In this context, AntioxH is termed a
“chain-breaking antioxidant”.2
Phenolic compounds are arguably the most common
type of chain-breaking antioxidants.3 Interest in their
natural occurring subset has increased considerably due
to restrictions imposed on synthetic antioxidants4 and
consumer perception about organic/natural additives in
the food and pharmaceutical industries.3 Natural phenolic
antioxidantsstudiedsofarcanberoughlydividedinto four
families: phenolic acids, phenolic mono- and diterpenes,
flavonoids, and some volatile oils.3
RH þ initiator f R• þ ::: (initiation)
•
R• þ O2 f RO2 (propagation)
•
RO2 þ RH f R• þ RO2H
In cases where this reaction is undesirable, the pro-
pagating step can be interrupted by introducing a molecule
† Universidad de Antioquia.
Phenylphenalenones are another group of phenolic pig-
ments isolated from Hemodoraceae, Musaceae, Pontederiaceae,
‡ Universidad Nacional de Colombia Sede Medellín.
§
€
Max-Planck Institut f€ur Chemische Okologie.
(1) (a) Newman, A. Free Radical Biol. Med. 2005, 39, 1265–1290.
ꢁ
ꢁ
(b) Pospı
´
sil, J.; Nespurek, S. Polym. Degrad. Stab. 1995, 49, 99–110.
(3) Brewer, M. Compr. Rev. Food Sci. Food Saf. 2011, 10, 221–247.
(4) (a) Velioglu, Y.; Mazza, G.; Gao, L.; Oomah, B. J. Agric. Food
ꢁ
´
Chem. 1998, 46, 4113–4117. (b) Pospısil, J.; Weideli, H.-J. Polym.
(2) Ba.rclay, L.; Vinqvist, M. The Chemistry of Phenols Part 1;
Rappoport, Z., Ed.; John Wiley & Sons, Ltd: Chichester, 2003; Chapter 12,
pp 839ꢀ908.
Degrad. Stab. 1996, 52, 109–117.
r
10.1021/ol400384z
XXXX American Chemical Society

and Strelitziaceae.5 These compounds are involved in the
defense of Musa plants against pathogens,5fꢀi and some
have shown pharmacological properties.6 However, there
are no reports concerning the antioxidant capacity
of phenylphenalenones, despite the fact that they possess
relevant structural characteristics like the capacity to
absorb in the UV region,5a metal coordination capacity,7
and in some cases, polyhydroxylated nuclei equipped with
diverse substitution patterns.5a
Table 1. Oxygen Radical Absorbance Capacity (ORAC), Ferric
Reducing Antioxidant Power (FRAP), and Bond-Dissociation
Enthalpy (BDE, OꢀH bond) Values for Compounds 1ꢀ8a
4-Hydroxyanigorufone (2,4-dihydroxy-9-phenyl-1H-
phenalen-1-one, 1), a compound first isolated from
Anigozanthos flavidus and later found in Monochoria
elata,8 represents one of the simplest polyhydroxylated
phenylphenalenones that have been prepared by total
synthesis.9 This precedent, together withits UVabsorption
spectrum8a and its potential to chelate metals by means of
its R-hydroxy ketone group, renders 1 an interesting target
for antioxidant studies (Figure 1).
Figure 1. Structure of 4-hydroxyanigorufone (2,4-dihydroxy-9-
phenyl-1H-phenalen-1-one, 1). Phenalenone chromophore high-
lighted in bold.
Here, we report the results of hydrophilic oxygen radical
absorbance capacity (ORAC) and ferric reducing anti-
oxidant power (FRAP) assays conducted on 1 and struc-
tural analogues systematically lacking one, two, or three
functional groups (functional group exclusion) together
with their in silico analysis of bond-dissociation enthalpy
(BDE).
Analyticalmethodswerechosen which belong tothe two
main categories of antioxidant assays. Therefore, ORAC
a Groups BꢀD represent analogues of 1 lacking one (B), two (C), or
three (D) functional groups.
(5) (a) Cooke, R.; Edwards, J. Prog. Chem. Org. Nat. Prod. 1981, 40,
153–190. (b) Greca, M.; Previtera, L.; Zarelli, A. Tetrahedron Lett. 2008,
49, 3268–3272. (c) Opitz, S.; Holscher, D.; Oldham, N.; Bartram, S.;
was selected asthe hydrogen atom transfer (HAT) reaction
based assay and FRAP as the single electron transfer
(SET) assay.10 The 2,20-azinobis-3-ethylbenzothiazoline-
6-sulfonic acid (ABTS) method was considered as another
possibility for a SET based assay. However, examination
of the operating pH for both techniques (3.6 vs neutral)
favored the FRAP method in order to evade ambiguity on
the predominant species of the tested compounds.
ORAC and FRAP measures on 1 afforded radical
scavenging capacities of 2.00 ( 3% and <0.01 (μmol
Trolox/ μmol compound), respectively (Table 1). For
comparative purposes, ellagic acid was measured giving
1.34( 4% and 0.04 ( 2% (μmol Trolox/ μmol compound)
in the same assays. In order to explore the structural
€
Schneider, B. J. Nat. Prod. 2002, 65, 1122–1130. (d) Fang, J.; Paetz, C.;
€
Holscher, D.; Munde, T.; Schneider, B. Phytochem. Lett. 2011, 4, 203–
204. (e) Luis, J.; Fletcher, W.; Echeverri, F.; Grillo, T. Tetrahedron 1994,
~
50, 10963–10970. (f) Luis, J.; Quinones, W.; Echeverri, F.; Grillo, T.;
Kishi, M.; Garcia-Garcia, F.; Torres, F.; Cardona, G. Phytochemistry
1996, 41, 753–757. (g) Tsunashi, K.; Nagomi, K.; Nobuhiro, H.;
Mitsuya, T.; Daie, F.; Hajime, O. Biosci. Biotechnol. Biochem. 1998,
62, 95–101. (h) Tsunashi, K.; Nobuhiro, H.; Kumiko, I.; Daie, F.;
Hajime, O. Tetrahedron 2001, 57, 7649. (i) Jitsaeng, K.; Schneider, B.
Phytochem. Lett. 2010, 3, 84–87.
~
ꢀ
(6) (a) Rosquete, L.; Cabrera, M.; Pinero, J.; Martı
L.; Luis, J.; McNaughton, G.; Grillo, T. Bioorg. Med. Chem. 2010, 18,
4530–4534. (b) Luque, J.; Martınez, S.; Saugar, J.; Izquierdo, L.; Abad,
´
n, P.; Fernandez,
´
~
T.; Luis, J.; Pinero, J.; Valladares, B.; Rivas, L. Antimicrob. Agents
Chemother. 2004, 48, 1534–1540.
(7) (a) Mohammad, A.; Satchell, D.; Satchell, R. J. Chem. Soc. B
1967, 723–725. (b) Deun, R.; Fias, P.; Nockemann, P.; Van Hecke, K.;
Van Meervelt, L.; Binnemans, K. Inorg. Chem. 2006, 45, 10416–10418.
€
(8) (a) Holscher, D.; Schneider, B. Phytochemistry 1999, 50, 155–161.
€
€
€
(b) Holscher, D.; Reichert, M.; Gorls, H.; Ohlenschlager, O.;
(10) (a) Huang, D.; Ou, B.; Prior, R. J. Agric. Food Chem. 2005, 53,
1841–1856. (b) Magalhaes, L.; Segundo, M.; Reis, S.; Lima, J. J. Anal.
Bringmann, G.; Schneider, B. J. Nat. Prod. 2006, 69, 1614–1617.
~
ꢀ
ꢀ
(9) Duque, L.; Restrepo, C.; Saez, J.; Gil, J.; Schneider, B.; Otalvaro,
Chim. Acta 2008, 613, 1–19. (c) Prior, R.; Wu, X.; Schaich, K. J. Agric.
Food Chem. 2005, 53, 4290–4302.
F. Tetrahedron Lett. 2010, 51, 4640–4643.
B
Org. Lett., Vol. XX, No. XX, XXXX

reasons for these radical scavenging properties, a system-
atic approach was adopted in which the functional groups
of 1 were replaced by hydrogen. Thus, three levels of
functionalgroupexclusioncan beenvisioned in1, account-
ing for the phenyl ring at C-9, and two hydroxyl groups at
C-2 and C-4, respectively (Table 1, groups BꢀD). The first
level (Table 1, group B) comprises those molecules lacking
only one of the functional groups mentioned above. A
second level can be conceived with those molecules lacking
simultaneouslytwo ofthethree functional groups(Table1,
group C), and a final level where all functional groups
considered are absent and only the tricyclic chromophore
(perinaphthenone, 8) remains.
consecutively to methylation, hydrolysis, FriedelꢀCrafts
cyclization, and dehydrogenation to afford 4-methoxy-
perinaphthenone.9 This compound together with commercial
perinaphthenone (8) served as templates for the installation
of the C-9 phenyl ring by means of a Grignard reaction
(compounds 1ꢀ3, 7, Scheme 1). Hydroxylation of the phe-
nalenone core at C-2 was achieved by means of the Yang-
Finnegan procedure11 (Scheme 1, compounds 3, 6) except for
the cases where a C-4 methoxyl group was present. In the
latter, the JacobsenꢀKatsuki epoxidation proved superior.9
Table 1 reports the ORAC and FRAP measures for all
compounds used in this study. Several conclusions were
inferredfrom the data obtained. Atfirst, itwas noticed that
all samples displayed very small antioxidant-reducing
capacity (small FRAP values) but solid hydrogen-atom-
donating capacity (high ORAC values) relative to Trolox
and ellagic acid, suggesting that HAT and not SET is the
dominat mechanism in these compounds. Compounds 7
and 8 were expected to give low ORAC values due to the
lack of phenolic groups. This was confirmed by experi-
mental data (Table 1). Interestingly, no correlation was
found between FRAP values and calculated ionization
potentials (Table S2, Supporting Information), a situation
that was ascribed to the low sensitivity of the FRAP
method in the present case.
Scheme 1. Synthesis of Compounds Envisioned in the
Functional Group Exclusion Analysis (in Bold)
Comparing ORAC data obtained for compounds 2 and
5 with ORAC data for 3 and 6 (Table 1) suggests that, in
light of otherwise identical structures, the hydroxyl group
at C-4 is more relevant for the hydrogen atom transfer
process than is its counterpart at C-2. However, a syner-
gistic effect between both hydroxyls can be observed by
comparing data from compounds 4, 5, and 6 or 1, 2, and 3.
Specifically compound 5 showed superior ORAC activity
compared to compound 6 (0.61 ( 8% vs 0.49 ( 2% (μmol
Trolox/ μmol compound), Table 1), with both structures
differing only in the position of the hydroxyl group
from C-4 to C-2. Moreover, compound 4, which has
both positions substituted by hydroxyls, showed superior
ORAC activity (1.06 ( 6% (μmol Trolox/ μmol com-
pound), Table 1). Similar results are obtained by the
interpretation of data from compounds 1, 2, and 3.
It is generally accepted that BDE is an important param-
eter for determining the efficacy of HAT-based antiox-
idants and the thermodynamic stability of the generated
radical.12 In the case of phenylphenalenones, BDE data
(Table 1) nicely agreed with experimental ORAC data.
For example, the calculation of C2- and C4-OH BDE in 4
afforded values of 87.92 and 76.72 kcal/mol, respectively
(Table 1), and this last value was significantly inferior to
the value obtained for compound 5 (80.10 kcal/mol),
which in turn was lower than the corresponding value for
compound 6 (89.30 kcal/mol). Similar results were ob-
tained by comparing BDE of compounds 1, 2 ,and 3. This
suggests that hydrogen atom transfer process in molecules
like 1 should start from the hydroxyl at C-4 to form a radical
that seems to be thermodynamically stabilized by the
Scheme 1 illustrates the synthetic sequence for the pre-
paration of the required compounds. This scheme shows
previously reported procedures which produced already
characterized products (except compounds 2 and 4, see
Supporting Information).9,11 In brief, refluxing acryloni-
trile with a basic solution of 2-naphthol afforded the cor-
responding propanenitrile. This compound was subjected
ꢀ
(11) (a) Hidalgo, W.; Duque, L.; Saez, J.; Arango, R.; Gil, J.; Rojano,
ꢀ
B.; Schneider, B.; Otalvaro, F. J. Agric. Food Chem. 2009, 57, 7417–
ꢀ
~
7421. (b) Otalvaro, F.; Quinones, W.; Echeverri, F.; Schneider, B.
ꢀ
J. Label. Compd. Radiopharm. 2004, 47, 147–159. (c) Otalvaro, F.;
Nanclares, J.; Vasquez, L.; Quinones, W.; Echeverri, F.; Arango, R.;
ꢀ
~
(12) Wright, J.; Johnson, E.; DiLabio, G. J. Am. Chem. Soc. 2001,
123, 1173–1183.
Schneider, B. J. Nat. Prod. 2007, 70, 887–890.
Org. Lett., Vol. XX, No. XX, XXXX
C

Figure 3. Structure of lachnanthocarpone (2,6-dihydroxy-9-
phenyl-1H-phenalen-1-one, 9).
Figure 2. Putative mechanism for the radical-chain-breaking
reaction of 1.
was indeed the observed behavior for the applied concentra-
tion range (Figure S29, Supporting Information).
The putative mechanism presented in Figure 2 offers
several opportunities for extrapolation. For example,
2,6-dihydroxy-9-phenyl-1H-phenalen-1-one (9, Figure 3)
would preserve all the structural features relevant for the
activity of 1 plus an extended delocalization of the initial
radical. This statement can be supported by calculations of
C6-OH BDE performed on 2,6-dihydroxy-1H-phenalen-1-
one (9a) for which a value of 74.12 kcal/mol was found
(2.6 kcal/mol less than C4-OH BDE in compound 4). Thus,
compound 9 (Figure 3), also known as lachnanthocarpone,
was synthesized and tested in the ORAC and FRAP assays.11b
We were delighted to find capacities of 2.39 ( 3% (ORAC)
and 0.23 ( 1% (FRAP) (μmol Trolox/ μmol compound) for
9 in these experiments. This ORAC value is significant super-
ior to the one found for compound 1 (2.00( 3%) and almost
twice the one found for ellagic acid (1.34 ( 4%).
It is worth noting that phenolic groups can be involved
in acid/base equilibriums and therefore, the antioxidant
capacity of compounds like 1 is expected to change with
variable pH. However, the majority of HAT based assays
apply a competitive reaction scheme at constant pH that
makes the study of this relationship not feasible at the
moment using standardized techniques.10
simultaneous presence of the carbonyl and the C-2 hydroxyl
groups taking advantage of a captodative effect.13
Comparing the activities of compounds with and with-
out a phenyl ringatC-9(compound1 vs4, 2 vs5, 3 vs6, 7 vs
8) reveals that the phenyl ring is another characteristic that
enhances ORAC activity (Table 1). Interestingly, BDE
calculations did not support the role of the phenyl ring as an
enhancer of thermodynamic stability of the radical gener-
ated from 1. Properly, C4-OH BDE for compounds 1 and 4,
differing only in the presence or absence of the phenyl ring at
C-9, afforded no differences in the values obtained. This
anomalous result hampered what would otherwise be a
linear correlation between C4-OH BDE of phenylphenale-
nones and ORAC activity. Recently, we reported a photo-
chemical step involved in the synthesis of musafluorone in
which 4-hydroxy-2-methoxy-9-phenyl-1H-phenalen-1-one
was transformed by the action of light in open air atmo-
sphere to 5-methoxy-3H-naphtho[2,1,8-mna]xanthen-3-
one.9 Although the mechanistic details of this reaction
remain to be investigated, the reaction conditions and the
structure of the product strongly suggest a radical type
reaction in which the oxygen of the carbonyl group of the
phenalenone nuclei interacts with the phenyl ring at C-9. This
provides a means for rationalizing the role of the phenyl group
in compounds like 1. Properly, the radical formed after
hydrogen abstraction from the hydroxyl group at C-4 could
attack the phenyl ring at C-9, which would constitute an
intramolecular addition reaction (Figure 2). This process
seems to be favorable from a stereoelectronic point of view,
if we take into account that the phenyl ring probably lies in a
plane that makes a dihedral angle of ∼120ꢀ130° with the
phenalenone nucleus.14 At this point, the newly formed
radical could be the substrate for a new hydrogen atom
abstraction that restores aromaticity and minimizes the ability
of 1 to participate in a chain propagation process. Figure 2
illustrates a mechanistic hypothesis for the chain-breaking
reaction of 1 that agrees with the above-mentioned evidence.
This mechanism predicts a linear increase in the antioxidant
capacity with increasing concentration of antioxidant 1. This
In summary, we have shown the high antioxidant po-
tential phenylphenalenones possess as HAT antioxidants.
Structural features responsible for these capacities were
identified. A radical captodative stabilization effect and a
nonpropagating quenching mechanism were proposed as
relevant properties offered by these types of compounds.
The model developed can be used for the rational design of
better antioxidants basedon the perinaphthenone nuclei as
illustrated with lachnanthocarpone (9).
Acknowledgment. We thank Emily Wheeler for edito-
rial assistance and Carlos Gaviria (UN), Marisol Cano
(UA), William Hidalgo (MPI), and Adrian Ramırez (UA)
´
for experimental collaboration. This research was finan-
cially supported by Universidad de Antioquia and the
€
Max-Planck-Institut fur Chemische Okologie.
€
Supporting Information Available. Experimental pro-
cedures, spectroscopic data, and computational details.
This material is available free of charge via the Internet at
http://pubs.acs.org.
ꢀ
(13) Viehe, H.; Janousek, Z.; Merenyi, R. Acc. Chem. Res. 1985, 18,
148–154.
(14) For examples of crystal structures of phenylphenalenones, see:
ꢀ
€
€
(a) Otalvaro, F.; Gorls, H.; Holscher, D.; Schmitt, B.; Echeverri, F.;
~
Quinones, W.; Schneider, B. Phytochemistry 2002, 60, 61–66. (b) Luis, J.;
ꢀ
Fletcher, W.; Echeverri, F.; Grillo, T.; Perales, A.; Gonzales, J. Tetra-
hedron 1995, 51, 4117–4130.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX