Design and evaluation of anacardic acid derivatives as anticavity agents

Article abstract of DOI:10.1016/j.ejmech.2007.08.012

On the basis of antibacterial anacardic acids, 6-pentadecenylsalicylic acids, isolated from the cashew apple, Anacardium occidentale L. (Anacardiaceae), a series of 6-alk(en)ylsalicylic acids were synthesized and tested for their antibacterial activity against Streptococcus mutans ATCC 25175. Among them, 6-(4′,8′-dimethylnonyl)salicylic acid was found to exhibit the most potent antibacterial activity against this cariogenic bacterium with the minimum inhibition concentration (MIC) of 0.78 μg/ml.

Full text of DOI:10.1016/j.ejmech.2007.08.012

Available online at www.sciencedirect.com
European Journal of Medicinal Chemistry 43 (2008) 1315e1320
http://www.elsevier.com/locate/ejmech
Short communication
Design and evaluation of anacardic acid derivatives
as anticavity agents
a
b
Ivan R. Green a, , Felismino E. Tocoli , Sang Hwa Lee ,
*
Ken-ichi Nihei b, Isao Kubo
b,**
a Department of Chemistry, University of the Western Cape, P/Bag X17, Bellville, 7530, South Africa
b Department of Environmental Science, Policy and Management, University of California, Berkeley, CA 94720-3112, USA
Received 26 April 2007; received in revised form 21 July 2007; accepted 9 August 2007
Available online 14 September 2007
Abstract
On the basis of antibacterial anacardic acids, 6-pentadecenylsalicylic acids, isolated from the cashew apple, Anacardium occidentale L.
(Anacardiaceae), a series of 6-alk(en)ylsalicylic acids were synthesized and tested for their antibacterial activity against Streptococcus mutans
ATCC 25175. Among them, 6-(40,80-dimethylnonyl)salicylic acid was found to exhibit the most potent antibacterial activity against this
cariogenic bacterium with the minimum inhibition concentration (MIC) of 0.78 mg/ml.
Ó 2007 Elsevier Masson SAS. All rights reserved.
Keywords: Anacardic acid; Streptococcus mutans; Antibacterial activity; 6-(40,80-Dimethylnonyl)salicylic acid
1. Introduction
of chemical methods are easily recognized [4]. Among them,
safety of the chemicals seems to be the most important, since
S. mutans resides in the mouth.
Dental caries is one of the most ubiquitous infectious disease
in developed countries. Interactive elements, including nutri-
tional intake and presence of cariogenic microflora, are contrib-
utors to this disease. Many recent studies have concluded that
Streptococcus mutans is the primary bacteria causing dental
caries [1,2]. This bacterium adheres firmly to smooth tooth sur-
faces and facilitates the accumulation of other oral microorgan-
isms. Predominant in plaque, S. mutans and other accumulated
microorganisms generate organic acids, primarily lactic acid,
that gradually destroy the enamel surface thereby creating an
opening susceptible to further bacterial degradation and ulti-
mately forming a cavity [1,3]. Theoretically, dental caries
should be preventable by eliminating S. mutans. The difficulties
associated with eliminating this cariogenic bacterium by means
Anacardic acids; 6[80(Z ),110(Z ),140-pentadecatrienyl]sali-
cylic acid (1), 6[80(Z ),110(Z )-pentadecadienyl]salicylic acid
(2), and 6[80(Z )-pentadecenyl]salicylic acid (3), were previ-
ously isolated from the cashew apple Anacardium occidentale
L. (Anacardiaceae) [5,6], and for the purposes of this paper are
referred to as anacardic acid (C15:3), anacardic acid (C15:2) and
anacardic acid (C15:1) for simplicity, respectively (see Fig. 1
for structures). Their diverse biological activities have been
reported which include their potent antibacterial activity
against the Gram-positive bacteria viz. S. mutans [7e11].
These anacardic acids, which are contained in a regularly im-
bibed beverage known as cashew apple juice, may be consid-
ered then as beneficial anticavity agents. However, the
anacardic acids are rather unstable for practical applications
because of their side chain unsaturation. This prompted us to
synthesize a series of anacardic acid analogues possessing
different side chains viz. phenolic, branched and alicyclic by
essentially a four-step protocol [12,13].
* Corresponding author. Tel.: þ27 21 9592262; fax: þ27 21 9593055.
** Corresponding author. Tel.: þ1 510 643 6303; fax: þ1 510 643 0215.
E-mail addresses: igreen@uwc.ac.za (I.R. Green), ikubo@calmail.berkeley.
edu (I. Kubo).
0223-5234/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2007.08.012

1316
I.R. Green et al. / European Journal of Medicinal Chemistry 43 (2008) 1315e1320
Table 1
Minimal inhibitory concentrations of anacardic acids against Streptococcus
mutans ATCC 25175
R1
R2O
Compounds tested
MIC and MBC (mg/ml)a
log P
Anacardic acid (C15:3) (1)
Anacardic acid (C15:2) (2)
Anacardic acid (C15:1) (3)
Anacardic acid (C15:0) (4)
Anacardic acid (C12:0) (11)
a
1.56 (6.25)
3.13 (3.13)
6.25 (6.25)
>800 (ꢀ)
6.62
6.89
7.21
7.53
6.28
1: R = COOH,
8: R1 = H, R
2 = H
9: R1 = COOCH
R2 = H
1
3, R2 = H
10: R1
= COO
H, R2 = Ac
1.56 (6.25)
The data were extracted from our previous papers [7,8,10].
COOH
COOH
COOH
HO
HO
HO
equally as potent activity as anacardic acid (C15:3) against S.
mutans, whereas anacardic acid (C15:0) (4) did not show any
activity up to 800 mg/ml [7]. Anacardic acids are salicylic
acid derivatives with a non-isoprenoid alkenyl side chain and
therefore the differences observed in the activity should be
interpreted to imply that changes in the hydrophobic side
chain tail portions are the major influential dictators to the ac-
tivity. Anacardic acids 1e3 can be isolated in fair quantities
from cashew nut shell liquid (CNSL) [5] which in turn, is
available in large quantities as agricultural waste. However,
the following two disadvantages of anacardic acids need to
be considered.
Firstly, the observed bactericidal activity of anacardic acid
(C15:3) against S. mutans may not be potent enough for practi-
cal applications. This is always a dilemma when the biological
activities of phytochemicals, particularly their antimicrobial
activity, are considered. Hence, studies to enhance their activ-
ity are always considered to be of the utmost importance and
one of the more promising strategies to enhance their activity
consists of combining two or more phytochemicals in their ap-
plication. Thus, combining two or more compounds might, in
general, be superior to the use of a single antimicrobial com-
pound in order to decrease the likelihood for the development
of resistance mechanisms by the microorganisms, in addition
to enhancing and/or broadening their total biological activity
spectrum. Antibacterial agents used against S. mutans were
previously isolated from edible plants [7,15]. The bactericidal
activity of the sweet tasting anethole (5) against this cariogenic
bacterium was previously confirmed by the time-kill curve
experiment [7]. However, bactericidal activity of anacardic
acid (C15:3) against S. mutans was enhanced 8-fold when in
combination with a sublethal amount (equivalent to 1/2
MBC) of anethole. Thus, in this way, the MBC of anacardic
acid (C15:3) was lowered from 6.25 to 0.78 mg/ml in combina-
tion with 200 mg/ml of anethole.
2
3
n
4: n = 4
11: n = 1
OH
O
O
n
H3CO
5
6: n = 1
7: n = 3
Fig. 1. Structures of natural occurring anacardic acids 1e3 and their related
compounds 4e11.
2. Results and discussion
In previous studies on the antibacterial activity of a number
of anacardic acids 1e3 against S. mutans ATCC 25175, the
emphasis was placed on anacardic acid (C15:3) (1) because
this compound showed the most potent bactericidal activity
with the minimum bactericidal concentration (MBC) of
6.25 mg/ml (Table 1). The bactericidal activity was confirmed
by time-kill curve experiments [7,10]. The antimicrobial activ-
ity of anacardic acids was proportional to the degree of the
side chain unsaturation and can be summarized in the follow-
ing decreasing order; C15:3 > C15:2 > C15:1. However, a recent
study indicates that the double bond is not essential in eliciting
the antibacterial activity but is more likely to be associated
with increasing the inherent activity of the compound [14]. In-
terestingly, anacardic acid (C15:2) (2) demonstrated an almost
Secondly, and more importantly, anacardic acid (C15:3) is
rather unstable for practical applications because of its side
chain unsaturation. The inter-relationship between the head
and tail moieties of anacardic acids isolated from the cashew
apple suggests that optimization of activity is a real possibility
through the synthetic approach. Since anacardic acids 1e4 are
salicylic acid derivatives with a pentadec(en)yl side chain, it
was reasoned that alkyl salicylates may show similar antibacte-
rial activity. Hence, both hexyl 2-hydroxybenzoate (6) and
nonyl 2-hydroxybenzoate (7) were synthesized and tested to
determine whether they exhibit antibacterial activity against

I.R. Green et al. / European Journal of Medicinal Chemistry 43 (2008) 1315e1320
1317
S. mutans. It was, however, demonstrated that, neither salicylate
showed this activity, which indicated that the 6-alkylsalicylic
acid scaffold structure is a prerequisite to elicit this particular
antibacterial activity against S. mutans. This suggestion can
be indirectly supported by the observation that cardanol
(C15:3) (8) [8], an artifact of the corresponding anacardic acid
(C15:3) obtained by pyrolysis [6], did not show any noticeable
antibacterial activity against S. mutans. In addition, the previ-
ous observation that neither methyl ether 9 nor acetate 10 of
anacardic acid (C15:3) exhibited any noticeable antibacterial ac-
tivity against S. mutans lent further support for this conclusion.
In general it is observed that antimicrobial activity is in-
versely proportional to the length of the C-6 side chain and
that at a particular critical length it reaches a maximum after
which activity greatly diminishes to finally become inactive.
If only 6-alkylsalicylic acids are compared, their calculated
partition coefficient (calculated log P) values seem to be in
proportion to their antibacterial activity and the maximum ac-
tivity against S. mutans seems to be at a value of log P of
around 6 [16]. Hence, a series of anacardic acids possessing
different lengths of the C-6 side chain were synthesized.
Among these compounds, anacardic acid (C12:0) (11) ex-
hibited the most potent bactericidal activity against S. mutans
viz. phenolic, branched and alicyclic were synthesized [12,13].
In brief, phosphonate ester 12 and commercially available al-
dehydes were condensed under basic Wittig type conditions to
afford the corresponding trans olefins 13 in average yields of
60%. Catalytic hydrogenation of these olefins afforded the sat-
urated aryl esters 14 (85%) which were hydrolysed by sodium
hydroxide in dimethyl sulfoxide (DMSO) at 110 ꢁC to produce
the corresponding aryl carboxylic acids 15 (85%). Demethyla-
tion of the aryl methoxy group was accomplished using boron
tribromide in 1,2-dichloroethane at low temperature to afford
the anacardic acid analogues (70%). Thus the average trans-
formation of phosphonate ester 12 into the anacardic acid
analogues was 26% for the 4 steps [12,13] (Scheme 1).
The three structurally different types of anacardic acid
analogues that were synthesized totaling sixteen were subse-
quently tested for their antibacterial activity against S. mutans
ATCC 25175 using a 2-fold serial broth dilution method. The
highest concentration tested was 200 mg/ml because of a solu-
bility limitation of some samples in the water based medium.
The results are listed in Table 2. The data obtained were com-
pared with that of anacardic acid (C15:3) which is usually
taken as the most active member of the naturally occurring
anacardic acids. All the synthesized compounds were also
tested for their antibacterial activity against Escherichia coli
ATCC 9637, but none of them exhibited any activity up to
200 mg/ml.
The synthetic anacardic acid analogues 16e22 representing
the hydroxyl- and non-hydroxyphenyl analogues at C-6
exhibited a varied activity against S. mutans ATCC 25175 in
which the MIC ranged between 6.25 and >200 mg/ml. The
maximum activity against S. mutans occurred for analogue
22 which had an MIC of 6.25 mg/ml. This activity of analogue
22 was 8-fold superior to that of 6-phenylethylsalicylic acid 21
which had an MIC of 50 mg/ml. It consequently appears that
the length of the linker alkyl side chain plays an important
role in increasing antibacterial activity. However, the activity
of analogue 22 against S. mutans was not superior to that of
anacardic acid (C15:3) which has an MIC of 1.56 mg/ml. The
most important observation in this series was that the antibac-
terial activity was much lower for the anacardic acid analogues
having a phenolic side chain. For example the MICs of the
compounds 16e22 that contain one to three hydroxyl groups
with an MBC of 6.25 mg/ml [8], while anacardic acid (C15:0
)
did not show any activity up to 800 mg/ml. It was important
to include anacardic acid (C15:0) since it can easily be obtained
by hydrogenation over 5% PdeC of the anacardic acid-
containing fraction of CNSL. It should be noted, however,
that although anacardic acid (C15:0) was not effective against
S. mutans it nevertheless exhibited potent antibacterial activity
against Propionibacterium acnes ATCC 11827 with a mini-
mum inhibitory concentration (MIC) of 0.78 mg/ml [17].
From the foregoing results it appears that the MBC of
6.25 mg/ml of anacardic acid (C12:0) seems to be a maximum
and thus optimization of activity through the synthetic ap-
proach would certainly offer an obvious solution for a more
comprehensive evaluation. However, in the initial optimization
investigation through the synthetic approach, only analogues
containing the non-branched aliphatic chains having various
lengths were synthesized [7,18]. This suggested that further
optimization is indeed still achievable and hence, a series of
anacardic acid analogues possessing different C-6 side chains
COOEt
COOEt
MeO
O
MeO
R1
aldehyde or ketone
H2, 5% Pd/C
MeOH
P
OEt
t-BuOK, THF
OEt
R2
12
13
COOEt
COOH
COOH
MeO
R1
MeO
R1
HO
R1
BBr3
0.5 M NaOH aq.
DMSO
CH2ClCH2Cl
R2
R2
R2
30
14
15
Scheme 1. Synthesis of anacardic acid derivatives 12e30.

1318
I.R. Green et al. / European Journal of Medicinal Chemistry 43 (2008) 1315e1320
Table 2
Minimal inhibitory concentrations of C-6 phenolic, alicyclic and branched
anacardic acid analogues against Streptococcus mutans ATCC 25175
COOH
HO
R
Compounds tested
MIC
(mg/ml)
log P
6-[20-(200,400,500-Trihydroxyphenyl)ethyl]salicylic acid (16)
6-[20-(200,500-Trihydroxyphenyl)ethyl]salicylic acid (17)
6-[20-(200,400-Trihydroxyphenyl)ethyl]salicylic acid (18)
6-[20-(300,400-Trihydroxyphenyl)ethyl]salicylic acid (19)
6-[20-(400-Hydroxyphenyl)ethyl]salicylic acid (20)
6-(20-Phenylethyl)salicylic acid (21)
200
200
2.54
2.93
2.93
2.93
3.32
3.71
>200
200
100
R2
R3
R4
R1
*
n
50
R =
R =
6-(40-Phenylbutyl)salicylic acid (22)
6.25 4.55
n
6-Cyclopentylmethylsalicylic acid (23)
6-Cyclohexylmethylsalicylic acid (24)
25
25
3.19
3.61
m
16: n = 1, R1 = R3 = R4 = OH, R2 = H
17: n = 1, R1 = R4 = OH, R2 = R3 = H
18: n = 1, R1 = R3 = OH, R2 = R4 = H
19: n = 1, R1 = R2 = H, R3 = R4 = OH
20: n = 1, R1 = R2 = R4 = H, R3 = OH
21: n = 1, R1 = R2 = R3 = R4 = H
23: m = n = 1
6-Cyclooctylmethylsalicylic acid (25)
6-Cyclododecylmethylsalicylic acid (26)
6-Cyclohexylethylsalicylic acid (27)
6-(40,80-Dimethylnonyl)salicylic acid (28)
6-(20-Ethylheptyl)salicylic acid (29)
6-(20-Methylhexyl)salicylic acid (30)
6.25 4.44
1.56 6.11
3.13 4.02
0.78 5.69
3.13 4.52
24: m = 2, n = 1
25: m = 2, n = 2
26: m = 4, n = 1
27: m = 8, n = 1
12.5
3.69
22: n = 3, R1 = R2 = R3 = R4 = H
on the aromatic ring of the side chain varied between 100e
200 mg/ml for S. mutans.
R =
Most probably, the presence of hydrophilic groups on both
phenyl rings of the molecule, referred to as the head and tail
structure, decreases the overall antibacterial activity against
S. mutans [9]. On the one hand the MICs of the compounds
16e19 suggest that the activity against S. mutans is not af-
fected by either the number or position of the hydroxyl
group(s) on the phenyl ring of the side chain. Thus, the activity
of compound 16, which contains three hydroxyl groups, was
comparable with that of the 17 and 19 both possessing two hy-
droxyl groups on the phenyl ring of the side chain. The MIC of
17 and 19 for S. mutans was 200 mg/ml. On the other hand,
compound 18 did not show any activity against this cariogenic
bacterium up to 200 mg/ml and this is due to the particular
oxygenation pattern on the phenyl ring.
28
R =
R =
29
30
Fig. 2. Structures of anacardic acid derivatives.
The maximum activity of the C-6 branched alkanyl side
chain anacardic acid analogues 28e31 against S. mutans
was found with 6-(40,80-dimethylnonyl)salicylic acid (28)
which had a MIC of 0.78 mg/ml. This rather good activity
of 28 was thus double that of the best natural anacardic
acid (C15:3). With an MIC of 12.5 mg/ml, 6-(20-methylhex-
yl)salicylic acid (30) was the least active of the branched
side-chain anacardic acid analogues studied in this project.
The MIC of analogue 29 having a C9H19 branched side-
chain was 3.13 mg/ml against S. mutans which is compara-
ble with that of anacardic acid (C12:0) [7]. It thus appears
that the presence of a branched side-chain at C-6 in the
salicylic acid nucleus plays a significant role in increasing
the antibacterial activity of the 6-alkylsalicylic acids
(Fig. 2).
The alicyclic anacardic acid analogues 23e27 showed far
more promising activity against S. mutans with MICs ranging
from 1.56 to 25 mg/ml. In general the activity of these syn-
thetic analogues increased in direct proportion to the number
of C-atoms in the cyclic ring. Among these compounds, 6-
cyclopentylmethylsalicylic acid (23) was the least active ana-
logue with an MIC of 25 mg/ml. The maximum activity for this
series was found for 6-cyclododecylmethylsalicylic acid (26)
which demonstrated an MIC of 1.56 mg/ml. The results clearly
show that the activity of the synthetic analogue 26 against
S. mutans is comparable with that of the standard naturally
occurring anacardic acid (C15:3).
Analogue 24 inhibited the growth of S. mutans at a con-
centration of 25 mg/ml, which was less active than 6-cyclohex-
ylethylsalicylic acid (27) that has an additional C-atom in the
linker chain between the phenyl and cyclohexyl rings in the C-
6 side chain. The MIC of 27 was 3.13 mg/ml for S. mutans. A
further most interesting observation was that 6-cyclooctylme-
thylsalicylic acid (25) with an MIC against S. mutans of
6.25 mg/ml, exhibited stronger activity than either that of cy-
clopentylmethyl (23) or cyclohexylmethylsalicylic acid (24)
(see Table 2).
3. Conclusions
On the basis of the data obtained, it may be concluded that
antibacterial anacardic acids act primarily as surface-active
agents to disrupt the native membrane-associated function of
the integral proteins (physical disruption of the membrane)
[11]. However, biochemical mechanisms may also be respon-
sible for eliciting this activity and should therefore not be
ruled out [14]. In addition, it appears that maximum activity

I.R. Green et al. / European Journal of Medicinal Chemistry 43 (2008) 1315e1320
1319
can be obtained when the most appropriate balance between
hydrophilic and hydrophobic moieties is attained. Therefore,
the alkyl side chain plays a direct role in the modulation of
the activity. Currently, however, the precise explanation for
the role of the alkyl side-chain remains unclear. It should fur-
ther be borne in mind that the relevance of the in vitro exper-
iments in simplified systems, when extended to include the
complex interaction between anacardic acids and bacterial
systems representing the real interactions, has to be carefully
considered before any final pronouncements can be made.
Anacardic acid (C15:0) (3) was previously reported to show
high selectivity toward Fe2þ and Cu2þ [19]. This implies that
metal ions might also play a significant role in antimicrobial
activity by reducing their availability for bacteria [20]. It is
considered that this possibility is rather unlikely since anacar-
dic acid (C15:0) (3) did not show any antibacterial activity
against S. mutans even up to 800 mg/ml. However, due to the
paucity of more investigative information, this hypothesis
may not be totally excluded. In addition, metal chelation
may play a significant role in determining the antioxidant ac-
tivity [21]. The chelation ability, rendering the metal ions
inactive to participate in free radical generating reactions,
should be of considerable importance to advance their antiox-
idant activities. Thus, the salicylic acid moiety is also respon-
sible for eliciting antioxidant properties without prooxidant
effects [22e25]. In view of the increasing importance of con-
trolling specific bacteria in the mouth, viz. S. mutans, studies
on the branched side chain anacardic acid analogues may lead
to new anticavity drugs with anti-S. mutans activity.
solid, which was used in the next step without further purifica-
tion. The ester was hydrogenated over 20% Pd(OH)2 on car-
bon (10 mg) in 1% AcOHeEtOAc (4 ml) for 12 h. Filtration
through Celite and concentration followed by silica gel chro-
matography with EtOAcehexane (1e40%) as eluant gave
pure hexyl 2-hydroxybenzoate (6) in 83% yield (2 steps) as
1
a colorless oil (95 mg). H NMR (500 MHz, CDCl3): d 0.91
(t, J ¼ 6.8 Hz, 3H), 1.34 (m, 4H), 1.45 (quin, J ¼ 6.8 Hz, 2H),
1.79 (quin, J ¼ 6.8 Hz, 2H), 4.35 (t, J ¼ 7.0 Hz, 2H), 6.88
(ddd, J ¼ 2.0, 7.0, 8.0 Hz, 1H), 6.98 (dd, J ¼ 2.0, 9.0 Hz, 1H),
7.45 (ddd, J ¼ 2.0, 7.0, 9.0 Hz, 1H), 7.84 (dd, J ¼ 2.0, 8.0 Hz,
1H), 10.84 (s, 1H) ppm; 13C NMR (125 MHz, CDCl3):
d 14.0, 22.5, 25.5, 28.5, 31.4, 65.5, 112.7, 117.6, 119.0,
129.4, 135.6, 161.7, 170.2 ppm; IR (CCl4): 3185, 2850, 1665,
1470, 1290, 1210, 1140, 1070 cmꢀ1; EI-MS (m/z): 222 (Mþ).
4.3. Microorganisms and media
S. mutans ATCC 25175 and E. coli ATCC 9637 were ob-
tained from American Type Culture Collection (Manassas,
VA). The freezeedried culture of S. mutans was inoculated
into 3.7% brain heart infusion (BHI) broth purchased from
Difco Lab. (Detroit, MI) and incubated stationary for 2 days
at 37 ꢁC before the assay. Although only one strain of S. mutans
was tested, compounds active against this strain are expected to
retain a similar order of activity against a variety of strains of
this species. In the case of E. coli, NYG broth, (0.8% nutrient
broth, 0.5% yeast extract, and 0.1% glucose) was used. Nutrient
broth was obtained from BBL Microbiology System (Cockeys-
ville, MD) and yeast extract was purchased from Difco Lab.
4. Experimental
4.4. Antibacterial assays
4.1. Chemicals
The assay was performed by a broth dilution method as pre-
viously reported [8]. Briefly, serial 2-fold dilutions of test com-
pounds were made in DMF and 30 ml of each dilution was
added to 3 ml of BHI broth. This test broth was then inoculated
with 30 ml of a 2-day-old culture of S. mutans. The highest con-
centration used for the assay was 200 mg/ml, unless otherwise
specified, because of solubility limitation in the water-based
media of some of the samples. The lowest concentration of
the test compound resulting in complete inhibition of visible
growth after 2 days of incubation at 37 ꢁC represented the
MIC. The MIC of each compound was determined at least in
triplicate on separate occasions.
Anacardic acids and their analogues were available in our
laboratory from previous investigations [6e9,18], and synthe-
sis of the 6-alkylsalicylic acid analog has been was previ-
ously described [12,13]. Nonyl 2-hydroxybenzoate (7) was
synthesized by the method previously reported [26]. Salicylic
acid and 1,1-diphenyl-2-picrylhydrazyl (DPPH) were pur-
chased from Sigma Chemical Co. (St. Louis, MO). N,N-
Dimethyl formamide (DMF) was obtained from EM Science
(Gibbstown, NJ). Other reagents were purchased from com-
mercial suppliers and used as received, unless otherwise
noted.
4.2. Preparation of hexyl 2-hydroxybenzoate (6)
4.5. log P calculation
Hexyl 2-hydroxybenzoate was synthesized as follows [26].
A mixture of 2-benzyloxybenzoic acid (200 mg, 0.88 mmol)
[27], 1-hexanol (108 mg, 1.06 mmol), and triphenylphosphine
(350 mg, 1.33 mmol) in tetrahydrofuran (4 ml) was cooled to
0 ꢁC and treated with diisopropyl azodicarboxylate (214 mg,
1.06 mmol). After being stirred for 2 h at room temperature,
the solvent was removed in vacuo. The residue was subjected
to silica gel chromatography and the product eluted with 1e
8% EtOAcehexane to give the corresponding ester as white
log P values were calculated by Chem Draw Pro version
4.5 developed by Cambridge Soft Co. (Cambridge, MA) using
Crippen’s fragmentation [16].
Acknowledgements
The authors are grateful to Dr. M. Himejima, Dr. H. Muroi
and Dr. A. Kubo for performing the antibacterial assay at an
earlier stage of the work.

1320
I.R. Green et al. / European Journal of Medicinal Chemistry 43 (2008) 1315e1320
[15] I. Kubo, H. Muroi, M. Himejima, J. Agric. Food Chem. 40 (1992)
245e248.
References
[16] A.K. Ghose, G.M. Crippen, J. Chem. Inf. Comput. Sci. 27 (1987) 21e35.
[17] I. Kubo, H. Muroi, A. Kubo, J. Nat. Prod. 57 (1994) 9e17.
[18] Y. Yamagiwa, K. Ohashi, Y. Sakamoto, S. Hirakawa, T. Kamikawa,
I. Kubo, Tetrahedron 43 (1987) 3387e3394.
[1] S. Hamada, H.D. Slade, Microbiol. Rev. 44 (1980) 331e384.
[2] M.H. de Jung, J.S. van der Hoeven, J.H. van Os, J.H. Olijve, Appl.
Environ. Microbiol. 47 (1984) 901e904.
[3] W.J. Loesche, Microbiol. Rev. 50 (1986) 353e380.
[4] R.J. Fitzgerald, Antimicrob. Agents Chemother. 1 (1972) 296e302.
[5] J.H.P. Tyman, Chem. Soc. Rev. 8 (1979) 499e537.
[19] K.S. Nagabhushana, S.V. Shobha, B. Ravindranath, J. Nat. Prod. 58
(1995) 807e810.
[20] A. Scalbert, Phytochemistry 30 (1991) 3875e3883.
[21] A. Arora, M.G. Nair, G.M. Strasburg, Free Radical Biol. Med. 24 (1998)
1355e1363.
[6] I. Kubo, S. Komatsu, M. Ochi, J. Agric. Food Chem. 34 (1986) 970e973.
[7] M. Himejima, I. Kubo, J. Agric. Food Chem. 39 (1991) 418e421.
[8] I. Kubo, H. Muroi, M. Himejima, Y. Yamagiwa, Y. Mera, K. Tokushima,
S. Otha, T. Kamikawa, J. Agric. Food Chem. 41 (1993) 1016e1019.
[9] I. Kubo, H. Muroi, A. Kubo, Bioorg. Med. Chem. 7 (1995) 873e880.
[10] H. Muroi, I. Kubo, J. Agric. Food Chem. 41 (1993) 1780e1783.
[11] H. Muroi, I. Kubo, J. Appl. Bacteriol. 80 (1996) 387e394.
[12] I.R. Green, F.E. Tocoli, J. Chem. Res., Synop. (2002) 105e106.
[13] I.R. Green, F.E. Tocoli, J. Chem. Res., Miniprint (2002) 346e375.
[14] I. Kubo, K. Nihei, K. Tsujimoto, J. Agric. Food Chem. 51 (2003)
7624e7628.
[22] N. Masuoka, I. Kubo, Biochim. Biophys. Acta 1688 (2004) 245e249.
[23] T.J. Ha, I. Kubo, J. Agric. Food Chem. 53 (2005) 4350e4354.
[24] I. Kubo, N. Masuoka, T.J. Ha, K. Tsujimoto, Food Chem. 99 (2006)
555e562.
[25] M.T.S. Trevisan, B. Pfundstein, R. Haubner, G. Wurtele, B. Spiegelhalder,
H. Bartsch, R.W. Owen, Food Chem. Toxicol. 44 (2006) 188e197.
[26] K. Nihei, A. Nihei, I. Kubo, J. Agric. Food Chem. 52 (2004) 5011e5020.
[27] S.M. Cohen, K.N. Raymond, Inorg. Chem. 39 (2000) 3624e3631.