Isoxazole analogs of curcuminoids with highly potent multidrug-resistant antimycobacterial activity

Article abstract of DOI:10.1016/j.ejmech.2010.07.003

Curcumin (1), demethoxycurcumin (2) and bisdemethoxycurcumin (3), the curcuminoid constituents of the medicinal plant Curcuma longa L., have been structurally modified to 55 analogs and antimycobacterial activity against Mycobacterium tuberculosis has been evaluated. Among the highly active curcuminoids, the isoxazole analogs are the most active group, with mono-O-methylcurcumin isoxazole (53) being the most active compound (MIC 0.09 μg/mL). It was 1131-fold more active than curcumin (1), the parent compound, and was approximately 18 and 2-fold more active than the standard drugs kanamycin and isoniazid, respectively. Compound 53 also exhibited high activity against the multidrug-resistant M. tuberculosis clinical isolates, with the MICs of 0.195-3.125 μg/mL. The structural requirements for a curcuminoid analog to exhibit antimycobacterial activity are the presence of an isoxazole ring and two unsaturated bonds on the heptyl chain. The presence of a suitable para-alkoxyl group on the aromatic ring which is attached in close proximity to the nitrogen function of the isoxazole ring and a free para-hydroxyl group on another aromatic ring enhances the biological activity.

Full text of DOI:10.1016/j.ejmech.2010.07.003

European Journal of Medicinal Chemistry 45 (2010) 4446e4457
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Isoxazole analogs of curcuminoids with highly potent multidrug-resistant
antimycobacterial activity
,
Chatchawan Changtam a, Poonpilas Hongmanee b, Apichart Suksamrarn a
*
a Department of Chemistry, Faculty of Science, Ramkhamhaeng University, Bangkok 10240, Thailand
b Department of Pathology, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok 10400, Thailand
a r t i c l e i n f o
a b s t r a c t
Article history:
Curcumin (1), demethoxycurcumin (2) and bisdemethoxycurcumin (3), the curcuminoid constituents of
the medicinal plant Curcuma longa L., have been structurally modified to 55 analogs and anti-
mycobacterial activity against Mycobacterium tuberculosis has been evaluated. Among the highly active
curcuminoids, the isoxazole analogs are the most active group, with mono-O-methylcurcumin isoxazole
(53) being the most active compound (MIC 0.09 mg/mL). It was 1131-fold more active than curcumin (1),
the parent compound, and was approximately 18 and 2-fold more active than the standard drugs
Received 19 May 2010
Received in revised form
20 June 2010
Accepted 3 July 2010
Available online 5 August 2010
kanamycin and isoniazid, respectively. Compound 53 also exhibited high activity against the multidrug-
resistant M. tuberculosis clinical isolates, with the MICs of 0.195e3.125 mg/mL. The structural require-
Keywords:
Curcuminoids
ments for a curcuminoid analog to exhibit antimycobacterial activity are the presence of an isoxazole ring
and two unsaturated bonds on the heptyl chain. The presence of a suitable para-alkoxyl group on the
aromatic ring which is attached in close proximity to the nitrogen function of the isoxazole ring and
a free para-hydroxyl group on another aromatic ring enhances the biological activity.
Ó 2010 Elsevier Masson SAS. All rights reserved.
Chemical modification
Isoxazole analogs
Antimycobacterial activity
1. Introduction
major constituents of C. longa and some other Curcuma species. It
has been known for some time that curcuminoids have been used
Tuberculosis is the leading cause of mortality among all
infectious diseases worldwide and is responsible for over two
million deaths annually [1]. The incident of the human immuno-
deficiency virus (HIV) pandemic and the increased prevalence of
multidrug-resistant strains of Mycobacterium tuberculosis caused
this disease to be more complicated [2,3]. The recent increase in
the number of multidrug-resistant clinical isolates of M. tubercu-
losis has created an urgent need for the evolution of new antitu-
berculosis therapeutics [4,5]. Naturally occurring compounds have
demonstrated significant activity in the in vitro assays against
M. tuberculosis [6]. The structural modification of natural products
is one of the potential strategies for the development of new anti-
TB drugs which are different from the drugs currently used. In the
as a natural food additive. The major curcuminoid isolated from
C. longa is curcumin (1). The minor constituents include deme-
thoxycurcumin (2) and bisdemethoxycurcumin (3). Curcuminoids
exhibited many interesting biological activities [7], for example,
antioxidant activity [8,9], anti-inflammatory activity [10,11], anti-
cancer activity [8,12,13], anti-trypanosomal activity [14] and
anti-HIV activity [15]. In the present work, the crude curcuminoids
were separated into compounds 1,
mycobacterial activity of these compounds against the non-viru-
lent M. tuberculosis H37Ra was 100, 50 and 25
g/mL, respectively.
2 and 3 and the anti-
m
These curcuminoids displayed much lower activity than the
standard antitubercular drugs kanamycin, which exhibited MIC of
2.5
mg/mL, and especially isoniazid and rifampin, which showed
search for compounds with anti-TB property,
a
number of
MIC of 0.06 and 0.004 mg/mL, respectively. However, the well
medicinal plants have been investigated by our group. Among the
tested extracts, the curcuminoid fraction of Curcuma longa L. was
shown to exhibit antimycobacterial activity. Curcuminoids are the
known medicinal use of curcuminoids, the relatively less toxicity
of these food ingredients and the unlimited availability of the
compounds from C. longa are of special interest. The unique
skeleton of curcuminoids which is different from the current
anti-TB drugs could also avoid any possible cross resistance of
the curcuminoid-derived compounds with the current anti-TB
drugs. This work deals with the structural modification of the
* Corresponding author. Tel.: þ662 3190931; fax: þ662 3108404.
E-mail addresses: s_apichart@ru.ac.th, asuksamrarn@yahoo.com (A. Suksamrarn).
0223-5234/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.07.003

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
4447
curcuminoids 1e3 to analogs with exceptionally high activity
against the multidrug-resistant (MDR) strains of M. tuberculosis.
to give mono-O-n-propylcurcumin (9) and di-O-n-propylcurcumin
(10) in 25 and 50% yield, respectively. The spectroscopic data of 9
(see Section 5) were in agreement with the structure. The spec-
troscopic data of 10 were identical to those of reported values [14].
Mono-O-(2-hydroxyethyl)curcumin (11), di-O-(2-hydroxyethyl)
curcumin (12), mono-O-allylcurcumin (13) and di-O-allylcurcumin
(14) were prepared by appropriate alkylation of curcumin (1) using
the literature procedures [14]. n-Pentylation of compound 1 was
achieved by reacting 1 with n-propyl iodide in acetone in the
presence of anhydrous potassium carbonate to afford mono-O-n-
pentylcurcumin (15) and di-O-n-pentylcurcumin (16) in 34 and 40%
yields, respectively. The spectroscopic data of 15 and 16 (see Section
5) were consistent with the structures.
2. Chemistry
2.1. Chemical modification of curcuminoids
The natural curcuminoids 1e3 obtained from C. longa were
subjected to chemical modifications for antimycobacterial
evaluations.
2.1.1. Demethylated analogs
The parent curcuminoids 1 and 2 were demethylated as
described previously [14,16] to yield the corresponding demethy-
lated analogs 4, 5 and 6 (Scheme 1).
2.1.3. Acetate analogs
In order to have both the mono and diacetate derivatives for
biological evaluation, the curcuminoids 1e3 were subjected to
partial acetylation as shown in Scheme 2. Reaction of 1 with acetic
anhydride in pyridineeCHCl3 afforded the monoacetate 17 and
diacetate 18 in 25 and 70% yields. The spectroscopic data of 17
(Section 5) were in agreement with the structure, and those of 18
were consistent with the literature values [18]. Compounds 2 and 3
were similarly subjected to partial acetylation to give the corre-
sponding mono and diacetate analogs 19, 20, 21, 22 and 23. The
spectroscopic data of these compounds (Section 5) were consistent
with the structures and were in agreement with the reported data
[14,18,19]. The isomeric monoacetates 19 and 20 were distin-
guished by 1H NMR spectral data. Thus, the presence of acetate
group at the 40-position was evident from the downfield shift of the
2.1.2. Methyl ether and higher alkyl ether analogs
Alkylation of the curcuminoid 1 is outlined in Scheme 1.
Methylation of 1 was achieved by modification of the literature
procedure [14,17] to afford mono-O-methylcurcumin (7) and di-O-
methylcurcumin (8) in 39 and 47%, respectively. The spectroscopic
(IR, 1H NMR and mass spectra) data of compounds 7 and 8 were
consistent with the literature values [17]. Compound 1 was sub-
jected to n-propylation by modification of the reported procedure
two aromatic protons, H-30 and H-50, at
d 7.11. This was different
1
from the H NMR spectrum of compound 20 that the H-30 and 50
signals appeared at
d d 7.02.
6.81, and H-500 at
2.1.4. Dihydro, tetrahydro, hexahydro and octahydro analogs
Dihydrocurcumin (24) was prepared in 30% yield by zinceacetic
acid reduction of the parent curcuminoid 1. The structure of the
product 24 was confirmed by 1H NMR spectral data. The differences
Scheme 1. Demethylation and alkylation of curcuminoids 1e3. Reagents and condi-
tions: (a) BBr3, CH2Cl2, 0 ꢀC, then ambient temp.; (b) MeI, K2CO3, acetone, reflux; (c) n-
propyl iodide, K2CO3, acetone, reflux; (d) 2-bromoethanol, K2CO3, acetone, reflux; (e)
allyl bromide, K2CO3, acetone, reflux; (f) n-pentyl iodide, K2CO3, acetone, reflux.
Scheme 2. Acetylation of curcuminoids 1e3. Reagents and condition: (g) Ac2O, pyri-
dine-CHCl3 (1:1), ambient temp.

4448
C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
from the starting curcuminoid 1 being the absence of two olefinic
protons of H-6 and H-7 at ca. 6.47 and 7.57, and the presence of
two methylene signals at 2.64 and 2.87. Tetrahydrocurcumin (25),
hexahydrocurcumin (26) and octahydrocurcumin (27) were
prepared by catalytic hydrogenation of 1 according to the literature
procedure [20,21] (Scheme 3).
2.1.8. Isoxazole analogs of dihydro and tetrahydrocurcumins
d
The isoxazole analog 40 was prepared in 63% yield by treatment
of dihydrocurcumin (24) with hydroxylamine hydrochloride in
pyridine (Scheme 5). The structure of the product was established
as 40, not the isomeric structure 41, by the spectroscopic evidence.
Thus, the 1H NMR spectrum showed a four-proton broad singlet
d
signal of methylene group at
olefinic protons at
6.73 (H-10) and 7.19 (H-20). The evidence was
supported by the 2D NMR experiments in which the HMBC corre-
lations between H-20 and C-3, C-2000, C-6000; H-200 and C-5, C-20000
d
2.92 (H-100 and H-200), and two
2.1.5. Unsaturated and saturated mono-keto analogs
The enone 28, the dienones 29 and 30, the trienone 31 and the
unconjugated ketone 32 were prepared by the literature methods
[14,22,23] (Scheme 4).
d
,
C-60000 were observed. The isoxazole analog 42 was similarly
prepared in 77% yield from tetrahydrocurcumin (25) (Scheme 5).
Thus, treatment of 25 in pyridine with hydroxylamine hydrochlo-
ride to give the intermediate monoxime, which was subsequently
treated with p-toluenesulfonic acid to yield 42 in 54% overall yield
from 25. The spectroscopic data of 42 were different from those of
2.1.6. Pyrazole analogs
The pyrazole analogues of the curcuminoids 1e3 were
prepared as outlined in Scheme 5. Using the literature procedure
[24] with a slight modification, curcumin pyrazole (33) was
obtained in 71% yield by treatment of 1 with hydrazine hydrate in
AcOH at 50 ꢀC. Reaction of 2 and 3 with the same reagent afforded
the pyrazole analogs 34 and 35 in 77% and 41% yields, respectively.
The spectroscopic data of these compounds were consistent with
the structures and were in agreement with the reported data
[20,24]. The N-phenyl substituted pyrazole 36 was similarly
prepared in 81% yield by reaction of compound 1 with phenyl-
hydrazine hydrate. The spectroscopic data (Section 5) were
consistent with the structure and were in agreement with the
reported data [24].
compound 40 by the absence of olefinic signal at
7.19 (H-20) and the presence of additional two methylene signals at
2.87e2.96 (m, 8H). The mass spectral data were also consistent
d
6.73 (H-10) and
d
with its structure.
2.1.9. Alkyl ethers and methyl ethers of isoxazole analogs
Antimycobacterial evaluation of the foregoing curcuminoid 1
and its mono-O-n-pentyl ether analog 15 indicated that introduc-
tion of an n-pentyl group resulted in 8-fold increase in activity
(Table 1). In order to see whether the n-pentyl group would
enhance the activity in the case of the isoxazole analog 37, the
mono-O-n-pentyl ether analog 43 and the isomeric isoxazole
analog 44 (1:1 mixture), were prepared in 24%, together with the
di-O-n-pentyl ether analog 45 in 62% (Scheme 6). The 43/44
mixture was separated by HPLC (see detail in Section 5). The
structures of the isomeric pentyl ether analogs 43 and 44 were
distinguished by 1H and 13C NMR data (see Section 5). The present
of the pentyl group was evident from the signals of the methyl
2.1.7. Isoxazole analogs
The isoxazole analog 37 was prepared in 70% yield by treatment
of 1 with hydroxylamine hydrochloride in pyridine (Scheme 5). The
spectroscopic data of 37 (see Section 5) were consistent with the
structure and were in agreement with those reported previously
[24]. Treatment of 2 with the same reagent afforded a 1:1 mixture
(65%) of two isomeric isoxazoles 38a and 38b, which could not be
separated by column chromatography (Scheme 5). However,
separation of the mixture was achieved by normal phase HPLC (see
Section 5) to give pure 38a and 38b. The structures of these iso-
xazoles were distinguished by 1H and 13C NMR analysis. Data
supporting this interpretation were provided by 2D NMR experi-
ments in which the HMBC correlations between H-20 and C-3, C-2000,
C-6000; H-200 and C-5, C-20000, C-60000 were observed. The isoxazole
analog 39 was similarly prepared in 68% yield from the curcumi-
noid 3. The spectroscopic data (Section 5) were consistent with the
structure.
group at
d
0.91 (t, J ¼ 6.8 Hz, H-500000), two methylene groups at
d 1.40
(m, H-300000 and H-400000), one methylene group at
d
1.84 (m, H-200000),
and another one methylene group at
d
4.02 (t, J ¼ 6.4 Hz, H-100000).
13
The C NMR data of C-500000, C-200000, C-300000, C-400000 and C-100000
appeared at
d 13.9, 22.4, 28.0, 28.8 and 69.0, respectively. The
evident was supported by 2D NMR experiments in which the HMBC
correlations between H-20 and C-3, C-2000, C-6000; H-200 and C-5, C-20000
,
C-60000; H-100000 and C-300000, C-3000 were observed. Although compound
43 was very active, this compound encountered solubility problem
during the dilution of sample solution for biological evaluation. The
Scheme 3. Reduction of curcumin (1). Reagents and condition: (h) Zinc powder, AcOH, ambient temp.; (i) H2/PdeC, EtOH.

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
4449
Scheme 4. Dehydration of hexahydrocurcumin (25), and dehydrogenation and catalytic hydrogenation of 28. Reagents and conditions: (j) p-TsOH, C6H6, reflux; (k) DDQ, THF; (l) H2/
PdeC, EtOH.
relatively more soluble and lower alkyl ether analogs, the mono-O-
(3,3-dimethylallyl) ether analogs 46a/46b (2:1 mixture) and the di-
O-(3,3-dimethylallyl) ether analogs 47, the mono-O-ethyl ether
analogs 48a/48b (5:2 mixture) and the di-O-ethyl ether analogs 49,
and the acetate analogs 50, 51 and 52, were prepared (Scheme 6).
The structures of the isomeric mono-O-(3,3-dimethylallyl) ether
analogs 46a/46b, the mono-O-ethyl ether analogs 48a/48b, and the
monoacetates 50 and 51 were distinguished by 1H and 13C NMR
spectral data in similar manner to those of 43 and 44. After the
assay results were obtained, the methyl ether analogs 53, 54, 55,
56a/56b (1:1 mixture) and 57 were also prepared (Scheme 6). The
structures of the isomeric methyl ether analogs 54 and 55, and 56a
and 56b were distinguished by 1H and 13C NMR data in similar
manner to those of 43 and 44 (see Section 5). The mono-deme-
thylated analogs 58a/58b (1:2 mixture) and the di-demethylated
analog 59 were also prepared (Scheme 5) and structural charac-
terization was achieved by similar spectroscopic spectral analysis.
activity, we therefore synthesized the higher alkyl ether analogs of
the parent curcuminoid 1 starting with the mono-O-n-propyl ether
9. As expected, the antimycobacterial activity of 9 was higher than
the parent compound 1; it was approximately 4-fold more active
than compound 1. However, the di-O-n-propyl analog 10 was very
weakly active (MIC 200 mg/mL). The relatively nonpolar nature of
the alkyl group is required for high activity of the analogs. This was
evident from the relatively low activity of the mono-O-(2-hydrox-
yethyl) analog (11) (MIC 200 mg/mL), which was more polar than
the n-propyl ether analog 9. The di-O-(2-hydroxyethyl) analog (12)
also showed low activity. The activity of the ether analogs was
sensitive to the nature of the alkyl group, as seen from the relatively
lower activity of the mono-O-allyl ether 13 and the di-O-allyl ether
14, the unsaturated ether analog of 9 and 10, than the analogs 9 and
10, respectively. To see whether saturated higher alkyl ether
analogs contributed to high activity of the curcuminoid, the mono
and di-O-n-pentyl ether analogs 15 and 16 were synthesized and
assessed for antimycobacterial activity and it was found that
3. Results and discussion
compounds 15 and 16 (MIC 12.5 and 100
than the corresponding n-propyl ether analogs 9 and 10 (MIC 25
and 200 g/mL), respectively. At this point, it is concluded that in
going from the mono alkyl analogs to dialkyl analogs, a sharp
decrease in activity was observed.
mg/mL) were more active
The antimycobacterial activity of the parent curcuminoids 1e3
against the non-virulent M. tuberculosis H37Ra was 100, 50 and
25 mg/mL, respectively (Table 1). The first type of analogs selected
m
for antimycobacterial evaluation was the demethylated analogs.
However, the mono-O-demethylated analog 4 and the di-O-
demethylated analog 5 were only as active as, or even less active
than, the parent compound 1 (see Table 1). The demethylated
analog 6 was much less active than its parent compound 2. The
results have indicated that increase in polarity of the curcuminoids
caused decrease in activity. We therefore moved to analogs which
were more lipophilic than their respective parent compounds. The-
mono-O-methyl analogs 7 was 4-fold more active than the parent
compound 1. However, further methylation to the di-O-methyl
We next evaluated the potency of the acetate analogs of cur-
cuminoids. The results have indicated that the monoacetates 17,
19, 20 and 22 exhibited comparable activity to that of their
respective parent compounds 1, 2, and 3. As expected, the corre-
sponding diacetate analogs 18 and 21 showed lower activity than
the monoacetate analogs 17, 19 and 20. The exception was for the
analog 23, which showed comparable activity to that of the parent
compound 3.
The chemical modification of the curcuminoids 1e3 to the
above corresponding ether and acetate derivatives did not seem to
give promising analogs with high antimycobacterial activity. We
therefore chose to modify the skeleton of the curcuminoids. The
reduced analogs of curcuminoid 1, viz. 24, 25, 26 and 27, were
analog 8 resulted in decrease in activity (MIC 100
compared with the mono-O-methyl analog 7 (MIC 50
m
m
g/mL) when
g/mL). Since
it seemed that increase in lipophilicity resulted in increase in

4450
C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
Scheme 5. Preparation of pyrazole and isoxazole analogs. Reagents and conditions: (m) NH2NH2, hydrate, AcOH; (n) phenylhydrazine hydrate, AcOH; (o) NH2OH.HCl, pyridine,
50 ꢀC; (p) (1) NH2OH.HCl, pyridine, ambient temp., (2) p-TsOH, C6H6, 60 ꢀC.
prepared and assessed for antimycobacterial activity. The assay
results indicated that, except for the tetrahydro analog 25 that was
approximately 2-fold more active than the parent compound 1,
the activity of the rest were approximately the same as (in the
case of the hexahydro analog 26) or less active (in the case of the
dihydro analog 24 and the octahydro analog 27) than the parent
compound 1.
We then explored the biological activity of the mono-keto
analogs of the parent compound 1. The enone 28, the dienones 29
and 30, the trienone 31 and the unconjugated ketone 32 were
assessed for antimycobacterial activity. Except for the ketone 32
which exhibited comparable activity to that of the parent curcu-
minoid 1, the rest showed 4-fold (compounds 28 and 30) and 8-fold
(compounds 29 and 31) more active than the parent compound 1.
Although the above conjugated mono-keto analogs exhibited
much improved antimycobacterial activity than that of the parent
compound 1, it was still less active than the standard drug kana-
mycin. We therefore changed our strategy to analogs with more
rigid heptyl chain. In order to keep the heptyl chain less flexible,
MICs of these compounds were 200, 100, and 25 mg/mL, respec-
tively. The N-substituted analog 36 was also prepared, but it was
only as active as compound 35.
We then turned our attention to the five-membered cyclic NeO
analogs, the isoxazole analogs. Curcumin isoxazole (37) exhibited
interesting biological activities, for example, anti-inflammatory
activity [25] and anticancer activity [26,27]. Potent anti-
mycobacterial compounds with the isoxazole functionality in the
molecules have also been reported [28e30]. In the present work,
the isoxazole analogs 37, 38a/38b (1:1 mixture), and 39 were
prepared from the parent compounds 1, 2, and 3, respectively, and
subjected to antimycobacterial evaluation. The results indicated
that the isoxazole analog 37 was highly active; its MIC was 1.56 mg/
mL, which was about 64-fold more active than the parent curcu-
minoid 1. The isoxazole analogs 38a/38b and 39 were less active;
their MICs were both 12.5 mg/mL. It was worth noting that the 3,4-
dioxygenated groups on both aromatic rings are required for an
isoxazole analog to exhibit high antimycobacterial activity. The
isoxazole 37 was thus selected as the lead compound.
a
cyclic structure was introduced into the chain. The five-
In order to see the influence of unsaturation at the alkyl chain,
the corresponding dihydro analog 40 and the tetrahydro analog 42
were prepared and their MICs were determined. The gradual
decrease in activity in going from the fully unsaturated analog 37
membered cyclic dinitrogen analog, the pyrazole analog, was first
chosen and the analogs 33, 34 and 35 were prepared from the
curcuminoids 1, 2 and 3, respectively. The antimycobacterial
activity of these cyclic analogs, however, was not promising; the
(MIC 1.56 mg/mL) to the dihydro analog 40 (MIC 6.25 mg/mL) and

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
4451
Table 1
The in vitro activity of the curcuminoid analogs against M. tuberculosis H37Ra strain.
Compound
MIC (
mg/mL)
Compound
MIC (mg/mL)
1
2
3
4
5
6
7
8
100
50
25
200
100
200
50
100
25
200
200
200
100
32
33
34
35
36
37
100
200
100
25
25
1.56
12.5
12.5
6.25
NDb
100
0.39
6.25
38a/38b (1:1)
39
40
41
42
43
44
45
46a/46b (2:1)
47
48a/48b (5:2)
49
50
51
52
53
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Inactivea
12.5
100
100
200
50
50
100
25
25
200
50
100
Inactivea
25
12.5
25
Inactivea
1.56
NDb
12.5
NDb
1.56
1.56
1.56
0.09
0.78
0.39
100
54
55
56a/56b (1:1)
57
58
Inactivea
6.25
200
2.5
0.06
0.004
59
Kanamycin
Isoniazid
Rifampin
12.5
a
Inactive at MIC > 200 mg/mL.
ND, not determined.
b
finally to the tetrahydro analog 42 (MIC 100
mg/mL) has led to
a conclusion that full conjugated system of the C7 chain is required
for high antimycobacterial activity of the isoxazole analogs.
Sincethemono-O-n-pentylanalog15 was8-foldmoreactivethan
the parent compound 1, the lead compound 37 was transformed to
the corresponding mono-O-n-pentyl analogs 43 and its isomeric
analog 44, and the di-O-n-pentyl ether analog 45. It was found that
the analog 43 was highly active (MIC 0.39 mg/mL) which was about
256-fold more active than the parent compound 1. However, its
isomeric analog 4 was less active than the analog 43 (MIC of
compound 44 was 6.25 mg/mL). As expected, compound 45 was
inactive to the test. The low activity of 45 was possibly, at least partly,
due to its low solubility. Since compound 43 (and 44) has some
solubility problem upon dilution with water, which was due to the
lipophilic nature of the n-pentyl group, we then further explored
other alkyl group that exhibited comparable activity to that of
compound 43, butwithhighersolubilitythanthatofcompound43. A
number of relatively less polar alkyl ether analogs, including the
mono-O-(2,2-dimethylallyl) ether analogs 46a/46b (2:1 mixture)
and the mono-O-ethyl ether analogs 48a/48b, and the acetate
analogs 50, 51 and 52, were prepared and assessed for anti-
mycobacterial activity. However, none of them was as active as the
analog43 (seeTable 1). Thedi-O-(3,3-dimethylallyl) etheranalogs 47
and the di-O-ethyl ether analogs 49 have not been subjected to
biological evaluation. We eventually discovered that the mono-O-
methyl analog 53 exhibited the highest activity; its the MIC was
Scheme 6. Alkylation and acetylation of isoxazole analogs 37, 38a, 38b and 39.
Reagents and conditions: (q) n-pentyl iodide, K2CO3, acetone, reflux; (r) 3,3-dime-
thylallyl bromide, K2CO3, acetone, reflux; (s) diethyl sulfate, K2CO3, acetone, reflux; (t)
Ac2O, pyridine; (u) MeI, K2CO3, acetone, reflux.
compound 53 with that of compound 54 and the activity of
compound 43 with that of 44 have led to a conclusion that one of the
structural requirements for a mono-O-alkylated analog of isoxazole
to exhibit high antimycobacterial activity is that the alkoxyl group is
located at the aromatic ring which is closer to the nitrogen function
than the oxygen function of the isoxazole ring. It should be noted
that, despite it being a di-O-methyl ether analog which was much
less active than the mono-O-methyl ether analog, the isoxazole
0.09
mg/mL or 0.24
mM, that is 1131-fold more active than the parent
analog52still exhibited highactivity, withMICof0.39 mg/mL, though
curcuminoid 1 (MIC 100
of the isoxazole 53 with those of the standard drugs, it was evident
that compound 53 wasapproximately18 and2-fold moreactive than
m
g/mL or 271.45 M). Comparisonof theMIC
m
it was less active than its mono-O-methyl ether analog. In contrast,
the activity of 56a/56b and 57, the fully methylated analogs of the
parent compounds 2 and 3, was very low (see Table 1). The findings
have pointed out that the number of oxygenation on the aromatic
rings of curcuminoids, the number and nature of the substituents on
the oxygen function of the aromatic rings are important for high
kanamycin sulfate (MIC 2.50
0.06 g/mL or 0.44 M), respectively. Its isomeric analog 54 was less
active, the MIC of which is 0.78 g/mL. Comparison of the activity of
mg/mL or 4.29 mM) and isoniazid (MIC
m
m
m

4452
C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
Table 2
5. Experimental
The in vitro activity of the isoxazole analogs 43, 44, 53 and 54 against multidrug-
resistant M. tuberculosis compared with H37Ra strain.
5.1. General
Entry Code
Isolate resistant profilea
MIC (
mg/mL)
Melting points were determined on an Electrothermal melting
43
44
53
54
0.78
0.78
6.25
point apparatus and are uncorrected. IR spectra were recorded in
KBr on a PerkineElmer FT-IR Spectrum BX spectrophotometer. 1H
and 13C NMR spectra were recorded on a Bruker AVANCE 400
spectrometer operating at 400 and 100 MHz, respectively. Electron
impact (EI) and electrospray (ES) mass spectra were obtained using
a Finnigan Polaris Q and a Finnigan LC-Q mass spectrometer. High
resolution mass spectra were obtained using a Bruker micrOTOF
mass spectrometer. Column chromatography and TLC were carried
out using Merck silica gel 60 (<0.063 mm) and precoated silica gel
60 F254 plates, respectively. Spots on TLC were detected under UV
light (254 nm) and by spraying with anisaldehydeeH2SO4 reagent
followed by heating.
1
2
3
4
5
6
7
8
H37Ra Pan sensitive
0.39
0.39
1.56
1.56
0.78
3.125 6.25
3.125 6.25
1.25
6.25
1.56
1.56
1.56
0.09
0.195
1.56
3.125 12.5
1.56
3.125 12.5
3.125 12.5
1.56
M3
M4
M5
M6
INH, RIF, SM
INH, RIF, EMB, SM
INH, RIF, SM
INH, RIF
3.125 0.39
M8
INH, RIF, EMB, SM
INH, RIF
INH, RIF, EMB
INH, RIF, EMB, SM
INH, RIF
M11
M16
M21
M22
M27
M46
M48
M53
1.56
12.5
9
3.125 6.25
0.78
0.39
3.125 12.5
10
11
12
13
14
6.25
1.56
1.56
0.39
0.195
1.56
1.56
0.78
6.25
INH, RIF
INH, RIF, EMB, SM, OFX, CIP 1.56
INH, RIF, EMB, SM, OFX, CIP 3.125 3.125 1.56
INH, RIF, EMB, SM, OFX, CIP 3.125 3.125 3.125 12.5
12.5
a
INH, isoniazid; RIF, rifampin; EMB, ethambutol; SM, streptomycin; OFX, oflox-
acin; CIP, ciprofloxacin.
5.2. Chemical modifications of curcuminoids
antimycobacterial activity. It should also be noted that, like the cur-
cuminoidanalogs, demethylationof theisoxazoleanalogsresulted in
decrease in activity. This was exemplified by the low activity of the
The three natural curcuminoids, curcumin (1), demethox-
ycurcumin (2) and bisdemethoxycurcumin (3), were obtained as
described previously [14].
isoxazoles 58a/58b and 59 (MICs 6.25 and 200 mg/mL, respectively).
The two most potent analogs, 43 and 53, together with their
isomers, 44 and 54, were assessed for various multidrug-resistant
(MDR) clinical isolates M. tuberculosis (Table 2, entries 2e14). For
the isoniazid (INH)- and rifampin (RIF)-resistant isolates (entries 5,
7, 10 and 11), the most active analog 53 were still very active (MICs
5.2.1. Demethylation of compounds 1 and 2
Compound 1 was demethylated in the same manner described
previously [14,16] to yield mono-O-demethylcurcumin (4) (42%)
and di-O-demethylcurcumin (5) (33%). Compound 2 was also sub-
jected to demethylation in similar manner to that of compound 1 to
give O-demethyldemethoxycurcumin (6) (64%). The spectroscopic
(IR, 1H NMR and mass spectra) data of these analogs were consis-
tent with the reported values [14,16].
0.195e3.125
mg/mL) whereas the analog 43 were also very active
(MICs 0.39e3.125
mg/mL). For the INH-, RIF-, and streptomycin
(SM)-resistant isolates (entries 2 and 4), the analog 53 also
exhibited comparable activity against those of the INH- and RIF-
resistant isolates (MICs 0.195 and 3.125
compound 43 showed similar activity (MICs 0.39 and 1.56
m
g/mL) whereas
5.2.2. Methylation of curcumin (1)
mg/mL).
Compound 1 was subjected to methylation by modification of
the literature procedure [14,17]. Thus, compound 1 (100 mg,
0.27 mmol) was dissolved in dry acetone (3 mL) and anhydrous
K2CO3 (100 mg) and MeI (0.8 mL) were added. The reaction mixture
was refluxed for 3 h; water was added and the mixture was
extracted with CH2Cl2 (100 mL ꢁ 2). The combined organic phase
was washed with H2O, dried over anhydrous Na2SO4 and the
solvent was removed under vacuum. The products were separated
by column chromatography using CH2Cl2 as eluent to yield curcu-
min mono-O-methyl ether (7) (40 mg, 39%) and curcumin di-O-
methyl ether (8) (50 mg, 47%). The spectroscopic (IR, 1H NMR, and
mass spectra) data were consistent with the reported values [17].
For the INH-, RIF-, and ethambutol (EMB)-resistant isolates (entry
8), and INH-, RIF-, EMB- and SM-resistant isolates (entries 3, 6 and
9), both the analogs 43 and 53 showed MICs in the range
1.56e3.125 mg/mL. More interestingly, the isolates that also resis-
tant to the second-line drugs ofloxacin (OFX) and ciprofloxacin
(CIP) (entries 12e14) were also sensitive to compounds 43 and 53
(MICs 1.56e3.125
active than its isomer 43 (MICs 1.56e6.25
exhibited varying activity against different MDR isolates (MICs
0.78e12.5 g/mL). The isoxazole analog 53 is, therefore, the potent
mg/mL). As expected, compound 44 was less
mg/mL). Compound 54
m
structure lead for the development of MDR antitubercular agents.
4. Conclusion
5.2.3. Propylation of curcumin (1)
Compound 1 (120 mg, 0.32 mmol) was dissolved in dry acetone
(2 mL) and anhydrous K2CO3 (40 mg) and n-propyl iodide (0.6 mL)
were added. The reaction mixture was reflux for 5 h, water was
added and the mixture was extracted with CH2Cl2. The combined
organic phase was washed with H2O, dried over anhydrous Na2SO4
and the solvent was removed under vacuum. The crude products
were isolated and purified by column chromatography using
CH2Cl2 to yield mono-O-n-propylcurcumin (9) (34 mg, 25%) and di-
O-n-propylcurcumin (10) (72 mg, 50%). The spectroscopic data of
compound 10 were consistent with the reported values [14].
Structural modifications of the natural curcuminoids, curcumin
(1), demethoxycurcumin (2) and bisdemethoxycurcumin (3), to 55
analogs have been undertaken and antimycobacterial activity
against M. tuberculosis has been evaluated. The isoxazole analogs
are the most active class of analogs, with mono-O-methylcurcumin
isoxazole (53) being the most active compound (MIC 0.09 mg/mL). It
was 1131-fold more active than the parent compound 1. Compound
53 also exhibited high activity against the multidrug-resistant
M. tuberculosis (MDR-TB) strains, with the MICs of 0.195e3.125 mg/
mL. The structural requirements for a curcuminoid analog to
exhibit antimycobacterial activity are the presence of an isoxazole
ring and two unsaturated bonds on the heptyl chain. The presence
of a suitable para-alkoxyl group on the aromatic ring which is
attached in close proximity to the nitrogen function of the isoxazole
ring and a free para-hydroxyl group on another aromatic ring
enhances the biological activity.
5.2.3.1. Mono-O-n-propylcurcumin (9). IR nmax: 3398, 2965, 2935,
1618, 1563, 1508, 1258, 1133, 1030, 968, 840 cmꢂ1 1H NMR
;
(400 MHz, CDCl3)
d
1.03 (t, J ¼ 7.3 Hz, 3H, H-3000), 1.86 (sextet,
J ¼ 7.1 Hz, 2H, H-2000), 3.89 and 3.93 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 4.00
(t, J ¼ 6.8 Hz, 2H, H-1000), 5.78 (s, 1H, H-4), 5.83 (br s, 1H, OH), 6.45
and 6.46 (each d, J ¼ 15.7 Hz, 2 ꢁ 1H, H-2 and H-6), 6.85 and 6.91

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
4453
(each d, J ¼ 8.2 Hz, 2 ꢁ 1H, H-50 and H-500), 7.03 and 7.06 (each br d,
J ¼ 1.4 Hz, 2 ꢁ 1H, H-20 and H-200), 7.15 (br d, J ¼ 8.2 Hz, 2 ꢁ 1H, H-60
and H-600), 7.57 and 7.58 (each d, J ¼ 15.7 Hz, 2 ꢁ 1H, H-1 and H-7);
ESMS (þve): m/z (% rel. abund.) 411 [M þ H]þ (100).
(19), mono-O00-acetyldemethoxycurcumin (20) and di-O-acetylde-
methoxycurcumin (21) in 28%, 21% and 33% yield, respectively. The
spectroscopic data of compound 21 were consistent with the
reported values [14].
5.2.4. 2-Hydroxyethylation and allylation of curcumin (1)
5.2.6.2. Mono-O0-acetyldemethoxycurcumin (19). Orange amor-
phous solid; m.p. 127e129 ꢀC; IR nmax: 3757, 3758, 1762, 1750, 1718,
1685, 1654, 1627, 1560, 1542, 1508, 1458, 1429, 1207, 1137, 968,
Compound 1 was subjected to 2-hydroxyethylation as described
previously [14] to afford mono-O-(2-hydroxyethyl)curcumin (11)
anddi-O-(2-hydroxyethyl)curcumin (12)in48and35%. Compound 1
was similarly subjected to allylation [14] to afford mono-O-allylcur-
cumin (13) and di-O-allylcurcumin (14) in 33 and 46%, respectively.
Thespectroscopicdatawereconsistentwiththereportedvalues[14].
838 cmꢂ1; 1H NMR (400 MHz, CDCl3)
d 2.29 (s, 3H, OAc), 3.92 (s, 3H,
OMe), 5.80 (s, 1H, H-4) and 5.85 (br s, 1H, OH), 6.47 (d, J ¼ 15.7 Hz,
1H, H-6), 6.54 (d, J ¼ 15.7, 1H, H-2), 6.91 (d, J ¼ 8.2 Hz, 1H, H-500),
7.03 (br s, 1H, H-200), 7.11 (d, J ¼ 8.5 Hz, 2H, H-30 and H-50), 7.11
(obscured signal, 1H, H-600), 7.55 (d, J ¼ 8.5 Hz, 2H, H-20 and H-60),
7.59 (d, J ¼ 15.7 Hz, 1H, H-7), 7.61 (d, J ¼ 15.8 Hz, 1H, H-1); ESMS
(þve) m/z (% rel. abund.) 403 [M þ Na]þ (29), 381 [M þ H]þ (100).
5.2.5. Pentylation of curcumin (1)
Compound 1 was subjected to pentylation in similar manner to
that of methylation of compound 1, but using n-pentyl iodide in
place of methyl iodide to give mono-O-n-pentylcurcumin (15) and
di-O-n-pentylcurcumin (16) in 34 and 40% yields, respectively.
5.2.6.3. Mono-O00-acetyldemethoxycurcumin (20). Yellow amor-
phous solid; m.p. 166e168 ꢀC; IR nmax: 3448, 1757, 1732, 1718, 1701,
1654, 1627, 1602, 1576, 1559, 1508, 1458, 1599, 1199, 1169, 1145,
1032, 961, 831 cmꢂ1; 1H NMR (400 MHz, CDCl3 þ 2 drops of CD3OD)
5.2.5.1. Mono-O-n-pentylcurcumin (15). IR nmax: 3420, 2934, 2869,
1625, 1582, 1511, 1465, 1424, 1262, 1136, 1032, 967, 847, 811 cmꢂ1
;
d 2.30 (s, 3H, OAc), 3.85 (s, 3H, OMe), 5.78 (s, 1H, H-4), 6.45 and 6.51
1H NMR (400 MHz, CDCl3)
d
0.91 (t, J ¼ 7.1 Hz, 3H, H-5000), 1.40 (m,
(each d, J ¼ 15.8 Hz, 2H, H-2 and H-6), 6.81 (d, J ¼ 8.5 Hz, 2H, H-30
and H-50), 7.02 (d, J ¼ 8.1 Hz,1H, H-500), 7.09 (br s,1H, H-200), 7.12 (dd,
J ¼ 8.1, 1.4 Hz, 1H, H-600), 7.42 (d, J ¼ 8.5 Hz, 2H, H-20 and H-60), 7.55
and 7.59 (each d, J ¼ 15.8 Hz, 2H, H-1 and H-7); ESMS (þve) m/z (%
rel. abund.), 403 [M þ Na]þ (27), 381 [M þ H]þ (100).
2 ꢁ 2H, H-3000 and H-4000), 1.84 (br q, 2H, H-2000), 3.89 and 3.93 (each s,
2 ꢁ 3H, 30-OMe and 300-OMe), 4.03 (t, J ¼ 6.8 Hz, 2H, H-1000), 5.79 (s,
1H, H-4), 5.84 (br s, 1H, 400-OH), 6.46 and 6.47 (each d, J ¼ 15.8 Hz,
2 ꢁ 1H, H-2 and H-6), 6.85 (d, J ¼ 8.2, 1H, H-50), 6.92 (d, J ¼ 8.2 Hz,
1H, H-500), 7.03 and 7.06 (each br s, 2 ꢁ 1H, H-20 and H-200), 7.10 (br d,
J ¼ 8.2 Hz, 2 ꢁ 1H, H-60 and H-600), 7.57 and 7.58 (d, J ¼ 15.8 Hz,
2 ꢁ 1H, H-1 and H-7); EIMS: m/z (% rel. abund.) 438 [M]þ (10), 420
(100), 350 (29), 349 (22), 191 (36), 190 (51), 177 (38).
Compound 3 was subjected to acetylation in the same manner to
that of compound 1 to give mono-O-acetylbisdemethoxycurcumin
(22) and di-O-acetylbisdemethoxycurcumin (23) in 25% and 60%
yield, respectively. The spectroscopic (IR, H NMR and mass spectra)
data of compound 23 were consistent with the reported values [19].
5.2.5.2. Di-O-n-pentylcurcumin (16). IR nmax: 2958, 2940, 2857,
1625, 1581, 1511, 1466, 1420, 1337, 1256, 1227, 1133, 1026, 969, 845,
5.2.6.4. Mono-O-acetylbisdemethoxycurcumin (22). Yellow powder,
m.p. 178e180 ꢀC; IR nmax: 3872, 3448, 1735, 1718, 1685, 1654, 1637,
793 cmꢂ1; 1H NMR (400 MHz, CDCl3)
d
0.91 (t, J ¼ 7.0 Hz, 2 ꢁ 3H, H-
5000 and H-50000), 1.40 (m, 4 ꢁ 2H, H-3000, H-4000, H-30000 and H-40000), 1.84
(br q, 2 ꢁ 2H, H-2000 and H-20000), 3.89 (s, 2 ꢁ 3H, 30-OMe and 300-
OMe), 4.03 (t, J ¼ 6.8 Hz, 2 ꢁ 2H, H-1000 and H-10000), 5.79 (s, 1H, H-4),
6.46 (d, J ¼ 15.8 Hz, 2 ꢁ 1H, H-2 and H-6), 6.85 (d, J ¼ 8.2 Hz, 2 ꢁ 1H,
H-50 and H-500), 7.05 (br s, 2 ꢁ 1H, H-20 and H-200), 7.09 (br d,
J ¼ 8.2 Hz, 2 ꢁ 1H, H-60 and H-600), 7.58 (d, J ¼ 15.8 Hz, 2 ꢁ 1H, H-1
and H-7); EIMS: m/z (% rel. abund.) 508 [M]þ (5), 420 (16), 344
(100), 234 (77), 220 (25), 177 (17).
1604, 1508, 1458, 1374, 1235, 1169, 972, 833 cmꢂ1 1H NMR
;
(400 MHz, CDCl3) d 2.30 (s, 3H, OAc), 5.78 (s, 1H, H-4), 6.48 and 6.54
(each d, J ¼ 15.8 Hz, 2 ꢁ 1H, H-2 and H-6), 6.83 (d, J ¼ 8.3 Hz, 2H, H-
300 and H-500), 7.11 (d, J ¼ 8.3 Hz, 2H, H-30 and H-50), 7.45 (d,
J ¼ 8.3 Hz, 2H, H-200 and H-600), 7.54 (d, J ¼ 8.3 Hz, 2H, H-20 and H-60),
7.60 (d, J ¼ 15.8 Hz, 2H, H-1 and H-7); ESMS (þve) m/z (% rel.
abund.) 723 [2M þ Na]þ (83), 373 [M þ Na]þ (8), 351 [M þ H]þ (56).
5.2.7. Reduction of curcumin (1) with zinceacetic acid
5.2.6. Acetylation of curcuminoids 1e3
To a solution of curcumin (1) (100 mg, 0.27 mmol) in acetic acid
(3 ml) was added Zn dust (20 mg) and the mixture was stirred at
ambient temperature for 4 h. The reaction was worked up with H2O
and the solution was extracted with EtOAc. The organic phase was
washed with H2O, dried over anhydrous Na2SO4 and the solvent
was evaporated to dryness. The crude products were purified by
column chromatography using n-hexaneeEtOAc (3:2) to give
dihydrocurcumin (24) (30 mg, 30%).
Ac2O (0.5 mL) was added to a solution of compound 1 (100 mg,
0.27 mmol) in pyridineeCHCl3 (1:1, 2 ml) and the reaction mixture
was stirred at ambient temperature for 2 h. After the usual work up,
the mixture was extracted with CH2Cl2 and the organic phase was
washed with H2O, dried over anhydrous Na2SO4 and the solvent
was evaporated to dryness. The crude product was purified by
column chromatography eluting with CH2Cl2 to afford mono-O-
acetylcurcumin (17) (28 mg, 25%) and di-O-acetylcurcumin (18)
(85 mg, 70%) as yellow amorphous solid; m.p. 163e165 ꢀC (lit. [18]
160 ꢀC). The spectroscopic (IR, 1H NMR and mass spectra) data of
compound 18 were consistent with the reported values [18].
5.2.7.1. Dihydrocurcumin (24). 1H NMR (400 MHz, CDCl3 þ 2 drops
of CD3OD)
d
2.64 (t, J ¼ 7.7 Hz, 2H, H-7), 2.87 (t, J ¼ 7.7 Hz, 2H, H-6),
3.83 and 3.90 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 5.56 (s, 1H, H-4), 6.27 (d,
J ¼ 15.7 Hz, 1H, H-2), 6.66 (partially obscured signal, 1H, H-500), 6.68
(s, 1H, H-200), 6.79 (d, J ¼ 7.8 Hz, 1H, H-600), 6.88 (d, J ¼ 8.2 Hz, 1H, H-
50), 6.98 (d, J ¼ 1.4 Hz, 1H, H-20), 7.04 (dd, J ¼ 8.2, 1.4 Hz, 1H, H-60),
7.48 (d, J ¼ 15.7 Hz, 1H, H-1); EIMS m/z (% rel. abund.): 370 [M]þ
(24), 352 (97), 305 (47), 177 (100), 150 (46), 137 (90), 135 (36).
5.2.6.1. Mono-O-acetylcurcumin (17). Orange foam: 1H NMR
(400 MHz, CDCl3)
d
2.30 (s, 3H, OAc), 3.86 and 3.93 (each s, 2 ꢁ 3H,
2 ꢁ OMe), 5.81 and 5.85 (each s, 2 ꢁ 1H, H-4 and OH), 6.44 (d,
J¼ 15.8Hz,1H,H-6), 6.35(d, J¼15.8Hz,1H, H-2), 6.92(d, J¼ca8.0Hz,
H-500), 7.03 (br s, 2H, H-20 and H-200), 7.11 (br d, J ¼ ca 8.0 Hz,1H, H-50),
7.59 (br d, J ¼ 15.8 Hz, 2H, H-1 and H-7); EIMS m/z (% rel. abund.) 410
[M]þ (17), 368 (40), 350 (100), 191 (33), 190 (57), 177 (24).
5.2.8. Catalytic hydrogenation of curcumin (1)
Tetrahydrocurcumin (25), hexahydrocurcumin (26) and octa-
hydrocurcumin (27) were prepared by catalytic hydrogenation of
curcumin (1) according to the literature procedure [20,21]. Their
Compound 2 was subjected to acetylation in the same manner
to that of compound 1 to give Mono-O0-acetyldemethoxycurcumin

4454
C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
identities were confirmed by 1H NMR spectroscopic data compar-
ison with those of the reported values [20,21].
flow rate: 1 mL/min, detector: 254 nm) to yield 38a (tR 13.25 min)
and 38b (tR 14.39 min).
5.2.9. Synthesis of unsaturated and saturated mono-keto analogs
The enone 28 was prepared by dehydration of hexahy-
drocurcumin (26) by the literature method [14]. The dienones 29 and
30, andthetrienone31 werepreparedbyDDQoxidationof theenone
28 using the literature procedure [14]. Catalytic hydrogenation of the
enone 28 using palladium on charcoal as a catalyst furnished the
saturated ketone 32. The spectroscopic data of the synthesized
compounds were consistent with the reported values [14,21].
5.2.12.2. Demethoxycurcumin isoxazole isomer1 (38a). White solid;
m.p. 198e200 ꢀC; IR nmax: 3406, 1645, 1606, 1590, 1513, 1433, 1277,
1171, 1033, 960, 820 cmꢂ1
;
1H NMR (400 MHz, CDCl3 þ 5 drops of
CD3OD)
d
3.91 (s, 3H, OMe), 6.38 (br s, 1H, H-4), 6.76 (d, J ¼ 16.4 Hz,
1H, H-10), 6.81 (br d, J ¼ 8.4 Hz, 2H, H-3000 and H-5000), 6.87 (d,
J ¼ 8.1 Hz, 1H, H-50000), 6.92 (d, J ¼ 16.4 Hz, 1H, H-100), 6.98 (br d,
J ¼ 8.1 Hz, 1H, H-60000), 7.05 (br s, 1H, H-20000), 7.07 (d, J ¼ 16.4 Hz, 1H,
H-200), 7.25 (obscured signal, 1H, H-20), 7.31 (br d, J ¼ 8.4 Hz, 2H, H-
2000 and H-6000); ESMS (ꢂve) m/z (% rel. abund.) 334 [M ꢂ H]ꢂ (100).
5.2.10. Synthesis of pyrazole analogs of the curcuminoids 1, 2 and 3
Curcumin (1) (5 g, 13.6 mmol) was dissolved in AcOH (150 ml)
and hydrazine hydrate (1.5 mL, 30.9 mmol) was added. The reaction
mixture was stirred at 50 ꢀC for 24 h; water (100 mL) was then
added and the mixture was extracted with EtOAc (150 mL ꢁ 3). The
combined organic phase was washed with H2O, dried over anhy-
drous Na2SO4. The solvent was evaporated and the residue was
purified by column chromatography using CH2Cl2eMeOH (50:2) to
yield compound 33 (3.5 g, 71%).
5.2.12.3. Demethoxycurcumin isoxazole isomer 2 (38b). White solid;
m.p. 194e195 ꢀC; IR nmax: 3406, 1645, 1606, 1590, 1513, 1433, 1277,
1171, 1033, 960, 820 cmꢂ1
;
1H NMR (400 MHz, CDCl3 þ 5 drops of
CD3OD)
d 3.90 (s, 3H, OMe), 6.37 (br s, 1H, H-4), 6.76 (partial
obscured signal, 1H, H-10), 6.79 (br d, J ¼ 8.4 Hz, 2H, H-30000 and H-
5
0000), 6.86e6.89 (m, 2H, H-10 and H-5000), 6.98 (br s, 1H, H-2000), 7.01
(br d, J ¼ 8.1 Hz, 1H, H-6000), 7.06 (d, J ¼ 16.4 Hz, 1H, H-200), 7.22
(obscured signal, 1H, H-20), 7.25 (obscured signal, 1H, H-20), 7.31 (br
d, J ¼ 8.4 Hz, 2H, H-2000 and H-6000); 13C NMR (100 MHz, CDCl3 þ 5
5.2.10.1. Curcumin
pyrazole
(33). Yellow
crystals
(from
drops of CD3OD) d
55.9 (3000-OMe), 97.5 (C-4), 108.9 (C-2000), 110.7 (C-
CH2Cl2eMeOH); m.p. 223e224 ꢀC (lit. [20] 215 ꢀC): 1H NMR data
were in agreement with those reported previously [24]; ESMS
(ꢂve) m/z (% rel. abund.) 363 [M ꢂ H]ꢂ (100).
10), 113.1 (C-100), 114.8 (C-5000), 115.7 (C-30000, C-50000), 121.4 (C-6000),
128.6 (C-20000, C-60000), 134.8 (C-20), 135.6 (C-200), 146.9 (C-3000, C-4000),
157.4 (C-40000), 162.3 (C-5), 168.4 (C-3); ESMS (ꢂve): m/z (% rel.
abund.) 334 [M ꢂ H]ꢂ (100).
By using the same procedure for the preparation of compound
33, compound 2 was converted to the corresponding pyrazole
analog 34 in 77%. 1H NMR data were in agreement with those
reported previously [20].
Compound 39 was prepared in 68% yield by the method
employed for compound 37.
By using the same procedure for the preparation of compound
33, compound 3 was converted to the corresponding pyrazole 35 as
white solid, m.p. >250 ꢀC (lit. [20] 272e273 ꢀC) in 41%. 1H NMR data
were in agreement with those reported previously [20].
5.2.12.4. Bisdemethoxycurcumin isoxazole (39). Needles (from
CH2Cl2eMeOH); m.p. >250 ꢀC; IR nmax: 3341, 1637, 1604, 1514, 1450,
1282, 1254, 970, 835 cmꢂ1; 1H NMR (400 MHz, CDCl3 þ 12 drops of
CD3OD)
d
6.35 (s, 1H, H-4), 6.72 (d, J ¼ 16.3 Hz, 1H, H-10), 6.77 (br d,
J ¼ 8.4 Hz, 4H, H-3000, H-5000, H-30000, H-50000), 6.85 (d, J ¼ 16.4 Hz, 1H, H-
100), 7.04 (d, J ¼ 16.4 Hz, 1H, H-200), 7.21 (d, J ¼ 16.3 Hz, 1H, H-20), 7.33
(d, J ¼ 8.4 Hz, 4H, H-2000, H-6000, H-20000, H-60000); ESMS (þve): m/z (% rel.
abund.) 633 [2M þ Na]þ (100); HR-TOFMS (ESþ): m/z 306.1130
[M þ H]þ; calcd for C19H15NO3 þ H, 306.1130.
5.2.11. Synthesis of phenylpyrazole analog of curcumin (1)
Compound 1 was converted to the corresponding phenyl-
pyrazole 36 in 81% by the same procedure for the preparation of
compound 33, but using phenylhydrazine in place of hydrazine
hydrate. 1H NMR data were in agreement with those reported
previously [24].
5.2.13. Synthesis of isoxazole analog of dihydrocurcumin (24) and
tetrahydrocurcumin (25)
5.2.12. Synthesis of isoxazole analog of the curcuminoids 1, 2 and 3
A solution of curcumin (1) (100 mg, 0.27 mmol) in pyridine
(3 mL) was treated with NH2OH$HCl (100 mg) and the mixture was
stirred at 50 ꢀC for 6 h. The reaction was worked up with H2O and
the solution was extracted with EtOAc. The organic phase was
washed with H2O, dried over anhydrous Na2SO4 and the solvent
was evaporated to dryness. The crude product was purified by
column chromatography using CH2Cl2eMeOH (100:1.5) to give
compound 37 (70 mg, 70%).
Starting from dihydrocurcumin (24), the isoxazole analog 40
was prepared in 63% yield by the same method employed for the
preparation of compound 37.
5.2.13.1. Dihydrocurcumin isoxazole (40). White solid; m.p.
168e170 ꢀC; 1H NMR (CDCl3)
d
2.92 (br s, 4H, H-100 and H-200), 3.84
and 3.92 (each s, 6H, 2 ꢁ OMe), 5.94 (br s, 1H, H-4), 6.68 (br s, 1H, H-
2
0000), 6.70 (obscured signal, H-60000), 6.73 (d, J ¼ 16.3 Hz,1H, H-10), 6.84
(d, J ¼ 8.5 Hz,1H, H-50000), 6.89 (d, J ¼ 8.1 Hz,1H, H-5000), 6.98 (br s,1H,
H-2000), 7.01 (d, J ¼ 8.1 Hz,1H, H-6000), 7.19 (d, J ¼ 16.3 Hz,1H, H-20); 13C
5.2.12.1. Curcumin isoxazole (37). Needles (from CH2Cl2en-
hexane); m.p. 177e178 ꢀC (lit. [24] m.p. 162 ꢀC); IR nmax: 3447, 1646,
NMR (100 MHz, CDCl3) d
28.2 (C-6), 34.2 (C-7), 55.9 (3000-OMe and
3
0000-OMe), 100.7 (C-4), 108.6 (C-2000), 110.9 (C-20000), 111.0 (C-2), 114.3
1606, 1575, 1513, 1433, 1369, 1277, 808, 734 cmꢂ1
;
1H NMR
(C-50000), 114.7 (C-5000), 121.0 (C-60000), 121.4 (C-6000), 128.2 (C-1000), 132.6
(C-10000), 134.6 (C-1), 144.0 (C-40000), 146.4 (C-30000), 146.8 (C-4000, C-3000),
163.6 (C-5), 168.4 (C-3); ESMS (þve): m/z (% rel. abund.) 368
[M þ H]þ (100); HR-TOFMS (ESþ): m/z 368.1505 [M þ H]þ; calcd for
C21H21NO5 þ H, 368.1498.
(400 MHz, CDCl3)
d
3.93 (s, 2 ꢁ 3H, 2 ꢁ OMe), 6.41 (s, 1H, H-4), 6.79
(d, J ¼ 16.3 Hz, 1H, H-10), 6.89 (br d, J ¼ 8.0 Hz, 1H, H-5000), 6.90 (br d,
J ¼ 8.0 Hz, 1H, H-50000), 6.94 (d, J ¼ 16.7 Hz, 1H, H-100), 7.01 (over-
lapping signals, H-2000, H-60000), 7.04 (m, 1H, H-6000), 7.06 (br s, 1H, H-
2
0000), 7.08 (d, J ¼ 16.7 Hz, 1H, H-200), 7.26 (d, J ¼ 16.3 Hz, 1H, H-20);
The isoxazole analog 42 was prepared as followed. A solution of
25 (116 mg, 0.31 mmol) in pyridine (3 mL) was treated with
NH2OH$HCl (60 mg) and the mixture was stirred at ambient
temperature for 1 h. The reaction was worked up with H2O and the
solution was extracted with EtOAc. The organic phase was washed
with H2O, dried over anhydrous Na2SO4 and the solvent was
ESMS (þve) m/z (% rel. abund.) 366 [M þ H]þ (100).
A 1:1 mixture of 38a and 38b was prepared from 2 in 65% yield
by the method employed for compound 37. The 38a/38b mixture
was further purified by normal phase HPLC (column: Luna Silica (2)
100 A, 5
mm, 4.6 ꢁ 250 mm, mobile phase: CH2Cl2eMeOH (99:1),

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
4455
evaporated to dryness. The crude product was purified by column
chromatography using CH2Cl2eMeOH (100:1.5) to give the mon-
oxime of 25 (85 mg, 70%), which was dissolved in benzene (2 mL)
and p-toluenesulfonic acid monohydrate (50 mg) was added. The
reaction mixture was stirred at 60 ꢀC for 1.5 h; water was added and
the mixture was extracted with EtOAc. The combined organic phase
was washed with water and dried over anhydrous Na2SO4. The
solvent was evaporated and the residue was chromatographed
using CH2Cl2eMeOH (10:0.1) to yield 42 (62 mg, 77%).
5.2.15. 3,3-Dimethylallylation of the isoxazole 37
The isoxazole 37 was subjected to 3,3-dimethylallylation in
similar manner to that of methylation of compound 1, but using
3,3-dimethylallyl bromide in place of methyl iodide, to afford
a mixture of 46a and 46b and 47 in 25 and 73% yields, respectively.
5.2.15.1. Compounds 46a/46b (2:1 mixture). IR nmax: 3427, 2934,
1646, 1559, 1511, 1430, 1269, 1137, 962, 810 cmꢂ1; 46a: 1H NMR
(400 MHz, CDCl3)
d
1.73 and 1.76 (each s, 2 ꢁ 3H, 2 ꢁ Me), 3.91 and
3.93 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 4.60 (br d, J ¼ 6.1 Hz, 2H, H-100000),
5.50 (m, 1H, H-200000), 5.74 (s, 1H, OH), 6.40 (s, 1H, H-4), 6.80 (d,
J ¼ 16.3 Hz, 1H, H-10), 6.86 (d, J ¼ 8.0 Hz, 1H, H-5000), 6.90 (d,
J ¼ 8.0 Hz, 1H, H-50000), 6.95 (d, J ¼ 16.4 Hz, 1H, H-100), 7.00e7.11 (m,
5H, H-200, H-2000, H-20000, H-6000, H-60000), 7.27 (d, J ¼ 16.3 Hz, 1H, H-20);
ESMS (þve): m/z (% rel. abund.) 434 [M þ H]þ (100); 46b: 1H NMR
5.2.13.2. Tetrahydrocurcumin isoxazole (42). White amorphous
solid. 1H NMR (CDCl3, 400 MHz):
d 2.87e2.91 (m, 6H), 2.96 (m, 2H)
(H-10, H-20, H-100, H-200), 3.82 and 3.83 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 5.50
and 5.51 (each s, 2 ꢁ 1H, 2 ꢁ OH), 5.67(s,1H, H-4), 6.61e6.67 (m, 4H,
H-2000, H-6000, H-20000, H-60000), 6.80 and 6.81 (each d, J ¼ 7.8 Hz, H-5000
and H-50000); ESMS (þve): m/z (% rel. abund.) 370 [M þ H]þ (100).
(400 MHz, CDCl3)
d
1.73 and 1.76 (each s, 6H, 2 ꢁ Me), 3.91 and 3.93
(each s, 6H, 2 ꢁ OMe), 4.60 (br d, J ¼ 6.1 Hz, 2H, H-100000), 5.50 (m, 1H,
H-200000), 5.77 (s, 1H, OH), 6.40 (s, 1H, H-4), 6.79 (d, J ¼ 16.4 Hz, 1H, H-
10), 6.85 (br d, J ¼ 8.0 Hz,1H, H-50000), 6.91 (br d, J ¼ 8.0 Hz, 1H, H-5000),
5.2.14. Pentylation of the isoxazole 37
The isoxazole 37 was subjected to n-pentylation in similar
manner to that of methylation of compound 1, but using n-pentyl
iodide in place of methyl iodide, to afford 43 and 44 (1:1 mixture)
and 45 in 24 and 62%, respectively. The 43/44 mixture was sepa-
6.98 (d, J ¼ 16.4 Hz, 1H, H-100), 7.00e7.07 (m, 5H, H-200, H-2000, H-20000
,
H-6000, H-60000), 7.26 (d, J ¼ 16.3 Hz, 1H, H-20); 46a/46b: ESMS (þve):
m/z (% rel. abund.) 434 [M þ H]þ (100).
rated by normal phase HPLC (column: Luna Silica (2) 100A, 5 mm,
4.6 ꢁ 250 mm, mobile phase: n-hexaneeCHCl3 (1:1), flow rate:
1.2 mL/min, detector: 254 nm) to yield 43 (tR 21.42 min) and 44 (tR
19.02 min).
5.2.15.2. Compound 47. White solid; m.p. 167e186 ꢀC; IR nmax
2928, 1645, 1581, 1512, 1432, 1312, 1136, 1028, 989, 818 cmꢂ1 1H
:
;
NMR (400 MHz, CDCl3)
d
1.73 and 1.76 (each s, 4 ꢁ 3H, 4 ꢁ Me), 3.90
and 3.91 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 4.60 (br d, J ¼ 6.1 Hz, 4H, H-100000
and H-1000000), 5.50 (m, 2H, H-200000 and H-2000000), 6.40 (s, 1H, H-4), 6.81
(d, J ¼ 16.3 Hz, 1H, H-10), 6.86 (br d, J ¼ 8.3 Hz, 2H, H-5000 and H-50000),
5.2.14.1. Compound 43. White solid; m.p.116e117 ꢀC; IR nmax: 3376,
2937, 1644, 1597, 1509, 1467, 1428, 1279, 1225, 1163, 1140, 1032, 962,
807, 734 cmꢂ1. 1H NMR (400 MHz, CDCl3)
5
d
0.91 (t, J ¼ 6.8 Hz, 3H, H-
6.97 (d, J ¼ 16.4 Hz, 1H, H-100), 7.01e7.11 (m, 5H, H-200, H-2000, H-20000
,
00000), 1.40 (m, 4H, H-300000 and H-400000), 1.84 (m, 2H, H-200000), 3.90 and
H-6000, H-60000), 7.27 (d, J ¼ 16.3 Hz, 1H, H-20); ESMS (ꢂve) m/z (% rel.
abund.) 500 [M ꢂ H]ꢂ (100); HR-TOFMS (ESꢂ): m/z 500.2436
[M ꢂ H]ꢂ; calcd for C31H35NO5eH, 500.2437.
3.93 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 4.02 (t, J ¼ 6.4 Hz, 2H, H-100000), 6.4000(0s,
1H, H-4), 6.80 (d, J ¼ 16.3 Hz,1H, H-10), 6.85 (d, J ¼ 8.0 Hz,1H, H-5 ),
6.90 (d, J ¼ 8.0 Hz, 1H, H-50000), 6.95 (d, J ¼ 16.5 Hz, 1H, H-100),
6.99e7.09 (m, 5H, H-200, H-2000, H-20000, H-6000, H-60000), 7.27 (d,
5.2.16. Ethylation of the isoxazole 37
J ¼ 16.3 Hz,1H, H-20); 13C NMR (100 MHz, CDCl3)
d
13.9 (C-500000), 22.4
The isoxazole 37 was subjected to ethylation in similar manner
to that of methylation of compound 1, but using diethyl sulfate in
place of methyl iodide, to afford a 5:2 mixture of 48a and 48b in 23%
and 49 in 60%.
(C-200000), 28.0 (C-300000), 28.8 (C-400000), 55.9 (30000-OMe), 56.0 (3000-OMe),
69.00(00C-100000) , 97.6 (C-4),108.2 (C-20000),109.6 (C-2000),110.0 (C0-0010 0),111.2
(C-5 ), 113.8 (C-100), 114.6 (C-50000), 121.0 (C-6000), 121.6 (C-6 ), 128.4
(C-10000), 128.5 (C-1000), 134.8 (C-20), 135.5 (C-200), 146.6 (C-40000), 146.8
(C-30000),149.6 (C-3000, C-4000),162.1 (C-5),168.5 (C-3); ESMS (ꢂve): m/z
(% rel. abund.) 434 [M ꢂ H]ꢂ (100); HR-TOFMS (APCIþ): m/z 436.2121
[M þ H]þ; calcd for C26H29NO5 þ H, 436.2118.
5.2.16.1. Compound 48a/48b (5:2 mixture). IR nmax: 3420, 2934,
1644, 1598, 1509, 1430, 1267, 1138, 1032, 961, 808 cmꢂ1; 48a: 1H
NMR (400 MHz, CDCl3)
d
1.46 (t, J ¼ 6.9 Hz, 3H, H-200000), 3.92 and
3.93 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 4.12 (m, 2H, H-100000), 5.75 (s, 1H, OH),
6.40 (s, 1H, H-4), 6.80 (d, J ¼ 16.3 Hz, 1H, H-10), 6.86 (d, J ¼ 8.0 Hz,
1H, H-5000), 6.90 (d, J ¼ 8.0 Hz, 1H, H-50000), 6.95 (d, J ¼ 16.4 Hz, 1H, H-
100), 7.00e7.11 (m, 5H, H-200, H-2000, H-20000, H-6000, H-60000), 7.27 (d,
5.2.14.2. Compound 44. White solid; m.p. 115e116 ꢀC; 1H NMR
(400 MHz, CDCl3)
d
0.91 (t, J ¼ 6.8 Hz, 3H, H-500000), 1.40 (m, 4H, H-
300000 and H-400000), 1.84 (m, 2H, H-200000), 3.90 and 3.93 (each s, 2 ꢁ 3H,
2 ꢁ OMe), 4.02 (t, J ¼ 6.4 Hz, 2H, H-100000), 6.40 (s, 1H, H-4), 6.79 (d,
J ¼ 16.3 Hz, 1H, H-10), 6.85 (d, J ¼ 8.2 Hz, 1H, H-50000), 6.91 (d,
J ¼ 8.0 Hz, 1H, H-5000), 6.97 (d, J ¼ 16.4 Hz, 1H, H-100), 7.01e7.08 (m,
5H, H-200, H-2000, H-20000, H-6000, H-60000), 7.26 (d, J ¼ 16.3 Hz, 1H, H-20);
ESMS (ꢂve): m/z (% rel. abund.) 434 [M ꢂ H]ꢂ (100); HR-TOFMS
(APCIþ): m/z 436.2125 [M þ H]þ; calcd for C26H29NO5þH, 436.2118.
J ¼ 16.3 Hz, 1H, H-20); 48b: 1H NMR (400 MHz, CDCl3)
d 1.46 (t,
J ¼ 6.9 Hz, 3H, H-200000), 3.91 and 3.94 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 4.12
(m, 2H, H-100000), 5.78 (s, 1H, OH), 6.40 (s, 1H, H-4), 6.79 (d,
J ¼ 16.4 Hz, 1H, H-10), 6.85 (br d, J ¼ 8.0 Hz, 1H, H-50000), 6.91 (br d,
J ¼ 8.0 Hz, 1H, H-5000), 6.98 (d, J ¼ 16.4 Hz, 1H, H-100), 7.00e7.11 (m,
5H, H-200, H-2000, H-20000, H-6000, H-60000), 7.26 (d, J ¼ 16.3 Hz, 1H, H-20);
48a/48b: ESMS (þve) m/z (% rel. abund.) 434 [M þ H]þ (100).
5.2.14.3. Compound 45. Aggregated needles (from n-hex-
aneeCH2Cl2); m.p. 130e131 ꢀC; IR nmax: 3446, 2952, 2869, 1634,
1596, 1581, 1558, 1514, 1467, 1393, 1325, 1268, 1241, 1138,
5.2.16.2. Compound 49. White plates (from CH2Cl2en-hexane);
m.p. 175 ꢀC; IR nmax: 2976, 1633, 1581, 1558, 1514, 1473, 1264, 1239,
1050 cmꢂ1; 1H NMR (400 MHz, CDCl3)
d
0.89 (t, J ¼ 6.9 Hz, 6H, H-
1136, 1026, 967, 800 cmꢂ1 1H NMR (400 MHz, CDCl3)
; d 1.46 (t,
500000 and H-5000000), 1.37 (m, 8H, H-300000, H-3000000, H-400000, H-4000000), 1.82
(m, 4H, H-200000 and H-2000000), 3.87 and 3.88 (each s, 2 ꢁ 3H, 2 ꢁ OMe),
4.60 (br t, J ¼ 5.4 Hz, 4H, H-100000 and H-1000000), 6.38 (s, 1H, H-4), 6.78
(d, J ¼ 16.4 Hz, 1H, H-10), 6.82 and 6.83 (each d, J ¼ 8.2 Hz, 2H, H-5000
and H-50000), 6.95 (d, J ¼ 16.4 Hz, 1H, H-100), 6.99e7.06 (m, 5H, H-200,
H-2000, H-20000, H-6000, H-60000), 7.25 (d, J ¼ 16.3 Hz, 1H, H-20); ESMS
(þve): m/z (% rel. abund.) 506 [M þ H]þ (100).
J ¼ 6.9 Hz, 6H, H-200000 and H-2000000), 3.91 and 3.92 (each s, 2 ꢁ 3H,
2 ꢁ OMe), 4.12 (m, 4H, H-100000 and H-1000000), 6.40 (s, 1H, H-4), 6.81 (d,
J ¼ 16.3 Hz,1H, H-10), 6.85 and 6.86 (each d, J ¼ 8.0 Hz, 2H, H-5000 and
H-50000), 6.97 (d, J ¼ 16.4 Hz, 1H, H-100), 7.01e7.11 (m, 5H, H-200, H-2000,
H-20000, H-6000, H-60000), 7.27 (d, J ¼ 16.3 Hz, 1H, H-20); ESMS (þve): m/z
(% rel. abund.) 422 [M þ H]þ (100); HR-TOFMS (ESþ): m/z 422.1974
[M þ H]þ; calcd for C25H27NO5 þ H, 422.1967.

4456
C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
5.2.17. Acetylation of the isoxazole 37
607 cmꢂ1; 1H NMR (400 MHz, CDCl3)
d
3.90 and 3.93 (each s, 2 ꢁ 3H
Ac2O (0.05 mL) was added to a solution of compound 37 (90 mg,
0.25 mmol) in pyridine (2 mL) and the reaction mixture was stirred
at ambient temperature for 1 h. The crude product which were
obtained by the usual work up followed by solvent extraction were
subjected to column chromatography eluting with CH2Cl2:MeOH
(100:0.8) to afford a mixture of 50 and 51 (31 mg, 30%), and 52
(62 mg, 55%). The 50/51 mixture was subjected to normal phase
HPLC separation (column: Luna Silica (2) 100A, 5
mobile phase: n-hexaneeCHCl3 (30:50), flow rate: 0.8 mL/min,
detector: 254 nm) to yield 47 (tR 30.99 min) and 48 (tR 33.62 min).
and 1 ꢁ3H, 3 ꢁ OMe), 6.40 (s,1H, H-4), 6.81 (d, J ¼ 16.3 Hz,1H, H-10),
6.86 (br d, J ¼ 8.3 Hz,1H, H-5000), 6.90 (br d, J ¼ 8.3 Hz,1H, H-50000), 6.94
(d, J ¼ 16.4 Hz, 1H, H-100), 7.00 (br d, J ¼ 8.3 Hz, 1H, H-60000), 7.04e7.08
(m, 4H, H-200, H-2000, H-20000, H-6000), 7.28 (d, J ¼ 16.3 Hz, 1H, H-20); 13C
NMR (100 MHz, CDCl3)
d
55.9 (3000-OMe and 30000-OMe), 56.0 (4000-
OMe), 97.7 (C-4),108.2 (C-20000),109.1 (C-2000),111.1 (C-10),111.2 (C-5000),
113.8 (C-100), 114.6 (C-50000), 121.1 (C-6000), 121.6 (C-60000), 128.5 (C-10000),
128.6 (C-1000), 134.7 (C-20), 135.6 (C-200), 146.7 (C-40000), 146.8 (C-30000),
149.3 (C-3000),150.2 (C-4000), 162.1 (C-5), 168.4 (C-3); ESMS (þve): m/z
(% rel. abund.) 380 [M þ H]þ (100); HR-TOFMS (APCIþ): m/z 380.1497
[M þ H]þ; calcd for C22H21NO5 þ H, 380.1492.
m
m, 4.6 ꢁ 250 mm,
5.2.17.1. Compound 50. White solid; m.p. 167e168 ꢀC; IR nmax
3421, 1759, 1647, 1602, 1561, 1513, 1429, 1371, 1276, 1220, 1121, 1031,
962, 828, 739 cmꢂ1; 1H NMR (400 MHz, CDCl3)
2.30 (s, 3H, OAc),
:
5.2.18.2. Compound 54. Colorless foam; 1H NMR (400 MHz, CDCl3)
d
d
3.89, 3.92 and 3.94 (each s, 3 ꢁ 3H, 3 ꢁ OMe), 6.40 (s,1H, H-4), 6.79
3.86 and 3.94 (each s, 2 ꢁ 3H, 2 ꢁ OMe), 6.41 (s, 1H, H-4), 6.79 (d,
J ¼ 16.3 Hz, 1H, H-10), 6.91 (d, J ¼ 8.1 Hz, 1H, H-50000), 7.00e7.14 (m,
7H, H-100, H-200, H-2000, H-20000, H-5000, H-6000, H-60000), 7.27 (d, J ¼ 16.3 Hz,
(d, J ¼ 16.4 Hz, 1H, H-10), 6.85 (br d, J ¼ 8.0 Hz, 1H, H-50000), 6.91 (br d,
J ¼ 8.0 Hz,1H, H-5000), 6.98 (d, J ¼ 16.4 Hz,1H, H-100), 7.04 (br s,1H, H-
2000), 7.06e7.09 (m, 4H, H-200, H-20000, H-6000, H-60000), 7.26 (d, J ¼ 16.3 Hz,
1H, H-20); 13C NMR (100 MHz, CDCl3)
d
20.6 (COMe), 55.9 (3000-
1H, H-20); 13C NMR (100 MHz, CDCl3) 55.9 (3000-OMe and 30000-OMe),
d
OMe), 56.0 (30000-OMe), 97.6 (C-4), 108.8 (C-2000), 110.2 (C-20000), 110.8
(C-10), 114.8 (C-50000), 116.5 (C-100), 120.0 (C-60000), 121.5 (C-6000), 121.6
(C-60000), 123.1 (C-5000), 128.1 (C-1000 and C-10000), 134.9 (C-200), 135.0 (C-
20), 146.8 (C-40000), 147.0 (C-30000), 151.1 (C-3000), 152.2 (C-4000), 161.8 (C-
5), 168.7 (C-3), 168.8 (COMe); ESMS (ꢂve): m/z (% rel. abund.) 406
[M ꢂ H]ꢂ (100); HR-TOFMS (APCIþ): m/z 408.1449 [M þ H]þ; calcd
for C23H21NO6 þ H, 408.1442.
56.0 (4000-OMe), 97.6 (C-4), 108.8 (C-2000), 108.9 (C-20000), 110.9 (C-10),
111.2 (C-50000), 114.2 (C-100), 114.7 (C-5000), 120.8 (C-60000), 121.5 (C-6000),
128.2 (C-1000), 129.0 (C-10000), 134.9 (C-20), 135.4 (C-200), 146.8 (C-4000),
146.9 (C-3000), 149.2 (C-30000), 150.0 (C-40000), 162.1 (C-5), 168.5 (C-3);
ESMS (þve): m/z (% rel. abund.) 380 [M þ H]þ (100); HR-TOFMS
(APCIþ): m/z 380.1501 [M þ H]þ; calcd for C22H21NO5 þ H, 380.1492.
5.2.18.3. Compound 55. White powder; m.p. 159e160 ꢀC; IR nmax
3449, 2999, 2936, 2838, 1646, 1601, 1561, 1513, 1428, 1266, 1224,
1140,1026, 966, 807, 767, 736 cmꢂ1; 1H NMR (400 MHz, CDCl3)
3.89
:
5.2.17.2. Compound 51. White solid; m.p.159 ꢀC; 1H NMR (400 MHz,
CDCl3)
d
2.30 (s, 3H, OAc), 3.87 and 3.92 (each s, 2 ꢁ 3H, 2 ꢁ OMe),
d
6.44 (s,1H, H-4), 6.87 (d, J ¼ 16.3 Hz,1H, H-10), 6.90 (d, J ¼ 8.1 Hz,1H,
H-50000), 6.95 (d, J ¼ 16.4 Hz, 1H, H-100), 6.99e7.10 (m, 6H, H-200, H-2000,
H-20000, H-5000, H-6000, H-60000), 7.29 (d, J ¼ 16.3 Hz,1H, H-20); ESMS (ꢂve):
m/z (% rel. abund.) 406 [M ꢂ H]ꢂ (100); HR-TOFMS (APCIþ): m/z
408.1444 [M þ H]þ; calcd for C23H21NO6 þ H, 408.1442.
and 3.91 (each s, 2 ꢁ 3H and 2 ꢁ 3H, 4 ꢁ OMe), 6.41 (s,1H, H-4), 6.81
(d, J ¼ 16.3 Hz, 1H, H-10), 6.84 (br d, J ¼ 8.2 Hz, 1H, H-5000), 6.86 (br d,
J ¼ 8.2 Hz,1H, H-50000), 6.98 (d, J ¼ 16.4 Hz,1H, H-100), 7.03e7.10 (m, 5H,
H-200, H-2000, H-20000, H-6000, H-60000), 7.27 (d, J ¼ 16.3 Hz, 1H, H-20); 13C
NMR (100 MHz, CDCl3) d
55.8 and 55.9 (3000-OMe, 30000-OMe, 4000-OMe,
4
0000-OMe), 97.6 (C-4), 108.7, 109.0 (C-2000, C-20000), 111.0 (C-10), 111.1 (C-
5.2.17.3. Compound 52. White solid; m.p.188e189 ꢀC; IR nmax: 3015,
5000, C-50000),114.0 (C-100),120.8,121.0 (C-6000, C-60000),128.5,128.8 (C-1000,
C-10000), 134.6 (C-20), 135.3 (C-200), 149.1, 149.8, 150.1 (C-3000, C-30000, C-
4000, C-40000), 162.0 (C-5), 168.3 (C-3); ESMS (þve): m/z (% rel. abund.)
394 [M þ H]þ (100); HR-TOFMS (APCIþ): m/z 394.1650 [M þ H]þ;
calcd for C22H23NO5 þ H, 394.1649.
2924, 1753, 1638, 1601, 1566, 1514, 1470, 1396, 1207, 1225, 1203, 971,
832, 740, 656 cmꢂ1
;
1H NMR (400 MHz, CDCl3)
d
2.31 (s, 2 ꢁ 3H,
2 ꢁ OAc), 3.87 (s, 2 ꢁ 3H, 2 ꢁ OMe), 6.47 (br s, 1H, H-4), 6.89 (d,
J ¼ 16.3 Hz,1H, H-10), 7.02e7.15 (m, 8H, H-100, H-200, H-2000, H-20000, H-5000,
H-50000, H-6000, H-60000), 7.31 (d, J ¼ 16.3 Hz,1H, H-20); 13C NMR (100 MHz,
CDCl3)
d
20.5 (COMe), 55.8(3000-OMe, 30000-OMe), 98.5(C-4),110.1,110.6
5.2.19. Methylation of a mixture of the isoxazole mixture 38a/38b
The 1:1 mixture of the isoxazoles 38a/38b was subjected to
methylation in similar manner to that of compound 37 to give a 1:1
mixture of 56a and 56b in 80%.
(C-2000, C-20000), 119.8, 120.0 (C-6000, C-60000), 123.0, 123.1 (C-5000, C-50000),
134.2 (C-20),134.4,134.7(C-1000, C-10000),135.0 (C-200),140.2, 140.4 (C-4000,
C-40000),151.3 (C-3000, C-30000),161.7 (C-5),168.0 (C-3),168.8 (2 ꢁ COMe);
ESMS(þve): m/z (%rel. abund.) 450 [Mþ H]þ (100);HR-TOFMS (ESþ):
m/z 450.1566 [M þ H]þ; calcd for C25H23NO7 þ H, 450.1553.
5.2.19.1. Compounds 56a/56b (1:1 mixture). IR nmax: 2962, 2838,
1644, 1603, 1577, 1509, 1430, 1307, 1263, 1139, 1025, 961, 821 cmꢂ1
;
5.2.18. Methylation of the isoxazole 37
56a: 1H NMR (400 MHz, CDCl3)
d
3.82, 3.89 and 3.92 (each s, 3 ꢁ 3H,
Compound 37 (1.5 g, 4.1 mmol) was dissolved in dry acetone
(45 mL) and anhydrous K2CO3 (1.1 g) and MeI (4.5 mL) was added.
The reaction mixture was reflux for 5 h. Water was added and the
mixture extracted with CH2Cl2. The combined organic phase was
washed with H2O, dried over anhydrous Na2SO4 and the solvent was
removed under vacuum. The crude products were separated and
purified bycolumn chromatography using CH2Cl2eMeOH (100:1) to
yield a 1:1 mixture of isomeric mono-methylated analogs 53 and 54
(590 mg, 38%) and the di-methylated analog 55 (805 mg, 50%). A
mixture of 53 and 54 was separated by normal phase HPLC (column:
3 ꢁ OMe), 6.41 (s, 1H, H-4), 6.83 (d, J ¼ 16.4 Hz, 1H, H-10), 6.86 (d,
J ¼ 8.2 Hz,1H, H-50000), 6.89 (br d, J ¼ 8.4 Hz, 2H, H-3000 and H-5000), 6.98
(d, J ¼ 16.4 Hz,1H, H-100), 7.03 (br s,1H, H-20000), 7.06e7.12 (m, 2H, H-200
and H-60000), 7.28 (d, J ¼ 16.4 Hz,1H, H-20), 7.45 (br d, J ¼ 8.4 Hz, 2H, H-
2
000 and H-6000); 56b: 1H NMR (400 MHz, CDCl3)
d 3.82, 3.89 and 3.92
(each s, 3 ꢁ 3H, 3 ꢁ OMe), 6.40 (s,1H, H-4), 6.83 (d, J ¼ 16.4 Hz,1H, H-
10), 6.85 (d, J ¼ 8.2 Hz,1H, H-5000), 6.89 (br d, J ¼ 8.4 Hz, 2H, H-30000 and
H-50000), 6.97 (d, J ¼ 16.4 Hz,1H, H-100), 7.03 (br s,1H, H-2000), 7.06e7.12
(m, 2H, H-200 and H-6000), 7.27 (d, J ¼ 16.4 Hz, 1H, H-20), 7.45 (br d,
J ¼ 8.4 Hz, 2H, H-20000 and H-60000); 56a/56b: ESMS (þve) m/z (% rel.
abund.) 749 [2M þ Na]þ (100).
Luna Silica (2) 100A, 5
aneeEtOAc (72:28), flow rate: 1 mL/min, detector: 254 nm) to yield
pure 53 (tR 31.38 min) and 54 (tR 32.01 min).
m
m, 4.6 ꢁ 250 mm, mobile phase: n-hex-
5.2.20. Methylation of the isoxazole 39
The isoxazole 39 was subjected to methylation in similar
manner to that of compound 37 to give the corresponding dimethyl
ether analog 57 in 78%.
5.2.18.1. Compound 53. Colorless foam; IR nmax: 3413, 2935, 2836,
1645, 1598, 1513, 1431, 1268, 1159, 1139, 1025, 961, 809, 736,

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 4446e4457
4457
5.2.20.1. Compound 57. Aggregated needles (from CH2Cl2en-
hexane), m.p.180e181 ꢀC; IR nmax:2934,1646,1604,1578,1559,1510,
which prevented the color change. NIH, RIF and kanamycin (Sigma)
were included as controls.
1307, 1248, 1174, 1092, 969, 819 cmꢂ1 1H NMR (400 MHz, CDCl3)
;
d
3.82 (s, 2 ꢁ 3H, 2 ꢁ OMe), 6.39 (s, 1H, H-4), 6.85 (d, J ¼ 16.4 Hz, 1H,
H-10), 6.90 (br d, J ¼ 8.4 Hz, 4H, H-3000, H-5000, H-30000, H-50000), 6.97 (d,
J ¼ 16.4 Hz, 1H, H-100), 7.10 (d, J ¼ 16.4 Hz, 1H, H-200), 7.28 (d,
J ¼ 16.4 Hz, 1H, H-20), 7.45 (d, J ¼ 8.4 Hz, 4H, H-2000, H-6000, H-20000, H-
Acknowledgements
This work was supported by the Research Team Strengthening
Grant of the National Center for Genetic Engineering and Biotech-
nology, National Science and Technology Development Agency.
Support from the Center of Excellence for Innovation in Chemistry
(PERCH-CIC), CommissiononHigherEducation,MinistryofEducation
is gratefully acknowledged. CC acknowledges a scholarship from the
Royal Golden Jubilee Ph.D. Program of The Thailand Research Fund.
6
0000); ESMS (þve): m/z (% rel. abund.) 334 [M þ H]þ (100); HR-TOFMS
(ESþ): m/z 334.1442 [M þ H]þ; calcd for C21H19NO3 þ H, 334.1443.
5.2.21. Synthesis of isoxazole analogs of mono-O-
demethylcurcumin (4)
A 1:2 mixture of the isoxazole analogs 58a and 58b was
prepared from 4 in 60% yield by the same method employed for the
preparation of compound 37 from compound 1.
References
5.2.21.1. Compounds 58a/58b (1:2 mixture). IR nmax: 3453, 3134,
[1] E. Valadas, F. Antunes, Eur. J. Radiol. 55 (2005) 154e157.
[2] R.A.M. Breen, R.F. Miller, T. Gorsuch, C.J. Smith, A. Schwenk, W. Holmes,
J. Ballinger, L. Swaden, M.A. Johnson, I. Cropley, M.C.I. Lipman, Thorax 61
(2006) 791e794.
[3] N.S. Shah, A. Wright, G.-H. Bai, L. Barrera, F. Boulahbal, N. Martin-Casabona,
F. Drobniewski, C. Gilpin, M. Havelková, R. Lepe, R. Lumb, B. Metchock,
F. Portaels, M.F. Rodrigues, S. Rüsch-Gerdes, A.V. Deun, V. Vincent, K. Laserson,
C. Wells, J.P. Cegielski, Emerg. Infect. Dis. 13 (2007) 380e387.
[4] D.F. Warner, V. Mizrahi, Clin. Microbiol. Rev. 19 (2006) 558e570.
[5] Y. Zhang, Annu. Rev. Pharmacol. Toxicol. 45 (2005) 529e564.
[6] A.L. Okunade, M.P.F. Elvin-Lewis, W.H. Lewis, Phytochemistry 65 (2004)
1017e1032.
[7] R.K. Maheshwari, A.K. Singh, J. Gaddipatti, R.C. Srimal, Life Sci. 78 (2006)
2081e2087.
[8] A.J. Ruby, G. Kuttan, K.D. Babu, K.N. Rajasekharan, R. Kuttan, Cancer Lett. 94
(1995) 79e83.
[9] L.N. Gringberg, O. Shalev, H.H. Tonnesen, E.A. Lachmillewitz, Int. J. Pharma-
ceut. 132 (1996) 251e257.
[10] M.M.-Y. Chan, C.-T. Ho, H.-I. Huang, Cancer Lett. 96 (1995) 23e29.
[11] M.M.-Y. Chan, H.-I. Huang, M.R. Fenton, D. Fong, Biochem. Pharmacol. 55
(1998) 1955e1962.
1645, 1601, 1561, 1509, 1434, 1376, 1277, 1118, 1023, 962, 816 cmꢂ1
;
58a: 1H NMR (400 MHz, CDCl3 þ 6 drops of CD3OD)
d 3.88 (s, 3H,
OMe), 6.37 (s, 1H, H-4), 6.74 (d, J ¼ 16.3 Hz, 1H, H-10), 6.81 and 6.84
(each d, J ¼ 7.9 Hz, 1H, H-50000 and H-5000), 6.87 (br d, J ¼ 8.1 Hz, 1H, H-
6
0000), 6.92 (d, J ¼ 16.3 Hz, 1H, H-100), 6.96e7.06 (m, 4H, H-200, H-2000,
H-20000, H-6000), 7.17 (d, J ¼ 16.3 Hz, 1H, H-20); 58b: 1H NMR (400 MHz,
CDCl3 þ 6 drops of CD3OD)
d 3.88 (s, 3H, OMe), 6.36 (s, 1H, H-4),
6.70 (d, J ¼ 16.3 Hz, 1H, H-10), 6.78 and 6.83 (each d, J ¼ 8.1 Hz, 1H,
H-50000 and H-5000), 6.87 (br d, J ¼ 8.1 Hz, 1H, H-6000), 6.92 (d,
J ¼ 16.3 Hz, 1H, H-100), 6.96e7.06 (m, 4H, H-200, H-2000, H-20000, H-60000),
7.16 (d, J ¼ 16.3 Hz, 1H, H-20); 58a/58b: ESMS (ꢂve): m/z (% rel.
abund.) 350 [M ꢂ H]ꢂ (100).
5.2.22. Synthesis of isoxazole analog of di-O-demethylcurcumin (5)
The isoxazole analog 59 was prepared in 67% yield by the same
method employed for the preparation of compound 37 from
compound 1.
[12] H. Inano, M. Onoda, N. Inafuku, M. Kubota, Y. Kamada, T. Osawa, H. Kobayashi,
K. Wakabayashi, Carcinogenesis 21 (2000) 1835e1841.
[13] R. Thapliyal, G.B. Maru, Food Chem. Toxicol. 39 (2001) 541e547.
[14] C. Changtam, H.P. de Koning, H. Ibrahim, M.S. Sajid, M.K. Gould,
A. Suksamrarn, Eur. J. Med. Chem. 45 (2010) 941e956.
[15] A. Mazumder, K. Raghavan, J. Weinstein, K.W. Kohn, Y. Pommier, Biochem.
Pharmacol. 49 (1995) 1165e1170.
[16] S. Venkateswarlu, M.S. Ramachandra, G.V. Subbaraju, Bioorg. Med. Chem. 13
(2005) 6374e6380.
[17] H. Ohtsu, Z. Xiao, J. Ishida, M. Nagai, H.K. Wang, H. Itokawa, C.Y. Su, C. Shih,
T. Chiang, E. Chang, Y. Lee, M.Y. Tsai, C. Chang, K.H. Lee, J. Med. Chem. 45
(2002) 5037e5042.
5.2.22.1. Compound 59. Pale white solid; m.p. >250 ꢀC; IR nmax
3367, 1647, 1603, 1565, 1520, 1440, 1288, 1113, 958, 808 cmꢂ1 1H
6.35 (s,1H, H-4), 6.69(d,
:
;
NMR (400MHz, CDCl3 þ6 drops of CD3OD)
d
J ¼ 16.3 Hz,1H, H-10), 6.71e6.87 and 6.96e7.06 (each m, 8H, H-100, H-
200, H-2000, H-20000, H-5000, H-50000, H-6000, H-60000), 7.15 (d, J ¼ 16.3 Hz,1H, H-
20); ESMS (ꢂve): m/z (% rel. abund.) 336 [M ꢂ H]ꢂ (100); HR-TOFMS
(ESꢂ): m/z 306.1130 [M ꢂ H]ꢂ; calcd for C19H15NO5eH, 306.1130.
[18] S. Mishra, U. Narain, R. Mishra, K. Misra, Bioorg. Med. Chem. 13 (2005)
1477e1486.
5.3. Biological activities
[19] K. Mohammadi, K.H. Thompson, B.O. Patrick, T. Storr, C. Martins, E. Polishchuk,
V.G. Yuen, J.H. McNeill, C.J. Orvig, J. Inorg. Biochem. 99 (2005) 2217e2225.
[20] J. Ishida, H. Ohtsu, Y. Tachibana, Y. Nakanishi, K.B. Bastow, M. Nagai,
H.-K. Wang, H. Itokawa, K.-H. Lee, Bioorg. Med. Chem. 10 (2002)
3481e3487.
[21] S.-L. Lee, W.-J. Huang, W.W. Lin, S.-S. Lee, C.-H. Chen, Bioorg. Med. Chem. 13
(2005) 6175e6181.
[22] H. Jiang, B.N. Timmermann, D.R. Gang, J. Chromatogr. A 1111 (2006) 21e31.
[23] S.-Y. Park, D.H.L. Kim, J. Nat. Prod. 65 (2002) 1227e1231.
[24] S. Mishra, K. Kamodiya, N. Surolia, A. Surolia, Bioorg. Med. Chem. 16 (2008)
2894e2902.
5.3.1. Mycobacterial strains
The reference strain M. tuberculosis H37Ra and the clinical
isolates of MDR-TB were obtained from Ramathibodi Hospital,
Mahidol University, Bangkok, Thailand.
5.3.2. Antimycobacterial assay
Antimycobacterial activities were determined by microplate
Alamar blue assay (MABA) [31]. Briefly, the compounds were dis-
solved in dimethyl sulfoxide (Sigma) and subsequently diluted
[25] C. Selvam, S.M. Jachak, R. Thilagavati, A.K. Chakraborti, Bioorg. Med. Chem.
Lett. 15 (2005) 1793e1797.
[26] M. Labbozzetta, R. Baruchello, P. Marchetti, M. Gueli, P. Poma, M. Notarbartolo,
D. Simoni, N. D’Alessandro, Chem. Biol. Interact. 181 (2009) 29e36.
[27] M. Labbozzetta, M. Notarbartolo, P. Poma, A. Maurici, L. Inguglia, P. Marchetti,
R. Paolo, M. Rizzi, R. Baruchello, D. Simoni, N. D’Alessandro, Ann. N. Y. Acad.
Sci. 1155 (2009) 278e283.
[28] J. Mao, H. Yuan, Y. Wang, B. Wan, M. Pieroni, Q. Huang, R.B. van Breemen,
A.P. Kozikowski, S.G. Franzblau, J. Med. Chem. 52 (2009) 6966e6978.
[29] A. Lilienkampf, M. Pieroni, B. Wan, Y. Wang, S.G. Franzblau, A.P. Kozikowski,
J. Med. Chem. 53 (2010) 678e688.
twofold in 100 mL of Middlebrook 7H9GC in clear flatbottom, 96-
well microplates. A mycobacterial suspension was prepared in
0.04% Tween 80 and diluted with sterile distilled water to
a turbidity of the McFarland no. 1. The suspension was then diluted
1:50 with 7H9GC, and 100
m
L was added to the wells. After incu-
bated at 37 ꢀC for 7 days, 12.5
m
L of 20% Tween 80 and 20 L of
m
[30] J. Mao, H. Yuan, Y. Wang, B. Wan, D. Pak, R. He, S.G. Franzblau, Bioorg. Med.
Chem. Lett. 20 (2010) 1263e1268.
[31] S.G. Franzblau, R.S. Witzig, J.C. McLaughlin, P. Torres, G. Madico, A. Hernandez,
M.T. Degnan, M.B. Cook, V.K. Quenzer, R.M. Ferguson, R.H. Gilman, J. Clin.
Microbiol. 36 (1998) 362e366.
Alamar blue (SeroTec Ltd., Oxford, UK) were added to all wells.
Growth of the organisms was determined after reincubation at
37 ꢀC for 16e24 h by visual determination of a color change from
blue to pink. The MIC was defined as the lowest concentration