Curcuminoid analogs with potent activity against Trypanosoma and Leishmania species

European Journal of Medicinal Chemistry 45 (2010) 941–956

Contents lists available at ScienceDirect

European Journal of Medicinal Chemistry

journal homepage: h ttp://www.elsevier.com/locate/ejmech

Original article

Curcuminoid analogs with potent activity against Trypanosoma and

Leishmania species

Chatchawan Changtam ^a, Harry P. de Koning ^b, Hasan Ibrahim ^b, M. Sohail Sajid ^b,

,

Matthew K. Gould ^b, Apichart Suksamrarn ^a

*

^aDepartment of Chemistry, Faculty of Science, Ramkhamhaeng University, Ramkhamhaeng Road, Bangkok 10240, Thailand

^bInstitute of Biomedical and Life Sciences, Division of Infection and Immunity, University of Glasgow, Glasgow G12 8TA, UK

a r t i c l e i n f o

a b s t r a c t

Article history:

The natural curcuminoids curcumin (1), demethoxycurcumin (2) and bisdemethoxycurcumin (3) have

been chemically modiﬁed to give 46 analogs and 8 pairs of 1:1 mixture of curcuminoid analogs and these

parent curcuminoids and their analogs were assessed against protozoa of the Trypanosoma and Leish-

mania species. The parent curcuminoids exhibited low antitrypanosomal activity (EC₅₀for our drug-

Received 11 May 2009

Received in revised form

15 July 2009

Accepted 17 November 2009

Available online 24 November 2009

sensitive Trypanosoma brucei brucei line (WT) of compounds 1, 2 and 3 are 2.5, 4.6 and 7.7 mM,

respectively). Among 43 curcuminoid analogs and 8 pairs of 1:1 mixture of curcuminoid analogs tested, 8

pure analogs and 5 isomeric mixtures of analogs exhibited high antitrypanosomal activity in sub-

micromolar order of magnitude. Among these highly active analogs, 1,7-bis(4-hydroxy-3-methoxy-

Keywords:

Curcuminoid

Chemical modiﬁcation

Antitrypanosomal activity

Antileishmanial activity

Structure–activity relationship

phenyl)hept-4-en-3-one (40) was the most active compound, with an EC₅₀value of 0.053 ꢀ 0.007

was about 2-fold more active than the standard veterinary drug diminazene aceturate (EC₅₀

M). Using a previously characterized diminazene-resistant T. b. brucei (TbAT1-KO) and

mM; it

0.12 ꢀ 0.01

m

a derived multi-drug resistant line (B48), no cross-resistance of curcuminoids was observed to the

diamidine and melaminophenyl arsenical drugs that are the current treatments. Indeed, curcuminoids

carrying a conjugated keto (enone) motif, including 40, were signiﬁcantly more active against T. b. brucei

B48. This enone motif was found to contribute to particularly high trypanocidal activity against all

Trypanosoma species and strains tested. The parent curcuminoids showed low antileishmanial activity

(EC₅₀values of compounds 1 and 2 for Leishmania mexicana amastigotes are 16 ꢀ 3 and 37 ꢀ 6

respectively) while the control drug, pentamidine, displayed an EC₅₀of 16 ꢀ 2 M. Among the active

curcuminoid analogs, four compounds exhibited EC₅₀values of less than 5 M against Leishmania major

mM,

m

promastigotes and four against L. mexicana amastigotes. No signiﬁcant difference in sensitivity to cur-

cuminoids between L. major promastigotes and L. mexicana amastigotes was observed. The parent cur-

cuminoids and most of their analogs were also tested for their toxicity against human embryonic kidney

(HEK) cells. All the curcuminoids exhibited lower toxicity to HEK cells than to T. b. brucei bloodstream

forms and only one of the tested compounds showed signiﬁcantly higher activity against HEK cells than

curcumin (1). The selectivity index for T. b. brucei ranged from 3-fold to 1500-fold. The selectivity index

for the most active analog, the enone 40, was 453-fold.

1. Introduction

depends on decades-old preparations based on toxic heavy metals:

arsenic in melarsoprol for trypanosomiasis and antimony in

Trypanosomiasis and leishmaniasis are among the most

neglected diseases in the world [1,2]. The diseases are caused by

related protozoan parasites, Trypanosoma and Leishmania species,

respectively, both of the order Kinetoplastida. Therapeutic options

for both diseases are severely limited and ﬁrst line treatment still

meglumine antimoniate (Glucantime) and sodium stibogluconate

(Pentostam) for leishmaniasis [3,4]. Any newer options such as

diﬂuoromethyl ornithine (DFMO) for late stage human trypanoso-

miasis or lipid formulations of amphotericin B (AmBisome) suffer

from serious drawbacks such as very high cost and/or extreme

dosage schemes [3,5,6]. Yet, human African trypanosomiasis (HAT,

or sleeping sickness) reached epidemic proportions during the

1990s [7], at the same time that treatment failure with the tradi-

tional treatment, melarsoprol, reached alarming rates of up to 30%

* Corresponding author. Tel.: þ662 3190931; fax: þ662 3108404.

E-mail addresses: s_apichart@ru.ac.th, asuksamrarn@yahoo.com (A. Suksamrarn).

doi:10.1016/j.ejmech.2009.11.035

942

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

[3,8]. In addition, the impact of trypanosomiasis on livestock

farming and development in sub-Saharan Africa is enormous,

responsible for losses topping 1 billion US dollars [9]. Trypanosoma

evansi is further responsible for infections in camels, buffalo and

other livestock in South Asia [10] and Trypanosoma vivax affects

livestock in South America [11]. Similarly, various forms of leish-

maniasis, ranging from cutaneous and muco-cutaneous to visceral

leishmaniasis or kala-azar, is a severe disease throughout Africa,

South Asia, Mid and South America, with antimony resistance

particularly acute in India. The estimated mortality of these tropical

diseases is exceeded only by malaria and tuberculosis [2]. Thus, new,

affordable drugs are urgently required, preferably with a broad

activity against the kinetoplastid parasites.

O

1

2'

5'

2"

5"

7

R¹

R³

4

6

2

6'

6"

R²O

OR⁴

1, R¹= R³= OMe, R²= R⁴= H

2, R¹= R²= R⁴= H, R³= OMe

3, R¹= R²= R³= R⁴= H

O

Curcuminoids are the major constituents of turmeric (Curcuma

longa L.) and some other Curcuma species, including a number of

other plant species. It has been known for some time that curcu-

minoids have been used as a natural food additive. The main cur-

cuminoid isolated from C. longa is curcumin (1). The minor

constituents include demethoxycurcumin (2) and bisdemethoxy-

curcumin (3). Curcuminoids exhibited many interesting biological

activities [12], for example, antioxidant activity [13,14], anti-

inﬂammatory activity [15,16], anticancer activity [13,17,18], anti-

protozoal activity [19] and anti-HIV activity [20]. In recent years, it

has been reported that curcumin (1) displays moderate-to-low

activity against Trypanosoma brucei brucei [21] and against pro-

mastigotes of Leishmania major [22] but no effort to systematically

screen for a curcuminoid analog with enhanced anti-kinetoplastid

activity has been reported. Here we describe the synthesis and

antiparasitic screening of a series of curcuminoid analogs. Compared

to the parent compound, curcumin (1), many of the new compounds

displayed a much-enhanced activity against kinetoplastid parasites

that cause human and veterinary disease, while displaying limited if

any toxicity against a human cell line. We propose that analogs of

curcuminoids are promising new lead compounds against leish-

maniasis and African trypanosomiasis.

R¹

R³

R²O

OR⁴

Reagent

4, R¹= OH, R²= R⁴= H, R³= OMe

5, R¹= R³= OH, R²= R⁴= H

a

b

6, R¹= R²= R⁴= H, R³= OH

7, R¹= R³= OMe, R²= R⁴= Me

8, R¹= R⁴= H, R³= OMe, R²= Me

9, R¹= H, R²= R⁴= Me, R³= OMe

10, R¹= R³= R⁴= H, R²= Me

11, R¹= R³= OMe, R²= R⁴=

12, R¹= R³= OMe, R²=

13, R¹= R³= OMe, R²= R⁴=

c

d

e

, R⁴= H

14, R¹= R³= OMe, R²=

15, R¹= R³= OMe, R²= R⁴=

, R⁴= H

16, R¹= H, R²= R⁴=

, R³= OMe

e

f

OH

2. Chemistry

, R⁴= H

17, R¹= R³= OMe, R²=

OH

18, R¹= R³= OMe, R²= R⁴=

f

In order to investigate the antitrypanosomal and antileishmanial

activities of curcuminoid analogs, the parent curcuminoids curcu-

min (1), demethoxycurcumin (2) and bisdemethoxycurcumin (3),

which were obtained from the rhizomes of C. longa in a ratio of

76.0:16.2:3.8, were subjected to structural modiﬁcation.

OH

19a, R¹= H, R²=

, R³= OMe, R⁴= H

OH

f

19b, R¹= R²= H, R³= OMe, R⁴=

f

Scheme 1. Demethylation and alkylation of curcuminoids 1-3. Reagents and conditions:

(a) BBr₃, CH₂Cl₂, 0 ^ꢁC, then ambient temp.; (b) CH₃I, K₂CO₃, acetone, reﬂux; (c) n-propyl

iodide, K₂CO₃, acetone, reﬂux; (d) allyl bromide, K₂CO₃, acetone, reﬂux; (e) 3,3-dime-

thylallyl bromide, K₂CO₃, acetone, reﬂux; (f) 2-bromoethanol, K₂CO₃, acetone, reﬂux.

2.1. Demethylated and methylated analogs

To study the effects of free hydroxyl group and hydrogen bonding

of the oxygen function at the aromatic rings, the parent curcumi-

noids 1–3 were subjected to demethylation and methylation to the

appropriate demethylated and methylated analogs (see Scheme 1).

Thus, starting from the parent compound 1, demethylation to the

corresponding mono-O-demethyl analog 4 and di-O-demethyl

analog 5 were accomplished in 43 and 33% yields, respectively, by

treatment of 1 with boron tribromide in dry dichloromethane.

Demethylation of the curcuminoid 2 was similarly achieved to the

corresponding O-demethyl analog 6 in 64% yield. The spectroscopic

(IR, ¹H NMR and mass spectra) data of 4, 5 and 6 were consistent

with the reported values [23]. Methylation of compounds 1, 2 and 3

was achieved by reacting with methyl iodide in acetone in the

presence of potassium carbonate to give the corresponding ether

analogs 7, 8, 9 and 10 in 80, 20, 54 and 27% yields, respectively. The

spectroscopic data of 7 were in agreement with the reported values

[24], and those of the ether analogs 8 and 10 were consistent with

those reported previously [25]. The IR, ¹H NMR and MS data of the

ether 9 were consistent with their structures (see Section 5).

2.2. Higher alkyl ether analogs

A number of higher alkyl ether analogs of curcumin (1) were

prepared to study the effect of different nature of the alkyl groups

on antiprotozoal activities. By employing a similar procedure for

the preparation of the foregoing methyl ethers, but using different

alkyl bromide or iodide (n-propyl iodide, allyl bromide and

3,3-dimethylallyl bromide), the corresponding alkyl ethers, di-O-n-

propylcurcumin (11) (85%), mono-O-allylcurcumin (12) (33%), di-O-

allylcurcumin (13) (46%), mono-O-(3,3-dimethylallyl)curcumin

(14) (43%) and di-O-(3,3-dimethylallyl)curcumin (15) (40%), were

prepared (see Scheme 1). The O-3,3-dimethylallyl analog of

compound 2, di-O-(3,3-dimethylallyl)demethoxycurcumin (16),

was also prepared in 43% yield. Compounds 12, 13 and 15 have

previously been prepared and have been used in some biological

evaluations [26,27]. By comparison with compound 15, the spec-

troscopic data of compounds 14 and 16 (see Section 5) were

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

943

compatible with their structures. The structure of 11 was conﬁrmed

O

by spectroscopic data (see Section 5). Moreover, by using 2-bro-

moethanol as the alkylating agent, the 2-hydroxyethyl ether

analogs, mono-O-(2-hydroxyethyl)curcumin (17) and di-O-(2-

hydroxyethyl)curcumin (18), were also prepared in 47 and 32%

yields. The presence of the 2-hydroxyethoxyl moiety was evident

R¹

R³

R²O

OR⁴

1, R¹= R³= OMe, R²= R⁴= H

2, R¹= R²= R⁴= H, R³= OMe

3, R¹= R²= R³= R⁴= H

from the two-proton signals at

hydroxyethyl) ether (17), and the four-proton signals at

d

4.11 and 3.92 for the mono-O-(2-

4.07 and

d

3.90 for the di-O-(2-hydroxyethyl) ether (18). The IR and mass

spectral data also conﬁrmed the structures of these two

compounds.

A

mixture of 4⁰⁰-O-(2-hydroxyethyl)demethoxy-

curcumin (19a) and 4⁰-O-(2-hydroxyethyl)demethoxycurcumin

(19b) was similarly obtained from compound 2. Pure 19a was

obtained in 28% yield after careful repeated column chromatog-

raphy followed by recrystallization. However, compound 19b could

not be puriﬁed in similar manner to that of compound 19a. The

structure of 19a was conﬁrmed by spectroscopic data

(see Section 5).

O

R¹

R³

R²O

OR⁴

2.3. Nitro analog

Reagent

21, R¹= R³= OMe, R²= R⁴= Ac

22, R¹= H, R²= R⁴= Ac, R³= OMe

23, R¹= R³= H, R²= R⁴= Ac

h

i

To see the inﬂuence of an extra substituent on the aromatic ring of

curcumin (1), the nitro analog 20 was prepared in 27% yield as

described in the Section 5 (Scheme 2). The presence of a nitro group

at the 5⁰-position was evident from the proton resonance of H-6⁰,

24, R¹= R³= OMe, R²= PhCO, R⁴= H

25, R¹= R³= OMe, R²= R⁴= PhCO

26, R¹= R⁴= H, R²= PhCO, R³= OMe

27, R¹= R²= H, R³= OMe, R⁴= PhCO

28, R¹= H, R²= R⁴= PhCO, R³= OMe

29, R¹= R³= R⁴= H, R²= PhCO

30, R¹= R³= H, R²= R⁴= PhCO

i

which has changed from a broad doublet signal (J ¼ 8.1 Hz) at

d 7.10

i

in compound 1 to a singlet signal at 7.87 in compound 20. The mass

d

i

spectral data also conﬁrmed the structure of the nitro compound 20.

i

2.4. Acetate and benzoate esters

i

Scheme 3. Acetylation and benzoylation of curcuminoids 1–3. Reagents and condi-

tions: (h) Ac₂O, pyridine; (i) (PhCO)₂O, pyridine.

The ester derivatives chosen for this study were the acetate and

benzoate esters. Acetylation of the parent curcuminoids 1, 2 and 3

by conventional method furnished the di-O-acetyl analogs 21 [28],

22 [29] and 23 [30] in good yields (Scheme 3). Benzoylation of 1

with benzoic anhydride in pyridine furnished the monobenzoate

24 and the dibenzoate 25 [31] in 47 and 33% yields, respectively.

The curcuminoid 2 was similarly benzoylated to the mono-

benzoates 26 and 27, and the dibenzoate 28 in 34, 29 and 21%

yields, respectively. The curcuminoid 3 was also benzoylated to the

monobenzoates 29 and the dibenzoate 30 in 53 and 31% yields,

respectively (Scheme 3). The spectroscopic data of the synthesized

benzoates (see Section 5) were in agreement with the structures.

2.5. Reduced curcuminoid analogs

A number of non-conjugated analogs of curcuminoids 1, 2 and 3

were prepared for biological evaluation (Scheme 4). Catalytic

hydrogenation of compound 1, with palladium on charcoal as

a catalyst, furnished tetrahydrocurcumin (31), hexahydrocurcumin

(32) and octahydrocurcumin (33) in 64, 20 and 15% yields, respec-

tively. The spectroscopic data of compounds 31–33 were consistent

with the reported values [32,33]. From the curcuminoids 2 and 3, the

tetrahydro analogs 34 and 35 were similarly prepared in 72 and 75%

yields, respectively. The spectroscopic data of compounds 34 and 35

were consistent with the reported values [34]. In addition, the

reduced analogs 36, 37a/37b mixture, 38 and 39 were similarly

prepared from the curcuminoid analogs 4, 5 and 7 in 40, 37, 65 and

62% yields, respectively. The spectroscopic data of compound 36

were in agreement with the structure (see Section 5). The spectro-

scopic data of compounds 38 and 39 were consistent with the

reported values [26,34,35]. The products 37a and 37b are a 1:1

inseparable mixture. One of them, 37a, has been isolated from Zin-

giber ofﬁcinale [35]. At this stage, no further investigation for the

separation of these two compounds and other mixture was

attempted.

O

MeO

HO

OMe

OH

1

g

O

MeO

HO

OMe

OH

2.6. Monoketo analogs

In order to prepare monoketo analogs of the curcuminoid 1,

compound 32 was subjected to dehydration, with p-toluenesulfonic

acid as a catalyst, to afford the enone 40 in quantitative yield

NO₂

20

Scheme 2. Nitration of curcumin (1). Reagents and conditions: (g) NaNO₂, AcOH, CH₂Cl_2.

(Scheme 4). The presence of the a,b-unsaturated keto system was

944

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

corresponding inseparable isomeric enone 41a and 41b were

O

obtained in 88% yield (Scheme 5). The ¹H NMR data of compounds

41a and 41b mixture (see Section 5) were consistent with the

structures and with those reported values of 41a [37]. The enone 42

was prepared by sodium borohydride reduction of the saturated

diketone 38 followed by dehydration of the corresponding hex-

ahydro analog to the required enone 42 in 65% (Scheme 6). The

spectroscopic data of compound 42 were consistent with the

reported values [38]. Methylation of compound 40 furnished

the mono-O-methyl analogs 43a and 43b (1:1 mixture) and di-O-

methyl analogs 29 in 30 and 57% yields, respectively. The ¹H NMR

spectrum of 43 was similar to that of 40, except for the presence of

additional methoxyl signal in that of compounds 43 and 43b

mixture. The ¹H NMR data of 44 were consistent with those

reported previously [39]. Following the method for the preparation

of the enone 42 from the dione 38, a 1:1 mixture of compounds 45a

and 45b was prepared from the diketo compound 34 in 62% overall

yield (Scheme 6). The spectroscopic data of the products 45a and

45b were consistent with the reported values [36,40]. Compound 46

was similarly prepared in 60% overall yield from compound 35. The

spectroscopic (IR, ¹H NMR and mass spectra) data of compound 46

were consistent with the reported values [41].

R¹

R³

R²O

OR⁴

1, R¹= R³= OMe, R²= R⁴= H

2, R¹= R²= R⁴= H, R³= OMe

3, R¹= R²= R³= R⁴= H

4, R¹= OH, R²= R⁴= H, R³= OMe

5, R¹= R³= OH, R²= R⁴= H

7, R¹= R³= OMe, R²= R⁴= Me

j

O

R¹

R³

R²O

OR⁴

2.7. Dienone and trienone analogs

31, R¹= R³= OMe, R²= R⁴= H

34, R¹= R²= R⁴= H, R³= OMe

35, R¹= R²= R³= R⁴= H

In order to see whether higher degree of unsaturation would

affect the biological activities, the enone 40 was subjected to

dehydrogenation with DDQ in THF to give the dienones 47 [42] and

48, and the trienone 49 [43] in 28, 23 and 32% yields, respectively

(Scheme 7). The ¹H NMR spectrum of 47 showed the characteristic

36, R¹= OH, R²= R⁴= H, R³= OMe

38, R¹= R³= OH, R²= R⁴= H

a

,

b

,

g

,

d

-unsaturated keto system at

d

6.22 (d, J ¼ 15.4 Hz, H-4), 7.28

+

(dd, J ¼ 15.4,10.7 Hz, H-5), 6.64–6.73 (obscured signal, H-6) and 6.83

(d, J ¼ 15.4 Hz, H-7). Compound 48 also exhibited two characteristic

O

OH

R¹

R³

O

OH

R¹

R³

R²O

OR⁴

32, R¹= R³= OMe, R²= R⁴= H

37a, R¹= OMe, R²= R⁴= H, R³= OH

37b, R¹= OH, R²= R⁴= H, R³= OMe

39, R¹= R³= OMe, R²= R⁴= Me

R²O

OR⁴

32, R¹= R³= OMe, R²= R⁴= H

37a, R¹= OMe, R²= R⁴= H, R³= OH

37b, R¹= OH, R²= R⁴= H, R³= OMe

+

k

OH

MeO

HO

OMe

OH

O

R¹

R³

33

R²O

OR⁴

Scheme 4. Catalytic hydrogenation of curcuminoids 1–7. Reagents and conditions: (j)

H₂/Pd-C, EtOH.

40, R¹= R³= OMe, R²= R⁴= H

41a, R¹= OMe, R²= R⁴= H, R³= OH

41b, R¹= OH, R²= R⁴= H, R³= OMe

43a, R¹= R³= OMe, R²= H, R⁴= Me

43b, R¹= R³= OMe, R²= Me, R⁴= H

44, R¹= R³= OMe, R²= R⁴= Me

evident from the IR absorption band at 1660 cm^ꢂ1and the broad

l

doublet signal (J ¼ 15.9 Hz, H-4) at

d 6.08 and a partially overlapping

signal at

d

6.77–6.84 (H-5) in the ¹H NMR spectrum. Other proton

resonances and mass spectral data (see Section 5) conﬁrmed the

structure of compound 40. The structure of this compound was

further conﬁrmed by comparison with the reported ¹H NMR data of

gingerenone, which was isolated from Z. ofﬁcinale [36]. Dehydration

of the isomeric mixture of the hexahydro analogs 37a and 37b, the

Scheme 5. Dehydration of hexahydrocurcuminoid analogues 37a and 37b, and

methylation of the enone 40. Reagents and conditions: (k) p-TsOH, C₆H₆, reﬂux;

(l) CH₃I, K₂CO₃, acetone, reﬂux.

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

945

O

R¹

R³

MeO

HO

OMe

OH

R²O

OR⁴

40

38, R¹= R³= OH, R²= R⁴= H

34, R¹= R²= R⁴= H, R³= OMe

35, R¹= R²= R³= R⁴= H

o

1) m

2) n

O

MeO

HO

OMe

OH

O

R¹

R³

47

R²O

OR⁴

+

42, R¹= R³= OH, R²= R⁴= H

45a, R¹= R²= R⁴= H, R³= OMe

45b, R¹= OMe, R²= R³= R⁴= H

46, R¹= R²= R³= R⁴= H

MeO

HO

OMe

OH

Scheme 6. Preparation of the enones 42, 45a, 45b and 46 from the diketones 38, 34

and 35. Reagents and conditions: (m) NaBH₄, EtOH; (n) p-TsOH, C₆H₆, reﬂux.

48

+

a,b

-unsaturated keto systems at

d

7.52 (d, J ¼ 15.9 Hz, H-1), 6.76 (d,

J ¼ 15.9 Hz, H-2), 6.42 (d, J ¼ 15.6 Hz, H-4) and 6.98 (m, H-5). The

spectroscopic data of compound 49 were in agreement with the

structure and were consistent with the reported values [43].

MeO

HO

OMe

OH

2.8. Alkyl ethers and acetate derivatives of enone

49

A number of alkyl derivatives of the enone 40 were prepared (see

Scheme 8). Thus pentylation of 40 with n-pentyl iodide afforded the

mono-O-n-pentyl analogs 50a and 50b (1:1 mixture) and di-O-n-

pentyl analog 51 in 53 and 27%, respectively. Each of the proton

resonances of these two isomeric analogs 50a and 50b (see Section 5)

were superimposed, except for the methoxy resonances which

Scheme 7. Dehydrogenation of the enone 40. Reagent and condition: (o) DDQ, THF.

and 56b mixture showed almost identical ¹H NMR data. The only

difference was the resonance positions of the methoxyl groups of

these two isomers, which appeared at

d 3.83 and 3.84. Finally, in

appeared at

d 3.81, 3.82, 3.83 and 3.84 and were the diagnostic

order to see the inﬂuence of the keto function in biological activity,

the keto group in the enone 40 was reduced with sodium boro-

hydride to the corresponding alcoholic group in 57 in 96% yield

(Scheme 9). The spectroscopic data (see Section 5) were consistent

with the structure of 57.

signals for the existence of two isomers. The mono-O-(2-hydroxy-

ethyl) analogs 52a and 52b (1:1 mixture) were prepared in 27%

from 40 in the same manner as that of compound 17. Like the

analogs 50a and 50b mixture, each of the proton resonances of the

two isomeric analogs 52a and 52b (see Section 5) were also

superimposed, except for the methoxy resonances at d 3.80 (s, 3H),

3.81 (2 ꢃ 3H) and 3.82 (3H) that revealed the existence of two

isomers. Acetylation of the enone 40 gave a 1:1 mixture of the

monoacetates 53a and 53b and the diacetate 54 (Scheme 8). For the

isomeric mixture 53a and 53b, the existence of two isomers was

3. Results and discussion

3.1. Trypanocidal effects of curcuminoid analogs

evident from the two AcO signals at

d 2.27 and 2.28, and four MeO

3.1.1. Activity against a standard T. brucei strain 427

signals at 3.78, 3.79, 3.84 and 3.85. The spectroscopic data of the

diacetate 54 were consistent with the reported values [44].

d

The three parent curcuminoids (compounds 1–3), 43 curcumi-

noid analogs (compounds 4–24, 26, 27, 29, 31–36, 38–40, 42, 44,

46–49, 51, 54, 55 and 57) and 8 pairs of 1:1 isomeric mixtures of

curcuminoid analogs (compounds 37a/37b, 41a/41b, 43a/43b, 45a/

45b, 50a/50b, 52a/52b, 53a/53b and 56a/56b) were assessed for in

vitro trypanocidal properties on the standard drug-sensitive labo-

ratory strain 427 (Table 1). It has previously been reported that

curcumin (1) has distinct trypanocidal properties, with an in vitro

2.9. Saturated keto analogs and enol analog

To see whether the oleﬁnic function in the conjugated enone

system is important for biological activity, compounds 40, and 45a

and 45b mixture were subjected to catalytic hydrogenation to the

corresponding dihydro analogs 55, and 56a and 56b (1:1 mixture)

in 86 and 86% yields (Scheme 9). The spectroscopic data of

compound 55 (see Section 5) were in agreement with the structure

and were consistent with the reported values [33]. Compounds 56a

EC₅₀value of 4.8 ꢀ 0.9

mM for bloodstream forms of the poly-

morphic strain GUTat 3.1 [21]. Our own value for compound 1

against bloodstream forms of the monomorphic strain 427 was

similar at 2.5 ꢀ 0.4 M (n ¼ 3). The two other parent curcuminoids,

m

946

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

O

R¹

R²

MeO

HO

OMe

OH

HO

OH

55, R¹= R²= OMe

56a, R¹= H, R²= OMe

56b, R¹= OMe, R²= H

40

s

O

MeO

R¹O

OMe

OR²

O

R¹

R²

Reagent

HO

OH

50a, R¹=

50b, R¹= H, R²=

51, R¹= R²=

52a, R¹=

52b, R¹= H, R²=

, R²= H

p

40, R¹= R²= OMe

45a, R¹= H, R²= OMe

45b, R¹= OMe, R²= H

p

q

OH

, R²= H

OH

q

t

53a, R¹= Ac, R²= H

53b, R¹= H, R²= Ac

54, R¹= R²= Ac

r

OH

MeO

HO

OMe

OH

Scheme 8. Alkylation and acetylation of the enone 40. Reagents and conditions:

(p) n-pentyl iodide, K₂CO₃, acetone, reﬂux; (q) 2-bromoethanol, K₂CO₃, acetone, reﬂux;

(r) Ac₂O, pyridine.

57

demethoxycurcumin (2) and bisdemethoxycurcumin (3) were

approximately 2 and 3-fold less active than compound 1. To see

whether the polarity contributed from the oxygen functions on the

aromatic ring exert any effect on the biological activity of curcu-

minoids, the demethylated analogs 4, 5 and 6, and the methylated

analogs 7, 8, 9 and 10 were evaluated for antitrypanosomal activity.

Scheme 9. Catalytic hydrogenation and sodium borohydride reduction of the ketone

40. Reagents and conditions: (s) H₂/Pd-C, EtOH; (t) NaBH₄, EtOH.

evaluated for antitrypanosomal activity and it turned out that the

presence of a nitro substituent resulted in a signiﬁcant (6-fold)

decrease in activity.

It was found that the mono-O-demethylated curcuminoid

4

exhibited approximately 3-fold higher activity than the parent

compound, curcumin (1). Further demethylation to the di-O-

demethylated analog 5, however, resulted in about 2-fold decrease

in activity, though it still was more active than compound 1.

Compound 6, the demethylated analog of the parent curcuminoid

2, was 7-fold more active than the parent compound. At this point it

seemed that increase in polarity in the molecule resulted in

increase in antitrypanosomal activity. We therefore assessed

whether increase in lipophilicity of the parent compounds caused

decrease in activity. Rather surprisingly, compound 7, the di-O-

methyl analog of compound 1, was almost 6-fold more active than

its parent compound 1, whereas the mono- and di-O-methylated

analogs 8 and 9 were almost 3-fold more active than their parent

compound 2. However, the O-methylated analog 10 was slightly

less active than the parent curcuminoid 3. In order to see the effect

of higher alkyl ether analogs on the biological activity of curcumi-

noid analogs, the n-propyl ether 11, allyl ethers 12 and 13,

3,3-dimethylallyl ethers 14, 15 and 16, and 2-hydroxyethyl ethers

17, 18 and 19a were evaluated for antitrypanosomal activity. The

assay results have demonstrated that the activity of these alkyl

ether analogs was 1 to 4-fold more active than their respective

parent curcuminoids, except for compound 19a that the activity

was 6.5-fold more active than the parent compound 2.

In addition to the ether derivatives, the ester analogs, i.e. the

acetates 21, 22 and 23, and the benzoates 24–30, were subjected to

biological evaluation. The results demonstrated that the acetate

analogs were about 1.5 to 3-fold more active than their parent

curcuminoids (see Table 1). However, it was worth noting that their

EC₅₀values were still in excess of 1 mM. For the benzoate analogs,

some displayed 2 to 4-fold higher activity (compounds 29 and 27)

while others displayed almost identical activity (compounds 24 and

26) to those of the parent curcuminoids. It should be mentioned

that compounds 25, 28 and 30 were insoluble under the condition

used for biological evaluation and this was due to the hydrophobic

nature of the two benzoyl groups in the molecule.

It should be noted that a decrease in oxygenated (hydroxyl and

alkyl) substituents in the aromatic rings of curcuminoids tended to

generally decrease antitrypanosomal activity of the curminoids and

analogs. However, the existing information did not permit a deﬁn-

itive conclusion for these observations.

The reduced analogs of curcuminoids were next tested against T.

b. brucei s427. For the tetrahydro analogs 31, 34–36, and 38, the ﬁrst

two analogs (i.e. 31 and 34) were approximately 8 and 5-fold less

active than their respective parent compounds. However, the

number and nature of the oxygen functions on the aromatic rings

strongly determined the antitrypanosomal activity of the tetrahy-

dro analogs. Thus, the analog 35 was 3-fold more active than its

To test for the effect of an additional polar group on the aromatic

ring of curcuminoid, the nitro analog 20 was synthesized and

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

947

Table 1

Antitrypanosomal, antileishmanial and cytotoxic activities of curcuminoids and analogs.

Compound

T. brucei strain 427

T. brucei

DTbat1

T. brucei clone B48

L. major promastigotes

L. mexicana amastigotes

HEK

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19a

20

21

22

23

24

2.5 ꢀ 0.4

4.6 ꢀ 0.8

7.7 ꢀ 1.1

0.87 ꢀ 0.06

1.8 ꢀ 0.1

0.65 ꢀ 0.02

0.45 ꢀ 0.07

1.7 ꢀ 0.7

1.6 ꢀ 0.4

8.9 ꢀ 1.4

1.3 ꢀ 0.4

2.4 ꢀ 0.1

1.6 ꢀ 0.1

1.4 ꢀ 0.2

1.2 ꢀ 0.2

1.9 ꢀ 0.3

2.9 ꢀ 1.1

1.8 ꢀ 0.8

0.7 ꢀ 0.3

15 ꢀ 2

4.7 ꢀ 0.3

5.9 ꢀ 0.5

9.5 ꢀ 0.3

1.1 ꢀ 0.1

2.5 ꢀ 0.2

1.8 ꢀ 0.1

2.9 ꢀ 1.1

2.2 ꢀ 0.24

4.5 ꢀ 0.46

2.7 ꢀ 0.5^b

3.4 ꢀ 0.1

1.0 ꢀ 0.1

0.77 ꢀ 0.02

2.1 ꢀ 0.1

1.9 ꢀ 0.1

2.9 ꢀ 0.4^b

1.6 ꢀ 0.1

12 ꢀ 1.6^c

5.8 ꢀ 0.6^c

2.0 ꢀ 0.2

1.3 ꢀ 0.1

3.1 ꢀ 0.1

1.8 ꢀ 0.4

1.6 ꢀ 0.5

1.1 ꢀ 0.1

25 ꢀ 8

33 ꢀ 4

37 ꢀ 1

72 ꢀ 3

4.3 ꢀ 0.5

26 ꢀ 0.1

22 ꢀ 2

2.8 ꢀ 0.4

23 ꢀ 3

23 ꢀ 6

87 ꢀ 5

28 ꢀ 2

5.7 ꢀ 0.7

25 ꢀ 9

7.6 ꢀ 0.7

>100

48 ꢀ 9

22 ꢀ 4

33 ꢀ 8

9.7 ꢀ 1.5

28 ꢀ 3

73 ꢀ 18

>100

16 ꢀ 3

37 ꢀ 4

63 ꢀ 3

3.2 ꢀ 0.5

12 ꢀ 3

21 ꢀ 2

10 ꢀ 1

38 ꢀ 6

61 ꢀ 3

32 ꢀ 2

34 ꢀ 0.2

12 ꢀ 4

>100

21 ꢀ 4

27 ꢀ 0.8

27 ꢀ 7

30 ꢀ 4

37 ꢀ 5

32 ꢀ 0.7

14 ꢀ 4

34 ꢀ 4

43 ꢀ 7

7.1 ꢀ 1

76 ꢀ 10

INS

37 ꢀ 6

40 ꢀ 5

200 ꢀ 40

36 ꢀ 4

39 ꢀ 7

30 ꢀ 3

200 ꢀ 50

79 ꢀ 8

690 ꢀ 230

400 ꢀ 0

>500

0.62 ꢀ 0.07

2.0 ꢀ 0.2

1.9 ꢀ 0.2

8.8 ꢀ 0.2

1.8 ꢀ 0.1

2.2 ꢀ 0.1

1.9 ꢀ 0.1

1.7 ꢀ 0.1

1.6 ꢀ 0.2

2.9 ꢀ 0.3

2.3 ꢀ 0.1

0.85 ꢀ 0.13

18 ꢀ 0.2

2.2 ꢀ 0.2

2.0 ꢀ 0.1

3.9 ꢀ 0.1

3.8 ꢀ 0.2

INS

270 ꢀ 130

>500

>1000

>500

25 ꢀ 3

20 ꢀ 1

46 ꢀ 3

82 ꢀ 5

80 ꢀ 18

49 ꢀ 6

>500

1.6 ꢀ 0.5

1.5 ꢀ 0.3

3.2 ꢀ 0.4

3.1 ꢀ 0.7

INS

1.2 ꢀ 0.2

0.92 ꢀ 0.08

3.2 ꢀ 0.3

2.1 ꢀ 0.5

INS

66 ꢀ 10

9.3 ꢀ 1.1

INS

94 ꢀ 11

25

INS

26

27

28

29

3.7 ꢀ 0.4

1.2 ꢀ 0.3

INS

7.8 ꢀ 0.8

1.6 ꢀ 0.16

INS

7.8 ꢀ 0.8

4.7 ꢀ 0.4

INS

20 ꢀ 1

42 ꢀ 12

INS

>100

INS

>100

>500

370 ꢀ 110

INS

4.1 ꢀ 1.7

10 ꢀ 2

12 ꢀ 2

60 ꢀ 10

270^a

30

INS

31

32

33

34

35

36

21 ꢀ 11

37 ꢀ 7

12 ꢀ 2

90 ꢀ 17

>100

43 ꢀ 9

120 ꢀ 10

370 ꢀ 60

ND

17 ꢀ 3

38 ꢀ 3

5.9 ꢀ 0.9^b

>100

78 ꢀ 5

61 ꢀ 5

>100

22 ꢀ 8

34 ꢀ 1

18 ꢀ 1

>100

>50^a

ND

2.5 ꢀ 0.4

0.50 ꢀ 0.10

3.0 ꢀ 0.7

1.9 ꢀ 0.6

33 ꢀ 13

4.0 ꢀ 0.3

0.52 ꢀ 0.10

2.9 ꢀ 0.5

1.5 ꢀ 0.2

22 ꢀ 5

5.1 ꢀ 0.9

0.43 ꢀ 0.01

2.5 ꢀ 0.4

1.7 ꢀ 0.1

19 ꢀ 4

80 ꢀ 12

11 ꢀ 1

>100

64 ꢀ 14

16 ꢀ 3

40 ꢀ 5

53 ꢀ 3

>100

350 ꢀ 90

130 ꢀ 19

77 ꢀ 2

150 ꢀ 8

370 ꢀ 15

24 ꢀ 2

63 ꢀ 9

220 ꢀ 60

20 ꢀ 2

26 ꢀ 2

40 ꢀ 5

45 ꢀ 3

76 ꢀ 3

8.1 ꢀ 1.0

22 ꢀ 1

30 ꢀ 0.3

43 ꢀ 2

26 ꢀ 1

25 ꢀ 1

30 ꢀ 2

140 ꢀ 13

200 ꢀ 2

130 ꢀ 2

37a D 37b

38

39

40

41a D 41b

42

43a D 43b

44

45a D 45b

46

47

48

>100

41 ꢀ 12

8.9 ꢀ 0.8

10 ꢀ 2

>100

0.053 ꢀ 0.007

0.75 ꢀ 0.13

2.5 ꢀ 0.5

0.14 ꢀ 0.05

1.9 ꢀ 0.2

3.0 ꢀ 0.3

2.6 ꢀ 0.5

4.1 ꢀ 1.6

0.22 ꢀ 0.09

1.1 ꢀ 0.3

0.86 ꢀ 0.3

1.1 ꢀ 0.3

0.27 ꢀ 0.06

0.089 ꢀ 0.027

0.052 ꢀ 0.015

32 ꢀ 0.4

45 ꢀ 1

0.082 ꢀ 0.005

0.73 ꢀ 0.08

3.2 ꢀ 0.1

0.29 ꢀ 0.10

1.6 ꢀ 0.1

2.1 ꢀ 0.2

1.9 ꢀ 0.1

9.7 ꢀ 1.7

0.22 ꢀ 0.02

1.4 ꢀ 0.2

0.62 ꢀ 0.1

1.2 ꢀ 0.2

0.27 ꢀ 0.07

0.079 ꢀ 0.024

0.042 ꢀ 0.017

28 ꢀ 1

0.023 ꢀ 0.004^b

0.30 ꢀ 0.06^b

ND

17 ꢀ 2

7.4 ꢀ 1

17 ꢀ 5

28 ꢀ 6

28 ꢀ 8

23 ꢀ 4

21 ꢀ 8

22 ꢀ 7

4.6 ꢀ 0.7

4.8 ꢀ 0.6

19 ꢀ 4

53 ꢀ 8

41 ꢀ 4

29 ꢀ 0.8

35 ꢀ 1

>50

0.032 ꢀ 0.004

0.023 ꢀ 0.003^d

0.049 ꢀ 0.008^d

0.12 ꢀ 0.02^b

2.2 ꢀ 0.4

0.10 ꢀ 0.03

0.58 ꢀ 0.08

0.89 ꢀ 0.2

1.8 ꢀ 0.1

0.11 ꢀ 0.04

0.045 ꢀ 0.015

0.021 ꢀ 0.007

27 ꢀ 2

13 ꢀ 1

12 ꢀ 1

15 ꢀ 2

9.9 ꢀ 1.2

30 ꢀ 2

2.7 ꢀ 0.7

7.4 ꢀ 0.1

4.0 ꢀ 1.3

6.6 ꢀ 1.0

15 ꢀ 0.1

7.5 ꢀ 0.8

7.8 ꢀ 1.0

37 ꢀ 3

55 ꢀ 3

50 ꢀ 4

49

50a D 50b

51

52a D 52b

53a D 53b

54

55

56a D 56b

57

Diminazene

Pentamidine

48 ꢀ 2

9.0 ꢀ 2.8^d

6.3 ꢀ 0.6^c

1.0 ꢀ 0.2

0.67 ꢀ 0.06

>100

64 ꢀ 15

40 ꢀ 3

42 ꢀ 4

0.12 ꢀ 0.01

2.4 ꢀ 0.3

3.8 ꢀ 0.4

16 ꢀ 2

Data are EC₅₀values in

m

M ꢀ standard errors and are the average and S.E.M. of 3–6 independent experiments except where indicated. ND ¼ Not done. INS ¼ Insoluble.

a

n ¼ 1.

b

P < 0.05 relative to strain 427.

P < 0.01 relative to strain 427.

P < 0.001 relative to strain 427.

c

d

parent curcuminoid 3 and was almost 9-fold more active than

compound 34, the 3⁰⁰-O-methoxy analog. Compound 36, the mono-

O-demethyl analog of 31, was even 42-fold more active than the

conjugated compound 4. The octahydro analog 33 exhibited

extremely low activity; it was 31-fold less active than the parent

conjugated compound 1 (Table 1).

methylated analog 31 (0.50 ꢀ 0.10

m

M). The hexahydro analogs 32

We next explored mono-oxygenated analogs of the heptyl linker

chain. We ﬁrst synthesized the monoketo analog 40, which was

obtained by catalytic dehydration of compound 32. We discovered

that the enone 40 exhibited very high antitrypanosomal activity of

and 39 exhibited very low activity, being 7 and 73-fold less active

than the conjugated compounds 1 and 7, respectively. The 1:1

mixture 37a/37b was 3.5-fold less active than the corresponding

948

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

0.053 ꢀ 0.007

m

M, which was 47-fold more active than the parent

a mutant line that is highly resistant to diminazene aceturate as well

as slightly resistant to pentamidine and melaminophenyl arsenicals.

compound 1, or 321-fold more active than its immediate precursor,

the hexahydro analog 32. The activity of 40 was superior to the

activity of the standard veterinary trypanocide diminazene acetu-

The second line, B48, was derived from

DTbat1 by incremental

exposure to pentamidine in vitro, and having lost or mutated

a second drug transporter, HAPT1, is highly resistant to most dia-

midines including diminazene, pentamidine and to the melamino-

rate, which displayed an activity of 0.12 ꢀ 0.01

mM against s427.

In order to study more about structure–activity relationship and

to ﬁnd any possible additional potent analogs by using the mono-

keto analog 40 as the lead structure, we further explored different

analogs. It was found that the mono-demethylated analog mixture

41a/41b, the completely demethylated analog 42, the methylated

analogs 43a/43b (1:1 mixture), 44, 45a/45b (1:1 mixture), and the

enone 46 were all more active than their respective parent curcu-

minoids 1–3, except for compound 42 whose activity was about the

same as that of its parent compound 1. However, none of them

showed higher activity than the enone 40.

phenyl arsenicals [49,50]. Whereas

diminazene aceturate (positive control) in this series of experiments

(0.12 ꢀ 0.01 (n ¼ 13) vs. 2.4 ꢀ 0.3 M (n ¼ 14); P < 0.001, Student

DTbat1 was 20-fold resistant to

m

T-test), there was no signiﬁcant resistance to curcuminoid analogs in

this line (Wilcoxon’s signed ranks test for two groups, paired

observations; P > 0.05). The curcuminoids and analogs exhibited

antitrypanosomal activity against these two clonal lines in similar

manner to that of the wild-type strain (Table 1). The activity of most

of the compounds appeared to be within a 2-fold difference in

sensitivityagainst the two lines and at most 2.7-fold for compound 6.

It should be emphasized that no resistance was observed for any of

the highly active analogs, such as 40, 48 and 54 and the 53a/53b

mixture.

With the framework of the most active compound 40, we

investigated whether additional oleﬁnic function would give rise to

an increase in biological activity. The dienones 47 and 48, and the

trienone 49 were synthesized and assessed for antitrypanosomal

activity. The results revealed that compounds 48 and 49 were

approximately 2 to 11-fold more active than the parent curcumi-

noid 1, whereas compound 47 displayed activity similar to 1. We

then investigated the alkyl ether and the acetate analogs of the

enone 40 (i.e. compounds 50–54) and it was found that all the

analogs exhibited high activity. The monoacetates 53a and 53b

mixture and the diacetate 54 displayed the highest activity, at

B48 is a multi-drug resistant line, with in vitro resistance to

pentamidine and melarsen oxide of 128 and 11-fold, respectively

[49]. Out of 57 compounds tested (Table 1), only three displayed

signiﬁcantly less activity against this line: compounds 4 (3.1-fold),

12 (5-fold) and 13 (3.6-fold). All are close analogs of 1, with conju-

gated diketo linkers, differing only in the substitutions at the

aromatic rings, in particular allyl ether substitutions (12, 13).

Unexpectedly, the B48 line proved to be signiﬁcantly more sensitive

to some of the most active curcuminoids (Table 1), including

compounds 40, 41a/41b, 44, 45a/45b and 46. These compounds all

share the enone motif of the most active compound for WT

trypanosomes, 40; compounds 44 and 45a/45b displayed a highly

signiﬁcant (P < 0.001, Student’s T-test) 83 and 60-fold higher

activity against the B48 strain.

The tests with the transporter mutant lines thus lead us to

conclude that there is little or no risk of cross-resistance between

curcumin (analogs) and existing therapies of the diamidine and

melaminophenyl arsenical categories, which form the mainstay of

trypanosomiasis treatment [1,2]. It is clear that the deletion of the

TbAT1 gene does not diminish uptake of the curcuminoids signiﬁ-

cantly. This strongly suggests that curcuminoids are not a substrate

for this transporter, in line with our understanding of the TbAT1/P2

recognition motif [51]. The result with the B48 line was unexpected,

as this line displays hypersensitivity to some of the most promising

curcuminoids. This would appear to be highly advantageous in

dealing with the current resistance problems in the ﬁeld. Further

investigations are required to fully understand this phenomenon,

and whether it extends to other pentamidine-resistant lines. One

possible explanation is a mutation in the HAPT transporter gene,

creating an altered transporter with lower afﬁnity for diamidines but

higher afﬁnity for some curcuminoids, speciﬁcally those with the

enone motif.

0.089 ꢀ 0.027 and 0.052 ꢀ 0.015

mM, or 28 and 48-fold more active

than the natural parent curcuminoid 1. These two compounds thus

displayed activity statistically identical to the enone 40 (P > 0.05,

Student’s T-test) and superior to that of diminazene aceturate.

In order to see the contribution of the oleﬁnic group in the

enone 40 to the antitrypanosomal activity, the corresponding

dihydro analog 55 was prepared and it was found that removal of

the conjugated oleﬁnic function resulted in almost complete loss of

activity. The activity of 55 was 607-fold less active than that of 40.

Similar result was also observed for the 1:1 mixture of the satu-

rated keto analogs 56a/56b (see Table 1). To prove that the keto

function was also essential for activity, the enol 57 was prepared;

this change resulted in an almost complete loss of activity. We

conclude that the most essential structural feature for a curcumi-

noid to exhibit strong antitrypanosomal activity is the presence of

an

a

,

b-keto system at the heptyl chain.

A logical next step would be to identify the mode of action of 40

on the trypanosome, and optimize further by studying the inter-

action with the cellular target. This work, although in progress, falls

outside the remit of the current report. It is relevant to reiterate that

a great many hypotheses have been proposed for the mechanism of

action of curcumin (1) and its analogs, and in many different cell

types. Indeed, the activity of curcuminoids is likely to be the result

of multiple cellular interactions [12,45]. Thus, in the absence of an

optimization strategy based on target structure, the structure–

activity analysis offered here, based on overall antiparasitic effects,

is the most direct approach to new and improved lead compounds

for the target diseases.

A sub-set of the analogs was also tested against T. evansi. Most of

the analogs were equally or more potent against this species than

against our T. b. brucei reference strain 427 (Table 2), but there was

no statistical signiﬁcant difference in overall curcuminoid sensi-

tivity between the two species (P > 0.05; Wilcoxon’s signed ranks

test for two groups, paired observations). Although the T. evansi

strain was signiﬁcantly more sensitive to diminazene than T. b.

3.1.2.

DTbat1 and B48 strains

As it is vital that any new drug developed against African

trypanosomiasis is not cross-resistant with the diamidine and

arsenic-based drugs currently in use, we tested the activity of all the

analogs on two additional clonal lines, derived from T. b. brucei strain

brucei WT (0.018 ꢀ 0.003 M, n ¼ 5; P < 0.001; unpaired Student’s

m

T-test) and compound 12 (P < 0.05), it was less sensitive to

compound 40 (P < 0.02) and compound 13 (P < 0.01). Fig. 1 illus-

trates the results obtained for T. b. brucei WT and T. evansi. Impor-

tantly, these experiments establish that the curcuminoids have

highly similar activity against multiple species of African trypano-

somes. Whereas human African trypanosomiasis is caused only by

427. The ﬁrst line,

DTbat1, was derived by the disruption of both

alleles coding for the TbAT1/P2 transporter [46]. The TbAT1/P2

transporter is the main transport protein mediating uptake of the

diamidine and melaminophenyl arsenical classes of trypanocides

[47,48] and the genetic deletion of the encoding gene thus creates

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

949

Table 2

statistically signiﬁcant (n ¼ 39, P < 0.2, Wilcoxon’s signed ranks test

for two groups, paired observations). Eight compounds, 4, 5, 7, 12,

20, 23, 48 and 49, and one mixture, 41a þ 41b, displayed in vitro

activity against the amastigote stage that was superior to the

Comparison of the activity of curcuminoids on T. b. brucei WT and T. evansi.

Compound

T. b. brucei WT

T. evansi

Student’s T-test

1

4

12

13

23

24

26

40

2.5 ꢀ 0.4

0.87 ꢀ 0.06

2.4 ꢀ 0.1

2.0 ꢀ 0.3

NS

P < 0.05

P < 0.01

NS

P < 0.02

NS

1.1 ꢀ 0.2

1.6 ꢀ 0.2

clinically used drug pentamidine (EC₅₀¼ 16 ꢀ 2 M; n ¼ 6) whereas

m

compound 36 and curcumin (1) were as active as pentamidine (see

Table 1).

1.6 ꢀ 0.1

5.6 ꢀ 0.7

3.2 ꢀ 0.4

2.5 ꢀ 0.6

It seemed that inﬂuence of polarity from the oxygen function on

the aromatic ring(s) has varying effects on antileishmanial activity.

This was exempliﬁed in the case of the parent curcuminoid 1 which

exhibited antileishmanial activity (against promastigotes) of

3.1 ꢀ 0.7

1.5 ꢀ 0.1

3.7 ꢀ 0.4

2.9 ꢀ 0.2

0.053 ꢀ 0.007

0.75 ꢀ 0.13

0.14 ꢀ 0.05

0.22 ꢀ 0.09

0.12 ꢀ 0.01

0.17 ꢀ 0.02

0.99 ꢀ 0.19

0.29 ꢀ 0.04

0.21 ꢀ 0.02

0.018 ꢀ 0.003

41

43

48

33 ꢀ 4

mM, whereas its mono-O-demethylated analog 4 was almost

NS

P < 0.001

8-fold more active. However, going from 4 to the more polar analog 5,

the di-O-demethylated analog of 1, resulted in a 6-fold decrease in

activity. The increase in lipophilicity in going from 1 to the di-O-

methylated analog 7 resulted in w12-fold increase in activity (see

Table 1). A larger O-alkyl group which increased the lipophilicity of

the molecule also improved the antileishmanial activity of the

analogs. For example, an increase of 5.7 and 4.3-fold in activity was

noted for the mono-O-alkyl analogs 12 and 14, respectively,

compared to the parent curcuminoid 1. However, addition of the

second alkyl group to the analogs 12 and 14 to give the corre-

sponding dialkyl analogs 13 and 15 caused sharp decrease in activity.

The conjugated keto system is also required for a curcuminoid

analog to exhibit high antileishmanial activity. However, analogs

with a conjugated diketo system (e.g. 4, 7), as well as analogs with

the conjugated monoketo system (enones, 48–54), displayed

similarly high antileishmanial activity.

Diminazene

Data are the average of 3–4 independent experiments, given as EC₅₀values in

M ꢀ S.E.M. NS ¼ not signiﬁcant.

m

T. brucei subspecies [6], the infection in domestic and wild animals

is caused by several related species, including T. evansi.

3.2. Antileishmanial activity of curcuminoid analogs

The same curcuminoids and analogs were tested on promasti-

gotes of L. major (Table 1). The published EC₅₀value for curcumin

against these cells is 37.6 ꢀ 3.5

mM [22]. Once again, our own value

at 33 ꢀ 4

m

M (n ¼ 3) is in very good agreement with this value. Most

analogs displayed higher activity than curcumin (1) against the

promastigotes. However, the activity was generally below that

displayed for T. b. brucei or T. evansi. For L. major promastigotes, four

curcuminoid analogs, 4, 7, 48 and 50a/50b, exhibited anti-

3.3. Effect of curcuminoid analogs on a human cell line

leishmanial activity with EC₅₀less than 5

minoid analogs, 12, 14, 19a, 24, 40, 46, 49, 51, 53a/53b, and 54,

showed activity within the EC₅₀range of 5–10 M. The remaining

compounds displayed IC₅₀values higher than 10 M. The most

active compounds, 7 and 48, exhibited activity of 2.8 ꢀ 0.4 and

2.7 ꢀ 0.7 M, respectively (Table 1). The compounds were also

mM, whereas ten curcu-

To assess whether the antiprotozoal activity described above

should be attributed to general toxicity, rather than speciﬁc anti-

protozoal activity, the analogs were also tested for their effect on

human embryonic kidney (HEK) cells (see Table 1). Without

exception, the toxicity to HEK cells was lower than toT. b. brucei WT

bloodstream forms. Out of 46 analogs and 8 pairs of two isomers,

only one compound, 48, had signiﬁcantly higher activity against

m

assessed against axenic Leishmania mexicana amastigotes and the

same analogs showed highest antileishmanial activity (4, 48, 49)

and had also displayed submicromolar activity against T. b. brucei

(see above). This suggests that these compounds act on a well-

conserved target in the kinetoplastids. However, the best per-

forming compound against T. b. brucei, the enone 40, performed

relatively poorly against L. mexicana amastigotes, with an EC₅₀

HEK cells than curcumin (1) itself (EC₅₀¼ 37 ꢀ 6

m

M) and was the

only compound that displayed an EC₅₀value below 20

mM. The ratio

of EC₅₀(HEK)/EC₅₀(T. b. brucei) could be described as the in vitro

selectivity index and ranged from w3-fold for compound 57 to

>1000-fold for compound 15 and 453-fold for the most active

analog, 40.

As the leishmanicidal activity of most analogs was less than

their trypanocidal activity, the therapeutic index calculated from

value of 17 ꢀ 2

mM. The general trend was that the curcuminoid

analogs were, on average, very slightly less active against the axenic

amastigotes than against promastigotes, but this was not

150

A

B

150

100

50

0

100

50

0

-10

-8

-6

-4

0

-10

-8

-6

-4

log[test compound] (M)

Fig. 1. Effect of some curcuminoid analogs on Trypanosoma brucei and Trypanosoma evansi. Curcuminoid analogs and a control drug (diminazene aceturate) were tested for effect

against the two trypanosome species in vitro, using the Alamar Blue protocol described in the Experimental section. A: T. b. brucei WT. B: T. evansi. (-) diminazene aceturate; (^B)

compound 40; (:) compound 1. Data are representative of 3–4 independent experiments.

950

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

the L. major EC₅₀values was correspondingly lower. The most

promising compound, by this measure was 7, with an index of just

over 70-fold.

curcuminoids, this was also observed for the B48 strain, which is

much more highly resistant to these drugs and has lost an addi-

tional drug transporter, HAPT [49]. Unexpectedly, some of the

curcumin analogs, bearing the enone motif, were much more active

against the B48 line. The reason for this is unclear at this moment,

but it can be speculated that the loss of both the P2/TbAT1 and

HAPT transporters could induce, in compensation, a higher level of

expression for another transport mechanism, capable of taking up

curcuminoids with this particular structural motif. Thus, a curcu-

min-based trypanocide would be at least equally effective against

strains sensitive and resistant to current trypanocides. In addition,

it appears that the analogs have similar activities for different

Trypanosoma and Leishmania species (compare T. b. brucei with T.

evansi and L. major with L. mexicana). This feature is of great

practical importance for the treatment of leishmaniasis or veteri-

nary trypanosomiasis, as it is often unknown which species has

caused the infection.

In conclusion, we have identiﬁed a class of curcuminoids with

potential applications against trypanosomiasis and/or leishmaniasis,

based on their strong antiparasitic activity and low toxicity. The

compounds compare, at least in vitro, favorably with existing

medications. In vivo experimentation, as well as further in vitro

testing of additional analogs, is expected to identify serious drug

candidates against these diseases.

4. Conclusion

The natural curcuminoids curcumin (1), demethoxycurcumin

(2) and bisdemethoxycurcumin (3) isolated from C. longa L. were

assessed against Trypanosoma and Leishmania species. Trypano-

soma species cause human African trypanosomiasis or sleeping

sickness, as well as the corresponding veterinary infections known

as nagana and surra. In this study we used three isogenic clonal

lines of increasing drug resistance. The T. b. brucei clonal line 427

was chosen as drug-sensitive standard (‘WT’). From this line,

a diminazene-resistant clone was derived by deletion of the TbAT1

gene that encodes the P2 aminopurine transporter (‘KO’) [46]. In

vitro exposure to increasing concentrations of pentamidine yielded

a new clonal line [49] highly resistant to the diamidine and mela-

minophenyl arsenical drugs that currently form the mainstay of

trypanosomiasis treatment [3]. This allowed us to not only assess

activity against a well-characterized line of this parasite, but also to

determine any issues about cross-resistance from the outset. For

the Leishmania species, L. major promastigotes and L. mexicana

amastigotes, have been used.

The natural curcuminoids exhibited fairly low antitrypanosomal

activity when compared with the drug diminazene aceturate. They

similarly showed somewhat lower antileishmanial activity than the

reference drug pentamidine. The natural curcuminoids 1–3 were

thus subjected to a series of chemical modiﬁcations to the corre-

sponding demethylated, methylated and higher alkylated analogs,

acetate and benzoate esters, tetrahydro and hexahydro analogs,

conjugated and non-conjugated monoketo analogs including their

demethylated, methylated and higher alkylated analogs, and

acetate derivatives. Out of 43 curcuminoid analogs and 8 pairs of

1:1 mixture of isomeric curcuminoid analogs tested, thirteen

analogs exhibited submicromolar antitrypanosomal activity,

including three with EC₅₀values below 100 nM (compounds 40,

53a/53b and 54). These last three share an enone motif that,

although not present in all the compounds with submicromolar

activity (e.g. 6, 7, 36), may be one structural motif associated with

particularly high trypanocidal activity. While the three most active

compounds displayed identical activity on HEK cells and thus

a similar in vitro selectivity index (250–500), we consider 40 as the

most promising lead compound because of its better solubility and

simpler synthetic route.

5. Experimental

5.1. General

Melting points were determined on an Electrothermal melting

point apparatus and are uncorrected. IR spectra were recorded in KBr

on a Perkin–Elmer FT-IR Spectrum BX spectrophotometer. ¹H NMR

spectra were recorded on a Bruker AVANCE 400 spectrometer.

Electron impact (EI) and electrospray (ES) mass spectra were

obtained using a Finnigan Polaris Q and a Finnigan LC-Q mass

spectrometers, respectively. 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 F₂₅₄plates, respectively.

Spots onTLC were detected under UV light (254 nm) and by spraying

with anisaldehyde–H₂SO₄reagent followed by heating.

5.2. Isolation and identiﬁcation of curcuminoids

The crude curcuminoids from Thai-China Flavours and

Fragrances (TCFF) were subjected to column chromatography using

silica gel as the adsorbent (CH₂Cl₂–MeOH, gradient elution) resul-

ted in the isolation of curcumin (1), demethoxycurcumin (2) and

bisdemethoxycurcumin (3) in 60.8, 16.2 and 3.0% yields, respec-

tively. The ratios of the curcuminoids 1, 2 and 3 are 76.0, 20.2 and

3.8 respectively. The spectroscopic (IR, ¹H NMR and mass spectra)

data were consistent with the reported values [23].

It should be noted that, although in many cases trypanocidal and

antileishmanial activities were linked, the association is not abso-

lute and of the four compounds displaying EC₅₀values below 5 mM

against promastigotes, two were diketones (4 and 7) and two were

enones (48 and 50a/50b). The situation with amastigotes was

similar. Nor were the conjugated ketones inherently more toxic to

human embryonic kidney (HEK) cells (e.g. 42). Some of the analogs

displayed no detectable effect on HEK until the limit of solubility

and compound 40 displayed a selectivity index of 453 using the

EC₅₀against WT trypanosomes and >1000 against B48. While none

of the curcuminoids displayed submicromolar activity against the

Leishmania promastigotes or amastigotes, 11 of them displayed

activity better or equal to the reference drug pentamidine, which is

in common clinical use [4].

Importantly, the activities against the WT and KO lines of T. b.

brucei were highly similar. This clearly shows that the activity of

this class of compounds is not dependent on the TbAT1/P2 amino-

purine transporter that has beenwidely associated with resistance to

some diamidines particularly diminazene [47] and furamidine [52]

and melaminophenyl arsenicals [3,50]. For most of the

5.3. Chemical modiﬁcations of curcuminoids

5.3.1. Demethylation of curcumin (1)

A solution of compound 1 (200 mg, 0.54 mmol) in dry CH₂Cl₂

(45 ml) was stirred at 0 ^ꢁC for 5 min and BBr₃(1.5 ml) was slowly

added. The reaction mixture was allowed to warm up to ambient

temperature and stirring was continued for 10 h. Water was added

and the mixture was extracted with EtOAc. The combined organic

phase was washed with H₂O, dried over anhydrous Na₂SO₄; the

solvent was evaporated and the residue chromatographed using

CH₂Cl₂–MeOH–AcOH (50:5:2) to yield mono-O-demethylcurcumin

(4) (80 mg, 43%) as orange amorphous solid, m.p. 168–170 ^ꢁC (lit.

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

951

[23] m.p.164–166 ^ꢁC) and di-O-demethylcurcumin (5) (60 mg, 33%)

as orange plates, m.p. 301–303 ^ꢁC (lit. [23] m.p. 302–304 ^ꢁC). The

spectroscopic (IR, ¹H NMR and mass spectra) data were consistent

with the reported values [23].

and H-6⁰⁰), 7.58 (d, J ¼ 15.7 Hz, 2H, H-1 and H-7); ESMS (þve): m/z

(% rel. abund.) 453 [M þ H]^þ(100).

5.3.7. Allylation of curcumin (1)

Curcumin (1) was subjected to allylation in similar manner to

that of methylation of compound 1, but using allyl bromide in place

of methyl iodide, to afford mono-O-allylcurcumin (12) in 33% and

di-O-allylcurcumin (13) [27] in 46%.

5.3.2. Demethylation of demethoxycurcumin (2)

Compound 2 was subjected to demethylation in similar manner

to that of compound 1 to give O-demethyldemethoxycurcumin (6)

(64%) as orange amorphous solid, m.p. 216–218 ^ꢁC (lit. [23]

218–220 ^ꢁC). The spectroscopic (IR, ¹H NMR and mass spectra) data

were consistent with the reported values [23].

Compound 12: Orange amorphous solid, m.p. 132–133 ^ꢁC; IR

n_max: 3293, 2927, 1623, 1587, 1512, 1256, 1135, 967, 840 cm^{ꢂ1 1}H

;

NMR (CDCl₃, 400 MHz):

d

3.90 and 3.92 (each s, 2 ꢃ 3H, 2 ꢃ OMe),

4.63 (br d, J ¼ 5.4 Hz, 2H, H-1%), 5.30 (br dd, J ¼ 10.5, 1.3 Hz, 1H, H-

3a%), 5.40 (br dd, J ¼ 17.2 Hz, 1.3, 1H, H-3b%), 5.78 (s, 1H, H-4), 5.85

(s, 1H, OH), 6.05 (m, 1H, H-2%), 6.45 and 6.47 (each d, J ¼ 15.7 Hz,

2 ꢃ 1H, H-2 and H-6), 6.86 (d, J ¼ 8.2 Hz, 1H, H-5⁰⁰), 6.91 (d,

J ¼ 8.2 Hz, 1H, H-5⁰), 7.02 and 7.06 (each d, J ¼ 1.6 Hz, 2 ꢃ1H, H-2⁰

and H-2⁰⁰), 7.09 and 7.11 (each dd, J ¼ 8.2 Hz, 1.6, 2 ꢃ 1H, H-6⁰and H-

6⁰⁰), 7.58 (d, J ¼ 15.7 Hz, 2 ꢃ1H, H-1 and H-7); EIMS: m/z (% rel.

abund.) 408 [M]^þ(12), 390 (27), 367 (5),178 (69),177 (61),139 (100).

Compound 13: Yellow needles (from n-hexane–CH₂Cl₂), m.p.

117–118 ^ꢁC; IR n_max: 2928, 1621, 1595, 1578, 1510, 1458, 1423, 1336,

5.3.3. Methylation of curcumin (1)

Compound 1 (50 mg, 0.14 mmol) was dissolved in dry acetone

(2 ml) and anhydrous K₂CO₃(40 mg) and MeI (0.6 ml) was 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 Na₂SO₄and the solvent

removed under vacuum. The crude products were puriﬁed by

column chromatography using CH₂Cl₂to yield di-O-methyl-

curcumin (7) (45 mg, 80%) as orange powder, m.p. 128–130 ^ꢁC (lit.

[24] m.p. 129–130 ^ꢁC). The spectroscopic (IR, ¹H NMR and mass

spectra) data were consistent with the reported values [24].

1254, 1232, 1132, 1011, 976, 845 cm^ꢂ1

;

¹H NMR (CDCl₃, 400 MHz):

d

3.91 (s, 6H, 2 ꢃ OMe), 4.64 (br d, J ¼ 5.3 Hz, 4H, H-1% and H-1&),

5.30 (br dd, J ¼ 10.5, 1.3 Hz, 2H, H-3a% and H-3a&), 5.40 (br dd,

J ¼ 17.2, 1.3 Hz, 2H, H-3b% and H-3b&), 5.79 (s, 1H, H-4), 6.06

(m, 2H, H-2% and H-2&), 6.47 (d, J ¼ 15.7 Hz, 2H, H-2 and H-6), 6.86

(d, J ¼ 8.2 Hz, 2H, H-5⁰and H5⁰⁰), 7.06 (d, J ¼ 1.7 Hz, 2H, H-2⁰and H-

2⁰⁰), 7.08 (dd, J ¼ 8.2, 1.7 Hz, 2H, H-6⁰and H-6⁰⁰), 7.57 (d, J ¼ 15.7 Hz,

2H, H-1 and H-7); EIMS: m/z (% rel. abund.) 448 [M]^þ(25), 430 (80),

407 (12), 389 (100), 230 (56), 217 (63), 189 (46), 177 (80).

5.3.4. Methylation of demethoxycurcumin (2)

Compound 2 was subjected to methylation in similar manner to

that of compound 1 to give 4⁰⁰-O-methyldemethoxycurcumin (8)

(20%) and di-O-methyldemethoxycurcumin (9) (54%).

Compound 8: Orange foam; IR n_max: 3420,1640,1585,1508,1263,

1136, 1020, 963, 832 cm^ꢂ^{1 1}H NMR (CDCl₃, 400 MHz):

; d 3.89 and

3.91 (each s, 2 ꢃ 3H, 2 ꢃ OMe), 5.78 (s, 1H, H-4), 6.46 and 6.47 (each

d, J ¼ 15.7 Hz, 2 ꢃ1H, H-2 and H-6), 6.84 (br d, J ¼ 8.2 Hz, 2H, H-3⁰

and H-5⁰), 6.85 (d, J ¼ 8.3 Hz, 1H, H-5⁰⁰), 7.05 (br s, 1H, H-2⁰⁰), 7.11

(br d, J ¼ 8.3 Hz, 1H, H-6⁰⁰), 7.43 (d, J ¼ 8.3 Hz, 2H, H-2⁰and H-6⁰),

7.58 (d, J ¼ 15.7 Hz, 2H, H-1 and H-7); ESMS (þve): m/z (% rel.

abund.) 727 [2M þ Na]^þ(100); HR-TOFMS (APCI^þ): m/z 353.1373

[M þ H]^þ; calcd for C₂₃H₂₄O₇þ H, 353.1384.

5.3.8. Dimethylallylation of curcumin (1)

Compound 1 was subjected to alkylation in similar manner to

that of methylation of compound 1, but using 3,3-dimethylallyl

bromide in place of methyl iodide, to afford mono-O-(3,3-dimethyl-

allyl)curcumin (14) in 43% as orange gum, and di-O-(3,3-dimethyl-

allyl)curcumin (15) in 40% as yellow solid, m.p. 123–124 ^ꢁC (lit. [26]

m.p. 124–125 ^ꢁC). The spectroscopic (IR, ¹H NMR and mass spectra)

data of the latter product were consistent with the reported values

[26].

Compound 9: Yellow needles (from n-hexane–CH₂Cl₂), m.p. 159–

161 ^ꢁC; IR n_max: 2937, 1624, 1601, 1515, 1458, 1256, 1138, 1027, 968,

840 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

d 3.83, 3.90 and 3.91 (each s,

3 ꢃ 3H, 3 ꢃ OMe), 5.78 (s, 1H, H-4), 6.48 (d, J ¼ 15.7 Hz, 2H, H-2 and

H-6), 6.86 (d, J ¼ 8.2 Hz,1H, H-5⁰⁰), 6.90 (br d, J ¼ 8.5 Hz, 2H, H-3⁰and

H-5⁰), 7.06 (br s, 1H, H-2⁰⁰), 7.12 (br d, J ¼ 8.2 Hz, 1H, H-6⁰⁰), 7.49 (d,

J ¼ 8.5 Hz, 2H, H-2⁰and H-6⁰), 7.58 and 7.60 (each d, J ¼ 15.7 Hz, 2H,

H-1 and H-7); ESMS (þve): m/z (% rel. abund.) 367 [M þ H]^þ(100).

Compound 14: IR n_max: 3413, 2927, 1624, 1582, 1508, 1261, 1134,

968 cm^ꢂ^{1 1}H NMR (CDCl₃, 400 MHz):

; d 1.72 and 1.76 (each s,

2 ꢃ 3H, 2 ꢃ Me) 3.89 and 3.92 (each s, 2 ꢃ 3H, 2 ꢃ OMe), 4.60 (br d,

J ¼ 6.4 Hz, 2H, H-1%), 5.49 (m, 1H, H-2%), 5.78 (s, 1H, H-4), 5.85 (brs,

1H, OH), 6.45 and 6.46 (each d, J ¼ 15.7 Hz, 2 ꢃ 1H, H-2 and H-6),

6.85 (d, J ¼ 8.2 Hz, 1H, H-5⁰⁰), 6.91 (d, J ¼ 8.2 Hz, 1H, H-5⁰), 7.02 and

7.05 (each m, 2 ꢃ1H, H-2⁰and H-2⁰⁰), 7.09 (br d, J ¼ 8.2 Hz, 2 ꢃ 1H, H-

6⁰and H-6⁰⁰), 7.56 and 7.58 (each d, J ¼ 15.7 Hz, 2 ꢃ 1H, H-1 and H-7);

ESMS (þve): m/z (% rel. abund.) 437 [M þ H]^þ(100); HR-TOFMS

(ES^þ): m/z 437.1982 [M þ H]^þ; calcd for C₂₆H₂₈O₆þ H, 437.1964.

5.3.5. Methylation of bisdemethoxycurcumin (3)

Compound 3 was subjected to methylation in similar manner to

that of compound 1 to give mono-O-methyldemethoxycurcumin

(10) (27%) as orange solid, m.p. 188–189 ^ꢁC. The spectroscopic (¹H

NMR and mass spectra) data of the later product were consistent

with the reported values [25].

5.3.9. Dimethylallylation of demethoxycurcumin (2)

Compound 2 was subjected to alkylation in similar manner to

that of methylation of compound 1, but using 3,3-dimethylallyl

bromide in place of methyl iodide, to afford di-O-(3,3-dimethy-

lallyl)demethoxycurcumin (16) in 43% as yellow aggregated nee-

dles (from n-hexane–CH₂Cl₂), m.p. 107–109 ^ꢁC.

5.3.6. Propylation of curcumin (1)

Compound 1 was subjected to n-propylation in similar manner

to that of methylation of compound 1, but using n-propyl iodide in

place of methyl iodide, to give di-O-n-propylcurcumin (11) (85%) as

yellow powder, m.p. 127 ^ꢁC.

Compound 16: IR n_max: 2920, 1623, 1600, 1509, 1254, 1170, 1134,

Compound 11: IR n_max: 2939, 2877, 1620, 1595, 1578, 1511, 1468,

971, 829 cm^ꢂ^{1 1}H NMR (CDCl₃, 400 MHz):

; d 1.72, 1.76, 1.78, 1.79

1423, 1334, 1273, 1234, 1131, 1011, 974 cm^ꢂ1

;

¹H NMR (CDCl₃,

(each s, 4 ꢃ 3H, 4 ꢃ Me), 3.89 (s, 3H, OMe), 4.52 (br d, J ¼ 6.4 Hz, 2H)

and 4.60 (br d, J ¼ 6.4 Hz, 2H) (H-1% and H-1&), 5.47 (m, 2H, H-2%

and H-2&), 5.77 (s, 1H, H-4), 6.47 (d, J ¼ 15.7 Hz, 2H, H-2 and H-6),

6.83 (d, J ¼ 7.9 Hz, 1H, H-5⁰⁰), 6.90 (br d, J ¼ 8.5 Hz, 2H, H-3⁰and H-

5⁰), 7.05 (br s, 1H, H-2⁰⁰), 7.11 (d, J ¼ 7.9, 1H, H-6⁰⁰), 7.47 (br d,

J ¼ 8.5 Hz, 2H, H-2⁰and H-6⁰), 7.57 and 7.59 (each d, J ¼ 15.7 Hz,

400 MHz):

d

1.03 (t, J ¼ 7.3 Hz, 2 ꢃ 3H, H-3% and H-3&),1.86 (sextet,

J ¼ 7.1 Hz, 2 ꢃ 2H, H-2% and H-2&), 3.89 (s, 6H, 2 ꢃ OMe), 4.00

(t, J ¼ 6.7 Hz, 4H, H-1% and H-1&), 5.79 (s, 1H, H-4), 6.46

(d, J ¼ 15.7 Hz, 2H, H-2 and H-6), 6.85 (d, J ¼ 8.4 Hz, 2 ꢃ 1H, H-5⁰and

H-5⁰⁰), 7.06 (br s, 2H, H-2⁰and H-2⁰⁰), 7.14 (br d J ¼ 8.4 Hz, 2H, H-6⁰

952

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

2 ꢃ1H, H-1 and H-7); ESMS (ꢂve): m/z (% rel. abund.) 473 [M ꢂ H]^ꢂ

(100); HR-TOFMS (ES^þ): m/z 497.2307 [M þ Na]^þ; calcd for

C₃₀H₃₄O₅þ Na, 497.2304.

(d, J ¼ 8.2, 1H, H-5⁰⁰), 7.03 (d, J ¼ 1.5 Hz, 1H, H-2⁰⁰), 7.10 (dd, J ¼ 8.2,

1.5 Hz, 1H, H-6⁰⁰), 7.25 (obscured signal, H-2⁰), 7.50 (d, J ¼ 15.7 Hz,

1H, H-7), 7.58 (d, J ¼ 15.7 Hz, 1H, H-1), 7.87 (d, J ¼ 1.6 Hz, 1H, H-6⁰);

ESMS (ꢂve): m/z (% rel. abund.) 412 [M ꢂ H]^ꢂ(100); HR-TOFMS

(APCI^þ): m/z 414.1168 [M þ H]^þ; calcd for C₂₃H₂₄O₇þ H, 414.1183.

5.3.10. Hydroxyethylation of curcumin (1)

Compound 1 was subjected to hydroxyethylation in similar

manner to that of methylation of compound 1, but using 2-bro-

moethanol in place of methyl iodide, to afford mono-O-(2-hydroxy-

ethyl)curcumin (17) and di-O-(2-hydroxyethyl)curcumin (18) in 47

and 32%.

5.3.13. Acetylation of curcumin (1)

Ac₂O (0.5 ml) was added to a solution of compound 1 (100 mg,

0.27 mmol) in pyridine (2 ml) and the reaction mixture was stirred

at ambient temperature for 3 h. After the usual work up, the

mixture was extracted with CH₂Cl₂and the organic phase was

washed with H₂O, dried over anhydrous Na₂SO₄and the solvent

evaporated to dryness. The crude product was puriﬁed by column

chromatography eluting with CH₂Cl₂to afford di-O-acetylcurcumin

(21) (95 mg, 85%) as yellow amorphous solid; m.p. 163–165 ^ꢁC (lit.

[28] 160 ^ꢁC). The spectroscopic (IR, ¹H NMR and mass spectra) data

of compound 21 were consistent with the reported values [28].

Compound 17: Yellow amorphous solid, m.p.154–156 ^ꢁC; IR n_max

3449, 2927, 1623, 1508, 1271, 1135, 1028, 960, 847 cm^ꢂ^{1 1}H NMR

(CDCl₃þ 1 drop of CD₃OD, 400 MHz): 3.92 (br s, 6H, 2 ꢃ OMe),

:

;

d

3.92 (obscured signal, 2H, H-2%), 4.11 (br s, 2H, H-1%), 5.78 (br s,

1H, H-4), 6.45 and 6.47 (each d, J ¼ 15.3 Hz, 2H, H-2 and H-6), 6.89

(m, 2H, H-5⁰and H-5⁰⁰), 7.02–7.10 (m, 4H, H-2⁰, H-2⁰⁰, H-6⁰, H-6⁰⁰),

7.56 (d, J ¼ 15.6 Hz, 2H, H-1 and H-7); ESMS (ꢂve): m/z (% rel.

abund.) 411 [M ꢂ H]^ꢂ(100); HR-TOFMS (ES^þ): m/z 435.1420

[M þ Na]^þ; calcd for C₂₃H₂₄O₇þ Na, 435.1418.

5.3.14. Acetylation of demethoxycurcumin (2)

Compound 18: Orange needles (from CH₂Cl₂–MeOH), m.p. 174–

175 ^ꢁC; IR n_max: 3538, 2934, 1623, 1585, 1509, 1455, 1423, 1309, 1261,

Compound 2 was subjected to acetylation in the same manner

to that of compound 1 to give di-O-acetyldemethoxycurcumin (22)

in 89% yield. The spectroscopic (IR, ¹H NMR and mass spectra) data

of compound 22 were consistent with the reported values [29].

Compound 22: Yellow aggregated needles (from n-hexane–

CH₂Cl₂), m.p. 140–141 ^ꢁC; IR n_max: 2942, 1760, 1634, 1597, 1559,

1134, 1083, 1031, 973, 838, 805 cm^ꢂ1

;

¹H NMR (CDCl₃þ 2 drops of

CD₃OD, 400 MHz):

d

3.90 (br s, 6H, 2 ꢃ OMe), 3.90 (obscured signal,

4H, H-2% and H-2&), 4.07 (br s, 4H, H-1% and H-1&), 5.78 (s, 1H, H-

4), 6.46 (d, J ¼ 15.7 Hz, 2H, H-2 and H-6), 6.85 (d, J ¼ 7.8 Hz, 2H, H-5⁰

and H-5⁰⁰), 7.04 (br s, 2H, H-2⁰and H-2⁰⁰), 7.15 (br d, J ¼ 7.8 Hz, 2H, H-6⁰

and H-6⁰⁰), 7.54 (d, J ¼ 15.7 Hz, 2H, H-1 and H-7); ESMS (ꢂve): m/z (%

rel. abund.) 455 [M ꢂ H]^ꢂ(100); HR-TOFMS (APCI^þ): m/z 457.1843

[M þ H]^þ; calcd for C₂₅H₂₈O₈þ H, 457.1857.

1541, 1508, 1368, 1257, 1201, 1014, 969, 835 cm^ꢂ1

;

¹H NMR (CDCl₃,

400 MHz):

d

2.29 and 2.31 (each s, 2 ꢃ 3H, 2 ꢃ OAc), 3.86 (s, 3H,

OMe), 5.82 (s,1H, H-4), 6.54 and 6.55 (each d, J ¼ 15.8 Hz, 2 ꢃ 1H, H-

2 and H-6), 7.04 (d, J ¼ 8.1 Hz, 1H, H-5⁰⁰), 7.11 (d, J ¼ 8.5 Hz, 2H, H-3⁰

and H-5⁰), 7.13 (partially obscured signal, 1H, H-6⁰⁰), 7.55 (d,

J ¼ 8.4 Hz, 2H, H-2⁰and H-6⁰), 7.60 and 7.62 (each d, J ¼ 15.8 Hz, 2H,

H-1 and H-7); ESMS (þve): m/z (% rel. abund.) 445 [M þ Na]^þ(34),

423 [M þ H]^þ(100).

5.3.11. Hydroxyethylation of demethoxycurcumin (2)

Following the same procedure for the preparation of compounds

17 and 18 from compound 1, the mono-4⁰⁰-O-(2-hydrox-

yethyl)demethoxycurcumin (19a) and its isomeric compound mono-

4⁰-O-(2-hydroxyethyl)demethoxycurcumin (19b) were obtained.

Careful repeated column chromatography followed by recrystalli-

zation resulted in the separation and puriﬁcation of compound 19a

in 28%. However, attempt to purify the contaminated 19b was

unsuccessful.

5.3.15. Acetylation of bisdemethoxycurcumin (3)

Compound 3 was subjected to acetylation in the same manner

to that of compound 1 to give di-O-acetylbisdemethoxycurcumin

(23) in 87% yield as yellow needles (from n-hexane–CH₂Cl₂), m.p.

172–174 ^ꢁC. The spectroscopic (IR, ¹H NMR and mass spectra) data

of compound 23 were consistent with the reported values [30].

Compound 19a: Yellow solid, m.p. 169–171 ^ꢁC; IR n_max: 3507,

2934, 1619, 1594, 1559, 1508, 1273, 1136, 970 cm^ꢂ^{1 1}H NMR

;

(CDCl₃þ 4 drops of CD₃OD, 400 MHz):

d

3.90 (s, 3H, OMe), 3.91 (m,

5.3.16. Benzoylation of curcumin (1)

2H, H-2%), 4.08 (m, 2H, H-1%), 5.75 (br s, 1H, H-4), 6.43 and 6.46

(each d, J ¼ 15.7 Hz, 2 ꢃ1H, H-2 and H-6), 6.80 (d, J ¼ 8.3 Hz, 2H, H-

3⁰and H-5⁰), 6.85 (d, J ¼ 8.1 Hz, 1H, H-5⁰⁰), 7.05 (br s, 1H, H-2⁰), 7.10

(d, J ¼ 8.1 Hz, 1H, H-6⁰⁰), 7.40 (d, J ¼ 8.3 Hz, H-2⁰and H-6⁰), 7.54 and

7.56 (each d, J ¼ 15.7 Hz, 2H, H-1 and H-7); ESMS (ꢂve): m/z (% rel.

abund.) 381 [M ꢂ H]^ꢂ(100); HR-TOFMS (ES^ꢂ): m/z 381.1320

[M ꢂ H]^ꢂ; calcd for C₂₂H₂₂O₆–H, 381.1338.

A solution of compound 1 (175 mg, 0.48 mmol) in pyridine

(5 ml) was treated with benzoic anhydride (90 mg) and the mixture

was stirred at ambient temperature for 4 h. The reaction was

worked up with H₂O and extracted with EtOAc; the organic phase

was washed with H₂O, dried over anhydrous Na₂SO₄and the

solvent was evaporated to dryness. The crude products were

separated and puriﬁed by column chromatography using CH₂Cl₂–

MeOH (200:1) to give mono-O-benzoylcurcumin (24) (105 mg,

47%) and di-O-benzoylcurcumin (25) (90 mg, 33%). The spectro-

scopic (IR, ¹H NMR and mass spectra) data of the dibenzoate 25

were consistent with the reported values [31].

5.3.12. Nitration of curcumin (1)

Compound 1 (300 mg, 0.81 mmol) was dissolved in CH₂Cl₂and

NaNO₂(100 mg) was added. The reaction mixture was stirred at

ambient temperature for 5 min, and 3–4 drops of AcOH was added.

The reaction mixture was stirred for another 30 min; the reaction

was worked up by H₂O and extracted with EtOAc. The organic

phase was washed with H₂O, dried over anhydrous Na₂SO₄and the

solvent was evaporated to dryness. The crude product was puriﬁed

by column chromatography using CH₂Cl₂–MeOH (200:2) to yield

5⁰-nitrocurcumin (20) (100 mg, 27%).

Compound 24: Yellow needles (from n-hexane–CH₂Cl₂), m.p.

134–136 ^ꢁC; IR n_max: 3448, 2920, 1735, 1628, 1560, 1508, 1458, 1265,

1204, 1123, 1024, 967, 843, 707 cm^ꢂ^{1 1}H NMR (CDCl₃, 400 MHz):

;

d

3.84 and 3.92 (each s, 2 ꢃ 3H, 2 ꢃ OMe), 5.82 (s, 1H, H-4), 6.47 and

6.55 (each d, J ¼ 15.7 Hz, 2 ꢃ 1H, H-2 and H-6) 6.91 (d, J ¼ 8.1 Hz,1H,

H-5⁰⁰), 7.03 (br s, 1H, H-2⁰⁰), 7.10 (br d, J ¼ 8.1 Hz, 1H, H-6⁰⁰), 7.14–7.19

(m, 3H, H-2⁰, H-5⁰, H-6⁰), 7.49 (t, J ¼ 7.4 Hz, 2H, m-ArH), 7.57–7.64

(m, 3H, H-1, H-7, p-ArH), 8.20 (d, J ¼ 7.4 Hz, 2H, o-ArH); ESMS (þve):

m/z (% rel. abund.) 473 [M þ H]^þ(100); HR-TOFMS (ES^þ): m/z

495.1436 [M þ Na]^þ; calcd for C₂₈H₂₄O₇þ Na, 495.1420.

Compound 20: Orange amorphous solid, m.p. 198 ^ꢁC (dec.); IR

n_max: 3447, 2924, 1654, 1627, 1601, 1541, 1508, 1458, 1266, 1206,

1137, 1026, 966, 843, 668 cm^ꢂ^{1 1}H NMR (CDCl₃, 400 MHz):

; d 3.92

and 3.98 (each s, 2 ꢃ 3H, 2 ꢃ OMe), 5.81 (s, 1H, OH), 6.46

Compound 25: Yellow amorphous solid, m.p. 228–230 ^ꢁC; IR

n_max: 2920,1734,1653, 1626, 1598, 1541, 1514,1456, 1302, 1269, 1122,

(d, J ¼ 15.7 Hz, 1H, H-6), 6.52 (d, J ¼ 15.7 Hz, 1H, H-2), 6.90

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

953

1082, 1027, 974, 854, 709 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

d

3.85 (s,

5.3.19. Catalytic hydrogenation of curcumin (1)

6H, 2 ꢃ OMe), 5.87 (s,1H, H-4), 6.59 (d, J ¼ 15.8 Hz, 2H, H-2 and H-6),

7.15–7.22 (m, 6H, H-2⁰, H-5⁰, H-6⁰, H-2⁰⁰, H-5⁰⁰, H-6⁰⁰), 7.50 (t, J ¼ 7.6,

4H, m-ArH), 7.63 (t, J ¼ 7.6, 2H, p-ArH), 7.64 (d, J ¼ 15.8 Hz, 2H, H-1

and H-7), 8.20 (d, J ¼ 7.6 Hz, 4H, o-ArH); ESMS (þve): m/z (% rel.

abund.) 599 [M þ Na]^þ(10), 577 [M þ H]^þ(10).

To a solution of compound 1 (100 mg, 0.27 mmol) in EtOH

(10 ml) was added 10% Pd-C (50 mg). After degassing, the mixture

was hydrogenated at room temperature and at atmospheric pres-

sure for 4 h. The mixture was ﬁltered through a column of Celite and

the solvent was evaporated. The residue was puriﬁed by column

chromatography using CH₂Cl₂–MeOH (gradient elution) to afford

tetrahydrocurcumin (31) (65 mg, 64%) as crystals from n-hexane–

CH₂Cl₂, m.p. 93–94 ^ꢁC (lit. [24] m.p. 92–93 ^ꢁC), hexahydrocurcumin

(32) (20 mg, 20%) as white amorphous solid, m.p. 81–82 ^ꢁC (lit. [33]

m.p. 80–82 ^ꢁC), and octahydrocurcumin (33) (15 mg, 15%) as color-

less sticky solid. The spectroscopic (IR, ¹H NMR and mass spectra)

data of compounds 31–33 were consistent with the reported values

[24,33].

5.3.17. Benzoylation of demethoxycurcumin (2)

Compound 2 was subjected to benzoylation in similar manner to

that of compound 1 to give the benzoates 26, 27and 28 in 34%, 29%

and 21% yields respectively.

Compound 26: Yellow amorphous solid, m.p. 183–184 ^ꢁC; IR

n_max: 3411, 1721, 1635, 1593, 1559, 1522, 1419, 1373, 1166, 1123, 1064,

972, 820, 709 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

d 3.93 (s, 3H, OMe),

5.81 (s, 1H, H-4), 5.87 (br s, 1H, OH), 6.48 and 6.57 (each d,

J ¼ 15.8 Hz, 2 ꢃ 1H, H-2 and H-6), 6.91 (d, J ¼ 8.1 Hz, 1H, H-5⁰⁰), 7.04

(br s, 1H, H-2⁰⁰), 7.11 (br d, J ¼ 8.1 Hz, 1H, H-6⁰⁰), 7.25 (d, J ¼ 8.4 Hz,

2H, H-3⁰and H-5⁰), 7.50 (t, J ¼ 7.6 Hz, 2H, m-ArH), 7.57–7.66 (m, 5H,

H-1, H-7, H-2⁰, H-6⁰, p-ArH), 8.19 (d, J ¼ 7.6 Hz, 2H, o-ArH); ESMS

(þve): m/z (% rel. abund.) 443 [M þ H]^þ(100); HR-TOFMS (ES^þ): m/z

465.1319 [M þ Na]^þ; calcd for C₂₇H₂₂O₆þ Na, 465.1314.

5.3.20. Catalytic hydrogenation of demethoxycurcumin (2)

Demethoxycurcumin was subjected to catalytic hydrogenation

in similar manner to that of compound 1 to afford tetrahy-

drodemethoxycurcumin (34) (72%) as colorless sticky solid and

a small quantity of a product, which was supposed to be the hex-

ahydro analog. The spectroscopic (IR, ¹H NMR and mass spectra)

data of compound 34 were consistent with the reported values [34].

Compound 27: Yellow needles (from n-hexane–CH₂Cl₂), m.p.

175–176 ^ꢁC; IR n_max: 3449, 2934, 1728, 1627, 1601, 1512, 1451, 1267,

1169, 1124, 1063, 967, 833, 707 cm^ꢂ1

;

¹H NMR (CDCl₃, 400 MHz):

5.3.21. Catalytic hydrogenation of bisdemethoxycurcumin (3)

Compound 3 was subjected to catalytic hydrogenation in similar

manner to that of compound 1 to give tetrahydrobisdemethoxy-

curcumin (35) (75%) as colorless crystal (from acetone–benzene),

m.p. 110–111 ^ꢁC (lit. [34] m.p. 101–102 ^ꢁC). The spectroscopic (IR, ¹H

NMR and mass spectra) data of compound 35 were consistent with

the reported values [34].

d

3.84 (s, 3H, OMe), 5.81 (s, 1H, H-4), 6.48 and 6.55 (each d,

J ¼ 15.8 Hz, 2x1H, H-2 and H-6), 6.82 (d, J ¼ 8.4 Hz, 2H, H-3⁰and H-

5⁰), 7.12–7.23 (m, 3H, H-2⁰⁰, H-5⁰⁰, H-6⁰⁰), 7.44 (d, J ¼ 8.4 Hz, 2H, H-2⁰

and H-6⁰), 7.50 (t, J ¼ 7.4 Hz, 1H, m-ArH), 7.61 (d, J ¼ 15.8 Hz, 2H, H-1

and H-7), 7.63 (obscured signal, p-ArH), 8.20 (d, J ¼ 7.4, 2H, o-ArH);

ESMS (þve): m/z (% rel. abund.) 465 [M þ Na]^þ(43), 443 [M þ H]^þ

(100); HR-TOFMS (ES^þ): m/z 465.1316 [M þ Na]^þ; calcd for

C₂₇H₂₂O₆þ Na, 465.1314.

5.3.22. Catalytic hydrogenation of mono-O-demethylcurcumin (4)

Compound 4 was subjected to catalytic hydrogenation in similar

manner to that of compound 1 to give mono-O-demethyltetrahy-

drocurcumin (36) (40%) as colorless sticky solid and a mixture of

mono-O-demethylhexahydrocurcumin (37a and 37b) (37%). The

identity of compounds 36, and 37a/37b mixture was conﬁrmed by

comparison of the spectroscopic (IR, ¹H NMR and mass spectra)

data with the reported values [34,35].

Compound 28: Yellow needles (from n-hexane–CH₂Cl₂), m.p.

177–179 ^ꢁC; IR n_max: 2934, 1734, 1629, 1597, 1559, 1508, 1209, 1165,

1060, 1025, 969, 808, 706 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

d 3.85

(s, 3H, OMe), 5.86 (s, 1H, H-4), 6.59 and 6.60 (each d, J ¼ 15.8 Hz,

2 ꢃ1H, H-2 and H-6), 7.16–7.27 (m, 5H, H-3⁰, H-5⁰, H-2⁰⁰, H-5⁰⁰, and

H-6⁰⁰), 7.50 (m, 2H, m-ArH), 7.60–7.69 (m, 6H, H-1, H-7, H-2⁰, H-6⁰

and p-ArH), 8.19 (m, 4H, o-ArH); ESMS (þve): m/z (% rel. abund.)

569 [M þ Na]^þ(100), 547 [M þ H]^þ(55); HR-TOFMS (APCI^þ): m/z

547.1744 [M þ H]^þ; calcd for C₃₄H₂₆O₇þ H, 547.1751.

Compound 36: ¹H NMR (CDCl₃þ1 drop of CD₃OD, 400 MHz):

d

2.33 (m), 2.82 (t, J ¼ 7.0 Hz) and 2.46–2.82 (m) (8H, H-1, H-2, H-6,

H-7), 3.80 (s 3H, OMe), 6.51 (m, 1H, H-6⁰⁰), 6.59–6.63 (m, 3H, H-2⁰,

H-2⁰⁰, H-6⁰), 6.70 (d, J ¼ 8.0 Hz, 1H, H-5⁰), 6.71 (d, J ¼ 7.9 Hz, 1H, H-

5⁰⁰); ESMS (ꢂve): m/z (% rel. abund.) 357 [M ꢂ H]^ꢂ(100).

5.3.18. Benzoylation of bisdemethoxycurcumin (3)

Compound 3 was subjected to benzoylation in similar manner to

that of compound 1 to give the benzoates 29 and 30 in 53 and 31%

yields respectively.

Compounds 37a and 37b (1:1 mixture): ¹H NMR (CDCl₃þ 3 drops

of CD₃OD, 400 MHz):

d

1.57 and 1.68 (each m, 2 ꢃ 2H, H-6a and H-

Compound 29: Yellow needles (from CH₂Cl₂–hexane), m.p.

228–230 ^ꢁC; IR n_max: 3440, 1723, 1635, 1601, 1541, 1438, 1276, 1221,

6b), 2.46–2.77 (m, 2 ꢃ 8H, H-1, H-2, H-4, H-7), 3.79 (s, 3H, OMe),

3.80 (s, 3H, OMe), 3.96 (m, 2 ꢃ1H, H-5), 6.47 (br dd, J ¼ 7.9, 1.8 Hz,

1H, H-6⁰⁰), 6.49 (br dd, J ¼ 7.9, 1.8 Hz, 1H, H-6⁰⁰), 6.57–6.64 (m,

2 ꢃ 3H, H-2⁰, H-2⁰⁰, H-6⁰), 6.69 (d, J ¼ 8.0 Hz, 2 ꢃ1H, H-5⁰), 6.75 (d,

J ¼ 7.9 Hz, 1H, H-5⁰⁰); 6.76 (d, J ¼ 7.9 Hz, 1H, H-5⁰⁰); ESMS (ꢂve): m/z

(% rel. abund.) 359 [M ꢂ H]^ꢂ(100).

1168, 1141, 1066, 972, 837, 707 cm^ꢂ1

;

¹H NMR (CDCl₃þ 2 drops of

CD₃OD, 400 MHz):

d 5.78 (s, 1H, H-4), 6.42 and 6.52 (each d,

J ¼ 15.7 Hz, 2 ꢃ1H, H-2 and H-6), 6.81 (br d, J ¼ 8.4 Hz, 2H, H-3⁰⁰and

H-5⁰⁰), 7.25 (obscured by CHCl₃signal, H-3⁰and H-5⁰), 7.42 (d,

J ¼ 8.4 Hz, 2H, H-2⁰⁰and H-6⁰⁰), 7.49 (m, 2H, m-ArH), 7.57–7.66 (m,

5H, H-1, H-7, H-2⁰, H-6⁰and p-ArH), 8.17 (d, J ¼ 7.6 Hz, 2H, o-ArH);

ESMS (þve): m/z (% rel. abund.) 413 [M þ H]^þ(100); HR-TOFMS

(ES^þ): m/z 435.1208 [M þ Na]^þ; calcd for C₂₆H₂₀O₅þ Na, 435.1208.

Compound 30: Yellow amorphous solid, m.p. 223–225 ^ꢁC; IR

n_max: 1739, 1654, 1636, 1595, 1560, 1508, 1458, 1270, 1213, 1167,

5.3.23. Catalytic hydrogenation of didemethylcurcumin (5)

Compound 5 was subjected to catalytic hydrogenation in similar

manner to that of compound

1 to give didemethyltetrahy-

drocurcumin (38) (65%) as colorless sticky solid. The spectroscopic

data of compound 38 were consistent with the reported values

[34,35].

1060, 970, 821, 704 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

d 5.84 (s, 1H,

H-4), 6.60 (d, J ¼ 15.8 Hz, 2H, H-2 and H-6), 7.25 (d, J ¼ 8.5 Hz, 4H,

H-3⁰, H-5⁰, H-3⁰⁰and H-5⁰⁰), 7.51 (t, J ¼ 7.5 Hz, 4H, m-ArH), 7.60–7.65

(m, 6H, H-2⁰, H-6⁰, H-2⁰⁰, H-6⁰⁰and p-ArH), 7.67 (d, J ¼ 15.8 Hz, 2H, H-

1 and H-7), 8.19 (d, J ¼ 7.5 Hz, 2H, o-ArH); ESMS (þve): m/z (% rel.

abund.) 517 [M þ H]^þ(100); HR-TOFMS (ES^þ): m/z 539.1472

[M þ Na]^þ; calcd for C₃₃H₂₄O₆þ Na, 539.1471.

5.3.24. Preparation of di-O-methylhexahydrocurcumin (39)

Compound 7 was subjected to catalytic hydrogenation in similar

manner to that of compound 1 to give di-O-methyltetrahy-

drocurcumin, which was directly subjected to sodium borohydride

reduction in ethanol to the corresponding hexahydro analog 39

954

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

(62% overall yield from compound 7) as white solid, m.p. 92–93 ^ꢁC

(lit. [26] m.p. 93–94 ^ꢁC). The spectroscopic (IR, ¹H NMR and mass

spectra) data of compound 38 were consistent with the reported

values [26].

2.68 (m, 2 ꢃ 2H, H-7), 2.81 (m, 2 ꢃ 4H, H-1 and H-2), 3.83 (s, 3H,

OMe), 3.84 (br s, 4 ꢃ 3H, 4 ꢃ OMe), 3.85 (s, 3H, OMe), 6.08 (d,

J ¼ 15.9 Hz, 2 ꢃ 1H, H-4), 6.63–6.70 (m, 2 ꢃ 4H, H-2⁰, H-2⁰⁰, H-5⁰, H-

5⁰⁰), 6.75–6.85 (m, 2 ꢃ 3H, H-5, H-6⁰, H-6⁰⁰); ESMS (þve): m/z (% rel.

abund.) 763 [2M þ Na]^þ(100); HR-TOFMS (ES^þ): m/z 393.1678

[M þ Na]^þ; calcd for C₂₂H₂₆O₅þ Na, 393.1678.

5.3.25. Dehydration of hexahydrocurcumin

Hexahydrocurcumin (32) (185 mg, 0.49 mmol) was dissolved in

benzene (45 ml) and p-toluenesulfonic acid monohydrate (50 mg)

was added. The reaction mixture was reﬂux for 3 h; H₂O was added

and the mixture was extracted with EtOAc. The combined organic

phase was washed with H₂O and dried over anhydrous Na₂SO₄; the

solvent was evaporated and the residue chromatographed using

hexane–EtOAc (3:2) to yield compound 40 (175 mg, quantitative

yield) as white amorphous solid. The spectroscopic (IR, ¹H NMR and

mass spectra) data of compounds 40 was consistent with the

reported values [36].

5.3.29. Preparation of enones 45a and 45b mixture

A mixture of the enones 45a and 45b was prepared from the

diketone 34 in similar manner to the preparation of compound 42

from compound 38 to give a mixture of the enones 45a and 45b

(67%) as colorless sticky solid. The spectroscopic (IR, ¹H NMR and

mass spectra) data of compounds 45a and 45b were consistent with

the reported values [36,40].

5.3.30. Preparation of compound 46

Compound 40: IR n_max: 3421, 2935, 2840, 1660, 1624, 1541, 1515,

By using the same procedure for the preparation of compound

42, the diketone 35 was converted to the corresponding a,b-

1457, 1362, 1268, 1238, 1154, 1037, 974, 816, 796 cm^ꢂ1

;

¹H NMR

(CDCl₃, 400 MHz):

d

2.47 (m, 2H), 2.67 (t, J ¼ 7.6 Hz, 2H) and 2.80

unsaturated ketone 46 as colorless sticky solid in 60% overall yield

from compound 35. The spectroscopic (IR, ¹H NMR and mass

spectra) data of compound 46 were consistent with the reported

values [41].

(m, 4H) (H-1, H-2, H-6, H-7), 3.84 (s, 3H, OMe), 3.85 (s, 3H, OMe),

5.47 (s, 1H, OH), 5.48 (s, 1H, OH), 6.08 (br d, J ¼ 15.9 Hz, 1H, H-4),

6.62–6.68 (m, 4H), 6.77–6.84 (m, 3H) (H-5, H-2⁰, H-5⁰, H-6⁰, H-2⁰⁰, H-

5⁰⁰and H-6⁰⁰); EIMS: m/z (% rel. abund.) 356 [M]^þ(60), 219 (17), 206

(49), 205 (14), 137 (100).

5.3.31. Dehydrogenation of compound 40

Compound 40 (80 mg, 0.22 mmol) was dissolved in THF (6 ml)

and 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) (40 mg,

0.18 mmol) was added. The reaction mixture was stirred at ambient

temperature for 4 h. The reaction was worked up by H₂O and

extracted with EtOAc. The organic phase was washed with H₂O,

dried over anhydrous Na₂SO₄and the solvent was evaporated to

dryness. The crude products were separated and puriﬁed by

column chromatography using n-hexane–EtOAc (2:1) to yield the

dienone 47 [42] (22 mg, 28%), dienone 48 (18 mg, 23%) and the

trienone 49 [43] (25 mg, 32%).

5.3.26. Dehydration of a mixture of 37a and 37b

A mixture of compounds 37a and 37b was subjected to dehy-

dration in similar manner to that of compound 31 to give a mixture

of products 41a and 41b (88%). The ¹H NMR data of compound 41a/

41b mixture were consistent with those reported values of 41a [37].

Compounds 41a and 41b (1:1 mixture): Colorless sticky solid; IR

n_max: 3384, 3019, 2936, 2853, 1655, 1516, 1450, 1366, 1273, 1152,

1113, 1034, 814, 754 cm^ꢂ1

400 MHz):

2.66 (m, 2 ꢃ 1H), 2.74 (m) and 2.78 (m) (2 ꢃ 4H, H-1, H-2, H-7), 3.82

(s, 3H, OMe), 3.83 (s, 3H, OMe), 6.04 (d, J ¼ 15.8 Hz, 2 ꢃ1H, H-4),

6.51 (m, 2 ꢃ 1H), 6.62 (m, 2 ꢃ 3H) and 6.70–6.80 (m, 2 ꢃ 3H) (H-5,

H-2⁰, H-2⁰⁰, H-5⁰, H-5⁰⁰, H-6⁰, H-6⁰⁰); ESMS (ꢂve): m/z (% rel. abund.)

341 [M ꢂ H]^ꢂ(100).

;

¹H NMR (CDCl₃þ 2 drops of CD₃OD,

d

2.44 (m, 2 ꢃ 2H, H-6), 2.60 (br t, J ¼ 7.3 Hz, 2 ꢃ 1H),

Compound 47: Yellow sticky solid; IR n_max: 3363, 3014, 2927,

1653, 1560, 1514, 1467, 1430, 1339, 1287, 1154, 1032, 858, 798 cm^ꢂ1

;

¹H NMR (CDCl₃, 400 MHz):

d 2.87 (br s, 4H, H-1 and H-2), 3.85 (s,

3H, OMe), 3.91 (s, 3H, OMe), 5.45 (s, 1H, OH), 5.77 (s, 1H, OH), 6.22

(d, J ¼ 15.4 Hz, 1H, H-4), 6.64–6.73 (m, 3H, H-6, H-5⁰or H-5⁰⁰and H-

2⁰or H-2⁰⁰), 6.82 (d, J ¼ 7.8 Hz, 1H, H-5⁰⁰or H-5⁰), 6.83 (d, J ¼ 15.4 Hz,

1H, H-7), 6.87 (d, J ¼ 8.1 Hz, 1H, H-6⁰or H-6⁰⁰), 6.95 (d, J ¼ 1.3 Hz, 1H,

H-2⁰⁰or H-2⁰), 6.98 (dd, J ¼ 8.1, 1.3 Hz, 1H, H-6⁰⁰or H-6⁰), 7.28

(dd, J ¼ 15.4, 10.7 Hz, 1H, H-5); ESMS (þve): m/z (% rel. abund.) 377

[M þ Na]^þ(100).

5.3.27. Preparation of compound 42

Compound 38 (50 mg, 0.15 mmol) in methanol (2 ml) was

treated with sodium borohydride (5 mg, 0.13 mmol) to give the

corresponding hexahydro analog which was directly dehydrated

without puriﬁcation to afford the product 42 (32 mg, 65% overall

yield from compound 38) as colorless sticky solid. The spectro-

scopic (IR, ¹H NMR and mass spectra) data of compound 42 were

consistent with the reported values [38].

Compound 48: Yellow needles (from n-hexane–CH₂Cl₂), m.p.

124–126 ^ꢁC; IR n_max: 3342, 3007, 2934, 1654, 1602, 1568, 1514, 1430,

1376, 1278, 1217, 1168, 1120,1030, 999, 860 cm^ꢂ^{1 1}H NMR (CDCl₃,

;

400 MHz):

d

2.55 (m, 2H, H-6), 2.74 (t, J ¼ 7.6 Hz, 2H, H-7), 3.85

(s, 3H, OMe), 3.92 (s, 3H, OMe), 5.47 (s, 1H, OH), 5.87 (s, 1H, OH),

6.42 (d, J ¼ 15.6 Hz, 1H, H-4), 6.67 (overlapping signal, 1H, H-2⁰⁰),

6.68 (overlapping signal, 1H, H-5⁰⁰), 6.76 (d, J ¼ 15.9 Hz, 1H, H-2),

6.83 (d, J ¼ 7.7 Hz, 1H, H-5⁰), 6.91 (d, J ¼ 8.1 Hz, H-6⁰⁰), 6.98 (m, 1H,

H-5), 7.04 (d, J ¼ 1.5 Hz, 1H, H-2⁰), 7.10 (dd, J ¼ 8.1, 1.5 Hz, 1H, H-6⁰),

7.52 (d, J ¼ 15.9 Hz, 1H, H-1); EIMS: m/z (% rel. abund.) 354 [M]^þ(6),

218 (13),137 (100); HR-TOFMS (ES^þ): m/z 377.1365 [M þ Na]^þ; calcd

for C₂₁H₂₂O₅þ Na, 377.1365.

5.3.28. Methylation of compound 40

Compound 40 (100 mg, 0.28 mmol) was dissolved in dry acetone

(3 ml) and anhydrous K₂CO₃(70 mg, 0.50 mmol) and MeI (0.8 ml,

1.27 mmol) were added. The reaction mixture was reﬂux for 6 h and

H₂O was added. The mixture was extracted with CH₂Cl₂; the

combined organic phase was washed with H₂O, dried over anhy-

drous Na₂SO₄and the solvent removed under vacuum. The crude

products were puriﬁed by column chromatography using CH₂Cl₂to

yield the mono-O-methylated products 43a/43b mixture in 29%

yield and di-O-methylated product 44 in 57% yield, the latter of

which was recrystallized from n-hexane–CH₂Cl₂to give 44 as nee-

dles, m.p. 68 ^ꢁC. The spectroscopic data of compound 44 were

consistent with the reported values [39].

5.3.32. Pentylation of the enone 40

Compound 40 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 give mono-O-n-pentyl analogs 50a/50b

mixture and di-O-n-pentyl analog 51 in 53 and 27%, respectively.

Compounds 50a and 50b (1:1 mixture): Colorless sticky solid; IR

n_max: 3438, 2934, 2869, 1667, 1626, 1515, 1464, 1366, 1262, 1153,

Compounds 43a and 43b (1:1 mixture): Colorless sticky solid; IR

n_max: 3431, 3017, 2936, 2836,1667, 1626, 1515,1262, 1154, 1029, 853,

808, 754 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

d

2.48 (m, 2 ꢃ 2H, H-6),

1035, 852, 800 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

d

0.90 (t, J ¼ 6.5 Hz,

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

955

2 ꢃ 3H, H-5%), 1.38 (m, 2 ꢃ 4H, H-3% and H-4%), 1.80 (m, 2 ꢃ 2H, H-

2%), 2.48 (m, 2 ꢃ 2H, H-6), 2.67 (m, 2 ꢃ 2H, H-7), 2.81 (m, 2 ꢃ 4H, H-

1 and H-2), 3.81 (s, 3H, OMe), 3.82 (s, 3H, OMe), 3.83 (s, 3H, OMe),

3.84 (s, 3H, OMe), 3.95 (t, J ¼ 6.5 Hz, 2 ꢃ 2H, H-1%), 5.51 (br s, 1H,

OH), 5.52 (br s, 1H, OH), 6.08 (d, J ¼ 15.8 Hz, 2 ꢃ1H, H-4), 6.63–6.67

(m, 2 ꢃ 4H, H-2⁰, H-2⁰⁰, H-5⁰, H-5⁰⁰), 6.75–6.81 (m, 2 ꢃ 3H, H-5, H-6⁰,

H-6⁰⁰); ESMS (þve): m/z (% rel. abund.) 875 [2M þ Na]^þ(100); HR-

TOFMS (ES^þ): m/z 449.2304 [M þ Na]^þ; calcd for C₂₆H₃₄O₅þ Na,

449.2304.

and 2.79 (br t, J ¼ 7.2 Hz, 2 ꢃ 2H) (H-1, H-2, H-4, H-5, H-6, H-7), 3.83

(s, 3H, OMe), 3.84 (s, 3H, OMe), 6.63 (m, 2 ꢃ 2H, H-2⁰⁰and H-5⁰⁰),

6.71 (d, J ¼ 7.9 Hz, 2 ꢃ 2H, H-3⁰and H-5⁰), 6.79 (d, J ¼ 7.7 Hz, 2 ꢃ 1H,

H-6⁰⁰), 6.98 and 6.99 (each d, J ¼ 7.9 Hz, 2 ꢃ 2H, H-2⁰and H-6⁰);

ESMS (ꢂve): m/z (% rel. abund.) 327 [M ꢂ H]^ꢂ(100).

5.3.37. Reduction of the enone 40

Compound 40 (25 mg, 0.07 mmol) was dissolved in EtOH (2 ml)

and NaBH₄(4 mg) was added. The reaction mixture was stirred at

ambient temperature for 30 min. The reaction was worked up by

H₂O and extracted with EtOAc. The organic phase was washed with

H₂O, dried over anhydrous Na₂SO₄and the solvent was evaporated

to dryness. The crude product was puriﬁed by column chroma-

tography using CH₂Cl₂to yield the enol 57 (24 mg, 96%) as colorless

sticky solid; IR n_max: 3423, 2935, 2853, 1603, 1515, 1452, 1428, 1364,

1271, 1231, 1152, 1123, 1033, 816 cm^ꢂ1; ¹H NMR (CDCl₃þ 2 drops of

Compound 51: White amorphous solid, m.p. 65–66 ^ꢁC; IR n_max

2958, 2931, 2861, 1670, 1627, 1589, 1517, 1469, 1277, 1241, 1141, 1035,

987, 850, 797 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

0.89 (m, 6H, H-5%

:

d

and H-5&), 1.38 (m, 8H, H-3%, H-4%, H-3& and H-4&), 1.80 (m, 4H,

H-2% and H-2&), 2.47 (m, 2H, H-6), 2.68 (m, 2H, H-7), 2.81 (m, 4H, H-

1 and H-2), 3.81 (s, 6H, 2 ꢃ OMe), 3.94 (m, 4H, H-1% and H-1&), 6.08

(d, J ¼ 15.8 Hz, 1H, H-4), 6.65 (br s, 2H, H-2⁰and H-2⁰⁰), 6.67 (d,

J ¼ 8.2 Hz, 2H, H-5⁰and H-5⁰⁰), 6.76 (br d, J ¼ 8.2 Hz, 2H, H-6⁰and H-

6⁰⁰), 6.81 (m, 1H, H-5); ESMS (þve): m/z (% rel. abund.) 497 [M þ H]^þ

(100); HR-TOFMS (ES^þ): m/z 519.3087 [M þ Na]^þ; calcd for

C₃₁H₄₄O₅þ Na, 519.3086.

CD₃OD, 400 MHz):

d 1.71 (m, H-2), 2.30 (m, 2H, H-6), 2.49–2.64

(m, 4H, H-1 and H-7), 3.83 (s, 6H, 2 ꢃ OMe), 4.01 (m, 1H, H-3), 5.47

(dd, J ¼ 15.3, 5.6 Hz, 1H, H-4), 5.63 (m, 1H, H-5), 6.64 (m, 4H, H-2⁰,

H-2⁰⁰, H-5⁰, H-5⁰⁰), 6.78 (br d, J ¼ 7.6 Hz, 2H, H-6⁰and H-6⁰⁰); ESMS

(þve): m/z (% rel. abund.) 359 [M þ H]^þ(100); HR-TOFMS (ES^þ): m/z

381.1677 [M þ Na]^þ; calcd for C₂₁H₂₆O₅þ Na, 381.1678.

5.3.33. Hydroxyethylation of the enone 40

Compound 40 was subjected to hydroxyethylation in similar

manner to that of compound 1 to afford mono-O-(2-hydroxyethyl)

analogs 52a and 52b mixture in 27%.

5.4. Biological activities

Compound 52a and 52b (1:1 mixture): Colorless sticky solid; IR

n_max: 3424, 2937,1661,1620,1515,1453,1453,1367,1263,1141,1033,

5.4.1. Parasite strains, human cell line and culture conditions

Bloodstream forms of T. b. brucei strain 427 (BS221) were used as

drug-sensitive control and the multi-drug resistant clonal line

899, 810 cm^ꢂ1

;

¹H NMR (CDCl₃þ 4 drops of CD₃OD, 400 MHz):

d

2.47 (m, 2 ꢃ 2H, H-6), 2.66 (m, 2 ꢃ 2H, H-7), 2.78 (m, 2 ꢃ 4H, H-1

and H-2), 3.80 (s, 3H, OMe), 3.81 (s, 2 ꢃ 3H, 2 ꢃ OMe), 3.82 (s, 3H,

OMe), 3.85 (m, 2 ꢃ 2H, H-2%), 4.01 (m, 2 ꢃ 2H, H-1%), 6.06 (d,

J ¼ 15.9 Hz, 2 ꢃ1H, H-4), 6.59–6.69 (m, 2 ꢃ 4H, H-2⁰, H-2⁰⁰, H-5⁰, H-

5⁰⁰), 6.76 (br d, J ¼ 8.2 Hz, 2 ꢃ 2H, H-6⁰and H-6⁰⁰), 6.79 (m, 2 ꢃ1H, H-

5); ESMS (ꢂve): m/z (% rel. abund.) 399 [M ꢂ H]^ꢂ(100); HR-TOFMS

(ES^þ): m/z 423.1783 [M þ Na]^þ; calcd for C₂₃H₂₈O₆þ Na, 423.1784.

427DTbAT1 was derived from this strain by deletion of the TbAT1

gene that encodes the P2 aminopurine transporter [46] responsible

for the uptake of melaminophenyl arsenical and diamidine drugs

[48,53]. The clonal line B48, which is highly resistant to melarsoprol

and pentamidine, was in turn derived from 427DTbAT1 by adap-

tation to high levels of pentamidine [49]. T. evansi strain

ITMAV170475 was obtained from Vincent Delespaux, Institute of

Tropical Medicine Antwerp, Belgium and adapted to culture in

HMI9/20% FBS following the procedure of Hirumi and Hirumi [54].

All T. b. brucei strains and T. evansi were cultured at 37 ^ꢁC in a CO₂

incubator using HMI9 media (BioSera, Ringmer, UK) supplemented

with 10% fetal bovine serum (FBS, BioSera) exactly as described

previously [55]. Amastigotes of L. mexicana (strain MNYC/BZ/62/

M379) were maintained in axenic culture at 33 ^ꢁC in a CO₂incu-

bator using Schneider’s Drosophila medium (Gibco) supplemented

with 20% FBS and 0.3% gentamycin as described [56]. Promastigotes

of L. major (Friedlin strain) were cultured in HOMEM medium

(Invitrogen) supplemented with 10% FBS at 25 ^ꢁC as described [57].

Human Embryonic Kidney (HEK) cells of strain 293 T were cultured

as described by Rodenko et al. [58] in a medium based on Dul-

becco’s modiﬁed Eagle’s medium (Sigma) with 10% newborn

bovine serum (Invitrogen).

5.3.34. Acetylation of the enone 40

Compound 40 was subjected to acetylation in the same manner

to that of compound 1 to give the monoacetates 53a/53b mixture

and diacetate 54 in 40 and 42%, respectively. The latter compound

has been synthesized previously [44].

Compound 53a and 53b (1:1 mixture): Colorless sticky solid; IR

n_max: 3444, 2938, 1763, 1664, 1604, 1514, 1453, 1369, 1272, 1199,

1152, 1122, 1034, 903, 822 cm^ꢂ1; ¹H NMR (CDCl₃, 400 MHz):

d 2.27

(s, 3H, OAc), 2.28 (s, 3H, OAc), 2.48 (m, 2 ꢃ 2H, H-6), 2.70 (m, 2 ꢃ 2H,

H-7), 2.81 (m, 2 ꢃ 4H, H-1 and H-2), 3.78, (s, 3H, OMe), 3.79, (s, 3H,

OMe), 3.84, (s, 3H, OMe), 3.85 (s, 3H, OMe), 6.08 (d, J ¼ 15.7 Hz, 1H,

H-4), 6.10 (d, J ¼ 15.7 Hz, 1H, H-4), 6.63–6.92 (m, 2 ꢃ 7H, H-5, H-2⁰,

H-2⁰⁰, H-5⁰, H-5⁰⁰, H-6⁰, H-6⁰⁰); ESMS (ꢂve): m/z (% rel. abund.) 397

[M ꢂ H]^ꢂ(100); HR-TOFMS (ES^þ): m/z 421.1627 [M þ Na]^þ; calcd for

C₂₃H₂₆O₆þ Na, 421.1627.

5.3.35. Catalytic hydrogenation of the enone 40

5.4.2. Cytotoxicity and antiprotozoal activity

Compound 40 was subjected to catalytic hydrogenation in

similar manner to that of compound 1 to afford the ketone 55

(43 mg, 86%). The spectroscopic (IR, ¹H NMR and mass spectra) data

of compound 55 were consistent with the reported values [33].

All compounds were screened for cytotoxicity against the

human cell line HEK293 exactly as described [58]. Brieﬂy, HEK cells

were grown in 96-well plates at 37 ^ꢁC under a 5% CO₂atmosphere

and to each well an equal volume of a doubling dilution of test

compound was added, after which the plate was incubated another

16 h prior to the addition of the Alamar Blue (resazurin, Sigma)

5.3.36. Catalytic hydrogenation of a mixture of 56a and 56b

A mixture of 45a and 45b was subjected to catalytic hydroge-

nation in similar manner to that of compound 1 to afford a mixture

of 56a and 56b (86%).

reagent. The highest concentration of test compound was 250

mM,

unless restricted by solubility. Fluorescence

(

l_exc¼ 530 nm;

l_em¼ 590 nm) was determined after an additional 24 h incubation,

in a Perkin–Elmer LS55 ﬂuorimeter. EC₅₀values (deﬁned as Effec-

tive Concentration for 50% reduction of ﬂuorescence in the Alamar

Blue assay, below) were determined by non-linear regression,

Compounds 56a and 56b (1:1 mixture): Colorless sticky solid; ¹H

NMR (CDCl₃, 400 MHz):

2 ꢃ 2H), 2.49 (br t, J ¼ 6.0 Hz, 2 ꢃ 2H), 2.65 (br t, J ¼ 7.0 Hz, 2 ꢃ 2H)

d

1.54 (m, 2 ꢃ 4H), 2.36 (br t, J ¼ 6.0 Hz,

956

C. Changtam et al. / European Journal of Medicinal Chemistry 45 (2010) 941–956

[21] M. Nose, T. Koide, Y. Ogihara, Y. Yabu, N. Ohta, Biol. Pharm. Bull. 21 (1998)

643–645.

[22] T. Koide, M. Nose, Y. Ogihara, Y. Yabu, N. Ohta, Biol. Pharm. Bull. 25 (2002)

131–133.

[23] S. Venkateswarlu, M.S. Ramachandra, G.V. Subbaraju, Bioorg. Med. Chem. 13

(2005) 6374–6380.

[24] 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) 5037–5042.

[25] W.-Y. Shao, Y.-N. Cao, Z.-W. Yu, W.-J. Pan, X. Qiu, X.-Z. Bu, L.-K. An, Z.-S. Huang,

L.-Q. Gu, A.S.C. Chan, Tetrahedron Lett. 47 (2006) 4085–4089.

[26] L. Lin, Q. Shi, A.K. Nyako, K.F. Bastow, C.-C. Wu, C.-Y. Su, C.C.-Y. Shih, K.-H. Lee,

J. Med. Chem. 49 (2006) 3963–3972.

[27] S. Barthelemy, L. Vergnes, M. Moynier, D. Guyot, S. Labidalle, E. Bahraoui, Res.

Virol. 149 (1998) 43–52.

[28] S. Mishra, U. Narain, R. Mishra, K. Misra, Bioorg. Med. Chem. 13 (2005)

1477–1486.

[29] A. Faridah, H. Lim Siang, A. Syahida, S. Johnson, A.D. Israf, S. Khozirah,

N.H. Lajis, J. Biosci. 61 (2006) 625–631.

[30] K. Mohammadi, K.H. Thompson, B.O. Patrick, T. Storr, C. Martins, E. Polishchuk,

V.G. Yuen, J.H. McNeill, C.J. Orvig, Inorg. Biochem. 99 (2005) 2217–2225.

[31] P.-H. Bong, Bull. Korean Chem. Soc. 21 (2000) 81–86.

[32] J. Ishida, H. Ohtsu, Y. Tachibana, Y. Nakanishi, K.F. Bastow, M. Nagai, H.-

K. Wang, H. Itokawa, K.-H. Lee, Bioorg. Med. Chem. 10 (2002) 3481–3487.

[33] S.-L. Lee, W.-J. Huang, W.W. Lin, S.-S. Lee, C.-H. Chen, Bioorg. Med. Chem. 13

(2005) 6175–6181.

[34] E. Portes, C. Gardrat, A. Castellan, Tetrahedron 63 (2007) 9092–9099.

[35] J. Ma, X. Jin, L. Yang, Z.-L. Liu, Phytochemistry 65 (2004) 1137–1143.

[36] K. Endo, E. Kanno, Y. Oshima, Phytochemistry 29 (1990) 797–799.

[37] H. Kikuzaki, M. Kobayashi, N. Nakatani, Phytochemistry 30 (1991) 3647–3651.

[38] M. Kuroyanagi, M. Shimomae, Y. Nagashima, N. Muto, T. Okuda, N. Kawahara,

T. Nakane, T. Sano, Chem. Pharm. Bull. 53 (2005) 1519–1523.

[39] C. Martin-Cordero, M. Lopez-Lazaro, M.A. Agudo, E. Navarro, J. Trujillo,

M.J. Ayuso, Phytochemistry 58 (2001) 567–569.

ﬁtting data to a sigmoid curve with variable slope (Prism version 4,

GraphPad).

Tests for trypanocidal (T. b. brucei) and leishmanicidal activity

utilized similar protocols based on Alamar Blue ﬂuorescence in

96-well plates, using the culture condition described in the previous

section and protocols exactly as described [49,55–57]. An Alamar

Blue procedure was validated for the culture adapted T. evansi, using

cell counts in a hemocytometer under phase-contrast microscopy to

conﬁrm correlation between Alamar Blue and cell survival after

exposure to various standard trypanocides. Test conditions were

optimized after (1) variation of trypanosome cell number at start of

drug test, (2) incubation of trypanosome culture with test

compound and (3) incubation time with the Alamar blue reagent.

These parameters were resolved as 1.5 ꢃ10⁵trypanosomes/ml and

48 h and 24 h, respectively, for the incubations at 37 ^ꢁC in a CO₂

incubator in HMI9/20% FBS.

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. CC

acknowledges a scholarship from the Royal Golden Jubilee Ph.D.

Program of The Thailand Research Fund. HI was supported by

a studentship from the Libyan government.

[40] K.-S. Lee, G. Li, S.H. Kim, C.-S. Lee, M.-H. Woo, S.-H. Lee, Y.-D. Jhang, J.-K. Son,

J. Nat. Prod. 65 (2002) 1707–1708.

References

[41] M. Nomura, T. Tokoroyama, T. Kubota, Phytochemistry 20 (1981) 1097–1104.

[42] H. Jiang, B.N. Timmermann, D.R. Gang, J. Chromatogr. A 1111 (2006) 21–31.

[43] S.-Y. Park, D.H.L. Kim, J. Nat. Prod. 65 (2002) 1227–1231.

[44] T. Murata, M. Shinohara, M. Miyamoto, Chem. Pharm. Bull. 20 (1972)

2291–2292.

[45] R.A. Sharma, A.J. Gesher, W.P. Steward, Eur. J. Cancer 41 (2005) 1955–1968.

[46] E. Matovu, M. Stewart, F. Geiser, R. Brun, P. Ma¨ser, L.J.M. Wallace,

R.J. Burchmore, J.C.K. Enyaru, M.P. Barrett, R. Kaminsky, T. Seebeck, H.P. De

Koning, Eukaryot. Cell 2 (2003) 1003–1008.

[47] H.P. De Koning, L.F. Anderson, M. Stewart, R.J.S. Burchmore, L.M.J. Wallace,

M.P. Barrett, Antimicrob. Agents Chemother. 48 (2004) 1515–1519.

[48] P.G. Bray, M.P. Barrett, S.A. Ward, H.P. De Koning, Trends Parasitol. 19 (2003)

232–239.

[49] D. Bridges, M.K. Gould, B. Nerima, P. Ma¨ser, R.J.S. Burchmore, H.P. De Koning,

Mol. Pharmacol. 71 (2003) 1098–1108.

[50] H.P. De Koning, Trends Parasitol. 24 (2008) 345–349.

[51] H.P. De Koning, S.M. Jarvis, Mol. Pharmacol. 56 (1999) 1162–1170.

[52] C.A. Lanteri, M.L. Stewart, J.M. Brock, V.P. Alibu, S.R. Meshnick, R.R. Tidwell,

M.P. Barrett, Mol. Pharmacol. 70 (2006) 1585–1592.

[53] H.P. De Koning, Int. J. Parasitol. 31 (2001) 512–522.

[54] H. Hirumi, K.J. Hirumi, J. Parasitol. 75 (1989) 985–989.

[55] L.J.M. Wallace, D. Candlish, H.P. De Koning, J. Biol. Chem. 277 (2002)

26149–26156.

[56] M.I. Al-Salabi, H.P. De Koning, Antimicrob. Agents Chemother. 49 (2005)

3682–3689.

[57] M.I. Al-Salabi, L.J.M. Wallace, H.P. De Koning, Mol. Pharmacol. 63 (2003)

814–820.

[1] M.P. Barrett, Curr. Opin. Infect. Dis. 13 (2000) 647–651.

[2] P. Olliario, J. Lazdins, F. Guhl, Expert Opin. Emerg. Drugs 7 (2002) 1–7.

[3] V. Delespaux, H.P. De Koning, Drug Resist. Updat. 10 (2007) 30–50.

[4] S.L. Croft, M.P. Barrett, J.A. Urbina, Trends Parasitol. 21 (2005) 508–512.

[5] R. Docampo, S.N.J. Moreno, Parasitol. Res. 90 (2003) S10–S13.

[6] J. Jannin, P. Cattand, Curr. Opin. Infect. Dis. 17 (2004) 565–570.

[7] A. Stich, M.P. Barrett, S. Krishna, Trends Parasitol. 19 (2003) 195–197.

[8] R. Brun, R. Schumacher, C. Schmid, C. Kunz, C. Burri, Trop. Med. Int. Health 6

(2001) 906–914.

[9] P.M. Kristjanson, B.M. Swallow, G.J. Rowlands, R.L. Kruska, P.N. De Leeuw,

Agric. Syst. 59 (1999) 79–98.

[10] S.A. Reid, Trends Parasitol. 18 (2002) 219–224.

[11] J.L. Gonzalez, E. Chacon, M. Miranda, A. Loza, L.M. Siles, Vet. Parasitol. 146

(2007) 9–16.

[12] R.K. Maheshwari, A.K. Singh, J. Gaddipatti, R.C. Srimal, Life Sci. 78 (2006)

2081–2087.

[13] A.J. Ruby, G. Kuttan, K.D. Babu, K.N. Rajasekharan, R. Kuttan, Cancer Lett. 911

(1995) 79–83.

[14] L.N. Gringberg, O. Shalev, H.H. Tonnesen, E.A. Lachmillewitz, Int. J. Pharm. 132

(1996) 251–257.

[15] M.M.-Y. Chan, C.-T. Ho, H.-I. Huang, Cancer Lett. 96 (1995) 23–29.

[16] M.M.-Y. Chan, H.-I. Huang, M.R. Fenton, D. Fong, Biochem. Pharmacol. 55

(1998) 1955–1962.

[17] H. Inano, M. Onoda, N. Inafuku, M. Kubota, Y. Kamada, T. Osawa, H. Kobayashi,

K. Wakabayashi, Carcinogenesis 21 (2000) 1835–1841.

[18] R. Thapliyal, G.B. Maru, Food Chem. Toxicol. 39 (2001) 541–547.

[19] C.A.C. Araujo, L.V. Alegrio, D.C.F. Gomes, M.E.F. Lima, L. Gomes-Cardoso,

L.L. Leon, Mem. Inst. Oswaldo Cruz 94 (1999) 791–794.

[20] A. Mazumder, K. Raghavan, J. Weinstein, K.W. Kohn, Y. Pommier, Biochem.

Pharmacol. 49 (1995) 1165–1170.

[58] B. Rodenko, A.M. Van der Burg, M.J. Wanner, M. Kaiser, R. Brun, M. Gould,

H.P. De Koning, G.J. Koomen, Antimicrob. Agents Chemother. 51 (2007)

3796–3802.

Article Doi

Curcuminoid analogs with potent activity against Trypanosoma and Leishmania species

DOI: 10.1016/j.ejmech.2009.11.035

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.ejmech.2009.11.035

Full text of DOI:10.1016/j.ejmech.2009.11.035

Products guided by the article

R&D Labs maybe for 1221590-24-4

Relevant to this article

Hot Product