2- and 3-substituted 1,4-naphthoquinone derivatives as subversive substrates of trypanothione reductase and lipoamide dehydrogenase from Trypanosoma cruzi: Synthesis and correlation between redox cycling activities and in vitro cytotoxicity

548

J . Med. Chem. 2001, 44, 548-565

2- a n d 3-Su bstitu ted 1,4-Na p h th oqu in on e Der iva tives a s Su bver sive Su bstr a tes

of Tr yp a n oth ion e Red u cta se a n d Lip oa m id e Deh yd r ogen a se fr om Tr ypa n osom a

cr u zi: Syn th esis a n d Cor r ela tion betw een Red ox Cyclin g Activities a n d in Vitr o

Cytotoxicity

Laurence Salmon-Chemin,^†Eric Buisine,^†Vanessa Yardley,^‡Sven Kohler,^§Marie-Ange Debreu,^†

Vale´rie Landry,^†Christian Sergheraert,^†Simon L. Croft,^‡R. Luise Krauth-Siegel,*^,§and

Elisabeth Davioud-Charvet*^,†

UMR 8525 CNRS - Universite´ Lille II, Institut de Biologie de Lille et Institut Pasteur de Lille, 1 rue du Professeur Calmette,

BP447, 59021 Lille Cedex, France, Department of Infections and Tropical Diseases, London School of Hygiene & Tropical

Medicine, Keppel Street, London WC1E 7HT, U.K., and Biochemie-Zentrum der Universita¨t Heidelberg,

Im Neuenheimer Feld 328, D-69120 Heidelberg, Germany

Received September 15, 2000

Trypanothione reductase (TR) is both a valid and an attractive target for the design of new

trypanocidal drugs. Starting from menadione, plumbagin, and juglone, three distinct series of

1,4-naphthoquinones (NQ) were synthesized as potential inhibitors of TR from Trypanosoma

cruzi (TcTR). The three parent molecules were functionalized at carbons 2 and/or 3 by various

polyamine chains. Optimization of TcTR inhibition and TcTR specificity versus human disulfide

reductases was achieved with the 3,3′-[polyaminobis(carbonylalkyl)]bis(1,4-NQ) series 19-20,

in which an optimum chain length was determined for inhibition of the trypanothione disulfide

reduction. The most active derivatives against trypanosomes in cultures were also studied as

subversive substrates of TcTR and lipoamide dehydrogenase (TcLipDH). The activities were

measured by following NAD(P)H oxidation as well as coupling the reactions to the reduction

of cytochrome c which permits the detection of one-electron transfer. For TcTR, 20_4-cproved

to be a potent subversive substrate and an effective uncompetitive inhibitor versus trypano-

thione disulfide and NADPH. Molecular modeling studies based on the known X-ray structures

of TcTR and hGR were conducted in order to compare the structural features, dimensions,

and accessibility of the cavity at the dimer interface of TcTR with that of hGR, as one of the

putative NQ binding sites. TcLipDH reduced the plumbagin derivatives by an order of

magnitude faster than the corresponding menadione derivatives. Such differences were not

observed with the pig heart enzyme. The most efficient and specific subversive substrates of

TcTR and TcLipDH exhibited potent antitrypanosomal activity in in vitro T. brucei and T.

cruzi cultures. The results obtained here confirm that reduction of NQs by parasitic

flavoenzymes is a promising strategy for the development of new trypanocidal drugs.

In tr od u ction

responsible for the resulting oxidative stress. The mo-

lecular basis of quinone toxicity is the enzyme-catalyzed

reduction to semiquinone radicals which then reduce

oxygen to superoxide anion radicals thereby regenerat-

ing the quinone.¹This futile redox cycling and oxygen

activation lead to increased levels of hydrogen peroxide

and GSSG. A well-studied example is menadione (2-

methyl-1,4-naphthoquinone), which is an acceptor of

electrons from numerous flavoproteins acting through

a one-electron mechanism, e.g. NADPH cytochrome c

reductase, NADH dehydrogenase, and ferredoxin NADP

reductase, or through a mixed mechanism (one- and

two-electron), e.g. NAD(P)H dehydrogenase (or DT-

diaphorase) or lipoamide dehydrogenase (LipDH).^3,4

The quinone structure is common to numerous natu-

ral products and is associated with anticancer, antibac-

terial, antimalarial, and fungicide activities.¹In most

cases, the biological activity is related to the ability of

quinones to accept one and/or two electrons to form the

corresponding radical anion or dianion species as well

as the acid-base properties of the compounds. The

cytoxicity of quinones is also often attributed to DNA

modification, thiol and amine depletion by alkylation,

oxidation of essential protein thiols by activated oxygen

species, and/or glutathione disulfide (GSSG).¹The vari-

able capacity of quinones to accept electrons is due to

the electron-attracting or -donating substituents at the

quinone moiety²which modulate the redox properties

Many naphthoquinone (NQ) drugs, such as menadi-

one, plumbagin, and lapachol, display notable trypano-

cidal activities upon different trypanosomes and Leish-

mania responsible for several human diseases including

African sleeping sickness (Trypanosoma brucei rhod-

esiense and Trypanosoma brucei gambiense), Chagas’

disease (Trypanosoma cruzi), and Kala-azar (Leishma-

nia donovani).^5-7These parasitic protozoa of the order

* To whom correspondence should be addressed. For E.D.-C.:

Phone: (33) 3 20 87 12 17. Fax: (33) 3 20 87 12 33. E-mail:

elisabeth.davioud@pasteur-lille.fr. For R.L.K.-S.: Phone: (49) 6221 54

41 87. Fax: (49) 6221 54 55 86. E-mail: krauth-siegel@urz.uni-

heidelberg.de.

^†UMR 8525 CNRS - Universite´ Lille II.

^‡London School of Hygiene & Tropical Medicine.

^§Biochemie-Zentrum der Universita¨t Heidelberg.

Published on Web 01/13/2001

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 549

Ch a r t 1. Structures of the Series of 1,4-NQs: 3-Polyamino-1,4-NQs 3-6, 3-(Polyaminocarbonylalkyl)-1,4-NQs 16-17,

2,3-Bis(polyaminocarbonylalkyl)-1,4-NQ 18, and 3,3′-[Polyaminobis(carbonylalkyl)]bis(1,4-NQ)s 19-20

Kinetoplastida are particularly sensitive to oxidative

stress and have developed specific antioxidant systems.⁸

In contrast to nearly all other organisms, they lack

glutathione reductase (GR) but have a trypanothione

reductase (TR) instead. The physiological substrate of

TR is the bis-glutathionyl spermidine disulfide (trypano-

thione disulfide, T(S)₂), which is reduced to the corre-

sponding dithiol trypanothione (T(SH)₂). Many am-

monium compounds possessing substituted side chains

have been described as effective and specific TR inhibi-

tors. The selectivity of inhibitors for parasite TR over

human GR (hGR) was mainly due to the charge differ-

ences at the catalytic sites of the two enzymes.⁹LipDH

is another flavoenzyme, structurally and mechanisti-

cally closely related to TR and GR, and all belong to

the family of FAD-disulfide oxidoreductases.¹⁰A well-

known function of LipDH is the NAD⁺-dependent

oxidation of protein-bound dihydrolipoamide in mito-

chondrial 2-oxoacid dehydrogenase complexes. TR and

LipDH have been purified^11,12from Trypanosoma cruzi

(TcTR and TcLipDH), and the genes have been cloned

and overexpressed.^13,14Besides their physiological reac-

tions, TR, GR, and LipDH catalyze the slow NAD(P)H-

dependent reduction of molecular oxygen.¹⁵In the case

of TcLipDH this intrinsic NADH oxidase activity

amounts to about 1% of the physiological activity of the

enzyme.^15,16All three flavoenzymes have been shown

to interact with NQs. In the case of TcTR, menadione,

550 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4

Ch a r t 2. Structures of the Polyamines a -h

Salmon-Chemin et al.

Sch em e 1. Synthesis of 3-Polyamino-1,4-NQs 3-6^a

plumbagin,¹⁷and other 1,4-NQs^18-20act as subversive

substrates (or turncoat inhibitors) by inhibiting the

physiological reduction of T(S)₂and promoting a non-

physiological reaction. In the presence of O₂, the one-

electron reduction of quinones can lead to the production

of superoxide anion radicals. Toward GR, 1,4-NQs act

mainly as weak reversible inhibitors.^19,21,22Since the

three parent NQs are known to interact unspecifically

with numerous NAD(P)H-dependent disulfide oxidoreduc-

tases, like TR,¹⁷GR,²³LipDH,⁴and thioredoxin reduc-

tase²⁴(TrxR), the aim of this work was to improve the

specificity of NQs for TR binding and to decrease

cytotoxicity against human cells.

In the present paper, we describe the synthesis of a

series of menadione, plumbagin, and juglone deriva-

tives, their inhibition potencies of TcTR, and their in

vitro trypanocidal effects. The approach included the

synthesis of cationic ligands displaying different redox

potentials. For this purpose, the three parent structures

were substituted at C-2 and C-3 of the quinone moiety

(Chart 1) by alkylamino side chains of varying length

(Chart 2), either attached directly or with an alkyl

spacer of variable length via a bridging amide bond.

Enzymatic and antitrypanosomal studies revealed that

several NQ derivatives with strong trypanocidal activity

were among the most effective TcTR inhibitors. How-

ever, the lack of correlation between inhibition of T(S)₂

reduction and antitrypanosomal action for several other

NQs suggested a study of the redox cycling properties

of the compounds toward TcTR and TcLipDH. These

studies revealed that the most potent trypanocidal NQs

displayed the highest and most specific redox cycling

activities toward TcTR or TcLipDH. In addition, mo-

lecular modeling studies based on the known X-ray

structures of TcTR and hGR^25-29were conducted to

obtain a deeper insight in the structural features,

dimensions, and accessibility of the cavity at the dimer

interface of TcTR and hGR, as one of the putative NQ

binding sites.

a

Reaction conditions: (a) R₂R′₂NH (5 equiv), EtOH-DCM; (b)

NH₂R′₂NH₂(0.5 equiv), EtOH-DCM.

allow the regeneration of the quinone moiety in the final

products 3-6.^30,31The ethanol-DCM (2:0.5) mixture

was found to be the solvent of choice for the reaction

since it favored the solubility of both starting NQs and

formed products. Attempts to improve the yields of the

3-polyaminoplumbagin derivatives 4 by changing the

reaction conditions (solvent, temperature) proved un-

successful. In comparison with EtOH alone or EtOH-

DCM, DCM, or DMF as sole solvent led to the formation

of many byproducts. Toluene was employed as a re-

placement for EtOH, but its handling proved difficult

under reduced pressure leading to NQ degradation.

Product mixtures were not less complex when the

reaction temperature was reduced from 25 to 4 °C. In

addition, product isolation and purification proved to

be difficult in the case of aliphatic tri- or tetraamines.

All compounds required purification by preparative TLC

with an appropriate elution solvent (acetone/NH₄OH for

mono derivatives and toluene/THF/TEA for bis deriva-

tives). In the case of the secondary amine h , in addition

to the expected product 3h , the minor compound 5h was

isolated. This was the result of an addition-elimination

reaction occurring during which the methyl group

behaves as an electron donor to oxygen.³²Nucleophilic

substitution of bromo intermediates was also consid-

ered.³³Three derivatives 3-bromo-2-methyl-1,4-NQ, 2-bro-

momethyl-1,4-NQ, and 3-bromo-2-bromomethyl-1,4-NQ

were prepared by the bromination of menadione.³¹With

the first two bromo derivatives, addition-elimination

Resu lts

Syn th esis of 3-P olya m in o-1,4-NQ. A first series of

3-polyamino-1,4-NQ derivatives 3-6 was synthesized

by direct 1,4-type addition of the quinone moiety of 1,4-

NQ to polyamines b-h (Scheme 1). The nonisolated

amino-substituted naphthohydroquinone intermediates

were spontaneously oxidized by molecular oxygen, to

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 551

Sch em e 2. Synthesis of Carboxylic Acids 9-11 and Boc-Protected Amines 14-15^a

a

Reaction conditions: (a) 9-10: diacid (3 equiv), AgNO₃catalytic, (NH₄)₂S₂O₈(1.3 equiv), aq 30% CH₃CN, 60-65 °C; 11: diacid (6

equiv), AgNO₃catalytic, (NH₄)₂S₂O₈(2.6 equiv), aq 30% CH₃CN, 60-65 °C; (b) 14-15: Boc-protected amines 13_2-5(3 equiv), AgNO₃

catalytic, (NH₄)₂S₂O₈(1.3 equiv), aq 30% CH₃CN, 60-65 °C.

Sch em e 3. Synthesis of Amides 16-18 and Bis(amides) 19-20^a

a

Reaction conditions: (a) 16-17: NHR₂R′₂(1 equiv), HBTU (1 equiv), DIEA (3 equiv), DCM-DMF, rt, 3 h; 18: NHR₂R′₂(2 equiv),

HBTU (2 equiv), DIEA (6 equiv), DCM-DMF, rt, 3 h; (b) NH₂R′₂NH₂(0.5 equiv), HBTU (1 equiv), DIEA (1 equiv), DCM-DMF, rt, 3 h.

did not lead to the major products described by Ohta et

al.³¹Only compound 5h was obtained as the major

product from the reaction of polyamine h and 3-bromo-

2-bromomethyl-1,4-NQ.³⁴However, in the plumbagin

series, 1,4-type addition followed by oxidation in air led

to low yields and many byproducts. Synthesis of 3,3′-

polyaminobis(1,4-NQ) 6 was achieved by inverse addi-

tion of polyamine (1 equiv) to menadione as starting

material (2 equiv).

Syn th esis of 3-(P olya m in oca r bon yla lk yl)-1,4-NQ.

Our aim was to synthesize in a high-yield two-step

procedure a second series of 1,4-NQ possessing an amino

side chain, with a nitrogen atom nonconjugated with

the quinone portion of 1,4-NQ. Since 1,4-NQs are highly

552 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4

Salmon-Chemin et al.

Ta ble 1. Inhibitory and Redox Cycling Activities toward TcTR and in Vitro Antiparasitic Activities of Compounds 3-6 and 16-18

x-fold^b

ED₅₀(µM)

a

IC₅₀

compd

n

series

(µM)

(100 µM)

L. donovani^c

T. cruzi^c

T. brucei

3_b

3_c

3_d

3_e

3_f

menadione

(R₁) H)

6

12

28

>57

×2.2

g12.5*

>12.5*

T/100^d

>12.5

g 6.2*

>12.5*

g12.5

T/+^g

>12.5

30

>12.5

nd

T/+^f

×2

g 6.2

>12.5

3_h

nd

4_b

4_c

4_d

4_e

4_f

plumbagin

(R₁) OH)

39

57

28

>57

×6.4

×2

T/+^d

>30

T/+^f

T/+^g

>12.5

T/+^e

>3

>12.5

>30

nd

>30

nd

>6.2*

>12.5

×2.8

T/100^d

5_h

6_b

6_c

6_d

6_e

menadione

(R₁) H)

>57

10

×3.9

>12.5*

T/+^e,*

>12.5

T/+^e

>12.5

>1

nd

9

×4

nd

g 12.5*

T/+^e

6.5

>57

T/100^d

T/+^g

>12.5*

>12.5

16_2-f

16_2-g

2

3

menadione

(R₁) H)

38

>50

×7

>30

T/+^d

>10

×19

87

70

9.5

17_2-f

17_2-g

plumbagin

(R₁) OH)

16

50

×10

nd

>88

T/+^e

>88

>10

52

×38

16_3-f

16_3-g

menadione

(R₁) H)

44

nd

×11

>76

32

>76

27

46

9.1

nd

17_3-f

17_3-g

plumbagin

(R₁) OH)

32

42

×40

66

82

50

>84

23

52

×14

18_2-f

18_2-g

18_3-f

18_3-g

18_4-f

18_4-g

18_5-f

18_5-g

2

3

4

5

juglone

>50

11

17.5

6.2

nd

>50

>62

>50

>62

59

>48

>46

T/+^e

T/+^d

T/+^f

21.5

22.2

259

31.4

>15

>56

10.1

13.1

×45

×220

nd

>59

>48

38.2

5.9

0.8

2.9

nd

T/100^f

T/+^e

T/+^f

a

b

In the presence of 57 µM T(S)₂. The redox cycling activities of compounds were measured at 25 °C in TR assay buffer, pH 7.25, at

an NADPH concentration of 250 µM, as described in the Experimental Section. The control assays which contained only NADPH and the

enzyme represent the intrinsic oxidase activity of TR. Oxidation of NADPH was monitored at 340 nm. The rate of NADPH oxidation by

TcTR in the presence of 100 µM NQ was x-fold increased, when compared to the intrinsic oxidase activity of the enzyme (17-fold for

menadione and 39-fold for plumbagin). nd, not determined. ^cT/100 means the compound was toxic to macrophages, 100% inhibition; T/+

e

f

g

means the compound was toxic to macrophages but parasites were still present: ^dat 30 µM, at 10 µM, at 3 µM, at 1 µM. *L. infantum.

reactive, it was necessary to find a general route which

allowed the preparation of intermediates under mild

conditions, which could be easily derivatized by parallel

synthesis. Both solid- and solution-phase methodologies

allowing parallel syntheses of a large array of diverse

amino-substituted NQs have recently been described.³⁵

Taking into account that many polyamines were com-

mercially available, we selected a versatile reaction to

prepare 1,4-NQ acid and amine derivatives as interme-

diates in the synthesis of a large array of amides by

standard coupling procedures in solution. Diacids of

varying length (succinic acid, glutaric acid, adipic acid,

and pimelic acid) were reacted with menadione or

plumbagin, via the oxidative decarboxylation of acids,

promoted by silver salts as an efficient catalyst in the

presence of ammonium peroxydisulfate³⁶(Scheme 2). In

this way, eight 1,4-NQ acid derivatives 9-10 were

prepared in bulk. HBTU was used as coupling reagent,

and the reaction was performed in solution (Scheme 3).

The 3-(polyaminocarbonylalkyl)-1,4-NQs 16-17 and the

3,3′-[polyaminobis(carbonylalkyl)]bis(1,4-NQ)s 19-20

were prepared from polyamines f-g and from polyamines

a -c, respectively, with the latter possessing two ter-

minal amino groups.

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 553

Ta ble 2. Inhibitory Activities toward TcTR and in Vitro

Antiparasitic Activities of Carboxylic Acids 9-11 and

Boc-Protected Amines 14-15

Syn th esis of 3-(N-Boc-a m in oa lk yl)-1,4-NQ. By the

same method, N-Boc-protected 1,4-NQ amines 14-15

were synthesized from N-Boc-protected amino acids 13,

previously prepared from the corresponding amino acids

12, â-alanine, γ-aminobutyric acid, 5-aminovaleric acid,

and 6-aminocaproic acid (Scheme 2).

Syn th esis of 2,3-Bis(p olya m in oca r bon yla lk yl)-

1,4-NQ. As described above for the synthesis of mena-

dione and plumbagin acid derivatives, four juglone

diacids 11 were synthesized from the four starting

diacids (Scheme 2). Then, 2,3-bis(polyaminocarbonyla-

lkyl)-1,4-NQs 18 were prepared by coupling in solution

diacids 11 with polyamines f-g in the presence of

HBTU as coupling reagent (Scheme 3). Noteworthy here

is that the structures of the synthesized juglone deriva-

tives resembled more closely the structures of the

plumbagin derivatives than that of juglone itself. J u-

glone derivatives can therefore be classified as plum-

bagin derivatives because of the presence of methylene

groups at C-2 and C-3.

a

IC₅₀

ED₅₀(µM)

compd

n

series

(µM) L. donovani T. cruzi T. brucei

9₂

9₃

9₄

9₅

2

3

4

5

menadione >50

>30

>116

39

>30

>116

44

>30

75

>110

(R₁) H)

>50

In h ibition of TR-Ca ta lyzed T(S)₂Red u ction . The

three series of 1,4-NQs 3-6, 9-11, and 14-20 were

tested in the T(S)₂reduction assay at varying concen-

trations (0.3-50 µM), by following the disappearance

of NADPH at 340 nm (Tables 1-3). For each NQ, we

checked at the concentration corresponding to the IC₅₀

value that the rate of NADPH oxidation was negligible

in the absence of T(S)₂, proving that the NQ used in

the presence of a low amount of TcTR (0.02 U/assay)

behaved as an inhibitor. When tested as potential

inhibitors of the TR-catalyzed T(S)₂reduction in the

presence of 57 µM T(S)₂and TR, the respective IC₅₀

values of the 3-polyamino-1,4-NQ derivatives 3_b, 3_c, 3_d

and 3,3′-polyaminobis(1,4-NQ) derivatives 6_b, 6_c, 6_d

were found to be lower than those of the parent

molecule, menadione (IC₅₀) 55 µM) (Table 1). In

contrast compound 6_elacking a protonatable nitrogen

atom in the side chain was not well-recognized by TcTR,

which illustrates the essential requirement of a posi-

tively charged amino group for TR recognition. None of

the plumbagin derivatives yielded an IC₅₀value lower

than that of plumbagin (IC₅₀) 28 µM). Using the

routine procedure, we also tested the 12 acid derivatives

9-11 at 50 µM and the 8 N-Boc-protected amine

derivatives 14-15 at either 25 or 50 µM (Table 2). No

inhibition of TcTR was observed. This was not so

surprising with the acid derivatives 9-11 since similar

results have already been reported with a variety of

molecules bearing a carboxylic group,^9,37proving once

more the importance of the charge as the major dis-

criminating factor for TR/GR specificity. Most of the

3-(polyaminocarbonylalkyl)-1,4-NQs 16-17 were rela-

tively ineffective with little TcTR inhibition evident at

concentrations up to 50 µM (16 < IC₅₀< 50 µM) (Table

1). When compared with g, amine f, in agreement with

the increased number of protonable amino groups of the

side chain (2 instead of 1), conferred a greater inhibitory

potency in the series of compounds 16-17. However,

the 2,3-polyaminocarbonylalkyl-1,4-NQ 18 (Table 1) and

the 3,3′-[polyaminobis(carbonylalkyl)]bis(1,4-NQ)s 19-

20 (Table 3) exhibited significant inhibition in the low-

micromolar range (IC₅₀< 5 µM), particularly when the

length of the alkyl linker reached a maximum (n ) 4 or

5 methylene groups). The bis(1,4-NQ) 20_4-cwas found

18

17

5.3

10₂

10₃

10₄

10₅

2

3

4

5

plumbagin >50

(R₁) OH) >50

>50

72

>109

56

46

>109

42

>11

78

>10

>3.3

>50

28

32

11₂

11₃

11₄

11₅

2

3

4

5

juglone

>50

>93

>87

>80

>74

>94

>87

80

59

61

>80

48

>74

14₂

14₃

14₄

14₅

2

3

4

5

menadione >50

8.8

9.5

0.9

(R₁) H)

>50

>25

>50

29.4

16.5

12.5

13.8

10.4

4.3

>3

2.7

15₂

15₃

15₄

15₅

2

3

4

5

plumbagin >25

(R₁) OH) >25

>25

11.8

10.0

14.2

33.8

3.5

3.7

7.0

0.6

1.4

2.5

>25

13.3

a

In the presence of 57 µM T(S)₂. The assays were carried out

as described in the legend of Table 1.

to be the most potent TcTR inhibitor with an IC₅₀of

0.45 µM in the presence of 57 µM T(S)₂, a value 62-fold

lower than the corresponding parent molecule, plum-

bagin (IC₅₀) 28 µM) (Table 3). To determine the

inhibition type, TcTR activity was measured in the

presence of varying concentrations of T(S)₂(25-200 µM)

and constant concentrations of the mono(NQ) 3_b(0-40

µM) and the bis(NQ) 20_4-c(0-2 µM), respectively. The

type of inhibition was deduced from Lineweaver-Burk,

Dixon, and Cornish-Bowden plots.³⁸While inhibition

of TcTR by 3_bwas noncompetitive (data not shown),

20_4-cacted as an uncompetitive inhibitor toward T(S)₂

(Figure 1). An inhibition constant of 0.5 ( 0.02 µM was

calculated by fitting the data to the appropriate equa-

tion by a computerized least-squares regression.³⁸With

NADPH (5-25 µM) as variable substrate and T(S)₂at

a constant concentration of 100 µM, inhibition of TR by

20_4-c(0-2.5 µM) was also uncompetitive.

TcTR ver su s h GR a n d h Tr xR Sp ecificity Stu d -

ies. The most potent TR inhibitor 20_4-cwas also

measured as inhibitor of hGR. In the presence of 50 µM

GSSG, the IC₅₀value was 10 µM which is 22-fold higher

than the value for TR. The corresponding IC₅₀values

of the parent molecules were quite similar for the

554 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4

Salmon-Chemin et al.

Ta ble 3. Inhibitory, Redox Cycling, and in Vitro Antiparasitic Activities of Compounds 19-20 toward TcTR

a

IC₅₀

x-fold^b

K_m

V_max

k_cat/K_m

ED₅₀(µM)

compd

n

series

(µM)

(25 µM)

(µM)

(µmol/min/mg)

(10⁴M^-1s^-1

)

L. donovani^c

T. cruzi^c

T. brucei

menadione

plumbagin

55

28

×15

21

15

0.26

0.31

1.0

1.7

T/100^d

T/100^f

T/100^e

inactive^d

1.5

×21

19_2-a

19_2-b

19_2-c

2

3

4

5

menadione

(R₁) H)

2.7

3.7

5

×117

×40^#

×69

×199

×158

×154

30

14

47

1.83

0.74

1.30

5.1

4.4

2.3

>50

>46

>51

13

40

47

4.9

3.2

4.5

20_2-a

20_2-b

20_2-c

plumbagin

(R₁) OH)

2.9

2.6

1.9

48

78

230

4.44

4.74

12.43

7.7

5.1

4.5

>48

>44

>49

46

44

13

13.8

30.3

1.7

19_3-a

19_3-b

19_3-c

menadione

(R₁) H)

4.0

4.5

11

×62

×50

×25

74

12

23

2.04

0.6

0.45

2.3

4.2

1.6

>48

14.7

>49

>48

13

>49

26.6

13.7

27.0

20_3-a

20_3-b

20_3-c

plumbagin

(R₁) OH)

2.3

2.0

×100

93

52

55

4.0

2.16

4.29

3.6

3.5

6.5

>46

38.5

46.7

46

28

47

>4.5

3.4

>4.6

×83

×165

19_4-a

19_4-b

19_4-c

menadione

(R₁) H)

1.4

1.1

18.7*

×76

×117

×16

16

37

1.11

1.61

0.34

5.8

8.4

0.8

11.9

31.3

>45

9

26

45

4.4

7.3

15.3

20_4-a

20_4-b

20_4-c

plumbagin

(R₁) OH)

1.7

2.5

0.45

×201

×335

×201

×69

×33

×8

×221

×121

×227

20

17

28

3.11

5.13

3.83

13.0

25.2

11.4

7.8

29.6

11.1

4.0

17.2

4.3

>1.5

>1.3

1.1

19_5-a

19_5-b

19_5-c

menadione

(R₁) H)

1.7

3.1

2.5*

21

12

3

1.2

0.47

0.09

4.8

3.3

2.3

5.4

14.3

22.8

4

11

12.5

2.6

5.7

>1.5

20_5-a

20_5-b

20_5-c

plumbagin

(R₁) OH)

1.6

2.0

37

8

28

5.91

1.61

4.87

13.3

16.8

14.5

34.8

17.4

10.3

24.4

11.4

3.0

0.3

>1.3

>1.4

a

#

b

In the presence of 57 µM T(S)₂. At 12.5 µM. *Precipitates in the assay buffer at the IC₅₀value. Reduction of the NQs by TcTR was

measured at 25 °C in TR assay buffer, pH 7.25, at an NADPH concentration of 250 µM as described in the Experimental Section. The

control assays which contained only NADPH and the enzyme represent the intrinsic oxidase activity of TR. Oxidation of NADPH was

followed at 340 nm. The rate of NADPH oxidation by TcTR in the presence of 25 µM NQ was x-fold increased, when compared to the

intrinsic oxidase activity of the enzyme. The values are the mean of at least two independent measurements which differed by less than

10%. ^cT/100 means the compound was toxic to macrophages, 100% inhibition; T/+ means the compound was toxic to macrophages but

parasites were still present: ^dat 30 µM, at 10 µM, at 3 µM.

e

f

parasite and host enzymes (for menadione: IC₅₀) 63

µM with hGR and 55 µM with TcTR; for plumbagin:

IC₅₀) 38 µM with hGR and 28 µM with TcTR). The

most potent series of TR inhibitors 19-20 (n ) 4-5)

were also tested in the standard hTrxR screening assay

using the recently reported 5,5′-dithiobis(2-nitrobenza-

mide) (DTNBA) as an alternative substrate for hTrxR³⁹

and in the standard hGR screening assay using GSSG

as substrate. At 25 µM NQ and 200 µM disulfide

substrate, both assays indicated a selective affinity of

compounds 20_4-cand analogues for the trypanosomal

enzyme compared to human GR and TrxR (data not

shown).

compounds 3-6, 16-18 (Table 1) or 25 µM compounds

19-20 (Table 3).

2O₂+ NAD(P)H + H⁺^enzymatic8

2O₂^-•+ NAD(P)⁺+ 2H⁺(1)

2NQ + NAD(P)H + H⁺^enzymatic8

2NQ^-•+ NAD(P)⁺+ 2H⁺(2)

For the first series of 3-polyamino-1,4-NQ derivatives

3-6 (Table 1), redox cycling activity proved to be lower

(2.0-6.4-fold) than that observed at 100 µM unsubsti-

tuted menadione (17-fold) and plumbagin (39-fold). This

may be due to the electron-donating effect of the

nitrogen atom bound directly to the quinone moiety.

Among the 3-(polyaminocarbonylalkyl)-1,4-NQs 16-17

(Table 1), only 16_2-g(19-fold) displayed an NADPH

oxidase activity comparable with that of menadione. In

the plumbagin and juglone series, 17_2-g(38-fold), 17_3-f

(40-fold), and 18_3-f(45-fold) showed NADPH oxidase

activities comparable with that of plumbagin. 18_3-g

Red u ction of NQ Der iva tives by TcTR a n d h GR.

Three series of NQs (menadione, plumbagin, and ju-

glone derivatives) were studied as subversive substrates

of TR by following the oxidation of NADPH in the

presence of 25-100 µM NQ. The NQ reductase activity

of TcTR (eq 2) was compared with the intrinsic NADPH

oxidase activity of the enzyme in the absence of NQ (eq

1). The data represent the increase of the rate of

NADPH oxidation (x-fold) in the presence of 100 µM

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 555

F igu r e 1. Lineweaver-Burk plot of 1/v versus 1/[S] (a) and Cornish-Bowden plot of [S]/v versus [I] (b) for the uncompetitive

inhibition of TcTR by 20_4-cagainst T(S)₂. The assays were measured at 22 °C at a constant concentration of 200 µM NADPH and

varying concentrations of 204-c 0 (O), 0.5 (b), 1.0 (0), 1.5 ([), 2.0 (4) µM. The concentrations of T(S)₂were 23 (1), 43 (3), 62 (2),

103 (4), 151 (9), 201 (0) µM. All assays contained a final concentration of 1% DMSO. The reaction was started by adding enzyme

(10^-2units in a total volume of 500 µL), and NADPH consumption was followed.

caused the highest NQ reductase activity (220-fold). In

comparison with the mono-NQ derivatives, the 3,3′-

[polyaminobis(carbonylalkyl)]bis(1,4-NQ)s 19-20 (Table

3) were much more effective subversive substrates of

TcTR. The rate of NADPH oxidation by TcTR increased

8-117-fold in the presence of 25 µM menadione deriva-

tives 19, when compared to the intrinsic NADPH

oxidase activity of the enzyme. Under identical condi-

tions, the rate of NADPH oxidation by TcTR increased

83-335-fold in the presence of the corresponding plum-

bagin derivatives. The K_mand k_catvalues were derived

from measurements at five different concentrations of

NQ. To gain insight of where the NQ reduction takes

place in the enzyme, we measured the reduction of 20_4-c

in the presence of mepacrine, a known competitive

inhibitor of T(S)₂reduction.⁴⁰Only a partial inhibition

was detected at 50 and 100 µM mepacrine in the

presence of varying concentrations of 20_4-c(3-50 µM),

which may be an indication that the disulfide binding

site of TR is not the major binding site of the bis(NQ)

superoxide anion radicals. A comparison of the amount

of cytochrome c reduced (measured at 550 nm) per

oxidized NADPH (measured at 340 nm) in the presence

of 12.5 µM 20_4-cyielded a 1:2 ratio (7.1 and 3.5 µmol/

min/mg, respectively).

To assess the preference of 20_4-cas subversive

substrate of TcTR over hGR, the NQ reductase activity

was measured in the cytochrome c reduction assay with

hGR in the presence of 12.5 µM 20_4-cand compared

with that observed in the presence of 12.5 µM plumba-

gin. The rate of cytochrome c reduction was not signifi-

cantly increased (1.1-fold) in the presence of 20_4-c, when

compared to that in the absence of NQ, which confirmed

that hGR does not reduce 20_4-c. Under identical condi-

tions, the rate of cytochrome c reduction increased 1.6-

fold in the presence of the parent plumbagin.

Mod elin g Stu d ies. The crystallographic analysis of

hGR has revealed that certain un/noncompetitive in-

hibitors bind in a cavity distinct from the disulfide and

NADPH binding sites.^25-29This “third site” is situated

at the 2-fold symmetry axis of the homodimeric protein²⁵

and is the locus of binding of 2-methyl-1,4-NQ (mena-

dione),²⁶3,7-diamino-2,8-dimethyl-5-phenylphenazini-

um chloride (safranin),²⁶6-hydroxy-3-oxo-3H-xanthene-

9-propionic acid,²⁷a series of 10-arylisoalloxazines,²⁸

and S-(2,4-dinitrophenyl)glutathione.²⁹As shown in

Figure 2, the structurally closely related parasite en-

zyme TcTR (35% sequence identity with hGR) possesses

a similar cavity. To elucidate possible discriminating

factors between the “third sites” in the human and

parasite enzymes, a detailed analysis has been under-

taken based on the available structural data.

In TcTR, the cavity is lined by 32 residues (Table 4)

of each monomer (41 in hGR). 27 of these residues are

conserved in all TRs of Crithidia fasciculata, Trypano-

soma brucei, Trypanosoma congolense, and Leishmania

donovani, suggesting a common mode of interaction

with molecules probably binding in this cavity. In

contrast, only 15 residues are conserved when compar-

ing TcTR with hGR. Moreover, none of the residues

20_4-c

.

The efficiency of NQs as turncoat inhibitors of TR can

be defined by the ratio (k_cat/K_m)/IC₅₀, combining the

redox cycling activity and inhibition of the physiological

reaction of TR. 20_4-cwas the most potent TcTR turncoat

inhibitor with the highest (k_cat/K_m)/IC₅₀value equal to

25.3 × 10¹⁰M^-2s^-1. Production of superoxide anion

radicals by this compound (eq 3) was evaluated by

coupling reaction 2 to cytochrome c (Fe³⁺) reduction (eqs

4 and 5).

NQ^-•+ O₂f NQ + O₂

(3)

(4)

(5)

-•

O₂^-•+ cyt c (Fe)³⁺f O₂+ cyt c (Fe)²⁺

NQ^-•+ cyt c (Fe)³⁺f NQ + cyt c (Fe)²⁺

Addition of SOD in the presence of 12.5 µM 20_4-c

decreased the rate of cytochrome c reduction by 88%,

which confirmed that reduction of 20_4-cby TR generates

556 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4

Salmon-Chemin et al.

F igu r e 2. Front (a,c, respectively) and back (b,d, respectively) views of the protein backbone depicted for TcTR and hGR, showing

the cavity at the dimer interface (green area). The cavity, also called “third site”, is connected to the two disulfide substrate (T(S)₂

and GSSG, respectively) binding sites through two short channels (blue area). These channels emerge at the bottom of the V-shaped

catalytic crevices in close proximity to the redox-active disulfide bridges. As indicated by the CAST calculation, two additional

channels which connect the cavity with the protein surface in a region distinct from the two catalytic sites are detectable in the

hGR structure (pink area). The redox-active disulfide bridges (Cys53 and Cys58 in TcTR; Cys58 and Cys63 in hGR) and the FAD

molecules are given as yellow and purple hard spheres, respectively.

involved in the binding of xanthene derivatives to hGR

(Trp70, Asn71, Val74, His75, Phe78, Met79, His82, and

Tyr407)²⁷is conserved in the five TR sequences. Figure

3 shows specific residues in the “third site” of TcTR. A

summation of net charges of acidic (-I) and basic (+I)

residues, considering histidine residues in their unpro-

tonated state, leads to an overall charge of -IV for the

interfacial cavity TcTR and only -II for hGR. Remark-

able is the high number of histidine residues in the

cavity of human GR with a total of 5/monomer while

TcTR has only 2/monomer. With regard to residues that

could be specifically involved in the binding of ligands,

Thr66, Gln69, Tyr70, Gln242, Ile369, Met400, Lys409,

Phe411, Asp432, and Pro435 are conserved in the

different TRs but not in hGR. Gln69, Met400, and

Asp432 are particularly accessible inside the interface

cavity and are characterized by high thermal param-

eters in the structure of free TcTR.⁴¹

As shown in Figure 2a,b, the “third site” of TcTR is

almost globular while that of hGR (Figure 2c,d) has a

rather cross-shaped form. The molecular surface and

volume of the cavity calculated with the CAST software

are 1755 Å²and 2280 Å³in TcTR and 1954 Å²and 2399

Å³in hGR. The surface and volume accessible to solvent

are 888 Å²and 511 Å³(887 Å²and 464 Å³in hGR),

respectively. The interface pocket of both enzymes is

connected to the two disulfide substrate binding sites

through short channels. There is no direct connection

with the FAD and NADPH binding sites. These “pri-

mary” channels open into the catalytic sites near the

redox-active disulfide bridges. The circumference of

these channel openings of TcTR enzyme is about 31 Å,

and the molecular area is about 46 Å². The openings

are delineated by hydrophobic residues (Phe396, Pro398,

Leu399, Pro462, Thr463, and Val59′ of the other sub-

unit) strictly conserved in the different TRs. In hGR, a

unique substitution is observed, Leu399 of TcTR being

replaced by Met406. In the structure of TcTR without

bound ligands,⁴¹Phe396, Pro398, and Leu399 at the

opening exhibit particularly high B-factors of 40-60 Å².

Each channel opening is 14 Å apart from the center of

the cavity, and its distance to the redox-active disulfide

bridge is about 9-10.5 Å, the shortest distance between

the catalytic sulfur atoms and atoms at the channel

entrance being 4.4 Å. In hGR, the CAST calculation

identified two additional channels, explaining the higher

number of residues lining the cavity (41/monomer in

hGR versus 32/monomer in TcTR). These “secondary”

channels provide direct access to the surface of the

protein in a region distinct from the two catalytic sites.

By neglecting the residues involved in these regions, the

structural comparison of the cavities in the human and

parasitic enzymes by superposition of main chain atoms

yielded a root-mean-square (rms) deviation of 0.91 Å.

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 557

Ta ble 4. Residues Lining the Interface Cavity of TcTR and the

Red u ction of Men a d ion e Der iva tives by T. cr u zi

a n d P ig Hea r t Lip DH. The ability of LipDH to reduce

menadione is well-known since the yeast enzyme was

first described as menadione reductase.⁴²Reduction of

the menadione derivatives was first studied by following

the oxidation of NADH which yields the sum of one- and

two-electron reduction rates of the compounds (eqs 2

and 6). As presented in Table 5, the unsubstituted

menadione is reduced by the parasite enzyme with a

k_cat/K_mvalue of 1 × 10⁵M^-1s^-1. The best new derivative

was 14₃which in the case of TcLipDH showed a

catalytic efficiency almost as high as that of menadione.

To estimate the amount of semiquinone formed in the

LipDH-catalyzed reduction of NQs, the reaction was

coupled to cytochrome c reduction as described for TR

(eqs 2-5). The assays were carried out at a constant

concentration of 50 µM NQ since at higher concentra-

tions the comproportionation of hydronaphthoquinones

(NQH₂) and oxidized NQ may become significant (eqs

6-7).

Corresponding Residues in Other TRs and hGR^a

rel acc^b

TcTR

(%)

TbTR TcgTR CfTR LdTR

hGR

V59

K62

L63

o

15

32

3

V64

K67

V68

V65

T66

Q69

Y70

7

W70

N71

V74

13

31

3

N

H75

E72

H73

E76

4

27

4

D

H

D

Q

D

T

D

L

E77

F78

D81

Q242

F 367

S368

I369

P 370

P 371

F 396

P 398

L399

M400

H401

K409

T410

F411

G431

D432

N433

P435

E436

P 462

T463

S464

1

10

1

10

5

10

37

21

32

37

4

7

31

2

E237

F 372

S373

H374

P 375

P 376

F 403

P 405

M406

Y407

H408

T415

K416

C417

G437

L438

G439

D441

E442

P 468

T469

S470

o

NQ + NADH + H⁺^{catalyzed by LipDH}8

NQH₂+ NAD⁺(6)

K

G

K

S

K

S

E

S

NQH₂+ NQ f 2NQ^-•+ 2H⁺

(7)

45

42

14

16

2

This nonenzymatic generation of semiquinones can

give rise to an erroneously high cytochrome c reductase

activity.³At concentrations lower than 50 µM, reduction

of the NQ by TcLipDH was too slow to allow detailed

kinetics. Therefore the rate of cytochrome c reduction

was measured in the presence of 50 µM NQ and

compared with the rate of NADH oxidation at the same

concentration (Table 5). Under these conditions TcLip-

DH catalyzed the one-electron reduction of the mena-

dione derivatives with an efficiency of about 50%. SOD

diminished the rate of cytochrome c reduction by 83%

in the case of 9₂and 16_2-fand by 67% in the presence

of menadione, which confirms the generation of super-

oxide anion radicals. The compounds 3_band 3_fand the

bis(NQ) derivative 6_cunderwent very little or no reduc-

tion at all by TcLipDH (see structures in Table 1). Only

6_cat 250 µM enhanced the rate of NADH oxidation by

about 25%. K_mand V_maxvalues were not determined.

The apparent second-order rate constants for the reduc-

tion of the menadione derivatives by pig heart LipDH

were similar to the values obtained with the parasite

enzyme (Table 5). Again the unsubstituted menadione

was the most effective substrate. Also in the case of the

mammalian enzyme a lower K_mvalue was always

accompanied by a lower V_maxvalue. The proportion of

one-electron reduction ranged from 21-84% of the total

electrons transferred from NADH to NQ. SOD dimin-

ished the rate of cytochrome c reduction by 86% in the

case of 9₂and by 35% in the case of menadione.

Reduction of cytochrome c mediated by 16_2-fwas

completely inhibited by SOD.

o

22

a

Residues conserved between TRs and hGR are given in bold

letters (residues corresponding to “secondary” channels in hGR

are not listed). The numbering used for TcTR residues refers to

that of the three-dimensional structure accessible at the Protein

Data Bank under 1aog refcode.⁴¹For the different TRs, only

mutations are specified. o denotes TcTR residues at the channel

b

openings. Relative accessibilities of side chains, e.g. surface_real

/

surface_maxratios as estimated by using the NACCESS software.⁶⁶

s^-1showed the highest catalytic efficiency. The parasite

enzyme reduced compound 10₂almost as effectively as

plumbagin, yet the much higher K_mvalue for this

derivative resulted in a relatively low k_cat/K_mratio.

When studied in the cytochrome c assay, plumbagin

yielded 80% one-electron reduction which was com-

pletely abolished in the presence of SOD. The deriva-

tives 14₃and 15₃caused a cytochrome c reduction in

the absence of LipDH and thus could not be studied for

their one-electron reduction capacity. Reduction of 17_2-f

and 10₂by TcLipDH occurred 50% and 66% as one-

electron reductions, respectively, and SOD diminished

the rate of cytochrome c reduction by about 75%. Also

pig heart LipDH exhibited the highest catalytic ef-

ficiency with the unsubstituted plumbagin resulting in

a k_cat/K_mof 2.9 × 10⁵M^-1s^-1. Interestingly, the

mammalian enzyme had a much lower activity with the

newly synthesized derivatives than the parasite enzyme.

For instance, reduction of 15₃occurred at a rate only

one-seventh that of the parasite enzyme. The apparent

second-order rate constants for 15₃, 17_2-f, and 10₂were

about an order of magnitude lower than that for

plumbagin. The proportion of one-electron reduction

ranged from 38% to 89%, and the reactions were

inhibited by SOD between 50% and 90%.

Red u ction of P lu m ba gin Der iva tives by T. cr u zi

a n d P ig Hea r t Lip DH. In the series of plumbagin

derivatives, the unsubstituted compound also proved to

be the best substrate of both LipDHs (Table 5). When

following NADH consumption, plumbagin was reduced

by TcLipDH with a k_cat/K_mof 4.8 × 10⁵M^-1s^-1. Of the

newly synthesized derivatives, 15₃with 2.3 × 10⁵M^-1

An titr yp a n osom a l Activities. Antitrypanosomal

activities of the compounds toward the intracellular

558 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4

Salmon-Chemin et al.

F igu r e 3. Enlarged view of the “third site” in TcTR with its connecting channels and corresponding openings. Only residues

specific to the parasite enzyme are shown (side chain); those common to human GR have been omitted. The two identical subunits

are pink and green. S, O1, and O2 designate the centers of the cavity and of the channel openings, respectively. Each channel

entrance is 14 Å away from the center of the cavity (S-O1 and S-O2 distances), and the distance between the two entrances is

27 Å (O1-O2 distance). The approximate distance between the center of the entrances (O1 or O2) and the redox-active cysteine

residues (Cys53 and Cys58) is 9-10.5 Å. The side chains of active cysteine residues are depicted as yellow (S_γatoms) and green

(C_âatoms) ball-and-sticks.

Ta ble 5. 1,4-NQ Derivatives Studied as Subversive Substrates of LipDH from T. cruzi and Pig Heart

TcLipDH

pig heart LipDH

cyt c reduction:

NADH oxidation,

(µmol/min/mg):

(µmol/min/mg)^b

cyt c reduction:

NADH oxidation,

(µmol/min/mg):

(µmol/min/mg)^b

NADH oxidation^a

V_max

(µM) (µmol/min/mg) (10⁴M^-1s^-1

NADH oxidation^a

V_max

(µM) (µmol/min/mg) (10⁴M^-1s^-1

)

K_m

k_cat/K_m

(% one-electron

reduction)

K_m

k_cat/K_m

(% one-electron

reduction)

compd

)

menadione

9₂

14₃

16_2-f

plumbagin

10₂

170

85

20

90

40

140

30

20

21.04

6.73

2.02

3.21

22.9

17.2

8.24

2.72

10

6.6

8.4

3

48

10.2

23

12.7:13.5 (47)

3.0:2.7 (56)

c

1.6:1.7 (47)

26.1:16.3 (80)

6.6:5.0 (66)

c

72

40

25

110

74

75

11.5

1.52

1.03

3.46

25.3

2.2

13

3.2

3.4

2.6

8.1:4.8 (84)

0.5:1.2 (21)

c

0.8:0.8 (50)

15.4:10.0 (77)

1.6:0.9 (89)

c

29

2.4

3.5

6.1

15₃

17_2-f

40

15

1.66

1.09

11.3

1.9:1.9 (50)

0.6:0.8 (38)

a

NADH oxidation was followed at 31 °C in 50 mM potassium phosphate, 1 mM EDTA, pH 7.0, at a constant concentration of 100 µM

b

NADH. The NQs were varied between 10 µM and 1.2 mM. The K_mand k_catvalues were derived from Lineweaver-Burk plots. Cytochrome

c reduction was measured at 100 µM NADH, 50 µM cytochrome c, and 50 µM NQ in 50 mM potassium phosphate, pH 7.0. NADH oxidation

was followed under identical assay conditions lacking cytochrome c. ^cThe compound could not be measured in the cytochrome c assay

because of a direct reduction of cytochrome c. All assays were carried out at least in duplicate, the values varied by less than 8%. The

intrinsic oxidase activity of T. cruzi LipDH is 1.2 µmol/min/mg when following NADH oxidation which corresponds to about 0.6% of the

LipDH activity. In the reaction coupled to cytochrome c the activity is 0.7 µmol/min/mg. The latter reaction is completely inhibited by

SOD. The oxidase activity of pig heart LipDH in the NADH assay is 0.8 µmol/min/mg (which corresponds to 0.4% of LipDH activity) and

0.4 µmol/min/mg in the cytochrome c assay. The latter activity is also completely abolished by SOD.

amastigote stage of both T. cruzi and L. donovani (or

L. infantum) and upon the T. brucei bloodstream form

trypomastigotes are given in Tables 1-3, as ED₅₀values

(when these could be calculated). Taking into account

the sensitivity of the different parasites, compounds

having ED₅₀values below 5 µM with T. cruzi or L.

donovani and values below 1 µM with T. brucei can be

considered as trypanocidal lead molecules. The standard

drugs, nifurtimox and pentamidine, were used as posi-

tive controls for T. cruzi and T. brucei and displayed

ED₅₀values in the range 2.2-4.4 and 0.03-0.1 µM,

respectively. The 3-polyamino-1,4-NQs 3-6, the 3-(poly-

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 559

F igu r e 4. Influence of the spacer length on inhibition of T(S)₂reduction, redox cycling activity, and trypanocidal effects of the

3,3′-[polyaminobis(carbonylalkyl)]bis(1,4-NQ)s 19-20 of the menadione and plumbagin series: (a) correlation between TR inhibition

and redox cycling activity; (b) TR inhibition and redox cycling activity as a function of the spacer length n; (c) trypanocidal effects

and redox cycling activity as a function of the spacer length n.

aminocarbonylalkyl)-1,4-NQs 16-17, the 2,3-bis(polyami-

nocarbonylalkyl)-1,4-NQ 18 (Table 1), and the starting

acid derivatives 9-11 (Table 2) exerted weak trypano-

cidal effects upon the different trypanosomatids in

corroboration with an observed weak TcTR inhibition

or, alternately, a cytotoxicity displayed upon human

macrophages. The cytotoxic effects of polyamine deriva-

tives upon human cells has often been encountered.^43,44

When molecules 3-6 proved noncytotoxic upon human

macrophages, the weak antitrypanosomal action could

be related to a low production of superoxide anion

radicals in agreement with the redox potential of the

molecule, affected by the donor effect of the nitrogen

atom linked directly to the quinone moiety. In contrast,

the 3,3′-[polyaminobis(carbonylalkyl)]bis(1,4-NQ)s 19-

20 (Table 3) displayed significant antitrypanosomal

effects, especially toward T. cruzi. Several derivatives

superoxide anions radicals in the parasites, due to the

redox cycling activity of the NQs.

In addition, the starting N-Boc-protected amine de-

rivatives 14-15 (Table 2) displayed potent trypanocidal

effects upon T. cruzi and T. brucei with ED₅₀values for

T. cruzi around 3-4 µM for the most active compounds

(see compounds 14₅, 15₂, 15₃) and around 1 µM for T.

brucei (see compounds 14₂, 15₂, 15₃, 15₄). The unex-

pected lack of TcTR inhibition by this series of 1,4-NQ

indicates the presence of other targets responsible for

the observed antitrypanosomal activities.

Cor r ela tion betw een Red ox Cyclin g Activities

a n d in Vitr o An titr yp a n osom a l Activities. In the

3,3′-[polyaminobis(carbonylalkyl)]bis(1,4-NQ) series 19-

20, the role of the parent NQ, the nature of the

polyamine, and the chain length were analyzed with

respect to the redox cycling properties of each derivative.

Some structure-activity relationships (SARs) for the

inhibition of T(S)₂reduction and redox cycling activities

could be derived (Figure 4). The highest redox cycling

activities correlate with strong inhibition of the physi-

ological T(S)₂reduction, particularly in the plumbagin

series (Figure 4a). For each group of three polyamine

yielded ED₅₀values of 3-4 µM (see compounds 20_4-a

,

20_4-c, 19_5-a, 20_5-c) comparable to that of nifurtimox.

T. brucei and L. donovani were less sensitive to these

1,4-NQ derivatives, with ED₅₀values for T. brucei

around 1 µM (see compounds 20_4-c, 20_5-a), and L.

donovani, with ED₅₀values of about 5 µM for the most

active compound (see compound 19_5-a). As expected, the

data revealed a clear tendency within the three series

of 1,4-NQ derivatives designed in Chart 1: the most

potent TcTR inhibitors (19-20: IC₅₀values below 5 µM)

were the most active antitrypanosomal compounds. It

should be noted that no cytotoxicity upon human

macrophages was observed above 25 µM for the com-

pounds of this 19-20 series. Nevertheless, within this

last series (Table 3), there was no straight correlation

between ED₅₀values from in vitro antitrypanosomal

assays and IC₅₀values from TS₂reduction assays. The

highest antiparasitic activity within the 3,3′-[polyami-

nobis(carbonylalkyl)]bis(1,4-NQ)s 19-20 was observed

with those compounds possessing the longest spacer

between the two NQ moieties (n ) 4-5). This finding

suggests that physical and molecular features other

than TR binding must be considered, for instance the

hydrophobicity of a long alkyl chain which may facilitate

cellular penetration. A second parameter that is likely

to play an important role for the trypanocidal activity

of the most potent derivatives is the specific release of

derivatives (19_{2a -c}, 19_{3a -c}, 19_{4a -b}, 19_{5a -c}, 20_{2a -c}, 20_{3a -c}

,

20_{4a -c}, 20_{5a -c}in Table 3) the mean values of the ratio

of (k_cat/K_m)/IC₅₀determined from the NADPH oxidase

assay and the T(S)₂reduction assay were calculated

(Figure 4b), as were the mean values of the ratio of (k_cat

/

K_m)/ED₅₀determined in the NADPH oxidase assay and

the T. cruzi trypanocidal tests (Figure 4c). From these

data, an effect of the spacer length can be deduced.

When TR inhibition, redox cycling activity of TR, and

trypanocidal activities were expressed as a function of

the spacer length, structure-activity curve profiles

attained a maximum when n ) 4 in both the plumbagin

and menadione series (Figure 4b,c, respectively). Inter-

estingly, plumbagin derivatives showed much higher

activities than their menadione counterparts (Figure

4b,c), which is in agreement with the one-electron

reduction potentials of the parent molecules (E₁values

) -0.20 and -0.16 V for menadione and plumbagin,

respectively).¹⁷

Three menadione and three plumbagin derivatives

displaying potent trypanocidal effects and no TR inhibi-

tion were also examined as redox cycling substrates of

560 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4

Salmon-Chemin et al.

LipDH, another flavoenzyme characterized from T.

cruzi.¹⁴As observed with the positively charged nitro-

furan chinifur, no inhibition of dihydrolipoamide dehy-

drogenation and relatively weak redox cycling activities

were found with menadione and plumbagin derivatives

carrying positively or negatively charged substituents

(9₂, 10₂, 16_2-f, 17_2-f). Interestingly, the most efficient

substrates of TcLipDH were the uncharged carbamates

14₃and 15₃which displayed potent antitrypanosomal

activities against T. brucei and T. cruzi in culture but

no inhibition of TR.

quinone moieties is essential for both T(S)₂reduction

inhibition and redox cycling activity. In addition, the

presence of protonatable amino groups in the spacer is

indispensable for TR recognition. Interestingly, N,N′-

bis-acylated polyamine derivatives are also potent and

specific inhibitors of TR from T. cruzi and Crithidia

fasciculata, i.e. the N,N′-1,5-pentylenediamide and N⁴,N⁸-

bis-acylated spermidine derivatives with 2-aminodi-

phenyl sulfide substituents^45,46as well as kukoamine

A,⁴⁷respectively. The bis(NQ) compounds presented in

this paper are structurally related to kukoamine A, a

potent mixed-type competitive TR inhibitor.⁴⁷Kukoam-

ine A (N¹,N¹²-bis(dihydrocaffeoyl)spermine), composed

of two catechol moieties linked by a 20-atom diacylsper-

mine chain, has been suggested to act through its

orthohydroquinone state as a subversive substrate of

TR. The mode of inhibition of the enzyme by kukoamine

A and by bis(NQ) derivatives could be in fact very

similar. Recently, a crystallographic study of TR crystals

soaked in a saturating solution of kukoamine A did not

yield the binding site of the compound but indicated

reduction of the active site disulfide bridge and probably

a modification of Cys52 (Cys53 in our amino acid

numbering, see Table 4).⁴⁸

Discu ssion

The aim of this study was to develop new NQs that

are specific and potent inhibitors of TcTR and are

reduced by the enzyme via a one-electron mechanism

leading to the generation of superoxide anion radicals.

The data obtained from the chemical synthesis and

enzymatic studies revealed that it is possible to design

specific and potent TcTR inhibitors from a lead struc-

ture, i.e. menadione and plumbagin, known to interact

with numerous NADPH-dependent disulfide oxidoreduc-

tases, like hGR, TcTR, and hTrxR. The chemistry which

has been developed has led to several building blocks

in the 1,4-NQ series, 9-11, which can easily be deriva-

tized with structurally diverse amines. This chemistry

has enabled us to design a series of 1,4-NQs, substituted

by amino side chains at C-2 and C-3 of the quinone

moiety of three natural 1,4-NQs, namely menadione,

plumbagin, and juglone already known for their try-

panocidal activities but lacking in specificity. Optimiza-

tion of TcTR inhibition and TcTR specificity versus

human disulfide reductase specificities was achieved

with the 3,3′-[polyaminobis(carbonylalkyl)]bis(1,4-NQ)

19-20 series, characterized by an optimal chain length

between the two NQ moieties (n ) 4-5). The results

presented in this paper suggest that inhibition of the

T(S)₂reduction alone is not sufficient for a strong

trypanocidal activity, as illustrated with several 3-poly-

amino-1,4-NQs (3_b, 6_c, 6_d) or 3,3′-[polyaminobis(carbo-

nylalkyl)]-bis(1,4-NQ)s (19-20; n ) 2 or 3), displaying

IC₅₀values below 10 µM. Indeed, the crucial parameter

for a high trypanocidal activity appears to be the

combination of both redox cycling activity and inhibition

of T(S)₂reduction. This point is clearly illustrated by

the properties of the 3,3′-[polyaminobis(carbonylalkyl)]-

bis(1,4-NQ) (20; n ) 4 or 5). The plumbagin derivatives

with n ) 4 were the most efficient subversive substrates

of TcTR, were the most potent inhibitors of the TR-

catalyzed T(S)₂reduction, and displayed the strongest

trypanocidal effects toward T. cruzi cultures. These

compounds combined the highest ratio (k_cat/K_m)/IC₅₀

(mean value from a -c derivatives ) 10.7 × 10¹⁰M^-2

s^-1) with the highest ratio (k_cat/K_m)/ED₅₀(mean value

from a -c derivatives ) 1.9 × 10¹⁰M^-2s^-1). Also in the

menadione series, derivatives with n ) 4 showed

maximum (k_cat/K_m)/IC₅₀and (k_cat/K_m)/ED₅₀mean values

of 5.7 × 10¹and 0.4 × 10¹⁰M^-2s^-1, respectively.

However, the menadione derivatives were much less

efficient subversive substrates than the corresponding

plumbagin derivatives, which also correlated with lower

antitrypanosomal activities and the redox potentials of

the parent molecules. Analyzing the structural features

of the NQ derivatives, our studies revealed that the

presence of a long alkylamino chain linking the two

Subversive substrates (turncoat inhibitors) of TR and

GR divert electrons from the physiologically catalyzed

reactions. Reduction of such molecules may occur at the

active site⁴⁹or at different site(s) as already suggested

for reversible inhibitors such as nitrofuran³⁷and quino-

ne.¹⁷In the case of the garlic-derived ajoene, the

compound reacts with the two-electron reduced enzymes

leading to a mixed disulfide at Cys58 (in hGR) and to

an increased NADPH oxidase activity of the modified

enzymes resulting in superoxide anion radical produc-

tion.⁴⁹This result fits with those obtained by Cenas and

co-workers who reported that TR alkylated by iodo-

acetamide at Cys52 maintained considerable quinone

reductase activity in the presence of juglone.¹⁷As TR

follows a ping-pong reaction mechanism,^11,50the un-

competitive inhibition type observed for numerous ni-

trofurans and quinones,³⁷and in particular for the

bis(NQ) 20_4-cstudied here, may indicate the accumula-

tion of dead-end NADP⁺‚TR‚T(S)₂complexes.⁵¹How-

ever, the binding mode of these compounds in TR is not

yet known. So far only one crystal structure of a TR/

inhibitor complex has been elucidated, namely with the

competitive inhibitor mepacrine.⁴⁰While no structure

of un/noncompetitive subversive substrate/TR complexes

is available, the structures of several complexes between

hGR and un/noncompetitive ligands have been described

in detail.^26-29These structures revealed that the inhibi-

tor molecules bind in a large cavity present at the dimer

interface, which is distinct from the GSSG, FAD, and

NADPH binding sites. Binding of these ligands at this

“third site” causes inhibition of the physiological GSSG

reduction, but none of the compounds is reduced by

hGR. In addition, ligands can have more than one

binding site, as illustrated for menadione which binds

to a lesser extent also at the NADPH and disulfide sites

in hGR,²⁶or as recently reported for acridine derivatives

in TcTR.⁵²Therefore it is tempting to speculate that the

respective “third site” in TR is the binding site of the

NQ molecules. The electron-transfer mechanisms oper-

ating between the flavin and NQ remain speculative

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 561

until structural studies will reveal if the compounds are

reduced at the NADPH site, the T(S)₂site, or the “third

site”. To gain insight in the locus where the NQ

reductase activity may take place, reduction of 20_4-c

was measured in the presence of mepacrine which acts

as a competitive inhibitor of TR versus T(S)₂. Only little

inhibition of NQ reduction was observed in the presence

of up to 100 µM mepacrine. This result may indicate

that the major binding locus of mepacrine in the T(S)₂

binding site is not the locus where the NQs are reduced,

as postulated by Cenas and co-workers.¹⁷

several NQs, among the most efficient subversive

substrates of TcTR and TcLipDH, suggest that NQ

reduction leading to the generation of superoxide anion

radicals is a promising strategy for the development of

new trypanocidal drugs. The design and synthesis of

more potent and specific NQs as turncoat inhibitors of

TcTR or TcLipDH are in progress, with a focus on the

preparation of new NQ moieties.

Exp er im en ta l Section

Abbr evia tion s: DMSO, dimethyl sulfoxide; DTNB, 5,5′-

dithiobis(2-nitrobenzoic acid); DTNBA, 5,5′-dithiobis{N-[3-

(dimethylamino)propyl]-2-nitrobenzamide}; GSH, reduced glu-

tathione; GSSG, glutathione disulfide; GR, glutathione reduc-

tase; hGR, human glutathione reductase; HBTU, 2-(1H-

benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophos-

phate; HOBt, N-hydroxybenzotriazole; LipDH, lipoamide de-

hydrogenase; TcLipDH, Trypanosoma cruzi lipoamide dehydro-

genase; SOD, superoxide dismutase; Trx, thioredoxin; TrxR,

thioredoxin reductase; hTrxR, human thioredoxin reductase;

T(SH)₂, reduced trypanothione; T(S)₂, trypanothione disulfide;

TR, trypanothione reductase; TcTR, Trypanosoma cruzi try-

panothione reductase.

A comparison of hGR and TcTR structures revealed

that the respective cavities at the 2-fold axis of both

enzymes have comparable dimensions but are clearly

different in shape. Our studies gained insight in the

accessibility of these cavities by a potential ligand. In

both enzymes, the interface site is connected to the two

catalytic sites by two short “primary” channels starting

near the disulfide substrate binding sites. In hGR, two

“secondary” channels were identified providing a direct

access to the surface of the protein. Several residues in

the cavity of TcTR show a marked flexibility, in par-

ticular those lining the “primary” channels which could

be crucial for large ligands to reach the “third site”.

En zym es. Recombinant TcTR (E.C. 1.6.4.8) was purified

from E. coli SG5 cells carrying the overproducing expression

vector pIBITczTR¹³following the published protocol.⁵³TcTR

concentration was determined by measuring the FAD absorp-

With respect to the specificity toward TR, only 47%

of the residues involved in the parasitic “third site” are

conserved in hGR. The most striking finding is that

none of the residues involved in the binding of un/

noncompetitive hGR ligands is conserved in the five

known TR sequences.²⁷As mentioned previously,⁹the

results obtained with the series of mono- and bis(NQ)

derivatives show that the presence of amino groups is

crucial for TR recognition. Under physiological condi-

tions, these groups are positively charged and could

correlate with the more pronounced acidity of the TcTR

cavity (overall charge of -IV for TcTR versus -II for

hGR). However this difference of charge must be

considered with caution due to the presence of several

histidine residues at the dimer interface. The preference

of TR over hGR for positively charged ligands even if

they bind at the “third site” may be due to the fact that

the cavity at the 2-fold axis - in contrast to hGR - is

obviously only accessible via the negatively charged TS₂

binding site. Comparison of the cavities may allow the

prediction that, in particular, Asp432, Gln69, and

Met400 may be important for ligand binding in TR.

These residues are easily accessible and display high

thermal parameters, as observed for Phe78 and Phe78′

in hGR which are directly involved in ligand binding.²⁶

tion at 461 nm (ꢀ₄₆₁

) 11.3 mM^-1cm^-1). 1 unit of TR

nm

corresponds to 1 µmol of T(S)₂reduced/min at 25 °C in assay

buffer A (20 mM Hepes, pH 7.25 containing 1 mM EDTA and

0.15 M KCl).¹¹The enzyme stock solutions used for kinetic

determinations were pure as judged from a silver stained

SDS-PAGE and had a specific activity of 137 U/mg for TcTR

in the T(S)₂reduction assay. Recombinant hGR (E.C. 1.6.4.2)

and native hTrxR (E.C. 1.6.4.5) were kindly provided by Prof.

Heiner Schirmer and Prof. Katja Becker, Biochemie-Zentrum,

Heidelberg.^54,55Using the DTNB assay, 1 unit of TrxR is

defined as the NADPH-dependent production of 2 µmol of

5-thio-2-nitrobenzoate/min (ꢀ₄₁₂

) 13.6 mM^-1cm^-1). 1 unit

nm

of GR activity is defined as the consumption of 1 µmol of

NADPH/min (ꢀ₃₄₀

) 6.22 mM^-1cm^-1). The enzyme stock

nm

solutions used for kinetic determinations were pure as judged

from a silver-stained SDS-PAGE and had a specific activity

of 27 U/mg for hTrxR in the DTNB assay,⁵⁶200 U/mg for hGR

in the GSSG assay, respectively. Recombinant TcLipDH (E.C.

1.8.1.4) was purified from E. coli J RG 1342 cells containing

the plasmid pQE/lpd/-mt gene¹⁴following a modified prepara-

tion described below. The specific activity in the forward

reaction was 490 U/mg which corresponds to that of the

authentic protein when taking into account that the activity

in the forward reaction is about 1.6 times higher than in the

reverse reaction.^12,14Cytochrome c and SOD (E.C. 1.15.1.1)

were purchased from Sigma. Pig heart LipDH (E.C. 1.8.1.4)

was obtained from Boehringer Mannheim. Q-Sepharose was

purchased from Pharmacia. Blue Sepharose is a Cibacrom blue

3GA-Sepharose 4B prepared as described.⁵⁷All other reagents

were of the highest available purity and were purchased from

Biomol, Boehringer, and Sigma. T(S)₂was purchased from

Bachem, Switzerland. ^DL-R-Dihydrolipoamide was prepared by

reduction of ^DL-lipoamide with NaBH₄. The procedure em-

ployed is a modification of that described by Reed et al.⁵⁸Using

10 mL of 1 M HCl in the acidification step yields crystalline

dihydrolipoamide directly.¹⁴Stock solutions of 40 mM lipo-

amide and 50 mM dihydrolipoamide were prepared in ethanol

and stored at -80 °C.

The lack of correlation between TR inhibition and the

trypanocidal effects among several NQs (14-15) sug-

gested a study of the redox cycling properties of several

menadione and plumbagin representatives with LipDH,

another flavoenzyme characterized in T. cruzi. Although

the redox cycling activities of both carbamates 14₃and

15₃were lower than those of the unsubstituted mena-

dione and plumbagin, the specificity of these derivatives

for the parasite versus mammalian enzyme was much

higher when compared with the parent molecules. The

data are consistent with the possible development of

potent and specific subversive substrates of TcLipDH

as potential trypanocidal drugs, rendering the enzyme

also an attractive target for antiparasitic chemotherapy.

TcTR, h Tr xR, a n d h GR In h ibition Stu d ies. 1. Tr yp a n o-

th ion e Disu lfid e Red u cta se Assa y. The standard assay

mixture contained 250 µM NADPH and 57 µM T(S)₂. IC₅₀

values were evaluated in duplicate in the presence of increas-

ing inhibitor concentrations (0.3-50 µM) and 1% DMSO as

final concentration. The reaction was started by adding

enzyme (10^-2units in 500-µL cuvette). Inhibition constants

and the type of inhibition were determined at 22 °C in

In conclusion, the strong antitrypanosomal action of

562 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4

Salmon-Chemin et al.

duplicate for compounds 3_b(at five different inhibitor concen-

trations 0, 5, 10, 20, 40 µM) and 20_4-c(at five different

inhibitor concentrations 0, 0.5, 1.0, 1.5, 2.0 µM) with six

different T(S)₂concentrations (20, 40, 60, 100, 150, 200 µM)

in the presence of 1% DMSO. The reaction was started by

adding enzyme (10^-2units in a total volume of 500 µL). When

inhibition of T(S)₂reduction by 20_4-cwas measured as a

function of T(S)₂concentration, the data were fitted by using

a nonlinear regression analysis software (Kaleidagraph) to the

following eq 8 for uncompetitive inhibition:⁵¹

and 25 µM 2,6-dichloroindophenol as reference inhibitor

containing 2% DMSO.

4. h GR In h ibition Stu d ies. The effect of the 1,4-NQ upon

hGR was studied in the standard assay^59,60developed for

microtiter plates, with the following slight modifications. All

enzymic and nonenzymic reactions were conducted in flat-

bottomed 96-well microtiter plates (Nunc Inc.) in a total

volume of 100 µL. The reactions were allowed to proceed for

20 min at room temperature (22-25 °C) and stopped by

addition of acetonitrile and DTNB. The plates were then read

using a 405-nm filter in the multiskan microplate reader, as

described above. All reactions with hGR were performed in

assay buffer C (100 mM sodium phosphate, 2 mM EDTA, pH

7.0). The components for the TrxR assay were added into each

well at the following final concentrations: 25 µM inhibitor,

200 µM GSSG, 500 µM NADPH, 1% DMSO. The reaction was

started by adding 80 µL of an enzyme solution containing 8 ×

10^-3units native hGR. Positive and negative controls were

made in duplicate by incubating the following components:

substrate solution (final concentration 2% DMSO) with and

without enzyme, substrate solution in the presence of enzyme,

and the reference inhibitor (25 µM isoalloxazine 3b²⁸(10-(3′,5′-

dichlorophenyl)-3-methylisoalloxazine), 2% DMSO).

Molecu la r Mod elin g. The protein structures were studied

on a Silicon Graphics work station using the Insight II module

of the MSI software. The CAST program was used for a

comparative analysis of the corresponding sites at the dimer

interface of TcTR and hGR.⁶¹This program identifies cavities,

pockets, and corresponding channel openings in protein struc-

tures. It provides also a measure of area, volume, and

circumference of each site and opening. The three-dimensional

structures of hGR and TcTR used were those of uncomplexed

enzymes accessible at the Protein Data Bank under 3grs²⁵and

1aog,⁴¹respectively, in which the TcTR structure is deposited

as a dimer, while for hGR crystallographic symmetry opera-

tions were applied to the monomer in order to obtain the

homodimer.

TcLip DH a n d P ig Hea r t Lip DH In h ibition Stu d ies. 1.

P u r ifica tion of Recom bin a n t TcLip DH. T. cruzi LipDH

was purified from E. coli J RG 1342 cells containing the

plasmid pQE/lpd/-mt¹⁴by a modified preparation scheme since

the purification scheme based on affinity chromatography on

AMP-Sepharose¹⁴yielded an enzyme species with a relatively

low specific activity when compared with the protein isolated

from T. cruzi epimastigotes.¹²In the procedure described here,

the AMP-Sepharose column was replaced by Q-Sepharose and

Blue Sepharose columns. The parasite enzyme was isolated

in an overall yield of 73% (Table 6) and the protein sample

proved to be much more stable during storage. Starting from

2.2 L of bacterial culture, the cell extract was prepared as

described.¹⁴The protein was precipitated by 42-70% am-

monium sulfate, centrifuged, and the pellet was dissolved in

buffer A (5 mM potassium phosphate, 1 mM EDTA, pH 7.5)

yielding fraction I. Following dialysis against buffer B (10 mM

potassium phosphate, 1 mM EDTA, pH 7.5) fraction I was

centrifuged and applied to a 50-mL Q-Sepharose column

equilibrated in buffer B. The column was washed with 150

mL of buffer B at a flow rate of 0.7 mL/min and TcLipDH was

eluted by 80 mM KCl in buffer B. Active fractions were pooled

and the protein was concentrated and washed with buffer C

(5 mM potassium phosphate, 1 mM EDTA, pH 7.0) in a

Centriprep 30 concentrator to give fraction II. Fraction II was

applied to an 8-mL Blue Sepharose column equilibrated in

buffer C at a flow rate of 0.8 mL/min. The column was washed

with 25 mL of buffer C, then with 25 mL of buffer D (20 mM

potassium phosphate, 1 mM EDTA, 1 mM dithiothreitol, pH

7.0) at a flow rate of 0.6 mL/min. TcLipDH was eluted with

buffer E (100 mM potassium phosphate, 100 mM KCl, 1 mM

EDTA, 1 mM dithiothreitol, pH 7.0). Active fractions were

pooled (fraction III) and concentrated by centrifugation in a

Centricon 30 concentrator. 50% Glycerol was added and the

pure protein was stored at -20 °C.

V_max[S]

V )

(8)

K_m+ [S] (1 + [I]/K_i)

2. Qu in on e Red u cta se Activity of TcTR a n d h GR. The

ability of TcTR to reduce NQs was assayed by monitoring the

substrate-dependent oxidation of NADPH at 340 nm (ꢀ₃₄₀

nm

) 6.22 mM^-1cm^-1). The assays, in a total volume of 120 µL,

contained 0.02 M Hepes, 1 mM EDTA, 0.15 M KCl, pH 7.25

at 22 °C, 200 µM NADPH, 0-100 µM NQ and 0.6 unit TR.

The NQs were dissolved in DMSO and the NADPH-oxidase

activity was measured at five different concentrations in the

presence of 1% DMSO. For the determination of K_mand V_max

values, the steady-state rates were graphically fitted by using

a nonlinear regression analysis software (Kaleidagraph) to the

Michaelis-Menten equation,⁵¹and the turnover number k_cat

and the dynamic specificity k_cat/K_mwere calculated. The initial

rate for intrinsic NADPH oxidase activity of TcTR has not been

subtracted from the rates measured in the presence of NQ,

since it proved negligible in comparison to the activity with

the bis(NQ). The amount of one-electron reduction of the NQs

by TR was monitored at 550 nm by coupling semiquinone

2+

formation to the reduction of 25 µM cytochrome c (ꢀ_550,cyt

c(Fe

)

) 18.91 mM^-1cm^-1). Inhibition of the reaction by SOD (0.6

unit in 120-µL cuvette) is a measure of superoxide anion

radical generation. The rate of cytochrome c reduction was

followed at 550 nm: in the absence of NQ without SOD

(NQ^-SOD^-) and with SOD (NQ^-SOD⁺), in the presence of NQ

without SOD (NQ⁺SOD^-) and with SOD (NQ⁺SOD⁺). As TR

reduces cytochrome c directly,¹¹the NQ-dependent cytochrome

c reductase activity of TcTR was calculated by subtracting the

rate of cytochrome c reduction in the absence of NQ (NQ^-SOD^-)

from the rate of cytochrome c reduction in the presence of NQ

(NQ⁺SOD^-). Addition of SOD allowed to calculate the NQ^-•

-

dependent cytochrome c reductase of TcTR by subtracting the

rate of cytochrome c reduction in the absence of NQ (NQ^-SOD⁺)

from the rate of cytochrome c reduction in the presence of NQ

(NQ⁺SOD⁺). Eq 9 gives the contribution of superoxide anion

radicals in the NQ-dependent cytochrome c reductase of TcTR:

[100(NQ⁺SOD⁺) - (NQ^-SOD⁺)]

100% -

(9)

(NQ⁺SOD^-) - (NQ^-SOD^-)

At the low-enzyme and high-cytochrome c concentrations

(25 µM) used in these experiments, the percentage of one-

electron reduction is calculated by dividing the rate of cyto-

chrome c reduction by twice the rate of NADPH oxidation in

the presence of 12.5 µM NQ.¹⁷The amount of one-electron

reduction of the NQs by hGR was monitored at 550 nm under

the conditions described above for TR.

3. h Tr xR In h ibition Stu d ies. The most potent TcTR

inhibitors were studied as inhibitors of human TrxR by using

the standard assay protocol developed in microtiter plates for

inhibitor screening.³⁹All reactions with hTrxR were performed

in assay buffer C (100 mM sodium phosphate, 2 mM EDTA,

pH 7.0). The components for the TrxR assay were added into

each well at the following final concentration: 25 µM inhibitor,

200 µM disulfide DTNBA, 500 µM NADPH, 2% DMSO. The

reaction was started by adding 80 µL of an enzyme solution

containing 8 × 10^-4units of native hTrxR. Positive and

negative controls were conducted in duplicate with the follow-

ing components: substrate (final concentration 2% DMSO)

with and without enzyme, substrate in the presence of enzyme,

2. Lip DH Assa y. LipDH activity was measured in the

forward reaction at 25 °C at 1 mM NAD⁺and 1.4 mM

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 563

Ta ble 6. Purification of 5.2 mg of TcLipDH from 2.2 L of Recombinant E. coli J RG1342 Cells^a

resulting

fraction

total protein

(mg)

total enzyme

activity (U)

specific activity

(U/mg)

overall yield

purification step

(%)

extraction

I

II

III

IV

V

4900

2500

1800

46

4800

4530

4450

3970

3500

1.0

1.8

2.5

86

490

100

94

93

83

73

42-70% (NH₄)₂SO₄precipitation, dissolved pellet

dialysis

ion-exchange chromatography on Q-Sepharose

affinity chromatography on Blue Sepharose,

concentration in an Amicon 30 concentrator

7.1

a

In crude fractions (I-III) the protein concentration was estimated by absorption measurements at 280 nm assuming that 1 mg/mL

protein corresponds to an absorbance of 1.0. The protein concentration of homogeneous TcLipDH (V) was determined at 280 nm with an

absorption of 1 corresponding to 1.8 mg/mL.

dihydrolipoamide in 50 mM potassium phosphate, 1 mM

EDTA, pH 7.0 (assay buffer) in a total volume of 1 mL.¹²The

assay mixture was preincubated in order to obtain a stable

baseline and the reaction was started by addition of the

enzyme. The absorption increase at 340 nm was monitored.¹⁴

Under these conditions, with about 50 ng/mL LipDH in the

assay the oxidase activity of the enzyme (see below) is

negligible.

from hamster spleen or at a ratio of 5:1 (2 × 10⁵/well) with T.

cruzi trypomastigotes derived from the MDCK fibroblast

overlay. Infected macrophages were then maintained in the

presence of the drug in a 3-fold dilution series, with quadru-

plicate cultures at each concentration, for 5 days for L.

donovani cultures and 3 days for T. cruzi cultures. After these

periods of drug exposure slides were methanol-fixed and

Giemsa-stained, and drug activity was determined by counting

the percentage of macrophages cleared of amastigotes in

treated cultures in comparison with untreated cultures.⁶⁴ED₅₀

values were determined by sigmoidal regression analysis (MS

xlfit). Sodium stibogluconate (NaSb^v) (Glaxo-Wellcome, Dart-

ford, U.K.) and nifurtimox (Bayer, U.K.) were used as the

respective control drugs.

3. Qu in on e Red u cta se Activity of Lip DHs. The NQ

reductase activity of LipDH was followed at 31 °C in a

Beckmann DU 65 spectrophotometer using microcuvettes. The

assay mixture of total volume of 90 µL contained 100 µM

NADH and 10 µM to 1 mM of the respective NQ in assay buffer

(50 mM potassium phosphate, 1 mM EDTA, pH 7.0). Several

stock solutions were prepared for each NQ. Thus all assays

contained 5 µL of DMSO which is important because of a slight

inhibition (about 20%) of the enzyme by DMSO. The reaction

was started by addition of the enzyme, and the absorption

decrease at 340 nm was monitored. The intrinsic oxidase

activity of LipDH was measured with 100 µM NADH in the

absence of any substrate. The assays contained between 0.1

and 2.5 µg/mL enzyme. The generation of semiquinones was

monitored by coupling the reduction of NQs to cytochrome c

reduction. The assay mixture contained in a total volume 90

µL, 50 mM potassium phosphate, pH 7.0, 100 µM NADH, 50

µM cytochrome c, and 50 µM of the respective NQ. The reaction

was started by addition of the enzyme and the absorption

increase at 550 nm was followed at 31 °C. At pH values < 7.2

reduction of cytochrome c by benzohydroquinones should be

negligible whereas reduction by the semiquinones is high.³

Thus the efficiency of one-electron reduction can be calculated

by dividing the rate of cytochrome c reduction by twice the

rate of NADH oxidation.⁶²At low enzyme activities and high

cytochrome c concentrations (30-40 µM) more than 90% of

the semiquinone formed is trapped.³To distinguish between

direct and superoxide-mediated reduction of cytochrome c by

the semiquinones, 5 µL of SOD (0.6 mg/mL) was added to the

assay after 3 min. The K_mand V_maxvalues were derived from

Lineweaver-Burk plots.

T. br u cei br u cei: Tests were performed in HMI-18 medium

as described above.⁶³Compounds were tested in triplicate in

a 3-fold dilution series from the highest concentration of 30

µM. Parasites were diluted to 2 × 10⁵/mL and added in equal

volume to the test compounds in 96-well, flat-bottom Microtest

III tissue culture plates (Becton Dickinson and Co., NJ ).

Appropriate controls with pentamidine isethionate (Rhone-

Poulonc-Rorer) as the positive were set up in parallel. Plates

were maintained for 3 days at 37 °C in a 5% CO₂-air mixture.

The antiparasitic activity of the compounds was determined

by use of a tetrazolium salt colorimetric assay⁶⁵on day 3. ED₅₀

values were determined by sigmoidal regression analysis.

Ack n ow led gm en t. The authors thank Prof. Heiner

Schirmer (Biochemie-Zentrum der Universita¨t Heidel-

berg) and Prof. Katja Becker who kindly provided

recombinant human GR and native human TrxR and

for many fruitful discussions. Anick Lemaire and Edith

Ro¨ckel are acknowledged for excellent technical as-

sistance, Ge´rard Montagne for NMR spectra, Dr. Steven

Brooks for proof-reading of the English, and Arnaud

Waeghe for his technical help in obtaining the protein

pictures presented in this paper. This work was sup-

ported by the Action Concerte´e Coordonne´e n°5 des

Sciences de la Vie (Interface Chimie-Biologie: “Biologie

Structurale et Pharmacochimie”) from the Ministe`re de

l’Education Nationale, de la Recherche et de la Tech-

nologie (fellowship attributed to L.S.) and by the Deut-

sche Forschungsgemeinschaft (DFG Kr 1242/1-4).

P a r a sites. L. donovani (strain MHOM/ET/67/L82) was

maintained routinely in special pathogen-free (SPF) female

Golden hamsters (Wright’s strain, Charles Rivers Ltd.) by

passage every 12 weeks. T. cruzi (strain MHOM/BR/OO/Y)

trypomastigotes were derived from MDCK fibroblasts in

Dulbecco’s modified Eagle medium (Life Technologies Ltd.,

Paisley, Scotland) with 10% heat-inactivated fetal calf serum

(HIFCS) (Harlan Sera-Lab., Crawley, U.K.) at 37 °C in an

atmosphere of 5% CO₂-air mixture. T. brucei brucei (strain

S427) bloodstream form trypomastigotes were maintained in

HMI-18 medium⁶³supplemented with 20% HIFCS at 37 °C

in a 5% CO₂-air mixture.

Su p p or tin g In for m a tion Ava ila ble: Details of chemical

procedures and analytical data. This material is available free

of charge via the Internet at http://pubs.acs.org.

In Vitr o An titr yp a n osom a l Bioa ssa ys. L. d on ova n i a n d

T. cr u zi: Peritoneal macrophages were harvested from female

CD1 mice (Charles River Ltd., Margate U.K.) by peritoneal

lavage 24 h after induction by soluble starch (Merk Ltd., Leics,

U.K.). After two washes in medium the exudate cells were

dispensed into 16-well Lab-tek tissue culture slides (Nunc Inc.,

IL) at 4 × 10⁵/mL in a volume of 100 µL of RPMI-1640 medium

(Sigma-Aldrich Co. Ltd., Dorset, U.K.) plus 10% HIFCS. After

24 h, macrophages were infected at a ratio of 10:1 (4 × 10⁶/

ml, 100 µL/well) with L. donovani amastigotes freshly isolated

Refer en ces

(1) O’Brien, P. J . Molecular mechanisms of quinone cytotoxicity.

Chem. Biol. Interact. 1991, 80, 1-41.

(2) Zuman, P. Substituent effects in organic polarography; Plenum

Press: New York, 1967.

(3) Iyanagi, T.; Yamazaki, I. One-electron-transfer reactions in

biochemical systems (V. Difference in the mechanism of quinone

reduction by the NADH dehydrogenase and the NAD(P)H

dehydrogenase (DT-diaphorase). Biochim. Biophys. Acta 1970,

216, 282-294.

564 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4

Salmon-Chemin et al.

(4) Nakamura, M.; Yamazaki, I. One-electron-transfer reactions in

biochemical systems. VI. Changes in electron-transfer mecha-

nism of lipoamide dehydrogenase by modification of sulfhyhdyl

groups. Biochim. Biophys. Acta 1972, 267, 249-257.

(5) Croft, S. L.; Evans, A. T.; Neal, R. A. The activity of plumbagin

and other electron carriers against Leihmania donovani and

Leishmania mexicana amazonensis. Ann. Trop. Med. Parasitol.

1985, 79, 651-653.

(28) Scho¨nleben-J anas, A.; Kirsch, P.; Mittl, P. R. E.; Schirmer, R.

H.; Krauth-Siegel, L. Inhibition of human glutathione reductase

by 10-arylisoalloxazines: crystallographic, kinetic, and electro-

chemical studies. J . Med. Chem. 1996, 39, 1549-1554.

(29) Krauth-Siegel, R. L.; Arscott, L. D.; Scho¨nleben-J anas, A.;

Schirmer, R. H.; Williams, C. H. Role of active site tyrosine

residues in catalysis by human glutathione reductase. Biochem-

istry 1998, 37, 13968-13977.

(6) Croft, S. L.; Hogg, J .; Gutteridge, W. E.; Hudson, A. T.; Randall,

A. W. The activity of hydroxynaphthoquinones against Leihma-

nia donovani. J . Antimicrob. Chemother. 1992, 30, 827-832.

(7) Boveris, A.; Docampo, R.; Turrens, J . F.; Stoppani, A. O. M. Effect

of â-lapachone on superoxide anion and hydrogen peroxide

production in Trypanosoma cruzi. Biochem. J . 1978, 175, 431-

439.

(8) Krauth-Siegel, R. L.; Coombs, G. H. Enzymes of parasite thiol

metabolism as drug targets. Parasitol. Today 1999, 15, 404-

409.

(9) Faerman, C. H.; Savvides, S. N.; Strickland, C.; Breidenbach,

M. A.; Ponasik, J . A.; Ganem, B.; Ripoll, D.; Krauth-Siegel, R.

L.; Karplus, P. A. Charge is the major discriminating factor for

glutathione reductase versus trypanothione reductase inhibitors.

Bioorg. Med. Chem. 1996, 4, 1247-1253.

(10) Williams, C. H., J r. In Chemistry and Biochemistry of Flavoen-

zymes; Mu¨ller, F., Ed.; CRC Press: Boca Raton, FL, 1992; Vol.

III.

(11) Krauth-Siegel, R. L.; Enders, B.; Henderson, G. B.; Fairlamb,

A. H.; Schirmer, R. H. Trypanothione reductase from Trypano-

soma cruzi: purification and characterization of the crystalline

enzyme. Eur. J . Biochem. 1987, 164, 123-128.

(12) Lohrer, H.; Krauth-Siegel, R. L. Purification and characterization

of lipoamide dehydrogenase from Trypanosoma cruzi. Eur. J .

Biochem. 1990, 194, 863-869.

(13) Sullivan, F. X.; Walsh, C. T. Cloning, sequencing, overproduction

and purification of trypanothione reductase from Trypanosoma

cruzi. Mol. Biochem. Parasitol. 1991, 44, 145-148.

(14) Scho¨neck, R.; Billaut-Mulot, O.; Numrich, P.; Ouaissi, M. A.;

Krauth-Siegel, R. L. Cloning, sequencing and functional expres-

sion of dihydrolipoamide dehydrogenase from the human patho-

gen Trypanosoma cruzi. Eur. J . Biochem. 1997, 243, 739-747.

(15) Blumenstiel, K.; Scho¨neck, R.; Yardley, V.; Croft, S. L.; Krauth-

Siegel, R. L. Nitrofuran drugs as common subversive substrates

of Trypanosoma cruzi lipoamide dehydrogenase and trypano-

thione reductase. Biochem. Pharmacol. 1999, 58, 1791-1799.

(16) Krauth-Siegel, R. L.; Scho¨neck, R. Trypanothione reductase and

lipoamide dehydrogenase as targets for a structure-based drug

design. FASEB J . 1995, 9, 1138-1146.

(17) Cenas, N. K.; Arscott, D.; Williams, C. H.; Blanchard, J . S.

Mechanism of reduction of quinones by Trypanosoma congolense

trypanothione reductase. Biochemistry 1994, 33, 2509-2515.

(18) Henderson, G. B.; Ulrich, P.; Fairlamb, A. H.; Rosenberg, I.;

Pereira, M.; Sela, M.; Cerami, A. Subversive substrates for the

enzyme trypanothione disulfide reductase: alternative approach

to chemotherapy of Chagas disease. Proc. Natl. Acad. Sci. U.S.A.

1988, 85, 5374-5378.

(19) J ockers-Scheru¨bl, M. C.; Schirmer, R. H.; Krauth-Siegel, R. L.

Trypanothione reductase from Trypanosoma cruzi. Catalytic

properties of the enzyme and inhibition studies with trypano-

cidal compounds. Eur. J . Biochem. 1989, 180, 267-272.

(20) Schirmer, R. H.; Mu¨ller, J . G.; Krauth-Siegel, R. L. Disulfide

reductase inhibitors as chemotherapeutic agents: the design of

drugs for trypanosomiasis and malaria. Angew. Chem., Int. Ed.

Engl. 1995, 34, 141-154.

(21) Bironaite, D. A.; Cenas, N. K.; Kulys, J . J . The inhibition of

glutathione reductase by quinones. Z. Naturforsch. C 1991, 46,

966-968.

(22) Lu¨o¨ng, R. M.; McKie, J . H.; Douglas, K. T.; Dascombe, M. J .;

Vale, J . Inhibitors of glutathione reductase as potential anti-

malarial drugs. Kinetic cooperativity and effect of dimethyl

sulphoxide on inhibition kinetics. J . Enzymol. Inhib. 1998, 13,

327-345.

(23) Cenas, N. K.; Rakauskiene´, G. A.; Kulys, J . J . One- and two-

electron reduction of quinones by glutathione reductase. Bio-

chim. Biophys. Acta 1989, 973, 399-404.

(24) Bironaite, D. A.; Anusevicius, Z.; J acquot, J .-P.; Cenas, N. K.

Interaction of quinones with Arabidopsis thaliana thioredoxin

reductase. Biochim. Biophys. Acta 1998, 1383, 82-92.

(25) Karplus, P. A.; Schulz, G. E. Refined Structure of glutathione

reductase at 1.54 Å Resolution. J . Mol. Biol. 1987, 195, 701-

729.

(30) Kallmayer, H. J . Synthese und Eigenschaften alkylierter Ami-

nonaphthochinone. Arch. Pharmaz. 1974, 37, 806-814.

(31) Ohta, S.; Hinata, Y.; Yamashita, M.; Kawasaki, I.; J inda, Y.;

Horie, S. One-step synthesis of 1,2,3,4-Tetrahydrobenzo[g]-

quinazoline-5,10-dione derivatives from vitamin K₃. Chem.

Pharm. Bull. 1994, 42, 1730-1735.

(32) Ohta, S.; Hinata, Y.; Kawasaki, I.; Yamashita, M. Photooxidation

of 2-Methylamino-3-(1-piperidinylmethyl)-1,4-naphthoquinone.

Chem. Pharm. Bull. 1994, 42, 2360-2362.

(33) Couladouros, E. A.; Plyta, Z. F.; Papageorgiou, V. P. A general

procedure for the efficient synthesis of (alkylamino)naphtho-

quinones. J . Org. Chem. 1996, 61, 3031-3033.

(34) Ohta, S.; Hinata, Y.; Yamashita, M.; Kawasaki, I.; Shoji, T.;

Yoshikawa, H.; Obana, Y. Synthetic approaches to 3-Amino-1,4-

naphthoquinone-2-carboxylic acid derivatives and photochemical

synthesis of novel 1,4,5,10-Tetrahydro-5,10-dioxo-2H-naphth[2,3-

d][1,3]oxazine derivatives. Chem. Pharm. Bull. 1994, 42, 1185-

1190.

(35) Salmon-Chemin, L.; Lemaire, A.; De Freitas, S.; Deprez, B.;

Sergheraert, C.; Davioud-Charvet, E. Parallel synthesis of a

library of 1,4-naphthoquinones and automated screening of

potential inhibitors of trypanothione reductase from Trypano-

soma cruzi. Bioorg. Med. Chem. Lett. 2000, 10, 631-635.

(36) Anderson, J . M.; Kochi, J . K. Silver(I)-catalyzed oxidative

decarboxylation of acids by peroxydisulfate. The role of silver-

(II). J . Am. Chem. Soc. 1970, 92, 1651-1659.

(37) Cenas, N.; Bironaite, D.; Dickancaite, E.; Anusevicius, Z.;

Sarlauska, J .; Blanchard, J . S. Chinifur, a selective inhibitor and

“subversive substrate” for Trypanosoma congolense trypano-

thione reductase. Biochem. Biophys. Res. Commun. 1994, 204,

224-229.

(38) Henderson, P. J . F. Statistical analysis of enzyme kinetic data.

In Enzyme assays: a practical approach; Eisenthal, R., Danson,

M. J ., Eds.; 1995.

(39) Davioud-Charvet, E.; Becker, K.; Landry, V.; Gromer, S.; Loge´,

C.; Sergheraert, C. Synthesis of alternative substrates for

trypanothione reductase and thioredoxin reductase: an inex-

pensive microtiter colorimetric assay for inhibitor screening.

Anal. Biochem. 1999, 268, 1-8.

(40) J acoby, E. M.; Schlichting, I.; Lantwin, C. B.; Kabsch, W.; and

Krauth-Siegel, R. L. Crystal structure of the Trypanosoma cruzi

trypanothione reductase-mepacrine complex. Proteins 1996, 24,

73-80.

(41) Zhang, Y.; Bond, C. S.; Bailey, S.; Cunningham, M. L.; Fairlamb,

A. H.; Hunter, W. N. The crystal structure of trypanothione

reductase from the human pathogen Trypanosoma cruzi at 2.3

Å resolution. Protein Sci. 1996, 5, 52-61.

(42) Misaka, E.; Kawahara, Y.; Nakanishi, K. Studies on menadione

reductase in yeast. III The identity of menadione reductase with

lipoamide dehydrogenase. J . Biochem. 1965, 58, 436-443.

(43) Bergeron, R. J .; Feng, Y.; Weimar, W. R.; McManis, J . S.;

Dimova, H.; Porter, C.; Raisler, B.; Phanstiel, O. A comparison

of structure-activity relationships between spermidine and

spermine analogue antineoplastics. J . Med. Chem. 1997, 40,

1475-1494.

(44) Bonnet, B.; Soullez, D.; Girault, S.; Maes, L.; Landry, V.;

Davioud-Charvet, E.; Sergheraert, C. Trypanothione reductase

inhibition/trypanocidal activity relationships in

a 1,4-bis(3-

aminopropyl)piperazine series. Biorg. Med. Chem. 2000, 8, 95-

103.

(45) Girault, S.; Baillet, S.; Horvath, D.; Lucas, V.; Davioud-Charvet,

E.; Tartar, A.; Sergheraert, C. New potent inhibitors of trypano-

thione reductase from Trypanosoma cruzi in the 2-aminodiphen-

ylsulfides series. Eur. J . Med. Chem. 1997, 32, 39-52.

(46) Bonnet, B.; Soullez, D.; Davioud-Charvet, E.; Landry, V.; Hor-

vath, D.; Sergheraert, C. New spermine and spermidine deriva-

tives as potent inhibitors of Trypanosoma cruzi trypanothione

reductase. Bioorg. Med. Chem. 1997, 5, 1249-1256.

(47) Ponasik, J . A.; Strickland, C.; Faerman, C.; Savvides, S., Karplus,

P. A.; Ganem, B. Kukoamine A and other hydrophobic acyl-

polyamines: potent and selective inhibitors of Crithidia fascicu-

lata trypanothione reductase. Biochem. J . 1995, 311, 371-375.

(48) Bonse, S.; Krauth-Siegel, R. L. Irreversible inhibitors of T. cruzi

trypanothione reductase: kinetic and crystallographic studies.

In Flavins and Flavoproteins; Rudolf Weber Agency for Scientific

Publication: Berlin, 1999.

(26) Karplus, P. A.; Pai, E. F.; Schulz, G. E. A crystallographic study

of the glutathione binding site of glutathione reductase at 0.3-

nm resolution. Eur. J . Biochem. 1989, 178, 693-703.

(27) Savvides, S.; Karplus, P. A. Kinetics and crystallographic

analysis of human glutathione reductase in complex with a

xanthene inhibitor. J . Biol. Chem. 1996, 271, 8101-8107.

1,4-NQs as Substrates of TR and LipDH

J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 4 565

(49) Gallwitz, H.; Bonse, S.; Martinez-Cruz, A.; Schlichting, I.;

Schumacher, K.; Krauth-Siegel, R. L. Ajoene is an inhibitor and

subversive substrate of human glutathione reductase and Try-

panosoma cruzi trypanothione reductase: crystallographic, ki-

netic, and spectroscopic studies. J . Med. Chem. 1999, 42, 364-

372.

(50) Leichus, B. N.; Bradley, M.; Nadeau, K.; Walsh, C. T.; Blanchard,

J . S. Kinetic Isotope effect analysis of the reaction catalyzed by

Trypanosoma congolense trypanothione reductase. Biochemistry

1992, 31, 6414-6420.

(51) Segel, I. H. Enzyme Kinetics; Wiley: London, 1975.

(52) Bonse, S.; Santelli-Rouvier, C.; Barbe, J .; Krauth-Siegel, R. L.

Inhibition of Trypanosoma cruzi trypanothione reductase by

acridines: kinetic studies and structure-activity relationships.

J . Med. Chem. 1999, 42, 5448-5454.

(53) Krauth-Siegel, R. L.; Sticherling, C., J o¨st, I.; Walsh, C. T., Pai,

E. F.; Kabsch, W., Lantwin, C. B. Crystallization and prelimi-

nary crystallographic analysis of trypanothione reductase from

Trypanosoma cruzi, the causative agent of Chagas’ disease.

FEBS Lett. 1993, 317, 105-108.

(54) Nordhoff, A.; Bu¨cheler, U. S.; Werner, D.; Schirmer, R. H.

Folding of the four domains and dimerization are impaired by

Gly446fGlu exchange in human glutathione reductase. Impli-

cations for the design of antiparasitic drugs. Biochemistry 1993,

29, 4022-4030.

(55) Gromer, S.; Arscott, L. D.; Williams, C. H.; Schirmer, R. H.;

Becker, K. Human placenta thioredoxin reductase. J . Biol.

Chem. 1998, 273, 20096-20101.

(58) Reed, J . W.; Koike, M.; Levitch, M. E.; Leach, F. R. Studies on

the nature and reactions of protein-bound lipoic acid. J . Biol.

Chem. 1958, 232, 143-158.

(59) Baker, M. A.; Cerniglia, G. J .; Zaman, A. Microtiter plate assay

for the measurement of glutathione and glutathione disulfide

in large numbers of biological samples. Anal. Biochem. 1990,

190, 360-365.

(60) Clarke, J . B.; Fortes, M. A.; Giovanni, A.; Brewster, D. W.

Modification of an enzymatic glutathione assay for the microtiter

plate and the determination of glutathione in rat primary

cortical cells. Toxicol. Methods 1996, 6, 223-230.

(61) Liang, J .; Edelsbrunner, H.; Woodward, C. Anatomy of protein

pockets and cavities: measurement of binding site geometry and

implications for ligand design. Protein Sci. 1998, 7, 1884-1897.

(62) Vienozinskis, J .; Butkus, A.; Cenas, N.; Kulys, J . The mechanism

of the quinone reductase reaction of pig heart lipoamide dehy-

drogenase. Biochem. J . 1990, 269, 101-105.

(63) Hirumi, H.; Hirumi, K. Continuous cultivation of Trypanosoma

brucei blood stream forms in

a medium containing a low

concentration of serum protein without feeder cell layers. J .

Parasitol. 1989, 75, 985-989.

(64) Neal, R. A.; Croft, S. L. An in vitro system for determining the

activity of compounds against the intracellular amastigote form

of Leishmania donovani. J . Antimicrobiol. Chemother. 1984, 14,

312-317.

(65) Ellis, J . A.; Fish, W. R.; Sileghem, M.; McOdimba F. A colori-

metric assay for trypanosome viability and metabolic function.

Vet. Parasitol. 1993, 50, 143-149.

(66) Hubbard S. J .; Thornton J . M. NACCESS Computer Program;

Department of Biochemistry and Molecular Biology, University

College London.

(56) Holmgren, A.; Bjo¨rnstedt, M. Thioredoxin and thioredoxin

reductase. Methods Enzymol. 1995, 252, 199-208.

(57) Dean, P. D.; Watson, D. H. Protein purification using im-

mobilized triazine dyes. J . Chromatogr. 1979, 165, 301-319.

J M001079L

Article Doi

2- and 3-substituted 1,4-naphthoquinone derivatives as subversive substrates of trypanothione reductase and lipoamide dehydrogenase from Trypanosoma cruzi: Synthesis and correlation between redox cycling activities and in vitro cytotoxicity

DOI: 10.1021/jm001079l

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm001079l

Full text of DOI:10.1021/jm001079l

Products guided by the article

R&D Labs maybe for 330451-76-8

Relevant to this article

Hot Product