role of classical nitroreductase and o-acetyltransferase on the mutagenicity of nifurtimox and eight...

Environmental and Molecular Mutagenesis 26:86-93 (1 995)

Role of Classical Nitroreductase and O-Acetyhransferase on the Mutagenicity of Nifurtimox and Eighi Derivatives

in Salmonella iyphimurium

Juan Jurado and Carmen Pueyo

Departamento de Bioquimica y Biologia Molecular, Universidad de Cordoba, Cordoba, Spain

This study investigates the mutagenicity of nifurtimox (NFX) and eight analogues in Salmonella ?.phi- murium indicator strains that possess different levels of classical nitroreductase or Oacetyltransferase activities. The NFX analogues tested replace the 3-methyl-4-yl-tetrahydro- 1 ,4-thiazine- 1 , l -dioxide group of the parent compound with the following other groups: indazol-1 -yl (1 G); pyrazol-l-yl ( 1 6); benzimidazol-l-yl ( 1 E); 1,2,4-triazo1-4-yl ( 1 D); 1- methyl-3-methylthio-1,2,4-triazol-4-yl-5-thione (1 I); 3,5-bis(methylthio)-l,2,4-triazol-4-yl (1 H); 1 -ado- mantyl (ADA); and 4,6diphenylpyridin-1 -yl-2-one (1 K). In the genetic backgrounds of the standard Ames tester strains TA98 and TAl 00, these bacteria combine the t-arabinose resistance forward mutation assay (Ara test) with a deficiency or overproduction of either nitroreduction or Oacetylation. The Ara test revealed, in agreement with previous findings, important differences between TA98 and TA100 and demonstrated, moreover, that these genetic differences are of significance in mutagenicity testing with nitrofuran compounds. The Ara test also indicated dissimilarities between the metabolic activation of NFX and its analogues, these compounds being classified in three different groups according

to their mutagenicity toward strain BA14 (genetic background of TA98) and its derivatives. The first group included analogues (1 G, 1 E, 1 I, and ADA) that showed similar mutagenic potency in all bacte rial strains. These compounds are considered not to be substrates for both classical nitroreductase and Oacetyltransferase. The second group included compounds (analogues 1 B and 1 K, and the reference drug NFX) with increased mutagenicity toward the strain overproducing the classical nitroreductase, and/or reduced mutagenicity toward the corresponding deficient bacteria. These compounds are considered to be activated by the classical nitroreductase. The third group [analogues 1 D and 1 H) was activated by bacterial Oaceiyltransfer- ase, and consequently showed increased and de creased mutagenicity with the particular overprc- ducer or deficient bacterial strain, as compared to their isogenic parentals. Previous reports have pointed out interest in NFX analogue 1 H as a promising candidate for the replacement of NFX. The present study further enhances the putative interest of compound 1 H, based on the different metabolic activation pathway exhibited by this analogue as compared to the parental drug, NFX. o 1995 Wiley-tiss, Inc.

Key words: nitroreductase, Oacetyltransferase, Ara test, nifurtimox, nifurtimox derivatives

INTRODUCTION

5-nitrofurans are a large group of synthetic nitro compounds used worldwide as human and veterinary drugs, food additives, or preservatives. Nifurtimox (NFX) is the most effective 5-nitrofuran for the treatment of Chagas disease (American trypanosomiasis), which affects 10 million people in Central and South America and whose etiological agent is Trypanosoma cruzi.

Nifurtimox shows genotoxic effects in numerous or- ganisms from prokaryotes to mammals [for summary see Alonso-Moraga and Graf, 19891. Besides the potential genotoxicity of NFX, its use in chemotherapy is severely limited by the occurrence of toxicity associated with treatment at high total dosage levels, and also because NFX has significant side effects and may not be curative in certain circumstances. For all these rea- sons, and in consideration of the enormous social and economical implications that the control of trypanoso- 0 1995 Wiley-Liss, Inc.

miasis would have in Central and South America, the availability of safe, effective, and inexpensive substi- tutes for NFX is important.

New derivatives structurally related to NFX have been synthesized from 5-nitrofurfural and the corresponding N-aminoheterocyclic compound, and have been tested for trypanocidal activity against T. cruzi as potential alterna- tive antichagasic drugs [Mester et al., 19871. Compounds in which the heterocyclic moieties are of the 1,2,4-triazol- 4-yl and pyridin-l-yl groups clearly show higher activity against T. cruzi than does NFX.

The genotoxicities of eight of the new derivatives, with

Received December 5, 1994; revised and accepted March 29, 1995.

Address reprint requests to Carmen Pueyo, Departamento de Bioquimica y Biologia Molecular, Universidad de Chdoba, Avda. de Medina Aza- hara sln, 14071 Cbrdoba, Spain.

Nitroreduction and Transacetylation in Nifurtimox and Derivative Mutagens 87

NFX as a reference, have been assessed both in bacterial [Alejandre-DurBn et al., 19881 and eukaryotic [Alonso- Moraga and Graf, 19891 test systems. All the NFX derivatives assayed are bacterial mutagens in the absence of an exogenous mammalian activating system such as the S9 activation mixture. Nevertheless, their mutagenicities differ markedly in the L-arabinose forward mutation assay with Salmonella typhimurium, covering a nearly 1,000- fold range in potency from very potent to weak mutagens [Alejandre-DurBn et al., 19881. Comparatively, only NFX and two of the analogues (compounds 1B and ADA) are clearly mutagenic and recombinogenic in the wing somatic test with Drosophila melanogaster, with the remaining compounds classified as nongenotoxic [Alonso- Moraga and Graf, 19891. Alejandre-DurBn et al. [1988] and Alonso-Moraga and Graf [ 19891 indicated that analogues with high antiparasitic activity but low bacterial mutagenicity and no genotoxicity in Drosophila are promising candidates for the replacement of NFX.

Nitrofurans do not interact with DNA per se, but they require metabolic conversion for exerting mutagenicity in bacteria. These compounds are metabolically activated to the corresponding hydroxylamines through the reduction of their functional nitro group [McCalla, 19831. Clas- sical nitroreductase (the major 02-insensitive nitrofurazone reductase activity of the cell) is thought to be respon- sible for the conversion of some nitrofurans to the proximate mutagens, because bacterial strains defective in this enzyme are resistant to the mutagenic effects of these compounds [Rosenkranz and Mermelstein, 19831. Comparatively, in the activation of nitroarenes, the resulting N-hydroxyarylamines are then esterified by 0- sulfonylation, N,O-transacetylation, or 0-acetylation [Weisburger, 19881. In bacteria, the prevalent metabolic pathway for the activation of N-hydroxyarylamines is 0- acetylation [Saito et al., 1983, 19851. Thus, bacteria defective in 0-acetyltransferase activity are not susceptible to the mutagenic action of typical nitroarenes [McCoy et al., 19831.

Bacterial strains that are deficient in or overproduce either classical nitroreductase or 0-acetyltransferase have recently been incorporated into the S. typhimurium arabinose assay in order to study the metabolism and mutagenicity of nitro-containing compounds. The resulting strains have been shown to be resistant or extremely sensitive to the mutagenic action of typical nitrofurans and nitroarenes [Jurado et al., 19941.

The present study investigated the role of nitroreduction as a metabolic activation pathway in the bacterial mutagenicity of NFX and eight of its analogues, since it has been proposed that bacterial mutagenicity assays might give an exaggerated picture of the mutagenic potency of nitrofurans and other nitro compounds with respect to the effect (mutation and tumor production) seen

in animals. The putative role of bacterial 0-acetylation was also studied for comparison.

MATERIALS AND METHODS

Chemicals

The names, molecular formulas, molecular weights (in parentheses), and structures of nifurtimox (NFX) and derivative compounds ( lG, lB, 11, IE, ID, IH, ADA, and 1K) are shown in Figure 1. NFX was obtained from Bayer (Wuppertal, Germany). The NFX analogues were synthesized as previously described [Mester et al., 19871 by R.M. Claramunt and co-workers in the Departamento de Quimica OrgBnica, Facultad de Ciencias, U.N.E.D., Madrid, Spain. Nitrofurantoin (NFA), nitrofurazone (NFZ), and N-methyl-"-nitro-N-nitrosoguanidine (MNNG) were from Sigma Chemicals (Munich, Germany); niridazole (NRZ) was from Ciba- Geigy (Barcelona, Spain); and 1,8-dinitropyrene (1,8DNP) and 4-nitro- quinoline-N-oxide (4NQO) were from Aldrich-Chemie (Steinheim, Ger- many). All compounds were dissolved in dimethylsulfoxide (Merck, Darmstadt, Germany), and used without further purification.

Bacterial Strains and Plasmids

The bacterial strains and the plasmids used in this study are listed in Table I. S. typhirnurium BA14 is as TA98 but araD.531; BA16 is as TAlOO but rough and uruD53Z. These two bacteria and the corresponding derivatives, which are deficient or else overproduce either classical nitroreductase or 0-acetyltransferase, were previously constructed in our laboratory [Jurado et al., 1993, 19941. The cloning vector plasmid and its derivatives carrying the gene encoding the classical nitroreductase or the 0-acetyltransferase enzyme were described in Watanabe et al. [1989, 19901. All bacteria carried the araD53Z allele used in the L- arabinose forward mutation assay. Master cultures were kept frozen at -80°C. and strains were reisolated as recommended by Maron and Ames [1983]. LB broth (1% Bacto tryptone, 0.5% Bacto yeast extract, 1% NaC1) was used for overnight cultures. Ampicillin (25 pg/ml) and/ or tetracycline (6.25 pg/ml) was added to the broth, as necessary.

Mutation Test

The L-arabinose mutation assay (Ara test) that provides a general test for all types of mutations [Pueyo and Ruiz-Rubio, 19841 was canied out as previously described [Hera and Pueyo, 19861. Briefly, the bacterial tester strain and the compound to be assayed were combined in molten top agar. A mean of 2.5 X lo7 viable bacteria was plated per selective agar plate. Forward mutations from L-arabinose sensitivity to L-arabinose resistance (Ara') were scored in selective plates supple- mented with 50 pg of D-glucose. Ara' mutant colonies were counted automatically (Analytical Measuring System Ltd., Shirehill, England, model 40-10) after 3 days at 37°C. All data represent averages from at least 2 duplicate plates. Each mutagenesis assay was repeated on at least 2 separate occasions, with a wide range of chemical concentrations. The number of induced mutants per nmol or pmol (defined as mutagenic potency) was estimated as the slope of the linear regression line fitted to the increasing portion of the corresponding dose-response curve.

RESULTS

The induction of forward mutations to L-arabinose resistance (Ara") by NFX and eight analogues was investigated with bacterial strains BA14 and BA16 and their deficient derivatives, and with the parentals BA140 and BA 160 carrying the cloning vector plasmid pBR322-Aps

88 Jurado and Pueyo

1G

-N

1E

-N

ADA

1B

1D

C7H5N503 (207) R = 1,2,4-triazol-4-~1

Ph 1K

-N>Ph

0

c g y 5 0 3 S2(299) R l-methyl-3-methylthio-

1,2,4-tria~ol-4-yl-5-thione

1H MeS \ hJ

-N I MeS P

CgHgN503 S2(299) R = 3,5-bis(methylthio)-

1,2,4-triazol-4-y1

NFX n

-N Me YSO*

C15H18N203 (274) C22H15N304 (385) Cl@13N305S (287) R = 1-adamantyl R = 4,6-diphenylpyridin- R = 3-methyl-4-yl-tetrahydro-

1 -yl-Zone 1 ,4-thiazine- 1 ,1 -dioxide Fig. 1 . structure to all NFX analogues is indicated at top.

Abbreviations, molecular formulas, and structures of compounds tested for mutagenicity. Common


TABLE 1. S. typhimurium Strains and Plasmids

Enzyme activity (nmolhidmg protein)"

Nitrofurazone 1soniazid-N- S traidplasmid Relevant genetic characteristics reductase acetvltransferase

Strains derived from BA14

BA14 BA14NR BA14DNP BA140 BA146 BA149

hisD3052, araD531. A(uvrB-bio), pKMlOl Defective in classical nitroreductase Defective in 0-acetyltransferase With pBR322-Ap' With pYG216 With pYG219

Strains derived from BA16

BA16 BA16NR BA 16DNP BA160 BA166 BA169

hisG46, araD531, A(uvrB-bio), pKMlOl Defective in classical nitroreductase Defective in 0-acetyltransferase With pBR322-Aps With pYG216 With pYG219

26.0 ? 1.2 0.1 2 0.1

22.4 f 2.5 21.2 ? 1.8

1,108.6 f 153.3 21.8 ? 1.6

66.9 2 2.8 0.1 ? 0.1

68.0 5 5.5 65.1 2 6.4

758.0 ? 136.2 66.9 2 1.3

3.8 2 0.8 3.1 ? 0.4 0.2 2 0.04 3.2 f 0.6 2.4 t 1.3

211.0 5 4.2

3.9 5 0.1 3.5 f 0.4 0.3 2 0.2 3.2 ? 1.1 2.6 f 0.9

291.0 5 3.3

Plasmids derived from pBR322-Ap'

pBR322-Ap' pYG216 pYG219

As pBR322 but sensitive to ampicillin With the gene encoding the classical nitroreductase With the gene encoding the 0-acetyltransferase

"Enzyme activity was assayed as previously described [Jurado et al., 19941. Data represent the average 2 SD from 2-3 independent determinations.

and their derivatives, which overproduce either classical nitroreductase or 0-acetyltransferase. BA14 and BA140 and derivatives, as well as BA16 and BA160 and derivatives, constitute isogenic sets of bacterial strains constructed in the genetic background of the well-known Ames tester strains TA98 and TA100, respectively [Jur- ado et al., 1993, 19941.

In Figure 2, comparative data are presented on the mutagenicity of NFX and its analogues in BA140 and BA160 and the corresponding bacteria with multicopy plasmids carrying either the classical nitroreductase or O-acetyltransferase gene of S. ryphimurium. The Ara test showed different mutagenic responses in the genetic background of BA14 as compared to BA16, the derivatives of strain BA14 being in most cases more sensitive than those of BA16. Of particular relevance are the differences found with NFX and its analogues, compounds 1K and 1H.

The mutagenicity of NFX was clearly enhanced in bacteria BA 146, overproducing classical nitroreductase. A less unequivocal effect of nitroreduction was detected with two of the NFX analogues, compounds 1G and 1K. The overproduction of 0-acetyltransferase had a pro- nounced effect on the mutagenic activity of compounds 1D and 1H in the BA149 strain. The mutagenicities of the remaining chemicals, lB, lE, 11, and ADA, were unaffected by the overproduction of either classical nitroreductase or 0-acetyltransferase. The enhanced mutagenicity of compounds lG, lK, and NFX, and of compounds

1D and lH, in bacteria with increased nitroreductase or 0-acetyltransferase levels, was less evident in the genetic background of BA16 as compared to that of BA14.

The mutagenic activities of NFX and its analogues in bacteria defective in either classical nitroreductase (BA14NR) or 0-acetyltransferase (BA14DNP) are shown in Figure 3. Table I1 compares these mutagenic activities with those of the corresponding overproducing strains (BA146 or BA149). To make the comparison easier, we calculated: 1) the numbers of induced Ara' mutants per nmol (defined as the mutagenic potency) of each chemical and strain, and 2) the relative sensitivity of each bacterial derivative by assigning the value 1.00 to the number of induced mutants per nmol detected in the corresponding parental strain (BA14 or BA140). BA16 and its defective derivatives with the genetic background of the standard TAlOO were, in general, less responsive than BA14 and derivatives, as mentioned previously in relation to the overproducing bacteria. For this reason, Table I1 shows only the results with BA14 and isogenic bacteria.

The defective bacterial strains confirmed the dependency of NFX and its analogue 1K on nitroreduction, and the dependency of compounds 1D and 1H on 0- acetylation. Two exceptional cases were the NFX derivatives 1G and 1B: 1) the small but consistent increased mutagenicity of 1G in bacteria overproducing the classical nitroreductase (Fig. 2) was not parallel to a decreased activity in the defective strain and 2) compound 1B with

90 Jurado and Pueyo

BA14 BA16 derivatives derivatives

BA14 BA16 derivatives derivatives

0.0 0.2 0.40.0 0.2 0.4

0 2 4 0 2 4

2500

0

18000

9000

0

5000

2500

0

10000

5000

0

12000

6000

0

0.0 2.5 5.0 0.0 2.5 5.0

0 5 1 0 0 5 10

4000 4000

0 10 20 0 10 20

800 6000

400 400 3000

0 0

0 10 20 0 10 20

0 6 1 2 0 6 12 0 15 30 0 15 30

8000

4000

0

6000

3000

0

B

0 15 30 0 15 30

Dose (nmol I plate) Fig. 2. Effect of overproducing classical nitroreductase or 0-acetyltransferase on mutagenic responses. The number of Ara' mutants was plotted as the function of the tested dose of the chemical, after subtracting the corresponding spontaneous values. Averages of the spontaneous numbers of h a ' mutantdplate were 794 -C 99 for BA140 (O), 739 2 94 for BA146 (W), and 852 2 104 for BA149 (A); 639 f 194 for BA160 (0). 578 2 101 for BA166 (0). and 700 + 115 for BA169 (A).

a near 4-fold reduction in mutagenicity in bacteria lacking the classical nitroreductase did not increase its mutagenic potency in the corresponding overproducer derivative.

The mutagenic responses in BA14 and derivative bacteria, of standard (4NQO and MNNG), nitroreductase-dependent (NFA, NFZ, and N U ) , or 0-acetyltransferase- dependent (1,8DNP) mutagens are shown in Table I11 for comparison to NFX and its analogues.

DISCUSSION

The Ara forward mutation assay with S. typhimurium has recently been combined with a deficiency or overproduction in either classical nitroreductase or O-acetyltransferase [Jurado et al., 19941. Two different sets of isogenic bacterial strains were constructed in the genetic background of Ames tester strains TA98 and TA100, respec-

Nitroreduction and Transaceiylation in Nifurtimox and Derivative Mutagens 91

4500

0

0 10 20

1000

0 5 10

5000

0.0 1.5 3.0

10000

5000

0

6000

0

0 1 2

0 10 20 0 15 30

Dose (nmole / plate) Fig. 3. Effect of deficiency in classical nitroreductase or 0-acetyltransferase on mutagenic responses. The number of Ara' mutants was plotted as the function of the tested dose of the chemical, after subtracting the corresponding spontaneous values. Averages of the spontaneous numbers of Ara' mutantdplate were 992 2 176 for BA14 (O), 922 5 227 for BA14NR (M), and 776 5 248 for BA14DNP (A).

TABLE I I . Mutagenic Potencv of NFX and Analogues in Different Bacterial Strains*

Control Nitroreductase Transacetylase

Chemical BA14 BA140 BAI4NR BA146 BA l4DNP BA149 ~~

IG 17,015 2 2,419 (1.00) IB 3,550 2 1,369 ( I .OO) IE 6,755 2 1,247 ( I .00) ID 474 2 175 (1.00) I I 579 L 15 (1.00) IH 254 2 84 ( 1 .OO) ADA 193 L 13 (1.00) IK 154 C 55 (1.00) NFX 27 5 8 (1.00)

~ ~~

26,565 ? 3,339 (1.00) 3,305 2 1,242 (1.00) 5,445 ? 101 (1.00)

409 2 2 (1.00) 510 2 35 (1.00) 208 ? 23 (1.00) 63 2 13 (1.00)

173 2 35 (1.00) 26 2 3 (1.00)

13,972 C 3,271 (0.82) 948 2 1 10 (0.27)

4,883 2 26 (0.72) 342 t I19 (0.72) 327 t 47 (0.56) 216 C 54 (0.85) 200 t 15 (1.04) 39 C 27 (0.25) 7 t 1 (0.26)

44,110 2 8,873 (1.69) 3,236 ? 274 (0.98) 5,129 C 1,850 (0.94)

534 C 213 (1.30) 690 ? 372 (1.35) 210 2 56 (1.01)

51 t 6 (0.81) 362 2 49 (2.09) 662 t 302 (25.46)

13,951 C 388 (0.82) 2,513 t 609 (0.71) 5,925 t 64 (0.88)

159 t 1 I (0.33) 413 t 77 (0.71) 134 2 23 (0.53) 162 t 17 (0.84) 159 t 9 (1.03) 26 t 7 (0.96)

18,796 t 2.763 (0.70) 3,050 t 1.150 (0.92) 4,083 2 202 (0.75) 2.227 t 350 (5.44)

475 t 97 (0.93) 1,438 2 392 (6.91)

58 2 19 (0.92) 140 t 47 (0.81) 24 t 9 (0.92)

*Numbers are the average 5 SD of the mutagenic potency, expressed as Arar-induced mutantshmol, from 2-3 independent experiments. These values were calculated from points within the linear initial portion of the corresponding dose-response curves (representative experiments are those of Figs. 2, 3). Numbers in parentheses are relative to BA14 for strains BA14NR and BA14DNP. and to BA140 for strains BA146 and BA149, respectively; bold numbers indicate S.D.

tively [Jurado et al., 1993, 19941. This study reports on the mutagenic responses of these new constructions to NFX and eight structurally related analogues.

Chemicals with reduced mutagenicity in the deficient strain displayed, as expected, increased mutagenicity in

the corresponding overproducer bacteria; the exception was analogue 1B. Einisto et al. [1991] have identified chemicals with reduced mutagenicity in the classical nitroreductase defective strain that do not show increased mutagenicity in bacteria overexpressing the correspond-

92 Jurado and Pueyo

TABLE 111. Mutagenic Potency of Reference Chemicals*

Control Nitroreductase Transacetylase

Chemical BA14 BA 140 BA I4NR BA 146 BA14DNP BA149

4NQO 51.161 (1.00) 59,533 ( 1 .00) 46.646 (0.91) 66.846 ( I . 12) 45.723 (0.89) 59.290 (0.99) MNNG 134 (1.00) 106 (0.80) 107 (0.80) NFA 350 (1.00) 448 ( 1.00) 31 (0.09) 2,709 (6.05) 371 (1.06) 370 (0.83) NFZ 343 (1.00) 243 ( 1 .00) 39 (0.11) 1,371 (5.64) 266 (0.78) 283 (1.16) NRZ 7,318 (1.00) 6,455 ( 1 .00) 682 (0.09) 377,709 (58.51) 4.586 (0.63) 7,260 (1.12) I,8DNP 435 ( I -00) 688 ( 1 .OO) 403 (0.93) 1,893 (2.75) 18 (0.04) 12,183 (17.71)

*Numbers are the mutagenic potency expressed as Ara'-induced mutantddose (nmol for 4NQ0, MNNG, NFA, NFZ, and NRZ. or pmol for 1 ,8DNP). These values were estimated from the corresponding dose-response curve of one representative experiment. Numbers in parentheses refer to BA14 for strains BA14NR and BAI4DNP. and to BA140 for strains BA146 and BA149, respectively; bold numbers indicate S.D.

ing encoding gene. This apparently contradictory result has been explained by supposing that such compounds are such good substrates for classical nitroreductase that the additional enzyme activity produced by the multicopy plasmid no longer affects their metabolic activation.

The Ara test revealed important differences between the genetic backgrounds of the tester strains TA98 and TAlOO with respect to the mutagenicity of NFX and most of its analogues, including such compounds as IK, activated by classical nitroreductase, or 1 H, activated by bacterial 0-acetyltransferase. These results are in agreement with previous data showing that the standard TA98 and TAlOO tester strains of the Ames Salmonella assay differ greatly in their responses to mutagens, in ways that are not associated with the mutagenic specificities of the orig- inal his mutations [Popkin et al., 1989; Jurado et al., 1993, 19941. This work demonstrates further that these genetic differences are of significance in mutagenicity testing with nitrofuran compounds. Since the genetic background of strain TA98 appears to be more sensitive than that of TAIOO, the set of isogenic bacteria derived from BA14 is recommended for studies investigating the role of nitroreduction and 0-acetylation as metabolic activation pathways in bacterial mutagenicity.

Based on the mutagenicity of each chemical toward strain BA I4 and derivatives, three different groups were identified. The first group of chemicals included the analogues lG, lE, 11, and ADA, which showed similar mutagenicity in all strains. These mutagens must be considered refractory to both classical nitroreductase and O-acetyltransferase activity. The second group included chemicals with increased mutagenicity, from 2-25-fold depending on the test compound, toward the strain overproducing classical nitroreductase andor reduced mutagenicity of about 4-fold toward the corresponding deficient bacterial strain. These compounds, 1 B, I K, and NFX, are considered to be activated by the classical nitroreductase enzyme. The third group, 1D and lH, was activated by bacterial 0-acetyltransferase, and consequently showed increased and decreased mutagenicity with the particular

overproducer or deficient bacterial strain, as compared to their isogenic parentals.

It has been argued that the bacterial mutagenicity of NFX and its analogues are due to nitroreductase activity that would metabolize the nitrofuran moiety of NFX and derivatives, which is identical for all the compounds [Na- gel, 1987; Alejandre-Duran et al., 19881. Thus, the situa- tion encountered in bacteria would not reflect the metabolic pathways present in eukaryotic systems, such as in Drosophila, where the nitroreductase activity is assumed to be of lesser importance [Alonso-Moraga and Graf, 19891. In contrast to this rather simple idea, the present study shows differences between the metabolic activation of NFX and its analogues.

Dependency on classical nitroreductase could be demonstrated with only 3 (NFX and the analogues IB and 1K) of the nitrofurans tested. The mutagenicities of the remaining 6 chemicals were independent of this major bacterial nitroreductase. In addition, 2 (compounds 1 D and 1H) out of the 3 analogues in which the heterocyclic moiety of the molecule was of the 1,2,4-triazol-4-y1 group showed a dependency on bacterial 0-acetylation. Al- though nitroreduction is considered the main route for the mutagenic activation of nitrofurans, evidence for the involvement of 0-acetylation has been previously re- ported in a series of 2-substituted 5-nitrofurans, because of the lowered mutagenic activity exhibited for the compounds in the corresponding Salmonella TA defective bacterial strain [Ni et al., 19871. In general, the dependency on bacterial nitroreduction or 0-acetylation of NFX and its analogues seemed less relevant than dependencies exhibited by typical nitrofurans such as NFA and NFZ, typical nitroarenes such as I,8DNP, or other nitro-containing compounds such as the nitrothiazole NRZ.

Previous studies have pointed out the possibility of chemicals 1H and IK as promising candidates for the replacement of NFX in clinical use. Thus, they show 1) an activity against Trypanosoma cruzi clearly superior to that of NFX [Mester et al., 19871; 2) low mutagenic poten- cies in bacteria as measured by the Ara forward mutation


test, though higher than those of NFX for all strains except BA146 [Alejandre-Durhn et al., 1988; the present study]; and 3) no genotoxicity in Drosophila [Alonso-Moraga and Graf, 19891. The present study further indicates that compound 1H is metabolically activated by a different pathway than its parental drug, NFX. Additional testing in other eukaryotic genotoxicity test systems would be of particular interest [Santos et al., 19941.

A C K N O W L E D G M E N T S

The authors thank R.M. Claramunt for generously pro- viding the nifurtimox analogues. This research was subsi- dized by the Junta de Andalucia (group no. 3167). J. Jurado received a fellowship from the Consejeria de Edu- caci6n y Ciencias of the Junta de Andalucia.

REFERENCES

Alejandre-Duran E, Claramunt RM, Sanz D, Vilaplana MJ, Molina P, Pueyo C (1988): Study on the mutagenicity of nifurtimox and eight derivatives with the L-arabinose resistance test of Salmo- nella typhimurium. Mutat Res 206: 193-200.

Alonso-Moraga A, Graf U ( 1 989): Genotoxicity testing of antiparasitic nitrofurans in the Drosophila wing somatic mutation and recom- bination test. Mutagenesis 4:105- 110.

Einisto P, Watanabe M, Ishidate M Jr, Nohmi T (1991): Mutagenicity of 30 chemicals in Salmonella typhimurium strains possessing different nitroreductase or 0-acetyltransferase activities. Mutat Res 259:95- 102.

Hera C, Pueyo C (1986): Conditions for the optimal use of the L- arabinose-resistance mutagenesis test with Salmonella typhimurium. Mutagenesis 1 :267-273.

Jurado J, Alejandre-Duran E, Pueyo C (1993): Genetic differences between the standard Ames tester strains TAl00 and TA98. Muta- genesis 8:527-532.

Jurado J, Alejandre-Duran E, Pueyo C (1994): Mutagenicity testing in Salmonella typhimurium strains possessing both the His rever- sion and Ara forward mutation systems and different levels of classical nitroreductase or 0-acetyltransferase activities. Environ Mol Mutagen 23:286-293.

Maron DM, Ames BN (1983): Revised methods for the Salmonella mutagenicity test. Mutat Res 113:173-215.

McCalla DR (1983) Mutagenicity of nitrofuran derivatives: Review.

McCoy EC, Anders M, Rosenkranz HS (1983): The basis of the insensi- tivity of Salmonella typhirnurium strain TA98/1,8DNP6 to the mutagenic action of nitroarenes. Mutat Res 121: 17-23.

Mester B, Elguero J, Claramunt RM, Castanys S, Mascard ML, Osuna A, Vilaplana MJ, Molina P (1987): Activity against Trypanosoma cruzi of new analogues of nifurtimox. Arch Pharm (Weinheim) 320: 1 15- 120.

Nagel R (1987): Genotoxicity studies with two antichagasic drugs. Mu- tat Res 191:17-20.

Ni Y-C, Heflich RH, Kadlubar FF, Fu PP (1987): Mutagenicity of nitrofurans in Salmonella typhimurium TA98, TA98NR and TA98/I,8DNP6. Mutat Res 192:15-22.

Popkin DJ, Davis VM, Prival MJ (1989): Isolation and characterization of an isogenic set of Salmonella typhimurium strains analogous to the Ames tester strains. Mutat Res 224:453-464.

Pueyo C, Ruiz-Rubio M (1984): The L-arabinose resistance test with Salmonella typhirnurium. In de Serres FJ (ed): “Chemical Muta- gens: Principles and Methods for Their Detection, Volume 9.” New York: Plenum Press, pp 89-109.

Rosenkranz HS, Mermelstein R (1983): Mutagenicity and genotoxicity of nitroarenes. All nitro-containing chemicals were not created equal. Mutat Res 114:217-267.

Saito K, Yamazoe Y, Kamataki T, Kato R (1983): Mechanism of activation of proximate mutagens in Ames tester strains: The acetyl- CoA dependent enzyme in Salmonella typhimurium TA98 deficient in TA98/1,8-DNP6 catalyzes DNA-binding as the cause of mutagenicity. Biochem Biophys Res Commun 116: 141-147.

Saito K, Shinohara A, Kamataki T, Kato R (1985): Metabolic activation of mutagenic N-hydroxyarylamines by 0-acetyltransferase in Salmonella typhimurium TA98. Arch Biochem Biophys 239:

Santos SJ, Takahashi CS, Natarajan AT (1994): Cytogenetic effects of the antichagasic benznidazole on human cells in vitro. Mutat Res

Watanabe M, Ishidate M Jr, Nohmi T (1989): A sensitive method for the detection of mutagenic nitroarenes: Construction of nitroreductase overproducing derivatives of Salmonella typhimurium strains TA98 and TA100. Mutat Res 216:211-220.

Watanabe M, Ishidate M Jr, Nohmi T (1990): A sensitive method for the detection of mutagenic nitroarenes and aromatic amines: New derivatives of Salmonella typhimurium tester strains possessing elevated 0-acetyltransferase levels. Mutat Res 234:337-348.

Weisburger JH (1988): Past, present, and future role of carcinogenic and mutagenic N-substituted aryl compounds in human cancer causation. In King CM, Romano LJ, Schuetzle D (eds): “Carci- nogenic and Mutagenic Responses to Aromatic Amines and Ni- troarenes.” New York: Elsevier, pp 3-19.

Accepted by-

286-295.

320:305-3 14.

Environ Mutagen 5:745-765. K. Mortelmans

role of classical nitroreductase and o-acetyltransferase on the mutagenicity of nifurtimox and eight...

Documents