Bacterial β-lyase mediated cleavage and mutagenicity of cysteine conjugates derived from the nephrocarcinogenic alkenes trichloroethylene, tetrachloroethylene and hexachlorobutadiene

Chem.-Biol. Interactions, 60 (1986) 31--45 Elsevier Scientific Publishers Ireland Ltd.

31

BACTERIAL ~-LYASE MEDIATED CLEAVAGE AND MUTAGENICITY OF CYSTEINE CONJUGATES DERIVED FROM THE NEPHROCARCINOGENIC ALKENES TRICHLOROETHYLENE, TETRACHLOROETHYLENE AND HEXACHLOROBUTADIENE*

WOLFGANG DEKANT**, SPYRIDON VAMVAKAS, KLEMENS BERTHOLD, SABINE SCHMIDT, DIETER WILD and DIETRICH HENSCHLERt Institute of Toxicology, University of Wiirzburg, Versbacherstr. 9, D-8700 Wiirzburg (F.R.G.) (Received May 18th, 1986) (Revision received August 15th, 1986) (Accepted August 20th, 1986)

SUMMARY

The metabolism of/~-lyase and the mutagenicity of the synthetic cysteine conjugates S-1,2-dichlorovinylcysteine (DCVC), S-1,2,2-trichlorovinylcysteine (TCVC), S-1,2,3,4,4-pentachlorobuta-l,3-dienylcysteine (PCBC) and S-3-chloropropenylcysteine (CPC) were investigated in Salmonella typhimurium strains TA100, TA2638 and TA98. The bacteria contained significantly higher concentrations of ~-lyase than mammalian subceUular fractions. Bacterial 100 000 X g supernatants cleaved benzthiazolylcysteine to equimolar amounts of mercaptobenzthiazole and pyruvate. DCVC, TCVC and PCBC produced a linear time-dependent increase in pyruvate formation when incubated with bacterial 100 000 X g supernatants; pyruvate formation was inhibited by the ~-lyase inhibitor arninooxyacetic acid (AOAA). CPC was not cleaved by bacterial enzymes to pyruvate. DCVC, TCVC and PCBC were mutagenic in three strains of S. typhimurium (TA100, TA2638 and TA98) in the Ames-test without addition of mammalian subcellular fractions; their mutagenicity was decreased by the addition of AOAA to *This work was supported by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 172), Bonn and the Doktor-Robert-Pfleger~qtiftung, Bamberg. **Present address: Department of Pharmacology, University of Rochester, 601 Elmwood Avenue, Rochester, NY 14642, U.S.A. t T o whom correspondence should be sent. Abbreviations: AOAA, aminooxyacetic acid; N-boc-cysteine, N-tert.-butoxycarbonyl-Lcysteine; BTC, S-(4,5-benz-3-thiazolyl)-cysteine; CPC, S-3-chloropropenylcysteine; DCVC, S-1,2-dichlorovinylcysteine; DMF, dimethylformanide; DMSO, dimethylsulfoxide; GSH, glutathione; HCBD, hexachlorobutadiene; MBT, 2-mercaptobenzothiazole; PCBC, S-1,2,3,4,4-pentachlorobuta-l,3-dienylcysteine; TBME, tert-butyl methyl ether; TCVC, S-1,2,2-trichlorovinylcyst eine; Tetra, tetrachloroethylene; Tri, trichloroethylene. 0009-2797/86/$03.50 © 1986 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

32 the preincubation mixture. CPC was not mutagenic in any of the strains of bacteria tested. These results indicate that ~-lyase plays a key role in the metabolism and mutagenicity of haloalkenylcysteines when tested in S. typhimurium systems. The demonstrated formation in mammals of the mutagens DCVC, TCVC and PCBC during biotransformation of trichloroethylene (Tri), tetrachloroethylene (Tetra) and hexachlorobutadiene {HCBD) may provide a molecular explanation for the nephrocarcinogenicity of these compounds.

Key words: ~-Lyase -- Halogenated alkenes -- Mutagenicity -- Nephrocarcinogenicity

INTRODUCTION

Halogenated alkenes are widely used industrial chemicals (e.g. Tri, Tetra) and persistent environmental contaminants (e.g. Tetra, HCBD). Tri and Tetra have been found to increase the rate of hepatocellular carcinoma in mice and to induce a low incidence of renal adenocarcinoma in rats [1,2]. Hexachlorobutadiene (HCBD) is a strong and specific nephrocarcinogen which causes identical types of tumors in high incidence [3]. Neither Tri nor Tetra are mutagenic when liver enzymes are used as the activating system, only a weak mutagenic activity was observed for HCBD, binding to DNA has not been demonstrated (for review see Refs. 4, 5 a n d 6). However, metabolic conversion by c y t o c h r o m e P-450 dependent monooxygenases to short-lived intermediates (oxiranes) has been postulated for Tri and Tetra both in vivo [7.8] and in vitro [9,10], whereas previous reports on the significance of oxidative metabolism of HCBD [6] could not be substantiated in other experiments [11]. The main pathway of HCBD-biotransformation follows a different route. HCBD is conjugated with glutathione (GSH) by the catalytic action of GSH-transferases to form both a monoand a diglutathione conjugate without prior oxidation [ 12]. The S-pentachlorobutadienyl-GSH formed in the liver is processed in the kidney by the enzymes of mercapturic acid pathway to S-pentachlorobutadienylcysteine, which is regarded as the penultimate metabolite in this metabolic pathway whose last step, ~-lyase mediated cleavage to reactive intermediate results in nephrotoxicity [13]. An identical activation mechanism is also found as a minor pathway in Tri and Tetra biotransformation [14,15]. The enzyme, cysteine conjugate~-lyase, which cleaves several aliphatic thioethers of cysteine [16,17] to pyruvate, ammonia and a reactive thiol, has been held responsible for the acute nephrotoxicity of DCVC and S-l-chloro-l,2,2trifluoroethylene [ 16,17 ]. ~-Lyase activity is present in rat liver and kidney, in bovine liver [18] and in m a n y species of bacteria [19] ; it is diminished by inhibitors of pyridoxal phosphate dependent enzymes such as AOAA [ 16] ; the free amino group of the cysteine conjugates is a prerequisite for e n z y m e activity; mercapturic acids are not accepted as substrates [18].

33 Recently, a role of ~-lyase in the formation of mutagens during biotransformation of Tri, Tetra and HCBD has been postulated on the grounds that the cysteine conjugates, DCVC (formed from Tri), TCVC (from Tetra) and PCBC (from HCBD) whose formation in the kidney has been demonstrated, are definitely mutagenic, particularly after addition of rat kidney S-9 in the Ames plate incubation assay [20]. This does, however, n o t provide p r o o f of the key role of ~-lyase in the bioactivation process of the three compounds. In similar investigations with TCVC in a modified Ames-test system (preincubation assay) we found a high direct mutagenic activity. This finding implies that either the test organisms display 5-1yase activity of their own, or other bioactivation mechanisms are operant. This p r o m p t e d us to investigate the role of ~-lyase in the metabolism and mutagenicity of TCVC, DCVC, PCBC and CPC by applying suitable inhibitors in a more sensitive assay and characterizing quantitatively the reaction products of the ultimate activation step. CPC was chosen for comparison, because it does n o t contain a vinylic moiety with high electronegativity b o u n d to the sulfur atom of cysteine. In our study these cysteine conjugates, whose metabolism by renal enzymes has already been demonstrated [20], were chosen to provide a basis for the use of bacterial ~-lyase as a model system. Our objective was to replace to some extent experiments with mammalian enzymes in the investigation of the metabolism of cysteine conjugates and the DNA binding of the metabolites formed. MATERIALS AND METHODS

Syntheses DCVC was synthesized b y the m e t h o d o f McKinney et al. [21], N-tert.butoxycarbonyl-L-cysteine (N-boc-cysteine) according to Moroder et al. [22]. S-(4,5-benz-3-thiazolyl)-cysteine (BTC), TCVC, CPC and PCBC were synthesized as follows: 210 mg (30 mmol) of lithium hydride were added under nitrogen to a solution o f 3.35 g (15 nmol) o f N-boc-cysteine in 20 ml o f dry dimethylsulfoxide (DMSO) or dimethylformamide (DMF). The mixture was stirred at room temperature until the evolution of hydrogen stopped {10--15 min). Then a solution of 2-chlorobenzthiazol or the halogenated alkene in 5 ml of DMSO or D M F was added dropwise (10 min). Stirring was continued for 3--15 h after which 40 ml o f water were added. The pH was adjusted to 2--3 for solvent extraction b y addition o f 1 M potassium hydrogen sulfate solution. The p r o d u c t was then extracted into 4 X 30 ml of tertbutyl methyl ether (TBME). The extracts were dried over sodium sulfate and the solvent evaporated under reduced pressure. The oily residues were purified b y preparative TLC (eluent-2-propanol/chloroform/hexane, 2 : 5 : 5) and reextracted with TBME (100 ml) in a Soxhlet extractor. After removal o f the solvent, the oil was treated with 20 ml 3 M HC1 in ethyl acetate for 2 h at room temperature. The crystalline precipitate was filtered and washed with

34 TBME. Identity of the reference c o m p o u n d s was checked by GC/MS~ purity b y HPLC (see Results). HPLC and GC/MS were performed under conditions described elsewhere [ 15 ].

Preparation of bacterial 100 000 X g homogenates The bacteria were harvested at the end of the exponential growth phase after aerobic incubation o f the cultures in Oxoid Nutrient Broth No. 2 for 10 h. Cell titers were determined by optical density measurements at 600 nm and were approx, equivalent to 3--4 X 109/ml. All preparative steps were carried o u t at 4°C. The bacterial cells were sedimented from t h e culture by centrifugation at 6000 X g for 10 min in a B20A centrifuge (DAMON/IEC, Needham, U.S.A.). The bacterial pellet was washed once by resuspending in 10 mM phosphate buffer (pH 7.4) and subsequent centrifugation at 6 0 0 0 X g for 10 min. The supernatant was discarded and the pellet was finally resuspended in 10 mM phosphate buffer (pH 7.4). The bacterial cells were lysed b y sonication. Sonication was carried o u t for a total of 4 rain (4 X 1 min with 2 min pauses to avoid protein denaturation). This procedure was followed by centrifugation at 100 000 X g for 20 min in a B6 ultracentrifuge (DAMON/IEC, Needham HTS., U.S.A.) to precipitate membranes and intact bacterial cells. Protein levels were determined with the Bio-Rad Protein Assay Kit using bovine gamma globulin as the standard (Bio-Rad Laboratories, Miinchen, F.R.G.).

Determination of 5-1yase activity Cysteine conjugate ~-lyase activity was determined in bacterial 100 000 X g supernatants and in liver and kidney microsomes and cytosol from male Wistar rats with BTC as substrate b y the formation of 2-mercaptobenzothiazole (MBT) and pyruvate as described in the literature [23]. Liver and kidney I 0 0 000 X g supernatants from male Wistar ratswere prepared according to Wolf et al. [ 11].

Cleavage o f synthetic cysteine conjugates by the 100 000 X g supernatant from S. typhimurium TA I O0 Incubations were carried out, after a series of model experiments, in 0.1 M phosphate buffer (at the optimum pH 7.4) containing substrate (2 mM), bacterial protein (5 mg/ml) and mercaptoacetic acid (5 mM) as a scavenger for reactive intermediates. To some incubations pyridoxal phosphate (0.5 raM) or aminooxyacetic acid (1 mM) were added. The final incubation volume was 5 ml. The reactions were carried o u t at 37°C in a shaker-water bath and samples (1 ml) were removed at 0, 10 and 20 min. The reaction was stopped b y the addition of ice-cold trichloroacetic acid (0.3 ml). The samples were then centrifuged for 4 min at 12 000 rev./min in a microrapid/k centrifuge (HETTICH, Tuttlingen, F.R.G.) to precipitate protein. Pyruvate was quantified colorimetrically in the protein-free solution using a commercial pyruvic acid determination kit (Sigma Chemical Company, St. Louis, U.S.A.).

35

Mu tagenicity assay The mutagenicity of the cysteine conjugates was tested in the S. typhio murium strains TA100, TA98 and TA2638 using the preincubation test as described by Maron and Ames [25]. Preincubations were performed at 37°C for 120 rain. The characteristic properties of these strains were checked regularly by testing the UV and crystal violet sensitivity, ampicillin resistance and mutability by UV light. RESULTS

Synthesis and characterisation of cysteine conjugates Since the data available in the literature do not provide unequivocal information as to the identity and the purity of the cysteine conjugates used, information which is of vital importance for the interpretation of the results obtained in different systems, we report here some details of our synthesis. The commonly used reaction of disodium cysteine in liquid ammonia did not yield reasonable amounts of pure TCVC and PCBC. Therefore, we developed a new method for synthesizing these adducts. We introduced the tert-butoxycarbonyl-protecting group (boc) to increase the solubility of the cysteine-moiety in organic solvents because it is stable to alkali and readily cleaved under mild conditions. It also inhibits a nucleophilic reaction of the amino function. We used LiH as a base because it prevents oxidation of the sulfide of N-boc-cysteine and it is easily removed from the reaction mixture. This procedure resulted in much higher yields (40-80%) of TCVC and PCBC as compared to the method using liquid ammonia (1--2%). The analytical characteristics and purity of the synthetic conjugates used for biochemical experiments are shown in Table I. The analytical data of BTC were identical to those described by Dohn and Anders [23]. #-Lyase activity in bacteria and rat tissue subcellular fractions The results summarized in Table II illustrate the ~-lyase activity in the 100 000 × g supernatants of homogenates of three S. typhirnuriurn strains and in liver and kidney microsomes and cytosol from male Wistar rats'as evaluated with BTC as the enzyme substrate. Mercaptobenzthiazole and pyruvate were formed in equimolar amounts. The S. typhirnurium strains contained significantly higher concentrations of/~-lyase than the rat organ homogenates; no ~-lyase activity was detectable in the microsomal pellet from rat liver or kidney. Cleavage of cysteine conjugates by bacterial ~-lyase Formation of pyruvate in a 100 000 × g supernatant from S. typhimuriurn TA100 homogenates incubated with DCVC, TCVC, PCBC and CPC was used to determine the extent of cleavage of the synthetic eysteine conjugates by bacterial/l-lyase. As shown in Fig. 1, incubation of DCVC, TCVC and PCBC with bacterial homogenate supernatants resulted in a linear increase in pyru-

H

Cl

Cl

R

TM

H

H

Cl

cl

R

CH2_R

cl

cl C c l ~ ~

I

Cl

]

T

Cl-~ ~

183--185 (dee.)

165--167 (dec.)

210 (dec.)

171

Not recorded

N-trifluoroacetyl- O-methyl ester: 59 (40), 69 (44), 79 (19), 117 (67), 138 (46), 150 (14), 170 (17), 185 (13), 198 (100), 220 (63 4C1), 255 (26 5 Cl), 269 (12 4C1), 305 (30 4Cl), 340 (20 5Cl), 453 (11 5C1M÷)

N,O-TMSb: 73 (82), 100 (25), 115 (25), 147 (15), 218 (100), 278 (57 3 Cl), 396 (0.2 3C1M÷)

N,O-TMSb: 59 (12), 73 (100), 100 (25), 115 {20), 147 (15), 218 (85), 242 (60 2Cl), 362 (0,01 2C1M ÷)

Mass spectrum c m/e (Intensity%)

(R = CYSTEINYL)

aIn D~O/DCI, standard: 3-trimethylsilyl-2,2,3,3-tetradeutero propionic acid sodium slat. bTMS, trimethylsilyl- derivative. COnly 3sCl-peaks are noted.

CPC

PCDC

TCVC

DCVC

R

M.P. (°C)

TABLE I A N A L Y T I C A L D A T A O F T H E SYNTHETIC CYSTEINE C O N J U G A T E S

2,8-3,4 ('H), 4,5 (4H) 4.5-4,8 ('H), 5,6-6,3 (2H)

3,4 (~H), 4,4 ('H) 5,4 ('H)

3,4 (2H), 4,6 (IH) 5,0 ('H)

3,5 (2H), 4,5('H) 4,8 ('H), 6,6 CH)

'H - N M R a 8-ppm (Intensity)

98%

99%

99%

98%

Purity exceeding

O'#

37 pyruvafe Formed (n~r)1 x mC1 }

DCVC

TCVC

/

mo

/

80

PCBC

a

/

/

40

70.

f

/ ~

,

,

lo

is

, >

20

,

t(mi.)

0

,

5

~'1'0

15

,

>

20 t(min:

,

0

5

,

,

10

15

,

>

20 t (rain)

Fig. 1. Cleavage of synthetic cysteine conjugates by bacterial homogenate 100 000 x g supernatant (a). Cysteine conjugates were incubated with TA100 100 000 x g supernatant (5 mg protein/ml). For some experiments A O A A (1 raM) was added to the incubation mixture (x), or the bacterial homogenate was incubated without the addition o f substrate (o). Mean of 4 determinations t S.D.

TABLE II #-LYASE ACTIVITY IN BACTERIA AND SUBCELLULAR FRACTIONS FROM MALE RATS ASSAYED WITH BENZTHIAZOLYLCYSTEINE AS SUBSTRATE Mean of 4 determinations ± S.D. MBT, mercaptobenzthiazole. Homogenates

MBT formed nmol × rain "1

Pyruvate formed nmol × m i n - '

#-Lyase nmol × m g prot. -~ × rain -~

18.2±3.3 17.6±~.0 32.2±3.5

18.6±3.3 16.8±2.8 32.2±3.5

18.4±3.3 17.2±2.9 32.2±3.5

6.1±1.1 3.9±1.2

6.2±1.1 4.5±1.1

6.1±1.1 4.2±1.2

S.~phimurium TA100 TA98 TA2638 100000xg~pernatant liver kidney 100 000 X g pellet liver kidney

0.01 0.01

Not detectable Not detectable

38 vate formation as compared to controls. Pyruvate formation was dependent on time, presence of cysteine conjugates and bacterial 100 000 X g supernatant (not shown). TCVC gave the highest rate of pyruvate production (14.5-fold increase over control after 20 min) and PCBC the lowest (6.8-fold increase), DCVC taking an intermediate position (8.3-fold increase over background after 20 min). There was no difference in pyruvate formation in the presence or absence of added pyridoxal phosphate {data not shown). Incubation of CPC with bacterial homogenate 100 000 × g supernatant did not increase the release of pyruvate over control values (data not shown). Addition of AOAA (1 mM) to the incubation mixtures decreased the rate of pyruvate formation by 80%.

Mutagenicity assay In preliminary experiments, the effect of mammalian microsomal and cytosolic enzymes on the mutagenicity of TCVC was tested in the S. typhimurium strain TA100. Neither addition of cytosolic enzymes (0.5 or 2 mg protein from rat liver or kidney cytosol to the preincubation mixture) nor microsomal enzymes (1 or 2 mg protein from rat liver or kidney microsomes to the preincubation mixture) increased the observed mutagenic activity (revertants/nmol) of this compound. Therefore, the series of experiments was performed without using mammalian enzymes for metabolic activation. When using the S. typhimurium TA100, we could demonstrate (Fig. 2) a high mutagenic potency of TCVC and less so of PCBC. Surprisingly, DCVC exerted a definite but unexpectedly low degree of mutagenicity only in a narrow concentration range one order of magnitude below that operant with the two other conjugates. This is obviously due to compound specific toxicity because we observed high toxicity expressed as increased formation SaLrnonetta typhimurium ~l his* reverfants per plate

/

500

5BO ,,00

-

300 200

TCV C

TAIO0

PC BC

DCVC

--LS-~.

/

/s

.0 / 0

~.O ~

--0 o

100

,

)

I0 nmol, perptate

,

5

>

nmo{

per plate

,

O~

013

)

05 nmql. ' per pLa;'e

Fig. 2. Mutagenicity of TCVC (~), PCBC (x) and DCVC (~) in S. typhimurium TA100. The inhibitor A O A A ( o ) w a s added to the preincubation mixtures at a concentration of 1 raM. Experiments were repeated several times with consistent results.

39 SaLmoneLLa

typhirnurium

TA2638

N hie

reverfants per late

PCBC

TCVC

DCVC

900 000 700 J fi00500 -

400•

o

/

/;

3OO

o

2O0 x

100 'c

.

,

10

,

15

o

>

nmot

1

per plate

o

/

5

10

15

> nrno perplatet

.

5

.

.

10

.

)

15

20 nmp[

per pta~e

Fig. 3. Mutagenicity of TCVC (a), PCBC (x) and DCVC ( o ) i n S. typhimurium TA2638. The inhibitor AOAA (o) was added to the incubation mixtures at a concentration of I raM. Experiments were repeated several times with consistent results. N His*, revertants/plate.

of microcolonies. We introduced T A 2 6 3 8 to attempt to overcome the problems related to primary, non-specific toxicity to the tester organisms. The results obtained in this series are presented in Fig. 3. In T A 2 6 3 8 , DCVC was the strongest mutagen of the compounds tested and resulted in a 17-fold increase in the frequency of revertants as compared to controls (solvents R = H,CL,C2CL3 R

CL

CL/~

S-CH 2 - CH - t O O -

I ~- Lyose

//~

R

S

+ CH3 - CO - C O O -

R

H

/ . ~

g

DNA- binding mutagenicity

Fig. 4. Metabolism o f haloalkenylcysteines to mutagenic intermediates by ~-lyase.

40 only). PCBC also exhibited some mutagenic properties when tested in TA2638, whereas no mutagenic response could be observed with TCVC. The cysteine conjugates, TCVC, DCVC and PCBC, also exerted a definite but weak mutagenicity in the S. typhimurium strain TA98 resulting in a 2--3fold increase in the frequency of revertants as compared to controls {data not shown). The addition of the ~-lyase inhibitor AOAA to the preincubation mixture significantly decreased the mutagenicity of DCVC, TCVC and less so of PCBC. AOAA did not reduce the revertants induced by UV-irradiation of the bacteria (not shown). Complete inhibition of the mutagenicity could not be achieved with AOAA since higher concentrations of the enzyme inhibitor had toxic effects on the bacteria. The cysteine conjugate of 1,3dichloropropene, CPC, which is not cleaved by bacterial ~-lyase to pyruvate, did not induce mutations in TA100, TA2638 and TA98 at concentrations of 0.1--100 nmol/plate. DISCUSSION The previously described synthetic method [21] for the preparation of cysteine conjugates using liquid ammonia as solvent and cysteine resulted, in our hands, in less than 2% yields of TCVC and PCBC. To avoid excessive formation of by-products and time-consuming purification procedures, we developed an alternative synthetic method to obtain TCVC and PCBC in reasonable yields. The use of organic solvents instead of liquid ammonia and the introduction of an easily removable protective group resulted in a more efficient and convenient procedure with considerable improvement of the yield of cysteine conjugates (40--80%) from Tetra and HCBD. Previous investigations of the mutagenicity of cysteine conjugates derived from halogenated alkenes indicated that bacterial enzymes might participate in the activation of this compound [20]. ~-Lyase is a common enzyme in bacteria; however, its presence in S. typhimurium and the ability of this enzyme to metabolise thioethers of cysteine had not yet been demonstrated. Our experiments indicate that S. typhimurium strains contain considerable amounts of ~-lyase activity and that the bacterial 5-1yase metabolises the model substrate [23] BTC as well as TCVC, DCVC and PCBC to pyruvate. Two facts support the conclusion that pyruvate is produced from the conjugates by a 5-1yase catalysed reaction: (a) from BTC, the expected thiol (mercaptobenzthiazole) and pyruvate were formed in equimolar amounts and (b) pyruvate and mercaptobenzthiazole formation were inhibited by addition of AOAA which is an inhibitor of pyridoxal phosphate dependent enzymes [16] and inhibits ~-lyase both in vivo [16] and in vitro [17]. /~-Lyase is a widely distributed enzyme found in several bacteria, in many cases with higher specific activity [ 19] than in rat liver and kidney homogenates. Recent investigations showed that the bacterial, the mammalian liver and kidney cytosolic enzymes and the kidney mitochondrial enzyme exhibit major differences in their properties: the cytosolic enzymes and the bacterial enzyme are soluble enzymes [18,19] whereas the mitochondrial

41 enzyme from rat kidney cortex is located in the outer mitochondrial membrane and is much more sensitive to inhibition b y AOAA [26]. Our results indicate that the relative activities (in terms of pyruvate formation) of bacterial ~-lyase parallel those of renal/~-lyases for DCVC, TCVC and PCBC [20]. TCVC is cleaved b y bacterial ~-lyase and by rat kidney ~-lyase to a much larger extent than DCVC and PCBC [20]. Cysteine conjugates with aromatic or vinylic (BTC, S-benzylcysteine, DCVC) [18] and/or electronegative (TCVC, PCBC, CTFE) substituents are good substrates for ~-lyase, whereas S-ethyl- and chloroethylcysteine are n o t cleaved [ 18], indicating that electronegative and/or unsaturated moieties bound to the sulfur atom are a pre~ requisite for enzyme activity. These structure-activity relationships may explain that CPC which does not contain an unsaturated highly electronegatire moiety bound to the sulfur of cysteine, is not cleaved b y 5-1yase. The conjugates cleaved b y ~-lyase (TCVC, DCVC and PCBC) were mutagenic in the Ames-test in three strains of bacteria. T A 1 0 0 gave the highest mutagenic response to TCVC and PCBC. The high toxicity observed even at low concentrations of DCVC prevented the demonstration of a more pronounced mutagenic response for this c o m p o u n d in TA100. On the other hand, in the strain TA2638, DCVC was found to be a strong mutagen at higher, nontoxic concentrations, whereas the mutagenicity of TCVC in this strain was very low. No explanation for the differences in response of the two strains to the mutagenicity and toxicity of these chemically similar c o m p o u n d s can be provided on the basis of the currently available data. The/~-lyase activities in both TA100 and T A 2 6 3 8 are considerable and DCVC and TCVC are intensively metabolised; therefore, the levels of reactive intermediates should be sufficient to exert a mutagenic effect. In view of the low effect in the frameshift strain TA98 and the potent effect in the base-pair substitution strains TA100 and/or TA2638 we conclude that the cysteine conjugates DCVC TCVC and PCBC induce base-pair substitution mutations. Our results confirm in principle those described in the literature for the mutagenicity of DCVC, TCVC and PCBC in TA100. Green and Odum [20] used the standard plate incorporation assay with T A 1 0 0 and they observed a higher mutagenicity in experiments performed with added rat kidney S-9 as compared to those using bacterial activation only. In our experiments, with a preincubation period of 2 h, we observed a higher mutagenicity without addition of rat kidney cytosol and a 10-fold increase in sensitivity in the strain TA100 as compared to the plate incubation assay [20]. The fact that bacteria exhibit much higher levels of/3-1yase activity than rat cytosolic fractions and the longer exposure of the bacteria in the preincubation assay m a y explain the differences in the results obtained. The ~-lyase inhibitor AOAA decreased the mutagenicity of DCVC, TCVC and PCBC. AOAA did n o t reduce the revertants induced by UV irradiation of bacteria, suggesting that the diminished mutagenicity of the cysteine conjugates in presence of AOAA is the result of enzyme inhibition. In our studies on the metabolism of the cysteine conjugates in bacterial homogenate supernatants AOAA did n o t completely inhibit pyruvate formation; the remaining

42 metabolism particularly at higher concentrations of the cysteine conjugates might produce levels of metabolites sufficient to cause mutagenicity. CPC, which is not a substrate for ~-lyase, was not mutagenic even at concentrations much higher than those used for the other conjugates. The lack of mutagenicity of this compound and the effect of AOAA indicate that ~-lyase mediated cleavage is the key event in the formation of mutagenic intermediates from haloalkenylcysteines. The structure of the ultimate mutagenic metabolite still needs to be elucidated. As shown in Fig. 4, the cleavage of cysteine conjugates by ~-lyase results in the formation of pyruvate, ammonia and a thiol [28], which is regarded as the ultimate metabolite which exerts acute nephrotoxicity by covalent interaction with cellular macromolecules [25,26]. Previous experiments on the DNA-binding of 3SS-labelled DCVC indicated that the radiolabelled sulfur atom is retained in the DCVC-fragment bound to DNA [29]. Considering the possible structure of the ultimate DNA-binding metabolite, it seems unlikely that the ethenethiol formed from the vinylic cysteine conjugates possesses the electrophilic potency to react with nucleophilic sites in DNA. We suggest two alternative structures for the ultimate binding metabolite. The initially formed thiol may form strong electrophiles by two different reactions {Fig. 4): (a) rearrangement of the enolic thiol to a thiocarbonic acid chloride or (b) elimination of a molecule of hydrogen chloride to form a thioketene. Both molecules are highly electrophilic structures [30,31], which should be expected to be able to react with DNA to produce the initial molecular lesion in the process leading to mutagenicity and carcinogenicity. In the in vivo situation, conjugation with glutathione resulting in the formation of haloalkenylcysteines has been shown to occur during the biotransformation of Tri, Tetra and HCBD [11,14,15]. GSH-conjugation most probably takes places in the liver; the GSH-conjugates formed are then transported to the kidney to be processed by glutamyltranspeptidase and cysteinylglycinase to yield the corresponding cysteine conjugates [32,33]. Two pathways for further reaction are then possible in vivo" ~-lyase .mediated cleavage to mutagenic intermediates, or acetylation to mercapturic acids which are finally excreted with the urine. Formation of haloalkenylcysteines and their ~-lyase mediated cleavage takes place in the proximal tubules of the kidney where the enzymes of mercapturic acid formation and ~-lyase are present in high concentrations [32, 33]. The proximal tubules are also the target cells for carcinogenicity of Tri, Tetra and HCBD [1--3]. Tri and Tetra induce only a low incidence of renal carcinoma (a rare spontaneous tumor) in rats, whereas HCBD is regarded as a strong and specific nephrocarcinogen. These differences in tumorigenic response may be explained by the extent of activation through GSH-conjugation during biotransformation. GSH-conjugation is only a minor pathway of biotransformation for Tri and Tetra [14,15] ; the main pathway for these compounds is P-450 mediated oxidation in the liver. In contrast, GSHconjugation is the only pathway of biotransformation of HCBD described so

43 far, producing much higher levels of PCBC in the kidney [ 34,35] as compared to the levels of DCVC or TCVC formed during biotransformation of Tri or Tetra, respectively. In conclusion, the formation of mutagenic cysteine conjugates during metabolism of Tri, Tetra and HCBD may provide a molecular explanation for the organ specific carcinogenicity of these compounds. A confirmation of this hypothesis needs further experiments on the binding of metabolites to renal DNA and the structure of the adducts formed. Our experiments show that the important step in the described activation mechanism, the formation of reactive intermediates from cysteine conjugates, can be simulated in bacteria. The bacterial/3-1yase, which is not identical in its properties to the renal enzyme, exhibits a similar activity towards the cysteine conjugates tested. The extent o f cleavage by the bacterial enzymes observed with DCVC, TCVC and PCBC parallels the activity of the enzymes present in rat kidney; bacterial enzymes also transform the cysteine conjugates to reactive and DNA damaging intermediates. Compared to rat kidney or liver homogenates as sources for/~-lyase, bacteria have two important advantages: the enzyme is present there with high specific activity, and can be obtained easily and purified in large quantities. Our results on the metabolism of the conjugates investigated here indicate that the use of the bacterial ~-lyase may be helpful in elucidating structure/activity relationships and in the identification of reactive intermediates and potential DNA-adducts. The use of this model system m a y supplement and simplify investigations with mammalian enzymes and with intact animals. ACKNOWLEDGEMENTS The skillful technical assistance of Ms. J. Ahamer and Ms. S. Seufert is gratefully acknowledged. REFERENCES

1 National Toxicology Program (NTP) Technical Report 232, U.S. Department of Health and Human Services, Public Health Service, 1986. 2 National Cancer Institute, Bioassay of Tetrachloroethylene for Possible Carcinogenicity, DHEW Publication No. 77-813, 1977. 3 R.J. Kociba, G.D. Keyes, G.C. Jersey, J.J. Ballard, D.A. Dittenber, J.F. Quast, C.E. Wade, C.G. Humiston and B.A. Schwetz, Results of a two year chronic toxicity study

with hexachlorobutadiene in rats, Am. Ind. Hyg. Assoc. J., 38 (1977) 589--602. 4 Environmental Health Criteria, 50, Trichloroethylene, WHO, Geneva, 1985. 5 D. Reichert, Biological actions and interactions of tetrachloroethylene, Mutat. Res., 123 (1983) 411--429. 6 D. Reichert, T. Neudecker and S. Schlitz, Mutagenicity of hexachlorobutadiene, perchlorobutenoic acid and perchlorobutenoic acid chloride, Mutat. Res., 137 (1984) 89--

93. 7 J.F. Powell, Trichloroethylene -- absorption, elimination and metabolism, Br. J. Indust. Med., 2 (1945) 142--145. 8 S. Yllner, Urinary metabolites of '4C-tetrachloroethylene in mice, Nature, 191 (1961)

820.

44 9 H. Allemand, B. Pessayre, V. Descatoire, C. Degott, G. Feldmann and J.P. Benhamou, Metabolic activation of trichloroethylene into a chemically reactive metabolite toxic to the liver,J. Pharmacol. Exp. Thera., 204 (1978) 714--723. 10 A.K. Costa and M. Ivanetich, Tetrachloroethylene metabolism by the hepatic microsomal cytochrome P-450 system, Biochem. Pharmacol., 29 (1980) 2863--2869. 11 C.R. Wolf, P.N. Berry, J.A. Nash, T. Green and E.A. Lock, Role of cytosolic glutathione S-transferases in the conjugation of hexachloro-1:3-butadiene and its possible relevance to toxicity, J. Pharmacol. Exp. Ther., 228 (1984) 202--208. 12 W.J. Jones, R.G. Gerdes, K. Orrnstad and S. Orrenius, The formation of both a monoand a bis-substituted glutathione conjugate of hexachlorobutadiene by isolated hepatocytes and following in vivo administration in the rat, Chem.-Biol. Interact., 56 (1985) 251--267. 13 J.A. Nash, L.J. King, E.A. Lock and T. Green, The metabolism and disposition of hexachloro-1:3-butadiene in the rat and its relevance to nephrotoxicity, Toxicol. Appl. Pharmacol., 73 (1984) 124--130. 14 W. Dekant, M. Metzler and D. Henschler, Identification of S-1,2-dichlorovinyl-Nacetylcysteine as a urinary metabolite of trichloroethylene: A possible explanation for its nephrocarcinogenicity in rats, Biochem. Pharmacol. (1986) in press. 15 W. Dekant, M. Metzler and D. Henschler, Identification of S-l,l,2-trichlorovinyl-Nacetylcysteine as a urinary metabolite of tetrachloroethylene -- Bioactivation through glutathione conjugation m a y explain its nephrocarcinogenicity, J. Biochem. Toxicol. (1986) in press. 16 A.A. Elfarra, I. Jakobson and M.W. Anders, Mechanism of S-(1,2-dichlorovinyl)glutathione-induced nephrotoxicity, Biochem. Pharmacol., 35 (1986) 283--288. 17 D.R. Dohn, J.R. Leininger, L.L. Lash, A.J. Quebbeman and M.W. Anders, Nephrotoxicity of S-(2-chloro-1,1,2 -trifluoroethyl)glutathione and S-(2 -chloro-1,2,2 -trifluoroethyl)-L-cysteine, the glutathione and cysteine conjugates o f chlorotrifluoroethene, J. Pharmacol. Exp. Ther., 235 (1985) 851--857. 18 M. Tateishi, Methylthiolated metabolites, Drug Metab. Rev., 14 (1983) 1207--1234. 19 G.L. Larsen, Distribution o f cysteine conjugate ~-]yase in gastrointestinal bacterial and the environment, Xenobiotica, 15 (1985) 199--209. 20 T. Green and J. Odum, Structure/activity studies of the nephrotoxic and mutagenic action o f cysteine conjugates of chloro- and fluoroalkenes, Chem.-Biol. Interact., 54 (1985) 15--31. 21 L.L. McKinney, J.C. Pickem, F.B. Weakley, A.C. Eldridge, R.E. Campell, J.C. Cowan and H.E. Biester, Possible toxic factor of trichloroethylene-extracted soybean oil meal, J. Am. Chem. Soc., 81 (1959) 909--915. 22 I. Moroder, A. Hallett, E. Wiinseh, O. Veller and G. Wersin, Di.-tert. Butyldicarbamat - - ein vorteilhaftes Reagenz zur Einfiihrung der t-Butyloxycarbonyl-Schutzgruppe, Hoppe-Seyler's Z. Physiol. Chem., 357 (1986) 1651--1655. 23 D.R. Dohn and M.W. Anders, Assay o f cysteine conjugate ~-lyase activity with S-(2benzthiazolyl)-cysteine as the substrate, Anal. Biochem,, 120 (1982) 379--386. 24 D. Maron and B. Ames, Revised methods for the Salmanella mutagenicity test, Mutat. Res., 113 (1983) 173--215. 25 L.H. Lash and M.W. Anders, Cytotoxicity of S-(1,2-dichlorovinyl)glutathione and S(1,2-dicblorovinyl)-L-cysteine in isolated rat kidney cells, J. Biol. Chem. (1986) in press. 26 L.H. Lash, A.A. Elfarra and M.W. Anders, Renal cysteine conjugate ~-lyase: Bioactivation of nephrotoxic cysteine S-conjugates in mitochondrial outer membrane, J. Biol. Chem. (1986) in press. 27 G.L. Larsen and J.L. Stevens, Cysteine conjugate ~-lyase in the gastrointestinal bacterium Eubacterium limosum, Mol. Pharmacol., 29 (1986) 97--103. 28 C.D. Hassall, A.J. Gandolfi and K. Brendel, Effect of halogenated vinyl cysteine conjugates on renal tubular active transport, Toxicology, 26 (1983) 285--294.

45 29 R.K. Bhattacharya and M.O. Schulze, Properties of DNA treated with S-(1,2-dichlorovinyl)-L-Cysteine and a lyase, Arch. Biochem. Biophys., 153 (1972) 105--115. 30 S. Scheithauer and R. Meyer, Thio and Dithiocarboxylic acids and their derivatives, in; A. Senning (Ed.), Topics in Sulfur Chemistry, Vol. 4, Thieme Verlag, Stuttgart, 1979. 31 G. Seybold, Aliphatische Thiocarbons~urechloride. Darstellung und Eigenschaften, Angew. Chem., 87 (1975) 710--711. 32 W.B. Jacoby, J. Stevens, M.W. Duffel and R.A. Weisinger, The terminal enzymes of mercapturate formation and the thiomethyl shunt, in: E. Hodgson, J.R. Bend and R.M. Philpot (Eds.), Reviews in Biochemical Toxicology, Vol. 6, Elsevier/NorthHolland Press, Amsterdam, 1985. 33 A.A. Elfarra and M.W. Anders, Renal processing of glutathione conjugates. Role in nephrotoxicity, Biochem. Pharmacol., 33 (1984) 3729--3732. 34 D. Reichert, S. Schiitz and M. Metzler, Excretion pattern and metabolism of hexachlorobutadiene in rats. Evidence for metabolic activations by conjugation reactions, Biochem. Pharmacol., 34 (1985) 499--505. 35 D. Reichert and S. Schiitz, Mercapturic acid formation is an activation and intermediary step in the metabolism o f hexachlorobutadiene, Biochem. Pharmacol. (1986) in press.