Enzyme-mediated dichloromethane toxicity and mutagenicity of bacterial and mammalian dichloromethane-active glutathione S-transferases

Article abstract of DOI:10.1007/s002040050589

The kinetic properties of bacterial and rat liver glutathione S- transferases (GST) active with dichloromethane (DCM) were compared. The theta class glutathione S-transferase (rGSTT1-1) from rat liver had an affinity for dihalomethanes lower by three orders of magnitude (K(app) > 50 mM) than the bacterial DCM dehalogenase/GST from Methylophilus sp. DM11. Unlike the bacterial DCM dehalogenase, the rat enzyme was unable to support growth of the dehalogenase minus Methylobacterium sp. DM4-2cr mutant with DCM. Moreover, the presence of DCM inhibited growth with methanol of the DM4-2cr transconjugant expressing the rat liver GSTT1-1. In Salmonella typhimurium TA1535, expression of rat and bacterial DCM-active GST from a plasmid in the presence of DCM yielded up to 5.3 times more reversions to histidine prototrophy in the transconjugant expressing the rat enzyme. Under the same conditions, however, GST-mediated conversion of DCM to formaldehyde was lower in cell-free extracts of the transconjugant expressing the rat GSTT1 than in the corresponding strain expressing the bacterial DCM dehalogenase. This provided new evidence that formaldehyde was not the main toxicant associated with GST-mediated DCM conversion, and indicated that an intermediate in the transformation of DCM by GST, presumably S-chloromethylglutathione, was responsible for the observed effects. The marked differences in substrate affinity of rat and bacterial DCM-active GST, as well as in the toxicity and genotoxicity associated with expression of these enzymes in bacteria, suggest that bacterial DCM dehalogenases/GST have evolved to minimise the toxic effects associated with glutathione-mediated catalysis of DCM conversion.

Full text of DOI:10.1007/s002040050589

Arch Toxicol (1999) 73: 71±79
Ó Springer-Verlag 1999
TOXICOKINETICS AND METABOLISM
Daniel Gisi á Thomas Leisinger á Stephane Vuilleumier
Â
Enzyme-mediated dichloromethane toxicity and mutagenicity
of bacterial and mammalian dichloromethane-active glutathione
S-transferases
Received: 14 September 1998 / Accepted: 11 January 1999
Abstract The kinetic properties of bacterial and rat liver Key words Dichloromethane á Dichloromethane
glutathione S-transferases (GST) active with dichloro- dehalogenase á Formaldehyde á Glutathione
methane (DCM) were compared. The theta class gluta- S-transferase á Bacterial genotoxicity
thione S-transferase (rGSTT1-1) from rat liver had an
anity for dihalomethanes lower by three orders of
magnitude (Kapp > 50 mM) than the bacterial DCM Introduction
dehalogenase/GST from Methylophilus sp. DM11. Un-
like the bacterial DCM dehalogenase, the rat enzyme Dichloromethane (DCM) is widely used as an industrial
was unable to support growth of the dehalogenase mi- solvent (Howard 1990). DCM has been the subject of
nus Methylobacterium sp. DM4-2cr mutant with DCM. extensive toxicological studies because of its demon-
Moreover, the presence of DCM inhibited growth with strated tumorigenicity in mouse lung and liver (Green
methanol of the DM4-2cr transconjugant expressing the 1997), and of its suspected carcinogenicity in human
rat liver GSTT1-1. In Salmonella typhimurium TA1535, erythrocytes, liver and kidney (Thier et al. 1998). Such
expression of rat and bacterial DCM-active GST from a tumours are believed to result from DCM conversion to
plasmid in the presence of DCM yielded up to 5.3 times formaldehyde by glutathione S-transferases (GST), via a
more reversions to histidine prototrophy in the trans- short-lived glutathione conjugate, S-chloromethylgluta-
conjugant expressing the rat enzyme. Under the same thione (reviewed in Green 1997). The genotoxic poten-
conditions, however, GST-mediated conversion of tial of formaldehyde is well documented (O'Donovan
DCM to formaldehyde was lower in cell-free extracts of and Mee 1993; Graves et al. 1996; Casanova et al. 1997).
the transconjugant expressing the rat GSTT1 than in the S-chloromethylglutathione, a known DNA-alkylating
corresponding strain expressing the bacterial DCM agent (Dechert 1995; Green 1997), is most probably also
dehalogenase. This provided new evidence that formal- mutagenic (Graves et al. 1994, 1996).
dehyde was not the main toxicant associated with GST-
Mammalian GST enzymes active with DCM
mediated DCM conversion, and indicated that an (GSTT1-1) have been puri®ed from rat (Meyer et al.
intermediate in the transformation of DCM by GST, 1991), mouse (Mainwaring et al. 1996a) and human
presumably S-chloromethylglutathione, was responsible (Schro
for the observed e€ects. The marked di€erences in sub- cloned and sequenced (Pemble et al. 1994; Mainwaring
strate anity of rat and bacterial DCM-active GST, as et al. 1996b; Schroder et al. 1996). Dihalomethane-spe-
È
der et al. 1996), and the corresponding genes were
È
well as in the toxicity and genotoxicity associated with ci®c GST, so-called dichloromethane dehalogenases,
expression of these enzymes in bacteria, suggest that had been characterised previously in methylotrophic
bacterial DCM dehalogenases/GST have evolved to bacteria growing with DCM as the sole carbon source
minimise the toxic e€ects associated with glutathione- (reviewed in Leisinger et al. 1994). The gene encoding
mediated catalysis of DCM conversion.
these enzymes was characterised in Methylobacterium
sp. DM4 (La Roche and Leisinger 1990) and in Meth-
ylophilus sp. DM11 (Bader and Leisinger 1994). Bac-
terial dichloromethane dehalogenases/GST are thought
to transform DCM by the same mechanism as the
mammalian DCM-active GST enzymes (Blocki et al.
D. Gisi á T. Leisinger á S. Vuilleumier (&)
Mikrobiologisches Institut, ETH Zurich, ETH Zentrum,
È
1
994). Bacterial and mammalian DCM-active GST
CH-8092 Zu
e-mail: svuilleu@micro.biol.ethz.ch,
Tel.: ++41 1 632 33 57; Fax: ++41 1 632 11 48
È
rich, Switzerland
share ꢀ25% sequence identity at the protein level, while
the mammalian sequences are ꢀ80% identical to one

72
another. DM4 and DM11 DCM dehalogenases are most BL21(DE3)pLysS for expression of bacterial DM11 dehalogenase
from plasmid pME1919 (Vuilleumier and Leisinger 1996). Trans-
conjugants of Methylobacterium sp. DM4-2cr (La Roche and Lei-
singer 1990) were constructed by biparental mating using E. coli
S17-1 (Simon et al. 1983) as the donor. Transconjugants of Sal-
closely related to each other in sequence databases (56%
amino acid identity), suggesting that bacterial dihalo-
methane-speci®c GST diverged from a common pre-
cursor hundreds of millions of years ago already (Bader monella typhimurium TA1535 (Maron and Ames 1983) were
obtained by heat shock transformation with plasmids isolated from
S. typhimurium LB5000 (Bullas and Ryu 1983) as the intermediate
host. The identity of all new strains was veri®ed by plating on
selective media, polymerase chain reaction (PCR) ampli®cation of
and Leisinger 1994). Even though DCM may have ap-
peared in large quantities on earth only since industrial
times, sucient amounts may have been generated nat-
urally to exert selective pressure for the evolution of the dcmA and rGSTT1 genes, and by measuring DCM dehaloge-
nation in cell-free extracts as described below.
DCM-speci®c GST. Production of DCM in volcanic
eruptions has been postulated (Isidorov 1990), and other
plausible mechanisms for the synthesis of DCM by
natural processes have recently been reviewed (Urhahn
Molecular cloning
and Ballschmiter 1998). Alternatively, DCM dehaloge-
Standard recombinant DNA techniques were used (Ausubel et al.
nases may have originally evolved to degrade other 1998). The broad host range plasmid pME1685 and the corre-
naturally occurring substrates structurally related to sponding transconjugant strain Methylobacterium sp. DM4-
2
cr(pME1685), designated here DM4-2cr(DM11), were described
DCM, such as dibromomethane (Goodwin et al. 1997)
or methyl chloride (Harper 1997).
There is an ongoing debate on the relevance of the
previously (Gisi et al. 1998). This transconjugant strain allowed the
constitutive expression of the DCM dehalogenase dcmA gene of
Methylophilus sp. DM11 in Methylobacterium sp. DM4-2cr. Plas-
mouse model for assessing DCM tumorigenicity in hu- mid pME1690, a plasmid otherwise identical to pME1685 but car-
rying rGSTT1, the gene for the rat liver GSTT1-1 (Pemble and
Taylor 1992), in place of the DM11 DCM dehalogenase, was con-
structed as follows. The HindIII site downstream of rGSTT1 in
pKK233-2/GST5(+) (Thier et al. 1993) was removed and a BamHI
mans. Rats and hamsters, unlike mice, do not form liver
or lung tumours when exposed to DCM (Mainwaring et
al. 1996b, 1998; Green 1997; Liteplo et al. 1998). DCM-
active GST appeared to be predominantly located in the site inserted by digestion with HindIII, end-®lling with Klenow
fragment and subsequent insertion of a BamHI linker. A 750 bp
fragment carrying rGSTT1 was excised from the resulting plasmid
nucleus in the mouse but not in the rat (Mainwaring
et al. 1996b, 1998), although recent data indicate that
the cellular location of the GSTT1-1 enzyme may
by digestion with NcoI and mung bean nuclease (Boehringer
Mannheim), followed by partial digestion with BamHI. The P
A
depend on the conditions that induce its expression promoter of the dcmA gene of Methylobacterium sp. DM4 was
Sherratt et al. 1998). Bacteria, lacking a nuclear com- PCR-ampli®ed from plasmid pME1540 (Gisi et al. 1998) using the
(
T3 universal primer ATTAACCCTCACTAAAGG and primer
partment would also be expected to be highly susceptible
to DCM turnover by DCM-active GST enzymes. Hence,
DCM exposure of Salmonella sp. Ames tester strains
CACGTTATCCTCCCTTACTG. The resulting PCR product was
treated with T4 DNA polymerase, cut with HindIII, and ligated
together with the blunt-BamHI rGSTT1 fragment into the BamHI-
expressing rat liver GSTT1-1 as intracellular enzyme and HindIII-restricted pBluescript-KSII(+) (Stratagene) derivative
pME1673 previously used in the cloning of dcmA genes (Gisi et al.
from plasmids had marked mutagenic e€ects (Thier
et al. 1993; Oda et al. 1996).
These studies prompted us to investigate the toxicity
1998), yielding plasmid pME1688. The fragment containing
rGSTT1 obtained by sequential digestion of plasmid pME1688 with
Bsp1201, Klenow treatment and XbaI digestion was next ligated
and mutagenicity associated with DCM turnover by into the broad host range plasmid pJB3Km1 (Blatny et al. 1997)
which had been sequentially cut with SphI, treated with Klenow
fragment and digested with XbaI, yielding plasmid pME1690. The
transconjugant strain Methylobacterium sp. DM4-2cr(pME 1690)
will hitherto be referred to as DM4-2cr(GsTT1) for simplicity.
Four derivatives of the expression vector pTrc99A (Pharmacia)
dichloromethane dehalogenase/GST during growth of
methylotrophic bacteria with DCM as the sole carbon
source. In this work, we compare the enzymatic prop-
erties of the bacterial DCM dehalogenase from Meth-
ylophilus sp. DM11 (Vuilleumier and Leisinger 1996) were constructed, which featured either the DCM dehalogenase
gene dcmA from Methylophilus sp. DM11 or rGSTT1 in both sense
and antisense orientations behind the isopropyl-b-D-thiogalacto-
and of rat liver GSTT1-1 (Meyer et al. 1991), and the
e€ects of the expression of these two enzymes in bacteria
exposed to DCM.
side (IPTG)-inducible Ptrc promoter. The dcmA gene from Meth-
ylophilus sp. DM11 was PCR-ampli®ed with primers
AGAGAAATTACCATGGGTACTAAAC and GACCATTA-
TAGGATCCTGATATTCAC containing NcoI and BamHI sites,
respectively, and inserted as NcoI-BamHI fragment into pTrc99A,
yielding plasmid pME1983. The latter encodes the S2G mutant of
the DM11 DCM dehalogenase gene behind the Ptrc promoter of
expression vector pTrc99A. The corresponding S. typhimurium
transconjugant strain was designated TA1535(DM11+). Insertion
of the NcoI-HindHIII rGSTT1 fragment from pKK233-2/GST5 (+ )
Materials and methods
Reagents
Restriction and DNA-modifying enzymes were purchased from
Fermentas (Maechler, Basel, Switzerland) if not noted otherwise.
Oligonucleotides were purchased from Microsynth (Balgach,
Switzerland). All other chemicals were analytical grade and were
purchased from Fluka (Buchs, Switzerland) except where noted.
(Thier et al. 1993) yielding plasmid pME1895, and the corre-
sponding strain TA1535(GSTT1+). Control plasmids with dcmA
and rGSTT1 genes in the antisense orientation behind the Ptrc
promoter of vector pTrc99A were obtained by ligating the GST
genes fragments from the (+ ) plasmids as NcoI-digested, T4 DNA
polymerase-treated, and HindIII-digested fragments with DNA
fragments from plasmids pME1983 and pME1895, which had been
Bacterial strains
Escherichia coli XL1-Blue (Stratagene) was used for cloning, plas- successively digested with PstI, treated with T4 DNA polymerase
mid ampli®cation, and expression of rat liver GSTT1-1, and E. coli and cut with HindIII. The resulting control plasmids pME1984 and

7
3
pME1896 were introduced in S. typhimurium yielding strains Enzyme measurements
TA1535(DM11-) and TA1535(GSTT1-), respectively.
The speci®c DCM dehalogenation in cell-free extracts was mea-
sured by chemical derivatization of formaldehyde as described
previously (Vuilleumier and Leisinger 1996). Kinetic parameters of
the puri®ed enzymes with DCM, dibromomethane, chloro¯uoro-
Media and growth conditions
E. coli strains were grown in Luria-Bertani medium at 37 °C. methane (Fluorochem, Old Glossop, Derbyshire, UK) and gluta-
Transconjugant derivatives of Methylobacterium sp. DM4-2cr were thione were determined using a previously described coupled assay
grown in minimal medium containing 25 lg kanamycin/ml as de- with formaldehyde dehydrogenase (Vuilleumier and Leisinger
scribed elsewhere (Gisi et al. 1998), with methanol (40 mM) or 1996). The reactions were performed at 30 °C in gas-tight spec-
DCM (10 mM) as the carbon source. Transconjugants of trophotometric cuvettes in 100 mM sodium/potassium phosphate,
S. typhimurium TA1535 were grown as described (Maron and pH 8.0.
Ames 1983), with 100 lg ampicillin/ml for plasmid selection.
Kinetic parameters V and Km for dihalomethanes were deter-
mined at a ®xed concentration of 5 mM glutathione, and the ap-
parent K
dihalomethane, respectively. Kinetic parameters were obtained by
tting the experimental data directly to the Michaelis-Menten
m
for glutathione at a ®xed concentration of 20 mM
Mutagenicity assays
®
Mutagenicity assays were performed according to standard pro- equation by non-linear least-square curve ®tting using Kaleida-
tocols for the Ames test (Maron and Ames 1983) with some graph (Synergy software). Values for kcat/Km of rat liver GSTT1-1
modi®cations. Brie¯y, transconjugants of S. typhimurium TA1535 were estimated from the initial rate of formaldehyde production
were grown at 37 °C in nutrient broth medium containing 100 lg with 2±20 mM dihalomethane was described (Danielson and
ampicillin/ml. The saturated culture was diluted 1000-fold into Mannervik 1985).
fresh medium and the incubation continued until the cultures had
The inhibition of DM11 dehalogenase and rGSTT1-1 activity
achieved an optical density (OD600) of 0.4. The expression of the with DCM by S-methylglutathione and chloroacetonitrile was de-
enzymes was then induced by adding IPTG (Biosynth AG, Staad, termined at 30 °C in 100 mM phosphate bu€er, pH 8.0, containing
Switzerland) to a ®nal concentration of 100 lM and the incubation 20 mM DCM and 1 mM GSH. The GST-dependent conjugation
temperature was lowered to 25 °C, to ensure expression of the of chloromethane to glutathione was determined by following
DM11 dehalogenase as soluble and active protein (Bader and chloromethane depletion by gas chromatography. Chloromethane
Leisinger 1994).
gas (60 ll) was added to 14 ml gas-tight vials containing 3 ml
After 6 h incubation at 25 °C, 100 ll aliquots of the cultures 100 mM sodium/potassium phosphate (pH 8.0) with 5 mM GSH
were used for mutagenicity assays, which were performed in du- and 5±10 lg puri®ed enzyme. This resulted in an initial calculated
plicate. Bacteria were resuspended and preincubated for 20 min at concentration of chloromethane of 3300 ppm in the gas phase and
37 °C in a total liquid volume of 200 ll of 200 mM sodium of 0.35 mM in the liquid phase, basing on a Henry constant of 0.43
phosphate bu€er, pH 7.4, in 10 ml closed glass vials containing for chloromethane at this temperature (Gossett 1987). Brie¯y,
varying amounts of a 0.26% (v/v; 40 mM) stock solution of DCM 100 ll of the headspace vial volume was injected into a Porapak P
dissolved in the same bu€er. The bacterial suspensions were then column (1800 ´ 2 mm, mesh 80/100, Supelco) connected to PE8700
diluted in liquid top agar and spread onto minimal medium agar gas chromatograph (Perkin-Elmer) equipped with a ¯ame ioniza-
plates. The plates were incubated for 72 h at 37 °C and the tion detector. Nitrogen at a ¯ow rate of 40 ml/min was used as
+
number of his -revertants was determined. Strain viability was the carrier. The temperature was set to 150 °C in the column
checked by plating dilutions of the cultures from the preincuba- and 250 °C in the detector. Under these conditions, the retention
tion step with DCM on minimal medium agar containing 50 lg time of chloromethane was 0.6 min with a detection limit of
histidine/ml without antibiotics. Colonies were counted after 72 h 10 lM. The activity was obtained from a plot of the natural
incubation at 37 °C and expressed as colony forming units (cfu) logarithm of chloromethane concentration against time and had
per ml culture.
units (h á mg protein).
Preparation and analysis of cell extracts
Cell-free extracts from Methylobacterium sp. DM4-2cr and
S. typhimurium TA1535 transconjugant strains were obtained by
repeated passage through a French press apparatus as described
previously (Gisi et al. 1998). Proteins were separated by sodium
Results
Characterization of the catalytic properties
dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) of DCM-dehalogenating GST
on minigels (CBS Scienti®c Company) by standard protocols
(
Ausubel et al. 1998). DM11 DCM dehalogenase and rGSTT1-1
bands were quanti®ed as the percentage of the soluble cell protein
detection limit 0.5%) using ImageQuant software (Molecular
The kinetic parameters of the bacterial DCM dehalo-
genase from Methylophilus sp. DM11 and of the rat liver
glutathione S-transferase rGSTT1-1 with di€erent
dihalomethanes are shown in Table 1. The bacterial
DCM dehalogenase feature Km values between 3.5 lM
and 100 lM for the three substrates tested, but the af-
(
Dynamics).
GST puri®cation
DM11 dehalogenase, expressed from E. coli BL21(pME1919) ®nity of rGSTT1-1 for the same substrates was weaker
(
Vuilleumier and Leisinger 1996), was puri®ed by anion exchange
chromatography (ResourceQ, Pharmacia) as described previously
Vuilleumier and Leisinger 1996), using a Biocad Sprint Perfusion
Chromatography System (Perseptive Biosystems). Rat liver theta
by several orders of magnitude (Fig. 1). Half-saturation
constants of rGSTT1-1 with dihalomethanes were esti-
mated to be >50 mM in all cases. The speci®c deha-
(
GSTT1-1 expressed from E. coli XL1-Blue(pME 1688) was puri®ed logenation activity increased up to the water solubility
essentially as described (Meyer et al. 1991; Meyer and Ketterer
1995), except that the anion exchange step was performed prior to
Matrex gel Orange (Amicon, Millipore) chromatography. GST
fractions of >95% (w/w) purity were pooled, ¯ash frozen and
stored in liquid nitrogen until further use.
limit of dihalomethanes in water (Fig. 1; approximately
7
0 mM for dibromomethane, 120 mM for chloro¯uo-
romethane, and 200 mM for the most soluble dihalo-
methane, DCM).

74
Table 1 Kinetic analysis of the conversion of CH
and CH
tathione S-transferase (rGSTT1-1). Puri®ed enzymes were assayed tivity, GSH glutathione)
2
Br
2
, CH
2
Cl
2
,
Materials and methods (DCM dichloromethane, K
m
Michaelis
2
ClF by bacterial DM11 dehalogenase and rat liver glu- constant, S0.5 50% saturation level, V velocity, kcat catalytic ac-
spectrophotometrically at 20 mM DCM as described in the
dihalomethane
GSH
0.5 (lM)
GSH
m
Substrate
K
m
(lM)
S
K
(lM)
V(lmol/min per mg)
m
kcat/K (M/s)
a
DM11
dehalogenase
rGSTT1-1 DM11
rGSTT1-1
DM11
dehalogenase
rGSTT1-1 DM11
rGSTT1-1
92 ‹ 14
dehalogenase
dehalogenase
b
CH
CH
CH
2
2
2
Cl
Br
ClF
2
59 ‹ 6*
3.5 ‹ 0.4* >50 000
100 ‹ 8 >50 000
>50 000
66 ‹ 3b
78 ‹ 3
82 ‹ 7
61 ‹ 6
1210 ‹ 98
60 ‹ 7
6.0 ‹ 0.4* 4.1 ‹ 0.2
10.0 ‹ 0.3* 3.5 ‹ 0.4
0.22 ‹ 0.01 4.3 ‹ 0.2
56 000 ‹ 1700*
1 570 000 ‹ 57 000* 89 ‹ 13
1 400 ‹ 20 95 ‹ 15
2
a
Although the rGSTT1-1 enzyme was not saturated with dihalomethane substrate, <0.5& of the dihalomethane was consumed during
for glutathione at 20 mM dihalomethane to be determined
Values obtained from Table 1 of Vuilleumier and Leisinger (1996)
the time course of the experiments, allowing an apparent K
m
b
The very low anity of the rGSTT1-1 enzyme for DCM-active GST perform such a role by testing inhi-
dihalomethanes was unexpected. Although the kinetics bition of glutathione-mediated dihalomethane conver-
of DCM conversion to formaldehyde in human hae- sion by S-methylglutathione. Neither rGSTT1-1 (IC50 of
molysate did suggest a low anity of GSTT1-1 enzymes 6.4 mM) nor the DM11 dehalogenase (IC50 of 5.5 mM),
for DCM (Hallier et al. 94), a K of 300 lM for the rat however, were signi®cantly inhibited by this compound,
m
GSTT1-1 enzymes with DCM was previously reported suggesting that S-methylglutathione was an inadequate
(Blocki et al. 1994). Our puri®ed preparation of mimic for a reaction intermediate in the conversion of
rGSTT1-1 had a speci®c activity with 1,2-epoxy-3-p-ni- DCM by GST. The production of S-methylglutathione
trophenoxypropane (ENPP) of 205 lmol/min per mg, by GSTT1-1 catalysed conjugation of glutathione to
which compared well with the published value of chloromethane, however, was previously demonstrated
220 lmol/min per mg (Meyer and Ketterer 1995). In in cell-free extracts of rat liver (Thier et al. 1998). In the
addition, the dependence of the dehalogenation rate of present work, the activity of the rat GSTT1-1 enzyme
rGSTT1-1 on the concentration of glutathione followed with chloromethane (211 h/mg) was higher than that of
apparent Michaelis-Menten kinetics (Table 1), indicat- the bacterial DCM dehalogenase from D111 (53 h/mg)
ing that the protein was properly folded. in contrast to what that observed with dichloromethane
The activity and catalytic eciency kcat/Km of the as the substrate (Fig. 1, Table 1). This con®rmed that
bacterial DCM dehalogenase varied signi®cantly with the two enzymes have markedly di€erent catalytic
di€erent dihalomethanes (Table 1), with CH ClF being
2
converted at a >30-fold slower rate than the other two
dihalomethanes tested. In contrast, the speci®c activity
of rGSTT1-1 and its catalytic eciency kcat/K , esti-
m
mated from the initial rates of dehalogenation at low
substrate concentrations (Danielson and Mannervik
1985), were very similar for the three dihalomethanes.
A further di€erence between the bacterial and the
mammalian DCM-active GST was noted for the inhi-
bition of enzyme-catalysed DCM conversion by
chloroacetonitrile in the presence of 20 mM DCM. The
previously reported tight binding of chloroacetonitrile to
the DCM dehalogenase from strain DM11 (Logan et al.
1993) was con®rmed [50% inhibitory concentration
(
IC ), 0.3 lM]. In the case of rGSTT1-1, however,
50
chloroacetonitrile had little e€ect on DCM conversion
under the same conditions (IC50 of 670 lM). Thus,
bacterial DCM dehalogenases, but not the rGSTT1-1
enzyme, may feature a speci®c well-de®ned binding site
for dihalomethanes. This is supported by the observa-
tion that the binding of chloroacetonitrile to the DM11
2 2 2 2
GST was previously shown to be competitive (Logan Fig. 1 Kinetic analysis of the conversion of CH Br , CH Cl and
2
CH ClF by DCM-active glutathione S-transferases. Formaldehyde
production by bacterial DM11 dehalogenase (open symbols) and rat
liver GSTT1-1 (®lled symbols) was measured at various concentrations
et al. 1993).
Another function of GST may be to sequester toxic
conjugates, thereby preventing their further conversion
2 2 2 2 2
of CH Br (h, j) CH Cl (s,d), and CH CIF (n, m) as described in
(Meyer 1993). We investigated the possibility that the Materials and methods

7
5
properties. Rates for the dehalogenation of chloro-
methane and dichloromethane, however, are not directly
comparable because of the di€erent order of the reac-
tions. Moreover, it is not yet known whether the
chloromethane concentration used (3300 ppm in the gas
phase, see the Materials and methods) represented sat-
urating conditions of substrate for either the bacterial
DCM dehalogenase from strain DM11 or the rat
GSTT1-1 enzyme, since the anity of both enzymes for
chloromethane remains to be determined.
E€ects of the expression of DCM-active GST
in Methylobacterium sp. DM4-2cr
and S. typhimurium TA 1535
The genes for Methylophilus sp. DM11 dcmA and rat
liver GSTT1, under the control of the dcmA promoter of
strain DM4 (Gisi et al. 1998) in otherwise identical
broad host range plasmids, were expressed constitutively
in the dehalogenase minus mutant Methylobacterium sp.
DM4-2cr (Table 2). The transconjugant DM4-
2cr(DM11) grew well in liquid minimal medium con-
taining either 40 mM methanol, 10 mM DCM or a
mixture of both substrates as carbon sources (Fig. 2A,
Fig. 2A, B Growth of Methylobacterium sp. DM4-2cr transconjug-
Table 2). The reasons for the low expression of DCM ants in minimal medium. Three independent cultures of Methylobac-
dehalogenase in the DM4-2cr(DM11) transconjugant terium sp. DM4-2cr(DM11) (A) and Methylobacterium sp.
DM4-2cr(GSTT1) (B) were grown at 30 °C in gas-tight ¯asks with
observed during growth with a methanol/DCM mixture
(
minimal medium containing 40 mM methanol (s, left curve). At
Table 2) are not yet understood. Loss of the dcmA gene
could be ruled out in this case, however, since plating
eciency of this strain after growth on a methanol/ (h), or a mixture of 40 mM methanol and 10 mM CH
t = 50 h (arrow), precultures were diluted 50-fold into fresh minimal
medium and grown with either 40 mM methanol (s), 10 mM CH Cl
Cl (d). Upon
reaching the stationary phase, the cultures grown with a methanol/
CH Cl mixture were diluted 50-fold (arrow) into fresh minimal
medium, again with a mixture of 40 mM methanol and 10 mM
CH Cl (d)
2
2
2
2
DCM mixture was the same with either DCM or
methanol as the carbon source (data not shown).
2
2
In contrast, the DM4-2cr(GSTT1) transconjugant
grew well with methanol only (Fig. 2B), despite ex-
pressing slightly higher levels of rGSTT1-1 enzyme than
the otherwise identical transconjugant expressing the
bacterial DCM dehalogenase (Table 2). The DM4-
2
2
extracts of DM4-2cr(GSTT1) cultures which harvested
at early stationary phase after exposure to DCM during
the exponential growth phase (Table 2), and speci®c
primers failed to amplify the rGSTT1 gene from bacte-
rial colonies recovered from the ®nal culture (data not
shown). After reinoculation, cultures that had lost the
rGSTT1 gene grew at the same rate with methanol/
2cr(GSTT1) transconjugant could not grow with DCM
as the sole source of carbon, and its growth rate with
methanol/DCM as carbon sources was reduced to about
half that observed with methanol alone. Moreover,
DCM dehalogenation activity was no longer detected in
Table 2 Growth rates and speci®c DCM dehalogenase activity
of Methylobacterium sp. DM4-2cr transconjugants expressing
DCM-active GST. Growth rates l (h ) of Methylobacterium sp.
DM4-2cr(DM11) and Methylobacterium sp. DM4-2cr(GSTT1)
were obtained from the data in Fig. 2 (t = 50±110 h). Bacteria
were harvested at early stationary phase, and the speci®c activity
and the percentage of total soluble cell protein of DCM-active
GST were determined as described in the Materials and methods
(GST Glutathione S-transferase)
)1
DM4-2cr(DM11)
DM4-2cr(GSTT1)
4
0 mM methanol 40 mM methanol 10 mM DCM 40 mM methanol 40 mM methanol/ 10 mM DCM
/
10 mM DCM
10 mM DCM
l (h)
1
)
0.124 ‹ 0.002
0.23 ‹ 0.01
3.8 ‹ 0.3
0.106 ‹ 0.003
0.05 ‹ 0.01
0.9 ‹ 0.2
0.094 ‹ 0.008 0.112 ‹ 0.001
0.22 ‹ 0.01
3.6 ‹ 0.4
0.053 ‹ 0.005
n.d
<0.5
n.d.
n.d
<0.5
a
b
V(lmol/min per mg)
Cell protein% (w/w)
0.07 ‹ 0.02
5.1 ‹ 0.4
a
With 10 mM DCM and 5 mM GSH
n.d., detectable activity (detection limit = 1 nmol/min per mg)
b

76
DCM as with methanol (Fig. 2B, t ˆ 120±14 h). This
suggested that the reduced growth rate of DM4-
2cr(GSTT1) observed with methanol/DCM (Fig. 2B,
t = 50±110 h) arose from a mixture of slowly dividing
cells which still expressed GSTT1-1 and of mutants that
had lost the rGSTT1 gene. These results constituted
clear evidence that expression of the rGSTT1 gene was
toxic to Methylobacterium sp. DM4-2cr(rGSTT1) in the
presence of DCM, and resulted in strong selection
pressure for loss of the rGSTT1 gene under those con-
ditions.
We wished to exclude the possibility that special
protection mechanisms were at work in Methylobacte-
rium, which protected this strain from the toxic e€ects
arising from DCM turnover by bacterial DCM dehalo-
genases. For this purpose, the bacterial DCM dehalo-
genase from strain DM11 and the rat liver GSTT1-1
enzyme were expressed in S. typhimurium TA1535, a
histidine-auxotrophic strain used to assess the mutagenic
+
Fig. 3 Ames his -reversion test of Salmonella typhimurium TA1535
transconjugants expressing DCM-active glutathione S-transferases
incubated with various concentrations of DCM. S. typhimurium
strains TA1535(DM11+) (j), TA1535(GSTT1+) (d) and the
potential of chemicals by the Ames test (Maron and control strains TA1535(DM11)) (h) and TA1535(GSTT1)) (s)
Ames 1983). In previous work, Thier et al. (1993) carrying the GST genes in opposite orientation to the Ptrc promoter
were incubated with DCM as described in the Materials and methods.
The given DCM concentrations refer to the assay preincubation
demonstrated that intracellular expression of the rat
GSTT1-1 enzyme in this strain by way of a derivative of
plasmid pKK233-2 carrying the rGSTT1 gene had a
volume according to the standard protocol (Maron and Ames 1983).
The numbers of his -revertants represent the mean of duplicate
+
strong mutagenic e€ect in the presence of DCM. In the determinations (variation <20%).
present study, the bacterial DCM dehalogenase from
strain DM11 and the DCM-active rat GSTT1-1 enzyme TA1535(GSTT1)) and TA1535(DM11)) with GST
were cloned in plasmid pTrc99A, a derivative of genes in reverse orientation to the Ptrc promoter did not
pKK233-2 featuring a lacl9 repressor gene that allows show any dehalogenation activity.
+
better control of gene expression by IPTG induction.
Expression of DCM-active GST was initiated by in- vertants in the presence of DCM was observed only with
A pronounced increase of the number of his -re-
duction with IPTG at 25 °C since the DCM dehaloge- the S. typhimurium TA1535(GSTT+) transconjugant
nase was expressed mainly in insoluble and inactive form (Fig. 3, Table 3). The numbers of his -revertants of
+
at 37 °C, as observed previously in E. coli (Bader and S. typhimurium TA1535(GSTT11+) were similar to
Leisinger 1994). Under these conditions, the two strains those previously measured by others with S. typhimur-
S. typhimurium TA1535(GSTT1+) and TA1535 ium expressing rGSTT1-1. (Thier et al. 1993; DeMarini
(DM1 1+ ) expressed active rat GSTT1-1 and DM11 de- et al. 1997), and arose in a dose-dependent manner, in
halogenase to 6 and 15% (w/w) of the soluble cell pro- good agreement with the kinetic properties of the
tein, respectively, as estimated from the speci®c activity rGSTT1-1 enzyme (Fig. 1). The transconjugant
in cell-free extracts relative to that of the puri®ed en- TA1535(DM11+) expressing the bacterial DCM deha-
+
zymes (Table 3). The control strains S. typhimurium logenase showed a much lower level of his -revertants
a
Table 3 DCM dehalogenase activity and viability of Salmonella phosphate bu€er, pH 8.0, containing 5 mM glutathione. Speci®c
typhimurium TA1535 transconjugants. Cultivation of S. typhimu- activity in cell-free extracts is presented as means ‹ S.D., n = 3,
rium transconjugants was performed as described in the Materials and the percentage of total soluble cell protein of DCM-active
and methods. Measurements of speci®c DCM activity in cell GST were determined as described in the Materials and methods.
extracts at 20 mM DCM (the highest concentration used in the Cell counts were performed in duplicate (variation <20%). (n.d. no
mutagenicity experiment shown in Fig. 2) at 30 °C in 100 mM detectable activity, cfu colony forming units)
9
cfu (10 /ml) after 20 min preincubation
with
S. typhimurium
transconjugant
V(lmol/min per mg)
DCM-active GST
in cell protein (%)
0
mM DCM
20 mM DCM
TA1535(GSTT1+)
TA1535(GSTT1))
TA1535(DM11+)
0.25 ‹ 0.03
n.d.
0.90 ‹ 0.07
15
n.d.
6
1.9
1.7
1.8
2.5
1.3
2.2
*
*
TA1535(DM11))
n.d.
n.d.
1.5
1.7
a
As a result of the cloning procedure, the DCM dehalogenase in these experiments was actually the S2G variant of the DM11 enzyme, but
showed properties indistinguishable from the wild-type enzyme (data not shown)

7
7
(Fig. 3). In this case as well, the reversion frequency controlled and speci®c way. The main function of
increased with DCM concentration and followed the mammalian GSTT1-1 enzymes in dihalomethane con-
saturation kinetics of the enzyme with DCM as a sub- version may be merely to activate the thiol group of
strate (Fig. 1), reaching its maximum at submillimolar the glutathione cofactor to a more nucleophilic state,
concentrations. thereby facilitating its reaction with dihalomethanes or
Treatment of transconjugants with 20 mM DCM in halomethanes which have di€used into the active site of
phosphate bu€er, pH 7.4 at 37 °C caused 5.3 times more the enzyme.
+
his -revertants in TA1535(GSTT1+) than in
It is noteworthy that the accessibility of the substrate-
TA1535(DM11+), although the viability of all trans- binding site of GST is de®ned to an important extent by
conjugants exposed to this treatment was within the the structure and length at the C-terminal end of the
same range (Table 3). The activity of the puri®ed bac- protein. A comparison of crystal structures revealed that
terial DCM dehalogenases (5.1 lmol/min per mg) and theta-class enzymes have the least accessible substrate-
of the puri®ed rat GSTT1-1 enzyme (5.4 lmol/min per binding site of the GST superfamily (Rossjohn et al.
mg) under these conditions was very similar to the ac- 1998). Bacterial DCM dehalogenase polypeptides are
tivity under standard assay conditions (30 °C, pH 8.0) at ~20 residues longer than the rGSTT1 protein at the C-
the same concentration of substrate (6 and 4.1 lmol/min terminus. Thus, the C-terminal end of the protein chain
per mg, respectively, Table 1). Thus, taking into account may be instrumental in de®ning a speci®c site for bind-
the expression levels of the two enzymes expressed in ing and conversion of dihalomethanes in bacterial DCM
Salmonella (Table 3), conversion of DCM by rGSTT1-1 dehalogenases.
under these conditions was ~12-fold [5.3 ´ (5.1 lmol/
min/mg ´ 15%) / (5.4 lmol/min/mg ´ 6%)] more muta-
genic per turnover than by the bacterial DCM dehalo- Nature of the toxic agent in GST-mediated
genase from Methylophilus sp. strain DM11 at 20 mM dichloromethane toxicity
dichloromethane substrate concentration, despite a
much lower rate of formaldehyde production in Formaldehyde could be ruled out as the toxic agent
TA1535(GSTT1+).
arising from DCM conversion by the rat GSTT1-1
enzyme in both Methylobacterium sp. DM4-2cr and
S. typhimurium TA1535 for several reasons. In both
backgrounds, formaldehyde production from GST-me-
diated DCM conversion was higher in the transconjug-
ants expressing the bacterial DCM dehalogenase from
Discussion
Rationale for the catalytic di€erences
between bacterial DCM dehalogenases and mammalian strain DM11 (Tables 2, 3), despite the fact that DCM
DCM-active GST
turnover by this enzyme was essentially non-toxic to these
bacteria. Also, formaldehyde does not accumulate in
Mammalian GST usually serve as broad spectrum en- methylotrophic bacteria such as Methylobacterium, as it
zymes for the detoxi®cation of xenobiotic compounds. is immediately used for bacterial growth. Moreover,
In contrast, bacterial DCM dehalogenases are key en- formaldehyde is known to be more genotoxic in Salmo-
zymes in the transformation of the growth substrate by nella strains other than S. typhimurium TA1535 that
the bacterial host, which are highly substrate-speci®c feature the R-factor plasmid pMK101 and a functional
(Vuilleumier 1997). We report here that the main dif- copy of the uvrB gene (O'Donovan and Mee 1993; Graves
ference between mammalian and bacterial DCM-active et al. 1994). Incidentally, the mutation spectrum of
GST, overlooked until now, resides in the anity of GST-mediated DCM turnover predominantly caused
these enzymes for dihalomethanes (Fig. 1). In addition, transition mutations in S. typhimurium TA1535 (DeMa-
our results demonstrate that bacterial DCM dehaloge- rini et al. 1997), as observed previously in Chinese hamster
nases also avoid the toxicity and genotoxicity associated ovary cells (Graves et al. 1996), whereas exposure to
with glutathione-dependent conversion of DCM previ- formaldehyde only produced transversions in the latter
ously observed with mammalian enzymes, and which study.
were con®rmed in the present work (Figs. 2, 3). The
The lack of mutagenicity observed here with the
substrate anity and maximal turnover rate of the control transconjugants S. typhimurium TA1535
bacterial DCM dehalogenase from strain DM11 varied (GSTT1)) and TA1535(DM11)) in the presence of
signi®cantly with di€erent dihalomethanes (Table 1). In DCM also ruled out signi®cant contributions to the
contrast, the rat GSTT1-1 enzyme had only very low toxication process of the abiotic DCM conjugation of
anity for dihalomethanes (Table 1), but achieved a glutathione (Osterman-Golkar et al. 1983), of endoge-
higher turnover rate than bacterial DCM dehalogenases nous P450 cytochromes and DCM dehalogenases
at very high substrate concentrations (Fig. 1). We (Graves et al. 1994; Green 1997), or of dichloromethane
thus speculate that bacterial DCM dehalogenases/ itself (Dillon et al. 1992). The comparison of the toxic
GST, but not mammalian GSTT-1 enzymes, are able to e€ects of DCM turnover by the bacterial or mammalian
process DCM and subsequent reaction intermediates representatives of DCM-converting GST provides new
arising in its transformation to formaldehyde in a and conclusive evidence that an intermediate in the

78
GST-catalysed degradation of DCM, rather than its Bullas LR, Ryu J-I (1983) Salmonella typhimurium LT2 strains
)
+
which are r
of DNA-restriction and modi®cation. J Bacteriol 156: 471±
74
m
for all three chromosomally located systems
product formaldehyde, is the main toxicant associated
with DCM conversion by GST.
4
There is good evidence in the literature that this toxic
intermediate may be S-chloromethylglutathione. The
glutathione adduct of DCM is extremely short-lived
(c. 4 s (Hashmi et al. 1994; Dechert 1995), and its
presence is not easily demonstrated. The formation of
Casanova M, Bell DA, Heck HdA (1997) Dichloromethane me-
tabolism to formaldehyde and reaction of formaldehyde with
nucleic acids in hepatocytes of rodents and humans with and
without glutathione S-transferase T1 and M1 genes. Fundam
Appl Toxicol 37: 168±180
Danielson UH, Mannervik B (1985) Kinetic independence of the
subunits of cytosolic glutathione transferase from the rat. Bio-
chem J 231: 263±7
S-¯uoromethylglutathione from CH ClF by DM11
2
19
DCM dehalogenase, however, was detected by F-
nuclear magnetic resonance (NMR; Blocki et al. 1994),
taking advantage of its much longer half-life (5.8 min)
Dechert S (1995) Untersuchungen zum Wirkmechanismus der
t von Dichlormethan und seinen
t Wurzburg
Mutagenita
È
Metaboliten. Thesis, Universita
t und Tumorigenita
È
È
È
compared to S-chloromethylglutathione. The reaction of DeMarini DM, Shelton ML, Warren SH, Ross TM, Shim J-Y,
S-chloromethylglutathione with DNA was demonstrat-
ed experimentally with an oligonucleotide in vitro (De-
chert 1995), supporting the idea that this compound may
be responsible for the observed mutagenic and other
genotoxic e€ects of DCM conversion by GST (reviewed
in Green 1997). S-chloromethylglutathione is considered
to be too short-lived to cross the cell membrane of
Salmonella (Green 1997; Josephy et al. 1997), and
Richard AM, Pegram RA (1997) Glutathione S-transferase-
mediated induction of GC ± AT transitions by halomethanes in
Salmonella. Environ Mol Mutagen 30: 440±447
Dillon D, Edwards I, Combes R, McConville M, Zeiger E (1992)
The role of glutathione in the bacterial mutagenicity of vapour
phase dichloromethane. Environ Mol Mutagen 20: 211±217
Gisi D, Willi L, Traber H, Leisinger T, Vuilleumier S (1998) E€ects
of bacterial host and dichloromethane dehalogenase on the
competitiveness of methylotrophic bacteria growing with di-
chloromethane. Appl Environ Microbiol 64: 1194±1202
genotoxic e€ects of dihalomethanes were much more Goodwin KD, North WJ, Lidstrom ME (1997) Production of
bromoform and dibromomethane by giant kelp: factors a€ect-
ing release and comparison to anthropogenic bromine sources.
Limnol Oceanogr 42: 1725±1734
Gossett JM (1987) Measurement of Henry's law constants for C1
extensive when the rat GSTT1-1 enzyme was expressed
within the cell (Thier et al. 1993; Oda et al. 1996), rather
than added exogenously to the bacteria, for example in
the form of the S9 fraction of rat liver (Dillon et al. 1992;
Graves et al. 1994).
and C2 chlorinated hydrocarbons. Environ Sci Technol 21:
202±208
Graves RJ, Callender RD, Green T (1994) The role of formalde-
hyde and S-chloromethylglutathione in the bacterial muta-
genicity of methylene chloride. Mutat Res 320: 235±243
Graves RJ, Trueman P, Jones S, Green T (1996) DNA sequence
analysis of methylene chloride-induced HPRT mutations in
Chinese hamster ovary cells: comparison with the mutation
spectrum obtained in 1,2-dibromoethane and formaldehyde.
Mutagenesis 11: 229±233
The conversion of DCM to formaldehyde by bacte-
rial DCM dehalogenases and mammalian DCM-active
GST probably is believed to proceed via the same toxic
S-chloromethylglutathione intermediate (Blocki et al.
1994). It is thus intriguing that the toxic e€ects associ-
ated with DCM conversion are not observed with bac-
terial DCM dehalogenases. The molecular basis for this
di€erence between bacterial DCM dehalogenases and
mammalian DCM-active GST is under investigation.
Green T (1997) Methylene chloride induced mouse liver and lung
tumours: an overview of the role of mechanistic studies in hu-
man safety assessment. Hum Exp Toxicol 16: 3±13
È
Hallier E, Schroder KR, Asmuth K, Dommermuth A, Aust B,
Acknowledgements We thank Ricarda Thier for discussions and
advice, John Taylor for providing the pKK233-2/GST5(+) plas-
mid derivative with the rGSTT1 gene, and the Ames laboratory for
providing S. typhimurium TA1535 tester strain. This work was
Goergens HW (1994) Metabolism of dichloromethane (methy-
lene chloride) to formaldehyde in human erythrocytes: in¯uence
of polymorphism of glutathione transferase Theta (GST T1-1).
Arch Toxicol 68: 423±427
supported by grant 5002-037905 from the Biotechnology Pro- Harper DB (1997) Halogenated methanes ± biological sources and
gramme of the Swiss National Science Foundation.
physiological role. In: Janssen DB, Soda K, Wever R (eds)
Mechanisms of biohalogenation and dehalogenation. Elsevier
North-Holland, Amsterdam, pp 15±31
Hashmi M, Dechert S, Dekant W, Anders MW (1994) Bioactiva-
tion of [ C]dichloromethane in mouse, rat, and human liver
References
13
13
cytosol: C nuclear magnetic resonance spectroscopic studies.
Chem Res Toxicol 7: 291±296
Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG,
Smith JA, Struhl K (1998) Current protocols in molecular bi- Howard PH (1990) Handbook of environmental fate and exposure
ology. John Wiley, New York
data for organic chemicals vol 2. Lewis, Chelsea, Michigan
Bader R, Leisinger T (1994) Isolation and characterization of the Isidorov VA (1990) Organic chemistry of the earth's atmosphere.
Methylophilus sp. strain DM11 gene encoding dichloromethane
Springer, Berlin
dehalogenase/glutathione S-transferase. J Bacteriol 176: 3466± Josephy PD, Gruz P, Nohmi T (1997) Recent advances in the
3
473
construction of bacterial genotoxicity assays. Mutat Res 386:
1±23
Blatny JM, Brautaset T, Winther-Larsen HC, Haugan K, Valla S
(
1997) Construction and use of a versatile set of broad-host- La Roche SD, Leisinger T (1990) Sequence analysis and expression
range cloning and expression vectors based on the RK2 rep-
licon. Appl Environ Microbiol 63: 370±379
Blocki FA, Logan MSP, Baoli C, Wackett LP (1994) Reaction of
of the bacterial dichloromethane dehalogenase structural gene,
a member of the glutathione S-transferase supergene family.
J Bacteriol 172: 164±171
rat liver glutathione S-transferases and bacterial dichloro- Leisinger T, Bader R, Hermann R, Schmid-Appert M, Vuilleumier
methane dehalogenase with dihalomethanes. J Biol chem 269:
826±8830
S (1994) Microbes, enzymes and genes involved in dichloro-
methane utilization. Biodegradation 5: 237±248
8

7
9
Liteplo RG, Long GW, Meek ME (1998) Relevance of carcino- Pemble SE, Taylor JB, (1992) An evolutionary perspective on
genicity bioassays in mice in assessing potential health risks
associated with exposure to methylene chloride. Hum Exp
Toxicol 17: 84±87.
glutathione transferases inferred from class-theta glutathione
transferase cDNA sequences. Biochem J 287: 957±963
Pemble S, Schroeder KR, Spencer SR, Meyer DJ, Hallier E, Bolt
HM, Ketterer B, Taylor JB (1994) Human glutathione
S-transferase theta (GSTT1): cDNA cloning and the charac-
terization of a genetic polymorphism. Biochem J 300: 271±276
Rossjohn J, MCKinstry WJ, Oakley AJ, Verger D, Flanagan J,
Chelvanayagam G, Tan K-L, Board PG, Parker MW (1998)
Human theta class glutathione transferase: the crystal structure
reveals a sulfate-binding pocket within the buried active site.
Structure 6: 309±322
Logan MSP, Blocki FA, Stimp¯ KJ, Wackett LP (1993) Halo-
acetonitriles are low K
methane dehalogenases. Biochem Biophys Res commun 197:
53±858
i
inhibitors of bacterial dichloro-
8
Mainwaring GW, Foster JR, Green T (1998) Nuclear and cellular
immunolocalization of theta-class glutathione S-transferase
GSTT1-1 in the liver and lung of the mouse. Biochem J 329:
4
31±432
Mainwaring GW, Nash J, Davidson M, Green T (1996a) Isolation Schro
È
È
der KR, Hallier E, Meyer DJ, Wiebel FA, Muller AMF, Bolt
of a mouse theta glutathione S-transferase active with methyl-
ene chloride, Biochem J 314: 445±448
Mainwaring GW, Williams SM, Foster JR, Tugwood J, Green T
HM (1996) Puri®cation and characterization of a new gluta-
thione S-transferase, class theta, from human erythrocytes.
Arch Toxicol 70: 559±566
(1996b) The distribution of theta-class glutathione S-trans- Sherratt PJ, Manson MM, Thomson AM, Hissink EAM,Neal GE,
ferases in the liver and lung of mouse, rat and human. Biochem
J 318: 297±303
Mainwaring GW, Foster JR, Green T (1998) Nuclear and cellular
van Bladeren PJ, Green T, Hayes JD (1998) Increased bioac-
tivation of dihaloalkanes in rat liver due to induction of class
theta glutathione S-transferase T1-1. Biochem J 3: 619±630
immunolocalization of theta-class glutathione S-transferase Simon R, Priefer U, Pu
GTT1-1 in the liver and lung of the mouse. Biochem J 329: 431± zation system for in vivo genetic engineering: transposon mu-
32
È
hler A (1983) A broad host range mobili-
4
tagenesis in Gram negative bacteria. Bio/Technology 1: 784±
790
Maron DM, Ames BN (1983) Revised methods for the Salmonella
mutagenicity test. Mutat Res 113: 173±215
Meyer DJ (1993) Signi®cance of an unusually low K
Thier R, Taylor JB, Pemble SE, Humphreys WG, Persmark M,
Ketterer B, Guengerich FP (1993) Expression of mammalian
glutathione S-transferase 5-5 in Salmonella typhimurium
TA1535 leads to base-pair mutations upon exposure to deha-
lomethanes. Proc Natl Acad Sci USA 90: 8576±8580
m
for gluta-
thione in glutathione transferases of the alpha, mu and pi
classes. Xenobiotica 23: 823±834
Meyer DJ, Ketterer B (1995) Puri®cation of soluble human gluta-
thione S-transferases. Methods Enzymol 252: 53±65
Meyer DJ, Coles B, Pemble SE, Gilmore KS, Fraser GM, Ketterer
B (1991) Theta, a new class of glutathione transferases puri®ed
from rat and man. Biochem J 274: 409±414
È
Thier R, Wiebel FA, Hinkel A, Burger A, Bruning T, Morgenroth
K, Senge T, Wilhelm M, Schulz TG (1998) Species di€erences in
the glutathione transferase GSTT1-1 activity towards the model
substrates methyl chloride and dichloromethane in liver and
kidney. Arch Toxicol 72: 622±629
Oda Y, Yamazaki H, Thier R, Ketterer B, Guengerich FP, Shi-
mada T (1996) A new Salmonella typhimurium NM5004 strain Urhahn T, Ballschmiter K (1998) chemistry of the biosynthesis of
expressing rat glutathione S-transferase 5-5: use in detection of
genotoxicity of dihaloalkanes using an SOS/umu test system.
Carcinogenesis 17: 297±302
halogenated methanes: C1-organohalogens as pre-industrial
chemical stressors in the environment? Chemosphere 37: 1017±
1032
O'Donovan MR, Mee MD (1993) Formaldehyde is a bacterial Vuilleumier S (1997) Bacterial glutathione S-transferases: what are
mutagen in a range of Salmonella and Escherichia indicator
strains. Mutagenesis 8: 577±581
they good for? J Bacteriol 79: 1431±1441
Vuilleumier S, Leisinger (1996) Protein engineering studies of di-
chloromethane dehalogenase/glutathione S-transferase from
Methylophilus sp. strain DM11. Ser12 but not Tyr6 is required
for enzyme activity. Eur J Biochem 239: 410±417
Osterman-Golkar S, Hussain S, Walles S, Anderstam B,
Sigvardsson K (1983) Chemical reactivity and mutagenicity of
some dihalomethanes. Chem-Biol interact 46: 121±130