Metabolism of tauromustine in liver and lung microsomes from various species

Article abstract of DOI:10.1080/004982599238236

1. The cytochrome P450 (CYP)-mediated metabolism of tauromustine has been evaluated in liver and lung microsomes from various species. Liver microsomes from rat pretreated with typical CYP inducers, human liver microsomes and cDNA-expressed human CYP enzymes were used to study the enzymatic basis of the metabolism. The further metabolism of the monodemethylated product of tauromustine and that of the denitrosated product were also investigated. 2. The major routes of tauromustine metabolism were demethylation to the alkylating active compound, R2, and denitrosation to the inactive metabolite, M3. The extent of metabolism and the activity of demethylation versus denitrosation varied among the species. The highest metabolism was found in mouse (BDF strain) followed by dog, rat and the human liver. Tauromustine was also metabolized to a low extent in lung microsomes from these species. 3. The further metabolism of R2 and M3 was ~ 100 times lower in activity than that of tauromustine. Both the demethylation and the denitrosation of tauromustine were increased 3-fold in liver microsomes from rat pretreated with phenobarbital, whereas treatment with cyanopregnenolone enhanced the denitrosation 11-fold, indicating the involvement of CYP3A. 4. Metabolism across a panel of 10 human liver microsomal samples demonstrated a correlation with testosterone 6β-hydroxylation of demethylation (r² = 0.86) and denitrosation of tauromustine (r² = 0.79). Among the human cDNA expressed CYP enzymes, not only was tauromustine determined to be catalysed predominantly by CYP3A4, but also to some extent by CYP2C19 and CYP2D6. 5. In conclusion, the present results indicate a major role of CYP3A enzymes in the metabolism of tauromustine.

Full text of DOI:10.1080/004982599238236

xenobiotica , 1999, vol. 29, no. 8, 783±792
Metabolism of tauromustine in liver and lung
microsomes from various species
H. TUVESSON‹Œ*, P. O. GUNNARSSON‹ and J. SEIDEGARDŒ
‹ Pharmacokinetics and Metabolism, Active Biotech Research AB, PO Box 724,
S-220 07 Lund, Sweden
ΠDepartment of Cell and Molecular Biology, Lund University, Solvegatan 21,
S-223 62 Lund, Sweden
Received 23 December 1998
1. The cytochrome P450 (CYP)-mediated metabolism of tauromustine has been
evaluated in liver and lung microsomes from various species. Liver microsomes from rat
pretreated with typical CYP inducers, human liver microsomes and cDNA-expressed
human CYP enzymes were used to study the enzymatic basis of the metabolism. The
further metabolism of the monodemethylated product of tauromustine and that of the
denitrosated product were also investigated.
2. The major routes of tauromustine metabolism were demethylation to the alkylating
active compound, R2, and denitrosation to the inactive metabolite, M3. The extent of
metabolism and the activity of demethylation versus denitrosation varied among the
species. The highest metabolism was found in mouse (BDF strain) followed by dog, rat and
the human liver. Tauromustine was also metabolized to a low extent in lung microsomes
from these species.
C
3. The further metabolism of R2 and M3 was 100 times lower in activity than that of
tauromustine. Both the demethylation and the denitrosation of tauromustine were
increased 3-fold in liver microsomes from rat pretreated with phenobarbital, whereas
treatment with cyanopregnenolone enhanced the denitrosation 11-fold, indicating the
involvement of CYP3A.
4. Metabolism across a panel of 10 human liver microsomal samples demonstrated a
correlation with testosterone 6b-hydroxylation of demethylation (r# 0.86) and de-
5
nitrosation of tauromustine (r# 0.79). Among the human cDNA expressed CYP enzymes,
not only was tauromustine determined to be catalysed predominantly by CYP3A4, but also
to some extent by CYP2C19 and CYP2D6.
5
5. In conclusion, the present results indicate a major role of CYP3A enzymes in the
metabolism of tauromustine.
Introduction
Chloroethylnitrosoureas have been useful in the treatment of a variety of
malignancies since the late 1960s (Mitchell and Schein 1986). The anti-tumour
eåects are generally considered to be exerted via alkylation and cross-linking of
DNA, a two-stage reaction which occurs when nitrosoureas are spontaneously
cleaved (Montgomery 1976, Kohn 1977). In addition to their non-enzymatic de-
nitrosation, and hence inactivation (Montgomery et al. 1975), nitrosoureas can be
enzymatically denitrosated, a reaction preferentially catalysed by cytochrome P450
(CYP) in the microsomal fraction of the liver (Potter and Reed 1982). It has also
been demonstrated that denitrosation of nitrosoureas can be catalysed by the
cytosolic fraction of the liver by glutathione transferase(s) (GST) (Smith et al.
1989).
* Author for correspondence at: Pharmacokinetics and Metabolism, Active Biotech Research AB,
PO Box 724, SE-220 07 Lund, Sweden; email: helen.tuvesson!activebiotech.com
}
’
http: www.tandf.co.uk JNLS xen.htm http: www.taylorandfrancis.com JNLS xen.htm
Xenobiotica ISSN 0049±8254 print ISSN 1366±5928 online
}} }}
1999 Taylor & Francis Ltd
}
}
}
}

784
H. Tuvesson et al.
"%
Figure 1. Structure of tauromustine, including the position of $H (E ) and C ( ), and the proposed
n
pathways of its metabolism.
Tauromustine, 1-(2-chloroethyl)-3-[2-(dimethylaminosulphonyl)ethyl]-1-ni-
trosourea, is a nitrosourea based upon the endogenous amino acid taurine with the
amino group nitrogen constituting one of the nitrogens of the nitrosourea (®gure 1).
Tauromustine shows a broad spectrum of anti-tumour activity in vitro and in vivo
(Hartley-Asp et al. 1988) and has been shown to be highly active against a panel of
murine nitrosourea-resistant tumours (Phillips et al. 1988, Pratesi and Savi 1989).
The toxicological pro®le of tauromustine has been shown to be similar to that of
other nitrosoureas with delayed haematological toxicity with thrombocytopenia
being the dose-limiting factor. Tauromustine has previously been clinically
evaluated in phase III studies (Smyth et al. 1995).
Tauromustine can be enzymatically degraded in vivo and in vitro. The
metabolism of tauromustine has been extensively studied in rat (Seidegard et al.
1990) and the proposed metabolic pathway is shown in ®gure 1. Two main metabolic
routes have been demonstrated for tauromustine. One route involves demethylation
to the monomethylated derivative R2, and further demethylation to the di-
demethylated derivative, R1. The other route involves denitrosation of tauro-
mustine, R2, and R1 to the corresponding denitrosated products M3, M2 and M1.
It has also been demonstrated that tauromustine can be denitrosated by GST in
liver cytosol from various species (Tuvesson et al. 1993).
Since the demethylated metabolites of tauromustine, R2 and R1, have been
shown to exert similar anti-tumour activity to tauromustine itself (Hartley-Asp et al.
1988, Matthew et al. 1989), a prolonged total alkylating eåect could be expected.
Thus, the extent of demethylation versus denitrosation might be an important
factor in determining the therapeutic as well as toxicological eåects of tauromustine.
In the present study the metabolism of tauromustine, and the further metabolism
of the metabolites of tauromustine, have been investigated in microsomal fractions
from liver and lungs from various species. The CYP enzymes responsible for the
metabolism of tauromustine in the microsomal fractions have also been evaluated.
Materials and methods
Radiochemicals
$H-tauromustine and "% C-tauromustine with a speci®c activity of 20.1 and 5.8 mCi mmol respectively
}
"
were synthesized at Pharmacia & Upjohn (Helsingborg, Sweden) with a radiochemical purity 95%.
Labelled R2 and M3 (®gure 1) were extracted from incubation mixtures containing rat liver microsomes
and $H-tauromustine. Brie¯y, this mixture was analysed by HPLC (see below) and the metabolites were
separated and collected in diåerent fractions. The fractions were then puri®ed by an additional separation
on the same HPLC-system.

Metabolism of tauromustine
785
Chemicals
Tauromustine, R2, R1 and M3 were all synthesized at Pharmacia & Upjohn with a chemical purity
95%. NADPH was obtained from the Sigma Chemical Co. (St Louis, MO, USA). Phenobarbital was
"
purchased from Apoteksbolaget (Stockholm, Sweden). All other chemicals were standard commercial
products of analytical grade.
Animals and human liver biopsies
Female and male mice of the Bom:NMRI (NMRI) strain, female mice of strains C57BL 6JBOM
}
(C57), DBA 2JBOM (DBA), B6D2F1 (BDF) and NMRI*, 8 weeks old, weighing 30 g, and Sprague-
}
Dawley rats (SpD), of both sexes, 6 weeks old, weighing 200 g, were used. Mice and rats were obtained
C
C
from Bomholtgaard (Ry, Denmark) except the NMRI* mice which were a gift from M. C. Bibby from
C
their colony at the University of Bradford, UK. Beagle dogs of each sex and
C
2 years old,
weighing 6±10 kg, were used. They were obtained from the Rahojden kennel (Orkelljunga, Sweden).
All animals were kept on a standard laboratory diet.
Human liver biopsies were obtained from the University Hospitals of Malmo and Lund, Sweden,
with approval from the local ethical committee in Malmo. The biopsies were stored at 70 C until used.
The human liver microsomes were prepared as described for the other species.
2
Ê
Microsomes
Rats were starved overnight before being killed by decapitation; the dogs were killed by an overdose
of mebumal. Livers and lungs were dissected and kept in ice-cold 0.25 m sucrose. The livers, the lungs
and the biopsies were homogenized and centrifuged at 105000 g and 4 C to isolate the microsomal
Ê
fraction. The microsomes were resuspended in ice-cold 0.25 m sucrose to a ®nal concentration of 2 g wet
tissue per 5 ml (Ernster et al. 1962). The protein content was assayed using bovine serum albumin as
standard (Lowry et al. 1951). Liver microsomes from male rats (SD-OFA) pretreated with CYP
}
prototype inducers (3-methylcholanthrene (3-MC, 20 mg kg day for 4 days), phenobarbital (PB,
50 mg kg day for 4 days), cyanopregnenolone (PCN, 300 mg kg day for 3 days) and clo®brate (CF,
}
}
}
}
}
}
}
500 mg kg day for 3 days)), were obtained from IFFA Credo (L’Albresle Cedex, France).
Human liver microsomes were also obtained from Pharmacia & Upjohn (Kalamazoo, MI, USA). The
speci®c content of the individual CYP enzyme was estimated from their catalytic activity toward a speci®c
substrate for each CYP. The following substrates and CYP were analysed; CYP1A2 by 7-ethoxyresoru®n
O-deethylation, CYP2A6 by coumarin 7-hydroxylation, CYP2C9 by tolbutamide hydroxylation,
CYP2D6 by dextromethorphan O-demethylation, CYP2E1 by p-nitrophenol hydroxylation and
CYP3A4 by testosterone 6b-hydroxylation. The cDNA-expressed human CYP enzymes were CYP1A2,
2C9, 2C19, 2D6 and 3A4, all obtained from Gentest (Woburn, MA, USA).
Assay conditions
Incubation of labelled tauromustine, R2, or M3 was performed at a substrate concentration of 1.9 mm
}
(Seidegard et al. 1990) and at a microsomal protein concentration of 0.5±7.5 mg ml in 50 mm phosphate
buåer of pH 7.4 at 37 C for 10 min. Preliminary experiments established the assay condition that a
Ê
reasonable substrate consumption (up to 10%), as well as time- and protein-linearity of metabolite
formation in the microsomal fraction was achieved. The reaction was started by the addition of NADPH
(1 mm) and the reaction was terminated after 10 min by adding an equal volume of ice-cold acetone,
followed by centrifugation at 10000g for 10 min. The supernatant was then mixed with saturated
amounts of non-radioactive reference standards, including tauromustine, R2 and R1 before being
analysed by HPLC. The non-enzymatic formation of metabolites from the nitrosoureas was subtracted
from the enzymatically formed metabolites to achieve the apparent speci®c activities. The `total’ non-
enzymatic degradation, including alkylating and carbamoylating moieties of tauromustine (Loftsson and
!
metabolites under the condition used in this study. Incubations with cDNA-expressed CYP were
Baldvinsdottir 1993), or its metabolites R1, R2 and R3, was
20% of the enzymatically formed
}
performed as described above, with a microsomal protein concentration of 5±13 mg ml, a CYP
}
concentration of 1±2 nmol ml and an incubation of 60 min.
HPLC analyses
Quanti®cation of tauromustine and its metabolites was obtained with an HPLC-system (Seidegard et
al. 1990). The samples were injected (10±100 ll) into a chromatographic system and separated on a radial
packed Nova-Pak C column, 4 lm (3.9 150 mm) with guard column Nova-Pak C (Waters
3
")
a linear gradient system from 4:1:95 to 70:30:0
Associates, Milford, ")MA, USA) and with
(acetonitrile:methanol:ammonium acetate buåer, pH 3.5). The analyses were performed at room
}
temperature at 0.55 ml min. Radioactivity was detected by a ¯ow detector (Flo-One b, Radiomatic
Instruments), with a 2.5 ml ¯ow cell. The mobile phase was mixed post-column with PicoFluor TM 40
}
(Packard Instrument Co., Meriden, CT, USA) at 1.9 ml min.

786
H. Tuvesson et al.
Results
Tauromustine was metabolized in microsomes prepared from the liver and lungs
of various species. Apart from the non-enzymatic degradation, tauromustine was
metabolized by two CYP-dependent metabolic pathways, namely demethylation
and denitrosation. The capacity to metabolize tauromustine, in all the diåerent
species, was higher in the liver than the lungs.
Metabolism of tauromustine in liver microsomes from various species
The diåerent speci®c activities of demethylation and denitrosation of tauro-
mustine in liver microsomes are presented in table 1. In general, the highest
demethylation and denitrosation activities were found in the various mouse strains.
The BDF mouse exhibited the highest activity, while the lowest activity was found
in the NMRI* mouse. The ratio between demethylation and denitrosation activities
C
was 1 in all the mouse strains with the exception of the NMRI* mouse where the
ratio was 2.3.
In rat, the activities of demethylation and denitrosation were slightly lower than
in mouse. The highest activity was the demethylation reaction in the male rat, and
the extent of demethylation versus denitrosation was 3.8 compared with 1.0 in the
female.
In liver microsomes from the beagle dog, the rate of demethylation of
tauromustine was considerably higher than that of denitrosation. Thus, the ratio
C
between demethylation and denitrosation was also considerably high, being 4.5.
The demethylation and denitrosation activities of tauromustine in liver
microsomes from humans were relatively low and the demethylation versus de-
nitrosation ratio was just over 1.0.
Metabolism of tauromustine in lung microsomes from various species
The metabolism of tauromustine in lung microsomes was qualitatively similar to
that in the liver microsomes, but quantitatively lower for both demethylation and
denitrosation. The various activities from diåerent mouse strains and the rat are
presented in table 2. The highest demethylation and denitrosation activity was
found in the BDF mouse. It had also the highest activity in the liver. The lowest
activity was in the NMRI* mouse, again correlating with the low activity in the liver
in this strain. Except for the NMRI strain, the denitrosation activity seemed
proportionately lower in the lung than the liver giving rise to a higher ratio between
"
C
demethylation and denitrosation: 3.5 in rat and 2±3 in the other mouse strains
examined.
Metabolism of R2 and M3
The primary metabolism of tauromustine is demethylation to R2 and de-
nitrosation to M3. Attempts to determine the further metabolism of R2 and M3 in
C
rat liver microsomes were performed and the demethylation of R2 to R1 was 140
times lower than the demethylation of tauromustine itself, while the denitrosation
C
of R2 to M2 was 15 times lower than the denitrosation of tauromustine to M3.
M3 was demethylated to M2 to a minor extent, being
C
100 times lower than the
!
demethylation of tauromustine corresponding to an activity 0.01 nmol min mg
}
}
protein.

Metabolism of tauromustine
787
Table 1. Speci®c activities of demethylation and denitrosation of tauromustine in liver microsomes
from various species. Values are the mean (range) of two to eight determinations. Activities are
}
}
expressed as nmol min mg protein. The ratio is the mean of the individual ratios between the
demethylation and denitrosation activity.
Demethylation
rate
Denitrosation
rate
Sexa
No.
Ratio
}
Species strain
Mouse
NMRI
NMRI
NMRI*
DBA
M
F
F
F
F
F
6
6
5
6
6
4
1.45 (1.07±1.66)
2.16 (1.93±2.47)
1.78 (0.89±2.64)
3.25 (2.41±3.77)
2.29 (1.25±3.45)
5.35 (2.57±7.52)
1.39 (1.12±1.56)
1.65 (1.41±2.10)
0.69 (0.40±1.27)
3.06 (2.73±3.31)
3.34 (2.87±3.80)
5.39 (3.06±7.34)
1.1
1.3
2.3
1.1
0.7
1.0
C57
BDF
Rat
SpD
SpD
M
F
8
8
1.95 (0.97±2.39)
0.29 (0.13±0.48)
0.51 (0.30±0.64)
0.38 (0.22±1.22)
3.8
1.0
Dog
Beagle
Beagle
M
F
2
2
2.23 (2.22±2.24)
2.22 (1.97±2.48)
0.47 (0.46±0.47)
0.49 (0.47±0.51)
4.8
4.5
Man
b
}
M F
7
0.65 (0.44- 1.20)
0.80 (0.27±1.14)
1.3
a F, females; M, males.
b
}
M F, mixed sexes.
Table 2. Speci®c activities of demethylation and denitrosation of tauromustine in lung microsomes
from various species. Values are the mean (range) of four to eight determinations. Speci®c
}
}
activities are expressed as nmol min mg protein. The ratio is the mean of the individual ratios
between the demethylation and denitrosation activity.
Demethylation
rate
Denitrosation
rate
Sexa
No.
Ratio
}
Species strain
Mouse
NMRI
NMRI
NMRI*
DBA
M
F
F
F
F
F
6
6
5
6
6
4
0.21 (0.17±1.269)
0.16 (0.06±0.24)
0.18 (0.08±0.28)
1.60 (1.05±1.93)
1.71 (1.49±1.93)
3.73 (3.09±4.36)
0.42 (0.35±0.52)
0.31 (0.24±0.39)
0.09 (0.04±0.13)
0.53 (0.31±0.67)
0.88 (0.82±0.94)
1.79 (1.77±1.80)
0.5
0.5
2.1
3.1
1.9
2.1
C57
BDF
Rat
SpD
SpD
M
F
8
8
0.66 (0.39±1.09)
0.59 (0.48±0.71)
0.23 (0.11±0.36)
0.13 (0.07±0.19)
3.6
4.9
a F, females; M, males.
Induction of diåerent CYP
The expression of diåerent forms of CYP can be relatively speci®cally enhanced
by pretreatment with various inducers. Pretreatment with PB, PCN or CF in the
SpD rat markedly increased the metabolism of tauromustine (®gure 2). In
microsomes from rat treated with PB, both the demethylation and the denitrosation
C
were increased 3-fold. Pretreatment with PCN or CF preferentially enhanced the
denitrosation of tauromustine, 11- and 6-fold respectively, whereas the de-
methylation remained unchanged. Pretreatment with 3-MC did not change the
metabolism of tauromustine to any measurable extent.

788
H. Tuvesson et al.
Figure 2. Eåect on the metabolism of tauromustine after pretreatment with various CYP inducers in
rat. Data are the average of two determinations and in percent of control activity. The rate of
}
demethylation and denitrosation in control microsomes was 19.2 and 5.9 nmol min mg
respectively. The ratios, demethylation denitrosation, are 3.3, 1.7, 2.7, 0.2, 0.7 respectively.
}
}
Control, no inducer; 3-MC, 3-methylcholanthrene; PB, phenobarbital; PCN, cyanopregneno-
lone; CF, clo®brate.
Figure 3. Correlation of activities of demethylation and denitrosation of tauromustine with
testosterone 6b-hydroxylation in 10 diåerent human liver microsomal preparations.
Metabolism of tauromustine in characterized human liver microsomes
The metabolism of tauromustine was determined in a panel of microsomal
preparations from 10 diåerent human livers. The liver microsomes were charac-
terized with respect to diåerent CYP activities, as exempli®ed by CYP3A in ®gure
3. Regression analysis of the formation of the demethylated and denitrosated
products of tauromustine and various CYP reactions are expressed as coeæcients of
correlation (r# ; table 3). The highest correlations were for the demethylation and de-
nitrosation of tauromustine and 6b-hydroxylation of testosterone r# 0.86 and 0.79
5
respectively. The denitrosation of tauromustine was furthermore well correlated to
7-hydroxylation of coumarine (CYP2A6), r# 0.74. Correlations between 0.45 and
5

Metabolism of tauromustine
789
Table 3. Correlation of tauromustine metabolite (M3, R2) formation with speci®c CYP activities, and
correlation of speci®c CYP activities against each other, across a panel of human liver microsomes
(n 10).
5
Correlation (r#)
Marker activity
CYP M3
R2
1A2 2A6 2C9 2C19 2D6 2E1 3A4
1.00
7-Ethoxyresoru®n O-dealkylation
Coumarin 7-hydroxylation
1A2 0.19 0.45
2A6 0.74a 0.60a 0.01 1.00
Tolbutamide methyl-hydroxylation 2C9 0.25 0.50
S-mephenytoin 4-hydroxylation 2C19 0.05 0.05
0.18 0.35 1.00
0.03 0.07 0.01 1.00
Dextromethorphan O-demethylation 2D6 0.50 0.70a 0.25 0.26 0.21 0.12 1.00
p-Nitrophenol hydroxylation
Testosterone 6b-hydroxylation
2E1 0.01 0.10
0.01 0.12 0.32 0.09 0.01 1.00
3A4 0.79a 0.86a 0.27 0.62 0.53 0.00 0.58 0.04 1.00
a
!
0.0001.
p
Table 4. Metabolism of tauromustine by cDNA-expressed human CYP.
Demethylation rate Denitrosation rate
}
}
}
}
nmol min pmol
P450
nmol min pmol
P450
}
nmol min mg
}
}
nmol min mg
}
CYP
1A2
2C9
2C19
2D6
3A4
0.3
0.2
3.7
0.3
1.4
1.9
1.1
16.3
1.5
3.1
0.1
0.9
1.8
4.6
19.0
0.7
4.1
8.3
6.7
22.1
0.60 were found when the metabolism of tauromustine was compared with N-
demethylation of dextromethorphan (CYP2D6). However, when the activities for
the speci®c CYP enzymes were compared with each other, correlations were found
between CYP3A4 and CYP2A6 (r# 0.62), and CYP2D6 (r# 0.58) as well as for
5
5
CYP2C9 (r# 0.53). Therefore, the contribution of CYP2A6 and CYP2D6 to the
metabolism of tauromustine cannot be suæciently evaluated from these data.
5
Metabolism of tauromustine by diåerent cDNA-expressed human CYP
Tauromustine was incubated with the cDNA-expressed human CYP enzymes,
CYP1A2, 2C9, 2C19, 2D6 and 3A4 (table 4). Both the demethylation and de-
nitrosation of tauromustine were observed to be catalysed by CYP3A4. The de-
nitrosation of tauromustine was also catalysed by CYP1A2 and to some extent by
CYP2D6. The demethylation of tauromustine was also seen in microsomes from
cells expressing CYP2C19.
Discussion
The present study demonstrates a qualitatively similar metabolism of tauro-
mustine in each of the various species investigated, with a similar metabolite pattern
found throughout. The major routes of metabolism were demethylation, principally
to the monomethylated product R2, which exerts alkylating activity, and de-
nitrosation to the inactive M3. Tauromustine was metabolized in a qualitatively
similar manner in both liver and lung microsomes, but with a rate of demethylation

790
H. Tuvesson et al.
and denitrosation that was lower in lung than the corresponding activities in the
liver. The results con®rm earlier studies in rat (Seidegard et al. 1990) that
demonstrated that tauromustine is metabolized in vitro via CYP enzymes in the liver
and to a lesser extent in the lung, in addition to its non-enzymatic degradation.
The extent of metabolism and the balance between demethylation and de-
nitrosation varied among the mouse strains and between the diåerent species. The
ratio was relatively high in the NMRI* mouse, the male SpD rat and the beagle dog.
This may be of great importance to the pharmacological response as well as to the
toxicological eåect of tauromustine. Tumour-bearing BDF and NMRI* mice have
been used in pharmacological studies to investigate the eåect of drug scheduling on
the anti-tumour eåect of tauromustine (Matthew et al. 1994). A therapeutic
advantage of divided doses of tauromustine was found in the BDF mouse that could
be due to a higher level of demethylation activity in this strain, following a prolonged
alkylating eåect. The maximum tolerated single oral dose of tauromustine is
C
}
}
60 mg kg in the BDF mouse and 40 mg kg in the NMRI* and NMRI mice, and
SpD rat (Bibby et al. 1988, Matthew et al. 1994). Thus, the higher activity of de-
nitrosation and hence detoxi®cation of tauromustine in the BDF mouse may account
for the increased tolerable oral dose in these animals. In the beagle dog no such dose
has been established but the toxicological eåects of tauromustine have been studied
}
after repeated doses to the SpD rat (up to 10 mg kg) and beagle dog (up to
1.5 mg kg), and it was found that the dog was even more sensitive to tauromustine
}
than the rat (Beryl Hartley-Asp, Pharmacia & Upjohn, personal communication).
This might be explained by the comparatively low denitrosation activity in the
microsomal fraction but also by a 10 times lower GST-mediated denitrosation
activity in the liver cytosol in the beagle dog (Tuvesson et al. 1993). In man, a
}
maximal tolerable oral single dose of 3.9 mg kg has been de®ned for use in clinical
studies (Smyth et al. 1995, Hartley-Asp et al. 1996).
The diåerent inducers used in this study, 3-MC, PB, PCN and CF, are typical
inducers of diåerent CYP (Soucek and Gut 1992). For example, in rat pretreatment
with 3-MC will increase the activity of CYP1A whereas PB preferentially increases
the activity of a number of subfamilies within the CYP2 and CYP3A family.
Pretreatment with PCN is known to increase the activity of CYP3A1 in rat. Both
C
microsomes from rat pretreated with PB. The denitrosation of tauromustine was
demethylation and denitrosation of tauromustine were increased
3-fold in
C
increased
11- and 6-fold after PCN and CF pretreatment respectively, whereas
the demethylation activity remained unchanged. No increase in the metabolism of
tauromustine was seen after pretreatment with 3-MC. These results indicate that
the CYP2 and CYP3 families are involved in the metabolism of tauromustine, at
least in rat. Interestingly, the denitrosation of tauromustine was induced in liver
microsomes from the CF pretreated rat. However, a possible involvement of
CYP4A in the de-nitrosation of tauromustine needs to be further evaluated.
For characterization of the speci®c CYP enzyme(s) responsible for the metab-
olism in man, 10 diåerent characterized human liver microsomal preparations were
used. The enzymes were characterized for seven diåerent CYP activities. By
regression analysis between these activities and the demethylation and de-
nitrosation of tauromustine, it was found that 6b-hydroxylation of testosterone was
best correlated to the metabolism of tauromustine, implicating CYP3A activity.
Moderately good correlations were also found for the CYP2A6 and CYP2D6
activities, but in the diåerent human microsomal fractions used in this study an

Metabolism of tauromustine
791
equally good correlation was also found between these activities and CYP3A. Thus,
data from this experiment indicate that CYP3A may be involved in the metabolism
of tauromustine in man. For further characterization of the responsible CYP
enzyme(s) ®ve diåerent cDNA-expressed CYP, 1A2, 2C9, 2C19, 2D6 and 3A4,
were examined for activity towards tauromustine as substrate. The highest activity
for both of the metabolic reactions of tauromustine was exhibited by CYP3A4. The
other CYP forms tested, 1A2, 2D6 and 2C19, could also be assigned some metabolic
activity towards tauromustine. However, in view of their relatively low expression
compared with CYP3A4, their contribution in vivo would probably be of lesser
signi®cance. This is also in agreement with the results above obtained with the
human liver microsomes where a mixture of CYP is expressed.
The results from this study demonstrate that further metabolism of the
demethylated product R2 and the denitrosated product M3 is up to 140 times
lower than that of tauromustine, probably due to a much lower aænity for those
metabolites for CYP3A enzyme(s).
The demethylation of tauromustine to two active and metabolically more stable
metabolites, R2 and R1, might account for prolonged alkylating activity. In a clinical
phase III study (Smyth et al. 1995), high plasma concentrations of tauromustine
have been determined after oral administration of tauromustine in patients, which
indicates that tauromustine possesses pharmacokinetics diåerent from those of
other nitrosoureas. However, the relatively complicated metabolism of tauromustine
C
and the fact that CYP3A4 is highly variably ( 20-fold) expressed in man (Watkins
1994) suggests that the metabolism of tauromustine needs to be further evaluated in
patients.
In conclusion, tauromustine was metabolized in a qualitatively similar manner in
the animals investigated and to a higher extent in liver microsomes compared with
lung microsomes. Tauromustine possesses two major primary metabolic pathways,
demethylation and denitrosation. Further metabolism of the primary metabolites
could also be demonstrated but to a much lower extent than that of the parent
compound. The major CYP enzyme responsible for both metabolic pathways is
CYP3A in both animals and man. In man, CYP2D6, 1A2 and to some extent 2C19
may play a role but probably of little or no signi®cance in vivo.
Acknowledgements
The authors thank Lillemor Halvarsson and Ulrika Larsson for excellent
technical assistance.
References
Bibby , M. C., Double , J. A. and Morris , C. M., 1988, Anti-tumour activity of TCNU in a panel of
transplantable murine colon tumours. European Journal of Cancer and Clinical Oncology, 24,
1361±1364.
Ernster , L., Siekevitz , P. and Palade, G. E., 1962, Enzyme±structure relationships in the endoplasmic
reticulum of rat liver. Journal of Cell Biology, 15, 541±562.
Hartley -Asp, B., Christensson , P. I., Gunnarsson , K., Gunnarsson , P. O., Jensen , G., Polacek, J.
and Stamvik , A., 1988, Anti-tumour, toxicological and pharmacokinetic properties of a novel
taurin-based nitrosourea (TCNU). Investigational New Drugs, 6, 19±30.
Hartley -Asp, B., Hansson , K., Vibe -Petersen , J., Hansen , H. H., Polacek, J., Ellman , M. and
Gunnarsson , P. O., 1996, Chromosome aberrations and pharmacokinetics in patients receiving
tauromustine as either a single or a repeated dose. Cancer Chemotherapy and Pharmacology, 38,
309±316.

792
Metabolism of tauromustine
Kohn , K. W., 1977, Interstrand cross-linking of DNA by 1,3-bis(2-chloro-ethyl)-1-nitrosourea and
other 1-(2-haloethyl)-1-nitrosoureas. Cancer Research, 37, 1450±1454.
Loftsson , T. and Baldvinsdo ttir , J., 1992, Degradation of tauromustine (TCNU) in aqueous solutions
Acta Pharmacologica Nordavica, 4, 129±132.
Lowry, O. H., Rosenbrough , N. J., Farr, A. L. and Randall , R. L., 1951, Protein measurement with
the Folin phenol reagent. Journal of Biological Chemistry, 193, 265±275.
Matthew , A., Bibby , M. C., Double , J. A. and Crawford, S. M., 1989, Anti-tumour activity and bone
marrow toxicity of the TCNU metabolites LS2724 and LS 2715. Investigational New Drugs, 7,
405.
Matthew , A. M., Christensson , P. I., Seidega rd, J., Tuvesson , H., Bibby , M. C., Hartley-Asp, B.
and Creaven, P. J., 1994, Eåect of drug scheduling on the anti-tumour activity and bone marrow
toxicity of tauromustine (TCNU). International Journal of Oncology, 4, 403±409.
Mitchell , E. P. and Schein , P. S., 1986, Contributions of nitrosoureas to cancer treatment. Cancer
Treatment Reports, 70, 31±41.
Montgomery , J. A., James , R., McCaleb, G. S., Kirk, M. C. and Johnston , T. P., 1975, De-
composition of N-(2-chloroethyl)-N-nitrosoureas in aqueous media. Journal of Medicinal
Chemistry, 18, 568±571.
Montgomery , J. A., 1976, Chemistry and structure±activity studies of the nitrosoureas. Cancer
Treatment Reports, 60, 651±654.
Phillips , R. M., Bibby , M. C. and Double , J. A., 1988, In vitro and in vivo responses of a panel of
murine colon tumours to tauromustine: a positive correlation. European Journal of Cancer and
Clinical Oncology, 24, 1365±1371.
Potter , D. W. and Reed, D. J., 1982, Denitrosation of carcinostatic nitrosoureas by puri®ed NADPH
cytochrome P-450 reductase and rat liver microsomes to yield nitric oxide under anaerobic
conditions. Archives of Biochemistry and Biophysics, 216, 158±169.
Pratesi , G. and Savi, G., 1989, Eåects of tauromustine, a water soluble nitrosourea compound, on
NMU-1 murine lung tumor. Oncology, 46, 132±135.
Seidega rd, J., Gronquist , L. and Gunnarsson , P. O., 1990, Metabolism of a novel nitrosourea,
tauromustine, in the rat. Biochemical Pharmacology, 39, 1431±1436.
Smith , M. T., Evans, C. G., Doane-Setzer, P., Castro, V. M., Tahir, M. K. and Mannervik , B.,
1989, Denitrosation of 1,3-bis(2-chloroethyl)-1-nitrosourea by class mu glutathione transferases
and its role in cellular resistance in rat brain tumour cells. Cancer Research, 49, 2621±2625.
Smyth , J. E., Hardcastle, J. D., Denton , G., Alderson , D., Grace, R. H., Mansi , J. L., Yosef , H.
M. A., Nordle, O., Lauri, H. and Wahlby , S., 1995, Two phase III trials of tauromustine
(TCNU) in advanced colorectal cancer. Annals of Oncology, 6, 948±949.
Soucek, P. and Gut, I., 1992, Cytochromes P-450 in rats: structures, functions, properties and relevant
human forms. Xenobiotica, 22, 83±103.
Tuvesson , H., Gunnarsson , P. O. and Seidega rd, J., 1993, Measurement and characterisation of the
denitrosation of tauromustine in liver cytosol from various species. Carcinogenesis, 14, 1143±1147.
Watkins , P. B., 1994, Noninvasive tests of CYP3A enzymes. Pharmacogenetics, 4, 171±184.