In vitro metabolism of two heterocyclic amines, 2-amino-9H-pyrido[2,3-b]indole (AαC) and 2-amino-3-methyl-9H-pyrido[2,3-b]indole (MeAαC) in human and rat hepatic microsomes

Article abstract of DOI:10.1034/j.1600-0773.2002.900303.x

2-Amino-9H-pyrido[2,3-b]indole (AαC) and 2-amino-3-methyl-9H-pyrido[2,3-b]indole (MeAαC) are two mutagenic and carcinogenic heterocyclic amines formed during ordinary cooking. In this study, we have investigated the in vitro metabolism of tritium-labelled AαC and MeAαC in hepatic microsomes from human pools, rats induced with polychlorinated biphenyl (PCB) (Aroclor 1254) and control rats. The microsomes were incubated with AαC and MeAαC and the detoxified and activated metabolites of AαC and MeAαC were separated and characterised by HPLC-MS. AαC is metabolised to two major and three minor detoxified metabolites, while MeAαC is metabolised to three major and one minor detoxified metabolites. Some AαC and MeAαC are activated by oxidation to the reactive metabolites N²-OH-AαC and N²-OH-MeAαC, respectively. These reactive N²-OH-metabolites react partially in the incubation system with formation of protein adducts, dimers and the parent compound by reduction of the N²-OH-metabolites. The distribution between the detoxified and activated metabolites in the different types of hepatic microsomes showed same pattern for both AαC and MeAαC. In PCB-induced rat microsomes, the major part of the metabolites are detoxified, only a little amount is activated. In control rat microsomes there is a fifty-fifty distribution between detoxification and activation, while the major part of the metabolites from the human microsomes are activated and reacts to form dimers and protein adducts. These data show that, in human hepatic microsomes compared to rat hepatic microsomes, a major part of AαC and MeAαC are metabolically activated to the reactive N²-OH-AαC and N²-OH-MeAαC.

Full text of DOI:10.1034/j.1600-0773.2002.900303.x

C Pharmacology & Toxicology 2002, 90, 127–134.
Copyright C
Printed in Denmark . All rights reserved
ISSN 0901-9928
In vitro Metabolism of Two Heterocyclic Amines, 2-Amino-
H-pyrido[2,3-b]indole (AaC) and 2-Amino-3-methyl-9H-
9
pyrido[2,3-b]indole (MeAaC) in Human and Rat Hepatic
Microsomes
Hanne Frederiksen and Henrik Frandsen
Institute of Food Safety and Toxicology, Danish Veterinary and Food Administration, Mørkhøj Bygade 19,
DK-2860 Søborg, Denmark
(
Received July 10, 2001; Accepted October 10, 2001)
Abstract: 2-Amino-9H-pyrido[2,3-b]indole (AaC) and 2-amino-3-methyl-9H-pyrido[2,3-b]indole (MeAaC) are two muta-
genic and carcinogenic heterocyclic amines formed during ordinary cooking. In this study, we have investigated the in
vitro metabolism of tritium-labelled AaC and MeAaC in hepatic microsomes from human pools, rats induced with
polychlorinated biphenyl (PCB) (Aroclor 1254) and control rats. The microsomes were incubated with AaC and MeAaC
and the detoxified and activated metabolites of AaC and MeAaC were separated and characterised by HPLC-MS. AaC
is metabolised to two major and three minor detoxified metabolites, while MeAaC is metabolised to three major and one
2
minor detoxified metabolites. Some AaC and MeAaC are activated by oxidation to the reactive metabolites N -OH-AaC
2
2
and N -OH-MeAaC, respectively. These reactive N -OH-metabolites react partially in the incubation system with forma-
2
tion of protein adducts, dimers and the parent compound by reduction of the N -OH-metabolites. The distribution
between the detoxified and activated metabolites in the different types of hepatic microsomes showed same pattern for
both AaC and MeAaC. In PCB-induced rat microsomes, the major part of the metabolites are detoxified, only a little
amount is activated. In control rat microsomes there is a fifty-fifty distribution between detoxification and activation,
while the major part of the metabolites from the human microsomes are activated and reacts to form dimers and protein
adducts. These data show that, in human hepatic microsomes compared to rat hepatic microsomes, a major part of AaC
2
2
and MeAaC are metabolically activated to the reactive N -OH-AaC and N -OH-MeAaC.
2-Amino-9H-pyrido[2,3-b]indole (AaC) and its methyl
compounds are often ring-hydroxylated followed by phase
homologue, 2-amino-3-methyl-9H-pyrido[2,3-b]indole (Me-
AaC), are two food-borne mutagenic and carcinogenic het-
erocyclic amines. AaC and MeAaC are often referred to as
a-carbolines (fig. 1). AaC and MeAaC are formed as pyrol-
ysis products of tryptophan-rich compounds and are also
found in cooked food such as meat, chicken and fish
II conjugation. Activated compounds are hydroxylated in
their characteristic exocyclic amino group, usually followed
by O-esterification catalysed by e.g. acetyltransferase or sul-
fotransferase (Eisenbrand & Tang 1993). Especially en-
zymes from the CYP1A-family are involved in the phase I
2
metabolism of AaC and MeAaC to their corresponding N -
(
Gross & Gruter 1992; Holder et al. 1997; Skog et al. 1998;
Solyakov et al. 1999) and in cigarette smoke condensates
Yoshida & Matsumoto 1980; Matsumoto et al. 1981; Man-
OH-derivates (Niwa et al. 1982; Sugimura 1985). Activated
heterocyclic amines are able to form adducts with macro-
3
2
(
molecules such as protein and DNA. It is shown by P-
abe et al. 1990; Wakabayashi et al. 1995). MeAaC is also
found in wine (Richling et al. 1997). AaC and MeAaC are
mutagenic in bacterial test systems (Nagao et al. 1983; Hol-
me et al. 1989; Wild et al. 1991). Dietary administration of
AaC and MeAaC to mice and rats induced liver tumours
post-labeling analysis that MeAaC and AaC form one
major DNA adduct (N -deoxyguanin-8-yl-(Me)AaC) in
2
primary hepatocytes from rats (Pfau et al. 1996 & 1997).
There are only few reports of the metabolism of AaC and
MeAaC compared with other heterocyclic amines. Recently
some of the major metabolites of AaC in human and rodent
hepatic microsomes were characterized by Raza et al. (1996)
and King et al. (2000). Previously, the metabolism of
MeAaC in hepatic microsomes from PCB-induced rat was
studied (Frandsen et al. 1998), but not the metabolism of
MeAaC in human microsomes.
(Ohgaki et al. 1984; Hasegawa 1992). AaC and MeAaC
have, like other heterocyclic amines, two metabolic path-
ways, detoxification or activation. The first step in the met-
abolism of the heterocyclic amines is a phase I hydroxyl-
ation catalysed by cytochrome P450 enzymes. Detoxified
The aim of this study was to investigate the in vitro met-
abolism of AaC and MeAaC in hepatic microsomes from
rat and man. In the present study we have investigated the
in vitro metabolism of AaC and MeAaC in hepatic micro-
Author for correspondence: Henrik Frandsen, Institute of Food
Safety and Toxicology, Danish Veterinary and Food Administra-
tion, Mørkhøj Bygade 19, DK-2860 Søborg, Denmark (fax π45 33
9
5 60 01, e-mail hf/fdir.dk).

1
28
HANNE FREDERIKSEN AND HENRIK FRANDSEN
and stored at –20æ. The synthesis product was diluted in 1% for-
somes from polychlorinated biphenyl (PCB)-induced rats,
control rats and two different human pools and compared
the distribution between activation and detoxification of the
metabolites. N -OH-derivaties of both AaC and MeAaC
were synthesized for comparison with metabolites of the
two heterocyclic amines.
mic acid in 10% acetonitrile and analysed by HPLC-MS. The
2
yield of N -OH-MeAaC was 192 mg (44%).
2
Synthesis of 2-nitro-9H-pyrido[2,3-b]indole (NO
AaC was synthesised by the same method as the synthesis of N -
2 2
-AaC). NO -
2
MeAaC. 3.6 mg AaC was used and the yield of NO
.6 mg (38%).
2
-AaC was
1
2
Materials and Methods
Synthesis of 2-hydroxyamino-9H-pyrido[2,3-b]indole (N -OH-
2
AaC). N -OH-AaC was synthesised by the same method as the
3
Chemicals. AaC, [ H]-AaC and MeAaC were obtained from To-
2
synthesis of N -OH-MeAaC. 0.4 mg NO
2
-AaC was used and the
ronto Research Chemicals (Toronto, Canada). Tritiation of
MeAaC is described previously (Frandsen et al. 1998). Pooled
human liver microsomes (from 15 donors) were obtained from in
vitro Technologies (Baltimore, MD, USA) and pooled human
liver microsomes (from 11 donors) was obtained from Gentest
2
yield of N -OH-AaC was 110 mg (59%).
2
Synthesis of MeAaC- and AaC-dimers. N -OH-MeAaC dissolved
in dimethyl formamid (DMF) was diluted to 50 mg/ml with ar-
gon-purged water. To the solution was added 5 ml acetic anhy-
dride under a stream of argon. The reaction mixture was kept
under argon and stirred at room temperature for 3 hr. MeAaC
treated in the same way was used as reference. The syntheses
products were diluted in argon-purged 1% formic acid in 10%
acetonitrile and analysed by HPLC-MS. The same procedure was
used to create AaC-dimers.
(
1
Woburn, MA, USA). Polychlorinated biphenyl (PCB: Aroclor
254) was obtained from Monsanto Industrial Chemical Co. (St.
Louis, MO, USA). Isocitricdehydrogenase (from pork heart), ni-
cotinamide adeninedinucleotide-phosphate and 3-[N-morpholino]
propanesulfonic acid hemi sodium salt (MOPS) were obtained
from Sigma Chemical Co. (St. Louis, MO, USA). HPLC grade
acetonitrile and formic acid (distillated before use) were obtained
from Romil (Cambridge, UK). Chromabond columns (C18ec and
2
OH) were obtained from Macherey-Nagel (Duren, Germany).
Soluene-350 and Hionic-Flour were obtained from Packhard
Meriden, CT, USA). All other chemicals were of analytical pu-
rity and obtained from Riedel-de Ha e¨ n (Seelze, Germany), Merck
Darmstadt, Germany), Fisher Scientific (Loughborough, UK)
Preparation of rat hepatic microsomes. Adult male Wistar rats (age
7–8 weeks, weight ∂ 200 g) were delivered from Møllegaard Breed-
ing Center (Lille Skensved, Denmark). Microsomes from PCB-in-
duced rats and control rats were prepared as previously described
(Frandsen et al. 1994). Four rats were used to prepare a pool of
PCB-induced hepatic rat microsomes. PCB (Aroclor 1254, 500 mg/
kg, dissolved in corn oil) was injected intraperitoneally 5 days be-
fore sacrifice. Two rats were used to prepare a pool of control he-
patic rat microsomes.
(
(
and Rathburn (Walkerburn, Scotland, UK).
2
Synthesis of 2-nitro-3-methyl-9H-pyrido[2,3-b]indole (NO -
MeAaC). A previously described method (Frandsen et al. 1998)
was modified in the following way: A sample of 3.4 mg MeAaC
was dissolved in 1 ml acetic acid and heated to 90æ. One hundred
ml of hydrogen peroxide (30%) was added drop-wise under stir-
ring. Heating and stirring were continued for 30 min. The reac-
tion mixture was diluted with 15 ml water and applied in small
portions onto an activated Chromabond C18ec column. The col-
umn was washed with 5 ml water and eluted with 2 ml methanol.
The eluate was placed in a 50æ water bath and evaporated to
dryness under a stream of nitrogen. The residue was dissolved in
In vitro metabolism. The microsomal incubations were performed as
previously described (Frandsen et al. 1998). Five mg test substance
3
3
(MeAaC, [ H]-MeAaC, AaC and [ H]-AaC dissolved in DMF) per
ml incubation mixture was added after preincubation for 5 min.
The reactions were terminated after 30 min. for the MeAaC incuba-
tions and after 45 min. for the AaC incubations by addition of one
volume of ice-cold argon-purged ethanol. After centrifugation the
supernatants were isolated and analysed by HPLC-MS.
2
ml of ethyl acetate and added 6 ml of heptane. The dissolved
part of the residue was applied onto a Chromabond 2OH column
equilibrated with heptane. The eluate was collected and the col-
umn was eluted with 2 ml ethyl acetate/heptane (1:1). The com-
bined eluate was evaporated as described above, redissolved in 1
ml dimethyl sulfoxide (DMSO) and stored at 4æ. The product was
diluted in 1% formic acid in 10% acetonitrile and analysed by
high performance liquid chromatography-mass spectrometry
Analytical. High performance liquid chromatography analyses were
performed on a Agilent Technologies model 1100 liquid chromato-
graph equipped with a photodiode array detector (Agilent Techno-
logies, Wallbronn,Germany). The products were separated on a
Zorbax SB-C3, 5 mm, 150¿3 mm column from Agilent Technolo-
gies. The flow rate was 0.4 ml/min. and the oven temperature was
40æ. Solvent were A: 1% formic acid and B: acetonitrile. Solvent
(
2
HPLC-MS). The yield of NO -MeAaC was 1.6 mg (41%).
programming was: 0–2 min., 2% B; 10 min., 25% B; 20 min., 80%
B; 22 min., 2% B.
Synthesis
(
2
of
N -OH-MeAaC). A previously described method (Agus et al.
000) was modified in the following way: All solvents were
2-hydroxyamino-3-methyl-9H-pyrido[2,3-b]indole
Positive ion electrospray mass spectra were obtained with an
Agilent Technologies MSD 1100 mass spectrometer equipped with
an electro-spray interface (Agilent Technologies, Wallbronn, Ger-
many). The following interphase settings were used: nebulizer press-
ure 60 psi; drying gas(nitrogen) 10 l/min., 350æ; capillary voltage
4000 V; fragmentor voltage 70 V.
2
purged with argon before use and all solutions were stored under
argon. Two hundred ml water was added to 1 ml 0.2 M ascorbic
acid in DMSO. The solution was stirred at room temperature for
5
min. Forty ml 0.3 M copper sulphate was added drop wise
under stirring. After 5 min. with stirring, 0.5 ml of 0.94 mg/ml
NO -MeAaC in DMSO was added drop-wise (30 sec.) under stir-
ring. After 45 sec. the reaction was stopped by addition of 2 ml
.1 M EDTA followed by 8 ml ice-cold water under stirring. The
When the incubations were conducted with tritiated materials the
supernatants were analysed by HPLC and 1 min. fractions were
collected (the same HPLC-conditions were used to analyse both
tritiated and not-tritiated materials). Four ml Hionic-Flour were
added to the collected fractions followed by liquid scintillation
counting. The precipitates from the incubations with tritiated ma-
terials were washed twice with acetone and dissolved in Soluene-
350. One hundred ml of the solutions were added 4 ml of Hionic-
Flour and analysed by liquid scintillation counting. Liquid scintil-
lation counting was performed on a Tri Carb 3100TR using a Hion-
2
0
reaction mixture was kept on ice until purification. The reaction
mixture was under argon applied in small portions on an acti-
vated Chromabond C18ec column. The column was washed with
4
ml ice-cold water, 2 ml ice-cold 10% methanol and 2 ml ice-
2
cold water. N -OH-MeAaC was eluted with 1.2 ml ice-cold DMF

IN VITRO METABOLISM OF TWO HETEROCYCLIC AMINES, AaC AND MeAaC
129
ic-Flour scintillation cocktail and with external standardization
Similar metabolite patterns of MeAaC are found in in-
cubation with control rat and human microsomes (fig. 2B
and 2C). However, the amounts of the late eluating metabo-
lites 4 and 5 in control rat and human microsomes are con-
siderable larger than the amount of metabolites 4 and 5 in
PCB-induced rat microsomes.
Fig. 2D, E and F shows the metabolism of AaC in he-
patic microsomes from PCB-induced rats, control rats and
man. Also here a similar metabolite pattern in the three
different types of microsomes is obtained. In PCB-induced
rat and control rat microsomes AaC are metabolised to
three major metabolites (labelled 1–3) and two minor meta-
bolites (labelled with arrows) eluting before the parent com-
pound, AaC. In human hepatic microsomes it is primarily
metabolite 1 and 3 that are dominating, but also metabolite
(
Packhard, Meriden, CT, USA).
All incubations were repeated three times with similar results.
Results
Metabolism of MeAaC and AaC. Several metabolites of
MeAaC and AaC were detected in microsomes from PCB-
induced rats, control rats and humans after incubation with
MeAaC and AaC. Fig. 2 shows the HPLC profiles of
MeAaC and AaC and their metabolites from the micro-
somal incubations. Two different pools of human micro-
somes were used, but there were no particular difference
between the metabolism of MeAaC and AaC in the two
human hepatic microsomal pools.
In PCB-induced hepatic rat microsomes MeAaC is
metabolised to three major metabolites (labelled 1–3) and
one minor metabolite (labelled with arrow) eluting before
the parent compound, MeAaC (fig. 2A). The mass spectra
2
is noticeable. The mass spectra of all the early eluting
π
metabolites showed molecular ions [MπH] at m/z 200, in-
dicating that they are hydroxylated metabolites. Raza et al.
(
1996) have characterized the major metabolites of AaC in
of all the early eluting metabolites showed molecular ions
π
human microsome incubation as 3-OH-AaC and 6-OH-
AaC. UV spectra of metabolite 1 and 3 were identical with
the published UV spectra of 6-OH-AaC and 3-OH-AaC,
respectively (data not shown). In the three chromatograms
[
MπH] at m/z 214, indicating that they are hydroxylated
metabolites. An earlier identification of the three major
1
metabolites by electro spray mass spectrometry and H-
NMR has shown, that metabolite 1, 2 and 3 can be char-
(
fig. 2D, E and F) two other metabolites (labelled 4 and 5)
acterized as 6-OH-MeAaC, 3-CH OH-AaC and 7-OH-
2
are eluting after AaC. The mass spectra of the late eluting
MeAaC, respectively (Frandsen et al. 1998). UV spectra of
the minor metabolite were nearly identical with the UV
spectra of 6-OH-MeAaC and 7-OH-MeAaC (data not
shown), therefore it is proposed that the minor metabolite
could be another ring-hydroxylated metabolite. The tree
early eluting major metabolites are all characterized as de-
toxified compounds (Frandsen et al. 1998). Small amounts
of two other metabolites (labelled 4 and 5) are eluting after
MeAaC. The mass spectra of the late eluting metabolites
π
metabolites showed molecular ions [MπH] at m/z 365,
both with daughter ions [MπH]π at m/z 183, indicating
that dimers of a reactive metabolites is formed. The last
metabolite (labelled 6), eluting right before the parent com-
pound, AaC, was found in the human microsomal incuba-
tions (fig. 2F). A small amount of metabolite 6 was also
detectable in the preparation with PCB-induced rat micro-
somes by analysis of the MS data using extracted ion chro-
matogram (data not shown). Metabolite 6 has a molecular
ion [MπH]π at m/z 200 with a daughter ion [MπH]π at m/
z 183. This metabolite was much more unstable, when
stored in air at room temperature, than the other metabo-
lites (labelled 1–5), which could indicate that this metabolite
π
showed molecular ions [MπH] at m/z 393, both with
π
daughter ions [MπH] at m/z 197. This high molecular
weight indicates that dimers of a reactive metabolite are
formed. The presence of a reactive metabolite was not de-
tected during HPLC analysis, but its presence is indicated
by binding of 8% of the radioactivity to the protein pellet.
2
was the reactive N -OH-AaC.
Synthesis of N-OH-MeAaC and N-OH-AaC. In order to
2
evaluate the possible presence of unstable N -hydroxylated
2
2
metabolites, N -OH-MeAaC and N -OH-AaC were syn-
thesised by two-step procedures. MeAaC and AaC were
oxidized to NO -derivates with H O in acetic acid. In an
2
2
2
attempt to increase the yield of NO -MeAaC, we have tried
2
to use different reaction conditions. However, oxidation of
MeAaC with dimethyldioxirane, a method developed to
synthesize NO -AaC (King et al. 2000), or oxidation with
2
H O of MeAaC diluted in triflour acetic acid, trichlor
2
2
acetic acid, formic acid, diluted sulphuric acid and diluted
phosphoric acid did not result in higher yields.
In the second step the NO -compounds were reduced to
2
2
2
N -OH-MeAaC and N -OH-AaC using copper sulphate in
ascorbic acid. These methods were modified from a method
developed to synthesize N -hydroxy-2-amino-3-methylimid-
Fig. 1. Structure and molecular weights of 2-amino-9H-pyrido[2,3-
b]indole (AaC) and 2-amino-3-methyl-9H-pyrido[2,3-b]indole
2
(
MeAaC).
azo[4,5-f]quinoline (Agus et al. 2000). HPLC-MS analysis

1
30
HANNE FREDERIKSEN AND HENRIK FRANDSEN
Fig. 2. HPLC-chromatograms monitored at 360 nm showing the metabolism of MeAaC- and AaC in hepatic microsomes (micr.) from PCB-
induced rats, control rats and man. Chromatographic conditions are described in Materials and Methods. The chromatograms show MeAaC
and its metabolites in hepatic microsomes from PCB-induced rats (A), control rats (B) and man (C), and AaC and its metabolites in hepatic
microsomes from PCB-induced rats (D), control rats (E) and man (F). Peaks labelled 1–3 and arrows without labels are detoxified metabo-
lites, while peaks labelled 4–6 indicate activation of the parent compound.
2
2
of the synthetic N -OH-AaC compound showed retention
The synthetic N -OH-MeAaC has a molecular ion
time, UV spectrum and molecular mass identical to the pro-
posed N -OH-AaC metabolite from the microsomal in-
cubation of AaC, showing that some of the AaC was acti-
vated to the reactive N -OH-compound.
[MπH]π at m/z 214 with a daughter ion [MπH]π at m/z
197 and elute in the HPLC-system at the retentiontime 14.7
min. The presence of a N -OH-MeAaC metabolite was not
detected in chromatogram A–C (fig. 2).
2
2
2

IN VITRO METABOLISM OF TWO HETEROCYCLIC AMINES, AaC AND MeAaC
131
the parent compound, dimers can not be formed from AaC
or MeAaC alone. This indicates that the dimers found in
2
the microsomal incubation mixtures are formed from N -
2
OH-AaC and N -OH-MeAaC, respectively. Further cha-
racterisation of the dimers was not attempted.
Metabolic distribution. Fig. 4 shows the percentage distri-
bution of the metabolised part of MeAaC and AaC in the
hepatic microsomes from PCB-induced rats, control rats
and humans. The calculations are made from liquid scintil-
lation counting data (not shown) on fractions collected
from HPLC analysis and dissolved protein precipitates. In
3
3
the microsomal incubation of [ H]-MeAaC and [ H]-AaC
in PCB-induced rat microsomes about 8 and 6%, respec-
tively, of the radioactivity were counted in the protein pellet,
indicating activation of the parent compound. In the con-
Fig. 3. HPLC-chromatogram monitored at 360 nm showing the
synthetic AaC-dimers formed from N-OH-AaC. Chromatographic
conditions are described in Materials and Methods.
3
3
trol rat microsomal incubation of [ H]-MeAaC and [ H]-
AaC about 21 and 13%, respectively, of the radioactivity
were counted in the protein pellet, and in the human micro-
2
Addition of acetic anhydride to solutions of N -OH-
2
3
3
MeAaC and N -OH-AaC resulted in formation of three
somal incubation of [ H]-MeAaC and [ H]-AaC about 9
and 11%, respectively, of the radioactivity were counted in
the protein pellet.
2
products, each. Fig. 3 shows the HPLC profile of N -OH-
AaC treated with acetic anhydride. AaC is the major prod-
uct obtained together with two proposed dimers having
identical retention times, UV spectra and mass spectra as
the dimers found in the microsomal incubation mixtures.
Table 1 show the distribution between detoxification and
activation of the total amount of metabolised MeAaC and
AaC in the three types of microsomes. In PCB-induced rat
microsomes about 82 and 92% of the metabolised MeAaC
and AaC were detoxified, respectively. In control rat micro-
2
N -OH-MeAaC treated with acetic anhydride gave similar
results (data not shown). Due to the lack of reactivity of
Fig. 4. Distribution of MeAaC- and AaC-metabolites in hepatic microsomes (micr.) from PCB-induced rats, control rats and man shown
in percent of the total amount of metabolised parent compound.

1
32
HANNE FREDERIKSEN AND HENRIK FRANDSEN
Table 1.
Distribution between detoxification and activation of MeAaC- and AaC-metabolites in hepatic microsomes (micr.) from PCB-induced rats,
control rats and man. The data are calculated as percent of the total amount of metabolised parent compound.
Total amount of detoxification and activation (%)
PCB rat micr. Control rat micr. Human micr.
MeAaC-detoxification
AaC-detoxification
MeAaC-activation
AaC-activation
82.2∫5.1
91.8∫0.7
17.8∫5.1
8.1∫0.7
46.0∫6.5
51.7∫3.1
54.0∫6.5
48.3∫3.1
35.8∫1.1
41.7∫2.9
64.2∫1.1
58.3∫2.9
Values are presented as means∫S.D., NΩ3.
somes about 54 and 48% of the metabolised MeAaC and
AaC were activated, and in the human hepatic microsomal
incubation 64 and 58% respectively of the metabolised
MeAaC and AaC were activated.
lites were found. Three of them are detoxified metabolites,
earlier characterized as 6-OH-MeAaC, 3-CH OH-AaC and
2
7-OH-MeAaC (Frandsen et al. 1998). A minor metabolite,
with the same mass spectrum and a similar UV-spectrum as
6-OH-MeAaC and 7-OH-MeAaC, is probably also a ring-
hydroxylated detoxified metabolite of MeAaC.
Discussion
In our study of AaC metabolism in rat hepatic micro-
somes we found three major detoxification products and
small amounts of two minor metabolites. A previous study
of AaC metabolism in hepatic PCB-induced rat microsomes
has shown that AaC is metabolised to five metabolites,
Using hepatic microsomes from PCB-induced rats, control
rats and two different human pools, the in vitro metabolism
of AaC and MeAaC was investigated. In the microsomal
incubation of MeAaC four early-eluting MeAaC-metabo-
Fig. 5. Metabolic pathway of MeAaC (only the major detoxified metabolites are shown).

IN VITRO METABOLISM OF TWO HETEROCYCLIC AMINES, AaC AND MeAaC
133
three non-mutagenic and two mutagenic metabolites. The
two were supposed to be a N -hydroxy-derivate and a ni-
and 18%, respectively). A previous study of MeAaC met-
abolism in PCB-induced hepatic rat microsomes has shown
similar results; 83% detoxification and 13% activation
(Frandsen et al. 1998). In the control rat microsomes there
is a nearly fifty-fifty distribution between detoxification and
activation, 52% of the metabolised part of AaC was detoxi-
fied and 48% was activated, while 46% MeAaC was detoxi-
fied and 54% was activated. The major part of the AaC and
MeAaC metabolites from the human microsomal incuba-
tion were activated, 58 and 64%, respectively. As regards the
distribution of the AaC-metabolites in human microsomes,
our data are not comparable with a previous study, where
Raza et al. (1996) found that only 15% of AaC was acti-
vated and 85% was detoxified in human hepatic micro-
somes. In this study, however, binding of the activated meta-
bolite to protein was not taken into consideration. In vitro
studies of another heterocyclic amine, 2-amino-1-methyl-6-
phenylimidazo[4,5-b]pyridine (PhIP) has shown species
variation in metabolic distribution similar to our results on
AaC and MeAaC. The ratio of N-hydroxylation to the non
genotoxic 4’-hydroxylated derivate of PhIP was 97:1, 3.3:1
and 1.7:1 for humans, rats and mice, respectively (Lin et al.
1995).
2
troso-derivate of AaC (Niwa et al. 1982). In the human
microsomal preparation we found two major detoxification
products and one minor, which were identical to the three
major products in the rat microsome preparations. In a
study of AaC metabolism in human hepatic microsomes,
Raza et al. (1996) found two major detoxified metabolites
characterized as 3-OH-AaC and 6-OH-AaC and four mi-
2
nor metabolites characterized as N -OH-AaC, 2-nitroso-
AaC and corresponding azoxy- and nitro-derivates. There
are good indications that the two major metabolites from
the human microsome preparation are 6-OH-AaC and 3-
OH-AaC, as characterized by Raza et al. (1996). In our
microsomal incubations of AaC we did not find nitroso-,
azozy- or nitro-derivates of AaC, but we found a much
2
higher amount of N -OH-AaC (16%) in the human micro-
somal incubation.
Another interesting finding was that in all microsomal
incubations of both AaC and MeAaC, a considerable
amount of the tritium labelled AaC and MeAaC were
found in the protein fractions (fig. 4), indicating that these
parts of the compounds also had been activated and bound
to proteins.
Conclusion. In this study we have investigated the in vitro
metabolism of the a-carbolines, AaC and MeAaC in hu-
In both the microsomal preparations of AaC and
MeAaC, two late-eluting metabolites with molecular
2
man and rat hepatic microsomes. We have synthesised N -
π
masses [MπH] at m/z 365 and 393, respectively, were
OH-derivates and dimers of both AaC and MeAaC for
comparison with the activated metabolites found in the
microsomal incubations. We have revised the methods for
found. The same two sets of late-eluting products as well as
2
the parent compounds were obtained, when N -OH-AaC
2
2
and N -OH-MeAaC were added acetic anhydride. This in-
the synthesis of the N -OH-derivates, and our HPLC-MS
2
2
dicates that N -OH-AaC and N -OH-MeAaC were reacting
to form two dimers with identical retention times, UV-spec-
tra and masses as found in the HPLC-MS profiles of the
microsomal metabolism of AaC and MeAaC. This suggests
that the late-eluting AaC and MeAaC metabolites are se-
condary metabolites, created from the primary and very re-
method is specially developed to conserve the activated
metabolites of the a-carbolines.
Our data show that AaC and MeAaC have similar meta-
bolic patterns. The results also show a remarkable differ-
ence in the metabolic pattern of the a-carbolines in human
hepatic microsomes, in control rat microsomes and in PCB-
induced rat microsomes respectively. In human hepatic
microsomes, a major part of the a-carbolines are meta-
bolically activated, in hepatic microsomes from control rats
there is a fifty-fifty distribution between activation and de-
toxification and in PCB-induced hepatic microsomes the a-
carbolines are mainly detoxified.
In the light of these results we have proposed a new in
vitro metabolic pathway for both AaC and MeAaC. Finally
it is concluded that, due to the higher degree of activation
in human microsomes, human exposure to AaC and
MeAaC may result in a higher cancer risk than estimations
based on long-term cancer experiment in rodents.
2
active N -OH-derivates and the parent compounds. Fig. 5
shows a suggested metabolic pathway for MeAaC, which is
oxidized to three major detoxified metabolites; 3-CH OH-
2
AaC, 6-OH-MeAaC and 7-OH-MeAaC, and to the very
2
2
reactive metabolite N -OH-MeAaC. N -OH-MeAaC can
react in three different ways; reduction back to MeAaC,
protein adduct formation, and formation of two MeAaC-
dimers with the masses of 392. Our results indicate that the
metabolic pathway for AaC seems to be similar, with two
major detoxification products, 3-OH-AaC and 6-OH-AaC,
and a similar activation/reaction pattern.
2
It is our experience that N -OH-AaC is much more stable
2
than N -OH-MeAaC. This could explain why we did not
2
find the primary N -OH-MeAaC in the microsomal incuba-
tions, but only the secondary metabolites, dimer 1, dimer 2
and the protein bound MeAaC-derivates.
Acknowledgements
The authors thank Joan Frandsen for skilled technical
assistance and Professor Ole Andersen, Roskilde Univer-
sity, for supervision during the work. This study has been
carried out with financial support from the commission of
the European Communities, specific RTD programme
‘‘Quality of life and management of living resources’’
The metabolism of AaC and MeAaC in the three differ-
ent types of microsomes results in the same metabolites,
however in different amounts. In PCB-induced rat micro-
somes, the major part of the metabolites is detoxified (92
and 82%, respectively), only a little amount is activated (8

1
34
HANNE FREDERIKSEN AND HENRIK FRANDSEN
gens, amino-alpha-carbolines in grilled foods and cigarette
smoke condensate. Cancer Lett. 1981, 12, 105–110.
Nagao, M., Y. Fujita, K. Wakabayashi & T. Sugimura: Ultimate
forms of mutagenic and carcinogenic heterocyclic amines pro-
duced by pyrolysis. Biochem Biophys. Res. Commun. 1983, 114,
QLK1-CT99–01197, Heterocyclic amines in cooked foods –
Role in human health. It does not necessarily reflect the
commission’s views and in no way anticipates its future pol-
icy in this area.
626–631.
Niwa, T., Y. Yamazoe & R. Kato: Metabolic activation of 2-amino-
9
1
H-pyrido[2,3-b]indole by rat-liver microsomes. Mutat. Res.
982, 95, 159–170.
References
Ohgaki, H., N. Matsukura, K. Morino, T. Kawachi, T. Sugimura &
S. Takayama: Carcinogenicity in mice of mutagenic compounds
from glutamic acid and soybean globulin pyrolysates. Carcino-
genesis 1984, 5, 815–819.
Agus, C., K. F. Ilett, F. F. Kadlubar & R. F. Minchin: Characteriza-
tion of an ATP-dependent pathway of activation for the hetero-
cyclic amine carcinogen N-hydroxy-2-amino-3-methylimidazo[4,
5
-f]quinoline. Carcinogenesis 2000, 21, 1213–1219.
Pfau, W., U. Brockstedt, C. Schulze, G. Neurath & H. Marquardt:
Characterization of the major DNA adduct formed by the food
mutagen 2-amino-3-methyl-9H-pyrido[2,3-b]indole (MeAal-
phaC) in primary rat hepatocytes. Carcinogenesis 1996, 17, 2727–
2732.
Pfau, W., C. Schulze, T. Shirai, R. Hasegawa & U. Brockstedt:
Identification of the major hepatic DNA adduct formed by the
food mutagen 2-amino-9H-pyrido[2,3-b]indole (A alpha C).
Chem. Res. Toxicol. 1997, 10, 1192–1197.
Raza, H., R. S. King, R. B. Squires, F. P. Guengerich, D. W. Miller,
J. P. Freeman, N. P. Lang & F. F. Kadlubar: Metabolism of 2-
amino-alpha-carboline. A food-borne heterocyclic amine muta-
gen and carcinogen by human and rodent liver microsomes and
by human cytochrome P4501A2. Drug Metab. Disp. 1996, 24,
395–400.
Eisenbrand, G. & W. Tang: Food-borne heterocyclic amines. Chem-
istry, formation, occurrence and biological activities. A literature
review. Toxicology 1993, 84, 1–82.
Frandsen, H., S. Damkjær, S. Grivas, R. Andersson, M. L.
Binderup & J. C. Larsen: Microsomal metabolism and activation
of the environmental carcinogen 2-amino-3-methyl-9H-pyri-
do[23-b]indole. Mutagenesis 1998, 13, 181–185.
Frandsen, H., P. A. Nielsen, S. Grivas & J. C. Larsen: Microsomal
metabolism of the food mutagen 2-amino-3,4,8-trimethyl-3H-im-
idazo[4,5-f]-quinoxaline to mutagenic metabolites. Mutagenesis
1
994, 9, 59–65.
Gross, G. A. & A. Gruter: Quantitation of mutagenic/carcinogenic
heterocyclic aromatic amines in food products. J. Chromatogr.
1
992, 592, 271–278.
Hasegawa, R.: Dose-dependent formation of preneoplastic foci and
DNA adducts in rat liver with 2-amino-3-methyl-9H-pyrido[2,3-
b]indole (MeA alpha C) and 2-amino-1-methyl-6-phenylimida-
zo[4,5-b]pyridine (PhIP). Carcinogenesis 1992, 13, 1427–1431.
Holder, C. L., S. W. Preece, S. C. Conway, Y. M. Pu & D. R. Doerge:
Quantification of heterocyclic amine carcinogens in cooked meats
using isotope dilution liquid chromatography/atmospheric press-
ure chemical ionization tandem mass spectrometry. Rapid Com-
mun. Mass Spectrom. 1997, 11, 1667–1672.
Richling, E., C. Decker, D. Haring, M. Herderich & P. Schreier:
Analysis of heterocyclic aromatic amines in wine by high-per-
formance liquid chromatography-electrospray tandem mass spec-
trometry. J. Chromatogr. A 1997, 791, 71–77.
Skog, K., A. Solyakov, P. Arvidsson & M. J a¨ gerstad: Analysis of
nonpolar heterocyclic amines in cooked foods and meat extracts
using gas chromatography-mass spectrometry. J Chromatogr. A
1998, 803, 227–233.
Solyakov, A., K. Skog & M. J a¨ gerstad: Heterocyclic amines in pro-
cess flavours, process flavour ingredients, bouillon concentrates
and a pan residue. Food Chem. Toxicol 1999, 37, 1–11.
Sugimura, T.: Carcinogenicity of mutagenic heterocyclic amines
formed during the cooking process. Mutat. Res. 1985, 150, 33–
41.
Holme, J. A., H. Wallin, G. Brunborg, E. J. Søderlund, J. K. Hongs-
lo & J. Alexander: Genotoxicity of the food mutagen 2-amino-
1
-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP): formation of 2-
hydroxamino-PhIP, a directly acting genotoxic metabolite. Car-
cinogenesis 1989, 10, 1389–1396.
King, R. S., C. H. Teitel & F. F. Kadlubar: In vitro bioactivation
of N-hydroxy-2-amino-alpha-carboline. Carcinogenesis 2000, 21,
Sugimura, T., M. Nagao & K. Wakabayashi: Carcinogenicity of
food mutagens. Environ. Health Perspect. 1996, 104, Suppl 3,
429–433.
1
347–1354.
Lin, D. X., N. P. Lang & F. F. Kadlubar: Species differences in
the biotransformation of the food-borne carcinogen 2-amino-1-
methyl-6-phenylimidazo[4,5-b]pyridine by hepatic microsomes
and cytosols from humans, rats, and mice. Drug Metab Disp.
Wakabayashi, K.: Identification of new mutagenic heterocyclic
amines and quantification of known heterocyclic amines. Princess
Takamatsu Symp. 1995, 23, 39–49.
Wild, D., B. E. Watkins & M. Vanderlaan: Azido- and nitro-PhIP,
relatives of the heterocyclic arylamine and food mutagen PhIP–
mechanism of their mutagenicity in Salmonella. Carcinogenesis
1991, 12, 1091–1096.
Yoshida, D. & T. Matsumoto: Amino-alpha-carbolines as muta-
genic agents in cigarette smoke condensate. Cancer Lett. 1980,
10, 141–149.
1
995, 23, 518–524.
Manabe, S., O. Wada & Y. Kanai: Simultaneous determination of
amino-alpha-carbolines and amino-gamma-carbolines in ciga-
rette smoke condensate by high-performance liquid chromato-
graphy. J Chromatogr. 1990, 529, 125–133.
Matsumoto, T., D. Yoshida & H. Tomita: Determination of muta-