In vitro hepatic microsomal metabolism of N-benzyl-N-methylaniline

Article abstract of DOI:10.1016/S0014-827X(99)00034-8

In the present study, the in vitro microsomal metabolism of a tertiary aniline, N-benzyl-N-methylaniline (NBNMA) was studied to determine whether this compound produces an amide derivative (benzoyl) together with N-dealkylation and C- and N-oxidation products as metabolites. The preparations of the corresponding potential metabolites were undertaken and were separated using TLC and HPLC. Incubations were performed using rat microsomal preparations fortified with NADPH. The substrate and its potential metabolites were extracted into dichloromethane in the presence of NaCl and examined by TLC and HPLC-UV. The results indicated that NBNMA did not produce the corresponding amide (benzoyl derivative) or N-oxide metabolite but was dealkylated to the corresponding secondary amine. Two p-hydroxylated phenolic metabolites were also observed. These findings support the concept that nitrones are essential intermediate metabolites for the formation of amides from secondary aromatic amines (chemical rearrangement to amide via an oxaziridine intermediate). The carbinolamine produced from NBNMA does not seem stable enough to allow further oxidation to the amide and therefore this intermediate is broken down to the dealkylation products. N-Dealkylations and p-hydroxylations are major metabolic reactions following in vitro hepatic microsomal metabolism of the benzylic tertiary aniline, NBNMA.

Full text of DOI:10.1016/S0014-827X(99)00034-8

www.elsevier.nl/locate/farmac
Il Farmaco 54 (1999) 331–337
In vitro hepatic microsomal metabolism of
N-benzyl-N-methylaniline
a
a,
b
8
Ilkay Ku¨c¸u¨kgu¨zel , Mert Ulgen *, John W. Gorrod
a Uni6ersity of Marmara, Faculty of Pharmacy, Department of Pharmaceutical Chemistry, 81010 Haydarpasa, Istanbul, Turkey
b Toxicology Unit, John Tabor Laboratories, Uni6ersity of Essex, Wi6enhoe Park, Colchester CO4 3SQ, Essex, UK
Received 3 July 1998; accepted 27 December 1998
Abstract
In the present study, the in vitro microsomal metabolism of a tertiary aniline, N-benzyl-N-methylaniline (NBNMA) was studied
to determine whether this compound produces an amide derivative (benzoyl) together with N-dealkylation and C- and
N-oxidation products as metabolites. The preparations of the corresponding potential metabolites were undertaken and were
separated using TLC and HPLC. Incubations were performed using rat microsomal preparations fortified with NADPH. The
substrate and its potential metabolites were extracted into dichloromethane in the presence of NaCl and examined by TLC and
HPLC–UV. The results indicated that NBNMA did not produce the corresponding amide (benzoyl derivative) or N-oxide
metabolite but was dealkylated to the corresponding secondary amine. Two p-hydroxylated phenolic metabolites were also
observed. These findings support the concept that nitrones are essential intermediate metabolites for the formation of amides from
secondary aromatic amines (chemical rearrangement to amide via an oxaziridine intermediate). The carbinolamine produced from
NBNMA does not seem stable enough to allow further oxidation to the amide and therefore this intermediate is broken down
to the dealkylation products. N-Dealkylations and p-hydroxylations are major metabolic reactions following in vitro hepatic
microsomal metabolism of the benzylic tertiary aniline, NBNMA. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Microsomes; Metabolism; Tertiary anilines; Amide formation
genation step on the constituent nitrogen giving an
N-hydroxy compound which could be further oxidised
to a nitrone, followed by conversion of the nitrone to
the corresponding amide via an oxaziridine intermedi-
ate (Fig. 2, R=H).
1. Introduction
Studies on the in vitro metabolism of substituted
N-benzylanilines by hepatic microsomal preparations
from various species have previously been performed to
establish the formation of the corresponding amides
[1–5]. Two mechanisms have been proposed for the
metabolic formation of amides [5]. One mechanism
suggests that initial oxygen attack results in hydroxyla-
tion of the benzylic carbon atom to form a carbino-
lamine intermediate, which is further oxidised to give
the amide (Fig. 1, R=H). If amide formation utilises
this route, then metabolism of N-benzyl-N-methylani-
line (NBNMA) may give a N-benzoyl metabolite by
further oxidation of the carbinolamine (Fig. 1, R=
CH3).
Previous studies on the in vitro hepatic metabolism
of N-benzylaniline (NBA) and N-benzyl-4-methylani-
line (NBPT) showed the formation of the correspond-
ing amide and nitrone [1–6] in a number of species.
Therefore, an in vitro metabolic study on NBNMA
(Fig. 3), the N-methyl derivative of NBA, was carried
out. NBA itself was previously shown to be one of the
best substrates for amide formation [1–4]. The aim was
to establish whether the corresponding amide (the N-
benzoyl derivative) is produced. This study with
NBNMA was planned to inhibit initial nitrone forma-
tion by introducing the methyl moiety into the sub-
strate molecule which would prevent oxygenation of the
constituent nitrogen to hydroxylamine and hence ni-
trone. If the corresponding amide is detected, this
Another route proposed for the formation of an
amide from a secondary aniline requires an initial oxy-
* Corresponding author.
0014-827X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0014-827X(99)00034-8

332
I. Ku¨c¸u¨kgu¨zel et al. / Il Farmaco 54 (1999) 331–337
Fig. 1. Possible formation of amides from benzylic secondary (R=H) andtertiary (R=CH3) amines via a carbinolamine intermediate.
Fig. 2. Possible formation of amides from benzylic secondary (R=H) amines via nitrones and oxaziridines (if R=CH3, this reaction is not
possible).
’
formaldehyde (m/z 183), benzaldehyde (m/z 107), CH3
would mean that the N-oxidative route proposed above
is probably not involved in the formation of amides
and benzylic carbon oxidation may be involved (Fig. 1,
if R=CH3). In contrast, a negative result for amide
formation from NBNMA will support the concept that
metabolic amide formation requires initial N-oxidation
to nitrone. From our previous experiments on the
metabolism of other benzylic anilines, it is predicted
that a number of other metabolic products, i.e. the
phenolic metabolites, debenzylated amines and the cor-
responding aldehydes, would be produced from this
substrate. The results of the above experiments are the
subject of this communication.
(m/z 198) and Ar–CH2 (m/z 106) from M+ and the
’
’
’
subsequent loss of CH3 (m/z 182) or Ar–CH2 (m/z
106) from the parent amine (m/z 197), was observed.
The structure of NBNM4HA was determined on the
1
basis of H NMR and EI-MS data besides elemental
analysis. Chemical shifts observed in the 1H NMR
spectrum were consistent with the structure (see text).
The fragmentation pathway for this compound was
similar to those reported for EI-MS of N-benzyl-N-
methyl-4-aminophenol [16] displaying the fragment ions
’
’
arising from the loss of CH3 (m/z 198), C6H5 (m/z
’
136) and Ar–CH2 (m/z 122).
3. Chemical experimental section
2. Chemistry
N-Methylaniline (NMA), N-methyl-N-benzylaniline
(NBNMA), benzyl and benzoyl chlorides were pur-
chased from Aldrich. 4-(Methylamino)phenol sulphate
was purchased from Fisons, UK. All solvents, glacial
acetic acid and hydrogen peroxide were obtained from
Merck.
The metabolic study on NBNMA required the syn-
thesis of three potential metabolites, namely, N-ben-
zoyl-N-methylaniline (NBZNMA), N-benzyl-N-methyl-
aniline-N-oxide (NBNMAO) and N-benzyl-N-methyl-
4-hydroxyaniline (NBNM4HA). NBZNMA was pre-
pared by the reaction of NMA and benzoyl chloride in
aqueous alkali [7,8]. Synthesis of the corresponding
N-oxide (NBNMAO) was achieved by the action of
H2O2 on NBNMA in acetic acid [9–11]. The 4-hydrox-
ylated derivative of NBNMA (NBNM4HA) was ob-
tained by direct benzylation of 4-(methylamino)phenol
in acetone [12–14]. Alkylation of the phenol function
to yield an ether was avoided by the use of NaHCO3
instead of strong alkali.
3.1. Preparation of potential metabolites of NBNMA
3.1.1. N-Benzoyl-N-methylaniline (NBZNMA) [8]
N-Benzoyl-N-methylaniline was prepared by the ben-
zoylation of N-methylaniline (1.07 g, 0.01 mol) in the
NBZNMA [8] and NBNMAO·HCl [10,11] had iden-
tical melting points to those reported previously. The
structure of NBNMAO was further confirmed by the
use of EI-MS which displayed the correct M+ peak
(m/z 213). In addition, the expected fragmentation
pathway [15], such as the loss of oxygen (m/z 197),
Fig. 3. Structure of N-benzylic tertiary aniline, NBNMA.

I. Ku¨c¸u¨kgu¨zel et al. / Il Farmaco 54 (1999) 331–337
333
Table 1
Chromatographic properties of NBNMA and its potential metabolites a
Compound
Abbreviation
TLC (Rf×100)
HPLC, Rt (min)
N-Benzyl-N-methylaniline
N-Benzoyl-N-methylaniline
N-Benzyl-N-methylaniline-N-oxide
N-Methylaniline
N-Benzylaniline
Benzaldehyde
NBNMA
NBZNMA
NBNMAO
NMA
NBA
B
74
49
0
61
68
19.67
4.84
16.59
3.57
10.29
2.97
N-Benzyl-4-hydroxyaniline
N-Benzyl-N-methyl-4-hydroxyaniline
NB4HA
NBNM4HA
33
44
4.47
5.88
a HPLC solvent system: acetonitrile–0.02 M phosphate buffer, 40:60, v/v (pH 7) (flow rate: 2 ml/min; umax: 254 nm). TLC solvent system:
petroleum ether–acetone (70:30, v/v) (b.p. of the petroleum ether is 40–60°C).
presence of NaOH solution (10 ml, 2N). The resulting
mixture was kept overnight and then extracted with
diethyl ether (3×20 ml). The ethereal solution was
subsequently washed with HCl solution (5%, 3×20 ml)
and distilled water in order to eliminate any unreacted
N-methylaniline. Evaporation of the solvent afforded
the amide (yield 62%). M.p. 64°C [8]. UV (ethanol),
umax (nm): 276.0
TLC and stopped when the level of a new substance
formed was maximal. Any remaining acetone was evap-
orated and the solid remaining was dissolved in diethyl
ether, dried over anhydrous sodium sulphate and dry
hydrogen chloride was passed through the solution.
The precipitated hydrochloride salt was washed with
diethyl ether and recrystallised from diethyl ether (yield
20%). M.p. 173–175°C. UV (ethanol), umax (nm): 270.0,
315.5. EI-MS, m/z (% relative intensity): 214 (23), 213
(M+, 100), 196 (1.8), 161 (2.6), 136 (13), 122 (23), 91
(11), 65 (2) [16].
3.1.2. N-Benzyl-N-methylaniline-N-oxide (NBNMAO)
[10,11]
N-Benzyl-N-methylaniline (1.97 g, 0.01 mol) was dis-
solved in acetic acid (10 ml). To this stirred solution,
portions of H2O2 (35%, 1.7 ml, 0.02 mol) were added
dropwise. The mixture was kept for 3 days without
stirring. The reaction mixture was cooled to −10°C (in
ice–salt mixture) and made alkaline by adding chilled
NaOH (20%) dropwise. The separated brownish oil was
then extracted with diethyl ether in order to eliminate
any unreacted amine. The aqueous phase was saturated
with NaCl and extracted with DCM (3×20 ml). Evap-
oration of the solvent gave crude N-oxide. Recrystalli-
sation from diethyl ether–methanol mixture afforded
NBNMAO in a pure state (yield 45%), m.p. 73–77°C.
After passing dry hydrogen chloride through a solution
of the N-oxide (in chloroform) the HCl salt formed,
m.p 128°C, as previously reported [10,11]. UV
(ethanol), umax (nm): 252.4, 269.2 (shoulder). EI-MS,
m/z (% relative intensity): 213 (M+, 100), 197 (28), 182
(4), 122 (52), 106 (16), 91 (40), 77 (16), 65 (6), 51 (5).
4. Analytical separation section
N-Benzylaniline (NBA) was obtained from British
Drug Houses (BDH), Poole, UK. 4-(Benzyl-
amino)phenol (NB4HA) was purchased from Eastman,
Rochester, NY, USA. N-Benzoyl-N-methylaniline
(NBZNMA),
N-benzyl-N-methyl-4-aminophenol
(NBNM4HA) and N-benzyl-N-methylaniline-N-oxide
(NBNMAO) were synthesised as reported in the text.
The separation techniques used were based on TLC and
an isocratic HPLC system. TLC was carried out using
precoated silica gel GF254 0.25 mm on glass (Merck)
using petroleum ether–acetone (70:30, v/v) as the sol-
vent system. The plates, after development, were exam-
ined under UV light (254 nm) and then sprayed with
diazotised sulphanilic acid reagent (specific for phenolic
compounds), followed by sodium carbonate. Metabolic
standards were eluted with a mobile phase composition
of acetonitrile–0.2 M phosphate buffer (40:60, v/v)
(final pH 7), at a flow rate of 2 ml/min. The metabolic
products were detected by their absorbance at 254 nm.
Table 1 shows TLC Rf and HPLC Rt values of the
substrate and its potential metabolites.
3.1.3. N-Benzyl-N-methyl-4-hydroxyaniline
(NBNM4HA)
4-Methylaminophenol sulphate was treated with
NaHCO3 solution (5%), followed by extraction with
diethyl ether. The solvent was then evaporated and the
remaining solid, 4-methylaminophenol (0.62 g, 0.005
mol) was dissolved in acetone (30 ml) and NaHCO3
(0.5 g, 0.006 mol) was added. This mixture was refluxed
and benzyl chloride (0.005 mol, 0.63 g) was dropped
through the condenser. The reaction was monitored by
All chromatographic solvents were obtained from
Merck. The HPLC column (Spherisorb C18 5 mm, 25
cm length×4.6 mm i.d.) was purchased from Phase
Separations Ltd., Deeside, UK. The guard column
packing material (Whatman Pellicular ODS) was pur-

334
I. Ku¨c¸u¨kgu¨zel et al. / Il Farmaco 54 (1999) 331–337
chased from Whatman International Ltd., Maidstone,
UK. The HPLC chromatograph consisted of an iso-
cratic system comprising one LCD analytical consta-
Metric 3200 solvent delivery system, a Rheodyne
syringe-loading sample injector valve (model 7125)
fitted with a 20 ml sample loop, a Milton ROY spec-
troMonitor-3100 variable wavelength UV detector, and
a Milton ROY integrator. A rapiscan SA6508 UV
detector was connected to the HPLC to directly obtain
UV spectra of eluting metabolites from NBNMA.
mono sodium salt (NADP) and glucose-6-phosphate
disodium salt were obtained from Sigma. Hepatic
washed rat microsomal preparations were prepared at
0°C using the calcium chloride precipitation method
[17]. Incubations were carried out in a shaking water
bath at 37°C for 30 min using a standard co-factor
solution at pH 7.4. Co-factor generating solutions con-
sisting of NADP (2 mmol), glucose-6-phosphate (10
mmol), glucose-6-phosphate dehydrogenase (1 unit),
MgCl2 (20 mmol) prepared in phosphate buffer (2 ml,
0.2 M, pH 7.4) were pre-incubated for 5 min before
addition of microsomes (1 ml), equivalent to 0.5 g
original liver and substrate (2 mmol in 50 ml methanol
per flask). Metabolic reactions were stopped by extrac-
tion with dichloromethane (DCM) (2×5 ml) after
adding NaCl (0.8 g). The DCM extracts were evapo-
rated to dryness using a stream of N2 at 20°C. Dry
5. Biological section
Glucose-6-phosphate dehydrogenase was purchased
from the Boehringer Mannheim Corporation (London)
Ltd. Nicotinamide adenine dinucleotide phosphate
Fig. 4. HPLC chromatogram obtained (A) from standards, (B) following extraction from male rat microsomal incubation mixture with NBNMA*
as substrate (1=B, 2=NMA, 3=NB4HA, 4=NBZNMA, 5=NBNM4HA, 6=NBA, 7=NMNBAO, 8=NBNMA (substrate)). * See Table
1 for abbreviations.

I. Ku¨c¸u¨kgu¨zel et al. / Il Farmaco 54 (1999) 331–337
335
organic residues were reconstituted in 200 ml of
methanol for HPLC and 100 ml of methanol for TLC
analysis.
6. Results and discussion
Examination of extracts obtained following the in
vitro metabolism of NBNMA failed to show the forma-
tion of the N-benzoyl derivative as a metabolite. How-
ever, benzaldehyde and both secondary amines, i.e.
NMA and NBA, were observed by HPLC (Fig. 4) and
TLC (Fig. 5). The formation of the N-oxide could not
be demonstrated. The two phenolic metabolites, i.e.
NBNM4HA and NB4HA, were also detected, having
identical HPLC and TLC chromatographic properties
to those of authentic standards (Figs. 4 and 5). These
metabolites also gave a positive response to diazotised
sulphanilic acid reagent (with an orange colour) which
supports our proposal. Further confirmation of the
formation of all metabolites from NBNMA was
achieved using a rapiscan UV detector (Fig. 6). From
N-dealkylated metabolites, only NBA was detected by
TLC (Fig. 5). Fig. 7 shows established metabolic path-
ways for NBNMA.
Fig. 5. TLC chromatogram obtained following extraction from male
rat microsomal incubation mixture with NBNMA* as substrate. 1,
NMA; 2, NBA; 3, NB4HA; 4, NBNM4HA; 5, NBZNMA; 6, NBN-
MAO; 7, NBNMA; 8, test; 9, control (denatured microsomes); 10,
control (without co-factors). * See Table 1 for abbreviations.
This present work, using the benzylic tertiary amine
NBNMA as a substrate, clearly supports the proposal
that the formation of amides requires initial nitrone
Fig. 6. UV spectra of NBNMA* metabolites. Authentics: 1a, B; 2a, NMA; 3a, NB4HA; 4a, NBNM4HA; 5a, NBA; 6a, NBNMA. Metabolites:
1b, B; 2b, NMA; 3b, NB4HA; 4b, NBNM4HA; 5b, NBA; 6b, NBNMA. * See Table 1 for abbreviations.

336
I. Ku¨c¸u¨kgu¨zel et al. / Il Farmaco 54 (1999) 331–337
Fig. 7. Possible metabolic intermediates involved in the biotransformation of NBNMA.
formation which can only be formed from secondary
amines (Fig. 2), as it did not indicate the formation of
the corresponding amide. The methyl moiety intro-
duced into the substrate prevented nitrone formation
which supports our previous proposal that a-C benzylic
oxidation leads to debenzylation, whereas nitrone for-
mation leads to amide formation [4–6,18]. It has previ-
ously been shown that nitrones can chemically
rearrange to oxaziridines that yield amides following
ring cleavage [18].
The two phenolic metabolites observed during the
metabolism of NBNMA are the first examples from a
tertiary benzylic amine substrate. However, secondary
benzylic anilines are known to form p-hydroxylated
metabolites in vitro as a major metabolic pathway [2,6].
In a previous study, N,N-dialkyl- and N-alkyl-4-
aminophenols derived from N-ethyl-N-methylaniline
have also been detected from incubation mixtures of
rabbit hepatic microsomes [19].
Acknowledgements
The authors thank King’s College London Mass
Spectrometry Services for EI-MS analysis.
References
[1] N.J. Gooderham, J.W. Gorrod, Microsomal N-oxidation of
secondary aromatic amines, in: J.W. Gorrod, L.A. Damani
(Eds.), Biological Oxidation of Nitrogen in Organic Molecules,
Ellis Horwood, Chichester, UK, 1985, pp. 81–95.
[2] J.W. Gorrod, N.J. Gooderham, Microsomal N- and C-oxidation
of 4-substituted N-benzylanilines, Xenobiotica 15 (1985) 1021–
1031.
[3] N.J. Gooderham, J.W. Gorrod, Microsomal formation of
amides from secondary aromatic amines, in: J.W. Gorrod (Ed.),
Development of Drugs and Modern Medicines, Ellis Horwood,
Chichester, UK, 1986, pp. 318–323.
[4] M. Ulgen, Metabolic and chemical studies on amide formation
from benzylic amines, Ph.D. Thesis, University of London, UK,
1992.
In the present study, the N-oxidation of NBNMA to
the corresponding N-oxide (NBNMAO) was prevented
by electronic and/or steric influences. In addition,
failure to detect this N-oxide metabolite in our experi-
mental conditions may be due to low extractability of
this metabolite from the aqueous phase or it may have
been formed in a very low amount. Alternatively, as
benzylic N-oxides often undergo the ‘Meishenheimer
rearrangement’ to produce the corresponding
alkoxyamines [8,9,14], this well-known reaction may
account for the failure to detect the N-oxide in the
present study. Experiments on the metabolism of NBN-
MAO are in progress to examine these proposals fur-
ther.
[5] M. Ulgen, J.W. Gorrod, D. Barlow, Structure–activity relation-
ships in the formation of amides from substituted N-benzylanili-
nes, Xenobiotica 24 (1994) 735–748.
[6] M. Ulgen, J.W. Gorrod, Microsomal formation of N-benzyl-4-
hydroxymethylaniline from N-benzyl-4-methylaniline, J. Pharm.
Pharmacol. 44 (1992) 865–866.
[7] A.I. Vogel, Practical Organic Chemistry, 3rd ed., Longmans,
London, 1961.
[8] I.A. Kaye, H.C. Klein, W.J. Burlant, Iodine-catalyzed benzoyla-
tions, J. Am. Chem. Soc. 75 (1953) 745–746.
[9] M.M. Davis, H.B. Hetzer, The relative basicities of tribenzy-
lamine and tribenzylamine oxide in benzene and water. Prepara-
tion and properties of tribenzylamine oxide, J. Am. Chem. Soc.
76 (1954) 4247–4260.
[10] A.H. Wragg, T.S. Stevens, D.M. Ostle, The rearrangement of
amine oxides, J. Chem. Soc. (1958) 4057–4064.

I. Ku¨c¸u¨kgu¨zel et al. / Il Farmaco 54 (1999) 331–337
337
[11] U. Schoellkopf, U. Ludwig, Umlagerung von N-Methyl-N-ben-
zyl-anilinoxiden zu N-Methyl-N-phenyl-O-benzyl-hydroxylami-
nen: Einflußn von Substituenten im stationaeren Rest auf die
Geschwindigkeit der Meisenheimer-Umlagerung, Chem. Ber. 101
(1968) 2224–2230.
[12] F.P. Ballistreri, E. Maccarone, A. Mamo, Kinetics and mechanism
of benzylation of anilines, J. Org. Chem. 41 (1976) 3364–3367.
[13] M. Ulgen, J.W. Gorrod, Synthesis, isolation and characterization
of some substituted N-benzyl and N-benzoyl anilines, J. Fac.
Pharm. Istanbul. 30 (1994) 1–9.
[15] D.A. Cowan, L.H. Patterson, L.A. Damani, J.W. Gorrod,
Metabolism of 4-substituted-N-ethyl-N-methylanilines, Biomedic.
Mass Spectrom. 9 (1982) 233–240.
[16] W.N. Wu, J.F. Pritchard, K.T. Ng, J.F. Hills, J.A. Uetz,
K.A. Yorgey, L.A. McKown, P.J. O’Neill, Disposition of bepridil
in laboratory animals and man, Xenobiotica 22 (1992) 153–
169.
[17] J.B. Schenkman, D.L. Cinti, Preparation of microsomes with
calcium, Methods Enzymol. 52 (1978) 83–89.
[18] M. Ulgen, J.W. Gorrod, Metabolic and chemical studies on
a,N-diarylnitrones, Drug Metab. Drug Interact. 11 (1994) 1–13.
[19] J.W. Gorrod, L.H. Patterson, Evidence for p-hydroxylation of
N-ethyl-N-methylaniline, Xenobiotica 10 (1978) 603–610.
[14] M. Ulgen, J.W. Gorrod, Synthesis and characterization of some
new 2,4,6-trisubstituted N,N-dibenzylanilines, J. Pharm. Univ.
Mar. 7 (1991) 25–32.
.