Human moricizine metabolism. I. Isolation and identification of metabolites in human urine

Article abstract of DOI:10.1080/004982597240703

1. Using synthetic standards and/or spectral data, seven moricizine metabolites were structurally identified in human urine. Two novel metabolites were identified as phenothiazine-2-carbamic acid and ethyl [10-(3-aminopropionyl) phenothiazin-2-yl] carbamate. Two novel human moricizine metabolites, 2-amino-10-(3-morpholinopropionyl) phenothiazine, a previously identified dog metabolite, and 2-aminophenothiazine, a previously identified rat metabolite, were also identified. Three additional human metabolites, phenothiazine-2-carbamic acid ethyl ester sulphoxide (P2CAEES), moricizine sulphoxide, and ethyl {10-[N-(2'-hydroxyethyl)3-aminopropionyl] phenothiazin-2-yl} carbamate, all previously described in the literature, were observed. 2. Both 2-amino-10-(3-morpholinopropionyl) phenothiazine and ethyl [10-(3-aminopropionyl) phenothiazin-2-yl] carbamate, and possibly ethyl {10-[N-(2'-hydroxyethyl)3-aminopropionyl]phenothiazin-2-yl} carbamate, possess the structural characteristics thought to be necessary for class 1 antiarrhythmic activity.

Full text of DOI:10.1080/004982597240703

, 1997,
. 27,
. 2, 217±229
no
xenobiotica
vol
Human moricizine metabolism. I. Isolation and
identi®cation of metabolites in human urine
L. E. RICHARDS , H. J. PIENIASZEK, JR *,
‹ ‹
S. SHATZM ILLER
, G. O. PAGE , K. F. BLOM
,
‹s
J. M. READ , A. F. DAVIDSON and P. N. CONFALONE
‹Ž
‹
‹
‹
‹
‹ Drug Metabolism and Pharmacokinetics and ‹Physical and Chemical Sciences, The
DuPont Merck Pharmaceutical Co., Stine- Haskell Research Center and Experimental
Station, Newark and Wilmington, DE, USA
‹University of Tel Aviv, Tel Aviv, Israel
Received 22 July 1996
1. Using synthetic standards and or spectral data, seven moricizine metabolites were
}
structurally identi®ed in human urine. Two novel metabolites were identi®ed as
phenothiazine- 2-carbamic acid and ethyl [10- (3- aminopropionyl) phenothiazin- 2- yl]
carbamate. Two novel human moricizine metabolites, 2-amino-10- (3-morpholino-
propion yl) phenothiazine, a previously identi®ed dog metabolite, and 2-aminopheno-
thiazine, a previously identi®ed rat metabolite, were also identi®ed. Three additional
human metabolites, phenothiazine- 2-carbamic acid ethyl ester sulphoxide (P2CAEES),
«
moricizine sulphoxide, and ethyl 10- [N -(2 -hydroxyethyl)3- aminopropionyl] pheno-
thiazin-2-yl carbamate, all previously described in the literature, were observed.
²
´
2. Both 2-amino- 10- (3-morpholinoprop ionyl) phenothiazine and ethyl [10- (3-
«
10- [N -(2 -
hydroxyethyl)3- aminopropionyl] phenothiazin- 2-yl carbamate, possess the structural
aminopropionyl) phenothiazin- 2-yl] carbamate, and possibly ethyl
²
´
characteristics thought to be necessary for class 1 antiarrhythmic activity.
Introduction
Moricizine HCl (USAN ; moracizine, INN, or Ethmozine, the proprietary name
frequently found in existing literature) (®gure 1) is currently marketed in the USA
and Europe as a class 1 antiarrhythmic agent. Mammalian metabolism of this
compound has been the object of a limited number of investigations over the past
two decades (Vikhlyaev et al. 1977, Sheehan et al. 1985, Sheehan 1990, Yang and
Chan 1995). In man, however, moricizine metabolism has not been fully charac-
terized. In addition, there is a lack of correlation between plasma moricizine
concentration and onset, oŒset and duration of antiarrhythmic activity (Fitton and
Buckley 1990). One explanation for this discrepancy is that certain moricizine
metabolites, due to their modi®ed chemical structures, may possess more potent
antiarrhythmic activity than parent moricizine.
Antiarrhythmic drugs are classi®ed based on their electrocardiogr aphic and
electrophysiolo gic eŒects. Class 1 antiarrhythmic agents are further subclassi®ed
into three groups : fast (1B), intermediate (1A) and slow (1C), based on the onset
and recovery kinetics of use- dependent block of the maximum rate of depolarization
(V ) of cardiac intracellular action potentials and changes in action potential
max
* Author for correspondence.
Ž Current address : Laboratory Management Systems, Wilmington, DE, USA.
Current address : IM I Institute for Research and Development, Israel Chemical Group, Haifa Bay,
s
Israel.
±
0049±8254 97 $12 00 ’ 1997 Taylor & Francis Ltd
}

218
L. E. Richards et al.
Figure 1. Moricizine HCl.
duration (APD) (Singh and Vaughan Williams 1970). This use- dependent block has
been attributed to selective binding of drug to inactivated sodium channels or
voltage-dependent binding (Hondeghem and Katzung 1977).
Moricizine has de®ed subclassi®catio n, having been described in the literature
as a class 1A (Waldo et al. 1988), 1B (Snyder 1989), and 1C antiarrhythmic drug
(Vaughan Williams 1992, Koller and Franz 1994). It has been suggested that the
uniqueness of the antiarrhythmic behaviour of moricizine is due in part to the low
pK (6±4) of the morpholine nitrogen (Schubert et al. 1986, Makielski et al. 1990).
a
This pK is well below the physiological pH of 7±4, and therefore the drug exists
a
largely as the lipophilic free base in vivo , although a small, changing fraction of the
drug would be protonated. In this way, moricizine could exhibit a chameleon- like
behaviour. In contrast, most class 1 antiarrhythmic drugs contain alkyl amines with
much higher pK ’s (8±10), and exist largely as the charged ammonium ion species
a
in vivo .
The morpholine nitrogen has a low pK relative to an alkyl amine because it has
a
two bonds that are attached by a two-carbon bridge to the oxygen atom, each
lowering the pK by approximately 1±1 pH units (Perrin et al. 1981). Therefore, any
a
metabolism of the morpholine ring that results in loss of one or both of the two-
carbon bridges between the morpholine nitrogen and oxygen results in a rise in the
pK of the nitrogen. This paper describes the structural characterization of
a
moricizine metabolites in human urine with particular emphasis on those meta-
bolites that could potentially exhibit antiarrhythmic activity.
Materials and methods
Chemicals
Sodium metaperiodate, hydrogen peroxide, activated charcoal and reagent grade solvents were
purchased from Aldrich Chemical Co. (Milwaukee, WI, USA). Acrylonitrile, t-butylammonium
hydrogen sulphate, benzyl chloroformate, morpholine, butyl lithium, potassium carbonate, magnesium
sulphate, sulphuric acid, hydrochloric acid, sodium hydroxide and reagent grade solvents were obtained
from Merck (Darmstadt, Germany). Analytical solvents and reagents : acetonitrile, methanol, acetic acid
(J. T. Baker, Inc., Phillipsburg, NJ, USA), ammonium acetate, sodium 1-pentanesulphon ate and
sodium bicarbonate (Sigma Chemical Co., St Louis, MO, USA) were all of hplc or analytical grade.
Deuterated water, acetone and DM SO were obtained from MSD isotopes (Dorval, Quebec, Canada).
Water was puri®ed using a Milli- Q system (Millipore Corp., Bedford, MA, USA).
Synthesis of reference compounds
A number of moricizine derivatives (including some known metabolites from animal studies) were
synthesized as reference compounds for metabolic investigation. These included moricizine HCl,
moricizine HCl sulphoxide, phenothiazine- 2-carbamic acid ethyl ester sulphoxide, phenothiazine- 2-
carbamic acid, 2-aminophenothiazine and 2-amino-10- (3-morpholinoprop ionyl) phenothiazine. These
compounds were used as authentic standards for the isolation of major metabolites by gradient hplc, and

Human moricizine metabolism
219
for their identi®cation by nmr and mass spectrometry. Depending on their chemical structure, they were
prepared either from moricizine or phenothiazine- 2-carbamic acid ethyl ester (P2CAEE) as starting
materials (Gritsenko et al. 1972).
Moricizine HCl sulphoxide
Moricizine HCl sulphoxide was prepared by slowly adding a cold aqueous solution of moricizine HCl
°
to a cold aqueous solution of sodium metaperiodate. The solution was stirred at 0±5 C until the starting
material had disappeared. The reaction mixture was extracted with methylene chloride, which was then
stripped and the remaining solids dissolved in re¯ uxing ethanol. After hot ®ltration, the ®ltrate was
C
®ltration, the solids were washed with cold ethanol and dried in a 60 C vacuum oven. "H- nmr (D O) :
±
3 0 and stirred at room temperature until crystals appeared. After
acidi®ed with HCl to a pH of
°
1 2 (t, 3H), 3 15 (t, 2H), 3 2 (s, 4H), 3 4 (t, 2H), 3 8 (s, 4H), 4 18 (q, 2H), 7 2±7 9 (m, 7H). Ms (NH -
#
±
±
DCI) : (M H) , m z
±
±
±
±
±
±
$
+
C
93 % by hplc.
444. Purity
¯
-
¯
}
Phenothiazine- 2-carbamic acid ethyl ester (P2CAEE ) sulphoxide
P2CAEE sulphoxide was prepared by dissolving P2CAEE in acetone and slowly adding 30 % H O at
#
#
room temperature until all of the starting material had reacted. The reaction mixture was then stripped
down and the solids dissolved in DM SO. After stirring with activated charcoal, the solution was ®ltered
and the product solids precipitated from the DM SO by water addition. After cooling and ®ltration, the
solids were stirred in re¯ uxing ethanol. The solids were again ®ltered, ground and dried overnight in a
°
110 C vacuum oven. M.P. 300 C. H- nmr (DMSO- d ) : 1 25 (t, 3H), 4 2 (q, 2H), 7 2±7 9 (m, 7H). Ms
(NH -DCI) : (M H) , m z
°
±
±
±
±
"
’
+
±
303. Purity
99 3 % by hplc.
-
¯
¯
}
$
Phenothiazine- 2-carbamic acid and 2- amino phenothiazine
Phenothiazine- 2-carbamic acid was prepared by adding 1 molar equivalent of NaOH to a solution of
±
P2CAEE (5 3 g, 18 5 mmol) in ethanol (95 %, 110 ml), under a nitrogen atmosphere. The mixture was
re¯ uxed for 1 5 h. The resulting precipitate was ®ltered and washed with ether. H- nmr (D O) : 6 64 (d,
±
±
±
"
#
±
±
±
±
±
±
±
1H), 6 70 (d of d, 1H), 7 25 (t, 1H), 7 36 (d, 1H), 7 65 (t, 1H), 7 72 (d, 1H), 7 92 (d, 1H), 8 44 (s, 1H). This
compound was decarboxylated to 2-aminophenothiazine by suspending it in water and neutralizing with
1 m HCl to pH 7. Carbon dioxide gas evolved during neutralization. A colour change was observed from
olive green to purple. The resulting solid was dried on a vacuum dessicator to give 2-aminophenothiazine
°
(3 5 g, 83 %), m.p. 182±4 C. "H- nmr (acetone-d ) : 2 8 (NH , 2H), 6 13 (s, 1H), 6 18 (d, 1H), 6 62 (d,
1H), 6 69 (d, 1H), 6 74 (t, 1H), 6 92 (d, 1H), 6 95 (t, 1H), 7 58 (NH, 1H). Ms (NH -DCI) : (M H) , m z
±
±
±
±
±
+
’
#
±
±
±
±
±
-
}
$
±
±
±
215. Elemental analysis : C, 66 98 ; H, 4 92 ; N, 13 43 ; S, 14 76 % ; C
±
± ±
H N S requires C, 67 26 ; H, 4 7 ;
¯
N, 13 07 ; S, 14 96 %.
"# "!
#
±
±
2-Amino- 10- (3-morpholinopropionyl) phenothiazine
2-Amino- 10- (3-morpholinoprop ionyl) phenothiazine was prepared from 2-aminophenothiazine by
protection of the amino group to give phenothiazine- 2-carbamic acid benzyl ester (P2CABE), followed by
addition to the morpholino side chain and subsequent deprotection. To
a solution of 2-
±
aminophenothiazine (3 9 g, 19 3 mmol) in 60 ml methylene chloride were added NaOH (0 95 g,
23 9 mmol), 15 ml water and tetrabutylammon ium hydrogen sulphate (5 mg). The mixture was cooled
±
±
±
°
under nitrogen to 10 C and a solution of benzylchloroformate (6 6 ml, 19 9 mmol) in toluene (50 % was
added dropwise. After 15 min the organic layer was separated and the aqueous layer was extracted with
±
±
methylene chloride. The organic layer was washed with 1 m HCl, dried over MgSO and evaporated
%
under reduced pressure. The resulting concentrate was chromatographed on silica gel (eluent
1 : 1
hexane : ethyl acetate) to yield P2CABE (2 4 g, 47 %). "H- nmr (DM SO- d ) : 5 12 (s, 2H), 6 63±6 92 (m,
¯
±
±
±
7H), 7 30±7 36 (m, 5H), 8 11 (s, 1H), 9 15 (s, 1H). Elemental analysis : C, 69 05 ; H, 4 87 ; N, 8 09 ; S, 9 43 ;
±
’
±
±
±
±
±
±
±
±
± ± ± ±
H O N S requires C, 68 95 ; H, 4 63 ; N, 8 04 ; S, 9 2 %. The 3-morpholinopropi onic acid ethyl ester
C
was prepared by addition of morpholine to acrylonitrile (Whitmore et al. 1944), followed by re¯ ux with
#! "’
#
#
±
ethanol in the presence of sulphuric acid (Weisel et al. 1945). Then, to a solution of P2CABE (1 09 g,
4 mmol) in 20 ml tetrahydrofuran under nitrogen at 65 C, BuLi (5 27 ml, 1 4 m, 8 mmol) was added
°
±
±
®
°
by a syringe. Deprotonation continued for 10 min at 65 C. Then a solution of 3-morpholinopropi onic
acid ethyl ester (1 5 g, 8 mmol) in 6 ml dry tetrahydrofuran was added dropwise and reacted for 1 h at
®
±
°
65 C, then 1 h at 25 C. The reaction was quenched with 20 ml water, then extracted twice with
diethyl ether and once with ethyl acetate. The combined organic layer was dried over MgSO and
°
®
%
evaporated under reduced pressure. The crude product was chromatographed on silica gel [eluent ; ethyl
acetate: methanol (9 : 1)]. The product was recrystallized from ethyl acetate hexane (1 : 9) to give the
}
±
benzyl ester analogue of moricizine (0 413 g, 25 %). This compound was treated with 40 % HBr in 2 ml
°
±
acetic acid at 25 C. A colour change to purple was observed. After 0 5 h, 10 ml dry diethyl ether was
added and the precipitate was cooled on a Buchner funnel under nitrogen, then was suspended in K CO
#
$

220
L. E. Richards et al.
Table 1. Metabolite retention times on hplc systems 1±3.
System 2a
d
3
±
(pH 4 5)
RT (min)
System 1
a
RT (min)
System 2
RT (min)
System
RT (min)
Urinary Synthetic
metabolite standard
Urinary
Synthetic
Urinary
Urinary
metabolite standard metabolite metabolite standard
Synthetic
Metabolite
b
±
±
±
(Moricizine)
33 78
32 92
nd
±
26 49
nd
nd
nd
nd
nd
nd
±
nd
±
6 78
±
±
±
1
2
3
4
5
6
7
10 35
10 79
23 08
±
23 05
7 21
±
±
±
±
12 44
nd
±
15 36
15 44
12 44
18 30
18 28
±
±
23 08
23 32
nd
nd
nd
nd
±
6 60
±
24 10
nd
nd
nd
na
na
nd
±
±
25 88
na
±
26 19
22 45
c
±
±
±
33 00
28 53
na
na
±
23 55
24 07
±
±
28 53
nd
±
24 98
nd
33 00
na
±
34 69
±
35 41
a
RT , retention time.
nd, not determined.
b
c
d
na, standard not available.
Ratios of ammonium acetate buŒer : methanol : water varied for diŒerent metabolites.
solution. The free amine was extracted into ethyl acetate. The organic layer was dried over MgSO and
%
evaporated to obtain 2- amino-10- (3-morpholinopropi onyl) phenothiazine (0 23 g, 80 %). H- nmr
±
(acetone-d ), 2 35 (s, 2H), 2 6±2 9 (m, 10H), 6 6 (d, 1H), 6 95 (s, 1H), 7 15 (d, 1H), 7 25 (t, 1H), 7 30
"
±
±
(t, 1H), 7 42 (d, 1H), 7 6 (d, 1H). Ms (NH -DCI) : (M H) , m z
±
±
±
+
±
±
±
’
±
±
H, 6 23 ; N, 11 56 ; S, 9 23 ; C
±
356. Elemental analysis : C, 64 45 ;
-
¯
}
$
±
±
±
± ± ± ±
H O N S requires C, 64 20 ; H, 5 95 ; N, 11 82 ; S, 9 02 %.
"* #"
#
$
Drug administration and sample collection
Healthy adult male and female subjects, between the ages of 19 and 30 years and weighing 50±86 kg,
signed informed consent and were entered into the study, which was approved by the Clinical Research
Institutional Review Committee (Needham, MA, USA) for Medical and Technical Research Associates
(Boston, MA, USA). Following overnight fasts, each subject received a single, oral 250 mg dose of
moricizine HCl. Urine collections were made at predose and during the 0±12- h interval post-dose.
During each collection period, the urine was stored in refrigerated vessels. All samples were stored at
°
20 C until analysed.
%®
Instrumentation conditions
}
Hplc equipment consisted of a Waters Associates (Milford, MA, USA) 600E pump and controller and
a Waters WISP sample injector. Column temperature was controlled using a Dual Zone Systec
(Thomson Instrumen t Co., Newark, DE, USA) temperature controller. Ultraviolet absorption was
monitored with an ABI (Foster City, CA, USA) 783A programm able absorban ce detector and a Hewlett-
Packard (Wilmington, DE, USA) 3392A Integrator.
Metabolite samples (10±20 lg) and synthetic samples (1±2 mg) for nmr analysis were dissolved in
deuterated water, DMSO or acetone. The spectra were obtained using a Varian VXR 400S, a Varian
VXR 500 instrument (Palo Alto, CA, USA), or a Bruker AC-200E instrument (Karlsruh e, Germany).
Chemical shifts are expressed in ppm, set relative to the solvent. Metabolites were isolated in the presence
±
of ammonium acetate and therefore the samples exhibited a resonance at d 1 9 from the acetate ion in
¯
±
4 7.
Low resolution mass spectra were obtained with a Finnigan MAT 8230 mass spectrometer (Bremen,
addition to the water resonance at d
¯
Germany) using a direct exposure probe and ammonia chemical ionization (NH -DCI). The samples
$
were deposited onto the probe from methanol methylene chloride solutions.
}
Identi®cation of urinary metabolites by chromatographic comparison with synthetic standards
Standards of known rat, dog and human moricizine metabolites (and their intermediates) were
prepared synthetically and characterized by nmr, ms, and elemental analysis. The relative retention time
(RRT) of each synthetic standard (versus moricizine) was compared with the relative retention times of
human plasma or urine metabolites under speci®c hplc conditions (table 1).
RRT
RRT
retention time of standard retention time of moricizine
}
¯
¯
d
st
retention time of metabolite retention time of moricizine.
}
met

Human moricizine metabolism
221
Figure 2. Representative hplc chromatogram (system 1) of a typical 0±12- h urine extract from a female
volunteer containing moricizine and its metabolites.
Figure 3.
H- nmr spectrum of moricizine HCl (approximately 1 mg in deuterated water).
"
Correlation was said to occur when the absolute value of
!
±
0 05.
[RRT
RRT
]
®
d
st
met
Isolation of metabolites
The metabolites were isolated using a combination of solid phase extraction (s.p.e.) and hplc.
Untreated urine (160 ml) was centrifuged and applied to a Mega BondElut (Varian Sample Preparation
Products, Harbor City, CA, USA) C- 18 s.p.e. column (pre- conditioned with 2-column vols methanol,
then 1-column vol. water) using nitrogen pressure. The column was washed three times with 5 ml water,
°
then eluted seven times with 5 ml methanol. The eluent was taken to dryness under nitrogen at 37 C.
The sample was reconstituted with hplc mobile phase and ®ltered with 0 45 lm Ultrafree ®lters
±
(Millipore Corp., Bedford, MA, USA). Three to four sequential hplc puri®cation steps were used. For
hplc system 1, a chromegabond MC- 18 (15 cm 14 mm, 5 lm, 60A, ES Industries, Berlin, NJ, USA)
¬
column at 35 C was used. S.p.e. fractions of urine were analysed using a linear gradient developed from
°
85 : 15 A : B to 10 : 90 A : B over 50 min (A
±
90 % H O, 10 % MeOH, 2 5 mm sodium- pentanesulphon ate,
¯
#
±
± ±
95 % MeOH, 5 % H O, 2 5 mm sodium- pentanesulphon ate, 0 1 % H PO ) (®gure 2).
0 1 % H PO ; B
$
¯
%
#
$
%
Metabolite fractions were collected from system 1, neutralized with sodium bicarbonate and dried under
vacuum.
For hplc systems 2 and 2a, a Beckman Ultrasphere ODS (25 cm, 5 lm ; Fullerton, CA, USA) column

222
L. E. Richards et al.
°
at 35 C was used. Fractions collected from system 1 were analysed using a linear gradient (varied for each
metabolite) of A and B [A
± ±
90 % 1 0 m ammonium acetate (neutral or pH 4 5 for 2 and 2a respectively),
¯
10 % MeOH ; B
For hplc system 3, a Beckman Ultrasphere ODS (25 cm, 5 lm) column was used. Metabolite fractions
100 % MeOH]. Analysis on system 2 was repeated if necessary.
¯
collected from system 2 were analysed using an isocratic elution of 7 : 27 : 66 (varied slightly for each
±
±
90 % 0 1 m ammonium acetate (neutral or pH 4 5), 10 % MeOH ; B
90 % H O, 10 % MeOH]. Metabolites were detected at a wavelength of 254 nm.
metabolite) A : B : C [A
100 %
¯
¯
MeOH ; C
¯
#
Results
Identi®cation of individual urinary metabolites
Metabolite 1. Synthetically prepared 2- amino-10- (3- morpholinopropionyl)
phenothiazine, a previously described dog metabolite (K. W. Sigvardson, A. F.
Davidson and H. J. Pieniaszek Jr, unpublished data), was shown to co-
chromatograph with metabolite 1 on hplc systems 1±3. The NH -DCI mass spectra
$
+
of both the standard and metabolite 1 (®gure 4A) exhibited the (M H) species at
-
m z
}
356 and little or no fragmentation.
¯
Metabolite 2. The synthetically prepared sodium salt of phenothiazine- 2- carbamic
acid was shown to co-chromatograph with metabolite 2 on hplc systems 1±3. In
"
addition, the H- nmr of about 50 g of metabolite 3 in D O [6±64 (d, 1H), 6±70 (d of
l
#
d, 1H), 7±25 (t, 1H), 7±36 (d, 1H), 7±65 (t, 1H), 7±72 (d, 1H), 7±92 (d, 1H), 8±44 (s,
1H)] (®gure 5) was identical to that of the synthetically prepared standard. When
metabolite 2 was isolated and exposed to acidic conditions (i.e. evaporation to
dryness in acidic mobile phase without prior neutralization), subsequent analysis
showed
a small amount of degradation to a compound identi®ed as 2-
aminophenothiazine (by hplc retention comparison with authentic standard).
Presumably this was the result of acid-catalysed decarboxylation of the carbamic
acid (analogous to the acid catalysed decarboxylation of synthetic phenothiazine- 2-
carbamic acid to give 2- aminophenothiazine described under Synthesis of reference
compounds).
Metabolite 3. Synthetically prepared 2- aminophenothiazine, a previously de-
scribed rat metabolite (Vikhlyaev et al. 1977), was shown to co-chromatograph with
metabolite 3 on hplc system 1. These data are consistent with the presence of 2-
aminophenothiazine in human urine, although unequivocal structural determi-
nation has yet to be completed.
Metabolite 4. Synthetically prepared moricizine sulphoxide, a previously de-
scribed, pharmacologically active human metabolite (Sheehan 1990, Pieniaszek et
al. 1994, Yang and Chan 1995), a rat metabolite (Vikhlyaev et al. 1977), and a dog
metabolite (Sigvardson, Davidson and Pieniaszek Jr, unpublished data), was shown
to co-chromatograph with metabolite 4 on hplc system 1. This metabolite has been
well- characterized in human urine and therefore was not extensively analysed in the
present study.
Metabolite 5. The "H- nmr spectrum of metabolite 5 was obtained in D O from
#
about 12 g isolated material [1 ±2 (t, 3H), 2±95 (t, 1H), 3±15 (t, 1H), 3±55 (t, 2H), 3±85
l
(m, 4H), 4±0 (m, 4H), 4±15 (q, 2H), 7±20 (d, 1H), 7±25 (t, 1H), 7±30 (t, 1H), 7±45 (d,

Human moricizine metabolism
223
+
+
Figure 4. NH -DCl m.s. of (A) metabolite 1 [(M H)
$
356], (B) metabolite 5 [(M H) 402], and
¯
358].
¯
-
(C) metabolite 6 [(M H)
¯
-
+
-

224
L. E. Richards et al.
Figure 5. The aromatic region of the H- nmr spectrum of metabolite 2 (approximately 20 lg in
"
deuterated water).
Figure 6.
H- nmr of metabolite 5 (approximately 12 lg in deuterated water).
"
1H), 7±50 (d, 2H), 7±65 (s, 1H), 8±35 (s, 1H)] (®gure 6). This showed that the
carbamate side chain of moricizine was intact but the morpholino side chain was
structurally altered. The NH -DCI mass spectrum of metabolite 5 exhibited
$
+
(M H) at m z
402 (®gure 4B). These data are consistent with an N -2-
-
¯
}
«
hydroxyethyl group in place of the morpholino ring (i.e. 10[3- N- (2 -hydroxyethyl)
amino propionyl] phenothiazin- 2-carbamic acid ethyl ester), a metabolite structure
that has previously been described in human urine (Sheehan et al. 1985).
Metabolite 6. The "H- nmr spectrum of metabolite 6 in D O was obtained from
#
about 10 g isolated material [1±2 (t, 3H), 3±1 (m, 1H), 3±2 (t, 2H), 3±3 (t, 1H), 4±2 (q,
l
2H), 7±22 (d, 1H), 7±35 (t, 1H), 7±42 (t, 1H), 7±5 (d, 1H), 7±58 (d, 2H), 7±7 (s, 1H), 8±4
(s, 1H) (®gure 7). It showed that the carbamate side chain was intact. The proton
resonances of the morpholino ring were absent but the resonances of the 3-
aminopropionyl section were present. The NH -DCI mass spectrum exhibited
$
+
(M H) at m z
358 (®gure 4C). These data are consistent with a primary amine
-
in place of the morpholino ring of moricizine (i.e., ethyl [10- (3- aminopropionyl)
¯
}
phenothiazin- 2- yl] carbamate).

Human moricizine metabolism
225
Figure 7. "H- nmr spectrum of metabolite 6 (approximately 10 lg in deuterated water).
Metabolite 7. Synthetically prepared phenothiazine- 2- carbamic acid ethyl ester
sulphoxide, a previously described human metabolite (Sheehan 1990), rat metabolite
(Vikhlyaev et al. 1977) and dog metabolite (Sigvardson, Davidson and Pieniaszek Jr,
unpublished data) was shown to co-chromatograph with metabolite 7 on hplc
system 1. This metabolite has been well characterized in human urine and therefore
was not extensively analysed in the present study.
Discussion
Moricizine metabolites described herein (table 2) provide an outline of the
metabolic pathways of this drug in man (®gure 8). Many of the human urine
metabolites of moricizine, although not structurally identi®ed, were quanti®ed in a
separate study employing "% C- moricizine (H. J. Pieniaszek, Jr., unpublished data.
In this latter radiolabelled study, the combined percent of dose for all identi®ed
!
urinary metabolites was very small ( 7 %), although several of these metabolites
achieved signi®cant plasma concentrations and were eliminated with longer half-
lives relative to parent drug. Additionally, the UV- detected chromatographic
metabolic pro®le in plasma correlated closely with the urine pro®le.
Earlier investigators of human moricizine metabolism have described the 2-
hydroxyethyl amine derivative (ethyl 10- [N -(2-hydroxyethyl)3- aminopropionyl]
²
phenothiazin- 2- yl carbamate), the sulphoxide of the bis- 2-hydroxyethylami ne
´
derivative (ethyl 10- [N ,N -(2- hydroxyethyl)3- aminopropionyl] phenothiazin- 2-yl
²
´
carbamate sulphoxide), ring- hydroxylated phenothiazine- 2- carbamic acid ethyl
ester, 2-acetamido-phenothiazine sulphoxide, 10- (3- acetylaminopropionyl)- 2-
aminophenothiazine sulphoxide and 10- (3-propionylmorpholin- 2- one)-2-amino-
phenothiazine sulphoxide (Sheehan et al. 1985, Sheehan 1990). A recent report
describes the identi®cation of moricizine sulphone (Yang and Chan 1995). In
addition, a glucuronide conjugate of 2-aminophenothiazine has been identi®ed (L.
Shum, unpublished data).
There appear to be three primary pathways of human biotransformation of
moricizine : oxidation of the phenothiazine sulphur, hydrolysis or hydrolysis

226
L. E. Richards et al.
Table 2. Structures of moricizine metabolites isolated from human urine.
Metabolite
number
Compound
R1
R2
R3
0
1
Moricizine
2-amino-10-
CO2CH2CH3
H
COCH2CH2N(CH2CH2)2O
COCH2CH2N(CH2CH2)2O
±
±
(3-morpholinoprop ionyl)-
phenothiazinea
2
3
4
5
phenothiazine-2-carbamic acid
2-aminophenothiazineb
moricizine sulphoxidea,b
ethyl (10-[N-(2¢-hydroxyethyl)
3-aminopropionyl]
CO2
H
H
H
±
±
H
CO2CH2CH3
CO2CH2CH3
COCH2CH2N(CH2CH2)2O
COCH2CH2NHCH2CH2OH
O
±
phenothiazin-2-yl)
carbamate
6
7
ethyl [10-(3-aminopropionyl)
phenothiazin-2-yl] carbamate
phenothiazin-2-carbamic acid
ethyl ester sulphoxidea,b
CO2CH2CH3
CO2CH2CH3
COCH2CH2NH2
H
±
O
a Previously identified as a dog metabolite (K. W. Sigvardson, A. F. Davidson and H. J. Pieniaszek Jr,
ÊÊÊ
unpublishe d data).
ÊÊÊÊb Previously reported rat metabolite (Vikhlyaev et al. 1977).
followed by decarboxylation of the carbamate side chains and oxidation and or
}
cleavage of the morpholino side chain. In addition, primary metabolites can undergo
further biotransformation by another of the above reactions or undergo phase II
metabolism to generate conjugated products (Vikhlyaev et al. 1977).
Biotransformations analogous to many of these have been reported for drugs
with similar structures. Oxidation of phenothiazine drugs to the corresponding
sulphoxide has been described frequently in the literature (Beckett et al. 1963,
Aravagiri et al. 1984). Ethacizin, a moricizine analogue, has been shown to undergo
hydrolysis of the carbamic acid ethyl ester to generate the free carbamic acid
metabolite in man (Beloborodov et al. 1989). Hydrolysis of the morpholino ring to
generate the bis- (2-hydroxyethyl) amine derivative or oxidation to the morpholino
a
oxygen followed by hydrolysis of the resulting ester has been described for
molsidomine metabolism in animals and man (Wilson et al. 1987). Oxidative mono-
or di- N -dealkylation of an analogous moricizine intermediate would generate the 2-
hydroxyethylami ne derivative (metabolite 5) (Sheehan et al. 1985), or the primary
amine derivative (metabolite 6), respectively. Interestingly, di- N -dealkylation of
ethacizin in man generates a metabolite structurally identical to metabolite 6
(Beloborodov et al. 1989).
Several of the moricizine metabolites described above (1, 3 and 4) have been
tested for antiarrhythmic activity in isolated guinea pig atria (B. S. Brown and A. L.
Cigan, unpublished data). Moricizine sulphoxide was found to be approximately
®ve times less potent than moricizine. All other metabolites tested did not contain
any part of the morpholine ring and did not exhibit any activity.

Human moricizine metabolism
227
Figure 8. Proposed pathways of moricizine biotransformation in man. (a) Sulphoxidation ; (b)
hydrolysis of the amide bond of the heterocyclic nitrogen atom ; (c) hydrolysis of the ester bond ;
(d) decarboxylation of carbamic acid ; (e) morpholine ring opening ; (f) oxidative N -dealkylation.
In the present study, several human moricizine metabolites were characterized
that contain either the intact morpholino side chain (metabolite 1) or the alkyl amine
in some form (metabolites 5 and 6). This alkyl amine functionality has been
proposed to be instrumental in the activity of class 1 antiarrhythmics by enhancing
the binding of drug to the sodium channel receptor (Sheldon et al. 1991). In
addition, the pK of the amine functionality has been shown to aŒect drug binding.
a
Ethacizin, a tertiary amine analogue of moricizine with a pK
9±3, is protonated at
¯
physiological pH. The sodium channel receptor oŒ-rate of ethacizin in perfused
a
cardiac Purkinje cells was greatly increased as compared with moricizine, which,
with a pK
6±4, exists largely as the free base at physiological pH (Makielski 1990).
¯
Ethacizin has been shown to be a more potent antiarrhythmic drug than moricizine
a
(Kaverina et al. 1984).
Based on these ®ndings, metabolite 1, (2- amino-10- (3- morpholinopropionyl)
phenothiazine), with the intact morpholine, would be expected to exhibit anti-
arrhythmic activity similar to that of moricizine. Likewise, metabolite 5, (ethyl
«
10- [N -(2 -hydroxyethyl)3- aminopropionyl] phenothiazin- 2- yl carbamate), would
be expected to have a pK between 7 and 8, and therefore an activity between that of
²
´
a

228
L. E. Richards et al.
moricizine and ethacizin, although the hydroxyl moiety, with its ability to hydrogen-
bond, could aŒect binding to the sodium channel in an unforeseen manner. The
primary amine functionality of metabolite 6, (ethyl[10- (3- aminopropionyl) pheno-
thiazin- 2- yl] carbamate), would have a pK between 8 and 9, and would be
a
protonated at physiological pH. Therefore, metabolite 6 would be expected to
exhibit potency similar to ethacizin.
In summary, moricizine is converted to multiple metabolites in man, some of
which contain the structural requirements for antiarrhythmic activity.
Acknowledgements
We would like to thank Dr M. Zinny (Medical and Technical Research
Associates, Brighton, MA) for supervision of the clinical study, F. V. Ferrari for
assisting with the acquisition of nmr spectra and K. N. Faulk for assistance in
preparation of this manuscript.
References
Aravagiri, M., Hawes, E. M. and M idha , K. K., 1984, Radioimmunoassay for the sulfoxide metabolite
of Tri¯ uoropera zin e and its application to a kinetic study in humans. Journal of Pharmaceutical
Science,
, 1383±1387.
73
Beckett , A. H., Beaven , M. A. and Robinson , A. E., 1963, Metabolism of chlorprom azine in humans.
Biochemical Pharmacology, , 779±794.
12
Beloborodov , V. L., Rodionov , A. P., T yukavkina , N. A., Klimov , A. V., Kaverina , N. V.,
Kolesnik , Y. A., Gritsenko , A. N. and L esina , V. P., 1989, Ethacizin metabolism in humans.
Xenobiotica,
, 755±767.
19
Fitton , A. and Buckley , M. M.- T., 1990, Moricizine : a review of its pharmacological properties and
therapeutic e cacy in cardiac arrhythmias. Drugs, , 138±167.
40
Gritsenko , A. N., Ermakova , Z. I. and Zhuravlev , S. V., 1972, The synthesis of ethmozine : A new
preparation with antiarrhythmic action. Khimiko -Farmatsevticheskii Zhurnal , , 17±19.
6
Hondeghem , L. M. and Katzung, B. G., 1977, Time- and voltage-dependent interactions of
antiarrhythmic drugs with sodium channels. Biochimica et Biophysica Acta,
, 373±398.
472
Kaverna, N. V., L yskovtsev , V. V., Senova , Z. P. and Grigoryeva , E. K., 1984, Ethacizin :
pharmacological properties and prospects of clinical employment. Kardiologiya , , 52±57 (in
24
Russian). Cumulated Index Medicus
, book 4.
25
Koller , B. and Franz, M. R., 1994, New classi®cation of moricizine and propafenone based on
electrophysiolo gic and electrocardiogr aphic data from isolated rabbit heart. Journal of Car -
diovascular Pharmacology,
, 753±760.
24
M akielski , J. C., 1990, State dependence of Ethacizin and Ethmozin block of sodium current in voltage
clamped and internally perfused cardiac Purkinje cells. Journal of Pharmacology and Experimental
Therapeutics,
Perrin , D. D., Dempsey , B. and S erjeant , E. P., 1981, pK Prediction for Organic Acids and Bases
, 1110±1117.
253
a
(London : Chapman & Hall), pp. 27±31.
Pieniaszek , H. J. Jr, Davidson , A. F., McEntegart , C. M., Quon, C. Y., Sampliner , R. E. and
M ayersohn , M., 1994, The eŒect of hepatic disease on the disposition of moricizine in humans.
Biopharmaceutics and Drug Disposition,
, 243±252.
Schubert , B., Hering , S., Bodewei , R., Rosenshtraukh , L. V. and W ollenberger , A. J., 1986, Use
15
and voltage-dependent depression by Ethmozine (moricizine) of the rapid inward sodium current
in single rat ventricular muscle cells. Journal of Cardiovascular Pharmacology, , 358±366.
8
Sheehan , E. W., 1990, Rat and human urinary metabolites of Ethmozine. PhD thesis, University of
Pittsburgh.
Sheehan , E. W., W ong, L., Zemaitis , M. A., Buchner , J. and L o, H. H., 1985, Studies of Ethmozine
metabolism by LC MS MS. In Proceedings of the 33rd Annual Conference on Mass Spectroscopy
}
}
and Allied Topics, San Diego, CA, 26±31 May, (Sante Fe, New Mexico : American Society for
Mass Spectroscopy and Applied Topics) 782±783 (abs).
Sheldon , R. S., Hill , R. J., T aovis , M. and W ilson , L. M., 1991, Aminoalkyl structural requirem ents
for interaction of lidocaine with the class 1 antiarrhythmic receptor on rat cardiac myocytes.
Molecular Pharmacology,
, 609±614.
39
Singh , B. N. and Vaughan Williams , E. M., 1970, The eŒect of Amiodarone, a new anti-anginal drug
on cardiac muscle. British Journal of Pharmacology, , 657±667.
39

Human moricizine metabolism
229
Snyder , D. W., 1989, Class 1B antiarrhythmic drugs : tocainide, mexilitine, and moricizine. Journal of
the Louisiana State Medical Society, , 21±25.
141
Vaughan W illiams , E. M., 1992, The relevance of cellular to clinical electrophysiolo gy classifying
antiarrhythmic actions. Journal of Cardiovascular Pharmacology, (suppl. 2), S1±7.
20
Vikhlyaev , Y. I., Zherdev , V. P., Gritsenko , A. N. and Ermakova , Z. I., 1977, Features of Ethmozin
metabolism [English translation]. Farmakologiya i Toksikologiya, , 19±22.
40
W aldo, A. L., Rogers , W. J., Pratt, C. M. and Capone , R. J., 1988, EŒects of encainide, ¯ ecainide,
imipramine, and moricizine on ventricular arrhythmias during the year after acute myocardial
infarction : the CAPS. American Journal of Cardiology,
, 501±509.
61
W eisel , C. A., T aylor , R. B., Mosher , H. S. and W hitmore , F. C., 1945, Substituted b-
aminopropionic esters. Journal of the American Chemical Society, , 1071±1072.
67
W hitmore , F. C., Mosher , H. S., Adams , R. R., T aylor , R. B., Capin , E. C., Weisel , C. and Yanko,
W., 1944, Basically substituted aliphatic nitriles and their catalytic reduction to amines. Journal
of the American Chemical Society,
W ilson , I. D., Fromsom , J. M. and Illing , H. P. A., 1987, The metabolism of [ C] N -ethoxycarbonyl-
, 725±731.
66
"%
, 93±104.
3-morpholinosydno nimine (molsidomine) in man. Xenobiotica,
17
Yang, J. M. and Chan, K., 1995, Isolation and characterization of sulphoxidation metabolites of
moricizine in rat and human biological ¯ uids. Methods and Findings in Experimental and Clinical
Pharmacology,
, 415±421.
17

Copyright of Xenobiotica is the property of Taylor & Francis Ltd and its content may not be copied or emailed
to multiple sites or posted to a listserv without the copyright holder's express written permission. However,
users may print, download, or email articles for individual use.