 |
Introduction |
Aminoalkylindoles
(AAIs2) are a structurally distinct group of
cannabinoid receptor agonists that were originally developed from
pravadoline (Ward et al., 1990
; Bell et al., 1991; D'Ambra et al.,
1992; Kuster et al., 1992; Eissenstat et al., 1995). AAIs have been
found to displace potent cannabinoid ligands such as CP-55940 in
competitive binding to the CB1 cannabinoid receptor (Ward et al.,
1990). The ability of AAI analogs to inhibit cerebella adenylate
cyclase activity and neuronally stimulated contractions in the mouse
vas deferens preparation, and lower intraocular pressure, suggests a
receptor-mediated mechanism of action (Pacheco et al., 1990; D'Ambra
et al., 1992; Song and Slowey, 1998). A widely used prototype of AAIs
is
R(+)-[2,3-dihydro-5-methyl-3-[(morpholinyl)methyl]pyrrolo[1,2,3-de]1,4-benzoxazinyl]-(1-naphthalenyl)methanone mesylate (WIN55212-2), a structurally constrained analog. WIN55212-2 has been used as a radiolabeled probe for CB1 binding assays (Kuster et
al., 1992).
Whereas the physiological effects of WIN55212-2 via CB1 receptor
mediation have been studied extensively (Aceto et al., 2001; Bridges et
al., 2001; Gardiner et al., 2001; Simoneau et al., 2001), to our
knowledge, there has been very little research on the metabolism of
WIN55212-2 and other aminoalkylindole analogs, either in vitro or in
vivo. Because of the remarkable structural difference between
aminoalkylindoles and other cannabinoid ligands, it is hypothesized
that AAIs could undergo biotransformations that also differ
significantly from the metabolic patterns of the classical and
nonclassical cannabinoids. Furthermore, some of the metabolic products
of WIN55212-2 may retain cannabimimetic activities of their parent
compound, thereby contributing to the overall physiological efficacy of
WIN55212-2. However, no metabolites of WIN55212-2 have been identified
or isolated to date, and their potential biological activities remain
unknown. Thus, the current study was undertaken to investigate the
metabolic fate of WIN55212-2 using liver microsome preparations. The
purpose of the study also included the isolation and purification of
major WIN55212-2 metabolites for detailed structural characterization
as well as for future cannabinoid receptor binding studies. Previous
metabolic studies have focused on naturally occurring cannabinoids and
structurally similar synthetic derivatives. The low polarity of the
cannabinoids makes them amenable to gas chromatographic analysis and
the metabolites, upon derivatization, have also been qualitatively
determined by gas chromatography-mass spectrometry (Harvey and Paton,
1984; Harvey, 1999). The highly polar nature of WIN55212-2 and its
metabolites does not allow direct analysis by gas chromatography, and
derivatization can be tedious and technically difficult. In this study,
high-performance liquid chromatography (HPLC) coupled with tandem mass
spectrometry (MS/MS) was used for separation and identification of
metabolic products. In addition, where possible, NMR spectroscopy was
employed to provide complementary structural information.
| |
Materials and Methods |
Materials.
WIN55212-2 was purchased from Sigma/RBI (Natick, MA). HPLC-grade
solvents (acetonitrile, methanol, and water) and deuterated solvents
(CDCl3 and CD3OD) were
purchased from Sigma-Aldrich (St. Louis, MO). All other
chemicals were obtained from Fisher Scientific Co. (Fair Lawn, NJ). Rat
liver microsomes were purchased from BD Gentest Corporation
(Woburn, MA) and stored at 
80°C prior to use.
Microsomal Incubations.
One microliter of a 20 mM WIN55212-2 stock solution, prepared in
dimethyl sulfoxide, was added as substrate to individual incubation
aliquots. Liver microsomes (protein concentration at 1.5 mg/ml) were
preincubated at 37°C for 3 min. The 0.2-ml incubation solution
consisted of 75 mM potassium phosphate (pH 7.4), 17 mM magnesium
chloride, 7 mM NADP+, 17 mM glucose 6-phosphate, and 1.2 units of
glucose-6-phosphate dehydrogenase. Reactions were initiated by the
addition of WIN-55212-2 (final concentration 100 µM). Incubation
times ranged from 1 to 4 h. Incubations were halted by placing the
incubation vials in an ice bath, with addition of an equal volume of
methanol (0.2 ml) and stored at 20°C until analysis. Prior to HPLC
analysis, microsomal proteins were precipitated by centrifugation at
room temperature, and the methanol was evaporated with a stream of
nitrogen at 37°C. The residual solution was applied to 6-ml SUPELCO
SPE C18 solid-phase extraction columns pretreated with water and methanol. The columns were washed with HPLC-grade water
and eluted with methanol; effluents were concentrated by a nitrogen
stream at 37°C. A number of control incubations were also performed
including incubation with microsomes inactivated by heating at 100°C
for 10 min, incubation in the absence of NADPH, incubation in the
absence of microsomes, and microsomes incubated in the absence of
WIN55212-2.
HPLC-UV Analyses.
Initial analyses of the incubation products were performed on a
Shimadzu QP8000 HPLC-MS system equipped with a parallel UV-Vis SPD-10ADVP detector (Shimadzu Instruments Co. Columbia, MD).
A 4.6 × 150 mm, 5-µm Supelco C8 HPLC
column (Supelco Corp., Bellefonte, PA), coupled to a Supelco
C18 guard column (4.0 × 18 mm, 5 µm) was
used for separation. Mobile phase flowrate was set at 1.0 ml/min, with
gradient elution starting at 10% acetonitrile and 90% water for 5 min, followed by a linear increase to 60% acetonitrile in 20 min, and
a linear change to 100% acetonitrile in 5 min, and a final linear
change back to 10% acetonitrile in 5 min. Eluted components were
detected by the UV detector (
max, 330 nm). To ensure reproducibility, at least three injections were performed for
each incubation aliquot. No significant qualitative or quantitative differences were found between runs.
Semipreparative HPLC separation.
Semipreparative HPLC separation of the metabolites was carried out on a
Phenomenex (Torrance, CA) ODS HPLC column (10.0 × 250 mm; 4-µm
pore size) coupled to a Phenomenex ODS guard precolumn (10 × 50 mm, 4 µm). A model 7125 Rheodyne manual injector with 500-µl loop
volume (Rheodyne Inc., Cotati, CA) was used for sample introduction.
Mobile phase flowrate was set at 5.5 ml/min, with gradient elution
starting at 10% acetonitrile and 90% water for 5 min, followed by a
linear increase of acetonitrile composition to 50% in 15 min and a
linear change to 100% acetonitrile in 7 min. Eluent absorbance was
monitored at 330 nm using a variable wavelength detector. Incubation
products were centrifuged to precipitate proteins, and the supernatant
was subjected to HPLC separation. Three peaks were collected
corresponding to metabolites M4 (m/z
461), M5 (m/z 461), and the unchanged
WIN55212-2 (m/z 427), respectively. Samples
corresponding to each metabolite were pooled and dried under vacuum and
used for NMR analyses.
LC/MS and LC/MS/MS Analyses.
A Phenomenex ODS HPLC column (2.1 × 150 mm; 4-µm pore size)
coupled to a Supelco C18 guard column (2 × 18 mm, 5 µm) was used for separation. Mobile phase flowrate was set
at 0.1 ml/min, with gradient elution starting at 10% acetonitrile and
90% water for 5 min, followed by a linear increase of acetonitrile
composition to 100% in 20 min. MS/MS experiments were performed on a
Quattro II triple quadrupole tandem mass spectrometer equipped with an electrospray ionization (ESI) source (Micromass Inc., Beverly, MA). The
ESI "needle" potential was set at 3.46 kV, and the orifice potential was set at 70 V for MS scans and 62 to 67 V for MS/MS measurements. In MS/MS experiments examining collision-induced dissociation of selected precursors taking place in the central hexapole collision cell, argon was used as the collision gas at collision energies between 17 to 20 eV and collision-induced
dissociation pressure 1.9 × 10
4 mbar. The
ion source was held at 250°C.
NMR Spectroscopy.
1H NMR spectra were recorded at 400 MHz on a
Varian Unity-400 spectrometer (Varian Medical Systems, Palo
Alto, CA). Data were processed on a SUN-5 computer using Varian VNMR
software version 6.1B. Each metabolite was dissolved in 0.5 ml
methanol-d4 (99.9 atom
%2H) or
chloroform-d1 (99.8 atom
%2H) for 1H NMR analysis.
Chemical shifts are reported on the 
scale (parts per million) by
assigning the residual solvent peak to 3.35 ppm for methanol and 7.26 ppm for chloroform, respectively. Typical data acquisition parameters
were as follows: data size, 32,000; sweep width, 8125 Hz, filter width,
8945 Hz; acquisition time, 2.38 s; flip angle, 90°; relaxation
delay, 1 s; temperature, 298 K. Two-dimensional COSY experiments
were performed on the purified metabolites to improve signal assignment
of the complex aromatic region. The parameters for the COSY experiments
were as follows. The number of scans per increment was 16, the spectral
width was 3881.32 Hz, and 1024 increments were performed in the F1
dimension. The free induction decays were collected into
1-kilobyte computer data points. The relaxation delay between
successive pulses was 1.5, with no zero-filling in F2. Unshifted
sinebell windows were applied before transformation.
| |
Results |
WIN55212-2 was subjected to enzymatic breakdown using rat liver
microsomes as outlined in the experimental section. Rat liver microsomes were chosen for their convenience and availability, and they
are considered to engage in metabolic processes conserved in a broad
spectrum of mammalian species.
HPLC-UV.
Shown in Fig. 1 is the HPLC chromatogram
of the products of microsomal incubation employing UV absorbance
detection. The parent compound, WIN55212-2 elutes at 29.1 min.
Preceding peaks observed in the chromatogram that are absent in control
incubation products and the standard solution of WIN55212-2 are
considered possible metabolites formed from microsomal incubation.
Thus, peaks at 14.8, 16.0, 16.5, 19.3, 19.9, 24.6, 25.0, and 25.4 min
are possible metabolic products. Peak areas of the 19.3 and 19.9 min
metabolites constitute 60 to ~75% of the total metabolite peak
areas. It is noted that two chromatographic peaks (31.8 and 44.8 min)
that elute after the parent compound are also present in control/blank samples, indicating that they are likely substances introduced by the
NADPH-regenerating solutions. However, no degradation product of
WIN55212-2 was found in any control incubations.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 1.
HPLC-UV chromatogram of a rat microsomal
incubation product of WIN55212-2.
Sample was collected at 4 h from the start of incubation. The
identified metabolites are labeled M1 through M8 in the order of
elution, and the parent compound is labeled as WIN55212-2.
|
|
HPLC-MS/MS and NMR Spectra of WIN55212-2.
To assist in the identification of metabolites that usually contain
diagnostic fragment ions that are related to those of the parent
compound, the product ion spectrum of the protonated parent compound,
WIN55212-2, was first acquired as shown in Fig. 2. In the conventional ESI mass spectrum,
intact protonated WIN55212-2 (FW 426) was observed at
m/z 427 as the dominant peak. Subsequently, this
ion was isolated by the first quadrupole (Q1) as the precursor for
collision induced dissociations. The fragment ions generated as a
result of collision with Ar gas in the radio frequency only hexapole (Q2) collision cell were then mass analyzed by the third quadrupole (Q3). The three characteristic fragment ions from WIN55212-2 are noted at m/z 155, 127, and 100. As
illustrated in Fig. 2, the fragment ion observed at
m/z 155 corresponds to moiety a via a
cleavage at the carbonyl carbon bonded to the indole ring. This fragment ion can then lose a CO molecule to yield a
resonance-stabilized ion at m/z 127. In a
separate fragmentation process, the morpholine moiety can also be
cleaved, resulting in a fragment ion (moiety c) at
m/z 100. These three ions were subsequently used
as diagnostic fragment ions for identification of metabolites, the
product ion spectra of which may contain one or more of the same
fragments or those that have undergone hydroxylation or other oxidative metabolism.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 2.
MS/MS spectrum obtained by collision induced
dissociation of WIN55212-2 [M + 1]+ ion and proposed
fragmentation pathway of WIN55212-2.
|
|
The 1H NMR chemical shifts and the
respective coupling patterns of WIN55212-2 and its metabolites M4, M5
obtained in the present investigation are summarized in Table
1. The numbering scheme for the protons on the WIN55212-2 structure is given in Fig. 2. All
proton assignments were derived from one dimensional and COSY experiments. Of the 10 aromatic protons in the molecule ( 6.4-8.0 ppm), two protons ( 6.43 and 6.64 ppm) show correlation peaks with the proton at 6.83 ppm, and these three protons show no correlation peaks with any other protons in the molecule. The values and the correlation pattern indicate that these are the three
aromatic protons on the indole ring and are assigned as follows: 6.83 ppm for proton I, 6.43 ppm for proton H,
and 6.64 ppm for proton J. The remaining seven aromatic
proton signals are assigned to the naphthalene ring. The COSY
experiments show that the two protons at 7.58 and 7.99 ppm give
rise to correlation peaks with the proton at 7.54 ppm, and the
three protons have no correlation peaks with the other four protons in
this region. Thus, 7.58, 7.54, and 7.99 ppm are the signals
from aromatic protons A, B, and C. For
the last four aromatic protons, correlation peaks are observed between
the two protons at 7.93 and 7.44 ppm and the proton at 7.51ppm, and between the proton at 7.44 ppm and the proton at 8.08 ppm. These correlations allow the assignments of the last four
protons as follows: 7.93 ppm for proton D, 7.44 ppm
for proton F, 7.51 ppm for proton E, and 8.08 ppm for proton G. All aromatic protons show coupling
patterns that are consistent with the above proton assignments
(J = 8 and 1.2 Hz).
In the aliphatic region (between 2.5 and 4.8 ppm), integration
of the singlet at 2.59 ppm indicates the presence of three protons,
linking the single peak to the methyl protons (R) on the
indole ring. The signals from the eight protons on the morpholine moiety are accounted for by protons at 2.57 ppm and at
3.73 ppm, which show correlation peaks with each other. The four
protons at 2.57 ppm are most likely the two
CH2 (N and Q) next to the
nitrogen atom, the other four at 3.73 ppm being the two
CH2 (O and P) next to the
oxygen. The proton at 2.76 ppm is coupled with the proton at 2.50 ppm, both of which are also coupled to the proton at 4.67 ppm.
Thus, 2.76 and 2.50 ppm proton peaks are those of the
CH2 (M) between indole and morpholine rings, and 4.67 ppm is assigned to the proton at the L
position. Clearly, the splitting of the CH2
(M) protons into a doublet reflects the chemical
nonequivalence of the two protons caused by the neighboring chiral
carbon of L. The remaining two aliphatic signals at 4.84 (1H) and 4.22 ppm (1H) are coupled to each other with a coupling constant of 11.6 Hz (clearly a germinal coupling), and they may be
assigned to the two protons on CH2 (K)
bonded to an oxygen atom.
HPLC-MS/MS Spectra of WIN55212-2 Metabolites.
Shown in Fig. 3E are the total ion
chromatogram obtained from an incubation product mixture and the
selected ion chromatograms for m/z 425, 443, 461, and 477 all reconstructed from a single run. The rationale for the
structural characterization of each proposed metabolite is given below
in the order of increasing retention time.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 3.
HPLC-MS chromatograms of WIN55212-2
metabolites from rat microsomal incubation: A, SIM chromatogram
of m/z 477; B, SIM chromatogram of m/z 461; C, SIM chromatogram of m/z
443; D, SIM chromatogram of m/z 425; and E, total ion chromatogram
(TIC). SIM, selected ion monitoring.
|
|
The selected ion chromatogram for m/z 477 is
shown in Fig. 3A, in which three peaks are observed at retention times
of 14.4, 15.9, and 17.0 min, respectively. The well resolved peaks
indicate three metabolites, designated as M1, M2,
and M3 that have the same molecular weight but differ in
positions of hydroxylation. These metabolites show a common [M + H]+ ion at m/z 477, 50 amu
higher than the [M + H]+ ion of WIN55212-2. It
is proposed that M1, M2, and M3 are
likely the products of trihydroxylation of WIN55212-2. The product ion
spectra of M1, M2, and M3 are
identical, and the spectrum is shown in Fig.
4. A comparison with the product ion
spectrum of WIN55212-2 and its proposed fragmentation pathways (Fig. 2)
suggests that the fragment ions at m/z 171 and 143 are related to the WIN55212-2 fragment ions at
m/z 155 and 127, respectively. The mass
difference of 16 suggests the involvement of a hydroxyl group on the
naphthyl moiety. The fragment ion at m/z 189 indicates the presence of two hydroxyl groups on the moiety a. Finally, the fragment ion at m/z
100 provides evidence that M1, M2, and
M3 all have an unaltered moiety c, leaving the
third hydroxyl group at moiety b. The results discussed above led us to propose the fragmentation pathways as illustrated in
Fig. 4.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 4.
MS/MS spectrum obtained by collision-induced
dissociation of the protonated ions at m/z 477 (M1, M2, and M3) and
their proposed fragmentation pathways.
|
|
Figure 3B is the selected ion chromatogram extracted for the ion at
m/z 461, in which two distinct chromatographic
peaks are observed, suggesting two isomeric metabolites (assigned as
M4 and M5) with hydroxylation occurring at
different positions. The daughter ion spectra for the protonated ions
of M4 and M5 are shown in Fig.
5, in which identical fragment ions with
varying relative abundances are noted. The protonated M4 and M5 ions,
both at m/z 461, are 34 amu higher than the
protonated WIN55212-2; thus, dihydrodiol products differing in sites of
hydroxylation are proposed for M4 and M5. Again,
fragment ions at m/z 189, 171, 143, and 115 suggest that the two hydroxyl groups are within moiety a.
Also illustrated in Fig. 5 are the proposed fragmentation mechanisms
for M4 and M5.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 5.
Product ion spectra obtained by collision
induced dissociation of the protonated ion (m/z 461) of M4 (A) and M5
(B) and their proposed fragmentation pathways.
|
|
The structural elucidation of M4 and M5 is also assisted by NMR
studies. From the 1H NMR spectra of M4 and M5
(Table 1), integration of the aromatic (and alkene, once dihydrodiol
products are formed) region give only 8 proton signals (as compared
with 10 found in WIN55212-2), indicating that hydroxylations have taken
place in this region. Whereas the chemical shifts of protons
H, I, and J remain in the aromatic
region, two new signals emerge in the aliphatic region. A comparison
with the NMR spectrum of WIN55212-2 shows that all original aliphatic
protons are found intact in both M4 and M5,
providing further evidence that the two metabolites have the intact
moiety c, consistent with the mass spectral findings (i.e.,
presence of the fragment ion at m/z 100 in the
product ion mass spectra of M4 and M5).
In the NMR spectra of M4, proton signals at 5.92 and 6.57 are coupled to each other with a coupling constant of 10 Hz (Table
2). The newly emerged aliphatic signals
at 4.87 (1H, J = 10.4 Hz) and 4.56 (1H,
J = 10.4 Hz) are also coupled to each other whereas the
4.56 proton shows additional coupling with proton signals at 5.92 and 6.57. These observations strongly suggest that
dihydroxylation has taken place on two adjacent naphthyl carbons
resulting in some loss in aromaticity. Moreover, the remaining two
alkene protons ( 5.92 and 6.57) should also be ortho to one
another and adjacent to one of the new aliphatic signals ( 4.56).
Thus, the possible hydroxylation sites are either D and E or F and G. Because all chemical
shifts of the aromatic protons on the indole ring of M4
remain unchanged compared with those of WIN55212-2, the only possible
dihydroxylation sites are D and E. Had the
hydroxylations taken place on F and G, the
chemical shift of the aromatic proton H on the indole ring
would have shifted to a lower field due to the effect of the OH group
on the G carbon (the two bonds connecting the naphthalene
ring, carbonyl group, and indole ring can both rotate freely).
As summarized in Table 2, the 1H NMR spectrum of
M5 also contains two newly emerged aliphatic signals at 4.99 ppm (1H, J = 10.8 Hz) and 4.67ppm (1H,
J = 10.8 Hz) coupled to each other. The proton
at 4.67 is also coupled to a single alkene proton signal at 6.22 ppm (J = 2.4 Hz). The coupling patterns suggest
that, like in M4, the two hydroxylation sites are ortho to
each other, and that unlike M4 only one alkene proton ( 6.22 ppm) is observed, the coupling pattern of which indicates that it
is adjacent to one of the new aliphatic proton signals ( 4.67).
Thus, the possible dihydroxylation positions are either A
and B or B and C. In contrast to the
case of M4, the chemical shift of the aromatic proton at
H on the indole ring does show a shift to the lower field
(from 6.43 to 7.30 ppm) as a result of interaction between the
proton at H and the OH group at A. Based on the
above information, the dihydroxylation positions are A and
B for M5.
At least two metabolic products exhibit an [M + H]+ ion at m/z 443, identified as M6 and M7 (Fig. 3C). At 16 atomic
mass units higher than the protonated ion of WIN55212-2, M6
and M7 are presumed to be two isomeric monohydroxylation products (OH substitution for H). As shown in Fig.
6B, the product-ion spectrum of
[M7 + H]+ at
m/z 443 yields fragment ions at
m/z 171 and 143, which are analogous to fragment
ions from WIN55212-2 at m/z 155 and 127, respectively, and indicate that monohydroxylation occurs on the naphthyl ring. It is confirmed by the observation of the intact moiety
c at m/z 100. However, it is noted
that the chromatographic peak representing M7 may contain
two or more unresolved isomeric metabolites because of possible
positional isomers of the ring hydroxylation.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 6.
Product ion spectra of the protonated M6(A)
and M7(B) ions (m/z 443) and their proposed fragmentation pathways.
|
|
Figure 6A shows the product-ion spectrum of M6, also
observed at m/z 443, in which fragment ions at
m/z 155, 127, and 100 are the same as those from
WIN55212-2. This monohydroxylated metabolite differs from M7
in that the site of hydroxylation is within moiety b. The
fragmentation pathways for M6 and M7 are
presented in Fig. 6A and 6B, respectively.
Figure 3D represents yet another metabolite, M8 with [M + H]+ ion observed at m/z
425, which is exactly 2 amu lower than the [M + H]+ ion of WIN55212-2. A comparison of the
product ion spectrum of M8 (Fig. 7) and
WIN55212-2 (Fig. 2) reveals that the fragment ion at
m/z 98 in Fig. 7 is also 2 amu lower than the
fragment ion at m/z 100 in Fig. 2, whereas all
other fragment ions are identical. Moreover, a minor fragment ion
appears at m/z 340, that is also present in the
product ion spectrum of WIN55212-2 (Fig. 2). Formation of
m/z 340 from M8 results from the loss
of a mass 85 neutral instead of 87 (from WIN55212-2). The above mass
spectral information indicates that moiety c of WIN55212-2 has been dehydrogenated to yield M8. The proposed
fragmentation pathways of M8 are given in the same Figure.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 7.
MS/MS spectrum of the protonated M8 (m/z
425) obtained by collision-induced dissociation and its proposed
fragmentation pathway.
|
|
| |
Discussion |
WIN55212-2 can undergo microsomal metabolic transformation to
yield several metabolites as detailed in Fig.
8. Based on HPLC and tandem mass
spectrometric analysis, eight different oxidative metabolites have been
identified. In one metabolic pathway, the product (M8) is
believed to have formed a carbon-carbon double bond in the morpholine
moiety (moiety c). Dehydrogenation metabolites have been
reported for the biotransformation of piperitenone in rats where ring
hydroxylation can be followed by dehydration to yield a double bond
(Madyastha and Gaikwad, 1999). M8 may have been formed via
an intermediate hydroxylation product on the morpholine ring. However,
it is noted that no trace of the precursor hydroxylation product has
been observed in LC chromatograms.
Two monohydroxylation metabolites have been identified based on
chromatography and tandem mass spectrometry analysis. In one case,
where hydroxylation takes place within moiety b, a
metabolite with molecular weight of 442 results (M6). The
exact site of hydroxylation cannot be ascertained because the fragment
information from the product ion spectrum of this metabolite is not
conclusive. Another monohydroxylation metabolite (M7) with
the same molecular weight of 442 was observed, the protonated ion of
which yields a product ion spectrum different from that of
M6 in that hydroxylation occurs in the aromatic naphthyl
ring for M7. The metabolic pathways for the two
monohydroxylation metabolites, however, are believed to be entirely
different. M6 is proposed to be the result of hydroxylation
within moiety b, whereas M7 is proposed to be a
rearrangement product of the epoxide intermediates that also lead to
dihydrodiol metabolites upon epoxide hydrolase action as discussed
later. Interestingly, the chromatographic peak corresponding to
M7 is not well resolved, suggesting the presence of possibly
two or more isomers, resulting from two isomeric epoxide intermediates
as illustrated in Fig. 8.
Dihydroxylation metabolites have been identified as M4 and
M5, the only two metabolites that were isolated and purified
in sufficient quantities for further structural studies by NMR.
M4 and M5 are characterized by the formation of a
dihydrodiol functional group at differing positions on the naphthyl
ring. The formation of such diols is most likely initiated by an
epoxidation process on the aromatic ring. The arene oxide mechanism was
first proposed by Jerina and coworkers (Guroff et al., 1967, Jerina et
al., 1968). This metabolic pathway involves the formation of an epoxide
on an aromatic ring, which can either be hydrolyzed to yield a
dihydrodiol product or undergo spontaneous rearrangement to give a
phenolic product (Daly et al., 1972). In the present study, it appears
that M4 and M5 are not metabolic products from
further hydroxylation of M7; rather, all three metabolites
are proposed to originate from the common epoxide intermediates via
either spontaneous rearrangement or the action of epoxide hydrolase.
At least three trihydroxylation metabolites, M1,
M2, and M3 have been identified that are believed
to be the products of further hydroxylation of M4 and
M5. Whereas there are three peaks, the product ion spectra
of which indicate that the third hydroxyl group must occur within
moiety b containing the indole ring structure, it should be
pointed out that there may be more than three isomeric trihydroxylated
metabolites that couldn't be resolved under the present HPLC
conditions. Finally, these three metabolites are not considered to be
products arising from further metabolism of the monohydroxylated
M6 via an epoxide mechanism, because if that were the case,
additional dihydroxylated metabolites ([M + H]+, m/z 459) should have
been observed, resulting from spontaneous rearrangement of the epoxide
intermediates of M6.
In summary, this study shows that WIN55212-2, a potent CB1
agonist that bears little structural similarity to classical
cannabinoids, indeed undergoes distinct metabolic pathways. For
example, the major microsomal metabolite of
1-tetrahydrocannabinol in rat is
7-hydroxy-1-tetrahydrocannabinol, a product of
alkyl hydroxylation (Harvey and Paton, 1984; Agurell et al., 1986). The
major metabolic pathway of WIN55212-2, however, is predominantly via
arene oxide formation to give dihydrodiols. Such metabolic differences
may have implications on the extent to which cannabimimetic properties
are retained or removed in WIN55212-2 metabolites. With two major
metabolites isolated in pure form, further physiological studies are
now possible to determine their agonistic activities.
Financial support was provided by the National Institute on
Drug Abuse through a research Grant DA07970, by National Institutes of
Health-Minority Biomedical Research Support through GM08008, and by the
National Science Foundation through CHE-9981948.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
