Identification of a 3-hydroxylated tacrine metabolite in rat and man: Metabolic profiling implications and pharmacology

Article abstract of DOI:10.1021/jm9602266

Discrepancies in urinary metabolic profiles in rats administered tacrine (1) suggested the presence of an unidentified metabolite of 1. Chromatographic methods were developed that allowed isolation of a metabolite fraction containing both 1-hydroxytacrine

Full text of DOI:10.1021/jm9602266

3014
J . Med. Chem. 1996, 39, 3014-3018
Id en tifica tion of a 3-Hyd r oxyla ted Ta cr in e Meta bolite in Ra t a n d Ma n :
Meta bolic P r ofilin g Im p lica tion s a n d P h a r m a cology1
William F. Pool,*, Thomas F. Woolf, Michael D. Reily,§ Bradley W. Caprathe,‡ Mark R. Emmerling,† and
J uan C. J aen‡
Departments of Pharmacology, Chemistry, Analytical Research, and Pharmacokinetics and Drug Metabolism,
Parke-Davis Pharmaceutical Research, Division of Warner-Lambert Company, 2800 Plymouth Road,
Ann Arbor, Michigan 48105
Received March 21, 1996X
Discrepancies in urinary metabolic profiles in rats administered tacrine (1) suggested the
presence of an unidentified metabolite of 1. Chromatographic methods were developed that
allowed isolation of a metabolite fraction containing both 1-hydroxytacrine (2) and an unknown
metabolite from rat urine. Mass spectral analysis indicated this metabolite to be a monohy-
droxylated derivative, which upon two dimensional COSY NMR analysis could be assigned as
3-hydroxytacrine (4). This structural assignment was confirmed by independent synthesis of
4. Compound 4 was also identified as a human urinary metabolite of 1. Biologically, 4 was
found to have in vitro human red blood cell acetylcholinesterase inhibitory activity similar to
that of 2 and 4-hydroxytacrine (5) and approximately 8-fold less than that of 1. These results
underscore the need to conduct rigorous structural identification studies, especially in cases
where isomeric metabolites are possible, in assessing the accuracy of chromatographic profiling
techniques.
Sch em e 1. Metabolism of 1 in Rat
In tr od u ction
Tacrine (1, Cognex) is a cholinesterase inhibitor with
demonstrated efficacy in treating patients with mild to
moderate Alzheimer’s disease (AD),1,2 a condition as-
sociated with decreased central cholinergic function.3,4
In addition to gastrointestinal side effects, tacrine
administration to AD patients has associated a rela-
tively high incidence, up to 50% of patients treated, of
asymptomatic elevations of liver marker enzymes (ala-
nine aminotransferase (ALT)).5-7 The metabolic fate of
1 in animals and man is characterized by rapid and
extensive oxidative metabolism with marked differences
in species plasma and urinary metabolic profiles.8-12
Clearly, identification of these metabolites and assessing
their pharmacological properties are necessary to gain
insight into the contribution of metabolites to the overall
pharmacological profile of 1. While investigating the
urinary metabolic profile of 1 in rat, we observed a
difference in relative amounts of 1-, 2-, and 4-hydroxy-
tacrine metabolites (2, 3, and 5, respectively) (Scheme
1) quantitated using our analytical method compared
to results reported by Hsu et al.12 This observed differ-
ence could not be explained by dose, route of adminis-
tration, or collection period. Hsu et al. found 2 and 3 as
major urinary excretion products, while we observed 2
and 5 as major urinary metabolites. To rationalize this
apparent discrepancy, we conducted a series of chro-
matographic and spectral investigations ultimately
leading to the identification and characterization of a
new tacrine metabolite, namely, 3-hydroxytacrine (4).
The identification and synthesis of 4 and its cholin-
esterase inhibition properties are described below.
Ch em istr y
The synthesis of 3-hydroxytacrine (4) is outlined in
Scheme 2. The general synthetic strategy utilized by
Shutske13 for the synthesis of dihydroxytacrine deriva-
tives was applied to the synthesis of 7. The tetrahy-
droaminoacridine ring system was built with a phenyl-
dimethylsilyl group as a masked form of the desired
hydroxy functionality. Attempts to build the ring
system with a protected alcohol in the 3-position con-
sistently led to the formation of 1,2-dihydro-9-amino-
acridine (an elimination byproduct) as the only isolated
product. Briefly, 1,3-dione 714 was condensed with
anthranilonitrile under Dean-Stark conditions to give
8 in 47% yield. Cyclization of 8 was accomplished with
zinc chloride as catalyst to produce 9 in 80% yield.
Reduction of the carbonyl group of 9 was efficiently
t
accomplished with BuNH2-BH3 and AlCl3 in CH2Cl2
o
at 0 C.15 Conversion of the phenyldimethylsilyl group
of 10 into the desired hydroxy group was accomplished
by a two-step sequence involving treatment of 10 with
HBF4‚Et2O to give crude 11 in 60% yield and oxidation
of 11 with basic H2O2/KF to give the desired compound
4 in 22% yield for the final step.
† Department of Pharmacology.
‡ Department of Chemistry.
§ Department of Analytical Research.
Department of Pharmacokinetics and Drug Metabolism.
X Abstract published in Advance ACS Abstracts, J uly 1, 1996.
S0022-2623(96)00226-9 CCC: $12.00 © 1996 American Chemical Society

Identification of 3-Hydroxylated Tacrine Metabolite
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 3015
Sch em e 2. Synthetic Route to 4a
a
t
Reagents: (a) p-TsOH‚H2O, benzene, reflux; (b) ZnCl2, toluene, reflux; (c) AlCl3, BuNH2‚BH3, CH2Cl2, 0 °C; (d) HBF4‚Et2O, CH2Cl2,
room temperature; (e) H2O2, KF, NaHCO3, 0 °C.
F igu r e 2. Representative HPLC radioactivity chromatogram
of 0-4 h urine from rats administered a single oral 18 mg/kg
dose of [14C]1 using method B.
F igu r e 1. Representative HPLC radioactivity chromatogram
of 0-4 h urine from rats administered a single oral 18 mg/kg
dose of [14C]1 using method A.
amounts of 3 and 5 (Figure 2). Again, unchanged 1 was
present in lesser amounts.
Bioch em istr y
The differences in metabolic profile between methods
A and B with respect to the amounts of 5 and 2 present
in rat urine suggested to us to examine the correspond-
ing isocratic and gradient chromatographic peaks by
thermospray (TSI)-LC/MS analysis. The relatively
“soft” sample ionization technique of TSI-LC/MS re-
sulted in abundant formation of protonated molecular
ions at m/ z 215 corresponding to monohydroxylated
derivatives of 1 at retention times consistent with
standards of 2 and 5. There was no evidence for any
additional metabolites of 1 in the peaks corresponding
to 2 and 5 using either method A or B by TSI-LC/MS.
These chromatographic and spectroscopic results sug-
gested that an unknown monohydroxylated derivative
of 1 was coeluting with either 5 or 2, depending on
whether method A or B, respectively, was used.
In order to investigate the relative contribution of 2-5
to the pharmacological action of 1, their inhibition of
red blood cell acetylcholinesterase from mouse, rat,
monkey, and man was evaluated using a modified
Ellman assay.16
Resu lts a n d Discu ssion
Bioa n a lytica l Da ta . Analysis of synthetic standards
of 2, 3, and 5 as well as 1, individually and as a mixture,
by an isocratic chromatographic method (method A)
resulted in an elution sequence of 3, 5, 2, and 1. HPLC
radioactivity profiling using method A of a directly
injected aliquot of 0-4 h urine from a rat administered
a single 18 mg/kg oral dose of [14C]1 showed a series of
radiolabeled peaks corresponding to 3, 5, and 2 and a
peak with identical retention properties as 1 (Figure 1).
Metabolites 5 and 2 were present in greatest amounts.
These results differed from those reported by Hsu et
al.12 where 3 and 2 were found as major tacrine-derived
metabolites in rat urine. Given similarities in the dose,
route of administration, and collection periods, we
surmised that this discrepancy may have been the result
of differences in HPLC methods where Hsu et al.12 used
a phenyl column while we used a cyano column. These
differences in the urinary metabolic profiles led us to
develop a C-18 reversed-phase gradient HPLC method
(method B) where the elution order was 3, 2, 5, and 1.
HPLC radioactivity profile analysis of a directly injected
aliquot of 0-4 h rat urine from the same animals
revealed 2 as the sole major metabolite with minor
To further elucidate the structure of this unknown
metabolite, 0-4 h urine from rats dosed orally with 2
mg/kg (13 µCi) [14C]1 was extracted and fractionated
by semipreparative HPLC (method C). A fraction with
similar elution properties as 2 was collected and con-
centrated. An aliquot of this isolate was chromato-
graphed on the analytical gradient system (method B)
and shown to be one peak coeluting with 2. The isolate
fraction was dissolved in deuteriomethanol and sub-
1
jected to H NMR analysis.
The various metabolites resulting from monohy-
droxylation of the aliphatic ring portion of tacrine are
distinguishable by the chemical shifts of the hydroxyl-
ated methine proton and the remaining methylene
1
protons. The H NMR spectrum of the isolate fraction

3016 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Pool et al.
F igu r e 3. 1H NMR spectra (500 MHz) of major metabolites
of 1 from rat and human. Bottom: human urine metabolite
fraction. Top: rat urine metabolite fraction. Inset: expansion
of the region near water in the rat metabolite spectrum after
resolution enhancement and spline base-line correction to
resolve the signal at 4.95 ppm. The numbering without and
with asterisks refers to 2 and 4, respectively. Solution and
spectral conditions are given in the material and methods
section.
F igu r e 4. Aliphatic regions of the 500 MHz two-dimensional
COSY spectra of 2 and 4: A, rat urine metabolite fraction; B,
human urine metabolite fraction; C, 2; D, 4. Cross-peaks
critical for assignment of the site of hydroxylation are indicated
on the plots. Solution and spectral conditions are given in the
material and methods section.
showed the presence of two multiplet resonances at δ
4.29 and 4.95 in the chemical shift region expected for
hydroxy-bearing methines (Figure 3, top trace). The
signal resonating at δ 4.95 and the chromatographic
retention time of the isolate were consistent with 2. The
signal resonating at δ 4.29 was close to the known
chemical shift of the 2-CH in 3 at δ 4.34. The identity
of this unknown as 3 was unlikely based on chromato-
graphic differences, and results from a titration experi-
ment where 3 was added to the rat isolate produced a
shoulder on the nonresolved multiplet at δ 4.29 as well
as an additional set of 1-methylenes. These peaks
increased in intensity with increasing amounts of added
3. The identity of the unknown as 5 was also ruled out
based on its chromatographic retention properties and
distinguishing chemical shifts, most notably the C4
proton at δ 4.89. Since mass spectral data indicated
that the unknown was a monohydroxylated derivative
of 1, it was suspected that the methine at δ 4.29 came
from 4. To further elucidate the structure of the
hydroxy isomer, two-dimensional COSY spectra were
obtained that allowed correlation of the methine protons
with the methylenes on adjacent carbons (Figure 4A).
This spectrum showed the expected coupling patterns
for 2 and 4.
HPLC radioactivity profiling of urine obtained from
human subjects administered a single oral 40 mg dose
of (100 µCi) [14C]1 using method B showed 1 to be
extensively metabolized with no single major metabolite
(Figure 5). A 0-4 h aliquot of urine was subjected to
the above extraction and purification techniques to
isolate the component coeluting with 2. The isolate was
dissolved in deuteriomethanol and subjected to 1H NMR
analysis. The 1H NMR spectrum of the isolated human
metabolite was similar to that observed in rat and is
shown in the bottom trace of Figure 3. Two-dimensional
COSY spectra recorded on the human isolate resulted
in coupling patterns consistent with the presence of 2
and 4 in the isolate and are shown in Figure 4B.
Comparative two-dimensional COSY experiments run
on synthetic 2 and 4 confirmed that the coupling
patterns were identical with those seen in the isolated
metabolite fractions (Figure 4C,D).
F igu r e 5. Representative HPLC radioactivity chromatogram
of 0-4 h urine from human subjects administered a single oral
40 mg dose of [14C]1 using method B. HPLC peak eluting at
53 min coelutes with 2. Compound 5 coelutes with peak eluting
at 54.5 min.
Bioch em istr y. In order to investigate the relative
contribution of 2-5 to the pharmacological action of 1,
in vitro IC50 values for the inhibition of red blood cell
acetylcholinesterase were obtained for 1 and the mono-
hydroxylated metabolites in mouse, rat, monkey, and
man (Table 1). The IC50 for 4 was 8-fold higher than
that of 1 in the human preparation and similar in
magnitude to that determined for 2 and 5. Metabolite
3 displayed the weakest inhibitory activity of the
hydroxylated regiomers in all preparations examined.
Con clu sion
A new monohydroxylated metabolite of 1, 4, was
identified in urine from rats and humans following oral
1 administration. This metabolite, depending on chro-
matographic conditions, can coelute with either 2 or 5.
Interestingly, the in vitro AchE inhibition activities of
4 are similar to those of 2 and 5 and differ significantly
from that of the less potent regiomer 3. Whether or not
this new metabolite significantly contributes to the
central cholinergic effects of 1 in vivo is not known.

Identification of 3-Hydroxylated Tacrine Metabolite
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 3017
Ta ble 1. IC50 Values (µM ( SE) for the Inhibition of Red Blood Cell Acetylcholinesterasea
1
2
3
4
5
human
monkey
rat
0.078 ( 0.005
0.075 ( 0.004
0.046 ( 0.003
0.143 ( 0.010
0.702 ( 0.054
0.954 ( 0.023
0.361 ( 0.023
1.000 ( 0.019
1.61 ( 0.137
2.30 ( 0.078
1.77 ( 0.097
4.82 ( 0.085
0.623 ( 0.035
0.759 ( 0.036
0.483 ( 0.072
1.210 ( 0.164
0.521 ( 0.017
0.476 ( 0.025
0.411 ( 0.012
1.43 ( 0.074
mouse
a
Inhibition assays were performed as described in ref 16.
min until a clear solution resulted. A solution of 4.48 g (12.93
mmol) of 9 in 150 mL of CH2Cl2 was added dropwise. The
reaction mixture was allowed to slowly warm to room tem-
perature overnight. The fluorescent yellow solution was
concentrated, the reaction carefully quenched with dropwise
addition of ice-cold water, and the mixture basified with
concentrated NH4OH to pH ∼9-10 and extracted into CHCl3.
The CHCl3 extract was dried (MgSO4), filtered, and concen-
trated, and the crude foamy yellow solid was chromatographed
(CHCl3:MeOH:NH4OH, 90:10:1) to give 3.54 g (82%) of 10 as
a light yellow solid, mp 141-150 °C. 1H NMR (DMSO-d6):
0.33 (6H, d, J ) 1.4 Hz), 1.23-1.28 (1H, m), 1.44-1.48 (1H,
m), 2.02-2.07 (1H, m), 2.42-2.45 (1H, m), 2.58-2.72 (2H, m),
2.76-2.82 (1H, m), 6.30 (2H, s), 7.26 (1H, dd, J ) 8 Hz), 7.38-
7.41 (3H, m), 7.46 (1H, dd, J ) 8 Hz), 7.55-7.60 (3H, m), 8.12
(1H, d, J ) 8 Hz). MS (CI): 333 (M + 1, 100). Anal.
(C21H24N2Si) C, H, N.
(()-9-Am in o-3-h yd r oxy-1,2,3,4-tetr a h yd r oa cr id in e (4).
The two-step general procedure of Oliver et al.14 for the
replacement of PhMe2Si by OH was followed. A solution of
2.51 g (7.55 mmol) of 10 in 150 mL of CH2Cl2 was treated with
10.0 mL (57.7 mmol) of 85% tetrafluoroboric acid-diethyl ether
complex. A solid immediately formed. After 2 h at room
temperature, all solids had gone into solution. Another portion
of 5.0 mL of 85% tetrafluoroboric acid-diethyl ether complex
was added; the solution was stirred for 2 h, concentrated,
suspended in ice-cold water, basified with saturated NaHCO3
solution, and extracted into CHCl3. The CHCl3 extract was
washed with brine, dried (MgSO4), filtered, and concentrated
to give 1.24 g (60%) of crude 11 as a white-yellow solid. This
material was used without further purification.
These results underscore the need to conduct rigorous
structural identification studies, especially in cases
where isomeric metabolites are possible, in assessing
the accuracy of chromatographic profiling techniques.
Exp er im en ta l Section
Gen er a l Meth od s. Air- or moisture-sensitive reactions
were carried out in flame-dried glassware under an atmo-
sphere of nitrogen or argon. Organic solutions were dried over
anhydrous MgSO4 and concentrated under reduced pressure
on a rotary evaporator. Thin-layer chromatography (TLC) was
carried out on E. Merck silica gel 60-F254 precoated glass
plates (0.25 mm thickness). Medium pressure liquid chroma-
tography (MPLC) was performed with E. Merck silica gel 60,
230-400 mesh ASTM. Melting points were determined on a
Thomas Hoover capillary melting point apparatus and are
uncorrected. 1H NMR spectra were recorded on a Varian
Unityplus-400 NMR spectrometer. Chemical shifts are re-
ported in ppm downfield from tetramethylsilane (internal
standard). Mass spectra were recorded on a Finnigan 4500
mass spectrometer or a VG analytical 7070E/HF mass spec-
trometer; the spectra are described by the molecular peak (M)
and its relative intensity as well as the base peak (100%).
Elemental analyses were performed on a CEC 240XA elemen-
tal analyzer. Where analyses are indicated by the symbols of
the elements, the results are within 0.4% of the theoretical
values. The syntheses of 2, 3, and 5 were accomplished using
the procedures detailed by Hsu et al.,12 while the syntheses of
[14C]1 and [14C]2 are described by McNally et al.17
Syn th etic Meth od s. N-[5-(Dim eth ylp h en ylsilyl)-3-oxo-
cycloh exen -1-yl]-2-a m in oben zon itr ile (8). A solution of 30
g (0.10 mol) of crude 5-(dimethylphenylsilyl)cyclohexane-1,3-
dione (7),14 14.1 g (0.10 mol) of anthranilonitrile (6), and 3.0 g
of p-toluenesulfonic acid monohydrate in 500 mL of benzene
was refluxed with a Dean-Stark trap for 2 h. The solution
was cooled, concentrated in vacuo, dissolved in EtOAc, and
washed with saturated NaHCO3 solution and brine. The
EtOAc extract was dried (MgSO4), filtered, concentrated, and
chromatographed (MPLC: gradient 50-75% EtOAc in hexane)
to give 19.3 g (47%) of 8 as a yellow solid. A portion was
recrystallized from EtOAc-hexanes to give a light yellow solid,
mp 127-130 °C. 1H NMR (DMSO-d6): 0.33 (6H, d, J ) 6.8
Hz), 1.58-1.67 (1H, m), 2.01-2.11 (2H, m), 2.37-2.48 (2H,
m), 4.84 (1H, s), 7.38-7.42 (5H, m), 7.55-7.57 (2H, m), 7.71
(1H, dd, J ) 8 Hz), 7.85 (1H, d, J ) 8 Hz), 9.03 (1H, s). MS
(CI): 347 (M + 1, 100). Anal. (C21H22N2OSi) C, H, N.
(()-9-Am in o-3-(d im eth ylp h en ylsilyl)-3,4-d ih yd r oa cr i-
d in -1(2H)-on e (9). A suspension of 10.0 g (28.86 mmol) of 8
and 11.80 g (86.59 mmol) of ZnCl2 in 750 mL of toluene was
refluxed under N2 for 12 h, with mechanical stirring. The
mixture was concentrated and partitioned between CHCl3 and
10% NH4OH. The CHCl3 extract was dried (MgSO4), filtered,
concentrated, and chromatographed (MPLC: EtOAc:hexane,
1:1) to give 8.03 g (80%) of 9 as a light yellow solid. A portion
was recrystallized from EtOAc-hexanes to give a white solid,
mp 184-187 °C. 1H NMR (DMSO-d6): 0.35 (6H, d, J ) 3.1
Hz), 1.68-1.76 (1H, m), 2.48-2.54 (2H, m), 2.77-2.90 (2H,
m), 7.37-7.44 (4H, m), 7.54-7.59 (2H, m), 7.63-7.70 (2H, m),
8.31 (1H, d, J ) 8.2 Hz), 8.37 (1H, broad s), 9.92 (1H, broad
s). MS (CI): 347 (M + 1, 100). Anal. (C21H22N2OSi‚0.25H2O)
C, H, N.
To an ice-cold suspension of 1.07 g (3.9 mmol) of crude 11,
2.1 g (36.1 mmol) of KF, and 2.6 g (30.9 mmol) of NaHCO3 in
30 mL of MeOH:THF (1:1) was added dropwise 13.5 mL (132.1
mmol) of 30 wt % H2O2 solution in water. The mixture was
stirred at room temperature for 12 h, filtered through a pad
of Celite, concentrated, and partitioned between 10% NH4OH
and CHCl3. The CHCl3 extract was dried (MgSO4), filtered,
concentrated, and chromatographed by MPLC (CHCl3:MeOH:
NH4OH, 90:10:1) to give 181 mg (22%) of 4 as a light yellow
solid, mp 171-178 °C dec. 1H NMR (DMSO-d6): 1.69-1.78
(1H, m), 1.98-2.01 (1H, m), 2.46-2.54 (1H, m), 2.66-2.77 (2H,
m), 3.02 (1H, dd, J ) 3.7, 16.8 Hz), 4.01-4.03 (1H, m), 4.84
(1H, broad s), 6.37 (2H, broad s), 7.28 (1H, dd, J ) 8 Hz), 7.48
(1H, dd, J ) 8 Hz), 7.63 (1H, d, J ) 8 Hz), 8.14 (1H, d, J ) 8
Hz). MS (CI): 215 (M + 1, 100). Anal. (C13H14N2O‚0.25H2O)
C, H, N.
An alytical an d Spectr oscopic Meth ods: HP LC Meth od
A. HPLC isocratic radioactivity analysis was performed with
an ALLTECH Econosphere CN analytical column (5 µm, 250
mm × 4.6 i.d.) in series with an Upchurch uptight precolumn
packed with C-18 pellicular media (30-40 µm). Radioactivity
was monitored online with a Radiomatic IC flow detector with
UV detection at 325 nm. The mobile phase consisted of 0.05
M ammonium formate, pH 3.1:acetonitrile (95:5, v/v). Ana-
lytes were eluted at 1 mL/min. The flow rate of the scintillant
(Flo-Scint III) was set at 3 mL/min.
HP LC Meth od B. HPLC gradient radioactivity profiling
was performed with a Waters µBONDAPAK C-18 column (10
µm, 300 mm × 3.9 i.d.) with the above precolumn, instruments,
and conditions. The mobile phase consisted of 0.1 M am-
monium acetate, pH 4.0, and acetonitrile. Analytes were
eluted at 1 mL/min with a linear gradient starting at 0%
acetonitrile for 8 min and increasing to an acetonitrile
concentration of 10% over 27 min. The acetonitrile concentra-
tion was maintained at 10% for an additional 55 min.
HP LC Meth od C. Metabolite isolation was accomplished
by gradient semipreparative HPLC with a Whatman M-9
(()-9-Am in o-3-(d im et h ylp h en ylsilyl)-1,2,3,4-t et r a h y-
d r oa cr id in e (10). The reductive deoxygenation procedure of
Lau et al.15 was followed. To a suspension of 10.34 g (77.55
mmol) of AlCl3 in 200 mL of CH2Cl2 was added in one portion
13.50 g (155.23 mmol) of borane-tert-butylamine complex, at
0 °C under nitrogen. The mixture was stirred at 0 °C for 30

3018 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Pool et al.
ODS-3 column (500 mm × 9.4 i.d.) in series with the above
precolumn. The fraction coeluting with 2 was separated from
other radiolabeled components with a linear gradient starting
at 0% acetonitrile for 10 min and increasing to a final
acetonitrile concentration of 10% over 30 min. The acetonitrile
concentration was maintained at 10% for an additional 80 min.
Analytes were eluted at 5 mL/min with UV monitoring at 325
nm.
equivalent doses. Urine collected through 24 h was analyzed
by HPLC radioactivity profiling.
Hu m a n E xp er im en t . As part of the drug development
process for 1, healthy male volunteers received single oral 40
mg free base equivalent (100 µCi) solution doses of [14C]1.
Urine collected through 24 h was analyzed by HPLC radio-
activity profiling.
LC/MS. Rat and human urine aliquots were analyzed by
thermospray LC/MS using HPLC conditions described for
methods A and B. Mass spectra were recorded on a VG Trio-2
spectrometer operated in the plasmaspray positive ion detec-
tion mode, with a source temperature of 250 °C, capillary
temperature of 225 °×b0C, and a discharge current of 0 µA.
Full scans (600-150 Da) were taken repetitively at a scan rate
Ack n ow led gm en t. We thank Susan M. Bjorge,
Robert Bonczyk, J ames Burleigh, Beverly Hanson, and
Richard Lister for their excellent technical assistance.
Refer en ces
of 1 s decade-1
.
(1) Knapp, M. J .; Knopman, D. S.; Soloman, P. R.; Pendlebury, W.
W.; Davis, C. S.; Gracon, S. I. A 30-week randomized controlled
trial of high-dose tacrine in patients with Alzheimer’s disease.
J . Am. Med. Assoc. 1994, 271, 985-991.
(2) Crismon, M. L. Tacrine: first drug approved for Alzheimer’s
disease. Ann. Pharmacother. 1994, 28, 744-751.
(3) Nordberg, A. Biological markers and the cholinergic hypothesis
in Alzheimer’s disease. Acta Neurol. Scand. 1992, Suppl. 85, 54-
58.
(4) Hakansson, L. Mechanism of action of cholinesterase inhibitors
in Alzheimer’s disease. Acta Neruol. Scand. 1993, Suppl. 88,
7-9.
(5) Watkins, P. B.; Zimmerman, H. J .; Knapp, M. J .; Gracon, S. I.;
Lewis, K. W. Hepatotoxic effects of tacrine administration in
patients with Alzheimer’s disease. J . Am. Med. Assoc. 1994, 271,
992-998.
NMR Sp ectr oscop y. Samples for NMR were prepared by
dissolving the lyophilized metabolite in 0.5 mL of 99.96%
deuteriomethanol and transferring into 5 mm Wilmad 528
NMR tubes. Synthetic reference standards were similarly
prepared by dissolving 0.5 mg of compound in 0.5 mL of 99.96%
deuteriomethanol. NMR spectra were collected on a Bruker
AMX500 spectrometer. One-dimensional 1H NMR spectra
were recorded at 25 and 30 °C using a sweep width of 5530
Hz and 8K complex data points. Signal averaging over 32-
1024 scans, depending on sample concentration, was used in
conjunction with application of an exponential weighting
function (LB ) 1.0 or 2.0 Hz) to enhance signal-to-noise in the
Fourier transformed data. When necessary, frequency domain
resolution was enhanced by application of a Lorentzian-to-
Gaussian weighting function (GB ) 0.07, LB ) -5.0 Hz). Two-
dimensional correlated spectroscopy (COSY) experiments18
were acquired at either 25 or 30 oC. The COSY spectra for
metabolite fractions were obtained by recording 128 or 256 t1
blocks of 96-112 transients each (2.0 s relaxation delay) with
a time-domain digital resolution of 10.8 Hz in t2 and 23.5 or
47 Hz in t1. The intense water signal was minimized by
application of a low-power, selective rf pulse during the
relaxation delay. COSY spectra for synthetic standards were
acquired by recording 256 t1 blocks of four transients each with
a time-domain digital resolution of 7.8 Hz in t2 and 15.6 Hz in
t1 and a relaxation delay of 1.3 s. In all cases the COSY data
were treated with a sine-bell window function and zero-filled
once in t2 and twice in t1 prior to Fourier transformation and
magnitude calculation.
(6) Gracon, S. I.; Hoover, T. M.; Lewis, K. W.; Dolan-Ureno, J .;
Rieger, M. M.; Mers, S. L.; Knapp, M. J . Tacrine in Alzheimer’s
disease: efficacy and safety in
a parallel-group study. In
Alzheimer’s disease: advances in clinical and basic research;
Corain, B., Iqbal, K., Nicolini, M., Winblad, B., Wisniewski, H.,
Zatta, P., Eds.; Wiley: New York, 1993; pp 551-557.
(7) Ames, D. J .; Bhathal, P. S.; Davies, B. M.; Fraser, J . R. E.;
Gibson, P. R.; Roberts, S. Heterogeneity of adverse hepatic
reactions to tetrahydroaminoacridine. Aust. NZ J . Med. 1990,
20, 193-195.
(8) Madden, S.; Spaldin, V.; Park, B. K. Clinical pharmacokinetics
of tacrine. Clin. Pharmacokinet. 1995, 28, 449-457.
(9) Pool, W. F.; Bjorge, S. M.; Chang, T.; Woolf, T. F. Effect of
3-methylcholanthrene induction on the disposition of the cogni-
tion activator tacrine in rat. ISSX Proceedings, Fourth North
American ISSX Meeting, 1992; p 163.
(10) Pool, W. F.; Bjorge, S. M.; Chang, T.; Woolf, T. F. Metabolic
disposition of the cognition activator tacrine in man: identifica-
tion of phenol glucuronide metabolites in urine. ISSX Proceed-
ings, Fourth North American ISSX Meeting, 1992; p 164.
(11) Pool, W. F.; Bjorge, S. M.; Reily, M. D.; Chang, T.; Woolf, T. F.
Metabolic disposition of the cognition activator tacrine in dog:
identification of dihydroxy metabolites including a dihydrodiol.
ISSX Proceedings, Fourth North American ISSX Meeting, 1992;
p 162.
(12) Hsu, R. S.; Shutske, G. M.; Dileo, E. M.; Chesson, S. M.; Linville,
A. R.; Allen, R. C. Identification of the urinary metabolites of
tacrine in the rat. Drug Metab. Dispos. 1990, 18, 779-783.
(13) Shutske, G. M.; Bores, G. M.; Bradshaw, K. C.; Huger, F. P.;
Kapples, K. J .; Larsen, R. D.; Rush, D. K.; Tomer, J . D. Synthesis
and biological activity of putative mono-hydroxylated metabolites
of velnacrine. Bioorg. Med. Chem. Lett. 1992, 2, 865-70.
(14) Oliver, J . E.; Waters, R. M.; Lusby, W. R. Synthesis of 2-do-
decanoyl-3,5-dihydroxy-2-cyclohexen-1-one. Tetrahedron 1990,
46, 1125-1130.
(15) Lau, C. K.; Tardiff, S.; Dufresne, C.; Scheigetz, J . Reductive
deoxygenation of aryl aldehydes and ketones by tert-butylamine-
borane and aluminum chloride. J . Org. Chem. 1989, 54, 491-
494.
(16) Ashour, M.-B, A.; Gee, S. J .; Hammock, B. D. Use of a 96-well
microplate reader for measuring enzyme activities. Anal. Bio-
chem. 1987, 166, 353-360.
(17) McNally, W. P.; Pool, W. F.; Sinz, M. W.; DeHart, P.; Ortwine,
D. F.; Huang, C. C.; Chang, T.; Woolf, T. F. Distribution of
tacrine and metabolites in rat brain and plasma following single
and multiple dose regimens: evidence for accumulation of
tacrine in rat brain tissue. Drug Metab. Dispos. 1996, 24, 628-
633.
(18) Bax, A.; Freeman, R. Investigation of complex networks of spin-
spin coupling by two-dimensional NMR. J . Magn. Reson. 1981,
44, 542-561.
Bioch em istr y Meth od s: Acetylch olin ester a se Assa y.
A microplate colorimetric Ellman assay16 was used for the
determination of enzyme activity. The substrate solution
contained 500 µM acetylthiocholine and 1 mM 5,5′-dithiobis-
(2-nitrobenzoic acid) in 50 mM sodium phosphate buffer, pH
8.0. The assays were initiated by adding 100 µL of an
appropriate dilution of acetylcholinesterase (AchE) (1:20 for
mouse and rat, 1:100 for monkey and human) to 200 µL of
substrate solution containing concentrations of inhibitor rang-
ing from 100 µM to 1 nM. Blank (background) values were
determined by the addition of buffer with no enzyme to the
substrate solution. All assays were performed in triplicate.
The microplates were read in the kinetic mode of a Molecular
Devices Thermomax microplate reader at room temperature
and set for 405 nm wavelength. The dose of each compound
producing 50% inhibition of enzyme activity (IC50) was ex-
trapolated from the dose-response curve. Human AchE (type
XIII) was obtained from human red blood cell ghosts purchased
from Sigma Chemical Co. Red blood cells from monkey, rat,
and mouse were collected after washing in isotonic saline and
then osmotically lysed in distilled water. The membranes were
pelleted at 20000g for 30 min at 4 °C. The AchE was extracted
from the membranes by homogenizing in 5 vol of 50 mM borate
buffer, pH 8.0, containing 1% Triton X-100. The homogenate
was centrifuged at 10000g for 30 min at 4 °C. The supernatant
was removed, aliquoted in 1 mL Eppendorf tubes, and stored
at -70 °C until needed. All other assay materials were
obtained from Sigma Chemical Co.
Meta bolism Stu d ies: Ra t Exp er im en ts. Fasted male
Wistar rats were administered [14C]1 at either 18 mg/kg (120
µCi) or 2 mg/kg (13 µCi) as single oral solution free base
J M9602266