Synthesis and biological activity of an optically pure 10-spirocyclopropyl analog of huperzine A

Article abstract of DOI:10.1039/a801748d

The synthesis of a spirocyclic analog of huperzine A that bears a cyclopropane ring at its 10-position has been carried out in an enantioselective manner using a diastereoselective Michael-aldol reaction; in assays of AChE inhibition, this compound was found to be nearly as active as huperzine A itself, with comparable on and off rates from the enzyme.

Full text of DOI:10.1039/a801748d

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and biological activity of an optically pure 10-spirocyclopropyl
analog of huperzine A
Alan P. Kozikowski,*a† K. R. C. Prakash,a Ashima Saxenab and Bhupendra P. Doctorb
a Georgetown University Medical Center, Drug Discovery Program, Institute for Cognitive and Computational Sciences, 3970
Reservoir Road, N.W., Washington, DC 20007-2197, USA
b Division of Biochemistry, Walter Reed Army Institute of Research, Washington, DC 20307, USA
The synthesis of a spirocyclic analog of huperzine A that
bears a cyclopropane ring at its 10-position has been carried
out in an enantioselective manner using a diastereoselective
Michael–aldol reaction; in assays of AChE inhibition, this
compound was found to be nearly as active as huperzine A
itself, with comparable on and off rates from the enzyme.
cyclohexanone 4 was transformed to the b-keto ester 7a
employing conditions identical to those reported previously.1,2
Transesterification of 7a with (2)-8-phenylmenthol gave the
desired ester derivative 7b in 75% overall yield.8 Next, a
diastereoselective Michael–aldol reaction of 7b with methacro-
lein was carried out. Under optimal conditions, the reaction was
run at 220 °C over a two day period. The alcohol mixture 8a
that formed was converted to a 12.5:1 mixture (ratio from 1H
NMR analysis) of the olefins 9a and 9b, respectively, in 70%
yield by NaOAc–HOAc induced elimination of the derived
mesylate 8b. These olefins were readily separable by column
chromatography on silica gel using 1:9 ethyl acetate–hexanes
as eluent. The major isomer 9a was subjected to a Wittig
olefination reaction to afford the Z-isomer 10a. Isomerization of
the double bond using thiophenol–AIBN gave a mixture of the
E- and Z-isomers 10b and 10a in a 6:1 ratio, respectively. The
ester group of this mixture was reduced to alcohols 11a and 11b
by treatment with LAH in THF, and these isomers were
separated by column chromatography. The enantiomeric purity
of the major isomer 11b was confirmed at this stage by
transforming it to its Mosher ester derivative. 1H and 19F NMR
spectroscopy revealed the compound to be of at least 98%
optical purity. Jones oxidation of the alcohol 11b afforded the
acid 12, which was converted to urethane 13 by Curtius
rearrangement. Lastly, the required compound 2 was obtained
in 75% yield by the removal of the protecting groups from 13
using TMSI in CHCl3 at reflux. The optical rotation of
crystalline 2 (mp 235 °C) was found to be 273 (c 0.83, CHCl3).
For purposes of biological comparison, we also prepared the
spirocyclopropyl analog in racemic form, using basically the
same approach as above but omitting the transesterification
reaction with 8-phenylmenthol.
Huperzine A (HA), a potent reversible inhibitor of acetyl-
Me
Me
7
8
6
H
N
H
N
1
10
O
O
9
11
2
Me
Me
5
3
4
NH2
NH2
Huperzine A (HA) 1
Spirocyclopropyl Analog 2
cholinesterase (AChE), is an important psychotherapeutic agent
for improving cognitive function in Alzheimer’s patients by
enhancement of central cholinergic tone.1 Because of the
tremendous promise this alkaloid holds for the palliative
treatment of a disease that afflicts millions of individuals
worldwide, we and others have been engaged in an intensive
effort to explore the structure–activity relationships of this
alkaloid.2 From biological studies with mammalian AChE,
Torpedo AChE, mammalian BChEs (butyrylcholinesterases)
and mouse AChE mutants, together with molecular modeling
studies, mammalian Tyr337(330) and Trp86(84) have been
implicated in the binding of HA to AChE.3 This particular
+
interaction is of the cation (NH3 )–p type,4 while other amino
acid residues appear to participate in hydrogen bonding to the
pyridone NH and the carbonyl group.5 The superior inhibition
properties of HA have been attributed to the very slow
dissociation (t0.5 = 35 min) of the AChE–huperzine A complex
in solution.6 To date, we have identified several analogs of HA
that have comparable or better activity than the parent structure.
In particular, the C-10 axial methyl analog of HA was found to
be about 8-fold more potent than HA, whereas the 10,10-dime-
thyl analog was found to possess comparable activity. In
exploring further modifications to this region of the molecule,
we felt that it would be of interest to examine a spirocyclic
analog bearing a cyclopropyl group at C-10. Such an analog can
be viewed as a ring constrained version of the dimethyl
derivative. The possibility exists, however, that its activity
might be improved due both to the smaller size of the
cyclopropyl group and possible electronic effects which may be
capable of enhancing interactions with the enzyme.
The chemical pathway that was followed to assemble the
cyclopropyl analog 2 is shown in Schemes 1 and 2. The key
intermediate 7b was obtained in good yield through a sequence
of steps starting from the reaction of cyclohexanedione
monoethylene ketal 3 with (2-chloroethyl)dimethylsulfonium
iodide/ButOK in ButOH at room temperature to afford the
spirocyclohexanone 4 in 60% yield (Scheme 1).7 The spiro-
The biological activity of the analogs of huperzine A was
evaluated using AChE purified from fetal bovine serum.9 AChE
O
O
H
N
O
i
ii
O
O
O
O
O
O
5
3
4
iii
N
OMe
N
OMe
iv
O
O
v
O
CO2R
7a R = Me
b R = 8-Phenylmenthyl
6
Scheme 1 Reagents and conditions: i, Me2S+CH2CH2Cl I2, KI, KOBut,
ButOH, room temp.; ii, ethyl propiolate, MeOH–NH3, 100 °C, 300 psi; iii,
MeI, Ag2CO3, CHCl3, reflux; iv, MeOH–HCl, reflux, then (MeO)2CO,
NaH, THF, reflux; v, (2)-8-phenylmenthol, TsOH, benzene, reflux
Chem. Commun., 1998
1287

View Article Online
Table 1 Kinetic and inhibition parameters for huperzine A, 10,10-dime-
thylhuperzine A and 10-spirocyclopropylhuperzine A
Me
N
OMe
kon/1026
m21
koff
/
KIa/ KIb/
Me
Inhibitor
min21
min21 nm
nm
O
R′O
CO2R
i
9a
N
OMe
7b
(2)-Huperzine A
4.2
0.76
1.0
0.016 3.9
0.01 13.2 17.0
0.014 14.0 12.4
5.6
iii
Me
(±)-10,10-Dimethylhuperzine A
(±)-10-Spirocyclopropylhuperzine A
O
CO2R
(2)-10-Spirocyclopropylhuperzine A 2.4
0.015 6.4
8.8
MeO
N
8a R′ = H
b R′ = Ms
ii
a KI = Koff/Kon b Determined by the steady state method.
.
O
RO2C
9b
Me
Me
inhibition constants and kinetic parameters, and as expected, it
is slightly more active than the 10,10-dimethyl analog (com-
parison of racemic materials). Further studies are now under-
way to examine whether these 10-substituted analogs show
greater elements of neuroprotection from glutamate toxicity
than does HA itself. The neuroprotective aspect of HA
represents a newly discovered property of this molecule.11 This
pharmacological property together with the AChE inhibitory
activity further enhances the value of huperzine A and its
analogs as therapeutic agents for the treatment of Alzheimer’s
disease.11
iv
vi
N
OMe
9a
N
OMe
Me
Me
CO2R
CH2OH
10a Z-isomer
b E-isomer
11a Z-isomer
b E-isomer
v
Me
Me
vii
viii
N
OMe
N
OMe
We are indebted to the Department of Defense (DMAD17-
93-V-3018) for partial support of these studies.
11b
Me
Me
CO2H
NHCO2Me
Notes and References
12
13
† E-mail: kozikowa@giccs.georgetown.edu
ix
Me
1 J.-S. Liu, Y.-L. Zhu, C.-M. Yu, Y.-Z. Zhou, Y.-Y. Han, F.-W. Wu and
B.-F. Qi, Can. J. Chem., 1986, 64, 837; Y. Xia and A. P. Kozikowski,
J. Am. Chem. Soc., 1989, 111, 4116 and references cited therein; S.-S.
Xu, Z.-X. Gao, Z. Weng, Z.-M. Du, W.-A. Xu, J.-S. Yang, M.-L. Zhang,
Z.-H. Tong, Y. S. Fang, X.-S. Chai and S.-L. Li, Acta Pharmacol. Sin.,
1995, 16, 391.
H
N
O
Me
NH2
2
2 G. Campiani, L.-Q. Sun, A. P. Kozikowski, P. Aagaard and M.
McKinney, J. Org. Chem., 1993, 58, 7660; A. P. Kozikowski, G.
Campiani, V. Nacci, A. Sega, A. Saxena and B. P. Doctor, J. Chem.
Soc., Perkin Trans. 1, 1996, 1287; A. P. Kozikowski, G. Campiani,
L.-Q. Sun, S. Wang, A. Saxena and B. P. Doctor, J. Am. Chem. Soc.,
1996, 118, 11 357.
Scheme 2 Reagents and conditions: i, CH2NCMeCHO, (Me2N)2CNNH,
CH2Cl2, 220 °C; ii, MeSO2Cl, Et3N, DMAP, CH2Cl2, room temp.; iii,
NaOAc, AcOH, reflux; iv, Ph3P+Et Br2, KOBut, THF, room temp.; v,
PhSH, AlBN, toluene, reflux; vi, LAH, THF, reflux; vii, Jones reagent,
acetone, room temp.; viii, (PhO)2PON3, Et3N, toluene, reflux, 2 h, then
MeOH, reflux; ix, TMSI, CHCl3, reflux, 6 h, then MeOH, reflux
3 A. Saxena, N. Qian, I. M. Kovach, A. P.Kozikowski, Y. P. Pang, D. C.
Vellom, Z. Radic, D. Quinn, P. Taylor and B. P. Doctor, Protein Sci.,
1994, 3, 1770. The dual numbering system provides the residue number
in the species designated followed by the corresponding residue in
Torpedo AChE.
4 D. A. Dougherty and D. A. Stauffer, Science, 1990, 253, 872.
5 A. P. Kozikowski and Y. P. Pang, in Trends in QSAR and Molecular
Modeling ’92, Proceedings of the 9th European Symposium on
Structure–Activity Relationships: QSAR and Molecular Modeling, ed.
C. G. Wermuth, ESCOM Science Publishers, Leiden, The Netherlands,
1993; Y. P. Pang and A. P. Kozikowski, J. Comput. Aided Mol. Design,
1994, 8, 669; M. Raves, M. Harel, Y. P. Pang, I. Silman, A. P.
Kozikowski and J. L. Sussman, Nature Struct. Biol., 1997, 4, 57.
6 Y. Ashani, J. O. Peggins and B. P. Doctor, Biochem. Biophys. Res.
Commun., 1992, 184, 719.
7 S. M. Ruder and R. C. Ronald, Tetrahedron Lett., 1984, 25, 5501.
8 F. Yamada, A. P. Kozikowski, E. R. Reddy, Y.-P. Pang, J. H. Miller and
M. McKinney, J. Am. Chem. Soc., 1991, 113, 4695. Selected data for 2:
m/z (%) 268 (M+, 10), 225 (100); dH(CDCl3) 0.97 (2 H, br m), 1.25 (1
H, br s), 1.57 (3 H, s), 1.66 (3 H, d, J 6.6), 1.70 (1 H, m), 2.12 (1 H, d,
J 16.5), 2.23 (1 H, d, J 16.8), 2.65 (1 H, d, J 4.5), 5.46 (1 H, br s), 5.54
(1 H, q, J 12.9, 6.6), 6.35 (1 H, d, J 9.6), 7.87 (1 H, d, J 9.3); dC(CDCl3)
164.4, 146.3, 142.3, 139.8, 134.1, 122.9, 122.3, 116.6, 111.1, 77.4, 55.3,
49.7, 42.8, 29.6, 26.8, 22.6, 17.1, 12.8, 12.4.
activity was measured in 50 mm sodium phosphate, pH 8.0, at
22 °C as described previously using acetylthiocholine as the
substrate.10 The interaction of HA and its analogs with AChE
can be described by eqn. (1):6
kon
ææÆ
+
E
HUP - A
E ◊ HUP - A
¨ææ
koff
The ratio koff/kon is the dissociation constant (KI). The KI values
for the inhibition of FBS AChE with analogs of HA were
determined by equilibrating a known amount of enzyme (1–2
units ml21) with various concentrations of the analog. Plots of
percent residual activity versus [analog] were used to calculate
KI by the steady state method. The rate constant for the
inhibition of AChE was determined by diluting an appropriate
volume of stock solutions (1–2 mm) of each analog of huperzine
A into the enzyme solution (5–10 units ml21 in 50 mm sodium
phosphate, pH 8.0, containing 0.05% BSA) and measuring the
residual enzyme activity at various time intervals. Plots of
percent residual activity versus time at each concentration were
used to calculate the rate of inhibition (kon). Direct measurement
of the rate constant of regeneration of enzyme activity (koff) was
initiated by > 10 000-fold dilution of HA-inhibited AChE (2–4
mm) to ascertain that the rate of inhibition by residual inhibitor
was negligible in the reactivation medium.
9 D. De La Hoz, B. P. Doctor, J. S. Ralston, R. S. Rush and A. D. Wolfe,
Life Sci., 1986, 39, 195.
10 G. L. Ellman, D. Courtney, V. Andres and R. M. Featherstone, Biochem.
Pharmacol., 1961, 1, 88.
11 H. S. Ved, M. L. Koeinig, J. R. Dave and B. P. Doctor, NeuroReport,
1997, 8, 963.
As is apparent from an examination of kon, koff and the KIS
reported in Table 1, the optically pure 10-spirocyclopropyl
analog of HA is comparable in activity to HA itself in both its
Received in Corvallis, OR, USA, 2nd March 1998; 8/01748D
1288
Chem. Commun., 1998