A practical conversion of natural physostigmine into the potent butyrylcholinesterase inhibitor N1,N8-bisnorcymserine

Article abstract of DOI:10.1016/S0040-4039(00)00740-1

A rapid novel synthetic route to the potent reversible butyrylcholinesterase inhibitor (-)-N¹,N⁸-bisnorcymserine (1) is reported from physostigmine (2) in a 20% total yield. Details on the formation of the imino-quinone 6 obtained in the oxidation of N¹-benzylnoresermethole (4) and its conversion into N¹-bisnoreseroline (7) are given. As expected, the product of this synthesis, (1), had identical biological activity to the same agent produced by total synthesis. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00740-1

Tetrahedron Letters 41 (2000) 4861±4864
A practical conversion of natural physostigmine into the
potent butyrylcholinesterase inhibitor N1,N8-bisnorcymserine
Xiaoxiang Zhu,a Nigel H. Greig,a,* Harold W. Holloway,a Noel F. Whittaker,b
Arnold Brossic and Qian-sheng Yua
aDrug Design & Development, Laboratory of Neurosciences, Gerontology Research Center (4E02), National Institute
on Aging Intramural Research Program, National Institutes of Health, 5600 Nathan Shock Dr., Baltimore,
Maryland 21224-6825, USA
bLaboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Disease, National
Institute of Health, Bethesda, Maryland 20892, USA
cSchool of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-7361, USA
Received 30 March 2000; revised 28 April 2000; accepted 2 May 2000
Abstract
A rapid novel synthetic route to the potent reversible butyrylcholinesterase inhibitor (^)-N1,N8-bis-
norcymserine (1) is reported from physostigmine (2) in a 20% total yield. Details on the formation of the
imino-quinone 6 obtained in the oxidation of N1-benzylnoresermethole (4) and its conversion into N1-bis-
noreseroline (7) are given. As expected, the product of this synthesis, (1), had identical biological activity to
the same agent produced by total synthesis. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: butyrylcholinesterase; Alzheimer's disease therapeutic; physostigmine; anticholinesterase; N1,N8-bisnor-
cymserine.
Neuroscience and molecular biology studies show that inappropriate butyrylcholinesterase
(BChE) activity increases the risk and/or progression of Alzheimer's disease.1 3 Based on this,
our group initiated studies to design and synthesize novel, potent, and highly selective reversible
inhibitors of BChE to provide a candidate for therapeutic development and to test the novel
hypothesis that BChE inhibitors are of bene®t in the treatment of Alzheimer's disease.4 (^)-
N1,N8-Bisnorcymserine (1) proved to be one of the most potent and interesting inhibitors of
human BChE in an initial pharmacological evaluation.4 However, until now, synthesis of this
optically pure compound could be accomplished only by tedious multiple step chemistry, which
included the resolution of the required 3S enantiomer at an intermediate stage.4 7 It was therefore
desirable to develop an ecient synthetic method to rapidly gain a large quantity of this optically
* Corresponding author. Fax: 410-558-8278; e-mail: greign@vax.grc.nia.nih.gov
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00740-1

4862
active compound for use in further pharmacological studies. We report, herein, the synthesis of
(^)-N1,N8-bisnorcymserine (1) from the commercially available alkaloid, physostigmine (2).
(^)-Physostigmine (2) was treated with sodium n-butoxide in n-butanol to give eseroline (3).8
(^)-Eseroline (3) then was puri®ed and isolated as its fumarate salt, and, thereafter, was converted
into N1-benzylnoresermethole (4), according to a known procedure9 11 (Scheme 1).
Scheme 1. (1) Ref. 7, 89%; (2) (a) CH3I, KOH, DMSO (b) C6H5CH2NH2, total 66%
In order to prepare the desired N1,N8-bisnor derivatives, selective N8-demethylation of
compound 4 is a key step. It was reported12 that oxidation of N8-methyl to N8-formyl, followed
by hydrolysis with diluted hydrochloric acid a€orded N8-norphysostigmine. Following the
previous methods, the oxidation of N1-benzylnoresermethole (4) gave the N8-formyl compound
5
13 in 25 and 36% yields, respectively, when pyridinium dichromate (PDC) and Collins' reagent14
were used as oxidants in dichloromethane. This low yield is an obstacle in the ecient conversion
of physostigmine (2) to the target compound (1), particularly in light of the fact that the N8-
formyl compound 5 proved to be a mixture of geometrical isomers in a ratio of 1:3. These
1
were clearly detected in the H NMR spectrum, and showed as two close spots by TLC, but,
nevertheless, were dicult to separate.6,7
With some modi®cation, the yield rose to 45% when the oxidation of N1-benzylnoresermethole
(4) was undertaken with PDC in an ice bath and in the presence of a weak base, such as pyridine
or sodium bicarbonate. We additionally found that a new compound, the imino-quinone 6,15 was
simultaneously generated and we isolated it from the reaction mixture by column chromato-
graphy in a 19% yield.
The following is a detailed experimental procedure for the oxidation of N1-benzylnoresermethole
(4): To a solution of 4 (1.54 g, 4.99 mmol) in CH2Cl2 (75 mL) was added NaHCO3 (1 g). The
mixture was stirred vigorously and cooled in an ice bath. Pyridinium dichromate (3.76 g, 9.99
mmol) was then added, and the mixture was stirred for 2 h. The reaction mixture was ®ltered and
the resulting solid was washed with CH2Cl2 (50 mL). The combined CH2Cl2 solution was washed
with water (3Â50 mL), dried over anhydrous Na2SO4 and concentrated under vacuum. The
residue was puri®ed by column chromatography (silica gel) using petroleum ether:ethyl acetate
(10:1 and 10:2) as the eluent to obtain starting material 4 (0.15 g) and 5 (0.65 g, 45%), then using
petroleum ether:ethyl acetate (4:1) as the eluent to obtain the more polar component 6 (0.235 g,
19%).
Compound 5 then was reacted with BBr3 in dichloromethane for one hour at room temperature
and methanol was added to destroy any remaining BBr3. The resulting solution was evaporated
to give a residue that was puri®ed by ¯ash chromatography to a€ord N1-benzylbisnorseroline (7),
with simultaneous hydrolysis of the N-formyl group. The latter was unexpected and provided a
valuable one pot reaction to e€ect both O-demethylation and N-deformylation, resulting in
compound 7 in a yield of 77% (Scheme 2).

4863
Scheme 2. (1) PDC, NaHCO3, CH2Cl2; (2) BBr3, CH2Cl2, CH3OH 77%; (3) NaBH4, THF, 85%; (4) BBr3, CH2Cl2,
91%; (5) Na, anhydrous ether, 4-isopropylphenyl isocyanate, 97%; (6) H2, Pd(OH)2/C, 70%
In the reaction of compound 5 with BBr3, carefully controlling the addition of methanol to
destroy remaining BBr3 (speci®cally, by using approximately an equivalent of methanol and
maintaining the temperature at 0ꢀC in an ice bath), followed by dilution of the reaction mixture
with dichloromethane and washing the CH2Cl2 solution with water to remove acid resulted in
O-demethylation alone to give compound 817 in 91% yield.
Compound 6 was reduced with sodium borohydride in THF to a€ord the useful intermediate
N1-benzylbisnorseroline (7)16 in 85% yield.
Compound 7 is a key intermediate from which the N1,N8-bisnorcymserine was readily
prepared. Reaction of 4-isopropylphenyl isocyanate with compound 7 a€orded N1-benzyl-N8-
norcymserine (9).18 N-Debenzylation of compound 9 was accomplished by catalytic hydrogenation
using Pd(OH)2/C as a catalyst and iso-propanol as a solvent to give N1,N8-bisnorcymserine (1).4
As both of the products of the oxidation of N1-benzylnoresermethole (4), compounds 5 and 6,
were converted into N1-benzylbisnorseroline (7), the synthesis of N1,N8-bisnorcymserine (1) from
physostigmine (2) was achieved in an overall yield that totalled 20%. Chemical characterization
demonstrated that compound 1, prepared as described herein, was chemically identical in every
respect to the same material prepared by total synthesis.4 An ex vivo assay4 11 then was under-
taken to quantify and compare the activity of N1,N8-bisnorcymserine (1), synthesized by both
routes, to inhibit human BChE and AChE prepared freshly from plasma and erythrocytes,
respectively. As expected, compound 1, prepared as described herein and by total synthesis,4 had

4864
identical anticholinesterase activity (with IC50, concentration required to inhibit 50% enzyme
action, values for BChE and AChE being 1.0 and 110 nM, respectively).
References
1. Guillozet, A.; Smiley, J. F.; Mash, D. C.; Mesulam, M. M. Ann. Neurol. 1997, 42, 909±918.
2. Barber, K. L.; Mesular, M. M.; Kraft, G. A.; Klein, W. L. Proc. Soc. Neurosci. 1996, 22, 1172.
3. Lehman, D. J.; Johnston, C.; Smith, A. D. Hum. Mol. Genet. 1997, 6, 1933±1936.
4. Yu, Q. S.; Holloway, H. W.; Utsuki, T.; Brossi, A.; Greig, N. H. J. Med. Chem. 1999, 42, 1855±1861.
5. Pei, X. F.; Yu, Q. S.; Lu, B. Y.; Greig, N. H.; Brossi, A. Heterocycles 1996, 42, 229±236.
6. Yu, Q. S.; Greig, N. H.; Holloway, H. W.; Brossi, A. J. Med. Chem. 1998, 41, 2371±2379.
7. Yu, Q. S.; Pei, X. F.; Holloway, H. W.; Greig, N. H.; Brossi, A. J. Med. Chem. 1997, 40, 2895±2901.
8. Yu, Q. S.; Schonenberger, B.; Brossi, A. Heterocycles 1987, 26, 1271±1275.
9. Pei, X. F.; Greig, N. H.; Flippen-Anderson, J. L.; Bi, S.; Brossi, A. Helv. Chim. Acta 1994, 77, 1412±1422.
10. He, X. S.; Greig, N. H.; Rapoport, S.; Brossi, A. Med. Chem. Res. 1992, 2, 229±237.
11. Pei, X. F.; Greig, N. H.; Bi, S.; Brossi, A.; Toome, V. Med. Chem. Res. 1995, 5, 455±461.
12. Takano, S.; Moriya, M.; Iwabuchi, Y.; Ogasawara, K. Chem. Lett. 1990, 109±112.
22
13. Compound 5 (pale brown crystal): mp 88±92ꢀC; ‰ꢀŠD =^42.2 (c=0.102, CH2Cl2); H NMR (300 MHz, CDCl3):
1.35 (s, 1/3 3H, C3a-CH3), 1.41 (s, 2/3 3H, C3a-CH3), 1.91±2.04 (m, 2H, H-3), 2.40±2.75 (m, 2H, H-2), 3.74 (s, 3H,
O-CH3), 3.85±4.15 (m, 2H, CH2Ph), 4.86 (s, 2/3H, H-8a), 5.22 (s, 1/3 H, H-8a), 6.61±6.71 (m, 2H), 8.12 (s, 2/3H,
N-CHO), 8.82 (s, 1/3H, N-CHO); HRMS: calcd: 322.1681 (M+), found: 322.1679.
1
14. Corey, E. J.; Balanson, R. D. J. Am. Chem. Soc. 1974, 96, 6516±6517.
22
1
15. Compound 6 (pale yellow syrup): ‰ꢀŠD =^201.4 (c=0.209, CH2Cl2); H NMR (300 MHz, CDCl3): 1.32 (s, 3H,
C3a-CH3), 1.63±1.71 (m, 1H, H-3a), 1.85±1.96 (m, 1H, H-3b), 2.20±2.31 (m, 1H, H-2a), 2.15±2.24 (m, 1H, H-2b),
3.80 (d, J=13.4 Hz, 1H, N-CH2Ph), 4.21 (d, J=13.4 Hz, 1H, N-CH2Ph), 5.02 (s, 1H, H-8a), 6.12 (d, J=1.7 Hz,
1H, H-4), 6.54 (dd, J=1.7, 20 Hz, 1H, H-6); 13C NMR (75 MHz, CDCl3): 23.30, 39.79, 49.70, 50.90, 54.42, 101.47,
121.65, 126.91, 128.19 (2C), 128.51 (2C), 133.37, 135.59, 138.85, 160.84, 165.63, 167.87; UV (CH2Cl2): lmax=268
nm, lmax=330 nm; HRMS: calcd: 278.1419 (M+), found: 278.1421.
22
1
16. Compound 7 (foam): ‰ꢀŠD =^224.1 (c=0.054, CH2Cl2); H NMR (300 MHz, CDCl3): 1.39 (s, C3a-CH3), 1.86±
1.94 (m, 2H, H-3), 2.58±2.73 (m, 2H, H-2), 3.71±3.84 (m, 2H, CH2Ph), 4.46 (s, 1H, H-8a), 6.40±6.53 (m, 3H, H-4,
H-6 and H-7); HRMS: calcd: 280.1576 (aM+), found: 280.1578.
1
17. Compound 8 (syrup): H NMR (300 MHz, CDCl3): 1.40 (s, 1/3 3H, C3a-CH3), 1.45 (s, 2/3 3H, C3a-CH3), 1.91±
2.04 (m, 2H, H-3), 2.40±2.75 (m, 2H, H-2), 3.85±4.21 (m, 2H, CH2Ph), 4.92 (s, 2/3H, H-8a), 5.34 (s, 1/3H, H-8a),
6.61±6.71 (m, 2H), 8.20 (s, 2/3H, N-CHO), 8.89 (s, 1/3H, N-CHO).
22
1
18. Compound 9 (foam): ‰ꢀŠD =^157.3 (c=0.066, CH2Cl2); H NMR (300 MHz, CDCl3): 1.25 (d, J=6.8 Hz, 6H,
(CH3)2CH), 1.46 (s, 3H, C3a-CH3), 1.98±2.11 (m, 2H, H-3), 2.70±2.98 (m, 3H, H-2 and CHMe2), 3.80±3.90 (m,
2H, CH2Ph), 4.54 (s, 1H, H-8a), 6.50±6.90 (m, 2H); HRMS: calcd: 441.2416 (M+), found: 441.2418. Fumarate of
compound 9: white crystal, mp 137±139ꢀC.