Efficient asymmetric synthesis of ABT-594; a potent, orally effective analgesic

Article abstract of DOI:10.1016/S0957-4166(98)00291-2

A concise asymmetric synthesis of (R)-2-chloro-5-(2- azetidinylmethoxy)pyridine (ABT-594) is presented in which the key intermediate t-butoxycarbonyl protected (2R)-azetidinylalcohol is obtained in three steps from the dibenzyl ester of D-aspartic acid in 44% yield and >99% ee.

Full text of DOI:10.1016/S0957-4166(98)00291-2

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 2791–2794
Efficient asymmetric synthesis of ABT-594; a potent, orally
effective analgesic
John K. Lynch,a,∗ Mark W. Holladay,a Keith B. Ryther,a Hao Bai,a Chi-Nung Hsiao,b
Howard E. Morton,b Daniel A. Dickman,b William Arnold b and Steven A. King b
aNeurological and Urological Diseases Research D-47W, Pharmaceutical Products Division, Abbott Laboratories,
Abbott Park, IL 60064-3500, USA
bProcess Research and Development D-45L, Pharmaceutical Products Division, Abbott Laboratories, North Chicago,
IL 60064-4000, USA
Received 14 July 1998; accepted 15 July 1998
Abstract
A concise asymmetric synthesis of (R)-2-chloro-5-(2-azetidinylmethoxy)pyridine (ABT-594) is presented in
which the key intermediate t-butoxycarbonyl protected (2R)-azetidinylalcohol is obtained in three steps from the
dibenzyl ester of D-aspartic acid in 44% yield and >99% ee. © 1998 Elsevier Science Ltd. All rights reserved.
The natural product (−)-epibatidine (1), a (chloropyridyl)azabicycloheptane alkaloid isolated from
the skin of an Ecuadoran poison frog, is a non-opioid analgesic that is 200-fold more potent than
morphine in mice.1 The antinociceptive properties of epibatidine are mediated through interactions with
nicotinic acetycholine receptors (nAChR),1–3 however, it is toxic at doses only slightly higher than its
effective analgesic dose.4–6 The search for nAChR modulator compounds that exhibit a larger separation
between analgesic doses and those that elicit toxic effects led to the identification of (R)-2-chloro-5-
(2-azetidinylmethoxy)pyridine (ABT-594, 2).7,8 ABT-594 is 30–100 times more potent than morphine in
animal models with a significantly improved therapeutic ratio and reduced side effect liabilities relative to
epibatidine. In this communication we describe an efficient synthesis of enantiomerically pure ABT-594
from the commercially available dibenzyl ester of D-aspartic acid and 2-amino-5-hydroxypyridine.
The synthesis of ABT-594 required the construction of two primary fragments, namely 2-chloro-
5-hydroxypyridine 3, and the t-butoxycarbonyl protected (2R)-hydroxymethylazetidine 4 (Scheme 1).
∗
Corresponding author. E-mail: John.K.Lynch@Abbott.com
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00291-2

2792
J. K. Lynch et al. / Tetrahedron: Asymmetry 9 (1998) 2791–2794
2-Chloro-5-hydroxypyridine 3 was readily produced in two steps from 2-chloro-5-aminopyridine 5 by
modification of a literature method for closely related chlorohydroxypyridines.9 In contrast, an efficient
synthesis of the azetidine portion of the molecule proved to be more challenging, with known routes
to appropriate (R)-azetidine precursors being problematic. Thus, as reported previously, a route to a
protected form of 6 from D-methionine10,11 resulted in substantial racemization in our hands.12 In
addition, a synthesis from γ-butyrolactone which required optical resolution13,14 gave enantiomerically
pure material but was lengthy and proceeded in poor overall yield. Furthermore, these routes required
three additional steps to convert the protected form of azetidine carboxylic acid 6 to the desired t-
butylcarbamate azetidinyl alcohol 4.
Scheme 1.
A better approach which utilized the commercially available or easily prepared dibenzyl ester of
D-aspartic acid15 7 and proceeded through the enantiomer of a known β-lactam intermediate was
then explored (Scheme 2). Treatment of 7 with trimethylsilylchloride and triethylamine in diethyl
ether, removal of the triethylamine hydrochloride by filtration, followed by addition of t-BuMgCl (2
M in Et2O) according to the literature precedent16 produced the β-lactam 8 in highly variable yields
(20–70%) on a small scale. The inconsistent yields were due to the high water sensitivity of the N-
trimethylsilyl intermediate, which necessitated filtration under anhydrous conditions.17 This proved even
more difficult on a large scale and therefore modifications were explored. After some experimentation,
it was discovered that filtration of the ammonium salt could be avoided if the solvent was changed
to CH2Cl2 and two equivalents of t-BuMgCl were used. Using this process, the β-lactam 8 could be
reproducibly formed in 70% yield. The ester and lactam carbonyls of 8 were reduced with excess lithium
aluminum hydride and the resulting amine protected as the t-butylcarbamate to give 4 in 60% yield18 and
>99% ee as demonstrated by chiral HPLC.19 Thus, this route to enantiomerically pure t-butoxycarbonyl
protected (2R)-hydroxymethylazetidine 4 proceeds in three steps and 44% overall yield, both of which
are significant improvements over previously reported syntheses.
Scheme 2.
Synthesis of 2-chloro-5-hydroxypyridine 3 and coupling of the primary fragments is shown in
Scheme 3. Treatment of 2-chloro-5-aminopyridine 5 with boron trifluoride etherate and t-butylnitrite
generated the diazonium salt which was decomposed in acetic anhydride9 to give 2-chloro-5-
acetoxypyridine. This acetate was then subjected to potassium carbonate in methanol to produce
2-chloro-5-hydroxypyridine 3 in 53% overall yield for the two steps, a considerable improvement over
published syntheses.8,20 In earlier accounts, the ether forming step to produce compounds such as 11
were always carried out under Mitsunobu conditions.8,12 This was due in part to a report in which the
tosylates of N-t-butoxycarbonyl prolinol underwent intramolecular cyclization and loss of t-butyl to
generate the fused 5,5-bicyclic carbamate under basic conditions.21 However, this undesired reaction
appears to be slow with the azetidinyl alcohols and we were able to mesylate 4 to give 10 followed by
coupling with 3 in the presence of NaOH in DMF to produced 11 in 92% yield for the two steps.

J. K. Lynch et al. / Tetrahedron: Asymmetry 9 (1998) 2791–2794
2793
Scheme 3.
Removal of the t-butylcarbamate of 11 using trifluoroacetic acid followed by chromatographic purifi-
cation provided the free amine of ABT-594 in moderate yield on a small scale (60–80%) and gave a
significant byproduct on a large scale which proved difficult to remove. In addition, it was discovered
that the initially selected salt forms of ABT-594 (mono or bishydrochloride salts) were hygroscopic and
partially decomposed after the absorption of water to give significant quantities of the same byproduct.
This undesired material was identified as dimer 13 which presumably resulted from the addition of 12
to the 4-position of another azetidine (Scheme 4). Therefore, it was critically important to find a stable,
non-hygroscopic salt form and to remove the t-butoxycarbonyl group under conditions which prevented
dimer formation. After generating and examining several salt forms, it was found that the tosylate salt
provided the desired properties.22 Subsequently, it was discovered that deprotection and salt formation
could be accomplished in a single step by heating 11 in ethanol with a slight excess of p-toluenesulfonic
acid hydrate. After allowing the reaction to cool and addition of ethyl acetate, crystalline toluenesulfonic
acid salt of ABT-594 was produced in 88% yield. This exceptionally efficient sequence produced pure
final product in 81% overall yield from t-butoxycarbonyl protected (2R)-hydroxymethylazetidine 4.
Scheme 4.
In conclusion, the synthesis of ABT-594 from readily available starting materials was accomplished
in six steps, in 36% overall yield and with an enantiomeric excess >99%. This route includes a concise
synthesis of a protected form of (2R)-hydroxymethylazetidine 4 from the dibenzyl ester of D-aspartic
acid, an improved construction of 2-chloro-5-hydroxypyridine 3, and an efficient assembly and final
deprotection sequence.
References
1. Spande, T. F.; Garaffo, H. M.; Edwards, M. W.; Yeh, H. J. C.; Pannell, L.; Daly, J. W. J. Am. Chem. Soc. 1992, 114,
3475–3478.
2. Qian, C.; Li, T.; Shen, T. Y.; Libertine-Garahan, L.; Eckman, J.; Biftu, T.; Ip, S. Eur. J. Pharmacol. 1993, 250, R13–R14.

2794
J. K. Lynch et al. / Tetrahedron: Asymmetry 9 (1998) 2791–2794
3. Badio, B.; Daly, J. W. Mol. Pharm. 1994, 45, 563–569.
4. Sullivan, J. P.; Decker, M. W.; Brioni, J. D.; Donnelly-Roberts, D.; Anderson, D. J.; Bannon, A. W.; Kang, C.-H.; Adams,
P.; Piattoni-Kaplan, M.; Buckley, M.; Gopalakrishnan, M.; Williams, M.; Arneric, S. P. J. Pharmacol. Exp. Ther. 1994,
271, 624–631.
5. Bonhaus, D. W.; Bley, K. R.; Broka, C. A.; Fontana, D. J.; Leung, E.; Lewis, R.; Shieh, A.; Wong, E. H. F. J. Pharmacol.
Exp. Ther. 1995, 273, 1199–1203.
6. Rao, T. S.; Correa, L. D.; Reid, R. T.; Lloyd, G. K. Neuropharmacology 1996, 35, 393–405.
7. Bannon, A. W.; Decker, M. W.; Holladay, M. W.; Curzon, P.; Donnelly-Roberts, D.; Puttfarcken, P. S.; Bitner, R. S.; Diaz,
A.; Dickenson, A. H.; Porsclt, R. D.; Williams, M.; Arneric, S. P. Science 1998, 279, 77–80.
8. Holladay, M. W.; Wasicak, J. T.; Lin, N.-H.; He, Y.; Ryther, K. B.; Bannon, A. W.; Buckley, M. J.; Kim, D. J. B.; Decker,
M. W.; Anderson, D. J.; Campbell, J. E.; Kuntzweiler, T. A.; Donnelly-Roberts, D. L.; Piattoni-Kaplan, M.; Briggs, C. A.;
Williams, M.; Arneric, S. P. J. Med. Chem. 1998, 41, 407–412.
9. Koch, V.; Schnatterer, S. Synthesis 1990, 499–501.
10. Miyoshi, M.; Sugano, H.; Fujii, T.; Ishihara, T.; Yoneda, N. A. Chem. Lett. 1973, 5–6.
11. Sugano, H.; Miyoshi, M. Bull. Chem. Soc. Jpn 1973, 46, 669–670.
12. Abreo, M. A.; Lin, N.-H.; Garvey, D. S.; Gunn, D. E.; Hettinger, A.-M.; Wasicak, J. T.; Pavlik, P. A.; Martin, Y. C.;
Donnelly-Roberts, D. L.; Anderson, D. J.; Sullivan, J. P.; Williams, M.; Arneric, S. P.; Holladay, M. W. J. Med. Chem.
1996, 39, 817–825.
13. Rodebaugh, R. M.; Cromwell, N. H. J. Heterocycl. Chem. 1969, 6, 435–437.
14. Rodebaugh, R. M.; Cromwell, N. H. J. Heterocycl. Chem. 1969, 6, 993–994.
15. Zervas, L.; Winitz, M.; Greenstein, J. P. J. Org. Chem. 1957, 22, 1515.
16. Salzmann, T. N.; Ratcliffe, R. W.; Christensen, B. G.; Bouffard, F. A. J. Am. Chem. Soc. 1980, 102, 6161–6163.
17. Baldwin, J. E.; Adlington, R. M.; Gollins, D. W.; Schofield, C. J. Tetrahedron 1990, 46, 4733–4748.
18. Alternatively, removal of the benzyl group of 8 by hydrogenation (Pd/C, 4 atm hydrogen, THF) followed by LAH reduction
produced 4 in 45–55% yield. Although lower yielding, this process eliminated the need for separation of benzyl alcohol
by flash silica gel chromatography.
19. Chiralpak AD column, 90:10 hexane:EtOAc (1.0 mL/min) at 210 nm. Michael Fitzgerald and Michael Rasmussen (D45L)
are gratefully acknowledged for developing the method and running the analyses.
20. Effenberger, F.; Krebs, A.; Willrett, P. Chem. Ber. 1992, 125, 1131–1140.
21. Curran, T. P.; Pollastri, M. P.; Abelleira, S. M.; Messier, R. J.; McCollum, T. A.; Rowe, C. G. Tetrahedron Lett. 1994, 35,
5409–5412.
22. Dr. Devalina Law of Abbott Laboratories (Pharmaceutics, D4P3) is gratefully acknowledged for evaluation of the physical
properties of the different salts of ABT-594.